Search

A genetic polymorphism of the newly discovered interferon-λ 4 (IFNL4) gene was associated with hepatitis C virus (HCV) clearance in individuals of African ancestry. To assess whether a dinucleotide variant of IFNL4 (ss469415590) also affected treatment outcome of antiviral therapy in Japan, we genotyped 213 patients with chronic genotype 1 HCV infection and 176 healthy subjects. The ΔG allele was associated with treatment failure [odds ratio (OR) 4.73, P = 0.019], as was the IFL3 rs8099917 single nucleotide polymorphism (SNP) (OR 5.06, P = 0.068). The correlation between ss469415590 and rs8099917 was high (r(2) = 0.92, D' = 0.98). Multivariate analysis revealed that the rs8099917 SNP was independently associated with treatment failure (OR 5.28, P = 0.009). Therefore, ss469415590 may be another predictive marker of antiviral therapy outcome in the Japanese population.

Super high-resolution single molecule sequence-based typing (SS-SBT) is a human leukocyte antigen (HLA) DNA typing method to the field 4 level of allelic resolution (formerly known as eight-digit typing) to efficiently detect new and null alleles without phase ambiguity by combination of long ranged polymerase chain reaction (PCR) amplification and next-generation sequencing (NGS) technologies. We previously reported the development and application of the SS-SBT method for the eight classical HLA loci, A, B, C, DRB1, DQA1, DQB1, DPA1 and DPB1. In this article, we describe the development of the SS-SBT method for three DRB1 linked loci, DRB3, DRB4 and DRB5 (DRB3/4/5) and characterization of DRB1-DRB3/4/5 haplotype structures to the field 4 level. Locus specific PCR primers for DRB3/4/5 were designed to amplify the gene regions from intron 1 to exon 6 [3' untranslated region (3'UTR)]. In total 20 DRB1 and 13 DRB3/4/5 allele sequences were determined by the SS-SBT to the field 4 level without phase ambiguity using 19 DR51, DR52 and DR53 positive genomic DNA samples obtained from Japanese. Moreover, 18 DRB1-DRB3/4/5 haplotypes were estimated to the field 4 level by the SS-SBT method in contrast to 10 haplotypes estimated by conventional methods to the field 1 level (formerly known as two digit typing). Therefore, DRB1-DRB3/4/5 haplotyping by SS-SBT is expected to provide informative data for improved HLA matching in medical research, transplantation procedures, HLA-related disease studies and human population diversity studies.

Current human leukocyte antigen (HLA) DNA typing methods such as the sequence-based typing (SBT) and sequence-specific oligonucleotide (SSO) methods generally yield ambiguous typing results because of oligonucleotide probe design limitations or phase ambiguity for HLA allele assignment. Here we describe the development and application of the super high-resolution single-molecule sequence-based typing (SS-SBT) of HLA loci at the 8-digit level using next generation sequencing (NGS). NGS which can determine an HLA allele sequence derived from a single DNA molecule is expected to solve the phase ambiguity problem. Eight classical HLA loci-specific polymerase chain reaction (PCR) primers were designed to amplify the entire gene sequences from the enhancer-promoter region to the 3' untranslated region. Phase ambiguities of HLA-A, -B, -C, -DRB1 and -DQB1 were completely resolved and unequivocally assigned without ambiguity to single HLA alleles. Therefore, the SS-SBT method described here is a superior and effective HLA DNA typing method to efficiently detect new HLA alleles and null alleles without ambiguity.

Behçet's disease (BD) is a chronic inflammatory disease characterized by oral aphthous ulcers, genital ulcers, uveitis and skin lesions. Etiology and pathogenesis of BD are not fully elucidated, but the association with human leukocyte antigen (HLA)-B51 or B*5101 has been repeatedly reported. Previous studies have shown that there are few sequence variations in the protein-coding region of B51, while there is a report on many variations in the 5'-flanking region and intron. In this study, HLA-B*5101 gene from 37 individuals including Japanese, Turkish, Jordanian and Iranian patients and healthy controls were fully sequenced to further clarify the B*5101 gene in association with BD. We found that all the patients and healthy controls carried B*510101 with no variation in the 5'-flanking region, exon and intron. However, seven polymorphisms were found in the 3'-flanking region. These polymorphisms composed of six haplotypes that were shared and stretched over the ethnic groups, suggesting that the susceptibility to BD was conferred by the B*510101 itself and not by any genes in linkage disequilibrium with B*510101. In addition, phylogenetic analyses of B*510101 showed that the 3'-flanking sequences followed an evolutional divergence differently from that of the other regions, implying that a unifying selection might operate to conserve B*510101.

