Role of Pseudomonas aeruginosa PhoP-PhoQ in resistance to antimicrobial cationic peptides and aminoglycosides

Emma L. A. Macfarlane1, Agnieszka Kwasnicka1 and Robert E. W. Hancock1

Department of Microbiology, #300, 6174 University Boulevard, University of British Columbia, Vancouver, British Columbia, Canada V6T 1Z31

Author for correspondence: Robert E. W. Hancock. Tel: +1 604 822 2682. Fax: +1 604 822 6041. e-mail: bob{at}cmdr.ubc.ca


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Resistance to the polycationic antibiotic polymyxin B and expression of the outer-membrane protein OprH in the opportunistic pathogen Pseudomonas aeruginosa both involve the PhoP-PhoQ two-component regulatory system. The genes for this system form an operon with oprH, oprH-phoP-phoQ, that responds to Mg2+ starvation and PhoP levels. In this study, the Mg2+-regulated promoter for this operon was mapped upstream of oprH by primer-extension experiments. An oprH::xylE-GmR mutant H855 was constructed and measurement of the catechol 2,3-dioxygenase activity expressed from this transcriptional fusion provided evidence for a second, weak promoter for phoP-phoQ. Wild-type P. aeruginosa PAO1 strain H103 was found to exhibit Mg2+-regulated resistance to the {alpha}-helical antimicrobial cationic peptide CP28 in addition to its previously characterized resistance to polymyxin B. Resistance to this peptide was unchanged in the OprH-null mutant H855 and a PhoP-null mutant H851. In contrast, PhoQ-null mutant H854 demonstrated constitutive CP28 resistance. Northern blot analysis revealed constitutive expression of phoP in this strain, implicating PhoP-PhoQ in the resistance of P. aeruginosa to cationic peptides. Furthermore, all three null-mutant strains demonstrated increased resistance to the aminoglycoside antibiotics streptomycin, kanamycin and amikacin. Two additional mutant strains, H895 and H896, were constructed that carried unmarked deletions in oprH and were found to exhibit aminoglycoside susceptibility equivalent to that of the wild-type. This result provided definitive evidence that OprH is not involved in P. aeruginosa aminoglycoside resistance and that the changes in resistance in strain H855 and a previously reported oprH mutant were due to polar effects on phoP-phoQ rather than loss of OprH expression. A role for PhoP-PhoQ in resistance to aminoglycosides is envisaged that is distinct from that in resistance to cationic peptides and polymyxin B.

Keywords: PhoP-PhoQ, antimicrobial cationic peptides, aminoglycoside resistance, Pseudomonas aeruginosa


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Antibiotic resistance is becoming increasingly prevalent in pathogenic bacteria. As research into the mechanisms behind this resistance intensifies, the complex nature of the regulation of antibiotic resistance is becoming evident. Bacteria frequently employ two-component regulatory systems as a means of co-ordinating their often widely disparate responses to environmental stimuli. A growing number of such systems are being discovered that play a global regulatory role within the bacterial cell, and factors that influence antibiotic resistance and/or virulence frequently fall under their control. One notable example is PhoP-PhoQ in Salmonella typhimurium, which regulates the expression of at least forty genes in response to extracellular Mg2+ concentrations. The sensor protein PhoQ responds to extracellular concentrations of Mg2+ ions and, under conditions of Mg2+ starvation, first auto-phosphorylates at a conserved histidine residue, then activates the regulator protein PhoP by a phospho-transfer reaction (Garcia Vescovi et al., 1996 ). The phoP-phoQ locus is subject to auto-regulation, such that low extracellular Mg2+ concentrations upregulate transcription of this operon from an inducible promoter in a PhoP-dependent manner (Soncini et al., 1995 ). Other regulators under the control of PhoP-PhoQ include PmrA-PmrB (Gunn & Miller, 1996 ; Soncini & Groisman, 1996 ), a two-component regulatory system responsible for structural changes in the outer-membrane lipopolysaccharide that lead to polymyxin B resistance (Gunn et al., 1998 ; Helander et al., 1994 ), and the transcriptional activator HilA, which is a key regulator of Salmonella pathogenicity island 1 and thus controls expression of several virulence factors (Bajaj et al., 1996 ). In addition, PhoP-PhoQ also directly regulates S. typhimurium resistance to antimicrobial peptides (Fields et al., 1989 ; Miller et al., 1989 ).

The Gram-negative bacterium Pseudomonas aeruginosa is an important opportunistic pathogen. Chronic infections due to this organism are prevalent in cystic fibrosis patients and are frequently recalcitrant to treatment. In addition to displaying high levels of intrinsic antibiotic resistance, P. aeruginosa frequently converts to a mucoid state resulting in a rapid adaptive resistance that accounts for the high failure rate of antibiotic therapy in eradicating these infections.

Sequencing of the P. aeruginosa PAO1 genome has recently been completed (http://www.pseudomonas.com) and should provide some insight into the pathogenesis of this bacterium. One striking feature of the predicted protein complement of P. aeruginosa is the presence of a disproportionately large number of regulatory proteins. Recently we identified the PhoP-PhoQ homologues in P. aeruginosa (Macfarlane et al., 1999 ). The genes for this regulatory system lie immediately downstream of the oprH gene, which encodes a small outer-membrane protein OprH that is highly expressed under Mg2+-starvation conditions. The three genes oprH-phoP-phoQ form an operon that is under the joint control of PhoP and Mg2+ ion concentrations. Similar to its counterpart in S. typhimurium, P. aeruginosa PhoP-PhoQ is highly expressed under Mg2+-starvation conditions and is involved in resistance to polymyxin B. P. aeruginosa PAK has been shown to exhibit LPS structural modifications when grown under Mg2+-deficient conditions (Ernst et al., 1999 ), and polymyxin resistance in P. aeruginosa, like that in S. typhimurium, may thus be related to altered outer-membrane permeability. In S. typhimurium, PhoP-PhoQ mediates resistance through activation of PmrA-PmrB, and a PhoP-null strain is therefore rendered supersusceptible to polymyxin B. However, our PhoP-null strain of P. aeruginosa PAO1 (strain H851) retained polymyxin B resistance under Mg2+-deficient growth conditions, and a PhoQ-null strain (H854) exhibited constitutive polymyxin B resistance. In accord with the large number of regulators in P. aeruginosa, the involvement of another regulatory system(s) in the resistance of this organism appears likely. However, in contrast to the situation in S. typhimurium, PhoP is not required for activation of this system(s) under low-Mg2+ conditions.

