Logo of embojLink to Publisher's site
EMBO J. Jul 11, 2007; 26(13): 3124–3131.
Published online Jun 7, 2007. doi:  10.1038/sj.emboj.7601747
PMCID: PMC1914095

The two chromosomes of Vibrio cholerae are initiated at different time points in the cell cycle


The bacterium Vibrio cholerae, the cause of the diarrhoeal disease cholera, has its genome divided between two chromosomes, a feature uncommon for bacteria. The two chromosomes are of different sizes and different initiator molecules control their replication independently. Using novel methods for analysing flow cytometry data and marker frequency analysis, we show that the small chromosome II is replicated late in the C period of the cell cycle, where most of chromosome I has been replicated. Owing to the delay in initiation of chromosome II, the two chromosomes terminate replication at approximately the same time and the average number of replication origins per cell is higher for chromosome I than for chromosome II. Analysis of cell-cycle parameters shows that chromosome replication and segregation is exceptionally fast in V. cholerae. The divided genome and delayed replication of chromosome II may reduce the metabolic burden and complexity of chromosome replication by postponing DNA synthesis to the last part of the cell cycle and reducing the need for overlapping replication cycles during rapid proliferation.

Keywords: C period, divided genomes, flow cytometry, initiation of DNA replication, replication time


Most bacteria contain only one circular chromosome and bidirectional replication is initiated at a single origin. However, a few bacteria contain a genome distributed on multiple chromosomes and some have unusual genome structures like linear chromosomes. What are the evolutionary advantages of dividing the genome into several chromosomes, when most bacteria have not evolved so? Multiple chromosomes are found in diverse prokaryotic phyla including the Deinococcus–Thermus, Spirochaetes and the α-, γ- and β-subgroups of Proteobacteria. Interestingly, most of the bacteria with multiple chromosomes have several different life cycles and interact with host organisms, like Vibrio cholerae that lives in brackish water and within human hosts, making it likely that dividing the genome into several different replicons may be favourable for organisms living under variable environmental conditions (Egan et al, 2005). The genome of V. cholerae is distributed onto two circular chromosomes (Trucksis et al, 1998; Heidelberg et al, 2000), with a large 2.96 Mb chromosome I and a small 1.07 Mb chromosome II. This division is seen in other Vibrio bacteria as well, but the relative sizes of the secondary chromosomes vary significantly between different Vibrio species (Okada et al, 2005). The splitting of the V. cholerae genome onto two chromosomes could facilitate an up- or downregulation of genes present on one of the chromosomes by an amplification or elimination of one of the chromosomes under certain environmental conditions (Heidelberg et al, 2000). A study of gene expression in V. cholerae shows a preferential increase in the transcription of genes from the small chromosome II during colon colonisation compared to cells growing in vitro under aerobic conditions (Xu et al, 2003). Thus, possessing multiple chromosomes may give an additional mechanism for switching between different portfolios of genes needed in various environments.

Bacteria tightly regulate genome replication and this process is primarily controlled at the level of initiation of chromosomal DNA replication. The presence of several chromosomes makes the coordination of genome replication more difficult compared to when only a single chromosome is present. Some bacteria grow very fast under optimal conditions, with a doubling time much shorter than the time required for DNA replication and segregation, and therefore have overlapping DNA replication cycles. V. cholerae cells are able to divide every 18 min under optimal conditions. How do bacteria with multiple chromosomes and fast cell cycles ensure the faithful replication and distribution of several chromosomes to daughter cells?

Most of the experimental data on chromosome replication and segregation in bacteria with divided genomes have been obtained using V. cholerae as model organism. A Meselson–Stahl density shift experiment showed that both V. cholerae chromosomes are replicated only once per cell cycle and with an inter-replication time equal to the doubling time of the cells (Egan et al, 2004). Thus, the replication of both chromosomes in V. cholerae are linked to the cell cycle (Egan et al, 2005). In contrast, plasmid replication is normally not tied to the cell cycle; plasmids tend to replicate at random time points during the cell cycle. Flow cytometry analysis of samples of slow growing cells treated with rifampicin and cephalexin indicated that the two chromosomes are initiated at the same time in the cell cycle (Egan et al, 2004). Examination of the replication control systems of the two V. cholerae chromosomes showed that each chromosome is controlled by its own initiator molecule: replication of chromosome I is controlled by the DnaA protein, similarly to the control of chromosome replication in Escherichia coli (Egan and Waldor, 2003). Increased expression of the DnaA protein stimulates initiation of chromosome I, but not of chromosome II (Duigou et al, 2006). Chromosome II has a plasmid-like origin, its replication is controlled by the RctB protein and requires the DnaA protein (Egan and Waldor, 2003). P1 and F plasmids shows a similar requirement for DnaA protein (Hansen and Yarmolinsky, 1986) despite that replication is controlled by the respective replication proteins RepA and RepE. The role of DnaA protein in these and other plasmids is strand opening, helicase loading or an architectural role in forming the replication complex (reviewed by Messer, 2002) and a similar role for the DnaA protein in chromosome II replication is probable. The RctB protein is autoregulated and the activity of RctB is further controlled by the rctA titration sites. The titration activity of rctA is controlled by the RctB protein (Venkova-Canova et al, 2006). Induction of the RctB protein stimulates initiation of chromosome II, but not chromosome I (Duigou et al, 2006). The separate control of replication initiation of the two V. cholerae chromosomes makes it difficult to understand how initiation synchrony can be obtained in this organism. This led us to re-evaluate the finding of initiation synchrony by Egan et al (2004). Using novel flow cytometry methods and marker frequency analysis, we show that initiation of the smaller chromosome II is delayed compared to initiation of chromosome I, resulting in completion of replication of the two chromosomes at approximately the same time in the cell cycle.

