Diet, Nutrients, and Bone Health - PDF Free Download (2024)

DIET,

NUTRIENTS, and BONE

HEALTH Edited by

John J.B. Anderson Sanford C. Garner Philip J. Klemmer

DIET,

NUTRIENTS, and BONE

HEALTH

DIET,

NUTRIENTS, and BONE

HEALTH Edited by

John J.B. Anderson Sanford C. Garner Philip J. Klemmer

Boca Raton London New York

CRC Press is an imprint of the Taylor & Francis Group, an informa business

CRC Press Taylor & Francis Group 6000 Broken Sound Parkway NW, Suite 300 Boca Raton, FL 33487-2742 © 2012 by Taylor & Francis Group, LLC CRC Press is an imprint of Taylor & Francis Group, an Informa business No claim to original U.S. Government works Version Date: 20110811 International Standard Book Number-13: 978-1-4398-1956-2 (eBook - PDF) This book contains information obtained from authentic and highly regarded sources. Reasonable efforts have been made to publish reliable data and information, but the author and publisher cannot assume responsibility for the validity of all materials or the consequences of their use. The authors and publishers have attempted to trace the copyright holders of all material reproduced in this publication and apologize to copyright holders if permission to publish in this form has not been obtained. If any copyright material has not been acknowledged please write and let us know so we may rectify in any future reprint. Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced, transmitted, or utilized in any form by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying, microfilming, and recording, or in any information storage or retrieval system, without written permission from the publishers. For permission to photocopy or use material electronically from this work, please access www.copyright.com (http:// www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood Drive, Danvers, MA 01923, 978-750-8400. CCC is a not-for-profit organization that provides licenses and registration for a variety of users. For organizations that have been granted a photocopy license by the CCC, a separate system of payment has been arranged. Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. Visit the Taylor & Francis Web site at http://www.taylorandfrancis.com and the CRC Press Web site at http://www.crcpress.com

Contents Preface...............................................................................................................................................ix Contributors.......................................................................................................................................xi

Part I  Introduction to Diet and Bone Chapter 1 Overview of Relationships between Diet and Bone...................................................... 3 John J.B. Anderson Chapter 2 Role of Lifestyle Factors in Bone Health.................................................................... 17 John J.B. Anderson and Philip J. Klemmer Chapter 3 Bone Marrow and Stem Cell Recruitment.................................................................. 23 Sumithra K. Urs and Clifford J. Rosen Chapter 4 Skeletal Tissues and Mineralization........................................................................... 33 Sanford C. Garner and John J.B. Anderson Chapter 5 Optimizing the Skeletal Benefits of Mechanical Loading and Exercise.................... 53 Stuart J. Warden and Robyn K. Fuchs Chapter 6 Hormone Actions in the Regulation of Calcium and Phosphorus Metabolism.......... 71 David A. Ontjes Chapter 7 Renal Regulation of Calcium and Phosphate Ions.................................................... 113 Philip J. Klemmer and John J.B. Anderson

Part II  Effects of Specific Nutrients on Bone Chapter 8 Calcium and Bone..................................................................................................... 121 John J.B. Anderson, Sanford C. Garner, and Philip J. Klemmer Chapter 9 Inorganic Phosphorus: Do Higher Dietary Levels Affect Phosphorus Homeostasis and Bone?............................................................................................. 141 Mona S. Calvo

vii

viii

Contents

Chapter 10 Vitamin D and Bone ................................................................................................ 157 Michael F. Holick Chapter 11 Vitamin A and Bone................................................................................................. 171 Håkan Melhus Chapter 12 Vitamin K and Bone ................................................................................................ 193 Cees Vermeer and Marjo H.J. Knapen Chapter 13 The Iron Factor in Bone Development...................................................................... 203 Denis M. Medeiros and Erika Bono Chapter 14 Micronutrients and Bone........................................................................................... 213 Elizabeth Grubert and Jeri W. Nieves Chapter 15 Dietary Protein’s Impact on Skeletal Health . .......................................................... 223 Anna K. Surdykowski, Anne M. Kenney, KarlL. Insogna, and Jane E. Kerstetter Chapter 16 Omega-3 Fatty Acids and Bone Metabolism............................................................ 233 Bruce A. Watkins, Kevin Hannon, Mark F. Seifert, and Yong Li Chapter 17 Is There a Role for Dietary Potassium in Bone Health?........................................... 259 Susan Joyce Whiting Chapter 18 Acid–Base Balance................................................................................................... 271 Susan A. Lanham-New Chapter 19 Antioxidants and Bone Health.................................................................................. 281 Martin Kohlmeier

Part III  Effects of Life Cycle Changes on Bone Chapter 20 Diet and Bone Changes in Pregnancy and Lactation................................................ 293 Frances A. Tylavsky Chapter 21 Calcium Intake Influences the Bone Response to Exercise in Growing Children..................................................................................................................... 301 Bonny L. Specker, Ramu Sudhagoni, and Natalie W. Thiex

ix

Contents

Chapter 22 Obesity, Adipose Tissue,andBone........................................................................... 325 Sue A. Shapses, Norman K. Pollock, and Richard D. Lewis Chapter 23 Exercise and Skeletal Growth................................................................................... 369 Adam D.G. Baxter-Jones and R.A. Faulkner Chapter 24 Peak Bone Mass: Influence of Nutrition and Lifestyle Variables............................. 381 Jennifer L. Bedford and Susan I. Barr Chapter 25 Calcium, Other Nutrients, Exercise, and Bone Health in Twins............................... 395 John D. Wark Chapter 26 Nutrition and Bone in Young Adults.........................................................................403 Christel Lamberg-Allardt and Merja Kärkkäinen Chapter 27 Nutrition and Bone Health in Older Adults.............................................................. 415 Connie W. Bales, Kenlyn R. Young, and John J.B. Anderson

Part IV Race, Ethnicity, and Bone Chapter 28 Bone Growth in African Children and Adolescents................................................. 433 Ann Prentice, Kate A. Ward, Inez Schoenmakers, and Gail R. Goldberg Chapter 29 Skeletal Racial Differences....................................................................................... 451 Felicia Cosman and Jeri W. Nieves Chapter 30 Nutrition and Bone Health of the Japanese............................................................... 465 Takuo Fujita Chapter 31 Diet and Bone Health of the Chinese Population...................................................... 477 Suzanne C. Ho and Yu-ming Chen

Part V  Osteopenia and Osteoporosis Chapter 32 Prevention of Bone Loss with Exercise..................................................................... 493 Anna Nordström and Peter Nordström

x

Contents

Chapter 33 The Bone–Vascular Axis in Chronic Kidney Disease.............................................. 509 Philip J. Klemmer and John J.B. Anderson Chapter 34 Contribution of Clinical Risk Factors to the Assessment of Hip Fracture Risk and Treatment Decision Making............................................................................... 517 Guizhou Hu, Martin M. Root, and John J.B. Anderson

Part VI  Conclusion Chapter 35 Nutrition and Bone Health: Promotion of Bone Gain and Prevention of Bone Loss across the Life Cycle......................................................................................... 531 John J.B. Anderson, Philip J. Klemmer, and Sanford C. Garner Index............................................................................................................................................... 541

Preface This expanded and advanced treatise, Diet, Nutrients, and Bone, is a follow-up a decade and a half later of our earlier book, Calcium and Phosphorus in Health and Disease. Considerable advances in our knowledge and understanding of the roles of nutrients in skeletal development and maintenance have been made since publication of that book in 1996. The current book is an expansion of the earlier one and, in essence, an update. Several new topics, however, have been introduced, whereas greater coverage has been given to other topics. Finally, based on their own research and expertise, an impressive group of scientists has made significant contributions to this offering that is intended for graduate students and established researchers in the bone field and related areas of investigation. Public health aspects of bone health are emphasized in this book, that is, health promotion and disease prevention from a nutritional perspective. The authors thank the following for their assistance in various aspects of this book: Boyd R. Switzer, Jean C. Brown, and the reference staff of the Health Sciences Library at the University of North Carolina. John J.B. Anderson Sanford C. Garner Philip J. Klemmer

xi

Contributors John J.B. Anderson Department of Nutrition Gillings School of Global Public Health University of North Carolina Chapel Hill, North Carolina Connie W. Bales Department of Medicine Duke University School of Medicine Durham, North Carolina

Felicia Cosman Department of Medicine College of Physicians and Surgeons Columbia University New York and Regional Bone and Clinical Research Centers Helen Hayes Hospital West Haverstraw, New York

Susan I. Barr Department of Food, Nutrition, and Health University of British Columbia Vancouver, British Columbia, Canada

R.A. Faulkner College of Kinesiology University of Saskatchewan Saskatoon, Saskatchewan, Canada

Adam D.G. Baxter-Jones College of Kinesiology University of Saskatchewan Saskatoon, Saskatchewan, Canada

Robyn K. Fuchs Department of Physical Therapy School of Health and Rehabilitation Sciences Indiana University Indianapolis, Indiana

Jennifer L. Bedford Department of Food, Nutrition, and Health University of British Columbia Vancouver, British Columbia, Canada

Takuo Fujita Calcium Research Institute Katsuragi Hospital Osaka, Japan

Erika Bono Department of Human Nutrition Kansas State University Manhattan, Kansas

Sanford C. Garner Integrated Laboratory Systems, Inc. Durham, North Carolina

Mona S. Calvo Office of Applied Research and Safety Assessment Center for Food Safety and Applied Nutrition Laurel, Maryland

Gail R. Goldberg Nutrition and Bone Health Research Group MRC Human Nutrition Research Elsie Widdowson Laboratory Cambridge, United Kingdom and Keneba, The Gambia

Yu-ming Chen Department of Community and Family Medicine Prince of Wales Hospital The Chinese University of Hong Kong Hong Kong, People’s RepublicofChina

Elizabeth Grubert Department of Epidemiology and Biostatistics School of Public Health University at Albany Rensselaer, New York

xiii

xiv

Contributors

Kevin Hannon Department of Basic Medical Sciences College of Veterinary Medicine Purdue University West Lafayette, Indiana

Marjo H.J. Knapen Maastricht University and BioPartner Center Maastricht Maastricht, the Netherlands

Suzanne C. Ho Department of Community and Family Medicine Prince of Wales Hospital The Chinese University of Hong Kong Hong Kong, People’s Republic of China

Martin Kohlmeier Department of Nutrition University of North Carolina Chapel Hill, North Carolina

Michael F. Holick Department of Medicine Vitamin D, Skin and Bone Research Laboratory Boston University Medical Center Boston, Massachusetts

Christel Lamberg-Allardt Department of Food and Environmental Sciences University of Helsinki Helsinki, Finland

Guizhou Hu Department of Research and Development Biosignia, Inc. Durham, North Carolina Karl L. Insogna Department of Internal Medicine Yale University School of Medicine New Haven, Connecticut Merja Kärkkäinen Department of Food and Environmental Sciences University of Helsinki Helsinki, Finland Anne M. Kenney Allied Health Sciences University of Connecticut Storrs, Connecticut Jane E. Kerstetter Allied Health Sciences University of Connecticut Storrs, Connecticut Philip J. Klemmer UNC Kidney Center University of North Carolina ChapelHill,North Carolina

Susan A. Lanham-New Nutritional Sciences Division University of Surrey Guildford, United Kingdom Richard D. Lewis Department of Foods and Nutrition University of Georgia Athens, Georgia Yong Li Molecular Biosciences Department of Nutritional Sciences University of Connecticut Storrs, Connecticut Denis M. Medeiros School of Biological Sciences The University of Missouri-Kansas City Kansas City, Missouri Håkan Melhus School of Biological Sciences The University of Missouri-Kansas City Kansas City, Missouri

xv

Contributors

Jeri W. Nieves Department of Epidemiology Mailman School of Public Health Columbia University, New York and Regional Bone and Clinical Research Centers Helen Hayes Hospital West Haverstraw, New York Anna Nordström Department of Community Medicine and Rehabilitation and Department of Surgical and Perioperative Sciences, Sports Medicine Umeå University Umeå, Sweden Peter Nordström Department of Surgical and Perioperative Sciences, Sports Medicine and Department of Community Medicine and Rehabilitation Umeå University Umeå, Sweden David A. Ontjes Department of Medicine University of North Carolina School of Medicine Chapel Hill, North Carolina Norman K. Pollock Department of Pediatrics Georgia Health Sciences University Augusta, Georgia Ann Prentice Nutrition and Bone Health Research Group MRC Human Nutrition Research Elsie Widdowson Laboratory Cambridge, United Kingdom and Keneba, The Gambia

Martin M. Root Department of Nutrition and Health Care Management Appalachian State University Boone, North Carolina Clifford J. Rosen Maine Medical Center Research Institute Scarborough, Maine Inez Schoenmakers Nutrition and Bone Health Research Group MRC Human Nutrition Research Elsie Widdowson Laboratory Cambridge, United Kingdom Mark F. Seifert Department of Anatomy and Cell Biology School of Medicine Indiana University Indianapolis, Indiana Sue A. Shapses Department of Nutritional Sciences Rutgers University New Brunswick, New Jersey Bonny L. Specker E.A. Martin Program in Human Nutrition South Dakota State University Brookings, South Dakota Ramu Sudhagoni E.A. Martin Program in Human Nutrition South Dakota State University Brookings, South Dakota Anna K. Surdykowski Allied Health Sciences University of Connecticut Storrs, Connecticut Natalie W. Thiex E.A. Martin Program in Human Nutrition South Dakota State University Brookings, South Dakota Frances A. Tylavsky Department of Preventive Medicine University of Tennessee Health Sciences Center Memphis, Tennessee

xvi

Sumithra K. Urs Maine Medical Center Research Institute Scarborough, Maine Cees Vermeer Maastricht University and BioPartner Center Maastricht Maastricht, the Netherlands Kate A. Ward Nutrition and Bone Health Research Group MRC Human Nutrition Research Elsie Widdowson Laboratory Cambridge, United Kingdom Stuart J. Warden Department of Physical Therapy School of Health and Rehabilitation Sciences Indiana University Indianapolis, Indiana

Contributors

John D. Wark Department of Medicine University of Melbourne and Bone & Mineral Service Royal Melbourne Hospital Parkville, Victoria, Australia Bruce A. Watkins Department of Nutritional Sciences University of Connecticut Storrs, Connecticut Susan Joyce Whiting College of Pharmacy and Nutrition University of Saskatchewan Saskatoon, Saskatchewan, Canada Kenlyn R. Young Department of Nutrition Meredith College Raleigh, North Carolina

Part I Introduction to Diet and Bone

1

Overview of Relationships between Diet and Bone John J.B. Anderson

CONTENTS Introduction......................................................................................................................................... 3 Nutrients Required for Bone Growth and Bone Maintenance............................................................ 4 Energy .......................................................................................................................................5 Protein .......................................................................................................................................5 Calcium......................................................................................................................................5 Phosphorus.................................................................................................................................6 Magnesium................................................................................................................................ 6 Vitamin D...................................................................................................................................6 Vitamin K...................................................................................................................................7 Vitamin A................................................................................................................................... 7 Vitamin C................................................................................................................................... 7 Antioxidant Nutrients................................................................................................................7 Fluoride......................................................................................................................................8 Phytomolecules.......................................................................................................................... 8 Summary....................................................................................................................................8 Calcium and Phosphorus Interrelationships........................................................................................8 Calcium and Vitamin D Interrelationships.......................................................................................... 9 Nutrition and Bone Health across the Life Cycle...............................................................................9 Nutrition and Skeletal Growth from Birth through Adolescence..............................................9 Nutrition and Bone Changes after Skeletal Growth (Length) Has Ceased.............................. 10 Nutrition and Skeletal Losses during Late Life....................................................................... 11 Vegetarian Diets................................................................................................................................ 11 Role of Physical Activity in Bone Development and Maintenance.................................................. 12 Summary........................................................................................................................................... 12 References......................................................................................................................................... 13

INTRODUCTION The consumption of adequate amounts of food ingredients, that is, nutrients and phytochemicals, is critical to bone and general health. The typical patterns of eating and living influence skeletal development and the maintenance of bone tissue throughout life. Because bone health depends on a few nutrients not so easily obtained in sufficient quantity from meals over short time spans, the overall dietary pattern of food consumption takes on great importance in making available sufficient amounts of the critical bone-building nutrients. Throughout the millennia, cultures have obtained these essential nutrients without knowledge of nutritional science. In recent years, technologically advanced nations have adopted eating patterns in which traditional foods have been replaced, in part, by overly processed foods. These convenient but less nutritious foods have become common in 3

4

Diet, Nutrients, and Bone Health

Western societies. Disappearing, however, is knowledge about healthier eating behaviors practiced by our ancestors. Information about the relationships between nutrients, including eating patterns, and bone development and maintenance from birth to late life is reviewed in this chapter. Emphasis is placed on recent research that has enhanced our understandings of diet–bone relationships. Calcium and phosphorus, two critical nutrients needed for the mineralization of the organic matrix of bone, receive greater emphasis than do other nutrients required for bone health. Coverage of osteoporosis and vitamin D deficiency diseases are limited mainly to their nutritional determinants rather than to other risk factors. This introductory chapter briefly highlights the contributions of dietary patterns and specific nutrients that have significant effects on bone health. Short sections on the roles of individual nutrients are covered in the early part of the chapter, and then integrative aspects of the diet are emphasized later in the chapter. This book provides updates on similar topics covered in our earlier publication (Anderson and Garner, 1996).

NUTRIENTS REQUIRED FOR BONE GROWTH AND BONE MAINTENANCE Most U.S. citizens are omnivorous in their eating habits, but a small percentage (1% to 5%) of the population is vegetarian in one form or another. A large percentage of the North American population fails to meet the currently recommended guidelines for optimal nutrient intakes; some nutrients are consumed in excess; others, in insufficient or even severely deficient quantities. Of particular concern is that intakes of calcium and vitamin D are lower than recommended by current guidelines (Ervin et al., 2004). Corrections of common deficits, such as of calcium and vitamin D, are well recognized as major adjustments needed by deficient adult and elderly individuals to help them maintain their bone mass. Improvement of deficient intakes of these nutrients in children and adolescents is clearly needed to support skeletal growth and to achieve optimal peak bone mass (PBM) by the end of the growth years. Other nutrient deficits, however, may be equally important (Ilich and Kerstetter, 2000; Nieves, 2005). For example, vitamin K and magnesium are also essential for bone health, and intakes of these by Americans are generally insufficient (less than 70% of Dietary Reference Intake or [DRI]) or even deficient (50% or less). Beyond nutritional deficits, excesses in dietary intakes of total calories (energy), protein, sodium, and phosphorus may have adverse effects on the bones of both children and adults. Nutritional factors affecting the bones health, both positively and negatively, are listed in Table1.1. Each of these dietary factors and a few other nutrients and phytochemicals are briefly noted in this section as preparation for subsequent chapters. TABLE 1.1 Nutritional Factors Affecting Bone Health: Beneficial and Adverse Effects Nutrient Calcium Phosphorus Vitamin D Animal protein Vitamin K Sodium Magnesium Vitamin A

Beneficial Intakes

Adverse When Intakes Are

RDA amounts RDA amounts RDA amounts ~15% of total intake >RDA amounts 2400 mg (recom. amt.) RDA amounts RDA amounts

Too little or too mucha Too much Too little or too muchb Too much animal (not plant) Too little Too much Too little Too little or too muchb

Notes:  RDA = recommended dietary allowance. a b

Excessive intakes of calcium may contribute to arterial calcification. Excessive intakes of vitamin D and vitamin A from supplements may be toxic.

5

Overview of Relationships between Diet and Bone

Energy Energy intake and bone mass are positively associated in all life stages. During childhood, energy intakes must be sufficient to support skeletal formation and growth (Ilich et al., 2003). In children, excessive weight gain, that is, overweight or obesity, typically does not favor optimal skeletal growth and has been reported to lead to an increase in fractures, especially of the wrist (Goulding et al., 2000). In adults, however, weight gain from excessive energy intake exerts a positive influence on bone mass, and, conversely, extreme weight loss and/or undernutrition may increase the risk of osteoporosis. Weight loss due to anorexia nervosa or other eating disorders is usually accompanied by bone loss. In anorexic women, bone loss results from the associated estrogen deficiency, as well as suboptimal energy intake. Loss of normal body fat may lead to a reduction in ovarian estrogen production amenorrhea and bone loss (Gordon et al., 2002).

Protein Bone formation during growth in early life requires sufficient protein intake to form the organic matrix of bone. Both high- and low-protein intakes are detrimental to bone. When a low-protein intake by adults is supplemented with protein (meat) in the face of a low-calcium intake, intestinal calcium absorption may be increased and presumably bone mass may be maintained rather than decreased (Kerstetter et al., 2005; Hunt et al., 2009). On the other hand, in growing children, high dietary intakes of protein, particularly animal protein containing acidic amino acids, may increase urinary calcium losses, leading to suboptimal skeletal growth and bone mass (Zhang et al., 2010). Optimal protein intakes, including that from animal sources, support healthy bone development and maintenance later in life (Promislow et al., 2002b; Ilich et al., 2003). Sufficient protein intake may reduce hip fractures of the elderly (Misra et al., 2009) (see Chapter 15).

Calcium The most calcium-rich food sources of calcium exist as low-fat dairy products. Typical intakes of calcium in the United States are less than the recommended amounts by current guidelines starting during early adolescence. Comparisons of dietary intakes and recommended amounts of calcium for U.S. men and women are listed in Table 1.2. A plant-based diet may be capable of supplying sufficient TABLE 1.2 Comparisons of Dietary Intakes and Recommended Amounts of Calcium and Phosphorus of U.S. Men and Women Dietary Intakes Age, Years 14–18 19–50 >51

Recommendations

Calcium

Phosphorus

Calcium

Phosphorus

787 ~750 619

1172 ~1245 1055

1300 1000 1200

1250   700   700

Source: Ervin, R.B., Wang, C.-Y., Wright, J.D., and Kennedy-Steenson, J. Dietary intake of selected minerals for the United States population: 1999–2000 (advance data no. 341). CDC, Vital and Health Statistics, US DHHS, Hyattsville, MD; and Food and Nutrition Board, Institute of Medicine. 1997. Dietary Reference Intakes for Calcium, Phosphorus, Magnesium, Vitamin D, and Fluoride. National Academy Press, Washington, DC. Notes: ~ means imputed data.       See also Chapter 8.

6

Diet, Nutrients, and Bone Health

calcium if appropriate foods are selected (Weaver et al., 1999). Calcium from dairy foods may be the best sources for skeletal development (Matkovic et al., 2004; Huncharek et al., 2008). Consumption of calcium-fortified foods taken as part of the regular diet may also effectively optimize or maintain skeletal health (Heaney, 2007). Calcium supplements may increase intake in those who cannot meet their needs by ingesting conventional or calcium-fortified foods alone, but such supplementation in premenarcheal females was effective for only 12 to 18 months (Cameron et al., 2004). These gains, however, may not be maintained after the calcium supplements are stopped, and bone losses may quickly offset the earlier gains, at least in growing children (Johnston et al., 1992). Supplementation alone may not be effective in promoting a gain in bone mass if the individual is already consuming an adequate amount of calcium, that is, near or above recommended intakes; in fact, bone loss may still occur in calcium-supplemented women but at a lower rate than in a comparable group of women on a placebo (Riis et al., 1987). A unified public health strategy is needed to ensure optimal calcium intakes from foods (both natural foods and fortified foods) and, if necessary, supplements in the North American population, while avoiding the excessive amounts that approach the tolerable upper intake levels for calcium (2500 mg/day) and raise the risk of soft tissue and vascular calcification by adults (Food and Nutrition Board, Institute of Medicine, 1997) (see below and Chapters 8 and 34).

Phosphorus Phosphorus, along with calcium, is present in the skeleton in large amounts as part of bone mineral. Therefore, dietary phosphate is required to support the growth and maintenance of the skeleton. However, the average Western diet rich in processed foods contains greater quantities of this essential element than in previous decades. Concern has been raised about the possible adverse effects of excessive intake rather than those of deficiency of this nutrient (Calvo and Park, 1996). Table 1.2 compares U.S. data for phosphorus intakes with recommended amounts. A few studies have shown that excessive phosphate intake from foods exerts adverse effects on the skeleton (Calvo et al., 1990; Kemi et al., 2008, 2009). In addition, because of the differing absorption efficiencies (70% for phosphorus and 30% for calcium) during adulthood, it is advisable to aim for an intake ratio of about 1:1 (see Chapters 8 and 9 for discussions of the calcium:phosphorus ratio).

Magnesium Magnesium is another mineral nutrient that has been shown in animal studies to be required for normal bone development and maintenance. Magnesium is not an integral part of bone mineral crystals composed of calcium and phosphorus, that is, hydroxyapatite. Approximately two-thirds of the total 25 g of magnesium in the average human body is bound to bone crystal surfaces. Magnesium supplementation has shown little or no effect in increasing bone mass (Rude et al., 2009). Low magnesium intakes are common among men and women of all ages in North America, primarily because of insufficient consumption of dark green vegetables (see Chapter 14).

Vitamin D Dietary vitamin D is converted into one of the essential hormones, 1,25-dihydroxyvitamin D, that, along with parathyroid hormone (PTH), regulates calcium metabolism. The importance of vitamin D in bone metabolism is evident in the diseases of vitamin D deficiency, that is, rickets in children and osteomalacia in adults (Kreiter et al., 2000; Holick, 2007; Stoffman and Gordon, 2009). Rickets results in deformities of bones. Osteomalacia in adults typically leads to osteopenia, fractures, poor fracture healing, and muscle weakness. Many elderly have serum vitamin D concentrations that are below optimal levels because of a decreased consumption of food sources and too little exposure to sunlight, which permits skin production of vitamin D. The cutaneous production of vitamin D decreases with age, as well as a consequence of the use of sun-blocking clothing and

Overview of Relationships between Diet and Bone

7

creams. Sun exposure is particularly low in northern latitudes during the winter months. Skin production of vitamin D in response to UVB light exposure has, in the past, been the major source of this molecule during the months of UVB penetration of the atmosphere, that is, late spring, summer, and early autumn in the northern hemisphere, whereas at present, dietary sources have taken on more prominence. Today, a large majority of North Americans fail to consume adequate amounts of this essential fat-soluble vitamin (Tylavsky et al., 2005; Holick, 2008) (see Chapter 10).

Vitamin K Vitamin K, a micronutrient, is now considered to be protective against the loss of bone mass and osteoporosis because of its role in the maintenance of the organic matrix of bone, through modification of specific amino acids in matrix proteins. Studies in postmenopausal women have shown that vitamin K may have modest beneficial effects on bone turnover and calcium metabolism (Braam etal., 2004). The actions of vitamin K may be responsible for other beneficial effects on the skeleton and calcium metabolism (Booth et al., 2000). At least one other study, however, does not support a benefit of vitamin K on bone mineral density (BMD) or fracture prevention (Binkley et al., 2009) (see Chapter 12).

Vitamin A Consumption of vitamin A within recommended levels is considered to be beneficial to bone health, but negative effects can result from either too little or too much vitamin A. Both types of bone cells, osteoblasts and osteoclasts, contain receptors for retinoic acid, which is derived from vitamin A. High vitamin A (retinol) intake may contribute to excessive resorption of bone resulting in bone loss (Promislow et al., 2002a) and even to hip fractures (Feskanich et al., 2002). Consumption of β-carotene, other carotenoids, and lycopene from fruits and vegetables, however, has been shown to have positive effects on bone health that reduce the risk of hip fracture (Sahni et al., 2009a), perhaps through their antioxidant roles (see below) rather than their conversion to retinol (see Chapter 11).

Vitamin C Like vitamin K, vitamin C or ascorbic acid plays a major role in maintaining bone health through its effect on modification of bone proteins. Bone matrix, composed of collagen, the major organic component of bone, serves to organize the three-dimensional lamellar structure of bone and is the major determinant of bone strength. Collagen molecules are cross-linked, which increases their strength and thus supports bone strength. Optimal amounts of vitamin C intake at reasonable levels have been shown to reduce the risk of hip fracture (Sahni et al., 2009b). In vitamin C deficiency, the organic structure of bone, that is, cross-linking, may be weakened because of suboptimal crosslinking of bone collagen. Fortunately, deficiency of vitamin C in the United States is rare because of fortification of many foods with vitamin C. Also, vitamin C has an important role as an antioxidant in bone cells, as well as in other cell types of the body’s tissues (see below) (see Chapter 14).

Antioxidant Nutrients Several antioxidant nutrients naturally occurring in foods act to lower the activity of free radicals in cells and, therefore, help to prolong the lives of the cells, including bone cells. Vitamin C has already been mentioned, but vitamin E, carotenoids and lycopene, selenium, and perhaps another trace element or two have important roles in diminishing the oxidative effects of free radicals— highly reactive oxygen species (Basu et al., 2001; Maggio et al., 2003). Finally, many phytochemicals from a wide variety of plant foods, especially polyphenolic molecules, have similar roles (see Chapter 19).

8

Diet, Nutrients, and Bone Health

Fluoride Fluoridation of drinking water has been very effective in reducing dental caries, but studies on the effects of this nutrient in increasing bone mineral content (BMC) and reducing bone fractures have been disappointing. Studies of fluoride have shown that protection is not obtained from the cariostatic levels found in municipal water supplies, that is, 1 ppm. Fluoride supplements at sufficiently high doses (>3 ppm) may even be deleterious to bone because of defects in the mineral phase of bone, causing it to be well mineralized but brittle, that is, hysteresis. At present, supplemental use of fluoride is not recommended for osteoporotic women or others because of the risk of adverse bone effects associated with higher exposures. The Food and Drug Administration is not likely to ever approve of fluoride use in amounts greater than 1 ppm for the intended benefit of bone mass because the quality of bone formed at higher intakes is structurally inferior, and fracture risk may actually be increased (Kleerekoper, 1996) (see Chapter 14).

Phytomolecules Phytoestrogens, including soy isoflavones, have been proposed to have osteoprotective actions because of their activation of estrogen receptors in bone cells, especially osteoblasts. These estrogen-like molecules, classified chemically as polyphenols, were reported to have positive effects on bone measurements in postmenopausal women and women with low circulating estrogens compared with those in placebo-treated women in some studies, but more recent studies have failed to support these findings. Isoflavones from soy products may have weak skeletal benefits in postmenopausal women (Arjmandi et al., 2003), but a few recent reports do not support such benefits (Alekel et al., 2010; Wong et al., 2009). In young adult women with normal estrogen status, isoflavones appear to have no effect on bone (Anderson et al., 2002).

Summary Many nutrients are required for optimal skeletal growth and maturation. The Dietary Reference Intakes of the nutrients, but not of phytomolecules, have been published by the National Academies Press (Food and Nutrition Board, Institute of Medicine, 1997, 2004). Only a few specific recommendations are given in this section, because typically, the quantities are readily available (see also Table 1.2).

CALCIUM AND PHOSPHORUS INTERRELATIONSHIPS Dietary calcium deficiency or insufficiency among adults is common, but phosphorus deficiency is extremely rare, because practically all natural foods contain phosphorus (see section “Phosphorus” above). Suboptimal phosphorus consumption from foods remains highly unlikely if adequate energy is consumed. Low phosphorus intake, however, may exist in a small percentage (<5%) of elderly living in poverty or near poverty and not consuming enough of foods containing adequate amounts of phosphorus. A very low phosphorus intake may result in hypophosphatemic osteomalacia, which decreases both bone density and bone strength. Supplementation with phosphate salts is not recommended for healthy individuals because high quantities of this mineral are already consumed in diets rich in processed foods in the United States. Phosphate excess—and calcium deficiency together—tends to increase PTH secretion and thereby decrease skeletal mass and density, partly because phosphate ions are absorbed much more rapidly than calcium ions (Anderson and Talmage, 1973). If chronic, the continuous hyperparathyroidism may lead to osteopenia and osteoporosis (see Chapters 8 and9 on calcium to phosphorus ratio). Phosphate toxicity resulting from excessive dietary phosphorus is extremely rare in individuals with normal kidney function. The recent decline in the dietary calcium-to-phosphorus ratio over

Overview of Relationships between Diet and Bone

9

the past several decades to less than 1:1, even as low as 0.5:1, has resulted from a declining consumption of dietary calcium associated with less milk intake and an increase in dietary phosphorus from cola-type beverages and foods processed with phosphate additives.

CALCIUM AND VITAMIN D INTERRELATIONSHIPS A deficient dietary calcium status has a large impact on the renal production of the hormonal form of vitamin D as well as on the increased serum concentration of PTH. Both of these hormonal adaptations are designed to enhance calcium absorption and release of calcium from bone to reestablish the normal serum concentration of calcium ions. If both dietary intakes of calcium and vitamin D are lower than recommended, too little calcium is absorbed and, thus, less goes into bone mineral (see section “Vitamin D” above). Extreme consequences of a deficiency of these two nutrients are rickets in children and osteomalacia in adults, and each of these diseases is reviewed in later chapters.

NUTRITION AND BONE HEALTH ACROSS THE LIFE CYCLE Nutrient requirements differ according to the stage of the life cycle. These needs are determined by the greater demands for growth of the skeleton during early life and the requirements to maintain the skeleton once PBM is achieved by about the age of 30 years. The early life accumulation of bone, that is,, the “early gain,” yields mean measurements of BMD of the healthy population of 20- to 29-year-old males and females that are used as the standards for comparison with individual bone measurements late in life (see World Health Organization standards for BMD in Chapter 32). The “later loss” of bone differs between men and women. Accelerated skeletal declines in bone mass and density of women occur characteristically beyond the age of menopause, that is, about aged 50 years, and these declines are associated first with osteopenia and then with osteoporosis. In men, the late-life declines start later, and they typically result in less overall loss prior to death at younger ages than that for women. The combination of the earlier period of life, that is, accumulation of bone mineral in the skeleton, and the later period as one of loss, has been aptly described as the “early gain and later loss” by an observant physical anthropologist (Garn, 1970). These changes in bone mass across the life cycle are illustrated later in this volume. The accumulation of bone mineral and the eventual loss of that mineral are intimately connected with the activity of living cells within bone. These cells are responsible for the activities of bone modeling (building predominantly), that is, bone formation and bone resorption (degradation), and remodeling (building and turning over the final model) that occur during skeletal development and throughout life. Insufficient intakes of calcium and vitamin D commonly exist among fairly high percentages of the U.S. population, and these high deficits extend across the life cycle, although they are higher in females than those in males (Food and Nutrition Board, 1997). Failure to achieve optimal PBM as a consequence of poor nutrition during the growth years cannot be reversed later in life through increased intakes of calcium and vitamin D.

Nutrition and Skeletal Growth from Birth through Adolescence Provision of an adequate diet containing all of the nutrients required for the development of healthy bones of children is obviously the responsibility of the parents or caretakers. Good nutrition during the postnatal growth years influences greatly the achievement of optimal skeletal growth, which, of course, is governed by genetic potential. Many studies throughout the world have shown that high-quality diets, in terms of protein, energy, and micronutrients, foster good growth of all tissues, including the skeleton. Thus, more affluent youth in developed Western countries and in Japan since the 1950s achieve or approach the mean heights of American children. Calcium and other minerals in the diet are retained in the skeleton, but calcium intake per se has little to do with skeletal growth acceleration. Adequate calcium intake is necessary, but not sufficient in itself, to achieve optimal

10

Diet, Nutrients, and Bone Health

bone mass. Adequate energy and high-quality protein are the critical nutrients needed for providing the stimulus for organic matrix production as part of skeletal growth from infancy through adolescence. Calcium intakes need to remain reasonably high to increase the mineral content and density of the skeleton both during growth (prepuberty and postpuberty) and in the early postgrowth period of the 20s (young adulthood). This need for calcium is especially evident for females, who consume too little calcium in the United States and most Western nations to accumulate adequate amounts of the mineral. Rapidly growing males in their late adolescent years also need an abundant amount of calcium, perhaps even more than the age-specific DRIs, for an optimal skeletal mass. The concept of a calcium requirement is complicated by the fact that, often despite somewhat poor dietary calcium intakes, American children have satisfactory skeletal growth if the intakes of energy and protein are satisfactory. At least during the adolescent period of growth, the efficiency of calcium absorption is high, perhaps 60%, to meet the skeleton’s needs for calcium in the formation of hydroxyapatite. Optimal development of PBM, however, is achieved by calcium intakes approximating the DRI (1300 mg/day) from dairy products and other calcium-containing foods throughout these early years. Premenarcheal girls (10 to 12 years old) typically have the greatest gains of bone mass and density of any stage of the life cycle because of their rapid growth, and with optimal calcium intakes, they continue gaining bone mass (Bonjour et al., 2001). Positive calcium balance is the rule during the growth phase; zero (or slightly positive) balance exists during the equilibrium phase of the decade of the 20s; and negative calcium balance dominates during the resorptive phases of the postmenopausal and elderly periods of life. A healthy state of calcium balance exists when formation equals resorption over a period of months, that is, zero balance, but after the decade of the 20s, zero balance probably rarely occurs (see Chapter 8). Skeletal growth in length (height) ceases in adolescent females within approximately 2 to 4 years following menarche (Bonjour et al., 1997; Jackman et al., 1997; Matkovic et al., 2004), whereas in males it continues into the early 20s. As stated above, however, calcium intakes continue to be important because of the accumulation of calcium in hydroxyapatite, the mineral salt in bone, during this early adult phase, sometimes called bone consolidation. Accrual of PBM is not achieved until after both growth in length and postgrowth mineral acquisition have been completed, that is, by approximately 30 years of age. Thus, adequate calcium intakes remain important for both the achievement of PBM and the maintenance of these skeletal tissues during the early adult decades. The age-specific recommended calcium allowances (DRIs) decrease from 1300 mg/day during adolescence to slightly less during adulthood (see Chapter 8). A high-calcium intake later in life, however, does not compensate for a deficient intake during childhood and adolescence.

Nutrition and Bone Changes after Skeletal Growth (Length) Has Ceased The mineral phase of the compact tissue of long-bone shafts increases its calcium and phosphorus content during this postgrowth period. For example, it has been estimated that young adult women between ages 20 and 30 years may gain an additional 5% to 10% of their skeletal mass over this decade. The widths of the long bones may continue to increase at very low rates throughout much of the remainder of life. Optimal calcium intakes then should maximize an individual’s PBM by about 30 years of age. In women, the maintenance of bone mass gained by early adulthood becomes more important with each passing decade after the menopause or equivalent age in males when the loss of bone mass continues, often at an increased rate. In females, an accelerated loss of bone occurs typically within the first decade following menopause. After growth ceases, physical activity (exercise) and pregnancy/lactation are the major factors that contribute to both higher rates of bone remodeling and increased BMC and BMD until approximately age 40 years (Tylavsky et al., 1989). After that time, both increased remodeling and bone loss typically occur until the age of 50 years, the typical age at the start of the menopausal transition that lasts for approximately a decade.

Overview of Relationships between Diet and Bone

11

Nutrition and Skeletal Losses during Late Life During the decade following menopause, resorption increases relative to formation (referred to as uncoupling) in the elderly period of life and contributes to reduced BMC and BMD. Males do not have an equivalent to menopause, and they lose bone gradually during the later decades of life. Mean estimates of calcium intakes of men and women typically are lower than the AI of 1200 mg/ day, starting at the age of 50 years. Slowing the dominance of resorption over formation in the later adult decades may be enhanced by exercise coupled with adequate calcium intake in addition to the use of bone-conserving drugs. Physicians previously prescribed hormone-replacement therapy, consisting of an estrogen and a progestin, during the early menopause, but these drugs may increase the risk of breast, endometrial, or other reproductive cancers, at least in older women. So, now, other drugs prescribed mostly for women to slow the loss of bone mass and density (secondary prevention) have been quite robust (see Chapter 5). Supplemental calcium has been used to slow bone loss by inhibiting PTH release in aged subjects (McKane et al., 1996) and to help maintain bone mass (see Chapter 27). The precise human requirements for calcium beyond the adolescent growth phase remain uncertain. During late adulthood, when sarcopenia, a decline in muscle mass, commonly accompanies osteopenia, elders have a high prevalence of both low calcium and vitamin D intakes, and supplements may improve bone measurements, that is, BMD, to a modest degree. Calcium supplementation of postmenopausal women has yielded very limited gains in BMD (see Chapters 8 and32). Too high an intake of calcium, however, may contribute to arterial calcification, especially when renal function declines (see above). The type of bone tissue that undergoes greater loss with aging is trabecular or cancellous bone, in large part because of greater surface areas. Much of the trabecular tissue is located in the vertebral bodies (spine) and the ends of the long bones (hip and wrist). Many individuals begin to lose height around the age of 50 years because of the shortening of the vertebral height (not the leg height) resulting from compression or even crushing of the vertebrae, particularly those in the lumbar region (lower spine) whose vertebral bodies are high in trabecular bone tissue. When bone mass declines below a certain low threshold in women within a decade or two after menopause and in men by their 60s and 70s, increased risk of fractures of the vertebrae and hips may occur in these individuals. In addition to dietary and activity factors, many other lifestyle variables contribute to osteoporotic fractures and subsequent survival, especially following hip fractures (see Chapter 32). Because of the high prevalence rates of hip fractures in the elderly, which result in significant mortality and cost for care, osteoporosis has become an enormous public health problem in the United States as well as in practically all other Western nations, especially as these populations are aging. In future decades, virtually all Asian, African, and Latin American countries will be similarly affected. Many reasons can be given for the increasing worldwide rates of osteoporotic hip fractures, but a major demographic factor is the increasing longevity of women and men in economically developed countries because of better nutrition, better health care, and generally improved living conditions. The aging of populations is also occurring in less developed nations, and the age-adjusted rates of osteoporotic fractures are anticipated to increase greatly in these nations.

VEGETARIAN DIETS Vegetarian diets come in different forms from strict (vegan) through lacto-, ovo-, or lacto-ovovegetarian to vegetarian diets with occasional fish. Flexitarians are those who try to eat reasonably cheaply, but on occasion, they may consume some poultry or red meat. Omnivores, of course, consume all groups of foods, and many processed food choices are made within this type of eating behavior. Emphasis in this book is based on results of bone studies of omnivorous eating patterns, although bone mass accumulation and loss appear to be similar in vegetarians.

12

Diet, Nutrients, and Bone Health

The few bone studies that have been performed with lacto-ovo-vegetarians and a recent metaanalysis confirm that vegetarians have similar bone measurements at all ages of the life cycle as omnivores—both within normal ranges of BMD during early adulthood—but that the mean measurements appear to be slightly lower for vegetarians (Ho-Pham et al., 2009), although only trivially lower (Lanham-New, 2009). The only prospective study that examined bone changes of elderly lacto-ovo-vegetarians showed that they lost bone mass over 5 years at approximately the same rate as omnivores (Reed et al., 1994).

ROLE OF PHYSICAL ACTIVITY IN BONE DEVELOPMENT AND MAINTENANCE Over the last decade or so, numerous reports have been published with findings on the beneficial effect of weight-bearing exercise on BMC and BMD. The most beneficial gains resulting from exercise occur in the early periods of life (premenarche in girls and slightly later in boys) (Welch and Weaver, 2005). Benefits of physical activities on the skeleton may occur at any age, although they are less robust after growth has ceased. The elderly show the least gains in BMC or BMD, but nevertheless, they may improve in these measurements as well as in their microarchitectural bone structures of the affected bones (see Chapters 3 and 33). What has recently become evident, however, is that the practices of upper body exercise, walking, and other activities must be continued on a regular basis or any gains of bone mass will be lost. The rates of osteoporotic fractures occurring during the postmenopausal decades have been shown to be greater in nonactive women than those among more active women. Physical activities exert forces on bone tissue that both enhance development during the bone modeling years and help maintain bone during the bone remodeling years. The benefits of activity on bone from roughly the age of 40 years and beyond have only small or no effects on BMD, but they do increase bone turnover—hence new bone—at critical sites of the skeleton and they may delay fractures at these sites (Dornemann et al., 1997). The combination of exercise plus AIs of calcium and other nutrients contribute to healthy bones.

SUMMARY A good diet containing all the nutrients at recommended intakes, plus phytochemicals from plant foods, is optimal for bone health, especially during the first two decades of life. A healthy dietary pattern coupled with regular physical activity maximizes skeletal development and PBM, as established for 20- to 29-year-old males and females. Later in life, low bone mass or osteopenia typically precedes the development of osteoporosis by approximately a decade. Osteoporosis reflects the long-term senescence of the skeletal tissues and losses of both bone mass and density, but it may occur prematurely if ovarian production of estrogens ceases because of premature menopause or oophorectomy. Osteoporosis is a multifactorial disorder with risk factors relating to nutrition; lifestyle, including cigarette smoking; hormonal status; and yet to be determined hereditary determinants. The risk of fracture is related to the degree of osteoporosis and factors related to risks of falls, including balance and agility, and to impairments in vision and cognition. This common bone disease among elders may be delayed—although not avoided—by a good diet, that is, calcium, vitamin D, and other nutrients, plus regular physical exercise, healthy lifestyle behaviors, and drugs. Adequate dietary calcium and vitamin D intakes and reduced consumption of phosphorus, particularly from processed foods and cola-type soft drinks, may help slow the loss of bone mass in the later decades of life by suppressing PTH secretion. Elders who consume little or no dairy products obtain most of their calcium from breads and baked goods, but not enough calcium exists in these foods to meet the recommended intakes for calcium. Problem nutrients for bone health continue to be too little consumption of calcium and vitamin D and of too much phosphorus (as phosphates) and sodium at all ages of the life cycle.

Overview of Relationships between Diet and Bone

13

REFERENCES Alekel, D.L., Van Loan, M.D., Koehler, K.J., et al. 2010. The Soy Isoflavones for Reducing Bone Loss (SIRBL) study: A 3-year randomized controlled trial in postmenopausal women. Am J Clin Nutr 91: 218–230. Anderson, J.J.B., Chen, X.W., Boass, A., et al. 2002. Soy isoflavones: No effects on bone mineral content and bone mineral density in healthy, menstruating young adult women after one year. J Am Coll Nutr 21: 388–393. Anderson, J.J.B., and Garner, S.C., eds. 1996. Calcium and Phosphorus in Health and Disease. CRC Press, Boca Raton, FL. Anderson, J.J.B., and Talmage, R.V. 1973. The Effect of calcium infusion and calcitonin on plasma phosphate in sham-operated and thyroparathyroidectomized dogs. Endocrinology 93: 1222–1226. Arjmandi, B.H., Khalil, D.A., Smith, B.J., et al. 2003. Soy protein has a greater effect on bone in postmenopausal women. J Clin Endocrinol Metab 88: 1048–1054. Basu, S., Michaelsson, K., Olofsson, H., et al. 2001. Association between oxidative stress and bone mineral density. Biochem Biophys Res Commun 288: 275–279. Binkley, N., Harke, J., Krueger, D., et al. 2009. Vitamin K treatment reduces undercarboxylated osteocalcin but does not alter bone turnover, density, or geometry in healthy postmenopausal North American women. JBone Miner Res 24: 983–991. Bolland, M.J., Barber, P.A., Doughty, R.N., et al. 2008. Vascular events in healthy older women receiving calcium supplementation: Randomised controlled trial. BMJ 336: 262–265. Bonjour, J.-P., Chevally, T., Ammann, P., et al. 2001. Gain in bone mineral mass in prepubertal girls 3–5 years after discontinuation of calcium supplementation: A follow-up study. Lancet 358: 1208–1212. Bonjour, J.-P., Carrie, A.L., Ferrari, S., et al. 1997. Calcium-enriched foods and bone mass growth in prepubertal girls: A randomized double blind placebo-controlled trial. J Clin Invest 99: 1287–1294. Booth, S.I., Tucker, K.L., Chen, H., et al. 2000. Dietary vitamin K intakes are associated with hip fractures but not with BMD in elderly men and women. Am J Clin Nutr 71: 1201–1208. Braam, L.A., Knapen, M.H., Geusens, P., et al. 2004. Vitamin K1 supplementation retards bone loss in postmenopausal women between 50 and 60 years of age. Calcif Tiss Int 73: 21–26. Calvo, M.S., Kumar, R., and Heath, H.H., III. 1990. Persistently elevated parathyroid hormone secretion and action in young women after four weeks of ingesting high phosphorus, low calcium diets. J Clin Endocrinol Metab 70: 1334–1340. Calvo, M.S., and Park, Y.K. 1996. Changing phosphorus content of the U.S. diet: Potential for adverse effects on bone. J Nutr 126: 1168S. Cameron, M.A., Paton, L.M., Nowson, C.A., et al. 2004. The effect of calcium supplementation on bone density in premenarcheal females: A co-twin approach. J Clin Endocrinol Metab 89: 4916–4922. Demer, L. A skeleton in the atherosclerotic closet. 1995. Circulation 92: 2029–2032. Dornemann, T.M., McMurray, R.G., Renner, J.B., et al. 1997. Effects of high-intensity resistance exercise on bone mineral density and muscle strength of 40–50 year old women. J Sports Med Phys Fitness 37: 246–251. Ervin, R.B., Wang, C.-Y., Wright, J.D., et al. 2004. Dietary intake of selected minerals for the United States population: 1999–2000 (advance data no. 341). CDC, Vital and Health Statistics, US DHHS, Hyattsville, MD, 8 pages. Feskanich, D., Singh, V., Willett, W., et al. 2002. Vitamin A intake and hip fractures among postmenopausal women. JAMA 287: 47–54. Food and Nutrition Board, Institute of Medicine. 1997. Dietary Reference Intakes for Calcium, Phosphorus, Magnesium, Vitamin D, and Fluoride. National Academy Press, Washington, DC. Food and Nutrition Board, Institute of Medicine. 2004. Dietary Reference Intakes for Electrolytes and Water. National Academy Press, Washington, DC. Garn, S. 1970. The Early Gain and Late Loss of Cortical Bone: In Nutritional Perspective. Thomas, C.C., Springfield, IL. Gordon, C.M., Goodman, E., Emans, S.J., et al. 2002. Physiologic regulators of bone turnover in young women with anorexia nervosa. J Pediatr 141: 64–70. Goulding, A., Taylor, I.E., Jones, F.A., et al. 2000. Overweight and obese children have low bone mass and area for their weight. Int J Obesity 24: 627–632. Heaney, R.P. 2007. Bone health. Am J Clin Nutr 85: 300S–303S. Holick, M.F. 2007. Vitamin D deficiency. New Engl J Med 357: 266–281. Holick, M.F. 2008. Vitamin D: A D-lightful health perspective. Nutr Rev 66: S182–S194.

14

Diet, Nutrients, and Bone Health

Ho-Pham, L.T., Nguyen, N.D., and Nguyen, T.V. 2009. Effect of vegetarian diets on bone mineral density: A Bayesian meta-analysis. Am J Clin Nutr 90: 943–950. Huncharek, M., Muscat, J., and Kupelnick, B. 2008. Impact of dairy products and dietary calcium on bonemineral content in children: Results of a meta-analysis. Bone 40: 312–321. Hunt, J.R., Johnson, L.K., and Roughead, Z.K.F. 2009. Dietary protein and calcium interact to influence calcium retention: A controlled feeding trial. Am J Clin Nutr 89: 1357–1365. Ilich, J.Z., Brownbill, R.A., and Tanborini, L. 2003. Bone and nutrition in elderly women: Protein, energy, and calcium as main determinants of bone mineral density. Eur J Clin Nutr 57: 554–565. Ilich, J.Z., and Kerstetter, J.E. 2000. Nutrition in bone health revisited: A story beyond calcium. J Am Coll Nutr 19: 715–737. Jackman, L.A., Millane, S.S., Martin, B.R., et al. 1997. Calcium retention in relation to calcium intake and postmenarcheal age in adolescent females. Am J Clin Nutr 66: 327–333. Johnston, C.C., Jr., Miller, J.Z., Slemenda, C.W., et al. 1992. Calcium supplementation and increases in bone mineral density of children. New Engl J Med 327: 82–87. Kemi, V.E., Karkkainen, M.U.M., Karp, H.J., et al. 2008. Increased calcium intake does not completely counteract the effects of increased phosphorus intake on bone: An acute dose-response study in healthy females. Br J Nutr 99: 832–839. Kemi, V.E., Rita, H.J., Karkkainen, M.U.M., et al. 2009. Habitual high phosphorus intakes and foods with phosphate additives negatively affect serum parathyroid concentration: A cross-sectional study on healthy premenopausal women. Public Health Nutr 12: 1885–1892. Kerstetter, J.E., O’Brien, K.O., Caseria, D.M., et al. 2005. The impact of dietary protein on calcium absorption and kinetic measures of bone turnover in women. J Clin Endocrinol Metab 90: 26–31. Kleerekoper, M. 1996. Fluoride: The verdict is in, but the controversy lingers. J Bone Miner Res 11: 565–567. Kreiter, S.R., Schwartz, R.P., Kirkman, H.N., Jr., et al. 2000. Nutritional rickets in African American breast-fed infants. J Pediatr 137: 153–157. Lanham-New, S.A. 2009. Is “vegetarianism” a serious risk factor for osteoporotic fracture? Am J Clin Nutr 90: 910–911. Maggio, D., Barabani, M., Pierandrei, M., et al. 2003. Marked decreases in plasma antioxidants in aged osteoporotic women: Results of a cross-sectional study. J Clin Endocrinol Metab 88: 1523–1527. Matkovic, V., Landoll, J.D., Badenhop-Stevens, N.E., et al. 2004. Nutrition influences skeletal development from childhood to adulthood: A study of hip, spine, and forearm in adolescent females. J Nutr 134: 701S–705S. McKane, W.R., Khosla, S., Egan, K.S., et al. 1996. Role of calcium intake in modulating age-related increases in parathyroid function and bone resorption. J Clin Endocrinol Metab 81: 1699–1703. Misra, D., Berry, S., Broe, K.E., et al. 2009. Does dietary protein reduce hip fracture risk in elders? The Framingham Osteoporosis Study. Osteoporos Int (in press). Nieves, J.W. Osteoporosis: The role of micronutrients. 2005. Am J Clin Nutr 81: 1232S–1239S. Promislow, J.H.E., Goodman-Gruen, D., Slymen, D.J., et al. 2002a. Retinol intake and bone mineral density in the elderly. J Bone Miner Res 17: 1359–1362. Promislow, J.H.E., Goodman-Gruen, D., Slymen, D.J., et al. 2002b. Protein consumption and bone mineral density in the elderly: The Rancho Bernardo Study. Am J Epidemiol 155: 636–644. Reed, J.A., Tylavsky, F.A., Anderson, J.J.B., et al. 1994. Camparative changes in radial bone density of elderly female lactoovovegetarians and omnivores. Am J Clin Nutr 59: 1197S–1202S. Riis, B., Thomsen, K., and Christiansen, C. 1987. Does calcium supplementation prevent postmenopausal bone loss? A double-blind, controlled clinical trial. New Engl J Med 316: 173–177. Rude, R.K., Singer, F.R., and Gruber, H.E. 2009. Skeletal and hormonal effects of magnesium deficiency. J Am Coll Nutr 28: 131–141. Sahni, S., Hannan, M.T., Blumberg, J., et al. 2009a. Protective effect of total carotenoid and lycopene intake on the risk of hip fracture: A 17-year follow-up from the Framingham Osteoporosis Study. J Bone Miner Res 24: 1086–1094. Sahni, S., Hannan, M.T., Gagnon, D., et al. 2009b. Protective effect of total and supplemental vitamin C intake on the risk of hip fracture—A 17-year follow-up from the Framingham Osteoporosis Study. Osteoporos Int 20: 1853–1861. Stoffman, N., and Gordon, C.M. 2009. Vitamin D and adolescents: What do we know? Curr Opin Pediatr 21: 465–471. Tylavsky, F.A., Bortz, A.D., Hanco*ck, R.L., et al. 1989. Familial resemblance of radial bone mass between premenopausal mothers and their college-age daughters. Calcif Tissue Int 45: 265–272.

Overview of Relationships between Diet and Bone

15

Tylavsky, F.A., Ryder, K.A., Lyyktikainen, A., et al. 2005. Vitamin D, parathyroid hormone, and bone mass in adolescents. J Nutr 135: 2735S–2738S. Weaver, C.M., Proulx, W.R., and Heaney, R. 1999. Choices for achieving adequate dietary calcium with a vegetarian diet. Am J Clin Nutr 70 (Suppl): 543S–548S. Welch, J.M., and Weaver, C.M. 2005. Calcium and exercise affect the growing skeleton. Nutr Rev 63: 361–373. Zhang, Q., Ma, G., Greenfield, H., et al. 2010. The association between dietary protein intake and bone mass accretion in pubertal girls with low calcium intakes. Br J Nutr 103: 714–723.

2

Role of Lifestyle Factors in Bone Health John J.B. Anderson and Philip J. Klemmer

CONTENTS Introduction....................................................................................................................................... 17 Nondietary Risk Factors Contributing to Osteoporosis.................................................................... 18 Thinness and Low LBM.......................................................................................................... 18 Cigarette Smoking................................................................................................................... 19 Excessive Alcohol Consumption............................................................................................. 19 Insufficient Physical Activity................................................................................................... 19 Drug Usage—Over-the-Counter and Prescription Drugs........................................................ 19 Decline of Sensory Perceptions............................................................................................... 19 Falls�����������������������������������������������������������������������������������������������������������������������������������������19 Other Adverse Risk Factors Contributing to the Pathogenesis of Osteoporosis...............................20 Summary........................................................................................................................................... 21 References......................................................................................................................................... 21

INTRODUCTION Although dietary factors remain the central focus of this book, lifestyle variables are also important determinants of bone health. Therefore, a brief account of the relevant lifestyle factors is provided as background for the later chapters that examine these factors in greater detail. Several nondietary risk factors that may increase the likelihood of osteoporotic fractures of the skeleton have been identified. These potentially adverse risk factors promote the loss of bone mass that contribute over time to fractures (Table 2.1). Each of these lifestyle variables and environmental factors has a risk by itself, and when coupled with one or more other risk factors, the risk typically increases exponentially. A classic example of multiple risk factors was reported in 1976 by an endocrine specialist in Charleston, SC, who had observed in 36 of his elderly female patients with vertebral fractures that they all exhibited thinness, excessive cigarette smoking, and possibly too much consumption of alcohol (Daniell, 1976). So, thin older (>70) postmenopausal women who have hip fractures will have been at risk over the previous decade or longer before their bone loss becomes clinically apparent by the development of pathologic bone fractures as a result of osteoporosis. In some instances, a few lifestyle practices may have positive effects on the skeleton and hence the prevention of osteoporosis-related fractures. For example, regular physical activity, including upper body weight exercise, may promote a small improvement in bone mass or at least maintain bone at a steady-state of skeletal integrity. This latter point is especially important in preventing hip fractures because maintenance of a stable bone mineral density at this site yields major benefits. The topic of exercise is covered in other chapters. The elderly commonly have losses of visual acuity, hearing, muscular strength, and equilibrium, any or all of which may contribute to falls and fractures (see below). The loss of muscle mass, also known as sarcopenia, typically goes hand in hand with osteopenia, or bone mass loss that is not as 17

18

Diet, Nutrients, and Bone Health

TABLE 2.1 Lifestyle or Nondietary Risk Factors for Osteoporosis Cigarette smoking (any amount, but pack a day more detrimental) Excessive alcohol consumption (exceeding two drinks a day for men and one drink a day for women) Insufficient physical activity and sedentary lifestyle Drugs—over-the-counter and prescription [vitamin A, corticosteroids, anabolic steroids, phenytoin (Dilantin), proton pump inhibitors, and others] Falls for many reasons (see Table 2.3)

TABLE 2.2 World Health Organization Definitions of Osteoporosis and Osteopenia: SDs Relative to Means of 20- to 29-Year-Old Controls Classification Normal BMD Osteopenic BMD Osteoporotic BMD

Definition of T Scorea Within ± 1 SD of mean Between 1 and 2.5 SDs below mean Greater than 2.5 SDs below mean

T-Score compares the current BMD measurement of adult with 20- to 29-year-old means. BMD = bone mineral density; SD = standard deviation. a

severe as in osteoporosis. Medications may also affect these same sensory organs and contribute to falls and fractures (see below). The one fracture that should be the focus of fracture prevention is of the hip because an individual with a broken hip has significantly greater morbidity and mortality over the next 6 to 12 months. The decline in body mass to the point of thinness may be the most important of the acquired risk factors. A decline in lean body mass (LBM) because of inactivity or too little food consumption is a major determinant of bone loss and fracture. Accompanying the leanness from sarcopenia, which translates to an overall decline in muscle strength, is the associated bone loss leading to osteopenia and eventually to osteoporosis, as defined by the World Health Organization (WHO) (Table 2.2). The decline in muscle strain on the skeleton at sites of muscle insertion is typically coupled with decreased bone formation and excessive bone resorption (Ontjes and Anderson, 2009).

NONDIETARY RISK FACTORS CONTRIBUTING TO OSTEOPOROSIS Each major risk factor for osteopenia and osteoporosis is briefly reviewed in this section. Fractures resulting from osteoporosis in the elderly are associated with a high risk of mortality (Cauley, 2000).

Thinness and Low LBM Maintenance of LBM, which reflects muscle mass, helps conserve skeletal mass because of the coordinated functions of the musculoskeletal system. Whereas obesity is not recommended for healthy individuals, some fat mass contributes to body mass (and body mass index [BMI]) that places individuals in the healthy weight range (BMI of 18.5 to 24.9). When BMI is below 18.5, individuals are classified as underweight, and underweight means less LBM. Elders with low LBM are at increased risk of fractures, as well as of many other chronic diseases. Low-LBM elders who also smoke tobacco and consume excessive amounts of alcohol have greatly increased risk of hip fractures (see below).

Role of Lifestyle Factors in Bone Health

19

Cigarette Smoking Cigarette smoking adversely affects practically every organ system of the body, mainly because of oxidative stress and accelerated atherosclerosis. Cigarette smoking may also increase oxidative damage within cells that contributes to reduced cellular functions. In women, cigarette smoking decreases serum estrogen concentrations by stimulating estrogen catabolism in the liver, and it may contribute to an early menopause. Bone loss typically follows declines in serum estrogen concentrations. In older men, current smoking adversely affects fractures of the vertebrae and hips (Jutberger et al., 2010).

Excessive Alcohol Consumption Too much alcohol has direct deleterious effects on cells of the body, including bone cells. Modest alcohol consumption (two drinks a day by men or one drink a day by women), however, is not considered a risk factor, and these small amounts may actually have a slightly positive effect on bone tissue, as it does in other tissues of the body.

Insufficient Physical Activity This topic is considered in considerable detail in Chapters 5, 20, and 30. Adequate physical activity when coupled with a healthy dietary pattern contributes to optimal skeletal development early in life and to maintenance later in life. Physical activity may be the most important lifestyle variable for boosting and maintaining bone mass in adult life. The types of beneficial activities include aerobic ones, such as walking, stair climbing, and gardening, but loading exercises, such as lifting weights or other activities, may also improve muscle and bone.

Drug Usage—Over-the-Counter and Prescription Drugs Excessive or inappropriate use of drugs, whether over-the-counter or prescription drugs, may have detrimental effects on the skeletal system. Vitamin A overuse has long been known to have adverse skeletal effects (see Chapter 11). A classic example of a prescription medication used to treat seizures, Dilantin, exists, and without adequate vitamin D supplementation, individuals taking this drug have been reported to develop vitamin D deficiency and osteomalacia which may cause growth abnormalities in children and pathologic fractures in adults. Prednisone and other corticosteroid hormones have adverse effects on bone tissue when chronically used over periods of months to years. Osteoporosis is a well-established adverse effect of these agents. The drug class known as proton pump inhibitors (PPIs) also has adverse effects on calcium metabolism in chronic users. PPIs reduce acid (H+) secretion by gastric cells and the reduced amounts of acid entering the small intestine decrease the absorption of calcium in the more alkaline intestinal fluid. Increased fracture rates have been reported in older patients taking PPIs (Yang etal., 2006). Illicit recreational drugs may also have adverse effects on bone, mainly because of poor dietary intakes.

Decline of Sensory Perceptions Limited vision, hearing, or equilibrium contributes to stumbling and falls (see below), which contribute to fractures in fragile elders. Correction of these losses may not be entirely possible, but new medical techniques and therapies may help mitigate these deficits.

Falls Falling has been established as a major risk factor for skeletal fractures, particularly among the elderly (Tinetti et al., 1988). Lifestyle rather than physical risk factors has been reported to greatly

20

Diet, Nutrients, and Bone Health

contribute to falls in the elderly (Faulkner et al., 2009). New approaches to preventing falls include home safety measures, such as avoiding throw rugs and loose wire cords, installing hand-hold bars in bathrooms and hallways, improving overhead lighting, and utilizing hip pads. Because poor coordination and balance contribute to falls, any intervention, such as walking-assist devices, helps to reduce fall-related fractures. Falls may also be reduced by avoiding drugs which alter cognitive function, such as sleep aids and tranquilizers, and even excessive alcohol and antihypertensive drugs which may cause orthostatic syncope (falling). Individuals with cognitive decline, memory loss, and Alzheimer’s disease may have fewer falls with improved home lighting. Individuals with depression are much more likely to fall than those without. Low serum vitamin D status, that is, 25-hydroxyvitamin D, has been shown at least in one study to contribute to poor clinical function and balance measurements and to falls, possibly by reducing muscle strength (Shahar et al., 2009). Vitamin D supplementation may reduce the risk of falls (Bischoff-Ferrari et al., 2004, 2009; Close, 2009; Pramyothin et al., 2009). Also, an adequate serum folate concentration has been found to be associated with a reduction in falls (Shahar et al., 2009).

OTHER ADVERSE RISK FACTORS CONTRIBUTING TO THE PATHOGENESIS OF OSTEOPOROSIS Additional biological risk factors contribute adversely to osteoporosis. Some of these factors relate to declines in organ system functions, particularly kidney function, whereas others relate to general lifestyle, such as inactivity and the consumption of excessive nutrient supplements. Several of these factors are listed in Table 2.3. In essence, these physiological declines result in the loss of bone mass over time (see Chapters 27 and 34). Individual elderly osteoporotic individuals at high risk of fracture, often referred to as the frail old, need to be placed on effective antiosteoporotic drug therapy. If not, these frail old individuals have approximately a 10% increased risk of death associated with fragility fracture, as reported in one meta-analysis (Bolland et al., 2010). Supplemental dietary calcium and/or vitamin D treatment, although possibly beneficial to bone, has not been shown to have as much of an effect in reducing falls or fractures, and hence deaths, in other meta-analyses (Bischoff-Ferrari, et al., 2005, 2009; Tang et al., 2007), including one that has been withdrawn (Shea et al., 2002).

TABLE 2.3 Adverse Biological Risk Factors Contributing to the Pathogenesis of Osteoporosis Thinness with low lean body mass (<18.5 BMI for low lean body mass) Age-related functional (organ system) declines  Low estrogen status: hypogonadism, early oophorectomy, premature    menopause   Androgen deficiency   Sarcopenia   Osteopenia Low bone mass at any age (relative to age-appropriate norms) Decline of sensory functions

Role of Lifestyle Factors in Bone Health

21

SUMMARY Several major nondietary risk factors or lifestyle variables that may increase the risk of fractures are briefly highlighted. Both nondietary and dietary risk factors take on greater significance in elderly subjects, whether they have already had a vertebral or hip fracture or not. Fragility fractures in osteoporotic elderly subjects resulting because of falls are a major reason for the high morbidity and mortality among this age group. Strategies designed for the prevention of fractures need to include consideration of all of these identified risk factors and any others that may arise.

REFERENCES Bischoff-Ferrari, H.A., Dawson-Hughes, B., Staehelin, H.B., et al. 2009. Fall prevention with supplemental and active forms of vitamin D: A meta-analysis of randomized controlled trials. BMJ 339: 3692–3703. Bischoff-Ferrari, H.A., Dawson-Hughes, B., Willett, W.C., et al. 2004. Effect of vitamin D on falls: A metaanalysis. JAMA 291: 1999–2006. Bischoff-Ferrari, H.A., Willett, W.C., Wong, J.B., et al. 2005. Fracture prevention with vitamin D supplementation: A meta-analysis of randomized controlled trials. JAMA 293: 2257–2264. Bolland, M.J., Grey, A.B., Gamble, G.D., et al. 2010. Effect of osteoporosis treatment on mortality: A metaanalysis. J Clin Endocrinol Metab 95: 1174–1181. Cauley, J.A. 2000. Risk of mortality following clinical fractures. Osteoporos Int 11: 556–561. Close, J.C.T. 2009. Falls in older people: Risk factors, assessment and intervention. IBMS BoneKEy 6: 368–384. Daniell, H.W. 1976. Osteoporosis of the slender smoker. Arch Intern Med 136: 298–304. [This classic paper was based on one endocrinologist’s clinical practice.] Faulkner, K.A., Cauley, J.A., Studenski, S.A., et al. 2009. Lifestyle predicts falls independent of physical risk factors. Osteoporos Int 20: 2025–2034. Jutberger, H., Lorentzon, M., Barrett-Connor, E., et al. 2010. Smoking predicts incident fractures in elderly men: Mr OS Sweden. J Bone Miner Res 25: 1010–1016. Ontjes, D.A., and Anderson, J.J.B. 2009. Nutritional and pharmacologic aspects of osteoporosis. In Handbook of Clinical Nutrition and Aging, Bales, C.W., and Ritchie, C.S., eds. Humana Press, Springer Science, New York. Pramyothin, P., Techasurungkul, S., Lin, J. et al. 2009. Vitamin D status and falls, frailty and fractures among postmenopausal Japanese women living in Hawaii. Osteoporos Int 20: 1955–1962. Shahar, D., Levi, M., Kurtz, I., et al. 2009. Nutritional status in relation to balance and falls in the elderly. Ann Nutr Metabol 54: 59–66. Shea, B., Wells, G.A., Cranney, A., et al. 2002. Meta-analysis of calcium supplementation for the prevention of postmenmopausal osteoporosis. Endocr Rev 23: 552–559. Tang, B.P.M., Eslick, G.D., Nowson, C., et al. 2007. Use of calcium or calcium in combination with vitamin D supplementation to prevent fractures and bone loss in people aged 50 years and older: A meta-analysis. Lancet 370: 657–666. Tinetti, M.E., Speechley, M., and Gintner, S.M. 1988. Risk factors for falls among elderly persons living in the community. New Engl J Med 319: 1701–1707. Yang, Y., Lewis, J., Epstein, S., et al. 2006. Long-term proton pump inhibitor therapy and risk of hip fracture. JAMA 296: 2947–2953.

3

Bone Marrow and Stem Cell Recruitment Sumithra K. Urs and Clifford J. Rosen

CONTENTS Introduction....................................................................................................................................... 23 Bone Cell Origin............................................................................................................................... 23 Molecular Regulation of Bone Formation........................................................................................24 Life Cycle Changes in Bone Marrow Cell Production..................................................................... 25 Aging��������������������������������������������������������������������������������������������������������������������������������������26 Drugs��������������������������������������������������������������������������������������������������������������������������������������� 27 Dietary and Other Influences on Bone Marrow Cell Production...................................................... 28 Summary...........................................................................................................................................30 References.........................................................................................................................................30

INTRODUCTION In the past decade, it has become clear that the skeleton is integrated with other metabolic tissues and that the skeletal microenvironment, which includes not only bone cells but also hematopoietic precursors, adipocytes, endothelial, and stem cells, is critical for regulating the process of bone remodeling. The niche that is established at various locales throughout the marrow requires the maintenance of all these cell types to preserve skeletal mass and respond to stressors that may be genetic, environmental, or pharmacologic. Dietalso plays a critical role in maintaining the health of the niche and influences the exit of stem cells from a quiescent to active state and their eventual differentiation. In this review, we discuss the origin of bone cells, their relationship to other marrow components, and the effects of diet on the proliferation and differentiation of these cells.

BONE CELL ORIGIN Bone-forming cells arise from a common progenitor cell type, the mesenchymal stem cells (MSCs), located in bone marrow. MSCs are multipotent stem cells that can differentiate into a variety of cell types including osteoblasts, chondrocytes, myocytes, and adipocyte (Pei and Tontonoz, 2004; Hong etal., 2005; Lin etal., 2008; Shockley etal., 2009). The lineage commitment of the MSCs depends on vital cues and delicate alterations in the microenvironment of the bone marrow niche in the form of signaling molecules, growth factors, and hormones that can affect their differentiation. These changes can occur as autocrine, paracrine, or endocrine signals and have a profound impact on the lineage preferences of the MSCs. Bone resorbing cells arise from hematopoietic precursors, and their differentiation is dependent on signals from MSCs that are undergoing differentiation. The major types of bone cells typically found in the bone marrow include osteoblasts, osteocytes, osteoclasts, and osteoprogenitor cells. (1) Osteoblasts, commonly called bone-forming cells, secrete osteoid, which forms the bone matrix. The process of mineralization of that matrix is also controlled by osteoblasts. Surprisingly, these cells are terminally differentiated and are unable to 23

24

Diet, Nutrients, and Bone Health

divide. (2) Osteocytes, which are mature osteoblasts, no longer secrete matrix, yet they are surrounded by it. These cells are responsible for maintaining skeletal metabolism through a series of canaliculi projecting from their cell body. They participate in nutrient/waste exchange via blood vessels and signaling of osteoblasts and resting cells. Their resting metabolic rate is low, and they contain few mitochondrial, but they are biologically active. (3) Osteoclasts, which are large, multinucleated cells, concentrate in the endosteum and function in the resorption and degradation of existing bone. (4) Osteoprogenitors are immature progenitor cells that have the capacity to divide and differentiate into osteoblasts or other mesenchymal cells.

MOLECULAR REGULATION OF BONE FORMATION Bone formation is a tightly regulated process characterized by a sequence of events starting from commitment of osteoprogenitor cells, their differentiation into preosteoblasts, and eventually maturity of osteoblasts that synthesize the collagenous bone matrix, which is progressively mineralized. These processes are tightly regulated by transcription factors and cytokines that play a pivotal role in initiating gene transcription and supporting osteoblastogenesis (Figure 3.1). Several key factors are responsible for the full differentiation of MSCs, including the following:

1. RUNX2/Cbfa1 (Runt-related transcription factor 2/core-binding factor-1) is the first transcription factor and perhaps the most important member of Runt family transcription factors expressed in mesenchymal cells at the onset of skeletal development and present throughout osteoblast differentiation; Runx2 stimulates early MSC differentiation but late in the process may downregulate osteoblast differentiation (Marie, 2008). 2. Osterix (Osx) is a zinc finger transcription factor specifically expressed by osteoblasts and an early marker of osteogenesis; it is vitally important for osteoblast maturation and acts downstream of Runx2 (Nakashima etal., 2002). 3. AP1, the activator protein 1 (AP-1), a transcription factor, is a heterodimeric protein composed of proteins belonging to the c-Fos, c-Jun, ATF, and JDP families. This family Mesenchymal cell Runx2

Msx2, Dlx3, Stat1, Sox8, Hoxa2, Sox9, MEF1, p53

Runx2

Dlx3, Dlx5, Rb, TAZ, β-catenin

Osteoprogenitor cell

Immature cell Runx2 Mature cell

Osteocyte

Runx2

FIGURE 3.1  Schematics of transcriptional regulation of osteogenesis occurring in the bone marrow.

Bone Marrow and Stem Cell Recruitment

25

regulates gene expression in response to a variety of stimuli, including cytokines, growth factors, stress, and bacterial and viral infections. AP-1 controls a number of cellular processes in the osteoblast differentiation scheme including differentiation, proliferation, and apoptosis (Hess etal., 2001). 4. ATF4/CREB2 (activating transcription factor 4) belongs to a family of DNA-binding proteins that includes the AP-1 family of transcription factors, cAMP-response element binding proteins (CREBs) and CREB-like proteins. ATF4 transcription factor is known to play a role in osteoblast differentiation along with RUNX2 and Osterix. ATF4 enhances bone formation by favoring amino acid import and collagen synthesis in osteoblasts, a function requiring its phosphorylation by RSK2 (Elefteriou etal., 2006). Further, CCAT/enhancer binding protein beta also promotes osteoblast differentiation by enhancing Runx2 activity with ATF4 (Tominaga et al., 2008). Terminal osteoblast differentiation, represented by matrix mineralization, can be significantly inhibited by the inactivation of JNK, and JNK inactivation downregulates expression of ATF4 and subsequently matrix mineralization, making ATF4 a critical factor in osteogenesis. 5. A recently recognized transcription factor is TAZ, the transcriptional coactivator with a PDZ motif acting as an important coactivator with the potential to bind to two critical transcription factors in the mesenchymal population, Runx2 and PPARγ. TAZ binding to Runx2 promotes osteogenesis, whereas binding to PPARg represses adipogenesis by suppressing PPARγ activation, thus playing an important role in balancing MSC differentiation (Hong etal., 2005; Hong and Yaffe, 2006; Rosen and MacDougald, 2006; Hong etal., 2009).

Apart from these transcription factors, β-catenin, an intracellular subunit of the cadherin protein complex, is implicated as an integral component in the Wnt signaling pathway controlling bone formation and bone mass. Interaction of Wnt proteins with Frizzled and LRP5/6 coreceptors leads to inhibition of GSK-3-mediated β-catenin phosphorylation, resulting in β-catenin accumulation and translocation to the nucleus, binding to LEF/TCF transcription factors, and activation of downstream genes (Hay etal., 2009). Inactivation of β-catenin blunts osteoblast differentiation from mesenchymal progenitors, indicating that β-catenin plays an essential role in osteoblast differentiation in vivo. Interactions between β-catenin and Runx2 are also essential for BMP9-induced osteogenic differentiation of MSCs by BMP9 (Tang etal., 2008). Other factors include the Homeobox proteins (Msx1, Msx2, Dlx5, and Dlx6), which function as transcriptional repressors during embryogenesis through interactions with components of the core transcription complex and other homeoproteins. These may also have roles in limb-pattern formation; craniofacial development, particularly odontogenesis; and tumor growth inhibition. Mutations in homeobox 7 have been associated with nonsyndromic cleft lip with or without cleft palate (Boersma etal., 1999). Peroxisome-proliferator-activated receptor gamma 2 (PPARγ2), CCAAT/enhancer-binding proteins (C/EBPs) delta, helix-loop-helix (HLH) proteins (Id and Twist), basic helix-loop-helix (bHLH), and transcription factors TRPS-1 (Piscopo etal., 2009) and Hey1 (Sharff etal., 2009) have all been implicated in cell lineage determination and differentiation and are critical components of osteogenesis in the bone marrow. These transcription factors and proteins are regulated by other factors such as parathyroid hormone (PTH), estrogen, glucocorticoids, vitamin D, bone morphogenetic protein (BMP-2, 9), and the growth factors TGF-β, IGF-1, FGF-2, and FGF-18.

LIFE CYCLE CHANGES IN BONE MARROW CELL PRODUCTION The most common factors influencing the dynamics of BM cell populations are diseases, drugs, and aging. As previously mentioned, MSCs residing in the bone marrow not only are a crucial cellular components in the microenvironment influencing the commitment and differentiation of osteogenesis, adipogenesis, and osteoclastogenesis but also are vital for supporting hematopoiesis (Guo etal., 2000). Active hematopoiesis itself is an essential requirement for maintaining skeletal homeostasis

26

Diet, Nutrients, and Bone Health

in the bone marrow niche. Clinical and experimental observations have revealed a reciprocal relationship between hematopoiesis and MSCs differentiation in the bone marrow (Shockley etal., 2007; Rodriguez etal., 2008; Shockley etal., 2009). In the normal bone marrow where active hematopoiesis takes place, the marrow appears red. In contrast, when hematopoietic tissues are damaged by irradiation or chemotherapeutic drugs, or otherwise impaired during the aging process, adipocytes begin to expand and accumulate lipid, and ultimately, the marrow takes on a yellow appearance. Lipid accumulation actively suppresses hematopoietic progenitor proliferation and osteoblast differentiation. Serum from aging women has been found to have an inhibitory effect on osteoblast differentiation but not on adipocyte differentiation (Abdallah etal., 2006). Because osteoblasts and adipocytes share an inverse relationship, especially in the bone marrow, the influence of adipocytes in the bone marrow itself becomes a critical determinant of skeletal remodeling (Rosen etal., 2009; Rosen and Klibanski, 2009). Comparison of the number and function of hematopoietic stem and progenitor cells in adipocyte-rich bone marrow with adipocyte-poor bone marrow identified a decrease in absolute numbers of hematopoietic progenitor cells. A decrease in the numbers of hematopoeitic progenitor cells relative to adipocytes in the adipocyte-rich bone marrow suppressed hematopoiesis (Naveiras etal., 2009). In addition, cell-cycle activities of hematopoietic stem and progenitor cells also vary between adipocyte-rich and adipocyte-poor bone marrow, with the hematopoietic cells staggering in the G0 phase in adipocyte-rich marrow (Sugimura and Li, 2009). Some of the factors that modulate MSCs and hematopoietic stem cells (HSCs) by initiating a proinflammatory immune response are Dlk-1/FA-1 (Delta like-1/fetal antigen-1) (Abdallah etal., 2007) and proinflammatory cytokines, such as interleukin 1 (IL-1beta) and tumor necrosis factor alpha (TNF-alpha), which also inhibit osteogenic differentiation and bone repair during inflammation (Lacey etal., 2009).

Aging Bone mass is maintained through a delicate balance between bone formation and resorption. During growth, bone formation exceeds resorption, and total bone mass increases. As adulthood progresses, the balance shifts and resorption exceeds formation. Bone is lost beginning early in adulthood and continues unabated into old age. Bone mass is maintained through a balance between osteoblast and osteoclast activity. Osteoblasts themselves regulate the number and activity of osteoclasts through expression of RANKL, osteoprotegerin (OPG), and macrophage-colony stimulation factor (M-CSF). Aging increases osteoclastogenesis and is accompanied by increased expression of RANKL and M-CSF expression, increased stromal/osteoblastic cell-induced osteoclastogenesis, and expansion of the osteoclast precursor pool. Aging results in decrease of proproliferative, antiapoptotic, and functional responses of growth factor signaling (IGF-1) with impaired receptor activation and signal transduction through MAPK (ERK½) and PI3K (AKT) pathways (Cao etal., 2005, 2007). Monocyte/macrophage population, as judged by expression of CD11b, increases with age by about 30%. The transcription factor hypoxia-inducible factor (HIF) plays an important role in maintenance of oxygen homeostasis and has an anabolic function in cartilage and bone development. Expression of HIF-1alpha is reduced with aging, thereby impairing the response to ischemia and production of angiogenic cytokines like VEGF (Wang et al., 2007; Shomento et al., 2009). Interestingly, hypoxia also induces an adipocyte-like phenotype, but not the expression of adipogenic markers (Fink etal., 2004). Aging is also associated with enhanced generation of reactive oxygen species (ROS) in the marrow which are generated in response to increased activity of enzymes like Alox12 and15. Fatty acid products, prostaglandins, and other products, such as HODE, can bind to PPARγ and enhance adipocyte differentiation (Jilka, R.L. 2002 personal communication). Growth hormone (GH) is an important physiological regulator of bone growth and remodeling. Osteoblasts and chondrocytes have receptors for GH, and addition of GH increases cell proliferation and osteoblast differentiation while suppressing bone marrow lipid accumulation. A recent study showed that GH replacement in hypophysectomized (HYPOX) rats could reverse bone loss and mineralizing perimeter, along with bone marrow adiposity and precursor pool size (Menagh etal., 2009).

Bone Marrow and Stem Cell Recruitment

27

However, PTH can also reverse the inhibitory effect of HYPOX on mineralizing perimeter but has no effect on marrow adiposity. Insulin-like growth factor (IGF-1) can accelerate bone marrow stem cell mobilization by paracrine activation of stromal-cell-derived factor SDF-1alpha/CXCR4 and contributes to MSC stimulation. Most of these molecules identified as modulators are growth factors and cytokines acting via the MAPK signaling pathway (Haider etal., 2008). Some genes like Dlk-1/Pref-(Delta-like 1/preadipocyte factor-1) and Zfp512 control adipocyte and osteoblast differentiation (Abdallah etal., 2004; Wu etal., 2009a). Dlk-1/Pref-1 acts by inhibiting mineralization, and new bone formation by inhibiting expression of known bone gene markers like Collagen 1 (Col1), alkaline phosphatase (ALP), and osteocalcin (OC), whereas zinc finger protein 512 (Zfp512) inhibits Runx2/Cbfa1 and suppresses transition of osteoprogenitor cells to immature and mature osteoblasts. Furthermore, Pref-1 has been implicated in bone marrow adiposity and low BMD in anorexia nervosa patients (Fazeli etal., 2009). Recently, negative regulators for osteoblast differentiation, such as integrin and noggin, have been recognized. Although there is a sufficient pool of osteoprogenitor cells in the bone marrow, it is possible that not all of these stem cells mature to form the osteoblasts that mediate bone formation. Furthermore, the microenvironment modulates every step of osteogenesis and adipogenesis occurring in the bone marrow milieu. While some in vitro studies have suggested that inhibition of the adipocyte differentiation pathway leads to enhanced osteoblastogenesis, this has not been supported by in vivo studies. However, early B-cell factor-1 (Ebf-1), a transcription factor expressed in MSCs and essential for B-cell fate specification and function has been reported to induce both osteogenesis and adipogenesis (Hesslein etal., 2009). Dozens of factors, including systemic, local, and inherent changes in the bone cell population, likely contribute to age-related bone loss. Whether the changes in stromal/osteoblastic cell function and osteoclast precursor number reflect inherent cell senescence or whether the extracellular milieu in the process of aging alters the metabolic activity of the maturing preosteoblast–osteoclast remains to be determined. On the other hand, bone marrow stromal cells also support osteoclast differentiation and express receptor activator of NF-kB ligand (RANKL) (Takagi and Kudo, 2008). Some bone marrow stromal cell lines, for example, TSB cell lines, support osteoclast differentiation and differentiate into osteoblasts, suggesting that osteoblast precursor cells can support osteoclast differentiation. Osteoclast inhibitory lectin (OCIL), a type II C-type lectin that binds to NK-cell-associated receptor Nkrp1d and sulfated glycosaminoglycans, is expressed by several cell types found in bone and inhibits osteoclast differentiation and osteoblast mineralization (Nakamura etal., 2007). Certain growth factor deficiencies like vitamin B12 deficiency can stimulate osteoclastogenesis, which manifests in the form of loss of bone mass during aging (Vaes etal., 2009). In sum, a balance of stimulatory and inhibitory signals determines MSC progression through a particular lineage and continued regeneration.

Drugs The use of nonsteroidal anti-inflammatory drugs (NSAIDs) is ubiquitous in contemporary medical practice as highly effective adjuncts for the amelioration of postoperative pain. NSAIDs act through inhibition of cyclooxygenase (COX-2) enzymes and therefore diminish prostaglandin production. However, eicosanoids (prostaglandins) are intimately involved in the modulation of bone metabolism and favor bone anabolism. Inhibition of cyclooxygenase-2 downregulates osteoclast and osteoblast differentiation while favoring adipocyte differentiation (Kellinsalmi et al., 2007). Insulin and/or glycemic status can also regulate osteogenesis. The consequences of diabetes and insulin treatment on bone formation and osteoblastogenesis show decreases in new bone formation. Insulin treatment can restore bone formation to levels observed in nondiabetic controls; however, it fails to exert a significant decrease in adipogenesis. RUNX2 and several RUNX2 target genes, including matrix metalloproteinase-9, ALP, integrin binding sialoprotein, Dmp1, Col1a2, Phex, VDR, osteocalcin, and osterix, are significantly downregulated during insulin deficiency, hyperglycemia, and diabetic conditions; all these contributing to diabetic bone diseases (Fowlkes etal., 2008). Rosiglitazone, a

28

Diet, Nutrients, and Bone Health

commonly used antidiabetic thiazolidinedione class of drug, has also been found to decrease bone formation and increase adipocyte and osteoclast numbers (Ali etal., 2005).

DIETARY AND OTHER INFLUENCES ON BONE MARROW CELL PRODUCTION Numerous nutrients and dietary components, ranging from macronutrients to micronutrients as well as bioactive food ingredients, influence bone health. Many of the nutrients and food components can potentially have a positive or negative impact on bone health (Table 3.1). They can affect bone turnover by various mechanisms, including alterations of bone structure, rates of bone metabolism, and responsiveness to endocrine and/or paracrine signals. These dietary factors range from inorganic minerals (e.g., calcium, magnesium, phosphorus, sodium, potassium, and various trace elements) and vitamins (vitamins A, D, E, K, C, and certain B vitamins) to macronutrients, such as protein and fatty acids. In addition, the relative proportions of these dietary factors derived from different types of diets (vegetarian vs. omnivorous) may also affect bone health and bone marrow cell production. In recent years, a number of bioactive food components have been proposed as being beneficial for bone health (Bonjour, 2005; Cashman, 2007). Below, several of those nutrients and their relationship to stromal cell differentiation are examined. Calcium is critical for normal growth and development of the skeleton and is essential for achieving optimal peak bone mass. Adequate calcium also modifies the rate of bone loss associated with aging. Over the past decade, convincing evidence has emerged with respect to the effects of dietary calcium on bone health across several age groups. Calcium in food sources generally is present as salts or associated with other dietary constituents in the form of complexes of calcium ions (Ca 2+). Calcium must be released in an ionized soluble form to be absorbed both by passive diffusion and by active transport mediated by vitamin D through transcellular and paracellular pathways (Cashman, 2007). Ionic calcium regulates chemotaxis of selective bone marrow cells through the Ca-sensing receptor CaR and CXCR4, an important receptor

TABLE 3.1 Nutritional Factors That Influence Bone Beneficial Nutrient Factors Calcium Copper Zinc Fluoride Magnesium Phosphorus Potassium Vitamin C Vitamin D

Potentially Detrimental Environmental Factors Excess alcohol Excess caffeine Excess sodium Excess fluoride Excess/insufficient proteins Excess phosphorus Excess/insufficient vitamin A Excess n-6 PUFA

Vitamin K B vitamins n-3 fatty acids Proteins Whey-derived peptides Phytoestrogens Nondigestable oligosaccharides (especially inulin-type fructans) Source: Adapted from Cashman, K.D., J Nutr 137: 2507S–2512S, 2007.

Bone Marrow and Stem Cell Recruitment

29

promoting stem cell mobilization and homing (Wu et al., 2009). Dietary zinc reduces osteoclast resorption and increases osteoblast differentiation, matrix maturation, and mineralization (Hadley etal., 2009). Among the vitamins, vitamin D is found naturally in very few foods; however, endogenous synthesis of vitamin D occurs when skin is exposed to UV radiation (290–320 nm) from sunlight. Vitamin D deficiency is characterized by inadequate mineralization, or demineralization, of the skeleton. In children, severe vitamin D deficiency results in rickets, whereas in adults, it leads to a mineralization defect in the skeleton causing osteomalacia. In addition, secondary hyperparathyroidism associated with low vitamin D status enhances mobilization of calcium from the skeleton. Vitamin D deficiency contributes to osteoporosis through less efficient intestinal absorption of calcium, increased bone loss, muscle weakness, and a weakened bone microstructure. Supplemental vitamin D can significantly reduce the risk of fracture in older people. Recently, BM-MSCs were shown to have the molecular machinery to metabolize and respond to vitamin D, wherein circulating 25OH-vitamin D signaling is amplified through IGF-1 upregulation and induced 1-alpha hydroxylase (CYP27B1). These processes also potentiate osteoblast differentiation by IGF-1 (Zhou etal., 2009). Vitamin K is a fat-soluble molecule essential for promoting the activity of Gla proteins in bone and elsewhere. Osteocalcin, a bone-specific protein, is gamma carboxylated by vitamin K. The circulating concentration of carboxylated versus uncarboxylated osteocalcin reflects the relative level of vitamin K nutritional status. In some studies, low vitamin K has been shown to be an indicator of hip fracture and a predictor of BMD. Relatively high-dose vitamin K1 supplementation (1 mg/day) for 3 years, if coadministered with calcium, magnesium, zinc, and vitamin D, reduced postmenopausal bone loss (Braam etal., 2003a, 2003b) and improved hematopoietic functions (Miyazawa and Aizawa, 2004), although no evidence from other studies supports a role for vitamin K in reducing fracture risk. Consumption of fish or n-3 fatty acids protects not only against cardiovascular and autoimmune disorders, but these long-chain n-3 fatty acids have beneficial effects on bone mineral density and bone accrual (Hogstrom etal., 2007) and they may decrease bone resorption markers (Sun etal., 2003; Griel etal., 2007). Estrogen deficiency is a major contributory factor to the development of osteoporosis in women, and estrogen and/or hormone replacement therapy (HRT), bisphosphonates, calcitonin, and raloxifene, as well as calcium and vitamin D supplementation, are the mainstays for prevention of bone loss in postmenopausal women. Recently, as a consequence of poor uptake and adherence of HRT as well as potential concerns over an increased risk of malignancy and other side effects associated with the use of HRT, attention has focused on the dietary phytoestrogens. Dietary phytoestrogens are nonsteroidal isoflavones naturally occurring in foods of plant origin (especially soy-based foods) that structurally resemble natural estrogens and compete with them for binding estrogen receptors. One such isoflavone is soy-derived genistein, which when supplemented (56 mg/day) significantly increased BMD in the femur and lumbar spine; these effects appeared to be of similar magnitude as those achieved with HRT in early postmenopausal women (Morabito etal., 2002). These compounds have been shown in vitro to have prodifferentiation effects on the preosteoblasts (Morris etal., 2006). However, large randomized placebo-controlled trials in postmenopausal women are just now finishing and will determine if a role for the isoflavones in the prevention of osteoporosis materializes. Factors that negatively regulate bone and marrow functions include oxidized low density lipoproteins, which have a direct inhibitory effect on differentiation of preosteoblasts and MSCs by inhibition of alkaline phosphatase activity, collagen I processing, and mineralization. These regulatory factors operate through a mitogen-activated protein (MAP) kinase-dependent pathway to direct progenitor marrow stromal cells to undergo adipogenic instead of osteogenic differentiation (Chu etal., 2009). Furthermore, along the same lines, a low carbohydrate–high fat (LC–HF) diet has been found to impair longitudinal growth, BMD, and mechanical properties, possibly mediated by reductions in circulating IGF-1 (Bielohuby etal., 2009). This finding was surprising and has lowered enthusiasm among advocates of an LC–HF diet, particularly because

30

Diet, Nutrients, and Bone Health

serum bone formation markers as well as expression of transcription factors influencing osteoblastogenesis are reduced with this diet. On the other hand, a high-fat diet can also decrease cancellous bone mass in certain circ*mstances (Cao etal., 2009). A balanced diet combined with adequate exercise is the best recommendation for a proper functioning of the body and maintaining a balance in the bone marrow stromal cells for cell recruitment and balance between resorption and formation.

SUMMARY Although knowledge about the bone marrow niche has grown dramatically in the last decade, many unanswered questions remain. The precise relationship between adipogenesis and osteogenesis remains to be clarified; similarly bone cell development is essential for hematopoiesis, but the factors that coordinate that response are not fully defined. The effects of dietary components on marrow cellular processes are important, yet the model systems are not perfect, and the randomized trials for nutrients other than calcium are lacking. Even the actions of vitamin D are controversial, and the effects of this hormone/vitamin on marrow elements need further clarification. Significant research imperatives relative to dietary factors and the bone marrow niche need further examination.

REFERENCES Abdallah, B.M., Boissy, P., Tan, Q., etal. 2007. Dlk1/FA1 regulates the function of human bone marrow mesenchymal stem cells by modulating gene expression of pro-inflammatory cytokines and immune responserelated factors. J Biol Chem 282: 7339–51. Abdallah, B.M., Haack-Sorensen, M., Fink, T., etal. 2006. Inhibition of osteoblast differentiation but not adipocyte differentiation of mesenchymal stem cells by sera obtained from aged females. Bone 39: 181–8. Abdallah, B.M., Jensen, C.H., Gutierrez, G., etal. 2004. Regulation of human skeletal stem cells differentiation by Dlk1/Pref-1. J Bone Miner Res 19: 841–52. Ali, A.A., Weinstein, R.S., Stewart, S.A., etal. 2005. Rosiglitazone causes bone loss in mice by suppressing osteoblast differentiation and bone formation. Endocrinology 146: 1226–35. Bielohuby, M., Matsuura, M., Herbach, N., etal. 2009. Short Term exposure to low-carbohydrate/high fat diets induces low bone mineral density and reduces bone formation in rats. J Bone Miner Res Aug 4. Epub ahead of print. Boersma, C.J., Bloemen, M., Hendriks, J.M., et al. 1999. Homeobox proteins as signal transduction intermediates in regulation of NCAM expression by recombinant human bone morphogenetic protein-2 in osteoblast-like cells. Mol Cell Biol Res Commun 1: 117–24. Bonjour, J.P. 2005. Dietary protein: An essential nutrient for bone health. J Am Coll Nutr 24: 526S–36S. Braam, L.A., Knapen, M.H., Geusens, P., etal. 2003a. Vitamin K1 supplementation retards bone loss in postmenopausal women between 50 and 60 years of age. Calcif Tissue Int 73: 21–6. Braam, L.A., Knapen, M.H., Geusens, P., etal. 2003b. Factors affecting bone loss in female endurance athletes: A two-year follow-up study. Am J Sports Med 31: 889–95. Cao, J.J., Gregoire, B.R., and Gao, H. 2009. High-fat diet decreases cancellous bone mass but has no effect on cortical bone mass in the tibia in mice. Bone 44: 1097–104. Cao, J.J., Kurimoto, P., Boudignon, B., etal. 2007. Aging impairs IGF-I receptor activation and induces skeletal resistance to IGF-I. J Bone Miner Res 22: 1271–9. Cao, J.J., Wronski, T.J., Iwaniec, U., etal. 2005. Aging increases stromal/osteoblastic cell-induced osteoclastogenesis and alters the osteoclast precursor pool in the mouse. J Bone Miner Res 20: 1659–68. Cashman, K.D. 2007. Diet, nutrition, and bone health. J Nutr 137: 2507S–12S. Chu, L., Hao, H., Luo, M., etal., 2009. Ox-LDL modifies the behavior of bone marrow stem cells and impairs their endothelial differentiation via inhibition of Akt phosphorylation. J Cell Mol Med. Epub ahead of print. Elefteriou, F., Benson, M.D., Sowa, H., etal. 2006. ATF4 mediation of NF1 functions in osteoblast reveals a nutritional basis for congenital skeletal dysplasiae. Cell Metab 4: 441–51. Fazeli, P.K., Bredella, M.A., Misra, M., etal. 2009. Preadipocyte factor-1 is associated with marrow adiposity and bone mineral density in women with anorexia nervosa. J Clin Endocrinol Metab. Epub ahead of print.

Bone Marrow and Stem Cell Recruitment

31

Fink, T., Abildtrup, L., Fogd, K., etal. 2004. Induction of adipocyte-like phenotype in human mesenchymal stem cells by hypoxia. Stem Cells 22: 1346–55. Fowlkes, J.L., Bunn, R.C., Liu, L., etal. 2008. Runt-related transcription factor 2 (RUNX2) and RUNX2-related osteogenic genes are down-regulated throughout osteogenesis in type 1 diabetes mellitus. Endocrinology 149: 1697–704. Griel, A.E., Kris-Etherton, P.M., Hilpert, K.F., etal. 2007. An increase in dietary n-3 fatty acids decreases a marker of bone resorption in humans. Nutr J 6: 2. Guo, Z., Tang, P., Liu, X., et al. 2000. Mesenchymal stem cells derived from human bone marrow support hematopoiesis in vitro. Zhongguo Shi Yan Xue Ye Xue Za Zhi 8: 93–6. Hadley, K.B., Newman, S.M., and Hunt, J.R. 2009. Dietary zinc reduces osteoclast resorption activities and increases markers of osteoblast differentiation, matrix maturation, and mineralization in the long bones of growing rats. J Nutr Biochem. Epub ahead of print. Haider, H., Jiang, S., Idris, etal. 2008. IGF-1-overexpressing mesenchymal stem cells accelerate bone marrow stem cell mobilization via paracrine activation of SDF-1alpha/CXCR4 signaling to promote myocardial repair. Circ Res 103: 1300–8. Hay, E., Laplantine, E., Geoffroy, V., etal. 2009. N-cadherin interacts with axin and LRP5 to negatively regulate Wnt/beta-catenin signaling, osteoblast function, and bone formation. Mol Cell Biol 29: 953–64. Hess, J., Porte, D., Munz, C., et al. 2001. AP-1 and Cbfa/runt physically interact and regulate parathyroid hormone-dependent MMP13 expression in osteoblasts through a new osteoblast-specific element 2/AP-1 composite element. J Biol Chem 276: 20029–38. Hesslein, D.G., Fretz, J.A., Xi, Y., etal. 2009. Ebf1-dependent control of the osteoblast and adipocyte lineages. Bone 44: 537–46. Hogstrom, M., Nordstrom, P., and Nordstrom, A. 2007. n-3 Fatty acids are positively associated with peak bone mineral density and bone accrual in healthy men: The NO2 Study. Am J Clin Nutr 85: 803–7. Hong, D., Chen, H.X., Xue, Y., etal. 2009. Osteoblastogenic effects of dexamethasone through upregulation of TAZ expression in rat mesenchymal stem cells. J Steroid Biochem Mol Biol 116: 86–92. Hong, J.H., Hwang, E.S., Mcmanus, M.T., etal. 2005. TAZ, a transcriptional modulator of mesenchymal stem cell differentiation. Science 309: 1074–8. Hong, J.H., and Yaffe, M.B. 2006. TAZ: A beta-catenin-like molecule that regulates mesenchymal stem cell differentiation. Cell Cycle 5: 176–9. Kellinsalmi, M., Parikka, V., Risteli, J., etal. 2007. Inhibition of cyclooxygenase-2 down-regulates osteoclast and osteoblast differentiation and favours adipocyte formation in vitro. Eur J Pharmacol 572: 102–10. Lacey, D.C., Simmons, P.J., Graves, S.E., etal. 2009. Proinflammatory cytokines inhibit osteogenic differentiation from stem cells: Implications for bone repair during inflammation. Osteoarthritis Cartilage 17: 735–42. Lin, Y.F., Jing, W., Wu, L., etal. 2008. Identification of osteo-adipo progenitor cells in fat tissue. Cell Prolif 41: 803–12. Marie, P.J. 2008. Transcription factors controlling osteoblastogenesis. Arch Biochem Biophys 473: 98–105. Menagh, P., Turner, R., Jump, D., etal. 2009. Growth hormone regulates the balance between bone formation and bone marrow adiposity. J Bone Miner Res. Epub ahead of print. Miyazawa, K., and Aizawa, S. 2004. Vitamin K2 improves the hematopoietic supportive functions of bone marrow stromal cells in vitro: A possible mechanism of improvement of cytopenia for refractory anemia in response to vitamin K2 therapy. Stem Cells Dev 13: 449–51. Morabito, N., Crisafulli, A., Vergara, C., etal. 2002. Effects of genistein and hormone-replacement therapy on bone loss in early postmenopausal women: A randomized double-blind placebo-controlled study. J Bone Miner Res 17: 1904–12. Morris, C., Thorpe, J., Ambrosio, L., etal. 2006. The soybean isoflavone genistein induces differentiation of MG63 human osteosarcoma osteoblasts. J Nutr 136: 1166–70. Nakamura, A., Ly, C., Cipetic, M., etal. 2007. Osteoclast inhibitory lectin (OCIL) inhibits osteoblast differentiation and function in vitro. Bone 40: 305–15. Nakashima, K., Zhou, X., Kunkel, G., etal. 2002. The novel zinc finger-containing transcription factor osterix is required for osteoblast differentiation and bone formation. Cell 108: 17–29. Naveiras, O., Nardi, V., Wenzel, P.L., etal. 2009. Bone-marrow adipocytes as negative regulators of the haematopoietic microenvironment. Nature 460: 259–63. Pei, L., and Tontonoz, P. 2004. Fat’s loss is bone’s gain. J Clin Invest 113: 805–6. Piscopo, D.M., Johansen, E.B., and Derynck, R. 2009. Identification of the GATA factor TRPS1 as a repressor of the osteocalcin promoter. J Biol Chem 284: 31690–703. Rodriguez, J.P., Astudillo, P., Rios, S., etal. 2008. Involvement of adipogenic potential of human bone marrow mesenchymal stem cells (MSCs) in osteoporosis. Curr Stem Cell Res Ther 3: 208–18.

32

Diet, Nutrients, and Bone Health

Rosen, C.J., Ackert-Bicknell, C., Rodriguez, J.P., et al. 2009. Marrow fat and the bone microenvironment: Developmental, functional, and pathological implications. Crit Rev Eukaryot Gene Expr 19: 109–24. Rosen, C.J., and Klibanski, A., 2009. Bone, fat, and body composition: Evolving concepts in the pathogenesis of osteoporosis. Am J Med 122: 409–14. Rosen, E.D., and MacDougald, O.A. 2006. Adipocyte differentiation from the inside out. Nat Rev Mol Cell Biol 7: 885–96. Sharff, K.A., Song, W.X., Luo, X., etal. 2009. Hey1 basic helix-loop-helix protein plays an important role in mediating BMP9-induced osteogenic differentiation of mesenchymal progenitor cells. J Biol Chem 284: 649–59. Shockley, K.R., Lazarenko, O.P., Czernik, P.J., etal. 2009. PPARgamma2 nuclear receptor controls multiple regulatory pathways of osteoblast differentiation from marrow mesenchymal stem cells. J Cell Biochem 106: 232–46. Shockley, K.R., Rosen, C.J., Churchill, G.A., etal. 2007. PPARgamma2 regulates a molecular signature of marrow mesenchymal stem cells. PPAR Res 2007: 81219. Shomento, S.H., Wan, C., Cao, X., etal. 2009. Hypoxia-inducible factors 1alpha and 2alpha exert both distinct and overlapping functions in long bone development. J Cell Biochem. Epub ahead of print. Sugimura, R., and Li, L., 2009. Shifting in balance between osteogenesis and adipogenesis substantially influences hematopoiesis. J Mol Cell Biol. Epub ahead of print. Sun, D., Krishnan, A., Zaman, K., etal. 2003. Dietary n-3 fatty acids decrease osteoclastogenesis and loss of bone mass in ovariectomized mice. J Bone Miner Res 18: 1206–16. Takagi, K., and Kudo, A. 2008. Bone marrow stromal cell lines having high potential for osteoclast-supporting activity express PPARgamma1 and show high potential for differentiation into adipocytes. J Bone Miner Metab 26: 13–23. Tang, N., Song, W.X., Luo, J., etal. 2008. BMP9-induced osteogenic differentiation of mesenchymal progenitors requires functional canonical Wnt/beta-catenin signaling. J Cell Mol Med. Epub ahead of print. Tominaga, H., Maeda, S., Hayashi, M., etal. 2008. CCAAT/enhancer-binding protein beta promotes osteoblast differentiation by enhancing Runx2 activity with ATF4. Mol Biol Cell 19: 5373–86. Vaes, B.L., Lute, C., Blom, H.J., etal. 2009. Vitamin B12 deficiency stimulates osteoclastogenesis via increased hom*ocysteine and methylmalonic acid. Calcif Tissue Int 84: 413–22. Wang, Y., Wan, C., Deng, L., etal. 2007. The hypoxia-inducible factor alpha pathway couples angiogenesis to osteogenesis during skeletal development. J Clin Invest 117: 1616–26. Wu, M., Hesse, E., Morvan, F., etal. 2009a. Zfp521 antagonizes Runx2, delays osteoblast differentiation in vitro, and promotes bone formation in vivo. Bone 44: 528–36. Wu, Q., Shao, H., Darwin Eton, D., etal. 2009. Extracellular calcium increases CXCR4 expression on bone marrow-derived cells and enhances pro-angiogenesis therapy. J Cell Mol Med. Epub ahead of print. Zhou, S., Leboff, M.S., and Glowacki, J. 2009. Vitamin D metabolism and action in human bone marrow stromal cells. Endocrinology. Epub ahead of print.

4

Skeletal Tissues and Mineralization Sanford C. Garner and John J.B. Anderson

CONTENTS Introduction....................................................................................................................................... 34 Bone as an Organ.............................................................................................................................. 34 Structure................................................................................................................................... 35 Gross Anatomy........................................................................................................................ 35 Bone as Tissue.................................................................................................................................. 36 Microscopic Anatomy of Bone Cells....................................................................................... 36 Bone-Forming Cells (Osteoblasts).............................................................................. 36 Osteoblast-Derived Cells of Mature Bone................................................................... 37 Bone-Resorbing Cells (Osteoclasts)............................................................................ 37 Chemical Structure.................................................................................................................. 38 Collagen and Noncollagenous Proteins of Bone Matrix......................................................... 39 Cell-Attachment Proteins............................................................................................. 39 Proteoglycans...............................................................................................................40 γ-Carboxylated (Gla) Proteins.....................................................................................40 Growth-Related Proteins..............................................................................................40 Bone Formation................................................................................................................................ 41 Intramembranous Bone Formation.......................................................................................... 41 Endochondral Bone Formation (Bone Modeling)................................................................... 41 Mineralization.......................................................................................................................... 41 Bone Resorption................................................................................................................................ 42 Bone Remodeling.............................................................................................................................. 43 The Bone Multicellular Unit and Bone Remodeling...............................................................44 Regulation of Bone Remodeling............................................................................................. 45 Lifestyle Effects................................................................................................................................46 Peak Bone Mass Accrual.........................................................................................................46 Effect of Exercise (Strain) on Bone.........................................................................................46 Aging and Osteoporosis....................................................................................................................46 Assessment of Bone Mass and Bone Turnover.................................................................................46 Bone Histomorphometry......................................................................................................... 47 In Vivo Measurements of BMC and BMD............................................................................... 47 Dual-Energy X-Radiographic Absorptiometry............................................................ 47 Quantitative Computed Tomography........................................................................... 48 Other Methods.............................................................................................................48 Summary......................................................................................................................48

33

34

Diet, Nutrients, and Bone Health

Bone Markers in Blood and Urine........................................................................................... 49 Markers of Bone Formation......................................................................................... 49 Markers of Bone Resorption........................................................................................ 49 Summary...................................................................................................................... 50 Summary........................................................................................................................................... 50 References......................................................................................................................................... 50

INTRODUCTION The skeleton conducts a number of essential functions for maintenance of life in vertebrates, some of which involve the obvious rigid structure provided by bone and others that result from the dynamic organization of bone at the microscopic level. Bone provides both mechanical support for the body, including attachment for locomotor muscles, and protection for portions of soft tissue, such as brain, spinal cord, heart, and lungs. In addition, all blood cell formation or hematopoiesis takes place in the bone marrow. A less obvious function of bone is that it serves as a metabolic reservoir for calcium and other minerals. A constant exchange of ions of calcium (Ca), phosphate (inorganic phosphorus or Pi), and other mineral elements occurs between bone mineral surfaces and extracellular fluid in bone tissue. This constant ebb and flow of Ca and Pi is the result primarily of the active exchange of these minerals that occurs at the interface of bone mineral surfaces with the extracellular fluid and secondarily of the remodeling processes within bone that renew bone microstructure throughout life. As described by Talmage and others based on the pioneering concepts of William F. Neuman, the control of free calcium ion concentration in the extracellular fluids is primarily regulated at the mineralized bone surfaces (Talmage and Talmage, 2006, 2007; Talmage and Mobley, 2008, 2009). The processes involved the regulation of the solubility of hydroxyapatite, which would otherwise result in an equilibrium between bone mineral and the fluid passing over it that would lower plasma calcium below the concentration necessary to maintain the normal function of mammalian tissue. Bone can be described at several levels. Bone may be considered first as a structural material, a mixture of organic matrix and inorganic (mineral) materials. Bone may also be viewed at the level of a mechanical device, with a characteristic resistance to bending, to compression, and to fracture. At a gross physical (organ) level, long bones such as the leg or arm bones may be separated into anatomical regions of epiphysis, metaphysis, and diaphysis. The total bone structure also consists of an arrangement of trabecular and cortical bone. Bone microstructure, that is, the mixture of trabecular and cortical bone in different bones or in different portions of the same bone, may be equally, if not more, important than the absolute amount of bone mineral in determining strength and resistance to fracture. The structural analysis of bone is beyond the scope of this chapter, but the reader is referred to references (Evans, 1973; Currey, 1984; Albright and Skinner, 1987; Lanyon, 1992; Mammone and Hudson, 1993) for additional information. At the microscopic or histological (tissue) level, bone consists of osteoblasts and osteoclasts that create and destroy bone, respectively. Bone as a metabolic entity serves as a storage reservoir for minerals, especially Ca and phosphorus (P), but also for other macrominerals and trace minerals. As a dynamic part of the body’s response to changing availability of these minerals, bone mineral is essential for maintaining Ca and Pi concentrations in the extracellular fluid compartment within limits acceptable for numerous functions essential to the continued well-being of the individual. Bone also undergoes extensive changes during growth, in response to physical exercise, and during maintenance of skeletal tissues throughout the life cycle.

