Autophagy Regulation on Cancer Stem Cell Maintenance, Metastasis, and Therapy Resistance

Simple Summary Certain types of cancer have higher relapse rates compared to others, and cancer stem cells (CSCs) have been shown as the main drivers of cancer relapse and cancer severity. This subpopulation of cells displays stem-like characteristics which bolster tumorigenesis along with metastasis and lead to poorer prognoses. Autophagy has been studied as a mechanism by which CSCs maintain stemness and acquire resistance to chemotherapy and radiation. The aim of this review is to condense and organize what has been recently published on the connection between cancer stem cells (CSCs) and autophagy. Multiple studies on autophagy have suggested that the pathway is a double-edged sword, which can either undermine or enhance CSC characteristics depending on interactions with different pathways. Thus, future research should investigate regulation of autophagy in combination with traditional cancer therapies as a possible method to effectively eliminate CSCs and minimize cancer relapse. Abstract Cancer stem cells (CSCs) are a subset of the tumor population that play critical roles in tumorigenicity, metastasis, and relapse. A key feature of CSCs is their resistance to numerous therapeutic strategies which include chemotherapy, radiation, and immune checkpoint inhibitors. In recent years, there is a growing body of literature that suggests a link between CSC maintenance and autophagy, a mechanism to recycle intracellular components during moments of environmental stress, especially since CSCs thrive in a tumor microenvironment that is plagued with hypoxia, acidosis, and lack of nutrients. Autophagy activation has been shown to aid in the upkeep of a stemness state along with bolstering resistance to cancer treatment. However, recent studies have also suggested that autophagy is a double-edged sword with anti-tumorigenic properties under certain circumstances. This review summarizes and integrates what has been published in the literature in terms of what role autophagy plays in stemness maintenance of CSCs and suggests that there is a more complex interplay between autophagy and apoptosis which involves multiple pathways of regulation. Future cancer therapy strategies are needed to eradicate this resistant subset of the cell population through autophagy regulation.


Cancer Stem Cells
CSCs, also known as tumor initiating cells (TIC) and cancer initiating cells (CIC), are a select subset within the tumor population that have been demonstrated to initiate tumor formation, tumorigenicity, resistance to different types of treatment, and metastasis [1]. CSCs have also been insinuated as the primary culprit for cancer severity and cancer relapse in various solid tumors, including breast, ovarian, brain, lung, pancreatic, liver, and colorectal cancers [2][3][4][5][6]. CSCs are capable of remodeling the tumor microenvironment (TME) through the use of matrix metalloproteinases (MMP), which are calcium-dependent zinc-containing endopeptidases induced by hypoxia and extracellular acidosis [1,3,[7][8][9]. TME reconstruction enhances cancer invasion and metastasis, allowing CSCs to undergo

Autophagy
Autophagy is a mechanism of energy metabolism where intracellular components including damaged organelles, pathogens, and non-essential proteins are sequestered into vesicles and recycled during moments of environmental stress such as nutrient deprivation, oxidative stress, hypoxia, and infection in order to provide nutrients and overcome new stressors [8,46]. Autophagy can be broken down into three main groups: macro-autophagy, micro-autophagy, and chaperone-mediated autophagy. Macro-autophagy is characterized by the envelopment of materials by endoplasmic reticulum-derived membranes followed by fusion with lysosomes for degradation [47]. Micro-autophagy is the direct engulfment of cytoplasmic material, which is mediated by the lysosomes [48]. Lastly, chaperone-mediated autophagy is when chaperone proteins bind and transport intracellular proteins to the lysosomes [49]. This paper will mainly focus on macro-autophagy and will subsequently use the convention of referring to this process as "autophagy." Interestingly, there are two different roles for macro-autophagy reported in the literature. Lethal/toxic autophagy, also known as autophagic cell death, results in increased cell death, which has been reported in glioblastoma and HCC [50,51]. Conversely, protective autophagy is a major cause of survival and chemotherapy resistance in CSC populations [15]. This form of autophagy promotes metastasis, allowing the cancer to spread and wreak havoc throughout the body. This was demonstrated in pancreatic, colorectal, and lung cancers [5,6,52].