Results and discussion

Models for coordinating replication of the two chromosomes in V. cholerae

The cell cycle of slowly growing bacteria is divided into three time periods with respect to chromosomal DNA replication: B is the period from cell birth to initiation of chromosome replication, C is the period from initiation until termination of chromosome replication and D is the period between termination and cell division (Figure 1). Thus, cells in the B period are small and contain one genome equivalent of DNA, cells in the C period are intermediate in size and contain between one and two genome equivalents of DNA and cells in the D period are large and contain two genome equivalents of DNA.

Figure 1
Models of the DNA replication patterns during the V. cholerae cell cycle assuming either initiation or termination synchrony. Shown are chromosome I (large circles) and chromosome II (small circles). Origins are marked with filled circles and termini ...

Both of the chromosomes in V. cholerae are replicated once and only once during the cell cycle and replication run-off experiments support an ‘initiation synchrony' model (see Figure 1), where the two origins are initiated at the same time and DNA replication proceeds bidirectionally on both chromosomes until the terminus of each chromosome is reached (Egan et al, 2004). Owing to the smaller size, replication of chromosome II will terminate earlier than replication of chromosome I. According to the initiation synchrony model, the number of oriCI will equal the number of oriCII, whereas the number of terI will be lower than the number of terII. The total rate of DNA accumulation will be highest at the onset of the C period when both chromosomes are replicating and decrease when replication of chromosome II has terminated.

Alternatively, initiation of the smaller chromosome II could take place at a later point in the cell cycle. In the ‘termination synchrony' model (see Figure 1), chromosome II is initiated when the size of the non-replicated part of chromosome I is roughly equivalent to the size of chromosome II. Replication of both chromosomes then proceeds until the replication forks reach the termini at approximately the same time in both chromosomes. According to the termination synchrony model, the number of terI will equal the number of terII, whereas the number of oriCI will be higher than the number of oriCII. The rate of DNA accumulation will be highest late in the C period when both chromosomes are replicating.

The DNA replication rate increases abruptly during the C period

To determine which of the two models in Figure 1 that best describes the timing of replication of the two chromosomes in V. cholerae, we measured the DNA accumulation rate as a function of cell age. The DNA content and cell size of exponentially growing cells were measured using flow cytometry and the data are visualised by the 3-dimensional (3D) presentations in Figure 2A. The ridges of the 3D distributions were computationally determined as described in Materials and methods. The graph for exponentially growing Caulobacter crescentus cells, a one-chromosome bacteria with distinct B, C and D periods, is shown in Figure 2A. The first horizontal part of the ridge graph represents the B period, where only cell size increases. This is followed by the C period where both cell size and DNA content gradually increase, and the final horizontal line is the D period. Similar results were obtained with an E. coli strain (Supplementary Figure S1). When the same analysis is performed on an exponentially growing V. cholerae culture, the graph still displays B, C and D periods, but a bend is evident in the C period part of the graph (Figure 2A and Supplementary Figure S1). Since the cell size most likely increases gradually during the cell cycle, this bend must correspond to a change in the rate of DNA synthesis (Figure 2B). The slope of the ridge plot increases in the last part of the C period, indicating that V. cholerae cells abruptly increase their rate of DNA synthesis late in the C period. We repeated this experiment with varying carbon sources to change the growth rate (Table I) and in all cases we observed a very similar bend of the ridge representing the cells in the C period (Supplementary Figure S1).

Figure 2
Flow cytometry analysis of chromosome replication. (A) 3D flow cytometry analysis of DNA replication patterns in C. crescentus and V. cholerae. The left panels show the raw 3D flow cytometry data with cell size on the x-axis, the DNA contents on the ...
Table 1
Estimates of the B, C and D periods of V. cholerae cells growing at moderate growth rates

Simultaneous initiation of the two chromosomes would have led to a high rate of DNA replication at the onset of the C period and a decreased rate of DNA replication after termination of chromosome II (Figures 1 and and2B,2B, Initiation synchrony) in contradiction with our results. In contrast, the increased rate of DNA replication late in the C period corresponds well with later initiation of chromosome II than chromosome I, leading to a doubling of the number of active DNA replication forks late in the C period (Figures 1 and and2B,2B, Termination synchrony).

