Transient receptor potential ankyrin-1 [transient receptor potential cation channel subfamily A member 1 (TRPA1)] and vanilloid-1 [transient receptor potential cation channel subfamily V member 1 (TRPV1)] detect inhaled irritants, including air pollutants and have roles in the development and exacerbation of asthma.
This study tested the hypothesis that increased expression of TRPA1, stemming from expression of the loss-of-function TRPV1 (I585V; rs8065080) polymorphic variant by airway epithelial cells may explain prior observations of worse asthma symptom control among children with the TRPV1 I585I/V genotype, by virtue of sensitizing epithelial cells to particulate materials and other TRPA1 agonists.
TRP agonists, antagonists, small interfering RNA (siRNA), a nuclear factor kappa light chain enhancer of activated B cells (NF-) pathway inhibitor, and kinase activators and inhibitors were used to modulate TRPA1 and TRPV1 expression and function. Treatment of genotyped airway epithelial cells with particulate materials and analysis of asthma control data were used to assess consequences of TRPV1 genotype and variable TRPA1 expression on cellular responses in vitro and asthma symptom control among children as a function of voluntarily reported tobacco smoke exposure.
A relationship between higher TRPA1 expression and function and lower TRPV1 expression and function was revealed. Findings of this study pointed to a mechanism whereby NF- promoted TRPA1 expression, whereas NF-–regulated nucleotide-binding oligomerization domain, leucine rich repeat and pyrin domain containing 2 (NLRP2) limited expression. Roles for protein kinase C and p38 mitogen activated protein kinase were also demonstrated. Finally, the TRPV1 I585I/V genotype was associated with increased TRPA1 expression by primary airway epithelial cells and amplified responses to selected air pollution particles in vitro. However, the TRPV1 I585I/V genotype was not associated with worse asthma symptom control among children exposed to tobacco smoke, whereas other TRPA1 and TRPV1 variants were.
This study provides insights on how airway epithelial cells regulate TRPA1 expression, how TRPV1 genetics can affect TRPA1 expression, and that TRPA1 and TRPV1 polymorphisms differentially affect asthma symptom control. https://doi.org/10.1289/EHP11076
Transient receptor potential ankyrin-1 [transient receptor potential cation channel subfamily A member 1 (TRPA1)] and vanilloid-1 [transient receptor potential cation channel subfamily V member 1 (TRPV1)] are cation-permeable (primarily calcium) channels involved in sensory physiology and pain.1,2 These receptors are also proximal sensors of selected inhaled chemicals and particulate materials (PMs), and coordinate responses of lung cells exposed to environmental irritants and pneumotoxins.3–6
Asthma is a chronic airway inflammatory disease that is affected by exposure to environmental irritants and pollutants, including volatile/semivolatile chemicals and PM released during the combustion of various materials (e.g., petroleum products/fuels, coal, biomass/biomass-based materials, cigarettes/E-cigarettes).7–13 TRPA1 and TRPV1 are important in the pathogenesis and exacerbation of asthma,1,2,14–17 and the expression and activity of TRPA118 and TRPV119,20 can be elevated in the airways of people with asthma and chronic cough. Studies in mice have demonstrated a need for TRPA1 in the development of Th2-high allergic asthma phenotypes,21 with similar findings reported for TRPV1.22,23 Several representative PMs, including diesel exhaust particles (DEP),4,5 wood smoke particles (WSPM),6,24,25 and coal fly ash (CFA),3,15,26 activate TRPA1 and TRPV1. Moreover, in airway epithelial cells (AECs), TRP activation by PM can trigger pro-inflammatory responses and pathological endoplasmic reticulum stress, leading to cell death. In addition, TRPA1 activation by WSPM can promote mucin 5AC (MUC5AC) expression and secretion by AECs, which could contribute to asthma exacerbation and symptomatology.24
Previous studies by our group demonstrated that the predominantly co-inherited TRPA1 gain-of-function polymorphisms R3C and R58T (rs13268757 and rs16937976) correlated with reduced likelihood (; ) of achieving optimal asthma symptom control in children.26 Studies have also shown the TRPV1 I585V (rs8065080) variant to be less responsive to canonical TRPV1 agonists and stimuli27 and to be associated with reduced cough in people with asthma.27,28 However, our research has shown that the TRPV1 I585I/V genotype occurs at a higher frequency in children with moderate-to-severe steroid-resistant asthma, specifically correlating with more frequent asthma symptoms (e.g., cough/wheeze, nighttime awakenings, inhaler use, care needs; odds ; ) and decreased overall asthma control when compared with those with either the I585I/I or I585V/V genotypes or combinations of two additional common TRPV1 polymorphisms (i.e., I315M; rs222747 and T469I; rs224534).15 A basis for this clinical association was hypothesized to result from higher TRPA1 expression by AECs of individuals with the I585I/V genotype, potentially increasing the sensitivity of such individuals to common asthma triggers that activate TRPA1.
Currently, neither how TRPA1 expression is regulated by AECs, nor the basis for how the TRPV1 I585I/V genotype promotes TRPA1 expression are understood. Moreover, the potential impact of elevated TRPA1 expression on asthma is not fully understood. Goals of this work were to a) understand how TRPA1 expression is regulated and specifically promoted by the TRPV1 I585I/V genotype in AECs, b) determine the importance of elevated TRPA1 expression in regulating responses of AECs to pneumotoxic PM treatment, and c) evaluate the contribution of the I585I/V genotype and TRPA1 expression on asthma symptom control as a function of tobacco smoke exposure among children with asthma.