Initiation of autophagy begins with the formation of a non-degradative vesicle called the autophagosome. This vesicle then fuses with a lysosome to form the autolysosome, where intracellular materials are degraded and released into the cytoplasm for recycling [4,46]. Autophagosome generation occurs via a three-step process; initiation starts with autophagic machinery assembling at the membrane source site to form the pre-autophagosome membrane. Two main multi-protein complexes, Unc-51-like autophagy activating kinase (ULK) complex and vacuolar protein sorting 34-beclin 1 (VSP34-BECN1) complex, are both needed for initiation and nucleation of the isolation membrane/phagophore, which is a double membrane that encloses cytosolic material [46,53]. Autophagy related protein 9 (ATG9) recruits lipids for the isolation membrane [54]. Protease ATG4 catalyzes the cleavage of microtubule-associated protein light chain 3 (LC3) to LC3-I. ATG7, ATG3, and the ATG5-ATG12-ATG16L1 complex covalently conjugate phosphatidyl-ethanolamine (PE) with LC3-I, forming LC3-II [2]. LC3-II relocates to the nucleation membrane to initiate elongation and closure of the autophagosome around intracellular materials. Finally, SNARE-like proteins promote the fusion of lysosomes with autophagosomes to form autolysosomes ( Figure 1) [46]. [4,46]. Autophagosome generation occurs via a three-step process; initiation autophagic machinery assembling at the membrane source site to form the pr gosome membrane. Two main multi-protein complexes, Unc-51-like autophagy kinase (ULK) complex and vacuolar protein sorting 34-beclin 1 (VSP34-BECN1 are both needed for initiation and nucleation of the isolation membrane/ph which is a double membrane that encloses cytosolic material [46,53]. Autopha protein 9 (ATG9) recruits lipids for the isolation membrane [54]. Protease ATG the cleavage of microtubule-associated protein light chain 3 (LC3) to LC3-I. AT and the ATG5-ATG12-ATG16L1 complex covalently conjugate phosphatidy mine (PE) with LC3-I, forming LC3-II [2]. LC3-II relocates to the nucleation me initiate elongation and closure of the autophagosome around intracellular m nally, SNARE-like proteins promote the fusion of lysosomes with autophag form autolysosomes (Figure 1) [46]. The autophagy pathway has many points of regulation. It is mainly sup mammalian target of rapamycin (mTOR), which inhibits ULK complex activat Additionally, Bcl-2 family proteins inhibit VSP34-BECN1 complex activation cent paper demonstrated in a colorectal carcinoma (CRC) model that the YAP/T plex binds to upregulate Bcl-2 expression, which leads to autophagy inhibitio suggests that additional pathways may play a role in autophagy regulation.
The role of autophagy in cancer has been proposed as a double-edged sw inhibits early-stage tumorigenesis. However, in the later stages, autophagy demonstrated to enhance tumor preservation, growth, and proliferation in a tumor types, including prostate, breast, and ovarian cancers [58][59][60]. S The autophagy pathway has many points of regulation. It is mainly suppressed by mammalian target of rapamycin (mTOR), which inhibits ULK complex activation [15,55]. Additionally, Bcl-2 family proteins inhibit VSP34-BECN1 complex activation [56]. A recent paper demonstrated in a colorectal carcinoma (CRC) model that the YAP/TEAD complex binds to upregulate Bcl-2 expression, which leads to autophagy inhibition [57]. This suggests that additional pathways may play a role in autophagy regulation.
