Joint effects of HLA, INS, PTPN22 and CTLA4 genes on the risk of type 1 diabetes

Deep integrative models for large-scale human genomics

Polygenic risk scores (PRSs) are expected to play a critical role in precision medicine. Currently, PRS predictors are generally based on linear models using summary statistics, and more recently individual-level data. However, these predictors mainly capture additive relationships and are limited in data modalities they can use. We developed a deep learning framework (EIR) for PRS prediction which includes a model, genome-local-net (GLN), specifically designed for large-scale genomics data. The framework supports multi-task learning, automatic integration of other clinical and biochemical data, and model explainability. When applied to individual-level data from the UK Biobank, the GLN model demonstrated a competitive performance compared to established neural network architectures, particularly for certain traits, showcasing its potential in modeling complex genetic relationships. Furthermore, the GLN model outperformed linear PRS methods for Type 1 Diabetes, likely due to modeling non-additive genetic effects and epistasis. This was supported by our identification of widespread non-additive genetic effects and epistasis in the context of T1D. Finally, we constructed PRS models that integrated genotype, blood, urine, and anthropometric data and found that this improved performance for 93% of the 290 diseases and disorders considered. EIR is available at https://github.com/arnor-sigurdsson/EIR.

CTLA4, SH2B3, and CLEC16A diversely affect the progression of early islet autoimmunity in relatives of Type 1 diabetes patients

The HLA region is the major genetic risk determinant of Type 1 diabetes. How non-HLA loci contribute to the genetic risk is incompletely understood, but there are indications that at least some impact progression of asymptomatic autoimmunity. We examined whether SNPs in 7 susceptibility loci (INS, SH2B3, PTPN2, PTPN22, CTLA4, CLEC16A, and IL2RA) could improve prediction of the progression from single to multiple autoantibody positivity, and from there on to diagnosis. SNPs were genotyped in persistently autoantibody positive relatives by allelic discrimination qPCR and disease progression was studied by multivariate Cox regression analysis. In our cohort, only the CTLA4 GA genotype (rs3087243, P = 0.002) and the CLEC16A AA genotype (rs12708716, P = 0.021) were associated with accelerated progression from single to multiple autoantibody positivity, but their effects were restricted to presence of HLA-DQ2/DQ8, and IAA as first autoantibody, respectively. The interaction of CTLA4 and HLA-DQ2/DQ8 overruled the effect of DQ2/DQ8 alone. The HLA-DQ2/DQ8-mediated risk of progression to multiple autoantibodies nearly entirely depended on heterozygosity for CTLA4. The SH2B3 TT genotype (rs3184504) was protective for HLA-DQ8 positive subjects (P = 0.003). At the stage of multiple autoantibodies, only the CTLA4 GA genotype was a minor independent risk factor for progression towards clinical diabetes (P = 0.034). Our study shows that non-HLA polymorphisms impact progression of islet autoimmunity in a subgroup-, stage- and SNP-specific way, suggesting distinct mechanisms. If confirmed, these findings may help refine risk assessment, follow-up, and prevention trials in risk groups.

The missing heritability in type 1 diabetes

Type 1 diabetes (T1D) is a complex autoimmune disease characterized by an absolute deficiency of insulin. It affects more than 20 million people worldwide and imposes an enormous financial burden on patients. The underlying pathogenic mechanisms of T1D are still obscure, but it is widely accepted that both genetics and the environment play an important role in its onset and development. Previous studies have identified more than 60 susceptible loci associated with T1D, explaining approximately 80%-85% of the heritability. However, most identified variants confer only small increases in risk, which restricts their potential clinical application. In addition, there is still a so-called 'missing heritability' phenomenon. While the gap between known heritability and true heritability in T1D is small compared with that in other complex traits and disorders, further elucidation of T1D genetics has the potential to bring novel insights into its aetiology and provide new therapeutic targets. Many hypotheses have been proposed to explain the missing heritability, including variants remaining to be found (variants with small effect sizes, rare variants and structural variants) and interactions (gene-gene and gene-environment interactions; e.g. epigenetic effects). In the following review, we introduce the possible sources of missing heritability and discuss the existing related knowledge in the context of T1D.

Non-HLA Gene Polymorphisms in the Pathogenesis of Type 1 Diabetes: Phase and Endotype Specific Effects

The non-HLA loci conferring susceptibility to type 1 diabetes determine approximately half of the genetic disease risk, and several of them have been shown to affect immune-cell or pancreatic β-cell functions. A number of these loci have shown associations with the appearance of autoantibodies or with progression from seroconversion to clinical type 1 diabetes. In the current study, we have re-analyzed 21 of our loci with prior association evidence using an expanded DIPP follow-up cohort of 976 autoantibody positive cases and 1,910 matched controls. Survival analysis using Cox regression was applied for time periods from birth to seroconversion and from seroconversion to type 1 diabetes. The appearance of autoantibodies was also analyzed in endotypes, which are defined by the first appearing autoantibody, either IAA or GADA. Analyzing the time period from birth to seroconversion, we were able to replicate our previous association findings at PTPN22, INS, and NRP1. Novel findings included associations with ERBB3, UBASH3A, PTPN2, and FUT2. In the time period from seroconversion to clinical type 1 diabetes, prior associations with PTPN2, CD226, and PTPN22 were replicated, and a novel association with STAT4 was observed. Analyzing the appearance of autoantibodies in endotypes, the PTPN22 association was specific for IAA-first. In the progression phase, STAT4 was specific for IAA-first and ERBB3 to GADA-first. In conclusion, our results further the knowledge of the function of non-HLA risk polymorphisms in detailing endotype specificity and timing of disease development.

