The role of proton motive force in expression of the Staphylococcus aureus cid and lrg operons

Authors

  • Toni G. Patton,

    1. Department of Microbiology, Molecular Biology and Biochemistry, University of Idaho, Moscow, ID 83844-3052, USA.
    2. Department of Pathology and Microbiology, University of Nebraska Medical Center, Omaha, NE 68198-6245, USA.
    Search for more papers by this author
  • Soo-Jin Yang,

    1. Department of Microbiology, Molecular Biology and Biochemistry, University of Idaho, Moscow, ID 83844-3052, USA.
    2. Department of Pathology and Microbiology, University of Nebraska Medical Center, Omaha, NE 68198-6245, USA.
    Search for more papers by this author
  • Kenneth W. Bayles

    Corresponding author
    1. Department of Microbiology, Molecular Biology and Biochemistry, University of Idaho, Moscow, ID 83844-3052, USA.
    2. Department of Pathology and Microbiology, University of Nebraska Medical Center, Omaha, NE 68198-6245, USA.
    Search for more papers by this author

Summary

The Staphylococcus aureus cidABC and lrgAB operons have been shown to play a key role in the regulation of murein hydrolase activity and cell death in a manner thought to be analogous to bacteriophage-encoded holins and anti-holins respectively. Because of these functions, it has been proposed that the regulation of these operons is tightly controlled and responsive to key metabolic signals. The current study revealed the presence of two overlapping regulatory pathways controlling cidABC and lrgAB expression, one dependent on acetic acid and the other dependent on proton motive force (PMF). The latter pathway was analysed using agents that affect various aspects of the PMF. Gramicidin and carbonyl cyanide m-chlorophenylhydrazone (CCCP), antimicrobial agents that dissipate the ΔpH and membrane potential (ΔΨ), both enhanced lrgAB expression. Restoration of the PMF by incubation of the bacteria in the presence of glucose restored lrgAB expression back to the uninduced state. In addition, valinomycin, which specifically collapses the ΔΨ, also induced lrgAB expression. In contrast, nigericin, which dissipates the ΔpH component of the PMF, was found to have a minimal effect on ΔΨ and lrgAB transcription. Finally, the ΔΨ-inducible expression of lrgAB was shown to be dependent on the previously characterized LytSR two-component regulatory system that is involved in the regulation of autolysis. The results of this study support a model in which the LytSR regulatory system responds to a collapse in ΔΨ by inducing the transcription of the lrgAB operon.

Introduction

Previously in our laboratory, the Staphylococcus aureus LytSR two-component regulatory system was identified and shown to affect murein hydrolase activity and tolerance to penicillin (Brunskill and Bayles, 1996a; Groicher et al., 2000). Although the lytS and lytR gene products share strong sequence similarities with sensor and response regulator proteins, respectively (Brunskill and Bayles, 1996a), the signal(s) to which LytS responds has not been reported. At least one target of LytSR regulation is a dicistronic operon, designated lrgAB, located immediately downstream of the lytR gene (Brunskill and Bayles, 1996b). Although a lrgAB mutant derivative of the laboratory strain, RN6390, displayed increased extracellular murein hydrolase activity and decreased penicillin tolerance, neither the lrgA nor the lrgB gene products were predicted to encode a murein hydrolase (Groicher et al., 2000). Other S. aureus genes homologous to lrgA and lrgB were subsequently identified and designated cidA and cidB (Rice et al., 2003). A cidA mutant derivative of RN6390, as well as the clinical isolate, UAMS-1, displayed phenotypes opposite to that of the lrgAB mutant, exhibiting decreased murein hydrolase activity and increased tolerance to antibiotics (Rice et al., 2003; 2005). Based on these results, as well as the structural similarities of CidA and LrgA to the holin family of proteins, it was proposed that cidA and lrgA encode a holin and an anti-holin, respectively (Groicher et al., 2000; Rice et al., 2003), whose functions are to regulate murein hydrolase activity and, ultimately, cell death (Bayles, 2003; Rice and Bayles, 2003).

A recent analysis of cid transcription revealed the presence of a third gene in this operon, designated cidC, encoding a pyruvate oxidase (Rice et al., 2004; Patton et al., 2005). This analysis also revealed that two transcripts are produced by the cid operon, one spanning all three cid genes and the other spanning cidB and cidC (Rice et al., 2004). The cidBC transcript is positively regulated by sigma factor B (Rice et al., 2004), whereas cidABC transcription is enhanced by acetic acid (Rice et al., 2005) and this enhanced expression is dependent on the recently described LysR-type transcriptional regulator (LTTR) encoded by the cidR gene (Yang et al., 2005). Furthermore, recent characterization of the cidC gene has revealed that it is involved in acetate metabolism in stationary-phase cells grown in the presence of excess glucose (Patton et al., 2005).

