Identification of conserved promoter elements for aldB and isozyme specific residues in aldolase B.

The comparison of three complete aldolase B genes-including known and putative regulatory elements-is presented. The third aldolase B gene was provided by the complete aldB gene sequence (14803 bp) encoding the rabbit aldolase B isozyme. The promoter sequence alignment included the nonmammalian chicken aldolase B gene and confirms the promoter sequence conservation of those elements where trans-factor binding has been demonstrated in rat aldB. Moreover, the alignment reveals conserved sequences that may represent previously unidentified promoter elements that are present in all aldBs or specifically in the mammalian aldB promoters. One remarkable feature is a poly-purine segment found between the CAAT and TATA elements. In the mammalian promoters, this is exclusively a 9-10 bp poly-dA stretch. The avian promoter has an additional stretch of eight dG-bases immediately upstream of the poly-dA. Alignment of a portion of intron 1 of the chicken, human, and rabbit aldB genes reveals conserved sequences that are likely candidates for a reported positive activation sequence. In addition, the amino acid sequences of all eight known aldolase B isozymes is compared to the other vertebrate aldolases. A number of aldolase B-specific residues are identified that cluster in the carboxyl-portion of the sequence. With the exception of residue C268, these residues are not found near the active site, although, they are likely to be responsible for the substrate specificity of aldolase B.

A rabbit AldA pseudogene derived from a partially spliced primary aldolase A transcript.

The entire AldA processed pseudogene of rabbit was isolated and characterized. The pseudogene encodes the C-terminal portion of the protein from amino acids (aa) 126-363. There are deletions, insertions and nucleotide (nt) substitutions distributed throughout the 931 bp of identity shared with the 1.4-kb mRNA. There are 21 replacement codon substitutions, including a clearly deleterious change in the stop codon. This processed pseudogene has several uncommon features: (i) it has a 5'-boundary coincident with an intron/exon junction and does not encode the entire mRNA, (ii) there is a broken direct repeat that overlaps the region of shared identity with the mRNA rather than flanking it, and (iii) there is no poly(A) sequence. This processed pseudogene probably arose by integration of a DNA copy of a partially spliced primary transcript. The structure of this gene has added implications for the timing of posttranscriptional processing events.

Purification, structure, and properties of Escherichia coli tRNA pseudouridine synthase I.

The RNA modification enzyme, tRNA pseudouridine synthase I has been isolated in 95% purity from an Escherichia coli strain harboring a multicopy plasmid with a 2.3-kilobase pair insert from the hisT operon. Its molecular size, amino acid composition, and amino-terminal sequence correspond to those predicted by the structure and expression of the hisT gene. Enzyme activity, as measured by a 3H release assay, is unaffected by pretreatment of tRNA pseudouridine synthase I with micrococcal nuclease and is optimized by the addition of a monovalent cation and thiol reductant. The activity is inhibited by all tRNA species tested, including substrates, modified tRNAs, nonsubstrates, or tRNAs containing 5-fluorouridine. Binding of tRNA pseudouridine synthase I occurs with both substrate and nonsubstrate tRNAs and does not require a monovalent cation. Our findings are consistent with a multistep mechanism whereby tRNA pseudouridine synthase I first binds nonspecifically and then forms transient covalent adducts with tRNA substrates. In the absence of other proteins, purified tRNA pseudouridine synthase I forms psi at all three modification sites known to be affected in hisT mutants. The 36.4-kDa polypeptide product of the gene adjacent to hisT, whose translation is linked to that of tRNA pseudouridine synthase I, is not a functional subunit for tRNA pseudouridine synthase I activity, nor is it a separate synthase acting at one of the three loci.

Characterization of the human aldolase B gene.

