Association of Increased Frequencies of HLA-DPB1*05∶01 with the Presence of Anti-Ro/SS-A and Anti-La/SS-B Antibodies in Japanese Rheumatoid Arthritis and Systemic Lupus Erythematosus Patients

Introduction Autoantibodies to ribonucleoprotein are associated with a variety of autoimmune diseases, including rheumatoid arthritis (RA). Many studies on associations between human leukocyte antigen (HLA) alleles and RA have been reported, but few have been validated in RA subpopulations with anti-La/SS-B or anti-Ro/SS-A antibodies. Here, we investigated associations of HLA class II alleles with the presence of anti-Ro/SS-A or anti-La/SS-B antibodies in RA. Methods An association study was conducted for HLA-DRB1, DQB1, and DPB1 in Japanese RA and systemic lupus erythematosus (SLE) patients that were positive or negative for anti-Ro/SS-A and/or anti-La/SS-B antibodies. Results An increased prevalence of certain class II alleles was associated with the presence of anti-Ro/SS-A antibodies as follows: DRB1*08∶03 (Pc = 3.79×10−5, odds ratio [OR] 3.06, 95% confidence interval [CI] 1.98–4.73), DQB1*06∶01 (Pc = 0.0106, OR 1.70, 95%CI 1.26–2.31), and DPB1*05∶01 (Pc = 0.0040, OR 1.55, 95%CI 1.23–1.96). On the other hand, DRB1*15∶01 (Pc = 0.0470, OR 3.14, 95%CI 1.63–6.05), DQB1*06∶02 (Pc = 0.0252, OR 3.14, 95%CI 1.63–6.05), and DPB1*05∶01 (Pc = 0.0069, OR 2.27, 95% CI 1.44–3.57) were associated with anti-La/SS-B antibodies. The DPB1*05∶01 allele was associated with anti-Ro/SS-A (Pc = 0.0408, OR 1.69, 95% CI 1.19–2.41) and anti-La/SS-B antibodies (Pc = 2.48×10−5, OR 3.31, 95%CI 2.02–5.43) in SLE patients. Conclusion HLA-DPB1*05∶01 was the only allele associated with the presence of both anti-Ro/SS-A and anti-La/SS-B antibodies in Japanese RA and SLE patients.


Introduction
Rheumatoid arthritis (RA) is a chronic systemic inflammatory disease susceptibility to which is associated with genetic and environmental factors [1,2,3]. Altered frequencies of human leukocyte antigen (HLA) alleles are known to be associated with RA in most ethnic groups studied. Some HLA-DR alleles are reported to be positively associated with RA susceptibility [4]. A conserved amino acid sequence at position 70-74 (QKRAA, RRRAA, or QRRAA) in the HLA-DRb chain is shared between the RA-associated HLA-DR alleles; this was therefore designated the shared epitope (SE) [4].
The presence of autoantibodies to ribonucleoprotein is associated with a variety of autoimmune diseases, including Sjögren's Syndrome (SS), systemic lupus erythematosus (SLE), and RA. Anti-La/SS-B antibodies share many features with anti-Ro/SS-A antibodies, and almost all anti-La/SS-B antibody-positive RA patients also have anti-Ro/SS-A antibodies, whereas about one fifth of anti-Ro/SS-A antibody-positive RA patients also have anti-La/SS-B antibodies. HLA-DR2 (DRB1*15 and *16) and DR3 are strongly associated with anti-Ro/SS-A or anti-La/SS-B antibodies in European primary SS populations [5,6,7,8]. On the other hand, DRB1*08:03 was reported to be associated with anti-La/SS-B antibodies and *15:01 with anti-Ro/SS-A antibodies in primary SS, SLE, and asymptomatic individuals in the Japanese population [9]. However, few studies have focused on the association of anti-La/SS-B and anti-Ro/SS-A antibodies with HLA alleles in RA [10]. Here, we elucidate HLA class II associations with the presence of autoantibody in Japanese RA patients.

