Counterselection of GATC sequences in enterobacteriophages by the components of the methyl-directed mismatch repair system

Summary Weak to severe deficit of GATC sequences in the DNA of enterobacteriophages appears to be correlated with their undermethylation during growth indam ⁺ (GATC ade-methylase) bacteria. This observation is corroborated by the sequence analysis showing no evidence for site-specific mutagenicity of 6meAde. The MutH protein of the methyl-directed mismatch repair system recognizes and cleaves the undermethylated GATC sequences in the course of mismatch repair. To enquire whether the MutH function of the methyldirected mismatch repair system participates in counterselection of GATC sequences in enterobacteriophages, we have studied the yield of bacteriophage X174 containing either 0, 1, or 2 GATC sequences, in wild type,dam, andmut (H, L, S, U) Escherichia coli. Following transfection with unmethylated DNA containing two GATC sequences, a net decrease in the yield of infective particles was observed in all bacterialmutH ⁺ dam^– strains, whereas no detectable decrease was observed in bacteria infected by DNA without GATC sequence. This effect of the MutH function is maximum in wild type andmutL andmutS bacteria whereas the effect is not significant inmutU bacteria, suggesting an interaction of the, helicase II with the MutH protein.However, indam ⁺ bacteria, the presence of GATC sequences leads to an increased yield of infective particles. The effect of GATC sequence and its Dam methylation system on phage yield inmutH ^– bacteria reveals that methylated GATC sequences are advantageous to the phage. These results suggest that the methyl-directed mismatch repair system, and in particular its MutH protein, may have participated in severe counterselection of GATC sequences from enterobacteriophages, presumably, by DNA cleavage or by interfering with DNA replication or packaging when GATC sequences are undermethylated. Coevolution of the Dam and MutH proteins could then account for the loss of GATC sequences from DNA of bacteriophages growing indam ⁺ hosts.     

High-frequency electroporation and maintenance of pUC- and pBR-based cloning vectors inPseudomonas stutzeri

A number ofEscherichia coli cloning vectors, based on ColE1-like replicons, were shown to be maintained inPseudomonas stutzeri ATCC 17588. A restrictionless mutant ofP. stutzeri was isolated, and this strain was used to develop an efficient electroporation system. With theE. coli cloning vector pHSG298, transformation frequencies of up to 2×10⁷ transformants/g DNA were achieved. This frequency is comparable to that obtained for CaCl₂-mediated transformation ofE. coli; thus, direct cloning of DNA intoP. stutzeri is feasible. As will be discussed, this may prove useful for cloning DNA from high mol% G+C genera in cases in whichE. coli is not a suitable heterologous cloning host.     

Establishment of a novel sensitive method for detecting methylation modification on DNA of escherichia coli cell

This study focused on finding a novel sensitive method to determine the methylation modification at DNA dam (GATC) sites in Escherichia coli. A new plasmid which contained three GATC sites recognized by restriction enzyme BclI and one GAATTC site recognized by EcoRI was transformed into E. coli stains AB1157(dam ⁺) and GM2929(dam ⁻) respectively. Then the plasmid DNA was digested by restriction enzyme BclI(T*GATCA), which was sensitive to methylation. The results showed that the plasmid derived from AB1157 was not digested while that from GM2929 was, for the methylation level of the former was high while the latter was low. So by detecting the methylation of plasmid transferred into the strain, we could determine whether methylaion existed at DNA dam (GATC) site in E. coli. This method was effective and rapid; moreover, the digested fragments were not dispersive. It also made a basis for the detection of whether methylation occurred in mode beings by low-energy ion beam. The article is published in the original.     

