Proteome-wide quantification and characterization of oxidation-sensitive cysteines in pathogenic bacteria

Abstract

Thiol-group oxidation of active and allosteric cysteines is a widespread regulatory posttranslational protein modification. Pathogenic bacteria, including Pseudomonas aeruginosa and Staphylococcus aureus, use regulatory cysteine oxidation to respond to and overcome reactive oxygen species (ROS) encountered in the host environment. To obtain a proteome-wide view of oxidation-sensitive cysteines in these two pathogens, we employed a competitive activity-based protein profiling approach to globally quantify hydrogen peroxide (H2O2) reactivity with cysteines across bacterial proteomes. We identified ∼200 proteins containing H2O2-sensitive cysteines, including metabolic enzymes, transcription factors, and uncharacterized proteins. Additional biochemical and genetic studies identified an oxidation-responsive cysteine in the master quorum-sensing regulator LasR and redox-regulated activities for acetaldehyde dehydrogenase ExaC, arginine deiminase ArcA, and glyceraldehyde 3-phosphate dehydrogenase. Taken together, our data indicate that pathogenic bacteria exhibit a complex, multilayered response to ROS that includes the rapid adaption of metabolic pathways to oxidative-stress challenge.

Figure 1. A modified isoTOP-ABPP to profile oxidation-sensitive cysteines in the whole P. aeruginosa and S. aureus proteomes

(A) Bacteria were treated with or without 10 mM H₂O₂ before protein lysates were isolated for subsequent labeling and LC-MS/MS quantification. Isotopic envelopes are shown for light- and heavy-labeled peptides with green lines representing predicted values. Sequences are shown for tryptic peptides containing IA-probe-labelled cysteines (marked by asterisks) in non-redox-sensitive DNA-directed RNA polymerase (Ratio = 0.93) and redox-sensitive pyridoxal biosynthesis lyase (Ratio = 0.38) in S. aureus. (B) Pie chart showing the percentage of four protein categories that have oxidation-sensitive cysteines in P. aeruginosa and S. aureus. (C) Pie chart illustrating the percentage of functionally annotated cysteines in P. aeruginosa and S. aureus. (D) Pie chart showing the percentage of conserved cysteines in P. aeruginosa and S. aureus. See also Figure S1.

Figure 2. Cys79 is involved in ligand binding and oxidation sensing in LasR

(A) Swarming motility of the following four P aerugionsa strains on swarming agar plates. WT, MPAOl wild-type-containing pAK1900; ΔlasR, ΔlasR containing pAK1900; Comp, ΔlasR containing pAK1900-lasR; C79A, ΔlasR containing pAK1900-C79A. (B) Relative lasI-lux activity of four P. aeruginosa strains. (C) Crystal structure of LasR ligand-binding domain (LBD) showing the vicinity (4.3 Å) between Cys79 and the terminal carbon of 3-oxo-C₁₂-HSL (Bottomley et al., 2007). (D) An AHL-bioassay determined the 3-oxo-C₁₂-AHL content of LasR-WT, LasR-C79A, and LasR-C79S. (E) Northern hybridization analysis showed the transcription level of lasI in both wild-type and ΔlasR, treated with increasing concentrations of H₂O₂ for 20 min. (F) Relative activity of lasI-lux in three P. aeruginosa strains (WT, ΔlasR containing pAK1900-lasR (WT); C79A, ΔlasR containing pAK1900-lasR-C79A; C79S, ΔlasR containing pAK1900-lasR-C79S) treated with 5 mM H₂O₂ for 20 min. Data are represented as mean +/− SEM. The asterisks denote that the differences are statistically significant (P < 0.05). See also Figure S2.