The structure of the human gene encoding the aldolase B isozyme has been determined, including the sequence of 14,887 base-pairs. The 5'- and 3'-ends have been determined by S1 mapping. There is a single gene for this enzyme in humans that was determined from the sequence and restriction enzyme digestions of genomic DNA. The gene is 14,500 base-pairs long containing nine exons. In addition, 924 and 208 base-pairs of the 5'- and 3'-flanking region, respectively, have been determined. There is a high degree of conservation of nucleic acid sequence between aldolase B genes of human, rat and chicken. The conservation extends to untranslated and flanking regions, and includes the derived protein structures. In the 5'-flanking region there are several sequence elements that are conserved in vertebrate aldolase B genes in addition to the T-A-T-A and C-C-A-A-T boxes. These sequences may be involved in the co-ordinate and tissue-specific control of expression of this gene. Several possible polymorphic sites were detected in the sequence of the human gene which may be useful for linkage mapping of the human genome and diagnostic analysis of alleles in families with hereditary fructose intolerance.

Structural features of the hisT operon of Escherichia coli K-12.

The DNA sequence of a 2,3-kilobase segment of the E. coli hisT operon was determined. Analysis of the sequence indicated that the upstream gene in the operon encodes a 36,364-dalton polypeptide, which runs aberrantly on SDS-polyacrylamide gels. The distal hisT gene encodes the tRNA modification enzyme, pseudouridine synthase I, which was shown to have a polypeptide molecular mass of 30,399 daltons. The DNA sequence was consistent with the phenotypes and hisT expression of mutant operons. Analysis of the sequence and genetic complementation experiments demonstrated that the upstream and hisT genes are evolutionarily, structurally, and functionally unrelated; however, translation signals for the two genes overlap, which is consistent with genetic evidence suggesting translational coupling. Codon usage in the upstream gene is radically different from the hisT gene and may underlie the differential expression observed from the operon. Gene-inactivation experiments and S1-mapping of in vivo transcripts indicated that the operon contains an additional upstream gene. S1-mapping experiments also confirmed the presence of an internal promoter, which might be stringently controlled. Taken together, these results show that the structure of the hisT operon is complex and suggest that the operon might be regulated at several levels.

Characterization of the chicken aldolase B gene.

Vertebrates possess three isozyme forms of fructose diphosphate aldolase. We have isolated two overlapping chicken genomic clones which encode the liver-specific form of this enzyme; we have identified the 5' and 3' ends of this gene by a combination of primer extension analysis and S1 mapping; and we have determined the entire nucleotide sequence of this gene including 1400 base pairs (bp) of sequence from the regions flanking the 5' and 3' ends of the gene. The transcriptional unit for the aldolase B gene spans 8700 bp and contains eight intervening sequences, including a 4600-bp intron in the 5' non-coding region. On the basis of results from Southern genomic hybridizations, the aldolase B gene appears to be present only once per haploid genome. No differences were detected in the mRNA structure between RNA from three tissues expressing aldolase B (liver, kidney, and small intestine). Various features of the 5' flanking region are discussed, including a partial homology with the 5' noncoding region from rabbit aldolase A.

hisT is part of a multigene operon in Escherichia coli K-12.

The Escherichia coli K-12 hisT gene has been cloned, and its organization and expression have been analyzed on multicopy plasmids. The hisT gene, which encodes tRNA pseudouridine synthase I (PSUI), was isolated on a Clarke-Carbon plasmid known to contain the purF gene. The presence of the hisT gene on this plasmid was suggested by its ability to restore both production of PSUI enzymatic activity and suppression of amber mutations in a hisT mutant strain. A 2.3-kilobase HindIII-ClaI restriction fragment containing the hisT gene was subcloned into plasmid pBR322, and the resulting plasmid (designated psi 300) was mapped with restriction enzymes. Complementation analysis with different kinds of hisT mutations and tRNA structural analysis confirmed that plasmid psi 300 contained the hisT structural gene. Enzyme assays showed that plasmid psi 300 overproduced PSUI activity by ca. 20-fold compared with the wild-type level. Subclones containing restriction fragments from plasmid psi 300 inserted downstream from the lac promoter established that the hisT gene is oriented from the HindIII site toward the ClaI site. Other subclones and derivatives of plasmid psi 300 containing insertion or deletion mutations were constructed and assayed for production of PSUI activity and production of proteins in minicells. These experiments showed that: (i) the proximal 1.3-kilobase HindIII-BssHII restriction fragment contains a promoter for the hisT gene and encodes a 45,000-dalton polypeptide that is not PSUI; (ii) the distal 1.0-kilobase BssHII-ClaI restriction fragment encodes the 31,000-dalton PSUI polypeptide; (iii) the 45,000-dalton polypeptide is synthesized in an approximately eightfold excess compared with PSUI; and (iv) synthesis of the two polypeptides is coupled, suggesting that the two genes are part of an operon. Insertion of mini-Mu d1 (lac Km) phage into plasmid psi 300 confirmed that the hisT gene is the downstream gene in the operon.

