HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE ADVANCED SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Morigen Articles by Skarstad, K. Search for Related Content PubMed Articles by Morigen, Articles by Skarstad, K.

Department of Cell Biology, Institute for Cancer Research, Norwegian Radium Hospital, Rikshospitalet, University of Oslo, 0310 Oslo, Norway

	Abstract

Top Abstract Introduction Results Discussion Experimental procedures References

 
When the bacterium Escherichia coli is grown in rich medium, the replication and segregation periods may span two, three or four generations and cells may contain up to 24 replication forks. The newly synthesized, hemimethylated DNA at each fork is bound by SeqA protein. The SeqA–DNA structures form distinct foci that can be observed by immunofluorescence microscopy. The numbers of foci were lower than the numbers of replication forks indicating fork co-localization. The extent of co-localization correlated with the extent of replication cycle overlap in wild-type cells. No abrupt increase in the numbers of foci occurred at the time of initiation of replication, suggesting that new replication forks bind to existing SeqA structures. Manipulations with replication control mechanisms that led to extension or reduction of the replication period and number of forks, did not lead to changes in the numbers of SeqA foci per cell. The results indicate that the number of SeqA foci is not directly governed by the number of replication forks, and supports the idea that new DNA may be ‘captured’ by existing SeqA structures.

	Introduction

Top Abstract Introduction Results Discussion Experimental procedures References

 
Some bacteria (e.g. Escherichia coli and Bacillus subtilis) can grow with overlapping replication cycles when the medium is rich. Initiation then occurs at two origins, if the replication cycle spans more than one generation, at four origins, if the replication cycle spans more than two generations, or at eight origins if growth is really fast and the replication cycle spans more than three generations. Depending on cell age and growth conditions, cells may contain from four to 24 replication forks.

In slowly growing cells initiation occurs at one origin and generates two replication forks. In both E. coli and B. subtilis initiation of replication occurs at midcell, and with labeled replisome components the presence of cells with one focus indicate that the two forks co-localize some of the time at midcell (Bates & Kleckner 2005; Berkmen & Grossman 2006). The majority of replicating cells in a slowly growing population contain two replisome foci, however, indicating that replication forks most of the time hold different positions. It thus seems that, at least in slowly growing cells, there is no evidence for the presence of a stationary replication factory consisting of four polymerases and two helicases, into which the old DNA is pulled, and out of which the new DNA is extruded (Dingman 1974; Lemon & Grossman 1998).

In rapidly growing cells evidence for co-localization of two, four and six forks has been presented (Hiraga et al. 1998; Lemon & Grossman 1998; Molina & Skarstad 2004; Adachi et al. 2008). The limitation in resolution of the light microscope has not allowed any conclusions about fixed replication factories here either. However, the multifork replication that occurs in rich media places other demands on organization of the chromosome than the simple situation with two replication forks in slowly growing cells. Recent studies indicate that immediately after the simultaneous initiation at several origins, a co-assembly of four or six replication forks is found (Molina & Skarstad 2004; Fossum et al. 2007; Adachi et al. 2008). In these studies it was shown that pairs of forks stayed co-localized for major parts, but not an entire round of replication, possibly implying that co-localization depends on dynamic structures rather than stable complexes (Bates 2008). It has also recently been reported that E. coli cells lacking the SeqA protein do not co-localize forks to the same extent as wild-type cells (Fossum et al. 2007).

The SeqA protein prevents immediate re-initiation at new origins by specifically binding to newly replicated, hemimethylated, origins (Lu et al. 1994; von Freiesleben et al. 1994; Slater et al. 1995). Microscopy studies have shown that BrdU-labeled new DNA and immunostained SeqA co-localize (Molina & Skarstad 2004), and indicate that much of the cellular SeqA is bound at replication forks. In vitro, SeqA dimers bind to pairs of hemimethylated GATC sites (Brendler & Austin 1999). The SeqA dimers form multimers that take the shape of a left-handed helix (Guarne et al. 2005; Odsbu et al. 2005). In vivo, the SeqA multimer-DNA complex gives the impression of being a quite compact structure forming distinct foci detectable by immunofluorescence (Hiraga et al. 1998; Onogi et al. 1999; Brendler et al. 2000; Molina & Skarstad 2004). It is thus possible that the formation of SeqA structures on new DNA behind the forks is involved in aiding the correct segregation of sister chromosomes.

Here, we have manipulated replication control mechanisms in order to change the numbers of replication forks per cell. We find that the numbers of SeqA structures per cell increase with increasing cell mass and increasing extent of replication cycle overlap, but are not directly affected by changed replication fork numbers.

	Results

Top Abstract Introduction Results Discussion Experimental procedures References

 
Shortening of the C period led to fewer forks but the same number of SeqA foci per cell

To find out whether a change in the number of replication forks per cell leads to a change in the number of SeqA foci, we investigated cells with ‘delayed’ or ‘too early’ initiation of replication. When DnaA protein is overproduced, initiation occurs earlier in the cell cycle (Atlung et al. 1987; Løbner-Olesen et al. 1989; Skarstad et al. 1989), and the C period is extended. The cell thus seems to ‘compensate’ for the early initiation by letting forks move more slowly or pause longer or more often. Conversely, if initiation of replication is compromised, and initiation therefore occurs later than normal in the cell cycle, the C period is shortened (Boye et al. 1996; Torheim et al. 2000; von Freiesleben et al. 2000b; Morigen et al. 2003). The cells then contain fewer forks than normal.

