Clinical Reviews in Bone and Mineral Metabolism

, Volume 7, Issue 3, pp 240–248

Adipokine Effects on Bone

Authors

    • Faculty of Medical and Health SciencesUniversity of Auckland
  • J. B. Richards
    • Departments of Medicine and Human Genetics, Lady Davis InstituteMcGill University
    • Department of Twin Research and Genetic EpidemiologySt. Thomas’ Hospital
Original Paper

DOI: 10.1007/s12018-009-9048-4

Cite this article as:
Reid, I.R. & Richards, J.B. Clinic Rev Bone Miner Metab (2009) 7: 240. doi:10.1007/s12018-009-9048-4
  • 85 Views

Abstract

The adipocyte is an important source of factors that act as circulating regulators of bone metabolism. These include estrogens, and the adipokines, leptin, resistin, adiponectin, and probably others. Leptin acts directly on bone cells, and in some experimental models these effects are modified by its actions on the central nervous system, which impact on appetite, body weight, and insulin sensitivity. While not strictly an adipokine, insulin circulates in increased concentrations in obesity and exerts anabolic effects on bone. Adipokine levels correlate with bone turnover, suggesting that they dynamically influence bone metabolism. In postmenopausal women, they may be among the principal regulators of bone turnover, accounting for their increasing importance as determinants of bone density with age. Of the adipokines, adiponectin appears to have the strongest relationships with bone parameters in postmenopausal women.

Keywords

AdiponectinLeptinInsulinFatLeanResistin

Introduction

Adipocyte biology has been an area of rapid recent advances. Whereas, 20 years ago the adipocyte was not generally considered to play a significant regulatory role, it is now well accepted that it is an important source of factors that act as circulating regulators of metabolism, a phenomenon recognized in the coining of the term “adipokine”. Leptin is the most widely studied of these, but the number of regulatory factors known to be secreted by adipocytes is steadily growing and their roles in bone biology remain to be elucidated.

Estrogens

The adipocyte has long been recognized as an estrogen-producing cell, particularly in postmenopausal women. Thus, early postmenopausal women who lose bone rapidly have lower levels of both estrone and estradiol than “slow losers”, and this may be contributed to by their lower fat mass [1]. Our own study has confirmed a relationship between circulating estrone levels and bone density, but has shown this to be both independent of the effects of fat mass and substantially weaker [2]. This implies that estrogen is not the only pathway by which fat influences bone density, a suggestion supported by the finding of a fat–bone relationship in premenopausal women, in whom the adipocyte is a relatively minor source of estrogens.

Leptin

Over the last decade, leptin has come to be regarded as the classic adipocyte hormone, and it is now clear that it can directly modulate the activity of bone cells, as well as indirectly influence the skeletal metabolism through its actions on the central nervous system.

It has been stated that the leptin receptor does not exist in bone cells [3] so that all effects of this hormone on bone are mediated through its central nervous system actions. However, the signaling form of the leptin receptor has now been shown by many groups to be expressed in both osteoblasts and chondrocytes [4, 5], meaning that the direct effects of leptin on bone need to be integrated with whatever it might do centrally. Leptin has now been shown to increase proliferation and differentiation in osteoblasts [57] and to promote mineralized nodule formation [8, 9]. It has similar stimulatory effects on chondrocytes, both in vitro and in vivo [5, 10], so probably acts in vivo to increase growth of long bones. This would account for the short limbs of animals with impaired leptin signaling, as reviewed by Hamrick in this volume.

Leptin also directly regulates osteoclast development, reducing production of RANK and RANK-ligand, and increasing osteoprotegerin in cultures of human bone marrow stromal cells and in peripheral blood mononuclear cells [11, 12]. Thus, leptin inhibits osteoclastogenesis [5, 11]. This would work in concert with its osteoblast effects to promote skeletal preservation, consistent with a need for individuals with high fat mass to have stronger skeletons to support their greater soft tissue mass.

