Escherichia coli Gene Expression Responsive to Levels of the Response Regulator EvgA

Bacterial strains and media.E. coli wild-type strain MG1655 (ATCC 47076) was used. DH5α competent cells (Invitrogen) were used for amplification of plasmids. Bacterial cells were grown in Luria-Bertani (LB) broth (10 g of tryptone, 5 g of yeast extract, and 10 g of NaCl per liter, pH 7.0) or on LB agar medium (Bio 101). Antibiotics were added, when required, at the following final concentrations: carbenicillin, 100 μg/ml; chloramphenicol, 20 μg/ml for AcrB-producing strains and 5 μg/ml for AcrB-deficient strains. LB agar was supplemented with 10% (wt/vol) sucrose, as required.

Molecular biology techniques.Chromosomal DNA was isolated with a DNeasy tissue kit (Qiagen). Plasmids were isolated with a HiSpeed plasmid purification kit or a QIAprep Miniprep kit (Qiagen). PCRs were performed in an MJ Research PTC-100 programmable thermal controller. PCRs for constructions of plasmids were performed with Pfu Turbo DNA polymerase (Stratagene) or Platinum Pfx DNA polymerase (Invitrogen) because of their high fidelity. Diagnostic PCRs were performed with Z-Taq polymerase (Panvera). PCR products were purified with a QIAquick PCR purification kit (Qiagen). Restriction endonucleases and alkaline phosphatase were obtained from New England Biolabs. The DNA-ligation kit was obtained from Panvera. Restriction fragments were purified with a QIAquick PCR purification kit or a MinElute reaction cleanup kit (Qiagen) or were isolated, as required, from agarose gels with a QIAquick gel extraction kit (Qiagen). Electroporation was performed with a Gene Pulser apparatus (Bio-Rad). All molecular biology techniques were carried out according to the manufacturer's instructions or as described by Sambrook et al. (27). Oligonucleotides were purchased from Operon.

Construction of plasmids for EvgA expression and allelic exchange.The EvgA overexpression plasmid was constructed as follows. The 884-bp evgA gene with the native promoter and EvgA-binding site, which is important to simulate the evgA promoter (36), was amplified by PCR from MG1655 by using primers 5′-TTCCTTAAGCTTCTAAGACTAAACCGTGGCTTTTGCAATAC-3′ and 5′-TTCCTTGAATTCTTAGCCGATTTTGTTACGTTGTGCG-3′. These primers contain HindIII and EcoRI sites (underlined) suitable for cloning. The PCR products were digested with restriction enzymes and then were ligated into the HindIII and EcoRI sites of pUC19. Sequence analysis confirmed that no mutations had been introduced into the evgA gene. The resulting plasmid was designated pUCevgA. Another EvgA expression plasmid was constructed as follows. The 613-bp evgA gene was amplified by PCR from MG1655 with primers 5′-CGCGGATCCAACGCAATAATTATTGATGACCATCCTCTTG-3′ and 5′-TTCCTTCTGCAGTTAGCCGATTTTGTTACGTTGTGC-3′. These primers introduced BamHI and PstI sites (underlined) at the 5′ and 3′ ends of the PCR-generated evgA gene, respectively, and enabled the amplified gene to be inserted into the BamHI and PstI sites of the expression vector pQE80L (Qiagen) in the correct reading frame cloning. Sequence analysis confirmed the in-frame insertion and confirmed that no mutations had been introduced by PCR. The resulting plasmid, designated pQEevgA, expresses an EvgA fusion protein (composed of full-length EvgA with an amino-terminal Met-Arg-Gly-Ser-His-His-His-His-His-His-Gly-Ser) under transcriptional control of an isopropyl-β-d-thiogalactopyranoside (IPTG)-regulated phage T5 promoter.

To construct plasmids for allelic exchange, approximately 1-kb fragments flanking the gene or tandem group of genes to be disrupted were amplified from MG1655 genomic DNA in two separate PCRs by using primer pair No and Ni or primer pair Ci and Co, respectively. The two PCR products were combined and amplified by PCR using primers No and Co. Resulting amplified products were digested with restriction enzymes and then were cloned into pKO3 (17) or pKOV (25), constructed from pKO3 by removing the extra NotI site and adding a stuffer. Table 1 and Co used to construct plasmids for deletion of each gene or tandem group of genes. The Ni and Ci primers were designed such that the ends of the PCR products facing the 5′ and 3′ ends of the gene to be disrupted contained a sequence suitable for in-frame fusion of the 5′ and 3′ flanking regions by annealing. Each gene or tandem group of genes was completely deleted, leaving only the start codon of the first ORF of a tandem group of genes and the stop codon of the last ORF, except in the case of deletions of ydeP, b1500, ydeO, yegR-b2084-yegZ, slp-yhiF, hdeB, yiiS, yjdE, and emrKY in which the 9, 51, 15, 33, 9, 132, 93, 111, and 6 nucleotides after the start codon were left, respectively. These exceptions were designed to minimize polar effects or to obtain primers with adequate melting temperatures. The No and Co primers contain restriction enzyme cutting sites suitable for cloning. pKO3 and pKOV were used for insertion of NotI-SalI- and NotI-BamHI-digested fragments, respectively.

Gene deletion and screening.Each plasmid for deletion was electroporated into cells, and they were allowed to recover for 1 h at 30°C with aeration. The cells were plated on LB plates containing chloramphenicol and were incubated at 43°C overnight to isolate integrants. The integrants were streaked on LB plates containing chloramphenicol and were incubated at 43°C for 12 h to prevent contamination of the parent strain. The purified integrants were subsequently grown in drug-free LB for 9 h for a second allelic exchange, serially diluted, plated on an LB plate supplemented with 10% (wt/vol) sucrose, and incubated at 30°C for 24 h. Chloramphenicol-susceptible and sucrose-resistant clones were selected and subjected to PCR for verification of gene deletion. The verification was done by direct amplification of genomic DNA from each colony by using the primer pair No and Co. A wild-type or deletion genotype was diagnosed according to the size of the PCR product.

RNA isolation, mRNA enrichment, and labeling.A single colony of E. coli harboring plasmid was inoculated in 1 ml of LB broth containing carbenicillin and was grown overnight with aeration at 37°C. The next day 30 ml of LB broth was inoculated with 0.1 ml of the overnight culture and was grown at 37°C with aeration to an optical density at 600 nm of 0.2. The expression of His-tagged EvgA was induced by adding 1 mM IPTG to the LB broth before inoculation of the culture, when required. Total RNA was isolated from the cells by extraction with hot acid phenol:chloroform. Briefly, samples of culture were transferred directly into acid phenol:chloroform (5:1; Ambion) at 65°C to ensure rapid lysis and inactivation of RNases. Two additional acid phenol:chloroform extractions were performed, followed by ethanol precipitation. The pellet was then resuspended in diethyl pyrocarbonate-treated water and was purified with an RNeasy Mini Kit (Qiagen). Isolated RNA was resuspended in diethyl pyrocarbonate-treated water, quantitated on the basis of its absorption at 260 nm, visualized on a denaturing polyacrylamide gel to check quality, and stored at −20°C until further use.

Enrichment of mRNA was done as described in the GeneChip expression analysis technical manual (Affymetrix). In brief, a set of oligonucleotide primers specific for either 16S or 23S rRNA was mixed with total RNA. After annealing at 70°C for 5 min, Moloney murine leukemia virus reverse transcriptase (New England Biolabs) was added to synthesize cDNA strands complementary to the two rRNA species. The cDNA strand synthesis allows for selective degradation of the 16S and 23S rRNAs by RNase H (Epicentre Technologies). Treatment of the RNA-cDNA mixture with DNase I (Amersham Pharmacia Biotech) removed the cDNA molecules and oligonucleotide primers, which resulted in an RNA preparation that was enriched for mRNA.

For direct labeling of RNA, 20 μg of enriched RNA was fragmented at 95°C for 30 min in a total volume of 88 μl of T4 polynucleotide kinase buffer (New England Biolabs). After being cooled to 4°C, 2 μl of 5 mM γ-S-ATP (Roche Molecular Biochemicals) and 10 μl of a 10-U/μl concentration of T4 polynucleotide kinase (New England Biolabs) were added to the fragmented RNA and the reaction mixture was incubated at 37°C for 50 min. To inactivate T4 polynucleotide kinase, the reaction mixture was incubated for 10 min at 65°C and the RNA was subsequently ethanol precipitated to remove excess γ-S-ATP. After centrifugation the RNA pellet was resuspended in 96 μl of 30 mM MOPS (morpholinepropanesulfonic acid, pH 7.5), and 4 μl of a 50 mM polyethylene oxide-iodoacetyl biotin (Pierce Chemical) solution was added to introduce the biotin label. The reaction mixture was incubated at 37°C for 1 h, and the labeled RNA was subsequently ethanol precipitated and washed twice with 70% ethanol to remove unreacted biotin compounds. The labeled RNA was quantitated on the basis of the absorption at 260 nm and was hybridized to an E. coli genome array (Affymetrix).

Microarray hybridization, scanning, and analysis.Hybridization was done as described in the GeneChip expression analysis technical manual. The hybridization solution contained 100 mM MES (N-morpholinoethanesulfonic acid), 1 M NaCl, 20 mM EDTA, and 0.01% Tween 20, pH 6.6 (referred to as 1× MES). In addition, the solution contained 0.1 mg of herring sperm DNA (Promega)/ml, 0.5 mg of acetylated bovine serum albumin (BSA; Invitrogen)/ml, and 50 pM control oligonucleotide B2 (Affymetrix). Hybridization was carried out at 45°C for 16 h with mixing on a rotary mixer at 60 rpm.

Following hybridization, the sample solution was removed and the array was washed and stained in a GeneChip fluidics station (Affymetrix). In brief, to enhance the signals 10 μg of streptavidin (Vector Laboratories)/ml and 2 mg of BSA/ml in 1× MES were used as the first staining solution. After the streptavidin solution was removed, an antibody mix was added as the second stain, containing 0.1 mg of goat immunoglobulin G (Sigma-Aldrich)/ml, 5 μg of biotinylated anti-streptavidin antibody (Vector Laboratories)/ml, and 2 mg of BSA/ml in 1× MES. Nucleic acid was fluorescently labeled by incubation with 10 μg of streptavidin R-phycoerythrin conjugate (Molecular Probes)/ml and 2 mg of BSA/ml in 1× MES.

The arrays were scanned at 570 nm with a resolution of 3 μm with a GeneArray scanner (Affymetrix). Data analysis was performed by using Affymetrix Microarray Suite 5.0 software. The software calculates change calls, change p values (statistical significance for change calls), and signal log ratio. Signal log ratio is the change in expression level for a transcript between a baseline and an experiment array, expressed as the log₂ ratio. Arrays were globally scaled to a target signal of 500 and were normalized. The default parameters for change calls were used.

Susceptibility testing.The antibacterial activities of the agents were determined by the broth microdilution method by following the recommendations of the National Committee for Clinical Laboratory Standards (19) with modifications as follows. LB broth was used instead of cation-adjusted Mueller-Hinton broth, and initial concentrations of the compounds may differ from those recommended. The expression of His-tagged EvgA was induced by adding 1 mM IPTG to the LB broth before inoculation of cells, when required. The lowest concentration of agent that completely inhibited growth was identified as the MIC. A fourfold-or-more difference in MICs was considered significant. All agents used for susceptibility testing were obtained from Sigma-Aldrich.

Acid resistance assay.To determine the acid resistance of exponential-phase cells, a single colony of E. coli harboring plasmid was inoculated in 1 ml of LB broth containing carbenicillin and was grown overnight with aeration at 37°C. The next day 20 ml of LB broth was inoculated with 0.1 ml of the overnight culture and was grown at 37°C with aeration. The expression of His-tagged EvgA was induced by adding 1 mM IPTG to the LB broth before inoculation of the culture, when required. When the cultures reached a cell density of 2 × 10⁸ CFU/ml, 50 μl of the culture was transferred to 2 ml of phosphate-buffered saline (PBS) (pH 7.2) and to 2 ml of warmed LB broth (pH 2.5, adjusted with HCl). Some experiments involving a challenge by dilution into Vogel-Bonner minimal E medium (39) containing 0.4% glucose (EG; pH 2.5, adjusted with HCl) were also performed. To determine the acid resistance of stationary-phase cells, a single colony of E. coli was inoculated into 1 ml of LB broth and was grown overnight with aeration at 37°C. The overnight (24-h) stationary-phase cultures were diluted 1:1,000 into PBS and warmed LB broth (pH 2.5). The number of CFU per milliliter in PBS was determined by plating serial dilutions in PBS buffer (pH 7.2) on LB agar and was used as initial cell populations. The LB broth (pH 2.5) inoculated with E. coli was incubated further at 37°C, and the number of CFU per milliliter in LB broth (pH 2.5) was determined as described above and was used as final cell populations. Percent acid survival was then calculated as the number of CFU per milliliter remaining after the acid treatment divided by the initial number of CFU per milliliter at time zero. Two or three repetitions were performed for each experiment. Percent survival values were converted to logarithmic values (log₁₀ x, where x equals the percent survival) for calculation of geometric means and standard errors.

Microarray analysis of EvgA regulon.To identify the target gene candidates for EvgA, we constructed evgAS deletion strains and EvgA expression plasmids and then compared the genome-wide transcription profiles of strains overexpressing and lacking EvgA by using oligonucleotide microarrays. In MG1655 cultured in rich medium the EvgA response regulator might be in an active form, resulting in the expression of some EvgA-regulated genes. To eliminate basal expression of EvgA regulon genes, a markerless in-frame evgAS deletion mutant, ΔevgAS, was constructed from MG1655. Nishino and Yamaguchi (20) reported that the overexpression of EvgA confers multidrug resistance in an E. coli acrAB-deficient strain. To investigate this phenomenon, we also constructed a ΔacrB strain from MG1655 and then constructed a ΔacrBΔevgAS strain from the ΔacrB strain. The pUC19 vectors, pUCevgA and pQEevgA, which are constitutive and conditional EvgA expression plasmids, respectively, were transformed into the ΔevgAS and ΔacrBΔevgAS strains by electroporation. Table 2 lists susceptibilities of constructed transformants lacking acrB. The ΔacrBΔevgAS strain harboring pUCevgA exhibited elevated resistance to erythromycin, novobiocin, sodium dodecyl sulfate (SDS), deoxycholate, and rhodamine 6G as reported for a ΔacrAB strain (20). The ΔacrBΔevgAS strain harboring pQEevgA showed elevated resistance to these agents in the presence of IPTG, suggesting that the His-tagged EvgA functions as EvgA.

The comprehensive transcript profiles of the ΔevgAS and ΔacrBΔevgAS strains bearing plasmid pUCevgA were compared to those of ΔevgAS and ΔacrBΔevgAS strains bearing plasmid pUC19, respectively. The comprehensive transcript profiles of the ΔevgAS and ΔacrBΔevgAS strains bearing plasmid pQEevgA grown in the presence of IPTG were also compared to those in the absence of IPTG. Altogether, four comparisons were performed with eight microarray data sets. The complete data set can be found at http://arep.med.harvard.edu/cgi-bin/ExpressDBecoli/EXDStart. Seventy-nine of the ORFs showed a change call of increase at all four comparisons, and 24 of the ORFs showed a change call of decrease at all four comparisons. Table 3 lists 37 of the 79 ORFs and 5 of the 24 ORFs that had an average log₂ ratio of greater than 2 or less than −2 over four comparisons (i.e., more than a fourfold change). The Microarray Suite 5.0 software uses statistical algorithms to calculate change p values (statistical significance for change calls). Among the 42 genes listed in Table 3 the least significant p value was 0.002211, suggesting that changes in expression levels of these genes are significant. Thus, these genes are target gene candidates for the EvgA response regulator.

EvgA overexpression confers acid resistance to exponential-phase cells.Five of the genes whose expression was induced by EvgA overexpression code for proteins that are known to provide protection against acid stress in E. coli. These are the gadA, gadBC, and hdeAB genes. gadA and gadB code for isozymes of glutamate decarboxylase, which catalyze the conversion of glutamate to γ-aminobutyrate (30). gadC is located downstream of gadB and is predicted to code for a glutamate:γ-aminobutyrate antiporter (10). The GadA and GadB decarboxylases and the GadC antiporter are proposed to function together to help maintain a near-neutral intracellular pH when cells are exposed to extremely acidic conditions (29). HdeA and HdeB are predicted to be periplasmic proteins. HdeA may act as a chaperone and may prevent the aggregation of periplasmic proteins denatured at low pH (7). HdeB is predicted to be a structural homologue of HdeA and to form heterodimers with HdeA (7). The fact that EvgA overexpression induced these acid protection genes prompted us to study the effect of EvgA overexpression on resistance to acidic stress. Exponential-phase ΔevgAS strains harboring pUC19, pUCevgA, and pQEevgA grown in LB broth (pH 7.0) and an exponential-phase ΔevgAS strain harboring pQEevgA grown in LB broth (pH 7.0) containing IPTG were incubated in LB broth (pH 2.5) for 1 h, and percent survival values were determined as described in Materials and Methods. The percent survival values of EvgA-lacking strains were reduced to less than 0.01% in 1 h, while EvgA overexpression strains exhibited elevated resistance to acidic conditions (Fig. 1). A time-course experiment was performed with MG1655 bearing pQEevgA (Fig. 2). In the absence of IPTG the percent survival value was reduced to less than 0.01% within 30 min. In the presence of IPTG 1.7% of the cells survived after 4 h of exposure to acidic conditions. These results indicate that EvgA overexpression confers acid resistance to exponential-phase E. coli.

• Open in new tab
• Download powerpoint

FIG. 1.

Acid resistance of EvgA overexpression strains. Various strains were grown to mid-log phase in LB broth (pH 7.0) with or without IPTG. Cells were diluted 40-fold into LB broth (pH 2.5) or EG (pH 2.5) and were incubated for 1 h at 37°C. Initial cell densities ranged from 1.2 × 10⁶ to 1.6 × 10⁷ CFU/ml. Error bars represent standard errors of the means.

• Open in new tab
• Download powerpoint

FIG. 2.

Acid resistance as a function of exposure time. The percentage of survival of exponential-phase MG1655/pQEevgA grown in the absence (○) or presence (•) of IPTG and stationary-phase MG1655 (▵) after acid exposure in LB broth (pH 2.5).

Bacteria have acid pH neutralization mechanisms based on the production of cytoplasmic amino acid decarboxylases and their cognate amino acid antiporters. Each system requires its cognate amino acid during acid challenge to allow the cells to survive. In E. coli glutamate decarboxylases GadA and GadB (30), glutamate:γ-aminobutyrate antiporter GadC (10) and arginine decarboxylase AdiA (32) have been identified. To confirm whether an acid resistance system(s) induced by EvgA overexpression requires amino acid or not, exponential-phase MG1655 harboring pQEevgA grown in LB broth (pH 7.0) containing IPTG was incubated in EG (pH 2.5) for 1 h and percent survival values were determined as described in Materials and Methods. EvgA overexpression conferred acid resistance in EG to the same level as that in LB broth (Fig. 1, bar 8). Thus, acid resistance conferred by EvgA overexpression includes amino acid-independent mechanism(s).

Identification of novel genes related to acid resistance.To identify genes responsible for the acid resistance caused by EvgA overexpression, we constructed deletion mutants of the genes that were induced more than eightfold on average by EvgA overexpression, transformed pQEevgA to the constructed mutants, and then tested acid resistance in the presence of IPTG. MG1655 harboring pQEevgA showed acid resistance in the presence of IPTG despite having a native evgS gene (Fig. 2 and 3). Therefore, MG1655 was used as the parent strain. Markerless in-frame appCBA, gadC, ydeP-b1500-ydeO, yegR-b2084-yegZ, yfdXWUVE, slp-yhiF, yhiD, hdeB, hdeA, hdeD-yhiE, yhiUV, yiiS, and yjdE deletion mutants were constructed from MG1655. Because GadC is an essential component of the glutamate-dependent acid resistance system, constructions of gadB and gadA deletion strains were omitted. Exponential-phase cultures of each strain harboring pQEevgA grown in LB broth (pH 7.0) containing IPTG were incubated in LB broth (pH 2.5) for 1 h, and percent survival values were determined as described in Materials and Methods. Although the deletions of gadC and hdeA decreased the acid resistance to 34 and 46% survival (Fig. 3, bars 3 and 13), respectively, these strains still showed more than 1,000-fold higher resistance than EvgA-lacking strains. The deletions of hdeD-yhiE and ydeP-b1500-ydeO decreased survival to 0.9% and to less than 0.01% (Fig. 3, bars 14 and 4), respectively. The deletions of yfdXWUVE slightly decreased survival to 56% (Fig. 3, bar 9), while the deletions of the other genes had no effect on survival. To identify genes responsible for the acid resistance in ydeP-b1500-ydeO and hdeD-yhiE, markerless in-frame ydeP, b1500, ydeO, hdeD, and yhiE deletion strains were constructed from MG1655 and were tested as described above. The deletions of ydeP and ydeO decreased survival to 0.7 and 0.08%, respectively (Fig. 3, bars 5 and 7), while the deletion of b1500 slightly decreased survival to 35% (Fig. 3, bar 6). The deletion of yhiE decreased survival comparable to that of hdeD-yhiE (Fig. 3, bar 16), while the deletion of hdeD had no effect on survival (Fig. 3, bar 15). Altogether, EvgA overexpression induced five genes, gadA, gadB, gadC, hdeA, and hdeB, known to provide acid protection and at least three novel acid resistance genes, ydeP, ydeO, and yhiE. These results suggest that EvgA is involved in acid resistance.

• Open in new tab
• Download powerpoint

FIG. 3.

Acid survival of exponential-phase cells. Various deletion mutants harboring pQEevgA were grown to mid-log phase in LB broth (pH 7.0) containing IPTG. Cells were diluted 40-fold into LB broth (pH 2.5) and were incubated for 1 h at 37°C. Initial cell densities ranged from 7.4 × 10⁵ to 1.3 × 10⁷ CFU/ml. Error bars represent standard errors of the means. wt, wild-type MG1655; ydeP-O, ydeP-b1500-ydeO; yegR-Z, yegR-b2084-yegZ; yfdX-E, yfdXWUVE.

E. coli in stationary phase can remain viable for several hours at pH 2.5 (9). In fact, 71% of MG1655 cells grown in LB broth (pH 7.0) to stationary phase survived even after 4 h of exposure in LB broth (pH 2.5) (Fig. 2). We also tested whether or not the genes induced by EvgA overexpression were involved in stationary-phase acid resistance. Each deletion strain grown in LB broth (pH 7.0) to stationary phase was incubated in LB broth (pH 2.5) for 2 h, and percent survival values were determined as described in Materials and Methods. The ΔgadC and ΔhdeA mutants showed four- and twofold decreases in acid resistance (Fig. 4, bars 4 and 14), respectively. Survival of the ΔhdeD-yhiE and ΔyhiE mutants was more than 4 log units lower than that of MG1655 (Fig. 4, bars 15 and 17), suggesting that YhiE is involved in stationary-phase acid resistance. The other deletion mutants showed survival rates comparable to that of MG1655.

• Open in new tab
• Download powerpoint

FIG. 4.

Acid survival of stationary-phase cells. Various deletion mutants were grown overnight in LB broth (pH 7.0). Cells were diluted 1,000-fold into LB broth (pH 2.5) and were incubated for 2 h at 37°C. Initial cell densities ranged from 2.3 × 10⁶ to 9.5 × 10⁶ CFU/ml. Error bars represent standard errors of the means. wt, wild-type MG1655; ydeP-O, ydeP-b1500-ydeO; yegR-Z, yegR-b2084-yegZ; yfdX-E, yfdXWUVE.

Identification of the efflux pump responsible for multidrug resistance.Thirty-seven drug transporter and seven membrane fusion protein (MFP) genes have been identified in the E. coli genome (15, 35, 37; also see http://www-biology.ucsd.edu/∼ipaulsen/transport/). The transporter genes are classified into the ABC (ATP-binding cassette) family (7 genes), MFS (major facilitator superfamily) (18 genes), RND (resistance nodulation cell division) family (6 genes), SMR (small multidrug resistance) family (4 genes), and MATE (multidrug and toxic compound extrusion) family (2 genes) (23). To identify the efflux pump(s) responsible for the multidrug resistance caused by EvgA overexpression, we compared the expression data of these efflux pump genes. The expressions of three efflux pump genes, emrK, yhiU, and yhiV, were significantly induced by EvgA overexpression, as shown in Table 3. The yhiU and yhiV genes form an operon and code for MFP and an RND family transporter, respectively (Fig. 1). The emrK gene coding for MFP is located in an operon immediately upstream of the emrY gene coding for an MFS transporter (Fig. 1). The expression of emrY was induced 3.0- to 9.8-fold, although only one of four change calls was increase, with the others called no change because their p values were only marginally significant (P = 0.0031 to 0.036; default threshold value, 0.0025). The induction of the emrKY operon by EvgA overexpression is consistent with the result reported by Kato et al. (13). The other 40 genes showed no significant change in expression by EvgA overexpression. Therefore, YhiUV and EmrKY are candidate contributors to the multidrug resistance.

To confirm the involvement of the YhiUV and EmrKY efflux systems in the multidrug resistance caused by EvgA overexpression, yhiUV and/or emrKY deletion mutants were isolated from the ΔacrB strain and then were transformed with pUCevgA. The ΔacrBΔevgAS strains harboring pUC19 and pUCevgA had almost the same susceptibilities to the agents tested as the ΔacrB strains harboring pUC19 and pUCevgA, respectively (Table 2). These results suggest that the level of EvgAS expressed in the ΔacrB strain has no effect on susceptibility to these agents. Therefore, we isolated the deletion strains from ΔacrB instead of ΔacrBΔevgAS. Table 2 shows susceptibilities of constructed strains to erythromycin, novobiocin, doxorubicin, crystal violet, ethidium bromide, SDS, deoxycholate, benzalkonium, and rhodamine 6G. The ΔacrBΔyhiUV strains showed no change in susceptibilities by EvgA overexpression, while the ΔacrBΔemrKY strain bearing pUCevgA showed the same level of resistance as the ΔacrB strain bearing pUCevgA. These results suggest that yhiUV is essential for the multidrug resistance caused by EvgA overexpression and that emrKY is not involved in the resistance.

The tolC gene encoding the multifunctional outer-membrane channel showed a change call of increase at all four comparisons, while it was induced only 1.2- to 2.5-fold. The tolC homologues yjcP, yohG, and ylcB (35) showed no significant change in expression by EvgA overexpression. In order to investigate the role of TolC in the multidrug resistance caused by YhiUV induction, a markerless in-frame tolC deletion mutant, ΔtolC, was constructed from the ΔacrB strain and was transformed with pUCevgA. The tolC-deficient strain harboring pUCevgA showed no increase in resistance (data not shown). These results suggest that YhiUV requires TolC for its efflux activity.