Impact of T2R38 receptor polymorphisms on Pseudomonas aeruginosa infection in cystic fibrosis

Andrew R. Turnbull1,2, Ronan Murphy1, Volker Behrends3, Helena Lund-Palau1, Ameze Simbo1, Myril

Mariveles4, Eric W.F.W. Alton1, Andrew Bush1,2, Amelia Shoemark1,2, Jane C. Davies1,2,

1National Heart & Lung Institute, Imperial College London, United Kingdom.

2Paediatric Respiratory Medicine, Royal Brompton & Harefield NHS Foundation Trust, London, UK

3Health Science Research Centre, Department of Life Sciences, University of Roehampton, UK

4Adult Cystic Fibrosis Centre, Royal Brompton & Harefield NHS Foundation Trust, London UK

Corresponding author:

Andrew R. Turnbull,

Population Health & Gene Therapy

National Heart & Lung Institute,

Imperial College,

London SW3 6LR,

a.turnbull14@imperial.ac.uk

Running title: T2R38 receptor polymorphisms in cystic fibrosis

Author contributions:

ART and JCD designed the study with input from AB, EWFWA and AS1,2. ART consented patients and

collected patient samples. AS1,2 provided expertise in immunocytochemistry and confocal

microscopy, and advised on data interpretation. VB advised and assisted with liquid chromatography

with tandem mass spectrometry. RM assisted with DNA extraction and genotyping. HLP, AS1 and

MM aided with clinical data acquisition and database support. ART analyzed the data and drafted

the manuscript with input from JCD, AS1,2 and AB. All authors contributed to manuscript drafts and

preparation. ART is custodian of the data and takes responsibility for its accuracy.

At a glance commentary

Scientific knowledge on the subject

Heterogeneity in cystic fibrosis (CF) lung disease is greater than can be explained by variability at the

CFTR locus, suggesting the involvement of modifier genes. Polymorphisms in the gene encoding the

T2R38 receptor (TAS2R38), expressed on airway epithelial cells, have been proposed as modifiers of

host response to Pseudomonas aeruginosa. Understanding the impact of TAS2R38 polymorphisms

on P. aeruginosa infection in CF could have important implications for patient risk stratification and,

as naturally-occurring and synthetic agonists to T2R38 are already in clinical use, could identify

promising therapeutic targets.

What this study adds to the field

We have shown that T2R38 is present in CF airway epithelium, where it localizes to the ciliary rootlet

in the same distribution as in non-CF cells. However, in a large study of children and adults with CF

we have found no association between TAS2R38 genotype and prevalence of intermittent or chronic

P. aeruginosa infection. Our findings suggest there to be no prognostic value of TAS2R38 genotyping

in patients with CF, and do not indicate the T2R38 receptor to be a promising drug target in CF

mucosal immunity.

Manuscript word count: 3,228

This article has an online data supplement, which is accessible from this issue's table of content

online at www.atsjournals.org

Abstract

Rationale

The T2R38 bitter taste receptor on respiratory epithelia detects P. aeruginosa N-acyl-L-homoserine

lactones (AHLs) . Activation leads to increased ciliary beat frequency and mucociliary clearance,

dependent on polymorphisms in the TAS2R38 gene. We hypothesized that TAS2R38 polymorphisms

would modify prevalence or impact of P. aeruginosa infection in patients with cystic fibrosis (CF).

Objectives

To characterise T2R38 airway localization and determine the relationship between TAS2R38

genotype and P. aeruginosa infection in CF.

Methods

T2R38 localization in respiratory epithelia was evaluated by immunocytochemistry. CF patients were

genotyped for TAS2R38 polymorphisms. Prevalence of intermittent or chronic P. aeruginosa

infection was analyzed by logistic regression. AHL production by P. aeruginosa isolates was

quantified with LC-MS/MS.

Measurements and Main Results

In CF nasal and bronchial epithelium T2R38 localized to the ciliary rootlet consistent with non-CF

cells. 225 patients had AVI/AVI (33%), AVI/PAV (49%) or PAV/PAV (18%) TAS2R38 genotypes and

well-defined P. aeruginosa infection status. Only age, but not TAS2R38 genotype, was associated

with P. aeruginosa infection status. Lung function was not different in P. aeruginosa-infected

patients of the differing genotypes. The proportion of AHL-deficient P. aeruginosa isolates did not

differ by TAS2R38 genotype.

Conclusions

TAS2R38 genotype does not influence prevalence or consequences of P. aeruginosa infection in CF

patients. This is not related to differences in T2R38 localization in the CF airway, or to predominance

of AHL-deficient P. aeruginosa strains. Our data indicate that the impact of TAS2R38 polymorphisms

on the host response to P. aeruginosa AHLs in CF is not clinically relevant.