The Karsenty group have dramatically changed views of the regulation of bone biology with an elegant series of papers showing that infusion of leptin into the third ventricle causes bone loss in leptin-deficient and wild-type mice through inhibition of bone formation [3] and stimulation of bone resorption [13]. Blockade of the sympathetic nervous system abrogates these effects, which appear to be mediated by β-adrenoreceptors on osteoblasts [14]. These findings, coupled with their belief that there is no direct action of leptin on bone, have given birth to a new paradigm of the regulation of bone mass by the brain. However, this paradigm overlooks the fact that leptin is a systemic hormone produced outside the central nervous system and that its central effects are balanced by its direct effects on bone cells. In fact, leptin is produced in the bone marrow environment in marrow adipocytes, but also in chondrocytes and cells of the osteoblast lineage [15], so its local effects on bone might be expected to be dominant. The second fact overlooked by the brain–bone hypothesis is that the central administration of leptin produces a dramatic reduction in appetite leading to profound weight loss, adipocyte shrinkage, reduced serum levels of both leptin and insulin, and increases in ghrelin [1618]. Thus, much of the bone loss observed with central administration of leptin might simply reflect the profoundly catabolic state of these animals, a suggestion supported by the finding that bone loss can be produced by caloric restriction alone [19]. Both central administration of leptin and weight loss cause reductions in circulating levels of insulin, which will also have secondary effects on bone. The potential interactions of these effects are depicted in Fig. 1. While effects of the sympathetic nervous system on bone have been known for many years, they may be relatively unimportant in most contexts since beta adrenergic blockade does not consistently influence bone density, rates of bone loss, or fracture [20]. In a prospective randomized controlled trial, we did not find that propranolol stimulated bone formation [21], as would be expected if sympathetic tone was a potent regulator of this function.
https://static-content.springer.com/image/art%3A10.1007%2Fs12018-009-9048-4/MediaObjects/12018_2009_9048_Fig1_HTML.gif
Fig. 1

Possible mechanisms for central leptin effects on bone, via reduced peripheral insulin and leptin levels. SNS = sympathetic nervous system. © IR Reid, used with permission

Studies of leptin deficiency and systemic leptin administration in animals support these general conclusions. Leptin deficiency is associated with reduced linear growth, reduced cortical bone mass, increased trabecular bone of the spine but reduced trabecular bone in the femora, where huge adipocytes occupy much of the marrow space [22]. Systemic administration of leptin in animals (intact or leptin-deficient) usually results in an improvement in bone formation, skeletal mass or strength [4, 5, 12, 23, 24]. In leptin-deficient animals, leptin replacement reverses the adipocyte phenotype and increases total body bone mineral content by >30% [23]. However, systemic administration of high doses of leptin may have negative effects on bone, similar to its central administration [25], because these doses are associated with weight loss and decreases in serum IGF-I levels.

Studies of leptin administration in humans are few. Farooqi [26] provided leptin replacement to a 9-year old leptin-deficient girl, and observed weight loss accompanied by bone gain. Welt et al. [27] treated eight women with hypothalamic amenorrhea with leptin for up to 3 months. Leptin administration resulted in significantly increased levels of estradiol, thyroid hormones, IGF1, insulin-like growth factor-binding protein-3, bone alkaline phosphatase, and osteocalcin, demonstrating the many indirect mechanisms by which this hormone can impact on the skeleton. Both these studies indicate that the skeletal effects of systemic leptin in humans are positive.

