Partial purification and characterization of mRNA guanylyltransferase from Saccharomyces cerevisiae.
Explore the source record for details and available documents.
Biomedical subjects
Publications and source records attributed to Y Kaziro.
Explore the source record for details and available documents.
An RNA 5'-triphosphatase activity hydrolyzing gamma-phosphate from pppN-RNA was found to be associated with mRNA guanylyltransferase partially purified from rat liver nuclei. The activity specifically removed 32P as inorganic phosphate from [gamma-32P]pppA(pA)n, but not from [beta-32P]pppA(pA)n or from [gamma-32P]ATP. Free SH group(s) were required for its activity, and the reaction was inhibited by N-ethylmaleimide. Divalent cations were not required, but were rather inhibitory for the reaction. The RNA 5'-triphosphatase activity could not be separated from the guanylyltransferase activity through successive chromatographies on Sephadex G-150, CM-Sephadex and blue dextran-Sepharose columns. Both activities remained physically associated during sedimentation in glycerol density gradients after high salt treatment. The heat stability of the RNA 5'-triphosphatase activity was almost identical with that of the guanylyltransferase activity. These results indicate that the 69000 mol. wt. protein purified from rat liver nuclei as guanylyltransferase possesses both mRNA capping and RNA 5'-triphosphatase activities.
The mRNA capping reaction catalyzed by rat liver mRNA guanylyltransferase proceeds through an enzyme-GMP intermediate in which GMP is linked to the enzyme by a phosphoamide linkage. The studies described here show that GMP is bound to the epsilon-amino group of lysine of rat liver guanylyltransferase. The enzyme-[32P]GMP intermediate was digested with pronase to a [32P]GMP-peptide which was then converted to [32P]phosphoryl-peptide through periodate oxidation followed by beta-elimination. After alkaline hydrolysis of the [32P]phosphoryl-peptide, the major radioactive product co-electrophoresed with the authentic N epsilon-phospholysine on DEAE-cellulose paper. Neither [32P]Nimid-phosphohistidine nor Nguanido-phosphoarginine was detected in the hydrolysates. Furthermore, formation of N epsilon-guanylyl-lysine linkage on the enzyme was more directly shown by isolation of [32P]GMP(5' leads to N epsilon)lysine when the steps of periodate oxidation and beta-elimination were omitted. The results indicate that the nucleophile in the guanylyltransferase to which the guanylyl residue is linked is the epsilon-amino group of a lysine residue. [32P]Phosphoryl-lysine was also isolated from the vaccinia virus capping enzyme-[32P]GMP intermediate. Guanylyltransferase from HeLa cells, wheat germ, Artemia salina and yeast also formed the enzyme-GMP complex and, from the stability of the complex, the linkage between the enzyme and GMP was suggested to be a phosphoamide.
We have studied the effect of guanosine-5'-diphosphate-3'-diphosphate (ppGpp) on the transcription of the E. coli tufB and recA operons in a cell-free system containing of purified RNA polymerase holoenzyme. The transcription of the tufB operon which is under stringent control, was markedly inhibited by 0.5 mM ppGpp, and the extent of this inhibition was found to be greatly influenced by the Mg2+ and K+ concentrations in the reaction mixture. Maximal inhibition was obtained in the presence of 2 mM Mg2+ and 80-120 mM K+, whereas at higher concentrations of Mg2+ or lower concentrations of K+, practically no inhibition was observed. In contrast, transcription of the recA operon which is not subject to stringent control, was little affected by ppGpp at any of Mg2+ and K+ concentrations tested. The nucleotide inhibited initiation of transcription of tufB, while the rate of RNA chain elongation was not greatly inhibited in the presence of ppGpp.
We have constructed recombinant plasmids that direct the synthesis of the Mr 19 000 protein encoded by the adenovirus type 12 (Ad12) E1b region as either a native protein or a protein fused to the amino-terminal portion of the elongation factor EF-TuB in Escherichia coli cells. Using these recombinants, we could synthesize a large amount of the fused protein, while only a small amount of the native Mr 19 000 protein was produced. The failure to synthesize the native Mr 19 000 protein in E. coli cells was ascribed to inefficient translation.
