This Article Abstract Full Text (PDF) Submit a related Letter to the Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Donner, H. Articles by Badenhoop, K. Search for Related Content PubMed PubMed Citation Articles by Donner, H. Articles by Badenhoop, K.

The Presence or Absence of a Retroviral Long Terminal Repeat Influences the Genetic Risk for Type 1 Diabetes Conferred by Human Leukocyte Antigen DQ Haplotypes

Center of Internal Medicine, Medical Department I, Division of Endocrinology, University Hospital (H.D., T.S., J.B., J.H. K.H.U., K.B.), D-60590 Frankfurt am Main; and the Paul Ehrlich Institut (R.R.T., R.K.), D-63225 Langen, Germany; and the Diabetes Research Center, Vrije Universiteit Brussels (B.V.d.A., I.W.), B-1090 Brussels, Belgium

Address all correspondence and requests for reprints to: Dr. Klaus Badenhoop, Center of Internal Medicine, Medical Department I, Division of Endocrinology, University Hospital, Theodor Stern Kai 7, D-60590 Frankfurt am Main, Germany. E-mail: badenhoop{at}em.uni-frankfurt.de

	Abstract

Top Abstract Introduction Experimental Subjects Materials and Methods Results Discussion References

 
Major genetic susceptibility to type 1 diabetes mellitus maps to the human leukocyte antigen (HLA) region on chromosome 6p. During evolution, endogenous retroviral long terminal repeats (LTR) have been integrated at several sites within this region. We analyzed the presence of a solitary HERV-K LTR in the HLA DQ region (DQ-LTR3) and its linkage to DRB1, DQA1, and DQB1 haplotypes derived from 246 German and Belgian families with a patient suffering from type 1 diabetes mellitus. Segregation analysis of 984 HLA DQA1/B1 haplotypes showed that DQ-LTR3 is linked to distinct DQA1 and DQB1 haplotypes but is absent in others. The presence of DQ-LTR3 on HLA DQB1*0302 haplotypes was preferentially transmitted to patients from heterozygous parents (82%; P < 10^-6), in contrast to only 2 of 7 DQB1*0302 haplotypes without DQ-LTR3. Also, the extended HLA DRB1*0401, DQB1*0302 DQ-LTR3-positive haplotypes were preferentially transmitted (84%; P < 10^-6) compared with 1 of 6 DR-DQ matched DQ-LTR3 negative haplotypes. DQ-LTR3 is missing on most DQB1*0201 haplotypes, and those LTR3 negative haplotypes were also preferentially transmitted to patients (80%; P < 10^-6), whereas DQB1*0201 DQ-LTR3-positive haplotypes were less often transmitted to patients (36%). Other DQA1/B1 haplotypes did not differ for DQ-LTR3 between transmitted and nontransmitted haplotypes. Thus, the presence of DQLTR3 on HLA DQB1*0302 and its absence on DQB1*0201 haplotypes are independent genetic risk markers for type 1 diabetes.

	Introduction

Top Abstract Introduction Experimental Subjects Materials and Methods Results Discussion References

 
TYPE 1 diabetes mellitus is a ß-cell selective autoimmune disease with a strong genetic background that is conferred by at least 14 gene loci (reviewed in Refs. 1, 2). The largest part of heritable susceptibility is marked by variation in the human leukocyte antigen (HLA) region on chromosome 6p, designated IDDM1 (3, 4), and is mainly due to HLA DRB1, DQA1, and DQB1 alleles (reviewed in Ref. 5). Also, an increased prevalence of HLA DRB1*0401 and DQB1*0302 in combination with the HLA class I B*39 allele has been reported (6). However, the HLA association with type 1 diabetes depends on the genetic background, i.e. the distribution of alleles in the population (7). Several alternative explanations of this genetic phenomenon have therefore been sought. Linkage of an unidentified susceptibility locus within the HLA region cannot be ruled out, e.g. in unmapped areas of the class II region, the class III region (8), or outside (9). Recently, two separate parts of the HLA complex were reported to contribute to susceptibility to or protection from IDDM (10).

