Identification of gene products involved in the oxidative stress response of Moraxella catarrhalis

Abstract

Moraxella catarrhalis is subjected to oxidative stress from both internal and environmental sources. A previous study (C. D. Pericone, K. Overweg, P. W. Hermans, and J. N. Weiser, Infect. Immun. 68:3990-3997, 2000) indicated that a wild-type strain of M. catarrhalis was very resistant to killing by exogenous hydrogen peroxide (H₂O₂). The gene encoding OxyR, a LysR family transcriptional regulator, was identified and inactivated in M. catarrhalis strain O35E, resulting in an increase in sensitivity to killing by H₂O₂ in disk diffusion assays and a concomitant aerobic serial dilution effect. Genes encoding a predicted catalase (KatA) and an alkyl hydroperoxidase (AhpCF) showed dose-dependent upregulation in wild-type cells exposed to H₂O₂. DNA microarray and real-time reverse transcription-PCR (RT-PCR) analyses identified M. catarrhalis genes whose expression was affected by oxidative stress in an OxyR-dependent manner. Testing of M. catarrhalis O35E katA and ahpC mutants for their abilities to scavenge exogenous H₂O₂ showed that the KatA catalase was responsible for most of this activity in the wild-type parent strain. The introduction of the same mutations into M. catarrhalis strain ETSU-4 showed that the growth of a ETSU-4 katA mutant was markedly inhibited by the addition of 50 mM H₂O₂ but that this mutant could still form a biofilm equivalent to that produced by its wild-type parent strain.

FIG. 1.

Schematic representation of the M. catarrhalis oxyR, ahpC, ahpF, and katA loci in wild-type and mutant strains. (A) The oxyR, ahpC, and ahpF genes in wild-type M. catarrhalis O35E; (B) the katA locus in wild-type M. catarrhalis O35E; (C) the M. catarrhalis O35EΔoxyR mutant; (D) the O35EΔahpC mutant; (E) the O35EΔkatA mutant. Bent arrows represent relevant oligonucleotide primers.

FIG. 2.

Structure-based sequence alignment of M. catarrhalis OxyR with OxyR proteins from both E. coli and P. aeruginosa. The numbering is given on each side of the sequences. Areas of predicted helices are shown in red, while areas of predicted β-strands are blue. (A) The secondary structural elements of E. coli OxyR, as assigned by DSSP (17) from PDB entry 1I69, are shown above the sequences as red cylinders (helices) and blue arrows (β-strands). Areas of identity between the two structures are marked with asterisks. Where no structural information is available, predictions from a separate PSIPRED (16) prediction are shown in the color code described above. Clustal W (23) was used for the DNA-binding domain alignment. PROMALS3D (37) was used for the remainder of the alignment. Residues shown to be important for DNA binding in E. coli OxyR are boxed. The vertical black arrows indicate the conserved cysteine residues critical to the function of E. coli OxyR. (B) Alignment with the PA0218 OxyR protein. PROMALS3D (37) was used for the entire alignment. All coloring and notation conventions are the same as for panel A.

FIG. 3.

Characterization of the M. catarrhalis O35EΔoxyR mutant and related strains. (A) Western blot analysis of whole-cell lysates of M. catarrhalis O35E (lane 1), O35EΔoxyR (lane 2), O35EΔoxyR(pWL01) (lane 3), O35EΔoxyR(pWW115) (lane 4), and O35EΔoxyR (repaired) (lane 5) using polyclonal rat OxyR antiserum as the primary antibody. The black arrow on the right indicates the position of the OxyR protein. (B) Comparison of the growth rates of M. catarrhalis O35E (▪), O35EΔoxyR (•), O35EΔoxyR(pWL01) (□), O35EΔoxyR(pWW115) (○), and O35EΔoxyR (repaired) (▾). Results of a representative experiment are shown. (C) Sensitivities of the five strains for which results are shown in panels A and B to 88 mM H₂O₂ in disk diffusion assays. (D) Diameter of the zone of growth inhibition for each strain for which results are shown in panel C. Statistical analysis utilizing two-way ANOVA was utilized to determine the significance of each strain's zone of growth inhibition relative to that of wild-type M. catarrhalis O35E. Significant differences from the diameter of the zone of growth inhibition of M. catarrhalis O35E (bar 1) were observed for the zones of growth inhibition obtained with M. catarrhalis O35EΔoxyR (bar 2) and O35EΔoxyR(pWW115) (bar 4).