Type 1 Diabetes and Autoimmune Thyroid Disease-The Genetic Link

Type 1 diabetes (T1D) and autoimmune thyroid disease (AITD) are the most frequent chronic autoimmune diseases worldwide. Several autoimmune endocrine and non-endocrine disorders tend to occur together. T1D and AITD often cluster in individuals and families, seen in the formation of autoimmune polyendocrinopathy (AP). The close relationship between these two diseases is largely explained by sharing a common genetic background. The HLA antigens DQ2 (DQA1*0501-DQB1*0201) and DQ8 (DQA1*0301-DQB1*0302), tightly linked with DR3 and DR4, are the major common genetic predisposition. Moreover, functional single nucleotide polymorphisms (or rare variants) of various genes, such as the cytotoxic T-lymphocyte- associated antigen (CTLA4), the protein tyrosine phosphatase non-receptor type 22 (PTPN22), the interleukin-2 Receptor (IL2Ra), the Vitamin D receptor (VDR), and the tumor-necrosis-factor-α (TNF) that are involved in immune regulation have been identified to confer susceptibility to both T1D and AITD. Other genes including cluster of differentiation 40 (CD40), the forkhead box P3 (FOXP3), the MHC Class I Polypeptide-Related Sequence A (MICA), insulin variable number of tandem repeats (INS-VNTR), the C-Type Lectin Domain Containing 16A (CLEC16A), the Erb-B2 Receptor Tyrosine Kinase 3 (ERBB3) gene, the interferon-induced helicase C domain-containing protein 1 (IFIH1), and various cytokine genes are also under suspicion to increase susceptibility to T1D and AITD. Further, BTB domain and CNC homolog 2 (BACH2), C-C motif chemokine receptor 5 (CCR5), SH2B adaptor protein 3 (SH2B3), and Rac family small GTPase 2 (RAC2) are found to be associated with T1D and AITD by various independent genome wide association studies and overlap in our list, indicating a strong common genetic link for T1D and AITD. As several susceptibility genes and environmental factors contribute to the disease aetiology of both T1D and AITD and/or AP subtype III variant (T1D+AITD) simultaneously, all patients with T1D should be screened for AITD, and vice versa.

Variants in the BACH2 and CLEC16A gene might be associated with susceptibility to insulin-triggered type 1 diabetes

AIM/INTRODUCTION

Insulin administration was found to trigger type 1 diabetes in six Japanese type 2 diabetes patients with type 1 diabetes high-risk human leukocyte antigen class II and the class I allele of the insulin gene variable number tandem repeat genotype. The objective of the present study was to assess the contribution of non-human leukocyte antigen single-nucleotide polymorphisms (SNPs) to the risk of developing insulin-triggered type 1 diabetes.

MATERIALS AND METHODS

We genotyped 13 type 1 diabetes susceptible SNPs in six patients and compared them with those in Japanese controls (Hap Map3-JPT). The SNPs that showed statistically significant results were further analyzed using non-diabetic control participants and participants with type 2 diabetes at the Ehime University Hospital.

RESULTS

The risk allele frequency of BACH2 rs3757247 in the six patients was significantly more frequent than that in 86 Japanese controls (P = 0.038). No significant difference in the allele frequency was observed in the other SNPs. This result was confirmed by the findings that the risk allele frequency of BACH2 in the six patients was significantly higher than that in the non-diabetic control participants (n = 179) and type 2 diabetes with or without insulin treatment (n = 154 or n = 152; P = 0.035, 0.034 or 0.037, respectively). Despite being statistically not significant, the six patients were all homozygous for the CLEC16A rs12708716 risk allele and five were homozygous for the CLEC16A rs2903692 risk allele.

CONCLUSIONS

In addition to type 1 diabetes high-risk human leukocyte antigen class II and the class I allele of the insulin gene variable number tandem repeat genotype, the possibility that the risk variants of BACH2 and CLEC16A could contribute to the development of insulin-triggered type 1 diabetes cannot be excluded.

Early prediction of autoimmune (type 1) diabetes

Underlying type 1 diabetes is a genetic aetiology dominated by the influence of specific HLA haplotypes involving primarily the class II DR-DQ region. In genetically predisposed children with the DR4-DQ8 haplotype, exogenous factors, yet to be identified, are thought to trigger an autoimmune reaction against insulin, signalled by insulin autoantibodies as the first autoantibody to appear. In children with the DR3-DQ2 haplotype, the triggering reaction is primarily against GAD signalled by GAD autoantibodies (GADA) as the first-appearing autoantibody. The incidence rate of insulin autoantibodies as the first-appearing autoantibody peaks during the first years of life and declines thereafter. The incidence rate of GADA as the first-appearing autoantibody peaks later but does not decline. The first autoantibody may variably be followed, in an apparently non-HLA-associated pathogenesis, by a second, third or fourth autoantibody. Although not all persons with a single type of autoantibody progress to diabetes, the presence of multiple autoantibodies seems invariably to be followed by loss of functional beta cell mass and eventually by dysglycaemia and symptoms. Infiltration of mononuclear cells in and around the islets appears to be a late phenomenon appearing in the multiple-autoantibody-positive with dysglycaemia. As our understanding of the aetiology and pathogenesis of type 1 diabetes advances, the improved capability for early prediction should guide new strategies for the prevention of type 1 diabetes.

HLA class II alleles in Norwegian patients with coexisting type 1 diabetes and celiac disease

BACKGROUND

Type 1 diabetes (T1D) and celiac disease (CeD) are 2 distinct diseases, but there is an increased risk of developing CeD for T1D patients. Both diseases are associated with HLA-class II alleles, such as DQB1 *02:01 and DQB1 *03:02; however, their risk contribution vary between the diseases.

MATERIALS AND METHODS

We genotyped HLA-DRB1 and - DQB1 in 215 patients with coexisting T1D and CeD identified from a T1D cohort, and compared them to patients with T1D (N = 487) and CeD (N = 327), as well as healthy controls (N = 368).

RESULTS

The patients with coexisting T1D and CeD had an intermediate carrier frequency (72.8%) of the DRB1 *03:01- DQB1 *02:01- DQA1 *05:01 haplotype compared to T1D (64.1%) and CeD (88.7%) patients. The DRB1 *03:01- DQB1 *02:01- DQA1 *05:01/ DRB1 *04- DQB1 *03:02- DQA1 *03 haplotype combination, encoding DQ2.5 and DQ8 molecules, was equally frequent among patients with both T1D and CeD (52.6%) and T1D patients (46.8%) but significantly lower in CeD patients (9.5%).

