JOURNAL OF BACTERIOLOGY, Dec. 2007, p. 8660–8666 0021-9193/07/$08.00⫹0 doi:10.1128/JB.01212-07 Copyright © 2007, American Society for Microbiology. All Rights Reserved.
Vol. 189, No. 23
Dynamics of Escherichia coli Chromosome Segregation during Multifork Replication䌤 Henrik J. Nielsen,1 Brenda Youngren,1 Flemming G. Hansen,2 and Stuart Austin1* Gene Regulation and Chromosome Biology Laboratory, National Cancer Institute, CCR, NCI-Frederick, Frederick, Maryland 21702-12011, and Microbial Genomics Group, Center for Biological Sequence Analysis, BioCentrum-DTU, Technical University of Denmark, DK-2800 Kgs. Lyngby, Denmark2 Received 27 July 2007/Accepted 18 September 2007
Slowly growing Escherichia coli cells have a simple cell cycle, with replication and progressive segregation of the chromosome completed before cell division. In rapidly growing cells, initiation of replication occurs before the previous replication rounds are complete. At cell division, the chromosomes contain multiple replication forks and must be segregated while this complex pattern of replication is still ongoing. Here, we show that replication and segregation continue in step, starting at the origin and progressing to the replication terminus. Thus, early-replicated markers on the multiple-branched chromosomes continue to separate soon after replication to form separate protonucleoids, even though they are not segregated into different daughter cells until later generations. The segregation pattern follows the pattern of chromosome replication and does not follow the cell division cycle. No extensive cohesion of sister DNA regions was seen at any growth rate. We conclude that segregation is driven by the progression of the replication forks. Escherichia coli chromosome replication initiates once per cell cycle from a unique origin of replication and proceeds bidirectionally to a terminus region on the opposite side of the circular DNA (16). Dingman proposed a novel model for E. coli chromosome segregation based on this observation (3). In this model, the replication machinery was tethered to the cell center. The chromosome was drawn through the anchored forks, and the newly replicated sister duplexes were pushed in opposite directions toward the cell poles to form two new chromosome masses. Thus, segregation occurred concomitantly with replication and proceeded progressively from origin to terminus. This idea received support from experiments that showed that the replicative DNA polymerase was located at the cell center in Bacillus subtilis (11). Evidence for a central replication “factory” has also been reported for E. coli (8, 9). The mode of segregation has become more accessible due to the development of marker-specific fluorescence labeling techniques (4, 12, 15, 17). Some results suggested a picture of chromosomal segregation in E. coli radically different from that predicted by Dingman. Segregation appeared to be a discontinuous process, resembling eukaryotic chromosome segregation (2, 6, 19). It was deduced that the chromosome replicated to form a joint structure, with most chromosomal loci remaining paired after replication. Segregation then occurred as an independent process, with all or much of the chromosome coming apart from its sister in a single concerted event. The evidence for concerted chromosome segregation following extensive sister chromosome cohesion was based on findings that, for most markers around the chromosome, there were far fewer fluorescent foci than the number of predicted
copies of the locus. It was concluded that sister regions of the chromosome remained paired after replication for an extended period so that the two sister loci would appear as one focus through much of the cell cycle (2, 6, 19). Because for different loci much the same number of foci were found in all but the oldest cells, it was also concluded that segregation of the cohesive sister loci occurred as a concerted event in which most markers came apart at the same time. Recent studies question these conclusions. Using the phage P1 green fluorescent protein (GFP)-ParB/parS labeling system, we showed that 14 markers spaced around the chromosome segregated in their order of replication in slowly growing cells and that segregation