Portal de Periódicos da CAPES

Stationary phase induction of dnaN and recF, two genes of Escherichia coli involved in DNA replication and repair

1998; Springer Nature; Volume: 17; Issue: 6 Linguagem: Inglês

10.1093/emboj/17.6.1829

1460-2075

Magda Villarroya,

Epigenetics and DNA Methylation

Article16 March 1998free access Stationary phase induction of dnaN and recF, two genes of Escherichia coli involved in DNA replication and repair Magda Villarroya Magda Villarroya Instituto de Investigaciones Citológicas, Fundación Valenciana de Investigaciones Biomédicas, Valencia, 46010 Spain Search for more papers by this author Ignacio Pérez-Roger Ignacio Pérez-Roger Present address: Imperial Cancer Research Fund, London, WC2A 3PX GB Search for more papers by this author Fernando Macián Fernando Macián Present address: Center for Blood Research, Boston, MA, 02115 USA Search for more papers by this author M.Eugenia Armengod Corresponding Author M.Eugenia Armengod Instituto de Investigaciones Citológicas, Fundación Valenciana de Investigaciones Biomédicas, Valencia, 46010 Spain Search for more papers by this author Magda Villarroya Magda Villarroya Instituto de Investigaciones Citológicas, Fundación Valenciana de Investigaciones Biomédicas, Valencia, 46010 Spain Search for more papers by this author Ignacio Pérez-Roger Ignacio Pérez-Roger Present address: Imperial Cancer Research Fund, London, WC2A 3PX GB Search for more papers by this author Fernando Macián Fernando Macián Present address: Center for Blood Research, Boston, MA, 02115 USA Search for more papers by this author M.Eugenia Armengod Corresponding Author M.Eugenia Armengod Instituto de Investigaciones Citológicas, Fundación Valenciana de Investigaciones Biomédicas, Valencia, 46010 Spain Search for more papers by this author Author Information Magda Villarroya1, Ignacio Pérez-Roger2, Fernando Macián3 and M.Eugenia Armengod 1 1Instituto de Investigaciones Citológicas, Fundación Valenciana de Investigaciones Biomédicas, Valencia, 46010 Spain 2Present address: Imperial Cancer Research Fund, London, WC2A 3PX GB 3Present address: Center for Blood Research, Boston, MA, 02115 USA ‡M.Villarroya and I.Pérez-Roger contributed equally to this work *Corresponding author. E-mail: [email protected] The EMBO Journal (1998)17:1829-1837https://doi.org/10.1093/emboj/17.6.1829 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info The β subunit of DNA polymerase III holoenzyme, the Escherichia coli chromosomal replicase, is a sliding DNA clamp responsible for tethering the polymerase to DNA and endowing it with high processivity. The gene encoding β, dnaN, maps between dnaA and recF, which are involved in initiation of DNA replication at oriC and resumption of DNA replication at disrupted replication forks, respectively. In exponentially growing cells, dnaN and recF are expressed predominantly from the dnaA promoters. However, we have found that stationary phase induction of the dnaN promoters drastically changes the expression pattern of the dnaA operon genes. As a striking consequence, synthesis of the β subunit and RecF protein increases when cell metabolism is slowing down. Such an induction is dependent on the stationary phase σ factor, RpoS, although the accumulation of this factor alone is not sufficient to activate the dnaN promoters. These promoters are located in DNA regions without static bending, and the −35 hexamer element is essential for their RpoS-dependent induction. Our results suggest that stationary phase-dependent mechanisms have evolved in order to coordinate expression of dnaN and recF independently of the dnaA regulatory region. These mechanisms might be part of a developmental programme aimed at maintaining DNA integrity under stress conditions. Introduction The 40.6 kDa β subunit of DNA polymerase III, the major replicase of the Escherichia coli chromosome, is a sliding DNA clamp responsible for tethering the polymerase to DNA and endowing it with high processivity (for a review, see Kelman and O'Donnell, 1995). The structural gene for β, dnaN, maps between the dnaA and recF genes, at 83 min on the E.coli chromosome map (Figure 1). The dnaA gene product is essential for