Autophagy and Tumorigenesis

Abstract

Autophagy is a catabolic process that captures cellular waste and degrades them in the lysosome. The main function of autophagy is quality control of cytosolic proteins and organelles, and intracellular recycling of nutrients in order to maintain cellular homeostasis. Autophagy is upregulated in many cancers to promote cell survival, proliferation and metastasis. Both cell-autonomous autophagy (also known as tumor autophagy) and non-cell autonomous autophagy (also known as host autophagy) supports tumorigenesis through different mechanisms, including inhibition of p53 activation, sustaining redox homeostasis, maintenance of essential amino acids levels in order to support energy production and biosynthesis, and inhibition of anti-tumor immune responses. Therefore, autophagy may serve as a tumor-specific vulnerability and targeting autophagy could be a novel strategy in cancer treatment.

1. Introduction

Macroautophagy (herein referred to as autophagy) is a major catabolic pathway for the delivery of dysfunctional or unnecessary cellular components to the lysosome for degradation and subsequent reuse, a process which is conserved from yeast to mammals [1]. Autophagy is essential for cellular homeostasis, and abnormal autophagy is associated with many human diseases [2]. We will focus on the role of autophagy in cancer, and in particular, we will elucidate how autophagy regulates the immune response and metabolism to support tumor growth and survival.

2. Autophagy Regulation and Machinery

Autophagy is induced by environmental stressors such as nutrient deprivation. It is negatively regulated by the Mammalian Target of Rapamycin (mTOR), a nutrient sensor for the cell, and positively regulated by the AMP-activated protein kinase (AMPK), a master regulator of energy metabolism (Fig. 1) [3].

Figure 1. Autophagy regulatory pathways.

The PI3K/AKT/mTOR pathway, activated by nutrients and growth factors, leads to downregulation of autophagy. On the other hand, the LKB1/AMPK/ULK pathway leads to upregulation of autophagy due to metabolic stressors.

2.1. The PI3K/AKT/mTOR Pathway

Phosphoinositide 3-kinases (PI3Ks) are a class of enzymes that generate secondary messengers known as 3′ phosphoinositide lipids [4]. PI3K enzymes are usually activated by metabotropic G-protein coupled receptors in response to intracellular growth factors, such as insulin-like growth factor-1 (IGF1) or interleukin 3 (IL-3) [5]. These 3’ phosphoinositide lipids are then able to activate Protein kinase B (PKB), also known as Akt, which phosphorylates downstream targets and activates mTOR pathways. More specifically, in nutrient rich conditions or stimulation by growth factors, Akt activates mTOR Complex 1 (mTORC1), suppressing autophagy by phosphorylation-dependent inhibition of Unc-51 like-autophagy activating kinase 1/2 (ULK1/2) and the Vacuolar protein sorting 34 (Vps34) complex [6].

2.2. The AMPK/ULK Pathway

AMPK is activated during energy crisis, such as during amino acid depletion or glucose deprivation, which leads to a large AMP:ATP ratio within the cell. More specifically, AMPK is allosterically activated by AMP and ADP, while competitively inhibited by ATP. AMPK, activated during low-charged cellular energy states, can then phosphorylate ULK1/2, two crucial mammalian enzymes responsible for stimulating autophagy. Activated AMPK can also coincidentally inhibit mTOR through stimulation of the tuberous sclerosis complex 1/2 (TSC1/2). Hence, AMPK possesses a unique dual ability to robustly activate autophagy within the cell in response to metabolic stress [3].

2.3. Autophagy Machinery

The process of autophagy consists of multiple distinct steps and is mediated by several autophagy-related (Atg) genes (Fig. 2). The preliminary step in the formation of autophagy vesicles is the generation of the ULK complex. The ULK complex consists of a ULK family kinase, the focal adhesion kinase interacting protein of 200 kDa (FIP200), and autophagy-related gene 13 (Atg13) [7]. This complex is doubly regulated by two important sensors of cellular stress, mTOR and AMPK. Thus, the ULK complex acts as the main “gatekeeping” protein that regulates whether or not the autophagic vesicles form.

Figure 2. Autophagy machinery.

Step 1, the phagophore is synthesized from ER membrane lipids upon activation by the Vps34 Complex. Step 2, the phagophore is elongated with the help of structural proteins such as LC3 and GABARAP. These Atg8 family of proteins are attached to the phagophore with the help of ATG complexes. Step 3, p62 and NBR1 serve as cargo receptor proteins that carry polyubiquitinated proteins destined for degradation to the phagophore. Once the phagophore closes itself around the protein, it is known as an autophagosome. Step 4, a neighboring lysosome then fuses with the autophagosome to create and autolysosome. Step 5, the polyubiquitinated protein is degraded using lysosomal enzymes.

The first and second step of autophagy activation is broadly known as the “elongation step” and involves the preparation of autophagic vesicle membranes. The ULK complex first activates the Vps34 complex, which is composed of Vps34 (a class III PI3K) and Beclin-1 (BECN1) [7]. Using the endoplasmic reticulum as a base, the Vps34 complex is able to form the autophagic vesicle membrane lipids. These vesicle membranes, known as a “phagophore”, generate a spherical vesicle. These lipids are then conjugated to microtubule-associated protein 1A/1B light chain 3B (LC3), a microtubule-associated protein that serves a structural purpose in vesicle formation [7]. Additionally, LC3 serves a functional role as a docking site for cargo receptor proteins carrying polyubiquitinated organelles or proteins.

There are two ubiquitin-like systems used during phagophore elongation and maturation. The first is Atg5-Atg12 conjugation, which is mediated by E1-like activating enzyme Atg7. The Atg5-Atg12 forms a complex with Atg16 to promote LC3 cleavage by ATG4B as well as LC3-phosphatidylethanolamine (LC3-PE) conjugation (the second ubiquitin-like system), which is also mediated by Atg7 [8]. It is believed that curvature of the phagophore membrane is driven by the binding of damaged cargo to receptors lined along the lipid membrane. Eventually, the phagophore completely wraps around the cargo to form a double-membrane “autophagosome” [9].

Cargo receptor proteins play important roles during autophagy. p62, also known as sequestosome1, is a cargo receptor protein that recognizes polyubiquitinated proteins and binds them with autophagosomes. Since most organelles targeted for autophagy are tagged with ubiquitin, p62 plays a crucial role in facilitating degradation of these organelles within autophagosomes [10]. Neighbor of BRCA1 gene 1 protein (NBR1) is another cargo receptor protein that binds polyubiquitinated proteins to autophagosomes. Both p62 and NBR1 contain an LIR motif that binds to the Atg8 family of proteins (such as GABARAP and LC3) found on autophagosomes [11]. Although p62 and NBR1 act similarly, having different cargo receptor proteins provides a high degree of selectivity in the autophagic process. In other words, having different cargo receptor proteins enables damaged organelles or protein aggregates to preferentially bind to their specific cargo receptor protein.