The definitive version is available at www.blackwell-synergy.com., Article, TISSUE ANTIGENS. 70(1-5) 423-426 (2007)

Unusual HLA-A'null' alleles because of an entire gene deletion were found in three apparently unrelated Japanese families with leukemia patients. Inclusion of the entire HLA-A gene in the deletion was confirmed by polymerase chain reaction direct sequencing of the surrounding regions of HLA-A. Further localization of the breakpoints of the HLA-A deletion at the centromeric and telomeric sides was performed, and these families were shown to possess the identical deletion. We then determined the genomic sequence of the HLA-A-deleted haplotype. Surprisingly, the haplotype turned out to carry an insertion of an SVA (SINE-VNTR-Alu) retrotransposon of 2 kb as well as the 14 kb deletion that included the entire HLA-A gene.

Cardiomyopathy is a heart muscle disease with impaired stretch response that can result in severe heart failure and sudden death. A small proportion of hepatitis C virus (HCV)-infected patients may be predisposed to develop dilated cardiomyopathy (DCM) and hypertrophic cardiomyopathy (HCM). The molecular mechanisms involved in the predisposition remain unknown due in part to the lack of information on their genetic background. Because the human leukocyte antigen (HLA) region has a pivotal role in controlling the susceptibility to HCV-induced liver disease, we hypothesized that particular HLA alleles and/or non-HLA gene alleles within the human major histocompatibility complex (MHC) genomic region might control the predisposition to HCV-associated DCM (HCV-DCM) and/or HCV-associated HCM (HCV-HCM). Here, we present mapping results of the MHC-related susceptibility gene locus for HCV-associated cardiomyopathy by analyzing microsatellite and single nucleotide polymorphism markers. To delineate the susceptibility locus, we genotyped 44 polymorphic markers scattered across the entire MHC region in a total of 59 patients (21 HCV-DCM and 38 HCV-HCM) and 120 controls. We mapped HCV-DCM susceptibility to a non-HLA gene locus spanning from NFKBIL1 to MICA gene loci within the MHC class III-class I boundary region. Our results showed that HCV-DCM was more strongly associated with alleles of the non-HLA genes rather than the HLA genes themselves. In addition, no significant association was found between the MHC markers and HCV-HCM. This marked difference in the MHC-related disease susceptibility for HCV- associated cardiomyopathy strongly suggests that the development of HCV- DCM and HCV-HCM is under the control of different pathogenic mechanisms.

In order to determine whether matching/mismatching for microsatellite polymorphism provides useful information on acute graft-vs-host disease (GVHD), survival, and leukemia relapse in hematopoietic stem cell (HSC) transplantation, we genotyped for polymorphisms at 13 microsatellite loci within the major histocompatibility complex (MHC) of 100 unrelated HSC transplant donor-recipient pairs who were matched at five classical human leukocyte antigen (HLA) loci. A high percentage of allele matching was obtained for five microsatellite loci, DQCARII (96%), MICA (93%), MIB (89%), C1-3-1 (93%), and D6S510 (97%), that are localized within 100 kb of the HLA-DR, HLA-DQ, HLA-B, HLA-C, or HLA-A locus. In contrast, the other eight microsatellites are located farther away from the HLA classical loci and have much lower percentages of allele matching [e.g. tumor necrosis factor a (TNFa) (73%), TNFd (74%), D6S273 (64%), C3-2-11 (46%), C5-3-1 (50%), C5-4-5 (63%), C5-2-7 (68%), and D6S265 (81%)]. Therefore, there were at least eight microsatellite markers with relatively high percentages of mismatches in the donor/recipient pairs with acute or chronic GVHD, poor graft survival, and leukemia relapse. However, there were no statistically significant associations between mismatched donor-recipient pairs at the 13 microsatellite loci and acute or chronic GVHD, graft survival, and leukemia relapse. Nevertheless, allele matching at the microsatellite TNFd locus near the TNFa gene was found by the Fisher's exact double-sided test to be significantly associated with decreased survival in the grade III/IV acute GVHD group. Overall, these results suggest that the matching of microsatellite polymorphisms within the HLA region, especially the ones farthest from the classical HLA loci, was not useful indicator for the outcome of HSC transplantation from unrelated donors. In this regard, the future determination of the genome-wide microsatellite genotypes in HLA-matched donor-recipient pairs, outside the MHC, may be a better possibility for identifying minor histocompatibility genes in linkage disequilibria with microsatellites as potential predictive markers for the occurrence of acute GVHD and survival rate in HSC transplantation.

