• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of jbacterPermissionsJournals.ASM.orgJournalJB ArticleJournal InfoAuthorsReviewers
J Bacteriol. Apr 2005; 187(7): 2297–2307.
PMCID: PMC1065229

Role of the Extracytoplasmic Function Protein Family Sigma Factor RpoE in Metal Resistance of Escherichia coli

Abstract

RpoE of Escherichia coli is a sigma factor of the extracytoplasmic function protein family and is required for the expression of proteins involved in maintaining the integrity of periplasmic and outer membrane components. RpoE of E. coli was needed for full resistance to Zn(II), Cd(II), and Cu(II). Promoter gene fusion and quantitative real time reverse transcription (RT)-PCR (qRT-PCR) assays demonstrated that expression of RpoE was induced by metals. Global gene expression profiles upon metal treatment of a ΔrpoE mutant strain and its wild-type strain were analyzed with microarrays, and selected genes were confirmed by qRT-PCR. The absolute number of genes that were changed in their expression upon metal stress was similar in both strains, but the increase or decrease in transcript levels upon metal treatment was smaller in the ΔrpoE mutant strain than in the wild type. Genes showing increased expression in the ΔrpoE mutant strain encoded proteins that belong to general defense systems against protein-denaturing agents. Genes showing decreased expression were part of the RpoE modulon itself plus the ompC gene, encoding a major outer membrane protein. A ΔompC deletion strain was as sensitive to Cu(II) and Cd(II) as the ΔrpoE mutant or a ΔrpoE ΔompC double mutant strain. In the case of Zn(II), the double mutant was more sensitive than either single mutant. This indicates that increased expression of OmpC contributes to the RpoE modulon-mediated response to metals.

Sigma factors of the extracytoplasmic function family are part of the bacterial stress response regulon (18, 43). They react to stress signals outside the cytoplasmic membrane by transcriptional activation of genes encoding products involved in defense or repair processes (15, 47, 54). RpoE from Escherichia coli, present at six (± three) copies per cell (38), is a paradigm of these sigma factors (54). The products of the genes controlled by RpoE (RpoE modulon) are required for proper folding of outer membrane proteins and their turnover, phospholipid and lipopolysaccharide biosynthesis, signal transduction, expression of putative inner and outer membrane proteins (57), other envelope proteins such as DsbC, FkbA, Skp, and SurA, and recently identified Ecf proteins involved in extracytoplasmic function that were essential for growth of E. coli (15). RpoE might be even more important for survival of E. coli in the stationary phase than the “starvation” sigma factor RpoS (65). Because of the essential character of RpoE, ΔrpoE mutant cells seem to acquire a suppression that allows them to grow (19). However, the suppressor mutation has never been identified, although some essential genes under RpoE control are known (15).

RseA and RseB, encoded together with RpoE in the rpoE-rseABC operon, form a signal pathway that allows E. coli to respond to protein unfolding upon periplasmic or envelope stress, especially under heat shock conditions that lead to controlled proteolysis of RseA (3). Under nonstress conditions, the membrane-bound anti-sigma factor RseA sequesters RpoE through its N-terminal domain, thereby decreasing the cytoplasmic availability of RpoE for transcription initiation (14). The C-terminal domain of RseA interacts with the periplasmic protein RseB, which is present in about half as many copies per cell (about three) as RpoE. RseB binds to misfolded periplasmic proteins. Since RseB increases the affinity of RseA for RpoE 2.5- ± 0.5-fold, this binding event might titrate RseB away from RseA and lead to a release of RpoE from RseA (14).

Bacterial metal resistance is the result of an interplay of several metal efflux systems, which can mainly be assigned to the protein families RND (resistance, nodulation, cell division), CDF (cation diffusion facilitator), and P-type ATPases (46). There is accumulating evidence that this interplay is a two-step process that affects both cellular compartments, the cytoplasmic and the periplasmic space. In gram-negative bacteria, P-type ATPases and CDF proteins export their substrates across the cytoplasmic membrane into the periplasm. While P-type ATPases focus on the sulfur-loving elements Ag(I) and Cu(I) or Zn(II), Cd(II), and Pb(II), the CDF proteins transport the divalent metal cations of the first transition period from Mn(II) to Zn(II) plus Cd(II). In the second step, RND-driven transenvelope efflux complexes seem to export the cations from the periplasm across the outer membrane (46).

E. coli does not contain an RND-driven transenvelope efflux system for Zn(II) or Cd(II) export but detoxifies these two metal cations with the P-type ATPase ZntA and the CDF protein ZitB (23, 46, 56). On the other hand, copper homeostasis in this bacterium involves three components, CopA, CueO, and Cus. The P-type ATPase CopA transports Cu(I) from the cytoplasm to the periplasm (55, 56). The toxic effects of the presence of Cu(I) in the latter compartment, which probably originate from the oxidation of Cu(I) to Cu(II), are relieved by the multicopper oxidase CueO (24, 26, 58). The third copper homeostasis system, CusCBA/CusF, becomes important in a ΔcueO deletion mutant and under anaerobic conditions because CueO is probably dependent on oxygen for its function (51). CusCBA is an RND-driven copper transenvelope efflux system and CusF is a periplasmic auxiliary small chaperone for the efflux process (21, 22, 44). Copper stress is relieved in a ΔcueO deletion strain by the Cus transenvelope efflux system, which provides evidence for the importance of the RND-driven Cus system for periplasmic detoxification of copper ions (25). This marks the periplasm as an important cellular compartment for metal homeostasis in a way that has not been considered before. This study elucidates the contribution of the RpoE modulon to metal homeostasis in E. coli.

MATERIALS AND METHODS

Bacterial strains and growth conditions.

All strains used in this study were derivatives of E. coli strain BW25113 (16) (see Table 8 in the supplemental material). For cultivations, either Tris-buffered mineral salts medium (40) containing 0.2% glycerol and 3 g/liter Casamino Acids (TGC) or Luria-Bertani broth (LB) (60) was used. Analytical-grade salts of metal chlorides were used to prepare 1 M stock solutions, which were sterilized by filtration. Solid Tris-buffered media contained 20 g agar/liter.

Genetic techniques.

Standard molecular genetic techniques were used (45, 60). The ΔrpoE deletion strain was constructed with the gene deletion system described by Datsenko and Wanner (16). In the first step, a chloramphenicol resistance cassette was amplified from plasmid pKD3 with the primers FRT-rpoE-Ec-down and FRT-rpoE-Ec-up (Table 8 in the supplemental material). Lambda Red recombinase-mediated recombination was used to replace the rpoE gene in E. coli BW25113 with the cat gene. In a second step, the cat gene was deleted from the chromosome. Generalized P1 phage transduction was performed with the lysate from ΔcopA::cat, ΔcueO::cat, and ΔcusCFBA::cat strains (25) and from ΔzitB::cat and ΔzntA::cat strains to yield multiple deletion strains.

For complementation of the ΔrpoE mutant strain, a 2,720-bp fragment containing the rpoE promoter, rpoE, rseA, and rseB was amplified from total DNA of E. coli K-12 with primers Ec-rpoE-Pst and rseB (Ec) pASK-Pst (Table 8 in the supplemental material), cloned into plasmid pAH125, and integrated into the chromosome of strain ECA101 as described (28). To construct promoter fusions with the lacZ gene, 500-bp promoter fragments of the genes rpoE (59) and cueR, cueO, and copA were amplified from total DNA of E. coli W3110 with primers Ec-rpoEp-Pst and Ec-rpoEp-Eco, cueRp-Pst and cueRp-Eco, cueOp-Pst and cueOp-Eco, and copAp-Pst and copAp-Eco, respectively (Table 8 in the supplemental material). These fragments were also cloned into lacZ reporter plasmid pAH125 and integrated into the chromosome of E. coli BW25113 as described (28).

