|Home | About | Journals | Submit | Contact Us | Français|
A genetic screen for cell division cycle mutants of Caulobacter crescentus identified a temperature-sensitive DNA replication mutant. Genetic complementation experiments revealed a mutation within the dnaE gene, encoding the α-catalytic subunit of DNA polymerase III holoenzyme. Sequencing of the temperature-sensitive dnaE allele indicated a single base pair substitution resulting in a change from valine to glutamic acid within the C-terminal portion of the protein. This mutation lies in a region of the DnaE protein shown in Escherichia coli, to be important in interactions with other essential DNA replication proteins. Using DNA replication assays and fluorescence flow cytometry, we show that the observed block in DNA synthesis in the Caulobacter dnaE mutant strain occurs at the initiation stage of replication and that there is also a partial block of DNA elongation.
The initiation of DNA replication is a critical step in the cell cycle and requires the cooperative activities of a large number of proteins. The molecular events leading to the initiation of DNA replication have been clearly defined in Escherichia coli (15, 23). The process begins with the DnaA protein binding to conserved regions within the oriC region (DnaA boxes), leading to the formation of an open complex (4, 14, 17, 21). Upon formation of this single-stranded region, the helicase, DnaB, is recruited to the origin locus, followed by the primase, DnaG (2, 4, 16). This enzyme forms short ribonucleotide primers upon which DNA polymerase III (Pol III) holoenzyme units can assemble to drive bidirectional chromosomal replication.
The E. coli Pol III holoenzyme is composed of 10 subunits that can be organized into three functional groups (24, 32, 36, 39, 41, 45). The Pol III core consists of the α-catalytic subunit, encoded by the dnaE gene (35, 59); the 3′-5′ exonuclease subunit, (52); and a third subunit, θ, which is not conserved in bacteria and whose function is unknown (6, 41, 54, 56). The core enzyme associates with the ring-shaped β2 sliding clamp that holds the enzyme to the DNA and increases its polymerizing efficiency (7, 18, 25, 31, 44, 49, 55, 57). The τ subunit associates with the DnaB helicase and the Pol III α subunit, which results in the connection of two core polymerases at the replication forks (30, 62) and hence facilitates simultaneous leading- and lagging-strand DNA synthesis (7, 29). It is proposed that during the initiation event, a β dimer assembles onto primed DNA template sites and that its subsequent association with the Pol III core enzyme tethers the polymerases to the DNA (31, 57).
The initiation step of DNA replication has not been as well characterized in the gram-negative, alpha-purple bacterium Caulobacter crescentus, where DNA replication is tightly linked to cell division and differentiation. In this organism, asymmetric cell division gives rise to a swarmer cell, with a single flagellum at one pole and a nonmotile stalked cell (5, 19). The stalked cell eventually forms a predivisional cell, synthesizing a flagellum at the pole opposite the stalk. There is no DNA replication in the swarmer cell. DNA replication is initiated during the stage in which the swarmer cell differentiates into the stalked cell, at which time the swarmer cell loses its flagellum and synthesizes a stalk.
We report here that during a screen for Caulobacter temperature-sensitive (TS) mutants for cell division, we isolated a DNA replication mutant with a filamentous cell division phenotype at the restrictive temperature of 37°C. Further characterization of this TS strain revealed the mutation to be in the Caulobacter homolog of the dnaE gene, encoding the α-catalytic subunit of Pol III. We show that the TS mutation within the dnaE gene effectively blocks DNA replication, and our data suggest that this block is at the initiation stage. The dnaE TS mutation lies within a region of the gene that has been demonstrated in E. coli to be important in the interaction of DnaE with other DNA replication proteins essential for the replication process.
