help button home button Endocrine Society JCEM
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Journal of Clinical Endocrinology & Metabolism, doi:10.1210/jc.2006-1134
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
92/4/1491    most recent
Author Manuscript (PDF)
Right arrow Submit a related Letter to the Editor
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Fang, Y.
Right arrow Articles by Uitterlinden, A. G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Fang, Y.
Right arrow Articles by Uitterlinden, A. G.
Right arrowPubmed/NCBI databases
*Gene*GEO Profiles
*HomoloGene*UniGene
Related Collections
Right arrow Calcium and Bone Metabolism
The Journal of Clinical Endocrinology & Metabolism Vol. 92, No. 4 1491-1501
Copyright © 2007 by The Endocrine Society

Vitamin D Receptor Gene Haplotype Is Associated with Body Height and Bone Size

Yue Fang, Joyce B. J. van Meurs, Fernando Rivadeneira, Natasja M. van Schoor, Johannes P. T. van Leeuwen, Paul Lips, Huibert A. P. Pols and André G. Uitterlinden

Departments of Internal Medicine (Y.F., J.B.J.v.M., F.R., J.P.T.v.L., H.A.P.P., A.G.U.), and Epidemiology and Biostatistics (F.R., H.A.P.P., A.G.U.), Erasmus Medical Center, 3015 GD Rotterdam, The Netherlands; Institute for Research in Extramural Medicine (N.M.v.S.), VU University Medical Center, 1081 BT Amsterdam, The Netherlands; and Department of Endocrinology (P.L.), VU University Medical Center, 1007 MB Amsterdam, The Netherlands

Address all correspondence and requests for reprints to: André G. Uitterlinden, Genetic Laboratory, Room Ee575, Department of Internal Medicine, Erasmus MC, P.O. Box 1738, 3000 DR Rotterdam, The Netherlands. E-mail: a.g.uitterlinden{at}erasmusmc.nl.


   Abstract
Top
 Abstract
Introduction
 Subjects and Methods
 Results
 Discussion
 References
 
Context: Adult stature is a complex genetic trait. The vitamin D endocrine system has pleiotropic effects on several physiological processes, especially on skeletal metabolism. We recently identified promoter and 3'-untranslated region (UTR) haplotype alleles that influence vitamin D receptor (VDR) mRNA expression.

Objective: We studied whether VDR gene variants contribute to the genetic variation in height.

Design and Subjects: We studied VDR haplotype alleles and body height in two independent populations (n = 7187). In a meta-analysis (n = 14,157 from 27 studies and our current data), we evaluated the effect of the Bsm I polymorphism.

Results: Haplotypes of the linkage disequilibrium block 3 and block 5 were associated with body height differences with evidence for additive effects in the Rotterdam Study (P = 0.00002) and the Longitudinal Aging Study Amsterdam study (P = 0.001). Height differences between the extreme genotypes were 1.4 and 2.7 cm, respectively. The relationship was independent of age, gender, presence of vertebral fractures, and age-related height loss. In the Rotterdam population, we found the combined genotype to be associated with decreased vertebral area (P = 0.03) and femoral narrow neck width (P = 0.002). In the meta-analysis, subjects with the "BB" genotype were 0.6 cm (95% confidence interval, 0.2–1.1 cm) taller than those with the "bb" genotype (P = 0.006).

Conclusion: VDR gene variants are associated with differences in body height as evidenced by our study and by a meta-analysis. It remains for further studies to confirm whether the underlying mechanism of the association involves lower VDR expression in cells important for determining bone size.


   Introduction
Top
 Abstract
 Introduction
Subjects and Methods
 Results
 Discussion
 References
 
HUMAN ADULT STATURE is one of the most heritable human traits (heritability up to 90%), as demonstrated in twin studies (1, 2) and family studies (3, 4), although ethnicity also influences body height. On the other hand, environmental changes (e.g. improved nutrition) have progressively increased height (as a secular trend) during the 20th century (5). Adult stature has been reported to be inversely associated with a number of common complex genetic diseases such as cardiovascular diseases, diabetes, pulmonary diseases, cancer, and osteoporosis as well as disease-specific mortalities in different populations (6, 7, 8, 9, 10, 11, 12). Therefore, stature appears to be an interesting model of a complex trait, because the understanding of the underlying genetic causes of height determination may provide insight into mechanisms of other related diseases.

The vitamin D endocrine system has been shown to have pleiotropic effects on a number of endocrine pathways, e.g. related to immune modulation, regulation of cell proliferation and differentiation, and on skeletal metabolism (13). It has been suggested that susceptibility to cardiovascular heart disease, cancer, and diabetes is affected by dietary vitamin D intake and status of exposure to sunlight (14). Long-term vitamin D deficiency results in rickets in children and osteomalacia in adults. The vitamin D receptor (VDR) gene is a central regulator in this endocrine system and therefore, it is an interesting candidate gene for genetic studies of stature. Mutations in the VDR gene such as in the DNA binding domain (15), the ligand binding domain (16), or splice sites (17) cause hereditary vitamin D-resistant rickets (HVDRR). HVDRR is commonly associated with clinical manifestations of impaired longitudinal growth resulting in short stature and/or regular bone deformities. Adult height was found to be correlated with severeness in 13 patients with HVDRR (18), whereas growth retardation nearly invariably accompanies vitamin D-resistant rickets (19). Also, VDR-null mice (with a deletion of the DNA binding domain) show growth retardation in addition to rickets (20). The phenotype of rickets or osteomalacia in both human and animal models shows the same skeletal abnormalities associated with defective mineralization in the growing skeleton, small body size and weaker bones. Recent genome-wide linkage analyses (21, 22) showed evidence of linkage between stature and a region on chromosome 12 (12p11.2-q14), where the VDR gene is located. Therefore, sequence variation of the VDR gene might influence body height differences in the normal population. Although several association studies have investigated the relationship between VDR polymorphisms and body height in the past, conflicting results were found, possibly because of small sample sizes, variations in study design, heterogeneous populations, and because mostly anonymous VDR polymorphisms were used so far.

Recently, we and others have resolved the linkage disequilibrium (LD) structure of the VDR gene (23, 24). In functional studies, we have identified promoter and 3'-UTR haplotype alleles in blocks 2 and 5, respectively, that influence VDR expression (23). In a subsequent association study, we observed these risk alleles in block 2, block 3, and block 5 to be associated with increased risk to develop an osteoporotic fracture. However, no association of the VDR genotype and bone mineral density (BMD) was found. Hence, the potential mechanism of the association with fracture might be in part through an influence on bone size rather than BMD effects. The hypothesis of the current study was that the VDR fracture risk alleles might be associated with decreased bone strength in the elderly through lifelong sustained differences in bone size/geometry and subsequently with decreased skeletal frame size/body height. We investigated the relationship of haplotypes across the VDR gene with height as well as bone geometry data as derived from hip structure analysis parameters (25) in two independent cohort populations. Finally, in a meta-analysis of published data, we compared our results with those of previous studies on VDR Bsm I and/or Taq I polymorphisms, which can be used to identify the most common haplotypes in the 3' area (23, 24, 26).


