HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE 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 Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via CrossRef Citing Articles via Google Scholar Google Scholar Articles by Narajczyk, M. Articles by Wegrzyn, G. Search for Related Content PubMed PubMed Citation Articles by Narajczyk, M. Articles by Wegrzyn, G. Agricola Articles by Narajczyk, M. Articles by Wegrzyn, G.

Modulation of plasmid and phage DNA replication by Escherichia coli SeqA protein

Sylwia Baraska¹

Alicja Wgrzyn²

Grzegorz Wgrzyn¹

¹ Department of Molecular Biology, University of Gdask, 80-822 Gdask, Poland
² Laboratory of Molecular Biology (affiliated with University of Gdask), Institute of Biochemistry and Biophysics, Polish Academy of Sciences, 80-822 Gdask, Poland

Correspondence
Grzegorz Wgrzyn
wegrzyn{at}biotech.univ.gda.pl

	ABSTRACT

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
SeqA protein, a main negative regulator of the replication initiation of the Escherichia coli chromosome, also has several other functions which are still poorly understood. It was demonstrated previously that in seqA mutants the copy number of another replicon, the plasmid, is decreased, and that the activity of the p_R promoter (whose function is required for stimulation of ori) is lower than that in the wild-type host. Here, SeqA-mediated regulation of phage and plasmid replicons was investigated in more detail. No significant influence of SeqA on ori-dependent DNA replication in vitro was observed, indicating that a direct regulation of DNA replication by this protein is unlikely. On the other hand, density-shift experiments, in which the fate of labelled DNA was monitored after phage infection of host cells, strongly suggested the early appearance of replication intermediates and preferential rolling-circle replication of phage DNA in seqA mutants. The directionality of plasmid replication in such mutants was, however, only slightly affected. The stability of the heritable replication complex was decreased in the seqA mutant relative to the wild-type host, but a stable fraction of the O protein was easily detectable, indicating that such a heritable complex can function in the mutant. To investigate the influence of seqA gene function on heritable complex- and transcription-dependent DNA replication, the efficiency of plasmid replication in amino acid-starved relA seqA mutants was measured. Under these conditions, seqA dysfunction resulted in impairment of plasmid replication. These results indicate that unlike oriC, SeqA modulates DNA replication indirectly, most probably by influencing the stability of the replication complex and the transcriptional activation of ori.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
Replication of genetic material is a fundamental process that occurs in all living organisms (Kornberg & Baker, 1992). To ensure survival, this process must be precisely regulated, and this regulation usually occurs at the stage of initiation. In the model prokaryotic organism Escherichia coli, initiation of chromosome replication is controlled in the cell cycle (Messer, 2002). The dnaA gene product is a replication initiator protein. It binds to the oriC region and positively regulates DNA replication initiation. However, in most biological systems, which require precise and versatile regulation to ensure the ability to respond to various growth conditions, there are both positive and negative regulators. The SeqA protein appears to be one of key factors involved in the negative control of oriC-initiated replication (Lu et al., 1994; von Freiesleben et al., 1994; Slater et al., 1995; Boye et al., 1996); namely, it is an inhibitor of the onset of chromosome replication in vivo (Slater et al., 1995; Boye et al., 1996).

It has been demonstrated that in vivo SeqA limits DnaA activity in replication from oriC (von Freiesleben et al., 1994). Nevertheless, it has been proposed that the main role of SeqA is sequestration of newly replicated origin sequences, as this protein interacts with hemimethylated GATC motifs, abundant in the oriC region, and prevents DnaA binding to this region (Lu et al., 1994; Slater et al., 1995). However, subsequent studies have revealed that SeqA-mediated regulation of replication is significantly more complex. Although SeqA inhibits replication from oriC at high concentrations of DnaA in vitro, it stimulates this process at low DnaA concentrations (Wold et al., 1998). Other experimental lines have indicated that SeqA inhibits open complex formation at the replication origin (Torheim & Skarstad, 1999).

Studies on DNA supercoiling and the localization of SeqA in cells have suggested that this protein may affect organization of the nucleoid (Hiraga et al., 1998; Weitao et al., 1999; Skarstad et al., 2001). In seqA mutants, negative supercoiling of DNA increases, and purified SeqA protein generates positive DNA supercoils in vitro (Klungsoyr & Skarstad, 2004). Therefore, it has been proposed that binding of SeqA changes either the twist or the writhe of the DNA (Klungsoyr & Skarstad, 2004). This ability to affect the topology of DNA suggests that SeqA can take part in the organization of the chromosome in vivo. However, it has been proposed that the direct effect of SeqA on replication initiation is rather the result of binding to hemimethylated oriC and releasing DnaA molecules from this region (Taghbalout et al., 2000). Interestingly, several lines of evidence lead to the conclusion that for its proper function, SeqA protein aggregation is required (Lee et al., 2001; Fossum et al., 2003; Han et al., 2003, 2004; Odsbu et al., 2005).

