Clinical and Experimental Rheumatology 2003; 21: 399-402. PEDIATRIC RHEUMATOLOGY Role of human leukocyte ABSTRACT in polyarticular and systemic disease. antigen DRB1*0307 and Objective. To study the prevalence of Generally, patients with the rheumatoid Human Leukocyte Antigen (HLA) DR factor (RF) positive form are consid- DRB1*0308 in suscepti- alleles in children with juvenile rheu- ered a separate group, with a similar bility to juvenile rheuma- matoid arthritis (JRA). genetic predisposition to rheumat o i d toid arthritis Methods. DNA samples from 64 chil - a rt h ritis seen in adult patients. A l s o dren with oligoarticular and seronega - ex cluded from JRA are cases with K.M. Alsaeid1, M.Z. Haider1, tive polyarticular JRA and 64 controls spondyloarthropathy, psoriatic arthritis A.M. Al-Awadhi2, of the same ethnic back ground we re or other known causes of ch ro n i c B.S. Srivastva1, E.M. Ayoub3 analyzed using PCR-sequence specific arthritis. p ri m e rs (PCR-SSP) method. A n a ly s i s Se ver al previous studies have repo rt e d 1Department of Pediatrics, 2Department took into account the onset subtype, the an association of JRA with differe n t of Medicine, Faculty of Medicine, Kuwait p resence of antinu clear antibodies human leukoc yte antigen (HLA) class I University and 3Department of Pediatrics, ( A NA) and the presence of ch ro n i c and class II alleles, whi c h underlines the University of Florida, Gainesville, Florida, anterior uveitis, a recognised serious clinical and immun o l o gical heterogen e - USA. complication of JRA. ity of the disease. Using serol o gical typ- K.M. Alsaeid, MD, FAAP, Associate Profes- R e s u l t s . A high prevalence of DR3 ing methods, HLA-DR8 and DR2 wer e sor, Consultant Pediatrician; M.Z. Haider, alleles were detected in children with found to confer susceptibility to this PhD, Associate Professor; A.M. Al-Awadhi, oligoarticular JRA compared to con - for m of the disease while DR4 and DR7 MD, FACP, FRCPC, Associate Professor, trols (p < 0.05). DR3 alleles were the an t i g ens wer e found to be prot e c t i ve (8). Consultant Physician and Rheumatologist; B.S. Srivastva, BS, Research technician; commonest also in patients with posi - Using DNA typing methods the oligo- E.M. Ayoub, MD, FAAP, Professor. tive ANA as well as those with chronic ar ticular type was shown to be associat- The research was supported by funds anterior uveitis. The interesting finding ed with DRB1*0801and DRB1*1301 provided by the Kuwait University. in this study is the absence of two DR3 h aplotypes (8, 9). Ploski showed that Please address correspondence and alleles, namely DRB1*0307 and DRB1 DR 3 , 5 or 6 together with DPB1*0201 reprint requests to: Dr. K. Alsaeid, *0308 in the control group while pre - co n fe r red susceptibility in Norwegia n Kuwait University, Faculty of Medicine, sent in significant proportion in chil - JRA patients (10). However , no associa- Pediatrics Department, PO Box 24923, dren with JRA. DRB1*0307 was pre - tion was detected for DRB1*08 allele in Safat 13110, Kuwait. sent in 16% of children with oligoartic - a study from Germa n y (11), nor from a E-mail: [email protected] ular subtype and 15% of those with la r ge study from Greece (12). Donn & Received on August 5, 2002; accepted p o lya rticular JRA. DRB1*0308 wa s Ollier have repo r ted a high preval e n c e in revised form on January 17, 2003. only detected in children with oligoar - of DRB1*0801, *1101 and *1301 alle- © Copyright CLINICAL AND ticular JRA, none of the children with les in early onset oligo a rticular JRA EXPERIMENTAL RHEUMATOLOGY 2003. polyarticular JRA or the controls had (13). Still others have failed to show any this allele. s i g n i ficant association with HLA-DR Key words: Allele, HLA, juvenile C o n cl u s i o n . These findings support alleles in their population (14). Recently rheumatoid arthritis, PCR. earlier observations