FIG. 4.

Catalase reverses the aerobic serial dilution defect of the M. catarrhalis O35EΔoxyR mutant. Serial dilutions of the wild-type strain O35E (A and B) and the O35EΔoxyR mutant (C and D) were spotted onto BHI agar either lacking catalase (−) or containing catalase (+) and were then incubated overnight.

FIG. 5.

Transcriptional responses of the M. catarrhalis O35E ahpC, ahpF, and katA genes to increasing concentrations of exogenous H₂O₂ as measured by real-time RT-PCR. Cultures of wild-type M. catarrhalis O35E growing in BHI broth were exposed to increasing concentrations of H₂O₂ (10, 20, and 50 mM), and total RNA was extracted and utilized in real-time RT-PCRs with primer pairs specific for the M. catarrhalis ahpC (filled bars), ahpF (open bars), or katA (shaded bars) gene. These data are reported as the fold change in transcription from that of M. catarrhalis cells that were exposed to water instead of H₂O₂.

FIG. 6.

Comparison of selected DNA microarray data with real-time RT-PCR results for gene expression as affected by the presence of H₂O₂ and OxyR. Shown are the effects of the presence of 50 mM H₂O₂ on the expression of selected genes by strain O35E in the presence and absence of the oxyR gene product, as measured by DNA microarray analysis (filled bars) and real-time RT-PCR (open bars).

FIG. 7.

Abilities of different M. catarrhalis strains and constructs to scavenge H₂O₂. H₂O₂ was added to a final concentration of 10 μM to ∼1.6 × 10⁷ CFU of bacteria and was incubated for 1 min at room temperature. A nonsignificant difference in the scavenging of H₂O₂ was seen between M. catarrhalis O35EΔoxyR (bar 2) and the wild-type strain O35E (bar 1). A statistically significant reduction in the scavenging of H₂O₂ was observed for O35EΔkatA (*, P = 0.01) (bar 3). Significant increases in the scavenging of H₂O₂ by O35EΔahpC (#, P = 0.002) (bar 4) and O35EΔkatA(pTH61-1) (bar 5) (##, P = 0.05) over that by the wild-type strain O35E were observed. O35EΔkatA(pWW115) (bar 6) degraded H₂O₂ to about the same extent as the ΔkatA mutant.

FIG. 8.

Effects of oxyR, ahpC, and katA mutations on the growth response to exogenous H₂O₂ and on biofilm formation. (A) Growth response of wild-type M. catarrhalis ETSU-4 (•), the ETSU-4ΔoxyR mutant (□), the ETSU-4ΔahpC mutant (*), the ETSU-4ΔkatA mutant (▵), ETSU-4ΔkatA(pTH61-1) (▪), and ETSU-4ΔkatA(pWW115) (○) after the addition of H₂O₂ to a final concentration of 50 mM. H₂O₂ was added to each culture when it reached an OD₆₀₀ of 0.7. No viable M. catarrhalis ETSU-4ΔkatA cells were isolated from the culture at the end of the experiment. (B) Measurement of biofilm formation by wild-type M. catarrhalis ETSU-4 (bar 1), M. catarrhalis ETSU-4ΔoxyR (bar 2), M. catarrhalis ETSU-4ΔkatA (bar 3), and M. catarrhalis ETSU-4ΔahpC (bar 4) in the crystal violet-based biofilm assay. Bar 5 shows the result obtained from a mock (uninoculated) well.