Cadaverine Inhibition of Porin Plays a Role in Cell Survival at Acidic pH
Department of Biology and Biochemistry, University of Houston, Houston, Texas 77204l(xn, 百拇医药
Received 8 August 2002/ Accepted 1 October 2002l(xn, 百拇医药
ABSTRACTl(xn, 百拇医药
When grown at acidic pH, Escherichia coli cells secrete cadaverine, a polyamine known to inhibit porin-mediated outer membrane permeability. In order to understand the physiological significance of cadaverine excretion and the inhibition of porins, we isolated an OmpC mutant that showed resistance to spermine during growth and polyamine-resistant porin-mediated fluxes. Here, we show that the addition of exogenous cadaverine allows wild-type cells to survive a 30-min exposure to pH 3.6 better than cells expressing the cadaverine-insensitive OmpC porin. Competition experiments between strains expressing either wild-type or mutant OmpC showed that the lack of sensitivity of the porin to cadaverine confers a survival disadvantage to the mutant cells at reduced pH. On the basis of these results, we propose that the inhibition of porins by excreted cadaverine represents a novel mechanism that provides bacterial cells with the ability to survive acid stress.
INTRODUCTIONug]v, 百拇医药
Bacteria respond and adapt to changing environmental conditions through the use of multiple mechanisms that range from the rapid opening of ion channels (23, 40) to the slower regulation of gene expression (31). Responses to pH changes have been one of the most extensively studied means of adaptation (32, 39). Neutralophiles, such as Escherichia coli, maintain an internal pH homeostasis of 7.4 to 7.9, even during growth at external pHs of 5 to 9 (44). The mechanisms for acidification and alkalinization of the cytoplasm in response to changes in external pH have not yet been completely elucidated, but correlations with K+ flux have been established in some cases (1). Better known are the regulatory mechanisms that affect gene expression in response to external pH, and a number of pH-responsive genes have been identified in E. coli and other species (13, 32, 39). The molecular basis for bacterial responses to external pH has been a topic of considerable clinical interest because of the intimate relationship between the onset of virulence of some pathogenic species and the acidity of the environment (7, 8, 11) and because of the required adaptation of food-borne pathogens as they transit through the stomach (10). In addition, the common practice of using organic acids as food preservatives is drawing attention to the mechanisms of the inducible acid tolerance response, whereby the growth of cells in a moderately acidic environment protects the organisms against more acidic conditions, triggering the expression of a specific battery of genes (2, 9, 12, 16).
A well-characterized pH-dependent response is the induction of the cadBA operon at acidic pH (32). The operon encodes a lysine decarboxylase (CadA) and a lysine-cadaverine antiporter (CadB). Upon lowering of the external pH, the CadC protein of the cytoplasmic membrane binds the cadBA promoter and activates the operon. Increased levels of CadA and CadB lead to the acid-induced synthesis of cadaverine from lysine and the subsequent excretion of the polyamine through the lysine-cadaverine antiporter. This response eventually results in some neutralization of the external pH, thus protecting the cell from the acidic conditions, although we have found no evidence for neutralization, at least not within the 2 to 4 h of cell growth in our experiments (see below and reference 36). The secretion of cadaverine, however, produces a rapid decrease in outer membrane permeability because cadaverine closes a number of porins (3, 36). Induction of cadA has also been found to contribute to the acid tolerance response in Vibrio species (27, 34).
Porins are pore-forming proteins of the outer membrane of gram-negative bacteria (29). The nonspecific or general diffusion porins OmpF and OmpC control the fluxes of nutrients and of ß-lactam antibiotics into the cell. The regulation of porin synthesis is tightly controlled by the environment, and changes in the ratio of OmpC to OmpF levels occur in response to pH, osmolarity, and temperature (15, 33). Only recently has the in vivo regulation of porin function been reported. Either the addition of polyamines to the external milieu or the excretion of cadaverine by the cells decreases the flux of ß-lactam antibiotics in intact cells (4, 36) because of the inhibition of OmpF and OmpC function by the polyamines (3, 18). Electrophysiological analysis of porin-mediated ion fluxes showed that the polyamines exert complex effects on the cation-selective porins but not on the anion-selective PhoE (6, 37). These effects include pore blockage and stabilization of closed states, and they appear to involve binding of the polyamine inside the pore (18, 19, 24).
On the basis of this mechanism to regulate porin function and the induction of cadaverine secretion upon a drop in pH, we have formulated the hypothesis that cadaverine inhibition of porin may play a role in the adaptive response to acidic conditions. Based on this hypothesis, we predicted that a cadaverine-resistant porin mutant is more sensitive to low pH levels than wild-type cells are. The results presented here confirm this prediction. They support the idea that E. coli cells gain a survival advantage under acidic conditions via a novel mechanism: inhibition of porin-dependent outer membrane permeability by cadaverine synthesized by acid-induced lysine decarboxylase. Most mechanisms of adaptation to external pH deal with either cytoplasmic or inner-membrane events. The results of experiments presented here suggest that we should extend the spectrum of survival strategies to include outer membrane processes.1, http://www.100md.com
MATERIALS AND METHODS1, http://www.100md.com
Growth media and chemicals. Luria broth medium (LB) contains 1% tryptone, 0.5% yeast extract (Difco Laboratories), and 1% NaCl, while modified LB (MLB) is LB with 0.5% NaCl. Agar plates contain 1.5% agar. IPTG (isopropyl ß-D-thiogalactopyranoside), azolectin (phosphatidylcholine, type IIS), and antibiotics were from Sigma. All other chemicals were from Fisher Scientific. Enzymes used in molecular biology protocols were purchased from Gibco or Promega. The DNA sequencing kit was from Perkin-Elmer, and the DNA purification kit was from Promega or Qiagen.
Strains and plasmids. pNLC10 is a pAC9-derived plasmid containing wild-type ompC under lac promoter regulation (24). All strains are E. coli K-12 derivatives. AW739 expresses only OmpC (17). Strain HS110 is an AW739 derivative lacking both OmpC and OmpF and was used as the host strain for the expression of the random mutant porins. Since strains lacking both OmpC and OmpF are notoriously unstable, we constructed HS110 by P1 transduction of a sequence with a Tn10-linked ompC deletion (obtained from AW738 [17]) into AW739 in the presence of pHS10, a pMAK705-derived, replication temperature-sensitive plasmid (14) carrying an ompF gene and expressing ompF at the permissive temperature (30°C). After transformation of HS110 with a plasmid harboring a mutated porin gene (see below), cells were cultured at 42°C in order to increase the loss of the pHS10 plasmid and thus express only the mutant porin gene. This strategy allows cells to contain porins at all times during the genetic manipulations. The AD100/pNLC10 and AD101/pNLC10 strains are isogenic LacZ+ and LacZ- derivatives of AW739, similar to HS110, but were made independently and carry pNLC10 instead of pHS10. The genotype of these strains is thus ompC ompF, but they express OmpC from the plasmid after induction by IPTG addition.
Selection of OmpC porin mutants. The G195D OmpC mutant used here was obtained after random mutagenesis followed by selection for improved growth on spermine-containing plates. pNLC10 was introduced into Epicurian Coli XL1-Red cells (Stratagene) deficient in three DNA repair genes (mutS, mutD, and mutT). Two hundred microliters of the transformation mixture was plated on MLB agar (containing 50 µg of kanamycin per ml) and incubated for 36 h at 37°C. Transformants were washed from the plates with 5 ml of MLB, and a 100-µl aliquot of the mixture was inoculated into 5 ml of liquid MLB and incubated for 16 h at 37°C. Propagation of the cloned porin genes in this strain allows the accumulation of random mutations. Twenty independent preparations of mutagenized plasmid DNA were purified (Promega Wizard Minipreps) from separate cultures, and 1 µl of each mutagenized plasmid preparation was transformed into 100 µl of competent HS110 cells. The transformants were grown at 37°C overnight on MLB plates with 50 µg of kanamycin per ml. All the transformants were washed from each plate with 5 ml of MLB, and 100 µl of the mixture was inoculated into fresh liquid MLB and grown for at least 5 h at 42°C to allow the loss of the pHS10 plasmid. The cultures were then diluted and plated on a selective medium (MLB agar with 5 mM spermine, 50 µg of kanamycin per ml, and 1 mM IPTG) for 12 to 14 h at 37°C to isolate porin mutants that are resistant to polyamines (during growth). One mutant per selection plate and per DNA preparation was selected and subjected to further testing. We verified that the mutant phenotype was due to a mutation in the plasmid-borne porin gene by purifying the plasmid from the mutant cells and retransforming the plasmid into HS110. Five transformants were picked, grown at 42°C for 5 to 7 h, and plated on selection plates again. If all tested colonies showed spermine resistance, the mutant ompC porin gene was sequenced (BigDye; Perkin-Elmer Applied Biosystems) on an ABI automated sequencer to identify the mutation. The plasmid carrying the G195D mutation in OmpC (selected as described above) was named pHSC515.
Preparation of porin extracts and SDS-polyacrylamide gel electrophoresis. Cells from an overnight culture were washed in 30 mM Tris (pH 8.1) and lysed by Tris-EDTA-lysozyme treatment followed by sonication (three times for 20 s on ice with a probe sonicator). After removal of the unlysed cells, the supernatant was centrifuged again to 180,000 x g for 15 min to collect the cell envelopes. The final pellet was resuspended in 50 mM KCl-5 mM HEPES-1 mM K-EGTA (pH 7.2). The determination of protein concentration was done by the bicinchoninic acid method (Pierce). Thirty micrograms of proteins was loaded on a 12% polyacrylamide-sodium dodecyl sulfate (SDS)-6 M urea gel. The level of porin expression was quantified with an Eagle Eye densitometer and MacBas software and was expressed as a percentage of the amount of OmpA (standard control).(\xpx7, 百拇医药
Phage test. Five hundred microliters of an overnight bacterial culture was mixed with 3 ml of soft MLB agar (0.3% agar) and overlaid on regular MLB plates. After the agar lawn solidified, a 10-µl drop of diluted SS4 phage stock (104 PFU/ml) was spotted on the plate. Plaques were detected after overnight incubation at 37°C.
