SEARCH

SEARCH BY CITATION

Abstract

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

σS (RpoS) is a highly unstable global regulatory protein in Escherichia coli, whose degradation is inhibited by various stress signals, such as carbon starvation, high osmolarity and heat shock. As a consequence, these stresses result in the induction of σS-regulated stress-protective proteins. The two-component-type response regulator, RssB, is essential for the rapid proteolysis of σS and is probably involved in the transduction of some of these stress signals. Acetyl phosphate can be used as a phosphodonor for the phosphorylation of various response regulators in vitro and, in the absence of the cognate sensor kinases, acetyl phosphate can also modulate the activities of several response regulators in vivo. Here, we demonstrate increased in vivo half-lives of σS and the RpoS742::LacZ hybrid protein (also a substrate for RssB-dependent proteolysis) in acetyl phosphate-free (ptaackA) deletion mutants, even though no sensor kinase was eliminated. The in vivo data indicate that acetyl phosphate acts through the response regulator, RssB. In vitro, efficient phosphotransfer from radiolabelled acetyl phosphate to the Asp-58 residue of RssB (the expected site of phosphorylation in the RssB receiver domain) was observed. Via such phosphorylation, acetyl phosphate may thus modulate RssB activity even in an otherwise wild-type background. While acetyl phosphate is not essential for the transduction of specific environmental stress signals, it could play the role of a modulator of RssB-dependent proteolysis that responds to the metabolic status of the cells reflected in the highly variable cellular acetyl phosphate concentration.


Introduction

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

Two-component systems are common signal-transducing devices in prokaryotes, which operate by reversible and serial protein phosphorylation in response to environmental or intracellular signals (for recent reviews, see Hoch and Silhavy, 1995). Typically, these systems consist of a sensor kinase, which autophosphorylates at a histidyl residue located in the C-terminal transmitter domain in response to a signal perceived by its N-terminal sensory domain. The phosphorylated sensor kinase then serves as a phosphodonor for the phosphorylation of an aspartyl residue in the N-terminal receiver domain of a response regulator. This phosphorylation in turn influences the activity of the C-terminal output or effector domain of the response regulator. In the absence of specific signals, many sensor kinases also stimulate dephosphorylation of their cognate response regulators (Stock et al., 1995).

The discovery that small phosphorylated compounds, such as acetyl phosphate, can act as phosphodonors for response regulators in vitro (Feng et al., 1992; Lukat et al., 1992) has lead to the suggestion that the cellular level of acetyl phosphate may also influence the activity of response regulators in vivo (Feng et al., 1992; McCleary et al., 1993; McCleary and Stock, 1994; Wanner, 1992; 1995). The cellular contents of acetyl phosphate, which is synthesized from acetyl-CoA and Pi by phosphotransacetylase (pta) and from acetate and ATP by acetate kinase (ackA), are indeed highly variable, depending on the nutritional status of the cells. For instance, growth on glucose, pyruvate or rich medium results in high levels of acetyl-CoA and acetyl phosphate and the production of acetate via the pta–ackA pathway, whereas growth on glycerol or entry into stationary phase in rich medium results in lower acetyl phosphate levels (McCleary and Stock, 1994; Prüß and Wolfe, 1994). However, the total absence of acetyl phosphate as a result of mutations in pta and ackA usually affects response regulator-controlled processes only in genetic backgrounds that also lack the cognate sensor kinases/phosphatases. This has been found for the control of the pho regulon (Wanner and Wilmes-Riesenberg, 1992; Wanner, 1995; Kim et al., 1996), the nitrogen regulatory system (ntr ) (Feng et al., 1992) and the chemotaxis system (Dailey and Berg, 1993). So far, the only exception to this apparent rule has been OmpR-mediated repression of flagella synthesis, which is modulated by acetyl phosphate in otherwise wild-type strains (Prüß and Wolfe, 1994; Shin and Park, 1995). Therefore, it is still under debate whether cellular acetyl phosphate levels are physiologically relevant for the activities of response regulator proteins in vivo.

In the present study, we investigated whether acetyl phosphate affects the activity of the response regulator, RssB. RssB (also termed SprE or MviA in Salmonella typhimurium) is essential for the rapid degradation of the rpoS-encoded σS subunit of RNA polymerase (Bearson et al., 1996; Muffler et al., 1996a; Pratt and Silhavy, 1996). This proteolysis of σS, a global regulatory protein, is tightly regulated by external signals. Inhibition of σS turnover can be observed in response to carbon starvation, shift to high medium osmolarity or heat shock (Lange and Hengge-Aronis, 1994; Takayanagi et al., 1994; Muffler et al., 1996b; 1997). Thus, all these environmental stress conditions lead to a rapid increase in cellular σS levels and, consequently, to the activation of σS-regulated genes, which have broadly stress-protective functions (for reviews, see Hengge-Aronis, 1993; Loewen and Hengge-Aronis, 1994).

