E-liquids were prepared as previously described (Fischman et al., 2020; Xu et al., 2022; Christian et al., 2024). Briefly, propylene glycol and vegetable glycerol were mixed at a 1:1 ratio (solvent) and spiked with 20 mg/mL nicotine and 5% v/v cinnamon or menthol flavors. E-liquids were stored at 4°C and diluted in A. actinomycetemcomitans growth medium (AAGM, 3 g trypticase soy broth and 0.6 g yeast extract in 100 mL DI water) for experiments.

A. actinomycetemcomitans (JP2 genotype, a gift from Dr. Edward T. Lally) was initially cultured as previously described (Kachlany et al., 2002). Briefly, bacteria were grown in 100 mL AAGM (3 g trypticase soy broth and 0.6 g yeast extract), along with 5 µg vancomycin, 75 µg bacitracin, 0.8 g dextrose, and 4 mL 10% sodium bicarbonate (Tsuzukibashi et al., 2008) in a candle jar, at 37°C with no shaking for 18 hr.

For the growth curve experiments, the starter culture was diluted to an optical density (OD₆₀₀) of 0.06 in a 96-well plate, along with menthol or cinnamon E-liquids in a range of concentrations. The plate was incubated at 37°C, and the OD₆₀₀ was recorded hourly using a Tecan plate reader. The change in OD₆₀₀ (ΔOD₆₀₀), relative to the initial reading was calculated hourly.

Starter cultures were diluted 6:100 in a larger volume of AAGM (with vancomycin, bacitracin, dextrose, and sodium bicarbonate) along with menthol or cinnamon E-liquid and then cultured for another 24 hrs at 37°C with no shaking. OMVs were collected from the bacterial supernatants following a published protocol (Kato et al., 2002). Briefly, bacterial cultures were centrifuged twice at 10, 000 × g for 10 min to pellet the cells. Supernatants were filtered using a 0.45-µm filter and then concentrated using 50-kDa molecular weight cutoff (MWCO) centrifugal filters. Filtered supernatants were ultracentrifuged twice at 105, 000 x g and 4°C for 30 min. Pellets, containing the OMVs, were resuspended in 2 ml of Tris buffer (150 mM NaCl, 10 mM Tris, pH 7.0). Finally, the OMV suspensions were centrifuged for 5 min at 5, 000 x g to remove any remaining debris, and the supernatant was then filtered through a 0.45-μm syringe membrane filter (ThermoFisher Scientific, polyethersulfone (PES) membrane). The purified OMVs were stored at −20°C until use (Kato et al., 2002).

The lipophilic FM™ 4–64 dye (ThermoFisher Scientific) was used to determine the relative lipid concentrations of the purified OMVs (MacDonald and Kuehn, 2013). Liposomes composed of 1-palmitoyl-2-oleoyl-glycero-3-phosphocholine (POPC) were prepared and diluted serially. The FM™ 4–64 dye (5 μg/L) was added to the serially diluted liposomes as well as to serially diluted OMV solutions. The fluorescence intensity (excitation: 515 nm, emission 640 nm, collected using a Tecan plate reader) of the OMV solution was compared to that of a calibration curve created using the POPC liposomes to determine the lipid concentration of the OMVs, as previously described (MacDonald and Kuehn, 2013).

To determine the amount of protein in each OMV sample, the absorbance at 280 nm (A₂₈₀) was measured using a NanoDrop spectrophotometer (ThermoFisher Scientific) (Simonian and Smith, 2006).

The thin film method was used to synthesize liposomes (Zhang, 2017). POPC (25 mg/mL in chloroform) was added to a glass vial in the required amount. The chloroform was then evaporated using a stream of air to leave a thin lipid film. The lipid film was placed under vacuum overnight to remove any residual chloroform. Tris buffer was added to the films, which were then vortexed and extruded 13 times through 100 nm polycarbonate filters to create unilamellar liposomes.