In this paper we have further characterized the oprH-phoP-phoQ operon and demonstrated that the two-component regulatory system PhoP-PhoQ influences the resistance of P. aeruginosa to both cationic antimicrobial peptides and aminoglycoside antibiotics.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bacterial strains, growth conditions and general procedures.
Bacterial strains and plasmids used in this study are listed in Table 1. All strains were maintained at 37 °C on LB agar, or grown in LB broth, supplemented with appropriate antibiotics unless otherwise stated. Both LB (Lennox) broth containing 5 g NaCl l-1, and LB (Miller) broth containing 10 g NaCl l-1 were used. For conjugation experiments, Escherichia coli strain S17-1 (Simon et al., 1983 ) was grown at 30 °C and P. aeruginosa strain H103 was grown at 42 °C. To study the effects of Mg2+ concentration, BM2-glucose minimal medium (Gilleland et al., 1974 ) containing 20 µM (low) or 2 mM (high) MgSO4 was used. Antibiotics were used at the following concentrations: for E. coli, 100 µg ampicillin ml-1, 10 µg gentamicin ml-1; for P. aeruginosa, 300–350 µg carbenicillin ml-1, 15 µg gentamicin ml-1. Unless otherwise specified, plasmids were transformed into bacterial strains by electroporation using a Gene Pulser (Bio-Rad) and 0·1 cm gap cuvettes following published protocols (Sambrook et al., 1989 ; Dennis & Sokol, 1995 ). For DNA manipulations, standard protocols were followed (Sambrook et al., 1989 ; Ausubel et al., 1987 ). Total cellular RNA for Northern blots and primer-extension analysis was isolated from mid-exponential phase cultures using the RNeasy Mini Kit (Qiagen). Oligonucleotides were synthesized on an Applied Biosystems model 392 DNA–RNA Synthesizer. Catechol 2,3-dioxygenase activity assays were carried out as described previously (Macfarlane et al., 1999 ).


View this table:
[in this window]
[in a new window]
 
Table 1. Bacterial strains and plasmids used in this study

 
Northern blots and primer-extension analysis.
Northern blot experiments were carried out using methods and primers described previously for strain H103 (Macfarlane et al., 1999 ). For the primer-extension experiments, primer OprH-rev2 (5'-TGGATGTTGTTGCTGGTCTC-3') was designed that was complementary to nt 115–96 of the oprH gene. This primer was end-labelled with [{gamma}-32P]dATP and hybridized at 60 °C to RNA (5 µg) from strain H103 in 50 mM Tris/HCl, pH 8·3, containing 20 mM KCl (total volume 10 µl). Primer extension was initiated by the addition of 100 U Superscript reverse transcriptase (Gibco-BRL) in the presence of 50 mM Tris/HCl (pH 8), 40 mM KCl, 6 mM MgCl2, 20 mM dNTPs and 10 U ribonuclease inhibitor (final volume 30 µl) and cDNA synthesis was allowed to proceed for 60 min at 45 °C. After treatment with DNase-free RNase for 15 min at 37 °C, the cDNA was isolated by ethanol precipitation followed by resuspension in 3–4 µl Promega fmol loading buffer. After denaturation, samples were loaded onto a 6% acrylamide sequencing gel together with dideoxy sequencing reactions covering the appropriate region of plasmid pGB22 (Bell et al., 1991 ). Sequencing reactions used the OprH-rev2 primer and the Promega fmol DNA cycle sequencing system. Control primer-extension reactions using tRNA alone or RNA treated with RNase prior to cDNA synthesis were performed in parallel.

Construction of strains H855 (oprH::xylE-GmR), H895 ({Delta}oprH73) and H896 ({Delta}oprHC).
To construct strain H855, a 1·2 kb SmaI–HincII fragment containing the oprH gene was excised from plasmid pAK9 (Macfarlane et al., 1999 ) and cloned into the SmaI site of the gene-replacement vector pEX100T (Schweizer & Hoang, 1995 ) to give pEXH. The xylE-GmR cassette from plasmid pX1918GT (Schweizer & Hoang, 1995 ) was then cloned between the two PstI sites that are situated 18 bp apart within the oprH gene. Gene replacement in strain H103 using this construct was carried out as described previously for strains H851 and H854 (Macfarlane et al., 1999 ). Southern blot analysis following standard protocols (Ausubel et al., 1987 ) and using probes complementary to either oprH or xylE confirmed the presence of the xylE-GmR cassette within oprH in strain H855. Plasmid pEMH4a was constructed by subcloning a 0·7 kb SstI–XbaI fragment carrying the oprH gene from pBHR20 (Rehm & Hancock, 1996 ) into pUCP21 (West et al., 1994 ).

To construct the oprH deletion strains H895 and H896, the gene-replacement vectors pEXH{Delta}73 and pEXH{Delta}C were constructed as follows. Primer {Delta}oprH2 was designed with the sequence 5'-CTGCTGGTGTCGGAGGCATTCTCGTAGGTC-3' that was complementary to nt 263–234 of oprH. The three underlined bases replaced CAT in the coding sequence and introduced a BsmI site. A 410 bp fragment containing the 5' end of oprH plus 144 bp of vector sequence was PCR amplified from pEMH4a using primer {Delta}oprH2 and a second primer, pUCPlac, that was complementary to the vector lac promoter. A fragment of 557 bp from the 5' end of oprH was removed from pEXH by SmaI/BsmI digestion and replaced with the HincII/BsmI-digested PCR amplicon. The resulting plasmid, pEXH{Delta}73, contained the oprH gene with a 219 bp in-frame deletion (nt 709–927 inclusive). For plasmid pEXH{Delta}C, primer {Delta}oprH1 (5'-CTGAGCAAGAATGCCTC|ACCAACGCCAGCACCGAG-3') was designed that contained a 28 bp deletion (nt 930–957 of oprH) indicated by the vertical line. A 337 bp fragment containing the 3' end of oprH was PCR amplified from pGB22 (Bell et al., 1991 ) using this primer and RT-PCR3' (Macfarlane et al., 1999 ). pGB22 was digested with SmaI and religated to remove the vector KpnI site and give plasmid pEMH2. Replacement of a KpnI–BsmI fragment containing the 3' end of oprH in pEMH2 with the KpnI–BsmI-digested PCR amplicon yielded plasmid pEMH2{Delta}C. The 28 bp deletion in the oprH gene resulted in a frameshift that would prematurely terminate translation 83 bp upstream of the regular oprH TAA stop codon. Finally, the modified oprH gene was excised from pEMH2{Delta}C by SmaI–HincII digestion and cloned into the SmaI site of pEX100T to give pEXH{Delta}C.

Gene replacement in strain H103 using pEXH{Delta}73 and pEXH{Delta}C was carried out as described previously (Macfarlane et al., 1999 ) with the exception that single crossover events were identified by selection on BM2-glucose minimal medium containing carbenicillin alone. Double crossover events were identified as sucrose-resistant, carbenicillin-sensitive colonies. Mid-exponential phase cultures of these strains were screened for OprH expression by Western immunoblot using anti-OprH specific antiserum (Macfarlane et al., 1999 ). Genomic DNA was subsequently isolated from OprH strains and the appropriate deletion in oprH was confirmed by PCR.

