Family-based association study of HLA class II with type 1 diabetes in Moroccans

Does HLA explain the high incidence of childhood-onset type 1 diabetes in the Canary Islands? The role of Asp57 DQB1 molecules

The Canary Islands inhabitants, a recently admixed population with significant North African genetic influence, has the highest incidence of childhood-onset type 1 diabetes (T1D) in Spain and one of the highest in Europe. HLA accounts for half of the genetic risk of T1D.

AIMS

To characterize the classical HLA-DRB1 and HLA-DQB1 alleles in children from Gran Canaria with and without T1D.

METHODS

We analyzed classic HLA-DRB1 and HLA-DQB1 alleles in childhood-onset T1D patients (n = 309) and control children without T1D (n = 222) from the island of Gran Canaria. We also analyzed the presence or absence of aspartic acid at position 57 in the HLA-DQB1 gene and arginine at position 52 in the HLA-DQA1 gene. Genotyping of classical HLA-DQB1 and HLA-DRB1 alleles was performed at two-digit resolution using Luminex technology. The chi-square test (or Fisher's exact test) and odds ratio (OR) were computed to assess differences in allele and genotype frequencies between patients and controls. Logistic regression analysis was also used.

RESULTS

Mean age at diagnosis of T1D was 7.4 ± 3.6 years (46% female). Mean age of the controls was 7.6 ± 1.1 years (55% female). DRB1*03 (OR = 4.2; p = 2.13-13), DRB1*04 (OR = 6.6; p ≤ 2.00-16), DRB1* 07 (OR = 0.37; p = 9.73-06), DRB1*11 (OR = 0.17; p = 6.72-09), DRB1*12, DRB1*13 (OR = 0.38; p = 1.21-05), DRB1*14 (OR = 0.0; p = 0.0024), DRB1*15 (OR = 0.13; p = 7.78-07) and DRB1*16 (OR = 0.21; p = 0.003) exhibited significant differences in frequency between groups. Among the DQB1* alleles, DQB1*02 (OR: 2.3; p = 5.13-06), DQB1*03 (OR = 1.7; p = 1.89-03), DQB1*05 (OR = 0.64; p = 0.027) and DQB1*06 (OR = 0.19; p = 6.25-14) exhibited significant differences. A total of 58% of the studied HLA-DQB1 genes in our control population lacked aspartic acid at position 57.

CONCLUSIONS

In this population, the overall distributions of the HLA-DRB1 and HLA-DQB1 alleles are similar to those in other European populations. However, the frequency of the non-Asp-57 HLA-DQB1 molecules is greater than that in other populations with a lower incidence of T1D. Based on genetic, historical and epidemiological data, we propose that a common genetic background might help explain the elevated pediatric T1D incidence in the Canary Islands, North-Africa and middle eastern countries.

Family-based association of HLA-DRB1 and DQB1 alleles and haplotypes in a group of Iranian Type 1 diabetes children

This family-based study was conducted in a group of Iranians with Type 1 diabetes (T1D) to investigate the transmission from parents of risk and non-risk HLA alleles and haplotypes, and to estimate the genetic risk score for this disease within this population. A total of 240 T1D subjects including 111 parent-child trios (111 children with T1D, 133 siblings, and 222 parents) and 330 ethnically matched healthy individuals were recruited. High-resolution HLA typing for DRB1/DQB1 loci was performed for all study subjects (n = 925) using polymerase chain reaction-sequence-specific oligonucleotide probe method. The highest predisposing effect on developing T1D was conferred by the following haplotypes both in all subjects and in probands compared to controls: DRB1*04:05-DQB1*03:02 (Pc = 2.97e-06 and Pc = 6.04e-10, respectively), DRB1*04:02-DQB1*03:02 (Pc = 5.94e-17 and Pc = 3.86e-09, respectively), and DRB1*03:01-DQB1*02:01 (Pc = 8.26e-29 and Pc = 6.56e-16, respectively). Conversely, the major protective haplotypes included DRB1*13:01-DQB1*06:03 (Pc = 6.99e-08), DRB1*15:01-DQB1*06:02 (Pc = 2.97e-06) in the cases versus controls. Also, DRB1*03:01-DQB1*02:01/DRB1*04:02|05-DQB1*03:02 and DRB1*03:01-DQB1*02:01/DRB1*03:01-DQB1*02:01 diplotypes conferred the highest predisposing effect in the cases (Pc = 8.65e-17 and Pc = 6.26e-08, respectively) and in probands (Pc = 5.4e-15 and Pc = 0.001, respectively) compared to controls. Transmission disequilibrium test showed that the highest risk was conferred by DRB1*04:02-DQB1*03:02 (Pc = 3.26e-05) and DRB1*03:01-DQB1*02:01 (Pc = 1.78e-12) haplotypes and the highest protection by DRB1*14:01-DQB1*05:03 (Pc = 8.66e-05), DRB1*15:01-DQB1*06:02 (Pc = 0.002), and DRB1*11:01-DQB1*03:01 (Pc = 0.0003) haplotypes. Based on logistic regression analysis, carriage of risk haplotypes increased the risk of T1D development 24.5 times in the Iranian population (p = 5.61e-13). Also, receiver operating characteristic curve analysis revealed a high predictive power of those risk haplotypes in discrimination of susceptible from healthy individuals (area under curve: 0.88, p = 5.5e-32). Our study highlights the potential utility of genetic risk assessment based on HLA diplotypes for predicting T1D risk in individuals, particularly among family members of affected children in our population.