Materials and Methods
Chemicals and Other Materials
-(4-tert-butylbenzyl)--(1-[3-fluoro-4-(methylsulfonylamino)phenyl]ethyl)thiourea (LJO-328) was provided by J. Lee of Seoul National University. The structure has been published.29 2-Mercaptoethanol, -vanillylnonanamide (nonivamide; a capsaicin analog), allyl isothiocyanate (AITC), ionomycin calcium salt, and phorbol 12-myristate 13-acetate (PMA) were from Sigma-Aldrich. Dimethyl sulfoxide (DMSO) was from Fisher Scientific. BMS-345541 and Go6983 were from Tocris, and recombinant human tumor necrosis factor-alpha (), interleukin-1 alpha (), , IL6, and IL13 were from Peprotech. 4-(4-Fluorophenyl)-2-(4-nitrophenyl)-5-(4-pyridyl)-1H-imidazole (PD169316) was from Cayman Chemical.
WSPM, DEP, and CFA
Preparation of pine WSPM,6 and the sources and properties of the DEP4,5 and CFA3 have been described. Briefly, WSPM was prepared by burning of Austrian pine (from a tree growing in the Salt Lake Valley; ) using a pipe furnace at 750°C with constant air flow. WSPM was collected using an Anderson cascade impactor operated at , and fractions 6 and 7 ( and ) were used. For experiments, WSPM concentrate was suspended in DMSO at and diluted to in media containing DMSO to achieve a area dose in a single well of a 6-well plate. Features of pine WSPM, and the effects that it has on AECs (i.e., calcium flux, pro-inflammatory, cytostatic, and cytotoxic) have been described.24,25,30,31 Specifically, the material was shown to contain TRPA1 agonists, including resin acids, perinaphthenone, coniferaldehyde, ethyl phenols, and substituted xylenols; ethyl phenols and xylenols also activate TRPV3.6,30 The DEP was collected from idling diesel-powered vehicles during emissions testing in the Salt Lake City area.4 The PM was suspended in media containing DMSO and applied to cells at a final area dose of . The DEP used in this study was shown to contain the TRPA1 agonists 2,4-di-tert-butylphenol (also a TRPV3 agonist) and several quinones (benzo and naptho), as well as perinanpthenone.4 Finally, CFA was collected from the Hunter power plant in Castle Dale, Utah, and fractionated to . The ash was from low-sulfur bituminous coal, which was previously reported to consist of primarily insoluble oxides and salts of silicon, calcium, aluminum, and iron, with elemental carbon and unspecified organic carbon32 (presumably polycyclic aromatic hydrocarbons as described for other CFA samples33). Effects of CFA on AECs and TRP channels have also been described.3,26,34
BEAS-2B and HEK-293 cells were from ATCC. These and other cells were maintained in a humidified cell culture incubator at 37°C with a 95% air:5% carbon dioxide () atmosphere. Human TRPV1-overexpressing HEK-293 and BEAS-2B cells were generated as previously described.15,35 Briefly, cells were transfected with a pcDNA3.1 plasmid (ThermoFisher) harboring human TRPV1, selected using Geneticin (), and expanded from a single colony. Overexpression was verified using calcium flux assays and western blots. TRPV1-overexpressing HEK-293 cells were cultured in Dulbecco’s Modified Eagle Medium/Ham’s F12 (DMEM:F12) media containing 5% fetal bovine serum, penicillin/streptomycin, and Geneticin (ThermoFisher), and BEAS-2B cells were cultured in LHC-9 medium fortified with Geneticin. Normal (NHBE; CC-2540 and CC-2541) and diseased (DHBE; CC-2540 and 00194911) human bronchial epithelial cells (HBEs) were from Lonza and were cultured in bronchial epithelial cell growth medium (Lonza). NHBEs immortalized with cyclin-dependent kinase 4 (CDK4) and telomerase reverse transcriptase (hTERT), or HBEC3-KT cells, were from ATCC (CRL-4051) and were grown in airway epithelial cell basal medium supplemented with bronchial epithelial cell growth kit, Geneticin, and puromycin (ATCC). Human small airway epithelial cells (SAECs; CC-2547) were from Lonza and were grown in BronchiaLife epithelial airway medium complete kit.
NHBE and DHBE cells were genotyped using TaqMan Genotyping Master Mix and assays for TRPV1 I585V (ThermoFisher #C_11679656_10) I315M (ThermoFisher #C_1093688_10), T469I (ThermoFisher #C_1093674), TRPA1 R3C (ThermoFisher #C_2175739_10), and R58T (ThermoFisher #C_25646603_10). Genomic DNA was isolated from cells using the GeneElute Mammalian Genomic DNA Miniprep Kit (Sigma-Aldrich). DNA was then quantified by ultraviolet (UV) absorbance on a Nanodrop Onec (ThermoFisher) and assayed using a Life Technologies QuantStudio 6 Flex instrument (ThermoFisher) and the polymerase chain reaction (PCR) program specified by the supplier for the TaqMan Genotyping Master Mix. Genotypes were differentiated by relative probe fluorophore intensity using pcDNA3.1 plasmids harboring human TRPV1, human TRPA1, and the target single nucleotide polymorphism (SNP) variant mutations.15,26 Genotyping results and donor lot identifications for the cells used in this work are provided in Table S1. HBEC3-KT cells have the TRPV1 I585I/V genotype, BEAS-2B cells the I585I/I genotype, SAECs the I585I/V genotype, and NHBEs have variable genotypes.
Quantitative Real-Time PCR
Cells were plated at in 6-well plates and treated at confluence (3 d post plating with feeding on day 2). After treatment, total RNA was isolated using the PureLink RNA Mini Kit (Invitrogen). Total RNA () was quantified by UV absorbance on a Nanodrop Onec (ThermoFisher) and complementary DNA (cDNA) was synthesized using the ABI High-Capacity cDNA Synthesis Kit with RNase inhibitor (Applied Biosystems). The cDNA was then subjected to analysis by quantitative real-time PCR (q-PCR) using TaqMan Gene Expression Master Mix (ThermoFisher) and a Life Technologies QuantStudio 6 Flex instrument. The following TaqMan probe-based assays were used: human glyceraldehyde 3-phosphate dehydrogenase (GAPDH; Hs99999905_m1), human TRPA1 (Hs00175798_m1), human TRPV1 (Hs00218912_m1), human TRPV3 (Hs00376854_m1), human NLRP2 (Hs01546932_m1), human IGFBP2 (Hs01040719_m1), human DNA damage-inducible transcript-3 (DDIT3; Hs_00358796_g1), and human interleukin-8 (IL8; Hs00174103_m1). The PCR programs used were according to the supplier for the TaqMan Gene Expression Master Mix and probe assays. mRNA expression was normalized to the housekeeping gene, human (; Hs00984230_m1), and the average value of control samples (i.e., the comparative method)36 with relative quantification (Rq) reported.
