HLA haplotypes in Type 1 (insulin-dependent) diabetes mellitus: molecular analysis of the HLA-DQ locus

Diabetologia, Mar 1992

P. J. Tienari, E. Tuomilehto-Wolf, J. Tuomilehto, L. Peltonen, H. K. Åkerblom, A. Fagerlund, M. Flittner, B. Gustafsson, A. Hakulinen, L. Herva, et al.

A PDF file should load here. If you do not see its contents the file may be temporarily unavailable at the journal website or you do not have a PDF plug-in installed and enabled in your browser.

Alternatively, you can download the file locally and open with any standalone PDF reader:


HLA haplotypes in Type 1 (insulin-dependent) diabetes mellitus: molecular analysis of the HLA-DQ locus

Diabetologia H L A haplotypes in Type 1 (insulin-dependent) diabetes mellitus: molecular analysis o f the H L A - D Q locus P. J. Tienari 0 1 E. Tuomilehto-Wolf 0 1 J. Tuomilehto 0 1 L. Peltonen I 0 1 the D I M E Study Group 0 1 0 LaboratoryofMolecularGeneticsand z DepartmentofEpidemiology,NationalPublicHealth Institute , Helsinki , Finland 1 Dr. E J. Tienari National Public Health Institute Laboratory of Molecular Genetics Mannerheimintie 166 SF-00300 Helsinki Finland Summary. In Caucasians the predisposition to Type i (insulin-dependent) diabetes mellitus has been shown to associate with HLA-DR3,DQw2 and DR4,DQw8 and with the presence of amino acids other than aspartic acid at position 57 on the HLA-DQ[3 chain. In Finland the haplotypespecific absolute risk for developing Type i diabetes differs between various DR3 and DR4 positive haplotypes. The aim of our present analysis was to find out whether this variation is attributable to polymorphism at the DQ locus. As part of a nationwide prospective study including 757 serologically HLA genotyped families, we determined HLA-DQc~ and DQI3 restriction fragment polymorphisms in 17 selected families with important susceptibility haplotypes. Additionally, the DQA1 alleles were determined from 19 haplotypes using sequence-specific oligonucleotide probes, and the DQB1 second exon was sequenced from nine haplotypes. The DR3 as well as DR4 positive haplotypes frequently Type 1 (insulin-dependent) diabetes mellitus; HLA haplotypes; HLA-DQ; restriction fragment length polymorphism; genetics; disease susceptibility - 9 Springer-Verlag1992 Type i (insulin-dependent) diabetes mellitus is one of the diseases in which the study of the H L A system at the molecular level has provided new insights into disease susceptibility. However, controversyexists as to whether one or several H L A loci contribute in determining predisposition to Type i diabetes. After the initial findings ofpositive associations with HLA-B8, B15, and B18, stronger associations were found with DR3 and DR4 [ 1 ]. Restriction fragment length polymorphism (RFLP) analyses of the H L A class II loci have suggested that DQ alleles might be even more precise markers of susceptibilityto the disease than DR alleles [ 2, 3 ]. Based on nucleotide sequence data and oligonucleotide typingit has been proposed that aspartic acid at position 57 (Asp57) on the HLA-DQ[3 chain may protect against Type i diabetes in Caucasians [ 4, 5 ]. In Japanese, however, Asp57 seems to have no protective effect [ 6 ]. The H L A region spans over 3800 kilobases (kb) [ 7 ] and maps to chromosome 6 (band 6p21.3). It consists of several highly polymorphic loci, of which HLA-A, C and * seeAcknowledgements found in Type i diabetic patients showed no variation at the HLA-DQ locus, and they were DQw2 and DQw8, respectively. The absolute risk for Type 1 diabetes for DR4,DQw8 positive haplotypes A2,Cw4,Bw35,DR4 A3,Cw3,Bw62,DR4, A24,Cw7,Bw39,DR4, A2,Cw3,Bw62, DR4, and A2,Cw1,Bw56,DR4 was 35/100,000, 130/100,000, 166/100,000, 196/100,000, and 218/100,000, respectively. The absolute risks for DR3,DQw2 positive haplotypes A1, Cw7,B8,DR3 and A2,Cw7,B8,DR3 were 68/100,000 and 103/100,000, respectively. These results provide further evidence that not only the polymorphism at the DQ locus but also other genes of the haplotypes contribute to susceptibility to Type i diabetes. B loci code for class I antigens and HLA-DR,DQ, and DP for class II antigens. Class III constitutes several genes including genes for the complement components