Autophagy in chronic lung disease

Abstract

The development of chronic lung disease occurs as a consequence of multiple cellular events that involve an initial insult which often leads to the development of chronic inflammation, and the dysregulation of cellular proliferation and cell death mechanisms. Multiple cell types in the lung are key to the respiratory and protective/barrier functions necessary to manage the chronic exposures to environmental, mechanical, and oxidative stressors. Autophagy is essential to lung development and homeostasis, as well as the prevention and development of disease. The cellular process involves the collection and removal of unwanted organelles and proteins through lysosomal degradation. In recent years, investigations have addressed the roles of autophagy and selective autophagy in numerous chronic lung diseases. Here, we highlight recent advances on the role of autophagy in the pathogenesis of asthma, chronic obstructive pulmonary disease and emphysema, pulmonary arterial hypertension, and idiopathic pulmonary fibrosis.

1. Introduction

Macroautophagy (autophagy) is a highly conserved cellular process that drives the degradation and removal of unwanted proteins and organelles. This occurs under basal conditions to maintain homeostasis and is also stimulated during periods of cellular stress (e.g., nutrient deprivation, hypoxia, growth factor depletion, and infection).¹ Activated autophagic machinery recover nutrients through non-specific breakdown of proteins and organelles by the formation of autophagosomes that are then guided toward lysosomal degradation. In contrast, a process known as selective autophagy, drives the removal of distinct cytosolic structures such as mitochondria (mitophagy), protein aggregates (aggrephagy), and bacteria (xenophagy).¹ Beyond its maintenance role, autophagy is key to normal and pathological conditions and has been described as fundamental to the severity of several human diseases.²

During nutrient deprivation, mechanistic target of rapamycin complex 1 (mTORC1), a negative regulatory of autophagy, is inhibited by the presence of abundant levels of adenosine monophosphate (AMP), through the activity of 5′ AMP-activated protein kinase (AMPK).³ Inhibition of mTORC1 leads to the activation and subcellular localization of the Unc-51-like kinase 1 (ULK1) complex to the endoplasmic reticulum. ULK1 complex then triggers the nucleation phase of autophagy, a critical step in phagophore formation, through phosphorylation of components of the class III phosphoinositide 3-kinase (PI3KC3) complex I that includes key autophagic proteins Beclin-1, ATG14 (autophagy-related 14), VPS34 (vacuolar protein sorting 34), p115, and AMBRA1 (autophagy and Beclin-1 regulator 1) (Fig. 1). Activation of Beclin-1 and associated proteins enhances phosphatidylinositol 3-phosphate (PI3P) activity leading to the recruitment of WD repeat domain phosphoinositide interacting proteins and zinc finger protein FYVE domain-containing protein-1 to the developing autophagosome via interaction with PI3P-binding domains.³ Phagophore expansion occurs with further recruitment of autophagy-related proteins ATG5, ATG12, ATG3, ATG16L1, and, eventually, microtubule-associated protein 1 light chain 3 (LC3) to the phosphatidylethanolamine located in the lipid membrane. LC3 recruitment facilitates further sequestration of ubiquitinated material into the developing autophagosome. Sealing of the autophagosomal membrane gives rise to the double membraned mature autophagosome which eventually fuses with the lysosome to allow degradation of the autophagic cargo (Fig. 1).³

Fig. 1.

Activation of autophagy and autophagosomal formation. In a cellular state of nutrient deprivation, activation of 5′ AMP-activated protein kinase (AMPK) occurs when cellular levels of adenosine monophosphate (AMP) accumulate. This leads to the inhibition of mTORC1 which in turn supports the activation Unc-51-like kinase 1 (ULK1) complex. The ULK1 complex facilitates the initiation of the nucleation phase of autophagy through phosphorylation of the Beclin-1-PI3KC3 complex which includes Beclin-1, ATG14 (autophagy-related 14, VPS34 (vacuolar protein sorting 34), p115, and AMBRA1 (autophagy and Beclin-1 regulator 1). Activation of these proteins enhances phosphatidylinositol 3-phosphate (PI3P) activity promoting autophagosomal membrane formation. Autophagosomal elongation is stimulated by conjugation of ATG5 to ATG12 and ATG16L. Through the function of the autophagy related proteins (ATG4, ATG12, ATG16L, ATG4, ATG3, and ATG7) the microtubule-associated protein 1 light chain 3 (LC3) is modified with the cellular lipid phosphatidylethnanolamine (PE) and converted to the conjugated form LC3-II. LC3-II recruitment facilitates further sequestration of ubiquitinated material into the developing autophagosome. As a final stage, the autophagosome fuses with the lysosome to allow degradation of autophagic cargo.

