Specific inhibition of TH2-type cytokine production from human peripheral T cells by terfenadine in vitro

BACKGROUND

Cytokine imbalance is thought to be one of the causes for allergic diseases. The effect of anti-allergic drugs on cytokine production from T cells should be examined in a convenient way.

OBJECTIVES

To study the in vitro effect of terfenadine, a prototype non-sedating H1 receptor antagonist, on cytokine production from activated T cells.

METHODS

T cells were cultured in the presence of terfenadine on anti-CD3 mAb and anti-CD26 mAb-coated wells, anti-CD3 mAb and anti-CD28 mAb-coated wells, and anti-CD3 mAb wells with PMA. T-cell proliferation, along with the concentrations of interleukin (IL) -2, interferon (IFN) -gamma, IL-4, and IL-5 were measured.

RESULTS

Terfenadine inhibited T-cell proliferation and IL-4 and IL-5 production under each costimulatory condition tested, whereas it had no effect on IL-2 and IFN-gamma production.

CONCLUSIONS

These results indicate that terfenadine has a specific inhibitory effect on TH2-type cytokine production induced by several ways of costimulatory activation.

Functional interaction of DFF35 and DFF45 with caspase-activated DNA fragmentation nuclease DFF40

DNA fragmentation factor (DFF) functions downstream of caspase-3 and directly triggers DNA fragmentation during apoptosis. Here we described the identification and characterization of DFF35, an isoform of DFF45 comprised of 268 amino acids. Functional assays have shown that only DFF45, not DFF35, can assist in the synthesis of highly active DFF40. Using the deletion mutants, we mapped the function domains of DFF35/45 and demonstrated that the intact structure/conformation of DFF45 is essential for it to function as a chaperone and assist in the synthesis of active DFF40. Whereas the amino acid residues 101-180 of DFF35/45 mediate its binding to DFF40, the amino acid residues 23-100, which is homologous between DFF35/45 and DFF40, may function to inhibit the activity of DFF40. In contrast to DFF45, DFF35 cannot work as a chaperone, but it can bind to DFF40 more strongly than DFF45 and can inhibit its nuclease activity. These findings suggest that DFF35 may function in vivo as an important alternative mechanism to inhibit the activity of DFF40 and further, that the inhibitory effects of both DFF35 and DFF45 on DFF40 can put the death machinery under strict control.

Correlation of the epitopes defined by anti-CD26 mAbs and CD26 function

To clarify the different anti-CD26 mAbs corresponding different functions of CD26, the correlation of the epitopes defined by anti-CD26 mAbs and the functions of CD26 have been studied. Using truncated, human-rat CD26 swap mutants and cross-blocking studies, 13 anti-CD26 mAbs were divided into 5 separate groups. These 5 epitopes were localized between the 1-247th, 248-358th, 359-449th (closer to the 359th amino acid), 450-577th and 359 653th amino acid regions. MAbs against two of these five epitopes, the 248-358th and 359-449th amino acid regions, were associated with inducing modulation of CD26 and T-cell costimulation through the CD3 pathway. Furthermore, mAbs against one of these epitopes, the 359-449th amino acid region, appeared to encompass the ADA binding domain. Analysing the avidity of each mAb to the CD26 molecule using DPPIV enzymatic activity as an indicator, we found that the function of CD26 had little correlation with the avidity of anti-CD26 mAbs, suggesting that distinct epitopes defined by anti-CD26 mAbs appeared to be associated with different functions of CD26. These results will be very useful in the further definition of the functional domains of CD26.

Determination of adenosine deaminase binding domain on CD26 and its immunoregulatory effect on T cell activation

CD26, a 110-kDa cell surface glycoprotein, exhibits dipeptidyl peptidase IV enzyme activity and plays an important role in T cell costimulation. In the present study, we examined both the exact adenosine deaminase (ADA) binding domain on CD26 and the functional consequences of mutated CD26 transfectants that were deficient for cell surface ADA. Using CD26 deletion, human-rat swap, and point mutations, we found that the residues of L340, V341, A342, and R343 on the CD26 molecule were essential amino acids for ADA binding. When these amino acids were mutated and transfected into Jurkat cells, the resultant CD26 transfectants expressed only CD26, not ADA, on the cell surface. The amount of IL-2 produced by wild-type and mutated CD26 transfectants was almost the same following stimulation with anti-CD3 plus PMA. However, the mutated CD26 transfectants were much more sensitive to the inhibitory effect of adenosine on IL-2 production than were the wild CD26 transfectants. These data suggest that ADA on the cell surface does not directly involve T cell activation. Conversely, CD26 alone does not result in modulating the inhibitory effect of adenosine. Only the ADA bound to CD26 on the cell surface was functional and could counteract the inhibitory effect of elevated extracellular adenosine.

