Strategies and technical challenges in allele level Class II typing in 2578 bone marrow transplantation donor-recipient pairs.

Human leukocyte antigen typing of 2578 donor-recipient pairs whose transplantation was facilitated by the National Marrow Donor Program allowed for an in-depth analysis of the accuracy of high-volume allele level testing data. The methods employed provided allele level typing at DRB1/3/5, DQA1, DQB1, DPA1, and DPB1 using sequence-specific oligonucleotide probe hybridization (SSOPH), polymerase chain reaction (PCR) restriction fragment length polymorphism analysis, sequence specific PCR, and direct sequence-based typing (SBT). Each typing was independently tested by two laboratories in Phase 1, and in subsequent phases targeted samples were typed in duplicate by SBT to monitor typing quality. Comparison with prior transplant center typing was also evaluated. SSOPH detected discrepancies ranged from 0.6% at DPB1 to 5.1% at DQB1 in Phase 1. The majority of discrepancies, 62%, resulted from human error such as sample handling, result interpretation, or clerical errors. Alleles that are frequently discrepant have been identified in this predominantly white population.

A genetic explanation for the rising incidence of type 1 diabetes, a polygenic disease.

We had earlier hypothesized, if parents originated from previously isolated populations that had selected against different critical susceptibility genes for a polygenic disease, their offspring could have a greater risk of that disease than either parent. We therefore studied parents of patients with type 1 diabetes (T1D). We found that parents who transmitted HLA-DR3 to HLA-DR3/DR4 patients had different HLA-A allele frequencies on the non-transmitted HLA haplotype than HLA-DR4-transmitters. HLA-DR3-positive parents also had different insulin (INS) gene allele frequencies than HLA-DR4-positive parents. Parent pairs of patients had greater self-reported ethnicity disparity than parent pairs in control families. Although there was an excess of HLA-DR3/DR4 heterozygotes among type 1 diabetes patients, there were significantly fewer HLA-DR3/DR4 heterozygous parents of patients than expected. These findings are consistent with HLA-DR and INS VNTR alleles marking both disease susceptibility and separate Caucasian parental subpopulations. Our hypothesis thus explains some seemingly disconnected puzzling phenomena, including (1) the rising world-wide incidence of T1D, (2) the excess of HLA-DR3/DR4 heterozygotes among patients, (3) the changing frequency of HLA-DR3/DR4 heterozygotes and of susceptibility alleles in general in patients over the past several decades, and (4) the association of INS alleles with specific HLA-DR alleles in patients with T1D.

Conserved extended haplotypes discriminate HLA-DR3-homozygous Basque patients with type 1 diabetes mellitus and celiac disease.

The major susceptibility locus for type 1 diabetes mellitus (T1D) maps to the human lymphocyte antigen (HLA) class II region in the major histocompatibility complex on chromosome 6p21. In southern European populations, like the Basques, the greatest risk to T1D is associated with DR3 homo- and heterozygosity and is comparable to that of DR3/DR4, the highest risk genotype in northern European populations. Celiac disease (CD) is another DR3-associated autoimmune disorder showing certain overlap with T1D that has been explained by the involvement of common genetic determinants, a situation more frequent in DR3-rich populations, like the Basques. As both T1D- and CD-associated HLA alleles are part of conserved extended haplotypes (CEH), we compared DR3-homozygous T1D and CD patients to determine whether CEHs were equally distributed between both disorders or there was a differential contribution of different haplotypes. We observed a very pronounced distribution bias (P<10(-5)) of the two major DR3 CEHs, with DR3-B18 predominating in T1D and DR3-B8 in CD. Additionally, high-density single nucleotide polymorphism (SNP) analysis of the complete CEH [A*30-B*18-MICA*4-F1C30-DRB1*0301-DQB1*0201-DPB1*0202] revealed extraordinary conservation throughout the 4.9 Mbp analyzed supporting the existence of additional diabetogenic variants (other than HLA-DRB1*0301-DQB1*0201), conserved within the DR3-B18 CEH (but not in other DR3 haplotypes) that could explain its enhanced diabetogenicity.

Mendelian inheritance of polygenic diseases: a hypothetical basis for increasing incidence.

