You are here

  • Home
  • Archive
  • Volume 70, Issue 12
  • A large-scale association study identified multiple HLA-DRB1 alleles associated with ACPA-negative rheumatoid arthritis in Japanese subjects

Article Text

Article menu

Download PDFPDF

Clinical and epidemiological research
Extended report
A large-scale association study identified multiple HLA-DRB1 alleles associated with ACPA-negative rheumatoid arthritis in Japanese subjects
  1. Chikashi Terao1,2,
  2. Koichiro Ohmura1,
  3. Yuta Kochi3,
  4. Katsunori Ikari4,
  5. Etsuko Maruya5,
  6. Masaki Katayama1,
  7. Kota Shimada6,
  8. Akira Murasawa7,
  9. Shigeru Honjo8,
  10. Kiyoshi Takasugi9,
  11. Keitaro Matsuo10,
  12. Kazuo Tajima10,
  13. Akari Suzuki3,
  14. Kazuhiko Yamamoto11,
  15. Shigeki Momohara4,
  16. Hisashi Yamanaka4,
  17. Ryo Yamada2,
  18. Hiroo Saji5,
  19. Fumihiko Matsuda2,12,13,
  20. Tsuneyo Mimori1
  1. 1Department of Rheumatology and Clinical Immunology, Kyoto University Graduate School of Medicine, Kyoto, Japan
  2. 2Center for Genomic Medicine, Kyoto University Graduate School of Medicine, Kyoto, Japan
  3. 3Laboratory for Autoimmune Diseases, Center for Genomic Medicine, RIKEN, Yokohama, Japan
  4. 4Institute of Rheumatology, Tokyo Women's Medical University, Tokyo, Japan
  5. 5HLA Laboratory, Kyoto, Japan
  6. 6Department of Rheumatology, Sagamihara National Hospital, National Hospital Organization, Sagamihara, Japan
  7. 7Department of Rheumatology, Niigata Rheumatic Center, Niigata, Japan
  8. 8Saiseikai Takaoka Hospital, Toyama, Japan
  9. 9Dohgo Spa Hospital, Matsuyama, Japan
  10. 10Aichi Cancer Center Hospital and Research Institute, Nagoya, Japan
  11. 11Department of Allergy and Rheumatology, Graduate School of Medicine, University of Tokyo, Tokyo, Japan
  12. 12CREST program, Japan Science and Technology Agency, Kawaguchi, Japan
  13. 13Institut National de la Sante et de la Recherche Medicale (INSERM) Unite U852, Kyoto University Graduate School of Medicine, Kyoto, Japan
  1. Correspondence to Koichiro Ohmura, Department of Rheumatology and Clinical Immunology, Graduate School of Medicine, Kyoto University, 54 Shogoin-Kawahara-cho Sakyo-ku, Kyoto 606–8507, Japan; ohmurako{at}kuhp.kyoto-u.ac.jp

Abstract

Background HLA-DRB1 is associated with rheumatoid arthritis (RA). However, it has recently been suggested that HLA-DRB1 is only associated with patients with RA who have anticitrullinated peptide/protein antibodies (ACPA), which are specific to RA.

Objective To elucidate whether specific HLA-DR alleles are associated with ACPA-negative RA development.

Methods HLA-DRB1 typing was carried out in 368 Japanese ACPA-negative patients with RA and 1508 healthy volunteers as the first set, followed by HLA-DRB1 typing of 501 cases and 500 controls as the second set. The HLA-DRB1 allele frequency and diplotype frequency were compared in each group, and the results of the two studies were combined to detect HLA-DRB1 alleles or diplotypes associated with ACPA-negative RA.

Results HLA-DRB1*12:01 was identified as a novel susceptibility allele for ACPA-negative RA (p=0.000088, OR=1.72, 95% CI 1.31 to 2.26). HLA-DRB1*04:05 and *14:03 showed moderate associations with ACPA-negative RA (p=0.0063, OR=1.26, 95% CI 1.07 to 1.49 and p=0.0043, OR=1.81, 95% CI 1.20 to 2.73, respectively). The shared epitope was weakly associated with ACPA-negative RA, but no dosage effect was detected (p=0.016, OR=1.17, 95% CI 1.03 to 1.34). A combination of HLA-DRB1*12:01 and DRB1*09:01 showed a strong association with susceptibility to ACPA-negative RA (p=0.00013, OR=3.62, 95% CI 1.79 to 7.30). Homozygosity for HLA-DR8 was significantly associated with ACPA-negative RA (p=0.0070, OR=2.16, 95% CI 1.22 to 3.82). It was also found that HLA-DRB1*15:02 and *13:02 were protective against ACPA-negative RA (p=0.00010, OR=0.68, 95% CI 0.56 to 0.83 and p=0.00059, OR=0.66, 95% CI 0.52 to 0.84, respectively).

