Influence of TNF microsatellite polymorphisms (TNFa) on age-at-onset of insulin-dependent diabetes mellitus

Influence of TNF Microsatellite Polymorphisms (TNFa) on Age-at-Onset of Insulin-Dependent Diabetes Mellitus Hiroshi Obayashi, Naoto Nakamura, Michiaki Fukui, Hisataka Tegoshi, Mitsuhiro Fujii, Masakazu Ogata, Goji Hasegawa, Hirofumi Shigeta, Yoshihiro Kitagawa, Koji Nakano, Motoharu Kondo, Iwao Fukui, Etsuko Maruya, Hiroh Saji, Mitsuhiro Ohta, and Masataka Nishimura ABSTRACT: The TNF-␣ gene is located in the HLA region and has been implicated in the pathogenesis of Type I (insulin-dependent) diabetes mellitus (IDDM). We investigated the frequency of TNFa microsatellite alleles in 76 young-onset IDDM patients, 65 adult-onset IDDM patients, and 90 control subjects. We also examined the association of these TNFa alleles with HLADRB1 alleles, HLA-class I alleles, and TNF-␣ production. The frequency of the TNFa2 and TNFa9 alleles was increased in the young-onset IDDM patients compared to control subjects, but the increased frequency of TNFa2 was not significant after the correction for the number of comparisons was made. We did not find any association of TNFa2 or TNFa9 with any of the HLA-DRB1 alleles. In contrast, the frequency of the TNFa13 allele was decreased in both the young-onset and the adult-onset

ABBREVIATIONS TNF tumour necrosis factor TNFa TNF microsatellite polyporphism IDDM insuline-dependent diabetes mellitus MHC major histocompatibility complex

INTRODUCTION Type I (insulin-dependent) diabetes mellitus (IDDM) is caused by autoimmune destruction of pancreatic ␤ cells From The First Department of Internal Medicine Kyoto Prefectural University of Medicine (H.O., N.N., M.F., H.T., M.F., M.O., G.H., H.S., Y.K., K.N., M.K.), Kyoto 602-0841; the Department of Clinical Research, Kyoto Microbiological Institute (H.O., I.F.), Kyoto 607-8482; Department of Clinical Research Center, Utano National Hospital (M.O., M.N.), Kyoto 616-8255, Japan, and Kyoto Red Cross Blood Center (E.M., H.S.), Kyoto 605-0941, Japan. Address reprint requests to: Naoto Nakamura, The First Department of Internal Medicine, Kyoto Prefectural University of Medicine, KawaramachiHirokoji, Kamikyo-ku, Kyoto 602-0841, Japan; Tel: 075-251-5505; Fax: 075-252-3721. Received February 25, 1999; accepted April 22, 1999. Human Immunology 60, 974 –978 (1999) © American Society for Histocompatibility and Immunogenetics, 1999 Published by Elsevier Science Inc.

IDDM patients compared to the control subjects, but the difference lost significance after the correction was made in the adult-onset IDDM. The TNFa13 allele was strongly associated with DRB1*1502. Patients with TNFa2 or TNFa9 had greater TNF-␣ production, while those positive for TNFa13 had lower TNF-␣ production than patients with non-TNFa2, a9, and a13 alleles. These results suggest that TNFa polymorphisms are associated with age-at-onset of IDDM and influence the inflammatory process of pancreatic ␤ cell destruction in the development of IDDM. Human Immunology 60, 974 –978 (1999). © American Society for Histocompatibility and Immunogenetics, 1999. Published by Elsevier Science Inc. KEYWORDS: TNF microsatellite polyporphisms; TNF-␣; IDDM; Adult-onset IDDM; HLA

PBMC PCR LPS

peripheral blood mononclear cells polymerase chain reaction lipoplysaccaride

in genetically susceptible individuals [1]. In humans, strong genetic susceptibility to IDDM has been mapped to the major histocompatibility complex (MHC) region on the short arm of chromosome 6 [2, 3]. Although different alleles and haplotypes are associated with IDDM in different ethnic groups, certain HLA-DR and -DQ alleles have been shown to be associated with IDDM [4]. The age of onset for IDDM is variable, and it has been reported that the frequency of some IDDM susceptibility HLA-class II alleles varies depending on the age of disease onset [5, 6]. Demaine et al. [7] and Fujisawa et al. [8] reported that some 0198-8859/99/$–see front matter PII S0198-8859(99)00086-5

