Activity in vitro of three replication origins of the antibiotic resistance plasmid RSF1040

Replicating molecules of plasmid RSF1040, a deletion mutant of R6K, were synthesized in vitro and analyzed by electron microscopy. Initiation of replication occurs at three unique sites, ori alpha, ori beta, and ori gamma, within a 3900-base pair segment of the R6K genome. These sites are indistinguishable from the origins that are active in vivo. Frequencies of initiation at these three origins, however, are different from those observed in vivo. Replication proceeds unidirectionally in either direction from ori beta and ori gamma and in one direction from ori alpha. The replication terminus of the R6K genome is inactive in the in vitro system.     

Extension of the host range of Escherichia coli vectors by incorporation of RSF1010 replication and mobilization functions.

The broad-host-range vectors pSUP104, pSUP106, pSUP204, pSUP304, and pSUP404 are based on conventional Escherichia coli vectors (such as pBR325 and pACYC184) which have been modified to include the mobilization and broad-host-range replication functions of the IncQ plasmid RSF1010. These vector plasmids now can be maintained in a wide range of bacterial genera including Rhizobium, Agrobacterium, and Pseudomonas. They are efficiently mobilized by RP4 and thus are of particular interest for bacteria refractory to transformation. They offer the selection markers and cloning sites characteristic of the basic E. coli vectors. Therefore, they can be applied and adapted to a variety of cloning strategies. However, the cloning of very large fragments (e.g., in cosmid hybrids of pSUP106) was found to affect the stability of the recombinant molecules in a Rec+ background. This instability was not observed with smaller inserts of about 5 kilobases.     

Replication of the nonconjugative plasmid RSF1010 in Escherichia coli K-12.

Replicating DNA molecules of the nonconjugative R plasmid RSF1010 (Smr Sur) were cleaved with the EcoRI restriction endonuclease and examined with the electron microscope. Results of this analysis indicated that replication is initiated from an origin located at about 19% of total genome size from one of the EcoRI ends. Replication proceeded either unidirectionally or bidirectionally with equal frequency. Results of the analysis of replicative intermediates of RSF1010 containing the Apr-transposable sequence (Tn) are also presented.     

TnA-directed deletion of the trp operon from RSF2124-trp in Escherichia coli

Plasmid pMT-trp was constructed by digestion of RSF2124-trp with restriction endonuclease PstI and ligation with T4 ligase. In pMT-trp about 78% of the DNA of transposon TnA from RSF2124-trp was deleted, and hence the gene for ampicillin resistance was lost. All Trp- segregants from pMT-trp carriers in Escherichia coli W3110 and its derivatives were found to have lost the entire plasmid. On the other hand, deletion plasmids which had lost the trp operon were found among Trp- segregants from RSF2124-trp carriers, particularly from the mutant strain trpAE1 trpR tnaA. The experimental fact that deletion occurred exclusively in RSF2124-trp suggests that the presence of TnA in the plasmid (RSF2124-trp) was responsible for the deletion.     

Colloidal clay inhibits conjugal transfer of R-plasmid R1drd-19 in Escherichia coli.

Colloidal clay (montmorillonite) strongly inhibited the conjugal transfer of R-plasmid R1drd-19 in Escherichia coli. This finding is discussed in the context of the feasibility of plasmid transfer in environmental waters.     

Control of cyclic chromosome replication in Escherichia coli.

The biochemical basis for cyclic initiation of bacterial chromosome replication is reviewed to define the processes involved and to focus on the putative oscillator mechanism which generates the replication clock. The properties required for a functional oscillator are defined, and their implications are discussed. We show that positive control models, but not negative ones, can explain cyclic initiation. In particular, the widely accepted idea that DnaA protein controls the timing of initiation is examined in detail. Our analysis indicates that DnaA protein is not involved in the oscillator mechanism. We conclude that the generations of a single leading to cyclic initiation is separate from the initiation process itself and propose a heuristic model to focus attention on possible oscillator mechanisms.     