There are now a number of cross-sectional studies in humans which relate leptin to bone parameters. Interpretation of such studies is problematic, since there are co-dependencies of the bone indices and of leptin on fat mass, which cannot be adequately disentangled by statistical methods. With these caveats in mind, it has been reported that circulating leptin concentrations are inversely related to bone resorption in young adults [28]. In the healthy human fetus, leptin is positively associated with indices of bone formation, and inversely with resorption [29]. In postmenopausal women, circulating leptin is inversely related to bone resorption after adjustment for fat mass [30] or BMI [31], and is directly correlated with bone density at the hip, spine, and forearm [31]. These results, again, suggest that the final integration of leptin actions on the skeleton is positive. Ultimately, this dominance is attested to by the consistent positive relationship between fat mass and bone density—if the central effects of leptin were dominant, there would be an inverse relationship.

Adiponectin

Adiponectin is now emerging as a very important adipokine. It is a 28 kD protein with close homology to compliment factor C1q and collagens VIII and X [32]. This protein circulates in humans as trimers, hexamers, and high-molecular weight oligomers [33, 34], in total concentrations of 0.5–30 μg/ml [35] and is almost exclusively secreted by adipocytes. Plasma concentrations of adiponectin are inversely related to visceral fat mass and body mass index [35]. This paradoxical relationship is yet to be fully explained but may be mediated by inhibition of adiponectin secretion by cytokines (tumor necrosis factor-α, interleukin-6) and hormones (cortisol, testosterone) that are increased in obesity, by adipose tissue hypoxia, or by direct inhibition of its own production [36]. Adiponectin regulates energy homeostasis, glucose and lipid metabolism, and inflammatory pathways [37]. Serum adiponectin levels correlate well with independent measures of insulin sensitivity, and low levels of adiponectin are excellent predictors of insulin resistance in diverse population settings [3840]. Hypoadiponectinemia in humans is associated with cardiovascular and metabolic disorders such as coronary artery disease and type 2 diabetes mellitus [41, 42].

Adiponectin is particularly well-suited to measurement in large epidemiologic studies due to its long half life and high ex vivo stability [43, 44]. While adiponectin is generally thought to be related to insulin resistance through adiposity, recent evidence suggests that adiponectin may act as a downstream marker of tissue insulin exposure. Evidence for this assertion comes from suppression of adiponectin with intravenous insulin infusion, and a decrease in adiponectin has been noted in insulin deficient states such as type 1 diabetes [4548] and anorexia nervosa [49, 50]. Consequently, this preliminary evidence suggests that adiponectin may be a tractable biomarker for insulin exposure in large epidemiologic studies.

There is a small but contradictory literature regarding adiponectins effects on bone in laboratory studies. The receptors for adiponectin, AdipoR1, and AdipoR2, have been identified on both osteoblasts and osteoclasts [51, 52] implying a potential direct influence of this hormone on bone. Adiponectin has been shown to increase osteoblast proliferation and differentiation whilst inhibiting osteoclastogenesis in vitro [53, 54]. These actions would be expected to result in a net increase in bone mass in vivo. In support of this possibility, transient over-expression of adiponectin in mice increased trabecular bone mass and inhibited osteoclast number and bone resorption [53]. However, others have reported that adiponectin reduces osteoblast growth [51], and other animal studies have produced inconsistent results. A study of transgenic mice either over-expressing or deficient in the adiponectin gene showed no abnormality in bone phenotype at 8 weeks of age [51]. Another study, reported only in abstract, found adiponectin-deficient mice to have increased bone volume at 9 months of age but normal values at 3 months [55].

We have studied adiponectin effects on bone in vitro, and in knockout mice [56]. We find that adiponectin is dose-dependently mitogenic to primary osteoblasts (60% increase at 10 μg/ml), and markedly inhibits osteoclastogenesis (by 26% and 54% at 1 and 5 μg/ml, respectively). It has no effect on bone resorption in isolated mature osteoclast assays. We have also carried out micro-computed tomography on 14-week old adiponectin knockout mice. Trabecular bone volume is increased by 30% and trabecular number by 38% in these animals. These data indicate that adiponectin acts directly on bone cells, but that these actions do not explain the bone phenotype of the knockout animals. Thus, adiponectin may also exert indirect effects on bone, possibly through modulating insulin sensitivity or growth factor binding. The magnitude of the changes seen in the data from the knockout mice indicates that adiponectin is likely to be a significant contributor to the fat-bone relationship.