Patients and Controls
Nine hundred twenty five RA and 622 SLE patients were recruited at Sagamihara Hospital, Nagasaki Medical Center, Yokohama Minami Kyosai Hospital, Tama Medical Center, Kitasato University, Komagome Hospital, Himeji Medical Center, Morioka Hospital, and Kyushu Medical Center. All patients were native Japanese living in Japan. All patients with RA fulfilled the 1988 American College of Rheumatology Criteria for RA [11] and did not overlap any other collagen diseases. All patients with SLE fulfilled the American College of Rheumatology criteria for SLE [12]. The RA patients with SS also fulfilled the Japanese Ministry of Health Criteria for the diagnosis of SS [13]. This   Genotyping Genotyping of HLA-DRB1, DQB1, and DPB1 was performed by polymerase chain reaction using sequence-specific oligonucleotide probes, WAKFlow HLA typing kits (Wakunaga, Hiroshima, Japan), using a Bio-Plex 200 system (Bio-Rad, Hercules, CA). HLA-DRB1 alleles encoding the SE are as follows: *01:01, *04:01, *04:04, *04:05, *04:10, *10:01, *14:02, and *14:06 [14]. One of each DRB1 and DQB1 locus could not be typed in the present study. These were revealed to be novel HLA alleles, DRB1* 08:36:02 and DQB1*06:51, by sequencing of the isolated alleles [15].

Statistical Analysis
Differences of RA characteristics, allele frequencies, or amino acid residue frequencies were analyzed by Student's t-test or Fisher's exact test using 262 contingency tables. Adjustment for multiple comparisons was performed using the Bonferroni method. Corrected P (Pc) values were calculated by multiplying the P value by the number of alleles or amino acid residues tested.  Table 1. Mean age and percentage of males in the Ro(+)La(2)RA and Ro(+)La(+)RA groups were lower than in the anti-Ro/SS-A-and anti-La/SS-B-negative [Ro(2)La(2)] patients. Percentage of secondary SS in the Ro(+)La(2)RA and Ro(+)La(+)RA was higher than in the Ro(2)La(2)RA. There were no significant differences in terms of disease duration, rheumatoid factor or anti-citrullinated peptide antibody positivity, or Steinbrocker stage.

Characteristics of Anti-Ro/SS-A and/or Anti-La/SS-B Antibody-positive RA and SLE Patients
Characteristics of Ro(+)La (2) and Ro(+)La(+) SLE patients are also given in Table 1. Disease duration in the Ro(+)La(+) groups was shorter than in the Ro(2)La (2). Percentage of secondary SS in the Ro(+)La(+)SLE was higher than in the Ro(2)La(2)SLE. There were no significant differences in terms of mean age,
We also examined whether the positive association of DRB1*15:01, DQB1*06:02, and DPB1*05:01 alleles is secondary to the decrease of SE in the Ro(+)La(+)RA patients. When the patients with SE were excluded from the analysis, these allele frequencies were still higher in the Ro(+)La(+)RA than Ro(2)-La(2)RA, although the effect was not statistically significant ( Table 5).

Effects of Secondary SS on the Association of HLA Class II Alleles
The associations of anti-Ro/SS-A and anti-La/SS-B antibodies with secondary SS were described in Table 1, and the association of secondary SS with HLA class II alleles was investigated. No association was observed between DPB1*05:01 or anti-Ro/La antibody-associated HLA alleles and secondary SS.  Figure 1A). The amino acid residue at position 74 associated with anti-Ro/SS-A is different from the SE at that position; these three amino acid residues (13, 16 and 74) are shared by DRB1*08:02, *08:03, and *08:23. The amino acid position 87 in the DQb chain showed associations with anti-Ro/ SS-A antibodies ( Figure 1B). Finally, amino acid positions 35, 55, and 56 in the DPb chain showed associations with anti-Ro/SS-A   Figure 2C). Thus, association analysis suggested roles for certain defined amino acid residues in DRb and DPb.