Small Interfering RNA Studies
Small interfering RNA (siRNA) transfections were performed according to the ThermoFisher Stealth/siRNA Transfection Lipofectamine 2000 Protocol. HBEC3-KT cells were plated in 6-well plates at and transfected with siRNA at 30%–50% confluence (2 d post plating), as per the manufacturer protocol. Briefly, siRNAs were reconstituted to a concentration of in nuclease-free water. An aliquot containing 100 or negative control (Silencer Negative Control siRNA No.1; ThermoFisher #AM4611), positive control GAPDH (ThermoFisher #4404024), NLRP2 siRNA-1 or -2 (ThermoFisher siRNA ID 25505; AM16708), or TRPV1 siRNA (ThermoFisher siRNA ID 105495; AM16708) was diluted in antibiotic and serum-free HBEC3-KT media and incubated in the dark at room temperature. After 5 min, the siRNA was mixed 1:1 with the diluted Lipofectamine 2000 and incubated for 25 min in the dark at room temperature. After incubation, the siRNA:Lipofectamine complex was diluted with of antibiotic-free and serum-free HBEC3-KT media to a final volume of . This solution was then added to an individual well in a 6-well plate and incubated for 6 h, upon which the transfection solution was replaced with fresh HBEC3-KT media. Isolation and quantification of mRNA after siRNA transfection was performed as described above, 24 h posttransfection. For western blotting, cells were plated at in flasks and the siRNA volumes and concentrations were adjusted according to the supplier protocol.
Western blots were performed as previously described by Nguyen et al.25 For isolating nuclear protein, the NE-PER Nuclear and Cytoplasmic Extraction Reagents kit and protocol (ThermoFisher) was used. Cells were grown to confluence in flasks. Total protein was harvested on ice using radioimmunoprecipitation buffer, supplemented with 6 M urea, 1% sodium dodecyl sulfate, and Halt protease inhibitor. Lysates were sonicated on ice using -s pulses at 100 W, repeated 10 times and clarified by centrifugation at for 15 min at 4°C. Protein concentrations were determined using the bicinchoninic acid protein assay kit (ThermoFisher), and was loaded into each well of a 4%–12% Bolt Bis-Tris 12-well gel and resolved by electrophoresis for 1.5 h at 120 V. The Precision Plus Protein Dual Color Standards () ladder was also used (BioRad). Following electrophoresis, the proteins were transferred to a polyvinylidene fluoride membrane using the iBlot 2 gel transfer device. After transfer, the membrane was incubated in SuperBlock (ThermoFisher) for 1 h at room temperature. Primary rabbit monoclonal antibodies against nuclear factor kappa light chain enhancer of activated B cells (NF-)/p65 (#3034; Cell Signaling Technology) and NF--phospho-S536 (#3033; Cell Signaling Technology) were used at 1:5,000 dilution. Polyclonal antibodies against NLRP2 (ThermoFisher #15182-1-AP) and rabbit nuclear matrix protein (p84; ThermoFisher, PA5-69083) were used at 1:1,000 dilution. A mouse monoclonal antibody against (8H10D10; Cell Signaling Technology #3700) was used at 1:10,000 dilution. Rabbit GAPDH (D16H11) monoclonal antibody (Cell Signaling Technology #5174) was used at 1:1000 dilution. All primary antibodies were prepared in 5% bovine serum albumin with 0.1% sodium azide and incubated with the membranes at 4°C for 16 h. Horseradish peroxidase-conjugated sheep-antimouse (NA931) and anti-rabbit (NA934) secondary antibodies (GE Health Sciences) were used at 1:10,000 in SuperBlock and were also applied for 16 h at 4°C. SuperSignal West Dura Extended Duration Substrate (ThermoFisher) was added to the membrane and visualized using a FluorChem M imager with the chemiluminescence plus markers setting. Bands were quantified using densitometry analysis of 8-bit images in ImageJ37 and normalized to the respective gene, as outlined in the figure legends.
RNA sequencing was performed as previously described15 at the High Throughput Genomics Core Facility at the Huntsman Cancer Institute, University of Utah. NHBE cells from the four donors (14359 and 14664 for TRPV1 I585I/I, and 9853 and unknown for TRPV1 I585I/V) were used. Total RNA was extracted using the RNeasy Mini kit (Qiagen) with on-column DNase digestion. RNA quality was assessed by RNA nanochip technology, and library construction was performed using the Illumina TruSeq Stranded mRNA Sample Preparation kit using established protocols. The sequencing libraries (18 pM) were then chemically denatured and applied to an Illumina TruSeq version 3 single-read flow cell using an Illumina cBot. Hybridized molecules were clonally amplified and annealed to sequencing primers with reagents from an Illumina TruSeq SR Cluster kit, version 3-cBot-HS. Following transfer of the flow cell to an Illumina HiSeq instrument, a 50-cycle single-read sequence run was performed using TruSeq SBS version 3 sequencing reagents. Data were processed at the University of Utah Bioinformatics core. The data presented in this publication have been deposited in the National Center for Biotechnology Information’s (NCBI’s) Gene Expression Omnibus (GEO) and are accessible through GEO Series accession no. GSE85447 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE85447).