Antibiotic permeation assays. The periplasmic ß-lactamase was expressed from the R471a factor obtained by conjugation with the donor strain HN37 (from H. Nikaido). Conjugants were selected on plates containing 50 µg of kanamycin per ml and 100 µg of ampicillin per ml. The rates of permeation of cephaloridine in live cells were determined with a Unikon 810 double-beam spectrophotometer (Kontron Instruments) (4), and the rates are expressed as 10-4 cm/min as described previously (30). Control experiments showed no enzyme leakage (degradation rates for cell supernatants were less than 5% of the experimental values).)|#7p, http://www.100md.com
Cadaverine excretion assay. Measurements of culture supernatants for cadaverine content were done as described previously (36). The data are given as A340 values of the extract, where a value of 1.0 corresponds to ~ 250 µM cadaverine. Appropriate medium blanks were used to verify that trace amounts of other amines and amino acids present in the supernatant did not interfere with the cadaverine measurement.
Patch clamp electrophysiology. Outer membrane fractions were purified by sucrose gradient centrifugation from bacterial cells grown exponentially in LB, and patch clamp experiments were performed on lipid blisters induced from artificial liposomes containing the reconstituted outer membrane fractions (5). Control experiments were done with symmetric solutions of 150 mM KCl, 5 mM HEPES, 0.1 mM K-EDTA, and 0.01 mM CaCl2 (pH 7.2). Fifteen-milliliter solutions of spermine in the same buffer were then applied to the periplasmic side of the patch with bath perfusion. Currents were recorded with an Axopatch-1D amplifier (Axon Instruments), filtered at 1 kHz (Frequency Devices), and digitized at 100-µs sampling intervals (Instrutech).%3!^}, http://www.100md.com
Competition experiment. We constructed four strains that do not express ompF or ompC from the chromosome but that express the wild-type ompC gene from the pNLC10 plasmid or the mutant ompC gene from the pHSC515 plasmid. The strains were two LacZ+ strains, AD100/pNLC10 (C+B) and AD100/pHSC515 (CrB), and two LacZ- strains, AD101/pNLC10 (C+W) and AD101/pHSC515 (CrW), where C+ and Cr refer to the wild-type and polyamine-resistant OmpC strains, respectively, and B and W refer to blue (LacZ+) and white (LacZ-) colonies, respectively. For each competition experiment, four different mixed cultures were started by adding a 100-µl aliquot from two separate cultures incubated overnight into the same 10 ml of MLB at pH 7.0 or 5.0: culture 1, C+B mixed with CrW; culture 2, CrB mixed with C+W; culture 3, C+B mixed with C+W; and culture 4, CrB mixed with CrW. Immediately following inoculation of the cultures, a portion of each mixture was removed for cell counting. The cultures were incubated at 37°C with shaking until an optical density at 650 nm of ~ 0.6 (3 to 4 h) was reached. Then, a fresh 10-ml culture was inoculated with 100 µl of this previous culture, and an aliquot was removed again for cell counting. This cycle was repeated six times. Cell counting was done by removing an aliquot of the culture, making serial dilutions, and plating 100 µl of each dilution on LB plates containing 50 µg of kanamycin per ml, 1 mM IPTG, and 40 µg of X-Gal (5-bromo-4-chloro-3-indoyl-ß-D-galactopyranoside) per ml. The numbers of blue and white colonies on each plate were counted after overnight incubation of the plates at 37°C.
Survival rate determination. Twenty-milliliter cultures of either AD100/pNLC10 (C+B) or AD101/pHSC515 (CrW) were grown at 37°C with shaking at ~ 250 rpm in MLB (pH 6.8) containing either 0 or 20 mM cadaverine, until an optical density at 650 nm of 0.6 to 0.7 was reached. The AD100/pNLC10 culture grown in the absence of cadaverine was mixed 1:1 with the AD101/pHSC515 culture under the same conditions. The same was done for the two cultures grown in the presence of cadaverine. Each mixture was then divided in half, and the pH of one-half of the mixture was lowered to 3.6 by using 1 M HCl. This procedure resulted in four cultures, all initially containing a 1:1 proportion of C+B and CrW: two cultures at pH 6.8 with either 0 or 20 mM cadaverine and two other cultures at pH 3.6 with either 0 or 20 mM cadaverine. The four cultures were incubated at 37°C with shaking for 30 min. Serial dilutions were made and plated on MLB agar plates containing X-Gal and the appropriate antibiotics. Colonies were counted after the plates were incubated at 37°C overnight.
RESULTS AND DISCUSSIONy4y+, 百拇医药
Mutant isolation and characterization. Polyamines have been shown to act as natural modulators of ion channel activities in eukaryotes (20, 26) and prokaryotes (3, 18). In an effort to define the molecular mechanism of inhibition of bacterial porins by polyamines and the physiological significance of this type of modulation, we developed a method of selection for porin mutants that have lost sensitivity to polyamines. Although not produced endogenously, spermine was chosen for the selection because it is the most potent polyamine in the inhibition of porin. Wild-type E. coli cells grow poorly on LB plates containing 5 to 10 mM spermine. This may be due to two reasons. First, spermine is small enough to permeate porins; import of spermine from the periplasm can occur via the spermidine-preferential system PotABCD (21, 22, 41), and high intracellular concentrations of spermine are toxic because of the ability of this polycationic molecule to bind nucleic acids and proteins. Second, spermine inhibits OmpF and OmpC as it permeates porins and as it interacts with the inwardly folded constriction loop (19); as inhibition occurs, cells experience a deficit in nutrient flux that slows growth on plates.
Our goal was to isolate porin mutants that allow growth in the presence of the polyamine, either because they provide reduced spermine entry, reduced porin inhibition by spermine, or both. Mutants which have lost porin inhibition by polyamines would give us a tool to address the physiological significance of this form of modulation. For this purpose, we subjected an ompC-carrying plasmid (pNLC10) to random mutagenesis and introduced it into a host cell deficient in OmpC and OmpF. Seven mutants displaying growth on spermine plates were obtained; six were shown to have a single missense mutation and one was an ochre mutation. Four mutants (with A129P, Q175ochre, D204N, and S274P mutations) displayed a reduced porin level (<25% of that of the wild type), and two others (with V96D and G251D mutations) showed polyamine resistance in growth but not in antibiotic flux or in electrophysiology. Presumably, these mutants were selected only because they provided a reduced level of entry of spermine inside the cell due to reduced porin levels or impaired pore properties. The G195D mutant is the only mutant that maintained reasonable levels of expression and displayed a robust polyamine resistance phenotype in flux assays and was therefore used in this study to investigate the physiological significance of porin inhibition by polyamines. Residue G195 is located in the barrel wall adjacent to the L3 loop and is in close proximity to this loop. As the conformation of the L3 loop has been shown to be important for polyamine inhibition (19), it is possible that the G195D mutant lost polyamine sensitivity because of secondary effects on the L3 loop conformation.
Several phenotypes of the G195D mutant were compared to those of the same strain carrying pNLC10 with the wild-type ompC. The pNLC10 plasmid was initially constructed to provide OmpC levels that are comparable to those of the haploid wild-type strain (24). Although the porin levels were 54% ± 12% of the levels of wild-type porin expressed from the wild-type plasmid in the presence of IPTG (porin levels are <5% in the absence of IPTG) and the sensitivity to phage SS4 was slightly reduced, the rates of cephaloridine flux were comparable for both strains (expressed as 10-4 cm/min): 5.7 ± 0.3 for the wild type (nine samples) and 6.3 ± 0.3 for the mutant (three samples). It is known that ß-lactam antibiotics such as cephaloridine require porins for entry, and thus the flux of these antibiotics is a reliable indication of the permeability of the outer membrane (30). When flux assays were performed in the presence of external polyamines, a decrease in cephaloridine flux through the wild-type porin was clearly seen, as previously reported (4), with 26% ± 9% inhibition in the presence of 1 mM external spermine (six samples) and 31% ± 1% inhibition in the presence of 200 mM external cadaverine (three samples). As exogenously added cadaverine is less potent than spermine, we purposely chose to add 200 mM cadaverine externally to obtain a substantial reduction in flux. We previously showed that much smaller concentrations of cadaverine cause much larger effects when the polyamine exerts its effect from the periplasmic side (18, 36). On the contrary, the G195D mutant displayed only a 5% ± 3% and a 7% ± 2% inhibition in the presence of 1 mM spermine and 200 mM cadaverine, respectively (three samples for each). The relative flux values obtained with the G195D mutant in the presence of polyamines were always significantly greater than in the wild-type strain (P < 0.05, two-sample t test). It is important to point out that the inhibitory mechanism, as revealed by electrophysiology, mostly involves the complete closure of the affected porin molecules rather than the partial or total plugging of the pores by the polyamine molecules. Thus, the lack of inhibition seen in the mutant porin is due to the insensitivity of the mutant porin to the polyamines rather than to a compensatory mechanism originating from the fact that the mutant pore might be wider (since the total amount of antibiotic flux is comparable to that of the wild-type strain, despite the reduction in porin number).
This insensitivity to modulation by polyamine was confirmed with the very sensitive technique of patch clamp electrophysiology. shows typical recordings of currents from membrane patches containing many wild-type or mutant porins. Because porins are open most of the time under control conditions, the current trace dwells at a baseline level, from which upward transitions represent the rapid and transient spontaneous closures of one or many channels. This electrophysiological behavior has been extensively studied (6), but its details are not the subject of this publication. Here, we qualitatively point out the differences that wild-type and mutant channels exhibit towards inhibition by spermine. shows that the wild-type channels displayed a greatly modified activity in the presence of 1 mM spermine. The frequency and duration of channel closures (upward deflections) were greatly increased, and the current trace dwelled at various levels, depending on how many channels were closed. This inhibitory action of spermine was not observed at all with the G195D mutant, since the kinetic signature of the current trace was essentially unchanged in the presence of the polyamine. These results were confirmed in three independent experiments. Together with the antibiotic-flux assays, these results show that the inhibition of porin by polyamine was abolished in the G195D mutant.
fig.ommittedi.45, http://www.100md.com
FIG. 1. Spermine insensitivity of the G195D OmpC mutant. Representative patch clamp recordings are shown for the wild-type and the mutant in the absence (control [CON]) and the presence of 1 mM spermine (SPM). BL indicates the baseline level representing the total current flowing through multiple OmpC monomers. Upward deflections represent transient closures of channels. The pipette voltage was -60 mV.i.45, http://www.100md.com
Survival of the mutant strain at acidic pH. One of the cellular responses of E. coli cells to acidic pH is the excretion of the polyamine cadaverine. The role that this molecule may play in adaptation to pH has remained elusive. The secretion of cadaverine leads to a rapid and sustained inhibition of porins in vivo (36). Cadaverine produced by the cell is able to act from the periplasm, and we have shown that cadaverine applied from the periplasmic side is much more effective than exogenously added cadaverine (18, 36). Our previous results showed that the inhibition of outer membrane permeability by secreted cadaverine is not complete and leaves enough pathways for nutrient entry; thus, the production of cadaverine itself is not deleterious for the cell. To understand whether a physiological link ties porin modulation by cadaverine to the secretion of cadaverine at low pH, we compared the G195D mutant cells to wild-type cells for their ability to survive a treatment at pH 5.0. We purposely chose this pH for the following reasons: (i) cadA expression is maximal at pH 5.0 to 6.0 (38); and (ii) our goal was to monitor cells while they were being challenged by an acid pH at which they could still grow and adapt to the conditions rather than to look for protection against the extreme acid stress at which cell death would occur.
