| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Division of Bone Diseases, World Health Organization Collaborating Center for Osteoporosis and Bone Diseases, Department of Internal Medicine, University Hospital, 1211 Geneva 14, Switzerland
Address all correspondence and requests for reprints to: René Rizzoli, M.D., Division of Bone Diseases, Department of Internal Medicine, University Hospital, 1211 Geneva 14, Switzerland. E-mail: rizzoli{at}cmu.unige.ch
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Peak bone mineral mass is considered to be mostly genetically determined (see Ref. 5 for review), and familial resemblance for areal (a) or volumetric (v) bone mineral density (BMD) and bone size can actually be detected from childhood (6). Allelic polymorphisms in the 3'-end region of the vitamin D receptor (VDR) gene were the first candidates to be investigated in relation to BMD (7). Whereas VDR-3' alleles have been associated with significant BMD differences in prepubertal girls (8, 9), there is no consistent association in adult women (10, 11). This apparent discrepancy might partly be explained by complex interactions with environmental factors, such as dietary calcium intake, as well as age-related factors, possibly involving pubertal hormonal changes (8). More recently, a polymorphic sequence in the VDR gene translation start codon, which is susceptible to being associated with alterations in the structure of the VDR molecule (12), has also been related to aBMD in adult women by some (13, 14), but not other (15, 16), groups. Recent in vivo and in vitro results indicate that interactions between VDR gene 3' and 5' polymorphisms should be considered in the association with bone mineral mass (15, 17).
In the present study, we analyzed the relationship between VDR-3' (BsmI) and -5' (FokI) allelic polymorphisms, and bone mineral mass and bone size in a cohort of young healthy Caucasian men. Biochemical parameters of calcium-phosphate metabolism were measured cross-sectionally and then with a challenge of dietary modifications to assess physiological mechanisms potentially related to bone mineral mass differences among VDR gene polymorphisms.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
One hundred and four men, aged 20.7–38.6 yr, were recruited among students from the University of Geneva (Geneva, Switzerland). Exclusion factors were known acute or chronic disease or medications that could affect intestinal absorption, kidney function, or bone turnover. aBMD (grams per cm2), bone mineral content (BMC; grams), and bone area (BA; square centimeters) were measured by dual energy x-ray absorptiometry at the level of the lumbar spine (LS) and proximal femur (FN, neck; FT, trochanter) as well as for the whole body (WB), using a Hologic QDR-2000 instrument (Hologic, Inc., Waltham, MA).
Genotyping
Five-milliliter saliva samples were collected to perform genetic analysis. Saliva collection is particularly convenient in young individuals. VDR-3' (as determined by the sensitivity to the restriction enzyme BsmI) and 5'-start codon (FokI) polymorphisms were determined by PCR amplification and enzymatic restriction according to standard procedures, as previously described (15, 18). Unequivocal genotypes were obtained with BsmI in 104 subjects and with FokI in 96, which resulted in 95 cross-genotypes between BsmI and FokI polymorphic sites. Alleles are designated by the first letter of the restriction enzyme by which they are identified. Small letters or capitals are used according to the presence or, respectively, the absence of a restriction site. The absence of the FokI restriction site (F) indicates a VDR 3 amino acids shorter than the alternate allele (f). A BsmI-negative site (B) refers to the absence of several sequence differences in the 3'-untranslated region.
Dietary assessment and biochemical measurements
Seventy-two subjects completed a food frequency questionnaire administered by a trained dietitian to estimate dietary calcium, phosphate, and protein intakes. Fasting blood and urine samples as well as 24-h urine samples were subsequently collected while subjects were consuming their regular diets.