Based on several studies demonstrating that weak acids can act as uncoupling agents (Baronofsky et al., 1984; Eklund, 1985; Herrero et al., 1985; Luli and Strohl, 1990; Cherrington et al., 1991; Axe and Bailey, 1995), we hypothesized that acetic acid induction of cidABC and lrgAB expression is not a general function of specific weak acids but, rather, occurs via a mechanism involving the proton motive force (PMF). To test this hypothesis, we analysed cidABC and lrgAB expression in cells exposed to both antimicrobial agents that affect PMF as well as various weak acids. Although these studies indicate that acetic acid-inducible expression of these operons is independent of PMF, the expression of lrgAB transcription was found to be induced by a second regulatory pathway involving changes in the membrane potential (Δψ). Furthermore, this second regulatory pathway was shown to be independent of CidR but dependent on the LytSR two-component regulatory system. Thus, these results demonstrate the presence of two independent, but overlapping, regulatory pathways that control cidABC and lrgAB expression, one involving changes in Δψ and the other that is mediated specifically by the products of carbohydrate metabolism.

Results

The effect of weak acids on cidABC and lrgAB transcription

Previous studies in our laboratory have revealed that acetic acid strongly induces expression of the S. aureus cidABC and lrgAB operons involved in the regulation of murein hydrolase activity (Rice et al., 2005; Yang et al., 2005). To determine whether this effect is specific to acetic acid or simply a non-specific response to weak acids, Northern blot analyses were performed on RNA isolated from S. aureus cells exposed to a variety of weak acids. Surprisingly, this analysis revealed that lrgAB and cidABC displayed different patterns of transcription in response to weak acids. As shown in Fig. 1A, acetic acid and lactic acid (lanes 3 and 7) were the only weak acids that stimulated cidABC expression. By comparison, lrgAB expression was increased to some degree in response to most of the weak acids tested, with the exception of ascorbic acid (Fig. 1B). The pH of the cultures after 2 h of incubation with the weak acids ranged from 5.3 to 5.8, with the exception of the boric acid culture which was pH 7.2. Therefore, the differences in transcript levels observed in Fig. 1 are unlikely a result of differences in pH of the cultures. Although the mechanism by which these weak acids increase cidABC and lrgAB expression is unclear, the differential response of cidABC and lrgAB expression to these weak acids suggests the presence of two distinct regulatory mechanisms for controlling cidABC and lrgAB expression.

Figure 1.

Effect of weak acids on lrgAB and cidABC transcription. S. aureus UAMS-1 cultures were incubated at 37°C for 2 h before the addition of boric (lane 2), acetic (lane 3), sorbic (lane 4), ascorbic (lane 5), pyruvic (lane 6) and lactic (lane 7) acids. The cultures were then incubated for an additional 2 h, and total cellular RNA was isolated from each culture. RNA samples (5 µg each) were separated in a 1% (w/v) agarose-formaldehyde gel, transferred to a nylon membrane and hybridized with DNA probes specific for either cidA (A) or lrgA (B). The sizes of each transcript were determined by comparison with an RNA ladder (Invitrogen) run on the same gel. The corresponding ethidium bromide-stained gel is shown underneath each blot.

The effect of PMF on lrg and cid transcription

Given that weak acids have been shown to act as uncoupling agents (Baronofsky et al., 1984; Eklund, 1985; Herrero et al., 1985; Luli and Strohl, 1990; Cherrington et al., 1991; Axe and Bailey, 1995), it is possible that the effect of acetic acid on expression of cidABC and lrgAB was due to its effect on PMF. To test this, we examined lrgAB and cidABC expression in the presence of several antimicrobial agents known to dissipate various aspects of the PMF, including carbonyl cyanide m-chlorophenylhydrazone (CCCP), gramicidin, nigericin, valinomycin, and dicyclohexylcarbodiimide (DCCD). CCCP is an uncoupler that can diffuse into the membrane in its protonated or non-protonated form. The protonated form of CCCP diffuses across the membrane in response to the pH gradient and releases a proton inside the cell. The non-protonated form is then driven back to the external surface of the membrane by the electrical potential where it is reprotonated. This cycle continues until both the ΔΨ and ΔpH are completely dissipated (Harold, 1970; Nicholls, 1982). Gramicidin is an antimicrobial peptide that also dissipates both the ΔΨ and ΔpH components of the PMF by forming transmembrane channels that increase the membrane's permeability to cations (Decoursey, 2003). Nigericin is a lipophilic molecule that dissipates the ΔpH with an electrically neutral exchange of potassium for protons across the cytoplasmic membrane (Harold, 1970). Valinomycin is a lipophilic potassium carrier which specifically increases the potassium permeability of the membrane and can cause depolarization or hyperpolarization depending on the internal and external potassium concentrations (Nicholls, 1982; Shapiro, 1994). If the external potassium concentration is lower than the internal concentration the cells will be hyperpolarized. In contrast, depolarization occurs if the external potassium concentration is higher than the internal potassium concentration (Nicholls, 1982; Shapiro, 1994). Finally, DCCD is an antimicrobial agent that inhibits the F0 sector of ATPase and, thus, reduces the ability of the cell to generate and maintain the proton gradient (Harold and Kakinuma, 1985). Northern blot analyses revealed that none of these antimicrobial agents increased cidABC transcription when cultures were exposed to these compounds for up to 2 h (data not shown), indicating that the induction of cidABC expression by acetic acid was independent of the PMF. In contrast, lrgAB transcription was greatly enhanced after a 15 min exposure of the cells to gramicidin and CCCP (Fig. 2). Nigericin appeared to have only a slight stimulatory effect on lrgAB transcription, while DCCD had no effect (Fig. 2). These results indicate that gramicidin and CCCP are powerful inducers of lrgAB transcription.