In recent years, alterations in the autophagy pathway have been implicated in an increasing number of human diseases, including cancer, neurodegenerative disease, cardiovascular disease, and inflammatory disease.² While a complete understanding of the role played in either prevention or progression of certain diseases remains to be fully understood, several lines of research have highlighted changes in autophagic cellular functions as key to the pathogenesis of numerous chronic lung diseases. In this review, we aim to highlight the importance of autophagy as both a friend and foe in the pathogenesis of several chronic lung diseases.

2. Autophagy and the lung

The primary function of the lungs is to facilitate the delivery of oxygen gas to the arterial system, allowing perfusion of distal tissues, and the elimination of waste gases from the body. The respiratory system is composed of nasal passages, conducting airways (trachea and bronchi), bronchioles, and the alveoli-capillaries network, making up the primary ventilatory unit.⁴ The right and left bronchi expand off the trachea and contain cartilage and submucosal mucus secreting glands and neuroendocrine cells (that release serotonin, calcitonin, and gastrin releasing peptide) within their walls. They branch further to become bronchioles, small airways that lack both glands and cartilage, that extend to form even smaller (<2 mm diameter) terminal bronchioles. Respiratory bronchioles stem from terminal bronchioles which form into alveolar ducts then sacs.⁴ The alveolar wall serves as the site of gas exchange and consists of capillary endothelium, basement membrane and surrounding interstitial tissue, alveolar epithelium (type I and type III), and alveolar macrophages.

The components of the lung are directly exposed to the outside environment and serve as the first lines of defense to mitigate the challenges presented by chronic exposure to environmental (bacteria, virus, toxins), mechanical, and oxidative stress.⁵ Antioxidant defenses, innate and adaptable immune responses, and several mechanisms of cell death play key roles in the maintenance of normal lung function. A growing body of evidence supports the paradigm that autophagy activated by environmental and cellular stresses represents a critical intracellular process for the maintenance of lung homeostasis. The cytoprotective functions of autophagy facilitate the removal of damaged organelles and protein aggregates that occur as a consequence of acute and/or chronic stressors, aging, and genetic or epigenetic predispositions. When these cellular processes are disrupted, pathological states along the components of the respiratory tree ensue. Significant roles for autophagy have been defined across various types of chronic lung disease (obstructive lung diseases, pulmonary vascular disease, and restrictive lung disease) and whether it is protective or harmful in certain disease states remains to be understood.

3. Obstructive lung diseases

3.1. Asthma

Asthma is an obstructive disease of the airways typified by chronic inflammation, hyperresponsiveness (driven by muscular bronchoconstriction), mucus overproduction, and airway remodeling. As a clinical entity, asthma is a debilitating disease that effects over 300 million people worldwide and patients suffer from recurrent episodes of wheezing, breathlessness, chest tightness, and cough.⁶ Symptoms are triggered by diverse environmental exposures, infections, stress, and exercise, where excessive type 2 helper T (Th2) cell activation drives cytokine and chemokine production leading to airway narrowing and mucus production. Over time, abnormal Th1, Th2, and Th17 responses lead to chronic persistent inflammation and airways remodel with bronchial smooth muscle hypertrophy and increased vascularity, collagen deposition, and presence of mucus glands.^6–8 The mechanism driving the acute and chronic phases of asthma are complex, but many studies have established cross-talk between intracellular autophagic events and the pathogenesis of asthma (Table 1).

Table 1.

Murine and human studies highlighting the deleterious role of autophagy in asthma, COPD, and PAH.

Disease	Autophagy deleterious
Obstructive
Asthma	Murine studies
	– IL-13 response and mucus hypersecretion in airway epithelial cells reduced in Atg16ll knockout mice9 – Autophagy inhibition blocks TGF-ß3 signaling and mucus production in bronchial epithelial cells10 – Autophagy blockade in airway B cells reduces asthma phenotype in vivo and in vitro11
	Human data
	– Bronchial fibroblasts and epithelial cells from severe asthmatics with increase autophagosomes12 – SNPs in SQSTM1, MAP1LC3B, BECN1, and ATG5 were found in severe asthmatics with reduced lung function12 – Elevated levels of autophagy in neutrophils and eosinophils isolated from severe asthmatics13,14
Chronic obstructive pulmonary disease (COPD)	Murine studies
	– Reduction in autophagy dampened inflammation, cellular dysfunction and apoptosis in chronic cigarette smoke exposure15,16 – Loss of Atg5 in bronchial epithelial cells reduced airway inflammation to particulate matter17 – Global loss of Beclin-1 and Lc3b blocked cigarette smoke induced emphysema15 – Cellular dysfunction limited in autophagy impaired mice18 – Inhibition of mitophagy via loss of Pink1 −/− reduced response to cigarette smoke19 – Cigarette smoke exposed neutrophils have increased autophagy and compromised ability to ingest pathogens20
	Human data
	– Increased autophagosomes in human tissue15 – Polymorphisms in EGR-1 and ATG16L1 identified as risk factors in COPD tissue15,21
Vascular
Pulmonary arterial hypertension (PAH)s	Murine studies
	– Loss of Ucp2 and increased mitophagy worsened hypoxia-induced pulmonary hypertension22 – Reduction in autophagy was linked to efficacy of liraglutide in monocrotaline model23 – Autophagy activation linked to progression of right ventricular remodeling and fibrosis24

