|Home | About | Journals | Submit | Contact Us | Français|
The MtrR transcriptional-regulatory protein is known to repress transcription of the mtrCDE operon, which encodes a multidrug efflux pump possessed by Neisseria gonorrhoeae that is important in the ability of gonococci to resist certain hydrophobic antibiotics, detergents, dyes, and host-derived antimicrobials. In order to determine whether MtrR can exert regulatory action on other gonococcal genes, we performed a whole-genome microarray analysis using total RNA extracted from actively growing broth cultures of isogenic MtrR-positive and MtrR-negative gonococci. We determined that, at a minimum, 69 genes are directly or indirectly subject to MtrR control, with 47 being MtrR repressed and 22 being MtrR activated. rpoH, which encodes the general stress response sigma factor RpoH (sigma 32), was found by DNA-binding studies to be directly repressed by MtrR, as it was found to bind to a DNA sequence upstream of rpoH that included sites within the rpoH promoter. MtrR also repressed the expression of certain RpoH-regulated genes, but this regulation was likely indirect and a reflection of MtrR control of rpoH expression. Inducible expression of MtrR was found to repress rpoH expression and to increase gonococcal susceptibility to hydrogen peroxide (H2O2) and an antibiotic (erythromycin) recognized by the MtrC-MtrD-MtrE efflux pump system. We propose that, apart from its ability to control the expression of the mtrCDE-encoded efflux pump operon and, as a consequence, levels of gonococcal resistance to host antimicrobials (e.g., antimicrobial peptides) recognized by the efflux pump, the ability of MtrR to regulate the expression levels of rpoH and RpoH-regulated genes also modulates levels of gonococcal susceptibility to H2O2.
Neisseria gonorrhoeae is a gram-negative strict human pathogen and the causative agent of the sexually transmitted disease gonorrhea. Worldwide, over 60 million cases of gonorrhea occur each year (7, 33). Although the incidence of disease in the United States has decreased since the 1970s, there has since been an increase in antibiotic-resistant strains reported in the United States and elsewhere (25, 33). Indeed, the prevalence of gonococci resistant to penicillin, tetracycline, and/or fluoroquinolones became so significant over the past 25 years that in 2007 the Centers for Disease Control and Prevention added N. gonorrhoeae to the list of “superbugs”.
The capacity of gonococci and other bacteria to employ drug efflux pumps to recognize and export antibiotics has attracted considerable attention, as these pumps can impact the efficacy of antibiotic therapy of infections (18, 19, 25, 30, 34, 35). Moreover, since certain efflux pumps recognize host-derived antimicrobial agents, such as antimicrobial peptides (24), there is increasing suspicion that they provide bacteria with a mechanism to escape innate host defenses (13, 21). The gonococcal MtrC-MtrD-MtrE efflux pump can export a variety of antibiotics (17, 28), including penicillin (30) and macrolides (35), and antimicrobial peptides, such as human LL-37 (24).
The mtrCDE operon is negatively regulated by the MtrR repressor (10, 16, 23), which is encoded by a gene (mtrR) immediately upstream of, but transcriptionally divergent from, mtrCDE (19). Two homodimers of MtrR bind to a DNA sequence within the mtrCDE promoter (12). Point mutations in the MtrR-binding site (16), a single-base-pair deletion within a 13-bp inverted-repeat sequence in the mtrR promoter (10), or missense mutations that cause radical amino acid replacements within the helix-turn-helix motif of MtrR all can result in derepression of mtrCDE expression (23). Derepression of mtrCDE results in decreased antimicrobial susceptibility of gonococci (10, 11, 23) and increased fitness of gonococci in an experimental murine lower genital tract infection model (31).
