HLA Class II Antigens Are Associated with Japanese Pemphigus Patients Hironori Niizeld, Hidetoshi Inoko, Hisashi Narimatsu, Hajime Takata, Akira Sonoda, Takushi Tadakuma, Asako Ando, Kimiyoshi Tsuji, Takashi Hashimoto, and Takeji Nishikawa

ABSTRACT: We investigated the HLA class II antigens in 30 Japanese cases of pemphigus, 17 cases of pemphigus vulgaris (PV) and 13 cases of pemphigus foliaceus (PF), by both serologic and restriction fragment length polymorphism (RFLP) analyses. We detected two major haplotypes susceptible to PV, i.e., DRwl2-DQw7 and DRw6-DQw5. In contrast, DR2 was absent in PV. RFLP analyses showed that DRw6 PV patients had a diseaseassociated restriction fragment representing DQw5, the same association as that found in DRw6 Jewish PV pa-

ABBREVIATIONS BPF Brazilian pemphigus foliaceus HTC homozygous typing cell IC intercellular IHWS International Histocompatibility Workshop

tients. However, DRwI2 Japanese PV patients had DQw7, whereas DR4 Jewish PV patients had DQw8. On the other hand, all 13 PF patients were serologically typed for DQwl, which could not be further subdivided into DQw5 by RFLP analyses. These results suggest that Japanese and Jewish PV patients may be immunogenetically closely related to each other, but Japanese PV patients appear to be immunogenetically different from Japanese PF patients. Human Immunology 3 1 , 2 4 6 - 2 5 0 (199l)

kb PF PV RFLP

kilobase pemphigus foliaceus pemphigus vulgaris restriction fragment length polymorphism

INTRODUCTION Pemphigus, an autoimmune bullous disease of the skin, is characterized clinically by flaccid bullae and weeping erosions, and immunopathologically by the presence of circulating autoantibodies against the intercellular (IC) substances of epidermis [1]. There are two major types of pemphigus, namely pemphigus vulgaris (PV) and From the Departmentsof Dermatology(H.N.: T.H.; T.N.) and Micmbiology ¢H.N.; T.T.), Keio UniversitySchoolof Medicine, the Department of Clinical Laboratories(H.T.; A.S.). Keio UniversityHospital, Shinjukuku. Tokyo, and the Department of Transplantation (H.I.: A.A.; K.T.), Tokai University Schoolof Medicine, BoseidaL Isehara,Japan. The present addressof Hajime Takata is Tissue Antigen Laboratory, Imperial CancerResearchFund, P.O. Box 123, Lincoln'sInn Fields, London WC2A 3PX, United Kingdom. Address reprint requeststo Takeji Nishikawa, Departmentof Dermatology, Keio University Schoolof Medicine, 35 Shinanomachi, Shinjukuku, Tokyo 160,Japan. ReceivedSeptember7, 1990: acceptedJanuary 23, 1991.

246 0198-8859/91/$3.50

pemphigus foliaceus (PF) [1]. The presence of anti-IC autoantibodies is diagnostic for pemphigus, and recent immunoblot studies enabled the subtyping of clinical forms of pemphigus [2]. Like many other autoimmune diseases, PV is associated with certain HLA-DR serotypes. Previous studies revealed an association with DR4 in Jewish [3, 4] and Japanese [5, 6] PV patients, and an association with DR4 and/or DRw6 in non-Ashkenazi Jewish patients [7]. However, restriction fragment length polymorphism (RFLP) analyses demonstrated the association of PV with DQ$3 rather than D Q a or DR3, indicating that D Q ~ may be the suseptibility locus [7]. We, therefore, undertook the serologic typing for HLA-DQ antigens in Japanese PV and PF, as well as D N A typing using RFLP analyses which can subdivide serologically defined D Q types. HumanImmunology3,1,246-250 (1991) © AmericanSocietyfor Histocompadbilit'/andlmmunogenedcs,1991

HLA Class II in Pemphigus Patients

MATERIALS AND METHODS

Patients and controls. Seventeen unrelated PV patients and 13 unrelated PF patients were subjected to serologic typing and were compared with 100 unrelated healthy controls. Diagnosis was based on clinical, histologic, and immunopathologic examinations. Circulating anti-IC autoantibodies were detected in all pemphigus patients. Pemphigus erythematosus, or Senear-Usher's syndrome, was included in PF because of lack of clearcut clinical distinction. Sixteen PV patients and 10 PF patients were subjected to RFLP analyses. Twenty DRmatched healthy individuals and three homozygous typing cells (HTCs) from the tenth International Histocompatibility Workshop (IHWS) were selected as controls for RFLP analyses.