Complete amino acid sequence for human aldolase B derived from cDNA and genomic clones.

Several aldolase B clones from a human liver cDNA library have been identified by using a rabbit aldolase A cDNA as a hybridization probe. The most complete of these, pHL413, is 1389 base pairs long and covers approximately equal to 80% of the length of the mRNA, including 90% of the translated region. The cDNA, pHL413, was used to identify a genomic clone, lambda HG313, which encoded the remaining amino acids of human aldolase B. We demonstrate that the amino acid and nucleotide sequences of aldolase are strongly conserved even between different isozymes. Furthermore, in the 3'-untranslated regions of the mRNAs for the B isozyme of human and rat there is an extensive stretch of homology. Aldolase B lacks a cysteine at positions 72 and 338 and lacks a histidine at position 361. These residues, which are present in rabbit aldolase A, have previously been proposed to take part in catalysis. Our findings suggest that this may not be the case.

The complete nucleotide sequence for rabbit muscle aldolase A messenger RNA.

The complete nucleotide sequence of rabbit muscle aldolase mRNA has been determined from recombinant cDNA clones and from a primer-extended cDNA synthesis on the mRNA template. The sequence is composed of 1375 nucleotides, exclusive of the poly(A) tails. The 5'-untranslated region contains 62 bases including a potential ribosome-binding site. The 3'-untranslated region is 221 bases long. The remaining 1092 nucleotides are in an open reading frame coding for 364 amino acids. There is a single methionine residue preceding the NH2-terminal proline. The deduced amino acid sequence corrects a number of discrepancies between published structures as well as assignments previously missed. The sequences of the cloned cDNAs suggests there may be microheterogeneity in the messenger RNA population.

Assembly of vesicular stomatitis virus: distribution of the glycoprotein on the surface of infected cells.

This study demonstrates that the glycoprotein of vesicular stomatitis virus clusters in the plasma membrane of infected Chinese hamster lung cells during morphogenesis and suggests that viral nucleocapsids are required for this clustering. A mutant virus (ts E-1) which is temperature sensitive for the synthesis of viral nucleocapsids but not viral membrane proteins was used. The surface distribution of the viral glycoprotein in cells infected by this virus was determined by a specific indirect immunoferritin stain. Early in infection at permissive temperatures, the glycoprotein was randomly distributed on membrane ghosts. Later, clusters of ferritin the size and shape of virus particles were seen. In contrast, ghosts prepared from virus-infected cells maintained at a restrictive temperature always had a random distribution of viral glycoprotein.

Purification and properties of a mammalian tRNA pseudouridine synthase.

A tRNA pseudouridine synthase has been extensively purified from steer thymus extracts, using undermodified tRNA from hisT- mutants of Salmonella typhimurium as a substrate. The enzyme synthesizes a group of psi residues in the anticodon region of various hisT- isoacceptors and behaves like a eukaryotic homologue of Salmonella tRNA psi synthase I. The thymus enzyme requires a thiol and a monovalent cation (NH4+ or K+) for optimum activity; no energy sources or cofactors are required. The activity is inhibited by single tRNAs or bulk tRNA from all sources tested, and by ribosomal RNAs, various polyribonucleotides, and DNA. The enzyme modifies the two hisT- tRNAPhe isoacceptors, both tRNATyr acceptors and at least five of the tRNALeu isoacceptors to products that coelute with the respective wild type species on reverse-phase columns. With pure hisT- tRNA2Phe as substrate, the enzyme specifically converts residue U39 to psi. Interestingly, a psi residue is still present at position 32, in the anticodon loop of hisT- tRNA2Phe, indicating the existence of other uncharacterized pseudouridylation enzymes in S. typhimurium. These composite results show that the thymus enzyme can form psi at residues 38, 39, and 40 in the anticodon region of appropriate hisT- isoacceptors. During the enzyme purification, a second activity is partially resolved, which releases 3H from wild type S. typhimurium [pyrimidine-5-3H]tRNA. This activity may be associated with an enzyme that pseudouridylates sites that are uniquely modified in eukaryotic tRNAs, but not in Salmonella tRNAs. Our observations support the view that the psi residues in tRNA are synthesized by a family of enzymes, whose members act on uridine residues in specific regions of the molecule.