   Subjects and Methods
Top
 Abstract
 Introduction
 Subjects and Methods
Results
 Discussion
 References
 
Subjects

The Rotterdam Study population. The Rotterdam Study is a single-center prospective population-based cohort study and includes 7983 individuals, with 3105 men (38.9%) and 4878 women (61.1%), designed to analyze determinants and prognoses of chronic and disabling diseases in the elderly (older than 55 yr) Caucasians (27). The baseline measurements were performed between 1990 and 1993. The latest follow-up period ended January 1, 2002, and the mean follow-up period was 7.4 ± 3.3 (SD) yr for clinical fracture. For the current study, 6580 DNA samples (82.4% of the whole cohort population) were available, and 6031 DNAs were successfully genotyped for all single nucleotide polymorphisms (SNPs) as described in our previous study (23). Most of the scarce genotyping failures were because of low-quality DNA of those samples. For the association analysis, we included 5931 subjects with baseline measurement data of BMD, 5933 subjects with clinical data of vertebral fracture, 3114 subjects with x-ray-confirmed data of vertebral fracture data, 5165 subjects with vertebral area data, and 4230 subjects with hip bone geometry data.

Longitudinal Aging Study Amsterdam (LASA) population. The LASA is an ongoing cohort study of older persons aged 55 to 85 yr. The sampling and data collection procedures have been described in detail elsewhere (28). In summary, a random sample stratified by age, sex, and expected 5-yr mortality rate was drawn from the population registers of 11 municipalities in three regions of The Netherlands. In total, 3107 persons were enrolled in the baseline examination in 1992 through 1993. For the current study, persons who participated in the medical interview and were born in or before 1930 (aged 65 yr and older as of January 1, 1996) were selected (n = 1509). Of these, 1352 blood samples were obtained, 956 DNAs were isolated, and 911 DNAs were successfully genotyped for all SNPs. The failure of genotyping in some samples was because of the low quality of DNA.

DNA isolation and genotyping

Genomic DNA was isolated from peripheral venous blood specimens according to standard protocols. For the association study, we genotyped seven haplotype tagging SNPs (htSNPs) as described in our recent haplotype analysis across the entire VDR gene (23) with TaqMan allelic discrimination assays. The genotype results were analyzed independently by two operators. Five percent random samples were independently repeated to confirm the genotyping results for all SNPs. The disagreement rate of the genotype results was 0.3 to 1.2% for five htSNPs, and the genotype results of other htSNPs were completely consistent.

LD haplotype analyses and cladogram construction

We currently combined the nested cladistic method, pairwise LD measurement and LD block definition, to construct a cladogram and haplotypes within the LD blocks 3 and 5 based on our previous study (23). For LD and haplotype analyses with the PHASE program (http://archimedes.well.ox.ac.uk/pise/PHASE.html), we included 19 SNPs (nine SNPs in block 3 and 10 SNPs in block 5). PHASE output was used to calculate the pairwise standardized disequilibrium coefficient (D') with the "haploxt" (http://archimedes.well.ox.ac.uk/pise/haploxt.html) to estimate the linkage magnitude between two SNPs. We represent haplotype diversity by means of a cladogram that was generated with a PHASE output linked to the "hapdist" program (http://archimedes.well.ox.ac.uk/pise/hapdist.html) and depicted the graphic overviews of the cladogram for each LD block with Cladogramer (32).

Clinical examination of association study

Height and weight were measured in a standing position wearing indoor clothing without shoes, and all height measurements were attained by a research assistant using a standard wall-mounted stadiometer. Height loss was calculated as the difference of height between the baseline and the ending of follow-up measurements. Body mass index (BMI) was calculated as weight (in kilograms) divided by height squared (in meters). BMD (in grams per square centimeter) was determined by dual-energy x-ray absorptiometry (Lunar DPX-L densitometer; Lunar Radiation Corp., Madison, WI) at the femoral neck and lumbar spine (vertebral L2–L4) as described before (33). Vertebral body area (in square centimeters) was measured over L2–L4 by posteroanterior scanning using the same dual-energy x-ray absorptiometry instrument. Hip structure analysis such as the section modulus (Z, an index of bending strength) and the buckling ratio (BR, in index of bone instability) were calculated as described in previous reports (25, 34). The narrow-neck width (in centimeters) of the hip is across the narrowest point of the femoral neck. Clinical vertebral fracture was diagnosed according to a previous description (35); x-ray-confirmed vertebral fractures at baseline were determined by the McCloskey-Kanis method (35, 36). Vertebral fractures were defined as "incident" if they occurred during the follow-up period.

Statistical and association analysis

We applied one-way ANOVA to investigate the relationship between genotypes and age, height, bone geometry variables, and other continuous outcomes. All genotyping results were tested for Hardy-Weinberg equilibrium (HWE). To clarify the possible confounding effect of potential factors on the association between VDR genotypes with height and bone geometry data, we used adjusted analysis of covariance for age, gender, and BMI, and we also stratified the analyses by gender, clinical, or x-ray-confirmed vertebral fracture. To test the association between the three fracture risk haplotype alleles in blocks 2, 3, and 5 with height, the normal significance cutoff P value (0.05) was used, because we had a clear upfront test hypothesis for those three haplotypes. We adjusted the significance cutoff P value using the Bonferroni correction in the association analyses of other seven common haplotypes (minor allele frequency > 5%) from LD block 1 and 4 as well as Fok I polymorphism. This resulted in a cutoff P value at 0.006 (0.05/8) considering that eight independent tests were performed.

Three possible genetic models were allowed to explain differences between genotype groups, i.e. an allele dose effect, a dominant effect, or a recessive effect. In case of a consistent trend reflected as an allele dose effect, a linear regression analysis was performed and a "trend" P value was calculated to quantify the association. In case of a recessive or dominant effect of the test allele, a 2 x 2 {chi}2 test for binary outcomes was performed to test for differences between two genotype groups. For dominant alleles, we compared the test group of "carrier" (e.g. heterozygous and homozygous carriers of the haplotype 3 allele in LD block 3, named "block 3-hap 3") vs. "noncarrier" (e.g. subjects without the block 3-hap 3 allele). For recessive effects, homozygous subjects for the test allele (e.g. homozygous carriers of the block 3-hap 3 allele) were compared with the combined group of heterozygous (e.g. heterozygous carriers of the block 3-hap 3 allele) and noncarriers (e.g. subjects without the block 3-hap 3 allele). The intragenic effect of combined VDR polymorphisms was defined as the combined genotypes for the block 3-hap 3 allele in the exon 1b promoter region and the block 5-hap 2 allele (the haplotype 2 of LD block 5) in the 3'-UTR. According to the genetic model observed for individual haplotype alleles in our study populations, we combined genotypes of the VDR gene with a dominant model for the block 3-hap 3 allele and an allele dose model for the block 5-hap 2 allele. "Zero" indicates noncarriers of the block 3-hap 3 allele and homozygous carriers of the block 5-hap 2 allele; "one" indicates either carrier of the block 3-hap 3 allele and homozygous carriers of the block 5-hap 2 allele or noncarriers of the block 3-hap 3 allele and heterozygous carriers of the block 5-hap 2 allele; "two" indicates noncarriers of the block 3-hap 3 allele and noncarriers of the block 5-hap 2 allele as well as carriers of the block 3-hap 3 allele and heterozygous carriers of the block 5-hap 2 allele; and "three" indicates carriers of the block 3-hap 3 allele and noncarriers of the block 5-hap 2 allele.

All statistical analyses of the association study were carried out with the SPSS software package (version 11.0 for Windows; SPSS, Inc., Chicago, IL), which was supported by the European Technical Support Team.

