Mechanisms and drivers of epithelial to mesenchymal transition in COPD

Chronic obstructive pulmonary disease (COPD) is a progressive and irreversible limitation of airflow which predominantly affectsa smokers. One of the mechanisms behind the development and progression of COPD is thought to be epithelial-mesenchymal transition (EMT). EMT is a process whereby epithelial cells, which normally act as the body’s primary means of defence against injury, external pathogens and other damaging factors, lose their ability to form a cohesive barrier and instead take on characteristics of a motile, fibrotic and mesenchymal phenotype. There is in vitro evidence that cigarette smoke extract (CSE) can directly induce EMT through production of transforming growth factor-β1 (TGF-β). However, isolated cells in these studies are typically sourced from people with established COPD or with lung cancer. It is not known whether this CSE-TGF-β-EMT pathway occurs at earlier stages of the disease, for example in smokers with normal lung function or in pre-COPD airflow obstruction. Also unclear is whether current therapies for COPD affect the CSE-TGF-β-EMT pathway and if they have beneficial effects early in the development of the disease. The aims of this study were to investigate the CSE-TGF-β-EMT pathway in isolated epithelial cells from healthy non-smokers, smokers with normal lung function and people with airflow limitation which does not reach the threshold for clinically diagnosed COPD. In doing so, this study attempted to determine whether EMT occurs at very early disease stages, perhaps even before manifestation of clinically significant airflow limitation, and to investigate the effects of CSE on healthy cells from non-smokers. At the same time, the effects of three current COPD therapies - a long acting β-agonist (salmeterol xinafoate), a long acting muscarinic agonist (tiotropium bromide) and a corticosteroid (fluticasone proprionate) – on both early disease stages and the CSE-TGF-β-EMT pathway were investigated. Two cell culture models were examined: primary bronchial epithelial cells (pHBECs) and immortalised BEAS-2B airway epithelial cells. The models were tested for expression of EMT markers, as well as markers for the EMT-associated TGF-β and TWIST signalling pathways, using real-time qPCR, immunocytochemistry/histochemistry and ELISA. Cells from non-smokers, current smokers with normal lung function and both smokers and ex-smokers with airflow limitation were shown to survive both isolation and cryopreservation. The protein expression of EMT markers in the cultured cells agreed with the signals seen in biopsies from the same groups, indicating that phenotype was maintained in culture. However, cells from people with chronic airflow limitation did not show evidence of active EMT, either in culture or in the biopsies, when compared to cells from non-smokers or smokers with normal lung function. Despite BEAS-2B cells being widely used to study healthy bronchial epithelial cells when primary cells are unavailable, this study demonstrated that the immortalised cell line is not an ideal model for primary cells. Not only were the BEAS-2B cells more mesenchymal than primary cells taken from healthy non-smokers, but they also demonstrated differences in how they responded both to stimulation with TGF-β or CSE and to the drug treatments used in this study. In particular, the molecular mRNA level differences between the immortalised and the primary cells suggested that BEAS-2B cells may not be good models for healthy bronchial epithelium. In this study, TGF-β also exhibited lower levels of activity in BEAS-2B cells than expected from the literature, not managing to fully induce EMT in the immortalised cell line, nor in the primary cells taken from non-smokers. However, the cells trended towards an EMT-like expression profile. Exposure to CSE failed to elicit a full EMT response in primary cells from non-smokers, although it did elicit a partial EMT-like phenotype without affecting the Smad signalling pathway. This further suggests that acute exposure to CSE is not sufficient to trigger the CSE-TGF-β-EMT pathway in healthy airway epithelial cells, although it may trigger other pathways to promote EMT-like changes. Of the three drugs studied, neither salmeterol, tiotropium or fluticasone had any significant suppressive effect on EMT in either primary or BEAS-2B cells. Tiotropium, in particular, appeared to actually promote EMT in primary cells taken from non-smokers and smokers with normal lung function. Overall, this study confirmed that primary bronchial epithelial cells can be successfully cultured and demonstrated that cryopreservation of these cells is possible, although smoking status had a negative impact on their viability. BEAS-2B cells did not appear to be a reliable model of healthy primary epithelial cells, particularly at the mRNA level of EMT-related expression. Cells from people with airflow limitation which did not reach the level of COPD did not appear to be undergoing active EMT, suggesting that EMT may only be active in fully developed COPD. Of the drugs used for treatment of COPD, only salmeterol indicating a trend towards suppression of EMT, while tiotropium and fluticasone appeared to have no suppressive effect on the process, with tiotropium promoting a partial EMT phenotype