MIC determinations and killing assays.
Aminoglycoside MIC values were determined using the standard two-fold microtitre broth dilution protocol (Amsterdam, 1991 ) starting with mid-exponential phase cultures. Peptide MIC values were determined using a modified version of this protocol (Wu & Hancock, 1999 ) to prevent adhesion of the peptides to the walls of the microtitre plates that would artificially elevate the MIC values. All MIC values were read after a 48 h incubation due to the slow growth rate of strain H854. Killing assays were carried out as described previously (Macfarlane et al., 1999 ) with the following exceptions. For CP28, cultures were diluted 1:100 into 30 mM sodium phosphate buffer, pH 7, containing 150 mM NaCl and 8 µg CP28 ml-1. For streptomycin, cultures were diluted 1:200 into pre-warmed LB (Miller) broth containing 16 µg streptomycin ml-1 and killing was carried out at 37 °C.

Streptomycin-assisted lysozyme lysis assays.
Aliquots (1 ml each) of mid-exponential phase cultures (OD600 0·4–0·8) grown in LB (Miller) broth were centrifuged and washed once with 30 mM sodium phosphate buffer, pH 7. The cells were then resuspended in 1 ml of the same buffer and placed in a cuvette. Streptomycin (450 µg ml-1) and lysozyme (50 µg ml-1) were added simultaneously to the cuvette and cell lysis was monitored by following the decrease in optical density at 600 nm.


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
A single Mg2+-regulated promoter situated upstream of oprH transcribes the oprH-phoP-phoQ operon
In an earlier study we used Western blot analysis of OprH expression in P. aeruginosa phoP and phoQ mutants to demonstrate that Mg2+-regulated transcription of the oprH-phoP-phoQ operon is dependent on the presence of PhoP. To confirm the effects of the phoP and phoQ mutations on transcription of this operon, we analysed total cellular RNA from strains H851 (phoP::xylE-GmR) and H854 (phoQ::xylE-GmR) (Table 1) by Northern blot hybridization. Primers complementary to regions internal to the oprH and phoP genes were used to probe RNA isolated from cultures of the two mutant strains grown to mid-exponential phase in high- and low-Mg2+ medium. Using this method, a high level of transcription was observed in the PhoQ-null mutant H854. Large quantities of transcripts corresponding to oprH and significant but diminishing amounts of oprH-phoP and oprH-phoP-phoQ::xylE transcripts were identified in RNA isolated from this strain under both high- and low-Mg2+ growth conditions (Fig. 1). We had previously observed by Western blot analysis that this mutant constitutively overexpressed OprH (Macfarlane et al., 1999 ). The results of the Northern blot analysis confirmed our assumption that phoP is also constitutively overexpressed in strain H854. We were unable to detect any transcripts containing phoP-phoQ alone in this strain or in the wild-type strain H103 (Macfarlane et al., 1999 ). In contrast to strain H854, no transcripts containing oprH, phoP or phoQ could be detected by Northern blot analysis of RNA isolated from the PhoP-null strain H851 under either growth condition (data not shown). These results were consistent with our previous observation that PhoP is an essential positive regulator of transcription for the oprH-phoP-phoQ operon.



View larger version (49K):
[in this window]
[in a new window]
 
Fig. 1. Northern blot analysis of total cellular RNA from the PhoQ-null mutant H854 isolated under low-Mg2+ (lane 1) or high-Mg2+ (lane 2, both slightly overloaded) growth conditions. Blots were probed with primers complementary to (a) oprH, or (b) phoP.

 
To determine the transcriptional start site of this operon, we conducted primer-extension experiments using total cellular RNA isolated from the wild-type strain H103 and a primer complementary to an internal region of the oprH gene (see Methods). A single major product was observed under low-Mg2+ growth conditions (Fig. 2a, lane 2), which was only faintly visible under high-Mg2+ conditions (Fig. 2a, lane 1), and indicated the presence of a single, Mg2+-regulated promoter upstream of oprH. Examination of the sequence upstream of the transcriptional start site revealed only a weak -10 consensus sequence beginning 12 bp downstream of the four hexanucleotide repeats found upstream of the operon (Fig. 2b). No -35 consensus sequence could be identified; however, this is a common finding for P. aeruginosa regulated promoters (Rothmel et al., 1991 ). Analogous primer-extension experiments performed with a primer complementary to phoP failed to detect any transcriptional start sites immediately upstream of the phoP gene (data not shown).



View larger version (18K):
[in this window]
[in a new window]
 
Fig. 2. (a) Primer-extension analysis using a primer complementary to oprH and total cellular RNA isolated from wild-type strain H103 grown under high-Mg2+ (lane 1) or low-Mg2+ (lane 2) growth conditions. Lanes G, A, T and C indicate the dideoxy sequencing reactions. The arrow indicates the transcriptional start. (b) Schematic representation of the oprH-phoP-phoQ operon showing the sequence immediately upstream of oprH. Hexanucleotide repeats are underlined, the putative -10 consensus sequence is shown in bold, the transcriptional start site determined by primer-extension experiments is shown underlined in bold and the oprH ATG start codon is in bold italics.

 
Transcription of an oprH::xylE-GmR fusion is weakly Mg2+ regulated
Previous studies from this laboratory (Nicas & Hancock, 1980 ; Young et al., 1992 ) ascribed a role for overexpressed OprH in polymyxin B and aminoglycoside resistance. However, our discovery that oprH formed an operon with phoP and phoQ cast some doubt on these conclusions. To further elucidate the role of each of the three genes in the oprH-phoP-phoQ operon, we constructed a chromosomal knockout of oprH. A xylE-GmR cassette was inserted into two PstI sites within the oprH gene resulting in the removal of 18 bp of oprH coding sequence and the creation of a transcriptional fusion (see Methods). Gene replacement in the resulting oprH::xylE-GmR mutant strain H855 (Table 1) was confirmed by Southern blot analysis using probes complementary to oprH and to xylE (data not shown). As expected, Western blot analysis showed strain H855 to be deficient in expression of OprH protein under normal inducing conditions (data not shown).

