Functional characterization of OR51B5 and OR1G1 in human lung epithelial cells as potential drug targets for non-type 2 lung diseases

Abstract

Background: Hypersensitivity to odorants like perfumes can induce or promote asthma with non-type 2 inflammation for which therapeutic options are limited. Cell death of primary bronchial epithelial cells (PBECs) and the release of the pro-inflammatory cytokines interleukin-6 (IL-6) and IL-8 are key in the pathogenesis. Extra-nasal olfactory receptors (ORs) can influence cellular processes involved in asthma. This study investigated the utility of ORs in epithelial cells as potential drug targets in this context.

Methods: We used the A549 cell line and primary bronchial epithelial cells using air-liquid interface culture system (ALI-PBECs). OR expression was investigated by RT-PCR, Western blot, and Immunofluorescence. Effects of OR activation by specific ligands on intracellular calcium concentration, cAMP, Phospholipase C (PLC), cell viability, and IL-6 and IL-8 secretion were analyzed by calcium imaging, enzyme immunoassays, Annexin V/ propidium iodide -based fluorescence-activated cell staining or by ELISA, respectively.

Results: By screening A549 cells, the OR51B5 agonists Farnesol and Isononyl Alcohol and the OR1G1 agonist Nonanal increased intracellular Ca2 + . OR51B5 and OR1G1 mRNAs and proteins were detected. Both receptors showed a preferential intracellular localization. OR51B5- but not OR1G1-induced Ca2 + dependent on both cAMP and PLC signaling. Farnesol, Isononyl Alcohol, and Nonanal, all reduced cell viability and induced IL-8 and IL-6 release. The data were verified in ALI-PBECs.

Conclusion: ORs in the lung epithelium might be involved in airway-sensitivity to odorants. Their antagonism could represent a promising strategy in treatment of odorant-induced asthma with non-type 2 inflammation.

Keywords: Inflammation; Lung epithelial cell; Odorant sensitivity; Olfactory receptor; Therapeutic targets.

Fig. 1

Farnesol, Isononyl alcohol and Nonanal stimulated calcium mobilization in A549 cells. A549 cells were loaded for 30 min with Fura 2-AM followed by measurement of response rate of Ca2 + influx (f 340nm/f 380nm); ratio of responsive cells to the total number of cells, by three times repetitive application of (A) Farnesol (300 µM), (B) Isononyl alcohol (1 mM) and (C) Nonanal (300 µM) compared to the vehicle (DMSO, 0.1%) as indicated. ATP (100 µM) was used as a positive control and was applied after the repetitive addition of the ligands to ensure cell viability. The bars in the graphs on the left side indicate the stimulus duration. Data are shown as mean ± SEM, Technical replicates (TN) = 9. Statistical comparison utilized paired student's t-test referring to the response rate induced by odorant to the DMSO **p ≤ 0.01 and ***p ≤ 0.001

Fig. 2

Detection of OR51B5 and OR1G1 expression in A549 cells. RT-PCR demonstrated the presence of (A) OR51B5 PCR fragment at 227 bp and (B) OR1G1 at 203 bp in A549. (C) mRNA transcripts of OR1A1 and OR10S1 were not detected but the genomic DNA was detected for both. The expression of the corresponding proteins was confirmed by (D, E) Western Blot at the estimated molecular mass of about 35 kDa and (F, G) immunofluorescence staining, specific antibodies were used against the ORs (red) and the plasma membrane marker E-cadherin (green). The nucleus was stained using 4′,6-Diamidin-2-phenylindol (Dapi; blue). The overlay between the OR (red) and nucleus (blue) staining indicated as violet color in the merged picture. Scale bars: 50 μm and the squares delineate the magnified region. Representative of TN = 3 individual experiments

Fig. 3

Farnesol-Isononyl alcohol (OR51B5 agonists)-triggered calcium mobilization is mediated by both cAMP/PLC. A549 cells were stimulated each measurement three times with (A-C) Farnesol (300 µM) or (D, E) Isononyl alcohol (1mM) in presence or absence of the inhibitors; (A, D) extracellular calcium chelator (EGTA 10mM), (B) Adenylate cyclase inhibitor (SQ 22536 10 µM) or (C, E) PLC inhibitor (U73122, 2 µM) in the second application, inhibitors were pre-incubated for three minutes before adding odorants in the second application. The bars of all experiments indicate the stimulus duration. Data are shown as mean ± SEM, TN = 7–9. Statistical comparison utilized paired student's t-test referring to the ratio of calcium amplitude between second to first applications in the experimental group (odorants-inhibitors application) vs. the corresponding control group (odorants alone). **p ≤ 0.01; ***p ≤ 0.001

Fig. 4

Nonanal, OR1G1 agonist, induced intracellular calcium increase is not from extracellular Ca2 + or intracellular ER. A549 cells were stimulated each measurement three times with Nonanal (300 µM) in presence or absence of (A) (EGTA 10mM), (B) endoplasmic reticulum Ca2 + ATPase inhibitor (Thapsigargin, 5 µM) or (C) PLC inhibitor (U73122, 2 µM) in the second application, inhibitors were also pre-incubated for three minutes before adding odorants. The bars on graphs of the left side indicate the stimulus duration. Data are shown as mean ± SEM, TN = 6–9. Statistical comparison utilized paired student's t-test referring to the ratio of calcium amplitude between second to first applications in the experimental group (odorants-inhibitors application) vs the corresponding control group (odorants alone). ns = non-significant

