Effect of sub-chronic exposure to cigarette smoke, electronic cigarette and waterpipe on human lung epithelial barrier function

Personal devices that deliver aerosolized nicotine for recreational purpose, known as Electronic Nicotine Delivery Systems or electronic cigarettes (EC) use an electric heating element to vaporize a liquid mixture (e-liquid) to produce an aerosol. Therefore, many potential harmful constituents from the combustion of tobacco in cigarette smoke (CS) are absent in EC. Thus, EC are currently accepted as a healthier alternative to ordinary cigarettes [1]. While long-term effects of EC in people are not known, animal studies have demonstrated that there are toxic effects of EC [2, 3], and in vitro studies have found some dysregulation in bronchial and lung epithelial cells, lung fibroblast, and alveolar macrophages [2].

Tobacco smoking is the primary risk factor of chronic obstructive pulmonary disease (COPD) development. COPD is a progressive illness caused by chronic lung injury, with clinical manifestations that can include bronchitis, airway reactivity, and emphysema [4]. Considering the increased incidence of e-cigarette or vaping product use-associated lung injury, the evaluation of the biological activity of EC aerosol in in vitro models has become a top priority [5]. Hence, we studied the effects of sub-chronic EC aerosol exposure on human bronchial epithelial function in an in vitro air-liquid interface model in comparison to CS, and tobacco waterpipe (TW) exposures, in similar conditions. As the lining of the respiratory tract, the epithelia form a physical barrier that effectively prevents inhaled environmental insults from damaging the sub-epithelial layers [6]. Not surprisingly, we and other researchers have demonstrated that CS exposure disrupts the connectivity between adjacent cells in the lung epithelium interfering with the function of cell-cell adhesion proteins, often members of the cadherin family of adhesion molecules [7‐12]. Additionally, there is evidence of detrimental effects of CS exposure on the mucociliary clearance of entrapped particulates in the lung track, which is mediated by the synchronous beats of protruding organelles in ciliated cells from the lung epithelium [13, 14]. Thus, our study informs on the effects of EC aerosol exposure on the lung epithelial function and a synergistic effect of nicotine in the e-liquid. We demonstrated that EC aerosol, as well as CS and TW can distinctly harm the lung epithelial barrier, an observation that in the context of previous experiments, supports a correlation with the time of exposure. Therefore, our report emphasizes the need for more studies of long-term health and harmful consequences that may affect users of ECs.

There is increasing recognition that CS can disrupt the epithelial barrier due to the toxicity of the mixture of chemicals acting through several molecular mechanisms that include MAPK (mitogen-activated protein kinase), TGFβ (transforming growth factor beta-1), and reactive oxygen species [25‐29]. EC aerosol produced from e-liquid (propylene-glycol and glycerol) and smoke from TW contain many of the same toxicants found in CS; therefore, it is not surprising that similar pathways may be activated by exposure to EC and TW [10, 30‐33]. For the determination of toxic effects, we used primary differentiated human lung epithelial cells as they replicate the human lung physiology well enough for assessments of health risks [34]. Two cell models were used for CS-EC and TW. Therefore, we minimized comparisons across the two sets of experiments to account for functional differences in cell barrier permeability observed in this study and as reported by other researchers [17]. The changes in fluorescence levels that resulted from CS exposure (FITC-Dextran assay decreased 5.6 times, and TEER decreased to 50% compared to control air) are in line with the previously observed loss of permeability after CS exposure [8, 11]. In the exposure to EC with or without nicotine, we observed a trend to decreased permeability with FITC-Dextran assay and when using TEER to measure barrier properties, the toxic effect of EC (1.2%) became significant, decreasing permeability by 60% compared to air control. TEER measures the movement of ions across an electric gradient applied parallel to the monolayer, with a decreased resistance reflecting a loss of monolayer integrity [35]. A significant change of TEER in cells exposed to TW may be masked by a low number of samples. Nevertheless, our previous studies indicate that the permeability of epithelial monolayers increase only after repeated exposure to cigarettes, suggesting that disruption of permeability, as observed in this study, is in proportion to the length of the exposure [11, 16, 36]. Indeed, an absence of epithelial barrier disruption in in vitro models of very short exposures to EC has been reported [37]. In addition, flavors in the e-liquid may positively modulate permeability as well, which was previously shown in cell lines exposed to EC [38]. In our exposure with EC tobacco flavor without nicotine we do see a tendency to a reduction in barrier integrity based on TEER measurements that is significant in the presence of nicotine. Furthermore, we demonstrated that the presence of nicotine in EC (1.2%) deteriorates mucociliary clearance by decreasing CBF by 59% with respect to the control air. Mucociliary clearance mechanism is one of the critical protective functions of the epithelial monolayer that is performed by the rhythmic and directional beating of cilia on a significant subpopulation of the monolayer [39]. Previous studies have shown shortening or loss of cilia, as well as decreases in CBF, are a direct result of the volatile organic compounds [40‐44]. Whereas CS has a high concentration of volatile organic compounds, EC aerosols have concentrations that vary with the abundance of flavoring components and the glycerol:propylene-glycol ratio in the e-liquid [45‐48]. Here, we show a significantly decreased CBF and pixel movement upon exposure to CS or EC (1.2%) by 62 and 47%, respectively. We and other researchers have shown that CS exposure leads to decreased CBF and number of cilia in epithelial cell monolayers that is in line with previous studies in which deteriorating effects of CS on ciliated cells were reported [11, 13, 49]; however, the results obtained from our chronic EC (1.2%) exposure are novel. The plausible additive effect of nicotine in decreasing CBF, suggested in our experiment, is in striking contrast with recent report that showed nicotine exposure of cultured epithelial cells has no effect on CBF [50]. Therefore, based on our results, a synergistic effect of nicotine and additional components of EC such as flavors, as well as other components of EC, need to be tested specifically in settings of the aerosolized EC exposure. For example, nicotine showed an additive effect in decreased mucociliary clearance mechanism and other lung epithelial lesions in cells and in vivo experiments of EC exposures [51, 52].