Figure 2.

PMF induction of lrgAB transcription. S. aureus UAMS-1 was incubated at 37°C for 4 h before the addition of various compounds as indicated. The cultures were then incubated for an additional 15 min, and total cellular RNA was isolated before and after addition of the antimicrobial agents. RNA samples (5 µg each) were separated in a 1% (w/v) agarose-formaldehyde gel, transferred to a nylon membrane and hybridized with a lrgA-specific probe. The sizes of each transcript were determined by comparison with an RNA ladder (Invitrogen) run on the same gel. The corresponding ethidium bromide stained gel is shown underneath the blot.

To assess the effects of the above compounds on ΔΨ, we utilized flow cytometry for measuring ΔΨ as described previously (Novo et al., 1999). This method has been used to determine ΔΨ of S. aureus with the cyanine dye diethyloxacarbocyanine [DIOC2(3)] (Novo et al., 1999). As shown in Fig. 3A, a 15 min treatment of the cells with CCCP and gramicidin, agents that strongly induced lrgAB expression (Fig. 2), almost completely dissipated ΔΨ of the population as indicated by the reduction of the mean fluorescence ratio (MFR) to 20% and 10% of the control cells respectively. Nigericin, which only slightly increased lrgAB expression (Fig. 2), decreased the MFR to levels that were 52% of the control cells (Fig. 3B). DCCD did not effect lrgAB expression (Fig. 2) and had no impact on the MFR of the population (Fig. 3B), indicating that it had no immediate effect on ΔΨ. However, extended incubation (2 h) of the cells in the presence of DCCD eventually decreased the ΔΨ of the population and also caused increased lrgAB expression (data not shown). Interestingly, the addition of acetic acid to the medium, which strongly induces lrgAB expression (Fig. 2), did not affect the ΔΨ (Fig. 3C). As controls, dimethyl sulphoxide (DMSO; the solvent used for CCCP and gramicidin) and 95% ethanol (the solvent used for nigericin and DCCD) were also examined and, as expected, were shown to have no effect on either ΔΨ (Fig. 3A) or lrgAB expression (Fig. 2). Taken together, these data indicate that the dissipation of the PMF leads to the induction of lrgAB transcription. Furthermore, the induction of cidABC and lrgAB expression by acetic acid appeared to be unrelated to changes in ΔΨ.

Figure 3.

The effect of different antimicrobial agents on ΔΨ.
A. S. aureus UAMS-1 cultures were incubated for 4 h and then exposed to gramicidin (black, MFR = 52.81), CCCP (blue, MFR = 102.29) or DMSO (red fill, MFR = 508.05).
B. S. aureus UAMS-1 cultures were exposed to nigericin (black, MFR = 315.11), DCCD (blue, MFR = 626.12) or ethanol (red fill, MFR = 609.71).
C. S. aureus UAMS-1 cultures were exposed to acetic acid (black, MFR = 583.73) or NZY (red fill, MFR = 588.95).
Cultures before and after addition of the antimicrobial agents were washed, resuspended and exposed to 30 µM DIOC2(3), and were analysed by flow cytometry. The data presented are representative of three experiments and the mean is the average fluorescence intensity of 5000 cells analysed.

To further demonstrate the importance of PMF on the induction of lrgAB expression, we re-established the ΔΨ by resuspending CCCP-treated cells in buffer containing glucose and assessed lrgAB transcription by Northern blot analysis. As seen in Fig. 4, an increase in lrgAB expression occurred 15 min after addition of CCCP, compared with control cells in which no CCCP was added (Fig. 4A, compare lanes 1 and 2). After the cultures were harvested and resuspended in buffer containing 1% (w/v) glucose, the transcription of lrgAB was reduced back to the original expression levels (Fig. 4A, lane 4). In the control cells that were resuspended in buffer without glucose, lrgAB expression was only slightly reduced in comparison with the cells that were resuspended in 1% glucose buffer (Fig. 4A, compare lanes 3 and 4). As shown in Fig. 4B, flow cytometry analyses revealed that the ΔΨ in the CCCP-treated cells (in this experiment the MFR of these cells was 19% of the control cells) could be almost completely restored when the CCCP was removed and the cells were resuspended in phosphate buffer containing 1% glucose (the MFR was 94% of the control cells). In contrast, the ΔΨ remained mostly dissipated in the CCCP-treated cells (the MFR was 31% of the control cells) that were washed and resuspended in buffer without glucose (Fig. 4B). These data further support the conclusion that the PMF is an important regulator of lrgAB transcription.

Figure 4.