TNFa Microsatellite Polymorphisms in IDDM

genes in HLA class I region were associated with the age-at-onset of IDDM. The TNF-␣ and TNF-␤ genes are located in the central region of MHC, and five polymorphic microsatellites have been identified in or near the TNF gene [9]. There is evidence to suggest that TNF-␣ plays a crucial role in the pathogenesis of autoimmune diabetes. Recent studies have shown that several TNF microsatellite polyporphisms are associated with IDDM and TNF-␣ production levels [10 –13]. Since these IDDM-associated TNF microsatellite alleles or haplotypes were closely linked to HLA-DR3 haplotypes, the positive association of the TNF microsatellite alleles or haplotypes with IDDM may be due to linkage disequilibrium of these alleles or haplotypes with HLA-DR3. However, HLADR3 is a very rare allele in the Japanese population and most of the reported data have been obtained in typical (young-onset) IDDM. We, therefore, investigated the frequency of TNFa alleles in 76 young-onset IDDM patients, 65 adult-onset IDDM patients and 90 normal control subjects. We also examined the association between these TNFa alleles and several HLA-DRB1 and HLA class I (HLA-A, -B, and -C) alleles. Finally, we investigated the association between the different TNFa alleles and TNF-␣ production using peripheral blood mononuclear cells (PBMC). SUBJECTS AND METHODS Subjects Seventy-six young-onset IDDM patients, 65 adult-onset IDDM patients, and 90 normal control subjects, previously assessed for their HLA-A, -B, -C, and HLA-DRB1 allells, were analysed for TNFa polymorphisms. The mean age-at-onset of young-onset IDDM was 11.6 years (range: 3–19) and that of adult-onset IDDM was 47.6 (range: 35– 65). All the subjects were Japanese and unrelated, and all residents of the Kyoto and Osaka areas. The diagnoses of IDDM was based on the criteria of the National Diabetes Data Group [14]. HLA-class I alleles were typed by the National Institute of Health microlymphocytotoxicity method and HLA-DRB1 alleles were typed using a polymerase chain reaction (PCR)-low ionic strength single-stranded conformation polymorphism method [15]. TNFa Microsatellite Polymorphisms Genomic DNA was extracted from peripheral blood with DNA extractor WB kit (Wako pure chemical Ltd., Osaka, Japan) according to the manufacturer’s instructions. TNFa microsatellite alleles were determined by PCR with the following primers: forward (5⬘-GC CTCTAGATTTCATCCAGCCACA-3⬘) and reverse (5⬘CCTCTCTCCCCTGCAACACACA-3⬘). The forward

975

primer was 5⬘ end labeled with a fluorescent dye (6Fam). The amplification was performed in a 50-␮L reaction with 50 ng of genomic DNA and 1.0 U of AmpliTaq Gold polymerase (Perkin-Elmer, Norwalk, CT). The initial denaturation step was 10 min, followed by 35 cycles of 94°C for 1 min, 60°C for 50 seconds, and 72°C for 1 min. The PCR products were loaded onto 6% denaturing polyacrylamide gels with an internal lane standard labeled with “Genescan-500 Rox” dye and analysed on ABI 373 DNA sequencing systems using Genescan 672 software (Applied Biosystems, Foster City, CA). Analysis of TNF-␣ Production by PBMC PBMC were separated immediately after blood collection by density gradient centrifugation of heparinized venous blood on Ficol-Paque威 Plus (Pharmacia, Uppsala, Sweden). The cells were washed twice and resuspended in RPMI 1640 supplemented with 2% glutamine, 1% human AB serum, 100 U/ml penicillin, and 100 ␮g/ml streptomycin, adjusted to a concentration of 1 ⫻ 106 cells/ml/well and cultured in duplicate under 5% CO2 atmosphere with 500 pg/ml lipopolysaccharide (LPS) from E. coli 055:B5 W (Difco Labs., Detroit, MI). All culture media were endotoxin-free. After incubation for 24 h at 37°C, the cell culture supernatants were harvested by centrifugation and stored at ⫺80°C untiled assay. The TNF-␣ concentrations in the culture supernatants were measured using an enzyme-linked immunosorbent assay kit (R & D Systems, Minneapolis, MN) according to the manufacturer’s instructions. Statistical Analysis The data were evaluated by Chi-square test with 2 ⫻ 2 table using the standard statistical software (Stat View 4.5; Abacus Concepts, Berkeley, CA). The corrected p values (Pc) were obtained by multiplying the uncorrected p values (Pu) with the number of comparisons; Pc ⬍ 0.05 was taken as the level of significance. The data of the LPS-induced TNF-␣ production are expressed as the mean ⫾ SEM. The significance of differences between experimental groups was determined by one-way analysis of variance followed by Scheffe’s multiple comparison test (StatView software), A p ⬍ 0.05 was considered significant. RESULTS Frequency of TNFa Microsatellite Alleles The frequency of TNFa microsatellite alleles in 76 young-onset IDDM, 65 adult-onset IDDM patients, and 90 control subjects are shown in Table 1. When compared with the control subjects, the frequency of the TNFa2 and TNFa9 alleles were increased (9.4 vs 19.1%,

