SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

CspD is a stationary phase-induced, stress response protein in the CspA family of Escherichia coli. Here, we demonstrate that overproduction of CspD is lethal, with the cells displaying a morphology typical of cells with impaired DNA replication. CspD consists mainly of β-strands, and the purified protein exists exclusively as a dimer and binds to single-stranded (ss)DNA and RNA in a dose-dependent manner without apparent sequence specificity. CsdD effectively inhibits both the initiation and the elongation steps of minichromosome replication in vitro. Electron microscopic studies revealed that CspD tightly packs ssDNA, resulting in structures distinctly different from those of SSB-coated DNA. We propose that CspD dimers, with two independent β-sheets interacting with ssDNA, function as a novel inhibitor of DNA replication and play a regulatory role in chromosomal replication in nutrient-depleted cells.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Expression of cspD is induced during stationary phase and upon nutrient starvation and, although regulated at the level of transcription, is independent of the stationary phase sigma factor, σs (Yamanaka and Inouye, 1997). The cspD gene, located at 19.9 min on the Escherichia coli chromosome immediately upstream of clpA, but transcribed in the opposite direction, encodes for the 74-amino-acid residue CspD protein (molecular mass of 7963 Da and calculated pI of 5.81; Yamanaka et al., 1998).

CspD is a member of the CspA family of proteins. Proteins homologous to CspA are found in a variety of prokaryotes; E. coli and Bacillus subtilis possess nine (CspA to CspI) and three CspA homologues respectively. However, some bacteria, such as Pyrococcus horikoshii and Mycoplasma pneumoniae, lack CspA homologues (Graumann and Marahiel, 1998; Yamanaka et al., 1998; Yamanaka, 1999). Furthermore, a domain called the cold shock domain (CSD), which is highly homologous to CspA (43% identity), is found in a group of eukaryotic proteins. These proteins, termed Y-box proteins, are involved in translational and transcriptional regulation. Among these, the human YB-1 protein and Xenopus FRGY2 protein have been well characterized (Matsumoto and Wolffe, 1998).

CspA, consisting of 70 residues, is the major cold shock protein produced upon temperature downshift (Yamanaka et al., 1998). It is composed of five antiparallel β-sheets, β1 to β5, which form a typical β-barrel structure (Newkirk et al., 1994; Schindelin et al., 1994; Feng et al., 1998). It exists as a monomer in solution (Newkirk et al., 1994; Schindelin et al., 1994) and binds co-operatively to RNA and single-stranded (ss)DNA (Jiang et al., 1997). As CspA binding to RNA makes the RNA more sensitive to ribonucleases, and CspA enhances translation in a cell-free translation system, it has been proposed that CspA may function as an RNA chaperone to facilitate translation at low temperatures (Brandi et al., 1996; Jiang et al., 1997).

Here, we have examined the properties of the CspD protein. We found that CspD exists exclusively as a dimer in solution and binds to RNA and ssDNA in a dose-dependent manner. Although the cspD gene was dispensable, CspD overproduction caused a change in cellular morphology, similar to that observed when DNA replication is impaired, and resulted in cell death. CspD effectively inhibited both the initiation and the elongation stages of minichromosomal DNA replication in vitro. Electron microscopic analysis showed that CspD tightly packed ssDNA. On the basis of these data, we propose that CspD dimers bind via two ssDNA-binding β-sheets to the single-stranded region of the replication fork and effectively block DNA replication.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

CspD null mutants are viable

Expression of cspD is specifically induced during the stationary phase or upon depletion of the growth medium carbon source, indicating that CspD plays a role in cells whose growth is impaired by nutrient limitation (Yamanaka and Inouye, 1997). However, the cspD gene is distinct from a number of the stationary phase-related genes, as its expression does not require σs, the stationary phase sigma factor. To elucidate the cellular role of CspD, a cspD null mutant was constructed (see Experimental procedures). The region containing the promoter and the coding sequence for cspD was replaced with the kanamycin-resistant gene (kan). The disrupted cspD mutation could be transduced efficiently into a wild-type strain in the absence of a plasmid harbouring cspD+. The cspD null mutant, KNJ92, grew well at 37°C in both rich (LB) and minimal (M9 supplemented with glucose, casamino acids and l-tryptophan) media without any discernible difference from the wild-type strain, MC4100; the doubling times for both wild-type and ΔcspD::kan strains were 27 and 48 min in LB medium and M9 medium respectively. The cspD null mutant also grew well over the temperature range of 15°C to 42°C without any detectable morphological changes (data not shown). Survival of the cspD null mutant after UV irradiation was the same as that for the wild-type strain (data not shown). Combined, these results indicate that the cspD gene is dispensable for cell growth.

It is well known that cells acquire tolerance to several challenges, such as thermal and oxidative tolerance, after entering stationary phase (Kolter et al., 1993). Several mutants of stationary phase-induced genes, such as rpoS itself (Lange and Hengge-Aronis, 1991; McCann et al., 1991) and rmf (Yamagishi et al., 1993), show sensitivity to those challenges and also lose viability after being maintained in the stationary phase for a long time. However, thermotolerance and oxidative tolerance of stationary phase cells were not affected by the cspD null mutation (data not shown). Moreover, the colony-forming ability of the cspD null mutant after a 7 day incubation at 37°C was similar to that of the wild-type strain (percentage viability: wild-type, 3.4% and 51% in LB and M9 media respectively; ΔcspD::kan, 6.2% and 73% in LB and M9 media respectively). Additionally, when 7-day-old cultures of both mutant and wild-type strain were diluted with fresh LB medium, they grew at the same rate after a lag period of 1.5–2 h. These results indicate that the lack of CspD does not cause loss of viability or stationary phase tolerance to environmental stresses. It was also found that rpoS expression in the stationary phase, as measured with a RpoS::LacZ translational fusion, was not affected by the cspD null mutation. The β-galactosidase activities of the wild-type and cspD null strains were very low when cells were in exponential phase (83 and 85 Miller units respectively), whereas they were induced during the stationary phase (2952 and 2629 Miller units respectively).

We also analysed the total protein content in the cspD null mutant by two-dimensional gel electrophoresis. The pattern of proteins from the cspD null mutant was almost identical to that of the wild-type strain in both the exponential (data not shown) and stationary phases (2 days old; Fig. 1). Note that CspD was clearly induced during the stationary phase in the wild-type cells, whereas it was undetectable in the mutant cells (compare Fig. 1A and B). These results suggest that CspD probably does not affect gene expression, including transcription, mRNA stability, translation or protein stability.

image

Figure 1. Two-dimensional analysis of total cellular proteins synthesized during stationary phase in wild-type (MC4100) and cspD null (KNJ92) cells. Cells of MC4100, the wild-type strain, and KNJ92, the cspD null mutant, were grown at 37°C for 2 days to stationary phase in M9 medium supplemented with glucose and 19 amino acids (no methionine). These cells were then labelled with [35S]-methionine for 1 h. Total-cell lysates were prepared and processed by two-dimensional gel (VanBogelen et al., 1990). The first dimension was carried out between the pH range of 3.5 (right) and 10 (left). The circle represents the CspD protein.

A. Wild-type MC4100 cells.

B. cspD null KNJ92 cells.

Download figure to PowerPoint

CspD overproduction blocks cell growth

Although the above results indicate that CspD is non-essential, the dispensability of a gene often results from the existence of another gene encoding for a redundant function. In such a case, functional analysis of the dispensable gene may be carried out by overproducing the gene product and examining for a distinct phenotype. To pursue this approach, the cspD open reading frame (ORF) was cloned under the control of the lpp-lac promoter in a pINIII vector system, yielding pKNJ8003, so that cspD expression is controlled by the inducer IPTG.