Serologic method. Serologic typing was performed with the National Institutes of Health standard microlymphocytotoxicity test [8]. The HLA antisera were commercially obtained from One-Lambda (Los Angeles, CA). Several alloantisera were also obtained from the 10th IHWS.

cDNA probe. A D Q B I c D N A probe, pDC3101, representing an almost full-length DQB 1 transcript was obtained from the consanguineous HLA-homozygous lymphoblastoid cell line AKIBA (HLA-Aw24, Bw52, DR2, Dwl2, D Q w l , and Cp63) [9]. The probe was labeled by "he random priming method [10] using (c~~zp)_dCl'p.

RFLP analyses. Genomic DNAs were prepared from peripheral leukocytes and Epstein-Barr virus-transformed lymphoblastoid cell lines (the 10th IHWS HTCs) by sodium dodecyl sulfate extraction, followed by proteinase K treatment and phenol extraction, as described earlier [ 11 ]. Ten micrograms of each D N A were digested with BamHl endonuclease (Toyobo Co, Tokyo) at 37°C overnight, and subjected to electrophoresis in 0 . 7 ~ agarose gel, and then hybridized with ~2p-labeled probe, as described earlier [11].

RESULTS

HLA class 11 antigen frequencies in pemphigus patients by serologic method. Thirty unrelated Japanese pemphigus patients were serologically typed for the DR, DRw52/ w53, and DQ antigen specificities. Table 1 shows the antigen frequencies in pemphigus patients compared with those in 100 healthy, unrelated Japanese controls. The HLA antigens with p value of less than 0.05 were

247

TABLE 1 Frequencies of the DR, DRw52/53, and D Q antigens in Japanese pemphigus patients by serologic method Patients (%) HLA antigens

PV (n = 17)

PF (n = 13)

Controls (%) (n ~ 100)

DRI DR2 DR3 DR4 DR5 DRw6 DR7

0.0 0.0 ~ 0.0 47.1 41.2 ~ 64.7' 0.0

30.8 38.5 0.0 61.5 7.7 23.1 7.7

13.0 39.0 0.0 44.0 I7.0 27.0 2.0

DRw8 DR9 DRw10 DRwl I DRwI2 DRblank DRw52 DRw53 OQwl DQw2 DQw3 DQw7

23.5 11.8 0.0 5.9 35.3~ 17.6 94.1' 52.9 64.7 0.0 70.6 41.2

23.1 7.7 0.0 0.0 7.7 0.0 46.2 76.2 100.0~ 7.7 38.5 7.7

17.0 27.0 0.0 3.0 10.0 12.0 56.0 65.0 74.0 2.0 62.0 20.9

• p = -0.00052, corrected p = 0.01573.

~p< 0.05. ,p < 0.005. indicated. Among the HLA-DR antigens, DR2 was absent in PV (p = -0.00052, corrected p = 0.01573). The frequencies of DR5, DRw6, DRwI2, and DRw52 were increased in PV patients compared with those in controls. Furthermore, the frequency of DR4 was slightly increased, which was consistent with the results of the previous reports [5, 6]. Among the HLA-DQ antigens, DQw7 showed a higher frequency in PV, although it was not statistically significant (p = 0.05923). Surprisingly, D Q w l was present in 100e~ of the PF patients (p = 0.02665).

