Genetic heterogeneity of diabetes and HLA

Type 1 diabetes in diverse ancestries and the use of genetic risk scores

Over 75 genetic loci within and outside of the HLA region influence type 1 diabetes risk. Genetic risk scores (GRS), which facilitate the integration of complex genetic information, have been developed in type 1 diabetes and incorporated into models and algorithms for classification, prognosis, and prediction of disease and response to preventive and therapeutic interventions. However, the development and validation of GRS across different ancestries is still emerging, as is knowledge on type 1 diabetes genetics in populations of diverse genetic ancestries. In this Review, we provide a summary of the current evidence on the evolutionary genetic variation in type 1 diabetes and the racial and ethnic differences in type 1 diabetes epidemiology, clinical characteristics, and preclinical course. We also discuss the influence of genetics on type 1 diabetes with differences across ancestries and the development and validation of GRS in various populations.

Role of Type 1 Diabetes-Associated SNPs on Autoantibody Positivity in the Type 1 Diabetes Genetics Consortium: Overview

OBJECTIVE

Type 1 diabetes (T1D) arises from the autoimmune destruction of the β-cells of the pancreas, resulting in dependence on exogenously administered insulin for survival. Key biomarkers of the autoimmune process in T1D are the occurrence of autoantibodies directed against β-cells and other antigens. The Type 1 Diabetes Genetics Consortium (T1DGC) assembled collections to 1) discover genes that modify the risk of T1D, 2) conduct phenotyping related to risk, and 3) make available biologic and genetic resources for research. The goal of the T1DGC Autoantibody Workshop was to use T1DGC phenotypic, genotypic, and autoantibody data on affected sibling pair (ASP) families to discover genes accounting for variation in presence of autoantibodies.

RESEARCH DESIGN AND METHODS

The T1DGC provided the working groups with autoantibody and genetic data on 9,976 subjects from 2,321 ASP families. Data were distributed to numerous working groups for analyses of specific autoantibody subsets and targets.

RESULTS

Seven groups analyzed the joint autoantibody and genetic data within the ASP families. Six reports are provided in this collection, ranging from candidate gene analyses of selected autoantibodies to evaluation of regions of genetic variants associated with autoimmunity on the collection of autoantibodies.

CONCLUSIONS

Although selected variants in the available genes remain important genetic predictors for prevalence of T1D, other genes and nongenetic factors are expected to contribute to the initiation of islet autoimmunity, the first step in the pathogenesis of T1D.

HLA-associated susceptibility to type 2 (non-insulin-dependent) diabetes mellitus: the Wadena City Health Study

Epidemiologic data suggest that a parental history of Type 2 (non-insulin-dependent) diabetes mellitus increases the risk of Type 1 (insulin-dependent) diabetes in siblings of a Type 1 diabetes proband. This increase in risk is consistent with a shared genetic susceptibility between Type 1 and Type 2 diabetes. We have previously reported evidence that HLA-DR4-linked factors may represent a homogeneous subset of diabetes susceptibility. First, HLA-DR4 frequency was higher in Type 1 diabetic study subjects with a Type 2 diabetic parent than in Type 1 diabetic subjects whose parents were not diabetic. Second, a DR4-haplotype was transmitted from the Type 2 diabetic parent to the Type 1 offspring more often than expected. These data are consistent with the hypothesis that families with a Type 2 diabetic parent and Type 1 diabetic child, heavily determined by HLA-DR4 linked factors, may represent a homogeneous subset of diabetes susceptibility. In this report, we further explore the relationship between the high-risk HLA antigen (HLA-DR4) in study subjects with differing glycaemic status (National Diabetes Data Group criteria). In this community-based study, we find evidence that HLA-DR4 is increased in study subjects with Type 2 diabetes and may be a marker for Type 2 diabetes susceptibility.

