Organization and structure of the human tissue transglutaminase gene.

Analysis of the genomic organization of tissue transglutaminase shows that the gene is 32.5 kilobases, contains 13 exons and 12 introns. Our results show that the sites for the two alternative splicing forms of tissue transglutaminase we reported earlier are located within exons 6 and 10 respectively. The 5'-upstream region of the gene has several potential regulatory promoter elements, and the 3'-exon contains about 50% of the total cDNA size and codes for the C-terminus of the protein. Alignment of deduced tissue transglutaminase amino acid sequence with other transglutaminases showed very similar intron splice positions.

A third human tissue transglutaminase homologue as a result of alternative gene transcripts.

A 2.4 kilobase (kb) cDNA encoding a new form of human tissue transglutaminase homologue (TGH2) was isolated from retinoic acid-induced human erythroleukemia cell (HEL) library. Full-length cDNA analysis gives an open reading frame coding for a polypeptide of 349 amino acid residues with a molecular mass of 38,700 Da. This variant differs from the previously reported homologue TGH in that it is 199 amino acids shorter and has an alternative, 63 amino acid COOH-terminal peptide. The 3'-untranslated region of the cDNA also differs from the previously reported sequences for both TGH and human tissue transglutaminase. The region coding for the first 286 amino acids of TGH2, which contains the active site is identical to TGH. Immunoprecipitation of the in vitro translation product from a synthetic TGH2 mRNA and immunoprecipitation of total protein of human heart, liver, kidney and cultured erythroleukemia HEL cell, revealed a protein with a molecular mass of 37,000 Da by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Comparison of the cDNA sequence for the previously known tissue transglutaminases with genomic DNA and the TGH2 cDNA described here indicate that the sequence divergence points correlate with known intron-exon boundaries. The smaller RNA species encode for truncated proteins with novel carboxyl termini. The TGH cDNA and the TGH2 cDNA both produce transcripts which start with the regular coding sequence for TGase and then fail to splice at specific donor sites, resulting in the use of an alternative exon that contains a stop codon.

GTP hydrolysis by human tissue transglutaminase homologue.

Human tissue transglutaminase homologue cDNA was expressed in E. coli to analyze the catalytic characteristics. The transglutaminase homologue was purified by immunoaffinity chromatography. Specificity of GTP binding by the homologue was demonstrated by photoaffinity labeling in the absence or presence of GTP-gamma-S. The homologue had GTPase activity with an apparent Km value of 1.8 microM, several-fold lower than the reported Km values for the native tissue transglutaminase. GTPase activity was inhibited by guanine nucleotides in order GTP-gamma-S > GDP > GMP. The higher GTPase activity of the homologue may be related to the signaling events function.

Site-directed mutagenesis of human tissue transglutaminase: Cys-277 is essential for transglutaminase activity but not for GTPase activity.

Transglutaminases (EC 2.3.2.13) catalyze an acyl-transfer reaction between peptide-bound glutamine residues and primary amines, including the epsilon-amino group of lysine residues in protein. Purified human erythrocyte transglutaminase was found to have another activity, i.e., GTP hydrolysis. Treatment of the enzyme with iodoacetamide, a cysteine-directed reagent, caused a 94% loss of TGase activity within 8 min, but no significant loss of GTPase activity. Cys-277, a known residue which is selectively modified by iodoacetamide, was replaced with Ser by site-directed mutagenesis to assess the role of the Cys-277 in the transglutaminase/GTPase activities. Wild-type cDNA, coding for human endothelial cell transglutaminase, and its C277S-mutated cDNA were cloned into a plasmid vector that contained a promoter from phage T7, and then expressed in Escherichia coli. The wild-type recombinant enzyme was indistinguishable from human erythrocyte transglutaminase in mobility on a SDS-polyacrylamide gel, immunoreactivity and catalytic activities for transglutaminase and GTPase. However, the recombinant enzyme was not blocked at the N-terminal alanine residue, as is the case in the naturally occurring erythrocyte enzyme. The C277S mutant enzyme showed no transglutaminase activity, but had Km and kcat values for GTPase activity that were comparable to those of wild-type recombinant and natural erythrocyte enzymes. These results demonstrate that Cys-277 is essential for transglutaminase activity, but not for GTPase activity, and that N-terminal blocking of tissue-type transglutaminase is not critical for either transglutaminase or GTPase activities.

A retinoic acid-inducible mRNA from human erythroleukemia cells encodes a novel tissue transglutaminase homologue.

A 1.9-kilobase (kb) cDNA for a new transglutaminase protein has been cloned and sequenced from retinoic acid-induced human erythroleukemia (HEL) cells. Full-length cDNA analysis reveals an open reading frame coding for a polypeptide of 548 amino acid residues with a molecular weight of 61,740. The deduced amino acid sequence exhibited 98% identity to the human cellular transglutaminase sequence. The cysteine at position 277 in the active site and the putative Ca(2+)-binding pocket at residues 446-453 of cellular transglutaminase are conserved. Such evidence predicts that the encoded protein product is likely to be a transglutaminase homologue (TGase-H). Immunoprecipitation of the in vitro translation products from a synthetic TGase-H mRNA and from total protein of cultured erythroleukemia HEL cells revealed a protein with a molecular weight of 63,000 by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Northern blot analysis of HEL cells and normal human fibroblast cells WI-38 using a cellular TGase probe detected the 1.9- and 4.0-kb RNA species at a relative abundance of 1:3 and 1:7, respectively. The 3'-end of the human cellular transglutaminase mRNA was also cloned and sequenced to allow comparison to the 3'-end of TGase-H reported here. This new piece gives a full length of 4012 nucleotides (4.0 kb) for human cellular transglutaminase. Comparison of the 5'-end (bases 1-1747) of the 1.9- and 4.0-kb cDNA sequences revealed a very high degree of identity. Beginning with base 1748, the sequences diverge showing no homology. The divergence point correlates with known intron-exon consensus boundaries indicative of alternative splicing.

Separation and identification of urinary proteins and stone-matrix proteins by mini-slab sodium dodecyl sulfate-polyacrylamide gel electrophoresis.

Mini-slab electrophoresis on sodium dodecyl sulfate/polyacrylamide gel can be used to detect protein in unconcentrated urines containing more than 100 mg of protein per liter. Several urinary proteins--including albumin, transferrin, and the Tamm-Horsfall mucoprotein--precipitate, partly or totally, if urine samples are stored at 4 degrees C for one or two days. The present procedure permits identification of the albumin and transferrin in unconcentrated urines, thus avoiding their partial loss. Stone-matrix proteins in 10 mg or more of urinary calculi also could be detected. Albumin, transferrin, and Tamm-Horsfall mucoprotein accounted for greater than 40% of the total matrix protein in most of the urinary stones analyzed. The precipitation properties of these proteins and their high concentrations in stone matrix relative to those in urine suggest that they may play a role in stone formation and growth.

Transferrin and albumin excretion as a measure of glomerular function.

Albumin and transferrin are relatively small protein molecules and highly negatively charged. Their levels in urine are a useful indicator of the integrity of membrane barriers of the kidney glomerular capillary wall. The present data shows that the excretion rates of albumin and transferrin and their kinetics of excretions are similar. Thus, their filtration mechanisms at the active site of the kidney membrane pores are similar. Total urinary protein/creatinine or albumin or transferrin/creatinine ratio were found to be linear and highly significant. Their measurement could indicate the degree of impaired glomerular permeability. Also, in the present study, a rapid biochemical method of measurement of the selectivity of proteinuria based on the transferrin/albumin ratios in random samples is reported. This method is particularly useful in the early stages of glomerular basement membrane damage.