Autophagy is a process of lysosomal self-degradation that helps maintain homeostatic balance between the synthesis, degradation, and recycling of cellular proteins and organelles [23]. At the cellular level, autodigestion takes place within lysosomes and proteasomes. Proteasomes are involved in the clearance of ubiquitin-conjugated soluble proteins, whereas autophagy delivers diverse cytoplasmic components to the lysosome, including soluble proteins, aggregate-prone proteins, and organelles [24]. Autophagy is not simply machinery for amino acid supply in response to energy demand; it is an adaptive pathway of cytoprotection from cellular stresses, involving starvation, reactive oxygen species (ROS), endoplasmic reticulum (ER) stress, and microbe infection [24].

Three forms of distinct autophagy have been demonstrated: chaperon (Hsc70)-mediated autophagy (CMA), microautophagy, and macroautophagy. CMA degrades soluble proteins only, via direct translocation to the lysosome through Lamp2A, a lysosomal transmembrane protein. During microautophagy, small components of the cytoplasm are engulfed by direct invagination into lysosomes. Engulfment of cytoplasmic components by the isolation membrane (phagophore) is the initial step in autophagy and is followed by elongation and fusion, resulting in formation of double-membranous vesicles (autophagosome). Subsequent fusion of the autophagosome with the lysosome to form the autolysosome is absolutely required for proper degradation [25]. Recent advances in the molecular mechanisms of autophagy have mainly focused on macroautophagy, specifically on the detection of a series of autophagy-related (Atg) genes. Among 35 autophagy-related, (Atg), proteins identified in yeast, there are core Atg proteins required for autophagosome formation, which are well conserved in mammals. Hence, in general in the literature, macroautophagy is designated as autophagy [25].

Initially, autophagy was proposed to be a nonselective bulk degradation system, but recent advances demonstrate that a variety of ubiquitinated cargos, including protein aggregates, mitochondria, and microbes, are selective targets for autophagic degradation [25]. Accordingly, ubiquitination is an important tag for not only proteasomal degradation but also for selective autophagy. The p62 protein / sequestosome 1 (SQSTM1) has been shown to be an adaptor protein for selective autophagy based on its ability to bind both ubiquitin and microtubule-associated protein 1A/1B-light chain 3 (LC3), a crucial component for autophagosome formation [26]. Because of the dynamic nature of autophagy, in which autophagosome can be formed within several minutes, it is difficult to distinguish between increased autophagy flux and impaired subsequent clearance when using electron microscopic detection of autophagosomes or when examining Atg expression levels. Therefore, to detect the conversion of LC3-I to LC3-II (which is conjugated to phosphatidylethanolamine (PE) to ensure stable association to the autophagosomal membrane), the use of protease inhibitors is generally accepted to be standard methodology for evaluation of autophagy flux. In addition, based on the findings of selective autophagic degradation, concomitant accumulation of p62 and ubiquitinated protein is also recognized to at least partly reflect autophagy activity.

Due to the large number of physiological and aberrant intracellular components that are potential targets for autophagic degradation, autophagy status is linked to a diverse array of cellular processes and cell fates, including energy supply, homeostatic turnover of organelles, cell fate, cellular senescence, and immune responses [23]. In terms of the pathogenic role of autophagy, excessive activation may be associated with disease progression in extraphysiologic condition, whereas impairment of autophagy activity has been widely implicated in the pathogenic sequence of a variety of human disorders [26]. Indeed, recent in vitro and in vivo gene knockout studies revealed that insufficient autophagy is involved in the development of lung diseases [27–30].

Continuous ventilation of large amounts of air with high oxygen concentration, which may contain noxious particles and harmful microbes, is a fundamental function of the lungs and is required for sufficient gas exchange. Subsequently, indicating lung cells are serially exposed to a diverse array of cellular stresses, and it is reasonable to speculate that autophagy-mediated alleviation of cellular stress plays a key regulatory role in lung pathophysiology.

