Interaction between HLA-B60 and HLA-B27 as a Better Predictor of Ankylosing Spondylitis in a Taiwanese Population

Abstract

Objective

Ankylosing spondylitis (AS) is a form of chronic inflammatory spondyloarthritis (SpA) that causes pain and stiffness in spines or joints. Human leukocyte antigen B27 (HLA-B27) and B60 (HLA-B60) have been reported as major genetic risk factors of AS. In addition, rs13202464, located on major histocompatibility complex (MHC) region, showed high sensitivity (98.7%) and specificity (98.0%) for HLA-B27.

Design

The aim of our study is to test whether the interaction between HLA-B60 and HLA-B27 (rs13202464) can serve as a better predictor of AS. We have genotyped HLA-B60 and rs13202464 among 471 patients with AS and 557 healthy subjects. Combined risk factors were investigated to test the biological interaction.

Results

Our results indicated that the relative risk (RR) for HLA-B27+/HLA-B60− was 152 (95% CI 91 to 255) and it increased to 201 (95% CI 85 to 475) in HLA-B27+/HLA-B60+ patients (with HLA-B27−/HLA-B60− as reference). Combinational analysis of two risk factors (HLA-B27+/HLA-B60+) showed a relative excess risk due to interaction (RERI) of 46.79 (95% CI: -117.58 to 211.16), attributable proportion (AP) of 0.23 (95% CI: -0.41 to 0.88) and a synergy index (S) of 1.31 (95% CI: 0.56 to 3.04).

Conclusion

In conclusion, genetic interaction analysis revealed that the interaction between HLA-B60 and HLA-B27 is a better marker for the risk of AS susceptibility in a Taiwanese population.

About the Authors:

James Cheng-Chung Wei

Contributed equally to this work with: James Cheng-Chung Wei, Henry Wong Sung-Ching

Affiliations Division of Allergy, Immunology and Rheumatology, Department of Medicine, Chung Shan Medical University Hospital, Taichung, Taiwan, Institute of Medicine, Chung Shan Medical University, Taichung, Taiwan, Institute of Intergrative Medicine, China Medical University, Taichung, Taiwan

Henry Wong Sung-Ching

Contributed equally to this work with: James Cheng-Chung Wei, Henry Wong Sung-Ching

Affiliation: Master Program for Clinical Pharmacogenomics and Pharmacoproteomics, School of Pharmacy, Taipei Medical University, Taipei, Taiwan

Yu-Wen Hsu

Affiliations Department of Clinical Pharmacy, School of Pharmacy, Taipei Medical University, Taipei, Taiwan, The Ph.D. Program for Translational Medicine, College of Medical Science and Technology, Taipei Medical University and Academia Sinica, Taipei, Tiawan

Ya-Feng Wen

Affiliation: Department of Clinical Pharmacy, School of Pharmacy, Taipei Medical University, Taipei, Taiwan

Wen-Chang Wang

Affiliations Master Program for Clinical Pharmacogenomics and Pharmacoproteomics, School of Pharmacy, Taipei Medical University, Taipei, Taiwan, The Ph.D. Program for Translational Medicine, Taipei Medical University, Taipei, Taiwan

Ruey-Hong Wong

Affiliations Department of Public Health, Chung Shan Medical University, Taichung, Taiwan, Department of Occupational Medicine, Chung Shan Medical University Hospital, Taichung, Taiwan

Hsing-Fang Lu

Affiliations Department of Clinical Pharmacy, School of Pharmacy, Taipei Medical University, Taipei, Taiwan, Department of Pharmacy, Taipei Medical University-Shuang Ho Hospital, Taipei, Taiwan

Floris A. van Gaalen

Affiliation: Department of Rheumatology, Leiden University Medical Center, Leiden, The Netherlands

Wei-Chiao Chang

* E-mail: [email protected]

Affiliations Master Program for Clinical Pharmacogenomics and Pharmacoproteomics, School of Pharmacy, Taipei Medical University, Taipei, Taiwan, Department of Clinical Pharmacy, School of Pharmacy, Taipei Medical University, Taipei, Taiwan, Department of Pharmacy, Taipei Medical University-Wanfang Hospital, Taipei, Taiwan, Center for Biomarkers and Biotech Drugs, Kaohsiung Medical University, Kaohsiung, Taiwan

Introduction

Ankylosing spondylitis (AS) is a chronic inflammatory disease leading to pain, stiffness and possible fusion of spinal segments. It is considered as a chronic, inflammatory disorder and affects sacroiliac joints, lumbar spine, and peripheral joints [1]. Progression of disease in AS patients often leads to limited of mobility, functional impairment and finally affects the patients’ well-being [2]. Prevalence of AS in men is higher than in women [3], while the pathological mechanisms of AS remain unclear [4].

