INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

The major response of Escherichia coli to hydrogen peroxide (H₂O₂) is governed by a 34-kDa transactivator, OxyR (5, 6, 30, 31). OxyR positively regulates katG (encoding hydroperoxidase I), gorA (encoding glutathione reductase), ahpCF (encoding alkyl hydroperoxide reductase), dps (encoding a nonspecific DNA-binding protein) (1, 5, 6), and fur (encoding ferric uptake regulatory protein) (31). Each of these genes is important in combating H₂O₂-mediated stress. OxyR acts as a transcriptional autorepressor under noninducing conditions. However, in the presence of its inducer, H₂O₂, an intramolecular disulfide bond that activates OxyR is formed, allowing it to then govern transcription of OxyR-dependent promoters (24, 26).

In contrast to the case for E. coli, genes under OxyR control in Pseudomonas aeruginosa include the katB-ankB operon, ahpB, and ahpCF (17). These genes encode catalase B (KatB), an ankyrin-like protein that is necessary for optimal KatB activity (AnkB), and two alkyl hydroperoxide reductases, AhpB and AhpCF, respectively (17). In this work, we report dramatic sensitivity of a P. aeruginosa oxyR mutant to H₂O₂ and aminoglycosides, despite possessing wild-type catalase activity. Growth of serially diluted oxyR organisms under aerobic conditions was found to require the major housekeeping catalase, KatA, which was released into the extracellular milieu upon cell lysis. KatA was found to be resistant to a number of proteases, including the human neutrophil serine protease cathepsin G. Our data suggest that released KatA could remain for extended periods, especially in P. aeruginosa biofilms of environmental, industrial, and clinical importance. Such resilience would afford viable organisms, especially those in a biofilm (10), additional protection against H₂O₂.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Bacterial strains, plasmids, and media. All P. aeruginosa and E. coli strains used in this study are listed in Table 1 and were maintained on Luria (L) agar (10 g of tryptone, 5 g of yeast extract, 10 g of NaCl) or M9 minimal medium agar (6 g of Na₂HPO₄, 3 g of KH₂PO₄, 1 g of NH₄Cl, 0.5 g of NaCl, 1 mM MgSO₄ · 7H₂O, and 0.2% glucose [per liter]) plates, with each medium solidified with 15 g of Bacto agar per liter. All strains were stored indefinitely at 80°C in either 12.5% glycerol or 10% skim milk.

Chemicals. Bovine serum albumin, carbenicillin, chloramphenicol, horseradish peroxidase, H₂O₂, cumeme hydroperoxide (CHP), potassium ferricyanide, ferric chloride, o-dianisidine dihydrochloride, pepsin, trypsin, and chymotrypsin were from Sigma Chemical Company (St. Louis, Mo.). Gentamicin, tetracycline and Triton X-100 were from Fisher Scientific. Human neutrophil cathepsin G was from Calbiochem (La Jolla, Calif.). Tobramycin was from Eli Lilly (Indianapolis, Ind.), and amikacin was from Bristol-Myers Squibb (Princeton, N.J.).

Growth conditions. All bacteria were propagated from single colonies or stationary-phase suspensions that had been grown in either L broth or M9 minimal medium. Liquid cultures were grown at 37°C with shaking at 300 rpm or on a roller wheel at 70 rpm unless otherwise indicated. Culture volumes were 1/10 of the total Erlenmeyer flask volume to ensure proper aeration. Bacteria were also grown under anaerobic conditions in an anaerobic chamber (model 1024; Forma Scientific, Marietta, Ohio).

Serial dilution experiments. Bacteria were grown aerobically in L broth or anaerobically in L broth containing 1% KNO₃ for 17 h at 37°C. Suspensions were serially diluted in sterile aerobic or anaerobic L broth or filter-sterilized (0.2-µm-pore-size filter) spent culture supernatants from various bacteria in microtiter dishes (Becton Dickinson, Franklin Lakes, N.J.). Five or 20 microliters of each dilution was spotted onto L-agar or L-agar-1% KNO₃ plates and incubated aerobically or anaerobically, respectively.

