Hidetoshi Inoko

NUCLEOTIDE SEQUENCE OF THE HUMAN MHC CLASS I MICA GENE

In addition to the closely related genes encoding the conventional class I peptide-presenting molecules, the major histocompatibility complex (MHC) of humans and most mammals contains one or several highly diverged class I-like genes representing a second class I gene family (Bahram et al. 1994). In humans, two members of this gene family, MICA and MICB, are encoded near HLA-B and encode mRNA transcripts with intact long open reading frames (Bahram et al. 1994; Bahram and Spies 1996). Additional similar sequences, MICC, MICD, and MICE are widely spaced throughout the 2 megabase class I region and are pseudogenes because of several point mutations and/or gross deletions (S. Bahram and T. Spies, unpublished). Here we report the complete nucleotide sequence of the MICA gene comprising 11 722 basepairs (bp) of DNA 40 kilobases centromeric of HLA-B. The sequence was obtained from single (M13)and double (pUC19)-stranded templates of mapped or randomly shot-gun subcloned DNA fragments that were derived from the cosmids M32A (Spies et al. 1989) and pM67 (N. Mizuki and H. Inoko, unpublished). The first exon encoding the leader peptide is followed by an intron of 6840 bp, which is unusually large for a class I gene. The remainder of the MICA gene shows an organization quite similar to that of conventional class I genes, except for the presence of a relatively long intron following the transmembrane exon and the fusion of the cytoplasmic tail and 39 untranslated sequence in a single last exon. The translated amino acid sequence (not shown) corresponds to the previously identified MICA4 allele. Together with the recently documented sequence of an MICB cDNA (Bahram and Spies 1996), these data complete the primary genetic characterization of the highly diverged MHC class I MICA and MICB genes.

Precise switching of DNA replication timing in the GC content transition area in the human major histocompatibility complex.

The human genome is composed of long-range G+C% (GC%) mosaic structures thought to be related to chromosome bands. We previously reported a boundary of megabase-sized GC% mosaic domains at the junction area between major histocompatibility complex (MHC) classes II and III, proposing it as a possible chromosome band boundary. DNA replication timing during the S phase is known to be correlated cytogenetically with chromosome band zones, and thus the band boundaries have been predicted to contain a switch point for DNA replication timing. In this study, to identify to the nucleotide sequence level the replication switch point during the S phase, we determined the precise DNA replication timing for MHC classes II and III, focusing on the junction area. To do this, we used PCR-based quantitation of nascent DNA obtained from synchronized human myeloid leukemia HL60 cells. The replication timing changed precisely in the boundary region with a 2-h difference between the two sides, supporting the prediction that this region may be a chromosome band boundary. We supposed that replication fork movement terminates (pauses) or significantly slows in the switch region, which contains dense Alu clusters; polypurine/polypyrimidine tracts; di-, tri-, or tetranucleotide repeats; and medium-reiteration-frequency sequences. Because the nascent DNA in the switch region was recovered at low efficiency, we investigated whether this region is associated with the nuclear scaffold and found three scaffold-associated regions in and around the switch region.

A boundary of long-range G+C% mosaic domains in the human MHC locus: Pseudoautosomal boundary-like sequence exists near the boundary

The human genome is composed of long-range G+C% (GC%) mosaic structures related to chromosome bands. We found the human MHC locus to be an example of megabase-level GC% mosaic structures and predicted a possible boundary of the megabase-level domains within an undercharacterized 450-kb region harboring the junction of MHC classes II and III. Chromosome walking of the 450-kb region and base-compositional analysis precisely located the boundary of the mosaic domains, disclosing a sharp GC% transition. Near the transition point there was a 20-kb dense Alu cluster, a 30-kb dense LINE-1 cluster, and a sequence highly homologous with the pseudoautosomal boundary of the short arms of human sex chromosomes (PAB1X and PAB1Y); PAB1X and PAB1Y are the interface between sex-specific and pseudoautosomal regions. Many PAB1XY-like sequences (PABLs) were detected by hybridization against genomic DNA, and the new sequences defined the complete form of PABLs to be about 650 nt.

