Rheumatology Advance Access published July 12, 2005 Rheumatology 2005; 1 of 4
doi:10.1093/rheumatology/kei013
Concise Report
Influence of VEGF gene polymorphisms on the severity of ankylosing spondylitis J. S. Seo, S.-S. Lee1, S. I. Kim2, W. H. Ryu3, K. H. Sa, S. U. Kim, S. W. Han, E. J. Nam, J. Y. Park, W. K. Lee4, S. Y. Kim5 and Y. M. Kang Objectives. To investigate the role of polymorphisms of the vascular endothelial growth factor (VEGF) gene in susceptibility to ankylosing spondylitis (AS), and their relationship to clinical features and radiographic severity. Methods. This study included 157 patients with AS and 140 healthy unrelated controls. Polymorphisms of the VEGF gene were analysed by the polymerase chain reaction (PCR)–restriction fragment length polymorphism assay and amplification refractory mutation system–PCR. Haplotypes were reconstructed using the Bayesian algorithm. Radiographic severity was assessed by the Bath Ankylosing Spondylitis Radiological Index (BASRI). Results. The genotype frequencies of the polymorphisms were in Hardy–Weinberg equilibrium. The distributions of genotypes and alleles did not differ between AS patients and controls. Among the six haplotypes reconstructed based on the tight linkage disequilibrium at positions 2578, 1154 and 634 (pairwise linkage disequilibrium coefficient, r ^ 0.361–0.706), no haplotype was associated with susceptibility to AS. Clinical features were analysed for the four haplotypes (CGC, CGG, AAG, AGG) which were prevalent. In carriers of the AGG haplotype, the frequency of cervical spine involvement was significantly higher (P ^ 0.002, Pcorr ^ 0.036) and that of patients showing a BASRI score `6 was also higher (P ^ 0.025, Pcorr ^ 0.45). Conclusions. This study demonstrates that polymorphisms of the VEGF gene may contribute to disease severity in AS. KEY WORDS: Vascular endothelial growth factor, Ankylosing spondylitis, Polymorphism.
Ankylosing spondylitis (AS) is a chronic inflammatory disorder in which genetic factors play a major role in susceptibility. Although HLA-B27 is important, non-major histocompatibility complex (MHC) genes are estimated to account for at least half of the genetic variability for AS [1]. Case–control studies using candidate non-MHC gene approaches revealed weak associations of several genes, including IL-1Ra, CYP2D6 and TGF- 1 [2–5]. As assessed by the Bath Ankylosing Spondylitis Functional Index (BASFI), disease severity has also been shown to be largely genetically determined [6]. However, only a few candidate gene association studies have been performed for disease severity or clinical features in AS, and these showed weak or negative associations [3, 7, 8]. Angiogenesis is an important process in the pathogenesis of chronic inflammatory disorders such as rheumatoid arthritis [9, 10]. The pathological findings of sacroiliitis and peripheral arthritis in AS demonstrated increased vascularity in the synovial tissues [11–13], indicating the importance of angiogenesis in this disease. Among the proangiogenic factors, VEGF plays a central role in human RA and animal models of arthritis [10, 14]. VEGF concentration has been reported to be increased in spondyloarthropathies including AS [15, 16], and showed significant correlations with the disease activity as indicated by the Bath Ankylosing Spondylitis Disease Activity Index (BASDAI) and the severity of disease as indicated by the Bath Ankylosing Spondylitis Metrology Index (BASMI).
Single nucleotide polymorphisms (SNPs) of the VEGF gene are associated with production of VEGF protein and are reported to be involved in susceptibility to several disorders in which angiogenesis may be critical in pathogenesis [17–24]. In rheumatoid arthritis, the T allele at position 936 of the VEGF gene is significantly more frequent and the carriers of the two susceptible haplotypes are associated with younger age at disease onset [23]. Considering the similar pathological findings of synovitis in AS compared with those of rheumatoid arthritis, it is possible that polymorphisms of the VEGF gene may have a role in susceptibility to and severity of AS. The purpose of the present study was to assess the potential association between polymorphisms in the VEGF gene and susceptibility to AS. We also examined the effect of the polymorphisms on clinical features, including radiographic severity of AS.