Meta-analysis

Identification of studies. We first searched publications on PubMed with a combination of keywords: [vitamin D receptor] and [polymorphism or genotype], and [height]. References from retrieved publications were checked for additional studies. In addition, we also reviewed 108 references from three meta-analysis studies (37, 38, 39) of VDR polymorphisms and different bone phenotypes to detect data of height distribution by VDR genotypes. In total, we identified 27 studies (40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66) with genotyping of Bsm I or Taq I or Bsm-Apa-Taq haplotyping and body height from 1995 to May 2005. We also included our current association results in the meta-analysis. Of studies from the same population, we retained for the meta-analysis only the study with the most extensive data.

Data analysis. Three VDR polymorphisms, Bsm I, Apa I, and Taq I- restriction fragment length polymorphisms, are commonly analyzed in genetic association studies, and Bsm I is the most consistently reported. Bsm I, Taq I, and haplotypes in block 5 are highly linked with a pairwise D' = (0.88–1.0) according to our recent study (23), "B" of Bsm I is linked to "t" of Taq I and the block 5-hap 2 allele. In the meta-analysis, we focused on analyzing the relationship of Bsm I with height; when Bsm I genotype was absent, Taq I and block 5-hap 2 genotype were used to predict the Bsm I genotype. We analyzed the available data using the RevMan 4.2 program from the Cochrane Collaboration (www.cochrane.dk). According to our cross-sectional analysis, the relationship between VDR Bsm I genotype and height followed an allele dose effect genetic model. We therefore evaluated the comparison between homozygous genotypes, BB vs. bb, BB vs. Bb, and Bb vs. bb in the meta-analysis. Because no or only one individual was observed in the BB genotype group from three Asian populations (52, 55, 63), we therefore only included these three studies in the comparison of Bb vs. bb. Because one study (65) only presented height data for B carriers vs. bb, we also analyzed height difference by BB + Bb vs. bb.

Sensitivity of meta-analysis. Between-study heterogeneity was evaluated by the {chi}2 distributed Q statistic with a cutoff P value level at 0.10 and the inconsistency index, I2 (suggesting inconsistency among studies’ results with values of 50% or higher and larger heterogeneity for values of 75% or higher), as estimated from the RevMan program. The standardized mean difference (SMD) and its confidence interval (CI) were estimated by both fixed effect and random effects models. We stratified studies by ethnicity and by age to attempt to clarify potential between-study heterogeneity. We carried out sensitivity analyses by excluding one study with the largest sample size and/or excluding studies deviating from HWE proportions. We also performed recursive cumulative meta-analysis to evaluate whether the summary SMD changed in the same direction over time and appraised whether the first published study gave different results from subsequent ones (67, 68, 69).


   Results
Top
 Abstract
 Introduction
 Subjects and Methods
 Results
Discussion
 References
 
Cladogram of VDR haplotype blocks

Nine SNPs around exon 1b and 10 SNPs from exon 4 to the end of the 3'-UTR of the VDR gene, as described in our recent study (23), were found to be highly linked in block 3 (D' is 0.93–1.00) and block 5 (D' is 0.88–1.00), respectively (Fig. 1Go). Twelve haplotype alleles were predicted by PHASE in block 3 and 18 in block 5 for these two Caucasian populations. The cladogram graphs showed that these haplotypes could be categorized as three main clusters (labeled as I, II, and III in Fig. 1Go) in the blocks with one or two common haplotype alleles (frequency > 10%). Some SNPs (with arrow below the figure) are specific for certain clusters, e.g. 1b-G-2528A and 1b-G-886A for cluster III in LD block 3 and E9-T-48G, U-A311C, and U-D796T for cluster I in LD block 5. We combined this feature with the potential functionality (possible transcriptional factor binding site) and availability of a reliable genotyping method together to determine haplotype tagging SNPs [htSNPs (23)]. We identified seven htSNPs, which can represent most (93–95%) of the estimated five common haplotype alleles (frequency > 3%) in block 3 (four htSNPs) and block 5 (three htSNPs), respectively. Therefore, these htSNPs were used in our present association study.


Figure 1
View larger version (23K):
[in this window]
[in a new window]

  		FIG. 1. Constructed haplotypes and cladograms of VDR LD blocks 3 and 5. Top, The genomic structure of the VDR gene. Bottom, Cladograms and haplotype constructs. The LD map shows five determined haplotype blocks with black lines based on our data and two expected blocks with gray lines based on other data (23 ). Analyzed SNPs for blocks 3 and 5 are presented in boxes; seven bold SNPs are htSNPs of the blocks and were used for the association study. Three main haplotype clusters (I, II, and III) are separated by two dashed lines for each block. Two SNPs (with arrow on the bottom of the figure) are specific for the cluster III in block 2 and three SNPs for cluster I in block 5. The most frequent haplotype alleles are presented on the top of the LD block structures; dashes in the haplotype alleles indicate that the SNP allele is the same as the one in the first row on top. "Hap*" indicates haplotype code according to all analyzed SNPs in the blocks; the most frequent haplotype allele was assigned as hap 1. "Hap#" indicates haplotype code according to htSNPs in the blocks. Haplotypes used in the association analyses are boxed.

 
VDR haplotypes and body height

We genotyped seven htSNPs for blocks 3 and 5 in the Rotterdam and LASA populations. The genotype distribution of all htSNPs followed HWE proportions. The baseline characteristics of each study population are shown in Table 1Go. Data from the Rotterdam and LASA populations are presented by the combined VDR genotype (as described in the Subjects and Methods section). No significant differences in baseline characteristics were found according to VDR genotype. The allele frequencies of block 3-hap 3 and block 5-hap 2 were similar between the Rotterdam and LASA populations.


View this table:
[in this window]
[in a new window]

  		TABLE 1. Characteristics by the VDR genotype in different study populations

 
We first investigated the relationship of body height in the Rotterdam population with the haplotype risk alleles for fracture (Table 2Go). Block 3-hap 3 carriers were found to have a 0.3% decreased body height, and the block 5-hap 2 homozygotes were found to have a 0.5% increased body height with evidence for an allele dose effect. The combined genotype of block 3-hap 3 and block 5-hap 2 had an additive effect on body height. The height difference between two extreme genotype groups was 1.4 cm (0.8%; P = 0.00002, corresponding to 0.15 SD values). The same effects were replicated in the LASA population using the same genetic models, which showed 0.7% decreased height for block 3-hap 3 carriers, 0.8% increased height for block 5-hap 2 homozygotes, and 1.6% difference for the combined genotype. We observed that the VDR genotype-dependent height difference by block 5-hap 2 genotype increased with increasing mean age of the study population. This increase was 0.9 cm (0.09 SD) with mean age 70 yr in the Rotterdam Study and 1.4 cm (0.15 SD) with mean age 76 yr in LASA. We then grouped genotype groups, which have a similar direction of the effect, i.e. carriers of the block 3-hap 3 allele and noncarriers of the block 5-hap 2 allele, which both are associated with decreased height. Together these subjects represent 10% of the population and have 0.8 and 1.4 cm decreased body height compared with the mean height in the Rotterdam and LASA populations, respectively. This corresponds to an attributable fraction of explained genetic variance of 0.1% (in the model with VDR genotype, age, and gender, the total R2 for height is 0.55).