occurred relatively soon after replication for most markers (14). Evidence for progressive segregation in slowly growing cells has also been presented recently, using a fluorescent repressor/operator detection system (20). It is likely that the apparent eukaryote-like segregation seen in the earlier studies was primarily an artifact of inefficient focus detection. Although marker segregation appears to occur concomitantly with replication, two alternatives seem evident. Segregation might divide the nucleoid mass in two in each cell division cycle, starting with the origin sequences and progressing to the terminus, in a fashion timed to follow replication but not directly governed by it. Alternatively, segregation might be directly coupled to replication, as proposed in the Dingman model, so that sister regions that emerge from the replication forks are always directed into separate masses. Here, we address this question by studying chromosome segregation in fast-growing cells, where replication is uncoupled from the cell division cycle. In slowly growing cells, only one origin is initiated and a single chromosome is duplicated, resulting in two chromosomes that are separated into two daughter cells. These cells have a combined replication period (C period) and postreplicational period (D period) that are equal to or less than the interdivision time (5). In order to produce sufficient numbers
* Corresponding author. Mailing address: Gene Regulation and Chromosome Biology Laboratory, National Cancer Institute, CCR, NCI-Frederick, Frederick, MD 21702-1201. Phone: (301) 846-1266. Fax: (301) 846-6988. E-mail:
[email protected]. 䌤 Published ahead of print on 28 September 2007. 8660
E. COLI CHROMOSOME SEGREGATION
VOL. 189, 2007
8661
of chromosomes to keep up with the cell mass increase at high growth rates, the cells initiate new rounds of replication scheduled for future cell division events before previous rounds are ended (5). Thus, the replication of many markers occurs in generations before the cell divides. The C-D period exceeds the interdivision time, and replication is initiated from two or more origins simultaneously (5, 18). Under such conditions, when the C period is longer than the generation time, the cells have chromosomes with multiple replication fork pairs and the chromosomes are constantly replicating. The region around the origin of replication may be present in 8 or even 16 copies in dividing cells. The patterns of chromosome segregation under these conditions are unknown. Do the highly reiterated regions stay together and does the total genomic content split progressively in two during each cell division cycle, or is each chromosome region segregated as it is replicated? In the latter case, segregation would occur in generations prior to the one in which the duplicated regions are placed in separate cells. MATERIALS AND METHODS Bacterial strains and media. The strains used were derivatives of E. coli MG1655 with P1 parS inserted at 13 different positions around the chromosome (14). The GFP-⌬30ParB protein was expressed from pALA2705 (14). Cells were grown at 32°C in LB medium or minimal AB medium supplemented with 0.2% glucose, 0.05% Casamino Acids, 1 g/ml thiamine, and 1 g/ml uracil. Both media were supplemented with 100 g/ml ampicillin to ensure retention of the GFP-⌬30ParB-expressing plasmid. Strains were grown exponentially for at least 6 generations to a final optical density at 600 nm of 0.1 before samples were taken for microscopy and flow cytometry. Microscopy. From a growing culture, 1.5 ml was harvested and centrifuged for 2 min at 13,200 rpm, and the cell pellet was resuspended in 20 l of medium. Eight microliters of the concentrated cells was placed on a polylysine-coated glass slide and covered with a coverslip, and the cells were immobilized by pressing the coverslip at ⬃9 kg/cm2 for 6 seconds. Microscopy was carried out on a Nikon Eclipse E-1000 microscope equipped with a Nikon C-CU Universal condenser, a Nikon Apo TIRF 1.49-numeric-aperture 100⫻ objective, a Semrock GFP-3035 brightline zero band pass filter cube, and a Hamamatsu Orca-ER c4742-95 charge-coupled device camera. Images were acquired using Openlab 4.2 software. Automatic measurement of cells and