Initial in vivo and in vitro studies directly correlated the degree of interleukin-13 (IL-13) response, a cytokine central to the Th2 response observed in asthma (mucus hypersecretion and ROS generation), with the level of autophagic activation observed in airway epithelial cells.^9,25 Specifically, mice deficient in autophagy through the deletion of Atg16l1 had a blunted IL-13 response, evidenced by less mucin MUC5AC production following cytokine release when compared to wild-type control animals.⁹ In vitro studies using human airway epithelial cells, where genetic (ATG5 and ATG14) or pharmacological (3-methyladenine or bafilomycin A₁) inhibition of autophagy also revealed a dampened IL-13 response.⁹ Additional studies also linked activation of autophagy to secretion of IL-18 in response to Alternaria alternata, an outdoor allergen that causes allergic airway disease.²⁶ Autophagic events in bronchial epithelial cells are now positioned downstream of transforming growth factor-beta 3 (TGF-ß3) signaling which drives MUC5AC hyperexpression and mucus production and inhibition of autophagy appears to halt the TGF-ß3 effects.¹⁰ In airway B cells, through functions of IL-4, autophagy is upregulated and when deficient in autophagic proteins the asthma phenotype is perturbed both in vivo and in vitro.¹¹ Of high clinical relevance are the findings from several studies in the murine ovalbumin model of asthma where autophagy inhibition, via 3-methyladenine treatment or Atg5 shRNA knockdown, showed a reduction in airway hyperresponsiveness, eosinophilia, and inflammation.^27,28

Human studies also demonstrate the relevance of autophagy to the pathogenesis of asthma, as autophagosomes are increased in bronchial fibroblasts and epithelial cells from patients with moderately severe asthma.¹² Additionally, genetic association identified single nucleotide polymorphisms (SNPs) in genes encoding for key autophagic proteins ULK1, sequestosome 1 (p62/SQSTM1), microtubule-associated protein 1 light chain 3 beta (MAP1LC3B), Beclin-1 (BECN1), and ATG5 that correlated with a reduction in lung function.¹² Likewise, a variant in ATG5 has also been identified in childhood asthma.²⁹ Sputum and peripheral blood studies revealed elevated levels of autophagy in sputum granulocytes and peripheral blood eosinophils and neutrophils isolated from patients with severe asthma when compared to non-severe asthmatics and healthy controls.^13,14 In these studies, persistent activation of autophagy was found when key cytokines IL-5 and IL-1β were elevated.¹³ Additionally, in severe asthmatics, ATG5 protein expression was positively associated with expression of collagen type V alpha I (COL5A1), suggesting a role for autophagy in airway remodeling and fibrosis observed in progressive asthma.³⁰

While the notion that autophagy is pathogenic and may contribute to the asthma phenotype is evidenced in both animal and human studies, there are suggestions that the “pro-asthma” autophagic function represents a cell type-specific response. For example, in certain studies a loss of autophagy-related proteins in bronchial epithelial cells, and hence, a reduction in autophagy, revealed increased airway hyper-responsiveness to anticholinergic exposure and oxidative stress-mediated apoptosis.^31,32 Similarly, a loss of autophagy in Cd11c+immune cells led to an augmentation of neutrophilic airway inflammation observed in the murine asthma model.³³ Furthermore, simvastatin-induced autophagy provided cytoprotection in murine bronchial smooth muscle cells by suppression of IL-4, IL-5, and IL-13, resulting in decreased hypertrophy and fibrosis when compared to controls.³⁴

Several efficacious asthma therapies (e.g., monoclonal anti-IL5 antibody, anti-nerve growth antibody, astragalin) have been shown to downregulate autophagy in vivo.^35–37 Moreover, direct inhibition of autophagy through intranasal administration of chloroquine was also found to thwart airway inflammation, hyper-responsiveness, and remodeling.³⁸ Additionally, ketamine, an advanced therapy for refractory asthma, was found to block cellular autophagic activity in diseased airways, providing a potential explanation for how this agent may assist in the treatment patients with symptoms of severe refractory asthma.^39,40 Taken together, the therapeutic targeting of autophagy may represent a feasible treatment in asthma, but it should be approached with caution as autophagy seems to play a protective role in airway inflammation and hyperreactivity.

