Logo of aemPermissionsJournals.ASM.orgJournalAEM ArticleJournal InfoAuthorsReviewers
Appl Environ Microbiol. 2011 Aug; 77(15): 5220–5229.
PMCID: PMC3147466

Role of rpoS in the Development of Cell Envelope Resilience and Pressure Resistance in Stationary-Phase Escherichia coli[down-pointing small open triangle]


This work investigated the role of rpoS in the development of increased cell envelope resilience and enhanced pressure resistance in stationary-phase cells of Escherichia coli. Loss of both colony-forming ability and membrane integrity, measured as uptake of propidium iodide (PI), occurred at lower pressures in E. coli BW3709 (rpoS) than in the parental strain (BW2952). The rpoS mutant also released much higher concentrations of protein under pressure than the parent. We propose that RpoS-regulated functions are responsible for the increase in membrane resilience as cells enter stationary phase and that this plays a major role in the development of pressure resistance. Strains from the Keio collection with mutations in two RpoS-regulated genes, cfa (cyclopropane fatty acyl phospholipid synthase) and osmB (outer membrane lipoprotein), were significantly more pressure sensitive and took up more PI than the parent strain, with cfa having the greatest effect. Mutations in the bolA morphogene and other RpoS-regulated lipoprotein genes (osmC, osmE, osmY, and ybaY) had no effect on pressure resistance. The cytoplasmic membranes of the rpoS mutant failed to reseal after pressure treatment, and strains with mutations in osmB and nlpI (new lipoprotein) were also somewhat impaired in the ability to reseal their membranes. The cfa mutant, though pressure sensitive, was unaffected in membrane resealing, implying that the initial transient permeabilization event is critical for loss of viability rather than the failure to reseal. The enhanced pressure sensitivity of polA, recA, and xthA mutants suggested that DNA may be a target of oxidative stress in pressure-treated cells.


Consumer demand for minimally processed, additive-free foods that have a long shelf life has stimulated research into new nonthermal methods of food preservation. High hydrostatic pressure (HHP) can inactivate nonsporing food-borne pathogens and spoilage organisms with minimal effects on food quality and is thus regarded as one of the more promising of the new technologies. The operation of HHP machinery requires relatively low energy consumption, produces little waste, and therefore, has a low impact on the environment (13, 37, 46). The use of HHP in the food industry is gradually increasing and is now being applied to a wide range of food products, including oysters, crabs and other crustaceans, cured hams, guacamole, salsas, fruit juices, desserts, and prepared meals (50).

In order to ensure the safety of pressure-treated foods, it is important not only to have quantitative information on the intrinsic pressure resistance of different microorganisms but also to understand the basis of cellular pressure resistance and how this is affected by conditions during growth, pressure treatment, and recovery. Three main processes are currently believed to contribute to the inactivation of pressure-treated cells. These are protein denaturation, oxidative damage, and loss of membrane integrity (34).

In all bacteria examined so far, pressure resistance is highly dependent on the physiological state of the cells when exposed to pressure. For example, in Escherichia coli, stationary-phase cells are very much more resistant to pressure than those in the exponential phase of growth (8, 44). This is due to the operation of the general stress response, since mutations in the rpoS gene encoding the RpoS sigma factor (σS) lead to a considerable loss of pressure resistance in stationary phase (35, 47). Natural isolates of E. coli O157 vary widely in their resistance to HHP, and this is due largely to heterogeneity in the rpoS genes and consequent differences in RpoS activity (8, 44, 47).

RpoS is involved in the regulation of around 11% of the genes in E. coli that have functions affecting stress resistance, cell morphology, metabolism, and virulence (29, 55). The particular RpoS-regulated genes responsible for increased pressure resistance in stationary phase are not known. It would be reasonable to suppose that some RpoS-regulated functions might have an influence in protecting pressure-treated cells against the supposed main causes of cell death described above: protein denaturation, oxidative stress, or membrane damage. The development of resistance to oxidative damage in stationary-phase cells is well known (38), and a number of genes known to play a protective role are at least partially regulated by RpoS, including katG, katE, dps, gor, sodC, and xthA (55). The possible role of RpoS in protecting cells against pressure-induced protein denaturation is presently unclear. The DnaK chaperone is necessary for the development of resistance to heat and oxidation in stationary-phase cells (48, 49) and is also induced in response to pressure (2). However, although the chaperones DnaK, GroEL, and HtpG are all induced in stationary phase, they are predominantly under the control of RpoH rather than RpoS (23). In nitrogen-starved cells, the expression of a number of heat shock proteins, including DnaK, DnaJ, and HtpG, was reduced in an rpoS mutant, so RpoS may play an indirect role in regulating the expression of the heat shock proteins under these conditions (27). Oxidative stress can cause carbonylation of proteins in stationary-phase E. coli cells (41) and, since the levels of carbonylation increase in rpoS mutants, there may also be an indirect role for RpoS in protecting against this form of protein damage in pressure-treated cells. Trehalose synthesis, which is under RpoS regulation, is involved in the development of thermotolerance in stationary-phase E. coli cells (20). Like other polyols, trehalose protects proteins against thermal denaturation (6) and may possibly protect proteins against denaturation by pressure as well.

In E. coli, the cytoplasmic membranes of stationary-phase cells are much more robust with respect to pressure than membranes of growing cells, presumably reflecting changes in the structure of the cell envelope (10, 36, 43). As cells of E. coli enter stationary phase, they undergo a reduction in cell size and a change in shape from rod to coccobacillus. There are also major changes in the composition of the cell envelope, including an increase in the level of cross-linking in the peptidoglycan, an increase in lipoprotein content, changes in phospholipid composition, and the conversion of monounsaturated fatty acids to the cyclopropane form (21). Some of these changes involve genes that are under RpoS regulation (e.g., cfa and bolA), but whether any of the changes contributes to increased resistance of the cell envelope toward pressure-induced damage is not yet known.