C2, C4 and Bf, which are located between class I and class II [ 8 ]. Many other loci are known and, additionally, numerous genes, which have not so far been characterized, exist in the H L A region [ 9, 10 ]. Linkage disequilibrium between the HLA-A,C,B,DR and DQ alleles greatly complicates any attempt to pinpoint the susceptibility gene(s) to a specific H L A locus. Consequently, it may be too simplistic to explain predisposition to Type 1 diabetes by the structure of the DQ molecules alone omitting the effect of other H L A gene products. Certain H L A haplotypes appear to be found particularly often in Type i diabetic subjects. In northern Europe, the haplotypes A1,Cw7,B8,DR3,DQw2 and A2,Cw3,Bw62,DR4,DQw8 are frequently found in patients with Type i diabetes. In Finland the haplotype A2,Cw1,Bw56,DR4 is the third most common haplotype in diabetic patients [ 11 ], and it has not so far been found in other populations. The absolute risk for Type 1 diabetes o f this h a p l o t y p e is t h e h i g h e s t o f all H L A h a p l o t y p e s f o u n d in F i n l a n d [ 12 ]. F i n l a n d h a s t h e h i g h e s t i n c i d e n c e o f T y p e 1 d i a b e t e s in t h e w o r l d [ 13 ], a n d t h e r e f o r e , t h e c o n s e r v e d a n t i g e n c o m b i n a t i o n o f t h i s p o p u l a t i o n - s p e c i f i c h a p l o t y p e m a y b e h i g h l y r e l e v a n t t o d i s e a s e s u s c e p t i b i l i t y . T h i s s t u d y w a s p a r t o f a n a t i o n w i d e p o p u l a t i o n - b a s e d p r o s p e c t i v e f a m i l y s t u d y o f T y p e i d i a b e t e s in F i n l a n d . O u t o f 757 s e r o l o g i c a l l y H L A g e n o t y p e d f a m i l i e s w e h a v e c h o s e n 17 f o r H L A - D Q l o c u s a n a l y s i s at t h e m o l e c u l a r level. W e w e r e p a r t i c u l a r i l y i n t e r e s t e d in d e f i n i n g t h e D Q l o c u s o f t h e n e w l y - f o u n d s u s c e p t i b i l i t y h a p l o t y p e A 2 , C w 1 , B w 5 6 , D R 4 . B y s e q u e n c i n g a n d o l i g o n u c l e o t i d e t y p i n g o f t h e D Q B 1 a n d D Q A 1 alleles, r e s p e c t i v e l y , a n d u s i n g t h e R F L P t e c h n i q u e w i t h DQc~ a n d DQ[3 c h a i n p r o b e s , w e h a v e t e s t e d w h e t h e r t h e d i f f e r e n c e s in h a p l o t y p e - s p e c i f i c a b s o l u t e r i s k [ 12 ] c o u l d b e a t t r i b u t a b l e to p o l y m o r p h i s m s at t h e D Q locus. Subjects and methods The families The Childhood Diabetes in Finland (DIME) study is a nationwide study into the genetic and environmental factors involved in Type 1 diabetes, which was carried out in Finland between September 1986 and April 1989. This study is the first population-based prospective family study which used H L A haplotypes as predictive markers for Type 1 diabetes. The genetic part of the study comprised 757 newlydiagnosed diabetic children aged 14 years or younger (probands). Together with their parents and siblings they were HLA-A,C,B,DR genotyped using conventional H L A serology [ 11 ]. From these genotyped families 17 were especiallyselected for this RFLP and sequence analysis. The families were selected for the presence of certain important H L A haplotypes such as the newlyfound susceptibility haplotype A2,Cw1,Bw56,DR4 [ 11 ]. Only families which were informative for the selected haplotypes were chosen; families where one parent was homozygous for a certain haplotype were excluded. Four of the familieswere multiplex families (Table 1). The four H L A haplotypes found in each family were divided into "diabetic" and "non-diabetic" haplotypes. "diabetic" haplotypes were the two haplotypes found in the diabetic probands. "Nondiabetic" haplotypes were defined as those parental haplotypes which were not found in probands or in parents or in siblings with Type i diabetes. The haplotype-specific absolute risk The haplotype-specificabsolute risk was calculated using the formula R = R1 (p2) + R2 (2pq) + R3 (q2), where R is the total incidence of Type i diabetes in the general population (age 0-14 years), which was 35/100,000 per year during 198%1988 in Finland [ 13 ]. R1 is the incidence in