976

H. Obayashi et al.

TABLE 1 Frequency of TNFa microsatellite alleles in young-onset and adult-onset IDDM patients and control subjects TNFa alleles

1 2 3 4 5 6 7 8 9 10 11 12 13

Control subjects

Young-onset IDDM patients

size (bp)

n ⫽ 180 (%)

n ⫽ 152 (%)

98 100 102 104 106 108 110 112 114 116 118 120 122

4 (2.2) 17 (9.4) 0 (0.0) 1 (0.6) 24 (13.3) 50 (27.8) 2 (1.1) 3 (1.7) 7 (3.9) 25 (13.9) 16 (8.9) 9 (5.0) 22 (12.2)

7 (2.0) 29 (19.1) 0 (0.0) 2 (1.3) 29 (19.1) 28 (18.4) 2 (1.3) 2 (1.3) 21 (13.8) 24 (15.7) 6 (3.9) 2 (1.3) 0 (0.0)

Pu NS 0.00885 NS NS NS NS NS NS 0.00107 NS NS NS ⬍0.00001

Adult-onset IDDM patients Pc

n ⫽ 130 (%)

Pu

Pc

NS NS NS NS NS NS NS NS 0.01391 NS NS NS 0.00001

7 (5.4) 14 (10.8) 0 (0.0) 2 (1.5) 24 (18.5) 27 (20.8) 2 (1.5) 3 (2.3) 10 (7.7) 17 (13.1) 8 (6.2) 11 (8.5) 5 (3.8)

NS NS NS NS NS NS NS NS NS NS NS NS* 0.00698

NS NS NS NS NS NS NS NS NS NS NS NS NS

n, number of alleles. Pu, uncorrected P values. Pc, corrected P values. NS, not significant. *, Pu ⫽ 0.00439 vs young-onset IDDM patients.

Pu ⫽ 0.00885 and 3.9 vs 13.8%, Pc ⫽ 0.01391, respectively) in the young-onset IDDM patients, but TNFa2 was not significant after the correction for the number of comparisons was made. In contrast, the frequency of the TNFa13 was decreased in both the youngonset IDDM (12.2 vs 0.0%, Pu ⬍ 0.00001, Pc ⫽ 0.00001) and the adult-onset IDDM patients (12.2 vs. 3.8%, Pu ⫽ 0.00698), but the difference lost significance after the correction made in the adult-onset IDDM patients. The frequency of TNFa12 alleles was higher in the adult-onset IDDM patients than in the young-onset IDDM patients (8.5 vs. 1.3%, Pu ⫽ 0.00439). The Association between TNFa Alleles and HLA Haplotypes We did not find an association between the TNFa9 allele with HLA-DRB1 and the HLA class I alleles in either IDDM patients or control subjects. There was an association between TNFa2 and HLA-B61 in both IDDM patients and control subjects, although no significant association between TNFa2 and HLA-DRB1 alleles was found. The TNFa13 allele was strongly (Pu ⬍ 1 ⫻ 10⫺5) associated with HLA-DRB1*1502 in control subjects (data not shown). Relationship between TNFa Alleles and TNF-␣ Production by PBMC TNF-␣ production was determined in the diabetic patients with TNFa2/non-TNFa9 and a13 (n ⫽ 12), the patients with TNFa9/non-TNFa2 and a13 (n ⫽ 12), the patients with TNFa13/non-TNFa2 and a9 (n ⫽ 12), and the patients without TNFa2, a9, a13 (n ⫽ 20, as a