In contrast to the variable findings in cell culture and animal studies, clinical studies demonstrate that circulating adiponectin concentrations are consistently inversely related to bone mineral density [5759]. For example, Lenchik [58] found that serum adiponectin was inversely associated with areal BMD (r = −0.20 to−0.3), volumetric BMD (r = −0.35 to−0.44), and visceral fat volume (r = −0.30) in 80 men and women, most of whom had diabetes. These associations remained significant after adjusting for whole body fat mass. Richards [59] reported a similar study in non-diabetic women, and found that each doubling of serum adiponectin was associated with decreases in BMD of 2.0–3.2%, at the sites assessed. This relationship persisted after adjustment for potential confounding factors, including BMI, serum leptin, central fat mass, and exercise. Further analyses of these data are presented below.

In summary, adiponectin has direct actions on bone cells, and adiponectin knockout mice have increased bone mass. Taken together with the clinical data, this suggests that the final sum of direct and indirect actions of adiponectin on bone mass in vivo is negative. Moreover, epidemiologic evidence suggests that the relationship between adiponectin and BMD appears to be distinct from other potential sources of influence arising from fat, such as body weight and leptin levels. Thus, adiponectin does appear to be an adipokine that independently impacts on bone mass.

Resistin

Resistin was identified as a product of the adipocyte during a search for genes which are down-regulated by thiazolidinedione anti-diabetic drugs. We have demonstrated that resistin modestly increases the proliferation of osteoblasts in both cell and organ culture systems [60]. It also increases the formation of osteoclasts in bone marrow culture, and their activity in organ culture. Thommesein has reported similar findings [61]. Whether these counter-balancing effects lead to any change in bone mass is not known at present.

Novel Adipokines

With the rapidly increasing knowledge of adipocyte biology, a number of new factors secreted from these cells have been identified. Those believed to act as adipokines include visfatin, retinol binding protein-4, fasting-induced adipocyte factor, and adipsin. To our knowledge, the effects on bone of these factors have not been studied. Since leptin and adiponectin appear to only partially account for the fat–bone relationship and since there remain significant unanswered questions in this area, casting the net wider for other mediating factors is important if this critical determinant of bone density is to be more fully understood.

Beta Cell Hormones

The previous chapter reviews the role of beta cell hormones on bone in the context of bone-active factors that are responsive to feeding. It is important to realize that insulin, and the factors co-secreted with it, amylin and preptin, also circulate in increased concentrations in obesity. Therefore, these factors are chronically elevated in obese individuals, quite independent of feeding effects. As outlined above, insulin, amylin, and preptin are directly stimulatory to osteoblast growth, and amylin also has a potent calcitonin-like effect, inhibiting osteoclastic bone resorption. However, insulin may have indirect effects to increase bone density. Hyperinsulinemic women display increased androgen and estrogen production in the ovary, and insulin directly inhibits production of sex hormone binding globulin in the liver. As a result, free concentrations of both androgens and estrogens are increased, with effects on both osteoblasts and osteoclasts which will tend to increase bone mass. The interaction of these activities is shown in Fig. 2. As a result, high bone density is a very consistent finding across a wide range of hyperinsulinemic states, including obesity, polycystic ovary syndrome [62], and congenital generalized lipodystrophy [63]. The latter is particularly significant because it represents a dissociation of fat mass and insulin levels. These findings are consistent with the literature showing lower bone densities in insulinopenic diabetes and increased facture risk in these subjects [64].
https://static-content.springer.com/image/art%3A10.1007%2Fs12018-009-9048-4/MediaObjects/12018_2009_9048_Fig2_HTML.gif
Fig. 2

Summary of the principal mechanisms by which the hyperinsulinemia associated with obesity contributes to increased bone mass. © IR Reid, used with permission

