Survival and mutagenic effects of 5-azacytidine in Escherichia coli

Survival and mutagenesis caused by 5-azacytidine was studied in Escherichia coli. Survival was partially lexA- and recA-dependent and was decreased by the presence of a DNA (cytosine-5)methyltransferase. The dcm, MspI, and EcoRII methyltransferase genes all decreased survival. There was no direct relationship between amount of methylase enzyme present and cell survival, but only plasmids containing a methylase gene sensitized cells to 5-azacytidine. Survival was not affected by uvrA, uvrB or umuCD mutations. Induction of sulA::lacZ fusions by 5-azacytidine was inhibited in strains containing elevated levels of DNA methylase. Cells resistant to 5-azacytidine when they contained a plasmid specifying the EcoRII methylase were sensitive if the plasmid specified the complete EcoRII restriction-modification system. The mechanism of cell death in these situations is therefore different. Mutation of the rpoB gene by 5-azacytidine was studied. The mutation rate was decreased by the presence of recA and lexA mutations. Mutation in umuCD had little effect on the mutation rate. The recA430 mutation, which does not support SOS-dependent mutagenesis induced by UV light, does support 5-azacytidine induced mutagenesis. The presence of DNA (cytosine-5)methyltransferase had no effect on the mutation rate caused by 5-azacytidine treatment. The mutagenic and lethal lesions caused by 5-azacytidine in the absence of methylase therefore differ from the lethal lesions that occur in the presence of methylase. The former could be due to the opening of the 5-azacytosine ring in DNA. Cell death in the presence of methylase could be due to tight binding of methylase to azacytosine containing DNA as well as inhibition of induction of the SOS response.     

Substitutions of a cysteine conserved among DNA cytosine methylases result in a variety of phenotypes.

The proposed mechanism for DNA (cytosine-5)-methyltransferases envisions a key role for a cysteine residue. It is expected to form a covalent link with carbon 6 of the target cytosine, activating the normally inactive carbon 5 for methyl transfer. There is a single conserved cysteine among all DNA (cytosine-5)-methyltransferases making it the candidate nucleophile. We have changed this cysteine to other amino acids for the EcoRII methylase; which methylates the second cytosine in the sequence 5'-CCWGG-3'. Mutants were tested for their methyl transferring ability and for their ability to form covalent complexes with DNA. The latter property was tested indirectly with the use of a genetic assay involving sensitivity of cells to 5-azacytidine. Replacement of the conserved cysteine with glycine, valine, tryptophan or serine led to an apparent loss of methyl transferring ability. Interestingly, cells carrying the mutant with serine did show sensitivity to 5-azacytidine, suggesting the ability to link to DNA. Unexpectedly, substitution of the cysteine with glycine results in the inhibition of cell growth and the mutant allele can be maintained in the cells only when it is poorly expressed. These results suggest that the conserved cysteine in the EcoRII methylase is essential for methylase action and it may play more than one role in it.     

Effect of 5-azacytidine on DNA methylation in Ehrlich's ascites tumor cells

5-Azacytidine inhibited in vivo DNA methylation in Ehrlich's ascites tumor cells depending upon the dose at which 5-azacytidine did not inhibit DNA synthesis significantly. This drug did not inhibit DNA methylation in vitro. The DNA methylase activity in ascitic cells decreased with the increasing dose of 5-azacytidine. Hypomethylated DNA was obtained from the 5-azacytidine treated ascitic cells.     

Cloning and characterization of the dcm locus of Escherichia coli K-12.

The dcm locus of Escherichia coli K-12 has been shown to code for a methylase that methylates the second cytosine within the sequence 5'-CC(A/T)GG-3'. This sequence is also recognized by the EcoRII restriction-modification system coded by the E. coli plasmid N3. The methylase within the EcoRII system methylates the same cytosine as the dcm protein. We have isolated, from a library of E. coli K-12 DNA, two overlapping clones that carry the dcm locus. We show that the two clones carry overlapping sequences that are present in a dcm+ strain, but are absent in a delta dcm strain. We also show that the cloned gene codes for a methylase, that it complements mutations in the EcoRII methylase, and that it protects EcoRII recognition sites from cleavage by the EcoRII endonuclease. We found no phage restriction activity associated with the dcm clones.     