RFLP analyses. An association of HLA with PV in Jewish patients was previously investigated by means of RFLP technique [7, 12]. Based on the results, 13amHl endonuclease for digestion and the D Q B i c D N A for hybridization probe were selected for typing of the DQw5 specificity [13] in the present study. Hybridization of the DQB1 cDNA probe with BamHI digested genomic DNAs from DQw5 HTCs showed the 3.0/6.3-kilobase (kb) fragments (Fig. 1, lanes a and b), which were equivalent to the 3.03/5.76 kb 10th IHWS fragments specific for DQw5 [13]. These fragments were found in all (10 of 10) DRw6 PV

248

H. Niizek;. et al.

r- HTC

t

fa (b-fc-ldJe If

DRw6 PV patients

Ig I " I ~ fj f k [ ~

i

t

[ml

B ~3.0

F I G U R E 1 DQB I RFLP pattern of BamHl-digested DNAs from DRw6 PV patients and the 10th IHWS HTCs. DNAs from HTCs were loaded in lane a (DRI-DQw5; KAS116), lane b (DRwl4-DQwS; EK), and lane c (DRw13-Dw18; WDV). DNAs from PV patients were loaded in lanes d-e (DRw6/DR4), lane f (DRw6/DRwl2), lanes g-h (DRw6/ DRw8), lane i (DRw6/DRblank), lane j (DRw6/DR4), lane k (DRw6/DRw12), lane I (DRw6/DR9), and lane m (DRw6/ DRblank). The DQw5-specific fragments (3.0/6.3 kb) are depicted in lanes a, b, d-m. The 2.89-kb 10th IHWS fragment, DQw6.1-specific fragment [ 12,13], is seen in lane c. Lanes ai and j-m are different filters. Lanes e and j-m are found to have the DQwS-specific fragments in another experiment (data not shown).

patients (Fig. 1, lanes d - m ) and in five of 10 DRw6 controls (Table 2, relative risk = 21.0, p = 0.016254). Because all PF patients had the D Q w l serologic specificity, we selected BamHI for digestion to identify the DQw5 specificity. Seven of 10 PF patients had DQw5specific fragments (Table 2). DISCUSSION In our serologic typing of 30 pemphigus patients, DR2 showed a negative association with PV (p < 0.001),

consistent with the results of the previous reports [% 6]. Furthermore, some new loci were shown to be positively associated with PV. In PV, 41.2% (p = 0.03058) of the patients were positive for DR5, which is in linkage disequilibrium with DQw7. Furthermore, the frequency of D R w l 2 was increased (35.3%, p = 0.01273) and all DRwl2 PV patients were typed for DQw7 (data not shown). Therefore, we concluded that one PV-associated haplotype was DRwl2-DQw7. The relationship of D R w l 2 with PV was stronger than DQw7. DRw6 was present in 64.7% (p = 0.00338) and DRw52 was present in 94.1% (p = 0.00166). Because DRw52 is in linkage disequilibrium with DR5 and DRw6, it is considered that the primary association might be caused by DR5 and DRw6 and the association with DRw52 is due to secondary association. On RFLP analyses, Jewish DRw6 PV patients were reported to have the D Q B I BamHI 2.5-kb fragment with increased frequency [7, 12]. This fragment corresponds to the 3.03-kb 10th IHWS fragment specific for DQew5 [13]. The RFLP analyses showed that this fragment was present in DQW5 HTCs and Jewish DRw6 PV patients [12]. Our analyses also revealed that 10 of 10 DRw6 Japanese PV patients had the same BamH1

HLA Class II in Pemphigus Patients

TABLE 2

2,69

Frequencies of the DQw5-specific fragments ~ in PV and PF patients compared with HLA-matched controls Phenotype

Patients

Controls

Type of pemphigus

DR

DQ

No.

¢,:

No.

f~

Relative risk

p value

PV PF

DRw6

DQwI DQwl

IO/I0 7/tO

100.0 70.0

5/ 10 6/15

'J0.0 40.0

21.0 3.1

0.016254 O.144192

BamHI-digested3.0/6.3-kbDNAfragmentshybridizedwithDQBIcDNAprobe.