The human endogenous retrovirus K (HERV-K) family is present in 30–50 full-length copies (11), whereas about 10,000 solitary long terminal repeats (LTRs) exist per haploid human genome (12). The evolutionary development of the MHC was accompanied by insertions of endogenous retro-viral DNA elements such as LTRs at several sites (9), which contribute to the heterogeneity of MHC haplotypes. Two LTRs with more than 90% sequence homology to the HERV-K LTR had earlier been described in the vicinity of HLA DQ (13). We recently demonstrated that the presence of one of these LTRs (DQ-LTR3) was significantly associated with type 1 diabetes (14) as well as with rheumatoid arthritis (15).

The aims of this study were to characterize the linkage of DQ-LTR3 to HLA DR-DQ haplotypes and to investigate whether DQ-LTR3 is an additional and independent genetic marker in type 1 diabetes.

	Experimental Subjects

Top Abstract Introduction Experimental Subjects Materials and Methods Results Discussion References

 
Altogether 246 families comprising 816 members (492 parents, 246 affected children, and 78 nonaffected siblings) with a type 1 diabetic proband (n = 246) were recruited in the out-patient endocrine and diabetes clinics of the Departments of Internal Medicine and Pediatrics, University Hospital Frankfurt am Main (130 families), or from the Belgian Diabetes Registry (116 families) (16). The diagnosis of type 1 diabetes was based on WHO or National Diabetes Data group criteria (17). The study was approved by the local ethical committees.

	Materials and Methods

Top Abstract Introduction Experimental Subjects Materials and Methods Results Discussion References

 
Linkage analysis of DQ-LTR3 and HLA class II haplotypes was achieved by designing a PCR assay that clearly distinguished between homo- and heterozygotes for DQ-LTR3. Segregation analysis in families allowed us to define which HLA DRB1*04, DQA1, DQB1, and DQ-LTR3 variants were inherited in combination. By defining transmitted haplotypes (inherited by patients) and those not transmitted (not inherited), we studied the presence of DQ-LTR3 on disease-associated chromosomes and their linkage with HLA DR and DQ. This analysis should enable us to distinguish whether DQ-LTR3 is a secondary association marker or provides additional information regarding susceptibility to type 1 diabetes.

HLA DQ, DR, and DQ-LTR3 genotyping

Genomic DNA was isolated from whole ethylenediamine tetraacetate blood samples and subjected to PCR analyses that always included a negative control standard with all reagents but genomic DNA. HLA DQA1 and DQB1 alleles were defined by sequence-specific primer or allele-specific oligonucleotide analysis based on the recent WHO HLA nomenclature (18, 19). Additionally, HLA DRB1*04 subtypes were studied in all DQA1*03- and DQB1*03-positive individuals (20). In this way we could distinguish 11 HLA DRB1*04 alleles.

The presence or absence of DQ-LTR3 was defined by a nested PCR. For DQ-LTR3, external primers were used (5'-AATGCTGATTAGAAGTAGCTCTG-3' and 5'-ACAAGGACATCTCCTGATCAG-3') to generate a 1285-bp fragment (4925–6210, clone HSE1448, GenBank Z80898) in the presence of the DQ-LTR3 or a 310-bp fragment in the absence. All amplifications were performed on a Multicycler PTC 200 (Biozym, Hess Oldendorf, Germany). All external fragments were amplified using the Expand High Fidelity PCR system (Boehringer Mannheim, Mannheim, Germany) in the following reaction: 250 ng genomic DNA, 8 mmol/L deoxy-NTPs, 20 pmol of each primer, 1.5 mmol/L MgCl₂, and 1.5 U of polymerase mix. For DQ-LTR3, 30 cycles (94 C for 1 min, 61 C for 1 min, and 72 C for 75 s) after 4-min initial denaturation and with a final extension for 4 min were applied. A 1008-bp internal fragment of DQ-LTR3 was amplified only in the presence of DQ-LTR3 with the following primers: 5'-GGTGGAGCAACAGCCCACCCGGG AAGT-3' and 5'-CCCCTTGTGACTTCTGTGGGGAAAAGC-3' (5056–6084, clone 1448, GenBank Z80898). The PCR reactions were carried out as for the external fragments but using Taq polymerase (Promega Corp., Madison, WI), 10 mmol/L Tris-HCl, 50 mmol/L KCl, and 1.5 mmol/L MgCl₂ in a final volume of 25 µL under the following conditions: initial denaturation (94 C) for 4 min and 30 cycles (94 C for 1 min, 62 C for 1 min, and 72 C for 1 min) and a final extension for 4 min.