A genome-wide association study (GWAS) conducted by The Australo-Anglo-American Spondyloarthritis Consortium (TASC) revealed the association of HLA-B27, IL-23 and IL-1 genes to AS [5]. HLA-B27 gene is the best-known genetic susceptibility marker for AS, however, it only explains for 16% of the genetic variability in AS [6,7]. Associations between AS and the HLA-B27 gene and HLA-B60 gene have also been revealed [8]. In addition, Wei et al. showed that HLA-B60 is a risk factor for HLA-B27 negative patients [9]. In 2013, epistasis between HLA-B27 and HLA-B60 has been reported to associate with increased risk of AS in Caucasians, with a very high relative excess risk [10].

rs13202464, located on major histocompatibility complex (MHC) region, showed high sensitivity (98.7%) and specificity (98.0%) to for HLA-B27 [11]. Another GWAS in Han populations also indicated that rs13202464 of HLA-B can represent the risk effects of HLA-B27 in a Chinese population [6]. In this study, we investigated the correlation between HLA-B27 and HLA-B60 and the risk of AS. The association between HLA-B27/HLA-B60 and disease severity of AS was also tested.

Materials and Methods

Subject recruitment

The patients with AS and the healthy subjects were from the Chung Shan Medical University Hospital. All of the participants recruited were ethnic Taiwanese. AS patients who met the New York AS diagnosis criteria were recruited to participate. Our study was approved by the institutional review boards of the Chung Shan Medical University Hospital in Taichung, Taiwan. Informed consent was obtained and be written before any data were collected from the subjects. Our study was approved by the institutional review boards of the Chung Shan Medical University Hospital in Taichung, Taiwan. Informed consent was obtained and be written before any data were collected from the subjects. This study has included the patients with age below 18 years old, with youngest subjects enrolled with age 17. These patients are considered adults by our Ethics Committee, and also the informed consent was obtained and be written before any data were collected from these subjects. The Bath AS Disease Activity Index (BASDAI), Bath AS Functional Index (BASFI), and Bath AS Global (BAS-G) which evaluate disease activity, physical function, and global well-being is collected by questionnaire. Modified Chinese versions of the BASDAI, BASFI, and BAS-G showed good intra-class correlations and Cronbach’s alpha values.

DNA extraction and HLA Genotyping

DNA of blood cells were extracted by first treating them with 0.5% sodium dodecylsulfate lysis buffer and then protease K (1 mg/ml) to digest nuclear proteins for 4 h at 60°C. Total DNA was harvested using a Gentra (Qiagen, Valencia, CA) extraction kit followed by 70% alcohol precipitation. DNA purification from buffy coat was carried out by using the Gentra Puregene Blood Kit (Qiagen, Valencia, CA, USA). rs13202464 (HLA-B27) were genotyped by using the TaqMan® Allelic Discrimination Assay (Applied Biosystems, Foster City, CA). HLA-B60 positivity is identified by two separated SYBR Green real-time PCRs. A 96-well micro-plate with an ABI9700 Thermal Cycler (Applied Biosystems) is used to perform polymerase chain reaction (PCR). After PCR, fluorescence was detected and analyzed by StepOne software vers. 2.2.2 (Applied Biosystems) [12].

Data analysis

Previous studies showed that genotypes of rs13202464 can tag to HLA-B27. Thus, we classified all subjects into HLA-B27 positive (GG and AG genotype) or negative (AA genotype) by rs13202464 genotypes [11]. To examine the interaction effect between HLA-B27 and HLA-B60, samples were categorized into four groups: HLA-B27+/HLA-B60+, HLA-B27+/HLA-B60-, HLA-B27-/HLA-B60+, HLA-B27-/HLA-B60- based on the genotyping results. The odds ratios (ORs) were calculated to assess the additivity of interaction effect, which use the sample of HLA-B27-/HLA-B60- as reference. The prevalence of AS in Taiwan is rare (0.167%) according to the definition of World Health Organization and this satisfies a rare disease assumption thereby replacing RRs with ORs. We used CaTS to calculate the power of association between HLA-B27/B60 to AS susceptibility [13].

