Genetic and environmental factors regulate the type 1 diabetes gene CTSH via differential DNA methylation.

Cathepsin H (CTSH) is a type 1 diabetes (T1D) risk gene; large-scale genetic and epidemiological studies found that T1D genetic risk correlates with high CTSH expression, rapid decline of beta-cell function, and early onset T1D. Counterintuitively, transcriptional downregulation of CTSH by proinflammatory cytokines has been shown to promote beta-cell apoptosis. Here, we potentially explain these observed contrasting effects, describing a new mechanism where proinflammatory cytokines and T1D genetic risk variants regulate CTSH transcription via differential DNA methylation. We show that, in human islets, CTSH downregulation by the proinflammatory cytokine cocktail interleukin 1ß + tumor necrosis factor α + interferon γ was coupled with DNA hypermethylation in an open chromatin region in CTSH intron 1. A luciferase assay in human embryonic kidney 293 cells revealed that methylation of three key cytosine-phosphate-guanine dinucleotide (CpG) residues in intron 1 was responsible for the reduction of promoter activity. We further found that cytokine-induced intron 1 hypermethylation is caused by lowered Tet1/3 activities, suggesting that attenuated active demethylation lowered CTSH transcription. Importantly, individuals who carry the T1D risk variant showed lower methylation variability at the intron 1 CpG residues, presumably making them less sensitive to cytokines, whereas individuals who carry the protective variant showed higher methylation variability, presumably making them more sensitive to cytokines and implying differential responses to environment between the two patient populations. These findings suggest that genetic and environmental influences on a T1D locus are mediated by differential variability and mean of DNA methylation.

Maternal Microchimerism in Cord Blood and Risk of Celiac Disease in Childhood.

OBJECTIVES: During pregnancy, small quantities of maternal cells are naturally transmitted to the fetus. This transmission, termed maternal microchimerism (MMc), has been implicated in autoimmune diseases but its potential role is unclear. We aimed to investigate if MMc at birth predicted childhood celiac disease (CD) risk, a common immune-mediated enteropathy often presenting in childhood. METHODS: We designed a case-control study, nested in the Norwegian Mother, Father and Child Cohort. Participants were HLA class II typed to determine noninherited, nonshared maternal alleles (NIMA). Droplet digital (dd) PCR assays specific for common HLA class II NIMAs (HLA-DQB103:01, 04:02 and 06:02/03) were used to estimate the quantity of maternal DNA, as a marker of maternal cells, in cord blood DNA from 124 children who later developed clinically diagnosed CD (median age at end of study 7.4 years, range 3.6-12.9) and 124 random controls. We tested whether presence of MMc was associated with CD using logistic regression, and compared ranks between cases and controls. RESULTS: MMc, for example, maternal HLA antigens not inherited by the child, was found in 42% of cases and 43% of controls, and not associated with CD (odds ratio [OR] 0.97, 95% confidence interval [CI] 0.58-1.60). The ranks of MMc quantities in cases and controls were also similar (Mann-Whitney U-test, P = 0.71). The subgroup with HLA-DQB1:03*01 as their NIMA had a potential association with MMc, where levels greater than median was associated with CD (OR 3.78, 95% CI 1.28-11.18). CONCLUSION: MMc measured in cord blood was not associated with later risk of CD.

Maternal microchimerism in cord blood and risk of childhood-onset type 1 diabetes.

BACKGROUND: Maternal microchimerism (MMc), the transmission of small quantities of maternal cells to the fetus, is relatively common and persistent. MMc has been detected with increased frequency in the circulation and pancreas of type 1 diabetes (T1D) patients. We investigated for the first time whether MMc levels at birth predict future T1D risk. We also tested whether cord blood MMc predicted MMc in samples taken at T1D diagnosis. METHODS: Participants in the Norwegian Mother and Child Cohort study were human leukocyte antigen (HLA) class II typed to determine non-inherited, non-shared maternal alleles (NIMA). Droplet digital (dd) polymerase chain reaction (PCR) assays specific for common HLA class II NIMA (HLADQB1*03:01, *04:02, and *06:02/03) were developed and validated. MMc was estimated as maternal DNA quantity in the fetal circulation, by NIMA specific ddPCR, measured in cord blood DNA from 71 children who later developed T1D and 126 controls within the cohort. RESULTS: We found detectable quantities of MMc in 34/71 future T1D cases (48%) and 53/126 controls (42%) (adjusted odds ratio [aOR] 1.27, 95% confidence interval (CI) 0.68-2.36), and no significant difference in ranks of MMc quantities between cases and controls (Mann-Whitney P = .46). There was a possible association in the NIMA HLA-DQB1*03:01 subgroup with later T1D (aOR 3.89, 95%CI 1.05-14.4). MMc in cord blood was not significantly associated with MMc at T1D diagnosis. CONCLUSIONS: Our findings did not support the hypothesis that the degree of MMc in cord blood predict T1D risk. The potential subgroup association with T1D risk should be replicated in a larger cohort.

