Lovastatin inhibits EGFR dimerization and AKT activation in squamous cell carcinoma cells: potential regulation by targeting rho proteins.

We recently showed the ability of lovastatin to inhibit the function of the epidermal growth factor receptor (EGFR) and its downstream signaling of the phosphatidylinositol-3 kinase/AKT pathway. Combining lovastatin with gefitinib, a potent EGFR inhibitor, induced synergistic cytotoxicity in various tumor-derived cell lines. In this study, lovastatin treatment was found to inhibit ligand-induced EGFR dimerization in squamous cell carcinoma cells and its activation of AKT and its downstream targets 4E-binding protein 1 and S6 kinase 1. This inhibition was associated with global protein translational inhibition shown by a decrease in RNA associated polysome fractions. The effects of lovastatin on EGFR function were reversed by the addition of geranylgeranyl pyrophosphate, which functions as a protein membrane anchor. Lovastatin treatment induced actin cytoskeletal disorganization and the expression of geranylgeranylated rho family proteins that regulate the actin cytoskeleton, including rhoA. Lovastatin-induced rhoA was inactive as EGF stimulation failed to activate rhoA and inhibition of the rho-associated kinase, a target and mediator of rhoA function, with Y-27632 also showed inhibitory effects on EGFR dimerization. The ability of lovastatin to inhibit EGFR dimerization is a novel exploitable mechanism regulating this therapeutically relevant target. To explore the potential clinical significance of this combination, we evaluated the effect of statin on the overall survival (OS) and disease-specific survival (DSS) of patients with advanced non-small-cell lung cancer enrolled in the NCIC Clinical Trials Group phase III clinical trials BR21 (EGFR tyrosine kinase inhibitor erlotinib versus placebo) and BR18 (carboplatin and paclitaxel with or without the metalloproteinase inhibitor BMS275291). In BR18, use of statin did not affect OS or DSS. In BR21, patients showed a trend for improvement in OS (HR: 0.69, P=0.098) and DSS (HR: 0.62, P=0.048), but there was no statin x treatment interaction effect (P=0.34 and P=0.51 for OS and DSS, respectively).

Expression of thymidylate synthase in human cells is an early G(1) event regulated by CDK4 and p16INK4A but not E2F.

Thymidylate synthase (TS) is the enzyme that catalyses the last step in de novo thymidylate synthesis. It is of interest clinically because it is an effective target for drugs such as 5-fluorouracil, often used in combination therapy. Despite a number of earlier reports indicating that TS is a cell cycle-dependent enzyme, this remains equivocal. Here, we show that in HCT116 cells synchronised by serum starvation, there is a clear dissociation between the expression of cyclin E (a well-characterised cell-cycle protein) and TS. Although both cyclin E and TS mRNA and protein increased during G(1), TS upregulation was delayed. Moreover, TS levels did not decrease following S-phase completion while cyclin E decreased sharply. Similarly, clear differences were seen between cyclin E and TS as asynchronously growing HCT116 cells were growth-inhibited by low-serum treatment. In contrast to previous reports using rodent cells, adenovirus-mediated over-expression of E2F1 and cyclin E in three human cell lines had no effect on TS. Cell-cycle progression was blocked by treatment of cells with pharmacological inhibitors of CDK2 and CDK4 and by ectopic expression of p16INK4A. Whereas CDK2 inhibition had no effect on TS levels, inhibition of CDK4 was associated with decreased TS protein levels. These results provide the first evidence that drugs targeting CDK4 may be useful with anti-TS drugs as combination therapy for cancer.