Consequently, it is important to define the regulatory properties of MtrR. DNA-binding proteins such as MtrR that negatively regulate bacterial efflux pump genes have been considered to be “local” gene regulators, although there is increasing evidence that they can directly or indirectly influence the expression of other genes (6, 15). In order to define the genes in gonococci regulated by MtrR, we employed a microarray analysis of total RNA extracted from mid-logarithmic-phase broth cultures of isogenic MtrR-positive and MtrR-negative gonococci. Through this analysis, we found that MtrR can directly or indirectly regulate 69 genes, which are distributed throughout the genome. We identified 47 MtrR-repressed genes (including the previously described mtrCDE operon) and 22 MtrR-activated genes. An MtrR-repressed gene of particular interest, which was the subject of this investigation, was rpoH. rpoH encodes the alternative stress response sigma factor RpoH (sigma 32), which appears to be essential for gonococcal viability even under normal growth conditions (3, 8). Earlier studies showed that rpoH expression is increased during exposure of gonococci to certain stress conditions, such as growth at elevated temperatures (8), and during bacterial contact with cultured epithelial cells (2, 3). Moreover, certain RpoH-regulated genes are differentially expressed during exposure of gonococci to H2O2 (29). Thus, apart from its regulation of mtrCDE and levels of gonococcal resistance to hydrophobic antimicrobial agents (20, 21, 24, 30), MtrR regulation of rpoH expression may have significance for gonococcal survival in vivo, particularly at sites rich in neutrophils producing large quantities of H2O2.
The bacterial strains used in this study are listed in Table Table1.1. Escherichia coli strain TOP10 (Invitrogen, Carlsbad, CA) or strain DH5α mcr was 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 (22). N. gonorrhoeae strains were grown on gonococcal medium base (GCB) agar (Difco Laboratories, Detroit, MI) containing glucose and iron supplements (22) at 37°C under 3.8% (vol/vol) CO2 or in GCB broth with supplements and sodium bicarbonate as described previously (22). The plasmids and oligonucleotide primers used in this investigation are listed in Tables Tables11 and and2.2. All chemicals were purchased from Sigma Biochemical (St. Louis, MO).
RNA was isolated from 50-ml GCB broth cultures of strains FA19 and JF1 (the same as FA19 but with mtrR deleted [ΔmtrR] ) grown to mid-log phase using a hot-phenol method as previously described (4). Samples were then DNase treated on column (Qiagen DNase kit), and the RNA was recovered using the Qiagen RNeasy Mini kit. RNA was quantified by NanoDrop1000 (NanoDrop Technolgies), and the RNA integrity was analyzed on an Agilent bioanalyzer (Agilent Technologies). The RNA samples were stored at −80°C until further use.
A custom multipathogen Affymetrix GeneChip, MPAUT1a520274F, was utilized for the microarray experiments in the present study. Probe pairs were designed for each gene, consisting of 11 probe pairs (22 features) with a perfect match and a single-nucleotide mismatch. The N. gonorrhoeae genome probe set (1,925 genes) was represented on the array.
Sample preparation and hybridization were done at the University of Iowa DNA Facility (http://dna-9.int-med.uiowa.edu/) according to the Affymetrix Genechip Expression Analysis technical manual prokaryotic protocol. Briefly, biotinylated cDNA was generated from 10 μg of total RNA and fragmented prior to hybridization to the custom chip at 45°C for 16 h. The arrays were stained using streptavidin-phycoerythrin (ProkGE-WS2v3_450 protocol; Affymetrix, Inc.), and the scanned images (Affymetrix GeneChip Scanner 3000) were processed using Genechip Operating Software version 1.4 (Affymetix, Inc.). Hybridizations were done on three separate biological replicates of each isolate.
Data files generated by Genechip Operating Software (Affymetix, Inc.) were imported into GeneSpring GX 7.3.1 (Agilent Technologies) and normalized per chip by dividing each measurement by the 50th percentile of all the measurements in that sample and per gene by dividing each gene by the median of its measurements across all samples. The cross-gene error model was operating based on replicates. The normalized data from all samples were filtered on genes flagged as present or marginal, with the resulting gene list used for further gene expression analysis and clustering. Change was expressed as the ratio of gene expression of the wild-type FA19 over the same gene expression for the mutant. Genes with a differential expression of ≥1.5-fold and a P value of 0.05 were selected.