Strain H855 was transformed with plasmids carrying the oprH, phoP, phoQ and phoP-phoQ genes (pEMH4a, pEMR3, pEMQ1a and pEMPQ2a, respectively; Table 1). Cultures of H855 alone and harbouring each of these plasmids were grown to mid-exponential phase in both high- and low-Mg2+ media and the catechol 2,3-dioxygenase activity expressed from the xylE transcriptional fusion was measured. The results, shown in Table 2, indicated that transcription of oprH was still Mg2+ regulated in strain H855 with the level of catechol 2,3-dioxygenase activity measured under low-Mg2+ growth conditions being between 16- and 23-fold higher than that measured under high-Mg2+ conditions. The xylE-GmR cassette in strain H855 is flanked by omega fragments (Schweizer & Hoang, 1995 ); therefore the oprH::xylE-GmR construct was assumed to exert a polar effect on the downstream phoP-phoQ genes. Since we have shown that transcription of oprH is dependent on the presence of PhoP (Macfarlane et al., 1999 ), the observed Mg2+ regulation of the transcriptional fusion in strain H855 strongly suggested that phoP-phoQ was being weakly transcribed from a second promoter. Presumably the level of this transcript was too low to be detected in the Northern blots and primer-extension experiments described above, even when the autoradiograms were overexposed (data not shown).


View this table:
[in this window]
[in a new window]
 
Table 2. Catechol 2,3-dioxygenase activity of the oprH::xylE-GmR transcriptional fusion in P. aeruginosa strain H855 harbouring PhoP, PhoQ and OprH expression plasmids

 
When introduced into strain H855, plasmids pEMR3 (phoP) and pEMPQ2a (phoP-phoQ) had opposing effects on the transcription of the oprH::xylE fusion. Overexpression of PhoP from pEMR3 led to a large increase in catechol 2,3-dioxygenase activity expressed in high-Mg2+ medium, and a slight decrease in low-Mg2+ medium, which essentially eliminated the Mg2+ regulation of oprH::xylE transcription. Conversely, the presence of both PhoP and PhoQ proteins expressed from pEMPQ2a led to a much higher degree of Mg2+ regulation than for strain H855 alone. A 759-fold difference in activity was measured between the two media, caused largely by the suppression of expression in high-Mg2+ medium. As would be expected, neither pEMQ1a (phoQ) nor pEMH4a (oprH) had any effect on transcription of the oprH::xylE fusion.

The OprH-null strain H855 retains Mg2+-regulated polymyxin B resistance
The resistance of the OprH-null strain H855 to the polycationic antibiotic polymyxin B was determined by a killing assay. Cultures of this strain with and without the OprH, PhoP and PhoQ expression plasmids were grown to mid-exponential phase in BM2-glucose minimal medium and treated with 8 µg polymyxin B ml-1. The number of survivors after 5 min was determined by a plate count (Table 3). Polymyxin B susceptibility in strain H855 remained similar to that of the wild-type strain H103 under both high- and low-Mg2+ growth conditions. Plasmids carrying the phoQ (pEMQ1a), phoP-phoQ (pEMPQ2a) or oprH (pEMH4a) genes had no effect on the polymyxin B resistance of strain H855. In contrast, constitutive overexpression of PhoP resulting from the presence of pEMR3 (phoP) led to constitutive polymyxin B resistance in this strain. This was a similar effect to that of pEMR3 on the polymyxin B resistance of the PhoP-null strain H851 (Macfarlane et al., 1999 ).


View this table:
[in this window]
[in a new window]
 
Table 3. Killing assays with polymyxin B (8 µg ml-1) for strain H855 (oprH::xylE-GmR) harbouring PhoP, PhoQ and OprH expression plasmids

 
P. aeruginosa resistance to the {alpha}-helical cationic peptide CP28 is Mg2+ regulated and involves PhoP-PhoQ
The PhoP-PhoQ two-component regulatory system of S. typhimurium has been shown to regulate resistance to defensins and other antimicrobial peptides in response to extracellular Mg2+ ion concentrations (Garcia Vescovi et al., 1996 ; Guo et al., 1998 ; Miller et al., 1990 ). The Mg2+ regulation of polymyxin B resistance in P. aeruginosa is well documented (Brown & Melling, 1969 ; Nicas & Hancock, 1980 ), and we have shown that this resistance is also affected by PhoP and PhoQ (Macfarlane et al., 1999 ). To determine whether antimicrobial-peptide resistance in P. aeruginosa was similarly dependent on PhoP-PhoQ and the Mg2+ ion content of the growth medium, we conducted killing assays on the mutant strains H855 (oprH::xylE-GmR), H851 (phoP::xylE-GmR) and H854 (phoQ::xylE-GmR) (Table 1) using the antimicrobial peptide CP28 (Piers et al., 1994 ). CP28 is an {alpha}-helical cationic peptide, the sequence of which is based on the cecropin-mellitin hybrid peptide CEME (Piers et al., 1994 ). For the killing assays, mid-exponential phase cultures grown in BM2-glucose medium containing high or low concentrations of Mg2+ ions were diluted 100-fold into 30 mM sodium phosphate buffer, pH 7, containing 150 mM NaCl and 8 µg CP28 ml-1 and survivors were assessed by plate counts. The antimicrobial action of CP28 is unaffected by NaCl concentrations up to 300 mM (Friedrich et al., 1999 ), and the addition of NaCl to the assay buffer helped to stabilize the cells under the assay conditions. The results of these experiments, given in Table 4, clearly indicated that P. aeruginosa resistance to CP28 is Mg2+ regulated. After a 5 min exposure to 8 µg CP28 ml-1, only a 3% survival rate was observed for the wild-type strain H103 grown in high-Mg2+ medium compared to a 63% survival rate for the same strain grown in low-Mg2+ medium. Results for the PhoP-null (H851) and OprH-null (H855) strains were comparable to the wild-type. However, the PhoQ-null mutant H854 demonstrated a higher level of resistance to CP28 under low-Mg2+ growth conditions (100% survival) together with significant resistance in high-Mg2+ medium (86% survival). These results were analogous to those previously reported for polymyxin B (Macfarlane et al., 1999 ).


View this table:
[in this window]
[in a new window]
 
Table 4. Killing assays for P. aeruginosa wild-type and mutant strains with CP28 or streptomycin

 
In view of the effect of PhoP-PhoQ on CP28 resistance in P. aeruginosa, we screened the oprH, phoP and phoQ mutants for changes in resistance to a series of cationic peptides of varying structure by measurement of the MICs. Divalent cations exert an antagonistic effect on the action of many cationic peptides (Friedrich et al., 1999 ), making Mg2+ regulation of resistance to these compounds difficult to detect by measurement of the MIC values. However, the results of killing assays with these mutant strains and either polymyxin B (Macfarlane et al., 1999 ) or CP28 indicated that PhoP-PhoQ related effects could be seen most clearly under high-Mg2+ conditions. Therefore, resistance to five peptides was determined for strains grown in BM2-glucose minimal medium supplemented with 2 mM MgSO4 (Table 5). Although the differences in MICs for the mutant strains were modest (twofold), they were consistently observed in four to six independent experiments. It is noteworthy that under these conditions, the mutant strains demonstrated the same pattern of resistance to CP28 as that to polymyxin B, consistent with the killing-assay results. Resistance to the {alpha}-helical peptide CP29 [an analogue of CP28 (Friedrich et al., 1999 )] and to the protegrin-like peptide IB367 also followed this pattern, with increased resistance being seen in the PhoQ-null strain H854. Peptide CP11CN is a synthetic analogue of the bovine neutrophil peptide indolicidin with improved activity towards Gram-negative bacteria (Falla & Hancock, 1997 ). Resistance to this peptide was increased in the PhoP-null H851 and OprH-null H855 strains, but remained the same as wild-type for strain H854. Resistance to polyphemusin, a synthetic version of the natural peptide isolated from the haemocytes of horseshoe crabs (Iwanaga et al., 1994 ), was unaffected in all three mutants.