linking these two we have repo r ted an interesting associa- DR3 alleles, namely 0307 and 0308, to tion with two of the DR3 alleles namely the genetic susceptibility to JRA. DRB1*0307 and DRB1*0308 in Kuw aiti Ar ab chi l d r en with RF negati ve Introduction o l i go a rticular and polya rticular sub- Juvenile rheumatoid arthritis (JRA) is a types of JRA (15). The high preval e n c e h e t e rogeneous ch ronic disease with of the specific DR3 alleles (DRB1* variability in its presentation, course, 0307 and *O308) have accounted in extra-articular manifestations and long- p a rt for the genetic susceptibility to t e rm prognosis (1, 2). The disease is these for ms of JRA in our patients. classified on the basis of presentation In order to further study the signifi- d u ring the fi rst six months from its cance of our finding, we analysed the onset into pauciarticular onset or prevalence of HLA-DR alleles includ- oligoarticular (with less than 5 joints ing the different DR3 alleles in children involved at onset), polyarticular (5 or with JRA from another racial back- m o re joints) and systemic (3-7). Pa - ground.We have included in the pre- tients with JRA, particularly girls with sent study only Caucasian ch i l d re n oligoarticular disease, develop chronic with RF negat ive oligo a rticular and anterior uveitis (4). Apart from uveitis, polyarticular JRA and racially matched ex t ra - a rticular manife s t ations are controls. unusual in oligoarticular but common Subjects and methods

399 PEDIATRIC RHEUMATOLOGY HLA-DRB1 *0307 and *0308 alleles in children with JRA / K.M. Alsaeid et al.

Patients more efficiently used in PCR reaction girls and 3 boys.The frequency of the HLA-DRB1 genotyping was per- than a primer with one or several mis- HLA-DR alleles were analyzed for the formed on 64 white children with JRA matches in the 3’ end. Allele specific polyarticular form alone, oligoarticular who were seen on regular basis in the PCR products we re analy zed by ge l alone and the two subtypes together. rheumatology outpatient clinic of the e l e c t ro p h o resis fo l l owed by staining The allelic frequency in each of the 3 University of Florida with the diagno- with ethidium bro m i d e. All analy s e s groups was compared to the frequency sis of JRA made at least 6 months we re carried out double blind and in a control group consisting of 64 before the initiation of the study. All app ro p ri a te positive and negati ve con- healthy children. Among the patients, p atients are from Caucasian fa m i l i e s tr ols wer e included for each sample run. 34 were positive for antinuclear anti- residing in the region of North Florida. body (53%) and 8 patients were diag- Detailed clinical information was avail- Statistical analysis nosed with ch ronic anterior uve i t i s able on all patients, including gender, The frequencies of DRB1 alleles were (12.5%). age of onset, joints affected, presence c a l c u l ated by a standard method. The frequency of HLA-DRB1 alleles of enthesitis, presence of extra-articular Throughout this report, the term “fre- d e t e rmined by the DNA-based PCR- m a n i fe s t ations including iri d o cy cl i t i s , quency” refers to the phenotypic fre- SSP method is shown in Table I. Figure and A NA and RF test results. Th e quency. The significance of the differ- 1 demonstrates the perc e n t age of p atients we re cl a s s i fied according to ences in allele/phenotypic frequencies patients with DR3 and its alleles in the the American College of Rheumatolo- was evaluated by chi-squared analysis c o n t rols and the diffe rent pat i e n t gy 1987 criteria. Inclusion criteria were with Yates correction. gro u p s , i . e. oligo a rt i c u l a r, p o lya rt i c u- the presence of oligo- or polyarthritis. lar, all JRA, ANA-positive, and those Patients with a positive rheumatoid fac- Results with uveitis. The incidence of all the tor