Cloning and sequencing of two genes fromStaphylococcus carnosus coding for glucose-specific PTS and their expression inEscherichia coliK-12

Phosphoenolpyruvate (PEP)-dependent phosphorylation experiments have indicated that the grampositive bacteriumStaphylococcus carnosus possesses an EIICBA fusion protein specific for glucose. Here we report the cloning of a 7 kb genomic DNA fragment containing two genes,glcA andglcB, coding for the glucose-specific PTS transporters EII^Glc1 and EII^Glc2 which are 69% identical. The translation products derived from the nucleotide sequence consist of 675 and 692 amino acid residues and have calculated molecular weights of 73 025 and 75 256, respectively. Both genes can be stably maintained inEscherichia coli cells and restore the ability to ferment glucose toptsG deletion mutants ofE. coli. This demonstrates the ability of the PTS proteins HPr and/or EIIA^Glc of a gram-negative organism (E. coli) to phosphorylate an EIICBA^Glc from a gram-positive organism (S. carnosus).     

Electrotransformation ofClostridium thermosaccharolyticum

Transformation of the thermophileClostridium thermosaccharolyticum ATCC 31960 was achieved using plasmid pCTC1 and electroporation. Evidence supporting transformation was provided by Southern blots, detection of the plasmid in 10 out of 10 erythromycin-resistant clones, retransformation ofE. coli andC. thermosaccharolyticum with plasmid DNA isolated fromC. thermosaccharolyticum, and a proportional relationship between the number of transformants and the amount of DNA added. Transformation efficiencies were very low for plasmid DNA prepared fromE. coli (0.6 transformants mg^–1 DNA), although somewhat higher for plasmid DNA prepared fromC. thermosaccharolyticum (52 transformants mg^–1 DNA). Transformation-dependent erythromycin resistance indicates that an adenosine methylase gene originating fromEnterococcus faecalis, a mesophile, is expressed inC. thermosaccharolyticum. The plasmid pCTC1 appears to be replicated independently of the chromosome, as indicated by visualization of recovered plasmid on gels, and retransformation using recovered plasmid. pCTC1 is maintained inC. thermosaccharolyticum at both 45 and 60°C. Restriction analysis showed little or no rearrangement occurred upon passage through the thermophile.     

Enhanced tolerance against freezing stress in<Emphasis Type="Italic">Escherichia coli</Emphasis> cells expressing an algal cyclophilin gene

Members of the cyclophilin (Cyp) family are known to function as co-chaperones, interacting with chaperones such as heat shock protein 90, and perform important roles in protein folding under high temperature stress. In addition, they have been isolated from a wide range of organisms. However, there have been no reports on the functions of algal Cyps under other stress conditions. To study the functions of the cDNAGjCyp-1 isolated from the red alga (Griffithsia japonica), a recombinant GjCyp-1 containing a hexahistidine tag at the amino-terminus was constructed and expressed inEscherichia coli. Most of the gene product expressed inE. coli was organized as aggregate insoluble particles known as inclusion bodies. Thus, the optimal time, temperature, and concentration ofl(+)-arabinose for expressing the soluble and nonaggregated form of GjCyp-1 inE. coli were examined. The results indicate that the induction of Cyp, at 0.2%l(+)-arabinose for 2 h at 25°C, had a marked effect on the yield of the soluble and active form of the co-chaperone as PPlase. An expressed fusion protein, H₆GjCyp-1, maintained the stability ofE. coli proteins up to-75°C. In a functional bioassay of the recombinant H₆GjCyp-1, the viability ofE. coli cells overexpressing H₆GjCyp-1 was compared to that of cells not expressing H₆GjCyp-1 at −75°C. For all the cycles of a freeze/thaw treatment, a significant increase in viability was observed in theE. coli cells overexpressing H₆GjCyp-1. The results of the GjCyp-1 bioassays, as well asin vitro studies, strongly suggest that the algal Cyp confers freeze tolerance toE. coli.     