View this table:
[in this window]
[in a new window]
 
Table 5. MIC values determined for P. aeruginosa wild-type and mutant strains grown in BM2-glucose minimal medium supplemented with 2 mM MgSO4

 
P. aeruginosa resistance to aminoglycosides involves PhoP-PhoQ
In a manner analogous to polymyxin B resistance, P. aeruginosa resistance to the aminoglycoside antibiotics streptomycin and gentamicin has previously been shown to be dependent on the Mg2+ concentration of the growth medium (Hancock et al., 1981 ). In view of the influence of PhoP-PhoQ on resistance to both polymyxin B (Macfarlane et al., 1999 ) and cationic peptides, we investigated whether our mutant strains also displayed changes in aminoglycoside resistance.

MIC determinations revealed increased resistance (two- to fourfold) to a number of aminoglycoside antibiotics for the three mutant strains grown under high-Mg2+ conditions (Table 5). As was the case for cationic peptides, high divalent cation concentrations also interfered with the action of aminoglycosides (Zimelis & Jackson, 1973 ), which accounts for the increased MIC values for streptomycin measured under high-Mg2+ conditions compared to those in LB medium, which contains only moderate concentrations of Mg2+. Although all three mutant strains carry the aacC1 gene encoding the acetyltransferase-3-1 (AAC(3)I) on the xylE-GmR cassette, this enzyme has been shown to have very narrow substrate specificity (Phillips & Shannon, 1984 ) and, therefore, is unlikely to affect resistance to other aminoglycosides. Of the aminoglycosides used in these experiments, streptomycin lacks the requisite substituent for modification by AAC(3)I, and no modification of kanamycin A or amikacin by this enzyme has been observed (Phillips & Shannon, 1984 ).

We chose to study the resistance of the strains H851, H854 and H855 to streptomycin in more detail. Since the increased resistance in all of these strains (fourfold over wild-type) was observed in LB medium as well as high-Mg2+ BM2-glucose, we conducted killing assays in the former medium. After treatment with 16 µg streptomycin ml-1, wild-type P. aeruginosa H103 was completely killed within 30 min (Table 4). All three mutant strains showed higher streptomycin resistance than the wild-type, but the extent of this resistance varied. The PhoP-null (H851) and OprH-null (H855) strains showed the highest levels of streptomycin resistance (100% and 70% survival, respectively, after 30 min), while the PhoQ-null strain H854 showed a level of resistance intermediate to these two strains and the wild-type (15% survival after 30 min).

To determine whether altered outer-membrane permeability could account for the observed increase in aminoglycoside resistance, we tested the ability of streptomycin to promote lysozyme lysis of the wild-type, and PhoP-, PhoQ- and OprH-null mutants. Lysozyme targets the peptidoglycan layer, but normally is unable to cross the outer-membrane barrier. Addition of aminoglycoside antibiotics disrupts the outer membrane sufficiently to allow the lysozyme molecules access to the cell. As can be seen from Fig. 3, all three mutant strains (H851, H854 and H855) were less susceptible to lysozyme lysis in the presence of 450 µg streptomycin ml-1 than wild-type H103. Of these mutants, the PhoQ-null strain H854 showed the highest level of resistance to streptomycin-assisted lysozyme lysis, and the OprH-null strain H855 the lowest. Treatment of cultures with 450 µg streptomycin ml-1 in the absence of lysozyme caused minimal cell lysis and no lysis was seen in cells treated with lysozyme alone (data not shown).



View larger version (14K):
[in this window]
[in a new window]
 
Fig. 3. Streptomycin-assisted lysozyme lysis of strains H103 (wild-type; {circ}), H851 (PhoP-null; {bullet}), H854 (PhoQ null; {square}) and H855 (OprH null; {blacksquare}). Results shown are the mean±SEM of three independent experiments using 450 µg streptomycin ml-1 and 50 µg lysozyme ml-1. Strains were grown in LB (Miller) broth and lysis was carried out in 30 mM sodium phosphate buffer, pH 7.

 
Aminoglycoside resistance of P. aeruginosa is unaffected by deletions in oprH
To confirm our hypothesis that the increased aminoglycoside resistance of strain H855 (oprH::xylE-GmR) was due to polar effects on the downstream phoP-phoQ genes rather than disruption of oprH, we constructed two strains, H895 and H896, with unmarked deletions in oprH (Table 1). In strain H895 ({Delta}oprH73), a 219 bp in-frame deletion within oprH was created that removed 73 amino acid residues, incorporating three of the eight proposed ß-strands (Rehm & Hancock, 1996 ), from the central region of OprH. The 28 bp deletion in strain H896 ({Delta}oprHC) was placed towards the 3' end of oprH, resulting in a frameshift mutation and creation of a premature stop codon 83 bp upstream of the regular TAA oprH stop codon.

Western blot analysis of mid-exponential phase cultures of both strains H895 and H896 grown under Mg2+-limiting (inducing) conditions indicated that no proteins were expressed that reacted with anti-OprH specific antiserum (data not shown). This suggests that if a mutant OprH protein was produced by these strains, it was rapidly degraded within the cell.

The unmarked deletions in strains H895 and H896 should permit normal transcription of phoP-phoQ from the inducible promoter upstream of oprH and thus allow us to determine the effects of an OprH-null mutation without concomitant effects on PhoP-PhoQ expression. The MIC values for several antibiotic compounds were determined for the two oprH deletion strains grown in BM2-glucose minimal medium supplemented with 2 mM (high) MgSO4. The values for polymyxin B, the aminoglycosides streptomycin, kanamycin and amikacin, and the {alpha}-helical cationic peptide CP28 are given in Table 5. No difference in the MIC value for any of these five antibiotics was observed between strains H895, H896 and the wild-type P. aeruginosa strain H103. These results strongly suggest that the increased resistance of strain H855 (oprH::xylE-GmR) to aminoglycosides was due to polar effects of the xylE-GmR cassette on the transcription of phoP-phoQ and not to the loss of OprH expression.


