Vitamin D Binding Protein Gene in Male Osteoporosis - Springer Link

Calcif Tissue Int (1999) 65:262–266

© 1999 Springer-Verlag New York Inc.

Vitamin D Binding Protein Gene in Male Osteoporosis: Association of Plasma DBP and Bone Mineral Density with (TAAA)n-Alu Polymorphism in DBP S. S. Papiha, L. C. Allcroft, R. M. Kanan, R. M. Francis, H. K. Datta Department of Clinical Biochemistry and Metabolic Medicine, The Medical School, University of Newcastle, Newcastle upon Tyne NE2 4HH, UK

Received: 5 May 1998 / Accepted: 10 April 1999

Abstract. Vitamin D binding protein (DBP) is a major carrier protein for the vitamin D metabolites, but may also play an important role in osteoclast differentiation. Polymorphisms of the DBP gene have been reported, including (TAAA) n-Alu repeat polymorphisms downstream of intron 8. We have examined the relationship between polymorphisms of the DBP gene and bone mineral density (BMD) and vertebral fractures in a group of 26 men with vertebral fractures but no underlying secondary cause of osteoporosis (median age 64, ages 27–72 years) and 21 male control subjects (median age 65, ages 40–77 years). There was no apparent effect of DBP phenotype on BMD, but there was a relationship between certain genotypes of (TAAA) n-Alu repeats and reduced BMD and vertebral fracture. Lumbar spine and femoral neck BMD were significantly lower in men with 10/8 genotype than 10/10 genotype (P < 0.05). Furthermore, the predominant genotype in men with vertebral fractures was 10/8, whereas the most common genotype in control subjects was 10/10 (odds ratio 56; 95% confidence interval 7–445). Plasma DBP was higher in men with 10/8 genotype than those with 10/10 genotype (P < 0.05), and patients with vertebral fractures were found to have higher levels than control subjects (P < 0.0005). Although our study is small because of the relative rarity of idiopathic osteoporosis in men, the results suggest that (TAAA) n-Alu polymorphism may have an important effect on plasma levels of DBP, bone density and fracture risk in men. Key words: Male osteoporosis — Vitamin D binding protein — (TAAA) n-Alu polymorphism — Bone mineral density — Fractures.

of influencing bone density. Another relevant polymorphic gene is the vitamin D binding protein (DBP), also known as group-specific component (GC). DBP is not only the major carrier protein for vitamin D metabolites, but may also play an important role in osteoclast differentiation [7–9]. Three common alleles and over 120 rare variant alleles have been observed, most due to single amino acid substitutions [10]. The molecular differences producing the three common alleles are seen in exon 11. Recently, another polymorphism in the DBP gene has been demonstrated in Alu elements located downstream of intron 8 showing a polymorphic poly (A) tail due to a variable number of tandem repeats (TAAA) n. Of the six possible Alu repeat alleles, only four have been observed in a single study [11]. Both DBP and Alu repeat polymorphism show independent Mendelian inheritance. Although studies have examined the association between traditional DBP gene polymorphism and disease [12], the relationship between AluVpA polymorphism [Alu repeat including a polymorphic variable poly(dA)] in the DBP gene and disease has not been explored previously. We have therefore investigated the effect of DBP isoforms and DBP-Alu repeat polymorphism on bone density in men with symptomatic vertebral fractures and male control subjects from North East England. Although we found no association between DBP isoforms and bone density or propensity to fracture, there was a significant effect of the (TAAA) nAlu genotype on circulating DBP concentration, bone density, and risk of vertebral fracture. Materials and Methods Subjects

Genetic factors are believed to account for much of the variance in peak bone mass and may therefore have a major effect on the development of osteoporosis and risk of fracture [1, 2]. Some of the recent studies suggest that vitamin D receptor (VDR) gene polymorphism may have a significant effect on bone mineral density (BMD) in women [3], but the association between the VDR gene and bone density remains controversial [4, 5]. Furthermore, our recent study failed to demonstrate a significant effect of VDR gene polymorphism on bone density in men [6]. These studies have stimulated interest in other possible candidate genes capable Correspondence to: H. K. Datta

