Redox-sensitive GPCR signaling drives Gq-dependent Ca²⁺ mobilization and cytokine production in human bronchial epithelial cells

Abstract

Organic dust (OD) from animal production facilities contains a complex mixture of microbial products, metabolites, and particulates that engage airway epithelial signaling pathways. Here we investigated the mechanisms by which OD extract (ODE) activates immune signaling pathways in human bronchial epithelial (hBE) cells. ODE rapidly stimulated reactive oxygen species (ROS) generation and a biphasic increase in intracellular Ca²⁺ concentration ([Ca²⁺]_i), consisting of an early transient peak followed by a smaller sustained phase. Antioxidant scavenger pretreatment (glutathione, N-acetylcysteine) markedly attenuated both ROS production and Ca²⁺ mobilization, whereas induction of endogenous antioxidant defenses with bardoxolone abolished the response, indicating redox sensitivity. Pharmacologic inhibition of Gqα with YM-254890 suppressed both phases of the Ca²⁺ response, implicating Gq-coupled receptor activation. Consistent with an autocrine amplification mechanism, selective antagonists of histamine (H1), cysteinyl leukotriene (CysLT1 and CysLT2), leukotriene B4 (BLT1), and prostaglandin receptors (EP1) each reduced ODE-evoked Ca²⁺ mobilization. In parallel, inhibitors of histidine decarboxylase, 5-lipoxygenase, and cyclooxygenases (COX-1/COX-2) attenuated Ca²⁺ signaling, supporting rapid endogenous ligand production and secretion. Downstream of Ca²⁺ mobilization, ODE activated PKCα/β and PKCδ and induced robust transcription of proinflammatory cytokine and chemokine mRNAs (IL1β, IL6, IL8, IL33, TNFα) within 2 hours of exposure. ELISA confirmed increased secretion of IL-1β, IL-6, IL-8, and IL-33, with differential sensitivity to PKC isoforms and NF-κB inhibition. These findings identify a redox-sensitive GPCR network that amplifies Gq-dependent Ca²⁺ signaling in airway epithelial cells and provides a mechanistic framework for epithelial inflammatory activation following ROS-inducing environmental exposures.

Graphical Abstract

Introduction

Agricultural workers are at increased risk of respiratory disease due to occupational exposure to organic dust (OD), a complex mixture derived from plant, animal, and microbial sources (1, 2). OD exposure has been associated with a spectrum of respiratory conditions, including rhinosinusitis, asthma, chronic bronchitis, chronic obstructive pulmonary disease (COPD), and hypersensitivity pneumonitis (Farmer’s lung) (2–5). Acute syndromes such as organic dust toxic syndrome may develop following exposure and are characterized by headache, malaise, and nausea, while mucous membrane irritation syndrome is associated with increased nasal inflammation (3). Controlled human exposure studies demonstrate that even short-term exposure to swine confinement environments induces significant reductions in expiratory flow rates and promotes airway hyperresponsiveness in previously unexposed individuals (3, 5, 6).

Air sampling studies of swine confinement facilities show that OD is composed primarily of grain particles and dried fecal matter, along with biologically active components including microbial agents, endotoxins, and proteins (7). Numerous fungal species have been identified in swine OD, which may elicit adverse health effects through immune activation, infection, or the production of mycotoxins (8). In addition, bacterial communities within swine OD—predominantly gram-positive species—closely resemble the porcine gastrointestinal microbiome (9). Elevated levels of ammonia and endotoxins in these environments have been strongly associated with respiratory symptoms and airway inflammation in exposed workers (3, 10). Importantly, both viable and nonviable microbial components contribute to these effects, indicating that biological activity is not limited to living organisms within the dust (9).

Despite well-established associations between OD exposure and respiratory disease, the underlying pathogenic mechanisms remain incompletely defined. Increasing evidence suggests that OD exposure induces the generation of reactive oxygen species (ROS), which function as signaling intermediates but also contribute to oxidative stress and inflammation (1, 11, 12). ROS-mediated signaling has been shown to regulate the expression of inflammatory mediators, in part through activation of transcription factors such as STAT3 (1). Excessive ROS production disrupts redox homeostasis, leading to oxidative stress that promotes inflammation, cellular injury, and cell death (11). Inhaled environmental pollutants represent a major exogenous source of ROS and contribute to both local airway and systemic inflammatory responses (13).

Calcium (Ca²⁺) signaling plays a central role in airway epithelial cell function and inflammatory responses. Intracellular Ca²⁺ ([Ca²⁺]_i) acts as a key second messenger regulating cytokine production, epithelial barrier function, and responses to environmental stressors (14). Crosstalk between oxidative stress and Ca²⁺ signaling pathways has been reported, with ROS capable of modulating Ca²⁺ channel activity and intracellular Ca²⁺ homeostasis. However, the role of Ca²⁺ signaling in mediating OD-induced inflammatory responses in airway epithelial cells remains incompletely understood.

Therefore, in the present study we hypothesized that oxidative stress induced by swine OD extract increases intracellular [Ca²⁺]_i and that Ca²⁺ mobilization is required for the induction of inflammatory autocrine signaling pathways that control cytokine transcription and secretion by airway epithelial cells. To test this hypothesis, we: 1) determined the effects of swine ODE exposure on [Ca²⁺]_i mobilization and assessed whether inhibition of oxidative stress using ROS scavengers or an NRF2 activator attenuated the response; 2) identified signaling pathways responsible for ODE-induced increases in [Ca²⁺]_i; and 3) evaluated the effects of ODE exposure on inflammatory cytokine/chemokine expression and secretion and determined whether these responses were dependent on oxidative stress. Elucidating these mechanisms is essential for improving our understanding of OD-induced airway inflammation and for developing strategies to mitigate respiratory risk in agricultural workers.

Materials and Methods

Organic Dust (OD) Collection:

Swine OD was collected from the University of Minnesota swine facility (Falcon Heights, MN, USA) using Nasco Sampling Whirl-Pak Dry collection sponges (Cat. # B01316; Whirl-Pac Filtration Group, Fort Atkinson WI) and phosphate buffered saline (PBS) (Cat. # 10010031; Thermo-Fisher Scientific, Waltham, MA, USA). OD that had settled from the air was collected from elevated horizontal surfaces perpendicular to the floor. Samples were obtained from two rooms at four locations within the swine facility. The collection procedure involved wetting a Whirl-Pak sponge with sterile PBS, then thoroughly wiping the surface in two different directions before placing the sponges back in the sealed sterile bags. At the lab, sponges were rehydrated with 25 mL of PBS each and the contents were expelled into a sterile 50 mL conical centrifuge tube (Cat. # 430828; Corning Life Sciences, Lowell, MA, USA). All samples were brought to a total volume of 20 mL with PBS and centrifuged at 1200 rpm for two minutes. Undissolved sediment collected at the bottom of the tube was discarded and the supernatants were pooled to create one combined OD extract (ODE) sample which was aliquoted and stored at −80°C until use.

