High-resolution molecular characterization of the HLA class I and class II in the Tarahumara Amerindian population

We describe for the first time the high-resolution profiling of HLA-A, -B, -C, -DRB1, -DQB1 and -DPB1 in a culturally and geographically distinct Mexican ethnic group, the Tarahumaras. The alleles most frequently found by reference strand-mediated conformational analysis in this population were for class I: HLA-A*240201, *020101/09, *0206, *310102, *680102; HLA-B*4002, *1501, *510201, *3501/02/03, *4005, *4801; HLA-Cw*0304, *0801, *0102, *040101; and for class II: HLA-DRB1*080201, *1402, *040701; HLA-DQB1*0402, *0301, *0302/07; HLA-DPB1*0402, *0401, *020102. In addition, a novel allele, HLA-A*0257, was found. Based on comparison of presently known HLA-DRB1 and -DQB1 allele frequencies in Amerindian groups and worldwide populations, the Tarahumaras are unexpectedly more related to the geographically and linguistically distant Aymara and Terena Amerindian groups than they are to neighbouring tribes.

Sequence of a novel HLA-A*0301 intronic variant (A*03010103)

We report here the full-length sequence of a novel HLA-A*0301 allele, A*03010103, which differs from A*03010101 by a single nucleotide substitution (G>T) at position 492 within intron 2. The variant was originally identified by Reference Strand-mediated Conformational Analysis (RSCA) and was confirmed by cloning and sequencing. The difference in RSCA mobility between A*03010101 and A*03010103 demonstrates the sensitivity of RSCA to detect single nucleotide polymorphisms.

A novel HLA-A allele: A*0257

A novel human leucocyte antigen-A*02 (HLA-A*02) allele was detected by reference strand-mediated conformation analysis (RSCA) of a DNA sample from a Tarahumara individual. Direct sequencing of HLA-A locus polymerase chain reaction products identified a mutation in one of the alleles. Cloning and sequencing confirmed the presence of a new allele, A*0257 which differed from A*0206 by two nucleotides at positions 355 and 362, inducing changes in residues 95 and 97, respectively, within the peptide-binding site. Those changes suggest that allele A*0257 may have resulted from an intralocus recombination event.

Further polymorphism of the MICA gene

The MHC class I chain-related (MIC) gene family constitutes an interesting genetic group that is related to major histocompatibility complex (MHC) class I genes and is located within the MHC. The MIC gene products, MICA and MICB, have similar structures to HLA class I molecules. So far over 50 MICA alleles have been reported, which suggests that this genetic system is highly polymorphic. In order to investigate further the extent of MICA polymorphism we have studied exons 2-5 of the MICA gene in over 200 homozygous and heterozygous cell lines. Altogether we have identified 11 new MICA alleles and report 13 new nucleotide variations, one in exon 2, four in exon 3, four in exon 4, two in intron 1, one in intron 4 and one (a deletion) in exon 4. Eight of the 10 exonic variations are non-synonymous. The deletion in exon 4 leads to a frame-shift mutation and the introduction of a repeat of 12 leucine residues encoded by the microsatellite in exon 5. This study provides further evidence that the MICA gene is highly polymorphic. In contrast to MHC class I molecules, the polymorphic sites in MICA are predominantly within the alpha2 and alpha3 domains. The distribution of synonymous and non-synonymous substitutions suggests that there is selection for the polymorphic positions, which therefore define potential functional sites in the protein. We were also able to determine the association between MICA and HLA-B alleles in a number of homozygous cell lines bearing extended haplotypes.

HLA typing by Reference Strand Mediated Conformation Analysis (RSCA)

Reference Strand Mediated Conformation Analysis (RSCA) is a novel conformational method that offers high resolution and high sample throughput for typing the HLA class I and class II genes. This technique differs from conventional sequence-based typing methodologies in that the HLA type is assigned on the basis of accurate measurement of conformation-dependent DNA mobility in polyacrylamide gel electrophoresis (AGE). RSCA utilises a fluorescent-labelled locus-specific reference DNA to selectively modify the molecular conformation of the tested DNA. The use of laser-based instrumentation and computer software, in addition to internal DNA markers for correction of gel variability, allows the discrimination of HLA alleles which differ by one nucleotide in a DNA fragment nearly as large as a kilobase in length. RSCA has been successfully applied in blind studies of HLA typing, demonstrating that is reproducible, able to identify new alleles and able to resolve ambiguous heterozygous combinations.

