Toxic cultures: E-cigarettes and the oral microbial exposome

doi:10.21203/rs.3.rs-4629512/v1

Download PDF

Article

Toxic cultures: E-cigarettes and the oral microbial exposome

https://doi.org/10.21203/rs.3.rs-4629512/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

E-cigarettes have emerged as an exposomal factor of great concern to human health. We aimed to test the hypothesis that e-cigarette aerosol is metabolized in the oral cavity by the indigenous microbiome, leading to structural and functional alterations in oral biofilms. We combined untargeted metabolomic analysis of in vitro commensal-rich and pathogen-rich microcosm communities with metatranscriptomics, and fluorescent microscopy, and verified the results in human samples. Spectral deconvolution of 4,215 peaks identified 969 exposomal and endogenous metabolites that mapped to 23 metabolic pathways. Aerosol characteristics and biofilm composition affected metabolite profiles. Metabolites generated by commensal-rich biofilms contained antimitic, anti-fungal and anti-bacterial compounds, while pathogen-rich biofilms metabolized nicotine-containing aerosol using the pyridine and pyrrolidine pathways. Both communities generated endogenous metabolites that mapped to quorum sensing functions. Several of these metabolites were verified in the saliva of current, never, and former smokers who vape. Metatranscriptomics revealed upregulation of xenobiotic degradation, capsule, peptidoglycan, and glycosaminoglycan biosynthesis in commensal-rich communities, while genes encoding organic carbon-compound metabolism, antimicrobial resistance and secretion systems were over-expressed in pathogen-rich biofilms. Topographical analysis revealed an architecture characterized by low surface-area to biovolume ratio, high biomass, and diffusion distance only in commensal-rich biofilms. 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. These findings explain the higher incidence of dental caries, gingival inflammation, and antimicrobial resistance observed in vapers.

Biological sciences/Microbiology/Microbial communities/Microbial ecology

Health sciences/Health care/Dentistry/Dental conditions/Plaque

In common with all living organisms, the human body is responsive to the surrounding environment; an ideology that led to a new field of study in 2005: Exposomal Biology¹. The environmental exposures that humans experience over their lifetime can be either internal, specific external and general external². Anthropogenic or lifestyle-based influences are examples of specific external exposures; and prominent among these are diet, tobacco and alcohol².

One such exposomal factor that has become a concern in recent times is the electronic cigarette (E-cig), because of its potential to have an impact from a very early age³. Also known as an Electronic Nicotine Delivery System (ENDS), it is a battery-operated device that heats up a mixture of humectant, flavoring agents and/or nicotine under high pressure and temperature to create an aerosol that is then inhaled through the mouth. These devices entered the US market a little over 15 years ago and have captured a large market share within this short period of time. Indeed, the CDC reports a 46.6% increase in sales and a 46% rise in the number of brands (from 184 to 269) between January 2020 and December 2022⁴. The National Youth Tobacco Survey of 2022 found that 2.5 million school children have used these devices, most of them on a daily basis, exponentially increasing the risk of lifetime nicotine addiction and creating a gateway to other forms of addiction⁵.

Emerging studies, as summarized in the Review of the Health Effects of Electronic Nicotine Delivery Systems by the National Academies of Sciences, Engineering, and Medicine (NASEM) Committee⁶, provide substantial evidence that e-cigarettes influence the pathophysiology of human diseases in several ways, for example, by promoting endothelial dysfunction and oxidative stress. Data from human studies also catalog their deleterious effects on the cardiovascular and respiratory systems⁷.

The oral cavity is the first point of contact, and therefore, the first organ system to be impacted by this aerosol. We have previously demonstrated that e-cigarette use is associated with higher virulence signatures and a brisk proinflammatory signal in the oral cavity of clinically healthy e-cigarette users⁸. However, the mechanisms by which e-cigarettes create this pathogen-enriched ecosystem are not known.

Man is a holobiont⁹, and therefore, the exposome has to be interrogated not only through the lens of adaptive/maladaptive changes in human cells, but also by quantifying changes in the microbiomes that we host. While this is conceptually attractive, the complexity of biological systems as well as the multiplicity of exposures over the lifetime create seemingly unsurmountable challenges to unraveling these interactions. However, the advent of high-dimension biology has enhanced our ability to identify the compounds that we are exposed to and the metabolic profiles that result from these exposures. Indeed, by mapping changes at the DNA, RNA, protein, or metabolite level in response to a specific environmental exposure, it is possible to define an “exposotype” and to explicate its molecular underpinnings.

In the present study, we hypothesized that bacteria metabolize e-cigarette aerosol and that these metabolites play important roles in altering community structure, function, and topography. We aimed to test this hypothesis using a combinatorial enumeration of transcriptional events and metabolic activity, quantification of biofilm topography and verification of the in vitro findings in the human exposome.

E-cigarette aerosol contains several toxins and compounds

We began our analysis by investigating the chemical composition of a commercial brand of tobacco-flavored nicotine-containing and nicotine-free e-cigarette aerosol using GCMS. Interestingly, both nicotine-free and nicotine-containing aerosol contained similar numbers of compounds (313 and 308 respectively), but less than half (188) of the compounds were common to both aerosols (Supplemental Table 1), indicating a large difference between the constituents. Notable among these common compounds were paraldehyde, acetyl chloride, allyl acetate, anabasine, dimethylphosphine, diacetyl sulphide, diglycerol, dimethyl sulfoxide, ethylhydroxylamine, erythritol, fluoro-acetic acid, glyceraldehyde, glycerin, glycol, nitro-methane, phosgene, propylene glycol, trimethylpropoxy silane, thioacetic acid, trimethylphosphine, thiodiglycol, and xylitol. Additionally, among the predominant compounds identified in nicotine-free aerosol were esters and salts of hexanoic and octanoic (caprylic acid) and propionic acid, while minor tobacco alkaloids (nornicotine, anabasine, cotinine, nicotine nitriles) and compounds containing butane and silanes were more frequently identified in nicotine-containing aerosol.