homozygotes for haplotype p, R2 the incidence in heterozygotes for haplotype p and haplotype q, and R3 the incidence in homozygotes for haplotype q. The R F L P analysis High molecular weight D N A was extracted from peripheral blood leucocytes using a standard method [ 14 ]. Six micrograms of genomic D N A was digested with the restriction endonucleases PstI, TaqI and BamHI (New England Biolabs, Beverly, Mass., USA) according to the manufacturers instructions. After digestion D N A was electophoresed in 0.6 % agarose gel for 16-20 h using 50 V. Hind III digested bacteriophage lamda D NA was run as a molecular weight marker in each gel. After an alkaline denaturation step D N A fragments were capillary-blotted overnight onto a nylon filter (Hybond-N, Amersham, Bucks., UK). Filterswere prehybridizedfor 6 h at 42 ~ 50 % deionized formamide, 6 % standard sodium citrate (SSC), 5 % Denhardt's solution (2 % Ficoll,2 % bovine serum albumin, 2 % polyvinylpyrrolidone),0.5 % sodium dodecyl sulphate (SDS) andherring sperm D N A (100 gg/ml). In the hybridization step a probe labelled with o~-32p-dCTP (Random prime, Boehringer, Mannheim, Germany) was added into the prehybridization solution and the tilters were hybridized overnight at 42 ~ The probes were cDNA probes for the DQc~andDQI3chains [ 15, 16 ].TheDQc~probewas usedinthe 10th Histocompatibility Workshop, and the DQ[3 probe was kindly provided by Dr. M. Trucco (University of Pittsburgh, Pa., USA). The tilters were washed in 2 x SSC at 25 ~ for 10 min, and at 60 ~ for 15min, in 2 x S S C + 0 . 1 SDS at 60~ for 15min, and finally in 0.3 SSC + 0.1 SDS at 60~ for 5-15 min. Filters were autoradiographed for 2 and 6 days. The restriction fragment sizes were determined as the mean size obtained from different blots. There was up to 10 % variationin fragment sizes between different blots. H L A - D Q RFLP fragments for each haplotype were determined by co-segregation of the fragments with the haplotype. The assignment of an RFLP fragment to a haplotype is clear when only one parent is positive for the fragment, and both of that parent's haplotypes are inherited by the children. It is also clear when both parents are positive, and at least one child is negative for the fragment. The assignment of a fragment to a haplotype is not certain when all family members are positive for the fragment. It is also uncertain when only one haplotype of a parent is inherited by the children. In the latter case it is possible that the non-inherited haplotype is positive for the fragment as well as the inherited haptotype. The H L A - D Q B1 sequence analysis Most of the variability of the DQB1 gene is in the second exon [ 8 ]. The HLA-DQB 1 second exon was amplifiedtwice using polymerase chain reaction (PCR) [ 17 ]. Prior to amplification genomic D N A was digested with restriction endonuclease BstEII (New England Binlabs) to avoid co-amplification of the DQB2 gene [ 18 ]. One microgram of DNA was amplified using a thermal cycler (Techne PHC1) under the following conditions: 0.2 mmol/l dNTPs, 20 mmol/1 TrisHC1, pHS.8, 15mmol/l (NH4)2804, 1.5mmol/1 MgC12, 0.1% Tween 20, 0.01% gelatin, 100 pmol of primers and 2.5 U of Taq D N A polymerase (Amplitaq, Perkin-Elmer Cetus, Emeryville, Calif., USA). Before adding the enzyme the reaction mixture was denaturated at 98 ~ for 5 rain and then kept at 80 ~ when enzyme was added. Thereafter, the denaturation step was 1 min at 95 ~ annealing 1 min at 60~ and extension 2 min at 70 ~ After 30 cycles there was an extension step 10 min at 70~ We used the primers GLPDQ[31 and GAMPDOX~2 [ 4 ] for the first amplification. The amplified product was run in a 3 % low melting point agarose gel (NuSieve, FMC, Rockland, Me., USA) and the 240 bp fragment was excised. The excised gel slice was diluted three-fold with distilled water, and 10 gl was amplified again using the same conditions as above except slightly modified primers with BamHI and HindIII sites: 5 ' - G A T T T C G T G G A T C C G T T r A A G - 3 ' and 5'-CCACCTCGAAGCTTTGTGTGCA-3'. The amplified product was run in a 3 % agarose gel (standard low-mr, Bio Rad, Richmond, Calif., USA). The 240 bp product was recovered from the gel using