CONCLUSION

Overall, the patients with coexisting T1D and CeD had an HLA profile more similar to T1D patients than CeD patients.

Clinical worthlessness of genetic prediction of common forms of diabetes mellitus and related chronic complications: A position statement of the Italian Society of Diabetology

AIM

We are currently facing several attempts aimed at marketing genetic data for predicting multifactorial diseases, among which diabetes mellitus is one of the more prevalent. The present document primarily aims at providing to practicing physicians a summary of available data regarding the role of genetic information in predicting diabetes and its chronic complications.

DATA SYNTHESIS

Firstly, general information about characteristics and performance of risk prediction tools will be presented in order to help clinicians to get acquainted with basic methodological information related to the subject at issue. Then, as far as type 1 diabetes is concerned, available data indicate that genetic information and counseling may be useful only in families with many affected individuals. However, since no disease prevention is possible, the utility of predicting this form of diabetes is at question. In the case of type 2 diabetes, available data really question the utility of adding genetic information on top of well performing, easy available and inexpensive non-genetic markers. Finally, the possibility of using the few available genetic data on diabetic complications for improving our ability to predict them will also be presented and discussed. For cardiovascular complication, the addition of genetic information to models based on clinical features does not translate in a substantial improvement in risk discrimination. For all other diabetic complications genetic information are currently very poor and cannot, therefore, be used for improving risk stratification.

CONCLUSIONS

In all, nowadays the use of genetic testing for predicting diabetes and its chronic complications is definitively of little value in clinical practice.

HLA-DQ genotypes - but not immune markers - differ by ethnicity in patients with childhood onset type 1 diabetes residing in Belgium

AIM

The aim of this study was to compare genetic (HLA-DQ) and immune markers in a large population of type 1 diabetic (T1D) children and adolescents residing in the same environment, but of different ethnic origin: European Caucasians (EC), Moghrabin Caucasians (MC), Black Africans (BA) and of Mixed Origin (MO).

METHODS

Retrospective study, including 452 patients with T1D aged 0.1-17.5 yr at diagnosis recruited at the Diabetology Clinic of the University Children's Hospital Queen Fabiola from May 1995 to March 2013. HLA-DQ genotyping, diabetes-associated autoantibodies, organ-specific autoantibodies, and other markers of autoimmunity were studied.

RESULTS

The proportion of the different ethnic groups was: 55% EC, 35% MC, 6% BA, and 4% MO. Between these four groups, there were no significant differences concerning age, hemoglobin A1c (HbA1c), presence of diabetic ketoacidosis, random C-peptide level at diagnosis and 2 yr later. The two most frequent haplotypes were DQA1*0501-DQB1*0201 and DQA1*0301-DQB1*0302 with a significant higher prevalence in MC and EC (p = 0.002 and 0.03, respectively). The high-risk heterozygous genotype DQA1*0301-DQB1*0302/DQA1*0501-DQB1*0201 was more frequent in EC than in MC, whereas the homozygous genotype DQA1*0501-DQB1*0201/DQA1*0501-DQB1*0201 was more prevalent in MC (p = 0.019). These susceptible genotypes were more frequent in youngest patients (p = 0.003). Diabetes-associated autoantibodies, organ-specific autoantibodies, and other immune markers did not statistically differ between ethnic groups.

CONCLUSIONS

These observations in a large population of T1D children and adolescents of different ethnic groups residing in Belgium show significant differences in HLA-DQ status, but not in diabetes-associated autoantibodies, organ-specific autoantibodies, or other immune markers.

GCORE-sib: An efficient gene-gene interaction tool for genome-wide association studies based on discordant sib pairs

BACKGROUND

A computationally efficient tool is required for a genome-wide gene-gene interaction analysis that tests an extremely large number of single-nucleotide polymorphism (SNP) interaction pairs in genome-wide association studies (GWAS). Current tools for GWAS interaction analysis are mainly developed for unrelated case-control samples. Relatively fewer tools for interaction analysis are available for complex disease studies with family-based design, and these tools tend to be computationally expensive.

RESULTS

We developed a fast gene-gene interaction test, GCORE-sib, for discordant sib pairs and implemented the test into an efficient tool. We used simulations to demonstrate that the GCORE-sib has correct type I error rates and has comparable power to that of the regression-based interaction test. We also showed that the GCORE-sib can run more than 10 times faster than the regression-based test. Finally, the GCORE-sib was applied to a GWAS dataset with approximately 2,000 discordant sib pairs, and the GCORE-sib finished testing 19,368,078,382 pairs of SNPs within 6 days.

CONCLUSIONS

An efficient gene-gene interaction tool for discordant sib pairs was developed. It will be very useful for genome-wide gene-gene interaction analysis in GWAS using discordant sib pairs. The tool can be downloaded for free at http://gcore-sib.sourceforge.net .

Insulin gene VNTR polymorphisms -2221MspI and -23HphI are associated with type 1 diabetes and latent autoimmune diabetes in adults: a meta-analysis

AIMS

A variable number of tandem repeat (VNTRs) region in the insulin gene (INS) possibly influences the progression of type 1 diabetes (T1D) and latent autoimmune diabetes in adults (LADA). However, effects of INS VNTR polymorphisms in these contexts remain inconclusive.

METHODS

We performed a systematic review of work on the INS VNTR -2221MspI and -23HphI polymorphisms to estimate the overall effects thereof on disease susceptibility; we included 17,498 T1D patients and 24,437 controls, and 1960 LADA patients and 5583 controls.

RESULTS

For T1D, the C allele at -2221MspI and the A allele at -23HphI were associated with estimated relative risks of 2.13 (95 % CI 1.94, 2.35) and 0.46 (95 % CI 0.44, 0.48), which contributed to absolute increases of 46.76 and 46.98 % in the risk of all T1D, respectively. The estimated lambda values were 0.44 and 0.42, respectively, suggesting that a co-dominant model most likely explained the effects of -2221MspI and -23HphI on T1D. For -23HphI, the A allele carried an estimated relative risk of 0.55 (95 % CI 0.50, 0.61) for LADA and increased the risk of all LADA by 36.94 %. The λ value was 0.43, suggesting that a co-dominant model most likely explained the effect of -23HphI on LADA.