For this, we performed competition experiments where wild-type and mutant cells were mixed together in a single flask and grown until mid-log phase. At the end of this growth, a cell count was performed, the pH of the culture was measured, the amount of cadaverine released was estimated, and an aliquot of this culture was used to start a fresh culture. This cycle was repeated six times. Thus, relative survival rates of wild-type cells and mutant cells were compared as the cultures went through several generations. To distinguish wild-type cells from mutant cells, we constructed LacZ+ and LacZ- marked strains that give rise to blue or white colonies on X-Gal plates, respectively. and B show the averaged results of three experiments. The ordinate axes represent the percentages of blue colonies, obtained by dividing the number of blue colonies by the total number of colonies on each plate; error bars are standard deviations. The data were normalized across experiments by setting the initial percentage of blue colonies to 50 for all curves (cycle 0). In each experiment, four cultures were investigated (as described in Materials and Methods): culture 1, C+B mixed with CrW; culture 2, CrB mixed with C+W; culture 3, C+B mixed with C+W; and culture 4, CrB mixed with CrW. Thus, each culture was internally controlled; the results from cultures 1 and 2 were complementary, and cultures 3 and 4 served as controls.
fig.ommitted}, 百拇医药
FIG. 2. Competition experiments. Measurements were performed on cultures grown at pH 7.0 (A, C, and E) and pH 5.0 (B, D, and F). (A and B) Averaged percentages of blue colonies in three separate experiments (error bars are standard deviations) after all values at cycle 0 were normalized to 50% to compare the results of the three experiments. The actual percentages at cycle 0 ranged between 40 and 57% for all three experiments. (C and D) pHs of the four cultures throughout a single experiment. (E and F) Amount of cadaverine excreted by the cells, as measured by the A340 in the cadaverine assay in a single experiment. For all panels, the cultures consisted of the following: culture 1, C+B cells mixed with CrW cells ; culture 2, CrB cells mixed with C+W cells ; culture 3, C+B cells mixed with C+W cells ; and culture 4, CrB cells mixed with CrW cells (where C+ and Cr refer to the wild-type and polyamine-resistant OmpC strains, respectively, and B and W refer to blue [LacZ+] and white [LacZ-] colonies, respectively).
through F show the results of the pH and cadaverine measurements for a single experiment. The pHs remained as set, at 7.0 and 5.0, for all four cultures throughout the cycles. The amount of cadaverine excreted by cells was negligible at pH 7.0 , but a significant amount was excreted when cells were grown at pH 5.0 . In this case, as the cultures went through more cycles, the excretion of cadaverine increased until the A340 reached about 1.0 (about 250 µM cadaverine) in the third cycle, where it remained for the rest of the experiment. shows that the number of blue colonies hovered around 50% throughout the entire experiment at pH 7.0 with no clear trend of the mutant or the wild type being preferred. Thus, in neutral environments, there was no excretion of cadaverine and the wild-type and mutant cells had equal levels of fitness. On the contrary, at pH 5.0, the percentage of wild-type cells increased and the percentage of mutant cells decreased as the culture went through more cycles. This trend became more significant after the fourth cycle. By the end of the sixth cycle, the percentage of wild-type cells was greater than 70% and the percentage of mutants was below 25%. However, the percentage of blue colonies in both control sets remained at ~ 50%. The wild type is clearly at an advantage at pH 5.0, unlike at pH 7.0.
We have previously found that the excretion of up to ~ 250 µM cadaverine correlates with an ~ 70% inhibition of antibiotic flux mediated by OmpF and OmpC (36). Thus, the measured levels of excreted cadaverine are effective to promote inhibition of the wild-type channel. We believe that most of this inhibition occurs while cadaverine transits through the periplasm, as it persists even after the external cadaverine is washed (36). Note that much higher concentrations of externally supplied cadaverine were needed in the characterization of the mutants by antibiotic flux assays, because cadaverine is much more potent when it is applied to the periplasmic side than to the extracellular side of porin molecules (18).;8, 百拇医药
These results suggest that the inhibition of porins by excreted cadaverine may be an adaptive mechanism that helps the E. coli cells survive acidic environments. It is important to note that the only significant difference between the wild-type and mutant cells is the polyamine-resistant phenotype of their OmpC porin. The reduced porin levels, as detected by SDS-polyacrylamide gel electrophoresis, are likely to be inconsequential since the flux efficiency is similar to that of the wild-type cells and since the mutant strain was not outcompeted at neutral pH. In addition, although the activation of CadC at low pH is known to decrease porin expression (36), this is not relevant in these experiments since porin expression is solely driven from the IPTG-dependent plasmid promoter. Finally, it is also important to consider whether the sensitivity of the mutant to acidic pH is due to the inability of this porin mutant to close at low pH. There have been numerous reports that acidic pH triggers closure of wild-type porins in electrophysiological experiments (25, 35, 42, 43). We have confirmed this mode of porin regulation in vivo by comparing the flux of cephaloridine at pH 3.6 with that at pH 6.0 (in the absence of added polyamines). Despite a 10-fold decrease in the activity of the released ß-lactamase, compared to levels at pH 6.0, antibiotic flux in the wild-type and G195D mutant cells at pH 3.6 was reduced 22% ± 3% and 26% ± 4%, respectively (three samples), indicating that the rate of antibiotic flux through porins was still rate limiting. Thus, the extents of inhibition by acidic pH of porin-dependent permeability were similar in wild-type and mutant cells, suggesting that the mutant porins had not lost their sensitivity to pH.
Cadaverine and pH sensitivity. The results of the competition experiments are substantiated by our finding that cadaverine supplied exogenously helps cells carrying the wild-type OmpC porin survive a 30-min exposure to acidic pH better than cells carrying the G195D mutant OmpC porin. Here, we aimed at obtaining some cell death; thus, a pH of 3.6 was chosen, despite the fact that this pH would produce little excreted cadaverine in vivo. Cultures containing LacZ+ and LacZ- cells expressing the wild-type and mutant OmpC porin, respectively, were grown in parallel in LB and in LB supplemented with 20 mM cadaverine to allow the accumulation of cadaverine in the periplasm and the inhibition of porins. Then we mixed equal amounts of LacZ+ and LacZ- cells expressing the wild-type and mutant OmpC porin, respectively, so that the experiments were internally controlled. Each mixture was exposed to pH 3.6 for 30 min, still in the presence or absence of 20 mM cadaverine. After the plating of the cells, the ratio of blue (wild-type) to white (mutant) colonies was calculated for each condition. We found that there were 16 ± 6 times more wild-type cells than mutant cells surviving acidic pH when cadaverine was present (three samples) (this value is not directly comparable to those of Fig. 2 because of the large difference in the pHs used in the two sets of experiments). In the absence of cadaverine, the ratio of wild-type to mutant cells after the acid treatment was 0.9 ± 0.5. Similarly, a ratio of ~ 1.0 was consistently found when the mixed cells remained at pH 6.8, regardless of the presence of cadaverine. Thus, in the absence of cadaverine, the survival of wild-type and that of mutant cells were equally affected. However, wild-type cells were able to survive acidic pH better than mutant cells specifically when cadaverine was present.
Conclusions. We have identified a mutation in the ompC gene that confers resistance to the inhibition of porin function by cadaverine. Taken together, our results suggest that this resistance of the mutant OmpC porin to cadaverine inhibition is correlated with the mutant cells being outcompeted by wild-type cells only in an acidic milieu and not at neutral pH. Thus, the excretion of cadaverine at low pH appears to play a role in the adaptation of cells to these conditions.x, 百拇医药
It is noteworthy that cells reduce their outer membrane permeability under acidic conditions by the use of multiple pathways that may act sequentially. The pH-induced closure of porins is immediate (25, 35, 42) and will rapidly lead to about a 20 to 25% reduction in permeability. This reduction is followed by the slower processes of regulation of gene expression, which affects the porin regulon and the cad operon. Acidic pH promotes the decrease of porin levels in the outer membrane through both an envZ-dependent pathway (15) and a novel mechanism involving the cad operon regulator CadC (36). Finally, the secretion of cadaverine as a result of increased CadA and CadB levels leads to the functional inhibition of up to ~ 70% of the porin-mediated permeability (36). This multiplicity of mechanisms points to the importance of reducing outer membrane permeability under acidic conditions, although it is difficult to assess whether such a reduction permits a better control of periplasmic composition. The work presented here, however, demonstrates that the impairment of one such pathway leads to a deficit in cell survival; thus, a tight control of influx—and possibly efflux—across the outer membrane is required for optimal adaptation to acidic conditions. It appears that cells might find a compromise whereby the incomplete shutdown of the porin-mediated pathways still provides sufficient nutrient flux, while reducing the amplitude or rate of acid flux to confer a survival advantage over time. The ~ 70% decrease in permeability found in the presence of secreted cadaverine (36) does not seem to impair the growth rate significantly in rich laboratory media. But by the same token, it appears that a minimum permeability is required, as porin-null mutants rapidly acquire compensatory mutations. Therefore, the necessity for striking the right balance in the interplay of nutrient and acid flux might be of even more significance under conditions of nutrient limitation, as is the case outside the host, than in rich media. The secretion of cadaverine is not the only example of regulation of polyamine extrusion in response to environmental conditions. A rapid rise in external osmolarity promotes the excretion of cellular putrescine (28), which inhibits porin-mediated fluxes (4). The use of natural modulators of porin activity to control the permeability of the outer membrane may be a general strategy used by cells in response to deleterious external conditions.