Evolutionary significance of intra-genome duplications on human chromosomes

Phylogenetic analyses indicated that a series of paralogous gene pairs, found in two extensive regions on human chromosomal bands 6p21.3 and 9q33-34, were created by at least two independent duplications. The duplicated genes on chromosomal band 6p21.3 include the genes for type 11 collagen alpha2 subunit (COL11A2), NOTCH4 (mouse int-3 homologue), 70 kDa heat shock protein (HSPA1A, HSPA1B, and HSPA1L), valyl-tRNA synthetase 2 (VARS2), complement components (C2 and C4), pre-B cell leukemia transcription factor 2 (PBX2), retinoid X receptor beta (RXRB), NAT/RING3, and four other proteins. Their paralogous genes on chromosomal band 9q33-34 are genes for type 5 collagen alpha1 subunit (COL5A1), NOTCH1, 78 kDa glucose-regulated protein (HSPA5), valyl-tRNA synthetase 1 (VARS1), complement component V (C5), PBX3, retinoid X receptor alpha (RXRA), ORFX/RING3L, and others. Among these, the genes for collagen, complement components, NAT/RING3, PBX, and RXR appear to have been duplicated around the time of vertebrate emergence, supporting the idea that they were duplicated simultaneously at that time. Another group of genes that includes NOTCH and HSP appear to have diverged long before that time. A comparison of the physical maps of these two regions revealed that the genes which duplicated in the same period were arranged in almost the same order in the two regions, with the assumption of a few chromosomal rearrangements. We propose a possible model for the evolution of these regions, taking into account the molecular mechanisms of regional duplication, gene duplication, translocation, and inversion. We also propose that a comparative mapping of paralogous genes within the human genome would be useful for identifying new genes.

Gene organization of human NOTCH4 and (CTG)n polymorphism in this human counterpart gene of mouse proto-oncogene Int3

The cDNA and genomic clones for the human counterpart of the mouse mammary tumor gene Int3 were isolated and sequenced. We designated this human major histocompatibility complex (MHC) class III gene as NOTCH4, since very recently, by sequencing cDNA clones, the complete form of the mouse proto-oncogene Int3 has been clarified and named Notch4. The present human NOTCH4 sequence is the first example of the genomic sequence for the extracellular portion of the mammalian Notch4, and by comparing it with the mouse Notch4 cDNA sequence, the exon/intron organization was clarified. The comparison of the predicted amino acid sequence of human NOTCH4 with those of other Notch homologues of a wide range of species revealed four subfamilies for mammalian Notch. In the protein coding region of human NOTCH4, we found (CTG)n repeats showing a variable number tandem repeat (VNTR) polymorphism for different human leukocyte antigen (HLA) haplotypes. Ten genes mapped on 6p21.3, including NOTCH4, were found to have counterparts structurally and functionally similar to those mostly mapped on 9q33-q34, indicating segmental chromosome duplication during the course of evolution. Similarity of genes on chromosomes 1, 6, 9 and 19 was also discussed.

Allelic repertoire of the human MICB gene.

A distinct family of major histocompatibility complex (MHC) class I genes has recently been identified within the human MHC. Members of this MHC class I chainrelated gene (MIC) family are dispersed over the 2 megabase HLA class I region (Bahram et al. 1994; Bahram and Spies 1996a; Campbell and Trowsdale 1997). The MICA and MICB transcripts in this family carry full-length openreading frames encoding typical MHC class I-like polypeptide chains with three (α1–3) extracellular domains, a transmembrane, and a cytoplasmic segment (Bahram et al. 1994; 1996a, b; Bahram and Spies 1996a). MICA and MICB are quite similar, sharing more than 90% similarity, but are very distantly related (having less than 30% overall similarity) to other MHC-I genes (Bahram et al. 1994; Bahram and Spies 1996a). MICC, MICD, and MICE are pseudogenes due to debilitating mutations or deletions (Bahram and Spies 1996b). This mirrors the composition of the HLA class I gene family, in which six functional genes (HLA-A to G) are interspersed between 12 pseudogenes or gene fragments (Geraghty 1993). MICs are conserved across mammalian evolution, which defines them as a second lineage of MHC class I genes (Bahram et al. 1994). Unlike typical MHC class I genes, MIC are not responsive to type I and II interferons but are regulated by cell stress through upstream heat shock response elements (Bahram et al. 1994; Groh et al. 1996). Generation of monoclonal antibodies demonstrated that MICA is a membrane-bound single-chain glycoprotein, expressed almost exclusively in the gastrointestinal epithelium (Groh et al. 1996). These data raised the possibility that MICA may serve as restriction elements for intestinal intra-epithelial T lymphocytes. In light of the astonishing degree of HLA polymorphism, it was important to measure the extent of MIC allelic variation (Parham and Ohta 1996). Recent work unveiled a relative high degree of polymorphism in the MICA gene, with 16 alleles reported thus far (Fodil et al. 1996). Remarkably, none of the amino acid replacements coincided with the well-defined HLA polymorphic residues concentrated at peptide or T-cell receptor contact sites (Bjorkman and Parham 1990). Interestingly, superimposition of MICA variable residues on an HLA-A2 tridimensional structure revealed their preferential positioning along the periphery of the putative antigen binding cleft, leaving an apparently invariant ligand binding site (Fodil et al. 1996). The close proximity of MICA to HLA-B warranted examination of the possible role of these alleles in MHC class I-associated disorders (Dausset and Svejgaard 1977). In fact, one transmembrane variant of MICA is preferentially associated with Behçet’s disease at a significantly higher rate than is the previously implicated HLA-B51 gene (Mizuki et al. 1997). Given these encouraging results, we embarked on defining potential MICB alleles. The allelic repertoire of MICB was evaluated using genomic DNA from 33 HLA homozygous typing cell lines (HTCL) collected during the 10th International Histocompatibility Workshop (Tsuji et al. 1992) and purchased from the European Collection of Animal Cell Cultures (England). These were, BM14, BM15, BM92, BTB, DBB, DEU, DHIF, DKB, EHM, EJ32B, FPAF, HOM2, JESTHOM, JO528239, JVM, KAS116, LWAGS, OMW, PF97387, PITOUT, PMG075, RSH, SA, SCHU, SPACH, TAB089, TEM, TISI, VAVY, WT24, YAR, SP0010, and The nucleotide sequence data reported in this paper have been submitted to the EMBL/GenBank nucleotide sequence databases and have been assigned the accession numbers U95729 (MICB002), U95730 (MICB003), U95731 (MICB004), U95732 (MICB005), U95733 (MICB006), U95734 (MICB007)