A 3.1-kilobase Bgl II fragment of Saccharomyces cerevisiae carrying the nuclear gene encoding the mitochondrial polypeptide chain elongation factor (EF) Tu has been cloned on pBR327 to yield a chimeric plasmid pYYB. The identification of the gene designated as tufM was based on the cross-hybridization with the Escherichia coli tufB gene, under low stringency conditions. The complete nucleotide sequence of the yeast tufM gene was established together with its 5'- and 3'-flanking regions. The sequence contained 1,311 nucleotides coding for a protein of 437 amino acids with a calculated Mr of 47,980. The nucleotide sequence and the deduced amino acid sequence of tufM were 60% and 66% homologous, respectively, to the corresponding sequences of E. coli tufA, when aligned to obtain the maximal homology. Plasmid YRpYB was then constructed by cloning the 2.5-kilobase EcoRI fragment of pYYB carrying tufM into a yeast cloning vector YRp-7. A mRNA hybridizable with tufM was isolated from the total mRNA of S. cerevisiae D13-1A transformed with YRpYB and translated in the reticulocyte lysate. The mRNA could direct the synthesis of a protein with Mr 48,000, which was immunoprecipitated with an anti-E. coli EF-Tu antibody but not with an antibody against yeast cytoplasmic EF-1 alpha. The results indicate that the tufM gene is a nuclear gene coding for the yeast mitochondrial EF-Tu.
The 3.6-kilobase Bgl II-EcoRI fragment from R1 plasmid containing copA, repA, and the replication origin (ori) was inserted into the ColE1-type plasmid pUC8. The resulting hybrid plasmid replicates in extracts prepared from both polA- and polA+ cells, whereas pUC8 replicates only in a polA+ extract. This characteristic provides a method for assaying the repA and ori functions. Hybrid plasmids that were either repA- or ori- were unable to replicate in a polA- cell extract. Replication of the repA- ori+ plasmid was restored by complementation of the repA defect by a repA+ ori- plasmid in vitro. Successful complementation of the repA function in vitro provides a method for assaying the repA protein. In order to define the minimum DNA segment with origin function (oriR), deletions were introduced starting from either side of the insert, and the replication properties of the plasmids carrying these deletions were examined in a polA- cell extract. The right end of oriR was located at position 1,611 in the nucleotide coordinates defined previously [Ryder, T., Rosen, J., Armstrong, K., Davidson, D. & Ohtsubo, E. (1981) in The Initiation of DNA Replication: ICN-UCLA Symposia on Molecular and Cellular Biology, ed. Ray, D.S. (Academic, New York), Vol. 22, 91-111]. By complementing repA- ori+ plasmids with the repA+ ori- plasmid, the left end of oriR was localized at position 1,424. Therefore, the oriR sequence, localized within a region of 188 base pairs, is separate from the repA gene. A hybrid plasmid carrying the 206-base-pair segment between positions 1,406 and 1,611 also replicates in a polA- cell extract when the repA function is supplied in trans. Removal of an additional 66 base pairs (positions 1,406-1,471) inactivates the function of the minimal oriR segment.
Brain microtubules purified by cycles of assembly and disassembly contained an ATPase activity in the fraction of microtubule-associated proteins (MAPs). This ATPase activity was found to be stimulated by 6S tubulin in the presence of Ca2+ ions, suggesting its functional association with brain microtubules (Ihara et al. (1979) J. Biochem. 86, 587-590). On further purification by DEAE-cellulose column chromatography, two peaks of ATPase activity were separated; one, eluted at 0.2 M KCl (ATPase I), was dependent on added 6S tubulin but the other, eluted at 0.5 M KCl (ATPase II), was not. ATPase I was highly unstable but could be stabilized by the addition of 0.1 mM ADP, 50% (v/v) glycerol or 0.3 mg/ml tubulin. ATPase I was further purified by CM-cellulose column chromatography, and by gel filtration on Sephacryl S-300. Its molecular weight, estimated by gel filtration, was 33,000. ATPase II had a high molecular weight and appeared to be associated with membrane vesicles. It sedimented on glycerol density gradient centrifugation with an s value of 27S. It was purified by high speed sedimentation and hydrophobic chromatography, and was observed under an electron microscope to consist of membrane vesicles of about 70 nm in diameter containing knob-like structures similar to those of H+-pump ATPase.
