| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Section on Comparative Medicine, Department of Pathology, Wake Forest University School of Medicine (T.C.R., M.J.J., M.S.A.), Winston-Salem, North Carolina 27103-1040; and Pathology Associates International (M.J.J.), Advance, North Carolina 27006
Address all correspondence and requests for reprints to: Thomas C. Register, Ph.D., Comparative Medicine Clinical Research Center, Wake Forest University School of Medicine, Medical Center Boulevard, Winston-Salem, North Carolina 27157-1040. E-mail: register{at}wfubmc.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
There has been considerable interest in the use of phytoestrogens as substitutes for traditional estrogen replacement therapies in menopausal women (7, 8, 9, 10). Phytoestrogens are plant products with estrogenic activity, and some evidence suggests these compounds may provide tissue-selective effects. Phytoestrogens come in several forms, a major subclass being the isoflavones. Isoflavones, in particular daidzein and genistein, are highly enriched in soy compared with other food sources. Cross-cultural population studies have provided circumstantial evidence for a relationship between the consumption of high levels of soy isoflavones and postmenopausal skeletal health. For example, hip fracture rates are 50–60% lower in Asian than in Western populations, and this has been suggested as evidence for a beneficial effect of high consumption of soy on the skeleton (11, 12, 13). However, a number of other factors are likely to play a role in this reduced propensity for hip fracture. Asian women tend to be more active and perhaps more physically fit, to have better balance, and to have a lower incidence of falls (14, 15). Genetic factors, such as skeletal shape and geometry, are also likely to play a role, as Asian women differ from Western women due to a shorter hip axis and femoral neck length (16, 17). Supporting the idea that fitness and/or skeletal geometric factors play a role in the protection of Asian women from hip fracture is the fact that spine fracture rates (18) as well as bone density have been shown to be comparable between Asian and Western populations (19, 20). Nevertheless, associations between measures of isoflavone intake and bone density have been made in several observational epidemiological studies in postmenopausal Asian women (21, 22, 23) as recently reviewed by Anthony et al. (24), although urinary excretion of isoflavones and bone density of the radius were not associated with one another in a study of postmenopausal Dutch women (25). Only a handful of randomized, prospective, double-blind, placebo-controlled clinical trials on the effects of soy phytoestrogens on bone have been performed (26, 27, 28, 29, 30, 31), and the results from these trials have been equivocal. Trials to date have been of limited duration (all <1 yr), with a relatively small number of subjects. The difficulties of performing long-term randomized trials of dietary treatments in humans can be circumvented through the use of animal models. The present study was carried out to determine the efficacy of dietary soy phytoestrogens relative to mammalian estrogens (conjugated equine estrogens) in preventing menopause-related osteopenia and bone loss in a long-term experiment (3 yr) in nonhuman primates.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Skeletally mature female cynomolgus macaques (Macaca fascicularis, n = 181) were obtained through our collaborative association with Institut Pertainian Bogor (Bogor, Indonesia) and were studied extensively in a premenopausal phase (32, 33). During the premenopausal phase, animals received placebo or an oral contraceptive (Triphasil, Wyeth Ayerst Laboratories, Inc., Philadelphia, PA) and were maintained on diets containing 17% of calories from protein, 45% of calories from fat, 38% of calories from carbohydrate, and 0.28 mg cholesterol/Cal. At the end of the premenopausal period, the animals underwent bilateral ovariectomies to make them surgically menopausal. All procedures were conducted in compliance with state and federal laws, standards of the U.S. DHHS, and regulations and guidelines established by the Wake Forest University animal care and use committee.
Study design
Specific details of the overall design of this study as well as outcomes with respect to plasma lipoproteins and coronary artery atherosclerosis have been previously described (34). Briefly, the postmenopausal phase of the study was designed as a three-group, parallel-arm design in which animals were randomized to treatment groups based upon social groupings and oral contraceptive exposures during the premenopausal phase of the study. Animals were ovariectomized and fed three different soy protein-based diets: 1) an alcohol-washed, soy phytoestrogen-depleted, soy protein-containing diet (control; n = 57), 2) a soy phytoestrogen-rich, soy protein-containing diet (SPE; n = 59), and 3) the phytoestrogen-depleted control diet along with estrogen replacement therapy in the form of conjugated equine estrogens (CEE; n = 64; Premarin, Wyeth-Ayerst Laboratories, Inc.).