Apoptosis is necessary for the maintenance of homeostatic plasticity in the lung. However, the cell-type-specific imbalance of positive and negative regulation of apoptosis has been proposed to be a critical determination of lung disease progression [31]. Although autophagy had been postulated to be responsible for type II programmed cell death, current understanding is mainly that autophagic cell death is attributable simply to overwhelming autophagosome formation as a part of stress responses, in which cytosol and organelles are destroyed to an unrecoverable degree. Autophagy is an adaptation pathway for cellular stress, including starvation, ROS, endoplasmic reticulum (ER) stress, and microbe infection; hence autophagy is generally considered to be a mechanism for cell survival. However, Atg-gene-dependent cell death has been reported in the setting of dysfunction of apoptosis machinery. Indeed, double knockout of Bax/Bak, essential components of the mitochondrial apoptotic pathway, demonstrated a distinct type of cell death, marked by accumulation of autophagosomes [32]. Atg5 knockdown ameliorated this cell death, indicating that autophagy may promote cell death in the setting of extraphysiologic apoptosis deficiency.

In addition, there is functional cross talk between apoptosis and autophagy. For instance, B-cell lymphoma-2 (Bcl-2), an anti-apoptotic protein, interferes with starvation-dependent autophagy by binding to Beclin1 (Atg6) [33], which is an important constituent of the class III phosphatidylinositol (Ptdlns) 3-kinase complex involved in nucleation and assembly of the initial phagophore membrane. During receptor-mediated apoptosis, cleavage of Atg3 protein by caspase-8 suppresses autophagy activation, resulting in increased cell death. Furthermore, irrespective of their known role in autophagy induction, several Atg proteins may exert diverse regulatory roles during both the apoptotic process and cell survival. Fragmented Atg5 cleaved by calpain promotes intrinsic mitochondrial apoptosis [34]. Atg5 on the autophagosomal membrane serves as a platform for an intracellular death-inducing signaling complex (iDISC) that recruits caspase-8 to initiate the apoptosis cascade [35]. LC3B has been demonstrated to regulate apoptosis through interaction with Fas and caveolin-1(Cav-1). Therefore, clarification of whether the autophagy process is involved in apoptosis regulation is a prerequisite for proper interpretation of experimental results in the setting of Atg manipulation.

Cellular senescence has been widely implicated in disease pathogenesis in terms of not only impaired cell repopulation but also aberrant cytokine secretions of SASP. SASP may exert deleterious effects on the tissue microenvironment of neighboring cell [36]. Increased cellular senescence is one of major features of aging, and hence cellular senescence has been widely implicated in age-associated disorders. The detailed molecular mechanism for regulation of cellular senescence is complex and incompletely understood, but one of the typical manifestations is accumulation of damaged proteins and organelles, occasionally associated with ubiquitinated aggregations [37]. Therefore, it has been proposed that functional insufficiencies in the cellular cleaning and housekeeping mechanisms of autophagy play a pivotal role in the accumulation of deleterious cellular components and therefore in the regulation of cellular senescence [37].

Indeed, autophagy diminishes with aging, and accelerated aging can be attributed to reduced autophagy. Thus, autophagy activation appears to be associated with longevity [38]. Pathologic premature aging due to autophagy malfunction has been intensively examined using animal models of autophagy inhibition by tissue-specific knockout of Atg genes. Those animal models with insufficient autophagy demonstrated a cellular phenotype of progressive accumulation of ubiquitinated aggregates and disorganized mitochondria, suggesting the causal relationship between loss of autophagy and aging-associated disease phenotypes [38]. However, those phenotypic alterations were mainly evaluated in the central nervous system and liver, not in other organs. Among the variety of targets for autophagic degradation, selective autophagy of mitochondria (mitophagy) has been widely implicated in cellular senescence in terms of regulation of reactive oxygen species (ROS) of oxidative stress. Mitochondria are the main organelle responsible for intrinsic ROS release through respiratory chain reactions, and insufficient mitophagy results in accumulation of damaged mitochondria accompanied by increased ROS production [39].