The catalytic properties of two ATPases which had been purified from bovine brain microtubules (Tominaga, S. & Kaziro, Y. (1983) J. Biochem. 93, 1085-1092) were studied. ATPase I, which had a molecular weight of 33,000, required the presence of 1.0 microM tubulin, 0.2 mM Mg2+, and 10 mM Ca2+ for maximal activity. The activation of ATPase I by tubulin was specific to the native form of tubulin, which could not be replaced by F-actin or tubulin denatured either by heat or more mildly by dialysis in the absence of glycerol. ATPase I was not specific to ATP, and GTP, and to a lesser extent, UTP and CTP were also hydrolyzed. Km for ATP of ATPase I was about 0.04 mM. ATPase I was inhibited by 5 mM Mg2+, 0.04 M K+, 10(-3) M vanadate, 10 mM N-ethylmaleimide, or 20% (v/v) glycerol. ATPase II, which was associated with membrane vesicles, required the presence of 0.2-2.0 mM Mg2+ and 20 mM KCl for activity. Tubulin stimulated the reaction of ATPase II only partially, and the addition of Ca2+ was rather inhibitory. ATPase II was specific to ATP with a Km value of 0.14 mM. It was inhibited by 1.6 mM N-ethylmaleimide and 20% (v/v) glycerol, but was not very sensitive to vanadate. Instead, ATPase II was inhibited by trifluoperazine, chlorpromazine, and nicardipin at 10(-3) M.
Of the two plasmids pTUB1 and pTUB2 constructed by cloning of the 8.9 kb EcoRI fragment carrying tufB (Miyajima, A., Shibuya, M., & Kaziro, Y. (1979) FEBS Lett. 102, 207-210), pTUB2 possesses a deletion of about 0.3 kb. Restriction and sequence analyses have located the deletion in the region of the four tRNA genes thrU-tyrU-glyT-thrT upstream of the tufB structural gene. As a result of homologous recombination between thrU and thrT, the four tRNA genes have been replaced by a single thrU-thrT hybrid gene. The deletion of the three tRNA genes does not significantly alter the in vivo expression of tufB as assessed by the kirromycin-sensitive phenotype of the transformant cells and by the synthesis of EF-Tu in mini-cells. Nor does the deletion affect the synthesis of beta-galactosidase in lysogens carrying a lambda transducing phage with a tufB-lacZ fusion. Transcription of tufB and synthesis of EF-Tu in a cell-free transcription-translation coupled system were essentially the same, regardless of whether pTUB1 or pTUB2 DNA was used as a template. Likewise, 0.2 mM ppGpp inhibits the synthesis of tufB mRNA on both pTUB1 and pTUB2 templates to the same extent. We concluded that the replacement by thrU-thrT hybrid gene of the four tRNA genes upstream of the tufB coding region does not significantly affect either in vivo or in vitro expression of tufB.
Explore the source record for details and available documents.
