|Home | About | Journals | Submit | Contact Us | Français|
Pan-aminoglycoside-resistant Pseudomonas aeruginosa mutants expressing the mexXY components of the aminoglycoside-accommodating MexXY-OprM multidrug efflux system but lacking mutations in the mexZ gene encoding a repressor of this efflux system and in the mexXY promoter have been reported (S. Fraud and K. Poole, Antimicrob. Agents Chemother. 55:1068–1074, 2011). Genome sequencing of one of these mutants, K2966, revealed the presence of a mutation within the predicted promoter region of the rplU-rpmA operon encoding ribosomal proteins L21 and L27, consistent with an observed 2-fold decrease in expression of this operon in the mutant relative to wild-type P. aeruginosa PAO1. Moreover, correction of the mutation restored rplU-rpmA expression and, significantly, reversed the elevated mexXY expression and pan-aminoglycoside resistance of the mutant. Reduced rplU-rpmA expression was also observed in a second mexXY-expressing pan-aminoglycoside-resistant mutant, K2968, which, however, lacked a mutation in the rplU-rpmA promoter region. Restoration of rplU-rpmA expression in the K2968 mutant following chromosomal integration of the rplU-rpmA operon derived from wild-type P. aeruginosa failed, however, to reverse the elevated mexXY expression and pan-aminoglycoside resistance of this mutant, although it did so for K2966, suggesting that the mutation impacting rplU-rpmA expression in K2968 also impacts other mexXY-related genes. Increased mexXY expression owing to reduced rplU-rpmA expression in K2966 and K2968 was dependent on PA5471, whose expression was also elevated in these mutants. Thus, mutational disruption of ribosome function, by limiting expression of ribosomal constituents, promotes recruitment of mexXY and does so via PA5471, reminiscent of mexXY induction by ribosome-disrupting antimicrobial agents. Interestingly, reduced rplU-rpmA expression was also observed in a mexXY-expressing pan-aminoglycoside-resistant clinical isolate, suggesting that ribosome-perturbing mutations have clinical relevance in the recruitment of the MexXY-OprM aminoglycoside resistance determinant.
Pseudomonas aeruginosa is a Gram-negative, opportunistic human pathogen that is often associated with life-threatening nosocomial infections in immunocompromised individuals, notably pulmonary infections in patients with cystic fibrosis (CF) (27). Unfortunately, antipseudomonal chemotherapy is complicated by the organism's intrinsic antimicrobial resistance and its ready acquisition of additional resistance mechanisms (38, 45). Major contributors to this intrinsic and acquired resistance are multidrug efflux systems of the RND (resistance nodulation division) family of which four, MexAB-OprM, MexCD-OprJ, MexEF-OprN, and MexXY-OprM, are reported to be significant determinants of multidrug resistance in laboratory and clinical isolates (38). MexXY-OprM accommodates a broad spectrum of antimicrobial substrates that include fluoroquinolones, tetracycline, tigecycline, macrolides (e.g., erythromycin), chloramphenicol, and selected β-lactams (1, 10, 30, 31, 33), though it is particularly noteworthy for its export of and contribution to resistance to aminoglycosides (1, 51). MexXY-OprM is somewhat unique in P. aeruginosa in that the mexXY operon is induced upon exposure to many of the antibiotics that this efflux system exports (30). Still, only ribosome-targeting agents promote mexXY expression (8, 10, 17), and this is compromised by so-called ribosome protection mechanisms (17), an indication that it is not the antimicrobial agents per se but rather their corresponding detrimental effects on the ribosome that drives expression of this efflux operon. This upregulation of mexXY by antimicrobials is dependent upon a gene, PA5471, encoding a conserved hypothetical protein whose expression is also promoted by ribosome-disrupting antimicrobials (35).