initiation of DNA replication at the bacterial chromosomal origin, oriC, serving as a major regulator of the timing of initiation (for a review, see Messer and Weigel, 1996). Gene recF codes for a DNA-binding protein involved in recombination, repair and in the resumption of DNA replication at disrupted replication forks (Courcelle et al., 1997). Recently, it has been reported that the dnaN gene contains an internal in-frame gene, termed dnaN*, that encodes a smaller form of the β subunit (Skaliter et al., 1996). The novel 26 kDa protein, called β*, is synthesized in much smaller amounts than the β subunit and it could be involved in a recovery function such as DNA repair. Figure 1.Schematic diagram of dnaA, dnaN, dnaN* and recF organization. Coordinate 0 is the middle of the EcoRI site located at the beginning of the dnaA structural gene (Ohmori et al., 1984). Structural genes dnaA, dnaN, dnaN* and recF start at −63, 1346, 1748 and 2446, respectively. The main transcriptional start points are at −297 and −214 (for dnaA promoters p1 and p2), 934, 1077, 1105 and 1319 (for dnaN promoters p1–p4), 1377 (for promoter p5), 1718 (for dnaN* promoter) and 1806 and 1826 (for recF promoters p2 and p1, respectively). T1, T2 and T3 are transcription termination sites responsible for transcriptional polarity of the dnaN–recF operon (Armengod et al., 1991). Download figure Download PowerPoint In exponentially growing cells, dnaN and recF are expressed predominantly from transcripts starting at the dnaA promoters (Pérez-Roger et al., 1991); however, four promoters for dnaN (dnaNp1–dnaNp4 in Figure 1) have been detected in the second half of the dnaA structural gene (Armengod et al., 1988; Quiñones and Messer, 1988) and two major recF promoters (recFp2p1 in Figure 1) were localized in the middle region of the dnaN gene (Armengod and Lambíes, 1986). We have previously described that the cumulative transcriptional activity of the dnaA promoters, when monitored in logarithmic phase, is 10 and three times higher than that shown by the DNA regions containing the dnaN and recF promoters, respectively (Macián et al., 1994). However, we observed in our experiments that transcriptional activity of these regions increased during transition from the exponential to the stationary phase of growth. It is well known that E.coli responds to the onset of stationary phase by inducing the synthesis of proteins which are thought to be important for viability during prolonged periods of starvation (for recent reviews, see Hengge-Aronis, 1996a; Huisman et al., 1996). Stationary phase induction of some genes depends on rpoS, which encodes an alternative sigma factor, σS or σ38, able to recognize a set of promoters that are poorly recognized in vivo by RNA polymerase containing the major vegetative σ-factor σD, or σ70. On the other hand, the promoter selectivity of RNA polymerase may also be controlled in the stationary phase by modulation of the core enzyme (Ozaki et al., 1992). In this work, we characterize the expression pattern of the dnaA operon genes throughout the growth curve. Interestingly, stationary phase-dependent mechanisms seem to have evolved in order to coordinate expression of the dnaN and recF genes independently of the dnaA regulatory region. Results Transcriptional activity of the dnaN and recF promoter regions is dependent on growth phase but independent of growth rate To study the activity of the dnaA, dnaN and recF promoter regions throughout the growth curve, we constructed lacZ transcriptional fusions in single copy on the chromosome as previously described (Macián et al., 1994). Figure 2A and B shows that the β-galactosidase activity of fusions containing the dnaN and recF promoter regions increased 100- and 44-fold, respectively (from 1 to 116 and from 5 to 220 Miller units), during transition into stationary phase. Interestingly, this induction is dependent mostly on rpoS in the case of the dnaN promoter region but not in the case of the DNA region containing the recF promoters. Figure 2C also shows that the β-galactosidase activity of the fusion carrying the dnaA promoters only increases ∼4-fold during transition into the stationary phase (from 60 to 250 