Nobuhiro Kamikawaji

EFFECT OF MATCHING OF CLASS I HLA ALLELES ON CLINICAL OUTCOME AFTER TRANSPLANTATION OF HEMATOPOIETIC STEM CELLS FROM AN UNRELATED DONOR

BACKGROUND The requirements with respect to HLA compatibility and the relative importance of matching for individual class I and class II HLA alleles in the transplantation of hematopoietic stem cells from unrelated donors have not yet been established. METHODS We performed retrospective DNA typing of alleles at 11 polymorphic loci of HLA genes in 440 recipients of hematopoietic stem cells from unrelated donors who were serologically identical with their respective recipients for HLA-A, B, and DR antigens. Of these recipients, 80 percent had leukemia; the rest had lymphoma, marrow failure, or a congenital disorder. RESULTS Multivariate analysis showed that incompatibility for HLA-A alleles and incompatibility for HLA-C alleles were independent risk factors for severe acute graft-versus-host disease (GVHD) (HLA-A, P=0.006; HLA-C, P=0.001). Mismatching of HLA-A, but not of HLA-C, alleles was an independent risk factor for death (P<0.001). Matching [corrected] of HLA-C alleles was a significant risk factor for relapse of leukemia (P=0.035). HLA-B disparity was a significant risk factor for both GVHD and death in the univariate analysis, but not in multivariate analysis. Disparities in class II HLA alleles of the DRB1, DQA1, DQB1, DPA1, and DPB1 loci were not identified as significant risk factors for acute GVHD or death in the multivariate analysis. CONCLUSIONS Genomic typing of class I HLA alleles adds substantially to the success of transplantation of hematopoietic stem cells from unrelated donors, even if the donors are serologically identical to their recipients with respect to HLA-A, B, and DR antigens.

Immune suppression gene on HLA-Bw54-DR4-DRw53 haplotype controls nonresponsiveness in humans to hepatitis B surface antigen via CD8+ suppressor T cells

The development of antiviral vaccines has been accelerated using monoclonal antibody and/or recombinant DNA techniques, the objective being to prevent grave viral infectious diseases, such as acquired immunodeficiency syndrome (AIDS), adult T-cell leukemia (ATL), and hepatitis B virus (HBV)-associated liver diseases. Certain proportions of individuals in the human population do not have any appreciable immune response to foreign antigens, either in cases of natural exposure or a planned immunization. Here we report that in the nonresponders to HB vaccine, there is an HLA-linked immune suppression gene for hepatitis B surface antigen (Is-HBsAg) controlling the nonresponsiveness to HBsAg through HBsAg-specific suppressor T cells. The Is-HBsAg is in strong linkage disequilibrium with the HLA-Bw54-DR4-DRw53 haplotype.

Analysis of HLA alleles in Japanese patients with cirrhosis due to chronic hepatitis C

Hepatitis C virus (HCV) leads to chronic liver disease in at least 50–60% of infected people and approximately 40–50% of these patients will go on to develop cirrhosis due to chronic hepatitis C (HCV‐C). The pathogenic mechanisms that result in HCV‐C are unknown. Sixty Japanese patients with HCV‐C were examined for HLA‐A, B, C and DR alleles by serologic typing and for HLA‐DQB1 alleles by DNA typing using the polymerase chain reaction‐restriction fragment length polymorphism (PCR‐RFLP) method. As the control population, 293 healthy un‐related Japanese were used. The frequencies of HLA‐B61, Cw3, DR4, DQB1*0401 and DQB1*0402 were increased, while those of HLA‐DR9, DQB1*0301 and DQB1*0303 were decreased in the patients. The co‐ordinate increase in the frequency of HLA‐DR4, DQB1*0401 or 0402 and decrease in the frequency of DR9 or DQB1*0303 were suggestive of a strong linkage disequilibrium between HLA‐DR4 and DQB1*0401 or 0402 and between HLA‐DR9 and DQB1*0303, respectively. From the odds ratio (OR) analysis, the combinations of HLA‐Cw3+DR4‐DQB1*0401 or 0402, or HLA‐B61+DR4‐DQB1*0401 or 0402 increased the risk for developing HCV‐C when compared to each HLA allele alone. This suggested an additive effect for these classes I and II HLA allele combinations in HCV‐C. In contrast, HLA‐DR9‐DQB1*0303 and DQB1*0301 may confer resistance to this disease. These results suggest the existence of HLA‐linked susceptibility genes to HCV‐C.

DNA typing of HLA in the patients with moyamoya disease.

SummaryMoyamoya disease is a clinical entity demonstrating a chronic occlusion of the cerebrovascular system. Although some possible etiological factors have been postulated, the etiology of this disease is still unknown. So far, some investigations have suggested the association between moyamoya disease and HLA in the serological typing. However, DNA typing of HLA have not been performed yet. Thus, we performed DNA-typing of HLA in the unrelated Japanese patients with definite moyamoya disease, using the polymerase chain reaction-sequence specific oligonucleotide probe (PCR-SSOP) technique. In the total patients, DQB1*0502 had a positive association with the disease. On the other hand,DRB1*0405 and DQB1*0401 showed a negative association. In comparing the early-onset and late-onset groups, two groups did not share the same disease associated alleles at all. Thus, the etiology of moyamoya disease seem to have a genetic background. Furthermore, different genetic factors might also be involved in the difference between the early-onset and late-onset groups.

Use of BONSAI decision trees for the identification of potential MHC Class I peptide epitope motifs.