Materials and methods Study population One hundred and fifty-seven Korean patients with AS who fulfilled the modified New York criteria [25] were recruited from outpatient clinics of four university hospitals, including Kyungpook
Department of Internal Medicine, School of Medicine, Kyungpook National University, 1Chonnam National University, 2Pusan National University, 3 Chonbuk National University, 4Health Promotion Research Center and 5Department of Orthopedic Surgery, School of Medicine, Kyungpook National University, Republic of Korea. Received 20 December 2004; revised version accepted 27 May 2005. Correspondence to: Y. M. Kang, Department of Internal Medicine, Kyungpook National University Hospital, Samduk 2-Ga, Junggu, Daegu, 700-721, Republic of Korea. E-mail:
[email protected] 1 of 4 ß The Author 2005. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For Permissions, please email:
[email protected]
2 of 4
J. S. Seo et al.
National University Hospital, Chonnam National University Hospital, Pusan National University Hospital and Chonbuk National University Hospital. A control group of 140 unrelated, healthy Korean individuals who had no known medical problems on a health screening questionnaire were enrolled. All individuals gave informed consent for study participation and the study was approved by the Institutional Review Board of each hospital. Clinical and demographic data, including peripheral arthritis, enthesopathies, uveitis, HLA-B27 and age at disease onset, were collected from AS patients. We determined the severity of AS using the Bath Ankylosing Spondylitis Radiological Index (BASRI) [26] in 147 patients whose X-ray films of sacroiliac joints, lumbar spine and cervical spine were available. The BASRI hip score was not included in this study because of the low frequency of hip involvement. AS patients included 132 men and 25 women between 17 and 69 yr of age (mean 29.8 yr). The mean duration of the disease and the age at onset were 6.25 and 23.1 yr respectively. Among AS patients, 147 (93.6%) had HLA-B27, 102 (65.2%) peripheral arthritis, 73 (46.5%) enthesitis and 28 (17.8%) uveitis. The control group included 101 men and 39 women between 19 and 52 yr of age (mean 28.8 yr).
Genotyping of the VEGF gene polymorphisms Genomic DNA was extracted from peripheral blood mononuclear cells (PBMC) by a standard extraction method. The biallelic transition polymorphisms at position 2578 and 1154 in the promoter region, at position 634 in the 50 -untranslated region (UTR) and position 936 in the 30 -UTR were genotyped according to a previously described method [23].
TABLE 1. Distribution of haplotypes at positions 2578, 1154 and 634 of the VEGF gene predicted by Bayesian algorithm 2578/1154/634 AS (%) Controls (%) haplotype (n ¼ 314) (n ¼ 280) P-value CGC CGG AAG AGG CAG AGC
139 94 46 30 5 0
(44.3) (29.9) (14.6) (9.6) (0.8) (0.0)
124 90 36 29 0 1
(44.3) (32.1) (12.9) (10.4) (0.0) (0.4)
1.000 0.594 0.553 0.784 0.063 0.471
Odds ratio (95% CI) 0.999 0.902 1.163 0.914 0.984 1.004
(0.722–1.382) (0.637–1.278) (0.728–1.860) (0.534–1.566) (0.970–0.998) (0.997–1.011)
TABLE 2. Demographic and clinical findings according to the presence of AGG haplotype of the VEGF gene in patients with AS
a
Age at onset (yr) Males:females Duration of disease (yr)a C-spine involvement (%) L-spine involvement (%) BASRI score 46 (%) BASRI score/yra Peripheral arthritis (%) Enthesitis (%) Uveitis (%) Renal involvement (%)
2578/1154/634 AGG () (n ¼ 129)
AGG (þ) (n ¼ 28)
P-value
23.39 0.80 108:21 6.12 0.47 44/120 (36.7) 88/120 (73.3) 37/120 (30.8) 1.66 0.13 82 (63.6) 57 (44.2) 24 (18.6) 33 (25.6)
21.79 1.36 24:4 6.83 1.00 19/27 (70.4) 22/27 (81.5) 15/27 (55.6) 1.76 0.28 20 (71.4) 16 (57.1) 4 (14.3) 5 (17.9)
0.382 1.000 0.522 0.002b 0.467 0.025c 0.734 0.515 0.296 0.787 0.472
a
Values are mean S.E. Pcorr ¼ 0.036, Bonferroni correction. Pcorr ¼ 0.45, Bonferroni correction.