Computer simulations of DNA contents distributions

To support our conclusion that chromosome II replicate during the late part of the C phase and to determine basic cell-cycle parameters of V. cholerae, we compared the experimental DNA histograms obtained by flow cytometry of exponentially growing cells to computer simulations of DNA contents in ideal cultures using the approach described by Michelsen et al (2003) that was modified to accommodate two-chromosome bacteria. In these simulations, the DNA histograms are resolved into the contributions from cells in the B, C and D periods by varying the basic cell-cycle parameters and the time of initiation of chromosome II. We simulated the DNA contents assuming that replication of the two chromosomes either initiate or terminate simultaneously. We only obtained good fits between experimental and simulated DNA histograms when the two chromosomes terminate at approximately the same time in the computer simulations (Figure 2C). Similar results were obtained when V. cholerae cells were grown in other media, resulting in different doubling times (Supplementary Figure S2).

The fitting of simulations of the DNA histograms to experimental data gave estimates of the basic cell-cycle parameters of V. cholerae cells growing at medium and slow growth rates (Table I). In cells growing at a moderate rate (42–55 min doubling time), the replication time for chromosome I was nearly 30 min (27–32 min) and the replication time for chromosome II was nearly 10 min (9–11 min), giving a replication speed of approximately 800 bp per second per replication fork, very similar to the replication speed in E. coli when grown at the same temperature (37°C). The D period was 12–14 min, which is similar to that observed in E. coli B/r cells and shorter than that observed in E. coli K-12 cells at similar growth rates (Michelsen et al, 2003).

The oriCI to oriCII ratio is increased in fast-growing V. cholerae cells and is growth rate dependent

The different sizes of the two chromosomes will lead to different copy numbers of either the origins or the termini of the two chromosomes (Figure 1). The initiation synchrony model predicts an oriCI to oriCII ratio of one and a terI to terII ratio less than one (Figure 1). The termination synchrony model predicts an oriCI to oriCII ratio higher than one and a terI to terII ratio of one (Figure 1). The differences in ter and oriC ratios, respectively, are most pronounced during rapid growth, where a larger fraction of the cells will be in the part of the cell cycle where chromosome I is under replication and replication of chromosome II is either finished or not initiated yet. This difference is modelled as described in Materials and methods and shown in Figure 3B and C.

Figure 3
Relative marker frequencies (oriCI/oriCII and terI/terII ratios) at different growth rates. (A) Wild-type V. cholerae cells were grown in media containing different carbon sources, resulting in a range of doubling times. Cells from exponentially growing ...

We measured the ratios of oriCI to oriCII and the ratios of terI to terII in V. cholerae cells where the doubling time was modulated between 18 and 164 min by varying the carbon source and the supplements of amino acids in the culture medium (Figure 3A). The oriCI to oriCII ratio dropped from ~2 at fast growth rates to just above one at doubling times longer than 100 min, whereas the terI to terII ratio was almost invariant and near 1 for all doubling times. The results from the marker frequency analysis fit a termination synchrony model significantly better than it fit an initiation synchrony model (Figure 3).

Initiation of replication at oriCI and oriCII responds differently to the transcription inhibitor rifampicin

Our finding of later initiation at oriCII than at oriCI contradicts with the initiation synchrony observed by Egan et al (2004). Their observation was based on flow cytometry analysis of slowly growing V. cholerae cells followed by treatment with rifampicin and cephalexin. Rifampicin inhibits transcription, which then inhibits initiation of DNA replication, and cephalexin inhibits cell division. If chromosome II is initiated later than chromosome I, it was expected to observe cells that had started replication of chromosome I but not replication of chromosome II in the rifampicin run-off experiments. These cells should contain approximately 7.2 Mb DNA after rifampicin incubation, corresponding to two copies of chromosome I and one copy of chromosome II. However, Egan et al (2004) observed that all cells contained either 4.1 or 8.3 Mb DNA after rifampicin treatment, corresponding to either one or two copies of each chromosome and they therefore concluded that the two chromosomes are initiated synchronously.