3.2. COPD and emphysema

COPD is the third leading cause of death in the United States and there is currently no curative treatment.⁴¹ COPD represents a chronic inflammatory disease state, often following long-term exposure to cigarette smoke, where involvement of both the proximal airways as well as the distal lung tissue, including the terminal alveolar subunits, drives the symptoms of chronic bronchitis and emphysema. Airway epithelial cells are primarily targeted by exposure to cigarette smoke and the unchecked chronic inflammation and oxidative stress that ensue drives the pathogenesis of COPD with the involvement of multiple forms of regulated cell death (i.e., apoptosis, necroptosis, and autophagy).⁵ Alveolar macrophages derived from circulating monocytes also participate in the pathogenesis of disease through the release of chemokines that lead to the recruitment of CD8 lymphocytes, monocytes, and neutrophils to the lung.⁴² A complete understanding of the molecular mechanisms underpinning the pathogenesis of COPD are not fully understood, but several lines of investigation have shown that dysregulated autophagy contributes to the pathogenesis of COPD (Table 1).

Lung tissue from COPD patients or from mice chronically exposed to cigarette smoke revealed increase expression of autophagic proteins. Increased autophagy was observed in human lung tissue across all GOLD stages, making this cellular process a marker of both early and late disease states.¹⁵ The initial suggestions of the importance of autophagy in COPD progression came from studies examining upstream regulators of autophagy (Toll-like receptor 4 (TLR4) and Early growth response-1 (EGR-1)), where dampening of the autophagic response limited the inflammation, cellular dysfunction, and apoptosis observed in response to chronic cigarette smoke exposure both in vitro and in vivo.^15,16 Loss of bronchial epithelial-specific Atg5 was also found to reduce airway inflammation in response to a murine model of particulate matter.¹⁷ Gene expression arrays have revealed activators of autophagy (e.g., EGR-1) to be upregulated in COPD tissue and several polymorphisms were identified in EGR-1 and ATG16L1 as significant risk factors for the development of COPD.^21,43 Furthermore, global loss of key autophagic proteins (Beclin-1, Lc3b) blocked cigarette smoke-induced epithelial cell death and the development of airspace enlargement.¹⁵ Likewise, autophagy impaired mice were also resistant to mucociliary clearance disruption in the airways of subchronic cigarette smoke exposure.¹⁸ Interestingly, studies have also shown that autophagy activation drives systemic complications of COPD, namely chronic kidney disease, when loss of autophagic proteins prevented the development of kidney abnormalities in a murine model of COPD.⁴⁴

A link between activation of selective autophagy of mitochondria, namely mitophagy, and other regulated forms of cell death (e.g., apoptosis and necroptosis) as drivers of the COPD phenotype stressed the importance of these processes in normal lung homeostasis and the pathogenesis of COPD^19,45–49 Inhibition of mitophagy through the genetic deletion of PTEN-induced kinase 1 (Pink1−/−) reduced the development of airspace enlargement, mucociliary clearance disruption, and mitochondrial dysfunction commonly observed in epithelial cells exposed to cigarette smoke.¹⁹ Genetic or pharmacological manipulation of autophagy activation in in vitro models linked activation of autophagy with chronic inflammation and mucus production.^17,50,51 More recent studies showed enhanced autophagic degradation of ferritin, a process known as ferritinophagy, in airway epithelial cells with enhanced ferroptosis, an additional form of regulated cell death mediated by phospholipid peroxidation in association with free iron, in the pathogenesis of COPD utilizing both in vitro and in vivo models.⁵²

While there are accumulating data that activation of autophagy and selective autophagy in epithelial cells are key to the pathogenesis of COPD, there are also several lines of evidence that support the notion that it is insufficient autophagy or mitophagy that drives the pathogenesis of COPD^53–57 (Table 2). Initial studies using human bronchial epithelial cells found that cigarette smoke-induced cellular senescence was blocked by the pharmacological activation of autophagy using the mTOR inhibitor, Torin-1.⁵⁶ Additionally, epithelial cells isolated from COPD patients had a blunted observable activation of autophagy in response to cigarette smoke exposure when compared to epithelial cells obtained from non-smokers or non-COPD smokers.⁵⁶

Table 2.