Quantitative nuclear DNA changes inEleusine (Gramineae)

2C nuclear DNA amounts were determined in 30 collections belonging to 10 species ofEleusine. About a 2.5-fold variation in genome size is evident in the genus. The 2C DNA amount in the diploid species ranged from 2.50 pg inE. verticillata to 3.35 pg inE. intermedia. In contrast, the tetraploid species showed a range from 4.95 pg inE. africana to 6.13 pg inE. floccifolia. At intraspecific level 10 collections ofE. coracana, 6 ofE. indica, 4 ofE. africana, 2 ofE. tristachya, and 2 ofE. kigeziensis did not show any significant variation. However, 2 collections ofE. floccifolia, connected with polyploidy, displayed about 90% variation. Polyploid species showed approximately double the genome size of that of their corresponding diploids. An evolutionary increase in DNA amount is evident inE. coracana during the course of its origin and domestication fromE. africana.     

Cloning and expression inEscherichia coli of the sucrase gene fromBacillus subtilis

Summary A recombinant cosmid carrying the sucrase gene (sacA) was obtained from a colony bank ofE. coli harboring recombinant cosmids representative of theB. subtilis genome. It was shown that thesacA gene is located in a 2 kbEcoRI fragment and that the cloned sequence is homologous to the corresponding chromosomal DNA fragment. A fragment of 2 kb containing the gene was subcloned in both orientations in the bifunctional vector pHV33 and expression was further looked for inB. subtilis andE. coli. Complementation of asacA mutation was observed in Rec⁺ and Rec^- strains ofB. subtilis. Expression of sucrase was also demonstrated inE.coli, which is normally devoid of this activity, by SDS-polyacrylamide gel electrophoresis, specific immunoprecipitation and assay of the enzyme in crude extracts. The specific activity of the enzyme depended on the orientation of the inserted fragment. The saccharolytic activity was found to be cryptic inE. coli since the presence of the recombinant plasmids did not allow the transport of [U¹⁴C] sucrose and the growth of the cells.It was shown also that the recombinant cosmid contained part of the neighboring locus (sacP) which corresponds to a component of the PEP-dependent phosphotransferase system of sucrose transport ofB. subtilis.     

Partial characterization ofPseudomonas fluorescens subsp.cellulosa endoglucanase activity produced inEscherichia coli

Summary The recombinant plasmid, pPFC4, which carriesPseudomonas fluorescens subsp.cellulosa chromosomal DNA was previously isolated on the basis of its ability to direct the expression of endoglucanase inEscherichia coli. In the present study, some physical and chemical properties of this activity were characterized. The major portion (78.4%) of the endoglucanase activity is found in the periplasmic space ofE. coli. This plasmid-encoded endoglucanase has a pH optimum of approximately 6.0 and a temperature optimum of approximately 50°C. With carboxymethylcellulose-zymograms, after polyacrylamide gel electrophoresis, periplasmic extracts fromE. coli carrying pPFC4 show six distinct bands with endoglucanase activity. The molecular mass of the major endoglucanase band is approximately 29 kDa while the remaining bands with endoglucanase activity range from 48 to 100 kDa. Although the basis of this heterogeneity is not known, the DNA insert of pPFC4 that encodes endoglucanase activity is not large enough to contain six separate genes; hence, the observed array of endoglucanases may result from post-translational modification of one or two primary gene products.     

Comparison of the cell surface barrier and enzymatic modification system inBrevibacterium flavum andB. lactofermentum

To investigate impediments to plasmid transformation inBrevibacterium flavum BF4 andB. lactofermentum BL1, cell surface barriers were determined by measuring growth inhibition whilst enzymatic barriers were determined by comparing DNA methylation properties.B. lactofermentum was more sensitive to growth inhibition by glycine thanB. flavum. Release of cellular proteins during sonication was more rapid forB. lactofermentum than forB. flavum. Plasmid DNA (pCSL17) isolated fromB. flavum transformed recipient McrBC⁺ strains ofEscherichia coli with lower efficiency than McrBC⁻.McrBC digestion of this DNA confirmed thatB. flavum contain methylated cytidines in the target sequence ofMcrBC sequences butB. lactofermentum contained a different methylation pattern. DNA derived from theB. lactofermentum transformed recipient EcoKR⁺ strains ofE. coli with lower efficiency than EcoKR⁻, indicating the presence of methylated adenosines in the target sequence of EcoK sequences. The present data describe the differences in the physical and enzymatic barriers between two species of corynebacteria and also provide some insight into the successful foreign gene expression in corynebacteria.     