The aims of this study were to investigate the relationship between loss of viability and loss of membrane integrity in strains of E. coli that are isogenic apart from rpoS and to identify RpoS-regulated genes that may contribute to any observed changes in membrane resilience toward pressure. This was accomplished by comparing the responses to pressure of parent and relevant mutant strains from the Keio collection (5). The effects of certain other genes that are upregulated in stationary phase but are not under RpoS control were also examined. For comparative purposes, we have also examined the effects of some other genes reported to affect pressure resistance.


Bacterial strains and growth conditions.

The strains of Escherichia coli used in the study are shown in Table 1. E. coli strains BW2952 and BW3709 were kindly provided by Thomas Ferenci, School of Molecular and Microbial Biosciences, The University of Sydney, Australia. E. coli BW2952 is a derivative of the E. coli K-12 strain MC4100 containing a wild-type rpoS gene and a malG-lacZ transcriptional fusion [genotype, F araD139 Δ(argF-lac)U169 rpsL150 deoC1 relA1 thiA ptsF25 flbB5301 rbsR mal::λplacMu55[var phi](malG::lacZ)]. E. coli BW3709, an rpoS null mutant strain, was constructed by using P1 transduction with P1 cml clr1000 grown on strain ZK1171 to introduce rpoS::Tn10 into E. coli BW2952. These two strains were used for studying the effect of the rpoS gene on pressure resistance and loss of membrane integrity. For comparing genes controlled by rpoS, E. coli BW25113 and its mutants from the Keio collection (5) were tested for resistance to pressure in paired comparisons. These strains were kindly provided by H. Mori, Keio University, Japan.

Table 1.
E. coli strains used in this study

Cultures for use in experiments were produced as follows: a single colony cultivated from frozen stock on tryptone soya agar (TSA) (product code CM 0131; Oxoid, Basingstoke, United Kingdom) was inoculated into tryptone soya broth (TSB) (product no. CM 0129; Oxoid, Basingstoke, United Kingdom) and incubated at 37°C for 6 h. Stationary-phase cultures were prepared by inoculating 0.1% (vol/vol) of the 6-h culture into fresh TSB and incubating under the same conditions for approximately 12 to 15 h. Cells in exponential phase were prepared by transferring 0.1% (vol/vol) of the stationary-phase cells into fresh TSB and incubating with shaking for 2 to 3 h at 37°C until the optical density at 680 nm (OD680) reached 0.2 to 0.3 (Cecil model CE 2020 spectrophotometer; Cecil Instruments Ltd., Cambridge, United Kingdom).

Cloning and sequencing of the rpoS gene.

The wild-type rpoS gene was cloned and used to complement BW3709 in order to confirm the recovery of pressure resistance. The rpoS coding and promoter region was PCR amplified using primers RPOS-1 (5′-AAG GCG ACA CAC TTT TCT ATA T-3′) and RPOS-2 (3′-ATG ACT CAG GGT TCT GGA TTG TGA CCA-5′). The gene fragment was cloned into the EcoRI and BamHI restriction sites of plasmid pSU18 (7) to generate plasmid pSU-rpoS. The sequence of the rpoS gene was confirmed using primers RPOS-1, RPOS-2, and RPOS-3 (5′-TGA TTA CCT GAG TGC CTA CG-3′) by the University of Reading BioCentre sequencing facility.

Detection of rpoS mutants.

The transformed cells were tested for rpoS status by iodine staining, catalase assay, and growth on LB-tetracycline medium (40). Escherichia coli strains were streaked onto Luria agar plates and incubated at 37°C for 24 h. Plates were left at 4°C for 24 h and flooded with concentrated iodine to detect the level of glycogen. Catalase activity was also examined by adding 3% H2O2 to colonies and observing the production of oxygen bubbles.

Pressure treatment.

Cultures were harvested by centrifugation at 2,750 × g for 10 min at 5°C. The cell pellets were resuspended in phosphate-buffered saline (PBS). Cell suspensions (2 ml) were placed into sterile high-density polyethylene plastic pouches of 65-μm film thickness (Seward Ltd., Worthing, West Sussex, United Kingdom) which were heat sealed after removal of air and stored on ice. The viable counts of non-pressure-treated cells were determined for each pouch before it was sealed. A model S-FL-850-9-W high-hydrostatic-pressure unit (Stansted Fluid Power, Stansted, United Kingdom) with a chamber volume of 300 ml was used for pressure treatments with monopropylene glycol/water (30%, vol/vol) as the pressure-transmitting fluid. The come-up rate was approximately 330 MPa/min. After treatment, the pressure was released in two steps. In the first step, the pressure decreases to 30 MPa in about 15 s. The total decompression takes about 35 s. Cell suspensions were pressure treated at 100 to 600 MPa for 8 min. The temperature in the transmitting fluid during the pressure treatment was measured with a thermocouple. The maximum temperature during the compression did not exceed 30°C.

Viable counts.

After pressurization, cell suspensions were serially diluted in maximum recovery diluent (MRD) (product no. CM 0733; Oxoid, Basingstoke, United Kingdom) and plated onto TSA containing 0.3% yeast extract and 0.1% (wt/vol) filter-sterilized sodium pyruvate. Colonies formed on the agar plates were counted after 24 to 48 h of incubation at 37°C. All experiments were performed at least three times.

Determination of membrane damage by PI uptake.

Samples of cell suspensions in PBS with an OD680 of 0.2 were mixed with propidium iodide (PI) solution to a final concentration of 2.9 μM. For evaluation of PI uptake after pressure treatment, cells were incubated with PI for 10 min and then centrifuged at 10,000 × g for 10 min at 4°C. When PI was present during pressure treatment, the cells were centrifuged and washed immediately after decompression. The cell pellet was washed twice with PBS at the same volume. The cell pellet was resuspended in PBS, and fluorescence was measured in a fluorimeter (model LS-5B; PerkinElmer, Waltham, MA) at an excitation wavelength of 495 nm and an emission wavelength of 615 nm. Fluorescence values from untreated cells were subtracted from those of pressure-treated cells, and the data were normalized against the OD680 of the cell suspensions (PI value = fluorescence value during or after treatment − fluorescence value of nontreated cells/OD680).

Measurement of protein release from pressure-treated cells.