Expression of mexXY is negatively regulated by MexZ, a TetR-family repressor protein that is encoded by the divergently expressed mexZ gene (1, 51) and that binds to the promoter region of this efflux operon (32, 52). Consistent with ribosome disruption and not antimicrobials being the trigger for mexXY expression, antimicrobials do not interact with MexZ to affect antimicrobial-inducible mexXY expression (32), although the antimicrobial-inducible PA5471 gene product has been shown to interact with this repressor and interfere with its DNA binding activity (52), providing a possible explanation for the antimicrobial induction of this efflux operon. Still, PA5471 alone is insufficient for maximal drug-inducible mexXY expression (34), an indication that an additional antimicrobial-dependent factor(s) is involved. Not surprisingly, many MexXY-expressing aminoglycoside-resistant isolates of P. aeruginosa carry mutations in mexZ (6, 26, 49). Consistent, too, with MexXY-OprM being the most common determinant of aminoglycoside resistance in CF isolates of P. aeruginosa, mexZ is the most commonly mutated gene in such isolates (38). Nevertheless, there are numerous reports of MexXY-expressing aminoglycoside-resistant isolates lacking mutations in mexZ (or the mexXY promoter region) (6, 11, 16, 17, 26, 36, 37, 44, 49, 51), an indication that additional genes/mutations are linked to mexXY expression and MexXY-OprM-mediated aminoglycoside resistance in this organism. Indeed, a recent study identified a gene, parR, encoding the response regulator of a two-component system (TCS), ParRS, whose mutation was responsible for mexXY upregulation and aminoglycoside resistance (36). However, the signal(s) to which this TCS responds and the details of its control of mexXY expression are unknown. In an ongoing attempt to elucidate the details of the recruitment of MexXY-OprM under conditions of ribosome disruption, including the identification of the inducing factor(s), the genomes of previously reported MexXY-expressing pan-aminoglycoside-resistant P. aeruginosa mutants lacking mutations in mexZ or the mexXY promoter region were sequenced. We report here a novel link between reduced expression of a ribosomal protein operon, rplU-rpmA, and mexXY expression, further solidifying the link between ribosome dysfunction and MexXY-OprM recruitment.
The bacterial strains used in this study are described in Table 1. Bacterial cells were cultured in Luria broth (L-broth) and on Luria agar (L-agar), with antibiotics as necessary, at 37°C. To generate P. aeruginosa strain K3216 in which the rplU-rpmA promoter mutation of strain K2966 was restored to the wild type, a 1,465-bp fragment encompassing the promoter region, including ca. 700 bp upstream and downstream of the mutation site, was first amplified from the chromosome of wild-type P. aeruginosa PAO1 strain K767 by PCR. Amplification was achieved using primers rplU-1465-F (F stands for forward) and rplU-1465-R (R stands for reverse) (Table 2) in a 50-μl reaction mixture containing 10 ng of chromosomal DNA, 1 U of Phusion high-fidelity DNA polymerase (New England BioLabs Ltd., Pickering, Ontario, Canada), 1× Phusion GC buffer, primers at a final concentration (each) of 0.6 μM, and deoxynucleoside triphosphates (dNTPs) at a final concentration (each) of 0.2 mM. The mixture was heated for 3 min at 98°C, followed by 35 cycles with 1 cycle consisting of 0.5 min at 98°C, 0.5 min at 70°C, and 0.75 min at 72°C, before finishing with 10 min at 72°C. The PCR product was subsequently cloned into plasmid pEX18Tc as an XbaI-HindIII-restricted fragment to yield plasmid pCL1. Plasmid pCL1 harboring the wild-type rplU-rpmA promoter sequence was then introduced into Escherichia coli S17-1 and mobilized into P. aeruginosa strain K2966 as described previously (7). Briefly, 700 μl of pCL1-carrying E. coli S17-1 (log phase, cultured at 37°C) was mixed with 300 μl of P. aeruginosa K2966 (stationary phase, cultured at 42°C) in a microcentrifuge tube, followed by centrifugation. The resultant cell pellet was then resuspended in 100 μl of L-broth and subsequently spotted onto the center of an L-agar plate. Following incubation at 37°C for 16 h, bacteria were recovered from the L-agar plate in 1 ml of L-broth, and P. aeruginosa K2966 transconjugants harboring chromosomal inserts of pCL1 were selected on L-agar plates containing tetracycline (50 μg/ml) and chloramphenicol (5 μg/ml) (to counterselect E. coli S17-1). These transconjugants were subsequently streaked onto L-agar plates containing sucrose (10% [wt/vol]), with sucrose-resistant colonies screened for restoration of the wild-type rplU-rpmA promoter sequence using high-resolution melt (HRM) analysis as described below. One such “revertant,” K3216, was saved for further study.