Growth curves.

Overnight cultures of the E. coli strains were diluted 1:400 into LB medium and incubated 2 h with shaking at 37°C. These cultures were diluted again 1:400 into fresh LB medium to give parallel cultures representing the test conditions and controls. These cultures were incubated at 37°C until stationary phase was reached. Growth was monitored as turbidity with a Klett photometer. To determine the plating efficiency, samples were diluted in LB, 0.1-ml volumes were plated onto LB plates, and the CFU were counted after 16 h of incubation at 37°C.

Dose-response curves describing the action of Cu2+, Zn2+, or Cd2+ on E. coli cells were also performed in LB medium. Overnight cultures of the E. coli strains were diluted 1:400 to inoculate parallel cultures with increasing metal concentrations. The cells were cultivated for 16 h with shaking at 37°C, and the turbidity was determined at 600 nm. The mean values of the data from at least three independent experiments were used to determine the 50% inhibitory concentration (IC50) values, which are the concentrations required to diminish cell density by 50% under the conditions tested. The formula used was the simplified version OD(c) = OD0/(1 + exp[(c − IC50)/b)] of a Hill-type equation as introduced by Pace and Scholtz (52) and described previously (5). OD(c) is the turbidity at a given metal concentration, OD0 is that at no added metal, b is the slope of the sigmoidal dose response curve, and c is the metal concentration.

Induction experiments.

E. coli cells with lacZ reporter gene fusions were cultivated in TGC medium with shaking at 37°C. After 2 h, metal chloride was added at various final concentrations, and the cells were incubated with shaking for a further 2 h. Promoter activity was measured as β-galactosidase activity as described previously (41).

Microarrays of E. coli.

E. coli cells, wild-type and ΔrpoE mutant strains, were treated for 10 min with 25 μM Cd(II), 100 μM Zn(II), or 250 μM Cu(II) or no metal as a control in TGC medium. The exposure time used was long enough to allow complete transcription of all E. coli operons up to 24 kb in size. However, it was short enough to see the initial response of the cells to the metal shock. The concentration was chosen to be high enough to see an effect at all but low enough to prevent artifacts resulting from a global metabolic breakdown rather than from a specific response of the cells to the metal treatment.

PAN E. coli K-12 V2 arrays (MWG Biotech, Ebersberg, Germany) were used, consisting of 4,239 gene-specific oligonucleotide spots and 48 spots with DNA from Arabidopsis thaliana that served as a negative control. Each set of conditions was performed four times, with two independent bacterial cultures and a dye swap according to the minimum information about a microarray experiment (9) protocol (see the supplemental material).

RNA isolation and preparation of labeled cDNA.

Total RNA was isolated as described (27). After isolation, DNase treatment was performed followed by purification with RNeasy columns (Qiagen, Hilden, Germany). To exclude experimental artifacts resulting from DNA contamination, only RNA preparations that did not generate PCR fragments in a PCR with chromosomal primers without a previous reverse transcription reaction were used. RNA concentration was determined photometrically, and RNA quality was checked on formamide gels (60). In a reverse transcription reaction, 50 μg of total RNA was labeled in a 40-μl labeling reaction with 9 μg of hexamer primers, 50 μM either dCTP labeled with either Cy3 or Cy5 (Amersham, Freiburg, Germany), 0.5 mM each dATP, dGTP, and dTTP, 0.2 mM nonlabeled dCTP, 10 mM dithiothreitol, and 200 U of reverse transcriptase in reaction buffer (Superscript II, Invitrogen, Karlsruhe, Germany). Primers and RNA were heated to 70°C for 5 min and snap-cooled in ice. Reverse transcription proceeded for 10 min at room temperature, followed by 2 h at 42°C. To denature the remaining RNA 10 μl of 1 M NaOH was added and incubated for 10 min at 65°C, followed by addition of 10 μl of 1 M HCl. Nonincorporated fluorescent nucleotides were removed with a Qiaquick PCR purification kit (Qiagen, Hilden, Germany). The amount of cDNA and the integration of fluorescent dye were determined as published (8).

Hybridization of DNA microarrays and image analysis.

Equal amounts of Cy3- and Cy5-labeled cDNA in 120 μl of hybridization buffer (salt-based, MWG Biotech, Ebersberg, Germany) were denatured for 3 min at 94°C and hybridized to the PAN E. coli K-12 V2 microarrays for 20 h at 42°C in a shaking waterbath. After hybridization, the arrays were washed in 2× SSC (1× SSC is 0.15 M NaCl plus 0.015 M sodium citrate)-0.1% sodium dodecyl sulfate and then in 1× SSC and 0.1× SSC. The slides were scanned with a laser scanner (Array Scanner 428, Affymetrix) seven times per channel with increasing photomultiplier settings to expand the dynamic range of measurements. The resulting images were analyzed with the ImaGene 4.2 software (BioDiscovery, Inc. El Segundo, Calif.). An average of the sensitivities was calculated by linear regression with MAVI Pro 2.6.0 software (MWG Biotech, Ebersberg, Germany).

Normalization.

For normalization and obtaining ratios, two strategies were used, which both led to identical results. First we performed the algorithm with GeneSight version 3.0 (BioDiscovery). The signal intensities were background subtracted and normalized by the total array intensity. Scatter plot and histogram analysis were performed to obtain the ratio for each experiment. Second, our own algorithm was used, which provided more control over the steps of data processing. For a few genes, the results obtained by our method were compared to that of the commercial software. In all cases tested, the results of both procedures were similar.

In our algorithm, the number one minus the regression coefficient of the linear regression used to calculate the optimal photomultiplier sensitivity for image analysis was defined as the read error of a single data read. These single reads were normalized not against the overall brightness of the slide but against a probable number of 3,880 total gene-specific mRNAs per cell. This gave the expression level per experiment. The mean values of the expression levels of all four experiments of each data set were calculated. As the deviation, the standard deviation of this mean value was added to the sum of the four read errors to obtain the experimental error. When the results of the different conditions were compared, Q was defined as the ratio of the mean expression levels condition 2/condition 1. Additionally, Q′ was used, with Q′ = Q if Q > 1, otherwise Q′ = 1/Q. The significance was defined as (condition 2 − condition 1)/(deviation 1 + deviation 2). Results with significance > 1 or significance < −1 and Q′ ≥ 2.0 were further considered, and results with 1.5 ≥ Q′ > 2.0 were kept as additional results if the significance met the same conditions.

Validation of microarray data by qRT-PCR.

Microarray data were confirmed by performing quantitative real-time reverse transcription (RT)-PCR (qRT-PCR) on the iCycler iQ real-time PCR detection system (Bio-Rad, Hercules, Calif.). The cDNA was produced with the same RNA used for microarray analysis. For real-time PCR duplicate reaction with 1 μl of template cDNA, 5 pmol of primers, and the QuantiTect SYBR Green PCR kit (Qiagen, Hilden, Germany) were used as well as a second independent approach including 1 μl of template cDNA, 5 pmol of primers, 0.5× SYBR Green (Fisher) and 1 U of Taq polymerase (Roche Diagnostics GmbH, Mannheim, Germany). Details for PCR protocols and primer sequences are available on request.