Diets
The compositions of the diets have been previously reported (34) and are comparable to a typical North American diet containing relatively high levels of fat and relatively low levels of calcium. Doses of hormones and isoflavones were administered on a caloric basis, with the assumption that the average postmenopausal woman consumes about 1800 Cal/d. The diets were formulated to be isocaloric for the macronutrients [protein (19% of calories), carbohydrate (37%), and fat (44%)] and comparable for cholesterol (0.28 mg/Cal), calcium (830 mg/1800 Cal), and phosphorus (820 mg/1800 Cal). The calcium content of the diet, while being at the higher end of the typical range of 400–800 mg calcium/d calcium consumed by an average adult postmenopausal woman, is below the National Academy of Sciences recommended doses of 1000 mg/d for women aged 19–50 yr and 1200 mg/d for women more than 50 yr of age, and well below the dose of 1500 mg/d recommended for women not taking estrogen and for adults over 65 yr by the NIH Consensus Conference and the National Osteoporosis Foundation (35). The monkeys were fed about 120 Cal/kg body weight/d in two daily feedings (one third in the morning, two thirds in the afternoon). All dietary protein was derived from soy protein isolates (SUPRO 670) provided by Protein Technologies International (St. Louis, MO). The control and the CEE groups were fed soy protein isolate that had been extracted by an aqueous-alcohol wash to deplete the SPEs (SUPRO 670-IF), and which contained, on the average, 0.04 mg genistein, 0.01 mg daidzein, and 0.01 mg glycitein/g isolate (expressed in aglycone units). The complete, unextracted soy protein isolate (SUPRO 670-HG) contained, on the average, 1.105 mg genistein, 0.365 mg daidzein, and 0.08 mg glycitein/g soy protein isolate (expressed in aglycone units). The SPE diet provided 35–40 mg isoflavones/monkey/d, comparable to a dose of 129 mg/d for women. The CEE (Premarin, Wyeth-Ayerst) were provided to mimic a woman’s oral dose of 0.625 mg/1800 cal/d.
Densitometry procedures
Lumbar spine (L2–L4) bone mineral content (BMC) and density (BMD), and whole BMC (BMCw) were measured by dual energy x-ray absorptiometry (DEXA) using a Norland XR26 instrument (Norland, Ft. Atkinson, WI). Precision and accuracy for this instrument with nonhuman primates have been previously reported (36). Baseline scans were taken 209 d before ovariectomy, then again at 11, 23, and 35 months after ovariectomy. Five animals received amputations, and their BMCw were not included for analyses.
Serum biomarkers of bone metabolism
Serum levels of total alkaline phosphatase (units per liter), acid phosphatase (ACP; units per liter), tartrate-resistant ACP (TRAP; units per liter), and calcium (Ca2+; milligrams per deciliter) were measured as previously described (33) at baseline and 3, 11, 18, 23, and 30 months. Intra- and interassay coefficients of variation were less than 8% for serum chemistries. Variations observed in ACP and calcium data may have resulted from differences in animal handling and serum collection procedures at DEXA sampling time points vs. the intermediate time points, which were collected at separate times under different conditions. This intertime point variation is probably not due to seasonal effects, as the animals were randomly staggered into the experiment and through time points over an 11- to 12-month period. Serum osteocalcin and C-terminal cross-link of type I collagen (Ctx) for baseline, 3-month, and 35-month samples were determined en bloc at the end of the study. Serum osteocalcin was determined using a double antibody RIA kit (DSL6900, Diagnostic Systems Laboratories, Inc., Webster, TX) specific for detection of the 1–43 amino acid fragment of osteocalcin or intact (1–49) osteocalcin. Serum Ctx was determined using a commercially available ELISA (Serum Crosslaps, Nordic Bioscience Diagnostics, Inc., Atlanta, GA) specific for the amino acid sequence EKAHD-ß-GGR derived from the C-terminal telopeptide region of type I collagen. Intraassay coefficients of variation for serum osteocalcin and Ctx were less than 7%.