Calcium Flux/TRP Activity Assays
Calcium flux in HBEC3-KT cells was measured using the Fluo-4 Direct assay kit and imaging on an EVOS FL Auto microscope at magnification using a green fluorescent protein (GFP) filter.3–5,25,26,30 Briefly, cells were plated in a flat-bottom 96-well plate at , fed on day 2 post plating, and assayed at 80%–90% confluence 3 d post plating. Pretreatment of cells with nonivamide (), LJO-328 (), pine WSPM (), (), and BMS-345541 () occurred 12 or 24 h before the assay. Prior to the assay, the cells were loaded with Fluo-4 diluted 1:1 in LHC-9 at 37°C for 1 h. After 1 h, the Fluo-4 was removed and the cells were washed with LHC-9 containing probenecid and trypan red (ATT Bioquest). The assays were performed in an on-stage environmental chamber maintained at 37°C in a 95% air:5% atmosphere. Agonist treatments (AITC 25 or ) were added to cells at the desired final concentration in LHC-9, which contained calcium. Images were captured every 6 s for 72 s. Changes in fluorescence were quantified using a custom MATLAB program, as previously described.3,5 Reported values are from the 60-s time point and were corrected by subtracting the fluorescence response to a blank media control (i.e., no agonist) and then normalized to the response value at 72 s following ionomycin () treatment applied after the 60-s image was taken.
Calcium flux in HEK-293 cells stably overexpressing TRPV1 (HEKV1OE cells) was measured essentially as above using a BMG Labtech NOVOStar fluorescence plate reader. The following differences applied: HEKV1OE cells were plated in flat-bottom 96-well plates coated with 1% gelatin at cells/well and assayed at 100% confluence 1–2 d post plating. Pretreatments (12 h) included Go6983 () or media containing DMSO (control). Changes in fluorescence were determined using the NOVOStar analysis software (MARS version 2.41). The values were normalized to the response elicited by LHC-9 media and are reported as the maximum change in fluorescence intensity from plots of change in fluorescence vs. time (60 s total assay time).
Asthma Cohort Studies
Details on the cohort used in this study have been published,15,26,38 and updated demographic and other information relevant to asthma are summarized in Table S2. Briefly, participants were recruited from the emergency department and inpatient wards at Primary Children’s Hospital, University of Utah, Salt Lake City, Utah, as part of an institutional review board–approved study of factors influencing asthma symptom control. All patients/parents/guardians provided informed consent prior to enrollment, DNA sampling, and data collection and analysis. Saliva samples were obtained prospectively from children 2–17 years of age with a physician-confirmed diagnosis of asthma. Information on chronic medical conditions, concomitant medication use, and the chief complaint at the time of enrollment was collected during enrollment and subsequent medical chart abstraction. The level of asthma control in each subject was assessed using a questionnaire based on guidelines modified from the National Heart, Lung, and Blood Institute’s Expert Panel Report 339 (Table S3), as described by Stockman et al.38 The questionnaire consisted of five questions scored on a four-point scale (i.e., 0, 1, 2 or 3). The asthma control score equals the sum of the five item scores, where 0 represents well controlled and 15 represents poorly/not controlled.
Saliva Collection, Genomic DNA Extraction, and SNP Genotyping
DNA was collected using either an Oragene DNA Kit (DNA Genotek) or Zymo DNA/RNA shield collection tube with swab (Zymo Research) and extracted using the GeneElute Mammalian Genomic DNA Miniprep Kit (Sigma-Aldrich). DNA was then quantified by Nanodrop and assayed for TRP SNPs using custom 64- or 128-feature TaqMan Open Array cards (ThermoFisher) containing assays for the TRPV1 I315M, T469I, I585V, and TRPA1 R3C and R58T SNPs, among others, at the University of Utah Genomics core facility. Prior to genotyping, of genomic DNA was amplified using a custom TaqMan PreAmp Mastermix specific to the array card features. The TaqMan reactions were cycled as recommended by the manufacturer on a Life Technologies QuantStudio 12K instrument. Data clustering and SNP identification analysis were performed using TaqMan Genotyper software (v1.3.1; Life Technologies).
Statistical Analysis and Graphics
Graphing and statistical analyses were performed using GraphPad Prism (version 7.03). Values are represented as the . One-tailed unpaired Student’s -tests; one-way analysis of variance (ANOVA) and Tukey’s multiple comparisons test; two-way ANOVA and either a Bonferroni, Tukey’s, or Dunnett’s multiple comparisons test; and repeated measures two-way ANOVA and Bonferroni’s multiple comparisons test were used as specified in the figure legends. A -value of was considered significant for all experiments. Schematics were prepared using Adobe Illustrator (v24.4.8).
TRPA1 mRNA Expression in AECs with the TRPV1 I585I/V Genotype
Transcriptome analysis comparing NHBEs with either the TRPV1 I585I/I or I585I/V genotype revealed an enrichment of TRPA1 mRNA in cells with the I585I/V genotype (Figure 1A), as previously reported.15 Using 13 unique donor lots of NHBEs and qPCR, cells with the TRPV1 I585I/I genotype expressed on average 3.9-fold () lower TRPA1 mRNA than cells with the I585I/V genotype, whereas TRPV1 was equivalent (Figure 1B,C). Stratification of these data based on the TRPV1 I315M and T469I SNPs demonstrated essentially no difference in TRPA1 mRNA expression, but TRPV1 mRNA was on average 1.8-fold () more abundant in cells with the TRPV1 T469I/T genotype (Figure S1). Agreement between changes in TRPA1 and TRPV1 mRNA and activity/protein in lung epithelial cells has previously been shown.15,25,29,40 In cells with the TRPV1 I585I/I genotype, mRNA for the NF--induced and NF- regulatory genes NLRP241,42 and IGFBP243–45 were enriched; NLRP2 and IGFBP2 mRNA expression was 2.2- () and 5.7-fold () lower in cells with the TRPV1 I585I/V genotype compared with cells with the I585 I/I genotype (Figure 1D,E).
