Complex multi-block analysis identifies new immunologic and genetic disease progression patterns associated with the residual β-cell function 1 year after diagnosis of type 1 diabetes

Characterizing T cell responses to enzymatically modified beta cell neo-epitopes

Introduction

Previous studies verify the formation of enzymatically post-translationally modified (PTM) self-peptides and their preferred recognition by T cells in subjects with type 1 diabetes (T1D). However, questions remain about the relative prevalence of T cells that recognize PTM self-peptides derived from different antigens, their functional phenotypes, and whether their presence correlates with a specific disease endotype.

Methods

To address this question, we identified a cohort of subjects with T1D who had diverse levels of residual beta cell function. Using previously developed HLA class II tetramer reagents, we enumerated T cells that recognize PTM GAD epitopes in the context of DRB1*04:01 or PTM IA2 epitopes in the context of DQB1*03:02 (DQ8).

Results

Consistent with prior studies, we observed higher overall frequencies and a greater proportion of memory T cells in subjects with T1D than in HLA matched controls. There were significantly higher numbers of GAD specific T cells than IA2 specific T cells in subjects with T1D. T cells specific for both groups of epitopes could be expanded from the peripheral blood of subjects with established T1D and at-risk subjects. Expanded neo-epitope specific T cells primarily produced interferon gamma in both groups, but a greater proportion of T cells were interferon gamma positive in subjects with T1D, including some poly-functional cells that also produced IL-4. Based on direct surface phenotyping, neo-epitope specific T cells exhibited diverse combinations of chemokine receptors. However, the largest proportion had markers associated with a Th1-like phenotype. Notably, DQ8 restricted responses to PTM IA2 were over-represented in subjects with lower residual beta cell function. Neo-epitope specific T cells were present in at-risk subjects, and those with multiple autoantibodies have higher interferon gamma to IL-4 ratios than those with single autoantibodies, suggesting a shift in polarization during progression.

Discussion

These results reinforce the relevance of PTM neo-epitopes in human disease and suggest that distinct responses to neo-antigens promote a more rapid decline in beta cell function.

Ketoacidosis at onset of type 1 diabetes in children up to 14 years of age and the changes over a period of 18 years in Saxony, Eastern-Germany: A population based register study

OBJECTIVE

The aim of this study was to examine the incidence trends of type 1 diabetes diagnosed with ketoacidosis in Saxony, Germany from 1999 to 2016.

METHODS

The population based Childhood Diabetes Registry of Saxony comprising valid data for all children aged 0-14 years diagnosed with type 1 diabetes from1999 to 2016 were used for the analyses. Direct age-standardized incidence rates were calculated and the effects of age, sex, calendar year, home districts and family history of any types of diabetes on the incidence were modelled using Poisson regression. Trend analyses for standard rate ratios of children with moderate and severe diabetic ketoacidosis versus children with type 1 diabetes with non-diabetic ketoacidosis were performed using join point regression.

RESULTS

The rate of ketoacidosis at the time of the type 1 diabetes diagnosis was high with 35.2% during the entire observation period in Saxony. The Poisson regression analysis indicated a statistically significant increased occurrence of diabetic ketoacidosis for younger age-groups, but no statistically significant differences between boys and girls. The join point trend analyses show that the proportion of severe and moderate ketoacidosis is increasing disproportionally to the increase in incidence of type 1 diabetes over the years.

CONCLUSION

Due to the observed increasing incidence of diabetes as well of diabetic ketoacidosis, an educational prevention campaign is needed in Saxony as soon as possible to aid pediatricians, general physicians as well as general public to identify the early signs of type 1 diabetes.

The association between rs1893217, rs478582 in PTPN2 and T1D risk with different diagnosed age, and related clinical characteristics in Chinese Han population

OBJECTIVE

To investigate the association between polymorphisms in PTPN2 (rs1893217 and rs478582) and type 1 diabetes (T1D) risk with different diagnosed age, as well as related clinical characteristics in Chinese Han population.