As shown in Fig. 2A, MC4100 cells harbouring pKNJ8003 were unable to form colonies on medium containing IPTG (0.1 mM), in contrast to those harbouring the vector pINIII-NdeI-Cm. Moreover, when IPTG (1 mM) was added to liquid cultures at mid-log phase, the cells harbouring the vector continued to grow with a doubling time (36 min) identical to that seen in the absence of IPTG (Fig. 2B and C). However, cells harbouring pKNJ8003 grew more slowly, even in the absence of IPTG (doubling time of 75 min), probably because of the leaky expression of cspD from the lpp-lac promoter (Fig. 2B and C). Importantly, the viability of cells with pKNJ8003 was reduced upon addition of IPTG (Fig. 2C); although the turbidity of the culture increased (Fig. 2B), by 8 h after the addition of IPTG, the cell viability dropped to < 1% (Fig. 2C).

image

Figure 2. Effects of CspD overproduction on cell growth. MC4100 cells were transformed with the vectors pINIII-NdeI-Cm or pKNJ8003 carrying the cspD gene.

A. Transformed cells were streaked on M9 plates containing glucose (0.4%), casamino acids (0.2%), l-tryptophan (50 µg ml−1), chloramphenicol (30 µg ml−1) and IPTG (0 or 0.1 mM) and incubated at 37°C for 24 h.

B and C. Transformed cells grown in M9 liquid medium containing glucose (0.4%), casamino acids (0.2%), l-tryptophan (50 µg ml−1) and chloramphenicol (30 µg ml−1) at 37°C overnight were diluted 100 times with the same medium. The optical density (OD600) of the culture (B) and viable cells (C) were measured every 2 h. At an OD600 of ≈ 0.2, indicated by arrows, cultures were divided into two. IPTG was added to a final concentration of 1 mM to one of the two portions. (○), MC4100 harbouring pINIII-NdeI-Cm without IPTG; (●), MC4100 harbouring pINIII-NdeI-Cm with IPTG; (Δ), MC4100 harbouring pKNJ8003 without IPTG; and (▴), MC4100 harbouring pKNJ8003 with IPTG.

D. MC4100 cells harbouring pKNJ8003 were harvested at the 12 h time point of (B) (8 h after the addition of IPTG) and stained with DAPI as described previously (Hiraga et al., 1989).

Download figure to PowerPoint

These results suggest that CspD overproduction does not affect transcription and translation, because protein production, as reflected by optical density, continued to increase in the presence of IPTG. Therefore, possible targets of CspD overproduction may be DNA replication, chromosome segregation or cell division. To address these possibilities, cells transformed with pKNJ8003 were grown for 8 h in the presence of IPTG and stained with DAPI according to the fluo-phase combined method (Hiraga et al., 1989). Cells overproducing CspD were elongated and contained a condensed chromosome with one or two unit size at the centre of the cell (Fig. 2D), a phenotype typically observed with a number of dna conditional mutants under non-permissive conditions (Hirota et al., 1968), suggesting that CspD may inhibit DNA replication.

CspD forms a stable homodimer

In order to elucidate the function of CspD further, the protein was purified, and its biochemical properties were characterized. CspD is distantly related to CspA of the E. coli CspA family, with only 32 identical residues along its 74-residue sequence (Fig. 3A). Nevertheless, the sequences in the putative β1, β2 and β3 strands of CspD retain great similarities to those of CspA and, importantly, all five hydrophobic amino acid residues forming a hydrophobic core (shown by dots in Fig. 3A) are conserved in both proteins. Indeed, highly purified CspD (Fig. 3B, lane 2) has a circular dichroism (CD) spectrum typical of a protein consisting mainly of β-strands (Fig. 3C), indicating that CspD is also composed of a β-barrel structure similar to that of E. coli CspA (Newkirk et al., 1994; Schindelin et al., 1994; Feng et al., 1998) and B. subtilis CspB (Schindelin et al., 1993; Schnuchel et al., 1993).

image

Figure 3. Purification of CspD.

A. Amino acid alignment of CspD with E. coli CspA and B. subtilis CspB. A gap is indicated by a dash. Identical and similar amino acid residues are indicated by vertical lines and dots respectively. Five β-strands (β1 to β5) are indicated by lines on the basis of previous studies (Schnuchel et al., 1993; Feng et al., 1998). RNP1 (KGFGFI) and RNP2 (VFVHF) motifs are boxed. Residues forming the highly conserved hydrophobic core are indicated by dots on top.

B. The purification procedures were as described in Experimental procedures. Purified CspD (lane 2) was analysed by 17.5% SDS–PAGE and stained with Coomassie brilliant blue R-250. Bovine serum albumin (66.2 kDa), trypsin inhibitor (21.5 kDa) and RNase A (13.7 kDa) were used as molecular weight standards (lane 1).

C. CD spectrum of CspD in 10 mM potassium phosphate, pH 7.0, containing 50 mM NaCl at 4°C.

D. Purified CspD was applied to a Superdex 75 HR 10/30 (10 mm × 300 mm; Pharmacia), eluted with 10 mM potassium phosphate buffer, pH 7.0, containing 50 mM NaCl and detected by UV absorbance. Bovine serum albumin (66.2 kDa), trypsin inhibitor (21.5 kDa), myoglobin (17.0 kDa), RNase A (13.7 kDa) and E. coli CspA (7.4 kDa) were used as standard molecular size markers. Their peak positions of elution are indicated by arrows.

Download figure to PowerPoint

Interestingly, in contrast to CspA, purified CspD (monomeric size of 8.0 kDa) is found exclusively as a dimer under a physiological condition with an apparent molecular size of 15.0 kDa (Fig. 3D). It is worth noting that CspD is unique among the E. coli CspA family in having an amino-terminal end that is shorter by three residues (see Fig. 3A). The extended N-terminal end of CspA has been shown to interact partially with the β4 strand, an association that probably prevents intermolecular antiparallel β-interaction between the β4 strands of two CspA molecules (Newkirk et al., 1994). Such an antiparallel β4–β4 interaction between two molecules, leading to the formation of a stable homodimer, has been observed for B. subtilis CspB (Schindelin et al., 1993), which, like CspD, has an amino-terminal structure that is shorter than the amino-terminus of CspA by three residues (see Fig. 3A).

CspD binds ssDNA and RNA, but not dsDNA

Gel shift assays were carried out to examine whether CspD binds to DNA or RNA. When a double-stranded (ds) fragment (141 bp) of DNA was used as the probe, migration of the fragment was not retarded (Fig. 4A). In contrast, when heat-denatured DNA was used, clear bandshifting was observed (Fig. 4B), indicating that CspD binds ssDNA, but not dsDNA. With increasing amounts of CspD, ssDNA migration decreased further, suggesting that CspD binds to ssDNA in a dose-dependent manner. It is important to note that, unlike CspA, CspD does not bind ssDNA in a co-operative manner. The dose-dependent binding of ssDNA was examined further with an 82 base ssDNA (Fig. 4C). With less than 3 ng of CspD (25 nM), no discernible retarded band was observed (Fig. 4C, lanes 2 and 3). With 10 ng of CspD, a small amount of the ssDNA was retarded (Fig. 4C, lane 4). With 30 ng (Fig. 4C, lane 5), ≈ 10% of the ssDNA migrated to the same position, and the remaining majority of the ssDNA migrated slightly more slowly than free ssDNA (cf. Fig. 4C, lanes 4 and 5). As the concentration of CspD increased further, 0.1 µg (Fig. 4C, lane 6), 0.3 µg (Fig. 4C, lane 7), 1 µg (Fig. 4C, lane 8) and 3 µg (Fig. 4C, lane 9), the migration of the ssDNA was retarded stepwise as a single band. Whether the decrease in migration with increased CspD concentrations is caused by multiple binding sites or oligomerization of CspD cannot be deduced from this experiment.

image

Figure 4. Binding assays of CspD to dsDNA, ssDNA and RNA. The binding assay was carried out in a reaction (15 µl) containing 10 mM Tris-HCl, pH 8.0, 1 mM EDTA, 50 mM KCl, 7.4% (w/v) glycerol, DNA or RNA probe and purified CspD for 20 min on ice.