test, enthesitis or history or a family The patient group consisted of 64 chil- tested DR3 alleles were higher in the h i s t o ry of psoriasis or infl a m m at o ry dren,of whom 51 had the oligoarticular group of patients with oligo a rt i c u l a r b owel disease we re ex cl u d e d. Only and 13 the polyarticular variety of JRA; disease compared to the control group p atients with oligo a rticular and all patients tested negative for rheuma- (43% ve rsus 25% re s p e c t ive ly, p < rheumatoid factor negative polyarticu- toid fa c t o r. The oligo a rticular gro u p 0.05). Two of the DR3 alleles, namely lar JRA course we re incl u d e d. A l l consisted of 37 girls and 14 boys while DRB1*0307 and DRB1*0308, we re patients were examined at 4-6 monthly the polyarticular group consisted of 10 not detected in any of the control sub- i n t e rvals by an ophthalmologist fo r anterior uveitis. HLA-DRB1 allele fre- quencies in JRA cases were compared Table I. HLA-DR frequencies in JRA patients and normal controls. with a group of 64 control ch i l d re n with a similar ethnic background and DR-type Controls Oligoarticular Polyarticular Total JRA N = 64 N = 51 N = 13 N = 64 f rom the same ge ographical regi o n . N (%) N (%) N (%) N (%) Verbal informed consent was obtained from all patients included in this study. DR1 11 (17.2) 12 (23) 3 (23) 15 (23) The antinuclear antibody in the serum DR3 16 (25) 22 (43)** 3 (23) 25 (39) was detected by immunofluorescence DRB1*0301 15 (23) 10 (20) 1 (7) 11 (17) as described by Donn; a titer of 1:40 or DRB1*0302 1 (1.5) 0 0 0 DRB1*0307 0 8 (16) 2 (15) 10 (16) m o re was considered positive (16). DRB1*0308 0 4 (8) 0 4 (6) R h e u m atoid factor in the serum wa s DR4 9 (14) 8 (16) 3 (23) 11 (17) measured by nephelometry (Beckman DR7 10 (16) 11 (22) 2 (15) 13 (20) A rray Nephelometer) according to DRB1*0701 10 (16) 11 (22) 2 (15) 13 (20) manufacturer’s recommendation. DR8 1 (1.5) 5 (10) 0 5 (8) DR9 1 (1.5) 2 (4) 1 (7) 3 (5) Genotyping DR10 2 (3) 0 1 (7) 1 (1.5) Blood samples we re collected fro m DR11 20 (31) 4 (9) 3 (23) 7 (12) p atients and control ch i l d ren after DR12 2 (3) 4 (8) 0 4 (6) obtaining verbal consent. Total genom- DR13 14 (22) 11 (22) 2 (15) 13 (20) ic DNA was isolated by a standard pro- DRB1*1301 14 (22) 11 (22) 2 (15) 13 (20) cedure as described by Sambrook (17). DR14 4 (6) 5 (10) 1 (7) 6 (10) HLA-DR alleles were identified by a DR15 21 (33) 11 (22) 4 (31) 15 (23) commercial kit (Dynal, Oslo, Norway) DRB1*1501 21 (33) 11 (22) 4 (31) 15 (23) which utilizes a DNA-based PCR-SSP DR16 2 (3) 1 (2) 0 1 (2) method described by Olerup (18). The DRB1*1601 2 (3) 1 (2) 0 1 (2) technique is based on the principle that ** p < 0.05 as compared to controls a completely matched primer will be

400 HLA-DRB1 *0307 and *0308 alleles in children with JRA / K.M. Alsaeid et al. PEDIATRIC RHEUMATOLOGY

allele detected. Statistical analy s i s c o m p a ring the incidence of DRB1* 0307 and DRB1*0308 in the patients and controls was not technically feasi- ble because of the absence of these alleles in the control group; neverthe- less, we believe that this is an important and relevant finding for understanding genetic susceptibility in this subgroup of JRA. Of the remaining DR alleles analyzed, HLA-DR8 (DRB1*0801) was more frequent in the patients with oligoartic- ular JRA than in the control gro u p . This represents a known and consistent association with oligoarticular JRA (8). Interestingly, none of the children with polyarticular JRA carried the DRB1* Fig. 1. The pr evalence (%) of DR3 and its alleles in the controls and patients (total JRA patients, 0801 allele. On the other hand, DRB1* oligoarticular subtype, polyarticular subtype, ANA positive and the presence of uveitis, respectively). 