Effect of TRPV1 Activity on TRPA1 mRNA Expression
The role of variable TRPV1 activity as a basis for differences in TRPA1 mRNA expression was tested by treating HBEC3-KT cells for 24 h with the TRPV1 agonist nonivamide (more active TRPV1, akin to TRPV1 I585I/I) or the antagonist LJO-328 (less active TRPV1, akin to TRPV1 I585I/V). Protracted stimulation of TRPV1 with nonivamide lowered TRPA1 mRNA expression 1.3-fold () compared with control cells, whereas 1.8-fold higher TRPA1 mRNA expression () was observed when TRPV1 was inhibited by LJO-328 (Figure 2A, white bars). The difference between nonivamide and LJO-328 treatment was 2.3-fold () and TRPV1 mRNA expression was not different between cells treated with nonivamide and those treated with LJO-328 (Figure 2B, white bars). Similar differences in TRPA1 mRNA expression were observed using TRPV1-overexpressing BEAS-2B cells (Figure S2); cells treated with nonivamide had 5.1-fold lower TRPA1 mRNA expression () and cells treated with LJO-328 had 12.5-fold higher expression ().
TRPA1 mRNA Expression and Effects of TRPV1 Agonists and Antagonists on AECs Treated with Pro-Inflammatory and Cytotoxic Agents
The effects of concurrent treatment of cells with pine WSPM, a pro-inflammatory and pneumotoxic environmental pollutant that activates TRPA1,6,24,25 on the ability of TRPV1 agonists and antagonists to alter TRPA1 mRNA expression was also evaluated. HBEC3-KT cells exposed to WSPM (24 h) alone had 10.9-fold higher TRPA1 mRNA expression (; Figure 2A, gray hashed bars). As above, cells co-treated with nonivamide and WSPM had lower TRPA1 mRNA expression (), whereas cells co-treated with LJO-328 and WSPM had 1.9-fold higher expression (; Figure 2A, gray hashed bars). Also as above, TRPV1 mRNA expression was similar to control cells for cells co-treated with nonivamide and WSPM, but it was 1.6-fold lower () in cells co-treated with LJO-328 and WSPM (Figure 2B, gray hashed bars). Cells were also treated with alternative inflammatory stimuli, including , , , IL6, and IL13. Cells treated with all but IL13 had higher TRPA1 mRNA expression 4 h posttreatment (Figure S3).
TRP mRNA Expression with WSPM/TRPA1 Agonist Challenge
The temporal profiles of TRPA1 and TRPV1 mRNA expression following WSPM treatment were distinct and opposite (Figure 3A,B), revealing an inverse relationship. As previously observed,25TRPA1 mRNA increased in control cells provided fresh media at , between 6 and 20 h. This did not occur with WSPM treatment. Accordingly, TRPA1 mRNA was markedly lower relative to the media-treated control cells when normalized at each time point. However, after , as the WSPM-treated cells visibly recovered from stress/damage and restored a monolayer, TRPA1 mRNA was higher (1.6-fold; ) compared with control cells, as in Figure 2A. The suppression of TRPA1 mRNA expression was also observed in cells treated with the TRPA1 agonists AITC, coniferaldehyde, and 2,4-di-tert-butylphenol (Figure S4), albeit the magnitude and duration of the effect varied by treatment, with WSPM producing the most robust effect and monolayer damage. Concurrently, TRPV1 mRNA was higher in cells treated with WSPM between 0 and 24 h (Figures 3B and S4), and similar time-dependent perturbations to mRNA expression for TRPV3 (higher expression across at all time points where TRPA1 was lower, similar to TRPV1), TRPV4 (higher expression at 0 to 8 h and lower at other times), and TRPM8 (lower at all time points up to 24 h, paralleling TRPA1) following WSPM treatment and treatment with other TRPA1 agonists. Like TRPA1 and TRPV1, variations in TRPV3 mRNA also manifest as changes in protein expression and activity.25,31
TRPA1 mRNA Expression in Cells with Suppression of TRPV1 Expression/Function Using siRNA
To further link TRPV1 expression/function with TRPA1 mRNA expression, TRPV1 siRNA was used. Cells treated with TRPV1 siRNA had 2.7- () and 3.4-fold () higher TRPA1 mRNA compared with the scramble and GAPDH siRNA controls, respectively (Figure 4, white bars). In this experiment, TRPV1 mRNA was () lower relative to both controls ( for both, gray hashed bars). For comparison, cells treated with GAPDH siRNA had 6.3-fold () and () lower GAPDH mRNA and protein expression (Figure S5).
NF- andTRPA1, TRPV1, and IL8 mRNA Expression
A role for NF- in regulating TRPA1 mRNA expression was tested by treating HBEC3-KT cells (24 h) with the Inhibitor of Nuclear Factor Kappa-B Kinase Subunits Alpha and Beta (; NF- pathway) inhibitor BMS-345541. Treated cells had 7.7-fold () lower TRPA1 mRNA expression compared with control cells (Figure 5A). Alternatively, cells treated with , an NF- pathway activator, had 2.6- () and () higher TRPA1 mRNA expression at 4 and 24 h, respectively (Figure 5B). Consistent with the inverse relationship between TRPA1 and TRPV1 mRNA expression, cells treated with expressed lower TRPV1 mRNA with time (1.8-fold lower with 24-h treatment; ; Figure 5C), consistent with reports that NF- inhibits TRPV1 expression.46 Simultaneously mRNA for another NF- target gene, IL8,47,48 was 24.3-fold () and 140.4-fold () higher at 4 and 24 h, respectively (Figure 5D). TRPA1 activity was also higher in HBEC3-KT cells treated for 24 h with , as indicated by 2.7-fold higher AITC-induced changes in intracellular calcium (; Figure 5E). In addition, pretreatment with BMS-345541 alone and in combination with , resulted in () and () lower AITC-induced calcium flux relative to the control, as well as 4.4- () and 4-fold () lower response compared with -treated cells. Figure S6 shows raw images and antagonist inhibition data confirming that the AITC-induced calcium flux was due to TRPA1.
