Oncogene-induced autophagy and the Goldilocks principle

Although several oncogenes enhance autophagic flux, the molecular mechanism and consequences of oncogene-induced autophagy remain to be clarified. We have recently shown that expression of oncogenic H-Ras^V12 promotes autophagy through upregulation of Beclin 1 and the BH3-only protein Noxa. H-Ras-expressing cells undergo autophagic cell death as a result of Noxa-mediated displacement of Mcl-1 and Bcl-xL from Beclin 1. Oncogenic H-Ras-induced death is attenuated through knockdown of BECLIN 1, ATG5, or ATG7, or through overexpression of Mcl-1, Bcl-2, Bcl-xL and their close relatives. These observations suggest that high-intensity oncogene activation may be selected against by promoting excessive autophagy, leading to cell death. Consequently, such oncogenes may select for cells with a reduced capacity for autophagy, either through loss of a BECLIN 1 allele or through upregulation of negative regulators of Beclin 1, such as Bcl-2 family members.     

Differential Dependence on Beclin 1 for the Regulation of Pro-Survival Autophagy by Bcl-2 and Bcl-xL in HCT116 Colorectal Cancer Cells

Autophagy is described to be involved in homeostasis, development and disease, both as a survival and a death process. Its involvement in cell death proceeds from interrelationships with the apoptotic pathway. We focused on survival autophagy and investigated its interplays with the apoptotic machinery. We found that while Mcl-1 remained ineffective, Bcl-2 and Bcl-xL were required for starved cells to display a fully functional autophagic pathway as shown by proteolysis activity and detection of autophagic vesicles. Such pro-autophagic functions of Bcl-2 and Bcl-xL were independent of Bax. However they appeared to operate through non redundant mechanisms as Bcl-xL wielded a tighter control than Bcl-2 over the regulation of autophagy: unlike Bcl-2, Bcl-xL and Atg7 manipulation yielded identical phenotypes suggesting they could be components of the same signalling pathway; Bcl-xL subcellular localisation was modified upon starvation, and importantly Bcl-xL acted independently of Beclin 1. Still an intact BH3-binding site was required for Bcl-xL to stimulate a fully functional autophagic pathway. This study highlights that, in addition to their well-established anti-death function during apoptosis, Bcl-2 and Bcl-xL have a broader role in cell survival. Should Bcl-2 and Bcl-xL stand at the cross-roads between pro-survival and pro-death autophagy, this study introduces the new concept that the regulation of autophagy by Bcl-2 and Bcl-xL is adjusted according to its survival or death outcome.     

Oncogenic Ras-induced expression of Noxa and Beclin-1 promotes autophagic cell death and limits clonogenic survival

Deregulated oncogenes such as MYC and RAS are typically insufficient to transform cells on their own due to the activation of pathways that restrain proliferation. Previous studies have shown that oncogenic H-Ras can induce proliferative arrest or senescence, depending on the cellular context. Here, we show that deregulated H-Ras activity can also lead to caspase-independent cell death with features of autophagy. Ras-induced autophagy was associated with upregulation of the BH3-only protein Noxa as well as the autophagy regulator Beclin-1. Silencing of Noxa or Beclin-1 expression reduced Ras-induced autophagy and increased clonogenic survival. Ras-induced cell death was also inhibited by coexpression of Bcl-2 family members that inhibit Beclin-1 function. Ras-induced autophagy was associated with Noxa-mediated displacement of the Bcl-2 family member, Mcl-1, from Beclin-1. Thus, Ras-induced expression of Noxa and Beclin-1 promotes autophagic cell death, which represents a mechanism to limit the oncogenic potential of deregulated Ras signals.     