Shared genetic susceptibility of type 1 (insulin-dependent) and type 2 (non-insulin-dependent) diabetes mellitus: contributions of HLA and haptoglobin

Epidemiologic data suggest that having a parent with Type 2 (non-insulin-dependent) diabetes mellitus increases the risk for Type 1 (insulin-dependent) diabetes in siblings of a Type 1 diabetes proband. This increase in risk is consistent with a shared genetic susceptibility between Type 1 diabetes and Type 2 diabetes. We contrast genetic risk factors in three sets of families, consisting of (1) a single Type 1 diabetic child (proband) and non-diabetic parents, (2) multiple Type 1 diabetic siblings and non-diabetic parents, and (3) at least one Type 1 diabetic child and at least one Type 2 diabetic parent. Previous studies have demonstrated that HLA region genes, which elevate the risk in Type 1 diabetes, have no significant effect with respect to the risk for developing Type 2 diabetes. An earlier report cited a contribution by the haptoglobin locus to genetic susceptibility for Type 2 diabetes. We provide evidence that a high risk HLA antigen (HLA-DR3) is decreased to a greater extent in Type 1 patients with a Type 2 parent than in Type 1 patients in which the parents are not diabetic. The role of HLA-DR4 is maintained in these families, with an unexpectedly significant increased rate of transmission of the HLA-DR4 allele from Type 2 parent to Type 1 offspring. The role of haptoglobin in these families does not appear to be important, either with respect to association with diabetes or with respect to linkage with a secondary susceptibility locus. These results indicate that families with a Type 2 parent and Type 1 child, heavily determined by HLA-DR4 linked factors, may represent a homogeneous subset of diabetes susceptibility.

HLA antigens in Spanish type 1 diabetic population. Correlations with clinical, biological and autoimmune markers

The HLA haplotype and its relationships with clinical, biological and immunological parameters were analyzed in a group of 87 Spanish type 1 diabetic patients at the clinical onset of the disease. The frequency of HLA-B18, DR3 and DR4 antigens was significantly increased whereas DR2, DR5 and DR7 were decreased in comparison with 189 healthy unrelated controls without family history of diabetes. DR3 showed a maximum relative risk for diabetes (5.5) whereas DR4 had a lower one (4.0). HLA-DR4 patients were younger at the time of diagnosis than DR4 negative (16.7 vs 21.4 years). We found no statistically significant relationship between HLA antigens and the other variables studied including the presence of islet cell antibodies, complement fixing islet cell antibodies, insulin autoantibodies, organ-specific antibodies, fasting and maximal glucagon stimulated C-peptide levels, initial glycemia and glycosylated hemoglobin.

The Swedish childhood diabetes study--results from a nine year case register and a one year case-referent study indicating that type 1 (insulin-dependent) diabetes mellitus is associated with both type 2 (non-insulin-dependent) diabetes mellitus and autoimmune disorders

From July 1, 1977 to July 1, 1986, 3,503 incident cases of Type 1 (insulin-dependent) diabetes mellitus were registered in the Swedish childhood diabetes study. Using data from this register and from a case-referent study, including all incident Type 1 diabetic children in Sweden during one year and, for each patient, two referent children matched according to age, sex and county, we have studied the associations between Type 1 diabetes and familial Type 1 and Type 2 (non-insulin-dependent) diabetes, thyroid, adrenal, allergic, rheumatic, heart and bowel disease. The mean annual incidence per 100,000 during the nine year period was 25.1 for boys and 23.5 for girls. In 8.5% of the patients, one parent had Type 1 diabetes, 73% of whom were fathers. Fifty-six of the patients (1.7%) had a parent with Type 2 diabetes. The prevalence of parental Type 1 diabetes tended to be higher in patients with younger age at onset; whereas, the opposite was found for patients with parental Type 2 diabetes. In the case-referent study, the age-adjusted odds ratio for Type 1 diabetes when a first and/or second degree relative had Type 1 diabetes was 5.5 (95% confidence limits 4.0-7.7), and in accordance with the findings of the case register, the odds ratio tended to be highest in patients with the youngest age at onset. Season at onset of the patients was not associated with parental Type 1 diabetes. The odds ratio for Type 1 diabetes was significantly increased 3.3 (95% confidence limits: 2.3-4.6) when Type 2 diabetes was reported in relatives (three generations).(ABSTRACT TRUNCATED AT 250 WORDS)

HLA-DQ rather than HLA-DR region might be involved in dominant nonsusceptibility to diabetes