Cloning of theHelicobacter pylori recA gene and functional characterization of its product

The RecA protein is a key enzyme involved in DNA recombination in bacteria. Using a polymerase chain reaction (PCR) amplification we cloned arecA homolog fromHelicobacter pylori. The gene revealed an open reading frame (ORF) encoding a putative protein of 37.6 kDa showing closest homology to theCampylobacter jejuni RecA (75.5% identity). A putative ribosome binding site and a near-consensus σ⁷⁰ promoter sequence was found upstream ofrec A. A second ORF, encoding a putative protein with N-terminal sequence homology to prokaryotic and eukaryotic enolases, is located directly downstream ofrecA. Compared to the wild-type strains, isogenicH. pylori recA deletion mutants of strains 69A and NCTC11637 displayed increased sensitivity to ultraviolet light and abolished general homologous recombination. The recombinantH. pylori RecA protein produced inEscherichia coli strain GC6 (recA ⁻) was 38 kDa in size but inactive in DNA repair, whereas the corresponding protein inH. pylori 69A migrated at the greater apparent molecular weight of approx. 40 kDa in SDS-polyacrylamide gels. However, complementation of theH. pylori mutant using the clonedrecA gene on a shuttle vector resulted in a RecA protein of the original size and fully restored the general functions of the enzyme. These data can be best explained by a modification of RecA inH. pylori which is crucial for its function. The potential modification seems not to occur when the protein is produced inE. coli, giving rise to a smaller but inactive protein.     

Clearance of false-positive antigen-antibody reactions of a diagnostic antigen produced inEscherichia coli with human sera

Although many pharmaceutically useful proteins are produced inE. coli expression system, it is very rare for the system to be used in the production of diagnostic antigen due to a major problem,i.e., false-positive reaction ofE. coli host-derived proteins contaminating purified diagnostic antigen with human sera. The N (nucleocapsid) protein of Seoul virus causing haemorrhagic fever with renal syndrome (HFRS) was produced inE. coli BL21 (DE3), and used for the detection of N protein-specific antibodies in human sera. Using the N protein as a diagnostic antigen of HFRS, the false-positive reaction was cleared by merely mixing the test sera with the extract ofE. coli host strain not harboring expression plasmid.     

Site-specific recombination in Escherichia coli between the att sites of plasmid pSE211 from Saccharopolyspora erythraea

Summary pSE211 fromSaccharopolyspora erythraea integrates site-specifically into the chromosome through conservative recombination betweenattP andattB, the plasmid and chromosomal attachment sites. Integration depends on the presence ofint, an open reading frame (ORF) that lies adjacent toattP and encodes the putative integrase. Immediately upstream ofint liesxis (formerly calledorf2) which encodes a basic protein that is thought to exhibit DNA binding.xis andint were cloned in various combinations in pUC18 and expressed constitutively inEscherichia coli from thelac promoter.attP andattB were cloned inStreptomyces orE. coli plasmids containing kanamycin resistance (Km^R) or chloramphenicol resistance (Cm^R) markers. Stable Km^R Cm^R cointegrates formed byattP ×attB orattP ×attP recombination (integration) were obtained inE. coli hosts that expressedint. Co-integrates were not found in hosts expressingint+xis. Excision (intraplasmidatt site recombination) was examined by constructing plasmids carryingattL andattR or twoattP sites separating Cm^R from Km^R and by following segregation of the markers in various hosts. BothattL ×attR andattP ×attP excision depended on bothxis andint inE. coli. pSE211att site integration and excision were not affected by a deletion inhimA, the gene encoding a subunit of integration host factor.     