Cell suspensions from pressure-treated and untreated samples were centrifuged at 2,750 × g for 10 min at 5°C. The supernatants were filtered through 0.2-μm Minisart high-flow syringe filters (Sartorius Mechatronics Limited, Epsom, Surrey, United Kingdom), and the protein concentrations in the supernatants were determined by using the Bradford assay (product no. B 6916; Sigma-Aldrich Company Ltd., Gillingham, Dorset, United Kingdom) with bovine serum albumin as the standard. Whole-cell protein was measured using Folin Ciocalteu phenol reagent (product no. F9252; Sigma-Aldrich Company Ltd. Dorset, United Kingdom). Whole-cell protein was first solubilized by adding 0.5 ml of 1 M NaOH to 0.5 ml of cell suspension and boiling for 5 min.

Statistical analysis.

Survival is represented as the log surviving fraction, this being log(Nt/N0), where Nt is the viable count (CFU/ml) after pressure treatment and N0 the initial viable count. All experiments were performed independently three times. When comparing the pressure resistance of parent and mutant strains, the parent was always tested in the same pressure run as the mutant strain being tested. The results are presented as average values with error bars representing the standard deviations. The pressure resistance of a mutant strain was expressed as the differential log survival ratio (DLSR), this being the log surviving fraction of the mutant minus the log surviving fraction of the parent (35). A negative value indicates that the mutant is more pressure sensitive than the parent. Significant differences between means were tested using one-way analysis of variance with the GenStat statistical analysis package (VSN International).


Effect of rpoS on pressure resistance of E. coli.

Before examining the possible effect of rpoS status on the susceptibility of the cell membrane to high-pressure treatment, it was necessary to characterize the effect of rpoS on cell survival after pressure treatment of the test strain. Accordingly, E. coli BW2952 (wild type) and BW3709 (rpoS mutant) were exposed to a range of pressures for 8 min in exponential and stationary phase. In stationary-phase cells, E. coli wild type had a much greater resistance to high pressure than the rpoS mutant (Fig. 1). There was no loss of viability in the wild type after pressure treatment between 100 and 400 MPa in stationary phase, whereas in the rpoS mutant, the viable numbers deceased by 2.0 log when subjected to high pressure at 300 MPa and by more than 6 log at 400 MPa. At 500 and 600 MPa, the reductions of the viable counts of the wild type were 0.5 and 2.1 log10 units, respectively, while the rpoS mutant was completely inactivated at 600 MPa. In exponential-phase cells, both strains were far more susceptible to pressure, and there were negligible differences in pressure resistance between them. This indicates that the rpoS status has a major impact on pressure resistance in stationary phase but not in exponential phase.

Fig. 1.
Loss of viability of E. coli parent BW2952 ([diamond with plus], [diamond]) and rpoS mutant W3709 (□, ■) in exponential phase ([open star], ■) and stationary phase ([diamond], ■). Cells were pressure treated at different pressure levels ...

Relationship between pressure resistance and growth phase.

The observed relationship, described above, between pressure resistance and growth phase was investigated in more detail. Samples were taken every 2 h during growth to examine the loss of viability after a fixed-pressure challenge of 300 MPa for 8 min. To avoid any effect of cell concentration on pressure resistance, the cultures were concentrated or diluted to give 108 cells/ml before pressure treatment. In relative terms, both strains underwent similar changes in pressure resistance with phase of growth (Fig. 2). Survival after the pressure challenge decreased rapidly (>6 logs reduction) when cells entered exponential phase but then increased rapidly at the onset of stationary phase (6 h) before increasing more slowly between 6 and 10 h to reach levels of resistance similar to those at 0 h. Although a similar pattern of growth phase-dependent change in resistance occurred in both strains, higher levels of resistance were reached in stationary phase in the parent than in the rpoS mutant, such that there was a difference in survival of approximately 1.0 log after the pressure challenge.

Fig. 2.
Changes in pressure resistance of E. coli parent BW2952 ([diamond], [division on times]) and rpoS mutant BW3709 (■, □) during growth. E. coli strains were subcultured into fresh TSB, and samples taken every 2 h to measure growth ([diamond with plus], □, ...

Determination of protein release after pressurization.

To test whether the loss of pressure resistance in the rpoS mutant might be associated with changes in cell membrane properties, the levels of release of cell protein under pressure were compared for the wild type and the rpoS mutant. The results depicted in Fig. 3 show that the rpoS mutant released a much larger amount of protein (up to 3-fold) than the wild type. Interestingly, the highest concentration of released protein from the rpoS mutant (176 μg/ml) occurred at 300 MPa. At higher pressure levels, the protein concentration gradually decreased (154, 118, and 98 μg/ml at 400, 500, and 600 MPa, respectively). This behavior was not observed in the wild type, in which the protein concentration (47 μg/ml) reached a maximum at 200 MPa and did not change when cells were exposed to higher pressures.

Fig. 3.
Determination of pressure-induced protein release from E. coli parent BW2952 (open columns) and rpoS mutant BW3709 (solid columns). Cell suspensions in stationary phase were pressure treated for 8 min at different pressures. Protein concentration in the ...

Determination of membrane damage by PI uptake.

As a further test of membrane damage, pressure-treated cells were stained with the fluorescent dye PI, which is excluded from cells with intact membranes. Figure 4 shows the PI uptake of E. coli strains during and after pressurization. The onset pressure for PI uptake was higher in the wild-type strain (200 to 300 MPa) than in the rpoS mutant (100 to 200 MPa). The degree of staining in the rpoS mutant was also of much greater extent than in the wild type both during and after pressurization. Although the wild type took up PI during pressurization, there was almost no uptake of PI after pressurization. The results indicate that the membrane of the wild type has a leaky cell membrane under pressurization and can largely restore its membrane after pressure treatment. In contrast, the rpoS mutant exhibited very little ability to reseal and, thus, had suffered permanent loss of membrane integrity.

Fig. 4.
Determination of pressure-induced loss of membrane integrity in E. coli parent BW2952 ([division on times], [diamond]) and rpoS mutant BW3709 (□, ■). Propidium iodide uptake in stationary-phase cells was examined during ([division on times], □) and ...

Recovery of pressure resistance and membrane resilience after complementation with rpoS.