The construction of pJF1, which contains the wild-type mtrR coding sequence and promoter sequence from strain FA19 cloned into pGCC3 (27), and strain JF6 (like JF1 but with this mtrR sequence, including its promoter, located between aspC and lctP) has been described by Folster and Shafer (5). In order to engineer an IPTG (isopropyl-β-d-thiogalactopyranoside)-inducible copy of mtrR, the mtrR coding sequence with its ribosome binding site was PCR amplified from chromosomal DNA prepared from wild-type strain FA19 using primers 5′mtrR and 3′mtrR (Table (Table2).2). The resulting DNA fragment was inserted into the PmeI and PacI sites of pGCC4, which contains the lacZ promoter (27) (kindly provided by A. Jerse and H. Seifert) to produce pJF2, and the correct orientation and nucleotide sequence were confirmed by DNA sequencing. pJF2 was digested with ClaI, and the fragment containing the gonococcal lctP gene, mtrR, ermC (an erythromycin [Ery] resistance cassette), and aspC was purified and used to transform gonococcal strain FA19 ΔmtrR. Transformations were performed as previously described (9). Transformants were selected on GCB agar containing 1 μg/ml of Ery, and the resulting strain was named JF7 (like JF1, but mtrR+).
For inducible expression of mtrR, strain JF7 was grown on GCB agar as described above. The overnight growth was then used to inoculate 50 ml of GCB broth, and the culture was grown at 37°C with shaking (200 rpm) until the optical density at 600 nm reached 0.2, at which time it was split into two equal samples. One sample received IPTG (1 mM), while the other received an equal volume of sterile distilled water. These cultures were grown for 1 h, and samples were then harvested by centrifugation. MtrR was detected in solubilized cell extracts by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and Western immunoblotting as described previously (6). MtrR was detected using a rabbit anti-MtrR antiserum (1:5,000 dilution) prepared by Protein Tech Group, Inc. (Chicago, IL), with purified MtrR fused to maltose-binding protein (MBP) (15) as the antigen and was visualized on the blot using goat anti-rabbit immunoglobulin G coupled to alkaline phosphatase (1:5,000 dilution). IPTG-induced and control samples were also evaluated for their susceptibility to Ery (0.5 μg/ml) by an efficiency-of-plating assay described previously (22), using agar plates with or without 1 mM IPTG. H2O2 susceptibility was determined using a disk diffusion assay containing filter disks (1 cm) presoaked in 3% (vol/vol) H2O2 that were placed on freshly inoculated GCB agar plates that contained or lacked 1 mM IPTG.
Translational lacZ fusions were constructed as previously described (5, 6). In brief, the promoter sequence of rpoH was amplified by PCR from strain FA19 using primers 5′PrpoH and 3′PrpoHB (Table (Table2).2). The resulting DNA fragment was inserted into the BamHI site of pLES94 (26), and the recombinant plasmid was introduced into E. coli DH5α mcr by transformation. The correct orientation of the insertion was confirmed by PCR analysis and DNA sequencing. The plasmid was used to transform strains FA19, JF1, and JF6 to allow insertion into the chromosomal proAB gene, and transformants were selected on GCB agar containing 1 μg/ml of chloramphenicol. Representative transformants obtained with the three recipient strains were termed JF8, JF9, and JF10, respectively (Table (Table1).1). Strains bearing a grpE-lacZ fusion were prepared essentially as described above, but the oligonucleotide primers 5′PgrpE and 3′PgrpE (Table (Table2)2) were used to PCR amplify a 300-bp fragment encompassing the upstream sequence of grpE and the first two codons. The strains bearing the grpE-lacZ fusion were termed JF11 (FA19 grpE-lacZ), JF12 (FA19 ΔmtrR grpE-lacZ), and JF13 (FA19 ΔmtrR mtrR+ grpE-lacZ), respectively (Table (Table1).1). For construction of an mtrR-lacZ fusion in strain FA19, oligonucleotide primers mtrC1 and pmtrR (Table (Table2)2) were used to PCR amplify mtrR; the strain bearing the mtrR-lacZ fusion was termed JF16 (FA19 mtrR-lacZ). β-Galactosidase (β-Gal) assays were conducted as previously described (5, 6).