Analysis of DBP isoforms and DBP-Alu repeat genotyping was carried out on 26 men with symptomatic vertebral fractures (median age 64, ages 27–72 years) and 21 male control subjects (median age 65, ages 40–77 years) who had been recruited from North East England as part of a case-control study of vertebral fractures in Caucasian men. Possible secondary causes of osteoporosis were excluded in men with vertebral fractures and control subjects by medical history, physical examination, and laboratory investigations [13]. In all male patients and control subjects, bone mineral density (BMD) of the lumbar spine (L1–L4) and femoral neck was determined by dual energy X-ray absorptiometry (DEXA) using Hologic QDR 2000W. The precision for in vivo measurements of BMD is 1% at the lumbar spine and 1.5% for the femur. Individual BMD results were obtained as an areal density in g/cm2, but have been compared with the normal range for young adult males and age-related normal range to give T- and Z-scores. The T-score is

S. S. Papiha et al.: Vitamin D Binding Protein Gene in Male Osteoporosis

263

Table 1. Age, height, and weight of male control subjects and patients with symptomatic vertebral fractures (mean ± SD)

Age (yr) Weight (kg) Height (cm)

Patients (26)

Controls (21)

59.8 ± 12.1 72.2 ± 9.7 1.70 ± 0.06

62.0 ± 10.3 78.5 ± 12.0 1.72 ± 0.06

Student’s unpaired t-test shows no significant difference between the two groups.

Fig. 1. (A) Partial sequence of intron 8 and exon 9 of the human DBP gene. The Alu sequence region is shown and repeats flanking the Alu sequence are underlined. The forward and reverse primer sequences used for the amplification of the fragment with eight TAAA repeats (GC-I8*8) are shown. (B) Two different genotypes of the polymorphism of Alu in intron 8 of the human DBP: 8/10 genotype is seen in lanes 1 & 2; 9/11 genotype is seen in lane 3, and lane 4 has a molecular weight marker.

the number of standard deviation (SD) units above or below the mean for normal young men, and the Z-score is the number of SD units above or below the mean for normal men of the same age. The protocol of the study was approved by the Newcastle Joint Ethics Committee and all subjects gave their consent.

Isoelectric Focusing of the DBP Phenotypes The vitamin D-binding protein or group-specific component (GC) was analyzed from plasma samples using the standard procedure based on isoelectric focusing migration, as previously described [12]. Genomic DNA was isolated from 10 ml EDTA blood samples using the Nucleon II isolation kit (Scotlab). The precipitated DNA was resuspended in 10 mM TRIS-HCl, 1 mM EDTA, pH 8 to a final concentration of 1 ␮g/␮l. Plasma DBP Radioimmunoassay The total plasma DBP levels were determined in 14 patients with vertebral fractures and 18 control subjects by competitive radioimmunoassay (RIA) using a slight modification of a previously described procedure [14]. In the assay, monospecific polyclonal goat antihuman DBP antibodies were employed (Diasorin, Berkshire, UK).

Polymerase Chain Reaction for (TAAA)n-Alu Polymorphism in Intron 8 of DBP The amplification of intron 8 was carried out using the following primers: two exon-specific oligonucleotides spanning 911–930

Fig. 2. Phenotype distribution of DBP (GC) isoforms in men with vertebral fractures and male control subjects. Chi square test between the observed and expected values for each phenotype was not significant. (䊐) Controls; ( ) patients.

nucleotides (A1) 5⬘-CAGCCATGGACGTTTTTGTG-3⬘) for the 5⬘ end and nucleotides 1069–1088 (A2) (5⬘-TTACTGAGGAATACTTCCGG-3⬘) for the 3⬘ end. A further primer (A3) (5⬘CAGCGAGCCAAGATGGCAC-3⬘) lying in the 3⬘ region of the ALU repeat was used. The total volume of the reaction was 40 ␮l and consisted of 1 ␮g of DNA, 200 ␮mol of each dNTP, 2 U Taq DNA polymerase, 1.5 MgCl2, and 50 ng each of the forward and either one of the reverse primers. The following thermocycling parameters were used to perform 30 amplifications: denaturation at 95°C for 30 seconds, annealing at 55°C for 1 minute, and extension at 72°C for 2 minutes. The detection of the different alleles was carried out using polyacrylamide gel electrophoresis followed by silver staining. Samples (18 ␮l of PCR product and 3 ␮l of loading buffer) were loaded onto nondenaturing polyacrylamide gels (8%). The gels were run at 115 V for 22 hours, fixed in an aqueous Solution1 (ethanol (10%) and acetic acid (0.5%)) for 3 minutes, followed by 15 minutes incubation in Solution2 (silver nitrate solution (0.1%)). The gels were copiously rinsed and a solution of NaOH (1.5%) and formaldehyde (0.15%) was then poured over the gel revealing the bands which were then fixed by 5 minutes of incubation in sodium carbonate solution (0.75%). A representative gel showing two different genotypes is shown, along with the partial nucleotide sequence of intron 8 and exon 9 of DBP containing the Alu repeat region (Fig. 1A, B).