Epithelial Cell Culture Conditions:

Human bronchial epithelial (hBE) cells (female) were originally immortalized following transfection of genes encoding cyclin-dependent kinase-4 and human telomerase reverse transcriptase (15) and obtained from American Type Culture Collection (ATCC; Cat. # CRL-4051). hBE cells were cultured in bronchial epithelial cell growth medium with growth factor supplements (Cat. # C-21160; PromoCell GmbH, Heidelberg, Germany) and incubated at 37 °C in a humidified atmosphere of 5% carbon dioxide for 48–72 h to achieve >90% confluence prior to all experiments. All experiments were performed using cells between passages 10-30. Cells were initially screened for mycoplasma contamination using an EZ-PCR Mycoplasma Detection Kit (Cat. # 20-700-20; Sartorius, Göttingen, Germany) and as an added precaution, media was treated with Plasmocin^® (Cat. # ant-mpp; InvivoGen, San Diego, CA) to prevent mycoplasma contamination. Cell monolayers were grown on 96-well plates (Cat. # 655160; Greiner Bio-One, Monroe, NC, USA) for [Ca²⁺] imaging, ROS measurements or DNA release experiments or on 12 well plates for ACh measurements and ELISA assays (Cat. # 3513; Thermo-Fisher Scientific, Waltham, MA, USA). Seeding densities for ATP release and Quantitative polymerase chain reaction (qRT-PCR) experiments: 5 x 10⁴ cells/dish, to reach 5 x 10⁵ cells/dish, DNA release, ROS and intracellular [Ca²⁺] measurements: 5 x 10³ cells/well, to reach 5 x 10⁴ cells/well, [ACh] concentration measurements and ELISA assays: 2 x 10⁴ cells/well, to reach 2 x 10⁵ cells/well. A list of compounds, solvents, final concentrations and incubation times for all experiments are provided in Table 1.

Table 1.

Compounds, solvents, final concentrations and incubation times.

Compounds	Cat. #	Sources	Solvents	[Final]	Times
GSH	G6013	Sigma	Water	5 mM	4 hr
N-acetylcysteine	A7250-5G	Sigma	Water	5 mM	4 hr
CDDO-Me	SMB00376	Sigma	DMSO	100 nM	24 hr
YM254890	AGCN20509	AdipoGen	Water	20 μM	30 min
DPH	HY-B0303	MCE	Water	50 μM	30 min
α-FMH	525854	MedKoo	Water	10 μM	24 hr
Montelukast	80064	AstaTech	DMSO	5 μM	30 min
BayCysLT2	10532	Cayman	DMSO	5 μM	30 min
LY293111	10009768	Cayman	DMSO	5 μM	30 min
Zileuton	S1443	Selleckchem	DMSO	10 μM	30 min
SC51322	10010744	Cayman	DMSO	10 μM	30 min
Indomethacin	I-7378	Sigma	DMSO	10 μM	30 min
FSLLRY-NH2	4751	Tocris	Water	20 μM	30 min
DAU5884	SML0845	Sigma	DMSO	10 μM	30 min
Scopolamine	1414	Tocris	Water	10 μM	30 min
EXP3174	40904	AstaTech	DMSO	5 μM	30 min
Icatibant Ac	HY-17446	MCE	Water	5 μM	30 min
Bosentan	HY-A0013	MCE	DMSO	5 μM	30 min
Go6976	13310	Cayman	DMSO	10 μM	30 min
Delcasertib	949100394	Tocris	DMSO	10 μM	30 min
Bay11-7082	196870	Sigma	DMSO	10 μM	1 hr
Histamine	EEL005	Thermo	Water	1-10 μM	None*
LTB4	20110	Cayman	EtOH	5 nM-5 μM	None*
LTC4	20210	Cayman	EtOH	5 nM-5 μM	None*
LTD4	20310	Cayman	EtOH	5 nM-5 μM	None*

Sigma-Aldrich (Sigma), St. Louis, MO.

AdipoGen life Sciences (AdipoGen), San Diego, CA.

MedChemExpress (MCE), Monmouth, NJ.

MedKoo Biosciences (MedKoo), Morrisville, NC.

AstaTech Inc. (AstaTech), Bristol, PA.

Cayman Chemical (Cayman), AnnArbor, MI.

Selleck Chemicals LLC (Selleckchem), Houston, TX.

Tocris Bioscience (Tocris), Minneapolis, MN.

Thermo-Fisher Scientific (Thermo), Waltham, MA.

^*None: Refers to no pretreatment with agonist. Acute responses were recorded

Protein Assay:

Protein concentration was measured using a Qubit Protein Assay kit (Cat. # Q33211; Invitrogen, Carlsbad, CA, USA) following the manufacturer’s protocol. A Qubit Flex Fluorometer (Cat. # Q33327; Thermo-Fisher Scientific) at 485/590 nm, was used to compare sample fluorescence intensity to a dilution standard curve for protein quantification. The protein concentration was 1.1 ± 0.12 mg/ml, averaged over 4 sites where OD samples were collected. Experiments involved exposure of hBE cells to a 1:20 dilution of ODE ([protein] = 55 μg/ml).

Intracellular [Ca²⁺] Measurements:

Changes in [Ca²⁺]_i were measured using a FLIPR Calcium 6 Assay kit (Cat. # R8190; Molecular Devices, San Jose, CA). hBE cells were grown on 96-well plates and a loading buffer (available as part of the kit) was initially added to each well followed by a 2-hour incubation period at 37 °C in a humidified atmosphere containing 5% carbon dioxide. A 96-well compound plate was then prepared with ODE, and both plates were loaded into a FlexStation 3 Microplate Reader (Cat. # Flex3; Molecular Devices). ODE was subsequently added to cell monolayers using the auto-titration feature of the FlexStation 3 Microplate Reader and fluorescence intensity measured at excitation/emission wavelengths of 485/525 nm. Normalizing intracellular [Ca²⁺] measurements in Figure 2 involved calculating the mean maximum relative fluorescence intensity (RFI) Ca²⁺ response for the control ODE treatment condition then dividing each of the RFI values by the mean maximum RFI and expressing the result as the fractional response applied to all pretreatment conditions in the experiment.

Figure 2.