Oral bacteria metabolize e-cigarette aerosol

We then investigated the metabolic byproducts produced by commensal-rich, intermediate, and pathogen-rich biofilms when exposed to these nicotine-containing or nicotine-free e-cigarette aerosols, or to clean-air (control). Overall, 4,474 peaks were generated and 4,215 were identifiable beyond a zero threshold. Spectral deconvolution and annotation to the molecular level identified 969 unique metabolites and 23 pathways corresponding to these metabolites after subtracting those identified in the clean-air and no-biofilm control groups (Supplemental Table 2).

Since metabolites can also be products of normal bacterial cell cycle, we investigated the percentage of naturally occurring metabolites versus those that are part of the human exposome. Over 82% of the metabolites that were identifiable to a molecular formula belonged to the category of human exposomes based on the Human Metabolite Database¹⁰. This category was also the most numerically abundant class of metabolites identified. A large fraction (27%) of the exposome family of compounds mapped to antimitic, anti-fungal and anti-bacterial agents.

Interestingly, the predominant bacterial metabolites were quorum sensing molecules and dipeptides, pointing to significant communications among the gram-positive and gram-negative bacteria in the multi-species biofilms.

To gain insights into the sources of variability in the e-cigarette metabolome, we used principal components analysis (PCA) on variance-stabilized abundances of peaks. PCA revealed nicotine concentration (0mg versus 6mg) as a source of variation along with biofilm diversity (Fig. 1A). Corroborating this, the Chao and Shannon indices demonstrated the greatest level of metabolite diversity when all biofilms were exposed to nicotine-containing aerosol (Figs. 1B and 1C). Furthermore, although enrichment analysis identified 879 metabolites common to both nicotine-free and nicotine-containing aerosol (Supplemental Table 2), DESeq revealed statistically significant differences in abundances between 95 of these common metabolites (Supplemental Table 3). Moreover, 887 metabolites were unique to nicotine-containing aerosol and 1027 metabolites were unique to nicotine-free aerosol. Time series analysis revealed that the majority of metabolites and compounds were generated within one hour of exposure to e-cigarette aerosol (Fig. 1D).

Byproducts of e-cigarette metabolism depend on bacterial community composition

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, while exposure to nicotine-containing aerosol accounted for 37.6% of the variation of component 1 (Figs. 2A and 2B), indicating that biofilm diversity was also a robust determinant of metabolite profile. Corroborating this, covariate analysis demonstrated a significant interaction between biofilm diversity and aerosol composition (Fig. 2C). Additionally, spectral deconvolution revealed that 423 metabolites were generated from nicotine-free aerosol when exposed to commensal-rich biofilms, while intermediate biofilms generated 505 metabolites and pathogen-rich generated 566 (Supplemental Table 2). Similarly, following exposure to nicotine-containing aerosol, 380, 436 and 597 compounds were generated by commensal-rich, intermediate, and pathogen-rich biofilms. Exposure to commensal-biofilms generated significantly higher levels of metabolites that mapped to antimitic, anti-fungal and anti-bacterial agents. However, pathogen-rich biofilms were exposed to nicotine-containing aerosol generated significantly greater amounts of compounds related to the pyridine and pyrrolidine pathways of nicotine metabolism (notably, 2-ketoglutaramate, 6-hydroxy-N-methylmyosmine, and ethyl 4-(acetylthio)butyrate) than commensal-rich biofilms. A machine learning algorithm trained on this dataset identified 15 compounds that were able to discriminate between biofilm diversity and aerosol composition (Fig. 2D). Notable among these were homo-serine-lactone, pyrollidine and dopamine, which showed high out-of-box (OOB) prediction for nicotine-containing aerosol and pathogen-rich biofilms.

Metabolite pathway enrichment analysis corroborated our findings that pathways regulated by these metabolites depend on biofilm diversity. For instance, the compounds generated from nicotine-free and nicotine-containing aerosol by commensal biofilms mapped to significantly fewer metabolic pathways when compared to those produced by intermediate and pathogen-rich biofilms (Figs. 3A-C). Globally, commensal-rich biofilms were significantly enriched in pathways of lipid, carbohydrate and energy metabolism, xenobiotic degradation and intermediate metabolites. E-cig metabolism by intermediary biofilms impacted metabolism of nucleotides, beta-alanine, caffeine, riboflavine, tyrosine, vancomycin, alanine, carbohydrate (fructose, galactose, ketone bodies), lipid and proteins, steroid biosynthesis, and glycan biosynthesis and degradation. Pathogen-rich biofilm metabolism demonstrated enrichment of pathways related to biosynthesis of flavenoids, indoles, ascorbates, ketones, proteo-glycans, steroid, inositol and pyridine alkaloids, and the pyrrolidine pathway, among others. Of note, several oligopeptide metabolites were also enriched, pointing to enrichment of proteolysis-related pathways. However, lack of a peptide pathway database precluded their inclusion into the pathway mapping.