The incidence of common polygenic diseases, such as type 1 diabetes, bronchial asthma, and gluten-sensitive enteropathy, is increasing. Although this is usually attributed to environmental factors, it is possible that this rising incidence also has a genetic basis. The hypothesis is put forth that, in the past, these diseases, with their increased morbidity and mortality, were selected against. In contrast to monogenic diseases, the incidence of polygenic diseases can be reduced by selection against susceptibility alleles of any of the genetic loci necessary for disease to occur. In different isolated populations, different disease susceptibility loci may have been selected against. Parents who derive from different isolated populations in which there are inversely different susceptibility allele frequencies because of selection or genetic drift, would be expected to have offspring with an increased risk for that polygenic disease. It is shown mathematically that the incidence of a hypothetical polygenic disease increases under these circumstances. The increased risk in these offspring results from a kind of genetic complementation in which they have inherited a more complete set of susceptibility alleles at all susceptibility loci than is carried by either of their parents. Hallmarks of this hypothesized phenomenon would be increased heterozygosity for specific population markers (whether susceptibility alleles or not) among the disease-affected offspring and a paucity of such heterozygotes among their parents. The parents and patients would also be expected to give more evidence of ethnic or subethnic disparity than that observed in controls.

Inheritable variable sizes of DNA stretches in the human MHC: conserved extended haplotypes and their fragments or blocks.

The difference in sizes of conserved stretches of DNA sequence within the major histocompatibility complex (MHC) in human individuals constitutes an underappreciated genetic diversity that has many practical implications. We developed a model to describe the variable sizes of stretches of conserved DNA in the MHC using the known frequencies of four different kinds of small (< 0.2 Mb) blocks of relatively conserved DNA sequence: HLA-Cw/B; TNF; complotype; and HLA-DR/DQ. Each of these small blocks is composed of two or more alleles of closely linked loci inherited as one genetic unit. We updated the concept of the conserved extended haplotype (CEH) using HLA allele identification and TNF microsatellites to show that specific combinations of the four blocks form single genetic units (>/= 1.5 Mb) with a total haplotype frequency in the Caucasian population of 0.30. Some CEHs extend to the HLA-A and -DPB1 loci forming fixed genetic units of up to at least 3.2 Mb of DNA. Finally, intermediate fragments of CEHs also exist, which are, nevertheless, larger than any of the four small blocks. This complexity of genetic fixity at various levels should be taken into account in studies of genetic disease association, immune response control, and human diversity. This knowledge could also be used for matching CEHs and their fragments for patients undergoing allotransplantation.

Polymorphism of the human retinoid X receptor beta and linkage disequilibrium with HLA-DPB1.

The human retinoid X receptor beta (RXRB) gene is localized in the major histocompatibility complex (MHC) region between DPB1 and RING2. The RXRB gene sequence reported by different investigators suggests that the gene may be polymorphic. In this study, we confirmed one polymorphism by sequencing genomic DNA from four Caucasian individuals. We also developed a restriction fragment length polymorphism (RFLP) analysis to detect this specific polymorphism. Linkage analysis studies between RXRB alleles and a number of HLA markers showed significant linkage disequilibrium between RXRB*T and HLA-DPB1*0401.

Hepatitis B surface antigen- and tetanus toxoid-specific clonal expansion of CD4+ cells in vitro determined by TCRBV CDR3 length and nucleotide sequence.

We demonstrate activation of primary human TCRBV-specific CD4+ cells in vitro towards hepatitis B surface antigen (HBsAg) and tetanus toxoid (TT) without the use of cell lines, clones or added cytokines. By multiplex PCR analysis and spectratyping, antigen-activated cells exhibited clonal T cell receptor expansion within specific and limited TCRBV families. The expanded CD4+ T cells were CD45RO. Three of four unrelated HBsAg responders showed CD4+ expansion within the TCRBV16 family. The response comprised predominantly single CDR3 sequences in all three donors and was completely monoclonal in one of them. However, the CDR3 lengths and sequences differed among the responders. Clonality induced by HBsAg in TCRBV16 was specific, reproducible and distinct from that induced by TT in terms of sequence, nucleotide addition and diversity (BD) or junctional (BJ) element usage. Thus, for the first time, we show monoclonal or oligoclonal expansion of primary human CD4- peripheral blood mononuclear cells (PBMC) in vitro in response to nominal protein antigen without manipulations utilizing exogenous IL-2. The ability to induce monoclonal/ oligoclonal responses to HBsAg now permits motif identification studies for determining the T cell role in nonresponsiveness to the HBsAg vaccine.