A complementation test was performed to confirm that the observed loss of pressure resistance in the rpoS mutant was indeed due to the rpoS mutation. The rpoS mutant was thus complemented with plasmid pSU-rpoS (rpoS+), and the rpoS status of the resulting Cmr transformants was confirmed by measuring catalase activity and glycogen levels. The pressure resistance and membrane resilience of the complemented rpoS mutant were then compared with these properties in the plasmid-free wild type and the rpoS mutant, as well as with the vector-bearing rpoS strain (Fig. 5 and and6).6). Complementation of the rpoS mutant with pSU-rpoS restored the pressure resistance of the mutant to a level similar to that of the wild type (Fig. 5) and also restored the ability to reseal its cytoplasmic membrane (Fig. 6); however, the pSU18 vector had no notable effect on the pressure sensitivity of the rpoS mutant, which had a log surviving fraction of −6.5, similar to that of the plasmid-free mutant (−7.0) (Fig. 5). These results show that the provision of a wild-type version of the rpoS gene restores both the pressure resistance and membrane resilience of the rpoS mutant to wild-type levels.

Fig. 5.
Effect of rpoS complementation on pressure resistance of the rpoS mutant. Loss of viability was determined in stationary-phase cells of parent (BW2952), rpoS mutant (BW3709), rpoS-complemented strain BW3709 (BW3709 rpoS), and vector-containing strain ...
Fig. 6.
Effect of rpoS-complementation on loss of membrane integrity of the rpoS mutant. Loss of membrane integrity was measured as uptake of propidium iodide in the parent (BW2952), rpoS mutant (BW3709), and complemented strain BW3709 (BW3709rpoS) during and ...

Effect of RpoS-regulated and other genes on pressure resistance.

The effects on pressure resistance of several RpoS-regulated genes affecting cell envelope composition or structure, as well as certain genes involved in stress responses, were studied by comparing survival after treatment at 300 MPa and 600 MPa in E. coli BW25113 (wild type) and otherwise isogenic mutants from the Keio collection. At 300 MPa, only the rpoS and nlpI mutants were significantly (P < 0.05) more sensitive to high pressure than the wild type, with differential log survival ratios of 3.1 and 1.0, respectively. The differences in loss of viability of the mutants compared with that in the parent strain after pressurization at 600 MPa are shown in Fig. 7. Strains with mutations in cfa, rpoS, trxB, otsA, recA, nlpI, osmB, otsB, trxA cspA, polA, and xthA were all significantly more pressure sensitive (P < 0.05) than the parent strain, with log surviving fractions of −6.74, −6.63, −6.47, −5.67, −5.48, −5.44, −5.25, −5.23, −5.21, −5.15, −4.93, and −4.90, respectively, compared with −3.50 for the parent. The mutations having the biggest effect were cfa, rpoS, and trxB. Four mutations (soxS, uspB, mltB, and sodA) caused a mean increase in sensitivity that was virtually identical to the effects of mutations in polA and xthA, but because of greater variation between replicates, the effect was not statistically significant. Table 2 lists the genes in functional categories with their effects on pressure resistance.

Fig. 7.
Effects on pressure resistance of mutations in selected RpoS-regulated genes and other genes with potential effects on pressure resistance. Cells in stationary phase were pressure treated at 600 MPa for 8 min. The differential log survival fraction is ...
Table 2.
Genes tested for their effect on pressure resistance

Effects of genes controlled by RpoS on loss of membrane integrity.

Mutations that had an effect on pressure resistance were further investigated to determine whether they also affected resistance to membrane disruption. The levels of uptake of PI by E. coli BW25113 and pressure-sensitive mutant strains during and after pressure treatment are shown in Fig. 8. Four mutant strains, with mutations in cfa (cyclopropane fatty acid synthetase), nlpI (new lipoprotein), osmB (osmotically inducible protein), and otsAB (trehalose synthesis), showed significantly (P < 0.05) higher PI uptake during pressure treatment than the parent. The highest uptake of PI was seen in the cfa mutant, which also had the greatest loss of viability during pressurization. The intensity of PI staining depends on both loss of membrane integrity and amount of dye bound to nucleic acid, so fluorescence intensity may not be directly related to the extent of membrane damage if strains contain different amounts of nucleic acid. However, tests of the ability of cells to reseal membranes after decompression are based on measurements of PI uptake during and after pressure treatment of samples taken from the same cell suspension, which will have same nucleic acid content. Interestingly, the PI uptake in the cfa mutant after decompression was not significantly different from that of the parent, indicating that membrane resealing was not impaired in this mutant. PI uptake after decompression in the nlpI, osmB, and otsA mutants was significantly higher than in the parent, suggesting that the products of these genes may facilitate membrane resealing.

Fig. 8.
Effects of mutations on loss of membrane integrity in pressure-treated cells. Mutants showing a significant loss of pressure resistance were examined for loss of membrane integrity by propidium iodide uptake during and after pressurization. Strains showing ...


Role of rpoS in the development of cell envelope resilience toward pressure.

This work aimed to establish whether the large increase in pressure resistance that occurs when E. coli cells enter stationary phase can be attributed in part to RpoS-regulated changes in the resilience of the membrane toward pressure damage. Loss of viability and membrane integrity after pressure treatment were therefore compared in the rpoS+ and rpoS strains BW2952 and BW3709, respectively (40). BW2952 is known to possess high levels of RpoS (28), and as expected, differences between the parent (BW2952) and the rpoS mutant (BW3709) in cell survival after pressure treatment were particularly large and, hence, convenient for comparing corresponding differences in loss of membrane integrity. Loss of viability and loss of membrane integrity, indicated by uptake of PI, both occurred at lower pressures in the rpoS mutant than in the parent, consistent with the idea that RpoS-regulated functions bring about an increase in membrane resilience as cells enter stationary phase, resulting in a major enhancement of pressure resistance.