To confirm our results, DQ-LTR3-positive and -negative PCR fragments were purified on agarose gels and cloned into the TA cloning vector pGem-T easy (Promega Corp.). Inserts devoid of LTR sequences were fully sequenced, and inserts containing LTR elements were partially sequenced using a cycle-sequencing protocol employing SP6 and T7 primers with Thermo Sequenase and Taq DyeDeoxy terminators on a 373A DNA Sequencing System according to the instructions of the manufacturer (Amersham, Aylesbury, UK). Primers were commercially purchased from Eurogentec (Seraing, Belgium) and Roth (Karlsruhe, Germany).

Segregation analysis

Genotypes of patients, their parents, and siblings were compared to define haplotypes. No recombination was observed, and all families displayed segregation patterns of Mendelian inheritance. Alleles that were found in patients were classified as transmitted, whereas those parental alleles not observed in patients were classified as not transmitted. Studying 246 nuclear families thus resulted in the analysis of 984 haplotypes. Segregation of DQ-LTR3 in families allowed us to assign their linkage to HLA DQA1 and DQB1 alleles on haplotypes.

Statistics

To analyze our data we performed different types of statistics. First, the combined transmissions of selected haplotypes, either DQ-LTR3 positive or negative, were analyzed by a transmission distortion test. The observed transmissions were compared to the random expected transmissions of 50% by the ² test (21) (Table 1). Only transmissions from heterozygous parents were counted, and statistical significance was defined as P < 0.05 (1 df) with Yates’ correction and Fisher’s exact test where appropriate. Additionally, we compared the distribution of DQ-LTR3 on transmitted with that on not transmitted DQ haplotypes by ² test (see Table 2). Third, the positive and negative predictive values (PPV and NPV) as well as the haplotype relative risk (HRR) were calculated (see Table 3). These latter comparisons included all transmissions, also from homozygous parents (22). The formats for the calculations of PPV and NPV values are shown in Table 3. In the analysis of DQ-LTR3 presence/absence, we tested the a priori hypothesis that the frequency of DQ-LTR3-positive haplotypes differed between selected haplotypes (e.g. HLA DQA1*03 and DQB1*0302) known to confer strong susceptibility to type 1 diabetes. Thus, P values generated from this comparison (DQ-LTR3 ± on these haplotypes) were not corrected.

View this table: [in this window] [in a new window]   Table 2. Format of the 2 test statistic to compare the different distributions of DQ-LTR3 on selected HLA DQ and DR haplotypes (data were deduced from Table 1)

	Results

Top Abstract Introduction Experimental Subjects Materials and Methods Results Discussion References

Table 1 displays the combined transmission of DQ-LTR3 with selected DQA1, DQB1, and DRB1*04 alleles from heterozygous parents in 492 transmitted and 492 nontransmitted haplotypes derived from 246 German and Belgian families with an type 1 diabetic patient as offspring. As expected, 78% of all DQB1*0302-positive parental haplotypes (P < 10^-6) were transmitted to patients. Subdividing DQB1*0302 haplotypes into DQ-LTR3 positive and negative, this transmission rate increased to 82% for all DQ-LTR3-positive haplotypes (P < 10^-6), whereas only 29% of all DQB1*0302 DQ-LTR3-negative haplotypes were transmitted. The transmission of HLA DQA1*0301- and DQB1*0301-positive haplotypes was reduced independently of that of DQ-LTR3 (40% for DQLTR3-positive and 35% for DQ-LTR3-negative haplotypes). We observed the highest transmission rate for the extended DRB1*0401- and DQB1*0302 DQ-LTR3-positive haplotype (84%; P < 10^-6), in contrast to only one of six DRB1*0401-, DQB1*0302-, and DQ-LTR3-negative haplotypes (17%).