The role of autophagy in cancer has been proposed as a double-edged sword which inhibits early-stage tumorigenesis. However, in the later stages, autophagy has been demonstrated to enhance tumor preservation, growth, and proliferation in a variety of tumor types, including prostate, breast, and ovarian cancers [58][59][60]. Studies on autophagy have developed various targets on different sections of autophagosome formation. For example, beclin1 KD or 3-MA (autophagy inhibitor) treatment inhibits early autophagy activation, while ATG7, ATG5, or ATG12 KD prevents elongation and sealing of the autophagosome. Drugs such as BafA1 and chloroquine (CQ) inhibit later stages of autophagy by preventing fusion of autophagosomes with lysosomes [61]. In fact, many clinical trials are looking at regulation of autophagy as a means to treat cancer (Table 1).

Pro-Survival Autophagy Promotes Stemness Maintenance
In recent years, research has looked at the complex role that autophagy plays in CSC function. Most of the literature suggests that CSCs utilize autophagy as a pro-survival mechanism to maintain a dormant-like stage, which allows for resistance against environmental stresses including hypoxia, chemotherapy, and radiotherapy [2,4,46].
A study on teratocarcinoma cells (cancer of the embryonic stem cells) investigated how alterations in autophagy levels affect CSCs. Inhibition of Nicotinamide phosphoribosyl transferase (NAMPT) activated PTEN, an inhibitor of mTOR, and subsequently induced autophagy activation, evidenced by elevated mRNA and protein levels of autophagy-related genes including ATG5, ATG7, and LC3-II. Subsequent alterations included reduced POU domain transcription factor (POU5F1) expression along with other pluripotency factors such as Nanog and SOX2, resulting in lower proliferation and augmented differentiation of CSCs. Inhibition of autophagy by knocking down ATG12 and ATG7 also reduced stemness and promoted differentiation and/or senescence, suggesting that autophagy supports CSC stemness and blocking autophagy may be a potential target to eliminate teratocarcinoma stem CSCs [62].
Pagotto et al., (2017) explored inhibition of autophagy via ATG5 KO or CQ treatment in CD118+CD44+ ovarian CSCs. The treatment not only impaired cell viability but also reduced spheroid formation, downregulated stem cell-associated markers (e.g., Nanog, SOX2, OCT4), and tumorigenic potential, suggesting that autophagy enhances ovarian CSC stemness [2]. The findings of Li et al., (2017) showed that autophagy inhibition via ATG3 KD, ATG7 KD, or CQ significantly reduced the proportion of hepatic Axin2+CD90+ CSCs. Autophagy inhibition was also found to reduce expression of hepatocyte growth factor (HGF). HGF activates the receptor c-Met to induce JNK/STAT3 signaling, which upregulates CSC self-renewal and tumorigenesis, further corroborating the fact that autophagy plays some role in CSC stemness maintenance [63,64]. The mechanism by which autophagy promotes CSC self-renewal is still under investigation; however, a recent study suggests that p53 may play an important role in this regard. It is well-known that changes in p53 intracellular localization affect autophagy regulation. P53 nuclear localization allows the transcription factor to upregulate expression of autophagy-related genes, while undermining negative regulators of autophagy (i.e., P13K, AKT, and mTOR) [65]. However, p53 cytoplasmic localization inhibits the activity of AMP-dependent kinase (AMPK). Thus, this allows for mTOR activation and prevents autophagy activation [66]. Recently, Wang et al., (2021) showed that autophagy post-transcriptionally regulates p53 levels by removing cytosolic ub-p53 in the lung CSC cell line A549 and augments CSC stemness and spheroid/tumor formation [61].