As can be concluded from the data mentioned above, the function of SeqA in replication regulation is definitely not simple. The same applies to the biochemical properties of this protein. Even the rules that govern the binding of SeqA to DNA are not completely clear. Using oligonucleotide templates, it has been demonstrated that SeqA binds specifically to hemimethylated GATC sequences, but not to fully methylated and unmethylated GATC motifs (Brendler & Austin, 1999; Kang et al., 1999). However, studies of larger DNA fragments indicate that SeqA may interact specifically with fully methylated oriC fragments (Slater et al., 1995; Skarstad et al., 2000). When other (non-oriC) DNA fragments, consisting of several hundred base pairs, are studied, specific binding of SeqA to fully methylated GATC sites is also evident (Somiska et al., 2001, 2003a, b).

Although the structure of a part of the SeqA protein has been resolved (Guarne et al., 2002), and structure–function studies have been performed (Fujikawa et al., 2003, 2004), some unexpected effects of mutations in the seqA gene have been reported, which strongly suggest that SeqA is involved in the regulation of processes as various as transcription initiation (Somiska et al., 2001, 2003a; Lobner-Olesen et al., 2003), protein degradation (Torheim et al., 2000; Somiska et al., 2003c), and biological functions of cellular membranes (Wgrzyn et al., 1999). Moreover, it has been proposed that the function of this protein in the regulation of DNA replication initiation may be linked to E. coli chromosome segregation and cell division (Bach et al., 2003).

Since viruses are intracellular parasites, they use various host proteins for their propagation. Although the use of products of cellular genes by viruses may sometimes appear unusual, studies of such processes have often led to the discovery of important properties of host proteins. This applies also to bacteriophages and their hosts, including one of the most intensively studied models in molecular biology, phage . Examples of the processes described above are the use of E. coli LamB protein, normally allowing uptake of maltose and maltodextrins, for phage adsorption on the cell surface (Vinga et al., 2006), employment of heat-shock proteins in DNA replication and morphogenesis (Polissi et al., 1995; Taylor & Wgrzyn, 1995), and involvement of the replication initiator protein DnaA in the regulation of transcription from the p_R promoter (Szalewska-Paasz et al., 1998; Glinkowska et al., 2003). Therefore, we aimed to investigate the role of SeqA in the control of replication of the bacteriophage genome and plasmids derived from this phage. In fact, previous studies have demonstrated that seqA gene function is involved in the regulation of bacteriophage development, particularly through the facilitation of transcription stimulation from certain promoters, including p_R, and that this is crucial for the expression of replication genes and for transcriptional activation of ori (Somiska et al., 2001, 2003a; Wgrzyn, 2006). Moreover, plasmid copy number is found to be decreased in seqA mutants relative to wild-type bacteria (Somiska et al., 2001), and it has been suggested that SeqA interferes with DnaA-mediated regulation of DNA replication initiation from ori (Somiska et al., 2003b; Glinkowska et al., 2001). On the other hand, the distribution of SeqA-binding sequences (GATC) at oriC and ori is substantially different (GATC motifs are abundant at oriC but not at ori; for reviews, see Messer, 2002; Wgrzyn, 2006). Thus, we assumed that studies of the effects of SeqA on DNA replication would provide important information, useful to understand in more detail both the functions of this protein and the regulation of bacteriophage development.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
Bacterial strains, phages, plasmids and culture media used in in vivo experiments.
Previously described E. coli wild-type strain MG1655 (Jensen, 1993) was used. The seqA : : Tn10 (called seqA in this report) and dnaA46 seqA : : Tn10 derivatives of this strain were constructed previously by Wgrzyn et al. (1999). The in-frame deletion mutant seqA10 has already been described (Slater et al., 1995). The MG1655-derived relA251 : : kan seqA : : Tn10 strain was constructed by P1 transduction of the seqA : : Tn10 allele from the above-described strain to the relA251 : : kan recipient (Xiao et al., 1991).

Bacteriophage cI857S7 (Goldberg & Howe, 1969) was used. The following plasmids derived from this phage were employed: pKB2 (Kur et al., 1987), pCB104 (Boyd & Sherratt, 1995), pRLM4 (Wold et al., 1982) and pKBlin (Herman-Antosiewicz et al., 1998b). All of them consist of the replication region and a different antibiotic-resistance gene.

LB medium (Sambrook et al., 1989) and minimal medium 2 (MM-2) (Wgrzyn & Taylor, 1992) were used in the in vivo experiments.

Proteins, antibodies and protein fraction.
Bacteriophage O and P proteins were prepared from E. coli strain MM294 (Tabor & Richardson, 1985) bearing plasmids pGP1-2 and pEW1. The latter plasmid was constructed by replacement of the EcoRI–SstII fragment from pIK12 (Konieczny & Marszaek, 1995) with an analogous fragment of pKB2 (Kur et al., 1987). Thus, overexpression of the O and P genes was possible in the T7 RNA polymerase/promoter system (Tabor & Richardson, 1985). O and P proteins were purified as described previously (ylicz et al., 1984). SeqA protein, purified according to a published method (Skarstad et al., 2000), was a kind gift from Dr Kirsten Skarstad, Institute of Cancer Research, Oslo, Norway, as were anti-SeqA antibodies.