The plasmids used in this study are listed in Table 1. Plasmids pEX18Tc and mini-CTX1 and their derivatives were maintained in Escherichia coli with 10 μg/ml of tetracycline. Plasmid pRK415 and its derivatives were maintained with 10 μg/ml (E. coli) or 50 μg/ml (P. aeruginosa) tetracycline. Plasmid pCL2 carrying the rplU-rpmA operon with the wild-type promoter region was constructed by amplifying the rplU-rpmA operon and upstream region from P. aeruginosa strain K767 as a 1,092-bp fragment by PCR, before cloning it into mini-CTX1. Amplification was carried out as described above using primers rplU-rpmA-F and rplU-rpmA-R (Table 2) with an annealing temperature of 69.1°C and an extension time of 0.55 min. The PCR product was then digested with HindIII and XbaI and cloned in HindIII-XbaI-restricted mini-CTX1. Plasmid pCL2 was subsequently introduced into E. coli S17-1 and mobilized in P. aeruginosa as described above for pCL1, with transconjugants harboring chromosomal inserts of the vector (at the phage D113 attB site) selected also as described above. Elimination of the mini-CTX1 plasmid backbone was achieved using the pFLP2-encoded Flp recombinase (13) to leave only the rplU-rpmA operon and wild-type promoter region in the chromosome. Thus, plasmid pFLP2 was introduced into P. aeruginosa via electroporation, and pFLP2-containing, carbenicillin (200 μg/ml)-resistant colonies were streaked onto L-agar containing 10% (wt/vol) sucrose to select for the loss of plasmid pFLP2 (following excision of the mini-CTX1 backbone from the chromosome). Sucrose-resistant colonies were then patched onto L-agar plates containing tetracycline (50 μg/ml) or carbenicillin (200 μg/ml) to confirm the loss of the mini-CTX1 backbone (tetracycline sensitive) and the pFLP2 plasmid (carbenicillin sensitive). Plasmid pCL3 carrying the rplU-rpmA operon with the wild-type promoter region was constructed by cloning the aforementioned 1,092-bp fragment into HindIII-XbaI-restricted pRK415.
Standard protocols were generally used for restriction endonuclease digestion, ligation, transformation, plasmid isolation, and agarose gel electrophoresis as described by Sambrook and Russell (39). Plasmid DNAs were also prepared from E. coli using a GeneJET plasmid miniprep kit (Fermentas Canada Inc., Burlington, Ontario, Canada) according to a protocol provided by the manufacturer. Chromosomal DNA from P. aeruginosa was extracted using a DNeasy blood and tissue kit (Qiagen Inc., Mississauga, Ontario, Canada) according to the manufacturer's protocol. DNA fragments used for cloning were extracted from agarose gels using a Wizard SV gel and PCR cleanup system (Fisher Scientific, Ltd., Nepean, Ontario, Canada) and, once cloned, sequenced to verify that no unintended mutations were introduced during PCR. Competent E. coli cells were prepared as described previously (15). Oligonucleotide synthesis was carried out by Integrated DNA Technologies (Coralville, IA), and nucleotide sequencing was carried out by ACGT Corp. (Toronto, Ontario, Canada) using universal and custom primers. Genome sequencing and polymorphism detection, annotation, and validation were conducted as described previously (4).