Also, the DQA1*0501 DQB1*0201 haplotype was preferentially transmitted to patients (78%; P < 10^-6). Comparing these DQB1*0201 haplotypes, the transmission rate was 80% (P < 10^-6) for all DQLTR3-negative haplotypes, whereas it was less frequent (36%) for all DQB1*0201 DQ-LTR3-positive haplotypes. HLA DQA1*0501 DQB1*0301 was less often transmitted to patients (16%; P < 2 x 10^-6) for both DQ-LTR3-positive (18%) and DQ-LTR3-negative haplotypes (16%). There were no significant differences for DQ-LTR3 frequencies among haplotypes with other HLA DQA1/B1 alleles.

To address the question of whether the frequencies of DQ-LTR3 differed significantly on DQB1*0302- or DQB1*0201-transmitted compared to nontransmitted haplotypes, we performed a ² test (see Table 2). DQ-LTR3 was significantly more frequent on DQA1*0301-transmitted haplotypes but was significantly less so on DQA1*0501-transmitted haplotypes (P < 2 x 10^-6 and P < 10^-6, respectively). Also, the extended haplotypes DQA1*0301, DQB1*0302 (P < 0.004), DRB1*0401, DQB1*0302 (P < 9 x 10^-4), as well as DQA1*0501, DQB1*0201 (P < 0.003) differed significantly when transmitted haplotypes were compared to nontransmitted haplotypes for DQLTR3. The DQ-LTR3 distribution on the protective DQA1*0301 DQB1*0301 and DQA1*0501 DQB1*0301 haplotypes was not different.

Calculation of the PPV and NPV as well as the HRR of selected haplotypes

We also tested whether the retroviral element would qualify as an additional risk marker. Table 3 summarizes the PPV and NPV as well as the HRR. When this analysis was performed, we compared haplotypes positive for a marker combination against the remaining haplotypes (either transmitted or nontransmitted). Therefore, the numbers of haplotypes differ in Table 3 from those in Table 1, where only transmissions from heterozygous parents were counted. High risk DQB1*0302 or DQB1*0201 haplotypes subdivided into DQ-LTR3-positive or -negative haplotypes differed in the PPV. The PPV of DQA1*03 DQ-LTR3-positive haplotypes was 0.74, whereas it was 0.39 on DQ-LTR3-negative haplotypes. Also, DQB1*0302 as well as the extended DRB1*0401-, DQB1*0302-, and DQ-LTR3-positive haplotypes showed a higher PPV (0.78 and 0.82) compared to DQ-LTR3-negative haplotypes (0.29 and 0.17). The absence of DQ-LTR3 on DQA1*0501 haplotypes was associated with a higher PPV (0.62), whereas it was lower in its presence (0.23). Similarly, the PPV for DQB1*0201 haplotypes without DQ-LTR3 was 0.75, and that for DQB1*0201 DQ-LTR3-positive haplotypes was 0.36. The PPV do not substantially vary when protective DQA1*0301, DQB1*0301 or DQA1*0501, DQB1*0301 haplotypes were analyzed for DQ-LTR3.

The HRR was 4.5 for DQB1*0302-positive haplotype, 4.9 for DQB1*0302 DQ-LTR3-positive haplotype, and 5.6 for the extended DRB1*0401, DQB1*0302 DQ-LTR3-positive haplotype. The HRR of DQB1*0201 DQ-LTR3-negative haplotype was 4.2. In contrast, DQB1*0302 DQ-LTR3-negative and DQB1*0201 DQ-LTR3-positive haplotypes conferred a HRR below 1.

Sequence analysis of DQ-LTR3 flanking sites

Furthermore, we sequenced the flanking chromosomal regions encompassing the LTR or, in their absence, their corresponding empty sites in selected DQ-LTR3^+/+, DQ-LTR3^+/-, and DQ-LTR3^-/- individuals and confirmed our results obtained by DQ-LTR3 PCR (data not shown). No variation within the DQ-LTR3 sequences was found.