b c
Statistical analysis The differences in genotype distribution and allele frequency among the groups were examined for statistical significance using the 2 test and Fisher’s exact test. Allele and genotype frequencies were tested for Hardy–Weinberg equilibrium using the 2 test. For the comparison of mean values, Student’s t-test and analysis of variance were performed. To calculate linkage disequilibrium (LD) coefficients between polymorphisms of the VEGF gene, we used SAS Genetics software (SAS Institute, Cary, NC, USA). Statistical reconstruction of haplotypes in this case–control study was performed with the Bayesian algorithm using the Phase program [27], which is available at http:// www.stat.washington.edu/stephens/phase.html. The odds ratios and 95% confidence intervals (CIs) were obtained by unconditional logistic regression analysis. When multiple comparisons were involved, corrected P values were calculated for multiple testing using Bonferroni’s method. All analyses were conducted using SPSS, v. 10.0 (SPSS, Chicago, IL, USA).
Results The genotype frequencies of the four polymorphisms of the VEGF gene showed no significant deviation from Hardy– Weinberg expectation. No significant differences in genotype and allele frequencies were observed between the AS patients and controls (data not shown). We compared the clinical and demographic features of the AS patients according to the genotype of respective polymorphism of the VEGF gene, and there were no significant differences (data not shown). The frequency of HLA-B27 was totally independent of the genotypes of the four VEGF polymorphisms. Because linkage disequilibrium has been suggested to be highly structured into conserved blocks of sequence separated by hotspots
of recombination, the final function of a conserved haplotype may be the result of interaction among polymorphisms within the block [28, 29]. We calculated the LD coefficient between polymorphisms which showed that 2578, 1154 and 634 polymorphisms are in tight linkage disequilibrium (pairwise LD coefficient r ¼ 0.361–0.706), and that 936 polymorphism was only weakly associated with the other polymorphisms (r ¼ 0.032–0.141). Based upon the LD coefficients, haplotype inference was restricted to three polymorphisms at positions 2578, 1154 and 634. Of the eight possible haplotypes, six were estimated to be present and four (CGC, CGG, AAG and AGG) accounted for more than 98% of patients and controls (Table 1). Frequencies of the inferred haplotypes were not significantly different between the AS group and controls. We then compared clinical features according to the carriers of each haplotype (Table 2). Of the six haplotypes, four (CGC, CGG, AAG and AGG), which were dominant in both groups, were used for the analysis. The frequency of cervical spinal involvement was significantly higher in the carriers of AGG (P ¼ 0.002, Pcorr ¼ 0.036), who also showed a higher frequency of BASRI score 46 (P ¼ 0.025, Pcorr ¼ 0.45). The BASRI score divided by disease duration was slightly higher in the AGG carriers, but the difference was not statistically significant. Clinical manifestations did not differ according to the presence of CGG, CGC or AAG haplotypes.
Discussion The present data demonstrate that no associations exist between the four functional polymorphisms of the VEGF gene and susceptibility to AS. Although haplotypes at positions 2578, 1154 and 634 were not associated with AS, carriers of the AGG
VEGF gene polymorphisms in AS
from lipopolysaccharide (LPS) stimulated PBMC in GG homozygotes than in AA homozygotes, and suggested that LPS may work through the myeloid zinc finger protein (MZF1) binding site, within which the þ405 (same as 634) polymorphism was predicted to be located. However, the serum concentration was higher in the CC homozygote in healthy individuals [21]. The discrepancy between these results may be partially explained by another study [33]. Among carriers of the 460C/þ450G haplotype, VEGF production was different according to the presence of another polymorphism in between, such as 116G/A. When we further analysed the present data, carriers of the 1154G/634G haplotype showed a higher frequency of cervical spine involvement (31.4 vs 49.0%, P ¼ 0.04), and additional carriage of 2578A was associated with more conspicuous differentiation in radiological severity (36.7 vs 70.4%, P ¼ 0.002). Because the VEGF gene is very highly polymorphic, there is the possibility that if the impact of a single polymorphism is not overwhelming, the net effect of multiple polymorphisms may determine the production of VEGF, thus affecting the clinical manifestations. We are currently evaluating the influence of haplotypes of the VEGF gene polymorphisms on the production of VEGF. Our study indicates that non-MHC region genes may influence the severity of AS. Further studies of the associations between severity indices, such as BASRI, BASFI and BASMI, and genes involved in the inflammatory processes of AS are necessary to clarify how genetic factors determine the severity of the disease.