Here, we first compared cells in which the availability of the DnaA initiator protein was limited by the presence of extra datA DnaA binding sites (Morigen et al. 2003), to wild-type cells. When grown in minimal salts medium supplemented with glucose, casaminoacids (Glu-CAA medium) and uridine, E. coli MG1655 cells have a cell cycle in which replication and segregation spans three generations. Initiation of replication occurred at four origins in the ‘grandmother’ generation (Fig. 1A; Supplementary Fig. S1 in Supporting Information/Supplementary Material; Skarstad et al. 1985; Molina & Skarstad 2004). Most of these cells (85%) contained 12 or 8 replication forks, and only 15% of the cells had four forks (Fig. 1A). The cells with extra datA and delayed initiation had a C period that was about 20% shorter. This led to an increase in the proportion of cells with four forks (to 46%) (Fig. 1B).

View larger version (56K): [in this window] [in a new window]   Figure 1  Replication patterns and SeqA focus distributions in cells with changed C or D periods. Exponentially growing cells in Glu-CAA medium with (A–D) or without (E) uridine were fixed and analyzed by flow cytometry and fluorescence microscopy. The first panels (left-most) illustrate replication patterns of the cells indicated. Cells (rods in shades of green and blue) with chromosomes (black lines) and origins (black dots) were drawn schematically to show the number of replication forks and origins at different stages of the cell cycles. (A) Wild-type MG1655 grown in Glu-CAA medium with uridine. The width of the rectangle indicates one generation, with the green, blue and light blue, indicating the periods from cell birth to initiation, from initiation to termination, and from termination to cell division, respectively. The horizontal black line indicates the C period which was found by the formula oriC/terC = 2C/, in which oriC/terC was measured by Southern hybridization. The horizontal white line indicates the D period. The C + D period was determined by analyzing the rifampicin-run-out DNA histogram obtained from flow cytometry (second panel). Here it is seen that cells contain four origins before, and eight origins after, initiation and thus C + D spans three generations. The calculation is described in detail in Supplementary Fig. S1. The duration of D was then found by subtraction ([C + D] – C = D). The numbers in blue indicate generation number (1: current generation, 2: ‘mother’ generation and 3: ‘grandmother’ generation). This culture thus contains cells with four origins and four forks at birth, eight origins and 12 forks after initiation, and then eight forks after termination of replication of the four oldest forks. These three phases of the cell's life with 4, 12 and 8 forks are indicated in green, blue, and light blue, respectively. The percentage of cells in each of the phases are shown in a bar diagram (third panel). Exponentially growing cells were fixed, immunostained with anti-SeqA antibody and visualized by fluorescence microscopy. Panel four shows the distribution of SeqA foci per cell, and panel five shows examples of images. The number of cells included in each microscopy experiment is shown. The numbers above the focus images are the average number of forks per SeqA focus. (B) MG1655/MiniR1-datA, (C) MG1655 dnaAT174P, and (D) MG1655 hns206 grown in Glu-CAA medium with uridine. (E) MG1655 grown in Glu-CAA medium without uridine. The cells shown in D and E have C + D periods that span two generations and thus have two, six or four forks.

In an earlier study of cells growing with a high degree of fork co-localization it was found that pairs of forks were co-localized for a major part of the replication cycle and in addition more extensive co-localization (of four or six forks) was suggested immediately after initiation of replication (Molina & Skarstad 2004). Here, we analyzed further wild-type cells with a moderate degree of co-localization (Fig. 1E) in order to find out whether forks were more extensively co-localized right after initiation also in this case. The replication pattern was such that the majority of these cells were either young with two forks or old with four forks (Fig. 1E, panels 1 and 3). Only 19% of the cells had six forks and had recently initiated replication. The cell length distributions of cells with one, two, three and four foci were measured (Fig. 2). Cell length was found to be increased with increasing number of SeqA foci. Thus, big cells with four forks had two, three or four foci and small cells with two forks had one or two foci. The result shows that pairs of forks are co-localized only part of the time under these growth conditions. At initiation of replication cells start four new forks in addition to the two old forks, but no abrupt increase in numbers of foci was observed at this cell length (Fig. 2). Most cells with six forks had two or three foci. Thus, the replication forks seem to be more extensively co-localized right after initiation also under these growth conditions. It is not possible from this analysis to determine whether all four of the new forks are co-localized or whether initiation at the two origins occurs in separate locations. If initiation occurs at separate locations and only two new forks are co-localized, the locations must also contain old forks.

View larger version (13K): [in this window] [in a new window]   Figure 2  The extent of replication fork co-localization is highest right after initiation. The lengths of the MG1655 cells described in Fig. 1E were measured and plotted as a function of the number of SeqA foci in each cell (bottom). The schematic drawing and percentages of cells with two, six or four forks were taken from Fig. 1E, but now shown in shades of grey instead of green (top). Young cells with two forks were assumed to be the smallest (light grey) and cells with four forks were assumed to be the largest (intermediate shade of grey). Most newly initiated cells (dark grey) had two or three SeqA foci.

Extension of the C period in hda mutant cells led to more forks but approximately the same number of SeqA foci per cell

Extended C periods have been observed in cells that initiate too early in the cell cycle or undergo excess initiation of replication (Atlung et al. 1987; Løbner-Olesen et al. 1989; Skarstad et al. 1989). We now investigated hda mutant cells that lack the RIDA regulatory system, and presumably contain a higher than normal proportion of ATP-form DnaA. These cells re-initiate, or initiate prematurely, and show an asynchrony phenotype after replication run-out (Katayama et al. 1998; Camara et al. 2003; Riber et al. 2006). The DNA histogram of exponentially growing cells was broad and indicated considerable cell-to-cell variability in replication pattern. Therefore, a C period common to all cycles was not possible to calculate. A lower and an upper estimate were made, yielding C periods from approximately 70 to 90 min. The Southern analysis gave an average C value of 81 min (Table 1). Assuming a D period of 30–40 min, cells contain from 4 to 28 forks with an estimated average of 16 (Table 1). The distribution of SeqA foci was found to be quite similar to that of wild type, however. This means that extra replication forks generated by uncontrolled initiation of replication do not lead to extra SeqA foci. In the Hda-less cells on average four forks were found to be co-localized (Table 1). The result is in accordance with earlier results from cells with mutant, non-sequesterable origins that undergo extra initiations (Fossum et al. 2007), and supports the idea that the distribution of SeqA foci is not directly dependent on the number of forks in the cell.