Statistical Analysis Results are presented as mean ± standard error of the mean. Significant differences in anthropometric measurements and BMD were analyzed using Student’s unpaired t-test and analysis of variance. The significance of differences in genotype frequency between control subjects and men with symptomatic vertebral fractures was determined using odds ratio. Odds ratios was calculated using GraphPad InStat version 3 for Windows 95 (GraphPad Soft-

264


Fig. 3. The relationship between spinal (A) and femoral neck (B) BMD expressed as a T score and DBP Alu genotype in men with vertebral fractures and male control subjects. Individual results are shown for 8/10 and 10/10 genotypes, with the horizontal line indicating 1 SEM above and below the mean value. Student’s un-

paired t-test showed that the results were significantly different for femoral neck (P ⳱ 0.037) and spinal BMD (P ⳱ 0.011); open circles are symptomatic patients with vertebral fractures and closed circles are control subjects.

ware, San Diego, California, USA). We calculated that this study has at least an 80% probability of detecting a 10% difference in BMD and 15% difference in circulating DBP levels at P < 0.05.

paired Student’s t-test was P ⳱ 0.011 (Fig. 3). Similarly, femoral neck BMD T-score was lower in subjects with the 10/8 genotype than the 10/10 group (−1.95 ± 0.31 and −1.08 ± 0.23, respectively, P ⳱ 0.037) (Fig. 3). The comparison of 8/10 and 10/10 genotypes for femoral Z-scores showed significantly higher value for the homozygotes (P ⳱ 0.032), though spinal Z scores did not show any significant difference (P ⳱ 0.114). There was no association between Alu genotypes and GC phenotypes. However, plasma DBP levels in men with the 10/8 genotype were found to be higher than in those with 10/10 genotype (96.3 ± 8.9 and 67.6 ± 8.2 mg/100 ml; P ⳱ 0.049). The comparison of plasma DBP levels between men with vertebral fractures and control subjects revealed a more marked difference, with the fracture group having much higher levels (105.3 ± 6.9 versus 62.9 ± 6.3 mg/100 ml; P ⳱ 0.0003) (Fig. 4).

Results

Student’s unpaired t-test showed no statistically significant difference in age, weight, or height between men with symptomatic vertebral fractures and male control subjects (Table 1). The distribution of the GC phenotypes in men with symptomatic vertebral fractures and control subjects is shown as a histogram (Fig. 2). In men with vertebral fractures, phenotype 2/2 is at a much lower frequency compared with control subjects and phenotype 1F1F was absent in the fracture patients. However, despite the observed variations, the phenotypic distribution did not show any significant association with vertebral deformation [patients versus male control subjects (␹2 ⳱ 1.3, df 2, and P > 0.05)]. With respect to the Alu repeat polymorphism studied, frequently observed alleles were GC-I8*8 and GC-I8*10 whereas alleles GC-I8*6, GC-I8*9, and GC-I8*11 were seen less frequently. Out of 26 men with vertebral fractures and 21 control subjects studied for Alu repeats, eight genotypes were detected. The genotype 8/10 was most frequent in men with vertebral fractures (15 patients versus 2 controls), and genotype 10/10 was more common in control subjects (2 patients and 15 controls) (odds ratio 56; 95% confidence interval 7-445). Other genotypes were found at lower frequencies: 6/8 (2 patients only), 6/10 (3 controls only), 8/8 (2 patients only), 9/11 (2 patients only), 10/11 (2 patients and 1 control), and 11/11 (1 patient only). Allele I*8 showed a significant association with symptomatic vertebral fracture compared with allele I*10 (␹2 ⳱ 12.99, df 1, and P < 0.001). The association of Alu genotype with spinal and femoral BMD expressed as a T-score is given in Figure 3. The mean ± SEM spinal T-score for the 10/8 genotype was −2.15 ± 0.54 and for the 10/10 genotype it was −0.15 ± 0.23; the statistical significance of the difference determined by un-