Key messages
Rheumatology
haplotype showed more severe radiological damage. These results indicate that genetic factors may influence the radiographic severity of AS. Histological studies of sacroiliitis, which is a fundamental clinical feature of AS, showed that synovitis and subchondral bone marrow changes may cause joint destruction [11]. Fibrovascular synovial tissue with infiltration of inflammatory cells invades the joint, suggesting the importance of new vessel formation in the pathogenesis of sacroiliitis. Synovial membranes of peripheral arthritis in AS and other spondyloarthropathies show inflammatory processes similar to those of rheumatoid arthritis [12, 13, 30]. In the peripheral arthritis of AS, the degree of increased vascularity is similar to that in RA. AS is characterized by new bone formation that results in ankyloses of intervertebral and sacroiliac joints. Angiogenesis also plays an important role in bone formation from cartilage, during which new vessels grow into cartilage before osteoblastic activation [31]. These findings suggest that angiogenesis is an essential component in the pathogenesis of AS. VEGF gene polymorphisms and haplotypes, which may have interactions between polymorphic sites, have been reported to differentially regulate the production of VEGF [18, 19, 21, 32–35]. A previous report showed that VEGF polymorphism at position 936 and haplotypes at positions 2578, 1154, 634 and 936 are associated with susceptibility to rheumatoid arthritis [23]. In the present study, we could not find any associations of the four functional polymorphisms or the statistically reconstructed haplotypes with susceptibility to AS. The susceptibility to AS has been estimated to be largely determined by genetic factors in a twin study [36]. Despite the high positivity rate in AS, HLA-B27 may contribute to no more than half of the overall genetic susceptibility. Previous results indicate that genes outside the HLA region must be involved [1, 37]. Genome-wide screenings have also revealed additional susceptibility regions on chromosomes other than the MHC region [38–40]. Although four selected polymorphisms of the VEGF gene, located on chromosome 6p12, were not associated with the susceptibility to AS, there remains the possibility that other polymorphisms on the VEGF gene might be associated with susceptibility to AS. Based on the observation of greater similarity of disease severity in monozygotic compared with dizygotic twins [36], Hamersma et al. [6] demonstrated that in AS disease severity is largely genetically determined and that shared environmental factors play little role in determining the disease severity. Several candidate gene association studies have been performed to assess disease severity and clinical features in AS. Studies of HLA-B27 and other MHC genes, such as HLA-DR and LMP2, showed no association with disease severity indices [41–44], supporting the idea that the genes determining the severity of AS may be encoded outside of the MHC region. IL-10 promoter polymorphism was reported to be weakly associated with BASFI [8], and vitamin D receptor polymorphism showed weak associations with inflammatory indices, such as C-reactive protein and ESR [45]. When AS patients were divided into two groups according to the presence of the AGG haplotype, carriers of the AGG haplotype had a significantly higher frequency of cervical vertebral involvement. The mechanisms that underlie the more extensive spinal involvement in carriers of the AGG haplotype are unknown. Because the VEGF concentration in AS has been reported to be elevated in axial spondyloarthropathy compared with controls [15] and there is a significant correlation with disease severity, indicated by BASMI [16], it is possible that the AGG haplotype is associated with higher production of VEGF in the inflammatory milieu of AS, resulting in more extensive radiographic change. Of the four selected SNPs on the VEGF gene reported to be associated with VEGF synthesis, the 634 polymorphism, which is located at the 50 -UTR, revealed controversial results. Watson et al. [32] showed that there was higher VEGF production
3 of 4
These data support the possibility that non-MHC genes influence the severity of AS. Carriers of AGG haplotype at positions 2578, 1154 and 634 of the VEGF gene had more extensive radiological change.
Acknowledgement This study was supported by a grant of the Korea Health 21 R & D project, Ministry of Health and Welfare, Republic of Korea (01-PJ3-PG6-01GN11-0002). The authors have declared no conflicts of interest.
References 1. Brown MA, Laval SH, Brophy S, Calin A. Recurrence risk modelling of the genetic susceptibility to ankylosing spondylitis. Ann Rheum Dis 2000;59:883–6. 2. McGarry F, Neilly J, Anderson N, Sturrock R, Field M. A polymorphism within the interleukin 1 receptor antagonist (IL-1Ra) gene is associated with ankylosing spondylitis. Rheumatology 2001;40: 1359–64. 3. Maksymowych WP, Reeve JP, Reveille JD et al. High-throughput single-nucleotide polymorphism analysis of the IL1RN locus in patients with ankylosing spondylitis by matrix-assisted laser desorption ionization-time-of-flight mass spectrometry. Arthritis Rheum 2003;48:2011–8. 4. Brown MA, Edwards S, Hoyle E et al. Polymorphisms of the CYP2D6 gene increase susceptibility to ankylosing spondylitis. Hum Mol Genet 2000;9:1563–6.