BONE AS AN ORGAN Bone is considered a component of the body’s connective tissue, but together with teeth and cartilage, it shares the ability to add mineral to the underlying organic matrix of collagen and noncollagenous proteins, that is, to become calcified. All bone consists of mineral (hydroxyapatite) deposited

35

Skeletal Tissues and Mineralization

within an organic (collagen) matrix. The mineral alone is hard, rigid, and brittle and is effective mainly in resistance to compression. The collagen matrix on the other hand remains flexible and is mainly tension resistant, that is, strong, reducing fractures. In combination, the organic and inorganic components of bone produce a material that is one-third the weight but ten times as flexible and with approximately equal tensile strength as that of cast iron (Albright and Skinner, 1987).

Structure Bone is organized into two types of structure that are intermixed within each bone (Clarke, 2008). These structures are compact (cortical) bone, which forms the shafts of the long bones and the outer surface of almost all bones, and trabecular bone, also called cancellous or spongy bone, which fills the ends of the shafts of the long bones as well as forming most of the structure of the vertebrae (Figure 4.1). Most of the volume occupied by compact bone consists of calcified matrix (~80%), whereas only a small portion (~15%) of trabecular bone volume is calcified; the space between the trabeculae is often filled with bone marrow, especially early in life, and later fat cells become prominent in this space.

Gross Anatomy Anatomists separate bone into two types, the flat bones, such as skull bones, scapula, mandible, and ilium, and the long bones, such as tibia, femur, and humerus. The long bones consist of three anatomic divisions, as illustrated in Figure 4.1. The inner surface of the bone is called the endosteum, and the outer surface, the periosteum. The shaft, or diaphysis, is a hollow cylinder of compact bone. The ends of the bone are known as the epiphyses and consist of a thin cortical layer of compact bone surrounding a region of trabecular bone. The metaphysis is a tapering, transitional region between the epiphysis and the diaphysis containing both cortical and trabecular bone. The epiphysis and metaphysis are separated by a plate of hyaline cartilage called the epiphyseal growth plate (or physis). The growth plate is the site at which elongation of the long bones occurs by proliferation of chondrocytes, cartilage-forming cells. Calcification of this cartilaginous area at puberty results in closure of the growth plate and cessation of longitudinal growth (Cormack, 1987; Fawcett, 1994).

Anatomy of long bones Epiphyseal growth plate Trabecular bone Periosteum Cortical bone

Epiphysis Metaphysis

Diaphysis

Endosteum Marrow cavity Epiphyseal growth plate

Metaphysis Epiphysis

FIGURE 4.1  Anatomy of long bones. The long bones, i.e., the tibia, femur, humerus, and radius, are divided anatomically to three segments: the epiphyses, the metaphyses, and the diaphysis. The composition of cortical and trabecular bone varies as one moves from segment to segment.

36

Diet, Nutrients, and Bone Health

BONE AS TISSUE Bone tissue contains a number of different cell types that perform the functions necessary to build and maintain bone structure and carry out the homeostatic functions of bone. Thus, there are osteoblasts, which are responsible for formation of bone; cells derived from the osteoblasts that are present in mature bone; and osteoclasts, which are capable of resorbing bone tissue during turnover of bone. In addition to the cellular components of bone tissue, the majority of bone tissue consists of hydroxyapatite (see section “Chemical Structure” below) crystals deposited within a protein matrix that consists primarily of type I collagen. However, the other protein constituents of bone organic matrix are quite varied, and the role of many of these molecules remains to be discovered.

Microscopic Anatomy of Bone Cells Bone cells can be divided into two general types, bone-forming cells and bone-resorbing cells, which are derived from different osteoprogenitor cells. The bone-forming cells, or osteoblasts, are believed to give rise to related cell types with different functions in bone, for example, the osteocyte and the resting osteoblast, or bone-lining cell. The osteoclast is the major cell responsible for bone resorption (see Chapter 3). Bone-Forming Cells (Osteoblasts) The osteoblasts are derived from mesenchymal cells and are related to fibroblasts and the cells forming the walls of blood vessels (Figure 4.2). Preosteoblasts are found in the tissue layers near bone-forming surfaces. They are elongated cells lacking the full ability of the mature osteoblasts to synthesize bone matrix proteins, but they retain the ability to divide. Osteoblasts are bone-forming cells that secrete collagen to form the osteoid or unmineralized bone matrix and also regulate the processes that initiate mineralization. Osteoblasts are metabolically active Osteoblast lineage Stem cell ? Preosteoblast

Osteoblast

Osteocyte

Bone-lining cell

FIGURE 4.2  Osteoblast lineage. The osteoblast is derived from the same stem cell population that gives rise to mesenchymal cells, including fibroblasts; however, the source of these stem cells has not been identified. Once the stem cell differentiates to the preosteoblast, the daughter cells of this line are committed to osteoblastic development. The fate of osteoblasts once bone has been formed is threefold: (1) they may undergo apoptosis; (2) remain on the surface of the bone as biosynthetically inactive bone-lining cells; or (3) completely enclose themselves in bone, forming an osteocyte.

Skeletal Tissues and Mineralization

37

cells and are characterized by an abundant rough endoplasmic reticulum and Golgi apparatus for synthesis and export of proteins (Figure 4.2). These cells also express characteristic membrane proteins, particularly the enzyme alkaline phosphatase, which has been proposed as a regulator of mineralization for many years, although its function remains undefined. Osteoblasts possess receptors for two of the major Ca and bone-regulating hormones, parathyroid hormone (PTH) and 1,25-dihydroxyvitamin D [1,25(OH)2D]. It is generally accepted that the major effects of these hormones on maintenance of plasma Ca and on bone turnover are controlled via osteoblastic cells (see Chapter 6). Osteoblast-Derived Cells of Mature Bone Osteoblasts give rise to two cell types within bone that maintain some osteoblastic characteristics but no longer synthesize collagen for formation of new bone. Osteocytes Osteocytes are the most common type of bone cells, making up 90% of the total number of cells in bone. These bone cells that lie within small cavities or lacunae are completely surrounded by bone. They are osteoblasts that have surrounded themselves with matrix and remain connected with each other and with cells on the bone surface by gap junctions at the ends of thin processes that pass through thin tubular channels or canaliculi. The canaliculi provide a passageway for the diffusion of nutrients from extracellular fluid to these cells. Osteocytes also have the capability of resorbing bone by a process known as osteocytic osteolysis. These cells have a role in the control of bone remodeling (Henriksen et al., 2009), but a full understanding of the importance of the osteocytic activities awaits further research. Bone-Lining Cells Bone-lining cells are also called resting osteoblasts or surface osteocytes. As the names imply, these cells are osteoblast-like cells that are no longer actively forming bone but are not completely surrounded by bone. These lining cells are central to maintenance of blood Ca levels, perhaps by actively pumping Ca ions from the bone fluid compartment to the extracellular fluid compartment. Although it is likely that these cells represent another developmental stage of the osteoblast, similar to the osteocyte, it has been suggested that these cells are formed directly from preosteoblasts, which they resemble histologically. It has been demonstrated that the bone-lining cells can be stimulated to differentiate into active osteoblasts. Bone-Resorbing Cells (Osteoclasts) Osteoclasts are multinucleated, giant cells that are capable of resorbing both bone mineral and matrix (Blair et al. 1993). The osteoclasts are formed from cells of the monocyte–phagocyte lineage (Figure 4.3). The cellular lineage of the osteoclast was defined in experiments with the disease osteopetrosis, a condition of dense bone growth without osteoclastic remodeling. Animal models with this disease either lack the mononuclear precursor cells or these cells fail to respond to the cytokines that signal cell proliferation and maturation. In either case, the number of functional osteoclasts is insufficient to resorb substantial amounts of bone, and hence, bone turnover is significantly depressed in osteopetrosis. (This disease in humans also may result from defects in the machinery of acidification [Stark and Savarirayan, 2009].) The osteoclast resorbs bone in a region of attachment of the cell to the bone surface where the cell membrane is elaborately folded, resulting in the characteristic appearance of this area called the “ruffled border” (Figure 4.4; Henriksen et al., 2008; Boyce et al., 2009). Bone resorption involves the lowering of the pH in the region between the osteoclast and the bone surface through secretion of hydrogen ions (H+), which are generated within the osteoclasts by carbonic anhydrase II (the same enzyme that is present in H+ (acid)-secreting cells of the stomach and kidney). The site of bone resorption is separated from the surrounding tissue by a ring of contractile proteins, the sealing zone, which is formed by actin filaments anchored through integrins to the bone surface. Within this zone, the bone surface is exposed to low pH and concentrated lysosomal enzyme activity, leading to

38

Diet, Nutrients, and Bone Health Osteoclast lineage CFU-GM (+) Promonocyte

IL-1; IL-6; TGF-α

(+)

(+)

Early pre-osteoclast

Monocyte

TGF-β

Tissue macrophage

Late pre-osteoclast

(+)

PTH; 1,25D

Osteoblast

(+)

(+)

PTH; 1,25D

IL-6

GM-CSF

Osteoclast

Giant cell

(+) (+)

(–)

CT

Active osteoclast

PTH; 1,25D; IL-1; TGF-α

FIGURE 4.3  Osteoclast lineage. Osteoclasts derive from a stem cell population (colony-forming unit–granulocyte–macrophage or CFU-GM) that is capable of giving rise to osteoclasts or monocytes/macrophages. CFU-GM is stimulated by the granulocyte–macrophage colony-stimulating factor to differentiate into premonocytes. The premonocyte may commit to either the osteoclastic or monocytic pathway under the stimulus of local factors (cytokines), including interleukin-1 (IL-1), interleukin-6 (IL-6), and transforming growth factor-β (TGF-β) for the osteoclastic lineage and transforming growth factor-α for the monocytic lineage. Further development of the osteoclastic cell is stimulated by PTH and 1,25-dihydroxyvitamin D, which may act directly on the differentiating osteoclasts or may act through the osteoblast. Other cytokines, such as IL-1 and TGF-β, also stimulate resorptive activity through the osteoblast. The active osteoclast may also be regulated directly through stimulatory cytokines such as IL-6 and by inhibitory agents such as the circulating peptide CT.

the description of this region as a “secondary lysosome.” However, Blair et al. (1993) have pointed out that the ability of the osteoclast to acidify the region of bone adjacent to the ruffled border vastly exceeds the transport of H+ into cytoplasmic lysosomes.

Chemical Structure Although bone is organized into different structural components, at the molecular level, all bone tissues share a common composition. Bone consists of an inorganic or mineral phase composed primarily of hydroxyapatite crystals deposited within an organic phase of cross-linked collagen fibers.

39

Skeletal Tissues and Mineralization – HCO3

Chloride–bicarbonate exchanger

CO2 Cl–

– HCO3 H2CO3 H+

RER

Chloride channel Cl–

ATP

CA II H2 O + CO2 ADP + Pi

Golgi

Proton pump H+

FIGURE 4.4  Osteoclast. The principal, and perhaps the only, bone-resorbing cell is the osteoclast. This multinucleated cell forms a “sealing zone” with the bone surface that allows a local area of bone to be exposed to low pH, which is produced by active pumping of hydrogen ions into the resorption pit, and a high concentration of proteolytic enzymes, which are released from lysosomes into the area of resorption.

The mineral phase provides stiffness and resistance to compression, whereas the organic phase gives the bone strength and resistance to breaking. Together, this composite structure has both rigidity and resistance to fractures. Hydroxyapatite has the chemical structure Ca10(PO4)6(OH)2 and makes up approximately 60%–65% of bone by weight. Type I collagen fibers make up 90% of total bone proteins. When bone is first formed as part of a newly modeled bone or as part of the repair of a fracture, the collagen fibers are randomly oriented, and this bone is referred to as woven bone. However, as bone is remodeled during growth or as part of normal bone turnover, the collagen fibers are laid down in parallel arrays that are visible under polarized light as layers within the bone. Thus, this mature bone is referred to as lamellar (layered) bone.

Collagen and Noncollagenous Proteins of Bone Matrix As stated above, type I collagen makes up approximately 90% of bone matrix protein. The remaining 10% of total bone proteins consists of a mixture of proteins that are either synthesized by osteoblasts or are serum derived (Termine, 1988). Whether the serum-derived proteins play a role in bone metabolism or whether they are simply trapped during the formation of bone matrix is not known. The remaining noncollagenous proteins of bone matrix (Table 4.1) are synthesized primarily by osteoblasts, and less so by osteoclasts. These additional matrix proteins, which function as either structural or chemical signaling proteins, can be divided into four general classes: Cell-Attachment Proteins Cell attachment is mediated by the family of proteins that constitute the integrins. Integrins are membrane-spanning proteins that recognize the amino acid sequence Arg–Gly–Asp (RGD) in extracellular proteins. Bone cells synthesize four proteins, that is, fibronectin, thrombospondin,

40

Diet, Nutrients, and Bone Health

TABLE 4.1 Possible Functions of Noncollagenous Proteins of Bone Matrix Noncollagenous Protein of Bone Matrix Thrombospondin Fibronectin Bone sialoprotein Osteopontin Proteoglycan I (biglycan) Proteoglycan II (decorin) Osteonectin Osteocalcin Matrix Gla protein Insulin-like growth factor-1 Platelet-derived growth factor Transforming growth factor-β

Possible Function Cell attachment Cell attachment Cell attachment Cell attachment Unknown Unknown Calcium binding Calcium binding Unknown Growth related Growth related Growth related

osteopontin, and bone sialoprotein (BSP), that are part of the extracellular matrix and serve as sites for cell attachment. Only BSP is confined primarily to bone, whereas the other proteins are found in many nonskeletal connective tissues. Proteoglycans Proteoglycans consist of a central core protein to which acidic polysaccharide side chains are attached. The side chains or glycosaminoglycans in bone are primarily chondroitin sulfate and heparin sulfate. The proteoglycans may facilitate osteoblast interaction with cell-attachment and other proteins, but their function remains unknown. γ-Carboxylated (Gla) Proteins The bone Gla proteins contain γ-carboxylated glutamine residues that are the product of a vitaminK-dependent enzymatic process. The addition of the second carboxyl group to the side chain of the glutamine residue produces a binding site for Ca ions. Ca ions are essential cofactors in the function of γ-carboxylated proteins of the blood-clotting cascade, and the ability of the bone Gla proteins to bind Ca plays a role in the regulation of bone mineralization (Krüger et al., 2009). Osteocalcin, a bone Gla protein, is found only in bone, where its synthesis by osteoblasts is regulated by 1,25(OH)2D. The serum concentration of osteocalcin is often regarded as a marker for bone turnover. A second γ-carboxylated protein, matrix Gla protein, is found in both bone and cartilage. Growth-Related Proteins Growth-related proteins represent the largest number of different protein molecules found in bone. In addition to the insulin-like growth factors (IGF-1 and IGF-2), bone also contains TGF-β1–5 and other proteins secreted by osteoblasts that stimulate osteoblast growth in a paracrine/autocrine manner (Hauschka et al., 1988). One role suggested for these factors after they are liberated by osteoclastic action has been in the recruitment of osteoblasts to areas of bone resorption. Other osteoblast-derived proteins found in bone matrix include alkaline phosphatase and osteonectin, a glycoprotein that makes up approximately 2% of bone protein in developing bone. Alkaline phosphatase has long been associated with osteoblasts and its synthesis, and secretion of this enzyme by these cells is considered a marker of bone formation. The function of alkaline phosphatase in bone cell biology, however, must be considered unknown. Osteonectin (also called secreted protein acidic and rich in cysteine or SPARC) is found in both proliferating bone and in

Skeletal Tissues and Mineralization

41

growing tissues and is the most abundant noncollagenous protein in bone (Kapinas et al., 2009). This glycoprotein promotes osteoblast differentiation and supports cell survival, as well as likely plays a role in maintaining the balance between bone resorption and formation.

BONE FORMATION All bone is formed by deposition of hydroxyapatite crystals within a collagen matrix. However, the process by which the collagen matrix is laid down differs according to the two major types of bone formation, intramembranous and endochondral bone formation (bone modeling on a cartilaginous structure) (Deschaseaux et al., 2009).

Intramembranous Bone Formation The flat bones of the skull, the jaw, and the ribs are formed by intramembranous bone formation. In this process, mesenchymal cells differentiate directly into preosteoblasts and then into osteoblasts in areas of highly vascular embryonic connective tissue. The bone matrix synthesized by these cells is randomly oriented (or woven), and calcification is accomplished in irregular patches rather than in the orderly fashion seen in remodeling bone. Eventually, this bone is remodeled to form lamellar bone.

Endochondral Bone Formation (Bone Modeling) The long bones are formed de novo by calcification of a cartilaginous structure that serves as a model for the new bone. Cartilage is formed by chondroblasts, cells that are derived from prechondroblasts arising from the same mesenchymal cells that form osteoblasts. The chondrocytes secrete a collagen matrix, eventually enclosing themselves into lacunae. Because cartilage consists primarily of collagen fibers and is not rigid, chondrocytes can continue to enlarge and divide, expanding the lacunar spaces. Cartilage also expands at the periphery by appositional (circumferential) growth. Cartilage is an avascular tissue, and the cells receive their nutrients by diffusion through the fluid surrounding the collagen fibers. Invasion of the cartilage by blood vessels initiates the process of mineralization of the organic matrix. As a result of the reduced diffusion of nutrients to the chondrocytes after mineralization, these cells begin to die. Not all chondrocytes are destroyed during initial calcification of the bone model, however. At each end of the bone, a layer of epiphyseal cartilage remains (see Figure 4.1), and this epiphyseal growth plate serves as the site of bone formation for elongation during growth. As chondrocytes proliferate, the oldest layers begin to calcify. This calcified cartilage is resorbed by osteoclasts, leaving calcified longitudinal septa. Osteoblasts form additional woven bone on these septa to form trabeculae of the primary spongiosa. As additional remodeling of these trabeculae takes place, lamellar bone replaces the woven bone and secondary spongiosa is formed. The epiphyseal cartilage calcifies after puberty as a result of increases at that time of estrogens in females and testosterone in males. Epiphyseal closure occurs at an earlier age in females, resulting in earlier cessation of longitudinal growth. Delayed puberty, for almost any reason except undernutrition, results in elongation of the long bones and increased height in both females and males.

Mineralization Bone mineralization involves the formation of hydroxyapatite crystals within calcifying cartilage or newly laid down organic matrix (osteoid) in remodeling bone or new, woven bone. Two distinct mechanisms operate to deposit mineral in bone (Bonucci, 1971; Skinner, 1979; Termine etal., 1980). In lamellar bone formed during remodeling, mineral is deposited in association with the tightly spaced collagen fibrils and associated noncollagenous proteins. In the more open environment of randomly oriented collagen fibrils in cartilage and in woven bone, mineralization is

42

Diet, Nutrients, and Bone Health

characterized by matrix vesicles that provide a site for hydroxyapatite crystal formation (Anderson et al., 2005; Nahar et al., 2008). These 50- to 200-nm diameter membrane-invested vesicles are formed by exocytosis from the plasma membrane of chondrocytes and osteoblasts. Crystallization eventually produces a crystal larger than the original vesicle and the vesicle is destroyed. Thus, the earliest bone formation either in endochondral bone formation or in woven bone proceeds by vesicle-driven mineralization. When bone matrix is first synthesized, it requires a period of time before mineralization can occur. Thus, sites of active bone formation are marked by a layer of unmineralized bone matrix, or osteoid, that is not yet calcified. It is not known whether this delay is due to removal of inhibitors of crystal formation or to formation of proteins that are necessary for initiation of mineralization. However, about 5 to 10 days after new matrix is laid down, an initial, rapid phase of bone mineralization is initiated (Bala et al., 2009), which results in mineralization at about 50% to 70% of the maximum level eventually achieved. A secondary phase follows this and is characterized by a slow, gradual maturation of the mineral component that takes up to 30 months to complete. Several inhibitors of bone mineralization have been identified, including pyrophosphate and phosphoproteins associated with calcified tissues (Termine et al., 1980; Orimo, 2010). The rate of mineralization appears to be related to the presence of these inhibitor molecules. Pyrophosphate is generated by NPP1 and is transported out of the cell by ANKH. The pyrophosphate is hydrolyzed by tissue nonspecific alkaline phosphatase to remove the inhibition and provide inorganic phosphate, which promotes mineralization. Bone modeling adapts structure to loading by changing bone size and shape and removes damage and so maintains bone strength. Remodeling is initiated by damage-producing osteocyte apoptosis, which signals the location of damage via the osteocyte–canalicular system to endosteal lining cells that form the canopy of a bone remodeling compartment (BRC). Molecular signaling within the BRC between precursors, mature cells, cells of the immune system, and products of the resorbed matrix titrates the birth, work, and lifespan of this remodeling machinery to either remove or form a net volume of bone (see also Chapter 3).

BONE RESORPTION Bone resorption is defined as removal of both bone mineral and organic matrix to produce a cavity in the bone structure, that is, a resorption pit or Howship’s lacuna. This process is carried out primarily by the osteoclast (Figure 4.3). Osteoclasts remove bone in a regulated fashion essential for both the remodeling process and for reconfiguring bone during the modeling process in endochondral bone formation. The major difference in osteoclastic regulation in these two processes is the time-dependent coupling of bone resorption with bone formation that is a characteristic of bone remodeling and the simultaneous, but physically separated, resorption and formation during modeling. Molecular signals regulate the activity of bone cell precursors and the mature bone cells to determine the extent of bone resorption and formation in the remodeling process (Seeman, 2009). Once bone is dissolved beneath the osteoclast, the Ca and Pi ions and amino acids resulting from protein hydrolysis are removed from the resorption pit (Howship’s lacunae) by several possible mechanisms: (1) translocation of individual components through the cytoplasm; (2) bulk transfer of these products via endocytotic vesicles; or (3) release of products when the osteoclast detaches along the sealing zone. The last mechanism is supported by microscopic evidence that osteoclasts are capable of moving along the surface of bone and forming multiple resorption pits. This suggests that the osteoclast is capable of continued bone resorption, but local factors may regulate the maximum size of a given resorption pit. Resorption of bone on the surface of either trabecular or compact bone results in excavation of shallow Howship’s lacuna. However, in compact bone, a group of osteoclasts may continue to resorb bone deep into the compact bone, forming a cylindrical opening through the bone; this cylindrical structure is known as an osteon or Haversian system. Osteoblasts will line the surface of this

43

Skeletal Tissues and Mineralization Haversian system of cortical bone

Osteoclasts

Cutting cone

Osteoblasts

Bone formation

Bone lining cells

Mature osteon

Haversian canal

Lamellar bone

FIGURE 4.5  Haversian systems. The Haversian systems (osteons) of compact bone are formed when a “cutting cone” of several osteoclasts excavates an area of bone. At the reversal zone between the cutting cone and area of bone formation, osteoblasts move in and begin replacing the bone with concentric layers of lamellar bone. An opening is left through the center of the Haversian canal through which blood vessels and nerves pass.

cylinder and lay down new bone in concentric rings until they have filled the excavation except for a central region through which blood vessels and nerves extend (Figure 4.5). The activity of the osteoclasts is stimulated by PTH and vitamin D, whereas osteoclasts are inhibited by calcitonin (CT) and estrogens. Thus, the regulation of osteoclastic activity by these hormones is believed to be mediated by cytokines secreted by the osteoblasts (McSheehy and Chambers, 1986). Osteoblasts are responsive to both PTH and vitamin D and are believed to produce factors with osteoclast osteoclast-resorption-stimulating activity that are either secreted into the surrounding extracellular fluid or are bound to the cell membrane of the osteoblast (or extracellular matrix) and affect the osteoclasts via direct contact with the cell (Fuller et al., 1991).

BONE REMODELING During growth, the bones must be constantly reshaped to maintain their optimal shape for support. As elongation occurs, the protuberance of the epiphysis must be reduced by osteoclastic resorption as this volume is filled first by the metaphyseal region and eventually by the cylindrical diaphysis (Figure 4.6). In growing bone, bone resorption and formation take place in separate regions of the bone, with formation exceeding resorption, so that gains in bone mass occur as part of early-life skeletal development. In mature bone, osteoblastic bone formation occurs only at sites of osteoclastic activity and ideally results in replacement of the exact quantity of bone removed by resorption (Figure 4.7). When bone formation fails to keep pace with bone resorption, a net decrease in bone mass results; osteoporosis is marked by such a discrepancy between removal of bone and its replacement. The rate of bone turnover is much higher in trabecular bone, where approximately 25% of bone is resorbed and replaced annually, whereas there is only about 3% of cortical bone turnover each year (Manolagas and Jilka, 1995).

44

Diet, Nutrients, and Bone Health

Areas requiring resorption during growth

The midshaft undergoes appositional growth at the periosteal surface and resorption at the endosteal surface to maintain cortical thickness

Areas requiring resorption during growth

FIGURE 4.6  Reshaping of a long bone during growth (modeling). During growth, the bones must be reshaped to maintain the bone model. The outer aspect of the epiphysis is resorbed as the bone length increases at the epiphyseal plate. The midshaft undergoes both resorption on the endosteal surface and growth on the periosteal surface and maintain a similar cortical thickness as the diameter of the bone increases. The process of modeling is distinct from that of remodeling (Figure 4.7), in which the resorption of bone and its subsequent formation are tightly coupled at individual sites. a

h

b

g

c

f

d

e

FIGURE 4.7  Coupling of remodeling in Howship’s lacunae. Remodeling of bone is a multistep process beginning with (a) activation at an area of bone surface covered by bone-lining cells. (b) Osteoclast(s) are recruited to this site and (c) excavate a resorption pit (Howship’s lacuna). During the (d) reversal phase, the osteoclast moves away from the resorption pit and (e) osteoblasts move in and (f) begin replacing the resorbed bone with new layers of bone. (g) Once this process is complete and the bone tissue has been completely restored, (h) the osteoblasts are replaced by quiescent bone-lining cells again. (Talmage, R.V., Grubb, S.A., and VanderWiel, C.J. 1983. Physiologic processes in bone. In The Musculoskeletal System: Basic Processes and Disorders, Wilson, F. C., ed., 2nd ed., Chapter 8. J.B. Lippincott Co., Philadelphia.)

The Bone Multicellular Unit and Bone Remodeling The concept of bone remodeling was advanced by Frost (1963). He later introduced the concept of the bone multicellular unit for the multicellular sequence of changes in bone that resulted in renewal of bone matrix and mineral in distinct units. As the organic matrix ages, it becomes brittle and weak and undergoes microfractures. Bone remodeling replaces it with new matrix. This process of remodeling of mature bone consists of

Skeletal Tissues and Mineralization

45

a predictable and ordered sequence of bone resorption followed by bone formation, and it can be divided into the following three stages:

1. Activation of osteoclasts 2. Resorption by osteoclasts 3. Formation by osteoblasts

The periods of resorption and formation are also separated by a period sometimes called the reversal phase, as osteoclasts are replaced by osteoblasts at the resorption site.

Regulation of Bone Remodeling The processes of bone remodeling also are involved in the response of bone to strain, such as repeated exercise, that results in greater bone mass and in stronger bone, and remodeling is essential for the healing of bone fractures. Although bone remodeling could occur in a random fashion, intersecting the site of microfracture damage by chance, a process has been described whereby bone cells, particularly osteocytes, act as mechanosensors to regulate these activities so that areas of damage are replaced preferentially (Robling and Turner, 2009). The process by which osteoblasts replace the bone removed from the bone surface is closely regulated or coupled (Martin et al., 2009) (see Chapter 3). The regulation of bone remodeling requires interactions between bone-resorbing and bone-forming cells. Experiments with isolated osteoblasts and osteoclasts have demonstrated that osteoclasts alone are not responsive to stimulation; however, when cocultured with osteoblasts, addition of PTH or 1,25(OH)2D can promote bone resorption by osteoclasts. In addition, exposing osteoclasts to medium conditioned by osteoblasts stimulated by PTH or 1,25(OH)2D has a similar stimulatory effect on bone resorption. Therefore, osteoblasts must synthesize a factor that promotes bone resorption by the osteoclast. Numerous proteins are present in bone along with collagen. Several of these are large glycoprotein molecules that likely serve primarily structural function. Many soluble proteins, however, are also incorporated into mineralized bone at the time of matrix formation. As discussed above (see the section on noncollagenous proteins of bone matrix), protein growth factors secreted by osteoblasts or adsorbed from plasma are almost always present in mature bone. Liberation of these factors during bone resorption may serve as signals for the recruitment of osteoblasts to resorption sites. The local factors that are secreted by osteoblasts and other cells of the bone marrow or stromal cells are proteins known collectively as cytokines (Chambers, 1988; Mundy, 1992; Roodman, 1993). Cytokines may either stimulate osteoclastic bone resorption or inhibit resorption and stimulate bone formation. Factors in the local environment of bone that stimulate osteoclastic bone resorption include interleukin-1, transforming growth factors-α (TGF-α), tumor necrosis factor-α (TNF-α), and interleukin-6. Bone resorption is inhibited by TGF-α, γ-interferon, and interleukin-4. Because these agents are secreted by cells of the lymphocyte and macrophage lineages, bone marrow is a rich source for these peptides. Osteoblast regulation of proliferation of osteoclast precursors is regulated by macrophage colonystimulating factor, which is secreted by the osteoblast. The receptor activator of NFκB (RANK) ligand (RANKL) is a member of the TNF superfamily that is expressed on the membrane of osteoblasts and activates receptor RANK on osteoclast precursors (Yavropoulou and Yolos, 2008; Seeman, 2009). Osteoblasts can also inhibit osteoclast formation through osteoprotegerin, a secreted protein that acts as a decoy receptor for RANKL. The investigation of the role of cytokines in bone remodeling and in the regulation of bone cell metabolism will be greatly accelerated in the near future by the development of specific animal models with selected deletion or overexpression of specific cytokines and/or their receptors. In many cases, researchers may be able to target specifically these mutations to cells of bone, or they may mutate the genes for these proteins in all cells.

46

Diet, Nutrients, and Bone Health

LIFESTYLE EFFECTS A few topics relating to bone formation and maintenance require special emphasis.

Peak Bone Mass Accrual At some time during the third decade of life, individuals will have accumulated the largest amount of bone tissue they will have at any time during their lifetime. This peak bone mass is a critical factor in contributing to the risk of osteoporotic fracture later in life. The attainment of peak bone mass by about the age of 30 years depends on the genetic inheritance of an individual, on nutrition from the prenatal period through adulthood, and on physical activity.

Effect of Exercise (Strain) on Bone As discussed in Chapter 5 of this volume, mechanical stress on bone influences skeletal adaptation, an effect that may be particularly important during the growth of the skeleton in early life and in the aged. Considerable evidence supports a role for mechanical loading (strain) or gravitational force in maintaining bone mass (Rubin and Lanyon, 1984). Individuals that are confined to long-term bed rest or that have participated in space flights lose bone. If bones are not subjected to load through mechanical usage, normal bone mass development will be limited to about 30% to 50% of normal, and even the cross-sectional shape of the long bones will fail to develop normally (Robling and Turner, 2009). Bone mineral density (BMD) in the spine of postmenopausal women can be increased by aerobic, weight-bearing, and resistance exercises (Iwamoto et al., 2009). Although extensive loss of bone can be observed in a relatively short period of time with inactivity or weightlessness (Donaldson et al., 1970; Vogel et al., 1977), the gain in bone mass with exercise is much smaller. Comparison of bone mineral content (BMC) and BMD in young adult women suggests that increased physical activity may offset, at least in part, an inadequate dietary Ca intake (Metz et al., 1993). Studies in experimental animal models generally support the concept that the response to mechanical loading is a local one and does not affect other bones, including the contralateral limb (Sugiyama et al., 2009). The adaptive response affects both the cortical and trabecular bone. Mechanical loading signals changes in bone cells that produce biochemical and eventually structural changes in bone (Robling and Turner, 2009). As theorized over 100 years ago by Julius Wolff, bone adjusts its structure to adapt to the loads placed upon it.

AGING AND OSTEOPOROSIS Aging is generally associated with a net loss of bone tissue at all skeletal sites. The bone appears normal, and excessive osteoclasts cannot be found, but too little bone, or osteopenia, exists. Thus, bone is more susceptible to fracture, especially the trabecular bone tissue. Osteoporotic fractures occur most commonly to the vertebrae, distal radius and ulna (wrist), proximal femur (hip), or humerus (upper arm). Bone loss in aging occurs mainly from the endosteal surface of the cortex, which would result in thinning and weakening of the structure; however, exercise can increase the rate of periosteal bone formation (Seeman, 2009) (see Chapter 5). This periosteal increase can result in a bone with larger diameters of both the internal and external part of the cross-section, but the mechanical strength of the bone is maintained.

ASSESSMENT OF BONE MASS AND BONE TURNOVER A great need exists for accurate methods of assessing bone mass and bone turnover for either diagnosis in subjects who are believed to be at risk for bone loss or assessment of treatment modalities in those who have already lost bone. Although the most accurate information on bone quality and bone

Skeletal Tissues and Mineralization

47

remodeling activity at a particular site can be obtained by bone histomorphometry, the invasive nature of this procedure limits its usefulness in diagnosis and treatment of bone disorders. Although X-rays can be used to provide some information about gross changes in bone structure, the need to measure much smaller changes in BMC or BMD has led to the development and refinement of instruments that can be used to noninvasively determine bone quality in a particular area with continually improving accuracy and reproducibility. Another goal has been to reduce the exposure to radiation as low as possible. Several general types of these instruments are discussed below. Although BMC and/or BMD can be measured with accuracy and reproducibility approaching 1%–2%, this provides information only on the current status of the bone, not on changes in bone formation and resorption that can be related to treatment. Some of the molecules that are present in blood and/or urine are discussed below as possible markers for the dynamic processes occurring within the bone.

Bone Histomorphometry The most accurate estimates of bone mass are obtained by direct sampling of the bone, for example, transiliac bone biopsy, followed by biochemical and histological analyses. Because of the invasive nature of this procedure, which requires removal of a cylindrical core of bone through the iliac crest, it is not in routine use. However, as a research tool, this procedure can provide valuable information about the rates of bone formation and turnover when bone is labeled in vivo with tetracycline, fluorescein, or other fluorescent markers of bone mineralization. Together with well-established staining techniques for bone, these fluorescent compounds can be used to measure the actual bone turnover rate during the period of study and define the status of the population of bone cells in the region examined. These methods are frequently used to assess bone abnormalities in patients with chronic kidney disease because of the effects of renal failure on bone health (Ott, 2008).

In Vivo Measurements of BMC and BMD The need to assess BMC and/or BMD in large numbers of individuals has prompted the development of several noninvasive methods that give quantitative results, as opposed to qualitative estimates from standard X-rays. The ideal for a useful method of measuring bone densitometry is a technique that is both accurate, that is, that provides an estimate of bone mass or density that is very close to the true value, and precise, that is, being able to measure reproducibly the same site in the skeleton. Because of the relatively slow rate of change in bone mass during most of the life cycle, precision in the range of 1%–2% is necessary to measure meaningful changes in bone mass or density even when studies extend over a year or more. Older methods, such as radiographic absorptiometry, single-photon absorptiometry, and dual-photon absorptiometry, have largely been replaced. Dual-Energy X-Radiographic Absorptiometry Dual-energy X-radiographic absorptiometry (DXA) was an important advance in the diagnosis of osteoporosis by measurement of BMD when it was first introduced in routine clinical practice in 1987 (Blake and Fogelman, 2009). DXA is generally considered the best method for diagnosing osteopenia or osteoporosis on the basis of the World Health Organization (WHO) T-score definition of osteopenia as −2.5 < T < 1.0 and osteoporosis as −2.5. The T score is calculated based on the mean BMD for healthy young adults (aged 20–29 years). In addition to the usefulness of DXA in diagnosing osteoporosis, its BMD results also can be used to predict fracture risk, to target therapies against fractures, and to monitor the response to those treatments. The basic principle of DXA is that it utilizes two monoenergetic X-rays (gamma photons) that are generated by an X-ray machine and then focuses on a target tissue site for a short period. A machine-driven tracking system allows the source to move across (laterally) different sites of interest or the entire body from head to toes. One of the X-rays is harder, and the other is soft. The harder

48

Diet, Nutrients, and Bone Health

X-ray is stopped to a large extent by the skeletal (mineralized) tissue, whereas most of the soft X-rays are stopped by the soft tissue, including fat. The algorithm uses the weight, height, sex, and age of the individual, as well as the X-ray attenuation data for calculating BMC or BMD, fat mass, and lean body mass, exclusive of bone, that is, a three-compartment model of body composition. DXA instruments are typically found in radiology departments of hospitals where radiation safety is carefully monitored. The radiation dosages to subjects with either of these methods are very low, for example, the amount received by a person flying at 40,000 ft from New York to Denver or about a tenth of the radiation dose of a standard chest X-ray. Babies, small infants, and even pregnant women can be measured with the DXA technique because of the low radiation doses, but of course, an experimental study measuring the bone mass of pregnant women would be almost impossible to do. Bone Measurements of BMC and BMD using DXA are considered very accurate (3% to 4%) for normalweight or thin individuals, but not for heavily muscled and obese subjects because of the greater attenuation by increased mass from soft tissues. Precision, however, remains good (2%) for all subjects. Measurements of the axial skeleton can be made with this method, as well as those of the appendicular skeleton. In addition, total body measurements from head to toe (or neck to toe) can be made. A special arm attachment (device) permits separate measurement of the radius and ulna. These measurements can be done quite rapidly (fast mode) or more slowly. Fat Therefore, the normalized data set used for comparison does not apply to very obese individuals (>50% body mass) or to sick patients who have lost much of their lean tissue, who may also be dehydrated, or who may have shifts of water (fluid) from blood to extravascular fluid compartments. The potential pitfalls of using DXA have not so dampened the enthusiasm of investigators that they are discarding this method; to the contrary, they are trying different ways to improve the method for evaluating body fat under conditions of excessive obesity. Quantitative Computed Tomography Although quantitative computed tomography (QCT) was introduced in the mid-1970s, it is less widely used than DXA (Adams, 2009). QCT does have several advantages, including the ability to measure cortical and trabecular BMD separately, in providing a measurement related to the volume of bone rather than the cross-sectional area and in providing information on geometric and structural characteristics of the bone that contribute to its strength. QCT requires a slightly higher radiation dose compared with DXA, but it is similar to the spinal radiographs that are used to diagnose osteoporosis. The technique, however, is only used routinely for estimating vertebral bone density, especially of the lumbar region of the spine. Comparison of single measurements using QCT and DXA cannot be made, and QCT cannot be used for calculation of the WHO T-score definitions of osteopenia and osteoporosis, but multiple measurement trends can be used for comparison. Other Methods Another method for the assessment of BMD and diagnosis of osteoporosis is quantitative ultrasound, which has advantages of low cost and lack of ionizing radiation compared with DXA and QCT (Guglielmi and de Terlizzi, 2009). Positron emission tomography has been proposed as a method for detection of microdamage in bone, but results have only been reported in a rodent study (Li et al., 2005). Initial results with a microindentation method for in vivo measurement of the mechanical properties of bone tissue were reported in 2010 (Diez-Perez et al., 2010). Summary The best experimental use of these bone-densitometric machines is for prospective investigations of BMC and BMD in the same individuals using the same machines and operators to limit errors,

Skeletal Tissues and Mineralization

49

such as geometry, positioning, and operator differences. DXA has become the method of choice for experimental studies because of its wide availability, high precision, and ease of use, despite some disadvantages of linear density estimates. Furthermore, the low dose of radiation of DXA machines makes them quite safe.

Bone Markers in Blood and Urine The importance of balanced bone resorption and formation in maintenance of bone health has resulted in continued efforts by researchers and clinicians to find molecules in urine or plasma that reflect bone resorption and/or formation (Delmas, 1993). These markers of bone turnover are likely to change over short periods, for example, hours or days, as opposed to the months or years required for the changes in bone mass or density (described above). Thus, these methods can be used to follow therapeutic intervention in metabolic bone diseases such as osteoporosis (Garnero, 2009). Candidates for these markers are molecules that reflect the enzymatic activities of osteoblasts or osteoclasts or the products resulting from the breakdown of bone tissue. Markers of Bone Formation Alkaline phosphatase is an enzyme that is characteristic of osteoblasts, and an increase in the bone-specific isoenzyme, such as b-s ALP, reflects the rate of osteoblastic bone formation. Total blood alkaline phosphatase measurement, however, is less useful because the synthesis of isozymes of alkaline phosphatase in liver, gastrointestinal tract, and other tissues makes measurement and interpretation of changes difficult. Osteocalcin, also known as bone Gla protein, is one of the proteins that make up bone matrix along with collagen, although its function is not known (see Chapter 12). Synthesis of this protein is dependent on vitamin K. Like collagen and alkaline phosphatase, osteocalcin is synthesized by the osteoblasts, and its blood levels may reflect the level of bone formation. Another molecule, procollagen type I N-terminal propeptide (PINP), is one of two propeptides (the other is C-terminal propeptide [PICP]) synthesized as part of type 1 procollagen that are cleaved by specific proteases to produce the mature type 1 collagen molecule. PICP is cleared more quickly than PINP (Lüftner et al., 2005). Newer markers for bone formation include BSP, osteopontin, periostin, and urinary midmolecule osteocalcin fragments (Garnero, 2009). Markers of Bone Resorption Hydroxyproline is a modified amino acid that is unique to collagen and is measurable in both blood and urine. When collagen is degraded during bone resorption, the hydroxyproline residues are released from the protein and are not reused. Therefore, the amount of hydroxyproline in the blood and/or urine reflects the level of osteoclastic bone resorption, assuming no alteration in turnover of other tissues containing collagen. Osteoclasts synthesize a tartrate-resistant acid phosphatase (TRAP) that may be measured in blood. Changes in serum concentrations of TRAP may be reflective of bone resorption by osteoclasts. Mature collagen contains cross-links between adjacent protein molecules. These pyridinoline and deoxypyridinilone intermolecular cross-links can be measured in urine and in some cases in serum, and they are indicative of resorption of collagen molecules because the cross-links are not hydrolyzed before excretion. However, collagen molecules other than bone collagen contain significant amounts of pyridinoline cross-links. The N-terminal (NTX) and C-terminal (CTX) telopeptides are generated from the collagen I molecule by cathepsin K digestion, whereas carboxy-terminal cross-linked telopeptide (CCXMMP) and carboxy-terminal telopeptide of type I collagen (ICTP) are generated by matrix-metalloproteases (MTX). Additional candidates for markers of bone resorption are tartrate-resistant acid phosphatase (TRACP5) and cathepsin K (Garnero, 2009). TRACP5 is a subform specific to osteoclasts, and cathepsin K is a bone-resorbing enzyme expressed selectively in osteoclasts.

50

Diet, Nutrients, and Bone Health

Summary Bone markers, although useful for estimating changes in turnover, have not yet been shown to be as accurate as would be desirable. Because this area of understanding is being actively investigated, better use of these markers should emerge in the future.

SUMMARY As an organ, bone can be described by its anatomic location, shape, and function. The organization of bone into cortical and trabecular microstructure is essential for the supportive and metabolic functions provided by the skeleton. At the tissue level, bone contains several different types of cells that are responsible for bone formation (osteoblasts) and bone resorption (osteoclasts) during both growth and maintenance of the skeleton. Although bone appears externally to be an unchanging structure, its microenvironment is remarkably dynamic. Not only does bone undergo microscopic remodeling processes that are constantly replacing small packets of bone throughout the skeleton, but at the chemical level, bone is constantly exchanging Ca and Pi with extracellular fluid and ultimately with the plasma. The regulation of Ca homeostasis is complex, involving interactions with kidney and gut and with hormonal factors, especially PTH and vitamin D. The control of bone turnover is even more complex as circulating hormones known to affect bone, including PTH, vitamin D, CT, and estrogen, change in their concentrations. The roles of local bone factors, for example, cytokines, growth factors, and noncollagenous matrix proteins, in the regulation of bone formation and remodeling are only now beginning to be elucidated. Once the bone matures and the individual attains peak bone mass, a phenomenon that generally occurs between 20 and 29 years of age for different parts of the skeleton, the maintenance of bone health is dependent on the balance between bone resorption and formation. When formation fails to replace the same amount of bone removed by osteoclastic resorption, bone mass decreases. In late life, the resulting low bone mass or osteopenia is eventually evidenced as osteoporosis when an increase in fracture risk occurs. Current research suggests that physical activity and adequate nutrition during periods of bone accretion and bone maintenance contribute to attaining optimal peak bone mass, as well as to minimizing the rate of loss with aging in older individuals.

REFERENCES Adams, J.E. 2009. Quantitative computed tomography. Eur J Radiol 71: 415–424. Albright, J.A., and Skinner, H.C.W. 1987. Bone: Structural organization and remodeling dynamics. In The Scientific Basis of Orthopaedics, Albright, J.A. and Brand, R.A., eds. Appleton & Lange, Norwalk, CT, 161 pp. Anderson, H.C., Garimella, R., and Tague, S.E. 2005. The role of matrix vesicles in growth plate development and mineralization. Frontiers Biosci 10: 822–837. Bala, Y., Farley, D., Delmas, P.D., et al. 2009. Time sequence of secondary mineralization and microhardness in cortical and cancellous bone from ewes. Bone 46: 1204–1212. Blair, H.C., Schlesinger, P.H., Ross, F.P., et al. 1993. Recent advances toward understanding osteoclast physiology. Clin Orthop Rel Res 294: 7–22. Blake, G.M., and Fogelman, I. 2009. The clinical role of dual energy X-ray absorptiometry. Eur J Radiol 71: 406–414. Bonucci, E. 1971. The locus of initial calcification in cartilage and bone. Clin Orthop Rel Res 78: 108–139. Boyce, B.F., Yao, Z., and Xing, L. 2009. Osteoclasts have multiple roles in bone in addition to bone resorption. Crit Rev Eukar Gene Exp 19: 171–180. Chambers, T.J. 1988. The regulation of osteoclastic development and function. In Cell and Molecular Biology of Vertebrate Hard Tissues, Ciba Foundation Symposium 136, Evered, D. and Harnett, S., eds. Wiley, Chichester, UK, pp. 92–107. Clarke, B. 2008. Normal bone anatomy and physiology. Clin J Am Soc Nephro 3(Suppl 3): S131–S139. Cormack, D.H. 1987. Bone. In Ham’s Histology, Cormack, D. H., ed. J.B. Lippincott Co., London, 273 pp.

Skeletal Tissues and Mineralization

51

Currey, J. 1984. The Mechanical Adaptations of Bones. Princeton University Press, Princeton, NJ, 294 pp. Delmas, P.D. 1993. Biochemical markers of bone turnover I: Theoretical considerations and clinical use in osteoporosis. Am J Med 95(Suppl 5A): 11S–16S. Deschaseaux, F., Sensébé, L., and Heymann, D. 2009. Mechanisms of bone repair and regeneration. Trends Mol Med 15: 417–429. Diez-Perez, A., Güerri, R., Nogues, X., et al. 2010. Microindentation for in vivo measurement of bone tissue mechanical properties in humans. J Bone Miner Res 25: 1877–1885. DOI 10.1002/jbmr.73. Donaldson, C.L., Hulley, S.B., Vogel, J.M., et al. 1970. Effect of prolonged bed rest on bone mineral. Metabolism 1: 1071–1084. Evans, F.G. 1973. Mechanical Properties of Bone, Charles C Thomas, Springfield, IL, 322 pp. Fawcett, D.W. 1994. Bone. In Bloom and Fawcett: A Textbook of Histology, Fawcett, D. W., ed. Chapman & Hall, New York, 194 pp. Frost, H.M. 1963. Bone Remodelling Dynamics. Charles C Thomas, Springfield, IL, 175 pp. Fuller, K., Gallagher, A.C., and Chambers, T.J. 1991. Osteoclast resorption-stimulating activity is associated with the osteoblast cell surface and/or the extracellular matrix. Biochem Biophys Res Commun 181: 67–73. Garnero, P. 2009. Bone markers in osteoporosis. Curr Ostoporosis Rep 7: 84–90. Guglielmi, G., and de Terlizzi, F. 2009. Quantitative ultrasound in the assessment of osteoporosis. Eur JRadiology 71: 425–431. Hauschka, P.V., Chen, T.L., and Mavrakos, A.E. 1988. Polypeptide growth factors in bone matrix. In Cell and Molecular Biology of Vertebrate Hard Tissues, Ciba Foundation Symposium 136, Evered, D. and Harnett, S., eds. Wiley, Chichester, UK, 207 pp. Henriksen, K, Neutzky-Wulff, A.V., Bonewald, L.F., et al. 2009. Local communication on and within bone controls bone remodeling. Bone 44: 1026–1033. Henriksen, K., Sørensen, M.G., Jensen, V.K., et al. 2008. Ion transporters involved in acidification of the resorption lacuna in osteoclasts. Calcif Tissue Int 83: 230–242. Iwamoto, J., Sato, Y., Takeda, T., et al. 2009. Effectiveness of exercise in the treatment of lumbar spinal stenosis, knee osteoarthritis, and osteoporosis. Aging Clin Exp Res (Epub ahead of print). Kapinas, K., Kessler, C.B., and Delaney, A.M. 2009. mIR-29 suppression of osteonectin in osteoblasts: Regulation during differentiation and by canonical Wnt signaling. J Cell Biochem 108: 216–224. Krüger, T., Westenfeld, R., Schurgers, L., et al. 2009. Coagulation meets calcification: The vitamin K system. Int J Artif Organs 32: 67–74. Lanyon, L.E. 1992. Control of bone architecture by functional load bearing. J Bone Min Res 7(Suppl 2): S369–S375. Li, J., Miller, M.A., Hutchins, G.D., et al. 2005. Imaging bone microdamage in vivo with positron emission tomography. Bone 37: 819–824. Lüftner, D., Jozereau, D., Schildhauer, S., et al. 2005. PINP as serum marker of metastatic spread to the bone in breast cancer patients. Anticancer Res 25: 1491–1500. Mammone, J.F., and Hudson, S.M. 1993. Micromechanics of bone strength and failure, J Biomechanics 26: 439–446. Manolagas, S.C. and Jilka, R.L. 1995. Bone marrow, cytokines, and bone remodeling. Emerging insights into the pathophysiology of osteoporosis. New Engl J Med 335: 305–311. Martin, T., Gooi, J.H., and Sims, N.A. 2009. Molecular mechanisms coupling bone formation to resorption. Crit Rev Eukaryot Gene Expr 19: 73–88. McSheehy, P.M.J., and Chambers, T.J. 1986. Osteoblastic cells mediate osteoclastic responsiveness to parathyroid hormone. Endocrinology 118: 824–828. Metz, J.A., Anderson, J J.B., and Gallagher, P.N., Jr. 1993. Intakes of calcium, phosphorus, and protein, and physical-activity level are related to radial bone mass in young adult women. Am J Clin Nutr 58: 537–542. Mundy, G.R. 1992. Cytokines and local factors which affect osteoclast function, Int J Cell Cloning 12: 215–222. Nahar, N.N., Missana, L.R., Garimella, R., et al. 2008. Matrix vesicles are carriers of bone morphogenetic proteins (BMPs), vascular endothelial growth factor (VEGF), and noncollagenous matrix proteins. J Bone Miner Metab 26: 514–519. Orimo, H. 2010. The mechanism of mineralization and the role of alkaline phosphatase in health and disease. J Nippon Med Sch 77: 4–12. Ott, S.M. 2008. Histomorphometric measurements of bone turnover, mineralization, and volume. Clin J Am Soc Nephrol 3(Suppl 3): S151–S156. Robling, A.G., and Turner, C.H. 2009. Mechanical signaling for bone modeling and remodeling. Crit Rev Eukar Gene Exp 19: 319–338.

52

Diet, Nutrients, and Bone Health

Roodman, G.D. 1993. Role of cytokines in the regulation of bone resorption, Calcif Tissue Int 53(Suppl 1): S94–S98. Rubin, C.T., and Lanyon, L.E. 1984. Regulation of bone formation by applied dynamic loads. J Bone Jt Surg 66-A: 397–402. Seeman, E. 2009. Bone modeling and remodeling. Crit Rev Eukar Gene Exp 19: 219–233. Skinner, H.C.W. 1979. Bone: Mineralization. In The Scientific Basis of Orthopaedics, Albright, J. A., and Brand, R. A., eds. Appleton & Lange, Norwalk, CT, 199 pp. Stark, Z., and Savarirayan, R. 2009. Osteoporosis. Orphanet J Rare Diseases 4: http://www.ojrd.com/ content/4/1/5. Sugiyama, T., Price, J.S., and Lanyon, L.E. 2009. Functional adaptation to mechanical loading in both cortical and cancellous bone is controlled locally and is confined to the loaded bones. Bone 46: 314–331. Talmage, R.V., Grubb, S.A., and VanderWiel, C.J. 1983. Physiologic processes in bone. In The Musculoskeletal System: Basic Processes and Disorders, Wilson, F. C., ed., 2nd ed., Chapter 8. J.B. Lippincott Co., Philadelphia. Talmage, R.V., and Mobley H.T. 2008. Calcium homeostatis: Reassessment of the actions of parathyroid hormone. Gen Comp Endocrinol 156: 1–8. Talmage, R.V., and Mobley H.T. 2009. The concentration of free calcium in plasma is set by the extracellular action of noncollagenous proteins and hydroxyapatite. Gen Comp Endocrinol 162: 245–250. Talmage, R.V., and Talmage, D.W. 2006. Calcium homeostasis: Solving the solubility problem. J Musculoskelet Neuronal Interact 6: 402–407. Talmage R.V., and Talmage D.W. 2007. Calcium homeostasis: How bone solubility relates to all aspects of bone physiology. J Musculoskelet Neuronal Interact 7: 108–112. Termine, J.D. 1988. Non-collagen proteins in bone. In Cell and Molecular Biology of Vertebrate Hard Tissues, Ciba Foundation Symposium 136, Evered, D. and Harnett, S., eds. Wiley, Chichester, UK, 178 pp. Termine, J.D., Eanes, E.D., and Conn, K.M. 1980. Phosphoprotein modulation of apatite crystallization, Calcif Tissue Int 31: 247–251. Vogel, J.M., Whittle, W.M., Smith, M.C., et al. 1977. Bone mineral measurement, experiment M078. In Biomedical Results from Skylab, Pool, S.L., ed. Administration, National Aeronautics and Space Administration, GPO, Washington, DC, 183 pp. Yavropoulou, M.P., and Yovos, J.G. 2008. Osteoclastogenesis—current knowledge and future perspectives. JMusculoskelet Neuronal Interact 8: 204–216.

5

Optimizing the Skeletal Benefits of Mechanical Loading and Exercise Stuart J. Warden and Robyn K. Fuchs

CONTENTS Introduction....................................................................................................................................... 53 Mechanical Loading Features Influencing Skeletal Adaptation....................................................... 54 Skeletal Adaptation Is Influenced by Load Magnitude............................................................ 54 Skeletal Adaptation Is Enhanced with Novel Mechanical Loading........................................ 55 Skeletal Adaptation Is Influenced by How Fast Loads Are Introduced................................... 56 Skeletal Adaptation Is Greatest with Brief Yet Often Mechanical Loading............................ 56 Skeletal Adaptation in Response to Mechanical Loading Is Site Specific.............................. 56 Implications of Mechanical Loading Features Influencing Skeletal Adaptation..................... 57 Skeletal Benefits of Exercise during Growth.................................................................................... 58 Growth Presents a “Window of Opportunity”......................................................................... 58 Exercise During Growth Optimizes Peak Bone Mass............................................................. 58 Exercise-Induced Optimization of Peak Bone Mass Is Not Maintained Long-Term.............. 59 Conventional Imaging Techniques Do Not Adequately Determine Skeletal Benefits of Exercise........................................................................................................................ 59 Exercise During Growth Encourages Structural Optimization................................................ 62 Exercise-Induced Structural Optimization May Last Lifelong............................................... 62 Contribution of Nutrition to Mechanically Induced Skeletal Adaptation during Growth....... 62 Effects of Exercise on the Aging Skeleton........................................................................................64 Focus Shifts to Maintaining Bone Health and Protecting Skeleton from Excessive Load......64 Exercise May Augment Skeletal Benefits of Pharmaceutical Agents for Osteoporosis.......... 65 Summary...........................................................................................................................................66 Acknowledgments.............................................................................................................................66 References.........................................................................................................................................66

INTRODUCTION Osteoporosis is a prominent and growing problem characterized by a reduction in bone strength and consequent increase in the risk for low-trauma fractures. Key determinants of bone strength and thus fracture risk include the amount of bone material present (quantity), and the spatial distribution (structure) and intrinsic properties (quality) of this material. There is clear evidence for a genetic contribution to these bone properties, with heritability estimates ranging from 60% to 90%, depending on the skeletal property and site assessed (Peaco*ck et al., 2002). The remaining variance in bone properties is accounted for by other factors. 53

54

Diet, Nutrients, and Bone Health Nondominant

Dominant

Control

Thrower

10 mm

FIGURE 5.1  The skeletal effects of mechanical loading and exercise are eloquently observed in individuals playing sports that expose their dominant upper extremity to mechanical overload. The images depict the structural features of the midshaft humerus in the nondominant and dominant upper extremities of a sedentary control individual and throwing (baseball) athlete, as acquired using peripheral quantitative computed tomography. Note the larger bone with thicker cortex in the dominant upper extremity of the thrower when compared with the nondominant upper extremity and the upper extremities in the control individual. (Reprinted from Bone, 45, Warden, S.J., et al., 931–41, 2009, with permission from Elsevier.)

The skeleton’s primary function is mechanical wherein it provides internal support to enable the force of gravity to be countered and presents attachment sites to allow muscle forces to generate motion at specialized bone-to-bone linkages. Given this mechanical role, it follows teleologically that skeletal tissue responds and adapts to its prevailing mechanical environment. This phenomenon is loosely referred to as Wolff’s law, named after the German anatomist/surgeon Julius Wolff who suggested that the form of bone is related to mechanical stress by a mathematical law (Wolff, 1892). Although basic tenets of Wolff’s law contain inaccuracies (Pearson and Lieberman, 2004), the general concept that bone adapts to its mechanical environment is widely accepted. Genetics may impact the magnitude of this adaptation with studies utilizing animal models demonstrating genotype influences on skeletal mechanosensitivity (Robling and Turner, 2002; Robling et al., 2007); however, co-twin studies and studies investigating bone health within individuals who unilaterally overload one extremity have confirmed that mechanical loading in the form of exercise is an important genetic-independent factor influencing skeletal properties (Figure 5.1) (Huddleston et al., 1980; Iuliano-Burns et al., 2005; MacInnis et al., 2003; Warden et al., 2009). This chapter discusses mechanical loading features that influence skeletal adaptation and the response of the skeleton to the mechanical loading associated with exercise during two critical phases of the life span—growth and aging. Included is a discussion of the contribution of nutrition and pharmaceutical agents to mechanically induced skeletal adaptation during growth and aging, respectively.

MECHANICAL LOADING FEATURES INFLUENCING SKELETAL ADAPTATION Skeletal Adaptation Is Influenced by Load Magnitude Animal studies introducing controlled loading to the skeleton have provided a wealth of knowledge regarding mechanical loading features contributing to skeletal adaptation. Initially, these studies

Optimizing the Skeletal Benefits of Mechanical Loading and Exercise

55

Physiological window

Bone gain Homeostasis Minimum effective strain (MES)

Bone loss 50–200 µε

1500–2500 µε Strain

FIGURE 5.2  Adaptation of the skeleton to mechanical loading according to Frost’s mechanostat theory. Frost mechanostat theory predicted mechanical strains to fall between two effective strain levels—the minimum effective strain (MES), speculated to be in the vicinity of 1500 to 2500 µε (Frost, 1987) and a lower effective strain level, suggested to be approximately 50–200 µε. When mechanical strains fell within this “physiological window,” bone resorption during remodeling equaled formation, resulting in mineral homeostasis and no bone adaptation. When mechanical usage caused strain levels to fall outside the window, an imbalance between resorption and formation was predicted. Bone loss (net mineral loss) was predicted for strains below the lower effective strain level (50–200 µε), whereas bone gain (net mineral gain) was predicted for strains above the MES (1500–2500 µε). For extremely high strains, microscopic trauma (microdamage) was predicted.

identified load magnitude as a key factor determining adaptation. When discussing the magnitude of loads applied to the skeleton, the internal strain that the bone experiences is most relevant. Strain refers to the change in length per unit length of a structure and is typically expressed in terms of microstrain (µε) in bone because of its small value. Bone strains during usual activities of daily living range from 400 to 1500 µε, although activities involving high impact loads result in higher strains (Burr et al., 1996). For bone to respond and adapt to an external load, the microstrain engendered within the bone needs to surpass a certain threshold that is greater than what is typically experienced (Rubin and Lanyon, 1985; Turner et al., 1994b). The importance of strain to skeletal mechanoresponsiveness was recognized by the pioneering work of Dr. Harold M. Frost who developed a mathematical model for describing the response of bone to mechanical loading—Frost’s mechanostat theory (Frost, 1987). This theory described a negative feedback control system where bone was maintained such that everyday mechanical strains fell between two effective strain levels—the minimum effective strain (MES), speculated to be in the vicinity of 1500 to 2500 µε, and a lower effective strain level, suggested to be approximately 50–200 µε (Frost, 1987, 1990). The combination of these two strain levels created a “physiological window” wherein mechanical strains within the window resulted in mineral homeostasis, whereas strains below and above the window resulted in net mineral loss and gain, respectively (Figure5.2).

Skeletal Adaptation Is Enhanced with Novel Mechanical Loading Frost’s mechanostat theory provided a quantum advance in understanding bone adaptation to loading, yet it was not without limitations (Martin, 2000; Turner, 1999). In particular, the theory assumed that bone cells were somehow preprogrammed with a MES value which set the threshold for a skeletal response. However, it is now understood that the MES must vary both between and within bones; otherwise, relatively nonloaded bones (i.e., cranium) and sites (i.e., along the neutral axis in a bending bone) would constantly be in states of net mineral loss. Such loss of bone tissue does not occur because bone cell mechanosensitivity is plastic.

56

Diet, Nutrients, and Bone Health

Plasticity forms the foundation of the cellular accommodation theory. The theory agrees that mechanosensitive cells must contain some strain threshold above which a mechanical signal can elicit a cellular response yet argues that the threshold is not a set value, rather is a product of local strain history (Turner, 1999). The cellular accommodation theory assumes that when a strain threshold is surpassed, the mechanosensor cells gradually accommodate to the new state, either by cytoskeletal reorganization or by change of the extracellular microenvironment. Given that bone adaptation is error driven (Lanyon, 1992), the bone response is then proportional to the difference between the new strain and the ever-changing set point, rather than being directly related to the absolute magnitude of the new strain induced by a loading stimulus. Therefore, adaptation to mechanical loading is greatest when strains differ most from usual strains (i.e., when novel loads are introduced). Preliminary evidence to support the cellular accommodation theory has been provided experimentally with bone formation in a mechanical loading study closely resembling the theory’s predicted results, but not those predicted by Frost’s mechanostat theory (Schriefer et al., 2005).

Skeletal Adaptation Is Influenced by How Fast Loads Are Introduced Bone adaptation to a loading stimulus is not dependent on strain magnitude alone, as indicated by the observation that dynamic loading induces significantly greater adaptation than if the same strain magnitudes are held statically (Lanyon and Rubin, 1984; Robling et al., 2001b). The preferential response of bone to dynamic stimuli, combined with its greater adaptation to increased strain magnitude, suggests that a bone’s adaptive response to loading is influenced by how fast strain is introduced. This relationship has been confirmed experimentally with higher strain rates generating greater bone adaptation (Mosley and Lanyon, 1998; Turner et al., 1994a, 1995a). As strain rate is the product of strain magnitude and loading frequency, it fits that increasing either component may contribute to the magnitude of bone adaptation. Frequency refers to the number of loading cycles per second. A positive relationship between loading frequency and cortical bone adaptation exists, with increasing loading frequency beyond a threshold of 0.5 Hz generating progressively greater adaptation (Hsieh and Turner, 2001; Turner et al., 1994a, 1995a; Warden and Turner, 2004).

Skeletal Adaptation Is Greatest with Brief Yet Often Mechanical Loading Further features influencing the skeletal adaptive response to loading include the duration of loading and length of rest between loading bouts. Extending the duration of loading does not necessarily yield proportional increases in bone mass (Rubin and Laynon, 1984; Umemura et al., 1997). As loading duration increases, the bone formation response tends to fade as the mechanosensitive cells accommodate to the prevailing environment. The decline in adaptation with ongoing loading cycles fits a logarithmic relationship such that after only 20 back-to-back loading cycles, bone has lost more than 95% of its mechanosensitivity (Turner and Robling, 2003). These observations indicate that loading programs do not need to be long to induce meaningful adaptation and, in addition, that bone cells need time to resensitize between loading bouts to be responsive to future loading bouts. The amount of rest time required between loading bouts depends on the nature of the loading stimulus. For instance, including a few seconds rest between consecutive loading cycles will result in greater bone adaptation than if the same strain stimulus is introduced with no rests in back-toback cycles (Robling et al., 2001a; Srinivasan et al., 2002). Rests of a number of hours between consecutive loading bouts also result in greater bone adaptation than if the same strain stimulus is introduced in back-to-back bouts (Robling et al., 2000, 2002a, 2002b).

Skeletal Adaptation in Response to Mechanical Loading Is Site Specific As the response of bone to mechanical loading is highly stimulus specific, it follows that its adaptive response is also highly site specific. This site specficity has been confirmed in individuals

57

Optimizing the Skeletal Benefits of Mechanical Loading and Exercise Caudal

–3000 –2000 –1000 0 1000 2000

Compressive surface

(c)

Neutral axis Tensile surface Caudal

Lateral

Lateral

(b)

Microstrain

(a)

Bone geometry at baseline Loading-induced adaptation

FIGURE 5.3  The adaptive response of bone to mechanical loading is site specific such that only those regions within a bone that experience sufficient microstrain will adapt. (a) Schematic diagram of the rodent ulna axial compression model. The distal forelimb is fixed between upper and lower cups. When force is applied to one of the cups, the ulna is caused to bow laterally. (b) The bending of the ulna under axial compression generates medial surface compression and lateral surface tension at the midshaft. There is no strain along the axis through which the bone is bending (neutral axis). (c) Loading of rat ulnas for 16 weeks using the axial compression model causes new bone to be formed on surfaces of high strain (medial and lateral surfaces). There is minimal new bone formation near the neutral axis (caudal and cranial surfaces) where there is the least microstrain during loading. (Data from Robling, A.G., et al., J Bone Miner Res, 17, 1545–54, 2002.)

playing sports that expose their dominant upper extremity to mechanical overload, with only the loaded bones undergoing adaptation (Figure 5.1) (Haapasalo et al., 1996; Huddleston et al., 1980; Jones et al., 1977; Warden et al., 2009). The site-specific nature of bone adaptation to mechanical loading can be localized further than the individual bone level. Long bones are curved such that they bend when axially loaded. Bending results in the exposure of different regions within the bone cross-section to different levels of microstrain. Only those regions within the bone that experience sufficient strain stimulus adapt, as clearly observed in the adaptive response to loading in the rodent ulna axial compression model (Figure 5.3).

Implications of Mechanical Loading Features Influencing Skeletal Adaptation Knowledge of the loading characteristics conducive to skeletal adaptation enables the development of appropriate interventions aimed at optimizing skeletal health (Turner and Robling, 2003; Warden et al., 2004). Specifically, exercises should introduce novel, high-magnitude, rapid strains at the specific sites and in the specific directions that adaptation is desired. Unfortunately, in humans, it is unknown what magnitude of external loading is required to induce a certain strain level or what range of strain magnitudes and rates result in adaptation at specific sites. Although these questions are easily examined in animal models, they are difficult to answer in humans due to difficulties in measuring bone strain in vivo. Laboratory-based studies have attached strain gauges to bone surfaces in humans to assess strains during various activities (Milgrom, 2001); however, measurements have only been performed in a small subgroup of the population and on localized sections of a few bones. It is important to appreciate that bone strains quantified at a specific skeletal location may not correspond to strains engendered at distant sites, in alternate bones, or in the wider population. Finite element models have been developed to model bone strains in response to given loads; however, these models remain in their infancy.

58

Diet, Nutrients, and Bone Health

Despite limitations in assessing bone loading, activities with high magnitude and rapid loading are known to result in higher strain magnitudes and rates. For example, the limited strain gauge data available do show that activities such as jumping and running result in greater bone strain magnitudes and rates than those of lower load and slower loading rate activities such as walking (Milgrom, 2001). Based on clinical and randomized control trial data, it is known that individuals involved in high and rapid load activities (i.e., jumping and running) exhibit greater lower extremity bone adaptation than that of individuals who are not (Guadalupe-Grau et al., 2009). Thus, exercises involving high magnitude and rapid loads are more conducive of skeletal adaptation. Exercises may include endurance activities, such as distance running; however, given the desensitization of the skeleton to prolonged loading and preference of the skeleton for novel short-duration loading, activities should also include jumping and landing activities and rapid changes in direction, such as occur during basketball, volleyball, soccer, and gymnastics. These activities expose the skeleton to short-duration, high-magnitude, rapid loading in multiple novel directions, and all have been associated with significant skeletal adaptation (Nichols et al., 2007).

SKELETAL BENEFITS OF EXERCISE DURING GROWTH Growth Presents a “Window of Opportunity” Growth is an opportune time to take advantage of the skeletal benefits of exercise due to the highly plastic skeletal state presented. Growth is primarily characterized by bone modeling, which involves spatially independent osteoblast-mediated bone formation and osteoclast-mediated bone resorption. These processes function to add or remove bone tissue on previously quiescent bone surfaces to effectively alter bone quantity and structure. Exercise during this period primarily influences the skeleton by stimulating de novo bone formation. Mechanical signals are transmitted to the bone surface via osteocyte signaling whereby they stimulate the differentiation of precursor cells into osteoblasts to transform the periosteal cellular layer from a quiescent to an osteogenic cell layer. As a result, periosteal mineral apposition increases, with formation rates rising largely due to alterations in the amount of bone surface undergoing formation. Observational and longitudinal cohort studies have demonstrated that children and adolescents who lead more physically active lifestyles typically have 10%–15% greater bone mass than that of their peers (Bailey et al., 1999; Janz et al., 2006; Tobias et al., 2007). These data are supported by prospective randomized controlled trials that have demonstrated that weight-bearing exercise in children and adolescents increases bone mass at loaded sites (lower extremities and spine) by up to 5% in <2 years (Hind and Burrows, 2007). The skeletal advantage of exercising during growth has also been eloquently shown in racquet sport players who began training during prepuberty and early puberty. Specifically, girls who began playing before puberty had more than twofold greater differences in bone mass between their playing and nonplaying arms, compared with those who began playing postpuberty (Figure 5.4) (Kannus et al., 1995). This bilateral difference has subsequently been confirmed by others (Bass et al., 2002; duch*er et al., 2009) and supports the presence of a “window of opportunity” during prepuberty and early puberty where the skeleton is most amenable to the influences of mechanical loading and exercise (MacKelvie et al., 2002).

Exercise During Growth Optimizes Peak Bone Mass The skeleton appears most receptive to the benefits of exercise during growth; however, reduced bone strength and the concomitant increase in the risk for low-trauma fractures is predominantly an age-related phenomenon (Warden and Fuchs, 2009). This age-related disconnection raises the question as to whether exercise-induced bone changes during growth persist into adulthood when they would be most advantageous in reducing fracture risk (Warden et al., 2005a). The traditional doctrine is that exercise during growth adds extra mineral to maximize peak bone mass, with the

Optimizing the Skeletal Benefits of Mechanical Loading and Exercise

59

Percent difference (playing vs. nonplaying arm)

25 20 15 10 5 0

Post Pre Peri (<10.5 yrs) (<10.5–18.5 yrs) (>18.5 yrs) Pubertal stage when commenced racquet sports

FIGURE 5.4  The growing skeleton is more responsive to mechanical loading than is the adult skeleton. This study of competitive female racquet sport players showed that those who started playing at an earlier age (several years before menarche [Pre]) had more than two times as much differential (playing vs. nonplaying arm) in mineral accrual than that of players who started playing during their adult years [Post]. (Data from Kannus, P., et al., Ann Intern Med, 123, 27–31, 1995.)

hypothesis being that the accrual of more bone mineral when young provides greater reserves against both the bone loss and compromised bone strength that develops during aging (Rizzoli et al., 2010). This hypothesis is logical considering that (1) approximately 95% of the adult skeleton is formed by the end of adolescence (Bailey et al., 1999; Bonjour et al., 1991); (2) approximately 25%–30% of the adult bone mineral is accrued during the 2–3 years around puberty (Bailey et al., 1999; Slemenda et al., 1994); (3) exercise enhances growth-related bone accrual and peak bone mass (Valimaki etal., 1994); (4) fracture risk during aging doubles for each standard deviation of bone lost from mean peak bone mass (Johnell et al., 2005); and (5) the onset of osteoporosis has been predicted to be delayed by 13 years, with a 10% increase in peak bone mass (Hernandez et al., 2003).

Exercise-Induced Optimization of Peak Bone Mass Is Not Maintained Long-Term Numerous studies have explored the sustainability of exercise-induced bone mass benefits acquired during growth. Follow-up assessments of former participants in randomized controlled trials have documented that cessation of an osteogenic exercise program is associated with short-term maintenance of exercise-induced bone mass benefits (Fuchs and Snow, 2002; Gunter et al., 2008; Kontulainen et al., 2002). However, these benefits do not appear to persist long-term. For instance, Gunter et al. (2008) demonstrated a jumping exercise intervention to increase bone mass by 3.6% in the exercise group compared with controls, but this difference declined by over 60% (to 1.4%) in the 7 years following intervention cessation (Figure 5.5a). The bone mass difference between groups remained statistically significant at 7 years, and the rate of loss of the intervention benefit appeared to decline over time; however, supplementary evidence provided by Karlsson et al. (2000) suggests that the bone mass benefit of exercise during growth ultimately disappears. These investigators demonstrated that exercise in the form of soccer playing conferred high peak bone mass, but its cessation resulted in accelerated bone loss during nonplaying aging (Figure 5.5b).

Conventional Imaging Techniques Do Not Adequately Determine Skeletal Benefits of Exercise The loss of the bone mass benefits induced with exercise during growth suggests that the skeletal changes generated by exercise when young do not persist into adulthood when they may influence

60

Diet, Nutrients, and Bone Health 4

(a) % ∆ Total HIP BMC over controls

3.5 3 2.5 2 1.5 1 0.5

Controls = 0

7

19

43

55

67

79

91

Months from baseline (b)

1.8 1.7

BMD (glcm2)

1.6 1.5 1.4 1.3 1.2 1.1 0 10

20

30

40

50

60

70

80

90

Age (years)

FIGURE 5.5  Exercise-induced optimization of bone quantity is not maintained long-term. (a) Jumping exercise intervention in growing children increased total hip bone mass by 3.6% in the exercise group compared with that in controls, but this difference declined to 1.4% in the 7 years following intervention cessation. (b)Former soccer players (closed circles) achieved a higher lower extremity peak bone mass but had more rapid loss of bone quantity with aging than controls (open circles). (Panel (a) Reproduced from Gunter, K., A.D. Baxter-Jones, R.L. Mirwald, et al. J Bone Miner Res 23:986–93, with permission from American Scciety of Bone and Mineral Research; Panel (b) Reprinted from The Lancet, 355, Karlsson, M.K., et al. Exercise during growth and bone mineral density and fractures in old age, 469–70, 2000, with permission from Elsevier.)

osteoporotic fracture risk. However, this conclusion is based on studies that used dual-energy x-ray absorptiometry (DXA) to assess the sustainability of the skeletal benefits of exercise. DXA is the gold standard in the clinical assessment of bone health and provides a picture of overall bone status, but it has limitations in assessing bone strength and fracture risk. Considerable overlap in DXA-derived measures has been found between people who fracture and those who do not (Stone et al., 2003). Also, DXA-derived measures only explain a fraction of the observed reduction in the risk for fracture associated with osteoporosis drug therapies (Cummings et al., 2002). A prominent reason for the limited usefulness of DXA scans is that bone strength, and the consequent risk for fracture, is dependent upon not only bone quantity but also bone structure. DXA does not provide an adequate measure of bone structure as it is a planar measurement and has low spatial resolution. These features allow DXA to provide a two-dimensional areal analysis of bone; however, this areal analysis can lead to size-related artifacts when compared with a true three-dimensional

Optimizing the Skeletal Benefits of Mechanical Loading and Exercise (a)

1 cm

0.92 cm

0.63 cm

0.5 cm

I

(b)

61

5 cm

II

5 cm

Direction of DXA beam

I

Projected bone image “seen” by DXA

II

Area = 5.0 cm2 aBMD = 0.5 g/cm2

Area = 4.6 cm2 aBMD = 0.54 g/cm2

(c)

I

Area moment of inertia = 320.4 mm4

II

Area moment of inertia = 412.8 mm4

FIGURE 5.6  Schematic illustration of the limitation of dual-energy x-ray absorptiometry (DXA) measures of bone and the influence of structural properties on bone strength. (a) The two bones (I and II) have the same bone mineral content (BMC = 2.5 g), volumetric bone mineral density (BMD = 1063 mg/cm3), and cortical area (0.47 cm2). The only difference between the bones is the distribution of the bone material from the middle of the bone, with the material of bone II being more distant. (b) DXA assessment of the two bones would indicate that the bones have equivalent BMC (2.5 g) and that bone I has a higher areal BMD (aBMD). This results from the fact that bone I has a smaller projected bone area on DXA and DXA-derived BMD is derived as BMC divided by projected bone area. Thus, it may be concluded following assessment of these two bones using DXA that they have equivalent strength (because of their equivalent BMC) or that bone I is stronger (because of its greater aBMD). (c) Bone II is structurally bigger than bone I, as evident by its greater periosteal circumference (3.14 cm vs. 2.89 cm) and total cross-sectional area (0.78 cm2 vs. 0.66 cm 2). Despite both bones having the same cortical bone cross-sectional area and BMC, the bone material in II is distributed further from the neutral axis than in I, as evident by its greater area moment of inertia (412.8 mm4 vs. 320.4mm4). This results in bone II possessing 29% greater resistance to bending than that of bone I purely because of a difference in its structure rather than material properties.

62

Diet, Nutrients, and Bone Health

analysis (Figure 5.6). As DXA does not adequately assess bone structure, it is particularly limited when assessing bone changes, or their maintenance, induced by mechanical loading. Mechanical loading associated with exercise predominantly influences bone structure rather than mass as the way of improving bone strength.

Exercise during Growth Encourages Structural Optimization Exercise during growth adds extra material to loaded sites to effectively increase the quantity of bone present; however, mechanical loading associated with exercise generates disproportionate increases in bone strength. For instance, it has been demonstrated in animal models that very small (<10%) changes in bone mass generated via mechanical loading result in very large (>60%) increases in skeletal mechanical properties (Robling et al., 2002a; Warden et al., 2005b). These gains result from the site-specific deposition of new bone tissue to regions where mechanical demands are greatest. The net result of site-specific deposition of new bone is structural optimization of the skeleton whereby bone material is distributed in such a way that it is better positioned to resist external loads. Typically, this relocation involves new bone being laid down as far as possible from the respective axis of bending or rotation, as observed in clinical trials whereby exercise during growth, especially before puberty, caused new bone to be preferentially laid down on the periosteal (outer) surface of loaded bones (Bass et al., 2002; duch*er et al., 2009; Kontulainen et al., 2003). Such site-specific accrual of new bone is functionally important as it increases bone mass and bone strength where they are needed most while not excessively increasing the overall mass (quantity) of the skeleton.

Exercise-Induced Structural Optimization May Last Lifelong Exercise-induced gains in bone mass appear to decline following exercise cessation; however, mechanisms exist for exercise-induced structural benefits to remain intact until senescence where they may have antifracture benefits. Age-related loss of bone quantity is principally mediated by endocortical and not periosteal surface changes (Figure 5.7a) (Ahlborg et al., 2003). During aging, progressive periosteal bone apposition occurs, but this circumferential gain is unable to sustain bone quantity because of the more rapid loss of endosteal bone, particularly during the menopausal transition. The net result is progressive structural decay and skeletal weakening during aging. As exercise during growth primarily induces periosteal adaptation and aging is not associated with loss of bone from the periosteal surface, the enhanced structure induced by exercise during growth may remain intact and have antifracture properties later in life (Figure 5.7b). The hypothesis for sustainability of the skeletal benefits of exercise is supported by an animal study which found that mechanical-loading-induced changes in bone structure and strength, but not quantity, persist lifelong following the cessation of the loading during growth (Warden et al., 2007). Further support is provided by clinical evidence demonstrating that former athletes >60 year of age have lower risk of fragility fractures than that of matched controls (Nordstrom et al., 2005). These data indicate that loading during growth can have antifracture benefits through induced structural changes, independent of any lasting effects on bone quantity.

Contribution of Nutrition to Mechanically Induced Skeletal Adaptation during Growth Nutrition is a key factor for skeletal development and determines whether the necessary building blocks are available for optimal growth. Thus, it is logical to hypothesize that the anabolic response of the skeleton to mechanical loading depends on the availability of essential nutrients. Anumber of randomized controlled trials in children have confirmed this by demonstrating significant interactions between calcium intake and the skeletal response to exercise (Bass et al., 2007; Specker and Binkley, 2003). In these studies, exercise and calcium supplementation introduced in isolation did

Optimizing the Skeletal Benefits of Mechanical Loading and Exercise

63

(a)

Periosteal bone formation Endocortical bone resorption Growth

Aging

(b)

Exercise-induced periosteal expansion Growth + exercise

Persistence of exercise-induced structural benefits

Aging

FIGURE 5.7  Bone structural changes associated with aging and exercise. (a) Bone loss during aging occurs primarily via bone resorption on the endocortical surface. There is concomitant bone formation on the periosteal surface, which helps to maintain bone structure, but this is insufficient to maintain bone mass. (b)Exercise during growth facilitates periosteal bone formation, which optimizes bone structure. As bone loss during aging occurs from the inside out, the enhanced structure induced by exercise during growth has the potential to remain intact irrespective of age-related changes in bone mass. (Reproduced from Warden, S.J., and Fuchs, R.K. Br J Sports Med 43:885–7, 2009, with permission from BMJ Publishing Group Ltd.)

not influence the skeleton, but when introduced simultaneously, the interventions generated significant skeletal benefits. The bone adaptation suggests that calcium intake needs to be higher than minimally required for growth for the skeletal benefits of exercise to be permitted. The importance of optimal nutrition was confirmed by way of meta-analysis, which concluded that exercise only had a beneficial skeletal effect in the presence of high calcium intakes (>1000 mg/day) (Specker, 1996). The adequacy of dietary calcium intake has been most extensively studied in terms of the permissive role of nutrition on the skeletal effects of exercise; however, the availability of other nutrients may also play a significant role. In particular, sufficient intake of protein has been linked with the skeletal effects of exercise. The relationship between adequate protein and bone has been shown to be independent of energy intake and calcium intake. Chevalley et al. (2008) demonstrated that protein intake higher than the usual recommended dietary allowance enhanced the positive skeletal effects of exercise before the onset of pubertal maturation. The mechanism for this permissive effect of protein is not clear but may relate to insulin-like growth factor, which is crucial for both muscle and bone tissue development.

64

Diet, Nutrients, and Bone Health

EFFECTS OF EXERCISE ON THE AGING SKELETON Focus Shifts to Maintaining Bone Health and Protecting Skeleton from Excessive Load While exercise is best understood in terms of its anabolic effects on bone modeling during growth, it also has a significant role in the mature skeleton. The primary tissue-level process taking place in the mature skeleton is bone remodeling. Remodeling involves the temporally and spatially coordinated actions of osteoclasts and osteoblasts who work in teams to remove and replace discrete, measurable “packets” of bone. During aging, a negative bone balance exists such that more bone is resorbed than is replaced by each team of cells. The result is a progressive net loss of bone mineral and the potential development of osteoporosis. Exercise during aging may limit the loss by reducing the amount of bone resorbed or increasing the amount of bone formed by each team of cells. However, the anabolic effect of exercise during remodeling appears negligible relative to its effects on bone formation during modeling. The limited gain in bone during adulthood has been demonstrated in animal studies which show that the adult skeleton has a lesser anabolic response to mechanical loading than the growing skeleton (Turner etal., 1995b). Instead, exercise during aging appears to exert its primary skeletal effects by reducing bone resorption. The suppression of bone resorption is most clearly evident when exercise levels fall below customary levels, such as during disuse where accelerated remodeling mainly on endosteal and trabecular surfaces contributes to a 0.5%–1% loss of bone mass per month (Pavy-Le Traon etal., 2007). The ability to perform highly osteogenic exercises (i.e., dynamic, high-impact loading) is limited in adults due to these types of loads being associated with a risk for osteoarthritis, bone microdamage, and stress fractures. Consequently, exercise of the adult skeleton typically involves activities with low- or moderate-impact loads, such as walking, running, aerobic exercise, resistance training, and stair climbing. The introduction of low- or moderate-impact loads to the adult skeleton has been met with variable success in modifying bone mass, with maximal increases being around 2%–3%; however, systematic review and meta-analysis of randomized controlled trial evidence have confirmed that low- or moderate-impact loads are effective in reducing age-related bone loss (Bérard et al., 1997; Bonaiuti et al., 2002; Wallace and Cumming, 2000; Wolff et al., 1999). Likewise, maintaining a consistent loading regimen preserves exercise-induced bone mass changes in animals (Shimamura et al., 2002; Wu et al., 2004) and humans (Heinonen et al., 1999) and offsets the loss of bone associated with aging in humans (Martyn-St James and Carroll, 2009). Thus, exercise in the adult skeleton should be encouraged as a means of preventing bone loss and maintaining bone health. In addition to maintaining bone health, overwhelming evidence exists that continued exercise throughout life has significant, bone-mass-independent, antifracture benefits. Reports have consistently shown that persons with a current or previous history of low activity levels have a higher incidence of hip fractures than that of persons with a higher activity level. In the Study of Osteoporotic Fractures trial, which longitudinally followed 9,704 women aged 65 years or older for 4 years, there was a 30% reduction in hip fracture risk associated with frequent walking (Cummings et al., 1995). Likewise, in a separate study, walking for at least 4 hours/week among women who did no other exercise was associated with a 41% lower risk of hip fracture compared with that of women who walked for less than 1 hour/week (Feskanich et al., 2002). The reason for the beneficial antifracture effects of exercise on the older skeleton despite its somewhat limited ability to produce significant changes in bone mass may relate to the multifactorial cause of bone fractures. Although low bone mass has been established as an important predictor of fracture risk, the results of many studies indicate that clinical risk factors related to risk of fall also serve as important predictors of fractures (Cummings et al., 1995; Geusens et al., 2002). Only 5%–10% of falls result in fracture; however, 95% of hip fractures result from falls. By way of its beneficial effects on muscle strength, balance,

Optimizing the Skeletal Benefits of Mechanical Loading and Exercise

65

and proprioception, exercise may reduce the risk for falling and, consequently, the risk for fracture by protecting the aging skeleton from excessive loads.

Exercise May Augment Skeletal Benefits of Pharmaceutical Agents for Osteoporosis Exercise may be used as an adjunct to pharmaceutical agents for preventing and/or treating osteoporosis. This hypothesis stems from preliminary evidence suggesting that exercise combined with either antiresorptive or anabolic pharmaceutical agents produces additive or even synergistic benefits to bone health. Animal studies using ovariectomized rats investigated whether the reduced endosteal and trabecular bone resorption associated with the use of antiresorptive agents combined with the enhanced periosteal bone formation associated with exercise enhanced bone health more than the introduction of either modality alone did (Fuchs et al., 2007; Tamaki et al., 1998). Tamaki et al. (1998) found etidronate and treadmill running to have independent beneficial effects on midshaft and distal femoral bone mineral density (BMD), whereas Fuchs et al. (2007) found independent beneficial effects for alendronate and treadmill running on midshaft femur mechanical properties. These findings indicate that, at a minimum, bisphosphonates and exercise may have additive beneficial effects. However, both studies also found statistical interactions between the modalities for some measures. These latter findings indicate that combined treadmill running and bisphosphonate therapy induced greater benefits than predicted by the summation of their observed individual effects. These results suggest some form of synergy between exercise and antiresorptive pharmaceutical agents wherein one of the interventions magnifies the effect of the other. A number of clinical studies have explored the role of combined exercise and antiresorptive therapy on bone health (Fuchs and Warden, 2008). These studies provide some supportive, albeit inconclusive, evidence for synergistic or at least additive effects between the two modalities. Primarily, Braith et al. (2003, 2007) investigated the combined effects of alendronate and resistance training on glucocorticoid-induced osteoporosis following lung or heart transplant in two separate studies. In both studies, participants treated with combination therapy had superior improvements in BMD than those of people treated with alendronate alone. Although these data suggest an additive or even synergistic beneficial effect of combination therapy, this conclusion could not be substantiated as an exercise-alone treated group was not investigated. Chilibeck et al. (2002) and Uusi-Rasi et al. (2003) addressed this limitation by performing randomized controlled trials designed to elicit both the additive and interactive effects of exercise and antiresorptive therapies on bone health in postmenopausal women. Neither study was able to find additive or interactive effects between the treatments for any skeletal measure; however, the resistance exercise program implemented by Chilibeck etal. (2002) was unable to elicit an independent effect, indicating that it may have been inadequately osteogenic, whereas Uusi-Rasi et al. (2003) observed antiresorptive therapy to increase distal tibial bone mass and jumping exercise to increase distal tibial polar section modulus, suggesting that the two interventions had additive effects in optimizing bone health. While preliminary clinical evidence supporting the combined introduction of exercise and antiresorptive agents for optimizing bone health during aging has been reported, no such evidence currently exists for the combined introduction of exercise and anabolic pharmaceutical agents. However, a number of preclinical cell- and animal-based studies have yielded interesting observations when exploring the combined effects of mechanical loading and parathyroid hormone (PTH) therapy. As both mechanical loading and PTH primarily target bone-forming osteoblasts, their simultaneous introduction may allow one modality to enhance the response of osteoblasts to the effects of the other modality. Such synergy appears to be present with PTH sensitizing the osteogenic response of osteoblasts to mechanical loading, possibly by enhancing the mobilization of intracellular calcium (Ryder and Duncan, 2001). When PTH was introduced prior to or at the commencement of mechanical loading in animal-based studies, synergy between the two modalities occurred such that their individual osteogenic effects were magnified (Kim et al., 2003; Li et al., 2003). These synergistic effects indicate a need for clinical studies investigating the combined effects of PTH

66

Diet, Nutrients, and Bone Health

and exercise. In designing these studies, a priori consideration needs to be given to the timing of the interventions. PTH has a short half-life (75 minutes) and reaches maximal serum concentrations within 15–45 minutes following subcutaneous injection (Lindsay et al., 1993). Exercise should be performed during this period immediately following PTH administration to optimize any synergistic effects between the treatments. Such timing is not as critical when exploiting synergy between bisphosphonates and exercise due to the prolonged skeletal half-life of the former (Khan et al., 1997; Mitchell et al., 1999).

SUMMARY Mechanical loading of the skeleton by way of exercise is an important intervention for the reduction in bone strength and consequent increase in the risk for low-trauma fractures associated with aging. Early in life, especially around puberty, exercise should be promoted to develop a skeleton with the most ideal structural design. The hope is that this design persists long-term to have antifracture benefits later in life. Exercises early in life that may facilitate structural optimization of the skeleton include impact activities introducing high magnitude loads at rapid rates in multiple directions, such as those experienced during participation in basketball and gymnastics. As nutrition is permissive of the skeletal effects of exercise, adequate availability of essential nutrients is necessary to enable structural optimization to occur. The objective of exercise during aging shifts towards preserving bone quantity so as to enter late adulthood with maximal bone stock and protecting the skeleton during late adulthood from excessive loads. Exercises during these phases may include low- or moderate-impact loads, such as walking, running, aerobic exercise, resistance training, and stair climbing. Consideration when prescribing these exercises during aging should be given to their coupling with pharmaceutical agents for preventing and/or treating osteoporosis. By appropriately timing exercise following drug administration, bone health may be optimized above that associated with either intervention in isolation.

ACKNOWLEDGMENTS This contribution was made possible by support from the National Institutes of Health (R01 AR057740 and R15 AR056858 [S.J.W.]; K01 AR054408 [R.K.F.])

REFERENCES Ahlborg, H.G., Johnell, O., Turner, C.H., et al. 2003. Bone loss and bone size after menopause. New Engl JMed 349:327–34. Bailey, D.A., McKay, H.A., Mirwald, R.L., et al. 1999. A six-year longitudinal study of the relationship of physical activity to bone mineral accrual in growing children: The university of Saskatchewan bone mineral accrual study. J Bone Miner Res 14:1672–9. Bass, S.L., Naughton, G., Saxon L., et al. 2007. Exercise and calcium combined results in a greater osteogenic effect than either factor alone: A blinded randomized placebo-controlled trial in boys. J Bone Miner Res 22:458–64. Bass, S.L., Saxon, L., Daly, R.M., et al. 2002. The effect of mechanical loading on the size and shape of bone in pre-, peri-, and postpubertal girls: A study in tennis players. J Bone Miner Res 17:2274–80. Bérard, A., Bravo, G., and Gauthier, P. 1997. Meta-analysis of the effectiveness of physical activity for the prevention of bone loss in postmenopausal women. Osteoporos Int 7:331–7. Bonaiuti, D., Shea, B., Iovine R., et al. 2002. Exercise for preventing and treating osteoporosis in postmenopausal women. Cochrane Database Syst Rev:CD000333. Bonjour, J.P., Theintz, G., Buchs, B., et al. 1991. Critical years and stages of puberty for spinal and femoral bone mass accumulation during adolescence. J Clin Endocrinol Metab 73:555–63. Braith, R.W., Conner, J.A., Fulton, M.N., et al. 2007. Comparison of alendronate vs alendronate plus mechanical loading as prophylaxis for osteoporosis in lung transplant recipients: A pilot study. J Heart Lung Transplant 26:132–7.

Optimizing the Skeletal Benefits of Mechanical Loading and Exercise

67

Braith, R.W., Magyari, P.M., Fulton, M.N., et al. 2003. Resistance exercise training and alendronate reverse glucocorticoid-induced osteoporosis in heart transplant recipients. J Heart Lung Transplant 22:1082–90. Burr, D.B., Milgrom, C., Fyhrie, D., et al. 1996. In vivo measurement of human tibial strains during vigourous activity. Bone 18:405–10. Chevalley, T., Bonjour, J.P., Ferrari, S., et al. 2008. High-protein intake enhances the positive impact of physical activity on BMC in prepubertal boys. J Bone Miner Res 23:131–42. Chilibeck, P.D., Davison, K.S., Whiting, S.J., et al. 2002. The effect of strength training combined with bisphosphonate (etidronate) therapy on bone mineral, lean tissue, and fat mass in postmenopausal women. Can J Physiol Pharmacol 80:941–50. Cummings, S.R., Karpf, D.B., Harris, F., et al. 2002. Improvement in spine bone density and reduction in risk of vertebral fractures during treatment with antiresorptive drugs. Am J Med 112:281–9. Cummings, S.R., Nevitt, M.C., Browner, W.S., et al. 1995. Risk factors for hip fractures in white women. NEngl J Med 332:767–73. duch*er, G., Daly, R., and Bass, S. 2009. The effects of repetitive loading on bone mass and geometry in young male tennis players: A quantitative study using magnetic resonance imaging. J Bone Miner Res 24:1686–92. Feskanich, D., Willett, W., and Colditz, G. 2002. Walking and leisure-time activity and risk of hip fracture in postmenopausal women. JAMA 288:2300–6. Frost, H.M. 1987. Bone “mass” and the “mechanostat”: A proposal. Anat Rec 219:1–9. Frost, H.M. 1990. Skeletal structural adaptations to mechanical usage (SATMU): 2. Redefining Wolff’s law: The bone remodeling problem. Anat Rec 226:414–22. Fuchs, R.K., Shea, M., Durski, S.L., et al. 2007. Individual and combined effects of exercise and alendronate on bone mass and strength in ovariectomized rats. Bone 41:290–6. Fuchs, R.K., and Snow, C.M.. 2002. Gains in hip bone mass from high-impact training are maintained: A randomized controlled trial in children. J Pediatr 141:357–62. Fuchs, R.K., and Warden, S.J. 2008. Combination therapy using exercise and pharmaceutical agents to optimize bone health. Clinic Rev Bone Miner Metab 6:37–45. Geusens, P., Autier, P., Boonen, S., et al. 2002. The relationship among history of falls, osteoporosis, and fractures in postmenopausal women. Arch Phys Med Rehabil 83:903–6. Guadalupe-Grau, A., Fuentes, T., Guerra, B., et al. 2009. Exercise and bone mass in adults. Sports Med 39:439–68. Gunter, K., A. D. Baxter-Jones, R. L. Mirwald, et al. 2008. Impact exercise increases BMC during growth: An 8-year longitudinal study. J Bone Miner Res 23:986–93. Haapasalo, H., Sievänen, H., Kannus, P., et al. 1996. Dimensions and estimated mechanical characteristics of the humerus after long-term tennis loading. J Bone Miner Res 11:864–72. Heinonen, A., Kannus, P., Sievanen, H., et al. 1999. Good maintenance of high-impact activity-induced bone gain by voluntary, unsupervised exercises: An 8-month follow-up of a randomized controlled trial. J Bone Miner Res 14:125–8. Hernandez, C.J., Beaupre, G.S., and Carter, D.R. 2003. A theoretical analysis of the relative influences of peak BMD, age-related bone loss and menopause on the development of osteoporosis. Osteoporos Int 14:843–7. Hind, K., and Burrows, M. 2007. Weight-bearing exercise and bone mineral accrual in children and adolescents: A review of controlled trials. Bone 40:14–27. Hsieh, Y.F., and Turner, C.H. 2001. Effects of loading frequency on mechanically induced bone formation. JBone Miner Res 16:918–24. Huddleston, A. L., Rockwell, D., Kulund, D.N., et al. 1980. Bone mass in lifetime tennis athletes. JAMA 244:1107–9. Iuliano-Burns, S., Stone, J., Hopper, J.L., et al. 2005. Diet and exercise during growth have site-specific skeletal effects: A co-twin control study. Osteoporos Int 16:1225–32. Janz, K.F., Gilmore, J.M., Burns, T.L., et al. 2006. Physical activity augments bone mineral accrual in young children: The Iowa Bone Development study. J Pediatr 148:793–9. Johnell, O., Kanis, J.A., Oden, A., et al. 2005. Predictive value of BMD for hip and other fractures. J Bone Miner Res 20:1185–94. Jones, H., Priest, J., Hayes, W., et al. 1977. Humeral hypertrophy in response to exercise. J Bone Joint Surg Am 59:204–7. Kannus, P., Haapasalo, H., Sankelo, M., et al. 1995. Effect of starting age of physical activity on bone mass in the dominant arm of tennis and squash players. Ann Intern Med 123:27–31. Karlsson, M.K., Linden, C., Karlsson, C., et al. 2000. Exercise during growth and bone mineral density and fractures in old age. Lancet 355:469–70.

68

Diet, Nutrients, and Bone Health

Khan, S.A., Kanis, J.A., Vasikaran, S., et al. 1997. Elimination and biochemical responses to intravenous alendronate in postmenopausal osteoporosis. J Bone Miner Res 12:1700–7. Kim, C.H., Takai, E., Zhou, H., et al. 2003. Trabecular bone response to mechanical and parathyroid hormone stimulation: The role of mechanical microenvironment. J Bone Miner Res 18:2116–25. Kontulainen, S.A., Kannus, P.A., Pasanen, M.E., et al. 2002. Does previous participation in high-impact training result in residual bone gain in growing girls? One year follow-up of a 9-month jumping intervention. Int J Sports Med 23:575–81. Kontulainen, S., Sievanen, H., Kannus, P., et al. 2003. Effect of long-term impact-loading on mass, size, and estimated strength of humerus and radius of female racquet-sports players: A peripheral quantitative computed tomography study between young and old starters and controls. J Bone Miner Res 18:352–9. Lanyon, L.E. 1992. Control of bone architecture by functional load bearing. J Bone Miner Res 7(Suppl. 2):S369–S375. Lanyon, L.E., and Rubin, C.T. 1984. Static vs dynamic loads as an influence on bone remodelling. J Biomech 17:897–905. Li, J., Duncan, R.L., Burr, D.B., et al. 2003. Parathyroid hormone enhances mechanically induced bone formation, possibly involving L-type voltage-sensitive calcium channels. Endocrinology 144:1226–33. Lindsay, R., Nieves, J., Henneman, J., et al. 1993. Subcutaneous administration of the amino-terminal fragment of human parathyroid hormone-(1–34): Kinetics and biochemical response in estrogenized osteoporotic patients. J Clin Endocrinol Metab 77:1535–9. MacInnis, R.J., C. Cassar, C. A. Nowson, et al. 2003. Determinants of bone density in 30- to 65-year-old women: A co-twin study. J Bone Miner Res 18:1650–6. MacKelvie, K.J., K. M. Khan, and H. A. McKay. 2002. Is there a critical period for bone response to weightbearing exercise in children and adolescents? A systematic review. Br J Sports Med 36:250–7. Martin, R.B. 2000. Toward a unifying theory of bone remodelling. Bone 26:1–6. Martyn-St James, M., and Carroll, S. 2009. A meta-analysis of impact exercise on postmenopausal bone loss: The case for mixed loading exercise programmes. Br J Sports Med 43:898–908. Milgrom, C. 2001. The role of strain and strain rates in stress fractures. In Musculoskeletal Fatigue and Stress Fractures, edited by D. B. Burr and C. Milgrom, 119–29. Bota Raton, Florida: CRC Press. Mitchell, D.Y., Heise, M.A., Pallone, K.A., et al. 1999. The effect of dosing regimen on the pharmaco*kinetics of risedronate. Br J Clin Pharmacol 48:536–42. Mosley, J.R., and Lanyon, L.E. 1998. Strain rate as a controlling influence on adaptive modeling in response to dynamic loading of the ulna in growing male rats. Bone 23:313–8. Nichols, D.L., Sanborn, C.F., and Essery, E.V. 2007. Bone density and young athletic women. An update. Sports Med 37:1001–14. Nordstrom, A., Karlsson, C., Nyquist, F., et al. 2005. Bone loss and fracture risk after reduced physical activity. J Bone Miner Res 20:202–7. Pavy-Le Traon, A., Heer, M., Narici, M.V., et al. 2007. From space to Earth: Advances in human physiology from 20 years of bed rest studies (1986–2006). Eur J Appl Physiol 101:143–94. Peaco*ck, M., Turner, C.H., Econs, M.J., et al. 2002. Genetics of osteoporosis. Endocr Rev 23:303–26. Pearson, O.M., and Lieberman, D.E. 2004. The aging of Wolff’s “law”: Ontogeny and responses to mechanical loading in cortical bone. Am J Phys Anthropol Suppl 39:63–99. Rizzoli, R., Bianchi, M.L., Garabedian, M., et al. 2010. Maximizing bone mineral mass gain during growth for the prevention of fractures in the adolescents and the elderly. Bone 46:294–305. Robling, A.G., Burr, D.B., and Turner, C.H. 2000. Partitioning a daily mechanical stimulus into discrete loading bouts improves the osteogenic response to loading. J Bone Miner Res 15:1596–1602. Robling, A.G., Burr, D.B., and Turner, C.H.. 2001a. Recovery periods restore mechanosensitivity to dynamically loaded bone. J Exp Biol 204:3389–99. Robling, A.G., Duijvelaar, K.M., Geevers, J.V., et al. 2001b. Modulation of appositional and longitudinal bone growth in the rat ulna by applied static and dynamic force. Bone 29:105–13. Robling, A.G., Hinant, F.M., Burr, D.B., et al. 2002a. Improved bone structure and strength after long-term mechanical loading is greatest if loading is separated into short bouts. J Bone Miner Res 17:1545–54. Robling, A.G., Hinant, F.M., Burr, D.B., et al. 2002b. Shorter, more frequent mechanical loading sessions enhance bone mass. Med Sci Sports Exerc 34:196–202. Robling, A.G., and Turner, C.H. 2002. Mechanotransduction in bone: Genetic effects on mechanosensitivity in mice. Bone 31:562–9. Robling, A.G., Warden, S.J., Shultz, K.L., et al. 2007. Genetic effects on bone mechanotransduction in congenic mice harboring bone size and strength quantitative trait loci. J Bone Miner Res 22:984–91.

Optimizing the Skeletal Benefits of Mechanical Loading and Exercise

69

Rubin, C.T., and Laynon, L.E. 1984. Regulation of bone formation by applied dynamic loads. J Bone Joint Surg Am 66:397–402. Rubin, C.T., and Lanyon, L.E. 1985. Regulation of bone mass by mechanical strain magnitude. Calcif Tissue Int 37:411–7. Ryder, K.D., and Duncan, R.L. 2001. Parathyroid hormone enhances fluid shear-induced [Ca2+]i signaling in osteoblastic cells through activation of mechanosensitive and voltage-sensitive Ca2+ channels. J Bone Miner Res 16:240–8. Schriefer, J.L., Warden, S.L., Saxon, L.K., et al. 2005. Cellular accomodation and the response of bone to mechanical loading. J Biomech 38:1838–45. Shimamura, C., Iwamoto, J., Takeda, T., et al. 2002. Effect of decreased physical activity on bone mass in exercise-trained young rats. J Orthop Sci 7:358–63. Slemenda, C.W., Reister, T.K., Hui, S.L., et al. 1994. Influences on skeletal mineralization in children and adolescents: Evidence for varying effects of sexual maturation and physical activity. J Pediatr 125:201–7. Specker, B., and Binkley, T. 2003. Randomized trial of physical activity and calcium supplementation on bone mineral content in 3- to 5-year-old children. J Bone Miner Res 18:885–92. Specker, B.L. 1996. Evidence for an interaction between calcium intake and physical activity on changes in bone mineral density. J Bone Miner Res 11:1539–44. Srinivasan, S., Weimer, D.A., Agans, S.C., et al. 2002. Low-magnitude mechanical loading becomes osteogenic when rest is inserted between each load cycle. J Bone Miner Res 17:1613–20. Stone, K.L., Seeley, D.G., Lui, L.Y., et al. 2003. BMD at multiple sites and risk of fracture of multiple types: Long-term results from the Study of Osteoporotic Fractures. J Bone Miner Res 18:1947–54. Tamaki, H., Akamine, T., Goshi, N., et al. 1998. Effects of exercise training and etidronate treatment on bone mineral density and trabecular bone in ovariectomized rats. Bone 23:147–53. Tobias, J.H., Steer, C.D., Mattocks, C.G., et al. 2007. Habitual levels of physical activity influence bone mass in 11-year-old children from the United Kingdom: Findings from a large population-based cohort. J Bone Miner Res 22:101–9. Turner, C.H. 1999. Toward a mathematical description of bone biology: The principal of cellular accommodation. Calcif Tissue Int 65:466–71. Turner, C.H., Forwood, M.R., and Otter, M.W. 1994a. Mechanotransduction in bone: Do bone cells act as sensors of fluid flow? FASEB J 8:875–8. Turner, C.H., Forwood, M.R., Rho, J.-Y., et al. 1994b. Mechanical loading thresholds for lamellar and woven bone formation. J Bone Miner Res 9:87–97. Turner, C.H., Owan, I., and Takano, Y. 1995a. Mechanotransduction in bone: Role of strain rate. Am J Physiol 269:E438–42. Turner, C.H., and Robling, A.G. 2003. Designing exercise regimens to increase bone strength. Exerc Sport Sci Rev 31:45–50. Turner, C.H., Takano, Y., and Owan, I. 1995b. Aging changes mechanical loading thresholds for bone formation in rats. J Bone Miner Res 10:1544–9. Umemura, Y., Ishiko, T., Yamauchi, T., M, et al. 1997. Five jumps per day increase bone mass and breaking force in rats. J Bone Miner Res 12:1480–5. Uusi-Rasi, K., Kannus, P., Cheng, S., et al. 2003. Effect of alendronate and exercise on bone and physical performance of postmenopausal women: A randomized controlled trial. Bone 33:132–43. Valimaki, M.J., Karkainen, M., Lamberg-Allardt, C., et al. 1994. Exercise, smoking and calcium intake during adolescence and early adulthood as determinants of peak bone mass. BMJ 309:230–5. Wallace, B.A., and Cumming, R.G. 2000. Systematic review of randomized trials of the effect of exercise on bone mass in pre- and postmenopausal women. Calcif Tissue Int 67:10–8. Warden, S.J., Bogenschutz, E.D., Smith, H.D., et al. 2009. Throwing induces substantial torsional adaptation within the midshaft humerus of male baseball players. Bone 45:931–41. Warden, S.J., and Fuchs, R.K. 2009. Exercise and bone health: Optimising bone structure during growth is key, but all is not in vain during ageing. Br J Sports Med 43:885–7. Warden, S.J., Fuchs, R.K., Castillo, A.B., et al. 2007. Exercise when young provides lifelong benefits to bone structure and strength. J Bone Miner Res 22:251–9. Warden, S.J., Fuchs, R.K., Castillo, A.B., et al. 2005a. Does exercise during growth influence osteoporotic fracture risk later in life? J Musculoskelet Neuronal Interact 5:344–6. Warden, S.J., Fuchs, R.K., and Turner, C.H. 2004. Steps for targeting exercise towards the skeleton to increase bone strength. Eura Medicophys 40:223–32. Warden, S.J., Hurst, J.A., Sanders, M.S., et al. 2005b. Bone adaptation to a mechanical loading program significantly increases skeletal fatigue resistance. J Bone Miner Res 20:809–16.

70

Diet, Nutrients, and Bone Health

Warden, S.J., and Turner, C.H. 2004. Mechanotransduction in cortical bone is most efficient at loading frequencies of 5–10 Hz. Bone 34:261–70. Wolff, I., van Croonenberg, J.J., Kemper, H.C.G., et al. 1999. The effect of exercise training programs on bone mass: A meta-analysis of published controlled trials in pre- and postmenopausal wmoen. Osteoporos Int 9:1–12. Wolff, J. 1892. Das Gesetz der Transformation der Knochen. Bei Hirschwald, Leipzig. Wu, J., Wang, X.X., Higuchi, M., et al. 2004. High bone mass gained by exercise in the growing male mice is increased by subsequent reduced exercise. J Appl Physiol 97:806–10.

6

Hormone Actions in the Regulation of Calcium and Phosphorus Metabolism David A. Ontjes

“The constancy of the internal environment is the condition for free and independent life.” Claude Bernard, in “Lectures on the Phenomena of Life Common to Animals and Plants” (1878)

CONTENTS Body Calcium and Phosphorus: Homeostasis and Balance.............................................................. 72 Need for Calcium Homeostasis............................................................................................... 72 Distribution of Body Calcium and Phosphorus....................................................................... 73 Organs Important in Regulating Calcium and Phosphorus Balance........................................ 73 Gastrointestinal Tract................................................................................................... 73 Kidney�������������������������������������������������������������������������������������������������������������������������75 Bone���������������������������������������������������������������������������������������������������������������������������� 76 Overall Requirements for Positive Calcium Balance.............................................................. 77 PTH–Vitamin D Endocrine System and Calcium Homeostasis....................................................... 77 PTH�����������������������������������������������������������������������������������������������������������������������������������������77 Structure and Biosynthesis of PTH.............................................................................. 78 Actions of PTH........................................................................................................................ 79 PTH Receptors and Intracellular Messengers..............................................................80 Actions of PTH on Bone.............................................................................................. 80 Actions of PTH on Kidney.......................................................................................... 81 Basis of Use of PTH as Antiosteoporosis Drug........................................................... 81 Control of PTH Secretion........................................................................................................ 81 Role of Calcium Sensing Receptor in Regulating PTH Secretion.............................. 82 Role of Calcitriol in Regulating PTH Secretion.......................................................... 82 Other Ions Affecting PTH Secretion............................................................................ 83 Drugs Affecting PTH Secretion................................................................................... 83 Calcimimetics..............................................................................................................84 Diseases Caused by Abnormal PTH Secretion or Action........................................................84 Primary Hyperparathyroidism.....................................................................................84 Hypercalcemia of Malignancy..................................................................................... 85 Congenital or Acquired Abnormalities of CaSR......................................................... 85 Hypoparathyroidism and Pseudohypoparathyroidism................................................. 86 71

72

Diet, Nutrients, and Bone Health

Steroid Hormones and Bone Metabolism................................................................................................. 86 General Features of Steroid Hormone Action................................................................................. 86 Estrogens......................................................................................................................................... 87 Sources of Estrogens in Women and Men.......................................................................... 87 Effects of Estrogen on Bone Metabolism and Calcium Balance........................................ 88 Effects of Estrogen Administration on Osteoporosis.......................................................... 89 Selective Estrogen Receptor Modulators........................................................................................ 91 General Properties of Selective Estrogen Receptor Modulators........................................ 91 Biological Effects of Raloxifene........................................................................................ 91 Effects of SERMs in Treatment of Osteoporosis................................................................ 92 Androgens....................................................................................................................................... 92 Sources in Men and Women............................................................................................... 92 Effects of Testosterone on Bone Metabolism and Calcium Balance.................................. 93 Effects of Androgen on Osteoporosis in Men..................................................................... 93 Glucocorticoids............................................................................................................................... 95 Sources of Glucocorticoids and Causes of Glucocorticoid Excess.................................... 96 Effects of Glucocorticoid Excess on Bone Metabolism and Calcium Balance.................. 96 Glucocorticoid-Induced Osteoporosis................................................................................ 98 Fracture Risk Reduction in Glucocorticoid-Induced Osteoporosis.................................... 98 Thyroid Hormones and Bone Metabolism.................................................................................... 100 Introduction...................................................................................................................... 100 Actions on Skeletal Tissue.................................................................................................101 Effects of Thyroid Hormone Excess on Bone Health........................................................101 Growth Hormone and IGF-I.......................................................................................................... 102 Introduction...................................................................................................................... 102 Effects of GH and IGF-I on Bone Metabolism................................................................ 103 Therapy for Osteoporosis................................................................................................. 104 Summary and Conclusions..................................................................................................................... 105 References............................................................................................................................................... 106

BODY CALCIUM AND PHOSPHORUS: HOMEOSTASIS AND BALANCE This chapter discusses the role of the endocrine system in regulating calcium homeostasis and calcium and phosphate balance. Hormones are the primary agents responsible for sustaining a number of essential conditions within our internal environment. These include plasma concentrations of glucose, amino acids, sodium, potassium, and not least, calcium ions. In this chapter, calcium homeostasis refers to the processes by which a constant concentration of ionized calcium is maintained in the extracellular fluid. The distribution of calcium between bone and other body compartments is in a dynamic equilibrium and is closely regulated by several hormones, the most important of which are parathyroid hormone (PTH) and calcitriol. Calcium balance refers to the processes leading to the gain or loss of all calcium from the entire body pool. Because most of the calcium pool resides in bone, calcium balance is synonymous with the net gain or loss of mineralized bone over time. In the hierarchy of conditions essential for survival, a constant concentration of ionized calcium commonly takes precedence over maintenance of an optimal bone mass. Some of the circ*mstances in which bone mass is sacrificed to maintain calcium homeostasis are discussed later in this chapter.