Abstract word count: 249

Key words: cilia; taste receptor, type 2; quorum sensing; mucociliary clearance

Introduction

Cystic fibrosis (CF) pulmonary disease is characterized by recurrent bacterial infection, an

exaggerated host inflammatory response, and progressive lung function decline. Pseudomonas

aeruginosa is the most prevalent pathogen in CF and chronic infection is associated with accelerated

rates of disease progression (1, 2). Understanding host factors that influence susceptibility to chronic

P. aeruginosa is important to identify high-risk populations and could lead to the discovery of novel

drug targets.

The respiratory epithelium expresses many surface receptors, including bitter taste receptors (T2Rs)

(3, 4). These G-protein-coupled receptors, first identified in the tongue, putatively protect against

bitter toxin ingestion (5). In airway epithelial cells the T2R38 bitter taste receptor detects N-acyl-L-

homoserine lactones (AHLs) secreted by P. aeruginosa (6). AHLs are a major class of quorum sensing

molecules governing bacterial gene expression and virulence (7). The most abundant P. aeruginosa

AHLs, N-butanoyl-L-homoserine lactone (C4-HSL) and N-(3-oxododecanoyl)-L-homoserine lactone (3-

oxo-C12-HSL), have been detected in sputum from patients with CF (8, 9). In vitro, T2R38 activation

by AHLs initiates calcium-dependent increases in nitric oxide production and ciliary beat frequency,

both of which could enhance bacterial clearance (6).

The T2R38 receptor exhibits high allelic variation (10), associating with taste perception of the bitter

compound, phenylthiocarbamide (11). Three common single nucleotide polymorphisms in the

coding gene (TAS2R38) result in two haplotypes with amino acid substitutions at positions 49, 262,

and 296 in the receptor protein. The ‘taster’ haplotype codes for proline-alanine-valine (PAV) and

the ‘non-taster’ haplotype codes for alanine-valine-isoleucine (AVI). In Europeans these haplotypes

are approximately balanced at 49% (PAV) and 47% (AVI), the remainder being rare haplotypes (AAV,

AAI, PVI) (12). One group has reported that cellular responses to AHLs in vitro are determined by

TAS2R38 genotype, with cells homozygous for the PAV allele having greater responses than cells

with the AVI allele (6).

Clinical relevance of TAS2R38 genotype is proposed from studies in chronic rhinosinusitis (CRS).

Isolation of P. aeruginosa from patients undergoing CRS surgery was significantly less frequent in

patients with PAV/PAV compared to AVI/AVI genotypes, suggesting that the former offered some

protection (6). The same group reported that the PAV/PAV genotype was associated with a lower

likelihood of requiring CRS surgery and with improved postoperative outcomes (13, 14). In a small

study in patients with CF and sinus disease, CRS symptom scores were lower in adults with PAV/PAV

vs other genotypes (15).

The potential that T2R38-dependent pathways might modulate mucosal immunity represents an

attractive therapeutic target. As several compounds already in clinical use are T2R agonists (16) the

identification of T2R38-mediated pathways relevant to P. aeruginosa clearance could lead to rapid

therapeutic trials.

We hypothesized that TAS2R38 genotype would have the same protective effects against P.

aeruginosa in the CF lower airway as in the upper airway in CRS. We aimed to characterize T2R38

localization and to determine whether TAS2R38 genotype modifies prevalence of P. aeruginosa

infection in CF patients. Some of these results have previously been reported as abstracts (17, 18).

Methods

Subjects

T2R38 localization studies - Nasal and/or bronchial brushings were obtained from patients aged 6

years or older with a confirmed diagnosis of CF (19) during a clinically-indicated bronchoscopy under

general anaesthesia. Control nasal brushings were obtained from healthy adult volunteers.

TAS2R38 genotyping - Patients with CF aged 6 years or older were recruited and provided 3 ml of

whole blood for genomic DNA extraction.

T2R38 localization

Methods are described in the online supplement (OLS). All slides were double-labeled with

antibodies to T2R38 and ciliary proteins (Table 1). Slides were visualized under a Zeiss Axioskop

fluorescent microscope at x40 magnification to identify cells staining for ciliary proteins. T2R38

staining was assessed in the second channel and considered present if ≥7 out of 10 ciliated cells

stained positively for T2R38, as described previously (20). Representative slides were imaged with a

Zeiss LSM-510 inverted confocal microscope at x63 magnification. T2R38 colocalization was

quantified using the thresholded method of Manders (21) in the JACoP plug-in for ImageJ (22).

TAS2R38 genotyping

Methods are described in the OLS. Genomic DNA was extracted from blood and genotyped for the

three common single nucleotide polymorphisms in the TAS2R38 gene (rs713598, rs1726866, and

rs10246939).

