Transcriptomic and barrier responses of human airway epithelial cells exposed to cannabis smoke

Abstract

Globally, many jurisdictions are legalizing or decriminalizing cannabis, creating a potential public health issue that would benefit from experimental evidence to inform policy, government regulations, and user practices. Tobacco smoke exposure science has created a body of knowledge that demonstrates the conclusive negative impacts on respiratory health; similar knowledge remains to be established for cannabis. To address this unmet need, we performed in vitro functional and transcriptomic experiments with a human airway epithelial cell line (Calu-3) exposed to cannabis smoke, with tobacco smoke as a positive control. Demonstrating the validity of our in vitro model, tobacco smoke induced gene expression profiles that were significantly correlated with gene expression profiles from published tobacco exposure datasets from bronchial brushings and primary human airway epithelial cell cultures. Applying our model to cannabis smoke, we demonstrate that cannabis smoke induced functional and transcriptional responses that overlapped with tobacco smoke. Ontology and pathway analysis revealed that cannabis smoke induced DNA replication and oxidative stress responses. Functionally, cannabis smoke impaired epithelial cell barrier function, antiviral responses, and increased inflammatory mediator production. Our study reveals striking similarities between cannabis and tobacco smoke exposure on impairing barrier function, suppressing antiviral pathways, potentiating of pro-inflammatory mediators, and inducing oncogenic and oxidative stress gene expression signatures. Collectively our data suggest that cannabis smoke exposure is not innocuous and may possess many of the deleterious properties of tobacco smoke, warranting additional studies to support public policy, government regulations, and user practices.

Keywords: Calu-3 cells; Marijuana; Transepithelial electrical resistance; interferon stimulated genes; oxidative stress.

Figure 1

Cannabis and tobacco smoke extract‐induced transcriptomic changes. Correlation of differential gene expression profiles (log₂FoldChange(FC)) versus control, comparing cannabis and tobacco exposures, as determined by Pearson’s correlation. Significantly differentially expressed genes in cannabis smoke extract (CSE) only (blue diamonds, n = 832), tobacco smoke extract (TSE) only (red dots, n = 190), and shared between CSE and TSE (purple triangles, n = 389) are highlighted. Only seven genes were identified as significantly differentially expressed between CSE and TSE (orange squares, n = 7).

Figure 2

Transcriptomic correlations of in vitro tobacco smoke exposure in Calu‐3 cells to human smokers and primary airway epithelial cells. Correlation of Calu‐3 cell differential gene expression profile (log2FoldChange (FC)) following exposure to tobacco smoke extract (TSE) and differential gene expression profiles between healthy controls and lifetime smokers in the publicly available datasets (A) GSE4498 and (B) GSE11784. Genes that are identified in both datasets and exhibit statistically significant differences in expression are included. Correlations with differential gene expression profiles from primary human airway epithelial cells grown under air‐liquid interface culture conditions and exposed to main‐stream tobacco smoke using the publicly available datasets (C) SRP096285 and (D) SRP126155.

Figure 3

Impact of cannabis smoke exposure on epithelial cell barrier function and transcriptomic profile. Genes involved in (A) airway epithelial repair and remodeling or epithelial junctions were curated from the RNA‐sequencing dataset performed at 10% CSE or TSE. The expression for each gene is presented for all experimental replicates, with expression for each replicate being scaled by the gene. Calu‐3 cells were exposed to increasing concentrations of CSE (orange) or TSE (blue) (control, 0.625%, 1.25%, 2.5%, 5%, 10%, and 20%) for 24h with outcome measurements of (B) cell viability assessed by lactate dehydrogenase (LDH) assay, (C) transepithelial electrical resistance – TEER (ohms*cm²), (D) transforming growth factor‐alpha (TGF‐α) (pg/mL), and (E) platelet derived growth factor‐AA (PDGF‐AA) (pg/mL). *=P < 0.05 relative to control untreated ‐ Tukey HSD. Error bars represent standard deviation (n = 4).

Figure 4

Impact of cannabis smoke exposure on epithelial cell antiviral and inflammatory expression and transcriptomic profile. Genes involved in (A) airway epithelial anti‐viral, pro‐inflammatory, or neutrophil mediated immunity were curated from the RNA‐sequencing dataset performed at 10% CSE or TSE. The expression for each gene is presented for all experimental replicates, with expression for each replicate being scaled by the gene. Calu‐3 cells were exposed to increasing concentrations of CSE (orange) or TSE (blue) (control, 0.625%, 1.25%, 2.5%, 5%, 10%, and 20%) for 24h with outcome measurements of (B) interferon gamma induced protein‐10 (CXCL10) (pg/mL), (C) regulated on activation, normal T cell expressed and secreted (CCL5) (pg/mL), (D) IL‐8 (pg/mL), and (E) IL‐6 (pg/mL). *=P < 0.05 relative to control untreated ‐ Tukey HSD. Error bars represent standard deviation (n = 4).