Regeneration of membrane potential. Northern blot of lrgAB expression (A) and flow cytometry analysis of UAMS-1 exposed to CCCP (B). S. aureus UAMS-1 cultures were incubated for 4 h and then exposed to CCCP for an additional 15 min. The cells were centrifuged and the pellets were resuspended in buffer with or without 1% glucose. Total cellular RNA was isolated from cells before exposure to CCCP (lane 1), after CCCP exposure (lane 2) and after 30 min in the buffer without (lane 3) or with (lane 4) 1% glucose. RNA samples (5 µg each) were separated in a 1% (w/v) agarose-formaldehyde gel, transferred to a nylon membrane and hybridized with a lrgA-specific probe. The sizes of each transcript were determined by comparison with an RNA ladder (Invitrogen) run on the same gel. The corresponding ethidium bromide stained gel is shown underneath the blot.
B. Flow cytometry analysis of ΔΨ with the cultures before exposure to CCCP (red fill, MFR = 481.32), after CCCP exposure (black, MFR = 91.99) and after 30 min in the buffer without (blue, MFR = 150.95) or with 1% glucose (grey, MFR = 451.37). The same samples for RNA isolation were collected for flow cytometry, washed, resuspended, and were analysed by flow cytometry. The data presented are representative of three experiments and the mean is the average fluorescence ratio calculated from 5000 cells analysed.

To determine whether changes in ΔΨ alone are responsible for the increased lrgAB transcription we treated the cells with valinomycin and assessed lrgAB expression (Fig. 5). In this analysis UAMS-1 cultures were incubated for 2 h before the addition of valinomycin and/or KCl and then incubated for an additional 2 h for optimal hyperpolarization and depolarization. As shown in Fig. 5B, treatment with valinomycin and KCl decreased the fluorescence intensity of the population, indicating a decrease in the ΔΨ. Although the cells were not uniformly depolarized as observed when comparing the peak width of this population with the control cells, it is clear that the majority of the cells exhibited some level of depolarization. Treatment of the cells with valinomycin without added KCl caused an obvious hyperpolarization of the cells, as indicated by an increase in MFR to levels that were 33% higher than the control cells (Fig. 5B). The Northern blot in Fig. 5A demonstrates that depolarization of the cells with valinomycin and KCl also caused an increase in lrgAB transcription. In contrast, hyperpolarization of the population with valinomycin had no detectable effect on lrgAB transcript levels (Fig. 5A). These data suggest that lrgAB expression is induced in response to a decrease in the ΔΨ component of the PMF.

Figure 5.

Effect of valinomycin on lrgAB transcription and ΔΨ.
A. S. aureus UAMS-1 cultures were incubated for 2 h and then exposed to valinomycin and/or KCl for an additional 2 h. Total cellular RNA was isolated from cells exposed to DMSO (lane 1), DMSO and KCl (lane 2), valinomycin (lane 3) and valinomycin and KCl (lane 4). RNA samples (5 µg each) were separated in a 1% (w/v) agarose-formaldehyde gel, transferred to a nylon membrane and hybridized with a lrgA-specific probe as described in the legend to Fig. 4.
B. Flow cytometry analysis of cells exposed to DMSO (red fill, MFR = 705.76, similar to untreated cells), DMSO and KCl (black, MFR = 592.24), valinomycin (blue, MFR = 941.13) and valinomycin and KCl (green, MFR = 540.01). The same samples for RNA isolation were collected for flow cytometry, washed, resuspended, and were analysed by flow cytometry. The data presented are representative of three experiments and the mean is the average fluorescence intensity of 5000 cells analysed.

Given that the PMF is required for the generation of adenosine 5′-triphosphate (ATP), it is possible that the effects of the depolarizing agents on lrgAB expression are a function of the secondary effects on ATP levels within the cell rather than being a direct effect of a decrease in the ΔΨ. Thus, intracellular ATP levels were monitored after treatment of the cells with gramicidin, CCCP and DCCD for 15 min. As shown in Table 1, treatment of the cells with gramicidin or CCCP had a minimal impact on the intracellular ATP levels compared with the control cells. In contrast, treatment of the cells with the ATPase inhibitor, DCCD, resulted in a drop in ATP to nearly undetectable levels. Thus, these data indicate that the induction of lrgAB expression is independent of changes in the intracellular ATP levels.

Table 1. Staphylococcus aureus ATP assays.
TreatmentnM ATP before treatmentStandard error (n = 4)nM ATP after treatmentStandard error (n = 4)
  • a. 

    Statistically significant (P = 0.029) compared with the DMSO-treated control (Mann–Whitney Rank Sum test, SigmaStat v3.1).