Discussion

The present study shows an association among the number of Alu repeats within the intraintronic region of the DBP gene, circulating DBP concentrations, BMD, and the development of vertebral fracture. In the population studied, the two major genotypes seen were 8/10 and 10/10. The subjects homozygous for allele 10 had a higher BMD than those with 8/10 genotype. Furthermore, most patients with vertebral fracture had the 8/10 genotype, whereas the majority of control subjects had the 10/10 genotype. Other genotypes were also present, however, due to their small number it is difficult to carry out statistical analysis to determine their association with BMD. In order to verify the observed association of GCI*8 alleles with BMD, a larger study comprising a larger group of subjects with a sufficient number of 8/8 homozygotes is needed. Recent important observations have shown that serum DBP plays an important role in macrophage activation [15] and osteoclast differentiation from monocytes, therefore may control bone morphogenesis and remodeling [7–9]. Deglycosylation of DBP, removal of galactose and sialic


Fig. 4. (A) Plasma DBP levels in men with vertebral fractures and male control subjects. The levels were found to be significantly higher in men with vertebral fractures than control subjects (P ⳱ 0.0003). (B) Plasma DBP levels in men with 8/10 and 10/10 genotypes were also found to be significantly different (P ⳱ 0.049). The horizontal lines indicate 1 SEM above and below the mean value; open circles denote symptomatic patients with vertebral fractures and closed circles are control subjects.

265

DBP action in the pathogenesis of osteoporosis should be viewed with caution, since in these studies in vitro DBP and Gc-MAF were found to be potent inhibitors at concentrations that were almost a million-fold lower than the normal plasma circulating levels [16]. Alu repeats may be affecting DBP gene expression by distinct interaction with the transcriptional apparatus; similar variation in transcriptional enhancer activity has recently been demonstrated for HRAS1 minisatellite [19]. The consensus sequence for the evolutionary recent major Alu subfamily contains a functional retinoic acid response element [20]. The interaction between Alu double half sites and bacterially synthesized retinoic acid receptors was demonstrated by electrophoretic mobility shift assay [20]. The sites were shown to function as a retinoic acid response element, increasing transcription of a reporter gene by as much as 35-fold in a transiently transfected cell. The Alu repeats could influence gene expression in a similar manner thus altering the level of DBP expressed by different genotypes. Indeed, it can be argued that minisatellite repeats may alter expression and activity of a variety of other impotant genes. Interestingly, a polymorphic AT-rich minisatellite repeat in 3⬘ flank of the interleukin-6 (IL-6) gene has been recently studied and showed an association between IL-6 gene polymorphism and bone density [21]. There are other possible explanations for the intriguing association between the Alu genotypes of DPB and susceptibility to low BMD and osteoporosis. In addition to binding vitamin D, another important function of DBP is the maintenance of microvascular stability by preventing damage to blood capillaries [22]. The expression of DBP can influence bone growth and repair by its effect on the microvascular stability of the bone, therefore affecting bone quality as well as quantity. It is also possible that observed variation in BMD in different Alu genotypes may in fact be due to linkage disequilibrium with other as yet unknown gene(s) located near the DBP locus on chromosome 4q11-q13. In view of the findings of this preliminary study, further work is now required to explore the relationship among Alu repeat polymorphism, circulating DBP levels, and BMD in a larger number of subjects. Acknowledgments. We are grateful to Rita Khurana and Pamela Trickett for their technical assistance. This work was supported by Newcastle Hospitals Trustees and National Osteoporosis Society. LCA was supported by Action Research. References

acid from the protein-bound trisaccharide leaving N-acetylgalactosamine, produces a potent macrophage activating factor (DBP-MAF, also called Gc-MAF). A defect in the ability of lymphocytes to convert DBP to Gc-MAF has been demonstrated in juvenile osteopetrosis [7]. Therefore, a converse situation may account for the accelerated bone loss seen in individuals with the 10/8 genotype, such that higher circulating DBP levels might lead to greater Gc-MAF generation. A recent study has suggested that DBP and Gc-MAF are potent inhibitors of extracellular calcium sensing in the osteoclast [16]; extracellular calcium sensing by osteoclasts has been shown to be a powerful antiresorptive signal [17, 18]. Therefore, the suppression of osteoclast sensitivity to extracellular calcium by DBP may promote bone resorption. However, the precise contribution of this mechanism of