A large proportion of Japanese with the HLA-B48 allele have a MICA gene deletion associated with a MICB null allele within the class I region of the Major Histocompatibility Complex (MHC). Here, we report for the first time a novel positive association between the presence of a polymorphic Alu insertion, AluyMICB, within the first intron of the MICB gene and the MICAdel/MICBnull/HLA-B48 haplotype for five of six well-characterized Japanese cell-lines. The AluyMICB insertion was found to be present at a frequency of 0.242 in 86 Japanese tissue donors and in four of the five individuals with the HLA-B48 allele. The AluyMICB insertion was also associated with at least three different MICB alleles, *0102, *0107N and *0105, and three different HLA-B alleles, B13, B48 and B57, respectively, in the seven Workshop cell-lines (the 4th Asia-Oceania Histocompatibility Workshop, and the 10th International Histocompatibility Workshop) and the six Japanese cell-lines that were selected for this study. Based on the analysis of associations between different polymorphic markers within the beta block, the MICB*0102 allele was inferred to be the ancestral form of the MICB*0105 and MICB*0107N alleles. The AluyMICB polymorphism can now be used to further investigate its relationship with other MICB alleles and consequently their origins. In addition, we have examined the absence and presence of three other polymorphic Alu markers distributed within the alpha block of the class I region of the HLA-B48/AluyMICB haplotype. We conclude that the extended HLA-B haplotypes are best defined by considering multiple genomic sites including the four polymorphic Alu insertions described in this study.

Hepatitis C virus (HCV) causes various grades of chronic liver disease, ranging from an asymptomatic state to cirrhosis. To assess genetic factors of disease severity, we selected two HCV patient groups according to the following stringent criteria: (i) asymptomatic carrier state (ASC) defined by HCV infection for more than 20 years, normal alanine aminotransferase levels for the past 5 years as well as normal liver histology and/or shape and (ii) liver cirrhosis (LC) as diagnosed by clinical symptoms, liver biopsy and/or ultrasonography. A total of 103 chronically infected Japanese HCV patients (43 ASC and 60 LC) were analyzed. HLA class I and II alleles were established using low resolution DNA typing. HLA-DRB1 and DQB1 genotypes were inferred upon polymerase chain reaction-restriction fragment length polymorphism analysis. Two hundred and one anti-HCV-negative ethnically matched controls were included. The frequencies of DRB1*12 (*1201 and *1202), DQB1*0301 and DRB3*03 alleles were higher in patients with ASC than in those with LC (odds ratio (OR) 11.23, OR 4.25, and OR 3.22, respectively). The frequency of DQB1*0503 were lower in ASC patients compared to LC patients (OR 0.05). No significant differences between groups were observed for age, sex, source of infection, HCV genotype or viral loads. Our findings establish that certain HLA class II alleles strongly influence disease progression following HCV infection.

We have previously suggested that in a Japanese population the susceptible locus for Behcet's disease (BD) is HLA-B51 itself. To confirm this finding in another population, we performed HLA class I typing using the PCR-SSP method and analyzed eight polymorphic markers distributed within 1100 kb around the HLA-B gene using automated sequencer and subsequent automated fragment detection by fluorescent-based technology with the DNA samples of 84 Iranian patients with BD and 87 healthy ethnically matched controls. As a result, three microsatellite alleles (MICA-A6, MIB-348, C1-4-1-217) and HLA-B51 were found to be strongly associated with BD. Of these alleles HLA-B51 is the most strongly associated allele. There were no alleles that were increased in allele frequency at any microsatellite loci centromeric of MICA or telomeric of HLA-B51. Therefore, HLA-B51 was confirmed to be by far the most strongly associated gene with BD in an Iranian population.

Non-obstructive azoospermia is a male infertility characterized by no or little sperm in semen as a result of a congenital dysfunction in spermatogenesis. Previous studies have reported a higher prevalence of particular human leukocyte antigen (HLA) antigens in non-obstructive azoospermia. As the expression of the RING3 gene located in the HLA class II region was predominant in the testis, mainly around spermatids and pachytene spermatocytes, it is tempting to speculate that RING3 is one of the strong candidate genes responsible for the pathogenesis of the disease. In this study, the genetic polymorphism in the RING3 gene was investigated by the direct sequencing technique. As a result, a total of 14 single nucleotide polymorphisms were identified. Among them, six were localized in the coding region but none of them was accompanied by an amino-acid substitution. No significant difference in the allelic distribution at these 14 polymorphic sites was observed between the patients and healthy controls, suggesting that the susceptible gene for non-obstructive azoospermia is not the RING3 gene. Then, in order to map the susceptibility locus for non-obstructive azoospermia precisely within the HLA region, 11 polymorphic microsatellite markers distributed from the SACM2L gene just outside the HLA class II region (187 kb telomeric of the DPB1 gene) to the OTF3 gene in the HLA class I region were subjected to association analysis in the patients. Statistical analysis of distribution in the allelic frequency at each microsatellite locus demonstrated that the pathogenic gene for non-obstructive azoospermia is located within the HLA-DR/DQ subregion. In fact, DRB1*1302 and DQB1*0604 were found to be strongly associated with non-obstructive azoospermia by polymerase chain reaction-based DNA typing. Further, haplotype analysis suggested that the DQB1*0604 allele may play a decisive role in the pathogenesis of non-obstructive azoospermia.

