Neuropeptide Y Reduces Nasal Epithelial T2R Bitter Taste Receptor–Stimulated Nitric Oxide Production

Bitter taste receptors (T2Rs) are G-protein-coupled receptors (GPCRs) expressed on the tongue but also in various locations throughout the body, including on motile cilia within the upper and lower airways. Within the nasal airway, T2Rs detect secreted bacterial ligands and initiate bactericidal nitric oxide (NO) responses, which also increase ciliary beat frequency (CBF) and mucociliary clearance of pathogens. Various neuropeptides, including neuropeptide tyrosine (neuropeptide Y or NPY), control physiological processes in the airway including cytokine release, fluid secretion, and ciliary beating. NPY levels and/or density of NPYergic neurons may be increased in some sinonasal diseases. We hypothesized that NPY modulates cilia-localized T2R responses in nasal epithelia. Using primary sinonasal epithelial cells cultured at air–liquid interface (ALI), we demonstrate that NPY reduces CBF through NPY2R activation of protein kinase C (PKC) and attenuates responses to T2R14 agonist apigenin. We find that NPY does not alter T2R-induced calcium elevation but does reduce T2R-stimulated NO production via a PKC-dependent process. This study extends our understanding of how T2R responses are modulated within the inflammatory environment of sinonasal diseases, which may improve our ability to effectively treat these disorders.

Antibiotic-resistant organisms are increasingly found in patients with CRS and other airway diseases [12][13][14][15][16][17][18][19][20]. An attractive alternative therapeutic approach to antibiotics is to stimulate receptors that control endogenous immune responses, thus enhancing innate defenses without increasing pressures for single-agent resistance. One pathway that emerged over the last decade is the bitter taste receptor (T2R) pathway. Initially identified on the tongue, T2Rs are G-protein-coupled receptors (GPCRs) also expressed in other organs, including the bronchial and nasal motile cilia [4,[21][22][23][24]. Several nasal cilia T2Rs detect bacterial products and activate local and rapid calcium-dependent nitric oxide (NO) synthase (NOS) activity, specifically via the eNOS isoform localized to the ingham, VT, USA). Each fluo-4 experiment was normalized to the baseline fluorescence at time = 0, as done previously [22,86,87] and is standard in the field with Fluo-4. Thus, normalized fluorescence was indicated by F/Fo. DAF-FM fluorescence is shown as raw increase after subtraction of baseline (time = 0), with care to keep loading and imaging parameters identical between all experiments.

Immunofluorescence
ALI cultures were fixed for 20 min in 4% formaldehyde at room temperature, followed by permeabilization and blocking for 1 h using PBS containing 5% normal donkey serum, 1% bovine serum albumin (BSA), 0.2% saponin, and 0.3% Triton X-100 at 4 • C. Primary antibody incubations (1:100 for anti-T2R antibodies, 1:250 for tubulin and eNOS antibodies) were performed at 4 • C overnight. Secondary antibody incubations with Alexa Fluor-labeled donkey anti-mouse or rabbit (1:1000) were performed at 4 • C for 2 h. Following primary and secondary antibody incubations, Transwell filters were cut from the plastic mounting ring with a razor blade and mounted onto glass slides with DAPI-containing Fluoroshield (Abcam, Cambridge, MA, USA). Zenon antibody labeling kits (Invitrogen/Molecular Probes/Thermo Scientific, Waltham, MA, USA) were used for direct labeling of primary antibodies. This was necessary to co-localize T2R38 and T2R14 as both antibodies were rabbit antibodies. Zenon labeling kits were used per the manufacturers' instructions. An Olympus Fluoview confocal system with IX-81 microscope and 60× (1.4 NA) objective was used to take images of ALIs. The 60× (1.4 NA oil) objective on an inverted Olympus IX-83 microscope with spinning disk (DSU) running Metamorph (Molecular Devices, San Jose, CA, USA) was used for imaging isolated ciliated cells. Analysis of images was performed using Fluoview software, Metamorph, and/or FIJI [88].