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

Since several peaks identified in the study mapped to bacterial quorum sensing molecules; and compounds and metabolites such as glycerol, acetaldehydes, glycol etc. are known to impact bacterial growth and development, we investigated gene expression profiles in commensal-rich, intermediate and pathogen-rich biofilms in response to nicotine-containing and nicotine-free aerosol exposure and compared them to clean-air exposure (Fig. 4A and Supplemental Table 4) using 86 million analyzable sequences that mapped to 8973 transcripts. Multivariable association analysis identified 2049 KEGG orthologs that were significantly upregulated following exposure to nicotine-free or nicotine-plus aerosol when compared to clean-air (Supplemental Table 4). 1872 of these were significantly upregulated in response to nicotine-plus aerosol exposure when compared to clean air, and 2257 in response to nicotine-free aerosol. Interestingly, 1810 genes were upregulated in response to both nicotine-free and nicotine-rich aerosol. Of these, over 100 transcripts encoded quorum sensing and competence, 37 coded for biofilm formation and 72 genes contributed to glycerol metabolism. When the analyses were directed to biofilm type, 458 and 475 genes were significantly overexpressed in intermediate and pathogen-rich biofilms when compared to commensal-rich biofilms respectively. Prominent among the genes upregulated in intermediate and pathogen-rich biofilms were those encoding organic carbon-compound metabolism, quorum sensing, antimicrobial resistance, secretion systems and transporters. Commensal-rich biofilms demonstrated upregulation of genes encoding capsule, peptidoglycan and glycosaminoglycan biosynthesis, rhamnose containing glycans, and extracellular polysaccharide biosynthesis, notably sialic acids, e.g., legionaminic acid and neuraminic acid. Moreover, graph theoretics revealed robust and statistically significant correlations between transcripts and metabolites (Figs. 4B-4G). In commensal-rich biofilms, two large hubs demonstrating high betweenness and degree centrality were evident, one that was anchored by genes encoding phosphotransferase systems, transporters (carbohydrate, oligopepetide) and metabolites corresponding to lipid, carbohydrate, and energy metabolism. Prominent anchors of the second large network were genes that corresponded to xenobiotic degradation, and stress response and metabolites that mapped to the xenobiotic degradation pathway. Pathogen-rich networks also demonstrated two large hubs, however, the hubs evidenced one-third (one-fourth) fewer connectivity between metabolites and transcripts than commensal-rich (intermediate) biofilms, as well as one-tenth (one-twelfth) betweenness centrality than commensal-rich (intermediate) biofilms. One hub consisted of networks between quorum sensing and bacterial motility genes with metabolites related to ketone and steroid biosynthesis, while bacterial metabolites related to quorum sensing correlated with genes encoding antimicrobial resistance, siderophores, osmotic stress, molecular chaperones and CRISPR in the second network neighborhood.

E-cigarette exposure alters biofilm topography

To explore the structural impact of these metabolic and transcriptional changes, we used confocal laser scanning microscopy to visualize the biofilms and computed topographical parameters with IMARIS. Exposure to nicotine-free and nicotine-plus aerosol increased the surface area of the biofilms within one hour of exposure, followed by a steady decline over 8 hours (Fig. 5, panel d(i) and d(ii)). For further experiments, we selected the 1-hour biofilm. Several topological features provided evidence of adaptation of biofilm to the environment even after a short period of exposure to aerosol (Fig. 5, panels e(i-iv)). The most salient feature was the significantly higher surface-to-volume ratio in aerosol-conditioned commensal-rich biofilms when compared to clean air control. In further confirmation of this, this difference in ratios was evident only in the live cells (Fig. 5e(i)), not in the dead cells (Fig. 5e(ii)), suggesting a dynamic rearrangement of growth patterns. More importantly, the average biomass (mass that is connected to the base or substratum, (Fig. 5e(iii))) and diffusion distances (Fig. 5e(iv)) were also significantly higher based on vaping exposure in commensal-rich communities.

Salivary metabolome profile of e-cigarette users recapitulates metabolism of nicotine-plus aerosol by pathogen-rich biofilms

We then investigated whether these metabolites could be identified in the saliva of a previously characterized cohort of e-cigarette users, dual users of e-cigarettes and cigarettes, and former smokers who currently use e-cigarettes⁸. We identified 3645 metabolites and compounds beyond a zero threshold. Spectral deconvolution and annotation to the molecular level identified 513 unique metabolites. As expected, psLDA revealed significant separation between the metabolomic profiles of pure e-cigarette users when compared to dual and former smokers. (Fig. 6 and Supplemental Table 5). We then compared these metabolome profiles with those generated by in vitro biofilms 196 of these were also identified in the in vitro analysis (Supplemental Table 5).

In the present investigation, we integrated metabolomics (a catalog of the complex chemical reactions elicited by a perturbation) and metatranscriptomics (a repository of all genes expressed by a community in response to a particular stressor), with topographical analysis of in vitro microcosm communities and validated the findings on human samples. By doing so, we obtained a real-time readout of changes in molecular pathways and biological functions following e-cigarette exposure and found evidence to support our hypothesis that oral bacteria metabolize e-cigarette aerosols into diverse compounds and that these compounds alter the geophysiology of oral microbial ecosystems.

While several studies in the literature have used single species biofilms or planktonic bacteria for exposomal studies^11–13, we used three prototypical, well-validated microcosm communities to mimic health-compatible and disease-associated oral microbiomes, since bacteria exist in nature in multi-species colonial relationships. It is well established that species belonging to the genera Streptococcus, Veillonella, Neisseria and Actinomyces are early colonizers and predominate in a health-compatible microbiome, and that this community increases in diversity through the bridging function mediated by F.nucleatum^14,15. Periodontitis is associated with a very diverse and heterogenous community that comprises species belonging to Selenomonas, Prevotella, Porphyromonas, Filifactor, Parvimonas and Tannerella among others. We have previously used such microcosms to investigate the impact of smoking¹⁶ and vaping⁸. Such an approach enabled us to identify the various tactics employed by bacteria to adapt to a change in environmental conditions.