Nuclear Localization of NF-/p65 and TRPA1 mRNA Expression
Nuclear localization of NF- is central to transcriptional regulation by NF-.49 Western blot analysis of NF- proteins from HBEC3-KTs treated with , nonivamide, and LJO-328 showed 2-fold () higher nuclear phospho-NF- (pNF-) with treatment and 1.8 () lower and essentially equivalent (i.e., 1.2-fold; ) nuclear pNF- with nonivamide and LJO-328 treatment, respectively (Figure 6A,B), consistent TRPV1-dependent modulation of NF- to alter TRPA1 expression. Unprocessed western blot images are shown in Figure S7.
Relationship between Protein Kinase C, p38 Mitogen-Activated Protein Kinase, and TRPA1 mRNA Expression
Protein kinase C (PKC)50 and p38 mitogen-activated protein kinase (MAPK)51,52 promote NF- pathway activity and enhance target gene (e.g., IL8) expression.47,48 PKC also acutely sensitizes TRPV1.53,54 HBEC3-KTs cells treated with the PKC activator and TRPV1 sensitizer PMA exhibited dose- and time-dependent lower TRPA1 mRNA expression at 4 and 12 h (Figure 7A) while simultaneously exhibiting higher TRPV1 and IL8 mRNA expression at 4 h, but less so at 12 h (Figure 7B,C). Consistent with a role for PKC and p38 MAPK activity in modulating NF- signaling, AECs treated for 12 h with the PKC inhibitor Go6983 and the p38 MAPK inhibitor PD169316 also had lower TRPA1 mRNA expression: 3.3- () and 2.9-fold (), respectively (Figure 7D), while simultaneously having 1.4-fold and 2.2-fold higher TRPV1 () and IL8 () mRNA expression (Figure 7E). Finally, it is possible that the increase in TRPV1 mRNA expression associated with protracted PKC inhibition in nonstimulated cells by Go6983 (Figure 7E) was driven in part by a reduction in basal TRPV1 activity given that 12-h Go6983 treatment attenuated nonivamide-induced TRPV1 activation in TRPV1-overexpressing HEK-293 cells (Figure 7F). This effect was characterized by a shift in the 50% effective concentration () from to and a decrease in the Hill slope from to (raw data are shown in Figure S8).
Relationship between NF-B, NLRP2, TRPA1, TRPV1, and IL8 mRNA Expression
Treatment of HBEC3-KT cells with the NF- pathway inhibitor BMS-345541 resulted in () lower NLRP2 mRNA expression (Figure 8A). The role for NLRP2 in TRPA1 mRNA expression was further evaluated. HBEC3-KT cells transfected with NLRP2 siRNA-1 and -2 exhibited 1.7 () and 2.2-fold () lower NLRP2 protein compared with the scramble siRNA control (Figure 8B,C) and 2.7-fold (; Figure 8D) lower NLRP2 mRNA (NLRP2 siRNA-2 shown). Unprocessed western blot images for NLRP2 are shown in Figure S9. Finally, cells transfected with NLRP2 siRNA-2 had 7.4-fold () higher TRPA1 (Figure 8E), 2-fold higher TRPV1 (), and 4.3-fold higher () IL8 mRNA (Figure 8G). Cells treated with TRPV1 siRNA also had () higher IL8 mRNA (Figure 8H), supporting a role for NLRP2 in regulating the NF--dependent expression of TRPA1 and IL8 and the suppression of TRPV1.
TRPV1 I585I/V Genotype and Responses of NHBE Cells to PM in Vitro
NHBE cells from multiple donors (Figure 1B–D) were treated with DEP and CFA. DDIT3 mRNA expression was used as a biomarker for TRPA1 activation and pathological endoplasmic reticulum stress,25 and IL8 mRNA as a marker of inflammation and likely NF- activity.29,40 Both DEP and CFA treatment resulted in higher IL8 and DDIT3 mRNA expression relative to the respective untreated donor line control (Figure 9A,B,D,E). In addition, the average fold difference in DDIT3 and IL8 mRNA was greater for cells with the I585I/V genotype: 2.3-fold () for DEP and 3.6-fold () for CFA (Figure 9C). For IL8, responses were also greater in I585I/V expressing cells, but not significant ( and 0.2029, respectively for DEP and CFA; Figure 9F). Correlation analysis confirmed an association between TRPV1 I585I/V genotype and TRPA1 mRNA expression (0.54), as well as for DEP-induced DDIT3 mRNA expression (0.93; Figure S10). A negative correlation between TRPV1 mRNA expression and the I585I/V genotype () was also observed. Finally, stratification of the DDIT3 and IL8 mRNA expression data as a function of either the TRPV1 I315M (Figure S11) or T469I (Figure S12) genotypes did not show evidence of an effect on in vitro cellular responses to DEP or CFA.
Gain and Loss-of-Function TRPA1 and TRPV1 SNPs and Asthma Symptom Control as a Function of Cigarette Smoke Exposure
Stratifying asthma control data for TRPV1 I585V allele expression and tobacco smoke exposure among children with asthma revealed worse asthma control for the genotype combination with smoke exposure (; Figure 10A). In fact, an unexpected trend toward improved control for the I585I/V genotype was observed for smoke exposure. In addition, the essentially co-inherited TRPA1 R3C and R58T SNPs,26 or allelic copy of the TRPV1 I315M or T469I SNPs were associated with poorer symptom control (, 0.0143, and 0.0031, respectively; Figure 10B–D) compared with individuals of the same genotype without smoke exposure. For comparison, the effects of several other SNPs previously reported to be associated with asthma risk or asthma symptoms associated with tobacco smoke including rs1871042 and rs6591255 (glutathione -transferase pi 1),55,56 rs4073 (IL8),57 rs1800925 (IL13),58,59 rs1042713 (adrenergic receptor beta-1),60 and rs2243250 (IL4)61 were also evaluated (Figure S13). In all cases, tobacco smoke exposure was associated with worse asthma symptom control, having similar effect sizes as the TRPA1 and TRPV1 SNPs, but genotype-specific associations were not observed as they were for the TRP SNPs.