Since HLA-DRw15 (a subdivision of the HLA-DR2 specificity previously called DR2 long) is associated with dominant nonsusceptibility to insulin-dependent diabetes mellitus (IDDM), while HLA-DRw16 (another subdivision of HLA-DR2, previously called DR2 short) is positively associated with the disease, we looked for particular characteristics of HLA products encoded by the DR2 haplotypes of IDDM patients. The results show the following: (i) HLA-DQ molecules of HLA-DRw15-positive IDDM patients are different from those of HLA-DRw15-positive controls, suggesting that the HLA-DQ gene of DRw15 haplotypes is involved in a protective effect. (ii) HLA-DR and -DQ products of DRw16-positive IDDM are functionally indistinguishable from those of HLA-DRw16-positive controls. Furthermore, our data provide evidence that the residue at position 57 on the DQ beta chain could play a crucial biological role in antigen presentation to T cells as far as the DRw16 haplotype is concerned. This observation fits with the recent observation of correlation between DQ beta allelic polymorphism at position 57 and both susceptibility and resistance to IDDM.

HLA A,B,C and DR association with insulin-dependent diabetes in Martinique

HLA-A,B,C, and DR frequencies have been determined in 34 Coloured Martinican IDDM patients to establish the HLA and IDDM associations. HLA A3, B15, B18, Cw3 and DR4 antigens associations with IDDM are confirmed by this study. We found an increase of B21 similar to that found in Asiatic Indians. As in some African Black populations and in Cape coloured people, A1, B8, and DR3 are not increased in our population. We should point out that our patients' ages of onset were low, and that some studies have found DR4 association in young patients and DR3 in older ones. The protective role of DR2 is confirmed here. B35 and Cw4 negative associations have been found. We have observed that the antigens associated with IDDM are decreased in our control population, except DR4, and that the negative associated DR/ and Cw4 antigens are increased compared to the Continental French population. This corresponds with the low IDDM incidence in Blacks and Coloured people.

HLA-DR4 in insulin-dependent diabetic parents and their diabetic offspring: a clue to dominant inheritance

Insulin-dependent diabetes mellitus (IDDM) susceptibility determinants are known to be associated with both HLA-DR3 and -DR4. We monitored the inheritance of HLA-DR alleles in 37 families in which IDDM affected one parent and at least one offspring in order to try to learn more about the modes of inheritance of IDDM determinants. Ninety-seven insulin-dependent diabetics whose parents did not have diabetes and 158 nondiabetics were used as control groups for estimates of DR allele frequencies in the overall diabetic and general populations. The proportion of diabetic parents who transmitted DR4 to diabetic offspring (78%) was significantly higher (P less than 0.001) than the gene frequency of DR4 in the overall diabetic population (43%). The proportion of nondiabetic parents who transmitted DR4 to diabetic offspring (22%) was not significantly different from the gene frequency of DR4 in the nondiabetic population (16%), but it was significantly lower (P less than 0.05) than the gene frequency in the overall IDDM population. These proportions suggest that inheritance of the DR4-associated IDDM susceptibility determinant is not recessive, because in recessive inheritance expression of a trait depends on each parent contributing a susceptibility determinant. The proportions of diabetic and nondiabetic parents who transmitted the DR allele associated with the susceptibility determinant would then equal one another. The transmission of predominantly DR4 from affected parents to affected offspring suggests that susceptibility to IDDM is inherited primarily via a single dose of a potent determinant associated with DR4, as in dominant inheritance. When DR3 was transmitted at all it was usually by the nondiabetic parent. Only 8% of diabetic parents transmitted DR3 but 35% of nondiabetic parents transmitted DR3. The proportion of nondiabetic parents who transmitted DR3 was similar to the gene frequency of DR3 in the overall diabetic population (29%), but it was significantly higher than the gene frequency of DR3 in the nondiabetic population (15%; P less than 0.005). The percentage of diabetic offspring with the genotype DR3DR4 (35%) was identical to the percentage of individuals in the overall IDDM population with this genotype (35%). Numerous population data indicate that the DR3DR4 genotype carries a higher relative risk for IDDM than any other genotype, which suggests synergism between the DR3- and DR4-associated determinants. The family data reported here support this synergism but suggest that the DR4-associated determinant can give substantial susceptibility independent of the DR3-associated determinant and that the DR3-associated determinant is often expressed as enhancing susceptibility in the presence of the dominant DR4- associated determinant.