The role of stress-induced autophagy activation in longevity has been mainly demonstrated in the case of caloric restriction (CR) [40]. CR induces autophagy through the inhibition of mammalian target of rapamycin (mTOR), an essential negative regulator of autophagy, and also through activation of adenosine monophosphate-activated protein kinase (AMPK) and sirtuin1 (SIRT1). In response to the rising AMP/ATP ratio during CR, AMPK induces autophagy via phosphorylation of ULK1, a mammalian orthologue of the yeast protein kinase Atg1. SIRT1 deacetylation of Atg proteins and transcription factors, including the FOXO family, is involved in autophagy induction [22]. The involvement of CR-induced autophagy in longevity was confirmed by inhibition of autophagy, and SIRT1-mediated longevity by CR is at least partly conferred by autophagy activation. Intriguingly, recent paper demonstrated that SIRT1 protects against emphysema by a FOXO3-mediated reduction of premature senescence in mice, but the involvement of autophagy was not examined.

Autophagy is getting more importance during the aging process, because proteasome pathway could not degrade protein aggregates in the presence of an enhanced prooxidant and aggregation-prone milieu characteristic of aging [41]. Pharmacological inhibition of autophagy prevented development of premature senescence but did lead to the enhanced rate of apoptosis in human umbilical vein endothelial cells. A subset of autophagy-related genes is upregulated during senescence [42]. Overexpression of one of those genes, ULK3, induces autophagy and senescence. Furthermore, inhibition of autophagy delays the senescence phenotype, including senescence-associated secretion [43]. Goehe et al. showed that autophagy and senescence tend to occur in parallel and furthermore that autophagy accelerates the development of the senescent phenotype [44]. Collectively, autophagy and cellular senescence are associated with each other in some situations, but these two important cellular processes may be interdependently involved in the pathophysiology of several diseases.

Chronic obstructive pulmonary disease (COPD) is one of the leading causes of death worldwide and is characterized by partially irreversible and progressive airflow limitation. Cigarette smoke, the major cause of COPD, is rich in toxic components including ROS, and a variety of biological responses to cigarette smoke exposure have been demonstrated. Although detailed molecular mechanisms for COPD development remain unclear, the possible participation of autophagy in the pathogenic sequence of COPD has been intensively explored (Fig. 6.2). It has been reported that autophagy in lung tissue from COPD patients is augmented by means of an increase in the LC3B-II/LC3B-I ratio, and Egr-1-induced LC3B expression is essential for autophagy activation [45]. LC3B−/− mouse experiments confirmed the pivotal role of LC3B in epithelial cell apoptosis induction by cigarette smoke exposure. The proposed mechanism of LC3B-induced apoptosis is attributed to the balance in a trimolecular interaction between LC3B with Fas and caveolin-1(Cav-1), a lipid raft protein. LC3B knockdown inhibits apoptosis by increasing Cav-1-dependent Fas sequestration, and dissociation of Fas and LC3B from Cav-1 in response to CSE exposure initiates apoptosis in epithelial cells. LC3B is a key component for autophagy machinery, and association between LC3B and Fas is an interesting observation; however it is still unclear whether autophagy activation by LC3B expression is crucial for apoptosis induction in these COPD models. Furthermore, in cases of hyperoxia-induced apoptosis in epithelial cells, LC3B interacts with Fas, resulting in prevention of apoptosis [46], suggesting that the role of association between LC3B and Fas in apoptosis regulation is dependent on the stimuli or experimental conditions. Intriguingly, decreased autophagy activity in alveolar macrophages derived from smokers has been reported in terms of impaired xenophagy. In spite of increased LC3B-II and autophagosomes in macrophages from smokers, impairment of autophagy flux was shown using protease inhibitors and also by detecting accumulation of p62 aggregates [47], indicating that autophagy activity in COPD lung may be regulated via cell-type-specific mechanisms.