Mitophagy is a subset of autophagy based on DPR1-driven targeted removal of dysfunctional mitochondria, often marked by increased oxidative stress, DNA damage, and accumulation of p62 protein [67]. Mitophagy works to maintain low ROS levels, which prevents excessive ROS-induced genome damage and consequent induction of mitochondriadependent apoptosis [68,69]. Cell plasticity has been shown to be accompanied with significant changes in mitochondrial composition, function, and maturation, suggesting that mitophagy is crucial for maintenance of stemness [69]. Pten-induced putative kinase 1 (PINK1) has been shown to maintain mitochondrial morphology and function along with mitophagy-dependent mitochondrial degradation [70]. Vazquez-Martin et al., (2016) showed that KD of PINK1 inhibited mitochondrial rejuvenation and prompted spontaneous differentiation of induced pluripotent stem cells (iPSCs). Thus, this suggests that PINK1-dependent mitophagy is required for iPSC to maintain an undifferentiated state [69]. Liu et al., (2017) illustrated that mitophagy regulates tumor suppressor p53 levels in hepatic CSCs. Blockade of autophagy (ATG5 KD, ATG7 KD, and 3-MA) results in increased PINK1-dependent phosphorylation of p53 at serine 392, which induces stabilization and tetramer formation. Phosphorylated p53 is able to translocate to the nucleus to bind to the Nanog promoter and act as a competitive inhibitor against OCT4-SOX2 attachment. Therefore, this suggests that mitophagy removes mitochondria-associated and phosphorylated p53 to allow for Nanog expression and enhances hepatic CSC self-renewal and stemness maintenance [71]. Together, these studies suggest that pro-survival autophagy can help enhance and maintain stemness in various CSC types (Table 2). Table 2. Preclinical data of the effects of pharmaceutical and genetic inhibition of autophagy on cancer stemness.

Mechanism of Action Cell Line Animal Results
Sharif et al.,

Pro-Survival Hypoxia-Induced Autophagy Promotes Metastasis
CSCs constantly modify their TME to create a more suitable niche for tumor growth. One common alteration is the creation of hypoxic areas. The high proliferation rate of tumor cells places great stress on the local vasculature and strips oxygen away from the environment. Thus, hypoxia often induces autophagy as a means to recycle nutrients. One study on CD133+ pancreatic CSCs demonstrated that a hypoxic environment induced greater spheroid formation and enhanced expression of OCT4 and SOX2, which indicates greater self-renewal capacity. Upregulated levels of hypoxia-inducible factor 1a (HIF-1α), the main transcription factor for hypoxia-induced genes [77], resulted in activation of autophagy, which was measured by increased expression of beclin1, ATG5, and LC3-II. In addition, HIF-1α siRNA along with 3-MA treatment resulted in mesenchymal-epithelial transition (MET), where the cells lost their mobile fibroblast phenotype and took on epithelial characteristics. These cells were observed to have increased expression of E-cadherin with downregulation of mesenchymal markers, vimentin and MMP-9. Thus, this study suggests that hypoxia-induced autophagy may play a role in both CSC self-renewal and metastasis [5] (Table 2).

Pro-Survival Autophagy Promotes Treatment Resistance
CSCs have been documented to resist conventional cancer therapies, such as chemotherapy and radiation, and subsequently are able to reestablish tumors after treatment when CSCs are motivated by minute stimuli in the TME that trigger signaling cascades [78]. In recent years, autophagy has been reported as a mechanism that grants resistance to cancer treatment in CSCs.
Radiotherapy (RT) targets tumor cells with photons (e.g., X-ray) and particles (e.g., electrons, protons, and heavy ions) to directly induce irreparable DNA damage by DNA ionization or indirectly through the generation of ROS by interacting with water molecules [72,73]. CSCs have been shown to elevate levels of the antioxidant glutathione (GSH) to protect against ROS-induced oxidative stress [79]. The CSC marker SLC3A2/CD98 heavy chain (CD98hc) associates with cysteine transporter SCL7A11 to synthesize GSH. Digomann et al., (2019) found that amino acid restriction in CD98hc KO HNSCC CSCs induced inhibition of mTOR/PI3K and subsequently activated autophagy. Autophagy inhibition via ATG5 KD or BafA1 increased radio sensitization and induction of apoptosis [72]. Another study on breast CSCs demonstrated increased autophagic vesicles paired with a delayed response to irradiation and reduced initial apoptosis induction compared to normal cancer cells [73]. Both studies suggest that autophagy may play a protective role against radiotherapy for CSCs.