While many of the metabolites in the present investigation are unknown and uncharacterized at the present time, a group of metabolites were identified as pesticides, fungicides, miticides and anti-bacterial agents, presumably from tobacco agriculture and agents used as preservatives in the e-cigarette liquid. These compounds were appreciably more abundant in commensal-rich biofilms; and mapped to the xenobiotic degradation pathway. Corroborating this, the metatranscriptome of commensal-rich biofilms demonstrated 2-7-fold overexpression of genes belonging to this functional family when compared to intermediate or pathogen-rich biofilms. Studies on the gut microbiome demonstrate that the indigenous bacteria respond to pesticide-like chemicals by upregulating pathways of xenobiotic degradation¹⁷, and that these altered bacterial metabolites create a proinflammatory response and modulate the permeability of the mucosal barrier^18,19. Most importantly, several of these metabolites were also identified in the salivary metabolome of 75 disease-naïve current, former or never smokers who used e-cigarettes. Taken together, these provide a plausible biological mechanism for the exaggerated gingival inflammation that has been reported in e-cigarette users^20,21, and deserves further investigation.

Another key finding was that e-cigarettes elicit a florid quorum sensing (QS) mediated response in all oral microbial communities. The first line of evidence was the detection of homoserine lactone and dipeptides in the metabolome of all three microcosms following exposure to nicotine-free and nicotine-plus aerosol, but not clean-air. The second was that these molecules were verified in the human salivary metabolome, confirming that these in vitro findings recapitulate real life events. The third line of evidence came from the upregulation of genes encoding QS functions in the metatranscriptome of all three biofilms; and from graph theoretics, which demonstrated robust and positive network hubs between QS molecules in the metabolome and several QS-regulated genes in the metatranscriptome. The fourth line of evidence came from the topological alterations in the biofilm architecture, where pronounced alterations were evident in the commensal-rich biofilm landscape, and to a lesser extent, in the intermediate biofilms. An important aspect of this finding was that both nicotine-free and nicotine-plus aerosols activated QS-responsive systems. While it is known that nicotine stimulates QS in bacterial biofilms²², we now demonstrate that other components of e-cigarettes are equally efficient in eliciting this response. This finding introduces a cautionary note to the perception that nicotine-free e-cigarettes are safer than nicotine-containing aerosol.

Although the aerosol metabolites produced by commensal-rich biofilms differed based on the presence or absence of nicotine, these biofilms responded to both types of e-cigarettes by increased secretion of public goods. This was supported by the following findings: (a) The prominent QS-regulated functions in the metatranscriptome were those contributing to cell wall and capsular biosynthesis and extra-cellular polysaccharides. (b) Robust correlations between these transcripts and compounds corresponding to carbohydrate metabolism. (c) Topographical features indicative of poor access to resources, and evidence of geomorphological restructuring to optimize access to nutrients²³.

Public goods are extracellular secretions that are produced in response to QS-regulated stress response, the most notable among which are exopolysaccharides, toxins, surfactants, and extracellular enzymes²⁴. Since these goods are readily shared among community members, it enables diversification of the community to include ‘cheaters’, that is, organisms that do not produce the goods, but benefit from it. This might explain our previous discovery that the oral microbiome of e-cigarette users is distinctly more diverse and includes more species than those of non-vapers⁸. However, the production of publicly shared goods leaves the producer vulnerable to invasion and take-over by the cheaters. One strategy employed by the cooperators is to modify the spatial structure of the biofilm to partition resources and limit diffusion²⁵. This is borne out by increase in surface area to biovolume ratio and large diffusion distance in commensal-rich biofilms. Bacteria increase surface to volume ratio in order to optimize access to the limited supply of nutrients. Together, the metabolomic, metatranscriptomic and topography suggest that health-compatible microbiomes that are exposed to nicotine-free or nicotine-containing e-cigarette aerosol protect themselves from stress by producing extracellular matrix and protect this exoproduct from exploitation by spatial restructuring. This possible mechanism provides a biological underpinning for recent reports of increased caries prevalence in these individuals^26–28.

QS-signaling can also induce bacteria to produce private goods, which are cytoplasmic or surface-attached proteins or metabolites²⁹. Pathogen-rich biofilms demonstrated upregulation of private goods (antimicrobial resistance, histidine kinases, metal transport), as well as siderophores, which have both public and private benefits. These correlated with the pyridine and pyrrolidine pathways of nicotine metabolism by this biofilm and suggest that pathogen-rich biofilms are influenced by different elements of the aerosol than commensal-rich biofilms. This provides an explanation for the increased antimicrobial resistance that has been reported in e-cigarette users^30,31, and highlights the potential risk for failure of antimicrobial therapy in these individuals. Interestingly, the pathogen-rich biofilm landscape did not change following e-cigarette exposure, indicating that mechanisms of QS-mediated stress response differ widely between health-compatible and disease-associated communities.

In summary, within the limitations of an in vitro biofilm model and a cross-sectional human study, we demonstrate that oral bacteria vigorously metabolize e-cigarette aerosol to generate multiple compounds and chemicals. The resultant metabolome depends on the e-cigarette aerosol as well as the denizens of the bacterial biofilm. We also demonstrate that bacteria utilize QS-regulated pathways to adapt and survive e-cigarette induced stress. These survival mechanisms vary based on the microbial denizens but are similar between nicotine-free and nicotine-containing aerosol, implicating the glycerol/glycol humectant as a driver of the microbial exposotype. Importantly, we find that xenobiotic degradation by health-compatible, commensal-rich biofilms has the potential to detrimentally impact the host-microbial interface. Further studies that interrogate metabolic and transcriptional events at the host-microbial interface are urgently needed to further explicate the mechanisms by which these products create at-risk-for-harm environments.