Fraction II, an extract containing proteins necessary for DNA replication, was prepared according to an already-published procedure (Fuller et al., 1981). E. coli C600 strain (Appleyard, 1954) was grown in LB medium at 37 °C to OD₆₀₀=1.0. The cells were collected by centrifugation, washed in buffer A (25 mM HEPES/KOH, pH 7.6, 5 mM EDTA) and then resuspended thoroughly in the same buffer (2 ml buffer A per 1 ml cell paste). The cell paste was transferred to a polypropylene tube, quickly frozen in liquid nitrogen and stored at –70 °C. The frozen cell suspension was thawed at 0–4 °C, and KCl, lysozyme and DTT were added to final concentrations of 150 mM, 0.5 mg ml^–1 and 2 mM, respectively. Following incubation at 0 °C for 30 min, the cells were frozen again in liquid nitrogen, thawed at 0–4 °C and centrifuged in a Beckman 50.2 Ti rotor at 30 000 r.p.m. for 30 min at 4 °C. The supernatant was collected and ammonium sulfate (0.28 g per ml cleared lysate) was slowly added to the lysate with stirring over a 60 min period at 0 °C. The precipitate was collected by centrifugation at 23 000 r.p.m. for 20 min at 4 °C and packed into a dialysis tube. The pellet was resuspended in 100 µl buffer B (25 mM HEPES/KOH, pH 8.0, 0.1 mM EDTA, 2 mM DTT) and dialyzed against 200 ml buffer B at 0 °C for 30 min. The fraction extract was frozen in aliquots in liquid nitrogen and stored at –70 °C.

In vitro DNA replication.
The standard reaction mixture (final volume 25 µl) consisted of: 40 mM HEPES/KOH, pH 7.6, 11 mM magnesium acetate, 50 µg BSA ml^–1, 40 mM creatine phosphate, 20 µg creatine kinase ml^–1, 2 mM ATP, 0.5 mM GTP, 0.5 mM CTP, 0.5 mM UTP, 0.1 mM dATP, 0.1 mM dGTP, 0.1 mM dCTP, 0.1 mM dTTP with [³H]dTTP (150 c.p.m. pmol^–1), 8.6 µg PEG 6000 ml^–1, 250 ng supercoiled pRLM4 plasmid DNA [purified by ultracentrifugation in a cesium chloride/ethidium bromide gradient, as described elsewhere (Sambrook et al., 1989)], 300 ng O protein and 80 ng P protein. SeqA protein was either absent or added to the concentrations indicated. The mixtures were assembled in an ice bath. The reactions were started by the addition of 1.5 µl fraction II and then the samples were incubated at 32 °C for 2 h. In some experiments, 1.5 µl undiluted anti-SeqA serum was added to fraction II and incubated for 12 min in an ice bath before mixing with other components. Total ³H-labelled nucleotide incorporation was measured by determining the radioactivity (in a scintillation counter) of samples after TCA precipitation and filter binding.

Density-shift experiments.
Density-shift experiments were performed according to a previously published method (Wgrzyn et al., 1995b). Briefly, bacteria were grown in a ‘light’ MM-2 overnight at 37 °C, and after dilution (1 : 50, v/v) with fresh medium the cultivation was prolonged to OD₅₀₀=0.2. The bacteria were sedimented, washed with TM buffer (10 mM Tris/HCl, pH 7.2, 10 mM MgSO₄) and suspended with 0.1 volume of this buffer. After 60 min incubation at 37 °C, the [³H]thymidine-labelled phage (9x10^–5 c.p.m. p.f.u.^–1) was added to m.o.i.=10 and incubation was continued for 15 min. The suspension was sedimented, resuspended in the original volume of prewarmed (to 37 °C) ‘heavy’ minimal medium (containing [¹⁵N]NH₄Cl and [¹³C]glucose instead of NH₄Cl and glucose, respectively), and further incubation was performed at 37 °C. Samples of the infected culture were withdrawn at the indicated times, and total DNA was isolated and ultracentrifuged in the CsCl density gradient, as described previously (Wgrzyn et al., 1995b). Fractions were collected from the bottom of the tube and the radioactivity of each fraction was measured in a scintillation counter.

2D-agarose gel electrophoresis (AGE).
Analysis of plasmid replication intermediates by 2D-AGE was performed according to a method described elsewhere (Viguera et al., 1996), with modifications described subsequently (rutkowska et al., 1999). Before electrophoresis, plasmid pCB104 was digested with EcoRI, EcoRV or SspI/PvuI.

Electron microscopy.
Electron microscopy analysis of replicating plasmid DNA molecules was performed as described previously (Burkardt & Lurz, 1984; rutkowska et al., 1998).