Behcet's disease (BD) is widely known to be strongly associated with human leukocyte antigen (HLA) B51 in many different ethnic groups.Recently, HLA-B51 allele typing of Greek BD patients was performed to study the distribution of B*5101-B*5107 alleles in this Greek population, the B51 antigen strongly associated with BD was found to be predominantly encoded by allele B*5101. As it is now known that the B51 antigen can be encoded by 21 alleles, B*5101-B*5121, we performed HLA-B*51 allele genotyping among 58 Greek patients with BD. After serological HLA typing, typing of HLA-B*51 alleles was performed using the polymerase chain reaction-sequencing-based typing (PCR-SBT) method. The frequency of the B51 antigen was found to be significantly higher in the patient group as compared with the control group (75.9% of patients vs 22.0% of controls. In the genotyping of B51 alleles, 34 out of 44 B51-positive patients possessed B*5101, 13 out of the 44 carried B*5108. In contrast, all of the 9 B51-positive normal controls carried B*5101. This study revealed a strong association between Greeks with BD, both B*5101, B*5108, provided important insights into the molecular mechanism underlying the association between HLA status, this disease.

Behcet’s disease (BD) is known to be associated with human leukocyte antigen (HLA) B51 in many different ethnic groups. An increased incidence of HLA-B51 in the patient group has also been reported in a Japanese population. Recently, the B51 antigen has been identified to comprise 21 alleles, B*5101–B*5121. Further, not only HLA-B*5101 but also HLA-B*5108 were found to be relatively increased in the patient groups among Italian and Saudi Arabian populations. Therefore, we performed HLA-B*51 allele genotyping by the polymerase chain reaction-sequencing based typing (PCR-SBT) method in order to investigate whether there is any correlation of one particular B51-associated allele with Japanese BD. Ninety-six Japanese patients with BD and 132 healthy Japanese volunteers were enrolled in this study. As a result, the phenotype frequency of the B51 antigen was confirmed to be remarkably increased in the patient group as compared to the ethnically matched control group (59.4% in patients vs. 13.6% in controls; Pc=0.0000000000098, R.R.=9.3). In the B*51 allele genotyping, 56 out of 57 B51-positive patients were defined as B*5101 and the remaining one was B*5102. In contrast, all of 18 B51-positive normal controls were B*5101. None of the Japanese patients and healthy controls carried the HLA-B*5108 allele. This study revealed that B*51 allelic distribution in Japanese was different from those in Italian and Saudi Arabian populations, and that the significantly high incidence of the HLA-B51 antigen in the Japanese BD patient group was mostly caused by the significant increase of the HLA-B*5101 allele.

We previously reported a conserved haplotype of HLA B52-DR2 and a significantly high frequency of the major histocompatibility complex (MHC) class I chain-related gene A (MICA) transmembrane-short tandem repeat (TM-STR) 6 allele in Japanese patients with ulcerative colitis (UC). To examine the predominance of the MICA TM-STR 6 allele as a marker of the susceptibility to UC within the susceptible haplotype, the association of each allele with UC was estimated following stratification of the patients to control for any possible confounding effects of other alleles positively associated with UC. Sixty-four patients with UC and 236 unrelated healthy controls were included in this study. All subjects were Japanese. HLA-A, -B, -C, and -DR antigens were determined serologically. A triplet repeat polymorphism of the MICA was determined by direct sequencing. To control for the effect of linkage disequilibrium, Mantel-Haenszel weighed odds ratios were calculated. Significantly higher phenotype frequencies of B52, MICA TM-STR 6, and DR2 were observed in patients with UC. Linkage disequilibria among alleles associated with UC revealed that a B52 –MICA TM-STR 6 – DR2 haplotype was conserved in patients with UC, as in controls. When the association of HLA-B52 was estimated after patient stratification for the possible confounding effect of MICA TM-STR 6 or DR2, a strong significant association of B52 with UC was still observed. In contrast, no association with UC was observed for MICA TM-STR 6 or DR2, after stratification of the possible confounding effect of HLA-B52. These results imply that the significant increase in MICA TM-STR 6 in Japanese patients with UC is attributable to linkage disequilibrium with HLA-B52.