CONCLUSIONS

Our results support the existence of associations of INS with T1D and LADA.

Saffold Virus, a Human Cardiovirus, and Risk of Persistent Islet Autoantibodies in the Longitudinal Birth Cohort Study MIDIA

The aim of this study was to describe the frequency and distribution of Saffold virus in longitudinal stool samples from children, and test for association with development of persistent autoantibodies predictive of type 1 diabetes. A cohort of Norwegian children carrying the HLA genotype associated with highest risk of type 1 diabetes (“DR4-DQ8/DR3-DQ2”) was followed with monthly stool samples from 3 to 35 months of age. Blood samples were tested for autoantibodies to insulin, glutamic acid decarboxylase₆₅ and Islet Antigen-2. 2077 stool samples from 27 children with ≥2 repeatedly positive islet autoantibodies (cases), and 53 matched controls were analysed for Saffold virus genomic RNA by semi-quantitative real-time reverse transcriptase PCR. Saffold virus was found in 53 of 2077 (2.6%) samples, with similar proportions between cases (2.5%) and controls (2.6%). The probability of being infected by 3 years of age was 28% (95% CI 0.18–0.40). Viral quantities ranged from <1 to almost 10⁵ copies/μl. Estimated odds ratio between islet autoimmunity and infection episodes prior to seroconversion was 1.98 (95% CI: 0.57–6.91, p = 0.29). Saffold virus had no statistically significant association with islet autoimmunity.

Improving prediction of type 1 diabetes by testing non-HLA genetic variants in addition to HLA markers

OBJECTIVE

The purpose of this study was to explore whether non-human leukocyte antigen (non-HLA) genetic markers can improve type 1 diabetes(T1D) prediction in a prospective cohort with high-risk HLA-DR,DQ genotypes.

METHODS

The Diabetes Autoimmunity Study in the Young (DAISY) follows prospectively for the development of T1D and islet autoimmunity (IA)children at increased genetic risk. A total of 1709 non-Hispanic White DAISY participants have been genotyped for 27 non-HLA single nucleotide polymorphisms (SNPs) and one microsatellite.

RESULTS

In multivariate analyses adjusting for family history and HLA-DR3/4 genotype, PTPN22 (rs2476601) and two UBASH3A (rs11203203 and rs9976767) SNPs were associated with development of IA [hazard ratio(HR)=1.87, 1.55, and 1.54, respectively, all p ≤ 0.003], while GLIS3 and IL2RA showed borderline association with development of IA. INS,UBASH3A, and IFIH1 were significantly associated with progression from IA to diabetes (HR=1.65, 1.44, and 1.47, respectively, all p ≤ 0.04), while PTPN22 and IL27 showed borderline association with progression from IA to diabetes. In survival analysis, 45% of general population DAISY children with PTPN22 rs2476601 TT or HLA-DR3/4 and UBASH3A rs11203203 AA developed diabetes by age 15, compared with 3% of children with all other genotypes (p<0.0001). Addition of non-HLA markers to HLA-DR3/4,DQ8 did not improve diabetes prediction in first-degree relatives.

CONCLUSION

Addition of PTPN22 and UBASH3A SNPs to HLA-DR,DQ genotyping can improve T1D risk prediction.

Assessment of type 1 diabetes risk conferred by HLA-DRB1, INS-VNTR and PTPN22 genes using the Bayesian network approach

BACKGROUND

Determining genetic risk is a fundamental prerequisite for the implementation of primary prevention trials for type 1 diabetes (T1D). The aim of this study was to assess the risk conferred by HLA-DRB1, INS-VNTR and PTPN22 single genes on the onset of T1D and the joint risk conferred by all these three susceptibility loci using the Bayesian Network (BN) approach in both population-based case-control and family clustering data sets.

METHODOLOGY/PRINCIPAL FINDINGS

A case-control French cohort, consisting of 868 T1D patients and 73 French control subjects, a French family data set consisting of 1694 T1D patients and 2340 controls were analysed. We studied both samples separately applying the BN probabilistic approach, that is a graphical model that encodes probabilistic relationships among variables of interest. As expected HLA-DRB1 is the most relevant susceptibility gene. We proved that INS and PTPN22 genes marginally influence T1D risk in all risk HLA-DRB1 genotype categories. The absolute risk conferred by carrying simultaneously high, moderate or low risk HLA-DRB1 genotypes together with at risk INS and PTPN22 genotypes, was 11.5%, 1.7% and 0.1% in the case-control sample and 19.8%, 6.6% and 2.2% in the family cohort, respectively.

CONCLUSIONS/SIGNIFICANCE

This work represents, to the best of our knowledge, the first study based on both case-control and family data sets, showing the joint effect of HLA, INS and PTPN22 in a T1D Caucasian population with a wide range of age at T1D onset, adding new insights to previous findings regarding data sets consisting of patients and controls <15 years at onset.

Pathogenesis of type 1 diabetes: lessons from natural history studies of high-risk individuals

Type 1 diabetes (T1D) is an autoimmune disease characterized by known genetic risk factors with T cell-mediated infiltration and destruction of the beta cells within pancreatic islets. Autoantibodies are the most significant preclinical marker of T1D, and birth cohort studies have provided important insights into the natural history of autoimmunity and T1D. While HLA remains the strongest genetic risk factor, a number of novel gene variants associated with T1D have been found through genome-wide studies, some of which have been linked to suspected environmental risk factors. Multiple environmental factors that have been suggested to play a role in the development of T1D await confirmation. Current risk-stratification models for T1D take into account genetic risk factors and autoantibodies. In the future, metabolic profiles, epigenetics, as well as environmental risk factors may be included in such models.

