Erik Thorsby

HLA susceptibility genes in celiac disease: Genetic mapping and role in pathogenesis

The overrepresentation of particular HLA alleles in patients with celiac disease was first noted two decades ago. Several lines of evidence obtained during the last years strongly suggest that a particular HLA-DQ heterodimer, encoded by the DQA1*0501 and DQB1*0201 genes in cis or trans configuration, confers the primary disease susceptibility. This paper reviews the evidence behind this concept and discusses how this particular DQ molecule may be involved in the pathogenesis.

HLA Antigens and Immunoregulatory T Cells in Ulcerative Colitis Associated with Hepatobiliary Disease

Serologic HLA typing was carried out in 20 patients with ulcerative colitis (UC) combined with hepatobiliary disease, in 34 UC patients without hepatobiliary disease, and in control subjects. Association with HLA-B8 and -DR3 was found in both groups of patients. HLA-B8 was found in 80% of patients with combined disease (p less than 0.0005 vs. controls; relative risk (RR), 12.0), whereas 32% of the patients with UC without hepatobiliary disease were HLA-B8-positive (not significant vs. controls). Concomitantly, HLA-DR3 was found in 70% of patients with combined UC and hepatobiliary disease (p less than 0.0005 vs. controls; RR 9.95) and in 35% of UC patients without hepatobiliary disease (p less than 0.05 vs. controls; RR, 2.33). HLA-B8 was found more frequently in UC patients with than without hepatobiliary disease (p less than 0.001; RR, 8.36), as was the case with HLA-DR3 (p less than 0.025; RR, 4.28). No indications of defects in immunoregulatory lymphocytes (T gamma and T mu cells) which could explain hyperactivity in the immune system were observed in the patients with combined UC and hepatobiliary disease. The present study gives support to the theory that UC and, particularly, UC combined with hepatobiliary lesion may be autoimmune diseases with a genetic predisposition.

Vaccination with mutant ras peptides and induction of T-cell responsiveness in pancreatic carcinoma patients carrying the corresponding RAS mutation

Mutations in codon 12 of K-RAS are frequently found in pancreatic adenocarcinomas. T-cell responses specific for individual RAS mutations can be elicited in vitro by stimulation of peripheral blood mononuclear cells with synthetic peptides. Mutant ras peptides are therefore a candidate vaccine for specific immunotherapy in pancreatic carcinoma patients. When vaccinated with a synthetic ras peptide representing the K-RAS mutation in their tumours, a transient ras-specific T-cell response was induced in two of five patients treated. The vaccination protocol involved multiple infusions of large amounts of peptide-pulsed antigen-presenting-cells obtained by leucapheresis. These results indicate that specific T-cell responses against mutations uniquely harboured in tumour cells can be induced in cancer patients by vaccination.

T cells from the small intestinal Mucosa of a DR4, DQ7/DR4. DQ8 celiac disease patient preferentially recognize gliadin when presented by DQ8

CD is an immunologic disease of the small intestine which is precipitated by ingestion of wheat gliadin. Most patients carry the HLA-DQ (alpha 1*0501, beta 1*0201) (DQ2) heterodimer. We recently reported that a preponderance of gliadin-specific T cells from the small intestinal mucosa of DQ2-positive CD patients were restricted by this DQ heterodimer. However, a small percentage of CD patients do not carry this DQ heterodimer, and most of them instead carry DQ (alpha 1*0301, beta 1*0302) (DQ8). Here we report that a majority of gliadin-specific T cells from the small intestinal mucosa of a DR4,DQ7/DR4,DQ8 heterozygous CD patient are restricted by DQ8. Thus, preferential presentation of gliadin-derived peptides to T cells by the CD-associated DQ2 and DQ8 molecules may be an initial and important immunopathogenic step in CD.

Association analysis of the 1858C>T polymorphism in the PTPN22 gene in juvenile idiopathic arthritis and other autoimmune diseases.

A functional single nucleotide polymorphism, 1858C>T, in the PTPN22 gene, encoding a tyrosine phosphatase, has been reported to be associated with type I diabetes and some other autoimmune diseases. To further investigate whether this polymorphism may be a general susceptibility factor for autoimmunity, we performed an association study in five different autoimmune diseases, three previously not tested. We found an association with juvenile idiopathic arthritis (OR=1.41; P=0.04), not previously reported, and a tendency for an association with coeliac disease (OR=1.35; P=0.08). In primary sclerosing cholangitis, no association was observed (OR=0.95; P=0.8). Furthermore, we confirmed the increased risk in rheumatoid arthritis (OR=1.58; P=0.001), but could not find support for an association with systemic lupus erythematosus (OR=0.94; P=0.8). Altogether, we have provided further evidence of an association between autoimmune diseases and the 1858C>T polymorphism in PTPN22.