Murine and human studies highlighting the protective role of autophagy in COPD, IPF, and PAH.

Disease	Autophagy protective
Obstructive
Chronic obstructive pulmonary disease (COPD)	Murine studies
	– deletion of Parkin (reduction in mitophagy) revealed worsening airway thickening and airspace enlargement54 – pharmacological activation of autophagy blocked the development of aggresomes in COPD mouse model58,59 – Cigarette smoke-exposed alveolar macrophages also show increased aggresomes and impaired autophagy60,61
	Human data
	– Bronchial epithelial cells isolated from COPD patients have a blunted observable activation of autophagy in response to cigarette smoke exposure in vitro56,57 – Parkin, a key regulator of mitophagy, is downregulated in human COPD tissue53 – Lung homogenates from COPD tissue have increased p62, suggesting a defective autophagy18
Restrictive
Idiopathic pulmonary fibrosis (IPF)	Murine studies
	– Autophagy decreased in airway epithelial cells and fibroblasts in fibrosis murine model62–66 – Induction of autophagy in bleomycin fibrosis model was found following neutralization of pro-inflammatory cytokine Il-17A, the effect was dependent on autophagy induction67 – Loss of Atg4, Pink1, Park2 response to bleomycin induced fibrosis was exaggerated64–66,68
	Human data
	– Autophagy decreased in human tissue (reduction in LC3 and p62 levels67 – Beclin 1 is downregulated in fibroblasts isolated from IPF patients69 – Nintedanib and pirfenidone treatment results in activation of autophagy through mitophagy and Beclin-1 and Atg7 activation70,71
Vascular
Pulmonary arterial hypertension (PAH)	Murine studies
	– Upregulation of autophagy in response to experimental PAH (hypoxia)72,73 – Loss of Beclin-1, Egr-1, Lc3b worsened PAH phenotype in experimental models72,73 – Rapamycin treatment perturbed the development of PAH in Sugen treated rats74
	Human data
	– Increase autophagosomes and autophagic proteins in pulmonary arterial hypertension tissue72,75

Parkin, a key regulator of mitophagy, is downregulated in human COPD tissue.⁵³ Genetic deletion of Parkin (Prkn−/−) in a murine model of COPD revealed worsening airway thickening and airspace enlargement.⁵⁴ In vitro experiments where Parkin was overexpressed in epithelial cells resulted in dampening of mitochondrial reactive oxygen species and cellular senescence observed in response to cigarette smoke extract.⁵⁴ These investigations suggest that increased levels of Parkin enhance mitophagy and perturb the development of COPD both in vitro and in vivo.

Cigarette smoke is known to induce proteostasis that drives an accumulation of ubiquitinated proteins (aggresomes) within epithelial cells and potentiates chronic inflammation.⁵⁸ Furthermore, decreased proteosomal function has been shown to accelerate cigarette smoke-induced airspace enlargement^58,59 in animal models. Lung homogenates from COPD patients show an increase in ubiquitinated proteins and p62/SQSTM1, a key adaptor protein responsible for delivering ubiquitinated substrates to LC3 and autophagosome formation.¹⁸ Defective autophagy was considered the cause of cigarette smoke-induced formation of aggresomes when pharmacological activation of autophagy blocked this phenotype in the COPD mouse model.⁶⁰ In those studies, the investigators surmised that the accumulation of p62 and ubiquitinated proteins observed in severe COPD-emphysema lungs indicated the pathogenic role of insufficient autophagic clearance in COPD.⁶⁰

While the relevance of autophagy to the inflammatory response and impaired mucociliary clearance observed in cigarette smoke exposed airway epithelial cells has been realized, the upregulation of autophagy in alveolar macrophages and neutrophils has also been observed.^20,61 Interestingly, impaired aggresome clearance and p62 accumulation are also key features of cigarette smoke-exposed alveolar macrophages, suggesting that impaired autophagy impacts homeostasis in these cells as well.⁶¹ In those same studies, cigarette smoke exposed alveolar macrophages were found to be defective in xenophagy, a selective autophagy responsible for the clearance of bacteria, suggesting an explanation for the increased infection rates observed in smokers with COPD.⁶¹ Further analysis revealed that alveolar macrophages from lung tissue or bronchoalveolar lavage fluid from patients with COPD are defective in phagocytosis and the ability to regulate mitochondrial reactive oxygen species production. However, how this relates to autophagy remains to be elucidated. Extensive studies in neutrophils are lacking, but initial investigation uncovered that cigarette smoke-exposed neutrophils have increased autophagy associated with a compromised ability to ingest the respiratory pathogen, Staphylococcus aureus.²⁰ Further studies are required to fully understand the contribution of the cigarette smoke-induced autophagy observed in alveolar macrophages and neutrophils to the pathogenesis of COPD.