Identification of OxyR-regulated genes that rescue the aerobic dilution defect. Two phenotypes of the oxyR mutant constructed in this study were a dramatic sensitivity to H₂O₂ and an inability to survive aerobic serial dilution in L broth. To identify OxyR-regulated or other genes that could rescue the dilution defect, the oxyR, ahpB-tdr, ahpCF, katB, dps, and recG genes were expressed constitutively in the oxyR mutant via the lac promoter of pUCP19 or pUCP22 (29). The construction of pUCP-oxyR, pUCP-katB, pUCP-ahpCF, and pUCP-recG is described in an accompanying paper by Ochsner et al. (17). Plasmid pUCP-ahpB-tdr was isolated from a pUCP19-based plasmid library of P. aeruginosa PAO1 and contained a 6-kb partial Sau3A fragment that carried the ahpB gene under the control of the plasmid-borne P_lac promoter and also harbored the tdr gene. The dps gene was PCR amplified using primers dps-845 (5'-XbaI-tctaGAGTGCCCCATTCACAAGG and dps-1377 (5'-HindIII-aaGCTTTCTATCGAGCGATCGA (bases in the restriction enzyme cleavage site are underlined, and designed and mismatched bases are in lowercase type). The 539-bp PCR product was then cloned into pCR-Blunt and ligated directionally as an XbaI-HindIII fragment into pUCP22, yielding pUCP-dps, where the dps gene is expressed from the plasmid-borne P_lac promoter. The strains containing these recombinant plasmids were grown aerobically for 17 h in L broth and serially diluted in sterile, aerobic L broth in microtiter dishes as described above. Suspensions (20 µl) of each dilution were spotted on L-agar plates containing 400 µg of carbenicillin per ml.

Manipulation of recombinant DNA. Plasmid DNA was transformed into E. coli DH5-MCR (Gibco-BRL, Gaithersburg, Md.) or SM10 (21). 5-Bromo-4-chloro-3-indolyl--D-galactopyranoside (X-Gal) (40 µg/ml) was routinely added to agar medium to detect the presence of insert DNA. Restriction endonuclease, alkaline phosphatase, Klenow fragment, T4 DNA polymerase, and T4 DNA ligase were used as specified by the vendor (Gibco-BRL). Plasmid DNA was isolated using miniprep kits from Qiagen. DNA fragments used for cloning or in the construction of radiolabeled probes were recovered from agarose gels using SeaPlaque low-melting-point agarose (FMC BioProducts, Rockland, Maine). Gene replacements were done using the pEX100T vector (19) as described in detail in the accompanying paper by Ochsner et al. (17).

H₂O₂, CHP, and antibiotic sensitivity assays. P. aeruginosa strains were grown aerobically in M9 minimal medium for 17 h at 37°C. Each culture was diluted in 3 ml of 0.8% low-melting-point M9 agarose (SeaPlaque) to a final optical density at 600 nm of 0.2 and maintained at 37°C. These suspensions were then distributed evenly on M9 agar plates, and the top agarose was allowed to solidify. Filter paper disks containing 10 µl of either 0.2% H₂O₂, 20% CHP, gentamicin (40 mg/ml), amikacin (40 mg/ml), tobramycin (25 mg/ml), carbenicillin (50 mg/ml), chloramphenicol (34 mg/ml in 95% ethanol), or tetracycline (15 mg/ml in 50% ethanol) solution were placed on the top agar surface, and the plates were incubated aerobically at 37°C for 17 to 24 h. The zones of growth inhibition were then recorded in triplicate.