ODE-induced increases in intracellular [Ca²⁺]. A. ODE produced a rapid increase in [Ca²⁺]_i that peaks within 60 seconds then decreases towards baseline before increasing again to a lower, sustained level for at least 5 minutes following stimulation. Pretreatment with oxidant scavengers including glutathione (GSH, 5 mM) or N-acetylcysteine (NAC, 5 mM) significantly reduced the ODC response and treatment with the Nrf2 activator CDDO-ME (100 nM for 24 hrs) abolished both the peak and sustained increases in [Ca²⁺]_I (*p < 0.0001; One way ANOVA with Tukey’s post-test, n = 4). B. ODE-stimulated Ca²⁺ mobilization was significantly inhibited with the Gqα inhibitor YM254890 (20 μM) (*p = 0,0021; Welch-corrected two-tailed t test, n = 4). C. Removal of extracellular Ca²⁺ significantly reduced both peak and sustained ODE Ca²⁺ responses (*p = 0.0017; **p = 0.0239 respectively; unpaired t test, n = 4) D. The H1 histamine receptor antagonist, diphenhydramine (50 μM), reduced the peak of the ODE effect on [Ca²⁺]_i by ~25%. Moreover, inhibition of endogenous histamine production by airway epithelial cells with the histidine decarboxylase inhibitor α-fluromethylhistamine (α-FMH;10 μM) caused a similar decrease in the [Ca²⁺]_i peak response (*p = 0.0313 and **p = 0.0157 respectively; One way ANOVA with Tukey’s post-test, n = 4). E. Antagonists targeting the BLT₁ (LY293111, 5 μM), CysLT₁ (Montelukast, 5 μM) and CysLT₂ (BayCysLT2, 5 μM) receptors each inhibited the ODE Ca²⁺ response ranging from ~34-54%. Furthermore, the 5-lipoxygenase inhibitor Zileuton, which blocks endogenous LTB₄, LTC₄ and LTD₄ production inhibited the ODE-induced increase in [Ca²⁺]_i by ~70% (*p < 0.0001; **p = 0.0007; One way ANOVA with Tukey’s post-test, n = 4). F. Pretreatment with SC51322, a selective antagonist for the EP₁ receptor inhibited the peak ODE response by ~46%, while pretreatment with the cyclooxygenase inhibitor indomethacin produced comparable inhibition that was not significantly different than SC51322 (*p < 0.0001 and **p = 0.0002 respectively; One way ANOVA with Tukey’s post-test, n = 4).

Reactive Oxygen Species:

ROS generation was measured using a ROS Assay kit (Cat. # STA-347; Cell Biolabs, San Diego, CA, USA). hBE cells were grown to confluence on 96-well plates (Cat. # 655160; Greiner Bio-One). The RFI value corresponds to a mean cell number/well od ~5 x 10⁴ cells. A loading buffer (available as part of the kit) was added to the cells and incubated at 37 °C for 60 minutes. Addition of ODE to the cell monolayer was performed by the FlexStation 3 Microplate Reader (Molecular Devices) and fluorescence was continuously measured at 480/530 nm at 1 minute intervals.

ATP release assay:

ATP release was measured in real time following stimulation with ODE or Alternaria extract using an ATP determination kit (Cat. # A2206; Invitrogen, Carlsbad CA) and a previously published protocol (16, 17). Briefly, hBE monolayers grown on 35 mm culture dishes (Cat. # 153066; Thermo-Fisher Scientific) were bathed in standard reaction solution containing luciferin and luciferase then placed into a Glomax 20/20 luminometer (Cat. # E5311; Promega, Madison WI). Luminescence was measured for 1 minute prior to addition of either ODE or Alternaria extract (50 μg/ml) to establish baseline luminescence, then changes in intensity were measured continuously for the next 5 minutes following extract treatments. Luminescence was converted to [ATP] using an ATP calibration curve (range, 0.1-100 nmol) and expressed as nmol/cm². An area of 1 cm² contains ~5 x 10³ cells.

DNA release assay:

DNA release was measured in hBE cells grown to confluence on 96 well plates using a SpectraMax Quant dsDNA assay kit (Cat. # R8358) and a FlexStation 3 Microplate Reader (Molecular Devices) according to the manufacturer’s instructions. Levels of extracellular DNA were expressed as the change in fluorescence intensity above the background fluorescence (18).

qRT-PCR analysis.

A RNeasy Mini Kit (Cat. # 74104) from Qiagen (Hilden, Germany) was used to extract total RNA. RNA was quantified by spectrophotometry then reverse-transcribed to cDNA as previously described (19). qRT-PCR was carried out using a Thermo-Fisher Step-One-Plus Real-Time qPCR System (Cat. # 4376592). The threshold cycle (Ct) number was used to determine the relative expression level of specific genes. The TaqMan probes (Thermo-Fisher Scientific) used for detecting mRNA transcripts are listed in Table 2 (Cat. #s in Table 2).

Table 2.

qRT-PCR TaqMan probes

Target	Cat. #
GAPDH	Hs02786624_g1
TNFAIP6	Hs00200180_m1
IL1-β	Hs01555410_m1
IL-6	Hs00174131_m1
IL-33	Hs00369211_m1
TSLP	Hs00263639_m1
IL-8	Hs00174103_m1

ELISA Assay.

The supernatant from hBE cells exposed to ODE was collected and used in enzyme-linked immunosorbent assays (ELISA) following a previously published protocol (20). The Legend Max Human IL-1β ELISA Kit (Cat. # 437004; BioLegend, San Diego, CA), Legend Max Human IL-6 ELISA Kit (Cat. # 430507; BioLegend), Legend Max Human IL-8 ELISA Kit (Cat. # 431507; BioLegend) and the Quantikine ELISA Human IL-33 kit (Cat. # D3300B; Bio-Techne, Minneapolis, MN) were used to analyze the samples. hBE cells were grown to confluency on 12-well plates. After ODE exposure for a period of 2 hours, 50 μL of supernatant from the cell culture wells was transferred to sample wells of an ELISA plate. The plates were then treated according to the manufacturer’s protocol and absorbance at 450 nm was measured within 30 minutes of the final step of the assay using a FlexStation 3 Microplate Reader (Molecular Devices).

Bacterial Characterization of ODE:

DNA from bacteria inherent to ODE was extracted using the Quick-DNA HMW MagBead Kit (Cat. # D6060; Zymo Research, Irvine, California, USA), with 3 negative controls made utilizing the same kit to account for any bacteria naturally occurring within the DNA extraction kit (21). Each of the four environmental samples of ODE were processed in triplicate, along with three replicates of the “master mix” of ODE that combined all four environmental sampling locations. Samples were prepared for sequencing using Oxford Nanopore’s 16S Barcoding Kit 24 V14 (Cat. # SQK-16S114.24; Oxford Nanopore Technologies, OX4 4DQ, UK), following the manufacturer’s protocol. Barcoded samples were pooled and sequenced on an Oxford Nanopore minION R10.4.1 flow cell utilizing minKNOW software (v6.2.14). Sequencing reads were base-called and quality-filtered utilizing dorado (v1.0.0, Oxford Nanopore Technologies) super-accurate model (v5.0) with default read quality thresholds.