Miller units) and that this small induction is not dependent on rpoS. Figure 2.Growth phase-dependent induction of the dnaA, dnaN and recF promoter regions. Strains were MC4100 (wild-type, □) and RH90 (rpoS359::Tn10, ▪) carrying lacZ transcriptional fusions λIC482 (A), λIC451 (B) and λIC549 (C). Optical densities (thin lines) and β–galactosidase activities (thick lines) were determined along the growth curve. Download figure Download PowerPoint Since expression of some stationary phase-inducible genes is inversely correlated with growth rate (Aldea et al., 1990; Yamagishi et al., 1993), we examined the transcriptional activity of the dnaN and recF promoter regions during exponential growth in various media. Thus, we measured the β-galactosidase production of MC1000 cells lysogenized with λIC482 (dnaN::lacZ), λIC451 (recF::lacZ) and, as a control, λMAV103 (bolA::lacZ) during exponential growth in LBT (doubling time: 25 min), M9Caa (doubling time: 60 min) and M9LT (doubling time: 120 min) media. The results obtained (data not shown) indicated that the activity of the dnaN and recF promoter regions is independent of growth rate, whereas expression of bolA increases with decreasing growth rates, as described previously by Aldea et al. (1990). Therefore, it can be concluded that the growth phase induction of the dnaN and recF promoter regions (Figure 2A and B) is not a consequence of the gradual reduction in growth rate that takes place during the growth phase transition. Dissectional analysis of the dnaN regulatory region Four promoters for dnaN (p1–p4) have been detected previously by using transcriptional fusions, S1 mapping of mRNAs, deletion analysis and in vivo dnaN complementation tests (Armengod et al., 1988). In the same work, the presence of a weak promoter, named p5, at the beginning of the dnaN structural gene was also reported. In order to identify promoters controlled by RpoS at the dnaN regulatory region, we have studied the growth phase induction of dnaN transcripts using the S1 nuclease protection assay. Results from Figure 3 indicate that the same start sites of transcription are detected when using total cellular RNA obtained during both exponential and stationary phases of growth. Therefore, stationary phase induction of dnaN depends on some of the previously detected promoters. The autoradiograph shown in Figure 3 also suggests that the activity of the dnaNp1 and dnaNp2 promoters is growth phase inducible and RpoS dependent. It is relevant to point out here that any growth phase effects on transcripts started at dnaNp4 might be underestimated from results such as those shown in Figure 3 since these transcripts seem to be particularly reluctant to S1 protection analysis; other authors have been unable to detect them by this method (Quiñones and Messer, 1988; Tadmor et al., 1994). Figure 3.Transcription initiation sites of the dnaN promoters. Total cellular RNA was isolated from MC4100 (rpoS+) and RH90 (rpoS−) cells during exponential (Ex) or stationary (St) growth in LBT. The DNA probe was a 1401 bp EcoRI–RsrII fragment (extending from nucleotide −3 to 1399 in Figure 1) labelled at the 5′ end of the RsrII site. In lane 1, a labelled HaeII digest of pBR322 DNA is shown as a reference marker, and the sizes (in base pairs) of some fragments are indicated on the left. Lane 2 shows the DNA probe without S1 treatment. Samples were run on a 5% polyacrylamide, 8 M urea gel. Download figure Download PowerPoint To investigate further which promoters at the dnaN regulatory region are RpoS dependent, we have carried out a dissectional analysis of this region so that the transcriptional activity of each one of the dnaN promoters could be monitored independently by using transcriptional fusions to the lacZ reporter gene. Data from Table I indicate that P1 (λIC715), P2 (compare λIC687 and λIC695) and P4 (compare λIC693 and λIC699) are strongly induced by entry into the stationary phase and that this induction is mostly RpoS dependent. Note that exclusion of the −35 region of P1, P2 and P4 in λIC797, λIC695 and λIC699 respectively, greatly reduces the RpoS induction of these constructs. This suggests that nucleotides in the −35 regions