Determination of adenosine deaminase binding domain on CD26 and its immunoregulatory effect on T cell activation

CD26, a 110-kDa cell surface glycoprotein, exhibits dipeptidyl peptidase IV enzyme activity and plays an important role in T cell costimulation. In the present study, we examined both the exact adenosine deaminase (ADA) binding domain on CD26 and the functional consequences of mutated CD26 transfectants that were deficient for cell surface ADA. Using CD26 deletion, human-rat swap, and point mutations, we found that the residues of L340, V341, A342, and R343 on the CD26 molecule were essential amino acids for ADA binding. When these amino acids were mutated and transfected into Jurkat cells, the resultant CD26 transfectants expressed only CD26, not ADA, on the cell surface. The amount of IL-2 produced by wild-type and mutated CD26 transfectants was almost the same following stimulation with anti-CD3 plus PMA. However, the mutated CD26 transfectants were much more sensitive to the inhibitory effect of adenosine on IL-2 production than were the wild CD26 transfectants. These data suggest that ADA on the cell surface does not directly involve T cell activation. Conversely, CD26 alone does not result in modulating the inhibitory effect of adenosine. Only the ADA bound to CD26 on the cell surface was functional and could counteract the inhibitory effect of elevated extracellular adenosine.

Different regulatory effects of pentoxifylline on human T cell activation pathways

Pentoxifylline (PTX), a methylxanthine derivative, was examined for its effects on T cell proliferation and cytokine production stimulated by cross-linking anti-CD3 alone, anti-CD3 with PMA, anti-CD3 with anti-CD26, or anti-CD3 with anti-CD28 mAb, respectively. PTX at a 3.5 x 10(-5) M concentration significantly inhibited T cell proliferation and the production of tumor necrosis factor-alpha, interleukin-2, and interleukin-4. Moreover, this effect was selective for stimulation by cross-linking anti-CD3 with PMA, or anti-CD3 with anti-CD26, but not by cross-linking anti-CD3 with anti-CD28. These results suggest that the inhibitory effect of PTX on T cell activation involves the CD3 and CD26, but not the CD28 signal pathway.

Cross-linking of CD26 by antibody induces tyrosine phosphorylation and activation of mitogen-activated protein kinase

CD26, a T-cell activation antigen that has dipeptidyl peptidase IV activity in its extracellular domain and has also been shown to play an important role in T-cell activation. The earliest biochemical events seen in stimulated T lymphocytes activated through the engagement of the T-cell receptor (TCR) is the tyrosine phosphorylation of a panel of cellular proteins. In this study we demonstrate that antibody-induced cross-linking of CD26-in CD26-transfected Jurkat cells induced tyrosine phosphorylation of several intracellular proteins with a similar pattern to that seen after TCR/CD3 stimulation. Herbimycin A, an inhibitor of the src family protein tyrosine kinases dramatically inhibited this CD26-mediated effect on tyrosine phosphorylation. Major tyrosine phosphorylated proteins were identified by immunoblotting, and included p56lck, p59fyn, zeta associated protein-tyrosine kinase of 70,000 MW (ZAP-70), mitogen-activated protein (MAP) kinase, c-Cb1, and phospholipase C gamma. CD26-induced tyrosine phosphorylation of MAP kinase correlated with increased MAP kinase activity. In addition, CD26 was costimulatory to CD3 signal transduction since co-cross-linking of CD26 and CD3 antigens induced prolonged and increased tyrosine phosphorylation in comparison with CD3 activation alone. We therefore conclude that CD26 is a true costimulatory entity that can up-regulate the signal transducing properties of the TCR.

Role of CD26 for CD4 memory T cell function and activation

CD26 is a 110 kDa T cell activation antigen and has been shown to have DPPIV enzyme activity which cleaves amino-terminal dipeptides with either L-proline or L-alanine at the penultimate position. Recent studies showed that CD26 plays an integral role in T cell activation. A partial explanation of the mechanism of CD26 mediated T cell signaling appears to be its association with CD45 tyrosine phosphatase, which may be importance in T cell activation and signal transduction. In addition, we showed that CD26 is a receptor for adenosine deaminase (ADA). Moreover, ADA on the cell surface is involved in an important immunoregulatory mechanism by which released ADA binds to cell surface CD26 and this complex is capable of reducing the local concentration of adenosine. Thus, CD26 is a multifunctional molecule controlling many key aspects of lymphocyte function.