Fork organization in wild-type cells when C and D periods span four generations

In LB medium cells had a replication pattern where C and D spanned more than three generations (Fig. 3 and Table 2). Cells contained 12, 8 or 24 forks and were found to have from 2 to 12 SeqA foci, with most cells containing from four to eight foci (Fig. 3). Thus, this change in growth medium led to both an increase in fork number and in SeqA focus number per cell. The average number of forks per focus was found to be 2.2 that is, somewhat lower than for Glu-CAA-uri grown cells (Table 2). This result shows that the extent of fork co-localization was about the same in cells with more than three generations replication cycle overlap compared to cells with almost three generations overlap, and indicates that extent of replication overlap is not the only factor that affects fork organization.

View larger version (16K): [in this window] [in a new window]   Figure 3  Replication pattern and SeqA focus distribution of wild-type MG1655 grown in LB medium. Exponentially growing cells in LB medium at 37 °C were fixed and analyzed by flow cytometry and fluorescence microscopy. Calculations and illustrations were made as described in legend to Fig. 1 and Experimental procedures.

Fork organization is impaired in cells with low concentrations of SeqA protein

The results described above indicate that the organization of replication forks into SeqA structures is not directly dependent on the exact replication pattern. We investigated whether the cellular SeqA protein concentration varied and found that it was the same in LB and Glu-CAA-uri medium, about 10% reduced in Glu-CAA medium (Table 2), and about 20% increased in the dnaA_T174P mutant (Table 1).

To investigate the importance of the SeqA concentration we analyzed the fork and focus distributions in strain MOR256 which has a chromosomal deletion of the seqA gene and harbors a plasmid with an inducible seqA gene. Expression of the seqA gene was induced with 0.05 and 0.5 mM IPTG for four generations, which resulted in 70% and sevenfold concentrations of SeqA, respectively (Table 3). Cells with a reduced concentration of SeqA showed some asynchrony of replication (Fig. 4B) and heterogeneous distributions of DNA contents and mass (data not shown). It is to be expected that the cells are not uniformly induced and therefore are heterogeneous with respect to SeqA content, with some cells containing very little SeqA and some containing wild-type levels, giving a population average of 70%. This was apparent from the microscopy images which showed cells with SeqA immunofluorescence of varying intensity (Fig. 4B). Cells with normal fluorescence intensity also had normal looking SeqA foci (Fig. 4B, cells labeled ‘a’). Cells with intermediate amounts of SeqA fluorescence seemed to have many faint foci (Fig. 4B, cells labeled ‘b’). The result indicates that a minimal amount of SeqA is required for fork co-localization, and is in accordance with earlier BrdU labeling experiments indicating that in SeqA less cells replication forks were not co-localized (Fossum et al. 2007).

View larger version (57K): [in this window] [in a new window]   Figure 4 Replication fork and SeqA focus distributions in cells with too much or too little SeqA protein. Exponentially growing cells of wild-type CM735 (A) and MOR256 (CM735seqA/pMAK7, see text) at 37 °C in Glu-CAA medium supplemented with tryptophan were analyzed by flow cytometry, fluorescence microscopy and immunoblotting. For MOR256, a 0.7-fold (B) and sevenfold (C) concentration of SeqA was obtained by induction with 0.05 and 0.5 mM IPTG, respectively (see Experimental procedures). The first and second panels show replication run-out histograms and replication fork distributions respectively. Calculations of the fork distributions were made using the parameters listed in Table 3. The third and fourth panels show SeqA focus distributions and images of foci.

SeqA structures containing pairs of forks stayed intact when DNA spilled out of broken cells

Occasionally the lysozyme treatment required for immunostaining led to an apparent weakening of the cell wall and about 1% of the cells in the microscopy preparations were found to be partial ‘ghosts.’ That is, the cells were broken and the contents spilled out, fully or partially. Such cells yielded a ‘paler’ than normal phase contrast image. We investigated cells that had low phase contrast and DNA partially on the outside to see if the SeqA foci were robust enough to still be present (Fig. 5). A small cell of this kind is indicated with a ‘c’ and a large cell with a ‘d’ (Fig. 5). The DNA that came from the small cell had two SeqA foci (Fig. 5). In this cell population most of the small cells have four forks (Fig. 4A). Thus, forks were still co-localized in SeqA structures when the DNA was no longer inside the cell. Although it is conceivable that three replication forks were co-localized at one SeqA structure and only one fork present at the other, we consider it most likely that a pair of forks from the same initiation event were co-localized in each of the two SeqA structures. Among all the broken cells studied, most of the small ones had all foci intact (i.e. same number as in intact cells) on the DNA outside the cell.

View larger version (30K): [in this window] [in a new window]   Figure 5 SeqA structures exist on DNA outside broken cells. Wild-type CM735 cells were fixed, immunostained with anti-SeqA antibody and visualized by fluorescence microscopy. About 1% of the cells suffered ruptures so that some of the contents partially escaped and are seen outside the cells.