METHODS

A total of 2270 Chinese Han individuals (1023 T1D patients and 1247 healthy controls) were genotyped for rs1893217 and rs478582. And 306 newly diagnosed T1D patients were measured for C-peptide levels based on a standard mixed-meal tolerance test. In addition, 40 healthy controls were analyzed for different T cell subsets by multi-color flow cytometry.

RESULTS

Neither rs1893217 nor rs478582 showed any association with T1D risk under an additive model. Stratified analysis for T1D diagnosed age revealed that rs1893217, but not rs478582, was significantly associated with T1D patients diagnosed age ≤18 (OR =0.80, 95% CI: 0.67-0.97, p = 0.02). For those diagnosed age >18, neither of them showed any association. We also found that rs1893217 had a higher positive rate of ZnT8A (CC vs. TT carrier, OR = 2.07, 95% CI: 1.07-4.03, p = 0.026) and IA-2A (CT vs. TT carrier, OR = 1.36, 95% CI: 1.02-1.80, p = 0.038). Furthermore, for rs478582, compared with TT, healthy individuals carrying CC/CT carriers had significantly lower frequency and Helios expression of naive Treg subsets (p = 0.049 and 0.048 respectively), but not secreting or activating Treg subsets. In addition, we did not find any association between these two polymorphisms and residual β-cell function in newly diagnosed T1D patients.

CONCLUSIONS

Our results suggest that rs1893217 may increase the risk of early-onset T1D and affect humoral immunity, while rs478582 may affect Treg subsets.

Capturing residual beta cell function in type 1 diabetes

Since the 1970s, C-peptide has been used as a surrogate marker for monitoring the progression of type 1 and type 2 diabetes and to determine the effects of interventions designed to preserve or improve residual beta cell function. C-peptide measurement is a well-established surrogate of residual beta cell activity and of clinical significance as it is associated with HbA_1c, risk for microvascular complications and the incidence of hyperglycaemia in longitudinal studies. Measurement of C-peptide after a mixed meal tolerance test is considered the gold standard of measuring beta cell function in type 1 diabetes, but the method is laborious and inconvenient. In this issue of Diabetologia, Wentworth et al ( https://doi.org/10.1007/s00125-018-4722-z ) report an algorithm for estimating C-peptide (CP_EST) based on six routine clinical measures. These do not include stimulated C-peptide measurement and outperform other prevailing algorithms for estimating residual beta cell function. Going forward it is very likely that this new algorithm will serve as a simple measure of beta cell function in routine practice and as a more acceptable primary outcome measure in future trials of disease-modifying therapies.

Type 1 diabetes genome-wide association studies: not to be lost in translation

Genetic studies have identified >60 loci associated with the risk of developing type 1 diabetes (T1D). The vast majority of these are identified by genome-wide association studies (GWAS) using large case-control cohorts of European ancestry. More than 80% of the heritability of T1D can be explained by GWAS data in this population group. However, with few exceptions, their individual contribution to T1D risk is low and understanding their function in disease biology remains a huge challenge. GWAS on its own does not inform us in detail on disease mechanisms, but the combination of GWAS data with other omics-data is beginning to advance our understanding of T1D etiology and pathogenesis. Current knowledge supports the notion that genetic variation in both pancreatic β cells and in immune cells is central in mediating T1D risk. Advances, perspectives and limitations of GWAS are discussed in this review.

Coincidence of PTPN22 c.1858CC and FCRL3 -169CC genotypes as a biomarker of preserved residual β-cell function in children with type 1 diabetes

BACKGROUND

Genotype-phenotype studies in type 1 diabetes (T1DM) patients are needed for further development of therapy strategies.

OBJECTIVE

Our aims were to investigate the distribution of selected PTPN22 and FCRL3 gene polymorphisms and their associations with clinical course of disease in children with newly diagnosed T1DM from the Pomeranian region of Poland.

