Targeting leucine addiction and autophagy in melanoma

Melanoma is an increasingly prevalent cancer that is known for being aggressive. At advanced stages, the prognosis is poor and patient survival rates are low. Melanoma quickly progresses to metastasis and becomes highly resistant to current therapies. Somatic mutations in B-RAF and RAS, with BRAFV600E being the most common one, have been found with high incidence in malignant melanoma, which results in constitutive activation of the MAPK signaling pathway. Consequently, unraveling the molecular mechanisms controlled by hyperactivation of the RAS/RAF/MAPK pathway is critical to gaining a better understanding of how melanoma tumors grow and progress and to identify new therapeutic strategies. Initial efforts in this direction have suggested that the targeted inhibition of components of this cascade is a promising approach. That is, RAF inhibitors display remarkable clinical activity in melanoma tumors with BRAFV600E mutations. However, resistance to these drugs invariably arises resulting in reduced clinical responses. Interestingly, there is mounting evidence that these oncogenic mutations could influence nutrient-sensing and metabolic pathways that modulate the survival of cancer cells (Mathew and White, 2011). In this regard, autophagy has emerged as a central player in this process and, therefore, as a potential cancer target. Autophagy is a lysosome-dependent mechanism for the degradation of cytoplasmic proteins, damaged organelles and aggregates that can be induced by nutrient deprivation (Mathew and White, 2011). There are data suggesting an important role for autophagy as a common mechanism underlying resistance to metabolic stress and cancer therapy.

A recent study from the Sabatini laboratory suggests a novel intervention to promote melanoma cell death by taking advantage of a metabolic disadvantage of these cells in terms of the activation of autophagy during nutrient stress. The authors focused on the response of melanoma cells, with regard to survival versus death, when starved of different essential amino acids. Sabatini and co-workers found that melanoma cells with hyperactivated RAS–MEK signaling were sensitive to leucine deprivation, which induced a mitochondrial apoptotic cascade triggered by a failure to activate autophagy (Fig. 1). This effect was selective for melanoma cells and did not occur in normal cells, which offers an interesting therapeutic opportunity. These authors went on to show, in in vivo melanoma xenografts, that dietary leucine deprivation combined with chloroquine, an autophagy inhibitor, had a synergistic effect on tumor size reduction concomitant with the induction of melanoma cell death. A critical question, from a mechanistic point of view, is how oncogenic activation of the RAS–MEK cascade regulates the melanoma cell’s dependence on leucine for survival. The authors propose that this effect is mediated through mTOR complex 1 (mTORC1), an amino acid sensor and a master-negative regulator of autophagy (Fig. 1). It is now very well established that nutrient starvation inhibits the activation of autophagy by mTORC1. However, in melanoma cells with activated RAS–MEK signaling, leucine deprivation is not sufficient to inhibit mTORC1 and therefore does not trigger the autophagic survival mechanism. This is a very important observation that should be considered in the context of another recent report demonstrating a MAPK-dependent mechanism for the regulation of autophagy independent of mTOR, through the control of the transcription factor EB (TFEB), a master gene for lysosomal biogenesis and autophagy (Settembre et al., 2011). According to these data, MAPK phosphorylates TFEB and blocks its nuclear translocation in response to nutrient starvation, which results in the reduction of its transcriptional activity and the consequent stimulation of the autophagic/lysosomal program (Settembre et al., 2011). It would be interesting to determine whether leucine starvation also disrupts this MAPK–TFEB mechanism in melanoma as a way to keep autophagy at bay.

Figure 1.

Melanoma cells with mutations in the RAS/MAPK signaling pathway are sensitive to combine therapy of leucine deprivation and inhibition of autophagy. In normal cells, the mTORC1 complex is a sensor of nutrient levels, and nutrient starvation induces its inhibition-activating autophagy, which in turn replenishes nutrient levels. In melanoma cells, however, the hyperactivation of the RAS/MAPK pathway cells keeps mTORC1 active and therefore autophagy is not turned on, which induces apoptosis.