control group). The LPS-induced TNF-␣ production in the PBMC relative to the TNFa microsatellite polymorphisms is shown in Fig. 1.3 The mean (⫾SEM) TNF-␣ concentration in the supernatants was higher in the patients with TNFa2 (3378 ⫾ 108 pg/ml, P ⬍ 0.05) or TNFa9 (3538 ⫾ 110 pg/ml, P ⬍ 0.01), and significantly lower in the patients with TNFa13 (2188 ⫾ 88 pg/ml, P ⬍ 0.05) than in the control subjects (2758 ⫾ 138 pg/ml). DISCUSSION In this study we found a positive association of the TNFa9 allele with young-onset IDDM in a Japanese population. Concerning the frequency and association with HLA of TNFa9, Mizuki et al. reported that the frequency of TNFa9 was significantly increased and was associated with HLA-B51 in Bahcet’s disease [16], while Asano et al. reported that the frequency of TNFa9 was significantly higher in the rejection group of renal transplant recipients compared to a rejection-free group. They also found a strong association between TNFa9 and HLA-B35 [17]. An association between TNFa9 and HLA-B51 was observed among the IDDM patients, although it was not statistically significant. No significant difference was found in the frequency of HLA-B51 between the IDDM patients and the control subjects. Although further research is necessary to clarify other genes and/or polymorphisms encoded within HLA region for responsibility to TNF-␣ production, we found that both TNFa2 and TNFa9 are associated with high TNF-␣ production. We did not find any association of TNFa9

TNFa Microsatellite Polymorphisms in IDDM

FIGURE 1 Lipopolysaccharide-stimulated TNF-␣ production by peripheral blood mononuclear cells in relation to TNFa microsatellite alleles. *p ⬍ 0.05, **p ⬍ 0.01 vs diabetic patients with non-TNFa2, a9, and a13.

with HLA-DRB1 and/or HLA class I genes in IDDM, suggesting that TNFa9 is possibly an HLA class IIindependent genetic factor contributing to young-onset IDDM susceptibility and that high TNF-␣ production may contribute to the inflammatory process of pancreatic ␤ cell destruction in IDDM. We also found that the frequency of TNFa2 was increased in young-onset IDDM patients, although it was not significant after a correction was made for the number of comparisons. It has been reported that the TNFa2-bearing haplotype (TNFa2b3) is associated with IDDM in Caucasian populations and that this haplotype is in linkage disequilibrium with HLA-DR3 [10 –12]. However, HLA-DR3 is a very rare both among the IDDM patients and the control subjects in a Japanese population. There was no evidence for linkage disequilibrium between TNFa2 and HLA-DRB1 alleles. Also, we have demonstrated that the TNFa13 allele showed a negative association with young-onset IDDM. This negative association is due to the strong linkage disequilibrium (⬍10⫺5) between the TNFa13 and DRB1*1502, which is known to be a major resistance allele to IDDM in Japanese populations. In our current study, none of the young-onset IDDM patients possessed DRB1*1502. In conclusion, the results in this study suggest that TNFa polymorphisms were associated with age-at-onset of IDDM, and influence the inflammatory process of pancreatic ␤ cell destruction in the development of IDDM.

