Inducción de apoptosis por el agente antileucémico trióxido de arsénico (Trisenox) y otros agentes mitocondriotóxicospotenciación por inhibidores metabólicos

Laburpena

One of the common characteristics of tumor cells is their altered metabolic rate. Cancer cells mainly rely in high glycolysis rates to produce lactate (“Warburg effect”) and increased lipid metabolism to sustain the high energy demand of proliferating cells. This fact provides us with new opportunities for pharmacologic intervention. Thus, variable antitumor efficacy has been described for several glycolytic and lipid metabolism inhibitors. These are mostly used in combination with conventional antitumor drugs, but they have variable efficiencies. Arsenic trioxide (ATO, Trisenox) is an antileukemic agent against acute promyelocytic leukemia (APL), used in the clinics as a single agent or in combination with other drugs (retinoic acid (ATRA) or anthracyclines). At low concentrations, (<1 μM) ATO induces the degradation of oncoprotein PML-RARα (promyelocytic leukemia- all trans retinoic acid receptor alfa), which is found in APL and responsible of differentiation blockade. At high concentrations, ATO induces apoptotic cell death in APL and other tumor cells, which offers us new possibilities of therapeutic intervention. However, these applications would require the design of chemo-sensitization strategies in order to increase the efficiency and maintain clinically well-tolerated doses. Based on this, we aimed to analyze the potential cooperation between ATO and three glycolysis inhibitors (lonidamine, 2-deoxyglucose and 3- bromopyruvate) or a fatty acid β-oxidation inhibitor (etomoxir) in the induction of apoptosis in AML cells and to study the possible mechanisms involved. The main cell model utilized was the HL60 cell line, that is derived from an AML patient, due to its low sensitivity to ATO, and occasionally mitogen-stimulated peripheral blood lymphocytes (PBL) were used as a nontumoral cell model. Results obtained can be summarized as follows: 1. Used at clinically well tolerated doses, lonidamine, 2-DG and etomoxir efficiently cooperated with ATO to induce apoptosis in HL60 cells and other leukemic cell models, with limited efficacy in PBLs. Moreover, cooperation of such inhibitors with other antitumor drugs (cisplatin, etoposide) was limited or null. In contrast, 3-BrP was highly citotoxic (apoptosis and necrosis), by itself and the cooperation with ATO was lower. 2. Glycolytic inhibitors as well as ATO are mitochondrion-targeted drugs. Thus, lonidamine and 2-DG induced mitochondrial inner membrane permeability (mIMP) and membrane potential dissipation (ΔΨm). These effects can be a pre-condition of apoptosis, but do not explain by themselves the potentiation of apoptosis in combined treatments. Moreover, these inhibitors triggered mitochondrial outer membrane permeability (mOMP) and subsequent activation of the intrinsic apoptotic pathway that includes exportation of cytochrome c and other mitochondrial factors to cytosol, activation/translocation of Bax to mitochondria and XIAP degradation, phenomenon that are clearly potentiated in combined treatments. 3. According to energetic parameters, variable responses have been obtained. Thus, (a) 2-DG and 3-BrP triggered variable ATP depletion, whereas lonidamine and etomoxir did not, and in all cases these effects were not altered by co-treatment with ATO. The observed ATP depletion may be irrelevant in terms of cell death induction, except for the total ablation observed at necrotizing 3-BrP concentrations. (b) 2-DG induced dose-dependently glycolysis inhibition without affecting mitochondrial respiration; etomoxir inhibited mitochondrial respiration, effect which was compensated by glycolysis activation; and 3-BrP inhibited both, glycolysis and mitochondrial respiration. (c) The determination of adenine nucleotide subpopulations after treatment with 2-DG, 3-BrP and etomoxir, alone or in combination with ATO, showed variable increases in the AMP/ATP ratio and a decrease of energy charge, whose functional significance is yet to be determined. 4. An early consequence of metabolic inhibition, mainly of mitochondrial respiration, is reactive oxygen species (ROS) generation and accumulation, which could have deleterious effects. For instance, lonidamine, 3-BrP and etomoxir induce a moderate increase in the intracellular ROS content. Although this increase is not sufficient to justify single-drug toxicity, it explains the increase in lethality observed in combinations with ATO, which is an oxidative stress-sensitive agent. In contrast, 2-DG does not increase, but instead decreases ROS content. Moreover, etomoxir (slightly) and 3-BrP (dose-dependently) trigger intracellular GSH depletion, which may also explain their increased lethality in combination with ATO and, in the case of 3-BrP, its toxicity at higher concentrations. 5. Analysis of multiple protein kinase signaling pathways yielded multiple results, including: (a) lonidamine, 2-DG and 3-BrP activated the defensive (antiapoptotic) signaling pathways PI3K/Akt/mTOR and MEK/ERK, which explains the limited efficacy of these agents in monotherapy. (b) In the case of 2-DG, such activation is mediated by the early activation of insulin growth factor receptor-1 (IGF-1R). (c) The activation of these pathways is prevented by cotreatment with ATO, which in part explains the higher efficacy of combined treatments. (d) While lonidamine and etomoxir induce LKB-1/AMPK activation, 2-DG and 3-BrP unexpectedly induced inactivation. (e) 2-DG and 3-BrP-mediated inactivation of LKB- 1/AMPK occurs as a consequence of Akt activation, since both LKB-1/AMPK and PI3K/Akt behave as antagonic pathways. (f) Apoptosis enhancement in combined treatments is favored by AMPK inactivation in the cases of 2-DG and 3-BrP, and favored by AMPK activation in the case of lonidamine and etomoxir. 6. It is possible to improve apoptotic efficacy by the double inhibition of both energetic pathways, glycolysis and fatty acid oxidation, as it was demonstrated in the combined treatment of etomoxir with 2-DG or lonidamine and, furthermore, the effect of combined treatments is enhanced by co-incubation with ATO. Nevertheless, the potential use of this combined treatment is subject to restrictions, such the intrinsic metabolic properties of the tumor cell type used, as well as the fact that certain cooperation was observed in non-tumor cells (PBLs). Moreover, the study of the potential mechanisms involved in such cooperation has not had concluding results (yet). In summary, although the results shown herein have been obtained from in vitro cell models, they allow us to suggest that adequate combinations of metabolic inhibitors and antitumor agents like ATO could be useful to increase the clinical efficacy of all these agents, which is normally limited in monotherapy.