Cilia Stimulatory and Antibacterial Activities of T2R Bitter Taste Receptor Agonist Diphenhydramine: Insights into Repurposing Bitter Drugs for Nasal Infections

Abstract

T2R bitter taste receptors in airway motile cilia increase ciliary beat frequency (CBF) and nitric oxide (NO) production. Polymorphisms in some T2Rs are linked to disease outcomes in chronic rhinosinusitis (CRS) and cystic fibrosis (CF). We examined the expression of cilia T2Rs during the differentiation of human nasal epithelial cells grown at air-liquid interface (ALI). The T2R expression increased with differentiation but did not vary between CF and non-CF cultures. Treatment with Pseudomonas aeruginosa flagellin decreased the expression of diphenhydramine-responsive T2R14 and 40, among others. Diphenhydramine increased both NO production, measured by fluorescent dye DAF-FM, and CBF, measured via high-speed imaging. Increases in CBF were disrupted after flagellin treatment. Diphenhydramine impaired the growth of lab and clinical strains of P. aeruginosa, a major pathogen in CF and CF-related CRS. Diphenhydramine impaired biofilm formation of P. aeruginosa, measured via crystal violet staining, as well as the surface attachment of P. aeruginosa to CF airway epithelial cells, measured using colony-forming unit counting. Because the T2R agonist diphenhydramine increases NO production and CBF while also decreasing bacterial growth and biofilm production, diphenhydramine-derived compounds may have potential clinical usefulness in CF-related CRS as a topical therapy. However, utilizing T2R agonists as therapeutics within the context of P. aeruginosa infection may require co-treatment with anti-inflammatories to enhance T2R expression.

Keywords: G protein-coupled receptors; Pseudomonas aeruginosa; Staphylococcus aureus; calcium; chronic rhinosinusitis; ciliary beat frequency; cystic fibrosis; nitric oxide.

Figure 1

Generation of air–liquid interface (ALI) cultures from residual surgical material. (A) Nasal turbinate mucosal tissue was removed from the bone, embedded in agarose, sliced and fixed. Immunofluorescence staining for β-tubulin IV (green) shows the highly ciliated mucosal surface. Nuclei are shown in blue. Representative tissue slices from three patients are shown. Spinning disk confocal, 10 × 0.4 NA objective. Scale bar is 50 µm. (B) non-fixed or embedded tissue is enzymatically dissociated and basal epithelial cells are propagated in submersion culture. Cells are seeded onto permeable Transwell filters at high density, and the next day, the medium on the apical surface is removed. The resulting air exposure on the apical side leads to the differentiation of cells into ciliated and goblet cells reminiscent of the in vivo epithelium; (C) representative immunofluorescence of ALI cultures derived from three different patients after three weeks of differentiation, showing an abundance of ciliated cells (β-tubulin IV; green). AlexaFluor 647-labeled phalloidin staining of actin (magenta) shows tight cobblestone epithelial pattern and apical microvilli. Nuclear DAPI stain shown in blue. Images taken on laser-scanning confocal, 60×. Images are maximum z-projections. Scale bar is 50 µm.

Figure 2

T2R expression in primary nasal air–liquid interface (ALI) cultures. (A) T2R expression increases during differentiation of nasal ALIs. Ciliary T2Rs, 4, 10, 14, 16 and 38 mRNA expression increased with at least 3 weeks of differentiation, as revealed by qPCR. Note that T2R receptors are encoded by corresponding TAS2R genes. Data points represent the average of 5 patients; significance was analyzed by one-way ANOVA using Dunnett’s post-test (* p < 0.05). (B) Immunofluorescence of MS4A8B (Mouse monoclonal antibody 3E6, Kerafast, Boston, MA) and β-tubulin IV (Rabbit monoclonal EPR16775, Abcam) showing cilia co-localization. Representative images shown from 3 ALIs from 3 individual patients. (C) Cystic fibrosis does not alter T2R mRNA expression. Cilia-localized T2Rs 4, 10, 14, 16 and 38 did not significantly alter the expression when comparing nasal ALIs derived from either non-CF or CF patients. Cilia marker MS4A8B and MUC5AC expression demonstrates that cultures were differentiated. Bar graphs represent data obtained from 5 non-CF and 3 CF patients. Data pairs were analyzed by Student’s t-test showing no significant differences.

