APAP- Acetaminophen; ALD- Alcoholic Liver Disease; ALT- Alanine Aminotransferase; CQ- Chloroquine; CMA- Chaperone-Mediated Autophagy; EFV- Efavirenz; CYP2E1- Cytochrome P450 2E1; GSH- Glutathione; KEAP1- Kelch-like ECHAssociated Protein 1; NAPQI- N-Acetyl-P-Benzoquinone Imine; NFE2L2/NRF2- Nuclear Factor Erythroid-Derived 2-Like 2; RIPReceptor Interacting Protein; SQSTM1- Sequestosome 1; TNF-a- Tumor Necrosis Factor-a

In toxicology- it has been traditionally thought that environmental toxicants and xenobiotics trigger tissue and cell injury by causing necrotic cell death- which is a sudden and unregulated process. This notion has been challenged in the past decade due to the expanding understanding of the mechanisms of apoptosis- a programmed cell death. Indeed, it has been demonstrated that a variety of environmental contaminants including dioxin, heavy metals (cadmium and methylmercury), organotin compounds and dithiocarbamates induce apoptosis in various in vitro models [1]. Moreover, recent evidence indicates that necrotic cell death, which has been long thought to be unregulated- can also be a well-regulated programmed process. When exposed to tumor necrosis factor-α (TNF-α) together with actinomycin D (ActD) or smac mimetics to block nuclear factor κB activation, cells are killed by apoptosis through activation of the death-receptor activation apoptotic pathway and caspases [2-4]. Intriguingly- when TNF-α/ActD-induced apoptosis is completely blocked by a general caspase inhibitor (Z-VAD-fmk), the cells are killed by alternative necrotic cell death [2-5]. This programmed necrotic cell death requires activation of the necrosome which includes the receptor interacting protein (RIP) kinases, Rip1 and Rip3, mixed lineage kinase domainlike (MLKL) and the mitochondrial phosphatase PGAM5 [5,6]. In addition to TNF-α-induced programmed necrosis in vitro, it has also been demonstrated that RIP3-mediated necrosis plays an important role in defending against viral infection-induced inflammation in mice- suggesting a physiological relevance of programmed necrosis [7].

Although environmental toxicant and xenobiotic induced cell death has been extensively studied, little attention has been paid to the role of cellular protective mechanisms in environmental toxicant and xenobiotic-induced cell death and tissue injury.

There are three types of autophagy in mammalian cells which include macroautophagy (here after referred to as autophagy), chaperone-mediated autophagy (CMA), and microautophagy. The three different modes of autophagy differ in how the target cargos are delivered to lysosomes [8]. Autophagy involves the formation of a double-membrane autophagosome which wraps cellular proteins and organelles. The autophagosome fuses with a lysosome to form an autolysosome and then uses lysosomal enzymes to degrade the cellular contents sequestered in the autophagosome. Microautophagy refers to a process where a lysosome directly uptakes cytoplasm or organelles at the lysosomal surface by protrusion of the lysosomal membrane. CMA selectively degrades a subset of proteins that have the specific motif KFERQ which can be transported through the lysosome membrane associated protein 2 (LAMP-2) at the lysosomal membrane [8-11].

After it was first described nearly 40 years ago by de Duve and Wattiaux, autophagy is now generally thought to be a genetically programmed and evolutionarily conserved catabolic process that serves as a cell survival mechanism in response to stress. Liverspecific deletion of either Atg5 or Atg7, two essential autophagy genes leads to severe liver injury [12,13]. These findings suggest that basal autophagy is an important cellular protective mechanism by regulating cellular homeostasis. Moreover these mice also eventually develop spontaneous liver tumors- supporting the notion that autophagy is a tumor suppressor [14,15].

In addition autophagy also plays role in development [16], clearance of misfolded protein aggregates, and defense against microbial infections [17]. Furthermore emerging evidence indicates that autophagy is also important for protecting cells against toxicant/ xenobiotic-induced toxicity.

Arsenic and cadmium are two widely distributed toxic metals that result in a variety of human diseases including cancer. Chronic exposure of arsenic induces liver injury, neurological disorders, and cancer. Cadmium induces genotoxicity in lung, kidney and prostate. Both arsenic and cadmium bind to sulfhydryl groups of many cellular proteins and trigger oxidative stress which may contribute to their toxicity.

Cadmium has been shown to induce autophagic changes in various cultured cells including skin epidermal cells [18], W318 human lung epithelial fibroblast cells [19], vascular endothelial cells [20] and MES- 13 mesangial cells [21]. Several mechanisms have been implicated in cadmium-induced autophagy including activation of both p38 and glycogen synthase kinase-3beta (GSK-3beta) and ROS-mediated downregulation of mTOR. However the role of autophagy in cadmiuminduced cell death is less clear. While some reports suggest autophagy may contribute to cadmium-induced cell death [21,22], one report suggests that autophagy may protect against cell death by relieving endoplasmic reticulum (ER) stress [19].