Aerosol generation and analysis via gas chromatography-mass spectrometry (GCMS):

E-cigarette aerosol was captured for GCMS analysis by connecting a 50 mL glass syringe (Micro-mate) to the e-cigarette mouthpiece of a moderate sized vape pen with 0.2 W resistance coil and 3000mAh/80W battery. Aerosol was generated and captured following the protocol by researchers at the ADA Foundation Volpe Research Center (Gaithersburg, MD) consisting of a custom acrylonitrile butadiene styrene (ABS) enclosure with a 510 adapter, precision wattage meter and power analyzer³². 100 mL of e-cigarette aerosol was generated following a physiologic puffing profile (3 second puff duration, 18 second interval, repeated for 32 puffs) and immediately injected into a 22 mL GC headspace vial with polytetrafluoroethylene (PTFE)/silicone rubber septa (Perkin Elmer) using an 18-gauge needle. 1 cm³ GC glass wool (Sigma Aldrich 20384) was used as a filter between the mouthpiece and syringe tip. Analysis was conducted using the PerkinElmer Clarus 680 Gas Chromatograph and Clarus SQ 8C Mass Spectrometer, equipped with a Phenomenex Zebron ZB-5MSplus capillary column (30 m L x 0.25 mm ID x 1.0 μm df), using the methods previously developed³². Briefly, 1 μL of sample was injected onto the GC column for separation, with an injector temperature of 250 °C. The oven temperature was held at the initial temperature of 40 °C for 2 min before increasing to 150 °C at a rate of 4 °C/min. This was held for 4 min before increasing to 290 °C at a rate of 6 °C/min. This final temperature was held for 2 min. Mass spectra were acquired in positive ion mode from m/z 45 – 350 with no solvent delay. Quantitation of analytes was accomplished by creating an external standard curve ranging from 5-750 ppm (detection limit:5 ppm) using the area under the peak corresponding to the analyte. The resulting linear trendline was used to determine the concentration of the analyte in each sample.

Multispecies biofilm model: Artificial saliva was made following the Marshall Group Research protocol for artificial saliva. SHI medium was prepared according to the protocol described by Tian et al.³³. Biofilms were developed using the modifications¹⁶ from the protocol established by Guggenheim et al.³⁴. Briefly, sterilized, sintered hydroxyapatite (HA) disks (Clarkson Chromatography Products, South Williamsport, PA) were incubated in artificial saliva for 24 hours to establish a pellicle coat, following which multispecies commensal primary biofilms were generated by seeding six pioneer species [Streptococcus oralis (ATCC 35037), S. sanguis (10556), S. mitis (49456), Actinomyces naeslundii (12104), Neisseria mucosa (25997), and Veillonella parvula (17745)] and incubating under aerobic conditions in a 1:1(v/v) mixture of SHI media and artificial saliva. Pathogen-rich biofilms were created by further seeding the commensal biofilms with an intermediate bridging colonizer [Fusobacterium nucleatum (10953), secondary biofilm] followed 24 hours later by Porphyromonas gingivalis (33277), Filifactor alocis (35896), Selenomonas sputigena (35185), S. noxia (43541), Campylobacter gracilis (33236), Prevotella intermedia (25611), Parvimonas micra (33270), and Tannerella forsythia (43037) and incubating under anaerobic conditions for a further 24 hours (tertiary biofilms).

ENDS exposure: Electronic cigarette vapor (ECV) was prepared following our previous protocol with minor modifications⁸: a moderate-sized e-cigarette pen was filled with either nicotine-free e-liquid or 6mg/ml nicotine e-liquid, both unflavored, and actuated by pressing “on” for 5 s then “off” for 25 s and repeated for a total of 10 minutes or 20 “puffs”. The e-cigarette was connected via Pasteur pipettes into 5 ml of artificial saliva. ECV was prepared immediately before each use. The nicotine-free and nicotine-containing ECV replaced the artificial saliva in the 50/50 saliva/SHI media incubating mixture following each comparative exposure condition. To maintain the consistency of ECV between experiments, an optical density of 0.15 at 600 nm represented 100%.

Bacterial metabolomics:Following growth of each respective biofilm, growth media was removed, and biofilms were exposed to 100% artificial saliva following each respective exposure condition: nicotine-free, nicotine-containing, and control conditions. Saliva supernatant was collected for 1-, 2-, 4-, and 8-hour timepoints. Samples were spun at 10,000rpm and decanted to remove cell debris. Saliva was analyzed by nuclear magnetic resonance (NMR) spectroscopy and trapped ion mobility spectrometry tandem time-of-flight (TIMS-TOF).

Nuclear magnetic resonance (NMR) spectroscopy: Untargeted one-dimensional (1D) 1H NMR of the bacterial supernatant was analyzed with 800 MHz spectrometer (Bruker,USA). Samples were analyzed using the first increment of NOESY pulse sequence with presaturation and the CPMG pulse sequence. 1H NMR spectra was acquired at 298K using 128 scans and 64K data points. 2D NMR was applied on selected samples to confirm the identity of the specific metabolites. Free induction decays (FIDs) were multiplied by a decaying exponential function with a 1 Hz line broadening factor prior to Fourier transformation. The 1H NMR spectra were corrected manually for phase and a polynomial fourth-order function applied for base-line correction in order to achieve accurate and reproducible measurements upon integration of the signals of interest. Chemical shifts were reported in ppm as referenced to TSP (δ = 0). Spectra were processed and analyzed using Topspin 3.2. Prior to statistical data analysis, each bucketed region was normalized to the total sum of the spectral intensities.