Estimation of O protein stability.
The stability of the O protein in E. coli cells was investigated, as described previously (Wgrzyn et al., 1995a), by [³⁵S]methionine labelling of bacteria growing in MM-2 medium for 20 min, followed by chasing with an excess of unlabelled L-methionine (1 mg ml^–1), cell lysis, immunoprecipitation with anti- O serum, SDS-PAGE, autoradiography and densitometry.

Plasmid DNA replication in amino acid-starved bacteria.
Bacteria were grown in MM-2 medium, and isoleucine starvation was induced by addition of L-valine to a final concentration of 1 mg ml^–1. Samples of bacterial cultures were withdrawn at indicated times, plasmids were isolated from cells by alkaline lysis, and the amount of plasmid DNA was estimated after AGE and densitometric analysis of plasmid bands on an electrophoregram, as described previously (Herman-Antosiewicz et al., 1998b).

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
In vitro DNA replication in the presence and absence of SeqA
The SeqA protein regulates replication initiation from oriC directly, as demonstrated in studies of the in vitro replication of E. coli minichromosomes (Slater et al., 1995; Wold et al., 1998; Torheim & Skarstad, 1999). Since a decreased copy number of plasmids derived from bacteriophage in seqA mutants has been reported previously (Somiska et al., 2001), we asked whether the influence of SeqA on replication of these replicons was direct or indirect. Note that the only specific effect of SeqA reported to date that could influence DNA replication was the regulation of p_R promoter activity (Somiska et al., 2001).

In vitro replication experiments were performed using a plasmid supercoiled DNA template, purified O and P proteins, and a fraction of replication proteins (called fraction II). In such an experimental system, transcription from p_R is effective due to the presence of RNA polymerase in fraction II (Wold et al., 1982; for a review, see Taylor & Wgrzyn, 1995). In our experiments we used plasmid DNA template isolated from wild-type E. coli cells growing exponentially under standard laboratory conditions. DNA isolated from such bacteria should theoretically be a mixture of hemimethylated and fully methylated molecules. However, assuming that replication of a plasmid molecule takes less than 1 min, and considering the kinetics of DNA methylation, it appeared that fully methylated plasmid DNA was predominant in the preparations. On the other hand, it has previously been demonstrated that both the in vivo and in vitro effects of the seqA mutation and SeqA protein on plasmid copy number and p_R activity, as well as the binding of SeqA to the DNA replication region, are of equal efficiency and of the same specificity with respect to both hemimethylated and fully methylated DNA (Somiska et al., 2001, 2003b).

We observed no significant effects of SeqA on in vitro replication initiated from ori. First, because our Western-blotting analysis revealed that fraction II contains some SeqA protein (data not shown), we tested the effects of neutralization of this protein by anti-SeqA antibodies on DNA replication. No significant differences were observed in the efficiency of DNA synthesis in the presence and absence of anti-SeqA serum (synthesis in the presence of the serum was 86±6 % of that measured in the absence of anti-SeqA antibodies). Second, in contrast to previous in vivo studies (Somiska et al., 2001), the in vitro results presented in this report indicated a slight inhibition rather than stimulation of DNA replication by SeqA. Similar results were obtained irrespective of whether SeqA was pre-incubated with DNA template or added together with other replication proteins (Fig. 1). These results show that a direct influence of SeqA on DNA replication is unlikely.

View larger version (13K): [in this window] [in a new window]   Fig. 1. In vitro replication of plasmid (pRLM4) in the presence of various amounts of the SeqA protein. SeqA was added to the reaction mixture simultaneously with other proteins () or incubated with DNA template for 15 min prior to addition of fraction II (). The amount of DNA synthesized in the absence of SeqA was taken as 100 % and corresponds to 75 pmol.

Bacteriophage replication in infected E. coli seqA mutant

View larger version (23K): [in this window] [in a new window]   Fig. 2. Fate of DNA of infecting phage in E. coli seqA+ and seqA hosts. (a) Diagram of experiments and possible results. Radioactively labelled parental phage DNA (dashed-line circles) starts its replication in bacteria growing in the heavy, non-radioactive medium. After the first round of replication, two molecules of DNA appear, each consisting of one radioactive light strand (dashed circles) and one non-radioactive heavy strand (solid circles). Next, several rounds of replication result in the formation of many (50–100 on average) DNA molecules, but only two of them can contain a radioactive strand, and thus radioactivity can be detected only at the heavy-light position after DNA isolation and ultracentrifugation on a CsCl density gradient. Since only a few of these 50–100 DNA molecules enter replication at late stages of phage development, there is a very low probability that any of these rolling-circle molecules would contain a radioactive strand, and as such they are not detected in these experiments [(a), upper panel]. The lower panel of (a) shows a situation in which replication starts early after infection or rolling-circle molecules are highly predominant in the population of DNA molecules. Such replication intermediates are likely to contain a light, radioactive parental DNA strand and a newly synthesized, heavy, non-radioactive tail, which leads to the transfer of the radioactivity toward the fully heavy position during density-gradient ultracentrifugation. (b) Experimental results. E. coli seqA+ and seqA strains growing in the minimal light medium (at 30 °C) were infected with phage cI857S7 (m.o.i.=10) labelled with [3H]thymidine (9x10–5 c.p.m. p.f.u.–1), and further incubation was perfomed in heavy medium (containing [13C]glucose and [15N]NH4Cl instead of glucose and NH4Cl, respectively). Samples were withdrawn at the indicated times, and total DNA was isolated and centrifuged in a CsCl density gradient. Five-drop fractions were collected from the bottom of the tube, and the radioactivity in each fraction was estimated in a scintillation counter. Arrows indicate the positions of fully heavy, heavy-light and fully light DNA (from left to right in each sub-panel). The results of the experiment with the seqA mutant at time 0 were analogous to those obtained for the wild-type host (i.e. only a single peak at the fully light position was detected; data not shown). The experiments were repeated three times with roughly the same distribution of the peaks.