Characterization of adenosine deaminase binding to human CD26 on T cells and its biologic role in immune response

CD26, a T cell activation Ag, also known as dipeptidyl peptidase IV, is directly associated with adenosine deaminase (ADA) on the surface of T cells and T cell lines. In the present study, we examined both the binding of ADA and CD26 and the functional consequences of this interaction. We found that ADA was associated with CD26 on T cell lines lacking either ADA or dipeptidyl peptidase IV enzymatic activity, indicating that the association between dipeptidyl peptidase IV and ADA did not require enzymatic activity. Moreover, using immunoelectron microscopy, we demonstrated that CD26 and ADA co-localized on the cell surface, but not inside cells, suggesting that CD26 did not transport ADA to the surface. In keeping with this observation, we showed that human CD26-transfected murine pre-B cell lines lacking human ADA acquired ADA from an extracellular source. More importantly, adenosine in the absence of cell surface ADA inhibited T cell proliferation and IL-2 production induced by various stimuli. On the other hand, cells expressing ADA and CD26 on the surface were much more resistant to the inhibitory effect of adenosine. These data suggest that ADA on the cell surface is involved in an important immunoregulatory mechanism by which released ADA binds to cell surface CD26, and this complex is capable of reducing the local concentration of adenosine.

Characterization of adenosine deaminase binding to human CD26 on T cells and its biologic role in immune response

CD26, a T cell activation Ag, also known as dipeptidyl peptidase IV, is directly associated with adenosine deaminase (ADA) on the surface of T cells and T cell lines. In the present study, we examined both the binding of ADA and CD26 and the functional consequences of this interaction. We found that ADA was associated with CD26 on T cell lines lacking either ADA or dipeptidyl peptidase IV enzymatic activity, indicating that the association between dipeptidyl peptidase IV and ADA did not require enzymatic activity. Moreover, using immunoelectron microscopy, we demonstrated that CD26 and ADA co-localized on the cell surface, but not inside cells, suggesting that CD26 did not transport ADA to the surface. In keeping with this observation, we showed that human CD26-transfected murine pre-B cell lines lacking human ADA acquired ADA from an extracellular source. More importantly, adenosine in the absence of cell surface ADA inhibited T cell proliferation and IL-2 production induced by various stimuli. On the other hand, cells expressing ADA and CD26 on the surface were much more resistant to the inhibitory effect of adenosine. These data suggest that ADA on the cell surface is involved in an important immunoregulatory mechanism by which released ADA binds to cell surface CD26, and this complex is capable of reducing the local concentration of adenosine.

Characterization of T cell epitopes restricted by HLA-DP9 in streptococcal M12 protein

Interaction of the HLA-DP9 (DPA1*0201/DPB1*0901) molecule and M protein of serotype 12 (SS95/12) streptococci, a main component of the streptococcal cell wall Ag, has been investigated to decipher peptide-binding capacity and T cell activation in the context of the HLA-DP molecule. Seven antigenic peptides (amino acids 19-25) restricted by the HLA-DP9 molecule were identified in M12 protein, using M12 protein- or peptide-specific T cell lines from naturally exposed individuals. The binding affinity of each peptide to the HLA-DP9 molecule was measured by fluorescence intensity of biotinylated peptides bound to L cell transfectants expressing HLA-DP9, followed by treatment with avidin-fluorescence. Binding of biotinylated peptides to the HLA-DP9 molecule was inhibited by an excess amount of corresponding nonbiotinylated peptides and other nonbiotinylated peptides, indicating that the peptides were bound to the HLA-DP9 molecule at a single binding site. Seven synthetic peptides containing the T cell epitopes restricted by the HLA-DP9 molecule had high binding affinity to the HLA-DP9 molecule. Comparison of the amino acid sequences of truncated analogues that could bind to the HLA-DP9 molecule and/or activate T cells suggested an HLA-DP9-specific binding motif, composed of a positively charged residue (R or K) at position 1, a hydrophobic residue (A, G, or L) at position 6, and another hydrophobic residue (L or V) at position 9. Analysis of single amino acid-substituted analogues suggested that the positively charged amino acid in the motif served as a key anchor residue for binding to the HLA-DP9 molecule, which differs from the binding motif to the HLA-DR molecules.