A and B. Two fentomol of 141 bp dsDNA (A) and 4 fmol of 141 base heat-denatured DNA (B) were used as probes. Lane 1, no CspD; lanes 2–4, 0.3, 1 and 3 µg of CspD respectively.

C and D. Twenty fentomol of 87 base heat-denatured DNA (C) and 10 fmol of 47 base heat-denatured DNA (D) were used as probes. Lane 1, no CspD; lanes 2–9, 0.001, 0.003, 0.01, 0.03, 0.1, 0.3, 1 and 3 µg of CspD respectively.

E. Two 26 base oligonucleotides (5′-ACGAAGGTCAATCCGTTCAGTTTGAT-3′ for lanes 1 and 2, and 5′-GGCCCCTGTTCGATGGTGAAGGACAC-3′ for lanes 3 and 4) were used (22.5 fmol/assay) as probes. Lanes 1 and 3, no CspD; lanes 2 and 4, 3 µg of CspD.

F. Ten fentomol of 125 base heat-denatured DNA was used as a probe. Lane 1, no CspD; lanes 2–7, 3 µg of CspD, with KCl at concentrations of 0, 50, 100, 200, 600 and 1000 mM respectively.

G. Fifty fentomol of RNA was used as a probe. Lanes 1–3, 0.3, 1 and 3 µg of CspD; lane 4, no CspD.

Download figure to PowerPoint

Both E. coli CspA (Jiang et al., 1997) and B. subtilis CspB (Graumann and Marahiel, 1994) also bind ssDNA and RNA, but not dsDNA. Bacillus subtilis CspB binds ssDNA with a preferential sequence (Graumann and Marahiel, 1994), whereas E. coli CspA binds co-operatively to ssDNA and RNA without apparent sequence specificity (Jiang et al., 1997). For the co-operative binding of E. coli CspA to RNA, the length of RNA has to be longer than 72 bases (Jiang et al., 1997). However, CspD can bind to ssDNA 82 bases (Fig. 4C), 47 bases (Fig. 4D) and even 24 bases in length (Fig. 4E). Furthermore, CspD can bind to two totally different sequences (Fig. 4E), suggesting that CspD binds to ssDNA without sequence specificity. Note that CspD binding to ssDNA was observed even in the presence of 1 M KCl, suggesting that the interaction between CspD and ssDNA is governed by hydrophobic rather than ionic forces (Fig. 4F).

In a gel shift assay, CspD also bound RNA in a dose-dependent manner, similar to its binding of ssDNA (Fig. 4G). All the RNA probes, which possess randomized sequences, were shifted simultaneously, albeit in a broader range, indicating that CspD binds RNA without sequence specificity. Taken together, it is concluded that CspD binds through hydrophobic interaction to ssDNA and RNA in a dose-dependent manner without apparent sequence specificity.

CspD inhibits DNA replication in vitro

Based on the above results, the effect of CspD on chromosomal replication was examined using a cell-free DNA synthesis system reconstituted with purified components (Crooke, 1995). The process of bacterial chromosomal replication can be divided into several stages: initiation, priming of chain starts, chain elongation and termination. Initiation occurs at the single origin (oriC) on the E. coli chromosome. Multiple copies of DnaA, the initiator protein, bind to four 9-mer DnaA boxes within the oriC sequence. In the presence of HU or IHF protein and 5 mM ATP, DnaA distorts the DNA, causing the neighbouring AT-rich duplex DNA region to separate, thus creating the ‘open complex’. DnaA then directs the DnaC-mediated delivery of DnaB helicase into the open complex at the site of the future bidirectional replication forks, forming the ‘prepriming complex’. With the addition of SSB, primase, gyrase, DNA polymerase III holoenzyme and deoxynucleotides (dNTPs), the priming and chain elongation steps of DNA synthesis commence.

CspD efficiently inhibited oriC-plasmid replication, either before or after prepriming complex formation (Fig. 5A). DNA synthesis decreased ≈ 60% in the presence of 1.7 pmol of CspD dimer (25 ng; 66 nM) and was completely inhibited with CspD dimer at levels of 6.7 pmol (100 ng; 267 nM) and higher. The indistinguishable influence of CspD when added before or after prepriming complex formation suggests that it affects a later stage of minichromosome replication. In agreement, CspD was equally efficient at inhibiting replication when added before or after an even earlier stage of initiation, open complex formation (data not shown).

image

Figure 5. Effect of CspD on in vitro DNA replication. The in vitro replication reaction (see Experimental procedures) was staged at prepriming complex formation (A), small bubble complex formation (B) or the completion of initiation, but before elongation (C) and (D) by the omission of certain components during the first incubation.

A. Effect of CspD on prepriming complex formation. The first incubation (38°C, 10 min) was of a mixture (17 µl) that contained 30 mM HEPES.KOH (pH 7.6), 9 mM magnesium acetate, 2 mM ATP, 200 ng of oriC plasmid, 10 ng of HU, 127 ng of DnaB–C complex, 130 ng of DnaA (○) or the indicated amount of CspD (●). DnaA (130 ng) was then added to the reactions that already contained CspD, the indicated amounts of CspD were added to the reactions that already contained DnaA, and all reactions were incubated further (38°C, 10 min). DNA synthesis from the prepriming complexes proceeded with the addition of the omitted components (7 µl containing: 450 ng of SSB, 14 ng of primase, 400 ng of gyrase A, 180 ng of gyrase B, 26 ng of β-subunit, 120 ng of DNA polymerase III*, 100 µM dNTPs), and the reactions were incubated further (30°C, 20 min). Before (●) indicates the addition of CspD before prepriming complex formation, and After (○) indicates the addition of CspD after prepriming complex formation.

B. Effect of CspD on small bubble complex formation. The reactions were staged as for prepriming complex formation (A), except that SSB (450 ng) was included in the first incubation. Before (●) indicates the addition of CspD before small bubble complex formation, and After (○) indicates the addition of CspD after small bubble complex formation.

C. Effect of temperature on initiation of oriC-plasmid in vitro replication. Initiation of replication, without chain elongation, was carried out by incubating (5 min) all the components except dNTPs at 38°C (●) or 20°C (○). The reactions were cooled by incubating for 2 min at 20°C, and DNA synthesis was allowed to proceed by the addition of dNTPs and the reactions being incubated (20°C) for 20, 50 or 70 min.

D. The effect of CspD on chain elongation. Initiation of replication, without chain elongation, was carried out by incubating all the components except dNTPs for 5 min at 38°C. Further initiations were prevented by cooling the reaction to 20°C. The indicated amounts of CspD were added, and DNA synthesis in the absence of further initiation was allowed to proceed (70 min, 20°C) by the addition of dNTPs.