Dietary intervention
Twenty-five males matched for height and weight were chosen on the basis of alternate homozygous BsmI genotypes (15 bb, 10 BB) to participate in a prospective short term dietary calcium-phosphorus modification study. Dietary intervention was performed on an ambulatory basis and consisted, after a 2-day run-in period during which subjects were allowed to eat their regular diets, of a 5-day period of calcium and phosphorus restriction, achieved through dietary counseling and the administration of a magnesium- and aluminum-containing phosphorus binder (Alucol, Novartis). Then, subjects were reequilibrated for 2 days on their regular diet and subsequently advised to consume a diet rich in calcium- and phosphorus-containing products for 5 days while they also received phosphorus supplements (1000 mg elemental phosphorus/day as potassium-phosphorus syrup). Thus, dietary challenge concerned primarily phosphorus intake. Fasting blood and urine samples as well as 24-h urine samples were collected for biochemical analysis at baseline and on the last day of the restriction and supplementation periods, when a steady state was expected to be reached. Dietary calcium and phosphorus intakes were assessed during the whole study period by a self-administered daily food diary in which all foods were weighed. Dietary intervention and food diary were repeatedly controlled during the trial by a trained dietitian. Compliance with the assigned regimen was assessed by 24-h urinary excretion of calcium and inorganic phosphate (Pi), performed during each dietary intervention period (namely the run-in, restriction, equilibration, and supplementation periods). Accordingly, a reliable analysis could be performed in 24 of 25 subjects (14 bb, 10 BB).
Biochemical determinations
Plasma and urinary solutes were measured using standard methods. Vitamin D metabolites were determined with RIA or protein binding assays (INCSTAR Corp., Stillwater, MN). The assay for intact PTH (Immulite) was obtained from Diagnostic Products (Los Angeles, CA). Circulating insulin-like growth factor I and osteocalcin levels were measured by RIAs from Nichols Institute Diagnostics (San Juan Capistrano, CA) and from CIS-Bio International (Gif-sur-Yvette, France), respectively. For the former, separation from binding proteins was obtained by acid-ethanol and cryoprecipitation extraction. Deoxypyridinoline excretion was measured in the second morning sample by fluorescence emission after acid hydrolysis. The renal tubular reabsorption of calcium and Pi was calculated using published nomograms (20, 21). The assay for cAMP was obtained from Immunotech (Marseille, France).
Statistics
aBMD, BMC, and BA as well as biochemical parameters were compared among VDR genotypes by ANOVA, followed by Fisher’s protected least significant difference (PLSD) test when P < 0.05 (by ANOVA). Further comparisons were made by calculating standardized BMD (z-scores) as BMC adjusted for age, height, weight, and site-specific bone area (19). Interactions between VDR gene polymorphisms and diet with regard to biochemical values were tested by two-factor ANOVA for repeated measures and by covariance analysis (ANCOVA), as indicated.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The frequency of VDR-3' (BsmI) alleles in young healthy
Caucasian men (mean age ± SD, 24.3 ± 3.1 yr)
was previously reported in various Western populations and was in
Hardy-Weinberg equilibrium (Table 1
).
Clinical characteristics were similar among VDR-3' BsmI
genotypes. No significant association was observed between aBMD, BMC,
or BA in lumbar spine, proximal femur, or whole body and
BsmI genotypes. However, after BMC adjustment for age, body
size, and bone area (standardized BMD), z-score differences were
significant at the levels of the LS and FT [LS: -0.31 ± 0.21,
0.29 ± 0.17, and -0.10 ± 0.14 z-scores in BB,
Bb, and bb, respectively (mean ±
SEM; P = 0.03); FT: -0.31 ± 0.21,
0.27 ± 0.19, and -0.06 ± 0.11 z-scores in BB,
Bb, and bb, respectively (P =
0.05)]. A nonsignificant trend was found at the femoral neck level.
There was no relationship between aBMD and FokI genotypes at
any skeletal site.
|
Among the 95 subjects for whom genotypes could be determined for
both BsmI and FokI polymorphic sites, there was
no evidence of linkage disequilibrium between the two loci
(
2 = 4.5; P = 0.34). Interactions were
detected between VDR-3' (BsmI) and -5' (FokI)
polymorphisms in association with aBMD and BMC (P =
0.019 for LS aBMD, 0.031 for LS BMC, and 0.012 for WB BMC). Thus, among
subjects carrying one or more f (FokI) alleles
(designated f+), standardized BMD (z-scores) were
significantly lower in BB compared to those in Bb
and bb men (Fig. 1). In
contrast, in the absence of an f allele (designated
f-), BMD (z-scores) were as high or higher in BB
as in the other genotypes.
|
Seventy-two men (69% of the cohort) reported their spontaneous dietary intake and underwent a biochemical evaluation of calcium and bone metabolism. Noteworthy, their calcium, phosphorus, and protein intakes were high, on the average (calcium, 1456 ± 643 mg/day; phosphate, 1532 ± 562 mg/day; proteins, 96.9 ± 27.6 g/day; mean ± SD).