Recognition of short peptides of 8 to 10 mer bound to MHC class I molecules by cytotoxic T lymphocytes forms the basis of cellular immunity. While the sequence motifs necessary for binding of intracellular peptides to MHC have been well studied, little is known about sequence motifs that may cause preferential affinity to the T cell receptor and/or preferential recognition and response by T cells. Here we demonstrate that computational learning systems can be useful to elucidate sequence motifs that affect T cell activation. Knowledge of T cell activation motifs could be useful for targeted vaccine design or immunotherapy. With the BONSAI computational learning algorithm, using a database of previously reported MHC bound peptides that had positive or negative T cell responses, we were able to identify sequence motif rules that explain 70% of positive T cell responses and 84% of negative T cell responses.

Genetic control of immune response and disease susceptibility by the HLA-DQ gene.

The particular alleles of the HLA-DQ locus may control the low immune response to natural antigens by a dominant genetic trait through the immune suppression mediated by CD8+ suppressor T cells. The suppressor T cells may be activated by DQ-restricted and antigen-specific CD4+ suppressor/inducer T cells, because (1) a statistically significant association and linkage between low immune responsiveness to the natural antigens and the HLA-DQ gene were observed; (2) antigen-specific CD4+ T cells restricted by the DQ molecules encoded for by the HLA-DQ allele associated with low responsiveness were evidenced in many low responders; and (3) anti-HLA-DQ mAb restored the immune response to natural antigens, in some low responders. This HLA-DQ-controlled polymorphism of immune response to the natural antigens may account for the association between HLA-DQ alleles and organ-specific autoimmune diseases.

An HLA-binding-motif-aided peptide epitope library: A novel library design for the screening of HLA-DR4-restricted antigenic peptides recognized by CD4 + T cells

AbstractSusceptibility to a series of autoimmune diseases is strongly associated with particular HLA class II alleles. Identification of T cell clones and antigenic epitopes bound by HLA class II molecules involved in autoimmune diseases is critical to understanding the etiology of these HLA class II-associated diseases. However, establishment of T cell clones in autoimmune diseases is difficult because the antigenic peptides are unknown. Peptide library methods which include all possible peptide sequences offer a potentially powerful tool for the detection of cross-reactive antigenic peptides recognized by T cells. Here, we reduced the number of peptides per mixture by utilizing the known binding motifs of peptides for the HLA-DRB1*0405 molecule and evaluated the effectiveness of this library design. Each library mixture evoked a strong proliferative response in the unprimed peripheral blood lymphocytes (PBL) from HLA-DRB1*0405-positive donors but little or no response in the PBL from HLA-DRB1*0405-negative donors. The library also detected antigenic peptides that activated three antigen-specific T cell lines restricted by HLA-DRB1*0405, with different specificities. The motif-based approach thus presents a powerful method for monitoring T cells in large, heterogeneous T cell populations and is useful for the identification of the mimic peptide epitopes of T cell lines and clones.

Identification of an epitope for T cells correlated with antibody response to hepatitis B surface antigen in vaccinated humans

Two antigenic T-cell epitopes of HBsAg, designated HBs 16-31 and HBs 81-99, were identified using synthetic peptides and HBsAg-specific T-cell lines. HBs 16-31 was recognized by five HBsAg-specific T-cell lines from vaccinees with both high and low antibody titers, whereas HBs 81-99 was recognized by two T-cell lines derived from vaccinees with high antibody titers. The antibody titer against HBsAg was correlated significantly with the proliferation of vaccinees PBLs in response to HBs 81-99 (r = 0.47) but not to HBs 16-31, suggesting that HBs 81-99 plays a critical role in anti-HBs antibody production in humans vaccinated with HBsAg.

Allele-specific expression of the cytoplasmic exon of HLA-DQB1 gene

The β chain of the HLA-DQ molecule is shorter by eight amino acid residues than other major histocompatibility complex class II β chains due to elimination of the fifth exon coding for part of the cytoplasmic domain. This elimination is caused by one base substitution in the splice accepter site of the exon. We found that two HLA-DQB1 alleles, DQB1*0503 and DQB1*0601, did not have this substitution, and the exon was utilized in these two alleles. However, two forms of HLA-DQB mRNA, with or without exon 5, were generated in Epstein-Barr virus-transformed cell lines homozygous for DQB1*0503 or DQB1*0601, indicating alternative mRNA splicing. The alternative splicing of DQB1*0601 mRNA was also found in peripheral blood lymphocytes and L cell transfectants. To investigate the functional relevance of the allele-specific long cytoplasmic tail of HLA-DQ β chain, we developed three types of L cell transfectants expressing exclusively the HLA-DQw6 molecules with short cytoplasmic tail, long cytoplasmic tail, or both forms of the β chain, and used them as antigen presenting cells for streptococcal cell wall antigen-specific T cell lines. These three types of transfectants could function almost equally well as antigen presenting cells. It was thus demonstrated that both forms of HLA-DQ β chain, with or without eight amino acid residues coded for by the exon 5, can be associated with the HLA-DQ α chain, be expressed on the cell surface, and function as restriction molecules in antigen recognition by the CD4+ T cells.

Persistent elevation of immunoglobulin G titer against the C region of recombinant group A streptococcal M protein in patients with rheumatic fever.

To analyze the immune response to the C region of group A streptococcal M protein in patients with rheumatic fever (RF), we cloned the structural gene for the C region of type 12 M protein and produced recombinant C region of M protein. IgG titers against the C region of M protein were measured by ELISA from sera of patients with RF (n = 10), uncomplicated streptococcal pharyngitis (n = 26), and age-matched controls (n = 49). IgG titers against the C region were significantly higher in patients with RF than in controls or patients with uncomplicated streptococcal pharyngitis (43versus 1.5 or 1.8 μg/mL, p < 0.01). Studies using overlapping synthetic peptides of the C region demonstrated that increased IgG reactivity was observed against the amino-terminal halves of the C repeat blocks (C1, C2) in RF, indicating that these domains are the main immunodominant epitopes in the C region.