PCR mutagenesis was used to delete the 172-bp rpoH leader sequence and was performed using the overlapping primers 5′ΔPrpoH and 3′ΔPrpoH (Table (Table2).2). First, two fragments were amplified by PCR from FA19 chromosomal DNA using the primer set pairs 5′PrpoH/3′ΔPrpoH and 3′PrpoHB/5′ΔPrpoH. The resultant fragments were purified after agarose gel electrophoresis using the Qiaquick purification kit (Qiagen Inc., Valencia, CA), and they then served as both primers and templates for a second PCR. After eight reaction cycles, primers 5′PrpoH and 3′PrpoHB were added to the PCR mixture, and amplification continued for an additional 25 cycles. The resulting DNA fragment containing the deletion was purified and served as the template for the last PCR using primers 5′PrpoH and 3′PrpoHB. The resulting DNA fragment was inserted into the BamHI site of pLES94, resulting in the rpoHΔ-lacZ construct. The recombinant plasmid was introduced into DH5α mcr by transformation. Correct insertion and orientation were confirmed by PCR analysis, and DNA-sequencing analysis confirmed the mutation of the MtrR-binding site (data not presented). The plasmid was used to transform strains FA19 and JF1 to allow insertion into the chromosomal proAB gene. Transformants were selected on GCB agar containing 1 μg/ml of chloramphenicol; representative transformants obtained with recipient strains FA19 and JF1 were termed JF14 and JF15, respectively (Table (Table11).
Electrophoretic mobility shift assays (EMSAs) using purified MBP fused to MtrR were performed as previously described (5, 6, 15). In brief, the 184-bp promoter region of rpoH lacking the leader sequence was amplified by PCR using 5′PrpoH and 3′PrpoHB from strain JF14 (Table (Table1).1). The 184-bp nonspecific probe was PCR amplified from FA19 chromosomal DNA using primers 5′ΔPrpoH and 3′PrpoHB (Table (Table2).2). The resulting PCR products were end labeled with [γ-32P]dATP using T4 polynucleotide kinase (New England Biolabs, Beverly, MA). The labeled DNA fragments (10 ng) were incubated with purified MtrR in 30 μl of reaction buffer [10 mM Tris-HCl (pH 7.5), 0.5 mM dithiothreitol, 0.5 mM EDTA, 4% (vol/vol) glycerol, 1 mM MgCl2, 50 mM NaCl, poly(dI-dC) (0.05 μg/ml), salmon sperm (200 ng/ml)] at 4°C for 20 min. Samples were subjected to electrophoresis on a 6% (wt/vol) polyacrylamide gel at 4°C, followed by autoradiography.
DNase I protection assays were performed as previously described (16). Target DNA sequences for DNase I footprints were generated by PCR using the oligonucleotide primers 5′PrpoH and 3′PrpoHB (Table (Table2)2) to generate a 184-bp promoter region of rpoH lacking the leader sequence from strain JF14. Target DNA was labeled at the 5′ end of one strand as described for EMSA. Purified MBP-MtrR protein was allowed to bind in the binding buffer as in the EMSA before CaCl2 and MgCl2 were added to final concentrations of 2.5 mM and 5 mM, respectively. Five units of DNase I (Promega, Madison, WI) were then added to the reaction mixture, which was incubated at room temperature for 1 min. Digestion was stopped by the addition of NaCl to 250 mM, and the reaction mixture was extracted with phenol-chloroform-isoamyl alcohol (25:24:1) before precipitation with 100% ethanol for 30 min at −20°C. The pellet was washed four times with 70% (vol/vol) ethanol, vacuum dried, and resuspended in gel loading buffer, which consisted of 0.1 M NaOH-formamide (1:2 [vol/vol]), 0.1% (vol/vol) xylene cyanol, and 0.1% (vol/vol) bromophenol blue. Regions protected from DNase I digestion were resolved on a 6% denaturing polyacrylamide gel, which was dried and exposed to X-ray film for autoradiography.