	Discussion

Top Abstract Introduction Experimental Subjects Materials and Methods Results Discussion References

 
By evaluating the cosegregation of DQ-LTR3 on HLA DQ haplotypes in type 1 diabetic families, we have shown that its presence or absence improves risk assessment. DQ-LTR3 cosegregates with particular HLA DQA1, DQB1 haplotypes, and its linkage to HLA DQB1*0302 and DQB1*0201 differs significantly between transmitted and nontransmitted haplotypes. These differences are obvious in the German families studied, but the lack of significance in Belgian families is due to the fact that only two DQ-LTR3-negative DQB1*0302 and one DQ-LTR3-positive DQA1*0501, DQB1*0201 haplotypes were found among them. However, the combined dataset still shows that DQB1*0302 DQ-LTR3-positive haplotypes were preferentially transmitted to patients, whereas those without the DQ-LTR3 were preferentially not transmitted. Thus, the DQLTR3 marks a haplotype that carries more susceptibility than defined by the sole allele HLA DQB1*0302.

Conversely, the HLA DQA1*0501, DQB1*0201 haplotypes transmitted to patients were rarely DQ-LTR3 positive. Therefore, the absence of DQLTR3 on HLA DQA1*0501 DQB1*0201 haplotypes marks a more susceptible haplotype than that defined by its DQ alleles alone.

Why some HLA haplotypes appear to have integrated the DQ-LTR3 more frequently than others is unclear at present. This integration must have occurred late in the evolution of the MHC (23). Our observations are therefore consistent with recent findings that the HLA DQB1 locus reveals the highest level of sequence variation found in the human genome to date and that most of the major insertions/deletions are of retroviral origin (24). Thus, haplotype- specific variation, as seen in our family study, extends beyond the exons and transcriptional units. Also, HLA DRB1 exons are surrounded by similar polymorphic microsatellites even in distant populations (25). Thus, exon variation has evolved in parallel with intron polymorphism.

The genetic risk of DR3 and DR4 haplotypes was recently fine mapped with HLA-B alleles and microsatellite markers (6, 10). In our study the presence of DQ-LTR3 clearly distinguishes HLA DR4, DQB1*0302 with regard to IDDM susceptibility, in contrast to the absence of LTR on transmitted DQB1*0201 haplotypes. We also observe a strong difference in the PPV of high risk DQB1*0302 or DQB1*0201 haplotypes when subdivided into DQ-LTR3-positive and -negative groups. As the presence or absence of DQ-LTR3 only splits susceptible haplotypes, protection seems to be dominant, as seen in HLA class II exon variants (26). It is conceivable that the presence of retroelements might correlate with the high degree of diversity that distinguishes the MHC from other regions. Thus, these elements might have helped to shape the immunogenetic repertoire in accordance with the infectious environment during evolution. The documented retroelements within the MHC are not known to have interfered with gene function. Nevertheless, they may affect gene transcription by direct, indirect, or tissue-specific regulation of DQB1, as has been reported for other loci (24, 27, 28, 29).

The significantly different transmission of DQ-LTR3 on HLA DRB1*0401, DQB1*0302 suggests that this LTR represents an additional susceptibility marker. In contrast, the absence of DQ-LTR3 enhances the risk on the other predisposing HLA DQA1*0501, DQB1*0201 haplotype.

This is the first report of heterogeneity within the DRB1*0401 DQB1*0302 haplotype, suggesting that the contribution of allelic variation of exon 2 of HLA DRB1, DQA1, and DQB1 genes cannot completely explain the association of this haplotype with type 1 diabetes. Thus, the DQ-LTR3 marks a DR4 haplotype that presumably contains a genetic factor with a strong influence on susceptibility to type 1 diabetes.