Characterization of T cell epitopes restricted by HLA-DP9 in streptococcal M12 protein

Interaction of the HLA-DP9 (DPA1*0201/DPB1*0901) molecule and M protein of serotype 12 (SS95/12) streptococci, a main component of the streptococcal cell wall Ag, has been investigated to decipher peptide-binding capacity and T cell activation in the context of the HLA-DP molecule. Seven antigenic peptides (amino acids 19-25) restricted by the HLA-DP9 molecule were identified in M12 protein, using M12 protein- or peptide-specific T cell lines from naturally exposed individuals. The binding affinity of each peptide to the HLA-DP9 molecule was measured by fluorescence intensity of biotinylated peptides bound to L cell transfectants expressing HLA-DP9, followed by treatment with avidin-fluorescence. Binding of biotinylated peptides to the HLA-DP9 molecule was inhibited by an excess amount of corresponding nonbiotinylated peptides and other nonbiotinylated peptides, indicating that the peptides were bound to the HLA-DP9 molecule at a single binding site. Seven synthetic peptides containing the T cell epitopes restricted by the HLA-DP9 molecule had high binding affinity to the HLA-DP9 molecule. Comparison of the amino acid sequences of truncated analogues that could bind to the HLA-DP9 molecule and/or activate T cells suggested an HLA-DP9-specific binding motif, composed of a positively charged residue (R or K) at position 1, a hydrophobic residue (A, G, or L) at position 6, and another hydrophobic residue (L or V) at position 9. Analysis of single amino acid-substituted analogues suggested that the positively charged amino acid in the motif served as a key anchor residue for binding to the HLA-DP9 molecule, which differs from the binding motif to the HLA-DR molecules.

HLA-A and -DRB4 genes in controlling the susceptibility to Hashimoto's thyroiditis

HLA-linked genetic factors involved in the pathogenesis of HT were studied in 71 patients with HT by serologic typing for HLA-A, -B, -C, -DR, and -DQ specificities and by DNA typing for HLA-DRB1, -DRB3, -DRB4, -DRB5, -DQA1, -DQB1, AND -DPB1 genes using the PCR-SSOP method. Typing results demonstrated significant positive associations of HT with HLA-A2 and -DRB4*0101 (DR53) (p < 0.01, RR = 2.03, EF = 0.61 and p < 0.0001, RR = 4.48, EF = 0.69, respectively). Although HLA-DR8, -DRB1*0403, -DQA*03, and -DQB1*0303 were statistically more prevalent in the patient group than in the controls, these associations were presumably due to the strong linkage disequilibria of these alleles with HLA-A2 or -DRB4*0101 in the Japanese population. Ninety-seven percent of the patients (69 out of 71) were positive for HLA-A2 or -DRB4*0101 compared to 79% in controls (RR = 8.7, p < 0.0005). The combination of HLA-A2 and -DRB4*0101 showed higher OR of risk for HT (OR = 12.8) than HLA-A2 (OR = 7.3) or DRB4*0101 (OR = 7.5) alone. These observations suggest that at least two loci, HLA-A and HLA-DRB4 together, may control the susceptibility to HT. On the other hand, the frequency of DQA1*0102 was significantly decreased in the patient group, suggesting that DQA1*0102 might confer resistance to HT.

Resistance to autoimmune thyroid disease is associated with HLA-DQ

Genotypes of DQA1 and DQB1 genes were determined by polymerase chain reaction followed by dot blot hybridization with sequence-specific oligonucleotide probes in 105 patients with goitrous autoimmune thyroiditis (Hashimoto's thyroiditis) and in 67 patients with Graves' disease to investigate whether specific DQ alleles were associated with susceptibility or resistance to autoimmune thyroid diseases. Hashimoto's thyroiditis was found to be negatively associated with DQA1*0102 and DQB1*0602 whereas Graves' disease showed a negative association with DQB1*0501. No strongly positive association with a specific DQ allele was found in either disease. These results suggest that the HLA-DQ gene may be a genetic marker for resistance to autoimmune thyroid diseases.