Gene expression data for all microarray experiments can be retrieved from the Gene Expression Omnibus (GEO) database at NCBI (http://www.ncbi.nih.gov/geo/) (accession number GSE12686).
To identify MtrR-regulated genes in N. gonorrhoeae, a microarray analysis was performed that permitted a comparison of gonococcal transcript levels in total RNA prepared from mid-logarithmic-phase cultures of wild-type strain FA19 and its isogenic mtrR deletion mutant, strain JF1 (Table (Table1).1). Using an expression difference of 1.5-fold (P ≤ 0.05), analysis of the microarray results allowed us to establish an MtrR regulon that at a minimum consists of 47 MtrR-repressed and 22 MtrR-activated genes (Table (Table3);3); these MtrR-regulated genes were distributed throughout the genome (Fig. (Fig.1).1). Four previously identified MtrR-repressed genes (mtrC, mtrD, mtrE, and mtrF) were confirmed as being MtrR repressed by the array, which validated the array as a means to identify MtrR-regulated genes. Based on the annotation of the FA1090 genome sequence (GenBank accession no. AE004969), the majority of MtrR-regulated genes were predicted to encode hypothetical proteins, but several genes encoding proteins with known or putative functions were identified (Table (Table3).3). In particular, we noted that expression of rpoH, which encodes the stress-related sigma factor RpoH, and some RpoH-regulated genes (grpE, clpB, and marR [8, 29]) were increased in strain JF1 relative to the MtrR-positive parental strain FA19, suggesting that they are MtrR repressed. Since RpoH expression in gonococci is increased during exposure to certain environmental stresses (8) when gonococci bind to cervical epithelial cells (2, 3) and is essential for viability (3, 8), we studied the ability of MtrR to regulate rpoH expression in N. gonorrhoeae.
To study the capacity of MtrR to regulate the expression of rpoH, a translational lacZ reporter fusion system was engineered. This translational fusion consisted of 365 bp of DNA sequence upstream of rpoH, which included the promoter region (14), the first two codons of rpoH (Fig. (Fig.2),2), and a promoterless lacZ gene. This construct was transformed into isogenic strains FA19 and JF1 to create the transformant strains JF8 (FA19 rpoH-lacZ) and JF9 (FA19 ΔmtrR rpoH-lacZ), respectively, resulting in a single copy of the rpoH promoter fused translationally to lacZ within the proAB chromosomal locus. β-Gal activity in cell lysates from these strains indicated that expression of rpoH increased more than eightfold in the MtrR-negative strain JF9 compared to the MtrR-positive strain JF8 (FA19 rpoH-lacZ) (Fig. (Fig.3A).3A). To confirm that these results were due to deletion of mtrR and not a polar effect on other genes, strain JF9 was complemented with the wild-type mtrR gene from strain FA19, which was inserted at a secondary site within the gonococcal chromosome (between the lctP and aspC genes) to create strain JF10 (FA19 ΔmtrR mtrR+ rpoH-lacZ) (Table (Table1).1). Using this strain, we found that complementation of the mtrR deletion in JF9 (FA19 ΔmtrR rpoH-lacZ) with the wild-type mtrR gene restored rpoH repression to a level similar to that of JF8 (like FA19, but rpoH-lacZ) (Fig. (Fig.3A3A).
