The association of HLA-G with type 1 diabetes in the South African black population

Abstract

Background: Type 1 diabetes mellitus (T1D) is an autoimmune disease that is characterized by a T cell mediated destruction of pancreatic β-cells which leads to an absolute insulin deficiency. The cause of T1D is multifactorial and HLA-G, a non- classical HLA class Ib gene, was identified as an independent locus for disease susceptibility. HLA-G has limited allelic and protein variability, and unique alternative splicing patterns that produce multiple isoforms. It has immune-suppressive properties which include altering cytokine secretion, downregulating the production of CD4+ T cells, inhibiting antigen-presenting cells, and inhibiting the activity and inducing apoptosis of cytotoxic CD8+ T cells and natural killer (NK) cells. The HLA-G molecule is constitutively expressed in the pancreas and is upregulated on the surface of islet cells that have been stimulated to secrete insulin. When there are more HLA-G molecules on the cell-surface, CD8+ T cell activation is suppressed, creating immune tolerance within the pancreas. Decreased expression of HLA-G therefore results in decreased tolerance to pancreatic autoantigens, and potentially the initiation and progression of T1D. Polymorphisms in the HLA-G gene, affect isoform production, peptide binding, mRNA expression and consequently the immunoregulatory functions of HLA-G. Polymorphisms in the 3’ untranslated region (UTR) of the HLA-G gene (a 14 bp insertion/deletion polymorphism and rs9380142) have been associated with an increased risk of developing T1D and a younger age at diagnosis of the disease. Therefore, the aim of this study was to sequence the HLA-G gene in black South Africans with T1D (cases) and controls and to determine whether polymorphisms in the HLA-G gene are associated with age at diagnosis and/or T1D in the South African black population. In addition, this study aimed to determine the effect of HLA-G genotypes on serum HLA-G concentrations. Methods: A total of 263 cases were previously recruited from the greater Johannesburg area. The eight exons of the HLA-G gene were sequenced and compared in 20 cases and 20 controls. The HLA-G sequences were aligned with a reference sequence (GRCh38.p14; NC_000006.12) to identify variants that had significantly different frequencies between cases and controls. Two variants in exon 8, rs1063320 and rs1710, were selected for further analysis. Participants were genotyped for the HLA-G rs1063320 and rs1710 polymorphisms using PCR-RFLP. In vi addition, participants were genotyped for the 14 bp indel using conventional PCR and the rs9380142 polymorphism using a predesigned Applied Biosystems TaqMan® SNP Genotyping assay. The concentration of serum HLA-G was measured by an enzyme linked immunosorbent assay. Statistical analysis was performed using Statistica software. Results: In our cohort the mean age at diagnosis of T1D was 19.3 ± 10.1 years with a median duration of 6.00 [2.00; 12.0] years. Blood glucose concentrations (p<0.001), systolic blood pressure (SBP) (p=0.021) and serum HLA-G concentrations (p<0.001) were significantly higher in cases than controls. There was no significant difference in the genotypic and allelic frequencies between cases and controls for any of the HLA- G polymorphisms studied (p>0.05). Similarly, there was no association between HLA- G polymorphisms and HLA-G concentrations (p>0.05). In a univariate analysis none of the four polymorphisms studied were associated with age at diagnosis, however, on multivariate analysis the 14 bp insertion/insertion genotype was associated with an earlier age at diagnosis (0.90 (0.81-1.00); p=0.042). Participants with T1D were 5.2 times more likely to have higher HLA-G concentrations compared to controls (p<0.001). In addition, participants with the 14 bp deletion/deletion genotype had a SBP, 4.00 mmHg higher than those with the insertion allele (p=0.036). Furthermore, for every one-year increase in age and one unit increase in BMI, the SBP increased by 0.412 mmHg (p<0.001) and 0.67 mmHg (p<0.001), respectively. Conclusion: The rs1063320, rs1710, 14 bp indel and rs9380142 polymorphisms are not associated with T1D in the black South African population. The association of the 14 bp homozygous insertion with an earlier age at diagnosis supports the involvement of HLA-G in the pathogenesis of T1D. Serum HLA-G concentrations were higher in cases than controls. Our findings contrast with that reported in literature where the 14 bp deletion and lower HLA-G concentrations have been associated with age at diagnosis and T1D, respectively. It has been hypothesised that increased HLA-G levels cause apoptosis of CD8+ cytotoxic T cells and natural killer cells, thus protecting pancreatic β-cells from autoimmune destruction. However, our study does not support this hypothesis and future studies should be performed to ascertain the involvement of HLA-G in T1D pathogenesis in the black South African population