PTPN22 gene polymorphism (C1858T) is associated with susceptibility to type 1 diabetes: a meta-analysis of 19,495 cases and 25,341 controls

The protein tyrosine phosphatase N22 (PTPN22) gene C1858T polymorphism has been reported to be associated with susceptibility to type 1 diabetes (T1D) in relatively small sample sizes. This study aimed at investigating the pooled association by carrying out a meta-analysis on the published studies. The Medline, EBSCO, and BIOSIS databases were searched to identify eligible studies published in English before June 2012. The association was assessed by odds ratio (OR) with 95% confidence intervals (CI). The presence of heterogeneity and publication bias was explored by using meta-regression analysis and Begg's test, respectively. A total of 28 studies were involved in this meta-analysis. Across all populations, significant associations were found between the PTPN22 C1858T polymorphism and susceptibility to T1D under genotypic (TT vs. CC [OR = 3.656, 95% CI: 3.139-4.257], CT vs. CC [OR = 1.968, 95% CI: 1.683-2.300]), recessive (OR = 3.147, 95% CI: 2.704-3.663), and dominant models (OR = 1.957, 95% CI: 1.817-2.108). In ethnicity- and sex-stratified analyses, similar associations were found among Caucasians and within Caucasian male and female strata. The meta-analysis results suggest that the PTPN22 C1858T polymorphism was associated with susceptibility to T1D among the Caucasian population, and males who carried the -1858T allele were more susceptible to T1D than females.

Evidence of gene-gene interaction and age-at-diagnosis effects in type 1 diabetes

The common genetic loci that independently influence the risk of type 1 diabetes have largely been determined. Their interactions with age-at-diagnosis of type 1 diabetes, sex, or the major susceptibility locus, HLA class II, remain mostly unexplored. A large collection of more than 14,866 type 1 diabetes samples (6,750 British diabetic individuals and 8,116 affected family samples of European descent) were genotyped at 38 confirmed type 1 diabetes-associated non-HLA regions and used to test for interaction of association with age-at-diagnosis, sex, and HLA class II genotypes using regression models. The alleles that confer susceptibility to type 1 diabetes at interleukin-2 (IL-2), IL2/4q27 (rs2069763) and renalase, FAD-dependent amine oxidase (RNLS)/10q23.31 (rs10509540), were associated with a lower age-at-diagnosis (P = 4.6 × 10⁻⁶ and 2.5 × 10⁻⁵, respectively). For both loci, individuals carrying the susceptible homozygous genotype were, on average, 7.2 months younger at diagnosis than those carrying the protective homozygous genotypes. In addition to protein tyrosine phosphatase nonreceptor type 22 (PTPN22), evidence of statistical interaction between HLA class II genotypes and rs3087243 at cytotoxic T-lymphocyte antigen 4 (CTLA4)/2q33.2 was obtained (P = 7.90 × 10⁻⁵). No evidence of differential risk by sex was obtained at any loci (P ≥ 0.01). Statistical interaction effects can be detected in type 1 diabetes although they provide a relatively small contribution to our understanding of the familial clustering of the disease.

Serum long chain n-3 fatty acids (EPA and DHA) in the pregnant mother are independent of risk of type 1 diabetes in the offspring

BACKGROUND

This article aims to study whether higher proportions of the long chain n-3 fatty acids eicosapentaenoic acid (EPA) or docosahexaenoic acid (DHA) in the phospholipid fraction of serum samples in pregnancy were associated with a lower risk of childhood onset type 1 diabetes in the offspring.

METHODS

In a prospective cohort of nearly 30 000 pregnant women who gave birth in Norway during 1992-1994, we analysed serum samples from 89 women whose child developed type 1 diabetes and was included in the nationwide Norwegian Childhood Diabetes Registry and 125 randomly selected women whose child did not develop type 1 diabetes before 15 years of age. Specific fatty acids were expressed as the proportion of total fatty acids (g/100 g) in the phospholipid fraction in serum analysed using solid phase extraction and gas chromatography with flame ionization detection.

RESULTS

There was no significant association between EPA or DHA in maternal serum and risk of type 1 diabetes in the offspring. Odds ratio (OR) for upper versus lower quartile of EPA was 0.75 [95% confidence interval (CI) 0.34-1.65], test for trend p = 0.4, and for DHA OR = 0.71 (95% CI 0.33-1.53), test for trend p = 0.6. No significant association was found for the sum of n-3 fatty acids, or for n-6/n-3 ratio in the mother with risk of type 1 diabetes in the offspring.

CONCLUSIONS

Our data did not support the hypothesis that higher proportions of maternal EPA or DHA during pregnancy are associated with a lower risk of type 1 diabetes in the offspring.

Improving power of genome-wide association studies with weighted false discovery rate control and prioritized subset analysis

The issue of large-scale testing has caught much attention with the advent of high-throughput technologies. In genomic studies, researchers are often confronted with a large number of tests. To make simultaneous inference for the many tests, the false discovery rate (FDR) control provides a practical balance between the number of true positives and the number of false positives. However, when few hypotheses are truly non-null, controlling the FDR may not provide additional advantages over controlling the family-wise error rate (e.g., the Bonferroni correction). To facilitate discoveries from a study, weighting tests according to prior information is a promising strategy. A 'weighted FDR control' (WEI) and a 'prioritized subset analysis' (PSA) have caught much attention. In this work, we compare the two weighting schemes with systematic simulation studies and demonstrate their use with a genome-wide association study (GWAS) on type 1 diabetes provided by the Wellcome Trust Case Control Consortium. The PSA and the WEI both can increase power when the prior is informative. With accurate and precise prioritization, the PSA can especially create substantial power improvements over the commonly-used whole-genome single-step FDR adjustment (i.e., the traditional un-weighted FDR control). When the prior is uninformative (true disease susceptibility regions are not prioritized), the power loss of the PSA and the WEI is almost negligible. However, a caution is that the overall FDR of the PSA can be slightly inflated if the prioritization is not accurate and precise. Our study highlights the merits of using information from mounting genetic studies, and provides insights to choose an appropriate weighting scheme to FDR control on GWAS.