Here, we demonstrate that, in vivo, acetyl phosphate affects the proteolysis of σS and the RpoS742::LacZ hybrid protein, which is also a substrate for RssB-dependent protein turnover. In contrast to most other systems, the in vivo effects of acetyl phosphate were observed without eliminating any sensor kinase, and thus seem physiologically relevant. In vitro, acetyl phosphate acts as an efficient phosphodonor for the phosphorylation of the Asp-58 residue of RssB. We, therefore, propose that acetyl phosphate modulates the activity of RssB and, thereby, proteolysis of σS and RpoS742::LacZ.

Results

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

A lack of acetyl phosphate results in reduced rates of proteolysis of σS and the RpoS742::LacZ hybrid protein

The cellular σS content is regulated at the levels of rpoS transcription and translation as well as degradation of σS protein (Lange and Hengge-Aronis, 1994). For distinguishing the effects of these different levels of control, reporter gene fusions between rpoS and lacZ have proved to be extremely useful. In particular, a comparison of the β-galactosidase activities produced from the translational fusions rpoS379::lacZ (inserted after nucleotide 379 within rpoS) and rpoS742::lacZ (inserted after nucleotide 742) allows detection of the effects on σS turnover, because the former encodes a stable hybrid protein whose activity reflects σS synthesis only, whereas the latter encodes a larger hybrid protein that contains a ‘turnover element’ and is therefore subject to regulated proteolysis just as σS itself (Muffler et al., 1996b). Only RpoS742::LacZ levels are affected in a mutant deficient for the response regulator, RssB, which is essential for σS degradation (Muffler et al., 1996a; Pratt and Silhavy, 1996), or by signals, such as carbon starvation or heat shock, that act exclusively on σS turnover (Lange and Hengge-Aronis, 1994; Muffler et al., 1997).

As acetyl phosphate can influence the activities of response regulator proteins, we wanted to know whether acetyl phosphate influences σS degradation. We tested the effect of a (ptaackA) deletion mutation, which results in a total lack of acetyl phosphate, on the β-galactosidase activities produced by strains carrying the two reporter gene fusions. Activities were clearly increased in the RpoS742::LacZ-containing strain (Fig. 1A) but were unaffected in the strain that expresses the shorter and stable RpoS379::LacZ hybrid protein (Fig. 1B). These data indicated that the absence of acetyl phosphate does not affect the synthesis but the degradation of RpoS742::LacZ and, therefore, probably of σS and suggested that, during exponential growth, the half-lives of these unstable proteins might be longer in acetyl phosphate-free mutants.

image

Figure 1. . A lack of acetyl phosphate affects the hybrid protein RpoS742::LacZ, which is subject to σS-like turnover, but not the stable RpoS379::LacZ hybrid. Strains carrying the translational fusions rpoS742::lacZ (A) and rpoS379::lacZ (B) in either pta+ackA+ (circles) or Δ(pta–ackA) backgrounds (triangles) were grown in M9 minimal medium with 0.025% glucose. Optical densities (closed symbols) and specific β-galactosidase activities (open symbols) were determined.

Download figure to PowerPoint

Using pulse-chase labelling experiments followed by immunoprecipitation of σS and RpoS742::LacZ, we determined the in vivo half-lives of these proteins. As the rpoS742::lacZ fusion is located in single copy on a λ phage inserted at the att(λ) site in the chromosome and the chromosomal copy of rpoS present in the strains used is a wild-type allele, both proteins can be monitored in parallel in the same strains. Quantitation of the autoradiographs obtained (Fig. 2A and D) demonstrates the instability of σS and the RpoS742::LacZ hybrid protein in the pta+ackA+ strain during exponential growth with glucose as a carbon source (which is known to sustain high cellular acetyl phosphate levels; McCleary and Stock, 1994). In the strain unable to synthesize acetyl phosphate, both proteins were already more stable during exponential growth (Fig. 2B and E). Half-lives of σS and the RpoS742::LacZ hybrid protein were calculated (Table 1). In the acetyl phosphate-free mutant background, the half-lives of the two proteins were increased by the factors 2.5 and 3, respectively, when compared with the otherwise isogenic parental strain. Thus, acetyl phosphate promotes rapid degradation of σS and RpoS742::LacZ in wild-type cells growing in glucose minimal medium. However, in contrast to the response regulator RssB, the presence of acetyl phosphate is not an absolute prerequisite for degradation, but rather is a modulatory factor.

image

Figure 2. . In vivo turnover of σS and RpoS742::LacZ. Cells were grown in M9 minimal medium with 0.1% glucose. Samples were withdrawn during late exponential phase (at an optical density of approximately 0.7) (A, B, D and E) and 20 min after the onset of glucose starvation (C and F) and pulse labelled with [35S]-methionine as described in Experimental procedures. σS and RpoS742::LacZ were immunoprecipitated from cell extracts prepared from pulse-labelled cells and were visualized by SDS–PAGE and autoradiography. A densitometric quantitation of the autoradiographic data is shown, with A, B and C representing the degradation of σS and D, E and F that of RpoS742::LacZ. A and D. The data obtained for the pta+ackA+ strain (RO91) during the exponential growth phase, whereas the data for the otherwise isogenic acetyl phosphate-free Δ(pta–ackA) mutant in exponential phase and 20 min after the onset of glucose starvation are shown in B and E and in C and F respectively. The values obtained are expressed relative to the values for the shortest chase time.