Need for Calcium Homeostasis A constant concentration of ionized calcium in the extracellular fluid is essential for several vital processes including the stability and permeability of plasma membranes. Consequently, the normal concentration of ionized calcium measured in plasma is narrow, ranging from 1.12 to 1.23 mmol/L

Hormone Actions in the Regulation of Calcium and Phosphorus Metabolism

73

(4.48 to 4.92 mg/dL). A reduction in the extracellular calcium concentration increases the permeability of cell membranes to sodium and increases the excitability of muscle and nerve tissues, causing nerve discharge and muscle contraction. The resulting clinical manifestation in humans with hypocalcemia is tetany. Extracellular calcium ions can bind to a G-protein-linked calcium-sensing receptor (CaSR) in parathyroid, renal epithelial, and other cells to control cellular function directly in these tissues. In addition to its role in the normal mineralization of bone matrix, extracellular calcium also activates numerous extracellular enzymes and clotting factors. Intracellular calcium is sequestered within mitochondria, endoplasmic reticulum, or sarcoplasmic reticulum, where its release can be stimulated by the activation of a number of cell-surface receptors by several hormones and neurotransmitters. In cells, calcium ions act as a second messenger by interacting with a large number of key enzymes and other proteins governing muscle contraction, microtubule and microfilament assembly, and membrane permeability.

Distribution of Body Calcium and Phosphorus The body of an average adult contains approximately 1 kg of calcium, 99% of which is deposited in bone as hydroxyapatite. About 1% of the calcium in bone is rapidly exchangeable with that in the extracellular fluid. The remainder may be mobilized more slowly when conditions require. The overall distribution is shown in Table 6.1. Approximately 85% of total body phosphorus resides in bone in the form of hydroxyapatite and 15% in soft tissues. In contrast to calcium, extraskeletal phosphorus is located primarily within cells where it may serve as a component of numerous complex organic molecules, such as nucleic acids and phospholipids, or in free ionic form as HPO42− or H2PO4−. The concentration of extracellular phosphate ions is influenced by many of the same hormones that regulate extracellular calcium ions. However, extracellular phosphorus concentration is much less tightly regulated than that of calcium, with normal concentrations ranging from 0.87 to 1.45 mmol/L (2.7 to 4.5 mg/dL).

Organs Important in Regulating Calcium and Phosphorus Balance Gastrointestinal Tract Gastrointestinal Calcium Absorption The gastrointestinal (GI) tract governs entry of both calcium and phosphorus into the body and plays an important role in overall calcium balance. Calcium is absorbed in its ionized form, mainly

TABLE 6.1 Distribution of Calcium in Body Tissues In bone   Non-labile hydroxyapatite   Labile skeletal calcium pool In extracellular fluid   Plasma In cells

1000 g (99%)   999 g   1g 1 g (0.1%)  0.2 g (48% ionized; 46% protein bound; 6% complexed) 1 g (0.1%)

Source: Bringhurst, F.R., and Leder, B.Z. 2006. Regulation of calcium and phosphate homeostasis. In Textbook of Endocrinology, DeGroot, L.J., and Jameson, J.L, eds., 5th ed., Chapter 74. Elsevier Saunders, Philadelphia.

74

Diet, Nutrients, and Bone Health

in the duodenum and jejunum. Insoluble salts of calcium are not well absorbed. Absorption occurs by two processes, one that is active and saturable and a second that is passive and nonsaturable. The saturable process is mediated by calcium-binding proteins existing in the duodenum and upper jejunum. One of these binding proteins, known as TRPV6, exists in the highest concentration on the apical cell surface in intestinal endothelial cells (den Dekker et al., 2003). The expression of the TRPV6 gene is stimulated by 1,25(OH)2D3 or calcitriol, acting through the vitamin D receptor. The TRPV6 protein appears to facilitate the movement of calcium ions through the apical brush border and across the endothelial cell membrane by formation of a selective calcium channel. Once calcium has entered the intestinal endothelial cell, it must be transported through the cytoplasm and moved out of the cell into the extracellular compartment against an electrochemical gradient. Transport across the cytoplasm is carried out by a cytosolic protein known as calbindin D9K. Once calcium ions reach the basolateral membrane, they dissociate from calbindin and are actively extruded out of the cell by high-affinity membrane Ca2+-ATPases to diffuse into the extracellular fluid and enter the portal circulation (Bronner et al., 1986). The expression of the calbindin gene and the activity of membrane Ca2+-ATPases are also governed by the vitamin D receptor in intestinal endothelial cells. Thus, calcitriol is the primary regulator of the active, saturable component of intestinal calcium transport (Bringhurst and Leder, 2006). This pathway accounts for most of the calcium transport occurring when calcium ion concentrations within the intestinal lumen are low. The efficiency of the active pathway can be regulated up or down, depending on the availability of dietary calcium and the prevailing need to maintain calcium homeostasis. The jejunum and ileum also provide for a passive, nonsaturable movement of calcium across the intestinal lumen through paracellular channels (Karbach, 1992). This mechanism predominates in delivering calcium from the gut to the systemic circulation under conditions where the intraluminal concentration of calcium is high, as for example in subjects taking oral calcium supplements. Vitamin D may also play a role in enhancing this process as well, although the mechanism is unclear. At the same time that absorption of calcium is occurring in the upper GI tract, there is a movement of calcium into the intestinal lumen in the form of bile and other digestive juices. These endogenous intestinal secretions normally contain 100 to 200 mg of calcium per day and are little affected by dietary or serum calcium. Net absorption represents the difference between active and passive calcium absorption and endogenous intestinal secretions. The efficiency of calcium absorption can vary greatly from 5% to 70% under different conditions in the same individual. The relationships between dietary calcium intake and absorption are illustrated in Figure 6.1. Under conditions of dietary calcium restriction and with adequate supplies of vitamin D, active absorption is maximized. At higher calcium intakes, passive absorption plays an increasing role. As the calcium supply increases, the overall efficiency of calcium absorption decreases, whereas the total amount of calcium absorbed continues to increase, but at a slower rate. At dietary calcium intakes below approximately 200 mg/day, the obligatory excretion of calcium from endogenous intestinal secretions exceeds the absorption of calcium from the upper GI tract so that the gut actually becomes a source of net calcium loss. Gastrointestinal Phosphorus Absorption The average dietary intake of phosphorus is 800–900 mg/day, usually exceeding the minimum requirement of 400 mg/day. Normally, about 70% of dietary phosphorus is absorbed in the duodenum and proximal jejunum. In the jejunum, a saturable, active, sodium-dependent process is responsive to vitamin D and a nonsaturable process is thought to represent passive paracellular transport. The active process is mediated by a sodium phosphate transporter in the luminal brush border of intestinal epithelial cells. Calcitriol stimulates the active process by increasing the expression of the transporter. Thus, both calcium and phosphorus absorption are stimulated by calcitriol. In contrast to calcium absorption, however, the basal absorption of phosphate in the absence of vitamin D is much higher, suggesting that the passive paracellular mechanism plays a more dominant role (Cross et al., 1990).

Hormone Actions in the Regulation of Calcium and Phosphorus Metabolism

75

600 500 400 Net absorption (mg/day)

300 200 100 0 –100 –200

200

400

600

800

1000

1200

Dietary calcium (mg/day)

FIGURE 6.1  Model relationship of dietary calcium intake and net gastrointestinal absorption of calcium in a healthy young individual. Note that the efficiency of absorption declines as dietary calcium intake increases. At dietary calcium intakes of less than ~200 mg/day, net absorption becomes negative due to fecal loss of calcium from intestinal secretions.

Kidney Renal Calcium Excretion At normal rates of glomerular filtration, 7000 to 10,000 mg of ionized calcium is delivered to the proximal renal tubules each day, yet only 100–300 mg is ultimately excreted in the urine. Approximately 98% of the filtered calcium load is normally reabsorbed at various sites along the nephron. About 60% to 70% is absorbed in the proximal tubules, and another 20% to 25% is absorbed in the loop of Henle, mainly by passive paracellular diffusion. In contrast, the 8% to 10% absorbed in the distal tubule is hormonally controlled and involves transcellular transport mechanisms. Epithelial cells in the distal tubules possess transport proteins similar to those seen in the intestinal epithelium where calcium is also actively absorbed. These proteins include an epithelial calcium channel, TPRV5, which is closely hom*ologous to TPRV6 in the gut, a calcium-binding protein, calbindin D, basolateral calcium ATPases, and a basolateral Na+/Ca++ exchanger. Similar to the process in the gut, calcium penetrates into the tubular epithelial cell through a TPRV5 calcium channel, moves across the cytoplasm bound to a calbindin protein, and is finally extruded against a gradient by the Na+/Ca++ exchanger and calcium ATPase located in the basolateral membrane (Hoenderop et al., 2002; Loffing and Kaissling, 2003). The regulation of calcium reabsorption in the nephron is a complex process involving both hormones and divalent cations. PTH is the main regulator of calcium absorption in the distal nephron. It increases calcium absorption in the distal tubules through several mechanisms including activating the TRPV5 channel, increasing calbindin expression, and increasing the affinity for calcium of basolateral Ca2+-ATPases (Hoenderop et al., 2002). Calcitriol augments the action of PTH, apparently by further increasing the expression of TRPV5, calbindin, and plasma-membrane-associated Ca2+-ATPases (Hoenderop et al., 2001). Calcium ion itself plays a role in the control of calcium reabsorption in those portions of the nephron possessing a CaSR. The ascending limb of the loop on Henle possesses CaSRs that are activated by the binding of calcium ions present in the tubular fluid. At this site in the nephron, activation of the CaSR results in a reduction of calcium and sodium transport and a decrease in urinary concentrating ability. Thus, a rise in serum Ca2+ concentration tends to be buffered by increased renal excretion of calcium and a more dilute urine (Tfelt-Hansen and Brown, 2005). This self-regulatory process can occur independently of PTH and calcitriol. Inactivating mutations in the CaSR tend to

76

Diet, Nutrients, and Bone Health

cause hypercalcemia due in part to impaired renal clearance of calcium in a syndrome known as familial hypocalciuric hypercalcemia (FHH) (Brown, 2007). The CaSR is also present in bone and parathyroid cells, where it plays an important regulatory role, as discussed later in this chapter. Renal Phosphate Excretion Eighty-five percent of the phosphorus present in the plasma is in the form of free ions, HPO42− and H2PO4− or complexed with sodium as NaHPO4 − and is thus ultrafiltrable at the renal glomerulus. The control of serum phosphate is accomplished mainly by alterations in the renal reabsorption of phosphate filtered at the glomerulus. At normal glomerular filtration rates, the filtered load of phosphate is 4,000 to 6,000 mg/day. With varying dietary phosphate intake, the reabsorption of phosphate from the glomerular filtrate can vary widely, from 70% to 95%. The fractional excretion of phosphate in the urine, determined by the ratio of phosphate to creatinine clearance, typically ranges from 10% to 15%. Most of the filtered phosphate is reabsorbed by the proximal tubule in an active process requiring sodium ions. At least three separate Na–P cotransporters have been identified at various locations on the nephron. The type II Na–P cotransporter accounts for most of the transport in the proximal tubule and is involved in the regulation of phosphate transport by PTH. A substantial part of the regulation of the renal handling of phosphate is dependent on the ambient concentration of PTH. Increased serum PTH rapidly leads to decreased expression of the type II Na–P cotransporter and thus to reduced phosphate reabsorption. Ablation of the gene coding for the type IIa Na–P cotransporter in mice results in a loss of 70% of proximal phosphate transport and loss of regulation by both PTH and dietary phosphorus intake (Murer et al., 2000). Renal phosphate excretion is highly responsive to variations in dietary phosphorus intake. With restricted intake, a rapid increase in the tubular reabsorption of phosphate follows, and conversely with a high intake, there is a decrease. Some of the adaptations of the kidney to variations in dietary phosphate intake occur independently of PTH. The mechanisms of this PTH-independent regulation are incompletely understood. However, one of the likely mediators is FGF-23, a member of the fibroblast growth factor (FGF) family of proteins with potent phosphaturic effects. FGF-23 is produced by certain tumors occurring in hypophosphatemic patients with tumor-induced osteomalacia (Shimada et al., 2001). In other patients with the autosomal dominant form of hypophosphatemic rickets, mutations may occur in the FGF-23 gene, rendering the FGF-23 molecule more resistant to proteolytic cleavage. This resistance leads to augmented FGF-23 activity in the kidney with increased phosphate excretion and an abnormally low serum phosphate. Bone The role of osteoblasts and osteoclasts in the deposition and resorption of bone mineral is discussed in detail in the preceding chapter (see Chapter 5). The hydroxyapatite composed of bone mineral contains 6mmol of phosphate for every 10 mmol of calcium (approximately 1 mg of phosphorus for 2 mg calcium). The estimated quantity of calcium released daily by osteoclastic bone resorption is 250 to 500 mg/day. Significantly higher estimates of calcium flux have been derived from calcium isotope studies measuring the movement of calcium between the extracellular fluid pool and the fluid enclosed by bone lining cells (Talmage et al., 1976). One of the well-known effects of PTH is to stimulate bone resorption through osteoclastic activity. This effect is most likely achieved indirectly through the actions of cytokines. Mature osteoclasts do not have PTH receptors. Furthermore, osteoclast-mediated release of calcium from bone into the extracellular fluid is too slow to serve the need for minute-to-minute maintenance of a constant extracellular concentration of ionized calcium. PTH in fact does act to increase extracellular calcium concentration within minutes, but the mechanisms for this rapid action are not entirely understood. PTH receptors exist on bone lining cells, osteocytes, and osteoblasts. Following fibroblast growth factor (PTH) administration, there is a rapid entry of calcium from the bone surface into bone lining cells and osteocytes followed by a net movement of calcium into the extracellular fluid (Talmage et al., 1976). Osteoclasts are directly inhibited by calcitonin, a hormone secreted by parafollicular cells located in the thyroid gland in mammals. Intravenous administration of calcitonin to rats, rabbits, and

Hormone Actions in the Regulation of Calcium and Phosphorus Metabolism

Dietary intake

GI tract

Bone

550 mg

1000 mg

500 mg

300 mg 150 mg

77

5000 mg

Plasma Ca

4900 mg GI excretion

Kidney

Urinary excretion

850 mg

100 mg Net positive balance: 50 mg

FIGURE 6.2  Optimal calcium balance in a healthy young adult. The numbers represent the estimated flux of calcium in milligrams per day in the gastrointestinal tract, kidney, and bone assuming an optimal dietary intake of calcium and normal hormonal regulation of calcium homeostasis.

humans produces a rapid decrease in serum calcium, mediated primarily by a decreased release of ionized calcium from bone. Osteoclasts do have calcitonin receptors. After exposure to calcitonin in vitro, these cells show almost immediate changes in structure accompanied by a reduction in their resorptive activity (Chambers et al., 1985). The pharmacologic effects of calcitonin are clinically useful in treating hypercalcemia in patients, but it is doubtful that calcitonin plays a physiological role in regulating calcium homeostasis in man. Thyroidectomized subjects lacking calcitonin have no recognizable impairment in regulating their serum calcium levels. Furthermore, subjects with persistently high serum levels of calcitonin, due to calcitonin-secreting thyroid tumors (medullary thyroid carcinoma), have no significant derangements in calcium homeostasis.

Overall Requirements for Positive Calcium Balance To achieve a steady increase in bone mass during childhood or to maintain a healthy bone mass during adulthood, the GI tract, the kidney, and bone tissue must all function effectively under normal hormonal regulation. In addition, a sufficient dietary intake of calcium must exist. The basic elements required for a positive calcium balance are illustrated in Figure 6.2. In an optimal state with a daily calcium intake of 1000 mg, approximately 30% or 300 mg of the ingested calcium might be absorbed. Of this amount, 150 mg would be secreted back into the intestinal lumen, leading to a loss of 850 mg/day in the stool. Thus, the net gain of body calcium from the intestinal tract would be 150mg. Approximately 5000 mg of ionized and free calcium would be filtered at the renal glomeruli, and 4900 mg would be reabsorbed back into the plasma leading to a net loss of 100 mg/day in the urine. The combined loss of calcium in the stool and the urine might therefore total 950 mg/day. The estimated flux of calcium into and out of mineralized bone might be on the order of 500 mg/ day. In an optimal situation, as illustrated in Figure 6.2, total calcium intake would exceed output by 50 mg/day, with the net gain being deposited in bone. Obviously diseases causing derangements in the function or control of the GI tract, the kidney, or bone itself can interfere with the delicate balance between calcium intake and loss. Some of these conditions are discussed in the sections below, dealing with the regulation of calcium metabolism by individual hormones.

PTH–VITAMIN D ENDOCRINE SYSTEM AND CALCIUM HOMEOSTASIS PTH PTH is the primary endocrine mediator of calcium homeostasis. The actions of PTH are enhanced and in some cases mediated by calcitriol, the active form of vitamin D.

78

Diet, Nutrients, and Bone Health 1 10 NH2-SER-VAL-SER-GLU-ILE-GLN-LEU-MET-HIS-ASN-LEU-GLY-LYS-HIS LEU 20 28 GLN-LEU-LYS-LYS-ARG-LEU-TRP-GLU-VAL-ARG-GLU-MET-SER-ASN 30 ASP 40 VAL-HIS-ASN-PHE-VAL-ALA-LEU-GLY-ALA-PRO-LEU-ALA-PRO-ARG ASP 50 LEU-VAL-ASN-ASP-GLU-LYS-LYS-ARG-PRO-ARG-GLN-SER-GLY-ALA 60 VAL 70 GLU-SER-HIS-GLU-LYS-SER-LEU-GLY-GLU-ALA-ASP-LYS-ALA-ASN VAL 84 80 O GLN-SER-LYS-ALA-LYS-THY-LEU-VAL-ASP C OH

FIGURE 6.3  Amino acid sequence of human PTH. The first 13 amino acids of the N-terminus are sufficient for partial biological activity, whereas the first 34 amino acids provide full biological activity. (Data from Keutmann, H.T., et al., Biochemistry, 17, 5723–5729, 1978.)

Structure and Biosynthesis of PTH PTH is a polypeptide hormone containing 84 amino acids, as shown in Figure 6.3. Biosynthesis occurs in parathyroid epithelial cells where the PTH gene encodes a larger precursor known as prepro-PTH. The precursor has an additional 29 amino acids at the amino-terminus of PTH, which are removed by the action of proteolytic enzymes before mature PTH is secreted. The intracellular processing of pre-pro-PTH occurs in two steps. In the first step, occurring in the endoplasmic reticulum, a signal peptidase removes the “pre” or leader sequence to yield pro-PTH. In the second step, occurring in the Golgi apparatus, a second peptidase removes the “pro” sequence before PTH is packaged into secretory granules to await secretion. Normally, the precursor forms of mature PTH are not secreted. Forms of Circulating PTH PTH 1-84, commonly referred to as intact PTH, is the main secretory product of the parathyroid glands. Normally, PTH 1–84 accounts for approximately 5%–30% of all PTH peptides in the circulation. The intact hormone has a plasma half life of only 2 to 4 minutes, being rapidly broken down by peptidases in the liver and kidney to form amino-terminal and carboxy-terminal fragments. Amino-terminal fragments containing only the first 13 amino acids of the structure of intact PTH 1-84 are still able to activate the PTH receptor and display biological activity. These aminoterminal fragments are very rapidly cleared from the circulation and constitute less than 10% of circulating PTH peptides. The carboxy-terminal fragments do not bind to the usual PTH receptors and lack a hypercalcemic effect. Carboxy-terminal peptides of PTH have a plasma half life five to ten times longer than that of intact PTH and can constitute 70% to 90% of all PTH peptides in the circulation. Assay of Circulating PTH PTH in the circulation is measured by radioimmunoassays using antibodies directed against antigenic sites located in either the amino-terminus or carboxy-terminus of intact PTH. The most commonly used assays for assessing PTH secretion in human subjects employ two antibodies, one directed at the amino-terminus and a second directed at the carboxy-terminus. In these assays, the carboxy-terminal antibody is typically bound to a solid support such as a polystyrene bead and then is exposed to a serum sample containing both intact PTH and various fragments. All polypeptides containing the carboxy-terminal amino acid sequence bind to the solid support. The PTH peptides bound to the solid support are then exposed to the second amino-terminal antibody, which is prelabeled with 125Iodine or a chemiluminescent molecule allowing detection of the bound antibody.

Hormone Actions in the Regulation of Calcium and Phosphorus Metabolism

79

Generally, such “sandwich” assays detect only the intact, biologically active PTH molecule, containing both carboxy-terminal and amino-terminal antigens (Wood, 1992). PTH-Related Peptide A paracrine factor produced by a variety of cell types and known as PTH-related peptide (PTHrP) has structural similarities to PTH. In the fetus, PTHrP plays a physiological role in directing placental calcium transfer. Normally, PTHrP does not play a significant role in controlling calcium homeostasis, but when secreted in large quantities by certain malignant tumors, it can cause hypercalcemia by stimulating calcium release from bone. PTHrP is the humoral factor most frequently associated with the hypercalcemia of malignancy (Stewart, 2005). PTHrP and PTH share a similar amino-terminal sequence capable of interacting with and stimulating a common receptor in bone tissue. The PTHrP gene is distinct from the PTH gene, and unlike PTH, its expression is not regulated by serum ionized calcium. The carboxyl-terminal sequences of PTH and PTHrP are quite dissimilar. Thus, PTHrP is not measured in most radioimmunoassays for intact PTH.

Actions of PTH PTH increases serum calcium by promoting the release of calcium from bone, by stimulating calcium reabsorption by the distal tubules of the kidney, and indirectly by promoting GI calcium absorption through the mediation of calcitriol (see Figure 6.4). Concurrent with these effects on calcium homeostasis, PTH increases phosphorus release from bone and increases urinary phosphate excretion. The net effect is usually a decrease in serum phosphorus concentration provided that renal responsiveness is normal.

Plasma Ca2+

Parathyroid glands PTH

Bone Increased resorption

Kidney Phosphate excretion

Calcium reabsorption

Release of Ca2+ and phosphate

Plasma Ca2+

Calcitriol formation Intestinal Ca2+ and PO4 absorption

Plasma Ca2+

Plasma Ca2+

FIGURE 6.4  The response of the PTH–calcitriol regulatory system to hypocalcemia. A decline in serum Ca2+ directly stimulates PTH secretion by the parathyroid glands. PTH acts on bone to cause immediate release of exchangeable Ca2+ from the bone compartment into the extracellular fluid. Later increased calcium absorption by osteoclasts releases additional calcium as well as phosphorus. PTH acts on the kidney to increase calcium flux from the glomerular filtrate back into the extracellular fluid and to increase phosphate excretion, thus increasing serum calcium and reducing serum phosphorus concentrations. At the same time, PTH action on the kidney increases the synthesis of calcitriol, which in turn stimulates more active absorption of calcium from the gastrointestinal tract. The net movement of Ca2+ into the extracellular fluid by the combination of these effects restores the serum calcium concentration to normal.

80

Diet, Nutrients, and Bone Health

PTH Receptors and Intracellular Messengers PTH receptors of several types exist in various tissues in the body. The classical receptor, PTH1R, mediates the actions of PTH on bone and kidney. This receptor binds both PTH and PTHrP and recognizes the common N-terminal sequence of both ligands. PTH1R is heavily expressed not only in bone and kidney, but it is also present in other tissues such as breast, skin, heart, blood vessels, and pancreas where its function is unknown. Binding of PTH to PTH1R activates multiple intracellular signaling pathways, including the adenylate cyclase–cyclic AMP system. PTH1R is a 500amino-acid protein located in the cell membrane with its ligand-binding region oriented toward the external cell surface. Like a number of other peptide hormone receptors, the intracellular domain of PTH1R binds G protein subunits that in turn transduce the hormone signal into cellular responses involving intracellular second messengers. Studies with cloned PTH1R indicate that the receptor can be coupled to more than one G protein and that other second messengers, in addition to cAMP, may be involved in the cellular response. Other mediators include the phospholipase C-protein kinase C system, which promotes increased intracellular concentrations of inositol triphosphate and diacylglycerol (Juppner et al., 2006). Although it is unclear which of these systems is paramount in regulating calcium homeostasis, an experiment of nature suggests that the cyclic AMP system may be the most critical. In patients with a genetic disorder known as pseudohypoparathyroidism, a mutation in the stimulatory G protein subunit, Gsα, leads not only to a failure of PTH to generate cAMP in bone and kidney cells but also to a condition of chronic hypocalcemia and hyperphosphatemia. Patients with this mutation fail to show a rise in serum calcium or an increase in urinary cAMP and phosphate after the administration of PTH (Thakker and Juppner, 2006). Actions of PTH on Bone PTH acts on bone to release calcium in two phases. The most immediate effect is to release calcium ions from a bone reservoir that is readily available and in equilibrium with the extracellular fluid. PTH increases the availability of Ca and P in the bone fluid compartment, a response mediated primarily by bone lining cells (Talmage, 1967). Within minutes after exposure to PTH, and before osteoclasts can be involved, there is a rise in extracellular Ca2+ concentration. The mechanism of this effect is still unknown. The later effects of PTH on osteoblasts and osteoclasts vary according to the concentration of PTH and the duration of exposure. Most evidence indicates that osteoblasts, but not mature osteoclasts, express PTH receptors (Murray et al., 2005), yet the action of osteoclasts is ultimately required for bone resorption. The prevailing view today is that the initial effect of PTH on bone cells is on osteoblasts, which have been shown to have a high density of PTHR1 receptors and to respond with a rapid increase in cAMP and inositol triphosphate. When exposed to low, intermittent concentrations of PTH, osteoblastic cells show increased maturation and bone-forming activity. With higher PTH concentrations and more prolonged exposure, osteoclast activation occurs and bone resorption is accelerated. The signal promoting this effect is probably mediated by cytokines stimulated by the action of PTH on osteoblasts and acting on osteoclasts through paracrine mediators. RANK is a cell-surface receptor expressed in osteoclast precursors that controls osteoclast maturation and activation. RANK ligand, a cell-surface protein from cells of the osteoblast lineage, binds to RANK and activates it. Osteoprotegerin (OPG), another protein secreted by osteoblasts, acts as a decoy receptor for RANK ligand, binding it and preventing it from activating RANK. PTH increases the expression of RANK ligand in osteoblastic cells and reduces the expression of OPG, thus promoting the activation of osteoclasts (Ma et al., 2001; Huang et al., 2004). Administration of OPG blocks the calcemic action of exogenous PTH in vivo. The functions of these mediators in bone cell biology are discussed more thoroughly in Chapters 3 and 4. The end result of prolonged, high concentrations of PTH is an increase in osteoclast numbers and activity accompanied by a reduction in the quantity of calcium and phosphorus stored in bone as hydroxyapatite. Other PTH receptors, including receptors recognizing the C-terminal portion of intact PTH as well as circulating C-terminal fragments, have been described in bone and other tissues. Distinct

Hormone Actions in the Regulation of Calcium and Phosphorus Metabolism

81

receptors for the C-terminus of PTH (CPTHRs) have been identified in both bone and renal tissues. The biological significance of these receptors is still unclear. However, in vivo studies suggest that CPTHRs may have a calcium-lowering effect, whereas in vitro studies on bone cells and tissues suggest that they exert antiresorptive effects (Murray et al., 2005). Actions of PTH on Kidney Tubular Handling of Calcium and Phosphorus PTH acts on PTH1R receptors in the kidney tubules to stimulate calcium reabsorption and to inhibit phosphate reabsorption. The distal tubule is the site where calcium reabsorption is actively regulated. As discussed earlier, PTH increases calcium reabsorption from the distal tubular fluid by mediating the activity of several transport proteins including the TRPV5 channel, calbindin, and Ca2+-ATPase (Hoenderop et al., 2002). Thus, when the serum concentration of Ca2+ declines, the resulting rise in serum PTH increases the reabsorption of calcium from distal tubular fluid and less calcium is excreted in the urine, correcting the hypocalcemia (Figure 6.4). The opposite sequence of events occurs when the serum concentration of Ca2+ rises. In addition, an elevated serum calcium acts directly on renal CaSRs to increase calciuresis (Hebert, 1996). PTH is also the major hormone controlling renal phosphate handling. It acts to inhibit mainly proximal but also distal tubular reabsorption of phosphate by affecting the activity of phosphate transport proteins, as discussed earlier. In the proximal tubule, the main effect of PTH is to decrease the activity of the sodium phosphate cotransporter (Murer et al., 2000). PTH Effects on Calcitriol Synthesis An additional important effect of PTH on the kidney is to stimulate the expression of 1-alpha hydroxylase in the proximal tubule, thus promoting the conversion of 25-hydroxyvitamin D to 1,25dihydroxyvitamin D (calcitriol). The factors affecting the interconversion of vitamin D metabolites are discussed in detail in Chapter 10. In the presence of normally functioning kidneys, a decline in serum calcium leads to an increase in serum PTH followed by increased serum calcitriol and an increase in the GI absorption of calcium and phosphorus (Figure 6.4). Basis of Use of PTH as Antiosteoporosis Drug Paradoxically, PTH can act either as an anabolic agent, promoting bone formation, or as a catabolic agent, promoting bone resorption, depending on its concentration and the duration of tissue exposure. When present at low concentrations for intermittent, brief intervals, the predominant effect of PTH, or its active analogs, is the stimulation of osteoblast activation and bone formation. When present in continuous high concentrations, the dominant effects of PTH are to promote RANK ligand expression by osteoblasts and to activate osteoclasts to resorb bone (Ma et al., 2001). Teriparatide (PTH 1–34), a synthetic peptide consisting of the amino-terminal 34 amino acids of PTH, has been shown in clinical trials to promote increased bone density and to reduce fracture risk in postmenopausal women and men with osteoporosis. Teriparatide, when given in low doses (20 mcg) by subcutaneous injections once a day, is therefore an effective antiosteoporosis drug (Neer et al., 2001; Cosman and Lindsay, 2008).

Control of PTH Secretion The concentration of extracellular ionized calcium regulates PTH secretion by a typical negative feedback effect on parathyroid cells. The relationship between serum ionized calcium and serum intact PTH is characterized by a sigmoidal curve, as shown in Figure 6.5. The midpoint on the steepest portion of the curve represents the concentration of ionized calcium at which PTH secretion is half maximal and represents the set point at which serum calcium tends to be maintained. In normal individuals, a decrease of as little as 0.1 mg/dL in serum ionized calcium leads to a rapid increase in PTH secretion, whereas an increase leads to rapid lowering (Brown, 1983). The most

82

Diet, Nutrients, and Bone Health

CaSR-regulated PTH release

Normal senstitivity to Ca2+

Decreased sensitivity to Ca2+

Setpoint

Increased sensitivity to Ca2+

4

6

8

10

12 mg/dL

Serum Ca2+ concentration

FIGURE 6.5  The relationship of PTH secretion to serum calcium concentration. The center curve shows the normal relationship, where the set point of half-maximal secretion is equal to a normal serum calcium concentration of approximately 9 mg/dL. The curve on the left shows an increased sensitivity to inhibition by calcium, as might be seen in a gain-of-function mutation in the CaSR. The curve on the right shows decreased sensitivity, as might be seen in an inactivating mutation in the CaSR in patients with familial hypocalciuric hypercalcemia or in some parathyroid adenomas.

immediate effect of hypocalcemia on parathyroid cells, occurring within minutes, is exocytosis of preformed PTH from secretory vesicles into the extracellular fluid. This is followed within hours by an increase in PTH gene expression, and within days to weeks by proliferation of parathyroid cells (Naveh-Many et al., 1989). The latter effects involve increased PTH gene expression and parathyroid cell hyperplasia and are also activated under conditions where extracellular concentrations of calcitriol are low. Role of Calcium Sensing Receptor in Regulating PTH Secretion The CaSR is a large cell-surface protein expressed in multiple tissues, including the parathyroid glands, kidneys, osteoblasts, and osteoclasts. The proposed structure of the CaSR in human parathyroid glands is shown in Figure 6.6. This receptor is the primary mediator of the negative feedback effect of ionized calcium on PTH secretion. The CaSR is highly expressed on the surface of parathyroid chief cells, where it senses small variations in calcium concentration. Binding of Ca2+ to the CaSR is thought to cause dimerization of monomeric CaSR in the cell membrane and to inhibit adenylate cyclase through action on a G protein. Reduced intracellular levels of cAMP then lead to reduced PTH secretion, although the pathways are not fully known. The CaSR also activates intracellular phospholipases, probably indirectly through a protein-kinase-C-mediated mechanism (Diaz and Brown, 2006). Role of Calcitriol in Regulating PTH Secretion The parathyroid glands possess vitamin D receptors that mediate a negative feedback of active vitamin D metabolites on PTH secretion. Calcitriol acts on parathyroid cells to downregulate the expression of the PTH gene, leading to a decrease in PTH messenger RNA and decreased PTH synthesis (Naveh-Many et al., 1989). Calcitriol also induces increased expression of the CaSR in parathyroid tissue, thereby increasing the sensitivity of parathyroid cells to inhibition by Ca2+ (Dusso et al., 2005). In chronic renal failure, prolonged deficiency of calcitriol leads to reduced CaSR

83

Hormone Actions in the Regulation of Calcium and Phosphorus Metabolism SP

60 100

NH2 HS

200 250

300

400

350 450 550

500

600

613

670

683

745

770

828

838

625

650

700

725

792

807

852

P

P

Conserved Cysteine Acidic

Alu

P

900 P

N-glycosylation P

P

1000

PKC site HOOC

FIGURE 6.6  Schematic representation of the proposed structure of the bovine calcium-sensing receptor. The amino acid sequence from 1–600 resides in the extracellular compartment outside the parathyroid cell plasma membrane and interacts directly with calcium ions. The transmembrane sequence from 613 to 852 is hydrophobic and resides within the cell membrane, whereas the C-terminal sequence resides within the cell. SP denotes a signal peptide that is cleaved in the process of biosynthesis. HS denotes hydrophobic substance. (Reproduced with permission from Brown, E.M., G. Gamba, D. Riccardi, et al. Nature, 366, 578, 1993.)

levels, requiring markedly higher Ca2+ levels to suppress PTH secretion and parathyroid cellular hyperplasia. Other Ions Affecting PTH Secretion Hyperphosphatemia stimulates PTH secretion and parathyroid cell hyperplasia, primarily by inducing a decline in the serum concentration of Ca2+. At high serum phosphorus concentrations, calcium ions move from the extracellular compartment into tissues, resulting in lower concentrations of Ca2+ to inhibit PTH secretion. Evidence exists that chronic hyperphosphatemia, as seen in chronic renal failure, can stimulate PTH secretion and parathyroid gland hyperplasia by a direct action on parathyroid cells (Slatopolsky et al., 1996). Other cations including magnesium, aluminum, and strontium bind to the CaSR, but their affinity is much lower than that of calcium so that their effects on PTH secretion are minimal under normal conditions. Drugs Affecting PTH Secretion Lithium is widely used as a mood-stabilizing drug for the treatment of bipolar disorders. From 10% to 25% of patients on chronic lithium therapy develop mild hypercalcemia and inappropriately high serum PTH concentrations (Kallner and Petterson, 1995). Lithium induces an abnormal response of parathyroid cells to calcium in vitro where the Ca2+–PTH response curve is shifted to the right, causing a higher set point for calcium inhibition (Brown, 1981). In lithium-treated human subjects, there is also a rightward shift of the response curve, suggesting altered calcium sensing at the level of the CaSR (Haden et al., 1997).

84

Diet, Nutrients, and Bone Health

(a)

H N

(b)

H N

OMe

OMe

HCl

HCl

Cl (c) CF3

H N

HCl

FIGURE 6.7  Chemical structure of three phenylalkylamine calcimimetic compounds. (a) NPS R-467. (b)Tecalcet hydrochloride, a first-generation calcimetic compound. (c) Cinacalcet, a second-generation calcimimetic compound. (Adapted from Nagano, N., Pharmacol Ther, 109, 339–365, 2006.)

Calcimimetics Following the identification and sequencing of the CaSR, pharmaceutical research has focused on finding compounds that might interact with the receptor and either stimulate it (calcimimetics) or inhibit it (calcilytics). Calcimimetics would be expected to “trick” the CaSR into responding as it would in the presence of increased Ca2+ and therefore reduce PTH secretion. Several calcimimetics have been described and tested, but only one such compound is currently approved in the United States for use as a drug in human subjects. Cinacalcet is a phenylalkylamine compound with the structure shown in Figure 6.7. In patients with both primary and secondary hyperparathyroidism, cinacalcet is effective in reducing serum PTH concentrations (Nagano, 2006). In one trial involving subjects with primary hyperparathyroidism, there was no improvement in bone mineral density (BMD), although both serum calcium and PTH concentrations were reduced (Peaco*ck et al., 2009). This is a promising field of pharmaceutical research where new compounds are under active investigation.

Diseases Caused by Abnormal PTH Secretion or Action Primary Hyperparathyroidism Primary hyperparathyroidism is diagnosed in patients whenever serum concentrations of both ionized calcium and PTH are abnormally elevated. Secondary hyperparathyroidism refers to conditions in which serum PTH levels are appropriately elevated in response to low serum concentrations of calcium or active vitamin D metabolites. The differences between primary and secondary hyperparathyroidism in terms of clinical laboratory findings are illustrated in Figure 6.8. Primary hyperparathyroidism is the most common cause of hypercalcemia in the United States. This condition is most frequently caused by a benign parathyroid adenoma having a higher than normal set point for negative feedback control by Ca2+. Less frequently, the condition is caused by primary hyperplasia of all of the parathyroid glands. The mechanism of inappropriate PTH secretion involves a decreased sensitivity of individual parathyroid cells to calcium, an increase in the number of parathyroid cells, or a combination of both. The cells in most parathyroid adenomas are monoclonal, suggesting that a mutation has occurred in a key growth-controlling gene. The gene mutations implicated in causing inappropriate cell proliferation are the subject of several reviews (Hendy, 2000; Arnold et al., 2002; Brown, 2002) and will not be discussed in detail here.

Hormone Actions in the Regulation of Calcium and Phosphorus Metabolism

85

500 400 300 200

Serum PTH (IRMA), pg/mL

180 160 Primary hyperparathyroidism Hypercalcemia of malignancy Hypoparathyroidism Normal

140 120 100 80 Normal

60 40 20 0

4

6

8

9

10 11 12 13 14 15 16 17 Serum total calcium, mg/dL

FIGURE 6.8  Serum PTH and calcium concentrations in various disease states. The normal range is illustrated by the sigmoid curve in the box. Serum PTH and calcium are both elevated in patients with primary hyperparathyroidism and are both low in patients with hypoparathyroidism. In patients with the hypercalcemia of malignancy, calcium is elevated but immunoreactive PTH is low because PTHrP is not recognized by most clinical immunoassays for PTH. (Adapted from Haden, S.T., et al., Clin Endocrinol, 52, 329–338, 2000.)

Hypercalcemia of Malignancy As discussed above, PTHrP is produced in large quantities by some malignant tumors, particularly those of squamous cell origin. In other malignancies, hypercalcemia may be mediated by other humoral factors, particularly tumor-produced cytokines (Stewart, 2006). Still, other tumors, particularly lymphomas, express the gene for 1-alpha hydroxylase and are associated with excessive production of calcitriol. In patients having tumor-associated hypercalcemia, serum levels of intact PTH are almost always low (see Figure 6.8). Ectopic production of PTH by nonparathyroid tumors is very rare. Congenital or Acquired Abnormalities of CaSR The density of extracellular CaSRs is known to be reduced in some parathyroid adenomas and may account in part for a decreased sensitivity to calcium feedback control (Kifor et al., 1996, Cetani et al., 2000). In a few uncommon but very informative familial disorders, both activating and inactivating mutations in the CaSR account for the observed abnormalities in calcium homeostasis. Inactivating mutations are responsible for a syndrome called familial hypocalciuric hypercalcemia in which the set point of the PTH response curve to serum Ca2+ is shifted to the right (see Figure6.5). Higher concentrations of Ca2+ are required to inhibit PTH secretion, resulting in hypercalcemia with inappropriately normal or elevated serum PTH. In patients with FHH, reduced activity of the CaSR in the kidney results in reduced tubular clearance of calcium, causing hypocalciuria.

86

Diet, Nutrients, and Bone Health

In contrast, activating mutations of the CaSR cause an increased sensitivity of parathyroid cells to inhibition by Ca2+ and hypocalcemia of varying severity (Brown, 2007). Acquired diseases associated with autoantibodies to the CaSR may affect calcium homeostasis by either activating or inactivating the receptor. Some patients with autoimmune hypoparathyroidism have anti-CaSR antibodies capable of activating the receptor, thus inhibiting PTH secretion, whereas a few patients with hypercalcemia apparently have antibodies that may inactivate the receptor, causing hyperparathyroidism (Gavalas et al., 2007; Pelletier-Morel et al., 2008; Brown, 2009). Hypoparathyroidism and Pseudohypoparathyroidism A deficiency of either PTH itself or an inability to respond to PTH causes hypocalcemia and hyperphosphatemia (Levine, 2006). Primary hypoparathyroidism, an absolute deficiency of PTH, may be caused by an autoimmune disease in which parathyroid cells are damaged or destroyed by autoantibodies. More commonly, the parathyroid glands are inadvertently destroyed by surgical procedures on the thyroid gland or by radiation therapy directed at malignant tumors in the neck. Patients with primary hypoparathyroidism present with hypocalcemia, often accompanied by symptoms of muscle cramps or tetany and an inappropriately low serum level of intact PTH. Patients with what is known as pseudohypoparathyroidism secrete PTH normally but have hypocalcemia due to target organ resistance to the actions of PTH. Detailed knowledge about the mechanisms of PTH resistance in patients with pseudohypoparathyroidism has provided greater insight into the mechanisms of action of PTH at the cellular level. Several variants of this syndrome have been described, but in the most common type, the defect is an inherited abnormality of the G protein (Gsα) linking the PTHR1 receptor to adenylate cyclase and cAMP formation (Levine, 2006). The G protein defect results in a failure of PTH to stimulate cAMP formation in bone and kidney cells. As one might expect, patients with pseudohypoparathyroidism have hypocalcemia and hyperphosphatemia with serum PTH levels that are generally high or high–normal.

STEROID HORMONES AND BONE METABOLISM A number of hormones do not play a direct role in maintaining calcium homeostasis, as does PTH, but do affect overall calcium balance in important ways. A deficiency (or in some cases an excess) of these hormones can lead to skeletal changes resulting in metabolic bone disease. Several classes of steroid hormones, including estrogens, androgens, and glucocorticoids, have important skeletal effects. In addition, thyroid hormones and certain pituitary hormones, especially growth hormone (GH), play a role in promoting bone growth and maturation. The purpose of this section will be to discuss the role of these hormones in bone metabolism and their role in the pathogenesis or treatment of osteoporosis.

General Features of Steroid Hormone Action Steroid hormones are derivatives of cholesterol synthesized by the gonads or the adrenal cortex. During the reproductive years, the principal steroid products of the ovaries are estradiol and progesterone. In men, the principal testicular steroid is testosterone. All steroid hormones are believed to have a common mechanism of action, as depicted in Figure 6.9. Typically, the steroid ligand circulates in the plasma in association with a binding protein. It dissociates from the binding protein to cross the cell membrane and bind to a specific cytoplasmic receptor within the cell. As a result, the receptor–steroid complex usually forms an active dimer that migrates to the cell nucleus and interacts with genomic DNA to regulate the expression of various steroid-responsive genes. Ultimately, the concentrations of proteins encoded by these genes are altered, leading to a cellular response. Steroid receptors belong to a family of related proteins with similar structure, each having its own preference for binding steroids of a particular class. Each receptor class has a ligandbinding region for attachment of the preferred steroid and a nuclear-binding region which interacts

Hormone Actions in the Regulation of Calcium and Phosphorus Metabolism Plasma

Receptor complex

87

Cytoplasm

SBG Nucleus

DNA SRE

Steroid

Receptors

Protein

mRNA

FIGURE 6.9  General mechanism of action for steroid hormones. The steroid ligand dissociates from its specific steroid-binding globulin (SBG) in plasma to enter the target cell by diffusion. Once inside the cell, the steroid binds to an intracellular receptor that is specific for each class of steroid hormones. The steroid–receptor complexes associate to form active dimmers and then enter the cell nucleus where they bind to the DNA of steroid-responsive genes at specific steroid response elements (SREs). Binding results in either enhancement or suppression of gene expression, resulting in alterations in the amounts of messenger RNA (mRNA) and protein formed. (From Ontjes, D.A., The role of estrogens and other steroid hormones in bone metabolism, in Calcium and Phosphorus in Health and Disease, Anderson, J.J.B., and Garner, S.C., eds., CRC Press, Boca Raton, FL. With permission.)

with specific nucleotide sequences in the DNA of steroid-responsive genes. The general structure of steroid receptors is illustrated in Figure 6.10. Note that the receptors for vitamin D and thyroid hormones are also members of the same family (Tsai and O’Malley, 1994).

Estrogens Sources of Estrogens in Women and Men The natural estrogens produced by the ovary are estradiol, estrone, and estriol (see Figure 6.11). Estradiol is the major secretory product of the ovary in women of child-bearing age and accounts for most of the estrogenic activity in the circulation. Estrone and estriol are weaker estrogens and are mainly produced by extragonadal conversion from estradiol in the liver. During pregnancy, large quantities of estrogens are produced by the placenta. Estradiol is synthesized in the body by conversion from testosterone by aromatization of the A ring of the steroid nucleus. This conversion is favored in the ovary, where the concentration of the aromatase enzyme is high, but also occurs to a minor degree in the testes. Testosterone may also be converted to estradiol in nongonadal tissues possessing an aromatase enzyme. After menopause, ovarian production of estradiol virtually ceases, but peripheral production from adrenal androgens continues, yielding mainly the weaker estrogen, estrone, as shown in Figure 6.12. In addition to the natural estrogens, there are a large number of synthetic estrogens, developed specifically by the pharmaceutical industry for use as oral contraceptives and for estrogen replacement therapy. The most commonly used synthetic estrogen is ethinyl estradiol, a compound having approximately 50 times the potency of natural estradiol when administered orally (Figure 6.11). Not all compounds having estrogenic activity are steroids—they need only to be able to bind to and activate the estrogen receptor to have estrogenic activity. An example of such a compound is diethylstilbestrol, also shown in Figure 6.11. Estrogen-like compounds may occur in nature in various foods, including soy protein, where two compounds, genistein and daidzein, have weak estrogenic activity.

88

Diet, Nutrients, and Bone Health hER hPR hGR hMR hTR

185

251

315

553

567

635

680

930

421

487

532

777

602

670

734

984

170 102 DNA

238

Hormone

456

Binding regions

FIGURE 6.10  Structure of the family of steroid hormone receptors. The domains responsible for binding to the steroid ligand to the steroid response elements of genomic DNA are indicated. The structure of the hormone binding domain confers specificity for binding of the steroid class. Human estrogen receptor (hER), human progestin receptor (hPR), human glucocorticoid receptor (hGR), human mineralocorticoid receptor (hMR), and human thyroid hormone receptor (hTR) are shown. (From Ontjes, D.A., The role of estrogens and other steroid hormones in bone metabolism, in Calcium and Phosphorus in Health and Disease, Anderson, J.J.B., and Garner, S.C., eds., CRC Press, Boca Raton, FL. With permission.) OH

Estradiol 11

1

O

Estrone

17

Estriol

OH OH

9

2 OH

OH

OH

4 OH

OH

C

CH

Ethinyl estradiol

OH

OH Diethylstilbestrol

OH

FIGURE 6.11  Structures of natural and synthetic compounds having estrogen activity. Estradiol, estrone, and estriol are all naturally produced. Ethinyl estradiol and diethylstilbesetrol are synthetic estrogens given orally as drugs.

Effects of Estrogen on Bone Metabolism and Calcium Balance As early as the 1940s, Fuller Albright showed that postmenopausal women were in negative calcium balance and that estrogen administration could correct the negative balance (Riggs et al., 2002). Estrogens, like other steroid hormones, act through specific intracellular receptors that are capable of regulating the expression of specific genes. High-affinity estrogen receptors have been found in all tissues usually considered to be targets for estrogen action including the oviducts, endometrium, and breast. Bone cells also contain estrogen receptors. Two types of estrogen receptors have been identified; estrogen receptor-alpha (ER-alpha) and estrogen receptor-beta (ER-beta). Both types of receptors are found in osteoblasts, whereas ER-alpha is also present in osteoclasts. Studies in knockout mice indicate that animals lacking the ER-alpha receptor but not the ER-beta receptor are short and have lower femoral bone densities (Couse and Korach, 1999; Vidal et al., 2000).

Hormone Actions in the Regulation of Calcium and Phosphorus Metabolism

89

Peripheral tissues Adipose cells

Ovary

Estradiol

Adrenal

Androstenedione

Androstenedione

Estrone

Estrone

Reproductive life

Postmenopausal life

FIGURE 6.12  Sources of estrogens in women during reproductive life and after menopause. Estradiol is the main estrogen produced by the ovaries during reproductive life. Both before and following menopause, the ovaries and the adrenal glands produce the weak androgen, androstenedione, which is converted in peripheral tissues to the weak estrogen, estrone. Estrone is the main circulating estrogen in postmenopausal women.

Estrogens tonically inhibit bone resorption mainly by influencing the cytokine network described earlier. More than one cytokine appears to be involved. There is good evidence that estrogen deficiency leads to the increased production of TNF-alpha and IL-1, both of which can stimulate RANK ligand synthesis and activate osteoclasts (Manolagos, 2000; Weitzmann and Pacifici, 2006). Other studies implicate a role for TNF-beta, OPG, and IL-6. An additional effect of estrogen deficiency may be reduced bone formation by osteoblasts. Both IL-7 and TNF-alpha, whose production is known to be increased in estrogen deficiency, have been shown to decrease the activity of osteoblasts. Whatever the cellular mechanisms may be, the net effect of estrogen deficiency is an increase in overall bone resorption and a failure of new bone formation to keep up—hence, a loss of bone mineral. Typically, the earliest phase of trabecular bone loss begins in the perimenopausal period when serum estrogen levels begin to decline and accelerate at menopause. This is followed by a period of slower bone loss involving both trabecular and cortical bone (Riggs et al., 2008). The accelerated phase lasts for 4–8 years before decelerating back to a slower continuous rate of loss characteristic of aging. During the accelerated phase, bone loss in one large cohort of women averaged 5.6% for the vertebrae and 2.9% for the proximal femur over a period of 4 years (Sowers et al., 2006). At the same time that bone density is being lost, changes occur in bone microstructure that further impair its strength. In trabecular bone, there is a disproportionate loss of structural cross struts and in cortical bone increased porosity, leading to increased risk of fragility fractures. Most of the effects of estrogen deficiency are exerted directly on bone cells, but adverse extraskeletal effects may also contribute (Riggs et al., 2008). During the accelerated phase of bone loss, calcium ions move out of bone into the extracellular space, slightly increasing serum calcium concentrations. This results in a slight suppression of PTH secretion and consequently a reduced rate of calcitriol formation. Lower levels of PTH allow an increased loss of calcium in the urine, due to reduced renal tubular reabsorption of calcium, whereas lower levels of calcitriol lead to reduced absorption of calcium and phosphorus from the GI tract. The overall effects of estrogen deficiency on calcium balance are illustrated in Figure 6.13. Effects of Estrogen Administration on Osteoporosis In normal women, the loss of bone mass usually accelerates at menopause, coinciding with the onset of menopausal symptoms such as hot flashes and vagin*l dryness. Early metabolic studies

90

Diet, Nutrients, and Bone Health

Bone

Dietary intake

Increased resorption GI tract

Plasma Ca Decreased Ca2+ absorption

Kidney Decreased PTH

GI excretion

Decreased calcitriol Urinary excretion

Net negative calcium balance

FIGURE 6.13  Results of estrogen deficiency on calcium balance. The primary effect of estrogen deficiency is accelerated bone resorption, leading to a net movement of Ca2+ from bone to serum. The slightly increased concentration of serum Ca2+ inhibits PTH secretion resulting in a decreased synthesis of calcitriol by the kidney. The reduction in calcitriol leads to reduced gastrointestinal absorption of calcium from the diet. The overall effect is a net negative calcium balance.

in postmenopausal women showed that a negative calcium balance could be prevented by estrogen administration (Albright, 1947). Later, when techniques for measuring bone density became available, it became apparent from a number of clinical trials that estrogens could indeed prevent the loss of bone mass and in some cases even increase it. None of these trials were large enough to give conclusive results about the effects of estrogen on fracture risk. Estrogen replacement therapy became a widespread clinical practice, not only for the relief of menopausal symptoms, but because of the widely held opinion that postmenopausal estrogens could prevent osteoporosis and reduce the risk of coronary heart disease. This practice changed with the publication of the findings of the Women’s Health Initiative (WHI) clinical trial. The WHI trial compared the effects of long-term therapy with estrogen plus progestin versus placebo in over 16,000 postmenopausal women. The WHI treatment regimen included 0.625 mg of conjugated equine estrogens and 2.5 mg of medroxyprogesterone given daily. As expected, hip fractures were significantly reduced in the treated subjects (relative risk 0.66; confidence interval 0.45–0.98). However, the risk of several adverse outcomes was increased in the estrogen–progestin-treated group, including myocardial infarction, breast cancer, stroke, and pulmonary embolism. All cause mortality was not significantly affected. The investigators in this study concluded that the net benefits of hormone replacement therapy did not justify its use in most postmenopausal women (Roussouw et al., 2002). As a result of this study, no estrogen product is approved by the Food and Drug Administration in the United States for the treatment of postmenopausal osteoporosis, only for prevention. Many investigators have published short-term studies of bone density outcomes using treatment regimens differing from those used in the WHI trial. Doses of conjugated estrogens as low as 0.3 mg/day are effective in preserving bone density (Lindsay et al., 2005). There is also evidence that estrogens and calcium are interactive. The addition of a calcium supplement can make a dose of 0.3 mg of conjugated estrogens as effective as a higher dose of 0.625 mg/day (Ettinger et al., 1987). Whether alternative doses of estrogen or the use of different progestins might decrease long-term fracture risk without having the same adverse effects as the WHI trial is unknown.

Hormone Actions in the Regulation of Calcium and Phosphorus Metabolism

91

TABLE 6.2 Activity Spectrum of Selective Estrogen Receptor Modulators Compound Clinical Use

Antiestrogen Effects

Proestrogen Effects

Clomiphene Fertility drug Tamoxifen Breast cancer drug

Hypothalamus/pituitary Hypothalamus/pituitary Breast

Raloxifene

Hypothalamus/pituitary Breast cancer

Uterus, breast Uterus Bone Serum lipids Bone Serum lipids

Antiosteoporosis drug

Selective Estrogen Receptor Modulators General Properties of Selective Estrogen Receptor Modulators A wide variety of compounds are capable of interacting with estrogen receptors. Complete estrogen agonists, such as estradiol itself, activate estrogen receptors in all tissues of the body that are normally estrogen responsive. These tissues include not only bone but also the endometrium, breast, and the hypothalamus where a negative feedback of estrogens normally inhibits gonadotropin secretion by the pituitary gland. To have an agonist or proestrogen effect in a specific tissue, a ligand must first bind to the estrogen receptor in that tissue. It must then cause a conformational change in the receptor, leading to receptor dimerization and interaction with DNA at estrogen response elements in specific genes. Only when all of these conditions are fulfilled does the estrogen-like ligand succeed in modifying the expression of the targeted genes. Selective estrogen receptor modulators (SERMs) are compounds that can bind to estrogen receptors in multiple target tissues but can only act as agonists in a limited subset of tissues by activating the receptors. In other tissues, SERMs can act as estrogen antagonists by binding to estrogen receptors, but failing to activate them can block the effects of natural active estrogens. The main effects of several SERMs in current clinical use are summarized in Table 6.2. Clomiphene has been used for many years as a drug to promote fertility in women. Its main therapeutic action is on estrogen receptors in the hypothalamus. These receptors mediate the normal negative feedback effect of estrogens on pituitary secretion of luteinizing hormone (LH) and follicle stimulating hormone (FSH). By inhibiting these receptors, the usual negative feedback is blocked and gonadotropin secretion is increased, thereby stimulating ovulation. Tamoxifen is a drug used primarily as an adjuvant in the treatment of breast cancer. It functions as an estrogen antagonist in breast tissue, inhibiting the estrogen-dependent stimulation of tumor growth in breast cancers having estrogen receptors. Biological Effects of Raloxifene Raloxifene is a SERM used primarily for the treatment of osteoporosis. Its mechanism of action is well understood from X-ray crystallographic studies comparing the conformation of the estrogen receptor after binding either estradiol or raloxifene (Brzozowski et al., 1997; Prince et al., 2008). The binding of raloxifene to the estrogen receptor causes displacement of a C-terminal helix in the receptor, leading to interference with the binding of coactivator proteins that are necessary for estrogen effects in certain tissues (breast) but not in others (bone). Thus, the estrogen receptor is effectively blocked in breast tissue but is activated in bone. Studies in rats and monkeys suggest that raloxifene has effects similar to estradiol in inhibiting bone loss after oophorectomy. Both raloxifene and estrogen inhibit increases in osteoclast number and bone turnover as well as the reduction in bone strength seen after oophorectomy (Turner et al., 1994). Raloxifene has no significant effects on the endometrium but acts as an estrogen antagonist in breast tissue and as an estrogen agonist in the liver, where it reduces synthesis of low-density lipoprotein cholesterol and increases the synthesis of certain clotting factors.

92

Diet, Nutrients, and Bone Health

Effects of SERMs in Treatment of Osteoporosis Both tamoxifen and raloxifene have similar effects of bone, but only raloxifene has been approved by the U.S. Food and Drug Administration for the treatment of osteoporosis. In the largest clinical fracture trial with raloxifene (the MORE Trial), 7705 postmenopausal women with low bone densities received either raloxifene, 60 or 120 mg/day, or a placebo and were followed for 36 months (Ettinger et al., 1999). Women receiving raloxifene had significantly fewer new vertebral fractures than those of control women, regardless of the dose or presence of prior vertebral fractures. Among the women receiving 60- and 120-mg raloxifene, the 3-year risk of vertebral fracture was 6.6% and 5.4%, respectively, compared with a risk of 10.1% in the placebo group. There was no difference among the groups in the incidence of hip or nonvertebral fractures. In the MORE trial and another large clinical trial primarily examining the effects of raloxifene on coronary artery disease (Barrett-Connor et al., 2006), no reduction in coronary events was seen. In both trials, the incidence of invasive breast cancer was reduced in the raloxifene-treated women, whereas the incidence of venous thromboembolic disease was increased. A number of other compounds with estrogen-like activity on bone are known and are being actively investigated as potential drugs for the prevention and treatment of osteoporosis, breast cancer, and possibly coronary disease.

Androgens Sources in Men and Women Androgenic steroids are produced in the gonads and adrenal glands of both men and women. Testosterone, the most potent androgen, is the major steroid secreted by the testes. Much smaller quantities of testosterone are produced by the ovaries, and by the adrenal glands in both sexes. The adrenal glands secrete large quantities of weak androgens, mainly androstenedione and dehydroepiandrosterone. These weak androgens can be converted by peripheral tissues into small quantities of testosterone. Thus, in adult men, over 90% of circulating androgen activity is produced directly by the testes in the form of testosterone. In adult women, circulating androgen activity is much lower and is derived equally from ovarian and adrenal sources. The pathways for peripheral interconversion of androgenic steroids are illustrated in Figure 6.14. Most of the testosterone produced in the Peripheral tissues Adipose and other tissues

Testis

Adrenal Testosterone

Androstenedione DHEA

Aromatization

Testosterone 6 mg/day

Estradiol estrone 0.1 mg/day

Androstenedione 3 mg/day

FIGURE 6.14  Sources of circulating androgens and estrogens in the adult male. The most potent androgen, testosterone, is produced directly in large amounts by the testes. Weaker androgens, including androstenedione and dehydroepiandrosterone, are produced by the adrenal glands. Both testosterone and androstenedione may be aromatized in peripheral tissues to small quantities of estradiol and other estrogens.

Hormone Actions in the Regulation of Calcium and Phosphorus Metabolism

93

ovaries is aromatized to estradiol before it is secreted. Testosterone is also aromatized in peripheral tissues to estradiol in both sexes. Aromatase enzyme activity is widely present in a number of tissues, including adipose cells, hepatocytes, and even bone cells. In the normal male, approximately 15% of circulating estrogen is produced by the testes, whereas the other 85% is produced by peripheral metabolism (Gennari et al., 2004). The clinical importance of aromatase activity for bone health in men is demonstrated by the effect of congenital aromatase deficiency on bone development and bone mass. In males with aromatase deficiency, there is a delay in skeletal development during puberty, lack of epiphysial closure, and osteopenia or osteoporosis. These skeletal defects are similar to those seen in males with a congenital deficiency of ER-alpha and imply that at least part of the effects of testosterone on bone may be mediated by the conversion of testosterone to estrogen (Smith et al., 1994; Gennari et al., 2004). Effects of Testosterone on Bone Metabolism and Calcium Balance The administration of testosterone to orchiectomized animals can stimulate bone formation and inhibit bone resorption. Although a part of this effect may be due to the peripheral conversion of testosterone to estrogen, there is good evidence that testosterone itself has direct effects on bone. An androgen receptor (AR) has been identified in a number of cell types and cloned (Chang et al., 1988; Lubahn et al., 1988). The AR is a typical member of the steroid hormone receptor family, with an overall structure and mode of action resembling that of the estrogen receptor. A high-affinity AR is present in osteoblastic cells of both men and women. These receptors bind other natural and synthetic androgens as well as testosterone, but they do not bind estrogens, progesterone, or glucocorticoids. ARs are also expressed in osteocytes, bone stromal cells, and osteoclasts (Wiren, 2008). The level of AR expression in osteoblasts increases as osteoblasts mature, suggesting that a key action of androgens occurs in mature, mineralizing osteoblasts. The effects of androgens on osteoblasts, as observed in tissue culture experiments, are complex and biphasic. Early exposure promotes osteoblast proliferation and maturation, but continued exposure can promote decreased cell viability and apoptosis. In isolated osteoclasts, androgens reduce bone resorption and reduce the stimulatory effects of PTH. It is likely that part of the inhibitory effect of androgens on bone resorption is mediated by the RANK ligand–OPG system through actions on osteoblasts. Increased expression of OPG mRNA occurs after the exposure of osteoblasts to testosterone in tissue culture (Chen et al., 2004). Other cytokines originating in osteoblasts may also be involved. Androgens, as well as estrogens, inhibit the expression of interleukin-6 (IL-6) by osteoblast cells. IL-6 is another cytokine associated with the activation of osteoclast activity (Wiren, 2008). The release of tonic inhibitory effects of androgens on osteoclastic activity may explain the rapid increase in bone resorption rates seen after orchiectomy in experimental animals. Effects of Androgen on Osteoporosis in Men Testosterone deficiency can occur as a result of congenital defects, as an acquired disease, or as a natural result of aging. In aging men, changes gradually occur in the hypothalamic–pituitary–gonadal axis, leading to decreased serum concentrations of total and free testosterone. Levels of free biologically active testosterone decline more than those of total testosterone do as a result of increasing levels of sex hormone binding globulin (SHBG) occurring with age. In the Baltimore Longitudinal Study of Aging, the fraction of men who were hypogonadal increased with each decade, as shown in Figure 6.15 (Harman et al., 2001). In men aged 80 years and older, the prevalence of hypogonadism was approximately 50% as measured by total testosterone and nearly 90% as measured by the free testosterone index (total testosterone/SHBG). Despite these significant declines, it is unclear whether testosterone deficiency is the dominant factor responsible for the increasing occurrence of osteoporosis in elderly men. Some cross-sectional studies have failed to show an association

94

Diet, Nutrients, and Bone Health 90 80 70

Free T index

60

Total T

% of 50 subjects 40 30 20 10 0

20–29

30–39

40–49

50–59

60–69

70–79

80+

Age decade

FIGURE 6.15  Increasing prevalence of hypogonadism in aging men. Bar height indicates the percent of men in each 10-year interval with total testosterone <325 ng/dL or free testosterone index below 0.153 nmol/nmol (below the normal range for each test). The number of men who were hypogonadal by either criterion increased after the age of 50 years. More men were hypogonadal by the free testosterone index than by total testosterone after the age of 50 years. (Data from Harman, S.M., et al., J Clin Endocrinol Metab, 86, 724–731. 2001.) 25

16 14

20

12

15

10

10

6

8 4

5 0

2

<200

2–300

3–400

4–500

Total testosterone, ng/dL

>500

<10

10–15

15–20

20–25

>25

Estradiol, pg/mL

FIGURE 6.16  Percentage of osteoporosis (T-score < 2.5, dotted bars) and rapid bone loss (>3% per year, cross-hatched bars) in relation to serum total testosterone and estradiol in men aged 65 years and older. The prevalence of low bone density and rapid bone loss both increase as estradiol concentration declines. Low bone density is also associated with the lowest testosterone concentrations of less than 200 ng/dL. (Data from Fink, H.A., et al., J Clin Endocrinol Metab, 91, 3908–3915, 2006.)

between low bone density and serum testosterone levels in older men after adjusting for other factors such as age, body mass index, and estrogen levels. In one large longitudinal study of 2447 men over the age of 65 years, the prevalence of osteoporosis of the hip increased as total or bioavailable estradiol levels declined (Fink et al., 2006). In the same study, there was also an increased incidence of osteoporosis in men whose total testosterone levels were less than 200 ng/dL, but no increasing risk of osteoporosis as testosterone declined above this threshold (see Figure 6.16). One interpretation of these data would be that estrogens are more directly involved than are testosterone in the maintenance of bone health in elderly men. Very low testosterone levels below 200 ng/dL could still have an adverse effect by lowering estradiol formation through a reduced peripheral conversion of testosterone to estrogen. In young men with testosterone deficiency, testosterone replacement clearly increases BMD whether the testosterone deficiency is due to a congenital defect or acquired disease. In one study of 72 men with hypogonadism, testosterone replacement for up to 16 years led to a sustained

95

Bone mineral density, mg/cm3

Hormone Actions in the Regulation of Calcium and Phosphorus Metabolism 240 200 160 120 100 80 60 Range of high fracture risk

40 20

Before therapy

2

4

6

8

10

12

14

16

Duration of testosterone therapy, years

FIGURE 6.17  Changes in spinal bone density in hypogonadal men receiving testosterone. Increases in bone mineral density, as measured by quantitative computed tomography, are shown in 23 men receiving long-term testosterone therapy. The data show measurements made before initiation of testosterone therapy and follow-up measurements made after 2 years of testosterone therapy. The greatest increase in bone density occurred in the first year of therapy. (Data from Behre, H.M., et al., J Clin Endocrinol Metab, 82, 2386–2390, 1997.)

improvement in BMD in the lumber spine, as measured by quantitative computed tomography (see Figure6.17). The mean bone density increased from 95 to 120 mg/cm3 in the first year of treatment (Behre et al., 1997). None of the observational studies in younger men have been placebo-controlled trials with a sufficient number of subjects to demonstrate a reduction in fracture risk. Clinical trials testing the effects of testosterone on BMD in older men have yielded mixed results. In one study of 108 men over the age of 65 years, using transdermal testosterone hip and spine bone density improved no more in the treated subjects than in the placebo group (Snyder et al., 1999). The mean serum testosterone in this trial was 367 ng/dL (still within the normal range). In another trial using intramuscular testosterone in a group of 70 men whose mean testosterone was <350 ng/dL (below the normal range), spine and hip densities were significantly improved in the treated group (Amory et al., 2004). The difference in results may be accounted for by the differences in initial testosterone concentration in the two studies or possibly by differences in the mode of testosterone administration. None of the clinical trial data in older men are sufficient to judge whether any form of testosterone therapy is effective in reducing fractures. Because of the continuing uncertainties regarding therapeutic effectiveness in elderly men, testosterone should probably be used for the treatment of osteoporosis only in combination with other proven drugs, such as bisphosphonates and teriparatide, in subjects where the serum testosterone is clearly low.

Glucocorticoids The adrenals secrete several types of steroids including glucocorticoids, mineralocorticoids, and androgens. Adrenal androgens have been discussed earlier in the sections on estrogens and androgens. The glucocorticoids, typified by cortisol, are essential for life because of their physiological roles in supporting vital functions such as the maintenance of blood pressure and normal glucose homeostasis. A temporary increase in the secretion of cortisol is part of the normal response of the body to stress. At high concentrations, glucocorticoids can suppress the inflammatory and immune responses of the body to a variety of injurious stimuli. Glucocorticoids do not play a significant

96

Diet, Nutrients, and Bone Health

physiological role in calcium homeostasis, but at persistent high concentrations, they do promote a strongly negative calcium balance. Thus, a chronic excess of cortisol or other more potent synthetic glucocorticoids is an important risk factor for the development of osteoporosis. Sources of Glucocorticoids and Causes of Glucocorticoid Excess More than 70 years ago, Harvey Cushing described osteoporosis in patients with pituitary tumors and adrenal hyperplasia. Patients with endogenous overproduction of cortisol (Cushing’s syndrome) usually have either adrenocorticotropin (ACTH)-producing pituitary tumors or cortisol-producing tumors of the adrenal glands. Osteoporosis in such patients can be prevented by appropriate therapy designed to reduce cortisol secretion. Endogenous Cushing’s syndrome is relatively uncommon. Unfortunately, iatrogenic Cushing’s syndrome, due to the administration of exogenous glucocorticoids by physicians, occurs quite frequently. Very potent synthetic glucocorticoids such as prednisone, methylprednisolone, and dexamethasone are widely used for treating a variety of inflammatory and immune-mediated diseases in patients with no underlying abnormalities in adrenal function. An estimated 0.2%–0.5% of the population of the United States is using glucocorticoids chronically at any given time (Adler et al., 2008). Examples of some of the conditions typically treated with glucocorticoids are listed in Table 6.3. Effects of Glucocorticoid Excess on Bone Metabolism and Calcium Balance Direct Effects on Bone Cells Glucocorticoids at supraphysiologic concentrations directly inhibit osteoblastic activity by interacting with cytoplasmic glucocorticoid receptors. In adult bone, glucocorticoid receptors are present in pre-osteoblast/stromal cells and osteoblasts (Abu et al., 2000). The predominant effect is reduced bone formation. There is a reduced expression of procollagen genes and reduced synthesis

TABLE 6.3 Uses of Glucocorticoids for Treating Inflammatory and Immune-Mediated Diseases Disease Category Allergic diseases

Blood dyscrasias

Collagen vascular diseases

Gastrointestinal diseases Pulmonary diseases Renal diseases Neurological diseases Skin disorders

Transplantation medicine

Examples Acute hypersensitivity reactions Serum sickness Allergic drug reactions Acute leukemias, lymphomas Autoimmune thrombocytopenia Autoimmune hemolytic anemias Rheumatoid arthritis Systemic lupus erythematosis Temporal arteritis Polymyalgia rheumatica Ulcerative colitis Crohn’s disease Asthma Chronic obstructive pulmonary disease Autoimmune glomerulitis Multiple sclerosis Myasthenia gravis Allergic dermatitis (various types) Pemphigus and other bullous diseases Poison ivy and other exposures Patients with heart, lung, kidney, or liver transplants

97

Hormone Actions in the Regulation of Calcium and Phosphorus Metabolism

of collagen and other matrix proteins. Osteoblast proliferation is inhibited, and there is increased osteoblast and osteocyte apoptosis, leading to a reduced number of active osteoblasts (Weinstein etal., 1998). Glucocorticoids increase bone resorption, probably through indirect mechanisms. Mature osteoclasts do not have glucocorticoid receptors and are not directly responsive, but glucocorticoids still increase overall osteoclastic activity by promoting osteoclast proliferation (Kaji et al., 1997). They suppress the synthesis of OPG, an inhibitor of osteoclast differentiation, and stimulate the synthesis of RANK ligand by pre-osteoblast/stromal cells (Khosla, 2001). Extraskeletal Effects of Glucocorticoids on Calcium Balance High concentrations of glucocorticoids exert indirect effects on bone metabolism through several other mechanisms in intact humans. They decrease the synthesis of androgens and estrogens primarily by inhibiting the secretion of hypothalamic gonadotropin-releasing hormone and pituitary gonadotropins. They decrease the GI absorption of calcium without impairing the production of calcitriol or binding to its intestinal receptors. The mechanism for this effect is poorly understood. However, in rats, glucocorticoids reduce the resistance to passive paracellular calcium diffusion, resulting in increased serosal-to-mucosal backflux (Yeh et al., 1984). This “leakage” of calcium into the intestinal lumen may account for a reduced net absorption from the GI tract, although calcitriolmediated absorption is not impaired. High serum concentrations of glucocorticoids also tend to increase urinary excretion of calcium. This effect may be due in part to an increase in the filtered load of calcium (Lemann et al., 1970) but may also be due to a decrease in renal tubular calcium reabsorption. In patients with hyperparathyroidism and a high filtered load of calcium, administration of glucocorticoids causes marked hypercalciuria (Breslau et al., 1982). The effects of high concentrations of glucocorticoids on bone cells as well as the extraskeletal effects on the GI tract and the kidney are summarized in Figure 6.18. The combined actions on bone, gut, and kidney can result in a rapid decrease in bone mass and a negative overall calcium balance. Bone

Decreased osteoblast activity

Dietary intake

Increased osteoclast activity

Increased Ca2+ backflux GI tract

Increased filtered load of Ca2+

Plasma Ca Decreased net Ca2+ absorption

Decreased reabsorption

Kidney

GI excretion Urinary excretion Net negative calcium balance

FIGURE 6.18  Results of glucocorticoid excess on calcium balance. The primary effect is on osteoblasts leading to reduced osteoblast numbers and reduced synthesis of bone matrix proteins. Osteoclast numbers and activity are increased. Reduced bone formation and increased bone resorption lead to a decrease in bone mass. Net gastrointestinal absorption of calcium is decreased, whereas urinary excretion of calcium is increased, also contributing to the net negative calcium balance.

98

Diet, Nutrients, and Bone Health 6 5 4

Relative risk of 3 fracture 2

No steroid Prednisolone 2.5–7.5 mg Prednisolone >7.5 mg

1 0

Hip

Spine

FIGURE 6.19  Effects of low-dose prednisolone on bone. The relative risk of both spine and hip fractures increases progressively, with daily prednisolone doses beginning with as little as 2.5–7.5 mg/day. This dose of prednisolone is equivalent to 12.5–37.5 mg/day of hydrocortisone. (Data from van Staa, T.P., et al., Arthritis Rheum, 48, 3224–3229, 2003.)

Glucocorticoid-Induced Osteoporosis Glucocorticoid-induced osteoporosis is the most common form of drug-induced osteoporosis. Depending on the glucocorticoid dose, there is typically a loss of 1.5%–3% of bone mass during the first 6 months of therapy (Adler et al., 2008). Trabecular bone is initially most affected, followed by a loss of cortical bone. After 2 years of continuous therapy, the rate of bone loss slows to 1.5%–3% per year but continues at a higher rate than normal for the age and sex of the patient. Fractures increase within the first 6 months of therapy and may even precede measurable changes in bone density. Several studies have observed that fractures begin to occur at a higher BMD threshold in patients with glucocorticoid-induced osteoporosis than that in patients with typical postmenopausal osteoporosis (van Staa et al., 2003; Kanis et al., 2004). This suggests that alterations in bone “quality” or microarchitecture may precede changes in overall bone mass. A safe dose of glucocorticoid low enough to avoid the increased risk of osteoporosis has never been clearly established. In an observational study of 240,000 glucocorticoid-treated patients in the United Kingdom, there was a trend toward increased hip and spine fractures even at “physiological” doses of prednisolone of 2.5 to 7.5 mg/day, as shown in Figure 6.19. Cumulative glucocorticoid dose over time appears to be the most important predictor of bone loss (van Staa et al., 2007). Limited information on alternate day therapy suggests that this regimen is not necessarily protective of bone (Gluck et al., 1981). Fracture Risk Reduction in Glucocorticoid-Induced Osteoporosis Glucocorticoids were first used over 60 years ago to treat inflammatory diseases. Although other anti-inflammatory and immunosuppressant drugs are now available, these drugs also have significant associated side effects. In the future, it is likely that glucocorticoids will continue to be widely used because of their low cost and therapeutic effectiveness. Several approaches should be effective in reducing fracture risk in patients requiring long-term glucocorticoid therapy. First, the dose of glucocorticoid should be reduced as quickly as possible to the lowest amount required to maintain control of the underlying disease. Topical or regional steroid administration is often a feasible alternative when the disease process is locally confined, for example, to the bronchial tree, a localized area of skin, or a single joint. Topical administration of glucocorticoid derivatives that are designed to act locally and to be poorly absorbed into the general circulation can yield therapeutic benefits and minimize systemic side effects. Lifestyle modifications including exercise and nutrition are effective in reducing the risk of fractures in patients who are expected to take systemic glucocorticoids for a period of 6 months or more. Smoking cessation, increased weight-bearing exercise, and reduction in alcohol intake are examples of modifiable risk factors that can reduce fracture risk, even in patients continuing on glucocorticoid therapy.

Hormone Actions in the Regulation of Calcium and Phosphorus Metabolism

99

Dietary Supplementation All guidelines for the prevention of glucocorticoid-induced osteoporosis recommend calcium intakes of at least 1200 mg/day. To achieve this, calcium supplements are usually required because the diets of most glucocorticoid users are insufficient in calcium. In addition, vitamin D supplements should be administered to maintain serum levels of 25-hydroxyvitamin D of at least 30 ng/mL. For most adults over the age of 50 years, the required amount will be at least 1000 IU/day of vitamin D3. Some patients with less efficiency in absorbing the vitamin will require more. This means that vitamin D therapy should be guided by actual serum measurements in individual patients and not by a predetermined dose. A number of clinical trials have used more active vitamin D derivatives such as alfacalcidiol and calcitriol to prevent bone loss in patients on long-term glucocorticoids. Meta-analyses of these trials have concluded that virtually all vitamin D analogs are more effective in reducing fracture risk than no vitamin D therapy at all (de Nijs et al., 2004; Richy etal., 2005). Regardless of the type of vitamin D preparation used, hip fracture rates remained high, especially in patients over 50 years old or with previous fragility fractures, suggesting that dietary supplementation, although useful, was not sufficient for alleviating all risks. Antiosteoporosis Drugs The availability of effective antiosteoporosis drugs, particularly the bisphosphonates, has greatly changed the management of patients on chronic glucocorticoid therapy. Several bisphosphonates, including alendronate, risedronate, and zoledronate, have been shown to protect against bone loss in patients on glucocorticoids and are approved by the Food and Drug Administration for this indication (Adler et al., 2008). Fracture risk reduction in trials with alendronate and risedronate ranged from 38% to 90%. Teriparatide (synthetic PTH 1-34) is also an effective agent in reducing fracture risk. In a 3-year clinical trial comparing the effects of teriparatide versus those of alendronate in 428 subjects with glucocorticoid-induced osteoporosis, both treatments significantly increased BMD, but teriparatide had a greater effect (Saag et al., 2009). Fewer subjects in the teriparatide group had new vertebral fractures than did subjects in the alendronate group (1.7% vs. 7.7%, P = .007). There was no difference in the incidence of new nonvertebral fractures. Sex steroid replacement may be appropriate for some patients. Estrogen replacement reverses or stabilizes bone loss in postmenopausal women on long-term glucocorticoids (Lukert and Raisz, 1990), although there is no information documenting a beneficial effect on fracture risk. As discussed earlier in the section on estrogen therapy, there are potential adverse effects from long-term estrogen replacement therapy that argue against the use of estrogens as first-line therapy when other effective agents are available. Testosterone replacement in men with glucocorticoid-induced hypogonadism is also effective in reducing bone loss. In one study of 15 asthmatic men on longterm glucocorticoid therapy, there was a 5% improvement of bone density in the lumbar spine after treatment for 12 months with intramuscular injections of testosterone (Reid et al., 1996). Guidelines for Dietary and Drug Management Based on the available evidence of fracture risk and the effects of various interventions, several organizations have issued guidelines for the prevention and treatment of glucocorticoid-induced osteoporosis. The American College of Rheumatology recommends that all patients receiving the equivalent of ≥5 mg/day of prednisone for 3 months or having a BMD T-score below −1.0 receive the following:

1. Calcium and vitamin D supplementation (1000 to 1500 mg/day and 800 IU/day, respectively). 2. Bisphosphonate therapy using an approved drug at the standard dose. Caution should be used in administering bisphosphonates to premenopausal women who might become pregnant. If bisphosphonates are not tolerated, the use of teriparatide or calcitonin should be considered. 3. Replacement of testosterone in men if deficient. 4. Annual measurement of BMD for follow-up.

100

Diet, Nutrients, and Bone Health

Thyroid Hormones and Bone Metabolism Introduction The thyroid gland produces two closely related hormones, thyroxine (T4) and triiodothyronine (T3). These hormones act on nuclear receptors belonging to the same family as the steroid hormone receptors and control the expression of thyroid-hormone-responsive genes in multiple body tissues, including bone. Both T4 and T3 are derived from thyroglobulin, an iodinated glycoprotein that is synthesized by thyroid follicular cells within the gland. The uptake of iodine by thyroid cells, the synthesis of thyroglobulin, and the cleavage and release of T4 and T3 are all controlled by thyroidstimulating hormone (TSH) from the pituitary gland. The thyroid gland secretes virtually 100% of circulating T4 but only 20% of circulating T3. The remaining 80% of T3 is derived from T4 by conversion in peripheral tissues, particularly the liver, where an enzyme (5′-deiodinase) removes an iodine atom from the outer ring of T4 to yield T3. An alternative enzyme (3′-deiodinase) removes an alternative iodine atom, yielding an inactive isomeric product known as reverse T3. The sources of circulating T4, T3, and reverse T3 are illustrated in Figure 6.20. Over 99% of T4 and T3 in the circulation are bound to thyroid-binding globulin and other binding proteins produced by the liver. A classical negative feedback relationship exists between circulating levels of T4 and T3 and pituitary secretion of TSH, which operates to maintain constant concentrations of both thyroid hormones within the circulation. When serum concentrations of free T3 and T4 rise, due to either endogenous thyroid secretion or to administration of exogenous thyroid hormones, pituitary TSH secretion is inhibited, resulting in decreased thyroid secretion. Small increases in thyroid hormone supply can be buffered by decreased stimulation by TSH up to a point. With further increases in supply, serum concentrations of T3 and T4 eventually rise. Thus, patients with mild hyperthyroidism typically have a suppressed serum TSH but maintain T3 and T4 concentrations within the normal range. Such patients have few symptoms and are described as having “subclinical hyperthyroidism.” Patients with more severe hyperthyroidism typically have not only suppressed serum TSH but also elevated serum levels of T3 and T4.

T4

Direct secretion from thyroid

Production per day

Serum levels

30 mcg

Hepatic conversion

T4

T3

rT3

80 mcg

Direct secretion from thyroid

Hepatic conversion

30 mcg

T3

T4

rT3

120 ng/dL

8 mcg/dL

30 ng/dL

FIGURE 6.20  Sources of thyroid hormones in the circulation. The thyroid gland is the sole source of T4 (levothyroxine). The thyroid secretes smaller quantities of T3 (3,5,3′-triiodothyronine) and reverse T3 (3,3′,5′triiodothyronine). The majority of circulating T3 and reverse T3 are produced in the liver from T4 by either a 5′-deiodinase (for T3) or a 5-deiodinase (reverse T3). T3 has a higher affinity for the thyroid hormone receptor than T4. Reverse T3 is biologically inactive.

Hormone Actions in the Regulation of Calcium and Phosphorus Metabolism

101

Thyroid hormones act on virtually every tissue in the body. There are two thyroid hormone receptor isoforms, TRα and TRβ. Both receptors are expressed in most tissues, but their levels of expression vary depending on the organ. TRα is most abundantly expressed in the brain, kidney, gonads, muscle, heart, and bone, whereas TRβ is more highly expressed in pituitary and liver. T3 is bound with 10–15 times greater affinity to both receptors than is T4, explaining its greater potency. Once thyroid hormone binds to its intracellular receptor, the receptors bind together as dimers and interact with thyroid hormone response elements in the promoter regions of target genes leading to either increased or decreased gene expression. Actions on Skeletal Tissue Thyroid hormones play a role in skeletal growth and maturation and are necessary for normal chondrocyte development (Baran, 2008). The mechanisms of action are complex and still incompletely understood. T4 and T3 regulate heparin sulfate proteoglycan expression in the growth plate during endochondral bone formation and play a role in bone maturation. After bone growth is completed, thyroid hormones increase the activity of both osteoblasts and osteoclasts. T3 can act on osteoblast cells in vitro, increasing their overall activity in part through mediation of growth factors including insulin-like growth factor-I (IGF-I) and fibroblast growth factor (FGF) (Pepene et al., 2001). Circulating IGF-I levels are decreased in patients with hypothyroidism, who typically show decreased rates of bone formation. T3 also increases osteoclast activity. The effect may be mediated in part by the RANK–RANK ligand system through effects on osteoblasts but could also represent a more direct action. T3 increases the expression of c-fos mRNA in osteoclast precursors, suggesting a mechanism independent of RANK ligand–RANK interaction (Kanatani et al., 2004). Effects of Thyroid Hormone Excess on Bone Health Overt untreated hyperthyroidism is associated with accelerated bone remodeling, reduced bone density, and an increased fracture rate. Although bone formation rates are increased, bone resorption is increased even further so that there is a net efflux of calcium and phosphorus from bone. PTH secretion is inhibited, and urinary calcium excretion is increased. If urinary clearance of the added calcium load is insufficient to maintain calcium homeostasis, hypercalcemia can develop. Histomorphometric studies of the bone remodeling cycle in hyperthyroid subjects have shown a pronounced shortening of the 200-day length of the normal cycle by as much as 50%. Because osteoclastic and osteoblastic activities are out of balance, significant bone loss occurs with every completed cycle (Eriksen, 1986). In hypothyroid patients, the length of the cycle can double. The degree of bone loss in several studies of patient cohorts with long-term hyperthyroidism is in the range of 10%–20%. Some studies have shown partial recovery of bone density after thyroid hormone levels are reduced by appropriate treatment (Nielsen et al., 1979; Rosen and Adler, 1992; Diamond et al., 1994; Karga et al., 2004). The risk of fractures is also increased in hyperthyroidism. In one prospective cohort study of 686 white women over the age of 65 years, those with hyperthyroidism were identified by the presence of a low TSH of less than 0.1 mU/L (normal 0.4–4.5 mU/L). (In this subgroup of women, excess circulating thyroid hormones had suppressed serum TSH levels by the negative feedback relationship discussed above.) The relative risks of hip and vertebral fractures were increased in the hyperthyroid subjects by factors of 3.6 and 4.5, respectively, after a mean follow-up of 3.7 years (Bauer et al., 2001). It is still unclear how severe and prolonged hyperthyroidism must be in order for adverse skeletal effects to occur. Many patients being treated with thyroid hormone for hypothyroidism actually become mildly hyperthyroid, due to excessive thyroid hormone administration. Typically, these patients have a moderately suppressed serum TSH and a serum T4 still within the normal range. Most of them have no symptoms of hyperthyroidism and are considered to have subclinical hyperthyroidism (see above). Several studies have shown that postmenopausal women with subclinical

102

Diet, Nutrients, and Bone Health

hyperthyroidism have accelerated bone loss (Ross et al., 1987; Lehmke et al., 1992; Franklyn et al., 1994), but adverse effects have been less apparent in premenopausal women and men. One metaanalysis suggested a significant reduction in bone mass only in postmenopausal women (Faber and Galloe, 1994), whereas a second meta-analysis showed adverse effects in premenopausal women as well (Uzzan et al., 1996). Fracture rates have been increased in some studies of patients with iatrogenic subclinical hyperthyroidism (Bauer et al., 2001; Flynn et al., 2010) but not in all (Leese et al., 1992). The risk is probably related not only to the severity and duration of TSH suppression but also to the age of the patients studied. Subclinical hyperthyroidism poses a greater risk for elderly women than it does for premenopausal women or men. Clearly, thyroid hormone replacement therapy per se does not increase the risk of fracture if not given in excess. In summary, hyperthyroidism has a deleterious effect on bone health proportionate to its severity and duration. Most patients with a suppressed serum TSH and elevated serum T3 and T4 (overt hyperthyroidism) should be treated promptly to reduce their thyroid hormone levels to normal. In patients with only a suppressed TSH who have no symptoms of hyperthyroidism (subclinical hyperthyroidism), immediate therapy may not be required unless the patient has other risk factors for osteoporosis. Postmenopausal women with subclinical hyperthyroidism should be treated because of the prevailing clinical evidence that they are more likely to suffer bone loss and fractures if their mild hyperthyroidism is left untreated.

Growth Hormone and IGF-I Introduction GH is most critical to bone health during childhood and adolescence when it plays a primary role in promoting linear bone growth. GH acts together with other hormones, including sex steroids and thyroid hormone, to promote formation and development of the growth plate and new bone formation. The rate of growth is most rapid during fetal life, and again at puberty, when a growth spurt typically occurs. GH secretion normally peaks during adolescence and then declines steadily during adult life. By middle age, GH secretion rates are typically only 15% of the rates during puberty (Melmed and Jameson, 2006). GH is a 191-amino-acid polypeptide whose secretion is controlled by two hypothalamic factors, growth hormone releasing hormone (GHRH) and somatostatin. GHRH and somatostatin are both secreted by specific hypothalamic neurons and transported to the anterior pituitary gland via portal blood vessels where they interact with specific receptors on GH-producing cells. GHRH, the dominant hypothalamic factor controlling GH secretion, has a stimulatory effect, whereas somatostatin has an inhibitory effect. The production and release of GHRH and somatostatin in the hypothalamus are controlled by a complex variety of CNS stimuli, including sleep, exercise, and hypoglycemia. Plasma concentrations of GH vary throughout the day, depending on the balance between the effects of GHRH and somatostatin at any given time. Peak levels are typically seen at night within an hour after the onset of deep sleep. The main factors involved in the regulation of GH secretion are illustrated in Figure 6.21. GH circulates in the plasma bound to a GH-binding protein with a structure similar to that of the extracellular domain of the GH receptor. To interact with its receptor, GH first dissociates from the binding protein and then binds to the extracellular domain of the receptor in target tissues. Receptor binding induces dimerization of GH–receptor complexes followed by signaling through the JAK/ STAT intracellular pathway. The activated STAT proteins translocate to the cell nucleus where they modulate the expression of GH-responsive genes. The metabolic effects of GH action in addition to linear growth in children include nitrogen retention with enhanced lean body mass and lipolysis with decreased fat mass in both children and adults (Melmed and Jameson, 2006). Most of the growth-promoting effects of GH on peripheral tissues are exerted via IGF-I. IGF-I is a polypeptide with some structural similarities to insulin, but with distinct receptors. It is

Hormone Actions in the Regulation of Calcium and Phosphorus Metabolism

103

CNS effects/diurnal rhythm/glucose GHRH

_

+

Pituitary

GH IGF-I

Growth

+

GH

IGF-I

Somatostatin + IGF-I & GH _

IGF-I

+

Liver IGF-I synthesis

FIGURE 6.21  The complex control of growth hormone and insulin-like growth factor I secretion. GH is secreted by the anterior pituitary under the influence of two hypothalamic hormones. GHRH is stimulatory, and somatostatin is inhibitory for GH release. GH acts on target tissues including liver and bone to stimulate the synthesis and release of IGF-I. IGF-I may act locally to promote mitogenesis in its tissue of origin, as in bone, or may be secreted into the circulation, as from the liver. Circulating IGF-I exerts a negative feedback directly on the pituitary gland to downregulate GH secretion. Together, IGF-I and GH also act on the hypothalamus to stimulate somatostatin secretion, thereby reducing GH secretion by the pituitary. GHRH and somatostatin secretion is regulated by a number of other physiological variables mediated by the central nervous system, including sleep, stress, and the prevailing blood glucose concentration.

a potent mitogenic agent produced by multiple tissues under the influence of GH. Most of the IGF-I in plasma originates from the liver and acts as a circulating growth factor. IGF-I can also be produced directly by bone and connective tissue cells and act locally on neighboring cells to promote growth. Under normal conditions, there is a close correlation between circulating levels of IGF-I and GH. Patients with a deficiency of GH will typically have low plasma levels of both GH and IGF-I. IGF-I participates together with GHRH and somatostatin in the feedback regulation of GH secretion (see Figure 6.21). When IGF-I levels rise, GH secretion normally tends to fall. In a rare condition known as Laron dwarfism involving a loss-of-function mutation in the GH receptor, IGF-I is not produced. Patients with this condition have short stature in spite of normal or high GH levels because their peripheral tissues fail to respond to GH. IGF-I is a single-chain polypeptide with 70 amino acids. It is structurally distinct from IGF-II, a related peptide that also has mitogenic activity, but is not GH dependent. IGF-I in the circulation is mostly bound to serum IGF binding proteins (IGFBPs). These binding proteins serve as a storage reservoir, prolonging the half life of circulating IGF and affecting its distribution among various tissues. The most important binding protein is IGFBP-3, whose synthesis is also stimulated by GH (Rosen and Niu, 2008). The IFG-I receptor is structurally hom*ologous to the insulin receptor and the IGF-II receptor but has a higher affinity for IGF-I than for insulin or IGF-II. This receptor has intrinsic tyrosine kinase activity which is critical for second messenger generation after IGF-I binding occurs. Intracellular signaling involves the JAK/STAT and mitogen-activated protein kinase (MAPK) pathways (Rosen and Niu, 2008). When activated, the IGF-I receptor has antiapoptotic effects in several cell types, including osteoblasts and osteocytes (Le Roith et al., 1997). Effects of GH and IGF-I on Bone Metabolism The overall effects of GH on bone are anabolic in adults as well as in children. However, the effects of GH on bone remodeling are complex because there are both circulating and local sources of IGF-I

104

Diet, Nutrients, and Bone Health

that may interact with bone cells. In vitro both GH and IGF-I have mitogenic effects on osteoblastic cell lines. GH induces the local synthesis of IGF-I by bone cells. The mitogenic effects of GH on osteoblasts can be blocked by the addition of specific monoclonal antibodies to IGF-I (Mohan and Baylink, 1991). IGF-I is mitogenic when added to cultures of rodent preosteoblasts, rapidly increasing the expression of the protooncogene c-fos (Merriman et al., 1990). IGF-I can increase collagen synthesis, alkaline phosphatase activity, and osteocalcin production in osteoblasts and can act as an antiapoptotic factor for differentiated osteoblasts (Rosen and Niu, 2008). Although a major effect of IGF-I is exerted on osteoblasts, it also plays a role in the coupling of osteoblast activity with osteoclast recruitment and activation, either directly through the IGF-I receptor or through the RANK–RANK ligand pathway (Mochizuki et al., 1992; Rubin et al., 2002). Both in vitro and in vivo, the absence of IGF-I impairs osteoclast recruitment and activation (Wang et al., 2006). The effects of GH deficiency on skeletal development and growth in children are clear cut. Linear bone growth is impaired and skeletal maturation is usually retarded. Children with GH deficiency have low bone mass, as measured by techniques such as single-photon absorptiometry (Wuster etal., 1992). In adults, the skeletal effects of GH deficiency are more difficult to define than those in children. This is due in part to the fact that serum GH and IGF-I levels normally decline with aging. In adults, severe GH deficiency usually occurs in association with panhypopituitarism and is accompanied by deficiencies of gonadal and adrenal steroids as well as thyroid hormone. In such patients, the frequent underreplacement of estrogen and testosterone, as well as the possibility of overreplacement with glucocorticoids and thyroid hormone, make it more difficult to identify skeletal effects due solely to GH deficiency. With these reservations in mind, there is much evidence suggesting that adult GH deficiency is associated with low bone turnover osteoporosis and increased fracture risk (Guistina et al., 2008; Rosen and Niu, 2008). Several cross-sectional studies of adults with GH deficiency have found lumbar spine BMD reduced in comparison with those of age- and sexmatched controls (Johansson et al., 1992; Holmes et al., 1994; Colao et al., 1999) and fracture rates increased (Wuster et al., 2001). Bone biopsies from adult males with GH deficiency show decreased osteoid and mineralizing surfaces, suggesting that impaired bone formation plays a dominant role (Bravenboer et al., 1996). Evidence that declining GH secretion causes bone loss in healthy older adults is more conflicted. Some investigators studying older adults without evidence of pituitary disease have found a significant relationship between bone mass and serum levels IGF-I. In the Framingham Heart Study, there was a strong correlation between the lowest quintile for serum IGF-I and BMD of the spine and hip (Langlois et al., 1998). In males, correlations between low BMD and low concentrations of IGF-I and IGFBP-3 have been reported by several investigators (Ljunghall et al., 1992; Kurland et al., 1997). In one study of young men with idiopathic osteoporosis and low serum IGF-I, dynamic tests of GH secretion were normal, suggesting that factors other than GH deficiency may have been responsible for the low IGF-I levels (Kurland et al., 1998). Others have concluded that even in adults with well-documented GH deficiency, a deleterious effect on BMD is apparent only in younger individuals. Beyond the age of 60 years, no differences were noted between the bone density scores of patients with a history of GH deficiency and control subjects (Murray et al., 2004). In summary, strong evidence supports that GH deficiency plays a role in the pathogenesis of osteopenia in younger subjects with clear-cut pituitary deficiency. However, in other subjects without a history of pituitary disease, where correlations between low BMD and low IGF-I may exist, it is still unclear whether GH deficiency plays a critical role. Therapy for Osteoporosis GH replacement therapy using recombinant human growth hormone (rhGH) is now an accepted standard of care for children with GH deficiency. The results in terms of skeletal growth and improvement in BMD have been quite successful. In one observational study, 26 GH-deficient

Hormone Actions in the Regulation of Calcium and Phosphorus Metabolism

105

children were given rhGH for 12 months. During that period, the bone density scores (Z-scores) were normalized in nearly 50% of subjects (Saggese et al., 1993). In children with short stature due to causes other than GH deficiency, GH treatment has not improved BMD despite increases in lean body mass (Rosen and Niu, 2008). In GH-deficient adults, GH replacement therapy causes an increase in serum IGF-I and in lean body mass within 2 weeks. Markers of bone metabolism, including serum osteocalcin and urinary hydroxyproline, routinely increase. The effects of GH replacement on bone density have varied depending on the age and sex of the patients and the dose of GH administered. In one of the few randomized, placebo-controlled clinical trials using “physiological” doses of GH to normalize serum IGF-I, spine BMD increased by nearly 4% in men but did not improve significantly in women. Neither men nor women showed significant improvement in hip BMD (Snyder et al., 2007). Thus far, no clinical trials of sufficient size to examine fractures as a clinical outcome have been reported. Only a few small trials of GH therapy have been reported in osteoporotic patients without preexisting GH deficiency. The results of these trials, some of which included combinations of GH with antiresorptive agents, have been inconclusive. Likewise, there have been only limited short-term trials testing the effects of recombinant IGF-I administration to adults with osteopenia. In one group of women with anorexia nervosa and low bone mass, IGF-I therapy for 9 months increased spine BMD by 1.1%, whereas a combination of IGF-I and estrogen yielded a 1.8% increase. Subjects receiving placebo experienced bone loss over the same period (Grinspoon et al., 2002). Currently, therapy with GH and IGF-I holds promise for the treatment of osteoporosis, particularly in subjects with well-defined GH deficiency. These agents are not currently approved for the treatment of osteoporosis because of their high cost, potential long-term side effects, and lack of demonstrated efficacy in fracture reduction. Active therapeutic research with anabolic agents is ongoing and may yield more favorable results in the future, particularly in combination with antiresorptive drugs.

SUMMARY AND CONCLUSIONS The maintenance of a constant concentration of calcium ions in the extracellular fluid is of critical importance in most multicellular organisms. In man, calcium homeostasis is regulated primarily by the endocrine system through the actions of PTH and 1,25-dihydroxyvitamin D (calcitriol). PTH secretion is regulated by a negative feedback of calcium ions acting through a CaSR in the plasma membrane of parathyroid cells and by a negative feedback of calcitriol acting to inhibit the synthesis of PTH mRNA. In the presence of reduced dietary intake of calcium and vitamin D, serum concentrations of PTH are typically elevated and renal conversion of 25-hydroxyvitamin D to calcitriol is stimulated. Conversely, an excess of dietary calcium or vitamin D downregulates both PTH secretion and calcitriol synthesis. The most important organs mediating calcium homeostasis are the GI tract, the kidney, and bone itself, which contains 99% of the calcium in the human body. GI absorption of dietary calcium is largely dependent on the presence of calcitriol. Under conditions where the dietary intake of calcium is limited, increased renal synthesis of calcitriol promotes more efficient absorption of calcium and phosphorus by the GI tract. Calcium deprivation also leads to hormonally mediated adjustments in the renal handling of calcium and phosphorus. Increased secretion of PTH stimulates increased renal tubular reabsorption of calcium from the glomerular filtrate, conserving calcium and helping to maintain calcium homeostasis. Bone cells, primarily osteocytes and bone lining cells, mediate a rapid flux of calcium ions between bone and the extracellular fluid under the influence of PTH. Osteoclasts mediate a slower process of calcium resorption from hydroxyapatite. Bone lining cells, osteocytes, and osteoblasts (but not osteoclasts) have PTH receptors. PTH has a complex, biphasic effect on bone mineral content and calcium balance. One of the earliest effects of PTH is a direct stimulation of osteoblasts resulting in increased bone formation. A more delayed effect, seen with sustained high concentrations of

106

Diet, Nutrients, and Bone Health

PTH, consists of an activation of osteoclasts resulting in increased bone resorption. The longer-term effect of PTH on bone resorption is not due to the direct action of PTH on osteoclasts but is mediated by cytokines including the RANK–RANK ligand system. The differing responses of bone cells to short-term and long-term PTH exposure explain why patients with primary hyperparathyroidism due to PTH-secreting parathyroid tumors tend to lose bone mass, whereas patients given intermittent injections of active PTH analogs gain bone mass. Several other hormones, although not primarily involved in maintaining calcium homeostasis, have significant effects on calcium balance and bone health. Estrogens play an important role in maintaining bone mass in both women and men largely by inhibiting bone resorption. In menopausal women, and also in elderly men, serum levels of estrogen decline, leading to increased bone loss and an increased risk of osteoporosis. The administration of estrogen or SERMs to postmenopausal women can increase BMD and reduce fracture risk. Testosterone and other androgens promote bone formation by acting directly on ARs located on osteoblasts. Testosterone is also aromatized to estradiol in peripheral tissues and can thus have indirect estrogen-like effects on bone. The administration of testosterone to young men with testosterone deficiency clearly promotes increased bone mass, but the skeletal benefits in elderly men are less certain. Glucocorticoids, including a number of synthetic derivatives used to suppress inflammatory and immune-mediated diseases, have deleterious effects on bone health. When administered at higher doses for several months, glucocorticoid derivatives lead to a loss of bone mass, largely through their direct inhibitory effects on osteoblast activity. They also reduce net calcium absorption from the GI tract. Glucocorticoid-induced osteoporosis is the most common form of drug-induced osteoporosis. Not only bone mineral content, but also bone microarchitecture is adversely affected, leading to increased fracture risk at a higher threshold of BMD. Dietary supplementation with calcium and vitamin D, as well as the use of bisphosphonates and other antiosteoporosis drugs, are effective in reducing fracture risk in patients who must be continued on glucocorticoid therapy. Thyroid hormones, T3 and T4, play a physiological role in skeletal growth and maturation. In adult life, thyroid hormones increase the activity of both osteoblasts and osteoclasts, probably acting both directly and through cytokine mediators. At excessive thyroid hormone concentrations, bone resorption is typically increased disproportionately to bone formation, leading to reduced bone mass. The bone remodeling cycle is shortened. The risk of fracture is most significant in postmenopausal women, who already have a tendency to lose bone more rapidly. For this reason, it is particularly important to correct hyperthyroidism and to avoid overtreatment of hypothyroidism in postmenopausal women. GH and IGF-I have significant anabolic effects on the skeleton. The growth-promoting effects on GH are mediated by IGF-I, a polypeptide produced in peripheral tissues under the influence of GH. GH plays a critical role in promoting linear bone growth and overall bone mass in childhood. During adult life, GH secretion and serum concentrations of IGF-I decline progressively with increasing age. Adults with severe GH deficiency due to pituitary disease typically have a low bone mass with evidence of impaired bone formation. It is less clear whether naturally declining GH secretion in older adults plays a critical role in the development of senile osteoporosis. Several small clinical trials of GH therapy in osteoporotic patients have been reported, but as yet none have demonstrated fracture reduction. Further evidence of efficacy will be required before GH, IGF-I, and related anabolic peptides can be approved as therapeutic agents for the treatment of osteoporosis.

REFERENCES Abu, E.O., Horner, A., Kusec, V., et al. 2000. The localization of the functional glucocorticoid receptor in human bone. J Clin Endocrinol Metab 85: 883–889. Adler, R.A., Curtis, J., Weinstein, R.S., et al. 2008. Glucocorticoid-induced osteopororosis. In Osteoporosis, 3rd Edition, Marcus, R., Feldman, D., Nelson, D.A., and Rosen, C.J., eds. Elsevier Academic Press, Burlington, MA, Chapter 44, 1135–1166.

Hormone Actions in the Regulation of Calcium and Phosphorus Metabolism

107

Albright, F. 1947. Osteoporosis. Ann Intern Med 27: 861–882. Amory, J.K., Watts, N.B., Easley, K.A., et al. 2004. Exogenous testosterone or testosterone with finasteride increases bone mineral density in older men with low serum testosterone. J Clin Endocrinol Metab 89: 503–510. Arnold, A., Shattuck, T.M., Mallya, S.M., et al. 2002. Molecular pathogenesis of primary hyperparathyroidism. J Bone Miner Res 17 (Suppl 2): N30–N36. Baran, D. 2008. Thyroid hormone and the skeleton. In Osteoporosis, 3rd Edition, Marcus, R., Feldman, D., Nelson, D.A., and Rosen, C.J., eds. Elsevier Academic Press, Burlington, MA, Chapter 48, 1195–1202. Barrett-Connor, E., Mosca, L., Collins, P., et al. 2006. Effects of raloxifene on cardiovascular events and breast cancer in postmenopausal women. New Engl J Med 355: 125–137. Bauer, D.C., Ettinger, B., Nevitt, M.C., et al. 2001. Risk for fracture in women with low serum levels of thyroidstimulating hormone. Ann Intern Med 134: 561–568. Behre, H.M., Kliesch, S., Leifke, F., et al. 1997. Long term effects of testosterone therapy on bone mineral density in hypogonadal men. J Clin Endocrinol Metab 82: 2386–2390. Bravenboer, N., Holzmann, P., De Boer, H, et al. 1996. Histomorphometric analysis of bone mass and bone metabolism in growth hormone deficient adult men. Bone 18: 551–557. Breslau, N.A., Zerwekh, J.E., Nicar, M.J., et al. 1982. Effects of short term glucocorticoid administration in primary hyperparathyroidism: Comparison to sarcoidosis. J Clin Endocrinol Metab 54: 824–830. Bringhurst, F. R., and Leder, B.Z. 2006. Regulation of calcium and phosphate homeostasis. In Textbook of Endocrinology, 5th Edition, DeGroot, L.J., and Jameson, J.L., eds. Elsevier Saunders, Philadelphia, Chapter 74, 1465–1498. Bronner, F., Pannu, D., and Stein, W.D. 1986. An analysis of intestinal calcium transport across rat intestine. Am J Physiol 250: G561–G569. Brown, E.M. 1981. Lithium induces abnormal calcium-regulated PTH release in dispersed bovine parathyroid cells. J Clin Endocrinol Metab 52: 1046–1048. Brown, E.M. 1983. Four parameter model of the sigmoidal relationship between PTH release and extracellular ionized calcium in normal and abnormal parathyroid tissue. J Clin Endocrinol Metab 56: 572–581. Brown, E.M. 2002. The pathophysiology of primary hyperparathyroidism. J Bone Miner Res 17 (Suppl 2): N24–N29. Brown, E.M. 2007. The calcium-sensing receptor: Physiology, pathophysiology and CaR-based therapeutics. Subcell Biochem 45: 139–167. Brown, E.M. 2009. Anti-parathyroid and anti-calcium sensing receptor antibodies in autoimmune hypoparathyroidism. Endocrinol Metab Clin North Am 38: 437–445. Brown, E.M., Gamba, G., Riccardi, D., et al. 1993. Cloning and characterization of an extracellular Ca2+sensing receptor. Nature 366: 575–580. Brzozowski, A.M., Pike, A.C., Dauter, Z., et al. 1997. Molecular basis of agonism and antagonism in the oestrogen receptor. Nature 389: 753–758. Cetani, F., Picone, L., Cerrai, P., et al. 2000. Parathyroid expression of calcium-sensing receptor protein and in vivo parathyroid hormone–Ca(2+) set-point in patients with primary hyperparathyroidism. J Clin Endocrinol Metab 85: 4789–4794. Chambers, T.J., McSheehy, P.M., Thomson, B.M., et al. 1985. The effect of calcium regulating hormones and prostaglandins on bone resorption by osteoclasts disaggregated from neonatal rabbit bones. Endocrinology 116: 234–239. Chang, C., Kokontis, J., and Liao, S. 1988. Structural analysis of complementary DNA and amino acid sequences of human and rat androgen receptors. Proc Nat Acad Sci (USA) 85: 7211–7215. Chen, Q., Kaji, H., Kanatani, M., et al. 2004. Testosterone increases osteoprotegerin mRNA expression in mouse osteoblast cells. Horm Metab Res 36: 674–678. Colao, A., Di Somma, C., Pivonello, R., et al. 1999. Bone loss is correlated to the severity of growth hormone deficiency in adult patients with hypopituitarism. J Clin Endocrinol Metab 84: 1919–1924. Cosman, F., and Lindsay, R.L. 2008. Treatment with PTH peptides. In Osteoporosis, 3rd Edition, Marcus, R., Feldman, D., Nelson, D.A., and Rosen, C.J., eds. Elsevier Academic Press, Burlington, MA, Chapter 78, 1793–1808. Couse, J.F., and Korach, K.S. 1999. Estrogen receptor null mice: What have we learned and where will they lead us? Endocrinol Rev 20: 358–417. Cross, H.S., Debiec, H., and Peterlik, M. 1990. Mechanism and regulation of intestinal phosphate absorption. Miner Electrolyte Metab 16: 115–124. de Nijs, R.N., Jacobs, J.W., Algra A., et al. 2004. Prevention and treatment of glucocorticoid-induced osteoporosis with active vitamin D3 analogs: A review with meta-analysis of randomized controlled clinical trials including organ transplantation studies. Osteoporos Int 15: 589–602.