of these promoters play a prominent role in RpoS-dependent transcription. Table 1. Dissectional analysis of the dnaN promoter region Phage Regulatory elementsa β-Galactosidase activity (Miller units)b rpoS+ rpoS− ODc 1.7 OD 1.7 λIC482 P1–P5 1.3 116 1.0 18 λIC715 P1 0.6 34 0.3 3 λIC797 (−10)P1 0.3 3 0.3 1 λIC687 P2, (−35)P3 0.7 26 1.0 4 λIC695 (−10)P2, P3, (−35)P4 0.2 5 0.4 3 λIC693 P4, P5 0.9 58 0.6 9 λIC699 (−10)P4, P5 0.8 7 0.9 6 λIC701 P5 0.2 4 0.5 4 a dnaN promoters present on the chromosomal fragment fused to lacZ. Some of the DNA fragments inserted into the promoter probe vector contain a truncated promoter; the −35 or −10 box remaining in the cloned fragment is shown in parentheses. b β-Galactosidase activity was measured in MC4100 (rpoS+) and RH90 (rpoS−). c Optical density at 600 nm. Recently, a promoter consensus for σ38 consisting of the −10 sequence CTATACT and an upstream region with intrinsic DNA curvature has been proposed (Espinosa-Urgel et al., 1996). Although we have not tested whether the dnaN promoters are recognized in vitro by σ38, we have examined these promoters for the presence of a curved DNA structure since homology to the consensus −10 can be found. To do this, we have analysed, according to Espinosa-Urgel and Tormo (1993), the electrophoretic mobility of DNA fragments encompassing the dnaA and dnaN promoters on polyacrylamide gel electrophoresis at low temperature, because it is well established that, under such conditions, DNA fragments containing curved DNA show anomalous electrophoretic mobility. Curiously, we have found that fragments containing the dnaA promoters exhibit anomalous mobility at 4°C but that those including the dnaN promoters do not (data not shown). Computer predictions of DNA bending performed according to Espinosa-Urgel and Tormo (1993) support the conclusion that the dnaA promoter region contains a curved DNA structure whereas the dnaN promoters seem to be located in DNA regions without static bending. Accumulation of RpoS is not sufficient for induction of dnaN The failure to induce a dnaN::lacZ fusion in the rpoS mutant (see Figure 2A) was complemented by the presence of a wild-type allele of rpoS in trans on plasmid pIC747 (data not shown). This agrees with the conclusion that the rpoS gene is directly or indirectly required for the growth phase-dependent activation of the dnaN promoter region. However, it should be pointed out that rpoS is under control of the tac promoter in pIC747 and that rpoS overexpression mediated by isopropyl-β-D-thiogalactopyranoside (IPTG) was not able to induce activity of the dnaN promoter region during the exponential phase. This result might be due to the fact that the RpoS protein is degraded rapidly by the ClpXP protease during exponential growth (Schweder et al., 1996); degradation slows down significantly during stationary phase, allowing RpoS to accumulate (Muffler et al., 1996; Pratt and Silhavy, 1996). Since it has been shown recently that RpoS accumulation also occurs at low temperature during exponential growth, and that this accumulation produces induction of the bolAp1 promoter (Sledjeski et al., 1996), we decided to test whether the RpoS-dependent activity of the dnaN promoter region was also induced under such conditions. We have found (data not shown) that, as expected, the exponential phase activity of a bolA::lacZ transcriptional fusion (λMAV103) increased with decreasing temperatures (from 45 to 325 Miller units) in an RpoS-dependent manner; however, the β-galactosidase activity of a dnaN:: lacZ transcriptional fusion (λIC482) during exponential growth was similar at 20 and 37°C (∼2 Miller units). Therefore, RpoS accumulation is not sufficient for induction of the dnaN promoter region, and some additional factor or condition may be required to activate the dnaN promoters. Osmotic induction of the dnaN promoters The cellular level of σ38 also increases during the exponential growth phase in response to high osmolarity, and this produces induction of many σ38-dependent genes (for a review, see Hengge-Aronis, 1996b). Therefore, we have analysed whether osmotic accumulation of RpoS is able to activate the dnaN promoters. As shown in Figure 