Statistical methods
Statistical analyses were performed using BMDP statistical software (version 7.0, BMDP, Los Angeles, CA) and Statistica (Statsoft, Tulsa, OK). All variables were evaluated for their distribution and equality of variances between groups, and log transformations were performed for variables not meeting the tests of equal variances. One-way ANOVA and analysis of covariance (ANCOVA) were used to test for differences among groups. For longitudinal DEXA and anthropometric measures, one-way repeated measures ANOVA was carried out using preovariectomy data as a covariate. Prior oral contraceptive treatment status (a randomization factor) was not a significant independent predictor of outcome, and it was not used in analysis. Serum chemistries were analyzed by ANOVA at each time point without covarying for baseline values due to incomplete baseline data. Results are presented as the mean and SE of original data for normally distributed variables or were retransformed into original units for transformed variables as appropriate, and t tests were used for post hoc between-group comparisons if the ANOVA or ANCOVA P value was significant. A level of 0.05 or less was used to determine statistical significance. Correlations between variables are represented by Pearson product-moment coefficients calculated by STATISTICA and BMDP.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Body weight increased significantly in all groups through time, such that by 35 months gains of approximately 13–16% were observed (Table 1A
). No significant effect of treatment on body weight was observed (P = 0.58; Table 1B).
|
|
Significant effects of treatment (P = 0.002) as well as time (P = 0.0001) were observed on BMCw, but there was no time x treatment interaction (Table 1B
). Control and SPE groups responded similarly to ovariectomy and were not different from one another (P = 0.44), whereas both had significantly lower BMCw relative to the CEE group (both P = 0.0005). Significant losses of whole body BMC (
4–5%) were observed in control and SPE groups during the first year after ovariectomy (Table 1A). In contrast, CEE-treated animals gained BMCw (3%) over the first year. All groups gained BMCw in parallel by approximately 9% over the final 2 yr of the study. After 35 months the CEE-treated group had approximately 5–7% greater BMCw than either the control or SPE group.
Lumbar spinal BMC and BMD
The effects of treatment on lumbar spinal BMCs and BMDs, presented in Table 1, A and B, and Fig. 1, were similar to those observed for the whole body. Significant effects of treatment (P = 0.0001) as well as time (P = 0.0001) were observed, with no significant interactive effect (P > 0.4; Table 1B). After 11 months, the control and SPE groups lost significant amounts of BMCs and BMDs (8% relative to preovariectomy levels) compared with the CEE group, which had essentially maintained or slightly increased BMCs and BMDs (Fig. 1). The response of BMDs to CEE treatment was significantly different from that of the control (P = 0.0001) or SPE (P = 0.0001) groups, which were not different from each other (P = 0.996). Outcomes for BMCs were similar to those of BMDs (Table 1, A and B, and Fig. 1), and no effects of treatment or time were observed on spinal areal measures (all P > 0.4; data not shown).
|
The results obtained from determinations of serum markers are shown in Table 2 and Fig. 2. Serum levels of osteocalcin, Ctx, total alkaline phosphatase (units per liter), ACP (units per liter), and TRAP (units per liter) were significantly lower in the CEE group compared with the SPE and control groups (all P < 0.005, with the exception of P < 0.05 for TRAP at 30 months, indicating a suppression of ovariectomy-induced bone turnover by CEE; Table 2). Modest, transient inhibitory effects of SPE were observed at 3 months after ovariectomy, when serum total alkaline phosphatase, calcium, and Ctx were reduced in the SPE group relative to the controls (Table 2).
|
|
Correlations between serum markers and DEXA end points are summarized in Table 3. There was a consistent, significant inverse relationship between the percent change in BMDs over the 35 months of the study and alkaline phosphatase, osteocalcin, Ctx, and TRAP regardless of time point of collection. A similar, less consistent, relationship with percent change in BMDs was observed for serum calcium and ACP.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Although some of the human trials suggest a modest benefit of phytoestrogens, few direct comparisons of the effects of phytoestrogens to mammalian estrogens on bone have been carried out in women. In the Mefis trial (31), a soy-enriched diet, self-administered from a provided menu and verified by urinary daidzein measurements, was less effective than hormone replacement therapy (HRT) in reducing bone turnover, although neither the diet group nor the HRT group lost bone over the short 6-month course of the study, whereas the placebo group did. In the single human trial examining the effects of a purified isoflavone, Morabito et al. (38) reported that oral genistein (54 mg/d) paradoxically increased serum bone formation markers (osteocalcin and bone-specific alkaline phosphatase) and decreased urinary free collagen cross-links pyridinoline and deoxypyridinoline, while increasing bone mass in osteopenic postmenopausal women over the course of 1 yr. The effects of genistein on bone mass were similar to those observed with HRT using estradiol plus norethindrone acetate, although the HRT group experienced the classic coupled reductions in the resorption and formation biomarkers routinely observed with estrogen therapies. The genistein dose of 54 mg/d was lower than that in the present study where the monkeys received about 91 mg genistein as aglycone/d woman’s equivalent dose, although the genistein in soy protein was primarily in glycoside form, whereas the human trial was presumably performed with genistein in the aglycone form. Verification of these effects of genistein in a larger, longer trial are warranted.