Clinical and microbiological data

Data from each CF patient’s 2014 annual assessment was obtained from medical records. P.

aeruginosa infection status was determined by review of all respiratory cultures during 2014 and

assigned according to modified Leeds criteria (23) as chronic (>50% positive), intermittent (≤50%

positive), free (previous P. aeruginosa but none for >12 months), or never. Only patients with ≥3

cultures during 2014 were included. Spirometry values at annual assessment were expressed as

percentage of predicted values according to age-appropriate reference equations (24).

Quantification of P. aeruginosa quorum sensing profiles

Methods are described in the OLS. Cryo-preserved P. aeruginosa isolates from TAS2R38-genotyped

CF patients (matched for age and FEV1) were passaged through Luria-Bertani broth. After

measurement of optical density at 600 nm (OD600), culture supernatants were filter-sterilized.

Quantitative analysis of N-acyl-L-homoserine lactones (AHLs) and 2-alkyl-4-(1H)-quinolones (AQs)

was performed by liquid chromatography with tandem mass spectrometry (LC-MS/MS). Limits of

detection (LOD) and limits of quantification (LOQ) were defined as signal:noise ratios 3:1 and 10:1

respectively, as previously described (25).

Statistical analysis

Power calculations predicted 250 patients would provide 80% power to detect a difference in

chronic P. aeruginosa infection of ≥20% in the PAV/PAV group compared to other genotypes at α of

5%. Between-group comparisons were by Kruskal-Wallis with Dunn’s post-hoc test or by chi-square

test. Analysis of P. aeruginosa infection by TAS2R38 genotype was by Chi-squared analysis.

Additionally, a binomial logistic regression model was constructed with ‘chronic’ and ‘intermittent’

groups combined, and ‘never’ and ‘free’ groups combined as the dependent variables, with age, sex,

CFTR genotype and TAS2R38 genotype as independent variables. Graphpad Prism (Version 7.0) and

SPSS (version 23) were used for analyses. The null hypothesis was rejected at p<0.05.

Ethics

The protocol was approved by ethical review committees (02-019 and 10/H0504/9) and written

consent was obtained from subjects or their parent/guardian.

Results

T2R38 is localized at ciliary rootlets in healthy nasal epithelium

T2R38 Immunostaining was present in nasal epithelial cells from all (n=4) healthy subjects. Of the

three T2R38 antibodies tested, specific immunostaining was observed only with AB130503, which

was therefore used in all further immunocytochemistry studies. T2R38 stained proximally to

acetylated α-tubulin (staining ciliary microtubules) and γ-tubulin (staining ciliary basal bodies), and

colocalized with rootletin (staining ciliary rootlets) (figure 1). The thresholded Manders’ correlation

coefficient (tM1; mean ± SD of 4 cells) for T2R38 and rootletin staining was 0.91 ± 0.07, indicating

that 91% of green (rootletin) pixels were positive for red (T2R38).

T2R38 is localized at ciliary rootlets in CF nasal and bronchial epithelium

T2R38 was identified in all nasal epithelial samples from CF patients (n=3). Similar to non-CF cells,

T2R38 stained proximally to acetylated α-tubulin and γ-tubulin, and colocalized with rootletin (tM1

0.90 ± 0.08) (Figure 2). T2R38 was present in all bronchial epithelial samples from CF patients (n=3),

with an identical staining pattern to CF and non-CF nasal epithelium (Fig 2B), and colocalization with

rootletin (tM1 0.90 ± 0.04).

TAS2R38 genotype distribution and patient demographics

TAS2R38 genotypes were obtained for 271 patients with CF aged over 6 years, of which 243 (89.7%)

patients were homozygous or compound heterozygous for the common AVI or PAV haplotypes. Of

these, 225 (92.6%) had the predefined number of respiratory cultures during 2014 (≥3), enabling P.

aeruginosa infection category to be assigned. Between AVI/AVI, AVI/PAV and PAV/PAV TAS2R38

genotype groups there was no significant difference in median age, gender, proportion of

p.Phe508del CFTR mutations, or body mass index (table 2).

TAS2R38 genotype is not associated with P. aeruginosa infection status

By chi-square analysis there was no association between TAS2R38 genotype and P. aeruginosa

infection status (P=0.46) (table 3). In the binomial logistic regression model only age was associated

with the combined outcome of intermittent or chronic P. aeruginosa infection, but not sex, CFTR

genotype or TAS2R38 genotype (table 4). In multivariate analysis the results remained unchanged.

TAS2R38 genotype is not associated with lung function in patients with P. aeruginosa infection

Among patients who met criteria for intermittent or chronic P. aeruginosa infection during 2014

(n=141), there was no significant difference by TAS2R38 genotype in the median percent-predicted

FEV1 (AVI/AVI 54.0%, AVI/PAV 62.0%, PAV/PAV 53.5%, p=0.3) or FVC (AVI/AVI 80.0%, AVI/PAV

87.5%, PAV/PAV 87.5%, p=0.2) (Figure 3).