Download figure to PowerPoint

DnaB helicase, as a component of the prepriming complex, is able to unwind the DNA further and, when SSB is present, generate a ‘small bubble’ of ssDNA. When CspD was added before the formation of small bubble complexes, it had an inhibitory effect similar to that seen with prepriming complexes. However, when CspD was added to already formed small bubble complexes, its inhibitory influence was reduced, but still significant (Fig. 5B). In the former reaction, CspD probably inhibited an aspect of initiation as well as subsequent chain elongation, whereas in the latter reaction, CspD was unable to affect initiation and only inhibited chain elongation. When affecting initiation, CspD may be binding to the single-stranded region generated by helicase during initiation and, thus, blocking proper assembly of replisomes at the replication forks. In agreement, CspD is able to bind ssDNA of 24 bases (Fig. 4E) and perhaps shorter.

For efficient in vitro DNA replication from oriC, certain events during initiation require the reaction to be incubated at 38°C. However, once replisomes have been successfully assembled at the replication forks and primers have been synthesized, chain elongation can proceed at temperatures as low as 16°C (Sekimizu et al., 1988; Messer and Weigel, 1996). This difference permits the elongation stage of DNA replication to be examined separately from initiation: if all the components required to reconstitute DNA replication, except the dNTPs, are mixed on ice and incubated at 38°C for 5 min, the initiation events can occur, and the system is poised for chain elongation. Upon shifting the temperature down to 20°C and with the addition of the dNTPs, chain elongation can commence. The temperature of 20°C blocks further initiations. The inhibitory influence of such a lower temperature on initiation was evident when the initial 5 min incubation was carried out at 20°C versus 38°C (Fig. 5C), with the effect becoming more apparent as the time for chain elongation was increased from 20 min to 70 min.

CspD was added just before the addition of the dNTPs to examine its effect on chain elongation (Fig. 5D). Over the same concentration range that impeded small bubble formation (Fig. 5B), CspD also blocked chain elongation (Fig. 5D). Taken together, CspD appears to inhibit DNA replication during both the initiation and the elongation stages, most probably by binding to the opened, single-stranded regions at replication forks.

It is interesting to compare how two different single-stranded, DNA-binding proteins, SSB and CspD, differ in their function in DNA replication. SSB is known to be required for DNA replication and, in the reconstituted in vitro DNA replication system, the highest level of DNA synthesis was obtained when 450 ng (1.5 pmol as a tetramer; 25 µl reaction volume) of SSB was present (data not shown). In contrast, at the same concentration (1.5 pmol as a dimer; 25 ng), CspD strongly inhibited DNA replication (Fig. 5A). Moreover, CspD was unable to replace SSB in the in vitro DNA synthesis reaction (data not shown).

CspD and SSB form distinctly different structures when binding ssDNA

Although both SSB and CspD are ssDNA-binding proteins, SSB is essential for DNA replication, whereas CspD acts as an inhibitor, suggesting that they bind to ssDNA in different ways. To address this question, their associations with ssDNA were examined by electron microscopy. When linearized, heat-denatured φX174 DNA was incubated with SSB, a smooth-contoured DNA–protein complex was observed (Fig. 6A). From measurements of the length of the SSB–ssDNA complexes, SSB appears to bind along the entire length of the ssDNA. SSB is known to bind co-operatively to ssDNA and to maintain it in a single-stranded state (Meyer and Laine, 1990). In contrast, when ssDNA was incubated with CspD, CspD locally packed the ssDNA, forming several unlinked dot-like structures (Fig. 6B). Note that ssDNA alone and a dimer of CspD by itself are undetectable by electron microscopy under the conditions used (data not shown).

image

Figure 6. Visualization of CspD–ssDNA complexes by electron microscopy. Binding reaction was carried out in a reaction mixture (15 µl) containing 20 mM HEPES-AcOH, pH 7.5, 50 mM NaCl, 0.6 µg of DNA (0.34 pmol of ssDNA) and 5 µg of SSB (66 pmol as tetramers) or 25 µg of CspD (1563 pmol as dimers) for 15 min on ice. PstI-linearized, heat-denatured φX174 DNA (5386 bases) was used as a ssDNA for substrate. The protein–ssDNA complexes were processed and visualized as described in Experimental procedures. ssDNA bound by SSB (A) or (B) CspD.

C. ssDNA incubated first with SSB, followed by incubation with CspD.

D. ssDNA incubated first with CspD followed by incubation with SSB. Some of the dot-like structures are indicated by arrows. The bar represents 0.3 µm.

Download figure to PowerPoint

Additional studies examined whether CspD is able to bind to SSB-coated ssDNA, or if SSB can bind to ssDNA already complexed with CspD. For the former possibility, the addition of CspD to SSB-covered ssDNA did not alter the structure of the SSB–ssDNA complex (Fig. 6C), indicating that, once SSB binds to ssDNA, CspD is no longer able to interact with the DNA. On the other hand, in the latter experiment, both proteins were able to bind to the ssDNA simultaneously, as both dot-like structures and smooth-contoured regions can be seen on an individual ssDNA molecule (Fig. 6D). In these experiments, CspD was added in large excess over the amount of the ssDNA molecules (see Experimental procedures), indicating that CspD does not bind co-operatively to ssDNA and binds in patches, allowing SSB to bind to regions between the patches. To confirm this, after ssDNA had been incubated with CspD, the CspD–ssDNA complexes were fixed with glutaraldehyde, fractionated and then incubated with SSB. The resulting complexes were quite similar to the CspD/SSB–ssDNA complexes (data not shown).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

CspD, a member of the large CspA family of E. coli, is a novel inhibitor of DNA replication based on the following observations: (i) the disruption of cspD has no effect on cell growth and morphology; (ii) the overproduction of CspD causes a cellular morphological change similar to those caused by mutations in genes encoding for essential replication proteins (Fig. 2); (iii) CspD binds to ssDNA without apparent sequence specificity (Fig. 4), resulting in a packed structure (Fig. 6); (iv) CspD inhibits DNA synthesis in an in vitro DNA replication assay at both the initiation and the elongation stages (Fig. 5).

Replication of the E. coli chromosome is initiated at a specific locus, oriC, and replication forks proceed bidirectionally (Kornberg and Baker, 1992; Messer and Weigel, 1996). Initiation of chromosomal replication is a critical step in the control of the cell cycle in prokaryotes as well as in eukaryotes. The reconstitution of DNA replication with purified proteins facilitated the identification and characterization of many E. coli replication factors (Kaguni and Kornberg, 1984; Funnell et al., 1986). Among them is the initiator protein, DnaA, which plays a key role in the regulation of replication (Kornberg and Baker, 1992; Messer and Weigel, 1996). DnaA protein binds to its recognition sites, DnaA boxes, within oriC. At the proper time in the cell cycle a neighbouring AT-rich region is opened (Bramhill and Kornberg, 1988), and chromosomal replication is initiated with the assembly of replisomes at the replication forks.

A number of mechanisms have been described for the restraint of E. coli chromosomal replication. A factor termed IciA (Inhibitor of chromosomal initiation) was seen to inhibit the initiation of replication in vitro by binding to the AT-rich region of oriC and preventing the strand opening promoted by DnaA; if strand separation had already occurred, IciA had no effect on subsequent replication steps (Hwang and Kornberg, 1990). Yet, both deletion of iciA and overproduction of IciA had no effects on growth, with the iciA null mutant and IciA-overproducing cells having the same growth rate as wild-type cells (Thöny et al., 1991). Another mechanism involves the datA locus, which has been reported to regulate the timing of DNA replication in the cell cycle by titrating DnaA protein (Kitagawa et al., 1996; 1998). A third negative-acting factor for DNA replication is SeqA protein, which binds to hemimethylated GATC sequences to prevent reinitiation of recently replicated origins (Lu et al., 1994; Slater et al., 1995) and to fully methylated GATC sequences possibly to help regulate the timing of primary initiation events (Boye et al., 1996). A recently reported mechanism for the negative regulation of chromosomal replication, termed RIDA (regulatory inactivation of DnaA) involves the β-subunit of DNA polymerase III holoenzyme and a novel factor IdaB, which function to convert the replicatively active ATP form of DnaA into inactive ADP-DnaA (Katayama et al., 1998).