Incomplete penetrance of MHC susceptibility genes: prospective analysis of polygenic MHC-determined traits.

We propose an approach to understanding incomplete penetrance of disease susceptibility genes as a method of studying the underlying mechanisms of polygenic diseases. Incomplete penetrance is the failure of genetically susceptible individuals to exhibit a trait. We define as baseline penetrance that which occurs in genetically identical (monozygotic) twins of an index subject with a major histocompatibility complex (MHC)-associated disease or trait. We consider two mechanisms for incomplete baseline penetrance: an extrinsic (environmental) trigger and an intrinsic stochastic, gene-associated process. The latter can be detected for dominant expression because susceptibility genes in homozygotes (with their two intrinsic triggers) will be up to twice as frequently penetrant as those in heterozygotes. The extent of MHC and non-MHC gene contribution determines differences between baseline penetrance and apparent penetrance in MHC-identical sib pairs, sib pairs in general and MHC-identical unrelated individuals. Inheritance patterns in families do not reveal modes of inheritance of incompletely penetrant polygenic MHC-determined traits. A method is proposed to study such traits prospectively in persons presumed to be homozygous, heterozygous or non-carrying for susceptibility genes by determining trait expression in homozygotes, heterozygotes or non-carriers of trait-associated conserved extended MHC haplotypes. The method provides direct estimates of apparent penetrance rates, modes of genetic determination, and, if the trait is dominant, the origin of penetrance. When applied to dominant MHC susceptibility gene-determined immunoglobulin deficiencies in two populations, the ratios of affected haplotype homozygotes to heterozygotes near 2.0 were consistent with an intrinsic mechanism for baseline penetrance acting on the MHC susceptibility genes.

Prospective analysis suggests susceptibility genes for deficiencies of IgA and several other immunoglobulins on the [HLA-B8, SC01, DR3] conserved extended haplotype.

The extended major histocompatibility complex (MHC) haplotype [HLA-B8, SC01, DR3] is increased in frequency among patients with immunoglobulin (Ig)A deficiency and common variable immunodeficiency. Because the genomic region from HLA-B to HLA-DR/DQ is virtually the same on all instances of the haplotype in the general population, we reasoned that all independent instances of [HLA-B8, SC01, DR3] carry MHC susceptibility genes for these disorders. To define immunoglobulin deficiencies determined by genes on this haplotype and their mode of expression and penetrance, serum immunoglobulin class and IgG subclass concentrations were determined prospectively in homozygotes and heterozygotes of this haplotype and in Caucasian controls. Prevalence of individual immunoglobulin deficiencies in persons with [HLA-B8, SC01, DR3] ranged from 13% to 37%, significantly higher than rates in non-carriers or general controls. We found significantly increased frequencies of IgA and IgG4 deficiency only in homozygotes (13.3% and 30%, respectively) compared with heterozygotes (1.7% and 3.4%) or non-carriers (1.6% each), suggesting recessive expression. In contrast, IgD and IgG3 deficiencies were significantly more common in both homozygotes (36.7% and 30%) and heterozygotes (20.3% and 17.5%) compared with controls (4.9% and 3.4%), suggesting dominant inheritance. These results indicate multiple distinct susceptibility genes, some recessive and others dominant, for deficiency of IgA, IgD, IgG3 or IgG4 (but not for IgE, IgG1, IgG2 or IgM) on [HLA-B8, SC01, DR3]. These observations may also help to explain the observed associations of [HLA-B8, SC01, DR3] with both IgA deficiency and common variable immunodeficiency and the common occurrence of IgG subclass deficiencies in some patients with IgA deficiency.

The [HLA-B18, F1C30, DR3] conserved extended haplotype carries a susceptibility gene for IgD deficiency.