electroelution [ 14 ]. The amplified D N A was purified with phenol and chloroform and digested with BamHI and HindIII. Nucleotide sequences were determined after subcloning fragments in the pGem7Z (Promega, Madison, Wiss., USA) by the chain-termination method [ 19 ] (Sequenase 2.0, USB, Cleveland, Ohio, USA). The amplification, cloning and sequencing was performed two-three times for each subject to rule out possible errors made by Taq-polymerase. The most recent names for DQB1 alleles recommended by the WHO Nomenclature Committee were used [ 20 ]. The H L A - D Q A 1 oligonucleotide analysis The amplification and the analysis of the alleles were performed using the AmpliType H L A - D Q ~ Forensic D N A Amplification and Typing Kit (Perkin-Elmer Cetus, Emeryville, Calif., USA), which " Subject with Type 1 (insulin-dependent) diabetes, b query homozygote, ~ subject not RFLP typed, x, undefineable antigen. A2,Cwl,Bw56,DR4 found in six families; A3,Cw4,B35,DR1 and A2,Cw3,Bw62,DR4 found in four families; A1,Cw7,B8,DR3 and detects six different alleles [ 21 ]. Basically, the procedure involved amplification of the DNA segment coding for the outer domain of the HLA-D Qc~chain by PCR using biotinylated primers. The amplified product was subsequently hybridized to a filter carrying immobilized sequence-specific oligonucleotide DNA probes. Detection of the hybridization reaction was enzymatically mediated by a streptavidine-horseradish peroxidase conjugate resulting in a visually detectable dye. The nomenclature for the DQA1 alleles has been previously published [ 21 ]. Two subjects (a parent and a child) per family were typed to confirm the co-segregation of the DQA1 alleles with the serologically defined HLA haplotype. R e s u l t s T h e 68 p a r e n t a l H L A h a p l o t y p e s f o u n d in the 17 families are p r e s e n t e d in T a b l e 1. Since s o m e i m p o r t a n t h a p l o t y p e s w e r e s e l e c t e d to b e p r e s e n t in s e v e r a l families 47 d i f f e r e n t h a p l o t y p e s w e r e found. In these families 35 of 64 (55 % ) of A3,Cw3,Bw62,DR4 found in three families; A2,Cw7,B7,DR2, A3,Cw7,B7,DR2, A2,Cw6,B13,DR7, A3,Cw3,Bw62,DR8, and A3,Cw7,B7,DR14 found in two families the h a p l o t y p e s h a d i n f o r m a t i v e DQ[~/TaqI R F L P s , 44 of 60 ( 7 3 % ) DQ[~/PstI R F L P s , 18 o f 48 ( 3 8 % ) D Q [ ~ / B a m H I R F L P s , 33 o f 44 ( 7 5 % ) DQc~/TaqI R F L P s and 30 of 48 (63 % ) D Qc~/PstI R F L P s . O f the h a p l o t y p e s , 9 of 68 (13 % ) r e m a i n e d u n i n f o r m a t i v e w i t h all p r o b e / e n z y m e c o m b i n a tions and t h e r e f o r e n o f r a g m e n t s could b e assigned to t h e s e h a p l o t y p e s . T h e R F L P results of 20 i m p o r t a n t h a p l o t y p e s are summ a r i z e d in T a b l e 2. E x a m p l e s of a u t o r a d i o g r a m s using e a c h p r o b e / e n z y m e c o m b i n a t i o n are illustrated in Figure 1. T h e D R 3 p o s i t i v e h a p l o t y p e s s h o w e d n o v a r i a t i o n in D Q w 2 associated R F L P patterns, n e i t h e r did D R 7 positive haplotypes. Two families ( F a m i l y 6 a n d 15) h a d n o n - i n f o r m a t i v e R F L P s for D R 4 p o s i t i v e h a p l o types. C o n s e q u e n t l y , no f r a g m e n t s c o u l d be assigned to h a p l o t y p e s A 2 , C w 4 , B w 3 5 , D R 4 , A 2 8 , C w 3 , B w 6 0 , D R 4 , A 2 4 , C w 7 , B w 3 9 , D R 4 , A l l , C w 3 , B w 6 0 , D R 4 , and A 2 , a: c: bc bc bd a: c: bc ad a: c: ad a: c: ac ad a: c: bc ad a: c: bd bd ad a: c: bc bd a: c: ad ad ad a: c: ac ac 0 I O I I 0 0 o O l O I I 0 I o o 0 0 0 Cw1,Bw56,DR4 in these families. T h e absence of fragments associated with D Q w 7 (DQ~/PstI4.3 and DQI3/BamHI 3.55 [ 22 ]), and the presence of fragments DQ~/Pst111.0 kb, 6.8 kb, 5.1 kb and DQ[~/BamH111.0 kb in all members of these families suggest that these haplotypes carried the specificity DQw8. Thus, all eight D R 4 positive haplotypes in Table 3 were DQw8. T h r e e of the families (Families 7, 8, and 11) were selected for probands who where neither D R 3 nor D R 4 positive. Two of these probands were D R 7 , D Q w 2 / D R S , D Q w 4 , and one was D R 1 , D Q w 5 / D R 1 , D Q w 5 . T h e remaining DR1, DR2, D R w 8 and D R w l 0 positive haplotypes in these families showed R F L P patterns corresponding to DQw5, DQw6, D Q w 4 and DQw5, respectively. DRw5 and D R w 6 positive haplotypes showed considerable variation both in DQ(z and DQ[3 R F L P s (data not shown). T h e D Q A 1 alleles were typed using sequence-specific oligonucleotide probes f r o m 19haplotypes, and the D Q B 1 second exon was sequenced from nine haplotypes (Table 3). These included the newly-found susceptibility haplotype A2,Cw1,Bw56,DR4, the two most c o m m o n haplotypes in Finnish diabetic children, A1,Cw7,B8,DR3 and A2,Cw3,Bw62,DR4, and the two most frequent"non-diabetic" haplotypes in the Finnish population, A3,Cw4,B35,DR1 and A3,CwT,B7,DR2. All DR1, DR2, DR3, DR4, DR7, and D R 8 positive haplotypes carried D Q A 1 alleles D Q A I . 1 , DQA1.2, D Q A 4 , D Q A 3 , D Q A 2 , and D Q A 4 , respectively. T h e D Q B 1 sequences correlated exactly with previously published sequences [ 5 ]. The haplotype-specific absolute risks developing Type i diabetes are shown in Table 4. T h e absolute risks have b e e n calculated in a sample of 757 families with a newly-diagnosed diabetic child, altogether 1,424 unequivocally defined haplotypes among patients and 1,254 haplotypes among non-diabetic family members. Because the H L A haplotype data were obtained f r o m a population-based family study, we were able to calculate the absolute risk of Type 1 diabetes associated with each haplotype found in the probands. T h e r e was a considerable variation in absolute risk among different H L A haplotypes although similar D Q ~ and DQ~3 RFLPs, the same D Q A 1 alleles, and, in the case of the four D R 4 positive haplotypes (Table 3) the D Q B 1 exon 2 sequence was shown to be identical. D i s c u s s i o n Since the families were selected for this molecular study, the H L A haplotypes of these 17 families were clearly defined by serology. The DQo~ and DQ[3 RFLPs were less informative than the serologically defined haplotypes, as would be expected when comparing polymorphic information content of haplotypes vs single loci. Segregation of the serologically defined H L A haplotypes in the families provided a solid basis, which greatly helped the assignment of restriction fragments to each chromosome. TaqI and PstI were the most informative restriction enzymes with DQo~ and DQ~, respectively. Only reproducible and clearly visible fragments were included in the R F L P analysis, and, RJ.Tienariet al.:DQ locusin high-riskhaplotypesfor Type1diabetes grees above.Molecularsizesof restrictionfragments(in kilobases) are indicatedon the rightofeachautoradiogram for instance, the DQw8 associated DQ~/TaqI 2.5 kb fragment was not clearly detected and was therefore excluded. DQw8 was in these families most readily defined by the presence of DQ[3/PstI 11.0 kb, 6.8 kb, and 5.1 kb fragments, which differentiate DQw8 from DQw7 and DQw4 [ 23 ]. In DR1, DR2, DR3, DR4, DR7, DRw8, and DRwl0 positive haplotypes the RFLPs were in agreement with the 10th HistocompatibilityWorkshop RFLP standardization reports [ 22-26 ], with the exception that certain ambiguous fragments were exluded in our analysis. In addition to the HLA-DQ RFLPs, which are mainly based on intron polymorphism, we selected certain haplotypes to study exon polymorphism of the DQA1 and DQB1 genes. For the determination of DQB1 alleles we chose to sequence the DQB1 second exon instead of oligonucleotide typing, because sequencing allows possible new alleles to be more accurately detected. Subjects heterozygous for DQ were chosen for sequencing, which made it possible to assign the DQB1 sequence to one or the other haplotype. In this study we analysed HLA-DQot and DQ[3 RFLPs, DQA1 oligonucleotide hybridizations, and the DQB1 exon 2 sequences of several haplotypes. We then combined these data with the haplotype-specific absolute risks estimated from previous data [ 12 ].The large population-based sample allowed us to calculate haplotype-specific absolute risks and directly assess the true probability of the disease for subjects carrying a certain haplotype. In these families the most important