PTH and osteocalcin were both significantly higher among BB
and Bb compared to bb men, whereas their serum
calcium and calcitriol [1,25-dihydroxyvitamin D3;
1,25-(OH)2D3] levels did not differ (Table 2). Differences in PTH levels among
BsmI genotypes were significant in the presence of an
f allele [mean ± SEM, 3.63 ± 0.49,
3.60 ± 0.30, and 2.46 ± 0.27 pmol/L in (f+)
BB, Bb, and bb, respectively;
P = 0.025] but not in its absence [3.68 ± 0.75,
3.78 ± 0.60, and 3.31 ± 0.55 pmol/L in (f-) BB,
Bb, and bb, respectively; P =
NS].
|
). During the dietary modifications, PTH
remained significantly higher in BB compared to
bb subjects. This was associated with significantly
decreased tubular reabsorption of phosphate and plasma Pi levels
despite increased 1,25-(OH)2D3 levels on the
low calcium-phosphorus diet. In contrast, neither urinary cAMP nor
indexes of bone remodeling were altered during changes in the
calcium-phosphate diet in any genotypic subgroup.
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
We had previously reported similar aBMD differences among BsmI genotypes in prepubertal girls, whereas in adult women with the same genetic background these differences were not detectable (8). Similarly, others have found a significant association of VDR-3' (BsmI) alleles with volumetric density in prepubertal girls (9) and with aBMD in younger women (22), whereas many studies reported a lack of detectable BMD differences among BsmI genotypes in pre- and postmenopausal women (8, 11, 23, 24, 25). Accordingly, it has been proposed that endogenous and environmental factors could partly explain the apparent attenuation of aBMD differences among VDR-3' allelic polymorphisms occurring in females with age, among which pubertal estrogens and dietary calcium intake might play a role (8, 26). It is likely therefore that gender as well as age, social, and genetic background homogeneity of our cohort allowed small bone mineral mass differences to be detected among VDR-3' gene polymorphisms (27).
The present study further suggests that interactions between VDR-3' and -5' polymorphisms should be considered in relation to peak BMD. Indeed, BMD at each skeletal site separately as well as for the whole body differed by about 1 z-score between BB subjects also carrying at least one f (FokI) allele [designated BB(f+)] and other cross-genotypic groups. These observations are compatible with previous comparisons in healthy premenopausal women showing a lower BMD in BB (f+) than in BB (f-) subjects (15). VDR-start codon polymorphisms are translated into VDR molecules of variable lengths, potentially with different activities (12, 17). Functional differences in VDR-3' polymorphisms remain controversial (7, 17, 28). Expression of the VDR or the response to calcitriol was not different in cultured fibroblasts with either the BB or bb polymorphic variants (29). Taking into consideration such interaction in various polymorphisms may explain apparently discordant results in the present and the two previously published studies in adolescent and young adult males, which found either no association of BMD with VDR-3' (BsmI) alleles (30) or lower BMD in bb subjects (31). Our observations support this hypothesis (17). This underlines the need for identifying multiple single base polymorphisms within candidate genes for any disease to increase the likelihood of a genetic disequilibrium between these markers and the mutations ultimately responsible for the disorder (32).
Concerning the physiological mechanisms supporting a relationship between peak bone mineral mass and VDR allelic polymorphisms, little is known to date beyond the fact that dietary calcium handling and possibly the response to vitamin D may be involved (8, 18, 33, 34, 35, 36). Indeed, longitudinal evaluations of calcium, phosphate, and bone metabolism according to VDR genotypes, particularly in men, are still missing.
Our results indicate that PTH and osteocalcin levels were significantly associated with VDR-3' alleles. Moreover, FokI alleles influenced the association of VDR-3' genotypes with PTH in a similar manner as with BMD. Osteocalcin has previously been reported to be higher in BB compared to Bb or bb women (37), but these data were not later confirmed (24). PTH has been found to be significantly higher in BB postmenopausal Mexican-American women (38). In contrast, the prevalence of bb alleles is higher in patients with primary hyperparathyroidism (39), and parathyroid tumors of patients homozygous for the b allele had higher PTH messenger ribonucleic acid levels (40).