Statistical Analyses
Analyses were performed using Prism (GraphPad, San Diego, CA, USA) with p < 0.05 considered to be statistically significant. One-way analysis of variance (ANOVA) was performed with appropriate post-tests. For comparisons to a control group, one-way ANOVA with Dunnett's posttest was used. Bonferroni posttest was used for preselected pair-wise comparisons. Tukey-Kramer posttest was used for comparisons of all samples within a data set. All additional data analyses were performed in Excel. All figures are displayed with mean ± SD; one asterisk (*) indicates p < 0.05. Two asterisks (**) indicate p < 0.01.
NPY (1 µM) was used in many experiments here as a saturating concentration to test effects of maximal stimulation of NPYRs. The effects of NPY were blocked by NPY2R antagonist BIIE-0246 (1 µM) or PKC inhibitor Gö6983 (1 µM) ( Figure 1E,F). Results are summarized in Figure 1G. This confirms our previous observations that NPYR2 activation reduces CBF via PKC activation [78]. PKC is one of the major negative regulators of CBF [89].

T2R38, T2R14, and eNOS Have Different Patterns of Cilia Localization
As described above, PKC also negatively regulates eNOS, likely the major NOS isoform in healthy airway ciliated cells [90]. In isolated ciliated cells brushed from the middle turbinate tissue, we observed cilia localization of both T2R14 and T2R38 (Figure 2A,B) and some ciliary but mostly sub-ciliary apical localization of eNOS ( Figure 2C,D). This confirms previous observations that eNOS is localized to the base of airway cilia [25,28]

T2R38, T2R14, and eNOS Have Different Patterns of Cilia Localization
As described above, PKC also negatively regulates eNOS, likely the major NOS isoform in healthy airway ciliated cells [90]. In isolated ciliated cells brushed from the middle turbinate tissue, we observed cilia localization of both T2R14 and T2R38 (Figure 2A,B) and some ciliary but mostly sub-ciliary apical localization of eNOS ( Figure 2C,D). This confirms previous observations that eNOS is localized to the base of airway cilia [25,28].

NPY Attenuates CBF Response to T2R14 Agonists via NPY2R Receptors
We tested T2R ciliary beat responses in nasal ALI cultures, which also express cilialocalized T2R14 and T2R38 ( Figure 3A and [23,24]). The flavone apigenin is a T2R14 and T2R39 agonist that increases nasal epithelial cell CBF in a manner dependent on NO [23]. Apigenin (100 µM) increased CBF ~15% after 8 min, while vehicle control (0.1% DMSO) had minimal effect ( Figure 3B,C). CBF increases in response to 100 µM apigenin were blocked by T2R14 and T2R39 antagonist [91] 4′-fluoro-6-methoxyflavanone (50 µM; Figure  3D,E). CBF responses to apigenin were inhibited with co-stimulation with 1 µM NPY (Figure 3D,E). This did not occur in the presence of 1 µM BIIE-0246 ( Figure 3D,E). Together, these data suggest NPY negatively regulates T2R-mediated CBF responses, which are NO dependent as they are blocked by NOS inhibitor L-NAME (10 µM; 30 min pre-treatment) Figure 2. Bitter taste receptors T2R38 and T2R14 are localized to motile cilia while endothelial nitric oxide synthase (eNOS) is primarily localized at the base of cilia. Isolated ciliated cells brushed from middle turbinate tissue were stuck to glass using CellTak and fixed as described in the text and stained for β tubulin IV (βTubIV) and T2R14 (A), T2R38 (B), eNOS (C), or rabbit IgG ((D) isotype control for nonspecific antibody binding). All images were taken at identical microscope settings (exposure, LED power, binning, etc.) with 60× 1.4 NA oil objective.
homozygous for the functional (PAV) TAS2R38 allele [93], while CBF did not increase in cultures homozygous for the non-functional (AVI) TAS2R38 allele ( Figure 3G,H). We tested co-stimulation with a lower concentration of NPY (10 nM) than that used above. We found that NPY reduced the CBF response to 3oxoC12HSL in PAV/PAV cultures (Figure 3G,H). Thus, NPY can inhibit T2R-activated CBF increases.