To detect derivatives of K2966 (e.g., K3216) in which the rplU-rpmA promoter mutation was restored to the wild type, single colonies were recovered from sucrose plates and resuspended in 30 μl of sterile distilled H2O, which was then heated for 5 min at 95°C to release the chromosomal DNA, followed by centrifugation to pellet the unwanted cell debris. A 48-bp segment of the rplU-rpmA upstream region that included the promoter mutation site was amplified from the chromosomal DNA with primers HRM-K3216-F and HRM-K3216-R (Table 2) in a 20-μl PCR mixture containing 1 μl of colony lysate, a 0.6 μM concentration of each primer, and 10 μl of the SsoFast EvaGreen supermix (Bio-Rad). The mixture was heated in a CFX96 real-time PCR detection system (Bio-Rad) for 3 min at 95°C, followed by 30 cycles, with 1 cycle consisting of 0.5 min at 95°C, 0.5 min at 58°C, and 5 s at 72°C, finishing with 10 min at 72°C. The melting curve profile of the resultant amplicons was subsequently determined over a range of 65°C to 95°C with a 0.5°C-stepwise increment using the same instrument. The peak melting temperatures (Tm) of the amplicons derived from the chromosomal DNA from the positive control (wild-type PAO1 strain K767) and negative control (K2966 with a point mutation) were 78.5°C and 79.5°C, respectively. The rplU-rpmA upstream region of candidates with a Tm of 78.5°C was PCR amplified and sequenced to confirm restoration of the wild-type promoter sequence.
Total bacterial RNA was isolated from 1 to 2 ml of log-phase (optical density at 600 nm [OD600] of 0.6 to 0.8) P. aeruginosa L-broth cultures using a High Pure RNA isolation kit (Roche Canada, Mississauga, Ontario, Canada) as described by the manufacturer. Remaining traces of DNA in the isolated RNA samples were eliminated by a 30-min treatment with Turbo DNA-free DNase (Applied Biosystems Canada, Streetsville, Ontario, Canada) again as described by the manufacturer. RNA yields were typically 3 to 9 μg RNA per extraction. The purity of the samples was verified spectrophotometrically (an OD260/OD280 ratio of 1.8 to 2.0 was required), and the absence of contaminating genomic DNA was confirmed by failure to amplify the housekeeping gene rpsL using PCR (44). RNA samples were immediately reverse transcribed into cDNA (see below) or stored at −20°C for later conversion.
Reverse transcription of RNA samples into cDNA was carried out using the iScript cDNA synthesis kit (Bio-Rad) in a 10-μl mixture containing 500 ng of RNA, 2 μl of the 5× iScript reaction mix, and 0.5 μl of the iScript reverse transcriptase. The reaction mixture was then heated for 5 min at 25°C, followed by 30 min at 42°C, finishing with 5 min at 85°C. The cDNA products were then quantitated immediately using real-time PCR or stored at −20°C for later quantitation.
The primers used in quantitative real-time PCR (Table 2; identified by qPCR prefix) were designed to amplify specific gene fragments with lengths of 142 bp (mexX), 89 bp (rplU), 77 bp (rpmA), 89 bp (PA5471), and 91 bp (rpoD). The linear dynamic range and amplification efficiency of each primer set were determined, using serially diluted cDNA samples derived from 5 to 10 different RNA samples as the templates, and reported below in accordance with the MIQE (minimum information for publication of quantitative real-time PCR experiments) guidelines (5). The amplification efficiencies of the quantitative real-time PCR primer sets for mexX, rplU, rpmA, PA5471, and rpoD were determined to be 99.7% (correlation coefficient, r2 = 0.993), 98.6% (r2 = 0.998), 101.3% (r2 = 0.997), 99.0% (r2 = 0.998), and 104.4% (r2 = 0.992), respectively. All quantitative real-time PCR primer sets used in the present study had a minimum 4 log10 dynamic range.