Analysis of negative and multiple HLA antigen disease associations

The nature and extent of negative and multiple HLA antigen disease associations are investigated theoretically under two models. The first model assumes that an HLA antigen is involved directly in predisposing individuals to disease. The second model assumes that the association of a particular HLA antigen(s) with a disease is the result of linkage disequilibrium between the allele determining the antigen and alleles at a nearby locus which confers susceptibility to disease. We determined whether observed decreases in antigen frequencies among a patient group are simply the inevitable result of the fact that if one or more alleles at a locus is increased in frequency, then others must be decreased. Under the antigen predisposing model exact predictions concerning allele and antigen class frequencies at the predisposing locus, and the non-predisposing loci, are given. The predictions are examined using HLA-DR data for multiple sclerosis.

Will a three-allele model of inheritance explain the HLA data for type 1 (insulin-dependent) diabetes?

The HLA data from nine published studies on Type 1 (insulin-dependent) diabetes were examined to see whether a three-allele model for the inheritance of Type 1 diabetes at the HLA-associated locus could be rejected. None of the data rejected the three-allele model. The data were also examined to see whether they would reject a recessive model. Out of the nine data sets, five rejected a recessive and four did not. The p value for all studies together rejected a recessive. Two of the data sets allowed us to test the hypothesis that multiplex and simplex families would exhibit different modes of inheritance. Multiplex data from both data sets rejected recessive inheritance while the multiplex data from only one data set also rejected three-allele inheritance. The results of assuming a recessive model and analyzing the data from simplex families led to different results from the two data sets. In addition, data from a non-European population were examined and found to reject both recessive and three-allele inheritance for Type 1 diabetes at the HLA-associated locus.

Insulin-dependent diabetes--associated HLA-D region encoded determinants

We have studied the relative frequency of Dw specificities (defined with homozygous typing cells or primed LD (lymphocyte defined) typing reagents) associated with DR4 and DR2 in the normal and insulin-dependent diabetic population. Our findings demonstrate that there is a highly significantly increased frequency of Dw4 in DR4 positive diabetics as compared with normals and a significantly decreased frequency of Dw2 and Dw12 in the few DR2 positive insulin-dependent diabetics that we have found. In addition, we have used PLT reagents to define a new LD specificity, LD-MN2, that is associated with DR2 and is found significantly more frequently in DR2+ IDD patients than in DR2+ normals. These results suggest that determinants of import in the association between HLA-D and IDD may be more closely related to Dw than to DR.

Segregation of HLA-DR2 among affected and non-affected offspring of 66 families with type 1 (insulin-dependent) diabetes

Segregation of HLA-DR2 among affected and unaffected offspring was studied in 66 HLA-genotypes families with Type 1 diabetes in whom at least one parent carried DR2. The frequency of DR2-positive parents (21%) was not different from that of control families (29%). Among the diabetic probands, the gene frequency of DR2 was significantly decreased compared with control subjects (0.05 versus 0.17, p less than 0.001) as were DR5 (0.07 versus 0.17, p less than 0.01) and DR7 (0.06 versus 0.13, p less than 0.003). Twenty probands carried DR2, in 11 or whom (55%) it was found in combination with either DR3 or DR4. The nine cases who carried another DR allele included one who was DR2 homozygous. Transmission of DR2 was reduced in affected offspring, and random in unaffected siblings, compared with the expected ratio. However, when the DR2 transmission was analysed separately for parents bearing DR2 with DR3, DR4 or another DR allele, it appeared that DR2 transmission to affected offspring was random when the parents carried neither DR3 or DR4, the transmission deficit being due to over-transmission of DR3 and DR4. The haplotype analysis showed that the haplotype A3, Cw7, B7, GfS, DR2, found in 19% of "non-diabetic" DR2 haplotypes was practically absent among "diabetic" DR2-haplotypes (4%). In conclusion, population and segregation analysis could not demonstrate a specific protective effect of DR2.