Conclusions In this large-scale association study multiple alleles and diplotypes were found to be associated with susceptibility to, or protection against, ACPA-negative RA.

https://doi.org/10.1136/annrheumdis-2011-200353

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Introduction

    Rheumatoid arthritis (RA) is one of the most common causes of chronic arthritis and results in severe joint damage and a shorter life span.1 Genetic factors have been shown to contribute to the onset of RA.2 Among the genetic susceptibility loci detected to date, HLA-DRB1 has a strong impact on the predisposition to RA and has been repeatedly shown to be associated with RA in an ethnicity-independent manner.3 It is widely accepted that the shared epitope (SE), a common amino acid sequence located from the 70th to the 74th amino acids of the HLA-DR β chain, explains the associations of specific HLA-DRB1 alleles with RA.4 Anticitrullinated protein antibodies (ACPA) are a highly specific marker of RA.5 6 Recent data have shown that the SE is associated with ACPA-positive RA but not associated or only weakly associated with ACPA-negative RA.7,–,9 Many of the non-HLA susceptibility genes for RA detected to date, such as PTPN2210 and CTLA411 have been shown to be associated with ACPA-positive RA alone, and no association between these genes and ACPA-negative RA has been detected. These findings suggest that ACPA-negative RA is genetically distinct from ACPA-positive RA.

    Among HLA-DRB1 molecules, HLA-DR312 and HLA-DR1313 were reported to be associated with ACPA-negative RA in populations of European descent, but the same results were not obtained in a meta-analysis of a large Caucasian cohort.14 In Asian populations, there has only been a small study which showed that HLA-DRB1*09:01 might be associated with ACPA-negative RA,15 while SEs, especially DRB1*04:05, *04:01 and *01:01, were associated with RA and ACPA-positive RA.15 16 Thus, no specific alleles that convey susceptibility to, or are protective against, ACPA-negative RA have been identified in populations of European or Asian descent. In this large-scale Japanese case–control association study, we show that HLA-DRB1*12:01, *14:03 and *04:05 are susceptibility alleles for ACPA-negative RA and that HLA-DRB1*13:02 and *15:02 are protective against ACPA-negative RA. We also identified multiple diplotypes that convey susceptibility to, or are protective against, ACPA-negative RA.

    Materials and methods

    Study subjects

    DNA samples were collected at Kyoto University Hospital from 184 patients with RA who were negative for ACPA, as reported previously,7 and another 184 patients with RA without ACPA were recruited at Tokyo Women's Medical University. These two sample groups were used as the first set. Independent DNA samples were collected from 501 ACPA‑negative patients with RA at RIKEN under the support of BioBank Japan and were used as the second set. The 501 cases in the second set are a fraction of 2410 RA cases included in another manuscript (K Shimane et al, unpublished data). All patients were Japanese and diagnosed by rheumatologists to fulfil the 1987 American College of Rheumatology revised criteria for RA.17 A first set of control DNA samples were collected from 1508 healthy control subjects at Aichi Cancer Center Hospital and from the DNA banks of the Pharma SNP Consortium, which contains DNA samples from healthy Japanese volunteers.18 The second set of control DNA samples were collected from 500 healthy volunteers at the HLA laboratory. This study was approved by the local ethical committees at each institution, and written informed consent was obtained from all patients. Basic information about cases and controls is shown in table 1.

    View this table:
    Table 1

    Basic information for ACPA-negative patients with RA and controls

    ACPA detection

    ACPA were detected with the MESACUP CCP ELISA kit (Medical and Biological Laboratories Co, Ltd, Nagoya, Japan) according to the manufacturer's instructions at each institution. A cut-off value of 4.5 U/ml was used to assess ACPA positivity.

    HLA-DRB1 genotyping

    HLA-DRB1 typing was carried out with the WAKFlow system and described in detail elsewhere.7 In the 184 cases collected at Kyoto University and all the controls in the two sets, genotyping was performed at the HLA laboratory (Kyoto, Japan), whereas it was carried out at RIKEN for all 501 cases in the second set. HLA-DRB1 genotyping of the 184 cases collected at Tokyo Women's Medical University was performed by a sequencing-based typing method using the AlleleSEQR HLA-DRB1 typing kit (Abbott, Tokyo, Japan), and allele assignment was performed using the Assign software.