SUBJECTS/METHODS

The prospective, longitudinal study of 147 children with newly diagnosed T1DM-autoimmune subtype was conducted. The PTPN22 c.1858T>C (rs2476601) and FCRL3 -169C>T (rs7528684) polymorphisms were analyzed using polymerase chain reaction-restriction fragment length polymorphism method (PCR-RFLP) and DNA sequencing. The frequencies of genotypes were compared between the study and population-matched control group (327 random anonymous samples from the Pomeranian region). Selected patients underwent a 24-monthly follow up [periodic re-evaluation of fasting C-peptide concentration (FCP) and hemoglobin A1c (HbA1c ) level].

RESULTS

A significantly lower coincidence of the PTPN22 c.1858CC and FCRL3 -169CC genotypes was found in the study group compared with controls (P = 0.04). The PTPN22 c.1858CC and FCRL3 -169CC genotype combination, restricted to female patients only, was associated with well-preserved residual β-cell function throughout the entire follow up (prolonged FCP level increase up to the sixth month of disease, with further very stable dynamics-FCP median level ≥0.67 ng/mL without significant decrease up to the 24th month). HbA1c levels in this subgroup also remained the lowest during the observation period.

CONCLUSIONS/INTERPRETATION

Ascertained phenomenon could be explained by an interacting mechanism of the two polymorphisms through estrogen-regulated nuclear factor kappa B signaling in regulatory T (T_reg ) lymphocytes. This hypothesis, if confirmed, may lead to further development of T_reg administration-based therapies.

The rs2292239 polymorphism in ERBB3 gene is associated with risk for type 1 diabetes mellitus in a Brazilian population

The Erb-b2 receptor tyrosine kinase 3 (ERBB3) belongs to a family of epidermal growth factor receptors of protein tyrosine kinases, and regulates cell survival, differentiation and proliferation in several cell types. Previous studies have suggested that ERBB3 contributes to T1DM pathogenesis by modulating antigen presenting cell function, autoimmunity and cytokine-induced beta-cell apoptosis. Accordingly, some genome-wide association studies identified ERBB3 gene as a susceptibility locus for T1DM, with the strongest association signal being observed for the rs2292239 single nucleotide polymorphism (SNP) in intron 7 of the gene. Therefore, the aim of the present study was to replicate the association of the ERBB3 rs2292239 SNP with T1DM in a Brazilian population. We analyzed 421 T1DM patients (cases) and 510 nondiabetic subjects (controls). All subjects were self-declared as white. The ERBB3 rs2292239 (A/C) SNP was genotyped by real-time PCR using TaqMan MGB probes. Genotype (P=0.001) and allele (P=0.002) frequencies of the ERBB3 rs2292239 SNP were differently distributed between T1DM patients and nondiabetic controls. Moreover, the A allele was significantly associated with risk for T1DM when considering recessive (OR=1.58, 95% CI 1.11-2.27; P=0.015), additive (OR=1.78, 95% CI 1.21-2.62; P=0.004), and dominant (OR=1.39, 95% CI 1.07-1.81; P=0.016) models of inheritance. However, after adjustment for presence of high-risk HLA DR/DQ genotypes, the rs2292239 SNP remained independently associated with T1DM only for the additive model (OR=1.62, 95% CI 1.02-2.59; P=0.043). Our results suggest that the A/A genotype of the ERBB3 rs2292239 SNP is associated with risk for T1DM in a white Brazilian population.

Circulating microRNA levels predict residual beta cell function and glycaemic control in children with type 1 diabetes mellitus

AIMS/HYPOTHESIS

We aimed to identify circulating microRNA (miRNA) that predicts clinical progression in a cohort of 123 children with new-onset type 1 diabetes mellitus.

METHODS

Plasma samples were prospectively obtained at 1, 3, 6, 12 and 60 months after diagnosis from a subset of 40 children from the Danish Remission Phase Cohort, and profiled for miRNAs. At the same time points, meal-stimulated C-peptide and HbA_1c levels were measured and insulin-dose adjusted HbA_1c (IDAA_1c) calculated. miRNAs that at 3 months after diagnosis predicted residual beta cell function and glycaemic control in this subgroup were further validated in the remaining cohort (n = 83). Statistical analysis of miRNA prediction for disease progression was performed by multiple linear regression analysis adjusted for age and sex.