Segregation Analysis of Genotyped and Family-Phased, Long Range MHC Classical Class I and Class II Haplotypes in 5 Families With Type 1 Diabetes Proband in the United Arab Emirates

The classical Human Leucocyte Antigen (HLA) class II haplotypes of the Major Histocompatibility Complex (MHC) that are associated with type 1 diabetes (T1D) were identified in five families from the United Arab Emirates (UAE). Segregation analyses were performed on these 5 families with the disease, 3 with one child and 2 with 2 children diagnosed with T1D. Three HLA-DR4 haplotypes were identified: HLA- DRB1^∗04:01:01-DQB1^∗03:02:01:01; HLA- DRB1^∗04:02:01- DQB1^∗03:02:01; and HLA -DRB1^∗04:05:01-DQB1^∗02:02:01:02. All have previously been identified to be associated with T1D in studies of the Arabian population. In the 10 parents from the 5 families, 9 had at least one HLA-DR4 and HLA-DR3 haplotype which potentially increases the risk of T1D. Of these 9 parents, 3 were heterozygous for HLA-DR4/HLA-DR3 and one was homozygous for HLA-DR3. Two haplotypes that were identified here extend to the HLA class I region were previously designated AH8.2 (HLA -A^∗26-B^∗08-DRB1^∗03) and AH50.2 (HLA -C^∗06-B^∗50-DRB1^∗03:01-DQ^∗02) and associated with diabetes in neighboring North Indian populations. This study provides examples of MHC haplotype analysis in pedigrees to improve our understanding of the genetics of T1D in the understudied population of the UAE.

Research of anti-GAD and anti-IA2 autoantibodies by ELISA test in a series of Moroccan pediatric patients with diabetes type 1

Introduction

Type I diabetes (T1D) is an autoimmune disease with a prediabetic, asymptomatic period characterized by the selective destruction of insulin-producing β cells. During the pre-clinical phase, various auto-antibodies are generated against several beta cell antigens such as anti glutamate acid decarboxylase (Anti-GAD), anti tyrosine phosphatase (Anti-IA2). Today, the coupled detection of Anti-IA2 with that of Anti-GAD proves its great importance in the diagnosis and prediction of type 1 diabetes. The combined positivity for both antibodies has a specificity and a positive predictive value of 100%.

Objectives

In this work, we evaluate the diagnostic value of anti-GAD and anti-IA2 antibodies in a series based on 78 Moroccan subjects initially under 16, suspected T1D.

Results and Discussion

Our series consists mainly of 74% of newly diagnosed patients for T1D and 26% of confirmed diagnostic patients, of whom 52% are females. The mean age of diagnosis is 7 ± 4 years, the mean of HbA1c at the time of diagnosis is 11.63 ± 2.16%, and the percentage of family history in our series is 69%. The proportion of positive results for anti-IA2 antibodies and anti-GAD antibodies are, respectively, 76.92% and 62.82%, and 52.56% of patients are positive for both auto-antibodies. This study confirms that anti-GAD and anti-IA2 auto-antibodies assays can detect patients early and the autoantibodies can persist several years after diagnosis of type 1 diabetes.