To investigate the regulatory mechanism of the tufB operon, we have constructed plasmids in which the lac structural genes have been fused to the regulatory region and the 5'-coding sequence of the tufB gene. The fusion was performed by incorporating the 6.6 kb EcoRI-HpaI fragment of plasmid pTUB1, which carried the tufB gene (Miyajima et al. 1979), into the EcoRI and SmaI sites of pMC1403 lac fusion vector (Casadaban et al. 1980). This gene fusion resulted in the production of a hybrid protein consisting of the N-terminal portion (12 amino acid residues) of EF-TuB and the enzymatically active C-terminal half of beta-galactosidase. Bacteria harboring the recombinant plasmid showed a strong Lac+ phenotype. In such a fusion, the lac gene expression was under the control of the tufB promoter. This was evidenced by the following observations; (i) the tufB-lacZ hybrid protein was synthesized constitutively; (ii) its production augmented in parallel with the increase in growth rate; and (iii) on carbon-source upshift, the hybrid protein was produced at a rate 2.5-fold higher than that of the mass increase. Several derivatives of this recombinant plasmid harboring deletions and/or inversions in the tufB regulatory region have been constructed and their properties are described.
Rat liver RNA guanylyltransferase catalyzes a GTP-PPi exchange reaction in the absence of acceptor RNA [Mizumoto, K. & Lipmann, F. (1979) Proc. Natl. Acad. Sci. USA 76, 4961-4965] suggesting that the reaction proceeds through the formation of a covalent guanylylated intermediate. We now present more direct evidence for the existence of the enzyme-GMP intermediate: (i) the enzyme-[32P]GMP intermediate was formed on incubation of rat liver guanylyltransferase with [alpha-32P]GTP and migrated as a single radioactive band with Mr 69,000 on NaDodSO4/polyacrylamide gel electrophoresis, and (ii) the intermediate isolated on gel filtration can transfer its GMP moiety to ppGpCpC-poly(A2,U2,G) to form the capped RNA molecule or it can react with PPi to regenerate GTP. The formation of the intermediate was dependent on Mg2+ and was strongly inhibited by PPi. The addition of pyrophosphatase markedly increased the amount of the intermediate complex. On blue dextran-Sepharose affinity column chromatography, the activity of guanylyltransferase to form an enzyme-[32P]GMP intermediate comigrated with activities of cap formation and GTP-PPi exchange. A phosphoamide type linkage between GMP and enzyme is suggested by its acidlabile and alkali-stable nature and also by the susceptibility to acidic hydroxylamine. These results indicate that the reaction catalyzed by rat liver guanylyltransferase occurs through the following two partial steps: (i) E + GTP in equilibrium E-pG + PPi; and (ii) E-pG + ppN .....leads to GpppN .....+ E.
The complete amino acid sequence of Fragment B obtained by the limited tryptic digestion of E. coli polypeptide chain elongation factor Tu (EF-Tu) was determined. Seven peptides formed from Fragment B by cleavage with cyanogen bromide (designated as CB1 to CB7 according to their order of alignment from N- to C-termini of Fragment B) were purified, and six of them were completely sequenced by the manual method of sequential Edman degradation with direct identification of the phenylthiohydantoin-amino acids. The remaining one cyanogen bromide peptide (CB6) containing 109 amino acid residues was further digested with trypsin. Twelve tryptic peptides (designated as T1 to T12 according to their order of alignment from N- to C-termini of CB6) were isolated, and their amino acid sequences were analyzed. The alignment of CB peptides was based on the results of the automated sequence analysis of Fragment B from its N-terminal, and the sequence analysis of the overlapping peptides containing sulfhydryl groups obtained by the complete tryptic digestion of Fragment B. The alignment of peptides T1 to T12 on CB6 was based on the result of the automated sequence analysis of CB6, and the sequence of the overlapping peptide obtained by the chemical cleavage of CB6 at the tryptophan residue using cyanogen bromide in heptafluorobutyric acid. The nucleotide sequence of the tuf A gene was also utilized for the alignment of these peptides. Fragment B comprises amino acid residues 59 to 263 of E. coli EF-Tu, which consists of 393 amino acids. It contains two functional (SH1 and SH2) and one non-functional (SH3) sulfhydryl groups of EF-Tu. All of the five histidine residues in Fragment B were distributed within the first N-terminal quarter, and three of them were found to be clustered around SH2. Although E. coli EF-Tu consists of two gene products (tuf A and tuf B), no microheterogeneity was found in the amino acid sequence of Fragment B.