TAS2R38 genotype is not associated with mucoid P. aeruginosa isolation

Among all patients with intermittent or chronic P. aeruginosa infection (n=141), there was no

significant difference in the proportion of patients who isolated mucoid P. aeruginosa (AVI/AVI 69%,

AVI/PAV 60%, PAV/PAV 68%, p=0.5).

Analysis of quorum sensing signalling molecules in P. aeruginosa clinical isolates

In 18 P. aeruginosa clinical isolates from TAS2R38-genotyped CF patients there was no difference

between genotypes in the proportion of isolates in which C4-HSL or 3-oxo-C12-HSL were

unquantifiable (C4-HSL: AVI/AVI 67%, AVI/PAV 50%, PAV/PAV 50%; 3-oxo-C12-HSL: AVI/AVI 67%,

AVI/PAV 67%, PAV/PAV 50%).

Discussion

We have demonstrated that the T2R38 receptor is present in CF nasal and bronchial epithelium and

that there is no difference in receptor localization between CF and non-CF cells. In this cohort of

children and adults with CF, there was no significant impact of TAS2R38 genotype on the prevalence

of intermittent or chronic P. aeruginosa infection. Further, in those patients with current P.

aeruginosa infection, we observed no differences in lung function or prevalence of mucoid P.

aeruginosa by TAS2R38 genotype. Finally, in a small sample of clinical isolates we observed no

relationship between TAS2R38 genotype on quorum sensing profiles, suggesting that polymorphisms

in this receptor are not exerting a selective pressure on P. aeruginosa isolates in the CF airway.

In agreement with other researchers (4, 6) our immunocytochemistry results confirm T2R38

localization in ciliated nasal and bronchial epithelial cells. However, whereas previous studies have

shown T2R38 localization at the distal cilia (4) or extending from the tip to below the base of the cilia

(6), our data indicate T2R38 to be localized at the ciliary rootlet. These differences may in part be

attributable to differences in primary antibodies and in cell samples used; Shah and colleagues used

SC-67108 (Santa Cruz Biotechnology) and AB65509 (Abcam; no longer in production) and showed

T2R38 staining at the distal portion of the cilia in tracheal- and bronchial-derived ALI cultures(4). Lee

and colleagues used SC-67108 and demonstrated staining throughout and below the cilia in

sinonasal ALI cell cultures and excised sinonasal tissue (6, 26). In our studies SC-67108 showed

diffuse and non-specific staining throughout the cell whereas AB130503 (the currently available

T2R38 antibody from Abcam) showed specific staining at an intracellular location. SC-67108 is raised

against N-terminal amino acids 1-220 of the T2R38 protein and is predicted to bind to cytoplasmic,

helical and extracellular domains of the receptor (UniProt.org). AB130503 is raised against amino

acids 304-332 corresponding to a cytoplasmic domain at the C- terminus of the protein

(UniProt.org). In neutrophils AB130503 been shown only to stain permeabilised cells suggesting it is

specific to an intracytoplasmic receptor domain (27). Our results of sequential double-staining of

AB130503 with three different antibodies (acetylated -tubulin, -tubulin and rootletin) consistently

showed T2R38 localization at the ciliary rootlet. Rootletin is a structural component of the ciliary

rootlet, originating from the ciliary basal body and extending toward the nucleus (28). It provides an

anchoring role in both motile and non-motile cilia, and is suggested to have a role in ciliary protein

trafficking via intraflagellar transport (29). As expected, our immunocytochemistry results show

rootletin to be localized in the subciliary compartment, consistent with earlier studies (30, 31).

An additional difference in our localization studies is the use of primary epithelial cells obtained by

nasal and bronchial brushing as opposed to tissue explants or epithelial cell cultures. While this

could plausibly lead to differences in receptor localization by immunocytochemistry, identification

and localization of ciliary proteins in brushed epithelial cells has proven reliable in clinical diagnosis

of primary ciliary dyskinesia (20). Moreover, we propose that these samples obtained directly from

patients are less subject to variation that may occur during culture. We have previously

demonstrated the validity of our method for detection of receptors of the ciliary membrane through

staining of the polycystin mechanosensing membrane channel (32).

Despite the differences we have observed in cellular localization of T2R38, we confirm that

localization of T2R38 is the same in CF and non-CF nasal epithelial cells, and between CF nasal and

bronchial epithelium. This finding is in keeping with recent data on gene expression profiles in which

no differences were seen across a panel of T2R transcripts between immortalized bronchial

epithelial cells from a non-CF and CF subject (33).