Fig. 5

Farnesol but not Nonanal increases the intracellular cAMP and PLC levels. A549 were stimulated with (A, C) Farnesol, (B, D) Nonanal or DMSO solvent for 20 min. Concentrations are indicated in the graph. Forskolin and Carbachol are positive controls in cAMP and PLC assay experiments respectively. Data are shown as mean ± SEM, TN = 4–5. Relative changes in (A, B) cAMP or (C, D) PLC vs. DMSO 0.1% (negative control) were analyzed by using One-way ANOVA test with post hoc “Two-stage-up method of Benjamini, Krieger, and Yekutieli,”. *p ≤ 0.05 and **p ≤ 0.01 and *** p ≤ 0.001. ns = non-significant

Fig. 6

Stimulation of OR51B5 and OR1G1 reduces A549 cell viability. A549 cells were exposed for 24h to increasing concentrations of (A, B) Farnesol, (C, D) Isononyl alcohol or (E, F) Nonanal or solvent DMSO (0.1%). Proportions of viable, apoptotic, and necrotic cells were determined by flow cytometry. Relative cell counts were calculated in comparison to the total cell population. Cells staining positive for Annexin V (AV +) were indicative of early apoptosis and were denoted in the lower right quadrant of the chart (R3), represented by the dark grey portion in the accompanying graphs. AV/propidium iodide (PI) double-positive staining, (AV + , PI +), signified necrotic cells and was in the upper right quadrant of the chart (R2), depicted by the light grey section of the graphs. Viable cells, defined as those unstained by AV and PI (AV − , PI −), were found in the lower left quadrant of the chart and were represented by the black bars in the graphs. Data are shown as mean ± SEM, TN = 5–7. Statistical significance vs. DMSO (0.1%) utilized One-way ANOVA. test with post hoc “Two-stage-up method of Benjamini, Krieger, and Yekutieli,”. *p ≤ 0.05, **p ≤ 0.01 and ***p ≤ 0.001

Fig. 7

Farnesol, Isononyl alcohol and Nonanal induced IL-8 and IL-6 secretion. A549 cells were treated for 24 h with increasing doses of the odorants; (A) Farnesol, (B) Isononyl alcohol, (C) Nonanal or DMSO solvent. IL-6 experiments cells were stimulated with (D-F) the same odorants, solvent ± LPS (1 µg/ml). Concentrations are indicated in the graphs. IL-8 and IL-6 in cell culture supernatants was quantified by commercial ELISA. Data were normalized to solvent controls (DMSO) and shown as mean ± SEM. Data significance of agonists vs. (A-C) DMSO alone or (D-F) the odorant/vehicle ± LPS was analyzed using One-way ANOVA test with post hoc “Two-stage-up method of Benjamini, Krieger, and Yekutieli,”. *p ≤ 0.05, **p ≤ 0.01 and ***p ≤ 0.001. TN = 5–15

Fig. 8

ɣ-Decalactone and 2-ethyl-1-hexanol induced IL-8 and IL-6 secretion. A549 cells were stimulated with ɣ-Decalactone (A, C) or 2-ethyl-1-hexanol (B, D), in the presence or absence of LPS, for 24 h with concentrations as indicated in the graphs. IL-8 and IL-6 were measured in culture supernatants by enzyme-linked immunosorbent assay (ELISA). Data have been normalized to solvent controls (DMSO) and are presented as mean ± SEM. TN = 6. Data significance was analyzed using One-way ANOVA test with post hoc “Two-stage-up method of Benjamini, Krieger, and Yekutieli,”. *p ≤ 0.05 and **p ≤ 0.01. ns = non-significant

Fig. 9

IL-8 induction by Farnesol and Isononyl alcohol is mediated by cAMP. SQ22536 was applied 30 min before stimulation of cells with Farnesol (100µM), Isononyl alcohol (500µM) or DMSO solvent for 24 h.IL-8 concentrations were determined in culture supernatants by ELISA. All data are normalized to solvent controls (DMSO) and presented as mean ± SEM. TN = 6. Data significance of odorants vs. DMSO or the odorant/SQ22536 was analyzed using One-way ANOVA test with post hoc “Two-stage-up method of Benjamini, Krieger, and Yekutieli,”. **p ≤ 0.01

Fig. 10

Expression and functional characterization of OR51B5 and OR1G1 ALI-cultured primary epithelial cell. RT-PCR revealed the presence of mRNA transcripts of (A) OR51B5 and (B) OR1G1. Validation of (C) OR51B5 and (D) OR1G1 protein expression by Western Blot. Evident bands for both receptors at the expected size of 35 kDa are shown. Representative of 3 donors. ALI-PBECs were stimulated with increasing concentrations of (E) Farnesol, (F) Isononyl alcohol or (G) Nonanal or solvent DMSO (0.1%). Proportions of viable, apoptotic, and necrotic cells were determined by flow cytometry using AV/ PI staining. Percentage of (E–G) necrotic cells were calculated versus total cell count of 5 individual experiments from five donors. ALI-PBECs were stimulated for 6 h with (H, K) Farnesol, (I, L) Isononyl alcohol and (J, M) Nonanal. After 6 h media was removed from apical to preserve cell integrity. Supernatant was collected from the basal part after 24 h. Concentrations are indicated in the graphs. IL-8 and IL-6 were measured by ELISA. Experiments were conducted with TN = 6–10 independent samples from six to tent different donors. Data were normalized to solvent controls (DMSO). (E-M) data are shown as mean ± SEM. Statistical significance vs. DMSO (0.1%) utilized Friedman test with post hoc “Two-stage-up method of Benjamini, Krieger, and Yekutieli,”. *p ≤ 0.05, **p ≤ 0.01 and ***p ≤ 0.001, ns = non-significant