108

Diet, Nutrients, and Bone Health

den Dekker, E., Hoenderop, J.G., Nilius, B., et al. 2003. The epithelial calcium channels, TRPV5 & TRPV6: From identification towards regulation. Cell Calcium 33: 497–507. Diamond, T., Vine, J., Smart, R. et al. 1994. Thyrotoxic bone disease in women: A potentially reversible disorder. Ann Intern Med 120: 8–11. Diaz, R. and Brown, E.M. 2006. Familial hypocalciuric hypercalcemia and other disorders dud to calcium-sensing receptor mutations. In Endocrinology, 5th Edition, DeGroot, L.J., and Jameson, J.L., eds. Elsevier Saunders, Philadelphia, Chapter 81, 1595–1609. Dusso, A.S., Brown, A.J., and Slatopolsky, E. 2005. Vitamin D. Am J Physiol 289: F8–F28. Eriksen, E.F. 1986. Normal and pathological remodeling of human trabecular bone: Three dimensional reconstruction of the remodeling sequence in normals and in metabolic bone disease. Endocrinol Rev 7: 379–408. Ettinger, B., Black, D.M., Mitlak, B.H., et al. 1999. Reduction of vertebral fracture risk in post-menopausal women with osteoporosis treated with raloxifene: Results from a 3-year randomized clinical trial. Multiple Outcomes of Raloxifene Evaluation (MORE). JAMA 282: 637–645. Ettinger, B., Genant, H.K., and Cann, C.E. 1987. Postmenopsusal bone loss is prevented by treatment with lowdosage estrogen with calcium. Ann Intern Med 106: 40–45. Faber, J., and Galloe, A.M. 1994. Changes in bone mass during prolonged subclinical hyperthyroidism due to L-thyroxine treatment: A meta analysis. Eur J Endocrinol 130: 350–356. Fink, H.A., Ewing, S.K., Ensrud, K.E., et al. 2006. Association of testosterone and estradiol deficiency with osteoporosis and rapid bone loss in older men. J Clin Endocrinol Metab 91: 3908–3915. Flynn, R.W., Bonellie, S.R., Jung, R.T., et al. 2010. Serum thyroid-stimulating hormone concentration and morbidity from cardiovascular disease and fractures in patients on long-term thyroxine therapy. J Clin Endocrinol Metab 95: 186–193. Franklyn, J., Betteridge, J., Holder, R., et al. 1994. Bone mineral density in thyroxine treated females with or without a previous history of thyrotoxicosis. Clin Endocrinol 41: 425–432. Gavalas, N.G., Kemp, E.H., Krohn, K.J., et al. 2007. The calcium-sensing receptor is a target of autoantibodies in patients with autoimmune polyendorrine syndrome type 1. J Clin Endocrinol Metab 92: 2107–2114. Gennari, L., Nuti, R., and Bilezekian, J. 2004. Aromatase activity and bone homeostasis in men. J Clin Endocrinol Metab 89: 5898–5907. Gluck, O.S., Murphy, W.A., Hahn, T.J., and Hahn, B. 1981. Bone loss in adults receiving alternate day glucocorticoid therapy. Arthritis Rheum 24: 892–898. Grinspoon, S., Thomas, L., Miller, K., et al. 2002. Effects of recombinant human IGF-I and oral contraceptive administration on bone density in anorexia nervosa. J Clin Endocrinol Metab 87: 2883–2891. Guistina, A., Mazziotti, G., and Canalis, E. 2008. Growth hormone, insulin-like growth factors, and the skeleton. Endocr Rev 29: 535–559. Haden, S.T., Brown, E.M., Hutwitz, S., et al. 2000. The effects of age and gender on parathyroid hormone dynamics. Clin Endocrinol 52: 329–338. Haden, S.T., Stoll, A.L., McCormick, S., et al. 1997. Alterations in parathyroid hormone dynamics in lithiumtreated subjects. J Clin Endocrinol Metab 82: 2844–2848. Harman, S.M., Metter, E.J., Tobin, J.D., et al. 2001. Longitudinal effects of aging on serum total and free testosterone levels in healthy men. Baltimore Longitudinal Study of Aging. J Clin Endocrinol Metab 86: 724–731. Hebert, S.C. 1996. Extracellular calcium-sensing receptor: Implications for calcium and magnesium handling in the kidney. Kidney Int 50: 2129–2139. Hendy, G.N. 2000. Molecular mechanisms of primary hyperparathyroidism. Rev Endocr Metab Disord 1: 297–305. Hoenderop, J.G., Muller, D., Van Der Kemp, A.W., et al. 2001. Calcitriol controls the epithelial calcium channel in kidney. J Am Soc Nephrol 12: 1342–1349. Hoenderop, J.G., Nilius, B., and Bindels, R.J. 2002. Molecular mechanism of active Ca2+ reabsorption in the distal nephron. Annu Rev Physiol 64: 529–549. Holmes, S.J., Economouu, G., Whitehouse, R.W., et al. 1994. Reduced bone mineral density in patients with adult onset growth hormone deficiency. J Clin Endocrinol Metab 78: 669–674. Huang, J.C., Sakata, L.L., Pfleger, M., et al. 2004. PTH differentially regulates expression of RANKL and OPG. J Bone Miner Res 19: 235–244. Johansson, A.G., Burman, P., Westermark, K., et al. 1992. The bone mineral density in acquired growth hormone deficiency correlates with circulating levels of insulin-like growth factor I. J Internal Med 232: 447–452. Juppner, H., Gardella, T.J., Brown, E.M., et al. 2006. Parathyroid hormone and parathyroid hormone-related peptide in the regulation of calcium homeostasis and bone development. In Endocrinology, 5th Edition, DeGroot, L., and Jameson, J.L., eds. Elsevier Saunders, Philadelphia, Chapter 71, 1377–1417.

Hormone Actions in the Regulation of Calcium and Phosphorus Metabolism

109

Kaji, H., Sugimoto, T., Kanatani, M., et al. 1997. Dexamethasone stimulates osteoclast-like cell formation by directly acting on hemopoietic blast cells and enhances osteoclast-like cell formation stimulated by parathyroid hormone and prostaglandin E-2. J Bone Miner Res 12: 734–741. Kallner, G., and Petterson, U. 1995. Renal thyroid and parathyroid function during lithium treatment: Laboratory tests in 207 people treated for 1–30 years. Acta Psychiatr Scand 91: 48–51. Kanatani, M., Sugimoto, T., Sowa, H., et al. 2004. Thyroid hormone stimulates osteoclast differentiation by a mechanism independent of RANKL–RANK interaction. J Cell Physiol 201: 17–25. Kanis, J.A., Johansen, H.K., Oden, A., et al. 2004. A meta-analysis of prior corticosteroid use and fracture risk. J Bone Miner Res 19: 893–899. Karbach, U. 1992. Paracellular calcium transport across the small intestine. J Nutr 122: 672–677. Karga, H., Papapetrou, P.D., Korakovouni, A., et al. 2004. Bone mineral density in hyperthyroidism. Clin Endocrinol (Oxf) 61: 466–472. Keutmann, H.T., Sauer, M.M., Hendy, G.N., et al. 1978. The complete amino acid sequence of human parathyroid hormone. Biochemistry 17: 5723–5729. Khosla, S. 2001. The OPG/RANKL/RANK system. Endocrinology 142: 5050–5055. Kifor, O., Moore, F.D., Jr., Wang, P., et al. 1996. Reduced immunostaining for the extracellular Ca2+sensing receptor in primary and uremic secondary hyperparathyroidism. J Clin Endocrinol Metab 81: 3130–3131. Kurland, E.S., Chan, F.K., Rosen, C.J., et al. 1998. Normal growth hormone secretory reserve in men with idiopathic osteoporosis and reduced circulating levels of insulin-like growth factor-I. J Clin Endocrinol Metab 83: 2576–2579. Kurland, E.S., Rosen, C.J., Cosman, F., et al. 1997. Insulin-like growth factor-I in men with idiopathic osteoporosis. J Clin Endocrinol Metab 82: 2799–2805. Langlois, J.A., Rosen, C.J., Visser, M., et al. 1998. Associations between insulin-like growth factor I and bone mineral density in older women and men: The Framingham Heart Study. J Clin Endocrinol Metab 83: 4257–4262. Le Roith, D. Parrizas, M., and Blakesley, V.A. 1997. The insulin-like growth factor-I receptor and apoptosis. Implications for the aging process. Endocrine 7: 103–105. Leese, G.P., Jung, R.T., Guthrie, C., et al. 1992. Morbidity in patients on L-thyroxine: A comparison of those with a normal TSH to those with a suppressed TSH. Clin Endocrinol 37: 500–503. Lehmke, J., Bogner, U., Felsenberg, D., et al. 1992. Determination of bone mineral density by qualitative computed tomography and single photon absorptiometry in subclinical hyperthyroidism: A risk of early osteopaenia in post-menopausal women. Clin Endocronol 36: 511–517. Lemann, J., Jr., Piering, W.F., and Lennon, E.J. 1970. Studies of the acute effects of aldosterone and cortisol on the interrelationship between renal sodium, calcium and magnesium excretion in normal man. Nephron 7: 117–130. Levine, M.A. 2006. Hypoparthyroidism and pseudohypoparathyroidism. In Endocrinology, 5th Edition, DeGroot, L., and Jameson, J.L., eds. Elsevier Saunders, Philadelphia, Chapter 82, 1611–1636. Lindsay, R., Gallagher, J.C., Kleerekoper, M., et al. 2005. Bone response to treatment with lower doses of conjugated estrogens with and without medroxyprogesterone acetate in early postmenopausal women. Osteoporos Int 16: 372–379. Ljunghall, S., Johansson, A.G., Burman, P., et al. 1992. Low plasma levels of insulin-like growth factor-I (IGF-I) in male patients with idiopathic osteoporosis. J Internal Med 232: 59–64. Loffing, J., and Kaissling, B. 2003. Sodium and calcium transport pathways along the mammalian distal nephron: From rabbit to human. Am J Physiol Renal Fluid Electrolyte Physiol 284: F628–F643. Lubahn, D., Joseph, D., Sullivan, P., et al. 1988. Cloning of human androgen receptor complementary DNA and localization to the X chromosome. Science 240: 324–326. Lukert, B.P., and Raisz, L.G. 1990. Glucocorticoid-induced osteoporosis: Pathogenesis and management. Ann Intern Med 112: 352–364. Ma, Y.L., Cain, R.L., Halladay, S., et al. 2001. Catabolic effects of continuous human PTH (1–38) in vivo is associated with sustained stimulation of RANKL and inhibition of osteoprotegerin and gene-associated bone formation. Endocrinology 142: 4047–4054. Manolagos, S.C. 2000. Birth and death of bone cells: Basic regulatory mechanisms and implications for the pathogenesis and treatment of osteoporosis. Endocrinol Rev 21: 115–137. Melmed, S., and Jameson, J.L. 2006. Disorders of the anterior pituitary and hypothalamus. In Harrison’s Endocrinology, Jameson, J.L., ed. McGraw Hill Companies, New York, Chapter 2, 17–56. Merriman, H.L., La Tour, D., Linkhart, T.A., et al. 1990. Insulin-like growth factor-I and insulin-like growth factor-II induce c-fos in mouse osteoblastic cells. Calcif Tissue Int 46: 258–262.

110

Diet, Nutrients, and Bone Health

Mochizuki, H., Hakeda, Y., Wakatsuki, N., et al. 1992. Insulin-like growth factor-I supports formation and activation of osteoclasts. Endocrinology 131: 1075–1080. Mohan, S., and Baylink, D.J. 1991. Bone growth factors. Clin Orthop Relat Res 263: 30–48. Murer, H., Hernando, N., Forster, I., et al. 2000. Proximal tubular phosphate reabsorption: Molecular mechanisms. Physiol Rev 80: 1373–1409. Murray, R.D., Columb, B., Adams, J.E., et al. 2004. Low bone mass is an infrequent feature of the adult growth hormone deficiency syndrome in middle-age adults and the elderly. J Clin Endocrinol Metab 89: 1124–1130. Murray, T.M., Rao, L.G., Divieti, P., et al. 2005. Parathyroid hormone secretion and action: Evidence for discrete receptors for the carboxyl-terminal region and related biological action of carboxyl-terminal ligands. Endocrinol Rev 26: 78–113. Nagano, N. 2006. Pharmacological and clinical properties of calcimimetics: Calcium receptor activators that afford an innovative approach to controlling hyperparathyroidism. Pharmacol Ther 109: 339–365. Naveh-Many, T., Friedlaender, M.M., Mayer, H., et al. 1989. Calcium regulates parathyroid hormone messenger ribonucleic acid (mRNA), but not calcitonin mRNA in vivo in the rat. Dominant role of 1,25dihydroxyvitamin D. Endocrinology 125: 275–280. Neer, R.M., Arnaud, C.D., Zanchetta, J.R., et al. 2001. Effect of parathyroid hormone (1–34) on fractures and bone mineral density in postmenopausal women with osteoporosis. New Engl J Med 344: 1434–1441. Nielsen, H.A., Mosekilde, L., and Charles, P. 1979. Bone mineral content in hyperthyroid patients after combined medical and surgical treatment. Acta Radiol Phys Biol 18: 122–128. Ontjes, D.A. 2004. The role of estrogens and other steroid hormones in bone metabolism. In Calcium and Phosphorus in Health, Anderson, J.J.B., and Garner, S.C., eds. CRC Press, Boca Raton, FL, 207–233. Peaco*ck, M., Bolognese, M.A., Borofsky, M., et al. 2009. Cinacalcet treatment of primary hyperparathyrodsm: Biochemical and bone densitometric outcomes in a five-year study. J Clin Endocrinol Metab 94: 4860–4867. Pelletier-Morel, L., Fabien, N., Mouhoub, Y., et al. 2008. Hyperparathyroidism in a patient with autoimmune polyglandular syndrome. Intern Med 47: 1911–1915. Pepene, C.E., Kasperk, C.H., Pfeilshifter, J., et al. 2001. Effects of triiodothyronine on the insulin-like growth factor system in primary human osteoblastic cells in vitro. Bone 29: 540–546. Prince, R. Muchmore, D.B., and Siris, E. 2008. Estrogen analogues: Selective estrogen receptor modulators and phytoestrogens. In Osteoporosis, 3rd Edition, Marcus, R., Feldman, D., Nelson, D.A., and Rosen, C.J., eds. Elsevier Academic Press, Burlington, MA, Chapter 73, 1705–1723. Reid, I.R., Wattie, D.J., Evans, M.C., and Stapleton, J.P. 1996. Testosterone therapy in glucocorticoid-treated men. Arch Intern Med 156: 1173–1177. Richy, F., Schacht, E., Bruyere, O., et al. 2005. Vitamin D analogs versus native vitamin D in preventing bone loss and osteoporosis-related fractures: A comparative meta-analysis. Calcif Tissue Int 76: 176–186. Riggs, B.L., Khosla, S., and Melton, L.J. 2002. Sex steroids and the construction and conservation of the adult skeleton. Endocrinol Rev 23: 279–302. Riggs, B.L., Khosla, S., and Melton, L.J. 2008. Estrogen, bone homeostasis and osteoporosis. In Osteoporosis, 3rd Edition, Marcus, R., Feldman, D., Nelson, D.A., and Rosen, C.J., eds. Elsevier Academic Press, Burlington, MA, Chapter 40, 1011–1039. Rosen, C.J., and Adler, R.A. 1992. Longitudinal changes in lumbar bone density among thyrotoxic patients after attainment of euthyroidism. J Clin Endocrinol Metab 75: 1531–1534. Rosen, C.J., and Niu, T. 2008. Growth hormone and insulin-like growth factors: Potential applications and limitation in the management of osteoporosis. In Osteoporosis, 3rd Edition, Marcus, R., Feldman, D., Nelson, D.A., and Rosen, C.J., eds. Elsevier Academic Press, Burlington, MA, Chapter 79, 1809–1836. Ross, D.S., Neer, R.M., Ridgway, E.C., et al. 1987. Subclinical hyperthyroidism and reduced bone density as a possible result of prolonged suppression of the pituitary–thyroid axis with L-thyroxine. Am J Med 82: 1167–1170. Roussouw, J.E., Anderson, G.L., Prentice, R.L., et al. 2002. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: Principal results from the Women’s Health Initiative randomized controlled trial. JAMA 288: 321–333. Rubin, J., Ackert-Bickness, C.L., Zhu, L., et al. 2002. IGF-I regulates osteoprotegerin (OPG) and receptor activator of nuclear factor-kappa B ligand (RANKL) in vitro and OPG in vivo. J Clin Endocrinol Metab 87: 4273–4279. Saag, K.G., Zanchetta, J.R., Devogelaer, J.P., et al. 2009. Effects of teriparatide versus alendronate for treating glucocorticoid-induced osteoporosis: Thirty-six month results of a randomized, double-blind, controlled trial. Arthritis Rheum 60: 3346–3355.

Hormone Actions in the Regulation of Calcium and Phosphorus Metabolism

111

Saggese, G., Baroncelli, G.I., Bertelloni, S., et al. 1993. Effects of long-term treatment with growth hormone on bone and mineral metabolism in children with growth hormone deficiency. J Pediatr 122: 37–45. Shimada, T., Mizutani, S., Muto, T., et al. 2001. Cloning and characterization of FGF23 as a causative factor of tumor-induced osteomalacia. Proc Nat Acad Sci (USA) 98: 6500–6505. Slatopolsky, E., Finch, J., Denda, M., et al. 1996. Phosphorus restriction prevents parathyroid gland growth. High phosphorus directly stimulates PTH secretion in vitro. J Clin Invest 97: 2534–2540. Smith, E., Boyd, J., Frank, H., et al. 1994. Estrogen resistance caused by a mutation in the estrogen-receptor gene in a man. New Engl J Med 331: 1056–1061. Snyder, P.J., Biller, B.M.K., Zagar, A., et al. 2007. Effect of growth hormone replacement on BMD in adultonset growth hormone deficiency. J Bone Miner Res 22: 762–770. Snyder, P.J., Peachey, H., Hannoush, P., et al. 1999. Effect of testosterone treatment on bone mineral density in men over 65 year of age. J Clin Endocrinol Metab 84: 1966–1972. Sowers, M.F.-R., Jannausch, M., McConnell, D., et al. 2006. Hormone predictors of bone mineral density changes during the menopausal transition. J Clin Endocrinol Metab 91: 1261–1267. Stewart, A.F. 2005. Hypercalcemia associated with cancer. New Engl J Med 352: 373–379. Stewart, A.F., and Broadus, A.E. 2006. Malignancy-associated hypercalcemia. In Endocrinology, 5th Edition, DeGroot, L., and Jameson, J.L., eds. Elsevier Saunders, Philadelphia, Chapter 78, 1555–1564. Talmage, R.V. 1967. A study of the effect of parathyroid hormone on bone remodeling and calcium homeostasis. Clin Orthop Rel Res 54: 163–173. Talmage, R.V., Doppelt, S.H., and Fondren, F.B. 1976. An interpretation of acute changes in plasma 45Ca following parathyroid hormone administration to thyroparathyroidectomized rats. Calcif Tissue Res 22: 117–128. Tfelt-Hansen, J., and Brown, E.M. 2005. The calcium-sensing receptor in normal physiology and pathophysiology: A review. Crit Rev Clin Lab Sci 42: 35–70. Thakker, R.V., and Juppner, H. 2006. Genetic disorders of calcium homeostasis caused by abnormal regulation of parathyroid hormone secretion or responsiveness. In Endocrinology, 5th Edition, DeGroot, L., and Jameson, J.L., eds. Elsevier Saunders, Philadelphia, Chapter 76, 1511–1531. Tsai, M.J., and O’Malley, B.W. 1994. Molecular mechanisms of action of steroid/thyroid receptor superfamily members. Annu Rev Biochem 63: 451–486. Turner, C.H., Sato, M., and Bryant, H.U. 1994. Raloxifene preserves bone strength and bone mass in ovariectomized rats. Endocrinology 135: 2001–2005. Uzzan, B., Campos, J., Cucherat, M., et al. 1996. Effects on bone mass of long term treatment with thyroid hormones: A meta-analysis. J Clin Endocrinol Metab 81: 4278–4289. van Staa, T.P., Laan, R.F., Barton, I.P., et al. 2003. Bone density threshold and other predictors of vertebral fracture in patients receiving oral glucocorticoid therapy. Arthritis Rheum 48: 3224–3229. van Staa, T.P., Leufkens, H., and Cooper, C. 2007. The epidemiology of corticosteroid-induced osteoporosis: A meta-analysis. Osteoporos Int 13: 777–787. Vidal, O., Lindberg, M.K., Hollberg, K., et al. 2000, Estrogen receptor specificity in the regulation of skeletal growth and maturation in male mice. Proc Nat Acad Sci (USA) 97: 5474–5479. Wang, Y., Nishida, S., Elalieh, H.Z., et al. 2006. Role of IGF-I signaling in regulating osteoclastogenesis. J Bone Miner Res 21: 1350–1358. Weinstein, R.S., Jilka, R.L., Parfitt, A.M., et al. 1998. Inhibition of osteoblastogenesis and promotion of apoptosis of osteoblasts and osteocytes by glucocorticoids. Potential mechanisms of their deleterious effects on bone. J Clin Invest 102: 274–282. Weitzmann, M.N., and Pacifici, R. 2006. Estrogen deficiency and bone loss: An inflammatory tale. J Clin Invest 116: 1186–1194. Wiren, K. 2008. Androgens and skeletal biology: Basic mechanisms. In Osteoporosis, 3rd Edition, Marcus, R., Feldman, D., Nelson, D.A., and Rosen, C.J., eds. Elsevier Academic Press, Burlington, MA, Chapter 15, 425–449. Wood, P.J. 1992. The measurement of parathyroid hormone. Ann Clin Biochem 29: 11–21. Wuster, C., Abs, R., Bengtsson, B.A., et al. 2001. The influence of growth hormone deficiency, growth hormone replacement therapy, and other aspects of hypopituitarism on fracture rate and bone mineral density. J Bone Miner Res 16: 398–405. Wuster, C., Duckeck, G., Ugurel, A., et al. 1992. Bone mass of spine and forearm in osteoporosis and in German normals: Influences of sex, age, and anthropometric parameters. Eur J Clin Invest 22: 366–370. Yeh, J.K., Aloia, J.F., and Semla, H.M. 1984. Interrelation of cortisone and 1,25-dihydroxycholecalciferol on intestinal calcium and phosphate absorption. Calcif Tissue Int 36: 608–614.

7

Renal Regulation of Calcium and Phosphate Ions Philip J. Klemmer and John J.B. Anderson

CONTENTS Introduction..................................................................................................................................... 113 Calcium........................................................................................................................................... 113 Phosphate........................................................................................................................................ 117 Summary......................................................................................................................................... 117 References....................................................................................................................................... 118

INTRODUCTION An understanding of normal mineral metabolism is essential to understand the classical hormonal endocrine feedback system, which maintains optimal concentrations of calcium and phosphorus in the extracellular fluids while simultaneously serving to regulate external calcium and phosphate balances to facilitate skeletal health. These classical control systems interrelate with each other by means of negative feedback to maintain optimal extracellular concentrations of calcium and inorganic phosphate. Bone not only provides an abundant endogenous source of these minerals to supply extracellular fluids, but it also functions to buffer excess supplies of these minerals entering extracellular fluids from external dietary sources. These same regulatory systems also help maintain skeletal integrity during adult life as well as facilitate skeletal growth during childhood and adolescence. Appropriate mineral balance is maintained across wide ranges of dietary calcium and phosphorus intakes. The kidney plays a pivotal role in the maintenance of divalent ion homeostasis by virtue of a finely tuned excretory capacity as well as its ability to synthesize the active 1,25-dihydroxycholecalciferol [1,25(OH)2vitamin D], which actively regulates calcium absorption in the small intestine. During fasting periods, the kidneys typically reabsorb about 99% of calcium ions (Ca++) and approximately 95% of the filtered phosphate ions (HPO4=). The bulk of reabsorption of these ions takes place in the proximal convoluted tubule, where calcium reabsorption is coupled with sodium reabsorption. Following meals, however, these resorptive efficiencies are reduced a few percentage points, and urinary excretions of the two ions are increased to maintain physiological concentrations of these two ions in extracellular fluids. Over a 24-hour period, the external balances of the two elements are maintained while at the same time optimal ionic concentrations of calcium and phosphate are maintained in extracellular fluids. This chapter reviews the renal regulation of calcium and phosphorus in relation to dietary intakes of these two minerals in health. Renal calcium and phosphate homeostasis are reviewed in this chapter.

CALCIUM The normal 70-kg adult possesses approximately 1.2 kg of calcium, of which 99% is located within bone and only 1.3 g (0.1%) is located in extracellular fluids. The human kidney is in a unique position 113

114

Diet, Nutrients, and Bone Health

to regulate calcium homeostasis. Although by weight the two kidneys represent less than 2% of the total body weight, this organ receives more than 20% of the cardiac output each minute and produces over 180 L of glomerular filtrate each day. The selective reabsorption of greater than 99% of this glomerular filtrate leaves behind approximately 1.4 L of urine each day, which contains not only metabolic waste products but also sufficient calcium and phosphate ions to maintain appropriate balance. The retained divalent ions facilitate growth in childhood and adolescence and steady-state skeletal maintenance in adults. The kidney also helps to regulate optimal concentrations of ionic calcium and phosphorus in extracellular fluids needed for normal neuromuscular and organ function. Calcium intake from food and supplemental sources varies with race, age, and gender in the United States, but it ranges between 700 and 1000 mg/day for adults. Figure 7.1a and 7.1b depicts calcium and phosphate balance in the normal adult. Using the daily 1,000 mg of calcium intake estimate, it can be seen that the sum of urinary excretion (200 mg) and fecal loss (800 mg) approximates dietary calcium intake (1000 mg/day). Thus, in adult life, until approximately the age of 50 years, zero calcium balance, that is, no net calcium retention, is maintained. The principal site of regulation of calcium balance is the small intestine. The absorption fraction of calcium is approximately 20% to 30% in adults ingesting 1000 mg of calcium in their daily diets. Under conditions (a) Calcium metabolism Food 1000 mg/d

Influx 6000 mg/d

Outflux 6000 mg/d

Digestive juice calcium Intestine 1200 mg

200 mg Total absorbed intestinal calcium

Calcium pool (8.8–10.4)

400 mg

Feces 800 mg

Food 1400 mg/d

Urine 200 mg (b) Phosphate balance Bone Influx 1000 mg/d

Outflux 1000 mg/d

Digestive juice phosphate Intestine 1600 mg

200 mg Total absorbed intestinal phosphate

Phosphate pool (3.5–5.0)

1000 mg

Feces 400 mg

Urine 1000 mg

FIGURE 7.1  (A) Daily calcium balance; (B) daily phosphorus balance. (Adapted from Nordin, B.E.C., ed. 1976. Calcium, Phosphate and Magnesium Metabolism. Churchill Livingstone, Edinburgh; modified calcium fluxes based on Talmage, R.V. 1996. Foreword. In Calcium and Phosphorus in Health and Disease, Anderson, J.J.B., and Garner, S.C., eds. CRC Press, Boca Raton, FL.)

115

Renal Regulation of Calcium and Phosphate Ions

of low dietary calcium intake, the intestinal absorption fraction of calcium increases to maintain appropriate calcium balance. Intestinal calcium absorption is primarily regulated by the serum concentration of the active form of vitamin D, that is, 1,25(OH)2vitamin D (calcitriol), which becomes critically important during periods of low dietary calcium intakes. The adaptive increase in enteric calcium absorption which occurs during periods of low (<500 mg/day) dietary calcium intake is regulated by increases in the renal synthesis of 1,25(OH)2vitamin D derived from 25-hydroxycholecalciferol (calcidiol) in response to increases in parathyroid hormone (PTH) secretion. PTH responds to the ionic concentration of calcium in extracellular fluids by means of the calcium-sensing receptor in cell membranes of the parathyroid gland. During periods of low dietary calcium intake, slight decreases in ionic calcium concentration in extracellular fluid indirectly promote increases in absorption of calcium at the intestinal mucosa by stimulating PTH production, which in turn promotes the formation of increased 1-alpha-hydroxylase activity in the proximal tubule of the kidney. The higher level of active 1,25(OH)2vitamin D promotes greater transcellular calcium absorption and helps to maintain zero calcium balance in adults and appropriate positive calcium balance in growing children and adolescents. Reciprocal signaling also occurs during periods of high dietary calcium intake and thus prevents positive calcium balance, which is inappropriate in the adults whose closed epiphyses prevent further skeletal growth. Under conditions of extremely high calcium intake, nonregulated calcium uptake may occur by unregulated paracellular diffusion, that is, absorption between intestinal cells (enterocytes). The kidney is limited in its capacity to compensate for excess enteric calcium absorption (Figure 7.2). In the healthy adult, daily urinary calcium excretion rarely exceeds 4 mg/kg lean body weight. As a consequence of limited urinary calcium excretory capacity, extreme levels of calcium intake, such as exceeding 2300 mg/day, may lead to net positive calcium balance and soft tissue calcification, a condition originally called the milk-alkali syndrome (Hardt and River, 1923), now also referred to as the calcium loading syndrome. While the kidney regulates intestinal calcium absorption by formation of the active vitamin D hormone, that is, 1,25(OH)2vitamin D, it also reclaims most of the massive quantity of calcium within the glomerular filtrate by regulating tubular reabsorption. The kidneys filter about 10 g of calcium each day, and tubular segments reabsorb all but approximately 200 mg of this massive filtered load in men and about 120 mg in women. This level of urinary calcium excretion is maintained across a wide range of calcium intakes by virtue of the tightly regulated calcium absorption in the small intestine. In general, the net amount of calcium absorbed in the small intestine equals the daily urinary calcium excretion (Figure 7.1a).

Disturbing signal

1-Inadequate dietary calcium intake 2-Excessive loss of calcium

Controlling systems

Controlled system

Serum ionic calcium concentration

PTH 1,25 vitamin D

Negative feedback

FIGURE 7.2  Serum ionic calcium concentration is maintained within a very narrow range (average 1.1 mM) throughout life by the rapid effects of PTH on calcium flux at the bone envelope and tubular calcium reabsorption in the kidney. Active vitamin D, 1,25 dihydroxycholecalciferol, acts more slowly with respect to maintaining this physiologic set point for ionic calcium concentration in extracellular fluids. (Adapted from Lobaugh, B. 1996. Blood calcium and phosphorus regulation. In Calcium and Phosphorus in Health and Disease, Anderson, J.J.B., and Garner, S.C., eds. CRC Press, Boca Raton, FL.)

116

Diet, Nutrients, and Bone Health

The regulators of renal calcium excretion include PTH, 1,25(OH)2vitamin D, and other hormones including, but not limited to, insulin, glucagons, and growth hormone. Expansion of the extracellular compartment by high dietary salt intake increases urinary calcium excretion. Sustained excessive intakes of sodium may lead to hypercalciuria and nephrolitiasis. Whether the high dietary sodium contributes to negative external calcium balance, bone loss, and osteoporosis has not been established. Even modest reductions in GFR cause a significant lowering of urinary calcium excretion. Diets rich in animal protein have been shown to cause hypercalciuria, possibly by increasing endogenous acid production which inhibits tubular calcium reabsorption. The ionic calcium concentration in blood and extracellular fluids is tightly regulated at a level between 1.17 and 1.33 mmol/L primarily by the action of PTH on bone. Bone contains greater than 99.5% of the body’s calcium reserve, mostly in the form of calcium apatite which forms structural bone. Although bone resorption and formation are normally tightly coupled (1:1), the calcium in apatite crystals is not readily available to buffer minute-to-minute changes in ionic calcium concentration in the extracellular fluid. Rather, the calcium ions in the bone fluid compartment serve to buffer minute-to-minute perturbations in the calcium ion concentration. A readily available source of exchangeable calcium, however, is found in the bone-lining space (bone envelope or bone compartment) which covers practically all bone surfaces (Talmage, 1996). Ionic calcium held in this reservoir is readily mobilized in response to minute-to-minute changes in the concentration of PTH (Figure 7.3). When ionic serum calcium falls below a physiologic level, the calcium-sensing receptors on the membranes of the parathyroid cells release PTH. This higher level of PTH secretion promotes the release of ionic calcium sequestered in the bone envelope until the serum ionic calcium concentration returns to the physiologic level. Reciprocal changes occur during periods of postprandial calcium absorption, and as the PTH level decreases, the bone Efflux— 6000 mg/day Influx— 6000 mg/day

Bone resorption— 400 mg/day

Bone formation— 400 mg/day

FIGURE 7.3  Exchange of calcium between bone and blood. The Ca in blood (and all extracellular fluid) undergoes constant change; both were depleted via Ca ion influx to bone (as Ca and Pi precipitate onto the surface of hydroxyapatite as calcium phosphate crystals) as blood passes through bone and via calcium ion influx from the bone fluid compartment. The tremendous influx of Ca (and Pi) into bone is balanced by an equal amount of Ca (and Pi) ion returned to blood (Ca efflux) by the action of the bone-lining cells and other osteoblast-derived bone cells. The amount of Ca and Pi ions leaving blood through bone formation is at least an order of magnitude less than that of the influx of these ions that results from exceeding the extracellular fluid Ca–Pi solubility product in relation to bone mineral. Similarly, the amount of Ca and Pi ions that can be returned from bone to blood by osteoclastic resorption is quite small, that is, about an order of magnitude lower, and should equal the amount of ions required for bone formation in an individual whose skeleton is undergoing no net change in bone mass, that is, in balance. (Adapted from Talmage, R.V. 1996. Foreword. In Calcium and Phosphorus in Health and Disease, Anderson, J.J.B., and Garner, S.C., eds. CRC Press, Boca Raton, FL.)

Renal Regulation of Calcium and Phosphate Ions

117

envelope buffers the rise in ionic calcium concentrations in extracellular fluids by removing calcium ions from the circulation. Thus, in healthy adults, zero calcium balance is maintained over the long term, whereas the set point of ionic calcium concentration in extracellular fluids is tightly regulated between 1.17 and 1.33 mmol/L by the action of PTH at the level of bone and, to a lesser degree, at the level of the renal tubule where PTH acts to increase calcium reabsorption. The major roles of PTH in calcium homeostasis are considered to be increasing the flux of calcium ions from the bone fluid compartment to extracellular fluids and regulating the synthesis of the active form of vitamin D, 1,25(OH)2vitamin D. Calcium homeostasis is thus maintained across a wide range of dietary calcium intakes by actions of the kidneys, small intestine, skeleton, and parathyroid glands.

PHOSPHATE Phosphate homeostasis in the healthy adult differs significantly from that of calcium in that the regulation of serum phosphate concentrations and maintenance of whole-body phosphate balance occur principally at the level of the proximal renal tubule rather than in the small intestine, as is the case for calcium (Anderson et al., 2006). In essence, serum phosphate concentration is regulated through regulated renal excretion rather than by means of regulated intestinal absorption (Lemann et al., 1979). In recent years, dietary phosphorus intakes have increased as a consequence of phosphate additives in processed foods (Bell et al., 1977). Figure 7.1b depicts phosphate balance in the normal adult with a phosphorus intake of 1400 mg/day, renal excretion of 900 mg/day, and fecal loss of 500 mg/day. Since the total intake and the combined urinary and fecal outputs in health are equal, zero metabolic balance is achieved by adults. Approximately 60% to 70% of dietary phosphorus is absorbed by the small intestine via a sodium/phosphate transporter in the mucosa. Phosphate absorption in the small intestine occurs largely independent of active vitamin D regulation. Instead, regulation of renal phosphate excretion occurs at the level of the proximal tubule in the kidney in response to PTH and FGF-23, a phosphatonin hormone produced by osteocytes in bone. FGF-23 has a phosphaturic effect, and it also reduces the activity of the renal enzyme, 1a-hydroxylase, which, in turn, reduces production of 1,25(OH)2vitamin D (Quarles, 2008). After a large phosphate intake from dietary protein sources and/or processed foods containing phosphate additives, PTH and FGF-23 concentrations increase in extracellular fluids. These hormonal increases lead to a reduction in proximal tubular phosphate reabsorption from the glomerular filtrate, thus enhancing urinary phosphate excretion. Thus, these hormonal actions restore balance between dietary phosphate intake and urinary and fecal phosphate excretion. The set point for ionic phosphate concentration in the extracellular fluids is less tightly defended than that of ionic calcium. Serum phosphate concentrations in adults range between 2.5 and 4.5 mg/ dL (0.83–1.5 mmol/L) in fasting adults and demonstrate diurnal and postprandial variations (Calvo et al., 1991). Phosphate flux across the bone envelope contributes to the maintenance of serum phosphate at physiologic concentrations, but this flux has not been adequately investigated.

SUMMARY Mineral homeostasis, bone metabolism, and kidney function are interrelated by common hormonal control systems. The kidneys tightly regulate the concentrations of calcium and inorganic phosphate ions in the extracellular fluids as well as the total body external balance of these ions across a wide range of dietary calcium and phosphorus intakes. With respect to calcium, these dual roles are achieved by the regulation of calcium flux from the skeleton by PTH and the regulation enteric calcium absorption by the hormonally active form of vitamin D produced in the kidney, 1,25(OH)2vitamin D. In contrast, phosphorus homeostasis is regulated almost entirely at the level of the kidney, specifically in the proximal convoluted tubule, which responds to variable levels of PTH and phosphatonins, including FGF-23, to alter tubular absorption of phosphorus from glomerular filtrate. The importance of adequate kidney function in the maintenance of normal mineral

118

Diet, Nutrients, and Bone Health

homeostasis, cardiovascular health, and skeletal integrity becomes evident when chronic kidney disease develops. These issues, discussed in Chapter 33, make us mindful of the exquisite precision of natural selection in the evolution of these tightly regulated control systems.

REFERENCES Anderson, J.J.B., Klemmer, P.J., Watts, M.E.S., et al. 2006. Phosphorus. In Present Knowledge of Nutrition, vol. I, 9th ed., Bowman, B.A., and Russell, R.M., eds. ILSI Press, Washington, D.C., 383–399. Bell, R.R., Draper, H.H., Tzeng, D.Y., et al. 1977. Physiological responses of human adults to foods containing phosphate additives. J Nutr 107: 42–50. Calvo, M.S., Eastell, R., Offord, K.P., et al. 1991. Circadian variation in ionized calcium and intact parathyroid hormone: Evidence for sex differences in calcium homeostasis. J Clin Endocrinol Metab 72: 69–76. Hardt, L., and River, A. 1923. Toxic manifestations following the alkali treatment of peptic ulcer. Arch Intern Med 31: 171–180. Lemann, J., Adams, N,D., and Gray, R.W. 1979. Urinary calcium excretion in human beings. New Engl J Med 301: 535–541. Lobaugh, B. 1996. Blood calcium and phosphorus regulation. In Calcium and Phosphorus in Health and Disease, Anderson, J.J.B., and Garner, S.C., eds. CRC Press, Boca Raton, FL, 27–43. Nordin, B.E.C., ed. 1976. Calcium, Phosphate and Magnesium Metabolism. Churchill Livingstone, Edinburgh. Quarles, L.D. 2008. Endocrine functions of bone in mineral metabolism regulation. J Clin Invest 118: 3820–3828. Talmage, R.V. 1996. Foreword. In Calcium and Phosphorus in Health and Disease, Anderson, J.J.B., and Garner, S.C., eds. CRC Press, Boca Raton, FL.

Part II Effects of Specific Nutrients on Bone

8

Calcium and Bone John J.B. Anderson, Sanford C. Garner, and Philip J. Klemmer

CONTENTS Introduction..................................................................................................................................... 121 Dietary Sources of Calcium............................................................................................................ 122 Dietary Reference Intakes for Calcium........................................................................................... 123 Recommended Calcium Intakes in the United States and Canada........................................ 125 Calcium and PBM Accrual.................................................................................................... 126 Adult Calcium Needs............................................................................................................. 128 Elderly Calcium Needs.......................................................................................................... 129 Calcium Metabolism....................................................................................................................... 130 Intestinal Absorption of Calcium........................................................................................... 130 Parathyroid Hormone and the Regulation of Serum Calcium Concentration....................... 131 Urinary Excretion of Calcium............................................................................................... 132 Calcium Balance.................................................................................................................... 132 Effects of Dietary Calcium and Phosphorus Ratios on Calcium Homeostasis............................... 133 Lactose Intolerance and Inadequate Calcium Intake...................................................................... 134 Deficient Calcium Intakes and the Need for Calcium Supplements............................................... 134 Potential Calcium Toxicity: Arterial Calcification and Renal Stones............................................. 135 Summary......................................................................................................................................... 136 References....................................................................................................................................... 136

INTRODUCTION In addition to fulfilling the needs for calcium ions required in numerous intracellular functions as well as for the regulation of blood clotting (hemostasis), practically all of the remainder of the body’s calcium exists in skeletal salts that support the body, enable ambulation, and protect internal organs. A fixed amount of calcium forms the teeth which, after formation, remain static in the oral cavity and which, unlike bone, do not participate in calcium metabolism. In the growing years, dietary calcium contributes greatly to the development of peak bone mass (PBM). Following the achievement of PBM in the second or third decade of life, dietary calcium is needed to replace calcium lost from bone tissue as part of the normal dynamic turnover of the skeleton. Adult bone mineral content (BMC) and bone mineral density (BMD) are better maintained by an adequate amount of calcium in the diet. Adequate amounts of dietary calcium are necessary for the maintenance of bone mass and calcium metabolic balance, that is, equality between dietary calcium intake and excretion of calcium in the stool and urine. Although dietary intake of calcium, therefore, has a major role in supplying this essential nutrient, hormonal regulation at the intestine and kidneys maintains neutral (zero) calcium balance over a wide range of dietary calcium intakes (Peaco*ck, 2010) (see also Chapter 6). Low or inadequate consumption of calcium may be partially compensated for by a healthy vitamin D status. Vitamin D enhances intestinal calcium absorption and thereby assists in adaptation to a low calcium intake. Dietary phosphate also interacts with calcium, especially when the phosphorus intake is excessive relative to the calcium intake, and this interaction may be detrimental to bone 121

122

Diet, Nutrients, and Bone Health

health. This chapter examines the dietary contributions of calcium and the metabolic and skeletal handling of calcium ions. In addition, the deleterious effects of deficient and excessive intakes of calcium are briefly discussed.

DIETARY SOURCES OF CALCIUM The amounts of calcium per serving of major calcium-containing foods are given in Table 8.1. A benefit of calcium-rich foods is that they provide good quantities of many additional nutrients besides calcium. Figure 8.1 illustrates the nutrient content of several calcium-rich food sources in comparison with a calcium carbonate supplement and a cola-type of soda drink. Except for dairy foods, few commonly consumed foods in the United States and most other Western nations contain much calcium (Table 8.2). Because of its limited distribution, consuming calcium in sufficient amounts is a challenge for many, and particularly for strict vegetarians or vegans (Havala Hobbs and Anderson, 2009). Based on food consumption surveys, approximately 60% of the calcium consumed by adults in the United States is derived from milk and dairy foods (Looker et al., 1993; Fleming and Heimbach, 1994; Ervin et al., 2004; Anon, 2009). The remainder of calcium is obtained from baked goods, dark green leafy vegetables, and relatively few other foods. For some elderly, who consume little or no dairy products, most of their calcium is obtained from breads and baked goods, which provide some of this mineral but not enough to meet the amount, that is, Dietary Reference Intake (DRI), recommended by the Food and Nutrition Board, Institute of Medicine (IOM) (1997). Fortified food products and beverages, such as orange juice, also provide calcium. The dietary reference intakes (DRIs) for calcium are given in Table 8.3. A cup (8 oz) of practically any kind of cow’s milk provides approximately 300 mg of calcium. This value is the standard to which other foods are compared. Table 8.2 lists the calcium content of selected food items from each food group. The calcium content of vegetarian diets, especially those consumed by vegans, may be sufficient in calcium to maintain good nutritional health if serving sizes are increased to compensate for lower calcium density in these foods (Weaver et al., 1999). Careful selection of calcium-containing plant foods by strict vegetarians is necessary for meeting calcium DRIs (Havala Hobbs and Anderson, 2009). Low fat dairy products, such as yogurts, are excellent sources of calcium; roughly 30% to 50% more calcium is provided in one serving of yogurt compared with that in a cup of milk. Many young TABLE 8.1 Approximate Amounts of Calcium per Serving of Major Calcium-Containing Foods Food Milk Yogurt Cheddar cheese American cheese Cottage cheese Tofu, calcium-set Salmon, canned, bones Kale Collards

Serving

Calcium, mg

8 oz 1 cup 1 oz 1 slice ½ cup 4 oz 3 oz

300 350 200 175 65 145 165

½ cup ½ cup

100 180

Source: USDA National Nutrient Database for Standard Reference, http://www.nal.usda.gov/fnic/ foodcomp/search/

123

Calcium and Bone

10

10

10

10

10

(b) 50 40 30 20

50

Soda, Cola type 16 ounces

40

Calcium carbonate 1 tablet

30

Vitamin B12

20

Riboflavin

20

30

Energy

20

30

Broccoli 6 ounces

40

Vitamin A

40

Phosphorus

30

50

Ice cream 8 ounces

Protein

50

Yogurt, fruit flavored 8 ounces

Percent of RDA

40 Percent of RDA

20

10

50

Percent of RDA

20

30

Calcium

20

30

Tofu, raw 4 ounces

40 Percent of RDA

30

50

Cheese, American 1 ounce

40

Percent of RDA

Percent of RDA

40

50

Milk, 2% fat 8 ounces Percent of RDA

50

Percent of RDA

(a)

20

10

10

FIGURE 8.1  Nutrient content of selected calcium-rich sources in comparison with a calcium carbonate supplement and a cola-type soft drink. Calcium-rich foods provide more nutrients than calcium. Note the number of other nutrients provided by the food items rich in calcium, but not by a supplement containing calcium alone or by a cola-type soft drink. (Adapted from Anderson, J.J.B., et al. 2005. Nutrition and Health, Carolina Academic Press.)

women find yogurt more appealing because they are concerned about dietary fat and its potential contribution to weight gain. Therefore, knowledge of the nutrient composition of various dairy products should help concerned individuals ingest enough calcium while at the same time avoid too much phosphorus or fat. The actual intakes of calcium and phosphorus, reported in a recent National Health and Nutrition Examination Survey (NHANES) survey based on assessed intakes of a large sample of U.S. adults, have been summarized (see Table 1.2). Compared with the age-specific DRIs, deficits in calcium intakes typically occur beyond 11 years of age, especially in females. Phosphorus intakes, which have been found to be considerably higher than calcium in the NHANES survey, are much higher than their DRIs. Based on NHANES data (Ervin etal., 2004), the adult Ca:P intake ratio of women is approximately 700:1200 (1:1.7 or 0.6:1) (Ervin et al., 2004), compared with the recommended ratio of 1:0.6 (Food and Nutrition Board, IOM, 1997).

DIETARY REFERENCE INTAKES FOR CALCIUM Calcium is often called a threshold nutrient because intakes need to meet a “threshold” or “window” of intake that satisfies body requirements at all stages of the life cycle. Although age-specific

124

Diet, Nutrients, and Bone Health

TABLE 8.2 Calcium Content per Serving of Most Calcium-Containing Foods, According to Food Groups Milk and Dairy Products

Calcium (mg)

Milk, whole (8 oz) Milk, skim (8 oz) Yogurt, fruit, low fat (6 oz) Ice cream (½ cup) Ice milk (½ cup) Colby cheese (1 oz) American cheese (1 oz) Cheddar cheese (1 oz) Swiss cheese (1 oz) Colby cheese (1 oz) Edam cheese (1 oz) Mozzarella cheese (1 oz) Mozzarella cheese, low fat (1 oz) Muenster cheese (1 oz) Provolone cheese (1 oz) Cottage cheese, creamed (½ cup) Ricotta cheese, full milk (½ cup)

276 301 258   92 107 194 159 204 224 194 207 143 222 203 214   68 255

Protein Foods Tofu, firm, nigari (4 oz) Salmon, canned, bones (3 oz) Sardines, with bones (3 oz) Almonds (2 oz) Fruits and Vegetables Orange (1 medium) Greens, kale (½ cup) Prunes (6) Green beans (½ cup) Squash, winter (½ cup) Cereal Grains and Bakery Products White bread, enriched (1 slice) Whole wheat bread (1 slice) Cornbread, enriched (2½″ sq) Pancake, enriched, (4″ diam.) Tortilla, corn, enriched (6″ diam.)

Calcium (mg) 198 212 325 141   52   90   25   28   23   45   36   88   68   19

Source: USDA National Nutrient Database for Standard Reference, http://www.nal.usda.gov/fnic/foodcomp/ search/

TABLE 8.3 Dietary Reference Intakes for Calcium Age Range 0 to 6 months 7 to 12 months 1–3 years 4–8 years 9–18 years 19–50 years 51–70 years >70 years

DRI (mg/day)   NA   NA   700 1000 1300 1000 1000 (M); 1200(F) 1200

UL (mg/day) 1000 1500 2500 2500 3000 2500 2000 2000

Notes: NA = not available. Values for pregnant and lactating women are not included.     Dietary Reference Intakes (DRIs) and tolerable upper intake levels (ULs) are the same for males and females. Source: Food and Nutrition Board, Institute of Medicine, 2011. Dietary Reference Intakes for Calcium and Vitamin D. National Academy Press, Washington, DC.

125

Calcium and Bone Input (1000 mg Ca)

Output

GUT

Intestinal absorption

Kidneys

Intestinal secretion

Skin

Urinary excretion

Net balance

Shedding of skin cells

A)

200 mg

(150 mg +

100 mg

+

15 mg)

=

–65 mg

B)

300 mg

(150 mg +

135 mg

+

15 mg)

=

0 mg

C)

400 mg

(150 mg +

170 mg

+

15 mg)

=

+65 mg

FIGURE 8.2  Three scenarios of calcium balance. Schematic diagrams illustrating three states of calcium balance that occur across the life cycle. In later life (A), 1000 mg of calcium a day may lead to a net negative balance; in early adult life (B), 1000 mg of calcium may be adequate and result in zero balance; and during growth (C), 100 mg of calcium per day may place children and adolescents in positive calcium balance.

calcium requirements cannot easily be determined, even by modern research methods, estimates of the amounts of calcium needed have been established using metabolic balance studies (Matkovic and Heaney, 1992; Hunt and Johnson, 2007) and skeletal calcium accrual analyses based on measurements of total body BMC (Vatanparast et al., 2010). Usual calcium intakes below the threshold or window and those above the threshold or window are considered unhealthy; a schematic diagram (Figure 8.2) illustrates these three conditions of calcium intake.

Recommended Calcium intakes in the United States and Canada The recommended calcium intakes across the life cycle, part of the DRIs, are set forth for the various stages stage of the life cycle. The calcium DRIs and tolerable upper limit intake levels (ULs) are set at estimated daily amounts across the life cycle (Food and Nutrition Board, IOM, 1997) (see Table 8.3). The DRIs for those boys and girls 8 to 19 years of age may be more than enough for optimal skeletal development. Few females in this age range consume 1,300 mg on a regular daily basis, yet most girls achieve reasonable skeletal development and height with lower intakes than their DRI (Bonjour et al., 1997) (see below). Precise requirements for calcium are not known for either males or females, and these recommendations (DRIs) may be set on the high side, especially for females, to optimize PBM development both during adolescent growth and the subsequent skeletal consolidation period of early adulthood, that is, from approximately 19 to 30 years of age (Hunt and Johnson, 2007). Therefore, the calcium DRIs across the life cycle appear to provide a considerable margin of safety, especially for women who typically develop less skeletal mass than do men. Figure 8.3 gives a scenario in which 1000 mg/day may be sufficient for optimal skeletal development from 8 to 19 years of age. To illustrate the discrepancy between calcium intake data from a U.S. Department of Agriculture survey and DRIs for calcium, the intakes are plotted against age in Figure 8.4. The median calcium intakes of females from 11 years and beyond are much lower than the age-specific DRIs. Males do better, but they still consume less than the DRIs. The typical low intakes of calcium, especially intakes lower than approximately 50% of the DRI, by peripubertal females suggest that PBM accrual may not be optimal in these girls. Similar low calcium intakes are common in similarly

126

Diet, Nutrients, and Bone Health ↑ Excess 1200

Window of optimal calcium intake

Ca, mg/dL 1000

800

↓ Deficiency

FIGURE 8.3  Optimal range of calcium intake. Scenario in which 1000 mg a day may be sufficient for optimal skeletal development from 8 to 19 years of age, but the current recommendations are 1300 mg/day for boys and girls. Many children and adolescents consume less than 800 mg/day, which implies that these individuals are consuming diets deficient in calcium. 1600

Adequate intakes (AIs)

Calcium, mg per day

1400

1300 mg/d

1200 1000

1200 mg/d 1000 mg/d

800

800 mg/d

600 400

500 mg/d

200 0

1–3

4–6

7–10 11–14 15–18 19–24 25–50 51–64

65+

Age, years

FIGURE 8.4  Discrepancies between calcium intakes (USDA survey) and AIs for calcium across the life cycle. Calcium intakes are plotted against age. AIs or Adequate Intakes were the IOM recommendations for calcium from 1997 to 2011. (Adapted from Anderson, J.J.B., et al. 2005. Nutrition and Health, Carolina Academic Press.)

aged females of China and other Asian nations, although Japanese girls have been improving their calcium intakes in recent decades (see Chapter 30).

Calcium and PBM Accrual Peripubertal calcium intakes have a powerful impact on PBM development, which is typically achieved by approximately the age of 20 years or somewhat later in the third decade, that is, at 30 years or so, when bone consolidation is completed. The World Health Organization recognizes the achievement of PBM and further skeletal mass accrual by the age of 30 years, since the 20- to 29-year-old means of BMC and BMD are used as the healthy adult standards for dual energy x-ray absorptiometry (DXA) measurements (Kanis et al., 1994). DXA scans at later ages are typically compared with those of young healthy adult mean values for BMD. Thus, as part of a healthy diet,

127

Calcium and Bone

calcium intakes during the first two decades of life set the stage for optimal skeletal development, and this optimal PBM accrual is presumed to serve until late in life when declines in bone mass occur (Figure 8.5) (see also Chapter 24). The concept that a greater PBM may help delay the late-life onset of osteopenia and osteoporosis, and its adverse consequences, has been argued for many years, although direct proof has not been established. Dietary factors, such as calcium, phosphorus, vitamin D, protein, and others, have long been considered as significant determinants of PBM (Heaney et al., 2000; Nieves, 2005; Vatanparast and Whiting, 2006). Results of many research studies have supported the calciumrelated skeletal gains in BMC during the peripubertal decade from roughly 8 to 18 years, starting earlier in girls than boys but ending later in boys, depending on pubertal development (Bonjour et al., 1991, 1997; Johnston et al., 1992; Lloyd et al., 1992, 1993; Abrams and Stuff, 1994; Matkovic et al., 1994, 2004; Young et al., 1995; Weaver et al., 1995, 2007 ; Teegarden et al., 1995; Cadogan et al., 1997; Jackman et al., 1997; Martin et al., 2007; Bailey et al., 2000; Abrams, 2005). Also, two studies have found that low calcium intakes during this important period of skeletal growth may contribute [to] fractures in children and adolescents (Goulding et al., 1998; Wyshak and Frisch, 1994). Studies of dizygotic and monozygotic twin subjects, which benefit from a reduction in genetic variance, have been instrumental in demonstrating the skeletal gains in mass and cross-sectional area or bone size of long bones of the twin receiving the calcium supplement in both the prepubertal and postpubertal periods (Johnston et al., 1992; Young et al., 1995) and in strength parameters (see Chapter 25). Twins, especially monozygotic pairs, have remarkably similar PBM development. This similar development illustrates a strong genetic contribution to PBM, perhaps as high as 80%, but dietary factors and other environmental factors still contribute to this accrual of bone mass (Slemenda et al., 1992). Gains of bone mass resulting from the addition of calcium supplements, especially during the growth years of early life, may be lost after supplementation ceases. Two reports, one of singleton 12-year-old girls and one of 10- to 12-year-old twins of both genders, have shown that the BMD that was gained by the calcium-supplemented children over 2 to 4 years was subsequently (b)

10

20 Age

30

Positive (+)

+

Average (0)

Negative (–)

40

Bone mass

Bone mass

(a)

30

Fracture risk range

40

50

60

70

80

90

Age

FIGURE 8.5  Early gain and later loss of bone mass across the life cycle. The early gain of bone mass in men and women is shown in the left-hand side of the figure up to about the age of 30 years when peak bone mass is achieved. The peak is higher for men than that for women. The later loss of bone mass is illustrated by the right-hand side of the figure from the age of 50 years and over. After the menopause, at approximately the age of 50 years, in females, and somewhat later in males, the loss of bone mass begins, and it continues at a slow rate of loss after the first postmenopausal decade in women and at about the age of 65 years in men for the rest of an individual’s life. Slowing of the bone loss by behavioral factors such as diet and activity retards or delays the onset of osteoporosis. (Adapted from Anderson, J.J.B., et al. 2005. Nutrition and Health, Carolina Academic Press.)

128

Diet, Nutrients, and Bone Health

lost after cessation of supplements as the treated individuals achieved approximately the same BMD values as their nonsupplemented comparators over the next few years (Lloyd et al., 1994; Slemenda et al., 1997). The conclusion of these short-term dietary studies is that little or no residual benefit continue after calcium supplements stop. Longer periods of calcium intakes that are at least as high as the DRI for growing children have not been examined to assess bone gains that are retained well into adulthood, the presumed benefit of high calcium intakes. In contrast, gains in bone mass related to physical activity during skeletal development are considered to continue into adulthood (see Chapter 23). Calcium supplementation along with an exercise program has also resulted in improved PBM as measured by BMC (Barr and McKay, 1996; French et al., 2000; Welch and Weaver, 2005; Specker and Vukovich, 2007). BMC is typically chosen because BMD is not as useful a measure in growing children as BMC. A major benefit of regular physical activity is the improvement of bone quality, especially of trabecular or cancellous bone tissue, and dietary calcium availability permissively facilitates this bone consolidation. Gains of bone size in response to DRIs of calcium occur in most ethnic groups (Weaver et al., 2007). Compared with whites and Asians, African Americans have higher PBM and a lower prevalence of osteoporosis, although gender differences are still evident as fracture rates are greater in black females than those in black males (see Chapter 29). A reason for this high bone mass (and muscle mass) among blacks relates in large part to unknown genetic factors. Because so many blacks are lactose intolerant and refrain from dairy products, they have difficulty obtaining the recommended amounts of calcium for skeletal development during childhood and adolescence. Yet, on average, they exhibit no apparent deleterious effects with respect to the development of PBM. The mechanism or mechanisms that allow blacks to adapt to low calcium intakes and develop superior BMD remain unknown, but a slower action of PTH has been suggested as a reason for increased bone mass. Reasonable calcium intakes, that is, DRI amounts (Food and Nutrition Board, IOM, 1997), however, are still recommended for black adolescents to optimize PBM. A greater “bone” bank may help black women protect against or delay the development of osteoporosis, as their life expectancy is increasing in the United States (Arias, 2010).

Adult Calcium Needs Adult women and men need adequate amounts of calcium each day to maintain their skeletal mass (BMC) and density (BMD). The calcium recommendations are 1000 mg/day from 19 to 50 years and 1200 beyond the age of 50 years for both genders. Young adult women and men typically continue to gain bone mass via the process of consolidation until approximately 30 years (Halioua and Anderson, 1989). Women, however, begin losing some bone mass during their last decade prior to menopause, that is, during their 40s, as their ovarian estrogen production declines. Men do not lose much bone mass, if any, until a decade or more later when their androgen production decreases. Women who continue reasonably good physical activity during their 30s and 40s may maintain their bone mass for several years longer than others who are less active (Tylavsky et al., 1989). Beyond 50 years of age, bone loss in women abruptly increases in women as a consequence of the menopausal cessation of ovarian estrogen production. For women, the amount of bone loss from 50 to 80 years has been estimated to be as much as 30% to 50% of their total bone mass (Table 8.4). Women and men beyond 50 years old need to consume sufficient calcium to replace bone loss of calcium, but this zero balance may not occur when bone resorption exceeds bone formation even with additional calcium as supplement (Riis et al., 1987). So, calcium intakes in excess of ~1000 mg/day may not be so well utilized by the skeleton, and neutral (zero) skeletal balance may be impossible to achieve in late life. The DRI of 1200 mg/day for older adults may be reasonably safe, but the excess calcium that is not taken up by the skeleton, may contribute to calcium loading, that is, excessive soft tissue calcification and renal stones (Anderson and Sjoberg, 2001). Some concern exists about excessive intakes of calcium, for example, greater than 1200 to 1500 mg/day, because of arterial

129

Calcium and Bone

TABLE 8.4 Estimated Relative Gains and Losses of Bone Mass in Females across the Life Cycle By Age

Gain (%)

10 years 15 years 30 years 40 years 50 years 60 years 80 years 100 years Total

Loss (%)

50–60 30–40 10    5–10   10–20   15–20 100

~30–50

Total bone mass

% 100

Estrogens

96

Calcium

92

Placebo 0

6

12

18

24

Months

FIGURE 8.6  Relative loss of bone mineral density (BMD) in postmenopausal women. Elderly women receiving a calcium supplement (1000 mg) daily over 2 years in comparison to those receiving estrogen therapy and placebo. (Adapted from Riis, B., et al., New Engl J Med, 316, 173–177, 1987.)

calcification and cardiovascular disease (Bolland et al., 2008) and renal stones (Jackson et al., 2006) (see the section “Potential Calcium Toxicity: Arterial Calcification and Renal Stones”). Lactose intolerance from lactase deficiency in a large percentage of the U.S. population, that is, almost 15% of white adults and a higher percentage of African American adults, may also contribute to inadequate calcium and vitamin D intakes and, hence, lower bone density (see the section “Lactose Intolerance and Inadequate Calcium Intake”).

Elderly Calcium Needs Low calcium intakes among the elderly usually result from reductions in milk and cheese consumption. So, meeting the DRI of 1200 mg/day remains almost impossible without the consumption of calcium supplements; 500 mg of calcium supplements per day may be sufficient for most women to achieve the recommended 1000 to 1200 mg/day. Even with consumption above 1500 mg/day from food and supplements, elderly Danish women studied over 3 years still lost BMD (Figure 8.6) (Riis etal., 1987). In older adults, as well as at earlier adult ages, calcium supplements depress PTH secretion and bone resorption; the decrease in bone resorption tends to suppress bone turnover (McKane et al., 1996), making it less dynamic and less able to repair microfractures (Heaney, 2007) (see the section “Potential Calcium Toxicity: Arterial Calcification and Renal Stones”).

130

Diet, Nutrients, and Bone Health

CALCIUM METABOLISM This section highlights major aspects of calcium metabolism across the life cycle. Three significant physiological changes occur in the elderly, especially in females, that influence calcium metabolism: intestinal calcium absorption declines, renal calcium reabsorption decreases, and PTH increases. Together, these changes suggest negative calcium balance.

Intestinal Absorption of Calcium Absorption of calcium ions occurs in the absorbing epithelial cells (enterocytes) via the two-step process common to water-soluble nutrients. Calcium cations, that is, Ca2+, are relatively poorly absorbed compared with inorganic phosphate anions, that is, HPO4=, also written as Pi. For adults, the net calcium absorption efficiency is approximately 20% to 30%, whereas for phosphate, it is typically about 50% to 70%. Calcium absorption is completed within 2 to 3 hours following a meal, a much slower entry than for phosphate ions (Anderson, 1991). When considering that almost all foods contain phosphorus and only a relatively few contain much calcium, the quantitative absorption of phosphate ions practically always significantly exceeds that of calcium ions from a meal. The absorption of calcium ions may be depressed a percentage point or two, if phosphate is excessive in the diet, because of precipitation of calcium ions by phosphate ions within the gut lumen (see also Chapter 6). One of the two important interactions that exist for calcium is the enhancement of intestinal calcium absorption, involving the transcellular route, by the hormonal form of vitamin D (see Chapter 10). The vitamin D hormone, derived from prior intake of the vitamin from the diet or from skin biosynthesis, increases the efficiency of calcium ion absorption by stimulating the synthesis of calbindins by the gut absorbing cells. The calbindins carry the calcium ions, four calcium ions per calbindin, across the absorbing cells of the small intestine to the extracellular fluids and blood. At the site of new bone formation in the skeleton, the hormonal form of vitamin D increases the uptake by osteoblasts of calcium ions that are used in making new bone mineral, that is, calcium hydroxyapatite. If both calcium and vitamin D in the diet are low, for example, in a young child who is no longer breast feeding, declines in intestinal calcium absorption and bone mineralization almost certainly occur. Severe deficiencies of each may lead to rickets. At low intakes of calcium, intestinal absorption percentages of calcium ions typically increase to greater than 30% because of the feedback regulation of the vitamin D hormonal mechanism that improves calcium absorption efficiency. However, when calcium intakes are adequate or high, the vitamin-D-mediated gut absorptive mechanism operates at a much greatly reduced efficiency. Furthermore, it is probable that this mechanism may become totally inactivated to prevent excessive calcium absorption and potential adverse effects of calcium. Thus, the vitamin D adaptation mechanism for calcium absorption depends on a low calcium load (amount) from the diet and increased absorptive efficiency of calcium ions from the diet, as mediated by the vitamin D hormone (see Chapter 10). At high intakes of calcium, the serum calcium concentration, both total calcium and ionic calcium, increases slightly after a meal. The increase in ionic calcium depresses PTH secretion and, hence, its action on osteoblasts that signal osteoclasts to resorb bone (Martini and Wood, 2002; McKane et al., 1996). A decrease in PTH also reduces proximal tubular calcium reabsorption and thereby increases urinary calcium excretion. Figure 8.7 illustrates the sequence of physiological events following a low intake of calcium from foods or supplements. To maintain a reasonably constant supply of calcium in blood for tissue functions, a highly regulated serum calcium ion concentration is activated to increase the flow of calcium ions to blood. Although not illustrated here, a calcium-rich meal or supplement reverses this homeostatic regulation by suppressing PTH secretion in this sequence during the first few hours after a calcium-rich meal. The sequence of steps is listed below:

1. Initial positive signal (+) for PTH secretion results from an decrease in serum Ca+2. 2. Parathyroid glands secrete PTH in response to the decrease in the serum Ca+2 concentration.

131

Calcium and Bone

PTH Parathyroid glands

2

(+)

3

To Gut

4

1,25-(OH) Vitamin D2 Ca Pi Transfer from Bone 1

(+) Blood calcium

6

Restoration of normal blood 5 calcium Normal blood Ca Pi calcium Reabsorption

Ca Excretion Pi Excretion

FIGURE 8.7  Regulation of blood calcium (Ca) concentration following intake of a meal low in calcium. The stimulation of parathyroid hormone (PTH) secretion and action on bone and the kidney during fasting. (Adapted from Anderson, J.J.B., et al. 2005. Nutrition and Health, Carolina Academic Press.)

3. PTH acts on bone to increase transfer of both Ca and Pi ions to blood. 4. It also acts on the kidneys to increase Ca reabsorption (or decrease Ca excretion). Also, PTH stimulates an increase in renal 1,25(OH)2vitamin D production, which may increase intestinal (gut) Ca absorption over the next 24 hours if calcium intakes are typically low. 5. The net effect of these actions is to return the serum total Ca to normal set level (~10 mg/ dL), and the calcium ionic concentration (Ca+2) rises accordingly. 6. A normal serum calcium concentration has little or no influence on PTH secretion by this feedback mechanism until a decline in the Ca ionic concentration triggers PTH secretion. The serum total Ca concentration is highly regulated by this homeostatic mechanism; only small deviations occur under normal conditions.

The absorption of calcium ions is significantly depressed by oxalic acid (oxalate anions) and phytates when they are consumed in the same meal. These binding molecules (anions) carry negative charges that combine with positively charged calcium ions (cations) to reduce the number of calcium ions that are bioavailable and ready for absorption within the gut lumen. Oxalates are high in only a few foods, such as spinach, rhubarb, and beet greens. Factors in spinach other than oxalate may also contribute to the low efficiency of calcium absorption (Heaney et al., 1988). Phytate anions in grains have a similar, but less severe, effect of depressing the absorption of calcium ions. Hence, compared with animal sources, calcium ions from calcium-rich plant sources may be less well absorbed, that is, less bioavailable, depending on the anions or other factors in the plant foods. Postmenopausal women are generally considered to have a decline in intestinal calcium absorption, which adversely affects calcium balance. This decline may worsen as women enter the decade of their 50s, that is, the post menopause. Because of this reduction in calcium absorption efficiency, the recommendation for dietary calcium was increased for women and men in this stage of the life cycle.

Parathyroid Hormone and the Regulation of Serum Calcium Concentration Parathyroid hormone (PTH) is a significant hormone involved in calcium regulation because of its action of removing (pumping) calcium ions from bone to blood during intermeal or fasting periods to maintain blood calcium at its set level, that is, 10 mg/dL (2.5 mmol/L) (Talmage and Talmage, 2006, 2007; Talmage and Mobley, 2008) (see also Chapter 6). Thus, through its major actions on

132

Diet, Nutrients, and Bone Health

bone and the kidneys, PTH is the major regulator of the blood concentration of calcium, which supplies sufficient amounts of calcium ions for the functional needs of cells. When PTH acts on bone, it stimulates osteoclasts to remove bone mostly from the trabecular surfaces of the vertebrae and the ends of the long bones, and it increases calcium effluxes from the bone extracellular compartment. When PTH acts on the renal tubules, it increases calcium reabsorption and it also blocks phosphate reabsorption, thereby increasing urinary phosphorus losses.

Urinary Excretion of Calcium Several factors contribute to an increase in urinary calcium losses in postmenopausal women after the menopause and older men. Urinary calcium excretion is significantly influenced by several nutrients in the usual diet. The average North American consumes about 15% of food energy from protein, a large portion of which is derived from animal sources, such as meat, fish, poultry, milk, cheese, and eggs. When the individual amino acids are metabolized, they generate acid equivalents, especially sulfuric and phosphoric acids, which must be buffered by serum bicarbonate, cellular proteins, and bone before being excreted by the kidneys. A modest hypercalciuria (excessive calcium in the urine) may be a normal part of aging, if renal function remains healthy, because of increasing resorption of bone, that is, declining bone mass. In economically advanced societies, however, where a high-meat diet is consumed or where more physical activity is not an everyday phenomenon, the “relative” acid-induced hypercalciuria may be even more significant. For consumers of low amounts of calcium, protein-induced hypercalciuria remains a potentially significant mechanism explaining calcium loss in adults. A high sodium diet also increases urinary calcium losses because renal calcium excretion is tightly linked to urinary sodium excretion. These minerals are reabsorbed by renal tubules, in part, by a common mechanism that favors sodium reabsorption over calcium (Lemann et al., 1979). As glomerular filtration rate begins to decline in the later years of the life cycle, lower quantities of the calcium ions are excreted, that is, lower amounts appear in the urine over 24 hours. This decline in urinary calcium reflects lower resulting from a reduction in the formation of the active vitamin D hormonal molecule in urinary calcium by the aging kidney. Under conditions of very high calcium intake, that is, greater than 1400 mg/day, intestinal calcium absorption by older adults may lead to greater retention of calcium, but little in the skeleton. Older adults may risk vascular consequences of a positive calcium balance, that is, arterial calcification (Demer, 1995).

Calcium Balance Calcium balance is determined to the extent by which calcium intake from foods and supplements offsets calcium output in urine, feces, and sweat. In calcium balance, the major input is diet, if adequate, and the major outputs are urine and feces. Hormonal control of calcium homeostasis is exerted at the gut, bone, and kidneys, primarily through the actions of PTH. In addition, the major action of the hormonal form of vitamin D is on the absorbing cells of the small intestine when usual calcium intakes are less than adequate. When calcium intakes are not adequate, bone serves as the major source of calcium ions in blood. During the growth years, this balance is generally positive. In the years from roughly 20 to 40, calcium balance generally remains neutral or zero. In later life, certainly after the age of 50years, calcium balance is usually negative. During late life, measurements of urinary calcium over a 24-hour period may be needed to assess calcium balance. An increase in urinary calcium in individuals with normal renal function suggests a negative shift in calcium balance and a probable loss of bone mass (see Chapter 7). In older individuals, calcium balance may seem neutral, but this assessment may be misleading because of ectopic calcification, that is, new bone formation at inappropriate sites, especially in arteries and heart valves (see Chapter 33). In the elderly with normal renal function, the calcification in inappropriate locations contributes to better calcium balance, as a result of reduced or

133

Calcium and Bone

normal urinary calcium losses. If renal function becomes compromised, urinary calcium losses may decline and calcification may substantially increase. Then, calcium balance becomes even better since much less calcium is excreted by the kidneys (see the section “Potential Calcium Toxicity: Arterial Calcification and Renal Stones”).

EFFECTS OF DIETARY CALCIUM AND PHOSPHORUS RATIOS ON CALCIUM HOMEOSTASIS Too much dietary phosphorus may adversely affect bone health by increasing bone resorption (see the section “Parathyroid Hormone and the Regulation of Serum Calcium Concentration”). Bone resorption is the process by which small packets of the bone are degraded, both organic matrix and hard mineral. This process allows calcium and phosphate ions and bone matrix components (biomarkers) to be released into the extracellular fluid and blood. Bone resorption is increased during periods of low calcium and high phosphate intakes, and if it continues for long periods, this diet may adversely affect bone maintenance and reduce BMD (Calvo et al., 1990; Calvo and Park, 1996; Kemi et al., 2009). Therefore, a reasonable intake of calcium in relation to phosphorus is necessary for good bone health. An optimal dietary ratio of calcium to phosphorus for adults ranges between 1:1 and 1:2; a value of 1:1.5 is considered good. Adults who consume too little milk or other dairy products have ratios that are lower than 1:2 and may even approach 1:4. Figure 8.8 illustrates Ca:P ratios for breast milk; usual dietary intakes during childhood, adolescence, and adulthood; and bone mineral in comparison with the AIs. Constant dietary ratios of 1:2 and less are considered to be potentially detrimental to skeletal health. Recommended daily amounts of calcium and phosphorus are given above (see the section “Dietary Reference Intakes for Calcium”). Increasing the intake of calcium as part of a high phosphate diet does not appear to correct entirely the adverse alterations in calcium metabolism although the Ca:P ratio is improved, as the phosphorus intake still remains too high (Kemi et al., 2010). Over a long period of time, such habitual diets may cause alterations in calcium metabolism that contribute to excessive loss of bone mass and possibly osteoporosis (Calvo et al., 1990; Kemi et al., 2009). It should be noted that these studies were conducted in healthy young women, and they have not been replicated in postmenopausal women. 2.5:1

Ca:P ratio

2:1 1.5:1 1:1 Typical dietary ratio

0.5:1

Human breast milk

Bone

Blood (adult)

FIGURE 8.8  Ca:P ratios for breast milk; usual dietary intakes during childhood, adolescence, and adulthood; and bone mineral in comparison to the range of ratios of recommended intake amounts (DRIs) of calcium and phosphorus (RDAs).

134

Diet, Nutrients, and Bone Health

LACTOSE INTOLERANCE AND INADEQUATE CALCIUM INTAKE Because many people throughout the world stop synthesizing the enzyme lactase early in life, they become at risk for lactose intolerance, insufficient calcium intake, and poor bone development. If vitamin-D-fortified milk is not consumed, then both calcium and vitamin D, as well as other important nutrients, will not be ingested in adequate amounts to support bone growth. Rickets, either mild or severe, is a likely consequence. Lactose intolerance almost certainly leads to the avoidance of consuming adequate amounts of calcium by affected individuals. Many African Americans, Asians, and whites who develop lactose intolerance at some time early in life typically avoid milk and often other dairy products. Those who are lactose intolerant may benefit from using lactase-treated milk or from taking calcium supplements beginning as early in life as possible to try to optimize the development of the skeleton, that is, PBM, and then continue supplement use to maintain their bone mass during the adult decades. Without additional calcium sources in the diet, lactose intolerance may have adverse effects on skeletal mass and density in individuals who inherit the gene controlling the synthesis of the intestinal lactase enzyme. Most ethnic groups throughout the world develop hypolactasia (low state of lactase enzyme) after the first few years of life, which typically leads to a lower efficiency of calcium absorption and to intestinal cramping and other GI symptoms upon milk consumption. Late in life, these lactose intolerants are at increased risk for osteoporotic fractures (Segal et al., 2003). Despite hypolactasia and low or very low calcium intakes because of low dairy consumption, African American children typically develop bones that are dense and strong. In general, then, African American children have sufficient vitamin D and calcium in their diets during childhood— well after weaning—to absorb the available dietary calcium efficiently and to put it in their skeletons. This pattern differs for white children of Northern European backgrounds and for Asian children who typically consume greater amounts of calcium throughout most of the first two decades. White children obtain most of their calcium from dairy products, whereas Asian children get most from plant sources (see also Chapter 29).

DEFICIENT CALCIUM INTAKES AND THE NEED FOR CALCIUM SUPPLEMENTS In general, calcium deficiency results from too little ingestion of calcium-rich foods, a fairly common finding in the United States. Such low consumption patterns may contribute to late-life osteopenia and possibly osteoporotic bone fractures, whereas high calcium intakes and regular physical activity will likely help prevent or delay those fractures in later life. Preventive programs for elderly women include recommendations for the consumption of calcium and vitamin D at DRI or greater levels from foods and, typically, supplements. Supplements of these two nutrients have been demonstrated to improve bone measurements in elderly women (Chapuy et al., 1992; Dawson-Hughes etal., 1997), but not in elderly men (Dawson-Hughes et al., 1997). Also, recommendations are made for regular weight-bearing exercises, especially walking at a good pace for 30 minutes four or more days a week, and strength exercises involving upper body muscle groups that help maintain the bone mass and quality of the proximal femur (see later in this chapter). In modest quantities, calcium supplements may help maintain bone health, that is, bone mass and density, and the reduction of fractures by increasing the Ca:P ratio as well as the amount of calcium in the diet (see Chapter 27). The accumulation of calcium in the body, however, does not necessarily mean that all the additional calcium is put in the skeleton (see below). In older individuals with low calcium intakes, calcium supplements in modest amounts typically improve bone measurements, but an uncertain amount of additional supplemental calcium is now also considered to increase arterial calcification (Persy and D’Haese, 2009; Reid et al., 2010). Therefore, among the elderly, calcium supplements in large amounts (>500 mg/day) may actually exert adverse effects on health and serve as an unintended risk factor for cardiovascular diseases. This effect of extra dietary calcium as a supplement may have a trade-off between the improvement of overall calcium balance of the elderly

Calcium and Bone

135

and the simultaneous adverse effect of increasing calcium updake in coronary and other arteries of the body. A decline in renal function also may contribute to this unhealthy state of arterial calcification. Supplementation with calcium may also improve the effectiveness of bone-conserving drugs during the menopausal transition and in later postmenopausal life, but the usual amount of supplemental calcium recommended per day, 1000 mg, may be higher than necessary without considering typical daily intakes of the individuals. Adults and elders who obtain 800 to 1000 mg of calcium per day from foods may be doing as well as possible with respect to skeletal maintenance. Yet, most consume less than 800 mg/day, especially the elderly, and they may need a daily supplement of calcium (500 mg or so), along with vitamin D (400 IU/day or more), in an attempt to try to minimize their loss of bone mass and density. The DRIs for calcium remain the best recommendations for the U.S. population, but practically all individuals have difficulty consuming these recommended amounts from foods alone, especially the elderly whose DRI is 1,200 mg/day. Calcium intakes at or above DRIs have relatively little effect on bone compared to antiresorptive drugs (Ontjes and Anderson, 2009). The concern about excessive calcium loading relates to the higher supplemental amounts that, once absorbed, may not go into bone but rather into arterial walls or other soft tissues. The UL for calcium is 2500 mg/day (Food and Nutrition Board, IOM, 1997), but this high amount may be too much of a risk for health. This issue of excessive supplemental calcium intakes by older adults needs to be more closely monitored in the elderly who have declining renal function.

POTENTIAL CALCIUM TOXICITY: ARTERIAL CALCIFICATION AND RENAL STONES Toxic effects of calcium may occur with excessive intakes, particularly in cases of overconsumption of calcium supplements. This condition has also recently been referred to as calcium loading (Anderson, 2009). Calcium intakes in excess of needs may contribute to excessive mineralization of soft tissues, that is, calcification of arterial walls in large arteries, such as the aorta; small arteries, such as the coronaries; and the heart valves (Demer, 1995). Calcium salts in arterial walls, as part of rigid bone, limit greatly the elasticity of arteries; the aorta is especially affected. Other than constipation and renal stones, which have been observed in many female calcium supplement users, arterial calcification may be a forthcoming epidemic in older adults, in part because of excessive calcium supplement usage and in part because of reduced renal function. Calcification of coronary arteries, renal arteries, and heart valves may increase the risk of cardiovascular diseases (Bolland et al., 2008) (see also Chapter 33). Despite the fairly liberal UL value of 2,500 mg of calcium a day, adults at risk for these abnormalities may require dietary calcium restriction in an attempt to lessen their condition. Older adults are often appropriately advised to consume calcium supplements, in amounts of 500 or 1,000 mg/day, to assure that the absorbed calcium will help maintain BMD throughout the skeleton and especially at common sites of fractures. This recommendation, however, may cause adverse effects on calcium balance if renal function has started to decline in older adults. A glomerular filtration rate below 60 is considered the cut point for the beginning of serious renal disease, and it probably also is the point when arterial calcification becomes prominent. In the presence of positive calcium balance, the additional calcium absorbed tends not to be retained by the skeleton, but rather it is retained in soft tissues and the vasculature, that is, new bone at inappropriate sites. Although research has not yet fully established a progression of calcification with age when dietary calcium may be increased (Hsia et al., 2007), evidence suggests that some of the extra calcium from diet and bone resorption is retained in arterial walls (McClelland et al., 2006; Reid et al., 2010). Data from prospective randomized controlled trials are needed to rule in or out this hypothesis. The tentative conclusion now for older adults is that they must have their renal function checked to obtain their estimated glomerular filtration rate before starting to take calcium supplements—and

136

Diet, Nutrients, and Bone Health

perhaps before taking vitamin D supplements as well—or even after such supplementation has already been initiated. Calcium from dietary sources may be more preferable than supplemental calcium, but high intakes from foods alone may also contribute to arterial calcification. Calcium loading of the body has long been recognized in the coronaries of the heart, the aorta, the carotid arteries, and heart valves, but the increased risk of death from cardiovascular diseases when this condition exists has only recently become a major concern.

SUMMARY As an essential nutrient, calcium needs to be consumed in sufficient amounts to meet the body’s requirements on a daily basis. Intake amounts need to be within the optimal window that confers health. Although precise estimates of calcium requirements at any stage of the life cycle have not been determined, reasonable estimates of what healthy intakes should actually be have been published by the Institute of Medicine (Food and Nutrition Board, IOM, 1997 and 2011). The calcium recommended DRIs across the life cycle appear to provide a considerable safety factor, especially for women. These DRIs, however, are the same for males and females throughout the life cycle, despite differences in skeletal mass between the genders. Also, they are the same for all ethnic groups, despite differences in bone metabolism. Interactions between calcium and inorganic phosphate and between calcium and vitamin D may potentially have large impacts on the skeleton during growth and young adulthood, and possibly even during bone maintenance in later adulthood. Lactose intolerance also may have adverse effects on calcium and bone metabolism. Finally, calcium supplementation which is generally beneficial in terms of achieving DRIs of calcium may also exert adverse effects on arterial tissue by promoting calcification when total calcium intakes are excessive. During the period of skeletal growth, physical activity, when regularly performed, has a major positive impact on the accrual of skeletal mass as long as dietary intakes of most nutrients, especially calcium and vitamin D, remain sufficient. Young adult women between 20 and 30 years old may gain an additional 5% to 10% of their skeletal mass over this decade. Optimal nutrient intakes, then, should maximize an individual’s PBM by approximately the third decade of life. The growth years may be viewed as a window of opportunity, though short, to obtain dietary calcium in the skeleton and achieve PBM. When epiphyses close, excess calcium from supplements may not improve BMC and BMD appreciably, but it may increase the likelihood of arterial calcification. Approximately 60% of the calcium consumed by adults in the United States is from milk and dairy products. Deficiencies and more modest insufficiencies of calcium are common in the United States because of insufficient consumption of calcium-rich foods. Women typically have lower calcium intakes than those of men, and many begin taking calcium supplements around the time of the menopause to help delay the adverse effects of low bone mass and density. Such intakes do tend to increase bone mass modestly, especially when consumed with sufficient vitamin D. Excessive calcium intakes, however, may contribute to inappropriate calcification in arterial walls and heart valves, which may be a risk factor for cardiovascular and cerebrovascular diseases. In conclusion, achieving an adequate calcium intake as part of a healthy diet is often difficult, especially during the growth periods of the lifecycle. Meeting the recommended window of calcium intake, that is, enough but not too much, may be more of a challenge now that calcium supplementation is so widespread, at least in the United States. Future research should generate better understandings of excessive intakes and arterial calcification.

REFERENCES Abrams, S.A. 2005. Calcium supplementation during childhood: Long-term effects on bone mineralization. Nutr Rev 63: 251–255. Abrams, S.A., and Stuff, J.E. 1994. Calcium metabolism in girls: Current dietary intakes lead to low rates of calcium absorption and retention during puberty. Am J Clin Nutr 60: 739–743.

Calcium and Bone

137

Anderson, J.J.B. 1991. Nutritional biochemistry of calcium and phosphorus. J Nutr Biochem 2: 300–307. Anderson, J.J.B. 2009. Calcium, vitamin D, and bone health: How much do adults need? Nutr Food Sci 39: 337–341. Anderson, J.J.B., and Sjoberg, H.E. 2001. Dietary calcium and bone health in the elderly: Uncertainties about recommendations. Nutr Res 21: 263–268. Anon. 2009. What We Eat in America, NHANES 2005–2006: Usual Nutrient Intakes from Food and Water Compared to 1997 Dietary Reference Intakes for Vitamin D, Calcium, Phosphorus, and Magnesium. U.S. Department of Agriculture, Agricultural Research Service, Beltsville Human Nutrition Research Center, Food Surveys Research Group, Washington, D.C. http://www.ars.usda.gov/ba/bhnrc/fsrg Arias, E. 2010. United States life tables, 2006. Nat Vital Stat Rep 58 (21), 40 pp. Bailey, D.A., Martin, A.D., McKay, H.A., et al. 2000. Calcium accretion in girls and boys during puberty: A longitudinal analysis. J Bone Miner Res 15: 2245–2250. Barr, S.I., and McKay, H.A. 1996. Nutrition, exercise, and bone status in youth. Int J Sport Nutr 8: 124–142. Bolland, M.J., Barber, P.A., Doughty, R.N., et al. 2008. Vascular events in healthy older women receiving calcium supplementation: Randomized controlled trial. BMJ 336: 262–265. Bonjour, J.P., Carrie, A.L., Ferrari, S., et al. 1997. Calcium-enriched foods and bone mass growth in prepubertal girls: A randomized, double-blind, placebo-controlled trial. J Clin Invest 99: 1287–1294. Bonjour, J.P., Theintz, G., Buchs, B., et al. 1991. Critical years and stages of puberty for spinal and femoral bone mass accumulation during adolescence. J Clin Endocrinol Metab 73: 555–563. Cadogan, J., Eastell, R., Jones, N., et al. 1997. Milk intakes and bone mineral acquisition in adolescent girls: Randomized, controlled intervention trial. BMJ 315: 1255–1260. Calvo, M.S., Kumar, R., and Heath, H., III. 1990. Persistently elevated parathyroid hormone secretion and action in young women after four weeks of ingesting high phosphorus, low calcium diets. J Clin Endocrinol Metab 70: 1334–1340. Calvo, M.S., and Park, Y.K. 1996. Changing phosphorus content of the U.S. diet: Potential for adverse effects on bone. J Nutr 126: 1168s–1180s. Chapuy, M.-C., Arlot, M.E., Duboeuf, F., et al. 1992. Vitamin D3 and calcium to prevent hip fractures in elderly women. New Engl J Med 327: 1637–1642. Dawson-Hughes, B., Harris, S.S., Krall, E.A., et al. 1997. Effect of calcium and vitamin D supplementation on bone density in men and women 65 years of age or older. New Engl J Med 337: 670–676. Demer, L. 1995. A skeleton in the atherosclerotic closet. Circulation 92: 2029–2032. Ervin, R.B., Wang, C.-Y., Wright, J.D., et al. 2004. Dietary intake of selected minerals for the United States population: 1999–2000. Advance Data No. 341. [CDC, Vital and Health Statistics, U.S. DHHS, Hyattsville, MD, 8 pages] Fleming, K.H., and Heimbach, J.T. 1994. Consumption of calcium in the US: Food sources and intake levels. J Nutr 124: S1426–S1430. Food and Nutrition Board, Institute of Medicine, 1997. Dietary Reference Intakes for Calcium, Phosphorus, Magnesium, Vitamin D, and Fluoride. National Academy Press, Washington, DC. Food and Nutrition Board, Institute of Medicine, 2011. Dietary Reference Intakes of Calcium and Vitamin D. National Academy Press, Washington, DC. French, S.A., Fulkerson, J.A., and Story, M. 2000. Increasing weight-bearing physical activity and calcium intake for bone mass growth in children and adolescents: A review of intervention trials. Prev Med 31: 722–731. Goulding, A., Cannan, R., Williams, S.M., et al. 1998. Bone mineral density in girls with forearm fractures. J Bone Miner Res 13: 143–148. Halioua, L., and Anderson, J.J.B. 1989. Lifetime calcium intake and physical activity habits: independent and combined effects on bone mass in healthy premenopausal Caucasian women. Am J Clin Nutr 49:534–541. Heaney, R.P. 2007. Bone health. Am J Clin Nutr 85: 300S–303S. Heaney, R.P., Abrams, S., Dawson-Hughes, B., et al. 2000. Peak bone mass. Osteoporos Int 11: 985–1009. Heaney, R.P., Weaver, C.M., and Recker, R.R. 1988. Calcium absorbability from spinach. Am J Clin Nutr 47: 707–709. Hobbs, S.H., and Anderson, J.J.B. 2009. Calcium and vitamin D. In The Complete Vegetarian, Carlson, P., ed. University of Illinois Press, Urbana and Chicago, IL, 78–82. Hsia, J., Heiss, G., Ren, H., et al. 2007. Calcium/vitamin D supplementation and cardiovascular events. Circulation 106: 856–854. Hunt, C.D., and Johnson, L.K. 2007. Calcium requirements: New estimations for men and women by cross-sectional statistical analyses of calcium balance data from metabolic studies. Am J Clin Nutr 86: 1054–1063.

138

Diet, Nutrients, and Bone Health

Jackman, L.A., Millane, S.S., Martin, B.R., et al. 1997. Calcium retention in relation to calcium intake and postmenarcheal age in adolescent females. Am J Clin Nutr 66: 327–333. Jackson, R.D., LaCroix, A.Z., Gass, M., et al. for the Women’s Health Initiative Investigators. 2006. Calcium plus vitamin D supplementation and the risk of fractures. New Engl J Med 354: 669–683. Johnston, C.C., Jr., Miller, J.Z., Slemenda, C.W., et al. 1992. Calcium supplementation and increases in bone mineral density in children. New Engl J Med 327: 82–87. Kanis, J.A., Melton, L.J., III, Christiansen, C., et al. 1994. The diagnosis of osteoporosis. J Bone Miner Res 9: 1137–1141. Kemi, V.E., Karkkainen, M.U.M., Rita, H.J., et al. 2010. Low calcium:phosphorus ratio in habitual diets affects serum parathyroid hormone concentration and calcium metabolism in healthy women with adequate calcium intake. Br J Nutr 103: 561–568. Kemi, V.E., Rita, H.J., Karkkainen, M.U., et al. 2009. Habitual high phosphorus intakes and foods with phosphate additives negatively affect serum parathyroid concentration: A cross-sectional study on healthy premenopausal women. Public Health Nutr 12: 1885–1892. Lemann, J., Adams, N.D., and Gray, R.W. 1979. Urinary calcium excretion in human beings. New Engl J Med 301: 535–541. Lloyd, T., Andon, M.B., Rollings, N., et al. 1993. Calcium supplementation and bone mineral density in adolescent girls. JAMA 270: 841–844. Lloyd, T., Rollings, N., Andon, M.B., et al. 1992. Determinants of bone density in young women. I. Relationships among pubertal development, total body bone mass, and total body bone density in premenarchal females. J Clin Endocrinol Metab 75: 383–387. Lloyd, T., Rollings, N., Andon, M.B., et al. 1994. Enhanced bone gain in early adolescence due to calcium supplementation does not persist in late adolescence. J Bone Miner Res 11: S154. [Abstract] Looker, A.C., Loria, C.M., Carroll, M.D., et al. 1993. Calcium intakes of Mexican Americans, Cubans, Puerto Ricans, non-Hispanic whites, and non-Hispanic blacks in the United States. J Am Diet Assoc 93: 1274–1279. Martin, B.R., Davis, S., Campbell, W.W., and Weaver, C.M. 2007. Exercise and calcium supplementation: Effects on calcium homeostasis in sportswomen. Med Sci Sports Exerc 39: 1481–1486. Martini, L., and Wood, R.J. 2002. Relative bioavailability of calcium-rich dietary sources in the elderly. Am J Clin Nutr 76: 1345–1350. Matkovic, V., and Heaney, R.P. 1992. Calcium balance during human growth: Evidence for threshold behavior. Am J Clin Nutr 55: 992–996. Matkovic, V., Jelic, T., Wardlaw, G.M., et al. 1994. Timing of peak bone mass in Caucasian females and its implication for the prevention of osteoporosis. J Clin Invest 93: 799–808. Matkovic, V., Landoll, J.D., Badenhop-Stevens, N.E., et al. 2004. Nutrition influences skeletal development from childhood to adulthood: A study of hip, spine, and forearm in adolescent females. J Nutr 14: 701S–705S. McClelland, R.L., Chung, H., Detrano, R., et al. 2006. Distribution of coronary artery calcium by race, gender, and age: Results from the Multi-Ethnic Study of Atherosclerosis (MESA). Circulation 113: 30–37. McKane, W.R., Khosla, S., Egan, K.S., et al. 1996. Role of calcium intake in modulating age-related increases in parathyroid function and bone resorption. J Clin Endocrinol Metab 81: 1699–1703. Nieves, J.W. Osteoporosis: The role of micronutrients. 2005. Am J Clin Nutr 81: 1232S–1239S. Ontjes, D.A., and Anderson, J.J.B. 2009. Nutritional and pharmacologic aspects of osteoporosis. In Handbook of Clinical Nutrition and Aging, 2nd ed., Bales, C.W., and Ritchie, C.S., eds. Humana Press, Springer Science, New York, NY. Peaco*ck, M. 2010. Calcium metabolism in health and disease. Clin J Am Soc Nephrol 5: S23–S30. Persy, V., and D’Haese, P. 2009. Vascular calcification and bone disease: The calcification paradox. Trends Mol Med 15: 405–416. Reid, I., Bolland, M.J., and Grey, A. 2010. Does calcium supplementation increase cardiovascular risk? Clin Endocrinol doi:10.1111/j.1365-2265.2010.03792.x Riis, B., Thomsen, K., and Christiansen, C. 1987. Does calcium supplementation prevent postmenopausal bone loss? A double-blind, controlled clinical trial. New Engl J Med 316: 173–177. Segal, E., Dvorkin, L., Levy, L., et al. 2003. Bone density in axial and appendicular skeleton in patients with lactose intolerance: Influence of calcium intake and vitamin D status. J Am Coll Nutr 22: 201–207. Slemenda, C.W., Christian, J.C., Williams, C.J., et al. 1992. Genetic determinants of bone mass in adult women: A reevaluation of the twin model and the potential importance of gene interaction on heritability estimates. J Bone Miner Res 6: 561–567.

Calcium and Bone

139

Slemenda, C.W., Peaco*ck, M., Hui, S., et al. 1997. Reduced rates of skeletal remodeling are associated with increased bone mineral density during the development of peak skeletal mass. J Bone Miner Res 12: 676–682. Specker, B.L., and Vukovich, M. 2007. Evidence for an interaction between exercise and nutrition for improved bone health during growth. Med Sport Sci 51: 50–63. Talmage, D.W., and Talmage, R.V. 2007. Calcium homeostasis: How bone solubility relates to all aspects of bone physiology. J Musculoskelet Neuronal Interact 7: 108–112. Talmage, R.V., and Mobley, H.T. 2008. Calcium homeostasis: Reassessment of the actions of parathyroid hormone. Gen Comp Endocrinol 156: 1–8. Talmage, R.V., and Talmage, D.W. 2006. Calcium homeostasis: Solving the solubility problem. J Musculoskelet Neuronal Interact 6: 402–406. Teegarden, D., Proulx, W.R., Martin, B.R., et al. 1995. Peak bone mass in young women. J Bone Miner Res 10: 711–715. Tylavsky, F.A., Bortz, A.D., Hanco*ck, R.L., et al. 1989. Familial resemblance of radial bone mass between premenopausal mothers and their college-age daughters. Calcif Tissue Int 45: 265–272. Vatanparast, H., Bailey, D.A., Baxter-Jones, A.D.G., et al. 2010. Calcium requirements for bone growth in Canadian boys and girls during adolescence. Br J Nutr 103: 575–580. Vatanparast, H., and Whiting, S.J. 2006. Calcium supplementation trials and bone mass development in children, adolescents and young adults. Nutr Rev 64: 204–209. Weaver, C.M., Martin, B.R., Plawecki, K.L., et al. 1995. Differences in calcium metabolism between adolescent and adult females. Am J Clin Nutr 61: 577–581. Weaver, C.M., McCabe, L.D., McCabe, G.P., et al. 2007. Bone mineral and predictors of bone mass in white, Hispanic, and Asian early pubertal girls. Calcif Tissue Int 81: 352–363. Weaver, C.M., Proulx, W.R., and Heaney, R. 1999. Choices for achieving adequate dietary calcium with a vegetarian diet. Am J Clin Nutr 70 (Suppl): 543S–548S. Welch, J.M., and Weaver, C.M. 2005. Calcium and exercise affect the growing skeleton. Nutr Rev 63: 361–373. Wyshak, G., and Frisch, R.E. 1994. Carbonated beverages, dietary calcium, the dietary calcium/phosphorus ratio, and bone fractures in girls and boys. J Adolesc Health 15: 210–215. Young, D., Hopper, J.L., Nowson, C.A., et al. 1995. Determinants of bone mass in 10 to 26 year old females: A twin study. J Bone Miner Res 10: 558–567.

9 Do Higher Dietary Levels

Inorganic Phosphorus Affect Phosphorus Homeostasis and Bone?* Mona S. Calvo

CONTENTS Introduction..................................................................................................................................... 141 Inorganic Phosphorus and Bone..................................................................................................... 142 Inorganic Phosphorus Balance............................................................................................... 142 Bone��������������������������������������������������������������������������������������������������������������������������������������143 Dietary Phosphorus......................................................................................................................... 143 Differences in Organic and Inorganic Phosphorus in Food................................................... 143 Phosphorus Additives in Food............................................................................................... 143 Dietary Guidelines for Phosphorus Intake............................................................................. 147 Total Dietary Phosphorus Intake............................................................................................ 147 Current Understanding of the Regulation of Phosphorus Homeostasis.......................................... 149 Regulation by Novel Intestinal Phosphate Sensor................................................................. 149 Classical Endocrine Regulation............................................................................................. 149 Phosphatonin Regulation....................................................................................................... 150 Physiologic Effects of High Inorganic Phosphorus Diet................................................................ 151 Effects on Bone Health.......................................................................................................... 151 Effects on Cardiovascular Health.......................................................................................... 153 Summary......................................................................................................................................... 153 References....................................................................................................................................... 154

INTRODUCTION The adult body contains approximately 600 g of phosphorus as both inorganic and organic phosphorus (Endres and Rude, 2006). Approximately 510 g or 85% of total body phosphorus is contained in the adult skeleton as organic and inorganic phosphates, and soft tissues contain 15% as both inorganic and organic phosphate, whereas the extracellular fluid contains 0.1% largely as inorganic phosphorus. Cellular phosphates function in many energy-intensive physiologic functions such as muscle contraction, nerve conduction, electrolyte transport, and energy production, in addition to *

Required Disclaimer: The findings and conclusions presented in this chapter are those of the author and do not necessarily represent the views and opinions of the U.S. Food and Drug Administration. Mention of trade names, product labels, or food manufacturers does not constitute endorsem*nt or recommendations or use by the U.S. Food and Drug Administration.

141

142

Diet, Nutrients, and Bone Health

providing the main structural support of the body as a component of bone mineral. Intracellular phosphates are critical to the regulation of intermediary metabolism of proteins, fats and carbohydrates, gene transcription, and cell growth (Endres and Rude, 2006). These tissue functions are very sensitive to fluxes of inorganic phosphate into soft tissues when blood levels increase due to dietary loading or failing kidney function. To maintain homeostasis of phosphorus, exquisitely sensitive physiologic mechanisms have evolved to tightly regulate the level of inorganic phosphorus in the extracellular fluid (Calvo and Carpenter, 2003). This chapter focuses on our current understanding of the role of bone and other endocrine organs in the regulation of phosphorus homeostasis. It addresses the question of how the changing inorganic phosphorus content of our food supply influences phosphorus homeostasis and impacts bone health and risk of hypertension and cardiovascular disease.

INORGANIC PHOSPHORUS AND BONE Inorganic Phosphorus Balance Extracellular fluid phosphorus is regulated within a very narrow concentration range by hormonal processes that control intestinal absorption and renal excretion (Schiavi and Kumar, 2004). Figure9.1 shows how balance is maintained between intake and excretion in a normal 70-kg adult consuming approximately 20 mg phosphorus/kg of body weight/day. Phosphorus is absorbed with much greater efficiency than are calcium and other minerals (Lemann, 1993). As much as 80% of ingested phosphorus from highly processed food is absorbed and enters the extracellular fluid pool from which it can be moved in and out of bone as needed (Schiavi and Kumar, 2004). Approximately 70% of the dietary phosphorus is absorbed, but almost 100% of this phosphorus is excreted per day by the kidneys. This excretion enables maintenance of phosphorus homeostasis, which is a balance between the amount absorbed and the amount excreted. Plasma phosphorus levels normally occur within a

Phosphate intake 1400 mg/d 210 mg/d Bone exchange

Intestinal absorption 1120 mg/d 210 mg/d Intestinal juices

490 mg/d Fecal loss

ECF PO4– Normal plasma range 2.5–4.5 mg/dL 0.81–1.45 nmol/L

910 mg/d Urine loss

FIGURE 9.1  Maintenance of phosphorus balance in a 70-kg adult consuming 1400 mg of phosphorus daily. Figure was drawn by the author based on schematic presented by Schiavi and Kumar (2004).

Inorganic Phosphorus

143

very narrow range, 2.5 to 4.5 mg/dL, and maintenance of this set point or narrow range of plasma phosphorus differs from the maintenance of whole body homeostasis. Plasma phosphorus is freely filtered at the glomerulus, and regulation of the amount excreted in the urine occurs at the level of the proximal tubule where it is controlled by a number of factors that either increase or decrease phosphorus reabsorption. Dietary phosphate loading and depletion are key factors regulating renal phosphate reabsorption. This is achieved by controlling the number of sodium–phosphate cotransporters (NaPiIIa) located on the surface of the renal proximal tubule brush border membrane (Takeda et al., 2004). Phosphorus depletion results in recruitment of existing NaPiIIa cotransporters from intracellular pools, whereas phosphorus loading results in endocytosis of the cotransporters (Takeda et al., 2004). Sabbagh et al. (2009) have recently demonstrated that an intestinal cotransporter (Npt2b) plays a major role in phosphate absorption and overall phosphorus homeostasis. Intestinal phosphate absorption occurs by both passive and active transport, with the active transport controlled by the intestinal cotransporter.

Bone The skeleton, which comprises most of the inorganic phosphate of the body, serves as a ready reserve of phosphate ions for biological functions, especially cellular uses such as ATP, DNA, RNA, and many other molecules critical for metabolism. Osteocytes, cells embedded in bone, are a source of endocrine factors, such as fibroblast growth factor 23 (FGF23), secreted into the circulation that function in the regulation of phosphorus balance. Bone thus functions in structural support and mineral storage and serves as an endocrine organ.

DIETARY PHOSPHORUS Differences in Organic and Inorganic Phosphorus in Food Phosphorus in the food supply is either organic, bound to a carbon compound, or inorganic, a phosphate acid or salt not bound to a carbon-containing compound. Examples of organic and inorganic phosphorus commonly found in food are shown in Figure 9.2. Organic forms, such as phytate, the plant storage form of phosphorus in whole grains, and phosphoproteins, such as casein or whey from milk, show a slower rate of phosphorus absorption than that of inorganic phosphate salts (Calvo and Carpenter, 2003; Karp et al., 2007; Uribarri, 2009). Organic phosphorus is generally less bioavailable and must be digested or degraded by enzymatic action such as phytase, which is not produced by the mammalian gastrointestinal tract (Calvo and Carpenter, 2003). Inorganic phosphate salts readily dissociate in the acidic environment of the stomach requiring no enzymatic digestion and therefore are more rapidly and efficiently absorbed and have greater metabolic affects. Evidence of this physiologic difference in the rate of absorption and metabolic affect was recently reported by Karp et al. (2007). They monitored parathyroid hormone (PTH) levels after consumption of high dietary phosphorus from different food sources: meat, whole grain (phytate source), cheese (calcium–phosphoprotein), and a dietary supplement (combination of disodium and trisodium phosphate salt). Inorganic phosphate salts are shown to have metabolic consequences (rise in PTH) in this study, whereas organic phosphorus from foods such as cheese, meat, and whole grains either depressed PTH (cheese affect) or did not differ from the control session where only 500-mg phosphorus was consumed.

Phosphorus Additives in Food The phosphorus content of the U.S. diet is increasing as a result of the growing consumption of processed foods containing phosphate additives (Calvo and Park, 1996; Uribarri and Calvo, 2003).

144

Diet, Nutrients, and Bone Health Inorganic phosphate salts

Organic phosphorus example: phytate OH

OH –O

P O

O O

O O P O O

P O–

O–

O

O– O– O P O– O P

O– O– O

O

O P

O–

OH

• Slower rate of absorption

HO

P

HO

P

O

O OH HO

O

P

O

P

OH

P

O

P

OH

P

OH

O O HO O OH OH OH P P Orthophosphoric Pyrophosphoric O OH acid acid O Trimetaphosphoric O O O acid O O HO P O P O P OH O P P OH OH OH O OO O Tripolyphosphoric acid P P O O O O O O O O P

O

OH OH OH OH Tetrapolyphosphoric acid

• Generally less bioavailable

• Rapidly absorbed

• Bioavailable when digested or degraded by enzymatic action

• Highly bioavailable

Phosphoric anhydride (P4O10)

• Rapidly dissociates in gut acidity • No enzymatic degradation

FIGURE 9.2  Structural and physiological differences between organic phosphorus and inorganic phosphate salts.

Almost two decades ago, we reported that the use of phosphorus containing food additives had increased by 17% over the previous report which estimated that phosphorus salts contributed to more than 30% of the adult phosphorus intake (Calvo, 1994). Phosphorus containing additives are inorganic forms of phosphorus widely used in the processing of a broad variety of food categories, ranging from baked goods to restructured meats and cola beverages as shown in Table 9.1 (International Food Additive Council, 2008). Greater use of these very desirable functions of phosphate additives is encouraged by our fast-paced culture’s ever-growing consumption of fast food and need for fast-cooking convenience foods (Calvo and Park, 1996; Sarathy et al., 2008). Convenience foods are more highly processed, and the added phosphate salts enable them to cook faster or to be “instant,” requiring no or little cooking. An example of such a convenience food product is shown in Table 9.2 which presents the ingredient labels from an “instant” lemon pudding mix, as well as the ingredient label from the “cooked” product made by the same manufacturer. These labels also illustrate that more than one phosphate additive can be found in a product due to their many valued functional properties. The U.S. Food and Drug Administration (FDA) has considered approved inorganic phosphates to be Generally Recognized as Safe since 1979, and a more recent toxicological review concluded that all four classes of inorganic phosphates exhibit low oral toxicity (Weiner et al., 2001). The toxicology reviewers stated that “humans [with normal renal function] are unlikely to experience adverse effects when the daily phosphorus consumption remains below 70 mg/kg/day.” This estimate is equivalent to 4.9 g of phosphorus per day by a 70-kg adult! Current reality is that we have no mechanism to accurately monitor the contributions of these phosphorus additives to total phosphorus intake. Phosphorus intakes from fast foods, convenience foods, or processed food in general are not captured by nutrient composition databases largely because the content changes with the constantly evolving processing techniques and phosphorus content is not required to be listed on the FDA’s Nutrition Facts panel (Sullivan et al., 2007; Uribarri, 2009).

Imitation cheese Cheese slices Starter cultures

Hard, soft, and imitation ice cream Imitation dairy products Nondairy creamer Cheese Cottage cheese Dips and sauces

Baked goods Baking powder Cakes, mixes Cake donuts Refrigerated dough Beverages Colas Chocolate milk Dry mixes Buttermilk Orange juice Strawberry-flavored milk Cereals and pasta Cooked cereals Extruded, dry cereals Pasta products Dairy products Instant puddings and cheesecakes

Finished Product

TABLE 9.1 Phosphate Use in Foods

Decrease cooking time; calcium and phosphorus fortification Aid in the flow through extruder, calcium and phosphorus fortification Decrease cooking time Salts to keep the thickened texture Prevent churning or gritty texture development of fat Buffer for smooth mixing into coffee Direct set by acidification Emulsifying action

Disodium phosphate, tricalcium phosphate Disodium phosphate, trisodium phosphate, tricalcium phosphate Disodium phosphate

Monocalcium phosphate, disodium phosphate, tetrasodium pyrophosphate Tetrasodium pyrophosphate

Dipotassium phosphate

Phosphoric acid Disodium phosphate, trisodium phosphate, sodium hexametaphosphate Disodium phosphate Disodium phosphate, dipotassium phosphate Disodium phosphate. Dipotassium phosphate

continued

Acidulates Suspend cocoa Prevent caking; clouding agent Maintain protein dispersion Calcium and phosphorus fortification Bind iron to maintain pink color

Phosphoric acid Tetrasodium pyrophosphate Tricalcium phosphate Tetrasodium pyrophosphate, disodium phosphate Tricalcium phosphate Tetrasodium phosphate

Emulsifying action Emulsifying action Inhibit phage growth

Acid-base reaction with sodium bicarbonate to produce CO2 Moderate action leavening; double-action leavening Fast-action leavening Heat-activated leavening

Phosphate Function

Sodium acid pyrophosphate 28, monocalcium phosphate Sodium acid pyrophosphate, monocalcium phosphate, salp Sodium acid pyrophosphate 40, sodium acid pyrophosphate 43 Sodium acid pyrophosphate 22

Phosphate Ingredient

Inorganic Phosphorus 145

Phosphate Ingredient

Phosphate Function

Moisture binding Mechanical peeling of shrimp Bind copper from blood to prevent discoloration Cryoprotectant to protein Moisture binding Remove salmonella and campylobacter Color stability Improve whipping and foam stability Maintain firmness when canned Create a bubbled surface Bind iron to inhibit iron induced blackening

Sodium tripolyphosphate Sodium acid pyrophosphate Sodium tripolyphosphate/tetrasodium pyrophosphate blends

Sodium tripolyphosphate, and blends with sodium hexametaphosphate Trisodium phosphate

Monosodium phosphate, monopotassium phosphate Sodium hexametaphosphate

Monocalcium phosphate Monocalcium phosphate Sodium acid pyrophosphate

Moisture binding Emulsion development; reduced sodium; cure color development

Develop characteristic surface

Maintain hom*ogeneity

Sodium tripolyphosphate, and blends with hexametaphosphate Sodium tripolyphosphate, tetrasodium pyrophosphate, tetrapotassium phosphate, sodium acid pyrophosphate Sodium tripolyphosphate

Disodium phosphate, sodium hexametaphosphate, sodium tripolyphosphate Monocalcium phosphate

Source: Data modified from International Food Additive Council. 2008. Phosphates Use in Foods. http:www.foodadditives.org/phosphates/phosphate_used_in_food.html (last accessed January 2, 2010).

Carcass washes Egg products Whole eggs Egg whites Fruit and vegetable products Tomatoes, berries Baked potato chips French fries, hash browns, potato flakes

Roast beef Seafood Shrimp Canned crab Surimi Poultry Poultry products

Baked chips Meat products Ham, corned beef Sausage franks, bologna

Chips dips

Finished Product

TABLE 9.1 (Continued) Phosphate Use in Foods

146 Diet, Nutrients, and Bone Health

Inorganic Phosphorus

147

TABLE 9.2 Example of Convenience Food Product Use of Phosphate Additives Instant pudding and pie filling Ingredients: sugar, modified cornstarch, contains less than 2% of natural flavor, disodium phosphate and tetrasodium pyrophosphate for thickening, monoglycerides and diglycerides (prevent foaming), yellow 5, yellow 6, BHA (preservative) Cook and serve pudding and pie filling Ingredients: cornstarch, sugar, dextrose, modified tapioca starch, fumaric acid (for tartness), contains less than 2% of natural flavor, salt, hydrogenated soybean oil, yellow 5, yellow 6, BHA (preservative) Ingredient Label Source:  JELL-O brand Lemon Puddings, Product of Kraft Foods Global, In C., Northfield, IL.