ACKNOWLEDGMENTS6]l1r8l, 百拇医药
We acknowledge Aparna Kumaraswamy for help with some of the experiments. We thank Linda Guynn for sequencing the mutant porin genes and William Widger for the use of the spectrophotometer.6]l1r8l, 百拇医药
This work was supported by NIH grant AI34905.6]l1r8l, 百拇医药
REFERENCES6]l1r8l, 百拇医药
Booth, I. R. 1985. Regulation of cytoplasmic pH in bacteria. Microbiol. Rev. 49:359-378.6]l1r8l, 百拇医药
Conner, D. E., and J. S. Kotrola. 1995. Growth and survival of Escherichia coli O157:H7 under acidic conditions. Appl. Environ. Microbiol. 61:382-385.6]l1r8l, 百拇医药
delaVega, A. L., and A. H. Delcour. 1995. Cadaverine induces closing of E. coli porins. EMBO J. 14:6058-6065.6]l1r8l, 百拇医药
delaVega, A. L., and A. H. Delcour. 1996. Polyamines decrease Escherichia coli outer membrane permeability. J. Bacteriol. 178:3715-3721.6]l1r8l, 百拇医药
Delcour, A. H., B. Martinac, C. Kung, and J. Adler. 1989. A modified reconstitution method used in patch-clamp studies of Escherichia coli ion channels. Biophys. J. 56:631-636.
Delcour, A. H. 1997. Function and modulation of bacterial porins: insights from electrophysiology. FEMS Microbiol. Lett. 151:115-123.{emoz, http://www.100md.com
DiRita, V. J. 1995. Three-component regulatory system controlling virulence in Vibrio cholerae, p. 351-365. In J. A. Hoch and T. J. Silhavy (ed.), Two-component signal transduction. ASM Press, Washington, D.C.{emoz, http://www.100md.com
Dziejman, M., and J. J. Mekalanos. 1995. Two-component signal transduction and its role in the expression of bacterial virulence factors, p. 305-317. In J. A. Hoch and T. J. Silhavy (ed.), Two-component signal transduction. ASM Press, Washington, D.C.{emoz, http://www.100md.com
Foster, J. W. 1999. When protons attack: microbial strategies of acid adaptation. Curr. Opin. Microbiol. 2:170-174.{emoz, http://www.100md.com
Gorden, J., and P. L. C. Small. 1993. Acid resistance in enteric bacteria. Infect. Immun. 61:364-367.{emoz, http://www.100md.com
Groisman, E. A., and F. Heffron. 1995. Regulation of Salmonella virulence by two-component regulatory systems, p. 319-332. In J. A. Hoch and T. J. Silhavy (ed.), Two-component signal transduction. ASM Press, Washington, D.C.
Guilfoyle, D. E., and I. N. Hirshfield. 1996. The survival benefit of short-term organic acids and the inducible arginine and lysine decarboxylase genes for Escherichia coli. Lett. Appl. Microbiol. 22:393-396.w9h$, 百拇医药
Hall, H. K., K. L. Karem, and J. W. Foster. 1995. Molecular responses of microbes to environmental pH stress. Adv. Microb. Physiol. 37:229-272.w9h$, 百拇医药
Hamilton, C. M., M. Aldea, B. K. Washburn, P. Babitzke, and S. R. Kushner. 1989. New method for generating deletions and gene replacements in Escherichia coli. J. Bacteriol. 171:4617-4622.w9h$, 百拇医药
Heyde, M., and R. Portalier. 1987. Regulation of major outer membrane porin proteins of Escherichia coli K-12 by pH. Mol. Gen. Genet. 208:511-517.w9h$, 百拇医药
Hill, C., B. O'Driscoll, and I. Booth. 1995. Acid adaptation and food poisoning microorganisms. Int. J. Food Microbiol. 28:245-254.w9h$, 百拇医药
Ingham, C., M. Buechner, and J. Adler. 1990. Effect of outer membrane permeability on chemotaxis in Escherichia coli. J. Bacteriol. 172:3577-3583.
Iyer, R., and A. H. Delcour. 1997. Complex inhibition of OmpF and OmpC bacterial porins by polyamines. J. Biol. Chem. 272:18595-18601.78, http://www.100md.com
Iyer, R., Z. Wu, P. M. Woster, and A. H. Delcour. 2000. Molecular basis for the polyamine—OmpF porin interactions: inhibitor and mutant studies. J. Mol. Biol. 297:933-945.78, http://www.100md.com
Johnson, T. D. 1996. Modulation of channel function by polyamines. Trends Pharmacol. Sci. 17: 22-27.78, http://www.100md.com
Karahalios, P., P. Mamos, D. H. Vynios, D. Papaioannou, and D. L. Kalpaxis. 1998. The effect of acylated polyamine derivatives on polyamine uptake mechanisms, cell growth, and polyamine pools in Escherichia coli, and the pursuit of structure/activity relationships. Eur. J. Biochem. 251:998-1004.78, http://www.100md.com
Kashiwagi, K., H. Kobayashi, and K. Igarashi. 1986. Apparently unidirectional polyamine transport by proton motive force in polyamine-deficient Escherichia coli. J. Bacteriol. 165:972-977.78, http://www.100md.com
Levina, N., S. Tötemeyer, N. R. Stokes, P. Louis, M. A. Jones, and I. R. Booth. 1999. Protection of Escherichia coli cells against extreme turgor by activation of MscS and MscL mechanosensitive channels: identification of genes required for MscS activity. EMBO J. 18:1730-1737.
Liu, N., M. J. Benedik, and A. H. Delcour. 1997. Disruption of polyamine modulation by a single amino acid substitution on the L3 loop of the OmpC porin channel. Biochim. Biophys. Acta 1326:201-212.2au00j6, http://www.100md.com
Liu, N., and A. H. Delcour. 1998. Inhibitory effect of acidic pH on OmpC porin: wildtype and mutant studies. FEBS Lett. 434:160-164.2au00j6, http://www.100md.com
Lopatin, A., E. N. Makhina, and C. G. Nichols. 1994. Potassium channel blocking by cytoplasmic polyamines as the mechanism of intrinsic rectification. Nature 372:366-369.2au00j6, http://www.100md.com
Merrell, D. S., and A. Camilli. 1999. The cadA gene of Vibrio cholerae is induced during infection and plays a role in acid tolerance. Mol. Microbiol. 34:836-849.2au00j6, http://www.100md.com
Munro, G. F., K. Hercules, J. Morgan, and W. Sauerbier. 1972. Dependence of the putrescine content of Escherichia coli on the osmotic strength of the medium. J. Biol. Chem. 247:1272-1280.2au00j6, http://www.100md.com
Nikaido, H. 1996. Outer membrane, p. 29-47. In F. C. Neidhardt et al. (ed.), Escherichia coli and Salmonella: cellular and molecular biology, 2nd ed., vol. 1. ASM Press, Washington, D.C.
Nikaido, H., E. Y. Rosenberg, and J. Foulds. 1983. Porin channels in Escherichia coli: studies with ß-lactams in intact cells. J. Bacteriol. 153:232-240.2#!f, http://www.100md.com
Ninfa, A. J. 1996. Regulation of gene transcription by extracellular stimuli, p. 1246-1262. In F. C. Neidhardt et al. (ed.), Escherichia coli and Salmonella: cellular and molecular biology, 2nd ed., vol. 1. ASM Press, Washington, D.C.2#!f, http://www.100md.com
Olson, E. 1993. Influence of pH on bacterial gene expression. Mol. Microbiol. 8:5-14.2#!f, http://www.100md.com
Pratt, L. A., and T. J. Silhavy. 1995. Porin regulon of Escherichia coli, p. 105-127. In J. A. Hoch and T. J. Silhavy (ed.), Two-component signal transduction. ASM Press, Washington, D.C.2#!f, http://www.100md.com
Rhee, J. E., J. H. Rhee, P. Y. Ryu, and S. H. Choi. 2002. Identification of the cadBA operon from Vibrio vulnificus and its influence on survival to acid stress. FEMS Microbiol. Lett. 208:245-251.2#!f, http://www.100md.com
Saint, N., A. Prilipov, A. Hardmeyer, K.-L. Lou, T. Schirmer, and J. P. Rosenbusch. 1996. Replacement of the sole histidinyl residue in OmpF porin from E. coli by threonine (H21T) does not affect channel structure and function. Biochem. Biophys. Res. Commun. 223:118-122.
Samartzidou, H., and A. H. Delcour. 1999. Excretion of endogenous cadaverine leads to a decrease in porin-mediated outer membrane permeability. J. Bacteriol. 181:791-798.3k, 百拇医药
Samartzidou, H., and A. H. Delcour. 1999. Distinct sensitivities of OmpF and PhoE to charged modulators. FEBS Lett. 444:65-70.3k, 百拇医药
Slonczewski, J. L., T. N. Gonzalez, F. M. Bartholomew, and N. J. Holt. 1987. Mu d-directed lacZ fusions regulated by low pH in Escherichia coli. J. Bacteriol. 169:3001-3006.3k, 百拇医药
Slonczewski, J. L., and J. W. Foster. 1996. pH-regulated genes and survival at extreme pH, p. 1539-1549. In F. C. Neidhardt et al. (ed.), Escherichia coli and Salmonella: cellular and molecular biology, 2nd ed., vol. 1. ASM Press, Washington, D.C.3k, 百拇医药
Sukharev, S. I., P. Blount, B. Martinac, and C. Kung. 1997. Mechanosensitive channels of Escherichia coli: the MscL gene, protein and activities. Annu. Rev. Physiol. 59:633-657.3k, 百拇医药
Tabor, C. W., and H. Tabor. 1966. Transport systems for 1,4-diaminobutane, spermidine and spermine in Escherichia coli. J. Biol. Chem. 241:3714-3723.3k, 百拇医药
Todt, J. C., W. J., Rocque, and E. J. McGroarty. 1992. Effects of pH on bacterial porin function. Biochemistry 31:10471-10478.3k, 百拇医药
Xu, G., B. Shi, E. J. McGroarty, and H. T. Tien. 1986. Channel-closing activity of porins from Escherichia coli in bilayer lipid membranes. Biochim. Biophys. Acta 862:57-64.3k, 百拇医药
Zilberstein, D., V. Agmon, S. Schuldiner, and E. Padan. 1984. Escherichia coli intracellular pH, membrane potential, and cell growth. J. Bacteriol. 158:246-252.(Hrissi Samartzidou Mahsa Mehrazin Zhaohui Xu Michael J. Benedik and Anne H. Delcour)
Received 8 August 2002/ Accepted 1 October 2002l(xn, 百拇医药
ABSTRACTl(xn, 百拇医药
When grown at acidic pH, Escherichia coli cells secrete cadaverine, a polyamine known to inhibit porin-mediated outer membrane permeability. In order to understand the physiological significance of cadaverine excretion and the inhibition of porins, we isolated an OmpC mutant that showed resistance to spermine during growth and polyamine-resistant porin-mediated fluxes. Here, we show that the addition of exogenous cadaverine allows wild-type cells to survive a 30-min exposure to pH 3.6 better than cells expressing the cadaverine-insensitive OmpC porin. Competition experiments between strains expressing either wild-type or mutant OmpC showed that the lack of sensitivity of the porin to cadaverine confers a survival disadvantage to the mutant cells at reduced pH. On the basis of these results, we propose that the inhibition of porins by excreted cadaverine represents a novel mechanism that provides bacterial cells with the ability to survive acid stress.