Reads were first filtered to remove human and pig contaminant DNA from samples by aligning reads to the human and pig (https://ftp.ncbi.nlm.nih.gov/genomes/all/GCF/000/001/405/GCF_000001405.26_GRCh38/); (https://ftp.ncbi.nlm.nih.gov/genomes/all/GCF/000/003/025/GCF_000003025.6_Sscrofa11.1/) reference genomes using minimap2 (22). Bacterial taxa in the remaining ODE reads were then classified at each level of phylogenetic classification and their abundances in samples were estimated using a combination of Kraken2 (23) and bracken (24) using the SILVA database (https://ftp.arb-silva.de/release_138.1/Exports/SILVA_138.1_SSURef_NR99_tax_silva.fasta.gz) (25). Sequencing reads stemming from bacteria already inherent to the extraction reagents (negative controls) were removed from estimates of relative abundances at the amplicon sequence variant (ASV) level through the R package microDecon (26) and Rstudio (v2024.04) utilizing default threshold settings at the ASV level and requiring the presence of an ASV in all 3 negative controls to be considered a contaminant. Because taxonomic classifications found no differences in composition or abundances of taxa between the four individual environmental sampling locations, characterizations were averaged values for the 3 replicates of the pooled “master mix”.

Statistical analysis:

Data are presented as means ± standard errors of the mean. Statistical analyses and curve fitting were performed using GraphPad Prism v10 software (GraphPad Software, La Jolla, Calif; https://www.graphpad.com/how-to-buy/). Specific statistical tests included unpaired, two-tailed t-tests, linear regression analysis, and one-way ANOVAs followed by Tukey’s post-test for multiple comparisons as indicated in the Results section and in the Figure Legends. A “p” value equal to or less than 0.05 was considered significant.

Results

ODE stimulated ROS production in hBE cells over a 10-minute time course (Fig. 1A). Heating the extract for 10 minutes at 80°C significantly reduced ROS production (ROS RFI/min before heating = 5.72 ± 0.23 and after heating = 1.49 ± 0.04, p = 0.0003; values determined by linear regression analysis and Welch’s unpaired t test, n = 4 each condition). Unlike the effects of extract derived from the common saprophytic fungus Alternaria alternata, which was previously shown (16, 27) to induce ROS production and ATP release in hBE cells, ODE did not stimulate the secretion of ATP (Fig 1B). Furthermore, results presented in Fig. 1C also shows that ODE does not induce release of DNA fragments into the extracellular media, in contrast to the effects of Alternaria, which stimulated DNA release at the same concentration used to induce ATP secretion with kinetics similar to previously published findings (18, 28).

Figure 1.

ODE exposure induces ROS production but not ATP or DNA release. A. ODE stimulated a rapid increase in the rate of intracellular ROS production. Preheating the extract to 80°C for 10 minutes significantly (p = 0.0003) reduced the rate of ROS production (Welch-corrected, two-tailed t test, n = 4). B. In contrast to Alternaria extract, ODE does not induce ATP release (n = 3) or DNA release (C.) into the extracellular media (n = 4).

ODE induced a rapid increase in [Ca²⁺]_i within 1 minute after exposure. This initial peak was followed by a more sustained secondary rise in [Ca²⁺]_i that varied in amplitude and duration (Fig 2A–C). Pretreatment of hBE cells with ROS scavengers including glutathione (GSH, 5 mM) and N-acetylcysteine (NAC, 5 mM) inhibited (p < 0.0001, n = 4) both the initial peak and secondary increase in [Ca²⁺]_i suggesting that Ca²⁺ mobilization was dependent on ODE-induced ROS production (Fig 2A). Moreover, pretreatment for 24 hr with the Nrf2 activator, CDDO-ME (100 nM; also known as bardoxolone) completely suppressed (p < 0.0001, n = 4) Ca²⁺ mobilization, providing further support for a ROS-dependent ODE response (Fig. 2A). The effects of GSH, NAC and CDDO-ME were not significantly different from each other (GSH vs CDDO-ME, p = 0.0752, n = 4; NAC vs. CDDO-ME, p = 0.9400, n = 4 and GSH vs. NAC, p = 0.2119, n = 4; one-way ANOVA followed by Tukey’s post-test). Additional experiments investigating the effects of the selective Gq inhibitor YM254890 (20 μM) showed nearly complete inhibition (p = 0.0021, n = 8, unpaired, two-tailed t-test) of the initial peak as well as the secondary increase in [Ca²⁺]_i suggesting that the Ca²⁺ response involved activation of Gq-coupled G-protein receptors (GPCRs; Fig. 2B). To determine the degree to which the rapid [Ca²⁺]_i peak and secondary rise of the Ca²⁺ response was dependent on extracellular [Ca²⁺], hBE cells were bathed in nominally Ca²⁺-free extracellular media for 60 minutes prior to ODE exposure. Lowering extracellular [Ca²⁺] reduced both the peak response (p = 0.0017, n = 4) and the secondary increase (p = 0.0239, n = 4) in [Ca²⁺]_i following treatment with ODE indicating that a portion of the effect was dependent on extracellular [Ca²⁺] (unpaired, two-tailed t-test). Table 3 summarizes the effects of GSH, NAC, bardoxolone and Ca²⁺-free conditions on the ODE Ca²⁺ response.

Table 3.

ROS scavengers, an Nrf2 activator and nominally Ca²⁺-free conditions inhibit Ca²⁺ mobilization induced by ODE exposure.

Nrf2- activator/ROS scavenger	[Compound]	% of Control ODE Response ± SE
Glutathione	GSH (5 mM)	34 ± 13 (n = 4)
N-acetylcysteine	NAC (5 mM)	5 ± 10 (n = 4)
Bardoxolone	CDDO-ME (100 nM)	3 ± 1 (n = 4)
Ca2+-free conditions	Initial peak	46 ± 17 (n = 4)
Ca2+-free conditions	Secondary increase	13 ± 11 (n = 4)

In order to identify which Gq-coupled GPCRs were involved in producing the initial increase in [Ca²⁺]_i, we examined several well characterized antagonists to known Gq receptors expressed by hBE cells. Initially the effects of the H1 histamine receptor antagonist diphenhydramine (DPH) was tested and found to decrease the peak amplitude of the response by ~25% (p = 0.0313) and also delayed the time needed to reach the maximum increase in [Ca²⁺]_i by ~10 seconds (Fig. 2D). As a second, orthogonal experiment, we investigated the effect of an inhibitor of histidine decarboxylase, (S)-α-fluoromethylhistidine (α-FMH), a rate-limiting enzyme involved in histamine synthesis by hBE cells. We found that like DPH, α-FMH produced a similar 25% decrease (p = 0.0157) in the peak Ca²⁺ response without altering the kinetics of Ca²⁺ mobilization (Fig 2D). The effects of DPH and α-FMH were not significantly different (p = 0.8456; one-way ANOVA followed by Tukey’s post-test).