Using the available commercial software for estimating nutrient intake, we compared the accuracy of these indirect methods of estimating phosphorus intake with that of the direct chemical analyses of the phosphorus content of a variety of diets. We found extreme underestimation of phosphorus intake (greater than 20%) when software relying on nutrient content databases was used (Oenning et al., 1988). Phosphorus intake has received little attention in the years since we explored the accuracy of estimating its intake using these databases. However, recent findings now challenge the safety of high inorganic phosphorus intake for bone, cardiovascular, and kidney health, all of which underscore the need for updating the phosphorus content of foods in our national nutrient databases and the need to require phosphorus content in the Nutrition Facts panel (Alonso et al., 2010; Dhingra et al., 2007; Foley et al., 2009; Giachelli, 2009; Isakova et al., 2009; Kemi et al., 2009; Uribarri and Calvo, 2003).

Dietary Guidelines for Phosphorus Intake Table 9.3 shows the U.S. Dietary Guidelines for phosphorus intakes from foods and dietary supplements (Institute of Medicine [IOM], 1997). These guidelines were last reviewed and changed in 1997 when both a Recommended Dietary Allowance (RDA) and an Estimated Average Requirement (EAR) were established for phosphorus intakes by specific age groups. Relative to the earlier 1989 RDA for phosphorus, changes were made for all age groups, except 9- to 18-year-olds, who require more of this nutrient during rapid bone accretion. Phosphorus intakes were lowered by 100 mg from 800 to 700 mg/day in adults. The EAR which is used to evaluate nutrient intake status of a population was lowered to 580 for adults (IOM, 1997). The Tolerable Upper Intake Level [UL], also set in 1997, is 4 g of phosphorus per day. As discussed above, this level has come into question, with the recent evidence linking high serum phosphorus to adverse health outcomes. The RDA and EAR differ from the labeling guidelines set by the FDA for phosphorus content labeling on food products. The FDA labeling guidelines are the Reference Daily Intake (RDI) or also termed Daily Value (DV); however, phosphorus content is not required on the Nutrition Facts panel of food labels. Listing phosphorus content is optional for food manufacturers and involves only one value, unlike the RDA. The RDI/DV for phosphorus is 1000 mg, which is usually expressed as a percent of the DV. When manufacturers opt to list phosphorus content of their product as a percent of the DV, it often leads to confusion and underestimation of the true phosphorus content. This information is critical in individuals who must closely monitor their phosphorus intake due to chronic renal failure (Kalantar-Zadeh et al., 2010; Uribarri, 2009).

Total Dietary Phosphorus Intake Figure 9.3 shows the median usual intake of phosphorus for various age groups of men and women taken from the most recent available data from the National Health and Nutrition Education Survey (NHANES) conducted in 2005 to 2006 (Moshegh et al., 2009). These nationally representative

148

Diet, Nutrients, and Bone Health

TABLE 9.3 U.S. Dietary Recommended Intake (DRI) Guidelines for Phosphorus (P) Intake from Foods and Dietary Supplements 1997 DRI, Recommended Dietary Allowance, and Estimated Average Requirement Years

mg/d

mg/d

1–3   460   380 4–8   500   405 9–18 1250 1055 19–50   700   580 51–70   700   580 70+   700   580 Tolerable Upper Intake Level (UL) for phosphorus 1997 DRI UL = 4.0 g Pi Source: Institute of Medicine. 1997. Dietary Reference Intakes for Calcium, Phosphorus, Magnesium, Vitamin D, and Fluoride. National Academy Press, Washington, DC.

PO4 (mg/d) consumed by 50% of the US population compared to the estimated average requirement (EAR) 1800 1600 1400 1200 EAR

1000 800

Women

600

Men

400 200 0

1 to 3

19 to 4 to 8 9 to 13 14 to 18 30 Age in years

31 to 50

51 to 70

>70

FIGURE 9.3  Median total phosphorus intake (mg/day) of men and women in various age groups compared with the Estimated Average Requirement (EAR) for each age group. Figure drawn by the author from data published for the NHANES 2005–2006 survey. (Moshegh, A., Goldman, J., Ahuja, J., et al. 2009. What we eat in America. NHANES 2005–2006. In Usual Nutrient Intakes from Food and Water Compared to 1997 Dietary Reference Intakes for Vitamin D, Calcium, Phosphorus, and Magnesium. U.S. Department of Agriculture Research Service.)

Inorganic Phosphorus

149

intake estimates are compared with the EAR for each age group. Traditionally, the EAR is used to evaluate intake adequacy of a nutrient at the 50th percentile (median) level of intake. For all ages and genders, except for adolescent girls, the estimated intake of phosphorus greatly exceeds the EAR. Despite the fact that total phosphorus intakes are generally in excess, these values most likely underestimate phosphorus intake in individuals with specific preferences for fast foods or other highly processed foods because the nutrient content databases used for these estimates do not reflect changes in phosphate additive use. Inaccuracy in these estimates presents a critical confounder in studies exploring the relationship between phosphorus intake and disease endpoints.

CURRENT UNDERSTANDING OF THE REGULATION OF PHOSPHORUS HOMEOSTASIS Our past understanding of phosphorus homeostasis was based on the concept that the body could tolerate wide variations in phosphate intake and plasma levels with little adversity as long as kidney function was maintained. However, current understanding is based on recent research findings indicating that extracellular fluid phosphate levels are tightly regulated within a narrow range and that even slight variations in these levels are associated with chronic disease development. Bergwitz and Juppner (2010), Berndt and Kumar (2009), Isakova et al. (2009), and Quarles (2008) presented valuable reviews of the newfound complexities of the endocrine regulation of phosphorus homeostasis, which is no longer limited to the classic hormones of the parathyroid glands and kidney (active metabolite of vitamin D) but also involves bone and the secretion of FGF23.

Regulation by Novel Intestinal Phosphate Sensor A simple overview of our present understanding of the hormonal factors involved in the regulation of phosphorus homeostasis is summarized in Figure 9.4. Phosphorus loading has been shown to immediately trigger an intestinal sensor to release an unidentified endocrine factor that has been shown to stimulate phosphaturia prior to the detection of phosphorus changes in plasma (Kumar, 2009; Berndt et al., 2007). The existence of the proposed phosphate-sensing mechanism within the intestine and the as yet unidentified endocrine factor that signals the kidney to increase phosphate excretion was discovered by administrating phosphate solutions directly into the duodenum of rats (Berndt et al., 2007). Immediate changes in the fractional excretion of phosphate were observed in the phosphate gavaged rats, but not in those rats gavaged with an equivalent amount of sodium chloride or infused with phosphate directly thus bypassing the intestine. Intestinal phosphate sensors offer a rapid response mechanism to maintain phosphorus balance without involving the classic endocrine hormones, PTH and calcitriol, the active form of vitamin D [1,25(OH)2D] (Berndt and Kumar, 2007).

Classical Endocrine Regulation The postulated intestinal phosphaturic factor provides a rapid response mechanism to adapt to changes in dietary phosphorus in contrast to the classical endocrine phosphorus regulating hormones, PTH and the vitamin D endocrine system shown in Figure 9.4 (Kumar, 2009). Classical endocrine feedback loops that function in the long-term adaptation to changes in dietary phosphate are illustrated in Figure 9.5. These classical endocrine changes occur over a longer period of time with chronic changes in the intake of phosphate (Calvo et al., 1988, 1990; Kemi et al., 2009). Oral loads of phosphate salts have been shown to depress ionized calcium and stimulate PTH release, an effect that appears to be dependent on the dose administered (Calvo and Heath, 1988; Kemi et al., 2009). PTH secretion has been shown to remain elevated and plasma calcitriol concentrations unstimulated with chronic consumption of high phosphorus, moderately low calcium diets (Calvoet al., 1990). Portale et al. (1989) demonstrated in humans that the normal stimulation of calcitriol synthesis by PTH is

150

Diet, Nutrients, and Bone Health

FGF-23

Dietary Inorganic Phosphate Load

PTH ECF PO4– 25OHD

1,25-(OH)2D Intestinal phosphaturic factor PO4 Excretion

FIGURE 9.4  Factors regulating phosphorus homeostasis in response to dietary loading with inorganic phosphorus. The intestinal phosphaturic factor is thought to be secreted within minutes of ingesting a phosphorus load and acts immediately to increase phosphate excretion. Direct stimulation of fibroblast growth factor 23 (FGF23) release from bone and parathyroid hormone (PTH) from the parathyroid glands requires only a slight elevation of phosphorus in the extracellular fluid which can occur postprandially within less than half an hour. Inhibition of calcitriol [1,25(OH)2D] occurs after persistent high phosphate loading over several days and results in lower phosphate absorption from the intestine and continued phosphaturia.

attenuated with chronic high dietary phosphorus intake. A decrease in plasma ionized calcium is the major stimulus for PTH release, but direct stimulation of PTH secretion by dietary phosphate loading was recently demonstrated in a rat model (Martin et al., 2005). Phosphate suppression of calcitriol was first observed with the chronic administration of therapeutic treatment (2 g phosphorus per day as inorganic phosphate salts) of patients with idiopathic hypercalciuria. When administered chronically, phosphorus only slightly elevated plasma parathyroid concentrations but significantly reduced plasma calcitriol (Van den Berg et al., 1980). Calcitriol functions to increase intestinal calcium absorption and to a lesser extent to increase phosphorus absorption. Significant reductions in calcitriol concentrations in the face of elevated parathyroid concentrations during chronic oral phosphate loading are puzzling because, until recently, PTH was considered the most potent stimulator of calcitriol synthesis (Calvo and Park, 1996). New factors, the phosphatonins, have recently been determined to also play a major role in the regulation of phosphorus homeostasis (Berndt et al., 2005).

Phosphatonin Regulation Emerging evidence suggests that specific factors secreted by osteocytes in bone also participate in maintaining phosphorus homeostasis, notably FGF23, shown in Figures 9.4 and 9.5. FGF23 is one of many factors, loosely termed the “phosphatonins,” which were discovered through studies of phosphate-wasting disorders (Schiavi and Kumar, 2004). Through classical endocrine feedback loops, dietary phosphate loading and calcitriol stimulate FGF23 secretion from osteocytes (Nishida et al., 2006; Saito et al., 2005). In turn, elevated FGF23 secretion inhibits renal synthesis of calcitriol

151

Inorganic Phosphorus

Dietary PO4

Parathyroid glands ds > Intakee Int

Dietary Ca Blood

+ +

Liver

[Ca++]

+

[PO4]

+ Intestine

PTH _ + 1,25 (OH)2D

_

25(OH)D

+ FGF23

Urine Ca Urine PO4 Urine PO4

Kidney

Bone

FIGURE 9.5  Schematic representation of the mechanisms regulating phosphorus homeostasis when dietary phosphorus intake exceeds dietary calcium intake. More rapid and more efficient absorption of phosphorus creates an imbalance in the blood, resulting in lower ionized calcium and higher serum phosphate levels, both of which stimulate parathyroid hormone release. Parathyroid hormone (PTH) acts immediately to stimulate renal reabsorption of mineral from bone. The less immediate action of PTH concerns the stimulation of renal alpha-1hydroxylase increasing the circulating level of calcitriol [1,25(OH)2D] which stimulates the intestinal absorption of both calcium and phosphorus. Slight elevation in blood phosphate concentration will directly stimulate osteocytes in bone to release fibroblast growth factor 23 (FGF23) and other phosphatonins, which inhibit PTH resorptive action in bone and renal synthesis of calcitriol but maintain phosphaturia. These actions can restore the balance between calcium and phosphorus in blood as long as there is sufficient renal function to excrete phosphorus.

from 25-hydroxy vitamin D (Shimada et al., 2004). Short-term phosphate loading in humans stimulates FGF23 secretion, phosphaturia, and suppression of renal calcitriol synthesis (Ferrari et al., 2005; Antoniucci et al., 2006; Burnett et al., 2006; Ito et al., 2007) through apparent parathyroidindependent mechanisms. Ben-Dov et al. (2007) demonstrated that FGF23 inhibits PTH secretion with chronic phosphate loading in rats. These investigators also showed that dietary phosphorus restriction exerts an opposite effect decreasing FGF23, increasing renal phosphate reabsorption, and indirectly increasing renal calcitriol synthesis purportedly by ceasing to inhibit parathyroid secretion. The importance of FGF23 to phosphorus homeostasis is specific to regulation of the wide fluctuations in dietary intake due to the extensive use of phosphate additives. High phosphorus intake is considered the main stimuli of FGF23 secretion as evidenced by the failure of FGF23 levels to increase when serum phosphate levels were raised by nondietary methods (Ito et al., 2007).

PHYSIOLOGIC EFFECTS OF HIGH INORGANIC PHOSPHORUS DIET Effects on Bone Health As world populations increasingly adopt Western culture and diet, such as highly processed convenience foods, we will see growing evidence of a disproportionate increase in phosphorus intake

152

Diet, Nutrients, and Bone Health

relative to calcium (Calvo and Carpenter, 2003). Few cross-sectional studies in adults relate high phosphorus intake to effects on bone mineral content or bone density, despite a well-established relationship between chronic dietary phosphate loading and adverse skeletal effects in animal models. Experimental studies examining low calcium, high phosphorus consumption have produced hormonal changes that are not conducive to the development or maintenance of peak bone mass, a situation that predisposes women to osteoporosis later in life (Calvo et al., 1990). Both acute and longer exposures to oral phosphorus loading in the presence or absence of adequate calcium intake produce a rise in PTH (Calvo and Carpenter, 2003). Bell et al. (1977) were the first to examine the physiologic effects of diets high in phosphorus containing food additives. Their findings are shown in Table 9.4 that demonstrate a clear effect of phosphorus intake on bone turnover (increased hydroxyproline excretion) and evidence of increased parathyroid activity (increased cyclic adenosine mono phosphate [AMP] excretion). Later, we also examined the effects of chronic high phosphate additive consumption using a more complex study design and direct measures of PTH and calcitriol (Calvo et al., 1988, 1990). In contrast to short-term feeding studies with phosphate salts, our chronic high phosphate additive consumption studies showed a significant reduction or no change in calcitriol levels, despite slight elevations in PTH. We speculate that in these earlier studies, persistent elevation in FGF23 may have inhibited PTH secretion and stimulatory action on calcitriol synthesis. Under conditions of low calcium intake, the dietary phosphate stimulation of FGF23 would impair the body’s main adaptive mechanism for adequate calcium absorption and optimal bone accretion, the increased synthesis of calcitriol. The effects of high phosphate additive intake on FGF23 secretion and function in phosphorus homeostasis merit further study. No prospective dietary studies have been long enough to accurately determine the effect of low calcium, high phosphorus intake on bone mass accretion in young adults. Some more recent crosssectional studies have shed light on our understanding of the potential adverse effects of some phosphate food additives. Tucker et al. (2006) measured bone mineral density of the spine and hip in older men and women in the Framingham Osteoporosis Study. They regressed bone density data on the frequency of soft drink consumption after multiple adjustment for confounding variables. They found that the intake of cola beverages that contain phosphoric acid but not other carbonated soft drinks, which do not contain phosphoric acid, is significantly associated with low bone mineral density in women but not men. In this study, total calcium intake was lower in women with the highest cola intakes and lowest bone mineral density. Again, we can only speculate that regular consumption of cola beverages with a dose of phosphoric acid may promote lower bone mass by evoking a strong FGF23 response impeding efficiency of calcium absorption by inhibiting calcitriol synthesis, a situation that would exacerbate bone loss if calcium intake was low. More recently, Pinheiro etal.

TABLE 9.4 Physiologic Effects of a Diet High in Foods Containing Phosphate Additives Biological Measures Calcium intake Phosphorus intake Urinary calcium Urinary phosphorus Urinary hydroxyproline Serum calcium Serum phosphorus Urinary cAMP

Control Diet

High PO4 Diet

677 mg/day 979 mg/day 179 mg/day 427 mg/day 27.9 mg/day 10.66 mg/100 ml 3.76 mg/100 ml 2.81 nmol/mg creatinine

745 2125 113 p < .01 1013 p < .01 33.4 p < .01 10.31 p < .01 4.43 3.44 p < .01

Source: Bell, R.R., et al. J Nutr, 107, 42–50, 1977.

Inorganic Phosphorus

153

(2009) evaluated the association between nutrient intake and osteoporotic fractures in a representative sample of older Brazilian men and women. They demonstrated a relationship between increasing phosphorus intake and bone fractures, showing a 9% increase in risk of fractures for every 100-mg increase in phosphorus intake. Given the rapid rate and high efficiency of absorption and frequency of consumption, specific phosphate additives such as phosphoric acid used in cola beverages merit further study of their influence on bone accretion. These current findings suggest that certain phosphate additives may significantly disrupt phosphorus homeostasis.

Effects on Cardiovascular Health Public health concern is growing over the association between subtle increases in serum phosphate levels within the normal range and increased risk of death and cardiovascular disease (Isakova et al., 2009). Hyperphosphatemia is a significant risk factor for kidney disease progression, vascular calcification, left ventricular hypertrophy, and mortality in chronic kidney disease (CKD) (Block et al., 1998; Giachelli et al., 2001; Jono et al., 2000). Moreover, Shuto et al. (2009) have demonstrated that dietary phosphorus can impair endothelial function. Key strategies to slowing the progression to cardiovascular disease in CKD focus on preventing vascular calcification and damage to the vascular endothelium. These strategies involve restricting phosphorus intake which is difficult given the hidden sources in our foods in the form of phosphorus food additives (Uribarri and Calvo, 2003). Vascular calcification is a major contributor to cardiovascular disease, and a growing body of evidence links small elevations in serum phosphate (high normal range, 3.5–4.5 mg/dl) in young adults with normal renal function with increased coronary artery calcium and increased risk of atherosclerosis (Foley et al., 2009; Giachelli, 2009). Other prospective studies in community-living adults with normal renal function (Dhingra et al., 2007) or patients with previous myocardial infarctions but normal kidney function (Tonelli et al., 2005) showed higher serum phosphorus levels (within normal range) associated with increased cardiovascular disease risk. A recent prospective study of 13,444 participants in the Atherosclerosis Risk in Communities and the Multi-Ethnic Study of Atherosclerosis cohorts reported higher phosphorus intakes associated with lower blood pressure levels when the dietary phosphorus was obtained through the intake of dairy products, but not phosphorus from other dietary sources (Alonso et al., 2010). These findings reinforce the need for better nutrient content data of foods to help distinguish effects of organic phosphorus from those of inorganic phosphorus on critical health measures. Many experts believe that we now have sufficient evidence that higher serum phosphorus within the established normal range promotes cardiovascular calcification, impaired endothelial function, and progression to cardiovascular disease in both normal and CKD patients to justify development of effective strategies to reduce serum phosphorus in the overall population (Tuttle and Short, 2009). Consideration should be given to the development of dietary strategies to reduce phosphorus intake in the overall population by adjusting the use of inorganic phosphorus additives and thus facilitate maintenance of phosphorus homeostasis.

SUMMARY New knowledge of the role of bone in the endocrine regulation of phosphorus homeostasis has brought new understanding of the mechanisms that could potentially lead to adverse health outcomes resulting from the continued increase in use of inorganic phosphorus in food processing. The phosphorus content of the U.S. diet continues to increase as a result of the growing consumption of highly processed foods such as fast foods and convenience foods. Greater use of phosphate additives in these foods is fueled by our fast-paced culture’s need for specific additive functions that allow us to speed the preparation, improve the texture, or restructure food. A growing number of studies are finding that consumption of specific foods or diets rich in phosphorus additives may significantly disrupt phosphorus homeostasis contributing to bone loss, impaired kidney function, and cardiovascular disease. Hyperphosphatemia occurring in chronic renal failure has long been

154

Diet, Nutrients, and Bone Health

recognized to be a serious risk factor for progression of kidney disease, vascular calcification, left ventricular hypertension, disruption of endothelial function, and mortality, with dietary phosphorus restriction considered to be the best corrective strategy. In both chronic renal failure patients and the general population, vascular calcification is a major contributor to cardiovascular disease. However, growing evidence now links small increases in serum phosphate in individuals with normal renal function to measures of increased coronary artery calcium and increased risk of atherosclerosis. Considering these findings associated with small changes in serum phosphorus, the development of strategies to reduce inorganic phosphorus intake in the general population merits consideration. Adjusting the use of inorganic phosphorus additives in food processing may be a simple approach to reducing phosphorus intake and optimizing the maintenance of phosphorus homeostasis.

REFERENCES Alonso, A., Nettleton, J.A., Ix, J.H., et al. 2010. Dietary phosphorus, blood pressure, and incidence of hypertension in Atherosclerosis Risk in Communities Study and the Multi-Ethnic Study of Atherosclerosis. Hypertension 55: 776–784. Antoniucci, D.M., Yamash*ta, T., and Portale, A.A. 2006. Dietary phosphorus regulates serum fibroblast growth factor-23 concentrations in healthy men. J Clin Endocrinol Metab 91: 3144–3149. Bell, R.R., Draper, H.H., Tzeng, D.M., et al. 1977. Physiological responses of human adults to foods containing phosphate additives. J Nutr 107: 42–50. Ben-Dov, I.Z., Gailtzer H., Lavi-Moshayoff, Y., et al. 2007. The parathyroid is a target organ for FG23 in rats. J Clin Invest 117: 403–4008. Bergwitz, C., and Juppner, H. 2010. Regulation of phosphate homeostasis by PTH, vitamin D and FGF-23. Annu Rev Med 61: 91–104. Berndt, T.J., and Kumar, R. 2007. Phosphatonins and the regulation of phosphate homeostasis. Annu Rev Physiol 69: 341–359. Berndt, T.J., and Kumar, R. 2009. Novel mechanisms in the regulation of phosphorus homeostasis. Physiology (Bethesda) 24: 17–25. Berndt, T.J., Schiavi, S., and Kumar, R. 2005. Phosphatonins and the regulation of phosphorus homeostasis. Am J Physiol (Renal Physiol) 289: F1170–F1182. Berndt, T.J., Thomas, L.F., Craig, T.A., et al. 2007. Evidence for a signaling axis by which intestinal phosphate rapidly modulates renal phosphate reabsorption. PNAS 104: 11085–11090. Block, G.A., Hulbert-Shearon, T.E., Levin, N.W., et al. 1998. Association of serum phosphorus and calcium X phosphate product with mortality risk in chronic hemodialysis patients: A national study. Am J Kidney Dis 31: 607–617. Burnett, S.M., Gunawardene, S.C., Bringhurst, F.R., et al. 2006. Regulation of C-terminal and intact FGF-23 by dietary phosphate in men and women. J Bone Miner Res 21: 1187–1196. Calvo, M.S. 1994. The effects of high phosphorus intake on calcium metabolism. In: Advances in Nutritional Research, ed. H.H. Draper, 183–207. Plenum Press, New York, NY. Calvo, M.S., and Carpenter, T.O. 2003. Influence of phosphorus on the skeleton. In: Nutritional Aspects of Bone Health, ed. S.I. New and J.-P. Bonjour, 229–265. Royal Chemistry Society, Cambridge, UK. Calvo, M.S., and Heath, H., III. 1988. Acute effects of oral phosphate–salt ingestion on serum phosphorus, serum ionized calcium and parathyroid hormone in young adults. Am J Clin Nutr 47: 1025–1029. Calvo, M.S., Kumar, R., and Heath, H., III. 1988. Elevated secretion and action of serum parathyroid hormone in young adults consuming high phosphorus, low calcium diets assembled from common foods. J Clin Endocrinol Metab 66: 823–829. Calvo, M.S., Kumar, R., and Heath, H., III. 1990. Persistently elevated parathyroid hormone secretion and action in young women after four weeks of ingesting high phosphorus, low calcium diets. J Clin Endocrinol Metab 70: 1334–1340. Calvo, M.S., and Park, Y.K. 1996. Changing phosphorus content of the U.S. diet: Potential for adverse effects on bone. J Nutr 126: 1168s–1180s. Dhingra, R., Sullivan, L.M., and Fox, C.S. 2007. Relations of serum phosphorus and calcium levels to the incidence of cardiovascular disease in the community. Arch Intern Med 167: 879–885. Endres, D.B., and Rude, R. 2006. Mineral and bone metabolism. In: Tietz Textbook of Clinical Chemistry and Molecular Diagnostics, ed. C.A. Burtis, E B. Ashwood, and D.E. Burns, 1891–1965. Elsevier Saunders, St. Louis, MO.

Inorganic Phosphorus

155

Ferrari, S.L., Bonjour, J.P., and Rizzoli, R. 2005. Fibroblast growth factor-23 relationship to dietary phosphate and renal phosphate handling in healthy young men. J Clin Endocrinol Metab 90: 1519–1524. Foley, R.N., Collins, A.J., Herzog, C.A., et al. 2009. Serum phosphorus levels associate with coronary atherosclerosis in young adults. J Am Soc Nephrol 20: 397–404. Giachelli, C.M. 2009. The emerging role of phosphate in vascular calcification. Kidney Int 75: 890–897. Giachelli, C.M., Jono, S., Shioi, A., et al. 2001. Vascular calcification and inorganic phosphate. Am J Kidney Dis 38: S34–S37. Institute of Medicine. 1997. Dietary reference intakes for calcium, phosphorus, magnesium, vitamin D, and fluoride. National Academy Press, Washington, DC. International Food Additive Council. 2008. Phosphates Use in Foods. http:www.foodadditives.org/phosphates/ phosphate_used_in_food.html (last accessed January 2, 2010). Isakova, T., Gutierrez, O.M., and Wolf, M. 2009. A blueprint for randomized trials targeting phosphorus metabolism in chronic disease. Kidney Int 76: 705–716. Ito, N., f*ckumoto, S., Takeuchi, Y., et al. 2007. Effect of acute changes of serum phosphate on fibroblast growth factor (FGF)-23 levels in humans. J Bone Miner Metab 25: 419–422. Jono, S., McKee, M.D., Murry, C.E., et al. 2000. Phosphate regulation of vascular smooth muscle cell calcification. Circul Res 87: E10–E17. Kalantar-Zadeh, K., Gutekunst, L., Mehrota, R., et al. 2010. Understanding sources of dietary phosphorus in the treatment of patients with chronic kidney disease. Clin J Am Soc Nephrol 5: 519–530. Karp, H.J., Vaihia, K.P., Kärkkäinen, M.J., et al. 2007. Acute effects of different phosphorus sources on calcium and bone metabolism in young women: A whole-foods approach. Calcif Tissue Int 80: 251–258. Kemi, V., Rita, H.J., Kärkkäinen, M.U.M., et al. 2009. Habitual high phosphorus intakes and foods with phosphate additives negatively affect serum parathyroid concentration: A cross-sectional study on healthy premenopausal women. Public Health Nutr 12: 1885–1892. Kumar, R. 2009. Phosphate sensing. Curr Opin Nephrol Hypertens 18: 281–284. Lemann, J. 1993. Intestinal absorption of calcium, magnesium and phosphorus. In: Primer on Metabolic Bone Disease and Disorders of Mineral Metabolism, ed. M.J. Favus, 46–50. Raven Press, New York, NY. Martin, D.R., Ritter, C.S., Slatopolsky, E., et al. 2005. Acute regulation of parathyroid hormone by dietary phosphate. Am J Physiol (Endocrinol Metab) 289: E729–E734. Moshegh, A., Goldman, J., Ahuja, J, et al. 2009. What we eat in America. NHANES 2005–2006. In: Usual Nutrient Intakes from Food and Water Compared to 1997 Dietary Reference Intakes for vitamin D, Calcium, Phosphorus, and Magnesium. U.S. Department of Agriculture Research Service. Nishida, Y., Taketani, Y., Yammanaka-Okumura, H., et al. 2006. Acute effects of oral phosphate loading on serum fibroblast growth factor 23 levels in healthy men. Kidney Int 20: 214–2147. Oenning, L.J., Vogel, J., and Calvo, M.S. 1988. Accuracy of methods estimating calcium and phosphorus intake in daily diets. J Am Diet Assoc 88: 1076–1078. Pinheiro, M.M., Schuch, N.J., Genaro, P.S., et al. 2009. Nutrient intakes related to osteoporotic fractures in men and women—The Brazilian Osteoporosis Study (Brazos). Nutr J 8(6): 1–8. Portale, A.A., Halloran, B.P., and Morris, R.C., Jr. 1989. Physiologic regulation of the serum concentration of 1,25-dihydroxyvitamin D by phosphorus in normal man. J Clin Invest 83: 1494–1499. Quarles, L.D. 2008. Endocrine functions of bone mineral metabolism regulation. J Clin Invest 118: 3820–3828. Sabbagh,Y., O’Brien, S.P., Song, W., et al. 2009. Intestinal Npt2b plays a major role in phosphate absorption and homeostasis. J Am Soc Nephrol 20: 2348–2358. Saito, H., Maeda, A., Ohtomo, S., et al. 2005. Circulating FGF-23 is regulated by 1 alpha, 25-dihydroxyvitamin D3 and phosphorus in vivo. J Biol Chem 280: 2543–2549. Sarathy, S., Sullivan, C., Leon, J.B., et al. 2008. Fast food phosphorus-containing additives, and the renal diet. J Renal Nutr 18: 466–470. Schiavi, S.C., and Kumar, R. 2004. The phosphatonin pathway: New insights in phosphate homeostasis. Kidney Int 65: 1–14. Shimada, T., Hasegawa, H., Yamazaki, Y., et al. 2004. FGF-a3 is a potent regulator of vitamin D metabolism and phosphate homeostasis. J Bone Miner Res 19: 429–435. Shuto, E., Taketani, Y., Tanaka, R., et al. 2009. Dietary phosphorus acutely impairs endothelial function. J Am Soc Nephrol 20: 1504–1512. Sullivan, C.M., Leon, J.B., and Sehgal, A.R. 2007. Phosphorus-containing food additives and the accuracy of nutrient databases: Implications for renal patients. J Renal Nutr 17: 350–354. Takeda, E., Yamamoto, H. Nashiki, K., et al. 2004. Inorganic phosphate homeostasis and the role of dietary phosphorus. J Cell Mol Med 8: 191–200.

156

Diet, Nutrients, and Bone Health

Tonelli, M., Sacks, F., Pfeffer, M., et al. 2005. Cholesterol and recurrent events trial investigators relation between serum phosphate levels and cardiovascular event rate in people with coronary disease. Circulation 112: 2627–2633. Tucker, K.L., Morita, K., Qiao, N., et al. 2006. Colas, but not other carbonated beverages, are associated with low mineral density in older women: The Framingham Osteoporosis Study. Am J Clin Nutr 84: 936–942. Tuttle, K., and Short, R. 2009. Longitudinal relationships among coronary artery calcification, serum phosphorus and kidney function. Clin J Am Soc Nephrol 4: 1968–1973. Uribarri, J. 2009. Phosphorus additives in food and their effect in dialysis patients. Clin J Am Soc Nephrol 4: 1290–1292. Uribarri, J., and Calvo, M.S. 2003. Hidden sources of phosphorus in the typical American diet. Does it matter in nephrology? Semin Dial 16: 186–188. Van den Berg, C.J., Kumar, R., Wilson, D.M., et al. 1980. Orthophosphate therapy decreases urinary calcium excretion and serum 1,25-dihydroxyvitamin D concentration in idiopathic hypercalciuria. J Clin Endocrinol Metab 51: 998–1001. Weiner, M.L., Salminen, W.F., Larson, P.R., et al. 2001. Toxicological review of inorganic phosphates. Food Chem Tox 39: 759–786.

10

Vitamin D and Bone Michael F. Holick

CONTENTS Introduction..................................................................................................................................... 157 Vitamin D, Calcium, and Phosphorus Metabolism......................................................................... 157 Vitamin D’s Effect on Bone Metabolism........................................................................................ 159 Causes and Consequences of Rickets/Osteomalacia...................................................................... 161 Historical Perspective................................................................................................................. 161 Vitamin D Deficiency................................................................................................................. 162 Inherited Disorders of Vitamin D Metabolism and Recognition.................................................... 163 Vitamin D–25-Hydroxylase Deficiency..................................................................................... 163 Vitamin-D–Dependent Rickets Type I: Pseudovitamin-D–Deficiency Rickets......................... 163 Vitamin-D–Dependent Rickets Type II: Hereditary Vitamin-D–Resistant Rickets................... 163 Vitamin-D–Dependent Rickets Type III.................................................................................... 165 Strategies for Treatment and Prevention of Rickets/Osteomalacia................................................. 165 Responsiveness to Calcium and Vitamin D................................................................................ 165 Treatment of Vitamin D Deficiency........................................................................................... 167 Conclusion...................................................................................................................................... 167 Acknowledgment............................................................................................................................ 168 References....................................................................................................................................... 168

INTRODUCTION Vitamin D deficiency causes rickets in children and osteomalacia in adults. It also can precipitate and exacerbate osteoporosis including risk of fracture (Holick, 2007). Rickets/osteomalacia by definition means that osteoblasts have laid down a collagen matrix, but a defect exists in its ability to be mineralized. In children, a defect in the mineralization of the osteoid in the long bones and the failure or delay in the mineralization of endochrondal bone formation at the growth plate leads to the classic skeletal deformities of rickets (Holick, 2006) (Figure 10.1). However, in adults, the mineralization defect takes on a different character because of the failure of mineralization of newly formed osteoid at sites of bone turnover of periosteal or endosteal apposition. Several possible causes of poor or absent skeletal mineralization may lead to both rickets and osteomalacia. Although the major cause of rickets/osteomalacia is a deficiency of vitamin D, rare causes include a defect in vitamin D metabolism or in its recognition by calcium-regulating tissues.

VITAMIN D, CALCIUM, AND PHOSPHORUS METABOLISM The major components of skeletal mineral are calcium and phosphate ions. Thus, any alteration in the calcium–phosphate product in the circulation can result in a mineralization defect of the skeleton. Vitamin D plays a critical role in maintaining both serum calcium and phosphate concentrations (Holick, 2006, 2007; Holick and Garabedian, 2006) (Figure 10.2). Vitamin D is obtained by exposure of the skin to UVB from sunlight, resulting in the biochemical conversion 157

158

Diet, Nutrients, and Bone Health

FIGURE 10.1  Child with rickets demonstrating the bowed leg deformity and rachitic rosary. The child is being held up due to severe muscle weakness. (Reproduced from Holick, M.F., New Engl J Med, 357, 266–281, 2004. With permission.)

of 7-dehydrocholesterol to previtamin D3 (Figure 10.2). Previtamin D3, being thermodynamically unstable, is rapidly converted to vitamin D3. Once formed, vitamin D is ejected by the epidermal cell into the extracellular space and, by diffusion, enters the circulation bound to the vitamin-Dbinding protein (DBP) (Holick, 2006, 2007; Holick and Garabedian, 2006). Vitamin D3 and vitamin D2 (D represents D2 or D3) in the diet are ingested, and the fat-soluble vitamins are incorporated into chylomicrons and absorbed into the lymphatics. The lymphatic drainage into the thoracic venous system permits the entrance of vitamin D into the circulation where it is bound to the DBP and lipoproteins (Haddad et al., 1993). Vitamin D is converted in the liver by a vitamin D-25-hydroxylase (25-OHase) to form the major circulating form of vitamin D, 25-hydroxyvitamin D [25(OH)D]. At least four different 25-OHases have been identified in both mitochrondia and in microsomes (Holick, 2006, 2007). 25(OH)D is, however, biologically inert and requires hydroxylation on carbon 1 in the kidneys by the mitrochrondial enzyme 25-hydroxyvitamin D-1α-hydroxylase (1-OHase), also known as cyp27B1. This hydroxylation step results in the formation of 1α,25-dihydroxyvitamin D [1,25(OH)2D], which is the biologically active form, that is, hormone, of vitamin D responsible for regulating calcium and phosphorus homeostasis (Holick, 2006, 2007; Holick and Garabedian, 2006). 1,25(OH)2D enters the circulation and is bound to the DBP and travels to its target tissues. In the small intestine, 1,25(OH)2D interacts with its vitamin D nuclear receptor (VDR) that results in the expression of several gene products including the epithelial calcium channel, calbindin9k, and a calcium-dependent ATPase (Christakos et al., 2003; Holick, 2006, 2007; Holick and Garabedian, 2006) (Figure 10.2). 1,25(OH)2D increases the efficiency of intestinal calcium absorption. In a vitamin-D-deficient state, the small intestine is able to passively absorb about 10%–15% of dietary calcium. Vitamin D sufficiency enhances the absorption of calcium in the small intestine to about 30%–40% (Holick, 2007). 1,25(OH)2D enhances the efficiency of intestinal calcium absorption principally in the duodenum and to a lesser degree in the jejunum and ileum. 1,25(OH)2D stimulates phosphate absorption

159

Vitamin D and Bone

Solar UVB radiation

Skin 7-DHC

PreD3 Heat Vitamin D2

Vitamin D

D Fat cell

Diet

25-OHase

Vitamin D3 Fish Milk OJ

Inactive photoproducts

Liver

Supplement

Pi, Ca, FGF23 & other factors

25(OH)D 1-OHase

+/–

Preosteoclast 1,25(OH)2D RANKL RANK

Kidneys

1,25(OH)2D

Osteoblast

Ha se

ECaC CaBP

PTH Parathy oid glands

tion cifica Cal

-O

Calcitroic acid

PTH Osteoclast

24

Ca2+ HPO42– Blood calcium phosphorus

Intestine

Bile

Ca2+ HPO42– Absorption

Excreted

FIGURE 10.2  Schematic representation of the synthesis and metabolism of vitamin D for regulating calcium, phosphorus, and bone metabolism. (Reproduced from Holick, M.F., New Engl J Med, AQ9 357, 266– 281, 2007. With permission.)

in the jejunum and ileum. The small intestine passively absorbs about 60% of dietary phosphate. 1,25(OH)2D enhances the efficiency of phosphate absorption by an additional 20% to about 80%. When adequate calcium and phosphate are consumed in the diet and vitamin D sufficiency exists, the healthy serum calcium normal range is ~8.6–10.2 mg/dL, and the healthy serum phosphate range is ~2.5–4.5 mg/dL. The calcium–phosphate product (Ca × P) in the circulation and in the extravascular space plays a major role in the normal mineralization of osteoid laid down by osteoblasts.

VITAMIN D’S EFFECT ON BONE METABOLISM In rodents and humans, vitamin D is not necessary for the mineralization of the osteoid matrix (Balsan et al., 1986; Holick, 2006). This lack of a direct skeletal effect was demonstrated when

160

Diet, Nutrients, and Bone Health

vitamin-D-deficient rats were either infused with calcium or phosphorus to maintain a normal calcium–phosphate product in the circulation or when they received high-calcium lactose, highphosphorus diet that maintained a normal serum calcium–phosphate product (Holick, 2006). In both circ*mstances, bone histology revealed that the mineralization occurred normally without any significant unmineralized osteoid. Vitamin-D-resistant rickets patients have a mutation of their VDR and have severe rickets and osteomalacia. When these patients were infused with calcium and phosphorus to maintain a normal calcium–phosphate product, the unmineralized osteoid became mineralized (Balsan et al., 1986). 1,25(OH)2D interacts with its VDR in osteoblasts to increase the expression of alkaline phosphatase, osteocalcin, and receptor activator of NFκB ligand (RANKL) (Khosla, 2001). Alkaline phosphatase produced by osteoblasts is important in bone mineralization because patients with a decrease in the bone-specific alkaline phosphatase known as hypophosphatasia suffer from a mineralization defect of the osteoid (Parfitt, 1998). Osteocalcin is the major noncollagenous protein in the skeleton. Although its function is not well understood, it appears to have a role in osteoclastic activity (Aubin et al., 2006). RANKL, once expressed on the surface of an osteoblast, interacts with its receptor RANK on osteoclast precursors. This intimate interaction leads to signal transduction that results in the formation of multinucleated mature osteoclasts (Khosla, 2001; Holick, 2006, 2007; Holick and Garabedian, 2006) (Figure 10.2). The osteoclasts, under the direction of a variety of cytokines (Khosla, 2001; Aubin et al., 2006), increase the destruction of the skeleton by releasing hydrochloric acid to degrade and dissolve the mineral matrix and collagenases and cathepsin K to dissolve the matrix. During exposure to sunlight, 7-dehydrocholesterol in the skin is converted to previtamin D3. PreD3 immediately converts by a heat dependent process to vitamin D3. Excessive exposure to sunlight degrades previtamin D3 and vitamin D3 into inactive photoproducts. Vitamin D2 and vitamin D3 from dietary sources are incorporated into chylomicrons, transported by the lymphatic system into the venus circulation. Vitamin D (D represents D2 or D3) made in the skin or ingested in the diet can be stored in and then released from fat cells. Vitamin D in the circulation is bound to the vitamin-D-binding protein which transports it to the liver where vitamin D is converted by the vitamin D-25-hydroxylase to 25-hydroxyvitamin D [25(OH)D]. This is the major circulating form of vitamin D that is used by clinicians to measure vitamin D status (although most reference laboratories report the normal range to be 20–100 ng/mL, the preferred healthful range is 30–60 ng/mL). It is biologically inactive and must be converted in the kidneys by the 25-hydroxyvitamin D-1α-hydroxylase (1-OHase) to its biologically active form 1,25-dihydroxyvitamin D [1,25(OH)2D]. Serum phosphorus, calcium fibroblast growth factors (FGF-23), and other factors can either increase (+) or decrease (−) the renal production of 1,25(OH)2D. 1,25(OH)2D feedback regulates its own synthesis and decreases the synthesis and secretion of parathyroid hormone (PTH) in the parathyroid glands. 1,25(OH)2D increases the expression of the 25-hydroxyvitamin D-24-hydroxylase (24-OHase) to catabolize, 1,25(OH)2D to the water-soluble biologically inactive calcitroic acid which is excreted in the bile. 1,25(OH)2D enhances intestinal calcium absorption in the small intestine by stimulating the expression of the epithelial calcium channel (ECaC) and the calbindin 9K (calcium-binding protein; CaBP). 1,25(OH)2D is recognized by its receptor in osteoblasts, causing an increase in the expression of receptor activator of NFκB ligand (RANKL). Its receptor RANK on the preosteoclast binds RANKL, which induces the preosteoclast to become a mature osteoclast. The mature osteoclast removes calcium and phosphorus from the bone to maintain blood calcium and phosphorus levels. Adequate calcium and phosphorus levels promote the mineralization of the skeleton. Thus, the effect of vitamin D on bone metabolism is to maintain normal serum calcium and phosphate ion concentrations. Vitamin D accomplishes this by increasing intestinal calcium and phosphate absorption and by mobilizing calcium and phosphorus from the skeleton (Holick, 2006, 2007; Holick and Garabedian, 2006).

Vitamin D and Bone

161

CAUSES AND CONSEQUENCES OF RICKETS/OSTEOMALACIA Historical Perspective Historians have reported bone deformities similar to rickets as early as the second century, but the disease was not considered a significant health problem until the industrialization of northern Europe. Whistler, Glissen, and DeBoot recognized in the mid-1600s that many children who lived in the crowded and polluted cities of northern Europe developed a severe bone-deforming disease (rickety bones) that was characterized by growth retardation, enlargement of the epiphyses of the long bones, deformities of the legs, bending of the spine, knobby projections of the ribcage, and weak and toneless muscles (Rajakumar, 2003; Holick, 2006, 2007) (see Figure 10.1). The incidence of the debilitating bone disease increased dramatically in northern Europe and North America during the industrial revolution, and by the latter part of the 19th century, autopsy studies done in Boston and Leyden, the Netherlands, showed that 80%–90% of children had rickets. In addition, the pelvic bone structure was flattened, and this resulted in a high incidence of infant and maternal morbidity and mortality that led to the widespread use of cesarian sectioning for infant delivery (Rajakumar, 2003; Holick, 2006). In 1822, Sniadecki (1939) was the first to recognize the importance of sun exposure for the prevention and cure of rickets. Cod liver oil had been intermittently used as a home remedy for rickets, but it often was ineffective. Because of its high prevalence and devastating consequences, many scientists and physicians became interested in finding the cause and cure for rickets. In 1919, Huldschinsky (1919, 1928) found that exposing children to radiation from a sun quartz lamp (mercury arc lamp) or carbon arc lamp was effective in treating rickets as demonstrated by improvement in the children’s x-rays (Figure 10.3). He concluded that exposure to ultraviolet radiation was an “infallible remedy” against all forms of rickets in children. Two years later, Hess and Unger (1921) exposed seven rachitic children in New York City to varying periods of sunshine and reported that x-ray examination showed marked improvement in the rickets of each child, as evidenced by calcification of the epiphyses. Great confusion was expressed as to how either exposure to ultraviolet radiation and sunlight or a dietary factor could prevent and cure rickets. Powers et al. (1921) treated rachitic rats with ultraviolet radiation or cod liver oil and observed the same effect. Hess and Weinstock (1924) and Steenbock and Black (1924) reported that ultraviolet irradiation of various foods and oils imparted antirachitic activity. This discovery led to enhancing the antirachitic activity of milk by exposing milk to ultraviolet radiation or feeding cows ultraviolet-irradiated yeast. This simple fortification practice eradicated rickets (Rajakumar, 2003; Holick, 2006).

FIGURE 10.3  Florid rickets of the hand and wrist (left panel) and the same wrist and hand radiograph taken after treatment with 1-hour UVR two times a week for 8 weeks. Note the mineralization of the carpel bones and epiphyseal plates (right panel). (Reproduced from Holick, M.F., New Engl J Med, 357, 266–281, 2006. With permission.)

162

Diet, Nutrients, and Bone Health

Many physicians have assumed that rickets no longer remains a health problem for children in the United States. However, rickets is becoming much more common due to the misconception that breast feeding provides infants with all of their nutritional requirements (Hollis and Wagner, 2004; Holick, 2006). Human breast milk is now known to contain little, if any, vitamin D and, thus, in neonates, especially those of color, are at high risk for vitamin-D-deficiency rickets (Kreiter et al., 2000; Rajakumar, 2003; Holick, 2006). Recently, this high risk of vitamin D deficiency has been recognized by the American Academy of Pediatrics, which now recommends that all infants from the time they are born receive 400 IU of vitamin D a day (Wagner and Greer, 2008).

Vitamin D Deficiency Vitamin D deficiency, the most common cause of rickets, prevents the efficient absorption of dietary calcium and phosphorus. In a vitamin-D-deficient state, only 10%–15% of dietary calcium and 50%–60% of dietary phosphorus are absorbed (Holick, 2007). The poor absorption of calcium causes a decrease in serum-ionized calcium. This decline is immediately recognized by the calcium sensor in the parathyroid glands, resulting in an increase in the expression, synthesis, and secretion of parathyroid hormone (PTH) (Brown et al., 1993; Holick, 2006, 2007). PTH conserves calcium by increasing tubular reabsorption of calcium in both the proximal and distal convoluted tubules. PTH, like 1,25(OH)2D, enhances the expression of RANKL by osteoblasts to increase the production of mature osteoclasts that mobilize calcium stores from the skeleton. PTH also decreases phosphate reabsorption in the kidney, causing loss of phosphorus into the urine. The serum calcium is usually normal in a vitamin-D-deficient infant or child. However, the serum phosphate is low, and thus, there is an inadequate calcium × phosphate product that is necessary to mineralize the osteoid laid down by osteoblasts (Figure 10.4). Thus, typically, infants with vitamin-D-deficiency rickets have a normal serum calcium, low normal or low fasting serum phosphorus, elevated alkaline phosphatase, and low 25(OH)D (<15 ng/mL) (Hess and Lundagen, 1922; (a)

(b)

(c)

FIGURE 10.4  Bone histology demonstrating (a) increased osteoclastic bone resorption due to secondary hyperparathyroidism, (b) normal mineralized trabecular bone, and (c) osteomalacia with widened unmineralized osteoid (light gray areas). (Reproduced from Holick, M.F., New Engl J Med, 357, 266–281, 2004. With permission.)

Vitamin D and Bone

163

Hess, 1929; Kreiter et al., 2000; Rajakumar, 2003; Holick, 2006; Wagner et al., 2008). The secondary hyperparathyroidism stimulates the kidneys to produce 1,25(OH)2D, and thus 1,25(OH)2D levels are normal or often elevated. Only when the calcium stores in the skeleton are totally depleted will the infant or child become hypocalcemic, which can lead to tetany, tetanic seizures, and death (Hess, 1929; Rajakumar, 2003; Holick, 2006).

INHERITED DISORDERS OF VITAMIN D METABOLISM AND RECOGNITION Vitamin D–25-Hydroxylase Deficiency At least four different hepatic enzymes have been found in the mitochondria and microsomes that are capable of metabolizing vitamin D to 25(OH)D (Holick, 2006, 2007; Jones, 2007). This is the likely reason there have only been rare reports of a 25-hydroxylase deficiency causing rickets in children (Casella et al., 1994).

Vitamin-D–Dependent Rickets Type I: Pseudovitamin-D–Deficiency Rickets Before the discovery that vitamin D needed to be metabolized in the liver and kidneys before it could carry out its biologic actions on calcium and phosphorus metabolism, reports were published of children who had rickets but did not respond to physiologic doses of vitamin D. The children had hypocalcemia, hypophosphatemia, elevated alkaline phosphatase, elevated PTH, and severe rachitic changes on x-ray. However, many of these children responded to very large pharmacologic doses of vitamin D. As a result, these children received the diagnosis of vitamin-D-dependent rickets because of their need for much higher amounts of vitamin D to treat your rickets. With the revelation that vitamin D needed to be metabolized in the kidneys to its active form, it was speculated that vitamin-D-dependent rickets was caused by a mutation of the 1-OHase, resulting in either inadequate production of 1,25(OH)2D or the lack of production of 1,25(OH)2D (Figure 10.5). The first insight into the cause of this disorder was when 1,25(OH)2D3 was chemically synthesized and provided to these patients (Fraser et al., 1973). Within several months, there was a dramatic improvement in their serum calcium and phosphorus level with a decrease in alkaline phosphatase and PTH. Thus, vitamin-D-dependent rickets was concluded to be caused by a hereditary defect in the 1-OHase. The cloning of the 1-OHase gene led to the identification of inactivating mutations that confirmed the hypothesis for the cause of rare genetic disorder (Kitanaka et al., 1998). Patients with pseudovitamin-D-deficiency rickets (PDDR) present in their first year of life with severe hypocalcemia that can cause seizures and carpal pedal spasms, hypophosphatemia, elevated alkaline phosphatase, and PTH. Their blood level of 25(OH)D is usually normal, and treating them with physiologic doses of vitamin D had little effect on correcting their abnormal biochemistries (Holick, 2007). The hallmark for making the diagnosis is a low or undetectable blood level of 1,25(OH)2D. If these patients are not appropriately treated with replacement doses of 1,25(OH)2D3, they will have the same skeletal deformities seen in children with severe vitamin D deficiency (Fraser et al., 1973; Holick, 2006, 2007). These patients respond well to replacement doses of 1–2 mcg of 1,25(OH)2D3 (calcitriol) along with adequate calcium intake. The serum calcium levels begin to rise within 24 hours, and radiologic healing is observed by 3 months (Fraser et al., 1973; Holick, 2006, 2007).

Vitamin-D–Dependent Rickets Type II: Hereditary Vitamin-D–Resistant Rickets Children who had severe biochemical and skeletal abnormalities associated with vitamin-D-deficiency rickets and who did not respond to physiologic doses of vitamin D and only rarely responded to pharmacologic doses of vitamin D were also considered to have vitamin-D-dependent rickets.

164

Diet, Nutrients, and Bone Health

Serum 25(OH)D (ng/mL)

(a) 20 18 16 14 12 10 8 6 4

Serum 25(OH)D (ng/mL)

(b)

2 0

25(OH)D3

25(OH)D2

34 32 30 28 26 24 22 20 18 16

Serum 25(OH)D (ng/mL)

20

3

4 5 6 7 Time (weeks)

8

9

10 11

25(OH)D2 0

25

2

25(OH)D3

(c)

1

1

2

3

4 5 6 7 Time (weeks)

8

9

10 11

8

9

10 11

25(OH)D3

15 25(OH)D2

10 5 0

1

2

3

4 5 6 7 Time (weeks)

FIGURE 10.5  Effect of vitamin D2 or vitamin D3 on serum 25-hydroxyvitamin D2 [25(OH)D2] and 25-hydroxyvitamin D3 [25(OH)D3] levels. Serum levels of 25(OH)D2 (-•-)and serum 25(OH)D3 (-▪-) were measured in healthy subjects who received 1000 IU of vitamin D2 (a), 1000 IU of vitamin D3 (b), or 500 IU of vitamin D2 + 500 IU of vitamin D3 (c) daily for 3 months. Results are presented as means ± SEM over time. In panel A, *p < .0001 comparing 25(OH)D2 from baseline to 3 months. In panel B, *p < .0001 comparing 25(OH)D3 from baseline to 3 months. In panel C, p = .0014 comparing 25(OH)D3 and placebo from baseline to 3 months, **p = 0.0031 comparing 25(OH)D2 and placebo from baseline to 3 months. Note that the serum 25(OH)D2 levels <4 ng/mL were determined by subtracting the total 25(OH)D3 from the total 25(OH)D levels. (Reproduced with permission from Holick, M.F., et al. J Clin Endocrinol Met 93, 667–668, 2008).

However, unlike children with PDDR who had a low or undetectable blood level of 1,25(OH)2D, these children had a markedly elevated blood level of 1,25(OH)2D (Brooks et al., 1978). It was assumed that these children must have a genetic defect causing a lack of responsiveness to the calcium and bone metabolism to 1,25(OH)2D. The most likely cause was a defective or absent recognition of 1,25(OH)2D due to a mutation of the VDR. These patients had variable responses to

Vitamin D and Bone

165

replacement doses of 1,25(OH)2D3. A multitude of point mutations of the VDR that cause disruption of hormone-binding or DNA-binding leading to partial resistance to 1,25(OH)2D3 replacement therapy have been identified (Brooks et al., 1978; Holick, 2006, 2007). In some cases, these patients respond to pharmacologic doses of 1,25(OH)2D3. However, other patients do not respond because they have a point mutation that prevents the production of VDR or prevents the VDR from either binding 1,25(OH)2D or permitting the VDR– retinoic acid X receptor (RXR)–1,25(OH)2D complex from binding to its responsive element within the DNA. Typically, these patients are resistant to both physiologic and pharmacologic doses of 1,25(OH)2D3 (Brooks et al., 1978; Holick, 2006, 2007). Hereditary vitamin-D-resistant rickets patients have all of the biochemical and clinical manifestations of those with PDDR, with the exception of often having a low blood level of 25(OH)D and a markedly elevated level of 1,25(OH)2D. Another clinical manifestation that is not seen in PDDR patients is that some patients develop alopecia beginning the first year of life that progresses to alopecia totalis (Brooks et al., 1978; Holick, 2006).

Vitamin-D–Dependent Rickets Type III One case of vitamin-D–resistant rickets has been reported in which an abnormal expression of the hormone-responsive element-binding protein that binds to the vitamin-D-responsive element exists. The abnormal expression prevents the VDR–RXR–1,25(OH)2D complex from binding to its responsive element. This patient had a normal VDR expression and was completely resistant to 1,25(OH)2D3 (Chen et al., 2003).

STRATEGIES FOR TREATMENT AND PREVENTION OF RICKETS/OSTEOMALACIA The distinction between vitamin D insufficiency and vitamin D deficiency needs to be kept in mind when recommending supplemental vitamin D, with or without calcium, for the general public to prevent rickets/osteomalacia. Treatment of this disorder typically requires significantly greater doses of vitamin D until the clinical measurements return to the normal ranges.

Responsiveness to Calcium and Vitamin D The major cause of rickets/osteomalacia is vitamin D deficiency (Hess, 1929; Kreiter et al., 2000; Rajakumar, 2003; Holick, 2007; Wagner and Greer, 2008). Vitamin D deficiency is defined as a serum 25(OH)D < 20 ng/mL. However, to maximize intestinal calcium absorption and to minimize circulating PTH levels, it is desirable to maintain a 25(OH)D > 30 ng/mL (Heaney et al., 2003; Holick, 2007). Thus, vitamin D insufficiency is defined as a 25(OH)D of 21–29 ng/mL (Holick, 2007). To achieve vitamin D sufficiency, at least 1000–2000 IU of vitamin D/day or its equivalent is needed (Heaney et al., 2003; Holick, 2006, 2007; Holick et al., 2008). For every 100 IU of vitamin D2 or vitamin D3 ingested, the blood level of 25(OH)D increases by 1 ng/mL (Heaney et al., 2003; Holick et al., 2008). When given daily, vitamin D2 is equally effective as vitamin D3 in maintaining serum 25(OH)D levels. However, healthy adults at the end of the winter in Boston receiving 1000 of vitamin D2 or 1000 IU of vitamin D3/day or 500 IU vitamin D2 and 500 IU vitamin D3/day were unable to achieve a sustained blood level of 25(OH)D > 30 ng/mL (Holick et al., 2008) (Figure 10.5). Most children and adults have a circulating blood level of 25(OH)D of between 15 and 25 ng/mL (Holick, 2006, 2007). Thus, to achieve a blood level of 25(OH)D > 30 ng/mL, it has been recommended that both children and adults take 1000 IU of vitamin D/day along with a multivitamin containing 400 IU of vitamin D. For obese children and adults, the amount of vitamin D ingested should be increased by two- to three-fold because body fat will sequester vitamin D, and obese children and adults need at least two to three times more vitamin D to sustain their blood level of 25(OH)D > 30 ng/mL (Holick, 2006, 2007).

166

Diet, Nutrients, and Bone Health (a) 65 60

n=43

25(OH)D (ng/mL)

55 50

n=40

45 40 35 30

n=55

n=4 n=7

n=58 n=55

n=3

n=5

n=86

25

Insufficient

20 0

n=8

n=29

Deficient 0

6

12

18

24

30

36

42

48

54

60

72

Time (Months) (b) 65

n=6

60 25(OH)D (ng/mL)

55 n=20 n=17

45 40 35 30

n=25

n=28

n=6

n=28

n=45

25

Insufficient

20 0

n=4 n=2

n=3

50

6

12

18

Deficient 24 30 36

42

48

60

72

Time (Months) 10.4 10.2 n=6 10 n=13 n=54 n=9 n=7 9.8 n=53n=39 n=2 n=74 n=57 n=29 9.6 n=5 9.4 9.2 8 8.8 8.6 8.4 8.2 8 0 6 12 18 24 30 36 42 48 54 60 72

Calcium (mg/dL)

(c)

Time (Months)

FIGURE 10.6  (a) Mean serum 25-hydroxyvitamin D [25(OH)D] levels in all patients. (b) Mean serum

25(OH)D levels in patients receiving maintenance therapy only. (c) Serum calcium levels: Results for all 86 patients who were treated with 50,000 IU of vitamin D2. (Adapted from Pietras, S.M., et al., Arch Intern Med, 169, 1806–1808, 2009 and reproduced from Holick, M.F., New Engl J Med, 357, 266–281, 2007. With permission.)

Vitamin D and Bone

167

Treatment of Vitamin D Deficiency To treat vitamin D deficiency, 50,000 IU of vitamin D2, which is the only pharmaceutical form of vitamin D available in the United States, once a week for 8 weeks will often fill the empty vitamin D tank and raise the blood level above 30 ng/mL (Malabanan et al., 1998). Patients who are severely vitamin D deficient with a 25(OH)D < 10 ng/mL or patients who are on medications that would enhance vitamin D destruction and obese patients may require an additional 8 weeks of therapy with 50,000 IU of vitamin D2 (Holick, 2007). Once the blood level of 25(OH)D is above 30 ng/mL, vitamin D sufficiency can be maintained by giving patients 50,000 IU of vitamin D once every 2 weeks. In our clinic, we have found over a 6-year period that the blood levels are sustained between 40 and 50 ng/mL on this regime (Pietras et al., 2009) (Figure 10.6). Alternatively, to treat vitamin D deficiency with vitamin D supplement, 5000 IU of vitamin D/day for 2 months followed by 2000 IU of vitamin D/day will treat vitamin D deficiency and maintain vitamin D sufficiency in most adults. The graph depicted in Figure 10.6a includes all patients treated with 50,000 IU vitamin D2 every 2 weeks (maintenance therapy, N = 86). Forty one of the patients were vitamin D insufficient or deficient and first received 50,000 IU vitamin D2 weekly for 8 weeks before being placed on maintenance therapy of 50,000 IU vitamin D2 every 2 weeks. Error bars represent the mean ± SEM. Time 0 is the initiation of treatment. Results are shown as mean values averaged for intervals of 6 months. The mean 25(OH)D of each 6-month interval was compared with initial mean 25(OH)D and showed a significant difference of p < .001 for all time points. In patients receiving maintenance therapy, as shown in Figure 10.6b, thirty-eight patients were vitamin D insufficient [25(OH)D levels <21–29 ng/ mL] and seven patients were vitamin D sufficient [25(OH)D levels ≥ 30 ng/mL], who were treated only with maintenance therapy of 50,000 IU vitamin D2 every 2 weeks. Error bars represent the mean ± SEM. Time 0 is the initiation of treatment. Results shown are interval mean values averaged for intervals of 6 months. The mean 25(OH)D in each 6-month interval was compared with the mean initial 25(OH)D and showed a significant difference, p < .001, for all time points up to 48 months. The data for interval months 60 and 72 were pooled, and there was a significant difference p < .01 compared with the baseline value. Error bars shown in Figure 10.6c represent standard error of the mean. Time 0 is the initiation of treatment. Results shown are mean values averaged for intervals of 6 months. Normal serum calcium: 8.5–10.2 mg/dL. The American Academy of Pediatrics recommends that all infants and children receive 400 IU of vitamin D/day (Wagner and Greer, 2008). This daily dose will prevent rickets in infants and children. Children and adults who have calcium-induced rickets/osteomalacia should be treated with the adequate intake recommendation for calcium, which is 800 mg/day for children under the age of 12 years, 1300 mg/day for teenagers, 1000 mg/day for adults 19–50 years, and 1200 mg/day for adults over the age of 50 years (Institute of Medicine, 1997). Individuals should also be taking an adequate amount of vitamin D, that is, 800–1000 IU per day, for teenagers and adults to maximize the benefit of the calcium on skeletal health.

CONCLUSION Vitamin D is critically important for bone health. Vitamin D is responsible for regulating intestinal calcium absorption and bone calcium mobilization for the purpose of maintaining serum calcium in a physiologic range to support neuromuscular activity, signal transduction, as well as maintenance of a mineralized skeleton. Rickets/osteomalacia is caused by a defect in the mineralization of osteoid laid down by mature osteoblasts. Once the osteoid is laid down, an adequate calcium– phosphate product in the circulation results in the normal mineralization, with deposition of calcium hydroxyapatite forming mineralized bone. The most common causes of rickets/osteomalacia result from either vitamin D deficiency or calcium deficiency. Any acquired or inherited disorder that alters vitamin D metabolism or its recognition or phosphorus metabolism also causes a mineralization defect of the skeleton. Osteomalacia cannot be distinguished from osteopenia or osteoporosis based either by x-ray examination or by bone mineral density determination. Physicians should be

168

Diet, Nutrients, and Bone Health

alert to rickets in children and osteomalacia in adults because these disorders are more common than expected. Treatment of vitamin D deficiency and correction of calcium and phosphorus intake inadequacies, that is, nutritional deficiencies, result in complete resolution of rickets/osteomalacia. To maintain maximum skeletal health, all children should ingest at least 400 IU of vitamin D a day and adults should ingest at least 1000 IU of vitamin D a day. However, to take advantage of all of the health benefits of vitamin D, children should be on at least 1000 IU of vitamin D a day and adults should be on 2000 IU of vitamin D a day. Vitamin D deficiency/insufficiency is now recognized as the most common medical condition worldwide. This deficiency/insufficiency not only has significant consequences for skeletal health of children and adults, but it also has been associated with many chronic diseases. An estimated 50% or more of children and adults in the United States are vitamin D deficient or insufficient. African American children and adults are at especially high risk because of the sun-screening effect of their skin pigment in reducing the efficiency of the sun to produce vitamin D in their skin. Vitamin D deficiency and insufficiency have been linked to increased risk of autoimmune diseases, including type I diabetes; multiple sclerosis and rheumatoid arthritis; cancers of the colon, prostate, and breast; hypertension; heart disease; and stroke, as well as to increased risk of developing infectious diseases including tuberculosis, upper respiratory tract infections, and influenza. To take advantage of the nonskeletal health benefits of vitamin D, children are recommended to receive at least 1000 IU of vitamin D/day, and up to 2000 IU of vitamin D/day is considered safe. Adults need at least 2000 IU of vitamin D/day to maintain a blood concentration of 25(OH)D between 30 and 100 ng/mL.

ACKNOWLEDGMENT This work was supported in part by a grant from the UV Foundation.

REFERENCES Aubin, J.E., Lian, J.B., Stein, G.S. 2006. Bone formation: Maturation and functional activities of osteoblast lineage cells. In: Primer on the Metabolic Bone Diseases and Disorders of Mineral Metabolism, 6th ed., Favus, M.J., ed. The American Society for Bone and Mineral Research, Washington, DC, 20–29. Balsan, S., Garabedian, M., Larchet, M., et al. 1986. Long-term nocturnal calcium infusions can cure rickets and promote normal mineralization in hereditary resistance to 1,25-dihydroxyvitamin D. J Clin Invest 77: 1661–1667. Brooks, M.H., Bell, N.H., Love, L., et al. 1978. Vitamin-D-dependent rickets type II: Resistance of target organs to 1,25-dihydroxyvitamin D. New Engl J Med 298: 996–999. Brown, E.M., Gamba, G., Riccardl, D., et al. 1993. Cloning and characterization of an extracellular Ca2+sensing receptor from bovine parathyroid. Nature 366: 575–580. Casella, S.J., Reiner, B.J., Chen, T.C., et al. 1994. A possible defect in 25-hydroxylation as a cause of rickets. J Pediatr 124: 929–932. Chen, H., Hewison, M. Hu, B., et al. 2003. Heterogeneous nuclear ribonucleoprotein (hnRNP) binding to hormone response elements: A cause of vitamin D resistance. PNAS (USA) 100: 6109–6114. Christakos, S., Dhawan, P., Liu, Y., et al. 2003. New insights into the mechanisms of vitamin D action. J Cell Biochem 88: 695–705. Fraser, D., Kooh, S.W., Kind, H.P., et al. 1973. Pathogenesis of hereditary vitamin-D-dependent rickets. An inborn error of vitamin D metabolism involving defective conversion of 25 hydroxyvitamin D to 1 alpha,25-dihydroxyvitamin D. New Engl J Med 289: 817–822. Haddad, J.G., Matsuoka, L.Y., Hollis, B.W., et al. 1993. Human plasma transport of vitamin D after its endogenous synthesis. J Clin Invest 91: 2552–2555. Heaney, R.P., Davies, K.M., Chen, T.C., et al. 2003. Human serum 25-hydroxycholecalciferol response to extended oral dosing with cholecalciferol. Am J Clin Nutr 77: 204–210. Hess, A.F. 1929. Rickets Including Osteomalacia and Tetany. Lea J. Febiger, Philadelphia, PA, 401–429. Hess, A.F., and Lundagen, M.A. 1922. A seasonal tide of blood phosphate in infants. J Am Med Assoc 79: 2210–2212. Hess, A.F., and Unger, L.J. 1921. The cure of infantile rickets by sunlight. J Am Med Assoc 77: 39–41.

Vitamin D and Bone

169

Hess, A.F., and Weinstock, M. 1924. Antirachitic properties imparted to inert fluids and to green vegetables by ultraviolet irradiation. J Biol Chem 62: 301–313. Holick, M.F. 2006. Resurrection of vitamin D deficiency and rickets. J Clin Invest 116: 2062–2072. Holick, M.F. 2007. Vitamin D deficiency. New Engl J Med 357: 266–281. Holick, M.F., Biancuzzo, R.M., Chen, T.C., et al. 2008. Vitamin D2 is as effective as vitamin D3 in maintaining circulating concentrations of 25-hydroxyvitamin D. J Clin Endocrinol Metab 93: 677–681. Holick, M.F., and Garabedian, M. 2006. Vitamin D: Photobiology, metabolism, mechanism of action, and clinical applications. In: Primer on the Metabolic Bone Diseases and Disorders of Mineral Metabolism, 6th ed., Favus, M.J., ed. American Society for Bone and Mineral Research: Washington, DC, 129–137. Hollis, B.W., and Wagner, C.L. 2004. Assessment of dietary vitamin D requirements during pregnancy and lactation. Am J Clin Nutr 79: 717–726. Huldschinsky, K. 1919. Heilung von Rachitis durch Kunstliche Hohensonne. Deutsche Med Wochenschr 45: 712–713. Huldschinsky, K. 1928. The Ultra-violet Light Treatment of Rickets. Alpine Press, NJ, 3–19. Jones, G. 2007. Expanding role for Vitamin D in chronic kidney disease: Importance of blood 25-OH-D levels and extra-renal 1α-hydroxylase in the classical and nonclassical actions of 1α,25-dihydroxyvitamin D3. Semin Dialysis 20: 316–324. Khosla, S. 2001. The OPG/RANKL/RANK system. Endocrinology 142: 5050–5055. Kitanaka, S., Takeyama, K.I., Murayama, A., et al. 1998. Inactivating mutations in the human 25-hydroxyvitamin D3 1α-hydroxylase gene in patients with pseudovitamin D-deficient rickets. New Eng J Med 338: 653–661. Kreiter, S.R., Schwartz, R.P., Kirkman, H.N., et al. 2000. Nutritional rickets in African American breast-fed infants. J Pediatr 137: 2–6. Malabanan, A., Veronikis, I.E., Holick, M.F. 1998. Redefining vitamin D insufficiency. Lancet 351: 805–806. Parfitt, A.M. 1998. Osteomalacia and related disorders. In: Metabolic Bone Disease and Clinically Related Disorders, 3rd ed., Avioli, L.V., and Krane, S.M., eds. Academic Press, San Diego, CA, 327–386. Pietras, S.M., Obayan, B.K., Cai, M.H., et al. 2009. Vitamin D2 treatment for vitamin D deficiency and insufficiency for up to 6 years. Arch Intern Med 169: 1806–1808. Powers, G.F., Park, E.A., Shipley, P.G., et al. 1921. The prevention of rickets in the rat by means of radiation with the mercury vapor quartz lamp. Proc Soc Exp Biol Med 19: 120–121. Rajakumar, K. 2003. Vitamin D, cod-liver oil, sunlight, and rickets: A historical perspective. Pediatrics 112: 132–135. Sniadecki, J. 1939. (1768–1838) On the cure of rickets. (1840) Cited by Mozolowski, W. Nature 143: 121–124. Steenbock, H., and Black, A. 1924. The reduction of growth-promoting and calcifying properties in a ration by exposure to ultraviolet light. J Biol Chem 61: 408–422. Wagner, C.L., and Greer, F.R., and the Section on Breastfeeeding and Committee on Nutrition. (2008). Prevention of rickets and Vitamin D deficiency in infants, children, and adolescents. Pediatrics 122:1142–1152.

11

Vitamin A and Bone Håkan Melhus

CONTENTS Introduction..................................................................................................................................... 171 Terminology.................................................................................................................................... 172 Sources of Vitamin A...................................................................................................................... 173 Recommended Dietary Intakes of Vitamin A................................................................................. 173 Skeletal Effects of Provitamin a Carotenoids................................................................................. 174 Skeletal Effects of Preformed Vitamin A and Retinoids in Animals.............................................. 175 Case Reports in Humans................................................................................................................. 176 Human Studies................................................................................................................................ 177 Population Studies...................................................................................................................... 177 Study Design.............................................................................................................................. 177 Dietary Assessment of Vitamin A.............................................................................................. 177 Assessment of Vitamin A Status................................................................................................ 180 Serum Retinol........................................................................................................................ 180 Serum Retinyl Esters............................................................................................................. 181 Assessment of Bone Health....................................................................................................... 181 Studies with Bone Mineral as Endpoint..................................................................................... 184 Studies with Fractures as Endpoint............................................................................................ 186 Summary......................................................................................................................................... 188 References....................................................................................................................................... 189

INTRODUCTION In 1998, we reported that intake of vitamin A at levels only slightly greater than the recommended daily intake was associated with reduced bone density and increased risk of hip fractures (Melhus et al., 1998). The study was prompted by consistent animal and in vitro data, reports on osteoporosis as a toxic effect of long-term therapy with synthetic retinoids, and the observation that Sweden and Norway, for unknown reasons, have the world’s highest incidence of osteoporotic fractures and that the vitamin A intake in these countries is unusually high due to a high consumption of fortified milk products, cod liver oil, and multivitamins (Melhus et al., 1998). The results of the observational studies published 1998–2002, together with the recognition that the large increase in vitamin A intake from fortified foods and supplements during the preceding decades had resulted in an oversupplementation especially among the elderly (Anderson, 2002), led to a reassessment of the levels of vitamin A supplementation and food fortification both in the United States and in Europe. Retinol levels have generally been reduced in multivitamins. In addition, in Sweden, AD-vitamins given to all children have been replaced by D-vitamin, and the retinol levels in low-fat dairy products have been reduced. In Norway, the retinol level in cod liver oil has been reduced by 75%. For ethical reasons, no randomized clinical trials evaluating the effects of vitamin A on bone have been undertaken, but since 1998, over 20 observational human studies have been published, 171

172

Diet, Nutrients, and Bone Health

and they are the focus of this chapter. Together, they can be said to illustrate the spectrum of difficulties in performing and interpreting the results of this kind of study. In this chapter, vitamin A, its structure, and food sources are described. Recommended intakes are given, and the effects of vitamin A on bone are summarized. Before describing the different observational studies, methods and difficulties in assessing vitamin A exposure and the outcome bone health are reviewed.

TERMINOLOGY The term vitamin A refers to a family of essential, fat-soluble dietary compounds that play an important role in vision, bone growth, reproduction, cell differentiation, and regulation of the immune system. The alcohol form of vitamin A, (all-trans-) retinol, and its fatty acid ester derivatives, retinyl esters (most commonly, retinyl palmitate), are referred to as preformed vitamin A (Figure 11.1). In published literature, the terms vitamin A, retinol, and preformed vitamin A are often used interchangeably. Whereas retinyl esters are the storage form of vitamin A, retinol functions as a prehormone, which is transported to target tissues where it is converted to the active forms of vitamin A (Figure 11.2). In the eye, the active metabolite is 11-cis retinal. In all other target organs, retinol is converted to all-trans-retinoic acid and possibly also to 9-cis retinoic acid. The term retinoids refers to retinol, its metabolites, and synthetic analogues that have a similar structure. Beta-carotene is a precursor for vitamin A (Figure 11.1). It is a tail-joined retinal dimer and is more efficiently converted to retinol than are other compounds of the class of plant pigments called carotenoids, owing to their relation to the carotenes. Beta-carotene and other carotenoids, such as β-carotene and β-cryptoxanthin, that the body can transform to active forms of vitamin A are referred to as provitamin A (McDowell, 2000).

Preformed vitamin A CH2OH

All-trans-retinol

O C15H31

Retinyl palmitate Provitamin A

Beta-carotene

FIGURE 11.1  Vitamin A overview.

173

Vitamin A and Bone

Storage Retinyl esters

Liver

Retinyl esters

Retinol–RBP4 Blood

STRA6 ? All trans-retinoic acid

9-cis retinoic acid

CYP26 Metabolites Provitamin A carotenoids Plants

Preformed vitamin A

Retinoic acid Retinoid X receptor receptor Nucleus

Animal tissue Intestine

Target organs, excluding the eye

FIGURE 11.2  Vitamin A overview. Vitamin A is obtained as provitamin A carotenoids in plant foods or as preformed vitamin A in the form of retinyl esters from animal sources. In the intestine, provitamin A carotenoids are cleaved to retinal and converted to retinol. Retinyl esters are hydrolyzed to retinol. Retinol is then reesterified and incorporated to chylomicrons, which circulate in the intestinal lymph before entering the general circulation. Chylomicron remnants are cleared by the liver, where the retinyl esters are stored. After hydrolysis of retinyl esters to retinol, retinol binds to retinol-binding protein (RBP) 4 and the vitamin is transported from the liver to target organs. Retinol is taken up by a specific RBP4 receptor, STRA6, and converted to all-trans retinoic acid (ATRA), and possibly also 9-cis RA. In the eye, retinol is converted to 11-cis retinal which binds to opsin. In other target organs, retinol is oxidized to ATRA, which binds to its specific nuclear RA receptor. The enzyme CYP26 converts ATRA to polar metabolites, which can be conjugated and eliminated.

SOURCES OF VITAMIN A Preformed vitamin A is found in foods of animal origin, such as liver, meats, milk products, eggs, and fatty fish. Because retinyl esters are stored in the liver of both fish and mammals, cod liver oil and other liver products contain very high levels. Common sources are also fortified foods (such as margarine and breakfast cereals) and dietary supplements. Many orange fruits and green vegetables are sources of provitamin A carotenoids, for example, carrots, spinach, sweet potatoes, cantaloupes, broccoli, squash, mango, apricots, peas, and papaya (Sp*rn et al., 1994; McDowell,2000).

RECOMMENDED DIETARY INTAKES OF VITAMIN A In the United States, recommendations for vitamin A are provided in the Dietary Reference Intakes (DRI) developed by the Institute of Medicine (IOM) (Food and Nutrition Board, IOM, 2001). Three important types of reference values included in the DRIs are Recommended Dietary Allowances (RDA), Adequate Intakes (AI), and Tolerable Upper Intake Levels (UL). The RDA recommends

174

Diet, Nutrients, and Bone Health

the average daily dietary intake level that is sufficient to meet the nutrient requirements of nearly all (97%–98%) healthy individuals in each age and gender group (Table 11.1A). An AI is set when there are insufficient scientific data to establish an RDA. AIs meet or exceed the amount needed to maintain nutritional adequacy in nearly all people (Table 11.1B). The UL is the highest level of daily vitamin A intake that is likely to pose no risk of adverse health effects in almost all individuals (Table 11.2) (Food and Nutrition Board, IOM, 2001).

SKELETAL EFFECTS OF PROVITAMIN A CAROTENOIDS There is no evidence of bone toxicity from provitamin A carotenoids in laboratory animals or humans (Armstrong et al., 1994; Sahni et al., 2009a, 2009b). Beta-carotene produces no bone TABLE 11.1A Recommended Dietary Allowances (RDAs) for Vitamin A Children

1–3 years 4–8 years

  300   400

Males

9–13 years 14 years and older

  600   900

Females

9–13 years 14 years and older

  600   700

Pregnancy

14–18 years 19 years and older

  750   770

Lactation

14–18 years 19 years and older

1200 1300

TABLE 11.1B Adequate Intakes (AIs) of Vitamin A for Infants (Males and Females) 0–6 months 7–12 months

400 500

Note: RDAs and AIs are listed as micrograms of retinol activity equivalents/day. 1 RAE = 3.3 IU. Source: Food and Nutrition Board, Institute of Medicine. 2001. Dietary Reference Intakes for Vitamin A, Vitamin K, Arsenic, Boron, Chromium, Copper, Iodine, Iron, Manganese, Molybdenum, Nickel, Silicon, Vanadium, and Zinc. National Academy Press, Washington, DC.

175

Vitamin A and Bone

resorption or other significant change in bone in vitro (Kamm et al., 1984). Moreover, these compounds do not cause skeletal malformations (Armstrong et al., 1994). Important reasons are that the cleavage of provitamin A carotenoids to retinal in the intestine is a highly regulated step (Bachmann et al., 2002; Takitani et al., 2006) and that an excess of carotenoids can be stored in the skin, resulting in a yellow to yellow-orange discoloration, or carotenodermia.

SKELETAL EFFECTS OF PREFORMED VITAMIN A AND RETINOIDS IN ANIMALS Evidence from animal studies has clearly shown that high doses of retinol can have adverse skeletal effects (Binkley and Krueger, 2000). This research on vitamin A toxicity was primarily carried out between 1925 and 1950. Nieman and Obbink (1954) provided an extensive review of these studies. Skeletal lesions were reported in 31 different studies, and laboratory animals included rat, mouse, guinea pig, rabbit, chicken, duck, and dog. The most prominent features of toxic doses of retinol were a thinning of the long bones and spontaneous fractures (Table 11.3). These findings have been confirmed in more recent studies (Frankel et al., 1986; Hough et al., 1988; Wu et al., 1996). The active form of vitamin A, retinoic acid, and synthetic retinoids induce the same characteristic bone thinning (Teelmann, 1981; Kneissel et al., 2005). The available data suggest that these adverse

TABLE 11.2 Tolerable Upper Intake Levels (ULs) for Preformed Vitamin A Age (Years) 0–1 1–3 4–8 9–13 14–18 Adults

ULs for Preformed Vitamin A (μg/day)   600   600   900 1700 2800 3000

Source: Food and Nutrition Board, Institute of Medicine. 2001. Dietary Reference Intakes for Vitamin A, Vitamin K, Arsenic, Boron, Chromium, Copper, Iodine, Iron, Manganese, Molybdenum, Nickel, Silicon, Vanadium, and Zinc. National Academy Press, Washington, DC.

TABLE 11.3 Major Findings Related to Bone in Animals with Hypervitaminosis A Spontaneous fractures Thinning of long bones Premature closure of epiphyses Retardation of growth No or little effect on mineralization, except when vitamin D is low. Toxicity is cumulative and reversible.

176

Diet, Nutrients, and Bone Health

effects occur as a result of increased bone resorption and decreased bone formation. It has been pointed out that most studies on laboratory animals have used very high doses administered to young growing animals and that results from such studies may not be directly relevant to human osteoporosis (Binkley and Krueger, 2000). However, subclinical hypervitaminosis A in mature rats leads to reduced bone diameter (Johansson et al., 2002), and the same changes, although smaller in magnitude, also occur in aged rats (Kneissel et al., 2005). There are several important observations in animal studies that may help us to interpret human studies. First, retinol has no or little effect on mineralization, except when vitamin D is present at a marginal level (Aburto and Britton, 1998a, 1998b). When there is sufficient vitamin D, retinol increases the risk of fracture by a mechanism that reduces the diameter of the bone but not the bone mineral density (BMD) (Lind et al., 2006). This is in contrast to vitamin D. If this is true also in humans, BMD measurements (which are used to diagnose osteoporosis) will not detect the adverse effects caused by an excessive intake of retinol. Second, vitamin A toxicity is cumulative. The duration of onset and frequency of fractures are dose related. This has been most clearly shown for the retinoid etretinate (Teelmann, 1981). Fractures occurred as early as 10–14 days at doses of 10–15 mg/kg/day. In contrast, with 3 mg/kg/day, fractures were not observed earlier than after 5–6 months. The fractures were preceded by a markedly intensified remodeling of the bones with deformations, increased periosteal osteoclastic activity, thickening of the periosteum, and accelerated ossification of the epiphyseal line. In addition, an augmented endosteal osteoblastic activity resulting in marked reduction of bone diameter and increasing porosity, particularly of the long bones, was noted. All side effects were reversible after cessation of treatment (Teelmann, 1981; Melhus, 2003).

CASE REPORTS IN HUMANS Case reports of children and adults who have ingested massive doses of vitamin A describe most of the findings that have been found in animals (Table 11.4) (Kamm et al., 1984; Armstrong et al., 1994). However, only one case of fracture due to hypervitaminosis A has been reported in man (Ruby and Mital, 1974). A girl 1 year and 3 months old had been given a daily dose of 15 mg in a water-soluble preparation for 7 months. Roentgenograms showed a fracture of the right humerus. Only remnants of the distal femoral epiphyses were visible, and both were seemingly impressed into the metaphyses, which were irregular. The serum vitamin A was elevated by a factor of five. Vitamin A has an effect on bone turnover also in humans (Jowsey and Riggs, 1968). The microradiographic appearance of bone taken during and after a period of high vitamin A intake revealed that there was an increase in the size of the osteocyte lacunae, resorption surfaces were a factor of six above the normal, and there was an absence of a comparative increase in bone formation. Similar findings with numerous resorptive lacunae and no index of bone formation have been described in a French patient (Baglin et al., 1986). The lowest dose inducing chronic hypervitaminosis A has not been defined. In some cases, doses as low as about 0.1–0.3 mg/kg/day (Schurr et al., 1983; Carpenter et al., 1987) have been suggested to cause toxicity in children.

TABLE 11.4 Major Findings Related to Bone in Human Subjects with Hypervitaminosis A Bone and joint pain Thinning of bone cortex Premature closure of epiphyses Retardation of growth Hyperostosis Calcification of tendons and ligaments

Vitamin A and Bone

177

As described above, preclinical studies demonstrated that synthetic retinoids have effects on bones that mimic the effects of hypervitaminosis A. When these retinoids were introduced in clinical dermatology in the late 1970s, relatively high doses were used, and premature closure of epiphyses (Milstone et al., 1982), hyperostosis (Pittsley and Yoder, 1983), calcification of ligaments and tendons (DiGiovanna et al., 1986), and thinning of long bones and traumatic fractures (Prendiville et al., 1986) were all reported within a few years.

HUMAN STUDIES Results from observational studies of the effect of retinol on bone health are inconsistent. This may be due, in part, to differences in population studied, study designs, assessment of retinol intake or vitamin A status, and method used to measure bone health. Some aspects of these will therefore be discussed.

Population Studies About half of all studies on vitamin A and bone health have been conducted in the United States and one third in the Nordic countries. The reason for this is that the highest incidence of osteoporotic fracture is found in these countries, especially Norway and Sweden (Johnell et al., 1992; EPOS Group, 2002; Kanis et al., 2002). The dietary intake of retinol is higher in Nordic countries than that in southern Europe and the United States (Cruz et al., 1991; Scientific Advisory Committee on Nutrition, 2005). One important reason for this is that the milk consumption is higher in Nordic countries (International Dairy and Federation, 1995). This is possibly due to lactose tolerance (Sahi, 1994), which is a genetic adaptation to vitamin D deficiency; at northern latitudes, the cutaneous synthesis of previtamin D during the long winter season is undetectable (Burgaz et al., 2007). Historically, the intake of dairy products, fatty fish, cod liver oil, and AD-vitamins has therefore been important in these countries. Low-fat milk products are not fortified with vitamin A in Sweden anymore, so the United States and Canada are now the only countries in which these products are fortified with AD-vitamins.

Study Design The majority of the studies on bone mineral are cross-sectional, whereas prospective studies dominate when fracture is the endpoint. For reasons discussed below, a cohort study with multiple assessments of retinol intake or vitamin A status, a good follow-up, and fracture as endpoint is important to obtain reliable results. Case–control studies have a lower quality grade on a standard scale of medical evidence (Phillips et al., 2001), and it is interesting to note that the results from these (Table 11.5) often show that vitamin A has positive effects on BMD. A problem with case–control studies is that of confounding. Cases with osteoporosis are more likely to be older, have lower BMI, and have poorer health compared with those of controls. If serum retinol is found to be lower in cases than in controls, it is impossible to separate whether this is the cause or a consequence of other factors. Differences in age and BMI can be adjusted for in the statistical analysis, but to control for differences in general health will require very detailed information about comorbidities and intricate statistical modeling.

Dietary Assessment of Vitamin A Multiple methods have been used to ascertain retinol intake, with the Food Frequency Questionnaire (FFQ) being the instrument of choice in large nutritional epidemiological studies. Two other dietary instruments that have been commonly used are the 24-h dietary recall and the dietary record (DR). It is generally accepted that all these methods have advantages and limitations, and none of them is

Country

United States

United States

United States

United States

United States

United States

United Kingdom

Sweden

United States

Iceland

United States

Year

1985

1985

1986

1990

1993

1995

1997

1998

2001

2001

2002

Four-year longitudinal

Cross-sectional

Cross-sectional

Cross-sectional

Cross-sectional

Physical exercise trial

Cross-sectional

Cross-sectional

Four-year calcium trial

Cross-sectional

Cross-sectional

Study Design

FFQ Three 4-day records Three-day record Four 1-week records Not assessed

246 ♀, 55–80 years 281 ♀, 50–60 years 66 ♀, 28–39 years 426 ♀, 45–59 years 175 ♀, 28–74 years 2888 ♀, 2902 ♂, 20–80+ years 408 ♀, 70 years 570 ♀, 388 ♂, 55–92 years

Multiple 24-hour records Not assessed

99 ♀, 35–65 years

FFQ

FFQ

24-hour recall + previous week 24-hour recall

Dietary Assessment

912 ♀, 1208 ♂ 43–81 years 324 ♀, 55–80 years

Study Sample

TABLE 11.5 Studies Examining the Association between Vitamin A Intake and Bone Mineral

DXA

DXA

DXA

DXA

DXA

DXA

SPA

SPA

SPA

SPA

SPA

Bone Measurement

Retinol

Serum retinyl esters Retinol

Total vitamin A Total vitamin A Retinol

Retinol

Total vitamin A Total vitamin A Total vitamin A Serum retinol

Nutrient/ Biomarker

Inverse U-shaped

0 (+)

0 (+)

0 (+ and −)

0 (+)

Association

Sowers et al., 1985 Freudenheim et al., 1986 Sowers and Wallace, 1990 Hernandez-Avila et al., 1993 Houtkooper et al., 1995 Earnshaw et al., 1997 Melhus et al., 1998 Ballew et al., 2001a, 2001b Sigurdsson et al., 2001 Promislow et al., 2002

Yano et al., 1985

Reference

178 Diet, Nutrients, and Bone Health

England

Canada

Scotland

Denmark

United States

United Kingdom United States

Sweden

Norway

2003

2003

2004

2004

2005

2005

2008

2008

Retrospective

Longitudinal

Nested case–control Case–Control

Cross-sectional

Cross-sectional + longitudinal

Longitudinal

Cross-sectional

Longitudinal

Case–Control

Not assessed Questionnaire

3052 ♀, 50–70y

Three-day food records

Not assessed

Seven-day records at baseline Retrospective questionnaires FFQ × 2 (5–7y apart) Four-day or 7-day food records (5y apart) FFQ

Not assessed

312 cases, 934 controls from 2606 ♀, >75y 27 ♀ with osteoporosis, 24 ♀ without osteoporosis, 48–83y 78 ♂, 21–25y at baseline

11,068 ♀, 50–79y

2016 ♀, median 50y

891 ♀, 45–55y

58 ♀, 45–75y

♀ ≥ 60y 75 with osteoporosis, 75 controls 470 ♂, 474 ♀, 67–79y

Childhood cod liver oil intake

Serum retinol

DXA × 2 (2y apart) SXA

DXA

DXA

Vitamin A, retinol, betacarotene Serum retinol, RE, betacarotene Vitamin A

Retinol, beta-carotene

Vitamin A/Retinol

DXA

DXA × 2 (5–7y apart) DXA × 2 (5y apart)

Vitamin A, beta-carotene

DXA × 2 (2–5y apart) DXA Retinol

Plasma retinol

DXA

Hogstrom et al., 2008 Forsmo et al., 2008

Barker et al., 2005 Penniston et al., 2006

+ 0

Wolf et al., 2005

Kaptoge et al., 2003 Suzuki et al., 2003 Macdonald et al., 2004 Rejnmark et al., 2004

Maggio et al., 2003

0 (−)

0 (−)

+

Notes: SPA = singe-photon absorptiometry, FFQ = Food Frequency Questionnaire, DXA = dual-energy X-ray absorptiometry. 0 = no significant association, + = positive association, − = negative association, (+) = weak positive or partly positive association (e.g., in subgroups), (−) = weak negative or partly negative association, DXA = dual-energy X-ray absorptiometry, SXA = single energy X-ray absorptiometry, FFQ = Food Frequency Questionnaire, RE = retinyl esters, y = years.

2006

Italy

2003

Vitamin A and Bone 179

180

Diet, Nutrients, and Bone Health

entirely satisfactory (Henriquez-Sanchez et al., 2009). Because retinol is found in very high concentrations in a limited number of foods (liver, liver products, fortified foods, and supplements), some of which may be consumed infrequently, it is difficult to obtain reliable estimates of average consumption (especially with the 24-hour recall method) and to identify consumers of high levels of retinol (Scientific Advisory Committee on Nutrition, 2005). The intraindividual variation can cause random error when classifying persons by intake of retinol and produce a bias that systematically underestimates the strength of associations to bone health. It has been estimated that three to nine independent measurements may be required to distinguish reliably even large differences (Tangney et al., 1987). Osteoporosis develops over many years, and repeated measurements are also required to identify changes in diet over time. It is important to recognize that the majority of published studies are based on one single dietary assessment and find no association with bone health. In fact, there are only four studies with three or more independent measurements of retinol intake (Tables 11.5 and 11.6). One study used multiple 24-hour records; one study, three 4-day records; one study, four 7-day records; and the best designed study performed five independent measurements of retinol intake by FFQs over 18 years. In this latter study with five FFQs, the association with hip fracture was not significant when only the baseline measurement was used for analyses (Feskanich et al., 2002), emphasizing the importance of repeated measurements. Another problem in correctly assessing the vitamin A intake is the bioconversion of carotenoids. For more than 30 years, it was generally accepted that 6 µg of beta-carotene and 12 µg of other provitamin A carotenoids had the same vitamin activity as 1 µg of retinol (FAO/WHO, 1967). However, several studies published in the 1990s showed that the bioefficacy of beta-carotene in plant foods was considerably less. Therefore, the U.S. Food and Nutrition Board, IOM (2001) introduced the term retinol activity equivalent (RAE), where 1 µg RAE = 1 µg retinol = 12 µg beta-carotene = 24 µg other provitamin A carotenoids. The development of techniques using isotope-labeled betacarotene and retinol has advanced our knowledge further and shown that this bioconversion is quite variable; 1 mol beta-carotene can provide anything between 0 and 0.27 mol retinol (Lin et al., 2000; van Lieshout et al., 2001; Hickenbottom et al., 2002). The activity of the enzyme responsible for the conversion of beta-carotene to retinal in the intestine, the β,β-carotene 15,15′-monooxygenase (βCMOOX), is upregulated in vitamin A deficiency and downregulated by an increased intake of retinyl esters or beta-carotene (van Vliet et al., 1996). It has been shown that retinoic acid reduces, and a retinoic acid receptor alpha antagonist significantly increases the intestinal enzyme activity. Moreover, a retinoic-acid-responsive element has been identified in the promoter of the µCMOOX gene. These results are consistent with a transcriptional feedback regulation of the enzyme by RA via the specific nuclear receptors (Bachmann et al., 2002; Takitani et al., 2006). Thus, the bioconversion of beta-carotene seems to be determined by the body’s needs for the vitamin. This inefficiency of bioconversion explains why provitamin A carotenoids do not cause hypervitaminosis A even when ingested in large amounts.

Assessment of Vitamin A Status Serum Retinol Serum retinol levels reflect the liver stores of vitamin A, but not in a linear fashion. The levels are homeostatically controlled (usually around 2 µmol/L) over the physiologic range of liver vitamin A concentrations and are therefore generally unaffected by normal retinol intakes (Willett et al., 1983; Krasinski et al., 1989). When liver stores are exhausted (<20 µg/g liver), serum retinol values tend to fall, and when the capacity for storage in the liver is exceeded (>300 µg/g liver) or the rate of intake is greater than the rate it can be removed by the liver, serum values tend to increase. Thus, except in cases of deficiency or excess, retinol levels in the serum are not good indicators of vitamin A status (Olson, 1984). This may, in part, explain why serum retinol does not correlate with normal dietary

Vitamin A and Bone

181

intakes (Garry et al., 1987; Roidt et al., 1988; Booth et al., 1997), but it does with supplement use (Garry et al., 1987; Roidt et al., 1988; Sowers and Wallace, 1990; Opotowsky and Bilezikian, 2004; Barker et al., 2005). An advantage of serum retinol as a measure of retinol exposure is that it tends to be stable for many months. In one study, serum retinol measurements taken 4 months apart correlated highly (r= 0.79), while estimates of vitamin A intake among the same group correlated less well (r = 0.47) (Kardinaal et al., 1995). A number of other factors may also influence the levels. Serum retinol has, for example, been positively associated with age, weight, serum lipids, socioeconomic status, and renal failure and negatively associated with smoking, alcohol consumption, infections, and chronic liver diseases. With the exception of serum lipids and infections, all the other confounding factors also influence the risk of fracture (Michaelsson et al., 2003) and need to be considered in the statistical analysis. Serum Retinyl Esters Retinyl esters from animal food sources are transported from the intestine via the lymph to the liver, where they are stored. In contrast to retinol, which is homeostatically controlled, the serum levels of retinyl esters will increase markedly after each vitamin-A-rich meal. In fasting serum, most of the circulating vitamin A is found in the form of retinol, and only a small proportion is retinyl esters. This may explain why total intake of vitamin has not been found to correlate with serum retinyl esters (Scientific Advisory Committee on Nutrition, 2005). By contrast, in some cases of vitamin A intoxication, the retinyl ester to retinol ratio was found to exceed 10% (Smith and Goodman, 1976), and long-term use of retinol supplements has been associated with higher serum retinyl esters in elderly subjects (Krasinski et al., 1989). Serum retinyl esters have therefore been used as a marker of retinol toxicity in two studies on BMD (Table 11.5) and two on fractures (Table 11.6). However, serum retinyl esters may more reflect a temporary excess in vitamin A intake rather than long-term vitamin A intake and storage. In patients with vitamin A toxicity, serum retinyl esters decrease faster than serum retinol after discontinuation of vitamin A supplements (Smith and Goodman, 1976). Studies of plasma kinetics have shown that the clearance of serum retinyl esters varies substantially from one person to another (Bitzen et al., 1994; Reinersdorff et al., 1996; Johansson and Melhus, 2001), with an average increase in clearance of more than 50% over a 12-hour period after an intake of vitamin A of 1.0 to 1.5 mg (Krasinski et al., 1990).

Assessment of Bone Health In studies of factors influencing bone health, the most clearly defined outcome of interest is bone fracture following minimal trauma. Most studies, however, use the intermediate outcome measure of BMD (Scientific Advisory Committee on Nutrition, 2005). Techniques to measure BMD, such as single-photon absorptiometry (SPA) and dual-energy X-ray absorptiometry (DXA), determine the amount of mineral per unit area and not volume. This means that BMD measurements are influenced by the size, shape, and orientation of the bone and therefore are more useful in prospective studies, where changes can be assessed over time. There are six such prospective studies (Table 11.5). If retinol reduces the size rather than the BMD in humans in the same way as in animals, peripheral quantitative computer tomography and other methods that can provide a measure of volumetric density may be necessary to more accurately assess the adverse skeletal effects of retinol. The most serious and important consequence of osteoporosis is hip fractures. In studies that use fracture as outcome, hip fractures are therefore of special relevance. They have generally been identified via hospital discharge records or are self-reported. Some studies have confirmed medical record data with radiographic records or home visits. A problem with self-reported fractures is selection bias. Fracture patients have a substantially increased risk of death that persists several years post fracture (Bliuc et al., 2009). This risk of death is most pronounced for hip fractures,

Country

United States

Sweden

Sweden

United States

Sweden

United States

Year

1990

1998

1998

2002

2003

2004

Prospective, 9.5y follow-up (IWHS)

Prospective, 30y follow-up (ULSAM)

Prospective, 2.4y follow-up (MDCS) Prospective, 18y follow-up (NHS)

Nested case– control, 2–64 months follow-up (SMC)

Retrospective

Study Design

34,703 ♂ + ♀, 6502 any fx, 525 hip, 55–69y

FFQ 127 items + suppl question

FFQs every 4y, mean intake determined from 5 FFQs Serum retinol, serum beta-carotene, 7-day food record in 1138 men 20y after entry

72,337 ♀, 603 hip fx, 34–77y

Cohort 2322 ♂, 266 any fx, 84 hip fx, 49–51y

Seven-day menu book, + FFQ

Serum retinol, 24-h food recall, interview for suppl FFQ of usual intake of 60 foods during previous 6 months

246 ♀, 56 cases (31 fx in the last 10y) 55–80y 66,651 ♀, 247 hip fx cases, 873 controls, 40–76y

6576♂, 160 any fx cases, 51 fragility fx, 46–68y

Vitamin A Assessment

Study Sample

TABLE 11.6 Studies Examining the Association between Vitamin A and Fracture Risk

Hospital discharge, orthopedic + radiographic records Self-reported

Local hospital registry + X-ray examinations Self-reported by questionnaire

Hospital discharge records

Self-reported by interview

Identification of Fractures

Tot vit A 4.33 mg RE, retinol food + suppl 1.16 mg

Total vit A food + suppl 0.9 mg RE Retinol from food: cases 0.96 mg, controls 0.8 mg RE Cases, 1.96 mg; controls, 1.82mg Retinol from food + suppl, 1.2 mg Retinol from food 1.1 mg

Mean Intake/ Level

Results

Rate ratio =2.5 for hip fx; >2.64 µM (5th) vs. 2.17–2.36 µM (3rd quintile) No association (RR hip fx for suppl users 1.18 (0.99–1.41)

RR = 1.89, ≥2 mg vs. <0.5 mg

No association

OR = 2.05, ≥1.5 mg vs. <0.5 mg

No association

Reference

Lim et al., 2004

Michaelsson et al., 2003

Feskanich etal., 2002

Elmstahl et al., 1998

Sowers and Wallace, 1990 Melhus et al., 1998

182 Diet, Nutrients, and Bone Health

United States

United Kingdom

United States

United States

2004

2005

2006

2009

Prospective, 6.6y follow-up (WHI)

Prospective, 17y follow-up (LWCS)

Nested case– control, 3.7y follow-up

Nested case– control, 5y follow-up (DOPS) Prospective, 22y follow-up (NHANES I)

Self-reported + hospital discharge records Questionnaire, medical records

Questionnaire about suppl

FFQ × 2

8877 ♀, 5101 ♂, mean 73.7 + 74.9y, 1227 hip fx, 445 wrist fx, 729 spine fx 75,747 ♀, 10,405 total fx, 588 hip fx, mean age 63y

Medical records, home visits

Self-reported, confirmed by hospital records

Hospital discharge records

Serum retinol, retinyl esters, beta-carotene

Serum vit A (retinol + retinyl esters)

Four- or 7-day record

2606 ♀, 312 any fx cases, 92 hip fx, 934 controls, >75y

Cohort 2016 ♀, 163 fx cases (extremities or spine), 978 controls, 45–58y 2799 ♂, 172 hip fx, 50–74y

1.95 µM any fx, 2.00 µM controls. 1.91 µM hip fx, 1.94 µM controls 68% of ♀ and 61% ♂, took vit A containing suppl Retinol from food + suppl, 0.98 mg

Serum retinol 2.0 µM

Retinol food 0.53 mg, food + suppl 1.21 mg

No association, RR = 1.15 in low vit D + high vit A

♀ suppl users wrist fx HR = 1.15 hip 1.07(1.00–1.15)

Opotowsky and Bilezikian, 2004

HR = 2.1 for ≥2.56 µM (5th) and 1.9 for ≤1.61 µΜ (1st) vs. 1.90–2.13 µΜ (3rd quintile) No association, supplement use HR = 0.76

Caire-Juvera et al., 2009

White et al., 2006

Barker et al., 2005

Rejnmark etal., 2004

No association

Notes: RE = retinol equivalents, fx = fracture(s), y = years, vit = vitamin, suppl = supplement, FFQ = Food Frequency Questionnaire, OR = odds ratio, RR = relative risk, HR = hazard ratio, SMC = Swedish Mammography Cohort, MDCS = Malmö Diet and Cancer Study, NHS = Nurses’ Health Study, ULSAM = Uppsala Longitudinal Study of Adult Men, IWHS = Iowa Women’s Health Study, DOPS = Danish Osteoporosis Prevention Study, NHANES = National Health and Nutrition Examination Survey, LWCS = Leisure World Cohort in Southern California, WHI = Women’s Health Initiative.

Denmark

2004

Vitamin A and Bone 183

184

Diet, Nutrients, and Bone Health

which have the highest mortality risks in the first 6 months post fracture (Farahmand etal.,2005).

Studies with Bone Mineral as Endpoint Before 1990, no human studies were reported that explicitly examined the relationship between vitamin A and bone health. Three studies in the 1980s had included vitamin A in investigations of diet, especially calcium, and bone health (Table 11.5). The association between diet and bone mineral content (BMC) was investigated in a population of elderly Japanese residents (1368 men and 1098 women) living in Hawaii (Yano et al., 1985). Dietary intake was obtained by the 24-hour recall method, and BMC was measured by SPA. In the analysis of data, all subjects who reported atypical dietary intakes during the previous 24 h were excluded, leaving 1208 men and 912 women for the analyses. If all subjects who consumed liver in the previous 24 h were excluded is unclear. Participants were also asked to estimate the frequency of eating selected food containing relatively large amounts of calcium during the previous week, and they received a mailed questionnaire in which they were asked to list the brand names and amounts of supplementary minerals and vitamins consumed in the past week. There were no correlations between vitamin A intake and BMC in men. Of the five examined sites, dietary plus supplementary vitamin A was very modestly associated with BMC at the distal radius and ulna in women. Covariables included age, weight, height, history of nonviolent fracture, current use of estrogen (women), and current use of thiazides. No difference in BMC was found between users and nonusers of supplements. With similar methods and only one single 24-h recall, it was found that “vitamin A intake was negatively correlated with bone density values at a level which approached significance” in 324 postmenopausal women in Iowa (Sowers et al., 1985). As part of a 4-year clinical trial of 99 women randomly assigned to placebo or calcium supplements, the effect of energy and 14 nutrients on BMC of the radius, humerus, and ulna was examined (Freudenheim et al., 1986). Up to seventy-two 24-h recall records were collected for each woman in this study from Wisconsin. A positive association between vitamin A and rate of humeral bone loss was found among the nine premenopausal women in the placebo group. In contrast, in the postmenopausal women of the calcium supplemented group, there was an inverse correlation between vitamin A intake and the rate of change in ulna BMC. In a single patient receiving a high supplemental dose (average intake 14,624 IU/day), bone loss was very rapid with no other reason apparent. Sowers and Wallace (1990) extended their initial study by evaluating vitamin A intake, serum retinol concentrations, and radial bone mass in 246 postmenopausal women. The correlation between serum retinol and dietary food intake estimated from 24-hour food recall was only 0.02, but serum retinol levels were higher among vitamin A supplement users (1.69 ± 0.52 µmol/L) compared with those among nonusers (1.55 ± 0.52 µmol/L, p < .03). After controlling for age, current estrogen replacement, and current thiazide use, they observed no association between vitamin A supplement use or serum retinol with radial bone mass (Sowers and Wallace, 1990). In a study from Massachusetts (Hernandez-Avila et al., 1993), the influence of dietary, anthropomorphic, and hormonal factors on bone density in a cross-sectional sample of 281 premenopausal and perimenopausal women was investigated. Dietary information was obtained using an FFQ (Willett et al., 1987). Bone density was measured using SPA in the midshaft and the ultradistal radius. The investigators observed no associations between dietary variables and midshaft bone density, but retinol, vitamin D, and vitamin C were all positively but weakly associated with ultradistal radius density. Women who used multivitamins had higher values than the values of those who did not. It was not possible to distinguish which nutrients were responsible for this association. Houtkooper et al. (1995) assessed the relationship between total energy intake, nutrient intake, body composition, exercise group status, and annual rates of change in BMD measured by DXA in

Vitamin A and Bone

185

66 premenopausal women living in Arizona and taking calcium supplements. Neither calcium nor other nutrients were significant variables in regression models predicting BMD slopes at the spine or at any femur site, but vitamin A or carotene was associated with slowing the annual rate of total body bone loss. Intakes from supplements were not included in the analyses. In the British Early Postmenopausal Intervention Cohort (EPIC) study (Earnshaw et al., 1997), BMD was measured with DXA. Dietary assessment was performed with a 3-day unweighed DR. No correlation between any dietary variable, including calcium, and BMD was found in 426 women. In the first study that distinguished between retinol and beta-carotene (Melhus et al., 1998), the association between dietary retinol or beta-carotene and BMD in 175 Swedish women was investigated. Diet was assessed from four 1-week DRs. In a multivariable analysis, retinol intake was negatively associated with BMD. When subjects with intakes >1.5 mg/day were compared with those with intakes <0.5 mg/day, BMD was significantly reduced at all five examined sites. Dietary intake of beta-carotene was not associated with BMD. In a cross-sectional analysis of the Third National Health and Nutrition Examination Survey (NHANES III) (Ballew et al., 2001a), the association between fasting serum retinyl esters and BMD was analyzed in 5790 adults. Although about one third of the participants had fasting serum retinyl esters ≥10% of total serum vitamin A and one quarter or more had osteopenia/ osteoporosis at one or more sites, the study showed no significant association between fasting serum retinyl ester concentration, or fasting serum retinyl esters as percent of total vitamin A, and BMD at any site. In a book chapter, a study of 70-year-old Icelandic women (n = 232) was published (Sigurdsson et al., 2001). Dietary intake was assessed using an FFQ, with 130 food items reflecting intake from the past 3 months. More than half of the retinol was obtained from cod liver oil and multivitamins. No association of retinol intake and BMD was observed. In the Rancho Bernardo Study (Promislow et al., 2002), the association between BMD and bone loss, and total and supplemental retinol intake was studied in a cohort of 570 elderly women and 388 men who were followed for 4 years. Dietary intake of vitamin A was assessed by a dietary questionnaire at baseline. Half of the women and more than a third of the men were taking retinol-containing supplements. After stratification by supplement use, regression analyses showed an inverse U-shaped association of retinol intake with baseline BMD and BMD change. In a case–control study of Italian women (Maggio et al., 2003), 75 subjects with osteoporosis (BMD T-score ≤ −3.5 at the femoral neck) and 75 controls were examined. Plasma retinol was found to be decreased in osteoporotic women (2.14 vs. 2.37 µM). In a subsequent analysis of 45 of these women with osteoporosis and 45 of the controls, plasma retinol, but not carotenoids, was positively correlated with femoral neck BMD (Maggio et al., 2006). In a study from England (Kaptoge et al., 2003), 470 men and 474 women that were recruited from a diet and cancer study (EPIC-Norfolk) were studied. Seven-day food diaries were used to calculate dietary intake of some 31 nutrients and 22 food groups. BMD loss was measured using DXA on two occasions about 3 years apart. In men, there was no evidence of an effect of any of the nutrients evaluated. In women, only low intake of vitamin C was associated with faster rate of BMD loss. In a small study from Canada (Suzuki et al., 2003), 58 postmenopausal women not taking estrogen were examined. Dietary intakes were assessed by retrospective questionnaires. Increased retinol intake from supplements had no adverse effect on BMD. In a longitudinal study of mainly premenopausal women (45–55 years at baseline and 50–60 years at follow-up 5–7 years later) in the Aberdeen Prospective Osteoporosis Screening Study (Macdonald et al., 2004), it was found that greater femoral neck BMD loss was associated with increased dietary intake of retinol, but this negative association was no longer significant when retinol from supplements was included.

186

Diet, Nutrients, and Bone Health

The relations between vitamin A intake and BMD and fracture risk were investigated in the Danish Osteoporosis Prevention Study—a prospective study on the effect of hormone replacement therapy (Rejnmark et al., 2004). Cross-sectional analysis at baseline and after 5 years showed no association between intake of retinol and BMD at any site. In a cross-sectional analysis of 11,068 women, who were participants in the Women’s Health Initiative (WHI) (Wolf et al., 2005), a negative association between total beta-carotene intake and femoral neck BMD was unexpectedly found. The finding was not consistent in that it was not significant for dietary intake alone, serum beta-carotene, or for other BMD sites. Retinol and other antioxidants were unrelated to BMD. In a case–control study of 312 incident osteoporotic fracture cases (92 hip fractures) and 934 controls, nested in a prospective study of 2606 British women over the age of 75 years (Barker et al., 2005), a tendency for increased serum retinol to predict benefit rather than harm in terms of BMD (r = 0.09, p = .002) was noted. Comorbidities or socioeconomic variables were not adjusted for in the analyses of this study. In a case–control study of 30 postmenopausal women with osteoporosis and 29 women with normal BMD from Wisconsin (Penniston et al., 2006), it was observed that serum retinyl esters were not elevated in either group, but serum retinol, total vitamin A (retinol + retinyl esters), use of supplements containing preformed vitamin A, and BMI were lower in women with osteoporosis. A trend existed for the association of serum retinyl esters as percentage of total vitamin A with osteoporosis (p = .07). Adjustment was only made for BMI and triacylglycerols. In a longitudinal study of 78 healthy young Swedish males (Hogstrom et al., 2008), the influence of retinol on peak bone mass was investigated. The mean serum retinol concentration was high (2.43 µM), but serum retinol was not associated with peak BMD or BMD changes during the 2-year follow-up. Finally, 3052 Norwegian women aged 50–70 years had forearm BMD measured with single X-ray absorptiometry in a substudy of the population-based Nord-Trøndelag Health Study (Forsmo et al., 2008). Women reporting childhood cod liver oil intake had significantly lower BMD than that of women with ingestion of cod liver oil. The authors described this result as unexpected and paradoxical, considering the good bone health intentions behind the long-standing cod liver oil recommendations in Norway. Thus, although methodological limitations exist in many of these studies, which may explain discrepancies and produce a bias that will systematically underestimate the strength of associations, the effect of vitamin A on bone mineral appears to be similar to that in animals, that is, no or little effect is observed.