Accumulating evidence argues for an essential role of autophagy in multiple cell types in the pathogenesis of COPD. It appears that both overactive and insufficient autophagy can drive the inflammation, cell death, and cellular dysfunction observed in COPD. Therapeutic options for the treatment of COPD continue to remain rather limited, and the potential of targeting autophagy as a treatment for COPD warrants further exploration. Pharmacological activation of autophagy has been used in murine models, but the clinical relevance of the observed protection remains unclear since therapy was administered prophylactically during the course of cigarette smoke exposure.^{18,19,60,76,77} Autophagy activation through mTOR inhibition, using rapamycin, has also been tested and the results are inconclusive.⁷⁸While administration of rapamycin to cigarette smoke-exposed mice appeared to reduce alveolar inflammation, the drug promoted cellular inflammation and apoptosis in room air-exposed mice. These findings underscore the complexity of the role of targeting autophagy as a therapeutic modality in COPD.

4. Pulmonary vascular disease

Pulmonary arterial hypertension (PAH) is a severely debilitating and often fatal disease typified by vascular remodeling and rise in pulmonary artery pressure associated with progression to right ventricular failure. Pulmonary hypertension affects more than 100 million people worldwide and the current treatments fail to halt or reverse ongoing vascular remodeling.⁷⁹ The distal arterioles are the targets of the disease and current tenet supports the notion that endothelial cell injury and apoptosis serve as an initial trigger, and vascular cells, namely endothelial cells and smooth muscle cells, progress to develop a pro-proliferative/anti-apoptotic phenotype that leads to unchecked vascular remodeling.

The importance of the initial endothelial injury and apoptotic cell death in the development of PAH is highlighted in several murine studies, where treatment with the vascular endothelial growth factor (VEGF) inhibitor SU5416, a strong inducer of endothelial apoptosis, resulted in PAH with features similar to that observed in human disease.⁸⁰ However, following the initial phase both endothelial and smooth muscle cells are highly proliferative in nature leading to vessel wall thickening and the formation of plexiform lesions. In this case, signaling driven by growth factors namely VEGF, platelet-derived growth factor (PDGF), and fibroblast growth factor (FGF) maintain cell cycle progression in vascular cells.^81,82 Pro-survival signaling pathways (AKT/ERK/NF-kB/STAT3) are upregulated in endothelial and smooth muscle cells that enhance persistent vascular disarray.⁸³ Although the molecular mechanisms involved in the pathogenesis of PAH have been extensively studied, new therapeutic targets that can halt and reverse progression remain lacking.

In recent years, several studies have questioned the role of autophagy in driving the pathogenesis of PAH and the conclusions remain controversial (Tables 1 and 2). Human PAH lung tissue samples show increase in autophagic markers, where LC3B and autophagosome numbers are elevated when compared to control lungs.⁷² Both in vivo and in vitro studies showed up-regulation of the same autophagic markers in vascular cells in response to the hypoxia-induced experimental PAH models.^72,75 Additionally, genetic deletion of Lc3b or Egr-1, a known regulator of LC3B, led to a rise in right ventricular systolic pressure and loss of LC3B in the vascular cells led to increased reactive oxygen species production and cellular proliferation in vitro.⁷² Additional studies revealed a loss of Beclin-1 enhanced PAH phenotype with increased angiogenesis and oxidative stress in pulmonary artery endothelial cells from fetal lambs with persistent pulmonary hypertension.⁷³ In the PAH rat model, using a VEGF inhibitor (SU5416), investigators highlighted the need for increased autophagy in endothelial cells as a mechanism to halt progression of pulmonary hypertension when treatment with rapamycin, an indirect activator of autophagy, perturbed the development of PAH in SU5416 treated rats.⁷⁴ In a separate study, investigators tested the ability of dihydroartemisinin (DHA), a compound with known anti-inflammatory and anti-tumor activities, to prevent the development of both hypoxia and monocrotaline-induced pulmonary hypertension.^84,85 In vitro studies clarified that this protection appears to be linked to the activation of endothelial cell autophagy in response to DHA.⁸⁶