Next we examined the effects of selective leukotriene receptor antagonists targeting BLT₁-R, CysLT₁-R and CysLT₂-R GPCRs expressed by hBE cells (29–32). Pretreatment with the highly selective CysLT₁-R antagonist Montelukast (33) inhibited the ODE Ca²⁺ response by ~54% (p < 0.0001) and like DPH, delayed the time to peak amplitude by ~15 seconds (Fig. 2E). Similarly, BayCysLT2, a selective CysLT₂-R antagonist (33, 34) reduced (p = 0.0007) the [Ca²⁺]_i peak by ~34% without altering the kinetics of the calcium response (Fig. 2E). Similarly, LY293111, which selectively antagonizes BLT₁-R (35), lowered (p < 0.0001) the ODE peak response by 41% and delayed the time-to-peak by ~6 seconds (Fig. 2E; one-way ANOVA followed by Tukey’s post-test). These findings indicated that multiple leukotriene receptors were activated by ODE exposure and that leukotriene release following ODE exposure may account for the activation of these receptors. To address this possibility, hBE cells were pretreated with the 5-lipoxygenase inhibitor zileuton to determine what effect it might have on Ca²⁺ mobilization and the kinetics of the ODE-induced increase in [Ca²⁺]_i.. Zileuton reduced (p < 0.0001) the [Ca²⁺]_i peak by ~70% and delayed the time-to-peak by ~12 seconds, consistent with involvement of ODE-evoked leukotriene secretion in autocrine activation of multiple leukotriene receptors (Fig. 2E; one-way ANOVA followed by Tukey’s post-test). Table 4 presents the inhibitory effects of specific Gq GPCR antagonists on ODE Ca²⁺ mobilization.

Table 4.

G_qα inhibition, GPCR antagonists and inhibition of endogenous agonist synthesis enzymes reduced ODE-stimulated Ca²⁺ mobilization.

Gq Receptor/Enzyme	[Antagonist]/[Enzyme inhibitor]	% of Control ODE Response ± SE
Gqα inhibitor	YM254890 (20 μM)	18 ± 28 (n = 4)
Histamine (H1)	DPH (50 μM)	77 ± 10 (n = 4)
Histidine decarboxylase	α-FMH (10 μM)	73 ± 11 (n = 4)
CysLT1 (LTD4)	Montelukast (5 μM)	46 ± 18 (n = 4)
CysLT2 (LTC4/LTD4)	BayCysLT2 (10 μM)	67 ± 8 (n = 4)
LTB4 receptor	LY293111 (10 μM)	59 ± 15 (n = 4)
5-Lipoxygenase	Zileuton (10 μM)	30 ± 16 (n = 4)
EP1 receptor	SC 51322 (10 μM)	67 ± 13 (n = 4)
Cyclooxygenase	Indomethacin (10 μM)	54 ± 10 (n = 4)

In addition to the effects of H1 and leukotriene receptor antagonists, pretreatment with the selective EP₁-R antagonist SC 51322 decreased the peak [Ca²⁺]_i response by ~46% (p < 0.0001, n = 4), without altering the kinetics of Ca²⁺ mobilization. Furthermore, pretreatment with indomethacin, a pan-specific cyclooxygenase inhibitor that blocks production of prostaglandins also reduced the [Ca²⁺]_i peak by ~33%, which was significantly lower than ODE exposure alone (p = 0.0002, n = 4) but not significantly different from the effects of SC 51322 (p = 0.1325, n = 4, one-way ANOVA followed by Tukey’s post-test). Like SC 51322, no effect was observed on the kinetics of the Ca²⁺ response. Figure 3 shows a series of experiments where H1 and leukotriene receptor agonists were used to provide additional supporting evidence for the expression and function of the Gq-coupled GPCRs previously shown to be activated by ODE (Fig. 3A–D). Measurements of peak [Ca²⁺]_i responses were performed with histamine, LTB₄, LTC₄ and LTD₄ and the resulting concentration-response relationships were fit with three-parameter logistic functions (Hill slopes = 1) where EC₅₀ values were derived from the fit (Fig. 3E, note that EC₅₀) values are reported in the legend).

Figure 3.

Agonist-induced activation of Gq-coupled GPCRs. A. Histamine, at progressively higher concentrations increased the peak [Ca²⁺]_i response (n = 4). B. LTB₄ concentration-dependent increase in [Ca²⁺]_I (n = 4). C. The LTC₄-induced increase in [Ca²⁺]_i was also concentration dependent (n = 4). D. The LTD₄-stimulated increase in [Ca²⁺]_i occurred over the same range of concentrations as observed for LTB₄ and LTC₄ (n = 4). E. Concentration-response curves for histamine (EC₅₀ = 2.2 μM; r² = 0.92, n = 4), LTB₄ (EC₅₀ = 132 nM; r² = 0.95, n = 4), LTC₄ (EC₅₀ = 116 nM; r² = 0.90, n = 4) and LTD₄ (EC₅₀ = 127 nM; r² = 0.94, n = 4). F. Alternaria and ODE-induced acetylcholine (ACh) secretion was measured at 15 and 30 minutes after stimulation Note that no significant difference was detected between the 15 and 30 minute exposure times for Alternaria or ODE, but ACh secretion in response to Alternaria was significantly greater than ODE at both time intervals (*p < 0.0001; one-way ANOVA followed by Tukey’s post-test, n = 4).

We also examined the effects of other Gq GPCRs expressed by hBE cells (Table 5) and found that they did not produce significant inhibition of the ODE Ca²⁺ response. However, it is noteworthy that the muscarinic receptor antagonist scopolamine reduced the [Ca²⁺]_i peak by ~25%, but the variability in the response presumably prevented a statistically significant outcome (unpaired, two-tailed t-test). Since we have previously shown that increases in oxidative stress in hBE cells triggers secretion of endogenously synthesized acetylcholine (ACh; 17), we examined the effect of ODE on ACh release (Fig. 3F). As a positive control, we also tested the effects of Alternaria and found an increase in extracellular [ACh] that was comparable to levels reported in our earlier study (36). In contrast, significantly lower amounts of ACh secretion were detected following ODE exposure, which may explain the variability of the scopolamine response.

Table 5.

