Deregulation of Biologically Significant Genes and Associated Molecular Pathways in the Oral Epithelium of Electronic Cigarette Users

Abstract

We have investigated the regulation of genes and associated molecular pathways, genome-wide, in oral cells of electronic cigarette (e-cigs) users and cigarette smokers as compared to non-smokers. Interrogation of the oral transcriptome by RNA-sequencing (RNA-seq) analysis showed significant number of aberrantly expressed transcripts in both e-cig users (vapers) and smokers relative to non-smokers; however, smokers had ~50% more differentially expressed transcripts than vapers (1726 versus 1152). Whereas the deregulated transcripts in smokers were predominately from protein-coding genes (79% versus 53% in vapers), nearly 28% of the aberrantly expressed transcripts in vapers (versus 8% in smokers) belonged to regulatory non-coding RNAs, including long intergenic non-coding, antisense, small nucleolar and misc RNA (P < 0.0001). Molecular pathway and functional network analyses revealed that "cancer" was the top disease associated with the deregulated genes in both e-cig users and smokers (~62% versus 79%). Examination of the canonical pathways and networks modulated in either e-cig users or smokers identified the "Wnt/Ca⁺ pathway" in vapers and the "integrin signaling pathway" in smokers as the most affected pathways. Amongst the overlapping functional pathways impacted in both e-cig users and smokers, the "Rho family GTPases signaling pathway" was the top disrupted pathway, although the number of affected targets was three times higher in smokers than vapers. In conclusion, we observed deregulation of critically important genes and associated molecular pathways in the oral epithelium of vapers that bears both resemblances and differences with that of smokers. Our findings have significant implications for public health and tobacco regulatory science.

Keywords: RNA-seq; carcinogenesis; gene regulation; transcription; vaping.

Figure 1

Aberrantly expressed transcripts detected by RNA-sequencing (RNA-seq) in electronic cigarette (e-cig) users and smokers as compared to controls. (a) Numbers of up-regulated and down-regulated transcripts in e-cig users and smokers are indicated. Fold-change: >1.5; P < 0.005. (b) Venn diagram of deregulated transcripts in e-cig users and smokers is shown.

Figure 2

Functional pathway analysis of differentially expressed genes in e-cig users and smokers relative to controls by IPA^®. In both groups, differentially expressed genes are predominantly associated with “cancer”. The “Wnt/Ca⁺ pathway” in e-cig users and the “integrin signaling pathway” in smokers are the most affected pathways.

Figure 3

Categories of the aberrant transcripts detected in e-cig users and smokers relative to controls by IPA^®. High proportions of the aberrantly transcribed DNA sequences in e-cig users and smokers, respectively, belong to regulatory non-coding RNAs and protein-coding genes.

Figure 4

Common canonical pathways disrupted in e-cig users and smokers relative to controls by IPA^®. (a) The canonical pathway heatmap visualizes the affected canonical pathways simultaneously in e-cig users and smokers, allowing a direct comparison between the two groups. (b) Predicting how up-regulated and down-regulated genes in the datasets (red and green nodes, respectively, on the pathway) can affect the activity of other molecules on the pathway. Orange nodes, prediction of activation; blue nodes, prediction of inhibition.

Figure 5

Gene ontology analysis of the differentially expressed genes in e-cig users and smokers relative to controls by Database for Annotation, Visualization and Integrated Discovery (DAVID). The top ten functional clusters in e-cig users and smokers are shown. Enrichment scores are indicated in parentheses.

Figure 6

Validation of RNA-seq gene-expression data by RT-qPCR. Transcription levels of several up- and down-regulated targets identified by RNA-seq in e-cig users and smokers were validated by RT-qPCR. Results are expressed as the median normalized expression levels + Standard Error (SE).

Figure 7

Validation of smoking/vaping status. Exhaled breath was measured in e-cig users, smokers, and controls (n = 18, 25, and 30, respectively) by a Bedfont Micro^+TM Smokerlyzer^® Breath CO monitor according to the instructions of the manufacturer (Bedfont Scientific Ltd., Kent, UK). Whisker box plots display distributions of carbon monoxide (CO) levels (parts per million, ppm) and percentage of carboxyhemoglobin (%COHb) in the respective groups.