From 2020 data, the Centers for Disease Control and Prevention estimates that people in the United States have asthma.62 Literature shows that exposure to environmental pollution increases risks for developing asthma among children and that asthmatics are more susceptible to exacerbation by environmental pollutants.7–13 Previous in vitro studies by our group have demonstrated that both TRPA1 and TRPV1 were variably activated by captured PM, including WSPM,24,25 DEP,4,5 CFA3,15 and cigarette smoke PM (CSPM)6. Depending upon source and composition, WSPM, DEP, and CSPM can be potent TRPA1 agonists, whereas CFA was shown to be a weak TRPA1 agonist,26 as well as an agonist of both TRPV13,15 and TRPM8.3,34 Because both TRPA1 and TRPV1 are implicated in the pathogenesis of asthma, this study tested the hypothesis that differences in TRPA1 expression and activity as a function of TRPV1 genetics may contribute to variations in AEC responses to PM challenge in vitro and asthma control, particularly as a function of one’s environmental exposure profile (i.e., their exposome or envirome), via altered sensitivity of AECs to PM and TRPA1 agonists.
The first objective of this study was to determine the mechanism driving higher TRPA1 expression by AECs having the TRPV1 I585I/V genotype. In vitro results suggest an integrated network consisting of PKC, p38 MAPK, NF-, and NLRP2 in regulating basal and dynamic TRPA1 expression in AECs. The activity of this network was TRPV1-genotype and -activity dependent, and TRPA1 activation itself served as a catalyst for variable expression. Multiple endogenous pro-inflammatory stimuli that directly or indirectly activate NF- (e.g., , , ) and orchestrate pulmonary inflammation, also affected TRPA1 expression. A hypothetical mechanism for how the TRPV1 genotype and associated changes in basal activity affect TRPA1 expression is presented in Figure 11. Further, the role of TRPA1 activation as a trigger for dynamic expression of TRPA1 and other TRPs (e.g., TRPV3) is shown in Figure 12.
The role of NF- in regulating TRPA1 expression has previously been shown in keratinocytes,63 A549 cells,64 and synoviocytes.63–66 The present study identified a comparable mechanism regulating basal and dynamic TRPA1 expression in AECs and further identified a relationship wherein loss or attenuation expression and function paradoxically promoted TRPA1 expression and function. Conversely, when TRPV1 was activated or more abundant, TRPA1 expression/function was lower. This relationship was also reciprocal in that TRPA1 activation using multiple agonists, including WSPM and AITC, led to transiently higher levels of TRPV1 mRNA and lower TRPA1 mRNA relative to control cells. Regarding this paradigm, the TRPV1 I585I/V genotype of HBEC3-KT cells was likely paramount, although dynamic expression of TRPA1 (and other TRPs) in BEAS-2B and NHBE AECs with the TRPV1 I585I/I genotype was/has also been observed.15,25,31 Regardless, this connection between TRP channels and TRP channel agonists is likely to have important consequences with respect to AEC and individual sensitivity to specific environmental stimuli and asthma triggers, as well as the ability to therapeutically manipulate TRPA1 or TRPV1 for pain, inflammation, and other purposes. Accordingly, a general recommendation is that this relationship be considered when studying TRPA1 and TRPV1 in AECs and possibly other cell types, particularly when evaluating the effects of pollutants or chemicals that may target one or both receptors.
A key aspect of TRPA1 regulation seemingly involved basal and temporal differences in NF- activity and associated NLRP2 expression as a function of cell status. In TRPV1 I585I/I cells, the markedly higher levels basal NLRP2 and IGFBP2 expression could suggest a higher level of basal NF- activity. Although not proven, it is tempting to hypothesize that higher NLRP2 and IGFBP2 expression may represent a mechanism to control basal inflammation mediated by NF- and TRPV1. Specifically, higher/more active TRPV1 could promote NF- signaling and increased IL8, NLRP2, and IGFBP2 expression, leading to the suppression of TRPA1 and cellular effects associated with changes in TRPA1 activity. Conversely, lower expression of NLRP2 in TRPV1 I585I/V cells could reflect a basally suppressed inflammatory state, due to less active TRPV1, and less need to negatively suppress NF-–regulated genes by NLRP2 feedback inhibition,41,42 including IL8 and TRPA1. Of significance, multiple TRP and pro-inflammatory stimuli that promote NF- signaling impacted this network, including , , and modifiers of PKC and p38 MAPK. PKC activation is regulated by intracellular calcium, which is a likely consequence of variable TRP activity. Here, activating PKC by PMA,53,54 which would also promote TRPV1 activity53,54 and NF- signaling led to a rapid but diminishing increase in IL8 and TRPV1 expression, as well as a delayed reduction in TRPA1 expression. This likely occurred by short term stimulation of NF--driven transcription, followed by a period of attenuated activity due to NLRP2-dependent negative feedback on NF-.41,42 Interestingly, inhibiting both PKC and p38 MAPK, which would attenuate basal NF- activity, also resulted in lower TRPA1 mRNA expression and higher TRPV1 expression, collectively demonstrating that the balance of NF- activation and the expression of NF-–regulated genes, including NLRP2, determined the TRPA1/TRPV1/IL8 dynamic. More work is necessary to validate and fully unravel the scope of interactions involved in the control of TRPA1 and TRPV1 expression/function by PKC, p38 MAPK, and presumably other kinases/phosphatases, and specifically how NRLP2 and NF- activities are altered. Such work should also address the limitation of this work by defining the kinetics of these interactions at the protein/activity level. Regardless, the finding that these entities communicated to affect TRP activity and expression basally and during inflammation/cell damage provides key insights into the complex, but orchestrated, regulation of TRP signaling during basal and pathological states, including following exposure of AECs to potential asthma triggers. Given the vital role of NF- in regulating pulmonary homeostasis, it is reasonable to conclude that the balance of this mechanistic hub may also play a key role in shaping cellular responses to various pro-inflammatory agents relevant to asthma.
