Polymorphisms of CXCR3-binding chemokines in type 1 diabetes

Type 1 diabetes (T1D) is a T-cell-mediated autoimmune disease. Although the precise mechanisms leading to the destruction of islet beta cells are unknown, diverse studies support a role of the CXCR3-binding chemokines. A combination of a case (n = 447)-control (n = 300) and family (n = 221) analysis was performed to investigate the role of the CXCL9 (rs10336, rs3733236) and CXCL10 (rs3921, rs35795399 and rs8878) polymorphisms and their interaction with HLA high-risk haplotypes DQ2(DQA*0501-DQB*0201)-DQ8(DQA*0301-DQB*0302) in T1D. In addition, the mRNA expression of these genes and of the CXCR3 in peripheral blood mononuclear cells (PBMCs) of T1D patients was studied. In the family analysis, an overtransmission of the allele T and G of the polymorphisms rs35795399 and rs8878 in the whole group (p = 0.0520 and p = 0.0290, respectively) as well as in combination with the HLA-high risk haplotypes (p = 0.0209 and 0.0340, respectively) were observed. In addition, the haplotype rs8878G-rs35795399T was more often transmitted from parents to affected offspring, whereas the haplotype rs8878A-rs35795399C was less often transmitted (p = 0.0130 and p = 0.0201, respectively). Nevertheless these associations did not remain significant after correction for multiple testing, and they could not be corroborated in the case-control analysis. Although we did not find an association of the CXCL9 and CXCL10 polymorphisms with type 1 diabetes in the German population, we cannot discard their role in other populations or other autoimmune diseases.

Mating in parents of type 1 diabetes families as a function of the HLA DR-DQ haplotype

AIM

The Major Histocompatibility Complex (MHC) region on chromosome 6p21 (IDDM1) contributes about half of the familial clustering of type 1 diabetes (T1D). Several studies have revealed that highly polymorphic genes within the MHC may associate with the mating choice. Our study should determine whether a specific mating effect is detectable in T1D families as a function of human leucocyte antigen (HLA) DR-DQ, which could contribute to disease susceptibility.

METHODS

We analysed the parental HLA-DR genotypes in 829 diabetic families. The families derive from the Type 1 Diabetes Genetics Consortium (T1DGC) in addition to those of our own centre and the original UK, US and SCAND diabetic families.

RESULTS

A total of 307 of 829 parental couples (37.0%) were matched for at least one known T1D risk haplotype (DR3 or DR4), which is significantly less than the expected 374.9 (45.2%), derived from population genotype frequencies (p < 0.0009). Parents share less susceptibility haplotypes and rather complement each other as both carry one different risk haplotype (DR3 or DR4). The number of such parental couples was significantly higher than expected (293 vs. 223.4; p < 0.0003). All non-transmitted DR haplotype pairs were also analysed. More often than expected, both parents did not transmit DR1 (94 vs. 59.1; p < 0.003) and DRy (y: not DR1, not DR3, not DR4; 63 vs. 30.3; p < 0.0005). In contrast, the parental non-transmitted pair of haplotypes DR1-DRy was observed to a far lesser extent than expected (26 vs. 84.7; p < 10(-8)). These observations were only made in multiplex families, whereas in simplex families, no deviation from the expected frequencies was observed.

CONCLUSIONS

Our data are consistent with the conclusion that genes in the HLA region may influence the mating choice in parents of T1D patients, thus contributing to familial clustering of T1D in multiplex families. This may indicate a different parental background of multiplex compared with simplex T1D families.

MHC-environment interactions leading to type 1 diabetes: feasibility of an analysis of HLA DR-DQ alleles in relation to manifestation periods and dates of birth

AIM

The region on chromosome 6p21 (IDDM1) confers the largest part of genetic susceptibility to type 1 diabetes (T1D) with particular human leucocyte antigen (HLA) alleles predisposing and others protecting from it. As T1D is primarily a "sporadic" disease, the pathophysiology must involve gene-environment interactions. We searched for indirect evidence for such major histocompatibility complex (MHC)-environment interactions by asking two questions: (i) can the degree of an HLA association vary over time periods? and (ii) if a prenatal event like an intrauterine infection - that might cluster in seasons - leads to differences of HLA associations in patients with particular birth months?

METHODS

We screened the Type 1 Diabetes Genetics Consortium (T1DGC) database (in addition our own database and the original UK, US and SCAND databases) for MHC DR-DQ and CTLA4 associations. First, we separated the groups of patients with onset of disease before 1980 in comparison with onset after 1980. Second, we analysed the data according to dates of birth (grouped in months). Not all patients' dates of birth or manifestation periods were available, leading to different group sizes. There were 282 patients analysed for manifestation periods and 329 for birth month.