Doublet preference and gene evolution

Summary Doublet preference analysis was carried out on coding and noncoding regions ofEscherichia coli, Saccharomyces cerevisiae, and human mitochondrial and nuclear DNA. The preference pattern in 1–2 and 2–3 doublets inE. coli andS. cerevisiae correlated with that in noncoding regions. The 3-1 doublet preference inE. coli genes with low optimal codon frequency and inS. cerevisiae genes also showed a correlation with each of their noncoding doublet preference. A mechanism to explain these double preference correlations in doublet preference is presented: mutational biases, the origin of the noncoding region doublet preference, evolved so as to maintain the 1–2 and 2–3 doublet preference, which is determined by codon usage. These biases then acted on the 3-1 doublet, which was almost free of coding constraints, resulting in a similar preference in this doublet.     

Inefficient mismatch repair: genetic defects and down regulation

The mismatch repair system is involved in the maintenance of genomic integrity by editing DNA replication and recombination. However, although most mutations are neutral or deleterious, a mutator phenotype due to an inefficient mismatch repair may generate advantageous variants and may therefore be selected for. We review the evidence for inefficient mismatch repair due either to genetic defects in mismatch repair genes or to physiological conditions. Among natural isolates ofEscherichia coli andSalmonella enterica, about 1% are mutator bacteria, mostly deficient in mismatch repair (most of them defective in themutS gene). Characterization of mutators derived from laboratory strains led also to the isolation of mismatch repair mutants in which the most frequently found defects are inmutL andmutS. The correlation of the size of the antimutator genes with the frequency of their defective alleles amongE. coli andSalmonella strains reveals thatmutU mutants are underrepresented. Analysis of the progeny of a defined M13 phage heteroduplex DNA transfected intoE. coli cells shows that mismatch repair efficiency progressively decreases from the end of the exponential growth in K-12 and is variable among natural isolates. Implications of this defective mismatch repair activity for evolution and tumorigenesis will be discussed.     

Isolation and molecular characterization of theSinorhizobium meliloti bet locus encoding glycine betaine biosynthesis

To cope with osmotic stress,Sinorhizobium meliloti accumulates organic compatible solutes such as glutamate, trehalose, N-acetylglutaminylglutamine amide, and the most potent osmoprotectant glycine betaine. In order to study the regulation of the glycine betaine biosynthetic pathway, a genetic and molecular analysis was performed. We have selected a Tn5 mutant ofS. meliloti which was deficient in choline dehydrogenase activity. The mutation was complemented using a genomic bank ofS. meliloti. Subcloning and DNA sequencing of a 8-6 kb region from the complemented plasmid showed four open reading frames with an original structural organization of thebet locus compared to that described inE. coli. (i) ThebetB and thebetA genes which encode a glycine betaine aldehyde dehydrogenase, and a choline dehydrogenase, respectively, are separated from thebetI gene (regulatory protein) by an additional gene namedbetC. The BetC protein shares about 30% identity with various sulphatases and is involved in the conversion of choline-O-sulphate into choline. Choline-O-sulphate is used as an osmoprotectant, or as a carbon or sulphur source and this utilization is dependent on a functionalbet locus. (ii) No sequence homologous tobetT (encoding a high-affinity choline transport system inE. coli) was found in the vicinity of thebet locus. (iii) ThebetB and thebetA genes, as well as thebetI and thebetC genes are, respectively, separated by 211 and 167 bp sequences containing inverted repeats. Southern blot analysis indicated that thebet locus is located on the chromosome, and not on the megaplasmids.     