Inhibition of recA-mediated strand exchange by adducts of azacytosine-containing DNA and the EcoRII methylase

Wild type Escherichia coli cells containing elevated levels of DNA (cytosine-5)methyltransferases have increased sensitivity to the toxic effects of 5-azacytidine. The methyltransferases form tight binding complexes with azacytosine in DNA which could interfere with the recA recBCD repair pathway which is largely responsible for cell survival after treatment with the drug. We therefore determined if these complexes interfered with recA-mediated strand exchange in vitro. 32P-Labeled DNA fragments containing a single EcoRII site, with cytosine in the (-) strand replaced by 5-azacytosine, were prepared. We investigated the effect of the EcoRII methyltransferase on recA-mediated strand exchange with homologous M13 DNA by electrophoresis in agarose gels. In the absence of the methylase the rate and extent of strand exchange of azacytosine-containing DNA is the same as control DNA. In the presence of the methyltransferase strand exchange is inhibited, but some incorporation of duplexes into recA-single-stranded DNA (ssDNA) complexes still occurs. The formation of these complexes is dependent on the length of the fragment 3' to the methylase binding site on the strand complementary to the ssDNA. The greater the length the greater the number of complexes that form. S-Adenosyl-L-methionine, which enhances binding of the methyltransferase to azacytosine-containing DNA, causes an increase in the inhibition of strand exchange and an increase in the number of inactive complexes formed. The complexes can be dissociated with guanidinium chloride which denatures the methyltransferase and leads to release of the (+) strand. The (-) strand remains associated with the ssDNA. This result implies that a plectonemic joint is formed between recA-ssDNA complexes and azacytosine-containing DNA-methyltransferase complexes. However, branch migration in these complexes is inhibited. Denaturation of the methyltransferase allows branch migration to proceed to completion, releasing the (+) strand.     

Nucleotide sequence of the EcoRII restriction endonuclease gene

The nucleotide sequence of a 1394 basepair (bp) DNA fragment containing the EcoRII restriction endonuclease (R.EcoRII) gene was determined. The endonuclease gene is 1206 bp in length (predicted 402 amino acids (aa) and Mr = 45 178) and is separated by 33 bp from the EcoRII modification methylase (M.EcoRII) gene. The EcoRII restriction-modification system has a tail-to-tail organization of the two genes.     

Genetic analysis of the 5-azacytidine sensitivity of Escherichia coli K-12.

DNA containing 5-azacytidine (5-azaC) has been shown to form stable detergent-resistant complexes with cytosine methylases. We reasoned that if 5-azaC treatment causes protein-DNA cross-links in vivo, then mutations in DNA repair and recombination genes may increase the sensitivity of a cell to 5-azaC. We found that although recA (defective) and lexA (induction-negative) mutants of Escherichia coli were very sensitive to the drug, mutations in uvrA and ung genes had little effect on cell lethality. The sensitivity of recA strains to 5-azaC was dose dependent and was enhanced by the overproduction of a DNA cytosine methylase in the cell. Unexpectedly, a strain of E. coli carrying a recA mutation and a deletion of the DNA cytosine methylase gene (dcm) was found to be significantly sensitive to 5-azaC. Study of mutations in the pyrimidine salvage pathway of E. coli suggests that direct phosphorylation of 5-azaC, rather than phosphorylation of its degradation products, is largely responsible for the lethal effects of the drug. The addition of uracil to the growth medium has little effect on cell lethality of recA mutants, but it partially reversed the inhibition of cell growth caused by 5-azaC. This reversal of the bacteriostatic effects of the drug could not be achieved by adding cytosine or orotic acid to the growth medium and required the presence of functional UMP-pyrophosphorylase (gene upp) in the cell.     