Studies with Fractures as Endpoint Eleven published studies have examined the most important endpoint, fracture (Table 11.6). In the first study by Sowers and Wallace (1990) described above, fracture history was also examined. Fracture history was self-reported and obtained by interview. Only 56 women reported an atraumatic fracture, 31 in the past 10 years, among these women. No linear association was found between serum retinol and fracture risk. (Nonlinear associations were not investigated.) The Swedish BMD study (Melhus et al., 1998) also included a nested case–control study of 247 women with a first hip fracture and 873 age-matched controls. Dietary vitamin A intake was estimated from a semiquantitative FFQ, and hip fracture was identified by using hospital discharge records and confirmed by record review. For every 1-mg increase in daily intake of retinol, the risk for hip fracture increased by 68%. For intakes >1.5 mg/day compared with intakes <0.5 mg/day, the risk for hip fracture was doubled. Dietary intake of beta-carotene was not associated with BMD or hip fracture risk. In another study from Sweden (Elmstahl et al., 1998), dietary risk factors for fracture in men aged 46–68 years were studied. The diet was assessed using a combined 7-day menu book for

Vitamin A and Bone

187

hot meals, beverages, and dietary supplements and a quantitative FFQ. The incident fractures that occurred during the mean observation time of 2.4 years were retrieved from the local hospital registry and registry of X-ray examinations. The study showed that the retinol intake in Sweden was high also in men, but there was no association between retinol, calcium, or vitamin D intake and fracture risk. A serious limitation of this study is that of the 160 men who had at least one fracture; only 51 of these had fragility fractures. This may explain why physical activity paradoxically was positively associated with fractures. The study with the best estimate of vitamin A intake is the Nurses’ Health Study (Feskanich etal., 2002). The intake was assessed every 4 years using an FFQ. Mean cumulative intake data from five FFQs were obtained. There were 603 incident hip fractures resulting from low or moderate trauma. Women in the highest quintile of total vitamin A intake from food and supplements (≥3mg retinol equivalents/day) had a significantly elevated relative risk of hip fracture (1.48 [95% confidence interval [CI] 1.05–2.07]) compared with that of women in the lowest quintile of intake (<1.25mg retinol equivalents/day). This increased risk was attributable primarily to retinol (relative risk 1.89 [95% CI 1.33–2.68] comparing ≥2 mg/day vs. <0.5 mg/day). Beta-carotene did not contribute significantly to fracture risk. A retinol intake ≥1.5 mg/day as compared with an intake of <0.5 mg/day was associated with a relative risk of 1.64 for hip fracture similar to the odds ratio of 1.54 reported from Sweden (Melhus et al., 1998). Multivitamins were the primary contributors to total retinol (35%–43% of intake). Liver was the primary food source of retinol (22% in 1980 and 15% in 1994). Multivitamins were used by 34% of the cohort in 1980 and 53% by 1996. Only three prospective studies have used a biological marker of vitamin A status to assess the risk of fractures. The first was published in the New England Journal of Medicine (Michaelsson et al., 2003). In a population-based study of 2322 men, serum retinol and beta-carotene levels were measured at baseline. Fractures were documented in 266 men during 30 years of follow-up. In multivariable analysis, the relative risk was 1.64 for any fracture and 2.47 for hip fracture among men in the highest quintile for serum retinol (>2.64 µmol/L), as compared with the middle quintile (2.17–2.36 µmol/L). Because serum retinol levels do not increase until the capacity for storage in the liver is exceeded, it is interesting that there was an exponential rise in the rate–ratio curve for men with serum levels above approximately 3 µmol/L. Men with >3.6 µmol/L had an overall risk of fracture that was seven-fold greater than that in those with lower concentrations. The serum betacarotene level was not associated with risk of fracture. The relationship between fracture risk and total vitamin A and retinol intake in a prospective study of 34,703 postmenopausal women from the Iowa Women’s Health Study (Lim et al., 2004) was examined. Intake at baseline was assessed using a semiquantitative FFQ. Fractures were self-reported. Thirty-five percent of participants reported using supplements containing retinol or beta-carotene. These users of supplements had a 1.18-fold increased risk of incident hip fracture compared with that of nonusers (95% CI 0.99–1.41), but there was no evidence of a dose–response relationship or of an increased risk for total fractures. Limitations of this study include the single FFQ, that fractures were self-reported, and that no distinction was made between fractures due to high trauma events and those due to low/moderate events. In the Danish study (Rejnmark et al., 2004) described above, a case–control study of fractures was included. During the 5 years of follow-up, 163 women sustained a fracture of the appendicular skeleton and/or the vertebrae. The number of hip fractures was not reported. Each case was matched to six controls. Average dietary intake was lower (0.53 mg/day) than that in Sweden and the United States, but not associated with fracture risk. The second study that used a biological marker of vitamin A status was the prospective analysis of the NHANES I follow-up study (Opotowsky and Bilezikian, 2004) on 2799 women aged 50–74 years at baseline. There were 172 incident hip fractures during the 22-year follow-up period. AU-shaped relationship between serum vitamin A (retinol + retinyl esters) and hip fracture risk. Fracture risk was significantly higher among subjects in the lowest (≤1.61 µmol/L) and highest quintile (≥2.56 µmol/L) compared to the risk among those in the middle quintile (1.90–2.13 µmol/L).

188

Diet, Nutrients, and Bone Health

The third study using a biomarker of vitamin A status was a nested case–control study conducted in the United Kingdom (Barker et al., 2005). The results differ from those in the Swedish and American studies. Within the placebo arm of a cohort of elderly women participating in a prospective study of hip fracture, they examined serum retinol, retinyl palmitate, and beta-carotene as predictors of incident hip and other fractures. After a mean follow-up of 3.7 years, 312 osteoporotic fracture cases, of which 92 had sustained a hip fracture, were identified and matched to 934 controls. Risk of any osteoporotic fracture was slightly less in the highest quartile (>2.42 µmol/L) of serum retinol compared with that in the lowest quartile (<1.66 µmol/L) (hazard ratio [HR] 0.85; 95% CI 0.69–1.05). The composition of the multivitamins was not recorded, but multivitamin or cod liver oil supplement users had higher serum retinol levels (mean 2.07 µmol/L) than those of nonusers (mean 1.95 µmol/L) and they had a lower risk of any fracture (HR 0.76; 95% CI 0.60–0.96). Comorbidities or socioeconomic variables were not adjusted for in the analyses of this study. Also, mean retinol intakes are higher in Sweden than in the United States and United Kingdom (Scientific Advisory Committee on Nutrition, 2005). This is also reflected in the serum retinol levels (mean levels among cases were 1.91 µmol/L and 1.94 µmol/L among controls in the U.K. study, 2.02 µmol/L in the U.S. study, and in the Swedish study, the median was 2.26 µmol/L). These differences may actually be even more pronounced because the women in this British study were about 15 years older than the women in NHANES I, and serum retinol has been shown to increase with age (McLaren, 1981; Sowers and Wallace, 1990; Ballew et al., 2001b). The lowest quartile (<1.66 µmol/L) was used as reference in the U.K. study. This level is close to the lowest quintile in the U.S. study (≤1.61 µmol/L), which was associated with an increased risk of hip fracture. This could, in part, be an explanation to why risk of any osteoporotic fracture was slightly less in the highest quartile compared with the lowest quartile in the U.K. study. In the Leisure World Cohort Study in southern California (White et al., 2006), incident fractures of the hip (n = 1227), wrist (n = 445), and spine (n = 729) were identified in 13,978 residents surveyed over the course of two decades. Mean age at entry was 74.9 years for men and 73.7 years for women. Information about supplement use was collected via a questionnaire, and fractures were identified from four follow-up surveys, hospital discharge records, and death certificates. Women, but not men, who used vitamin A supplements had modestly increased rates of hip fracture (HR 1.07; 95% CI 1.00–1.15) and wrist fracture (HR 1.15; 95 CI 1.07–1.23). Finally, the relation between total vitamin A and retinol intakes and the risk of incident total and hip fracture was also examined in the WHI Observational Study (Caire-Juvera et al., 2009). Fractures were self-reported or reported by proxy respondents, and 78% of self-reported hip fractures and 71% of self-reported single-site fractures could be confirmed by medical records. Dietary intake was assessed at baseline and at year 3 of follow-up with the WHI FFQ. There were 10,405 total fractures and 588 hip fractures among the 75,747 participants during 6.6 years of follow-up. There was no association between total vitamin A or retinol and fracture risk, but women with vitamin D intake below the mean (11 µg/day) in the highest quintile of intake of both vitamin A (HR 1.19; 95% CI 1.04–1.37) and retinol (HR 1.15; 95% CI 1.03–1.29) had a modest increased risk of total fracture. The total average of vitamin D intake (from food and supplements) was 2.5–4 µg higher in this and the Iowa Women’s Health Study compared with that in the Nurses’ Health Study (Caire-Juvera et al., 2009).

SUMMARY In animals, beta-carotene and other provitamin A carotenoids had no adverse effects on bone, whereas high doses of retinol and other retinoids induced spontaneous fractures by a mechanism that reduces the bone diameter. No or little effect on mineralization has been observed, except when vitamin D was low. Consistent with animal studies, no negative effects of provitamin A carotenoids have been found in man, and most human observational studies of the relationship between retinol intake and BMD

Vitamin A and Bone

189

find no association. The methods that have been used to measure BMD determine the amount of mineral per unit area, but not volume, and it is therefore unknown if retinol reduces the bone diameter. Results from studies with fracture as endpoint show some heterogeneity, but the majority of prospective studies found that an excessive intake of vitamin A is associated with increased risk of fracture. Currently, insufficient evidence exists to determine at what level of intake the risk increases. Prospective studies with more reliable and long-term assessments of retinol intake are necessary to clarify this. High levels of retinol are only found in liver and dietary supplements containing retinol, including cod liver oil. Because the intake of liver has decreased over the last 30 years and the retinol levels in supplements recently have been reduced, the negative effects of retinol on bone health should diminish in the future. Vitamin A and bone is perhaps best summarized with the classical words of Paracelsus (1493– 1541): Alle Ding’ sind Gift, und nichts ohn’ Gift; allein die Dosis macht, daß ein Ding kein Gift ist. “All things are poison and nothing is without poison, only the dose permits something not to be poisonous.”

REFERENCES Aburto, A., and Britton, W.M. 1998a. Effects and interactions of dietary levels of vitamins A and E and cholecalciferol in broiler chickens. Poult Sci 77: 666–673. Aburto, A., and Britton, W.M. 1998b. Effects of different levels of vitamins A and E on the utilization of cholecalciferol by broiler chickens. Poult Sci 77: 570–577. Anderson, J.J. 2002. Oversupplementation of vitamin A and osteoporotic fractures in the elderly: To supplement or not to supplement with vitamin A. J Bone Miner Res 17: 1359–1362. Armstrong, R., Ashenfelter, K., Eckoff, C., et al. 1994. General and reproductive toxicology of retinoids. In The Retinoids. 2nd ed., Sp*rn, M., Roberts, A., and Goodman, D., eds. Raven Press, New York, NY. Bachmann, H., Desbarats, A., Pattison, P., et al. 2002. Feedback regulation of beta,beta-carotene 15,15′monooxygenase by retinoic acid in rats and chickens. J Nutr 132: 3616–3622. Baglin, A., Hagege, C., Franc, B., et al. 1986. A systemic-like disease: Chronic vitamin A poisoning. Ann Med Interne (Paris) 137: 142–146. Ballew, C., Bowman, B.A., Russell, R.M., et al. 2001a. Serum retinyl esters are not associated with biochemical markers of liver dysfunction in adult participants in the third National Health and Nutrition Examination Survey (NHANES III), 1988–1994. Am J Clin Nutr 73: 934–940. Ballew, C., Bowman, B.A., Sowell, A.L., et al. 2001b. Serum retinol distributions in residents of the United States: Third National Health and Nutrition Examination Survey, 1988–1994. Am J Clin Nutr 73: 586–593. Barker, M.E., Mccloskey, E., Saha, S., et al. 2005. Serum retinoids and beta-carotene as predictors of hip and other fractures in elderly women. J Bone Miner Res 20: 913–920. Binkley, N., and Krueger, D. 2000. Hypervitaminosis A and bone. Nutr Rev 58: 138–144. Bitzen, U., Winqvist, M., Nilsson-Ehle, P., et al. 1994. Retinyl palmitate is a reproducible marker for chylomicron elimination from blood. Scand J Clin Lab Invest 54: 611–613. Bliuc, D., Nguyen, N.D., Milch, V.E., et al. 2009. Mortality risk associated with low-trauma osteoporotic fracture and subsequent fracture in men and women. JAMA 301: 513–521. Booth, S.L., Tucker, K.L., Mckeown, N.M., et al. 1997. Relationships between dietary intakes and fasting plasma concentrations of fat-soluble vitamins in humans. J Nutr 127: 587–592. Burgaz, A., Akesson, A., Oster, A., et al. 2007. Associations of diet, supplement use, and ultraviolet B radiation exposure with vitamin D status in Swedish women during winter. Am J Clin Nutr 86: 1399–1404. Caire-Juvera, G., Ritenbaugh, C., Wactawski-Wende, J., et al. 2009. Vitamin A and retinol intakes and the risk of fractures among participants of the Women’s Health Initiative Observational Study. Am J Clin Nutr 89: 323–330. Carpenter, T.O., Pettifor, J.M., Russell, R.M., et al. 1987. Severe hypervitaminosis A in siblings: Evidence of variable tolerance to retinol intake. J Pediatr 111: 507–512. Cruz, J. A., Moreiras-Varela, O., Van Staveren, W.A., et al. 1991. Intake of vitamins and minerals. Euronut SENECA investigators. Eur J Clin Nutr 45(Suppl 3): 121–138. Digiovanna, J.J., Helfgott, R.K., Gerber, L.H., et al. 1986. Extraspinal tendon and ligament calcification associated with long-term therapy with etretinate. New Engl J Med 315: 1177–1182.

190

Diet, Nutrients, and Bone Health

Earnshaw, S.A., Worley, A. and Hosking, D.J. 1997. Current diet does not relate to bone mineral density after the menopause. The Nottingham Early Postmenopausal Intervention Cohort (EPIC) Study Group. Br J Nutr 78: 65–72. Elmstahl, S., Gullberg, B., Janzon, L., et al. 1998. Increased incidence of fractures in middle-aged and elderly men with low intakes of phosphorus and zinc. Osteoporos Int 8: 333–340. EPOS Group. 2002. Incidence of vertebral fracture in Europe: Results from the European Prospective Osteoporosis Study (EPOS). J Bone Miner Res 17: 716–724. FAO/WHO. 1967. Requirements of vitamin A, thiamine, riboflavin and niacin. Report of a Joint FAO/WHO Expert Group. WHO Technical Report Series, Rome, Italy. Farahmand, B.Y., Michaelsson, K., Ahlbom, A., et al. 2005. Survival after hip fracture. Osteoporos Int 16: 1583–1590. Feskanich, D., Singh, V., Willett, W.C., et al. 2002. Vitamin A intake and hip fractures among postmenopausal women. JAMA 287: 47–54. Food and Nutrition Board, Institute of Medicine. 2001 Dietary Reference Intakes for Vitamin A, Vitamin K, Arsenic, Boron, Chromium, Copper, Iodine, Iron, Manganese, Molybdenum, Nickel, Silicon, Vanadium, and Zinc. National Academy Press, Washington, DC. Forsmo, S., Fjeldbo, S.K., and Langhammer, A. 2008. Childhood cod liver oil consumption and bone mineral density in a population-based cohort of peri- and postmenopausal women: The Nord-Trondelag Health Study. Am J Epidemiol 167: 406–411. Frankel, T.L., Seshadri, M.S., Mcdowall, D.B., et al. 1986. Hypervitaminosis A and calcium-regulating hormones in the rat. J Nutr 116: 578–587. Freudenheim, J.L., Johnson, N.E., and Smith, E.L. 1986. Relationships between usual nutrient intake and bonemineral content of women 35–65 years of age: Longitudinal and cross-sectional analysis. Am J Clin Nutr 44: 863–876. Garry, P.J., Hunt, W.C., Bandrofchak, J.L., et al. 1987. Vitamin A intake and plasma retinol levels in healthy elderly men and women. Am J Clin Nutr 46: 989–994. Henriquez-Sanchez, P., Sanchez-Villegas, A., Doreste-Alonso, J., et al. 2009. Dietary assessment methods for micronutrient intake: A systematic review on vitamins. Br J Nutr 102(Suppl 1): S10–S37. Hernandez-Avila, M., Stampfer, M. J., Ravnikar, V. A., et al. 1993. Caffeine and other predictors of bone density among pre- and perimenopausal women. Epidemiology 4: 128–134. Hickenbottom, S.J., Follett, J.R., Lin, Y., et al. 2002. Variability in conversion of beta-carotene to vitamin A in men as measured by using a double-tracer study design. Am J Clin Nutr 75: 900–907. Hogstrom, M., Nordstrom, A. and Nordstrom, P. 2008. Retinol, retinol-binding protein 4, abdominal fat mass, peak bone mineral density, and markers of bone metabolism in men: The Northern Osteoporosis and Obesity (NO2) Study. Eur J Endocrinol 158: 765–770. Hough, S., Avioli, L.V., Muir, H., et al. 1988. Effects of hypervitaminosis A on the bone and mineral metabolism of the rat. Endocrinology 122: 2933–2939. Houtkooper, L.B., Ritenbaugh, C., Aickin, M., et al. 1995. Nutrients, body composition and exercise are related to change in bone mineral density in premenopausal women. J Nutr 125: 1229–1237. International Dairy and Federation. 1995. Consumption statistics for milk and milk products. Bull Int Dairy Fed 301: 14–16. Johansson, S., Lind, P.M., Hakansson, H., et al. 2002. Subclinical hypervitaminosis A causes fragile bones in rats. Bone 31: 685–689. Johansson, S., and Melhus, H. 2001. Vitamin A antagonizes calcium response to vitamin D in man. J Bone Miner Res 16: 1899–1905. Johnell, O., Gullberg, B., Allander, E., et al. 1992. The apparent incidence of hip fracture in Europe: A study of national register sources. Osteoporos Int 2: 298–302. Jowsey, J. and Riggs, B. 1968. Bone changes in a patient with hypervitaminosis A. J Clin Endocrinol Metab 28: 1833–1835. Kamm, J., Ashenfelter, K., and Ehmann, C. 1984. Preclinical and clinical toxicology of selected retinoids. In The Retinoids. 1st ed., Sp*rn, M., Roberts, A., and Goodman, D., eds. Academic Press, London. Kanis, J.A., Johnell, O., De Laet, C., et al. 2002. International variations in hip fracture probabilities: Implications for risk assessment. J Bone Miner Res 17: 1237–1244. Kaptoge, S., Welch, A., Mctaggart, A., et al. 2003. Effects of dietary nutrients and food groups on bone loss from the proximal femur in men and women in the 7th and 8th decades of age. Osteoporos Int 14: 418–428. Kardinaal, A.F., Van ‘t Veer, P., Brants, H.A., et al. 1995. Relations between antioxidant vitamins in adipose tissue, plasma, and diet. Am J Epidemiol 141: 440–450.

Vitamin A and Bone

191

Kneissel, M., Studer, A., Cortesi, R., et al. 2005. Retinoid-induced bone thinning is caused by subperiosteal osteoclast activity in adult rodents. Bone 36: 202–214. Krasinski, S.D., Cohn, J.S., Schaefer, E.J., et al. 1990. Postprandial plasma retinyl ester response is greater in older subjects compared with younger subjects. Evidence for delayed plasma clearance of intestinal lipoproteins. J Clin Invest 85: 883–892. Krasinski, S.D., Russell, R.M., Otradovec, C.L., et al. 1989. Relationship of vitamin A and vitamin E intake to fasting plasma retinol, retinol-binding protein, retinyl esters, carotene, alpha-tocopherol, and cholesterol among elderly people and young adults: Increased plasma retinyl esters among vitamin A-supplement users. Am J Clin Nutr 49: 112–120. Lim, L.S., Harnack, L.J., Lazovich, D., et al. 2004. Vitamin A intake and the risk of hip fracture in postmenopausal women: The Iowa Women’s Health Study. Osteoporos Int 15: 522–529. Lin, Y., Dueker, S.R., Burri, B.J., et al. 2000. Variability of the conversion of beta-carotene to vitamin A in women measured by using a double-tracer study design. Am J Clin Nutr 71: 1545–1554. Lind, P. M., Johansson, S., Ronn, M., et al. 2006. Subclinical hypervitaminosis A in rat: Measurements of bone mineral density (BMD) do not reveal adverse skeletal changes. Chem Biol Interact 159: 73–80. Macdonald, H.M., New, S.A., Golden, M.H., et al. 2004. Nutritional associations with bone loss during the menopausal transition: Evidence of a beneficial effect of calcium, alcohol, and fruit and vegetable nutrients and of a detrimental effect of fatty acids. Am J Clin Nutr 79: 155–165. Maggio, D., Barabani, M., Pierandrei, M., et al. 2003. Marked decrease in plasma antioxidants in aged osteoporotic women: Results of a cross-sectional study. J Clin Endocrinol Metab 88: 1523–1527. Maggio, D., Polidori, M.C., Barabani, M., et al. 2006. Low levels of carotenoids and retinol in involutional osteoporosis. Bone 38: 244–248. McDowell, L. 2000. Vitamins in Animal and Human Nutrition. Iowa State University Press, Ames, Iowa. McLaren, D.S. 1981. The luxus vitamins—A and B12. Am J Clin Nutr 34: 1611–1616. Melhus, H. 2003. Vitamin A and fracture risk. In Nutritional Aspects of Bone Health. J.-P. Bonjour, and New, S., eds. The Royal Society of Chemistry, Cambridge, UK, 369–372. Melhus, H., Michaëlsson, K., Kindmark, A., et al. 1998. Excessive dietary intake of vitamin A is associated with reduced bone mineral density and increased risk for hip fracture. Ann Intern Med 129: 770–778. Michaelsson, K., Lithell, H., Vessby, B., et al. 2003. Serum retinol levels and the risk of fracture. New Engl J Med. 348: 287–294. Milstone, L.M., Mcguire, J., and Ablow, R.C. 1982. Premature epiphyseal closure in a child receiving oral 13-cis-retinoic acid. J Am Acad Dermatol 7: 663–666. Nieman, C., and Obbink, H. 1954. The biochemistry and pathology of hypervitaminosis A. Vitam Horm 12: 69–99. Olson, J.A. 1984. Serum levels of vitamin A and carotenoids as reflectors of nutritional status. J Natl Cancer Inst 73: 1439–1444. Opotowsky, A.R., and Bilezikian, J.P. 2004. Serum vitamin A concentration and the risk of hip fracture among women 50 to 74 years old in the United States: A prospective analysis of the NHANES I follow-up study. Am J Med 117: 169–174. Penniston, K.L., Weng, N., Binkley, N., et al. 2006. Serum retinyl esters are not elevated in postmenopausal women with and without osteoporosis whose preformed vitamin A intakes are high. Am J Clin Nutr 84: 1350–1356. Phillips, B., Ball, C., Sackett, D., et al. 2001. Levels of evidence. Oxford Centre for Evidence-Based Medicine. www.cebm.net/index.aspx?o+1047. Accessed March 10, 2010. Pittsley, R.A., and Yoder, F.W. 1983. Retinoid hyperostosis. Skeletal toxicity associated with long-term administration of 13-cis-retinoic acid for refractory ichthyosis. New Engl J Med. 308: 1012–1014. Prendiville, J., Bingham, E.A., and Burrows, D. 1986. Premature epiphyseal closure—A complication of etretinate therapy in children. J Am Acad Dermatol 15: 1259–1262. Promislow, J.H., Goodman-Gruen, D., Slymen, D.J., et al. 2002. Retinol intake and bone mineral density in the elderly: The Rancho Bernardo Study. J Bone Miner Res 17: 1349–1358. Reinersdorff, D.V., Bush, E. and Liberato, D.J. 1996. Plasma kinetics of vitamin A in humans after a single oral dose of [8,9,19–13C]retinyl palmitate. J Lipid Res 37: 1875–1885. Rejnmark, L., Vestergaard, P., Charles, P., et al. 2004. No effect of vitamin A intake on bone mineral density and fracture risk in perimenopausal women. Osteoporos Int 15: 872–880. Roidt, L., White, E., Goodman, G.E., et al. 1988. Association of Food Frequency Questionnaire estimates of vitamin A intake with serum vitamin A levels. Am J Epidemiol 128: 645–654. Ruby, L.K., and Mital, M.A. 1974. Skeletal deformities following chronic hypervitaminosis A: A case report. J Bone Joint Surg Am 56: 1283–1287.

192

Diet, Nutrients, and Bone Health

Sahi, T. 1994. Genetics and epidemiology of adult-type hypolactasia. Scand J Gastroenterol Supplement 202: 7–20. Sahni, S., Hannan, M.T., Blumberg, J., et al. 2009a. Inverse association of carotenoid intakes with 4-y change in bone mineral density in elderly men and women: The Framingham Osteoporosis Study. Am J Clin Nutr 89: 416–424. Sahni, S., Hannan, M.T., Blumberg, J., et al. 2009b. Protective effect of total carotenoid and lycopene intake on the risk of hip fracture: A 17-year follow-up from the Framingham Osteoporosis Study. J Bone Miner Res 24: 1086–1094. Schurr, D., Herbert, J., Habibi, E., et al. 1983. Unusual presentation of vitamin A intoxication. J Pediatr Gastroenterol Nutr 2: 705–707. Scientific Advisory Committee on Nutrition. 2005. Review of dietary advice on vitamin A. The Stationary Office, London. Sigurdsson, G., Franzon, L., Thorgeirsdottir, H., et al. 2001. A lack of association between excessive dietary intake of vitamin A and bone mineral density in seventy-year old Icelandic women. In Nutritional Aspects of Osteoporosis. Burckhardt, P., Dawson-Hughes, B., and Heaney, R.P., eds. Academic Press, San Diego, CA. Smith, F., and Goodman, D. 1976. Vitamin A transport in human vitamin A toxicity. New Engl J Med 294: 805–808. Sowers, M.F., and Wallace, R.B. 1990. Retinol, supplemental vitamin A and bone status. J Clin Epidemiol 43: 693–699. Sowers, M.R., Wallace, R.B. and Lemke, J.H. 1985. Correlates of mid-radius bone density among postmenopausal women: A community study. Am J Clin Nutr 41: 1045–1053. Sp*rn, M., Roberts, A., and Goodman, D. 1994. The Retinoids: Biology, Chemistry and Medicine. Raven Press, New York, NY. Suzuki, Y., Whiting, S.J., Davison, K.S., et al. 2003. Total calcium intake is associated with cortical bone mineral density in a cohort of postmenopausal women not taking estrogen. J Nutr Health Aging 7: 296–299. Takitani, K., Zhu, C.L., Inoue, A., et al. 2006. Molecular cloning of the rat beta-carotene 15,15′-monooxygenase gene and its regulation by retinoic acid. Eur J Nutr 45: 320–326. Tangney, C.C., Shekelle, R.B., Raynor, W., et al. 1987. Intra- and interindividual variation in measurements of beta-carotene, retinol, and tocopherols in diet and plasma. Am J Clin Nutr 45: 764–769. Teelmann, K. 1981. Experimental toxicology of the aromatic retinoid Ro 10–9359 (etretinate). In Retinoids: Advances in Basic Research and Therapy. Orfanos, C., Braun-Falco, O., Farber, E., Grupper, C., Polano, M., and Schuppli, R., eds. Springer-Verlag, Berlin and Heidelberg. van Lieshout, M., West, C.E., Muhilal, et al. 2001. Bioefficacy of beta-carotene dissolved in oil studied in children in Indonesia. Am J Clin Nutr 73: 949–958. van Vliet, T., Van Vlissingen, M.F., Van Schaik, F., et al. 1996. Beta-carotene absorption and cleavage in rats is affected by the vitamin A concentration of the diet. J Nutr 126: 499–508. White, S.C., Atchison, K.A., Gornbein, J.A., et al. 2006. Risk factors for fractures in older men and women: The Leisure World Cohort Study. Gend Med 3: 110–123. Willett, W.C., Reynolds, R.D., Cottrell-Hoehner, S., et al. 1987. Validation of a semi-quantitative Food Frequency Questionnaire: Comparison with a 1-year diet record. J Am Diet Assoc 87: 43–47. Willett, W.C., Stampfer, M.J., Underwood, B.A., et al. 1983. Vitamins A, E, and carotene: Effects of supplementation on their plasma levels. Am J Clin Nutr 38: 559–566. Wolf, R.L., Cauley, J.A., Pettinger, M., et al. 2005. Lack of a relation between vitamin and mineral antioxidants and bone mineral density: Results from the Women’s Health Initiative. Am J Clin Nutr 82: 581–588. Wu, B., Xu, B., Huang, T.Y., et al. 1996. [A model of osteoporosis induced by retinoic acid in male Wistar rats]. Yao Hsueh Hsueh Pao 31: 241–245. Yano, K., Heilbrun, L.K., Wasnich, R.D., et al. 1985. The relationship between diet and bone mineral content of multiple skeletal sites in elderly Japanese-American men and women living in Hawaii. Am J Clin Nutr 42: 877–888.

12

Vitamin K and Bone Cees Vermeer and Marjo H.J. Knapen

CONTENTS Introduction..................................................................................................................................... 193 Characteristics of Vitamin K........................................................................................................... 193 Vitamin-K–Dependent Proteins in Bone........................................................................................ 194 Vitamin K Status and Bone Health................................................................................................. 195 Intervention Trials........................................................................................................................... 196 Vitamin K in Childhood.................................................................................................................. 197 Safety of High Vitamin K Intake.................................................................................................... 198 Conclusions..................................................................................................................................... 198 References....................................................................................................................................... 199

INTRODUCTION The human diet contains different forms of vitamin K. Green vegetables, notably spinach, kale, broccoli, and sprouts, are the main sources of phylloquinone (vitamin K1); menaquinones (vitamin K2) form a family of closely related compounds produced by a number of microorganisms including lactic acid and propionic acid bacteria, as well as by Bacillus subtilis natto. In Europe and Northern America, cheese and curd cheese are the main dietary sources of vitamin K2, whereas in Japan, natto (fermented soy beans), a popular food, is extremely rich in menaquinone-7, one of the long-chain menaquinones (Booth et al., 1993; Shearer and Bolton-Smith, 2000; Schurgers and Vermeer, 2000). Also, bacteria in the gut produce large amounts of vitamin K2, but hardly any of it is absorbed because of the lack of bile salts at the site of production. Dietary vitamin K2 accounts for approximately 10% of the total vitamin K intake, but its better absorption and much longer half-life than K1 make it an important contributor to total human vitamin K status.

CHARACTERISTICS OF VITAMIN K A common characteristic of all forms of vitamin K is the methylated naphthoquinone ring system substituted with a variable aliphatic side chain at the 3-position. In K1, this side chain is constituted of four isoprenoid residues, the last three of which are saturated. In K2, the number of isoprenyl residues is variable, but all are unsaturated. The nomenclature is MK-n, where n stands for the number of isoprenyl residues in the side chain. Obviously, the lipophilic character of the menaquinones increases at increasing side-chain length. The most common menaquinones in the human diet are MK-4 through MK-6 (classified as the short-chain menaquinones) and MK-7 through MK-9 (the long-chain menaquinones). The function of all forms of vitamin K is to serve as a cofactor for the endoplasmic reticulum (ER) enzyme gammaglutamate carboxylase (GGCX), which converts certain polypeptide-bound glutamate residues into gammacarboxyglutamate (Gla). This vitamin-K-dependent carboxylation is a posttranslational modification accomplished during the maturation of secretory proteins that are equipped with a GGCX recognition sequence. Presently, 15 Gla-containing proteins have been 193

194

Diet, Nutrients, and Bone Health

identified, including four blood coagulation factors (all synthesized in the liver), the bone Glaprotein osteocalcin (exclusively synthesized by osteoblasts) and matrix Gla-protein (MGP). MGP is mainly synthesized by chondrocytes in cartilage and by smooth muscle cells in the arterial vessel wall. In these proteins, the Gla residues form strong calcium-binding groups which are essential for their biological function. During vitamin K deficiency or after intake of vitamin K antagonists, for example, warfarin (coumadin), uncarboxylated Gla-proteins devoid of biological activity, is formed (Berkner, 2005; Shearer and Newman, 2008). Following intestinal absorption, the K vitamins are taken up in triglycerides to be transported to the liver, where they can be utilized for the synthesis of the clotting factors. Notably, the longchain menaquinones, however, are subsequently incorporated into low-density lipoproteins which are transported to extrahepatic tissues such as bone and vessel wall (Schurgers and Vermeer, 2002). During the last decade, experiments in vitamin-K-deficient animals and in coumarin-treated subjects have unfolded a general principle in the pharmaco*kinetics of K vitamins: the liver takes what it needs and the excess (mainly K2) is redistributed to other tissues. In coumarin-treated volunteers treated with increasing doses of vitamin K, this principle is exemplified by the preferential carboxylation of the clotting factors (hepatic) well before the first sign of osteocalcin (extrahepatic) carboxylation was observed (Schurgers et al., 2004). These data were found for both K1 and K2 and are consistent with the recently published triage theory, which holds that during episodes of dietary vitamin K insufficiency, the body selectively transports vitamins to tissues carrying out functions that are vital for immediate survival (McCann and Ames, 2009). Obviously, prevention of bleeding is more important for survival than bone loss or vascular calcification in old age. In this way, it can be understood why in the healthy population having undercarboxylation of the clotting factors does not exist, whereas 20%–30% of the osteocalcin and MGP circulate in their uncarboxylated, inactive forms. Although high vitamin K intake was shown to be associated with low risk for osteoporosis (Feskanich et al., 1999), cardiovascular disease (Gast et al., 2009), and cancer (Nimptsch et al., 2008), research has not examined whether long-term subclinical vitamin K deficiencies contribute to age-related diseases. This concept of chronic vitamin K deficiency also raises the question of how to define human vitamin K requirement. Dietary reference intakes (DRIs) for vitamin K in various countries range between 1 and 1.5 µg /kg body weight/day or between 80 and 120 µg/day (for adults). These figures are based on the doses required to ensure complete carboxylation of the clotting factors; many studies have shown that these amounts are insufficient to support full carboxylation of the extrahepatic Gla-proteins. Here, we define vitamin K deficiency as a state in which the hepatic vitamin K status is too low to allow complete carboxylation of all clotting factors, whereas vitamin K insufficiency is defined as a state in which more than 5% of an extrahepatic Gla-protein occurs in its uncarboxylated form. According to this definition, nearly all nonsupplemented healthy adults are vitamin K insufficient. Before addressing the question of whether increased vitamin K intake—for instance, by using nutrient supplements or functional foods—may have benefits for bone health, we describe the various Gla-proteins reported to reside in bone.

VITAMIN-K–DEPENDENT PROTEINS IN BONE Thus far, five Gla-proteins have been identified in or associated with bone: osteocalcin (Hauschka and Reid, 1978), MGP (Price et al., 1983), growth-arrest-specific gene 6 protein (Gas6) (Katagiri et al., 2001), protein S (Maillard et al., 1992), and periostin, a protein abundantly found in the periosteum which was recently found to contain four widely separated Gla domains (Coutu et al., 2008). MGP is the most potent inhibitor of soft tissue calcification presently known, and it is synthesized in chondrocytes and, under certain conditions, also in osteoblasts. Gas6 is a cell growth regulator also involved in bone differentiation and resorption (Katagiri et al., 2001). One of the functions of protein S is to serve as a cofactor for protein C in the inhibition of the blood coagulation cascade; in the circulation, a large fraction of serum proteins is bound to the C4-binding protein of the

Vitamin K and Bone

195

complement system, but the function for the C4BP/PS complex remains unclear. No specific role for protein S in bone has been reported, but it is synthesized in various cells including osteoblasts (Maillard et al., 1992). Hereditary protein S deficiency has been associated with osteopenia (Pan et al., 1990). Periostin has primarily been studied as a recombinant protein produced in the absence of the vitamin-K-dependent carboxylase system. In this form, in vitro studies have demonstrated a potential role in extracellular matrix mineralization, whereas the role of carboxylated periostin remains to be elucidated. The most recently discovered Gla-protein is the Gla-rich protein (GRP), which is expressed in many vertebrate tissues, but most abundantly in cartilage. GRP has been suggested to play a role in calcium regulation in the extracellular environment (Viegas et al., 2008). Remarkably, no clear function has been described for osteocalcin, the most abundant noncollagenous protein in bone. Transgenic mice showed that it is a negative regulator of bone growth and that it may have a role in the appropriate deposition of hydroxyapatite crystals in bone and that it most likely functions as a regulator of bone mineral maturation (Ducy et al., 1996). Other researchers have found that osteocalcin has chemoattractant activity for osteoclasts. Recently, osteocalcin has also been postulated to act as a hormone that stimulates fat metabolism in adipose tissue (Lee et al., 2007). At this time, however, the precise function of osteocalcin is unclear, and its main practical importance is gained from its use as a circulating biomarker for osteoblast activity (total osteocalcin antigen) and for the vitamin K status of bone tissue (fraction of osteocalcin that is uncarboxylated). Since so many vitamin-K-dependent proteins have been identified in bone, and because at least two of them (osteocalcin and MGP) occur in a partly uncarboxylated form in the majority of the adult population, vitamin K intake is likely associated with bone health. Cell culture experiments in primary osteoblasts showed a marked stimulation of calcification by all forms of vitamin K, and these studies suggest that vitamin K promotes the transition of osteoblasts into osteocytes while at the same time decreasing the osteolytic potential of these osteocytes. The mechanisms by which vitamin K optimizes calcification, as part of the process of bone formation, and integrity in vivo may help explain the net positive effect of vitamin K on bone formation (Atkins et al., 2009).

VITAMIN K STATUS AND BONE HEALTH Many observational studies have demonstrated that poor vitamin K status is associated with low bone mineral density (BMD) and increased fracture risk. In these studies, vitamin K status was assessed in one of three ways: by dietary intake (food frequency questionnaires), by circulating vitamin K concentrations (only vitamin K1 was measured), or by the degree to which circulating osteocalcin had been carboxylated (reflecting vitamin K status of bone tissue). Pioneers in this fields were Hart and colleagues, who observed unusually low circulating vitamin K levels in patients with bone fractures and in osteoporotic patients (Hart et al., 1984). Feskanich and colleagues and Booth and coworkers first associated dietary K intake with bone health (Feskanich et al., 1999; Booth et al., 2003). Szulc and coinvestigators and Luukinen and coresearchers found strong correlations between the circulating levels of uncarboxylated osteocalcin (ucOC) and bone health (Szulc et al., 1996; Luukinen et al., 2000). These studies have been confirmed by other researchers. Remarkably, all the investigators mentioned above first found the relation of vitamin K deficiency with fracture risk, and only years later with low BMD. This paradox may be related to the role of vitamin K in regulating the synthesis of bone matrix proteins, especially collagen which contributes to bone strength independently of BMD. An important question is what is the most reliable method to establish vitamin K status. Circulating vitamin K species mostly reflect the diet of the day prior to blood sampling, but analytical procedures are insufficiently sensitive to detect the low levels of circulating long-chain menaquinones, which also have K activity and which are especially important for extrahepatic tissues. Food frequency questionnaires give an overall picture of dietary intakes and a fair estimate of dietary preferences, but they are not highly accurate and do not correct for food matrix. K1 from green vegetables, for instance, is absorbed at efficiencies of only 5% to 15% (depending on concomitant

196

Diet, Nutrients, and Bone Health

fat intake), whereas the absorption of K2 from cheese is almost complete, that is, ~100%. It seems, therefore, that tissue-specific Gla-proteins, like osteocalcin of bone, form the most reliable markers for vitamin K status of the tissue in which they were produced, especially if the noncarboxylated protein is expressed as a fraction of the total protein antigen or as a ratio between uncarboxylated and carboxylated protein.

INTERVENTION TRIALS The effect of vitamin K on bone strength remains a matter of controversy, mainly because many of the studies showing a beneficial effect were underpowered or too short in duration. Only a limited number of clinical intervention trials published today meet the criteria of more than 2 years of treatment and more than 100 subjects investigated. These trials are summarized in Table 12.1. The study reported by Shiraki et al. (2000) included 241 osteoporotic women receiving either calcium (150 mg/day) or calcium plus vitamin K 2 (45 mg/day) over a period of 2 years. Lumbar BMD decreased in the control group (calcium alone), but it remained constant in the vitamin-K2-treated group. Also, the incidence of clinical fractures was significantly higher in the control group than that in K2-treated group. Braam et al. (2003) included 180 nonosteoporotic postmenopausal women randomized into three groups of 60 each. They were treated for 2 years with placebo, calcium + vitamin D, or calcium+ vitamin D + vitamin K1 (1 mg/day). Vitamin K1 decreased the rate of bone loss, that is, BMD, as measured by dual-energy x-ray absorptiometry, by about 30% as compared with the calcium + vitamin D group. Knapen et al. (2007) included 325 nonosteoporotic postmenopausal women randomized to receive either placebo or vitamin K2 (45 mg/day) for 3 years. No effect on BMD was observed, but it was shown that vitamin K2 induced a beneficial change in the geometry of the hip. The calculated bone strength of the hip remained constant during the entire study period in the K2 group, whereas a significant decrease of the calculated bone strength was observed in the placebo group. Similar to the results of Braam et al. (2003), Bolton-Smith et al. (2007) showed a synergetic TABLE 12.1 Overview of Randomized Clinical Intervention Studies Testing Effects of K Vitamins on Bone Quality and Bone Strength Investigator (Year of Publication) Shiraki et al. (2000) Braam et al. (2003)

Bone Quality (BMD/BMC)

Cheung et al. (2008)

Positive effect on lumbar BMD Positive effect on hip BMD Rate of bone loss decreased No effect on BMD Positive effect on hip BMC Positive effect on BMC and BMD Only seen at ultradistal radius No effect on BMD or BMC

Booth et al. (2008) Inoue et al. (2009)

No effect on BMD or BMC Not measured

Knapen et al. (2007) Bolton-Smith et al. (2007)

Bone Strength (Fractures, Geometry) Fracture incidence decreased Not measured Femoral neck width increased Positive effect on indices for hip bone strength Not measured Decreased fracture incidence Study not designed for fracture risk monitoring Not measured Fracture risk decreased Only found in patients with >4 vertebral fractures

Notes: The references listed in this table describe all vitamin K intervention trials published until December 2009 meeting the criteria of a treatment period of at least 2 years and at least 100 subjects included in the trial. Conclusions in this paper are based on these seven trials.

Vitamin K and Bone

197

effect of 200 µg/day of vitamin K and calcium + vitamin D on bone metabolism in a 2-year study among 244 postmenopausal women, This study, however, was hampered by the fact that in the placebo group no bone loss was observed at the most critical site, that is, the hip. Cheung et al. (2008) reported a study among 440 osteopenic women receiving either placebo or vitamin K1 (5 mg/day) for 2 years. These researchers did not find an effect of vitamin K on BMD, but—although the study was not powered to examine fractures or cancers—significant decreases of the incidence of both fractures and cancers in the vitamin-K-treated group were found. The reduced cancer incidence is remarkable, because recently a large population-based study also showed an inverse correlation between vitamin K intake and prostate cancer incidence (Nimptsch et al., 2008). Booth et al. (2008) enrolled 452 elderly men and women in a 3-year study who received either a multivitamin containing calcium and vitamin D or the same multivitamin plus 500 µg/day of vitamin K1. No effect of vitamin K treatment was found. Finally, Inoue et al. (2009) reported a study among over 4000 postmenopausal women receiving either calcium or calcium plus vitamin K 2 (45 mg/day) for 3 years. Although the cumulative 48-month incidence rate of new clinical fractures was lower in the combined therapy group, the difference was only significant in the subgroup with at least five baseline fractures. Also, the loss of height was less with the combined therapy than with monotherapy among patients 75 years of age or older at enrollment and those with at least five vertebral fractures at enrollment. The authors concluded that vitamin K2 may especially prevent vertebral fractures in patients with more advanced osteoporosis. Throughout the literature, the effects of vitamin K status on the reduction of fracture incidence are remarkably more evident than improvements of BMD or bone mineral content (BMC), even in trials not designed to monitor fractures (Cheung et al., 2008). One explanation for this persistent observation of a lowered fracture incidence is that BMD and BMC are not appropriate endpoints to monitor the effect of vitamin K on bone strength or fracture risk. Another explanation for the lack of effect in some epidemiologic studies is that only poor vitamin K status is associated with increased fracture risk. It would be logical to investigate the effect of vitamin K on bone health in subjects preselected for poor dietary vitamin K status, but none of the published reports have included subjects with poor vitamin K intakes. We recommend that a placebo-controlled study be performed in early postmenopausal women, who are at great risk of rapid bone loss and who have established poor dietary vitamin K status as a prerequisite risk factor.

VITAMIN K IN CHILDHOOD Children’s bone is different from adult bone in that it continues to grow in length, and it rapidly accumulates mass up to late adolescence. These characteristics are demonstrated by the high activity of the osteoblasts that generate the high concentrations of circulating bone formation markers including osteocalcin. Serum concentrations of both inactive ucOC and the active form, carboxylated osteocalcin (cOC), are high, as is the ratio ucOC/cOC, suggesting a relative excess of ucOC which is characteristic of poor vitamin K status during bone growth (Van Summeren et al., 2007). Apparently, vitamin K is depleted from bone tissue because of the 10-fold higher osteocalcin synthesis rate of children compared with adult bone, resulting in a 10-fold higher vitamin K requirement for the growth of bone compared with the maintenance needs of vitamin K by adult bone. The increased requirement is compensated for neither by a higher dietary vitamin K intake nor by increased vitamin K uptake by osteoblasts from the blood circulation, resulting in a marked vitamin K insufficiency of the bone tissue. Recent studies have demonstrated that vitamin K insufficiency is more pronounced during childhood than in any other stage of life and that an inverse relationship exists between serum ucOC levels and both the rate of bone turnover (Kalkwarf et al., 2004) and BMD (Van Summeren et al., 2008). Before answering the question of whether children will benefit from increasing their intakes of vitamin K, it is important to learn if we can correct the apparent vitamin K insufficiency of

198

Diet, Nutrients, and Bone Health

children’s bone by giving them vitamin K supplements in doses not exceeding the DRIs. The first placebo-controlled vitamin K2 intervention study in healthy children between 6 and 10 years of age was published by Van Summeren et al. (2009). The children received either vitamin K2 (45 µg/day of menaquinone-7 as MenaQ7 capsules) or placebo during 8 weeks, and it was demonstrated that after the 8-week period the ucOC/cOC ratio rapidly declined in the vitamin K-treated group. The dose of 45 µg/day is comparable with the DRI for children of that age. Now that it has been established that the vitamin K status in children’s bone is extremely low and that it can be corrected by vitamin K supplements, the question arises whether higher vitamin K intakes should be recommended for children. Although no obvious dangers or disadvantages exist, it is not known whether children would benefit from vitamin K supplements given during their entire growth period. Vitamin K intake by children has declined substantially during the past half century (Prynne et al., 2005). Present strategies of bone health promotion and osteoporosis prevention include aiming to achieve a high peak bone mass in the young adults so that they are optimally prepared to withstand bone loss during the later stages of life without fracture. Whether or not vitamin K should be included in this strategy is not clear at this time.

SAFETY OF HIGH VITAMIN K INTAKE The best known function of vitamin K is its requirement for the synthesis of functional blood clotting factors and therefore normalization of hemostasis in vitamin-K-deficiency bleeding. During hepatic vitamin K deficiency or when vitamin K antagonists are used, uncarboxylated species of the clotting factors are produced, which lack procoagulant activity. Experimental animals receiving vitamin-Kdeficient diets develop severe vitamin K deficiency and exsanguinate within 3 weeks. Under these conditions, administration of vitamin K quickly normalizes hemostasis and prevents death (Groenenvan Dooren et al., 1993). A widespread misperception is that, in subjects with a normal hemostasis, extra vitamin K intake would lead to a hypercoaguble state and hence to an increased risk of thrombosis and thromboembolism. No scientific support for this notion, however, exists. This scenario is evidently not true. The most sensitive technique presently available to assess thrombosis risk is the endogenous thrombin potential (ETP) in which platelet-free plasma is activated to form thrombin; the amount of thrombin formed during a 30-minute period is then monitored using a chromogenic method (Hemker et al., 2006). Adverse effects monitoring in several of our intervention trials included ETP measurement, and in this way, several hundreds of participants treated with 10 mg/day of vitamin K1 or even 45 mg/day of vitamin K 2 were evaluated for increased thrombosis tendency during intervention periods of 2 to 3 years. Neither increased thrombosis risk nor other adverse effects were found in these studies. These results are consistent with the statement of the Institute of Medicine (2001) that vitamin K has a very wide safety range and that no evidence of toxicity has been associated with the intake of either phylloquinone (vitamin K1) or menaquinone (vitamin K2). Although no upper tolerable levels for vitamin K have been defined, it is well known that subjects using oral anticoagulants, such as warfarin, acenocoumarol, or phenprocoumon, should avoid taking high doses of vitamin K (either by diet or as supplements), especially if the intake fluctuates from day to day. Oral anticoagulant drugs act as vitamin K antagonists and are prescribed during episodes of high thrombosis risk. Obviously, such drugs should not be combined with high doses of vitamin K. In a study among anticoagulated volunteers, low-dose vitamin K intake (100 µg/day of K1 or 50 µg/day of K2) was demonstrated not to affect the level of anticoagulation (international normalized ratio) in a clinically relevant way (Schurgers et al., 2007).

CONCLUSIONS Although the results remain inconclusive, the majority of published studies have shown a mild positive effect of high vitamin K intake, that is, an adequate vitamin K status, on bone strength,

Vitamin K and Bone

199

especially if taken in combination with calcium and vitamin D (co*ckayne et al., 2006; Iwamoto et al., 2009). Most negative outcomes were found in trials of 1 year or shorter, whereas long-term intervention trials and retrospective studies of life-long dietary habits almost invariably showed a positive correlation between high vitamin K intake and bone health. Many indications suggest that BMD, BMC, or circulating biomarkers for bone metabolism are not suitable endpoints to establish an effect of vitamin K on bone health. Since the molecular functions of vitamin K–dependent proteins in bone are largely unknown, it is difficult to determine the appropriate surrogate markers indicative of bone health in intervention studies. Therefore, large trials in which fracture risk or calculated bone strength are used as endpoints may be the one way to demonstrate unequivocally the importance of vitamin K for bone health. Moreover, such trials should be performed using selected subjects with poor bone vitamin K status, based on their assessment of low osteocalcin carboxylation. Vitamin K supplements may also be important in another way. Women at risk of rapid bone loss or with high fracture risk often receive supplements containing calcium, vitamin D, or both. Recently, it was suggested that either of these regimens induces coronary artery calcification and cardiovascular mortality (Bolland et al., 2008). MGP is the only known calcification inhibitor in the arterial vessel wall; hence, it is the major tool for the vessel wall to prevent local arterial calcification. Like other extrahepatic Gla-proteins, however, MGP is not fully carboxylated in the majority of the people examined (Schurgers et al., 2008), which means that a substantial fraction of MGP produced in the vessel wall remains inactive. Hence, the calcification-inhibitory potential of the vasculature can be improved by increased vitamin K intake, which is especially important during periods of increased calcium loading, to increase carboxylation of the GRPs. Therefore, we recommend that extra vitamin K should always accompany calcium supplements or calcium plus vitamin D supplements.

REFERENCES Atkins, G.J., Welldon, K.J., Wijenayaka, A.R., et al. 2009. Vitamin K promotes mineralization, osteoblast to osteocyte transition and an anti-catabolic phenotype by {gamma}-carboxylation dependent and independent mechanisms. Am J Physiol Cell Physiol 297: C1358–C1367. Berkner, K.L. 2005. The vitamin K-dependent carboxylase. Annu Rev Nutr 25: 127–149. Bolland, M.J., Barber, P.A., Doughty, R.N., et al. 2008. Vascular events in healthy older women receiving calcium supplementation: Randomised controlled trial. Brit Med J 336: 262–266. Bolton-Smith, C., McMurdo, M.E.T., Paterson, C.R., et al. 2007. Two-year randomized controlled trial of vitamin K1 (phylloquinone) and vitamin D3 plus calcium on the bone health of older women. J Bone Miner Res 22: 509–519. Booth, D.L., Dallal, G., Shea, M.K., et al. 2008. Effect of vitamin K supplementation on bone loss in elderly men and women. J Clin Endocrinol Metab 93: 1217–1223. Booth, S.L., Broe, K.E., Gagnon, D.R., et al. 2003. Vitamin K intake and bone mineral density in women and men. Am J Clin Nutr 77: 512–516. Booth, S.L., Sadowski, J.A., Weihrauch, J.L., et al. 1993. Vitamin K1 (phylloquinone) content of foods: Aprovisional table. J Food Comp Anal 6: 109–120. Braam, L.A.J.L.M., Knapen, M.H.J., Geusens, P., et al. 2003. Vitamin K1 supplementation retards bone loss in postmenopausal women between 50 and 60 years of age. Calcif Tissue Int 73: 21–26. Cheung, A.M., Tile, L., Lee, Y., et al. 2008. Vitamin K supplementation in postmenopausal women with osteopenia (ECKO trial): A randomized controlled trial. PLoS Med 5: 1461–1472. co*ckayne, S., Adamson, J., Lanham-New, S., et al. 2006. Vitamin K and the prevention of fractures: Systematic review and meta-analysis of randomized controlled trials. Arch Intern Med 166: 1256–1261. Coutu, D.L., Wu, J.H., Monette, A., et al. 2008. Periostin, a member of a novel family of vitamin K-dependent proteins, is expressed by mesenchymal stromal cells. J Biol Chem 283: 17991–18001. Ducy, P., Desbois, C., Boyce, B., et al. 1996. Increased bone formation in osteocalcin-deficient mice. Nature 382: 448–452. Feskanich, D., Weber, P., Willett, W.C., et al. 1999. Vitamin K intake and hip fractures in women: A prospective study. Am J Clin Nutr 69: 74–79.

200

Diet, Nutrients, and Bone Health

Gast, G.C.M., de Roos, N.M., Sluijs, I, et al. 2009. A high menaquinone intake reduces the incidence of coronary heart disease. Nutr Metab Cardiovasc Dis 19: 504–510. Groenen-van Dooren, M.M.C.L., Soute, B.A.M., Jie, K-S.G., et al. 1993. The relative effects of phylloquinone and menaquinone-4 on the blood coagulation factor synthesis in vitamin K-deficient rats. Biochem Pharmacol 46: 433–437. Hart, J.P., Catterall, A., Dodds, R.A., et al. 1984. Circulating vitamin K1 levels in fractured neck of femur. Lancet 324 (8397): 283. Hauschka, P.V., and Reid, M.L. 1978. Vitamin K dependence of a calcium-binding protein containing gammacarboxyglutamic acid in chicken bone. J Biol Chem 253: 9063–9068. Hemker, H.C., Al Dieri, R., De Smedt, E., et al. 2006. Thrombin generation, a function test of the haemostatic– thrombotic system. Thromb Haemost 96: 553–561. Inoue, T., Fujita, K., Kishimoto, H., et al. 2009. Randomized controlled study on the prevention of osteoporotic fractures (OF study): A phase IV clinical study of 15-mg menatetrenone capsules. J Bone Miner Metab 27: 66–75. Institute of Medicine. 2001. Vitamin K. In: Dietary Reference Intakes for Vitamin A, Vitamin K, Arsenic, Boron, Chromium, Copper, Iodine, Iron, Manganese, Molybdenum, Nickel, Silicon, Vanadium, and Zinc. National Academy Press, Washington, DC. Iwamoto, J., Sato, Y., Takeda, T., et al. 2009. High-dose vitamin K supplementation reduces fracture incidence in postmenopausal women: A review of the literature. Nutr Res 29: 221–228. Kalkwarf, H.J., Khoury, J.C., Bean, J., et al. 2004. Vitamin K, bone turnover, and bone mass in girls. Am J Clin Nutr 80: 1075–1080. Katagiri, M., Hakeda, Y., Chikazu, D., et al. 2001. Mechanism of stimulation of osteoclastic bone resorption through Gas6/Tyro 3, a receptor tyrosine kinase signaling, in mouse osteoclasts. J Biol Chem 276: 7376–7382. Knapen, M.H.J., Schurgers, L.J., and Vermeer, C. 2007. Vitamin K2 supplementation improves bone geometry and bone strength indices in postmenopausal women. Osteoporos Int 18: 963–972. Lee, N.K., Sowa, H., Hinoi, E., et al. 2007. Endocrine regulation of energy metabolism by the skeleton. Cell 130: 456–469. Luukinen, H., Käkönen, S.-M., Petterson, K., et al. 2000. Strong prediction of fractures among older adults by the ratio of carboxylated to total serum osteocalcin. J Bone Miner Res 15: 2473–2478. Maillard, C., Berruyer, M., Serre, C.M., et al. 1992. Protein S, a vitamin K-dependent protein, is a bone matrix component synthesized and secreted by osteoblasts. Endocrinology 130: 1599–1604. McCann, J.C., and Ames, B.N. 2009. Vitamin K, an example of triage theory: Is micronutrient inadequacy linked to diseases of aging? Am J Clin Nutr 90: 889–907. Nimptsch, K., Rohrmann, S., and Linseisen, J. 2008. Dietary intake of vitamin K and risk of prostate cancer in the Heidelberg cohort of the European Prospective Investigation into Cancer and Nutrition (EPICHeidelberg). Am J Clin Nutr 87: 985–992. Pan, E.Y., Gomperts, E.D., Millen, R., et al. 1990. Bone mineral density and its association with inherited protein S deficiency. Thromb Res 58: 221–231. Price, P.A., Urist, M.R., and Otawara, Y. 1983. Matrix Gla-protein, a new gammacarboxyglutamic acid-containing protein which is associated with the organic matrix of bone. Biochem Biophys Res Commun 117: 765–771. Prynne, C.J., Thane, C.W., Prentice, A., et al. 2005. Intake and sources of phylloquinone (vitamin K(1)) in 4-year-old British children: Comparison between 1950 and the 1990s. Publ Health Nutr 8: 171–180. Schurgers, L.J., Cranenburg, E.C., and Vermeer, C. 2008. Matrix Gla-protein: The calcification inhibitor in need of vitamin K. Thromb Haemost 100: 593–603. Schurgers, L.J., Shearer, M.J., Hamulyák, K., et al. 2004. Effect of vitamin K on the stability of oral anticoagulant treatment: Dose response relationship in healthy subjects. Blood 104: 2682–2689. Schurgers, L.J., Teunissen, K.J., Hamulyák, K., et al. 2007. Vitamin K-containing dietary supplements: Comparison of synthetic vitamin K1 and natto-derived menaquinone-7. Blood 109: 3279–3283. Schurgers, L.J., and Vermeer, C. 2000. Determination of phylloquinone and menaquinones in food: Effect of food matrix on circulating vitamin K concentrations. Haemostasis 30: 298–307. Schurgers, L.J., and Vermeer, C. 2002. Differential lipoprotein transport pathways of K-vitamins in healthy subjects. Biochim Biophys Acta 1570: 27–32. Shearer, M.J., and Bolton-Smith, C. 2000. The U.K. food data-base for vitamin K and why we need it. Food Chem 68: 213–218. Shearer, M.J., and Newman, P. 2008. Metabolism and cell biology of vitamin K. Thromb Haemost 100: 530–547.

Vitamin K and Bone

201

Shiraki, M., Shiraki, Y., Aoki, C., et al. 2000. Vitamin K2 (menatetrenone) effectively prevents fractures and sustains lumbar bone mineral density in osteoporosis. J Bone Miner Res 15: 515–521. Szulc, P., Chapuy, M.-C., Meunier, P.J., et al. 1996. Serum undercarboxylated osteocalcin is a marker of the risk of hip fracture: A three year follow-up study. Bone 18: 487–488. Van Summeren, M.J.H., Braam, L.A.J.L.M., Lilien, M.R., et al. 2009. The effect of menaquinone-7 (vitamin K2) supplementation on osteocalcin carboxylation in healthy prepubertal children. Br J Nutr 102: 1171–1178. Van Summeren, M.J.H., Braam, L.A.J.L.M., Noirt, F., et al. 2007. Pronounced elevation of undercarboxylated osteocalcin in healthy children. Pediatr Res 61: 366–370. Van Summeren, M.J.H., van Coeverden, S.C.C.M., Schurgers, L.J., et al. 2008. Vitamin K status is associated with childhood bone mineral content. Br J Nutr 100: 852–858. Viegas, C.S., Simes, D.C., Laizé, V., et al. 2008. Gla-rich protein (GRP), a new vitamin K-dependent protein identified from sturgeon cartilage and highly conserved in vertebrates. J Biol Chem 283: 36655–36664.

13

The Iron Factor in Bone Development Denis M. Medeiros and Erika Bono

CONTENTS Introduction..................................................................................................................................... 203 Theoretical Underpinnings..............................................................................................................204 Collagen Cross-Linking.........................................................................................................204 Role of Iron in Collagen Cross-Linking................................................................................204 Copper Deficiency and Bone..........................................................................................................205 Iron Deficiency and Bone...............................................................................................................206 Animal Studies.......................................................................................................................206 Human Studies.......................................................................................................................209 Summary......................................................................................................................................... 211 Acknowledgments........................................................................................................................... 211 References....................................................................................................................................... 211

INTRODUCTION Osteoporosis is a major public health threat for an estimated 44 million Americans or 55% of the people 50 years of age and older. In the United States today, 10 million individuals are estimated to already have the disease, and almost 34 million more are estimated to have low bone mass, placing them at increased risk for osteoporosis. In 2005, osteoporosis-related fractures were responsible for an estimated $19 billion in costs. By 2025, experts predict that these costs will rise to approximately $25.3 billion (National Osteoporosis Foundation, 2008). Calcium and vitamin D are crucial for maintaining bone health and preventing diseases of aging such as osteoporosis. Deficiency of these nutrients increases the risk for developing this disease. These two nutrients are critical in promoting bone growth and development in the formative years. Other nutrients have also received attention because of their roles in bone health. For instance, the balance between dietary phosphorus and calcium is important for bone. Also, excess dietary intakes of protein and sodium may also have negative effects upon bone. Other micronutrients, such as magnesium, ascorbic acid, selenium, copper, vitamin A, boron, zinc, and vitamin K, also impact upon bone, as has been demonstrated when one of these nutrients is either lacking or deficient in a diet (Anderson et al., 1998; Ilich and Kerstetter, 2000; Palacios, 2006). For instance, in animals, magnesium deficiency results in decreased bone strength and volume, poor bone development, and uncoupling of bone formation and resorption (Ilich and Kerstetter, 2000). Manganese supplementation along with calcium, copper, and zinc resulted in a greater gain in bone compared with supplementation with calcium alone (Ilich and Kerstetter, 2000). Another nutrient, iron, has been demonstrated to play a role in bone strength primarily because of experimental animal research. A few human studies also suggest an association between iron and bone health. Although iron itself is not as important as are calcium and vitamin D in promoting 203

204

Diet, Nutrients, and Bone Health

bone health, nevertheless, iron has an essential role in collagen formation, which directly impacts on bone integrity. Because iron deficiency is the second most common public health nutrition problem in the United States, it may have relevance to suboptimal bone development. Insufficient iron intake may foster the detrimental effects on bone health that result from suboptimal calcium and/ or vitamin D intakes.

THEORETICAL UNDERPINNINGS Collagen Cross-Linking Our group has investigated trace elements, such as copper and iron, as they relate to bone health. We initially observed that copper deficiency led to increased fragility in bone, an observation also reported by others (Rucker et al., 1969a, 1969b, 1975; Rucker, 1972; Opsahl et al., 1982; Jonas etal., 1993). A key role for copper exists as a component in a cupro-enzyme, lysyl oxidase, which is a rate-limiting enzyme in collagen strengthening. This enzyme catalyzes the production of aldehydes from lysine. Specifically, the epsilon amino group of lysine and hydroxylysine residues is converted to the reactive aldehydes, allysine and hydroxyallysine, respectively (Figure 13.1). These products then form an aldol cross-link, as shown in Figure 13.2. Collagen and elastin cross-linking are dependent upon this reaction, an important reaction from a biological perspective because it gives these connective tissue proteins their strength. Without active lysyl oxidase, tissues rich in collagen and elastin decline in strength, that is, reduced response to physical stress. The weakened connective tissues are one reason aneurysms and cardiac valve abnormalities occur in copper deficiency (Medeiros et al., 1991). With respect to bone, type I collagen is likely compromised in copper deficiency, and this leads to more fragile bones.

Role of Iron in Collagen Cross-Linking A major reason to investigate iron in bone tissue is that iron is required for lysine to be hydroxylated prior to reacting with lysyl oxidase. This hydroxylation step requires a unique reaction and enzyme, in this case, lysyl hydroxylase. Another similar enzyme also involved with collagen metabolism and thereby bone is prolyl hydroxylase, which catalyzes the formation of hydroxyproline from proline. In both reactions, ascorbate, molecular oxygen, α-ketoglutarate, and ferrous iron are required for this reaction (Figure 13.3). Both iron and ascorbate are required for several hydroxylase reactions (Switzer and Summer, 1972, 1973; Anderson et al., 1998). NH2

O

CH2 CH2

O2

CH Lysyl oxidase H2O CH2

CH2

CH2

CH2

CH2

–NH–CH–CO2

–NH–CH–CO

Lysine

Allysine

H2O2

NH3

FIGURE 13.1  Reaction of lysine and the copper containing enzyme lysyl oxidase in producing allysine prior to cross-linking reaction in collagen synthesis.

205

The Iron Factor in Bone Development

HN

NH H3N–CH2–CH2–(CH2)2–CH

HC–(CH2)2–CH2–CH2–NH3 O C

C O Lysine residues Lysyl oxidase

HN HC–(CH2)2–CH2–C O C

O

O

H

H

NH C–CH2–(CH2)2–CH C O

Aldehyde derivatives (allysine)

HN

H HC–(CH2)2–CH2–C

O C

NH C–(CH2)2–CH C O

C O

H

Aldol cross-link

FIGURE 13.2  Complete cross-linking reactions mediated by lysyl oxides for collagen development.

COPPER DEFICIENCY AND BONE To provide background for understanding the role of iron in bone development and remodeling, a review of our knowledge of copper and bone provides comparison to the role of iron. Much of our understanding of copper as it relates to bone comes from the studies of Rucker (1972), Rucker et al. (1969a, 1969b, 1975), Opsahl et al. (1982), and Jonas et al. (1993). In these studies, elastinand collagen-containing tissues, such as blood vessels, tendon, and bone, demonstrated decreased mechanical strength in copper deficiency that was associated with deficient collagen and elastin cross-linking. Torsional strength of bone in copper deficiency was noted to be decreased in chick tibia (Opsahl et al., 1982). Decreased activity of lysyl oxidase was also reported in this study. Jonas et al. (1993), who reported decreased torsional loading in femurs from copper-deficient rats, suggested that because the calcium content of the femurs did not differ by copper treatment, the difference in torsional strength was likely due to a decrease in collagen cross-linking. However, the issue of decreased bone mineralization in copper deficiency was reported earlier by Smith and colleagues (2002) who demonstrated lower bone mineral density (BMD), as measured by dual-energy x-ray absorptiometry (DXA) in the fifth lumbar vertebra and the proximal femur in copper-deficient rats. Furthermore, biochemical markers of bone formation, such as changes in blood alkaline

206

Diet, Nutrients, and Bone Health O

HN H

O–

C HC

CH2

C

C

H

H

O

C

O–

CH2

CH2

O2

HO HO

O

O

C C

O

C C

O

HC

CH2

H

C

HO

OH

O–

α-Ketoglutarate

Proline

CH

Ascorbic acid

Prolyl hydroxylase Fe2+

O

HN HO

O–

C HC

CH2

C

C

H

H

O

Hydroxyproline

O– HO

O

CH

O

HC

CH2 C

HO CH2

CH2

CO2

H

C

C O–

Succinate

O

O

C C O

Dehydroascorbic acid

FIGURE 13.3  A role for iron in collagen. Here proline is hydroxylated to hydroxyproline in the presence of an iron-dependent enzyme, prolyl hydroxylase, and ascorbic acid. α-Ketoglutarate is used as a substrate in the reaction with a succinate product, both of which are tricarboxylic acid cycle (TCA) intermediates.

p­ hosphatase activity, were not noted, which indirectly suggested an accelerated bone resorption as the reason for decreased BMD in the bones of copper-deficient rats.

IRON DEFICIENCY AND BONE Animal Studies We were the first laboratory to report a possible role for iron in bone integrity. Initially, our laboratory had been studying the role of copper and iron deficiency upon cardiac metabolism and cardiac hypertrophy. In the course of a usual study, we noted that the bones from iron-deficient rats appeared to be more fragile upon dissection of the carcass. A review of the literature suggested that iron could play a role in bone via hydroxylase coenzyme partner, as reviewed above. In our first study, we compared rats fed either control diets, iron-deficient diets, or copper-deficient diets from weaning until 5 weeks thereafter. Animals were sacrificed and assessed for iron stores and had their femurs removed and x-rays developed for morphometric analysis and then submitted to a threepoint bending test for strength analysis (Medeiros et al., 1997). We reported that in iron-deficient and copper-deficient rats, the breaking strength decreased in femurs. Iron and copper deficiency both resulted in smaller cortical and larger medullary areas in a part of the femur.

207

The Iron Factor in Bone Development

With this new information, more defined studies were planned to confirm and explain in greater detail the changes that occur in bone of rats that were fed iron-deficient diets. One issue that needed to be addressed appeared to be the relative perturbation of iron deficiency upon bone as compared with that of calcium restriction; the latter, of course, is a commonly accepted culprit in the genesis of osteoporosis. We conducted a study in which four groups of weanling male rats were fed diets either deficient in iron (5 to 8 mg Fe/kg diet), low in calcium (1.0 g calcium/kg diet), or deficient in both minerals as compared with a control diet that contained adequate amounts of both nutrients (Medeiros et al., 2002). Extra care was taken in preparing the diets as we wanted to rigorously control the dietary intake levels of both calcium and iron. Because of our concern about avoiding any iron contamination in this and other studies, we used 5% avicel as a source of fiber rather than cellulose, as cellulose has a rather high iron background content. For the low-calcium diets, we used chemically pure calcium phosphate dibasic and magnesium carbonate. Also, because of concern about the phosphate components of the diet, we elected to control the phosphorus level in the diet at 0.56%. This diet was very low in iron content (5 to 8 mg Fe/kg diet). Rats were fed the control or iron-deficient diets for 5 weeks. Results revealed that cortical widths were reduced in all experimental groups, with the calcium-restricted and the calcium/iron-deficient groups having the greatest reductions in cortical width. Total femur and tibia widths were decreased in all experimental groups. The iron-deficient group had an increase in the medullary width. Calcium restriction and iron deficiency either alone or in combination resulted in reduced BMD and cortical bone area. In this animal model, iron deficiency clearly had a negative impact upon bone health, and in combination with calcium restriction, even greater negative effects were observed. A fundamental issue in the study previously cited (Medeiros et al., 2002) was that the low iron content of the diet resulted in severe iron deficiency and multiple adverse skeletal and nonosseous effects. The animals had greatly reduced body weight, and thus the issue of food intake during iron deficiency had to be taken into consideration. Another study was designed with similar diets, and rats were allocated to one of four groups: control (ad libitum), calcium-restricted, iron-deficient, or a control group pair fed to the iron-deficient group (Medeiros et al., 2004). The pair-fed control group simply meant that the amount of food consumed by the iron-deficient group was determined daily and an equal amount of control diet was provided to the pair-fed rats. The results revealed that while the pair-fed rats did have small but greater decreases in whole body and femur BMD and bone mineral content (BMC) than those of the control group, both the calcium-restricted and the iron-deficient groups had much greater reductions in BMD and BMC at the same measurement sites than the pair-fed group (Table 13.1). These results suggested that the impact of iron deficiency was independent of low caloric intake and a reduced body weight resulting from iron deficiency. In this same study, we examined the lumbar region of the animals as well as the microarchitecture of lumbar vertebrae. The third lumbar vertebra (L3) revealed decreases in bone volume and total bone volume, decreased trabecular number TABLE 13.1 Whole Body and Femur BMD of Rats That Were Fed Control, CalciumDeficient, Iron-Deficient, or Pair-Fed Diets1,2 (Mean ± SEM) Variable Whole-body BMD, g/cm   3 weeks   5 weeks Femur BMD, g/cm2 1 2

Control

Ca−

Fe−

Pair-Fed

2

0.122 ± 0.0025a 0.142 ± 0.0018a 0.183 ±.0042a

0.093 ± 0.0023b 0.091 ± 0.0014d 0.096 ± 0.0015d

0.119 ± 0.0034a 0.129 ± 0.0019c 0.161 ± 0.0030c

0.117 ± 0.0031a 0.135 ± 0.0012b 0.173 ± 0.0021b

Pair-fed group was fed the control diet in an amount consumed by the iron-deficient group (Fe−). Means in a row without a common superscript differ, p ≤ .05.

208

Diet, Nutrients, and Bone Health

and thickness, a lower structural model index, and increased trabecular separation in each of the nutrient-deficient groups compared with the control and pair-fed groups. Finite element analysis was used to further analyze the data. This analytic method uses a computer simulation technique, often employed in engineering and other physical sciences, to predict how materials and structures will respond when loads are placed upon them. Finite element analysis revealed that a lower force was needed to compress the vertebrae that had lower stiffness, but that a greater von Mises stress was needed in the calcium-restricted and iron-deficient groups compared with both the control and pairfed groups. Von Mises stress is the internal stress in the bone when a constant force is applied. These findings mean that greater internal stresses existed in the lumbar vertebrae of the calcium-restricted and iron-deficient groups. Urine concentrations of deoxypyridinium cross-links, serum osteocalcin, and 1, 25-dihydroxycholecalciferol were increased in calcium-restricted rats compared with the other three groups. Iron deficiency apparently did not affect these measures. The lack of change in serum 1,25-dihydroxycholecalciferol concentrations were unexpected because it had been reported that 1α-hydroxylase involves a reaction that is dependent upon a flavoprotein, that is, an iron-sulfur protein, and cytochrome P-450, also iron containing (DeLuca, 1976). Regardless, this study provided further evidence that iron deficiency had a detrimental impact upon the microarchitecture of L3 and other vertebrae and that the bone effects were independent of the amount of food ingested. The final study from our laboratory sought to answer the question as to whether marginal iron intake, as opposed to a very severe iron deficiency (as used in the above studies), would produce similar adverse effects on bone (Parelman et al., 2006). Our goal was to address more specifically the likelihood of iron deficiency in humans might be a factor affecting bone health. This investigation, therefore, evaluated marginal iron and calcium intakes upon the same bone parameters as in our previous study (Medeiros et al., 2004). We assigned 32 male weanling rats to one of four diets: control (ad libitum), marginal calcium in the amount of 2.5 g Ca/kg diet (compared with 1.0 g Ca/kg diet in previous studies), marginal iron in the amount of 12 mg Fe/kg diet (compared with 5 to 8 mg Fe/kg diet in previous studies), or a diet marginal in both calcium and iron. Because the diets were higher in calcium and iron, we elected to feed these diets to animals for 10 weeks instead of the usual 5 weeks. As would be expected, rats fed the marginal iron diet did have lower hematocrit readings compared with those of iron-adequate rats. Using similar measures as reported above, the whole-body BMD was lower in the marginal calcium group, and whole-body BMC was lower in the marginal iron group when compared with the controls (Medeiros et al., 2004). Marginal calcium or marginal iron intake resulted in decreased BMD of the femur. There was a trend (p = .06) for the doubly deficient group to have lower BMD than those of the other three treatment groups. Furthermore, the most revealing findings concerned the L4 vertebrae, in that BMD was lower in all three experimental groups compared with the control group. The overall BMC was lower in both the marginal iron and the marginal calcium groups than in the groups with adequate iron and the control rats. Microcomputed tomography analysis revealed that marginal iron L4 and marginal calcium L4 had reduced connectivity of the trabeculae. Moreover, in the marginal iron rats, trabecular number was decreased and trabecular separation was increased, which results in enhanced porosity (Figure 13.4). Finite element analysis suggested that the marginal iron group was less likely to withstand compression force and could break at lower external stress than was the control group. Control

Iron-restricted

FIGURE 13.4  Example of trabecular bone of L 4 in rats fed either a control diet containing adequate levels of iron or a diet marginal in iron (12 mg Fe/kg diet). The arrow denotes increased porosity in the vertebrae for a rat fed an iron-restricted diet.

The Iron Factor in Bone Development

209

A follow-up study of iron chelation in hFOB osteoblast cells, that is, iron deficiency induced by an iron chelator, was designed to test whether iron deficiency in bone-forming cells influences type I collagen formation (Parelman et al., 2006). No differences in type I collagen levels were found between iron-deficient cells and control cells. Interestingly, we did observe decreased mineralization associated with iron chelation treatment. These in vitro results supported, at least in part, the in vivo studies, which demonstrated decreased mineralization in rats consuming an iron-restricted diet. McClung et al. (2008) did not report any differences in bone-breaking threshold as a result of marginal iron intake. However, a major difference in their study compared with ours (Parelman et al., 2006) was that their rats began the iron-deficient diets at 10 weeks of age. Although they were fed these diets for 12 weeks, this was an insufficient period of time for animals to develop irondeficiency anemia. Presumably, the animals in this report had stored up sufficient iron reserves by the time the study started. In contrast, our study provided an iron-deficient diet starting at weaning (when limited iron accretion by the liver occurs in comparison with an older rat), and all the rats developed iron-deficiency anemia. Another critical issue may be the specific period in the life cycle that the limited intake of iron is fed. Dietary iron deficiency might be more detrimental to the bones of rapidly growing rats after weaning than later in life. Other laboratories have been able to corroborate our finding of the negative impact of a postweaning iron-deficient diet upon bone development (Katsumata et al., 2006, 2009). For example, Katsumata et al. (2009) investigated whether iron deficiency would decrease BMD in weanling male rats. Two groups of six rats each were fed a control (ad libitum) or an iron-deficient diet, and a third group, similar to our previous study (Medeiros et al., 2004), was pair-fed a control diet in an amount consumed by the iron-deficient rats. The diet was the AIN-93G diet, a modification of the AIN-77 diet previously used by our laboratory. The rats were allowed to consume their respective diets for 4 weeks. BMC and BMD were measured, and bone histomorphometry was assessed. The iron-deficient rats did develop both anemia and greater heart weights, common signs of severe dietary iron deficiency. Supporting our previous studies, BMC and BMD measurements were significantly lower in the femurs of the iron-deficient groups compared with the control and pair-fed groups. Lower bone-volume-to-total-bone-volume ratio in the iron-deficient group was also demonstrated. Whereas trabecular thickness did not differ among the three groups, trabecular number was significantly decreased in both iron-deficient groups, and trabecular separation was higher as well. These findings are consistent with increased porosity in iron-deficient rats and are in agreement with our previous studies (Medeiros et al., 2004; Parelman et al., 2006). These investigators also assessed osteoid volume, osteoid surface, and osteoid thickness. All of these parameters were reduced in the iron-deficient groups compared with those of controls. Mineralizing surface, mineral apposition rate, bone formation rate, and adjusted apposition rate were also significantly lower in the iron-deficient group. The percentage of bone surface occupied by osteoclasts was decreased in the iron-deficient group as was osteoclast number. Biochemical measures, such as blood osteocalcin concentrations and urinary deoxypyridinoline levels, were lower in the iron-deficient group, whereas the C-terminal telopeptide of type I collagen was higher in the iron-deficient group. The lower osteocalcin levels suggested lower bone formation in the iron-deficient group. The reductions of deoxypyridinoline and C-terminal telopeptide of type I collagen suggested that bone resorption is decreased in iron deficiency. We did not find changes of osteocalcin and urinary deoxypyridinoline in our study, nor did we report differences in 1,25-dihydroxycholecalciferol, as these authors did (Medeiros et al., 2004). McClung and coinvestigators (2008) pointed out that the degree of anemia, a major determinant of the results, in their study was much greater than in ours.

Human Studies No conclusive data exist that dietary iron or adequate iron status may be beneficial in humans as it relates to bone integrity and BMD, but a few studies suggest the benefit of sufficient iron intake for

210

Diet, Nutrients, and Bone Health

skeletal health. The first study published by Harris and colleagues (2003) assessed 242 women 40 to 66 years of age for a variety of bone-related variables as part of the Bone, Estrogen, and Strength Training (BEST) Study. This randomized clinical trial examined the impact of exercise on BMD in early postmenopausal women. As part of this study, the investigators had access to 3-day dietary records and BMD measurements of the lumbar spine at L2 to L4, greater trochanter, femur neck, Ward’s triangle, and whole body (Figure 13.5). Using multiple regression techniques and adjustments for pertinent confounding variables, dietary iron intake was associated with greater BMD at all skeletal sites. This relationship was significant even after adjustment for dietary calcium and protein. Specifically, subjects who consumed greater than 20 mg of dietary iron per day and had a calcium intake between 800 and 1200 mg per day had the most significant increases in BMD. A second investigation by the same group using subjects from the BEST study (Maurer et al., 2005) provided additional support to the idea that iron may be related to bone mass and density. Here, hormone replacement therapy was considered in a 1-year longitudinal study of 116 women who received hormone replacement and 112 women who did not receive replacement therapy. This study used 8-day dietary records and had the same BMD measurements as in the previous report (Harris et al., 2003). Iron intake was shown to be positively related to BMD in the greater trochanter and Ward’s triangle only in women receiving hormone replacement therapy. For those women in the lowest tertile of calcium intake, femur neck BMD increased as iron intake increased. A study of a British group of 32 women aged 46 to 55 years who were not on hormone replacement therapy suggested an independent positive association between dietary iron and BMD (Abraham et al., 2006). The women were followed for a period of 11 to 14 years and had periodic weighed-food intakes and BMD measurements at L2 to L4 of the vertebral column. Again, dietary iron intake was positively related to BMD, even after adjusting for calcium and protein intakes. A larger crosssectional study of the same subjects enrolled in this study revealed that, among 244 females within the same age range, a positive association between dietary iron and BMD was found at all bone sites (Farrell et al., 2009). Perhaps the ultimate test short of a randomized clinical trial with iron supplementation is illustrated by a study conducted by Moran et al. (2008), in which young Israeli soldiers in basic training were evaluated for risk factors that might predict stress fractures. The study included 227 females and 83 males. None of the males developed stress fractures during basic training, but 27 females did develop stress fractures during the 4-month period of basic training. Females with iron deficiency Ward’s triangle

Greater trochanter

Femoral neck

Femur

FIGURE 13.5  Diagram of a femur in which DXA measures are often made for bone mineral density measurements. Note that the greater trochanter and an area on the femur neck called Ward’s triangle are often measured in bone studies.

The Iron Factor in Bone Development

211

had a greater risk of developing stress fractures. These results are intriguing and should be further investigated because of relevance to other military units as well as to athletes who may be consuming an iron-poor diet.

SUMMARY Experimental studies have revealed that dietary factors other than protein, calcium, and vitamin D, that is, other micronutrients, may have a role in skeletal development and maintenance. Magnesium, ascorbic acid, selenium, vitamin A, boron, zinc, and vitamin K are some of the most common nutrients known to have roles in bone growth and maintenance. Also, copper is known to promote bone integrity as a cofactor for the enzyme lysyl oxidase, which promotes cross-linking among collagen fibrils, thereby enhancing strength in bone as well as in other collagen-containing tissue. Iron has been shown to play a role in supporting bone strength in both animal and human studies. From a theoretical perspective, two iron-dependent enzymes act in support of collagen cross-linking: lysyl hydroxylase and prolyl hydroxylase. Experimentally, data from animal models indicate that bone mineralization appears to be compromised as a result of either severe iron deficiency or marginal iron deficiency. Human and animal studies are consistent in finding that bone mineral density and content, as measured by DXA, are decreased with lower iron intake. Stress fractures may occur in young women athletes and military recruits with iron deficiency. One important study revealed an increased incidence of stress fractures among iron-deficient female military recruits compared with none in ironreplete female military recruits in basic training. These findings have implications from a practical perspective because iron insufficiency, and less commonly, iron-deficiency anemia, is a common U.S. and worldwide problem. A precise mechanism regarding the linkage between iron deficiency and poor bone development and bone loss in young adults has not yet been determined. Randomized clinically controlled trials of BMD in females of various ages, with or without iron supplementation, appear to be warranted.

ACKNOWLEDGMENTS This was supported in part by K-State Research and Extension Multistate project number W2002; “Nutrient Bioavailabilty—Phytonutrients and Beyond.”

REFERENCES Abraham, R., Walton, J., Russell, L., et al. 2006. Dietary determinants of post-menopausal bone loss at the lumbar spine: A possible beneficial effect of iron. Osteoporos Int 17: 1165–1173. Anderson, J.J.B., Stender, M., Rondano, P., et al. 1998. In Nutrition in Exercise and Sport, I. Wolinsky, ed., 219–244. CRC Press, Boca Raton, FL. DeLuca, H.F. 1976. Metabolism of vitamin D: Current status. Am J Clin Nutr 29: 1258–1270. Farrell, V., Harris, M., Lohman, T.G., et al. 2009. Comparison between dietary assessment methods for determining associations between nutrient intakes and bone mineral density in postmenopausal women. J Am Diet Assoc 109: 899–904. Harris, M.M., Houtkooper, L.B., Stanford, V.A., et al. 2003. Dietary iron is associated with bone mineral density in healthy postmenopausal women. J Nutr 133: 3598–3602. Ilich, J.Z., and Kerstetter, J.E. 2000. Nutrition in bone health revisited: A story beyond calcium. J Am Coll Nutr 19: 715–737. Jonas, J., Burns, J., Abel, E.W., et al. 1993. Impaired mechanical strength of bone in experimental copper deficiency. Ann Nutr Metab 37: 245–252. Katsumata, S., Katsumata-Tsuboi, R., Uehara, M., et al. 2009. Severe iron deficiency decreases both bone formation and bone resorption in rats. J Nutr 139: 238–243. Katsumata, S., Tsuboi, R., Uehara, M., et al. 2006. Dietary iron deficiency decreases serum osteocalcin concentration and bone mineral density in rats. Biosci Biotechnol Biochem 70: 2547–2550.

212

Diet, Nutrients, and Bone Health

Maurer, J., Harris, M.M., Stanford, V.A., et al. 2005. Dietary iron positively influences bone mineral density in postmenopausal women on hormone replacement therapy. J Nutr 135: 863–869. McClung, J.P., Andersen, N.E., Tarr, T.N., et al. 2008. Physical activity prevents augmented body fat accretion in moderately iron-deficient rats. J Nutr 138: 1293–1297. Medeiros, D.M., Bagby, D., and Ovecka, G. 1991. Myofibrillar, mitochondrial and valvular morphological alterations in cardiac hypertrophy among copper deficient rats. J Nutr 121: 815–824. Medeiros, D.M., Ilich, J., Ireton, J., et al. 1997. Femur from rats fed diets deficient in copper or iron have decreased mechanical strength and altered mineral composition. J Trace Elem Exp Med 10: 197–203. Medeiros, D.M., Plattner, A., Jennings, D., et al. 2002. Bone morphology, strength and density are compromised in iron-deficient rats and exacerbated by calcium restriction. J Nutr 132: 3135–3141. Medeiros, D.M., Stoecker, B., Plattner, A., et al. 2004. Iron deficiency negatively affects vertebrae and femurs of rats independently of energy intake and body weight. J Nutr 134: 3061–3067. Moran, D.S., Israeli, E., Evans, R.K., et al. 2008. Prediction model for stress fracture in young female recruits during basic training. Med Sci Sports Exerc 40: S636–S644. National Osteoporosis Foundation. 2008. Fast Facts on Osteoporosis. http://www.nof.org/osteoporosis/diseasefacts.htm (accessed October 29, 2009). Opsahl, W., Zeronian, H., Ellison, M., et al. 1982. Role of copper in collagen cross-linking and its influence on selected mechanical properties of chick bone and tendon. J Nutr 112: 708–716. Palacios, C. 2006. The role of nutrients in bone health, from A to Z. Crit Rev Food Sci Nutr 46: 621–628. Parelman, M., Stoecker, B., Baker, A., et al. 2006. Iron restriction negatively affects bone in female rats and mineralization of hFOB osteoblast cells. Exp Biol Med 231: 378–386. Rucker, R.B. 1972. The effect of copper deficiency on bone strength and the metabolism of aortic elastin. In Trace Substances in Environmental Health VI. R.S. Riggins ed., 153–157. University of Missouri Press, Columbia, MO. Rucker, R.B., Parker, H.E., and Rogler, J.C. 1969a. Effect of copper deficiency on chick bone collagen and selected bone enzymes. J Nutr 98: 57–63. Rucker, R.B., Parker, H.E., and Rogler, J.C. 1969b. The effects of copper on collagen cross-linking. Biochem Biophys Res Comm 34: 28–32. Rucker, R.B., Riggins, R.S., Laughlin, R., et al. 1975. Effects of nutritional copper deficiency on the biomechanical properties of bone and arterial elastin metabolism in the chick. J Nutr 105: 1062–1070. Smith, B.J., King, J.B., Lucas, E.A., et al. 2002. Skeletal unloading and dietary copper depletion are detrimental to bone quality of mature rats. J Nutr 132: 190–196. Switzer, B.R., and Summer, G.K. 1972. Collagen synthesis in human fibroblasts: Effects of ascorbate, α-ketoglutarate and ferrous ion on proline hydroxylation. J Nutr 102: 721–728. Switzer, B.R., and Summer, G.K. 1973. Inhibition of collagen synthesis by α, α′-dipyridyl in human skin fibroblasts in culture. In Vitro 9: 160–166.

14

Micronutrients and Bone Elizabeth Grubert and Jeri W. Nieves

CONTENTS Introduction..................................................................................................................................... 213 Sodium............................................................................................................................................ 213 Vitamin C........................................................................................................................................ 214 Magnesium...................................................................................................................................... 215 Fluoride........................................................................................................................................... 216 B Vitamins....................................................................................................................................... 217 Other Nutrients................................................................................................................................ 218 Boron.......................................................................................................................................... 218 Copper........................................................................................................................................ 218 Zinc............................................................................................................................................ 218 Silicon........................................................................................................................................ 218 Combinations of Trace Minerals..................................................................................................... 219 Conclusions..................................................................................................................................... 219 References....................................................................................................................................... 219

INTRODUCTION The most recent definition of osteoporosis is a disease characterized by loss of bone mass, accompanied by microarchitectural deterioration of bone tissue, that leads to an unacceptable increase in the risk of skeletal failure (fracture). Osteoporosis and low bone mass are currently estimated to be a major public health threat for almost 44 million U.S. men and women aged 50 years and older, or 55% of the population in that age range (America’s Bone Health: The State of Osteoporosis and Low Bone Mass in Our Nation, 2002). In fact, one in two women and one in four men over the age of 50years will fracture at some point in their lifetime. The costs to the healthcare system associated with osteoporotic fracture are approximately 17 billion dollars annually (Gabriel et al., 2002; Ray et al., 1997; Tosteson et al., 2001), with each hip fracture having total medical costs of $40,000. Adequate nutrition plays a major role in the prevention and treatment of osteoporosis; the nutrients of greatest importance are calcium and vitamin D, and this will be discussed in Chapters 8 and 10. In this chapter, we cover various micronutrients including sodium, vitamin C, magnesium, fluoride, boron, silicon, copper, zinc, and the B vitamins. It is increasingly clear that exposures to a complex of nutrients and food constituents interact to affect bone status. In addition to identifying the role of individual components, a great need exists to understand the interactions of these factors within diets to make effective recommendations for prevention of bone loss and osteoporosis in the aging population (Tucker, 2003).

SODIUM Sodium may lead to an increase in renal calcium excretion. The mean urinary calcium loss is 1mmol/100 mmol sodium for males and females at all ages (Teucher and Fairweather-Tait, 2003). 213

214

Diet, Nutrients, and Bone Health

If the amount of calcium consumed and absorbed is less than the amount needed to offset these obligatory urinary calcium losses associated with sodium intake, then bone mass will be negatively impacted. In observational studies, higher salt intakes can lead to higher levels of circulating parathyroid hormone and greater rates of bone resorption in postmenopausal women and men (Harrington and Cashman, 2003; Jones et al., 1997). Furthermore, those with low-calcium and high-salt diets may have lower bone mineral density (BMD), although this is not always the case (Devine et al., 1995; Harrington and Cashman, 2003; Massey and Whiting, 1996; Mizushima et al., 1999). The effect of dietary sodium on calcium retention and the influence of race were studied in 35 adolescent girls (22 black and 13 white) who participated in two 20-day metabolic summer camps. At the end of the intervention, both race and sodium intake significantly correlated with calcium retention (p < .01). The high-sodium diet (3.96 g/day) significantly reduced calcium retention in both whites and blacks (p < .01), primarily through a decrease in net calcium absorption. However, black girls excreted significantly less calcium in the urine than did white girls, regardless of diet (p < .05) (Wigertz et al., 2005). The relationship between sodium intake, sodium and calcium excretion, and BMD of the total hip was measured in 50 Caucasian and 39 African American postmenopausal women. Sodium excretion was found to be a significant predictor of calcium excretion in both postmenopausal African American and Caucasian postmenopausal women. The relationship between sodium and calcium excretion is modulated by calcium intake, and the relationship is strongest at low calcium intakes (≤1000 mg/day). However, sodium excretion was not a significant predictor of total hip BMD in elderly African American and Caucasian postmenopausal women (Carbone et al., 2003). In postmenopausal women who had baseline sodium excretions equal to or greater than the average sodium intake in the United States (≥3.4 g/day), a low-sodium diet (2 g/day) for 6 months resulted in significant decreases in sodium excretion (p = .01), in calcium excretion (p = .01), and in bone turnover (p = .04). However, no significant changes were found in intact PTH (p = .97) or 1,25 dihydroxyvitamin D (p = .49) with the low-sodium diet. These findings suggest that in postmenopausal women with sodium intakes 3.4 g/day or more, a low sodium diet may have benefits for skeletal health (Carbone et al., 2005). In a study of 1098 men and women over the age of 65 years, sodium intake, reflected by urinary-sodium-to-creatinine ratio (Na/Cr), was the major factor linking blood pressure and osteoporosis, as shown by the inverse relationship with BMD. These findings lend further emphasis to the health benefits of salt reduction in our population both in terms of hypertension and osteoporosis (Woo et al., 2009). Further support of the role of salt in bone health comes from the finding that the Dietary Approaches to Stop Hypertension diet was shown to reduce bone turnover (Lin et al., 2003). Results of cross-sectional and prospective investigations on high salt intake and long-term bone health are inconclusive. In general, markers of bone resorption do relate to sodium intake, but BMD does not typically relate to sodium intake. Clearly, increased sodium intake will lead to increased renal calcium excretion. This relationship may be modified by protein or potassium intake as well as by genotype or salt sensitivity (Harrington and Cashman, 2003; Harrington et al., 2004a, 2004b). In addition, adequate intake of calcium may allow a more liberal use of sodium in the diet. The anion is also important, with sodium chloride increasing urinary calcium more than other salts such as sodium bicarbonate or sodium acetate (Carbone et al., 2003; Frassetto et al., 2001). In general, sodium intake will not be a problem in the face of adequate calcium intake (Carbone et al., 2003) or potassium intake (Harrington and Cashman, 2003). Furthermore, if American Heart Association guidelines are followed (1500 mg sodium/day), there should be no negative impact of sodium on bone health. However, further study of the impact of sodium consumption on peak bone mass is warranted.

VITAMIN C Vitamin C is an essential cofactor for collagen formation and synthesis of hydroxyproline and hydroxylysine (see Chapter 13). Vitamin C may also reduce oxidative stress. Rich dietary sources

Micronutrients and Bone

215

of vitamin C include citrus fruit and juices, peppers, broccoli and tomato products, and green leafy vegetables. The dietary reference intakes for vitamin C are 75 mg/day for adult women and 90 mg/ day for adult men. In an observational study of 533 women, results suggest that antioxidant vitamin E or C supplements may suppress bone resorption in nonsmoking postmenopausal women (Pasco et al., 2006). Other epidemiological studies have shown a positive association between vitamin C and bone mass, although the results are not always significant. In these studies, low intakes of vitamin C are associated with a faster rate of bone loss in women of varying ages, and one study found that higher vitamin C was associated with fewer fractures; however, randomized clinical trials are lacking (Freudenheim et al., 1986; Hall & Greendale, 1998; Hernandez-Avila et al., 1993; Kaptoge et al., 2003; Leveille et al., 1997; Macdonald et al., 2004; New et al., 1997; Odland et al., 1972; Prynne et al., 2006; Sowers et al., 1985; Weber, 1999). After adjustment for important BMD-related covariates, increasing intakes of antioxidants (including vitamin C) were not independently associated with BMD, but a significant interaction between intake of vitamin C and hormone therapy on BMD was found in a cohort from the Women’s Health Initiative Observational Study and Clinical Trial (Wolf et al., 2005). Tucker et al. (2002) evaluated the bone density of 213 men and 393 women, with the average age among them of 75 years, at the start in the Framingham Osteoporosis study, over a 4-year period to determine if any association existed between vitamin C intake and bone health. Using a diet questionnaire given to participants at baseline and again 4 years later, the researchers evaluated the change in bone density in the hip, spine, and arm over the follow-up period. Vitamin C and vitamin E intake were of primary interest, but whether participants smoked or were on hormone replacement therapy was taken into account. Men in the highest vitamin C intake (>300 mg/day; >3 times the recommended intake in men) had less bone loss than that of men in the lowest group of vitamin C intake (106 mg). A similar finding in women was not significant, and it was shown that vitamin C may interact with estrogen use, calcium, and vitamin E (Sahni et al., 2008). Because it is not possible to separate the effects of vitamin C supplements from vitamin C in fruits and vegetables in some studies (Macdonald et al., 2004; Sahni et al., 2008), the recommendation is to obtain the needed amounts of vitamin C only from fruits and vegetables for s`keletal health rather than from supplements. In fact, higher fruit and vegetable intakes may have positive effects on bone mineral status in both younger and older age groups so that this recommendation should cover the lifespan (Prynne et al., 2006). Recommended intakes of five or more servings of fruits and vegetables per day should supply enough vitamin C for bone health. Future studies of vitamin C and bone health need to take into account gender, estrogen use, and intake of other nutrients to assess any potential interactions.

MAGNESIUM Magnesium, when complexed with adenosine-5′-triphosphate, takes part in many intracellular enzyme reactions, including the synthesis of proteins and nucleic acid. Magnesium is an electrolyte mineral that contributes to alkalinity and is important in acid base balance. The current intakes recommended for healthy adult males are 420 mg and those for women are 320 mg/day. Because magnesium is present in most foods, particularly legumes, vegetables, nuts, seeds, fruits, grains, fish, and dairy, severe magnesium deficiency is, therefore, rarely seen in healthy people. However, many intakes in the United States fall below the recommended intake levels, as the mean magnesium intakes for males and females are 323 and 228 mg/day, respectively. Magnesium deficiency can also occur if concomitant disorders exist and/or medications place individuals at further risk for magnesium depletion (Rude and Gruber, 2004; Rude et al., 2004). Magnesium deficiency can be detected with biochemical measurements, that is, low serum magnesium, low serum calcium, and resistance to vitamin D, or via clinical symptoms, that is, muscle twitching, muscle cramps, high blood pressure, and irregular heartbeat. A small metabolic study showed that consuming a diet

216

Diet, Nutrients, and Bone Health

d­ eficient in magnesium, which resulted in negative magnesium balance, may affect calcium, potassium, and cholesterol metabolism (Nielsen, 2004). In preadolescent girls, a positive relationship between ultrasound determination of bone mass of the calcaneus and dietary magnesium intake was found (Wang et al., 1999), suggesting that this mineral was important in skeletal growth and development. Magnesium in serum and hair was associated with greater BMD in premenopausal women, and the ratio of serum calcium to magnesium was a significant indicator of BMD (Song et al., 2007). The intake of fruits and vegetables, containing vitamin C, magnesium, and potassium, may protect against premenopausal bone loss, but magnesium alone may not be protective against declines in BMC and BMD (Macdonald et al., 2004). In other studies of premenopausal women, magnesium intake was related to lumbar spine BMD in a cross-sectional evaluation (New et al., 2000), and a significant relationship was found over a 1-year period between dietary magnesium intake and rate of change of BMD of the lumbar spine and total body calcium (Houtkooper et al., 1995). A study of postmenopausal women showed a positive correlation between BMD of the forearm and magnesium intake (Tranquilli et al., 1994). Several small epidemiological studies have found that higher magnesium intakes are associated with higher BMD in elderly men and women, although rates of bone loss over 4 years were only related to magnesium intake in men (Tucker et al., 1999). In 2038 older black and white men and women (aged 70 to 79years) enrolled in the Health, Aging and Body Composition Study, a greater magnesium intake was significantly related to higher BMD in white women and men, but not in black women and men, which may have been related to differences in dietary assessment (Ryder et al., 2005). Only a few small controlled clinical trials of magnesium supplementation on bone have been conducted (Nielsen, 1990; Stendig-Lindberg et al., 1993), and they were primarily effective in ­magnesium-depleted subjects. Overall, observational and clinical trial data concerning magnesium and BMD or fractures are inconclusive, and in fact, one recent study from the Women’s Health Initiative surprisingly reported that higher intakes of magnesium were associated with a higher risk of wrist fracture (Durlach et al., 1998; Jackson et al., 2003; Nielsen, 1990; Rude and Olerich, 1996; Stendig-Lindberg et al., 1993; Tucker et al., 1999). Little evidence has been reported that magnesium is needed to prevent osteoporosis in the general population. Results relating magnesium to BMD are often confounded by coexisting limited intakes of other nutrients that are important for skeletal health. Clearly, a magnesium supplement may be required in frail elderly with poor diets (Durlach et al., 1998) and in persons with intestinal disease, alcoholics, or persons on treatment with diuretics or chemotherapy that depletes magnesium. In addition, as calcium supplements sometimes result in constipation, a supplement with magnesium might be useful in helping to keep bowel habits regular.

FLUORIDE Fluoride is an essential trace element that is required for skeletal and dental development. The adequate daily adult intake is 4 mg for males and 3 mg for females. The concentration of fluoride in the soil, water, and many foods varies by geographic region. Major dietary sources include drinking water, tea, coffee, rice, soybeans, spinach, onions, and lettuce. Fluoride interacts with mineralized tissues in a number of ways. At low doses, the fluoride may be passively incorporated into the mineral, stabilizing mineral surfaces against dissolution. At higher doses, such as those used previously for treatment of osteoporosis, the fluoride may alter the amount and structure of tissue present, including altering the interface between the collagen and mineral and causing a painful condition associated with extraosseous calcification and brittle bones. At very high doses, skeletal fluorosis may occur, characterized by debilitating changes in the skeleton; this can even occur in communities where the local drinking water has naturally high fluoride levels, much greater than 1 ppm in fluoridated waters (Chachra et al., 2008). High doses of fluoride can stimulate osteoblasts; however, the quality of bone that is formed may be abnormal and the effect on fracture rates is unclear (Riggs, 1993).

Micronutrients and Bone

217

The impact of more typical intakes of fluoride on the skeleton was evaluated in the Iowa Fluoride Study/Iowa Bone Development Study. Longitudinal fluoride intake at levels of intake typical in the United States (mean 0.68 mg per day) is only weakly associated with BMC or BMD in boys and girls at age 11 years. Additional research is warranted to better understand possible gender- and agespecific effects of fluoride intake on bone development (Levy et al., 2009). No benefit of adding fluoride supplements to an adult diet exists for skeletal health. The lower doses of fluoride typically found in drinking water have no effect on bone density or on reducing fractures (Cauley et al., 1995; Kurttio et al., 1999; Suarez-Almazor et al., 1993).

B VITAMINS The role of B vitamins in skeletal health has been the subject of many recent research studies. Elevated serum hom*ocysteine levels may be caused by deficiencies of folate (folic acid), vitamin B12, or vitamin B6. hom*ocysteine has recently been linked to fragility fractures, including hip fractures in older men and women (Gjesdal et al., 2006; van Meurs et al., 2004). Incident osteoporosis-related fractures were assessed in 702 Italian participants aged 65–94 years, with a mean follow-up of 4 years, and it was found that low serum folate was responsible for the association between hom*ocysteine and risk of osteoporosis-related fractures in elderly persons (Ravaglia etal., 2005). In two recent studies, poor vitamin B12 status was associated with low BMD in men and women and osteoporosis in elderly women but not men (Dhonukshe-Rutten et al., 2003; Tucker et al., 2005). It is unclear whether associations such as this are only a result of B12 deficiency or an indication of overall poor nutrition and frailty. Based on data collected on older men and women, that is, aged >55 years, who underwent dual-energy x-ray absorptiometry (DXA) scans of the hip as participants in phase 2 of the third U.S. National Health and Nutrition Examination Survey (n = 1550), the prevalence of osteoporosis in those with serum B12 in the lowest quartile was two times greater than that in individuals with serum B12 in the highest quartile (Morris etal., 2005). hom*ocysteine and vitamin B12 status indicators were negatively associated with BMD in older Americans (Stone et al., 2004). Whether this association reflects a causal relation remains unclear. In 1002 men and women (mean age 75 years) in the Framingham Osteoporosis study, low serum B vitamin concentrations were noted to be a risk factor for decreased bone health, although these low concentrations did not fully explain the relationship between elevated hom*ocysteine and hip fracture (McLean et al., 2008). In fact, controlled trials are clearly needed to determine whether treatment with B vitamins would reduce fracture rates among community-dwelling cohorts (McLean and Hannan, 2007). The Hordaland hom*ocysteine Study is a population-based study of more than 18,000 men and women in Western Norway, and among women in this study, raised hom*ocysteine levels were associated with decreased BMD and increased risk of osteoporosis (Refsum et al., 2006). Low folate concentrations were related to more than a two-fold elevation in hip fracture risk versus higher folate concentrations (McLean et al., 2008). In some studies, folate is more strongly related to BMD than any other B vitamin (Baines et al., 2007; Cagnacci et al., 2003, 2008; Rejnmark et al., 2008). However, in a 2-month study of folic acid supplementation in 61 healthy individuals, short-term folic acid supplementation did not affect biochemical bone markers in subjects without osteoporosis but who had a low folate status (Herrmann et al., 2006), and combined administration of folic acid, B6, and B12 over 1 year had no effect on bone turnover (Herrmann et al., 2007). However, in older Japanese, folic acid and [me]cobalamin reduced hip fracture as compared with placebo (Sato et al., 2005). In the Rotterdam study, those with the highest quartile of B6 intake had a lower risk of fracture than that of individuals with low B6 intakes (Yazdanpanah et al., 2007). Clearly, a need exists for long-term clinical trial data to determine the role of each B vitamin in skeletal health.

218

Diet, Nutrients, and Bone Health

OTHER NUTRIENTS Boron Boron is not an essential nutrient, so no recommended intakes have been published. Boron is present in several foods, such as fruits, vegetable, nuts, eggs, wine, and dried foods (Nielsen, 2008). A significant number of people, however, do not consistently consume more than 1 mg a day of boron (Nielsen, 1998), and whether such a low intake is of clinical concern is unknown. A small study noted that urinary boron excretion changes rapidly with changes in boron intake, indicating that the kidney is the site of homeostatic regulation (Sutherland et al., 1998). Although a few studies have found that 3 mg daily of boron may have a positive effect on bone (Nielsen, 1990; Nielsen etal., 1987), controlled trials are lacking.

Copper Copper is an essential element required by many enzymes including lysyl oxidase, which is required for cross-linking of collagen. Severe deficiency does have profound effects on bone (see also Chapter 13). A small study of 25 females with low bone mass found a correlation between the plasma copper concentrations and BMD of the lumbar spine as measured using DXA and quantitative computerized tomography (Chaudhri et al., 2009). Clearly, further study is needed regarding the role of copper in bone health.

Zinc A few intervention trials using zinc supplements have generated variable results on bone turnover and BMD (Gur et al., 2002; Strause et al., 1994); in one study, a mixture of trace elements on bone was investigated (Strause et al., 1994). Profound zinc deficiency leads to reduced bone growth and maturation, but little evidence has been reported showing that dietary zinc levels have an effect on bone mass or fractures related to osteoporosis. In a metabolic study of 21 women, low dietary zinc (3 mg/day) resulted in undesirable changes in circulating calcitonin and osteocalcin. However, ahigh intake of zinc may be a health concern for individuals consuming less than the recommended amounts of magnesium because of a zinc–magnesium interaction (Nielsen and Milne, 2004).

Silicon Cereals provide the greatest amount of silicon in the U.S. diet (about 30%), followed by fruit, beverages (hot, cold, and alcoholic beverages combined), and vegetables; together, these foods provided over 75% of the silicon intake. Silicon intakes may be suboptimal (McNaughton et al., 2005); a reported decrease of silicon concentrations occurs with age, especially in women (Bisse et al., 2005). Dietary silicon may be beneficial to bone and connective tissue health on the basis of recently reported strong positive associations between dietary silicon intake and BMD in U.S. and U.K. cohorts (Jugdaohsingh, 2007). The biological role of silicon in bone health remains unclear, although a number of possible mechanisms, including effects on the synthesis of collagen and/or its stabilization and on matrix mineralization, have been suggested. In a cross-sectional study of a sample of 2847 participants in the Framingham Offspring cohort (aged 30–87 years), dietary silicon correlated positively and significantly with BMD at all hip sites in men and premenopausal women, but not in postmenopausal women (Jugdaohsingh et al., 2004). Silicon appears to mediate the positive association between beer and BMD, but not of wine or liquor, in the Framingham Offspring cohort (aged 29–86 years) (Tucker et al., 2009). Results associating silicon to skeletal health require further follow-up.

Micronutrients and Bone

219

COMBINATIONS OF TRACE MINERALS Subclinical zinc and/or copper deficiency, resulting from a reduced dietary intake of micronutrients and reduced absorption (Thomson and Keelan, 1986), may result in bone loss. Both zinc and copper are essential cofactors for enzymes involved in the synthesis of various bone matrix constituents. Calcium supplementation may accentuate the problem of reduced serum zinc and copper levels by impairing the absorption and retention of these nutrients (Snedeker et al., 1982). The relationships among BMD and zinc, copper, and calcium in the meal and hair of 470 urban and rural elderly people were studied, and the amount of zinc, copper, and calcium in the meal was positively correlated with BMD (Li et al., 2005). Although significant correlations were found between serum elements, such as calcium, sodium, potassium, magnesium, zinc, iron, copper, and selenium, no significant differences in the concentrations of these elements were found in 290 women assigned to groups with osteoporosis, low bone mass, or normal bone mass (Liu et al., 2009). Two studies have shown that a combination of several minerals (zinc, manganese, and copper) with calcium was able to reduce spinal bone loss in postmenopausal women (Gur et al., 2002; Strause et al., 1994). These data indicate that individual or combinations of trace elements do not have a clear impact on skeletal health; hence, the effects of trace minerals on bone remain unknown. Therefore, the best advice is for individuals to consume a varied diet with adequate intakes of fruits, vegetables, cereals, and proteins to obtain enough of all the trace elements.

CONCLUSIONS The known benefits of calcium and vitamin D on bone cannot be considered in isolation from the other components of the diet, especially vitamins and the trace minerals. However, the needs of the other micronutrients for optimizing bone health can be easily met by a healthy varied diet that is high in fruits, vegetables, legumes, cereals, and adequate amounts of animal protein. In elderly individuals, greater attention should be placed on B vitamin status and hom*ocysteine levels.

REFERENCES National Osteoporosis Foundation, America’s Bone Health: The State of Osteoporosis and Low Bone Mass in Our Nation. Report of the National Osteoporosis Foundation, Washington, DC, 2002. Baines, M., Kredan, M.B., Davison, A., et al. 2007. The association between cysteine, bone turnover, and low bone mass. Calcif Tissue Int 81: 450–454. Bisse, E., Epting, T., Beil, A., et al. 2005. Reference values for serum silicon in adults. Anal Biochem 337: 130–135. Cagnacci, A., Bagni, B., Zini, A., et al. 2008. Relation of folate, vitamin B12 and hom*ocysteine to vertebral bone mineral density change in postmenopausal women. A five-year longitudinal evaluation. Bone 42: 314–320. Cagnacci, A., Baldassari, F., Rivolta, G., et al. 2003. Relation of hom*ocysteine, folate, and vitamin B12 to bone mineral density of postmenopausal women. Bone 33: 956–959. Carbone, L.D., Barrow, K.D., Bush, A.J., et al. 2005. Effects of a low sodium diet on bone metabolism. J Bone Miner Metab 23: 506–513. Carbone, L.D., Bush, A.J., Barrow, K.D., et al. 2003. The relationship of sodium intake to calcium and sodium excretion and bone mineral density of the hip in postmenopausal African-American and Caucasian women. J Bone Miner Metab 21: 415–420. Cauley, J.A., Murphy, P.A., Riley, T.J., et al. 1995. Effects of fluoridated drinking water on bone mass and fractures: The study of osteoporotic fractures. J Bone Miner Res 10: 1076–1086. Chachra, D., Vieira, A.P., and Grynpas, M.D. 2008. Fluoride and mineralized tissues. Crit Rev Biomed Eng 36: 183–233. Chaudhri, M.A., Kemmler, W., Harsch, I., et al. 2009. Plasma copper and bone mineral density in osteopenia: An indicator of bone mineral density in osteopenic females. Biol Trace Elem Res 129: 94–98.