HLA antigens in Japanese patients with systemic lupus erythematosus

To determine the association of HLA antigens with SLE and the clinical findings of the disease, HLA antigens were tested in 58 Japanese patients with SLE, who fulfilled the ARA diagnostic criteria, along with 97 normal controls. HLA class I and II antigens were typed serologically using the antisera provided by the 11th HLA Workshop. Among the HLA class II antigens, further DRB, DQ and DP alleles were defined by DNA typing using the PCR/SSOP method. There were significantly more SLE patients with HLA-B39, DRB1*1501, DRB5*0101 and DQB1*0602 than normal controls. This result suggested that the haplotype of HLA-DRB1*1501-DRB5*0101-DQA1*0102-DQB1*0602 consists of the SLE-associated MHC markers in Japan. There were some positive and negative associations between the HLA antigens and clinical or serological findings in SLE. There is a possibility that some HLA alleles might be related to the clinical and/or serological subsets of SLE.

HLA-linked susceptibility and resistance to Takayasu arteritis

To investigate genetic factors involved in the pathogenesis of Takayasu arteritis, patients in the Japanese population were examined for HLA-A, -B, and -C alleles by serological typing and for HLA-DR, DQ, and DP alleles by DNA typing using polymerase chain reaction (PCR)/sequence-specific oligonucleotide probe (SSOP) analysis. The frequencies of HLA-Bw52, DRB1*1502, DRB5*0102, DQA1*0103, DQB1*0601, and DPB1*0901 alleles were significantly increased and the frequencies of HLA-Bw54, DRB1*0405, DRB4*0101, DQA1*0301, and DQB1*0401 alleles were significantly decreased. Strong linkage disequilibria among the increased alleles and among the decreased alleles were evident in the Japanese population. Therefore, the haplotype of HLA-B252-DRB1*1502-DRB5*0102-DQA1*0103-DQB1++ +*0601-DPA1*02-DPB1*0901 may confer susceptibility to Takayasu arteritis while another haplotype of HLA-Bw54-DRB1*0405-DRB4*0101-DQA1*0301-DQB1++ +*0401 may confer resistance to the disease. These observations clearly indicate that HLA-linked gene(s) are involved in the development of Takayasu arteritis.

Immunogenetic analysis of silicosis in Japan

We previously reported that a gene in linkage disequilibrium with HLA-Bw54, DR4, and DRw53 might control the susceptibility to silicosis (K. Honda et al. 1988. N. Engl. J. Med. 319:1610). To further define the HLA-linked gene and other genetic factors for predisposition of silicosis, we determined for HLA-DQ and DP alleles using the polymerase chain reaction and sequence-specific oligonucleotide probes and made a restriction fragment length polymorphism (RFLP) analysis of the fourth component of complement (C4) genes, immunoglobulin lambda variable chain (IGLV) gene, and T-cell receptor alpha and beta genes in 46 Japanese patients with silicosis. The frequency of DQB1*0401 (relative risk [RR] = 2.2, P < 0.02) was increased and that of DQB1*0601 (RR = 0.36, P < 0.01) was decreased in the patients. RFLP analysis of C4 and IGLV genes showed significant association between silicosis and a specific RFLP pattern of C4A3-C4B5 allotype (RR = 2.3, P < 0.05) and that of IGLV 5.3 kb (RR = 0.33, P < 0.003). No other genetic markers showed significant association. Statistical analyses of the associated genetic markers revealed that the HLA-Bw54 was the allele that showed primary association with silicosis and the frequencies of the C4 and HLA-DQ alleles were suggested to be increased due to their linkage disequilibrium with the HLA-Bw54. We conclude that the major gene for silicosis may be mapped near the HLA-B locus.

Difference in HLA-linked genetic background between mixed connective tissue disease and systemic lupus erythematosus

We have typed 64 Japanese patients with mixed connective tissue disease (MCTD) and 53 Japanese patients with systemic lupus erythematosus (SLE) for HLA-DRB1, DRB3, DRB4, DRB5, DQA1, DQB1, and DPB1 genes by the HLA-DNA typing method using the PCR-SSOP technique. Frequencies of HLA-DRB1*0401, DRB1*0901, DRB4*0101, and DQA1*03 were increased and those of HLA-DRB1*0405 and DQB1*0401 were decreased in the patients with MCTD, while the frequencies of HLA-DRB1*1501, DRB5*0101, and DQB1*0602 were increased in the patients with SLE. The typing results suggest that susceptibility to MCTD is strongly associated with the HLA-DRB1*0401-DRB4*0101-DQA1*03-DQB1*0301 haplotype, and that to SLE is associated with the HLA-DRB1*1501-DRB5*0101-DQA1*0102-DQB1*0602 haplotype. The observation that the MCTD-associated HLA alleles are distinct from the SLE-associated ones may support the clinical entity of MCTD different from SLE.