Fluorescence was measured at the end of each 72°C incubation and analyzed with iCycler iQ software (version 3.0). Melting curves analysis (60 to 95°C, 0.5°C increments) were performed to ensure PCR specificity. For quantification, standard curves of cDNA dilutions (1:10 to 1:10,000) were performed as duplicates. The crossing point for each reaction converted to log copy numbers was determined with the standard curve algorithm and arithmetic baseline adjustment. Expression ratios were obtained by dividing the copy numbers of two corresponding strains. An average of four copy numbers per cDNA as well as an average of two independent biological examples were calculated.

RESULTS AND DISCUSSION

RpoE of E. coli is required for full copper tolerance.

The rpoE gene was deleted from the chromosome of E. coli strain BW25113 in a way that prevented polar effects on downstream genes. The copper resistance of a ΔrpoE mutant strain of E. coli was tested in solid (Table (Table1)1) and liquid (Fig. (Fig.1A)1A) media. In general, and in agreement with the results of other groups, the MICs of copper were higher than those of other metals of the first transition period of the periodic system of the elements (22, 24, 25, 44, 48, 51, 55). To our current knowledge, E. coli does not contain cytoplasmic copper-containing proteins or a copper uptake system, but only CopA as a copper efflux system. Thus, due to the resulting slow import of copper combined with an efficient export and binding of copper to complexing components of this cellular compartment, the “free” copper in the cytoplasm reaches only zeptomolar concentrations (12). Therefore, usage of copper by E. coli and toxic actions of this element on E. coli may be exclusively periplasmic events without touching the cytoplasm, which could explain the comparable low toxicity of copper in growth experiments.

FIG. 1.
Dose-response curves for growth of wild-type E. coli and its ΔrpoE derivative in the presence of metal cations. Parallel cultures of the ΔrpoE (open circles), ΔompC (open squares), and ΔrpoE ΔompC (open triangles) ...
TABLE 1.
Resistance of E. coli mutant strains to coppera

Deletion of rpoE led to a decrease in copper resistance. This suggested that RpoE-dependent genes were required for full cellular defense against excess copper. The differences in growth of a ΔrpoE mutant and its wild-type strain in dose-response curves (Fig. (Fig.1A)1A) were the result of a longer lag phase rather than a lower growth rate (Fig. (Fig.2,2, Table Table2)2) or growth yield (Fig. (Fig.2).2). The CFU count of ΔrpoE cells cultivated in the presence of 2 mM Cu(II) decreased after about 1 h for 2 h before it increased again (data not shown). The viability of wild-type cells, in contrast, was not influenced by 2 mM Cu(II). This demonstrated that the ΔrpoE mutant cells were killed by Cu(II) until a subpopulation of about 1% of the ΔrpoE mutant cells was able to mount a defense against the toxic metal. This subpopulation, however, was not that of a suppressor mutant, because cultivation of ΔrpoE cells in the presence of 2 mM Cu(II) followed by growth without added copper did not increase the copper resistance of these cells (data not shown).

FIG. 2.
Influence of copper on growth of a ΔrpoE strain of E. coli. Wild-type E. coli strain BW25113 (circles), its ΔrpoE deletion strain (squares), and a strain with the ΔrpoE deletion complemented in trans (triangles) were cultivated ...
TABLE 2.
Parameters of growth of a ΔrpoE mutant strain in the presence of metal cations in liquid culturea

Retarded growth of the ΔrpoE mutant strain in the presence of Cu(II) could be partially complemented by a single copy of the rpoE-rseAB operon inserted at the λ attachment site, λatt (Fig. (Fig.2,2, Table Table2).2). Partial complementation and retarded growth of the cells cultivated without added metals may be explained by the presence of the rseAB genes downstream of the deleted rpoE gene. This may lead to an imbalance between the RpoE sigma factor and its anti-sigma factor, resulting in an incomplete regulatory response to metal and other stresses.

RpoE is essential (15, 19), and an unknown suppression has to occur to allow growth of the ΔrpoE mutant strain. However, the stability of the copper-sensitive phenotype and partial restoration of copper resistance of ΔrpoE by complementation of rpoE-rseAB in trans indicated that the suppression allowed growth without added copper but did not restore copper resistance. This made it possible to analyze the influence of RpoE on metal resistance in E. coli.

The influence of a ΔrpoE deletion on copper resistance was weaker than that of a deletion of the copper-exporting P-type ATPase CopA (55) but stronger than that of an interruption of the periplasmic stress protein CueO (24) (Table (Table1).1). When the metal resistance of several different combinations of copper resistance gene deletions were analyzed, lack of all four genes (ΔrpoE ΔcopA ΔcusCFBA ΔcueO::cat) led (as expected) to the lowest level of copper resistance (Table (Table1).1). Surprisingly, the presence of CopA alone (ΔrpoE ΔcusCFBA ΔcueO::cat mutant) did not increase copper resistance compared to that of the quadruple mutant. This indicates the importance of at least one copper homeostasis factor acting in the periplasm.

Regulation of copper resistance genes by RpoE.

Gene fusions of lacZ with cueO, the gene for its regulator, cueR, and copA were constructed and inserted as single copies into the λatt site of the bacterial chromosome of the ΔrpoE mutant and its wild-type strain. In both types of reporter strains, expression of all three proteins was induced by copper (Fig. (Fig.3).3). Thus, RpoE was not required for this process. However, the copper concentration needed for optimum induction of cueR or copA was lower in mutant cells, about 1.0 to 1.5 mM Cu(II) (Fig. 3A and C) than in wild-type cells, about 2.0 to 2.25 mM Cu(II). Cells able to survive at copper concentrations above a “threshold” concentration of about 1.5 mM Cu(II) needed a functional Cus system or CueO or RpoE (Table (Table1)1) to minimize copper toxicity in the periplasm. At copper concentrations above this threshold, a strong increase in copper-dependent expression of CueO was visible, but only in wild-type cells (Fig. (Fig.3B).3B). Induction of CueO by copper followed a two-phase response that, below the threshold, was RpoE independent but, above it, was RpoE dependent (Fig. (Fig.3B).3B). This inability to respond might be due to copper damage above the threshold concentration, which agrees with the bactericidal activity of 2 mM Cu(II) on ΔrpoE cells as observed in the growth experiments (Fig. (Fig.22).

FIG. 3.
Induction of copper homeostasis genes by metal cations in cells with and without RpoE. E. coli cells containing a cueRp::lacZ (panel A and circles), a cueOp::lacZ (panel B and squares) or a copAp::lacZ fusion (panel C and triangles) were treated with ...

RpoE is also required for full cadmium and zinc tolerance.

When zinc or cadmium stress was analyzed on solid medium and in liquid culture (Fig. (Fig.1B1B and and1C1C and Tables Tables22 and and3),3), deletion of rpoE also showed a clear effect on resistance to either metal, indicating the necessity of RpoE-dependent factors for full tolerance against these two metals, as well. The effect of a ΔrpoE deletion was smaller than that of a deletion of the P-type ATPase ZntA but higher than that of a loss of the CDF protein ZitB (Table (Table3).3). However, in contrast to copper, the presence of ZntA alone led to a strong increase in zinc and cadmium resistance (ΔrpoE ΔzitB mutant), while the effect of the presence of RpoE or of ZitB alone (the other two double mutants) on metal resistance was similar. Nickel, cobalt, and lead resistance was unchanged in the ΔrpoE mutant strain (data not shown).

