PLEIOTROPIC EFFECTS OF PERIFOSINE ON GLIOBLASTOMA CELLS SURVIVAL: ALTERED MEMBRANE LIPID METABOLISM AND CELL SIGNALING

Glioblastoma multiforme (GBM) is the most frequent and aggressive malignant tumor of the central nervous system in adults. Despite decades of experimentation to improve the outcome of patients with GBM, this type of neoplasm remains one of the most lethal human cancers.Therefore, the need to test different and new agents for efficacy and safety is urgent. Perifosine (PF) is a synthetic lipid analogue belonging to a relatively new class of structurally related antitumor agents: the alkylphospholipids (APLs). PF exhibits potent antineoplastic activity against a multitude of cancer cell lines and different tumor models and is currently being tested in phase II clinical trial against major human tumors. However, the effect of PF against gliomas is poorly investigated. PF can induce apoptosis and/or cell growth arrest in tumor cells, but the details of its molecular mechanism is still to be elucidated. To date, the Ser/Thr kinase Akt, which is a key regulator of multiple survival pathways, is considered as the most important molecular target of PF. However, PF can induce also Akt-independent effects and the contribution of Akt inhibition to the clinical activity of PF remains to be assessed. As other ALPs, PF may alter the structure and function of cell membranes directly by inducing a biophysical disturbance of cell membranes where it accumulates and/or indirectly by interfering with the metabolism and transport of membrane lipids. In particular, alterations in the properties of lipid rafts, ordered membrane lipid domains enriched in cholesterol and sphingolipids (SLs), could affect numerous signaling pathways crucial to cell survival and proliferation that are dependent on these structures. On these premises, the purpose of this study was to investigate the sensitivity of GBM cells to PF treatment and to provide a contribution to the understanding of its molecular mechanism by focusing on the ability of PF to target membrane lipid metabolism and content, and, as a consequence, membrane-related signaling pathways crucial in the regulation of cell demise. At first, we evaluated the effect of PF on cell survival in several human GBM cell lines. We demonstrated that in these cell lines PF inhibits cell viability in a dose-dependent manner and that its cytotoxic effects are not solely due to Akt inhibition. Furthermore, we found that in glioma cells PF maintains ERK in its phosphorylated/active state in a sustained manner over time. Treatment with the MAPK inhibitor PD98059 potentiates PF toxicity, and strongly reduces PF-induced LC3B-II increase. This could thus represent a molecular mechanism for self-defense from PF, at least in part due to the induction of protective autophagy. Moreover, in cells exposed to PF we found a time-dependent increase in the number of giant and multinucleated cells with an irregular shape, these morphological changes resembling those described for mitotic catastrophe, suggesting that this could be the mechanism of PF-induced cells death, while apoptosis was undetectable. Accumulating literature indicates that in several tumor cell lines PF inhibits the rate limiting step of phosphatidylcholine (PC) synthesis, which is catalyzed by CTP:phosphocholine cytidylyltransferase, and this was associated to cell death with still unclear mechanisms. Furthermore, PC is the donor of phosphocholine in the reaction catalyzed by the enzyme sphingomyelin synthase (SMS), so the inhibition of the PC biosynthesis may also affect the sphingomyelin (SM) biosynthesis from ceramide (Cer). We found that also in GBM cells PF affects PC biosynthesis. In addition, in our model PF inhibits SM biosynthesis by affecting SMS activity, and the reduced PC seems not to represent a limiting factor for SM synthesis. The decreased utilization of Cer for SM biosynthesis results in a modest increase of Cer, which can accumulate at the endoplasmic reticulum (ER). Therefore, both the increased Cer levels, both the inhibition of PC synthesis in the ER could trigger ER stress ultimately leading to cell death. However, it is known from literature that in our GBM cells model PF fails to provoke ER stress, suggesting that other aspects of membrane lipid homeostasis, such as that involved in membrane functionality, are most probably involved in PF-induced cell death. Indeed, PC is the most abundant phospholipid in eukaryotic cellular membranes, essential for new membrane formation. On the other hand, SM together with glycosphingolipids (GSLs) represent the major SLs of the plasma membrane, displaying an asymmetric or polarized distribution, and play important roles in the regulation of membrane fluidity and sub-domain structures involved in cell signaling. We demonstrated that PF affected the endogenous levels of phospholipids (PLs), inducing a decrease in the total PLs/protein ratio accompanied by a change in the PLs pattern. PF-treated cells were poorer in PC, SM, and phosphatidylserine (PS), and richer in phosphatidylethanolamine (PE). In addition, PF induces an about 60% increase of endogenous gangliosides content compared to untreated cells. The dramatic changes in PLs and SLs endogenous content induced by PF could significantly affect membrane composition and functionality, which in turn can be involved in the biological response of cells to PF treatment. Indeed, we found that the inhibition of SM synthesis mimicked and potentiate PF cytotoxic effects; in addition the inhibition of gangliosides biosynthesis reverses cytotoxic action of PF thus suggesting that increase of gangliosides content is essential for its cytotoxic action. In conclusion, PF treatment in GBM cells results in a complex network of effects where the alteration of the metabolism and content of membrane-lipid components (and maybe their related secondary messengers), seems to play a key role in determining cell death. This study indicates that PF is decisive in its target to fight GBM cells, this representing a critical push to study new aspects in its cellular mechanisms, implying PF as anti-GBM agent.