	Acknowledgments

 
The technical assistance of G. Braun is gratefully acknowledged. The following members of the Belgian Diabetes Registry are gratefully acknowledged for contributing to the accumulation of patient samples: J. Beirinckx (Izegem), L. Claeys (Zoersel), M. Coeckelberghs (Antwerp), R. Craen (Gent), P. Decraene (Bonheiden), I. De Leeuw (Antwerp), J. De Schepper (Brussels), H. Dorchy (Brussels), M. Du Caju (Antwerp), L. Emsens (Knokke), A. Eykens (Herentals), K. Garmyn (Lier), J. Gérard (Liege), C. Herbaut (Mons), B. Keymeulen (Brussels), G. Krzentowski (Jumet), J. Monballyu (Ekeren), F. Nobels (Aalst), P. Taelman (Gent), J. Teuwen (Kapellen), J. Tits (Genk), P. Van Crombrugge (Aalst), E. Vandenbussche (Herentals), and D. Van Doorn (Duffel).

Received November 6, 1998.

Revised January 7, 1998.

Accepted January 20, 1998.

	References

Top Abstract Introduction Experimental Subjects Materials and Methods Results Discussion References

 

1. Vyse TJ, Todd JA. 1996 Genetic analysis of autoimmune disease. Cell. 85:311–318.[CrossRef][Medline]
2. Tisch R, McDevitt H. 1996 Insulin-dependent diabetes mellitus. Cell. 85:291–297.[CrossRef][Medline]
3. Mein CA, Esposito L, Dunn MG, et al. 1998 A search for type 1 diabetes susceptibility genes in families from the United Kingdom. Nat Genet. 19:297–300.[CrossRef][Medline]
4. Concannon P, Gogolin-Ewens KJ, Hinds DA, et al. 1998 A second-generation screen of the human genome for susceptibility to insulin-dependent diabetes mellitus. Nat Genet. 19:292–296.[CrossRef][Medline]
5. She J-X. 1996 Susceptibility to type 1 diabetes: HLA-DQ and DR revisited. Immunol Today. 17:323–329.[CrossRef][Medline]
6. Nejentsev S, Reijonen H, Adojaan B, et al. 1997 The effect of HLA-B allele on the IDDM risk defined by DRB1*04 subtypes and DQB1*0302. Diabetes. 46:1888–1892.[Abstract]
7. Dorman JS, LaPorte RE, Stone RA, Trucco M. 1990 Worldwide differences in the incidence of type I diabetes are associated with amino acid variations at position 57 of the HLA-DQ ß chain. Proc Natl Acad Sci USA. 87:7370–7374.[Abstract/Free Full Text]
8. Badenhoop K, Schwarz G, Trowsdale J, et al. 1989 TNF- gene polymorphisms in type 1 (insulin-dependent) diabetes mellitus. Diabetologia. 32:445–448.[CrossRef][Medline]
9. Beck S, Abdulla S, Alderton RP, et al. 1996 Evolutionary dynamics of non-coding sequences within the class II region of the human MHC. J Mol Biol. 255:1–13.[CrossRef][Medline]
10. Moghaddam PH, de Knijf P, Roep BO, et al. 1998 Two seperate regions in the major histocompatibility complex contributes to susceptibility or protection. Diabetes. 47:263–269.[Abstract]
11. Ono M. 1986 Molecular cloning and long terminal repeat sequences of human endogenous retrovirus genes related to type A and B retrovirus genes. J Virol. 58:937–944.[Abstract/Free Full Text]
12. Leib-Mösch C, Haltmeier M, Werner T, et al. 1993 Genomic distribution and transcription of solitary HERV-K LTRs. Genomics. 18:261–269.[CrossRef][Medline]
13. Kambhu S, Falldorf P, Lee JS. 1990 Endogenous retroviral long terminal repeats within the HLA-DQ locus. Proc Natl Acad Sci USA. 87:4927–4931.[Abstract/Free Full Text]
14. Badenhoop K, Tönjes RR, Rau H, et al. 1996 Endogenous retroviral long terminal repeats of the HLA-DQ region are associated with suceptibility to insulin-dependent diabetes mellitus. Hum Immunol. 50:103–110.[CrossRef][Medline]
15. Seidl C, Donner H, Petershofen E, et al. 1999 An endogenous retroviral long terminal repeat at the HLA-DQB1 gene locus confers susceptibility to rheumatoid arthritis. Hum Immunol60 :63–68.