A natural BH3 mimetic induces autophagy in apoptosis-resistant prostate cancer via modulating Bcl-2–Beclin1 interaction at endoplasmic reticulum

A natural BH3-mimetic, small-molecule inhibitor of Bcl-2, (−)-gossypol, shows promise in ongoing phase II and III clinical trials for human prostate cancer. In this study we show that (−)-gossypol preferentially induces autophagy in androgen-independent (AI) prostate cancer cells that have high levels of Bcl-2 and are resistant to apoptosis, both in vitro and in vivo, but not in androgen-dependent (AD) cells with low Bcl-2 and sensitive to apoptosis. The Bcl-2 inhibitor induces autophagy through blocking Bcl-2–Beclin1 interaction, together with downregulating Bcl-2, upregulating Beclin1, and activating the autophagic pathway. The (−)-gossypol-induced autophagy is dependent on Beclin1 and Atg5. Our results show for the first time that (−)-gossypol can also interrupt the interactions between Beclin1 and Bcl-2/Bcl-xL at endoplasmic reticulum, thus releasing the BH3-only pro-autophagic protein Beclin1, which in turn triggers the autophagic cascade. Oral administration of (−)-gossypol significantly inhibited the growth of AI prostate cancer xenografts, representing a promising new regimen for the treatment of human hormone-refractory prostate cancer with Bcl-2 overexpression. Our data provide new insights into the mode of cell death induced by Bcl-2 inhibitors, which will facilitate the rational design of clinical trials by selecting patients who are most likely to benefit from the Bcl-2-targeted molecular therapy.     

Upregulation of Beclin 1 in the ischemic penumbra

Here we discuss the probable role of autophagy in cerebral ischemia based on our own recent data and the literature. We examined the protein level of Beclin 1 (Bcl-2 interacting protein) and microtubule-associated protein 1 light chain 3 (LC3) which were previously found to promote autophagy. We found a dramatic elevation in Beclin 1 levels and LC3 in the penumbra of rats challenged by cerebral ischemia. We found also that a subpopulation of Beclin 1-upregulating cells is also expressing the active form of caspase-3, and that all Beclin 1 upregulating cells display dense staining of LC3. Neuronal cells that overexpress Beclin 1 may exhibit damaged DNA but without changes in nuclear morphology, which indicates that not all the Beclin 1-upregulating cells are predestined to die. We conclude that the cell death in the penumbra bears a resemblance not only to necrosis, apoptosis, or a compromise between the two, but exhibits also biochemical and morphological characteristics of autophagic cell death. The question that constantly arises, however, is whether autophagic activity in damaged cells is the cause of death or is actually an attempt to prevent it as a part of an endogenous neuroprotective response.

Addendum to: Rami A et al. Focal cerebral ischemia induces upregulation of Beclin 1 and autophagy-like cell death. Neurobiol Dis 2007; In press.     

Autophagy induced by ionizing radiation promotes cell death over survival in human colorectal cancer cells

Autophagy is commonly described as a cell survival mechanism and has been implicated in chemo- and radioresistance of cancer cells. Whether ionizing radiation induced autophagy triggers tumor cell survival or cell death still remains unclear. In this study the autophagy related proteins Beclin1 and ATG7 were tested as potential targets to sensitize colorectal carcinoma cells to ionizing radiation under normoxic, hypoxic and starvation conditions. Colony formation, apoptosis and cell cycle analysis revealed that knockdown of Beclin1 or ATG7 does not enhance radiosensitivity in HCT-116 cells. Furthermore, ATG7 knockdown led to an increased survival fraction under oxygen and glutamine starvation, indicating that ionizing radiation indeed induces autophagy which, however, leads to cell death finally. These results highlight that inhibition of autophagic pathways does not generally increase therapy success but may also lead to an unfavorable outcome especially under amino acid and oxygen restriction.     

Silencing of Bcl-2 expression by small interfering RNA induces autophagic cell death in MCF-7 breast cancer cells