Mechanism of action of nalidixic acid on Escherichia coli. 3. Conditions required for lethality

Deitz, William H. (Sterling-Winthrop Research Institute, Rensselaer, N.Y.), Thomas M. Cook, and William A. Goss. Mechanism of action of nalidixic acid on Escherichia coli. III. Conditions required for lethality. J. Bacteriol. 91:768-773. 1966.-Nalidixic acid selectively inhibited deoxyribonucleic acid (DNA) synthesis in cultures of Escherichia coli 15TAU. Protein and ribonucleic acid synthesis were shown to be a prerequisite for the bactericidal action of the drug. This action can be prevented by means of inhibitors at bacteriostatic concentrations. Both chloramphenicol, which inhibits protein synthesis, and dinitrophenol, which uncouples oxidative phosphorylation, effectively prevented the bactericidal action of nalidixic acid on E. coli. The lethal action of nalidixic acid also was controlled by transfer of treated cells to drug-free medium. DNA synthesis resumed immediately upon removal of the drug and was halted immediately by retreatment. These studies indicate that nalidixic acid acts directly on the replication of DNA rather than on the "initiator" of DNA synthesis. The entry of nalidixic acid into cells of E. coli was not dependent upon protein synthesis. Even in the presence of an inhibiting concentration of chloramphenicol, nalidixic acid prevented DNA synthesis by E. coli 15TAU.     

In vivo methylation of bacteriophage phi X174 DNA.

A mutant (designated mec(-)) has been isolated from Escherichia coli C which has lost DNA-cytosine methylase activity and the ability to protect phage lambda against in vivo restriction by the RII endonuclease. This situation is analogous to that observed with an E. coli K-12 mec(-) mutant; thus, the E. coli C methylase appears to have overlapping sequence specificity with the K-12 and RII enzymes; (the latter methylases have been shown previously to recognize the same sequence). Covalently closed, supertwisted double-standed DNA (RFI) was isolated from C mec(+) and C mec(-) cells infected with bacteriophage phiX174. phiX. mec(-) RFI is sensitive to in vitro cleavage by R.EcoRII and is cut twice to produce two fragments of almost equal size. In contrast, phiX.mec(+) RFI is relatively resistant to in vitro cleavage by R.EcoRII. R.BstI, which cleaves mec(+)/RII sites independent of the presence or absence of 5-methylcytosine, cleaves both forms of the RFI and produces two fragments similar in size to those observed with R. EcoRII. These results demonstrate that phiX.mec(+) RFI is methylated in vivo by the host mec(+) enzyme and that this methylation protects the DNA against cleavage by R.EcoRII. This is consistent with the known location of two mec(+)/ RII sequences (viz., [Formula: see text]) on the phiX174 map. Mature singlestranded virion DNA was isolated from phiX174 propagated in C mec(+) or C mec(-) in the presence of l-[methyl-(3)H]methionine. Paper chromatographic analyses of acid hydrolysates revealed that phiX.mec(+) DNA had a 10-fold-higher ratio of [(3)H]5-methylcytosine to [(3)H]cytosine compared to phiX.mec(-). Since phiX.mec(+) contains, on the average, approximately 1 5-methylcytosine residue per viral DNA, we conclude that methylation of phiX174 is mediated by the host mec(+) enzyme only. These results are not consistent with the conclusions of previous reports that phiX174 methylation is mediated by a phage-induced enzyme and that methylation is essential for normal phage development.     

Restriction-modification systems of type II in Shigella strains

Two restriction-modification systems specified by two plasmids are discovered in the clinical species of Shigella. The plasmids are designated pKMR114 and pKMR115. Both are of 60.800 bp and belong to the IncN incompatibility group. The EcoRI, EcoRV, HindIII restriction patterns of both plasmid DNAs are identical. As shown by efficiency of plating of bacteriophage lambda vir on the strains harbouring plasmids encoding EcoRI, EcoRII, EcoRIII, EcoRIV, EcoRV systems and plasmids studied, the discovered plasmids control synthesis of EcoRII specificity enzymes. The main distinctive feature of the pKMR114 is the ability to decrease efficiency of plating of bacteriophage T4 having glycolised DNA.     