TIMS-TOF Data Processing and Metabolite Annotations: Bacterial supernatant was subjected to matrix-assisted laser desorption ionization–trapped ion mobility spectrometry time-of-flight mass spectrometry (TIMS-TOF) (Bruker,USA). The acquired raw datasets were initially processed by using SCiLS lab 2021a software (Bruker,USA). Mass range was selected between 20-2000 m/z for assigning regions. Files were then exported in Metaboscape 2021b (Bruker,USA) for annotations and further downstream analysis. After checking the regions, all m/z points were annotated by using eleven libraries from analyte list of libraries i.e HMDB library 2.0_KEGG, Lipids Human Brain metabolites library, Lipids Mouse Kidney metabolites library, Small Molecules metabolites library, N-Glycan human library, Cell culture nutrient library, Fatty acids library, HMDB plasma metabolites library, Lipid maps library, Natural products metabolites library and CCS compendium library. Also, annotations were carried out by using a range of Mass spectral libraries provided by Metaboscape like Bruker Sumner MetaboBASE plant library, Bruker NIST 2020 MSMS Spectral Library hr-2, MSDIAL- TandemMassSpectralAtlas libraries for both positive and negative ions. The parameters (tolerances and scoring) used for annotations are as follows m/z: 2.0 - 5.0ppm, msigma 25 – 500 and CCS 2.0- 5.0%. Annotations of metabolites against all Lipid classes available in Metaboscape was also carried out with the same m/z and mSigma values.

Statistical analysis: Bucket tables from each experiment was exported from Metaboscape 2021b for further statistical data analysis in R. Nonmetric Dimensional scaling was carried out in R by using vegan package. In order to determine the statistical difference between the metabolites concentration between groups, ANOSIM was employed. For heatmap generation p values (FDR) based on t-test between the two groups were also calculated and table was exported in R to further plot the heatmap between the top 50 most significantly different metabolites between the two comparison groups. Pheatmap library was used to plot the heatmap. After Peak intensity table was imported, the uploaded data was log-transformed, and normalization was done by mean subtraction. Other parameters that were set included the use of the correlation-based clustering of the columns. To simplify the visualization of the abundances of the metabolites across the treatments, the top 50 metabolites ranked by t-test are shown. Metaboanalyst (v6.0) was used to analyze the role of microbial community composition via partial least squares discriminant analysis (PLSDA)³⁵.

RNA isolation, metagenomic sequencing, and analysis: Biofilms were harvested and RNA isolated using the mirVana miRNA isolation kit (Applied Biosystems). Ribosomal RNA was depleted, and mRNA was enriched by modified capture hybridization approach (MICROBExpress mRNA enrichment kit, Thermo Fisher Scientific). Enriched mRNA served as a template for the polyadenylation reaction and complementary DNA synthesis. Microbial libraries were clustered on the Illumina HiSeq 4000 platform, and 150-bp paired-end sequencing was performed. The Illumina base-calling pipeline was used to process the raw fluorescence images and call sequences. Raw reads with >10% unknown nucleotides or with >50% low-quality nucleotides (quality value, <20) were discarded. Microbial transcripts were quality-filtered using Sickle v1.33 (default parameters) and aligned against the RefSeq nonredundant proteins database using DIAMOND v0.8.3.65³⁶. Aligned sequences were annotated to the KEGG database using MEGAN 6³⁷.

Quality control: All samples were sequenced in two runs; and to minimize batch effects, samples were randomly assigned to each run. Replicate sequencing was carried out for two samples in each batch, and the replicates showed good reliability across the five batches, with coefficient of variability (SD/mean) ranging from 0.26 to 1.3% for alpha diversity of taxonomy and 3.4 to 6.3% for predominant functions (carbohydrate metabolism, respiration, and virulence, disease, and defense).

Biofilm imaging: To enable confocal microscopic imaging, we stained the biofilms using the BacLight kit (Life Technologies, NY) according to the manufacturer’s instructions. Briefly, the biofilms were incubated in 1.5 ml of 0.3% SYTO 9 and propidium iodide, and the fluorescence was measured at 486 and 520 nm using a Spectral FlowView confocal microscope at 10× magnification. The ratio of green to red fluorescence was computed, and Z-stack images were obtained. A minimum of eight images per group was obtained to generate volume and area graphs. Total surface area and volume were determined using Imaris v9 (http://bitplane.com) from the constructed three-dimensional images. Boxplot comparisons of areas and volumes were visualized using Seaborn v0.9.0, and the significance of pairwise differences was determined using Tukey’s post hoc test (JMP statistical software v13.0).

Clinical correlation: Clinical samples were collected in a previously published study⁸. We obtained approval for this study from the Office of Responsible Research Practices at The Ohio State University [IRB (Institutional Review Board) protocol number 2014H0062 and e-IBC protocol number 2015R00000005], and the study was conducted in accordance with approved guidelines. We recruited 123 systemically [ASA I (American Society of Anesthesiologists Physical Status Classification I)] and periodontally healthy individuals [attachment loss ≤ 1; less than three sites with 4 mm of probe depths (PD); bleeding index (BOP) ≤ 20%] following informed consent and clinical and radiographic examination to each of five groups: (i) smokers (25), (ii) nonsmokers (25), (iii) e-cigarette users (20), (iv) former smokers currently using e-cigarettes (25), and (v) concomitant cigarette and e-cigarette users (28). Current smokers were those who had at least a five pack-year history and had no prior history of e-cigarette use. Never smokers were those who had smoked less than 100 cigarettes in their lifetime and none in the past year, and e-cigarette users were those who used e-cigarettes daily for at least 3 months, with at least one cartridge per day or 1 ml of liquid per day. Former smokers were those who had quit smoking for at least 1 year. Exclusion criteria for all groups included controlled or uncontrolled diabetes, HIV infection, use of immunosuppressant medications, bisphosphonates, or steroids, antibiotic therapy or oral prophylactic procedures within the preceding 3 months, and fewer than 20 teeth in the dentition. Saliva was collected from each participant and analyzed via TIMS-TOF.