The rpoS mutant and the wild type underwent similar fluctuations in pressure resistance during the growth cycle, but the increase in resistance in stationary phase was much smaller in the rpoS mutant than in the parent strain. Other regulatory elements must also affect stationary-phase resistance, but these evidently have a much smaller effect than RpoS because the maximum level of pressure resistance in the rpoS mutant, measured as survival after the pressure challenge at 300 MPa, was 10-fold less than that in the parent (Fig. 2). The rpoS mutant also released much higher concentrations of protein under pressure than the parent. Measurement of total protein released does not distinguish between proteins originating from the cytoplasm, periplasm, or the membranes themselves but does provide an indication of gross loss of membrane integrity that supports the conclusion obtained from uptake of propidium iodide. The interesting pattern of increasing and then decreasing protein release with increasing pressure has been reported in other strains of E. coli and has been attributed to the denaturation of cellular proteins at high pressure that prevents their passing through the peptidoglycan to the outside (30).

Genes associated with morphology and cell envelope composition.

In an attempt to identify genes responsible for the stationary-phase increase in membrane stability and pressure resistance, we examined the effects on pressure resistance of a number of RpoS-regulated genes known to be associated with cell morphology and envelope composition. Strains with mutations in two of these, cfa and osmB, were significantly more pressure-sensitive than the parent (differential log survival ratios, −3.24 and −1.76, respectively), with cfa having the bigger effect. The cfa gene encodes cyclopropane fatty acyl phospholipid synthase, which converts unsaturated fatty acids to their cyclopropane form as cells enter into stationary phase (18). Membrane cyclopropane fatty acid is a major factor in acid resistance in E. coli and also contributes to resistance to freezing and freeze drying (9, 11, 17, 39). The cyclopropane fatty acids appear to stack more compactly in the membrane than unsaturated fatty acids, resulting in a stiffer membrane (39, 51). This structural alteration also appears to have an effect on the resilience of the membrane toward pressure. Casadei et al. (10) previously observed a qualitative relationship between pressure resistance and membrane cyclopropane fatty acid content in stationary-phase cells grown at different temperatures.

The osmB gene encodes an outer membrane lipoprotein that is induced in stationary phase or by hyperosmotic conditions (26). Curiously, an osmB mutant grew better than the parent in medium of high osmolarity (25). It was speculated that the OsmB lipoprotein may cross-link the outer membrane and peptidoglycan in a way that inhibits growth but aids survival under hyperosmotic conditions. Our observation that the osmB mutant was somewhat sensitive to pressure would be consistent with a role for the OsmB lipoprotein in cell survival. Mutations in the other RpoS-regulated lipoprotein genes, osmC, osmE, osmY, ybaY, and mltB, had no effect on pressure resistance, nor did mutations in yhiU (multidrug efflux) or ybgA (hypothetical inner membrane protein); however, in agreement with Malone et al. (35), we found that a mutant with inactivation of nlpI, encoding an outer membrane lipoprotein which is upregulated in response to pressure but is not under RpoS control, was sensitive to pressure. The conversion of rod-shaped exponential-phase cells of E. coli to the smaller and more rounded cells seen in stationary-phase cultures occurs under the influence of the bolA morphogene, which is itself regulated by RpoS. The bolA mutant was no more pressure sensitive than the parent strain, although examination by light microscopy revealed little difference in cell shape between the wild type and the bolA mutant under the rich-medium growth conditions used here (data not shown). As E. coli enters stationary phase, the enzyme cardiolipin synthase catalyzes the conversion of phosphatidylglycerol to cardiolipin. Mutations in cls increase membrane fluidity (45) and contribute to increased survival after repeated cycles of freezing and thawing (52); however, the cls mutation had no effect on pressure resistance.

Membrane resealing.

Strains with mutations in nlpI and osmB were somewhat impaired in their ability to reseal their membranes after pressure treatment as judged by the extent of PI uptake by cells after decompression. These two lipoproteins may play a role in maintaining membrane structure or, as suggested by Malone et al. (35) for NlpI, may assist in repair of damage to the cell envelope. Interestingly, the membrane of the cfa mutant was not impaired in its ability to reseal after pressure treatment, implying that the initial transient permeabilization event is critical for loss of viability rather than the failure to reseal. This is in agreement with the findings of Klotz et al. (30) working with strains that had natural differences in resistance to high pressure.

It is interesting that two lipoproteins located in the outer membrane, one under RpoS control (OsmB) and the other inducible by pressure (NlpI), had an effect on resistance to pressure, since previous work has shown that damage to the outer membrane, at least as measured by increased uptake of N-phenyl-1-naphthylamine (NPN) (16) or increased sensitivity to sodium deoxycholate (12), is not lethal to E. coli and can be repaired during incubation in growth medium after decompression. Clearly the roles of outer membrane damage and periplasmic stress responses in loss of viability merit further investigation.

Stress responses.

Cold shock and pressure share some common features in that both cause a decrease in membrane fluidity and an inhibition of translation, and cold shock proteins are induced in response to pressure (56). The cold-shock protein CspA is an RNA chaperone believed to be essential for efficient translation at low temperatures (24). Malone et al. (35) showed that cspA is upregulated 3.9-fold in response to pressure, but a cspA mutant was not significantly sensitive to pressure. In this work, the cspA mutant did show enhanced pressure sensitivity, possibly reflecting differences between strains or conditions. Taken together, our results and those of Malone et al. (35) suggest that the cold-shock response may contribute to enhanced pressure resistance. UspB (universal stress protein B) is believed to be a membrane protein under RpoS regulation that confers resistance to ethanol and freeze-thaw stress (15, 52). The mean differential survival ratio of the uspB mutant, although similar to that of cspA, was not statistically significant.


Inactivation of otsA and otsB, encoding trehalose-6-phosphate synthase and trehalose-6-phosphate phosphatase, respectively, caused a significant decrease in pressure resistance, in agreement with the results of Malone et al. (35). Trehalose is known to protect proteins against thermal denaturation but may also stabilize membranes (14). Both otsA and otsB are regulated by RpoS, so this would be consistent with a further role for RpoS-regulated functions in stabilizing membranes against pressure damage. osmB, otsA, and otsB are all expressed in response to osmotic pressure, suggesting a possible link between osmotic and pressure stresses.

Oxidative stress.