CspD is unique compared with these previously identified inhibitors of DNA replication in that: (i) it inhibits DNA synthesis at both the initiation and the elongation steps of replication, whereas the other inhibitors act only upon initiation; (ii) it binds to ssDNA without apparent sequence specificity, whereas IciA binds to dsDNA with sequence specificity (Hwang and Kornberg, 1990), and SeqA binds to fully and hemimethylated GATC sequences (Lu et al., 1994; Slater et al., 1995); (iii) it is a protein factor, whereas datA is a DNA element that titrates DnaA protein (Kitagawa et al., 1998); and (iv) it is not a required DNA replication factor, whereas RIDA includes the β-subunit of DNA polymerase III (Katayama et al., 1998).

CspD is induced during the stationary phase and upon carbon starvation (Yamanaka and Inouye, 1997). Several stationary phase-inducible proteins are known to possess activities to lower cellular metabolism. For example, a nucleoid protein, Dps, which exists at 150 000–200 000 molecules cell−1 at stationary phase (Almirón et al., 1992; Martinez and Kolter, 1997; Talukder et al., 1999), binds to DNA without apparent sequence specificity (Martinez and Kolter, 1997; Talukder and Ishihama, 1999) and packs the chromosome tightly (Grant et al., 1998). Another example is a ribosome modulation factor, Rmf, which is responsible for the formation of a translationally inactive 100S dimer from two 70S ribosomes during the stationary phase (Yamagishi et al., 1993; Wada et al., 1995). It is apparent that various cellular activities are co-ordinately inhibited during stationary phase by different factors, with CspD being a factor that blocks DNA synthesis.

E. coli possesses nine and B. subtilis three CspA-like proteins that might have overlapping functions (Graumann et al., 1996; Yamanaka et al., 1998). It is possible that they may complement each other functionally within each species. In E. coli, depletion of CspA did not affect cell growth, even at low temperature (Bae et al., 1997). Similarly, in B. subtilis, depletion of CspB did not affect cell growth, although the cells became more sensitive to freezing (Willimsky et al., 1992). However, in E. coli cspA null cells, as well as in B. subtilis cspB null cells, other CspA homologues (CspB, CspE and CspG for E. coli; CspC and CspD for B. subtilis) became overproduced upon cold shock, probably to compensate for the respective cspA and cspB defects (Graumann et al., 1996; Bae et al., 1997). Furthermore, it has been reported that at least one csp gene is required for the viability of B. subtilis (Graumann et al., 1997). Although the deletion of either cspD or iciA has little effect on cell growth, and the mechanisms by which CspD and IciA inhibit DNA replication in vitro are quite different, it is possible that simultaneous disruption of iciA and cspD may have a pronounced effect on cellular physiology.

CspA of E. coli and CspB of B. subtilis form a typical β-barrel structure, which has an RNA or ssDNA interacting β-sheet consisting of β1, β2 and β3 strands (Schindelin et al., 1993; 1994; Schnuchel et al., 1993; Newkirk et al., 1994; Feng et al., 1998). As with all known CspA homologues, CspD is highly conserved in the critical residues required for the function of the hydrophobic core of the β-barrel structure (Yamanaka et al., 1998). Thus, CspD is likely also to interact with RNA and ssDNA via the β-sheet consisting of the β1, β2 and β3 strands. The binding of ssDNA and RNA is a feature CspD shares in common with E. coli CspA, CspB, CspC and CspE (Jiang et al., 1997; Phadtare and Inouye, 1999) and B. subtilis CspB, CspC and CspD (Graumann and Marahiel, 1994; Graumann et al., 1997). Moreover, the CSD of Xenopus FRGY2 can also bind to RNA (Bouvet et al., 1995). The present results support the notion that the ability to bind ssDNA and RNA is a conserved function of CSDs throughout evolution (Graumann and Marahiel, 1998; Yamanaka et al., 1998). Thus, the possibility exists in eukaryotes for a CSD-containing protein that functions in the regulation of DNA replication.

CspD exists as a dimer (Fig. 3), binds to ssDNA and RNA in a dose-dependent manner without apparent sequence specificity (Fig. 4) and functions as an inhibitor of DNA replication (Fig. 5). In contrast, CspA exists as a monomer (Newkirk et al., 1994; Schindelin et al., 1994), binds to ssDNA and RNA co-operatively without apparent sequence specificity and functions as an RNA chaperone to facilitate translation at low temperature (Jiang et al., 1997). Although CspA is produced at levels of 800 000–1 000 000 molecules cell−1 upon cold shock (Jiang et al., 1997; Yamanaka and Inouye, 2001), the amount of CspD induced during the stationary phase is low, estimated to be less than 2000 dimers cell−1 (data not shown). Furthermore, although CspD overproduction was lethal (Fig. 2), CspA overproduction did not affect cell growth (data not shown). Thus, even though CspD and CspA are homologous and evolutionarily related, their biological functions and biochemical properties are quite different. An analysis of chimeric proteins of E. coli CspA and CspD is likely to provide insight into what differentiates the functions and roles of individual CspA homologues.

Experimental procedures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Bacterial strains, plasmids and media

Strains JM83 and BL21(DE3) were used for DNA cloning and protein production respectively. Plasmids pUC9 (Vieira and Messing, 1982), pHSG575 (Takeshita et al., 1987) and pET11a (Studier et al., 1990) were used. LB medium and M9 medium supplemented with 0.4% glucose, 0.4% casamino acids and l-tryptophan (25 µg ml−1) were used for bacterial growth. When necessary, antibiotics were added to the following final concentrations: ampicillin, 50 µg ml−1; kanamycin, 30 µg ml−1; and chloramphenicol, 30 µg ml−1.

Construction of the cspD deletion strain

The 5.9 kb EcoRI fragment containing the cspD gene was isolated from λ #213 of the Kohara library (Kohara et al., 1987) and cloned into pHSG575 (Takeshita et al., 1987), yielding pKNJ8001. The 1.4 kb fragment encompassing the region upstream of cspD was amplified by polymerase chain reaction (PCR), with pKNJ8001 as a template, with the following primers: #7811 (5′-GGGCGAGCCGTCGACCATTTGAC-3′) (the sequence underlined represents a SalI site added to the PCR fragment) and #7812 (5′-GTGTCCTGCTGCAGCTGCTTG-3′) (the sequence underlined represents a PstI site added to the PCR fragment). The 1.6 kb fragment encompassing the region downstream of cspD was also amplified in the same way with the following primers: #7809 (5′-CGGTGGTCAGGGAATTCTTG-3′) (the sequence underlined represents an EcoRI site added to the PCR fragment) and #7810 (5′-TTGTGCCCGTCGACGTAGAAGC-3′) (the sequence underlined represents a SalI site added to the PCR fragment). Both PCR fragments were digested with appropriate restriction enzymes and cloned into pUC9 digested with EcoRI and PstI, yielding pKNJ8005. The kanamycin resistance gene (kan) from pUC7Km(P) (Zhang et al., 1996) was inserted at the SalI site of pKNJ8005, yielding pKNJ8006. The linear pKNJ8006 DNA fragment containing ΔcspD::kan was introduced into the chromosome of the recD mutant FS1576 (Stahl et al., 1986). The ΔcspD::kan mutation was then transduced with phage P1vir into the wild-type strain MC4100. Kanamycin-resistant transductants carrying the cspD null mutation on the E. coli chromosome were confirmed by Southern hybridization (data not shown). One of them, KNJ92, was used in this study.