RESULTS

The cohorts of manifestation before 1980 demonstrated a significantly lower frequency of DQ2/X (2 vs. 14.2%; p = 0.03). There was a trend for DQ8/x to be more frequent for manifestations before 1980 (34 vs. 21.6%; p < 0.10). Other alleles did not differ significantly. The months of birth were not evenly distributed. Significant deviations from the whole group were seen in August (DQ2/8 trough and DQx/x high), whereas birth in September was more frequent in DQ8/x or DQ8/8 carriers. This pattern was significantly different from the expected distribution of months at birth (13.9 vs. 7.6%; p < 0.04).

CONCLUSIONS

We demonstrate the feasibility of an analysis that searches for indirect evidence of gene-environment interactions. These preliminary data need to be confirmed in larger data sets.

A new HLA-B*08 allele, HLA-B*0828, found in two voluntary stem cell donors

The novel allele HLA-B*0828 differs from HLA-B*080101 by three nucleotide exchanges at codon 113, 114, and 116 in exon 3.

HLA-DQ haplotypes in Spanish and German families with Graves' disease: contribution to DQA1*0501-DQB1*0301 mediated genetic susceptibility from fathers

Genetic and environmental factors are involved in the pathogenesis of Graves' disease. The human leukocyte antigen (HLA) locus is considered to be one risk factor for Graves' disease but parent of origin effects have not been studied. Therefore, we investigated the transmission of HLA risk haplotypes DQA1*0501, DQA1*0501-DQB1*0201 (DQ2), and DQA1*0501-DQB1*0301 (DQ7) in two Graves' disease family-cohorts from Spain and Germany. Altogether 208 trio-families (109 from Spain and 99 from Germany; n = 624 individuals) with Graves' disease were genotyped for HLA-DQ alleles DQA1*0501 and the haplotypes DQA1*0501-DQB1*0201 (DQ2) and DQA1*0501-DQB1*0301 (DQ7). Since both family groups-German and Spanish-showed the same pattern of HLA transmission and nontransmission, they were analyzed together. HLA DQA1*0501 and DQA1*0501-DQB1*0201 (DQ2) were significantly overtransmitted from the parents to the affected offspring (204 vs. 131, p = 0.0057, pc = 0.0228 and 109 vs. 55, p = 0.0036, pc = 0.0144, respectively). These haplotypes were preferentially transmitted from fathers and DQA1*0501-DQB1*0301 (DQ7) was also more prevalent in fathers (24.0% vs. 17.1%, p = 0.0162, pc = 0.0648). We conclude, that HLA DQA1*0501 and DQA1*0501-DQB1*0201 (DQ2) are strongly associated with Graves' disease in both populations. A parent of origin effect of risk haplotypes can not be excluded at present, warranting further family studies.

Genetic and biological characterization of recombinant HIV type 1 with Env derived from long-term nonprogressor (LTNP) viruses

Multiple factors are known to contribute to nonprogressive disease in long-term nonprogressors (LTNP). We previously selected LTNPs, in which broadly neutralizing antibodies against HIV-1 very likely contribute to disease prevention. Here, we characterize those LTNPs further. We analyzed sequences of the viral genes env, nef, vpr, tat, and rev as well as the cellular ccr5, HLA-B*5701, and HLA-B*27 genes derived from eight LTNPs, as mutations in these genes have been associated with the LTNP status in some studies. Furthermore, we compared the replication rates of recombinant reporter viruses carrying envelope proteins from LTNPs to control viruses from patients with similar CD4 count and viral load. Concerning the cellular factors, none of the eight LTNPs showed the 32-base pair deletion in the ccr5 gene, and HLA-B*5701 and HLA-B*27 alleles were detected in only one LTNP, respectively. The reading frames for the regulatory genes nef, vpr, tat, and rev were all open. Although Env sequences from LTNPs differed from those of control patients with respect to the length of variable domains and the number of N-glycosylation sites, these differences were not statistically significant and did not lead to differences in infectivity of recombinant reporter viruses.