RESULTS

In the discovery analysis, six miRNAs (hsa-miR-24-3p, hsa-miR-146a-5p, hsa-miR-194-5p, hsa-miR-197-3p, hsa-miR-301a-3p and hsa-miR-375) at 3 months correlated with residual beta cell function 6-12 months after diagnosis. Stimulated C-peptide at 12 months was predicted by hsa-miR-197-3p at 3 months (p = 0.034). A doubling of this miRNA level corresponded to a sixfold higher stimulated C-peptide level. In addition, a doubling of hsa-miR-24-3p and hsa-miR-146a-5p levels at 3 months corresponded to a 4.2% (p < 0.014) and 3.5% (p < 0.022) lower IDAA_1c value at 12 months. Analysis of the remaining cohort confirmed the initial finding for hsa-miR-197-3p (p = 0.018). The target genes for the six miRNAs revealed significant enrichment for pathways related to gonadotropin-releasing hormone receptor and angiogenesis pathways.

CONCLUSIONS/INTERPRETATION

The miRNA hsa-miR-197-3p at 3 months was the strongest predictor of residual beta cell function 1 year after diagnosis in children with type 1 diabetes mellitus.

C1858T Polymorphism of Protein Tyrosine Phosphatase Non-receptor Type 22 (PTPN22): an eligible target for prevention of type 1 diabetes?

INTRODUCTION

In type 1 diabetes (T1D), several genetic factors are associated to β-cell autoimmunity onset and clinical progression. HLA-genes play a major role in susceptibility and initiation of β-cell autoimmunity, whereas non-HLA genes may influence the destruction rate. Areas covered: Our review focuses on the possible role of the PTPN22 C1858 T variant as a prognostic factor, given its influence on disease variability. Moreover, we present the potential role of C1858 T as a target for tertiary prevention trials and new therapeutic strategies, such as the LYP inhibitors. We used PubMed for literature research; key words were 'PTPN22', 'C1858 T polymorphism', 'lymphoid-specific tyrosine phosphatase' and 'type 1 diabetes'. We selected publications between 2000 and 2016. Expert commentary: Current data suggest that PTPN22 can be a promising target for therapeutic interventions and identification of at-risk subjects in autoimmune diseases such as T1D.

Islet biology, the CDKN2A/B locus and type 2 diabetes risk

Type 2 diabetes, fuelled by the obesity epidemic, is an escalating worldwide cause of personal hardship and public cost. Diabetes incidence increases with age, and many studies link the classic senescence and ageing protein p16(INK4A) to diabetes pathophysiology via pancreatic islet biology. Genome-wide association studies (GWASs) have unequivocally linked the CDKN2A/B locus, which encodes p16 inhibitor of cyclin-dependent kinase (p16(INK4A)) and three other gene products, p14 alternate reading frame (p14(ARF)), p15(INK4B) and antisense non-coding RNA in the INK4 locus (ANRIL), with human diabetes risk. However, the mechanism by which the CDKN2A/B locus influences diabetes risk remains uncertain. Here, we weigh the evidence that CDKN2A/B polymorphisms impact metabolic health via islet biology vs effects in other tissues. Structured in a bedside-to-bench-to-bedside approach, we begin with a summary of the evidence that the CDKN2A/B locus impacts diabetes risk and a brief review of the basic biology of CDKN2A/B gene products. The main emphasis of this work is an in-depth look at the nuanced roles that CDKN2A/B gene products and related proteins play in the regulation of beta cell mass, proliferation and insulin secretory function, as well as roles in other metabolic tissues. We finish with a synthesis of basic biology and clinical observations, incorporating human physiology data. We conclude that it is likely that the CDKN2A/B locus influences diabetes risk through both islet and non-islet mechanisms.