To resolve this contradiction, we measured the ratio of oriCI to oriCII in very fast-growing V. cholerae cells (doubling time of 18 min) at different time points after addition of rifampicin (Figure 4). We found that the oriCI to oriCII ratio was ~2.6 before addition of rifampicin as expected from the experiment in Figure 3. After addition of rifampicin, this ratio dropped to near 1 after 10 min of incubation with rifampicin. This indicates that initiation at oriCII continued for a longer period after addition of rifampicin than initiation of chromosome I, until the two chromosomes had the same copy number. A double-sided t-test shows that the means of the 1 and 30 min points are different with 99% confidence. Similarly, the ratio of oriCI to oriCII in V. cholerae cells growing with doubling time of 55 min dropped from 1.4 to 1.0 after addition of rifampicin (data not shown). These results can only be explained if rifampicin has a delayed effect on initiation of chromosome II or if the chromosome I DNA is degraded during the rifampicin incubation. This second option is highly unlikely as the raw data show a constant amount of oriCI and an increasing amount oriCII. We therefore conclude that the initiations at the two origins are not equally sensitive to rifampicin, which explains the discrepancy between our results and the finding of initiation synchrony by Egan et al (2004).

Figure 4
Differential effect of rifampicin addition on the two V. cholerae chromosomes. The oriCI/oriCII ratios at various time points after addition of rifampicin to a culture growing in LB glucose media (τ=18 min). The measurements were repeated at least ...

Bremer and Churchward (1977b) measured a delay of 6–11 min before the inhibition of initiation of E. coli chromosomal DNA replication by either rifampicin (inhibiting transcription) or chloramphenicol (inhibiting translation) was complete, indicating that cells near initiation can complete this process in the presence of the drugs. Further, there are several demonstrations of rifampicin-resistant initiation in E. coli strains where DnaA is in surplus (Pierucci et al, 1987; Atlung and Hansen, 1993; von Freiesleben et al, 2000; Morigen et al, 2005), indicating that the primary action of rifampicin in chromosome replication is the inhibition of de novo DnaA protein synthesis. As initiation at oriCI is controlled by the activity of the DnaA protein and initiation at oriCII is controlled by the activity of the RctB protein (Duigou et al, 2006), initiation at both origins is expected to require ongoing protein synthesis and therefore they should both be sensitive to rifampicin. Our demonstration that rifampicin inhibition is delayed for oriCII may indicate that the RctB protein bound to the rctA titration sites can be released during transcription arrest and initiate replication.

Mechanisms ensuring approximately simultaneous termination of the two chromosomes

A mechanistically simple mechanism for ensuring approximately simultaneous termination of the two chromosomes would be to link initiation of both chromosomes to cell mass, but with a different initiation mass for the two chromosomes. The two chromosomes are independently controlled by distinct replication initiators, DnaA for chromosome I and RctB for chromosome II (Duigou et al, 2006). The activity of RctB is controlled in much of the same fashion as the activity of DnaA is controlled in E. coli with both initiator autorepression and titration (Pal et al, 2005; Egan et al, 2006; Venkova-Canova et al, 2006). Thus, it is likely that chromosome II initiation is linked to cell mass in a similar fashion as the chromosome of E. coli (Donachie, 1968) or to a lesser degree plasmid P1 (Keasling et al, 1992; Bogan et al, 2001).

Another possible molecular mechanism ensuring termination of the two chromosomes at approximately the same time during the cell cycle could be the presence of a timer on chromosome I that when replicated sends a signal to chromosome II. In other bacteria, a change in methylation status is known to effect transcription of a number of genes. In C. crescentus, the passage of the replication fork leads to the presence of hemimethylated DNA, which activates the promoter of the cell-cycle control gene ctrA (Reisenauer and Shapiro, 2002; reviewed by Wion and Casadesus, 2006). Dam methylation could have a similar function at an unknown activator gene in V. cholerae.

Concluding remarks

In the current work, we have determined the replication timing for both of the chromosomes in V. cholerae by flow cytometry analysis of samples harvested from exponentially growing cultures. Two different data analysis methods led to the conclusion that the rate of DNA replication increases late in the C period in congruence with a model where approximately two-thirds of the large chromosome I is replicated before the smaller chromosome II is initiated, leading to approximately simultaneous termination of replication of the two chromosomes. We have further tested this conclusion independently by measuring the ratio of the copy numbers of the two chromosomal origins, oriCI to oriCII, and the ratio of the two termini, terI to terII, as a function of the doubling time. These data show that the origin ratio varies with growth rate, whereas the termini ratio is invariant. This finding further supports that replication of the two chromosomes is initiated at different points of the cell cycle, whereas termination of the two chromosomes happens at approximately the same time in the cell cycle. Our finding of approximate termination synchrony conflicts with the finding of initiation synchrony by Egan et al (2004) based on flow cytometry analysis of a rifampicin-treated sample. We show here that it takes longer time before rifampicin inhibits initiation at oriCII than at oriCI and that the copy number of the two chromosomes will be the same after a long rifampicin treatment, explaining the discrepancy between the results.