Conclusion

This study confirmed the diagnosis and classification of T1D (type 1A) in 87.18% of patients, and we reported that the prevalence of anti-GAD and anti-IA2 is higher in girls than in boys.

Diverse human leukocyte antigen association of type 1 diabetes in north India

BACKGROUND

Type 1 diabetes (T1D) is a complex disease, with involvement of various susceptibility genes. Human leukocyte antigen (HLA) on chromosome 6p21 is major susceptibility region. This study examined genetic association of HLA genes with T1D.

METHODS

The study recruited 259 T1D patients and 706 controls from north India. PCR-SSP and LiPA were used to type HLA Class I and II alleles.

RESULTS

At HLA Class I locus, HLA-A*02, A*26, B*08 and B*50 were significantly increased in patients vs controls (39.8% vs 28.9% [Bonferroni-corrected P {P_c } = 0.032], 24.7% vs 9.6% [P_c  = 4.83 × 10^-8 ], 37.2% vs 15.7% [P_c  = 1.92 × 10^-9 ], and 19.4% vs 5.5% [P_c  = 4.62 × 10^-9 ], respectively). Similarly, in Class II region, DRB1*03 showed a strong positive association with T1D (78.7% vs 17.5% in controls; P = 1.02 × 10^-9 ). Association of DRB1*04 with T1D (28.3% vs 15.5% in controls; P_c  = 3.86 × 10^-4 ) was not independent of DRB1*03. Negative associations were found between T1D and DRB1*07, *11, *13, and *15 (13.8% vs 26.1% in controls [P_c  = 0.00175], 3.9% vs 16.9% in controls [P_c  = 6.55× 10^-6 ], 5.5% vs 21.6% in controls [P_c  = 2.51 × 10^-7 ], and 16.9% vs 43.9% in controls [P_c  = 9.94× 10^-10 ], respectively). Compared with controls, patients had significantly higher haplotype frequencies of A*26-B*08-DRB1*03-DQA1*05-DQB1*02 (10.43% vs 1.96%; P = 7.62 × 10^-11 ), A*02-B*50-DRB1*03-DQA1*05-DQB1*02 (6.1% vs 0.71%; P = 2.19 × 10^-10 ), A*24-B*08-DRB1*03-DQA1*05-DQB1*02 (4.72% vs 0.8%; P = 5.4 × 10^-7 ), A*02-B*08-DRB1*03-DQA1*05-DQB1*02 (2.36% vs 0.18%; P = 3.6 × 10^-5 ), and A*33-B*58-DRB1*03-DQA1*05-DQB1*02 (4.33% vs 1.25%; P = 0.00019).

CONCLUSIONS

In north India, T1D is independently associated only with HLA-DRB1*03 haplotypes, and is negatively associated with DRB1*07, *11, *13, and *15.

Association of HLA-DRB1 and -DQB1 alleles with type 1 (autoimmune) diabetes in African Arabs: systematic review and meta-analysis

Several studies confirmed the association of HLA-DRB1 and -DQB1 alleles with altered risk of type 1 diabetes (T1D). However, data from individual studies based on small sample sizes yielded often conflicting findings in African Arabs. This is a systematic review and meta-analysis aimed at comprehensively evaluating this association with T1D, using molecular HLA data. Relevant studies were identified through systemic search of Medline/PubMed, Cochrane, Science Direct, ResearchGate, and EMBASE databases. Statistical analysis was carried out using RevMan, and Comprehensive Meta-analysis programs. Given the heterogeneity of African Arabs, we also performed subgroup analysis according to ethnicity. Analysis of sensitivity, heterogeneity, and publication bias were performed to validate the outcome of the findings. This meta-analysis included 862 T1DM cases, along with 1,390 normoglycemic control, and comprised ten comparisons. Our study indicates that DRB1*03 (OR = 2.86), DRB1*04 (OR = 2.78), and DQB1*02 (OR = 2.29), are positively associated with increased risk of T1DM, while DRB1*07 (OR = 0.48), DRB1*11 (OR = 0.20), DRB1*13 (OR = 0.47), DRB1*15 (OR = 0.30), DQB1*05 (OR = 0.39), and DQB1*06 (OR = 0.27) were negatively associated with T1D, suggesting a protective role against T1D. This meta-analysis was characterized by low heterogeneity, sensitivity, and publication bias, indicating the robustness and reliability of the results.