220

Diet, Nutrients, and Bone Health

Devine, A., Criddle, R.A., Dick, I.M., et al. 1995. A longitudinal study of the effect of sodium and calcium intakes on regional bone density in postmenopausal women. Am J Clin Nutr 62: 740–745. Dhonukshe-Rutten, R.A., Lips, M., de Jong, N., et al. 2003. Vitamin B-12 status is associated with bone mineral content and bone mineral density in frail elderly women but not in men. J Nutr 133: 801–807. Durlach, J., Bac, P., Durlach, V., et al. 1998. Magnesium status and ageing: An update. Magnes Res 11: 25–42. Frassetto, L., Morris, R.C., Jr., Sellmeyer, D.E., et al. 2001 Diet, evolution and aging—The pathophysiologic effects of the post-agricultural inversion of the potassium-to-sodium and base-to-chloride ratios in the human diet. Eur J Nutr 40: 200–213. Freudenheim, J.L., Johnson, N.E., and Smith, E.L. 1986. Relationships between usual nutrient intake and bonemineral content of women 35–65 years of age: Longitudinal and cross-sectional analysis. Am J Clin Nutr 44: 863–876. Gabriel, S.E., Tosteson, A.N., Leibson, C.L., et al. 2002. Direct medical costs attributable to osteoporotic fractures. Osteoporos Int 13: 323–330. Gjesdal, C.G., Vollset, S.E., Ueland, P.M., et al. 2006. Plasma total hom*ocysteine level and bone mineral density: The Hordaland hom*ocysteine Study. Arch Intern Med 166: 88–94. Gur, A., Colpan, L., Nas, K., et al. 2002. The role of trace minerals in the pathogenesis of postmenopausal osteoporosis and a new effect of calcitonin. J Bone Miner Metab 20: 39–43. Hall, S.L., and Greendale, G.A. 1998. The relation of dietary vitamin C intake to bone mineral density: Results from the PEPI study. Calcif Tissue Int 63: 183–189. Harrington, M., Bennett, T., Jakobsen, J., et al. 2004a. Effect of a high-protein, high-salt diet on calcium and bone metabolism in postmenopausal women stratified by hormone replacement therapy use. Eur J Clin Nutr 58: 1436–1439. Harrington, M., Bennett, T., Jakobsen, J., et al. 2004b. The effect of a high-protein, high-sodium diet on calcium and bone metabolism in postmenopausal women and its interaction with vitamin D receptor genotype. Br J Nutr 91: 41–51. Harrington, M., and Cashman, K.D. 2003. High salt intake appears to increase bone resorption in postmenopausal women but high potassium intake ameliorates this adverse effect. Nutr Rev 61: 179–183. Hernandez-Avila, M., Stampfer, M.J., Ravnikar, V.A., et al. 1993. Caffeine and other predictors of bone density among pre- and perimenopausal women. Epidemiology 4: 128–134. Herrmann, M., Stanger, O., Paulweber, B., et al. 2006. Folate supplementation does not affect biochemical markers of bone turnover. Clin Lab 52: 131–136. Herrmann, M., Umanskaya, N., Traber, L., et al. 2007. The effect of B-vitamins on biochemical bone turnover markers and bone mineral density in osteoporotic patients: A 1-year double blind placebo controlled trial. Clin Chem Lab Med 45: 1785–1792. Houtkooper, L.B., Ritenbaugh, C., Aickin, M., et al. 1995. Nutrients, body composition and exercise are related to change in bone mineral density in premenopausal women. J Nutr 125: 1229–1237. Jackson, R., Bassford, T., Cauley, J., et al. 2003. The impact of magnesium intake on fractures: Results from the Women’s Health Initiative Observational Study (WHI-OS). Abstract presented at Annual Meeting of American Society for Bon and Mineral Research. Jones, G., Beard, T., Parameswaran, V., et al. 1997. A population-based study of the relationship between salt intake, bone resorption and bone mass. Eur J Clin Nutr 51: 561–565. Jugdaohsingh, R. 2007. Silicon and bone health. J Nutr Health Aging 11: 99–110. Jugdaohsingh, R., Tucker, K.L., Qiao, N., et al. 2004. Dietary silicon intake is positively associated with bone mineral density in men and premenopausal women of the Framingham Offspring cohort. J Bone Miner Res 19: 297–307. Kaptoge, S., Welch, A., McTaggart, A., et al. 2003. Effects of dietary nutrients and food groups on bone loss from the proximal femur in men and women in the 7th and 8th decades of age. Osteoporos Int 14: 418–428. Kurttio P., Gustavsson N., Vartiainen T., et al. 1999. Exposure to natural fluoride in well water and hip fracture: A cohort analysis in Finland. Am J Epidemiol 150: 817–824. Leveille, S.G., LaCroix, A.Z., Koepsell, T.D., et al. 1997. Dietary vitamin C and bone mineral density in postmenopausal women in Washington State, USA. J Epidemiol Community Health 51: 479–485. Levy, S.M., Eichenberger-Gilmore, J., Warren, J.J., et al. 2009. Associations of fluoride intake with children’s bone measures at age 11. Community Dent Oral Epidemiol 37: 416–426. Li, W., Tian, Y., Song, X., et al. 2005. Relationship between BMD and Zn, Cu, Ca levels in the hair and meal in elderly people. J Huazhong Univ Sci Technolog Med Sci 25: 97–99.

Micronutrients and Bone

221

Lin, P.H., Ginty, F., Appel, L.J., et al. 2003. The DASH diet and sodium reduction improve markers of bone turnover and calcium metabolism in adults. J Nutr 133: 3130–3136. Liu, S.Z., Yan, H., Xu, P., et al. 2009. Correlation analysis between bone mineral density and serum element contents of postmenopausal women in Xi’an urban area. Biol Trace Elem Res 131: 205–214. Massey, L.K., and Whiting, S.J. 1996. Dietary salt, urinary calcium, and bone loss. J Bone Miner Res 11: 731–736. McLean, R.R., and Hannan, M.T. 2007. B vitamins, hom*ocysteine, and bone disease: Epidemiology and pathophysiology. Curr Osteoporos Rep 5: 112–119. McLean, R.R., Jacques, P.F., Selhub, J., et al. 2008. Plasma B vitamins, hom*ocysteine, and their relation with bone loss and hip fracture in elderly men and women. J Clin Endocrinol Metab 93: 2206–2212. McNaughton, S.A., Bolton-Smith, C., Mishra, G.D., et al. 2005. Dietary silicon intake in post-menopausal women. Br J Nutr 94: 813–817. Mizushima, S., Tsuchida, K., and Yamori, Y. 1999. Preventive nutritional factors in epidemiology: Interaction between sodium and calcium. Clin Exp Pharmacol Physiol 26: 573–575. Morris, M.S., Jacques, P.F., and Selhub, J. 2005. Relation between hom*ocysteine and B-vitamin status indicators and bone mineral density in older Americans. Bone 37: 234–242. New, S.A., Bolton-Smith, C., Grubb D.A., et al. 1997. Nutritional influences on bone mineral density: A crosssectional study in premenopausal women. Am J Clin Nutr 65: 1831–1839. New, S.A., Robins, S.P., Campbell, M.K., et al. 2000. Dietary influences on bone mass and bone metabolism: Further evidence of a positive link between fruit and vegetable consumption and bone health? Am J Clin Nutr 71: 142–151. Nielsen, F.H. 1990. Studies on the relationship between boron and magnesium which possibly affects the formation and maintenance of bones. Magnes Trace Elem 9: 61–69. Nielsen, F.H. 1998. The justification for providing dietary guidance for the nutritional intake of boron. Biol Trace Elem Res 66: 319–330. Nielsen, F.H. 2004. The alteration of magnesium, calcium and phosphorus metabolism by dietary magnesium deprivation in postmenopausal women is not affected by dietary boron deprivation. Magnes Res 17: 197–210. Nielsen, F.H. 2008. Is boron nutritionally relevant? Nutr Rev 66: 183–191. Nielsen, F.H., Hunt, C.D., Mullen, L.M., et al. 1987. Effect of dietary boron on mineral, estrogen, and testosterone metabolism in postmenopausal women. Faseb J 1: 394–397. Nielsen, F.H., and Milne, D.B. 2004. A moderately high intake compared to a low intake of zinc depresses magnesium balance and alters indices of bone turnover in postmenopausal women. Eur J Clin Nutr 58: 703–710. Odland, L.M., Mason, R.L., and Alexeff, A.I. 1972. Bone density and dietary findings of 409 Tennessee subjects. 1. Bone density considerations. Am J Clin Nutr 25: 905–907. Pasco, J.A., Henry, M.J., Wilkinson, L.K., et al. 2006. Antioxidant vitamin supplements and markers of bone turnover in a community sample of nonsmoking women. J Womens Health (Larchmt) 15: 295–300. Prynne, C.J., Mishra, G.D., O’Connell, M.A., et al. 2006. Fruit and vegetable intakes and bone mineral status: A cross sectional study in 5 age and sex cohorts. Am J Clin Nutr 83: 1420–1428. Ravaglia, G., Forti, P., Maioli, F., et al. 2005. Folate, but not hom*ocysteine, predicts the risk of fracture in elderly persons. J Gerontol A Biol Sci Med Sci 60: 1458–1462. Ray, N.F., Chan, J.K., Thamer, M., et al. 1997. Medical expenditures for the treatment of osteoporotic fractures in the United States in 1995: Report from the National Osteoporosis Foundation. J Bone Miner Res 12: 24–35. Refsum, H., Nurk, E., Smith, A.D., et al. 2006. The Hordaland hom*ocysteine Study: A community-based study of hom*ocysteine, its determinants, and associations with disease. J Nutr 136: 1731S–1740S. Rejnmark, L., Vestergaard, P., Hermann, A.P., et al. 2008. Dietary intake of folate, but not vitamin B2 or B12, is associated with increased bone mineral density 5 years after the menopause: Results from a 10-year follow-up study in early postmenopausal women. Calcif Tissue Int 82: 1–11. Riggs, B.L. 1993. Formation-stimulating regimens other than sodium fluoride. Am J Med 95: 62S–68S. Rude, R.K., and Gruber, H.E. 2004. Magnesium deficiency and osteoporosis: Animal and human observations. J Nutr Biochem 15: 710–716. Rude, R.K., Gruber, H.E., Norton, H.J., et al. 2004. Bone loss induced by dietary magnesium reduction to 10% of the nutrient requirement in rats is associated with increased release of substance P and tumor necrosis factor-alpha. J Nutr 134: 79–85. Rude, R.K., and Olerich, M. 1996. Magnesium deficiency: Possible role in osteoporosis associated with glutensensitive enteropathy. Osteoporos Int 6: 453–461.

222

Diet, Nutrients, and Bone Health

Ryder, K.M., Shorr, R.I., Bush, A.J., et al. 2005. Magnesium intake from food and supplements is associated with bone mineral density in healthy older white subjects. J Am Geriatr Soc 53: 1875–1880. Sahni, S., Hannan, M.T., Gagnon, D., et al. 2008. High vitamin C intake is associated with lower 4-year bone loss in elderly men. J Nutr 138: 1931–1938. Sato, Y., Honda, Y., Iwamoto, J., et al. 2005. Effect of folate and mecobalamin on hip fractures in patients with stroke: A randomized controlled trial. JAMA 293: 1082–1088. Snedeker, S.M., Smith, S.A., and Greger, J.L. 1982. Effect of dietary calcium and phosphorus levels on the utilization of iron, copper, and zinc by adult males. J Nutr 112:136–143. Song, C.H., Barrett-Connor, E., Chung, J.H., et al. 2007. Associations of calcium and magnesium in serum and hair with bone mineral density in premenopausal women. Biol Trace Elem Res 118: 1–9. Sowers, M.R., Wallace, R.B., and Lemke, J.H. 1985. Correlates of mid-radius bone density among postmenopausal women: A community study. Am J Clin Nutr 41: 1045–1053. Stendig-Lindberg, G., Tepper, R., and Leichter, I. 1993. Trabecular bone density in a two year controlled trial of peroral magnesium in osteoporosis. Magnes Res 6: 155–163. Stone, K.L., Bauer, D.C., Sellmeyer, D., et al. 2004. Low serum vitamin B-12 levels are associated with increased hip bone loss in older women: A prospective study. J Clin Endocrinol Metab 89: 1217–1221. Strause, L., Saltman, P., Smith, K.T., et al. 1994. Spinal bone loss in postmenopausal women supplemented with calcium and trace minerals. J Nutr 124: 1060–1064. Suarez-Almazor, M.E., Flowerdew, G., Saunders, L.D., et al. 1993. The fluoridation of drinking water and hip fracture hospitalization rates in two Canadian communities. Am J Public Health 83: 689–693. Sutherland, B., Strong, P., and King, J.C. 1998. Determining human dietary requirements for boron. Biol Trace Elem Res 66: 193–204. Teucher, B., and Fairweather-Tait, S. 2003. Dietary sodium as a risk factor for osteoporosis: Where is the evidence? Proc Nutr Soc 62: 859–866. Thomson, A.B., and Keelan, M. 1986. The aging gut. Can J Physiol Pharmacol 64: 30–38. Tosteson, A.N., Gabriel, S.E., Grove, M.R., et al. 2001. Impact of hip and vertebral fractures on quality-­adjusted life years. Osteoporos Int 12: 1042–1049. Tranquilli, A.L., Lucino, E., Garzetti, G.G., et al. 1994. Calcium, phosphorus and magnesium intakes correlate with bone mineral content in postmenopausal women. Gynecol Endocrinol 8: 55–58. Tucker, K.L. 2003. Dietary intake and bone status with aging. Curr Pharm Des 9: 2687–2704. Tucker, K.L., Chen, H., Hannan, M.T., et al. 2002. Bone mineral density and dietary patterns in older adults: The Framingham Osteoporosis Study. Am J Clin Nutr 76: 245–252. Tucker, K.L., Hannan, M.T., Chen, H., et al. 1999. Potassium, magnesium, and fruit and vegetable intakes are associated with greater bone mineral density in elderly men and women. Am J Clin Nutr 69: 727–736. Tucker, K.L., Hannan, M.T., Qiao, N., et al. 2005. Low plasma vitamin B12 is associated with lower BMD: The Framingham Osteoporosis Study. J Bone Miner Res 20: 152–158. Tucker, K.L., Jugdaohsingh, R., Powell, J.J., et al. 2009. Effects of beer, wine, and liquor intakes on bone mineral density in older men and women. Am J Clin Nutr 89: 1188–1196. van Meurs, J.B., Dhonukshe-Rutten, R.A., Pluijm, S.M., et al. 2004. hom*ocysteine levels and the risk of osteoporotic fracture. New Engl J Med 350: 2033–2041. Wang, M.C., Moore, E.C., Crawford, P.B., et al. 1999. Influence of pre-adolescent diet on quantitative ultrasound measurements of the calcaneus in young adult women. Osteoporos Int 9: 532–535. Weber, P. 1999. The role of vitamins in the prevention of osteoporosis—A brief status report. Int J Vitam Nutr Res 69: 194–197. Wigertz, K., Palacios, C., Jackman, L.A., et al. 2005. Racial differences in calcium retention in response to dietary salt in adolescent girls. Am J Clin Nutr 81: 845–850. Wolf, R.L., Cauley, J.A., Pettinger, M., et al. 2005. Lack of a relation between vitamin and mineral antioxidants and bone mineral density: Results from the Women’s Health Initiative. Am J Clin Nutr 82: 581–588. Woo, J., Kwok, T., Leung, J., et al. 2009. Dietary intake, blood pressure and osteoporosis. J Hum Hypertens 23: 451–455. Yazdanpanah, N., Zillikens, M.C., Rivadeneira, F., et al. 2007. Effect of dietary B vitamins on BMD and risk of fracture in elderly men and women: The Rotterdam study. Bone 41: 987–994.

15

Dietary Protein’s Impact on Skeletal Health Anna K. Surdykowski, Anne M. Kenney, KarlL. Insogna, and Jane E. Kerstetter

CONTENTS Introduction..................................................................................................................................... 223 Protein Requirements and Intakes.................................................................................................. 223 Protein-Induced Metabolic Acidosis and Bone Loss......................................................................224 Epidemiological Studies................................................................................................................. 225 Isotopic Studies............................................................................................................................... 226 Potential Protein-Induced MechanismsOperating to Improve Calcium Balance.......................... 227 Summary......................................................................................................................................... 229 References....................................................................................................................................... 230

INTRODUCTION Both dietary calcium and vitamin D are undoubtedly beneficial to skeletal health. In contrast, despite intense investigation, the impact of dietary protein on calcium metabolism and bone balance remains controversial. Further complicating this debate is the potential difference that animal and vegetable protein sources may have on skeletal health. One previously held view is that diets high in protein were considered to be detrimental to bone, because the inorganic acids generated from the metabolism of amino acids increase urinary calcium excretion. According to this hypothesis, continued loss of calcium in the urine eventually results in negative calcium balance and loss of calcium from skeletal stores, including osteopenia if systemic acidosis is chronic. An alternative hypothesis is that a higher intake of protein increases circulating levels of insulin-like growth factor-1 (IGF-1), increases intestinal calcium absorption and improves muscle strength and mass, all of which may potentially benefit skeletal health. This review provides the scientific evidence supporting the latter hypothesis that dietary protein works synergistically to support both calcium retention and bone metabolism.

PROTEIN REQUIREMENTS AND INTAKES The Food and Nutrition Board, Institute of Medicine, the National Academies and Health Canada set dietary reference intakes (DRIs) for all of the macronutrients and micronutrients in our diet (Institute of Medicine, Standing Committee on the Scientific Evaluation of Dietary Reference Intakes, Food and Nutrition Board, 2002). The recommended dietary allowance (RDA) is defined as the intake that is sufficient to meet the needs of nearly all of the population (97.5%) and is perhaps the most commonly used reference value for protein. The RDA for protein in individuals aged 14 years and higher is 0.8 g/kg body weight. The acceptable macronutrient distribution range (AMDR) is defined as “a range of intakes that is associated with reduced risk of chronic diseases 223

224

Diet, Nutrients, and Bone Health

TABLE 15.1 Dietary Protein Intake (g/kg) in the United States from the NHANES 2003–2004 Database (Usual Intakes from Food Compared to Estimated Average Requirement [EAR])a Percentiles

Males   19–30 (n = 470)   31–50 (n = 624)   51–70 (n = 555)   71 + (n = 391) Females   19–30 (n = 393)   31–50 (n = 612)   51–70 (n = 606)   71 + (n = 406)

Mean ± SD

5

10

25

50

75

90

95

EAR

Percentage less than EAR

1.5 ± 0.4 1.4 ± 0.3 1.2 ± 0.3 1.0 ± 0.3

0.9 0.9 0.8 0.7

1.0 1.0 0.8 0.7

1.2 1.2 1.0 0.9

1.4 1.4 1.1 1.0

1.7 1.6 1.3 1.2

2.0 1.8 1.5 1.4

2.2 1.9 1.7 1.5

0.7 0.7 1.7 0.7

<3 <3 <3 <3

1.2 ± 0.3 1.1 ± 0.3 1.1 ± 0.3 1.0 ± 0.3

0.7 0.7 0.6 0.6

0.8 0.8 0.7 0.7

1.0 0.9 0.9 0.8

1.2 1.1 1.0 0.9

1.4 1.3 1.2 1.1

1.6 1.5 1.5 1.3

1.8 1.6 1.6 1.4

0.7 0.7 0.7 0.7

<3 4.0 7.2 8.6

Data from individuals with two days of reliable intake from NHANES 2003–2004. Body weights adjusted to nearest ideal body weight for children and adults. Results generated using SIDE program. Source: Adapted from Fulgoni, V.L., 3rd, Am J Clin Nutr, 87, 1554S–1557S, 2008. a

while providing adequate intakes of essential nutrients.” The protein AMDR for adults is 10%–35% of total caloric intake. Although Americans are typically considered to consume a high-protein diet, the 2003–2004 National Health and Nutrition Examination Survey (NHANES) dietary data do not support this notion for everyone. Fulgoni (2008) analyzed the latest NHANES data to characterize mean protein intakes according to age and sex categories (Table 15.1). These data show a trend toward decreased protein intake with age. Men, on average, consume more protein than women at all stages of life. Whereas 10% of women between the ages of 19 and 50 years had a protein intake at or below the RDA, 25% of the women over the age of 70 years were eating 0.8 or fewer g/kg (body weight) of protein each day. Overall, 10% of the entire population over 70 years did not meet the RDA for protein. In addition, 50% of adults aged 71 years and older consumed less than 1 g/kg protein, an amount slightly above the RDA (Fulgoni, 2008). Therefore, not all adults consume a high-protein diet, and individuals that consume the lowest protein diets, that is, the elderly, are at the highest risk of bone loss.

PROTEIN-INDUCED METABOLIC ACIDOSIS AND BONE LOSS An increase in dietary protein results in greater calcium excretion in the urine (Kerstetter and Allen, 1994). The source of this urinary calcium is not completely clear. Early research, up to the mid-1970s, suggested that intestinal calcium absorption was modulated by dietary protein (Wapnir, 1990). However, subsequent calcium balance studies failed to duplicate the original findings. Because the skeleton contains approximately 99% of the body’s calcium, the increased losses of urinary calcium (in excess of absorption) would support the notion that the skeleton was the most likely source of the extra urinary calcium. Therefore, the traditional hypothesis held that a high intake of protein, particularly from animal sources, generates a high fixed metabolic acid load, because the animal proteins contain higher amounts of sulfur- and phosphorus-containing amino acids. Dietary protein is the major contributor to endogenous acid production; the U.S. diet can generate 100 mEq of acid daily, primarily

Dietary Protein’s Impact on Skeletal Health

225

phosphate and sulfate anions (Barzel and Massey, 1998). Should the kidneys and lungs be unable to completely handle this diet-induced acid load, a source of additional buffer would be necessary via osteoclast-activated bone resorption. The large bicarbonate reservoir of the skeleton would provide this buffer, and calcium would consequently be released with the carbonate (Kerstetter et al., 2003). This hypothesis is supported by both cellular and animal studies (Arnett, 2003). Several human intervention trials demonstrate that the addition of a base, such as bicarbonate or citrate, suppresses bone resorption (Sebastian et al., 1994; Jehle et al., 2006), further supporting the hypothesis. The central question becomes, does this endogenous acid production from a high-protein diet have a sufficient magnitude to adversely impact bone? In the healthy individual, the lungs work to regulate pH by immediately expiring carbon dioxide, a metabolic by-product, while the kidney helps with buffering by excreting excess hydrogen ions primarily as ammonium ions and secondarily as phosphates, the so-called titratable acidity. The tandem actions of the kidneys and lungs very tightly and effectively regulate the pH of arterial blood. This tightly regulated homeostatic mechanism defends normal blood pH at 7.40 (constant) within a narrow pH range of 7.38 to 7.42, even in the face of variable day-to-day acid loads from dietary sources (Bonjour, 2005). Because food proteins are typically consumed throughout the daytime hours, acid generation occurs during the postprandial periods, thus providing ample time for neutralization during the fasting periods. The mechanism by which acidosis induces bone loss is through activation of osteoclasts via a slight decline in extracellular pH. However, the pH of extracellular fluid bathing cells deviates little from 7.40, and the initial activation of osteoclastic resorption requires a decline in the systemic pH of only approximately 0.2 units (Arnett, 2003). Small pH changes between 7.25 and 7.15 showed the greatest changes in osteoclast-mediated bone resorption (Arnett, 2003). It is not known if the extracellular pH in bone tissue is within this range or whether it changes following an animal-protein–containing meal. Given that the induced acid load from food is distributed over the course of a day, and in view of the robust buffering capacity of the renal and respiratory systems, it seems unlikely that increasing dietary (animal) protein would lead in healthy patients to osteoclast-dependent bone resorption, a conclusion also reached by Bonjour (2005). A recent meta-analysis by Fenton et al. (2009) evaluated the relationship between the acid­generating capacity of the diet and urinary calcium, calcium balance, and markers of bone resorption. Five studies were selected based on preset methodological criteria. Although a significant positive relationship between net acid excretion (NAE) and urinary calcium did exist, NAE was not associated with calcium balance or markers of bone resorption. The findings of this meta-analysis suggest that the increased acid-generating capacity of a high-protein diet may lead to increased urinary calcium loss, but this loss does not necessarily translate to negative calcium balance or bone loss.

EPIDEMIOLOGICAL STUDIES Epidemiological data largely support a positive association between protein intake and bone health. A few studies conducted in premenopausal adult women and adolescents have found a positive linkage between high dietary protein and bone (Beasley et al., 2010; Zhang et al., 2010). The vast majority of studies, however, have examined older adults. For example, Hannan et al. (2000) evaluated the relationship between baseline protein intake and 4-year change in bone mineral density (BMD) in 615 subjects with a mean age of 75 years. When percent protein intake was divided into quartiles, the group with the lowest protein intake (ranging from 0.21 to 0.71 g protein/kg) demonstrated the greatest loss in BMD. The highest quartile consumed 1.24 to 2.78 g/kg protein and demonstrated the least loss in BMD over the 4-year period (Hannan et al., 2000). Promislow et al. (2002) investigated the association between protein intake and BMD in a community-dwelling cohort of 572 women and 388 men aged 55–92 years living in Rancho Bernardo, California. For each increment of 15 g of animal protein daily, small but significant increases were found in women only over a 4-year period in BMD at the hip, femoral neck, spine, and total body. Interestingly, this association did not hold for consumers of vegetable protein nor was it observed in

226

Diet, Nutrients, and Bone Health

men. Nonetheless, the authors concluded that animal protein improved the skeletal health of elderly women! In a 5-year cohort study of 862 elderly women, food frequency questionnaires and dual-energy x-ray absorptiometry scans were used to examine the relationship between dietary protein at baseline and body composition 5 years later (Meng et al., 2009). After 5 years, there was greater bone mineral content (BMC) in those consuming the highest amount of protein (>87g/day) than in those consuming a moderate-protein (66–87g/day) or low-protein (<66g/day) diet. Whole-body BMC and appendicular BMC were 5.3% and 6.0% greater in the highest versus lowest tertile of protein intake, respectively. Subjects consuming the highest amount of protein also had significantly higher whole body lean muscle mass than that of subjects consuming the moderate or low levels of protein. These data support the hypothesis that protein intake positively impacts bone and muscle while also suggesting that the greater BMC may be due in part to an interaction between muscle and bone (Meng et al., 2009). A systematic review and meta-analysis of protein intake and bone health were recently reported by Darling et al. (2009). These investigators initially collected over 2000 potential studies, of which 61 met the inclusion criteria for the systematic review in that they measured both dietary protein and bone parameters (BMD or BMC, bone turnover, or fracture) in healthy adults. Overall, the authors could find little support for a negative relationship between dietary protein and bone. In fact, from the cross-sectional surveys, the pooled r values did not identify any negative association between protein intake and BMD/BMC at the clinically important skeletal sites. If anything, a slight positive association showed that protein was able to account for 1% to 2% of BMD measurements. Darling’s group further studied 19 randomized, placebo-controlled trials and found an overall slightly positive impact of protein supplementation (from all different sources) on lumbar BMD. These small changes in bone, however, did not translate to a beneficial association between dietary protein and fracture rates. In other words, no significant association (either positive or negative) of protein intake with fracture incidence was found in either the qualitative review or the meta-analysis. This meta-analysis does not support the contention that higher dietary protein is detrimental to bone, but it does suggest that a small yet potentially important positive effect may result from a higher protein intake. Although these epidemiological findings of positive protein-bone associations are important, they cannot establish a causal relationship between protein and bone.

ISOTOPIC STUDIES Several recent short-term feeding studies used calcium isotopes (generally considered the most sensitive and specific method) to evaluated calcium metabolism with different levels of dietary protein in humans. In a randomized crossover study of 15 healthy postmenopausal women, Roughead etal. (2003) assigned participants to low-meat-protein (12% of energy) and high–meat-protein (20% of energy) diets, each containing 600-mg calcium for 8-week periods. After a 4-week adjustment period on each diet, 2-day diets were labeled with 47Ca, and whole-body scintillation counting was performed over the subsequent 28 days. If the traditional hypothesis were correct, one would expect to see lower calcium retention among the group consuming the higher protein level. However, no significant difference was seen in calcium retention between the groups. Rather, a trend toward better calcium retention was observed on the higher protein diet. The high– and low–meat-protein diet also did not adversely affect biochemical markers of bone turnover (Roughead et al., 2003). In a follow-up randomized, controlled feeding study conducted by this same team of investigators, 27 postmenopausal women were assigned to either a low-calcium (675 mg Ca/day) or highcalcium (1510 mg Ca/day) diet. Subjects consumed low-protein (10% of energy) and high-protein (20% of energy) diets containing their assigned calcium level for 7 weeks each, with a 3-week washout period in between. After 3 weeks on each diet, 2-day diets were labeled with 47Ca isotopes, and whole-body scintillation counting followed. On the lower calcium diet, fractional calcium

Dietary Protein’s Impact on Skeletal Health

227

absorption increased with the higher protein diet (in comparison with the low-protein diet); however, on the higher calcium diet, this effect was not seen. Of further importance, the higher protein diet significantly increased serum IGF-1, an anabolic hormone that may be beneficial to bone. The higher protein diet also reduced urinary deoxypyridinoline, a marker of bone collagen breakdown (Hunt et al., 2009). In a final study, dual stable isotopes were used to evaluate the effect of a 10-day dietary intervention containing a moderate-protein (1.0g/kg) or high-protein (2.1g/kg) diet and a low intake of calcium (800 mg) in healthy women (Kerstetter et al., 2005). The high-protein diet resulted in a significant, 42% relative increase (7.7% raw) in intestinal calcium absorption and a significant increase in calcium excretion. No statistically significant differences were seen in kinetic measures of bone turnover between the moderate- and high-protein diets. However, the higher protein diet caused a significantly lower urinary fraction of calcium from bone origin. These data suggest that, at least acutely, hypercalciuria secondary to increased dietary protein is, in fact, the result of increased intestinal calcium absorption. Further, although not significant, there was a trend toward lower bone turnover in the high protein group, which may positively impact bone (Kerstetter et al., 2005). In all of the above isotopic studies (Roughead et al., 2003; Kerstetter et al., 2005; Hunt et al., 2009) where dietary protein had a positive effect on calcium and bone, dietary calcium was limited to 600–800mg. At higher calcium intakes, the impact is less evident. However, in a recent pilot feeding study, Ceglia et al. (2009) observed no increase in intestinal calcium absorption on a high-protein (1.5g/kg) versus low-protein (0.5g/kg) diet using dual-tracer stable isotopes. This intervention did not keep phosphorous constant. The high-dietary-phosphorus load that naturally accompanies a high–dietary-protein diet may blunt any change in intestinal calcium absorption (Ceglia et al., 2009). In addition, these subjects also received 1200-mg elemental calcium, which could have masked any effect of a change in dietary protein, as it did in the study by Hunt et al. (2009). The strength of the design of the feeding studies, with each subject serving as his or her own control, and the methods used to measure calcium metabolism make these findings an important addition to the epidemiological data supporting a positive relationship between long-term higher protein intake and bone health. However, the dietary feeding studies are limited by their short-term nature and small sample sizes. If experimental diets contain high levels of calcium, the impact of protein on absorption may not be evident (Ceglia et al., 2009; Hunt et al., 2009). On the other hand, when dietary calcium is limited (Kerstetter et al., 1998, 2005; Hunt et al., 2009), the effect of protein on calcium absorption becomes apparent. Because dietary calcium is inadequate in many older individuals, inadequate dietary protein may compound the problem of calcium bioavailability, whereas increasing protein may help to rectify it. The age of subjects may be important as older adults often have modest declines in glomerular filtration rate, which means that they retain more acid and calcium.

POTENTIAL PROTEIN-INDUCED MECHANISMSOPERATING TO IMPROVE CALCIUM BALANCE Several potential mechanisms might explain how increasing the amount of protein in the diet could potentially benefit calcium retention and bone homeostasis, including improved calcium absorption, increased production of IGF-1, and gain in lean body mass (Figure 15.1). These potential mechanisms probably overlap and are not necessarily exclusive. Experimental feeding studies demonstrate that increases in dietary protein enhance intestinal calcium absorption beginning at 1 week and at least to 7 weeks (Kerstetter et al., 2005; Hunt et al., 2009). If more calcium is absorbed from the intestine on a higher protein diet, parathyroid hormone (PTH) would be expected to decrease, resulting in reduced rates of bone resorption and bone loss. A significantly lower level of PTH observed in the trial with 10 individuals on a 10-day high-protein (1.5g/kg) versus low-protein (0.5g/kg) diet supports this hypothesis (Ceglia et al., 2009). In another

228

Diet, Nutrients, and Bone Health

Increased dietary protein

Increased growth hormone

Improved intestinal Ca absorption

Increased IGF–1

Suppressed PTH

Stimulated bone formation

Decreased skeletal turnover

Improved synthesis of collagen and noncollagen bone matrix proteins

Improved muscle mass, strength

Increased anabolic stress on bone

Decreased falls

Improved bone density, strength Decreased bone fracture

FIGURE 15.1  Potential mechanisms by which increased dietary protein positively impacts bone health. (Adapted from Gaffney-Stomberg, E., et al., J Am Geriatr Soc, 57, 1073–1079, 2009.)

short-term study, serum PTH was 1.6 to 2.7 times higher in women consuming low-protein (0.7g/kg) versus moderate-protein (1.0g/kg) diets for 14 days (Kerstetter et al., 1997). One mechanism by which dietary protein may increase intestinal calcium absorption is by an effect on gastric acid secretion. Calcium is absorbed in the ionized form by the small intestines. When ingested, calcium is in a food matrix or it is complexed with an anion, but not in an ionized form or readily available. Adequate gastric acid must be secreted to facilitate the release of calcium from these complexes and matrices, which allows for ionic calcium absorption. (Older adults may not have optimal gastric acid secretion.) However, the clinical intervention trials addressing this question, which usually used a proton pump inhibiting (PPI) drug to increase gastric pH, did not generate consistent conclusions. For example, Recker (1985) was the first to observe that fasting patients with achlorhydria absorbed less calcium (from a calcium carbonate supplement) than did control subjects with normal gastric acid production. In agreement with this finding are the results of a randomized crossover trial in which a significant decrease in fasting calcium absorption (from calcium carbonate) was observed when elderly women were given a PPI drug which blocked gastric acid production (O’Connell et al., 2005). On the other hand, during fed conditions, the blockage of gastric acid excretion by a PPI drug did not impact calcium absorption (Serfaty-Lacrosniere et al., 1995). Therefore, the influence of gastric acid on intestinal calcium absorption may be dependent on whether subjects are fed or fasted. To explore further the mechanisms underlying the effect of dietary protein on intestinal calcium absorption, we developed a rat model that simulates the acute response in humans (GaffneyStomberg et al., 2010). Female Sprague-Dawley rats were fed a control (20%), low-protein (5%), or high-protein (40%) diet for 1 week at which point duodenal mucosa was harvested and brush border membrane vesicles (BBMV) were prepared to evaluate calcium uptake. By day 7, urinary calcium was more than two-fold higher in the 40% protein group compared with that in the control group (4.2 mg/day vs. 1.7 mg/day, p < .05). Rats consuming the 40% protein diet both absorbed and retained more Ca compared with the 5% protein group in which absorption was 48.5% vs. 34.1% and retention was 45.8% vs. 33.7%, respectively (p < .01). The BBMV calcium uptake results suggest

Dietary Protein’s Impact on Skeletal Health

229

that a higher protein intake improves Ca absorption, at least in part, by increasing cellular calcium uptake. Another potential mechanism involves IGF-1, a key mediator of bone growth (Geusens and Boonen, 2002), that is regulated by dietary protein (Bonjour et al., 1997, 2001). In two studies employing high protein intakes, significantly greater levels of IGF-1 were found in subjects consuming the higher protein diets (Dawson-Hughes et al., 2004; Hunt et al., 2009). The anabolic effect of IGF-1 on muscle, rather than bone, may help further explain the positive relationship, though indirect, between dietary protein and bone. Thus, protein-induced increases in IGF-1 may indirectly benefit bone because of a direct enhancing action on muscle tissue and strength, which also increases bone strength. A frequently overlooked fact is that changes in bone mass and muscle strength tend to track together over the life span (Wolfe, 2006). The maintenance of bone strength is dependent upon maintenance of muscle mass, and a trophic effect of muscle contraction has a direct anabolic effect on bone (Wolfe, 2006). Like bone, muscle mass decreases with age; after the age of 40 years, skeletal muscle loss occurs at a rate of approximately 0.5%–1.0% per year (Paddon-Jones et al., 2008). Houston et al. (2008) evaluated the association between protein intake and lean body mass, largely skeletal muscle, over a 3-year period in 2066 black and white individuals between the ages of 70 and 79 years. Subjects with protein consumption in the highest quintile (median intake 1.1 g/kg/day) had significantly lower rates of loss of muscle mass (p < .05) than those in subjects in the lowest quintile of protein intake (median intake 0.7g/kg/day) (Houston et al., 2008). The relationship between protein intake and muscle mass in the aging population may be modulated by timing of food consumption, frequency of intake, quality of protein, presence or absence of other macronutrients, and physical activity. A recent hypothesis suggests that the consumption of high-quality protein at each meal (25–30 g) may combat sarcopenia by slowing or preventing muscle loss in the aging elderly population (Paddon-Jones and Rasmussen, 2009). The maintenance of good skeletal health in late life, including the avoidance of falls associated with aging, may be highly dependent on the maintenance of adequate muscle mass and function, which is in turn dependent to some extent on an adequate daily intake of dietary protein. The relationships among dietary protein, muscle mass, muscle strength and function, risk of falls and fracture, and bone density and bone health are complex and not yet well delineated. Because 50% of bone volume represents protein in the organic matrix, especially collagen, and continuous bone remodeling occurs, a higher intake of protein may positively impact bone formation and help maintain a greater bone mass (Heaney and Layman, 2008). Clearly, further findings from cross-sectional population trials and well-designed, long-term protein interventions are needed to define the relationships among dietary protein, bone health, and fractures.

SUMMARY The current RDA for protein of 0.8g/kg for adults was designed to provide the minimum amount of protein necessary to prevent a deficiency. The amount of protein that is required in the diet to optimize bone and muscle health is quite different and likely to be higher. Some groups of individuals in the United States consume inadequate protein diets; for example, 25% of women over the age of 70 years consume 0.8 g/kg protein or less per day, and overall, 10% of the population over 70 years does not meet the RDA for protein. Many epidemiological studies have found a significant positive relationship between protein intake and bone mass or density. Similarly, isotopic studies have also demonstrated greater calcium retention and absorption by individuals consuming high-protein diets, particularly when the calcium content of the diet was limiting. High-protein intake may positively impact bone health by several mechanisms, including calcium absorption, stimulation of the secretion of IGF-1, and enhancement of lean body mass. Clearly, long-term clinical intervention trials in which dietary protein is increased to 1.0–1.5 g/kg in healthy and well-nourished older individuals should be conducted to assess the effect of the level of dietary protein on muscle, bone, and fracture

230

Diet, Nutrients, and Bone Health

risk. Meanwhile, a limitation of protein in the diets of older individuals to improve bone health does not appear to be scientifically warranted. A widely held view is that high intakes of animal protein result in increased bone resorption, reduced BMD, and increased fractures because of its ability to generate a high fixed metabolic acid load. The hypothesis is supported by cellular and animal studies and human trials showing the addition of a base such as potassium bicarbonate or citrate which suppresses bone resorption (see review by Arnett, 2003 and Pizzorno et al., 2010). However, the consistency of the controlled isotopic studies as well as the recent epidemiological meta-analyses calls into question this conventional view. Clearly, further protein controlled human intervention trials are needed to establish causality and more clearly define the role of dietary protein and skeletal health.

REFERENCES Arnett, T. 2003. Regulation of bone cell function by acid-base balance. Proc Nutr Soc 62: 511–520. Barzel, U.S., and Massey, L.K. 1998. Excess dietary protein can adversely affect bone. J Nutr 128: 1051–1053. Beasley, J.M., Ichikawa, L.E., Ange, B.A., et al. 2010. Is protein intake associated with bone mineral density in young women? Am J Clin Nutr 91: 1311–1316. Bonjour, J.P. 2005. Dietary protein: An essential nutrient for bone health. J Am Coll Nutr 24: 526S–536S. Bonjour, J.P., Ammann, P., Chevalley, T., et al. 2001. Protein intake and bone growth. Can J Appl Physiol 26 Suppl: S153–S166. Bonjour, J.P., Schurch, M.A., Chevalley, T., et al. 1997. Protein intake, IGF-1 and osteoporosis. Osteoporos Int 7 Suppl 3: S36–S42. Ceglia, L., Harris, S.S., Abrams, S.A., et al. 2009. Potassium bicarbonate attenuates the urinary nitrogen excretion that accompanies an increase in dietary protein and may promote calcium absorption. J Clin Endocrinol Metab 94: 645–653. Darling, A.L., Millward, D.J., Torgerson, D.J., et al. 2009. Dietary protein and bone health: A systematic review and meta-analysis. Am J Clin Nutr 90: 1674–1692. Dawson-Hughes, B., Harris, S.S., Rasmussen, H., et al. 2004. Effect of dietary protein supplements on calcium excretion in healthy older men and women. J Clin Endocrinol Metab 89: 1169–1173. Fenton, T.R., Lyon, A.W., Eliasziw, M., et al. 2009. Meta-analysis of the effect of the acid-ash hypothesis of osteoporosis on calcium balance. J Bone Miner Res 24: 1835–1840. Fulgoni, V.L., III. 2008. Current protein intake in America: Analysis of the National Health and Nutrition Examination Survey, 2003–2004. Am J Clin Nutr 87: 1554S–1557S. Gaffney-Stomberg, E., Insogna, K.L., Rodriguez, N.R., et al. 2009. Increasing dietary protein requirements in elderly people for optimal muscle and bone health. J Am Geriatr Soc 57: 1073–1079. Gaffney-Stomberg, E., Sun, B., Cucchi, C., et al. 2010. The effect of dietary protein on intestinal calcium absorption in rats. Endocrinology 151: 1071–1078. Geusens, P.P., and Boonen, S. 2002. Osteoporosis and the growth hormone-insulin-like growth factor axis. Horm Res 58 Suppl 3: 49–55. Hannan, M.T., Tucker, K.L., Dawson-Hughes, B., et al. 2000. Effect of dietary protein on bone loss in elderly men and women: The Framingham Osteoporosis Study. J Bone Miner Res 15: 2504–2512. Heaney, R.P., and Layman, D.K. 2008. Amount and type of protein influences bone health. Am J Clin Nutr 87: 1567S–1570S. Houston, D.K., Nicklas, B.J., Ding, J., et al. 2008. Dietary protein intake is associated with lean mass change in older, community-dwelling adults: The Health, Aging, and Body Composition (Health ABC) Study. Am J Clin Nutr 87: 150–155. Hunt, J.R., Johnson, L.K., and Fariba Roughead, Z.K. 2009. Dietary protein and calcium interact to influence calcium retention: A controlled feeding study. Am J Clin Nutr 89: 1357–1365. Institute of Medicine, Standing Committee on the Scientific Evaluation of Dietary Reference Intakes, Food and Nutrition Board. 2002. Dietary Reference Intakes for Energy, Carbohydrates, Fiber, Fat, Protein and Amino Acids (Macronutrients). National Academy Press, Washington, DC. Jehle, S., Zanetti, A., Muser, J., et al. 2006. Partial neutralization of the acidogenic Western diet with potassium citrate increases bone mass in postmenopausal women with osteopenia. J Am Soc Nephrol 17: 3213–3222. Kerstetter, J.E., and Allen, L.H. 1994. Protein intake and calcium homeostasis. Adv Nutr Res 9: 167–181.

Dietary Protein’s Impact on Skeletal Health

231

Kerstetter, J.E., Caseria, D.M., Mitnick, M.E., et al. 1997. Increased circulating concentrations of parathyroid hormone in healthy, young women consuming a protein-restricted diet. Am J Clin Nutr 66: 1188–1196. Kerstetter, J.E., O’Brien, K.O., Caseria, D.M., et al. 2005. The impact of dietary protein on calcium absorption and kinetic measures of bone turnover in women. J Clin Endocrinol Metab 90: 26–31. Kerstetter, J.E., O’Brien, K.O., and Insogna, K.L. 1998. Dietary protein affects intestinal calcium absorption. Am J Clin Nutr 68: 859–865. Kerstetter, J.E., O’Brien, K.O., and Insogna, K.L. 2003. Dietary protein, calcium metabolism, and skeletal homeostasis revisited. Am J Clin Nutr 78: 584S–592S. Meng, X., Zhu, K., Devine, A., et al. 2009. A 5-year cohort study of the effects of high protein intake on lean mass and BMC in elderly postmenopausal women. J Bone Miner Res 24: 1827–1834. O’Connell, M.B., Madden, D.M., Murray, A.M., et al. 2005. Effects of proton pump inhibitors on calcium carbonate absorption in women: A randomized crossover trial. Am J Med 118: 778–781. Paddon-Jones, D., and Rasmussen, B.B. 2009. Dietary protein recommendations and the prevention of sarcopenia. Curr Opin Clin Nutr Metab Care 12: 86–90. Paddon-Jones, D., Short, K.R., Campbell, W.W., et al. 2008. Role of dietary protein in the sarcopenia of aging. Am J Clin Nutr 87: 1562S–1566S. Pizzorno, J., Frassetto, L.A., and Katzinger, J. 2010. Diet-induced acidosis: Is it real and clinically relevant? Br J Nutr 103:1185–1194. Promislow, J.H., Goodman-Gruen, D., Slymen, D.J., et al. 2002. Protein consumption and bone mineral density in the elderly: The Rancho Bernardo Study. Am J Epidemiol 155: 636–644. Recker, R.R. 1985. Calcium absorption and achlorhydria. N Engl J Med 313: 70–73. Roughead, Z.K., Johnson, L.K., Lykken, G.I., et al. 2003. Controlled high meat diets do not affect calcium retention or indices of bone status in healthy postmenopausal women. J Nutr 133: 1020–1026. Sebastian, A., Harris, S.T., Ottaway, J.H., et al. 1994. Improved mineral balance and skeletal metabolism in postmenopausal women treated with potassium bicarbonate. N Engl J Med 330: 1776–1781. Serfaty-Lacrosniere, C., Wood, R.J., Voytko, D., et al. 1995. Hypochlorhydria from short-term omeprazole treatment does not inhibit intestinal absorption of calcium, phosphorus, magnesium or zinc from food in humans. J Am Coll Nutr 14: 364–368. Wapnir, R. 1990. Calcium, magnesium and phosphorus absorption, nutritional status and effect of proteins. In Wapnir, R., ed. Protein Nutrition and Mineral Absorption. CRC Press, Boca Raton, FL. Wolfe, R.R. 2006. The underappreciated role of muscle in health and disease. Am J Clin Nutr 84: 475–482. Zhang, Q., Ma, G., Greenfield, H., et al. 2010. The association between dietary protein intake and bone mass accretion in pubertal girls with low calcium intakes. Br J Nutr 103: 714–723.

16

Omega-3 Fatty Acids and Bone Metabolism Bruce A. Watkins, Kevin Hannon, Mark F. Seifert, and Yong Li

CONTENTS Introduction..................................................................................................................................... 233 Muscle and Bone Relationships and Omega-3 Fatty Acids............................................................ 238 Research Findings of n-3 PUFA and Bone.....................................................................................240 PUFA Actions on Bone Cell Functions and Bone Modeling................................................240 Bone and Muscle Development and Their Interdependence: The Muscle–Bone Unit.......... 243 Role of Maternal n-3 PUFA Status on Fetal and Postnatal Development............................. 245 Diets Rich in the n-3 PUFA DHA Attenuate Loss of Bone and Muscle Mass Associated with Skeletal Unloading..........................................................................246 Endocannabinoids and the Musculoskeletal System.............................................................246 Human Studies Evaluating Dietary n-3 PUFA on Bone........................................................ 247 Studies on n-3 PUFA Actions in Bone Cell Cultures............................................................ 249 Recent Animal Studies Supporting Benefits of n-3 PUFA on Bone...................................... 250 Studies on n-3 PUFA in Models of Arthritis and Inflammation............................................ 251 Conclusions..................................................................................................................................... 251 References....................................................................................................................................... 252

INTRODUCTION The purpose of this chapter is to review the literature on omega-3 fatty acids and bone biology based on in vivo investigations and bone cell cultures. A recent search for publications indexed from January 1, 1999, to the present in PubMed using specific key words, resulted in the following number of refereed publications. The following key words were searched: PUFA and bone, 57 publications; omega-3 and bone, 89 publications; fatty acids and bone, 315 publications; and lipids and bone, 7891 publications. Publications for this review were selected based on human and animal studies that used or investigated the actions of omega-3 fatty acids or utilized bone cell cultures to understand their actions on bone formation. Herein, we explain the actions of omega-3 fatty acids on (1) the relationships between muscle and bone, (2) bone formation and bone mineral status in vivo, (3) recent findings in human and animal models, and (4) arthritis. The current findings of omega-3 fatty acids and their effects on bone modeling and remodeling are summarized and presented in table format (Tables 16.1 and 16.2). A major limitation of the research on omega-3 fatty acids and bone is the lack of consistent test mixtures and protocols to evaluate bone formation and resorption in modeling and remodeling bone. However, the future is promising because of the positive association between docosahexaenoic acid (DHA) and bone mineral density (BMD) in children and, in some studies, in adults. Fundamental knowledge on the differences between the actions of eicosapentaenoic acid (EPA) and DHA in bone cell cultures and animal models is now developing. The research is also testing various mechanisms 233

Treated with ALA, 45 RA patients (43 females and EPA, DPA(n-3), and 2 males) DHA Age 57.9 ± 10.8 yr (for the 39 subjects who completed the study)

Analyzed for ALA, EPA, and DHA

Dawczynski et al., 2009

Griffith et al., 2009

Weiler et al., 2005

Gronowitz et al., 2006

126 subjects (94 females, 32 males, mean age 69.7 ± 10.5 yr)

14 men with cystic fibrosis (aged 21.6 ± 2.2 yr), 42 healthy men as control (19.5 ± 0.2 yr) Analyzed for DHA 78 healthy young men with a and total n-3 mean age of 16.7 yr at baseline Analyzed for DHA 54 patients (25 males and 29 females, 35 children and 19 adults) with cystic fibrosis (6–33 yr, median 16 yr) Analyzed for EPA and Healthy infants at birth (16 DHA females and 14 males) and at 15 days

Hogstrom et al., 2007

Analyzed for ALA, EPA, and DHA

Gronowitz et al., 2008

85 healthy Caucasian 8-yr-olds (50 boys and 35 girls)

Human Subject

Analyzed for DHA, ALA

Type of n-3 PUFA

Eriksson et al., 2009

References

TABLE 16.1 Reported Observations of n-3 PUFA on Bone in Human Subjects

Remodeling

N/A

Modeling

Modeling

Modeling

Modeling

Modeling

Bone Modeling or Remodeling Observations

• LA, total n-6 PUFA, and the high ratio of n-6:n-3 PUFA were negatively associated with BMD. • AA was positively correlated with BMC and BMD of total body. • The endosteal circumference of radius was positively associated with serum PL DHA, while it was negatively associated with the ratio of n-6:n-3 PUFA in the patient group. • No such association was found in the control group. • Concentrations of DHA and total n-3 PUFA were positively associated with total and spine BMD at 22 yr of age. • The lumbar spine BMD Z-score correlated negatively with the ratio of AA:DHA. • Fatty acid status influenced BMD in cystic fibrosis children, but not in adults. • Cord RBC AA was positively correlated with infant whole-body BMC. • AA:EPA positively correlated with lumbar spine 1–4 BMC and femur BMC. • Maternal RBC AA was positively correlated with whole-body BMC. • Femur BMC was negatively predicted by maternal DHA. • The n-3 PUFA (ALA, EPA, and DHA) suppressed COX-2 expression. • n-3 PUFA did suppress the immune response as lymphocytes and monocytes were found to be significantly decreased. • Supplementing increased total plasma levels of EPA and DHA. • No improvement in the duration of morning stiffness, number of tender joints, and number of swollen joints was seen by dietary n-3 supplementation. • Research was done in patients undergoing orthopedic surgery with various underlying conditions. • No correlation was detected between marrow fatty acid concentrations (AA, EPA, and DHA) and bone mineral status in patients with normal, low, or osteoporotic BMD. • No dietary intake information was obtained from these patents. • Mixing postmenopausal women with similar aged men confounds the result.

234 Diet, Nutrients, and Bone Health

23 human subjects (20 males of 48.6 ± 1.6 yr and 3 females of 58.3 ± 2.7 yr) 167 patients aged 65 yr or more (mean age 73.2 yr, 80% women) with a low-energy fracture + matched controls at 1:1 Analyzed for EPA 256 men (22–59 yr of age) and 95 women (22–66 yr) Analyzed for EPA, 30 renal transplant patients (19 DHA, and total n-3 males; mean age 44 yr, range 22–65 years) Analyzed for total n-3 247 men (118) and women (129) aged 60 yr and older (overall 78.9 ± 6.8 yr) Treated with fish Bed rest study: 16 healthy subjects Short-duration spaceflight: (7 male and 3 female astronauts, aged 36–54 yr) Long-duration spaceflight: 24 crew members Reported total n-3 293 healthy subjects (63% female, 58.0 ± 5.5 yr) Analyzed for ALA, Community-dwelling men EPA, and total n-3 (n = 642, 72.9 ± 9.3 yr) and women (n = 564, 74.0 ± 9.2 yr for those without hormone therapy, n = 326, 69.4 ± 8.6 yr for those with hormone therapy)

Treated with ALA (walnuts and flaxseed oil) Analyzed for n-3

Remodeling

Remodeling

Remodeling

Remodeling

Remodeling

Remodeling

Remodeling

Remodeling

• Intake of n-6 PUFA was associated with an increased risk of bone marrow lesions. • There was a significant inverse association between the ratio of dietary LA:ALA and BMD at the hip. • An increasing ratio of total dietary n-6:n-3 PUFA was significantly and independently associated with lower BMD at the hip in all women.

• A higher intake of n-3 PUFA was associated with less N-telopeptide excretion during bed rest. • Higher fish consumption was associated with reduced BMD loss after space flight.

• The beta coefficient of the numbers of remaining teeth and EPA concentrations in the fraction was 0.89 (per 1% EPA, p = .007). • The ratio of n-3 PUFA:AA was positively correlated to BMD. • BMD improvement was positively related to EPA but negatively related to plasma phospholipid AA modification. • There was an association between greater reported n-3 PUFA intake and higher BMD.

• Serum NTx levels were significantly lower following the ALA diet. • There was no change in levels of bone-specific alkaline phosphatase activity across the three dietary treatments. • A higher ratio of MUFA to PUFA was associated with a reduced risk of fracture. • The intake of n-6 PUFA was associated with an elevated risk of fracture.

Notes: AA = arachidonic acid; ALA = alpha-linolenic acid; BMC = bone mineral content; BMD = bone mineral density; COX = cyclooxygenase; DHA = docosahexaenoic acid; DPA = docosapentaenoic acid; EPA = eicosapentaenoic acid; LA = linoleic acid; PL = phospholipids; PUFA = polyunsaturated fatty acids; MUFA = monounsaturated fatty acids; N/A = not applicable; NTx = cross-linked N-telopeptides of type I collagen; RBC = red blood cell; yr = year.

Weiss et al., 2005

Wang et al., 2008

Zwart et al., 2009

Rousseau et al., 2009

Baggio et al., 2005

Hamazaki et al., 2006

Martinez-Ramirez et al., 2007

Griel et al., 2007

Omega-3 Fatty Acids and Bone Metabolism 235

Animal Model

Analyzed for ALA, EPA, and DHA

Treated with menhaden oil Female rats (dams and pups to 21 wk) Treated with tuna fish oil 6- to 8-wk-old female BALB/c mice

Lau et al., 2009a, 2009b

Nielsen and Stoecker, 2009 Bendyk et al., 2009

Male and female fat-1 and wild type mice (3-wk-old)

Treated with menhaden oil Weanling SpragueDawley rats

Watkins et al., 2000

Weanling 21-day-old Long-Evans rats

Treated with ALA and DHA

Male low-birthweight (3-day-old) and very-low-birthweight Cotswold piglets (5-day-old) Treated with ALA Male Wistar rats (soybean oil) and fish oil (6-wk-old) Treated with ALA (linseed Rat dams (day 7 of oil and soybean oil) gestation) and 3-wk-old female pups Modeling

Treated with DHA

Type of n-3 PUFA

Reinwald et al., 2004

Korotkova et al., 2004

Lobo et al., 2009

Kohut et al., 2009

Reference

Modeling and remodeling Remodeling

Modeling and remodeling

Modeling

Modeling

Modeling

Modeling

Bone Modeling or Remodeling

TABLE 16.2 Reported Observations of n-3 PUFA on Bone in Animal Models Observations

• Femur length and cortical cross-sectional bone area and bone mineral content were significantly higher in the n-6 + n-3 PUFA group than in the other groups. • Cortical bone thickness in the n-6 + n-3 PUFA group was increased compared with the n-3 PUFA group. • Regulatory mechanisms were influenced by the ratio of n-6:n-3 PUFA early in life and not compensated for by the introduction of an ordinary diet after weaning. • n-3 PUFA deficiency diminished structural integrity in rat tibia. • Rats repleted with n-3 PUFA demonstrated accelerated bone modeling (crosssectional geometry) and an improved second moment in tibiae compared with those of control n-3-adequate rats. • Bone PGE2 was positively correlated with the ratio of AA:EPA. • The ratio of AA:EPA or PGE2 in bone was negatively correlated to bone formation rate. • Serum bone-specific alkaline phosphatase activity was greater in rats fed a diet high in n-3. • The ratio of n-6:n-3 PUFA in the femur and vertebra was negatively correlated with BMC and peak load. • ALA, EPA, and DHA were positively correlated with BMC (or BMD) and peak load. • Fish oil instead of safflower oil increased the maximum force to break and the bending moment of the femur, especially in rats fed adequate boron. • Tuna fish oil treatment reduced alveolar bone loss in induced periodontitis. • Alveolar bone loss was inversely related to n-3 PUFA tissue levels.

• Feeding fish oil to rats significantly increased BMC.

• Higher intake of AA and DHA lowered bone resorption relative to controls. • Dietary treatments did not change the bone formation rate.

236 Diet, Nutrients, and Bone Health

Treated with EPA and DHA

Treated with menhaden oil Male F344 BNF1 rats (12-months-old)

Treated with menhaden oil Sham or OVX CD-1 mice (6-months-old) Treated with menhaden oil Male F344 × BNF1 rats (12-months-old)

Treated with DHA

Poulsen et al., 2007

Shen et al., 2007

Ward and Fonseca, 2007

Watkins et al., 2006 Remodeling

Remodeling

Remodeling

Remodeling

Remodeling

Remodeling

Remodeling

Remodeling

Remodeling

• The OVX-induced decrease in lumbar spine BMC was significantly attenuated by DHA but not by EPA or GLA supplementation or supplementation with a mixture of all three long-chain PUFA. • Endosteal circumferences of tibiae were significantly greater in DHA and EPA compared with OVX. • Rats fed the n-3 PUFA diet had the highest values for peak load, ultimate stiffness, and Young’s modulus. • Rats fed the n-3 PUFA diet had lower values for formation rate, osteoclast number, and eroded surface in proximal tibia but higher values for periosteal mineral apposition and formation rates in tibia shaft. • Fish oil, either alone or combined with isoflavone, resulted in a higher BMD of LV. • Fish oil + isoflavone resulted in a higher peak load of LV4. • Rats fed the n-3 PUFA diet had the highest bone mineral content and cortical + subcortical BMD. • Rats fed the n-3 PUFA diet had higher values for serum insulin-like growth factor-I, parathyroid hormone, 1,25-(OH)2 vitamin D3, and bone-specific alkaline phosphatase activity but lower bone NO production and urinary Ca. • DHA-rich diets resulted in a significantly lower bone loss among the OVX rats. • DHA in the diet preserved rat femur BMC in the absence of estrogen. • Among the OVX rats, those fed the n-3 PUFA + isoflavone diet had a significantly higher value for tibial BMC. • The concentration of serum pyridinoline cross-links was significantly lower in the n-3 PUFA+ isoflavone group.

• Flaxseed + low-dose estrogen therapy resulted in the highest bone mineral density and peak load at the lumbar vertebrae, with no effect on bone mineral density or strength in the tibia and femur. • Combined treatment with 17beta-estradiol + DHA was more effective than was either treatment alone at preserving femur BMC and lowering circulating concentrations of proinflammatory IL-6. • Selective cyclooxygenase-2 inhibitor, prophylactic n-3 PUFA, and a combination of these two agents can inhibit gingival tissue MMP-8 expression.

Notes: AA = arachidonic acid; ALA = alpha-linolenic acid; BMC = bone mineral content; BMD = bone mineral density; DHA = docosahexaenoic acid; EPA = eicosapentaenoic acid; GLA = γ-linolenic acid; IL = interleukin; LV = lumbar vertebra; MMP = matrix metallopeptidase; NO = nitric oxide; OVX = ovariectomized; PGE2 = prostaglandin E2; PUFA = polyunsaturated fatty acids; wk = week.

Watkins et al., 2005

3-month-old female OVX rats Treated with menhaden oil 3-month-old female OVX rats

Treated with EPA and DHA

Vardar-Sengul et al., 2008

Shen et al., 2006

DHA

Poulsen et al., 2008a

3-month-old female Sprague-Dawley OVX rats 7-month-old female Sprague-Dawley OVX rats Adult male SpragueDawley (205 ± 29.3 g) rats 7-month-old female Sprague-Dawley OVX rats

Treated with ALA (flaxseed)

Sacco et al., 2009

Omega-3 Fatty Acids and Bone Metabolism 237

238

Diet, Nutrients, and Bone Health

of omega-3 fatty acid actions on gene expression. One evolving area of research in bone that has great potential to improve bone health is identifying how omega-3 fatty acids may alter receptor functions of the endocannabinoid signaling system. These aspects of omega-3 fatty acids and bone are also described.

MUSCLE AND BONE RELATIONSHIPS AND OMEGA-3 FATTY ACIDS Healthy and optimally functioning muscle and bone are vital for normal musculoskeletal growth, beginning at birth, and proceeding into adolescence and adulthood, and muscles are also necessary for maintaining the integrity of the skeletal system in the adult. Muscles exert biomechanical force-generated signals of the muscle/bone unit described by Schoenau and Frost (2002) for bone growth in children up to adults and in remodeling the skeletal system of the adult to minimize risk of osteoporosis and bone fractures (Figure 16.1). The mechanostat theory best describes how the biomechanical signals that emanate from muscle affect bone, that is, as part of the muscle/ bone unit. Substantial clinical and epidemiological evidence supports the premise that osteoporosis prevention must incorporate strategies to enhance peak bone mass early in life and to limit bone loss with aging. To understand the current science of the biomechanical and biochemical factors controlling bone growth in the young and which support musculoskeletal health throughout life, the fundamental principles of muscle and bone growth are examined along with environmental agents, mainly polyunsaturated fatty acids (PUFA), that influence this process. The relatively new nutrient candidates that are intimately involved in this process are the omega-3 (or n-3) PUFA. Specifically, DHA (22:6n-3) has been shown to support bone growth (Hogstrom et al., 2007), minimize muscle and bone loss (Poulsen et al., 2008a; Watkins et al., 2006, 2007), and improve bone architecture (mechanical properties testing) to resist long-bone fractures (Reinwald et al., 2004). Our research, and that of others, demonstrates important actions for n-3 PUFA in supporting optimal bone modeling and osteoblast functions, normalizing muscle insulin response to reduce the risk of obesity and diabetes, and minimizing chronic inflammation in the musculoskeletal system with estrogen decline and aging. The principle n-3 PUFA involved in this aspect of nutrition and Dietary PUFA, local and systemic factors

Activation of muscle mechanoreceptors

Muscle growth

Increased biomechanical strain

Synthesis of endocrine factors responsible for bone maintenance

Bone modeling

FIGURE 16.1  Muscles cause the largest loads and strains on bone to model and remodel bone tissue throughout life. These strains help control the biological mechanisms that determine whole bone strength and architecture, as explained by the mechanostat theory. Several dietary factors including essential fatty acids and the amounts of PUFA of the n-6 and n-3 families can influence muscle health. See text for details.

Omega-3 Fatty Acids and Bone Metabolism

239

health appears to be DHA, an essential fatty acid for neural growth and retinal development in the infant (Uauy et al., 2001). DHA has been associated with elevated bone mass (Watkins et al., 2006) and enhanced intestinal calcium absorption and retention (Haag et al., 2003; Poulsen et al., 2007), as well as reduced prostaglandin biosynthesis and bone resorption (Kang and Weylandt, 2008; Rahman et al., 2008). DHA improved bone mass in young rats with diabetes (a risk factor for low bone mass) (Yamada et al., 1995). Ovariectomized (OVX) rats fed either fish oil or DHA experienced attenuated loss of bone mineral, and in some cases, this observation was associated with increased bone formation and decreased bone resorption biomarkers (Watkins et al., 2005, 2006). In advanced aging, n-3 PUFA found in menhaden oil (which includes DHA) also attenuated bone loss in intact male rodents (Shen et al., 2006, 2007). Thus, in several studies, dietary sources of DHA benefit the musculoskeletal system by enhancing peak bone mass during growth and preserving bone mass during estrogen loss, as well as supporting muscle mass during advanced aging to prevent muscle atrophy and osteopenia. Healthy muscle is crucial in providing proper biomechanical signals needed to facilitate bone growth, proper architecture, and maintenance of mineral throughout life (Fricke and Schoenau, 2007). Muscle loss is followed by osteopenia at all ages (Carmeli et al., 2002; Giangregorio and McCartney, 2006). Moreover, substantial research has demonstrated that n-3 PUFA, including DHA, reduces muscle loss during cachexia (defined as the massive loss of up to 80% of both adipose tissue and skeletal muscle in cancer patients) (Khal and Tisdale, 2008; Smith et al., 2004; Tisdale, 1996, 2002). Thus, dietary approaches to attenuate age-associated muscle loss or losses of muscle and bone because of disuse may result in numerous health benefits beyond improvement of bone mineral content (BMC). Other aspects of the actions of n-3 PUFA include observations from studies conducted with DHA on bone modeling in rats. DHA supplemented to n-3-PUFA-deficient growing rats restored the n-3 PUFA content (including DHA) of bone tissue compartments including femoral marrow and cortical bone (Li et al., 2003a) and resulted in compensatory bone modeling and improvements in femur and tibia bone architecture (Reinwald et al., 2004). Moreover, Watkins et al. (2000) found that DHA was more efficiently incorporated into femoral marrow polar lipids than was EPA in growing rats and that the periosteum of the femur was highly enriched with DHA (10%–14%) compared with EPA (0.4%–0.3%). Hence, DHA appears to be more important than EPA for the periosteum, which supports osteoblastic activity, and enriching or restoring DHA concentrations in the periosteum, which has an abundance of osteoblasts, is rich with blood and lymph vessels and has a complex nerve supply. In a lipopolysaccharide (LPS)-induced experimental periodontitis model using adult male Sprague-Dawley rats (205 ± 29.3 g), feeding n-3 PUFA (40 mg/kg/day, 60% EPA + 40% DHA) for 14 days inhibited gingival matrix metallopeptidase (MMP)-8 (a collagenase regulated by proinflammatory cytokines) expression and increased tissue inhibitor of MMP (TIMP)-1 (endogenous downregulators of MMPs) expression, but the DHA-rich diet did not affect alveolar bone loss (Vardar-Sengul et al., 2008). OVX rats supplemented with DHA (expressed as a low ratio of n-6:n-3 PUFA of 5:1) had higher femur BMC compared with that of rats receiving a lower amount of DHA, that is, a ratio of n-6:n-3 PUFA of 10:1 (Watkins et al., 2006). DHA supplementation in OVX rats correlated with higher bone-specific alkaline phosphatase and lower pyridinoline (Pyd) cross-links, a marker of bone resorption. The diet containing DHA moderated the actions of linoleic acid (LA, the essential dietary n-6 PUFA) to support a bone biomarker profile that is associated with greater BMC in the femur of estrogen-deficient rats (Watkins et al., 2006). A lower total dietary PUFA content (both n-6 and n-3 PUFA and their ratio) led to a significantly higher osteocalcin level but lower values for Pyd and deoxypyridinoline in OVX rats compared with that of rats given a higher dietary PUFA content. With respect to the ratio of n-6:n-3 PUFA in this study, an excess of n-6, typical of the American diet, may exceed the capacity of n-3 PUFA to moderate the potential negative effect of n-6 PUFA.

240

Diet, Nutrients, and Bone Health

This research supports the concept that n-3 PUFA, specifically DHA in this case, maintains higher BMC in estrogen deficiency that correlates with higher biomarkers of bone formation and lower biomarkers of bone resorption compared with n-6 PUFA. Further, the actions of n-3 PUFA and lower total dietary n-6 PUFA both diminish the negative impact of excess n-6 PUFA on bone. In OVX rats (female Sprague-Dawley, 7-month-old) fed DHA (DHA in ethyl ester form, 0.5 g/ kg/day), BMC was conserved in lumbar spine, femur, and tibial trabeculae, whereas feeding EPA had no such effect on these bone sites (Poulsen et al., 2007). Feeding OVX rats (female SpragueDawley, 7-month-old) DHA (DHA in ethyl ester form, 0.5 g/kg/day) combined with 17β-estradiol conserved femur BMC and lowered circulating proinflammatory interleukin (IL)-6 better than did individual treatment of either DHA or 17β-estradiol (Poulsen et al., 2008a). However, combining DHA with a phytoestrogen, either genistein or daidzein, did not show any beneficial effect on bone mineral conservation in this animal model. Rather interesting relationships between dietary n-3 PUFA and cyclooxygenase-2 (COX-2) expression in human subjects have been reported. An n-3 PUFA supplement (1.1 g alpha-linolenic acid (ALA), 0.7 g EPA, 0.1 g docosapentaenoic acid, and 0.4 g DHA) in yogurt significantly decreased LPS-stimulated COX-2 expression in a randomized double-blind placebo-controlled crossover study (Dawczynski et al., 2009). In children, serum fatty acid profiles were found associated with bone mineralization in healthy 8-year-old boys and girls (Eriksson et al., 2009). Consistent with the findings in rats in our laboratory (Watkins et al., 2000), serum phospholipid LA was found to be negatively associated with BMD of the total body and lumbar spine, and this same pattern was found for both the total serum n-6 PUFA concentration and the ratio of n-6:n-3 PUFA. The correlations between serum phospholipid EPA, DHA, the ratio of arachidonic acid (AA) to EPA (AA:EPA) or ratio of AA:DHA, and bone parameters were not statistically significant in this subject group (Eriksson et al., 2009). In human infants, whole-body BMC was positively predicted by cord red blood cell (RBC) AA content, and lumbar vertebrae (L1–4) and femur BMC were positively associated with cord ratio of AA:EPA but negatively related to maternal DHA (Weiler et al., 2005). Clearly, blood and other tissue analyses of PUFA need to be considered as potential markers of bone status, and as the research advances in this area, a clearer relationship will be found for specific fatty acids, their ratios, and bone status.

RESEARCH FINDINGS OF N-3 PUFA AND BONE PUFA Actions on Bone Cell Functions and Bone Modeling PGE2 is a robust modulator of biochemical activity in bone and has been reported to influence osteoblast function (Watkins et al., 2001). At moderate levels, PGE2 supports bone formation, but at high concentrations, it promotes bone resorption (Watkins et al., 2000). In growing rats, a high dietary ratio of n-6:n-3 PUFA was positively correlated with lower bone formation rates and higher capacity for ex vivo PGE2 production in bone (Figure 16.2) (Watkins et al., 2000). Our laboratory confirmed that n-3 PUFA not only reduced the production of PGE2 in vivo but also affected the expression of COX-2, a key enzyme that catalyzes the biosynthesis of PGE2 from AA, in osteoblastic bone cell cultures (Watkins et al., 2003). Furthermore, n-3 PUFA feeding helped maintain BMC in OVX rats (Watkins et al., 2005, 2006) and mice (Fernandes et al., 2003). Moreover, n-3 PUFA levels in bone and osteoblastic cell cultures have been shown to coincide with the elevated levels of bone formation markers that are indicative of enhanced bone formation activity (Watkins et al., 2000, 2001,2003). Generally, DHA can function to attenuate AA conversion to prostanoids, and although DHA is not an eicosanoid precursor, it would provide an important means for modulating AA prostanoidmediated physiological and pathological processes (Mori and Beilin, 2001; Raisz et al., 1989). The need for understanding the role of n-3 PUFA in musculoskeletal health is further justified by the most recent dietary reference intakes published by the Institute of Medicine of the National Academies for these fatty acids (Institute of Medicine, 2002).

241

Omega-3 Fatty Acids and Bone Metabolism 35

6

30

5

20 3

15

6

10

2

5 0

2

4

5

0 10

20

40 PGE2 15

60 20

1

80 25

BFR (µm/d)

4

BFR

PGE2(ng/g)

25

30

35

Ratio of AA:EPA

FIGURE 16.2  Relationships between the ratio of AA:EPA in bone and ex vivo bone PGE2 production and bone formation in the rat. A positive correlation (shown as solid linear trend line that can be expressed as y = 0.439x + 17.389) was observed between ex vivo bone PGE2 (y) and the ratio of AA:EPA (x) in bone (▪). A negative correlation (shown as dotted linear trend line that can be expressed as y = 20.0396x + 4.705) was found between bone formation rate (BFR) (y) and the ratio of AA:EPA (x) (♢). The panel insert shows a negative correlation (shown as solid linear trend line that can be expressed as y = 20.0194x + 4.819) between BFR (y) and ex vivo PGE2 (x) production in bone. Bone alkaline phosphatase activity (bone formation marker) was greater in rats fed a diet high in n-3 PUFA or a low dietary ratio of n-6:n-3 PUFA. (Adapted from Watkins, B.A., Li, Y., Allen, K. G. D., et al., J Nutr 130, 2274–84, 2000.)

The evidence suggests that the n-3 PUFA (EPA and DHA) increase bone formation rates in growing rodents, and the response is associated with lowering PGE2 ex vivo in bone (Watkins et al., 2000) (see Figure 16.2). Furthermore, consistent with elevating n-3 PUFA in bone tissues by feeding dietary sources of EPA and DHA to rodents, a reduction of COX-2 protein but higher bone alkaline phosphatase (ALP) activity was observed in osteoblast-like cells (Watkins et al., 2003). What is not clear is how n-3 PUFA may influence genes and transcription factors associated with osteoblasts and osteoclasts. The core binding factor α1 (Cbfa1), a transcription factor of osteoblast differentiation, was higher in MC3T3-E1 cells after 7 days of exposure to EPA at different doses (1 and 10 µM) compared with AA treatment but lower compared with AA treatment (10 and 100 µM) at 14 days (Watkins et al., 2003). The expression values for Cbfa1 were normalized to the vehicle control (VC), and generally all fatty acid treatments except conjugated LA increased expression compared with the VC. Hence, these data suggest that fatty acids may be stimulatory or inhibitory to osteoblast expression of Cbfa1. A way in which n-3 PUFA may be beneficial to bone formation, related to the relationships demonstrated in Figure 16.2, is to reduce bone tissue concentrations of AA for PGE2 production as a means to control osteoclastic activity and bone loss. For example, PGE2 has been implicated in osteoclastogenesis in that this prostanoid promotes osteoclast formation by increasing the expression of receptor activator of nuclear factor kappa B ligand (RANKL) (Li et al., 2002), and Suda etal. (2004) reported that PGE2 decreases osteoblastic expression of osteoprotegerin (OPG), which is a decoy receptor for RANKL. In addition, PGE2, a proposed receptor agonist to bone morphogenetic protein-2 (BMP-2), may potentiate the effects of BMP-2 on osteoclastogenesis (Blackwell et al., 2009). Therefore, a major action of n-3 PUFA on bone can be the direct effects on reducing AA concentrations as well as decreasing the production of PGE2 and attenuating its downstream actions (cytokines and transcription factors) to influence both osteoblasts and osteoclasts, as shown in Figure 16.3.

242

Diet, Nutrients, and Bone Health

Dietary sources AA and EPA, DHA

Osteoblast AA-PL

EP1 prostanoid receptor Early differentiation Bone loading

EPA-PL DHA-PL COX-2

− Inflammatory agents

Bone resorption and osteoclastogenesis −

PGE2

IL-1 TNF

− Collagen ALP Osteocalcin

Osteoblast

Osteoclast and progenitor

− OPG ?

DHA

n-3 PUFA effects PGE2 OB gene expression Cytokine production Osteoclastogenesis

FIGURE 16.3  In bone, some actions of EPA and DHA are reducing the production of PGE2 either by lowering the substrate levels (AA in phospholipids) or effects on the activity or expression of COX in osteoblasts. In addition, substitution of EPA for AA in phospholipids can result in the production of PGE3. These actions are, in part, how n-3 PUFA influences the activity of osteoblasts and osteoclasts. EPA was reported to upregulate Cbfa1 (osteoblast differentiation); however, the responses appear to be related to both the concentration of fatty acids in and duration of culture time in osteoblast-like cells. In addition, reducing PGE2 has direct effects on osteoblasts that appear related to their functions in bone formation and to the downregulation of genes of pathways inducing osteoclastogenesis (osteoprotegerin [OPG] and receptor activator of nuclear-κB ligand [RANKL]). Some evidence in osteoblast-like cell cultures indicates that EPA and DHA decrease PGE2induced RANKL expression. Thus, the current findings suggest that with respect to osteoclastogenesis, n-6 PUFA via PGE2 and n-3 PUFA (EPA and DHA) may exert their actions through different genes and proteins. We have shown that elevated PGE2 production in bone (resulting from a diet rich in n-6 PUFA and low in n-3 PUFA) is associated with decreased in vivo bone formation in rodents (as shown in Figure 16.1). The beneficial actions of DHA in bone appear to be similar in muscle to prevent its atrophy in rodent models of hindlimb suspension and its association with higher BMD in growing individuals as described in the text.

During growth and development, the collective behavior of skeletal elements is organized to facilitate the bone modeling process in children. In long bones, surface drifts (the net loss [net resorption] or gain [net formation] of bone) resulting from physical strain on bone (Epker and Frost, 1965) occur during appositional growth that influences the bone modeling process in the young. Other long bone dimensions including cortical thickness, marrow cavity diameter, external diaphyseal and metaphyseal diameters, longitudinal curvatures, total cortical mass, and cross-sectional geometries can also be attributed to bone modeling activity. Bones of equal material quality but of differing geometries can vary in their degree of stiffness and potential to withstand stress or fracture (Ehrlich and Lanyon, 2002). Hence, bone modeling is a physiological activity significant to skeletal competency, whereas remodeling processes throughout adult life determine fracture risk commensurate with skeletal disease (Ehrlich and Lanyon, 2002). Until skeletal maturity is attained, over 90% of the periosteal and endosteal bone surfaces are continuously involved in bone appositional and resorptional activities that result in morphological

Omega-3 Fatty Acids and Bone Metabolism

243

changes pertinent to growth and reshaping (Ehrlich and Lanyon, 2002; Frost, 2003b). During bone modeling (e.g., long bone appositional growth), surface drifts occur, which alter bone cortical thickness, marrow cavity diameter, external diaphyseal and metaphyseal diameters, longitudinal curvatures, total cortical mass, and cross-sectional geometries (Frost, 2003b). These changes in the geometrical character of bone tissue govern the quality of bone, which is determined not only by its material properties but also by architectural, physical, and biological factors that influence its mechanical properties (Ehrlich and Lanyon, 2002; Frost, 2003b; Judex et al., 2003). Investigators have demonstrated that dietary factors including lipids can impact bone morphology and mechanical properties (Mori and Beilin, 2001; Wohl et al., 1998). Evaluation of bone mechanical properties, both structural and material, can be instrumental in elucidating the quality of bone architecture during bone modeling, as demonstrated with n-3 repletion in rats by our laboratory (Reinwald etal.,2004).

Bone and Muscle Development and Their Interdependence: The Muscle–Bone Unit Muscle and bone form an operational unit that controls the growth, maintenance, and operation of these organs. Muscles generate the largest loads and strains on the modeling and remodeling of bone throughout life (see Figure 16.1). These strains help control the biological mechanisms that determine whole bone strength (Schoenau and Frost, 2002). This relationship between muscle and bone is best described by the mechanostat theory, as previously explained, which states that increasing muscle mass and force during development creates the stimulus for the increase in bone mass and strength. Most hormones and other nonmechanical agents can modify the relationship between bone strength and muscle strength but cannot replace it (Schoenau and Frost, 2002). This codependence of muscle and bone means that the study of these two organ systems must be done as a unit and cannot be achieved practically as independent entities. As stated by Rauch and colleagues, “if muscle forces drive bone development, then analyses of muscle function should be added to the armamentarium of clinicians diagnosing bone disorders” (Rauch et al., 2004). Postnatally, the mechanostat theory is supported by a large volume of work that correlates muscle strength with bone density and strength. For example, peak velocity in lean body mass accretion precedes that of peak velocity of BMC accretion by a few months (Rauch et al., 2004). Very active humans without exceptionally strong muscles, such as marathon runners, lack the whole bone strength that weight lifters obtain (Frost, 2003a; Schoenau and Frost, 2002). Forearm muscle strength is significantly correlated with BMD (Ozdurak et al., 2003). Also supporting the theory is the observation that when muscle mass is decreased in children, such as those with cerebral palsy or muscular dystrophy, in comparison with their normal peers, BMD is significantly reduced (Gray et al., 1992; Lingam and Joester, 1994). According to Frost, the mechanostat theory (see Figure 16.1) is a result of muscle-derived flexural loads that are symmetrically oriented around the cross-sectional circumference of bone, causing a uniform increase in bone diameter (Frost, 1973). However, repetitive, similar dynamic flexural straining asymmetrically around the bone activates the flexure-drift feedback system and causes drifts of lamellar bone surfaces in tissue space. Bone surfaces will move toward the flexural concavity that arises when the flexural loads are applied, with convex-tending surfaces activating an osteoclastic drift (osteoclasts solubilize the basic organic and inorganic constituents of the bone and return them to the blood, thus facilitating bone resorption) and concave-tending ones activating osteoblastic drift (bone deposition by osteoblasts). Therefore, muscle forces have the capacity to control the shape and density of long bones and, hence, their architecture to resist fracture. As an example of diet effects on postnatal bone development, our laboratory examined repletion of n-3 PUFA in n-3-PUFA-deficient rats. In this investigation, three groups of rats were studied, second-generation n-3-deficient, n-3-repleted, and a control n-3-adequate rats (Reinwald etal., 2004). The n-3-adequate diet contained ALA (2.6% of total fatty acids) and DHA (1.3% of total fatty acids). Fatty acid composition of the hindlimb tissues (bone and muscle) of chronically n-3-deficient rats

244

Diet, Nutrients, and Bone Health

revealed a marked increase in n-6 PUFA (20:4n-6 or AA, 22:4n-6, and 22:5n-6) and a corresponding decrease in n-3 PUFA (18:3n-3, EPA, 22:5n-3, and DHA). Measurement of bone mechanical properties (energy to peak load) of tibiae showed that n-3 deficiency diminished structural integrity (Reinwald et al., 2004). Rats repleted with n-3 PUFA demonstrated accelerated bone modeling (cross-sectional geometry) and an improved second moment in tibiae compared with control n-3­adequate rats after 28 days of dietary treatment. This study showed that repletion with dietary n-3 PUFA restored the ratio of n-6:n-3 PUFA in bone compartments (compared with the adequate n-3 PUFA group) and reversed the compromised bone modeling in n-3-deficient rats (Reinwald et al., 2004). Although this study showed the important relationship between bone modeling and improved bone architecture in rats, it is consistent with the concept that bone strength and BMD are related, and in our study, bone strength (based on mechanical property testing) was higher with n-3 PUFA repletion in rats. Thus, indirectly, because muscle-derived loads model bone and determine bone strength or BMD (Ozdurak et al., 2003), repleting n-3 PUFA to n-3-deficient rats in our study is suggestive that n-3 PUFA would improve this muscle–bone relationship to model bone and increase bone strength. In support of the role n-3 PUFA plays in muscle–bone relationship, it is reported that feeding fish oil (containing EPA and DHA) prevented lean body mass loss during a state of cachexia in a mouse model (Tisdale and Dhesi, 1990). Increased circulating cytokines such as tumor necrosis factor-alpha (TNFα), IL-6, and IL-1β occur in chronic disease states (Levine et al., 1990; Valdez and Lederman, 1997). However, local expression of proinflammatory cytokines could exert a more significant effect on muscle atrophy than could systemically derived cytokines, as skeletal muscle can synthesize TNFα (De Bleecker etal., 1999), IL-1 (Belec et al., 1997), and IL-6 (Bartoccioni et al., 1994; Li et al., 2003b; Saghizadeh et al., 1996; Schulze et al., 2003). Recently, Spate and Schulze reported that the cytokines IL-1β and TNFα activate the ATP-dependent proteolytic pathway and increase proteolysis in muscle (Spate and Schulze, 2004). TNFα is thought to activate proteolysis through a pathway that involves stimulation of reactive oxygen species (ROS) production and ROS signaling in muscle (Garg and Aggarwal, 2002; Goossens et al., 1995; Li et al., 2000). This activated ROS system then stimulates ATP-dependent proteolysis (Gomes-Marcondes and Tisdale, 2002; Li et al., 2005). Increases in ROS activity and oxidative stress to levels greater than that can be normally neutralized by intracellular antioxidant defenses (which remain unchanged) (Barreiro et al., 2005) are associated with muscle mass loss in catabolic states such as cancer (Tisdale, 2001), sarcopenia (Spiers et al., 2000), and immobilization (Kondo et al., 1993). Similar to the prostaglandin story, the effects of cytokines on muscle protein synthesis and degradation depend on the specific cytokine. Although the above-mentioned cytokines are deleterious for muscle health, IL-15 positively affects parameters associated with skeletal muscle fiber hypertrophy (Quinn et al., 1995). In a tumor-bearing rat model (AH-130 ascites hepatoma), IL-15 treatment significantly slowed the rates of protein degradation to spare skeletal muscle mass from cancer-induced wasting (Carbo et al., 2000). The underlying mechanism on these alterations was reported to be linked to the inhibition of the ATP–ubiquitin-dependent proteolytic pathway (Carbo et al., 2000). Dietary sources of n-3 PUFA have been found to inhibit the negative effects of cytokines in muscle but may have a significant role in this capacity by moderating PGE2 production in bone (Figure 16.3). In contrast, dietary lipids rich in n-6 PUFA enhance IL-1 production and tissue responsiveness to cytokines, whereas those rich in n-3 PUFA have the opposite effect (Grimble and Tappia, 1998). The exact mechanism behind this selective effect of n-3 PUFA is not clear. In addition, it is not known what effects n-3 PUFA exert on the synthesis and activity of the positive-acting cytokines, such as IL-15. Clearly though, n-3 PUFA have the ability to control muscle protein synthesis and degradation by regulating the action of muscle-degrading prostanoids and cytokines. DHA is likely vital to normal bone physiology as a physical constituent of bone compartments, periosteum (nerve abundant tissue), and cortical bone (Li et al., 2003a) for biological maintenance in muscle (in preventing muscle atrophy) and muscle–bone interdependence in modeling bone

Omega-3 Fatty Acids and Bone Metabolism

245

(Reinwald et al., 2004). We hypothesize that the endocannabinoid (EC) system, which is fundamentally involved in bone cell differentiation and mature functioning of the skeletal system (Bab etal., 2008), provides some aspect of muscle-to-bone communications yet identified but highly likely because this system influences obesity and insulin sensitivity in muscle and bone cell functions. The EC system, through activation of its receptors, is responsible for differentiation of osteoblasts and osteoclastogenesis (Idris et al., 2008; Ofek et al., 2006) and is a target for modulation by dietary n-3 PUFA that would decrease the AA-derived agonists and may also influence the receptors in muscle and bone.

Role of Maternal n-3 PUFA Status on Fetal and Postnatal Development Barker and his colleagues postulated that some chronic diseases associated with aging may be programmed in very early life (Godfrey and Barker, 2001). This supposition was initially termed “the fetal origins of adult disease hypothesis.” One of the major ideas of this hypothesis is that the effects of early programming are most pronounced when there is a mismatch between early nutritional deprivation and later nutritional affluence. Many examples of metabolic programming exist. For example, premature infants who consumed breast milk (generally higher in DHA content than formula) had lower mean arterial blood pressure at 13 to 16 years of age than that of age-matched, previously premature individuals given standard infant formula (Singhal et al., 2001). The influence of infant growth on long-term muscle strength appears to continue into adult age. Sarcopenia, considered an age-related loss in muscle mass and strength, has its origins in early life (Sayer etal., 2004), as does compromised BMC (Sayer and Cooper, 2005). Muscle metabolism appears to be programmed early in development (Taylor et al., 1995), and nutrition during neonatal life can alter muscle metabolic properties (Harrison et al., 1996). It has become clear that n-3 PUFA are important for optimal fetal development, and these fatty acids appear to be involved in metabolic programming. This programming is relevant for industrialized countries where consumption of n-6 PUFA has greatly eclipsed the intake of n-3 PUFA from plants and marine food products. The dietary ratio of n-6:n-3 PUFA has now risen to more than 10:1 (Kris-Etherton et al., 2000), and because the ratio of n-6:n-3 PUFA in the milk of women is dictated by diet (Fidler and Koletzko, 2000), the loss of n-3 PUFA can be expected to have serious implications for the health of the neonate and adult. Probably the most important issue regarding metabolic programming is that any early deficiencies appear to be irreversible, even following repletion with the missing nutrient. For instance, n-3 PUFA deficiency in the perinatal period results in elevated blood pressure later in life, even when animals were subsequently repleted with these fatty acids (Weisinger et al., 2001). Nutrition during the gestational and perinatal period seems to be a critical imprinting regulator of growth and development and, consequently, in the risk for diet-related adult chronic diseases (Godfrey and Barker, 2000). For example, during late gestation and throughout lactation, female rats fed an isocaloric diet containing 7% (by weight) fat as linseed oil (n-3 PUFA diet, ratio of n-6:n-3 PUFA = 0.4), soybean oil (ratio of n-6:n-3 PUFA = 9), or sunflower oil (n-6 diet PUFA, ratio of n-6:n-3 PUFA = 216) produced male and female pups having lower mean body weights than those of dams fed the n-6 PUFA diet containing soybean oil. As both male and female pups grew to 28weeks of age, those given the sunflower diet (rich in n-6 PUFA) ended up being the heaviest, and the male pups from the sunflower-oil-fed dams had the highest systolic blood pressure (Korotkova et al., 2005). Femur length, cortical cross-sectional area, and BMC were significantly higher in the rat pups fed sources of n-3 PUFA (Korotkova et al., 2004). As BMC may be programmed early in life, so too does skeletal muscle metabolism and fatigue resistance appear to be programmed during early life. Adults who were low-birth-weight infants and therefore had less muscle mass as children had significantly less resistance during adulthood to muscle fatigue than that of size-matched adults who were significantly larger at birth (Taylor etal.,1995).

246

Diet, Nutrients, and Bone Health

Diets Rich in the n-3 PUFA DHA Attenuate Loss of Bone and Muscle Mass Associated with Skeletal Unloading In our laboratory, 15-week-old male NIH Swiss mice (Harlan Labs, Indianapolis, IN) were fed either a diet supplemented with n-6 PUFA, one containing a combination of n-3 (DHA) and n-6 PUFA, or a control diet for 7 days (Watkins et al., 2007). Diets (AIN-93G basal diet) were isonitrogenous and isocaloric, varying only in the fat source. Skeletal-unloading-associated atrophy was then initiated in mice using hindlimb suspension (Warren et al., 1994), and mice continued on their respective diets during suspension. After 9 days of hindlimb suspension, mice were scanned by dual-energy X-ray absorptiometry (DXA), and regions of interest were defined as the whole body and a trapezoid area, which included the region of the hindlimb below the knee to the tarsus. In the n-6 PUFA diet group of mice, the bone mineral ash, as a percentage of dry weight in the femur, was 11.1% less in the suspended group as compared with the weight-bearing group. In contrast, in the n-3-PUFA-fed group of mice, the bone mineral ash/dry weight (femur) was reduced only 3.2% following suspension. These results demonstrate that, in comparison with an n-6 PUFA diet, consumption of an n-3PUFA-supplemented diet significantly reduces the catabolic effect of hindlimb suspension on bone mineral loss. The bone-sparing effects of n-3 PUFA were also observed by DXA analysis. After 9days of hindlimb suspension, a significant loss of whole-body and hindlimb BMD and BMC was observed. In contrast to the diet rich in n-6 PUFA, reduction in BMD and BMC was significantly attenuated in mice receiving the diet rich in DHA, the sole n-3 PUFA. In addition to attenuating the loss of bone mass associated with skeletal unloading, consuming a diet rich in DHA also suppressed the loss of muscle associated with hindlimb suspension in mice (Watkins et al., 2007). Mice consuming the n-6 PUFA control diet lost approximately 26% of the gastrocnemius/soleus muscle wet weight following 9 days of suspension (p < .05 weight-bearing n-6 group versus the suspended n-6 group). Mice consuming the diet supplemented with n-3 PUFA lost only 11% of the gastrocnemius/soleus muscle wet weight following 9 days of suspension. Paralleling these wet weights, consumption of a diet rich in the n-3 PUFA DHA also attenuated the loss in fiber diameter observed with hindlimb suspension. These results demonstrate that consumption of diets containing DHA attenuates loss of muscle associated with hindlimb suspension. The same experimental design was repeated with the n-3 PUFA EPA and in contrast to the results observed with DHA, EPA supplementation had no sparing effect with respect to bone loss associated with hindlimb suspension. Supplementing with DHA in the diet for only 7 days has been found in our hands to attenuate significantly the loss of bone mineral and muscle mass associated with disuse. Although these findings in mice are promising and potentially have significant and immediate importance to humans to help prevent adult bone loss associated with casting of limbs and in astronauts undergoing reduced gravity during spaceflight, much research is needed to establish a possible mechanism of action.

Endocannabinoids and the Musculoskeletal System Recently, we found that DHA exerts actions on the endocannabinoid signaling pathway to attenuate the stimulation of the receptors and their actions on osteoclast numbers and osteoclastogenesis so as to favor bone formation during growth (Hutchins et al., 2009). The cannabinoid receptors, CB1 and CB2, were identified in the early and mid-1990s, and much attention has focused on the CB1 in brain because of its involvement in food intake and the use of antagonists to control weight and reverse obesity. The endogenous agonists for these receptors include N-arachidonoylethanolamide (anandamide or AEA) and 2-arachidonoylglycerol (2-AG). Both AEA and 2-AG are synthesized on demand from AA. The actions include food intake and influence obesity, insulin resistance, and osteoblast and osteoclast actions. We hypothesize that cross-talk between muscle and bone is communicated through the endocannabinoid signaling system, much like that between the recently described muscle and adipose interactions (Eckardt et al., 2008).

Omega-3 Fatty Acids and Bone Metabolism

247

Although the EC system has been shown to regulate food intake, it has recently been found to be of great consequence to the skeletal system, especially to regulate bone mass (Idris et al., 2005, 2008; Tam et al., 2008). As described above, the EC is composed of two receptors, CB1 and CB2, and the main endogenous agonists for the receptors are AEA and 2-AG. Both agonists are derived from AA and can be synthesized in osteoblasts and muscle tissue. In vitro and in vivo modulation of the EC receptors has been shown to influence bone modeling. In fact, knockout mice that lack the CB2 endocannabinoid receptor have low BMC and lack capacity for bone formation (Ofek et al., 2006). The enzymes for the synthesis of agonists are expressed in osteoblasts, osteocytes, and bone lining cells (Bab and Zimmer, 2008). The CB2 receptor has been identified in osteoblasts grown in an osteogenic medium for up to 28 days, whereas CB1 was not expressed in osteoblasts grown for this duration (Ofek et al., 2006; Tam et al., 2006). However, CB1 is expressed in sympathetic nerve fibers of trabecular bone (Tam et al., 2006). CB1 and CB2 are found in osteoclasts (Idris et al., 2005; Ofek et al., 2006; Scutt and Williamson, 2007). The expression of CB2 in osteoblasts mirrors that of osteoblast gene expression for tissue nonspecific ALP, runt-related transcription factor 2 (RUNX2), and parathyroid hormone receptor-1 (Ofek et al., 2006). The actions of the EC system in bone support our hypothesis that EC-related proteins are expressed during the development and growth of the musculoskeletal system and point to the need to investigate how DHA influences the actions of this system in bone and muscle. The use of dietary manipulation of fatty acids that decreases AA levels and increases DHA in bone promotes bone formation (Watkins et al., 2000) and bone strength (Reinwald et al., 2004) in growing rats, and both dietary EPA + DHA (menhaden oil) and DHA alone conserve bone mineral in tibia and femur, respectively, in OVX rats (Watkins et al., 2005, 2006). Recently, we have reported that n-3 PUFA modifies the mRNA of receptors and biosynthetic and degradation enzymes of endogenous agonists of the EC system in osteoblast-like cells (Hutchins et al., 2009). In addition, DHA was found to reduce AA in specific cell and tissue compartments of murine muscle and bone (Watkins, unpublished findings) and to reduce cellular levels of AEA (Matias et al., 2008). Moreover, DHA may decrease osteoclast activity by decreasing AEA because this agonist is reported to stimulate osteoclast formation (Idris et al., 2008). The new findings demonstrating n-3 PUFA influence on the expression of EC receptors and enzymes are a promising area of investigation and will likely enhance our understanding of muscle and bone interactions to optimize health through lipid nutrition.

Human Studies Evaluating Dietary n-3 PUFA on Bone Epidemiological studies on the relationship between dietary fatty acid intake and bone status in human subjects of various age groups confirm the concept that fatty acids are important modulators of skeletal health and that specific fatty acids, for example, n-3 and n-6 PUFA, affect bone metabolism differently. Rousseau and colleagues (2009) observed that in older adults aged 60 years and older (118 men and 129 women), higher self-reported n-3 PUFA intakes were positively related to femoral neck BMD. The mean reported intake of n-3 PUFA in this group of subjects was 1.27g/­day. In a 6-year follow-up study of a cohort of 78 healthy young men (mean age 16.7 years at baseline), to evaluate factors that determine BMD, measurements of BMD (measured at baseline and at 22years of age) of total body, hip, and spine and fatty acid concentrations (measured in the phospholipid fraction in serum at 22 years of age) showed that n-3 PUFA, especially DHA, were positively associated with BMD, and represented the peak BMD, in young men during this longitudinal study (Hogstrom et al., 2007). In a follow-up study that measured BMD at baseline and 2 years later in 54 cystic fibrosis male and female patients (aged 6–33 years) and fatty acid composition of serum phospholipids, both EFA status and AA and DHA were associated with lumbar spine BMD Z-score (Gronowitz et al., 2006).

248

Diet, Nutrients, and Bone Health

The BMD Z-score of lumbar spine was negatively correlated with the ratio of AA:DHA in serum phospholipids of children (n = 35, mean age in years after 2 years was 15.3 for boys and 13.9 for girls) but not in adults (n = 19). The lack of fatty acid associations with bone in adults is most likely because of significantly lower rates of bone growth, or a longer duration of study must be conducted (Gronowitz et al., 2006). In a 2-year (mean 24.4 months) follow-up study of 22 recipients (22–65 years of age) of a first renal allograft, after 2 years, a strong positive correlation between the ratio of n-3 PUFA:AA in plasma phospholipids and BMD was observed (Baggio et al., 2005). The value calculated for n-3 PUFA in the ratio included EPA, DHA, and 22:5n-3. Based on multivariate regression analysis of the data, the femoral BMD change was negatively associated with AA in plasma phospholipids of the renal transplant patients (Baggio et al., 2005). Several other human studies have examined the relationship of n-6 PUFA and bone mineral status. Eriksson et al. (2009) reported that serum fatty acid profiles are associated with bone mineralization in healthy 8-year-old children (50 boys and 35 girls). A general trend was that serum phospholipid LA (18:2n-6) was found to be negatively associated with BMD of total body and lumbar spine, and this same pattern was found for the total n-6 PUFA concentration and the ratio of n-6:n-3 PUFA. Based on dietary intake from the Melbourne Collaborative Cohort Study, investigators examined dietary PUFA intakes with marrow and knee cartilage (Wang et al., 2008) in a cohort of 293 healthy adult subjects without knee pain or injury. The purpose of this study was to evaluate dietary fat intakes with predisease and early stages of osteoarthritis using magnetic resonance imaging as a diagnostic assessment of knee health. The researchers reported a significant association between n-6 PUFA intake and an increased risk of bone marrow lesions, suggesting the importance of maintaining an optimal balance between n-3 and n-6 PUFA in the diet (Wang et al., 2008). In Spain, a case–control study (n = 167 with 1:1 matched controls) was conducted to examine the association of fatty acid intake and fracture risk in patients aged 65 years or older (80% women) suffering a low-energy fracture. Higher intakes of total PUFA and n-6 PUFA were reported to be associated with an elevated risk of fracture (p = .01 for the trend test), and a lower ratio of monounsaturated fatty acid:PUFA was associated with decreased fracture risk in these elderly subjects (Martinez-Ramirez et al., 2007). Most studies examining the relationships of dietary intake or tissue levels of n-3 and n-6 PUFA in cell cultures or animals report some aspect of potential or confirmed regulatory function of these fatty acids on bone metabolism. In human subjects, especially in young individuals when bone formation rates are higher compared with adults, n-3 PUFA or a low ratio of n-6:n-3 PUFA in blood is positively associated with BMD and a negative association between n-6 PUFA or ratio of n-6:n-3 and BMD. In contrast, however, a study of aged Chinese patients undergoing elective orthopedic surgery (Griffith et al., 2009) did not find the same relationship between marrow fatty acids and BMD. In this study, the fatty acid composition of tissue samples of marrow fat and subcutaneous fat from 126 subjects (98 females, 34 males, mean age 69.7 ± 10.5 years) showed no difference in marrow fatty acid composition between subjects of varying BMD (normal, low bone mass, and osteoporosis). This study is confounded for any dietary association that could have taken place during bone growth that occurred in childhood and adolescence and remodeling of bone in adults. Also, a one-time measurement at the time of disease onset and orthopedic surgery is more likely to indicate changes after disease progression. An analysis of diet history or comparison with subjects who consumed diets that varied in types of fats would have improved the study, and it is less likely to observe a difference when BMD of lumbar spine is compared with the fatty acid composition of marrow samples taken from the femur. Intervention studies in human subjects provide further support that n-3 PUFA intake is beneficial in reducing bone resorption and maintaining bone mass. Fish intake attenuated bone loss during long-term space flight, and higher n-3 PUFA intake led to lower N-telopeptide (NTx) urinary excretion during bed rest (Zwart et al., 2009). In a 6-week randomized 3-period crossover design study with 23 subjects, Griel et al. (2007) reported that supplementing diets with ALA from sources such as walnuts, walnut oil, and flaxseed oil significantly lowered the circulating NTx level (a reliable

Omega-3 Fatty Acids and Bone Metabolism

249

bone resorption marker), reflecting a protective effect of plant sources of dietary n-3 PUFA on bone metabolism via a decrease in bone resorption. Serum bone specific ALP activity, a bone formation marker, was not affected by the dietary treatments, which included an average American diet (control, ratio of n-6:n-3 PUFA = 9) and two high PUFA diets (one high in LA with a ratio of n-6:n-3 PUFA = 3.5 and the other high in ALA with a ratio of n-6:n-3 PUFA = 1.6).

Studies on n-3 PUFA Actions in Bone Cell Cultures To elucidate the mechanism of n-3 PUFA action on bone metabolism, numerous research projects have been performed using the two major bone cell types (osteoblasts and osteoclasts) to reveal how these fatty acids affect bone metabolism on both the cellular and the genetic levels. Several studies have reported findings on the positive actions of n-3 PUFA as well as the negative influences asserted by the n-6 PUFA when in dietary abundance, specifically AA, on various cell culture models of osteoblasts. In a human osteoblast-like cell culture model, investigators (Musacchio et al., 2007) showed that both EPA and oleic acid increased gene expression of type I collagen and fibronectin, whereas AA diminished bone cell adhesion. In a study by Shen and coresearchers (2008), mouse bone marrow stromal cells (ST-2) were treated with AA, EPA, or ethanol as a control. AA treatment resulted in the highest value for PGE2 production and elevated COX-2 mRNA expression in these ST-2 stromal cells. AA treatment also increased nitric oxide (NO) production in 7-day culture of these cells (early stage of osteoblastogenesis), whereas EPA showed a stronger stimulatory effect on NO production relative to AA treatment at near-mature osteoblast stages. The research implies that EPA could be supportive during early osteoblastogenesis through its suppressive actions on PGE2 and the NO pathways. Osteoblast-derived NO has been suggested to mediate the localized bone destruction in certain inflammatory bone diseases (Hukkanen et al., 1995). The process of osteoclastogenesis is known to be regulated, in part, by osteoblasts via the OPG/ RANKL signaling system. OPG is a decoy receptor for RANKL, both produced by osteoblasts, but if RANK on preosteoclasts does not bind to RANKL, osteoclastogenesis is interrupted. In a study with MC3T3-E1 osteoblast-like cells (Coetzee et al., 2007), AA (5–20 µg/mL) treatment inhibited OPG secretion while stimulating the production of RANKL; however, this effect was attenuated by pretreatment with the COX inhibitor indomethacin, implying that PGE2 is involved in the regulation of OPG/ RANKL balance. In another study, EPA and DHA were shown to inhibit PGE2-induced RANKL expression in MC3T3-E1 cells, implicating their antiosteoclastogenesis influence (Poulsen et al., 2008b). In cells not preinduced by PGE2, neither EPA nor DHA showed any effect on RANKL expression. Some of these relationships and the role of n-3 PUFA in osteoclastogenesis are presented in Figure 16.3. Many cell culture studies are showing a trend of distinct properties of n-3 and n-6 PUFA on osteoblast cell activity. However, because culture conditions are not consistent in the literature, some interpretation of data is difficult. As an example, one report that showed that ALP activity was inhibited when MC3T3-E1 osteoblast-like cells were exposed to either AA or DHA after shortand long-term exposure, although the mineralizing properties of these cells were not compromised under these conditions (Coetzee et al., 2009). Although cells were cultured for numerous days, and it is not clear if the high concentrations of fatty acids used in cell culture included albumin, the extended culture periods could have diminished osteoblast cell functions and protein synthesis because fatty acids (not bound to albumin) are toxic to cells in culture. More specific culturing conditions and physiological levels of fatty acids will help to resolve some of the inconsistent findings reported in the literature. In addition to its influence on osteoclasts via its actions on osteoblasts, n-3 PUFA may also act upon osteoclasts directly by controlling their inflammatory responses in various testing models. Osteoclasts are recruited and activated upon the stimulation from osteoclastogenic cytokines that are increased in inflammatory joint disease, which leads to eventual bone destruction. In a cell culture study using the murine monocytic cell line RAW 264.7 (commonly used as an osteoclast precursor cell line), DHA was demonstrated to be much more effective than EPA in reducing RANKL-induced

250

Diet, Nutrients, and Bone Health

proinflammatory responses by decreasing TNFα production and the expressions of nuclear factor kappa B (NFκB) and p38 mitogen-activated protein kinases, factors that regulate the responses to inflammatory stimuli (Rahman et al., 2008). These findings suggest a great potential for DHA in decreasing osteoclast activation and bone resorption. In an experiment using the same preosteoclastic murine cell line RAW264.7, Zwart et al. (2009) reported that EPA inhibited NFκB activity and protein expression in a routine stationary culture as well as in a simulated microgravity system by culturing cells in a bioreactor that was rotated on a high aspect vessel that provides an environment in which the cells are in a continuous state of free fall, mimicking the state of weightlessness.

Recent Animal Studies Supporting Benefits of n-3 PUFA on Bone In animal feeding studies, n-3 PUFA have been shown to act alone or in concert with other dietary factors, such as isoflavones, estrogen, boron, and inulin, a prebiotic, to promote bone health. The animal models adopted in these experiments are mostly mouse and rat, and occasionally, pig and chicken. Although pure n-3 fatty acids (ALA, EPA, and DHA) are the best forms to be used in these feeding experiments (void of any confounding factors that originate from the diverse source of these fatty acids), for practical and economical reasons, oils (either of vegetables or marine fish) are the most common sources of n-3 PUFA used in the diets given to animals. The effectiveness of various n-3 PUFA species, in particular, ALA, compared with the long chain, more biologically active EPA and DHA, is controversial because of its low conversion rate (0.2%) from ALA to EPA in healthy human subjects (Pawlosky et al., 2001). However, Sacco and coworkers demonstrated that ALA-rich flaxseed diet to OVX 3-month-old rats resulted in significantly higher levels of ALA and EPA and lower levels of LA, AA, and the ratio of n-6:n-3 PUFA compared with those of the control basal AIN-93M diet (Sacco et al., 2009). Furthermore, feeding these rats flaxseed (10% in the diet) plus a low-dose estrogen therapy (13 µg, 90-day release) led to the highest BMD and peak load at the lumbar vertebrae among the treatment groups; however, no such effect on BMD or bone strength was observed in the tibia and femur (Sacco et al., 2009). Another study (Poulsen et al., 2008a) found similar findings in an 18-week feeding experiment showing that dietary DHA (0.5 g/kg body weight/day) potentiated the mineral-sparing effect of 17β-estradiol (1 µg/day) in OVX rats. In either case, the combination treatment was more effective than either the n-3 PUFA or estrogen alone at maintaining bone mineral. Fish oils are rich sources of long-chain n-3 PUFA EPA and DHA and have been widely used in both human and animal studies on their effectiveness in promoting and maintaining bone health. Most recent studies emphasize the synergistic or additive actions of n-3 PUFA from fish oil together with other nutritional factors. Feeding rats fish oil plus the prebiotic inulin-type fructans enhanced mineral absorption and resulted in increased BMC and bone strength in the tibia (Lobo et al., 2009). Others (Nielsen and Stoecker, 2009) showed that rats (either adequate or deficient in boron intake) fed fish oil had mechanically superior bones compared with the bones of those given safflower oil in the diet. Feeding OVX CD-1 mice fish oil (7% menhaden oil in the diet) led to higher BMD in lumbar vertebra, whereas combining fish oil with isoflavones (250 mg of genistein + 250 mg of daidzein/kg diet) resulted in a higher peak load of lumbar vertebra 4, showing that fish oil can act alone or potentiate the bone-protective effect of isoflavones (Ward and Fonseca, 2007). Unlike any other animal model mentioned previously, the fat-1 mouse, a transgenic model that synthesizes n-3 PUFA from n-6 PUFA in its own tissue, provides a unique tool to illustrate how the same dietary source of fatty acids (safflower oil that contains about 70% LA and essentially void of any n-3 fatty acids) affects bone metabolism because of the different metabolic pathways the ingested fatty acids go through. In experiments with fat-1 mice and their wild-type counterparts (Lau et al., 2009a, 2009b), feeding an AIN-93 G diet containing 10% safflower oil from weaning until they were 12 weeks of age revealed that a lower ratio of n-6:n-3 PUFA was found in the vertebrae of fat-1 mice compared with the wild-type, and this ratio was negatively correlated with BMD and peak load in mechanical property testing, whereas total n-3 PUFA, including ALA, EPA, and

Omega-3 Fatty Acids and Bone Metabolism

251

DHA, were positively correlated with BMD and the peak load in the vertebrae (Lau et al., 2009b). Similar findings were reported for the femur, suggesting that n-3 PUFA have a favorable effect on mineral accumulation and functional measures of bone in young mice (Lau et al., 2009a). Differences have been established in how n-6 and n-3 PUFA affect bone metabolism in young modeling bone systems versus in older animals in which bone remodeling is the predominant process. In both animal and human studies, not only is n-3 PUFA (especially DHA) critical for bone mineral mass maintenance, but also AA is critical in bone mineral accretion during the rapid phase of growth that characterizes active bone modeling. Weiler et al. (2005) reported that whole-body BMC in human infants was positively predicted by cord RBC AA and that lumbar spine 1–4 and femur BMC was positively associated with cord RBC ratio of AA:EPA but negatively related to maternal DHA. This observation suggests the importance of a balanced maternal diet with respect to the n-6 and n-3 PUFA content of the diet. Furthermore, AA was found to be positively correlated with both total-body BMC and BMD in fast-growing Caucasian 8-year-old children (Eriksson et al., 2009). In animal studies, Kohut and coworkers reported that feeding low and very-low-birth-weight piglets a diet containing AA and DHA at a ratio of 6:1 (1.2 g AA + 0.2 g DHA/100 g dietary fat) for 15 days resulted in lowered bone resorption relative to the control group, whereas bone formation was not affected (Kohut et al., 2009).