View this table:
[in this window]
[in a new window]

  		TABLE 2. Height by VDR genotype in different study populations

 
An inverse correlation between body height and age was observed for both men and women in the Rotterdam Study (r = –0.28; P < 0.001) and the LASA cohort (r = –0.21; P < 0.001). We then analyzed the relationship between VDR genotypes and body height by 5-yr age strata to investigate the contribution of the VDR genotypes to age-related genotype-dependent height differences (secular trends). In all the age categories of the two populations, VDR genotype was associated with body height. In addition, we analyzed "height loss" longitudinally during a 7.4-yr follow-up in the Rotterdam Study. Height loss was not significantly different by VDR genotype. We also examined the relationship between body height and VDR haplotype alleles from other LD blocks, but no evidence of association was observed (data not shown).

Vertebral fracture is known to be an important confounder to influence associations regarding body height in elderly subjects. In the Rotterdam population, there were 98 cases [of 6031 analyzed (1.6%)] with incident clinical vertebral fracture during the follow-up period and 232 x-ray-confirmed vertebral fractures [of 3114 screened (7.5%)] at baseline. In the LASA cohort, seven cases [of 902 analyzed (0.8%)] of incident clinical vertebral fracture were reported during the follow-up time. When we only analyzed subjects without clinical or x-ray-confirmed vertebral fracture, the association of VDR genotype with height remained essentially the same (Table 3Go). We did not observe dietary vitamin D intake or serum vitamin D level to influence the association of VDR genotype and height in the complete study or a subset of 1331 individuals (for vitamin D levels) from the Rotterdam Study population (data not shown).


View this table:
[in this window]
[in a new window]

  		TABLE 3. No influence of vertebral fractures on the association between height and combined VDR genotypes in the Rotterdam Study

 
We considered the hip structure analysis indices as a parameter of bone geometry and looked for an association with VDR genotype in an attempt to look for biological consistency in the pattern of association we found with height. Thus, we also analyzed bone geometry parameters by VDR genotype available in the Rotterdam Study (Table 4Go). The combined VDR genotype was associated with decreased vertebral area of L2–L4 and decreased femoral size parameters (external and internal diameters) at the narrow neck and at the shaft and the lower section modulus (index of bending strength) of the femoral neck. Associations remained when we adjusted for age, gender, vertebral fracture, height, and weight. We did not see significant differences in cortical thickness by VDR genotype.


View this table:
[in this window]
[in a new window]

  		TABLE 4. Vertebral and hip bone geometry by combined VDR genotype in the Rotterdam Study

 
Meta-analysis of VDR genotype and body height

We identified 27 eligible published studies and the two current studies, in total reporting data from 35 different comparisons, which together included 14,157 subjects from different ethnic populations (Table 5Go). Study design contained cohort, case-control, normal subject-based, and hospital-based cross-sectional studies. The mean age for each study was 2 to 75 yr for both males and females. The sample size of studies varied from 24 to 6154. Twenty-one comparisons presented Bsm I and height data, four comparisons had Taq I genotype data, whereas six studies had Bsm I and Apa I and Taq I or Bsm-Apa-Taq haplotype data. Because of the high LD, we assumed the "t"-allele (or the C nucleotide) of Taq I, the haplotype 2 allele of Bsm-Apa-Taq (54), or the block 5-hap 2 allele (Ref. 26 and the current study) to represent the "B" allele (or the A nucleotide) of Bsm I. The frequency of the B-allele ranged from 3 to 49%.


View this table:
[in this window]
[in a new window]

  		TABLE 5. Studies of VDR Bsm I genotype and body height included in the meta-analysis

 
In our current association analysis, an allele-dose genetic model was observed for the block 5-hap 2 allele. In the meta-analysis, we therefore compared the extreme genotype groups, BB (2447 subjects) vs. bb (5329 subjects) based on an allele-dose genetic model. A significant between-study heterogeneity was observed (P = 0.0002; 52%). The random effects model was used to test the overall genetic effect of Bsm I genotype on body height from all comparisons (Fig. 2Go, bottom). Subjects with BB genotype were found to be 0.63 cm (95% CI, 0.17–1.08) higher compared with bb genotype carriers (P = 0.006). We also performed comparisons of BB vs. Bb and Bb vs. bb. A trend of decreasing height was observed for BB, Bb, and bb genotypes, respectively (data not shown). This indicates that the B-allele of VDR Bsm I is associated with increased body height with both models.


Figure 2
View larger version (43K):
[in this window]
[in a new window]

  		FIG. 2. Meta-analysis of the relationship between BB and bb genotypes of the VDR Bsm I restriction fragment length polymorphism and body height. Point estimates of the SMD of height between the genotypes and 95% CIs are presented for each study group. Summary estimates of SMD and their 95% CIs (diamonds) are given by random effects models for all studies combined at the bottom, as well as for subtotals in "young" (younger than 20 yr) and "adult" (older than 20 yr) strata, with the adult studies being further stratified by B-allele frequency being low (3–9%) or high (27–49%).

 
Body height changes remarkably during puberty, and the age of puberty differs between individuals. To investigate the source of the heterogeneity, we first stratified the analyses by age (younger than 20 yr or older than 20 yr; Table 6Go). Evidence of strong heterogeneity was found in the young (younger than 20 yr) group (P = 0.002; I2 = 76%, in the upper part of Fig. 2Go), and no association was found. In the adult (older than 20 yr) group, a significant but mild heterogeneity was found (P = 0.003; I2 = 47%) and a significant and strong association was observed [P = 0.002; height difference, 0.75 (0.34, 1.15) cm]. Five comparisons for Asian populations (51, 52, 55, 63), which had a significantly lower B-allele frequency (3–9%; P < 0.001), showed modest heterogeneity (P = 0.04; I2 = 61%). Other studies with a higher B-allele frequency [27–49%, including one study (60) of African-Americans with a B-allele frequency of 36%] showed a borderline significant heterogeneity (P = 0.05; I2 = 34%). The point estimates of the height difference by genotype were similar between the random effects model and the fixed effect model, and the genetic effect of Bsm I genotype on height remained. After stratifying for gender, status of menopause (pre- vs. postmenopausal), and study design, we still could see heterogeneity in all strata (data not shown).


View this table:
[in this window]
[in a new window]

  		TABLE 6. Mean difference in body height between VDR BsmI BB and bb genotype in the meta-analysis

 
To test the sensitivity of the meta-analysis, we replaced the study with the largest sample size, i.e. the current complete Rotterdam Study population with previously published data using a subset (1004 subjects) of the same population (54). This did not have any impact on the conclusion of association and heterogeneity, but the height difference by VDR genotype became somewhat smaller [P = 0.03; mean difference (95% CI), 0.57 (0.04–1.09) cm in the random effects model]. When we excluded two studies that deviated from HWE [Garnero et al. (57) and Laaksonen et al. (48)], the association remained the same.


