Combination Therapy Strategy of Quorum Quenching Enzyme and Quorum Sensing Inhibitor in Suppressing Multiple Quorum Sensing Pathways of P. aeruginosa

Abstract

The threat of antibiotic resistant bacteria has called for alternative antimicrobial strategies that would mitigate the increase of classical resistance mechanism. Many bacteria employ quorum sensing (QS) to govern the production of virulence factors and formation of drug-resistant biofilms. Targeting the mechanism of QS has proven to be a functional alternative to conventional antibiotic control of infections. However, the presence of multiple QS systems in individual bacterial species poses a challenge to this approach. Quorum sensing inhibitors (QSI) and quorum quenching enzymes (QQE) have been both investigated for their QS interfering capabilities. Here, we first simulated the combination effect of QQE and QSI in blocking bacterial QS. The effect was next validated by experiments using AiiA as QQE and G1 as QSI on Pseudomonas aeruginosa LasR/I and RhlR/I QS circuits. Combination of QQE and QSI almost completely blocked the P. aeruginosa las and rhl QS systems. Our findings provide a potential chemical biology application strategy for bacterial QS disruption.

Figure 1

Simulation results of stationary AHL concentration to QQ and QSI. (A) η(QQ), (B) QSI, (C) η(QQ) at different QSI concentrations, and (D) QSI at different η(QQ) values.

Figure 2

Simulation QS states to QQ and QSI. (A) 3D stationary AHL concentration to η(QQ) and QSI, (B) 2D map of QS on and off states to η(QQ) and QSI.

Figure 3

(A) Structure of G1. (B) Combination effects of G1 and AiiA on PAO1-lasB-gfp reporter strain. (C) Dose-response effects of G1 (50 µM) added with AiiA (32 µg/ml and 16 µg/ml) on lasB-gfp. (D) IC₅₀ values of the single and combination treatments on lasB-gfp, G1 (1.36 ± 0.08 µM), AiiA (6.88 ± 0.46 µg/ml), G + A 32 µg/ml (4.45 ± 0.15 nM). Calculation was performed using GraphPad Prism 6 software, taken at the time point between 4–6 hours where the inhibition started to occur. (E) Dose-pair effects of G1 and AiiA on lasB-gfp. The concentration of DMSO and buffer correspond to the highest test concentration used in the assays. All experiments were done in triplicate manner, only representative data are shown.

Figure 4

(A) Combination effects of G1 and AiiA on PAO1-pqsA-gfp reporter strain. (B) Dose-response effects of G1 (50 µM) added with AiiA (32 µg/ml and 16 µg/ml) on pqsA-gfp. (C) IC₅₀ values of the single and combination treatments on pqsA-gfp, G1 (3.08 ± 0.72 µM), AiiA (15.58 ± 0.17 µg/ml), G + A 32 µg/ml (0.63 ± 0.06 µM). Calculation was performed using GraphPad Prism 6 software, taken at the time point between 4–6 hours where the inhibition started to occur. Experiments were done in triplicate manner, only representative data are shown.

Figure 5

(A) Dose-dependent curves of QS deficient ΔlasIΔrhlI double mutant harboring lasB-gfp supplemented with 3-oxo-C12-HSL (AHL). (B) Dose-dependent curves of QS deficient ΔlasIΔrhlI double mutant harboring rhlA-gfp supplemented with C4-HSL (BHL). (C) Expression of the ΔlasIΔrhlI double mutant harboring lasB-gfp when G1 was added together with AHL. (D) Expression of the ΔlasIΔrhlI double mutant harboring rhlA-gfp when G1 was added together with BHL. Experiments were done in triplicate manner, only representative data are shown. Error bars are means ± SDs. ** = p < 0.01, *** = p < 0.001, **** = p < 0.0001, two-way ANOVA test.

Figure 6

Effects of QQ and QSI on the rhl system. (A) Combination effects of G1 and AiiA on PAO1-rhlA-gfp reporter strain. (B) Dose-response effects of G1 (50 µM) added with AiiA (32 µg/ml) on rhlA-gfp. (C) Effects on rhamnolipid production when tested at final concentration of 50 µM (for G1), and 32 µg/mL (for AiiA). Same amount of DMSO and buffer were used as positive control. PAO1 ΔlasIΔrhlI was used as negative control. Experiments were done in triplicate manner. Error bars are means ± SDs. *** = p < 0.001. **** = p < 0.0001, Student’s t test.