Figure 3

Changes in T2R expression with inflammatory mediators. (A) Flagellin decreases T2R expression in nasal ALIs. Human nasal epithelial cells were isolated and grown in ALI cultures and then exposed to air to induce differentiation for 4 weeks prior to treatment. Cultures were treated with various concentrations of flagellin (0.1, 1 or 10 µg/mL) for 24 h prior to RNA analysis via qPCR. Bar graphs represent combined data from 4 patients. Data were analyzed by one-way ANOVA, Dunnett’s post-test; * p < 0.05. (B) IL-13 treatment decreases T2R expression in nasal ALIs. Human nasal epithelial cells were isolated and grown in ALI cultures then exposed to air to induce differentiation for 3 weeks prior to treatment. Data were analyzed by one-way ANOVA, Bonferroni post-test; * p < 0.05 and ** p < 0.01. Bar graph represents combined data from 4 patients. (C) Expression of T2Rs in 3-week post-air primary nasal ALIs normalized to housekeeping gene UBC using 2^−ΔCt method where ΔCt = Ct_GAPDH − Ct_{target gene}. Data shown are the mean ± SEM from ALIs grown from 15 different individual patients. TAS2R14 expression was significantly higher than other TAS2R genes examined by one-way ANOVA with Bonferroni post-test; * p < 0.05.

Figure 4

Diphenhydramine induces nitric oxide (NO) production in nasal ALIs. (A) T2R14 localizes to cilia in nasal ALIs. Representative immunofluorescence image of nasal cilia (β-tubulin IV; green) and T2R14 (magenta) from ALI cultures from two individual donors. No staining was observed with rabbit serum plus secondary control (bottom). (B) NO production, as revealed by DAF-FM fluorescence, increased in a dose-dependent manner with diphenhydramine (DPD). (C) Anti-histamines ranitidine and cetirizine did not initiate NO production, while equimolar diphenhydramine did. (D) NOS inhibitor L-NAME (10 µM) blocked NO production while negative control D-NAME (10 µM) did not. (E) Flagellin pre-treatment resulted in less NO production in response to DPD. Representative traces from ≥3 experiments using ALIs from different non-CF patients are shown. Bar graphs are mean ± SEM analyzed via one-way ANOVA (B–D) or Student’s t test (E); ** p < 0.01.

Figure 5

Flagellin represses DPD enhancement of ciliary beat frequency. (A) T2R14 agonist diphenhydramine (100 µM) increases CBF by up to 1 Hz. (B) Fully differentiated (day 21) nasal ALI cultures were treated with media containing 0.1 µg/mL flagellin for 72 h to ensure reduction in both RNA and protein expression. Flagellin-treated cultures did not reveal a 1 Hz increase in CBF when treated with diphenhydramine. Traces are from 1 patient representative of results from ≥3 patients. Bar graphs are combined data from a ≥3 ALIs from ≥3 individual patients. * p < 0.05.

Figure 6

Diphenhydramine represses the planktonic growth of S. aureus and P. aeruginosa. (A) S. aureus strain M2 growth (measured by OD₆₀₀) was reduced by 1–5 mM diphenhydramine. (B) DPD (5 mM) significantly decreased the growth of Pseudomonas lab strain PAO-1. (C–F) The growth of clinical strains of Pseudomonas 2338 (C), L3847 (D), P11006 (E) and 1580 (F) were all impaired by 5 mM DPD. (G) Bar graph summarizing results from (C–F). Traces are representative of ≥3 experiments with mean ± SEM in bar graphs analyzed via one-way ANOVA using Sidak’s post-test (* p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001).

Figure 7

Diphenhydramine (DPD) inhibits P. aeruginosa biofilm production and attachment to CFBE cells. (A,B) Representative images of crystal violet staining showing diphenhydramine-reduced biofilm mass at 5 mM for lab strain PAO1 (A) and 1 mM for clinical strain P11006 (B). Other clinical strains used in Figure 4 did not form biofilms on 96-well plates and so could not be used in this assay. Representative images of the 96-well plate are shown. Data from three experiments are combined in bar graphs analyzed by one-way ANOVA using Dunnett’s post-test (* p < 0.05, *** p < 0.001). (C) Representative images of CFU counts following CFBE attachment assays as described in the text. DPD (500 µM) reduced the amount of bacteria recovered from CFBE cells while equimolar cetirizine did not. Control cells stimulated with media only. Lab strains PAO-1, ATCC27853 and PA-14 and clinical strain P11006 were used. (D) Quantification of results from independent experiments as in C. Significance by one-way ANOVA with Bonferroni post-test comparing all values to respective control; * p < 0.5, ** p < 0.01. (E) Representative phase contrast images of CFBE ALIs incubated in HBSS ± 500 µM DPD (1 h; 37 °C). No signs of epithelial break down were observed. Representative of 6 ALIs per condition from independent experiments. (F) Transepithelial resistance was measured ±DPD stimulation as in E. No significant difference was observed by one-way ANOVA.