   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
In a previous paper we reported that resistance of P. aeruginosa PAO1 to polymyxin B involves the two-component regulatory system PhoP-PhoQ (Macfarlane et al., 1999 ). In this study we have demonstrated that resistance to the {alpha}-helical cationic peptide CP28 is likewise regulated by both Mg2+ ion concentrations and by PhoP-PhoQ. Analogous to our findings with polymyxin, resistance to CP28 under Mg2+-deficient conditions does not appear to depend on a functional PhoP protein. Under low-Mg2+ growth conditions, the PhoP-null mutant H851 exhibited a level of resistance equivalent to that of the wild-type strain H103 (Table 4). However, the PhoQ-null strain H854, in which PhoP is constitutively overexpressed, displayed constitutive resistance to CP28. Based on our prior observations of the transcriptional regulation of the oprH-phoP-phoQ operon (Macfarlane et al., 1999 ), we have proposed that PhoQ acts primarily to dephosphorylate and thus downregulate the activity of PhoP under Mg2+-sufficient conditions. In strain H854, constitutive high levels of PhoP expressed in the absence of a PhoQ protein may therefore result in the interaction of this response regulator with other resistance genes, most probably through cross-talk with another regulatory system. The observed resistance to CP28, like that to polymyxin B, is consistent with this proposal. As PhoP is presumably in an activated form in strain H854, phosphorylation of PhoP by a phospho-donor other than PhoQ remains a possibility.

OprH is not involved in polymyxin B resistance
In addition to the P. aeruginosa phoP and phoQ mutants, we also constructed an analogous chromosomal knockout of oprH, the first gene of the oprH-phoP-phoQ operon. Killing assays with polymyxin B revealed levels of resistance in this oprH::xylE-GmR strain (H855) similar to those of the wild-type strain H103 (Table 3), supporting our previous suggestion that OprH is not essential for polymyxin B resistance but may play an accessory role in stabilizing the outer membrane under Mg2+-starvation conditions (Macfarlane et al., 1999 ). The only plasmid-borne gene to affect the polymyxin B resistance of strain H855 was phoP (pEMR3) (Table 3), the presence of which resulted in significant resistance in high-Mg2+ medium. This pattern of polymyxin B resistance closely resembles that of the PhoP-null strain H851 (Macfarlane et al., 1999 ), and thus supported our initial assumption that the polar nature of the xylE-GmR insertion in oprH rendered strain H855 PhoP- and PhoQ-null. This assumption was based on our inability to detect a second promoter for phoP-phoQ by Northern blotting (Macfarlane et al., 1999 ) or primer-extension analysis (data not shown) of RNA transcripts in the wild-type strain H103, or by dot-blot analysis of RNA from strain H855 using phoP and phoQ primers (data not shown). Quantitation of the RNA dot blots indicated that if any phoP-phoQ transcript was present in strain H855, it occurred at a level at least twofold lower than that of any phoP-containing transcripts in strain H103 under high-Mg2+ (uninduced) conditions (data not shown).

The region between the oprH and phoP genes in the P. aeruginosa genome is 79 bp in length. A possible ribosome-binding site occurs 5 bp upstream of the phoP ATG start codon, but no promoter consensus sequences were identified. There are also no repeat sequences in this region resembling those found upstream of oprH. However, the Mg2+ regulation of catechol 2,3-dioxygenase expressed from the oprH::xylE transcriptional fusion in strain H855 (Table 2), together with our observation that transcription of oprH is wholly dependent on the presence of PhoP (Macfarlane et al., 1999 ), provided indirect evidence for a very low level of phoP-phoQ transcription in this strain from an as yet unidentified second promoter. The presence of two promoters – one constitutive and one inducible – allowing a basal level of transcription is a common feature of two-component regulatory systems and has been reported for phoP-phoQ in E. coli (Kato et al., 1999 ), and for both phoP-phoQ and pmrA-pmrB in S. typhimurium (Gunn & Miller, 1996 ; Soncini et al., 1995 ). For the latter system, the second promoter lies within the 3' region of pmrC, the first gene of the pmrCAB operon (Gunn & Miller, 1996 ). The oprH-phoP intergenic region was also inspected for possible secondary structures. Although no classic rho-independent terminator sequences were identified, an inverted repeat sequence predicted to be capable of forming a hairpin structure (predicted {Delta}G=-14·3 kcal mol-1) was identified 36 bp downstream of the oprH stop codon. Northern blot analysis of strains H103 (Macfarlane et al., 1999 ) and H854 (Fig. 1) revealed high levels of oprH transcripts and diminishing amounts of oprH-phoP and oprH-phoP-phoQ transcripts. Formation of a hairpin structure that functions as a transcriptional attenuator downstream of oprH and provides additional regulation of oprH-phoP-phoQ transcription would be consistent with these results. It is worth noting also that our results do not eliminate the possibility that P. aeruginosa oprH-phoP-phoQ may be subject to some form of global regulation, for example by another two-component regulatory system, that acts in conjunction with PhoP-mediated regulation.

PhoP-PhoQ selectively regulates cationic peptide resistance
The polymyxin B and CP28 resistance of the OprH-null strain H855 supported our conclusions that PhoP is not essential for resistance to these two antibiotics under Mg2+-starvation conditions. Although transcription of phoP-phoQ presumably occurs at a very low level in strain H855, phoP transcription cannot be induced through the promoter upstream of oprH and levels of PhoP protein, therefore, would be minimal in this strain.

Interestingly, a phoP::Gm mutant of P. aeruginosa PAK has recently been reported that is supersusceptible to polymyxin B, but resistant to the cationic peptide C18G, under Mg2+-starvation conditions (Ernst et al., 1999 ). The possibility that the Gm cassette inserted into the PAK phoP gene was non-polar and allowed transcription of phoQ could account for the discrepancy in polymyxin B resistance between the phoP mutants in the two strains (Macfarlane et al., 1999 ).

The results of MIC determinations for cationic peptides other than CP28 (Table 5) indicated that P. aeruginosa resistance to a second {alpha}-helical peptide CP29 and the protegrin-like peptide IB-367 (currently undergoing clinical trials against P. aeruginosa lung infections in cystic fibrosis patients; see http://www.intrabiotics.com) probably involves a pathway similar to that for polymyxin B and CP28 resistance. However, resistance to the indolicidin analogue CP11CN, although still affected by mutations in PhoP-PhoQ, appears to be subject to different regulatory mechanisms. Increased resistance to this peptide was seen only in the two mutants, H851 and H855, that could not induce expression of PhoP. Resistance to polyphemusin, on the other hand, appeared to be independent of PhoP-PhoQ. While this was the only peptide we tested that was unaffected by the antagonistic effects of Mg2+ ions, resistance still appeared to be Mg2+ regulated (data not shown), indicating that other regulatory systems in P. aeruginosa must also respond to extracellular Mg2+ ion concentrations.