The structure and synthesis of influenza virus phosphoproteins.

The synthesis and phosphorylation of influenza virus nucleoprotein and nonstructural protein were analyzed. The nucleoprotein (NP) was found to be phosphorylated in both infected cells and in isolated virions. The phosphate is in a monoester linkage to a serine residue. Two-dimensional tryptic peptide maps of the 32P-labeled protein, as well as measurements of specific activity, suggests that NP is phosphorylated at one site per molecule. The viral nonstructural (NS 1) protein is also phosphorylated, but on threonine residues. Up to a maximum of two sites per NS 1 molecule could be so modified in infected cells, as demonstrated by two different methods of tryptic peptide analysis and by measurements of the ratio of 32P to 3H-amino-acids incorporated into NS 1 protein species. The NS 1 protein is resolved into four major species of differing isoelectric point in a two-dimensional electrophoretogram. The most acidic species was found to have two phosphorylated sites per molecule, and the next most acidic species contained on the average one phosphate per molecule. Treatment of the phosphorylated species with bacterial alkaline phosphatase demonstrated that the level of phosphorylation is the only identifiable difference between the phosphorylated and unphosphorylated NS 1 species. The distinction between the two unphosphorylated species could not be determined. The distribution of the un-, mono-, and diphosphorylated NS 1 species was characterized at different times after synthesis. These modifications were found to occur very rapidly after translation (30 to 60 s), after transport of the unmodified species from cytoplasm to nucleus of the infected cell. The phosphorylation of NP also takes place rapidly after its synthesis; the site within the cell of the NP phosphorylation has not been unambiguously determined.

Biochemical characterization of the tsE1 mutant of vesicular stomatitis virus (New Jersey). Alterations in the NS protein.

Alterations in the NS protein of the tsE1 mutant of vesicular stomatitis virus (New Jersey serotype) appear to be responsible for its temperature-sensitive phenotype. The NS proteins of thermostable revertants of tsE1 migrated in polyacrylamide gels containing sodium dodecyl sulfate with apparent sizes which were identical to tsE1 NS, or which were 5% or 14% larger than tsE1 NS. These novel differences persisted during electrophoresis in 10% and 12.5% acrylamide gels, and in gels with gradients of acrylamide, suggesting that aberrant sodium dodecyl sulfate binding was not involved. Co-infection of cells with pairs of viruses resulted in the synthesis of both types of NS protein, suggesting that no trans-acting phenomenon was involved. Two-dimensional gel electrophoresis demonstrated that each of the NS proteins consisted of several species, but the isoelectric points of the proteins from different viruses overlapped. Furthermore, all of the NS species from a particular virus migrated with the same apparent molecular weight, suggesting that aberrant phosphorylation was not responsible for the apparent differences in size. Finally, tryptic peptide maps of amino acid and 32Pi-labeled NS proteins demonstrated that the revertant NS proteins contained all of the peptides and phosphopeptides of tsE1 NS, but each revertant NS with an apparently larger protein also contained an extra nonphosphorylated peptide. These data are consistent with the idea that the reversion of the temperature-sensitive phenotype of tsE1 can be accompanied by production of a significantly larger NS protein.

Quantification of influenza virus messenger RNAs.

The amounts of specific influenza virus cRNAs present on polyribosomes of cells treated with cycloheximide at various times after infection were measured by quantitative hybridization with 125I-labelled virus RNA segments. All species of cRNA were found associated with polyribosomes at every time point analysed. The relative abundance of the specific RNAs was in the same order as the reported relative synthesis of the influenza virus proteins. However, the concentrations of the cRNAs spanned a considerably smaller range than published protein synthetic rates, suggesting that both transcriptional and translational controls operate to regulate the final levels of influenza virus polypeptide synthesis.