977

REFERENCES 1. Eisenbarth GS: Type I diabetes mellitus. A chronic autoimmune disease. N Engl J Med 314:1360, 1986. 2. Sheehy MJ: HLA and insulin-dependent diabetes. A protective perspective. Diabetes 41:123, 1992. 3. Davies JL, Kawaguchi Y, Bennett ST, Copeman JB, Cordell HJ, Pritchard LE, Reed PW, Gough SCL, Jenkins SC, Palmer SM, Balfour KM, Rowe BR, Farrall M, Barnett AH, Bain SC, Todd JA: A genome-wide search for human type 1 diabetes susceptibility genes. Nature 371:130, 1994. 4. She JX: Susceptibility to type I diabetes: HLA-DQ and DR revisited. Immunol Today 17:323, 1996. 5. Karjalainen J, Salmela P, Ilonen J, Surcel HM, Knip M: A comparison of childhood and adult type I diabetes mellitus. N Engl J Med 320:881, 1989. 6. Caillat-Zucman S, Garchon HJ, Timsit J, Assan R, Boitard C, Djilali-SAiah I, Bougneres P, Bach JF: Agedependent HLA genetic heterogeneity of type 1 insulindependent diabetes mellitus. J Clin Invest 90:2242, 1992. 7. Demaine AG, Hibberd ML, Mangles D, Millward BA: A new marker in the HLA class I region is associated with the age at onset of IDDM. Diabetologia 38:623, 1995. 8. Fujisawa T, Ikegami H, Kawaguchi Y, Yamato E, Takekawa K, Nakagawa Y, Hamada Y, Ueda H, Shima K, Ogihara T: Class I HLA is associated with age-at-onset of IDDM, while class II HLA confers susceptibility to IDDM. Diabetologia 38:1493, 1995. 9. Udalova IA, Nedospasov SA, Webb GC, Chaplin DD, Turetskaya RL: Highly informative typing of the human TNF locus using six adjacent polymorphic markers. Genomics 16:180, 1993. 10. Pociot F, Briant L, Jongeneel CV, Molvig J, Worsaae H, Abbal M, Thomsen M, Nerup J, Combon-Thomsen A:

978

Association of tumor necrosis factor (TNF) and class II major histocompatibility complex alleles with the secretion of TNF-␣ and TNF-␤ by human mononuclear cells: a possible link to insulin-dependent diabetes mellitus. Eur J Immunol 23:224, 1993. 11. Monos DS, Kamoun M, Udalova IA, Csanky E, Cizman B, Turetskaya RL, Smirnova JB, Zharkov VG, Gasser D, Zmijewski CM, Spielman RS, Nedospasov SA: Genetic polymorphism of the human tumor necrosis factor region in insulin-dependent diabetes mellitus. Linkage disequilibrium of TNFab microsatellite alleles with HLA haplotypes. Hum Immunol 44:70, 1995. 12. Garcia-Merino A, Alper CA, Usuku K, Marcus-bagley D, Lincoln R, Awdeh Z, Yunis EJ, Eisenbarth GS, Brink SJ, Hauser SL: Tumor necrosis factor (TNF) microsatellite haplotypes in relation to extended haplotypes, susceptibility to diseases associated with the major histocompatibility complex and TNF secretion. Hum Immunol 50:11, 1996. 13. Moghaddam PH, Zwinderman AH, de Knijff P, Rope BO, Schipper RF, Van der Auwera B, Naipal A, Gorus F, Schuit F, Giphart MJ: TNFa microsatellite polymorphism

H. Obayashi et al.

modulates the risk of IDDM in Caucasians with the high-risk genotype HLADQA1*0501-DQB1*0201/ DQA1*0301-DQB1*0302. Belgian Diabetes Registry. Diabetes 46:1514, 1997. 14. National Diabetes Data Group: Classification and diagnosis of diabetes mellitus and other categories of glucose intolerance. Diabetes 28:1039, 1979. 15. Maruya E, Saji H, Yokoyama S: PCR-LIS-SSCP (Low ionic strength single-stranded conformation polymorphism)—a simple method for high-resolution allele typing of HLADRB1, -DQB1, and -DPB1. Genome Res 6:51, 1996. 16. Mizuki N, Ohno S, Sato T, Ishihara M, Miyata S, Nakamura S, Naruse T, Mizuki H, Tsuji K, Inoko H: Microsatellite polymorphism between the tumor necrosis factor and HLA-B genes in Behcet’s disease. Hum Immunol 43:129, 1995. 17. Asano H, Kobayashi T, Uchida K, Hayashi S, Yokoyama I, Inoko H, Takagi H: Significance of tumor necrosis factor microsatellite polymorphism in renal transplantation. Tissue Antigens 50:484, 1997.