Interaction is defined as a departure from additivity of effects. Three indices have been used to evaluate the biological interaction between HLA-B27 and HLA-B60: (1) RERI: the relative excess risk due to interaction. (2) AP: the proportion of disease among those with two risk factors that is attributable to its interaction. (3) S: synergy index. RERI and AP should be zero and S should be 1 when no interaction was detected between two exposures. To obtain the parameter estimates needed for calculating these three measures, a logistic regression model was fitted [14]. The Statistical Package for the Social Sciences (SPSS) V. 20.0 (SPSS, Chicago, Illinois, USA) and R software (http://cran.r-project.org/), were used to analyse the data. Package epiR was used for biological interaction analysis. In all tests, p values less than 0.05 were considered significant.

Results

As shown in Table 1, a total of 1028 subjects were recruited including 471 patients with AS, and 557 healthy subjects. The number of male in AS patients and normal subjects was 320 (67.9%) and 435 (78.0%). The mean of age in both groups were 39.0 years. HLA-B27 and HLA-B60 genotype data were collected from both patients and control subjects. Four hundred and thirty one (91.5%) AS patients and forty three (7.7%) control subjects were categorized as HLA-B27+. Besides, the number of HLA-B60+ was one hundred (21.3%) and seventy three (13.1%), respectively. The ORs of HLA-B27 and HLA-B60 to AS have been 120.80 (95% CI = 83.31 to 204.82) and 1.79 (95% CI = 1.29 to 2.49), respectively (Table 2).

[Figure omitted. See PDF.]

Table 1. Basal characteristics of patients with ankylosing spondylitis (AS) and control subjects.

https://doi.org/10.1371/journal.pone.0137189.t001

[Figure omitted. See PDF.]

Table 2. HLA association with AS in Taiwanese population.

https://doi.org/10.1371/journal.pone.0137189.t002

We further assessed the independent effect and gene-gene interaction effect of HLA-B27 and HLA-B60 to AS susceptibility by categorizing our samples into four strata: HLA-B27+/HLA-B60+, HLA-B27+/HLA-B60-, HLA-B27-/HLA-B60+, HLA-B27-/HLA-B60-. Samples with HLA-B27-/HLA-B60- were considered as reference. As the prevalence of AS is relatively rare in Asian (0.167%), odds ratio was calculated to substitute relative risk in this study. Fig 1 showed that both HLA-B27 and HLA-B60 were disease-susceptibility gene for AS, with odds ratio (OR) 152 (95% confidence interval (CI) 91 to 255, Fisher’s-exact p = 0.0072) and OR 2.9 (95% CI 1.4 to 6.0, Fisher’s exact p = 1.222×10⁻¹⁵⁷) respectively. With HLA-B27−/HLA-B60− as the reference, patient who carried both HLA-B27 and HLA-B60 (HLA-B27+/HLA-B60+) showed a high susceptibility to AS, with the OR increased to 201 (95%CI 85 to 475, Fisher’s exact p = 2.5007×10⁻⁶⁹). Besides, CaTS power calculator revealed an expected power for a one stage study of 1.000 for both genetic risk factors to the AS susceptibility.

[Figure omitted. See PDF.]

Fig 1. Interaction between HLA-B60 and HLA-B27 in ankylosing spondylitis susceptibility in a Taiwanese population with contribution of the different genes to the odds ratio (OR) marked.

¹Odds ratio is calculated by unconditional maximum likelihood estimation and 95% confidence intervals are calculated using normal approximation (Wald method). ²P-value is calculated by Fisher-exact test. AS: ankylosing spondylitis; CI: confidence interval; No: number of individuals; OR: odds ratio.

https://doi.org/10.1371/journal.pone.0137189.g001

To confirm the independent effect of HLA-B27and HLA-B60 to AS susceptibility, logistic regression analysis were performed and the result indicated that two genetic risk factors were independent for the susceptibility of AS (HLA-B27: P-value < 0.001 and HLA-B27 P-value = 0.0148) (Table 3).