Directionality of DNA replication in the seqA mutant
It has been proposed previously that initiation of the mode of bacteriophage DNA replication is preceded by one round of unidirectional replication (in contrast to bidirectional replication, which predominates at early stages of phage development) (Baraska et al., 2001). Therefore, to determine whether the SeqA-mediated delay in the to switch results from a SeqA-driven stimulation of bidirectional replication, we investigated the directionality of plasmid replication in wild-type cells and seqA mutants. One should note that plasmid copy number is decreased in seqA mutants (Somiska et al., 2001) and that plasmids form multimers. Nevertheless, we assumed that these facts did not influence the interpretation of the results significantly.

Analysis of the results of 2D-AGE experiments indicated that there were few differences in the directionality of plasmid replication between wild-type and seqA strains (Fig. 3a, b). Interestingly, the seqA mutation partially suppressed the effects of dnaA gene dysfunction on DNA replication directionality demonstrated previously (Baraska et al., 2001). In contrast to the dnaA46 single mutant, in which a high predominance of unidirectional replication has been reported (Baraska et al., 2001), in the double mutant dnaA46 seqA, both unidirectional and bidirectional replication intermediates could be clearly visualized (data not shown). The suppression was however partial, as in contrast to the wild-type host, in which roughly equal frequencies of unidirectional and bidirectional replication could be deduced (Fig. 3a), in the dnaA46 seqA cells the accumulation of simple-Y and late double-Y structures, as well as short bubble arcs, suggested a somewhat more frequent unidirectional (preferentially leftward) replication than a bidirectional one (data not shown).

View larger version (30K): [in this window] [in a new window]   Fig. 3. 2D-AGE of the plasmid (pCB104). The results obtained after isolation of plasmid DNA from cultures of E. coli seqA+ (a) and seqA (b) hosts bearing pCB104 and growing at 30 °C are presented in the left-hand panels. Plasmid DNA was digested with EcoRI before electrophoresis. Schemes of the electrophoretic profiles are shown in the corresponding right-hand panels. Theoretical schemes based on the electrophoretic profiles of replication according to the unidirectional leftward mode (Uni-L), the unidirectional rightward mode (Uni-R), the bidirectional mode (Bi) and a combination of molecules replicating according to all these modes (All) are shown below. The schemes were obtained using a computer model for the analysis of DNA replication intermediates by 2D-AGE (Viguera et al., 1998). Consistent data were obtained after digestion of pCB104 with EcoRV or SspI and PvuI (data not shown). (c) Schemes of pCB104 digested with particular enzymes, with ori marked as a bubble. The theoretical patterns of electrophoretic profiles in the case of bidirectional and unidirectional replication of pCB104 were predicted based on the assumption that replication forks initiate synchronously at ori and travel at the same rate in each direction.

Effect of seqA mutation on stability of the heritable replication complex
It has been demonstrated previously that after initiation of a new round of DNA replication, the once-formed replication complex is not disassembled but rather inherited by one of two daughter DNA copies (Wgrzyn et al., 1992, 1996a; Wgrzyn & Taylor, 1992). Interestingly, subsequently, a similar phenomenon has been proposed to occur in eukaryotic cells also (Duncker et al., 2002; Li & DePamphilis, 2002). In , the heritable replication complex consists of O and P proteins and at least two host-encoded proteins, DnaB and DnaK (Potrykus et al., 2002), and occurs in both bacteriophage -infected cells and those bearing plasmids (Wgrzyn et al., 1995a, 1996a). The existence of the heritable replication complex can be monitored by measurement of the stability of the O protein, which is rapidly degraded in a free form in E. coli cells but is protected from proteolysis by other components of the complex (Wgrzyn et al., 1992, 1995a, 1996b, 1998).

To monitor the stability of the O protein in plasmid-bearing wild-type and seqA strains, the cells were pulse-labelled with radioactive methionine, and following chasing with an excess of unlabelled methionine, samples of bacterial culture were withdrawn and analysed. After cell lysis, immunoprecipitation with anti-O serum, and SDS-PAGE, the radioactivity of protein bands was estimated. We found that a stable fraction of the O protein, corresponding to the heritable replication complex, existed in both wild-type and seqA mutant hosts. However, this fraction was less abundant in the mutant (Fig. 4), suggesting that SeqA is required for fully pronounced stabilization of the replication complex.