It is well known that Behcet's disease (BD) is strongly associated with human leukocyte antigen (HLA) B51 in many ethnic groups. However, there has been no published report as yet with respect to this association among the Iranian people. Furthermore, since it is now known that the B51 antigen can be encoded by 21 alleles, B*5101-B*5121, we performed HLA-B*51 allele typing as well as HLA class I genotyping of 48 Iranian patients with this disease. As a result, the frequency of the B*51 allele was significantly higher (62.1%) in the patient group as compared with the ethnically matched control group (31.8%) (Pc=0.067, R.R.=3.51). In the genotyping of B*51 alleles, 33 out of the 36 B*51-positive patients possessed B*5101 and the remaining 3 carried B*5108. This study revealed that Iranian patients with BD also had a strong association with HLA-B51. In addition, this significantly high incidence of HLA-B*51 was found to be caused by an increase in both the HLA-B*5101 and HLA-B*5108 alleles. However, there was no significant difference in the HLA-B*51 allelic distribution between the patient and control groups.

Major histocompatibility complex (MHC) class I chain-related gene A (MICA) has been found near the HLA-B gene. The MICA molecule is exclusively expressed on gastrointestinal epithelium and recognized by intestinal epithelial gamma delta T cells, where it exhibits a triplet repeat polymorphism in the transmembrane region. We investigated the possible correlation between MICA genetic polymorphism and ulcerative colitis (UC). Eighty-three patients with UC and 132 unrelated controls were included in this study. All subjects were Japanese. A triplet repeat polymorphism in the transmembrane region of the MICA was determined by direct sequencing procedures after amplification by a polymerase chain reaction. A significantly higher allele and phenotype frequencies of MICA A6 allele were observed in patients with UC than controls (allele frequency: P(c)=0.000011, phenotype frequency: P(c)=0.0049 odds ratio=2.62). A6 homozygous patients with UC showed significantly earlier onset of UC than patients without the A6 allele ((P)c=0.0042). Phenotypes of MICA A6 allele in Japanese are closely related to the disease susceptibility and behavior in UC. Examinations of MICA polymorphism in other ethnic groups may provide important information about the locus of primary responsible gene for UC.

A 100-kb deletion including the MICA gene was recently reported in the HLA-B48 (B*4801)-associated haplotype in Japanese. Interestingly, this MICA deletion is accompanied by a MICB null allele, MICB0107N. In order to further investigate the universality of the apparent tight linkage between these two events, we present data on high-resolution deletion mapping of eight HLA-B48-homozygous individuals. Among these, five carried the MICA deletion linked to MICB0107N, as originally reported. Conversely, the remaining three possessed an intact MICA gene of MICA008 or MICA010 allelic variant associated this time with a putative expressed MICB allele, MICB0102. These results may imply that the expression of both MICA and MICB molecules is indispensable to viability through a yet-to-be understood mutual interaction in immune surveillance.

Cardiac sarcoidosis occurs in 1-5% of sarcoidosis patients. We previously reported a significant increase of the uncommon TNFA (tumor necrosis factor alpha) allele, TNFA2 with cardiac sarcoidosis in Japanese. In order to precisely localize the susceptible locus for cardiac sarcoidosis within the HLA region, genetic polymorphisms of classical HLA genes, non-classical HLA class II genes such as HLA-DMA and -DMB genes and several genes involved in the class I-mediated antigen presentation pathway (TAP1, TAP2, LMP2 and LMP7) were investigated. Further, association analyses using four polymorphic microsatellite markers located around the TAP1 and TNFA genes were also carried out. As a result, HLA-DQB1*0601 was found to be the most significantly associated allele, being more significantly increased than TNFA2. No significant increase of the DR52-associated DRB1 alleles (DRB1*03, 05, 06 and 08), which was suggested to be primarily associated with lung sarcoidosis, was observed in cardiac sarcoidosis. A primary role of DQB1*0601 in determination of the susceptibility to cardiac sarcoidosis was supported by association analysis using four polymorphic microsatellite markers, in which only the TAP1 microsatellite locus, the nearest marker to the DQB1 gene among the microsatellites tested, displayed a significant positive association with cardiac sarcoidosis. On the other hand, the HLA-DQB1*0501-DQA1*0101-DRB1*0101-B7 haplotype showed a negative association with the disease, as similarly observed in lung sarcoidosis. Thus, molecular mechanism for controlling the development of the disease related to HLA molecules are different between cardiac and lung sarcoidosis, whereas those for conferring a resistant trait may be similar to each other.