Download figure to PowerPoint

Table 1. . In vivo half-lives of σS and the RpoS742::LacZ hybrid protein. a. At an OD578 of 0.7 of a culture growing in M9 medium with 0.1% glucose.b. Twenty minutes after the onset of glucose starvation.c. Values in parentheses are the factors by which half-lives are increased in comparison to the half-lives determined in exponentially growing wild-type cells.Thumbnail image of

The response regulator RssB is required for the acetyl phosphate effect in vivo and is phosphorylated by acetyl phosphate in vitro

As RssB is essential for the degradation of σS and RpoS742::LacZ, acetyl phosphate as a putative phosphodonor for response regulators might act through RssB. In order to test this hypothesis, we analysed whether the (ptaackA) mutation also affected RpoS742::LacZ in a genetic background that is deficient for RssB. As shown in Fig. 3, this was not the case. We concluded that acetyl phosphate requires RssB in order to exert its effect.

image

Figure 3. . In the rssB mutant background, the absence of acetyl phosphate does not affect the hybrid protein RpoS742::LacZ. Strains carrying the translational fusion rpoS742::lacZ and the ΔrssAB::cat mutation in either pta+ackA+ (circles) or Δ(pta–ackA) backgrounds (triangles) were grown in M9 minimal medium with 0.025% glucose. Optical densities (closed symbols) and specific β-galactosidase activities (open symbols) were determined.

Download figure to PowerPoint

In order to demonstrate in vitro phosphorylation of RssB by acetyl phosphate, a thioredoxin-(His)6-tagged variant of RssB protein was purified (see Experimental procedures for details of construction and purification). After purification, the thioredoxin-(His)6 moiety was cleaved off from the hybrid protein using an enterokinase cleavage site. As a control, a similarly purified RssB mutant protein was used, in which the Asp-58 residue (which was expected to be the phosphorylated amino acid in RssB by homology with other response regulators) was replaced by a proline residue. This RssB(D58P) variant was chosen among a series of mutants with exchanges at Asp-58, because it complemented a rssB::Tn10 mutation (resulting in low levels of σS during exponential growth), i.e. RssB(D58P) is active in promoting σS degradation. As the receiver domain of RssB seems to play a positive role in the activation of RssB, the D58P receiver cannot exhibit strong structural disruptions that might interfere with phosphorylation in general (D. Fischer and R. Hengge-Aronis, unpublished results).

Phosphotransfer from acetyl [32P]-phosphate was assayed with crude extracts from strains producing the two hybrid proteins, with the purified hybrid proteins and with the RssB proteins from which the tag had been cleaved off (Fig. 4). In all cases, 32P-radiolabelled products of the expected sizes were observed with the RssB variants carrying Asp-58, whereas no phosphotransfer was obtained with the corresponding D58P mutants. In addition, phosphotransfer to RssB was extremely rapid (Fig. 4B). Under similar conditions, phosphotransfer from acetyl phosphate to the response regulators, OmpR (Kenney et al., 1995), PhoB (McCleary and Stock, 1994), NarL (Schröder et al., 1994) and KdpE (M. Lucassen and K. Jung, unpublished results), occurs with significantly slower kinetics.

image

Figure 4. . In vitro phosphorylation of RssB with acetyl [32P]-phosphate as a phosphodonor. Total cellular extracts from cells expressing the thioredoxin-(His)6-RssB or thioredoxin-(His)6-RssB(D58P) hybrid proteins (A, lanes 1–4), the purified hybrid proteins (A, lanes 5–8) and purified RssB and RssB(D58P) (B) were tested (for details of protein purification and the phosphorylation assay, see Experimental procedures). RssB variants with the wild-type Asp-58 residue are shown in lanes 1, 2, 5 and 6 in (A), and in lanes 1–6 in (B). The corresponding RssB(D58P) variants are shown in lanes 3, 4, 7 and 8 in (A) and in lane 7 in (B). Incubation times with acetyl [32P]-phosphate were 5 min (lanes 1, 3, 5 and 7) and 45 min (lanes 2, 4, 6 and 8) in (A), and 1 min (lane 1), 3 min (lane 2), 5 min (lane 3), 10 min (lane 4), 15 min (lane 5) and 30 min (lanes 6 and 7) in (B). Purified protein amounts applied per lane were 11 μg [hybrid proteins, RssB(D58P)] or 7.5 μg (RssB).

Download figure to PowerPoint

These data are the first reported evidence that RssB, which so far qualified as a response regulator by sequence criteria, can indeed be phosphorylated at the expected Asp-58 residue. In addition, it has been shown that acetyl phosphate can serve as an efficient phosphodonor for the phosphorylation of RssB at this position.