Assay for β-galactosidase activity

The cspD null mutation was transduced into strain SG22500 (ΔlacλGN272 containing an RpoS::LacZ translational fusion; Sledjeski et al., 1996), yielding KNJ81. SG22500 was a kind gift from Dr S. Gottesman (NIH). Cells were grown in LB medium at 30°C to different growth phases, and β-galactosidase activity was measured as described by Miller (1992).

Protein labelling experiment

Cells of MC4100, the parental strain, and KNJ92, the cspD null mutant, were grown in M9 medium supplemented with glucose and 19 amino acids (no methionine) at 37°C. Cells were labelled with trans-[35S]-methionine (1092 Ci mmol−1; Amersham) for 5 min at the mid-exponential phase of growth (OD600 of ≈ 0.5) or for 30 min when in stationary phase. Cell lysates were prepared and processed by two-dimensional gel electrophoresis as described previously (VanBogelen et al., 1990).

Microscopic observation of cells and nucleoids

Cells were fixed on a glass slide, stained with DAPI and observed under the microscope according to the fluo-phase combined method (Hiraga et al., 1989).

Production and purification of CspD

The cspD gene was amplified by PCR with the primer #4558, 5′-ggaattccatATGGAAAAGGGTACTG-3′, complementary to nucleotides −688 to −673 of the sequence reported by Gottesman et al. (1990), and the primer #4552, 5′-cgGGATCCAGTAGATGCTCTG-3′, nucleotides −998 to −980, where 5′ tails are shown in lower case and NdeI and BamHI sites are underlined. The PCR fragment was digested with NdeI and BamHI and cloned into the pET11a vector digested with NdeI and BamHI, yielding pET11-cspD. The DNA sequence was confirmed by DNA sequencing using Sequenase version 2.0 (USB).

E. coli strain BL21(DE3) harbouring pET11-cspD was grown in 3 l of M9-casamino acid medium supplemented with ampicillin to early log phase, and the production of CspD was induced for 2 h in the presence of 1 mM IPTG. The cells were harvested by centrifugation (5000 g) for 20 min at 4°C and suspended in 20 ml of 10 mM Bis-Tris buffer, pH 7.0. The cells were broken by three passages through a French press at 14 000 p.s.i. Unlysed cells and debris were removed from the lysate by centrifugation (8000 g for 15 min at 4°C). The supernatant fraction was subjected to high-speed centrifugation at 100 000 g for 1.5 h at 4°C to remove membranes and insoluble material. Ammonium sulphate was then added to the soluble fraction to a final concentration of 65% (w/v). The suspension was kept on ice, with stirring, for 2 h and then centrifuged (10 000 g for 20 min at 4°C). The pellet was dissolved in 20 ml of 20 mM Tris-HCl buffer, pH 8.0, dialysed twice against 2 l of the same buffer, and then applied onto a Q-Sepharose (Pharmacia) anion exchange column (2.5 cm diameter × 14 cm length) equilibrated with the same buffer. Proteins were eluted in the same buffer with a gradient of 0–1.0 M sodium chloride, and fractions were analysed by 17.5% SDS–PAGE. The fractions containing CspD were pooled, dialysed twice against 2 l of 10 mM potassium phosphate buffer, pH 7.0, and applied to a hydroxyapatite column (Bio-Rad; 2.5 cm diameter × 4 cm length) equilibrated with the same buffer. Proteins were eluted with a gradient of 10–500 mM potassium phosphate buffer, pH 7.0, and analysed by 17.5% SDS–PAGE. The fractions were pooled, concentrated with a Centricon 3 (Amicon) and applied onto an S-100 (Pharmacia) gel filtration column (1.5 cm diameter × 110 cm length) equilibrated with 10 mM potassium phosphate buffer, pH 7.0, containing 50 mM KCl. The fractions containing CspD were pooled, concentrated and stored at −20°C. All purification steps were carried out at 4°C. Protein concentrations were measured according to the method of Bradford (1976) using Protein Assay Reagent (Bio-Rad) with bovine serum albumin (BSA) as a standard.

Circular dichroism (CD) measurement

CD measurement was performed on an automated AVIV 60DS spectrophotometer fitted with a thermostated cell holder controlled by an online temperature control unit. A quartz rectangular cell (Precision Cells) with a path length of 1 mm was used. The sample (CspD in 20 mM potassium phosphate buffer, pH 7.0, containing 50 mM NaCl) was filtered through a 0.22 µm filter before measurement. The scan was carried out at wavelengths between 260 and 190 nm in a cuvette with a 1 mm path length maintained at 4°C.

Preparation of DNA probes

A 141 bp HindIII–PvuII fragment was isolated from pUC9 and end labelled by Klenow filled in with [α-32P]-dATP to label only one strand of DNA. Alternatively, PCR was carried out with a 32P-labelled T7 promoter primer and a T7 terminator primer using pET-11a plasmid DNA as a template. Again, only one strand of DNA was labelled. The PCR product was then digested with BamHI for the 145 bp fragment, NdeI for the 87 bp fragment or XbaI for the 47 bp fragment, and the resulting fragments were purified. When the ssDNA was used as a probe, the labelled dsDNA was heat denatured.

Preparation of RNA probe

A DNA template containing the promoter sequence for T7 RNA polymerase was prepared according to the method of Phadtare and Inouye (1999). Oligonucleotide #8150, 5′-TGGGCACTATTTATATCAACATGATCTCAAGGCAGNNNNNNNNNNNNNNNACCCCAAGTGTCGTGAATGTCGTTGGTGGCCC-3′, where N indicates a mixture of A, G, C and T, was used as a template and amplified by PCR with primer #8151, 5′-CGCGGATCCTAATACGACTCACTATAGGGGCCACCAACGACATT-3′, and primer #8152, 5′-CCCGACACCCGCGGATCCATGGGCACTATTTATATCAAC-3′, where underlining indicates the BamHI site and letters in italics indicate the promoter for T7 RNA polymerase. The PCR fragment was digested with BamHI, purified by polyacrylamide gel and used as a template to prepare the labelled RNA using T7 RNA polymerase and [α-32P]-UTP.

DNA and RNA binding assay

The binding assay was carried out in a 15 µl reaction containing 10 mM Tris-HCl, pH 8.0, 1 mM EDTA, 50 mM KCl, 7.4% (w/v) glycerol, DNA or RNA probe and purified CspD as described previously (Jiang et al., 1997). The reaction mixtures were incubated on ice for 20 min and analysed by an 8% polyacrylamide gel. TBE (50 mM Tris-boric acid, 1 mM EDTA) was used as a running buffer.

In vitro DNA replication assays

In vitro DNA replication using the oriC plasmid as a template was carried out as described previously (Crooke, 1995). Briefly, assays were carried out in a 25 µl reaction mixture containing 30 mM HEPES-KOH, pH 7.6, 9 mM magnesium acetate, 2 mM ATP, 100 µM each dATP, dCTP, dGTP and [α-32P]-dTTP, 200 ng (600 pmol as a nucleotide) oriC plasmid, 525 ng of SSB, 14 ng of primase, 400 ng of gyrase A subunit, 180 ng of gyrase B subunit, 26 ng of β-subunit, 127 ng of DnaB–DnaC complex, 130 ng of DnaA, 120 ng of DNA polymerase III* and 10 ng of HU. Reaction mixtures were incubated at 30°C for 20 min, and the reaction was stopped by the addition of 10% trichloroacetic acid and 100 mM pyrophosphate. Incorporation of [α-32P]-dTMP into acid-insoluble materials was measured by liquid scintillation counting.