Following these steps, organelle degradation occurs when the autophagosomes fuse with lysosomes to form the “autolysosome”. The autolysosome contains the lysosomal enzymes and acidic lumen necessary for protein degradation [9]. The acidic inner environment that is generated by this process causes destruction of the inner membrane of the autolysosome but not the outer membrane. This allows the autolysosome to still retain its shape and function [9]. Once degradation is complete, the broken-down contents are released and recycled for subsequent use by the cell [7].

3. Autophagy Function

Autophagy is mainly responsible for the breakdown and recycling of long-lived cell components and cytosolic organelles and is present in a wide range of organisms and cell types. Its functions are broad but mainly include organelle and protein turnover, tissue remodeling, and survival during starvation (Fig. 3) [12, 13].

Figure 3. Common functions of autophagy.

Autophagy degrades damaged or dysfunctional cytosolic organelles and proteins, serving as a quality control and waste disposal mechanism for the cell. Autophagy-mediated degradation also serves as a way for the cell to recycle nutrients and other substrates for use in biosynthesis and energy production - especially during times of stress or starvation.

3.1. Protein and Organelle Quality Control

One of the major functions of autophagy is to remove and degrade damaged cytosolic components such as protein aggregates and organelles, which enables the cell to maintain quality control of cellular contents [14, 15]. Elimination of toxic protein aggregates or damaged organelles by autophagy can prevent reactive oxygen species (ROS) production, making autophagy essential for sustaining genomic stability [16]. Mitophagy is another important type of cargo-specific autophagy in which damaged and dysfunctional mitochondria are selectively captured in autophagosomes for degradation to ensure mitochondrial quality and prevent buildup of defective mitochondria [17, 18]. Both non-selective and selective mechanisms of autophagy are responsible for the turnover and quality control of several other organelles as well, including endoplasmic reticulum, ribosomes, centrosomes, and lipid droplets [19–23].

3.2. Intracellular Nutrient Recycling

Autophagy plays an essential role in recycling cellular components to maintain metabolism, especially during metabolic stress such as starvation. Autophagy can breakdown cytosolic components into amino acids, nucleic acids, sugars, and fatty acids, which can then be recycled into carbon metabolism for energy homeostasis and biosynthesis [15, 24]. Autophagy is also responsible for maintaining amino acid levels during the neonatal starvation period in mice [25, 26], and is required to maintain glucose homeostasis for adult mice to survive fasting conditions [27]. Autophagy also promotes survival in cells undergoing high levels of metabolic stress, such as rapidly proliferating tumor cells [28–30]. Thus, autophagy provides metabolic substrates required for the growth and survival of both healthy cells and cancer cells [29, 31–33].

3.3. Non-Canonical Roles of Autophagy Proteins

The proteins involved in autophagy pathways also play non-canonical roles in other cellular contexts. Beginning with LC3, there are two different LC3 family members that play a role in cancer cells. LC3C plays a tumor suppressive role through regulation of Met/hepatocyte growth factor receptor tyrosine kinases (Met/HGF RTKs). Met/HGF RTKs are implicated in the increased migratory invasive responses of cancer cells [34]. However, complex formation with LC3C selectively degrades these Met/HGF RTKs through autophagic pathways. Therefore, LC3C plays an important role in reducing tumor metastasis. In fact, decreased LC3C expression in cancer cells causes a loss of autophagic degradation of Met/HGF RTKs, resulting in increased cancer cell invasion and metastatic progression [34, 35]. On the other hand, LC3B is believed to promote tumor growth by initializing a survival role for cancer cells once they reach metastatic competence. Tumor microarrays of various cancer types, such as beast carcinomas and melanomas, have shown to stain positively for LC3B in moderate to high levels [36]. An explanation for this paradoxical role of LC3 is that LC3C reduces metastasis through selective degradation of Met/HGF RTKs, while LC3B promotes metastasis via eliciting autophagy in response to environmental stressors [36].

ATG proteins also play non-canonical functions in pathogen replication and immune response. For example, the Atg5-Atg12 conjugate is crucially involved in phagophore elongation and maturation. However, studies show that the Atg5-Atg12 conjugate can also suppress innate immune responses by downregulating type I IFN production. This is accomplished since the Atg5-Atg12 conjugate can inhibit the function of cytoplasmic RNA helicases responsible for neutralizing virus-derived immunostimulatory RNA structures (isRNA), such as double-stranded RNA or 5’-triphosphorylated RNA [37]. In fact, mouse embryonic fibroblasts deficient in Atg5 and/or Atg12 have shown hyperproduction of type I IFNs in response to isRNA structures, affirming the Atg5-Atg12 conjugate’s role in suppressing innate immune responses to viral replication [38]. In response to mouse norovirus (MNV) infection, instead of inducing autophagy, IFN-gamma activated phagocytes recruited Atg5-Atg12-Atg16L1 conjugate to suppress expression of MNV polymerases, thereby inhibiting the viral replication complex [39, 40]. Other studies have shown that decreasing levels of Atg13 and FIP200 have led to reduced viral replication of encephalomyocarditis virus, two picornaviruses, and coxsackievirus (CV) B3, although an exact mechanism of action remains to be elucidated [41]. In addition, FIP200 suppresses immune checkpoint therapy responses in breast cancers by limiting AZI2/TBK1/IRF signaling [42]. Finally, the Atg8-related family of proteins, such as Atg8 in yeast and LC3 in mammals, also play non-canonical roles in antiviral immune signaling. More specifically, IFN production in human peripheral blood mononuclear cells (PBMCs) is dependent on Toll-like receptor 9 (TLR9) trafficking into a specialized interferon signaling compartment via a LC3-associated phagocytosis (LAP) process [43, 44].

4. Autophagy, the Immune System, and Cancer

4.1. Autophagy in Immune Cell Differentiation

Findings have established a key role of autophagy in cellular homeostasis and survival, including in immune cells. Monocytes circulating in the blood typically undergo apoptosis in the absence of stimulation or differentiation. However, in the presence of activating signals such as inflammation, they become viable, migrate to tissues, and differentiate into macrophages. Autophagy activation is observed in monocytes survival and differentiation induced by Granulocyte-macrophage colony-stimulating factor (GM-CSF) [45]. Inhibition of autophagy with 3-methyladenine (3-MA), chloroquine (CQ), or BECN1 knockdown in monocytes reduces survival, increases apoptosis, and blocks macrophagic differentiation and acquisition of phagocytic functions even when treated with stimulatory factors [45–47]. Inhibition of the P2RY6-PRKAA1-ULK1 pathway blocked both macrophagic differentiation and autophagy in stimulated cells, suggesting a mechanism by which differentiation signals induce autophagy [48]. Thus, autophagy is essential for proper differentiation of monocytes into macrophages.