We showed previously that the conserved extended MHC haplotype [HLA-B8, SCO1, DR3] carries recessive susceptibility genes for IgA and IgG4 deficiency and dominant genes for IgD and IgG3 deficiency. [HLA-B18, F1C30, DR3] has similar class II and III regions to [HLA-B8, SC01, DR3] and is common in the Basques. We therefore studied serum immunoglobulin concentrations in Basque homozygotes, heterozygotes, and noncarriers of (FIC30, DRB1*0301, DRB3*02, DQA1*0501. DQB1*0201) (F1C30, DR3). As shown by others, no subjects were deficient in IgA, IgM, or IgG subclasses. In contrast, 29% of homozygotes and three of seven double heterozygotes with (SC01, DRB1*0301, DRB3*0101, DQA1*0501, DQB1*0201) (presumed homozygotes for IgD deficiency susceptibility genes) were IgD deficient. Thus, 32% of presumed homozygotes were IgD deficient compared with 1.6% of noncarriers. Of haplotype heterozygotes, 25% were IgD deficient. The high frequency of IgD deficiency in both homozygotes and heterozygotes for (F1C30, DR3) suggests a partially penetrant dominant susceptibility gene for IgD deficiency on [HLA-B18, F1C30, DR3].

A simple estimate of the general population frequency of the MHC susceptibility gene for autoimmune polygenic disease.

We wished to determine the frequencies of the MHC and non-MHC susceptibility genes for polygenic autoimmune diseases like type 1 diabetes (IDDM). We used Mendelian inheritance and the Hardy-Weinberg equilibrium to calculate the frequencies of mating pairs and susceptible offspring under classical recessive and dominant inheritance of the MHC susceptibility gene. We then analyzed the distribution of haplotype sharing by affected sib pairs of the 4 MHC haplotypes in each of the kinds of mating pairs in terms of the frequency of the disease susceptibility gene. For IDDM, the analysis was consistent with a recessive, but not a dominant, MHC susceptibility gene of frequency 0.525 at a distribution of 55, 38 and 7% of affected sib pairs who share 2, 1 and 0 MHC haplotypes, respectively. A simple relationship was obtained: if inheritance is recessive, the MHC susceptibility gene frequency is the square root of the fraction of affected sib pairs who share no MHC haplotypes multiplied by 4. For recessive inheritance, affected sib pairs who share no haplotypes are solely in families where both parents are homozygous MHC-susceptible. Although homozygous MHC susceptibles represent over 25% of the population, only 2-3% of them are IDDM-susceptible at non-MHC susceptibility loci, also required for disease expression. Predictions from our analysis fit all published observations of the familial occurrence of disease. The analysis is general, simple and provides a single estimate (not a range) of the MHC susceptibility gene frequency. This approach should be applicable to other MHC-determined polygenic diseases.

The extent of HLA class II allele level disparity in unrelated bone marrow transplantation: analysis of 1259 National Marrow Donor Program donor-recipient pairs.

A comprehensive analysis of the HLA-D region loci, DRB1, DRB3, DRB5, DQA1, DQB1, DPA1 and DPB1, was performed to determine allelic diversity and underlying HLA disparity in 1259 bone marrow recipients and their unrelated donors transplanted through the National Marrow Donor Program. Although 43.0% of DRB1 alleles known to exist at the beginning of the study were found in this predominantly Caucasian transplant population, a few alleles predominated at each locus. In recipients, 67.1% of DRB1 alleles identified were one or two of six common DRB1 alleles. Only 118 (9.4%) donor-recipient pairs were matched for all alleles of DRB1, DQA1, DQB1, DPA1 and DPB1. While 79.4% of the pairs were matched for DRB1, only 13.2% were matched for DPB1 alleles. Almost 66% of pairs differed by more than one allele mismatch and 59.0% differed at more than one HLA-D locus. DQB1 was matched in 85.9% of DRB1-matched pairs. In contrast, only 13.9% of the pairs matched for DRB1, DQA1 and DQB1 were also matched for DPA1 and DPB1. This database, highlighting the underlying HLA disparity within the pairs, forms the foundation of an ongoing study to establish the relationship between HLA matching and successful outcome in unrelated allogeneic stem cell transplant.

HLA-DRB1 leprogenic motifs in nigerian population groups.

Amino acid residues involved in the peptide binding groove of HLA-DRB1 alleles were examined in three Nigerian ethnic groups with leprosy (n = 287) and 170 controls to determine the role of DRB1 alleles in disease outcome with Mycobacterium leprae. Nine positively charged motifs and two others with neutral charge to the binding groove were detected. These motifs occurred more frequently in leprosy (leprogenic) than was expected by chance (P < 0.0001). In contrast, five motifs with net negative or 'modified' neutral charges to the pocket were negatively associated with leprosy. We conclude that clinical outcome of infection with M. leprae is largely determined by a shared epitope in DRB1 alleles marked by several motifs. These motifs occur in otherwise normal DRB1 alleles, characterized by net positive or neutral charges in the binding groove. We hypothesize that these polarities cause poor binding of DRB1 to M. leprae. On presentation, the signal via the T cell receptor results in muted cell-mediated immunity. The resulting response translates to various forms of leprosy depending on degree of charge consonance between M. leprae and host DRB1 allele. Other factors within or without the HLA complex, such as the T cell receptor repertoire, may also influence the resulting disease.