HLA-A and DPB1 loci confer susceptibility to Grave's disease

To investigate HLA-linked genetic factors involved in the pathogenesis of Graves' disease, 76 patients and 317 healthy controls in the Japanese population were examined for HLA-A, B, C, DR, and DQ specificities by serologic typing and for HLA-DPB1 alleles by DNA typing by using the PCR-SSOP method. The frequencies of HLA-A2, B46, Cw11, and DPB1*0501 were increased and those of HLA-A24, B7, Bw52, and DR1 were decreased in the patients. The increased frequencies of HLA-A2 and DPB1*0501 in the patients were statistically significant when the corrected p value (pc) was applied (pc < 0.02 and pc < 0.002, respectively). ORs for a risk to develop the disease were calculated among individuals positive for DPB1*0501 and/or HLA-A2, and the highest OR (10.5) was observed in individuals possessed both DPB1*0501 and HLA-A2. This observation suggests a synergic involvement of a HLA class II allele (DPB1*0501) and an HLA class I allele (HLA-A2) in the pathogenesis of Graves' disease.

Sequencing and population analysis of four novel HLA-DPA1 alleles

We determined the base sequences of the HLA-DPA1 gene from four B-lymphoblastoid cell lines (CB6B, LKT3, AMAI, and T7526) that showed distinct electrophoretic patterns of single-stranded polymerase chain reaction products of the HLA-DPA1 gene. The novel HLA-DPA1 alleles of CB6B, LKT3, AMAI, and T7526 were designated DPA1*02021, DPA1*02022, DPA1*0301, and DPA1*0401, respectively. Although there was only one base substitution between DPA1*02021 and DPA1*02022, the single-strand conformation polymorphism of these two alleles was clearly demonstrated by electrophoresis in a nondenaturing polyacrylamide gel containing 10% glycerol. In addition, we genotyped for the HLA-DPA1 gene of healthy unrelated Oriental individuals--i.e., 227 Japanese, 88 Papua New Guineans, and 41 Buyi-Chinese--to demonstrate the ethnic distribution of the HLA-DPA1 alleles.

DNA typing of HLA class II genes in B-lymphoblastoid cell lines homozygous for HLA

The HLA class II genotypes were determined in the B-lymphoblastoid cell lines selected for the Tenth International Histocompatibility Workshop. The HLA class II genes were determined by the PCR-SSOP method using the reagents provided by the Eleventh Histocompatibility Workshop. Additional studies have been performed for further characterization of HLA class II polymorphism on these cell lines. It is observed that several cell lines have HLA class II haplotypes with the same DRB1, DQA1 and DQB1 alleles on both haplotypes but different alleles at the other class II loci, confirming that these cell lines are not truly HLA class II-homozygous. Other cell lines carried HLA class II haplotypes which were only different at the DRB1 gene. These results suggest double recombination events or gene conversion-like events in generation of HLA DR, DQ haplotypes. These cell lines provide an important tool as references for HLA DNA typing.

HLA-DP antigen and Takayasu arteritis

Sixty-four patients with Takayasu arteritis and 317 healthy individuals in the Japanese population were examined for HLA-A, -B and -C alleles by serological typing and for HLA-DR, DQ and DP alleles by DNA typing using PCR/SSOP analysis. The frequencies of HLA-Bw52, DRB1*1502, DRB5*0102, DQA1*0103, DQB1*0601 and DPB1*0901 alleles were significantly increased and the frequencies of HLA-Bw54, DRB1*0405, DRB4*0101, DQA1*0301, DQB1*0401 alleles were significantly decreased. Strong linkage disequilibria among the increased alleles and among the decreased alleles were evident in the Japanese population. Therefore, the combination or haplotype of HLA-Bw52-DRB1*1502-DRB5*0102-DQA1*0103-DQB1*0601 -DPA1*02-DPB1*0901 may confer susceptibility to Takayasu arteritis while another combination or haplotype of HLA-Bw54-DRB1*0405-DRB4*0101-DQA1*0301-DQB1++ +*0401 may confer resistance to the disease. Because this is the first evidence for the association between an HLA-DP allele and Takayasu arteritis, we examined the nucleotide sequences of the DPB1*0901 allele from a patient and her healthy relatives and found no difference. The disease is therefore not caused by a mutated DPB1 gene.