In our study of P. aeruginosa infection status by TAS2R38 genotype, our cohort was representative

of the UK CF population and the distribution of TAS2R38 genotypes was consistent with a

predominantly European population. A further strength of our study was the classification of P.

aeruginosa infection based on complete respiratory culture results for a calendar year with exclusion

of patients who had fewer than the recommended number of samples. Our data on P. aeruginosa

infection categories are therefore as representative of the spectrum of infections stages as possible

in a clinical patient cohort. The absence of any association between TAS2R38 genotype and P.

aeruginosa infection was consistent by chi-square analysis and logistic regression. Further, our

results show only age to be associated with intermittent or chronic infection, consistent with data

from CF registry studies and serving as a useful positive control (2, 34). Among those patients who

met criteria for intermittent or chronic infection, the lack of any difference between genotype

groups in spirometry or prevalence of mucoid P. aeruginosa adds further evidence to the lack of a

protective effect of the PAV/PAV genotype.

P. aeruginosa infection in CF occurs early in childhood. Bronchoscopic surveillance studies in children

with CF have shown at least intermittent positive P. aeruginosa cultures in up to 53% of patients by 5

years of age (35, 36) and by adulthood, up to 60% of patients with CF are chronically infected. It is

plausible that T2R38-dependent responses to P. aeruginosa may be most relevant at the initial

stages of infection when inflammation in the CF airway is less marked than it is in advanced disease.

A limitation of our study is the inclusion of subjects only 6 years and older, with a median age of 29

years. It is possible that studying a younger group of patients may reveal differences in P. aeruginosa

acquisition by TAS2R38 genotype, but given the absence of any effect in this study, we hypothesize

that even were TAS2R38 differences apparent early, they become less relevant with age. Analyzing

age of first P. aeruginosa isolation would potentially address any difference in bacterial clearance in

the early stage of disease, but would require prospective data collection. A further limitation of our

study is sample size; it is possible that there are subtle differences in P. aeruginosa infection that are

smaller than those our study is powered to detect. Given the exploratory nature of this study we

determined a pragmatic sample size based on what we deemed to be a clinically relevant difference

in outcomes of P. aeruginosa infection, and we speculate that a difference smaller than this may be

of limited clinical relevance. Our results showing a consistent absence of any genotype-related effect

on infection prevalence or severity measures (spirometry and mucoid prevalence) leads us to infer

that an increase in sample size would not have changed our findings. Furthermore, differences in

sinonasal P. aeruginosa infection by TAS2R38 genotype in subjects with CRS had been detectable in a

substantially smaller clinical cohort of 56 patients (6).

The relevance of quorum sensing (QS) to the establishment of chronic P. aeruginosa in the CF airway

has been well described (8, 9, 37). Multiple virulence factors have been shown to be under QS

control by AHLs, including growth of P. aeruginosa in biofilms (38, 39) – microcolonies of sessile

bacteria encased in a self-produced polymeric matrix with enhanced resistance to antimicrobial

attack. However, while QS systems appear to be important in the early stages of CF airway infection,

mutations in QS-regulatory genes and reduced levels of AHLs have been observed in P. aeruginosa

isolates from patients with chronic infection (40, 41) suggesting these signals to be non-essential

once chronic infection is established. As AHLs are the ligands for T2R38-receptor activation in airway

epithelial cells, we considered it plausible that T2R38-mediated host response to P. aeruginosa could

exert selective pressure favoring AHL-deficient bacterial strains in chronic infection. To address this

possibility we quantified AHL concentrations in clinical P. aeruginosa isolates from TAS2R38-

genotyped patients, and although our sample size was small, we observed no trend towards AHL

non-producing isolates in patients with PAV/PAV genotypes.

The results of our study suggest the in vitro response of epithelial cells to P. aeruginosa AHLs do not

translate into meaningful differences in clinical outcomes in a cohort of patients with CF. There are

several possible reasons for this. Although other studies suggest that T2R38 is located at the ciliary

tip, our experiments clearly demonstrate that in fresh, non-cultured cells, T2R38 is in fact located

within an intracellular compartment in airway epithelial cells. This finding is consistent with studies

in pancreatic cells, where T2R38 has been localized to intracellular compartments (42). An

intracellular location of T2R38 does not preclude receptor binding by AHLs, which are highly

lipophilic, and which have been found to diffuse rapidly through mammalian cell membranes into

the cytoplasmic compartment (43). However, in the context of the abnormal airway surface liquid

and mucus of the CF airway, whether AHL diffusion through the cell membrane occurs is unknown.

Furthermore, evidence that in the CF airway P. aeruginosa can persist in organized biofilms

embedded within mucus and not in direct contact with the airway epithelium (44) may limit

exposure of AHLs to the epithelial surface.