Behcet's disease has been known to be strongly associated with human leukocyte antigen (HLA) B51, one of the split antigens of HLA-B5. An increased incidence of HLA-B51 in the patient group has also been reported in an Italian population. Since the B51 antigen has been recently identified to comprise nine alleles, B*5101-B*5109, we performed HLA-B51 allele genotyping by the polymerase chain reaction-sequencing based typing (PCR-SBT) method as well as serological HLA-A and -B typing among 21 Italian patients with Behcet's disease in order to investigate whether there is any correlation of one particular B51-associated allele with Behcet's disease. In addition, HLA class II genotyping was performed by the PCR-restriction fragment length polymorphism (RFLP) method. As a result, only the phenotype frequency of the B51 antigen was found to be significantly increased in the patient group as compared to the ethnically matched control group by the corrected P-value analysis (71.4% in patients vs. 17.9% in controls; chi2 = 14.26, Pc = 0.0042, R.R. = 11.5). In the B51 allele genotyping, 11 out of 15 B51-positive patients were B*5101 and the remaining four were B*5108, whereas all of 5 normal controls were B*5101, showing significant association of each allele with Behcet's disease. No significant difference was observed between the patient and control groups in the HLA class II allelic distribution. This study revealed a strong association of Behcet's disease in Italian with B*5108 as well as B*5101, providing important insight into the molecular mechanism underlying an HLA association with Behcet's disease.

Thirteen Saudi Arabian patients with Behcet's disease (BD) were typed for HLA-A and -B alleles by the conventional serologic typing and for HLA-DRB1, -DQB1 and -DPB1 alleles by the polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) method. As a result, the phenotype frequency of the B51 antigen was significantly increased in the patient group as compared to the ethnically matched control group (76.9% in patients vs. 22.20% in controls), but no significant difference was observed in HLA-A, -DRB1, -DQB1 or -DPB1 alleles between the patients and controls, as previously observed in Japanese BD. Further, by HLA-B51 allelic genotyping performed by the polymerase chain reaction-sequence specific primers (PCR-SSP) method, all of the B51-positive patients and controls were found to carry one particular allele, B*5101, except one patient with B*5108.

We have established a sequencing based typing (SBT) method for detection of genetic polymorphism in the exon 2 to 4 domains of the major histocompatibility complex (MHC) class I chain-related gene A (MICA) and applied it to allele typing of 130 healthy Japanese individuals. A 2.2-kb segment including exons 2, 3 and 4 of the MICA gene was amplified by a pair of generic primers followed by cycle sequencing using exon-specific nested primers. In total, 8 alleles were observed in a Japanese population and the most frequent allele was MICA008 with the gene frequency of 30.8%. MICA009 was the second most frequent (16.5%), while the rarest one was MICA007 (1.2%). MICA alleles displayed strong linkage equilibria with HLA-B antigens (i.e. MICA008 with B7, B48, B60 and B61; MICA009 with B51 and B52; MICA002 with B35, B39, B58 and B67; MICA004 with B44, MICA007 with B13 and B27; MICA010 with B46, B62 and B48, MICA012 with B54, B55, B56 and B59; MICA019 and B70, B71 and B62). Recently, the B48 haplotype has been reported to lack the entire MICA gene by a large-scale deletion in a Japanese population. Among 8 serologically B48 homozygous individuals, 4 were found to represent this MICA null allele as assessed by no polymerase chain reaction (PCR) amplification using MICA-specific primers, while the remaining four possessed the intact MICA gene with MICA008 or MICA010.

Genetic polymorphism of the HLA class II loci including the DRB1, DQA1, DQB1 and DPB1 genes was investigated by the polymerase chain reaction-restriction fragment-length polymorphism (PCR-RFLP) method in a Kazak population inhabiting the most northwestern part of China, Urümqi in the Xinjiang Uygur Zizhiqu as well as in a Han population in the same area. Forty-two Kazak and 59 Han unrelated volunteers were enrolled in this study. Among 51 DRB1 alleles tested, 29 alleles were detected, and DRB1*0301 (13.1%) and DRB1*07 (10.7%) in Kazak and DRB1*0901 (11.9%), DRB1*1501 (11.0%) and DRB1*07 (11.0%) in northwestern Han were highly predominant. In 8 DQA1 alleles detected, DQA1*0501 (29.8%) and DQA1*0301 (23.8%) in Kazak, and DQA1*0301 (28.8%) and DQA1*0102 (19.5%) in northwestern Han were the most and the second most common alleles, respectively. Of 18 DQB1 alleles tested, 14 were observed, among which DQB1*0201 and DQB1*0301 were very frequent both in Kazak (23.8% and 21.4%, respectively) and northwestern Han (18.6% and 16.9%, respectively) populations. Of 37 DPB1 alleles tested, 14 were detected. Among them, the frequencies of DPB1*0401 (21.4%), DPB1*0501 (20.2%), DPB1*0402 (19.0%) and DPB1*0201 (16.7%) in Kazak, and those of DPB1*0501 (38.1%) and DPB1*0201 (16.1%) in northwestern Han were highly increased. Several three-locus haplotypes were recognized to predominate significantly, namely DRB1*0301-DQA1*0501-DQB1*0201 (13.1%) and DRB1*0701-DQA1*0201-DQB1*0201 (8.3%) in Kazak; and DRB1*0901-DQA1*0301-DQB1*0303 (11.9%) and DRB1*0701-DQA1*0201-DQB1*0201 (10.2%) in northwestern Han. The dendrogram constructed by the neighbor-joining (NJ) method based on the allele frequencies of the DRB1, DQA1, DQB1 and DPB1 genes of 12 representative populations all over the world including northern Han, southern Han, Manchu and Japanese suggested that Kazak and northwestern Han were the closest to each other, but Kazak was a little farther from the Asian ethnic groups than northwestern Han.

