Recs preventing wrecks

The asexual cell cycle of E. coli produces two genetically identical clones of the parental cell through processive, semiconservative replication of the chromosome. When this process is prematurely disrupted by DNA damage, several recF pathway gene products play critical roles processing the arrested replication fork, allowing it to resume and complete its task. In contrast, when E. coli cultures are starved for thymine, these same gene products play a detrimental role, allowing replication to become unregulated and highly recombinagenic, resulting in lethality after prolonged starvation. Here, I briefly review the experimental observations that suggest how RecF maintains replication in the presence of DNA damage and discuss how this function may relate to the events that lead to a loss of viability during thymine starvation.     

Chromosome replication does not trigger cell division in E. coli

An essential part of the chromosome replication origin of E. coli K-12 and B/r was replaced by the plasmid pOU71. The average initiation mass of replication for pOU71 decreases with increasing temperature. The constructed strains were grown exponentially at different temperatures, and cell sizes and DNA content were measured by flow cytometry. The average DNA content increased with increasing temperature, but the cell size distribution was largely unaffected. Furthermore, cells in which DNA replication had not yet initiated (cells in the B period) became less abundant with increasing temperature. The increased DNA content could not be explained by an increase in the length of the C period. It is concluded that chromosome replication does not trigger cell division in E. coli, but that the chromosome replication and cell division cycles of E. coli run in parallel independently of each other.     

Anucleate cell production and surface extension in a temperature-sensitive chromosome initiation mutant of Bacillus subtilis.

At 45 C, in a temperature-sensitive initiation mutant (TsB134) of Bacillus subtilis 168 Thy- tryp-, growing in a glucose-arginine minimal medium, chromosome completion occurred over a period of 80 to 90 min, after which there was no further nuclear division. Normal symmetrical cell divisions continued for a generation afterwards, so that nuclei were segregated into separate cells. During this period asymmetric divisions started to occur. Septa appeared at 25 to 30% from one end of the cell, giving a small anucleate cell and a larger nucleate cell. During inhibition of deoxyribonucleic acid (DNA) synthesis by thymine starvation under the restrictive conditions, asymmetrical division also occurred until there was approximately one nucleus per cell (about one generation time). Asymmetric division, giving anucleate cells, then occurred. Similar results were obtained when DNA synthesis was inhibited by nalidixic acid. After 3 h at 45 C, the rate of anucleate cell production in the presence and absence of thymine was constant at one division per 85 min per chromosome terminus present when DNA synthesis stopped. In the absence of DNA synthesis (during thymine starvation) at 35 C, growth in cell length was linear (i.e., the rate was constant), but at 45 C during thymine starvation the rate gradually increased by more than twofold. It is suggested that this was due to the establishment of new sites of growth associated with anucleate cell production. In the presence of thymine at 45 C, the rate of length extension increased by more than fourfold, which it is suggested was caused by the appearance of new growth zones as a result of chromosome termination and a contribution associated with anucleate cell production. If the mutant was incubated at 45 C for 90 min, both in the presence and absence of thymine, then anucleate cell formation could continue on restoration to 35 C in the absence of thymine...     

DNA replication initiation, doubling of rate of phospholipid synthesis, and cell division in Escherichia coli.

In synchronized culture of Escherichia coli, the specific arrest of phospholipid synthesis (brought about by glycerol starvation in an appropriate mutant) did not affect the rate of ongoing DNA synthesis but prevented the initiation of new rounds. The initiation block did not depend on cell age at the time of glycerol removal, which could be before, during, or after the doubling in the rate of phospholipid synthesis (DROPS) and as little as 10 min before the expected initiation. We conclude that the initiation of DNA replication is not triggered by the preceding DROPS but requires active phospholipid synthesis. Conversely, when DNA replication initiation was specifically blocked in a synchronized culture of a dnaC(Ts) mutant, two additional DROPS were observed, after which phospholipid synthesis continued at a constant rate for at least 60 min. Similarly, when DNA elongation was blocked by thymine starvation of a synchronized culture, one additional DROPS was observed, followed by linear phospholipid accumulation. Control experiments showed that specific inhibition of cell division by ampicillin, heat shock, or induction of the SOS response did not affect phospholipid synthesis, suggesting that the arrest of DROPS observed was due to the DNA replication block. The data are compatible with models in which the DROPS is triggered by an event associated with replication termination or chromosome segregation.     

Inhibition and restart of initiation of chromosome replication: effects on exponentially growing Escherichia coli cells.