CD4 TCRBV CDR3 analysis in prevalent SLE cases from two ethnic groups.

We examined CD4+ T cell TCRBV-CDR3 transcripts from 19 lupus patients and 16 controls to test the hypothesis that CD4+ TCRBV-CDR3 expression in SLE differs from normals. Within the disease group we also performed exploratory analyses to determine the association between risk of oligoclonality and HLA-DRB specificities and the duration of the CDR3 patterns. Oligoclonal patterns consistent with CDR3 restriction were three times more likely in SLE than in controls (OR = 3.7). TCRBV1, BV4, BV5.1, BV7, BV9, BV18 and BV22 gene segment CDR3 patterns of oligoclonality were seen exclusively among lupus patients. HLA-DRB3 increased the risk of oligoclonal expression in SLE. In four patients studied over time, the pattern of TCRBV-CDR3 expression was stable in a second sample obtained 6-14 months later. The increased frequency of CD4+ T cell TCRBV-CDR3 oligoclonal expression in SLE when compared to controls and the persistence of these patterns are consistent with an expanded pool of autoreactive CD4 T cells in SLE which recognize peptides derived from autoantigens. The association of HLA-DRB3 genes with increased risk of CDR3 oligoclonality among the SLE subjects is compatible with the hypothesis that molecules encoded by HLA-DRB3 may facilitate autoantigen recognition by CD4 T cells.

HLA-Cw alleles associated with HLA extended haplotypes and C2 deficiency.

There are four MHC-linked complement genes, BF, C2, C4A and C4B, that are inherited as single DNA units, known as complotypes. Extended haplotypes were initially defined by studying the distribution of complotypes in relation to HLA-B and HLA-DR loci in Caucasian families. In order to analyze the distribution of HLA-Cw alleles in relation to extended haplotypes, we studied a large panel of MHC homozygous and heterozygous cell lines representing previously described Caucasian-derived extended haplotypes and 14 patients with complete C2 deficiency. HLA alleles were assigned using sequence-specific oligonucleotide probe hybridization (SSOP). Family analysis served to assign haplotypes for heterozygous samples. We found distinctive HLA-Cw alleles for each independent extended haplotype. Their association in each instance was statistically significant. All patients with C2 deficiency carrying the haplotype [HLA-B18, S042, DR2] were associated with HLA-Cw*1203. These conserved allelic combinations may become an important tool for the study of human evolution and may contribute to the expeditious selection of prospective donors in clinical transplantation.

Characterization of non-expressed C4 genes in a case of complete C4 deficiency: identification of a novel point mutation leading to a premature stop codon.

The genetic basis of complete C4 deficiency in a patient with SLE was investigated. Previous studies have demonstrated that this patient has two different major histocompatibility complex (MHC) haplotypes that each contain a major deletion and a non-expressed C4 gene. In the present study, non-expression of the C4 genes was explained by the finding of two distinct C4 gene mutations. A previously described two base pair insertion in exon 29 of the C4 gene was detected in the paternal MHC haplotype [HLA-A2, B40, SC00, DR6]. The maternal haplotype [HLA-A30, B18, F1C00, DR3] carried a C4 gene with a one base pair deletion in exon 20 generating a premature stop codon. This mutation was neither found in 10 individuals with known non-expressed C4 genes nor in 9 individuals homozygous for the complotype F1C30. The isotype and allotype specific regions of the patient's C4 genes were sequenced, and both contained C4A3a sequence. In conclusion, two different MHC haplotypes resembling the extended haplotypes [HLA-A2, B40, SC02, DR6] and [HLA-A30, B18, F1C30, DR3] both contained a non-expressed C4A gene that was due to either of two distinct mutations, demonstrating the heterogeneous genetic background of C4 deficiency.

Relationship between protein complotypes and DNA variant haplotypes: complotype-RFLP constellations (CRC).