4, dnaNp1, dnaNp2 and dnaNp4 are induced in exponentially growing cells in response to moderate hyperosmotic conditions that only slightly reduce the growth rate (the doubling time changes from 40 to 45 min). The same kind of experiments performed in strain RH90 (rpoS359::Tn10) revealed that rpoS is required for osmotic induction of the dnaN promoters (data not shown). The bolAp1 promoter was used as a positive control in our experiments (data not shown) because osmotic induction of this σ38-dependent promoter has been reported previously (Hengge-Aronis et al., 1993). Figure 4.Osmotic induction of the dnaN promoters. Cells were grown overnight at 37°C in rich A medium with or without 15% sucrose, subcultured 1:100 into the same medium, and grown until they reached the cell densities indicated on the abscissa. The strain used was MC4100 lysogenized with λIC482 (A), λIC715 (B), λIC687 (C) and λIC693 (D). □: A media; ▪: A media + 15% sucrose. Download figure Download PowerPoint Taken together, our results indicate that the RpoS-dependent induction of dnaN may be triggered by both stationary phase and hyperosmotic conditions but not by incubation at low temperature. dnaN promoters are used to accomplish discoordinate regulation of the dnaA operon genes Transcription initiated at the dnaA promoters is controlled by a complex network of positive and negative regulatory elements. Several sequences have been found in the dnaA promoter region which are recognized by specific proteins capable of repressing or activating transcription from the dnaA promoters (Froelich et al., 1996; Lee et al., 1996; Messer and Weigel, 1996). In addition, termination of transcription occurs along dnaA (Wende et al., 1991; Pérez-Roger et al., 1995; compare also λIC549 and λIC638 in Figure 5). In spite of this, expression of dnaN depends mainly on transcription started at the dnaA promoters in exponentially growing cells because, under such conditions, the activity of the dnaN promoters is very low (Macián et al., 1994). As shown in Figure 5, this situation changes drastically when cells enter the stationary phase. At that time, the transcriptional activity of the dnaN promoter region increases 100-fold and becomes higher than that produced from the dnaA promoters (compare λIC482 and λIC638). This suggests that in stationary phase cells, expression of dnaN is largely dependent on transcription started at its own promoters. Figure 5.The dnaN promoters are used to accomplish discoordinate regulation of the dnaA operon genes. The horizontal bars below the map represent the chromosomal fragment fused to lacZ in λRZ5 derivatives. aExpressed as Miller units. The strain used was MC1000. bOptical density at 600 nm. Download figure Download PowerPoint Translational expression of the dnaN and recF genes To analyse further the response of the dnaA operon distal genes to entry into stationary phase, we have made use of translational fusions to lacZ. As shown in Table II, the β-galactosidase activity of pIC785 and pIC371, two translational dnaN::lacZ fusions, increases >60-fold in starved (stationary) cells. Note that pIC371 is a late translational dnaN::lacZ fusion containing most of the dnaN open reading frame (ORF); therefore, it could synthesize both DnaN::LacZ and DnaN*::LacZ proteins (see Figure 1). Since it has been reported that accumulation of β* occurs during stationary phase (Skaliter et al., 1996), we have examined induction of protein fusions by Western blot analysis using polyclonal antibodies against β-galactosidase; no DnaN*::LacZ protein could be detected from extracts of pIC371-bearing cells in conditions which allowed clear detection of the DnaN::LacZ protein (data not shown). This indicates that most of the β-galactosidase activity exhibited by pIC371 is due to the DnaN::LacZ protein. Therefore, considering that the induced activities of the early and late translational DnaN::LacZ fusions are similar (compare pIC785 and pIC371 in Table II), and assuming that the level of β-galactosidase activity reflects the translation of β, it may be concluded that accumulation of the processivity factor of DNA polymerase III holoenzyme, the β subunit, occurs in cells whose