Antagonists/enzyme inhibitors of other G_q-coupled receptors did not inhibit ODE-evoked Ca²⁺ responses

Gq Receptor	[Antagonist]	% of Control ODE Response ± SE
Protease Activated PAR2	FSLLRY-NH2 (20 μM)	91 ± 16 (p = 0.9968)
Protease Activated PAR2	FSLLRY-NH2 (40 μM)	86 ± 14 (p = 0.5454)
Muscarinic M3	DAU5884 (10 μM)	93 ± 11 (p = 8678)
Muscarinic M1	Scopolamine (10 μM)	75 ± 11 (p = 0.3113)
Angiotensin AT1	EXP3174 (5 μM)	92 ± 8 (p = 0.6646)
Bradykinin B2	Icatibant Ac (5 μM)	98 ± 11 (p = 0.8598)
Endothelin ETA	Bosentan (5 μM)	91 ± 3 (p = 0.5413)

ODE exposure also stimulated mRNA transcription of various inflammatory cytokines/chemokines and the anti-inflammatory protein TNFAIP6, produced in response to TNFα as well as other inflammatory cytokines within 2 hours after treatment (Fig. 4). Transcription of TNFAIP6, IL-1β, IL-6, IL-33 and IL-8 all increased between ~3–25 fold, but no increase in TSLP mRNA was observed (Fig. 4A). Suppression of ROS production with NAC effectively blocked mRNA transcription, of each of these immune response mediators including TSLP (Fig. 4A). Similarly, replacement of extracellular media with nominally Ca²⁺-free media also inhibited transcription, except for TSLP and had only a minor inhibitory effect on the level of IL-8 mRNA. Furthermore, inhibition of nuclear NF-κB translocation (37) with Bay11-7082 inhibited ODE-stimulated transcription of IL-6 and IL-8 and reduced basal transcription of IL1-β and IL-33 (Fig 4B–E). Pretreatment with Go6976, an inhibitor PKCα/β activity (38), and delcasertib, a selective PKCδ inhibitor (20) each reduced basal and ODE-stimulated increases in transcription of IL-1β, IL33 and IL-6, (Fig. 4B–D) whereas ODE-stimulated transcription of IL-8 was reduced by ~50% (Fig. 4E).

Figure 4.

ODE exposure stimulated mRNA production of multiple cytokines/chemokines within two hours of exposure. A. Increased mRNA levels of TNFAIP6, IL1β, IL-6, IL-33 and IL-8 were inhibited following pretreatment with the ROS scavenger NAC (5 mM). Similarly, Ca²⁺-free media conditions inhibited the increase in TNFAIP6, IL1β, IL-6 and IL-33, and IL-8 mRNA levels. B. ODE stimulated an increase in IL1-β mRNA that was reduced below unstimulated levels with inhibitors of Ca²⁺-dependent PKCs (Go6976), the novel and Ca²⁺-independent PKCδ isoform (delcasertib) and the NF-κB inhibitor, BAY 11-7082 which blocks transcriptional activity and inflammatory signaling. C. ODE also stimulated IL-33 mRNA expression and pretreatment with Go6976, delcasertib and BAY 11-7082 inhibit expression to below baseline. D. ODE treatment increased IL-6 mRNA expression while pretreatment with the PKC inhibitors Go6976 and delcasertib blocked expression below basal levels while BAY 11-7082 blocked only the increase in mRNA produced by ODE but note the baseline level of expression. E. ODE stimulated IL-8 mRNA expression to the greatest extent relative to the cytokines measured in the present study and PKC inhibitors (Go6976 and delcasertib) similarly reduced the increase by ~40%, whereas BAY 11-7082 inhibited the increase IL-8 mRNA by ~72%. One-way ANOVAs followed by Tukey’s post-test where n = 4 were used for statistical analysis of data in panels A-E.

Follow-up ELISA assays revealed that pretreatment with Bay11-7082 inhibited ~30% of IL1-β and 35% of IL-33 secretion after 2 hours of ODE exposure (Fig. 5A–B). Moreover, 71% of IL-6 and 82% of IL-8 secretion were also blocked by pretreatment with Bay11-7082 (Fig. 5C–D). Inhibition of PKCα/β with Go6976 reduced IL1-β and IL-33 secretion by 66 and 69% respectively whereas ~90% inhibition was observed for IL-6 secretion (Fig. 5A–C). In contrast, the effect of Go6976 on IL-8 secretion was significantly lower (~24%) compared to each of the other three cytokines (IL-1b:IL-8 = p = 0.0004, IL-33:IL-8 = p = 0.0002 and IL-6:IL-8 = p < 0.0001) (Fig. 5D). Finally, reducing extracellular [Ca²⁺] by bathing cells in nominally Ca²⁺-free media also partially inhibited IL-1β (~81%), IL-33 (~63%), IL-6 (~57%) and IL-8 (~61%) secretion (Fig. 5A–D).

Figure 5.

ELISA assays confirmed the increase IL1-β, IL-33, IL-6, and IL-8 secretion after 2 hours of exposure to ODE. A. The amount of IL1-β after ODE exposure was reduced following pretreatment with Go6976 (~66%), delcasertib (44%) and BAY 11-7082 (30%), but removal of extracellular Ca²⁺ inhibited IL1-β to the greatest extent (~81%). B. ODE-stimulated IL-33 release was inhibited after pretreatment with Go6976 (69%), delcasertib (62%) and BAY 11-7082 (35%) and following removal of extracellular Ca²⁺ by 63%. C. The increase in IL-6 evoked by ODE was significantly blocked by Go6976 (90%), delcasertib (89%), BAY 11-7082, (71%) and by removal of extracellular Ca²⁺ (~57%). D. ODE increased levels of IL-8 and only a modest inhibition of the effect was observed with Go6976 (~24%) and delcasertib (~16%) while pretreatment with BAY 11-7082 (~82%) and removal of extracellular Ca²⁺ (61%) produced greater inhibition. One-way ANOVAs followed by Tukey’s post-test where n = 4 were used for statistical analysis of data in panels A-D.

Analysis of microbial DNA within ODE revealed that it was mainly comprised of gram-positive taxa belonging to the phylum Bacillota and its classes Clostridia and Bacilli (Fig. 6). Ten of the top 25 most abundant taxa at the ASV level were species of Clostridium with ~38% of taxa collapsed at the genus level belonging to Clostridium (Fig. 6A). Other prominent genera within ODE included Romboutsia (12%), Terrisporobacter (10%), Paraclostridium (5%), and Aerococcus (3%). The most abundant species in descending order are shown in Fig. 6B.

Figure 6.

Analysis of the bacterial DNA content of ODE (n = 3). A. Percent abundance of the major phyla. B. Percent abundance of Class representation. C. Percent abundance of Orders in ODE. D. Percent abundance of the major families. E. Percent abundance of bacterial genera identified within the ODE. B. Percent abundance of bacterial species.