Whether AHL binding to T2R38 occurs in the CF airway, the basic defect in CF is an important

potential reason for the absence of TAS2R38-dependent differences in P. aeruginosa infection. It is

broadly accepted that the reduction of functional CFTR at the epithelial surface results in reduced

chloride secretion via CFTR, in addition to unregulated sodium absorption via the epithelium

sodium channel (ENaC), leading to a low volume airway surface liquid and consequent

impairment of mucociliary clearance (MCC) (45). In the context of this secondary defect in MCC any

additional stimulus of CBF may be insufficient to exert a meaningful change in bacterial clearance.

The processes which enable P. aeruginosa to establish persistent and treatment-resistant infection

within the CF airway are complex. Given the prognostic importance of this organism in the

progression of CF lung disease the study of pathways that could reveal therapeutic targets or

identify patients at high risk of chronic infection is valuable. The discovery of bitter taste receptors in

the airway and their potential role in pathogen recognition and modulation of mucosal immunity has

prompted investigation of the role of these receptors in different airway diseases. Investigation of

this mechanism is particularly appealing given the existence of natural and synthetic T2R38 agonists,

which could lead to rapid therapeutic trials of repurposed drugs. Additionally, the high frequency of

TAS2R38 polmorphisms makes the study of this potential modifier gene feasible within a well-

defined patient population. Unfortunately however, our data indicate that TAS2R38 polymorphisms

do not appear to be clinically important modifiers of P. aeruginosa infection in patients with CF.

Acknowledgements

This project was supported by the NIHR Respiratory Disease Biomedical Research Unit at the Royal

Brompton and Harefield NHS Foundation Trust and Imperial College London. The P. aeruginosa

clinical isolate repository used in this study was established as part of the Strategic Research Centre

for Pseudomonal infection in CF, funded by the Cystic Fibrosis Trust (UK). The Facility for Imaging by

Light Microscopy (FILM) at Imperial College London is part-supported by funding from the Wellcome

Trust (grant 104931/Z/14/Z) and BBSRC (grant BB/L015129/1). We thank Neil Madden and the

microbiology department at the Royal Brompton Hospital for the processing of clinical samples. We

thank Prof. Miriam Moffat and Kenny Wong from the Molecular Genetics group at NHLI, Imperial

College London, for the use of equipment and technical assistance. We thank Stephen Rothery from

FILM for advice and technical support in confocal microscopy. We thank the Statistical Advisory

Service at Imperial College London and Dr Stephanie MacNeill for statistical advice. We also thank

Prof. Mike Cheetham at the Institute of Ophthalmology, University College London, for the gift of a

rootletin antibody.

References

1. Konstan MW, Morgan WJ, Butler SM, Pasta DJ, Craib ML, Silva SJ, et al. Risk factors

for rate of decline in forced expiratory volume in one second in children and adolescents

with cystic fibrosis. The Journal of pediatrics. 2007;151(2):134-9, 9.e1.

2. Kerem E, Viviani L, Zolin A, MacNeill S, Hatziagorou E, Ellemunter H, et al. Factors

associated with FEV1 decline in cystic fibrosis: analysis of the ECFS patient registry. The

European respiratory journal. 2014;43(1):125-33.

3. Finger TE, Bottger B, Hansen A, Anderson KT, Alimohammadi H, Silver WL. Solitary

chemoreceptor cells in the nasal cavity serve as sentinels of respiration. Proceedings of the

National Academy of Sciences of the United States of America. 2003;100(15):8981-6.

4. Shah AS, Ben-Shahar Y, Moninger TO, Kline JN, Welsh MJ. Motile cilia of human

airway epithelia are chemosensory. Science (New York, NY). 2009;325(5944):1131-4.

5. Mueller KL, Hoon MA, Erlenbach I, Chandrashekar J, Zuker CS, Ryba NJP. The

receptors and coding logic for bitter taste. Nature. 2005;434(7030):225-9.

6. Lee RJ, Xiong G, Kofonow JM, Chen B, Lysenko A, Jiang P, et al. T2R38 taste receptor

polymorphisms underlie susceptibility to upper respiratory infection. The Journal of clinical

investigation. 2012;122(11):4145-59.

7. Williams P, Camara M. Quorum sensing and environmental adaptation in

Pseudomonas aeruginosa: a tale of regulatory networks and multifunctional signal

molecules. Current opinion in microbiology. 2009;12(2):182-91.

8. Singh PK, Schaefer AL, Parsek MR, Moninger TO, Welsh MJ, Greenberg EP. Quorum-

sensing signals indicate that cystic fibrosis lungs are infected with bacterial biofilms. Nature.

2000;407(6805):762-4.

9. Erickson DL. Pseudomonas aeruginosa Quorum-Sensing Systems May Control

Virulence Factor Expression in the Lungs of Patients with Cystic Fibrosis. Infection and

Immunity. 2002;70(4):1783-90.

10. Kim U, Wooding S, Ricci D, Jorde LB, Drayna D. Worldwide haplotype diversity and

coding sequence variation at human bitter taste receptor loci. Human mutation.