Asymmetrically positioned SeqA foci in dnaA mutant cells with asynchronous replication

The dnaA_A345S mutant has a similar deficiency as the dnaA_T174P mutant in that the proportion of ADP- to ATP-form DnaA is aberrantly high (Gon et al. 2006), and initiation therefore ‘delayed’ and C shortened. The replication pattern of the dnaA_A345S mutant was less normal with more severe under-initiation and asynchrony, compared to dnaA_T174P, but the growth rate was only slightly reduced (Fig. 6). The C period was found to be extremely short (19 min determined by Southern blot), and most cells contained two, three or four origins (Fig. 6A,B). The asynchrony in initiation, which probably is due to a considerable time interval between initiation of one origin and the next, led to different cell cycles in different cells. A particular cell's cycle depended on the variable timing of initiation, and so cell cycles differed from generation to generation. It was therefore not possible to calculate a common replication pattern for all cells. A calculation based on the information that 22% of the cells had initiated one origin but not the other (i.e. contained three origins), was performed (Fig. 6B,C; Experimental Procedures). The calculation was limited to cells with two, three and four origins (about 85% of the population). An estimate of the amount of cells that contained (i) two non-replicating chromosomes, (ii) one non-replicating chromosome and one replicating chromosome, (iii) two replicating chromosomes, (iv) one replicating chromosome and two non-replicating chromosomes, and (v) four chromosomes with no forks, was made, and a replication cycle estimated (Fig. 6B). On the basis of this a distribution of replication forks for the culture was made (Fig. 6C), and shows that 40% of the cells have two forks and 40% have four forks. Since 15% of the population (which contained five-, six- and seven-origin cells) was not included this is a slight underestimate.

View larger version (30K): [in this window] [in a new window]   Figure 6 Asymmetrically localized SeqA foci in dnaAA345S mutant cells. Cells were grown exponentially at 37 °C in Glu-CAA medium with uridine. (A) Localization of SeqA foci in two-focus-cells of the dnaAA345S mutant (lower panel) is compared with that of the control (upper panel). Each violet or purple triangle shows the position (relative to cell length) of one SeqA focus in a cell. The rifampicin–run-out histograms (small panels) are shown. Arrows indicate cells with asymmetrically localized foci. (B) Cells at different stages of the dnaAA345S cell cycle were schematically drawn as described in Fig. 1, but showing asynchronous initiations. The C (average of three Southern blot experiments), individual asynchronous C1 and C2 (dashed and dotted lines), C + D and D periods were determined as described in Experimental procedures. (C) Replication fork (left panel) and SeqA focus (right panel) distributions were determined as described. The average numbers of forks per cell, SeqA foci per cell and forks per focus are given.

In the calculation of the distribution of forks per cell (Fig. 6C) we found that about 20% of the cells have no forks. However, the microscopy images show that almost all cells have SeqA foci. Some of the foci were very weak, however, as in the cell labeled ‘e’ (Fig. 6A). The dnaA_A345S mutant cells were 70% larger than wild-type cells. It is possible that in these large cells with a surplus of SeqA, the SeqA structures exist even after replication is finished.

In these cells, initiation is a severe limitation. It is possible that the SeqA focus that is formed on the new forks immediately splits and the two foci move in opposite directions more quickly to ‘catch up’ with other cellular processes. These cells have only a fraction of the forks they should have had, and SeqA and segregation determinants might be expected to just process what is present. The result is in accordance with earlier observations of only one fork per SeqA focus in mutant cells where replication is uncoupled from cell growth (Bach et al. 2003).

Higher mutation frequency in cells with rapid replisomes

In several of the mutants studied here, the C periods were shorter than normal (Table 4 and Supplementary Table S3). The shortest C period was found in dnaA_A345S mutant cells, which are severely compromised in initiation. To find out whether the fast-going replisomes replicate with the same fidelity as wild-type replisomes, we measured rifampicin resistant colony formation. The dnaA_A345S mutant cells exhibited a 20-fold increase in spontaneous mutation rate relative to the wild-type cells (Table 4). The result is in accordance with previous findings (Gon et al. 2006), and is not due to an induction of the SOS response (I. Flåtten, L. Johnsen and K. Skarstad, unpublished). Also the hns206 and dnaA_T174P mutant cells, and cells with extra datA showed somewhat elevated (threefold to fivefold) mutation rates (Table 4). It was previously suggested by Gon et al. that a high concentration of RNR might be the cause of the increased frequency of mutation in the two dnaA mutants (Gon et al. 2006). Here, we consider it possible that all the four types of cells tested in Table 4 have an increased dNTP availability at the fork, caused either by increased RNR expression, reduced DNA concentration, or both, and that this leads to the increased rate of DNA polymerization by the replisome. If so, the increased polymerization rate may occur at the expense of the proofreading exonuclease activity of the replisome. The observations that the fastest fork (that of dnaA_A345S) led to the highest mutation rate, may support this idea.

	Discussion

Top Abstract Introduction Results Discussion Experimental procedures References

In the present work we find that the extent of fork co-localization increased with increasing replication cycle overlap. However, the distributions of SeqA foci did not seem to be governed by the exact replication patterns. The presence of extra datA sites caused no change in growth rate, but a relatively large change in replication pattern, which led to a relatively large reduction in numbers of replication forks. The distribution of SeqA foci remained unchanged, however. Likewise, deletion of the hda gene caused over-initiation and a large increase in the numbers of replication forks per cell, but the average number of SeqA foci did not increase. At the time of initiation of replication no abrupt increase was seen in the numbers of SeqA foci per cell. These results could indicate that the new replication forks that are generated at initiation become included in existing SeqA structures. Recent data that support this idea show that plasmids carrying GATC clusters that stay bound by SeqA most of the cell cycle also seem to be included in existing SeqA structures (Bach et al. 2008).