Neither an intronic CA repeat within the CD48 gene nor the HERV-K18 polymorphisms are associated with type 1 diabetes

Type 1 diabetes is an autoimmune heterogeneous disease that is determined by environmental and genetic factors. A possible retroviral etiology has been inferred from the observation that human endogenous retrovirus (HERV)-K18 encoding a superantigen (SAg) has a polymorphism associated with this disease. Type 1 diabetes families from Germany and Belgium were genotyped for the novel HERV-8914 (303 families) and for the known HERV-8594 (284 families) polymorphisms within the SAg-coding region on the HERV-K18. Case-control analysis was performed for the HERV-8914 polymorphism (506 patients) and for the HERV-8594 polymorphism (370 patients) and compared with 350 German controls. Haplotypes were constructed. Additionally, a microsatellite within the CD48 gene was analyzed in German type 1 diabetes families (n=125) as well as in patients (n=375) and in healthy controls (n=350). No association was found for HERV-K18 polymorphisms or the CA repeat within the CD48 gene with type 1 diabetes mellitus either in families or by comparing patients and controls. In conclusion, we cannot confirm a role of HERV-K18 polymorphisms -HERV-8914 and HERV-8594- or of the CD48 CA repeat for type 1 diabetes susceptibility.

HLA-DRB1*0826 and HLA-DQB1*0627, two novel class II alleles identified in blood stem cell donors of Caucasian origin

This report describes two novel HLA class II alleles, HLA-DRB1*0826 and HLA-DQB1*0627, that have been identified in two unrelated voluntary blood stem cell donors of Caucasian origin. HLA-DRB1*0826 is characterized by a nucleotide substitution (G to T) in exon 2 at position 163, leading to an amino acid exchange from argenine to leucine. The donor phenotype is HLA-A*0301,*2902; B*3501,*4403; Cw*0401,*1601; DRB1*0101,*0826; DQB1*0402, *0501. The HLA-DQB1*0627 alleles contain a nucleotide substitution at position 184 (T to C) resulting in an amino acid exchange from tyrosine to histidine. Family segregation analysis revealed that the HLA-DQB1*0627 allele belongs to the haplotype A*0101, B*1517, Cw*0701, DRB1*1302, DQB1*0627. The donor phenotype is HLA-A*0101; B*0801,*1517; Cw*0701; DRB1*1302,*1501; DQB1*0602,*0627.

Quantification of chimerism within peripheral blood, bone marrow and purified leukocyte subsets: comparison of singleplex and multiplex PCR amplification of short tandem repeat (STR) loci

OBJECTIVE AND METHODS

Chimerism analysis has become a routine diagnostic procedure after haematopoietic allogeneic stem cell transplantation for early detection of relapse of disease or graft failure. Whereas some centres developed individual in-house short tandem repeat (STR) systems, others prefer commercial multiplex PCR systems. However, little is known about inter-assay variation, which could have a significant impact on treatment decision. We therefore compared two commercial multiplex PCR kits with our in-house STR system using different sample sources, such as peripheral blood (PB), bone marrow (BM) and specific leukocyte subsets.

RESULTS

Fifty samples of eighteen paediatric patients were analysed. For neither material, PB, BM and leukocyte subtypes, a significant difference between the STR systems tested was observed. Chimerism analyses of each single STR primer, which is component of both the in-house and the commercial STR system, did not reveal significant differences.

CONCLUSION

Our analysis demonstrates that similar results can be obtained with both assays, even when using various sample sources. Further evaluation of different test systems will help to increase interlaboratory standardisation of chimerism analyses for early clinical intervention.

Intergenic recombination with HLA-C leads to a novel HLA-A*19 allele, HLA-A*2910+, that is characterized by a functionally inactive amino acid exchange in the loop connecting the alpha2 and alpha3 domains

This report describes the HLA-A*29 allele (A*2910) that has been identified by sequence-based typing in an 8-year-old Turkish female with leukaemia during search for a family-related stem cell donor. The allele is characterized by a nucleotide substitution (Guanine to Adenine) in exon 3 at position 258, leading to an amino acid exchange from glutamic acid to lysine at position 177. From family analysis and sequence comparison, the HLA-A*2910 allele has arisen from intergenic recombination with HLA-C. Structurally, the amino acid exchange at position 177 is probably functionally inactive due to the location of this amino acid exchange in the loop connecting the alpha(2) and alpha(3) domains.