Acetyl phosphate is not essential for acute stress signal transduction in the control of σS

In the acetyl phosphate-free mutant, proteolysis of σS and RpoS742::LacZ occurs at a reduced rate during exponential growth, but turnover was further inhibited in response to glucose starvation (Fig. 2C and F, Table 1[link]). The stationary phase half-lives of the two proteins were similar to the corresponding half-lives in wild-type strains (Table 1; Lange and Hengge-Aronis, 1994). This means that acetyl phosphate-free cells can sense glucose starvation and react appropriately. Consequently, a drop in the cellular acetyl phosphate level elicited by the exhaustion of glucose in the growth medium cannot be the (only) starvation signal that triggers σS stabilization.

We also wanted to know whether acetyl phosphate is essential for the inhibition of σS and RpoS742::LacZ degradation, which is found after a shift to high osmolarity. Figure 5 demonstrates that the osmotic induction of RpoS742::LacZ (which reflects both increased synthesis and reduced turnover of σS) is very similar in acetyl phosphate-producing and acetyl phosphate-free strains, even though in the latter β-galactosidase activities measured before the osmotic shift are higher (as was expected from the data shown above; Fig. 1).

image

Figure 5. . Osmotic induction of rpoS is not affected by a lack of acetyl phosphate. The rpoS742::lacZ-carrying strain RO91 (diamonds, squares) and an otherwise isogenic Δ(pta–ackA) derivative (circles, triangles) were grown in M9 medium with 0.4% glycerol. At optical densities of approximately 0.3, the cultures were split into two aliquots each, one of which was supplemented with 0.3 M NaCl. Optical densities (closed symbols) and specific β-galactosidase activities (open symbols) were determined in the absence (diamonds, circles) and in the presence of NaCl (squares, triangles).

Download figure to PowerPoint

Discussion

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

Since the discovery that acetyl phosphate can be used as a phosphodonor in the in vitro phosphorylation of response regulator proteins (Feng et al., 1992; Lukat et al., 1992), it has been speculated that acetyl phosphate, whose cellular concentration is highly variable depending on nutritional conditions (McCleary and Stock, 1994; Prüß and Wolfe, 1994), might also be a modulator of response regulator function in vivo (Feng et al., 1992; McCleary et al., 1993; McCleary and Stock, 1994; Wanner, 1995). Unfortunately, attempts to demonstrate this experimentally have mostly been disappointing. In several cases studied, the lack of acetyl phosphate caused by mutations in the genes encoding phosphotransacetylase (pta) and acetate kinase (ackA) had an effect on response regulator-controlled processes — but only when the cognate sensor kinases were eliminated by mutation (Feng et al., 1992; Wanner and Wilmes-Riesenberg, 1992; Dailey and Berg, 1993; Wanner, 1995; Kim et al., 1996; McCleary, 1996). It seems that, in the absence of the specific signals that normally activate these systems, phosphotransfer from acetyl phosphate to the response regulators is rapidly counteracted by the activity of phosphatases, which are the non-phosphorylated forms of the sensor kinases in many of these systems. The actual occurrence of such phosphotransfer is suggested by the finding that a strain deficient for the sensor kinase NtrB contains eightfold elevated acetyl phosphate levels (Prüß and Wolfe, 1994). Thus, in the wild-type strain, there may be a futile cycle of NtrC phosphorylation and dephosphorylation at the expense of acetyl phosphate.

In a few systems, however, acetyl phosphate affects response regulator output even in otherwise wild-type backgrounds. One is the repression of flagella synthesis by OmpR, which is relieved in acetyl phosphate-free mutants (Prüß and Wolfe, 1994; Shin and Park, 1995). OmpR represses the flhDC operon, which encodes the master regulator in the regulatory hierarchy governing the expression of flagellar and motility genes. Two regions located upstream and downstream of the promoter of the flhDC operon are bound by OmpR with phosphorylation-dependent affinity (Shin and Park, 1995). The other system affected by acetyl phosphate is RssB-dependent σS proteolysis as reported in this study. In vivo half-lives of σS and the RpoS742::LacZ hybrid protein (also a substrate for RssB-dependent proteolysis) are 2.5- to threefold increased in acetyl phosphate-free mutants (Fig. 2, Table 1[link]). Acetyl phosphate affects the turnover of σS and RpoS742::LacZ but not the synthesis of these proteins, because the somewhat shorter and stable RpoS379::LacZ hybrid protein (whose activity reflects rpoS transcription and translation, but not σS turnover) was not affected (Fig. 1B). To our knowledge, these two systems are the only response regulator-controlled processes for which a dependency upon acetyl phosphate in vivo has been demonstrated in genetic backgrounds not deficient for any sensor kinase, indicating that acetyl phosphate plays a physiologically relevant modulatory role in these systems.

It has been proposed that OmpR activity is sensitive to acetyl phosphate because, in contrast to other systems, its cognate sensor kinase and response regulator phosphatase, EnvZ, is present at a cellular concentration that is approximately two orders of magnitude below that of the response regulator, OmpR (Shin and Park, 1995). At present, we do not know of a sensor kinase acting as a phosphodonor for the response regulator RssB. It is even conceivable that RssB is phosphorylated mainly by cross-regulation, i.e. by other non-cognate sensor kinases as well as by acetyl phosphate, and that a still unknown RssB phosphatase is the main target for regulation under stress conditions that interfere with RssB activity and, thereby, with σS proteolysis.

