Plant flavones enhance antimicrobial activity of respiratory epithelial cell secretions against Pseudomonas aeruginosa

Abstract

Flavones are a class of natural plant secondary metabolites that have anti-inflammatory and anti-bacterial effects. Some flavones also activate the T2R14 bitter taste receptor, which is expressed in motile cilia of the sinonasal epithelium and activates innate immune nitric oxide (NO) production. Flavones may thus be potential therapeutics for respiratory infections. Our objective was to examine the anti-microbial effects of flavones on the common sinonasal pathogens Candida albicans, Staphylococcus aureus, and Pseudomonas aeruginosa, evaluating both planktonic and biofilm growth. Flavones had only very low-level antibacterial activity alone. They did not reduce biofilm formation, but did reduce production of the important P. aeruginosa inflammatory mediator and ciliotoxin pyocyanin. However, flavones exhibited synergy against P. aeruginosa in the presence of antibiotics or recombinant human lysozyme. They also enhanced the efficacy of antimicrobials secreted by cultured and primary human airway cells grown at air-liquid interface. This suggests that flavones may have anti-gram-negative potential as topical therapeutics when combined with antibiotics or in the context of innate antimicrobials secreted by the respiratory or other epithelia. This may have an additive effect when combined with T2R14-activated NO production. Additional studies are necessary to understand which flavone compounds or mixtures are the most efficacious.

Fig 1. Flavone backbone structure and specific compounds used in this study.

Fig 2. Synergistic anti-bacterial effects of flavones in combination with antibiotics.

(A) Planktonic growth traces (OD₆₀₀) of 2 strains of P. aeruginosa (PAO1 and ATCC 27853) under the indicated conditions. Note reduction of OD₆₀₀ in the presence of penicillin/streptomycin plus flavone mixture (apigenin, chrysin, wogonin; 100 μM each). (B) Bar graphs showing ΔOD₆₀₀ over 5.25 hrs from A (n = 4 experiments for each condition). Asterisks denote significance vs. control (LB only; one-way ANOVA, Dunnett’s post-test; * = p <0.05, ** = p <0.01); ## indicates p <0.01 between bracketed bars (one-way ANOVA, Bonferonni post-test).

Fig 3. Synergistic anti-bacterial effects of flavones in combination with the airway antimicrobial protein lysozyme.

(A) Traces of planktonic growth of P. aeruginosa in the presence of flavones ± lysozyme. Note the greatest OD₆₀₀ decrease (bacterial lysis) occurred with lysozyme and flavone mix combined. (B) Bar graphs of the initial OD decrease rate (OD₆₀₀ units/min) from A (n = 3–6 experiments for each condition). (C) Bar graphs showing ΔOD₆₀₀ after 2 hours from A. For B and C, asterisks denote significance vs. control (LB only) by one-way ANOVA, Dunnett’s post-test (* = p <0.05, ** = p <0.01); # and ## indicates p <0.05 and 0.01, respectively, between bracketed bars (one-way ANOVA, Bonferonni post-test).

Fig 4. Confirmation of bacterial lysis by measurement of GFP release from PAO-GFP.

Bar graph showing normalized fluorescence (control = 1) from PAO-GFP cultures incubated in lysozyme lysis buffer with addition of lysozyme and/or flavones as indicated and described in the text. Flavone mix contained 100 μM each apigenin, chrysin, and wogonin. Apigenin and chrysin were used alone at 300 μM to compare an equal number of moles of flavone molecules. Synergistic effects of the flavone mixture were observed both alone and combined with lysozyme. Significance determined by one-way ANOVA with Bonferonni post-test; # and ## indicate p <0.05 and p <0.01, respectively, compared with control; * indicates p <0.05 between bracketed groups.

Fig 5. Confirmation of bacterial cell wall damage by NPN uptake.

Bacteria were incubated for 5 min with lysozyme ± flavones as indicated and described in the text, followed by measurement of NPN fluorescence, reflecting uptake of NPN into the bacterial phospholipid membrane. Data are expressed as fold increase in NPN fluorescence. Raw fluorescence values are in S1 File. Significance determined by one-way ANOVA with Bonferonni post-test; # and ## indicate p <0.05 and p <0.01, respectively, compared with control; ** indicates p <0.01 between bracketed groups.

Fig 6. Antimicrobial effects of Calu-3 and primary cell airway surface liquid (ASL) are enhanced by flavones.

(A) Calu-3 air-liquid interface cultures (ALIs) recapitulate a polarized secretory epithelium with polarized secretion of antimicrobial peptides and mucus, similar in composition to that of airway submucosal exocrine gland serous acinar cells [65, 72, 89]. (B) Bar graph shows number of colony forming units (CFUs) recovered from bacteria mixed with dilutions of Calu-3 ASL washings. As negative control, bacteria were incubated with 25% PBS (first column) not in contact with Calu-3 cells; 50 μg/mL gentamicin in 25% PBS was used as positive control. Antimicrobial activity was enhanced at lower dilutions of Calu-3 ASL (12.5% and 6.25%) in the presence of the flavone mix (50 μM each apigenin, chrysin, and wogonin). (C) Primary sinonasal epithelial cultures recapitulate the surface airway epithelium, with differentiated ciliated, goblet, and solitary chemosensory cells, likewise with polarized secretion of antimicrobial peptides and mucus. (D) Bar graph shows CFUs when P. aeruginosa were mixed with ASL washings from primary sinonasal ALI cultures stimulated with denatonium benzoate (10 mM). Asterisks denote significance determined by one-way ANOVA, Bonferroni post-test of paired columns (each condition ± flavone; * p <0.05 and ** p <0.01).

Fig 7. Multiple mechanisms of modulation of respiratory epithelial innate immunity by flavones.

(A) Lysozyme is primarily secreted by serous cells of airway submucosal exocrine glands [72]. Defensins are secreted by surface epithelial cells as well as glands. Here, we show that flavones increase the efficacy of these and possibly other secreted antimicrobial peptides (AMPs). (B) We showed previously that flavones also activate the bitter taste receptor T2R14, expressed in both sinonasal [32] and bronchial cilia [99]. T2R14 activation in sinonasal cilia increases nitric oxide synthase (NOS)-mediated production of NO, which increases ciliary beating through protein kinase G (PKG) to promote bacterial clearance and directly diffuses into the airway surface liquid to kill bacteria [53].