Thus, it seems that the distribution of SeqA structures is dictated by something other than the replication cycle. It is possible that successful segregation of sister chromosomes is important in this context. All copies of one sister's new DNA must be directed into the correct cell half. It could be that new forks latching on to nearby SeqA structures would aid the process. How and when the structures split and become independent of each other may be governed by events relevant to the segregation required prior to cell division.

During slow growth no SeqA focus exists at the time of initiation, but as new DNA is generated a SeqA structure is built. It is possible that the determinants that interact with this structure and cause it to split after about 10–20 min, are the same as the determinants that govern numbers and movement of SeqA foci during rapid growth. Alternatively, slow growing cells may not form the same kind of SeqA structures as rapidly growing cells. Deletion of the seqA gene does not lead to growth impairment in poor media, only to changes in control of initiation (Boye et al. 1996). Thus, it may be that only rapidly growing cells require a choreography of SeqA structures throughout a round of replication.

Co-localization of replication forks involves active tethering of separate DNA molecules

So far, the limitations of the resolution of light microscopy have not allowed any conclusions about whether pairs of forks in a single SeqA focus are part of the same structure or whether they are in separate structures and just happen to be near each other by chance. Here, we found that the SeqA structures held pairs of forks co-localized even when the DNA escaped from the cell. This shows that pairs of forks do not just happen to be in the same place in the cell by chance, but are in close proximity because SeqA and/or other determinants keep them so. The result also indicates that SeqA may not be heavily anchored to the membrane as has been suggested by results from earlier work (Ogden et al. 1988; Slater et al. 1995). The SeqA structures seemed to easily escape from the cell together with the DNA. Alternatively, it could be that the lysozyme treatment necessary for immunostaining may have caused disintegration of such putative anchoring structures.

SeqA structures depend on a minimum amount of SeqA

We found here that the formation of SeqA foci was dependent on a minimum amount of SeqA protein. Normal SeqA foci and foci distributions were not found in cells with low concentrations of SeqA. In such cells faint continuous fluorescence or multiple faint foci were seen, indicating a lack of fork co-localization. The result supports earlier findings of aberrant fork localization in SeqA-less cells (Fossum et al. 2007), and indicates that SeqA protein not only labels new DNA, but is required for the organization of the new DNA.

The spatial organization of new DNA reported here may, or may not, reflect the spatial organization of the replisomes. This presumably depends on whether there are newly replicated GATC sites near the replisomes or not. In some regions of the chromosome long stretches of DNA without GATC sites exist. In such areas the replisomes could be far away from the hemimethylated DNA-SeqA structures, whereas in areas with high numbers of GATC sites replisomes would co-localize with the SeqA structures.

	Experimental procedures

Top Abstract Introduction Results Discussion Experimental procedures References

 
Bacterial strains, media and plasmids

All bacteria used were E. coli K-12 and are listed in Table 5. The hns206:: Ap and hda:: Tet were transferred into MG1655 by P1 transduction. Cells were grown in AB minimal medium (Clark & Maaløe 1967) supplemented with 10 µg/mL of thiamine, 0.2% glucose, 0.5% casaminoacids and 100 µg/mL of uridine (Glu-CAA-uri medium). Wild-type MG1655 was also grown in LB, or AB with supplements mentioned above except uridine (Glu-CAA medium). CM735 and derivative were grown in Glu-CAA medium with 25 µg/mL of tryptophan. Ampicillin, chlorampenicol and tetracycline were added to final concentrations of 25, 25, and 15 µg/mL, respectively when required for selection. Plasmids MiniR1-datA (Morigen et al. 2001) and pMAK7 (von Freiesleben et al. 2000a) were described previously.

At OD₄₅₀ (or OD₆₀₀) = 0.15, exponentially growing cells were harvested or treated with 300 µg/mL of rifampicin (Fluka, Buchs, Switzerland) and 10 µg/mL of cephalexin (Eli Lilly, Indianapolis, IN, USA) for four to five generations to allow completion of ongoing rounds of replication. Cells were collected after the drug treatment. Rifampicin inhibits transcription, which results in inhibition of initiation, but not elongation of replication (Skarstad et al. 1986). Cephalexin inhibits cell division (Boye & Løbner-Olesen 1991). In the presence of these drugs, cells with synchronous initiation end up with an integral number (2, 4, 8 or 16 chromosomes) which represents the number of origins in each cell at the time of drug action (Skarstad et al. 1986). In cultures of cells with asynchronous initiation, additional peaks representing cells with three, five, six, seven etc. origins appear.

Collected cells were resuspended in TE buffer, fixed in 70% of ethanol. The cells were washed in 0.1 M phosphate buffer (PB) pH 9.0, and stained overnight (4 °C) in 1.5 µg/mL of fluorescein isothiocyanate (FITC) in the same buffer (Wold et al. 1994). The cells were then washed in 0.02 M Tris–buffered saline pH 7.5, and stained in the same buffer containing 1.5 µg/mL of Hoechst 33258 which stains DNA. Staining of the DNA standard was performed as described previously (Torheim et al. 2000).

Flow cytometry analysis was then performed using a LSRII (Becton Dickinson) equipped with an Argon ion laser and a Krypton laser (both Spectra Physics). The data obtained from the flow cytometry measurement was analyzed by WinMDI software. The average DNA content per cell was determined by taking the average of the Hoechst fluorescence intensity distribution as average Hoechst fluorescence per cell. The average cell mass, determined as average FITC fluorescence per cell, was calculated by taking the average of the FITC fluorescence intensity distribution. The average DNA per mass was found by dividing average DNA content per cell by average cell mass.