We have also shown here that RssB can be readily phosphorylated in vitro using acetyl [32P]-phosphate. Phosphorylation most probably occurs at the Asp-58 residue, as replacing Asp-58 by proline eliminated phosphorylation of RssB (but not the ability of the protein to promote σS degradation). This is also the first experimental demonstration that RssB is indeed phosphorylated at an aspartyl residue, as expected for a response regulator protein. In view of these in vitro data, it seems likely that the in vivo effect of acetyl phosphate on σS and RpoS742::LacZ proteolysis results from an influence of acetyl phosphate on the phosphorylation of RssB. Consistent with this, RssB belongs to a subset of response regulators that exhibit very rapid kinetics of phosphorylation by acetyl phosphate, whereas others are phosphorylated more slowly under similar conditions (McCleary and Stock, 1994; Schröder et al., 1994; Kenney et al., 1995). This may indicate a relatively high specificity for the reaction of RssB with acetyl phosphate. Nevertheless, it can also not be excluded that a certain metabolic imbalance created by a disruption of the pta–ackA pathway represents some intracellular stress that might somehow indirectly affect or contribute to a reduction in RssB activity.

A lack of acetyl phosphate increases the in vivo half-lives of σS and RpoS742::LacZ but does not totally abolish turnover of these proteins (Table 1). Acetyl phosphate is thus not an essential component for RssB-dependent protein degradation (and cannot be the only phosphodonor for RssB phosphorylation), but rather modulates or fine tunes the rate of RssB-mediated proteolysis. In agreement with such a role for acetyl phosphate, we have observed that the cells do not need acetyl phosphate for sensing acute environmental stress. In the acetyl phosphate-free pta–ackA mutant, σS and RpoS742::LacZ are stabilized further in response to sudden glucose starvation (Fig. 2, Table 1[link]) and, also, the reaction to hyperosmotic shift is normal in the absence of acetyl phosphate (Fig. 5). A decrease in the cellular acetyl phosphate level as a consequence of glucose exhaustion thus cannot be the (only) starvation signal. Rather, the physiological role of acetyl phosphate may be that of a modulator of RssB-dependent proteolysis that responds to the metabolic status of the cell, which is reflected in the highly variable cellular acetyl phosphate concentration. Alternatively, acetyl phosphate may just contribute to a more or less constitutive phosphorylation of RssB, with the tightly regulated process in the control of RssB being its dephosphorylation in response to certain stress conditions.

Experimental procedures

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

Bacterial strains and growth conditions

The bacterial strains used in this study are listed in Table 2. Strains were constructed by P1 transduction as described previously (Miller, 1972). The Δ(pta–ackA) mutation was introduced by co-transduction with a zfa::Tn10 insertion.

Table 2. . Bacterial strains. a. rssB::Tn10 and ΔrssAB::cat confer identical phenotypes (no degradation of σS), i.e. in a rssB null mutant background, eliminating rssA is of no consequence (Muffler et al., 1996a).Thumbnail image of

The rpoS::lacZ fusion constructs used in this study are present in single copy on λ phages located at the att(λ) site in the chromosome and contain more than 1 kb of the DNA upstream of rpoS, which includes all the promoters that contribute to rpoS expression (see Lange and Hengge-Aronis, 1994 for a detailed description of the construction of these fusions). The rpoS379::lacZ and rpoS742::lacZ translational fusions (the numbers indicate the nucleotides within rpoS after which the fusions are inserted) exhibit identical transcriptional and translational regulation. However, they differ in the stability of the gene products. Whereas the RpoS379::LacZ hybrid protein is stable, RpoS742::LacZ, which contains a ‘turnover element’ not present in the shorter fusion, is subject to the same degradation as σS itself (Muffler et al., 1996b).

Cultures were grown at 37°C under aeration in Luria–Bertani (LB) medium or minimal medium M9 (Miller, 1972) supplemented with glucose or glycerol in concentrations as indicated. Antibiotics were added as recommended by Miller (1972). Growth was monitored by measuring the optical density at 578 nm (OD578).

SDS–PAGE and immunoblot analysis

Sample preparation for SDS–PAGE (Laemmli, 1970) and immunoblot analysis were performed as described previously (Lange and Hengge-Aronis, 1994). A polyclonal serum against σS, a goat anti-rabbit IgG alkaline phosphatase conjugate (Sigma) and a chromogenic substrate (BCIP/NBT; Boehringer Mannheim) were used for visualization of σS bands.