4 of 4
J. S. Seo et al.
5. Jaakkola E, Crane AM, Laiho K et al. The effect of transforming growth factor beta1 gene polymorphisms in ankylosing spondylitis. Rheumatology 2004;43:32–8. 6. Hamersma J, Cardon LR, Bradbury L et al. Is disease severity in ankylosing spondylitis genetically determined? Arthritis Rheum 2001;44:1396–400. 7. Khan MA, Ball EJ. Genetic aspects of ankylosing spondylitis. Best Pract Res Clin Rheumatol 2002;16:675–90. 8. Goedecke V, Crane AM, Jaakkola E et al. Interleukin 10 polymorphisms in ankylosing spondylitis. Genes Immun 2003;4:74–6. 9. Folkman J. Angiogenesis in cancer, vascular, rheumatoid and other disease. Nat Med 1995;1:27–31. 10. Koch AE. Review: angiogenesis: implications for rheumatoid arthritis. Arthritis Rheum 1998;41:951–62. 11. Francois RJ, Gardner DL, Degrave EJ, Bywaters EG. Histopathologic evidence that sacroiliitis in ankylosing spondylitis is not merely enthesitis. Arthritis Rheum 2000;43:2011–24. 12. Kidd BL, Moore K, Walters MT, Smith JL, Cawley MI. Immunohistological features of synovitis in ankylosing spondylitis: a comparison with rheumatoid arthritis. Ann Rheum Dis 1989; 48:92–8. 13. Revell PA, Mayston V. Histopathology of the synovial membrane of peripheral joints in ankylosing spondylitis. Ann Rheum Dis 1982;41:579–86. 14. Paleolog EM. Angiogenesis in rheumatoid arthritis. Arthritis Res 2002;4(Suppl. 3):S81–90. 15. Drouart M, Saas P, Billot M et al. High serum vascular endothelial growth factor correlates with disease activity of spondylarthropathies. Clin Exp Immunol 2003;132:158–62. 16. Goldberger C, Dulak J, Duftner C, Weidinger F, Falkenbach A, Schirmer M. Vascular endothelial growth factor (VEGF) in ankylosing spondylitis—a pilot study. Wien Med Wochenschr 2002;152:223–5. 17. McCarron SL, Edwards S, Evans PR et al. Influence of cytokine gene polymorphisms on the development of prostate cancer. Cancer Res 2002;62:3369–72. 18. Krippl P, Langsenlehner U, Renner W et al. A common 936 C/T gene polymorphism of vascular endothelial growth factor is associated with decreased breast cancer risk. Int J Cancer 2003;106:468–71. 19. Shahbazi M, Fryer AA, Pravica V et al. Vascular endothelial growth factor gene polymorphisms are associated with acute renal allograft rejection. J Am Soc Nephrol 2002;13:260–4. 20. Lin CC, Wu HC, Tsai FJ, Chen HY, Chen WC. Vascular endothelial growth factor gene-460 C/T polymorphism is a biomarker for prostate cancer. Urology 2003;62:374–7. 21. Awata T, Inoue K, Kurihara S et al. A common polymorphism in the 50 -untranslated region of the VEGF gene is associated with diabetic retinopathy in type 2 diabetes. Diabetes 2002;51:1635–9. 22. Howell WM, Bateman AC, Turner SJ, Collins A, Theaker JM. Influence of vascular endothelial growth factor single nucleotide polymorphisms on tumour development in cutaneous malignant melanoma. Genes Immun 2002;3:229–32. 23. Han SW, Kim GW, Seo JS et al. VEGF gene polymorphisms and susceptibility to rheumatoid arthritis. Rheumatology 2005;43:1173–7. 24. Salvarani C, Boiardi L, Casali B et al. Vascular endothelial growth factor gene polymorphisms in Behcet’s disease. J Rheumatol 2004;31:1785–9. 25. van der Linden S, Valkenburg HA, Cats A. Evaluation of diagnostic criteria for ankylosing spondylitis. A proposal for modification of the New York criteria. Arthritis Rheum 1984;27:361–8. 26. MacKay K, Mack C, Brophy S, Calin A. The Bath Ankylosing Spondylitis Radiology Index (BASRI): a new, validated approach to disease assessment. Arthritis Rheum 1998;41:2263–70.