In order to address the possibility that the MICA gene located 47 kb upstream from HLA-B is involved in the pathogenesis of ankylosing spondylitis (AS), we have investigated microsatellite polymorphism in the transmembrane region of MICA in Caucasian patients with AS. The microsatellite allele consisting of 4 repetitions of GCT/AGC was present at significantly higher frequency in the patient group (Pc

A simple and novel genotyping method was developed to detect alleles at the swine leukocyte antigen (SLA)-DRB1 and -DQB1 class II loci by using polymerase chain reaction (PCR)-fluorescently labeled sequence-specific oligonucleotide probes (SSOPs) and Luminex 100 xMAP detection. The PCR-SSOP-Luminex method exhibited accuracy of 95% for both SLA-DRB1 and -DQB1 in 6 homozygous and 16 heterozygous pig samples as confirmed by sequencing the PCR products of the same samples. In addition, 12 low-resolution SLA class II haplotypes consisting of 7 and 9 DRB1 and DQB1 alleles were identified, respectively, in one population of 283 Landrace pigs. This genotyping method facilitates the rapid and accurate identification of two- or four-digit alleles at the SLA-DRB1 and -DQB1 loci.

We previously developed a new technique for HLA class II genotyping by digestion of polymerase chain reaction-amplified genes with restriction endonucleases (PCR-RFLP method). This PCR-RFLP method is an efficient and convenient typing technique for class II alleles. However, small fragments or bands located close to each other on polyacryl-amide gels sometimes prevent precise analysis of the RFLP bands. Furthermore, the restriction enzymes we have reported in the previous papers are not sufficient to identify the genotypes of all heterozygous individuals. Here, we report an improved PCR-RFLP method using some informative restriction enzymes which have either a single cleavage site or, alternatively, no cleavage site in the amplified DNA region, depending on the HLA alleles, making reading of RFLP band patterns much easier. Each second exon of the HLA-DQA1 or -DPB1 gene was selectively amplified from genomic DNAs of 70 HLA-homozygous B-cell lines and 100 healthy Japanese by PCR. Amplified DNAs were digested with restriction endonucleases and then subjected to electrophoresis assaying simply for cutting, or no cutting, of the DNA. ApaLI, HphI, BsaJI, Fokl, MboII and Mnll can discriminate eight alleles of the DQA1 gene. Similarly 19 alleles of the DPB1 gene can be discriminated with Bsp1286I, Fokl, Ddel, BsaJI, BssHII, Cfr13I, Rsal, EcoNI, and AvaII enzymes. This modified PCR-RFLP method can be successfully applied to heterozygotes. Thus, the method is technically simpler and more practical for routine HLA typing work than our previous PCR-RFLP method.

We previously introduced HLA-DQA1,-DPB1 and DQB1 genotyping with the modified PCR-RFLP method using some informative restriction enzymes which have either a single cleavage site or alternatively no cleavage site in the amplified DNA region, depending on the HLA alleles, making reading of RFLP band patterns much easier. In this study, 43 HLA-DRB1 alleles, excluding DRB1*1103 and *1104 for which no restriction enzymes are available to distinguish each from the other, could be defined by this modified PCR-RFLP method combined with 7 pairs of group-specific primers. It is impossible to distinguish DRB 1*0701 and DRB1*0702 as they are identical for the second exon of DRB1. For DRI-DRB1, DR2-DRB1, DR4-DRB1, DR7-DR1, DR9-DRB1, DRw10-DRB1 or DRw52 associated antigens (DR3, w11, w12, w13, w14, and DRw8)-DRB1 gene amplification, the second exon of the DRB1 gene was selectively amplified using each group-specific primer from genomic DNAs of 70 HLA-homozygous B-cell lines and healthy Japanese by PCR. Amplified DNAs were digested with restriction endonucleases and then subjected to electrophoresis assaying simply for cutting, or no cutting, of the DNA, although some alleles can be distinguished only after examination of RFLP band patterns generated and in some cases using double digestion technique with two restriction enzymes. This modified PCR-RFLP method can be successfully applied to all possible DRB1 heterozygotes, despite the fact that 15 pairs of heterozygotes among them cannot be distinguished theoretically by the PCR-SSO method, because the PCR-RFLP method can tell whether two polymorphic sites are linked to each other (cis position) or located on a different chromosome (trans position) by checking the length of RFLP bands generated with double digestion. Thus, the PCR-RFLP method is technically simple, practical and inexpensive for determination of the HLA-DRB1 alleles for routine HLA typing work.