1. Slemenda CW, Christian JC, Williams CJ, Norton JA, Johnston CC Jr (1991) Genetic determinants of bone mass in adult women: a re-evaluation of the twin model and the potential importance of the gene interaction on heritability estimates. J Bone Miner Res 6:561–567 2. Christian JC, Yu PL, Slemenda CW, Johnston CC Jr (1989) Heritability of bone mass: a longitudinal study in ageing male twins. Am J Hum Genet 44:429–433 3. Morrison NA, Qi JC, Tokita A et al (1994) Prediction of bone mineral density from vitamin D receptor alleles. Nature 367: 284–287 4. Eisman JA (1995) Vitamin D receptor alleles and osteoporosis: an affirmative view. J Bone Miner Res 10:1289–1293 5. Peacock M (1995) Vitamin D receptor alleles and osteoporosis: a contrasting view. J Bone Miner Res 10:1294–1297 6. Francis RM, Harrington F, Turner E, Papiha SS, Datta HK (1997) Vitamin D receptor gene polymorphism in men and its effect on bone density and calcium absorption. Clin Endocrinol 46:83–86

266

7. Yamamoto N, Naraparaju VR, Orchard PJ (1996) Defective lymphocyte glycosidases in the macrophage activation cascade of juvenile osteopetrosis. Blood 88(4):1473–1478 8. Schneider GB, Benis KA, Flay RA, Popoff SN (1996) Effect of vitamin D binding protein-macrophage activating factor (DBP-MAF) infusion on bone resorption in two osteoporotic mutations. Bone 16(6):657–662 9. Benis KA, Schneider GB (1995) The effect of vitamin D binding protein-macrophage activating factor and colonystimulating factor-1 on hemopoietic cells in normal and osteopetrotic rats. Blood 88:2898–2905 10. Cleve H, Constans J (1988) The mutants of vitamin D binding protein: more than 120 variants of the Gc/dDBP system. Vox Sang 54:215–225 11. Braun A, Bichlmaier R, Muller B, Cleve H (1993) Molecular evaluation of an Alu repeat including a polymorphic variable poly(dA) (AluVpA) in the vitamin D protein (DBP) gene. Hum Genet 90:526–532 12. Papiha SS, Pal B (1985) Gc (vitamin D binding protein) subtypes in rheumatoid arthritis. Hum Hered 70:278–280 13. Baillie SP, Davison CE, Johnson FJ, Francis RM (1992) Pathogenesis of vertebral fractures in men. Age Ageing 21: 139–141 14. Haddad JG Jr, Walgate J (1976) Radioimmunoassay for the binding protein for vitamin D and its metabolism in human serum: concentrations in normal subjects and patients with disorders of mineral homeostasis. J Clin Invest 58:1217–1222


15. Yamamoto N, Naraparaju VR (1996) Role of mouse vitamin D3-binding protein in activation of macrophages. J Immunol 157:1744–1751 16. Adebanjo OA, Moonga BS, Haddad JG, Huang L-H, Zaidi M (1997) A possible new role for vitamin D-binding protein in osteoclast control: inhibition of extracellular Ca2+ sensing at low physiological concentrations. Biochem Biophys Res Commun 249:668–671 17. Datta HK, MacIntyre I, Zaidi M (1989) The effect of extracellular calcium elevation on the morphology and function of isolated rat osteoclasts. Biosci Rep 9:747–751 18. Zaidi M, Datta HK, Patchell A, Moonga B, MacIntyre I (1989) ‘Calcium-activated’ intracellular calcium elevation: a novel mechanism of osteoclast regulation. Biochem Biophys Res Commun 163:1461–1465 19. Greene M, Krontiris TG (1993) Allelic variation of reporter gene activation by the HRAS1 minisatellite. Genomics 17: 429–434 20. Vansant G, Reylonds WF (1993) The consensus sequence of a major Alu subfamily contains a functional retinoic acid response element. Proc Natl Acad Sci 92:8229–8233 21. Murray RE, McGuigan F, Grant SFA, Reid DM, Ralston SH (1997) Polymorphisms of the interleukin-6 gene are associated with bone mineral density. Bone 21:89–92 22. Van Baelen H, Allewaert K, Bouillon R (1998) New aspects of the plasma carrier protein for 25-hydroxy-cholecalciferol in vertebrates. Ann NY Acad Sci 538:60–68