DMSO113.4342.28103.2158.56
Gramicidin168.5851.37220.8311.93
CCCP146.7778.7297.4056.28
DCCD125.4823.630.74a0.74

PMF induction is lytSR dependent

The cidR gene has recently been shown to be required for the acetic acid induction of the cidABC (Yang et al., 2005). More recently, results from our laboratory have shown that cidR is also involved in the acetic acid induction of the lrg operon (data not shown). Given that our data have demonstrated that induction of lrgAB transcription by acetic acid is independent of the PMF, we wanted to determine whether cidR was required for the induction of lrgAB by agents that affect the PMF. As shown in Fig. 6A, the induction of lrgAB transcription by gramicidin and CCCP in a cidR mutant was comparable to that of the parental strain, indicating that cidR is not involved in the PMF induction of lrgAB transcription. In contrast, the lytSR-encoded two-component regulatory system, which was previously shown to regulate lrgAB transcription and autolysis (Brunskill and Bayles, 1996a,b), was required for the PMF-inducible expression of lrgAB. Although we were unable to generate lytS or lytSR mutations in UAMS-1, we were able to make a lytS mutation in the S. aureus laboratory strain, RN6390. As shown in Fig. 6B, lrgAB transcription was induced by gramicidin or CCCP in RN6390, but was uninducible in its lytS mutant derivative, KB333. These data suggest that the LytSR two-component regulatory system functions to monitor changes in the PMF associated with the cytoplasmic membrane.

Figure 6.

Roles of cidR and lytS in ΔΨ-inducible lrgAB expression.
A. S. aureus UAMS-1 (lane 1) and its cidR mutant derivative (KB1090, lane 2) were grown for 4 h and then treated with either gramicidin or CCCP.
B. S. aureus RN6390 (lane 1) and its lytS mutant derivative (KB333, lane 2) were incubated for 4 h and then treated with either gramicidin or CCCP. All strains were incubated for an additional 15 min, harvested by centrifugation, and total cellular RNA was isolated. RNA samples (5 µg each) were separated in a 1% (w/v) agarose-formaldehyde gel and analysed by Northern blotting with a lrgA-specific probe as described in the legend to Fig. 4.

Discussion

In this study it was hypothesized that acetic acid induction of cidABC and lrgAB was due to its ability to affect the PMF. In general, high concentrations of acetic acid and other weak acids have been shown to act as uncoupling agents by shuttling protons across the membrane and therefore dissipating the ΔpH (Rottenberg, 1979; Cherrington et al., 1991). Acetic acid was previously shown to collapse the ΔpH and have little effect on ΔΨ in Clostridium thermoaceticum (Baronofsky et al., 1984). Analysis by Axe and Bailey (1995) revealed that acetate can permeate the membrane of Escherichia coli at comparable rates in the dissociated and undissociated forms, supporting the model that acetic acid and other weak acids can act as uncoupling agents. The observation that nigericin, an antimicrobial agent that dissipates the ΔpH by an electrical neutral exchange of potassium and protons across the cytoplasmic membrane (Harold, 1970) did not have a major impact on cidABC or lrgAB expression suggests that this pathway does not respond to a change in ΔpH. Although nigericin caused a slight increase in lrgAB expression (Fig. 2) this was probably a result of the minor effects on ΔΨ that this compound induced (Fig. 3), similar to what has been observed in Bacillus subtilis (Jolliffe et al., 1981). Thus, these results suggest that changes in the ΔpH component of the PMF are probably not responsible for the acetic acid-induced expression of cidABC and lrgAB.

Although the acetic acid induction of cid and lrg expression is likely to be PMF independent, we have identified in this study a second regulatory pathway controlling lrgAB expression that involves the ΔΨ component of the PMF. The existence of this pathway is supported by several observations. First, the PMF dissipating agents, CCCP and gramicidin, strongly induce lrgAB transcription (Fig. 2) but have no effect on the expression of the cid operon (data not shown). Second, the effect of CCCP on lrgAB transcription could be reversed by repolarizing the cells in the presence of glucose (Fig. 4). Third, dissipation of the ΔΨ component of the PMF with valinomycin (Fig. 5) induced lrgAB transcription, whereas increasing the ΔΨ had no effect. Interestingly, the magnitude of the decrease in ΔΨ resulting from exposure to the different antimicrobial agents (Figs 2 and 4) roughly corresponded to the levels of lrgAB transcription induced by these agents (Figs 3 and 4), suggesting that lrgAB expression may be induced to different extents depending on the magnitude of the ΔΨ that is associated with the cytoplasmic membrane. Furthermore, studies performed to assess the downstream effects of ΔΨ disruption indicated that changes in ATP levels are not responsible for lrgAB induction (Table 1). Treatment of the cells for 15 min with gramicidin or CCCP (agents that induce lrgAB transcription) did not alter ATP levels significantly, while treatment with DCCD (which did not effect lrgAB transcription) reduced ATP levels (Table 1). Thus, based on the data presented in this study, we propose that a decrease in the ΔΨ is the primary signal leading to the induction of lrgAB transcription.

The molecular mechanism by which lrgAB expression responds to changes in membrane potential is currently unknown, although the data presented in this report (Fig. 6B) suggest the involvement of the previously characterized LytSR two-component regulatory system (Brunskill and Bayles, 1996a,b; Groicher et al., 2000). A hydropathy analysis of the LytS amino acid sequence revealed that it contains six potential transmembrane domains (Brunskill and Bayles, 1996a), unlike many sensor proteins, which contain only two (Stock et al., 1989). These multiple transmembrane domains could be ideally suited to sense a membrane-associated signal such as ΔΨ. Current studies are focused on investigating this hypothesis. The data presented here also indicate that the recently characterized LysR-type regulator (LTTR), designated CidR (Yang et al., 2005), is not involved in the ΔΨ-dependent induction of lrgAB transcription (Fig. 6A).