[Figure omitted. See PDF.]

Table 3. Logistic regression analysis to identify the independency of two risk factors HLA-B27 and HLA-B60.

https://doi.org/10.1371/journal.pone.0137189.t003

The risk for AS in HLA-B27+/HLA-B60+ exceeded the sum of the risks (201 > (152+2.9)) in HLA-B27−/HLA-B60+ and HLA-B27+/HLA-B60− (Fig 1) but not the product of the risks (201 < (152×2.9)). Calculated biological interaction measures show a departure from additivity of the two risk factors combined (HLA-B27+/HLA-B60+) with a RERI of 46.79 (95% CI: -117.58 to 211.16), AP of 0.23 (95% CI: -0.41 to 0.88) and S of 1.31 (95% CI: 0.56 to 3.04). These results implied the positive gene-gene interaction effects between HLA-B27 and HLA-B60 in a Taiwanese population.

In order to investigate the association between HLA antigens and AS severity, further analysis was conducted. We investigated whether HLA-B27 or HLA-B60 associated with clinical phenotypes including Bath Ankylosing Spondylitis Disease Activity Index (BASDAI), Bath Ankylosing Spondylitis Functional Index (BASFI), and Bath Ankylosing Spondylitis Global Index (BAS-G), which represent disease activity, physical function, and global well-being respectively. Although a deviation form additivity of HLA-B27 and HLA-B60 to AS susceptibility has been detected, neither association between HLA-B27 nor HLA-B60 and AS disease severity was observed (Table 4).

[Figure omitted. See PDF.]

Table 4. Difference in the scores of BASDAI, BASFI, and BAS-G among AS patients stratified by different HLA-B27/ HLA-B60 genotype.

https://doi.org/10.1371/journal.pone.0137189.t004

Discussion

HLA has been known to involve in the antigen recognition process and is a well-known susceptibility factor for the pathogenesis of AS. HLA-B27 is considered as the major susceptibility factor of AS [8,14,15]. HLA-B27 is a highly polymorphic gene, with 105 subtypes: HLA-B*27:01 to HLA-B*27:106 [16,17]. Indeed, comparison of HLA-B27−/HLA-B60− group and HLA-B27+/HLA-B60− group (Fig 1), our results confirmed that HLA-B27 plays an important role in the risk of AS in Taiwanese patients.

To measure biological interaction between HLA-B27 and HLA-B60, we calculated three parameters that measure the departure from additivity of risk effects of each risk factor, i.e. RERI, the relative excess risk due to attributable to HLA-B27 and HLA-B60 interaction; AP[AB], the proportion of AS among those with both exposures that is attributable to HLA-B27 and HLA-B60 interaction; and synergy index, which measures the interaction between two risk factors expressed as the ratio of relative excess risk for the combined effect of the risk factors and the sum of the relative excess risks for each separate effect of the two risk factors. In our study, odds ratio was calculated to estimate the value of risk ratio, and further, calculate the value of RERI, AP and S. Our results showed a departure from additivity of the two risk factors combined (HLA-B27+/HLA-B60+) with a RERI of 46.79 (95% CI: -117.58 to 211.16), AP of 0.23 (95% CI: -0.41 to 0.88) and S of 1.31 (95% CI: 0.56 to 3.04). The RERI, AP and S indices showed a positive additive interaction, indicating that the combination effects of HLA-B27 and HLA-B60.

HLA-B27 and HLA-B60 genes are located on chromosome 6p21.3, both coding proteins involve in antigen presenting functions. HLA presents endogenous antigens to T-cells and further triggers the autoimmune responses. Recent studies suggested that peptide motifs of HLA-B60 may different from HLA-B27 [18–20], thus these HLA-B subtypes might involve in different antigen-triggered pathologic pathways. Therefore, the epistatic effects between HLA-B27 and HLA-B60 may be due to the similar downstream T-cell mediated immune responses. Kirsten Falk et al. (1995) proposed that the peptide binding motif of HLA-B60 is different from HLA-B27 [20], thus it is unlikely that HLA-B27 and HLA-B60 can bind with the same AS pathogenic peptides and trigger disease onset⁹. However, López D et al. (1994) showed the cross-reactions of T cell epitope from CTL clones between HLA-B27 with HLA-B60/61 [21]. Thus, similar T-cell epitopes may be a key factor to explain the epistatic effects between HLA-B27 and HLA-B60.