In-vivo murine models have demonstrated that cotreatment with Irinotecan (chemotherapy) and autophagy inhibition, CQ, significantly reduced tumor size compared to any one therapy alone [6]. Irinotecan is hydrolyzed into SN-38, a topoisomerase I inhibitor. By preventing the function of topoisomerase I, this leads to inhibition of DNA replication and transcription, thus preventing tumor cells from replicating [80]. A recent study showed that Taxol (chemotherapy) treatment induced activation of the ERK pathway and increased expression of autophagy-related proteins in radio-resistant bladder CSCs but not in regular cancer cells [74]. Taxol works by stabilizing microtubules and prevents progression of mitosis. In addition, this hinderance induces the cell to undergo apoptosis or transition to the G0 phase [81]. Taxol paired with autophagy inhibition by beclin1 and ATG5 KD resulted in increased apoptosis [74]. Interestingly, Brunel et al., (2021) noted that Temozolomide (chemotherapy) treatment reduced proliferation of beclin1 KD cells glioblastoma stem cells (GSCs) without enhancing apoptosis, which suggests that there may be other pathways besides autophagy that promote resistance to chemotherapy [4]. Temozolomide works as a DNA methylguanine-DNA methyltransferase (MGMT); however, in many tumor cells, these genes are epigenetically silenced. Thus, the hypermethylation cannot be removed, which induces DNA damage and triggers apoptosis [82,83]. Zhu et al., (2021) elucidated that CSCs can resist chemotherapy through SOX2-dependent upregu-lation of ATP-binding cassette (ABC) transporters, which serve as chemotherapy efflux pumps [84,85]. SOX2 synergistically co-localizes with β-catenin to enhance ABC transporter promoter activity but also binds to the beclin1 promoter to induce autophagosome formation. Thus, CSCs are able to effectively pump out chemotherapy drugs before the drugs are able to take effect. Further, autophagy activation by rapamycin augmented chemoresistance, stemness, proliferation, and invasion, while autophagy inhibition with 3-MA diminished the malignant cancer phenotype [6]. Thus, both studies corroborating that autophagy strengthens CSC resistance to chemotherapy.
Targeted treatments are another subset of cancer treatment. A recent paper suggests that in HNSCC CSCs, the tyrosine kinase inhibitor afatinib increased ROS levels, which activated REDD1-TSC1 signaling to inhibit mTOR complex 1 (mTORC1) and subsequently induces autophagy activation. Afatinib coincubation with autophagy inhibitors, 3-MA or CQ, illustrated augmented levels of afatinib-induced apoptosis, suggesting that afatinib can induce pro-survival autophagy in CSCs by reducing sensitivity to apoptosis.This validates the need for further research into autophagy inhibition and conjoint therapies to more effectively eradicate CSCs [75]. Overall, these studies have demonstrated that autophagy enhances resistance toward conventional cancer treatment modalities ( Table 2).

Lethal Autophagy and Cancer Stem Cells
Some studies have indicated that autophagy is involved with suppression of cancer progression as demonstrated in gastric cancers and NSCLC [86,87]. Recent studies have also denoted that autophagy may negatively impact CSCs and cancer growth, hence termed lethal autophagy [15]. Curcumin, an anti-inflammatory spice, exhibited induction of lethal autophagy, followed by reduced glioblastoma SC (GSC) self-renewal and induction of differentiation in mice [76]. Tao et al., (2018) noted that mTOR inhibition (e.g., AZD8055 and rapamycin) suppressed self-renewal ability and tumorigenicity in CD133+Nestin+ GSCs, while 3-MA treatment reversed the mTOR inhibition phenotypes. Additionally, invivo mTOR suppression treatment in mice demonstrated reduced tumor size and prolonged survival through degradation of Notch1, which is involved in cancer cell tumorigenicity, self-renewal, and other various biological functions [50]. Similarly, autophagy inhibition via beclin1 KD, ATG5 KD, or CQ treatment on GSCs enhanced expression of stemness markers (e.g., CD133, POU5F1, SOX2, BMI1, LGR5, and NANOG), along with increased proliferation and clonogenicity. This suggests a departure from a dormant stemness status to an active, proliferating state when autophagy is inhibited [4,50].