DR3 and DR4 positive susceptibility haplotypes were DQw2 and DQw8, respectively. Interestingly, considerable variation could be seen in the absolute risk for developing Type 1 diabetes between haplotypes which were similar at the DQ locus. Striking differences in absolute risk were found between D R4 positive haplotypes (from 35 of 100,000 to 218 of 100,000) as well as between DR3 positive haplotypes (from 68 of 100,000 to 103 of 100,000). Unfortunately, a DR3 positive haplotype A28,Cw7,B8,BR3, which proved the highest absolute risk of all DR3 positive haplotypes (180 of 100,000) [ 12 ]was not included in the molecular analysis. The differences in haplotype-specific absolute risks suggest that there are other important genes in the H L A system, outside the DQ region, which modify the risk and make certain haplotypes more permissive to the disease. Finland has the highest incidence of Type 1 diabetes in the world [ 13 ], and since the Finnish population is genetically homogeneous [ 27 ], it is ideally suited for the study of genetic factors in this disease. A "new" Finnish susceptibility haplotype, A1,Cwl,Bw56,DR4, has been described, which is the third most common haplotype in patients with Type 1 diabetes in Finland [ 11 ]. This population-specific haplotype may provide a partially genetic explanation for E J. Tienari et al.:D Q locus in high-risk haplotypes for Type i diabetes the high incidence of the disease in Finland. Other haplotypes, which are frequently found in Finnish patients are typical susceptibility haplotypes found in northern E u r o p e such as A2,Cw3,Bw62,DR4 and A28,Cw3,Bw60,DR4. W h e n molecular analyses are carried out in unrelated subjects without H L A haplotype data, it is difficult to conclude whether the increased frequency of specific alleles is secondary to the increased frequency of certain haplotypes. A m o n g patients in the United States with Type 1 diabetes a 10 kb DR~/TaqI fragment has been reportedly increased [ 28 ]. This fragment has been shown to be in linkage disequilibrium with B8,DR3, and therefore serves as an additional haplotype m a r k e r rather than as a new marker for susceptibility to Type i diabetes [ 29 ]. In a French series the R F L P pattern defining D R 3 , D Q w 2 , D w 2 5 was most significantly associated with Type 1 diabetes but it was also associated with B18, and therefore primarily differentiates the southern E u r o p e a n B18,DR3 haplotypes (not found in Finland) from B8,DR3 haplotypes [ 30 ]. Consequently, since certain H L A haplotypes occur m o r e frequently among Type 1 diabetic patients than among control subjects, alleles detected with the more specific molecular methods provide relevant new markers for the disease only when analysed with adequate haplotype data. Whatever is detected by the molecular methods is still a part of the entire H L A haplotype due to linkage disequilibrium. Since the H L A haplotypes may cover about 3000 kb of D N A [ 7 ] they are excellent tools for studying genetic predisposition to Type 1 diabetes. W h e n H L A haplotypes are used as markers for disease susceptibility it is possible to detect the effect of as yet unidentified genes of the H L A system as well as the combined effect of alleles from different loci. O u r present data indicates that in Cw7,B8,DR3 and Cw3,Bw62,DR4 positive haplotypes the genes modifying the risk may be linked to the H L A - A locus since the absolute risk associated with these haplotypes varied when different A locus antigens were present. However, since the complement loci (located between class I and class II loci) were not tested variation in class I I I genes in these haplotypes is also possible, although it is not probable because of linkage disequilibrium between alleles at class I, class II and class I I I loci. T h o m s e n et al. [ 31 ] have proposed that in D R 4 positive haplotypes the complement C4 locus has an effect, independent of DQw8, on susceptibility to Type i diabetes. Sheehy et al. [ 32 ] have found that D Q w 8 positive haplotypes carrying D R 4 subtypes Dw4 or D w l 0 conferred the highest risk for Type i diabetes, which may suggest that susceptibility requires specific products of