Airway epithelial cells exposed to wildfire smoke extract exhibit dysregulated autophagy and barrier dysfunction consistent with COPD

The airway inflammation and remodelling observed during chronic obstructive pulmonary disease (COPD) is driven by exogenous noxious stimuli that activate the airway epithelium, which responds by releasing alarmins and pro-inflammatory factors that promote the recruitment of inflammatory cells [1, 2]. Consequently, identifying discrete biochemical effects caused by the exacerbating factors that potentiate this chronic airway wounding scenario is a priority for COPD research [3, 4]. In addition to cigarette smoke, exposure to smoke from burning wood and the combustion of biomass for cooking and heating which is prevalent in underdeveloped countries are associated with the development of COPD [5]. Further to this, the extensive clouds of soot and gas produced during a wildfire event or prescribed burns are a far less avoidable form of smoke exposure, and have the potential to exacerbate the respiratory symptoms of the surrounding population for an extended period of time (reviewed in [6]).

There are a number of reasons why wildfire smoke is an important consideration for individuals with COPD; 1) the rising incidence of COPD, which is already the third leading cause of death in the USA [7], 2) the expansion of metropolitan populations in proximity to forested regions and prescribed burns [8], 3) the increasing frequency of wildfire events associated with elevated atmospheric temperatures and reduced precipitation [9], and 4) the pressure on hospital resources during large scale crises. Further, advanced degeneration of the airways generally effects elderly sufferers of COPD who may be unable to escape the rapid onset of wildfire smoke, and are therefore more likely to be exposed to higher levels of the particulate matter that can penetrate into the lower airways [6, 9‐11]. While exposure to wildfire smoke is less frequent than, for example, exposure to biomass smoke within dwellings, wildfire smoke can cover vast populated regions with highly concentrated and volatile chemicals (e.g. acrolein, benzene, and phenols) produced after the combustion of unprocessed organic material which can persist for several hours [5, 6, 9, 10]. As a result, there is a high probability that individuals suffering from COPD will experience an exacerbation of their symptoms that requires clinical support, and during a period when hospital resources may be limited by other emergencies related to the wildfire. This has been reported in several epidemiological studies which show a statistically significant increase in COPD exacerbations as a result of exposure to wildfire smoke, and an increase in hospital admissions to manage the respiratory symptoms beyond the efficacy of self-administered medications [6, 9, 12, 13]. Indeed, in situations where people are exposed to intense heat or have a severe respiratory infection, co-exposure to prolonged periods of smoke may cause damage to the airway epithelium that promotes the pathogenesis of COPD [9]. This scenario is particularly relevant in countries such as Australia, which has a high prevalence of respiratory disease and where agricultural communities are required to conserve highly combustible adjacent bushlands that have evolved with bushfires as a normal phenomenon that promotes revegetation [8].

Dysfunction of the epithelial compartment in response to cigarette smoke exposure is known to contribute to the hall mark features of COPD, such as the protracted secretion of inflammatory factors, recruitment of harmful immune cells and production of cytokines such as TGF-β that remodel the airways (e.g. [3, 14]). Given that airway epithelium is the immunological interface for harmful environmental stimuli presented by the atmosphere, it then follows that airway epithelial cells (AEC) are major orchestrators of the nature and magnitude of the inflammatory response and exacerbations suffered by individuals with COPD that are exposed to wildfire smoke. Further to this, any deficits in mucociliary function caused by the extreme air temperatures associated with wildfire smoke will have immediate consequences for effective gas exchange at the alveoli. Surprisingly, there is a paucity of data describing how the airway epithelium responds to wildfire or any other form of environmental smoke exposures. Limited studies have reported that smoke exposures related to the combustion of wood or biomass initiate cytotoxic changes in the epithelium caused by the generation of reactive oxygen species, often due to dysregulation of the enzymes that counter the accumulation of intracellular oxygen radicals (e.g. [15‐17]). One study has shown that chronic exposure to biomass smoke causes morphological changes to the epithelial compartment consistent with emphysema and the small airway damage elicited by cigarette smoke [18]. In line with this, another group has shown that the epithelium can undergo metaplastic and dysplastic changes in response to chronic exposure to the pollutants in biomass smoke [19]. However, more information is needed to further our understanding of how the smoke generated by the combustion of forested regions, which has a different composition to other forms of environmental smoke, impacts the viability and function of the immunological barrier imparted by the airway epithelium.

Central to the normal function of the airway epithelium is the maintenance of a semi-permeable barrier, appropriate immunological surveillance and reactivity to environmental stimuli, and normal viability and turnover with the underlying progenitor cells. We and others have reported that these characteristics are perturbed during COPD and in various models of cigarette smoke exposure, and is instead replaced by a leaky, fragile barrier which is composed of AEC with a low activation threshold [14, 20, 21]. We and others also showed that a disease-related block in autophagy is a fundamental phenomenon that potentiates many of the pathological alterations ascribed to the epithelium (and other airway cells) in the context of COPD, such as susceptibility to oxidative stress, accelerated senescence and apoptosis, and a pro-inflammatory phenotype [22, 23].