foci. Phase-contrast images were used for cells grown in glucose medium. A single fluorescence exposure was sufficient to see the fluorescence signal from all foci in these cells. Due to the additional thickness of cells grown in LB medium, three Z planes were used in order to ensure detection of the signal from all foci present. The three layers were merged into a single image using the maximum-intensity extended depth of field procedure in Image Pro Plus 6.1. Differential interference contrast microscopy was used to visualize the cells grown in LB medium. The length of each cell, the number of foci, and the positions of the focus centers relative to the long cell axis were measured automatically using the Image Pro Plus 6.1 image analysis program. The relevant features of the program were linked by macro programming to adapt them to the specific task at hand. The counted cells were then inspected manually, and changes were made to the data when cells with an incorrect focus count were found. At least 200 cells were inspected for each strain and growth condition. Calculating separation time. The following equation from Helmstetter gives the average number of loci/genes per cell in the population (5): average number of loci ⫽ 2[C(1 ⫺x) ⫹ D]/ where C is the period of replication, D is the time between termination and cell division, x is the fraction of the C period at which the locus replicates, and is the generation time. By rearrangment, we get [C(1 ⫺ x) ⫹ D]/ ⫽ ln(average number of loci)/ln(2), where the left-hand side of the equation is the number of generations from locus replication to cell division. By replacing the average number of loci with the average number of observed foci on the right-hand side, the equation expresses the number of generations from focus separation to cell division instead. The number of generations from focus separation to cell division is then ln(average number of foci)/ln(2). Flow cytometry. Cells for flow cytometry were harvested directly from the exponentially growing culture. For runoff samples, cells were treated with rifampin and cephalexin to final concentrations of 200 g/ml and 36 g/ml, re-
FIG. 1. The cell division cycle. (A) The map positions of the 13 parS inserts used in this study. (B and C) Graphical presentations of the DNA replication cycle for cells grown in glucose-Casamino Acid medium (B) and cells grown in LB medium (C). The upper rows show chromosome configurations at division and at intermediate cell ages. Beneath, the timing and extent of the C and D periods with respect to cell division are shown. The time of initiation, period of replication (C period), and termination and postreplication period (D period) leading to the cell division in question are shown in bold color. The relevant preceding and subsequent C and D periods are shown in shaded color. The time scale is in generations before cell division.
spectively, and incubated for 4 h at 32°C. The runoff cells were used directly for microscopy or washed twice in buffer (1 mM EDTA, 10 mM Tris HCl, pH 7.4), fixed in 70% ethanol, and stored at 5°C for analysis by flow cytometry. These cells were subsequently washed once in 1 ml of 0.01 M MgCl2, 0.01 M Tris, pH 7.5. The cells were then centrifuged, and the pellet was suspended in 100 l of buffer. Mithramycin A and ethidium bromide were added to final concentrations of 100 g/ml and 20 g/ml, respectively. The cells were left at 4°C in the dark for 1 h. The DNA content distributions were then determined in a flow cytometer (Bryte SH; Bio-Rad, Hercules, CA) equipped with a 100-W Osram mercury short-arc
8662
NIELSEN ET AL.
J. BACTERIOL.
FIG. 2. Labeling of loci with the GFP-ParB/parS system. Image overlays of phase-contrast images and fluorescent images are shown. (A and B) Cells growing in glucose-Casamino Acid medium labeled at the origin of replication (84⬘) (A) and 54⬘ (B). (C and D) Cells growing in LB medium labeled at the origin (84⬘) (C) or the 54⬘ locus (D). The average number of foci depends both on the chromosomal locus and on the growth rate, due to the multiforked nature of the replicating chromosomes.
HBO lamp. The C and D periods were determined as described by Skarstad et al. (18) with the modification of Michelsen et al. (13).