Author Contributions:

Metabolomics: PPC, SK, EC, GR, BNDS, MY, MLSB, PSK, IAM Metatranscriptomics: MLSB, SMD, PSK Cell-culture and Imaging: MLSB, SMD, PSK Bioinformatics and data analysis: PPC, SMD, SMG, SK, EC,, MLSB, PSK, IAM Clinical study: PPC, SMG, SMD, MLSB, PSK

Competing interests:

All authors declare no financial or non-financial competing interests.

Acknowledgements:

The study was funded by NIDCR R01-DE DE027857 to Purnima Kumar and F30 DE032895 to Michelle Beverly.

Data availability:

All data generated or analyzed during this study are included in this published article and its supplementary information files.

Wild, C. P. Complementing the Genome with an “Exposome”: The Outstanding Challenge of Environmental Exposure Measurement in Molecular Epidemiology. Cancer Epidemiology, Biomarkers & Prevention 14, 1847-1850, doi:10.1158/1055-9965.Epi-05-0456 (2005).
Wild, C. P. The exposome: from concept to utility. International Journal of Epidemiology 41, 24-32, doi:10.1093/ije/dyr236 (2012).
Wang, T. W. et al. E-cigarette Use Among Middle and High School Students - United States, 2020. Mmwr-Morbid Mortal W 69, 1310-1312, doi:DOI 10.15585/mmwr.mm6937e1 (2020).
Ali, F. R. M. et al. E-cigarette Unit Sales by Product and Flavor Type, and Top-Selling Brands, United States, 2020-2022. Mmwr-Morbid Mortal W 72, 672-677 (2023).
Cooper, M. et al. E-cigarette Use Among Middle and High School Students - United States, 2022. Mmwr-Morbid Mortal W 71, 1283-1285 (2022).
St. Helen, G. & Eaton, D. L. Public Health Consequences of e-Cigarette Use. JAMA Internal Medicine 178, 984-986, doi:10.1001/jamainternmed.2018.1600 (2018).
Marques, P., Piqueras, L. & Sanz, M.-J. An updated overview of e-cigarette impact on human health. Respiratory research 22, 151, doi:10.1186/s12931-021-01737-5 (2021).
Ganesan, S. M. et al. Adverse effects of electronic cigarettes on the disease-naive oral microbiome. Science Advances 6, eaaz0108, doi:10.1126/sciadv.aaz0108 (2020).
Postler, T. S. & Ghosh, S. Understanding the Holobiont: How Microbial Metabolites Affect Human Health and Shape the Immune System. Cell Metab 26, 110-130, doi:10.1016/j.cmet.2017.05.008 (2017).
Wishart, D. S. et al. HMDB 5.0: the Human Metabolome Database for 2022. Nucleic Acids Res 50, D622-D631, doi:10.1093/nar/gkab1062 (2022).
Bagaitkar, J. et al. Tobacco-induced alterations to Porphyromonas gingivalis-host interactions. Environmental microbiology 11, 1242-1253, doi:10.1111/j.1462-2920.2008.01852.x (2009).
Ertel, A., Eng, R. & Smith, S. M. The differential effect of cigarette smoke on the growth of bacteria found in humans. Chest 100, 628-630 (1991).
Huang, R., Li, M. & Gregory, R. L. Effect of nicotine on growth and metabolism of Streptococcus mutans. Eur J Oral Sci 120, 319-325, doi:10.1111/j.1600-0722.2012.00971.x (2012).
Kolenbrander, P. E. Oral microbial communities: biofilms, interactions, and genetic systems. Annual review of microbiology 54, doi:10.1146/annurev.micro.54.1.413 (2000).
Kolenbrander, P. E., Andersen, R. N. & Moore, L. V. Coaggregation of Fusobacterium nucleatum, Selenomonas flueggei, Selenomonas infelix, Selenomonas noxia, and Selenomonas sputigena with strains from 11 genera of oral bacteria. Infect Immun 57, 3194-3203 (1989).
Shah, S. A. et al. The making of a miscreant: tobacco smoke and the creation of pathogen-rich biofilms. NPJ Biofilms Microbiomes 3, 26, doi:10.1038/s41522-017-0033-2 (2017).
Abou Diwan, M. et al. Impact of Pesticide Residues on the Gut-Microbiota–Blood–Brain Barrier Axis: A Narrative Review. International Journal of Molecular Sciences 24, 6147 (2023).
Djekkoun, N., Lalau, J.-D., Bach, V., Depeint, F. & Khorsi-Cauet, H. Chronic oral exposure to pesticides and their consequences on metabolic regulation: Role of the microbiota. European Journal of Nutrition 60, 4131-4149 (2021).
Joly Condette, C. et al. Use of molecular typing to investigate bacterial translocation from the intestinal tract of chlorpyrifos-exposed rats. Gut Pathogens 8, 1-12 (2016).
Park, B. et al. The mediating roles of the oral microbiome in saliva and subgingival sites between e-cigarette smoking and gingival inflammation. BMC Microbiology 23, 35, doi:10.1186/s12866-023-02779-z (2023).
Xu, F. et al. Comparative Effects of E-Cigarette Aerosol on Periodontium of Periodontitis Patients. Frontiers in Oral Health 2, doi:10.3389/froh.2021.729144 (2021).
Tang, H. et al. Regulation of Nicotine Tolerance by Quorum Sensing and High Efficiency of Quorum Quenching Under Nicotine Stress in Pseudomonas aeruginosa PAO1. Front Cell Infect Microbiol 8, 88, doi:10.3389/fcimb.2018.00088 (2018).
Reichhardt, C. & Parsek, M. R. Confocal Laser Scanning Microscopy for Analysis of Pseudomonas aeruginosa Biofilm Architecture and Matrix Localization. Frontiers in Microbiology 10, doi:10.3389/fmicb.2019.00677 (2019).
García-Contreras, R. et al. Quorum sensing enhancement of the stress response promotes resistance to quorum quenching and prevents social cheating. Isme j 9, 115-125, doi:10.1038/ismej.2014.98 (2015).
Özkaya, Ö., Xavier, K. B., Dionisio, F. & Balbontín, R. Maintenance of Microbial Cooperation Mediated by Public Goods in Single- and Multiple-Trait Scenarios. Journal of Bacteriology 199, 10.1128/jb.00297-00217, doi:doi:10.1128/jb.00297-17 (2017).
Irusa, K. F., Finkelman, M., Magnuson, B., Donovan, T. & Eisen, S. E. A comparison of the caries risk between patients who use vapes or electronic cigarettes and those who do not: A cross-sectional study. The Journal of the American Dental Association 153, 1179-1183, doi:10.1016/j.adaj.2022.09.013 (2022).
Vemulapalli, A., Mandapati, S. R., Kotha, A. & Aryal, S. Association between vaping and untreated caries: A cross-sectional study of National Health and Nutrition Examination Survey 2017-2018 data. The Journal of the American Dental Association 152, 720-729, doi:10.1016/j.adaj.2021.04.014 (2021).
Irusa, K. F., Vence, B. & Donovan, T. Potential oral health effects of e-cigarettes and vaping: A review and case reports. Journal of Esthetic and Restorative Dentistry 32, 260-264, doi:https://doi.org/10.1111/jerd.12583 (2020).
Schuster, M., Sexton, D. J. & Hense, B. A. Why Quorum Sensing Controls Private Goods. Frontiers in Microbiology 8, doi:10.3389/fmicb.2017.00885 (2017).
Laura, E. C. A., Shymaa, E. & Elisa, M. in Frontiers in <i>Staphylococcus aureus</i> (eds Enany Shymaa & E. Crotty Alexander Laura) Ch. 4 (IntechOpen, 2017).
Martínez-Solís, E. A. et al. Comparison of the phenotypic profile of antimicrobial resistance in the oral microbiota of non-smokers, tobacco smokers, and electronic cigarette vapers-a pilot study. Proceedings of Scientific Research Universidad Anáhuac. Multidisciplinary Journal of Healthcare 1, 5-13 (2021).
Kim, J. J. et al. Universal electronic-cigarette test: physiochemical characterization of reference e-liquid. Tobacco induced diseases 15, 14, doi:10.1186/s12971-017-0119-x (2017).
Tian, Y. et al. Using DGGE profiling to develop a novel culture medium suitable for oral microbial communities. Molecular oral microbiology 25, 357-367, doi:10.1111/j.2041-1014.2010.00585.x (2010).
Guggenheim, B. et al. In vitromodeling of host-parasite interactions: the 'subgingival' biofilm challenge of primary human epithelial cells. BMC Microbiology 9, 280, doi:10.1186/1471-2180-9-280 (2009).
Chong, J. & Xia, J. MetaboAnalystR: an R package for flexible and reproducible analysis of metabolomics data. Bioinformatics (Oxford, England) 34, 4313-4314, doi:10.1093/bioinformatics/bty528 (2018).
Buchfink, B., Xie, C. & Huson, D. H. Fast and sensitive protein alignment using DIAMOND. Nat Methods 12, 59-60, doi:10.1038/nmeth.3176 (2015).
Huson, D. H., Auch, A. F., Qi, J. & Schuster, S. C. MEGAN analysis of metagenomic data. Genome Res 17, 377-386, doi:10.1101/gr.5969107 (2007).