    The following HLA-DRB1 alleles were classified as belonging to the SE: DRB1*01:01, *01:02, *04:01, *04:04, *04:05, *04:08, *04:10, *04:13, *04:16, *10:01, *13:03, *14:02 and *14:06.

    Statistical analysis

    The frequency of each genotype or diplotype among the ACPA-negative patients with RA was compared with that in the controls using a χ2 test or Fisher's exact test. Ninety-five percent CIs, p values and ORs were also calculated. The relative risk (RR) of ACPA-negative susceptibility induced by homozygosity for each allele was calculated to estimate the dosage effect. We performed 1000 permutation tests to confirm the associations found for each allele. Logistic regression analysis was used to evaluate the effects of alleles by adjusting for the influence of other alleles. Statistical analysis was performed using the R statistic system (http://www.R-project.org) or SPSS (version 18). The power calculation was performed using an online power calculator (http://pngu.mgh.harvard.edu/∼purcell/gpc/).

    Results

    Genotyping of the first set

    We performed HLA-DRB1 genotyping in the 368 ACPA-negative patients with RA and 1508 healthy controls in the first set to compare the allele frequency of each genotype between the cases and controls (table 1). Tables 2 and 3 show the main results of our association study for single alleles and diplotypes, respectively. More detailed results are given in the online supplementary tables 1 and 2.

    View this table:
    Table 2

    Association of the HLA-DRB1 allele with ACPA-negative RA

    The SE showed a weak association with moderate effect (p=0.039), mainly due to HLA-DRB1*04:05. Among the other HLA-DRB1 alleles, HLA-DRB1*14:03, *12:01, and *09:01 resulted in moderate to potential susceptibility to ACPA-negative RA (p=0.022, 0.10, and 0.10, respectively). DRB1*13:02, *04:03, and *15:02 showed moderate to potentially protective effects (p=0.0072, 0.059, and 0.12, respectively).

    Replication in the second set and combined analysis

    We performed HLA-DRB1 genotyping of samples in the second set to replicate the results found in the first set, using the DNA samples from 501 ACPA-negative patients with RA and 500 sex-matched healthy controls and combined the results of the two association studies.

    Among the susceptibility alleles found in the first set, HLA-DRB1*12:01 was confirmed to display a susceptible association (p=0.010 and 0.000088 for the second set and combined study, respectively; table 2). The susceptibility tendencies of *04:05 and *14:03 were replicated in the second set, and these alleles showed moderate associations with susceptibility to ACPA-negative RA in the combined analysis (p=0.0063 and 0.0043, respectively). DRB1*09:01 and *14:05 showed potential susceptibility to ACPA-negative RA in the pooled study (p= 0.062 and 0.080, respectively). The SE showed a weak association with susceptibility to ACPA-negative RA in the combined study (p=0.016), but we could not detect any dosage effect (table 3 and figure 1). Among the protective alleles detected in the first set, the protective effect of DRB1*15:02 was successfully replicated (p=0.002 and 0.00010 in the second set and combined study, respectively; table 2). Although the protective effect of DRB1*13:02 was not replicated in the second set, the combined analysis showed a significant protective effect (p=0.00059). The protective effect of DRB1*04:03 was confirmed in the second set, and the combined study demonstrated a weak protective association (p=0.038). To exclude the possibility that the associations of the susceptibility alleles were induced by the absence of protective alleles or vice versa, we applied logistic regression analysis. The logistic regression analysis suggested that none of the allelic associations—namely, those of HLA-DRB1*12:01, *14:03, *04:05, *13:02, and *15:02, depended on the effects of other alleles (online supplementary table 3). In addition, the permutation tests confirmed the associations of these five alleles (permutation p<0.0070, data not shown).

    Figure 1
    Figure 1

    Suggestive dosage effect of associated alleles on anticitrullinated peptide/protein antibody (ACPA)-negative rheumatoid arthritis susceptibility. The OR for each genotype is shown. Different colours indicate the number of copies of each allele. The numbers of homozygotes of *12:01, *14:05, *15:02, and *13:02 in cases are limited (2, 2, 3 and 3, respectively). Since no patients in this study were homozygous for DRB1*14:03, only the result for *14:05 is shown in this figure. SE in the figure includes DRB1*04:05, which is shown separately.