Based on the results of the current study, we hypothesize the existence of two regulatory pathways that control cidABC/lrgAB expression (Fig. 7) and, subsequently, murein hydrolase activity, cell death and lysis. The first involves the induction of cidABC and lrgAB by acetic acid, or a by-product of acetic acid catabolism. When S. aureus cultures are grown in the presence of excess glucose, acetate is generated and accumulates in the culture supernatant. Acetic acid diffuses back into the cell where it, or a by-product of acetic acid catabolism, interacts with and activates CidR. Activated CidR then binds to the cidABC and lrgAB promoter regions and enhances the transcription of these operons. The second regulatory pathway is postulated to involve a decrease in ΔΨ, which causes the autophosphorylation of LytS. LytS phosphorylates LytR, which then binds to the lrgAB promoter region and induces lrgAB transcription. Although the details of these two pathways are currently being investigated, we speculate that they are required for assessing the overall health of the bacterial cell, and ultimately in making the commitment to cell death.

Figure 7.

A model for the ΔΨ- and acetic acid-dependent regulation of cidABC and lrgAB transcription. First, a decrease in ΔΨ is sensed by LytS, resulting in its autophosphorylation. LytS then phosphorylates LytR, activating and allowing it to bind to the lrgAB promoter region where it increases lrgAB transcription. Second, when the S. aureus cultures are grown in the presence of excess glucose, the accumulation of acetic acid leads to the activation of CidR. Activated CidR binds to the cidABC and lrgAB promoter regions and increases transcription of these operons.

Experimental procedures

Bacterial strains, growth conditions and DNA manipulations

The bacterial strains used in this study are listed in Table 2. S. aureus cultures were maintained on tryptic soy agar (TSA; Difco Laboratories, Detroit, MI) plates containing erythromycin (Em; 2 µg ml−1) or tetracycline (Tc; 5 µg ml−1) where appropriate. Overnight cultures were grown in filter-sterilized NZY broth [3% (w/v) N-Z Amine A (Sigma Chemical, St Louis, MO), 1% (w/v) yeast extract (Fisher Scientific, Fair Lawn, NJ) adjusted to pH 7.5]. Unless otherwise stated, cultures were incubated at 37°C with shaking at 250 r.p.m. in volumes that did not exceed 10% of the flask volume. S. aureus genomic DNA was isolated using a method previously described (Dyer and Iandolo, 1983). Plasmid DNA was purified using the Wizard®Plus SV DNA purification kit from Promega (Madison, WI). Restriction endonucleases used in this study were purchased from Invitrogen Life Technologies (Carlsbad, CA). Bacteriophage-mediated transduction and transformation of S. aureus was performed as previously described (Shafer and Iandolo, 1979; Kraemer and Iandolo, 1990; Schenk and Laddaga, 1992).

Table 2. Bacterial strains and plasmids.
Strain or plasmidDescriptionReference or source
S. aureus strains
 RN4220Highly transformable strain; restriction-deficient Kreiswirth et al. (1983)
 UAMS-1Clinical isolate Gillaspy et al. (1995)
 KB1090UAMS-1 cidR::Tet; Tcr Yang et al. (2005)
 RN6390Parental laboratory strain Peng et al. (1988)
 KB333RN6390 lytS::Em; EmrThis study
Plasmids
 pDG647Source of erythromycin cassette; Emr or Apr Guerout-Fleury et al. (1995)
 pCL52.2Temperature-sensitive shuttle vector, Tcr and Spr Sau et al. (1997)
 pTP200Derivative of pCL52.2 containing the Em cassette flanked by lytS fragments; Tcr and EmrThis study

