HLA class II alleles and haplotypes in Lithuanian children with type 1 diabetes and healthy children ( HLA and type 1 diabetes )

Erika Skrodenienė1, Dalia Marčiulionytė1, Žilvinas Padaiga2, Edita Jašinskienė3, Vaiva Sadauskaitė-Kuehne1, Carani B. Sanjeevi4, Johnny Ludvigsson5 Institute of Endocrinology, Kaunas University of Medicine, Department of Preventive Medicine, Kaunas University of Medicine, Department of Endocrinology, Kaunas University of Medicine, Lithuania, Department of Molecular Medicine, Karolinska Institute, Stockholm, Sweden, Division of Pediatrics and Diabetes Research Center, Department of Clinical and Experimental Medicine, Linkoping University, Sweden


Introduction
Type 1 diabetes mellitus (T1D) is an autoimmune disease caused by the selective destruction of the insulin-producing pancreatic beta cells (1).The initiation of this process may sometimes occur early in life.Genetic predisposition is an important part of the etiology.The genetic background of T1D is polygenic (2).The human leukocytes antigen (HLA) region on the short arm of chromosome 6p21.3contains genes that encode for class I, class II, and class III antigens.
Associations have been first established between T1D and HLA class I HLA-B8 and HLA-B15 genes, but the strongest genetic associations with T1D so far identified are the associations with HLA class II genes (3).HLA genes are thought to contribute to as much as 50% of the genetic risk for T1D.Several HLA genotypes have been associated with susceptibility and/or protection for T1D (4)(5)(6)(7).High-risk alleles, haplotypes, and genotypes as well as protective alleles, haplotypes, and genotypes have been identified.Among Caucasians, T1D is positively associated with DR3-DQ2 and DR4-DQ8 haplotypes and negatively associated with DR2-DQ6 (6,(8)(9)(10).Recent studies from different European countries have confirmed that the HLA DR3-DQ2/DR4-DQ8 genotype is associated with the highest diabetes risk (8,9,11).DQ8(DQA1*0301-DQB1*0302) (74%) and DQ2(DQA1*0501-DQB1*0201) (52%) or both are found in 89% of the Caucasian patients with T1D before the age of 15 years (12).It is possible that different genetic background may contribute to both different incidence and clinical manifestation of the disease.As an example DR3-and DR4-containing haplotypes are not associated with T1D in Southeast Asian communities, including Japan and Korea (13,14), where the incidence is much lower and clinical manifestation tends to differ (15)(16)(17).There is a big difference in the incidence of T1D between Lithuania and closely situated countries such as Nordic countries (17).In the previous study, we have also shown that clinical manifestations and prevalence of antibodies differ comparing Lithuania and Sweden (18).Thus, we found it important to determine the frequency of HLA class II alleles and haplotypes in Lithuanian children with diabetes and healthy children and compare these results with the figures seen in neighbor countries.

Material and methods
This study was a part of the case-control study mainly aimed at studying environmental factors important for the development of T1D.All 286 newly diagnosed T1D children during the period of August 1, 1996, and August 1, 2000, and 813 age-and sexmatched double randomly selected healthy controls participated in that study (18).Blood samples were obtained from children with diabetes as well as control children and stored at -20°C.
HLA was determined in 124 Lithuanian children with diabetes (55 males and 69 females; mean age, 9.2±3.9years) and compared with 78 Lithuanian controls (43 males and 35 females; mean age, 10.8±3.4 years).In both groups, the age varied between 0 and 15 years.
DNA was extracted from blood leukocytes by standard phenol-chloroform method (19).DNA was dissolved in sterile double distilled water.HLA DRB1, DQA1, and DQB1 alleles for diabetic children were genotyped using a polymerase chain reaction (PCR) with amplification of the second exon of the genes as described earlier (19).An amplified product was manually dot blotted onto nylon membranes.Synthetic sequence-specific oligonucleotide (SSO) probes were 3´-end-labeled with αP32-dCTP and used for hybridization followed by stringency washes and autoradiography (20).Laboratory analysis was carried out at the Department of Molecular Immunogenetics, Karolinska Institute, Stockholm, Sweden.
HLA DRB1, DQA1, and DQB1 alleles for control children were genotyped using the PCR with sequence specific primers (SSP-PCR) supplied by Protrans and following manufacture's recommendations (PRO-TRANS Medizinische Diagnostische Produkte GmbH, Germany).The amplified products were determined by means of agarose gel electrophoresis.Laboratory analysis was carried out at the Laboratory of Immunology and Genetics, Hospital of Kaunas University of Medicine, Kaunas, Lithuania.
The study was approved by the Research Ethics Committee of Kaunas University of Medicine, Lithuania.
Statistical analysis.Comparison of means between groups of cases and controls were performed by the Student's t test or Mann-Whitney U test (nonparametric values).Proportions were compared using chisquare or Fisher's exact test.Differences were considered significant at P<0.05.Odds ratios (OR) with 95% confidence intervals (CI) were calculated.Statistical package SPSS 13.0 for Windows release was used for the data analysis.
The distribution of HLA (DR)-DQA1-DQB1 haplotypes among Lithuanian children with T1D and nondiabetic control children (Table 4) matched closely with the Caucasoid individuals in other European countries.
For analysis, the children were divided into three groups according to their age: 0-4, 5-9, and 10-15 years old.There were no differences in the frequency of risk and protective haplotypes related to age and sex (data not shown).

Discussion
There is a great difference in the incidence of T1D between different countries, even neighboring countries such as Lithuania from the Baltic states and Sweden from the Nordic countries.Furthermore, the clinical manifestations seem to differ too (18).One explanation for this could that there is a different genetic background in the populations and/or different genetic traits among those who get T1D.were found significantly more frequently among children with diabetes than among control children in Lithuania.These alleles and haplotypes conferred the highest risk of the disease.Similar data were obtained in neighboring countries with rather low incidence of T1D -Estonia, Latvia, Poland, Russia (10,12,(21)(22)(23) -as well as in other countries in Europe -Sweden, Finland, France, Greece (9,(24)(25)(26), some of them having much higher incidence and somewhat different clinical manifestations.In some studies, (DR1)-DQA1*0101-04-DQB1*0501 has been found to be protective (22,25).We could not confirm these findings possibility due to a small sample size in our study.
A majority of studies have found that diabetic children have diabetes risk alleles and haplotypes (23,
We found no differences in the frequency of risk and protective haplotypes comparing males and females.

Conclusions
HLA class II haplotypes associated with type 1 diabetes mellitus positively or negatively were the same in Lithuanian children as in other European Caucasian populations, including the neighboring Nordic countries.Differences in incidence and clinical manifestations of type 1 diabetes might be due to different environmental factors and/or lifestyle.
HLA class II alleles and haplotypes in Lithuanian children with type 1 diabetes and healthy children

Table 1 .
Distribution of DRB1 alleles among children with type 1 diabetes (T1D) and control children

Table 2 .
Distribution of DQA1 alleles among children with type 1 diabetes (T1D) and control children

Table 3 .
Distribution of DQB1 alleles among children with type 1 diabetes (T1D) and control children