Lung Toxicity of Condensed Aerosol from E-CIG Liquids: Influence of the Flavor and the In Vitro Model Used

Abstract

The diffusion of e-cigarette (e-CIG) opens a great scientific and regulatory debate about its safety. The huge number of commercialized devices, e-liquids with almost infinite chemical formulations and the growing market demand for a rapid and efficient toxicity screen system that is able to test all of these references and related aerosols. A consensus on the best protocols for the e-CIG safety assessment is still far to be achieved, since the huge number of variables characterizing these products (e.g., flavoring type and concentration, nicotine concentration, type of the device, including the battery and the atomizer). This suggests that more experimental evidences are needed to support the regulatory frameworks. The present study aims to contribute in this field by testing the effects of condensed aerosols (CAs) from three main e-liquid categories (tobacco, mint, and cinnamon as food-related flavor), with (18 mg/mL) or without nicotine. Two in vitro models, represented by a monoculture of human epithelial alveolar cells and a three-dimensional (3D) co-culture of alveolar and lung microvascular endothelial cells were used. Cell viability, pro-inflammatory cytokines release and alveolar-blood barrier (ABB) integrity were investigated as inhalation toxicity endpoints. Results showed that nicotine itself had almost no influence on the modulation of the toxicity response, while flavor composition did have. The cell viability was significantly decreased in monoculture and ABB after exposure to the mints and cinnamon CAs. The barrier integrity was significantly affected in the ABB after exposure to cytotoxic CAs. With the exception of the significant IL-8 release in the monoculture after Cinnamon exposure, no increase of inflammatory cytokines (IL-8 and MCP-1) release was observed. These findings point out that multiple assays with different in vitro models are able to discriminate the acute inhalation toxicity of CAs from liquids with different flavors, providing the companies and regulatory bodies with useful tools for the preliminary screening of marketable products.

Keywords: condensed aerosol; e-cigarette; e-liquid; in vitro systems; inhalation toxicology.

Figure 1

The Vaping machine, custom-made apparatus for generating and condensing vapors from e-CIGs. In the picture it is showed the four channel vaping machine (M) set with four atomizers (A) through 510 socket adapter. Each atomizer is controlled by one Evolv DNA 75 circuit (D) and is plugged to a collection system for toxicological studies (T) and a condensation cilinder (C) for chemical assays.

Figure 2

Cytotoxic effects of condensed aerosols (CAs) from e-liquid samples on monoculture A549 cells. MTT viability test was performed after 24 h of exposure to CAs belonging to e-liquids with (18 mg/mL, black bars) and without (grey bars) nicotine. Bars represent the percentage of viable cells with respect to the control (unexposed cells, white bar), considered as 100%. Data are presented as mean±SEM of at least 5 independent experiments. * p < 0.05; unpaired t-test over the control; @ p < 0.05; unpaired t-test over Base 0; § p < 0.05; unpaired t-test over Base 18.

Figure 3

Pro-inflammatory cytokines released by monoculture A549 cells exposed to CAs from e-liquid samples with (18 mg/mL, black bars) and without (grey bars) nicotine; white bars represent control samples. (A), IL-8 release; (B), MCP-1 release. Bars represent the concentration (pg/mL) of the released protein. Data are presented as mean ±SEM of at least five different independent experiments. * p < 0.05; unpaired t-test over the control; @ p < 0.05; unpaired t-test over Base 0; § p < 0.05; unpaired t-test over Base 18.

Figure 4

Barrier integrity and cell viability in the in vitro alveolar-blood barrier (ABB) model exposed to CAs from different e-liquids with 18 mg/mL nicotine. (A) trans-epithelial electrical resistance (% TEER day13/day12) measured across the barrier; (B) cells viability percentage measured in the alveolar NCI-H441 cells (apical, grey histograms) and in the endothelial HPMEC cells (basal, black histograms). Data are presented as mean ± SEM of at least three different independent experiments. * p < 0.05; unpaired t-test over the control (untreated cells); § p < 0.05; unpaired t-test over Base 18.

Figure 4

Barrier integrity and cell viability in the in vitro alveolar-blood barrier (ABB) model exposed to CAs from different e-liquids with 18 mg/mL nicotine. (A) trans-epithelial electrical resistance (% TEER day13/day12) measured across the barrier; (B) cells viability percentage measured in the alveolar NCI-H441 cells (apical, grey histograms) and in the endothelial HPMEC cells (basal, black histograms). Data are presented as mean ± SEM of at least three different independent experiments. * p < 0.05; unpaired t-test over the control (untreated cells); § p < 0.05; unpaired t-test over Base 18.

Figure 5

Pro-inflammatory cytokines release in the ABB model. (A) IL-8 release by the alveolar NCI-H441 cells (apical, grey bars) and by the endothelial lung microvascular endothelial cells (HPMEC) cells (basal, black bars); (B) MCP-1 release by alveolar NCI-H441 cells (apical, grey bars) and by the endothelial HPMEC cells (basal, black bars). Bars = pg/mL of IL-8 and MCP-1 released by co-cultures. Data are presented as mean ± SEM of at least three different independent experiments. * p < 0.05; unpaired t-test over the control; § p < 0.05; unpaired t-test over basal Base 18.