In clinical studies, circulating insulin levels, both fasting and following a glucose load, are related to bone density. We have shown this in normal postmenopausal women and found the effect to be partly independent from that of adiposity [65]. Similar effects have been demonstrated in both men and women in the Rotterdam study, where again they were partly independent of the effects of BMI [66]. The San Antonio Heart Study has produced comparable findings in women [67]. Abrahamsen [68] studied these relationships in men, and found that insulin sensitivity (measured from an intravenous glucose tolerance test) was inversely related to bone density independent of weight and fat mass. In addition, they found that the dependence of bone density on fat mass was lost when insulin sensitivity was entered into the multiple regression analysis, suggesting that this relationship was mediated through insulin sensitivity. Recently, the Tromso study has found that non-vertebral fracture risk progressively declines with increasing insulin resistance (inferred from the severity of the metabolic syndrome), being reduced by 50% in those with greatest insulin resistance [69].

One way to assess the effect of these hormones on bone is to reduce their levels through the use of an insulin sensitizing agent. We have recently done this using rosiglitazone, in a 3-month randomized controlled trial in normal postmenopausal women [70]. This study demonstrated reductions in bone formation markers within weeks of initiating treatment, and reductions in hip BMD of 2% at 3 months, suggesting that increased insulin sensitivity in humans has profound effects on bone within a short period. Glitazones are also likely to influence bone via their stimulation of peroxisome-proliferator-activated receptor-gamma, which directs precursor cells into the adipocyte lineage in preference to becoming osteoblasts [71]. This is discussed in more detail by Kawai and by Schwartz, elsewhere in this volume. Some fatty acids are converted into agonists for this receptor, indicating another potential route for adipose tissue to impact on bone [72].

Interactions of Hormones and Fat Mass in Humans

In view of the multiple relationships demonstrated between bone density on the one hand, and soft tissue mass and fat-related hormones on the other, it is of interest to assess as many of these factors as possible within the same subjects. The data previously published by Richards et al. [59] provides an opportunity to do this. These data were collected from a population-based cohort of non-diabetic women. The published analyses were directed at determining whether serum adiponectin was related to bone density, and reached the conclusion that an inverse relationship existed. The purpose of the present secondary analysis is to identify the principal correlates of bone density and bone turnover from the variables in this dataset. These include: the standard demographic descriptors, body composition, and serum levels of leptin, adiponectin, and insulin. Because sex hormones have profound effects on bone and are likely to interact with some of the other ‘independent’ variables, the analyses have been carried out separately in pre- and post-menopausal women, and subjects taking hormone replacement have been excluded.

Premenopausal Women

Four hundred and fifty three premenopausal were available for this analysis. They were aged 18–52 (mean 34) years with a mean BMI of 24.0 (SD 4.4). The simple correlations with variables related to soft tissue mass are shown in Table 1. Bone density was consistently, positively related to fat mass, lean mass, and serum leptin, and inversely related to serum adiponectin. The correlations with insulin tended to be positive but were only significant at the hip. Fractional soft tissue composition is represented by the ratio of fat mass to body weight (%fat), and tends to be positively related to bone densities. This indicates that there is a genuine link between fat mass and bone density, not merely a spurious correlation arising out of the dependence of fat mass on body size. Osteocalcin, a marker of osteoblast activity, was inversely related to fat mass and leptin. Deoxypyridinoline (DPD), a marker of bone resorption, showed a weak inverse relationship with serum adiponectin only. Fat mass was most closely related to leptin, and much more weakly to adiponectin and insulin.
Table 1

Correlations in premenopausal women (n = 453)

 

Arm

Spine

Total hip

Femoral neck

Osteocalcin

DPD

Fat mass

Fat mass

0.17

0.32

0.27

0.22

−0.22

0.03

 

% Fat

0.07

0.17

0.07

0.03

−0.24

0.03

 