3.0-kb DQB1 fragment specific for DQwS. Therefore, it is suggested that the disease-associated RFLP pattern in both Japanese and Jewish PV patients is typed for DQw5. Results from D N A typing of PV patients using RFLP analyses in conjunction with the serotypes indicate two major haplotypes in Japanese PV patients: one is DRw12-DQw7 and the other is DRw6-DQw5 haplotype. These haplotypes possibly contribute to the susceptibility to PV. In the present study, DR2 was absent in PV, which was consistent with the results of the previous reports on a total of 37 Japanese PV patients [5, 6]. The negative association between PV and DR2 alleles leads us to speculate the possible existence of the genetic resistance against PV. These resistant haplotypes may be DR2-DQw6 (Dw2, Dw12) among Japanese, which are the common alleles in the Japanese DR2-associated haplotypes. However, Jewish DR2 patients exist, and the DR2 association with PV is neutral [3]. This indicates that the D Q antigen in Jewish DR2 PV patients may be typed for DQw5, and the DR2DQw5 haplotype may contribute to the susceptibility, but this haplotype is very rare among Japanese population (less than 1%). On the other hand, Scharf et al. [14-16] reported that the DR4 association with Jewish PV can account for the Dwl0-DRB1 (DRBI*0402) allele but not for DQw8 allele (DQBI*0302) (14). In Japanese PV, the frequency of DR4 is slightly increased as shown in the previous [5, 6] and pre~.nt reports. However, we found that the frequency of DRw 12 was statistically increased, which was strongly associated with DQw7. These findings suggested that one of the PV-associated haplotypes was not identical between Japanese and Jews. On looking at PF, all 13 PF patients had the DQw 1 serologic specificity. Seven of 10 PF patients had the DQw5 specificity by RFLP analyses (Table 2; p > 0.05). Three PF patients not tested by RFLP analyses were judged to carry the DQw6 specificity because these patients had DR2/DR4, DR2/DR4, and DR2/DRwl2, respectively, and DQw5 is very rare among DR2 Japanese (less than 1%). Therefore, these results suggested that

seven of 13 PF patients were considered to have the DQw5 specificity. PF is associated with D Q w l (including DQw5 and DQw6), but not solely with DQwS. In addition, any DR alloantigens associated with PF could not be detected. Thus it is suggested that the D Q w l associated allele in the D Q A I gene, but not DQB1 gene, may contribute to the susceptibility to PF, because the D Q w l serologic specificity is defined by the D Q A 1 gene product, but DQw5 and DQw6 are defined by the DQB1 gene product. It is also indicated that, in Japanese patients, PF can be distinguished from PV by the disease-associated phenotypes, as well as by the disease-specific autoantigens reacting with the patients" sera [1]. Recently, Petzl-Erler and Santamaria [17] reported that, in Brazilian pemphigus foliaceus (BPF), two haplotypes (DR1-DQwl and DR4-DQw3) are positively associated and one haplotype (DR3- or DR7DQw2) is negatively associated. BPF and PF are considered to be closely related diseases because the clinical features and the antigenic specificity of the autoantibodies are similar in both conditions [ 18]. Therefore, it is of interest to note that the disease-specific haplotypes of BPF are different from those of Japanese PF. This suggests that a relationship between HLA and the susceptibility or resistance to PF varies muong the races or populations. The molecular mechanism by which particular DR or D Q alleles increase the susceptibility to certain autoimmune disease is unknown. However, PV patients may be immunogenedcally identical over the races and populations, ~nd appear to be different from Japanese PF patients, lmmunogenedc studies at a molecular level are in progress to further analyze HLA-D region associations with pemphigus.

ACKNOWLEDGMENTS We thank M. Haranaka, C. Ono, M. Yoshida, and S. Komatsu for technical assistance and T. Ando, K. Nakazawa, K. Sato, and M. lnoue for their helpful discussions and advice. This work was partly supported by research grants from the Ministry of Health and Welfare of Japan.