4.2. Autophagy in Pathogen Clearance

Within the immune system, one of the main routes of pathogen removal from tissues is phagocytosis and degradation. However, several intracellular pathogens have adapted mechanisms to evade destruction such as blocking lysosomal degradation or migrating out of phagocytic vesicles into the cytosol [46]. When typical phagocytic pathways fail to clear pathogens, autophagy’s function in degradation of cytosolic cellular components makes it well-suited to target pathogens that escape phagocytic vesicles and replicate in the cytoplasm [49]. Autophagy also enhances phagocytic pathways to clear pathogens [50–52]. LC3-associated phagocytosis (LAP) is a noncanonical form of autophagy where the core autophagy pathway component LC3 conjugates to phagosomes [53]. LAP plays an important role in regulating inflammatory responses and clearing cell debris, and it is required for clearance of Aspergillus fumigatus infection [50, 52]. LAP is also involved in the elimination of apoptotic cells by macrophages in a process known as efferocytosis to dampen proinflammatory responses [50, 52]. Interferon-stimulated gene 15 (ISG15) recruits LC3 to phagosomes for the IFN-gamma mediated restriction of Toxoplasma gondii infection, which involves engulfment of the parasite through LAP. This shows that ISG15 serves as one of the functional links connecting LAP and phagocytosis in host defense, suggesting its importance for future studies [54].

4.3. Autophagy Suppresses Inflammation

Autophagy is a potent anti-inflammatory process that inhibits inflammasome activation and modulates type I interferon responses [55, 56]. The interaction between autophagy and inflammation is complex and mediated by a number of mechanisms. Autophagy degrades proinflammatory signaling molecules and regulates inflammatory responses from innate immune cells [57]. Autophagy also inhibits inflammation through regulation of inflammasome complexes, which are required to process and activate proinflammatory signals such as procaspase 1 or pro-IL-1β [58]. Additionally. autophagy is a key regulator of RIP homotypic interaction motif (RHIM) domain proteins, which are molecules important for inflammatory signaling and cell death. Indeed, defective autophagy causes a decrease in turnover in RHIM domain proteins in cells, leading to an increase in necroptosis and inflammatory signaling [59]. Defective mitochondrial function is associated with accumulation of ROS inside the cell as well as inflammasome activation, suggesting that impaired mitochondrial quality control as a consequence of autophagy inhibition could be another mechanism connecting autophagy and inflammasome activation [60].

Autophagy has the paradoxical effects of promoting tumor survival while also suppressing initiation of new tumors [61]. A typical inflammatory response contributes to increased cell proliferation, cell survival, cell migration, and angiogenesis, which are critical in wound responses [62]. However, the upregulation of these processes in conjunction with the metabolic stress and DNA damage resulting from production of ROS that occurs in persistent and chronic inflammation contribute to tumorigenesis [62–66]. Therefore, suppression of inflammation is a mechanism by which autophagy inhibits tumor initiation. This effect is contrast to autophagy’s protective role in existing tumors and demonstrates that the role of autophagy in cancer is context-dependent.

4.4. Non-Cell Autonomous Autophagy Inhibits Type I and II IFN Response

Non-cell autonomous autophagy outside of the tumor microenvironment can inhibit an anti-tumor immune response [67]. In particular, autophagy-mediated “hepatic autophagy immune tolerance” is sufficient to inhibit anti-tumor T-cell responses against tumor growth. Systemic or liver-specific Atg7 or Atg5 ablation results in mitochondrial DNA (mtDNA) release, upregulating IFN type I and II responses and antigen presentation. This increased immune response leads to reduced tumor growth, which was restored by the loss of IFNγ or STING [67]. Overall, non-cell autonomous autophagy promotes tumor growth by reducing IFN levels and regulating the anti-tumor immune response.

4.5. Autophagy Inhibits Antigen Presentation

Antigen presentation is a critical process responsible for the body’s adaptive immune response against cancer growth. The primary responsibility of T cells, both CD8+ cytotoxic (CTLs) and CD4+ helper (Th cells), is to recognize foreign protein antigens docked on major histocompatibility complexes (MHCs) on the surface of malignant cells [68]. However, cancers can evade such immunity by immunodominance, display of immune checkpoints, or immunoediting for loss of specific tumor antigens [69].

The proteasome plays an essential role in immune surveillance mechanisms by generating peptides from intracellular antigens that are then “presented” to T cells [70]. Autophagy has also been identified as a route to display antigens on MHC class II molecules to CD4+ T cells, and is implicated in MHC class I cross-presentation of tumor antigen and the activation of CD8+ T cells [71, 72]. The expression of MHC I is commonly low in pancreatic ductal adenocarcinoma (PDAC), leading to defective antigen presentation that prevents T cell-mediated tumor killing. MHC I molecules are selectively targeted for lysosomal degradation by an autophagy-dependent mechanism that involves NBR1 [72]. Autophagy inhibition by genetically knocking-out essential autophagy genes or pharmacologically using autophagy-inhibitors such as CQ results in enhanced antigen peptide presentation, higher levels of T cells producing pro-inflammatory cytokines, such as IFNγ and tumor necrosis factor (TNF), along with overall lower metastatic burden [72]. An effective strategy of synergizing autophagy-inhibitor drugs with dual immune checkpoint blockade (ICB) therapy (anti-PD1 and anti-CTLA4 monoclonal antibodies) shows great promise at stemming PDAC metastasis [72]. Similarly, autophagy inhibits MHC I presentation in Kras-mutant, Lkb1-deficient (KL) lung cancer cells. Upon treatment with autophagy inhibitors, such as CQ or ULK1 inhibitors MRT68921, KL cells show restored sensitivity to host immune responses due to enhanced immunoproteasome activity and restoration of antigen processing [73]. As a result, autophagy inhibition increases the sensitivity of KL tumors to anti-PD-1 therapy [73].

T cell immunoglobulin and mucin domain protein-4 (TIM-4), which is expressed on myeloid cells, regulates T cell homeostasis, thereby impacting host immunity within the tumor microenvironment. TIM-4 is induced on tumor-infiltrating myeloid cells by tumor-derived danger-associated molecular patterns (DAMPs), which subsequently interacts with AMPKɑ1 and activates autophagy. Autophagy activation impedes tumor-antigen presentation, leading to impaired antitumor immunity [3]. Therefore, targeting the TIM-4-AMPKα1 interaction could be a unique strategy for augmenting anti-tumor T-cell responses in tumor microenvironment.

Besides regulating antigen presentation in tumor cells, autophagy supports proper antigen phagocytosis and presentation to MHC class II via modulation of CD36 in dendritic cells [74]. Moreover, autophagy is involved in B cell polarization and presentation of particulate antigens [75].

4.6. Autophagy Negatively Regulates Cytotoxic T Lymphocytes

IFNγ resistance is a conserved tumor cell-autonomous mechanism of CTL evasion in cancer. Autophagy within various cancer cells, such as renal and breast carcinomas, acts as a cell-autonomous negative regulator of CD8+ T cell metabolism and anti-tumor immunity [76, 77]. A combination of mapping cytokine- and CTL-based genetic interactions and in vivo CRISPR screens identified that autophagy is a conserved mediator of both evasion of CTLs by cancer cells and resistance to cytotoxicity induced by the cytokines IFNγ and TNF. Such regulation occurs through the autophagy–NF-κβ axis. Both genetic ablation (by Atg12 deletion) and pharmacological inhibition (using the VPS34 inhibitor Autophinib) to block autophagy sensitized cancer cells to TNF-induced CTL-mediated cell death [77]. Ablation of FIP200 in mice resulted in defective autophagy, increased production of chemokines in tumor cells, increased infiltration of effector T cells in the tumor microenvironment, and reduced mammary tumorigenesis [78]. Moreover, a non-canonical autophagy function of FIP200 is responsible for limiting T-cell recruitment and activation of the TBK1-IFN signaling axis [42].