Interaction of adenosine-5'-O-(3-thiotriphosphate) (ATP gamma S) with Ca2+,Mg2+-ATPase of sarcoplasmic reticulum was studied. The nucleotide was slowly hydrolyzed by the ATPase at 30 degrees C at a rate of about 0.5% that of ATP hydrolysis. Whereas at 0 degrees C, ATP gamma S showed only a limited reactivity toward the ATPase in that a thiophosphorylated intermediate was formed and ADP was released, but hydrolysis of the intermediate to complete the catalytic cycle did not occur. A fairly stable analog of the E-P intermediate could thus be obtained. Presence of the thiophosphorylated intermediate was indicated by the [3H]ADP in equilibrium ATP gamma S exchange reaction and also by using [35S]ATP gamma S. When the ATPase was reacted with ATP gamma S at 0 degrees C in the presence of ferricyanide, EP-forming activity was rapidly lost. Free Ca2+ ions were required for this inactivation. Disulfide bond formation between a cysteinyl residue located near the substrate binding site and the enzyme-bound ATP gamma S or the thiophosphorylated intermediate was suggested by the fact that 2-mercaptoethanol reversed the inactivation. The reaction may prove to be a useful tool for affinity labeling of the active site of the ATPase.
Explore the source record for details and available documents.
The transcription of the tufB gene by purified RNA polymerase holoenzyme was studied using the transducing phage lambda rifd 18 DNA and the hybrid plasmid pTUB1 DNA (Miyajima et al. 1979) as templates. The size of tufB mRNA synthesized in this system was about 1,700 nucleotides, and the same strand as for rrnB was transcribed. By electron microscopic examination of the R-loop formed between lambda fus3 DNA and tufB mRNA synthesized under the direction of pTUB1 DNA, it was found that the untranslated sequence of about 500 nucleotides is at the 5' end of tufB mRNA. The sequencing of the 5' region of tufB mRNA synthesized on the truncated template has revealed that the tufB gene is cotranscribed with its upstream genes for four tRNAs (thrU, tyrU, glyT, and thrT). The synthesis of this mRNA molecule is completely abolished by low concentrations of ppGpp. Neither pppGpp, ppGp, nor pGpp was effective as inhibitor in this cell-free system.
When EF-Tu was photooxidized for 20 min at 0 degrees C in the presence of 10 microM GDP and 5 microM rose bengal, the activity to promote the binding of [14C]Phe-tRNA to ribosomes was rapidly lost, while the activity to bind [3H]GDP remained intact. The activity of EF-Tu to interact with Phe-tRNA and ribosomes, as assessed by protection of [14C]Phe-tRNA against RNase A digestion and by methanol-induced uncoupled GTPase activity, respectively, was also inactivated under the above conditions. It was found, however, that these activities were fully protected in the presence of aminoacyl-tRNA and GTP, indicating that the active site(s) of EF-Tu for interaction with aminoacyl-tRNA and ribosomes could be protected against photooxidation in the ternary aminoacyl-tRNA . EF-Tu . GTP complex. Comparison of the amino acid composition of EF-Tu photooxidized in the form of EF-Tu . GDP with that of the intact EF-Tu revealed that only 1.4 residues of histidine were damaged. On the other hand, no histidine residue was lost when EF-Tu was oxidized in the presence of both aminoacyl-tRNA and GTP. The photooxidized EF-Tu . GDP was then partially degraded with trypsin and each of the resulting tryptic fragments, D, B, and C (Arai, Nakamura, Arai, Kawakita, and Kaziro (1976) J. Biochem. 79, 69-83), was analyzed for histidine content. The results indicated that fragments B, C, and D had lost 0.7, 0.5, and 0.2 residues of histidine, respectively. Since fragment B contains the cysteine residue which is essential for interaction with aminoacyl-tRNA and ribosomes, the above results suggest that a histidine residue in fragment B may also play an essential role in the interaction with aminoacyl-tRNA and ribosomes.