TABLE 3.
Resistance of E. coli mutant strains to cadmium and zinca

RpoE was induced by metals.

A rpoEp::lacZ fusion was inserted as a single gene copy into the λatt site of the chromosome of wild-type E. coli cells. When the resulting strain was treated with various metal cations in TGC medium, Cu(II) and Zn(II) resulted in a strong increase in reporter activity (data not shown). The copper-mediated increase in reporter gene activity was sigmoidal, with the strongest increase at the copper threshold concentration of about 1.5 mM Cu(II). The maximum induction coefficient reached was 26.3- ± 14.5-fold at 2.5 mM Cu(II). Zinc showed the strongest effect, with a maximum induction coefficient of 42.5- ± 16.6-fold at 0.8 mM Zn(II) and a sharp increase in reporter activity above 0.4 mM Zn(II). In contrast, reporter expression increased only slightly and as a linear function with Cd(II), Ni(II), Mg(II), or Na(I), and the last was used to induce osmotic stress (data not shown). The maximum induction coefficients obtained with these metals were 3.86- ± 1.91-fold at 100 μM Cd(II), 2.13- ± 0.34-fold at 1 mM Ni(II), 2.02- ± 0.69-fold at 50 mM Mg(II), and 4.52- ± 2.40-fold at 500 mM Na(I).

Global gene expression analysis.

To analyze the full impact of the ΔrpoE deletion on metal homeostasis in E. coli, a global cellular expression analysis was performed with microarrays. Cells of the E. coli wild-type strain and its ΔrpoE deletion strain were treated in TGC medium for 10 min with 25 μM Cd(II), 100 μM Zn(II), 250 μM Cu(II), or without added metals prior to RNA isolation. This time should provide the cells sufficient opportunity to induce metal tolerance systems. The concentrations used were 10% of the maximally tolerated concentrations of the ΔrpoE mutant strain (Tables (Tables11 and and3).3). In each microarray, 4,239 gene-specific spots were analyzed. Each condition was performed four times, which includes two independent cultivations of the E. coli strains and a dye swap experiment. The mean values of the four normalized expression values were then analyzed from different perspectives.

The mean expression values were evaluated in 10 comparisons focusing on (i) the comparison of ΔrpoE mutant and wild-type cells without added metals as a “baseline” (0±rpoE), (ii) the influence of copper on wild-type (WT±Cu) and mutant cells (ΔrpoE±Cu) considered separately before the global gene expression profile of copper-treated cells plus and minus RpoE was directly compared (Cu±rpoE), and (iii) the same procedure for cadmium (WT±Cd and ΔrpoE±Cd followed by Cd±rpoE) and (iv) zinc (WT±Zn, ΔrpoE±Zn, Zn±rpoE). In each comparison, a change in expression of more than 2.0 in either direction (Q ≥ 2.0 or Q ≤ 0.5) was counted as a significant change. Moreover, a significant change required that the difference between the two signals had to be larger than the sum of both signal deviations. Of the 4,239 genes analyzed, a quarter (1,069 genes) showed significant upregulation or downregulation with in at least one of the 10 comparisons (data not shown).

Changes in transcript levels in selected genes were validated with quantitative real-time qRT-PCR (Table (Table4).4). Due to the shorter incubation times and the lower metal concentrations used in the qRT-PCR experiments compared to the reporter gene experiments, only zinc was able to upregulate rpoE transcription (Table (Table4),4), while the rpoEp::lacZ reporter fusion was induced by all three metals. Compared to the microarray analysis, qRT-PCR provided consistent data. One exception, ompA, was 2.9-fold downregulated (Q = 0.35, Table Table4)4) in the ΔrpoE mutant strain upon cadmium treatment in the microarray approach. However, by generalized standards (a gene is defined as regulated when it shows more than a twofold change) it was not significantly regulated (Q = 0.67, Table Table4)4) as determined by qRT-PCR. For all other genes tested, the qRT-PCR data confirmed differences in mutant and wild-type strain expression. The biases of regulation were similar, but the magnitude of regulation (up to 10-fold between microarray and qRT-PCR) varied notably. Differences in expression between microarray and qRT-PCR data were greater when regulation was stronger, with qRT-PCR usually giving the larger change in regulation. This observation has been reported by others (10, 20, 39, 42). As an independent technique, qRT-PCR generates more accurate and sensitive data and gives a wider dynamic range. No false positives could be detected among the genes tested. This and the narrow threshold of variability of the analysis marked the microarray data presented here as reliable.

TABLE 4.
Verification of microarray data by quantitative real-time RT-PCR analysis

Copper.

When copper-treated cells of the ΔrpoE mutant cells were compared to copper-treated wild-type cells, mRNAs of only four genes appeared in lower quantities in the mutant cells (Table (Table5).5). The decreased concentration of rpoE-specific mRNA was a direct (and trivial) result of the method of rpoE deletion. Since RpoE controls expression of the rpoErseABC operon (2), the transcript level of rseA (encoding the RpoE-specific anti-sigma factor) was also affected. The gene with the strongest (33-fold, Q = 0.03, Table Table5)5) decrease in mRNA concentration was the ompC gene, which encodes an outer membrane porin protein. Another gene encoding an outer membrane protein, ompA, was also downregulated, but only twofold. Validation with qRT-PCR showed strongly decreased ompC mRNA levels (140- to 320-fold) in untreated and metal-treated ΔrpoE mutant cells, but ompA transcript levels were also decreased only by half under the conditions tested (Table (Table4).4). A ΔompC deletion strain displayed the same degree of copper sensitivity as a ΔrpoE single and also a ΔompC ΔrpoE double deletion strain (Fig. (Fig.1A).1A). Therefore, RpoE apparently contributes to copper resistance via increased expression of the outer membrane protein OmpC.

TABLE 5.
Changes in transcript levels in the ΔrpoE mutant strain and its wild-type strain after treatment with 250 μM Cu(II)a

The contribution of OmpC to copper resistance is puzzling, because a porin is supposed to allow transport of substances across the outer membrane to the periplasm. Loss of OmpF and OmpC or of OmpF only (but not of OmpC alone) led to increased silver resistance in E. coli (36), which was based on diminished accumulation of Ag(I). If OmpC is able to transport Cu(II), the absence of OmpC should lead to decreased copper uptake and thus to increased resistance to this cation rather than increased sensitivity. Spontaneous copper-resistant E. coli mutants did not contain “outer membrane protein b” (37). However, OmpC and another major E. coli porin, OmpF, run close together on sodium dodecyl sulfate gels (36), so lack of “protein b” in this publication may indicate a loss of OmpC and of OmpF in these copper-resistant mutants (compare Fig. Fig.11 in reference 37) with the corresponding figure in reference 36). One explanation of this complicated situation could be that E. coli is able to control copper transport by OmpC to some extent but not by other porins.

Six genes were upregulated in the Cu±rpoE comparison (Table (Table5).5). Four genes encode the components (PspA, PspB, PspC, and PspD) of the phage shock response system Psp. The phage shock system Psp is strongly expressed in response to stressful environmental conditions, such as heat shock, ethanol treatment, osmotic shock, filamentous phage infection (64), and secretion defects (30). While PspB and PspC are involved in regulation of the Psp system, PspA protects the cell against dissipation of the proton motive force under stress conditions, probably by maintaining the integrity of the inner membrane (1, 33).

