Toxic cultures: e-cigarettes and the oral microbial exposome

Abstract

We tested the hypothesis that e-cigarette aerosol is metabolized by the indigenous oral microbiome, leading to structural and functional alterations. We combined untargeted metabolomics of in vitro commensal-rich and pathogen-rich biofilms with metatranscriptomics and fluorescent microscopy and verified the results in human samples. Spectral deconvolution of 4215 peaks identified 969 exposomal and endogenous metabolites that mapped to 23 metabolic pathways. The metabolites clustered by both aerosol characteristics and biofilm composition; and several were verified in human saliva of vapers. E-cigarette exposure upregulated xenobiotic degradation, capsule, peptidoglycan biosynthesis, organic carbon-compound metabolism, antimicrobial resistance, and secretion systems. E-cigarette exposure also altered biofilm architecture characterized by low surface-area to biovolume ratio, high biomass, and diffusion distance. In conclusion, our data suggest that bacterial metabolism of e-cigarette aerosol triggers a quorum-sensing-regulated stress response which mediates the formation of dense, exopolysaccharide-rich biofilms in health-compatible communities and antibiotic resistance and virulence amplification in disease-associated communities.

Fig. 1. Metabolite diversity and generation time.

Principal Components analysis (PCA) of the metabolites generated over 8 h are shown in (A). Data was acquired through untargeted one-dimensional (1D) 1H NMR and each bucketed region was normalized to the total sum of the spectral intensities prior to analysis. All samples were run in triplicates, and the experiments were duplicated. Only peaks that exceeded the zero threshold were inducted into analysis. The majority of the metabolites were generated within one hour of exposure to e-cigarette vapor (p = 0.008, ANOSIM). PCA also revealed nicotine concentration (0 mg versus 6 mg) and biofilm diversity as sources of variation along with (B). Kernel density plots of Chao and Shannon indices (C, D respectively) demonstrated the significant differences based on type of aerosol (p = 0.012, Wilcoxon signed rank test).

Fig. 2. Microbial community composition is also a determinant of metabolite profiles.

Univariate analysis of nicotine-free and nicotine plus aerosol using partial least squares discriminant analysis (PLSDA) demonstrated that 39.1% of the variation between biofilms was explained by component 1 following exposure to nicotine-free ENDS (Q² = 0.75525, R² = 0.80226), while exposure to nicotine-containing vapor accounted for 37.6% of the variation of component 1 (Q² = 0. 73132, R² = 0. 79832, A, B). indicating that biofilm diversity was also a robust determinant of metabolite profile. A machine-learning algorithm trained on the metabolite profiles identified 15 compounds as discriminants of biofilm composition and vapor type (C). Covariate analysis using two-way ANOVA demonstrated a significant difference in 13 metabolites based on biofilm exposure alone, 17 metabolites for ENDS exposure alone, and 17 metabolites for the interaction between biofilm and ENDS exposures (p < 0.05, ANOVA, D). All samples were run in triplicates, and the experiments were duplicated. Only peaks that exceeded the zero threshold were inducted into analysis. Data supporting this figure can be found in Supplemental File 2.

Fig. 3. Pathway enrichment analysis identifies key biological functions impacted by vapor exposure.

A–C compares pathways that are upregulated in commensal-rich, intermediate, and pathogen-rich communities following exposure to nicotine-free and nicotine plus e-cigarette vapor. D–F compares the pathways upregulated in the three biofilms by nicotine-free vapor and G–I compares the pathways upregulated in the three biofilms by nicotine-plus vapor. All samples were run in triplicates, and the experiments were duplicated. Only peaks that exceeded the zero threshold were inducted into analysis.

Fig. 4. E-cigarettes induce quorum-sensing regulated gene expression in oral biofilms.

Heatmap of KEGG orthologs that were upregulated following exposure to nicotine-free or nicotine plus vapor or clean-air is shown in (A). All samples were run in triplicates, and the experiments were duplicated. Data supporting (A) can be found in Supplemental Table 4. Co-occurrence networks between metabolites and microbial transcripts in each group are shown in (B–G). Commensal-rich biofilms are shown in (B, E), Intermediate in (C, F), and pathogen-rich in (D, G). Nicotine-free vapor exposure is represented in (B–D) and nicotine-plus vapor exposure is shown in (E–G). Each network graph contains nodes (circles) and edges (lines). Nodes represent metabolites (pink) and KEGG-annotated transcripts (blue), and edges represent Spearman’s rho. Edges are colored green for positive correlation and red for negative correlation. Only significant correlations (P < 0.05, t test) with a coefficient of at least 0.80 are shown.

Fig. 5. E-cigarette exposure alters biofilm topography.

Representative confocal images (selected randomly from those used in the analysis) of commensal-rich biofilms consisting of S. oralis, S. sanguis, S. mitis, A. naeslundii, N. mucosa, and V. parvula, intermediate biofilms (including intermediate colonizer (F. nucleatum) to the aforementioned species) and pathogen-rich biofilms (intermediate biofilms that were further colonized by P. gingivalis, F. alocis, Selenomas sputigena, S. noxia, C. gracilis, P. intermedia, P. micra, and T. forsythia) following exposure to nicotine-free or nicotine-plus to e-cigarette vapor and clean air controls are shown in a(i–iii), b(i–iii), c(i–iii). The change in biofilm area following exposure to nicotine-free and nicotine-plus vapor over 8 h are shown in d(i, ii) respectively. Surface area to biovolume ratio of dead and live cells are shown in e(i, ii) respectively, average biomass in e(iii) and diffusion distance in e(iv). Biofilms were visualized using confocal laser scanning microscopy in (B), and surface area and volume were computed with IMARIS. In all figures, groups connected by the same symbol are significantly different (P < 0.001, Dunn’s test with joint ranks).

Fig. 6. Salivary metabolome profiles of e-cigarette users recapitulate metabolism of nicotine-plus vapor by pathogen-rich biofilms.

Partial least squares discriminant analysis (PLSDA) revealed significant separation between the metabolomic profiles of pure e-cigarette users when compared to dual and former smokers (Q2 = 0. 45781, R2 = 0. 51411).