Regulation of the microenvironment and chemoresistance in liver therapy

Abstract

[eng] Hepatocellular carcinoma (HCC) is the most common liver cancer and currently the fourth cause of cancer-related death. Since its appearance, the multityrosine kinase inhibitor (MKI) sorafenib has been the standard systemic treatment for HCC patients. Recently, novel MKI have become available, such as lenvatinib, regorafenib and cabozantinib, as well as immune-based therapies. Still, the efficacy of MKI should be improved. On the other hand, the BCL-2 family of proteins tightly regulate programmed cell death. Therefore, BH3 mimetics, small molecules which mimic BCL-2 proteins, have been proposed as chemotherapeutic compounds to trigger cell death, although their efficacy against solid tumors has not been widely explored. Firstly, we identified that regorafenib upregulated the mRNA expression of the antiapoptotic BCL-xL in murine liver tumors. Taking these results into account, we proposed that regorafenib anti-tumor action could be enhanced by the addition of the BH3 mimetic A-1331852, which specifically targets the BCL-xL protein. When hepatoma cells were treated with this combination of agents, a potent cytotoxic effect was observed. Contrarily, ABT-199 administration, which blocks BCL-2, did not cooperate with regorafenib to increase cell death in liver cancer cells. The dual treatment of regorafenib and A-1331852 induced a loss in mitochondrial membrane potential, the release of cytochrome c and an increase in caspase-3 activity, indicating that the mitochondrial pathway of apoptosis was activated in liver cancer cells. Furthermore, regorafenib treatment was found to decrease MCL-1 protein levels in hepatoma cells. Hence, a specific inhibitor of MCL-1 together with A-1331852 was administered to HCC cells, resulting in strong cell death. Regorafenib and A-1331852 also reduced tumor liver spheroids growth. In a PDX mouse model, the co-administration of regorafenib and the BCL-xL antagonist A-1331852 reduced tumor volume and decreased its proliferation. This combination also proved to be effective even in regorafenib-resistant cells and animal models. Clinical data showed that the ratio of BCL-xL/MCL-1 was increased in HCC patients, including early and advanced stages of the disease. In short, the dual treatment of regorafenib and the BH3 mimetic A-1331852 was highly effective against HCC preclinical models. MKI therapy has been described to increase reactive oxygen species (ROS) production. We wondered whether the employment of a pro-oxidant compound, like BSO, which provokes an intracellular GSH depletion, could aid in sorafenib, regorafenib and cabozantinib effectiveness, or, on the contrary, the use of antioxidant supplements might counteract its efficacy. We observed that BSO-treated hepatoma cells were much more sensible to the administration of sorafenib and regorafenib. Both MKIs increased mitochondrial ROS generation in HCC cells, and that effect was enhanced with BSO pretreatment. Likewise, the cytotoxic capacity of regorafenib and A-1331852 was found to be ROS-mediated. Hepatoma cells were treated with two different MKIs and BH3 mimetic combinations and the addition of the antioxidants MnTBAP and GSHe increased cell viability, suggesting that their use could considerably interfere with chemotherapy effectiveness. Liver cancer spheroids displayed an increase in mitochondrial ROS production when treated with BSO and sorafenib. Again, the use of MnTBAP and GSHe blocked sorafenib effect on tumor liver spheroids. Finally, an induction of mitophagy was observed with the depletion of GSH and sorafenib/regorafenib treatment in he patoma cells. In conclusion, MKIs exert their cytotoxicity via a mitochondrial ROS generation and antioxidants supplementation may restrain MKI chemotherapeutic effect in liver cancer cells.