metabolism is slowing down. Table 2. Growth phase effects on expression of dnaN::lacZ and recF::lacZ translational fusions Plasmid Fragmenta Fusionb β-Galactosidase/β-lactamasec ODd 1.7 pIC785 (−3)–1725 dnaN 25 1700 pIC371 (−3)–2419 dnaN,dnaN* 23 1600 pIC356 (−3)–2586 recF 0.5 6 pIC361 1490–2586 recF 0.03 0.03 a Chromosomal fragment inserted into pMLB1034. The numbering system is that of Figure 1. b Gene translationally fused to lacZ. c The ratio of β-galactosidase units to β-lactamase units. The strain used was MC4100 (rpoS+). d Optical density at 600 nm. pIC356 is a translational recF::lacZ fusion where both the dnaN and recF promoters are present. As shown in Table II, synthesis of RecF::LacZ protein from pIC356 is induced during entry into the stationary phase of growth. Although we have found that the transcriptional activity of the recF promoter region is induced in starved cells (Figure 2B), induction of the RecF::LacZ protein from pIC356 seems to be entirely dependent on the dnaN promoters (compare pIC356, which contains the dnaN and recF promoters, with pIC361, which only includes the recF promoters). This is because terminators T1, T2 and T3 (see Figure 1), which drastically reduce transcription started at the recF promoters (Armengod et al., 1991), remain active during transition into the stationary phase (data not shown). On the other hand, it should be noted that the β-galactosidase activity mediated by the dnaN promoters is ∼50- to 300-fold lower when determined with pIC356 (the recF::lacZ protein fusion) than when using pIC785 or pIC371 (the dnaN::lacZ protein fusions). This suggests that the RecF and β (DnaN) proteins are differentially expressed from transcripts started at the dnaN promoters and that additional regulatory mechanisms modulate the synthesis of the recF gene product in both the exponential and stationary phases of growth. Discussion In the present work, we have characterized the pattern of expression of the dnaA–dnaN–recF operon throughout the growth curve, and demonstrate that the dnaN promoters are used to accomplish discoordinate regulation of the operon genes. We began this investigation with the observation that the transcriptional activity of the dnaN and recF promoter regions increased during transition from the exponential to the stationary phase of growth. Our previous results indicated that, in exponentially growing cells, the dnaA promoters are ∼10- and 4-fold stronger than the promoters for the dnaN and recF genes, respectively (Macián et al., 1994). This was in agreement with the assessment that, under such conditions, expression of dnaN and recF depends mostly on transcription started at the dnaA promoters (Pérez-Roger et al., 1991). Here we show that transcriptional activity of the dnaN promoter region increases 100-fold during transition into stationary phase, which has important effects on the expression pattern of the dnaA operon genes. Dissectional analysis of the dnaN regulatory region indicates that the dnaNp1, dnaNp2 and dnaNp4 promoters are induced strongly during entry into the stationary phase (Table I) or under hyperosmotic conditions (Figure 4) and that this induction is mostly RpoS dependent. Our results are consistent with the observation that although RpoS accumulation is necessary for expression of target promoters, it is not always sufficient (for references, see Sledjeski et al., 1996; Muffler et al., 1997). We propose that some specific factor(s) and/or condition(s) that are generated by osmotic stress or starvation (stationary phase), but not by incubation at low temperature, are needed in addition to RpoS accumulation in order to stimulate the activity of the dnaN promoters. It is noteworthy that trehalose and K+ glutamate, which accumulate under stationary phase and hyperosmotic conditions, specifically stimulate the activity of σ38-containing RNA polymerase at σ38-dependent promoters (Ding et al., 1995; Kusano and Ishihama, 1997). The induction of the dnaN gene reported here suggests that the β clamp, or the DNA polymerase III holoenzyme, is required for maintenance of DNA integrity under stationary phase and high osmolarity conditions. The establishment of a consensus