Hence, here for the first time we employ primary human AEC to examine the effects of wildfire smoke using an acute exposure model. We report that while AEC exposed to wildfire smoke extract (WFSE) exhibited a similar tendency for unscheduled apoptosis when compared with cells exposed to cigarette smoke extract (CSE), WFSE produced a significant increase in epithelial permeability and blockade in autophagic flux. These findings suggest that individuals who suffer from COPD have a heightened risk of exacerbations during a wildfire event which can be worsened by the increasing use of therapeutics such as macrolides that block autophagic flux.

There is a paucity of information describing the influence of wildfire smoke exposure as a contributing factor in respiratory disease. Effective function of the respiratory epithelium is a central requirement for the airways to counter the immediate challenges presented by the airborne particles and gases produced during a forest fire. This is an even more important consideration when the airway epithelium is dysfunctional, as is the case for COPD, where it exhibits a complex wound healing phenotype. Here, for the first time, we applied WFSE to primary models of human AEC in the context of a moderate exposure period (6 h) to approximate the situation AEC may encounter during exposure to wildfire smoke (e.g. farmers defending a property). We found that while treatment with either WFSE or CSE produces similar cytotoxic and programmed death responses, exposure to WFSE elicited autophagic insufficiency and a significantly greater magnitude of barrier dysfunction in the primary human epithelial cultures.

We first asked whether epithelial cell survival was affected by WFSE, as per the situation during COPD where aberrant turnover of AEC is a frequently observed phenomenon elicited by the cigarette smoke and inflammation [28, 29]. WFSE induced a significant cytotoxic response in the normally resilient 16HBE14o- cell line (Fig. 1a). This was reduced to background levels of cell necrosis after limiting the exposure period from 24 h to 6 h for the 16HBE14o- cells and in the primary SAEC models (Fig. 1b). However, a significant apoptotic response was observed for the SAEC after 6 h exposure to 10% WFSE, as evidenced by the cleavage of the caspae-3 substrate PARP (Fig. 2). These results were consistent with a report by Pavagadhi et al. (2013) who identified significant apoptosis due to the factors generated during the combustion of forest biomass materials, albeit in the A549 alveolar cancer cell line [17]. The shift from a necrotic response to programmed cell death is in keeping with the normal (if unscheduled) clearance of damaged AEC that enables their efficient replacement by the underlying progenitor stem cells [20, 30]. This apoptotic response was accompanied by a significant reduction in the potent inhibitor of mitochondrial depolarisation, Bcl2, which protects against extrinsic apoptotic stimuli such the toxins and oxidants contained in wildfire smoke. Of note, the induction of apoptosis and the corresponding decrease in Bcl2 was similar for WFSE and CSE. Hence, an important follow-up inquiry would be to co-treat AEC with both WFSE and CSE to model the exposure of a current smoker to wildfire smoke, and examine the interaction between the signalling pathways that promote cellular survival and apoptosis in the context of COPD. Further to this, given that the treatments between 1 and 5% WFSE did not promote apoptosis suggests that the 10% WFSE exposure may overcome or otherwise inhibit the survival mechanisms that protect against programmed cell death in AEC.

Bcl2 is also an inhibitor of autophagy [31], which is a fundamental cell survival process that is central in maintaining cellular homeostasis during periods of cell stress and starvation, and its dysregulation has been consistently associated with the pathogenesis of COPD (e.g. [32, 33]). We observed a significant increase in LC3-II (an essential component of the autophagic apparatus) following 5–10% WFSE and 10% CSE treatment (Fig. 2a), which was likely a homeostatic response to the accumulation of damaged organelles and misfolded proteins that are generated by (for example) the oxidants and toxins that are present in these exposures [22, 32, 33]. Of note, and in contrast to the other exposures, we observed a significant increase in Sequestosome following 10% WFSE treatment, concomitant with the elevation in LC3-II abundance (Fig. 2a, b). As Sequestosome (an adapter protein) is normally co-degraded with the cargo it shuttles to the autophagosome, our findings indicate a block in autophagic flux beyond the homeostatic influence imparted by the induction of LC3-II. Indeed, a block in autophagic flux was also exclusively observed for the 10% WFSE exposure in THP1 macrophages (Additional file 2: Figure S2). Hence, determining whether the autophagic insufficiency is a response related to an excess of accumulating cargo or occurred due to a defect in the regulators of autophagy may be an important consideration for COPD in the context of wildfire smoke exposure. This may be particularly critical for COPD patients that are prescribed macrolide antibiotics which are potent inhibitors of autophagic flux (e.g. [34]). The consequences for AEC with reduced autophagic potential is a diminished capacity to recycle potentially harmful cytosolic debris generated by the wildfire smoke. In addition, autophagy is an influential cytoprotective process that counters oxidative stress, inflammation, and senescence which are all central aspects in the pathogenesis of COPD [32]. Further, defective autophagy is also associated with a poorly understood link with aberrant apoptosis (“autophagic cell death” [35, 36]). We have found that unto itself CSE exposure is not enough to initiate defects in autophagy in primary models of human airway epithelium without further COPD-related stimuli such as nutrient deprivation [21]. Hence, given WFSE elicited a block in autophagic flux suggests this form of environmental exposure presents a toxin (or group of toxins) that prevents effective autophagy beyond the influence imparted by a similar level of cigarette smoke.