Expression of R-plasmid functions during anaerobic growth of an Escherichia coli K-12 host.

The normal habitat of enteric bacteria is largely anaerobic. Expression of the three characteristic properties of R-plasmids, drug resistance, vegetative replication, and fertility, was therefore studied in Escherichia coli K-12 during anaerobiosis. Replication and drug resistance functions were not altered in the 45 R-plasmids studies, whereas the expression of fertility varied considerably among different R-plasmids during anaerobiosis. The R-plasmids could be divided into three groups, one showing a strong, the second a moderate, and the third little or no reduction of fertility by anaerobiosis. Plasmid-determined sensitivity to F-, N-, and I-specific phage, respectively, was well, although not absolutely, correlated with each of three groups mentioned. Anaerobiosis-aerobiosis appears to change the fertility of type F R-plasmids by influencing the degree of repression of their fertility functions such as the formation of sex pili. Although the minimum inhibitory concentrations of ampicillin, chloramphenicol, streptomycin, and tetracycline were unaltered by anaerobiosis, sulfonamide was found to be four- to eightfold less active under this condition in both resistant and sensitive strains. A surprisingly high frequency and uniformity of minimal inhibitory concentrations was observed for R-plasmid-mediated resistance to streptomycin and chloramphenicol.     

Map location of the Escherichia coli origin of replication.

Summary Working with restriction fragments obtained directly from the Escherichia coli K12 chromosome, the EcoRI-HindIII restriction map of the section of the chromosome containing the replication origin has been extended by 14 kilobase pairs (kb) to cover 56kb. Within this newly mapped portion, the liv and rrnC cistrons have been identified by (1) hybridization of individual restriction fragmenents to the ilv-transducing phage dilv5 and (2) a comparison of the restriction map of this region with the EcoRI map of dilv5 and the HindIII map of the plasmid pJC110, a ColE1-ilv hybrid. The replication origin is located approximately 30 kb from the ilvE gene and 20 kb from the rrnC 16S rRNA cistron. This places the origin near 82.7 min on the genetic map, close to uncA.     

Nonrandom F-plasmid replication in Escherichia coli K-12.

Both the autonomous and chromosomally integrated F plasmids were found to replicate in a nonrandom fashion after a density transfer from heavy medium ([13C]glucose, 15NH4) to light medium ([12C]glucose, 14NH4). The data are consistent with the hypothesis that both the chromosome and the F plasmid are replicated in a cell cycle-specific manner. Thus, these data support the proposal (J. D. Keasling, B. O. Palsson, and S. Cooper, J. Bacteriol. 173:2673-2680, 1991) that plasmids replicate in a cell cycle-specific manner.     

Nonrandom minichromosome replication in Escherichia coli K-12.

The intervals between rounds of chromosome and minichromosome replication were measured by density shift experiments and found to be similar. Thus the minichromosome, a lambda asnA oriC bacteriophage, mostly replicates once each division cycle rather than randomly, despite its high copy number. Slight differences between the chromosome and the oriC plasmid are explained.     

Nucleotide sequence of the Escherichia coli replication gene dnaZX.

The Escherichia coli 2.2 kilobase dnaZX region contains one 1929 nucleotide reading frame which directs the synthesis of two protein products involved in DNA polymerization. The larger consists of 643 amino acids in a deduced 71,114 dalton chain which could be the tau subunit of DNA polymerase III. The smaller, the DNA polymerase III gamma subunit, is encoded by the same reading frame as the larger. The dnaZX sequence contains a region homologous to ATP binding sites, suggesting that these replication factors are adenine nucleotide binding proteins.     

R-plasmid transfer frequencies from environmental isolates of Escherichia coli to laboratory and fecal strains.