0401 was more common in this subtype of JRA. In the control group DR 15 was found to be more prevalent, but no jects; in contrast DRB1*0307 and ed to the pathogenesis of JRA. Report- allele was found to be pro t e c t ive DRB1*0308 were detected in 10% and ed associations between JRA and HLA against JRA in our present study. 5% of the JRA patients, respectively. alleles are variable, partly due to the DR3 alleles were also the commonest 16% of patients with oligo a rt i c u l a r methods used for analysis and partly to alleles in patient with A NA positive JRA and 15% of ch i l d ren with pol- the ethnic groups studied.The impor- JRA, perhaps adding to the importance yarticular disease had the DRB1*0307 tant finding in this study is the associa- of DR3 in the genetic susceptibilty to a l l e l e. None of the controls nor the tion of DR3 alleles, n a m e ly HLA- JRA as a consistent finding. Despite children with polyarticular JRA had the DRB1*0307 and DRB1*0308, with the the relatively small number of patients DRB1*0308 allele, while 8% of those oligoarticular subtype of JRA. This is with uveitis, DR3 alleles were also the with the oligoarticular subtype tested not a novel fi n d i n g, as it has been commonest detected.This is unlike the positive for this allele. DRB1*08 was described in some earlier reports. Plos- previously reported association mainly higher in patients with oligo a rt i c u l a r ki reported an association of DR3 with with DRB1*1104 in children with this JRA but did not re a ch a signifi c a n t DPB*0201 in JRA patients from Nor- serious complication of JRA (19). level, while DRB1*04 was present in way (10). We have also reported in a The difference in the HLA-DR alleles higher fre q u e n cy in the polya rt i c u l a r p revious study a similar associat i o n found to be associated with JRA should group but without statistical signifi- with DR3 alleles in Kuwaiti children not imply a lack of importance when cance. with JRA (15). In that study two DR3 linking inheritance of HLA to suscepti- Of the 8 children with uveitis, 4 had the alleles, DRB1*0307 and *0308, were bility to JRA. In fact, these findings are DR3 alleles (2 were DRB1*0301 and 2 found to account for the increased inci- not unlike those reported for the adult were DRB1*0307), 2 were positive for dence of DR3 in the JRA pat i e n t s . fo rm of rheumatoid art h ritis wh e re DRB1*1101, and 2 were positive for While in this study DR3 was found to more than one allele is associated with DRB1*1301. When patients were ana- be associated with oligoarticular JRA, a single well-defined entity of the dis- lyzed according to the presence of anti- in accordance with previous reports, a ease. Rheumatoid factor positive adult nuclear antibody (ANA), 34 tested pos- significant finding was the incidence of rheumatoid arthritis has been associat- itive (53%). Among all the DR alleles the two DR3 alleles DRB1*0307 and ed with the DRB1*0401, *0 4 0 4 , *0 4 0 5 , tested in this study, DR3 alleles were DRB1*0308. The DRB1*0308 allele *0 4 0 8 , *0 1 0 1 , *0102 and *1001 alleles. the commonest alleles detected in the was only present in the oligoarticular These alleles wer e found to encode a group of patients with a positive ANA group, and was not detected in the con- sp e c i f ic epitope shared among these alle- test. Of all patients with a positive t rol nor the polya rticular group. Th e les. Indeed it was found that the effect of ANA, 11 were DR3 positive (32%), of DRB1*0307 allele was also ab s e n t these shared epitopes confer red the ris k those 6 had DRB1*0301 (18%), 3 had from the control group but accounted in a dose-dependent way (20, 21). DR3 DRB1*0307 (9%) and 2 had DRB1* for 16% and 15% of the oligoarticular alleles have in common with the other 0308 (6%). and the polyarticular subtypes respec- alleles linked to oligoarticular JRA that Discussion t ive ly. In the control group HLA- they too are included in the DR52 super- HLA has long been thought to be relat- DRB1*0301 was the predominant DR3 sp e c i fi c i t y .