Reference strand mediated conformation analysis resolves HLA-DRB1 typing ambiguities when matching for unrelated bone marrow transplantation

We show here the use of reference strand mediated conformation analysis (RSCA) to unambiguously resolve the HLA-DRB1 typing of two individuals which were selected as potential unrelated donors for bone marrow transplantation (BMT). In the first case, both sequence-specific primer (SSP) amplification and sequence-specific oligonucleotide probing (SSO), routinely used in different tissue typing laboratories gave, for the two unrelated donors, the same ambiguous typing of HLA-DRB1*04011+*0403 or DRB1*0407+*0413. In this case sequence-based typing (SBT) was not the method of choice to resolve the situation, due to the sequence ambiguities of these two given combinations. RSCA of both samples, using homozygous typing cells (HTCs) for DRB1*04011, *0403 and *0407 as internal controls, gave the unambiguous result that both donors were HLA-DRB1*04011+*0403. In the second case, a donor was typed as DRB1*1102+1103 by SSP, while SSO excluded the DRB1*1102 allele. The patient was unambiguously typed as DRB1*1101+1103 by both techniques. RSCA, using DNA from reference cell lines as internal controls, gave the unambiguous typing that the donor was DRB1*1103 homozygous.

A new MICA allele with ten alanine residues in the exon 5 microsatellite

The MICA and MICB genes code for protein products that have structural similarities to major histocompatibility complex (MHC) class I genes. These genes are upregulated by heat stress. They have been shown to interact with a common receptor (NKG2D/DAP10) on gammadelta T cells, CD8+ T cells and natural killer (NK) cells. The MICA gene has an expressed microsatellite, GCT, within the exon 5 which encodes for alanine. So far, four different repetitions of this short tandem have been reported. Also one non-synonymous, one synonymous substitution and a 1-bp insertion within this region have also been described. An association of Behcet's disease with the microsatellite A9 has been reported. Here we report a novel allele with 10 GCT repetitions (A10) which was detected by reference strand mediated conformation analysis and confirmed by DNA sequencing.

Three novel MICB alleles

The two members of the MHC class I chain-related (MIC) gene family, MICA and MICB, have been shown by several investigators to be polymorphic. Most of the research effort so far has focussed on MICA, so less is known about the extent of polymorphism in the MICB gene. Here we report three novel MICB alleles, which had been detected in the course of an SSOP typing study on a large cohort of cell lines. Two of these alleles are formed by a non-synonymous nucleotide variation. Our results confirm previous findings that most of the polymorphisms in the MICB gene, as in MICA, are coding and suggest that the extent of polymorphism in the two genes might be comparable.

Novel intronic variants of MICB (MHC class I chain-related gene B)

We report an eight-nucleotide duplication in intron 4 of the MICB allele 01021, which was found in samples from different ethnic backgrounds and in association with several HLA-B alleles. We suggest that this new MICB allele is evolutionarily older than HLA-B alleles.

HLA-A typing by reference strand-mediated conformation analysis (RSCA) using a capillary-based semi-automated genetic analyser

HLA typing of class I loci by reference strand-mediated conformation analysis (RSCA) using a slab gel genetic analyser has been described. This study adapted the method for use in the capillary based ABI PRISM 310. Control DNA samples were used to create a database of mobility values for 37 HLA-A alleles. The technique was validated by comparing RSCA and sequence-specific oligonucleotide probe (SSOP)/sequence-specific primer amplification (SSP) HLA-A locus typing results from 214 cord blood samples. Of the samples tested, 6.5% required confirmatory typing by SSP, compared with a repeat rate of 10-40% for SSOP. In 200 samples where no SSP was necessary, there was 100% concordance between RSCA and previous results. The ABI PRISM 310 RSCA method defines HLA-A types at medium resolution and is quick and easy to implement.