A solution to the genetic and environmental puzzles of insulin-dependent diabetes mellitus

Studies of the segregation of heterozygous immunoglobulin allotypes in families with several cases of insulin-dependent diabetes mellitus (IDDM) show that germline heavy-chain V (variable region) genes are not major genetic determinants for IDDM, but data for IDDM and Graves' disease together suggest involvement of kappa light-chain V genes. Absence of IDDM at birth, the semi-random age of onset, and the 50% discordance of identical twins suggest that somatic mutation of germline V genes is involved in the development of the pathogenetic anti-beta-cell clones. The effect of histocompatibility and other alloantigens on the prevalence of IDDM is readily accounted for by the effect of the "holes" they induce, by natural tolerance, in the immune response repertoire; these alterations apparently affect the chance of emergence of anti-beta-cell clones by the somatic mutations and network of interclonal deletions that constantly change the fringes of the repertoire. Histocompatibility antigens can also influence repertoire development by changing the specificity of conjoint presentation of foreign antigens by macrophages. Antigenic stimulation by particular environmental microorganisms is probably essential to the repertoire development necessary for the occurrence of IDDM. Additionally, beta-cell damage by local infection may play a part by facilitating autoantigen presentation to the immune system.

HLA and disease

This review describes the greater portion of a large number of new studies on HLA and disease association which has appeared in the literature since 1979. The majority of these are concerned with the association of certain diseases with class II major histocompatibility complex antigens. The possible biologic significance of these associations in terms of their probable etiology appears to be the prevailing theme. Current thinking regarding certain heritable diseases is described. It seems that although much has been done to resolve the genetics of insulin-dependent diabetes mellitus, other diseases such a multiple sclerosis still remain a mystery. Doubtlessly, much will be gained from DNA cloning and sequencing studies proposed for the future. A great deal of new information has been obtained relative to HLA itself. New loci have been postulated in the HLA-D/DR region through the use of powerful immunochemical procedures made possibly by the advent of modern technological advances. The impact of these developments on our understanding of the function of the MHC in man and its possible relationship to disease are discussed.

The Pittsburgh Insulin-Dependent Diabetes Mellitus (IDDM) study. HLA antigens and haplotypes as risk factors for the development of IDDM in IDDM patients and their siblings

The relationships between HLA antigens, sex, age at diagnosis, season of onset and insulin-dependent diabetes mellitus (IDDM) were studied in a consecutive admissions series of newly-diagnosed IDDM patients at Children's Hospital of Pittsburgh. In agreement with the findings of others, the strongest positive associations between IDDM and HLA antigens were seen with DR3 and DR4 (odds ratios (OR) of 3.5 and 4.4, respectively), while a very strong negative association was observed with DR2 (OR of 0.1). Male patients were significantly more likely than female patients to possess DR3 while female patients were significantly more likely to be DR4+. No consistent relationships were found between HLA antigens and either age at diagnosis or season of onset. Using a life table approach, the cumulative risk of IDDM by age 24 in HLA-identical siblings of IDDM patients was estimated to be 10.3%. This risk was significantly greater than the risks to either HLA-haploidentical siblings (2.2%) or to HLA-nonidentical siblings (1.0%), whose risks were not significantly different. B7 + HLA-identical siblings of patients appeared to be protected from the effects of HLA-identicality--their cumulative risk by age 24 was estimated to be only 2.3%.

A two-susceptibility-allele model for genetic diseases and associated marker loci: differences and similarities to a one-s-allele model

A model for genetic diseases and associated markers is defined where two distinct susceptibility alleles are possible, each associated with a different marker allele. Marker genotype distributions in a disease population are then expressed in terms of haplotype frequencies and penetrance parameters. It is shown that, if the heterozygote with two different disease alleles has a higher penetrance than the two disease homogzygotes, the observed to 'Hardy-Weinberg-expected' ratio of associated marker genotypes (the alpha/beta ratio of Falk, Mendell & Rubinstein, 1983) will always be greater than or equal to one. When all disease penetrances are equal, the model becomes indistinguishable from a recessive one-s-allele model with alpha/beta = 1. Application of these observations to several data sets for insulin dependent diabetes mellitus suggests the possibility that different marker genotype distributions in different samples may be due to different penetrances of the disease genotypes in the samples. If a particular environment causes the heterozygote disease genotype (with two different disease alleles) to have the highest penetrance, the marker genotype distribution would be compatible with the 2-s-allele model. In other environments where the three disease genotypes have essentially equal penetrances, the marker distribution would be compatible with the 1-s-allele model.