Escherichia coli strains in which initiation of chromosome replication could be specifically blocked while other cellular processes continued uninhibited were constructed. Inhibition of replication resulted in a reduced growth rate and in inhibition of cell division after a time period roughly corresponding to the sum of the lengths of the C and D periods. The division inhibition was not mediated by the SOS regulon. The cells became elongated, and a majority contained a centrally located nucleoid with a fully replicated chromosome. The replication block was reversible, and restart of chromosome replication allowed cell division and rapid growth to resume after a time delay. After the resumption, the septum positions were nonrandomly distributed along the length axis of the cells, and a majority of the divisions resulted in at least one newborn cell of normal size and DNA content. With a transient temperature shift, a single synchronous round of chromosome replication and cell division could be induced in the population, making the constructed system useful for studies of cell cycle-specific events. The coordination between chromosome replication, nucleoid segregation, and cell division in E. coli is discussed.     

Cell division after inhibition of chromosome replication in Escherichia coli

The residual cell divisions after thymine starvation of exponential cultures of TJK16, a thymine-requiring derivative of Escherichia coli B/r, were evaluated. The results indicate that under the conditions used (glucose minimal medium 37 °), (1) only cells that had terminated a round of replication divided; (2) once termination had occurred, thymine starvation and replication no longer affected the time of cell division; (3) synchronously terminating subpopulations of cells began to divide about 17 min after termination; after that time, the rate of division decreased exponentially. The results confirm the previously inferred asymmetric distribution of D-periods in an exponential population of E. coli bacteria and suggest that an event associated with termination of replication is required for cell division. The method of data evaluation presented can be used to determine the duration of the D-period and to find the parameter values (halflife and onset) of the stochastic phase of the D-period in exponential cultures, eliminating the need for synchronization procedures.     

Inititation and termination of chromosome replication in Escherichia coli subjected to amino acid starvation.

Initiation and termination of chromosome replication in an Escherichia coli auxotroph subjected to amino acid starvation were examined by following the incorporation of [3H]thymidine into the EcoRI restriction fragments of the chromosome. The pattern of incorporation observed upon restoration of the amino acid showed that starvation blocks the process of initiation prior to deoxyribonucleic acid synthesis within any significant portion of the EcoRI fragment which contains the origin of replication, oriC. In this experiment, no incorporation of [3H]thymidine into EcoRI fragments from the terminus of replication was observed, nor was it found when a dnaC initiation mutant was used to prevent incorporation at the origin which might have obscured labeling of terminus fragments. Thus amino acid starvation does not appear to block replication forks shortly before termination of replication. Attempted synchronization of replication initiation by including a period of thymine starvation subsequent to the amino acid starvation led to simultaneous incorporation of [3H]-thymidine into all EcoRI fragments within the 240-kilobase region that surrounds oriC. It is shown that the thymine starvation step allowed initiation and a variable, but limited, amount of replication to occur.     

Host controlled plasmid replication: Escherichia coli minichromosomes

Escherichia coli minichromosomes are plasmids replicating exclusively from a cloned copy of oriC, the chromosomal origin of replication. They are therefore subject to the same types of replication control as imposed on the chromosome. Unlike natural plasmid replicons, minichromosomes do not adjust their replication rate to the cellular copy number and they do not contain information for active partitioning at cell division. Analysis of mutant strains where minichromosomes cannot be established suggest that their mere existence is dependent on the factors that ensure timely once per cell cycle initiation of replication. These observations indicate that replication initiation in E. coli is normally controlled in such a way that all copies of oriC contained within the cell, chromosomal and minichromosomal, are initiated within a fairly short time interval of the cell cycle. Furthermore, both replication and segregation of the bacterial chromosome seem to be controlled by sequences outside the origin itself.     

Correlation between size and age at different events in the cell division cycle of Escherichia coli.

The variability of (i) the B period between birth and initiation of chromosome replication, (ii) the U period between initiation of chromosome replication and initiation of cell constriction, and (iii) the interdivision period (tau) have been estimated for slowly growing Escherichia coli B/r F. Cultures synchronized by the membrane elution technique were pulse-labeled with [3H]thymidine or continuously labeled with [3H]thymine. After fixation, the pattern of deoxyribonucleic acid replication was analyzed by electron microscopic radioautography. Cell length was found to increase exponentially with age at two different slow growth rates. The coefficient of variation of the B period was estimated to be 60%, that of the U period was 29%, and that of the interdivision period was 12%. From these values and the coefficient of variation of length at different cell cycle events were calculated a negative correlation between the B and U period (r = -0.9) and a positive correlation between length at birth and cell separation (r = 0.6). Initiation of chromosome replication and cell constriction were strictly correlated both with respect to age (r = 0.7) and length (r = 0.8). On the other hand, length at initiation of chromosome replication was distantly correlated with age (r = 0.1) or length at birth (r = 0.3). This low correlation excludes a model in which chromosome initiation is controlled by a random event in the B period. It favors a model in which chromosome initiation occurs at a particular distributed size independent of cell division.     