View larger version (13K): [in this window] [in a new window]   Fig. 4. Relative O protein levels in E. coli seqA+ () and seqA () hosts bearing plasmid (pKBlin). Bacterial cultures were pulse-labelled with [35S]methionine and chased with an excess of non-radioactive methionine at time 0. The y axis shows the O protein level as a percentage of that of the corresponding E. coli strain at time 0.

ori

Replication of plasmids in the seqA mutant during the relaxed response
During amino acid starvation, when new replication complexes cannot be formed due to inhibition of protein synthesis, replication of plasmids may occur solely due to the activity of the heritable replication complex (Wgrzyn & Taylor, 1992). However, transcriptional activation of ori (transcription initiated at the p_R promoter and passing through the origin region) is necessary for initiation of plasmid replication in vivo. Thus, synthesis of DNA is inhibited in amino acid-starved wild-type cells, due to guanosine tetraphosphate (ppGpp)-mediated impairment of p_R activity, but proceeds in relA (relaxed) mutants, unable to produce ppGpp in starved cells (Szalewska-Paasz et al., 1994). Because of this, plasmid replication in the amino acid-starved relA host reflects the efficiency of functioning of the heritable replication complex and the effectiveness of p_R-initiated transcriptional activation of ori.

We found that in relA seqA double-mutant cells growing exponentially in a minimal medium, the relative kinetics of replication of plasmid DNA was at a level similar to that observed in the relA seqA⁺ host [one should note a lower plasmid copy number in seqA cells, as demonstrated previously (Somiska et al., 2001) and confirmed in this report]. However, amino acid starvation caused an impairment in the plasmid DNA synthesis in the double (relA seqA) mutant relative to the relA seqA⁺ bacteria (Fig. 5). In contrast to the almost complete inhibition of plasmid replication in the wild-type host (caused by accumulation of ppGpp and inhibition of transcription from p_R), some increase in the amount of DNA occurred in the relA seqA host during the relaxed response, but the efficiency of this process was significantly lower than that in the relA seqA⁺ bacteria (Fig. 5).

View larger version (17K): [in this window] [in a new window]   Fig. 5. Replication of plasmid (pCB104) in amino acid-starved E. coli wild-type (relA+ seqA+) host (), and relA seqA+ () and relA seqA () mutants. The results of experiments with the relA+ seqA strain were analogous to those obtained for relA+ seqA+ (data not shown). Isoleucine starvation was induced by the addition of L-valine up to 1 mg ml–1 at the time indicated by the arrow. The y axis shows the amount of plasmid DNA relative to that of the corresponding E. coli strain at time 0.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Since SeqA regulates replication from oriC directly (by binding specifically to the origin region, and influencing initiator protein function and changes in nucleoprotein structures in this region), which can be demonstrated in vitro (Slater et al., 1995; Wold et al., 1998; Torheim & Skarstad, 1999), we asked whether the effect of this protein on DNA replication is also direct. However, in contrast to a stimulation of plasmid replication in vivo, deduced from the negative effects of the seqA mutation on plasmid copy number, a small negative effect of SeqA was observed in in vitro experiments (Fig. 1). Thus, in the absence of stimulation of DNA replication by SeqA in vitro, we conclude that this protein controls the replicon indirectly. One might argue that since SeqA interacts with cell membranes, which are missing in the in vitro replication assay, such an experimental system could be incomplete. Although we cannot exclude such a possibility, the fact that SeqA can clearly influence oriC-initiated replication in vitro (Wold et al., 1998; Torheim & Skarstad, 1999), together with our results (this report), makes the putative hypothesis of a direct involvement of this protein in DNA replication unlikely.

Despite the apparent lack of direct effects, there appears to be a significant role for SeqA in the regulation of DNA replication. This can be concluded from the results of our density-shift experiments, which indicated considerable differences in the distribution of replication intermediates between wild-type and seqA hosts infected by bacteriophage (Fig. 2). The movement of the parental radioactively labelled phage DNA toward the fully heavy position suggests that replication intermediates appear earlier and are more abundant in the mutant, implying a role for SeqA in the switch from circle-to-circle to rolling-circle replication during bacteriophage development. The lack of the signal at the fully light position in the experiment with the seqA mutant (60 min after infection), in comparison with the wild-type host, is also intriguing. In theory, this could arise from the stimulation of either or replication. However, since seqA inactivation has a negative effect on the replication of plasmids (Somiska et al., 2001; see also Fig. 5 in this report), which replicate exclusively according to the mode, these results may support the hypothesis that SeqA delays the rolling-circle replication of phage DNA.