Fogel and Waldor (2005) studied the intracellular locations of the two origins in fast-growing cells and found that 78% of the cells had two oriCI foci and only 30% of the cells had two oriCII foci. Consistent with this, Fiebig et al (2006) found that mainly mid-sized cells have two oriCI foci and one oriCII focus, long cells had two foci of each origin and short cells had one focus of each origin. The finding of fewer oriCII foci than expected from the initiation synchrony model was explained by much earlier segregation of oriCI than of oriCII during the cell cycle. Our finding of later initiation at oriCII than at oriCI and the subsequently higher copy number of oriCI compared to oriCII is consistent with the origin localisation data. Thus, both oriCI and oriCII foci are likely to segregate away from each other shortly after initiation.

Srivastava et al (2006) studied the intracellular locations of the two termini in cells growing in either L broth or minimal glucose medium. They observed that most cells had only one focus of each terminus and that these foci colocalised in these cell. Only 2–3% of the cells had two terI foci and 17–18% of the cells had two terII foci, indicating that terII foci separates earlier than terI foci. With the age distribution (Powell, 1958), the cell age at terI and terII foci separation can be calculated to be later in the cell cycle than the termination of chromosome replication (calculations not shown). The earlier segregation of terII is therefore not in conflict with an approximately termination synchrony of the two chromosomes.

Evolutionary advantages of having a divided genome

Over-initiation in E. coli leads to a decreased replication speed (Atlung et al, 1987) and under-initiation results in an increased DNA replication rate (Atlung and Hansen, 2002; Morigen et al, 2003; Skovgaard and Løbner-Olesen, 2005), demonstrating that the normal replication rate is limited by the supply of energy, nucleotides, replication factors or other components. By delaying initiation at oriCII until as late as possible in the cell cycle, the cell postpones DNA synthesis until it has increased its metabolic capacity for providing the required energy, dNTPs and/or replication factors, potentially making replication more efficient.

The simulations of DNA content distributions provide estimates of the cell-cycle parameters for V. cholerae (Figure 2C and Table I and Supplementary Figure S2). For the fastest growing cells that can be analysed using this approach (42 min doubling time), a C period of approximately 27 min and a D period of approximately 13 min was observed (Table I). Since there is a short B period at a doubling time of 42 min, the minimal C+D period is 40 min or a little shorter. This is very fast compared to E. coli, in which the minimum C+D period is 50–60 min (Cooper and Helmstetter, 1968; Michelsen et al, 2003). Bacteria growing with a doubling time shorter than the minimal C+D period initiate replication in the previous cell cycle, so cells are born with a partially duplicated genome. Division of the genome into several replicons shortens the C+D period, and thereby reduces the number of overlapping replication cycles needed during fast growth. When grown in rich media, V. cholerae only need two overlapping replication cycles of chromosome I and no overlapping replication cycles of chromosome II, whereas E. coli need three overlapping rounds of replication. This reduces the number of oriCI copies in newborn fast-growing V. cholerae cells to two, compared to four oriC copies in newborn fast-growing E. coli cells. This also reduces the number of replication forks from fourteen to six (eight when both chromosome I and II replicate) after initiation of a new round of DNA replication. The lower number of overlapping replication cycles eases the regulatory problem of ensuring initiation at multiple identical origins within a very short time period (the initiation cascade; Løbner-Olesen et al, 1994) and makes chromosome segregation less complicated by reducing the complexity of properly localising and detangling branched, partially replicated chromosomes during rapid growth.

Materials and methods

Bacterial strains and growth conditions

The V. cholerae strain used was AC-V304 (Hava and Camilli, 2001). The cells were grown in AB minimal medium (Clark and Maaløe, 1967) supplemented with different carbon sources: 0.2% glutamate; 0.2% glycerol; 0.2% glycerol, 100 μg/ml serine; 0.2% fructose; 0.2% maltose; 0.2% sucrose; 0.2% glucose; or 0.2% glucose, 100 μg/ml serine, 0.5% casamino acids. Alternatively, cells were grown in LB media with 0.2% glucose added. The E. coli strain FH1218 (Løbner-Olesen et al, 1989) was grown in AB minimal medium containing 0.2% glycerol, 25 μg/ml histidine, 25 μg/ml tryptophan, 25 μg/ml uracil and 10 μg/ml thiamine. All growth experiments were conducted at 37°C in shaking water baths. The C. crescentus strain used was CB15N (Evinger and Agabian, 1977) and the cells were grown in PYE (Peptone-Yeast Extract) medium (Ely, 1991) at 30°C. Cell growth was monitored by measuring optical density at 450 nm (OD450).