27. Stephens M, Smith NJ, Donnelly P. A new statistical method for haplotype reconstruction from population data. Am J Hum Genet 2001;68:978–89. 28. Daly MJ, Rioux JD, Schaffner SF, Hudson TJ, Lander ES. Highresolution haplotype structure in the human genome. Nat Genet 2001;29:229–32. 29. Goldstein DB. Islands of linkage disequilibrium. Nat Genet 2001;29:109–11. 30. Ceponis A, Konttinen YT, Imai S et al. Synovial lining, endothelial and inflammatory mononuclear cell proliferation in synovial membranes in psoriatic and reactive arthritis: a comparative quantitative morphometric study. Br J Rheumatol 1998;37:170–8. 31. Gerber HP, Vu TH, Ryan AM, Kowalski J, Werb Z, Ferrara N. VEGF couples hypertrophic cartilage remodeling, ossification and angiogenesis during endochondral bone formation. Nat Med 1999; 5:623–8. 32. Watson CJ, Webb NJ, Bottomley MJ, Brenchley PE. Identification of polymorphisms within the vascular endothelial growth factor (VEGF) gene: correlation with variation in VEGF protein production. Cytokine 2000;12:1232–5. 33. Stevens A, Soden J, Brenchley PE, Ralph S, Ray DW. Haplotype analysis of the polymorphic human vascular endothelial growth factor gene promoter. Cancer Res 2003;63:812–6. 34. Yang B, Cross DF, Ollerenshaw M, Millward BA, Demaine AG. Polymorphisms of the vascular endothelial growth factor and susceptibility to diabetic microvascular complications in patients with type 1 diabetes mellitus. J Diabetes Complications 2003;17:1–6. 35. Renner W, Kotschan S, Hoffmann C, Obermayer-Pietsch B, Pilger E. A common 936 C/T mutation in the gene for vascular endothelial growth factor is associated with vascular endothelial growth factor plasma levels. J Vasc Res 2000;37:443–8. 36. Brown MA, Kennedy LG, MacGregor AJ et al. Susceptibility to ankylosing spondylitis in twins: the role of genes, HLA, and the environment. Arthritis Rheum 1997;40:1823–8. 37. Reveille JD, Ball EJ, Khan MA. HLA-B27 and genetic predisposing factors in spondyloarthropathies. Curr Opin Rheumatol 2001;13: 265–72. 38. Laval SH, Timms A, Edwards S et al. Whole-genome screening in ankylosing spondylitis: evidence of non-MHC genetic-susceptibility loci. Am J Hum Genet 2001;68:918–26. 39. Zhang G, Luo J, Bruckel J et al. Genetic studies in familial ankylosing spondylitis susceptibility. Arthritis Rheum 2004;50:2246–54. 40. Miceli-Richard C, Zouali H, Said-Nahal R et al. Significant linkage to spondyloarthropathy on 9q31-34. Hum Mol Genet 2004; 13:1641–8. 41. Brown MA, Kennedy LG, Darke C et al. The effect of HLA-DR genes on susceptibility to and severity of ankylosing spondylitis. Arthritis Rheum 1998;41:460–5. 42. Khan MA, Kushner I, Braun WE. Comparison of clinical features in HLA-B27 positive and negative patients with ankylosing spondylitis. Arthritis Rheum 1977;20:909–12. 43. Maksymowych WP, Adlam N, Lind D, Russell AS. Polymorphism of the LMP2 gene and disease phenotype in ankylosing spondylitis: no association with disease severity. Clin Rheumatol 1997;16:461–5. 44. Khan MA, Kushner I, Braun WE, Zachary AA, Steinberg AG. HLA– B27 homozygosity in ankylosing spondylitis: relationship to risk and severity. Tissue Antigens 1978;11:434–8. 45. Obermayer-Pietsch BM, Lange U, Tauber G et al. Vitamin D receptor initiation codon polymorphism, bone density and inflammatory activity of patients with ankylosing spondylitis. Osteoporos Int 2003;14:995–1000.