INTRODUCTIONug]v, 百拇医药
Bacteria respond and adapt to changing environmental conditions through the use of multiple mechanisms that range from the rapid opening of ion channels (23, 40) to the slower regulation of gene expression (31). Responses to pH changes have been one of the most extensively studied means of adaptation (32, 39). Neutralophiles, such as Escherichia coli, maintain an internal pH homeostasis of 7.4 to 7.9, even during growth at external pHs of 5 to 9 (44). The mechanisms for acidification and alkalinization of the cytoplasm in response to changes in external pH have not yet been completely elucidated, but correlations with K+ flux have been established in some cases (1). Better known are the regulatory mechanisms that affect gene expression in response to external pH, and a number of pH-responsive genes have been identified in E. coli and other species (13, 32, 39). The molecular basis for bacterial responses to external pH has been a topic of considerable clinical interest because of the intimate relationship between the onset of virulence of some pathogenic species and the acidity of the environment (7, 8, 11) and because of the required adaptation of food-borne pathogens as they transit through the stomach (10). In addition, the common practice of using organic acids as food preservatives is drawing attention to the mechanisms of the inducible acid tolerance response, whereby the growth of cells in a moderately acidic environment protects the organisms against more acidic conditions, triggering the expression of a specific battery of genes (2, 9, 12, 16).
A well-characterized pH-dependent response is the induction of the cadBA operon at acidic pH (32). The operon encodes a lysine decarboxylase (CadA) and a lysine-cadaverine antiporter (CadB). Upon lowering of the external pH, the CadC protein of the cytoplasmic membrane binds the cadBA promoter and activates the operon. Increased levels of CadA and CadB lead to the acid-induced synthesis of cadaverine from lysine and the subsequent excretion of the polyamine through the lysine-cadaverine antiporter. This response eventually results in some neutralization of the external pH, thus protecting the cell from the acidic conditions, although we have found no evidence for neutralization, at least not within the 2 to 4 h of cell growth in our experiments (see below and reference 36). The secretion of cadaverine, however, produces a rapid decrease in outer membrane permeability because cadaverine closes a number of porins (3, 36). Induction of cadA has also been found to contribute to the acid tolerance response in Vibrio species (27, 34).
Porins are pore-forming proteins of the outer membrane of gram-negative bacteria (29). The nonspecific or general diffusion porins OmpF and OmpC control the fluxes of nutrients and of ß-lactam antibiotics into the cell. The regulation of porin synthesis is tightly controlled by the environment, and changes in the ratio of OmpC to OmpF levels occur in response to pH, osmolarity, and temperature (15, 33). Only recently has the in vivo regulation of porin function been reported. Either the addition of polyamines to the external milieu or the excretion of cadaverine by the cells decreases the flux of ß-lactam antibiotics in intact cells (4, 36) because of the inhibition of OmpF and OmpC function by the polyamines (3, 18). Electrophysiological analysis of porin-mediated ion fluxes showed that the polyamines exert complex effects on the cation-selective porins but not on the anion-selective PhoE (6, 37). These effects include pore blockage and stabilization of closed states, and they appear to involve binding of the polyamine inside the pore (18, 19, 24).
On the basis of this mechanism to regulate porin function and the induction of cadaverine secretion upon a drop in pH, we have formulated the hypothesis that cadaverine inhibition of porin may play a role in the adaptive response to acidic conditions. Based on this hypothesis, we predicted that a cadaverine-resistant porin mutant is more sensitive to low pH levels than wild-type cells are. The results presented here confirm this prediction. They support the idea that E. coli cells gain a survival advantage under acidic conditions via a novel mechanism: inhibition of porin-dependent outer membrane permeability by cadaverine synthesized by acid-induced lysine decarboxylase. Most mechanisms of adaptation to external pH deal with either cytoplasmic or inner-membrane events. The results of experiments presented here suggest that we should extend the spectrum of survival strategies to include outer membrane processes.1, http://www.100md.com
MATERIALS AND METHODS1, http://www.100md.com
Growth media and chemicals. Luria broth medium (LB) contains 1% tryptone, 0.5% yeast extract (Difco Laboratories), and 1% NaCl, while modified LB (MLB) is LB with 0.5% NaCl. Agar plates contain 1.5% agar. IPTG (isopropyl ß-D-thiogalactopyranoside), azolectin (phosphatidylcholine, type IIS), and antibiotics were from Sigma. All other chemicals were from Fisher Scientific. Enzymes used in molecular biology protocols were purchased from Gibco or Promega. The DNA sequencing kit was from Perkin-Elmer, and the DNA purification kit was from Promega or Qiagen.
Strains and plasmids. pNLC10 is a pAC9-derived plasmid containing wild-type ompC under lac promoter regulation (24). All strains are E. coli K-12 derivatives. AW739 expresses only OmpC (17). Strain HS110 is an AW739 derivative lacking both OmpC and OmpF and was used as the host strain for the expression of the random mutant porins. Since strains lacking both OmpC and OmpF are notoriously unstable, we constructed HS110 by P1 transduction of a sequence with a Tn10-linked ompC deletion (obtained from AW738 [17]) into AW739 in the presence of pHS10, a pMAK705-derived, replication temperature-sensitive plasmid (14) carrying an ompF gene and expressing ompF at the permissive temperature (30°C). After transformation of HS110 with a plasmid harboring a mutated porin gene (see below), cells were cultured at 42°C in order to increase the loss of the pHS10 plasmid and thus express only the mutant porin gene. This strategy allows cells to contain porins at all times during the genetic manipulations. The AD100/pNLC10 and AD101/pNLC10 strains are isogenic LacZ+ and LacZ- derivatives of AW739, similar to HS110, but were made independently and carry pNLC10 instead of pHS10. The genotype of these strains is thus ompC ompF, but they express OmpC from the plasmid after induction by IPTG addition.
Selection of OmpC porin mutants. The G195D OmpC mutant used here was obtained after random mutagenesis followed by selection for improved growth on spermine-containing plates. pNLC10 was introduced into Epicurian Coli XL1-Red cells (Stratagene) deficient in three DNA repair genes (mutS, mutD, and mutT). Two hundred microliters of the transformation mixture was plated on MLB agar (containing 50 µg of kanamycin per ml) and incubated for 36 h at 37°C. Transformants were washed from the plates with 5 ml of MLB, and a 100-µl aliquot of the mixture was inoculated into 5 ml of liquid MLB and incubated for 16 h at 37°C. Propagation of the cloned porin genes in this strain allows the accumulation of random mutations. Twenty independent preparations of mutagenized plasmid DNA were purified (Promega Wizard Minipreps) from separate cultures, and 1 µl of each mutagenized plasmid preparation was transformed into 100 µl of competent HS110 cells. The transformants were grown at 37°C overnight on MLB plates with 50 µg of kanamycin per ml. All the transformants were washed from each plate with 5 ml of MLB, and 100 µl of the mixture was inoculated into fresh liquid MLB and grown for at least 5 h at 42°C to allow the loss of the pHS10 plasmid. The cultures were then diluted and plated on a selective medium (MLB agar with 5 mM spermine, 50 µg of kanamycin per ml, and 1 mM IPTG) for 12 to 14 h at 37°C to isolate porin mutants that are resistant to polyamines (during growth). One mutant per selection plate and per DNA preparation was selected and subjected to further testing. We verified that the mutant phenotype was due to a mutation in the plasmid-borne porin gene by purifying the plasmid from the mutant cells and retransforming the plasmid into HS110. Five transformants were picked, grown at 42°C for 5 to 7 h, and plated on selection plates again. If all tested colonies showed spermine resistance, the mutant ompC porin gene was sequenced (BigDye; Perkin-Elmer Applied Biosystems) on an ABI automated sequencer to identify the mutation. The plasmid carrying the G195D mutation in OmpC (selected as described above) was named pHSC515.
Preparation of porin extracts and SDS-polyacrylamide gel electrophoresis. Cells from an overnight culture were washed in 30 mM Tris (pH 8.1) and lysed by Tris-EDTA-lysozyme treatment followed by sonication (three times for 20 s on ice with a probe sonicator). After removal of the unlysed cells, the supernatant was centrifuged again to 180,000 x g for 15 min to collect the cell envelopes. The final pellet was resuspended in 50 mM KCl-5 mM HEPES-1 mM K-EGTA (pH 7.2). The determination of protein concentration was done by the bicinchoninic acid method (Pierce). Thirty micrograms of proteins was loaded on a 12% polyacrylamide-sodium dodecyl sulfate (SDS)-6 M urea gel. The level of porin expression was quantified with an Eagle Eye densitometer and MacBas software and was expressed as a percentage of the amount of OmpA (standard control).(\xpx7, 百拇医药
Phage test. Five hundred microliters of an overnight bacterial culture was mixed with 3 ml of soft MLB agar (0.3% agar) and overlaid on regular MLB plates. After the agar lawn solidified, a 10-µl drop of diluted SS4 phage stock (104 PFU/ml) was spotted on the plate. Plaques were detected after overnight incubation at 37°C.