Apoptosis (programmed cell death type I) and autophagy (type II) are crucial mechanisms regulating cell death and homeostasis. The Bcl-2 proto-oncogene is overexpressed in 50-70% of breast cancers, potentially leading to resistance to chemotherapy, radiation and hormone therapy-induced apoptosis. Here, we investigated the role of Bcl-2 in autophagy in breast cancer cells. Silencing of Bcl-2 by siRNA in MCF-7 breast cancer cells downregulated Bcl-2 protein levels (>85%) and led to inhibition of cell growth (71%) colony formation (79%), and cell death (up to 55%) by autophagy but not apoptosis. Induction of autophagy was demonstrated by acridine orange staining, electron microscopy and an accumulation of GFP-LC3-II in autophagosomal membranes in MCF-7 cells transfected with GFP-LC-3(GFP-ATG8). Silencing of Bcl-2 by siRNA also led to induction of LC-3-II, a hallmark of autophagy, ATG5 and Beclin-1 autophagy promoting proteins. Knockdown of ATG5 significantly inhibited Bcl-2 siRNA-induced LC3-II expression, the number of GFP-LC3-II-labeled autophagosome positive cells and autophagic cell death (p < 0.05). Furthermore, doxorubicin at a high dose (IC(95), 1 microM) induced apoptosis but at a low dose (IC(50), 0.07 microM) induced only autophagy and Beclin-1 expression. When combined with Bcl-2 siRNA, doxorubicin (IC(50)) enhanced autophagy as indicated by the increased number cells with GFP-LC3-II-stained autophagosomes (punctuated pattern positive). These results provided the first evidence that targeted silencing of Bcl-2 induces autophagic cell death in MCF-7 breast cancer cells and that Bcl-2 siRNA may be used as a therapeutic strategy alone or in combination with chemotherapy in breast cancer cells that overexpress Bcl-2.     

Bcl-2 and Bcl-xL play important roles in the crosstalk between autophagy and apoptosis

Autophagy and apoptosis play important roles in the development, cellular homeostasis and, especially, oncogenesis of mammals. They may be triggered by common upstream signals, resulting in combined autophagy and apoptosis. In other instances, they may be mutually exclusive. Recent studies have suggested possible molecular mechanisms for crosstalk between autophagy and apoptosis. Bcl-2 and Bcl-xL, the well-characterized apoptosis guards, appear to be important factors in autophagy, inhibiting Beclin 1-mediated autophagy by binding to Beclin 1. In addition, Beclin 1, Bcl-2 and Bcl-xL can cooperate with Atg5 or Ca(2+) to regulate both autophagy and apoptosis. Thus, Bcl-2 and Bcl-xL represent a molecular link between autophagy and apoptosis. Here, we discuss the possible roles of Bcl-2 and Bcl-xL in apoptosis and autophagy, and the crosstalk between them.     

Beclin 1-independent autophagy induced by a Bcl-XL/Bcl-2 targeting compound,Z18

Inhibitors of Bcl-X_L/Bcl-2 can induce autophagy by releasing the autophagic protein Beclin 1 from its complexes with these proteins. Here we report a novel compound targeting the BH3 binding groove of Bcl-X_L/Bcl-2, Z18, which efficiently induces autophagy-associated cell death in HeLa cells, without apparent apoptosis. Unexpectedly, the inhibition of Beclin 1 and phosphatidylinositol 3-kinase have no obvious effect on Z18-induced autophagy in HeLa cells, implying that it is a non-canonical Beclin 1-independent autophagy. Meanwhile, the accumulation of autophagosomes is positively correlated with Z18-induced cell death and the full flux of autophagy is not necessary.     

Differential interactions between Beclin 1 and Bcl-2 family members

Autophagy, a cellular degradation system, promotes both cell death and survival. The interaction between Bcl-2 family proteins and Beclin 1, a Bcl-2 interacting protein that promotes autophagy, can mediate crosstalk between autophagy and apoptosis. We investigated the interaction between anti-and pro-apoptotic Bcl-2 proteins with Beclin 1. Our results show that Beclin 1 directly interacts with Bcl-2, Bcl-x(L), Bcl-w and to a lesser extent with Mcl-1. Beclin 1 does not bind the pro-apoptotic Bcl-2 proteins. The interaction between Beclin 1 and the anti-apoptotic protein Bcl-x(L) was inhibited by BH3-only proteins, but not by multi-domain proteins. Sequence alignment and structural modeling suggest that Beclin 1 contains a putative BH3-like domain which may interact with the hydrophobic grove of Bcl-x(L). Mutation of the Beclin 1 amino acids predicted to mediate this interaction inhibited the association of Beclin 1 with Bcl-x(L). Our results suggest that BH3 only proapoptotic Bcl-2 proteins may modulate the interactions between Bcl-x(L) and Beclin 1.     