The human major histocompatibility complex (MHC) class I region is believed to contain a large number of disease-related loci for diseases such as Behcet's disease and psoriasis vulgaris. Although many novel genes have recently been identified in this region, it still appears to be difficult to relate any of these new genes to MHC class I-associated diseases as causative genetic factors. During the course of large-scale genomic sequencing of the human MHC class I region, we identified 262 microsatellite sequences with dinucleotide to pentanucleotide repeats around the HLA-B, -C and HLA-E genes. Of these, 26 microsatellites were investigated for repeat polymorphism using 60 HLA homozygous B-cell lines and 60 healthy random individuals. The average number of alleles at these microsatellite loci was 9.6 with a PIC (polymorphism content value) of 0.69. These new polymorphic microsatellite markers will probably be very useful for precise mapping of disease-related genes within the HLA class I region in linkage analysis. Moreover, they will provide a powerful tool to study recombination events in this region, which contributes to haplotypic diversification.

HLA class II (DRB1, DQA1, DQB1 and DPB1) genotyping was performed in 57 unrelated Uygur individuals inhabiting the northwestern China area by the polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) method. Among 98 DRBI alleles tested, 23 alleles were detected, and DRB1*0701 (16.7%) and DRB1*0301 (14.0%) were the most and the second most common alleles, respectively. In 8 DQA1 alleles detected, DQA1*05 (26.3%), DQA1*03 (21.9%) and DQA1*0201 (21.1%) were very frequent. Of 21 DQB1 alleles tested, 13 were observed. Among them, DQB1*02 was highly predominant with the gene frequency of 32.5%. Of 46 DPB1 alleles tested, 15 were detected, among which DPB1*0401 (31.6%) was the most frequent. Two haplotypes predominate clearly; DRB1*0701-DQA1*0201-DQB1*02 (15.5%) and DRB1*0301-DQA1*05-DQB1*02 (12.6%). The dendrogram constructed by the neighbour-joining (NJ) method based on the allele frequencies of the DRB1, DQA1, DQB1 and DPB1 genes of 13 representative populations all over the world suggested that Uygur belonged to the Asian group and lay at the closest genetic distance to a Kazak population inhabiting the same area.

We recently identified a trinucleotide repeat polymorphism, (GCT)n, within the transmembrane (TM) segment of the human MHC class I MICA gene (MHC class I chain-related gene A). Five distinct alleles (A4, A5, A5.1, A6, A9) corresponding to 4, 5, 5 with one nucleotide insertion, 6 and 9 repetitions, respectively, have been detected in various HLA-homozygous B cell lines. Here we present allele frequencies for this trimeric short tandem repeat (STR) in 604 unrelated individuals collected from nine human populations (Japanese, Northern Han, Hui, Uygur, Kazakhstan, Iranian, Saudi Arabian, Greek and Italian) determined using the polymerase chain reaction (PCR) combined with fluorescent-based automated fragment detection technology. All alleles were present in each population, but allelic distributions varied from one population to another. No new alleles (such as A7 or A8) were identified. The evolutionary and structural significance of these data as well as the potential application to forensic medicine is discussed.

The polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) method, which we previously reported as an efficient and convenient typing technique for accurate definition of the HLA-DQA1 and -DPB1 alleles, is now extended and applied to HLA-DRB and -DQB typing. The second exon of the HLA-DRB (B1 and B3 or B4) and DQB (B1 and B2) genes was selectively amplified from genomic DNAs of 70 HLA-homozygous B cell lines by PCR. Amplified DNAs were digested with the restriction endonucleases, which can recognize allelic variations specific for HLA-DR, -DQ, and -Dw allospecificities and then subjected to electrophoresis in polyacrylamide gel. Of DRB genes, FokI, HinfI, HhaI, HphI, KpnI and SacII were selected and the 20 different polymorphic patterns of the restriction fragments thus obtained were found to correlate with each HLA-DR and -Dw type defined by serological and cellular typing. Of the DQB genes, FokI, HaeIII, HhaI, RsaI and Sau3AI produced nine different polymorphic patterns of the restriction fragments, correlating with the HLA-DQ and -Dw types. This PCR-RFLP method provides a simple and rapid technique for accurate definition of the HLA-DR, -DQ and -Dw types at the nucleotide level, eliminating the need for radioisotope as well as allele specific oligonucleotide probes.