Chromosomal localization, gene structure, and expression pattern of DDAH1: comparison with DDAH2 and implications for evolutionary origins.

Endogenously produced asymmetrically methylated arginine residues are competitive inhibitors of all three isoforms of nitric oxide synthase (NOS). The enzyme dimethylarginine dimethylaminohydrolase (DDAH) specifically hydrolyzes these asymmetrically methylated arginine residues to citrulline and methylamines. Previously we have proposed that regulation of asymmetric methylarginine concentration by DDAH may provide a novel mechanism for the regulation of NOS activity in vivo. Recently we reported the cloning of human DDAH and identified a novel human DDAH isoform (DDAH I and DDAH II, respectively). Here we report that the DDAH1 gene maps to chromosome 1p22 and confirm that DDAH2 maps to the MHC III region of chromosome 6p21.3. Extensive analysis of the distribution of DDAH1 and DDAH2 mRNA in 50 human tissues indicates differential expression of DDAH isoforms in brain regions, in immune cells, and during development. DDAH2 expression predominates in highly vascularized tissues that express the endothelial NOS isoform and in immune tissues that can express iNOS. Whereas DDAH2 is expressed at relatively high levels in all fetal tissues examined, DDAH1 expression varies little between fetal and adult tissues. The chromosomal localization of the DDAHs is consistent with gene duplication, and consistent with this, comparison of the gene structures indicates that the intron/exon organization is highly conserved. Phylogenetic analysis of DDAH sequences from diverse species suggests that DDAH gene duplication occurred prior to the emergence of bony fish some 400 million years ago. Overall the data suggest that DDAH2 may be the more ancient of the two genes.

Mapping of human non-muscle type cofilin (CFL1) to chromosome 11q13 and muscle-type cofilin (CFL2) to chromosome 14.

Cofilin is a widely-distributed, intracellular, actin binding protein which is involved in the translocation of actin-cofilin complex from cytoplasm to nucleus. We have cloned a non-muscle-type cofilin (CFL1) from a human promyelocytic cDNA library and mapped this to human chromosome 11 by PCR amplification of 3' untranslated sequence in a panel of rodent-human somatic cell hybrids, and to the interval 11q12-q13.2 in a chromosome 11 somatic cell hybrid mapping panel. Confirmation of regional localisation to 11q13 has been obtained by fluorescent in situ hybridisation of genomic cosmid clones, by demonstration of the presence of both SEA (the human homologue of avian retrovirus proviral tyrosine kinase, 11q13) and CFL1 in some of these clones and by close linkage of CFL1 to SEA in a panel of high-dose irradiation hybrids. We have identified human muscle-type cofilin sequences by comparison of human expressed sequence tags with M-type cofilins of other species and we have mapped the human M-type cofilin, CFL2, to chromosome 14.

Chromosomal localisation of genes coding for human and mouse liver cytosolic cysteine dioxygenase.

A panel of 22 hybrids was tested for the presence of the gene coding for human cysteine dioxygenase (CDO) by using human specific oligonucleotide primers in the polymerase chain reaction. Detection of human CDO completely correlated with the presence of human chromosome 5. A human total genome cosmid library was screened with a PCR product from the coding region of human CDO cDNA and the two positive clones identified were used in fluorescent in situ hybridisation (FISH) analysis on metaphase chromosome spreads. Fluorescent signals were seen on chromosome 5q22-23. Interspecific backcross mapping in the mouse indicated that Cdo, the mouse homologue of CDO, is situated in the central region of mouse chromosome 18 which shares a region of homology with human chromosome 5.

Mutations of the microsomal triglyceride-transfer-protein gene in abetalipoproteinemia.