We had originally designed a genetic screen to isolate mutants of Caulobacter that are TS for the essential cell division gene ftsZ (11, 34, 51). This genetic screening involved propagating the plasmid pΔN, a pBGST18 derivative containing the ftsZ gene missing its promoter and the first 63 codons (51) in the XL-1 Red mutator E. coli strain (Stratagene Chemical Co., La Jolla, Calif.). The resulting plasmids were introduced into the wild-type Caulobacter strain NA1000 (12) through conjugation (53). A screen of 5,000 integrants for TS mutants resulted in the isolation of seven TS mutants. Five of the mutants generated had cell division defects, and one of them, YB551 (dnaE1100), was TS lethal and exhibited the strongest cell division phenotype at 37°C. Cells of this strain were slightly longer than wild-type Caulobacter cells at 30°C (Fig. (Fig.1C)1C) but were filamentous at 37°C in liquid cultures (Fig. (Fig.1D).1D). As a control, nonmutated pΔN was integrated into NA1000 (dnaE+) to generate YB1719 (dnaE+), which was similar to wild-type Caulobacter cells incubated at 30 and 37°C (Fig. (Fig.1A1A and B, respectively). Subsequent experiments suggested that the TS lethal mutation did not lie in the ftsZ gene. Introduction of the cosmid T46 (51) (carrying 30 kb of the C. crescentus genome, including ftsZ) into the TS strain YB551 (dnaE1100) to generate YB543 (dnaE1100) failed to rescue the TS cell division phenotype at 37°C. In addition, we were unable to recover cell division TS transductants when we attempted to transduce the ftsZ allele from YB551 (dnaE1100) into a wild-type, NA1000 (dnaE+), background. This indicated that the TS lethal mutation of YB551 (dnaE1100) was in another gene or was due to a combination of mutations, one in ftsZ and one unlinked to ftsZ. Hence, we used a C. crescentus plasmid library in an attempt to rescue the TS mutation present in YB551 (dnaE1100). This plasmid library (a gift from Jim Gober, University of California, Los Angeles) carries ~5-kb fragments from a partial Sau3A digest of C. crescentus genomic DNA cloned into the vector pJS14, a derivative of pBBR1MCS (33). YB543 (dnaE1100) was used in this screen to ensure that if a mutation existed in ftsZ, it was being complemented by cosmid T46. Biparental matings were carried out with a pool of the E. coli library strain and YB543 (dnaE1100). These matings were plated at the restrictive temperature of 37°C. Of the 88 colonies that appeared at 37°C, 5 grew equally well on plates incubated at 30 and 37°C. These five transconjugants were picked for further analysis. At 30°C, the cells were slightly longer than wild-type cells (Fig. (Fig.1E),1E), but at 37°C, the majority of cells of the five transconjugants were able to divide (Fig. (Fig.1F).1F). Introduction of the vector pBBR1MCS alone into YB543 (dnaE1100) did not confer growth on plates at 37°C, and liquid cultures exhibited the same filamentous cell phenotype as YB543 (dnaE1100) at 37°C (data not shown). The plasmids from the five transconjugants were isolated and reintroduced into the YB543 (dnaE1100) strain by conjugation to confirm that the rescue of the TS phenotype was due to the plasmids. Again, all the transconjugants obtained from these matings grew equally well on plates incubated at 30 and 37°C and exhibited the nonfilamentous phenotype at 37°C when they were grown overnight in liquid culture. Restriction digests of the five plasmids showed that they contained common fragments, and hence, all subsequent experiments were carried out with the plasmid pJS14M. pJS14M also rescued the cell division phenotype of the original TS YB551 (dnaE1100) mutant, indicating that the presence of cosmid T46 was not required for rescue. This finding suggested that the TS phenotype of YB551 (dnaE1100) was likely due to a mutation at a single locus and not within ftsZ.
DNA sequencing of pJS14 M indicated that the 4,596-bp insert carries the C. crescentus dnaE homolog, encoding the α subunit of Pol III (Fig. (Fig.2A).2A). The recent sequencing of the entire genome of C. crescentus (42) has shown that the dnaE gene encodes a protein of 1,143 residues (Fig. (Fig.2C);2C); pJS14 M carries all but the last 84 codons of dnaE. The Caulobacter DnaE protein has significant sequence similarity to the E. coli DnaE, which is responsible for the polymerase activity of the replicative enzyme. The overall identity between the two proteins is 41%; the highest sequence identity occurs over approximately the first 800 amino acids (48% identity). The sequence identity falls to 24% over the remaining C-terminal residues. The N-terminal half of the E. coli DnaE has been shown to be required for polymerase activity (28), and the key conserved active-site residues that have been identified within this region are also found in the Caulobacter DnaE (Fig. (Fig.2C2C).
In addition to dnaE, three additional open reading frames (CC1929, CC1928, and CC1927) (Fig. (Fig.2A)2A) are contained within the pJS14M insert. The start codon of CC1928 lies 290 bp upstream of the dnaE start codon and is transcribed in the orientation opposite to that of dnaE. The predicted product of CC1928 is 319 amino acids in length and is homologous to inosine-uridine-preferring nucleoside hydrolases (3). CC1929 is transcribed in the same direction as dnaE. pJS14M encodes 47 amino acids of the C-terminal region of CC1929, which is homologous to ATP-binding proteins of ABC transporters (8). Within the gap between dnaE and CC1928 there is another short (120-bp) open reading frame, CC1927, which encodes a hypothetical protein with no significant similarity to known proteins.