Discussion
Several studies have shown that certain HLA-DR alleles are associated with the presence of anti-Ro/SS-A or anti-La/SS-B antibodies in patients with autoimmune diseases. However, few studies have focused on the association of HLA alleles with anti-Ro/SS-A or anti-La/SS-B antibodies in RA. To the best of our knowledge, this is the first report of a positive association of HLA-DPB1*05:01 with anti-Ro/SS-A and anti-La/SS-B antibodies in RA, although a tendency towards a higher frequency of this allele in Japanese patients with anti-Ro/SS-A or anti-La/SS-B antibodies has been reported before [9]. Recent studies have noted associations of DPB1 alleles with several diseases [16,17,18,19,20], but here we report an association of DPB1*05:01 with anti-Ro/ SS-A and anti-La/SS-B antibodies in Japanese RA and SLE patients.
It was reported that RA patients with anti-Ro/SS-A antibodies had a more severe disease course, and that they were less frequently DR4-positive than patients without such antibodies [21,22]. In contrast, the presence of anti-Ro/SS-A antibodies in RA was reported to be positively associated with DR4 in some other studies [10,23]. Here, we found that frequencies of DR4 and the SE were lower in anti-Ro/SS-A-positive RA patients ( Figure 2). Although the implications of this finding are not clear, it might suggest that the role of SE may not be as important in anti-Ro/SS-A-positive RA. Alternatively, the genetic background of anti-Ro/ SS-A-positive and -negative RA may be different, and genetic factors other than SE may play a significant role in the former.
It was reported that anti-Ro/SS-A-positive patients were more frequently DR3-or DR2-positive in the context of other autoimmune diseases like primary SS and SLE in European populations [5]. An association of DRB1*15:01 and anti-Ro/SS-A antibodies has been reported in the Japanese population [9]. Such an association of DR2 with the presence of anti-Ro/SS-A was not confirmed in our study on RA patients, but an association of DR2 with the presence of anti-La/SS-B was observed. Although DRB1*08:03 was reported to be associated with anti-La/SS-B in Japanese [9], we observed here that it was associated with the presence of anti-Ro/SS-A antibodies in our RA patients. These could be explained by differences in the pathogenesis of RA and SLE.
Amino acid residues 13, 16 and 74 of the HLA-DRb chain were found to be associated with the presence of anti-Ro/SS-A  (2) antibodies ( Figure 1). Residues 13 and 74 form the HLA-DR peptide-binding groove [24]. Amino acid residues 84-87 and 96 in the DPb chain were associated with anti-La/SS-B antibodies. Similarly, amino acid residues 85 and 86 form the peptide-binding grooves of HLA-DP molecules. These data suggest the involvement of peptide antigens bound to specific HLA molecules in controlling the production of anti-Ro/SS-A or anti-La/SS-B antibodies. It has been determined that patients with anti-Ro/SS-A antibodies are more prone to develop adverse effects when treated with gold salts and other drugs [25,26,27]. Recent studies have shown that adverse drug reactions are associated with drugspecific HLA class I alleles, for example A*31:01 and B*15:02 with carbamazepine, and A*31:01 with methotrexate [28,29,30]. Furthermore, A*31:01 is in LD with DRB1*08:03 and *15:01 (0.56% and 0.63% of haplotype frequency, respectively, see http://hla.or.jp/haplo/haplonavi.php?type = haplo&lang = en). Frequencies of these class II alleles were higher in anti-Ro/SS-A or anti-La/SS-B antibody-positive RA patients (Tables 2, 3). Large scale association studies for HLA class I and anti-Ro/SS-A or anti-La/SS-B antibodies should be performed. Because of the limited sample size of the present study, the observed significance of the statistical association was modest. The association with HLA-DP needs to be confirmed in future independent studies. Because the allelic distribution of HLA in other ethnic populations is different from that in the Japanese, the role of HLA-DP in anti-Ro/SS-A or anti-La/SS-B antibody production in RA in other populations should be determined. This is the first identification of an association of HLA-DPB1*05:01 with positivity for anti-Ro/SS-A or anti-La/SS-B antibodies in RA. Our findings support the role of HLA-DP, as well as DR, in the pathogenesis of autoantibody production.