From the study of 52 families and 15 homozygous typing cells, 234 MHC complement haplotypes were characterized for features in the DNA of the complotype region: C2/Sst I (2.75, 2.70, 2.65, and 2.40 kb), BF/Taq I (6.6 and 4.5 kb), C4 5'/Bgl II (15 and 4.5 kb), C4 5'/Taq I (7.0, 6.4, 6.0 and 5.4 kb) and C4 3'/Xba I/BamH I (11 and 4 + 7 kb) restriction fragment length polymorphisms (RFLP's), by the presence or absence of C4A, C4B, CYP21A and CYP21B genes and by duplications. Nineteen (of over 1000 theoretically possible) complotype-RFLP constellations (CRC's) were found. The 9 CRC's with two C4 and CYP21 genes were designated A through I. CRC's Bdup and Ddup were like B and D but had duplicated C4B-CYP21B genes. The remaining CRC's had deletions of C4 and/or CYP21 genes and were designated Bdel, Cdel and the like. Individual complement alleles and complotypes were nor randomly distributed among the CRC's. Some complotypes, such as SC01, SC02 and FIC30, were restricted to only 1 CRC; others, such as SC31, FC31, and SC30, were found in several CRC's. Some of the CRC's contained a single complotype, others contained several. Remarkably, there are about 30 CRC-specified complotypes with frequencies of .01 or higher and 14 of .02 or higher. A number of evolutionary origins of complement alleles and complotypes are suggested by the relationships among CRC's. Approximate normal frequencies of the undeleted CRC's were A = .27, B = .19, Bdup = .02, C = .17, D = .07, Ddup = .02, E = .06, F = .05, and G = .02. Thus, CRC's without deletions accounted for 88% of normal complotypes. Since the frequency of Bdel, with a deletion of C4A, was .12, 10 CRC's accounted for all observed normal caucasian MHC haplotypes.

Defect in Th1-like cells of nonresponders to hepatitis B vaccine.

Peripheral blood lymphocytes from nonresponders to hepatitis B vaccine (HBsAg) failed to undergo a proliferative response to recombinant HBsAg in vitro, whereas cells from responders proliferated vigorously. The lack of proliferative response was not due to defective antigen presentation in that MHC-identical responder and nonresponder antigen presenting cells were equally effective in stimulating responder T cells. Nonresponder T cells did not proliferate in response to antigen-pulsed MHC identical responder antigen presenting cells. The present study demonstrated that: 1) there were no detectable (1 in < 20 x 10(4) HBsAg-precursor T cells in any of the nonresponders, while in responders the frequency of HBsAg-precursor T cells ranged from 1 in 3.2 x 10(3) to 1 in 40 x 10(3); 2) nonresponder cell cultures did not secrete IL-2 in response to HBsAg stimulation; 3) exogenous recombinant IL-2 did not restore the proliferative response of the T cells in HBsAg-pulsed cultures of nonresponders. These results suggest that the cellular basis for the lack of response to HBsAg is a defect in HBsAg-specific Th1-like cells; either there is an absence of the Th1 cells or cells with TCR specificity for HBsAg are present but are unresponsive to the HBsAg peptide-MHC complex (i.e., anergy or tolerance).

Tumor necrosis factor (TNF) microsatellite haplotypes in relation to extended haplotypes, susceptibility to diseases associated with the major histocompatibility complex and TNF secretion.

TNFabc microsatellite haplotypes were determined on normal, type I diabetes and multiple sclerosis Caucasian MHC haplotypes in family studies. Although independent examples of conserved extended haplotypes usually had the same TNFabc haplotypes, there were a number of exceptions, suggesting that these loci are more mutable than most loci in the human MHC. Some TNFabc haplotypes were characteristic of only one extended haplotype, whereas others were shared by several different extended haplotypes. From the analysis of TNFabc on extended haplotype fragments, and assuming that the fragments arose by ancient homologous crossing over, it was possible to "map" TNF and how that it was somewhat closer to HLA-B than the complement region, corresponding to the physical map of this region. TNF haplotype associations with type I diabetes and multiple sclerosis were attributable to the known extended haplotype associations of these diseases. There was also a trend for higher TNF-alpha secretion by peripheral blood mononuclear cells from individuals homozygous for [HLA-B8, SC01, DR3] than from individuals homozygous for [HLA-B7, SC31, DR2].