Southern hybridization

An amount of 50 mL of exponentially growing cells (OD₄₅₀, or OD₆₀₀ = 0.15) were lysed by adding SDS buffer (1.4% SDS, 4 mM EDTA) at 60 °C for 10 min to stop cellular reactions (Bach et al. 2003). Total DNA was precipitated by addition of NaCl (0.2 M) and isopropanol (0.7 volume) to the mixture, and pelleted by centrifugation. The resulting DNA was resuspended in TE buffer, treated with RNase (100 µg/mL) to remove RNA, followed by two phenol-extractions and chloroform-treatments, and then precipitated in ethanol and NaCl. The precipitated DNA was pelleted by centrifugation, washed in 80% and 96% ethanol, respectively, and subsequently resuspended in TE buffer. To determine the ratio of oriC per terC, the DNA was double-digested with SalI and XhoI prior to separation on 1% agarose gel by electrophoresis. Preparation of the sequence-specific oriC and terC probes, subsequent ³²P-labeling, as well as visualization, quantification and normalization of radioactive bands were performed as described previously (Morigen et al. 2001).

Determination of the C and D periods, numbers of forks per cell and calculation of theoretical distributions

The C period was determined by oriC/terC = 2^C/ (Sueoka & Yoshikawa 1965) in which oriC/terC was measured by Southern hybridization as mentioned above. The C + D period was determined by the rifampicin–run-out DNA histogram which gives the distribution of origins at the time of drug addition (see section of Flow cytometry above), the doubling time of the culture (), and the exponential age distribution (see Supplementary Fig. S1 Supporting Information/Supplementary Material for details about calculations). To facilitate the calculation, rounded numbers given in the run-out histograms were used (for instance 15% and 85% four and eight origin cells in Fig. 1A, second panel, rather than the measured 14% and 84%). Initiation age was determined by the percentage of cells which have not initiated (for instance 15% four origin cells in Fig. 1A, second panel) and termination age was then obtained by the length of the C period (Fig. 1A, first panel; Supplementary Fig. S1 Supporting Information/Supplementary Material). To convert from percentage of cell culture to part of cell cycle (time) the exponential age distribution was taken into consideration (Supplementary Fig. S1 Supporting Information/Supplementary Material) (Skarstad et al. 1985). The D period was simply C + D minus the C period. The numbers of forks per cell are given by the durations of the C and D periods (Supplementary Fig. S1 Supporting Information/Supplementary Material). Theoretical histograms were calculated with the cell cycle parameters obtained (see legend to Supplementary Fig. S1 Supporting Information/Supplementary Material) and compared to the experimental histograms to verify that the calculated parameters were correct. Only distributions of cells in balanced growth can be used for calculations.

Parameters for the hda and dnaA_A345S mutant cells were made as average estimates. For the culture of dnaA_A345S mutant cells where initiation of replication was asynchronous (Fig. 6A, lower small panel), an ‘extended’ of C was calculated since initiation occurred first at one origin and some time later at the second origin (i.e. each of the two chromosomes in the same cell had separate C periods, and ‘extended C’ just means the total time of replication). The first initiation age was found assuming that the cells with two origins have not yet initiated (i.e. using the percentage of cells with two origins in the calculation described in Supplementary Fig. S1 Supporting Information/Supplementary Material) and the second initiation age was found assuming that the cells with two and three origins have not initiated (i.e. using the fraction of cells with two and three origins in the calculation). The termination ages were found by assuming termination to occur 19 min after initiation. An average C period was used to determine an average D period. The fork distribution was then calculated and the average number of forks per cell determined.

Immunostaining and fluorescence microscopic visualization of SeqA foci

Immunostaining and microscopy of SeqA was performed as described previously (Fossum et al. 2003) using cell samples fixed in ethanol (see section of Flow cytometry above).

Measurement of cell lengths and SeqA focus positions in microscopy images

Measurement of the positions of foci in Fig. 6 was performed in such a way that the pole with the shortest distance to the foci was chosen to be at the zero position.

Induction of seqA gene expression

Strain MOR256 has a chromosomal deletion of the seqA gene (Torheim et al. 2000) and harbors a plasmid with an IPTG-inducible seqA gene (von Freiesleben et al. 2000a). After addition of 0.05 and 0.5 mM IPTG to cultures of MOR256 at OD₄₅₀ = 0.01, the cells were incubated for four generations. At OD₄₅₀ = 0.16, samples were taken for flow cytometry, microscopy and immunoblotting.

Immunoblotting

Exponentially growing cells (OD₄₅₀ = 0.2 or 0.15) were harvested by centrifugation at 4 °C and resuspended in Tris–EDTA buffer containing 1% SDS and glycerol and boiled for 5 min. Total protein concentration was determined by a colorimetric assay (BCA kit from Pierce). The same amount of total protein from each strain or treatment was subjected to SDS-PAGE (12%) after addition of a buffer containing SDS and β-mercaptoethanol. The proteins were transferred to a polyvinylidene difluoride (PVDF) membrane by semi-dry blotting. The membrane was probed with SeqA, DnaA or NrdA polyclonal antibody using standard procedures. Detection was performed with a fluorescence kit (ECF kit from Amersham Pharmacia Biotech). The membrane was scanned on a Storm 840 (Molecular Dynamics), and quantification was carried out using ImageQuant software (Molecular Dynamics).

Measurement of spontaneous mutation

Mutation frequency was measured by growing cells on rifampicin-containing AB plates supplemented with glucose and casaminoacids or LB plates. At OD₄₅₀ or OD₆₀₀ = 1.1, 1 mL of cells (about 10⁸ cells) growing in Glu-CAA medium or LB at 37 °C was plated on AB or LB plate containing 300 µg/mL of rifampicin and grown overnight. Formed colonies including small ones were scored up after growth of about 20 h at 37 °C. A dilution (about 200 cells) of each measurement was plated on AB or LB plate to determine the number of cells per 1 mL of the culture.