A general model for disease-marker association

A general model for analysing disease-marker associations from a random sample of patients and controls is given, assuming an arbitrary number of marker and disease susceptibility alleles. A method for testing the goodness-of-fit of various disease susceptibility models to the observed distribution of genotypes at the marker locus in patient and control samples is given. The method is demonstrated using a recently published data set on type I diabetes.

A search for heterogeneity in insulin dependent diabetes mellitus (IDDM): HLA and autoimmune studies in simplex, multiplex and multigenerational families

HLA antigens (A, B, C and DR loci), serum islet cell antibodies, thyrogastric antibodies, and insulin antibodies were studied in 77 families (25 simplex, 42 multiplex, and 10 multigenerational). In order to test for intrafamilial constancy and intergroup variation, we compared simplex with multiplex families, HLA identical and non identical siblings within families, as well as groups of families characterized by different DR alleles (DR3, DR4, and DR3/DR4) for various immunologic and clinical characteristics. These comparisons did not reveal all the distinct subgroups suggested by different cross-sectional population studies, but did provide evidence for a compound form having an aggregation of different high risk alleles. This study suggests that in many cases (and possibly especially in families with multiple affected individuals), there are several different genetic influences leading to IDDM.

HLA and complement allotypes in Type 1 (insulin-dependent) diabetes

A group of patients with Type 1 (insulin-dependent) diabetes mellitus was investigated for HLA-A, B and DR antigens as well as C4 and factor B polymorphism. A significant excess of DR3/DR4 heterozygotes was observed (27% versus 17% by Hardy-Weinberg expectation). The factor B allele BfF1 was present in 13% of patients with Type 1 diabetes (gene frequency of 0.08 versus 0.01 in control subjects). A rare C4 B allele, C4 B2.9, was found in 18% of patients with Type 1 diabetes (n = 63) compared with 1.1% of control subjects (n = 176). Total C4 deficiency at the C4 A locus (C4 AQ0,0) was present in 10% of patients with Type 1 diabetes compared with 0% of control subjects. Examination of HLA, C4 and Bf phenotypes in patients with Type 1 diabetes suggested that three high risk supratypes, HLA-A1 B8 BfS C4 AQ0 C4 B1 DR3; HLA-B18 BfF1 C4 A3 C4 BQ0 DR3; HLA-A2 CW3 BW62 BfS C4 A3 C4 B2.9 DR4 are markers for susceptibility alleles.

HLA and susceptibility to type I diabetes. Hypothesis

Positive associations between the antigens D(R)3 and D(R)4 and negative associations of D(R)2 have been reported with insulin dependent or type I diabetes mellitus. It has been suggested that susceptibility factor(s) associated with the D(R)3 and D(R)4 haplotypes and a resistance factor associated with the D(R)2 haplotype may be involved in the pathogenesis of the disease. We propose a hypothesis herein which attempts to unify these findings based on our present understanding suggesting (a) the existence of multiple antigenic determinants associated with any one D(R) haplotype and (b) the sharing of "single" D region encoded determinants between what we now refer to as different D(R) haplotypes. The hypothesis, in its simplest form, focuses on a single D region determinant which can be found associated at different frequencies with the various D haplotypes as potentially explaining the findings.

Heterogeneity of insulin-dependent diabetes-new evidence

We have performed complex segregation and linkage analysis in 182 families with at least one insulin-dependent diabetic proband. All families were typed for B histocompatibility (HLA) antigens and 118 for DR. The recessive model fit the data best, with maximum likelihood estimates of recombination between HLA DR and the susceptibility factor of 0.019. Substantial heterogeneity was suggested, with smallest estimated recombination for pedigrees whose probands have two high-risk DR alleles. The results are compatible with a strong, tightly HLA-linked susceptibility factor and evidence for additional non-HLA linked genetic factor(s).