Pulse-labelling of cells and immunoprecipitation

Pulse labelling of cells with L-[35S]-methionine and immunoprecipitation of σS has been described previously (Lange and Hengge-Aronis, 1994). The OD578 of the samples was adjusted by dilution with supernatant from the same cultures obtained by centrifugation immediately before taking the samples for pulse labelling. For the determination of σS and RpoS742::LacZ half-lives, the pulse time was 60 s and 120 s for exponentially growing and starved cells, respectively, and the chase times varied between 0.25 min and 12 min. As a σS-deficient control, strain RH90 was used (labelled in exponential phase samples harvested at an OD578 between 0.5 and 0.7). For immunoprecipitation, polyclonal sera against σS andβ-galactosidase were used. Protein bands on autoradiographs were quantitated densitometrically. The intensity of the bands representing σS and the RpoS742::LacZ hybrid protein was calculated relative to the intensity of the bands representing stable proteins that cross-reacted weakly with the antisera used.

DNA manipulations

For restriction digests, ligation, transformation and agarose gel electrophoresis, standard procedures were followed (Silhavy et al., 1984; Sambrook et al., 1989). Plasmid DNA preparations were performed with the Qiagen plasmid kit; for the recovery of DNA from agarose gels, the Qiaquick gel extraction kit was used (Qiagen). Bacterial chromosomal DNA was prepared as described previously (Silhavy et al., 1984). PCR amplification of DNA was performed according to standard procedures (Innis et al., 1990) with primers purchased from MWG Biotech in a Perkin-Elmer GeneAmp PCR System 9600. Non-radioactive double-strand DNA sequencing was performed with digoxygenin (DIG)-labelled primers (MWG Biotech) using a cycle sequencing protocol in the Perkin-Elmer GeneAmp PCR System 9600, followed by electrophoresis and direct blotting onto nylon membranes (GATC) in the GATC 1500 Direct Blotting Electrophoresis DNA Sequencer.

Cloning of the rssB gene

For the purification of RssB protein, the rssB coding region was cloned into the pET32a vector (Novagen; a NcoI/XhoI-digested PCR fragment obtained with MC4100 chromosomal DNA as the template and the primers 5′-AACGTCCATGGCTACGCAGCCATTGGTCGGAAAACAG-3′ and 5′-TGCACTCGAGTCATTATTCTGCAGACAACATCAAGCG-3′ was used, with underlined nucleotides indicating additions to or deviations from the wild-type sequence introduced in order to create NcoI and XhoI restriction sites, respectively, as well as an additional stop codon in the second primer). The resulting plasmid encodes a thioredoxin(Trx)-(His)6-RssB hybrid protein and was termed pEK9. As a control, a corresponding plasmid encoding Trx-(His)6-RssB(D58P) was constructed. The rssB mutant allele specifying proline instead of Asp-58 (D58P) is one of a series of mutants with exchanges specifically in the codon for Asp-58 originally obtained with an rssB-carrying pBAD18 derivative (D. Fischer and R. Hengge-Aronis, unpublished results). The D58P mutation was transferred onto pEK9 by subcloning a NcoI/SnaBI-digested PCR fragment (obtained with the rssBD58P-carrying pBAD18 derivative as a template and the first primer mentioned above and a second primer with the sequence 5′-GATAGAGACAGGCAAAAACCATCTCGCGC-3′) into pEK9 also digested with NcoI and SnaBI. The resulting plasmid was termed pEK10. The gene products of the two plasmids, i.e. Trx-(His)6-RssB and Trx-(His)6-RssB(D58P), exhibited the expected sizes on SDS gels, and both complemented a chromosomal rssB::Tn10 mutation (cellular levels of σS were assayed by immunoblot analysis). Relevant parts of plasmid sequences were verified by dideoxy sequencing.

Purification of RssB and RssB(D58P)

Overnight cultures of strain BL21(DE3) carrying either pEK9 or pEK10 grown in LB medium with 200 μg ml−1 ampicillin were diluted 100-fold and grown in LB medium with 100 μg ml−1 ampicillin at 28°C to an OD578 of 0.5–0.6. IPTG (1 mM) and 3% ethanol (to minimize the formation of inclusion bodies) were then added. After 2.5 h, cells were harvested and washed twice with 50 mM Tris-HCl, pH 7.5, 50 mM NaCl and 1 mM EDTA. All the following purification steps were carried out at 4°C. Cells were resuspended in French press buffer (50 mM Tris-HCl, pH 7.5, 50 mM KCl, 5 mM MgCl2 and 1 mM EDTA) and disrupted in a French pressure cell. The lysate was cleared by centrifugation (30 min at 30 000 × g) and loaded onto a Ni-NTA agarose column (Qiagen) previously equilibrated in French press buffer. The column was washed with wash buffer (50 mM Tris-HCl, pH 7.5, 300 mM KCl, 40 mM imidazole and 5 mM MgCl2) until the flow through exhibited an adsorption at 280 nm (A280) of less than 0.02. The Trx-(His)6-RssB proteins were eluted at room temperature with wash buffer containing 150 mM imidazole. The protein solution was dialysed overnight at room temperature against wash buffer containing no imidazole.

The Trx-(His)6 tag was cleaved off by adding recombinant enterokinase (Novagen) directly to Trx-(His)6-RssB or Trx-(His)6-RssB(D58P) bound to Ni-NTA agarose in a buffer containing 50 mM Tris-HCl, pH 7.5, 200 mM KCl, 2 mM CaCl2 and 5 mM MgCl2. Cleavage was performed overnight at room temperature. Enterokinase was removed with EKapture Agarose (Novagen) according to the directions given by the manufacturer.