4.7. Autophagy Supports Treg Lineage Stability and Survival Fitness, while Preventing Treg Infiltration into Tumors

Regulatory T cells (Tregs) are characterized by the expression of the master transcription factor forkhead box protein p3 (FoxP3). FoxP3 expression in Tregs is crucial for reinforcing Treg functional integrity and lineage stability [79]. In tumor immunity, FoxP3+ Tregs can promote the development and progression of tumors by preventing effective anti-tumor immune responses in tumor-bearing hosts [80]. Autophagy acts as a central and intrinsic regulator of Treg cell maintenance and immune homeostasis. Deletion of Atg7 or Atg5 specifically in Treg cells results in increased apoptosis and impaired lineage stability characterized by the loss of FoxP3 expression. Thus, autophagy in Tregs acts to support their functionality and stability. Mechanistically, autophagy protects Treg cell stability by restraining mTORC1-dependent c-Myc expression and function [79]. Treg infiltration in tumors has been associated with adverse prognosis of cancer patients. In FoxP3CreAtg7^fl/fl mice inoculated with MC38 colon adenocarcinoma cells, tumor growth was severely inhibited [79, 81] .

While autophagy ablation within Tregs leads to their impaired lineage stability, autophagy inhibited in certain cancers leads to a distinct immune response. In a genetically engineered mouse model (GEMM) for Kras-driven non-small cell lung cancer (NSCLC), conditional deletion of Atg5 in tumors markedly increased numbers of FoxP3+ Tregs infiltrating the lung tumor [82]. It has also been shown that late metastatic lung cancers treated with autophagy inhibitors such as CQ triggers upregulation of CD4+, Foxp3+ tumor infiltrating lymphocytes in late metastatic lung cancer tissues, leading to the induction of chemosensitization to carboplatin, immune activation and cell cycle arrest [83]. Thus, data from these studies point to the role that autophagy could prevent FoxP3+ Treg lymphocyte infiltration in tumor microenvironment.

Tumor cells deploy multiple mechanisms to escape from immunosurveillance. Autophagy promotes tumor immune tolerance and tumor growth via hepatic autophagy mediated suppression of STING and IFNγ activation, reduction of antigen presentation, inhibition of cytotoxic T cell infiltration and enforcing functional integrity of Tregs. However, autophagy ablation in certain cancers/settings have also shown to increase Treg infiltration, which could increase the sensitivity to chemotherapy (Fig. 4).

Figure 4. Autophagy modulates immune response for tumor progression.

A. Non-cell autonomous autophagy supports immune evasion for tumor growth. Hepatic autophagy suppresses STING and IFNγ activation to prevent tumor killing by T cells [67].

B. Cell autonomous autophagy promotes tumor growth via: 1) reduction of antigen presentation [71, 72] and 2) inhibition of cytotoxic T cell infiltration into the tumor [76, 77]. In certain types of lung cancers, 3) ablation of autophagy has been shown to activate regulatory T cells, thus promoting tumorigenesis [82, 83].

5. Autophagy and Cancer Metabolism

Cancer cells alter their metabolism to meet metabolic demands of high proliferation, malignance, and metastasis. Recent studies using genetically engineered mouse models (GEMMs) demonstrate that autophagy supports different types of tumor growth through distinct mechanisms, including both cell autonomous (tumor cell-intrinsic) and non-cell autonomous (host) autophagy [28, 30, 84–88] (Fig. 5).

Figure 5. Autophagy Supports Cancer Cell Metabolism.

A. Ablation of cell autonomous autophagy in tumor cells leads to a variety of outcomes: 1) lowered levels of amino acids, ATP production, and nucleotides synthesis due to reduced substrate recycling [29, 89, 90], 2) defective FAO and increased sensitivity to FAO inhibition [28, 30], 3) p53 induction [28, 95–97], and 4) higher basal ROS levels [29].

B. Ablation of non-cell autonomous autophagy causes various outcomes in the host and the TME: 1) Ablation of host autophagy or liver autophagy causes the release of Arginase I from hepatocytes, thereby reducing circulating arginine that is essential for tumor growth [31], 2) Ablation of systemic autophagy also leads to deduced metabolites in the TME, leading to tumor death [31], 3) Ablation of systemic autophagy is associated with reduced mTOR activity, impairing tumor growth [27, 31], and 4) ablation of autophagy in stromal cells of PDAC restrains the supply of metabolites to tumor cells [33].

Figure Template Taken from BioRender.com [169].

Tumor Cell-Intrinsic Autophagy Supports Tumor Growth

5.1. Autophagy Maintains Energy Homeostasis.

Autophagy is known to promote cell survival through a number of cell autonomous mechanisms such as mitigating metabolic stress, in part by recycling intracellular components to provide substrates for metabolic and biosynthetic pathways [29, 33]. Pulse-chase studies with isotope-labeled nutrients revealed that autophagy is essential to maintain levels of amino acids when cancer cells are starved [29]. This “autophagy addiction” may arise out of the role of autophagy in sustaining mitochondrial metabolism, which is critical for cancer cell survival and tumorigenesis, through mitophagy and recycling of substrates for the TCA cycle [89, 90]. Autophagy ablation significantly reduced mitochondrial respiration and ATP production during starvation, leading to fatal nucleotide depletion, which was rescued by exogenous glutamine supplementation [29]. This suggests that glutamine recycled through autophagy is essential to maintain proper mitochondrial function for energy production and homeostasis. Autophagy defective tumor cells also accumulate structurally abnormal mitochondria and exhibit other defects in mitochondrial respiration [84]. Autophagy is thus required for proper mitochondrial function and tumor cell metabolism under conditions of metabolic stress.

5.2. Autophagy Regulates Lipid Metabolism

Fatty acids are essential metabolic intermediates for several biosynthetic pathways, and in cancer cells continuous fatty acid synthesis is required for proliferation and maintaining metabolism [91]. Autophagy deficient Kras-mutant p53-deficient (KP) lung tumors show increased lipid droplet (LD) accumulation due to defective lipophagy and impaired fatty acid oxidation (FAO) [28]. However, in Kras-driven Lkb1-deficient (KL) lung tumors, autophagy ablation upregulates lipolysis, and autophagy deficient cancer cells rely more on FAO to survive starvation compared to autophagy intact cells. This has shown to lead to excessive FAO and energy crisis from depletion of LDs [30]. Despite exhibiting paradoxical responses to autophagy deficiency, both autophagy-deficient KP and KL tumor derived cell lines are significantly more sensitive to FAO inhibition during starvation as well as starvation-induced cell death compared with autophagy competent cells [28, 30]. Interestingly, inactivation of autophagy by FIP200 ablation reduced fatty acid release from LDs, resulting in decreased mTORC1 hyperactivation in TSC-deficient neural stem cells and a reversal of neural defects [92].