Cell extract preparation and biochemical assays. Cell extracts were prepared from bacteria harvested by centrifugation at 10,000 × g for 10 min at 4°C. Bacteria were washed twice in ice-cold 50 mM sodium phosphate buffer (pH 7.0) and sonicated in an ice-water bath for 10 s with a Heat Systems-Ultrasonics (Farmingdale, N.Y.) model W-225 sonicator equipped with microtip at output setting 5. The sonicate was clarified by centrifugation at 13,000 × g for 10 min at 4°C. Cell extract preparation for native gel electrophoresis was performed as described above except that 50 mM Tris-HCl (pH 7.4) was used as the diluent. Catalase activity was assayed spectrophotometrically at 240 nm by monitoring the decomposition of 18 mM H₂O₂ using a Spectronic Genesys 5 spectrophotometer (Spectronic Unicam, Rochester, N.Y.) (2). Catalase activity gels (5%) were stained according to the method of Wayne and Diaz (28). One unit of catalase activity was that which degraded 1 µmol of H₂O₂ per min at 23°C. The production of H₂O₂ in aerobic L broth was monitored as follows. For determination of H₂O₂ formation by autoxidation of aerobic L broth, L broth was diluted 1:1 with a solution containing 0.01% horseradish peroxidase, 0.2% Triton X-100, 0.63 mM o-dianisidine dihydrochloride, and 1 mM EDTA in 50 mM acetate buffer, pH 5.0. The production of H₂O₂ was monitored spectrophotometrically at 460 nm by the oxidation of o-dianisidine dihydrochloride. Bovine liver catalase (100 U/ml) (Boehringer-Mannheim) was used to inhibit H₂O₂ production. Susceptibility of P. aeruginosa KatA to proteolysis was assayed by incubating sterile spent culture supernatant containing KatA activity with an equal volume of various proteases (1 mg/ml) or 1 mM dithiothreitol for 1 h at 37°C, after which catalase activity was recorded. Neutrophil cathepsin G was used at 100 mU/ml. Protein concentrations in cell extracts were estimated by the method of Bradford (3), using bovine serum albumin (fraction V; Sigma) as a standard.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Phenotypes of a P. aeruginosa oxyR mutant. (i) H₂O₂ sensitivity despite possession of normal catalase activity. In our accompanying work (17), we describe the cloning and characterization of genes under the control of the global regulator OxyR in P. aeruginosa. These were found to include the katB-ankB operon, encoding a 228-kDa catalase and an ankyrin-like protein, AnkB, that is essential for optimal KatB activity, and ahpB and ahpCF, encoding two alkyl hydroperoxide reductases, AhpB and AhpCF, respectively. When we initially constructed an isogenic P. aeruginosa oxyR mutant (17), we postulated that it would be very sensitive to H₂O₂. Although the oxyR mutant possessed wild-type catalase activity (Table 2), we found the mutant to be to H₂O₂ sensitive but intermediate in sensitivity with respect to a catalase-deficient katA katB mutant when grown in L broth (Table 2) (10). Interestingly, the oxyR mutant was more sensitive to H₂O₂ than the katA katB mutant when grown in M9 minimal medium (Table 2).

(ii) Aerobic serial dilution defect of the oxyR mutant and protection by P. aeruginosa catalase KatA. During our selection for a double-crossover event while constructing our oxyR mutant, there were virtually no isolated colonies on the final selection medium, only patches of growth where high cell densities were distributed on the agar surface (data not shown). Hence, the oxyR mutant could not survive aerobic serial dilution, but it could in the presence of bovine liver catalase (Fig. 1B), in the presence of sterile spent culture supernatants of wild-type or oxyR mutant bacteria (Fig. 1), or when grown anaerobically. To determine the OxyR-dependent or -independent gene product(s) that could rescue this defect, a panel of mutants that had been compromised in some facet of oxidative stress machinery was screened for a potential aerobic serial dilution defect. These organisms were also serially diluted in sterile spent culture supernatants from seven different bacteria that lacked one or both catalases, alkyl hydroperoxide reductases, or various combinations of both. As shown in Fig. 1A, when diluted in L broth, the oxyR (lane 10) and oxyR katA (lane 11) mutants could survive at only a 10-fold or less serial dilution. Interestingly, serial dilution in sterile supernatants of wild-type, katB, and ahpB ahpCF bacteria rescued both oxyR and oxyR katA bacteria. Sterile spent supernatant from katA, ahpB katA, or katA katB mutants could not completely rescue diluted oxyR, katA oxyR, katA ahpB, or katA katB mutants. In fact, even the ahpB mutant alone could not be fully serially diluted in the latter supernatants, yet the strain was fully viable when diluted in L broth. Thus, an autoxidizable factor present in the spent culture supernatant of katA, katA katB, and katA ahpB strains sensitized ahpB, ahpB katA, and katA katB bacteria. Although this factor is unknown at present, we believe it to be the redox-cycling antibiotic pyocyanin.