A second objective of this study was to determine whether the I585I/V genotype and elevated TRPA1 expression would be consequential with respect to the effects of environmental pollutants on AECs in vitro and, ultimately, on asthma control. A theme that was consistent throughout this study was the inverse relationship between TRPA1 and TRPV1 mRNA expression and activity. The discovery of this balance, and the degree to which it is influenced by multiple TRPV1 genotypes and TRPA1 and TRPV1 agonists is unique, intriguing, and seemingly indicative of a broad mechanism for regulating inflammation and general cellular homeostasis in which TRPA1 and TRPV1 may play different, but defined roles in a context-dependent manner. Specifically, cellular calcium homeostasis is critical, with acute and chronic perturbations serving as a catalyst for many effects of stimuli, including those driven by PKC, p38 MAPK, and NF-. Results here suggest that regulation of an intracellular calciome is in part dependent upon TRPA1, TRPV1, and other dynamically regulated TRPs such as TRPV3, which seemingly act in concert to shape how cells respond to TRPA1 agonists, environmental stimuli, and even endogenous pro-inflammatory mediators relevant to asthma and airway physiology/pathophysiology in general. This conceptual framework is supported by both our prior work demonstrating the modulation of TRPA1, endoplasmic reticulum stress, and growth/repair in HBEC3-KT and other AECs by TRPV3,25,31 as well as by findings here that TRP genetics, and changes in TRPA1 and TRPV1 expression and function affect acute cellular responses to PM in vitro and asthma control as a function of tobacco smoke exposure.
Regarding the consequences of variable TRPA1 or TRPV1 expression/function on asthma control, results from a cohort analysis did not support the hypothesized association between the TRPV1 I585I/V genotype and worse asthma control as a function of smoke exposure. Rather, an unexpected trend indicative of better symptom control for individuals with the I585I/V genotype was observed, opposite that of individuals with the and other TRPA1 and TRPV1 genotypes evaluated. Despite limitations related to the use of self- and parental/guardian-reported smoke exposure (i.e., exposure is likely underreported), and the levels of expression/activity of the TRPs and other elements of the proposed regulatory network were not evaluated, several findings suggest the observed genotype–phenotype trends could be clinically relevant. First, higher TRPA1 expression and activity associated with the TRPV1 I585I/V genotype and tobacco smoke exposure would be expected to promote calcium-dependent PKC activation,67 acute sensitization of TRPV1,53,54,67 and TRPA1 suppression. With time, protracted PKC activation could also desensitize TRPV1. Second, despite the possibility that TRPV1 may become induced by TRPA1 agonists, at least with acute exposures, loss of function associated with TRPV1 I585V expression may render this effect irrelevant, whereas higher levels of expression of more active TRPV1 I315M and T469I15 may sensitize asthmatics to a broader array of stimuli, including non-TRPA1 agonists. This could explain why the TRPV1 I585V/V genotype has previously been associated with improved asthma symptom control15,27,28 and why the I315M and T469M genotypes have been associated with poorer symptom control with smoking. Interestingly, the effects of several other SNPs previously reported to be associated with asthma risk or asthma symptoms also showed associations with tobacco smoke exposure, but genotype-specific associations with asthma control were not observed, suggesting unique effects of TRP SNPs in regulating asthma control.
Finally, a person’s individual exposures must be considered with respect to the mechanisms described herein, and in understanding discrepancies between the in vitro observations related to the TRPA1/TRPV1 dynamic and asthma control. Specifically, the in vitro studies provide insight into how potential asthma-exacerbating stimuli could acutely perturb the NF-/TRP/IL8 balance, which may be more applicable to people who are not regularly exposed to TRPA1 agonists such as tobacco smoke but then are subsequently exposed. Alternatively, the cohort data provides insights on how chronic exposure to tobacco smoke (or perhaps other pollutants/TRPA1 agonists) may impact this dynamic; specifically, that TRPA1 signaling may become suppressed, whereas TRPV1 may become amplified.
To summarize, asthma symptom control is the product of an individual’s genetics, responses to therapeutics, medication adherence, lifestyle (diet and exercise), infection status, and exposure to indoor and outdoor triggers. Air pollutants can promote and exacerbate existing asthma by activating TRPA1 and TRPV1. A mechanism was demonstrated by which TRPV1 genotype and TRPV1 activity, as well as activation of TRPA1 itself, influenced TRPA1, TRPV1, other TRP expression by AECs through NF- signaling. Further, it was shown that the balance of TRPA1 and TRPV1 expression and activity was consequential in that higher TRPA1 expression was associated with higher acute cellular responses to selected model pollutants in vitro in AECs with the I585I/V genotype. However, analysis of asthma cohort data indicated a more complex relationship between TRPA1 and TRPV1, likely driven by variations in exposure to distinct types of asthma triggers. Overall, this study provides early but intriguing insight into how the TRPV1 I585I/V genotype and other SNPs, variable TRP expression, and exposure to certain types of pollutants may coordinately affect in vitro cellular responses to pollutant challenges and a person’s asthma in a variable environment.
This work was supported by the National Institutes of Health/National Institute of Environmental Health Sciences [grants ES017431 and ES027015 (to C.A.R.)] and National Institute of General Medical Sciences [grant GM121648 (to C.A.R.)].
The authors declare they have nothing to disclose.
Address correspondence to Christopher A. Reilly, Department of Pharmacology and Toxicology, Center for Human Toxicology, University of Utah, 30 S. 2000 E., Room 201 Skaggs Hall, Salt Lake City, UT 84112 USA. Telephone: (801) 581-5236. Email: [email protected]
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click DOWNLOAD.