It is important to note that the current study was designed to determine the effects of phytoestrogens per se, rather than the protein component of soy, on multiple end points of importance to postmenopausal women. Protein obtained from animal sources has a higher content of sulfur-containing amino acids (cysteine and methionine) than protein from vegetable sources, resulting in increased acidification of the urine and greater calcium excretion. Outcomes from studies in which control groups receive animal protein may well be influenced by this and other characteristics. Although the alcohol extracted soy protein used in the control diet in the present study was different from the SPE diet in other respects besides the phytoestrogen content (i.e. reductions in saponins and other alcohol-soluble components), these differences did not seem to influence bone mass. The modest increases in BMC and BMD that follow the marked initial bone loss in the control and SPE groups could be due to a number of factors. The nutrition of the animals was excellent, and calcium content was enriched relative to that of most postmenopausal women, as noted in Materials and Methods. In addition, the animals’ body weight increased throughout the study, and body weight as well as changes in body weight have a positive relationship with bone mass.
Additional results from this study demonstrated that CEE and, to a lesser extent, SPE inhibited coronary and carotid artery atherosclerosis relative to controls (34), whereas other data demonstrated minimal estrogenic activity of SPE on uterine or breast cell proliferation (39, 40). The isoflavone effects on bone biomarkers, albeit small, suggest that these compounds may influence bone metabolism to a limited extent in vivo in a primate model. Genistein and daidzein have recently received much attention because they have shown diverse biological activities, including the ability to bind to ER
and ERß and induce ERE-mediated transcription. Although ER has approximately twice the affinity toward 17ß-estradiol as ERß, ERß has up to 3- to 5-fold higher affinity than ER for some phytoestrogens (41). Thus, tissue-specific expression of ER and ERß combined with different affinities and actions of ligands suggest that differential targeting of the receptors could lead to improved therapies for postmenopausal women. The modest effects of SPE on serum biomarkers in the present study could be mediated by either ER or ERß, which are both expressed in monkey bone at the mRNA level (data not shown). It is conceivable that there are ethnic and species differences in ERs in bone (and other organs for that matter), and that certain populations of women might benefit from isoflavones more than others. Adaptations and genetic selection resulting from generations of people consuming high levels of phytoestrogens are perhaps even likely in some populations (i.e. Asian), but are as yet unproven. Definitive quantitative studies of ER and ERß protein levels in bone in monkey or any other species have not been reported. Nevertheless, phytoestrogen action is not restricted to ERß in vivo, and several studies in transgenic mouse models suggest that their mechanisms of action are sometimes dependent upon the presence of ER and not necessarily ERß (42, 43, 44, 45). In either case, the mild transient SPE effects on bone turnover were insufficient to be translated into protection of the skeleton from ovariectomy-induced bone loss, particularly compared with the effects of CEE. The mechanisms underlying the transient nature of the phytoestrogen effect on bone biomarkers are not known, but could relate to adaptations of the animal that lead to alterations in the metabolism of the isoflavone compounds in the gut or even after absorption. Alternatively, the sensitivity of the skeleton to phytoestrogen effects may be greatest soon after ovariectomy and may diminish with time, or the kinetics of the increase in bone turnover may be slowed by SPE relative to that in the control group, an effect that is overcome with time. It also is conceivable that the disappearance of estradiol after ovariectomy may be slowed in the presence of SPE, although we have no evidence for such an effect. Ultimately, the transient nature of this effect implies that one should be wary of biomarker data alone when assessing treatment influences on the skeleton.