Although the mechanism of action of antimicrobial cationic peptides remains uncertain, the {alpha}-helical peptides CP28 and CP29, like polymyxin B, are believed to initially interact with and disrupt the outer membrane of Gram-negative bacteria (Hancock & Chapple, 1999 ). In S. typhimurium, increased resistance to polymyxin B and certain antimicrobial peptides is observed under Mg2+-starvation conditions and is partly due to structural alterations of the outer membrane resulting from activation of genes in the PhoP-PhoQ regulon. Similarly, a phoP mutant of P. aeruginosa PAK was shown to lack certain modifications to core lipid A that were seen in the wild-type strain under Mg2+-starvation conditions (Ernst et al., 1999 ). Resistance of P. aeruginosa PAO1 to polymyxin B and CP28 likewise may be due to structural alterations to the outer membrane that are regulated in part by PhoP-PhoQ.

The small differences in MIC values measured for the oprH, phoP and phoQ mutants compared to wild-type indicate that PhoP-PhoQ is not a major factor in P. aeruginosa cationic peptide resistance. Our results are consistent with an indirect role for this regulatory system in the Mg2+ regulation of cationic peptide resistance. This conclusion contrasts peptide resistance in P. aeruginosa to that in S. typhimurium, where the PhoP-PhoQ two-component regulatory system has been proposed to play a key role. However, although killing assays comparing S. typhimurium phoP mutants with wild-type have showed substantial differences in peptide susceptibility (Fields et al., 1989 ), examination of the susceptibility of S. typhimurium strain MS7953 (phoP::Tn10; Fields et al., 1989 ) to the peptides CP28, CEME and CP11CN revealed decreases in MIC values of only two- to eightfold relative to the values for wild-type (Falla & Hancock, 1997 ; Piers et al., 1994 ).

Aminoglycoside-resistance regulation
The differing effects of the P. aeruginosa phoP and phoQ mutations on cationic peptide resistance are indicative of the complex regulation of antibiotic resistance in this bacterium. The observed increase in resistance to aminoglycosides for all three mutants (PhoP-, PhoQ- and OprH-null) is a further indication of the intricacies of the PhoP-PhoQ regulatory system. The only feature common to strains H851, H854 and H855 is the inability to induce expression of phoQ. Strains H895 and H896, which carry deletions in oprH but which should allow fully inducible expression of phoP-phoQ, showed identical aminoglycoside resistance to the wild-type strain H103 (Table 5). The increased resistance observed in strain H855 (oprH::xylE-GmR), therefore, can be directly attributed to the polar effect of the xylE-GmR cassette on the downstream phoP-phoQ genes rather than to loss of OprH expression. Similarly, previous results that implicated OprH in P. aeruginosa resistance to aminoglycosides (Hancock et al., 1981 ; Young et al., 1992 ) can now be explained in terms of polar effects of the tet cassette inserted into oprH in strain H703 (Young et al., 1992 ) on phoP-phoQ expression. Our results imply that PhoQ is responsible, either directly or through interaction with other regulatory systems, for the downregulation of resistance to the aminoglycosides streptomycin, kanamycin and amikacin. In this case, such a downregulation effect must be overridden by another system under Mg2+-starvation conditions, as these conditions are known to induce both PhoP-PhoQ expression (Macfarlane et al., 1999 ) and resistance to streptomycin and gentamicin (Hancock et al., 1981 ). Interestingly, in killing assays with streptomycin, the overexpression of PhoP in the PhoQ-null mutant H854 was seen to partially counteract the positive effect of a PhoQ-null phenotype (Table 4). Hence it is feasible that high levels of PhoP expression in the absence of PhoQ affect the other system(s) involved in streptomycin resistance, most probably through cross-talk.

Decreased permeability of the outer membrane to the aminoglycoside streptomycin was observed for all three mutants (Fig. 3). However, the high concentrations of streptomycin (>100-fold higher than the MIC) that were required for this experiment probably do not reflect the normal killing mechanism, and are consequently more indicative of a general increase in outer-membrane stability in the mutants. A greater resistance to perturbation of the outer membrane was observed in strain H854, which overexpresses OprH, than in strains H851 and H855 (both unable to express OprH), lending support to our suggestion of an outer-membrane-stabilization role for OprH.

We noted earlier that an exceptionally large number of two-component regulatory systems have been revealed in P. aeruginosa by the Pseudomonas genome sequencing project. Our results support the concept of P. aeruginosa antibiotic resistance being the result of a complex interplay of several of these systems. Further studies aimed at unravelling such interactions, as well as defining genes that fall into the PhoP-PhoQ regulon, should greatly aid our understanding of antibiotic resistance in this important pathogen.


   ACKNOWLEDGEMENTS
 
This work was supported by grants from the Canadian Cystic Fibrosis Foundation and the MRC. R.E.W.H. has an MRC Distinguished Scientist Award.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Amsterdam, D. (1991). Antibiotics in Laboratory Medicine. Baltimore, MD: Williams & Wilkins.

Ausubel, F. M., Brent, R., Kingston, R. E., Moore, D. D., Seidman, J. G., Smith, J. A. & Struhl, K. (editors) (1987). Current Protocols in Molecular Biology. New York: Greene Publishing and Wiley Interscience.

Bajaj, V., Lucas, R. L., Hwang, C. & Lee, C. A. (1996). Co-ordinate regulation of Salmonella typhimurium invasion genes by environmental and regulatory factors is mediated by control of hilA expression. Mol Microbiol 22, 703-714.[Medline]

Bell, A., Bains, M. & Hancock, R. E. (1991). Pseudomonas aeruginosa outer membrane protein OprH: expression from the cloned gene and function in EDTA and gentamicin resistance. J Bacteriol 173, 6657-6664.[Medline]

Brown, M. R. & Melling, J. (1969). Role of divalent cations in the action of polymyxin B and EDTA on Pseudomonas aeruginosa. J Gen Microbiol 59, 263-274.[Medline]

Dennis, J. J. & Sokol, P. A. (1995). Electrotransformation of Pseudomonas. Methods Mol Biol 47, 125-133.[Medline]

Ernst, R. K., Yi, E. C., Guo, L., Lim, K. B., Burns, J. L., Hackett, M. & Miller, S. I. (1999). Specific lipopolysaccharide found in cystic fibrosis airway Pseudomonas aeruginosa. Science 286, 1561-1565.[Abstract/Free Full Text]

Falla, T. J. & Hancock, R. E. (1997). Improved activity of a synthetic indolicidin analog. Antimicrob Agents Chemother 41, 771-775.[Abstract]

Fields, P. I., Groisman, E. A. & Heffron, F. (1989). A Salmonella locus that controls resistance to microbicidal proteins from phagocytic cells. Science 243, 1059-1062.[Medline]

Friedrich, C., Scott, M. G., Karunaratne, N., Yan, H. & Hancock, R. E. (1999). Salt-resistant alpha-helical cationic antimicrobial peptides. Antimicrob Agents Chemother 43, 1542-1548.[Abstract/Free Full Text]