Upregulation of Beclin 1 in the ischemic penumbra

Here we discuss the probable role of autophagy in cerebral ischemia based on our own recent data and the literature. We examined the protein level of Beclin 1 (Bcl-2 interacting protein) and microtubule-associated protein 1 light chain 3 (LC3) which were previously found to promote autophagy. We found a dramatic elevation in Beclin 1 levels and LC3 in the penumbra of rats challenged by cerebral ischemia. We found also that a subpopulation of Beclin 1-upregulating cells is also expressing the active form of caspase-3, and that all Beclin 1 upregulating cells display dense staining of LC3. Neuronal cells that overexpress Beclin 1 may exhibit damaged DNA but without changes in nuclear morphology, which indicates that not all the Beclin 1-upregulating cells are predestined to die. We conclude that the cell death in the penumbra bears a resemblance not only to necrosis, apoptosis, or a compromise between the two, but exhibits also biochemical and morphological characteristics of autophagic cell death. The question that constantly arises, however, is whether autophagic activity in damaged cells is the cause of death or is actually an attempt to prevent it as a part of an endogenous neuroprotective response.     

The Bcl-2-Beclin 1 interaction in (-)-gossypol-induced autophagy versus apoptosis in prostate cancer cells

Bcl-2 is a key dual regulator of autophagy and apoptosis, but how the level of Bcl-2 influences the cellular decision between autophagy and apoptosis is unclear. The natural BH3-mimetic (-)-gossypol preferentially induces autophagy in androgen-independent (AI) prostate cancer cells that have high levels of Bcl-2 and are resistant to apoptosis, whereas apoptosis is preferentially induced in androgen-dependent or -independent cells with low Bcl-2. (-)-Gossypol induces autophagy via blocking Bcl-2-Beclin 1 interaction at the endoplasmic reticulum (ER), together with downregulating Bcl-2, upregulating Beclin 1 and activating the autophagic pathway. Furthermore, (-)-gossypol-induced autophagy is Beclin 1- and Atg5-dependent. These results provide new insights into the mode of cell death induced by Bcl-2 inhibitors, which could facilitate the rational design of clinical trials by selecting patients who are most likely to benefit from the Bcl-2-targeted molecular therapy.     

The autophagic response to nutrient deprivation in the hl-1 cardiac myocyte is modulated by Bcl-2 and sarco/endoplasmic reticulum calcium stores

Macroautophagy is a vital process in the cardiac myocyte: it plays a protective role in the response to ischemic injury, and chronic perturbation is causative in heart disease. Recent findings evidence a link between the apoptotic and autophagic pathways through the interaction of the antiapoptotic proteins Bcl-2 and Bcl-XL with Beclin 1. However, the nature of the interaction, either in promoting or blocking autophagy, remains unclear. Here, using a highly sensitive, macroautophagy-specific flux assay allowing for the distinction between enhanced autophagosome production and suppressed autophagosome degradation, we investigated the control of Beclin 1 and Bcl-2 on nutrient deprivation-activated macroautophagy. We found that in HL-1 cardiac myocytes the relationship between Beclin 1 and Bcl-2 is subtle: Beclin 1 mutant lacking the Bcl-2-binding domain significantly reduced autophagic activity, indicating that Beclin 1-mediated autophagy required an interaction with Bcl-2. Overexpression of Bcl-2 had no effect on the autophagic response to nutrient deprivation; however, targeting Bcl-2 to the sarco/endoplasmic reticulum (S/ER) significantly suppressed autophagy. The suppressive effect of S/ER-targeted Bcl-2 was in part due to the depletion of S/ER calcium stores. Intracellular scavenging of calcium by BAPTA-AM significantly blocked autophagy, and thapsigargin, an inhibitor of sarco/endoplasmic reticulum calcium ATPase, reduced autophagic activity by approximately 50%. In cells expressing Bcl-2-ER, thapsigargin maximally reduced autophagic flux. Thus, our results demonstrate that Bcl-2 negatively regulated the autophagic response at the level of S/ER calcium content rather than via direct interaction with Beclin 1. Moreover, we identify calcium homeostasis as an essential component of the autophagic response to nutrient deprivation.     