NPY Does Not Alter T2R-Induced Calcium Elevation but Does Reduce T2R-Stimulated NO Production
Because T2R-stimulated NO production via eNOS is driven by calcium signaling [22][23][24], one potential explanation for the above observations is that NPY reduces the T2Rstimulated calcium response. To test this, we imaged T2R-induced calcium responses in ALIs loaded with fluorescent calcium-sensitive dye Fluo-4. Both P. aeruginosa [22] and apigenin [23] activate calcium responses in primary sinonasal cells grown at ALI. We confirmed that 100 µM 3oxoC12HSL and 100 µM apigenin activated calcium responses in ALIs derived from patients homozygous for functional (PAV) T2R38, while the 0.1% We also observed that Pseudomonas aeruginosa quorum-sensing molecule 3-oxododecanoylhomoserine lactone (3oxoC12HSL; 100 µM), also a T2R agonist [22,92], increased CBF in a T2R38-dependent manner as previously reported [22]. CBF increased in cultures homozygous for the functional (PAV) TAS2R38 allele [93], while CBF did not increase in cultures homozygous for the non-functional (AVI) TAS2R38 allele ( Figure 3G,H). We tested co-stimulation with a lower concentration of NPY (10 nM) than that used above. We found that NPY reduced the CBF response to 3oxoC12HSL in PAV/PAV cultures ( Figure 3G,H). Thus, NPY can inhibit T2R-activated CBF increases.

NPY Does Not Alter T2R-Induced Calcium Elevation but Does Reduce T2R-Stimulated NO Production
Because T2R-stimulated NO production via eNOS is driven by calcium signaling [22][23][24], one potential explanation for the above observations is that NPY reduces the T2R-stimulated calcium response. To test this, we imaged T2R-induced calcium responses in ALIs loaded with fluorescent calcium-sensitive dye Fluo-4. Both P. aeruginosa [22] and apigenin [23] acti- vate calcium responses in primary sinonasal cells grown at ALI. We confirmed that 100 µM 3oxoC12HSL and 100 µM apigenin activated calcium responses in ALIs derived from patients homozygous for functional (PAV) T2R38, while the 0.1% DMSO vehicle control solution had no effect on calcium ( Figure 4A). As previously reported [22], we found that 3oxoC12HSL activated calcium responses in PAV/PAV ALIs but not in ALIs derived from patients homozygous for non-functional (AVI) T2R38 ( Figure 4B). The calcium response in PAV/PAV ALIs was not reduced in the presence of 1 µM NPY ( Figure 4B), which is likely a saturating, supraphysiological, maximally activating concentration of NPY. Likewise, the calcium response to T2R14 agonist apigenin (100 µM) was not reduced by 1 µM NPY but was reduced by T2R14 antagonist 4 -fluoro-6-methoxyflavanone (50 µM; Figure 4C). Thus, the reduced CBF response during NPY co-stimulation was not likely due to reduced calcium signaling.
We also found that 50 nM of NPY was sufficient to reduce the NO produced during application of calcium ionophore ionomycin combined with calcium ATPase inhibitor thapsigargin (1 µg/mL each; Figure 5J). Because these compounds will elevate calcium independent of GPCR signaling, the inhibition of NO production here supports that the inhibitory effect of NPY was likely downstream of calcium.

NPY Effects on T2R-Mediated NO and CBF Responses Are Dependent on PKC
Data above show that NPY stimulation reduces T2R-mediated NO responses without a change in upstream calcium signaling. To find whether this was due to protein kinase C activation, We imaged DAF-FM-loaded PAV/PAV ALIs stimulated with 3oxoC12HSL (100 µM). Stimulated NO production was reduced in the presence of 1 µM NPY but was not reduced in the presence of NPY plus either of two PKC inhibitors, 1 µM Gö6983 or 1 µM calphostin C ( Figure 6A). Apigenin (100 µM)-induced NO production was likewise inhibited by 1 µM NPY but restored in the presence of 1 µM Gö6983 ( Figure 6B). Apigenininduced CBF increase was likewise reduced by 1 µM NPY but unaffected in the presence of 1 µM Go6983 or calphostin C ( Figure 6C).