Studies on n-3 PUFA in Models of Arthritis and Inflammation The beneficial effects of n-3 PUFA EPA and DHA in alleviating symptoms of rheumatoid arthritis (RA) reside not only in their ability to change the eicosanoid precursors to those which are less inflammatory, compared with those derived from AA, that is, PGE2, but also because both EPA and DHA give rise to resolvins that are anti-inflammatory and inflammation resolving (Calder, 2008). In addition, n-3 PUFA affect several other important aspects of immunity such as antigen presentation, T-cell reactivity, and inflammatory cytokine production that are important in RA development (Calder, 2008). Both intervention studies with n-3 long-chain PUFA and cross-sectional surveys support this notion. In a randomized double-blind placebo-controlled crossover study, 45 patients (43females and 2 males) were divided into two groups and were given a placebo or n-3 long-chain PUFA-supplemented dairy products. At the end of the supplementation period (2 × 12 weeks), the n-3 long-chain PUFA supplements lowered triacylglycerol, decreased LPS-stimulated COX-2 expression, and suppressed the immune response as lymphocytes and monocytes were found to be significantly decreased in subjects who consumed the enriched dairy products (Dawczynski et al., 2009). In rodent models of periodontitis, characterized by inflammation and infection of the ligaments and bones that support the teeth, n-3 PUFA have been shown to attenuate inflammation and minimize bone loss in the diseased area. In a feeding study using adult mice, diets containing 10% tuna oil resulted in significantly higher tissue n-3 PUFA concentrations in oral soft tissues compared with those in the control diet. Tuna oil also led to reductions in alveolar bone loss by more than 50% in mice with induced periodontitis compared with the control group that were fed a high n-6 PUFA Sunola oil (a sunflower oil with higher amount of monounsaturated fatty acids) diet, which exhibits an inverse relationship between alveolar bone loss and n-3 PUFA tissue levels (Bendyk et al., 2009). In a rat study using an LPS-induced periodontitis model, selective COX-2 inhibitor (Celecoxib), prophylactic n-3 PUFA, and a combination of these two agents inhibited gingival tissue MMP-8 expression, whereas the individual administration of therapeutic n-3 PUFA increased gingival TIMP-1 expression (Vardar-Sengul et al., 2008). Finally, one group (Hamazaki et al., 2006) reported a correlation between decreased tooth loss and EPA concentration in RBCs of humans.

CONCLUSIONS Benefits of n-3 PUFA on bone modeling and remodeling have been shown in a number of human (Table 16.1) and animal (Table 16.2) investigations. A consistent observation in many, but not all, of

252

Diet, Nutrients, and Bone Health

these in vivo studies is that a negative relationship exists between n-6 PUFA and bone formation or BMD. It is not clear if the positive benefits from n-3 PUFA result from a reduction in the potency of the n-6 PUFA, that is, their negative effects on bone formation and resorption. Based on decades of research, the following questions must be addressed about the role of n-3 PUFA and bone. What do we know after years of research on n-3 PUFA and bone biology? Early exposure of infants to DHA is vital for neural and retinal development, and in the case for children and young adults, DHA appears to support bone formation during modeling and the quality of mineral content. In mature adults, several positive relationships between bone quality and n-3 PUFA intakes are acknowledged, and in some cases, biomarkers of bone resorption are attenuated with the intake of n-3 PUFA. Can we explain or better define the differences between n-3 PUFA (ALA, DHA, and EPA) and their individual actions on bone remodeling and modeling? Although an answer to this question is not yet possible, experimental data do support a role for DHA in promoting early bone growth, based on the positive observations in animals and what limited data are available for children. Certainly, the dietary studies in human subjects make a positive correlation between n-3 PUFA and bone quality measurements, and the data in animal studies suggest a biological basis for DHA in certain bone compartments and for bone formation. In contrast, some data in animals indicate that EPA is a potent n-3 PUFA that moderates the actions of AA and prostanoid production in bone that support osteoclastic activity (biochemical and gene expression). So, DHA and EPA appear to have different, but supporting, roles in bone metabolism throughout the life cycle. Also, an important balance between DHA and AA is vital to bone formation in the young so that replacing all n-6 PUFA with n-3 PUFA is not appropriate. The most optimal dietary ranges of n-6 and n-3 PUFA at the different stages of the life cycle remain to be determined. What is the translational impact of the research on n-3 PUFA and bone? With convincing evidence that long-chain n-3 PUFA reduce cardiovascular disease, that DHA is essential for neural and retina development in the infant, and that positive associations of n-3 PUFA enhance bone health, a key role is emerging for the benefits of n-3 PUFA in bone metabolism. Future research with well-characterized test mixtures of PUFA and appropriate protocols for bone measurement endpoints should facilitate a better understanding of n-3 PUFA actions on bone in human subjects. Investigations of how EPA and DHA affect gene expression in both osteoblastogenesis and osteoclastogenesis are needed to advance the field. Discerning the relationships between the long-chain n-3 PUFA and eicosanoid production and the downstream signaling for these compounds may be critical for improving bone health. A final direction for this research is the study of n-3 PUFA actions in the endocannabinoid signaling pathways that impact muscle and bone physiology.

REFERENCES Bab, I., Ofek, O., Tam, J., et al. 2008. Endocannabinoids and the regulation of bone metabolism. JNeuroendocrinol 20 Suppl 1:69–74. Bab, I., and Zimmer, A. 2008. Cannabinoid receptors and the regulation of bone mass. Br J Pharmacol 153:182–8. Baggio, B., Budakovic, A., Ferraro, A., et al. 2005. Relationship between plasma phospholipid polyunsaturated fatty acid composition and bone disease in renal transplantation. Transplantation 80:1349–52. Barreiro, E., de la Puente, B., Busquets, S., et al. 2005. Both oxidative and nitrosative stress are associated with muscle wasting in tumour-bearing rats. FEBS Lett 579:1646–52. Bartoccioni, E., Michaelis, D., and Hohlfeld, R. 1994. Constitutive and cytokine-induced production of interleukin-6 by human myoblasts. Immunol Lett 42:135–8. Belec, L., Authier, F. J., Chazaud, B., et al. 1997. Interleukin (IL)-1 beta and IL-1 beta mRNA expression in normal and diseased skeletal muscle assessed by immunocytochemistry, immunoblotting and reverse transcriptase-nested polymerase chain reaction. J Neuropathol Exp Neurol 56:651–63. Bendyk, A., Marino, V., Zilm, P. S., et al. 2009. Effect of dietary omega-3 polyunsaturated fatty acids on experimental periodontitis in the mouse. J Periodontal Res 44:211–6.

Omega-3 Fatty Acids and Bone Metabolism

253

Blackwell, K. A., Hortschansky, P., Sanovic, S., et al. 2009. Bone morphogenetic protein 2 enhances PGE(2)stimulated osteoclast formation in murine bone marrow cultures. Prostaglandins Other Lipid Mediat 90:76–80. Calder, P. C. 2008. Session 3: Joint Nutrition Society and Irish Nutrition and Dietetic Institute Symposium on ‘Nutrition and autoimmune disease’ PUFA, inflammatory processes and rheumatoid arthritis. Proc Nutr Soc 67:409–18. Carbo, N., Lopez-Soriano, J., Costelli, P. et al. 2000. Interleukin-15 antagonizes muscle protein waste in tumour-bearing rats. Br J Cancer 83:526–31. Carmeli, E., Coleman, R., and Reznick, A. Z. 2002. The biochemistry of aging muscle. Exp Geront 37:477–89. Coetzee, M., Haag, M., and Kruger, M. C. 2007. Effects of arachidonic acid, docosahexaenoic acid, prostaglandin E(2) and parathyroid hormone on osteoprotegerin and RANKL secretion by MC3T3-E1 osteoblastlike cells. J Nutr Biochem 18:54–63. Coetzee, M., Haag, M., and Kruger, M. C. 2009. Effects of arachidonic acid and docosahexaenoic acid on differentiation and mineralization of MC3T3-E1 osteoblast-like cells. Cell Biochem Funct 27:3–11. Dawczynski, C., Schubert, R., Hein, G., et al. 2009. Long-term moderate intervention with n-3 long-chain PUFA-supplemented dairy products: Effects on pathophysiological biomarkers in patients with rheumatoid arthritis. Br J Nutr 101:1517–26. De Bleecker, J. L., Meire, V. I., Declercq, W., et al. 1999. Immunolocalization of tumor necrosis factor-alpha and its receptors in inflammatory myopathies. Neuromuscul Disord 9:239–46. Eckardt, K., Sell, H., and Eckel, J. 2008. Novel aspects of adipocyte-induced skeletal muscle insulin resistance. Archives Physiol Biochem 114:287–98. Ehrlich, P. J., and Lanyon, L. E. 2002. Mechanical strain and bone cell function: A review. Osteoporos Int 13:688–700. Epker, B. N., and Frost, H. M. 1965. Correlation of bone resorption and formation with the physical behavior of loaded bone. J Dental Res 44:33–41. Eriksson, S., Mellstrom, D., and Strandvik, B. 2009. Fatty acid pattern in serum is associated with bone mineralisation in healthy 8-year-old children. Br J Nutr 102:407–12. Fernandes, G., Lawrence, R., and Sun, D. 2003. Protective role of n-3 lipids and soy protein in osteoporosis. Prostaglandins Leukot Essent Fatty Acids 68:361–72. Fidler, N., and Koletzko, B. 2000. The fatty acid composition of human colostrum. Eur J Nutr 39:31–7. Fricke, O., and Schoenau, E. 2007. The ‘functional muscle–bone unit’: Probing the relevance of mechanical signals for bone development in children and adolescents. Growth Horm Igf Res 17:1–9. Frost, H. M. 1973. Bone Modeling and Skeletal Modeling Errors, vol. IV. Springfield, IL, Charles C. Thomas. Frost, H. M. 2003a. Bone’s mechanostat: A 2003 update. Anat Rec A Discov Mol Cell Evol Biol 275:1081–101. Frost, H. M. 2003b. On the pathogenesis of osteogenesis imperfecta: Some insights of the Utah paradigm of skeletal physiology. J Musculoskel Neuron Interact 3:1–7. Garg, A. K., and Aggarwal, B. B. 2002. Reactive oxygen intermediates in TNF signaling. Mol Immunol 39:509–17. Giangregorio, L., and McCartney, N. 2006. Bone loss and muscle atrophy in spinal cord injury: Epidemiology, fracture prediction, and rehabilitation strategies. J Spinal Cord Med 29:489–500. Godfrey, K. M., and Barker, D. J. 2000. Fetal nutrition and adult disease. Am J Clin Nutr 71:1344S–52S. Godfrey, K. M., and Barker, D. J. 2001. Fetal programming and adult health. Public Health Nutr 4:611–24. Gomes-Marcondes, M. C., and Tisdale, M. J. 2002. Induction of protein catabolism and the ubiquitin–proteasome pathway by mild oxidative stress. Cancer Lett 180:69–74. Goossens, V., Grooten, J., De Vos, K., et al. 1995. Direct evidence for tumor necrosis factor-induced mitochondrial reactive oxygen intermediates and their involvement in cytotoxicity. Proc Natl Acad Sci USA 92:8115–9. Gray, B., Hsu, J. D., and Furumasu, J. 1992. Fractures caused by falling from a wheelchair in patients with neuromuscular disease. Dev Med Child Neurol 34:589–92. Griel, A. E., Kris-Etherton, P. M., Hilpert, K. F., et al. 2007. An increase in dietary n-3 fatty acids decreases a marker of bone resorption in humans. Nutr J 6:2. Griffith, J. F., Yeung, D. K., Ahuja, A. T., et al. 2009. A study of bone marrow and subcutaneous fatty acid composition in subjects of varying bone mineral density. Bone 44:1092–6. Grimble, R. F., and Tappia, P. S. 1998. Modulation of pro-inflammatory cytokine biology by unsaturated fatty acids. Z Ernahrungswiss 37:57–65. Gronowitz, E., Lorentzon, M., Ohlsson, C., et al. 2008. Docosahexaenoic acid is associated with endosteal circumference in long bones in young males with cystic fibrosis. Br J Nutr 99:160–7.

254

Diet, Nutrients, and Bone Health

Gronowitz, E., Mellstrom, D., and Strandvik, B. 2006. Serum phospholipid fatty acid pattern is associated with bone mineral density in children, but not adults, with cystic fibrosis. Br J Nutr 95:1159–65. Haag, M., Magada, O. N., Claassen, N., et al. 2003. Omega-3 fatty acids modulate ATPases involved in duodenal Ca absorption. Prostaglandins Leukot Essent Fatty Acids 68:423–9. Hamazaki, K., Itomura, M., Sawazaki, S., et al. 2006. Fish oil reduces tooth loss mainly through its antiinflammatory effects? Med Hypotheses 67:868–70. Harrison, A. P., Rowlerson, A. M., and Dauncey, M. J. 1996. Selective regulation of myofiber differentiation by energy status during postnatal development. Am J Physiol 270:667–74. Hogstrom, M., Nordstrom, P., and Nordstrom, A. 2007. n-3 Fatty acids are positively associated with peak bone mineral density and bone accrual in healthy men: The NO2 Study. Am J Clin Nutr 85:803–7. Hukkanen, M., Hughes, F. J., Buttery, L. D., et al. 1995. Cytokine-stimulated expression of inducible nitric oxide synthase by mouse, rat, and human osteoblast-like cells and its functional role in osteoblast metabolic activity. Endocrinology 136:5445–53. Hutchins, H. L., Zhao, D., Hannon, K. M., et al. 2009. Culture duration and PUFA treatment influence expression of endocannabinoid proteins in MC3T3-E1 osteoblast-like cells. FASEB J 23:543. Idris, A. I., Sophocleous, A., Landao-Bassonga, E., et al. 2008. Regulation of bone mass, osteoclast function, and ovariectomy-induced bone loss by the type 2 cannabinoid receptor. Endocrinology 149:5619–26. Idris, A. I., van’t Hof, R. J., Greig, I. R., et al. 2005. Regulation of bone mass, bone loss and osteoclast activity by cannabinoid receptors. Nat Med 11:774–9. Institute of Medicine. 2002. Dietary References Intakes for Energy, Carbohydrate, Fiber, Fat, Fatty Acids, Cholesterol, Protein, and Amino Acids. Washington, DC, The National Academies Press. Judex, S., Boyd, S., Qin, Y., et al. 2003. Combining high-resolution micro-computed tomography with material composition to define the quality of bone tissue. Curr Osteoporos Rep 1:11–9. Kang, J. X., and Weylandt, K. H. 2008. Modulation of inflammatory cytokines by omega-3 fatty acids. Subcell Biochem 49:133–43. Khal, J., and Tisdale, M. J. 2008. Downregulation of muscle protein degradation in sepsis by eicosapentaenoic acid (EPA). Biochem Biophys Res Commun 375:238–40. Kohut, J., Watkins, B., and Weiler, H. 2009. Enhanced lumbar spine bone mineral content in piglets fed arachidonic acid and docosahexaenoic acid is modulated by severity of growth restriction. Br J Nutr 102:1117–20. Kondo, H., Nakagaki, I., Sasaki, S., et al. 1993. Mechanism of oxidative stress in skeletal muscle atrophied by immobilization. Am J Physiol 265:839–44. Korotkova, M., Gabrielsson, B. G., Holmang, A., et al. 2005. Gender-related long-term effects in adult rats by perinatal dietary ratio of n-6/n-3 fatty acids. Am J Physiol Regul Integr Comp Physiol 288:575–9. Korotkova, M., Ohlsson, C., Hanson, L. A., et al. 2004. Dietary n-6:n-3 fatty acid ratio in the perinatal period affects bone parameters in adult female rats. Br J Nutr 92:643–8. Kris-Etherton, P. M., Taylor, D. S., Yu-Poth, S., et al. 2000. Polyunsaturated fatty acids in the food chain in the United States. Am J Clin Nutr 71:179S–88S. Lau, B. Y., Ward, W. E., Kang, J. X., et al. 2009a. Femur EPA and DHA are correlated with femur biomechanical strength in young fat-1 mice. J Nutr Biochem 20:453–61. Lau, B. Y., Ward, W. E., Kang, J. X., et al. 2009b. Vertebrae of developing fat-1 mice have greater strength and lower n-6/n-3 fatty acid ratio. Exp Biol Med (Maywood) 234:632–8. Levine, B., Kalman, J., Mayer, L., et al. 1990. Elevated circulating levels of tumor necrosis factor in severe chronic heart failure. N Engl J Med 323:236–41. Li, X., Moody, M. R., Engel, D., et al. 2000. Cardiac-specific overexpression of tumor necrosis factor-alpha causes oxidative stress and contractile dysfunction in mouse diaphragm. Circulation 102:1690–6. Li, X., Pilbeam, C. C., Pan, L., et al. 2002. Effects of prostaglandin E2 on gene expression in primary osteoblastic cells from prostaglandin receptor knockout mice. Bone 30:567–73. Li, Y., Greiner, R. S., Salem, N., Jr., et al. 2003a. Impact of dietary n-3 FA deficiency on rat bone tissue FA composition. Lipids 38:683–6. Li, Y. P., Chen, Y., John, J., et al. 2005. TNF-alpha acts via p38 MAPK to stimulate expression of the ubiquitin ligase atrogin1/MAFbx in skeletal muscle. Faseb J 19:362–70. Li, Y. P., Lecker, S. H., Chen, Y., et al. 2003b. TNF-alpha increases ubiquitin-conjugating activity in skeletal muscle by up-regulating UbcH2/E220k. Faseb J 17:1048–57. Lingam, S., and Joester, J. 1994. Spontaneous fractures in children and adolescents with cerebral palsy. BMJ 309:265. Lobo, A. R., Filho, J. M., Alvares, E. P., et al. 2009. Effects of dietary lipid composition and inulin-type fructans on mineral bioavailability in growing rats. Nutrition 25:216–25.

Omega-3 Fatty Acids and Bone Metabolism

255

Martinez-Ramirez, M. J., Palma, S., Martinez-Gonzalez, M. A., et al. 2007. Dietary fat intake and the risk of osteoporotic fractures in the elderly. Eur J Clin Nutr 61:1114–20. Matias, I., Carta, G., Murru, E., et al. 2008. Effect of polyunsaturated fatty acids on endocannabinoid and N-acyl-ethanolamine levels in mouse adipocytes. Biochim Biophys Acta 1781:52–60. Mori, T. A., and Beilin, L. J. 2001. Long-chain omega 3 fatty acids, blood lipids and cardiovascular risk reduction. Curr Opin Lipidol 12:11–7. Musacchio, E., Priante, G., Budakovic, A., et al. 2007. Effects of unsaturated free fatty acids on adhesion and on gene expression of extracellular matrix macromolecules in human osteoblast-like cell cultures. Connect Tissue Res 48:34–8. Nielsen, F. H., and Stoecker, B. J. 2009. Boron and fish oil have different beneficial effects on strength and trabecular microarchitecture of bone. J Trace Elem Med Biol 23:195–203. Ofek, O., Karsak, M., Leclerc, N., et al. 2006. Peripheral cannabinoid receptor, CB2, regulates bone mass. Proc Natl Acad Sci USA 103:696–701. Ozdurak, R. H., Duz, S., Arsal, G., et al. 2003. Quantitative forearm muscle strength influences radial bone mineral density in osteoporotic and healthy males. Technol Health Care 11:253–61. Pawlosky, R. J., Hibbeln, J. R., Novotny, J. A., et al. 2001. Physiological compartmental analysis of alphalinolenic acid metabolism in adult humans. J Lipid Res 42:1257–65. Poulsen, R. C., Firth, E. C., Rogers, C. W., et al. 2007. Specific effects of gamma-linolenic, eicosapentaenoic, and docosahexaenoic ethyl esters on bone post-ovariectomy in rats. Calcif Tissue Int 81:459–71. Poulsen, R. C., Moughan, P. J., and Kruger, M. C. 2008a. Docosahexaenoic acid and 17 beta-estradiol cotreatment is more effective than 17 beta-estradiol alone in maintaining bone post-ovariectomy. Exp Biol Med (Maywood) 233:592–602. Poulsen, R. C., Wolber, F. M., Moughan, P. J., et al. 2008b. Long chain polyunsaturated fatty acids alter membrane-bound RANK-L expression and osteoprotegerin secretion by MC3T3-E1 osteoblast-like cells. Prostaglandins Other Lipid Mediat 85:42–8. Quinn, L. S., Haugk, K. L., and Grabstein, K. H. 1995. Interleukin-15: A novel anabolic cytokine for skeletal muscle. Endocrinology 136:3669–72. Rahman, M. M., Bhattacharya, A., and Fernandes, G. 2008. Docosahexaenoic acid is more potent inhibitor of osteoclast differentiation in RAW 264.7 cells than eicosapentaenoic acid. J Cell Physiol 214:201–9. Raisz, L. G., Alander, C. B., and Simmons, H. A. 1989. Effects of prostaglandin E3 and eicosapentaenoic acid on rat bone in organ culture. Prostaglandins 37:615–25. Rauch, F., Bailey, D. A., Baxter-Jones, A., et al. 2004. The ‘muscle–bone unit’ during the pubertal growth spurt. Bone 34:771–5. Reinwald, S., Li, Y., Moriguchi, T., et al. 2004. Repletion with (n-3) fatty acids reverses bone structural deficits in (n-3)-deficient rats. J Nutr 134:388–94. Rousseau, J. H., Kleppinger, A., and Kenny, A. M. 2009. Self-reported dietary intake of omega-3 fatty acids and association with bone and lower extremity function. J Am Geriatr Soc 57:1781–8. Sacco, S. M., Jiang, J. M., Reza-Lopez, S., et al. 2009. Flaxseed combined with low-dose estrogen therapy preserves bone tissue in ovariectomized rats. Menopause 16:545–54. Saghizadeh, M., Ong, J. M., Garvey, W. T., et al. 1996. The expression of TNF alpha by human muscle. Relationship to insulin resistance. J Clin Invest 97:1111–6. Sayer, A. A., and Cooper, C. 2005. Fetal programming of body composition and musculoskeletal development. Early Hum Dev 81:735–44. Sayer, A. A., Syddall, H. E., Gilbody, H. J., et al. 2004. Does sarcopenia originate in early life? Findings from the Hertfordshire cohort study. J Gerontol A Biol Sci Med Sci 59:930–4. Schoenau, E., and Frost, H. M. 2002. The “muscle–bone unit” in children and adolescents. Calcif Tissue Int 70:405–7. Schulze, P. C., Gielen, S., Adams, V., et al. 2003. Muscular levels of proinflammatory cytokines correlate with a reduced expression of insulinlike growth factor-I in chronic heart failure. Basic Res Cardiol 98:267–74. Scutt, A., and Williamson, E. M. 2007. Cannabinoids stimulate fibroblastic colony formation by bone marrow cells indirectly via CB2 receptors. Calcif Tissue Int 80:50–9. Shen, C. L., Peterson, J., Tatum, O. L., et al. 2008. Effect of long-chain n-3 polyunsaturated fatty acid on inflammation mediators during osteoblastogenesis. J Med Food 11:105–10. Shen, C. L., Yeh, J. K., Rasty, J., et al. 2006. Protective effect of dietary long-chain n-3 polyunsaturated fatty acids on bone loss in gonad-intact middle-aged male rats. Br J Nutr 95:462–8. Shen, C. L., Yeh, J. K., Rasty, J., et al. 2007. Improvement of bone quality in gonad-intact middle-aged male rats by long-chain n-3 polyunsaturated fatty acid. Calcif Tissue Int 80:286–93.

256

Diet, Nutrients, and Bone Health

Singhal, A., Cole, T. J., and Lucas, A. 2001. Early nutrition in preterm infants and later blood pressure: Two cohorts after randomised trials. Lancet 357:413–9. Smith, H. J., Greenberg, N. A., and Tisdale, M. J. 2004. Effect of eicosapentaenoic acid, protein and amino acids on protein synthesis and degradation in skeletal muscle of cachectic mice. Br J Cancer 91:408–12. Spate, U., and Schulze, P. C. 2004. Proinflammatory cytokines and skeletal muscle. Curr Opin Clin Nutr Metab Care 7:265–9. Spiers, S., McArdle, F., and Jackson, M. J. 2000. Aging-related muscle dysfunction. Failure of adaptation to oxidative stress? Ann N Y Acad Sci 908:341–3. Suda, K., Udagawa, N., Sato, N., et al. 2004. Suppression of osteoprotegerin expression by prostaglandin E2 is crucially involved in lipopolysaccharide-induced osteoclast formation. J Immunol 172:2504–10. Tam, J., Ofek, O., Fride, E., et al. 2006. Involvement of neuronal cannabinoid receptor CB1 in regulation of bone mass and bone remodeling. Mol Pharmacol 70:786–92. Tam, J., Trembovler, V., di Marzo, V., et al. 2008. The cannabinoid CB1 receptor regulates bone formation by modulating adrenergic signaling. FASEB J 22:285–94. Taylor, D. J., Thompson, C. H., Kemp, G. J. et al. 1995. A relationship between impaired fetal growth and reduced muscle glycolysis revealed by 31P magnetic resonance spectroscopy. Diabetologia 38:1205–12. Tisdale, M. J. 1996. Inhibition of lipolysis and muscle protein degradation by EPA in cancer cachexia. Nutrition 12:S31–33. Tisdale, M. J. 2001. Loss of skeletal muscle in cancer: Biochemical mechanisms. Front Biosci 6:164–74. Tisdale, M. J. 2002. Cachexia in cancer patients. Nat Rev Cancer 2:862–71. Tisdale, M. J., and Dhesi, J. K. 1990. Inhibition of weight loss by omega-3 fatty acids in an experimental cachexia model. Cancer Res 50:5022–6. Uauy, R., Hoffman, D. R., Peirano, P., et al. 2001. Essential fatty acids in visual and brain development. Lipids 36:885–95. Valdez, H., and Lederman, M. M. 1997. Cytokines and cytokine therapies in HIV infection. AIDS Clin Rev187–228. Vardar-Sengul, S., Buduneli, E., Turkoglu, O., et al. 2008. The effects of selective COX-2 inhibitor/celecoxib and omega-3 fatty acid on matrix metalloproteinases, TIMP-1, and laminin-5gamma2-chain immunolocalization in experimental periodontitis. J Periodontol 79:1934–41. Wang, Y., Wluka, A. E., Hodge, A. M., et al. 2008. Effect of fatty acids on bone marrow lesions and knee cartilage in healthy, middle-aged subjects without clinical knee osteoarthritis. Osteoarthritis Cartilage 16:579–83. Ward, W. E., and Fonseca, D. 2007. Soy isoflavones and fatty acids: Effects on bone tissue postovariectomy in mice. Mol Nutr Food Res 51:824–31. Warren, G. L., Hayes, D. A., Lowe, D. A., et al. 1994. Eccentric contraction-induced injury in normal and hindlimb-suspended mouse soleus and EDL muscles. J Appl Physiol 77:1421–30. Watkins, B. A., Jones, R., Li, Y., et al. 2007. Dietary long chain n-3 PUFA attenuates musculoskeletal atrophy associated with disuse in mice. FASEB J 21:A728. Watkins, B. A., Li, Y., Allen, K. G. D., et al. 2000. Dietary ratio of (n-6)/(n-3) polyunsaturated fatty acids alters the fatty acid composition of bone compartments and biomarkers of bone formation in rats. J Nutr 130:2274–84. Watkins, B. A., Li, Y., Lippman, H. E., et al. 2003. Modulatory effect of omega-3 polyunsaturated fatty acids on osteoblast function and bone metabolism. Prostaglandins Leukot Essent Fatty Acids 68:387–98. Watkins, B. A., Li, Y., and Seifert, M. F. 2006. Dietary ratio of n-6/n-3 PUFAs and docosahexaenoic acid: Actions on bone mineral and serum biomarkers in ovariectomized rats. J Nutr Biochem 17:282–9. Watkins, B. A., Lippman, H. E., Le Bouteiller, L., et al. 2001. Bioactive fatty acids: Role in bone biology and bone cell function. Prog Lipid Res 40:125–48. Watkins, B. A., Reinwald, S., Li, Y., et al. 2005. Protective actions of soy isoflavones and n-3 PUFA on bone mass in ovariectomized rats. J Nutr Biochem 16:479–88. Weiler, H., Fitzpatrick-Wong, S., Schellenberg, J., et al. 2005. Maternal and cord blood long-chain polyunsaturated fatty acids are predictive of bone mass at birth in healthy term-born infants. Pediatr Res 58:1254–8. Weisinger, H. S., Armitage, J. A., Sinclair, A. J., et al. 2001. Perinatal omega-3 fatty acid deficiency affects blood pressure later in life. Nat Med 7:258–9. Weiss, L. A., Barrett-Connor, E., and von Muhlen, D. 2005. Ratio of n-6 to n-3 fatty acids and bone mineral density in older adults: The Rancho Bernardo Study. Am J Clin Nutr 81:934–8. Wohl, G. R., Loehrke, L., Watkins, B. A., et al. 1998. Effects of high-fat diet on mature bone mineral content, structure, and mechanical properties. Calcif Tissue Int 63:74–9.

Omega-3 Fatty Acids and Bone Metabolism

257

Yamada, Y., Fushimi, H., Inoue, T., et al. 1995. Effect of eicosapentaenoic acid and docosahexaenoic acid on diabetic osteopenia. Diabetes Res Clin Pract 30:37–42. Zwart, S. R., Pierson, D., Mehta, S., et al. 2009. Capacity of omega-3 fatty acids or eicosapentaenoic acid to counteract weightlessness-induced bone loss by inhibiting NF-kappaB activation: From cells to bed rest to astronauts. J Bone Miner Res 10.1359/JBMR091041.

17

Is There a Role for Dietary Potassium in Bone Health? Susan Joyce Whiting

CONTENTS Introduction..................................................................................................................................... 259 Effects of Potassium on Calcium Metabolism................................................................................260 Observational Studies of Potassium Intake and Bone Health....................................................260 The Paleolithic Diet Implicates a High Potassium Diet............................................................. 262 Role of the Anion in Potassium Studies.......................................................................................... 262 Studies Using Different Potassium Salts.................................................................................... 262 The Alkaline Potassium Hypothesis..........................................................................................264 Can Potassium Be Replaced by Other Cations?........................................................................264 Potassium’s Role in Bone Examined.............................................................................................. 265 The AI Value for Potassium....................................................................................................... 265 Potassium Salts versus Fruits and Vegetables............................................................................266 Does Potassium Intake Alone Predict Bone Health?.................................................................266 Conclusions..................................................................................................................................... 267 References.......................................................................................................................................268

INTRODUCTION This chapter explores the question of the role of potassium in bone health, despite the specific reference to such an effect used in setting the current dietary reference intake recommendation for potassium by the Institute of Medicine (2005). Potassium has been associated with bone health (e.g., Lemann et al., 1989, 1991; Green and Whiting, 1994; Tucker et al., 1999; New et al., 2000; Jones etal., 2001; Sebastian et al., 2006) through its contribution to dietary alkalinity. Potassium is ingested mainly as a component of fruits and vegetables, but it is also found in lesser amounts in dairy foods, meats, and grains, especially whole grains. In Table 17.1, selected food sources of potassium in various food groups are shown. Unlike several nutrients that affect bone health directly through incorporation in bone mineral, such as calcium, phosphorus, magnesium, and fluoride, or through beneficial effects in bone cells, such as vitamins A, D, and K, potassium is classified as a nutrient which indirectly promotes bone health. Thus, uncertainty lies in whether a specific direct hypocalciuric effect of potassium exists or whether potassium serves primarily as a carrier for anions that promote bone health and also may possibly reduce sodium intake, assuming that potassium is consumed in its stead. In the former role, optimal potassium intake, but only as an alkaline salt, promotes the retention of calcium through countering the adverse effects of a mild metabolic acidosis that stimulates bone resorption. The latter role suggests that potassium itself plays no role and that several other cations (except sodium) are just as effective. This hypothesis is explored in this chapter in the context of bone health. 259

260

Diet, Nutrients, and Bone Health

TABLE 17.1 Intakes of Nutrients Important for Bone Health according to Serving Sizes of Foods of the DASH Diet DASH Diet Food Groups (Minimum (Examples of a Serving) Servings/Day) Milk products 2 • Milk, 1% (250mL—1 cup) • Cheese (50 g) Grain products 7 • Bread (1 slice) • Cereal (30 g—1 cup) • Rice (1 cup) Vegetable group 4 (raw leafy vegetable) • Lettuce (1 cup) • Spinach (1 cup) Fruit group 4 • Banana (medium size) • Orange (medium size) • Orange juice (1/2cup) Meat 2 or less • Lean meat (80 g) • Fish (80 g) • Poultry (80 g) Alternatives 0.6 (4 servings / • Egg (1) week from nuts, • Cooked dry seeds, and dry beans(125 g) beans) • Tofu (100 g) • Peanut butter (2tbsp—30mL) Total –

Approximate Calcium Intake (NotFortified)(mg)

Approximate Approximate Sodium Potassium Intake(mg) Intake (mg)

Approximate Protein Intake(g)

575

720

520

17

160

950

560

21

200

100

970

6

95

10

1610

4

50

135

550

19

30

35

180

8

1110

1950a

4390

75

Source: Modified from Institute of Medicine. 2005. Dietary Reference Intakes: Sodium, Chloride, Potassium and Sulphate. National Academy Press, Washington, DC. Notes: Calcium, potassium, and protein almost meet or exceed current recommendations, whereas sodium intake falls below its upper level. DASH diet = Dietary Approaches to Stop Hypertension diet.        a Consuming unsalted products reduces sodium intake further.

EFFECTS OF POTASSIUM ON CALCIUM METABOLISM Observational Studies of Potassium Intake and Bone Health A role for dietary vegetables and fruit on bone health emerged in the literature. All age groups were implicated. For adults, several population-based studies reported that increased potassium intake through vegetables and fruits is associated with increased bone mineral density (BMD) (Tucker et al., 1999; New, 2003). Studies by New et al. (1997, 2000) of premenopausal British

Is There a Role for Dietary Potassium in Bone Health?

261

women found potassium as well as current and past intakes of fruit or milk to be positively associated with BMD. In the cross-sectional analyses performed by New and colleagues, generally, there were significant correlations between potassium intake (determined by food frequency questionnaire) and BMD at several skeletal sites, in contrast to calcium intake where a significant correlation was found at only one site. This and other findings indicated that fruit and vegetable intake was an important determinant for bone, possibly due to changes in acid–base status. Tucker et al. (1999) provided an epidemiologic examination of fruit and vegetable intake, as well as potassium content of the diet, on BMD. They used a cross-sectional as well as a prospective design, utilizing Framingham data from the 20th and 22nd visits. The age group was 69 to 97 years, and both men and women were included. Cross-sectional analyses showed that a greater potassium intake, a greater magnesium intake, as well as fruit and vegetable intake were significantly associated with greater BMD. The positive skeletal effects were observed at more bone sites for potassium, and magnesium intake was highly related to potassium intake (as their food sources are similar). Rates of change in bone loss were also examined. Potassium, magnesium, and fruit and vegetable intakes were related to a slower bone loss, but only in men. Thus, both cross-sectional and prospective research approaches provide strong evidence for a positive effect of potassium on BMD from fruit and vegetable intake. A dietary plan promoting fruit and vegetable intake, the DASH (Dietary Approaches to Stop Hypertension) diet, emphasizes intake of vegetables, fruits, and low-fat diary products, and it avoids the consumption of processed foods (Table 17.1). A 3-month trial among 186 middle-aged men and women (Lin et al., 2003) reported that the DASH diet significantly reduced bone turnover, indicating that the DASH diet has beneficial effects on bone health (Doyle and Cashman, 2004). One disadvantage of using the DASH diet as evidence for a potassium effect is that the DASH diet incorporates many important healthful changes, that is, meeting calcium and protein recommendations (1000 mg and > 75 g, respectively) and keeping sodium intakes below the upper level of 2300 mg. High dietary sodium intakes cause hypercalciuria, which, in the absence of compensatory increases in enteric calcium absorption, would lead to negative calcium balance and loss of skeletal calcium. Overall, a diet rich in fruits and vegetables also supplies magnesium, vitamin C, polyphenols, and other plant compounds, which play roles in bone metabolism. As such, potassium may represent a marker for the beneficial components of such a healthful diet. Alternatively, potassium may be a marker for an alkaline-DASH diet, which has also been associated with both skeletal and cardiovascular health. A recent study corroborates this dietary effect of increasing fruits and vegetables, thus improving potassium intakes. McTiernan and coworkers (2009) examined whether dietary advice to improve fruit and vegetable intake could improve indices of bone, that is, BMD, and falls. In the Women’s Health Initiative Dietary Modification study, a low-fat and increased fruit, vegetable, and grain educational intervention in close to 50,000 postmenopausal women was evaluated with respect to incident hip, other site-specific, and total fractures and self-reported falls, and, in a subset, BMD. After an 8-year follow-up, the intervention group had a lower rate of reporting two or more falls than that of the comparison group. Thus, a dietary intervention which promoted increased fruit and vegetable consumption modestly reduced the risk of multiple falls and slightly lowered hip BMD of women. In children, Jones et al. (2001) first reported cross-sectional data that showed a positive link between potassium from fruit and vegetable consumption and BMD in 10-year-old girls. Tylavsky etal. (2004), studying girls aged 8 to 13 years, also found a positive relationship of fruit and vegetable consumption to bone area and BMD. McGartland et al. (2004) examined whether usual intake of fruit and vegetable influenced BMD in boys and girls aged 12 and 15 years; they found a significantly higher heel BMD only in 12-year-old girls who consumed high amounts of fruit. In contrast, however, vegetable and fruit intake had a significant independent effect on total body bone mineral content (BMC), development in boys but not girls (Vatanparast et al., 2005). In the latter study, it was predicted that total body (BMC) could be increased by about 50 g in a boy who had 10 servings/day intake of vegetable and fruit compared with a boy who had only one serving/day of

262

Diet, Nutrients, and Bone Health

these plant sources. Together, these studies suggest that both boys and girls likely benefit from the added fruit and vegetable intake, and potentially a lifetime beneficial effect may result.

The Paleolithic Diet Implicates a High Potassium Diet The diet of Paleolithic humans has been proposed as a model for the amounts and types of foods and nutrients that humans have evolved to consume (Eaton, 2006). Many differences exist between that diet, which is high in plant-based foods and meat but low in grains and dairy, compared with modern diets. Relevant to the potassium hypothesis, the Paleolithic diet was high in potassium, high in bicarbonate-producing anions, extremely low in sodium (<5 mEq/day), yet high in protein (Sebastian et al., 2006). The Paleolithic diet is calculated to have a net base urinary production that induces a low-grade metabolic alkalosis, in contrast to the modern diet which is net acid producing and may generate a low-grade metabolic acidosis (Frassetto et al., 2008). For most of the past 200,000 years, hominids and modern humans have consumed the diet of hunter-gatherers; uncultivated plant food sources, intermittent animal sources of protein, and rarely cereal grains or legumes. With the advent of the agricultural age approximately 10,000 years ago, cereal-based diets became established, along with the ingestion of protein from domesticated animals. It is generally believed that the humans from earlier times such as the Paleolithic era had low incidence of chronic diseases including osteoporosis (Eaton, 2006; Sebastian et al., 2006). The Paleolithic diet has been estimated to contain 400 mEq (400 mmol) per day of potassium (Sebastian et al., 2006). In contrast, the contemporary North American diet induces a mild metabolic acidosis through foods typically containing low amounts of bicarbonate precursors, high amounts of acid-forming protein, and a high sodium intake that can alter acid–base balance through “dilutional-type” acidosis caused by the expansion of extracellular fluid volume. Even cultivated cereal grains yield net acid on metabolism, and these foods displace potassium- and bicarbonate-precursor-rich plant foods (Frasseto et al., 2008). The ratio of inorganic salts of potassium to sodium chloride has also been reversed with this dietary shift. As potassium- and bicarbonate-precursor-rich plant food consumption has declined, sodium chloride has been increasingly consumed largely because of its use in food processing for specific functions, such as preservation and enhancement of taste (Frassetto etal., 2008). Thus, not only has a deficit of dietary potassium become commonplace in the modern diet, but also an excess intake of sodium chloride has become the rule, and an altered sodium-topotassium ratio exists (see below).

ROLE OF THE ANION IN POTASSIUM STUDIES Studies Using Different Potassium Salts In the early 1980s, several reports indicated that potassium salts were effective in the treatment of renal stone disease. For example, five male patients, with documented uric acid lithiasis, were given 60 mmol/day of potassium citrate for 3 weeks (Sakhaee et al., 1983), and during this period, their 24-hour urinary calcium significantly declined from the control level of 154 ± 47 mg/day by one third. Further investigation revealed, however, that providing any source of potassium would not produce hypocalciuria. Then, they compared the effects of different anions, that is, 80 mmol of potassium bicarbonate, potassium citrate, and potassium chloride, on the urine chemistries of eight patients with kidney stones (Sakhaee et al., 1991). After 2 weeks, 24-hour urinary calcium fell significantly during both potassium bicarbonate and potassium citrate administration: from 4.5 ± 1.0 to 3.0 ± 0.8 and 3.6 ± 1.0 mg/day, respectively. This study revealed that potassium chloride had no significant effect on reducing urinary calcium excretion, whereas bicarbonate and citrate salts had similar hypocalciuric actions. The finding that potassium chloride had no effect on urinary calcium loss suggests that either it is the anion associated with potassium that is the major dietary factor or that the anion accompanying the potassium can modify the response to potassium in some way.

263

Is There a Role for Dietary Potassium in Bone Health?

One might predict that providing a source of citrate or bicarbonate would cause hypocalciuria, but that did not happen either. Lemann and coworkers compared potassium and sodium bicarbonate salts and then evaluated potassium versus sodium with either a fixed anion (chloride) or a metabolizable anion (bicarbonate). In their first study (Lemann et al., 1989), they gave 60 mmol/ day of potassium bicarbonate to nine healthy males for 12 days. Twenty-four-hour urinary calcium declined significantly during treatment with potassium bicarbonate, that is, from 4.4 ± 2.0 (control) to 3.5 ± 1.9 mmol/day. Daily calcium balance also significantly improved from −2.3± 1.9 to 1.4± 1.8 mg with potassium but not sodium bicarbonate treatment. Lemann et al. (1991) then compared potassium and sodium bicarbonate salts, evaluating each cation (potassium vs. sodium) with either a fixed anion (chloride) or a metabolizable anion (bicarbonate). They gave 13 healthy adults 90 mmol/ day in a crossover design, measuring 24-hour urinary excretion at day 5. A schematic depiction of the results is shown in Table 17.2. Differences in 24-hour urinary calcium excretion were significant in rank order of lowest calcium excretion: potassium bicarbonate < potassium chloride = sodium bicarbonate < sodium chloride. Fasting calcium excretion, a measure of bone resorption, was also significantly reduced during potassium bicarbonate administration. These data support the contention that alkaline potassium, but not potassium paired with chloride, may slow bone breakdown and conserve bone mass. Depleting the body of potassium, independent of anion, promotes urinary calcium excretion. Lemann et al. (1991) examined the effect of removing either potassium bicarbonate or potassium chloride from the diet. Eight subjects were fed a formula diet low in potassium (2 mmol) for 5 days, which met their nutritional needs and contained added potassium as either potassium chloride or potassium bicarbonate at 1.0 mmol/kg/day. After a 5-day control period, the potassium salts were removed from the diet for 5 days and then reintroduced for a final 5-day recovery period. The removal of either potassium chloride or potassium bicarbonate from the diet resulted in a significant increase in urinary calcium excretion, which averaged 1.31 ± 0.25 mmol/day (p < .005) for both groups when potassium-depleted. Further evidence for the effect of potassium on urinary calcium was generated in a study of the effect of potassium depletion on essential hypertension (Krishna and Kapoor, 1991). In this randomized, crossover study, 12 patients were given either a placebo or potassium chloride (80 mmol/day) for 10 days, with a background dietary intake of only 16 mmol potassium. Urinary calcium was significantly lower during the potassium depletion phase (placebo)

TABLE 17.2 The Effect of Various Combinations of Salts from Two Cations, Potassium or Sodium and Two Anions, a Nonmetabolizable One (Chloride) or a Metabolizable One (Bicarbonate) Cation Sodium Sodium Potassium Potassium

Anion

Effect on Urinary Calcium Excretion

Chloride Bicarbonate Chloride Bicarbonate

↑ ↔ ↔ ↓

Note: Data represent results from a study of these salts on calcium balance and calcium excretion in human volunteers. Source: Lemann, J.J., Pleuss, J.A., Gray, R.W., and Hoffman, R.G., Kidney Int, 39, 973–83, 1991.

264

Diet, Nutrients, and Bone Health

than during potassium chloride administration. These studies suggest that potassium deficiency may be an important determinant of bone loss and that this role is independent of its anion.

The Alkaline Potassium Hypothesis The Paleolithic diet is suggested to be the model for preserving bone integrity through its high potassium and bicarbonate precursors. This idea has led to the alkaline potassium hypothesis in which dietary potassium at levels of 400 mmol/day (close to five times the current adequate intake [AI] for potassium) are advocated to maintain bone integrity due to counter the high acid-forming potential of the modern Western diet (Sebastian et al. 2002, 2006). Frassetto and coworkers (2008) have suggested that the modern diet is low in alkaline potassium which in turn is coupled with an excess of acid-forming compounds as well as of sodium chloride, which promotes an increase in urinary potassium excretion. These factors are considered to make major contributions to the pathogenesis of age-related disorders, including hypertension, renal stones, and osteoporosis. To return to a more favorable ratio of potassium (K) to sodium (Na) in the diet, that is, to a ratio that is comparable with what humans ate in preagricultural diets, both an increase in potassium and an avoidance or severe restriction of sodium chloride are necessary (Frassetto et al., 2008). They proposed that the diet can return to its evolutionary norms of net base production inducing low-grade metabolic alkalosis and a high potassium-to-sodium ratio by (1) greatly reducing content of energy-dense nutrient-poor foods and potassium-poor acid-producing cereal grains, which would entail increasing consumption of potassium-rich net-base-producing fruits and vegetables for maintenance of energy balance, and (2) greatly reducing sodium chloride consumption. Recent food guide revisions have reduced the amount of cereal grains servings and increased the amounts of fruit and vegetable servings, for example, Eating Well with Canada’s Food Guide (released in 2007) and MyPyramid (released in 2005). Both of these food guides resemble the DASH diet. As illustrated in Table 17.1, the DASH diet (and food guides that resemble it) is probably the “best” dietary pattern that can be achieved today using largely unprocessed foods available in the marketplace. By following the DASH diet, one can attain a dietary Na:K ratio of approximately 1:1. This ratio is a marked improvement over dietary intakes that typically occur, i.e., ~2:1, in the United States and Canada. Recent dietary surveys indicate that sodium intakes are higher than the upper level for sodium of 2.3 g and much lower than the AI for potassium of 4.7 g in Canada (Dolega-Cieszowski et al., 2006) and the United States (U.S. Department of Agriculture, 2008).

Can Potassium Be Replaced by Other Cations? The adverse effect of chloride on calcium retention has been related to the provision of a nonmetabolizable anion, in contrast to citrate which is a metabolic bicarbonate precursor. Barzel (1995), in making a recommendation for taking calcium carbonate, indicated that both the calcium and the carbonate would be beneficial to bone. This recommendation leads to the question: can the cation in a alkaline potassium salt be replaced by other cations (except sodium, as noted above)? Recently, the question of potassium versus calcium has been investigated. Karp and coworkers (2009) compared the effects of calcium carbonate, calcium citrate, and potassium citrate on markers of calcium and bone metabolism in young women. At the beginning of each of four 24-hour measurements, subjects received a single dose of calcium carbonate (calcium= 1000 mg), calcium citrate (calcium = 1000 mg), potassium citrate (potassium = 2250 mg; citrate equal to that of calcium citrate), or a placebo in random order. Potassium citrate and calcium carbonate significantly decreased the bone resorption marker N-terminal collagen peptide (Figure17.1). The results suggest that potassium citrate has a positive effect on bone despite the low calcium intake during the session. These results suggest that a diet high in fruits and vegetables is protective to bone in the short-term. However, a low calcium diet cannot be replaced by alkaline

265

Is There a Role for Dietary Potassium in Bone Health? (a)

28

S-BALP (U/l)

27 26 25 24 23 22 21

2

4

6

8

20

24

Time (h)

24 h U-NTx:Crea (nmol BCE/mmol Crea)

(b)

26 24 22 20 18 16 14 Control

Calcium citrate

Calcium carbonate

Potassium citrate

FIGURE 17.1  Changes in the marker of bone formation, serum bone-specific alkaline phosphatase (S-BALP) activity (a), and the marker of bone resorption 24-hour urinary excretion of N-terminal telopeptide of typeI collagen (U-NTx/Crea) (b) during the four study sessions in young women. A: control (□), calcium citrate (▲), calcium carbonate (*), potassium citrate (●). The supplement administration time is indicated with an arrow. The supplements did not affect S-BALP (marker of bone formation) but did affect 24-hour U-NTx/ Crea (marker of bone resorption as indicated by symbol [+]); (a) indicates significantly different from control session. Crea = creatinine. (From Karp, H.J., et al., Br J Nutr 102, 1341–1347, 2009. With permission.)

potassium in the long-term, as indicated by studies in which vegetarians having a low calcium intake were found to have low bone mass (Barr et al., 1998).

POTASSIUM’S ROLE IN BONE EXAMINED The AI Value for Potassium The current recommended intake for potassium for those aged 14 years and over is 4.7 g (120 mmol) (Institute of Medicine 2005). This value is an AI, meaning that not enough evidence was available to set an estimated average requirement. The criterion for setting this high AI value was to help blunt the salt sensitivity prevalent in African American men and to decrease the risk of kidney stones in the entire population. Also, an indication that “higher levels of potassium intake from foods are associated with decreased bone loss” had an impact in arriving at the potassium AI (Institute of Medicine, 2005). A further statement emphasized that “the beneficial effects of potassium . . . [are] from the forms of potassium that are associated with bicarbonate precursors.” Although these

266

Diet, Nutrients, and Bone Health

caveats are important, they do not get readily incorporated into tables where dietary recommendations are provided. As shown in Table 17.1, it is possible to achieve the AI for potassium, but it requires a diet emphasizing fruits and vegetables and lacking in packaged and most processed foods, such as the DASH diet, to achieve an intake even close to the AI value.

Potassium Salts versus Fruits and vegetables Few studies have directly compared the effects of an alkaline potassium salt associated with the ingestion of fruits and vegetables. Macdonald and coworkers (2008) completed a trial of potassium from either citrate or fruit and vegetable consumption in postmenopausal women. This randomized controlled trial, blinded for two levels of potassium citrate against a placebo, involved 276 women and lasted 2 years. Results from the trial were disappointing in that neither the potassium salt nor the additional fruits and vegetables improved bone turnover markers. One possible reason for the lack of an effect of the potassium salt was that habitual dietary calcium was likely to be adequate, that is, atypically low. Similarly, the women may have already been consuming fairly adequate numbers of servings of fruits and vegetable, which meant that the effect of ingesting more could not be detected. Compliance of study participants is also an issue for investigating dietary change. Thus, it may be difficult to conclude that an increase in fruit and vegetable intake, or an increase in alkaline potassium salts, has beneficial long-term effects on bone. Hopefully, better study designs in the future will be able to overcome some of the common limitations of these investigations.

Does Potassium Intake Alone Predict Bone Health? Several lines of evidence suggest that the relationship between potassium and preservation of bone mass is not straightforward. In establishing the definition of the AI value, the Institute of Medicine stressed that potassium intake should be obtained from foods, that is, fruits and vegetables, that provide potassium with bicarbonate precursors. Nevertheless, one would still expect that if potassium intake is protective to bone mass, then a good relationship between potassium intake and maintenance of bone should exist during aging. Rafferty et al. (2005) tested this relationship under steady-state conditions, using data from a total of 644 inpatient balance studies. Dietary potassium was highly significantly associated with urinary calcium excretion in a negative direction; yet, dietary potassium was negatively correlated with calcium absorption, and the effect of potassium on urinary calcium was lost when adjusted for enteric calcium absorption. The authors concluded that dietary potassium does not seem to exert a net positive influence on calcium retention. The source of potassium in these subjects was primarily from milk and meat, indicating that when the alkaline precursors were not with potassium, then potassium has no net effect on bone. A second line of evidence suggests that potassium itself is not the “active” part of alkaline potassium salts, according to a study from Dawson-Hughes and coinvestigators (2009). Unlike other studies in which bone loss or gain was attributed to changes in urinary calcium excretion alone, the Dawson-Hughes et al. study used changes in bone resorption markers to determine the effects of potassium on bone. In this study, 171 older adults (>50 years) were randomized to receive 67.5 mmol potassium bicarbonate (equivalent to 2600 mg potassium), sodium bicarbonate, potassium chloride, or a placebo. Subjects were measured at entry and after 3 months. The investigators were thus able to examine the effects of the anion (bicarbonate vs. chloride) and of the cation (sodium vs. potassium). Results are shown in Figure 17.2. Potassium chloride clearly did not prevent bone resorption, in contrast to potassium bicarbonate. The authors concluded that bicarbonate alone is the active component protecting against the loss of bone mineral, that is, calcium. However, the results with respect to sodium bicarbonate were equivocal because the skeletal effect of the bicarbonate was less than that of potassium bicarbonate, as previously reported (Lemann et al., 1989, 1991; Green and Whiting, 1994). Theoretically, the sodium cation may have caused an increase in urinary calcium excretion, which would nullify the beneficial effect of the

267

Is There a Role for Dietary Potassium in Bone Health? 3.0 2.0

Change in NTX/Cr (nmol/mmol)

1.0 0.0 –1.0 –2.0 –3.0 –4.0 –5.0 –6.0 –7.0 –8.0 –9.0 –10

Placebo

KCI

KHCO3

NaHCO3

FIGURE 17.2  Mean 3-month change in the marker of bone resorption urinary N-terminal telopeptide of type I collagen NTX per creatinine (Cr) excretion by treatment group, adjusted for sex and baseline NTX/Cr. Subjects were men and women over 50 years old. (From Dawson-Hughes, B., et al., JClin Endocrinol Metab 94, 96–102, 2009. With permission.)

bicarbonate anion on bone remodeling. Obviously, the use of sodium bicarbonate as an alkalizing food ingredient or supplement would not be desirable.

CONCLUSIONS The question addressed in this chapter was whether dietary potassium intake plays a role in bone health. From depletion studies, it is apparent that low levels of potassium, which could arise from low intakes, contribute to an increase in calcium excretion and, thus, promote bone loss via resorption should a negative calcium balance persist. This effect is independent of anion. Therefore, adequate dietary potassium is required for optimal bone health. However, the question is often phrased as being whether a higher than typical potassium intake is required for bone health, that is, one that is similar to or higher than the current recommendation of 4700 mg. Here, attention has been given to the concept that alkaline potassium salts promote calcium retention. This concept arose from two lines of research: (1) epidemiological studies in which fruit and vegetable intake was demonstrated to be positive for BMD or other markers of bone health and (2) experimental studies either in laboratory animals or human volunteers showing that alkaline potassium salts promoted a more positive calcium balance or reduced markers of bone resorption, in comparison with alkaline sodium salts or with acid-forming potassium salts. The data, however, are not so clear-cut as higher potassium intake does not always protect against bone loss. Also, the question has arisen as to whether the alkaline salt must have potassium as the cation, although sodium clearly would not benefit bone health because of other mechanisms. Thus, current recommended intake (AI) for potassium of 4700 mg (120 mmol) for 14 years and older, which is based, in part, on the alkaline potassium hypothesis, is providing an optimal cation level, but this cation does not have to be potassium. Meeting this high level of potassium necessitates choosing a healthy diet emphasizing fruits and vegetables, such as the DASH diet, Canada’s Food Guide, and MyPyramid. This dietary choice, and not intake of an alkaline potassium salt, ensures that bone health is optimized.

268

Diet, Nutrients, and Bone Health

REFERENCES Barr, S.I., Prior, J.C., Janelle, K.C., et al. 1998. Spinal bone mineral density in premenopausal vegetarian and nonvegetarian women: Cross-sectional and prospective comparisons. J Am Diet Assoc 98:760–765. Barzel, U. 1995.The skeleton as an ion exchange system: Implications for the role of acid–base imbalance in the genesis of osteoporosis. J Bone Miner Res 10:1431–1436. Dawson-Hughes, B., Harris, S.S., Palermo, N.J., et al. 2009. Treatment with potassium bicarbonate lowers calcium excretion and bone resorption in older men and women. J Clin Endocrinol Metab 94:96–102. Dolega-Cieszowski, J., Bobyn, P.J., and Whiting, S.J. 2006. Dietary Intakes of Canadians in the 1990s using population weighted data derived from the Provincial Nutrition Surveys. Appl Physiol Nutr Metab 31:753–758. Doyle, L., and Cashman, K.D. 2004. The DASH diet may have beneficial effects on bone health. Nutr Rev 62:215–220. Eaton, S.B. 2006. The ancestral human diet: What was it and should it be a paradigm for contemporary nutrition? Proc Nutr Soc 65:1–6. Frassetto, L.A., Morris, C.R., Sellmeyer, D.E., et al. 2008. Adverse effects of sodium chloride on bone in the aging human population resulting from habitual consumption of typical American diets. J Nutr 138:419S–422S. Green, T.J., and Whiting, S.J. 1994. Potassium bicarbonate reduces high protein-induced hypercalciuria in adult men. Nutr Res 14:991–1002. Institute of Medicine. 2005. Dietary Reference Intakes: Sodium, Chloride, Potassium and Sulphate. National Academy Press, Washington, DC. Jones, G., Riley, M.D., and Whiting, S.J. 2001. Association between urinary potassium, urinary sodium, current diet, and bone density in prepubertal children. Am J Clin Nutr 73:839–844. Karp, H.J., Ketola, M.E., and Lamberg-Allardt, C.J. 2009. Acute effects of calcium carbonate, calcium citrate, and potassium citrate on calcium and bone metabolism in young women. Br J Nutr 102:1341–1347. Krishna, G.G., and Kapoor, S.C. 1991. Potassium depletion exacerbates essential hypertension. Ann Int Med 115:77–83. Lemann, J.J., Gray, R.W., and Pleuss, J.A. 1989. Potassium bicarbonate, but not sodium bicarbonate, reduces urinary calcium excretion and improves calcium balance in healthy men. Kidney Int 35:688–695. Lemann, J.J., Pleuss, J.A., Gray, R.W., and Hoffman, R.G. Potassium administration decreases and potassium deprivation increases urinary calcium excretion in healthy adults. Kidney Int 39:973–983. Lin, P.H., Ginty, F., Appel, L.J., et al. 2003. The DASH diet and sodium reduction improve markers of bone turnover and calcium metabolism in adults. J Nutr 133:3130–3136. Macdonald, H.M., Black, A.J., Aucott, L., et al. 2008. Effect of potassium citrate supplementation or increased fruit and vegetable intake on bone metabolism in healthy post menopausal women: A randomized controlled trial. Am J Clin Nutr 88:465–474. McGartland, C.P., Robson, P.J., Murray, L.J., et al. 2004. Fruit and vegetable consumption and bone mineral density: The Northern Ireland Young Hearts Project. Am J Clin Nutr 80:1019–1023. McTiernan, A., Wactawski-Wende, J., Wu, L.L., et al. 2009. Low-fat, increased fruit, vegetable, and grain dietary pattern, fractures, and bone mineral density: The Women’s Health Initiative Dietary Modification Trial. Am J Clin Nut 89:1864–1876. New, S.A. 2003. Intake of fruit and vegetables: Implications for bone health. Proc Nutr Soc 62:889–899. New, S.A., Boulton-Smith, C., Grubb, D.A., et al. 1997 Nutritional influences on bone mineral density: A cross-sectional study in premenopausal women. Am J Clin Nutr 65:1831–1839. New, S.A., Robins, S.P., Campbell, M.K., et al. 2000. Dietary influences on bone mass and bone metabolism: Further evidence of a positive link between fruit and vegetable consumption and bone health. Am J Clin Nutr 71:142–151. Rafferty, K., Davies, M., and Heaney, R.P. 2005. Potassium intake and the calcium economy. J Am Coll Nutr 24:99–106 Sakhaee, K., Alpern, R., Jacobson, H.R., et al. 1991. Contrasting effects of various potassium salts on renal citrate excretion. J Clin Endocrinol Metab 72:396–400. Sakhaee, K., Nicar, M., Hill, K., et al. 1983. Contrasting effects of potassium citrate and sodium citrate on urinary chemistries and crystallization of stone-forming salts. Kidney Int 24:348–352. Sebastian, A., Frassetto, L.A., Sellmeyer, D.E., et al. 2002. Estimation of the net acid load of the diet of ancestral preagricultural hom*o sapiens and their hominid ancestors. Am J Clin Nutr 76:1308–1316. Sebastian, A., Frassetto, L.A., Sellmeyer, D.E., et al. 2006. The evolution-informed optimal dietary potassium intake of human beings greatly exceeds current and recommended intakes. Sem Nephrol 26:447–453.

Is There a Role for Dietary Potassium in Bone Health?

269

Tucker, K.L., Chen, H., Hannan, M.T., et al. 1999. Potassium, magnesium, and fruit and vegetable intakes are associated with greater bone mineral density in elderly men and women. Am J Clin Nutr 69:727–736. Tylavsky, F.A., Holliday, K., Danish, R.K., et al. 2004. Fruit and vegetable intake is an independent predictor of bone mass in early-pubertal children. Am J Clin Nutr 79:311–317. U.S. Department of Agriculture, Agricultural Research Service. 2008. Nutrient Intakes from Food: Mean Amounts Consumed per Individual, One Day, 2005–2006. Available at www.ars.usda.gov/ba/bhnrc/fsrg Vatanparast, H., Whiting, S.J., Baxter-Jones, A., et al. 2005. The positive effect of vegetable and fruit consumption on bone mineral accrual of boys during growth from childhood to adolescence in the University of Saskatchewan Pediatric Bone Mineral Accrual Study. Am J Clin Nutr 82:700–706.

18

Acid–Base Balance Susan A. Lanham-New

CONTENTS Introduction . .................................................................................................................................. 271 Fundamentals of Acid–Base Maintenance: Criticality to Health.................................................... 272 A Role for the Skeleton in Acid–Base Homeostasis: Early Reports.......................................... 273 Quantitative Loss of Bone: Ratio of Calcium to Fixed Acid Load............................................ 273 Systemic Acidosis and the Skeleton: Potential Mechanisms of Action?........................................ 273 Vegetarianism and Skeletal Health................................................................................................. 274 Omnivores and Bone Health...................................................................................................... 275 Quantifying the Acidity of Foods: Potential Renal Acid Load.................................................. 275 Net Endogenous Acid Production and Its Effects on Bone........................................................ 275 Interpretations................................................................................................................................. 276 Conclusions..................................................................................................................................... 276 References....................................................................................................................................... 277

INTRODUCTION Nutritional strategies for optimizing bone health throughout the life cycle are vital because prevention of osteoporosis rather than its treatment is the preferred approach. As an exogenous factor, nutrition is amenable to change and has relevant public health implications and hence deserves special attention. With the growing increase in the age of life expectancy, that is, 1:4 in the adult population will be aged 65 years and over by 2030, hip fractures are predicted to rise exponentially in the next decade and hence an urgent need for the implementation of public health strategies to target prevention of poor skeletal health on a population-wide basis. The role that the skeleton plays in acid–base homeostasis has been gaining increasing prominence in the literature over the last few decades. Theoretical considerations of the role alkaline bone mineral may play in the defense against acidosis date as far back as the late 19th century. Natural, pathological, and experimental states of acid loading/acidosis have long been associated with hypercalciuria and negative calcium balance. More recently, the detrimental effects of acid from the diet on bone mineral have been demonstrated. At the cellular level, a reduction in extracellular pH has been shown to have a direct effect on osteoclastic activity, resulting in increased resorption pit formation in bone. Although scientists thought that vegetarianism may be protective against bone loss, studies over the last two decades have demonstrated that such an eating practice is not protective. The amount of dietary alkali in vegetarian diets, however, may be a key component obtained from the consumption of high-­potassium, high-bicarbonate foods, such as fruits and vegetables. A low intake of dietary acidity has been shown to be protective to the skeleton, and numerous observational, experimental, clinical, and intervention studies, over the last decade have supported a positive link between fruit-and-vegetable

271

272

Diet, Nutrients, and Bone Health

consumption and preservation of the skeleton. Further research on fracture prevention, particularly with respect to the influence of dietary manipulation using alkali-forming foods, is required.

FUNDAMENTALS OF ACID–BASE MAINTENANCE: CRITICALITY TO HEALTH Life is a struggle, not against sin, not against the money power, not against malicious animal magnetism, but against hydrogen ions. H.L. Mencken, 1880–1956

As noted by Kraut and Coburn (1994), these famous words by Mencken in the early 20th century about the meaning of life and death may also apply to the struggle of the healthy skeleton against the deleterious effects of retained acid. As shown in Figure 18.1, acid–base homeostasis is critical to health. The pH of extracellular fluid ranges between 7.38 and 7.42, and it is a major challenge to the body’s balance to defend against changes in hydrogen ion (H+ ion) concentrations between 0.035 and 0.045 mEq/L (Green and Kleeman, 1991). Maintaining the H+ concentration within such narrow limits is essential to survival, and the body’s adaptive response involves three specific mechanisms: (1) buffer systems, (2) exhalation of CO2, and (3) kidney excretion. On a daily basis, humans eat substances that both generate and consume H+ ions (protons), and as a net result, adult humans on a normal Western diet generate ~1 mEq of acid (as H+ ions) per kilogram lean body mass (LBM) per day. The more acid precursors a diet contains, the greater the degree of systemic acid load (Kurtz et al., 1983). Furthermore, as people age, the overall renal function declines, including the ability to excrete hydrogen ion (acid) in the form of ammonium ion (Frassetto et al., 1996a). Thus, with increasing age, humans become significantly (albeit slightly) more acidic (Frassetto et al., 1996b), that is, less than pH 7.38, which is the lower end of the normal range. The question arises: Should not the elderly with reduced ammoniagenesis and decreased H+ secretion defend normal pH by increasing pulmonary ventilation and thus decrease their pCO2? Probably not. So, the elderly with decreased H+ ion excretory capacity likely depend on bone to buffer the retained H+ ions and

Importance of acid–base regulation and health/disease outcome When the human body is confronted with an excess of H2 ions from the diet, it employs a number of strategies to maintain normal blood pH(7.4).

Diet

Intestine

Lung Blood H/OH

AA-SH 0A

H/OH

Acid– base pool

H/OH

Pulmanary regulation of the biocarbonate-carbonic acid buffer system (respiratory compenstion)

H/OH

Kidney

Liver H/OH

H/OH

Elimination of nonvolatile acids or NaHCO3

Bone

FIGURE 18.1  Acid–base homeostasis: The components of the control of blood and tissue pH are critical to life. (Source: Green, J. and Kleeman, R., Kidney Int, 39, 9–26, 1991. With permission.)

Acid–Base Balance

273

p­ resent with hypercalciuria even in the face of a normal arterial pH of 7.40. This point, however, has not been adequately addressed by current research.

A Role for the Skeleton in Acid –Base Homeostasis: Early Reports The theoretical considerations of the role alkaline bone mineral may play in the defense against acidosis go as far back as the late 1880s (Goto, 1918; Irving and Chute, 1933; Albright and Reifenstein, 1948). In those early years, a number of studies were published that provided evidence that in natural (e.g., starvation), pathological (e.g., diabetic acidosis), and experimental (e.g., ammonium chloride ingestion) metabolic acidosis caused by fixed acid loading, an association exists between H+ ion loading on one hand and hypercalciuria and negative calcium balance on the other (Gastineau etal., 1960; Reidenberg et al., 1966). The pioneering work of Jacob Lemann and Uriel Barzel over 30years ago showed extensively the effects of acid from the diet on bone mineral in both man and in the animal model (Lemann et al., 1966; Barzel, 1969). In the 1960s, Wachman and Bernstein (1968) put forward a hypothesis linking the daily diet to the development of osteoporosis based on the role of bone in acid–base balance. They hypothesized “the increased incidence of osteoporosis with age may represent, in part, the results of a life-long utilisation of the buffering capacity of the basic salts of bone for the constant assault against pH homeostasis” (Wachman and Bernstein, 1968). The intake of acid is a way of everyday life, and it is known that animal proteins but also cereals and grains are rich sources of phosphoric and sulfuric acid and are recognized as acid ash foods (Barzel and Massey, 1998). Sources of the net production of acid are H+ ions produced by endogenous cell metabolism and from dietary proteins and other components. A gross quantitative relationship exists between the amount of acid generated from dietary protein sources (as reflected by urine pH) and the amount of acid ash consumed in the diet.

Quantitative Loss of Bone: R atio of Calcium to Fixed Acid Load Chronic metabolic acidosis exists, such as resulting from distal tubular renal acidosis (d-RTA), and is associated with hypercalciuria and loss of considerable amounts of skeletal calcium. Under such high degrees of hydrogen ion retention and resultant systemic metabolic acidosis, 2 mEq of Ca/kg/ day is required to buffer each mEq of fixed acid/kg/day; over a period of 10 years (and assuming a total body Ca of approximately 1 kg), this would account for 15% loss of inorganic bone mass in an average individual (Widdowson et al., 1951). To lessen the skeletal calcium loss from high dietary acid exposure in d-RTA, the consumption of a diet consisting of large quantities of fruit and vegetables, that is, alkaline ash diet, may be important for bone health maintenance. Although unproven, this diet may also benefit skeletal health in older persons in the general population with age-related defects in renal H+ ion secretion (Barzel, 1995).

SYSTEMIC ACIDOSIS AND THE SKELETON: POTENTIAL MECHANISMS OF ACTION? The novel work by Arnett and Dempster in 1986 demonstrated a direct enhancement of osteoclastic activity following a reduction in extracellular pH, an effect that was independent of parathyroid hormone (Arnett and Dempster, 1986, 1990; Arnett et al., 1994). Osteoclasts and osteoblasts appear to respond independently to small changes in pH in the culture media in which they are growing (Kreiger et al., 1992), and evidence exists that a small drop in pH, close to the lower limit of the physiological range, causes a tremendous burst in bone resorption (Arnett and Spowage, 1996; Bushinsky, 1996). The recent work by Arnett’s group has shown that metabolic acidosis stimulates resorption by activating mature osteoclasts already present in calvarial bone rather than by inducing

274

Diet, Nutrients, and Bone Health

formation of new osteoclasts (Meghji et al., 2001). Almost all the bone mineral, that is, calcium and phosphate ions, released in response to acidosis results from osteoclast activation and increased formation of resorption pits on bone surfaces. Also, the organic matrix is destroyed at the same time (Dr. T.R. Arnett, University College London, personal communication). Some evidence has been reported that excess hydrogen ions directly induce physicochemical calcium release from bone (Bushinsky et al., 1994), but most agree that osteoclasts must be involved.

VEGETARIANISM AND SKELETAL HEALTH Following the recognition of the role that bone plays in acid–base balance and the hypothesis linking diet to osteoporosis, it was proposed that long-term ingestion of vegetable-based diets may have a beneficial effect on bone mineral mass. In general, the earlier studies (published before 1990) appeared to support the hypothesis that plant foods had positive effects on bone, that is, bone mineral mass was found to be higher in the vegetarian group compared with their omnivorous counterparts (Table 18.1) (Ellis et al., 1972; Marsh et al., 1980, 1983, 1988; Tylavsky and Anderson, 1988; Hunt et al., 1989). However, two important points concerning these data require consideration. First, a fundamental error existed in the interpretation of the photographic density measurements in the first article published by Ellis et al. (1972); their conclusions should have been the opposite of what they claimed (Meema, 1973; Ellis et al., 1974; Meema, 1996; Barzel, 1996). Second, many subjects in several of the published studies were Seventh-Day Adventists who had a different lifestyle compared with that of the omnivorous group. The lifestyle variable was likely to have had an important confounding influence because almost all SDAs refrained from smoking and taking alcohol and caffeine and their physical activity levels were higher than those of comparator omnivores. Studies published in the last two decades suggest no differences in BMD between vegetarians and omnivores (Lloyd et al., 1991; Tesar et al., 1992; Reed et al., 1994) (Table 18.2). In a 5-year prospective study of changes in radial bone density of elderly white American women, no differences were seen in bone loss rates between the lacto-ovo vegetarians and the omnivorous group (Reed et al., 1994). Furthermore, in the most recently published studies, bone mass was found to be significantly lower in the Asian vegetable-based dietary groups (Chiu et al., 1997; Lau et al., 1998), although it is likely that protein undernutrition may account for some of these differences (Rizzoli et al., 1998). In a recent meta-analysis of vegetarianism and bone, Ho-Pham and colleagues (2009) found no clinically significant effects of vegetarianism on bone health, and in TABLE 18.1 Vegetarianism and Bone Health—Review of Earlier Work (Pre-1989) Author

Year

Journal Source

Ellis et al. Ellis et al. Mazess and Mather

1972 1974 1974

Am J Clin Nutr 25: 555–558 Am J Clin Nutr 27: 769–770 Am J Clin Nutr 27: 916–919

Mazess and Mather Marsh et al. Marsh et al. Marsh et al. Tylavsky et al.

1975 1980 1983 1988 1988

Human Biol 47: 45 JAMA 76: 148–151 Am J Clin Nutr 37: 453–456 Am J Clin Nutr 48: 837–841 Am J Clin Nutr 48: 842–849

Hunt et al.

1989

Am J Clin Nutr 50: 517–523

Key Findings BMD ↑ in vegetarian group BMD ↓ in vegetarian group BMC ↓ in North Alaskan Eskimos BMC ↓ in Canadian Eskimos Bone loss ↑ in omnivores BMD ↑ in vegetarians BMD ↑ in elderly vegetarians No difference in BMD between groups No difference in BMD between groups

Overall Summary ✓ ⨯ ✓ ✓ ✓ ✓ – –

275

Acid–Base Balance

TABLE 18.2 Vegetarianism and Bone Health: Summary of Studies—Later Work (1990 and Beyond) Author

Year

Source

Lloyd et al.

1991

Am J Clin Nutr 54: 1005–1010

Tesar et al.

1992

Am J Clin Nutr 56: 699–704

Reed et al. Chui et al. Lau et al. Ho-Pham et al.

1994 1997 1998 2009

Am J Clin Nutr 59: 1997–1202 Calcified Tissue Int 60: 245–249 Eur J Clin Nutr 52: 60–64 Am J Clin Nutr 90: 943–950

Findings No difference in BMD between groups No difference in BMD between groups Bone loss rates similar BMD ↓ in vegan group Hip BMD lower in vegetarian group Meta-analysis—no clinically significant difference

Summary – – – ⨯ ⨯ –

an accompanying editorial, it was concluded that vegetarianism is not a significant risk factor for osteoporosis (Lanham-New, 2009).

Omnivores and Bone Health Very few studies have focused attention with respect to bone health on populations consuming a diet highly dependent on animal foods, particularly that of meat (Hammond and Storey, 1970). Mazess and Mather (1974) examined the bone mineral content of forearm bones in a sample of 217 children, 89adults, and 107 elderly Eskimo natives of the north coast of Alaska (Mazess and Mather, 1974). After the age of 40 years, the Eskimos of both sexes were found to have a deficit of bone mineral of 10% to 15% relative to white standards. An even greater aging bone loss was found in Canadian Eskimos (Mazess and Mather, 1975). The issue of dietary change among the Eskimo population, particularly the increased utilization of refined carbohydrates, was raised (Mann, 1975), but the high-acid ash diet based on animal meat was considered a major factor contributing to bone loss of Eskimo adults. Clearly, these findings are of considerable interest to the interaction between diet and bone in the regulations of systemic acid–base balance, and further work in this area is clearly warranted (New, 2001a).

Quantifying the Acidity of Foods: Potential Renal Acid Load Of considerable interest is the finding that vegetable-based proteins generate a large amount of acid in the urine. The work by Remer and Manz (1995) examining the potential renal acid loads (PRALs) of a variety of foods has found that many grain products and some cheeses have a high PRAL level (Remer and Manz, 1995) (Table 18.3). These foods, which are likely to be consumed in large quantities by lacto-ovo vegetarians, may provide an explanation for the lack of a positive effect on bone health indices in studies comparing lacto-ovo vegetarians versus omnivores.

Net Endogenous Acid Production and Its Effects on Bone Determination of the acid–base content of diets consumed by individuals and populations groups is a useful way forward concerning the role of the skeleton in acid–base hom*oeostasis. Because 24-hour urine collections, considered as the gold standard for acid–base research, are difficult to perform in large population-based studies, an alternative approach is to examine the net acid content of the diet. Research from Sebastian et al. (1990) has found that the protein-to-potassium ratio predicts net acid excretion, and in turn, net renal acid excretion predicts calcium excretion.

276

Diet, Nutrients, and Bone Health

TABLE 18.3 Potential Renal Acid Load (PRAL) Values of a Variety of Foods and Food Groups Food/Food Group Cheese with high protein content Meat and meat products Cheese with low protein content Bread Milk and non-cheese products Fruits and fruit juices (without dried fruit) Vegetables

PRAL mEq/100 g Edible Portion 26.4 9.5 8.0 3.5 1.0 –3.1 –2.8

Source: Adapted from Remer, T. and Manz, F., J Am Diet Assoc, 95, 791–7, 1995.

Theyproposed a simple algorithm to determine the net rate of endogenous noncarbonic acid production (NEAP) from considerations of the acidifying effect of protein and the alkalizing effect of potassium (Frassetto et al., 1998). High intakes of NEAP have been shown to be associated with poorer indices of bone health throughout the life cycle (New et al., 2004). In the most recent metaanalysis of protein and bone, Darling and coworkers (2009) have shown that the overriding finding is that animal protein is positively associated with bone health (Kerstetter, 2009), yet other reports have shown that too much animal protein in the diet may also result in increased acid generation and bone loss.

INTERPRETATIONS The diet of the modern-age human is widely believed to be considerably different from that consumed by our Paleolithic ancestors (Eaton and Konner, 1985). Reflections on the dietary content of preagricultural man estimate intakes of sodium to be 2 to 5 mEq/day and potassium to be at levels reaching 200 mEq/day. This ratio is in stark contrast to current dietary data that estimate population intakes of sodium and potassium at levels of approximately 170 and 65 mEq/day, respectively, in the United Kingdom, United States, and Australia. Eaton and coinvestigators (1996) stated that the kidneys of our hunter-gatherer ancestors were designed to excrete potassium and conserve sodium. This evolutionary mechanism still exists despite the almost total dietary reversal of consuming more sodium than potassium; hence the term “today’s diet, yesterday’s genes” is most fitting. Although our Paleolithic forbears occasionally ate animal protein, their choice of foods was predominantly fruits, vegetables, and other plant sources. This alkali ash diet meant that they had lower renal H+ excretory demands than that of humans living over the last 10,000 years when agriculture and animal husbandry has flourished.

CONCLUSIONS The evidence currently available from experimental, clinical, and observational studies suggests a role for the skeleton in acid–base balance. An acidic dietary load appears to have an adverse effect on the retention of bone mass, whereas an alkaline dietary load appears to help conserve bone mass. Future research should focus attention on intervention trials centered specifically on fruits and vegetables as the supplementation vehicle and assessing a wide range of bone health indices, including

Acid–Base Balance

277

fracture risk. Reanalyses of existing dietary and bone mass/metabolism datasets to examine, in particular, the impact of dietary acidity on the skeleton, also are valuable data sources for future research.

REFERENCES Albright, F., and Reifenstein, E.C., Jr. 1948. The Parathyroid Glands and Metabolic Bone Disease. Williams and Wilkins, Baltimore, 241–247. Arnett, T.R., Boyde, A., Jones, S.L., et al. 1994. Effects of medium acidification by alteration of carbon dioxide or bicarbonate concentrations on the resorptive activity of rat osteoclasts. J Bone Miner Res 9: 375–379. Arnett, T.R., and Dempster, D.W. 1986. Effect of pH on bone resorption by rat osteoclasts in vitro. Endocrinology 119: 119–124. Arnett, T.R., and Dempster, D.W. 1990. Perspectives: Protons and osteoclasts. J Bone Miner Res 5: 1099–1103. Arnett, T.R., and Spowage, M. 1996. Modulation of the resorptive activity of rat osteoclasts by small changes in extracellular pH near the physiological range. Bone 18: 277–279. Barzel, U.S. 1969. The effect of excessive acid feeding on bone. Calcif Tissue Res 4: 94–100. Barzel, U.S. 1995. The skeleton as an ion exchange system: Implications for the role of acid–base imbalance in the genesis of osteoporosis. J Bone Miner Res 10: 1431–1436. Barzel, U.S. 1996. Nevertheless, an acidogenic diet may impair bone. J Bone Miner Res 11: 704. [Letter] Barzel, U.S., and Massey, L.K. 1998. Excess dietary protein can adversely affect bone. J Nutr 128: 1051–1053. Bernstein, D.S., Wachman, A,, and Hattner, R.S. 1970. Acid–base balance in metabolic bone disease. In Osteoporosis, Barzel, U.S., ed. Grune and Stratton, New York, 207–216. Bushinsky, D.A. 1996. Metabolic alkalosis decreases bone calcium efflux by suppressing osteoclasts and stimulating osteoblasts. Am J Physiol (Renal Fluid Electrolyte Physiol) 271: F216–F222. Bushinsky, D.A., Lam, B.C., Nespeca, R., et al. 1993. Decreased bone carbonate content in response to metabolic, but not respiratory, acidosis. Am J Physiol (Renal Fluid Electrolyte Physiol) 265: F530–F536. Chiu, J.F., Lan, S.J., Yang, C.Y., et al. 1997. Long term vegetarian diet and bone mineral density in postmenopausal Taiwanese women. Calcif Tissue Int 60: 245–249. Darling, A.L., Millward, D.M., Torgerson, D.T., et al. 2009. Dietary protein and bone health: A systematic review and meta-analysis. Am J Clin Nutr 90: 1674–1692. Eaton, B.S., Eaton, B.S., III, Konner, M.J., et al. 1996. An evolutionary perspective enhances understanding of human nutritional requirements. J Nutr 126: 1732–1740. Eaton, B.S., and Konner, M. 1985. Paleolithic nutrition. A consideration of its nature and current implications. New Engl J Med 312: 283–290. Ellis, F.R., Holesh, S., and Ellis, J.W. 1972. Incidence of osteoporosis in vegetarians and omnivores. Am J Clin Nutr 25: 555–558. Ellis, F.R., Holesh, S., and Sanders, T.A 1974. Osteoporosis in British vegetarians and omnivores. Am J Clin Nutr 27: 769–770. Frassetto, L.A., Morris, R.C., Jr., and Sebastian, A. 1996a. Effect of age on blood acid–base composition in adult humans: Role of age-related renal functional decline. Am J Physiol (Renal Fluid Electrolyte Physiol) 271: F1114–F1122. Frassetto, L.A., and Sebastian, A. 1996b. Age and systemic acid–base equilibrium: Analysis of published data. J Gerontol 51A: B91–B99. Frassetto, L., Todd, K., Morris, R.C., Jr., et al. 1998. Estimation of net endogenous noncarbonic acid production in humans from dietary protein and potassium contents. Am J Clin Nutr 68: 576–583. Gastineau, C.F., Power, M.H., and Rosevear, J.W. 1960. Metabolic studies of a patient with osteoporosis and diabetes mellitus: Effects of testosterone enanthate and strontium lactate. Proc Mayo Clin 35: 105–111. Goto, K. 1918. Mineral metabolism in experimental acidosis. Journal Biol Chem 36: 355–376. Green, J., and Kleeman, R. 1991. Role of bone in regulation of systematic acid–base balance. Kidney Int 39: 9–26. [Editorial Review] Hammond, R.H., and Storey, E. 1970. Measurement of growth and resorption of bone in rats fed meat diet. Calcif Tissue Res 4: 291. Ho-Pham, L.T., Nguyen, D.N., and Nguyen, T.V. 2009. Effect of vegetarian diets on bone mineral density: ABayesian meta-analysis. Am J Clin Nutr 90: 943–950.

278

Diet, Nutrients, and Bone Health

Hunt, I.F., Murphy, N.J., Henderson, C, et al. 1989. Bone mineral content in postmenopausal women: Comparison of omnivores and vegetarians. Am J Clin Nutr 50: 517–523. Irving, L., and Chute, A.L. 1933. The participation of the carbonates of bone in the neutralisation of ingested acid. J Cell Comp Physiol 2: 157. Kerstetter, J.E. 2009. Dietary protein and bone: A new approach to an old problem. Am J Clin Nutr 90: 1451– 1452. [Editorial] Kraut, J.A., and Coburn, J.W. 1994. Bone, acid and osteoporosis. New Engl J Med 330: 1821–1822. Kreiger, N.A., Sessler, N.E., and Bushinsky, D.A. 1992. Acidosis inhibits osteoblastic and stimulates osteoclastic activity in vitro. Am J Physiol 262: F442–F448. Kurtz, I., Maher, T., Hulter, H.N., et al. 1983. Effect of diet on plasma acid–base composition in normal humans. Kidney Int 24: 670–680. Lanham-New, S.A. 2009. Is “vegetarianism” a serious risk factor for osteoporotic fracture? Am J Clin Nutr 90: 910–911. [Editorial] Lau, E.M., Kwok, T., Woo, J., et al. 1998. Bone mineral density in Chinese elderly female vegetarians, vegans, lactoovovegetarians and omnivores. Eur J Clin Nutr 52: 60–64. Lemann, J., Jr., Adams, N.D., and Gray, R.W. 1979. Urinary calcium excretion in humans. New Engl J Med 301: 535–541. Lemann, J., Jr., Litzow, J.R., and Lennon, E.J. 1967. Studies of the mechanisms by which chronic metabolic acidosis augments urinary calcium excretion in man. J Clin Invest 46: 1318–1328. Lloyd, T., Schaeffer, J.M., Walker, M.A., et al. 1991. Urinary hormonal concentrations and spinal bone densities of premenopausal vegetarian and nonvegetarian women. Am J Clin Nutr 54: 1005–1010. Mann, G. 1975. Bone mineral content of North Alaskan Eskimos. Am J Clin Nutr 28: 566–567. [Letter] Marsh, A.G., Sanchez, T.V., Chaffee, F.L., et al. 1983. Bone mineral mass in adult lactoovovegetarian and omnivorous males. Am J Clin Nutr 83: 155–162. Marsh, A.G., Sanchez, T.V., Micklesen, O., et al. 1980. Cortical bone density of adult lactoovovegetarians and omnivorous women. J Am Diet Assoc 76: 148–151. Marsh, A.G., Sanchez, T.V., Michelsen, O., et al. 1988. Vegetarian lifestyle and bone mineral density. Am J Clin Nutr 48: 837–841. Mazess, R.B., and Mather, W.E. 1974. Bone mineral content of North Alaskan Eskimos. Am J Clin Nutr 27: 916–925. Mazess, R.B., and Mather, W.E. 1975. Bone mineral content in Canadian Eskimos. Human Biol 47: 45. Meema, H.E. 1973. Photographic density versus bone density. Am J Clin Nutr 26: 687. [Letter] Meema, H.E. 1996. What’s good for the heart is not good for the bones? J Bone Miner Res 11: 704. [Letter] Meghji, S., Morrison, M.S., Henderson, B., et al. 2001. PH dependence of bone resorption: Mouse calvarial osteoclasts are activated by acidosis. Am J Physiol (Endocrinol and Metab) 280: E112–E119. Mencken, H. L. 1880–1956. www.quotationspage.com New, S.A. 2001a. Nutrition, exercise and bone health. Proc Nutr Soc 60(2): 265–274. New, S.A. 2001b. Impact of food clusters on bone. In Nutritional Aspects of Osteoporosis 2000, DawsonHughes, B., Burckhardt, P., and Heaney, R.P., eds. Challenges of Modern Medicine. Proceedings of the 4th International Symposium on Nutritional Aspects of Osteoporosis. Ares-Serono Symposia Publications, Academic Press, London, 379–397. New, S.A., MacDonald, H.M., Campbell, M.K., Martin, J.C., Garton, M.J., Robins, S.P., and Reid, D.M. 2004. Lower estimates of net endogenous non-carbonic acid production are positively associated with indexes of bone health in premenopausal and perimenopausal women. AM J Clin Nutr 79: 131–138. Reed, J.A., Anderson, J.J.B., Tylavsky, F.A., et al. 1994. Comparative changes in radial bone density of elderly female lactoovovegetarians and omnivores. Am J Clin Nutr 59: 1197S–1202S. Reidenberg, M.M., Haag, B.L., Channick, B.J., et al. 1966. The response of bone to metabolic acidosis in man. Metabolism 15: 236–241. Remer, T., and Manz, F. 1995. Potential renal acid load of foods and its influence on urine pH. J Am Diet Assoc 95: 791–797. Rizzoli, R., Schurch, M.A., Chevalley, T., et al. 1998. Protein intake and osteoporosis. In Nutritional Aspects of Osteoporosis 1997, Burckhardt, P., Dawson-Hughes, B., Heaney R.P., eds. Proceedings of the 3rd International Symposium on Nutritional Aspects of Osteoporosis. Ares-Serono Symposia Publications, Springer-Verlag, New York, 141–154. Sebastian, A., Hernandez, R.E., Portale, A.A., et al. 1990. Dietary potassium influences kidney maintenance of serum phosphorus concentrations. Kidney Int 37: 1341–1349. Tesar, R., Notelovitz, M., Shim, E., et al. 1992. Axial and peripheral bone density and nutrient intakes of postmenopausal vegetarian and omnivorous women. Am J Clin Nutr 56: 699–704.

Acid–Base Balance

279

Tylavsky, F., and Anderson, J.J.B. 1988. Bone health of elderly lactoovovegetarian and omnivorous women. Am J Clin Nutr 48: 842–849. Wachman, A., and Bernstein, D.S. 1968. Diet and osteoporosis. Lancet I: 958–959. Widdowson, E.M., McCance, R.A., and Spray, C.M. 1951. The chemical composition of the human body. Clin Sci 10: 113–125. Wood, R.J. 1994. Potassium bicarbonate supplementation and calcium metabolism in postmenopausal women: Are we barking up the wrong tree? Nutr Rev 52: 278–280.

19

Antioxidants and Bone Health Martin Kohlmeier

CONTENTS Introduction..................................................................................................................................... 281 Physicochemical and Metabolic Features of Antioxidants............................................................. 281 Functions of Antioxidants in Bone Metabolism............................................................................. 283 Vitamin C................................................................................................................................... 283 Carotenoids................................................................................................................................ 283 Vitamin E................................................................................................................................... 283 Selenium.....................................................................................................................................284 Human Data....................................................................................................................................284 Vitamin C...................................................................................................................................284 Carotenoids................................................................................................................................284 Vitamin E...................................................................................................................................286 Selenium.....................................................................................................................................286 A Well Balanced Antioxidant Diet.................................................................................................. 286 Conclusions..................................................................................................................................... 287 References....................................................................................................................................... 287

INTRODUCTION Antioxidants have long been touted as an insurance against cancer, cardiovascular disease, and other chronic ailments. Much less attention has been given to the ability of antioxidants to protect against accelerated bone mineral loss and bone fracture. Now that robust evidence has burst the bubble of exuberant expectations for the health benefits of antioxidants, more sober considerations still point to an important role of antioxidants in foods for the protection of bone health. Consistent experimental and clinical evidence leaves little doubt that several food-derived antioxidants protect critical events in bone cell differentiation and function. The most important question for every individual is then whether additional intakes can improve bone health and reduce fracture risk. Aslowly growing number of high-quality human studies can now provide some answers and guide the selection of healthy foods for strong bones. This chapter reviews the known effects of the antioxidants in human bone, and some questions about what is not known are raised.

PHYSICOCHEMICAL AND METABOLIC FEATURES OF ANTIOXIDANTS Antioxidants are defined by their ability to neutralize or otherwise counteract oxidants, such as reactive oxygen molecules (superoxide anion, hydrogen peroxide, and the hydroxyl radical), transition metals (particularly, iron and copper), hypochlorite, and reactive nitrogen compounds (e.g., peroxynitrite). Some of these oxidants are free radicals, meaning that they contain unpaired electrons. All of them are highly reactive, with the potential for damaging DNA, proteins, and 281

282

Diet, Nutrients, and Bone Health

lipids. Several vital processes naturally generate significant amounts of free radicals. The superoxide anion, for instance, is a regular side product of oxidative phosphorylation due to unavoidable mechanistic inefficiencies (Sun and Trumpower, 2003). Intense and prolonged physical exertion (Hattori et al., 2009) slightly increases the release of reactive oxygen species (ROS). Significant amounts of ROS also come from the breakdown of purines (superoxide anion generated by xanthine oxidase), the production of prostanoid mediators (peroxides leaked from lipoxygenases), and the cellular immune response (myeloperoxidase-generated hypochlorous acid and hypobromous acid). Macrophages and eosinophils release highly corrosive bursts of hypochlorous acid and hypobromous acid, which can neutralize bacteria and parasites. These natural immune defense reactions highlight the fact that several oxidants have important physiological functions. Harmful amounts of free radicals also come from exposure to external influences. Both active (Bloomer, 2007) and passive (Valkonen and Kuusi, 2000) cigarette smoke inhalation constitutes a particularly consequential source. Exposure to industrial pollutants, such as polycyclic aromatic hydrocarbons and toluene, is another potential inducer of ROS, usually in a work-related context. Oxidant free radicals also function as modulators of intracellular signaling events. In particular, receptor activator of NF-κB ligand (RANKL) uses ROS to induce long-lasting oscillations of intracellular concentration of ionized calcium, which promotes the differentiation of monocytes from bone marrow into osteoclasts (Ha et al., 2004; Kim et al., 2010) and increases the rate of bone resorption (Darden et al., 1996). It is no surprise, therefore, that antioxidants slow osteoclast activity and motility by their ability to remove ROS (Bax et al., 1992). For example, the potent endogenous antioxidant lipoic acid was found to inhibit osteoclast differentiation (Kim et al., 2006). Exposure of osteoblasts, on the other hand, to ROS appears to slow proliferation by inducing cell cycle arrest (Li et al., 2009) and limit differentiation by heightened NF-κB p65 phosphorylation (Zhong etal.,2009). Additional complexity comes from extracellular ROS signaling. Superoxide anion generated during muscle contraction, for instance, appears to enhance maximal force generation (GomezCabrera et al., 2010), which might impact bone health indirectly. Furthermore, ROS can increase expression of free-radical scavenging enzymes, such as superoxide dismutase, through activation of specific redox-sensitive transcription factors (Hollander et al., 2001). Antioxidants have diverse physicochemical properties. Vitamin E and the carotenoids are highly lipophilic due to their aliphatic nature. They are transported in blood with lipoproteins, fit well into cell membranes, and accumulate in adipose tissue. Vitamin C, on the other hand, is a hydrophilic sugar alcohol, which is transported in blood plasma and distributes in cells mainly to the cytosol. Uric acid is another potent hydrophilic antioxidant, particularly in blood (Beretta et al., 2006). The major antioxidant metabolite in cells is glutathione, which mediates the enzyme-catalyzed reactivation of antioxidants. The metabolic interactions of antioxidants are often complex. For example, the reaction of alphatocopherol with superoxide anion neutralizes the free radical but converts alpha-tocopherol itself to tocopherylquinone epoxides, potent free radicals themselves (Podmore et al., 1998). Similarly, the reaction of ascorbic acid with a hydroxyl radical gives rise to semidehydroascorbic acid, which is a free radical itself. This may explain why excessive intakes of both vitamin C and vitamin E appear to increase free-radical load under some circ*mstances and promote diseases usually associated with free-radical exposure (Terentis et al., 2002; Levy et al., 2004). Selenium is an essential cofactor of several enzymes with antioxidant properties and of others that reactivate antioxidant compounds. Both tocopheryl quinone and the ascorbyl radical have to be detoxified by a system that relies on the reductant glutathione and the selenoenzyme thioredoxin reductase. The importance of the glutathione redox system for bone health is exemplified by the observation that consumption of N-acetylcysteine, which increases tissue glutathione levels, inhibits bone loss after ovariectomy (Lean et al., 2003).

Antioxidants and Bone Health

283

Many natural food constituents also are potent antioxidants. The overwhelming majority comes from plant foods and herbal teas and includes hundreds of different molecular species of carotenoids and flavonoids. The vibrant colors of fruits and vegetables usually indicate the presence of carotenoids, anthocyanines, and other antioxidants.

FUNCTIONS OF ANTIOXIDANTS IN BONE METABOLISM The four major nutrient antioxidants, that is, vitamin C, carotenoids, vitamin E, and selenium, have important functions in bone cells that impact on both bone formation and resorption.

Vitamin C Ascorbic acid is needed for collagen maturation. The reduced form acts in concert with iron as an essential cofactor of procollagen-proline dioxygenase and procollagen-lysine 5-dioxygenase. Inadequate vitamin C availability slows collagen synthesis and thereby bone growth and remodeling. Vitamin C is also the precursor of ascorbate 2-sulfate, which is needed for the synthesis of bone glucosaminoglycans, such as chondroitin and dermatansulfate.

Carotenoids An enzyme in the intestinal wall, 16,16′-oxygenase, cleaves a small percentage of ingested betacarotene, alpha-carotene, and a few other provitamin A carotenoids into retinol (vitamin A). An adequate supply of the vitamin A is important for bone health, not the least because its metabolite retinoic acid acts on multiple gene targets, either alone through retinoic acid receptors or through retinoic acid X receptors in combination with vitamin D and other cofactors. Even a slight excess of preformed vitamin A, on the other hand, is likely to promote osteoclastic activity and ultimately jeopardize bone health (Crandall, 2004). The majority of carotenoids in foods are not vitamin A precursors because their cleavage product does not yield retinol. An important example of such non-provitamin A carotenoids is lycopene, which lacks the conjugated rings of beta-carotene. Lycopene appears to promote proliferation and differentiation of osteoblast-like cells in vitro (Park et al., 1997; Kim et al., 2003). Osteoclastic activity of cultured cells, on the other hand, was found to be inhibited by lycopene (Rao et al.,2003).

Vitamin E At least 10 natural compounds possess characteristic vitamin E features, including their lipophilic character and their strong antioxidant activity. The major forms in plant-based foods are alpha-tocopherol and gamma-tocopherol, whereas beta- and delta-tocopherol are present in minor amounts. There are also four tocotrienols, which are tocopherol analogs with three instead of two double bonds in their side chain. Rice oil and palm oil are particularly rich in tocotrienols. People who maintain a low-fat diet and avoid fatty foods cannot meet recommended intake levels because these vitamin E-like compounds come mainly from high-fat foods, such as oils, seeds, and nuts. Much of the vitamin E added to foods (to protect them against oxidation) and dietary supplements is synthetic alpha-tocopherol. The chemical synthesis produces a racemic mixture of stereoisomers with reduced bioavailability and activity. Just like other antioxidants, vitamin E appears to limit ROS-induced RANKL signaling and thereby slow excessive osteoclast activation. The hydrophilic vitamin E analog Trolox was found to counteract the induction of RANKL, possibly due to inhibition of cyclooxygenase-2 activity (Lee etal., 2009). Judging by cell culture studies, vitamin E also may protect against the inhibitory effect of oxidized lipoproteins on osteoblast differentiation and osteopontin expression (Maziere et al.,

284

Diet, Nutrients, and Bone Health

2010). These findings are somewhat at odds with other in vitro observations suggesting an inhibitory effect of vitamin E on early osteoblast differentiation (Soeta et al., 2010).

Selenium Selenium, in the form of the amino acid selenocysteine, is present in several enzymes that maintain normal redox status, reduce oxidized lipids, and reactivate antioxidants. Several versions of glutathione peroxidase defang hydrogen peroxide and other ROS, repair lipids, and reactivate oxidized vitamin C. A closely related mitochondrial enzyme, phospholipid hydroperoxide glutathione peroxidase (GPX4), uses glutathione to detoxify lipid hydroperoxides. This enzyme protects cell membranes against damage from the highly corrosive effects of oxidative phosphorylation. Thioredoxin reductases are unusually ubiquitous enzymes that all known organisms use. The mammalian versions are selenoproteins and contribute to the reactivation of vitamin C, maintain the functional thiol groups of proteins in their optimal redox state, and activate NF-κB and other transcription factors. Which of these enzyme functions specifically contributes to bone health remains unclear. Protection against oxidative damage and ensuing cell death is likely to play a factor. The recycling of oxidized vitamin C may be of even greater importance. A strict ceiling effect of selenium intake on enzyme function can be expected a priori because the selenium content of the enzymes is governed by constant stochiometric ratios.

HUMAN DATA Human investigations of the four types of nutrient antioxidants are reviewed.

Vitamin C Mechanistic considerations certainly argue for an important role of vitamin C status in maintaining bone health. However, it is not even clear whether people with scurvy have an increased risk of osteoporosis or bone fracture. Data from healthy populations are sparse. The prospective observation of more than 11,000 postmenopausal women indicated a statistically significant effect of vitamin C consumption levels on bone mineral density, which disappeared after adjustments for bone-related factors (Wolf et al., 2005). Nonetheless, women with the highest quartile of vitamin C intake had a persistently greater bone-density-promoting effect of hormone replacement therapy than women with lower intakes. Low vitamin C concentration in blood, but not vitamin C intake level, appeared to be related to increased risk of osteoporotic fractures (MartinezRamirez et al., 2007). In a cross-sectional study of postmenopausal women in Japan, radial bone mineral density was found to be highest in participants with high vitamin C consumption (Sugiura et al., 2010). A 4-year prospective cohort study found that higher vitamin C intake slowed bone mineral loss in elderly men, but not in women (Sahni et al., 2008). To put these effects into perspective, it has been suggested that 1% decrease in BMD translates into a 12% increase of hip fracture risk (Nguyen et al., 2005). One might then assume that when men with relatively low vitamin C intake get an extra 60 mg, their hip fracture risk might decrease by as much as 18%. Adding about 200 mg per day might lower hip fracture risk by 34% (see Table 19.1).

Carotenoids Data from most prospective cohort studies are compatible with a beneficial effect of high carotenoid intake with fruits and vegetables on bone health. The strongest effect appears to be attributable to lycopene, and more modest effects, if any, may be associated with beta-carotene. In an investigation of healthy postmenopausal women, for instance, women with osteoporosis had lower serum

285

Antioxidants and Bone Health

TABLE 19.1 Daily Food Amount That on Its Own Might Reduce Hip Fracture Risk by 20% When Added to a Diet with Modest Vitamin C Content Food Amount

Vitamin C (mg)

Energy (kcal)

126   87   85   93   69   87

37 55 46 85 64 32

Guava, 1/3 cup (83 g) Red papaya, 1 cup cubes (140 g) Strawberries, 1 cup (144 g) Orange juice, medium glass (180 ml) Grapefruit, whole small fruit (200 g) Broccoli, 3 spears (96 g)

TABLE 19.2 Daily Food Amount That on Its Own Might Reduce Hip Fracture Risk by 20% When Added to a Low-Carotenoid Diet Food Amount Water melon, 1/4 of a wedge (75 g) Red papaya, 1 cup cubes (140 g) Guava, 1/3 cup (83 g) Tomato juice, 2 tablespoons (30 ml) Spaghetti sauce, 1 tablespoon (15 ml) Pepperoni pizza, 2 slices

Lycopene (µg)

Carotenoids (µg)

Energy (kcal)

3241 3122 2862 2745 2628 2536

3520 4678 3068 2845 2681 2696

  22   55   37    5    6 394

lycopene and cryptoxanthin concentrations than those in women with higher bone mineral density, whereas the concentration of beta-carotene tended to be higher in women with osteoporosis (Yang et al., 2008). In the Women’s Health Initiative study (Wolf et al., 2005), on the other hand, only betacarotene intake was associated with bone mineral density at the femoral neck and at all sites combined, whereas no association was observed for alpha-carotene, lycopene, lutein, or cryptoxanthin. The benefits from other carotenoids may well be obscured by a lack of reliable dietary data. Smokers may have the greatest bone health benefit from carotenoids (Melhus et al., 1999; Zhang etal., 2006), but high carotenoid intake from foods also was related to bone mineral density in nonsmoking elderly women (Barker et al., 2005; Pasco et al., 2006). The strongest and practically most applicable data come from the Framingham cohort study (Sahni et al., 2009b). These findings suggest that adding about one table spoon of tomato sauce daily (providing about 2600 µg lycopene/day) to the diet of someone with low intake may decrease the long-term hip fracture risk by about 20% (see Table 19.2). Adding daily half a cup (10,000 µg) might cut this risk nearly in half. The same 50% risk reduction might be achieved by adding to the daily menu two medium carrots or half a cup of cooked spinach (each providing 15,000 µg total carotenoids). In real life, of course, the benefit should accrue from any combination of foods rich in total carotenoids and/or lycopene. A correspondingly slower rate of bone mineral loss in the same cohort fully explains the reduced fracture risk (Sahni et al., 2009a). A recent clinical trial investigated the potential clinical benefit of several dietary supplements providing 15,000–70,000 µg of lycopene to postmenopausal women for 4 months. The reportedly lower excretion of N-telopeptide is consistent with slower bone resorption as a potential mechanism (Mackinnon et al., 2010). Although at least some carotenoids may lower osteoporosis risk, it is important to remember that several studies have observed accelerated bone mineral loss in people with high retinol intake (Melhus et al., 1998; Feskanich et al., 2002; Promislow et al., 2002), although the absence of such an

286

Diet, Nutrients, and Bone Health

effect was also found (Ballew et al., 2001). This issue becomes important when choosing a multivitamin supplement because both retinyl esters and beta-carotene are available as vitamin A sources in commercial products.

Vitamin E Age-adjusted data from the Women’s Health Initiative (Wolf et al., 2005) demonstrated higher bone mineral density in women with higher than average vitamin E intake (mostly from dietary supplements). However, further adjustment for multiple relevant factors abolished the statistical significance of this association. A greatly underpowered randomized controlled study of vitamin C (1000 mg/day) and vitamin E (400 mg) of only 6 months duration failed to show clinically relevant benefit (Chuin et al., 2009).

Selenium Among the participants of a large population-based case–control study, risk of osteoporotic hip fracture was inversely associated with selenium intake in smokers, but not at all in those who never smoked (Zhang et al., 2006).

A WELL BALANCED ANTIOXIDANT DIET Making a habit of including fruits and vegetables with most meals is a solid basis for protecting bone health. Diverse population studies consistently find that high fruit and vegetable intakes protect bone mineral density (Tucker et al., 1999; Macdonald et al., 2004; Prynne et al., 2006). The high content of several antioxidant nutrients, particularly of vitamin C and lycopene, explains the beneficial effect of fruits and vegetables on bone health to a significant degree, in addition and in synergy with other favorable nutrients. This is the main reason foods should be the preferred source of these antioxidants. The availability of vitamin E from foods alone is much more limited (Chun et al., 2010). Americans get on average less than 8 mg of alpha-tocopherol from foods, about half of what is recommended. Vegetables and whole-grain foods are an important source of vitamin E, but it would be difficult to meet requirements with these foods alone. Additional vitamin E sources are oils, nuts, and seeds. People who avoid fatty foods have a particularly hard time to get enough vitamin E from foods alone. Although moderate amounts of supplemental vitamin E are likely to be beneficial, high-dosed products (>200 IU/day) should be avoided in light of slightly increased mortality and health risks (Bjelakovic et al., 2007). A mixed diet is also a solid foundation to get enough selenium, but the amounts in foods vary greatly by region of origin. Americans get on average 109 µg selenium from foods, which is well above the 55 µg that most adults should get. Good foods for boosting selenium intakes are ocean fish and other sea food. For foods and beverages rich in phenolic antioxidants, the picture is more mixed and often inconclusive due to confounding by other food constituents. The polyphenols in soy (mainly the isoflavones genistein, daidzein, and glycetin) and other legumes have received probably the most attention and are discussed elsewhere. Drinkers of tea (Camellia sinensis), who consume several hundred milligrams of polyphenols per cup (including ellagic acid, catechins, theaflavins, and thearubigins), may protect the structural integrity of their bones, most importantly at the hip, nonetheless (Hegarty et al., 2000; Devine etal., 2007). No such beneficial effect can be expected from coffee despite its very high content of phenolic antioxidants (a hundred milligrams or more of chlorogenic acid per cup). If anything, consumption of several cups of caffeinated coffee may accelerate bone mineral loss and increase bone fracture

Antioxidants and Bone Health

287

risk, particularly in women with low calcium intake (Rapuri et al., 2001; Hallstrom et al., 2006). However, the results of the numerous reported studies are divergent and do not allow for a final conclusion, yet (Higdon and Frei, 2006). Particularly disappointing for many may be the apparently unfavorable effect of chocolate consumption, which was associated in a linear fashion with lower bone density in one prospective cohort study despite its very high content of catechins, epicatechins, procyandins, and other flavonoids with antioxidant properties (Hodgson et al., 2008).

CONCLUSIONS What we know are the following: ROS participate in cell signaling and other physiological events, but exposure to excessive ROS concentrations endangers bone health by inappropriately promoting osteoclastic activity and interfering with normal osteoblast differentiation and function. Various antioxidant nutrients help to inactivate ROS and protect normal function and regulation of bone cells. Generous intakes of vitamin C and lycopene with a diet rich in fruits and vegetables reduce the risk of accelerated bone mineral loss and osteoporosis more than generally appreciated. Even modest targeted intake adjustments that are easy to implement will reduce bone fracture risk by more than 20%. Such benefits are less ensured if most of these nutrients come from dietary supplements. What we do not know are the following: Adequate amounts of vitamin E and selenium may also promote bone health, but the available human data are weak and inconsistent across populations and bone sites. Troubling questions about the use of high-dosed supplements as the main source of these nutrients, particularly for vitamin E, remain. Some flavonoids and other polyphenols with antioxidant properties in fruits, vegetables, and green tea may well have beneficial effects, but more high-quality studies will be required to substantiate claims.

REFERENCES Ballew, C., Galuska, D., and Gillespie, C., 2001. High serum retinyl esters are not associated with reduced bone mineral density in the Third National Health And Nutrition Examination Survey, 1988–1994. J Bone Miner Res 16, 2306–2312. Barker, M.E., McCloskey, E., Saha, S., et al., 2005. Serum retinoids and beta-carotene as predictors of hip and other fractures in elderly women. J Bone Miner Res 20, 913–920. Bax, B.E., Alam, A.S., Banerji, B., et al., 1992. Stimulation of osteoclastic bone resorption by hydrogen peroxide. Biochem Biophys Res Commun 183, 1153–1158. Beretta, G., Aldini, G., Facino, R.M., et al., 2006. Total antioxidant performance: A validated fluorescence assay for the measurement of plasma oxidizability. Anal Biochem 354, 290–298. Bjelakovic, G., Nikolova, D., Gluud, L.L., et al., 2007. Mortality in randomized trials of antioxidant supplements for primary and secondary prevention: Systematic review and meta-analysis. JAMA 297, 842–857. Bloomer, R.J., 2007. Decreased blood antioxidant capacity and increased lipid peroxidation in young cigarette smokers compared to nonsmokers: Impact of dietary intake. Nutr J 6, 39. Chuin, A., Labonte, M., Tessier, D., et al., 2009. Effect of antioxidants combined to resistance training on BMD in elderly women: A pilot study. Osteoporos Int 20, 1253–1258. Chun, O.K., Floegel, A., Chung, S.J., et al., 2010. Estimation of antioxidant intakes from diet and supplements in U.S. adults. J Nutr 140, 317–324. Crandall, C., 2004. Vitamin A intake and osteoporosis: A clinical review. J Womens Health (Larchmt) 13, 939–953. Darden, A.G., Ries, W.L., Wolf, W.C., et al., 1996. Osteoclastic superoxide production and bone resorption: Stimulation and inhibition by modulators of NADPH oxidase. J Bone Miner Res 11, 671–675. Devine, A., Hodgson, J.M., Dick, I.M., et al., 2007. Tea drinking is associated with benefits on bone density in older women. Am J Clin Nutr 86, 1243–1247. Feskanich, D., Singh, V., Willett, W.C., et al., 2002. Vitamin A intake and hip fractures among postmenopausal women. JAMA 287, 47–54. Gomez-Cabrera, M.C., Close, G.L., Kayani, A., et al., 2010. Effect of xanthine oxidase-generated extracellular superoxide on skeletal m