401 PEDIATRIC RHEUMATOLOGY HLA-DRB1 *0307 and *0308 alleles in children with JRA / K.M. Alsaeid et al.

P rahalad used a diffe rent analy t i c a l T-cell receptor (TCR) families. A char- erogeneity of pathogenic mechanisms of the method to demonstrated the linkage of a c t e ristic of such ab n o rmal immu n e disease. Hum Immunol 1995; 42: 343-7. 11. PAUL C, SCHOENWALD U, TRUCKENBRODT oligoarticular JRA with HLA-DR and responses is the development of specif- H et al. : HLA-DP/DR interaction in early the alleles most like ly invo l ve d. He ic autoreactivity to self antigens such as onset pauciarticular juvenile chronic arthritis. studied allelic sharing in 80 pairs of h e at shock protein (HSP) and the Immunogenetics 1993; 37: 442-8. a ffected sibl i n g s , and showed that nu clear protein DEK antige n s , 12. P R AT S I D O U - G E RT S I P, K A NA KO U D I - TSAKA-LIDOU F, S P Y ROPOULOU M et al. : affected sibpairs with JRA share HLA- although whether the response to these Nationwide collaborative study of HLA class DR alleles signifi c a n t ly more often a n t i gens is disease-promoting or in- II association with distinct types of juvenile than ex p e c t e d, p roviding add i t i o n a l stead protective is unknown in the set- r h e u m atoid art h ritis in Gre e c e. Eur J evidence for a linkage between the ting of diffe rent fo rms of JRA (24). Immunogenet 1999; 26: 299-310. 13. DONN RP, OLLIER W E R:Ju venile ch ro n i c HLA region and JRA. Excess allelic However, on the basis of our data from arthritis Ð A time for change ? Eur J Immuno - s h a ring was again demonstrated in children with JRA, it would be fair to genet 1996; 23: 245-60. affected sibpairs who were concordant s u ggest that an individual with DR3 14. RUMBA I, DENISOVA A, SOCHNEV A, NILS- SON B, SANJEEV B: HLA class II genes in in their JRA cours e, whether this (DRB1*0307 and *0308 alleles) would L atvian patients with juvenile rheumat o i d course was oligoarticular or polyarticu- have a somewhat higher degree of pre- arthritis. Tissue Antigens 1997; 49: 56-60. lar. HLA-DR8 was shared significantly disposition for the disease, although the 15. ALSAEID K, HAIDER MZ, KAMAL H, S R I- m o re fre q u e n t ly in the affected sib- study must be extended to larger num- VASTVA BS, AYOUB EM: Prevelence of hu- man leukocyte antigen (HLA) DRB1 alleles p a i rs. HLA-DR pairs linked to JRA bers. in Kuwaiti children with juvenile rheumatoid i n cluded DR 8/11 and DR8/3 (22). arthritis. Eur J Immunogenet 2002; 1: 1-5. Both DR8 and DR3 were also increas- Acknowledgements 16. DONN RP, THOMPSON W, PEPPER L et al.: ed in our patients with oligoarticular We would like to thank all the families Antinuclear antibodies in early onset paucia- who participated in this study and Dr. rticular juvenile chronic arthritis (JCA) are JRA. associated with HLA-DQB1*0603: A possi- Ploski found a statistically significant P.N. Sharma for his help with the statis- ble JCA-associated human leukocyte antigen increase of DPB1*0201 among early tical analysis. haplotype. Br J Rheumatol 1995; 34: 461-5. onset oligoarticular JRA patients carry- 17. S A M B ROOK J, FRITIESCH F, M A N I ATIS T: References Molecular Cloning: A Laboratory Manual, ing at least one DRB1 allele encoding 2nd ed.,1989,Cold Spring Harbor Laborato- 1. NEPOM BS:The immunogenetics of juvenile DR3, DR5 or DR6, although in their ry, NY, USA. rheumatoid arthritis. Rheum Dis Clin North 18. OLERUP O, ALDENER A, FOGDELL A: HLA- data none of the DRB1 alleles showed Am 1991;17: 825-42. DQB1 and DQA1 typing by PCR amplifica- a stat i s t i c a l ly significant intera c t i o n 2. FINK CW: Proposal for the development of tion with sequence specific primers (PCR- cl a s s i fi c ation cri t e ria for idiopathic art h ri- with DPB1*0201 when analyzed sepa- SSP) in 2 hours. Tissue Antigens 1993; 41: tides of childhood. J Rheumatol 1995; 22: 119-34. rately (10). Van Kerckhove and Ploski 1566-9. 