Discussion
We show an inhibitory effect of NPY on T2R-mediated NO production and CBF responses. This likely occurs through activation of NPY2R receptors and subsequent PKC activation. We previously showed that activation of PKC downstream of fungal aflatoxins can reduce T2R-mediated NO production [78]. We hypothesize this is because of eNOS T495 phosphorylation by PKC, which impairs calcium-bound calmodulin binding to eNOS [94]. NO increases cilia beating and mucociliary clearance through activation of guanylyl cyclase and protein kinase G [89]. Thus, elevated NPY in CRS or asthma may impact cilia function, perhaps contributing to susceptibility to infection by reducing mucociliary clearance. While T2R-induced changes in CBF are somewhat small (10-20%), the relationship between CBF and mucociliary clearance rate (as measured in fluorescent particle transport assays) is not linear. We previously showed that these T2R-induced 10-20% changes in CBF equate to ~50-80% increases in mucociliary transport [22,29]. Thus, the blunting of T2R responses by NPY likely has even greater effects on mucociliary clearance than the changes in CBF observed here.
We observe effects of NPY on T2R-induced NO responses as low as 1 nM. The pKd of recombinant NPY2R vs I125-NPY was reported to be ~10.17 (~0.068 nM) [95] and EC50 of ~0.3 nM in a heterologous expression system [96]. The apparent EC50 for NPY is likely higher than the binding of NPY to its receptor due to downstream nonlinearity. Thus, while many experiments in this study and other studies (e.g., [97,98]) test high saturating concentrations of NPY (e.g., 100 nM-1 µM) to maximally activate receptors, we also see