Cloning of a new kinesin-related gene located at the centromeric end of the human MHC region

We previously reported the presence of a new gene (HSET) with an unknown function, in the centromeric side of the class II gene region of the human major histocompatibility complex (MHC). cDNA clones corresponding to the HSET gene were isolated from a human testis cDNA library. A 2.4 kilobase transcript from the HSET gene was abundantly expressed in testis, B-cell, T-cell, and ovary cell lines but was not detected in lung or stomach. Analysis of the nucleotide sequence of the HSET cDNA clones revealed significant similarity to kinesin-related proteins in yeast, Drosophila, and human. Its predicted amino acid sequence contains a domain with strong sequence similarity to the ATP-binding and motor domains of a plus end-directed microtuble motor protein, kinesin, which might be involved in mitotic chromosome segregation, suggesting that the HSET gene encodes a novel kinesin-related protein.

GENOMIC STRUCTURE OF THE HUMAN MHC CLASS I MICB GENE

The human major histocompatibility complex ( MHC) contains, in addition to the well known HLA class I genes, the highly divergentMHC class I chain-related ( MIC) gene family (Bahram et al. 1994; Bahram and Spies 1996a). Within this family, the closely relatedMICA and MICB genes encode transcripts of 1382 and 2376 base pairs (bp), respectively (Bahram et al. 1994; Bahram and Spies 1996b). The MICA molecule is encoded by an unusually large gene of 11722 bp, located about 40 kilobases (kb) centromeric to theHLA-B locus (Bahram et al. 1996). Here we report the 12930 bp nucleotide sequence of the MICB gene, located approximately 70 kb centromeric to MICA. The sequence was obtained from shot-gun subcloned DNA fragments derived from the cosmid TY2A9, a member of a cosmid contig covering this HLA sub-region and generated from a YAC clone isolated from the B-LCL, BOLETH (H. Inoko, unpublished data). The genomic organization of MICB is similar to that ofMICA (Bahram et al. 1996; Fig. 1) and distinct from other class I genes (Malissen et al. 1982). In particular, a large intron of 7352 bp (6840 bp in MICA) separates the first two exons, and a single six exon of 1338 bp (302 bp inMICA) encodes both the cytoplasmic tail and the 39 untranslated (3 9UT) sequence. Excess 3 9UT sequence is therefore responsible for the approximately 1 kb length difference between the MICA and MICB transcripts (Bahram et al. 1994, Bahram and Spies 1996b). As expected, the high degree of similarity observed within the coding sequences of MICA and MICB extends throughout both genes. Indeed, comparing individual exons and introns of MICB with their counterparts inMICA provides the following degrees of sequence similarity (based on an alignment employing the default parameters of the Clustal method of the MegAlign program, Lasergene Navigator (DNASTAR, Madison, WI): 80% for exon 1, 52% for intron 1, 91% for exon 2, 86% for intron 2, 90% for exon 3, 87% for intron 3, 98% for exon 4, 97% for intron 4, 73% for exon 5, 83% for intron 5, and finally 82% for exon 6. Comparing the MICB genomic sequence with the previously published cDNA sequence reveals four non-synonymous nucleotide substitutions [numbers correspond to nucleotide positions reported in Bahram and Spies (1996b)]: 121: GAA(E)?GGA(G), 243: AAG(K)?GAG(E) (both within α1), 411: GAT(D)?AAT(N) (α2) and 904 GTG(V)?GCG(A) (TM). Interestingly, none of these positions match the previously defined MICA polymorphic residues (Fodil et al. 1996). Availability of the nucleotide composition of the MICA andMICB genes adds to our knowledge of this particular segment of the human MHC, which is specifically associated with susceptibility to a number of rheumatic disorders.