Lean mass

0.24

0.42

0.42

0.47

0.03

0.03

0.50

Adiponectin

0.07

−0.10

−0.12

−0.12

0.02

−0.11

−0.27

Leptin

0.13

0.20

0.17

0.16

−0.24

0.03

0.76

Insulin

0.05

0.08

0.18

0.21

0.07

0.07

0.22

Data are Pearson correlation coefficients. Bolded figures are significant, P < 0.05

Stepwise multiple regression analysis was then undertaken to determine which variables were the principal determinants of bone density. The results of this analysis are shown in Table 2, and demonstrate marked variability from one skeletal site to another. Lean mass is consistently positively related to bone density, but adiponectin, insulin and fat mass variably appear. When the R2 procedure of SAS is used to generate a number of different regression equations to explain each dependent variable, it becomes apparent that fat mass, leptin, adiponectin, and insulin are partially interchangeable in these equations, in that the inclusion of one leads to the exclusion of the others. One of the reasons for the consistent presence of lean mass in all of these equations is that there are no other independent variables in this database which are highly correlated with it.
Table 2

Determinants of bone density and turnover by stepwise multiple regression analysis in premenopausal women

 

Independent variables

R2

Arm

Lean (0.07); age (0.06)

0.13

Spine

Lean (0.21); fat (0.03)

0.23

Total hip

Lean (0.26); height (0.03); age (0.01); insulin (0.01)

0.31

Femoral neck

Lean (0.24); age (0.04); adiponectin (0.02); insulin (0.02)

0.31

Osteocalcin

Leptin (0.07); height (0.02); age (0.02)

0.11

DPD

Age (0.02)

0.02

Partial R2 values for each variable are shown in parentheses. Parameter estimates are positive except where indicted by a superscript (). fat = total body fat mass, lean = total body lean mass

When a similar procedure was undertaken for the bone turnover markers, the inverse relationship between leptin and serum osteocalcin was again found. No biochemical or body composition values were related to DPD, but both turnover markers were inversely related to age, probably reflecting the decline in turnover markers that occurs during the third decade of life as the effects of puberty on bone turnover wane.

Postmenopausal Women

There were 215 postmenopausal women in the cohort who had adequate data for these analyses. They were aged 45–79 (mean 61) years with a mean BMI of 26.2 (SD 4.9). The simple correlations with variables related to soft tissue mass are shown in Table 3 and are broadly similar to those in the premenopausal women. Fat mass and lean mass are both positively related to bone densities, as is leptin. There is an inverse relationship between adiponectin and bone density, which is substantially stronger in postmenopausal women than in premenopausal women. As in premenopausal women, osteocalcin is inversely related to fat mass, and DPD is related only to adiponectin—positively in the case of postmenopausal women, the opposite of what was found in premenopausal women. Again, leptin is much more closely related to fat mass than is either of the other fat-related hormones.
Table 3

Correlations in postmenopausal women (n = 215)

 

Arm

Spine

Total hip

Femoral neck

Osteocalcin

DPD

Fat mass

Fat mass

0.23

0.27

0.35

0.29

−0.22

0.14

 

% Fat

0.07

0.11

0.17

0.13

−0.24

0.04

 

Lean mass

0.35

0.39

0.48

0.45

0.12

0.09

0.48

Adiponectin

−0.31

−0.29

−0.31

−0.30

0.31

0.21

0.13

Leptin

0.15

0.22

0.32

0.25

−0.23

0.06

0.73

Insulin

0.08

0.03

-0.02

0.01

0.07

0.01

0.04

Data are Pearson correlation coefficients. Bolded figures are significant, P < 0.05