RESULTS The cell cycle. Strains carrying parS sites at various chromosomal locations were grown in either minimal AB glucose medium supplemented with Casamino Acids or in LB medium. In the former, the cells grew with a doubling time of 55 min at 32°C. Flow cytometry gave a C period of 55 min and a D period of 44 min (Fig. 1). Thus, the majority of these cells have four origins and two termini. The chromosomes replicate continuously and have two to six replication forks. The average dividing cell contains two chromosomes, each with two origins. The two chromosomes are approximately 80% replicated (Fig. 1). Note that the D period in E. coli K-12 strains is considerably longer than that observed in strain B/r in the classic studies of Helmstetter and Michelsen et al. (5, 13). Cells grown in LB medium had a doubling time of 27 min at 32°C. Flow cytometry gave C and D periods of approximately 41 and 47 min. The majority of these cells have eight origins and two or four termini. Chromosomes replicate continuously and have 4 to 12
replication forks (a total of 8 to 24 forks per cell). The average dividing cell contains four chromosomes, each having four origins, two replication forks that have traversed 80% of the distance to the terminus, and four forks that have traversed 20% of the distance (Fig. 1). Strains containing only the parS site without a plasmid, as well as a wild-type MG1655 strain, were also analyzed in the flow cytometer and gave results identical to those for strains containing GFP-ParB/parS (data not shown). Thus, as we have seen in minimal glycerol medium, the labeling system does not affect the cell cycle at higher growth rates at low induction. The numbers derived from the flow cytometry are population averages. There will be considerable cell-to-cell variation in the timing of cell cycle events. Thus, for example, some cells will initiate early and have 16 origins in the case of LB medium. We demonstrate this below. At both growth rates, a variation coefficient on the timing of initiation of 20% was found. This is in agreement with our previous studies, as well as earlier studies on cell-to-cell variations (7). Detection of foci. The P1 GFP-ParB/parS labeling system is very efficient, and the number and positions of fluorescent foci
VOL. 189, 2007
E. COLI CHROMOSOME SEGREGATION
8663
FIG. 3. Labeled cells before and after replication runoff. Cells labeled at the terminus (33⬘) were grown in glucose medium as described in Materials and Methods and treated with rifampin and cephalexin to allow ongoing rounds of replication to run to termination. Fields of labeled cells before and after the treatment are shown in panels A and B, respectively.
in slowly growing cells can be determined automatically with high efficiency (14). However, the fastest-growing cells studied here had considerably more foci (Fig. 2), and the automatic counting method that we used failed to resolve some of these. We therefore visually inspected and manually corrected the automatic counts where necessary. The efficiency of focus detection can be determined by counting foci in cells treated with rifampin and cephalexin (the runoff procedure). These cells have fully completed chromosomes and have completed segregation so that all loci are equally represented and should contain corresponding and predictable numbers of foci (14). Figure 3 shows the results of runoff for a marker near the terminus of replication in cells growing in glucose-Casamino Acid medium. During the runoff period, the chromosomes transition from their partially replicated state to produce four or, in a few cells, two complete, separate chromosomes. If each locus is visible as a focus, there should be equivalent numbers of foci in runoff cells irrespective of whether the origin or the terminus is marked. The origin marker should show only a modest increase in the number of foci during the runoff period, reflecting any completion of segregation in the absence of replication. The number of foci for the terminus marker should increase sharply, however, reflecting completion of both replication and segregation. This effect is clearly seen in Fig. 3. From the flow cytometry measurements, there should be an average of 3.48 chromosomes per cell in the population. The observed values were 3.1 for the origin marker foci and 3.2 for the terminus. Thus, the efficiency of focus detection in these cells was approximately 90%. Microscopy of runoff cells grown in LB medium revealed an average of 8.3 foci for the origin-labeled strain and 8.5 for the terminus-labeled strain. These numbers are close to the 9.1 origins per cell we found from flow cytometry of the same cells, giving an efficiency of focus detection of approximately 90% also. Number of foci. We counted the foci in 13 strains carrying the parS site at 13 different positions on the chromosome. Microscopy of the cells growing exponentially in the glucoseCasamino Acid medium revealed cells containing from one to