View larger version (124K): [in this window] [in a new window]   FIG. 1.   Aerobic serial dilution defect of P. aeruginosa oxyR, oxyR katA, and katA katB strains. (A) Bacteria were grown aerobically in L broth until stationary phase. Organisms were serially diluted in either L broth or filter-sterilized spent culture supernatants (Supe) from different organisms to determine the effect of various gene products on rescue of the serial dilution defect of the oxyR mutant. Five-microliter suspensions of 101 through 108 serial dilutions were spotted on square L-agar plates and incubated at 37°C under aerobic conditions for 17 h. The top of each plate was removed, the plate was placed top-side-down on a scanner, and a black sheet of paper was laid carefully on top of each plate to enhance contrast. The images were scanned using ScanGal software and finally stored as a PowerPoint 98 file. Lanes: 1, PAO1; 2, katA; 3, katB; 4, katA katB; 5, ahpB; 6, ahpB katA; 7, ahpCF; 8, ahpB ahpCF; 9, recG; 10, oxyR; 11, oxyR katA; 12, lasI rhlI. (B) Bacteria harboring various pUCP-Plac based plasmids containing the genes listed to the left were grown aerobically in L broth containing 400 µg of carbenicillin for 17 h per ml at 37°C. Serial dilutions were made in either fresh L broth, filter-sterilized (0.2-µm-pore-size filter) spent culture supernatants of wild-type PAO1 or the oxyR mutant, or L broth containing 100 U of bovine liver catalase per ml. Twenty microliters of each dilution was spotted onto square L-agar-carbenicillin plates and incubated aerobically for 17 h at 37°C.

(iii) Constitutive expression of OxyR-controlled AhpB and AhpCF partially rescues the aerobic dilution defect but not H₂O₂ or CHP resistance. One simple means to identify OxyR-regulated genes that restore the aerobic dilution defect and/or wild-type H₂O₂ resistance was to express ahpB, ahpCF, and katB in the oxyR mutant via the lac promoter of pUCP19 (29). This allows for constitutive expression of what are otherwise stringently regulated loci. Predictably, the dilution defect (Fig. 1B) and H₂O₂ and CHP resistance (Table 2) were rescued in trans by oxyR itself. The recG, ahpB, and ahpCF genes provided in trans could partially restore the dilution defect (Fig. 1B). Another OxyR-controlled gene, katB (4, 17), encoding an H₂O₂-inducible catalase, KatB, could not rescue the dilution defect, nor could the Dps protein. The dps gene is regulated by OxyR in E. coli (1) but not in P. aeruginosa (data not shown). Interestingly, Dps, AhpB-Tdr, and AhpCF could provide some protection against H₂O₂, with only OxyR providing full protection. CHP resistance was more than fully restored by provision of oxyR in trans. Interestingly, AhpB and AhpCF overexpression in the oxyR mutant did not significantly alter sensitivity to CHP, and overexpression of KatB and Dps actually increased sensitivity to CHP (Table 3).

(iv) Hypersusceptibility to aminoglycoside but not to other antibiotics that do not cause an oxidative stress. Aminoglycosides are a class of antibiotics that require oxidative phosphorylation for antimicrobial activity. In fact, it was recently shown that gentamicin exposure evokes production of reactive oxygen species (20, 27). Thus, if oxidative stress defense systems are compromised, as in the oxyR mutant, we predicted that it would be more sensitive to aminoglycosides. To test this hypothesis, an unmarked oxyR deletion strain, oxyR-um, was constructed. As shown in Table 3, the oxyR-um mutant was far more sensitive than wild-type bacteria to several aminoglycosides, including gentamicin, tobramycin, and amikacin. In contrast, there was no difference in sensitivity to antibiotics that do not evoke an oxidative stress, including carbenicillin, tetracycline, and chloramphenicol.

Catalase (KatA) activity is present in spent culture supernatants of the oxyR mutant. To identify the catalase released upon cell lysis of stationary-phase organisms, culture supernatants from several strains were assayed for catalase activity. Table 2 demonstrates that wild-type and oxyR mutant bacteria possess 1,391 ± 216 and 1,858 ± 230 U of catalase activity per mg in their spent culture supernatants, respectively, while the katA katB mutant possesses none. In either case, supernatant catalase activity was attributable to KatA, since KatB was predictably absent because bacteria were not treated with H₂O₂ (4). As expected, no catalase activity was present in the katA (Fig. 2, lane 2), katA katB (lane 4), and ahpB katA (lane 5) strains relative to the wild-type (lane 1), katB (lane 3), or ahpB ahpCF (lane 6) strains. Thus, KatA is likely released into the culture supernatant upon cell lysis, a result confirmed by detection of fumarase (a cytoplasmic enzyme) activity, in the supernatants (data not shown). Additional evidence that KatA was not secreted came from construction of a KatA-PhoA translational fusion plasmid that yielded no alkaline phosphatase activity in the periplasmic space (data not shown). Interestingly, KatA was found to be a very stable enzyme, since it was highly resistant to sodium dodecyl sulfate, proteinase K, sodium dodecyl sulfate-proteinase K, pepsin, chymotrypsin, and trypsin but was inactivated by dithiothreitol and boiling. KatA activity was also retained after exposure to the human neutrophil protease cathepsin G (data not shown).