Allele replacement of the lytS gene

A lytS mutation was generated in RN6390 using an allele replacement strategy. A 558 bp DNA fragment spanning a region 3′ to lytS was amplified using polymerase chain reaction (PCR) with the forward primer, lytS-cla (5′-TCCACATTTTTTCTTCAAATCGATTAACACGATTTCAGC-3′) and the reverse primer, lytS-pst (5′-CCAAAAAGTCTGCA GGCTCGATGTCGATTCAAATTGTAATCG-3′), incorporating ClaI and PstI restriction sites. This fragment was then ligated into the ClaI and PstI sites of pDG647 (Guerout-Fleury et al., 1995) downstream of the Em cassette. Next, a 542 bp 5′ lytS fragment was amplified using the forward primer, lytSR-Eco (5′-CCCGAATTCTGCAACGGGACAATTGTTAG-3′) and the reverse primer, lytSR-Bam (5′-CCCGGATCCCAACGTGC TTTCCATGTACG-3′), incorporating EcoRI and BamHI restriction sites. This was subsequently ligated upstream of the Em cassette into the EcoRI and BamHI sites of pDG647 containing the 3′lytS fragment. The 1.3 kb Em cassette flanked by lytS sequences was liberated by digestion with EcoRI and PstI and then ligated into the plasmid, pCL52.2 (Sau et al., 1997). This plasmid, designated pTP200, was then transformed into S. aureus strain RN4220 by electroporation, spread onto TSA plates containing Em and incubated at 37°C overnight. Colonies were transferred to TSA plates containing Em and incubated at 45°C to select for cells in which the plasmid had integrated into the chromosome via homologous recombination. To promote a second recombination event, a single colony was inoculated into TSB without antibiotic and grown at 30°C for 5 days, with 1:1000 dilutions into fresh antibiotic free media each day. After the fifth day, dilutions of the culture were spread on TSA plates containing Em to yield isolated colonies, which were subsequently screened for EmR and TcS PCR and Southern blot analyses confirmed the lytS deletion in RN4220 (data not shown). This mutation was then transferred to RN6390 by bacteriophage φ11-mediated transduction to generate strain KB333.

RNA isolation

Overnight S. aureus cultures were inoculated to an OD600 of 0.1 into 20 ml of NZY broth and incubated at 37°C with shaking at 250 r.p.m. In one experiment UAMS-1 cultures were incubated for 2 h and then supplemented with 30 mM of acetic, boric, sorbic, ascorbic, lactic or pyruvic acids. The cultures were then incubated for an additional 2 h at 37°C with shaking at 250 r.p.m and RNA was isolated as indicated below. In a second experiment UAMS-1 cultures were incubated for 2 h and subsequently supplemented with 300 mM KCl, 15 µM valinomycin and/or 250 µl of DMSO (as the control). These cultures were then incubated for an additional 2 h at 37°C with shaking at 250 r.p.m and RNA was isolated as indicated below. In a third experiment, the cultures were inoculated as above and allowed to grow for 4 h before the addition of gramicidin (25 µg ml−1; mixture of the A, S and D forms, Sigma Chemical), CCCP (100 µM), nigericin (25 µM), DCCD (200 µM) or acetic acid (30 mM). The cultures were incubated for an additional 15 min at 37°C with shaking at 250 r.p.m. before harvesting the RNA. As controls, cultures treated with equal volumes of the solvents used for these compounds (e.g. DMSO and ethanol) were also included. Finally, in a fourth experiment, the UAMS-1, KB1090 and KB333 cultures were inoculated as above and allowed to grow for 4 h before the addition of gramicidin (25 µg ml−1) or CCCP (100 µM). The cultures were incubated for an additional 15 min at 37°C with shaking at 250 r.p.m. before harvesting the RNA.

RNA was isolated from the treated S. aureus cultures using an RNeasy Mini RNA Purification Kit (Qiagen, Valencia, CA) with modifications of the manufacturer's instructions. Briefly, 3 ml of the bacterial cultures were harvested at 3000 r.p.m. in a Sorvall Legend table-top centrifuge (Newtown, CT). The resulting bacterial pellets were resuspended in 900 µl of RLT buffer and the suspensions were transferred to FASTPREP tubes containing glass beads (lysing matrix B, Q.Biogene, Carlsbad, CA). The cells were lysed in a FASTPREP FP120 instrument (Bio 101 Thermo Savant, Vista, CA) at a speed of 6.0 for 23 s. Cell supernatants were harvested by centrifugation at 13 000 r.p.m. in a Biofuge pico (Heraeus Instruments, South Plainfield, NJ) at 4°C. The resulting supernatants were transferred to clean eppendorf tubes, followed by the addition of 500 µl of room temperature 100% ethanol. This mixture was placed in a mini-spin column provided by Qiagen (Valencia, CA), and centrifuged at room temperature for 20 s at 13 000 r.p.m. The remainder of the RNA isolation procedure, including DNase treatment was performed according to the manufacturer's instructions (Qiagen, Valencia, CA).

Northern blot analysis

RNA samples (5 µg) were separated by electrophoresis in a 1% (w/v) agarose gel containing 0.66 M formaldehyde in morpholine propane sulphonic (MOPS) acid running buffer (20 mM MOPS, 10 mM sodium acetate, 2 mM EDTA, pH 7.0). This was followed by capillary transfer of the RNA to a nylon membrane (Micron Separations, Westboro, MA) in 20× SSC buffer (0.3 M Na3-citrate, 3.0 M NaCl, pH 7.0). The transferred RNA was fixed to the nylon membrane using the ‘autocrosslink’ setting in a UV Stratalinker 1800 (Stratagene, Cedar Creek, TX). Hybridization and processing of the blots were carried out using the DIG system (Roche Applied Science, Indianapolis, IN) according to the manufacturer's recommendations for Northern blot analyses. Transcript sizes were estimated by comparison with a RNA molecular mass ladder (Sigma). Probes were synthesized using the PCR-based DIG probe Synthesis Kit (Roche) following the manufacturer's instructions. The primer pair, CidA1-F and CidA1-R (Rice et al., 2005), was used to generate a DIG-labelled probe corresponding to the cidA gene. The primer pair, lrgA1-F and lrgA1-R (Rice et al., 2004), was used to generate a DIG-labelled probe corresponding to the lrgA gene.