Effects of non-HLA gene polymorphisms on development of islet autoimmunity and type 1 diabetes in a population with high-risk HLA-DR,DQ genotypes

We assessed the effects of non-HLA gene polymorphisms on the risk of islet autoimmunity (IA) and progression to type 1 diabetes in the Diabetes Autoimmunity Study in the Young. A total of 1,743 non-Hispanic, white children were included: 861 first-degree relatives and 882 general population children identified as having high-risk HLA-DR/DQ genotypes for type 1 diabetes. Of those, 109 developed IA and 61 progressed to diabetes. Study participants were genotyped for 20 non-HLA polymorphisms, previously confirmed as type 1 diabetes susceptibility loci. PTPN22 and UBASH3A predicted both IA and diabetes in regression models controlling for family history of type 1 diabetes and presence of HLA-DR3/4-DQB1*0302 genotype. In addition, PTPN2 predicted IA whereas INS predicted type 1 diabetes. The final multivariate regression models for both IA and type 1 diabetes included PTPN22, UBASH3A, and INS, in addition to family history of type 1 diabetes and HLA-DR3/4. In general population children, the most frequent combinations including these five significant predictors conferred hazard ratio of up to 13 for IA and >40 for type 1 diabetes. Non-HLA susceptibility alleles may help estimate risk for development of type 1 diabetes in the general population. These findings require replication in different populations.

Statistical analysis of genetic interactions

Many common human diseases and complex traits are highly heritable and influenced by multiple genetic and environmental factors. Although genome-wide association studies (GWAS) have successfully identified many disease-associated variants, these genetic variants explain only a small proportion of the heritability of most complex diseases. Genetic interactions (gene-gene and gene-environment) substantially contribute to complex traits and diseases and could be one of the main sources of the missing heritability. This paper provides an overview of the available statistical methods and related computer software for identifying genetic interactions in animal and plant experimental crosses and human genetic association studies. The main discussion falls under the three broad issues in statistical analysis of genetic interactions: the definition, detection and interpretation of genetic interactions. Recently developed methods based on modern techniques for high-dimensional data are reviewed, including penalized likelihood approaches and hierarchical models; the relationships between these methods are also discussed. I conclude this review by highlighting some areas of future research.

HLA-DQ haplotypes differ by ethnicity in patients with childhood-onset diabetes

AIM

To understand the etiology of childhood-onset diabetes, we examined genetic risk markers, autoantibodies, and β-cell function in a mixed race group of young patients.

METHODS

One hundred and forty-five patients aged 0-17 at diagnosis (54% African American, 22% Caucasian, 16% Latino, 8% mixed-other) were studied at mean duration 6.9 ± 5.7 (range 0.1-28.5) yr, including human leukocyte antigen (HLA)-DQA1-DQB1 genotyping, stimulated C peptide (CP), glutamic acid decarboxylase, and insulinoma-associated antigen 2 antibodies (ABs). Based on no residual β-cell function (CP-) and islet autoantibodies (AB+), 111 patients were classified with type 1 diabetes mellitus (T1DM), 22 were CP+ and AB- and thus considered to have type 2 diabetes mellitus (T2DM), and 12 patients had features of both T1DM and T2DM or mixed phenotype.

RESULTS

Based on the presence of two high-risk HLA-DQA1/B1 haplotypes, 39% of African Americans, 81% of Caucasians, 70% of Latinos, and 67% of mixed-others were at high genetic risk. In patients with T1DM, 41% of African Americans, 80% of Caucasians, 73% of Latinos, and 63% of mixed-others were genetically susceptible. Thirty-one percent of African Americans, including 29% of those with T1DM, could not be characterized because their haplotypes had unknown T1DM associations. These unusual haplotypes comprised 11% in T1DM, 14% in T2DM, and 8% in patients with mixed phenotype.

CONCLUSIONS

Fifty-nine percent of childhood-onset patients with T1DM were identified with high genetic risk based on known HLA-DQA1/B1 associations. Many non-Caucasian patients carry HLA-DQ alleles whose association with T1DM is undetermined. Genetic approaches can provide insights into the etiology and appropriate treatment of childhood-onset diabetes but only if sufficient data are available in diverse ethnic groups.

Does the relative risk for type 1 diabetes conferred by HLA-DQ, INS, and PTPN22 polymorphisms vary with maternal age, birth weight, or cesarean section?

BACKGROUND AND OBJECTIVE

Maternal age at birth, birth weight, and cesarean section has been associated with a weak but significant increase in risk of type 1 diabetes. The objective was to assess whether the relative risk for type 1 diabetes conferred by established susceptibility loci human leukocyte antigen (HLA)-DQ, INS, and PTPN22 differed depending on these perinatal factors.

METHODS

We employed a case-control study with 456 cases of type 1 diabetes diagnosed before 15 yr of age and 1377 population-based control children. HLA genotypes were divided into high to moderate risk (DQ8/DQ2, DQ8/DQ8, DQ8/X, DQ2/DQ2) vs. all other genotypes. Case-only analysis using logistic regression was used to test for significant interaction.

RESULTS

There was no significant difference in the relative risks conferred by HLA-DQ, INS, or PTPN22 by maternal age, birth weight, or mode of delivery, except the relative risk conferred by PTPN22 which was 2.11 [95% confidence interval (CI): 1.64-2.72] for those born vaginally and 0.99 (95% CI: 0.50-1.99) for those born by cesarean section [p(interaction) = 0.028].

CONCLUSION

The relative risks conferred by the three established susceptibility genes investigated here were independent of the perinatal factors, apart from a possible interaction between PTPN22 and mode of delivery.

Type 1 diabetes associated with Hashimoto's thyroiditis and juvenile rheumatoid arthritis: a case report with clinical and genetic investigations

Autoimmune diseases are initiated by interaction between genetic and environmental factors and caused by the loss of immunologic tolerance to self-antigens. They cluster within families and individuals, but the aggregation in a triad is quite rare. We report a case of a young girl affected by three organ-specific autoimmune disorders, from which type 1 diabetes developed first, then Hashimoto's thyroiditis and juvenile rheumatoid arthritis were diagnosed. Hitherto unreported detailed genetic studies included genotyping of HLA class II, CTLA4, and PTPN22 gene regions. These genes have been associated with autoimmunity in general and some of their variants confer increased risk to all three diseases. Our results - with the limitation of reporting only on a single patient - contribute to the complex genetic background of these clustering organ-specific autoimmune diseases and the analysis of further similar cases might help to reveal how the major and minor genetic factors determine the individual clinical phenotype.