The expression change of only one gene, sohA, was specific for copper. This gene encodes a putative protease and an htrA suppressor. It allows cold-sensitive htrA mutant cells to grow at 42°C (6). However, the increase in expression of sohA in the Cu±rpoE comparison was just above the twofold cutoff level.

Thus, only 10 genes were changed in expression when copper-treated rpoE mutant and wild-type cells were compared. All 10 showed a similar change mRNA concentration in the 0±rpoE comparison (sohA, but only weakly, with Q = 1.8). Therefore, the altered mRNA concentrations of these genes were an effect of the ΔrpoE deletion and not of the copper treatment.

Copper treatment of wild-type cells resulted in upregulation of 61 genes (see the supplemental material). The cus copper resistance operon genes and genes involved in enterobactin synthesis and iron uptake showed the strongest upregulation (Table (Table5).5). In the presence of copper, CueO protects E. coli cells by oxidizing enterobactin, the catechol iron siderophore of E. coli (26), which the cell might counteract by increased enterobactin synthesis under copper stress. Cus mediates transenvelope efflux of copper by the CusCBA protein complex (21, 22, 25, 51). CusR is the response regulator of the determinant (21, 44), and its histidine sensor kinase CusS (YbcZ) was induced 1.7-fold by copper. CusF (YlcC) showed the strongest response upon copper shock and is a periplasmic copper-binding protein and maybe a periplasmic copper chaperone (22). Vigorous induction of the cus determinant by copper matches data obtained elsewhere (21, 22, 51).

Induction of the cus determinant by copper was accompanied by upregulation of other genes putatively involved in copper homeostasis. The yedWV genes encoding the putative copper response two-component regulatory system CopRS were also induced two- to threefold after copper treatment. The gene for the periplasmic oxidase CueO (YacK) was induced 1.6-fold, and that of the copper-exporting P-type ATPase CopA (YbaR) was induced 5-fold. In contrast, the genes that were downregulated in copper-treated wild-type cells (total number, 229 genes) did not show any evident connection to copper homeostasis (supplemental material and Table Table55).

All genes displaying a change in expression upon copper treatment in wild-type cells also showed a similar change in expression in the ΔrpoE mutant cells, but the change of gene expression in the mutant cells seemed to be smaller than that in the wild-type cells (Table (Table4).4). To analyze this in more detail for all E. coli genes, the change in gene expression upon copper treatment of ΔrpoE mutant cells in the ΔrpoE±Cu comparison (QCuMut) was plotted against the corresponding value in the WT±Cu comparison (QCuWT) in a double log10 plot (Fig. 4A in the supplemental material). Most of the gene values clustered around log10 = 0, meaning no change in expression has occurred in either strain. Linear regression of these more than 4,000 data points resulted in a slope of 0.845. This means that the change in gene expression into both directions is smaller in ΔrpoE mutant cells than in the wild type, with QCuMut = QCuWT0.845. Similar results were obtained for the data of the cadmium- and zinc-specific comparisons (Fig. 4B and C in the supplemental material), which indicates a similar effect of the ΔrpoE deletion on resistance to all three metals tested.

Thus, E. coli responded to copper stress primarily with synthesis of the Cus transenvelope efflux complex, the CusF periplasmic copper binding protein, and some genes involved in iron metabolism. Deletion of rpoE led to decreased flexibility in the global transcriptional adaptation to copper stress and to copper-mediated killing of E. coli, which affected the lag phase of growth. Second, it led to decreased expression of ompC in copper-treated as well as untreated cells, which also resulted in decreased copper resistance, maybe by enhanced copper transport by other porins, as discussed in detail above.

Cadmium.

When the global transcription profile of cadmium-treated ΔrpoE mutant cells was compared to that of cadmium-treated wild-type cells, decreased expression of eight genes was found (Table (Table6),6), including ompC, rpoE, and ompA, which also appeared in the Cu±rpoE comparison and the 0±rpoE comparison and were discussed above. As in the case of copper, deletion of ompC led to decreased cadmium resistance, and the diminished resistance of the ΔompC, ΔrpoE, and ΔompCΔrpoE mutant strains was similar (Fig. (Fig.1B1B).

TABLE 6.
Changes in transcript levels in the ΔrpoE mutant strain and its wild-type strain after treatment with 25 μM Cd(II)a

The genes katG, grxA, and sbp showed a unique regulatory pattern, because they were upregulated in cadmium-treated wild-type cells but only upregulated by cadmium to a significantly smaller extent in the ΔrpoE mutant cells. Five genes were upregulated in the Cd±rpoE comparison; four of them (yjbE and three psp genes) were similarly upregulated in the Cu±rpoE comparison.

A total of 133 genes were upregulated in cadmium-treated wild-type cells (supplemental material). The ycfR gene showed the strongest upregulation. YcfR might be an outer membrane protein with an unknown function. Open reading frame b1973 in Table Table66 is yodA and encodes a zinc/cadmium-binding, lipocalin-like protein (17). This gene is also known as zinT, encoding a zinc-binding protein, which likely functions as an alternative or additional zinc-chelating component of the zinc ZnuABC transporter (53). These two genes are followed by yebL, encoding ZnuA, the periplasmic binding protein of the ZnuABC zinc uptake system, and ftnA, which encodes the iron storage protein ferritin. The next gene is zntA, encoding the zinc- and cadmium-exporting P-type ATPase ZntA (56), which is involved in cadmium resistance. These data may indicate an action of cadmium on cellular zinc and iron metabolism, since E. coli responded to cadmium stress with increased expression of zinc uptake and cadmium efflux systems. Cadmium also induced expression of ZinT (YodA), ZnuA, Ftn, and ZntA in the ΔrpoE mutant strain (Table (Table66).

In E. coli wild-type cells, cadmium also induced genes required for cysteine biosynthesis and redox metabolism (grxA and katG). A total of 302 genes were downregulated in wild-type cells after cadmium treatment (Table (Table66 and supplemental material). Most of the genes with a change in expression in wild-type cells were also differentially expressed in ΔrpoE mutant cells as a response to cadmium (supplemental material).

Thus, E. coli responded to cadmium stress with increased synthesis of the cadmium efflux pump ZntA, increased zinc and iron binding and uptake, and cysteine, glutaredoxin 1, and hydroperoxidase I production. Deletion of rpoE diminished adaptation specifically to cadmium by preventing increased cysteine, glutaredoxin 1, and hydroperoxidase I synthesis. The flexibility of general stress adaptation and ompC mRNA levels were also decreased as in the case of copper stress.

Zinc.

In the direct comparison of zinc-treated cells with and without rpoE, expression of 17 genes was down- and that of 16 genes was upregulated (Table (Table7).7). Expression of 17 of these 33 genes was also changed in the 0±rpoE comparison. Therefore, their change in expression was a result of the rpoE deletion rather than of zinc treatment. This was also true for the first two members of the list of downregulated genes, rpoE and ompC. In contrast to the situation with copper and cadmium, a ΔrpoE ΔompC double-deletion strain was more sensitive to zinc than either single-deletion strain (Fig. (Fig.1C).1C). This suggests two independent pathways of OmpC- and RpoE-mediated resistance to zinc.

TABLE 7.
Changes in transcript levels in the ΔrpoE mutant strain and its wild-type strain after treatment with 100 μM Zn(II)a

In the list of downregulated genes in the Zn±rpoE comparison, rpoE and ompC were followed by htrA (Table (Table77 and supplemental material). Expression of the heat shock serine protease HtrA (= DegP) is under RpoE control and required under various stress conditions such as hyperosmotic shock, virulence, oxidative stress, and dead cell lysis (7, 29, 49, 61, 62). HtrA is essential for E. coli at elevated temperatures and is also involved in regulation of activity of RpoE because it destabilizes the RpoE regulatory component RseA (4, 35). Proteins related to HtrA are widely distributed, from bacteria to plants (13, 31), and the crystal structure indicates that a temperature-dependent conformational change might be required to activate the protease function of this protein (32, 34). The htrA gene was induced by Zn(II) 11-fold in the wild type, but expression was decreased 6-fold when zinc-treated mutant and wild-type cells were compared (Table (Table7,7, Q = 0.16). This may indicate that htrA transcription is regulated by both RpoE and Zn(II).

The expression of the HtrA target, the RpoE regulator RseA, was also zinc dependent. The transcript level of rseA was not changed in the 0±rpoE comparison, fivefold increased in the WT±Zn comparison, but not increased at all under the same conditions in the mutant (Table (Table7).7). Consequently, this gene appeared to be sixfold downregulated (Q = 0.16) in the Zn±rpoE comparison. This was also verified by qRT-PCR (Table (Table44).

As in the Cu±rpoE, Cd±rpoE, and 0±rpoE comparisons, psp transcripts were increased in the Zn±rpoE comparison. The remaining Zn±rpoE upregulated genes have mostly unknown functions. OmpX (threefold upregulated as determined by microarrays, Table Table5,5, and sixfold by qRT-PCR, Table Table4)4) was named because of its similarity to OmpX from Enterobacter cloacae (63). Thus, in addition to OmpC, other outer membrane proteins might also be involved in metal homeostasis.

In wild-type cells, 256 genes were upregulated upon addition of zinc (Table (Table77 and supplemental material). The highest increase was an exceptional 327-fold for yjaI (zraP), which encodes a zinc-dependent periplasmic protein (50). The yjaI (zraP) gene was also induced by zinc in the ΔrpoE mutant strain (176-fold). A strong increase in the zraP mRNA under this condition in wild-type and mutant cells was verified by qRT-PCR (Table (Table44).

Genes involved in ferrous iron transport were upregulated by zinc treatment in wild-type cells: feoA, entA, entB, entE, and entF, involved in ferrous iron uptake and biosynthesis of the catecholate siderophore enterobactin, and the genes for outer the membrane receptors, fhuE and fepA, involved in iron-siderophore uptake, were all upregulated (Table (Table7).7). The same genes were also upregulated in ΔrpoE cells, but to a smaller degree. All the data together may indicate that zinc treatment results in iron depletion and that the rpoE mutant can only partially adjust to this stress condition.

The zntA gene was induced only twofold in wild-type cells (supplemental material), and expression of the zitB (ygbR) gene, encoding the second, proton-driven Zn(II) efflux system of E. coli (23), was not changed under these conditions. This matched the result of the mutant studies (Table (Table3),3), which indicated that increased ZitB expression may not be required for zinc resistance when RpoE and ZntA were present.

The 115 genes downregulated in wild-type cells upon addition of zinc (Table (Table7)7) were led by ompF (10-fold downregulation, Q = 0.1), which was also number one in the corresponding list of the ΔrpoE strain (Table (Table7).7). Its downregulated expression after zinc treatment of both types of cells was also verified by qRT-PCR (Table (Table4).4). The expression level of both genes, ompC and ompF, is determined by a two-component regulatory system composed of the membrane-bound histidine kinase EnvZ and the response regulator OmpR (11). The amount of ompR-specific mRNA increased twofold in ΔrpoE and wild-type cells (only by zinc, not by cadmium or copper), but that of envZ remained unchanged (supplemental material). Thus, concerning porins, the cellular response to zinc seems to be more complicated than that to copper. The decrease in the OmpC level as a consequence of the rpoE deletion was connected to a zinc-dependent downregulation of ompF in both cell types. Additionally, ompX messages increased threefold in the Zn±rpoE comparison. These data show that E. coli may changed its porin composition as a response to metal stress.

Conclusion.

Expression of RpoE was induced by metals and a ΔrpoE mutant strain showed decreased metal resistance. Two possible modes of involvement of RpoE in metal homeostasis were found, (i) flexibility of the cell to adapt the transcriptome to stress conditions (Fig. 4 and supplemental material) and (ii) production of outer membrane proteins. The first mode would explain the increased lag phase of growth of ΔrpoE mutant cells in the presence of copper (Fig. (Fig.2).2). From copper via cadmium to zinc, the situation becomes more and more complicated, with additional categories of genes being involved and changed in expression after rpoE deletion in different ways. However, decreased transcriptome flexibility and decreased expression of OmpC are two effects of ΔrpoE deletion on cellular copper homeostasis in E. coli that were clearly demonstrated.

Supplementary Material

[Supplemental material]

Acknowledgments

This work was supported by Graduiertenkolleg “Stress” of the Deutsche Forschungsgemeinschaft and UMTS funds supplied to the Institute of Eco-Remediation.

We thank G. Schleuder for skillful technical assistance, especially for the ΔompC growth experiments and the additional work for the revised version, and S. Franke for some mutant strains. Thanks are also due B. Wanner for plasmids pAH125 and pINT-ts, S. Schornack and AG Humbeck for qRT-PCR support, and F. Stahl for advice on microarray technology. We thank Gerhard Mittenhuber for bioinformatic calculations.

Footnotes

Supplemental material for this article may be found at http://jb.asm.org/.