Characterization of the MICA polymorphism by sequence-specific oligonucleotide probing

A large number of diseases occur in association with specific HLA-B or -C alleles. Recently a new gene, termed major histocompatibility complex class I chain-related gene A (MICA), has been identified in close proximity to HLA-B. The function of this gene is still unknown, but, it is structurally related to HLA class I genes, is polymorphic, and is potentially associated with several diseases. Some DNA-based techniques have previously been described to type for MICA including sequencing and single-strand conformational polymorphism. In this paper we describe the application of sequence-specific oligonucleotide probe based typing for the analysis of the MICA gene. We used a set of 30 oligonucleotide probes to screen for the polymorphisms in exons 2, 3, and 4, which account for the 16 known alleles. We report here the typing results of MICA for 103 B-cell lines that have been well characterized for HLA and describe the linkage disequilibrium between MICA and HLA-B. Unequivocal MICA typing was achieved for 85 of the 103 cells tested, 6 cells gave ambiguous MICA types, and a further 12 cells showed patterns consistent with them expressing at least one new MICA allele.

Sequence of HLA-A*6808

A novel HLA-A*68 allele was identified by reference strand conformation analysis of a DNA sample studied for the International Cell Exchange program. Direct sequencing of HLA-A locus PCR products confirmed the presence of A*6602 and an A*68 allele which differed from A*68011 by two nucleotides at positions 538 and 539. The presence of this sequence motif in this allele which has been named A*6808 was verified by DNA sequencing of full-length A*6808 clones. No other coding differences between A*68011 and A*6808 were identified. The two nucleotide substitutions found in A*6808 effect an amino acid difference in the encoded protein at residue 156 within the peptide binding site which may evoke alloreactive immune response after HLA-A68 mismatched transplants.

HLA-A, -B, -C polymorphism in a UK Ashkenazi Jewish potential bone marrow donor population

To further our knowledge of HLA polymorphism in different ethnic populations and to increase the number of full HLA class I typed potential bone marrow donors on the Anthony Nolan Bone Marrow Trust register, HLA-A, -B and -C polymorphism was characterised in 412 Ashkenazi Jewish potential donors. Serological typings and limited molecular analysis was performed for HLA-A and -B, and molecular typings were performed for HLA-C. Gene and haplotype frequencies were calculated using the maximum likelihood method and compared with UK Caucasoid and other Jewish populations. While the specificities identified were in general overlapping with the UK Caucasoid data, a difference in the frequencies of individual specificities was observed. For example, HLA-B62, a common serotype found in the UK Caucasoid population, is almost absent in the Ashkenazim. HLA-A, -C, -B haplotype frequencies also differ between the two populations with A26-Cw*1203-B38 and A24-Cw*04-B35 significant in the Ashkenazim, whilst A1-Cw*07-B8, a common Caucasoid haplotype, was found to be less frequent. Overall the results for the UK Ashkenazi population were most similar to previous reports on Polish/Russian Jews.

Molecular typing shows a high level of HLA class I incompatibility in serologically well matched donor/patient pairs: implications for unrelated bone marrow donor selection

In comparison with HLA-matched sibling bone marrow transplants, unrelated donor transplants are associated with increased graft-versus-host disease and graft failure. This is likely in part due to HLA incompatibilities not identified by current matching strategies. High resolution DNA-based typing methods for HLA class II loci have improved donor selection and treatment outcome in unrelated donor bone marrow transplantation. By using DNA-based typing methods for HLA-A and -B on a cohort of 100 potential bone marrow donor/patient pairs, we find that serological typing for HLA class I is limited in its ability to identify incompatibilities in unrelated pairs. Furthermore, the incompatibilities identified are associated with the presence at high frequency of alloreactive cytotoxic T-lymphocyte precursors. DNA typing also indicates that HLA-C mismatches are common in HLA-A and -B serologically matched pairs. Such mismatches appear to be significantly less immunogenic with respect to cytotoxic T-lymphocyte recognition, but are expected to influence natural killer cell activity. Thus, improved resolution of HLA class I shows many previously undisclosed mismatches that appear to be immunologically functional. Use of high resolution typing methods in routine matching is expected to improve unrelated donor selection and transplant outcome.