Identification of a DNA restriction-modification system in Pectobacterium carotovorum strains isolated from Poland

AIMS: Polish isolates of pectinolytic bacteria from the species Pectobacterium carotovorum were screened for the presence of a DNA restriction-modification (R-M) system. METHODS AND RESULTS: Eighty-nine strains of P. carotovorum were isolated from infected potato plants. Sixty-six strains belonged to P. carotovorum ssp. atrosepticum and 23 to P. carotovorum ssp. carotovorum. The presence of restriction enzyme Pca17AI, which is an isoschizomer of EcoRII endonuclease, was observed in all isolates of P. c. atrosepticum but not in P. c. carotovorum. The biochemical properties, PCR amplification, and sequences of the Pca17AI restriction endonuclease and methyltransferase genes were compared with the prototype EcoRII R-M system genes. Only when DNA isolated from cells of P. c. atrosepticum was used as a template, amplification of a 680 bp homologous to the gene coding EcoRII endonuclease. CONCLUSIONS: Endonuclease Pca17AI, having a relatively low temperature optimum, was identified. PCR amplification revealed that the nucleotide sequence of genes for EcoRII and Pca17AI R-M are different. Dcm methylation was observed in all strains of Pectobacterium and other Erwinia species tested. The sequence of a DNA fragment coding Dcm methylase in P. carotovorum was different from that of Escherichia coli. SIGNIFICANCE AND IMPACT OF THE STUDY: Pca17AI is the first psychrophilic isoschizomer of EcoRII endonuclease. The presence of specific Dcm methylation in chromosomal DNA isolated from P. carotovorum is described for the first time. A 680 bp PCR product, unique for P. c. atrosepticum strains, could serve as a molecular marker for detection of these bacteria in environmental samples.     

Primary sequence of the EcoRII endonuclease and properties of its fusions with beta-galactosidase

The EcoRII endonuclease cleaves DNA containing the sequence CC(A/T)GG before the first cytosine. The methylation of the second cytosine in the sequence by either the EcoRII methylase or Dcm, a chromosomally coded protein in Escherichia coli, inhibits the cleavage. The gene for the EcoRII endonuclease was mapped by analysis of derivatives containing linker insertions, transposon insertions, and restriction fragment deletions. Surprisingly, plasmids carrying the wild-type endonuclease gene and the EcoRII methylase gene interrupted by transposon insertions appeared to be lethal to dcm+ strains of E. coli. We conclude that not all the EcoRII/Dcm recognition sites in the cellular DNA are methylated in dcm+ strains. The DNA sequence of a 1650-base pair fragment containing the endonuclease gene was determined. It revealed an open reading frame that could code for a 45.6-kDa protein. This predicted size is consistent with the known size of the endonuclease monomer (44 kDa). The endonuclease and methylase genes appear to be transcribed convergently from separate promoters. The reading frame of the endonuclease gene was confirmed at three points by generating random protein fusions between the endonuclease and beta-galactosidase, followed by an analysis of the sequence at the junctions. One of these fusions is missing 18 COOH-terminal amino acids of the endonuclease but still displays significant ability to restrict incoming phage in addition to beta-galactosidase activity. No striking similarity between the sequence of the endonuclease and any other protein in the PIR data base was found. The knowledge of the primary sequence of the endonuclease and the availability of the various constructs involving its gene should be helpful in the study of the interaction of the enzyme with its substrate DNA.     

Very short patch mismatch repair activity associated with gene dcm is not conferred by a plasmid coding for EcoRII methylase.