   Discussion
Top
 Abstract
 Introduction
 Subjects and Methods
 Results
 Discussion
References
 
Although several polymorphisms of the VDR gene have been assessed in the association analyses of body height (45, 46, 48, 70, 71), conflicting results were reported. One reason for this might be that nonfunctional or noncausal polymorphisms were used in the studies. Recently, we systematically analyzed the LD across the whole VDR gene and determined five haplotype blocks and 15 htSNPs to represent all common haplotypes in each block (23). The cladogram enables inference of evolutionary history patterns and identifies recombination events (72). "Old" alleles were mostly common alleles, whereas recombination is common but concentrates into hot spots and recurrent mutations at multiple sites may have occurred (73). We combined cladistic and haplotype analyses and identified a small number of common haplotype alleles (frequency > 10%) in LD blocks, which contain polymorphisms after recombination. In an LD block, haplotype alleles within the clusters of the cladogram are more similar than haplotypes between clusters. The hypothesis of the common disease/common variation hypothesis (74) indicates that a common risk allele accounts for substantial contribution to disease in a population. In the association study, we therefore only focused on common haplotype alleles in each block. In our previous study, the homozygous block 3-hap 3 genotype and the block 5-hap 2 carriers were found to be associated with 74% increased and 23% decreased fracture risk, respectively. In the consequent functionality analysis, we found that in vitro assays of the block 5-hap 2 allele showed a 15% higher VDR mRNA level compared with the block 5-hap 1 allele in five different cell lines (including one osteoblast cell line). We here demonstrate an association between these two alleles and adult body height in two independent large populations of elderly subjects.

Interestingly, the summarized genetic effect of Bsm I genotype on height from our meta-analysis was in line with our association results, although a heterogeneity among studies was found. The height difference between BB and bb genotype groups was 0.63 cm. In the sensitivity analysis, we found that the result of the Rotterdam Study did not substantially influence the significance of the association and heterogeneity in the meta-analysis, although the magnitude of the height difference decreased when we removed our current study from the meta-analysis. Taken together, this indicates that the VDR polymorphisms have a modest and constant genetic effect throughout life to influence the body height.

To investigate the source of the between-study heterogeneity, we first stratified the analysis by age in young (younger than] 20 yr) or adult (older than 20 yr) groups. In young populations, large heterogeneity was found, because the direction of the effects is different between studies. Childhood and adolescence is the period of most rapid skeletal growth in an individual’s lifetime. After this period, the stature tends to be relatively stable and is more suitable for investigation of the relationship between adult body height and genetic effects of VDR genotypes. However, in the adult group, we still observed significant but mild heterogeneity. In the elderly population of the Rotterdam Study, the VDR genotype effect is proposed to be transmitted to body height through a complex mechanism that also involves the strength of the vertebrae. To what extent this effect might be different in younger populations has to be established in additional genetic association studies of VDR polymorphisms and height in younger subjects. Further stratification by the VDR Bsm I B-allele frequency was carried out in the adult group. In the low B-allele frequency group (frequency 3–9%), a modest heterogeneity resulted from a different magnitude of genetic effects between studies, although a similar direction of the effect was found between studies. This probably is because of insufficient statistical power to detect the genetic effect in the small number of BB homozygotes at this low allele frequency. In the high B-allele frequency group (frequency, 27–49%), a borderline significant heterogeneity was found, and other factors of heterogeneity need to be further analyzed in this group. According to our previous study (23), only Asian populations have a lower B-allele frequency (5%) compared with Caucasian and African-American populations (39 and 36%, respectively). Therefore, age and B-allele frequency variation could both influence heterogeneity of this meta-analysis. Although heterogeneity between studies was observed, the genetic effect of Bsm I genotype on body height remained, especially in the adult population.

As we demonstrated, adult body height (this study) and fracture risk (23) are two complex skeletal phenotypes related to the variations of the VDR gene. We demonstrated fracture risk alleles of the VDR gene to be associated with decreased body height, decreased bone size (e.g. vertebral area, femoral external and internal diameters) and lower bending strength (section modulus) at the femoral neck. Similarly, the VDR genotype effect we observed on vertebral area could contribute to the VDR genotype effect we observed on stature. Hence, the correlations of VDR genotype with body height and bone size are in line with each other. Therefore, the findings of our current study together with those of our previous study (26) indicate a genetic effect of VDR genotype on related skeletal phenotypes, e.g. fracture risk, short stature, and small bone size. However, we did not observe an association between VDR genotype and BMD (nor with bone mineral content), and BMD was not observed to influence the association of VDR genotype with fracture risk in our population. Therefore, the mechanism of the association between VDR genotype and fracture is speculated to be through bone size effects rather than BMD.

Another fracture-risk haplotype allele found in the Rotterdam Study population (23), i.e. the block 2-hap1 allele, was not found to be related to height differences in the Rotterdam Study and the LASA population. However, this haplotype allele was recently found to be associated with decreased body height in French adolescent girls (75), whereas this haplotype allele was also associated with decreased serum 25-hydroxyvitamin D3 level and decreased serum IGF-I level in this population. Interestingly, the French population has a relatively low dietary calcium intake (<865 mg/d), whereas the elderly Dutch populations have a relatively high dietary calcium intake (mean, 1117 mg/d). The VDR Cdx-2 polymorphism is one of the htSNPs for this LD block 2 and has been implicated in vitamin D-regulated genetic differences in calcium absorption through the intestine (76). Therefore, the interaction between the block 2-hap 1 allele and dietary calcium intake and/or serum vitamin D levels could influence bone phenotypes, and thus, this aspect is of interest for further investigation in other young and old populations.

In our functionality study of VDR alleles, we observed the block 5-hap 2 allele to result in 15% higher mRNA level and stability compared with the block 5-hap 1 allele in several cell lines, including an osteoblast cell line (23). Although further studies remain to confirm whether this difference can influence VDR protein expression and the sensitivity of the cells to vitamin D, a possible underlying mechanism might be that for individuals with the block 5-hap 2 haplotype allele, bone tissue (especially osteoblast cells) has a higher sensitivity to vitamin D ligands because of higher expression of VDR and higher osteoblast activity. As a result, block 5-hap 2 allele carriers might have relatively higher bone formation to make somewhat bigger and stronger bones that are more resistant to fracture.

In addition, the intervertebral disc height and toughness might influence the body height, especially in the elderly populations. In different populations, the "t" allele of the VDR Taq I polymorphism was reported to be associated with more degeneration of the thoracic and lumbar discs (29), increased risk of disc bulges (30), and disc degeneration and herniation (31). Unfortunately, in our study population, we do not have data on intervertebral disc characteristics to investigate possible differences in disc parameters by VDR genotype.

In conclusion, we observed haplotype alleles in the promoter region and the 3'-end of the VDR gene to be associated with decreased body height in two elderly populations. This association was independent of age, gender, and presence of vertebral fracture. A meta-analysis of published data corroborated the relationship between these alleles at the 3'-end of the VDR gene and decreased "adult" height. The underlying mechanism of the association might involve slightly lower copy numbers of VDR protein (in carriers of these risk haplotypes) in cells important for determining bone size and strength.


   Acknowledgments
 
We thank all participants of the Rotterdam, LASA, and EPOS studies. In particular, we acknowledge the general practitioners, pharmacists, and the many field workers at the research center in Ommoord, Rotterdam, The Netherlands, for their help in collecting the data for the Rotterdam Study. We thank Dr. Thomas J. Beck (Johns Hopkins University, Baltimore, MD), who provided the hip structure analysis of dual-energy x-ray absorptiometry scans.


   Footnotes
 
This project was funded by the Dutch Research Organisation (NWO 903-46-178, 925-01-010, 014-90-001, and 911-03-012), the European Commission under grant QLK6-CT-2002-02629 ("GENOMOS"), and the Dutch Ministry of Health, Welfare, and Sports.

Disclosure Statement: The authors have nothing to declare.