structure for promoters recognized by σ38 is still a controversial matter. In contrast to what has been described for other RpoS-dependent promoters (Espinosa-Urgel et al., 1996), we have not found static bending associated with the dnaN promoters. Moreover, the fact that the same S1 nuclease-resistant DNA fragments were detected when using total RNA obtained from both rpoS+ and rpoS− strains (Figure 3) indicates that the dnaNp1, dnaNp2 and dnaNp4 promoters can be recognized in vivo by Eσ70. Curiously, DNA fragments including the dnaN promoters exhibited very weak binding to RNA polymerase containing σ70 (Armengod and Lambíes, 1986), probably because they deviate ∼50% from the consensus sequences recognized by this holoenzyme (Armengod et al., 1988), and one of them, dnaNp4, has been shown to be unable to direct in vitro transcription in the presence of 100 mM KCl (Ohmori et al., 1984). Therefore, it is possible that Eσ70 requires ancillary factors and/or specific conditions in order to initiate transcription at the dnaN promoters. Since we have not tested whether the dnaN promoters are recognized in vitro by Eσ38, we cannot exclude the possibility that RpoS plays an indirect role in regulation of the dnaN promoters; thus, a gene(s) under the control of RpoS might encode a transcriptional factor(s) able to help Eσ70 to initiate transcription at the dnaNp1, dnaNp2 and dnaNp4 promoters. It is noteworthy in this respect that deletion of the −35 consensus region, which is usually needed for the recognition of promoters by Eσ70, greatly reduces the RpoS induction of the dnaN promoters. Alternatively, Eσ38 might recognize the dnaN promoters in spite of their lack off static bending. In relation to this, it has been shown that the presence of upstream curved sequences is not essential for recognition of the galP1 promoter by Eσ38 and that DNA curvature probably contributes to promoter strength rather than to σ38-specific recognition (Kolb et al., 1995). Our results also show that recF expression is induced during entry into stationary phase and that such an induction is dependent on the dnaN promoters (Table II). This coordinate induction of dnaN and recF suggests that both genes are involved in the same process. Recently, Courcelle et al. (1997) have shown that when replication is disrupted by UV lesions, recF is required for the resumption of replication at DNA replication forks. These authors propose that the replisome disintegrates at a DNA lesion and then recF is required to re-assemble a replication holoenzyme at the site of a DNA replication fork. Since re-assembly of a processive holoenzyme should require the participation of β, coordinate induction of dnaN and recF might facilitate recovery of DNA replication at stalled replication forks. Interestingly, expression of β has been shown to be induced by DNA-damaging agents (Kaasch et al., 1989; Quiñones et al., 1989; Tadmor et al., 1994). On the other hand, there are some indications that a replication fork stalls at some frequency and disintegrates during a normal course of replication (Bierne and Michel, 1994; Canceill and Ehrlich, 1996). Moreover, Courcelle et al. (1997) have presented results suggesting that replication can be disrupted in the absence of DNA damage. Thus, disruption appears more frequent in a thy− background, presumably because the efficiency of processing and delivery of thymine nucleotides to the replication machinery is compromised in the auxotrophs. Also here, recF is required for the re-assembly of a replication holoenzyme at a DNA fork. If stalling of DNA replication was more frequent during entry into the stationary phase, then coordinate induction of dnaN and recF would be required to guarantee that the round of replication already under way can be completed. Kolter et al. (1993) have postulated that the cessation of growth could halt key metabolic processes, DNA replication in particular, at stages at which severe and irreparable damage could occur. Then, in order to ensure their survival, E.coli cells should be able to make an orderly transition into stationary phase such that the cell cycle would not be arrested randomly. If this were the case, coordinate indu