The only cytosine methylase in Escherichia coli K-12 methylates the second cytosine in the sequence CC (A/T)GG and is encoded by gene dcm. Methylation and very short patch mismatch repair activities lacking in a dcm mutant of E. coli were restored by a plasmid containing the cloned dcm gene. In contrast, plasmids with the gene for EcoRII methylase, which is a homolog of dcm, restored only cytosine methylase activity and not mismatch repair.     

Conjugal Deoxyribonucleic Acid Replication by Escherichia coli K-12: Effect of Nalidixic Acid

During the conjugal transfer of the R64-11 plasmid at 42 C from donor cells thermosensitive for vegetative deoxyribonucleic acid (DNA) synthesis to recipient minicells, the plasmids are conjugally replicated in the donor cells. This conjugal replication is inhibited by nalidixic acid, and the degree of inhibition is comparable to the reduction in the amount of plasmid DNA transferred to the recipient minicells in the presence of the drug. In addition, the size of DNA transferred to the minicells and the fraction of conjugally replicated DNA in the donor cells that can be isolated as closed-circular plasmid DNA under alkaline conditions are both reduced by nalidixic acid. When the drug is added to a mating that is underway, the rate of conjugal replication is immediately reduced. This change is accompanied by a reduction in the amount of conjugally replicated DNA in the donor cells that can be isolated as closed-circular plasmid DNA. Furthermore, conjugally replicated plasmid DNA that is not associated with the donor cell membrane becomes membrane bound after the addition of nalidixic acid.     

Maintenance forced by a restriction-modification system can be modulated by a region in its modification enzyme not essential for methyltransferase activity

Several type II restriction-modification gene complexes can force their maintenance on their host bacteria by killing cells that have lost them in a process called postsegregational killing or genetic addiction. It is likely to proceed by dilution of the modification enzyme molecule during rounds of cell division following the gene loss, which exposes unmethylated recognition sites on the newly replicated chromosomes to lethal attack by the remaining restriction enzyme molecules. This process is in apparent contrast to the process of the classical types of postsegregational killing systems, in which built-in metabolic instability of the antitoxin allows release of the toxin for lethal action after the gene loss. In the present study, we characterize a mutant form of the EcoRII gene complex that shows stronger capacity in such maintenance. This phenotype is conferred by an L80P amino acid substitution (T239C nucleotide substitution) mutation in the modification enzyme. This mutant enzyme showed decreased DNA methyltransferase activity at a higher temperature in vivo and in vitro than the nonmutated enzyme, although a deletion mutant lacking the N-terminal 83 amino acids did not lose activity at either of the temperatures tested. Under a condition of inhibited protein synthesis, the activity of the L80P mutant was completely lost at a high temperature. In parallel, the L80P mutant protein disappeared more rapidly than the wild-type protein. These results demonstrate that the capability of a restriction-modification system in forcing maintenance on its host can be modulated by a region of its antitoxin, the modification enzyme, as in the classical postsegregational killing systems.     

Mechanism of action of nalidixic acid on Escherichia coli. IV. Effects on the stability of cellular constituents

Cook, Thomas M. (Sterling-Winthrop Research Institute, Rensselaer, N.Y.), William H. Deitz, and William A. Goss. Mechanism of action of nalidixic acid on Escherichia coli. IV. Effects on the stability of cellular constituents. J. Bacteriol. 91:774-779. 1996.-Treatment of Escherichia coli 15TAU with nalidixic acid resulted in degradation of the nucleic acids of the cells, whereas protein was unaffected. Deoxyribonucleic acid (DNA) degradation appeared to be more extensive than ribonucleic acid degradation during periods of comparable bactericidal action. The onset of DNA degradation was evident prior to a measurable bactericidal effect. However, within the range of 2 to 20%, DNA degradation was accompanied by a decrease in viable cell numbers. Degradation of DNA to acid-soluble material occurred only under conditions permitting the bactericidal action of nalidixic acid. Arrest of the bactericidal action of nalidixic acid by the addition of dinitrophenol or chloramphenicol also inhibited DNA degradation. The acid-soluble products, which were excreted into the medium, have not been characterized completely, but probably were not phosphorylated.     