Survival of apurinic SV40 DNA in the d-complementation group of xeroderma pigmentosum.

The survival of depurinated Form I SV40 DNA was studied in normal human fibroblasts and in D-complementation Xeroderma pigmentosum (XP) fibroblasts. Survival was measured with an infective center assay. Heat-acid and methyl methanesulfonate (MMS) were used as depurinating agents. After 3 hrs of depurination by heat--acid treatment, infectivity in normal cells was less than 15% of the controls compared to more than 50% for the XP D cell strains. Similar results were obtained with MMS-treated DNA. These results are contrary to expectation since apurinic endonuclease activity, which is presumed to be involved in the repair of apurinic sites, is much lower in XP D cell strains than in normal cell strains. Our results indicate that another mechanism for the repair of apurinic sites could exist.

Viral gene expression in murine sarcoma virus(murine leukemia virus)-infected cells.

NIH 3T3 cells infected with Moloney murine sarcoma virus (murine leukemia virus) produce virions which contain about 99% murine sarcoma virus RNA and 1% murine leukemia virus RNA. This report describes experiments which measured intracellular concentrations of proviral DNA and RNA transcripts for each of the viruses. We found that three to four copies of proviral DNA from each virus were integrated into the cellular DNA. Measurements of RNA specific for each of the genomes by hybridization to specific cDNA reagents revealed a 10- to 15-fold difference in concentration in both nuclear and polysomal RNA fractions, with murine sarcoma virus RNA predominating in both cases. Unless there are major differences in stability between the two viral RNAs, our results suggest that transcriptional control is responsible for much of the difference in final levels of virus synthesis.

Influenza virus proteins: identity, synthesis, and modification analyzed by two-dimensional gel electrophoresis.

A modification of the two-dimensional protein electrophoresis system of O'Farrell was used to resolve influenza A virus proteins from each other and from host proteins in infected cells. Viral protein spots corresponding to the hemagglutinin proteins, neuraminidase, nucleocapsid protein, and nonstructural protein, were identified on the two-dimensional electrophoretogram. Use of the two-dimensional separation has allowed us to identify glycoprotein heterogeneity, to demonstrate directly the synthesis of neuraminidase, to analyze some viral proteins early after infection, and to demonstrate that the influenza virus NP and NS proteins are phosphorylated in infected MDCK cells.

Xeroderma pigmentosum fibroblasts of the D group lack an apurinic DNA endonuclease species with a low apparent Km.

Apurinic DNA endonuclease activity from cultured human fibroblasts was resolved into two species by phosphocellulose chromatography. The species had sedimentation coefficients of 3.3 S and 2.8 S and apparent Km's for apurinic sites of 5 and 44 nM, respectively. The low Km species was absent from extracts of cell lines XP5BE, XP6BE and XP7BE of xeroderma pigmentosum complementation group D.

Cellular RNA synthesis in normal and mengovirus-infected L-929 cells.

Cellular RNA synthesis was studied in mouse L-929 cells and in these cells infected with mengovirus. RNA polymerases I, II, and III were partially purified and their chromatographic properties were analyzed by DEAE-Sephadex A-25 chromatography. RNA polymerase II was purified from mouse liver and its subunit structure was compared to that of normal and virus-infected L-929 cells by two-dimensional gel electrophoresis. By these criteria, the enzymes from all three sources were identical. The RNA synthetic activities and capacities of chromatins from normal and virus-infected cells were compared under a variety of conditions. The endogenous activity in chromatin from infected cells was inhibited relative to controls but the residual activity responded normally to stimulation by ammonium sulfate, heparin, and Sarkosyl. The template capacity of the chromatins was compared with added RNA polymerase II and by a rifampicin challenge assay utilizing Escherichia coli RNA polymerase. Identical results were obtained in each case. The number of growing RNA chains and the rates of their elongations were determined. The results showed that nuclei and chromatin from infected cells have a smaller number of RNA polymerase II molecules engaged in RNA synthesis than normal cells do but that the active molecules elongate RNA chains at the same rate.