The consistency of higher PTH levels among BB subjects was further assessed by a short prospective dietary modification study performed in a subgroup of males (n = 25) from the present cohort. In these subjects, we found that PTH differences between alternate homozygotes BB and bb persisted throughout the 15-day period of the study under either a low or a high calcium and phosphorus diet. This was accompanied by decreases in renal tubular reabsorption of Pi and plasma Pi levels, which were more prominent relative to 1,25-(OH)2D3 levels, under dietary calcium and phosphorus restriction. These observations support the hypothesis that VDR gene polymorphisms may be markers of subtle differences in the molecular actions of the vitamin D-receptor complex, particularly at the level of the parathyroid gland and intestine (26, 34). Whether peak bone mineral mass differences among VDR gene polymorphisms might therefore be more pronounced in males whose dairy intakes are lower than those in our cohort remains to be investigated. Furthermore, the limited power of our study should be acknowledged. Indeed, only about two thirds of the enrolled volunteers agreed to blood and urine samplings.
In conclusion, osteoporosis is increasingly recognized as a health problem in elderly males (1, 2), but its genetic determinants are still poorly understood. Our results suggest subtle differences in BMD and bone mineral metabolism among VDR gene polymorphisms in young males together with a possible interaction between 3' and 5' polymorphisms as well as between these genetic markers and dietary factors.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Current address: Division of Bone and Mineral Metabolism, Beth
Israel Deaconess Medical Center, Harvard Institutes of Medicine,
Boston, Massachusetts 02446.
Received January 15, 1999.
Revised March 2, 1999.
Accepted March 11, 1999.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  L. Richert, T. Chevalley, D. Manen, J.-P. Bonjour, R. Rizzoli, and S. Ferrari Bone Mass in Prepubertal Boys Is Associated with a Gln223Arg Amino Acid Substitution in the Leptin Receptor J. Clin. Endocrinol. Metab., November 1, 2007; 92(11): 4380 - 4386. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. L. Ferrari, J.-P. Bonjour, and R. Rizzoli Fibroblast Growth Factor-23 Relationship to Dietary Phosphate and Renal Phosphate Handling in Healthy Young Men J. Clin. Endocrinol. Metab., March 1, 2005; 90(3): 1519 - 1524. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D-H Xiong, F-H Xu, P-Y Liu, H Shen, J-R Long, L Elze, R R Recker, and H-W Deng Vitamin D receptor gene polymorphisms are linked to and associated with adult height J. Med. Genet., March 1, 2005; 42(3): 228 - 234. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Yamada, F. Ando, N. Niino, and H. Shimokata Association of Polymorphisms of Interleukin-6, Osteocalcin, and Vitamin D Receptor Genes, Alone or in Combination, with Bone Mineral Density in Community-Dwelling Japanese Women and Men J. Clin. Endocrinol. Metab., July 1, 2003; 88(7): 3372 - 3378. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J R Ortlepp, R Hoffmann, F Ohme, J Lauscher, F Bleckmann, and P Hanrath The vitamin D receptor genotype predisposes to the development of calcific aortic valve stenosis Heart, June 1, 2001; 85(6): 635 - 638. [Abstract] [Full Text]
|
|
|
|
 |
|
 D A Nelson, P J Vande Vord, and P H Wooley Polymorphism in the vitamin D receptor gene and bone mass in African-American and white mothers and children: a preliminary report Ann Rheum Dis, August 1, 2000; 59(8): 626 - 630. [Abstract] [Full Text]
|
|
|
|
 |
|
 O. Tajima, N. Ashizawa, T. Ishii, H. Amagai, T. Mashimo, L. J. Liu, S. Saitoh, K. Tokuyama, and M. Suzuki Interaction of the effects between vitamin D receptor polymorphism and exercise training on bone metabolism J Appl Physiol, April 1, 2000; 88(4): 1271 - 1276. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Lorentzon, R. Lorentzon, and P. Nordström Vitamin D Receptor Gene Polymorphism Is Associated with Birth Height, Growth to Adolescence, and Adult Stature in Healthy Caucasian Men: A Cross-Sectional and Longitudinal Study J. Clin. Endocrinol. Metab., April 1, 2000; 85(4): 1666 - 1671. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