Similar to cadmium, arsenic has also been reported to induce autophagy in human leukemia cells [23,24], human bronchial epithelial cells [25], and human malignant glioma cell lines [26]. Mechanistically arsenic-induced autophagy may require activation of the MEK/ERK pathway [19], ROS generation [25] and induction of Bnip3 protein [26]. Arsenic has been used to treat tumor cells due to its ability to induce apoptosis. The induction of autophagy in arsenic-treated tumor cells has also been suggested to promote cell death, a process called “autophagic cell death” [24-26]. However one recent report suggests that autophagy can instead remove arsenic-induced damaged mitochondria to serve as a protective mechanism against arsenic induced bronchial epithelial cell transformation [26]. The controversial findings in regard to whether autophagy is protective or detrimental to either cadmium or arsenic-induced toxicity could be due to the lack of accurate assays for autophagy in these studies (see below discussion).

Alcohol is the most consumed beverage worldwide and excessive or chronic consumption of alcohol can lead to Alcoholic Liver Disease (ALD). ALD has a wide range of pathogenic features ranging from early steatosis to more severe acute alcoholic hepatitis, fibrosis, cirrhosis and even hepatocellular carcinoma [27]. It has long been known that only a small portion of heavy drinkers eventually develop fibrosis and cirrhosis. Our lab and others have recently found that acute ethanol treatment can modulate the autophagy process in cultured cells and animal models [28-30]. Autophagy is activated and serves as a compensatory mechanism to remove ethanol-induced damaged mitochondria and excessive lipid droplets but not protein aggregates. Subsequent studies reveal that ethanol-induced autophagic changes require its metabolism which is mainly mediated by cytochrome P450 2E1 (CYP2E1) and alcohol dehydrogenase (ADH) [28-32]. More importantly- pharmacological activation of autophagy significantly inhibits ethanol-induced steatosis and liver injury [28-30].

In addition to improvement of ethanol-induced liver steatosis and injury, induction of autophagy can also protect against druginduced hepatotoxicity. Since mitochondrion is a common target for various drug-induced hepatotoxicities and damaged mitochondria are removed by autophagy (a process referred to as mitophagy), it is not surprising that induction of autophagy would be a promising approach to attenuate drug-induced liver injury.

Acetaminophen (APAP) is a widely used antipyretic and analgesic drug and overdose of APAP can cause severe liver injury in animals and in humans [33]. APAP is also metabolized by CYP2E1 in the liver which generates the reactive metabolite N-acetyl-p-benzoquinone imine (NAPQI). NAPQI depletes hepatic glutathione (GSH) to induce hepatocellular oxidative stress [34]. We recently found that pharmacological induction of autophagy by rapamycin reduces APAPinduced necrosis significantly [35]. More importantly administration of rapamycin 2 hours following APAP treatment still significantly suppressed APAP-induced liver injury [35] and APAP is completely metabolized at this two hour time point. Therefore pharmacologic induction of autophagy may have potential therapeutic applications in human APAP-overdose patients because most APAP-patients enter the emergency room already passed the metabolism phase. In addition, Efavirenz (EFV), a widely used non-nucleoside reverse transcriptase inhibitor for HIV induces mitochondria-mediated cell death in hepatocytes. Pharmacological suppression of autophagy enhances EFV-induced cell death due to impaired removal of damaged mitochondria [36].

How does mitophagy protect against cell death? Although the mechanisms by which mitophagy protects against cell death are not clear, we found that APAP-induced reactive oxygen species (ROS) are suppressed by induction of autophagy but exacerbated by suppression of autophagy [35]. Thus one possibility for protection against cell death by mitophagy may be due to attenuation of mitochondria-derived ROS formation and release of pro-cell death factors from mitochondria. More future studies are needed to determine if induction of autophagy would protect against other drug-induced toxicities that are mediated by damaged mitochondria.