Neurodegeneration induces upregulation of Beclin 1

Autophagy, a bulk degradation of subcellular constituents, is activated in normal cell growth and development, and represents the major pathway by which the cell maintains a balance between protein synthesis and protein degradation. Autophagy was documented in several neurodegenerative diseases, and under stress conditions the autophagic process can lead to cell death (type II programmed cell death). Beclin 1 is a Bcl-2 interacting protein that was previously found to promote autophagy. We have used Beclin 1 protein as a marker for autophagy following traumatic brain injury in mice. We demonstrated a dramatic elevation in Beclin 1 levels near the injury site. Interestingly Beclin 1 elevation starts at early stages post injury (4 h) in neurons and 3 days later in astrocytes. In both cell types it lasts for at least three weeks. Neuronal cells, but not astrocytes, that overexpress Beclin 1 may exhibit damaged DNA but without changes in nuclear morphology. These observations may indicate that not all the Beclin1 overexpressing cells will die. The elevation of Beclin 1 at the site of injury may represent enhanced autophagy as a mechanism to discard injured cells and reduce damage to cells by disposing of injured components.     

Autophagy is induced by ischemic preconditioning in human livers formerly treated by chemotherapy to limit necrosis

The effectiveness of ischemic preconditioning (IP) against hepatic ischemia/reperfusion injury during human liver surgery is linked to decreased apoptotic cell death as well as preservation of the ATP content in liver tissue. Overproduction of Bcl-2 is reported in preconditioned organs. In human liver biopsies exhibiting steatosis and/or vascular injuries (mainly peliosis) induced by chemotherapy, we find that the expression of Bcl-2 in centrolobular and peliotic areas colocalizes with the autophagy protein Beclin 1 in IP livers. Increased expression of phosphorylated Bcl-2 in preconditioned livers is associated with a decreased immunoprecipitation of Beclin 1 and increased expression of LC3-II. The increased number of autophagic vacuoles seen by electron microscopy confirmed that IP could trigger autophagy in chemotherapy-injured livers, probably to reduce the pro-inflammatory necrotic cell death of hepatocytes or endothelial cells and to increase ATP levels. Indeed, necrosis is less frequent (p = 0.04) in IP livers than in the others although no change in apoptosis as assessed by TUNEL assay or caspase-3, -8 and -9 expressions is observed. In conclusion, Bcl-2 and Beclin 1 could be major targets in the regulation of cell death during ischemia/reperfusion injury modulating autophagy to switch on/off necrosis and/or apoptosis.     

Role of Bcl-2 family proteins in a non-apoptotic programmed cell death dependent on autophagy genes

Programmed cell death can be divided into several categories including type I (apoptosis) and type II (autophagic death). The Bcl-2 family of proteins are well-characterized regulators of apoptosis, and the multidomain pro-apoptotic members of this family, such as Bax and Bak, act as a mitochondrial gateway where a variety of apoptotic signals converge. Although embryonic fibroblasts from Bax/Bak double knockout mice are resistant to apoptosis, we found that these cells still underwent a non-apoptotic death after death stimulation. Electron microscopic and biochemical studies revealed that double knockout cell death was associated with autophagosomes/autolysosomes. This non-apoptotic death of double knockout cells was suppressed by inhibitors of autophagy, including 3-methyl adenine, was dependent on autophagic proteins APG5 and Beclin 1 (capable of binding to Bcl-2/Bcl-x(L)), and was also modulated by Bcl-x(L). These results indicate that the Bcl-2 family of proteins not only regulates apoptosis, but also controls non-apoptotic programmed cell death that depends on the autophagy genes.     

HSpin1, a transmembrane protein interacting with Bcl-2/Bcl-xL,induces a caspase-independent autophagic cell death

The Drosophila spinster (spin) gene product is required for programmed cell death in the nervous and reproductive systems. We have identified a human homologue of the Drosophila spin gene product (HSpin1). HSpin1 bound to Bcl-2 and apoptosis regulator Bcl-X (Bcl-xL), but not to proapoptotic members such as Bcl-2-associated X protein and Bcl-2 homologous antagonist killer, in cells treated with TNF-alpha. Exogenous expression of HSpin1 resulted in the cell death without inducing a release of cytochrome c from mitochondria. Overexpression of Bcl-xL inhibited the HSpin1-induced cell death. Interestingly, a necrosis inhibitor, pyrrolidine dithiocarbomate, but not the pancaspase inhibitors, carbobenzoxy-VAD-fluoromethyl ketone and p35, blocked the HSpin1-induced cell death. HSpin1-induced cell death increases autophagic vacuole and mature form of cathepsin D, suggesting a novel caspase-independent cell death, which is link to autophagy.     