Antibiotic permeation assays. The periplasmic ß-lactamase was expressed from the R471a factor obtained by conjugation with the donor strain HN37 (from H. Nikaido). Conjugants were selected on plates containing 50 µg of kanamycin per ml and 100 µg of ampicillin per ml. The rates of permeation of cephaloridine in live cells were determined with a Unikon 810 double-beam spectrophotometer (Kontron Instruments) (4), and the rates are expressed as 10-4 cm/min as described previously (30). Control experiments showed no enzyme leakage (degradation rates for cell supernatants were less than 5% of the experimental values).)|#7p, http://www.100md.com
Cadaverine excretion assay. Measurements of culture supernatants for cadaverine content were done as described previously (36). The data are given as A340 values of the extract, where a value of 1.0 corresponds to ~ 250 µM cadaverine. Appropriate medium blanks were used to verify that trace amounts of other amines and amino acids present in the supernatant did not interfere with the cadaverine measurement.
Patch clamp electrophysiology. Outer membrane fractions were purified by sucrose gradient centrifugation from bacterial cells grown exponentially in LB, and patch clamp experiments were performed on lipid blisters induced from artificial liposomes containing the reconstituted outer membrane fractions (5). Control experiments were done with symmetric solutions of 150 mM KCl, 5 mM HEPES, 0.1 mM K-EDTA, and 0.01 mM CaCl2 (pH 7.2). Fifteen-milliliter solutions of spermine in the same buffer were then applied to the periplasmic side of the patch with bath perfusion. Currents were recorded with an Axopatch-1D amplifier (Axon Instruments), filtered at 1 kHz (Frequency Devices), and digitized at 100-µs sampling intervals (Instrutech).%3!^}, http://www.100md.com
Competition experiment. We constructed four strains that do not express ompF or ompC from the chromosome but that express the wild-type ompC gene from the pNLC10 plasmid or the mutant ompC gene from the pHSC515 plasmid. The strains were two LacZ+ strains, AD100/pNLC10 (C+B) and AD100/pHSC515 (CrB), and two LacZ- strains, AD101/pNLC10 (C+W) and AD101/pHSC515 (CrW), where C+ and Cr refer to the wild-type and polyamine-resistant OmpC strains, respectively, and B and W refer to blue (LacZ+) and white (LacZ-) colonies, respectively. For each competition experiment, four different mixed cultures were started by adding a 100-µl aliquot from two separate cultures incubated overnight into the same 10 ml of MLB at pH 7.0 or 5.0: culture 1, C+B mixed with CrW; culture 2, CrB mixed with C+W; culture 3, C+B mixed with C+W; and culture 4, CrB mixed with CrW. Immediately following inoculation of the cultures, a portion of each mixture was removed for cell counting. The cultures were incubated at 37°C with shaking until an optical density at 650 nm of ~ 0.6 (3 to 4 h) was reached. Then, a fresh 10-ml culture was inoculated with 100 µl of this previous culture, and an aliquot was removed again for cell counting. This cycle was repeated six times. Cell counting was done by removing an aliquot of the culture, making serial dilutions, and plating 100 µl of each dilution on LB plates containing 50 µg of kanamycin per ml, 1 mM IPTG, and 40 µg of X-Gal (5-bromo-4-chloro-3-indoyl-ß-D-galactopyranoside) per ml. The numbers of blue and white colonies on each plate were counted after overnight incubation of the plates at 37°C.
Survival rate determination. Twenty-milliliter cultures of either AD100/pNLC10 (C+B) or AD101/pHSC515 (CrW) were grown at 37°C with shaking at ~ 250 rpm in MLB (pH 6.8) containing either 0 or 20 mM cadaverine, until an optical density at 650 nm of 0.6 to 0.7 was reached. The AD100/pNLC10 culture grown in the absence of cadaverine was mixed 1:1 with the AD101/pHSC515 culture under the same conditions. The same was done for the two cultures grown in the presence of cadaverine. Each mixture was then divided in half, and the pH of one-half of the mixture was lowered to 3.6 by using 1 M HCl. This procedure resulted in four cultures, all initially containing a 1:1 proportion of C+B and CrW: two cultures at pH 6.8 with either 0 or 20 mM cadaverine and two other cultures at pH 3.6 with either 0 or 20 mM cadaverine. The four cultures were incubated at 37°C with shaking for 30 min. Serial dilutions were made and plated on MLB agar plates containing X-Gal and the appropriate antibiotics. Colonies were counted after the plates were incubated at 37°C overnight.
RESULTS AND DISCUSSIONy4y+, 百拇医药
Mutant isolation and characterization. Polyamines have been shown to act as natural modulators of ion channel activities in eukaryotes (20, 26) and prokaryotes (3, 18). In an effort to define the molecular mechanism of inhibition of bacterial porins by polyamines and the physiological significance of this type of modulation, we developed a method of selection for porin mutants that have lost sensitivity to polyamines. Although not produced endogenously, spermine was chosen for the selection because it is the most potent polyamine in the inhibition of porin. Wild-type E. coli cells grow poorly on LB plates containing 5 to 10 mM spermine. This may be due to two reasons. First, spermine is small enough to permeate porins; import of spermine from the periplasm can occur via the spermidine-preferential system PotABCD (21, 22, 41), and high intracellular concentrations of spermine are toxic because of the ability of this polycationic molecule to bind nucleic acids and proteins. Second, spermine inhibits OmpF and OmpC as it permeates porins and as it interacts with the inwardly folded constriction loop (19); as inhibition occurs, cells experience a deficit in nutrient flux that slows growth on plates.
Our goal was to isolate porin mutants that allow growth in the presence of the polyamine, either because they provide reduced spermine entry, reduced porin inhibition by spermine, or both. Mutants which have lost porin inhibition by polyamines would give us a tool to address the physiological significance of this form of modulation. For this purpose, we subjected an ompC-carrying plasmid (pNLC10) to random mutagenesis and introduced it into a host cell deficient in OmpC and OmpF. Seven mutants displaying growth on spermine plates were obtained; six were shown to have a single missense mutation and one was an ochre mutation. Four mutants (with A129P, Q175ochre, D204N, and S274P mutations) displayed a reduced porin level (<25% of that of the wild type), and two others (with V96D and G251D mutations) showed polyamine resistance in growth but not in antibiotic flux or in electrophysiology. Presumably, these mutants were selected only because they provided a reduced level of entry of spermine inside the cell due to reduced porin levels or impaired pore properties. The G195D mutant is the only mutant that maintained reasonable levels of expression and displayed a robust polyamine resistance phenotype in flux assays and was therefore used in this study to investigate the physiological significance of porin inhibition by polyamines. Residue G195 is located in the barrel wall adjacent to the L3 loop and is in close proximity to this loop. As the conformation of the L3 loop has been shown to be important for polyamine inhibition (19), it is possible that the G195D mutant lost polyamine sensitivity because of secondary effects on the L3 loop conformation.
Several phenotypes of the G195D mutant were compared to those of the same strain carrying pNLC10 with the wild-type ompC. The pNLC10 plasmid was initially constructed to provide OmpC levels that are comparable to those of the haploid wild-type strain (24). Although the porin levels were 54% ± 12% of the levels of wild-type porin expressed from the wild-type plasmid in the presence of IPTG (porin levels are <5% in the absence of IPTG) and the sensitivity to phage SS4 was slightly reduced, the rates of cephaloridine flux were comparable for both strains (expressed as 10-4 cm/min): 5.7 ± 0.3 for the wild type (nine samples) and 6.3 ± 0.3 for the mutant (three samples). It is known that ß-lactam antibiotics such as cephaloridine require porins for entry, and thus the flux of these antibiotics is a reliable indication of the permeability of the outer membrane (30). When flux assays were performed in the presence of external polyamines, a decrease in cephaloridine flux through the wild-type porin was clearly seen, as previously reported (4), with 26% ± 9% inhibition in the presence of 1 mM external spermine (six samples) and 31% ± 1% inhibition in the presence of 200 mM external cadaverine (three samples). As exogenously added cadaverine is less potent than spermine, we purposely chose to add 200 mM cadaverine externally to obtain a substantial reduction in flux. We previously showed that much smaller concentrations of cadaverine cause much larger effects when the polyamine exerts its effect from the periplasmic side (18, 36). On the contrary, the G195D mutant displayed only a 5% ± 3% and a 7% ± 2% inhibition in the presence of 1 mM spermine and 200 mM cadaverine, respectively (three samples for each). The relative flux values obtained with the G195D mutant in the presence of polyamines were always significantly greater than in the wild-type strain (P < 0.05, two-sample t test). It is important to point out that the inhibitory mechanism, as revealed by electrophysiology, mostly involves the complete closure of the affected porin molecules rather than the partial or total plugging of the pores by the polyamine molecules. Thus, the lack of inhibition seen in the mutant porin is due to the insensitivity of the mutant porin to the polyamines rather than to a compensatory mechanism originating from the fact that the mutant pore might be wider (since the total amount of antibiotic flux is comparable to that of the wild-type strain, despite the reduction in porin number).
This insensitivity to modulation by polyamine was confirmed with the very sensitive technique of patch clamp electrophysiology. shows typical recordings of currents from membrane patches containing many wild-type or mutant porins. Because porins are open most of the time under control conditions, the current trace dwells at a baseline level, from which upward transitions represent the rapid and transient spontaneous closures of one or many channels. This electrophysiological behavior has been extensively studied (6), but its details are not the subject of this publication. Here, we qualitatively point out the differences that wild-type and mutant channels exhibit towards inhibition by spermine. shows that the wild-type channels displayed a greatly modified activity in the presence of 1 mM spermine. The frequency and duration of channel closures (upward deflections) were greatly increased, and the current trace dwelled at various levels, depending on how many channels were closed. This inhibitory action of spermine was not observed at all with the G195D mutant, since the kinetic signature of the current trace was essentially unchanged in the presence of the polyamine. These results were confirmed in three independent experiments. Together with the antibiotic-flux assays, these results show that the inhibition of porin by polyamine was abolished in the G195D mutant.
fig.ommittedi.45, http://www.100md.com
FIG. 1. Spermine insensitivity of the G195D OmpC mutant. Representative patch clamp recordings are shown for the wild-type and the mutant in the absence (control [CON]) and the presence of 1 mM spermine (SPM). BL indicates the baseline level representing the total current flowing through multiple OmpC monomers. Upward deflections represent transient closures of channels. The pipette voltage was -60 mV.i.45, http://www.100md.com
Survival of the mutant strain at acidic pH. One of the cellular responses of E. coli cells to acidic pH is the excretion of the polyamine cadaverine. The role that this molecule may play in adaptation to pH has remained elusive. The secretion of cadaverine leads to a rapid and sustained inhibition of porins in vivo (36). Cadaverine produced by the cell is able to act from the periplasm, and we have shown that cadaverine applied from the periplasmic side is much more effective than exogenously added cadaverine (18, 36). Our previous results showed that the inhibition of outer membrane permeability by secreted cadaverine is not complete and leaves enough pathways for nutrient entry; thus, the production of cadaverine itself is not deleterious for the cell. To understand whether a physiological link ties porin modulation by cadaverine to the secretion of cadaverine at low pH, we compared the G195D mutant cells to wild-type cells for their ability to survive a treatment at pH 5.0. We purposely chose this pH for the following reasons: (i) cadA expression is maximal at pH 5.0 to 6.0 (38); and (ii) our goal was to monitor cells while they were being challenged by an acid pH at which they could still grow and adapt to the conditions rather than to look for protection against the extreme acid stress at which cell death would occur.