While the data supporting a protective role of autophagy in the pathogenesis of PAH is robust, there are several studies that identify cases where autophagy may be pathological. In the case of HIV-related PAH, severe and exaggerated pulmonary vascular disease in patients with concomitant opioid usage was observed.⁸⁷ In vitro studies linked the presence of HIV virus and opioid dosing with an induction of key autophagic markers and autophagosome production in human arterial endothelial cells. Investigators linked the aggressive nature of this HIV-related PAH to the observed impact on cellular autophagy.⁸⁷ In recent years, roles for mitochondrial and metabolic dysfunction in the pathogenesis of PAH has been established.⁸⁸ The abnormalities identified in PAH include, but are not limited to, alterations in the pentose phosphate pathway, glutaminolysis, fatty acid oxidation.⁸⁸ A link between excessive mitophagy, a selective form of autophagy, mitochondrial dysfunction, and a pathogenic endothelial cell phenotype in response to hypoxia was observed in an in vitro experimental model of pulmonary hypertension.⁸⁹ Additionally, loss of the key mitochondrial protein, uncoupling protein 2, revealed increased mitophagy and worsening hypoxia-induced pulmonary hypertension.²² Given the relevance of metabolic derangements to the pathogenesis of PAH, metabolic therapies are being tested in pre-clinical models of pulmonary hypertension. Liraglutide, a glucagon-like peptide-1 receptor (GLP-1), is widely used to treat diabetes, and was found to prevent and reverse monocrotaline-induced PAH.²³ Follow up studies found that the observed effect appears to be related to decrease in mitochondrial fusion involving dynamin-related protein-1 (Drp1) and a reduction in cellular autophagic activity in the smooth muscle cells.⁹⁰ Notably, recent studies have also linked autophagy activation to the progression of right ventricular remodeling and fibrosis.²⁴

Further studies are needed to clarify the conditions where autophagy protects against or augments the pathogenesis of PAH. Murine models utilizing cell-specific genetic deletion strategies will likely assist in clarifying these conditions in experimental models of PAH. Pre-clinical animal studies examining the impact of pharmacological activation of autophagy on the severity of the observed PAH phenotype have been inconclusive.^91,92 Clinically, everolimus, an mTOR inhibitor and activator of autophagy, has been used in cases of severe PAH secondary to chronic thromboembolic disease and there was measurable improvement in exercise tolerance and a decrease in pulmonary vascular resistance.⁹³ Currently, there are Phase I clinical trials underway testing the utility of mTOR inhibitors, thus, autophagy activators, in the treatment of patients with severe PAH.

5. Restrictive lung disease

Idiopathic pulmonary fibrosis (IPF) is a chronic, progressive, and fatal fibroproliferative disorder characterized by loss of alveolar structure, accumulation of myofibroblasts, and extensive remodeling of lung parenchyma. It is the most common and severe form of restrictive lung disease with over three million people worldwide who are afflicted with IPF with limited available effective therapies. Current dogma supports the idea that endogenous or exogenous injury to the alveolar epithelium drives cell-specific apoptosis and activation of fibroblasts/myofibroblasts that leads to fibroblast deposition and extensive lung remodeling. TGF-ß1 is a pro-fibrotic cytokine that is increased in IPF lung tissues and has been shown to enhance production of type I collagen from myofibroblasts, as well as, induce apoptosis of epithelial cells in culture. In murine models of fibrosis, overexpression of TGF-ß1 causes alveolar epithelial apoptosis and fibrosis.

In recent years, several investigations have focused on the role of autophagy in the pathogenesis of IPF (Table 2). These data support the current paradigm that autophagy represents a protective mechanism in normal wound healing and limiting fibroblast activation and collagen deposition, while promoting autophagy associated cell death with resolution of the injury. In contrast, in the case of IPF, suppressed or defective autophagy has been shown to facilitate epithelial cell dysfunction and myofibroblast activation and production.

Autophagic pathways are decreased in lung epithelial cells and fibroblasts from IPF patients and in murine models of IPF.^62–65,67 Human tissues of patients with IPF demonstrated evidence of decreased autophagic activity via reduction in LC3 and p62 expression.⁶⁷ Initial murine studies using bleomycin model of lung fibrosis found that neutralizing the pro-inflammatory cytokine, IL-17A, led to induction of autophagy and resolution of acute inflammation, attenuation of pulmonary fibrosis, and increased cellular survival.⁹⁴ When autophagy was blocked by the administration of 3-methyladenine the protective effects of IL-17A blockade did not persist.⁹⁴ In addition to IL-17-mediated inhibition of autophagy, bioactive lipids such as sphingosine 1-phosphate and lysophosphatidic acid are also known to inhibit autophagy and to promote fibrogenesis.⁹⁵ Beclin-1, a key autophagic protein, is downregulated in fibroblasts isolated from patients with IPF.⁶⁹ Similarly, human fibroblasts from IPF donors displayed reduction in autophagic proteins and the unfolded protein response in parallel with increased collagen production in vitro.⁹⁶ Additionally, when autophagy is inhibited in epithelial cells the epithelial-to-mesenchymal transition is enhanced and fibrosis ensues.⁹⁷