four foci depending on the labeled locus. Cells labeled at or close to the origin had more foci than cells labeled closer to the terminus. This pattern was also observed in cells grown in LB medium, even though they had more than twice the number of foci. Figure 4 shows histograms of the numbers of cells containing a given number of foci for the cells grown in LB medium. As expected, we see that the number of foci decreases as the distance of the marker from the origin increases. This applies to both the left and right arms of the chromosome. Thus, the number of foci present in the cells reflects the replication order of the markers. Note that the cells do not always contain a number of foci corresponding to 2n, as would be the ideal case if detection was 100% efficient and all initiation, replication progression, and segregation events were perfectly synchronous. This is expected, however, as the progression of replication is a stochastic process and the exact times of replication of a given marker are bound to differ for different replication forks. Furthermore, the efficiency of detection, even when it is as high as 90%, has a dramatic effect when there are so many foci in the cell. For an eight-focus cell, the 90% focus detection efficiency results in a 53% chance of scoring the cell as a seven- or six-focus cell. Segregation is progressive in fast-growing cells. We previously determined the time of segregation for markers in slowly growing cultures by plotting the fraction of cells with segregated foci versus the cell age for several cell age classes. This created sigmoidal separation curves. By fitting these to a cumulative normal distribution, the mean value was then the average age at focus separation (14). This method works well for slowly growing cells where separation is simply indicated by the presence of two foci rather than one. At higher growth rates, this method is problematic, however, because many cells have segregated some of the sister focus pairs but not others. We therefore used an alternative method. We took the overall average number of foci per cell in each population and used the equations derived by Helmstetter (5) to calculate the time of focus separation (see Materials and Methods). This method give results for slowly growing cells almost identical to those
8664
NIELSEN ET AL.
FIG. 4. Histograms of detected foci. The distributions of the numbers of foci detected for the 13 different strains are presented as histograms. The histograms are organized according to the chromosomal locations of the labeled loci with the origin-proximal loci at the top. The position of each locus on the chromosome is shown on the chromosome diagram in the center.
obtained from the sigmoidal segregation curves as previously published (14). The times of segregation calculated in this way for the 13 different loci are shown in Fig. 5 for cells grown in glucoseCasamino Acid medium and for cells grown in LB medium. Even though replication now spanned more than a generation, segregation was clearly progressive from origin to terminus, as
J. BACTERIOL.
FIG. 5. Average cell age at focus separation. The time from focus separation to termination of the D period (cell division) was calculated from the average number of foci, as described in Materials and Methods, and is plotted as a function of chromosomal insertion positions for cells grown in glucose medium (A) and LB medium (B). The values for cell age are shown as a proportion of the generation time. The time of replication (the time at which a marker is replicated) as given by the C and D periods established by flow cytometry is shown with solid lines. Separation regression lines are shown as hatched lines.
was previously observed in slowly growing cells with a simple chromosome cycle. Kinetics of segregation. It is apparent from Fig. 5 that segregation is not only progressive, but also linear with time. Linear regressions to the time of segregation are shown in the figure. The terminus is not included in the regression plots, as it is subject to special conditions delaying its segregation in a fraction of the cells (1). The result is regression lines with R values of 0.97 for glucose cells and 0.99 for the LB cells. Clearly
E. COLI CHROMOSOME SEGREGATION