Morphological analysis of the division cycle of two Escherichia coli substrains during slow growth.

Morphological parameters of the cell division cycle have been examined in Escherichia coli B/r A and K. Whereas the shape factor (length of newborn cell/width) of the two strains was the same at rapid growth (doubling time, tau, less than 60 min), with decreasing growth rate the dimensions of the two strains did change so that B/r A cells became more rounded and B/r K cells became more elongated. The process of visible cell constriction (T period) lasted longer in B/r A than in B/r K during slow growth, reaching at tau = 200 min values of 40 and 17 min, respectively. The time between termination of chromosome replication and cell division (D period) was found to be longer in B/r A than in B/r K. As a result, in either strain completion of chromosome replication seemed always to occur before initiation of cell constriction. Nucleoplasmic separation did not coincide with termination as during rapid growth but occurred in both strains within the T period, about 10 min before cell division.     

Effect of p-Fluorophenylalanine on Chromosome Replication in Escherichia coli

The effect of p-fluorophenylalanine (FPA) on deoxyribonucleic acid (DNA) synthesis and chromosome replication was studied in a thymine-requiring mutant of Escherichia coli. The rate and extent of chromosome replication were followed by labeling the DNA with isotopic thymine and a density marker, bromouracil. The DNA was extracted and analyzed by CsCl gradient centrifugation. The block in chromosome replication caused by high concentrations of FPA occurred at the same point on the chromosome as that caused by amino acid starvation. In a random culture, DNA in cells treated with FPA replicated only slightly slower than the DNA from cells that were not exposed to the analogue. In cultures which had been previously starved for thymine, however, the DNA from the cells treated with FPA showed a marked decrease in the rate and extent of replication. It was concluded that the E. coli cell is most sensitive to FPA when a new cycle of chromosome replication is being initiated at the beginning of the chromosome.     

Changes in cell dimensions during amino acid starvation of Escherichia coli.

Electron microscopic analysis was used to study cells of Escherichia coli B and K-12 during and after amino acid starvation. The results confirmed our previous conclusion that cell division and initiation of DNA replication occur at a smaller cell volume after amino acid starvation. Although during short starvation periods, the number of constricting cells decreased due to residual division, it appears that during prolonged starvation, cells of E. coli B and K-12 were capable of initiating new constrictions. During amino acid starvation, cell diameter decreased significantly. The decrease was reversed only after two generation times after the resumption of protein synthesis and was larger in magnitude than that previously observed before division (F. J. Trueba and C. L. Woldringh, J. Bacteriol. 142:869-878, 1980). This decrease in cell diameter correlates with synchronization of cell division which has been shown to occur after amino acid starvation.     

The E. coli cell cycle and the plasmid R1 replication cycle in the absence of the DnaA protein.

In E. coli strain EC::71CW chromosome replication is under the control of the R1 miniplasmid pOU71. A dnaA850::Tn10 derivative of EC::71CW was viable, which confirmed that R1 can replicate in the absence of the DnaA protein. The frequency of initiation of replication was, however, lowered and cell division was severely disturbed due to underreplication of the chromosome. Both replication and cell division could be restored to normal by increasing the production of RepA, the rate-limiting protein for initiation of replication from the integrated R1 origin. Therefore, the RepA protein seems to compensate for the absence of DnaA in the initiation of replication and assembly of replisomes. The role of the DnaA protein in the initiation of DNA replication, and as an overall regulator of the chromosome replication and cell division cycles of E. coli, is discussed in view of these results.     

Cell Division in Escherichia coli: Analysis of Double Mutants for Filamentation and Abnormal Septation of Deoxyribonucleic Acid-Less Cells

The link between chromosome termination, initiation of cell division, and choice of division sites was studied in Escherichia coli by preparing double mutants. Hybrid mutants containing div52-ts, a cell division initiation mutation, and min, mutations which affect the choice of division sites resulting in the septation of minicells, were characterized. The mutants produced minicells and normal cells coordinately under all conditions studied, although the fraction of minicells is half that of the parental minicell strain. The mutant gradually stopped dividing at both the median and minicell septation sites when transferred from 30 to 41 C in rich medium. A synchronous cell division of filaments was induced 15 min after addition of chloramphenicol to the medium, even at 41 C. Divisions were observed at both normal and minicell sites. These results indicate that div52-ts and min functions share a common step in a cell division pathway. A double mutant containing div52-ts and div27-ts, a dnaB mutant which divides in the absence of DNA synthesis, was characterized. The mutant continues to divide after a shift to the high temperature, although at a reduced rate. The behavior of this hybrid mutant suggests a hypothesis that the chromosome termination signal and div52-ts division initiation signal act on a single membrane site which is altered in div27-ts strains.     

Control of cell length in Bacillus subtilis.