Elevated plasma levels of apolipoprotein B (apoB)-containing lipoproteins constitute a major risk factor for the development of coronary heart disease. In the rare recessively inherited disorder abetalipoproteinemia (ABL) the production of apoB-containing lipoproteins is abolished, despite no abnormality of the apoB gene. In the current study we have characterized the gene encoding a microsomal triglyceride-transfer protein (MTP), localized to chromosome 4q22-24, and have identified a mutation of the MTP gene in both alleles of all individuals in a cohort of eight patients with classical ABL. Each mutant allele is predicted to encode a truncated form of MTP with a variable number of aberrant amino acids at its C-terminal end. Expression of genetically engineered forms of MTP in Cos-1 cells indicates that the C-terminal portion of MTP is necessary for triglyceride-transfer activity. Deletion of 20 amino acids from the carboxyl terminus of the 894-amino-acid protein and a missense mutation of cysteine 878 to serine both abolished activity. These results establish that defects of the MTP gene are the predominant, if not sole, cause of hereditary ABL and that an intact carboxyl terminus is necessary for activity.

Structure of the human N-cadherin gene: YAC analysis and fine chromosomal mapping to 18q11.2.

The cadherins are a large family of cell adhesion molecules involved in calcium-dependent recognition and adhesion events. We have used YAC analysis to determine the structure of the human N-cadherin gene. An 850-kb YAC was isolated and the entire N-cadherin gene mapped to a 250-kb region spanning three putative CpG islands. A PCR and cosmid subcloning strategy was used to define the boundaries for the 16 exons that compose the gene. These were shown to be not only highly conserved between mouse and human N-cadherin genes, but also similar to other cadherins. The first and second introns of the gene are large, a property conserved between the mouse and human genes. In situ hybridization with YAC DNA refined the map position of N-cadherin to human chromosome 18q11.2.

Molecular cloning, genomic organization, and chromosomal localization of the human pancreatitis-associated protein (PAP) gene.

Pancreatitis-associated protein (PAP) is a secretory pancreatic protein present in small amounts in normal pancreas and overexpressed during the acute phase of the pancreatitis. In this paper, we describe the cloning, characterization, and chromosomal mapping of the human PAP gene. The gene spans 2748 bp and contains six exons interrupted by five introns. The gene has a typical promoter containing the sequences TATAAA and CCAAT 28 and 52 bp upstream of the cap site, respectively. We found striking similarities in genomic organization as well as in the promoter sequences between the human and rat PAP genes. The human PAP gene was mapped to chromosome 2p12 using rodent-human hybrid cells and in situ chromosomal hybridization. This localization coincides with that of the reg/lithostathine gene, which encodes a pancreatic secretory protein structurally related to PAP, suggesting that both genes derived from the same ancestral gene by duplication.

Regional localization of the lactase-phlorizin hydrolase gene, LCT, to chromosome 2q21.

The gene LCT which codes for the intestinal disaccharidase lactase-phlorizin hydrolase has previously been mapped, using somatic cell hybrids and linkage analysis, using the CEPH families, to chromosome 2. We describe here the regional localization of LCT to chromosome 2q21 by polymerase chain reaction (PCR) analysis of somatic cell hybrids and in situ hybridization. LCT is closely linked to D2S44, with a lod score of 30.6 at theta = 0.10.

Regional localization of the gene coding for the GM2 activator protein (GM2A) to chromosome 5q32-33 and confirmation of the assignment of GM2AP to chromosome 3.

The gene coding for the GM2 activator protein (GM2A) was previously mapped by us to chromosome 5 by an ELISA-based technique. Here we confirm this assignment using a PCR analysis of somatic cell hybrids and describe a regional localization to chromosome 5q32-33 by in situ hybridization. We also confirm the assignment of a pseudogene GM2AP to chromosome 3.

Chromosome 12-containing markers, including two dicentrics, in three i(12p)-negative testicular germ cell tumors.

A chromosome 12-derived marker was seen in each of 3 testicular germ cell tumors that lacked the i(12p). An interesting feature of 2 of the markers was that the major part, including the centromere, of an acrocentric (a #13 and #14, respectively) was translocated onto 12p, resulting in a dicentric. In the third tumor, 13q (translocated onto 12q) was again probably involved in the rearrangement. The findings support the view that the amplification of genes on 12p represents a significant step in the development of germ cell tumors.