The sensitivity of the alkaline comet assay in detecting DNA lesions induced by X rays, gamma rays and alpha particles

Experiments were designed and performed in order to investigate whether or not the different cellular energy deposition patterns of photon radiation with different energies (29 kV, 220 kV X rays; Co-60, Cs-137-gamma-rays) and alpha-radiation from an Am-241 source differ in DNA damage induction capacity in human cells. For this purpose, the alkaline comet assay (single cell gel electrophoresis) was applied to measure the amount of DNA damage in relation to the dose received. The comet assay data for the parameters '% DNA in the tail' and 'tail moment' for human peripheral lymphocytes did not indicate any difference in the initial radiation damage produced by 29 kV X rays relative to the reference radiations, 220 kV X rays and the gamma rays, whether for the total mean dose range of 0-3 Gy nor in the low-dose range. In contrast, when the 'tail length' data were analysed saturation of the fitted dose response curve appeared for X rays at about 1.5 Gy but was not apparent for gamma rays up to 3 Gy. Preliminary data for alpha exposures of HSC45-M2 cells showed a significant increase in DNA damage only at high doses (>2 Gy Am-241), but the damage at 2 Gy exceeded the damage induced at 2 Gy by Cs-137-gamma-rays by a factor of 2.5. In contrast, other experiments involving different cell systems and DNA damage indicators such as chromosomal aberrations have detected a significant increase in DNA damage at much lower doses, that is at 0.02 Gy for Am-241 and depicte a higher biological effectiveness. These results indicate that differences in biological effects arise through downstream processing of complex DNA damage.

Vitamin D receptor polymorphisms are associated with Graves' disease in German and Polish but not in Serbian patients

Diverse genes are candidates for susceptibility to Graves' disease, including the vitamin D receptor (VDR), which regulates the transcription of target genes in response to the active metabolite 1,25(OH)(2)D(3). We analyzed four polymorphisms of the VDR gene (ApaI, TaqI, BsmI, and FokI) in patients with Graves' disease (n = 789) and healthy controls (n = 823) from three European populations (German, Polish, and Serbian). The VDR ApaI (rs7975232) and TaqI (rs731236) polymorphisms showed no significant difference in any population. The BsmI (rs1544410) variant "b" was associated with Graves' disease in the Polish population (p = 0.0070). The FokI (rs10735810) variant "f " was found to be associated with Graves' disease in Germans and "F" in Polish patients (p = 0.0024 and 0.0049, respectively). Construction of haplotypes for TaqI, ApaI, and BsmI showed the haplotype "Tab" to be the most frequent in the German and Polish population as well as in the Serbian patients, while "tAB" in Serbian controls. Our results show an association of VDR gene polymorphisms in the German and Polish population but not in the Serbian. Furthermore, the VDR polymorphisms are differentially distributed in the three populations. Therefore, VDR polymorphisms analysis needs to be stratified according to the population background.

Prolonged preterm rupture of fetal membranes, a consequence of an increased maternal anti-fetal T cell responsiveness

A fetus, although semi-allogeneic, is usually accepted by the maternal immune system. However, complications, including alloresponsive mechanisms, are thought to be potentially detrimental for a successful pregnancy. Therefore, we compared allogeneic T cell responses of nonpregnant women with the response of healthy pregnant women and pregnant women who have various gestation-associated diseases. Peripheral blood mononuclear cells (PBMCs) of all three groups were stimulated with PBMCs from unrelated volunteers. Pregnant women had significantly reduced stimulation indices (SIs) compared with nonpregnant women. Exposing PBMCs from pregnant women to PBMCs of their own fetus led to a further significant decrease of SIs. Among the two groups of pregnant individuals, SIs of women with prolonged preterm rupture of fetal membranes (PPROM) were significantly higher when the maternal PBMCs were stimulated with PBMCs of their own fetus. This phenomenon could not be observed after stimulation with PBMCs from unrelated volunteers. In addition, an increased humoral immune response was assessed for women with PPROM in comparison with women with uncontrollable preterm labor. Our results revealed a strongly reduced allogeneic T cell response of PBMCs from pregnant women that was further down-regulated when PBMCs from their own fetus were used as stimulators. By contrast, data from women with PPROM suggest an increased maternal T cell response specifically toward the fetal HLA antigens.