View larger version (73K): [in this window] [in a new window]   FIG. 2.   Catalase activity and gel staining of spent culture supernatants of P. aeruginosa. P. aeruginosa strains were grown aerobically in L broth for 17 h at 37°C. The bacteria were harvested by centrifugation at 10,000 × g for 10 min, and the supernatants were filtered through 0.2-µm-pore-size filters. Culture supernatants were subjected to electrophoresis using 6% nondenaturing acrylamide gels and stained for catalase activity. The number with each lane is the percent catalase activity with respect to that of wild-type bacteria.

How much KatA is needed for rescue of the oxyR mutant in L broth? To determine the precise amount of KatA required for aerobic rescue of the oxyR mutant, wild-type spent culture supernatants were serially diluted in L broth and plated on L-agar plates as described above. As shown in Fig. 3, a 10-fold dilution of the oxyR mutant in L broth alone allowed for survival. A weakening of the rescue effect could be seen at the 10⁶ to 10⁸ dilutions when the supernatant was diluted 10-fold. Still, the most dramatic effect was observed when the supernatant was diluted 20-fold or greater. In this particular experiment, the stationary-phase suspension contained 2 × 10¹⁰ CFU/ml and the supernatant (~120 µg of protein per ml) contained ~688 U of KatA activity per mg. Thus, the total number of organisms present in the first 5-µl spot of the 10¹ dilutions is ~1 × 10⁷ CFU/ml. On a per-organism basis in a microtiter well volume of 200 µl, and based upon the near-complete inhibition by 20-fold-diluted supernatant of the growth of the oxyR mutant, we calculate that at least 16.5 U of KatA activity per well is required for dilution beyond 10⁷ CFU/ml. This corresponds to a spent culture supernatant protein requirement, based upon this experiment, of at least 2.4 µg/200-µl well volume.

View larger version (115K): [in this window] [in a new window]   FIG. 3.   Titration of KatA activity required for aerobic rescue of serially diluted oxyR bacteria. P. aeruginosa oxyR was grown aerobically to stationary phase in L broth. The bacteria were harvested by centrifugation at 10,000 × g for 10 min, and the supernatants were filtered through 0.2-µm-pore-size filters. The bacterial suspension was serially diluted in either L broth or various dilutions of wild-type spent culture supernatant (Supe) in microtiter dishes with a well volume of 200 µl. In this particular experiment, the catalase specific activity of the culture supernatant was 688 U/mg.

H₂O₂ is produced as an autoxidation product of aerobic L broth. Because of the dramatic sensitivity of oxyR bacteria to H₂O₂, coupled with its inability to be serially diluted in aerobic L broth, we predicted that H₂O₂ was generated in L broth via autoxidation. To test this hypothesis, H₂O₂ production in aerobic L broth was monitored by the horseradish peroxide-catalyzed oxidation of o-dianisidine (16). As shown in Fig. 4, a linear rate of 1.2 nmol of H₂O₂ s/min/ml (or 1.2 µM H₂O₂/min) was produced by aerobic L broth, an amount which gradually reached a plateau as oxygen became depleted in the cuvette (data not shown). Not surprisingly, H₂O₂ production was inhibited by the addition of 100 U of bovine liver catalase.