The expression of mexX, rplU, rpmA, and PA5471 (target genes) was assessed by quantitative real-time PCR. The rpoD gene (reference gene) was chosen as an internal reference to normalize the measured amounts of target gene mRNA, as recommended elsewhere (40). The abundance of target and reference gene mRNA was measured (in triplicate) in a 20-μl reaction mixture containing 5 μl of a 200-fold-diluted cDNA template (an amount corresponding to 1.25 ng of total RNA), a 0.3 μM (mexX) or 0.6 μM (rplU, rpmA, PA5471, rpoD) concentration of each primer (Table 2), and 10 μl of the SsoFast EvaGreen supermix (Bio-Rad). The reaction mixtures were heated for 3 min at 95°C, followed by 40 cycles of 10 s at 95°C and 15 s at 60°C, finishing with 10 min at 95°C in clear 96-well plates sealed with Microseal B Adhesive Seal (Bio-Rad) using a CFX96 real-time PCR detection system (Bio-Rad). To verify the specificity of each amplification reaction, the melting curve profile of the resultant amplicons was determined over a range of 65°C to 95°C with a 0.5°C stepwise increment upon completion of the amplification cycle. For each gene studied, at least one control reaction with no cDNA template was included in each experiment to check for contamination of reagent(s) and to identify unintended amplification products (e.g., primer dimers). The levels of expression of the target genes in each strain studied, normalized against that of the reference gene, were calculated using the standard analysis feature of the CFX-manager software version 1.6 (Bio-Rad) and were reported herein as fold change relative to that in the P. aeruginosa PAO1 wild-type strain K767.
Overnight cultures of P. aeruginosa were diluted into fresh L-broth (200 μl) to a final OD600 of 0.1 in flat-bottom 96-well plates, and growth was assessed by measuring the OD600 every 30 min for 18 h in a Varioskan microplate reader (Thermo Electron Corporation). The growth assay was carried out at 37°C with constant agitation.
The susceptibility of P. aeruginosa to antimicrobial agents was assessed using the 2-fold serial microtiter broth dilution method described previously (19), with an inoculum of ~5 × 105 cells per ml. MICs were recorded as the lowest concentration of antibiotic inhibiting visible growth after 18 h of incubation at 37°C.
Previously, we described a collection of amikacin-selected, pan-aminoglycoside-resistant mutants derived from peroxide (H2O2)-exposed wild-type P. aeruginosa PAO1 strain K767 (12). Two of these strains, K2966 and K2968, were pan-aminoglycoside-resistant mutants owing to constitutive mexXY expression, although the mutation(s) responsible was unknown (no mutations were detected in the mexZ repressor gene or the mexXY promoter) (12). Whole-genome sequencing of these mutants revealed a number of alterations relative to strain K767 (data not shown), one of which was an adenine-to-guanine substitution in the intergenic region preceding the 50S ribosomal protein L21-encoding rplU gene (PA4568) in strain K2966. rplU is the first gene of an apparently 2-gene operon that also includes the ribosomal protein L27-encoding gene rpmA. Intriguingly, the mutation upstream of this operon occurred within the putative −10 Pribnow box of the only predicted promoter for rplU-rpmA (SoftBerry BPROM promoter prediction software; SoftBerry, Inc., Mount Kisco, NY), 135 bp upstream of the rplU start codon (TTGCCTCGGCCTGGACCTTTCCGTACAATCGCGGGCc [the predicted −35 region is double underlined, the −10 region is underlined, the predicted +1 site indicated in lowercase type, and the site of A-G mutation is shown in bold italic type]), offering the possibility that strain K2966 was compromised for rplU-rpmA expression. Consistent with the mutation in K2966 occurring in the rplU-rpmA promoter and with these genes residing within an operon, expression of both rplU and rpmA was reduced approximately 2-fold relative to the wild-type parent strain K767 (Fig. 1A and andBB).
To assess whether reduction in expression of one or both of the rplU and rpmA genes was responsible for the increased mexXY expression and pan-aminoglycoside resistance of strain K2966, attempts were made to delete these genes in wild-type strain K767 and to assess the impact on mexXY expression and aminoglycoside resistance. Despite repeated attempts, elimination of rplU in strain K767 was not successful (rpmA deletion was thus not attempted), suggesting that this gene was likely essential in P. aeruginosa. Attempts at reducing rplU-rpmA expression in K767 via introduction of the rplU-rpmA promoter mutation of strain K2966 into this strain were similarly unsuccessful. These results suggested that reduction or loss of rplU-rpmA strongly negatively impacted P. aeruginosa, and in agreement with this, growth of K2966 was severely compromised relative to K767 (Fig. 2A). Thus, to ascertain the significance of reduced rplU-rpmA expression in promoting increased mexXY expression and pan-aminoglycoside resistance in strain K2966, the rplU-rpmA promoter sequence of this strain was restored to the wild type. As seen in Fig. 1A, correcting the promoter mutation in the K3216 strain, which was derived from strain K2966, restored expression of rplU and rpmA to wild-type levels, and this correlated with a major improvement in growth capability, though not to wild-type levels (Fig. 2A). Presumably, additional defects in this mutant also impact its growth capability. Most significantly, the elevated expression of mexXY (ca. 6-fold) seen in strain K2966 was returned to wild-type levels in strain K3216 (Fig. 1C), confirming that reduced expression of rplU and/or rpmA was responsible for the elevated mexXY expression observed in K2966. As expected, too, restoration of rplU-rpmA and mexXY expression to wild-type levels in strain K3216 also generally restored susceptibility to aminoglycosides (Table 3).