(Not answered)

SupplementalTable1.xlsx
Supplemental Table 1: The chemical composition of a commercial brand of tobacco-flavored nicotine-containing and nicotine-free e-cigarette aerosol using GCMS. Compounds that are common to both e-liquids are highlighted in red.
SupplementalTable2.xlsx
Supplemental Table 2: Spectral deconvolution and annotation to the molecular level of TIMS-TOF analysis of a commercial brand of tobacco-flavored nicotine-containing and nicotine-free e-cigarette aerosol
SupplementalTable3.xlsx
Supplemental Table 3: Metabolites that were differentially abundant and uniquely identified in nicotine-free and nicotine-plus vapor
SupplementalTable4.xlsx
Supplemental Table 4: Gene expression profiles in commensal-rich, intermediate and pathogen-rich biofilms in response to nicotine-containing and nicotine-free aerosol exposure and clean-air exposure
SupplementalTable5.xlsx
Supplemental Table 5: Spectral deconvolution and annotation to the molecular level of metabolites identified in the saliva of a previously characterized cohort of e-cigarette users, dual users of e-cigarettes and cigarettes, and former smokers who currently use e-cigarettes. Metabolites that were common to those obtained from vitro biofilms are highlighted in orange

Download PDF

Editorial decision: revise
26 Aug, 2024
Review #3 received at journal
16 Aug, 2024
Review #2 received at journal
15 Aug, 2024
Review #1 received at journal
30 Jul, 2024
Reviewer #3 agreed at journal
23 Jul, 2024
Reviewer #2 agreed at journal
23 Jul, 2024
Reviewer #1 agreed at journal
20 Jul, 2024
Reviewers invited by journal
20 Jul, 2024
Editor assigned by journal
14 Jul, 2024
Submission checks completed at journal
01 Jul, 2024
First submitted to journal
24 Jun, 2024

You are reading this latest preprint version

Toxic cultures: E-cigarettes and the oral microbial exposome

Status:

Version 1

Abstract

Figures

INTRODUCTION

RESULTS

DISCUSSION

METHODS

Declarations

Author Contributions:

Competing interests:

Acknowledgements:

Data availability:

References

Additional Declarations

Supplementary Files

Status:

Version 1