VOL. 189, 2007
the chromosome segregates linearly with time with a high degree of statistical significance. From the slope of the regressions, we calculated the rate of progression of separation around the chromosome for the two growth rates. In glucose-Casamino Acid, this rate was 38.3 kb/min. The rate of replication under these conditions was 42.2 kb per fork per minute. In LB medium, 60.2 kb of DNA was separated per minute at each fork compared to a replication rate of 56.5 kb/min. These values for rates of replication and segregation progression are very similar, as is illustrated by the almost parallel separation and replication lines in Fig. 5. Delay between replication and segregation. As seen in slowly growing cells (14), a delay was observed between replication and segregation of loci at both of the higher growth rates (Fig. 5). This delay was very similar for all markers except the terminus. Figure 5 shows the deduced time-of-replication curves for the respective growth conditions as determined by flow cytometry. By comparing the regression lines in Fig. 5 with the timing of replication, we get an average delay between replication and segregation of 12.8 min, or 0.23 generation, in glucose-Casamino Acid medium and 5.9 min, or 0.22 generation, in LB medium. We previously found a period of sister locus cohesion of 0.17 generation (20 min) in minimal glycerol medium (14). Although these delays are quite different in duration, they are very similar as a proportion of the cell cycle. This could reflect a common separation mechanism linked to the growth rate of the cell, or as discussed below, it could be a direct consequence of the efficiency of focus detection. DISCUSSION We have shown here that separation of sister regions of the chromosome follows soon after their replication, even when the replication event occurs in one or even two generations prior to that in which those regions end up in separate cells. The same rule also applies to low growth rates: generation times of 115 min (14) and greater than 180 min at 32°C (data not shown). Moreover, the rate at which the separation progresses around the chromosome is equal or very close to that at which the replication forks proceed, and the delays between replication and visible segregation are short and very similar for all markers except the terminus. Segregation is therefore directly coupled to replication and could well be driven by it, as originally suggested (3, 10). A short delay between replication and visible separation of the chromosome is inevitable, as the local regions coated with ParB-GFP are quite extensive (14) and some time will pass between replication and the formation of two visibly separate foci. Why, then, have other studies indicated that segregation is a discontinuous process, resembling eukaryotic chromosome segregation? The key to these conclusions has been evidence for extensive sister chromosome cohesion following replication. Using fluorescence in situ hybridization, it was found that there were far fewer fluorescent foci for most markers around the chromosome than the number of predicted copies of the locus. It was therefore concluded that sister regions of the chromosome remained paired after replication for an extended period, so that the two sister loci would appear as one focus through much of the cell cycle (2, 19). As for most loci much the same numbers of foci were found in all but the oldest cells,
8665
it was also assumed that segregation of the cohesive sister loci occurred as a concerted event in which most markers came apart at the same time. We believe that our conclusions differ from those described previously largely because of a better estimate of the number of fluorescent foci present in the cells. Any underestimation will result in an apparent delay of segregation. Even with a 90% focus detection efficiency, as achieved here, the missing foci would result in an apparent delay of 15% of the cell cycle without any actual delay. A focus detection efficiency of 50% would result in an apparent delay of an entire generation and would greatly complicate interpretation of the results. As it is inherently easier to detect multiple foci in large cells than in small ones, inefficiency of detection will also give the false impression that most segregation events occur relatively late in the cell cycle. For all growth rates we have looked at, we observed a delay between replication and segregation of approximately 20% of the generation time. Most of this (15%) is simply due to the efficiency of focus detection. The remaining 5% likely reflects the time it takes for