both D R and D Q loci. In both above-mentioned studies the H L A - A , C and B locus antigens were not determined, which causes a dilemma: it is impossible to conclude whether the class III or the D R polymorphisms were relevant themselves, or whether they only served as additional markers for high-risk haplotypes as interpreted by Sheehy et al. [ 32 ]. This illustrates the difficulties in pinpointing the susceptibility loci in the H L A region, and stress the importance of studying whole haplotypes from A to D Q locus for m o r e definite conclusions. M a n y studies indicate that the H L A - D Q polymorphism confers the major H L A linked susceptibility to Type 1 diabetes [ 2-5, 30 ]. Our findings indicate that the most c o m m o n DR3 and D R 4 positive haplotypes found in Type 1 diabetic patients (typed as D Q w 2 and DQw8, respectively) do not show variation at the D Q locus. O n the other hand, these haplotypes varied in the absolute risk they conferred, which suggests that other genes in these h a p l o t y p e s m i g h t also p l a y an i m p o r t a n t p a r t in m o d i f y i n g g e n e t i c s u s c e p t i b i l i t y to T y p e i d i a b e t e s . T h u s , t h e D Q m o l e c u l e s w h i l s t i m p o r t a n t s e e m t o a c c o u n t f o r o n l y p a r t o f t h e H L A l i n k e d p r e d i s p o s i t i o n to t h e d i s e a s e . M o r e k n o w l e d g e a b o u t t h e i m m u n o b i o l o g y o f H L A is n e e d e d to u n d e r s t a n d t h e c o n n e c t i o n b e t w e e n s t r u c t u r a l p o l y m o r p h i s m s a n d i m m u n o l o g i c a l f u n c t i o n s in t h e d i s e a s e p r o c e s s . 1. Svejgaard AP , Platz P , Ryder LP ( 1980 ) Insulin-dependent diabetes mellitus . In: Terasaki PC (ed) Histocompatibility 1980 . University of California Press, Los Angeles, pp 638 - 656 2. Owerbach D , Lernmark A , Platz P e t al. ( 1983 ) HLA-D region beta-chain D N A endonuclease fragments differ between HLADR identical healthy individuals and insulin-dependent diabetic individuals . Nature 303 : 815 - 817 3. Cohen-Haguenauer O , Robbins E , Massart C et al. ( 1985 ) A systematic study of H L A class II-beta DNA restriction fragments in insulin-dependent diabetes mellitus . Proc Natl Acad Sci USA 82 : 3335 - 3339 4. Todd JA , Bell JI , McDevitt HO ( 1987 ) HLA-DQB gene contributes to susceptibility and resistance to insulin-dependent diabetes mellitus . Nature 329 : 599 % 04 5. Morel PA , Dorman JS , Todd JA , McDevitt HO , Trucco M ( 1988 ) Aspartic acid at position 57 of the H L A - D Q beta -chain protects against type I diabetes: A family study . Proc Natl Acad Sci USA 85 : 8111 - 8115 6. Yamagata K , Nakajima H , Hanafusa T et al. ( 1989 ) Aspartic acid at position 57 of DQB chain does not protect against Type i (insulin-dependent) diabetes mellitus in Japanese subjects . Diabetologia 32 : 762 - 764 7. Ziegler A , Bloemer K , Pohla H , Weiss E , Schneider R Ragoussis J ( 1989 ) Towards a physical map of the H L A complex . In: Dupont B (ed) Immunobiology o f H L A Vol II . Springer, New York, pp 75 - 78 8. Auffray C , Strominger JL ( 1986 ) Molecular genetics of the human major histocompatibility complex . Adv Hum Genet 15 : 197 - 247 9. Sargent CA , Dunham I , Campbell RD ( 1989 ) Identification of multiple HTF-island associated genes in the human major histocompatibilitycomplex class III region . EMBO J 8 : 2305 - 2312 10. Hanson IM , Poustka A , Trowsdale J ( 1991 ) New genes in the class II region of the human major histocompatibility complex . Genomics 10 : 417 - 424 11. Tuomilehto-WolfE , Tuomilehto J , Cepaitis Z , Lounamaa R and the DIME study group ( 1989 ) New susceptibility haplotype for Type i diabetes . Lancet II : 299 - 302 12. Tuomilehto-WolfE , Tuomilehto J , Lappivaara J , Cepaitis A and the DIME study group ( 1990 ) Estimation of haplotype specific absolute risk in Type 1 diabetes in Finland . Diabetologia 33 : A44 (Abstract) 13. Diabetes Epidemiology Research International Group ( 1988 ) Geographic patterns of childhood insulin-dependent diabetes mellitus . Diabetes 37 : 113 - 119 14. Maniatis T , Fritsch EF , Sambrook J (eds) ( 1982 ) Molecular cloning . Cold Spring Harbor, New