During inhibition of deoxyribonucleic acid synthesis in Bacillus subtilis 168 Thy-minus Tryp-minus, the rate of length extension is constant. A nutritional shift-up during thymine starvation causes an acceleration in the linear rate of length extension. During a nutritional shift-up in the presence of thymine, the rate of length extension gradually increases, reaching a new steady state at about 50 min before the new steady-state rate of cell division is reached. The steady-state rates of nuclear division and length extension are reached at approximately the same time. The ratio of average cell length to numbers of nuclei per cell in exponential cultures is constant over a fourfold range of growth rates. These observations are consistent with: (i) surface growth zones which operate at a constant rate of length extension under any one growth condition, but which operate at an absolute rate proportional to the growth rate of the culture, (ii) a doubling in number of growth zones at nuclear segregation, and (iii) a requirement for deoxyribonucleic acid replication for the doubling in a number of sites.     

Control of F′lac Replication in Escherichia coli B/r

The timing of replication of an F'lac plasmid during the division cycle of Escherichia coli B/r lac(-)/F'lac was examined in relation to the timing of initiation of chromosome replication. This was accomplished by measuring the induction of beta-galactosidase and the incorporation of radioactive thymidine into cells at different ages in cultures growing exponentially at various rates. In cells growing with interdivision times of 27, 36, and 55 min, the F'lac replicated at various stages in the division cycle but always at approximately the same time as initiation of chromosome replication. In cells growing with an interdivision time of 85 min, the F'lac episome replicated midway through the division cycle, whereas chromosome replication initiated at the start of the cycle. Measurements of absorbance at 450 nm per cell suggested that the F'lac replicated when the cells reached a mass which was a constant multiple of the number of episomes per cell at each growth rate. In contrast, the mass per cell at initiation of chromosome replication in cells with an 85-min interdivision time was significantly lower than this constant value. A possible explanation for the apparent coupling between F'lac replication and initiation of chromosome replication at the higher growth rates, and the lack of coupling at the lowest growth rate, is discussed.     

Deoxyribonucleic acid-membrane interactions near the origin of replication and initiation of deoxyribonucleic acid synthesis in Escherichia coli.

A previously reported salt-sensitive binding of deoxyribonucleic acid (DNA) to the cell envelope in Escherichia coli, involving approximately one site per chromosome near the origin of DNA replication, is rapidly disrupted in vivo by rifampin or chloramphenicol treatment and by amino acid starvation. DNA replication still initiates with this origin-specific binding disrupted, even when the disruption extends over the period of obligatory protein and ribonucleic acid synthesis that must precede initiation after release of cells from amino acid starvation. Thus the origin-associated membrane-DNA interaction is not necessary either for the initiation event itself or for the maturation of a putative initiation apparatus in E. coli.     

A calcium flux at the termination of replication triggers cell devision in Escherichia coli : Hypothesis

Cell division in Escherichia coli is coupled to chromosome replication. Even in the absence of known inducible division inhibitors, perturbations of chromosome replication affect cell division. Early studies suggested that a signal at the termination of replication might trigger subsequent division. Although later studies have suggested that fork encounter during termination is an active process involving specific termination sites and the tus protein, the coupling mechanism between termination and cell division remains to be elucidated. Recently it has been shown that the chromosome of a bacterium, Pseudomonas tabaci, contains a high proportion of calcium. E. coli maintains an intracellular concentration of free calcium identical to that of higher organisms and in dividing cells of E. coli a twenty-fold increase in the level of total calcium in the cytoplasm, a flux, occurs. In this article I propose that during the replication of the chromosome calcium entry balances calcium binding to DNA. At the termination of replication, there is a brief interval between the end of calcium binding to the chromosome and the end of calcium entry or release into the cytoplasm. During this interval the level of free calcium therefore rises. This rise may result in the observed flux by triggering the entry of calcium directly via voltage-gated calcium channels or indirectly via changes in phospholipid configurations. Mechanisms whereby these changes in calcium levels might be coupled to cell division and to a phospholipid control of the cell cycle are discussed.     

Initiation of Deoxyribonucleic Acid Synthesis After Thymine Starvation of Bacillus subtilis

Evidence for premature initiation of deoxyribonucleic acid (DNA) replication after thymine starvation of Bacillus subtilis W23T(-) is presented, based on (i) increase in the number of ade(+) relative to met(+) transformants yielded by the DNA isolated from cultures after starvation (the ade(-) marker being near the origin of replication, whereas met(-) is close to the terminus), and (ii) increase in both the initial rate and final level of tritiated thymine incorporation in the presence of chloramphenicol after release from starvation. The marker ratio data agree quantitatively with the hypothesis that the initiation is induced only on one arm of each chromosome which was replicating prior to starvation.