Electron microscopy

PstI-linearized, heat-denatured φX174 DNA (5386 bases) was used as a ssDNA for substrate. The binding reaction was carried out in a 15 µl reaction mixture containing 20 mM HEPES-AcOH, pH 7.5, 50 mM NaCl, 0.6 µg of DNA (0.34 pmol of ssDNA), and 25 µg of CspD (1563 pmol as dimers) or 5 µg of SSB (66 pmol as tetramers) for 15 min on ice. The complexes were fixed with 0.6% glutaraldehyde (v/v) for 10 min at room temperature, followed by filtration through a 2 ml column of Bio-Gel A5m (Bio-Rad) to remove excess glutaraldehyde and free proteins. The fractions containing DNA–protein complexes were mixed with a buffer containing 2 mM spermidine, applied to glow-charged carbon-coated grids and washed with a sequential water–ethanol series. The samples were then air dried and rotary shadowcast with tungsten (Griffith and Christiansen, 1978). Samples were examined and micrographs were taken in a Philips 420 electron microscope.

Chemicals and enzymes

All restriction enzymes and DNA modification enzymes were purchased from New England Biolabs, Boehringer Mannheim or Gibco BRL. T7 RNA polymerase and RNase inhibitor were from Boehringer Mannheim. Radioisotopes were from Amersham or New England Nuclear. All reagents used were analytical grade.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

We thank Dr J.-P. Etchegaray, Dr S. Gottesman, Dr S. Ujwal and Ms N. Wang for constructing pET11-cspD, strain SG22500, CD measurement and preparing Fig. 4F respectively. This work was supported by grants from the National Institute of Health (GM19043 to M.I. and GM49700 to E.C.).