Discussion

ODE-induced ROS production and alarmin secretion:

In an earlier study, swine barn dust extract was shown to induce ROS production (including generation of superoxide, hydrogen peroxide, nitric oxide, 8-isoprostane and hydroxyl radicals) in human airway epithelial cells (39). Similarly, poultry dust extracts stimulated ROS production, lipid peroxidation, and overall oxidative stress in airway epithelial cell lines (BEAS-2B cells, A549 cells) and in primary human bronchial epithelial cells (1, 12). The present study also demonstrated that ROS production was stimulated by ODE and provided additional kinetic data showing that ROS production steadily increased during the initial 10 minutes of exposure. Furthermore, heating the extract to 80°C for 10 minutes prior to cell treatment substantially reduced ROS generation, indicating that one or more heat-labile components (presumably proteins or enzymes subject to heat inactivation/denaturation) were involved in regulating the response. Although we were unable to find any direct evidence for ODE-dependent temperature sensitivity of ROS production in earlier investigations, Natarajan et al., (2019) showed that ODE-induced inhibition of surfactant protein (SP-A and SP-B) expression in H441 cells was heat labile (40). Moreover, in our previous studies of airway epithelial cell responses to Alternaria extract, we showed that like ODE, ROS production rapidly increased after exposure and that the response was nearly abolished after heating to 80°C for 10 minutes (17).

In our previous studies, we reported that Alternaria exposure triggered alarmin secretion, specifically the release of ATP and IL-33 and that the response was dependent on oxidative stress (16, 27). In the present investigation, ODE exposure failed to induce ATP secretion even though ODE treatment induced ROS production and oxidative stress. Furthermore, previous studies showed that Alternaria stimulated the release of mitochondrial and genomic DNA fragments, ranging in size from ~0.5–5K bp, yet hBE cells remained viable and capable of normal cell division (28). DNA release experiments involving treatment with ODE did not produce DNA release despite the earlier observation that the Alternaria DNA response was blocked by ROS scavengers. These findings indicate that ODE- and Alternaria-induced oxidative stress are fundamentally distinct, perhaps involving production of different ROS species or combination of species that differentially activate downstream signaling mechanisms linked to alarmin secretion.

ROS-dependent GPCR activation and Ca²⁺ signaling:

Shortly after ODE induction of ROS production, a rapid increase in YM254890-sensitive Ca²⁺ mobilization occurred that was inhibited by ROS scavengers and Nrf2 activation, providing compelling evidence for oxidative stress as critical for stimulation of Gq-dependent GPCR signaling. Further analysis of the ODE Ca²⁺ response showed that at least five distinct Gq-coupled GPCRs were activated and that treatment with enzyme inhibitors responsible for the endogenous production of histamine (α-FMH)), various leukotrienes (zileuton) and prostaglandins (indomethacin) each partially inhibited the overall increase in [Ca²⁺]_i. Furthermore, concentration-response experiments with agonists including histamine, LTB₄, LTC₄ and LTD₄ provided additional evidence for the expression and coupling of these GPCRs to Ca²⁺ mobilization in hBE cells. These results support our speculation that ODE-induced oxidative stress triggers the release of multiple inflammatory signaling molecules that induce autocrine activation of Gq-coupled GPCRs leading to mobilization intracellular Ca²⁺. Moreover, the results revealed that release of Ca²⁺ from internal stores (e.g., endoplasmic reticulum, ER) led to activation of Ca²⁺ uptake from the extracellular fluid, resulting in a secondary more sustained increase in [Ca²⁺]_i. We propose that the initial increase in [Ca²⁺]_I triggers cytokine/chemokine mRNA transcription and that the secondary, more prolonged rise in [Ca²⁺]_I may be necessary to sustain release into the extracellular media.

Analysis of the decrease in peak Ca²⁺ amplitude caused by DPH and α-FMH consistently indicated a partial autocrine histamine contribution to ODE-stimulated Ca²⁺ mobilization from internal stores. The shift in the DHP time to peak response suggested competitive antagonism at the H1 receptor, which would be expected to reduce receptor occupancy, slow PLCβ activation and IP₃ accumulation, and delay Ca²⁺ release from the ER (41–44). In contrast α-FMH acts upstream of the receptor by inhibiting histamine production and release, which would not be expected to alter the kinetics of Ca²⁺ release (45). Our observation that multiple leukotriene receptors along with the EP₁-R prostaglandin receptor EP₁-R were activated by ODE exposure highlights the complexity of ODE Ca²⁺ response. The rightward shifts in time-to-peak kinetics of the CysLT₁ antagonist montelukast and the shorter delay observed for the leukotriene synthesis inhibitor zileuton suggests that CysLT₁ substantially contributes to the early phase of Ca²⁺ mobilization from the ER. Blockade of CysLT₂ and BLT₁ receptors produced minor delays in kinetics, consistent with secondary effects on Ca²⁺ signal amplification. Inhibition of prostaglandin production with indomethacin and EP₁-R antagonism with SC53122 produced comparable reductions in [Ca²⁺]_i, without altering Ca²⁺ kinetics, indicating that prostaglandins serve primarily as signal modulators and are less involved with driving the early Ca²⁺ mobilization response to ODE. Another notable observation was that adding the decreases in amplitude of the ODE Ca²⁺ response produced by each of the five Gq-receptor GPCR antagonists amounts to a value greater than 100% (DPH, 25%; montelukast, 54%; BayCysLT2, 34%; LY293111, 41% and SC61322, 46%). This implies that these receptors do not make strictly independent contributions to the ODE Ca²⁺ response. Instead, ODE appears to activate an interconnected autocrine network where histamine, cysteinyl leukotrienes, LTB4 and prostaglandin signaling converge on a shared Gq-PLCβ-IP₃-activated Ca²⁺ release mechanism.

Mechanisms of histamine, leukotriene and prostaglandin release in response to ODE exposure

To our knowledge, there is no direct evidence for a histamine-specific transport mechanism in the apical membrane of hBE cells. However, histamine has been shown to be transported by poly-specific organic cation transporters in the brain (46), specifically OCT 2, OCT3 (SLC22) and the Plasma Membrane Monoamine Transporter (PMAT, SLC29). OCT 1 and OCT2 are expressed in the apical membrane of hBE cells where they function in the release of acetylcholine (37, 47); however direct evidence for histamine transport in these cells is lacking.

The most likely mechanism responsible for apical leukotriene and prostaglandin (LTB₄, LTC₄ and PGE₂) transport in hBE cells is the MRP4/ABCC4 organic anion transporter, which is apically expressed (46) and requires ATP to transport LTB₄, LTC₄ or PGE₂ (49, 50). In contrast to LTB₄ and LTC₄, LTD₄ is typically produced extracellularly by sequential cleavage of LTC₄ by ectoenzymes including γ-glutamyl transpeptidase (51). Thus, LTD₄ is regarded as an extracellular metabolite rather than an endogenously synthesized leukotriene requiring transport across the apical membrane. At this time, there appears to be no evidence for ROS or Ca²⁺-dependent regulation of OCT, PMAT or MRP4/ATCC4 transport function. However, γ-glutamyl transpeptidase (identified as γ-GT in Fig. 7) expression and activity increase in response to oxidative stress suggesting that the conversion of LTC₄ to LTD₄ may be amplified following ODE exposure (52).