Autophagy plays a pivotal regulatory role for cellular senescence; hence we have attempted to elucidate the involvement of autophagy in the regulation of cigarette smoke extract (CSE)-induced human bronchial epithelial cell (HBEC) senescence [27]. CSE transiently induces autophagy activation followed by accumulations of p62 and ubiquitinated proteins accompanied by an increase in HBEC senescence. Autophagy inhibition by 3MA, a specific inhibitor of autophagic sequestration, or by LC3B and Atg5 knockdown further enhanced HBEC senescence with concomitant accumulation of p62 and ubiquitinated proteins [27]. In contrast, autophagy activation by Torin1, a mammalian target of rapamycin (mTOR) inhibitor, suppressed p62 and ubiquitinated protein accumulations and also inhibited HBEC senescence. In line with the previous finding of increased autophagy activation in COPD epithelial cells, we observed an increase in baseline autophagy but also found significantly decreased autophagy induction in response to CSE exposure in HBEC isolated from COPD patients compared to those from nonsmokers [27]. We speculated that the mechanism for enhanced baseline autophagy flux was attributed to increased oxidative stress, which was demonstrated by the accumulation of carbonylated proteins in HBEC from COPD patients [18]. Therefore, it is probable that the attenuation of autophagy flux in response to CSE exposure may reflect an insufficient reserve of autophagy activation in HBEC from COPD patients. Concomitant accumulation of p62 and ubiquitinated protein is also recognized to at least partly reflect autophagy activity. Increased accumulations of p62 and ubiquitinated proteins detected in lung homogenates support the notion that insufficient autophagic clearance is involved in accelerated cellular senescence in COPD [27] (Fig. 6.3).

SIRT6 has been demonstrated to regulate longevity by modulating insulin-like growth factor (IGF)-I signaling. IGF-I signaling activates mTOR, and a recent paper demonstrated that IGF-I exposure was sufficient to induce cellular senescence through inhibition of baseline autophagy [48]. Intriguingly, we demonstrated that CSE decreased the SIRT6 expression in HBEC and that CSE-induced HBEC senescence was inhibited by SIRT6 overexpression and that histone deacetylase (HDAC) activity of SIRT6 was indispensable for inhibition of CSE-induced HBEC senescence through autophagy activation, which was mainly attributed to attenuation of IGF-Akt-mTOR signaling [49]. Decreased expression levels of SIRT6 found in lung homogenates from COPD patients support the hypothesis that reduced SIRT6 expression with accompanying autophagy insufficiency may be associated with COPD development through the enhancement of cellular senescence, especially in the setting of increased IGF signaling. Furthermore, Decreased SIRT6 expression was significantly correlated with the decrease of FEV₁% [49]. As IGF-I shares receptors and signaling pathways with insulin, and type II diabetes mellitus with hyperinsulinemia is a common comorbidity in COPD, it may be associated with COPD development via increased IGF/insulin signaling and autophagy inhibition, especially in cases of decreased SIRT6 expression.

Mitochondria are the main organelle producing ATP as well as reactive oxygen species (ROS) and play central roles in cell fate regulation. Mitochondria also release mitochondrial DNA as one of damage-associated molecular pattern. Therefore, maintenance of mitochondrial homeostasis is prerequisite for cellular homeostasis [50]. Mitochondria are dynamic organelles, which continuously change their shape through fission and fusion. Damaged and fragmented mitochondria are removed through mitochondria-specific autophagic degradation (mitophagy). Disruption of mitochondrial dynamics is involved in disease pathology through excessive reactive oxygen species (ROS) production [51] (Fig. 6.4). In electron microscopic examination of lung tissues, we demonstrated that mitochondria in bronchial epithelial cells tended to be fragmented in COPD, suggesting the fission process dominancy of mitochondrial dynamics in COPD pathogenesis [52]. In vitro studies further confirmed that CSE-induced excessive fragmentation of mitochondria is associated with mitochondrial ROS production, resulting in HBEC senescence. Autophagy inducer Torin1 accelerate degradation of damaged mitochondria in autophagosome, resulted in the increase of healthy mitochondria [52].