We used the plasmids pJSI, pJPI, pJBI, and pJXI to determine which genes were able to complement the YB551 (dnaE1100) TS mutation (Fig. (Fig.2B).2B). These plasmids were constructed by respectively subcloning 2.0-kb SacI, 2.0-kb PstI, 2.5-kb BamHI, and 2.3-kb XhoI fragments from pJS14 M into pBBR1MCS. All failed to rescue the YB551 (dnaE1100) TS phenotype, suggesting that CC1927, CC1928, and CC1929 did not contain the TS mutation. This finding indicated that the cell division phenotype of YB551 (dnaE1100) was due to a TS mutation in DNA replication since inhibition in DNA replication blocks cell division in Caulobacter (9, 10, 43, 46, 47).
pJS14 M is a medium-copy-number plasmid, and it was possible that multiple copies of dnaE suppressed a mutation at another locus. To determine if a single copy of the pJS14M insert could rescue the TS cell division phenotype of the YB551 (dnaE1100) mutant, the entire insert from pJS14M was subcloned into the plasmid pGSZ, a gentamicin derivative of pGMTZI (1), to create pG17. A single-crossover event resulting in the integration of the entire plasmid at the dnaE locus was selected for by plating the cells on gentamicin following introduction of the plasmid into YB551 (dnaE1100). YB551 (dnaE1100)::pG17 integrants had wild-type morphology at 30 and 37°C (data not shown), indicating that cell division was taking place and that the TS cell division phenotype could be rescued by dnaE. Southern blot hybridization with a 2.5-kb XhoI fragment of the pJS14M insert as a probe confirmed that the YB551 (dnaE1100)::pG17 constructs whose phenotypes were observed were integrants at the dnaE locus (data not shown).
We used transduction to confirm that the TS mutation in YB551 (dnaE1100) was linked to dnaE. The dnaE allele in YB551 (dnaE1100) cells was linked to gentamicin resistance by cloning an approximately 2-kb N-terminal PstI fragment of pJS14M into pGSZ to generate pGZPΔC. This plasmid was transferred into both C. crescentus YB551 (dnaE1100) and NA1000 (dnaE+). YB551 (dnaE1100)::pGZPΔC, in which gentamicin resistance is closely linked to the dnaE locus thought to contain the TS mutation, was then used as the donor strain to transduce the gentamicin resistance into NA1000. Eighty-nine percent of the tested transductants were TS, confirming the presence of a TS mutation linked to dnaE in YB551 (dnaE1100). The resulting strain, YB1804 (dnaE1100), was filamentous when it was grown in peptone yeast extract (PYE) medium at 37°C (Fig. (Fig.1H).1H). These same cells grew normally at 30°C; however, they were slightly longer than NA1000 (dnaE+) cells grown at the same temperature (Fig. (Fig.1G),1G), as had been observed with the original mutant, YB551 (dnaE1100).
We carried out a reverse transduction experiment in which NA1000::pGZPΔC was used as the donor strain to transduce the wild-type dnaE allele into the YB551 (dnaE1100) recipient strain. All of the 48 gentamicin-resistant transductants tested were able to divide normally at 37°C (data not shown).
The smooth, filamentous phenotype of the Caulobacter dnaE TS mutant is similar to the phenotypes observed in other DNA replication mutants where DNA synthesis is halted (47). The same filamentous phenotype is observed when the Caulobacter DnaA protein is depleted (20).