The size of the OMVs was determined by dynamic light scattering (DLS). Samples were diluted 1:20 in Tris buffer and passed through a 0.45-µm pore-size PES membrane to minimize contamination. Measurements were performed on an ALV/CGS-3 compact goniometer at 632.8 nm wavelength, with a 90° scattering angle and a run time of 180 s. Each sample was analyzed in triplicate. Data acquisition and processing were carried out using the ALV proprietary software, which generated the number weighted size distribution using a regularized fit and assuming a membrane thickness of 5 nm (r* = 5 nm) (Nagle and Tristram-Nagle, 2000; Pencer and Hallett, 2003).

The DNA-intercalating dye TOTO-1 (ThermoFisher Scientific) was used to assess the relative surface-associated DNA levels of the purified OMVs (Hirons et al., 1994). OMVs were diluted to a final lipid concentration of 0.2 mM and mixed with an equal volume with TOTO-1 dye. The samples were incubated under dark conditions prior to fluorescence measurements for 15 minutes. The fluorescence was measured using a PTI fluorescence spectrometer [Photon Technology Inc. (PTI)] with excitation at 514 nm and emission collected at 533 nm.

Laurdan fluorescence was used to assess the membrane fluidity of purified OMVs. OMVs were prepared at a lipid concentration of 1 mM in Tris buffer and incubated with Laurdan (100 µM, 1:5) at 37 °C for 2 h. Following incubation, the samples were centrifuged using Millipore 30 kDa molecular weight cutoff membrane filters to remove excess unbound dye, washed, and resuspended in fresh Tris buffer. For fluorescence measurements, 100 µL of each sample was transferred to a black 96-well plate, and fluorescence intensity was measured using a Tecan plate reader (excitation: 368 nm; emission: 400–500 nm). The resulting spectra were normalized to the maximum emission intensity of each sample. The Laurdan emission spectrum exhibits two characteristic peaks corresponding to distinct membrane states; the emission at 440 nm (I₄₄₀) represents the gel (ordered) state, while the emission at 490 nm (I₄₉₀) represents the liquid-crystalline (disordered) state. Generalized polarization (GP) values were calculated using Equation 1, where higher GP values indicate lower membrane fluidity (Parasassi et al., 1991).

Control and treated A. actinomycetemcomitans OMV samples were collected as above, lyophilized and then resuspended in 20 µL water to measure protein concentration using the NanoDrop. Based on protein concentrations, 100 µg of each sample were added to a mixture of 100 µL final volume, containing SDS sample buffer and 2.5 µM dithiothreiotol to reduce disulfide bonds. Samples were denatured at 100°C for ten minutes. After denaturing and reducing, 40 µg of each protein sample were loaded and separated in a 4% - 20% gradient gel (GenScript Biotech, Piscataway, NJ, USA). The electrophoresis settings were 140 V, 80 mA and 150 Watts for 1 hour and then reduced to 100 V, 50 mA and 100 Watts for 45 minutes. Gels were then washed and stained in Coomassie overnight. For silver nitrate enhancement, the gel was stained as previously established (Merril et al., 1982). Bands were visualized using the ChemiDoc MP Imaging System (Bio-Rad Laboratories Inc. Philadelphia, PA, USA). Band densities were quantified using ImageJ (v. 1.54g), and the consistent 35 kDa bands were used as a loading control to normalize protein percentages for each treatment or control.

Statistical analyses were performed using MATLAB (MathWorks, Natick, MA, USA). Comparisons between two independent groups were conducted using a two-tailed Welch’s two-sample t-test to account for potential unequal variances and unequal sample sizes. Summary statistics (mean, standard deviation, and sample size) were used to compute the t-statistic, degrees of freedom based on the Welch–Satterthwaite approximation, and corresponding p-values. Results were considered statistically significant at p < 0.05.

The effect of menthol- or cinnamon-flavored E-liquids on A. actinomycetemcomitans growth was determined by measuring A. actinomycetemcomitans growth in AAGM alone (untreated) or AAGM supplemented with the E-liquids at four concentrations (0.25%, 0.5%, 0.75%, and 1%). As shown in Figure 1, both E-liquid treatments inhibited bacterial growth, with menthol E-liquid treatment (Figure 1A) inhibiting growth at concentrations above 0.5% and cinnamon E-liquid treatment (Figure 1B) inhibiting growth at all tested concentrations. The cinnamon E-liquid exhibited a stronger inhibitory effect than menthol E-liquid; concentrations greater than 0.5% led to complete growth suppression. Based on these findings, media containing 0.75% menthol or 0.25% cinnamon E-liquids were selected for further experiments on OMV production, as these conditions produced similar levels of growth inhibition while remaining clearly distinct from the untreated control (Figure 1C ).