5.3. Autophagy Inhibits p53-mediated Tumor Suppression

p53 is a well-known tumor suppressive gene that inhibits tumor growth through several mechanisms, and there is growing evidence for tumor promoting interactions between p53 and autophagy. Under nutrient deprivation, AMPK activation induces p53 activation [93]. Subsequently, induction of p53 activates the transcription of genes involved in autophagy activation or induces autophagy via transcriptional activation of damage-regulated autophagy modulator (DRAM-1) in human and mouse cell lines [94]. On the other hand, in various mouse models, p53 induction was observed when autophagy was ablated, suggesting that autophagy also regulates p53 and may serve as a resistance mechanism against p53 activators [95].

In normal cells, autophagy protects tissues from damage resulting from p53 overactivation, but in tumor cells this same function prevents p53-induced apoptosis and promotes tumor cell survival. p53 induction was observed in Atg7-deficient Kras-driven lung tumors, accompanied by accumulation of defective mitochondria, reduction of tumor burden and suppression of cell proliferation. This suppression was partially relieved by the deletion of p53 [28]. This is supported by similar findings in models of PALB2-associated breast cancer, where autophagy ablation reduced tumorigenesis in wild type models but not in models with p53 conditionally deleted [96]. Thus, autophagy promotes tumorigenesis by inhibiting p53-mediated tumor suppression [97].

This conclusion is complicated by findings that in p53-deficient mice containing oncogenic KRAS, autophagy inhibition accelerates pancreatic tumor growth, suggesting that the role of autophagy in pancreatic tumor progression is context dependent on the status of p53 in tumor cells [98]. On the other hand, another study showed contradictory findings in which autophagy deletion in the setting of p53 loss impaired pancreatic tumor progression [99]. These studies utilized different models of p53 deletion which could have contributed to conflicting results. Overall, the interactions between autophagy and p53 are complex and further investigation can elucidate the role of these interactions in various cancers.

5.4. Autophagy Attenuates Oxidative Stress

ROS are constantly generated in the cell from a number of processes including cellular respiration, and excessive levels of ROS result in oxidative stress that can interfere with vital functions [100]. Due to differences in metabolism from normal cells, including increased mitochondrial dysfunction, cancer cells have elevated levels of ROS, which if unmitigated can lead to damage to cellular components and induce cell death [101, 102]. Autophagy is activated in response to ROS production and oxidative stress, suggesting that it plays an antioxidant role, particularly in tumor cells [103–105]. Indeed, Atg7 deficient tumor cells showed higher basal ROS levels in nutrient rich conditions and even higher levels in starvation conditions, which was attenuated by glutamine supplementation [29]. Mitochondrial dysfunction leads to elevated ROS production, so autophagy’s role in organelle quality control and clearance of dysfunctional mitochondria is also important to mitigate oxidative stress [106, 107]. Thus, autophagy is required to sustain redox balance, maintain mitochondrial function, and suppress ROS production during metabolic stress, and these are all functions that promote tumor cell survival.

5.5. Autophagy Prevents Genomic Instability

While autophagy may play a tumor-promoting role as a survival pathway for tumor cells, it also acts as a suppressor of tumor initiation. The metabolic stress caused by starvation or other stressors can induce DNA damage and chromosomal instability. Autophagy-competent cells can mitigate this metabolic stress and limit genomic instability [108]. However, in autophagy-deficient cells, an accumulation of oncogenic mutations resulting from metabolic stress can serve to promote chromosomal abnormalities and tumorigenesis [16]. Treatment of cells with an autophagy inhibitor led to elevated levels of micronuclei, a sensitive indicator of genomic instability and DNA damage [109]. Hence, autophagy can suppresses tumor initiation by suppressing known inducers of genomic instability and tumor initiation, such as reactive oxygen species (ROS), tissue damage, and inflammation [15].

Several nontraditional autophagy functions represent mechanisms by which autophagy plays an important role in maintaining genomic stability. Through noncanonical piecemeal microautophagy (PMN), cells can specifically remove parts of the nuclei containing damaged DNA to mitigate effects of oxidative stress [105]. Autophagy is also linked to the DNA damage response (DDR) in a number of ways, such as induction of AMPK by DDR, suggesting that autophagy plays a role in maintaining genomic stability through DNA repair [110, 111]. Autophagy is required for the selective degradation of RNA granules containing localized retrotransposon replication intermediates. Retrotransposons are a major source of genetic variation chromosomal instability, and mice lacking essential autophagy genes accumulate both retrotransposon RNA and genomic insertions [112]. Appropriative activation of RhoA GTPase is essential in proper cytokinesis at the end of mitosis, and autophagy deficient cells demonstrate cytokinesis failure, chromosomal instability, and multi-nucleation [113].

Host Autophagy Supports Tumor Growth

5.6. Autophagy Maintains Circulating Arginine

Tumor survival heavily relies on a variety of circulating nutrients present in the host bloodstream. Besides tumor-intrinsic autophagy, systemic host autophagy is indispensable for tumorigenesis [27, 31]. Systemic ablation of autophagy leads to the depletion of a number of amino acids and other essential metabolites in serum, including leucine, arginine and methionine, some of which are essential for mTORC1 activation [27, 31, 95, 114–116]. Indeed, acute, systemic deletion of Atg7 or Atg5 impaired tumor growth, which was accompanied by reduced mTOR activity [27, 31]. Among the altered circulating metabolites caused by the loss of autophagy, reduction of serum arginine levels is the most striking [27, 31]. “Arginine-auxotroph” tumors are particularly sensitive to depleting arginine levels and are characterized by their lack of expression for argininosuccinate synthase, which furthers their dependence on environmental arginine [31]. Systemic or liver specific autophagy ablation leads to liver damage, which triggers the release of arginine-degrading enzyme arginase I (ARG1) from the liver into the serum and causes the degradation of arginine to ornithine. Dietary supplementation of arginine to the autophagy deficient host or liver-specific autophagy deficient host partially restored levels of circulating arginine and rescued tumor growth [31].

5.7. Autophagy Provides Nourishment in the Tumor Microenvironment

Not only do tumor cells rely on host metabolites for nutrition, but the tumor microenvironment also serves as a potent source of nourishment for cancer cells. Tumor microenvironment is composed of endothelial cells, fibroblasts, inflammatory cells, extracellular matrix, cytokines, and growth factors, which support commencement and progression of many types of tumors. Pancreatic cancer has a robust stromal component that damages vasculature, thereby creating a nutrient poor environment. Acute, systemic expression of a dominant-negative ATG4b in mice led to regression of KRAS-driven pancreatic cancer, which is caused by the elimination of autophagy-dependent metabolic cross-talk between tumor cells and nearby stromal cells [33]. Autophagy activation in stromal-associated pancreatic stellate cells promotes the secretion of alanine and other amino acids into the tumor microenvironment to provide substrates to neighboring PDAC cells for lipid and protein biosynthesis [32, 33].