Synthesis of acetyl [32P]-phosphate

Acetyl [32P]-phosphate was prepared after a modification of the procedure described by Stadtman (1957). The following were combined in a 16 × 100 Pyrex tube and stirred on ice: 0.19 ml of pyridine, 0.3 ml of 0.33 M K2HPO4 and 0.1 ml of carrier-free [32P]-orthophosphate (specific activity 10 mCi ml−1; Amersham Buchler). Acetic anhydride (22 μl) was added slowly over a period of 3 min and allowed to incubate on ice for an additional 3 min. Then LiOH (4 N) was added to adjust the pH to 7. After an additional 3 min incubation on ice, acetyl [32P]-phosphate was precipitated with 4.5 ml of 100% ethanol (−20°C) and kept on ice for 1 h. The precipitated acetyl [32P]-phosphate was collected by centrifugation, washed twice with 5 ml of cold ethanol and dried over KOH and CaCl2. The dried pellet was resuspended in 180 μl of 50 mM Tris-HCl, pH 7.5, 5% glycerol, 0.1 mM EDTA and 1 mM dithiothreitol. Acetyl [32P]-phosphate was assayed spectrophotometrically according to the procedure of Stadtman (1957) using acetyl phosphate (Sigma) as a standard. This procedure gives a yield of 60–70% with a specific radioactivity of 10–15 mCi mmol−1 acetyl [32P]-phosphate.

In vitro phosphorylation of RssB by acetyl [32P]-phosphate

Proteins were incubated in 50 mM Tris-HCl, pH 7.5, 20 mM MgCl2 and 300 mM KCl with 20 mM acetyl [32P]-phosphate at 30°C. At the times indicated, aliquots were removed, and the reaction was stopped with SDS–PAGE sample buffer (Laemmli, 1970). Samples were immediately subjected to SDS–PAGE. Phosphorylation of the proteins was detected on the dried gels using a phosphorimager (Molecular Dynamics).

β-Galactosidase assay

β-Galactosidase activity was assayed by use of ONPG as a substrate and is reported as μm of ONP min−1 mg−1 of cellular protein (Miller, 1972).