In conclusion, we have no evidence to support the idea that dietary isoflavones are as effective as CEE in preventing the ovariectomy-induced bone loss in monkeys. Nevertheless, although the consumption of SPE as components of soy showed little or no benefit to the skeleton in these animals, it also caused no apparent harmful effects to the skeleton or the reproductive system, while providing some benefit against atherosclerosis. These findings have important implications for postmenopausal women who are seeking alternatives to traditional pharmaceutical therapies to counteract the adverse effects of estrogen loss. Many supplements containing isoflavones/phytoestrogens purport to have beneficial effects on the skeleton, some citing evidence from studies of the compound ipriflavone, a synthetic isoflavone. In a large, multicenter trial in Europe, ipriflavone treatment did not prevent bone loss in postmenopausal women, although it did cause subclinical lymphocytopenia in about 13% of subjects (46). Dietary supplement use has grown dramatically in recent years, and these products are under less investigational scrutiny by federal agencies due to several factors. First, the Federal Food, Drug, and Cosmetic Act does not provide a statutory definition of functional foods, thus the FDA has no authority to establish a formal regulatory category for such foods. In addition, there has not been a requirement for exact compositional data of food supplements. Thus, many questions remain about the usefulness of many food supplements for health claims as well as their safety. In particular, little is known of the effects of isoflavone extracts in humans, and these supplements should be taken with caution until more complete evidence of their safety and efficacy is obtained.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Abbreviations: ACP, Acid phosphatase; ANCOVA, analysis of covariance; BMC, bone mineral content; BMCw, whole body BMC; BMD, bone mineral density; CEE, conjugated equine estrogens; Ctx, C-terminal cross-link of type I collagen; DEXA, dual energy x-ray absorptiometry; ER, estrogen receptor; HRT, hormone replacement therapy; SPE, soy phytoestrogens; TRAP, tartrate-resistant ACP.
Received March 17, 2003.
Accepted June 10, 2003.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and ß. Endocrinology 138:863–70 KO, mice. Nutr Cancer 39:226–232[CrossRef][Medline]
. Exp Biol Med 226:301–306 expression and multioocyte follicles in the maturing mouse ovary: evidence for ERß-mediated and nonestrogenic actions. Biol Reprod 67:1285–1296. Arterioscler Thromb Vasc Biol 22:1859–1864This article has been cited by other articles:
 |
|
 |
|
  C. J Lees, J. R Kaplan, H. Chen, C. P Jerome, T. C Register, and A. A Franke Bone mass and soy isoflavones in socially housed, premenopausal macaques Am. J. Clinical Nutrition, July 1, 2007; 86(1): 245 - 250. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. M. Sacks, A. Lichtenstein, L. Van Horn, W. Harris, P. Kris-Etherton, M. Winston, and for the American Heart Association Nutrition Commi Soy Protein, Isoflavones, and Cardiovascular Health: An American Heart Association Science Advisory for Professionals From the Nutrition Committee Circulation, February 21, 2006; 113(7): 1034 - 1044. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. M. Weaver and J. M. K. Cheong Soy Isoflavones and Bone Health: The Relationship Is Still Unclear J. Nutr., May 1, 2005; 135(5): 1243 - 1247. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. C. Register, J. A. Cann, J. R. Kaplan, J. K. Williams, M. R. Adams, T. M. Morgan, M. S. Anthony, R. M. Blair, J. D. Wagner, and T. B. Clarkson Effects of Soy Isoflavones and Conjugated Equine Estrogens on Inflammatory Markers in Atherosclerotic, Ovariectomized Monkeys J. Clin. Endocrinol. Metab., March 1, 2005; 90(3): 1734 - 1740. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. E. Wood, T. C. Register, M. S. Anthony, N. D. Kock, and J. M. Cline Breast and Uterine Effects of Soy Isoflavones and Conjugated Equine Estrogens in Postmenopausal Female Monkeys J. Clin. Endocrinol. Metab., July 1, 2004; 89(7): 3462 - 3468. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. E. Wood, J. M. Cline, M. S. Anthony, T. C. Register, and J. R. Kaplan Adrenocortical Effects of Oral Estrogens and Soy Isoflavones in Female Monkeys J. Clin. Endocrinol. Metab., May 1, 2004; 89(5): 2319 - 2325. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