IL-2 activated NK cell immunotherapy of three children after haploidentical stem cell transplantation

Natural killer (NK) cells are thought to be of benefit in HLA-mismatched hematopoietic transplantation (H-SCT). Therefore, we developed a protocol for clinical-use expansion of highly enriched and IL-2-stimulated NK cells. Purification of unstimulated leukaphereses by a two-step T cell depletion with a final CD56 enrichment procedure leads to a mean purity of 95% CD56(+)CD3- NK cells with a four- to five-log depletion of T cells. So far, three pediatric patients with multiply relapsed acute lymphoblastic leukemia (ALL) or acute myelogenous leukemia (AML) were treated with repeated transfusions post-H-SCT. Directed killer immunoglobulin-like receptor (KIR) mismatches were demonstrated in all three cases. Although all patients showed blast persistence at the time of transplant, they reached complete remission and complete donor chimerism within 1 month post-H-SCT. NK cell therapy was tolerated well without graft-versus-host disease (GvHD) induction or other adverse events. The AML patient died of early relapse on day +80, while the ALL patients died of thrombotic-thrombocytopenic purpura and atypical viral pneumonia on days +45 and +152, respectively. This initial trial showed the feasibility of good manufacturing practice (GMP)-compliant NK cell isolation and expansion for clinical applications. We now launch a clinical phase I trial with activated NK cells post-H-SCT.

ABCB1 genotype of the donor but not of the recipient is a major risk factor for cyclosporine-related nephrotoxicity after renal transplantation

Cyclosporine (CsA) nephrotoxicity is a severe complication in organ transplantation because it leads to impaired renal function and chronic allograft nephropathy, which is a major predictor of graft loss. Animal models and in vivo studies indicate that the transmembrane efflux pump P-glycoprotein contributes substantially to CsA nephrotoxicity. It was hypothesized that the TT genotype at the ABCB1 3435C-->T polymorphism, which is associated with decreased expression of P-glycoprotein in renal tissue, is a risk factor for developing CsA nephrotoxicity. In a case-control study, 18 of 97 patients developed CsA nephrotoxicity and showed complete recovery of renal function in all cases when switched to a calcineurin inhibitor-free regimen. Both recipients and donors were genotyped for ABCB1 polymorphisms at the positions 3435C-->T and 2677G-->T/A. For controlling for population stratification, two additional polymorphisms, CYP2D6*4 and CYP3A5*3, with intermediate allelic frequencies were studied. The P-glycoprotein low expressor genotype 3435TT only of renal organ donors but not of the recipients was overrepresented in patients with CsA nephrotoxicity as compared with patients without toxicity (chi2 = 10.5; P = 0.005). CsA dosage, trough levels, and the concentration per dose ratio were not different between the patient groups. In a multivariate model that included several other nongenetic covariates, only the donor's ABCB1 3435TT genotype was strongly associated with CsA nephrotoxicity (odds ratio, 13.4; 95% confidence interval, 1.2 to 148; P = 0.034). A dominant role of the donor's ABCB1 genotype was identified for development of CsA nephrotoxicity. This suggests that P-glycoprotein is an important factor in CsA nephrotoxicity.

Anti-fetal immune response mechanisms may be involved in the pathogenesis of placental abruption

Placental abruption is an unpredictable severe complication in pregnancy. In order to investigate the possibility that the activation of the fetal nonadaptive immune system may be involved in the pathogenesis of this disease, IL-6 release from cord blood monocytes was examined by intracellular cytokine staining and flow cytometric analysis. Our results demonstrate that preterm placental abruption (n = 15) in contrast to uncontrollable preterm labor (n = 33) is associated with significantly (P < 0.001) increased release of IL-6 from the fetal monocytes. The same holds true for rhesus disease (n = 9, P < 0.001) that is characterized by a maternal production of antibodies against the rhesus-D antigen expressed by the fetal erythrocytes. This suggests that during rhesus disease, IL-6 release of monocytes is induced by antibody-mediated cross-linking of these cells to the erythrocytes in the fetal circulation. Hence, this assumption favors the idea that also in case of placental abruption, an increased maternal antibody production against paternal antigens leads to an elevated IL-6 release by the fetal monocytes. To elucidate this potential mechanism, the presence of anti-HLA-antibodies was assessed in the maternal circulation of patients with placental abruption (n = 17) and patients with uncontrollable preterm labor (n = 29). The percentage of women producing anti-paternal HLA-antibodies was significantly (P < 0.01) increased in the group of women with preterm placental abruption (47%) in comparison to women with uncontrollable preterm labor (14%). Therefore, our results suggest that an increased humoral immune response of the mother against the fetus may be decisively involved in the pathogenesis of placental abruption.