First Published Online January 9, 2007

Abbreviations: BMD, Bone mineral density; BMI, body mass index; CI, confidence interval; htSNPs, haplotype tagging SNPs; HVDRR, hereditary vitamin D-resistant rickets; HWE, Hardy-Weinberg equilibrium; I2, inconsistency index; LASA, Longitudinal Aging Study Amsterdam; LD, linkage disequilibrium; SMD, standardized mean difference; SNP, single nucleotide polymorphism; UTR, untranslated region; VDR, vitamin D receptor.

Received May 25, 2006.

Accepted January 2, 2007.


   References
Top
 Abstract
 Introduction
 Subjects and Methods
 Results
 Discussion
 References
 

  1. Silventoinen K, Kaprio J, Lahelma E, Viken RJ, Rose RJ 2003 Assortative mating by body height and BMI: Finnish twins and their spouses. Am J Hum Biol 15:620–627[CrossRef][Medline]
  2. Schousboe K, Visscher PM, Erbas B, Kyvik KO, Hopper JL, Henriksen JE, Heitmann BL, Sorensen TI 2004 Twin study of genetic and environmental influences on adult body size, shape, and composition. Int J Obes Relat Metab Disord 28:39–48[CrossRef][Medline]
  3. Leiberman E, Pesler D, Parvari R, Elbedour K, Abdul-Latif H, Brown MR, Parks JS, Carmi R 2000 Short stature in carriers of recessive mutation causing familial isolated growth hormone deficiency. Am J Med Genet 90:188–192[CrossRef][Medline]
  4. Schwinger E, Kirschstein M, Greiwe M, Konermann T, Orth U, Gal A 1996 Short stature in a mother and daughter with terminal deletion of Xp22.3. Am J Med Genet 63:239–242[CrossRef][Medline]
  5. Niewenweg R, Smit ML, Walenkamp MJ, Wit JM 2003 Adult height corrected for shrinking and secular trend. Ann Hum Biol 30:563–569[CrossRef][Medline]
  6. Sorlie P, Gordon T, Kannel WB 1980 Body build and mortality. The Framingham study. JAMA 243:1828–1831[Abstract]
  7. Hebert PR, Rich-Edwards JW, Manson JE, Ridker PM, Cook NR, O’Connor GT, Buring JE, Hennekens CH 1993 Height and incidence of cardiovascular disease in male physicians. Circulation 88:1437–1443[Abstract/Free Full Text]
  8. Rich-Edwards JW, Manson JE, Stampfer MJ, Colditz GA, Willett WC, Rosner B, Speizer FE, Hennekens CH 1995 Height and the risk of cardiovascular disease in women. Am J Epidemiol 142:909–917[Abstract/Free Full Text]
  9. Han TS, Feskens EJ, Lean ME, Seidell JC 1998 Associations of body composition with type 2 diabetes mellitus. Diabet Med 15:129–135[CrossRef][Medline]
  10. Albanes D, Jones DY, Schatzkin A, Micozzi MS, Taylor PR 1988 Adult stature and risk of cancer. Cancer Res 48:1658–1662[Abstract/Free Full Text]
  11. Jousilahti P, Tuomilehto J, Vartiainen E, Eriksson J, Puska P 2000 Relation of adult height to cause-specific and total mortality: a prospective follow-up study of 31,199 middle-aged men and women in Finland. Am J Epidemiol 151:1112–1120[Abstract/Free Full Text]
  12. Song YM, Smith GD, Sung J 2003 Adult height and cause-specific mortality: a large prospective study of South Korean men. Am J Epidemiol 158:479–485[Abstract/Free Full Text]
  13. Uitterlinden AG, Fang Y, Van Meurs JB, Pols HA, Van Leeuwen JP 2004 Genetics and biology of vitamin D receptor polymorphisms. Gene 338:143–156[CrossRef][Medline]
  14. Grimes DS 2006 Are statins analogues of vitamin D? Lancet 368:83–86[CrossRef][Medline]
  15. Hughes MR, Malloy PJ, Kieback DG, Kesterson RA, Pike JW, Feldman D, O’Malley BW 1988 Point mutations in the human vitamin D receptor gene associated with hypocalcemic rickets. Science 242:1702–1705[Abstract/Free Full Text]
  16. Kristjansson K, Rut AR, Hewison M, O’Riordan JL, Hughes MR 1993 Two mutations in the hormone binding domain of the vitamin D receptor cause tissue resistance to 1,25 dihydroxyvitamin D3. J Clin Invest 92:12–16[Medline]
  17. Cockerill FJ, Hawa NS, Yousaf N, Hewison M, O’Riordan JL, Farrow SM 1997 Mutations in the vitamin D receptor gene in three kindreds associated with hereditary vitamin D resistant rickets. J Clin Endocrinol Metab 82:3156–3160[Abstract/Free Full Text]
  18. Herweijer TJ, Steendijk R 1985 The relation between attained adult height and the metaphyseal lesions in hypophosphataemic vitamin-D resistant rickets. Acta Paediatr Scand 74:196–200[Medline]
  19. Chan JC, Bartter FC 1979 Hypophosphatemic rickets: effect of 1 {alpha}, 25-dihydroxyvitamin D3 on growth and mineral metabolism. Pediatrics 64:488–495[Abstract/Free Full Text]
  20. Erben RG, Soegiarto DW, Weber K, Zeitz U, Lieberherr M, Gniadecki R, Moller G, Adamski J, Balling R 2002 Deletion of deoxyribonucleic acid binding domain of the vitamin D receptor abrogates genomic and nongenomic functions of vitamin D. Mol Endocrinol 16:1524–1537[Abstract/Free Full Text]
  21. Hirschhorn JN, Lindgren CM, Daly MJ, Kirby A, Schaffner SF, Burtt NP, Altshuler D, Parker A, Rioux JD, Platko J, Gaudet D, Hudson TJ, Groop LC, Lander ES 2001 Genomewide linkage analysis of stature in multiple populations reveals several regions with evidence of linkage to adult height. Am J Hum Genet 69:106–116[CrossRef][Medline]
  22. Xu J, Bleecker ER, Jongepier H, Howard TD, Koppelman GH, Postma DS, Meyers DA 2002 Major recessive gene(s) with considerable residual polygenic effect regulating adult height: confirmation of genomewide scan results for chromosomes 6, 9, and 12. Am J Hum Genet 71:646–650[CrossRef][Medline]
  23. Fang Y, van Meurs JB, d’Alesio A, Jhamai M, Zhao H, Rivadeneira F, Hofman A, van Leeuwen JP, Jehan F, Pols HA, Uitterlinden AG 2005 Promoter and 3'-untranslated-region haplotypes in the vitamin D receptor gene predispose to osteoporotic fracture: the Rotterdam Study. Am J Hum Genet 77:807–823[CrossRef][Medline]
  24. Nejentsev S, Godfrey L, Snook H, Rance H, Nutland S, Walker NM, Lam AC, Guja C, Ionescu-Tirgoviste C, Undlien DE, Ronningen KS, Tuomilehto-Wolf E, Tuomilehto J, Newport MJ, Clayton DG, Todd JA 2004 Comparative high-resolution analysis of linkage disequilibrium and tag single nucleotide polymorphisms between populations in the vitamin D receptor gene. Hum Mol Genet 13:1633–1639[Abstract/Free Full Text]
  25. Beck TJ, Looker AC, Ruff CB, Sievanen H, Wahner HW 2000 Structural trends in the aging femoral neck and proximal shaft: analysis of the Third National Health and Nutrition Examination Survey dual-energy X-ray absorptiometry data. J Bone Miner Res 15:2297–2304[CrossRef][Medline]