16. Vandewalle CL, Coeckelgerghs MI, De Leeuw IH, et al. 1997 Epidemiology, clinical aspects, and biology of IDDM patients under age 40. Diabetes Care. 20:1556–1561.[Abstract]
17. National Diabetes Data Group. 1979 National Diabetes Data Group classification and diagnosis of diabetes mellitus and other categories of glucose intolerance. Diabetes. 28:1039–1057.[Medline]
18. Olerup O, Aldener A, Fogdell A. 1993 HLA-DQB1 and -DQA1 typing by PCR amplification with sequence-specific primers (PCR-SSP) in 2 hours. Tissue Antigens. 41:119–134.[Medline]
19. Marsh SGE. 1998 HLA class II region sequences, 1998. Tissue Antigens. 51:467–507.[Medline]
20. Zetterquist H, Olerup O. 1992 Idetification of the HLA-DRB1*04, -DRB1*07 and -DRB1*09 alleles by PCR amplification with sequence-specific primers (PCR-SSP) in 2 hours. Hum Immunol. 34:64–74.[Medline]
21. Spielman RS, Ewens WJ. 1996 The TDT and other family-based tests for linkage disequilibrium and association. Am J Hum Genet. 59:983–989.[Medline]
22. Falk CT, Rubinstein P. 1987 Haplotype relative risk: an easy reliable way to construct a proper control sample for risk calculations. Ann Hum Genet. 51:227–233.[Medline]
23. Mayer J, Meese E, Mueller-Lantzsch N. 1998 Human endogenous retrovirus K homologous sequences and their coding capacity in Old World primates. J Virol. 72:1870–1875.[Abstract/Free Full Text]
24. Horton R, Niblett D, Milne S, et al. 1998 Large-scale sequence comparisions reveal unusually high levels of variation in the HLA-DQB1 locus in the class II region of the human MHC. J Mol Biol. 282:71–97.[CrossRef][Medline]
25. Epplen C, Santos EJM, Guerreiro JF, van Helden P, Epplen JT. 1997 Coding vs. intron variability: extremely polymorphic HLA-DRB1 exons are flanked by specific composite microsatellites, even in distant populations. Hum Genet. 99:399–406.[CrossRef][Medline]
26. Cucca F, Todd JA. 1996 HLA susceptibility to type 1 diabetes: methods and mechanisms. In: Browning M, McMichael A, eds. HLA and MHC genes, molecules and function, 1st ed. Oxford: BIOS Scientific Publishers; 383–406.
27. Kato N, Pfeifer-Ohlsson S, Larsson E, Rydnert J, Ohlsson R, Cohen M. 1987 Tissue-specific expression of human provirus ERV 3 mRNA in human placenta: two of the three ERV 3 mRNAs contain human cellular sequences. J Virol. 61:2182–2191.[Abstract/Free Full Text]
28. Feuchter-Murthy AE, Freeman JD, Mager DL. 1993 Splicing of a human endogenous retrovirus to a novel phospholipase A2 related gene. Nucleic Acids Res. 21:135–143.[Abstract/Free Full Text]
29. Dangel AW, Mendoza AR, Baker BJ, et al. 1994 The dichotomous size variation of human complement C4 genes is mediated by a novel family of endogenous retroviruses, which also established species-specific genomic patterns among Old World primates. Immunogenetics. 40:425–436.[Medline]
30. Sackett DL. 1996 Evaluation of clinical method. In: Weatherall DJ, Ledingham JGG, Warrell DA, eds. Oxford textbook of medicine, 1st ed. Oxford: Oxford University Press; 15–21.

This article has been cited by other articles:

				G. Gambelunghe, I. Kockum, V. Bini, G. D. Giorgi, F. Celi, C. Betterle, R. Giordano, R. Libe, A. Falorni, and on behalf of the Umbria Type 1 Diabetes Registry a Retrovirus-Like Long-Terminal Repeat DQ-LTR13 and Genetic Susceptibility to Type 1 Diabetes and Autoimmune Addison's Disease Diabetes, March 1, 2005; 54(3): 900 - 905. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Submit a related Letter to the Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Donner, H. Articles by Badenhoop, K. Search for Related Content PubMed PubMed Citation Articles by Donner, H. Articles by Badenhoop, K.