    View this table:
    Table 3

    Associations between HLA-DRB1 allele diplotypes and ACPA-negative RA

    Next, we analysed the dosage effects of each protective or susceptibility allele. DRB1*12:01 showed a potential dosage effect, but only two patients were homozygous for DRB1*12:01 (figure 1). We could not detect any dosage effects of HLA-DRB1*04:05 or the SE. No patients were homozygous for *14:03 in the cases or controls. Both DRB1*13:02 and *15:02 showed potential dosage effects.

    Diplotype analysis

    When we analysed the effects of HLA-DRB1 allele diplotypes on the predisposition to ACPA-negative RA, we found that a combination of DRB1*09:01 and *12:01 demonstrated susceptible effects in both sets (p=0.025, 0.020 and 0.00013 in the first, second and combined study, respectively; table 3). DRB1*08:03 homozygosity showed a weak susceptible association without any dosage effects (table 3,supplementary table 1). Although we found no susceptibility effect of DRB1*08:02 homozygosity, the combination of DRB1*08:02 and *08:03 also resulted in weak susceptibility (supplementary table 2). When we analysed DR8 allele homozygosity, we found that it displayed a moderate susceptibility association in the combined analysis (p=0.0070, table 3). Any combination of two of the three susceptibility alleles—namely, HLA-DRB1*12:01, *14:03, and *04:05, showed a potentially susceptible effect (supplementary table 2).

    The HLA-DRB1*08:03 and *15:02 diplotype showed the strongest protective effect (p=0.00011, table 3). We found that the diplotypes with protective effects (*08:03/*15:02, *15:02/*15:02 and *13:02/*15:02) all included HLA-DRB1*15:02 (table 3).

    Discussion

    Recent studies have suggested that ACPA-negative RA is a genetically different subset of RA.7 8 While SE is very strongly associated with ACPA-positive RA, it is reported as not associated or only weakly associated with ACPA-negative RA. In populations of European descent, HLA-DR3 and DR13 were reported to be susceptibility alleles,12 13 but a recent meta-analysis of a large Caucasian cohort did not find any such association.14 In Japanese subjects, only DRB1*09:01 was reported to be associated with ACPA-negative RA, using small numbers of patients and controls (28 and 265, respectively).15 16 HLA-DR3 is rare in the Japanese population, and we found only one HLA-DR3 allele in our cohorts.

    Although genetic factors contribute to the development of ACPA-negative RA as much as ACPA-positive RA,19 little is known about the ACPA-negative RA susceptibility alleles of HLA and non-HLA genes.

    Here, we performed a case–control association study using a large number of ACPA-negative patients with RA and controls and showed that multiple alleles and diplotypes are associated with ACPA-negative RA in Japanese people. Although the controls in the first set had different age and sex ratio values from those of the patients and we could not obtain age data for the 500 controls in the second set, the effects of the above-mentioned difference and lack of data on our results were considered to be limited. The HLA locus is located on chromosome 6 and is not affected by sex or age. Indeed, regression analysis did not significantly alter our association results (data not shown).

    Our study showed that HLA-DRB1*12:01 is strongly associated with ACPA-negative RA and that HLA-DRB1*14:03 and HLA-DRB1*04:05 in SE are moderately associated with ACPA-negative RA in Japanese people. All three susceptibility alleles showed susceptibility associations with ACPA-negative RA when found in combination with one of the other two alleles. Our data also suggested a dosage effect of HLA-DRB1*12:01, while no dosage effect of HLA-DRB1*04:05 was detected, with decreased OR of DRB1*04:05 in homozygotes compared with heterozygous patients. In addition, we showed that the HLA-DRB1*09:01 and HLADRB1*12:01 diplotype and HLA-DR8 homozygosity are strong susceptibility combinations for ACPA-negative RA. We also determined HLA-DRB1*13:02 and *15:02 as protective alleles against ACPA-negative RA with a potential dosage effect. The combination of DRB1*08:03 and *15:02 had a strong protective effect in our study. Using logistic regression analysis, we confirmed that the effects of these susceptibility and protective alleles do not depend on each other (supplementary table 3). Although we searched for common amino acid sequences among the susceptibility alleles, we could not detect any meaningful sequences common to HLA-DRB1*12:01, *14:03, and/or *04:05. We also failed to detect a common amino acid sequence among the protective alleles HLA-DRB1*15:02 and *13:02.