We expected that if there was a TS mutation in dnaE, then DNA synthesis levels at the restrictive temperature should be decreased in the TS mutant strains YB551 (dnaE1100) and YB1804 (dnaE1100). We measured the rate of DNA synthesis of the parent wild-type strain used in the original genetic screen, YB1719 (dnaE+), as well as in YB551 (dnaE1100) and PC2179, a holB mutant that is TS for DNA replication elongation (46). Cells were grown in Hutner base-imidazole-buffered glucose-glutamate minimal medium (48) at 30°C until mid-exponential phase and shifted to 37°C, and 1.0-ml samples were removed from the cultures at specific times to be pulse-labeled at 37°C for 2 min with 1 μCi of [8-3H]dGTP diluted in 0.1 ml of medium. The samples were then processed as previously described to measure DNA synthesis at each time point (37). The doubling time of the PC2179 cells in the Hutner base-imidazole-buffered glucose-glutamate medium at 37°C was approximately 180 min, compared to a doubling time of approximately 140 min for both YB551 (dnaE1100) and YB1719 (dnaE+). DNA synthesis remained constant in YB1719 (dnaE+), whereas it was rapidly inhibited in PC2179 (8% at 180 min). In contrast, YB551 (dnaE1100) exhibited a slow reduction in DNA synthesis (56% at 180 min), suggesting that the dnaE mutation did not significantly affect DNA replication elongation but that it might affect initiation (Fig. (Fig.3A).3A). In order to investigate this possibility, swarmer cells were collected by Ludox density gradient centrifugation (12) and released into prewarmed 30 and 37°C M2G medium (22) at an optical density at 660 nm of approximately 0.1. The cells were allowed to recover for 5 min before 1.0-ml samples were removed at different time points to be pulse-labeled with 1 μCi of [8-3H]dGTP and processed to measure DNA synthesis in NA1000 (dnaE+) and YB1804 (dnaE1100) cells at both 30 and 37°C. As expected, wild-type NA1000 at 30 and 37°C and YB1804 (dnaE1100) at 30°C had significantly higher DNA synthesis levels than YB1804 (dnaE1100) at 37°C (Fig. (Fig.3B).3B). The pattern of DNA synthesis in the NA1000 (dnaE+) cells at 30°C mimicked the cell cycle; DNA synthesis was initiated as swarmer cells differentiated into stalked cells and continued to increase until prior to cell division, which occurred after approximately 240 min. DNA synthesis levels in NA1000 (dnaE+) incubated at 37°C followed a similar pattern, except that cells seemed to be lagging slightly behind the 30°C cells in their cell cycle, indicating a slightly lower DNA synthesis rate at the elevated temperature. Cell division in these cells had not occurred by 360 min, the time at which the experiment was stopped. YB1804 (dnaE1100) cells incubated at 30°C began synthesis of DNA as cells differentiated into stalked cells; however, the overall synthesis rate was diminished in these cells compared to that of NA1000 (dnaE+) grown at either 30 or 37°C. Cell division in these cells occurred at approximately 240 min at 30°C. In contrast, DNA synthesis was completely inhibited when YB1804 (dnaE1100) swarmer cells were grown at 37°C, suggesting that the TS mutation in dnaE affected DNA replication initiation (Fig. (Fig.3B).3B). We obtained results similar to those with YB1804 (dnaE1100) and NA1000 (dnaE+) with swarmer cells from YB551 (dnaE1100) and YB1719 (dnaE+), respectively (data not shown). As had been observed previously with YB551 (dnaE1100) synchronized cells, the YB1804 (dnaE1100) swarmer cells continued growth without division, resulting in filamentous cells (data not shown).
A previous study where DNA methylation experiments were used to look at the progression of chromosome replication in YB1804 (dnaE1100) (50) indicates that immediately shifting synchronized cells to 37°C prevented cells from initiating replication but that when swarmer cells were first incubated at the permissive temperature for 45 min and then shifted, the majority of YB1804 (dnaE1100) cells had replicated past a marker 2 kb from the origin of replication by 90 min. Hence, these data support the idea that initiation of DNA replication is inhibited in these cells but that once initiated, the YB1804 (dnaE1100) cells are still proficient for DNA elongation.
We used fluorescence flow cytometry to investigate whether the observed block in DNA synthesis in YB1804 (dnaE1100) cells at the restrictive temperature was at the initiation step of replication (60). YB1804 (dnaE1100) cells, grown overnight at 30°C in PYE medium, were used to inoculate fresh medium in flasks placed at 30 and 37°C to a final optical density at 660 nm of 0.05, and growth was allowed to proceed at these temperatures. Samples were removed from the flasks at 0, 2, 4, and 6 h after the start of the experiment and treated with chloramphenicol at a final concentration of 25 μg/ml for 4 h. The cells were then fixed and stained with chromomycin A3 as described previously (20, 60) and analyzed by using a Becton Dickinson FACStar Plus machine (Stanford University shared fluorescence-activated cell sorter [FACS] facility). The resulting data were analyzed with the FLOWJO program by Tree Star. Treatment of cells with chloramphenicol inhibits the initiation of chromosome replication by interfering with the peptidyl transferase step in protein synthesis but allows ongoing DNA replication to continue to termination. Wild-type cells show two distinct peaks when analyzed by flow cytometry, representing populations of cells with one and two chromosomes (first and second peaks in Fig. Fig.4A,4A, respectively). YB1804 (dnaE1100) cells incubated at the permissive temperature displayed this pattern of distribution of DNA staining (Fig. (Fig.4A).4A). In contrast, although YB1804 (dnaE1100) cells incubated at 37°C originally had populations of cells with one or two chromosomes (Fig. (Fig.4B),4B), within 4 h at this restrictive temperature the majority of cells had only one chromosome (Fig. (Fig.4B),4B), suggesting that these cells were blocked at the initiation step of DNA replication. It is also apparent that there was a smaller population of cells which had not completed DNA replication after 4 h at 37°C, suggesting that there was also a partial block of DNA elongation at the nonpermissive temperature in the YB1804 (dnaE1100) cells.