Given the link between reduced rplU-rpmA expression and mexXY hyperexpression and aminoglycoside resistance in strain K2966, expression of the ribosomal protein operon was also assessed in other MexXY-hyperexpressing pan-aminoglycoside-resistant mutants previously recovered following peroxide exposure (12). One of these mutants, strain K2968 (Fig. 1C and Table 3), also demonstrated reduced (ca. 3-fold) expression of rplU and rpmA relative to wild-type strain K767 (Fig. 1A and andB)B) and a concomitant growth defect (Fig. 2B), although no mutations were present in the upstream promoter region of rplU-rpmA in this mutant. In an attempt to restore rplU-rpmA expression in K2968 and assess the impact on mexXY expression and aminoglycoside resistance, the rplU-rpmA operon was cloned into plasmid pRK415, and the resultant plasmid, pCL3, was introduced into strain K2968. As a control and because increased rplU-rpmA expression in strain K3216, which was derived from strain K2966, was sufficient to restore mexXY expression and aminoglycoside susceptibility to wild-type levels, pCL3 was also introduced into K2966. In neither case did pCL3 have any impact on aminoglycoside resistance, nor did it reverse the growth defect of these strains (data not shown). One possibility was that multicopy rplU-rpmA and possibly higher than normal production of ribosomal proteins L21 and L27, just as apparently with reduced production of these proteins, adversely impacts ribosome assembly or function. To address this, the rplU-rpmA operon, including the wild-type promoter region, was cloned into plasmid mini-CTX1, a vector capable of inserting into the P. aeruginosa genome at the phage D113 attB site (14), providing a means for delivery of cloned genes into the chromosome and, so, single-copy expression. Plasmid pCL2, a mini-CTX1 derivative carrying rplU-rpmA, was thus constructed and introduced into the chromosomes of strains K2966 and K2968. Excision of plasmid backbone sequences from strains K2966 and K2968 yielded strains K3217 and K3218, respectively, in which chromosomal rplU-rpmA was expressed from a wild-type promoter. The K3217 strain, derived from strain K2966, showed improved growth (Fig. 2A), reduced expression of mexXY (Fig. 1C), and enhanced aminoglycoside susceptibility relative to strain K2966 (Table 3), reminiscent of strain K3216 in which the rplU-rpmA promoter mutation of K2966 was corrected. This was consistent with rplU-rpmA expression being restored in K3217 (Fig. 1A and andB)B) and, so, reducing expression of mexXY. In contrast, strain K3218, derived from strain K2968, did not show any improvement in growth (Fig. 2B) or change in mexXY expression (Fig. 1C) or aminoglycoside susceptibility (Table 3) relative to K2968. Expression of rplU-rpmA was, however, restored to wild-type levels in K3218 (Fig. 1A and andB),B), suggesting that the mutation(s) responsible for reduced rplU-rpmA expression in K2968 influences expression of an additional gene(s) whose altered expression also negatively impacts growth and elevates mexXY expression. At present, the mutation responsible for the reduced expression of rplU-rpmA in K2968 has yet to be identified. In examining previously reported pan-aminoglycoside-resistant CF isolates in which MexXY contributed to the aminoglycoside resistance (44), it was noted that one mutant, K2154, showing elevated (2-fold) mexXY expression relative to the wild type (Fig. 3C) and lacking mutations in mexZ and the mexXY promoter region (44), also showed reduced (2-fold) expression of rplU-rpmA (Fig. 3A and andB).B). Like K2968, this isolate also lacked mutations in the rplU-rpmA upstream/promoter region.