For this, we performed competition experiments where wild-type and mutant cells were mixed together in a single flask and grown until mid-log phase. At the end of this growth, a cell count was performed, the pH of the culture was measured, the amount of cadaverine released was estimated, and an aliquot of this culture was used to start a fresh culture. This cycle was repeated six times. Thus, relative survival rates of wild-type cells and mutant cells were compared as the cultures went through several generations. To distinguish wild-type cells from mutant cells, we constructed LacZ+ and LacZ- marked strains that give rise to blue or white colonies on X-Gal plates, respectively. and B show the averaged results of three experiments. The ordinate axes represent the percentages of blue colonies, obtained by dividing the number of blue colonies by the total number of colonies on each plate; error bars are standard deviations. The data were normalized across experiments by setting the initial percentage of blue colonies to 50 for all curves (cycle 0). In each experiment, four cultures were investigated (as described in Materials and Methods): culture 1, C+B mixed with CrW; culture 2, CrB mixed with C+W; culture 3, C+B mixed with C+W; and culture 4, CrB mixed with CrW. Thus, each culture was internally controlled; the results from cultures 1 and 2 were complementary, and cultures 3 and 4 served as controls.
fig.ommitted}, 百拇医药
FIG. 2. Competition experiments. Measurements were performed on cultures grown at pH 7.0 (A, C, and E) and pH 5.0 (B, D, and F). (A and B) Averaged percentages of blue colonies in three separate experiments (error bars are standard deviations) after all values at cycle 0 were normalized to 50% to compare the results of the three experiments. The actual percentages at cycle 0 ranged between 40 and 57% for all three experiments. (C and D) pHs of the four cultures throughout a single experiment. (E and F) Amount of cadaverine excreted by the cells, as measured by the A340 in the cadaverine assay in a single experiment. For all panels, the cultures consisted of the following: culture 1, C+B cells mixed with CrW cells ; culture 2, CrB cells mixed with C+W cells ; culture 3, C+B cells mixed with C+W cells ; and culture 4, CrB cells mixed with CrW cells (where C+ and Cr refer to the wild-type and polyamine-resistant OmpC strains, respectively, and B and W refer to blue [LacZ+] and white [LacZ-] colonies, respectively).
through F show the results of the pH and cadaverine measurements for a single experiment. The pHs remained as set, at 7.0 and 5.0, for all four cultures throughout the cycles. The amount of cadaverine excreted by cells was negligible at pH 7.0 , but a significant amount was excreted when cells were grown at pH 5.0 . In this case, as the cultures went through more cycles, the excretion of cadaverine increased until the A340 reached about 1.0 (about 250 µM cadaverine) in the third cycle, where it remained for the rest of the experiment. shows that the number of blue colonies hovered around 50% throughout the entire experiment at pH 7.0 with no clear trend of the mutant or the wild type being preferred. Thus, in neutral environments, there was no excretion of cadaverine and the wild-type and mutant cells had equal levels of fitness. On the contrary, at pH 5.0, the percentage of wild-type cells increased and the percentage of mutant cells decreased as the culture went through more cycles. This trend became more significant after the fourth cycle. By the end of the sixth cycle, the percentage of wild-type cells was greater than 70% and the percentage of mutants was below 25%. However, the percentage of blue colonies in both control sets remained at ~ 50%. The wild type is clearly at an advantage at pH 5.0, unlike at pH 7.0.
We have previously found that the excretion of up to ~ 250 µM cadaverine correlates with an ~ 70% inhibition of antibiotic flux mediated by OmpF and OmpC (36). Thus, the measured levels of excreted cadaverine are effective to promote inhibition of the wild-type channel. We believe that most of this inhibition occurs while cadaverine transits through the periplasm, as it persists even after the external cadaverine is washed (36). Note that much higher concentrations of externally supplied cadaverine were needed in the characterization of the mutants by antibiotic flux assays, because cadaverine is much more potent when it is applied to the periplasmic side than to the extracellular side of porin molecules (18).;8, 百拇医药
These results suggest that the inhibition of porins by excreted cadaverine may be an adaptive mechanism that helps the E. coli cells survive acidic environments. It is important to note that the only significant difference between the wild-type and mutant cells is the polyamine-resistant phenotype of their OmpC porin. The reduced porin levels, as detected by SDS-polyacrylamide gel electrophoresis, are likely to be inconsequential since the flux efficiency is similar to that of the wild-type cells and since the mutant strain was not outcompeted at neutral pH. In addition, although the activation of CadC at low pH is known to decrease porin expression (36), this is not relevant in these experiments since porin expression is solely driven from the IPTG-dependent plasmid promoter. Finally, it is also important to consider whether the sensitivity of the mutant to acidic pH is due to the inability of this porin mutant to close at low pH. There have been numerous reports that acidic pH triggers closure of wild-type porins in electrophysiological experiments (25, 35, 42, 43). We have confirmed this mode of porin regulation in vivo by comparing the flux of cephaloridine at pH 3.6 with that at pH 6.0 (in the absence of added polyamines). Despite a 10-fold decrease in the activity of the released ß-lactamase, compared to levels at pH 6.0, antibiotic flux in the wild-type and G195D mutant cells at pH 3.6 was reduced 22% ± 3% and 26% ± 4%, respectively (three samples), indicating that the rate of antibiotic flux through porins was still rate limiting. Thus, the extents of inhibition by acidic pH of porin-dependent permeability were similar in wild-type and mutant cells, suggesting that the mutant porins had not lost their sensitivity to pH.
Cadaverine and pH sensitivity. The results of the competition experiments are substantiated by our finding that cadaverine supplied exogenously helps cells carrying the wild-type OmpC porin survive a 30-min exposure to acidic pH better than cells carrying the G195D mutant OmpC porin. Here, we aimed at obtaining some cell death; thus, a pH of 3.6 was chosen, despite the fact that this pH would produce little excreted cadaverine in vivo. Cultures containing LacZ+ and LacZ- cells expressing the wild-type and mutant OmpC porin, respectively, were grown in parallel in LB and in LB supplemented with 20 mM cadaverine to allow the accumulation of cadaverine in the periplasm and the inhibition of porins. Then we mixed equal amounts of LacZ+ and LacZ- cells expressing the wild-type and mutant OmpC porin, respectively, so that the experiments were internally controlled. Each mixture was exposed to pH 3.6 for 30 min, still in the presence or absence of 20 mM cadaverine. After the plating of the cells, the ratio of blue (wild-type) to white (mutant) colonies was calculated for each condition. We found that there were 16 ± 6 times more wild-type cells than mutant cells surviving acidic pH when cadaverine was present (three samples) (this value is not directly comparable to those of Fig. 2 because of the large difference in the pHs used in the two sets of experiments). In the absence of cadaverine, the ratio of wild-type to mutant cells after the acid treatment was 0.9 ± 0.5. Similarly, a ratio of ~ 1.0 was consistently found when the mixed cells remained at pH 6.8, regardless of the presence of cadaverine. Thus, in the absence of cadaverine, the survival of wild-type and that of mutant cells were equally affected. However, wild-type cells were able to survive acidic pH better than mutant cells specifically when cadaverine was present.
Conclusions. We have identified a mutation in the ompC gene that confers resistance to the inhibition of porin function by cadaverine. Taken together, our results suggest that this resistance of the mutant OmpC porin to cadaverine inhibition is correlated with the mutant cells being outcompeted by wild-type cells only in an acidic milieu and not at neutral pH. Thus, the excretion of cadaverine at low pH appears to play a role in the adaptation of cells to these conditions.x, 百拇医药
It is noteworthy that cells reduce their outer membrane permeability under acidic conditions by the use of multiple pathways that may act sequentially. The pH-induced closure of porins is immediate (25, 35, 42) and will rapidly lead to about a 20 to 25% reduction in permeability. This reduction is followed by the slower processes of regulation of gene expression, which affects the porin regulon and the cad operon. Acidic pH promotes the decrease of porin levels in the outer membrane through both an envZ-dependent pathway (15) and a novel mechanism involving the cad operon regulator CadC (36). Finally, the secretion of cadaverine as a result of increased CadA and CadB levels leads to the functional inhibition of up to ~ 70% of the porin-mediated permeability (36). This multiplicity of mechanisms points to the importance of reducing outer membrane permeability under acidic conditions, although it is difficult to assess whether such a reduction permits a better control of periplasmic composition. The work presented here, however, demonstrates that the impairment of one such pathway leads to a deficit in cell survival; thus, a tight control of influx—and possibly efflux—across the outer membrane is required for optimal adaptation to acidic conditions. It appears that cells might find a compromise whereby the incomplete shutdown of the porin-mediated pathways still provides sufficient nutrient flux, while reducing the amplitude or rate of acid flux to confer a survival advantage over time. The ~ 70% decrease in permeability found in the presence of secreted cadaverine (36) does not seem to impair the growth rate significantly in rich laboratory media. But by the same token, it appears that a minimum permeability is required, as porin-null mutants rapidly acquire compensatory mutations. Therefore, the necessity for striking the right balance in the interplay of nutrient and acid flux might be of even more significance under conditions of nutrient limitation, as is the case outside the host, than in rich media. The secretion of cadaverine is not the only example of regulation of polyamine extrusion in response to environmental conditions. A rapid rise in external osmolarity promotes the excretion of cellular putrescine (28), which inhibits porin-mediated fluxes (4). The use of natural modulators of porin activity to control the permeability of the outer membrane may be a general strategy used by cells in response to deleterious external conditions.