The entire dnaE gene in YB1804 (dnaE1100) was sequenced in order to determine the location of the TS mutation. Direct sequencing of both strands of overlapping PCR fragments (approximately 500 bp in length) was used. Primers were designed by using the wild-type sequence of the dnaE gene from the published sequence of the Caulobacter genome (42). Two separate PCRs were set up for each primer pair for individual sequence analysis.
A single base pair substitution was found at position 1706 (a change from T to A) within residue 569 in the DnaE protein (Fig. (Fig.2C),2C), which changed the valine residue in the wild-type sequence to a glutamic acid residue in YB1804 (dnaE1100). This sequence change was observed on both strands and was confirmed to be a change from the wild-type sequence when this region of the dnaE gene from NA1000 (dnaE+) DNA was sequenced. Figure Figure2C2C illustrates that this valine lies within a region near the middle of the protein and corresponds to residue 564 of E. coli DnaE (58). In order to confirm that the TS phenotype observed in YB1804 (dnaE1100) cells is indeed due to the single base pair substitution at position 1706 of the dnaE gene, we replaced the wild-type dnaE gene in NA1000 (dnaE+) with the dnaE1100 allele. We used a high-fidelity DNA polymerase (Pfu Turbo; Stratagene) to amplify the dnaE1100 gene from YB1804 genomic DNA. The fragment was then cloned into the vector pNPTS138, which carries both the gene for kanamycin resistance and the sacB gene, which encodes sucrose sensitivity. The resulting plasmid (pNPTS138/dnaE1100) was sequenced to confirm that there were no other mutations introduced into the cloned dnaE1100 gene through PCR and that the single base pair substitution of T to A was still present at position 1706. The plasmid pNPTS138/dnaE1100 was then introduced into wild-type C. crescentus NA1000 (dnaE+) cells by conjugation. pNPTS138 is unable to replicate in Caulobacter; hence, selection for kanamycin-resistant transconjugants selects for integration of the plasmid at the dnaE locus by homologous recombination. This integration creates a merodiploid with a wild-type allele and the dnaE1100 allele. Selection for a second recombination event was carried out by growing kanamycin-resistant transconjugants overnight in PYE medium and then plating cells on PYE-3% sucrose plates. This process selected for loss of the plasmid and created strains that had either a wild-type copy of the dnaE gene or the dnaE1100 allele. Of the 25 sucrose-resistant, kanamycin-sensitive (Sucr Kans) strains that were screened for the TS phenotype, 11 were nonfilamentous when they were grown in PYE medium at 30°C (Fig. (Fig.1I)1I) but were filamentous and showed the same characteristics as the YB1804 (dnaE1100) cells when they were grown at 37°C (Fig. (Fig.1J).1J). Four of these 11 TS strains were sequenced to confirm the presence of the T-to-A substitution at position 1706 within their dnaE alleles. Sequencing of two of the non-TS strains also confirmed that they had a wild-type allele of the dnaE gene. We cloned the wild-type dnaE allele from NA1000 (dnaE+) cells into the plasmid pNPTS138 and integrated the resulting plasmid (pNPTS138/dnaE+) into both YB1804 cells and the four Sucr Kans TS strains (that had been used for sequencing) through homologous recombination, as described above. The resulting transconjugants were complemented by the single copy of the wild-type dnaE gene, with the strains showing normal phenotypes at both the permissive and nonpermissive temperatures (data not shown).
Here, we have isolated and characterized a TS mutant of the C. crescentus dnaE gene encoding the α subunit of Pol III, the principal chromosomal replication enzyme in many bacteria. Since our screen had been designed to obtain mutations in ftsZ, which is not linked to dnaE, it is surprising that we obtained a mutation in dnaE. It is possible that the conjugation of pΔN into NA1000 (dnaE+) transiently increased the mutation rate, giving rise to the dnaE TS mutation (38).