Novel gene associations in type 1 diabetes

Recent genome-wide association studies have been able to identify multiple new gene loci affecting type 1 diabetes susceptibility, but the impact of these new defined loci seems to decrease in parallel with their number. The HLA gene region remains the main nominator of genetic susceptibility, although the identity of important genes and especially the mechanisms of their action are still largely unclear. Products of HLA and most other known risk genes are involved in regulation of the immune system in accordance with the autoimmune nature of the disease. The multitude of genes involved in the pathogenesis implies complex pathways where multiple steps in each may be essential in turning the balance of immune response to beta-cell destructing autoimmunity.

Bayesian analysis of genetic interactions in case-control studies, with application to adiponectin genes and colorectal cancer risk

Complex diseases such as cancers are influenced by interacting networks of genetic and environmental factors. However, a joint analysis of multiple genes and environmental factors is challenging, owing to potentially large numbers of correlated and complex variables. We describe Bayesian generalized linear models for simultaneously analyzing covariates, main effects of numerous loci, gene-gene and gene-environment interactions in population case-control studies. Our Bayesian models use Student-t prior distributions with different shrinkage parameters for different types of effects, allowing reliable estimates of main effects and interactions and hence increasing the power for detection of real signals. We implement a fast and stable algorithm for fitting models by extending available tools for classical generalized linear models to the Bayesian case. We propose a novel method to interpret and visualize models with multiple interactions by computing the average predictive probability. Simulations show that the method has the potential to dissect interacting networks of complex diseases. Application of the method to a large case-control study of adiponectin genes and colorectal cancer risk highlights the previous results and detects new epistatic interactions and sex-specific effects that warrant follow-up in independent studies.

An inverse association between history of childhood eczema and subsequent risk of type 1 diabetes that is not likely to be explained by HLA-DQ, PTPN22, or CTLA4 polymorphisms

BACKGROUND

Established genetic susceptibility loci for type 1 diabetes are important in immune regulation and may play a role also in atopic disorders, potentially explaining the inverse association between childhood eczema and subsequent risk for type 1 diabetes previously reported.

OBJECTIVE

We aimed to directly assess whether HLA-DQ, CTLA4, and PTPN22 genes could explain the putative association between childhood eczema and lower subsequent risk of type 1 diabetes observed in several case-control studies.

METHODS

We designed a case-control study with 339 incident cases of type 1 diabetes identified in the Norwegian childhood diabetes registry, and 985 population-based control children. DNA was collected, and physician-diagnosed childhood eczema was ascertained by a questionnaire administered to the parents of children with and without type 1 diabetes.

RESULTS

The previously reported association between childhood eczema and lower risk of type 1 diabetes was confirmed (odds ratio,OR, 0.61, 95% confidence interval, CI, 0.40-0.95] and this was consistent in subgroups defined by HLA-DQ, CTLA4, and PTPN22 genotypes. The OR was essentially not influenced by adjustment for genetic variation at these loci (OR simultaneously adjusted for the three genetic loci: 0.55, 95% CI: 0.32-0.92). The ratio of the unadjusted to adjusted OR was 1.12, with a corresponding 95% CI from 0.84 to 1.50.

CONCLUSION

In this first study of its kind, we demonstrated directly that the observed inverse association between childhood eczema and type 1 diabetes is not likely to be explained by the established diabetes susceptibility genes HLA-DQ, CTLA4, or PTPN22.

Banting Lecture 2009: An unfinished journey: molecular pathogenesis to prevention of type 1A diabetes

The Banting Medal for Scientific Achievement Award is the American Diabetes Association's highest scientific award and honors an individual who has made significant, long-term contributions to the understanding of diabetes, its treatment, and/or prevention. The award is named after Nobel Prize winner Sir Frederick Banting, who codiscovered insulin treatment for diabetes.

Dr. Eisenbarth received the American Diabetes Association's Banting Medal for Scientific Achievement at the Association's 69th Scientific Sessions, June 5–9, 2009, in New Orleans, Louisiana. He presented the Banting Lecture, An Unfinished Journey—Type 1 Diabetes—Molecular Pathogenesis to Prevention , on Sunday, June 7, 2009.

Using lifetime risk estimates in personal genomic profiles: estimation of uncertainty

Personal genome tests are now offered direct-to-consumer (DTC) via genetic variants identified by genome-wide association studies (GWAS) for common diseases. Tests report risk estimates (age-specific and lifetime) for various diseases based on genotypes at multiple loci. However, uncertainty surrounding such risk estimates has not been systematically investigated. With breast cancer as an example, we examined the combined effect of uncertainties in population incidence rates, genotype frequency, effect sizes, and models of joint effects among genetic variants on lifetime risk estimates. We performed simulations to estimate lifetime breast cancer risk for carriers and noncarriers of genetic variants. We derived population-based cancer incidence rates from Surveillance, Epidemiology, and End Results (SEER) Program and comparative international data. We used data for non-Hispanic white women from 2003 to 2005. We derived genotype frequencies and effect sizes from published GWAS and meta-analyses. For a single genetic variant in FGFR2 gene (rs2981582), combination of uncertainty in these parameters produced risk estimates where upper and lower 95% simulation intervals differed by more than 3-fold. Difference in population incidence rates was the largest contributor to variation in risk estimates. For a panel of five genetic variants, estimated lifetime risk of developing breast cancer before age 80 for a woman that carried all risk variants ranged from 6.1% to 21%, depending on assumptions of additive or multiplicative joint effects and breast cancer incidence rates. Epidemiologic parameters involved in computation of disease risk have substantial uncertainty, and cumulative uncertainty should be properly recognized. Reliance on point estimates alone could be seriously misleading.