In consistence with the Floris A van Gaalen et al (2012) study, our study confirmed that HLA-B60 can be used as an independent risk factor for AS susceptibility. In addition, HLA-B60 showed a modest positive biological interaction effects with HLA-B27. However, the difference between two studies is that our results showed that the risk ratio attributed by HLA-B27 is greater than that of HLA-B60 suggesting that HLA-B27 remains the most important genetic predictor on AS susceptibility in Taiwanese population. Comparatively, HLA-B60 plays as a minor additive role. Because of the difficulties in early diagnosis of AS in Taiwan, our study revealed a possibility to the implementation of combining HLA-B60 with HLA-B27 screen in future clinical practice.

There are some limitations in this study. First, larger sample size is needed to confirm our finding about the interaction effect between HLA-B27 and HLA-B60. Second, the mechanisms for addressing the epistatic effects between HLA-B27 or HLA-B60 is still unclear. Third, the interaction between genes and environment is not further investigated in this study. Despite the strong correlation between HLA-B27 and HLA-B60 to the AS susceptibility was found, none of significant correlation between HLA-B27, HLA-B60 and clinical manifestations of AS, i.e. BASDAI, BASFI, and BAS-G [2] was observed in this study. Indeed, similar findings were observed in macrophage migration inhibitory factor (MIF) gene polymorphism. MIF is associated with the susceptibility but not severity of polyarthritis [22]. In short, HLA-B27 and HLA-B60 are strong genetic determinant of susceptibility to AS. Therefore, combination of these two HLA antigens can be applied as a better clinical tool to detect the risk of AS.

In summary, we revealed that HLA-B60 is an independent risk factor for AS in Taiwanese population. In addition, a trend of positive interaction effects of HLA-B27 (rs13202464) and HLA-B60 was identified in a Taiwanese population, which is consistence with Caucasians population. Furthermore, we have shown that polymorphism between HLA-B27 and HLA-B60 antigens are associated with susceptibility to, but not severity of AS. As a results, Taiwanese individuals with the HLA-B27+/HLA-B60+ genotype have high risk of developing AS.

Author Contributions

Conceived and designed the experiments: JCCW RHW WCC. Performed the experiments: YFW YWH FAvG HFL. Analyzed the data: HWSC WCW FAvG. Contributed reagents/materials/analysis tools: YFW YWH FAvG WCC JCCW RHW. Wrote the paper: YFW HWSC YWH WCC HFL.

Citation: Wei JC-C, Sung-Ching HW, Hsu Y-W, Wen Y-F, Wang W-C, Wong R-H, et al. (2015) Interaction between HLA-B60 and HLA-B27 as a Better Predictor of Ankylosing Spondylitis in a Taiwanese Population. PLoS ONE 10(10): e0137189. https://doi.org/10.1371/journal.pone.0137189

References

1. Braun J, Sieper J. Ankylosing spondylitis. Lancet 2007 Apr 21; 369 (1379–1390). pmid:17448825

2. Madsen OR, Rytter A, Hansen LB, Suetta C, Egsmose C. Reproducibility of the Bath Ankylosing Spondylitis Indices of disease activity (BASDAI), functional status (BASFI) and overall well-being (BAS-G) in anti-tumour necrosis factor-treated spondyloarthropathy patients. Clin Rheumatol 2010 Aug; 29 (849–854). pmid:20306214

3. Calin A, Brophy S, Blake D. Impact of sex on inheritance of ankylosing spondylitis: a cohort study. Lancet 1999 Nov 13; 354 (1687–1690). pmid:10568571

4. Duftner C, Goldberger C, Falkenbach A, Wurzner R, Falkensammer B, Pfeiffer KP, et al. Prevalence, clinical relevance and characterization of circulating cytotoxic CD4+CD28- T cells in ankylosing spondylitis. Arthritis Res Ther 2003 5 (R292–300). pmid:12932293

5. Australo-Anglo-American Spondyloarthritis C, Reveille JD, Sims AM, Danoy P, Evans DM, Leo P, et al. Genome-wide association study of ankylosing spondylitis identifies non-MHC susceptibility loci. Nat Genet 2010 Feb; 42 (123–127). pmid:20062062

6. Lin Z, Bei JX, Shen M, Li Q, Liao Z, Zhang Y, et al. A genome-wide association study in Han Chinese identifies new susceptibility loci for ankylosing spondylitis. Nat Genet 2012 Jan; 44 (73–77).