We also examined the effects of mutations in genes associated with resistance to oxidative stress. The two main RpoS-regulated defenses against oxidative stress are Dps, a DNA-binding protein that sequesters iron and protects DNA from damage, and KatE, one of two hydroperoxidases found in E. coli. KatE is induced in stationary phase and is also regulated by OxyR. Surprisingly, neither of these mutants was pressure sensitive in E. coli BW25113 strains, which differs from the results of Malone et al. (35) (dps) and Aertsen et al. (3) (katE and oxyR). In agreement with Malone et al. (35), however, we found that trxA (thioredoxin I) and trxB (thioredoxin reductase) mutants are more pressure sensitive than the parent. The sodA and soxS mutants showed differential log survival ratios of −1.39 and −1.53 which were marginally nonsignificant but indicative of a role in pressure resistance. This is in agreement with the results of Aertsen et al. (3), who found soxS and sodAB mutants to be more sensitive to pressure than parent strains. We found no effect of sodC or sodB on pressure resistance, suggesting that the manganese-containing superoxide dismutase encoded by sodA that is induced under conditions of oxidative stress may be more important for pressure resistance than the constitutive iron-containing enzyme (sodB) or the periplasmic copper-zinc enzyme (sodC).

Sublethally injured cells in general tend to be hypersensitive to oxidative stress (32, 54), but the degree of sensitivity is highly dependent on the physiological state of the cells and the conditions during recovery. This can lead to variability in survival after stress and may account for certain differences between the results of this work and previously reported studies. Nevertheless, these results confirm the importance of oxidative stress as a mechanism leading to the inactivation of cells by high pressure but do not support a special role for the two RpoS-regulated protective functions, Dps and KatE, in preventing oxidative damage and death in pressure-treated cells. This does not rule out a role for RpoS in protecting against pressure-induced oxidative stress, but its role may vary between strains and may be influenced by conditions during exposure to pressure and recovery.

DNA repair.

Mutants with mutations in polA, recA, and xthA, encoding DNA repair enzymes, all showed significantly enhanced sensitivity to pressure. However, of these, only xthA, encoding exonuclease III, is regulated by RpoS. One interesting feature of these DNA repair mutants is that they are all hypersensitive to hydrogen peroxide (19). Hydrogen peroxide is believed to damage DNA by the generation of hydroxyl free radicals in the presence of traces of iron via the Fenton reaction (22). The sensitivity of these mutants to pressure would thus be consistent with a mechanism wherein damage to DNA arises in pressure-treated cells by the generation of reactive oxygen species (ROS). However, other mechanisms are possible; for example, xthA mutants are defective in the expression of heat shock proteins (42), and this could have an effect on protection against protein denaturation. Nevertheless, the production of ROS under pressure has been demonstrated by Aertsen et al. (3), using a leaderless alkaline phosphatase as probe, and Malone et al. (35) suggested a plausible mechanism based on the release of free iron from iron-sulfur proteins and the production of ROS from metabolically generated hydrogen peroxide by a Fenton-type reaction.

An alternative explanation is that pressure causes sublethal damage of a type that sensitized the DNA repair mutants to traces of peroxide present in agar recovery medium. This effect has been seen previously with heat-injured recA, polA, and lexA mutants of E. coli (33). These were apparently more sensitive to mild heat than wild-type strains when plated on tryptone soya agar, but the sensitivity was reversed by the addition of catalase to the recovery medium. Low levels of peroxide (15 μM) were found to be present in the medium, and similar levels of authentic hydrogen peroxide caused death of heat-injured cells. In the experiments described here, the addition of catalase or sodium pyruvate to the agar medium to decompose hydrogen peroxide did not reverse the sensitivity of polA or recA mutants (data not shown), suggesting that endogenously generated ROS may have been responsible for cell death in these strains. The present results suggest that DNA might be a target of oxidative damage in pressure-treated cells but that in the parent strain, any damage thus produced can be repaired.

A different mechanism of DNA damage was indicated by the work of Aertsen et al. (4) and Aertsen and Michiels (1), who discovered a novel pressure-induced SOS response in E. coli in which double-strand breaks in DNA were believed to be the inducing signal. The SOS response was brought about by activation of the Mrr restriction endonuclease. Inactivation of Mrr slightly improved the survival of E. coli after high-pressure treatment (1, 4). These studies show that DNA damage may arise indirectly by different mechanisms following pressure treatment, but such damage appears to play a relatively minor role in cell death.

In E. coli, approximately 11% of RpoS-regulated genes are involved in protecting cells against starvation, acid, heat, oxidative, and osmotic stresses (29, 55). Although RpoS has a very large effect on pressure resistance, the identities of the particular RpoS-regulated functions responsible for enhanced resistance were not previously known. Some aspects of bacterial physiology where RpoS may have an effect were examined here: cell envelope resilience, protection against osmotic and oxidative stresses, and DNA repair. Of these, cell envelope resilience appeared to be particularly important, and two RpoS-regulated genes, cfa and osmB, were found to play a significant role in the development of this resilience.


We are grateful to the Royal Thai Government for funding a sponsorship for Duangkamol Charoenwong.

We thank Thomas Ferenci, University of Sydney, Australia, and H. Mori, Keio University, Japan, for providing E. coli strains. We are very grateful to Hanna Mourand-Agha for advice and help with molecular biology techniques.


[down-pointing small open triangle]Published ahead of print on 24 June 2011.