Another study reinforces that autophagy inhibits dedifferentiation in a hepatocyte model. Ductular reaction is liver regeneration following chronic liver injury where the hepatocyte regeneration capacity has been impaired. Hepatocytes are able to dedifferentiate into ductular cells, which are liver progenitor cells (LPCs) [88]. Autophagy inhibition (ATG5/ATG7 KO) upregulated expression of YAP and Transcriptional coactivator PDZ-binding motif (TAZ), which are essential for controlling organ size and stemness in hepatocytes. Autophagy-deficient livers illustrated that strong YAP and TAZ co-expression induces ductular cell formation and carcinogenesis of HCC. Additionally, autophagydeficient livers had significant accumulation of cleaved caspase-3 and ductular markers, including SOX9, cytokeratin 19, panCK, suggesting that autophagy prevents apoptosis in hepatocytes and undermines dedifferentiation of hepatocytes into LPCs [51] (Table 2). Together, these studies support the idea that autophagy impedes CSC stemness and proliferation ( Figure 2).
Cancers 2022, 14, x FOR PEER REVIEW Figure 2. Relationship between autophagy with CSCs. Research has demonstrated that au may be separated into two distinct categories in cancer stem cells (CSCs). Protective autop hances CSC characteristics, including stemness maintenance through Pten-induced putativ 1 (PINK1)-dependent inactivation of p53 but also degradation of cytosolic p53. CSC self-re dependent on Jun N-terminal kinase (JNK) and signal transducer and activator of transc (STAT3). Hypoxia-inducible factor-1α (HIF-1α)-dependent autophagy has been shown to CSC stemness while promoting metastasis. CSC resistance to various treatment modalities around upregulation of antioxidants to remove reactive oxygen species (ROS) and ATP cassette (ABC) transporters to efflux anti-cancer drugs. On the other hand, lethal autophagy CSC self-renewal through degradation of Notch1 and reduction of stemness markers. Inhi this form of autophagy has been shown to induce tumorigenesis. (Created with BioRen accessed on 11 October 2021).

Conclusions and Future Directions
Clearly, there is a discrepancy in the literature over the role of autophagy i This review mainly focused on how autophagy strengthens CSC self-renewal, nance of a pluripotent state, and resistance against cancer treatment. However, th growing body of literature that argues that autophagy has the opposite effect a inhibit tumorigenesis. We hypothesize that the type of autophagy differs based on type, tumor stage, TME, and interaction with the apoptosis pathway. Multiple pa are known to regulate autophagy and apoptosis; e.g., PI3K/AKT/mTOR axis inhi tophagy but also phosphorylates caspase-3, caspase-9, and Bad to hinder apopto thermore, autophagy and apoptosis can work synergistically to induce programm death or antagonistically to allow for cell survival [89]. Therefore, the effect of aut on CSC characteristics may depend on which pathways interact with each other.
Given the amount of research on the interaction between autophagy and CSC progression, there is a subset of ongoing clinical trials focused on regulating aut in combination with other cancer treatments. Once again, there is a divide in th ments where some studies focus on autophagy inhibition with CQ/hydroxychlor while others inhibit mTOR with rapamycin, which subsequently allows for activ autophagy. Only one clinical trial specifically focuses on CSCs, while the majority study drug efficacy in cancer in general (Table 1). Thus, there is an urgent need to elucidate the role of autophagy in CSCs. Interestingly, many regulatory factors of agy, e.g., beclin1, p62, and Bcl-2 family proteins, also play a role in apoptosis regu Research has demonstrated that autophagy may be separated into two distinct categories in cancer stem cells (CSCs). Protective autophagy enhances CSC characteristics, including stemness maintenance through Pten-induced putative kinase 1 (PINK1)-dependent inactivation of p53 but also degradation of cytosolic p53. CSC self-renewal is dependent on Jun N-terminal kinase (JNK) and signal transducer and activator of transcription 3 (STAT3). Hypoxia-inducible factor-1α (HIF-1α)-dependent autophagy has been shown to enhance CSC stemness while promoting metastasis. CSC resistance to various treatment modalities revolves around upregulation of antioxidants to remove reactive oxygen species (ROS) and ATP-binding cassette (ABC) transporters to efflux anti-cancer drugs. On the other hand, lethal autophagy inhibits CSC self-renewal through degradation of Notch1 and reduction of stemness markers. Inhibition of this form of autophagy has been shown to induce tumorigenesis. (Created with BioRender.com, accessed on 11 October 2021).