Figure 7.

Proposed model highlighting G-protein receptor stimulated signaling pathways activated by ODE exposure. A detailed description is provided in the Conclusions section of the Discussion. Biorender was used to create Fig. 7.

Induction of inflammatory cytokine/chemokine production and secretion:

Earlier investigations of swine and poultry ODE exposure demonstrated that the induction of oxidative stress in human airway epithelial cells drives expression and secretion of multiple inflammatory cytokines and chemokines, primarily IL-8 and IL-6 through activation of NF-kB, STAT-3 and MAPK signaling pathways (1, 12, 53–55). Initial studies by Romberger et al. (2002) showed that hog barn dust extract triggered protein kinase C (PKC) activation followed by IL-6 and IL-8 secretion, establishing PKC as a critical component of the signaling cascade linking ODE exposure to epithelial innate immune responses (54). A subsequent study extended this observation by revealing that poultry dust extract activates PKC-dependent MAPK signaling, causing increases in NK-κB and AP-1 transcriptional activity that promoted IL-8 transcription and secretion by A549 and BEAS-2B lung epithelial cells (56). More recent investigations showed that poultry dust extract-induced ROS production acts upstream of inflammatory signaling such that antioxidant treatment suppressed PKC phosphorylation, NF-kB activation, and inflammatory gene expression. Corresponding studies further showed that poultry dust-induced ROS generation activated STAT3 signaling, which coupled oxidative stress to cytokine production, supporting the role of oxidant-driven kinase signaling in ODE-stimulated airway inflammation (1).

In the present study we provide direct measurements of ODE-evoked Ca²⁺ mobilization and show that the Ca²⁺ response is driven by ROS-induced secretion of specific inflammatory mediators (histamine, leukotrienes, prostaglandins) that activate Gq-coupled GPCR signaling, ultimately regulating cytokine and chemokine transcription and secretion. These findings extend previous work showing that mobilization of [Ca²⁺]_i was an early event in the activation of Ca²⁺-dependent PKCs (56–58). In addition, earlier investigations also showed that swine and poultry dust extracts contain proteases capable of activating PAR-1 and PAR-2 GPCRs expressed by airway epithelial cells (59–61). These Gq-coupled receptors stimulate PLCβ activity leading to IP₃ generation, Ca²⁺ release from internal stores and protease-driven kinase activity that promotes pro-inflammatory cytokine and chemokine expression and secretion (62). Based on strong evidence from these previous studies, we tested the effects of FSLLRY-NH₂, a selective peptide antagonist (63) to determine whether ODE in the present study was capable of activating PAR-2. Negative results were obtained at two antagonist concentrations sufficient to block PAR-2 receptors, suggesting that they were not involved in mediating Ca²⁺ mobilization and associated stimulation of innate immune signaling in hBE cells. This result was also consistent with our conclusion that ROS production functions as the primary trigger for autocrine, inflammatory mediator stimulated GPCR signaling that promotes cytokine/chemokine production and release by hBE cells.

Bacterial composition of ODE:

The lack of observable differences in bacterial composition of ODE from different sampling locations in the swine facility suggested that the bacteria found in ODE originated from a common source and were likely aerosolized prior to settling. A prior study in swine production systems found taxonomic parallels between air particulate samples, pig nasal microbiomes, and the nasal microbiomes of workers (64). Similarly, a prior study of swine facility air quality observed parallels between the microbes found in air samples and those common to the gastrointestinal tract of swine (9). Here, we observed many of the same taxa reported by these studies and others characterizing the typical gut microbiome of swine (65, 66), providing more evidence that the microorganisms found in pig gastrointestinal tracts are pervasive throughout the environment in which they live and the surrounding air.

The use of industry-standard cleaning techniques and disinfectants is known to reduce but not completely eliminate microorganisms from the environment (67–69). Microorganisms have a variety of mechanisms to survive common management practices used in swine production systems, including the formation of biofilms, resistance to desiccation, and antimicrobial resistance (70, 71). Although the top 25 taxa at the species level found in ODE were all gram-positive, this observation echoes that of prior work (9). A potential explanation for these observations may be the observed resistance of gram-positive bacteria, such as Clostridium disporicum and Aerococcus viridans found in ODE, to degradation and aerosolization (9, 72–74). While members of the genus Clostridium can be nonpathogenic and commensal in nature (75), taxa such as Clostridium perfringens are associated with animal and human disease (76). Additionally, Paraclostridium sordellii and Aerococcus viridans are known pathogens that can adversely affect human and animal health (74, 77). Therefore, the pervasiveness of these taxa within ODE found in swine facilities represents a potential concern for human health.

Conclusions:

The results of the present study demonstrate that ODE induces a complex, multidimensional response in human airway epithelial cells, largely driven by oxidative stress. ODE-induced reactive oxygen species (ROS) production acts as a central initiating event that triggers downstream signaling pathways that include Gq-coupled GPCR activation and intracellular calcium mobilization, ultimately promoting inflammatory cytokine and chemokine release. The study highlights important distinctions between ODE- and Alternaria-induced oxidative responses, suggesting that different ROS profiles lead to divergent cellular outcomes, particularly in alarmin secretion. Furthermore, although we cannot exclude the possibility of non-specific effects of ODE components on Ca²⁺ and/or cytokine signaling, our results support the conclusion that ODE components activate a highly interconnected signaling network in which multiple inflammatory mediators, particularly histamine, leukotrienes, and prostaglandins are simultaneously produced and released, collectively amplifying the cellular response. Instead of acting through isolated or linear pathways, these mediators converge on shared Gq-coupled GPCR signaling mechanisms that regulate intracellular calcium mobilization. Notably, the combined effects of antagonizing these pathways revealed substantial overlap and cross-talk between signaling routes. This coordinated interaction suggests that ODE does not trigger a single dominant pathway but instead induces a networked inflammatory program where multiple mediators reinforce one another, ultimately enhancing the strength and persistence of airway inflammatory signaling. Together, these findings provide key mechanistic insights into how organic dust exposure contributes to innate immune responses and ultimately, airway inflammation.

New and Noteworthy.

This study investigates how agricultural organic dust promotes airway inflammation. We show that organic dust extract rapidly induces ROS production along with a biphasic Ca²⁺ response through a redox-sensitive, Gq-coupled GPCR network involving histamine, leukotrienes, and prostaglandins. This signaling cascade activates PKC isoforms that trigger production and secretion of inflammatory cytokines, linking oxidative stress to Ca²⁺ dependent inflammatory responses by the airway epithelium following organic dust exposure.