The tumor microenvironment is also associated with an increase in macrophage levels, possibly due to immune responses that aim to clear dying tumor cells [33]. The inflammatory effects of these macrophages can be suppressed via activation of LAP in the myeloid compartment of the tumor microenvironment, which additionally aims to suppress surrounding T lymphocytes [117]. Inhibition of LAP consequently provides a potential avenue for augmenting the anti-tumor effects of T lymphocytes.

The tumor microenvironment also provides exogenous signals to promote tumor autophagy during times of starvation. For RAS-driven Drosophila melanogaster cancer models, the release of TNF or IL-6 by starving tumor cells into the microenvironment acted as exogenous signals to induce autophagy, both in the microenvironment and distal tissues. Normal epithelial cells are engaged as an active part of the microenvironment by providing amino acids, supporting tumor proliferation [118]. Moreover, allografted tumor implants placed into autophagy-deficient and control models demonstrated poorer tumor growth in autophagy-deficient models, regardless of where in the body the tumor was implanted. Tumor growth can subsequently be rescued if the tumor implants are transplanted into a model where autophagy is sustained [118]. Thus, systemic autophagy plays a major role in determining growth, regardless of tumor location or primary tumor microenvironment.

6. Autophagy and Metastasis

While autophagy is known to promote tumor cell survival through a number of mechanisms, the role of autophagy in cancer metastasis is also being examined in depth. Emerging evidence indicates that autophagy is upregulated during tumor metastasis and suggests that autophagy promotes several key processes during metastasis. In general, autophagy’s importance in cellular survival in response to metabolic stress makes it well-suited to promote metastasis, where tumor cells must survive and adapt to altered extracellular conditions [119]. The transition of epithelial to more mobile and invasive mesenchymal cells through the epithelial-mesenchymal transition (EMT) often occurs during embryonic development, wound healing, and the early stages of metastasis [120]. Induction of autophagy promotes EMT through TGF-beta dependent signaling, while ablation of autophagy diminishes EMT and cancer cell invasiveness- which can be rescued through addition of recombinant TGF-beta [121]. A recent study suggests that autophagy may promote EMT through myeloid cell activity, a hypothesis supported by findings that myeloid-specific autophagy ablation decreased tumor cell invasiveness, suppressed EMT, and reduced TGF-beta production [122]. In addition to directly promoting tumor metastasis and invasiveness, autophagy plays a key role in enabling malignant cancer cells to enter a reversible dormant state and then reactivate at a later time [123]. These cells are often resistant to cancer treatments which tend to target actively proliferating tumor cells. Through a number of mechanisms, autophagy promotes not only cancer cell survival but also metastasis and treatment resistance [124].

Though autophagy seems to promote cancer metastasis through a number of mechanisms, several studies show that autophagy suppresses growth of migratory tumor cells into lethal macrometastases [125]. In a GEMM with transplanted tumor cells, the size of metastatic lesions was significantly increased in mice with defective autophagy compared to controls. Interestingly, results from the study indicate that autophagy only meaningfully suppressed metastasis in the initial stages of outgrowth [126]. Surprisingly, pharmacological inhibition of autophagy in CQ does not result in enhanced tumor metastasis. This may be explained by the fact that in CQ treatment, autophagosomes still form since core autophagy machinery is functional, but after CQ treatment NBR1 remains sequestered in undigested autophagosomes [127]. This hypothesis is supported by the finding that preventing cytoplasmic accumulation of NBR1 in autophagy-deficient tumor cells reversed increases in metastasis [126, 127]. Autophagy is a pleiotropic cellular process with a number of diverse and seemingly contradictory effects, and its elucidation can aid in the development of new and more effective therapeutics.

7. Autophagy and Cancer Treatment

Given the significant role autophagy plays in tumor progression and the effects of its ablation on tumor survival, autophagy has been explored as a key target for cancer treatments (Fig. 6). The modulation of autophagy can enhance patient outcomes when combined with current treatments [128].

Figure 6. Autophagy in cancer treatment.

Autophagy inhibition in combination with other cancer treatments can potentiate overall treatment effectiveness and overcome resistance. One common cancer treatment involves the use of RAS/RAF/MEK/ERK pathway inhibitors, but tumor cells can adapt resistance to this therapeutic strategy. ERK inhibition can partially downregulate mTOR activity, leading to de-repression and activation of protective autophagy, which may partially explain tumor adaptation to ERK pathway inhibitors such MEK inhibitors. Combining an ERK inhibitor with both autophagy inhibition and mTOR inhibition can therefore overcome resistance pathways and increase overall response to these treatments.

7.1. Autophagy Inhibition combined with MAPK/ERK Inhibitors

RAS proteins are central mediators downstream of growth factor receptor signaling and are critical for cell proliferation, survival, and differentiation. In addition, RAS mutations are frequently detected in many types of cancers [129, 130]. Therefore, targeting components of the Ras-mediated MAPK cascade, also known as the RAF/MEK/ERK pathway, is an attractive strategy in the development of novel therapeutic approaches to treat RAS-driven cancers [131]. With a near 100% KRAS mutation frequency, PDAC is considered the most RAS-addicted of all cancers [132]. In particular, the RAF/MEK/ERK pathway is central to the initiation and maintenance of PDAC [133]. Attempts have been made to treat PDAC through the use of ERK/MAPK pathway inhibitors. However, such treatments generally fail to produce durable responses in patients [134], indicating that cancer cells develop inhibitor bypass mechanisms that limit the effectiveness of this approach.

A combinatorial siRNA approach to simultaneously target multiple genes in KRAS-driven cells showed that targeting BRAF and CRAF in combination with Atg7 is a promising therapeutic approach for RAS-driven cancers [135]. One possible reason for this is that the inhibition of the RAF/MEK/ERK pathway activates autophagy as a survival response to ERK inhibition, which is supported by findings that genetic or pharmacological inhibition of autophagy enhanced the anti-tumor activity of ERK inhibition in KRAS-driven PDAC [133, 136, 137]. ERK pathway inhibition is also associated with decreased mTORC1 signaling, and since mTOR activity regulates autophagy activation, inhibition of ERK leads to a de-repression of autophagy [138, 139]. This is consistent with findings that ERK inhibition elicits autophagy as a protective survival mechanism and may explain resistance to ERK inhibition based treatment [136]

Inhibition of MEK and autophagy had anti-proliferative effects against PDAC cells both in vitro and in mice xenografted tumors [138]. This anti-proliferative effect of combined ERK pathway and autophagy inhibition persisted for other cancer types including melanoma and colorectal cancer [136, 140]. Combinational treatment of MEK inhibitor Trametinib with autophagy inhibitor hydroxychloroquine (HCQ) in a human patient with PDAC even resulted in a partial but nonetheless striking disease response [136, 137].