The sequence of the gonococcal rpoH promoter, as well as the transcriptional start point and ribosome binding site, were defined previously (8) (Fig. (Fig.2).2). This region contains an extended 172-bp RNA leader sequence, which was previously observed to be involved in rpoH repression during growth under normal conditions (14). To determine if the leader sequence was involved in MtrR repression of rpoH, an internal deletion was engineered within the rpoH-lacZ construct and transformed into strains FA19 and JF1 (FA19 ΔmtrR) to create strains JF14 (FA19 rpoH-lacZ) and JF15 (FA19 ΔmtrR rpoH-lacZ) (Table (Table1),1), respectively, and expression was measured. We found that rpoH-lacZ expression from the wild-type promoter and the promoter with the leader deleted did not differ in either the MtrR-positive (JF14) or MtrR-negative (JF15) background (Fig. (Fig.3B),3B), indicating that MtrR-mediated repression of rpoH is independent of the leader sequence.
Gunesekere et al. (8) identified a panel of RpoH-regulated genes in gonococci, and Stohl et al. (29) showed that some of these were upregulated during bacterial exposure to H2O2. Since our microarray analysis (Table (Table3)3) suggested that a subset of these RpoH-regulated genes (clpB, grpE, and marR) is also subject to MtrR regulation, we examined the ability of MtrR to control the expression of RpoH-regulated genes in gonococci and used grpE for this purpose. We examined its expression in isogenic MtrR-positive and MtrR-negative gonococci utilizing a grpE-lacZ construct introduced into strains FA19, JF1 (FA19 ΔmtrR), and JF6 (FA19 ΔmtrR mtrR+); the transformed strains were termed JF11, JF12, and JF13, respectively (Table (Table1).1). The results indicated that expression of grpE (Fig. (Fig.4)4) was elevated in the absence of MtrR (see strain JF12) and restored to near-wild-type levels (strain JF11) by complementation with the intact mtrR gene (strain JF13).
As expression of RpoH-regulated grpE is induced during exposure of gonococci to low levels of H2O2 (29), we next evaluated whether expression of mtrR, rpoH, and grpE was altered when gonococci were exposed to the oxidizing agent. Using the lacZ fusion strains JF8 (FA19 rpoH-lacZ), JF11 (FA19 grpE-lacZ), and JF16 (FA19 mtrR-lacZ) (Table (Table1),1), we monitored the expression of rpoH, grpE, and mtrR during a brief (15-min) exposure of gonococci to 1.5 mM H2O2; this concentration of H2O2 did not reduce viability (data not presented). As is shown in Fig. Fig.5,5, we found that under these conditions, expression of both rpoH and grpE was increased significantly (P < 0.0001 and 0.002, respectively), while mtrR expression was slightly but significantly (P = 0.049) dampened. In order to determine if H2O2 induction of rpoH expression would be enhanced in the absence of MtrR, we next compared its expression when strains JF8 (FA19 rpoH-lacZ) and JF9 (like JF8, but ΔmtrR) were exposed to 1.0 mM H2O2 for 15 min. Although rpoH expression was nearly 15-fold greater in strain JF9 than in JF8, the changes in expression after H2O2 exposure were nearly identical (data not presented). We propose that although MtrR can control constitutive levels of rpoH and grpE exposure, it is not necessarily involved in their H2O2-inducible expression. Alternatively, under the experimental conditions employed, rpoH expression in JF9 reached maximal levels, and induction above the observed level was not possible or detectable.
Initial EMSA experiments showed that MtrR bound to the DNA sequence upstream of rpoH encompassing its promoter region in a specific, concentration-dependent manner and did not require the 172-bp leader sequence (data not presented). In order to determine if MtrR directly or indirectly regulates the RpoH-regulated grpE gene, an EMSA was performed using MtrR and a 305-bp probe containing the upstream DNA (including a putative promoter element) of the grpE coding sequence. Unlike its binding to the DNA sequence upstream of mtrC and rpoH, MtrR failed to bind the grpE target DNA sequence (data not presented). Accordingly, we believe that the ability of MtrR to regulate grpE is indirect and due to its capacity to repress rpoH.