We have identified specific trends in effect with TW, albeit with weak statistical power due to a small sample number (3 replicas) in the TW experiments. While this does hamper any comprehensive conclusions, we do believe that this in vitro model is ideal for sub-chronic exposures and can be adapted for chronic exposures of TW and other tobacco products in need of urgent toxicological assessment (e.g. EC) [53, 54]. We note that whereas a (nonsignificant) increase in CBF may be a distinct effect of TW in our study, is not unprecedented. We have recognized several instances of similar responses to CS in vitro [42, 55], ex-vivo [56, 57], and in vivo [58]. Moreover, we have previously shown that continuous monitoring of CBF in lung epithelial cells exposed to CS revealed temporal fluctuations, with the appearance of CBF curve inflections in cells, minutes and hours into the recovery time [20]. Thus, while it could be attributed to distinct components present in the TW smoke [59], it is also entirely plausible that chronic exposures to TW would cause decreases in CBF, similar to that seen with EC and CS. We are currently planning to test this in our laboratory.

We have previously demonstrated a 5 to 10 times increase in cellular velocity of epithelial cells exposed to CS compared to control [20], and here we showed evidence of 2.5 and 2.6 times increase after exposure to CS and EC (1.2%), respectively. The underlying molecular mechanisms that control cell velocity after CS or EC (1.2%) exposures are beyond the scope of this report, and it will be examined in the future, but this observation correlates with the loss of cell-cell adhesion proteins and altered actin-myosin contractility [11]. Nevertheless, accumulating evidence suggests that a decreased expression of the CDH1 gene, the main component of adhesin junctions in lung epithelial cells, is a marker of the loss of barrier function in response to CS [60, 61]. Here, we show that EC (1.2%) tended to decrease CDH1 expression, while CS decreased by 39% compare to the control air. Therefore, future work is needed to explore if exposure to EC (1.2%) alike that of CS, impairs the stability of adherence junctions that are necessary to keep adequate cell-cell interactions in the lung epithelium.

We devised a sub-chronic exposure system to measure the toxic effects of EC and other tobacco products by monitoring cell barrier disruption. Our results, albeit with limited number of samples, support that sub-chronic exposure to EC with 1.2% nicotine may disrupt the airway epithelia, similar to CS. More research with larger number of donor samples is being planned for the future. While, in our experimental conditions, EC 0% and TW caused less epithelial barrier disruption, future experiments need to address the toxicological significance of prolonged exposure and nicotine concentration as well as other components of these tobacco products as determinants of harmful effects. Specifically, for EC and TW, the addition of a variety of flavors needs to be addressed, as they are not proven safe when inhaled. Our system is adequate for future studies of chronic exposures that are required to understand the health effects of long-term users of EC and TW.