Discussion
We show an inhibitory effect of NPY on T2R-mediated NO production and CBF responses. This likely occurs through activation of NPY2R receptors and subsequent PKC activation. We previously showed that activation of PKC downstream of fungal aflatoxins can reduce T2R-mediated NO production [78]. We hypothesize this is because of eNOS T495 phosphorylation by PKC, which impairs calcium-bound calmodulin binding to eNOS [94]. NO increases cilia beating and mucociliary clearance through activation of guanylyl cyclase and protein kinase G [89]. Thus, elevated NPY in CRS or asthma may impact cilia function, perhaps contributing to susceptibility to infection by reducing mucociliary clearance. While T2R-induced changes in CBF are somewhat small (10-20%), the relationship between CBF and mucociliary clearance rate (as measured in fluorescent particle transport assays) is not linear. We previously showed that these T2R-induced 10-20% changes in CBF equate tõ 50-80% increases in mucociliary transport [22,29]. Thus, the blunting of T2R responses by NPY likely has even greater effects on mucociliary clearance than the changes in CBF observed here.
We observe effects of NPY on T2R-induced NO responses as low as 1 nM. The pKd of recombinant NPY2R vs I125-NPY was reported to be~10.17 (~0.068 nM) [95] and EC 50 of~0.3 nM in a heterologous expression system [96]. The apparent EC 50 for NPY is likely higher than the binding of NPY to its receptor due to downstream nonlinearity. Thus, while many experiments in this study and other studies (e.g., [97,98]) test high saturating concentrations of NPY (e.g., 100 nM-1 µM) to maximally activate receptors, we also see that lower, likely more physiological concentrations are also sufficient to reduce T2R responses. Thus, while in vivo and other further experimentation is needed to understand the potential role of this mechanism in sinonasal disease, our data suggest that the inhibition of T2Rs reported here may be activated by physiologically relevant levels of NPY that activate NPY2R. This is particularly true in cells in the nose in close proximity to NPYergic fibers. Local release of neurotransmitters in the airway could generate high nM to µM concentrations within close proximity to the neuron, albeit likely for short periods of time [99].
Notably, we show that NPY blunts responses to P. aeruginosa 3oxoC12HSL and PQS, detected largely by T2R38 [22] and T2R4 [24], respectively. We also show blunted responses to apigenin, which we previously concluded was mediated by T2R14 [23]. However, apigenin and other types of flavones also activate T2R39 [100][101][102]. The 4 -fluoro-6methoxyflavanone antagonist used here also blocks T2R39 [91]. We have not detected T2R39 in differentiated nasal cilia [23,24], but it is expressed in bronchial cilia [21]. We thus cannot fully rule out a potential role for T2R39 in the apigenin response. Nonetheless, this does not change any of the conclusions of this paper. NPY still blunts the T2R-mediated response to apigenin, whether through T2R14 alone or a combination of T2R14 and T2R39. We did not test agonists for T2R16, which is also in nasal cilia. As we have shown that all of these T2Rs signal identically [4,24], we believe it is likely that T2R16 responses are likewise reduced by NPY.
Beyond ciliary beating, NO is also important to airway immunity because it directly kills or inactivates pathogens. NO damages the DNA and cell walls of bacteria [103][104][105][106][107]. Replication of many respiratory viruses is also NO sensitive, including influenza, parainfluenza, rhinovirus [108], and SARS-COV1 and 2 [109][110][111][112]. Reduced NO output by the nasal epithelium with NPY elevations may contribute to susceptibility to both bacterial and viral infection. Plant flavonoids such as the flavones apigenin and chrysin are attractive molecules to target nasal T2Rs because they also have intrinsic antibacterial properties [23,113]. However, there ability to activate T2R14 may be limited in patients with elevated NPY levels.
As noted above, NPY may be elevated in asthma, which does carry a greater risk of respiratory infection [114][115][116]. As T2Rs are also expressed in bronchial cilia [21], impairment of T2R function by NPY in asthma may also contribute to innate immune dysfunction in this disease. Future studies of bronchial epithelial cells are needed to clarify the role of NO in the bronchial epithelial T2R CBF response. PKC inhibitors have been suggested as therapeutic modalities for several types of chronic inflammatory diseases [117] and might be useful in chronic inflammatory airway diseases where NPY is elevated.
In mice, NPY may be critical in type 2 inflammatory responses in the airways [118][119][120]. However, while some studies have demonstrated elevated human sinonasal NPY expression in different types of sinonasal diseases [62][63][64][65]121,122], a greater study of changes in NPY levels and localization in the nose in specific CRS patient subsets is needed to better understand whether and how NPY fits into the context of CRS pathophysiology. In particular, future measurement of NPY levels, localization of NPY production, and elucidation of potential changes in normal vs. diseased inflamed sinonasal tissue are all needed. While sensory neurons and macrophages are a potential source of NPY production, other immune cells such as neutrophils or eosinophils might also produce NPY and alter epithelial responses to bitter bacterial or fungal molecules. The T2R-to-eNOS signaling pathway also functions in macrophages, where it regulates phagocytosis [123], and macrophageproduced NPY may feed back onto this pathway in an autocrine signaling loop. An overall better understanding of neuropeptides in both upper and lower airway inflammation is needed to understand whether targeting these pathways can reduce inflammation and/or boost innate immunity.
We previously showed that NPY is a negative regulator of nasal turbinate submucosal gland secretion via the activation of NPY1R receptors [75]. Nasal and bronchial submucosal glands are important for the secretion of not only fluid for airway surface liquid and mucus but also antimicrobial peptides [76,124]. Reducing both glandular antimicrobial peptide secretion and antimicrobial NO production suggests that NPY has multiple negative effects on sinonasal innate immunity. However, future studies are needed to clarify what relevance the in vitro observations here have to in vivo sinonasal pathophysiology. It may be that the function of NPY as a "brake" on ciliated cell NO production and CBF is somehow beneficial. Regardless, the studies here demonstrate that NPY is a potent regulator of epithelial T2R NO, warranting further investigation into the role of NPY in sinonasal innate immunity.
NPYRs can also be activated by pancreatic polypeptide (PP) and peptide YY (PYY). While PYY is generally thought to be only produced by PP cells (also known as gamma or F cells) in pancreatic islets, some PYY is found in gastrointestinal epithelial neurons [60]. To our knowledge, studies of PP and PYY expression in the nasal cavity are extremely limited [125]. Future studies must explore these other NPY family polypeptides. Future studies are also needed to determine whether and how other neuropeptides sch as substance P and VIP may also affect T2R signaling in nasal epithelial ciliated cells.
Author Contributions: R.M.C.: conceptualization, methodology, investigation, data acquisition, formal analysis, and writing; N.D.A. and J.N.P.: resources and critical review; R.J.L.: conceptualization, methodology, investigation, data acquisition, formal analysis, writing, and funding acquisition. All authors have read and agreed to the published version of the manuscript.

Funding:
The study was supported by R01DC016309 to R.J.L.

Institutional Review Board Statement:
The study was conducted in accordance with code of federal regulation Title 45 CFR 46.116 by the U.S. Department of Health and Human Services and was approved by the University of Pennsylvania Institutional Review Board (protocol #800614).

Informed Consent Statement:
Informed consent was obtained from all subjects involved in the study.
Data Availability Statement: All data supporting the results are contained within the article figures. Raw numerical values used to generate the bar graphs or traces are available upon request to corresponding author.