Our data support the hypothesis that the defect in the mutated dnaE gene blocks the initiation stage of DNA replication and still allows DNA elongation to occur, but at a reduced rate, with the mutated DnaE protein being only partially active at 37°C. We hypothesize that in the dnaE TS strains, replication forks that have already initiated DNA replication prior to the shift of cells to 37°C can continue elongation because although the DnaE protein becomes unstable at this temperature, it is stabilized by its interactions with the other subunits of the polymerase complex. Once elongation has been completed, however, further initiation events are blocked in these cells.
Mutations in the dnaE gene might be manifested by the direct loss of catalytic activity or the loss of interaction with another holoenzyme component that is important for productive DNA replication. Experiments with E. coli DnaE have clearly demonstrated that the amino-terminal portion of the protein is responsible for its catalytic activity (28). Additionally, amino- and carboxy-terminal deletions of the E. coli α subunit were used to demonstrate that interactions between the α subunit and the β and τ subunits of the holoenzyme lie within the carboxy terminus of the protein (26, 27). These experiments fused the truncated α subunits to a short biotinylation sequence, enabling the immobilization of the truncated proteins to a streptavidin-coated biosensor chip. The ability of these immobilized truncated fusion proteins to bind the β and τ subunits was measured by using surface plasmon resonance. Gel filtration experiments were also used to probe the α and β interaction. Specifically, the interaction between the α and β subunits was localized to the region between amino acids 542 and 991 of the α subunit (27), and the region between amino acids 542 and 1160 was shown to be important in the interaction of the α subunit with the τ subunit. While residues beyond amino acid 542 were shown to be involved in helping to stabilize the interaction of the α subunit with the τ subunit, a C-terminal deletion of 48 amino acids of the α subunit abolished its ability to interact with the τ subunit (26). Hence, the β- and the τ-subunit-interacting regions of the E. coli α subunit lie within the last 620 amino acids of the protein. The TS mutation in Caulobacter strain YB1804 (dnaE1100) is located at codon 569 of dnaE, corresponding to codon 564 in the E. coli dnaE, and therefore is also found within these important ~620 C-terminal amino acids of the protein.
In the Pol III core, the α subunit associates tightly with the and θ subunits (40). Thus, another possibility is that the defect of the DnaE TS mutant is due to the inability of the core to assemble during DNA replication. Of course, it is possible that, at the restrictive temperature, the DnaE mutation causes the loss of an as-yet-unidentified interaction of DnaE with another DNA replication protein.
Interestingly, DNA replication studies of Streptomyces coelicolor have shown that a TS mutation in the E. coli type α subunit of Pol III present in this organism affects the initiation of DNA replication (13). The position of the mutation corresponds to residue 781 of the E. coli DnaE, within the τ- or β-subunit-interacting regions of the protein, and those researchers speculate that this mutation either prevents assembly of the core after ongoing rounds of DNA replication have been completed or is due to the loss of interaction of α with the τ subunit. Furthermore, the original clone, pJS14M, that complemented the YB551 (dnaE1100) TS strain is missing 84 amino acids of Caulobacter DnaE. Flett et al. made similar observations in their studies with Streptomyces, where a fragment of DnaE lacking the last 117 amino acids was able to complement their dnaE TS mutant strains (13). These data seem to suggest that the corresponding C-terminal amino acids in the Caulobacter and Streptomyces DnaEs are not required for the activity of the α subunits in their interaction with other components of the holoenzyme.
To our knowledge, this is one of the first descriptions of a dnaE mutation that affects the initiation process, and it should help to distinguish the domains and functions of DnaE that are involved in initiation and elongation, respectively. In addition, the isolation of a dnaE mutant deficient in the initiation of DNA replication has allowed us to investigate the role of DNA replication in the regulation of cell division in Caulobacter. We have shown that the initiation of DNA replication is necessary for proper localization of the cell division FtsZ ring but not for its assembly (50) and that transcription of the late cell division genes ftsQ and ftsA requires DNA replication (61).
We thank members of the Brun laboratory and Terry H. Bird for critical reading of the manuscript and Mike O'Donnell for helpful discussions. We especially thank Ann Reisenauer for her invaluable help with flow cytometry.
This work was supported by a National Institutes of Health grant (GM51986) and a National Science Foundation CAREER award (MCB-9733958) to Y.V.B. and by summer development grants through the College of Arts and Sciences at Loyola College to N.D.