Alterations in differentiation and pyrimidine pathway enzymes in 5-azacytidine resistant variants of a myoblast line.

5-azacytidine at concentrations higher than 5 muM inhibited the differentiation of a rat myoblast line in vitro. It was also somewhat cytotoxic at this level. Variants resistant to the cytotoxic effect of 5-azacytidine were obtained which were simultaneously unable to differentiate into myotubes and exhibited altered morphology. These characteristics were retained by the variants when subcultured in the absence of the drug for over 700 generations. Several of the azacytidine resistant cells were more susceptible than the parental line to the lethal action of 5-bromodeoxyuridine and adenosine, but not that of cytosine arabinoside, ouabain or 8-azaguanine. The variants were capable of transporting uridine, thymidine and 5-azacytidine. The uridine kinase activity was one-half to one-third of than in the parental cells but it was not missing completely in any of the variants. Two independently isolated variants selected for detailed study showed a 2- to 3-fold increase in the activity of orotidylic acid decarboxylase. This enzyme in the variants in contrast to that of the parental cells was completely insensitive to the inhibitory effect of a nucleotide generated from ATP and 5-azacytidine in cell extracts (probably 5-azacytidine monophosphate). These observations point to the possibility the 5-azacytidine resistance arises in myoblasts due to an alteration of the components of two target pathways of this drug, viz., the de novo pyrimidine pathway and an undefined sequence leading to the synthesis of membrane components.     

Nucleotide sequence of the DdeI restriction-modification system and characterization of the methylase protein.

The DdeI restriction-modification system was previously cloned and has been maintained in E. coli on two separate and compatible plasmids (1). The nucleotide sequence of the endonuclease and methylase genes has now been determined; it predicts proteins of 240 amino acids, Mr = 27,808, and 415 amino acids, Mr = 47,081, respectively. Inspection of the DNA sequence shows that the 3' end of the methylase gene had been deleted during cloning. The clone containing the complete methylase gene was made and compared to that containing the truncated gene; only clones containing the truncated form support the endonuclease gene in E. coli. Bal-31 deletion studies show that methylase expression in the Dde clones is also dependent upon orientation of the gene with respect to pBR322. The truncated and complete forms of the methylase protein were purified and compared; the truncated form appears to be more stable and active in vitro. Finally, comparison of the deduced amino acid sequence of M. DdeI with that of other known cytosine methylases shows significant regions of homology.     

Transfer of recombinant plasmids containing the gene for DpnII DNA methylase into strains of Streptococcus pneumoniae that produce DpnI or DpnII restriction endonucleases.

Plasmid transfer via the transformation pathway of Streptococcus pneumoniae was weakly restricted by the DpnI or DpnII restriction endonuclease, either of which gave a reduction only to 0.4, compared with phage infection, which was restricted to 10(-5). The greater sensitivity of plasmid transfer compared with chromosomal transformation, which was not at all restricted, can be attributed to partially double-stranded intermediates formed from two complementary donor fragments. However, clustering of potential restriction sites in the plasmids increased the probability of escape from restriction. The recombinant plasmid pMP10 , in which the gene for the DpnII DNA methylase was cloned, can be transferred to strains that contain neither restriction enzyme or that contain DpnII as readily as can the vector pMP5 . Introduction of pMP10 raised the level of methylase by five times the level normally present in DpnII strains. Transfer of pMP10 to DpnI -containing strains was infrequent, presumably owing to the suicidal methylation of DNA which rendered it susceptible to the host endonuclease. The few clones in which pMP10 was established had lost DpnI . Loss of the plasmid after curing of the cell eliminated the methylase but did not restore DpnI . Although this loss of DpnI could result from spontaneous mutation, its relatively high frequency, 0.1% suggested that the loss was due to a regulatory shift.