Induction of mexXY by ribosome-targeting antimicrobials (8, 10) or reactive oxygen species (ROS) (12) is dependent on the antimicrobial- and ROS-inducible PA5471 gene product (8, 12, 35) which functions, at least in part, as an antirepressor for the MexZ regulator of mexXY expression (35, 52). As with antimicrobial and ROS exposure, reduced expression of rplU-rpmA in strains K2966 and K2968 yields an increase, albeit modest, in PA5471 expression (Fig. 1D), which was reversed in K2966 upon restoration of rplU-rpmA expression (see strains K3216 and K3217 in Fig. 1D). In addition, the increased mexXY expression seen in K2966 and K2968 is lost upon deletion of PA5471 (Fig. 4, strains K2973 and K2975), an indication that increased expression of this efflux operon owing to reduced rplU-rpmA expression is similarly PA5471 dependent.
As reported here, a modest 2-fold reduction in expression of the rplU-rpmA ribosomal protein operon is enough to perturb translation sufficiently so as to negatively impact P. aeruginosa growth and to stimulate PA5471 and mexXY expression. This presumably speaks to the essential nature of ribosomal proteins L21 and L27 in this organism. Indeed, numerous attempts to generate an rplU deletion derivative of P. aeruginosa were unsuccessful, in contrast with E. coli where the gene has been successfully deleted (42). That restoration of rplU-rpmA expression in strain K2966 by correcting a single base mutation in the putative promoter region of this operon was sufficient to reverse the increased PA5471 and mexXY expression of this mutant confirmed both that the mutation was the primary cause of reduced rplU-rpmA expression and that reduced expression of this ribosomal protein operon was responsible for elevated mexXY and aminoglycoside resistance. The fact that restoration of rplU-rpmA expression in strain K2968 was unable to do the same indicates that reduced rplU-rpmA expression is not the sole determinant of increased PA5471 and mexXY expression in this particular mutant and that the mutation responsible for the reduced rplU-rpmA expression has as its primary effect reduced expression and/or activity of some other, probably ribosomal, constituent(s), with reduced rplU-rpmA expression being a secondary effect. Expression and assembly of ribosomal constituents are highly regulated and involve a myriad of protein-protein and protein-RNA interactions (41) such that a reduction in the expression or activity of one constituent may well yield similar reductions in others. In any case, the mutation ultimately responsible for reduced rplU-rpmA expression and enhanced PA5471 and mexXY expression remains to be elucidated.
In E. coli, ribosomal protein L27 has been shown to occur in the peptidyl transferase center (PTC) of the 50S ribosomal subunit where it interacts with tRNA substrates (50) and is implicated in the proofreading step of acyl-tRNA selection (18). Thus, L27-deficient mutants of E. coli show reduced rates of peptidyl transferase activity and, as with the P. aeruginosa K2966 and K2968 mutants showing reduced rplU-rpmA expression, exhibit a significant growth defect (28). Less is known about the rplU-encoded L21, although it has been reported to interact with 23S rRNA (2, 48) and to bind Sepharose-immobilized erythromycin and chloramphenicol (47), both of which suggest that it may be accessible to the PTC of the 50S ribosome (where erythromycin and chloramphenicol act ). In any case, reduced L27 resulting from reduced rpmA expression in P. aeruginosa strains K2966 and K2968 is likely to interfere with normal peptidyl transferase activity of the ribosome, possibly resulting in the generation of truncated and, so, aberrant polypeptides (as occurs with exposure to ribosome-disrupting antimicrobial agents) and certainly the slowing down of protein translation. The latter is relevant in the context of the observed increase in PA5471 and mexXY in these mutants, since induction of PA5471 by ribosome-targeting antimicrobials has been shown to occur via a transcriptional attenuation mechanism that is based on antimicrobial-dependent stalling of the ribosome on a leader peptide sequence (PA5471.1) upstream of PA5471, this stalling being the signal for formation of an antiterminator structure that ultimately permits transcription of the downstream PA5471 (34). Presumably, slowing of peptidyl transfer activity owing to reduced rplU-rpmA expression also promotes ribosome stalling on PA5471.1 sequences and, so, enhances PA5471 and, ultimately, mexXY expression.