    Although the association of SE with ACPA-negative RA cannot be concluded, our large-scale study showed that it is weakly associated with ACPA-negative RA. As we observed a lower OR of the SE in homozygotes than in heterozygous patients, confirmation of this association in other studies are needed. We consider that the SE is associated with ACPA-negative RA but has a much weaker effect than in ACPA-positive RA. Both the relatively small effect of SE on ACPA-negative RA and the small number of cases in previous reports might have resulted in non-significant p values for such tendencies.

    HLA-DRB1*12:01, which was found to be associated with ACPA-negative RA susceptibility in our study, was reported to be associated with type 1 diabetes mellitus (T1D) in Latin America, but no similar association has been reported in Japan.20 21 While a Japanese study showed RA with the anti-glucose-6-phosphate isomerase antibody is associated with HLA-DRB1*12:01,22 no large-scale studies have reported an association between HLA-DRB1*12:01 and RA. As RA shares susceptibility genes with T1D such as PTPN22,23 the determination of HLA-DRB1*12:01 as a potential common risk allele for both T1D and ACPA-negative RA is interesting. Although HLA-DRB1*12:01 showed a possible dosage effect, further confirmation is necessary as only two homozygous patients were among the cases. The allele frequency of HLA-DRB1*12:01 in a European population is 1–4%,24 and so far there are no reports showing an association with ACPA-negative RA.14 HLA-DRB1*12:02, the other allele of HLA-DR12, showed no association with ACPA-negative RA.

    HLA-DRB1*14:03 was reported to be associated with Grave's disease in Japanese patients,25 but its role in RA is unknown. Although our samples did not contain any patients who were homozygous for the allele owing to its low allele frequency, it showed a moderate association with ACPA-negative RA susceptibility. Among the other non-SE DR14 alleles, DRB1*14:05 displayed a tendency towards ACPA-negative RA susceptibility, while *14:01 and *14:07 did not. In total, DR14 alleles, including *14:06 in SE, showed moderate susceptibility effects on ACPA-negative RA (supplementary table 1).

    Although one European study suggested that HLA-DR15 has a protective effect against ACPA-negative RA, its effect on ACPA-negative RA has not been fully examined.13 We showed that HLA-DR15 has strong protective effect against ACPA-negative RA and a possible dosage effect. HLA-DRB1*15:02 is reported to be associated with Japanese T1D in a protective manner.21

    Among HLA-DR13 alleles, HLADRB1*13:02 was reported to be protective against ACPA-positive RA.26 27 Its protective effect was also reported in Japanese patients with RA.16 Its effect on ACPA-negative RA has not been established.13 14 Our study suggested that HLA-DRB1*13:02 has a protective effect against ACPA-negative RA. As the second set in our study did not show any differences in allele frequency between the patients and controls, further validation of our findings is necessary. HLA-DRB1*13:01, a major component of DR13 in populations of European descent, had no effect in our study, where we included DRB1*13:01 in eight alleles in cases and 23 alleles in controls (p=0.59).

    HLA-DR8 has also been reported to be associated with some arthropathic autoimmune diseases, such as juvenile idiopathic arthritis28 and psoriatic arthritis29 in European subjects. The associations indicate that these arthropathies share common pathological mechanisms. Interestingly, the combination of DR8 and DR15 had a strong protective effect against ACPA-negative RA. Considering that DR8 did not show susceptibility association as a single allele, it seems to induce ACPA-negative RA susceptibility in a recessive manner. Among the DR8 alleles, DRB1*08:03 appeared to have a strong effect on ACPA-negative RA susceptibility.

    Although we did not detect a dosage effect of HLA-DRB1*04:03, it showed a potentially protective effect against ACPA-negative RA in the combined study. Further studies are necessary to confirm the association.

    As DRB1*09:01 has been shown to be associated with a decreased ACPA titre in ACPA-positive RA,30 it is likely to be associated with ACPA-negative RA. While DRB1*09:01 showed a potential susceptibility association (p=0.062), the combination of DRB1*09:01 and *12:01 showed strong susceptibility association (p=0.00013). DRB1*09:01 also showed a possible dosage effect. From this viewpoint, we consider that DRB1*09:01 has a potential susceptibility effect on ACPA-negative RA. Owing to the relatively high allele frequency of DRB1*09:01, another independent association study or appropriate classification of ACPA-negative RA could produce significant results.