replicated loci to separate enough that they can be distinguished. Thus, segregation appears to follow replication as closely as is reasonably possible to detect. We conclude that segregation either occurs as a direct consequence of replication or is rate limited by it. Segregation does not conform to the cell division cycle but rather follows the replication cycle of the chromosome, even if it extends over more than one cell generation. Our observations are incompatible with an idea of extended sister chromosome cohesion or concerted segregation of much of the chromosome. Thus, E. coli chromosome segregation does not appear to resemble eukaryotic chromosome segregation, as has previously been suggested (2), but may be more analogous to the production of separable sister chromatids that occurs during replication in eukaryotes. ACKNOWLEDGMENTS This research was supported in part by the Intramural Research Program of the National Cancer Institute, NIH. H.J.N. was supported in part by the Oticon, Otto Mønsted, Ulla and Mogens Folmer Andersens, and Frant Allings foundations in Denmark. REFERENCES 1. Aussel, L., F. X. Barre, M. Aroyo, A. Stasiak, A. Z. Stasiak, and D. Sherratt. 2002. FtsK is a DNA motor protein that activates chromosome dimer resolution by switching the catalytic state of the XerC and XerD recombinases. Cell 108:195–205. 2. Bates, D., and N. Kleckner. 2005. Chromosome and replisome dynamics in E. coli: loss of sister cohesion triggers global chromosome movement and mediates chromosome segregation. Cell 121:899–911. 3. Dingman, C. W. 1974. Bidirectional chromosome replication: some topological considerations. J. Theoret. Biol. 43:187–195. 4. Gordon, G. S., D. Sitnikov, C. D. Webb, A. Teleman, A. Straight, R. Losick, A. W. Murray, and A. Wright. 1997. Chromosome and low copy plasmid segregation in E. coli: visual evidence for distinct mechanisms. Cell 90:1113– 1121. 5. Helmstetter, C. E. 1996. Timing of synthetic activities in the cell cycle, p. 1627–1639. In F. C. Neidhardt et al. (ed.), Escherichia coli and Salmonella, 2nd ed., vol. 2. ASM Press, Washington DC. 6. Hiraga, S., C. Ichinose, T. Onogi, H. Niki, and M. Yamazoe. 2000. Bidirectional migration of SeqA-bound hemimethylated DNA clusters and pairing of oriC copies in Escherichia coli. Genes Cells 5:327–341. 7. Koch, A. L. 1996. Similarities and differences of individual bacteria within a clone, p. 1640–1651. In F. C. Neidhardt et al., Escherichia coli and Salmonella: cellular and molecular biology. ASM Press, Washington, DC. 8. Koppes, L. J., C. L. Woldringh, and N. Nanninga. 1999. Escherichia coli contains a DNA replication compartment in the cell center. Biochimie 81: 803–810.
8666
NIELSEN ET AL.
9. Lau, I. F., S. R. Filipe, B. Soballe, O. A. Okstad, F. X. Barre, and D. J. Sherratt. 2003. Spatial and temporal organization of replicating Escherichia coli chromosomes. Mol. Microbiol. 49:731–743. 10. Lemon, K. P., and A. D. Grossman. 2001. The extrusion-capture model for chromosome partitioning in bacteria. Genes Dev. 15:2031–2041. 11. Lemon, K. P., and A. D. Grossman. 1998. Localization of bacterial DNA polymerase: evidence for a factory model of replication. Science 282:1516– 1519. 12. Li, Y., K. Sergueev, and S. Austin. 2002. The segregation of the Escherichia coli origin and terminus of replication. Mol. Microbiol. 46:985–996. 13. Michelsen, O., M. J. Teixeira de Mattos, P. R. Jensen, and F. G. Hansen. 2003. Precise determinations of C and D periods by flow cytometry in Escherichia coli K-12 and B/r. Microbiology 149:1001–1010. 14. Nielsen, H. J., Y. Li, B. Youngren, F. G. Hansen, and S. Austin. 2006. Progressive segregation of the Escherichia coli chromosome. Mol. Microbiol. 61:383–393.
J. BACTERIOL. 15. Niki, H., and S. Hiraga. 1998. Polar localization of the replication origin and terminus in Escherichia coli nucleoids during chromosome partitioning. Genes Dev. 12:1036–1045. 16. Prescott, D. M., and P. L. Kuempel. 1972. Bidirectional replication of the chromosome in Escherichia coli. Proc. Natl. Acad. Sci. USA 69:2842–2845. 17. Sherratt, D. J., I. F. Lau, and F. X. Barre. 2001. Chromosome segregation. Curr. Opin. Microbiol. 4:653–659. 18. Skarstad, K., E. Boye, and H. B. Steen. 1986. Timing of initiation of chromosome replication in individual Escherichia coli cells. EMBO J. 5:1711– 1717. (Erratum, 5:3074.) 19. Sunako, Y., T. Onogi, and S. Hiraga. 2001. Sister chromosome cohesion of Escherichia coli. Mol. Microbiol. 42:1233–1241. 20. Wang, X., X. Liu, C. Possoz, and D. J. Sherratt. 2006. The two Escherichia coli chromosome arms locate to separate cell halves. Genes Dev. 20:1727– 1731.