York 15. Auffray C , Lillie JW , Korman AJ et al. ( 1987 ) Structure and expression of HLA-DQc~ and -DXc~ genes: interallelic alternate splicing of the HLA-DQc~ gene and functional splicing of the HLA-DXc~ gene using a retroviral vector . Immunogenetics 26 : 63 - 73 16. Turco E , Care A , Compagnone-Post , Robinson C , Cascino I , Trucco M ( 1987 ) Allelicforms of the alpha- and beta-chain genes encoding DQwl-positive heterodimers . Immunogenetics 26 : 282 - 290 17. Saiki RK , Gelfand DH , Stoffel S e t al. ( 1988 ) Primer-directed enzymatic amplification of DNA with a thermostable DNA polymerase . Science 239 : 487 - 491 18. Gu XF , Elion J , Ouagued M , Clauser E , Assan R , Krishnamoorthy R ( 1988 ) A simple strategy to amplify specifically the HLA-DQ~ gene region with genomic DNA as template . Febs Lett 236 : 23 - 26 19. Sanger F , Nicklen S , Coulson A R (1977) D N A sequencing with chain-terminating inhibitors . Proc Natl Acad Sci USA 74 : 5463 - 5467 20. Bodmer JG , Marsh SGE , Albert ED et al. ( 1991 ) Nomenclature for factors of H L A system , 1990 . Tissue Antigens 37 : 97 - 104 21. Saiki R , Walsh PS , Levenson CH , Erlich H A ( 1989 ) Genetic analysis of amplified D N A with immobilized sequence-specific oligonucleotide probes . Proc Natl Acad Sci USA 86 : 6230 % 234 22. Hylding-Nielsen JJ , Boehm BO , Wood KJ , Bushell AR , Singal DR Svejgaard A ( 1989 ) RFLP standardization report for DQ beta/BamHI . In: Dupont B (ed) Immunobiology of H L A Vol I. Springer , New York, pp 634 - 636 23. Keller E , Andreas-Zietz A , Albert ED et al. ( 1989 ) RFLP standardization report for DQ beta/PstI . In: Dupont B (ed) Immunobiology o f H L A Vol I. Springer , New York, pp 613 - 614 24. Marcadet A , LeGall I , Cohen D et al. ( 1989 ) RFLP standardization report for DR beta/TaqI . In: Dupont B (ed) Immunobiology of H L A Vol I. Springer , New York, pp 58 % 589 25. Bontrop RE , Carpenter CB , Walford R , Sekiguchi S , Bignon JD , Cohen D ( 1989 ) RFLP standardization report for DQ alpha/TaqI . In: Dupont B (ed) Immunology of H L A Vol I. Springer , New York, pp 815 - 816 26. Hyldig-NielsenJJ , Trucco M , Coppin H et al. ( 1989 ) RFLP standardization report for DQ alpha/PstI . In: Dupont B (ed) Immunobiology o f H L A Vol I. Springer , New York, pp 809 - 810 27. Nevanlinna H ( 1972 ) The Finnish population structure. A genetic and genealogical study . Hereditas 71 : 195 - 236 28. Arnheim N , Strange C , Erlich H ( 1985 ) Use of pooled samples to detect linkage disequilibrium of polymorphic restriction fragments in human disease: Studies of the H L A class II loci . Proc Natl Acad Sci USA 82 : 6970 % 975 29. Serjeantson SW , Kohonen-Corish MRJ , Dunckley H , Reid MA (1986 ) H L A class II RFLPs are haplotype-specific . Cold Spring Harbor Symposia on Quantitative Biology , Vol LI: 83 - 89 30. Martell M , Marcadet A , Moine A et al. ( 1990 ) Heterogeneity of H L A genetic factors in IDDM susceptibility . Immunogenetics 31 : 233 - 240 31. Thomsen M , Molvig J , Zerbib A e t al. ( 1988 ) The susceptibilityto insulin-dependent diabetes mellitus is associated with C4 allotypes independently of the association with HLA-DQ alleles in HLA-DR3,4 heterozygotes . Immunogenetics 28 : 320 - 327 32. Sheehy MJ , Scharf SJ , Rowe JR et al. ( 1989 ) A diabetes-susceptible H L A haplotype is best defined by a combination of HLADR and -DQ alleles . J Clin Invest 83 : 830 . 835

This is a preview of a remote PDF: https://link.springer.com/content/pdf/10.1007%2FBF00400926.pdf

P. J. Tienari, E. Tuomilehto-Wolf, J. Tuomilehto, L. Peltonen, H. K. Åkerblom, A. Fagerlund, M. Flittner, B. Gustafsson, A. Hakulinen, L. Herva, P. Hiltunen, T. Huhtamäki, N. P. Huttunen, T. Huupponen, M. Hyttinen, Ch. Häggqvist, T. Joki, R. Jokisalo, S. Kallio, E. A. Kaprio, U. Kaski, M. Knip, M. L. Käär, L. Laine, J. Lappalainen, J. Mäenpää, A. L. Mäkelä, K. Niemi, A. Niiranen, P. Ojajärvi, T. Otonkoski, K. Pihlajamäki, S. Pöntynen, J. Sankala, J. Schumacher, M. Sillanpää, C. H. Stråhlmann, M. R. Ståhlberg, T. Uotila, P. Varimo, M. Väre. HLA haplotypes in Type 1 (insulin-dependent) diabetes mellitus: molecular analysis of the HLA-DQ locus, Diabetologia, 1992, 254-260, DOI: 10.1007/BF00400926