Because it has been proposed previously that replication of DNA is preceded by one round of unidirectional replication (Baraska et al., 2001), the lack of a marked effect of the seqA mutation on the directionality of plasmid replication (Fig. 3) might appear surprising. Nevertheless, this suggests that other processes and/or factors can also influence the mechanisms of the switch from an early to a late mode of DNA replication during bacteriophage lytic development. One might be the stabilization of the replication complex, which has in fact been postulated previously (ylicz et al., 1998). The nucleoprotein complex formed at ori contains the O protein, which is protected from proteolysis by other components of this complex, and inherited by one of the two daughter copies of DNA after each round of replication (Wgrzyn et al., 1992, 1996a; Wgrzyn & Taylor, 1992; Potrykus et al., 2002). Since a free (unbound) form of the O gene product is rapidly degraded in E. coli cells (t ~1–2 min) (Wgrzyn et al., 1992), the efficiency of formation and persistence of the heritable replication complex may be deduced from the stability of the O protein in in vivo pulse–chase experiments. Interestingly, the stable fraction of the O protein was significantly less abundant in the seqA mutant relative to wild-type bacteria (Fig. 4), indicating an involvement of SeqA in either the formation of the stable replication complex or its survival in cells. The mechanism of this phenomenon remains to be elucidated; nevertheless, it is worth noting that SeqA binds to a region of the p_R promoter and regulates its activity (Somiska et al., 2001), and that a decrease in the abundance of the stable O protein fraction is also observed in cells bearing a derivative of plasmid (pTC2) in which p_R is replaced by another promoter (p_tet) (Herman-Antosiewicz et al., 1998a) [the activity of the p_tet promoter has been demonstrated to be SeqA independent (Somiska et al., 2001)]. Moreover, another regulator of p_R activity, the DnaA protein (Szalewska-Paasz et al., 1998; Glinkowska et al., 2003), has also been suggested to be a factor that influences the stability of the replication complex (Herman-Antosiewicz et al., 1998a), and an interplay between DnaA and SeqA at the p_R promoter region has been clearly demonstrated (Somiska et al., 2003b). In this light, seqA dysfunction-mediated suppression of the effects of certain dnaA alleles on the transformation of E. coli cells with plasmids, demonstrated previously (Glinkowska et al., 2001), and on the directionality of plasmid replication, suggested in this report, may be important to understand the regulation of DNA replication initiation. It is tempting to speculate that a nucleoprotein structure much larger than that suggested previously is formed at the DNA region encompassing p_R and ori. SeqA and DnaA proteins might be involved in the formation of such a structure, which would control replication initiation from ori by both facilitating stabilization of the heritable replication complex and regulating p_R-dependent transcriptional activation of the origin. A precedent for the formation of this kind of ‘super-complex’ has already been described; namely, the ParB protein of phage P1 is capable of forming large complexes, due to spreading along DNA, which affect promoters located downstream (Rodionov & Yarmolinsky, 2004). In this light, it is worth mentioning that the filamentous form of SeqA is able to induce negative DNA supercoiling (Odsbu et al., 2005). This might potentially affect both p_R activity and O protein stability.

The lack of drastic effects of SeqA on in vitro DNA replication (Fig. 1), together with the moderate effect of seqA dysfunction on plasmid and phage replication in vivo, suggests that this protein functions in the modulation of this process rather than being essential. However, such modulation can play an important regulatory role under different environmental and physiological conditions. Interestingly, bacteriophage DNA replication based on an unstable, rather than stable, replication complex has been reported in UV-irradiated bacteria (Wgrzyn & Wgrzyn, 2000), strengthening the hypothesis presented above.

If SeqA indeed modulates the efficiency of DNA replication by regulating the stability of the replication complex and the transcriptional activation of ori, one should observe the most significant effects of seqA mutations on replication, which is totally dependent on these processes. Such replication occurs as the sole mode of plasmid DNA synthesis during the relaxed response of bacterial cells to amino acid starvation (Taylor & Wgrzyn, 1995; Szalewska-Paasz et al., 1994). Under these conditions, new replication complexes cannot be formed due to the lack of amino acids, and the previously formed, heritable replication complex requires p_R-initiated transcription to initiate DNA synthesis. In accordance with the above hypothesis, we observed a significant impairment of plasmid replication in the amino acid-starved relA seqA double mutant relative to the relA seqA⁺ host (Fig. 5).

In conclusion, unlike the E. coli chromosome, the replication of which is negatively regulated by the sequestering of oriC due to binding of the SeqA protein to hemimethylated GATC motifs (Taghbalout et al., 2000; Guarne et al., 2005), replication of DNA is moderately stimulated by SeqA rather than inhibited. This stimulation is indirect, brought about by stabilization of the heritable replication complex and stimulation of the transcriptional activation of ori, but may nevertheless play an important regulatory role, especially under various stress conditions. This also indicates that SeqA may differentially control the replication of various replicons, employing a large spectrum of molecular mechanisms. Such a function supports the proposal that this protein is a global regulator of processes occurring in bacterial cells. In fact, types of processes that have been demonstrated to be SeqA regulated in studies of bacteriophage , such as transcription regulation and the stabilization of protein complexes, may represent more general phenomena occurring in the host (E. coli) cells rather than being restricted solely to its parasite (phage ).