    In addition to the different associations of the SE with ACPA-negative RA and ACPA-positive RA, we found multiple alleles associated with ACPA-negative RA that are not shared by ACPA-positive RA. These showed that ACPA-negative RA is a distinct subset of RA. Moreover, when we focused on ACPA-negative erosive RA to exclude the possibility of our results being affected by non-RA arthritic diseases, the effects of all the following alleles were maintained: *12:01, *14:03, *04:05, *13:02 and *15:02 (data not shown).

    This is the first large-scale association study involving Japanese ACPA-negative patients with RA and the detection of multiple alleles and diplotypes associated with susceptibility to, or protection against, ACPA-negative RA. To evaluate whether our cohort had sufficient power to detect HLA-DRB1 genotype associations, we applied a risk allele with 5% frequency in the general population (see ‘Materials and methods’). Our power calculation showed that this study had power values of 81% for finding genotype associations with an OR of 1.4 at the 0.05 significance level. When we set the OR to 1.2, our study had power values of 31%. These results suggest that our study has sufficient power to detect associated alleles that are present in relatively high frequencies (such as 5%) and a moderate OR of 1.4. On the contrary, our study has insufficient power to detect associations involving a weak OR such as 1.2. There is a possibility that ACPA-negative RA is associated with more HLA-DRB1 alleles or diplotypes that display a low allele frequency and/or a low OR. Further studies using ACPA-negative RA samples in Japan are necessary to find such associations.

    While association studies using ACPA-negative patients with RA of European descent only found a few weak associations and none of them were subsequently replicated, our study successfully determined multiple alleles with relatively strong effects on ACPA-negative RA. From this viewpoint, we suppose that Japanese ACPA-negative patients with RA have a relatively similar genetic background compared to European patients. Population stratification within European population may also be assumed. Nevertheless, the validation of our results in Asian countries is necessary, and large-scale genome-wide association studies of ACPA-negative RA are also required to elucidate the pathogenesis of ACPA-negative RA.

    Acknowledgments

    The authors thank Dr Naoichiro Yukawa, Dr Hajime Yoshifuji, Dr Daisuke Kawabata, Dr Takaki Nojima, Dr Takashi Usui and Dr Takao Fujii at Kyoto University for collecting the DNA samples. We thank Mr Taishi Shigeki for developing the clinical database software used in Dohgo Spa hospital. We also thank all the doctors and medical staff who collected the patients' samples. This study was performed under the support of Genetics and Allied research in Rheumatic diseases Networking (GARNET) consortium.