	Acknowledgements

 
We are grateful to Kirsti Solberg Landsverk of the Department of Radiation Biology Flow Cytometry Core Facility for her excellent technical assistance in performance of flow cytometry. We also thank Felipe Molina for stimulatory discussions and Trond Bach and Torsten Waldminghaus for critical reading of the manuscript. Tove Atlung provided strain TC4136. This work was supported by the Norwegian Research Council (FRIBIOMOL and FUGE).

	Footnotes

 
Communicated by: Nancy Kleckner

^* Correspondence: kirsten.skarstad{at}rr-research.no

	References

Top Abstract Introduction Results Discussion Experimental procedures References

 
Adachi, S., Fukushima, T. & Hiraga, S. (2008) Dynamic events of sister chromosomes in the cell cycle of Escherichia coli. Genes Cells 13, 181–197.[Abstract/Free Full Text]

Atlung, T. & Hansen, F.G. (2002) Effect of different concentrations of H-NS protein on chromosome replication and the cell cycle in Escherichia coli. J. Bacteriol. 184, 1843–1850.[Abstract/Free Full Text]

Atlung, T., Løbner-Olesen, A. & Hansen, F.G. (1987) Overproduction of DnaA protein stimulates initiation of chromosome and minichromosome replication in Escherichia coli. Mol. Gen. Genet. 206, 51–59.[CrossRef][Medline]

Bach, T., Krekling, M.A. & Skarstad, K. (2003) Excess SeqA prolongs sequestration of oriC and delays nucleoid segregation and cell division. EMBO J. 22, 315–323.[CrossRef][Medline]

Bach, T., Morigen & Skarstad, K. (2008) The initiator protein DnaA contributes to keeping new origins inactivated by promoting the presence of hemimethylated DNA. J. Mol. Biol. 384, 1076–1085.[CrossRef][Medline]

Bates, D. (2008) The bacterial replisome: back on track? Mol. Microbiol. 69, 1341–1348.[CrossRef][Medline]

Bates, D. & Kleckner, N. (2005) Chromosome and replisome dynamics in E. coli: loss of sister cohesion triggers global chromosome movement and mediates chromosome segregation. Cell 121, 899–911.[CrossRef][Medline]

Berkmen, M.B. & Grossman, A.D. (2006) Spatial and temporal organization of the Bacillus subtilis replication cycle. Mol. Microbiol. 62, 57–71.[CrossRef][Medline]

Boye, E. & Løbner-Olesen, A. (1991) Bacterial growth control studied by flow cytometry. Res. Microbiol. 142, 131–135.[Medline]

Boye, E., Stokke, T., Kleckner, N. & Skarstad, K. (1996) Coordinating DNA replication initiation with cell growth: differential roles for DnaA and SeqA proteins. Proc. Natl. Acad. Sci. USA 93, 12206–12211.[Abstract/Free Full Text]

Brendler, T. & Austin, S. (1999) Binding of SeqA protein to DNA requires interaction between two or more complexes bound to separate hemimethylated GATC sequences. EMBO J. 18, 2304–2310.[CrossRef][Medline]

Brendler, T., Sawitzke, J., Sergueev, K. & Austin, S. (2000) A case for sliding SeqA tracts at anchored replication forks during Escherichia coli chromosome replication and segregation. EMBO J. 19, 6249–6258.[CrossRef][Medline]

Camara, J.E., Skarstad, K. & Crooke, E. (2003) Controlled initiation of chromosomal replication in Escherichia coli requires functional Hda protein. J. Bacteriol. 185, 3244–3248.[Abstract/Free Full Text]

Clark, D.J. & Maaløe, O. (1967) DNA replication and the division cycle in Escherichia coli. J. Mol. Biol. 23, 99–112.[CrossRef]

Dingman, C.W. (1974) Bidirectional chromosome replication: some topological considerations. J. Theor. Biol. 43, 187–195.[CrossRef][Medline]

Fossum, S., Crooke, E. & Skarstad, K. (2007) Organization of sister origins and replisomes during multifork DNA replication in Escherichia coli. EMBO J. 26, 4514–4522.[CrossRef][Medline]

Fossum, S., Soreide, S. & Skarstad, K. (2003) Lack of SeqA focus formation, specific DNA binding and proper protein multimerization in the Escherichia coli sequestration mutant seqA2. Mol. Microbiol. 47, 619–632.[CrossRef][Medline]

Gon, S., Camara, J.E., Klungsoyr, H.K., Crooke, E., Skarstad, K. & Beckwith, J. (2006) A novel regulatory mechanism couples deoxyribonucleotide synthesis and DNA replication in Escherichia coli. EMBO J. 25, 1137–1147.[CrossRef][Medline]

Guarne, A., Brendler, T., Zhao, Q., Ghirlando, R., Austin, S. & Yang, W. (2005) Crystal structure of a SeqA-N filament: implications for DNA replication and chromosome organization. EMBO J. 24, 1502–1511.[CrossRef][Medline]

Guyer, M.S., Reed, R.R., Steitz, J.A. & Low, K.B. (1981) Identification of a sex-factor-affinity site in E. coli as gamma delta. Cold Spring Harb. Symp. Quant. Biol. 45, 135–140.[Abstract/Free Full Text]

Hansen, F.G. & von Meyenburg, K. (1979) Characterization of the dnaA, gyrB and other genes in the dnaA region of the Escherichia coli chromosome on specialized transducing phages tna. Mol. Gen. Genet. 175, 135–144.[CrossRef][Medline]