Garcia Vescovi, E., Soncini, F. C. & Groisman, E. A. (1996). Mg2+ as an extracellular signal: environmental regulation of Salmonella virulence. Cell 84, 165-174.[Medline]

Gilleland, H. E.Jr, Stinnett, J. D. & Eagon, R. G. (1974). Ultrastructural and chemical alteration of the cell envelope of Pseudomonas aeruginosa, associated with resistance to ethylenediaminetetraacetate resulting from growth in a Mg2+-deficient medium. J Bacteriol 117, 302-311.[Medline]

Gunn, J. S. & Miller, S. I. (1996). PhoP-PhoQ activates transcription of pmrAB, encoding a two-component regulatory system involved in Salmonella typhimurium antimicrobial peptide resistance. J Bacteriol 178, 6857-6864.[Abstract]

Gunn, J. S., Lim, K. B., Krueger, J., Kim, K., Guo, L., Hackett, M. & Miller, S. I. (1998). PmrA-PmrB-regulated genes necessary for 4-aminoarabinose lipid A modification and polymyxin resistance. Mol Microbiol 27, 1171-1182.[Medline]

Guo, L., Lim, K. B., Poduje, C. M., Daniel, M., Gunn, J. S., Hackett, M. & Miller, S. I. (1998). Lipid A acylation and bacterial resistance against vertebrate antimicrobial peptides. Cell 95, 189-198.[Medline]

Hancock, R. E. & Chapple, D. S. (1999). Peptide antibiotics. Antimicrob Agents Chemother 43, 1317-1323.[Free Full Text]

Hancock, R. E., Raffle, V. J. & Nicas, T. I. (1981). Involvement of the outer membrane in gentamicin and streptomycin uptake and killing in Pseudomonas aeruginosa. Antimicrob Agents Chemother 19, 777-785.[Medline]

Helander, I. M., Kilpelainen, I. & Vaara, M. (1994). Increased substitution of phosphate groups in lipopolysaccharides and lipid A of the polymyxin-resistant pmrA mutants of Salmonella typhimurium: a 31P-NMR study. Mol Microbiol 11, 481-487.[Medline]

Iwanaga, S., Muta, T., Shigenaga, T., Seki, N., Kawano, K., Katsu, T. & Kawabata, S. (1994). Structure-function relationships of tachyplesins and their analogues. Ciba Found Symp 186, 160-174.[Medline]

Kato, A., Tanabe, H. & Utsumi, R. (1999). Molecular characterization of the PhoP-PhoQ two-component system in Escherichia coli K-12: identification of extracellular Mg2+-responsive promoters. J Bacteriol 181, 5516-5520.[Abstract/Free Full Text]

Macfarlane, E. L., Kwasnicka, A., Ochs, M. M. & Hancock, R. E. (1999). PhoP-PhoQ homologues in Pseudomonas aeruginosa regulate expression of the outer-membrane protein OprH and polymyxin B resistance. Mol Microbiol 34, 305-316.[Medline]

Miller, S. I., Kukral, A. M. & Mekalanos, J. J. (1989). A two-component regulatory system (PhoP PhoQ) controls Salmonella typhimurium virulence. Proc Natl Acad Sci U S A 86, 5054-5058.[Abstract]

Miller, S. I., Pulkkinen, W. S., Selsted, M. E. & Mekalanos, J. J. (1990). Characterization of defensin resistance phenotypes associated with mutations in the phoP virulence regulon of Salmonella typhimurium. Infect Immun 58, 3706-3710.[Medline]

Nicas, T. I. & Hancock, R. E. (1980). Outer membrane protein H1 of Pseudomonas aeruginosa: involvement in adaptive and mutational resistance to ethylenediaminetetraacetate, polymyxin B, and gentamicin. J Bacteriol 143, 872-878.[Medline]

Phillips, I. & Shannon, K. (1984). Aminoglycoside resistance. Br Med Bull 40, 28-35.[Medline]

Piers, K. L., Brown, M. H. & Hancock, R. E. (1994). Improvement of outer membrane-permeabilizing and lipopolysaccharide-binding activities of an antimicrobial cationic peptide by C-terminal modification. Antimicrob Agents Chemother 38, 2311-2316.[Abstract]

Rehm, B. H. & Hancock, R. E. (1996). Membrane topology of the outer membrane protein OprH from Pseudomonas aeruginosa: PCR-mediated site-directed insertion and deletion mutagenesis. J Bacteriol 178, 3346-3349.[Abstract]

Rothmel, R. K., Chakrabarty, A. M., Berry, A. & Darzins, A. (1991). Genetic systems in Pseudomonas. Methods Enzymol 204, 485-514.[Medline]

Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Schweizer, H. P. (1991). Escherichia-Pseudomonas shuttle vectors derived from pUC18/19. Gene 97, 109-121.[Medline]

Schweizer, H. P. & Hoang, T. T. (1995). An improved system for gene replacement and xylE fusion analysis in Pseudomonas aeruginosa. Gene 158, 15-22.[Medline]

Simon, R., Preifer, V. & Puhler, A. (1983). A broad host range mobilization system for in vivo genetic engineering: transposon mutagenesis in Gram negative bacteria. Bio/Technology 1, 784-790.

Soncini, F. C. & Groisman, E. A. (1996). Two-component regulatory systems can interact to process multiple environmental signals. J Bacteriol 178, 6796-6801.[Abstract]

Soncini, F. C., Garcia Vescovi, E. & Groisman, E. A. (1995). Transcriptional autoregulation of the Salmonella typhimurium phoPQ operon. J Bacteriol 177, 4364-4371.[Abstract]

West, S. E., Schweizer, H. P., Dall, C., Sample, A. K. & Runyen-Janecky, L. J. (1994). Construction of improved Escherichia-Pseudomonas shuttle vectors derived from pUC18/19 and sequence of the region required for their replication in Pseudomonas aeruginosa. Gene 148, 81-86.[Medline]

Wu, M. & Hancock, R. E. (1999). Interaction of the cyclic antimicrobial cationic peptide bactenecin with the outer and cytoplasmic membrane. J Biol Chem 274, 29-35.[Abstract/Free Full Text]

Young, M. L., Bains, M., Bell, A. & Hancock, R. E. (1992). Role of Pseudomonas aeruginosa outer membrane protein OprH in polymyxin and gentamicin resistance: isolation of an OprH-deficient mutant by gene replacement techniques. Antimicrob Agents Chemother 36, 2566-2568.[Abstract]

Zimelis, V. M. & Jackson, G. G. (1973). Activity of aminoglycoside antibiotics against Pseudomonas aeruginosa: specificity and site of calcium and magnesium antagonism. J Infect Dis 127, 663-669.[Medline]

Received 6 April 2000; revised 19 June 2000; accepted 29 June 2000.