  26. Uitterlinden AG, Pols HA, Burger H, Huang Q, Van Daele PL, Van Duijn CM, Hofman A, Birkenhager JC, Van Leeuwen JP 1996 A large-scale population-based study of the association of vitamin D receptor gene polymorphisms with bone mineral density. J Bone Miner Res 11:1241–1248[Medline]
  27. Hofman A, Grobbee DE, de Jong PT, van den Ouweland FA 1991 Determinants of disease and disability in the elderly: the Rotterdam Elderly Study. Eur J Epidemiol 7:403–422[CrossRef][Medline]
  28. Deeg DJH, Beekman ATF, Kriegsman DMW, Westendorp de Seriere M 1994 Autonomy and well-being in the Aging population I. Report from the Longitudinal Aging Study Amsterdam 1992–1993. Amsterdam: VU University Press 1998:9–22.
  29. Videman T, Leppavuori J, Kaprio J, Battie MC, Gibbons LE, Peltonen L, Koskenvuo M 1998 Intragenic polymorphisms of the vitamin D receptor gene associated with intervertebral disc degeneration. Spine 23:2477–2485[CrossRef][Medline]
  30. Videman T, Gibbons LE, Battie MC, Maravilla K, Vanninen E, Leppavuori J, Kaprio J, Peltonen L 2001 The relative roles of intragenic polymorphisms of the vitamin d receptor gene in lumbar spine degeneration and bone density. Spine 26:E7–E12
  31. Kawaguchi Y, Kanamori M, Ishihara H, Ohmori K, Matsui H, Kimura T 2002 The association of lumbar disc disease with vitamin-D receptor gene polymorphism. J Bone Joint Surg Am 84:2022–2028[Abstract/Free Full Text]
  32. Zhang L, Heng CK, Tan TW 2001 Cladogramer: incorporating haplotype frequency into cladogram analysis. Bioinformatics 17:481–482[Abstract/Free Full Text]
  33. Burger H, van Daele PL, Algra D, van den Ouweland FA, Grobbee DE, Hofman A, van Kuijk C, Schutte HE, Birkenhager JC, Pols HA 1994 The association between age and bone mineral density in men and women aged 55 years and over: the Rotterdam Study. Bone Miner 25:1–13[Medline]
  34. Rivadeneira F, Houwing-Duistermaat JJ, Beck TJ, Janssen JA, Hofman A, Pols HA, Van Duijn CM, Uitterlinden AG 2004 The influence of an insulin-like growth factor I gene promoter polymorphism on hip bone geometry and the risk of nonvertebral fracture in the elderly: the Rotterdam Study. J Bone Miner Res 19:1280–1290[CrossRef][Medline]
  35. Fang Y, van Meurs JB, Bergink AP, Hofman A, van Duijn CM, van Leeuwen JP, Pols HA, Uitterlinden AG 2003 Cdx-2 polymorphism in the promoter region of the human vitamin D receptor gene determines susceptibility to fracture in the elderly. J Bone Miner Res 18:1632–1641[CrossRef][Medline]
  36. McCloskey EV, Spector TD, Eyres KS, Fern ED, O’Rourke N, Vasikaran S, Kanis JA 1993 The assessment of vertebral deformity: a method for use in population studies and clinical trials. Osteoporos Int 3:138–147[CrossRef][Medline]
  37. Cooper GS, Umbach DM 1996 Are vitamin D receptor polymorphisms associated with bone mineral density? A meta-analysis. J Bone Miner Res 11:1841–1849[Medline]
  38. Gong G, Stern HS, Cheng SC, Fong N, Mordeson J, Deng HW, Recker RR 1999 The association of bone mineral density with vitamin D receptor gene polymorphisms. Osteoporos Int 9:55–64[CrossRef][Medline]
  39. Thakkinstian A, D’Este C, Eisman J, Nguyen T, Attia J 2004 Meta-analysis of molecular association studies: vitamin D receptor gene polymorphisms and BMD as a case study. J Bone Miner Res 19:419–428[CrossRef][Medline]
  40. Garnero P, Borel O, Sornay-Rendu E, Delmas PD 1995 Vitamin D receptor gene polymorphisms do not predict bone turnover and bone mass in healthy premenopausal women. J Bone Miner Res 10:1283–1288[Medline]
  41. Keen RW, Egger P, Fall C, Major PJ, Lanchbury JS, Spector TD, Cooper C 1997 Polymorphisms of the vitamin D receptor, infant growth, and adult bone mass. Calcif Tissue Int 60:233–235[CrossRef][Medline]
  42. Sainz J, Van Tornout JM, Loro ML, Sayre J, Roe TF, Gilsanz V 1997 Vitamin D-receptor gene polymorphisms and bone density in prepubertal American girls of Mexican descent. N Engl J Med 337:77–82[Abstract/Free Full Text]
  43. Suarez F, Zeghoud F, Rossignol C, Walrant O, Garabedian M 1997 Association between vitamin D receptor gene polymorphism and sex-dependent growth during the first two years of life. J Clin Endocrinol Metab 82:2966–2970[Abstract/Free Full Text]
  44. Ferrari SL, Rizzoli R, Slosman DO, Bonjour JP 1998 Do dietary calcium and age explain the controversy surrounding the relationship between bone mineral density and vitamin D receptor gene polymorphisms? J Bone Miner Res 13:363–370[CrossRef][Medline]
  45. Gennari L, Becherini L, Masi L, Mansani R, Gonnelli S, Cepollaro C, Martini S, Montagnani A, Lentini G, Becorpi AM, Brandi ML 1998 Vitamin D and estrogen receptor allelic variants in Italian postmenopausal women: evidence of multiple gene contribution to bone mineral density. J Clin Endocrinol Metab 83:939–944[Abstract/Free Full Text]
  46. Gomez C, Naves ML, Barrios Y, Diaz JB, Fernandez JL, Salido E, Torres A, Cannata JB 1999 Vitamin D receptor gene polymorphisms, bone mass, bone loss and prevalence of vertebral fracture: differences in postmenopausal women and men. Osteoporos Int 10:175–182[CrossRef][Medline]
  47. Lorentzon M, Lorentzon R, Nordstrom P 2000 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 85:1666–1670[Abstract/Free Full Text]
  48. Laaksonen M, Karkkainen M, Outila T, Vanninen T, Ray C, Lamberg-Allardt C 2002 Vitamin D receptor gene BsmI-polymorphism in Finnish premenopausal and postmenopausal women: its association with bone mineral density, markers of bone turnover, and intestinal calcium absorption, with adjustment for lifestyle factors. J Bone Miner Metab 20:383–390[CrossRef][Medline]
  49. Taverna MJ, Sola A, Guyot-Argenton C, Pacher N, Bruzzo F, Slama G, Reach G, Selam JL 2002 Taq I polymorphism of the vitamin D receptor and risk of severe diabetic retinopathy. Diabetologia 45:436–442[CrossRef][Medline]
  50. Grundberg E, Brandstrom H, Ribom EL, Ljunggren O, Kindmark A, Mallmin H 2003 A poly adenosine repeat in the human vitamin D receptor gene is associated with bone mineral density in young Swedish women. Calcif Tissue Int 73:455–462[CrossRef][Medline]