Application of RSCA for the typing of HLA-DPB1

We describe the application of RSCA, for the high resolution typing of alleles encoded at the HLA-DPB1 locus. RSCA differs from other sequence based typing methodologies in that the HLA type is assigned on the basis of differences in DNA conformation between different alleles. A total of 251 samples were typed in a blind study, of these 109 samples had been typed previously by conventional techniques. A comparison of the RSCA data with the historical typing results showed a concordance over 93%. Seven samples initially had discordant results, however, when these samples were typed by direct sequencing, the type assigned by RSCA was found to be correct in all but one case, indicating a concordance over 99%. RSCA has proved to be a simple reliable technique for the typing of the HLA-DPB1 locus, and is not limited by the ambiguous combinations of alleles determined in other conventional techniques.

A high resolution HLA class I and class II matching method for bone marrow donor selection

HLA matching in bone marrow transplantation has an important role in determining successful outcome. However HLA typing of both potential related and unrelated donors can be both time-consuming and laborious, and does not always resolve accurately the true level of histocompatibility. We have utilised a method, reference strand mediated conformation analysis (RSCA), which is technically simple and allows high resolution matching for all HLA loci, for the typing of 48 patients and their potential 120 donors. The results indicate that RSCA can detect many mismatches that are not routinely identified by conventional HLA typing methods. In addition, RSCA can be applied for the simultaneous analysis of multiple potential BM donor samples in order to quickly identify the best match for each patient.

High resolution HLA class I typing by reference strand mediated conformation analysis (RSCA)

We describe the application of RSCA for the high resolution typing of alleles encoded within the highly polymorphic HLA class I system. RSCA differs from other sequence based typing methodologies in that the HLA type is assigned on the basis of differences in DNA conformation between different HLA alleles. HLA class I locus specific PCR was performed on the sample to be HLA typed and the PCR product mixed with a fluorescent labelled locus specific reference strand. DNA molecules were allowed to reanneal and the duplexes formed were separated by non-denaturing PAGE. Only duplexes which possess the fluorescent labelled reference strand were detected using an automated DNA sequencer. The mobilities of duplexes formed for 131 HLA class I alleles have been calculated and HLA type assigned. The use of a laser scanning system for accurate detection of migrating DNA, in combination with specific software for data analysis and correction of gel variablility, make RSCA sufficiently robust to detect alleles differing by as little as one nucleotide, resulting in a simple and automated technique for high resolution HLA typing.

Mutation detection and typing of polymorphic loci through double-strand conformation analysis

Variations, such as nucleotide substitutions, deletions and insertions, within genes can affect the function of the gene product and in some cases be deleterious. Screening for known allelic variation is important for determining disease and gene associations. Techniques which target specific mutations such as restriction enzyme polymorphism and oligonucleotide probe or PCR primer reactivity are useful for the detection of specific mutations, but these techniques are not generally effective for the identification of new mutations. Approaches for measuring changes in DNA conformation have been developed, based on the principle that DNA fragments which differ in nucleotide composition exhibit different mobilities after separation by polyacrylamide gel electrophoresis (PAGE). Here we describe a conformation-based mutation detection system, double-strand conformation analysis (DSCA), which provides a simple means to detect genetic variants and to type complex polymorphic loci. We demonstrate the application of DSCA to detect genetic polymorphisms such as a single-nucleotide difference within DNA fragments of up to 979 base pairs in length. We present the application of DSCA in detecting four different mutations in the cystic fibrosis gene (CFTR) and 131 different alleles encoded by HLA class I genes.