The glycine cleavage system: structure of a cDNA encoding human H-protein, and partial characterization of its gene in patients with hyperglycinemias.

We have isolated a 1,192-base-long cDNA which encodes the entire structure of a precursor form of human H-protein. The tentatively calculated number of copies for this cDNA appeared to be about four times as many as that of the antithrombin III gene specified by a single locus in the human haploid genome. Southern analysis using H-protein cDNA probe demonstrates a deletion of the 5.0-kb SacI fragment in the genome of a patient with an atypical nonketotic hyperglycinemia in whom there was an inactive H-protein. This SacI fragment was also deleted from the genome of one of seven patients with nonketotic hyperglycinemia resulting from the lesion of glycine decarboxylase. The remaining six patients had common aberrations identified with the 5.2-kb EcoRI and 5.5-kb SacI fragments. Although implication of these defective fragments in pathogenesis is unclear at present, it is suggested that rearrangements occur in multiple genomic loci of patients with nonketotic hyperglycinemia and that this H-protein cDNA can be used for carrier screening.     

DNA length polymorphism located 5'' to the human myelin basic protein gene.

A region of DNA 5' to the human myelin basic protein (MBP) gene, located on the long arm of chromosome 18, is a site of restriction-fragment-length polymorphism (RFLP) showing 37% heterozygosity in 40 subjects studied. Southern transfer analysis using a 0.9-kb genomic fragment encompassing the first exon of the human MBP gene reveals this polymorphism with at least nine restriction enzymes, indicating that insertion, deletion, or both is the basis for the DNA length variation. Double restriction-enzyme digest analysis suggests that this polymorphism is within the region 0.5-2.0 kb upstream of the coding region of the first exon of the human MBP gene. Eleven different allelic RFLPs were identified, differing in size by as many as 450 bp. The distribution of insertion/deletion-size variants from this region is bimodal, with most restriction fragments varying in size over a 0.1-kb range. Pedigree analysis of polymorphism at this site in one three-generation family shows Mendelian assortment of parental haplotypes. The form and frequency of polymorphism generated by this site is similar to that reported for human DNA regions comprised of homologous short tandem repeats.     

One of the two genomic copies of the glycine decarboxylase cDNA has been deleted at a 5' region in a patient with nonketotic hyperglycinemia

One of eight patients with nonketotic hyperglycinemia resulted by the lesion in glycine decarboxylase showed the deletion of 0.6-kb SacI and 1.5-kb PstI fragments identified by the cDNA for this protein. A genomic clone, lambda HGDG10, encodes a 5' region of this cDNA in an organized structure and can produce these two fragments. The other clone, lambda HGDG8, carries a processed gene. Southern analysis using a limited segment of this cDNA demonstrated that the 1.7-kb and 1.5-kb PstI fragments predicted from its recognition sites in both genomic clones occur actually in the human genome, indicating that at least two copies of glycine decarboxylase cDNA exist in the haploid genome, and the patient has the glycine decarboxylase gene deleted at a 5' region.     

Restriction fragment length polymorphisms in bovine major histocompatibility complex class II ß-chain genes using bovine exon-containing hybridization probes

Restriction fragment length polymorphisms (RFLPs) have been identified in the bovine MHC class II region using five hybridization probes constructed from two bovine genomic clones. Four probes were constructed from a bovine DR beta-like gene, BoLA-DRB2. These included a probe containing the complete beta 1 exon (R2-beta 1), a probe containing the last 129 base pairs of the beta 2 exon (R2-beta 2), a probe containing intron immediately 5' of the beta 2 exon (R2-5' beta 2), and a probe containing the complete transmembrane exon (R2-TM). A fifth probe was constructed from a novel bovine beta-chain gene, BoLA-DIB, and contained the entire TM exon (I1-TM). R2-beta 1 defined very little polymorphism. R2-beta 2 hybridized to several fragments but one or two fragments hybridized much stronger on all Southern blots and it was presumed these corresponded to BoLA-DRB2 fragments. By using R2-5' beta 2 as a probe, these BoLA-DRB2 fragments were confirmed: 6.4 and 2.7-kb Eco RI alleles, 1.7- and 1.5-kb Pvu II alleles, 5.9-, 5.4-, 3.7- and 1.9-kb TaqI alleles, and a non-polymorphic 22.5-kb BamHI fragment. I1-TM identified three alleles with TaqI. To investigate the linkage between the RFLP alleles, 166 offspring of five sires were tested. Complete linkage was found for all RFLPs identified with the BoLA-DRB2 probes. However, the RFLP patterns of 13 calves out of 58 indicated recombination between BoLA-DRB2 and BoLA-DIB.     

A human metallothionein pseudogene containing AG/CT repetitive elements

Summary A lambda phage recombinant clone, 25 S, which contains a 15.5-kb EcoRI human genomic DNA fragment, has been characterized. Restriction mapping and Southern blot hybridization indicated a 3.0-kb HindIII fragment containing metallothionein (MT)-like sequences. Several interesting features were found upon comparison of this nucleotide sequence with that of other human MT genes: (1) sequences representing the 5 regulatory region, the 5 untranslated region, and the first exon are not contained in the 3.0-kb HindIII fragment; (2) the coding sequence of the second exon (amino acids 10–31 encoding a portion of the -domain of the MT protein) has 11 amino acid changes out of a total of 21, whereas, the third exon (amino acids 32–61, representing the complete -domain of the MT protein) has only 4 amino acid substitutions; however, all cysteine residues are conserved; (3) this MT-like gene retains intron sequences and processing signals; (4) Southern blot analysis of human genomic DNA indicated this MT-like gene is located on a 10.5-kb EcoRI genomic DNA fragment; and (5) unusual AG/CT-rich repetitive elements are located within the second intron and upstream of the second exon of this MT-like gene. This gene is not expressed in response to metal induction in two human cell lines, as shown by northern blot analyses. Based on these observations, this MT-like gene represents a unique nonprocessed pseudogene of the human MT multigene family.     