2005;26(3):199-204.

11. Bufe B, Breslin PA, Kuhn C, Reed DR, Tharp CD, Slack JP, et al. The molecular basis of

individual differences in phenylthiocarbamide and propylthiouracil bitterness perception.

Current biology : CB. 2005;15(4):322-7.

12. Kim UK, Jorgenson E, Coon H, Leppert M, Risch N, Drayna D. Positional cloning of the

human quantitative trait locus underlying taste sensitivity to phenylthiocarbamide. Science

(New York, NY). 2003;299(5610):1221-5.

13. Adappa ND, Zhang Z, Palmer JN, Kennedy DW, Doghramji L, Lysenko A, et al. The

bitter taste receptor T2R38 is an independent risk factor for chronic rhinosinusitis requiring

sinus surgery. International forum of allergy & rhinology. 2014;4(1):3-7.

14. Adappa ND, Farquhar D, Palmer JN, Kennedy DW, Doghramji L, Morris SA, et al.

TAS2R38 genotype predicts surgical outcome in nonpolypoid chronic rhinosinusitis.

International forum of allergy & rhinology. 2016;6(1):25-33.

15. Adappa ND, Workman AD, Hadjiliadis D, Dorgan DJ, Frame D, Brooks S, et al. T2R38

genotype is correlated with sinonasal quality of life in homozygous DeltaF508 cystic fibrosis

patients. International forum of allergy & rhinology. 2016;6(4):356-61.

16. Meyerhof W, Batram C, Kuhn C, Brockhoff A, Chudoba E, Bufe B, et al. The molecular

receptive ranges of human TAS2R bitter taste receptors. Chemical senses. 2010;35(2):157-

70.

17. Turnbull A, Lund-Palau H, Murphy R, Simbo A, Wong K, Bush A, et al. The T2R38

bitter taste receptor as a modifier of host response to Pseudomonas aeruginosa in cystic

fibrosis: does T2R38 genotype impact on clinical infection? Pediatric pulmonology.

2016;51(S45):S81-S114.

18. Turnbull A, Shoemark A, Palau HL, Murphy RA, Simbo A, Bush A, et al. Determining

the impact of the T2R38 bitter taste receptor on P. aeruginosa infection in the cystic fibrosis

airway. Journal of Cystic Fibrosis. 2017;16:S88.

19. Farrell PM, Rosenstein BJ, White TB, Accurso FJ, Castellani C, Cutting GR, et al.

Guidelines for diagnosis of cystic fibrosis in newborns through older adults: Cystic Fibrosis

Foundation consensus report. The Journal of pediatrics. 2008;153(2):S4-s14.

20. Shoemark A, Frost E, Dixon M, Ollosson S, Kilpin K, Patel M, et al. Accuracy of

Immunofluorescence in the Diagnosis of Primary Ciliary Dyskinesia. American journal of

respiratory and critical care medicine. 2017.

21. Manders EMM, Verbeek FJ, Aten JA. Measurement of co-localization of objects in

dual-colour confocal images. Journal of microscopy. 1993;169(3):375-82.

22. Bolte S, Cordelieres FP. A guided tour into subcellular colocalization analysis in light

microscopy. J Microsc. 2006;224(Pt 3):213-32.

23. Lee TWR, Brownlee KG, Conway SP, Denton M, Littlewood JM. Evaluation of a new

definition for chronic Pseudomonas aeruginosa infection in cystic fibrosis patients. Journal

of Cystic Fibrosis. 2003;2(1):29-34.

24. Rosenthal M, Bain SH, Cramer D, Helms P, Denison D, Bush A, et al. Lung function in

white children aged 4 to 19 years: I--Spirometry. Thorax. 1993;48(8):794-802.

25. Ortori CA, Dubern JF, Chhabra SR, Camara M, Hardie K, Williams P, et al.

Simultaneous quantitative profiling of N-acyl-L-homoserine lactone and 2-alkyl-4(1H)-

quinolone families of quorum-sensing signaling molecules using LC-MS/MS. Analytical and

bioanalytical chemistry. 2011;399(2):839-50.

26. Yan CH, Hahn S, McMahon D, Bonislawski D, Kennedy DW, Adappa ND, et al. Nitric

oxide production is stimulated by bitter taste receptors ubiquitously expressed in the

sinonasal cavity. American journal of rhinology & allergy. 2017;31(2):85-92.

27. Maurer S, Wabnitz GH, Kahle NA, Stegmaier S, Prior B, Giese T, et al. Tasting

Pseudomonas aeruginosa Biofilms: Human Neutrophils Express the Bitter Receptor T2R38 as

Sensor for the Quorum Sensing Molecule N-(3-Oxododecanoyl)-l-Homoserine Lactone. Front

Immunol. 2015;6:369.