Identification of loci where DNA methylation potentially mediates genetic risk of type 1 diabetes.

The risk of Type 1 Diabetes (T1D) comprises both genetic and environmental components. We investigated whether genetic susceptibility to T1D could be mediated by changes in DNA methylation, an epigenetic mechanism that potentially plays a role in autoimmune diabetes. From enrichment analysis, we found that there was a common genetic influence for both DNA methylation and T1D across the genome, implying that methylation could be either on the causal pathway to T1D or a non-causal biomarker of T1D genetic risk. Using data from a general population comprising blood samples taken at birth (n = 844), childhood (n = 846) and adolescence (n = 907), we then evaluated the associations between 64 top GWAS single nucleotide polymorphisms (SNPs) and DNA methylation levels at 55 non-HLA loci. We identified 95 proximal SNP-cytosine phosphate guanine (CpG) pairs (cis) and 1 distal SNP-CpG association (trans) consistently at birth, childhood, and adolescence. Combining genetic co-localization and Mendelian Randomization analysis, we provided evidence that at 5 loci, ITGB3BP, AFF3, PTPN2, CTSH and CTLA4, DNA methylation is potentially mediating the genetic risk of T1D mainly by influencing local gene expression.

Attenuated humoral responses in HLA-A*24-positive individuals at risk of type 1 diabetes.

AIMS/HYPOTHESIS: The rate of progression from islet autoimmunity to clinical type 1 diabetes depends on the rate of beta cell destruction. The HLA-A*24 gene is associated with early diabetes onset, but previous studies have shown attenuated humoral responses to islet antigens in individuals with both recent and long-standing type 1 diabetes carrying HLA-A*24. We aimed to establish whether HLA-A*24 is also associated with attenuated humoral responses in individuals at high risk of type 1 diabetes. METHODS: We established HLA-A*24, DQ and rs9258750 (an HLA-A*24 tagged single-nucleotide polymorphism) genotype, as well as GAD, zinc transporter 8 (ZnT8), insulin, islet antigen-2 (IA-2), and IA-2ß autoantibody status in 373 islet cell antibody-positive first-degree relatives participating in the European Nicotinamide Diabetes Intervention Trial. RESULTS: Univariate regression analyses showed that humoral responses to GAD, ZnT8 and insulin were less common in relatives carrying HLA-A*24. The prevalence of GAD and ZnT8 autoantibodies remained negatively associated with HLA-A*24 and rs9258750 after adjusting for age, sex, proband relationship and HLA class II genotype. CONCLUSIONS/INTERPRETATION: HLA-A*24 is associated with attenuated humoral responses in individuals at high risk of type 1 diabetes, and this may reflect a distinct phenotype of rapid beta cell loss.

Maternal micro-chimeric cells in the multiple sclerosis brain.

Maternal microchimeric cells (MMC) pass across the placenta from a mother to her baby during pregnancy. MMC have been identified in healthy adults, but have been reported to be more frequent and at a higher concentration in individuals with autoimmune diseases. MMC in brain tissue from individuals with autoimmune neurological disease has never previously been explored. The present study aims to identify and quantify MMC in adult human brain from control and multiple sclerosis (MS) affected individuals using fluorescent in situ hybridization (FISH) with a probe for the X and Y chromosomes. Post mortem brain tissue from 6 male MS cases and 6 male control cases were examined. Female cells presumed to be MMC were identified in 5/6 MS cases and 6/6 control cases. Cell specific labeling identified female cells of neuronal and immune phenotype in both control and active MS lesion tissue. This study shows that female cells presumed to be MMC are a common phenomenon in adult human brain where they appear to have embedded into brain tissue with the ability to express tissue specific markers.

