Low Frequencies of Autoimmunity-Associated <b><i>PTPN22</i></b> Polymorphisms in MODY Patients, Including Those Transiently Expressing Islet Cell Autoantibodies

Abstract

Background: The protein tyrosine phosphatase nonreceptor type 22 (PTPN22) gene encodes lymphoid tyrosine phosphatase (LYP), which is expressed primarily in lymphoid tissues. The functional but geographically highly variable PTPN22 single-nucleotide polymorphisms (SNPs), particularly c.1858C>T, contribute to the onset and progression of autoimmunity-associated diseases and facilitate the expression of disease-associated autoantibodies. In Central Europe, 17-25% of patients with monogenic diabetes (maturity-onset diabetes of the young, MODY) transiently express islet cell autoantibodies. Methods: We addressed the links between the functional and geographically variable PTPN22 SNPs with MODY manifestation and the expression of islet cell autoantibodies in 276 MODY patients who originated from four regions (the Czech Republic, Israel, Japan and Brazil). Results: The frequency of PTPN22 polymorphisms in the MODY patients was similar to those in geographically matched healthy populations, with the exception of c.788G>A, the minor allele frequency of which was significantly elevated in the Czech hepatocyte nuclear factor 1-α (HNF1A) MODY patients [odds ratio (OR) 4.8, 95% confidence interval (CI) 2.2-10.7] and the Brazilian MODY patients (OR 8.4, 95% CI 1.8-39.1). A barely significant increase in the c.788G>A minor allele was also detected in the islet cell autoantibody-positive Czech MODY patients. However, c.788A behaves as a loss-of-function mutant in T cells, and thus protects against autoimmunity. Conclusions: MODY patients (including islet cell autoantibody-positive cases) do not display any increase in autoimmunity-associated PTPN22 alleles. The absence of autoimmunity-associated PTPN22 alleles was also demonstrated in latent autoimmune diabetes in adults, which suggests that the slow kinetics of the onset of autoantibodies is subject to a regulation that is different from that experienced in type 1 diabetes and other autoimmune disorders.

Autoimmunity-Associated <b><i>PTPN22</i></b> Polymorphisms in Latent Autoimmune Diabetes of the Adult Differ from Those of Type 1 Diabetes Patients

“…The effects of the last polymorphism analyzed in this study, c.1970-852T>C, on the activity of LYP is unknown; however, its frequencies were reported to be increased in T1DM and other autoimmune disorders [50-54]. A detailed overview of disease associations of the analyzed PTPN22 polymorphisms has recently been published by Heneberg et al [28]. We tested all SNPs using a polymerase chain reaction restriction fragment length polymorphism (PCR-RFLP), as previously described [28], with randomly chosen PCR products validated by Sanger sequencing ( n = 9 for each SNP).…”

“…We recruited about a half of the study subjects for the purpose of this study only; however, polymorphisms in selected loci and clinical parameters of some study subjects have been described previously [36-41]. For the purpose of calculating canonical correspondence analyses (CCA) and odds ratios (ORs), we matched the data obtained in the course of this study with previously published T1DM and control cohorts of subjects of European descent [32, 42] and compared them against Czech cohorts of HNF4A -, GCK - and HNF1A - maturity onset diabetes of the young (MODY) subjects [28]. We obtained written informed consent from each subject.…”

“…This polymorphism affects the binding site of transcription factor AP-4 within the PTPN22 sequence [46, 47] and thus likely affects the transcription of PTPN22. The c.788G>A polymorphism is a rare polymorphism of negligible frequency in healthy people of Asian origin, but it occurs in 2–3% of people of European descent [28]. It behaves as a loss-of-function mutant in T cells and decreases LYP phosphatase activity toward its physiological substrates in vitro [48].…”

“…A detailed overview of disease associations of the analyzed PTPN22 polymorphisms has recently been published by Heneberg et al [28]. We tested all SNPs using a polymerase chain reaction restriction fragment length polymorphism (PCR-RFLP), as previously described [28], with randomly chosen PCR products validated by Sanger sequencing ( n = 9 for each SNP). We retested all individuals revealed as heterozygous in the first round of PCR-RFLP to exclude incomplete cleavage.…”

“…Here, we tested the following hypotheses: (1) LADA behaves as a hybrid form of diabetes when concerning PTPN22 risk SNPs [27]; (2) carriers of PTPN22 risk SNPs manifest LADA more frequently when compared to the general population as suggested by the association of these SNPs with multiple human autoimmune disorders [28]; (3) the associations of PTPN22 risk SNPs in LADA and type 2 diabetes mellitus with LADA-like progression to insulinotherapy (ID-T2DM) patients are gender-specific as suggested previously for 1858T in T1DM [29-31, but compare 32, 33]; (4) the PTPN22 risk SNPs are associated with increased serum levels of GADA and IA2A in Czech LADA and ID-T2DM patients as suggested previously for 1858T with serum levels of autoantibodies to islet antigens in T1DM patients [18, 34-35] and high GADA serum levels in LADA patients [17]. …”