1. Aertsen A., Michiels C. W. 2005. Mrr instigates the SOS response after high pressure stress in Escherichia coli. Mol. Microbiol. 58:1381–1391 [PubMed]
2. Aertsen A., et al. 2004. Heat shock protein-mediated resistance to high hydrostatic pressure in Escherichia coli. Appl. Environ. Microbiol. 70:2660–2666 [PMC free article] [PubMed]
3. Aertsen A., De Spiegeleer P., Vanoirbeek K., Lavilla M., Michiels C. W. 2005. Induction of oxidative stress by high hydrostatic pressure in Escherichia coli. Appl. Environ. Microbiol. 71:2226–2231 [PMC free article] [PubMed]
4. Aertsen A., Van Houdt R., Vanoirbeek K., Michiels C. 2004. An SOS response induced by high pressure in Escherichia coli. J. Bacteriol. 186:6133–6141 [PMC free article] [PubMed]
5. Baba T., et al. 2006. Construction of Escherichia coli K-12 in-frame, single-gene knockout mutants: the Keio collection. Mol. Syst. Biol. 2:1–11 [PMC free article] [PubMed]
6. Back J. F., Oakenfull D., Smith M. B. 1979. Increased thermal stability of proteins in the presence of sugars and polyols. Biochemistry 18:5191–5196 [PubMed]
7. Bartolomé B., Jubete Y., Martínez E., de la Cruz F. 1991. Construction and properties of a family of pACYC184-derived cloning vectors compatible with pBR322 and its derivatives. Gene 102:75–78 [PubMed]
8. Benito A., Ventoura G., Casadei M., Robinson T., Mackey B. M. 1999. Variation in resistance of natural isolates of Escherichia coli O157 to high hydrostatic pressure, mild heat, and other stresses. Appl. Environ. Microbiol. 65:1564–1569 [PMC free article] [PubMed]
9. Brown J. L., Ross T., McMeekin T., Nichols P. D. 1997. Acid habituation of Escherichia coli and the potential role of cyclopropane fatty acids in low pH tolerance. Int. J. Food Microbiol. 37:163–173 [PubMed]
10. Casadei M. A., Mañas P., Niven G. W., Needs E., Mackey B. M. 2002. The role of membrane fluidity in pressure resistance of Escherichia coli 8164. Appl. Environ. Microbiol. 68:5965–5972 [PMC free article] [PubMed]
11. Chang Y. Y., Cronan J. E. 1999. Membrane cyclopropane fatty acid content is a major factor in acid resistance of Escherichia coli. Mol. Microbiol. 33:249–259 [PubMed]
12. Chilton P., Isaacs N. S., Mañas P., Mackey B. M. 2001. Metabolic requirements for the repair of pressure-induced damage to the outer and cytoplasmic membranes of Escherichia coli. Int. J. Food Microbiol. 71:101–104 [PubMed]
13. Considine K. M., Kelly A. L., Fitzgerald G. F., Hill C., Sleator R. D. 2008. High-pressure processing-effects on microbial food safety and food quality. FEMS Microbiol. Lett. 281:1–9 [PubMed]
14. Crowe J. H., Crow L. M., Chapman D. 1984. Preservation of membranes in anhydrobiotic organisms: the role of trehalose. Science 223:701–703 [PubMed]
15. Farewell A., Kvint K., Nystrom T. 1998. uspB, a new σS-regulated gene in Escherichia coli which is required for stationary-phase resistance to ethanol. J. Bacteriol. 180:6140–6147 [PMC free article] [PubMed]
16. Gänzle M. G., Vogel R. F. 2001. On-line fluorescence determination of pressure-mediated outer membrane damage in Escherichia coli. Syst. Appl. Microbiol. 24:477–485 [PubMed]
17. Grogan D. W., Cronan J. E. 1986. Characterization of Escherichia coli mutants completely defective in synthesis of cyclopropane fatty acids. J. Bacteriol. 166:872–877 [PMC free article] [PubMed]
18. Grogan D. W., Cronan J. E. 1997. Cyclopropane ring formation in membrane lipids of bacteria. Microbiol. Mol. Biol. Rev. 61:429–441 [PMC free article] [PubMed]
19. Hagensee M. E., Moses R. E. 1989. Multiple pathways for repair of hydrogen peroxide-induced DNA damage in Escherichia coli. J. Bacteriol. 171:991–995 [PMC free article] [PubMed]
20. Hengge-Aronis R., Klein W., Lange R., Rimmele M., Boos W. 1991. Trehalose synthesis genes are controlled by the putative sigma factor encoded by rpoS and are involved in stationary-phase thermotolerance in Escherichia coli. J. Bacteriol. 173:7918–7924 [PMC free article] [PubMed]
21. Huisman G. W., Siegele D. A., Zambrano M. M., Kolter R. 1996. Morphological and physiological changes during stationary phase, p. 1672–1682 In Neidhardt F. C., editor. (ed.), Escherichia coli and Salmonella, cellular and molecular biology. ASM Press, Washington, DC
22. Imlay J. A., Chin S. M., Linn S. 1988. Toxic DNA damage by hydrogen peroxide through the Fenton reaction in vivo and in vitro. Science 240:640–642 [PubMed]
23. Jenkins D. E., Auger E. A., Matin A. 1991. Role of RpoH, a heat shock regulator protein, in Escherichia coli carbon starvation protein synthesis and survival. J. Bacteriol. 173:1992–1996 [PMC free article] [PubMed]
24. Jiang W., You Y., Inouye M. 1997. CspA the major cold shock protein of Escherichia coli is an RNA chaperone. J. Biol. Chem. 272:196–202 [PubMed]
25. Jung J. U., Gutierrez C., Villarejo M. 1989. Sequence of an osmotically inducible lipoprotein gene. J. Bacteriol. 171:511–520 [PMC free article] [PubMed]
26. Jung J. U., Gutierrez C., Martin F., Ardourel M., Villarejo M. 1990. Transcription of osmB, a gene encoding an Escherichia coli lipoprotein, is regulated by dual signals—osmotic-stress and stationary phase. J. Biol. Chem. 265:10574–10581 [PubMed]
27. Kabir M. S., et al. 2004. Effects of mutations in the rpoS gene on cell viability and global gene expression under nitrogen starvation in Escherichia coli. Microbiology 150:2543–2553 [PubMed]
28. King T., Ishihama A., Kori A., Ferenci T. 2004. A regulatory trade-off as a source of strain variation in the species Escherichia coli. J. Bacteriol. 186:5614–5620 [PMC free article] [PubMed]
29. Klauck E., Typas A., Hengge A. 2007. The σS subunit of RNA polymerase as a signal integrator and network master regulator in the general stress response in Escherichia coli. Sci. Prog. 90:103–127 [PubMed]
