Naringenin Inhibition of the Pseudomonas aeruginosa Quorum Sensing Response Is Based on Its Time-Dependent Competition With N-(3-Oxo-dodecanoyl)-L-homoserine Lactone for LasR Binding

Abstract

Bacterial quorum sensing (QS) is a cell-to-cell communication system that governs the expression of a large set of genes involved in bacterial-host interactions, including the production of virulence factors. Conversely, the hosts can produce anti-QS compounds to impair virulence of bacterial pathogens. One of these inhibitors is the plant flavonoid naringenin, which impairs the production of QS-regulated Pseudomonas aeruginosa virulence factors. In the present work, we analyze the molecular basis for such inhibition. Our data indicate that naringenin produces its effect by directly binding the QS regulator LasR, hence competing with its physiological activator, N-(3-oxo-dodecanoyl)-L-homoserine lactone (3OC12-HSL). The in vitro analysis of LasR binding to its cognate target DNA showed that the capacity of naringenin to outcompete 3OC12-HSL, when the latter is previously bound to LasR, is low. By using an E. coli LasR-based biosensor strain, which does not produce 3OC12-HSL, we determined that the inhibition of LasR is more efficient when naringenin binds to nascent LasR than when this regulator is already activated through 3OC12-HSL binding. According to these findings, at early exponential growth phase, when the amount of 3OC12-HSL is low, naringenin should proficiently inhibit the P. aeruginosa QS response, whereas at later stages of growth, once 3OC12-HSL concentration reaches a threshold enough for binding LasR, naringenin would not efficiently inhibit the QS response. To test this hypothesis, we analyze the potential effect of naringenin over the QS response by adding naringenin to P. aeruginosa cultures at either time zero (early inhibition) or at stationary growth phase (late inhibition). In early inhibitory conditions, naringenin inhibited the expression of QS-regulated genes, as well as the production of the QS-regulated virulence factors, pyocyanin and elastase. Nevertheless, in late inhibitory conditions, the P. aeruginosa QS response was not inhibited by naringenin. Therefore, this time-dependent inhibition may compromise the efficiency of this flavonoid, which will be effective just when used against bacterial populations presenting low cellular densities, and highlight the importance of searching for QS inhibitors whose mechanism of action does not depend on the QS status of the population.

Keywords: LasR; Pseudomonas aeruginosa; naringenin; quorum sensing; virulence inhibitor.

FIGURE 1

LasR DNA-binding capacity requires 3OC12-HSL during protein expression. E. coli SHA011 LasR producing strain was grown in the presence or absence of 20 μM 3OC12-HSL. As a control, E. coli SHA010, LasR non-producing strain was grown in same conditions. Protein extracts were tested for their capacity to bind to the lasI* probe by EMSA assays in the presence of 10 μM 3OC12-HSL. As shown, the presence of 3OC12-HSL in the culture is essential for purifying an active form of LasR (able to bind to the lasI* probe) from the E. coli SHA011 LasR producing strain. A non-specific retarded band was also observed in extracts obtained either from the control strain that does not produce LasR (SHA010) or from the SHA011 LasR-producing strain.

FIGURE 2

Docking of naringenin and 3OC12-HSL to LasR. The cartoon representation shows the best binding conformation for the analyzed ligands, naringenin (orange) and 3OC12-HSL (green), in the ligand binding pocket of the protein. In yellow, the aminoacid residues Tyr56, Trp60 and Asp73, which interact with the polar homoserine lactone head group of the autoinducer and with the hydrocarbon chain of the autoinducer (Tyr47), are shown. The prediction model supports that naringenin may be accommodated in the binding pocket where 3OC12-HSL binds, suggesting a possible competition of both molecules for LasR binding.

FIGURE 3

Partial inhibition of LasR binding activity by naringenin. The EMSA assays were carried out using a purified active protein obtained from SHA011 strain grown in the presence of 3OC12-HSL. The capacity of naringenin to displace 3OC12-HSL from LasR was analyzed by adding increasing concentrations of the flavonoid (10, 50, 500 μM) to the binding reaction. The figure shows that neither 3OC12-HSL (10 μM) nor naringenin (10 μM) addition to the binding buffer alters the LasR-DNA binding complex. However, increased concentrations of naringenin (50 and 500 μM) in the binding buffer were able to partially inhibit the LasR-DNA interaction, even in the presence of 3OC12-HSL (10 μM), evidencing that the flavonoid may partially displace the 3OC12-HSL from LasR.

FIGURE 4

Analysis of naringenin and 3OC12-HSL competition for the binding of LasR using a biosensor strain of E. coli. The E. coli JM109 (pSB1142) LasR-based reporter strain was used to analyze the capacity of naringenin and 3OC12-HSL to displace each other when the first molecule has already bound the LasR regulator. The luminescence produced by E. coli JM109 (pSB1142) was recorded and it is represented as relative light units normalized by growth (OD₆₀₀). 3OC12-HSL or naringenin were added in combination or separately, at different times. No differences in growth were observed between the analyzed conditions. As shown, the highest or the lowest luminescence were emitted in presence of 1 μM of 3OC12-HSL or 1 μM naringenin, respectively, when were independently added at time zero (t = 0 h). Moreover, intermediate luminescence was observed when the two compounds were added at time zero, demonstrating that there is a competition between them for LasR binding. Addition of 3OC12-HSL after 2 h of incubation with naringenin (added at time = 0 h) was unable to increase LasR activity, while addition of naringenin after 2 h of incubation with 3OC12-HSL (added at time = 0 h) was enough to partially decrease the LasR-dependent luminescence emitted by E. coli JM109 (pSB1142). Altogether, these results suggest a more efficient displacement of 3OC12-HSL by naringenin compared with the displacement of naringenin by 3OC12-HSL. Error bars represent standard error of three independent replicates.

FIGURE 5

Analysis of early and late inhibition of the expression of the QS-regulated genes and virulence factors by naringenin in P. aeruginosa PAO1. (A) The expression of the lasA and lasB QS-regulated genes was analyzed by real time PCR in cells treated with naringenin at t = 0 h (early inhibition) or at t = 4 h (late inhibition) after 4 h of growth. Results are represented as the fold change in the level of expression compared to same strain grown in presence of the solvent of naringenin, DMSO. The results show that inhibition of LasR-dependent regulation by naringenin is more efficient when this flavonoid is added at t = 0 h. Contrary, there is no inhibition by naringenin of lasA and lasB expression when this flavonoid was added 4 h after incubation (early stationary phase), when 3OC12-HSL production by P. aeruginosa is high. (B) Elastase activity and pyocyanin production were determined at the late stationary phase (OD₆₀₀ = 4.4; t = 8 h) in cells treated, at t = 0 h (early inhibition) or at t = 4 h (late inhibition), with naringenin or its solvent, DMSO. These graphs show that inhibition of LasR-dependent regulation and QS-dependent production of virulence factors by naringenin is more efficient when this flavonoid is added at t = 0 h, a cellular status where the production of 3OC12-HSL is practically zero. Contrary, there is no inhibition by naringenin of QS-regulated genes or of virulence factors production when this flavonoid was added 4 h after incubation (early stationary phase), when 3OC12-HSL production by P. aeruginosa is high and naringenin cannot displace the AI from LasR. In both figures, error bars represent standard error of three independent replicates.