Isolation of the osteonectin gene: evidence that a variable region of the osteonectin molecule is encoded within one exon

A complementary DNA clone for bovine osteonectin was used to isolate the osteonectin gene from two libraries of bovine genomic DNA fragments. Two overlapping clones were obtained whose relationship was determined by restriction mapping and sequence analysis. The two clones contain the entire osteonectin coding region spanning approximately 11 kilobases of genomic DNA. The coding region of the gene was determined, by electron microscopy and DNA sequencing, to reside in nine exons. In addition, there is at least one 5' exon interrupted by an intron in the 5'-nontranslated sequence of the gene. Excluding this 5' exon and the 3'-terminal exon, the exons are small and approximately uniform in size, averaging 130 +/- 17 base pairs. Three of the exons at the 5' end of the gene were sequenced and appear to encode discrete protein domains. For example, the putative exon 2 contains the coding region for the leader peptide of the molecule. The amino-terminal protein sequence was determined for osteonectin extracted from human, rabbit, and chicken bone and compared with those for bovine, mouse, and pig osteonectin. These data suggest that osteonectin is highly conserved between species, interspecies changes being seen primarily at the amino terminus of the protein and specifically in the region encoded by putative exon 3 in the bovine gene.     

Mutation in the alpha 5(IV) collagen chain in juvenile-onset Alport syndrome without hearing loss or ocular lesions: detection by denaturing gradient gel electrophoresis of a PCR product.

A single base mutation was identified in the type IV collagen alpha 5 chain gene (COL4A5) of a Danish kindred with Alport syndrome. The 27-year-old male proband developed hematuria in childhood and terminal renal failure at the age of 25 years. He has no hearing loss or ocular lesions. Electron microscopy demonstrated splitting of the lamina densa of the glomerular basement membrane. The proband's mother has had persistent microscopic hematuria since the age of 40 years, but no other manifestations. Southern analysis of MspI-digested genomic DNA from the proband showed the absence of 1.3-kb and 0.9-kb fragments present in control DNA but the presence of a 2.2-kb variant fragment, indicating the loss of an MspI restriction site in the 3' end of the gene. The mother had all three fragments, indicating heterozygosity. PCR amplification of exon 14 (counted from the 3' end) and subsequent denaturing gradient gel electrophoresis analysis suggested a sequence variant in the proband and his mother. This was confirmed by sequencing of the PCR-amplified exon 14 region of the hemizygous proband, which demonstrated the base change G----A abolishing an MspI restriction site. Hybridization analysis with allele-specific probes confirmed the inheritance of the mutation with the phenotype. The mutation changed the GGC codon for glycine-1143 to GAC for aspartate. Substitution of glycine-1143, located in the collagenous domain of the alpha 5(IV) chain, for any other amino acid can be expected to interfere with the maintenance of the triple-helical conformation of the collagen molecule. This could, in turn, weaken the glomerular-basement-membrane framework and lead to increased permeability.     

Cloning and nucleotide sequence of the murine interleukin-3 gene

Southern hybridization analysis using a probe derived from a murine interleukin-3 (IL-3) cDNA clone revealed the presence of a single IL-3 gene in the haploid murine genome. An 8600-base-pair (8.6-kb) murine genomic EcoRI fragment containing the IL-3 gene was isolated by screening a library of size-fractionated genomic EcoRI fragments cloned in lambda gtWES X lambda B. The nucleotide sequence of a 3.5-kb region of the cloned DNA encompassing the IL-3 gene was determined. The gene contains four introns of 96, 993, 135 and 122 base pairs (bp), located within the coding region. The large intron contains 12 copies of a 14-15-bp tandem repeating sequence which resembles a human cellular homologue of a BKV enhancer sequence. The nucleotide sequence of the exons agrees exactly with that of an IL-3 cDNA cloned from WEHI-3, a tumorigenic cell line which over-produces IL-3, establishing that the unprocessed primary structure of IL-3 is identical in WEHI-3 and in BALB/c mice. Southern hybridization has revealed genomic alteration in the vicinity of the IL-3 gene in WEHI-3 cells.     

Human acyl-CoA:cholesterol acyltransferase-1 (ACAT-1) gene organization and evidence that the 4.3-kilobase ACAT-1 mRNA is produced from two different chromosomes

Acyl-CoA:cholesterol acyltransferase (ACAT) plays important roles in cellular cholesterol homeostasis. Four human ACAT-1 mRNAs (7.0, 4.3, 3.6, and 2.8 kilobases (kb)) share the same short 5'-untranslated region (exon 1) and coding sequence (exons 2-15). The 4.3-kb mRNA contains an additional 5'-untranslated region (1289 nucleotides in length; exons Xa and Xb) immediately upstream from the exon 1 sequence. One ACAT-1 genomic DNA insert covers exons 1-16 and a promoter (the P1 promoter). A separate insert covers exon Xa (1277 base pairs) and a different promoter (the P7 promoter). Gene mapping shows that exons 1-16 and the P1 promoter sequences are located in chromosome 1, while exon Xa and the P7 promoter sequence are located in chromosome 7. RNase protection assays demonstrate three different protected fragments, corresponding to the 4.3-kb mRNA and the two other mRNAs transcribed from the two promoters. These results are consistent with the interpretation that the 4.3-kb mRNA is produced from two different chromosomes, by a novel RNA recombination mechanism involving trans-splicing of two discontinuous precursor RNAs.