Induction of oxidative stress by xenobiotics, drugs, heavy metals and ionizing radiation has been known to have profound effects on cell survival, growth, development and cell death. NF-E2-related factor 2 (NFE2L2/Nrf2), a nuclear transcription factor is a cellular sensor of chemical and radiation induced oxidative and electrophilic stress. Nrf2 regulates the expression of many cytoprotective genes such as NAD (P) H dehydrogenase quinone 1 (NQO1), glutamate-cysteine ligase catalytic subunit (GCLC) and glutamate-cysteine ligase modifier subunit (GCLM) which are critical for protection against electrophilic and oxidative stress. Nrf2 is also an important protective mechanism against xenobiotic-induced DNA damage and carcinogenesis [37]. Under basal conditions, Nrf2 is normally inhibited by Kelch-like ECHassociated protein 1 (Keap1). When cells are exposed to oxidative stress, electrophiles or xenobiotics, Nrf2 dissociates from Keap1 and moves to the nucleus to activate antioxidant response element (ARE)- dependent gene expression to maintain cellular redox homeostasis. Interestingly it was recently found that Nrf2 is persistently activated in autophagy-deficient mouse livers (liver-specific Atg5 or Atg7 knockout mice) due to the accumulation of an autophagy substrate protein p62/ Sqstm1 [12,13]. p62/Sqstm1 interacts with Keap1 which releases its inhibitory effects on Nrf2 and causes persistent Nrf2 activation [38]. We found that there is a significant increase in expression of the Nrf2- target genes Nqo1- Gclc and Gclm in mouse livers with the loss of Atg5. As a result the hepatic glutathione (GSH) levels were almost two times greater in the Atg5-knockout mouse livers compared to their wild type litter mates. Paradoxically liver-specific Atg7 and Atg5 knockout mice have severe liver injury. Deletion of p62 partially reduced, but deletion of Nrf2 completely suppressed liver injury in the liver-specific Atg7 and Atg5 knockout mice. These results indicate that the constitutive activation of Nrf2 is the dominant mechanism for the liver injury induced by impaired autophagy. However, how constitutive activation of Nrf2 causes liver injury is not clear at present. Moreover Atg5 and Atg7 liver-specific knockout mice all eventually develop spontaneous liver tumors and deletion of p62 can partially reduce this tumor incidence [15]. Activation of Nrf2 has been found in liver tumor cells but not in adjacent normal tissues [14]. Moreover Nrf2 activation which increases expression of antioxidant genes has been shown to detoxify reactive oxygen species to promote cell survival and oncogenedriven tumorigenesis [39]. Together these findings indicate that Nrf2 has a dual role in tissue injury and cancer. It seems that a balanced level of Nrf2 is critical for regulating the homeostasis of cellular redox; too much or too little Nrf2 may both lead to detrimental effects on animals and humans. More importantly it appears that autophagy can play a role in the regulation of Nrf2 activation in addition to traditional oxidants and electrophilic reagents.

It is well known that cancer cells use autophagy as a cell survival mechanism in response to a variety of stresses including hypoxia, growth factor deprivation, starvation and endoplasmic reticulum (ER) stress as well as proteasome inhibition [40-43]. Moreover genetic deletion of Atg5 or Atg7 from mouse livers leads to increased cell death and liver injury suggesting that even basal autophagy is an important cell survival pathway for normal liver cells [12,13]. Together these evidence strongly support the notion that autophagy is a cellular protective mechanism. In contrast, it has also been hotly debated that autophagy could be a cell death mechanism, the so-called “autophagic cell death” mechanism. However, it should be noted that many early studies only employed morphological approaches and criteria for autophagic cell death- and these may not be appropriate for defining autophagic cell death because co-existence of autophagy with cell death does not guarantee that autophagy contributes to cell death. In this case autophagy could either be detrimental, protective or just a by-stander.

However with better understanding of the autophagy machinery and molecular pathways, it is now relatively easier to address this issue. Autophagy has been found to contribute to cell death when cells are exposed to certain chemotherapeutic drugs [26-44], radiation [45], hypoxia [46] and cytokines such as INF-γ [47]. In all of these cases, siRNA knockdown or genetic deletion of key autophagy genes suppressed cell death while overexpression of these genes promoted cell death. How exactly autophagy induces cell death is not clear, although it is generally thought that excessive autophagy may nonselectively degrade essential cell components which could lead to cell death. Moreover whether there is a real “autophagic cell death” is also debatable because the presence of autophagy may actually just promote either apoptosis or necrosis [48]. More work is definitely needed to further clarify why certain cell types behave differently in response to the modulation of autophagy. However it should be noted that in most cases inhibition of autophagy can further promote cell death.

Although toxicologists have to often deal with cell death by necrosis and apoptosis, emerging evidence suggests that cell survival mechanisms such as autophagy can also be induced when various environmental contaminants and xenobiotics exert their toxicity. As autophagy is an important anti-tumorigenesis mechanism, it will be interesting to see what the association of autophagy is in many environmental carcinogen-induced tumorigeneses such as carcinogenesis induced by cadmium and arsenic. The rapidly expanding understanding of basic autophagy pathways and the availability of novel technology and animal models for autophagy study will definitely help us to better understand the role of autophagy in toxicology.

The research work in W.X Ding’s lab was supported in part by the NIAAA funds R01 AA020518-01- R21 AA017421 National Center for Research Resources (5P20RR021940-07) and P20 RR016475 from the IDeA Networks of Biomedical Research Excellence (INBRE) program of the National Center for Research Resources. The author thanks J. A. Williams for her critical reading of this manuscript.