Next, we investigated the effect of menthol and cinnamon flavored E-liquids on OMV production. Figure 1D shows the relative lipid mass of OMVs collected from A. actinomycetemcomitans grown in AAGM (control) or AAGM supplemented with 0.75% menthol or 0.25% cinnamon E-liquids. Both E-liquids reduced OMV production relative to the control, with cinnamon treatment causing the strongest effect, an approximately five-fold reduction. All OMV-related measurements presented here are reported as relative comparisons across experimental conditions, enabling evaluation of treatment-dependent trends despite the absence of quantitative particle counting.

Prior work in our lab has found that A. actinomycetemcomitans JP2 releases two populations of OMVs, one highly abundant population of small OMVs (radius < 112 nm), and a second population of larger OMVs (radius > 112 nm). Importantly the leukotoxin (LtxA) concentration of the OMV populations differs, with LtxA being predominately located on the larger OMVs (Collins et al., 2021; Nice et al., 2024; Singh et al., 2024). Here, we evaluated the effect of E-liquids on OMV size distributions using DLS. We observed similar overall OMV size distributions across all three treatments (Figure 2A). Comparison of the average size of “small” and “large” OMVs indicated a modest but significant decrease in size of small OMVs from the cinnamon E-liquid-treated cells relative to the control (Figure 2B). In terms of relative proportions, cinnamon E-liquid-treated bacteria produced fewer small OMVs and more large OMVs compared with both the menthol-treated and control groups (Figure 2C).

The protein concentration of the OMVs was measured using a NanoDrop spectrophotometer and normalized to the molar lipid concentration of each preparation. The resulting protein content per 1 mM lipid is shown in Figure 3A. OMVs collected from both menthol and cinnamon E-liquid-treated cultures had significantly higher protein concentrations than untreated control samples. The OMVs collected from the cinnamon-supplemented culture exhibited the highest protein-to-lipid ratio, exceeding both control and menthol-treated groups.

TOTO-1 fluorescence staining was used to assess relative surface-associated DNA levels among the OMV samples. OMVs derived from cinnamon E-liquid–treated cultures (Figure 3B) exhibited relatively higher TOTO-1 fluorescence compared to the other groups, indicating increased surface-associated DNA. In contrast, OMVs derived from menthol E-liquid-treated cultures contained a similar amount of surface-associated DNA as the untreated control.

To determine if large-scale changes in lipid composition occur upon E-liquid exposure, we used Laurdan fluorescence to determine changes in membrane fluidity. Figure 3C shows the resulting normalized fluorescence spectra revealing no significant differences in membrane fluidity among the groups. The GP values, which indicate membrane fluidity, with lower values indicating more gel-like membranes, are shown in Figure 3D and also indicate no significant differences.

Separation and Coomassie/silver staining of OMV proteins via SDS-PAGE revealed differences in the expression of several proteins within the A. actinomycetemcomitans OMVs between control and E-liquid treatments (Figure 4A). We observed that the 35 kDa molecular weight band was consistent across all OMV preparations. This band is most likely OmpA which is commonly used to normalize OMV concentration (Ojima et al., 2018) because of its consistent expression and localization in OMVs. We therefore used this band as a “loading control” and used it to determine the relative expression of other bands, including the 20 kDa, 32 kDa, 40 kDa, 42 kDa, and 180 kDa bands (Figure 4B). This analysis demonstrated that the 20, 32, 40, and 42 kDa bands increased in intensity in OMVs produced by cinnamon-treated cultures, while that of the 120 kDa band decreased. In OMVs produced by menthol-treated cultures, the 40 and 42 kDa bands increased and the 180 kDa band decreased. Altogether, the data indicate that E-liquids induce differential OMV proteome patterns, visible without an extremely sensitive assay (i.e., Western blot).