  51. Kim JG, Kwon JH, Kim SH, Choi YM, Moon SY, Lee JY 2003 Association between vitamin D receptor gene haplotypes and bone mass in postmenopausal Korean women. Am J Obstet Gynecol 189:1234–1240[CrossRef][Medline]
  52. Kurabayashi T, Matsushita H, Tomita M, Kato N, Kikuchi M, Nagata H, Honda A, Yahata T, Tanaka K 2004 Association of vitamin D and estrogen receptor gene polymorphism with the effects of longterm hormone replacement therapy on bone mineral density. J Bone Miner Metab 22:241–247[CrossRef][Medline]
  53. Garnero P, Munoz F, Borel O, Sornay-Rendu E, Delmas PD 2005 Vitamin D receptor gene polymorphisms are associated with the risk of fractures in postmenopausal women, independently of bone mineral density. The OFELY study. J Clin Endocrinol Metab 90:4829–4835[Abstract/Free Full Text]
  54. Uitterlinden AG, Weel AE, Burger H, Fang Y, van Duijn CM, Hofman A, van Leeuwen JP, Pols HA 2001 Interaction between the vitamin D receptor gene and collagen type I{alpha}1 gene in susceptibility for fracture. J Bone Miner Res 16:379–385[CrossRef][Medline]
  55. Tsai KS, Hsu SH, Cheng WC, Chen CK, Chieng PU, Pan WH 1996 Bone mineral density and bone markers in relation to vitamin D receptor gene polymorphisms in Chinese men and women. Bone 19:513–518[Medline]
  56. Need AG, Horowitz M, Stiliano A, Scopacasa F, Morris HA, Chatterton BE 1996 Vitamin D receptor genotypes are related to bone size and bone density in men. Eur J Clin Invest 26:793–796[CrossRef][Medline]
  57. Garnero P, Borel O, Sornay-Rendu E, Arlot ME, Delmas PD 1996 Vitamin D receptor gene polymorphisms are not related to bone turnover, rate of bone loss, and bone mass in postmenopausal women: the OFELY Study. J Bone Miner Res 11:827–834[Medline]
  58. Vandevyver C, Wylin T, Cassiman JJ, Raus J, Geusens P 1997 Influence of the vitamin D receptor gene alleles on bone mineral density in postmenopausal and osteoporotic women. J Bone Miner Res 12:241–247[CrossRef][Medline]
  59. Alahari KD, Lobaugh B, Econs MJ 1997 Vitamin D receptor alleles do not correlate with bone mineral density in premenopausal Caucasian women from the southeastern United States. Metabolism 46:224–226[CrossRef][Medline]
  60. Zmuda JM, Cauley JA, Danielson ME, Wolf RL, Ferrell RE 1997 Vitamin D receptor gene polymorphisms, bone turnover, and rates of bone loss in older African-American women. J Bone Miner Res 12:1446–1452[CrossRef][Medline]
  61. Hansen TS, Abrahamsen B, Henriksen FL, Hermann AP, Jensen LB, Horder M, Gram J 1998 Vitamin D receptor alleles do not predict bone mineral density or bone loss in Danish perimenopausal women. Bone 22:571–575[Medline]
  62. Tsuritani I, Brooke-Wavell KS, Mastana SS, Jones PR, Hardman AE, Yamada Y 1998 Does vitamin D receptor polymorphism influence the response of bone to brisk walking in postmenopausal women? Horm Res 50:315–319[CrossRef][Medline]
  63. Kikuchi R, Uemura T, Gorai I, Ohno S, Minaguchi H 1999 Early and late postmenopausal bone loss is associated with BsmI vitamin D receptor gene polymorphism in Japanese women. Calcif Tissue Int 64:102–106[CrossRef][Medline]
  64. Holmberg-Marttila D, Sievanen H, Jarvinen TL, Jarvinen TA 2000 Vitamin D and estrogen receptor polymorphisms and bone mineral changes in postpartum women. Calcif Tissue Int 66:184–189[CrossRef][Medline]
  65. Remes T, Vaisanen SB, Mahonen A, Huuskonen J, Kroger H, Jurvelin JS, Rauramaa R 2005 Bone mineral density, body height, and vitamin D receptor gene polymorphism in middle-aged men. Ann Med 37:383–392[CrossRef][Medline]
  66. Nguyen TV, Esteban LM, White CP, Grant SF, Center JR, Gardiner EM, Eisman JA 2005 Contribution of the collagen I {alpha} 1 and vitamin D receptor genes to the risk of hip fracture in elderly women. J Clin Endocrinol Metab 90:6575–6579[Abstract/Free Full Text]
  67. Ioannidis JP, Contopoulos-Ioannidis DG, Lau J 1999 Recursive cumulative meta-analysis: a diagnostic for the evolution of total randomized evidence from group and individual patient data. J Clin Epidemiol 52:281–291[CrossRef][Medline]
  68. Ioannidis JP, Ntzani EE, Trikalinos TA, Contopoulos-Ioannidis DG 2001 Replication validity of genetic association studies. Nat Genet 29:306–309[CrossRef][Medline]
  69. Ioannidis J, Lau J 2001 Evolution of treatment effects over time: empirical insight from recursive cumulative metaanalyses. Proc Natl Acad Sci USA 98:831–836[Abstract/Free Full Text]
  70. Fischer PR, Thacher TD, Pettifor JM, Jorde LB, Eccleshall TR, Feldman D 2000 Vitamin D receptor polymorphisms and nutritional rickets in Nigerian children. J Bone Miner Res 15:2206–2210[CrossRef][Medline]
  71. Minamitani K, Takahashi Y, Minagawa M, Yasuda T, Niimi H 1998 Difference in height associated with a translation start site polymorphism in the vitamin D receptor gene. Pediatr Res 44:628–632[Medline]
  72. Kroon LP, Bakker FT, van den Bosch GB, Bonants PJ, Flier WG 2004 Phylogenetic analysis of Phytophthora species based on mitochondrial and nuclear DNA sequences. Fungal Genet Biol 41:766–782[CrossRef][Medline]
  73. Templeton AR, Weiss KM, Nickerson DA, Boerwinkle E, Sing CF 2000 Cladistic structure within the human lipoprotein lipase gene and its implications for phenotypic association studies. Genetics 156:1259–1275[Abstract/Free Full Text]
  74. Collins FS, Guyer MS, Charkravarti A 1997 Variations on a theme: cataloging human DNA sequence variation. Science 278:1580–1581[Free Full Text]
  75. d’Alesio A, Garabedian M, Sabatier J, Guaydier-Souquieres G, Marcelli C, Lemacon A, Walrant-Debray O, Jehan F 2005 Two new SNPs in the human vitamin D receptor promoter switch transcription factor binding, change promoter activity and are associated with growth in female adolescents. J Bone Miner Res 20:S184
  76. Arai H, Miyamoto KI, Yoshida M, Yamamoto H, Taketani Y, Morita K, Kubota M, Yoshida S, Ikeda M, Watabe F, Kanemasa Y, Takeda E 2001 The polymorphism in the caudal-related homeodomain protein Cdx-2 binding element in the human vitamin D receptor gene. J Bone Miner Res 16:1256–1264[CrossRef][Medline]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
92/4/1491    most recent
Author Manuscript (PDF)
Right arrow Submit a related Letter to the Editor
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted