Regulation of mtrF Expression in Neisseria gonorrhoeae and Its Role in High-Level Antimicrobial Resistance

Abstract

The obligate human pathogen Neisseria gonorrhoeae uses the MtrC-MtrD-MtrE efflux pump to resist structurally diverse hydrophobic antimicrobial agents (HAs), some of which bathe mucosal surfaces that become infected during transmission of gonococci. Constitutive high-level HA resistance occurs by the loss of a repressor (MtrR) that negatively controls transcription of the mtrCDE operon. This high-level HA resistance also requires the product of the mtrF gene, which is located downstream and transcriptionally divergent from mtrCDE. MtrF is a putative inner membrane protein, but its role in HA resistance mediated by the MtrC-MtrD-MtrE efflux pump remains to be determined. High-level HA resistance can also be mediated through an induction process that requires enhanced transcription of mtrCDE when gonococci are grown in the presence of a sublethal concentration of Triton X-100. We now report that inactivation of mtrF results in a significant reduction in the induction of HA resistance and that the expression of mtrF is enhanced when gonococci are grown under inducing conditions. However, no effect was observed on the induction of mtrCDE expression in an MtrF-negative strain. The expression of mtrF was repressed by MtrR, the major repressor of mtrCDE expression. In addition to MtrR, another repressor (MpeR) can downregulate the expression of mtrF. Repression of mtrF by MtrR and MpeR was additive, demonstrating that the repressive effects mediated by these regulators are independent processes.

Neisseria gonorrhoeae is an obligate human pathogen and the causative agent of the sexually transmitted disease gonorrhea. In the United States alone, over 350,000 cases of gonorrhea were reported in 2002 (3). Recent studies have shown that untreated cases exceed treated cases, suggesting that the number of gonococcal infections is much greater than reported (36). Although the number of gonorrhea cases in the United States has declined since the 1970s, the trend of increasing antimicrobial resistance has become a significant problem.

In the last two decades, antibiotic resistance has dramatically increased for several pathogenic bacteria, including Escherichia coli, Pseudomonas aeruginosa, Vibrio cholerae, and N. gonorrhoeae (1, 12, 14, 26). The dramatic rise in microbial antibiotic resistance has become a major public health concern worldwide. The active pumping of these antibacterial agents out of the cell by efflux pump systems has been recognized as a major contributor to bacterial resistance to antibiotics. Although high-level expression of efflux pumps may permit clinically significant levels of resistance of bacteria to antimicrobial agents, the expression of these pumps and the active removal of these agents is an energy-expensive process. Therefore, the expression of efflux systems is usually tightly regulated (7).

Resistance of N. gonorrhoeae to structurally diverse hydrophobic agents (HAs) can be mediated by the mtr (multiple transferable resistance) locus (39, 40). HAs include antibiotics (including penicillin and erythromycin), nonionic detergents (including the spermacide, nonoxynol-9), and certain antimicrobial peptides that are produced at host mucosal surfaces. The mtr locus encodes an energy-dependent efflux system composed of three membrane proteins (MtrC, MtrD, and MtrE) that form the core components of the efflux pump, a transcriptional repressor (MtrR), and a gene (mtrF) that encodes an inner membrane protein (9, 37). MtrD is located in the inner membrane and functions as the transporter component of the pump. MtrE is an outer membrane protein whose function is similar to that of the Escherichia coli TolC protein and forms the channel for export of agents to the extracellular milieu. MtrC is a periplasmic protein with significant homology to a class of proteins termed membrane fusion proteins. MtrC functions as a bridge contacting the inner membrane component, MtrD, and outer membrane component, MtrE, of the efflux apparatus.

The mtrCDE operon is regulated by both positive and negative control mechanisms. These systems serve to tightly regulate the expression of the pump apparatus and to allow induction during exposure to HAs (10). The mtrR gene encodes a transcriptional repressor of mtrCDE expression (27), whereas the mtrA gene encodes a transcriptional activator similar to members of the AraC/XylS family and is required for inducible HA resistance. Mutations in mtrR or its promoter can enhance constitutive levels of HA resistance in gonococci due to increased expression of mtrCDE. In contrast, mutations in mtrA inhibit the ability of gonococci to express inducible levels of HA resistance.

The mtrF gene, which is required for high-level constitutive HA resistance mediated by the MtrC-MtrD-MtrE efflux pump, is located just downstream and transcribed divergent from mtrR. MtrF is a putative 58-kDa inner membrane protein composed of 12 transmembrane domains. Homologs of mtrF were identified in a number of diverse bacteria. With the exception of the AbgT transporter of E. coli, a transporter of p-aminobenzoyl-glutamate (13), all identified homologs were hypothetical proteins with unknown functions. Inactivation of mtrF had no discernible effect on the HA susceptibility property of gonococcal strain FA19 to hydrophobic agents, including erythromycin, Triton X-100 (TX-100), and crystal violet (37). However, inactivation of mtrF in an mtrR mutant resulted in a significant decrease in resistance to HAs. These results suggested that mtrF is necessary for high-level resistance in gonococcal strains lacking MtrR. The capacity of bacteria to express inducible levels of resistance to antimicrobial agents through efflux-dependent processes requires certain transcriptional regulatory proteins (6, 9) and membrane proteins other than those that are thought to be core components of the pump (24, 25). In order to gain insight regarding the involvement of MtrF in efflux of HAs, we examined whether it is required for inducible HA resistance and studied the regulation of mtrF expression. We also report that the expression of mtrF is negatively regulated by MtrR and a newly described transcriptional regulator, MpeR.

(A preliminary report of these findings was presented at the 14th International Pathogenic Neisseria Conference held in Milwaukee, Wis., September 5 to 10, 2004.)

MATERIALS AND METHODS

Bacterial strains, culture conditions, and HA susceptibility testing.

The bacterial strains used in the present study are listed in Table 1. E. coli strain TOP10 (Invitrogen, Carlsbad, CA) and E. coli DH5α mcr (31) were used in all cloning experiments. E. coli strains were grown in Luria-Bertani broth or on Luria-Bertani agar plates at 37°C. N. gonorrhoeae strain FA19 was used as the primary gonococcal strain. Gonococcal strains were grown on gonococcal medium base (GCB) agar (Difco Laboratories, Detroit, MI) containing glucose and iron supplements at 37°C under 3.8% (vol/vol) CO₂. All chemicals were purchased from Sigma Biochemical (St. Louis, MO). The MICs of selected antimicrobial agents against all strains were determined as previously described (8).

TABLE 1.

Bacterial strains used in this study

Strain	Relevant genotypea	Source or reference
Strains
Neisseria gonorrhoeae
FA19	Wild type	8
JF1	ΔmtrR	9
FA19 mtrF::Kmr	Inactivation of mtrF with insertion of aphA-3	This study
FA19 mtrC-lacZ	Translational fusion of the promoter region of mtrC to the lacZ gene and inserted into the chromosome (Cmr)	This study
FA19 mtrF::KmrmtrC-lacZ	Inactivation of mtrF with insertion of aphA-3, translational fusion of the promoter region of mtrC to the lacZ gene and inserted into the chromosome (Cmr)	This study
FA19 mtrF-lacZ	Translational fusion of the promoter region of mtrF to the lacZ gene and inserted into the chromosome (Cmr)	This study
JF1 mtrF-lacZ	ΔmtrR, translational fusion of the promoter region of mtrF to the lacZ gene and inserted into the chromosome (Cmr)	This study
FA19 mpeR::KmrmtrF-lacZ	Inactivation of mpeR with insertion of aphA-3, translational fusion of the promoter region of mtrF to the lacZ gene and inserted into the chromosome (Cmr)	This study
FA19 mpeR::KmrmtrC-lacZ	Inactivation of mpeR with insertion of aphA-3, translational fusion of the promoter region of mtrC to the lacZ gene and inserted into the chromosome (Cmr)	This study
JF1 mpeR::KmrmtrF-lacZ	ΔmtrR, translational fusion of the promoter region of mtrF to the lacZ gene and inserted into the chromosome (Cmr), inactivation of mpeR with the insertion of aphA-3	This study
Escherichia coli
DH5α mcr	[F− φ80dlacZΔM15 Δ(lacZYA-argF)U169 endA1 recA1 hsdR17(rK− mK+) deoR thi-1 supE44 λ−gyrA96 relA1]	31
TOP10	(F−mcrA Δ(mrr-hsdRMS-mcrBC)φ80 lacZΔM15 ΔlacX74 recA1 deoR araD139 Δ(ara-leu)7697 galU galK rpsL (StrrendAI nupG)	Invitrogen
Plasmids
pBAD-TOPO-T/A	pUC-derived protein expression vector; Apr	Invitrogen
pLES94	pUC18-derivative allowing a chromosomal lacZ fusion at the proAB site in Ng; Apr Cmr	34
pPmtrC	pLES94 containing the 248-bp upstream sequence of mtrC at the BamHI site	This study
pPmtrF	pLES94 containing the 138-bp upstream sequence of mtrF at the BamHI site	This study

^aCm^r, chloramphenicol resistance; Ap^r, ampicillin resistance.

Construction of the insertional inactivation of mtrF and mpeR.

The mtrF gene was previously inactivated in FA19 by the insertion of a kanamycin resistance (Km^r) cassette (aphA-3) to generate strain WV9 (22). Chromosomal DNA from strain WV9 was used to transform FA19, and transformants were selected on GCB agar plates containing 50 μg of kanamycin/ml. A single colony was selected and sequencing analysis of the mtrF gene confirmed the genotype of FA19 mtrF::Km^r. DNA sequencing was performed by using the Nucleic Acid Sequencing Core Facility of Emory University. The nucleotide sequence for mpeR is available through GenBank accession number AY941321. The mpeR gene was inactivated by the insertion of a kanamycin resistance cassette (aphA-3). Primers 5′mpeR (5′-ATGAACACCGCCGCCATCT-3′) and 3′mpeR (5′-GCACTTTTTCACATCCGAAGG-3′) were used to PCR amplify mpeR from FA19 chromosomal DNA. The gene was inserted into pBAD-TOPO-T/A and transformed into E. coli TOP10 as described in the manufacturer's protocol (Invitrogen). The nonpolar aphA-3 cassette (22) was removed from pUC18K after digestion with SmaI and cloned into a NaeI restriction site of mpeR. This recombinant plasmid was introduced into DH5α TOP10 by transformation. The plasmid was purified, and the inactivated mpeR::Km^r sequence was PCR amplified by using the primers 5′mpeR and 3′mpeR. The amplified product was used to transform FA19 and JF1 and selected for by growth on GCB containing 50 μg of kanamycin/ml. PCR and sequencing analysis confirmed the insertion of the kanamycin cassette into the chromosomal mpeR gene.

Construction of the mtrC-lacZ and mtrF-lacZ fusions in gonococci.

The translational lacZ fusions were constructed as previously described (34). In brief, the promoter sequence of mtrF was PCR amplified from strain FA19 by using the primers 5′PmtrF (5′-TTGGATCCGAATAACGATGTGGGCATTTTC-3′) and 3′PmtrF (5′-TTGGATCCCGACTCATCTGCTTCTCCTTAA-3′). The promoter sequence of mtrC was PCR amplified from strain FA19 by using the primers 5′PmtrC (5′-TTGGATCCCGTCTCATAATGGCGTTTTCGT-3′) and 3′PmtrC (5′-CGGGATCCCGAGCCATTATTTATCCTATCTG-3′). The resulting DNA fragments were inserted into the BamHI site of pLES94. These recombinant plasmids were introduced into DH5α TOP10 by transformation. Correct insertion and orientation was confirmed by PCR analysis and DNA sequencing analysis. The plasmids were used to transform strains FA19, JF1, FA19 mtrF::Km^r, FA19 mpeR::Km^r, and JF1 mpeR::Km^r to allow insertion into the chromosomal proAB gene. Transformants were selected on GCB agar containing 1 μg of chloramphenicol/ml.

Preparation of cell extracts and β-galactosidase assays.

The strains containing lacZ translational fusions were grown overnight on GCB agar plates containing 1 μg of chloramphenicol/ml and 50 μg of TX-100/ml where indicated. Cells were scraped, washed once with phosphate-buffered saline (pH 7.4), and resuspended in lysis buffer (0.25 mM Tris [pH 8.0]). Cells were broken by repeated freeze-thaw cycles. The cell debris was removed by centrifugation at 15,000 × g for 8 min at 4°C. β-Galactosidase assays were performed as previously described (35).

RESULTS

mtrF is required for inducible high-level resistance to TX-100 but does not influence expression of mtrCDE.

Previous studies in our laboratory demonstrated that mtrF was required for constitutive high-level HA resistance in gonococcal strains that lacked a functional MtrR repressor (37). High-level resistance can arise by either mutations in the expression or activity of MtrR or by induction of the efflux pump operon during growth of gonococci in sublethal concentrations of TX-100; both mechanisms result in increased expression of mtrCDE (9, 29). To determine whether mtrF was required for inducible high-level resistance to TX-100, the mtrF gene was inactivated by insertion of a nonpolar Km^r cassette in strain FA19. Strains FA19 and FA19 mtrF::Km^r were grown overnight on GC agar plates with or without the addition of a sublethal concentration (50 μg/ml) of TX-100. Cells were collected and spotted onto GC agar plates containing increasing concentrations of TX-100 to determine the MIC of this HA against each strain (Table 2). We noted no difference in the MICs of TX-100 for FA19 and FA19 mtrF::Km^r under noninducing conditions, confirming our earlier result that mtrF is not required for basal levels of TX-100 resistance (37). However, under inducing conditions, the TX-100 MIC against FA19 mtrF::Km^r was significantly (>20-fold) decreased from those observed for FA19, demonstrating that mtrF is required for high-level, inducible resistance of gonococci to TX-100.

TABLE 2.

MtrF is required for maximal levels of inducible resistance to TX-100

Strain	MIC (μg of TX-100/ml)a
	Without TX-100 induction	With TX-100 induction
FA19	125	> 10,000
FA19 mtrF::Kmr	125	500
FA19 mpeR::Kmr	125	> 10,000

^aAll values represent average results from at least three independent experiments.

The high-level constitutive HA resistance of gonococci that results due to mutations in mtrR or inducible resistance that can occur in the presence of sublethal concentrations of TX-100 requires increased expression of the mtrCDE operon (29, 38). In order to determine whether mtrF expression impacts expression of the mtrCDE operon under constitutive or inducible conditions, a translational reporter fusion system was used. For this purpose, the upstream region of mtrCDE, containing the promoter sequences, ribosome-binding site, and sequence encoding the first two codons of mtrC, were PCR amplified and cloned into pLES94 (34), which contains a promoter-less lacZ gene located between the proAB genes of N. gonorrhoeae. The mtrC-lacZ fusion was transformed into FA19 and FA19 mtrF::Km^r, which resulted in a single copy of the promoter of mtrC fused translationally to lacZ within the proAB chromosomal site. Strains FA19 mtrC-lacZ and FA19 mtrF::Km^r mtrC-lacZ were grown overnight on GC agar alone or GC agar with the addition of 50 μg of TX-100/ml. Cell extracts were harvested, and the expression of mtrC-lacZ was determined by measuring the β-galactosidase activity (Fig. 1). The expression of mtrC significantly increased in both FA19 and FA19 mtrF::Km^r under inducing versus noninducing conditions. No significant difference was observed for the expression of mtrC between FA19 and FA19 mtrF::Km^r under both noninducing and inducing conditions. These results demonstrate that mtrF is not required for the increased expression of mtrCDE under inducing conditions.

FIG. 1.

Regulatory affect of the mtrF mutation on the expression of mtrC. Shown is the specific β-galactosidase activity per mg of total protein in cell extracts of reporter strains, FA19 mtrC-lacZ and FA19 mtrF::Km^r mtrC-lacZ, containing the mtrC-lacZ fusion. −, Growth on GCB agar plates; +, growth on GCB agar plates containing 50 μg of TX-100/ml. The results shown are the average of three independent experiments. Error bars represent one standard deviation. The P value (Student t test) between the “−” and “+” TX-100 for both strains was <0.0001.

Expression of mtrF is induced by growth on sublethal concentrations of TX-100.

We next examined the regulation of mtrF expression during growth of gonococci in the presence of a sublethal concentration of TX-100. Expression of the mtrCDE operon is regulated by at least two independent mechanisms: induction by growth in sublethal concentrations of TX-100 and repression under constitutive conditions by MtrR. To monitor the regulation of mtrF expression, the translational lacZ fusion system described above was utilized (34). For this purpose, the upstream region of mtrF, containing the putative promoter sequences, ribosome-binding site, and sequence encoding the first two codons of mtrF, were PCR amplified and cloned into pLES94. The mtrF-lacZ fusion was transformed into strain FA19. To determine whether expression of mtrF was inducible by TX-100, strain FA19 mtrF-lacZ was grown overnight on GC agar alone or GC agar with the addition of 50 μg of TX-100/ml. Cell extracts were harvested, and the expression of mtrF-lacZ was determined by measuring the β-galactosidase activity in whole-cell extracts. Expression of mtrF increased more than threefold in the presence of TX-100, compared to GC agar alone, in FA19 (Fig. 2A). These results demonstrate that the expression of mtrF is inducible by growth of gonococci in the presence of TX-100.

FIG. 2.

Regulatory affects of TX-100 induction and the mtrR mutation on mtrF expression. Shown is the specific β-galactosidase activity per mg of total protein in cell extracts of reporter strains, FA19 mtrF-lacZ and JF1 mtrF-lacZ, containing the mtrF-lacZ fusion. (A) Effect of TX-100 induction on mtrF expression. FA19 mtrF-lacZ was grown overnight on GCB agar plates alone or containing 50 μg of TX-100/ml. (B) Effect of the mtrR mutation and TX-100 induction on mtrF expression. JF1 mtrF-lacZ was grown overnight on GCB agar plates alone or containing 50 μg/ml of TX-100. Strain JF1 has a deletion of mtrR. The results shown are the average of three independent experiments. Error bars represent one standard deviation. The P value (Student t test) between FA19 and JF1 under TX-100 induction was <0.015. The P value (Student t test) between the rest of the strains and conditions was <0.0001.

Expression of mtrF is repressed by MtrR.

Since the mtrF gene is located adjacent to mtrR and the mtrCDE operon (37), we tested whether MtrR can regulate mtrF expression during growth of gonococci under normal or inducing conditions. The mtrF-lacZ fusion described above was transformed into strain JF1, which is derived from FA19 but contains a deletion of 98% of the mtrR coding sequence (see Table 1). To determine whether expression of mtrF was repressed by MtrR, strains FA19 mtrF-lacZ and JF1 mtrF-lacZ were grown overnight on GC agar alone or GC agar with the addition of 50 μg of TX-100/ml. Cell extracts were prepared, and the expression of mtrF-lacZ was determined by measuring the β-galactosidase activity. In both the presence or the absence of TX-100, expression of mtrF increased ∼2-fold in strain JF1 over strain FA19 (Fig. 2A and B). These data demonstrate that mtrF, like mtrCDE, is subjected to repression by MtrR. However, in both strains FA19 and JF1, expression of mtrF increased ∼3-fold in the presence of TX-100 over that observed for the absence of TX-100 (Fig. 2A and B). Since no significant difference was observed for the induction of mtrF expression in strain FA19 versus JF1, these data also demonstrate that inducible expression of mtrF is an MtrR-independent process.

Previous studies have demonstrated that the negative regulation of mtrCDE, mediated by MtrR, is due to specific binding of MtrR to the promoter of the mtrCDE operon (19). Electrophoretic mobility shift analysis (EMSA) coupled with DNase footprinting analysis has identified the MtrR binding region within the mtrCDE promoter (19). Since MtrR repressed mtrF expression (Fig. 2), we next sought to determine whether MtrR could bind in a specific manner to the mtrF promoter region. As previously described (16), MtrR was expressed and purified as a fusion protein to MalE and used in EMSA. In competitive EMSA, purified MalE-MtrR demonstrated specific binding of MtrR to the mtrCDE promoter (data not shown). However, competitive EMSA analysis failed to demonstrate specific binding of MtrR to the promoter region of mtrF, suggesting that the MtrR-specific regulation of mtrF was indirect (data not shown). These data indicated that other transcriptional regulatory proteins could regulate mtrF expression.

Identification of mpeR, an AraC-like transcriptional regulator that represses mtrF.

The lack of specific binding of MtrR to the mtrF promoter prompted us to search for other regulatory proteins that control mtrF expression. A search of the N. gonorrhoeae genomic database revealed an open reading frame that would encode a member of the AraC family of transcriptional regulators, which we termed MpeR (NG0025 [http://www.stdgen.lanl.gov/stdgen/bacteria/ngon/]), and it was identified by its high homology to the conserved helix-turn-helix motif located at its C terminus (HTH-ARAC, SMART database) (Fig. 3) (17, 33). The chromosomal location of mpeR is shown in Fig. 3. Several members of the AraC family of transcriptional regulators are involved in the regulation of homologous efflux-pump systems, including MarA, SoxS, and Rob, which are involved in the regulation of the acrAB-encoded efflux pump of E. coli (5, 23). The mpeR gene encodes a 318-amino-acid protein with the predicted mass of 35.7 kDa. The predicted helix-turn-helix motif of MpeR has significant amino acid sequence identity to helix-turn-helix motifs of other AraC-like proteins including YbtA (27% identity) in E. coli (4), PchR (32%) in Pseudomonas aeruginosa (11), and to AlcR (29%), an AraC-like activator of iron transport in Bordetella pertussis (2, 28). The mpeR gene could be PCR amplified from eight additional gonococcal strains (FA1090, FA62, FA889, DG1 1918, EU75, RD5, and UU1), suggesting that it is a conserved gonococcal gene (data not shown). A search of meningococcal genomic databases (www.tigr.org and www.sanger.ac.uk) identified mpeR homologs in serogroups A (Z2491), B (MC58), and C (FAM18). However, the homologous sequence identified in MC58 is predicted to encode two polypeptides (NMB1878 and NMB1879), which may or may not encode functional proteins.

FIG. 3.

(A) Genetic organization of mpeR and adjacent genes. (B) Predicted domain structure of MpeR and amino acid alignment of the C-terminal region of MpeR to the consensus helix-turn-helix (HTH-ARAC) motif of AraC transcriptional regulators (17, 33). The top sequence represents MpeR, and the bottom sequence represents the consensus HTH motif. The asterisks the indicate the locations of identical amino acids.

To determine whether MpeR is needed for mtrF expression, mpeR was inactivated by the insertion of the nonpolar aphA-3 cassette, and the resulting plasmid construct was used to transform strain FA19. To measure the expression of mtrF, the previously described mtrF-lacZ fusion was inserted into the chromosome of strains FA19 and FA19 mpeR::Km^r. Strains FA19 mtrF-lacZ and FA19 mpeR::Km^r mtrF-lacZ were grown overnight on GC agar alone or GC agar with the addition of 50 μg of TX-100/ml. Cell extracts were harvested and the expression of mtrF-lacZ was determined by measuring the β-galactosidase activity (Fig. 4). Expression of mtrF increased approximately twofold (<0.0001) under both inducing (+TX-100) and noninducing (−TX-100) conditions in the mpeR-deficient strain, suggesting that mpeR encodes a repressor of mtrF expression.

FIG. 4.

Regulatory affects of TX-100 induction and the mpeR mutation on mtrF expression. Shown is the specific β-galactosidase activity per mg of total protein in cell extracts of reporter strains, FA19 mtrF-lacZ and FA19 mpeR::Km^r mtrF-lacZ, containing the mtrF-lacZ fusion. (A) Effect of TX-100 induction on mtrF expression. FA19 mtrF-lacZ was grown overnight on GCB agar plates alone or containing 50 μg of TX-100/ml. (B) Effect of the mtrR mutation and TX-100 induction on mtrF expression. FA19 mpeR::Km^r mtrF-lacZ was grown overnight on GCB agar plates alone or containing 50 μg of TX-100/ml. The results shown are the average of three independent experiments. Error bars represent one standard deviation. The P value (Student t test) between all of the strains and conditions tested was <0.0001.

Evidence that repression of mtrF expression by MtrR and MpeR are independent regulatory processes.

Our studies have identified two repressors of mtrF expression, MtrR and MpeR. Since we were unable to demonstrate specific binding of MtrR to the promoter region of mtrF, we suspected that MtrR regulation of mtrF is indirect. To determine whether the repressive regulation observed for MtrR was dependent on MpeR, a double mutant FA19 strain was engineered. The mpeR gene of the mtrR deletion strain, JF1, was inactivated by insertion of a nonpolar kanamycin cassette (Km^r), resulting in strain JF1 mpeR::Km^r. To measure the expression of mtrF in these mutant strains, the previously described mtrF-lacZ fusion was inserted into the chromosome of strains FA19, JF1, FA19 mpeR::Km^r, and JF1 mpeR::Km^r. All four strains were grown overnight on GC agar. Cell extracts were harvested, and the expression of mtrF-lacZ was determined by measuring the β-galactosidase activity. As previously observed, the expression of mtrF increased in an mtrR-deficient strain (JF1) and increased in an mpeR-deficient strain (FA19 mpeR::Km^r) (Fig. 5). In the strain lacking both MtrR and MpeR, expression increased greater than when either single mutation was present. Repression of mtrF by MtrR and MpeR was additive, suggesting that MtrR- and MpeR-mediated repression of mtrF expression are independent processes.

FIG. 5.

Regulatory affect of the mpeR/mtrR double mutation on mtrF expression. The specific β-galactosidase activity per milligram of total protein in cell extracts of reporter strains, FA19 mtrF-lacZ, FA19 mpeR::Km^r mtrF-lacZ, JF1 mtrF-lacZ, and JF1 mpeR::Km^r mtrF-lacZ, containing the mtrF-lacZ fusion, is shown. Strain JF1 has a deletion of mtrR. The results shown are the average of three independent experiments. Error bars represent one standard deviation. The P value (Student t test) between all of the strains tested was <0.0001.

DISCUSSION

Production of efflux pumps can result in elevated resistance of bacteria to antimicrobials recognized by the given pump (18, 20). Such pump expression can be constitutive, usually the result of loss of production of a functional transcriptional repressor that controls efflux gene expression (7). Elevated resistance of bacteria to antimicrobials can also occur transiently in the presence of sublethal levels of antimicrobials recognized by certain efflux pumps or environmental signals. This inducible resistance seems to require transcriptional activators and certain cell envelope proteins (20, 29). This inducible resistance of bacteria to antimicrobials would allow for a quick response to potentially lethal levels of an antimicrobial that is given therapeutically or via a natural host product. In the latter situation, inducible resistance of gonococci to HAs through the mtr efflux system would enhance its growth capability at mucosal surfaces that contain antimicrobial peptides and other antimicrobial agents. Indeed, recent studies by Jerse et al. (15) showed that a functional mtr efflux system was required for gonococci to cause a sustained vaginal infection in mice.

Although constitutive HA resistance in gonococci due to overproduction of the MtrC-MtrD-MtrE has been extensively studied (21, 38), considerably less is known about the molecular mechanisms that mediate inducible resistance. However, some progress in understanding inducible HA resistance has been made through studies that revealed the necessity for the transcriptional activator, MtrA (29), and energy supplied by the TonB-ExbB-ExbD system (30). In the present study we document a role for MtrF, a putative cytoplasmic membrane protein, in such resistance. Previous studies by Veal and Shafer (37) showed that MtrF is also required for high-level HA resistance that occurs in MtrR-deficient strains. In the work presented here, we set out to further define the role of mtrF in high-level HA resistance in gonococci and began to examine the regulation of mtrF expression.

MtrF was discovered by the observation that mutations in mtrF could phenotypically suppress mutations in mtrR that normally result in constitutive HA resistance (37). The mechanism of MtrF and how it modulates levels of constitutive or inducible HA resistance is not yet known. MtrF shares homology with AbgT of E. coli, a transporter of p-aminobenzoyl-glutamate (13). However, plasmid expressed recombinant MtrF failed to complement an AbgT-inactivated strain of E. coli, and therefore we do not believe that is the function of MtrF (37). Previous studies demonstrated that disruption of mtrF does not affect the transport of proteins across the inner membrane, change the lipooligosaccharide profile, or change the membrane phospholipid profile (37). One possible role for MtrF is that it directly interacts with one or more of the efflux pump components, and these interactions are necessary for high-level activity of the pump. It is interesting that putative homologs of mtrF have been identified in many diverse species of bacteria (37). The majority of these bacteria have at least one putative RND-transporter similar to MtrD, suggesting the possibility that MtrF may be a conserved member of RND-transporters.

The MIC results presented here demonstrate that mtrF is required for high-level HA resistance of gonococci due to TX-100 induction (Fig. 1). However, inactivation of mtrF had no effect on basal levels of HA resistance under noninducing conditions. Therefore, mtrF is required for high-level HA resistance that results from inactivation of MtrR or TX-100 induction. To our surprise, inactivation of mtrF had no effect on the expression of mtrC when induced by growth on sublethal concentrations of TX-100 (Fig. 2). Similar results were observed for inactivation of mtrF in an MtrR mutant gonococcal strain (data not shown). Taken together, these data suggest that mtrF does not participate in the regulation of expression of the mtrCDE-encoded efflux pump. Therefore, we believe that mtrF is required for the proper mechanism of the pump, during conditions which would normally result in high-level activity of the pump.

To gain a better understanding of the function of mtrF, we have begun to examine the regulation of mtrF expression. The major repressor of mtrCDE, MtrR, was shown to repress the expression of mtrF (Fig. 2). The regulation of mtrF by MtrR seems to be indirect because the repressor does not bind in a specific manner to a DNA sequence upstream of mtrF, as was previously seen for MtrR repression of mtrCDE. Expression of the mtrF gene was also shown to be inducible upon growth of gonococci on sublethal concentrations of TX-100 (Fig. 2).

Due to our lack of evidence for MtrR binding to the mtrF promoter, we sought to identify other transcriptional regulators that might be participating in controlling mtrF expression. We described previously an AraC-like protein (MtrA) in gonococci (29). However, results from lacZ expression experiments revealed that loss of MtrA did not impact mtrF expression (data not presented). We subsequently identified a second putative AraC regulator, MpeR (for Mtr protein efflux regulator). β-Galactosidase assays suggested that mpeR encodes a repressor of mtrF expression (Fig. 4). Although AraC-like regulators are commonly activators of transcription, several AraC-like regulators have been identified that function as both activators and repressors, including MarA in E. coli (32), YbtA (4), and PchR (11). Also similar to the repression of mtrF by MtrR, MpeR-dependent repression had no effect on the TX-100 induction of mtrF expression (Fig. 4). The level of repression of mtrF mediated by MpeR was very similar to that observed for MtrR (Fig. 2 versus Fig. 4). That result, coupled with the lack of binding of MtrR to the mtrF upstream region, suggested that these repressors may function via a single regulatory process. The repression of mtrF by MtrR and MpeR was additive, and therefore, independent processes (Fig. 5). β-Galactosidase assays showed no difference in the expression of mtrC in wild-type FA19 versus FA19 mpeR::Km^r, demonstrating that mpeR is not involved in the regulation of mtrCDE (data not shown). Moreover, MICs for TX-100 against strains FA19 and FA19 mpeR::Km^r showed no difference in resistance, confirming that MpeR is not involved in the regulation of mtrCDE expression (Table 2). Taken together, these results demonstrate that besides the regulatory mechanisms previously observed for mtrCDE (9), mtrF has at least one additional level of regulation and that this regulatory process involves MpeR. The mechanism by which MpeR modulates mtrF expression is now under investigation.