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  • Almirón, M., Link, A.J., Furlong, D. & Kolter, R. (1992) A novel DNA binding protein with regulatory and protective roles in starved Escherichia coli. Genes Dev 6: 26462654.
  • Bae, W., Jones, P.G. & Inouye, M. (1997) CspA, the major cold-shock protein of Escherichia coli, negatively regulates its own gene expression. J Bacteriol 179: 70817088.
  • Bouvet, P., Matsumoto, K. & Wolffe, A.P. (1995) Sequence-specific RNA recognition by the Xenopus Y-box proteins: an essential role for the cold shock domain. J Biol Chem 270: 2829728303.
  • Boye, E., Stokke, T., Kleckner, N. & Skarstad, K. (1996) Coordinating DNA replication initiation with cell growth: differential roles for DnaA and SeqA proteins. Proc Natl Acad Sci USA 93: 1220612211.DOI: 10.1073/pnas.93.22.12206
  • Bradford, M.M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72: 248254.
  • Bramhill, D. & Kornberg, A. (1988) Duplex opening by DnaA protein at novel sequences in initiation of replication at the origin of the E. coli chromosome. Cell 52: 743755.
  • Brandi, A., Pietroni, P., Gualerzi, C.O. & Pon, C.L. (1996) Post-transcriptional regulation of CspA expression in Escherichia coli. Mol Microbiol 19: 231240.
  • Crooke, E. (1995) DNA synthesis initiated at oriC: in vitro replication reactions. Methods Enzymol 262: 500506.
  • Feng, W., Tejero, R., Zimmerman, D.E., Inouye, M. & Montelione, G.T. (1998) Solution NMR structure and backbone dynamics of the major cold-shock protein (CspA) from Escherichia coli: evidence for conformational dynamics in the single-stranded RNA-binding site. Biochemistry 37: 1088110896.
  • Funnell, B.E., Baker, T.A. & Kornberg, A. (1986) Complete enzymatic replication of plasmids containing the origin of the Escherichia coli chromosome. J Biol Chem 261: 56165624.
  • Gottesman, S., Clark, W.P. & Maurizi, M.R. (1990) The ATP-dependent Clp protease of Escherichia coli: sequence of clpA and identification of a Clp-specific substrate. J Biol Chem 265: 78867893.
  • Grant, R.A., Filman, D.J., Finkel, S.E., Kolter, R. & Hogle, J.M. (1998) The crystal structure of Dps, a ferritin homolog that binds and protects DNA. Nature Struct Biol 5: 294303.
  • Graumann, P. & Marahiel, M.A. (1994) The major cold shock protein of Bacillus subtilis CspB binds with high affinity to the ATTGG and CCAAT sequences in single stranded oligonucleotides. FEBS Lett 338: 157160.
  • Graumann, P. & Marahiel, M.A. (1998) A superfamily of proteins that contain the cold-shock domain. Trends Biochem Sci 23: 286290.DOI: 10.1016/s0968-0004(98)01255-9
  • Graumann, P., Schröder, K., Schmid, R. & Marahiel, M.A. (1996) Cold shock stress-induced proteins in Bacillus subtilis. J Bacteriol 178: 46114619.
  • Graumann, P., Wendrich, T.M., Weber, M.H.W., Schröder, K. & Marahiel, M.A. (1997) A family of cold shock proteins in Bacillus subtilis is essential for cellular growth and for efficient protein synthesis at optimal and low temperatures. Mol Microbiol 25: 741756.
  • Griffith, J.D. & Christiansen, G. (1978) Electron microscopic visualization of chromatin and other DNA protein complexes. Annu Rev Biophys Bioeng 7: 1935.
  • Hiraga, S., Niki, H., Ogura, T., Ichinose, C., Mori, H., Ezaki, B. & Jaffé, A. (1989) Chromosome partitioning in Escherichia coli: novel mutants producing anucleate cells. J Bacteriol 171: 14961505.
  • Hirota, Y., Ryter, A. & Jacob, F. (1968) Thermosensitive mutants of E. coli affected in the processes of DNA synthesis and cellular division. Cold Spring Harbor Symp Quant Biol 33: 677693.
  • Hwang, D.S. & Kornberg, A. (1990) A novel protein binds a key origin sequence to block replication of an E. coli minichromosome. Cell 63: 325331.
  • Jiang, W., Hou, Y. & Inouye, M. (1997) CspA, the major cold-shock protein of Escherichia coli, is an RNA chaperone. J Biol Chem 272: 196202.
  • Kaguni, J.M. & Kornberg, A. (1984) Replication initiated at the origin (oriC) of the E. coli chromosome reconstituted with purified enzymes. Cell 38: 183190.
  • Katayama, T., Kubota, T., Kurokawa, K., Crooke, E. & Sekimizu, K. (1998) The initiator function of DnaA protein is negatively regulated by the sliding clamp of E. coli chromosomal replicase. Cell 94: 6171.
  • Kitagawa, R., Mitsuki, H., Okazaki, T. & Ogawa, T. (1996) A novel DnaA protein-binding site at 94.7 min on the Escherichia coli chromosome. Mol Microbiol 19: 11371147.
  • Kitagawa, R., Okazaki, T., Moriya, S. & Ogawa, T. (1998) Negative control of replication initiation by a novel chromosomal locus exhibiting exceptional affinity for Escherichia coli DnaA protein. Genes Dev 12: 30323043.
  • Kohara, Y., Akiyama, K. & Isono, K. (1987) The physical map of the whole E. coli chromosome: application of a new strategy for rapid analysis and sorting of a large genomic library. Cell 50: 495508.
  • Kolter, R., Siegele, D.A. & Tormo, A. (1993) The stationary phase of the bacterial life cycle. Annu Rev Microbiol 47: 855874.
  • Kornberg, A. & Baker, T.A. (1992) DNA Replication, 2nd edn. New York: W.H. Freeman.
  • Lange, R. & Hengge-Aronis, R. (1991) Identification of a central regulation of stationary phase gene expression in Escherichia coli. Mol Microbiol 5: 4959.
  • Lu, M., Campbell, J.L., Boye, E. & Kleckner, N. (1994) SeqA: a negative modulator of replication initiation in E. coli. Cell 77: 413426.
  • McCann, M.P., Kidwell, J.P. & Matin, A. (1991) The putative σ factor KatF has a central role in development of starvation-mediated general resistance in Escherichia coli. J Bacteriol 173: 41884194.
  • Martinez, A. & Kolter, R. (1997) Protection of DNA during oxidative stress by the nonspecific DNA-binding protein Dps. J Bacteriol 179: 51885194.
  • Matsumoto, K. & Wolffe, A.P. (1998) Gene regulation by Y-box proteins: coupling control of transcription and translation. Trends Cell Biol 8: 318323.DOI: 10.1016/s0962-8924(98)01300-2
  • Messer, W. & Weigel, C. (1996) Initiation of chromosome replication. In Escherichia coli and Salmonella: Cellular and Molecular Biology, 2nd edn. Neidhardt, F.C., Curtiss, R., III, Ingraham, J.L., Lin, E.C.C., Low, K.B., Magasanik, B. (eds). Washington, DC: American Society for Microbiology Press, pp. 15791601.
  • Meyer, R. & Laine, P.S. (1990) The single-stranded DNA-binding protein of Escherichia coli. Microbiol Rev 54: 342380.
  • Miller, J.H. (1992) A Short Course in Bacterial Genetics: A Laboratory Manual and Handbook for Escherichia coli and Related Bacteria. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press.
  • Newkirk, K., Feng, W., Jiang, W., Tejero, R., Emerson, S.D., Inouye, M. & Montelione, G.T. (1994) Solution NMR structure of the major cold shock protein (CspA) from Escherichia coli: identification of a binding epitope for DNA. Proc Natl Acad Sci USA 91: 51145118.
  • Phadtare, S. & Inouye, M. (1999) Sequence-selective interactions with RNA by CspB, CspC and CspE, members of the CspA family of Escherichia coli. Mol Microbiol 33: 10041014.
  • Schindelin, H., Marahiel, M.A. & Heinemann. U. (1993) Universal nucleic acid-binding domain revealed by crystal structure of the B. subtilis major cold-shock protein. Nature 364: 164168.
  • Schindelin, H., Jiang, W., Inouye, M. & Heinemann. U. (1994) Crystal structure of CspA, the major cold shock protein of Escherichia coli. Proc Natl Acad Sci USA 91: 51195123.
  • Schnuchel, A., Wiltscheck, R., Czisch, M., Herrler, M., Willimsky, G. & Graumann, P., et al. (1993) Structure in solution of the major cold-shock protein from Bacillus subtilis. Nature 364: 169171.
  • Sekimizu, K., Bramhill, D. & Kornberg, A. (1988) Sequential early stages in the in vitro initiation of replication at the origin of the Escherichia coli chromosome. J Biol Chem 263: 71247130.
  • Slater, S., Wold, S., Lu, M., Boye, E., Skarstad, K. & Kleckner, N. (1995) E. coli SeqA protein binds the DNA replication origin in two different methyl-modulated reactions appropriate to its roles in replication initiation and origin sequestration. Cell 82: 927936.
  • Sledjeski, D.D., Gupta, A. & Gottesman, S. (1996) The small RNA, DsrA, is essential for the low temperature expression of RpoS during exponential growth in Escherichia coli. EMBO J 15: 39934000.
  • Stahl, F.W., Kobayashi, I., Thaler, D. & Stahl, M.M. (1986) Direction of travel of RecBC recombinase through bacteriophage lambda DNA. Genetics 113: 215227.
  • Studier, F.W., Rosenberg, A.H., Dunn, J.J. & Dubendorff, J.W. (1990) Use of T7 RNA polymerase to direct expression of cloned genes. Methods Enzymol 185: 6086.
  • Takeshita, S., Sato, M., Toba, M., Masahashi, W. & Hashimoto-Gotoh, T. (1987) High-copy-number and low-copy-number plasmid vectors for lacZα-complementation and chloramphenicol- or kanamycin-resistance selection. Gene 61: 6374.
  • Talukder, A.A. & Ishihama, A. (1999) Twelve species of the nucleoid-associated protein from Escherichia coli: sequence recognition specificity and DNA binding affinity. J Biol Chem 274: 3310533113.
  • Talukder, A.A., Iwata, A., Nishimura, A., Ueda, S. & Ishihama, A. (1999) Growth phase-dependent variation in protein composition of the Escherichia coli nucleoid. J Bacteriol 181: 63616370.
  • Thöny, B., Hwang, D.S., Fradkin, L. & Kornberg, A. (1991) iciA, an Escherichia coli gene encoding a specific inhibitor of chromosomal initiation of replication in vitro. Proc Natl Acad Sci USA 88: 40664070.
  • VanBogelen, R.A., Hutton, M.E. & Neidhardt, F.C. (1990) Gene–protein database of Escherichia coli K-12: edition 3. Electrophoresis 11: 11311166.
  • Vieira, J. & Messing, J. (1982) The pUC plasmids, an M13mp7-derived system for insertion mutagenesis and sequencing with synthetic universal primers. Gene 19: 259268.
  • Wada, A., Igarashi, K., Yoshimura, S., Aimoto, S. & Ishihama, A. (1995) Ribosome modulation factor: stationary growth phase-specific inhibitor of ribosome function from Escherichia coli. Biochem Biophys Res Commun 214: 410417.DOI: 10.1006/bbrc.1995.2302
  • Willimsky, G., Bang, H., Fischer, G. & Marahiel, M.A. (1992) Characterization of cspB, a Bacillus subtilis inducible cold shock gene affecting cell viability at low temperatures. J Bacteriol 174: 63266335.
  • Yamagishi, M., Wada, A., Sakagami, M., Fujita, N. & Ishihama, A. (1993) Regulation of the Escherichia coli rmf gene encoding the ribosome modulation factor: growth phase- and growth rate-dependent control. EMBO J 12: 625630.
  • Yamanaka, K. (1999) Cold shock response in Escherichia coli. J Mol Microbiol Biotechnol 1: 193202.
  • Yamanaka, K. & Inouye, M. (1997) Growth-phase dependent expression of cspD, encoding a member of the CspA family in Escherichia coli. J Bacteriol 179: 51265130.
  • Yamanaka, K. & Inouye, M. (2001) Induction of CspA, an E. coli major cold-shock protein, upon nutritional upshift at 37°C. Genes Cells (in press).
  • Yamanaka, K., Fang, L. & Inouye, M. (1998) The CspA family in Escherichia coli: multiple gene duplication for stress adaptation. Mol Microbiol 27: 247255.DOI: 10.1046/j.1365-2958.1998.00683.x
  • Zhang, W., Inouye, M. & Inouye, S. (1996) Reciprocal regulation of the differentiation of Myxococcus xanthus by Pkn5 and Pkn6, eukaryotic-like Ser/Thr protein kinase. Mol Microbiol 20: 435447.
Footnotes
  1. Present address: Department of Molecular Cell Biology, Institute of Molecular Embryology and Genetics, Kumamoto University, 4-24-1 Kuhonji, Kumamoto 862-0976, Japan.