Unravelling the Roles of Susceptibility Loci for Autoimmune Diseases in the Post-GWAS Era.

Although genome-wide association studies (GWAS) have identified several hundred loci associated with autoimmune diseases, their mechanistic insights are still poorly understood. The human genome is more complex than single nucleotide polymorphisms (SNPs) that are interrogated by GWAS arrays. Apart from SNPs, it also comprises genetic variations such as insertions-deletions, copy number variations, and somatic mosaicism. Although previous studies suggest that common copy number variations do not play a major role in autoimmune disease risk, it is possible that certain rare genetic variations with large effect sizes are relevant to autoimmunity. In addition, other layers of regulations such as gene-gene interactions, epigenetic-determinants, gene and environmental interactions also contribute to the heritability of autoimmune diseases. This review focuses on discussing why studying these elements may allow us to gain a more comprehensive understanding of the aetiology of complex autoimmune traits.

Fluorescence In Situ Hybridization with Concomitant Immunofluorescence in Human Pancreas.

The ability to identify the presence of non-host cells in human pancreas with concomitant characterization of cell phenotype is particularly important to facilitate studies of transplantation and microchimerism resulted from pregnancy. The steps involved in processing tissue for fluorescence in situ hybridization (FISH) can however remove epitopes that are crucial for immunofluorescence and antigen retrieval strategies for immunofluorescence can negatively influence FISH. We describe a robust method to analyze X/Y chromosome constitution and cell phenotype simultaneously on the same pancreatic tissue section.

Laser Capture and Single Cell Genotyping from Frozen Tissue Sections.

There is an increasing requirement for genetic analysis of individual cells from tissue sections. This is particularly the case for analysis of tumor cells but is also a requirement for analysis of cells in pancreas from individuals with type 1 diabetes where there is evidence of viral infection or in the analysis of chimerism in pancreas; either post-transplant or as a result of feto-maternal cell transfer.This protocol describes a strategy to isolate cells using laser microdissection and to run a 17plex PCR to discriminate between cells of haplo-identical origin (i.e., fetal and maternal cells) in pancreas tissue but other robust DNA tests could be used. In short, snap-frozen tissues are cryo-sectioned and mounted onto membrane-coated slides. Target cells are harvested from the tissue sections by laser microdissection and pressure catapulting (LMPC) prior to DNA profiling. This is based on amplification of highly repetitive yet stably inherited loci (short tandem repeats, STR) as well as the amelogenin locus for sex determination and separation of PCR products by capillary electrophoresis.

Maternal microchimerism: friend or foe in type 1 diabetes?

Increased levels of non-inherited maternal HLA alleles have been detected in the periphery of children with type 1 diabetes and an increased frequency of maternal cells have been identified in type 1 diabetes pancreas. It is now clear that the phenotype of these cells is pancreatic, supporting the hypothesis that maternal cells in human pancreas are derived from multipotent maternal progenitors. Here we hypothesize how increased levels of maternal cells could play a role in islet autoimmunity.

Maternal microchimerism: increased in the insulin positive compartment of type 1 diabetes pancreas but not in infiltrating immune cells or replicating islet cells.

BACKGROUND: Maternal microchimeric cells (MMc) transfer across the placenta during pregnancy. Increased levels of MMc have been observed in several autoimmune diseases including type 1 diabetes but their role is unknown. It has been suggested that MMc are 1) effector cells of the immune response, 2) targets of the autoimmune response or 3) play a role in tissue repair. The aim of this study was to define the cellular phenotype of MMc in control (n = 14) and type 1 diabetes pancreas (n = 8). METHODS: Using sex chromosome-based fluorescence in-situ hybridization, MMc were identified in male pancreas and their phenotype determined by concomitant immunofluorescence. RESULTS: In normal pancreas, MMc positive for endocrine, exocrine, duct and acinar markers were identified suggesting that these cells are derived from maternal progenitors. Increased frequencies of MMc were observed in type 1 diabetes pancreas (p = 0.03) with particular enrichment in the insulin positive fraction (p = 0.01). MMc did not contribute to infiltrating immune cells or Ki67+ islet cell populations in type 1 diabetes. CONCLUSION: These studies provide support for the hypothesis that MMc in human pancreas are derived from pancreatic precursors. Increased frequencies of MMc beta cells may contribute to the initiation of autoimmunity or to tissue repair but do not infiltrate islets in type 1 diabetes.