Flow cytometry

Membrane potential (ΔΨ) analyses were performed essentially as previously described by Novo et al. (1999). Briefly, 2 ml samples from the cultures used for RNA isolation that were untreated or treated with gramicidin, CCCP, nigericin, DCCD, valinomycin, DMSO and ethanol were centrifuged at room temperature at 4150 r.p.m. in a Sorvall Legend table-top centrifuge (Newtown, CT) and the resulting bacterial pellets were washed once with 5 ml of a buffer containing 60 mM Na2HPO4 and 60 mM NaH2PO4 (pH to 7.5), 5 mM KCl, 130 mM NaCl, 1.3 mM CaCl and 0.5 mM MgCl2. The pellets were then resuspended in this same buffer to an OD540 of between 0.5 and 0.6 and 1 ml of this cell suspension was placed in 5 ml polystyrene round-bottom tubes (Becton Dickinson Labware, Franklin Lakes, NJ). Next, DIOC2(3) (Molecular Probes, Eugene, OR) was added to a final concentration of 30 µM, incubated at room temperature for 10 min and analysed using a BD FACSAria flow cytometer (Becton Dickinson Labware, San Jose, CA). The DIOC2(3) was excited at 488 nm and detected at 530 nm (green fluorescence in the monomeric form) and 610 nm (red fluorescence in the aggregated form) and analyses were performed with a ratio of red to green fluorescence as described by Novo et al. (1999). The red fluorescence, which is due to aggregation of the dye, is dependent on cell size/clumping and ΔΨ. The green fluorescence is dependent on cell size/clumping and independent of ΔΨ. Thus, a red to green fluorescence ratio will reduce cell size/clumping as a factor and provide a more accurate measure of ΔΨ (Novo et al., 1999). The MFR and histograms of the data were produced on Windows Multiple Document Interface for Flow Cytometry 2.5 (http://www.cyto.purdue.edu/flowcyt/software/winmdi.htm).

For experiments in which the ΔΨ was re-established, overnight S. aureus cultures were inoculated to an OD600 of 0.1 into 20 ml of NZY broth and incubated for 4 h at 37°C with shaking at 250 r.p.m. After 4 h of incubation, cultures were supplemented with 100 µM CCCP and incubated for an additional 15 min. The CCCP was removed by harvesting the cells at 4150 r.p.m. in a Sorvall Legend table-top centrifuge (Newtown, CT) at room temperature for 3 min and discarding the supernatant. The pellets were resuspended in 10 ml of a buffer containing 60 mM Na2HPO4 and 60 mM NaH2PO4 (pH to 7.5), 5 mM KCl, 130 mM NaCl, 1.3 mM CaCl2, 0.5 mM MgCl2 and 1% (w/v) glucose, followed by a 30 min incubation at 37°C with shaking at 250 r.p.m. Flow cytometry analyses and RNA isolation were performed on samples collected before exposure to CCCP, 15 min after CCCP exposure and 30 min after resuspension of the cells in phosphate buffer with or without 1% (w/v) glucose.

ATP assays

Overnight S. aureus UAMS-1 cultures were inoculated to an OD600 of 0.1 into 20 ml of NZY broth and incubated at 37°C with shaking at 250 r.p.m. for 4 h. UAMS-1 cultures were then supplemented with DMSO, gramicidin (25 µg ml−1), CCCP (100 µM) or DCCD (200 µM) and incubated at 37°C with shaking at 250 r.p.m. for 15 min. The pre- and post-treated bacterial cultures (1.0 ml each) were centrifuged at 13 000 r.p.m. in a Biofuge pico (Heraeus Instruments, South Plainfield, NJ). The resulting bacterial pellets were resuspended in 1 ml of sterile water and the suspensions were transferred to FASTPREP tubes containing glass beads (lysing matrix B, Q.Biogene, Carlsbad, CA). The cells were lysed in a FASTPREP FP120 instrument (Bio 101 Thermo Savant, Vista, CA) at a speed of 6.0 for 23 s. Cell supernatants were harvested by centrifugation at 13 000 r.p.m. in a Biofuge pico (Heraeus Instruments, South Plainfield, NJ) at 4°C. The resulting supernatants (100 µl) were transferred to clean Eppendorf tubes and ATP concentrations were determined using an ATP bioluminescent assay kit (Sigma) according to the manufacturer's instructions. Bioluminescence was measured using a TopCount NXT microplate scintillation and luminescence counter (PerkinElmer, Downers Grove, IL).

Acknowledgements

We would like to thank Dr Kelly Rice, Linda Liou, Keun Seok Seo, Lakshmi Chandramohan and Dev Ranjit for technical assistance. This work was funded by NIH Grant No. R01AI038901, NIH-NRRI Grant No. P20RR15587 and DOD Grant No. DAAD 19-03-1-0191.

Ancillary

Advertisement