REFERENCES

1. Adams, H., W. Teertstra, J. Demmers, R. Boesten, and J. Tommassen. 2003. Interactions between phage-shock proteins in Escherichia coli. J. Bacteriol. 185:1174-1180. [PMC free article] [PubMed]
2. Ades, S. E., L. E. Connolly, B. M. Alba, and C. A. Gross. 1999. The Escherichia coli σE-dependent extracytoplasmic stress response is controlled by the regulated proteolysis of an anti-sigma factor. Genes Dev. 13:2449-2461. [PMC free article] [PubMed]
3. Ades, S. E., I. L. Grigorova, and C. A. Gross. 2003. Regulation of the alternative σE during initiation, adaptation, and shutoff of the extracytoplasmic heat shock response in Escherichia coli. J. Bacteriol. 185:2512-2519. [PMC free article] [PubMed]
4. Alba, B. M., H. J. Zhong, J. C. Pelayo, and C. A. Gross. 2001. degS (hhoB) is an essential Escherichia coli gene whose indispensable function is to provide σE activity. Mol. Microbiol. 40:1323-1333. [PubMed]
5. Anton, A., A. Weltrowski, J. H. Haney, S. Franke, G. Grass, C. Rensing, and D. H. Nies. 2004. Characteristics of zinc transport by two bacterial cation diffusion facilitators from Ralstonia metallidurans and Escherichia coli. J. Bacteriol. 186:7499-7507. [PMC free article] [PubMed]
6. Baird, L., and C. Georgopoulos. 1990. Identification, cloning, and characterization of the Escherichia coli sohA gene, a suppressor of the htrA (degP) null phenotype. J. Bacteriol. 172:1587-1594. [PMC free article] [PubMed]
7. Bianchi, A. A., and F. Baneyx. 1999. Hyperosmotic shock induces σ32 and σE stress regulons of Escherichia coli. Mol. Microbiol. 34:1029-1038. [PubMed]
8. Bowtell, D., and J. Sambrook. 2003. DNA microarrays: a molecular cloning manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.
9. Brazma, A., P. Hingamp, J. Quackenbush, G. Sherlock, P. Spellman, C. Stoeckert, J. Aach, W. Ansorge, C. A. Ball, H. C. Causton, T. Gaasterland, P. Glenisson, F. C. P. Holstege, I. F. Kim, V. Markowitz, J. C. Matese, H. Parkinson, A. Robinson, U. Sarkans, S. Schulze-Kremer, J. Stewart, R. Taylor, J. Vilo, and M. Vingron. 2001. Minimum information about a microarray experiment (MIAME)-towards standards for microarray data. Nat. Genet. 29:365-371. [PubMed]
10. Brokx, S. J., M. Ellison, T. Locke, D. Bottorff, L. Frost, and J. H. Weiner. 2004. Genome-wide analysis of lipoprotein expression in Escherichia coli MG1655. J. Bacteriol. 186:3254-3258. [PMC free article] [PubMed]
11. Cai, S. J., and M. Inouye. 2002. EnvZ-OmpR interaction and osmoregulation in Escherichia coli. J. Biol. Chem. 277:24155-24161. [PubMed]
12. Changela, A., K. Chen, Y. Xue, J. Holschen, C. E. Outten, T. V. O'Halloran, and A. Mondragon. 2003. Molecular basis of metal-ion selectivity and zeptomolar sensitivity by CueR. Science 301:1383-1387. [PubMed]
13. Clausen, T., C. Southan, and M. Ehrmann. 2002. The HtrA family of proteases: Implications for protein composition and cell fate. Mol. Cell 10:443-455. [PubMed]
14. Collinet, B., H. Yuzawa, T. Chen, C. Herrera, and D. Missiakas. 2000. RseB binding to the periplasmic domain of RseA modulates the RseA: σE Interaction in the cytoplasm and the availability of σE · RNA polymerase. J. Biol. Chem. 275:33898-33904. [PubMed]
15. Dartigalongue, C., D. Missiakas, and S. Raina. 2001. Characterization of the Escherichia coli σE regulon. J. Biol. Chem. 276:20866-20875. [PubMed]
16. Datsenko, K. A., and B. L. Wanner. 2000. One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products. Proc. Natl. Acad. Sci. USA 97:6640-6645. [PMC free article] [PubMed]
17. David, G., K. Blondeau, M. Schiltz, S. Penel, and A. Lewit-Bentley. 2003. YodA from Escherichia coli is a metal-binding, lipocalin-like protein. J. Biol. Chem. 278:43728-43735. [PubMed]
18. De Las Penas, A., L. Connolly, and C. A. Gross. 1997. The σE-mediated response to extracytoplasmic stress in Escherichia coli is transduced by RseA and RseB, two negative regulators of σE. Mol. Microbiol. 24:373-385. [PubMed]
19. De las Penas, A., L. Connolly, and C. A. 1997. Gross. σE is an essential sigma factor in Escherichia coli. J. Bacteriol. 179:6862-6864. [PMC free article] [PubMed]
20. Firoved, A. M., and V. Deretic. 2003. Microarray analysis of global gene expression in mucoid Pseudomonas aeruginosa. J. Bacteriol. 185:1071-1081. [PMC free article] [PubMed]
21. Franke, S., G. Grass, and D. H. Nies. 2001. The product of the ybdE gene of the Escherichia coli chromosome is involved in detoxification of silver ions. Microbiology 147:965-972. [PubMed]
22. Franke, S., G. Grass, C. Rensing, and D. H. Nies. 2003. Molecular analysis of the copper-transporting CusCFBA efflux system from Escherichia coli. J. Bacteriol. 185:3804-3812. [PMC free article] [PubMed]
23. Grass, G., B. Fan, B. P. Rosen, S. Franke, D. H. Nies, and C. Rensing. 2001. ZitB (YbgR), a member of the cation diffusion facilitator family, is an additional zinc transporter in Escherichia coli. J. Bacteriol. 183:4664-4667. [PMC free article] [PubMed]
24. Grass, G., and C. Rensing. 2001. CueO is a multi-copper oxidase that confers copper tolerance in Escherichia coli. Biochem. Biophys. Res. Commun. 286:902-908. [PubMed]
25. Grass, G., and C. Rensing. 2001. Genes involved in copper homeostasis in Escherichia coli. J. Bacteriol. 183:2145-2147. [PMC free article] [PubMed]
26. Grass, G., K. Thakali, P. E. Klebba, D. Thieme, A. Müller, G. F. Wildner, and C. Rensing. 2004. Linkage between catecholate siderophores and the multicopper oxidase CueO in Escherichia coli. J. Bacteriol. 186:5826-5833. [PMC free article] [PubMed]
27. Grosse, C., G. Grass, A. Anton, S. Franke, A. Navarrete Santos, B. Lawley, N. L. Brown, and D. H. Nies. 1999. Transcriptional organization of the czc heavy metal homoeostasis determinant from Alcaligenes eutrophus. J. Bacteriol. 181:2385-2393. [PMC free article] [PubMed]
28. Haldimann, A., and B. L. Wanner. 2001. Conditional-replication, integration, excision, and retrieval plasmid-host systems for gene structure-function studies of bacteria. J. Bacteriol. 183:6384-6393. [PMC free article] [PubMed]
29. Humphreys, S., A. Stevenson, A. Bacon, A. B. Weinhardt, and M. Roberts. 1999. The alternative sigma factor σE is critically important for the virulence of Salmonella typhimurium. Infect Immun. 67:1560-1568. [PMC free article] [PubMed]
30. Jones, S. E., L. J. Lloyd, K. K. Tan, and M. Buck. 2003. Secretion defects that activate the phage shock response of Escherichia coli. J. Bacteriol. 185:6707-6711. [PMC free article] [PubMed]
31. Kieselbach, T., and C. Funk. 2003. The family of Deg/HtrA proteases: from Escherichia coli to Arabidopsis. Physiol. Plant. 119:337-346.
32. Kim, D. Y., D. R. Kim, S. C. Ha, N. K. Lokanath, C. J. Lee, H. Y. Hwang, and K. K. Kim. 2003. Crystal structure of the protease domain of a heat-shock protein HtrA from Thermotoga maritima. J. Biol. Chem. 278:6543-6551. [PubMed]