30. Klotz B., Mañas P., Mackey B. M. 2010. The relationship between membrane damage, release of protein and loss of viability in Escherichia coli exposed to high hydrostatic pressure. Int. J. Food Microbiol. 137:214–220 [PubMed]
31. Lacour S., Landini P. 2004. σS-dependent gene expression at the onset of stationary phase in Escherichia coli: function of σS-dependent genes and identification of their promoter sequences. J. Bacteriol. 186:7186–7195 [PMC free article] [PubMed]
32. Mackey B. M. 2000. Injured bacteria, p. 315–341 In Lund B. M., Baird-Parker A. C., Gould G. W., editors. (ed.), The microbiological safety and quality of food. Aspen Publishers, Inc., Gaithersburg, MD
33. Mackey B. M., Seymour D. A. 1987. The effect of catalase on recovery of heat-injured DNA-repair mutants of Escherichia coli J. Gen. Microbiol. 133:1601–1610 [PubMed]
34. Mackey B. M., Mañas P. 2008. Inactivation of Escherichia coli by high pressure, p. 53–85 In Michiels C., Aertsen A., Bartlett D., Yayanos Y., editors. (ed.), High pressure microbiology. ASM Press, Washington, DC
35. Malone A. S., Chung Y.-K., Yousef A. E. 2006. Genes of Escherichia coli O157:H7 that are involved in high-pressure resistance. Appl. Environ. Microbiol. 72:2661–2671 [PMC free article] [PubMed]
36. Mañas P., Mackey B. M. 2004. Morphological and physiological changes induced by high hydrostatic pressure in exponential- and stationary-phase cells of Escherichia coli: relationship with cell death. Appl. Environ. Microbiol. 70:1545–1554 [PMC free article] [PubMed]
37. Mañas P., Pagan R. 2005. A review: microbial inactivation by new technologies of food preservation. J. Appl. Microbiol. 98:1387–1399 [PubMed]
38. 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. 173:4188–4194 [PMC free article] [PubMed]
39. Munoz-Rojas J., et al. 2006. Involvement of cyclopropane fatty acids in the response of Pseudomonas putida KT2440 to freeze-drying. Appl. Environ. Microbiol. 72:472–477 [PMC free article] [PubMed]
40. Notley-McRobb L., King T., Ferenci T. 2002. rpoS mutations and loss of general stress resistance in Escherichia coli populations as a consequence of conflict between competing stress responses. J. Bacteriol. 184:806–811 [PMC free article] [PubMed]
41. Nystrom T. 2004. Stationary phase physiology. Annu. Rev. Microbiol. 58:161–181 [PubMed]
42. Paek K. H., Walker G. C. 1986. Defect in expression of heat-shock proteins at high temperature in xthA mutants. J. Bacteriol. 165:763–770 [PMC free article] [PubMed]
43. Pagán R., Mackey B. M. 2000. The relationship between membrane damage and cell death in pressure-treated Escherichia coli cells: differences between exponential and stationary phase cells and variation among strains. Appl. Environ. Microbiol. 66:2829–2834 [PMC free article] [PubMed]
44. Patterson M. F., Quinn M., Simpson R., Gilmour A. 1995. Sensitivity of vegetative pathogens to high hydrostatic pressure treatment in phosphate-buffered saline and foods. J. Food Prot. 58:524–529
45. Pluschke G., Hirota Y., Overath P. 1978. Function of phospholipids in Escherichia coli. Characterization of a mutant deficient in cardiolipin synthesis. J. Biol. Chem. 25:5048–5055 [PubMed]
46. Rastogi N. K., Raghavarao K. S. M. S., Balasubramaniam V. M., Niranjan K., Knorr D. 2007. Opportunities and challenges in high pressure processing of foods. Crit. Rev. Food Sci. Nutr. 47:69–112 [PubMed]
47. Robey M., et al. 2001. Variation in resistance to high hydrostatic pressure and rpoS heterogeneity in natural isolates of Escherichia coli 0157:H7. Appl. Environ. Microbiol. 67:4901–4907 [PMC free article] [PubMed]
48. Rockabrand D., et al. 1998. Roles of DnaK and RpoS in starvation-induced thermotolerance of Escherichia coli. J. Bacteriol. 180:846–854 [PMC free article] [PubMed]
49. Rockabrand D., Arthur T., Korinek G., Livers K., Blum P. 1995. An essential role for the Escherichia coli DnaK protein in starvation-induced thermotolerance, H2O2 resistance, and reductive division. J. Bacteriol. 177:3695–3703 [PMC free article] [PubMed]
50. San Martín M. F., Barbosa-Cánovas G. V., Swanson B. G. 2002. Food processing by high hydrostatic pressure. Crit. Rev. Food Sci. Nutr. 42:627–645 [PubMed]
51. Shabala L. 2008. Cyclopropane fatty acids improve survival in acidified minimal media by reducing membrane permeability to H+ and enhancing ability to extrude H+. Res. Microbiol. 159:458–461 [PubMed]
52. Sleight S. C., Orlic C., Schneider D., Lenski R. E. 2008. Genetic basis of evolutionary adaptation by Escherichia coli to stressful cycles of freezing, thawing and growth. Genetics 180:431–443 [PMC free article] [PubMed]
53. Somalinga R. V., Vijayakumar V., Kirchhof M. G., Patten C. L., Schellhorn H. E. 2004. RpoS-regulated genes of Escherichia coli identified by random lacZ fusion mutagenesis. J. Bacteriol. 186:8499–8507 [PMC free article] [PubMed]
54. Stephens P. J., Druggan P., Nebe-von Caron G. 2000. Stressed Salmonella are exposed to reactive oxygen species from two independent sources during recovery in conventional culture media. Int. J. Food. Microbiol. 60:269–285 [PubMed]
55. Weber H., Polen T., Heuveling J., Wendisch V. F., Hengge R. 2005. Genome-wide analysis of the general stress response network in Escherichia coli: σS-dependent genes, promoters, and sigma factor selectivity. J. Bacteriol. 187:1591–1603 [PMC free article] [PubMed]
56. Welch T. J., Farewell A. F., Neidhardt F. C., Bartlett D. H. 1993. Stress response of Escherichia coli to elevated hydrostatic pressure. J. Bacteriol. 175:7170–7177 [PMC free article] [PubMed]

Articles from Applied and Environmental Microbiology are provided here courtesy of American Society for Microbiology (ASM)
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...