Genetic Risk Score Modelling for Disease Progression in New-Onset Type 1 Diabetes Patients: Increased Genetic Load of Islet-Expressed and Cytokine-Regulated Candidate Genes Predicts Poorer Glycemic Control

Genome-wide association studies (GWAS) have identified over 40 type 1 diabetes risk loci. The clinical impact of these loci on β-cell function during disease progression is unknown. We aimed at testing whether a genetic risk score could predict glycemic control and residual β-cell function in type 1 diabetes (T1D). As gene expression may represent an intermediate phenotype between genetic variation and disease, we hypothesized that genes within T1D loci which are expressed in islets and transcriptionally regulated by proinflammatory cytokines would be the best predictors of disease progression. Two-thirds of 46 GWAS candidate genes examined were expressed in human islets, and 11 of these significantly changed expression levels following exposure to proinflammatory cytokines (IL-1β + IFNγ + TNFα) for 48 h. Using the GWAS single nucleotide polymorphisms (SNPs) from each locus, we constructed a genetic risk score based on the cumulative number of risk alleles carried in children with newly diagnosed T1D. With each additional risk allele carried, HbA1c levels increased significantly within first year after diagnosis. Network and gene ontology (GO) analyses revealed that several of the 11 candidate genes have overlapping biological functions and interact in a common network. Our results may help predict disease progression in newly diagnosed children with T1D which can be exploited for optimizing treatment.

The genetic and regulatory architecture of ERBB3-type 1 diabetes susceptibility locus

The study aimed to explore the role of ERBB3 in type 1 diabetes (T1D). We examined whether genetic variation of ERBB3 (rs2292239) affects residual β-cell function in T1D cases. Furthermore, we examined the expression of ERBB3 in human islets, the effect of ERBB3 knockdown on apoptosis in insulin-producing INS-1E cells and the genetic and regulatory architecture of the ERBB3 locus to provide insights to how rs2292239 may confer disease susceptibility. rs2292239 strongly correlated with residual β-cell function and metabolic control in children with T1D. ERBB3 locus associated lncRNA (NONHSAG011351) was found to be expressed in human islets. ERBB3 was expressed and down-regulated by pro-inflammatory cytokines in human islets and INS-1E cells; knockdown of ERBB3 in INS-1E cells decreased basal and cytokine-induced apoptosis. Our data suggests an important functional role of ERBB3 and its potential regulators in the β-cells and may constitute novel targets to prevent β-cell destruction in T1D.

Partial remission definition: validation based on the insulin dose-adjusted HbA1c (IDAA1C) in 129 Danish children with new-onset type 1 diabetes

OBJECTIVE

To validate the partial remission (PR) definition based on insulin dose-adjusted HbA1c (IDAA1c).

SUBJECTS AND METHODS

The IDAA1c was developed using data in 251 children from the European Hvidoere cohort. For validation, 129 children from a Danish cohort were followed from the onset of type 1 diabetes (T1D). Receiver operating characteristic curve (ROC) analysis was used to evaluate the predictive value of IDAA1c and age on partial C-peptide remission (stimulated C-peptide, SCP > 300 pmol/L).

RESULTS

PR (IDAA1c ≤ 9) in the Danish and Hvidoere cohorts occurred in 62 vs. 61% (3 months, p = 0.80), 47 vs. 44% (6 months, p = 0.57), 26 vs. 32% (9 months, p = 0.32) and 19 vs. 18% (12 months, p = 0.69). The effect of age on SCP was significantly higher in the Danish cohort compared with the Hvidoere cohort (p < 0.0001), likely due to higher attained Boost SCP, so the sensitivity and specificity of those in PR by IDAA1c ≤ 9, SCP > 300 pmol/L was 0.85 and 0.62 at 6 months and 0.62 vs. 0.38 at 12 months, respectively. IDAA1c with age significantly improved the ROC analyses and the AUC reached 0.89 ± 0.04 (age) vs. 0.94 ± 0.02 (age + IDAA1c) at 6 months (p < 0.0004) and 0.76 ± 0.04 (age) vs. 0.90 ± 0.03 (age + IDAA1c) at 12 months (p < 0.0001).

CONCLUSIONS

The diagnostic and prognostic power of the IDAA1c measure is kept but due to the higher Boost stimulation in the Danish cohort, the specificity of the formula is lower with the chosen limits for SCP (300 pmol/L) and IDAA1c ≤9, respectively.