In multiple in vivo studies, where genetic deletions led to autophagy deficient (Atg4−/−) or mitophagy deficient (Pink1−/−, Prkn−/−) mice, the fibrotic phenotype of the lung in response to bleomycin was exaggerated.^64–66,68 Interestingly, mice deficient in matrix metalloproteinase 19 (Mmp19−/−) have exaggerated fibrotic response to bleomycin which was linked to autophagic suppression via downregulation of ATG4C, a cysteine peptidase necessary for autophagosomal production.⁹⁸ Furthermore, genome-wide association studies identified the autophagy gene Cep55 as having a polymorphism significantly associated with susceptibility to bleomycin-induced lung fibrosis in mice.⁹⁹ Other studies focused on epithelial cells have shown that Atg4b-deficient mice have a greater inflammatory response 7 days post bleomycin and augmented apoptosis of alveolar and bronchial epithelial cells leading to more extensive fibrosis and collagen accumulation.⁶⁸ More recently, using 3D lung organoids, investigators showed that inhibition of vimentin intermediate filament assembly promoted autophagic clearance of type I collagen.¹⁰⁰ Aging studies revealed that mice with loss of autophagic proteins (Lc3b−/− and Atg4b−/−) were more susceptible to bleomycin-induced lung fibrosis and linked cathepsin A, a binding partner to LC3B, and ER stress to increased apoptosis of epithelial cells in culture.^101,102

The role of mTOR activation in dampening the autophagic response in IPF is supported by evidence that fibroblasts harvested from IPF patients were found to have aberrant PTEN/AKT/mTOR pathway that permits unchecked fibroblast proliferation through the suppression of autophagy.⁶² Accordingly, inhibition of autophagy in vitro led to increase in the development of myofibroblast phenotype in vitro, whereas, autophagy activation through mTOR inhibition with rapamycin reversed these effects.⁶² Similarly, conditional knockdown of the tuberous sclerosis-1 (Tsc1) gene in epithelial cells caused mice to be more susceptible to bleomycin-induced lung fibrosis, a condition that was reversible when rapamycin or chloroquine was administered to stimulate autophagy.¹⁰³

Targeting autophagy as a therapy for patients with IPF may have great potential and pre-clinical and clinical studies are underway. Immunomodulators including IL-17A neutralizing antibodies, mir449A, and PDGF4B inhibitors are known activators of autophagy and show promise in pre-clinical models.^65,94,104 Additionally, rapamycin, through the activation of autophagy, appears to blunt the development of pulmonary fibrosis in murine models.^66,103 Berberine, an isoquinolone alkaloid, known to prevent activation of TGF-ß1 signaling, has been shown to ameliorate IPF in the bleomycin model.¹⁰⁵ Further studies revealed that berberine leads to inhibition of mTOR and stimulation of autophagy. Interestingly both nintedanib and pirfenidone, two recently FDA approved therapies offered to patients with IPF, show anti-fibrotic activity through the induction of both non-selective autophagy through Beclin-1 and ATG7 and mitophagy, respectively.^70,71 Further studies are required to evaluate the benefits of targeting autophagy in IPF and whether or not these targeting strategies should be cell type-specific.

6. Concluding remarks

The cellular components of the lung are challenged by oxidative and environmental stressors that require appropriate cellular responses to maintain homeostasis. Whether autophagy functions as a protector or promoter of a given lung disease is a complex question that needs further investigations to better understand the physiologic functions of autophagy in lung disease. The multiple cell types, and spatial and temporal heterogeneity of lung tissue injury mediating the development of lung diseases such as asthma, COPD, PAH, and IPF make it even more challenging to interpret confidently the results from studies using global knockout mouse models. Further clarity will likely evolve with future studies utilizing cell-specific autophagy gene targeted mice in a given disease model. Additionally, further defining phenotypes of the heterogenous disease states of asthma and COPD will likely uncover new roles of autophagy and selective autophagy in the pathogenesis of these diseases. While pre-clinical models have advanced our understanding of the functional roles of autophagic pathways in multiple chronic lung diseases, further investigations in pre-clinical and, importantly, clinical studies are needed to translate the findings to potential therapeutic targets.