7. Khan MA, Ball EJ. Genetic aspects of ankylosing spondylitis. Best Pract Res Clin Rheumatol 2002 Sep; 16 (675–690). pmid:12406434

8. Brewerton DA, Hart FD, Nicholls A, Caffrey M, James DC, Sturrock RD. Ankylosing spondylitis and HL-A 27. Lancet 1973 Apr 28; 1 (904–907). pmid:4123836

9. Wei JC, Tsai WC, Lin HS, Tsai CY, Chou CT. HLA-B60 and B61 are strongly associated with ankylosing spondylitis in HLA-B27-negative Taiwan Chinese patients. Rheumatology (Oxford) 2004 Jul; 43 (839–842).

10. van Gaalen FA, Verduijn W, Roelen DL, Bohringer S, Huizinga TW, van der Heijde DM, et al. Epistasis between two HLA antigens defines a subset of individuals at a very high risk for ankylosing spondylitis. Ann Rheum Dis 2013 Jun; 72 (974–978). pmid:22887649

11. Evans DM, Spencer CC, Pointon JJ, Su Z, Harvey D, Kochan G, et al. Interaction between ERAP1 and HLA-B27 in ankylosing spondylitis implicates peptide handling in the mechanism for HLA-B27 in disease susceptibility. Nat Genet 2011 Aug; 43 (761–767). pmid:21743469

12. Schlosstein L, Terasaki PI, Bluestone R, Pearson CM. High association of an HL-A antigen, W27, with ankylosing spondylitis. N Engl J Med 1973 Apr 5; 288 (704–706). pmid:4688372

13. Skol AD, Scott LJ, Abecasis GR, Boehnke M. Joint analysis is more efficient than replication-based analysis for two-stage genome-wide association studies. Nat Genet 2006 Feb; 38 (209–213). pmid:16415888

14. Andersson T, Alfredsson L, Kallberg H, Zdravkovic S, Ahlbom A. Calculating measures of biological interaction. Eur J Epidemiol 2005 20 (575–579). pmid:16119429

15. Brown MA. Breakthroughs in genetic studies of ankylosing spondylitis. Rheumatology (Oxford) 2008 Feb; 47 (132–137).

16. Khan MA. Polymorphism of HLA-B27: 105 subtypes currently known. Curr Rheumatol Rep 2013 Oct; 15 (362). pmid:23990399

17. Gran JT, Husby G. Clinical, epidemiologic, and therapeutic aspects of ankylosing spondylitis. Curr Opin Rheumatol 1998 Jul; 10 (292–298). pmid:9725089

18. Weiss EH, Bloemer K, Doerner C, Kuon W, Lang M, Pohla H, et al. Molecular biology of the HLA-B27 locus. Br J Rheumatol 1988 27 Suppl 2 (12–18). pmid:3042071

19. Madden DR, Gorga JC, Strominger JL, Wiley DC. The structure of HLA-B27 reveals nonamer self-peptides bound in an extended conformation. Nature 1991 Sep 26; 353 (321–325). pmid:1922337

20. Falk K, Rotzschke O, Takiguchi M, Gnau V, Stevanovic S, Jung G, et al. Peptide motifs of HLA-B58, B60, B61, and B62 molecules. Immunogenetics 1995 41 (165–168). pmid:7806292

21. Lopez D, Garcia-Hoyo R, Lopez de Castro JA. Clonal analysis of alloreactive T cell responses against the closely related B*2705 and B*2703 subtypes. Implications for HLA-B27 association to spondyloarthropathy. J Immunol 1994 Jun 1; 152 (5557–5571). pmid:8189072

22. Barton A, Lamb R, Symmons D, Silman A, Thomson W, Worthington J, et al. Macrophage migration inhibitory factor (MIF) gene polymorphism is associated with susceptibility to but not severity of inflammatory polyarthritis. Genes Immun 2003 Oct; 4 (487–491). pmid:14551601