	ACKNOWLEDGEMENTS

 
We are grateful to Kirsten Skarstad for providing the seqA10 mutant, SeqA protein and anti-SeqA antibodies, and to Roger McMacken, Johns Hopkins University, and Igor Konieczny and Krzysztof Liberek, University of Gdask, for discussions and for providing samples of proteins at early stages of this study. The assistance of Ewa Wójtowicz and Jacek Trzeszczyski in the construction of plasmid pEW1 and electron-microscopic studies, respectively, is acknowledged. This work was supported by the Polish Ministry of Science and Higher Education (project grant no. N301 122 31/3747).

Edited by: L. Jannière

	REFERENCES

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
Appleyard, R. K. (1954). Segregation of new lysogenic types during growth of a doubly lysogenic strain derived from Escherichia coli K12. Genetics 39, 440–452.[Free Full Text]

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]

Baraska, S., Gabig, M., Wgrzyn, A., Konopa, G., Herman-Antosiewicz, A., Hernandez, P., Schvartzman, J. B., Helinski, D. R. & Wgrzyn, G. (2001). Regulation of the switch from early to late bacteriophage DNA replication mode. Microbiology 147, 535–547.[Abstract/Free Full Text]

Boyd, A. C. & Sherratt, D. J. (1995). The pCLIP plasmids: versatile cloning vectors based on the bacteriophage origin of replication. Gene 153, 57–62.[CrossRef][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 U S A 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., Abeles, A. & Austin, S. (1995). A protein that binds to the P1 origin core and the oriC 13mer region in a methylation-specific fashion is the product of the host seqA gene. EMBO J 14, 4083–4090.[Medline]

Burkardt, H. & Lurz, R. (1984). Electron microscopy. In Advanced Molecular Genetics, pp. 281–313. Edited by A. Puhler & K. N. Timmis. Berlin, Heidelberg: Springer Verlag.

Duncker, B. P., Shimada, K., Tsai-Pflugfelder, M., Pasero, P. & Gasser, S. M. (2002). An N-terminal domain of Dbf4p mediates interaction with both origin recognition complex (ORC) and Rad53p and can deregulate late origin firing. Proc Natl Acad Sci U S A 99, 16087–16092.[Abstract/Free Full Text]

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]

Fujikawa, N., Kurumizaka, H., Yamazoe, M., Hiraga, S. & Yokoyama, S. (2003). Identification of functional domains of the Escherichia coli SeqA protein. Biochem Biophys Res Commun 300, 699–705.[CrossRef][Medline]

Fujikawa, N., Kurumizaka, H., Nureki, O., Tanaka, Y., Yamazoe, M., Hiraga, S. & Yokoyama, S. (2004). Structural and biochemical analyses of hemimethylated DNA binding by the SeqA protein. Nucleic Acids Res 32, 82–92.[Abstract/Free Full Text]

Fuller, R. S., Kaguni, J. M. & Kornberg, A. (1981). Enzymatic replication of the origin of the Escherichia coli chromosome. Proc Natl Acad Sci U S A 78, 7370–7374.[Abstract/Free Full Text]

Glinkowska, M., Konopa, G., Wgrzyn, A., Herman-Antosiewicz, A., Weigel, C., Seitz, H., Messer, W. & Wgrzyn, G. (2001). The double mechanism of incompatibility between plasmids and Escherichia coli dnaA(ts) host cells. Microbiology 147, 1923–1928.[Abstract/Free Full Text]

Glinkowska, M., Majka, J., Messer, W. & Wgrzyn, G. (2003). The mechanism of regulation of bacteriophage p_R promoter activity by Escherichia coli DnaA protein. J Biol Chem 278, 22250–22256.[Abstract/Free Full Text]

Goldberg, A. R. & Howe, M. (1969). New mutations in the S cistron of bacteriophage affecting host cell lysis. Virology 38, 200–202.[CrossRef][Medline]

Guarne, A., Zhao, Q., Ghirlando, R. & Yang, W. (2002). Insights into negative modulation of E. coli replication initiation from the structure of SeqA-hemimethylated DNA complex. Nat Struct Biol 9, 839–843.[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]

Han, J. S., Kang, S., Lee, H., Kim, H. K. & Hwang, D. S. (2003). Sequential binding of SeqA to paired hemi-methylated GATC sequences mediates formation of higher order complexes. J Biol Chem 278, 34983–34989.[Abstract/Free Full Text]

Han, J. S., Kang, S., Kim, S. H., Ko, M. J. & Hwang, D. S. (2004). Binding of SeqA protein to hemi-methylated GATC sequences enhances their interaction and aggregation properties. J Biol Chem 279, 30236–30243.[Abstract/Free Full Text]

Hardy, C. D. & Cozzarelli, N. R. (2005). A genetic selection for supercoiling mutants of Escherichia coli reveals proteins implicated in chromosome structure. Mol Microbiol 57, 1636–1652.[CrossRef][Medline]

Herman-Antosiewicz, A., Wgrzyn, A., Taylor, K. & Wgrzyn, G. (1998a). DnaA-mediated regulation of phage <