Another disease-related phenomenon in COPD is the diminished efficacy of the junctional apparatus that maintains epithelial continuity, and which perpetuates the chronic would-repair process and airway remodelling [37]. Disruption of the sophisticated communication mechanisms that integrate adjacent AEC and progenitor cell populations (e.g. tight junctions, adherens junctions, and gap junctions) has serious consequences for the array of immunological activities that are involved in clearing noxious particles contained in environmental smoke. To examine this for WFSE, we employed an ex vivo model which closely approximates the epithelium observed in vivo (Additional file 1: Figure S1). An increase in ionic conductance of the paracellular pathway became evident after six hours for the WFSE exposures ranging from 2.5–10% (Fig. 3a). The 10% WFSE treatment elicited a significantly greater reduction in electrical impedance than 10% CSE, and this relationship became more evident beyond six hours when the effects of apoptosis and necrosis contribute to a reduction in epithelial integrity. In perhaps a more physiologically relevant analysis using fluorescent tracers (i.e. which mimics a situation whereby smoke particles passage the epithelial layer), an increase in paracellular molecular permeability was readily evident after exposure to 5 and 10% WFSE at six hours, while the effect of 10% CSE was comparable to the control (Fig. 3b). After six hours there was a vast increase in permeability for 5–10% WFSE, that was significantly greater than the effect of 10% CSE (Additional file 3: Table S1). Also, while the apoptosis and necrosis induced by 10% WFSE undoubtedly contributed to the reduction in epithelial integrity (Figs. 1 and 2), six hours treatment with 5% WFSE elicited a biochemical effect (prior to significant anoikis) that destabilised the TJ junction apparatus. In support of this, we observed a reduction in the abundance of the TJ proteins ZO-1 and Claudin-1 following six hours exposure to 5% WFSE, and in the absence of programmed cell death (Fig. 4a, b). Furthermore, while we did not detect increased activation of the potent pro-inflammatory transcription factor NFκB p65 (Fig. 2a), we observed a significant increase in IL-6 in response to WFSE (Fig. 5). IL-6 is an influential component of the inflammation in COPD, which is also secreted from AEC in response to the toxins contained in environmental smoke [38, 39]. However, in the normal situation IL-6 also plays an important role in the regeneration of AEC and co-operates with the inflammatory cells to repair the epithelial layer [40, 41]. Hence, in a scenario where wildfire smoke potentiates epithelial fragility, the secretion of IL-6 in response to defects in TJ integrity may be an important contributing factor in the chronic wound healing phenotype that is frequently ascribed to the epithelium during COPD.

Hence, we show that factors presented by wildfire smoke contribute to the deterioration of the airway epithelium, to a greater degree than cigarette smoke, most notably by blocking the fundamental cell survival processes governed by autophagy and disruption of epithelial barrier activity. To our knowledge, this is the first investigation that describes the effects of wildfire smoke using models of the epithelium in the context of COPD or any other respiratory disease. One limitation of our model is the use of the genera Eucalyptus and Acacia as a surrogate biomaterial. While these genera constitute the majority of Australia’s flora, it may be important to compare the effects elicited by wildfire products generated in countries with distinctly different vegetation. Indeed, a “standard blend” of forest-related biomass could serve as an important reference exposure in independent studies. Further, while we have progressed this inquiry beyond secondary cell lines, another important question is how AEC derived from smokers and COPD sufferers respond in similar models. Indeed, a greater level of utility can be achieved by applying an aerosol exposure system to ex vivo models to mitigate the issues related to dissolving smoke into a liquid vehicle [5], and which may be used to control particle size, gas composition, and exposure durations that vary with the distance wildfire smoke travels through the atmosphere [42]. In conjunction with a mouse model of emphysema or COPD, an aerosol system can also provide information relating to exacerbations, inflammatory responses and the consequences of smoke particle deposition in the airways, all of which potentiate the epithelial remodelling that is a hallmark feature of COPD. Given the increasing incidents of wildfires in proximity to an ever aging population who carry a high burden of respiratory disease, such inquires have significant potential to inform therapeutic and clinical strategies to ease the burden of hospitalisation caused by large scale wildfire events.

Wildfire smoke has a significant negative impact on the survival of AEC and the maintenance of the immunological barrier imparted by the airway epithelium. Perhaps central to these is the disruption to the fundamental and multifaceted survival processes governed by autophagy, which are already dysregulated in COPD. Hence, targeting deficits in autophagic activity may prevent or resolve the toxic influences presented by wildfire smoke for individuals who are susceptible to exacerbations of COPD.