Acknowledgements

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

We thank J. Cronan for strain DC1185. This study was carried out in the laboratories of W. Boos and K. Altendorf, whose support is gratefully acknowledged. Financial support was provided by the Deutsche Forschungsgemeinschaft (Schwerpunkt-Programm ‘Regulatory Networks in Bacteria’, He-1556/5 to R.H.-A., and AL118/18-3 to K.J.) and the Fonds der Chemischen Industrie (to R.H.-A.). M.L. was the recipient of a fellowship (Graduiertenkolleg) from the Deutsche Forschungsgemeinschaft.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  • 1
    Bearson, S.M.D., Benjamin, Jr, W.H., Swords, W.E., Foster, J.W. (1996) Acid shock induction of RpoS is mediated by the mouse virulence gene mviA of Salmonella typhimurium. J Bacteriol 178: 25722579.
  • 2
    Chang, Y.-Y., Wang, A.-Y., Cronan, Jr, J.E. (1994) Expression of Escherichia coli pyruvate oxidase (PoxB) depends on the sigma factor encoded by the rpoS (katF ) gene. Mol Microbiol 11: 10191028.
  • 3
    Dailey, F.E. & Berg, H.C. (1993) Change in direction of flagellar rotation in Escherichia coli mediated by acetate kinase. J Bacteriol 175: 32363239.
  • 4
    Feng, J., Atkinson, M.R., McCleary, W., Stock, J.B., Wanner, B.L., Ninfa, A.J. (1992) Role of phosphorylated metabolic intermediates in the regulation of glutamine synthetase synthesis in Escherichia coli. J Bacteriol 174: 60616070.
  • 5
    Hengge-Aronis, R. (1993) Survival of hunger and stress: the role of rpoS in stationary phase gene regulation in Escherichia coli. Cell 72: 165168.
  • 6
    Hoch, J.A. & Silhavy, T.J. (1995) Two-Component Signal Transduction. Washington, DC: American Society for Microbiology Press.
  • 7
    Innis, M.A., Gelfand, D.H., Sninsky, J.J., White, T.J.E. (1990) PCR Protocols. San Diego, CA: Academic Press.
  • 8
    Kenney, L.J., Bauer, M.D., Silhavy, T.J. (1995) Phosphorylation-dependent conformational changes in OmpR, an osmoregulatory DNA-binding protein of Escherichia coli. Proc Natl Acad Sci USA 92: 88668870.
  • 9
    Kim, S.-K., Wilmes-Riesenberg, M.R., Wanner, B.L. (1996) Involvement of the sensor kinase EnvZ in the in vivo activation of the response-regulator PhoB by acetyl phosphate. Mol Microbiol 22: 135147.
  • 10
    Laemmli, U.K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227: 680685.
  • 11
    Lange, R. & Hengge-Aronis, R. (1991) Identification of a central regulator of stationary-phase gene expression in Escherichia coli. Mol Microbiol 5: 4959.
  • 12
    Lange, R. & Hengge-Aronis, R. (1994) The cellular concentration of the σS subunit of RNA-polymerase in Escherichia coli is controlled at the levels of transcription, translation and protein stability. Genes Dev 8: 16001612.
  • 13
    Loewen, P.C. & Hengge-Aronis, R. (1994) The role of the sigma factor σS (KatF) in bacterial global regulation. Annu Rev Microbiol 48: 5380.
  • 14
    Lukat, G.S., McCleary, W.R., Stock, A.M., Stock, J.B. (1992) Phosphorylation of bacterial response regulator proteins by low molecular weight phospho-donors. Proc Natl Acad Sci USA 89: 718722.
  • 15
    McCleary, W.R. (1996) The activation of PhoB by acetylphosphate. Mol Microbiol 20: 11551163.
  • 16
    McCleary, W.R. & Stock, J.B. (1994) Acetyl phosphate and the activation of two-component response regulators. J Biol Chem 269: 3156731572.
  • 17
    McCleary, W.R., Stock, J.B., Ninfa, A.J. (1993) Is acetyl-phosphate a global signal in Escherichia coli? J Bacteriol 175: 27932798.
  • 18
    Miller, J.H. (1972) Experiments in Molecular Genetics. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press.
  • 19
    Muffler, A., Fischer, D., Altuvia, S., Storz, G., Hengge-Aronis, R. (1996a) The response regulator RssB controls stability of the σS subunit of RNA polymerase in Escherichia coli. EMBO J 15: 13331339.
  • 20
    Muffler, A., Traulsen, D.D., Lange, R., Hengge-Aronis, R. (1996b) Posttranscriptional osmotic regulation of the σS subunit of RNA polymerase in Escherichia coli. J Bacteriol 178: 16071613.
  • 21
    Muffler, A., Barth, M., Marschall, C., Hengge-Aronis, R. (1997) Heat shock regulation of σS turnover: a role for DnaK and relationship between stress responses mediated by σS andσ32 in Escherichia coli. J Bacteriol 179: 445452.
  • 22
    Pratt, L.A. & Silhavy, T.J. (1996) The response regulator, SprE, controls the stability of RpoS. Proc Natl Acad Sci USA 93: 24882492.
  • 23
    Prüß, B.M. & Wolfe, A.J. (1994) Regulation of acetyl phosphate synthesis and degradation, and the control of flagellar expression in Escherichia coli. Mol Microbiol 12: 973984.
  • 24
    Sambrook, J., Fritsch, E.F., Maniatis, T. (1989) Molecular Cloning. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press.
  • 25
    Schröder, I., Wolin, C.D., Cavicchioli, R., Gunsalus, R.P. (1994) Phosphorylation and dephosphorylation of the NarQ, NarX, and NarL proteins of the nitrate-dependent two-component regulatory system of Escherichia coli. J Bacteriol 176: 49854992.
  • 26
    Shin, S. & Park, C. (1995) Modulation of flagellar expression in Escherichia coli by acetyl phosphate and the osmoregulator OmpR. J Bacteriol 177: 46964702.
  • 27
    Silhavy, T.J., Berman, M.L., Enquist, L.W. (1984) Experiments with Gene Fusions. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press.
  • 28
    Stadtman, E.R. (1957) Preparation and assay of acetyl phosphate. Methods Enzymol 3: 228231.
  • 29
    Stock, J.B., Surette, M.G.l., Levit, M., Park, P. (1995) Two-component signal transduction systems: structure–function relationships and mechanisms of catalysis. In Two-Component Signal Transduction. Hoch, J.A., and Silhavy, T.J. (eds). Washington, DC: American Society for Microbiology Press, pp. 2551.
  • 30
    Studier, F.W. & Moffatt, B.A. (1986) Use of bacteriophage T7 RNA polymerase to direct selective high level expression of cloned genes. J Mol Biol 189: 113130.
  • 31
    Takayanagi, Y., Tanaka, K., Takahashi, H. (1994) Structure of the 5′ upstream region and the regulation of the rpoS gene of Escherichia coli. Mol Gen Genet 243: 525531.
  • 32
    Wanner, B.L. (1992) Is cross regulation by phosphorylation of two-component response regulator proteins important in bacteria? J Bacteriol 174: 20532058.
  • 33
    Wanner, B.L. (1995) Signal transduction and cross regulation in the Escherichia coli phosphate regulon by PhoR, CreC, and acetyl phosphate. In Two-Component Signal Transduction. Hoch, J.A., and Silhavy, T.J. (eds). Washington DC: American Society for Microbiology Press, pp. 203221.
  • 34
    Wanner, B.L. & Wilmes-Riesenberg, M.R. (1992) Involvement of phosphotransacetylase, acetate kinase, and acetyl-phosphate synthesis in the control of the phosphate regulon in Escherichia coli. J Bacteriol 174: 21242130.