ACKNOWLEDGMENTS6]l1r8l, 百拇医药
We acknowledge Aparna Kumaraswamy for help with some of the experiments. We thank Linda Guynn for sequencing the mutant porin genes and William Widger for the use of the spectrophotometer.6]l1r8l, 百拇医药
This work was supported by NIH grant AI34905.6]l1r8l, 百拇医药
REFERENCES6]l1r8l, 百拇医药
Booth, I. R. 1985. Regulation of cytoplasmic pH in bacteria. Microbiol. Rev. 49:359-378.6]l1r8l, 百拇医药
Conner, D. E., and J. S. Kotrola. 1995. Growth and survival of Escherichia coli O157:H7 under acidic conditions. Appl. Environ. Microbiol. 61:382-385.6]l1r8l, 百拇医药
delaVega, A. L., and A. H. Delcour. 1995. Cadaverine induces closing of E. coli porins. EMBO J. 14:6058-6065.6]l1r8l, 百拇医药
delaVega, A. L., and A. H. Delcour. 1996. Polyamines decrease Escherichia coli outer membrane permeability. J. Bacteriol. 178:3715-3721.6]l1r8l, 百拇医药
Delcour, A. H., B. Martinac, C. Kung, and J. Adler. 1989. A modified reconstitution method used in patch-clamp studies of Escherichia coli ion channels. Biophys. J. 56:631-636.
Delcour, A. H. 1997. Function and modulation of bacterial porins: insights from electrophysiology. FEMS Microbiol. Lett. 151:115-123.{emoz, http://www.100md.com
DiRita, V. J. 1995. Three-component regulatory system controlling virulence in Vibrio cholerae, p. 351-365. In J. A. Hoch and T. J. Silhavy (ed.), Two-component signal transduction. ASM Press, Washington, D.C.{emoz, http://www.100md.com
Dziejman, M., and J. J. Mekalanos. 1995. Two-component signal transduction and its role in the expression of bacterial virulence factors, p. 305-317. In J. A. Hoch and T. J. Silhavy (ed.), Two-component signal transduction. ASM Press, Washington, D.C.{emoz, http://www.100md.com
Foster, J. W. 1999. When protons attack: microbial strategies of acid adaptation. Curr. Opin. Microbiol. 2:170-174.{emoz, http://www.100md.com
Gorden, J., and P. L. C. Small. 1993. Acid resistance in enteric bacteria. Infect. Immun. 61:364-367.{emoz, http://www.100md.com
Groisman, E. A., and F. Heffron. 1995. Regulation of Salmonella virulence by two-component regulatory systems, p. 319-332. In J. A. Hoch and T. J. Silhavy (ed.), Two-component signal transduction. ASM Press, Washington, D.C.
Guilfoyle, D. E., and I. N. Hirshfield. 1996. The survival benefit of short-term organic acids and the inducible arginine and lysine decarboxylase genes for Escherichia coli. Lett. Appl. Microbiol. 22:393-396.w9h$, 百拇医药
Hall, H. K., K. L. Karem, and J. W. Foster. 1995. Molecular responses of microbes to environmental pH stress. Adv. Microb. Physiol. 37:229-272.w9h$, 百拇医药
Hamilton, C. M., M. Aldea, B. K. Washburn, P. Babitzke, and S. R. Kushner. 1989. New method for generating deletions and gene replacements in Escherichia coli. J. Bacteriol. 171:4617-4622.w9h$, 百拇医药
Heyde, M., and R. Portalier. 1987. Regulation of major outer membrane porin proteins of Escherichia coli K-12 by pH. Mol. Gen. Genet. 208:511-517.w9h$, 百拇医药
Hill, C., B. O'Driscoll, and I. Booth. 1995. Acid adaptation and food poisoning microorganisms. Int. J. Food Microbiol. 28:245-254.w9h$, 百拇医药
Ingham, C., M. Buechner, and J. Adler. 1990. Effect of outer membrane permeability on chemotaxis in Escherichia coli. J. Bacteriol. 172:3577-3583.
Iyer, R., and A. H. Delcour. 1997. Complex inhibition of OmpF and OmpC bacterial porins by polyamines. J. Biol. Chem. 272:18595-18601.78, http://www.100md.com
Iyer, R., Z. Wu, P. M. Woster, and A. H. Delcour. 2000. Molecular basis for the polyamine—OmpF porin interactions: inhibitor and mutant studies. J. Mol. Biol. 297:933-945.78, http://www.100md.com
Johnson, T. D. 1996. Modulation of channel function by polyamines. Trends Pharmacol. Sci. 17: 22-27.78, http://www.100md.com
Karahalios, P., P. Mamos, D. H. Vynios, D. Papaioannou, and D. L. Kalpaxis. 1998. The effect of acylated polyamine derivatives on polyamine uptake mechanisms, cell growth, and polyamine pools in Escherichia coli, and the pursuit of structure/activity relationships. Eur. J. Biochem. 251:998-1004.78, http://www.100md.com
Kashiwagi, K., H. Kobayashi, and K. Igarashi. 1986. Apparently unidirectional polyamine transport by proton motive force in polyamine-deficient Escherichia coli. J. Bacteriol. 165:972-977.78, http://www.100md.com
Levina, N., S. Tötemeyer, N. R. Stokes, P. Louis, M. A. Jones, and I. R. Booth. 1999. Protection of Escherichia coli cells against extreme turgor by activation of MscS and MscL mechanosensitive channels: identification of genes required for MscS activity. EMBO J. 18:1730-1737.
Liu, N., M. J. Benedik, and A. H. Delcour. 1997. Disruption of polyamine modulation by a single amino acid substitution on the L3 loop of the OmpC porin channel. Biochim. Biophys. Acta 1326:201-212.2au00j6, http://www.100md.com
Liu, N., and A. H. Delcour. 1998. Inhibitory effect of acidic pH on OmpC porin: wildtype and mutant studies. FEBS Lett. 434:160-164.2au00j6, http://www.100md.com
Lopatin, A., E. N. Makhina, and C. G. Nichols. 1994. Potassium channel blocking by cytoplasmic polyamines as the mechanism of intrinsic rectification. Nature 372:366-369.2au00j6, http://www.100md.com
Merrell, D. S., and A. Camilli. 1999. The cadA gene of Vibrio cholerae is induced during infection and plays a role in acid tolerance. Mol. Microbiol. 34:836-849.2au00j6, http://www.100md.com
Munro, G. F., K. Hercules, J. Morgan, and W. Sauerbier. 1972. Dependence of the putrescine content of Escherichia coli on the osmotic strength of the medium. J. Biol. Chem. 247:1272-1280.2au00j6, http://www.100md.com
Nikaido, H. 1996. Outer membrane, p. 29-47. In F. C. Neidhardt et al. (ed.), Escherichia coli and Salmonella: cellular and molecular biology, 2nd ed., vol. 1. ASM Press, Washington, D.C.
Nikaido, H., E. Y. Rosenberg, and J. Foulds. 1983. Porin channels in Escherichia coli: studies with ß-lactams in intact cells. J. Bacteriol. 153:232-240.2#!f, http://www.100md.com
Ninfa, A. J. 1996. Regulation of gene transcription by extracellular stimuli, p. 1246-1262. In F. C. Neidhardt et al. (ed.), Escherichia coli and Salmonella: cellular and molecular biology, 2nd ed., vol. 1. ASM Press, Washington, D.C.2#!f, http://www.100md.com
Olson, E. 1993. Influence of pH on bacterial gene expression. Mol. Microbiol. 8:5-14.2#!f, http://www.100md.com
Pratt, L. A., and T. J. Silhavy. 1995. Porin regulon of Escherichia coli, p. 105-127. In J. A. Hoch and T. J. Silhavy (ed.), Two-component signal transduction. ASM Press, Washington, D.C.2#!f, http://www.100md.com
Rhee, J. E., J. H. Rhee, P. Y. Ryu, and S. H. Choi. 2002. Identification of the cadBA operon from Vibrio vulnificus and its influence on survival to acid stress. FEMS Microbiol. Lett. 208:245-251.2#!f, http://www.100md.com
Saint, N., A. Prilipov, A. Hardmeyer, K.-L. Lou, T. Schirmer, and J. P. Rosenbusch. 1996. Replacement of the sole histidinyl residue in OmpF porin from E. coli by threonine (H21T) does not affect channel structure and function. Biochem. Biophys. Res. Commun. 223:118-122.
Samartzidou, H., and A. H. Delcour. 1999. Excretion of endogenous cadaverine leads to a decrease in porin-mediated outer membrane permeability. J. Bacteriol. 181:791-798.3k, 百拇医药
Samartzidou, H., and A. H. Delcour. 1999. Distinct sensitivities of OmpF and PhoE to charged modulators. FEBS Lett. 444:65-70.3k, 百拇医药
Slonczewski, J. L., T. N. Gonzalez, F. M. Bartholomew, and N. J. Holt. 1987. Mu d-directed lacZ fusions regulated by low pH in Escherichia coli. J. Bacteriol. 169:3001-3006.3k, 百拇医药
Slonczewski, J. L., and J. W. Foster. 1996. pH-regulated genes and survival at extreme pH, p. 1539-1549. In F. C. Neidhardt et al. (ed.), Escherichia coli and Salmonella: cellular and molecular biology, 2nd ed., vol. 1. ASM Press, Washington, D.C.3k, 百拇医药
Sukharev, S. I., P. Blount, B. Martinac, and C. Kung. 1997. Mechanosensitive channels of Escherichia coli: the MscL gene, protein and activities. Annu. Rev. Physiol. 59:633-657.3k, 百拇医药
Tabor, C. W., and H. Tabor. 1966. Transport systems for 1,4-diaminobutane, spermidine and spermine in Escherichia coli. J. Biol. Chem. 241:3714-3723.3k, 百拇医药
Todt, J. C., W. J., Rocque, and E. J. McGroarty. 1992. Effects of pH on bacterial porin function. Biochemistry 31:10471-10478.3k, 百拇医药
Xu, G., B. Shi, E. J. McGroarty, and H. T. Tien. 1986. Channel-closing activity of porins from Escherichia coli in bilayer lipid membranes. Biochim. Biophys. Acta 862:57-64.3k, 百拇医药
Zilberstein, D., V. Agmon, S. Schuldiner, and E. Padan. 1984. Escherichia coli intracellular pH, membrane potential, and cell growth. J. Bacteriol. 158:246-252.(Hrissi Samartzidou Mahsa Mehrazin Zhaohui Xu Michael J. Benedik and Anne H. Delcour)