View larger version (13K): [in this window] [in a new window]   FIG. 4.   H2O2 is produced as an autoxidation product of aerobic L broth. Ten milliliters of L broth was shaken vigorously at room temperature for 1 h. Samples were added to a solution containing 0.01% horseradish peroxidase, 0.2% Triton X-100, 0.63 mM o-dianisidine dihydrochloride, and 1 mM EDTA in 50 mM acetate buffer, pH 5.0. The production of H2O2 by autoxidation was monitored spectrophotometrically at 460 nm by the oxidation of o-dianisidine dihydrochloride. The addition of 100 U of bovine liver catalase per mg inhibited H2O2 production.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

H₂O₂ is used to treat problematic bacterial biofilms in environmental and industrial settings (7, 23). It is also produced by phagocytic cells to kill invading microbes during the infection process (8-10, 14). In its presence, however, survival of P. aeruginosa would depend upon an ability to rapidly mount a strategic defense against it. Such a response is governed, in part, by OxyR, which controls the katB-ankB operon, ahpB, and ahpCF (17). One likely reason for the dramatic H₂O₂ sensitivity of the oxyR mutant is that each OxyR-regulated gene product contributes some measure of protection against it (17). We have recently shown that KatB is required for optimal H₂O₂ resistance and that KatB activity is increased ~50-fold in its presence (4, 13). Because KatB can be detected in the cytoplasmic membrane and periplasmic fractions (13), it could act to protect H₂O₂-sensitive respiratory chain components. Optimal KatB-mediated protection also requires AnkB, a cytoplasmic membrane-periplasmic ankyrin-like protein that we believe anchors KatB to the inner membrane and/or stabilizes it is the periplasmic space (13). Although AhpB and AhpCF are predicted to encode proteins that are important in the detoxification of organic hydroperoxides, both AhpB and AhpCF appear to be important for some resistance to H₂O₂ when they are overexpressed in an oxyR mutant (Table 3). Interestingly, both ahpCF and ahpB are activated by H₂O₂ and CHP, and an ahpB mutant is very sensitive to H₂O₂ (17). AhpB could be the first line of defense in H₂O₂ detoxification, because AhpB-PhoA analyses indicate that it is a periplasmic protein (data not shown). Thus, AhpB, like KatB-AnkB (13), could also serve to protect H₂O₂-sensitive membrane components from oxidation by H₂O₂ or more hazardous downstream radicals (e.g., HO^·).

Haemophilus influenzae, Xanthomonas campestris (15a), and E. coli oxyR mutants require catalase for normal aerobic growth. In this study, we found that the aerobic dilution defect was due to the production of ~1.2 µM H₂O₂ min¹ via low-level autoxidation of components present in L broth. The H₂O₂ produced simply by aerobic L broth was sufficient to kill serially diluted oxyR and oxyR katA bacteria, and KatA, when present in the spent culture supernatant, and to a lesser extent periplasmic AhpB protected the diluted organisms. This suggests that P. aeruginosa possesses an exquisitely H₂O₂-sensitive protein that is likely in the cytoplasmic membrane. Interestingly, the  subunit of F₁F_o-ATPase of E. coli is oxidized by H₂O₂ and could represent such a target in P. aeruginosa (25). Because we also found that KatA is very resistant to multiple proteases, including the relatively nonspecific proteinase K and the clinically relevant neutrophil cathepsin G, it could persist for extended periods when release from lysed cells. Thus, KatA could prevent H₂O₂ penetration and killing of bacteria, especially those growing as a biofilm, because it would essentially be an antioxidant "land mine" (7, 23). KatA-mediated protection against H₂O₂ is also clinically relevant because human phagocytic cells produce copious quantities of H₂O₂ as part of an oxygen-dependent antimicrobial regimen against bacterial infections. In seven different sputum samples from cystic fibrosis patients, we were unable to detect P. aeruginosa KatA activity, but we did find high levels of neutrophil catalase (data not shown). The likely reason that we could not detect KatA activity in the sputum samples is that P. aeruginosa titers are being controlled to ~10⁴ organisms/ml of sputum with very aggressive antibiotic regimens.

Another phenotype of the oxyR mutant was a dramatic sensitivity to aminoglycosides, compounds whose antibacterial activity is dependent upon oxidative phosphorylation. Interestingly, aminoglycosides are used as part of aggressive antibiotic regimens (typically with -lactams and/or quinolones) in the treatment of P. aeruginosa infections, especially in patients with cystic fibrosis (22). Thus, our results suggest that OxyR is essential for resistance to not only H₂O₂ but also aminoglycosides. Finally, this study implies that OxyR, which is present in a myriad of different bacteria, could be a target for novel pharmaceuticals. Successful compounds would essentially render bacteria oxyR mutants that would be unable to successfully colonize new niches in the presence of micromolar levels of H₂O₂, aminoglycosides, or a combination of both.