Localization of genes encoding three distinct flavin-containing monooxygenases to human chromosome 1q.

We have used the polymerase chain reaction to map the gene encoding human flavin-containing monooxygenase (FMO) form II (N. Lomri, Q. Gu, and J. R. Cashman, 1992, Proc. Natl. Acad. Sci. USA 89: 1685-1689) to chromosome 1. We propose the designation FMO3 for this gene as it is the third FMO gene to be mapped. The two other human FMO genes identified to date, FMO1 and FMO2, are also located on chromosome 1 (C. Dolphin, E. A. Shephard, S. Povey, C. N. A. Palmer, D. M. Ziegler, R. Ayesh, R. L. Smith, and I. R. Phillips, 1991, J. Biol. Chem. 266: 12379-12385; C. Dolphin, E. A. Shephard, S. F. Povey, R. L. Smith, and I. R. Phillips, 1992, Biochem. J. 286: 261-267). The localization of FMO1, FMO2, and FMO3 has been refined to the long arm of chromosome 1. Analysis of human metaphase chromosomes by in situ hybridization confirmed the mapping of FMO1 and localized this gene more precisely to 1q23-q25.

Irradiation hybrids for human chromosome 11: characterization and use for generating region-specific markers in 11q14-q23.

High-dose irradiation hybrids containing fragments of chromosome 11 have been generated, with a view to isolating new region-specific markers. Forty-seven lines were scored for 34 markers: average retention was 6%. Fourteen lines contain markers from 11q14 to 11q23. One of these, Jo12, has 11q markers extending from tyrosinase (q14-q21) to PBGD (q23.3) plus one marker (TYRL, p11.2) from 11p. In situ hybridization using Alu PCR products from Jo12 as probe confirmed that the human DNA is derived from two regions, one in proximal 11p and a second, larger region in 11q23. Plasmid libraries of Alu PCR products from this and three other hybrids have been made. Six of eight recombinants identified as having single-copy inserts were mapped back to the regions of 11q22-q23 detected in the originating hybrid; only one mapped to a region not originally detected, and one was of hamster origin. These six clones provide new markers in 11q22-q23 that can be used directly for polymorphism studies. This series of hybrids is therefore a valuable resource for the rapid generation of markers from specific, defined regions of chromosome 11.

Human regeneration protein/lithostathine genes map to chromosome 2p12.

The pancreatic stone protein (lithostathine) secreted by the exocrine pancreas is an inhibitor of CaCO3 crystal growth. This protein, which is also present in endocrine pancreas, has also been called the regeneration protein (reg). Here we report the mapping of the REG gene to chromosome 2 using the polymerase chain reaction for the specific amplification of human reg sequences in rodent/human somatic cell hybrid DNA. A regional assignment has been made by in situ hybridization to metaphase chromosomes using two different fluorescently labelled genomic probes corresponding to the REG gene and a related gene REGL. Both probes hybridized to chromosome 2p12 suggesting the tandem organization of these genes.

Regional localization of the intestinal mucin gene MUC3 to chromosome 7q22.

The gene MUC3 which codes for a mucin expressed in intestine (Gum et al. 1990) has previously been mapped, using somatic cell hybrids, to chromosome 7. We describe here the regional localization of MUC3 to chromosome 7q22 by in situ hybridization. Preliminary linkage analysis using CEPH (Centre d'Etude du Polymorphisme Humain) families supports this assignment and places MUC3 in the same linkage group as COL1A2 and CFTR.

Structure and chromosomal localization of the human 2',3'-cyclic nucleotide 3'-phosphodiesterase gene.

Human brain cDNA clones for the myelin associated enzyme 2',3'-cyclic nucleotide 3'-phosphodiesterase (CNPase) have been isolated and sequenced. The only 5' untranslated region (UTR) sequence found was that of a human CNPII mRNA, with no direct evidence for a CNPI mRNA. Human CNPase cDNAs were used to isolate genomic clones containing the human CNPase gene which is 9 kb long. Four exons were identified, separated by three introns, and the sequence of each exon and intron/exon boundary has been established. The polymerase chain reaction (PCR) was used to detect the presence of the human CNPase gene in DNA from a panel of rodent/human somatic cell hybrids. By this means the human CNPase gene was mapped to chromosome 17. In situ hybridization of a human CNPase genomic clone to metaphase chromosomes further localized this gene to chromosomal band 17q21.