In order to identify the MtrR-binding site within the rpoH promoter region, a DNase I protection assay was performed. The rpoH promoter probe described above was radiolabeled separately at its 5′ and 3′ ends and utilized in DNase I protection assays that employed increasing amounts (0 to 20 μg) of MtrR (Fig. (Fig.6).6). With increasing amounts of MtrR, an area of protection of seventeen nucleotides on the sense strand (Fig. (Fig.2)2) could be identified as being involved in binding MtrR. A second area of protection of sixteen nucleotides was observed on the antisense strand (Fig. (Fig.2),2), which overlapped the area observed on the sense strand. The region of protection indicates that MtrR binds to the rpoH promoter at an area that overlaps the −35 hexamer sequence of the rpoH promoter. Using the nucleotide sequence of the MtrR-binding site on the noncoding strand within the mtrCDE promoter (5′-GGCACGTTAGCACATA-3′) defined by Hoffman et al. (12), we were able to identify a region of similarity (73.3% identity) (Fig. (Fig.2)2) within the rpoH promoter region, which was also found by DNase I protection to bind MtrR.
In order to further connect MtrR expression with levels of rpoH expression and to assess the biologic consequence of MtrR gene control in gonococci, we next inserted an IPTG-inducible copy of the wild-type mtrR gene from strain FA19 into its transformant strain JF9 (FA19 ΔmtrR rpoH-lacZ) to create strain JF7 (FA19 ΔmtrR mtrR+ rpoH-lacZ). Strain JF7 was grown in GCB broth to mid-log phase and split into two samples, one receiving 1 mM IPTG and one left untreated. After 1 h of induction, MtrR was clearly produced by the IPTG-treated culture (Fig. (Fig.7B),7B), and rpoH expression was decreased by nearly twofold (Fig. (Fig.7C)7C) compared to the untreated sample. In order to determine whether induction of mtrR expression resulted in biological changes of gonococci, we next examined the susceptibility of the IPTG-treated and control cultures to Ery and H2O2. Ery susceptibility analysis was performed because MtrR repression of the mtrCDE-encoded efflux pump increases gonococcal susceptibility to this antibiotic and other hydrophobic antimicrobial agents (10, 11, 19). In support of this, we found that inducible expression of mtrR resulted in increased susceptibility of gonococci to Ery (Fig. (Fig.7D).7D). H2O2 assays were performed because loss of RpoH production was found to enhance the susceptibility of Brucella melitensis (1) to the oxidizing agent. Since we (unpublished observations) and others (3, 8, 14) have been unsuccessful in obtaining an rpoH null mutant, we hypothesized that under conditions where rpoH expression is dampened due to the repressive action of MtrR, gonococcal susceptibility to H2O2 would increase. Indeed, we found that inducible expression of mtrR increased gonococcal susceptibility to H2O2 (Fig. (Fig.7E7E).
We believe that the MtrR regulon defined in this study represents the minimum number of gonococcal genes under MtrR control, as other genes (farR, ponA, and pilMNOPQ) known (6, 15) to be controlled by MtrR were not identified as being regulated by ≥1.5-fold in the microarray data set. In this respect, the MtrR-repressed farR (15) and pilM (6) genes were measured as being repressed by 1.44- and 1.40-fold, respectively, in the microarray data set, while the MtrA-activated ponA gene was measured to be 1.24-fold activated. Other MtrR-controlled genes might be uncovered under different growth conditions or by exposure of gonococci to stress conditions or antimicrobial agents.
Although the microarray results indicated that expression of rpoH was modestly repressed by MtrR, the level of repression (1.58-fold) was only slightly less than those of the mtrCDE genes (1.7- to 2.2-fold). Since loss of MtrR repression of mtrCDE in strain JF1 (FA19 ΔmtrR) can have important consequences for certain gonococcal properties (levels of antimicrobial susceptibility and in vivo fitness ), we elected to use rpoH as a model gene outside of the mtrCDE locus to study the global regulatory properties of MtrR. The combined translational fusion data that examined rpoH expression in the presence or absence of MtrR, especially those involving the inducible production of MtrR (Fig. (Fig.7),7), the increased expression of three RpoH-activated genes (clpB, marR, and grpE) in the MtrR-deficient strain JF1, and the ability of MtrR to bind within the rpoH promoter (Fig. (Fig.6),6), led us to conclude that MtrR acts as a direct repressor of rpoH. As EMSA experiments failed to show MtrR binding to grpE, we also conclude that the ability of MtrR to control the expression of this RpoH-regulated gene is likely indirect and is most easily explained by the ability of MtrR to dampen rpoH expression.