Pediatric central nervous system (CNS) tumor cells with BRAF(V600E) (but not wild-type) cells display high rates of induced autophagy, are sensitive to pharmacologic and genetic autophagy inhibition, and show synergy when the clinically used autophagy inhibitor CQ was combined with the RAF inhibitor Vemurafenib or standard chemotherapeutics [141]. A patient with BRAF-mutant brain cancer who had progressed while on anti-BRAF therapy experienced more than 2.5 years of disease regression on a combination of CQ plus a BRAF-targeted therapy[141]. Hence, autophagy inhibition in combination with or RAF/MEK/ERK pathway inhibitors presents promising possibilities for future therapies.

While autophagy inhibition in combination with ERK inhibition has shown promising results for PDAC patients, other cancers mediated by RAS activation have also shown sensitivity to this combination therapy [136, 137, 140]. Therefore, it would be worth determining if all RAS-driven cancers are susceptible to this therapeutic modality. Furthermore, studies should be conducted to determine whether the effectiveness of autophagy inhibition is specific to RAS-driven cancers or if it can be effective for other cancer types as well.

7.2. Autophagy inhibition combined with mTOR Inhibitors

mTOR regulates cell growth, proliferation, and survival via mTORC1 and mTORC2, which are often aberrantly activated in cancers. mTOR-based targeting strategies exhibit potent anti-cancer effects in many preclinical models. However, development of resistance can occur, which limits its success in the clinic [142, 143]. There are multiple mechanisms underlying resistance to mTOR inhibitors, including the rewiring of metabolic pathways [143]. Autophagy activation is tightly regulated by mTOR, so induction of autophagy by mTOR inhibitors may lead to the drug resistance and survival of cancer cells. Therefore, studies of mTOR inhibitors combined with autophagy inhibitor HCQ have been tested in several different cancer types. Myeloma cells treated with both cyclophosphamide and rapamycin have shown to “doubly” induce autophagy-dependence, providing a more robust platform for HCQ to work on [144]. Moreover, multiple patients experienced prolonged stable disease and less adverse growth rates; however, it is not to the degree that would provide any meaningful effect on patient outcomes [144]. In patients with clear-cell renal carcinoma (ccRCC), treatment with the mTOR inhibitor Everolimus along with HCQ was seen to prolong stable disease [145]. mTOR inhibitor Temsirolimus, when combined with HCQ, has been proven to be safe and tolerable in a phase I clinical trial of patients with either advanced solid malignant tumors or metastatic melanoma. In fact, a median of 3.5 months of PFS was achieved in approximately 68% of the melanoma patients [146]. A HCQ dose of 600mg twice daily in combination with a Temsirolimus dose of 25mg weekly was ultimately recommended as approximately 67% of solid tumor patients and 74% of melanoma patients achieved stable disease with this dosing regimen [146].

7.3. Autophagy Inhibition combined with Immunotherapy

Recent studies have demonstrated that both tumor-intrinsic autophagy and host autophagy play essential role for tumor cells to escape immunosurveillance [67, 72]. Moreover, HCQ synergizes with ICB therapy to inhibit pancreatic tumor growth [72]. Host autophagy inhibition also promotes inflammation in the tumor microenvironment to enhance immune surveillance [67]. A noncanonical function of the essential autophagy gene FIP200 was found to limit T-cell recruitment in mammary tumors and promote tumor progression. When combined with immune checkpoint inhibitors, ablation of FIP200 enhanced efficacy of this therapy and slowed tumor growth [42]. Based on these animal experiments, studies combining autophagy inhibitors and immune checkpoint inhibitors may hold promise for human trials.

7.4. Autophagy Inhibition combined with Other Therapies

Many forms of cancer therapy such as chemotherapy and radiotherapy rely on inducing stress in cancer cells in order to cause tumor death. However, the protective role of autophagy under stress conditions mitigates these effects and can contribute to therapy resistance. Autophagy is upregulated in response to many chemotherapeutic drugs, suggesting that autophagy inhibition in combination with chemotherapy may yield more effective treatments than chemotherapy alone [147, 148]. Indeed, studies combining autophagy inhibition with chemotherapy agents have shown promising results [149]. A novel inhibitor of autophagy acting on ULK1 sensitized NSCLC cells to cisplatin and suppressed cancer cell growth [150]. Promising results have also been shown in human clinical trials utilizing an autophagy inhibition and chemotherapy combination [149]. A phase Ib/II single-arm study for metastatic NSCLC utilized carboplatin and paclitaxel chemotherapies combined with HCQ. Not only was the treatment found to be generally safe and tolerable, but patients’ ORR and PFS showed marked improvement [151].

Radiation therapy often works by inducing DNA damage in cancer cells, causing them to undergo cell death. However, autophagy may serve as a resistance mechanism against this kind of therapy by removing damaged DNA and promoting DNA repair, preventing cell death [152]. Therefore, autophagy inhibition may also potentiate radiotherapy, and indeed combining autophagy inhibitors with radiation has shown improved anti-tumor effects [149]. Treatment with chloroquine sensitized bladder cancer cells to radiation, elevating apoptosis markers, inhibiting proliferation, and impeding tumorigenesis of xenograft tumors [153]. A study on whole brain radiation therapy combined with chloroquine showed that treatment was well-tolerated and suggested a survival benefit compared to radiotherapy alone [154]. Autophagy inhibition may play role as a treatment modulator even for radiotherapy.

7.5. Developing New Autophagy Inhibitors

So far, CQ and its derivative HCQ are the only FDA approved autophagy inhibitors for clinical usage. Indeed, besides blocking the fusion of autophagosomes with lysosomes using CQ and HCQ, autophagy processes can be targeted at multiple steps, leading to other autophagy inhibition strategies that are currently under investigation, including Vps34 inhibitors, ULK1 inhibitors, and Palmitoyl-Protein Thioesterase1 (PPT1) inhibitors. ULK1 inhibitors, such as SBI-0206965, aim to suppress ULK1’s function in phosphorylating enzymes required for autophagy initiation [155]. Vps34 inhibitors, such as SAR405, aim to inhibit Vps34’s role in vesicle trafficking and autophagosome formation [156]. Finally, PPT1 inhibitors such as Lys05 aim to deacidify autophagic lysosomes by inhibiting the action of the lysosomal enzyme PPT1 [157].