Assessment of killer cell immunoglobulinlike receptor expression and corresponding HLA class I phenotypes demonstrates heterogenous KIR expression independent of anticipated HLA class I ligands

Natural killer (NK) cell-mediated cytolysis is stimulated and downregulated through the interaction of distinct human leukocyte antigen (HLA) class I molecules on target cells with specific killer cell immunoglobulinlike receptors (KIRs) on NK cells. Killer cell immunoglobulinlike receptors are highly polymorphic and are clonally distributed on NK cell populations within individuals. However, the regulation of KIR expression by individual HLA class I phenotypes is not well understood. To examine a potential influence of the HLA class I phenotype on KIR expression patterns we studied the KIR expression in individuals that were subgrouped according to the major HLA-C encoded KIR-epitopes (group C1 versus C2). In these individuals, NK cells were analyzed for KIR expression using flow cytometry and RNA-based expression analysis. Our results demonstrate that KIR genes are transmitted very heterogeneously with two main patterns of KIR genotypes as previously described; group A and group B (with 21 different genotypes). There are distinct populations exhibiting different densities of CD158a and/or CD158b positive NK cells that coexist in all individuals. A clear correlation between KIR expression and the currently known HLA class I ligands was not observed. In conclusion, the surface expression of KIRs in individuals with different HLA class I genotypes indicates that other non-HLA class I encoded factors contribute to the shaping of the KIR repertoire.

Unbalanced amounts of HLA-DQA1 allele mRNA: DQA1*03 shows high and DQA1*0501 low amounts of mRNA in heterozygous individuals

Genes of the HLA-DR, DQ region confer strong susceptibility to type 1 diabetes mellitus (IDDM). A possible mechanism of susceptibility is a difference in the amounts of transcripts of predisposing and neutral or protective haplotypes. In this study we developed an assay to compare the amounts of mRNA of two distinct HLA-DQA1 alleles in peripheral blood lymphocytes (PBLs) of heterozygous individuals, using a quantitative RT-PCR with an internal standard covering all HLA-DQA1 specifities. We also developed an algorithm to calculate the amounts of mRNA for two distinct alleles in heterozygous individuals based on the comparison to the same internal standard. In total, 37 HLA-DQA1 heterozygous individuals were analysed, including patients with IDDM (n = 14) and healthy controls (n = 23). Intra-individually, we observed different amounts of mRNA for different HLA-DQA1 alleles in the order: HLA-DQA1*03 > *01 > *0201 > *05. This order was observed in all individuals. We also observed a variation in the ratio of these unbalanced amounts of mRNA in individuals with the same HLA-DQA1 allele combinations. In all allele combinations the average ratio was increased in patients with IDDM compared to the control samples. HLA-DQA1*03 positive and DQA1*03, *05 heterozygous patients had the highest average ratios. Nevertheless, based on limited sample numbers, these differences did not reach significance. We therefore conclude that variations between HLA-DQA1 alleles are not limited to the nucleotide sequence but are also found at the level of amounts of mRNA.

Preliminary evidence that an endogenous retroviral long-terminal repeat (LTR13) at the HLA-DQB1 gene locus confers susceptibility to Addison's disease

OBJECTIVE

Addison's disease is associated with particular haplotypes of the human leucocyte antigen (HLA) region [DQA1*0501-DQB1*0201 (DQ2) and DQA1*0301-DQB1*0302 (DQ8)]. This locus harbours several human endogenous retroviral (HERV) long-terminal repeats (LTR). LTRs within the HLA region have been shown to confer additional susceptibility to type 1 diabetes and rheumatoid arthritis.

DESIGN

We investigated the role of LTR3 and LTR13, both of which are located adjacent to the DQB1 gene, in Addison's disease.

PATIENTS

Eighty-seven patients and 160 controls were genotyped for HLA-DQA, -DQB, and the presence or absence of LTR3 and LTR13.

RESULTS

Significantly more patients' HLA alleles than those of controls carried the LTR13 insertion (19.0% vs. 10.6%, P = 0.0143), whereas there was only a trend for LTR3 (allele-wise chi-squared test: P = 0.0941). Both, LTR3 and LTR13 are in strong linkage disequilibrium with DQ8, which itself was significantly more frequent in patients than in controls (29.9% vs. 15.0%, P = 0.0089). However, significantly more alleles of DQ8+ patients than of DQ8+ controls carried the LTR13 insertion (44.2% vs. 18.8%, P = 0.0119), whereas we did not observe any difference for LTR3 in the DQ8+ subset (30.5 vs. 23.1%, P = 0.9416).

CONCLUSIONS

We have found preliminary evidence that the endogenous retroviral element DQ-LTR13, but not LTR3, is associated with Addison's disease. LTR13 appears to enhance HLA-DQ8 mediated disease risk. This retroviral insertion therefore might represent a novel susceptibility factor in Addison's disease, but these findings need to be confirmed in a larger data set.