    References

    1. ↵
      1. Mok CC,
      2. Kwok CL,
      3. Ho LY,
      4. et al
      . Life expectancy, standardized mortality ratios, and causes of death in six rheumatic diseases in Hong Kong, China. Arthritis Rheum 2011;63:1182–9.
      OpenUrlCrossRefPubMed
    2. ↵
      1. Firestein GS
      . Evolving concepts of rheumatoid arthritis. Nature 2003;423:356–61.
      OpenUrlCrossRefPubMed
    3. ↵
      1. Deighton CM,
      2. Walker DJ,
      3. Griffiths ID,
      4. et al
      . The contribution of HLA to rheumatoid arthritis. Clin Genet 1989;36:178–82.
      OpenUrlPubMedWeb of Science
    4. ↵
      1. Gregersen PK,
      2. Silver J,
      3. Winchester RJ
      . The shared epitope hypothesis. An approach to understanding the molecular genetics of susceptibility to rheumatoid arthritis. Arthritis Rheum 1987;30:1205–13.
      OpenUrlPubMedWeb of Science
    5. ↵
      1. Schellekens GA,
      2. Visser H,
      3. de Jong BA,
      4. et al
      . The diagnostic properties of rheumatoid arthritis antibodies recognizing a cyclic citrullinated peptide. Arthritis Rheum 2000;43:155–63.
      OpenUrlCrossRefPubMedWeb of Science
    6. ↵
      1. van Venrooij WJ,
      2. Hazes JM,
      3. Visser H
      . Anticitrullinated protein/peptide antibody and its role in the diagnosis and prognosis of early rheumatoid arthritis. Neth J Med 2002;60:383–8.
      OpenUrlPubMedWeb of Science
    7. ↵
      1. Ohmura K,
      2. Terao C,
      3. Maruya E,
      4. et al
      . Anti-citrullinated peptide antibody-negative RA is a genetically distinct subset: a definitive study using only bone-erosive ACPA-negative rheumatoid arthritis. Rheumatology (Oxford) 2010;49:2298–304.
      OpenUrlAbstract/FREE Full Text
    8. ↵
      1. Huizinga TW,
      2. Amos CI,
      3. van der Helm-van Mil AH,
      4. et al
      . Refining the complex rheumatoid arthritis phenotype based on specificity of the HLA-DRB1 shared epitope for antibodies to citrullinated proteins. Arthritis Rheum 2005;52:3433–8.
      OpenUrlCrossRefPubMedWeb of Science
    9. ↵
      1. Morgan AW,
      2. Thomson W,
      3. Martin SG,
      4. et al
      . Reevaluation of the interaction between HLA-DRB1 shared epitope alleles, PTPN22, and smoking in determining susceptibility to autoantibody-positive and autoantibody-negative rheumatoid arthritis in a large UK Caucasian population. Arthritis Rheum 2009;60:2565–76.
      OpenUrlCrossRefPubMedWeb of Science
    10. ↵
      1. Kallberg H,
      2. Padyukov L,
      3. Plenge RM,
      4. et al
      . Gene-gene and gene-environment interactions involving HLA-DRB1, PTPN22, and smoking in two subsets of rheumatoid arthritis. Am J Hum Genet 2007;80:867–75.
      OpenUrlCrossRefPubMedWeb of Science
    11. ↵
      1. Plenge RM,
      2. Padyukov L,
      3. Remmers EF,
      4. et al
      . Replication of putative candidate-gene associations with rheumatoid arthritis in >4,000 samples from North America and Sweden: association of susceptibility with PTPN22, CTLA4, and PADI4. Am J Hum Genet 2005;77:1044–60.
      OpenUrlCrossRefPubMedWeb of Science
    12. ↵
      1. Verpoort KN,
      2. van Gaalen FA,
      3. van der Helm-van Mil AH,
      4. et al
      . Association of HLA-DR3 with anti-cyclic citrullinated peptide antibody-negative rheumatoid arthritis. Arthritis Rheum 2005;52:3058–62.
      OpenUrlCrossRefPubMedWeb of Science
    13. ↵
      1. Lundström E,
      2. Källberg H,
      3. Smolnikova M,
      4. et al
      . Opposing effects of HLA-DRB1*13 alleles on the risk of developing anti-citrullinated protein antibody-positive and anti-citrullinated protein antibody-negative rheumatoid arthritis. Arthritis Rheum 2009;60:924–30.
      OpenUrlCrossRefPubMedWeb of Science
    14. ↵
      1. van der Woude D,
      2. Lie BA,
      3. Lundström E,
      4. et al
      . Protection against anti-citrullinated protein antibody-positive rheumatoid arthritis is predominantly associated with HLA‑DRB1*1301: a meta-analysis of HLA-DRB1 associations with anti-citrullinated protein antibody-positive and anti-citrullinated protein antibody-negative rheumatoid arthritis in four European populations. Arthritis Rheum 2010;62:1236–45.
      OpenUrlPubMed
    15. ↵
      1. Furuya T,
      2. Hakoda M,
      3. Ichikawa N,
      4. et al
      . Differential association of HLA-DRB1 alleles in Japanese patients with early rheumatoid arthritis in relationship to autoantibodies to cyclic citrullinated peptide. Clin Exp Rheumatol 2007;25:219–24.
      OpenUrlPubMedWeb of Science
    16. ↵
      1. Wakitani S,
      2. Murata N,
      3. Toda Y,
      4. et al
      . The relationship between HLA-DRB1 alleles and disease subsets of rheumatoid arthritis in Japanese. Br J Rheumatol 1997;36:630–6.
      OpenUrlAbstract/FREE Full Text