To further explore how leucine deprivation fails to inhibit mTORC1, Sabatini’s group examined the spatial localization of mTOR. This group has previously presented convincing evidence linking mTORC1 activation to lysosomal targeting in response to amino acids. The Rag GTPases and the Ragulator complex have been shown to be critical mediators in this activation event via lysosomal recruitment. Now, Sheen and co-workers suggest that, in melanoma cells, hyperactivation of the RAS–MEK pathway keeps mTORC1 constitutively associated with the lysosome rendering it insensitive to leucine deprivation. Consistent with this notion, it has been suggested that lysosomal positioning coordinates nutrient responses and influences both mTOR activation and autophagy. The exact mechanism whereby the oncogenic RAS–MEK cascade makes the mTOR complex insensitive to leucine deprivation and whether or not this pathway is connected with the Rag GTPase–Ragulator system that links nutrient starvation to mTOR activation and autophagy are questions that remain unanswered. Another important unresolved question is the identity of the signals that discriminates between leucine deprivation and that of the other branched amino acids that also activate mTORC1. These are important pieces of the puzzle that need to be filled in to further advance our understanding of this complex process. It would also be interesting to determine whether manipulating leucine levels could be a general paradigm for therapy in other types of cancer in which the RAS–MEK pathway is activated such as lung, pancreatic, colon and breast cancer or if this strategy is specific to melanoma. In addition, based on evidence that connects the RAS–MEK pathway with the LKB– AMPK cassette, it could be worth exploring how these cells handle energy stress or glucose deprivation as an alternative pathway to control cell death and autophagy (Zheng et al., 2009).

The potential importance of this new study lies in the suggestion that the combined inhibition of autophagy and leucine levels could be an effective novel strategy to treat melanoma tumors. The authors propose that inhibition of basal autophagy, which contributes to the recycling of essential amino acids, is required to reach leucine levels low enough to trigger apoptosis and that cannot be achieved by leucine-free diets. In this context, inhibition of autophagy seems beneficial. However, the role of autophagy in cancer is not entirely clear. Its inhibition could be a double-edged sword in that its actual effects are most likely context dependent (Mathew and White, 2011). Indeed, recent studies point to conflicting dual roles for autophagy acting both as a tumor suppressor and as an oncogene (Mathew and White, 2011). In fact, it has been suggested that the induction of autophagy can induce cell death in melanoma cells (Tormo et al., 2009). Experiments involving mice with Atg (autophagy-related) genes knocked out showed that impaired autophagy leads to tumor development. That is the case for knockout mice with an absence of Atg6/Beclin 1, Bif, Atg4c, Atg5 or Atg16L1, in which autophagy defects promote tumorigenesis (Mathew and White, 2011). This antitumorigenic role of autophagy could be mediated by accumulation of the scaffold protein p62, which is turned over via autophagy (Moscat and Diaz-Meco, 2009). Increased levels of p62 induce reactive oxygen species (ROS) accumulation, ER stress, DNA damage and genome instability, contributing to tumor promotion (Moscat and Diaz-Meco, 2009). This is also consistent with p62 being required for tumor survival by controlling NF-κB activation. Whether the accumulation of p62 that occurs in melanoma cells upon combine treatment with chloroquine and leucine deprivation could have undesirable longterm effects is not known and should be explored.

On the other hand, recent studies have shown that tumor cells, including melanoma cells, can escape the effects of cancer treatments by activating autophagy. In this context, autophagy can support cellular metabolism to help the growth of advanced tumors with high metabolic demands. Consistent with this notion, it has been shown that high levels of autophagy are associated with poor survival in patients with metastatic melanoma, suggesting that autophagy could be a prognostic factor and, in agreement with Sabatini’s paper, a therapeutic target in this type of tumor.

More work is undoubtedly needed to reconcile all of the apparently conflicting findings in the literature. In any case, it is clear that autophagy is a tightly regulated process and that its role in cancer is complex and probably cell-type dependent. Therefore, therapeutic strategies that involve inhibition of autophagy must be carefully tailored to uncouple the survival role for autophagy in response to nutrient starvation from its role in enhancing p62 levels and thus promoting oxidative stress, NF-κB activation, chronic inflammation and tissue damage. In the former context, inhibition of autophagy could be effective in slowing tumor growth and preventing therapeutic resistance. In the latter context, it could lead to undesirable protumorigenic effects.