Conclusions and Future Directions
Clearly, there is a discrepancy in the literature over the role of autophagy in CSCs. This review mainly focused on how autophagy strengthens CSC self-renewal, maintenance of a pluripotent state, and resistance against cancer treatment. However, there is a growing body of literature that argues that autophagy has the opposite effect and may inhibit tumorigenesis. We hypothesize that the type of autophagy differs based on the cell type, tumor stage, TME, and interaction with the apoptosis pathway. Multiple pathways are known to regulate autophagy and apoptosis; e.g., PI3K/AKT/mTOR axis inhibits autophagy but also phosphorylates caspase-3, caspase-9, and Bad to hinder apoptosis. Furthermore, autophagy and apoptosis can work synergistically to induce programmed cell death or antagonistically to allow for cell survival [89]. Therefore, the effect of autophagy on CSC characteristics may depend on which pathways interact with each other.
Given the amount of research on the interaction between autophagy and CSC/cancer progression, there is a subset of ongoing clinical trials focused on regulating autophagy in combination with other cancer treatments. Once again, there is a divide in the treatments where some studies focus on autophagy inhibition with CQ/hydroxychloroquine, while others inhibit mTOR with rapamycin, which subsequently allows for activation of autophagy. Only one clinical trial specifically focuses on CSCs, while the majority of trials study drug efficacy in cancer in general (Table 1). Thus, there is an urgent need to further elucidate the role of autophagy in CSCs. Interestingly, many regulatory factors of autophagy, e.g., beclin1, p62, and Bcl-2 family proteins, also play a role in apoptosis regulation.
Autophagy genes have also been observed to have non-autophagy related functions such as LC3-associated phagocytosis (LAP), which is when LC3 is conjugated to phagosome membranes. LAP activation is independent of nutrient deprivation, whereas canonical autophagy requires starvation. LAP has been shown to regulate inflammation and the immune response in response to fungal infection. Moreover, beclin1 has been demonstrated to have non-autophagy functions such as regulation of endocytic receptor trafficking, apoptosis, and inflammation [90]. One study illustrated that beclin1 is able to regulate growth factor and nutrient receptors, thus inhibiting tumor proliferation [91]. Therefore, it is crucial to clarify the role that autophagy plays in CSCs and cancer development.
Instead of knocking out entire autophagy genes, future studies are needed to focus on creating specific autophagy-deficient mutant alleles where non-autophagy functions are preserved, while autophagy is inhibited. This process will clarify the true role autophagy plays in CSCs. Additionally, the interaction between autophagy and apoptosis and how this affects CSCs should be investigated. Together, these studies should elucidate whether autophagy regulation paired with traditional cancer therapies are beneficial for patient outcome. Overall, unraveling autophagy's role in CSCs may offer novel treatments and antineoplastic combination therapies to effectively eradicate the CSC population and improve the success rate of adjuvant therapies.
Author Contributions: Conceptualization: C.X.; information collection, X.W.; original draft preparation, X.W.; writing-review and editing; X.W., C.X. and J.L. All authors have read and agreed to the published version of the manuscript.
Funding: This research was funded by the Physician-Scientist Early Investigator Program at CCR of NIH/NCI to CX (ZIA BC 011888).

Conflicts of Interest:
The authors declare that they have no conflict of interest.