What is the biological significance of MtrR regulation of rpoH? We considered the possibility that because rpoH expression can enhance bacterial resistance to killing by H2O2 (1) and because certain genes under RpoH control are upregulated in gonococci during bacterial exposure to H2O2 (29), MtrR expression in gonococci would impact the levels of such resistance. The results presented here support this hypothesis. Moreover, based on these results, we suggest that MtrR is a general regulator of levels of gonococcal susceptibility to antimicrobial agents of the innate host defense, particularly the nonoxidative and oxidative killing systems of neutrophils. Thus, by repressing mtrCDE expression, MtrR can enhance gonococcal susceptibility to LL-37, which is produced by neutrophils, epithelial cells, and certain organs (e.g., testis) (24) and is a mediator of nonoxidative killing of bacteria. As MtrR production also increases bacterial susceptibility to H2O2, its control of gene expression, along with other resistance systems, may be important in the ability of gonococci to survive oxidative killing by neutrophils. We do not yet know if the observed decrease in H2O2 susceptibility in mtrR mutants directly or indirectly involves RpoH, but the ability of MtrR to control rpoH expression suggests that this regulatory scheme is important in the general stress response. We also noted that hsp33 expression is also subject to MtrR repression (2.24-fold) (Table (Table3).3). This is of interest because Hsp33 has been implicated in bacterial responses to heat and peroxide (32). Unlike, grpE, MtrR regulation of hsp33 may be direct, as a potential MtrR-binding site consisting of 31 bp is positioned from nucleotides −110 to −141 (data not presented); this putative binding site is 61.3% identical to the 31-bp region upstream of mtrCDE shown by Lucas et al. (16) to bind MtrR. Taking these data together, we propose that naturally occurring mtrR mutants (22, 23, 34, 35) may have an advantage during infection, based on their increased resistance to LL-37 (24) and H2O2 (Fig. (Fig.7).7). This linkage of bacterial resistance to distinct mediators of innate host defense through MtrR may help to explain the recent observation of Warner et al. (31) that mtrR mutants have a competitive advantage over otherwise wild-type gonococci in a female mouse lower genital tract infection model. The potential roles of other transcriptional regulators similar to MtrR in modulating bacterial resistance to innate host defense mechanisms should be considered.
Against this background, it should be asked why mtrR mutants are not the dominant strain in the community. At first glance, it seems to be a disadvantage for gonococci to have maintained mtrR, as its presence would downregulate the expression of genes needed for resistance to antimicrobials. We propose that because MtrR can also transcriptionally activate certain genes (Table (Table3)3) involved in metabolism (e.g., abpE, glnE, and rfbB), the need to resist host-derived antimicrobials through constitutive MtrR-repressible systems may be essential only at those sites rich in these agents. For most strains and infections, the need to have expression of MtrR-activated genes is more critical and overrides any advantage afforded by mtrR mutations, particularly since the mtrCDE efflux operon can be transcriptionally induced by MtrA in the presence of sublethal levels of antimicrobials (20), even in the presence of a functional MtrR.
We thank L. Pucko for help in manuscript preparation and submission, N. Kamal and Y. Zalucki for critical reading of the paper, M. Apicella for facilitating access to the microarray slides, and the staff at the University of Iowa DNA Facility for sample preparation and hybridizations.
This work was supported by NIH grants AI-21150 (W.M.S.), RR-015564 (J. Iandolo and D.W.D.), and RR016478 (F. Waxman and D.W.D.) and by a VA Merit Award to W.M.S. W.M.S. was supported by a Senior Research Career Scientist Award from the Department of Veterans Affairs Medical Research Service.
Published ahead of print on 31 October 2008.