Possibly identical marker chromosome der(16)t(?13;16)(?q13or14;q22) in a squamous cell carcinoma of the skin and larynx.

A possibly identical marker chromosome was seen in two squamous cell carcinomas, of the face and larynx respectively, in direct preparations or 24-hour cultures. The probable designation of the marker, which may represent a significant cytogenetic event contributing to the evolution of these tumors, was der(16)t(?13;16)(?q13or14;q22).

Phosphoglucomutase 1: complete human and rabbit mRNA sequences and direct mapping of this highly polymorphic marker on human chromosome 1.

A cDNA clone encoding the mRNA for the highly polymorphic human enzyme phosphoglucomutase 1 (PGM1; EC 5.4.2.2) has been isolated and characterized. This was achieved indirectly by first isolating a rabbit cDNA from an expression library using anti-rabbit PGM antibodies. A comparison of the nucleotide sequences shows that the homologies between human and rabbit PGM1 mRNAs are 92% and 97% for the coding nucleotide sequence and the amino acid sequence, respectively. The derived rabbit amino acid sequence is in complete agreement with the published protein sequence for rabbit muscle PGM. A physical localization of the human PGM1 gene to chromosome 1p31 has been determined by in situ hybridization. Analysis of DNA from a wide variety of vertebrates indicates a high level of PGM1 sequence conservation during evolution.

The gene for human neurone specific ubiquitin C-terminal hydrolase (UCHL1, PGP9.5) maps to chromosome 4p14.

Ubiquitin carboxy terminal hydrolase 1, UCHL1, is a neurone-specific protein involved in the ubiquitin-mediated proteolytic pathway. The gene for human UCHL1 has been mapped to chromosome 4 using the polymerase chain reaction to amplify specifically the human UCHL1 sequences in rodent/human somatic cell hybrid DNA. A regional assignment of this locus to 4p14 has been made by in situ hybridization to metaphase chromosomes using both tritium and fluorescently labelled probes.

Chromosome changes in a squamous cell carcinoma of the vagina.

The findings on direct chromosome preparations of a moderately differentiated squamous cell carcinoma of the vagina are described. Most counts were in the range 82-86. Eight markers were present in at least five of six metaphases karyotyped: 3p- (1-2 copies); i(5p) or possibly 5q- (2-3 copies); i(8q) (2 copies); 11q- (2-4 copies); 15p+ (1-2 copies), a probable 18q- (1 copy), 22p+ (1 copy), and minute acrocentric (2-4 copies). Numerical changes included extra copies of chromosomes 7 and 13, but only one copy of chromosome 11 was present.

Mapping the human CAS2 gene, the homologue of the mouse brown (b) locus, to human chromosome 9p22-pter.

Melanin biosynthesis is a multistep process with the first step being the conversion of L-tyrosine to L-Dopa catalyzed by the enzyme tyrosinase. The enzymes which catalyze the other steps of melanogenesis are not known. One murine pigmentation gene, the brown (b) locus, when mutated, leads to a brown or hypopigmented coat. The b-locus protein has been shown to display catalase activity. The human b-locus, therefore, is designated as CAS2. We used the mouse b-locus cDNA to isolate the human homologue, which in turn, was used to map the CAS2 locus to a human chromosome. The potential CAS2 protein codes for 527 amino acids containing a putative signal sequence and transmembrane domain. The CAS2 protein has primary and probably secondary structures similar to human tyrosinase. The CAS2 was mapped to human Chromosome 9 by somatic cell hybridization and, more specifically, to 9p22-pter by in situ hybridization. The assignment of CAS2 on the human Chromosome 9 extends this region of known homology on mouse Chromosome 4.