28. Yang J, Liu X, Yue G, Adamian M, Bulgakov O, Li T. Rootletin, a novel coiled-coil

protein, is a structural component of the ciliary rootlet. The Journal of cell biology.

2002;159(3):431-40.

29. Mohan S, Timbers TA, Kennedy J, Blacque OE, Leroux MR. Striated rootlet and

nonfilamentous forms of rootletin maintain ciliary function. Current biology : CB.

2013;23(20):2016-22.

30. Wallmeier J, Al-Mutairi DA, Chen CT, Loges NT, Pennekamp P, Menchen T, et al.

Mutations in CCNO result in congenital mucociliary clearance disorder with reduced

generation of multiple motile cilia. Nature genetics. 2014;46(6):646-51.

31. Akiyama T, Inoko A, Kaji Y, Yonemura S, Kakiguchi K, Segawa H, et al. SHG-specificity

of cellular Rootletin filaments enables naive imaging with universal conservation. Scientific

reports. 2017;7:39967.

32. Adams L, Loebinger M, Shoemark A. P239 A role for polycystins in airway mucociliary

clearance? Thorax. 2013;68(Suppl 3):A185.

33. Jaggupilli A, Singh N, Upadhyaya J, Sikarwar AS, Arakawa M, Dakshinamurti S, et al.

Analysis of the expression of human bitter taste receptors in extraoral tissues. Molecular

and cellular biochemistry. 2017;426(1-2):137-47.

34. UK Cystic Fibrosis Registry. Annual data report 2015: Cystic Fibrosis Trust; 2016

[Available from:

https://www.cysticfibrosis.org.uk/the-work-we-do/uk-cf-registry/reporting-and-resources.

35. Kidd TJ, Ramsay KA, Vidmar S, Carlin JB, Bell SC, Wainwright CE, et al. Pseudomonas

aeruginosa genotypes acquired by children with cystic fibrosis by age 5-years. Journal of

Cystic Fibrosis. 2015;14(3):361-9.

36. Douglas TA, Brennan S, Gard S, Berry L, Gangell C, Stick SM, et al. Acquisition and

eradication of P. aeruginosa in young children with cystic fibrosis. The European respiratory

journal. 2009;33(2):305-11.

37. Storey DG, Ujack EE, Rabin HR, Mitchell I. Pseudomonas aeruginosa lasR

transcription correlates with the transcription of lasA, lasB, and toxA in chronic lung

infections associated with cystic fibrosis. Infect Immun. 1998;66(6):2521-8.

38. Davies DG. The Involvement of Cell-to-Cell Signals in the Development of a Bacterial

Biofilm. Science (New York, NY). 1998;280(5361):295-8.

39. Whiteley M, Lee KM, Greenberg EP. Identification of genes controlled by quorum

sensing in Pseudomonas aeruginosa. Proceedings of the National Academy of Sciences of

the United States of America. 1999;96(24):13904-9.

40. Smith EE, Buckley DG, Wu Z, Saenphimmachak C, Hoffman LR, D'Argenio DA, et al.

Genetic adaptation by Pseudomonas aeruginosa to the airways of cystic fibrosis patients.

Proceedings of the National Academy of Sciences of the United States of America.

2006;103(22):8487-92.

41. Jiricny N, Molin S, Foster K, Diggle SP, Scanlan PD, Ghoul M, et al. Loss of social

behaviours in populations of Pseudomonas aeruginosa infecting lungs of patients with cystic

fibrosis. PloS one. 2014;9(1):e83124.

42. Gaida MM, Mayer C, Dapunt U, Stegmaier S, Schirmacher P, Wabnitz GH, et al.

Expression of the bitter receptor T2R38 in pancreatic cancer: localization in lipid droplets

and activation by a bacteria-derived quorum-sensing molecule. Oncotarget.

2016;7(11):12623-32.

43. Ritchie AJ, Whittall C, Lazenby JJ, Chhabra SR, Pritchard DI, Cooley MA. The

immunomodulatory Pseudomonas aeruginosa signalling molecule N-(3-oxododecanoyl)-L-

homoserine lactone enters mammalian cells in an unregulated fashion. Immunology and cell

biology. 2007;85(8):596-602.

44. Bjarnsholt T, Jensen PO, Fiandaca MJ, Pedersen J, Hansen CR, Andersen CB, et al.

Pseudomonas aeruginosa biofilms in the respiratory tract of cystic fibrosis patients. Pediatric

pulmonology. 2009;44(6):547-58.

45. Matsui H, Grubb BR, Tarran R, Randell SH, Gatzy JT, Davis CW, et al. Evidence for

periciliary liquid layer depletion, not abnormal ion composition, in the pathogenesis of cystic

fibrosis airways disease. Cell. 1998;95(7):1005-15.