19. MELIN-ALDANA H, GIANNINI EH, TAYLOR suggested that DR3, 5 or 6 and DPB1* 3. C A S S I DY JT: R h e u m atic diseases of the J et al. : Human leuko cyte antige n - D R B 1 * childhood. In KELLY WN et al. (Eds.): Text - 0201 can potentiate each other’s effects 1104 in the chronic iridocyclitis of pauciar- book of Rheumatology , Philadelphia, Saun- whereas, in contrast, the effect of DR8 ticular juvenile rheumatoid arthritis. J Pedi - ders 1993, 1189-208. atr 1992; 121: 56-60. is independent from any intera c t i o n s 4. PACHMAN LM, P O Z NANSKI A K:Ju ve n i l e 20. GREGERSEN PK,SILVER J, WINCHESTER RJ: (10,23). This, they thought, may sug- r h e u m atoid art h ritis. I n MCCART Y D a n d The shared epitope hypothesis:An approach C O O P M A N W ( E d s . ) : A rt h ritis and A l l i e d gest that a difference exit in the mecha- to understanding the molecular genetics of C o n d i t i o n s, P h i l a d e l p h i a , Lea & Feb i ge r nisms by which DR8 on the one hand, susceptibility to rheumatoid arthritis. Arthri - 1993: 1021-38. tis Rheum 1987; 30: 1205-13. and the DPB1*0201 allele toge t h e r 5 . W H I T E P:Ju venile ch ronic art h ritis. I n K L I P- 21. JAWAHEER D, THOMSON W, MACGREGOR AJ with DR3, 5 or 6 on the other, predis- P E L J H and D I E P P E J H ( E d s . ) : R h e u m at o l o - et al. : “ H o m o z y go u s i t y ” for the HLA-DR gy, P h i l a d e l p h i a ,M o s by, 1 9 9 4 :3 . 1 7 . 1 - 1 0 . pose to early onset oligoarticular JRA. shared epitope contributes to the highest risk 6. C A S S I DY J T, L E V I N S O N J E , BREWER EJ: It has been postulated that for complex for rheumatoid arthritis concordance in iden- The development of classification criteria for tical twins. Arthritis Rheum 1994; 37: 681-6. t raits like JRA the diffe rences in children with juvenile rheumatoid arthritis. 22. P R A H A L A D S, RYAN MH, SHEAR ES, i m mune responses directed towa rd s Bull Rheum Dis 1989; 38: 1-7. THOMPSON SD, GIANNINI EH, GLASS DN: 7. S A K K A S L I , P L ATSOUCAS CD: I m mu n o - self antigens are likely to be central to Ju venile rheumatoid art h ritis. Linkage to pathogenesis of juvenile rheumatoid arthri- the pathogenesis of the disease.This HLA demonstrated by allelic sharing in tis: Role of T cells and MHC. Immunol Res affected sibpairs. Arthritis Rheum 2000; 43: re s p o n s iveness will most like ly be 1995; 14: 218-36. 2335-8. determined by the particular array of 8. F E R NA N D E Z - V I NA M , F I N K C W, STA S T N Y 23. VAN KERCKHOVE C,LUYRINK L, TAYLOR J P: HLA associations in juvenile arthritis. Clin genes present in an individual, i.e spe- et al.:HLA-DQA1*0101 haplotypes and dis- Exp Rheumatol 1994; 12: 205-14. cific HLA alleles may favour the pre- ease outcome in early onset pauciarticular 9. C E R NA M, VAV R I N C OVA P, H AVELKA S, j u venile rheumatoid art h ritis. J Rheumat o l sentation of one peptide antigen over IVASKOVA E, STASTNY P: Class II alleles in 1990; 18: 874-9. juvenile arthritis in Czech children. J Rheu- another to a specific T-cell repertoire 24. M U R R AY K, THOMPSON SD, GLASS DN: matol 1994; 21: 159-64. p re d e t e rmined ge n e t i c a l ly at the Pat h ogenesis of juvenile ch ronic art h ri t i s : 10. PLOSKI R, MCDOWELL TL, SYMONS JA et genetic and environmental factors. Arch Dis thymic level (at least in part by HLA al.: Interaction between HLA-DR and HLA- Child 1997; 77: 530-4. alleles as well) and modified by the DP, and between HLA and interleukin-1a in presence of null allele, in particular the j u venile rheumatoid art h ritis indicates het-

402