Flow cytometry

Samples were taken from exponentially growing cells (OD450~0.2), incubated on ice to arrest growth and fixed in 75% ethanol. Samples for rifampicin arrest were harvested from exponentially growing cultures at OD450~0.2 and incubated with rifampicin (300 μg/ml; Novartis Pharma Inc.) and cephalexin (36 μg/ml; Sigma Chemical Co.) to inhibit initiation of DNA replication and cell division, respectively (Skarstad et al, 1986; Boye and Løbner-Olesen, 1991). Samples were taken at 1, 5, 10, 20 and 30 min after arrest and fixed in 75% ethanol. Flow cytometry was performed as described by Løbner-Olesen et al (1989) using an Apogee A10 instrument (Apogee Flow Systems Inc.). The ridge in the 3D flow cytometry graphs was extracted using a custom Perl script. For each light scatter channel, the fluorescence in a sliding window covering 11 fluorescence channels was calculated and the midpoint channel of the window with the highest fluorescence was plotted as the ridge of the graph.

Cell-cycle simulations

The DNA distributions in ideal cultures were simulated using a custom Excel worksheet based on the method described by Skarstad et al (1985) and Michelsen et al (2003) that was modified to accommodate for the presence of two chromosomes. The age distributions of ideal bacterial cultures growing without overlapping DNA replication cycles were calculated as described (Powell, 1958). B- and D-phase cells were assumed to contain one or two genome equivalents, respectively. To calculate the total DNA amount originating from the two chromosomes during the DNA replication phase of the cell cycle, the C period cells were divided into 100 sub-populations, where the amount of chromosome I DNA increased linearly from one to two during the C period. The time of replication of chromosome II was controlled through an input parameter. Before this time was reached, the amount of chromosome II DNA was set at one. Then it increased linearly until it reached two chromosome II equivalents and stayed at two until the end of the C period. For the simulations, the progression rates of all replication forks were equal and constant during the entire C period. Instrumental variability was included into the simulations as described by Michelsen et al (2003) by approximating peaks with normal distributions. The input parameters to the simulations are the lengths of the C and D periods, the relative timing of chromosome II initiation during the C period and the standard deviations of the one and two chromosome peaks. These parameters were varied to obtain the best possible fits between experimental DNA histograms and simulated DNA distributions.

Marker frequency determination

DNA templates were prepared from ethanol-fixed cells. The cells were harvested, lysed by resuspension in TE-buffer and vigorously vortexing, and the DNA template was diluted 10-fold with H2O before use. Quantitative PCR was conducted on a Light-Cycler 2.0 instrument (Roche Inc.) using the LightCycler FastStart DNA MasterPLUS SYBR Green I PCR kit (Roche Inc.) according to the instructions.

The following primer pairs amplifying ~130 bp fragments located near the origin (oriC) or terminus (ter) on chromosome I (I) or chromosome II (II) were used for marker frequency determinations: oriCI (VC2775, gidA) CGCCAACCGAGTTTGGATTC and AAAAAGCGCGTGAGCTTGG; terI (VC1410, vceA) CTGAGGCGGATTTGGCACTC and GCTTGCGCCGCTTTTAACTG; oriCII (VCA1114, parB) GCTCCACCTTCGGTGTTTCG and TGGTTTCGTGTGGCAGCAAT; terII (VCA0563) TATCCGCACAGCCTCAGCAA and CACGCAAACAGACCGACACC.

Each measurement was made in triplicate, Cp (crossing point) values were extracted and used for calculating the oriCI/oriCII and terI/terII ratios. Marker ratios were normalised to a culture grown to stationary phase in AB+0.2% glucose medium, where flow cytometry showed that DNA replication was complete and the cells had the same number of chromosome I and chromosome II.

Marker frequency model

Marker frequencies generated by quantitative PCR are compared with a model based on the equation of Bremer and Churchward (1977a):

equation image

where τ is the doubling time and C is the replication period. It is assumed that the C period is constant at different doubling times.

For bacteria with two chromosomes, I and II, we get:

equation image

Assuming initiation synchrony, oriCI/oriCII will equal 1 (see Figure 1) and can be removed:

equation image

Assuming termination synchrony, terI/terII will equal 1 (see Figure 1) and can be removed.

equation image

Equations (3) and (4) are used to predict the changes in oriI/oriII and terI/terII ratios at different doubling times. The model is based on the assumption of a constant difference in replication time (CICII), which is likely to be constant at 18–20 min in the doubling time range 18–55 min as observed in E. coli (Donachie, 1968). At doubling times >55 min, the C period increase. This probably also results in an increase in CICII, but the differences is negligible due to the comparatively larger increase in doubling time.

Supplementary Material

Supplementary Figure S1

Supplementary Figure S2


We thank Tove Atlung for critical reading of the paper and for scientific discussions, Dhruba Chattoraj for sharing data before publication and Kristine S Madsen for graphical assistance with Figure 1. This work was supported by grants from The Danish Natural Sciences Research Council and The Carlsberg Foundation.