7.6. Identifying Potential Biomarkers for Autophagy Inhibition

RAS activation promotes tumor cell reliance on autophagy [90, 158, 159]. Studies from pre-clinical mouse models and clinical trials demonstrated autophagy inhibition could be effective in treating pancreatic cancer when combined with MEK inhibition or ICB therapy [72, 136, 137]. Similar to PDAC, KRAS mutations are frequently detected in NSCLC patients and autophagy is essential for Kras-driven lung tumor growth [27, 28, 30]. LKB1, a metabolic sensor and master modulator of AMPK, is the third most mutated gene detected in NSCLC. Interestingly, ablation of the LKB1 causes Kras-driven NSCLC to become much more sensitive to autophagy inhibition compared with LKB1 WT lung tumors [30]. Autophagy temporally compensates for acute systemic LKB1 loss for adult mouse survival [95]. In the case of Kras- driven lung tumors, studies suggest that the loss of LKB1 leads to the reduced metabolic plasticity in response to metabolic stress during tumorigenesis, which is further exaggerated by autophagy ablation, leading to energy crises in the tumor cell [30]. LKB1 deficiencies are also correlated with elevated tumor mutational burdens (TMBs) in NSCLCs. However, NSCLC patients bearing co-mutations of LKB1 and KRAS are resistant to immunotherapy [160]. Recent studies found that, compared to LKB1 WT lung tumors, LKB1-deficient cancers treated with autophagy inhibitors show higher sensitivity to immune checkpoint blockade anti-PD-1 treatment [73]. As a result, LKB1 mutations could consequently be explored as a predictive biomarker for the treatment of lung cancers with autophagy inhibition. However, the potential of LKB1 biomarkers still requires further investigation.

8. Addressing Arguments that Autophagy is Anti-Tumorigenic

While the above studies certainly point to the autophagy-mediated tumor promotion and promising effects of autophagy inhibition in cancer treatment, this contention is certainly up for debate. Other studies report that autophagy can actually be anti-tumorigenic and thus oppose inhibiting autophagy in cancer lines [161–163]. For example, one study has shown that immunosurveillance against transplanted carcinomas can be improved by pharmacologically inducing autophagy under specific caloric restrictions [164]. The role of autophagy in cancer and tumor progression seems paradoxical due to apparently contradictory effects of autophagy dysfunction. It has been shown that autophagy serves as a survival mechanism for tumor cells in response to metabolic stress, but there is growing evidence for autophagy as a tumor suppression mechanism. This evidence includes findings that deletion of essential autophagy genes suppressed tumor initiation in mouse models of breast cancer and that defects in autophagy resulting from allelic loss of Beclin1 promote tumorigenesis and are commonly observed in multiple human cancers [161, 162]. However, this evidence is complicated by studies that demonstrate that loss of BECN1 in breast cancer does not occur independently of deletion of BRCA1, a known tumor suppressor gene [163, 165]. This suggests that BRCA1 is the main driver or tumorigenesis in these cases [61]. Indeed, core autophagy machinery and genes are generally not mutated across most cancers [166].

Another argument against autophagy inhibition is tolerability. Some studies state that human cancer lines grown in vitro remain unaffected and show no size reduction when Atg genes are deleted [167]. However, it has been shown that autophagy-ablated tumors indeed display growth defects and can be restored upon nutrient-rich conditions despite being sensitive to starvation [28, 30]. Additionally, the in vitro environment may lack crucial anti-tumor effects that are prominent within the in vivo environment, such as immune responses, stromal cells in tumor microenvironment, or development of resistance mechanisms. Indeed, by claiming to observe tumor reduction in mice which have compromised immune systems [27, 67, 72], critics argue that immunocompromised mice models neglect the realistic activation of the host immune system compared to autochthonous models [167]. However, the use of nude mice is critical to investigate tumor progression dependency on oncogenic signaling. Additionally, nude mice do not necessarily neglect in vivo host immune responses, as activation of innate and adaptive immune responses are defining characteristics of autophagy-deficient tumor cells [64].

An additional argument proposes that it is only host autophagy that promotes tumor growth, not tumor cell-autonomous autophagy. Proponents of this argument conclude that inhibition of autophagy in cancer cells therefore is a fruitless endeavor [168]. However, extensive data supports the importance of tumor-intrinsic autophagy, in addition to host autophagy, in promoting cancer energy homeostasis, metabolism, attenuation of oxidative stress and immune tolerance [29, 30, 72, 73, 103–105].

9. Concluding Remarks

This review seeks to provide a broad overview of the interactions of autophagy and cancer as well as examine potential clinical applications of findings in this area. Cell-autonomous autophagy supports tumor growth via intracellular recycling to maintain essential nutrients, removing toxicity caused by the accumulation of protein aggregates and damaged organelles, inhibiting p53 function, and preventing cell surface expression of MHC-I for antigen presentation to escape immunosurveillance. Host autophagy support tumor growth by sustaining necessary levels of essential amino acids and metabolites for tumor cell metabolism and inhibiting innate and adaptive anti-tumor immune responses. Given its role in cell metabolism and immune responses, autophagy inhibition can potentially be combined with MAPK/ERK inhibitors, mTOR inhibitors or immunotherapy as novel cancer therapies. Autophagy and its role in cancer continues to be uncovered, and further research in this area can reveal new ways to target various kinds of cancer and disease.

Abbreviations:

3-MA

3-methyladenine

AMPK

AMP-activated protein kinase

ARG1

arginine-degrading enzyme arginase I

ATG

autophagy-related gene

BECN1

beclin-1

ccRCC

clear-cell renal carcinoma

CNS

central nervous system

CQ

chloroquine

CTL

cytotoxic T lymphocyte

CV

coxsackievirus

DAMP

danger-associated molecular patterns

DDR

DNA damage response

DRAM

damage-regulated autophagy modulator

EMT

epithelial-mesenchymal transition

FAO

fatty acid oxidation

FIP200

focal adhesion kinase interacting protein of 200kDa

GEMM

genetically engineered mouse model

GM-CSF

granulocyte-macrophage colony-stimulating factor

HCQ

hydroxychloroquine

HGF

hepatocyte growth factor

ICB

immune checkpoint blockade

IGF1

insulin-like growth factor-1

IL-3

interleukin 3

ISG15

interferon-stimulated gene 15

isRNA

immunostimulatory RNA

KL

Kras-mutant LKB1-deficient

KP

Kras-mutant p53-deficient

LAP

LC3-associated phagocytosis

LC3

microtubule-associated protein 1A/1B light chain 3B

LC3-PE

LC3-phosphatidylethanolamine

LD

lipid droplet

LIR

LC3-interaction region

MHC

major histocompatibility complex

MNV

mouse norovirus

mtDNA

mitochondrial DNA

mTOR

mammalian target of Rapamycin

mTORC1

mTOR complex 1

NBR1

neighbor of BRCA1 gene 1 protein

NSCLC

non-small cell lung cancer

PBMC

peripheral blood mononuclear cell

PDAC

pancreatic ductal adenocarcinoma

PI3K

phosphoinositide 3-kinase

PKB

protein kinase B

PMN

piecemeal microautophagy

PPT1

palmitoyl-protein thioesterase1

RHIM

RIP homotypic interaction motif

ROS

reactive oxygen species

RTK

receptor tyrosine kinase

TIM-4

T cell immunoglobulin and mucin domain protein-4

TLR

toll-like receptor

TSC1/2

tuberous sclerosis complex 1/2

ULK1/2

Unc-51 like autophagy activating kinase 1/2

Vps34

vacuolar protein sorting 34 complex