HLA-Encoded Genetic Predisposition in IDDM: DR4 Subtypes May Be Associated With Different Degrees of Protection

Recent studies have shown that the risk conferred by the high-risk DQAl*03-DQBl*0302 (DQ8) haplotype is modified by the DRB1*O4 allele that is also carried by this haplotype. However, many of these studies suffer from lack of sufficient numbers of DQ-matched control subjects, which are necessary because there is a strong linkage disequilibrium between genes in the HLA complex. In the present study, using a large material of IDDM patients and DQ-matched control subjects, we have addressed the contribution of DR4 subtypes to IDDM susceptibility. Our data, together with recent data from others, clearly demonstrate that some DR4-DQ8 haplotypes are associated with disease susceptibility, while others are associated with protection, depending on the DRB1*O4 allele carried by the same haplotype. In particular, our data demonstrate that DRBl*0401 confers a higher risk than DRBl*0404. Based on combined available data on the genetic susceptibility encoded by various DR4-DQ8 haplotypes and the amino acid composition of the involved DRβ*04 chains as well as the ligand motifs for these DR4 subtypes, we have developed a unifying hypothesis explaining the different risks associated with different DR4-DQ8 haplotypes. We suggest that disease susceptibility is mainly conferred by DQ8 while DR4 subtypes confer different degrees of protection. Some DR4 subtypes (i.e., DRB1*0405, 0402, and 0401) confer little or no protection, while others (i.e., DRB 1*0404, 0403, and 0406) cause an increasing degree of protection, possibly by binding a common protective peptide. Features of a protective peptide that fit such a model are briefly discussed.

HLA‐DR‐like Antigens in the Epithelium of the Human Small Intestine

Sections of ethanol‐fixed, paraffin‐embedded tissue specimens from different parts of the human gastrointestinal tract were stained by an indirect immunofluorescence method with a rabbit antiserum to HLA‐DR‐antigens from B lymphocytes. A specific staining reaction was seen in a patchy pattern apically in the columnar cells of the normal small intestine, decreasing in intensity from the top of the villi towards the crypts. No HLA‐DR‐like antigens could he detected in colon or stomach epithelium, whereas cellS with the morphology of lymphocytes and histiocytes in the lamina propria and also capillary walls were specifically stained throughout the gastrointestinal tract.

HLA-DQA1 and HLA-DQB1 genes may jointly determine susceptibility to develop multiple sclerosis

Serologic DR typing and genomic DRB1, DQA1, DQB1, DPA1, and DPB1 typing using sequence-specific oligonucleotides were performed in 69 multiple sclerosis (MS) patients and 181 healthy controls on in vitro amplified DNA. The frequencies of DR2 as well as the DR2-associated DQA1*0102 and DQB1*0602 alleles were increased whereas DR7 was decreased among MS patients. The distribution of DR4 subtypes as well as DP alleles were similar in patients and healthy controls. All but one of 23 DR4-positive MS patients carried the DQB1*0302 allele, whereas five of five DR7-positive MS patients carried the DQB1*0303 allele. Of the MS patients, 99% compared to 79% of the controls carried DQA1 alleles encoding glutamine at residue 34, while 97% of the MS patients compared to 72% of the controls carried DQB1 alleles encoding DQ beta chains sharing long polymorphic stretches. A combination of such DQA1 and DQB1 alleles was carried by 96% of the MS patients and 60% of the controls, suggesting an association between MS and a combination of particular DQA1 alleles and DQB1 alleles. The corresponding DQ alpha beta heterodimers may have in common an ability to bind a particular peptide.

The amino acid at position 57 of the HLA-DQB chain and susceptibility to develop insulin-dependent diabetes mellitus

In Caucasoids HLA-DQB1 genes encoding amino acids other than aspartic acid at position 57 of the DQ beta chain (non-Asp-57) are associated with susceptibility to develop insulin-dependent diabetes mellitus (IDDM), while resistance is associated with aspartic acid at this residue (Asp-57). Following amplification of genomic DNA by the polymerase chain reaction, the DQB1 alleles of 87 random Norwegian IDDM patients and 187 healthy controls were investigated with 11 different sequence-specific oligonucleotide probes. Of these patients 82% carried DQB1 alleles encoding non-Asp-57 at both of their DQ beta chains, compared to 27% of the controls (relative risk = 12.2, p less than 0.0001). Sixteen percent of the patients (versus 51% of the controls) were heterozygous Asp-57/non-Asp-57. Two percent of the patients (22% of the controls) were apparently Asp-57 homozygous. The results demonstrate that non-Asp-57 DQ beta chains are associated with susceptibility to develop IDDM but also indicate that the protection associated with DQ beta Asp-57 may not be as dominant as reported by others.

Gliadin specific, HLA DQ2-restricted T cells are commonly found in small intestinal biopsies from coeliac disease patients, but not from controls.

The authors have analysed gliadin specific, CD4+ T cells isolated from small intestinal biopsies of 23 adult coeliac disease patients (20 on a gluten‐free diet and three untreated) and nine control patients. The biopsies were stimulated ex vivo with a peptic/tryptic digest of gliadin for 24 h, and activated T cells were positively selected with paramagnetic beads coated with an antibody against the interleukin‐2 receptor. The T cells were expanded and tested for gliadin reactivity and HLA restriction. Gliadin specific, polyclonal T cell lines were recovered from biopsies of all 23 patients. Inhibition studies of T cell lines from 21 patients with anti‐HLA monoclonal antibodies indicated predominant presentation of the gliadin antigen by HLA‐DQ2 in T cell lines from 11 patients (lines from seven patients with complete MoAb inhibition, the remaining with incomplete inhibition) and incomplete inhibition by HLA‐DR3 in lines from three patients. Nine gliadin specific T cell clones from six patients were established; all of these were HLA‐DQ2 restricted. Gliadin specific T cells were not found in biopsies from the non‐coeliac controls. Our findings demonstrate that gliadin reactive T cells are commonly found in the intestinal mucosa of CD patients and they support the notion that the majority of T cells recognize gliadin peptide(s) when presented by the disease associated DQ2 molecules.