  • Atlung T, Hansen FG (1993) Three distinct chromosome replication states are induced by increasing concentrations of DnaA protein in Escherichia coli. J Bacteriol 175: 6537–6545 [PMC free article] [PubMed]
  • Atlung T, Hansen FG (2002) Effect of different concentrations of H-NS protein on chromosome replication and the cell cycle in Escherichia coli. J Bacteriol 184: 1843–1850 [PMC free article] [PubMed]
  • Atlung T, Løbner-Olesen A, Hansen FG (1987) Overproduction of DnaA protein stimulates initiation of chromosome and minichromosome replication in E. coli. Mol Gen Genet 206: 51–59 [PubMed]
  • Bogan JA, Grimwade JE, Thornton M, Zhou P, Denning GD, Helmstetter CE (2001) P1 and NR1 plasmid replication during the cell cycle of Escherichia coli. Plasmid 45: 200–208 [PubMed]
  • Boye E, Løbner-Olesen A (1991) Bacterial growth control studied by flow cytometry. Res Microbiol 142: 131–135 [PubMed]
  • Bremer H, Churchward G (1977a) An examination of the Cooper-Helmstetter theory of DNA replication in bacteria and its underlying assumptions. J Theor Biol 69: 645–654 [PubMed]
  • Bremer H, Churchward G (1977b) Deoxyribonucleic acid synthesis after inhibition of initiation of rounds of replication in Escherichia coli B/r. J Bacteriol 130: 692–697 [PMC free article] [PubMed]
  • Clark DJ, Maaløe O (1967) DNA replication and the division cycle in Escherichia coli. J Mol Biol 23: 99–112
  • Cooper S, Helmstetter CE (1968) Chromosome replication and the division cycle of Escherichia coli B/r. J Mol Biol 31: 519–540 [PubMed]
  • Donachie W (1968) Relationship between cell size and time of initiation of DNA replication. Nature 219: 1077–1079 [PubMed]
  • Duigou S, Knudsen KG, Skovgaard O, Egan ES, Løbner-Olesen A, Waldor MK (2006) Independent control of replication initiation of the two Vibrio cholerae chromosomes by DnaA and RctB. J Bacteriol 188: 6419–6424 [PMC free article] [PubMed]
  • Egan ES, Duigou S, Waldor MK (2006) Autorepression of RctB, an initiator of Vibrio cholerae chromosome II replication. J Bacteriol 188: 789–793 [PMC free article] [PubMed]
  • Egan ES, Fogel MA, Waldor MK (2005) Divided genomes: negotiating the cell cycle in prokaryotes with multiple chromosomes. Mol Microbiol 56: 1129–1138 [PubMed]
  • Egan ES, Løbner-Olesen A, Waldor MK (2004) Synchronous replication initiation of the two Vibrio cholerae chromosomes. Curr Biol 14: R501–R502 [PubMed]
  • Egan ES, Waldor MK (2003) Distinct replication requirements for the two Vibrio cholerae chromosomes. Cell 114: 521–530 [PubMed]
  • Ely B (1991) Genetics of Caulobacter crescentus. Methods Enzymol 204: 372–384 [PubMed]
  • Evinger M, Agabian N (1977) Envelope-associated nucleoid from Caulobacter crescentus stalked and swarmer cells. J Bacteriol 132: 294–301 [PMC free article] [PubMed]
  • Fiebig A, Keren K, Theriot JA (2006) Fine-scale time-lapse analysis of the biphasic, dynamic behaviour of the two Vibrio cholerae chromosomes. Mol Microbiol 60: 1164–1178 [PMC free article] [PubMed]
  • Fogel MA, Waldor MK (2005) Distinct segregation dynamics of the two Vibrio cholerae chromosomes. Mol Microbiol 55: 125–136 [PubMed]
  • Hansen EB, Yarmolinsky MB (1986) Host participation in plasmid maintenance: dependance upon dnaA of replicons derived from P1 and F. Proc Natl Acad Sci USA 83: 4423–4427 [PMC free article] [PubMed]
  • Hava DL, Camilli A (2001) Isolation and characterization of a temperature-sensitive generalized transducing bacteriophage for Vibrio cholerae. J Microbiol Methods 46: 217–225 [PubMed]
  • Heidelberg JF, Eisen JA, Nelson WC, Clayton RA, Gwinn ML, Dodson RJ, Haft DH, Hickey EK, Peterson JD, Umayam L, Gill SR, Nelson KE, Read TD, Tettelin H, Richardson D, Ermolaeva MD, Vamathevan J, Bass S, Qin H, Dragoi I, Sellers P, McDonald L, Utterback T, Fleishmann RD, Nierman WC, White O (2000) DNA sequence of both chromosomes of the cholera pathogen Vibrio cholerae. Nature 406: 477–483 [PubMed]
  • Keasling JD, Palsson BO, Cooper S (1992) Replication of mini-F plasmids during the bacterial division cycle. Res Microbiol 143: 541–548 [PubMed]