Previously described ribosomal protein mutations that influence mexXY expression include rplY (encoding ribosomal protein L25) (11) and rplA (encoding ribosomal protein L1) (51). Little is known about the function of L25, although the rplY knockout did exhibit a growth defect (less pronounced than that seen here) (11), while L1 interacts with tRNAs and has been implicated in tRNA movement through the ribosome during translation in E. coli (21), and E. coli rplA mutants lacking L1 show reduced rates of protein synthesis (46). Similar to our mutants reported herein, the P. aeruginosa rplA mutant hyperexpressing mexXY (dubbed amrAB in that study) shows a substantial growth defect (51), which is consistent with the presumed perturbation of translation in this mutant. In the case of the previously described rplY mutant, both PA5471 and mexXY were upregulated, with the latter dependent on PA5471, just as seen here for strains K2966 and K2968. In contrast with the rplY mutant, however, where the bulk of the aminoglycoside resistance was attributed to the rplY mutation itself and not the subsequent upregulation of mexXY (11), the aminoglycoside resistance of strains K2966 and K2968 was wholly attributable to MexXY—deletion of mexXY in these strains rendered them as susceptible to aminoglycosides as a mexXY deletion derivative of wild-type strain K767 (12). As with the rplY mutant, the increased expression of mexXY in an rplA mutant also did not appear to fully explain increased aminoglycoside resistance, since elimination of mexXY in this mutant reduced aminoglycoside MICs to or above that of the wild-type parent (51). How mutations in rplY and rplA promote aminoglycoside resistance independently of their induction of MexXY is unclear (aminoglycosides target the small ribosomal subunit and, so, mutations in large subunit components are not expected to influence aminoglycoside interaction with ribosomes).
While curtailing ribosome function may be the signal for PA5471 and mexXY recruitment, what is as yet unclear is the function of the MexXY-OprM efflux system in the context of ribosome stress and the attendant downstream adverse consequences of protein translation perturbation. Antimicrobials that target the ribosome and induce mexXY expression characteristically generate mistranslated and/or truncated polypeptides (9), and these aberrant polypeptides may well be additional triggers for mexXY expression (PA5471 is required but insufficient for this ) and, possibly, targets for expulsion by MexXY-OprM. Oxidative stress, another signal for recruitment of PA5471 and MexXY-OprM (12), also promotes mistranslation (25), and the ribosomal protein mutations described here can be expected to generate truncated polypeptides, as a result of their probable defects in elongation, making aberrant polypeptides possible candidates for mexXY effectors and MexXY-OprM substrates. Nevertheless, it cannot be ruled out that some “downstream stress” caused by these aberrant polypeptides might be the additional signal for mexXY and provide the “substrates” for this efflux system. Aminoglycoside-generated aberrant polypeptides are, for example, known to interact with and damage bacterial membranes (9), a significant factor in the lethality of these agents (23). Still, MexXY-OprM seems not to be associated with the envelope stress response that is controlled by the AmgRS two-component system (TCS) in P. aeruginosa. While this response is triggered by aminoglycoside exposure and protects cells from the adverse impact of these agents, mexXY is not regulated by AmgRS, and MexXY-OprM contributes to aminoglycoside resistance independent of this TCS (24). In any case, continued identification and study of mutations that activate mexXY expression independent of antimicrobial exposure may well provide essential clues to the signal(s) to which mexXY responds and the cellular product(s) that are the intended substrates for MexXY-OprM.
We thank Sam Basta (Queen's University) for assistance with the Varioskan microplate reader in assessing the growth of P. aeruginosa over time.
This work was funded by an operating grant from Cystic Fibrosis Canada (to K.P.). The genome sequencing was supported by the Pathogen Functional Genomics Resource Center (PFGRC) (contract N01-AI15447).
Published ahead of print 23 July 2012