Genome-wide association study and meta-analysis find that over 40 loci affect risk of type 1 diabetes

Type 1 diabetes (T1D) is a common autoimmune disorder that arises from the action of multiple genetic and environmental risk factors. We report the findings of a genome-wide association study of T1D, combined in a meta-analysis with two previously published studies. The total sample set included 7,514 cases and 9,045 reference samples. Forty-one distinct genomic locations provided evidence for association with T1D in the meta-analysis (P < 10(-6)). After excluding previously reported associations, we further tested 27 regions in an independent set of 4,267 cases, 4,463 controls and 2,319 affected sib-pair (ASP) families. Of these, 18 regions were replicated (P < 0.01; overall P < 5 × 10(-8)) and 4 additional regions provided nominal evidence of replication (P < 0.05). The many new candidate genes suggested by these results include IL10, IL19, IL20, GLIS3, CD69 and IL27.

Mannose binding lectin gene polymorphisms and associated auto-immune diseases in type 1 diabetes Brazilian patients

In our study we investigated the possible role of MBL2 functional single nucleotide polymorphisms (SNPs) in the augmented susceptibility to develop other autoimmune diseases in presence of type 1 diabetes (T1D) in a group of Brazilian patients. Patients were stratified for the presence of autoimmune diseases known to be associated with T1D, such as autoimmune thyroid disease (AITD) and celiac disease (CD), and compared with healthy controls (HC). Our findings suggest that MBL2 functional SNPs are more closely related to AITD than to T1D, being MBL2 SNPs frequencies in T1D patients not affected by AITD comparable to the HC ones, while significantly different between AITD patients and patients not affected by the disease. Thus, the association between MBL2 polymorphisms and T1D that we previously reported, seems to result from the stronger association of MBL2 SNPs with another autoimmune disease, the AITD, frequently associated with T1D.

A TLR2 polymorphism is associated with type 1 diabetes and allergic asthma

Type 1 diabetes (T1D) and allergic asthma are immune-mediated diseases. Pattern recognition receptors are proteins expressed by cells in the immune system to identify microbial pathogens and endogenous ligands. Toll-like receptors (TLRs) and CD14 are members of this family and could represent a molecular link between microbial infections and immune-mediated diseases. Diverging hypotheses regarding whether there exists a common or inverse genetic etiology behind these immune-mediated diseases have been presented. We aimed to test whether there exist common or inverse associations between polymorphisms in the pattern recognition receptors TLR2, TLR4 and CD14 and T1D and allergic asthma. Eighteen single nucleotide polymorphisms (SNPs) were genotyped in TLR2 (2), TLR4 (12) and CD14 (4) in 700 T1D children, 357 nuclear families with T1D children and 796 children from the 'Environment and Childhood Asthma' study. Allele and haplotype frequencies were analyzed in relation to diseases and in addition transmission disequilibrium test analyses were performed in the family material. Both T1D and allergic asthma were significantly associated with the TLR2 rs3804100 T allele and further associated with the haplotype including this SNP, possibly representing a susceptibility locus common for the two diseases. Neither TLR4 nor CD14 were associated with T1D or allergic asthma.

A coding polymorphism in NALP1 confers risk for autoimmune Addison's disease and type 1 diabetes

Variants in the gene encoding NACHT leucine-rich-repeat protein 1 (NALP1), an important molecule in innate immunity, have recently been shown to confer risk for vitiligo and associated autoimmunity. We hypothesized that sequence variants in this gene may be involved in susceptibility to a wider spectrum of autoimmune diseases. Investigating large patient cohorts from six different autoimmune diseases, that is autoimmune Addison's disease (n=333), type 1 diabetes (n=1086), multiple sclerosis (n=502), rheumatoid arthritis (n=945), systemic lupus erythematosus (n=156) and juvenile idiopathic arthritis (n=505), against 3273 healthy controls, we analyzed four single nucleotide polymorphisms (SNPs) in NALP1. The major allele of the coding SNP rs12150220 revealed significant association with autoimmune Addison's disease compared with controls (OR=1.25, 95% CI: 1.06-1.49, P=0.007), and with type 1 diabetes (OR=1.15, 95% CI: 1.04-1.27, P=0.005). Trends toward the same associations were seen in rheumatoid arthritis, systemic lupus erythematosus and, although less obvious, multiple sclerosis. Patients with juvenile idiopathic arthritis did not show association with NALP1 gene variants. The results indicate that NALP1 and the innate immune system may be implicated in the pathogenesis of many autoimmune disorders, particularly organ-specific autoimmune diseases.

Joint genetic susceptibility to type 1 diabetes and autoimmune thyroiditis: from epidemiology to mechanisms

Type 1 diabetes (T1D) and autoimmune thyroid diseases (AITD) frequently occur together within families and in the same individual. The co-occurrence of T1D and AITD in the same patient is one of the variants of the autoimmune polyglandular syndrome type 3 [APS3 variant (APS3v)]. Epidemiological data point to a strong genetic influence on the shared susceptibility to T1D and AITD. Recently, significant progress has been made in our understanding of the genetic association between T1D and AITD. At least three genes have been confirmed as major joint susceptibility genes for T1D and AITD: human leukocyte antigen class II, cytotoxic T-lymphocyte antigen 4 (CTLA-4), and protein tyrosine phosphatase non-receptor type 22. Moreover, the first whole genome linkage study has been recently completed, and additional genes will soon be identified. Not unexpectedly, all the joint genes for T1D and AITD identified so far are involved in immune regulation, specifically in the presentation of antigenic peptides to T cells. One of the lessons learned from the analysis of the joint susceptibility genes for T1D and AITD is that subset analysis is a key to dissecting the etiology of complex diseases. One of the best demonstrations of the power of subset analysis is the CTLA-4 gene in T1D. Although CTLA-4 showed very weak association with T1D, when analyzed in the subset of patients with both T1D and AITD, the genetic effect of CTLA-4 was significantly stronger. Gene-gene and genetic-epigenetic interactions most likely play a role in the shared genetic susceptibility to T1D and AITD. Dissecting these mechanisms will lead to a better understanding of the etiology of T1D and AITD, as well as autoimmunity in general.