  • Løbner-Olesen A, Hansen FG, Rasmussen KV, Martin B, Kuempel PL (1994) The initiation cascade for chromosome replication in wild-type and Dam methyltransferase deficient Escherichia coli cells. EMBO J 13: 1856–1862 [PMC free article] [PubMed]
  • Løbner-Olesen A, Skarstad K, Hansen FG, von Meyenburg K, Boye E (1989) The DnaA protein determines the initiation mass of Escherichia coli K-12. Cell 57: 881–889 [PubMed]
  • Messer W (2002) The bacterial replication initiator DnaA. DnaA and oriC, the bacterial mode to initiate DNA replication. FEMS Microbiol Rev 26: 355–374 [PubMed]
  • Michelsen O, Teixeira de Mattos MJ, Jensen PR, Hansen FG (2003) Precise determinations of C and D periods by flow cytometry in Escherichia coli K-12 and B/r. Microbiology 149: 1001–1010 [PubMed]
  • Morigen, Løbner-Olesen A, Skarstad K (2003) Titration of the Escherichia coli DnaA protein to excess datA sites causes destabilization of replication forks, delayed replication initiation and delayed cell division. Mol Microbiol 50: 349–362 [PubMed]
  • Morigen, Molina F, Skarstad K (2005) Deletion of the datA site does not affect once-per-cell-cycle timing but induces rifampin-resistant replication. J Bacteriol 187: 3913–3920 [PMC free article] [PubMed]
  • Okada K, Iida T, Kita-Tsukamoto K, Honda T (2005) Vibrios commonly possess two chromosomes. J Bacteriol 187: 752–757 [PMC free article] [PubMed]
  • Pal D, Venkova-Canova T, Srivastava P, Chattoraj DK (2005) Multipartite regulation of rctB, the replication initiator gene of Vibrio cholerae chromosome II. J Bacteriol 187: 7167–7175 [PMC free article] [PubMed]
  • Pierucci O, Helmstetter CE, Rickert M, Weinberger M, Leonard AC (1987) Overexpression of the dnaA gene in Escherichia coli B/r: chromosome and minichromosome replication in the presence of rifampin. J Bacteriol 169: 1871–1877 [PMC free article] [PubMed]
  • Powell EO (1958) An outline of the pattern of bacterial generation times. J Gen Microbiol 18: 382–417 [PubMed]
  • Reisenauer A, Shapiro L (2002) DNA methylation affects the cell cycle transcription of the CtrA global regulator in Caulobacter. EMBO J 21: 4969–4977 [PMC free article] [PubMed]
  • Skarstad K, Boye E, Steen HB (1986) Timing of initiation of chromosome replication in individual E. coli cells. EMBO J 5: 1711–1717 [PMC free article] [PubMed]
  • Skarstad K, Steen HB, Boye E (1985) Escherichia coli DNA distributions measured by flow cytometry and compared with theoretical computer simulations. J Bacteriol 163: 661–668 [PMC free article] [PubMed]
  • Skovgaard O, Løbner-Olesen A (2005) Reduced initiation frequency from oriC restores viability of a temperature-sensitive Escherichia coli replisome mutant. Microbiology 151: 963–973 [PubMed]
  • Srivastava P, Fekete RA, Chattoraj DK (2006) Segregation of the replication terminus of the two Vibrio cholerae chromosomes. J Bacteriol 188: 1060–1070 [PMC free article] [PubMed]
  • Trucksis M, Michalski J, Deng YK, Kaper JB (1998) The Vibrio cholerae genome contains two unique circular chromosomes. Proc Natl Acad Sci USA 95: 14464–14469 [PMC free article] [PubMed]
  • Venkova-Canova T, Srivastava P, Chattoraj DK (2006) Transcriptional inactivation of a regulatory site for replication of Vibrio cholerae chromosome II. Proc Natl Acad Sci USA 103: 12051–12056 [PMC free article] [PubMed]
  • von Freiesleben U, Rasmussen KV, Atlung T, Hansen FG (2000) Rifampicin-resistant initiation of chromosome replication from oriC in ihf mutants. Mol Microbiol 37: 1087–1093 [PubMed]
  • Wion D, Casadesus J (2006) N6-methyl-adenine: an epigenetic signal for DNA-protein interactions. Nat Rev Microbiol 4: 183–192 [PMC free article] [PubMed]
  • Xu Q, Dziejman M, Mekalanos JJ (2003) Determination of the transcriptome of Vibrio cholerae during intraintestinal growth and midexponential phase in vitro. Proc Natl Acad Sci USA 100: 1286–1291 [PMC free article] [PubMed]

Articles from The EMBO Journal are provided here courtesy of The European Molecular Biology Organization
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...