BACKGROUND

Several studies confirmed the association of HLA-DRB1 and -DQB1 alleles with altered risk of type 1 diabetes (T1D). However, data from individual studies based on small sample sizes yielded often conflicting findings in African Arabs. This is a systematic review and meta-analysis aimed at comprehensively evaluating this association with T1D, using molecular HLA data.

METHODS

Relevant studies were identified through systemic search of Medline/PubMed, Cochrane, Science Direct, ResearchGate, and EMBASE databases. Statistical analysis was carried out using Revman, and Comprehensive Meta-analysis programs. Given the heterogeneity of African Arabs, we also performed subgroup analysis according to ethnicity. Analysis of sensitivity, heterogeneity, and pub¬lication bias were performed to validate the outcome of the findings. This meta-analysis included 862 T1DM cases, along with 1,390 normoglycemic control, and comprised ten comparisons.

RESULTS

Our study indicates that DRB1*03 (OR = 2.86), DRB1*04 (OR = 2.78), and DQB1*02 (OR = 2.29), are positively associated with increased risk of T1DM, while DRB1*07 (OR = 0.48), DRB1*11 (OR = 0.20), DRB1*13 (OR = 0.47), DRB1*15 (OR = 0.30), DQB1*05 (OR = 0.39), and DQB1*06 (OR = 0.27) were negatively associated with T1D, suggesting a protective role against T1D.

CONCLUSION

This meta-analysis was characterized by low heterogeneity, sensitivity, and publication bias, indicating the robustness and reliability of the results.

Is HLA the cause of the high incidence of type 1 diabetes in the Canary Islands? Results from the Type 1 Diabetes Genetics Consortium (T1DGC)

INTRODUCTION

Incidence of childhood-onset type 1 diabetes mellitus in the Canary Islands is the highest reported so far in Spain, and among the highest worldwide. The HLA region accounts for approximately half the genetic risk of type 1 diabetes. Our aim was to assess distribution of high-risk and protective HLA haplotypes in the Canarian families included in the T1DGC, as compared to the rest of Spain.

METHODS

The T1DGC study, an international project to study the genetics and pathogenesis of type 1 diabetes, enrolled more than 3000 families with type 1 diabetes worldwide. Spain provided 149 of these families, of whom 42 were from Tenerife and Gran Canaria. HLA was genotyped centrally using a PCR-based, sequence-specific oligonucleotide probe system. Haplotypes were reconstructed using the deterministic algorithm alleHap in the R programming environment. Based on prior T1DGC results in Caucasian population, haplotypes DRB1*0405-DQA1*0301-DQB1*0302, DRB1*0401-DQA1*0301-DQB1*0302, DRB1*0301-DQA1*0501-DQB1*0201, DRB1*0402-DQA1*0301-DQB1*0302 and DRB1*0404-DQA1*0301-DQB1*0302 were considered high-risk. DRB1*0701-DQA1*0201-DQB1*0303, DRB1*1401-DQA1*0101-DQB1*0503, DRB1*1501-DQA1*0102-DQB1*0602, DRB1*1101-DQA1*0501-DQB1*0301, DRB1*1104-DQA1*0501-DQB1*0301, DRB1*1303-DQA1*0501-DQB1*0301, DRB1*1301-DQA1*0103-DQB1*0603 and DRB1*0403-DQA1*0301-DQB1*0302 were considered protective. The distribution of protective, high-risk, and other haplotypes in the (first two) affected siblings and unaffected parents from Canarian and non-Canarian Spanish families was compared (Chi-square test).

RESULTS

No significant differences were found between the regions in distribution of the HLA haplotypes in the affected siblings or in the non-affected parents.

CONCLUSIONS

The high incidence of childhood-onset type 1 diabetes in the Canarian population does not appear to be explained by a greater prevalence of high-risk class II HLA haplotypes in families with the disease. However, sample size limits the differences that can be detected in this study.