The stepwise multiple regression analysis for postmenopausal women is shown in Table 4. As expected, age has a consistent negative effect on bone density in postmenopausal women, much more so than in premenopausal women. Lean mass is consistently positively related to bone density, though this relationship is substantially less in postmenopausal than in premenopausal women. Adiponectin is consistently negatively related to bone density, in marked contrast to the data shown in Table 2 from premenopausal women. In postmenopausal women, adiponectin appears to account for virtually all of the “fat-related” variance in bone density, such that fat mass itself is not significant in any of these analyses. In contrast, the bone turnover indices (osteocalcin and DPD) are related to both adiponectin and fat mass, though lean mass (as in the premenopausal women) has no influence on either.
Table 4

Determinants of bone density and turnover by stepwise multiple regression analysis in postmenopausal women

 

Independent variables

R2

Arm

Adiponectin (0.12); lean (0.09); age (0.07)

0.29

Spine

Lean (0.13); adiponectin (0.11)

0.23

Total hip

Lean (0.14); adiponectin (0.10); age (0.05); leptin (0.03)

0.31

Femoral neck

Lean (0.12); adiponectin (0.09); age (0.05);

0.26

Osteocalcin

Adiponectin (0.11); fat (0.04)

0.15

DPD

Adiponectin (0.08); fat (0.07); age (0.04)

0.20

Partial R2 values for each variable are shown in parentheses. Parameter estimates are positive except where indicted by a superscript (). fat = total body fat mass, lean = total body lean mass

Discussion

As noted in chapter 1 of this volume, there are major pitfalls in inferring cause and effect from cross-sectional data. With this caveat in mind, some general inferences can be made. Bone density is primarily dependent on indices of fat and lean mass, as has been noted in many previous studies. This dataset has only one measure of lean mass, but several measures of fat mass, in that adiponectin, leptin, and insulin are related to adiposity to varying degrees. In the multiple regression analyses in premenopausal women, one of these fat-related parameters is found in most of the regression equations, consistent with the general principle of both fat and lean masses being important. The same is true in postmenopausal women, but adiponectin appears to be a much more consistent predictor of bone density in this group, reflected in the higher partial R2 values in the multiple regression analysis. The reason for this is not immediately apparent, since the means and ranges of values are comparable in pre- and post-menopausal women in this study. However, the absence of the powerful effects of estrogen on bone density in postmenopausal women may allow the influence of adiponectin to be more apparent. Alternatively, adiponectin may simply be a surrogate for fat, and the greater influence of adiponectin, a reflection of the greater influence of fat mass in postmenopausal women. This, in turn, could be related to the fact that adipose tissue produces a greater proportion of the circulating estrogen in postmenopausal women than in premenopausal women.

In contrast to lean mass, fat mass, and its correlates are related to osteocalcin in both groups of women, and to DPD in postmenopausal women. This suggests that fat mass is related to bone metabolism whereas lean mass is not. Thus, the two soft tissue compartments could interact with the skeleton through quite different mechanisms. The common origins of osteoblasts, myocytes, and other connective tissue cells, together with their response to similar growth regulatory signals and to loading, probably underlie the bone-lean correlation, which is strongest in children and young adults. However, fat tissue regulates bone turnover in both premenopausal and postmenopausal life, so its influence on bone mass is likely to increase over time, particularly in the absence of other major players such as estrogen. This would account for the frequent observation that fat mass is a more significant correlate of bone density later in life, particularly in women.

The present data suggest that adiponectin is the principal factor through which fat mass influences bone density in postmenopausal women, since its correlation with bone is greater than that of other fat-related indices (including fat mass). It is not simply a reflection of total fat mass, since leptin is more closely related to this, yet has only a small effect on bone density. Adiponectin may be a better reflector of metabolically active adipocytes than the other adipokines or than is fat mass itself. As noted above, there is evidence for many adipokines influencing bone metabolism and serum adiponectin may simply be a convenient way of quantifying their integrated effects. Much more sophisticated experimental designs will be necessary to tease apart the interactions of the multiple, metabolically-active factors produced in adipocytes.

Copyright information

© Humana Press Inc. 2009