    17. ↵
      1. Arnett FC,
      2. Edworthy SM,
      3. Bloch DA,
      4. et al
      . The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum 1988;31:315–24.
      OpenUrlPubMedWeb of Science
    18. ↵
      1. Kamatani N,
      2. Sekine A,
      3. Kitamoto T,
      4. et al
      . Large-scale single-nucleotide polymorphism (SNP) and haplotype analyses, using dense SNP Maps, of 199 drug-related genes in 752 subjects: the analysis of the association between uncommon SNPs within haplotype blocks and the haplotypes constructed with haplotype-tagging SNPs. Am J Hum Genet 2004;75:190–203.
      OpenUrlCrossRefPubMedWeb of Science
    19. ↵
      1. Ding B,
      2. Padyukov L,
      3. Lundström E,
      4. et al
      . Different patterns of associations with anti-citrullinated protein antibody-positive and anti-citrullinated protein antibody-negative rheumatoid arthritis in the extended major histocompatibility complex region. Arthritis Rheum 2009;60:30–8.
      OpenUrlCrossRefPubMedWeb of Science
    20. ↵
      1. Rojas-Villarraga A,
      2. Botello-Corzo D,
      3. Anaya JM
      . HLA-Class II in Latin American patients with type 1 diabetes. Autoimmun Rev 2010;9:666–73.
      OpenUrlCrossRefPubMed
    21. ↵
      1. Katahira M,
      2. Segawa S,
      3. Maeda H,
      4. et al
      . Effect of human leukocyte antigen class II genes on acute-onset and slow-onset type 1 diabetes in the Japanese population. Hum Immunol 2010;71:789–94.
      OpenUrlCrossRefPubMed
    22. ↵
      1. Furuya T,
      2. Matsumoto I,
      3. Tsuchiya N,
      4. et al
      . Anti-glucose-6-phosphate isomerase, anti‑cyclic citrullinated peptide antibodies and HLA-DRB1 genotypes in Japanese patients with early rheumatoid arthritis. Clin Exp Rheumatol 2008;26:918–21.
      OpenUrlPubMed
    23. ↵
      1. Bottini N,
      2. Musumeci L,
      3. Alonso A,
      4. et al
      . A functional variant of lymphoid tyrosine phosphatase is associated with type I diabetes. Nat Genet 2004;36:337–8.
      OpenUrlCrossRefPubMedWeb of Science
    24. ↵
      1. Gonzalez-Galarza FF,
      2. Christmas S,
      3. Middleton D,
      4. et al
      . Allele frequency net: a database and online repository for immune gene frequencies in worldwide populations. Nucleic Acids Res 2011;39(Database issue):D913–9.
      OpenUrlAbstract/FREE Full Text
    25. ↵
      1. Katsuren E,
      2. Awata T,
      3. Matsumoto C,
      4. et al
      . HLA class II alleles in Japanese patients with Graves' disease: weak associations of HLA-DR and -DQ. Endocr J 1994;41:599–603.
      OpenUrlPubMed
    26. ↵
      1. de Vries N,
      2. Tijssen H,
      3. van Riel PL,
      4. et al
      . Reshaping the shared epitope hypothesis: HLA-associated risk for rheumatoid arthritis is encoded by amino acid substitutions at positions 67-74 of the HLA-DRB1 molecule. Arthritis Rheum 2002;46:921–8.
      OpenUrlCrossRefPubMedWeb of Science
    27. ↵
      1. van der Helm-van Mil AH,
      2. Huizinga TW,
      3. Schreuder GM,
      4. et al
      . An independent role of protective HLA class II alleles in rheumatoid arthritis severity and susceptibility. Arthritis Rheum 2005;52:2637–44.
      OpenUrlCrossRefPubMedWeb of Science
    28. ↵
      1. Smerdel A,
      2. Ploski R,
      3. Flatø B,
      4. et al
      . Juvenile idiopathic arthritis (JIA) is primarily associated with HLA-DR8 but not DQ4 on the DR8-DQ4 haplotype. Ann Rheum Dis 2002;61:354–7.
      OpenUrlAbstract/FREE Full Text
    29. ↵
      1. Queiro-Silva R,
      2. Torre-Alonso JC,
      3. Tinturé-Eguren T,
      4. et al
      . The effect of HLA-DR antigens on the susceptibility to, and clinical expression of psoriatic arthritis. Scand J Rheumatol 2004;33:318–22.
      OpenUrlCrossRefPubMed
    30. ↵
      1. Okada Y,
      2. Suzuki A,
      3. Yamada R,
      4. et al
      . HLA-DRB1*0901 lowers anti-cyclic citrullinated peptide antibody levels in Japanese patients with rheumatoid arthritis. Ann Rheum Dis 2010;69:1569–70.
      OpenUrlFREE Full Text

    Footnotes

    • Funding CT is an associate fellow of the global COE program supported by the Ministry of Education, Culture, Sports, Science and Technology (MEXT), Japan. This work was supported by grants-in-aid from the Ministry of Health, Labor and Welfare of Japan and from the Ministry of Education, Culture, Sports, Science and Technology of Japan as well as by research grants from the Japan Rheumatism Foundation, the Waksman Foundation and the Mitsubishi Pharma Research Foundation.

    • Competing interests None.

    • Patient consent Obtained.

    • Ethics approval This study was approved by the local ethical committees at each institution.

    • Provenance and peer review Not commissioned; externally peer reviewed.