250

REFERENCES 1. Stanley JR: Pemphigus and pemphigoid as paradigms of organ-specific, autoantibody-mediated diseases. J Clin Invest 83:1443, 1989. 2. Hashimoto T, Ogawa M, Konohana A, Nishikawa T: Detection of pemphigus vulgaris and pemphigus foliaceus antigens by immunoblot analysis using different antigen sources. J Invest Dermato194:327, 1990. 3. Brautbar C, Moscovitz M, Livshits T, Haim S, HachamZadeh S, Cohen HA, Sharon R, Nelken D, Cohen T: HLA-DRW4 in pemphigus patients in Israel. Tissue Antigens 16:238, 1980. 4. Park MS, Terasaki Pl, Ahmed AR, Tiwari JL: HLADRW4 in 91% of Jewish pemphigus vulgaris patients, Lancet 2:441, 1979. 5. Matsuyama M, Hashimoto K, Yamasaki Y, Shirakura R, Higuchi R, Miyajima T, Amemiya H: HLA-DR antigens in pemphigus among Japanese. Tissue Antigens 17:238, 1981. 6. Sakurai M, Takigawa M, Terasaki PI, Imarnura S, Sakurami T, Ueno Y, Park MS: Absence of HLA-DRW2 in Japanese pemphigus vulgaris. J Invest Dermato176:70, 198!. 7. Szafer F, Brautbar C, Tzfoni E, Frankel G, Sherman L, Cohen I, Hacham-Zadeh S, Aberer W, Tappeiner G, Holubar K, Steinman L, Friedmann A: Detection of diseasespecific restriction fragment length polymorphisms in pemphigus vulgaris linked to the DQwl and DQw3 alleles of the HLA-D region. Proc Natl Acad Sci USA 84:6542, 1987. 8. Terasaki Pl, Bernoco D, Park MS, Ozturk G, lwaki Y: Microdroplet testing for FILA-A, -B, -C, and -D antigens. AmJ Clin Patho169:103, 1978. 9. Inoko H, Ando A, Kimura M, Tsuji K: Isolation and characterization of the cDNA and genomic clones of the HLA class II antigens. In Tsuji K (eds): Proceedings of the Twentie:h Japan Transplantation Society Symposium. Tokyo, Tokai University Press, 1985, p 188.

FI. Niizeki et al.

10. Peinberg AP, Vogelstein B: A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Anal Biochem 132:6, 1983. 11. Maeda M, Inoko H, Ando A, Uryu N, Nagata Y, Tsuji K: HLA-DP typing by analysis of DNA restriction fragment length polymorphisms in the HLA-DPfl subregion. Hum Immunol 21:239, 1988. 12. Sinha AA, Brautbar C, Szafer F, Friedmann A, Tzfoni E, Todd JA, Steinman L, McDevitt HO: A newly characterized HLA DQfl allele associated with pemphigus vulgaris. Science 239: I026, 1988. 13. Simons MJ, Wheeler R, Lalouel JM, Dupont B: Restriction fragment length polymotphisms of HLA genes: Summary of the tenth International Workshop southern blot analysis. In Dupont B (eds): Immunobiology of HLA, Vol 1. New York, Springer-Verlag, 1989, p 959. 14. Scharf SJ, Friedmann A, Steinman L, Brautbar C, Erlich HA: Specific HLA-DQB and HLA-DRB1 alleles confer susceptibility to pemphigus vulgaris. Proc Natl Acad Sci USA 86:6215, 1989. 15. Schaff SJ, Friedmann A, Brautbar C, Szafer F, Steinman L, Horn G, Gyllensten U, Erlich HA: HLA class 11 allelic variation and susceptibility to pemphigus vulgaris. Proc Natl Acad Sci USA 85:3504, 1988. 16. SchaffSJ, Long CM, Erlich HA: Sequence analysis of the HLA-DRfl and HLA-DQfl loci from three pemphigus vulgaris patients. Hum Immunol 22:61, 1988. Petzl-Erler ML, Santamaria J: Are HLA class II genes controlling susceptibility and resistance to Brazilian pemphigus foliaceus (logo selvagem)? Tissue Antigens 33:408, 1988. 18. Stanley JR, Klaus-Kovtun V, Sampaio SAP: Antigenic specificity of fogo selvagem autoantibodies is similar to North American pemphigus foliaceus and distinct from pemphigus vulgaris autoantibodies. J Invest Dermatol 87:197, 1986.

HLA class II antigens are associated with Japanese pemphigus patients.

We investigated the HLA class II antigens in 30 Japanese cases of pemphigus, 17 cases of pemphigus vulgaris (PV) and 13 cases of pemphigus foliaceus (...
544KB Sizes 0 Downloads 0 Views