Secretory expression in Escherichia coli and Bacillus subtilis of human interferon alpha genes directed by staphylokinase signals

Summary A DNA segment covering the signal sequence coding region, the ribosome binding site, and the promoter of the staphylokinase (sak) 42D gene (Behnke and Gerlach 1987) was cloned into pUC19 to form a portable expression-secretion unit (ESU). Fusion of human interferon α₁ (hIFNα₁) and hybrid hIFNα_1/2 genes to thissak ESU resulted in secretory expression of the two gene products in bothEscherichia coli andBacillus subtilis. While most of the IFNα was exported to the periplasmic space ofE. coli, about 99% was secreted to the culture medium by recombinantB. subtilis strains. The total yield inE. coli was 1.2×10⁵ IU/ml. This level of expression and export led to instability of the recombinant strains that was spontaneously relieved in vivo by inactivation of thesak ESU through insertion of an IS1 element. No such instability was observed withB. subtilis although expression and secretion levels reached even 3×10⁶ IU/ml. Proteolytic degradation of IFNα by extracellular proteases was avoided by a combination of constitutive expression and secretion during the logarithmic growth phase and the use of exoprotease-reduced host strains. The IFNα₁ protein purified fromB. subtilis culture supernatant was correctly processed, carried the expected 11 amino acid N-terminal elongation that resulted from DNA manipulations and proved to be homogenous in Western blotting experiments. The same recombinant plasmid that directed efficient secretion of hIFNα₁ inB. subtilis gave poor yields when introduced intoStreptococcus sanguis.     

AtJ1, a mitochondrial homologue of theEscherichia coli DnaJ protein

The nucleotide sequence of a cDNA clone fromArabidopsis thaliana ecotype Columbia was determined, and the corresponding amino sequence deduced. The open reading frame encodes a protein, AtJ1, of 368 residues with a molecular mass of 41 471 Da and an isoelectric point of 9.2. The predicted sequence contains regions homologous to the J- and cysteine-rich domains ofEscherichia coli DnaJ, but the glycine/phenylalanine-rich region is not present. Based upon Southern analysis,Arabidopsis appears to have a singleatJ1 structural gene. A single species of mRNA, of 1.5 kb, was detected whenArabidopsis poly(A)⁺ RNA was hybridized with theatJ1 cDNA. The function ofatJ1 was tested by complementation of adnaJ deletion mutant ofE. coli, allowing growth in minimal medium at 44°C. The AtJ1 protein was expressed inE. coli as a fusion with the maltose binding protein. This fusion protein was purified by amylose affinity chromatography, then cleaved by digestion with the activated factor X protease. The recombinant AtJ1 protein was purified to electrophoretic homogeneity.In vitro, recombinant AtJ1 stimulated the ATPase activity of bothE. coli DnaK and maize endosperm cytoplasmic Stress70. The deduced amino acid sequence of AtJ1 contains a potential mitochondrial targeting sequence at the N-terminus. Radioactive recombinant AtJ1 was synthesized inE. coli and purified. When the labeled protein was incubated with intact pea cotyledon mitochondria, it was imported and proteolytically processed in a reaction that depended upon an energized mitochondrial membrane.Abbreviations MBP maltose binding protein - PCR polymerase chain reaction - Stress70c the cytosolic member of the 70 kDA family of stress-related proteins     

Immunoblot analysis of the expression of genes for barley rubisco activase inE. coli

Rubisco activity during photosynthesis is regulated by the rubisco activase, which facilitates the dissociation of RuBP and other inhibitory sugar phosphates from the active site of rubisco in an ATP-dependent reaction. In this paper, barleyRca genes (RcaA1,RcaA2 andRcaB) were expressed inE. coli and the activity of rubisco activase expressed was assayed biochemically by chromatography. Then the protein was identified electrophoretically by SDS-PAGE and detected immunologically by Western blot analysis using polyclonal antibodies raised against the kidney bean rubisco activase as probe. The band pattern of purified proteins on the polyacrylamide gel showed two polypeptides of 46 kD and 42 kD. Anti-rubisco activase antibodies reacted specifically with both polypeptides of 46 kD and 42 kD present in the crude extracts ofE. coli transformants. Therefore, it was found that the genes of barley rubisco activase was successfully expressed inE. coli as active forms of 46 kD and 42 kD.