Figure 3. Oxidative stress inhibits acetaldehyde dehydrogenase (PA1984, ExaC) activity

(A) ExaC catalyzes the NAD(P)-dependent oxidation of acetaldehyde that produces acetic acid. (B) Enzymatic activity of purified ExaC-WT, ExaC-C301S, and ExaC-C303S in the presence of 10 µM DTT or 10 µM H₂O₂. (C) In vitro acetaldehyde dehydrogenase activity of bacterial lysates of four P. aeruginosa strains. WT/EV, wild-type MPAOl containing empty pAK1900; ΔexaC/EV, ΔexaC containing empty pAK1900; ΔexaC/p-exaC, ΔexaC containing pAK1900-exaC; ΔexaC/p-C303S, ΔexaC containing pAK1900-exaC-C303S. (D) Mass spectrometric mapping of the disulfide bond in oxidized ExaC. ESI-Q-TOF mass spectrum (m/z 200–1,000) of an unfractionated tryptic digest. The 3⁺ charge state (m/z 737) corresponding to the disulfide-containing peptide of interest (theoretical molecular mass, 2,210.04 Da). (E) Graphical fragment map correlating the fragmentation ions to the sequence of the disulfide-containing peptide. The disulfide-linked cysteines are circled. (F) Proposed model showing that oxidation of ExaC forms a disulfide between catalytically active Cys301 and Cys303, which leads to the inhibition of the ExaC activity. Data are represented as mean +/− SEM. The asterisks denote that the differences are statistically significant (P < 0.05). See also Figure S3.

Figure 4. Oxidation-induced activation of arginine deiminase (PA5171, ArcA)

(A) ArcA catalyzes the production of citrulline from arginine. (B) Enzymatic activity of purified ArcA-WT, ArcA-C37S, ArcA-C169S, ArcA-C286S, and ArcA-C363S in the presence or absence of 0.5 mM TCEP. (C) Effect of 1 mM cysteamine on the in vivo activity of ArcA in the wild-type MPAOl containing empty pAK1900 (WT), ΔarcA/pAK1900 (ΔarcA), ΔarcA/p-arcA (Comp), ΔarcA/p-arcA-C286S (C286S), and ΔarcA/p-arcA-C363S (C363S), respectively. Bacteria were cultured anaerobically at 37 °C in the M9 medium containing 10 mM L-arginine and PBS. (D) DTNB-based quantification of free thiols in anaerobically reduced or air-oxidized ArcA, ArcA-C286S, ArcA-C363S, and ArcA-C286S-C363S, respectively. (E) Enzymatic activity of these four purified proteins in different concentrations of GSH/GSSG. (F) Macrophage killing assay. P. aeruginosa strains were incubated with murine alveolar macrophages (AM) for 90 min and the surface bacteria were killed by polymyxin B for 1 h. CFUs were measured after lysing the cells and the plate was left in polymyxin B overnight. Data are represented as mean +/− SEM. The asterisks denote that the differences from the wild type are statistically significant (P < 0.05). See also Figure S4.

Figure 5. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) functions as a redox-sensing metabolic switch

(A) GAPDH catalyzes the conversion of glyceraldehyde 3-phosphate into glycerate 1,3-bisphosphate, the sixth and committed reaction in the glycolysis. (B) The relative activities of recombinant GAPDHs (from P. aeruginosa or S. aureus) treated with and without 10 µM H₂O₂ for 1 min. (C) NADPH accumulates upon H₂O₂ treatment in both pathogens. Data are represented as mean +/− SEM. The asterisks denote that the differences between H₂O₂-treated and untreated samples are statistically significant (P < 0.05). Also see Figure S5.

Figure 6. A summary showing the effects of ROS on protein examples identified and characterized in this study

ROS causes conformational change of the oxidized LasR, and thus reduces the promoter activity of its downstream gene, lasI. Oxidative inhibition of ExaC involves the formation of a disulfide between catalytic Cys301 and a neighbor Cys303. For ArcA, ROS induces activation of its enzymatic activity by oxidizing Cys286 and Cys363 away from the active site. GAPDH with catalytic active cysteines that are oxidation-sensitive can be inhibited by ROS-mediated direct oxidation, which leads to metabolic reroute. SH, reduced cysteine; SX, oxidized cysteine; S-S, disulfide.