Differential Interactions Between Beclin 1 and Bcl-2 Family Members

Autophagy, a cellular degradation system, promotes both cell death and survival. The interaction between Bcl-2 family proteins and Beclin 1, a Bcl-2 interacting protein that promotes autophagy, can mediate crosstalk between autophagy and apoptosis. We investigated the interaction between anti-and pro-apoptotic Bcl-2 proteins with Beclin 1. Our results show that Beclin 1 directly interacts with Bcl-2, Bcl-x_L, Bcl-w and to a lesser extent with Mcl-1. Beclin 1 does not bind the pro-apoptotic Bcl-2 proteins. The interaction between Beclin 1 and the anti-apoptotic protein Bcl-x_L was inhibited by BH3-only proteins, but not by multi-domain proteins. Sequence alignment and structural modeling suggest that Beclin 1 contains a putative BH3-like domain which may interact with the hydrophobic grove of Bcl-x_L. Mutation of the Beclin 1 amino acids predicted to mediate this interaction inhibited the association of Beclin 1 with Bcl-x_L. Our results suggest that BH3 only proapoptotic Bcl-2 proteins may modulate the interactions between Bcl-x_L and Beclin 1.     

The end of autophagic cell death?

In the mammalian system, cell death is often preceded or accompanied by autophagic vacuolization, a finding that initially led to the widespread belief that so-called "autophagic cell death" would be mediated by autophagy. Thanks to the availability of genetic tools to disable the autophagic machinery, it has become clear over recent years that autophagy usually constitutes a futile attempt of dying cells to adapt to lethal stress rather than a mechanism to execute a cell death program. Recently, we systematically addressed the question as to whether established or prospective anticancer agents may induce "autophagic cell death". Although a considerable portion among the 1,400 compounds that we evaluated induced autophagic puncta and actually increased autophagic flux, not a single one turned out to kill tumor cells through the induction of autophagy. Thus, knockdown of essential autophagy genes (such as ATG5 and ATG7) failed to prevent and rather accelerated chemotherapy-induced cell death, in spite of the fact that this manipulation efficiently inhibits autophagosome formation. Herein, we review these finding and--polemically--raise doubts as to the very existence of "autophagic cell death".     

Hypoxia induces autophagic cell death in apoptosis-competent cells through a mechanism involving BNIP3

Hypoxia (lack of oxygen) is a physiological stress often associated with solid tumors. Hypoxia correlates with poor prognosis since hypoxic regions within tumors are considered apoptosisresistant. Autophagy (cellular "self digestion") has been associated with hypoxia during cardiac ischemia and metabolic stress as a survival mechanism. However, although autophagy is best characterized as a survival response, it can also function as a mechanism of programmed cell death. Our results show that autophagic cell death is induced by hypoxia in cancer cells with intact apoptotic machinery. We have analyzed two glioma cell lines (U87, U373), two breast cancer cell lines (MDA-MB-231, ZR75) and one embryonic cell line (HEK293) for cell death response in hypoxia (<1% O(2)). Under normoxic conditions, all five cell lines undergo etoposide-induced apoptosis whereas hypoxia fails to induce these apoptotic responses. All five cell lines induce an autophagic response and undergo cell death in hypoxia. Hypoxia-induced cell death was reduced upon treatment with the autophagy inhibitor 3-methyladenine, but not with the caspase inhibitor z-VAD-fmk. By knocking down the autophagy proteins Beclin-1 or ATG5, hypoxia-induced cell death was also reduced. The pro-cell death Bcl-2 family member BNIP3 (Bcl-2/adenovirus E1B 19kDainteracting protein 3) is upregulated during hypoxia and is known to induce autophagy and cell death. We found that BNIP3 overexpression induced autophagy, while expression of BNIP3 siRNA or a dominant-negative form of BNIP3 reduced hypoxia-induced autophagy. Taken together, these results suggest that prolonged hypoxia induces autophagic cell death in apoptosis-competent cells, through a mechanism involving BNIP3.