Multiple-drug-resistant strains of Escherichia coli were isolated from the water at an estuarine site. They represented about 8.3% of the total E. coli population. Fifty-five strains, representing each of the 32 resistance patterns identified, were mated with an E. coli K-12 F- strain. Matings were performed on membrane filters, and the cells were washed to remove any colicins produced by the donors. Thirty-one strains, about 5% of the mean E. coli density in the samples, transferred drug resistance and, hence, posessed conjugative R plasmids. Of these, 80% transferred drug resistance at a frequency of about 10(-4) or less. Nine environmental R+ strains were mated with three fecal recipients. The R-plasmid transfer frequencies to the fecal strains from the environmental donors correlated well with those from a derepressed K-12 R+ laboratory donor. The R+ X K-12 F- lac- transconjugants from 16 environmental strains were "backcrossed" to a lac+ K-12 F- strain. All transfer frequencies were higher in the backcrosses than in the original matings from the environmental donor. Furthermore, 7 of 13 different transconjugants, which accepted plasmids at repressed frequencies of less than 10(-3), donated them at frequencies greater than 10(-2). This suggests that these were derepressed plasmids in a repressed host.     

On the termination of the DNA replication cycle in Escherichia coli.

The initiation of the DNA replication cycle in Escherichia coli requires protein synthesis. Marunouchi &; Messer (1973) have hypothesized that an additional protein synthesis step is required for the replication of the terminal segment of the chromosome, and that replication of this segment is a prerequisite for subsequent cell division. We have not confirmed the existence of a unique terminal segment using a protocol designed to label the hypothesized segment with [³H]dThd². Our protocol avoids the increased incorporation of [³H]dThd into DNA caused by abrupt increases in temperature, a complication implicit in the technique of Marunouchi &; Messer (1973).Treatment with nalidixic acid (an inhibitor of semiconservative DNA synthesis) in sufficient concentration to prevent replication of the postulated terminal segment prevents cell division but also causes loss of viability. This makes it difficult to correlate the effect of nalidixic acid on cell division with DNA synthesis inhibition alone.     

Perturbed chromosomal replication in recA mutants of Escherichia coli.

When initiation of DNA replication is inhibited in wild-type Escherichia coli cells by rifampin or chloramphenicol, completion of ongoing rounds of replication (runout of replication) leads to cells containing two, four, or eight fully replicated chromosomes, as measured by flow cytometry. In recombination-deficient recA strains, a high frequency of cells with three, five, six, or seven fully replicated chromosomes was observed in addition to cells with two, four, or eight chromosomes. recA mutants affected only in the protease-stimulating function behaved like wild-type cells. Thus, in the absence of the recombinase function of RecA protein, the frequency of productive initiations was significantly reduced compared with that in its presence. DNA degradation during runout of replication in the presence of rifampin was about 15%. The DNA degradation necessary to account for the whole effect described above was in this range or even lower. However, a model involving selective and complete degradation of partially replicated chromosomes is considered unlikely. It is suggested that the lack of RecA protein causes initiations or newly formed replication forks to stall but remain reactivatable for a period of time by functional RecA protein.     

The amino acid sequence of the trimethoprim-resistant dihydrofolate reductase specified in Escherichia coli by R-plasmid R67.

The amino acid sequence of a trimethoprim-resistant dihydrofolate reductase (EC 1.5.1.3) specified by the R-plasmid R67 is described. The sequence was deduced from automatic and manual sequence analysis of the intact protein, the fragments produced by cyanogen bromide cleavage, and peptides derived from the largest cyanogen bromide fragment by digestion with trypsin, Staphylococcus aureus V8 proteus, chymotrypsin, and Lysobacter enzymogenes alpha-lytic protease. The complete sequence comprises 78 residues in a single polypeptide chain of molecular weight 8444. No evidence of heterogeneity was obtained, indicating that all subunits of the native enzyme are identical. Comparison of the sequence with that of all known dihydrofolate reductases shows no significant sequence homology.