Hiraga, S., Ichinose, C., Niki, H. & Yamazoe, M. (1998) Cell cycle-dependent duplication and bidirectional migration of SeqA-associated DNA-protein complexes in E. coli. Mol. Cell 1, 381–387.[CrossRef][Medline]

Jensen, K.F. (1993) The Escherichia coli K-12 "wild types" W3110 and MG1655 have an rph frameshift mutation that leads to pyrimidine starvation due to low pyrE expression levels. J. Bacteriol. 175, 3401–3407.[Abstract/Free Full Text]

Katayama, T., Kubota, T., Kurokawa, K., Crooke, E. & Sekimizu, K. (1998) The initiator function of DnaA protein is negatively regulated by the sliding clamp of the E. coli chromosomal replicase. Cell 94, 61–71.[CrossRef][Medline]

Lemon, K.P. & Grossman, A.D. (1998) Localization of bacterial DNA polymerase: evidence for a factory model of replication. Science 282, 1516–1519.[Abstract/Free Full Text]

Løbner-Olesen, A., Skarstad, K., Hansen, F.G., von Meyenburg, K. & Boye, E. (1989) The DnaA protein determines the initiation mass of Escherichia coli K-12. Cell 57, 881–889.[CrossRef][Medline]

Lu, M., Campbell, J.L., Boye, E. & Kleckner, N. (1994) SeqA: a negative modulator of replication initiation in E. coli. Cell 77, 413–426.[CrossRef][Medline]

Molina, F. & Skarstad, K. (2004) Replication fork and SeqA focus distributions in Escherichia coli suggest a replication hyperstructure dependent on nucleotide metabolism. Mol. Microbiol. 52, 1597–1612.[CrossRef][Medline]

Morigen, Boye, E., Skarstad, K. & Løbner-Olesen, A. (2001) Regulation of chromosomal replication by DnaA protein availability in Escherichia coli: effects of the datA region. Biochim. Biophys. Acta 1521, 73–80.[Medline]

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.[CrossRef][Medline]

Odsbu, I., Klungsoyr, H.K., Fossum, S. & Skarstad, K. (2005) Specific N-terminal interactions of the Escherichia coli SeqA protein are required to form multimers that restrain negative supercoils and form foci. Genes Cells 10, 1039–1049.[Abstract/Free Full Text]

Ogden, G.B., Pratt, M.J. & Schaechter, M. (1988) The replicative origin of the E. coli chromosome binds to cell membranes only when hemimethylated. Cell 54, 127–135.[CrossRef][Medline]

Onogi, T., Niki, H., Yamazoe, M. & Hiraga, S. (1999) The assembly and migration of SeqA–Gfp fusion in living cells of Escherichia coli. Mol. Microbiol. 31, 1775–1782.[CrossRef][Medline]

Riber, L., Olsson, J.A., Jensen, R.B., Skovgaard, O., Dasgupta, S., Marinus, M.G. & Løbner-Olesen, A. (2006) Hda-mediated inactivation of the DnaA protein and dnaA gene autoregulation act in concert to ensure homeostatic maintenance of the Escherichia coli chromosome. Genes Dev. 20, 2121–2134.[Abstract/Free Full Text]

Skarstad, K., Boye, E. & Steen, H.B. (1986) Timing of initiation of chromosome replication in individual Escherichia coli cells. EMBO J. 5, 1711–1717.[Medline]

Skarstad, K., Løbner-Olesen, A., Atlung, T., von Meyenburg, K. & Boye, E. (1989) Initiation of DNA replication in Escherichia coli after overproduction of the DnaA protein. Mol. Gen. Genet. 218, 50–56.[CrossRef][Medline]

Skarstad, K., Steen, H.B. & Boye, E. (1985) Escherichia coli DNA distributions measured by flow cytometry and compared with theoretical computer simulations. J. Bacteriol. 163, 661–668.[Abstract/Free Full Text]

Slater, S., Wold, S., Lu, M., Boye, E., Skarstad, K. & Kleckner, N. (1995) E. coli SeqA protein binds oriC in two different methyl-modulated reactions appropriate to its roles in DNA replication initiation and origin sequestration. Cell 82, 927–936.[CrossRef][Medline]

Sueoka, N. & Yoshikawa, H. (1965) The chromosome of Bacillus subtilis. I. Theory of marker frequency analysis. Genetics 52, 747–757.[Free Full Text]

Torheim, N.K., Boye, E., Løbner-Olesen, A., Stokke, T. & Skarstad, K. (2000) The Escherichia coli SeqA protein destabilizes mutant DnaA204 protein. Mol. Microbiol. 37, 629–638.[CrossRef][Medline]

von Freiesleben, U., Krekling, M.A., Hansen, F.G. & Løbner-Olesen, A. (2000a) The eclipse period of Escherichia coli. EMBO J. 19, 6240–6248.[CrossRef][Medline]

von Freiesleben, U., Rasmussen, K.V., Atlung, T. & Hansen, F.G. (2000b) Rifampicin-resistant initiation of chromosome replication from oriC in ihf mutants. Mol. Microbiol. 37, 1087–1093.[CrossRef][Medline]

von Freiesleben, U., Rasmussen, K.V. & Schaechter, M. (1994) SeqA limits DnaA activity in replication from oriC in Escherichia coli. Mol. Microbiol. 14, 763–772.[Medline]

Wold, S., Skarstad, K., Steen, H.B., Stokke, T. & Boye, E. (1994) The initiation mass for DNA replication in Escherichia coli K-12 is dependent on growth rate. EMBO J. 13, 2097–2102.[Medline]

Received: 12 September 2008
Accepted: 22 February 2009

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Morigen Articles by Skarstad, K. Search for Related Content PubMed Articles by Morigen, Articles by Skarstad, K.