33. Kleerebezem, M., W. Crielaard, and J. Tommassen. 1996. Involvement of stress protein PspA (phage shock protein A) of Escherichia coli in maintenance of the protonmotive force under stress conditions. EMBO J. 15:162-171. [PMC free article] [PubMed]
34. Krojer, T., M. Garrido-Franco, R. Huber, M. Ehrmann, and T. Clausen. 2002. Crystal structure of DegP (HtrA) reveals a new protease-chaperone machine. Nature 416:455-459. [PubMed]
35. Langen, G. R., J. R. Harper, T. J. Silhavy, and S. P. Howard. 2001. Absence of the outer membrane phospholipase A suppresses the temperature-sensitive phenotype of Escherichia coli degP mutants and induces the Cpx and σE extracytoplasmic stress responses. J. Bacteriol. 183:5230-5238. [PMC free article] [PubMed]
36. Li, X. Z., H. Nikaido, and K. E. Williams. 1997. Silver-resistant mutants of Escherichia coli display active efflux of Ag+ and are deficient in porins. J. Bacteriol. 179:6127-6132. [PMC free article] [PubMed]
37. Lutkenhaus, J. F. 1977. Role of a major outer membrane protein in Escherichia coli. J. Bacteriol. 131:631-637. [PMC free article] [PubMed]
38. Maeda, H., M. Jishage, T. Nomura, N. Fujita, and A. Ishihama. 2000. Two extracytoplasmic function sigma subunits, σE and σFecI of Escherichia coli: promoter selectivity and intracellular levels. J. Bacteriol. 182:1181-1184. [PMC free article] [PubMed]
39. McHugh, J. P., F. Rodriguez-Quinones, H. Abdul-Tehrani, D. A. Svistunenko, R. K. Poole, C. E. Cooper, and S. C. Andrews. 2003. Global iron-dependent gene regulation in Escherichia coli. A new mechanism for iron homeostasis. J. Biol. Chem. 278:29478-29486. [PubMed]
40. Mergeay, M., D. Nies, H. G. Schlegel, J. Gerits, P. Charles, and F. van Gijsegem. 1985. Alcaligenes eutrophus CH34 is a facultative chemolithotroph with plasmid-bound resistance to heavy metals. J. Bacteriol. 162:328-334. [PMC free article] [PubMed]
41. Miller, J. H. 1972. Experiments in molecular genetics. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.
42. Mimmack, M. L., J. Brooking, and S. Bahn. 2004. Quantitative polymerase chain reaction: validation of microarray results from postmortem brain studies. Biol. Psychiatry 55:337-345. [PubMed]
43. Missiakas, D., and S. Raina. 1998. The extracytoplasmic function sigma factors: role and regulation. Mol. Microbiol. 28:1059-1066. [PubMed]
44. Munson, G. P., D. L. Lam, F. W. Outten, and T. V. O'Halloran. 2000. Identification of a copper-responsive two-component system on the chromosome of Escherichia coli K-12. J. Bacteriol. 182:5864-5871. [PMC free article] [PubMed]
45. Nies, D., M. Mergeay, B. Friedrich, and H. G. Schlegel. 1987. Cloning of plasmid genes encoding resistance to cadmium, zinc, and cobalt in Alcaligenes eutrophus CH34. J. Bacteriol. 169:4865-4868. [PMC free article] [PubMed]
46. Nies, D. H. 2003. Efflux-mediated heavy metal resistance in prokaryotes. FEMS Microbiol. Rev. 27:313-339. [PubMed]
47. Nies, D. H. 2004. Incidence and function of sigma factors in Ralstonia metallidurans and other bacteria. Arch. Microbiol. 181:255-268. [PubMed]
48. Nies, D. H. 1999. Microbial heavy metal resistance. Appl. Microbiol. Biotechnol. 51:730-750. [PubMed]
49. Nitta, T., H. Nagamitsu, M. Murata, H. Izu, and M. Yamada. 2000. Function of the σE regulon in dead-cell lysis in stationary-phase Escherichia coli. J. Bacteriol. 182:5231-5237. [PMC free article] [PubMed]
50. Noll, M., K. Petrukhin, and S. Lutsenko. 1998. Identification of a novel transcription regulator from Proteus mirabilis, PMTR, revealed a possible role of YjaI protein in balancing zinc in Escherichia coli. J. Biol. Chem. 273:21393-21401. [PubMed]
51. Outten, F. W., D. L. Huffman, J. A. Hale, and T. V. O'Halloran. 2001. The independent cue and cus systems confer copper tolerance during aerobic and anaerobic growth in Escherichia coli. J. Biol. Chem. 276:30670-30677. [PubMed]
52. Pace, C. N., and M. J. Scholtz. 1997. Measuring the conformational stability of a protein, p. 299-322. In T. Creighton (ed.), Protein structure: a practical approach. IRL Press, Oxford, United Kingdom.
53. Panina, E. M., A. A. Mironov, and M. S. Gelfand. 2003. Comparative genomics of bacterial zinc regulons: Enhanced ion transport, pathogenesis, and rearrangement of ribosomal proteins. Proc. Natl. Acad. Sci. USA 100:9912-9917. [PMC free article] [PubMed]
54. Raivio, T. L., and T. J. Silhavy. 2001. Periplasmic stress and ECF sigma factors. Annu. Rev. Microbiol. 55:591-624. [PubMed]
55. Rensing, C., B. Fan, R. Sharma, B. Mitra, and B. P. Rosen. 2000. CopA: an Escherichia coli Cu(I)-translocating P-type ATPase. Proc. Natl. Acad. Sci. USA 97:652-656. [PMC free article] [PubMed]
56. Rensing, C., M. Ghosh, and B. P. Rosen. 1999. Families of soft-metal-ion-transporting ATPases. J. Bacteriol. 181:5891-5897. [PMC free article] [PubMed]
57. Rezuchova, B., H. Miticka, D. Homerova, M. Roberts, and J. Kormanec. 2003. New members of the Escherichia coli σE regulon identified by a two-plasmid system. FEMS Microbiol. Lett. 225:1-7. [PubMed]
58. Roberts, S. A., A. Weichsel, G. Grass, K. Thakali, J. T. Hazzard, G. Tollin, C. Rensing, and W. R. Montfort. 2002. Crystal structure and electron transfer kinetics of CueO, a multicopper oxidase required for copper homeostasis in Escherichia coli. Proc. Natl. Acad. Sci. USA 99:2766-2771. [PMC free article] [PubMed]
59. Rouviere, P. E., A. De Las Penas, J. Mecsas, C. Z. Lu, K. E. Rudd, and C. A. Gross. 1995. rpoE, the gene encoding the second heat-shock sigma factor, sigma E, in Escherichia coli. EMBO J. 14:1032-1042. [PMC free article] [PubMed]
60. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning, a laboratory manual, 2nd. ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.
61. Sassoon, N., J. P. Arie, and J. M. Betton. 1999. PDZ domains determine the native oligomeric structure of the DegP (HtrA) protease. Mol. Microbiol. 33:583-589. [PubMed]
62. Skorko-Glonek, J., D. Zurawa, E. Kuczwara, M. Wozniak, Z. Wypych, and B. Lipinska. 1999. The Escherichia coli heat shock protease HtrA participates in defense against oxidative stress. Mol. Gen. Genet. 262:342-350. [PubMed]
63. Stoorvogel, J., M. J. van Bussel, J. Tommassen, and J. A. van de Klundert. 1991. Molecular characterization of an Enterobacter cloacae outer membrane protein (OmpX). J. Bacteriol. 173:156-160. [PMC free article] [PubMed]
64. Weiner, L., and P. Model. 1994. Role of an Escherichia coli stress-response operon in stationary-phase survival. Proc. Natl. Acad. Sci. USA 91:2191-2195. [PMC free article] [PubMed]
65. Yoon, S. H., M. J. Han, S. Y. Lee, K. J. Jeong, and J. S. Yoo. 2003. Combined transcriptome and proteome analysis of Escherichia coli during high cell density culture. Biotechnol. Bioeng. 81:753-767. [PubMed]

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

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...