In this paper, we have shown that P61A6 (GGTI) has significant anti-tumor effects on NSCLC cells in vitro and in vivo. Detailed analyses of the effects of P61A6 on one of the NSCLC cell lines, H358, showed that P61A6 inhibited anchorage-dependent and -independent growth of the cells, caused cell cycle effects, and inhibited the growth of mouse xenograft tumors whose treatment was initiated after the tumors became palpable. In GGTI-treated tumors, membrane association of RhoA was dramatically reduced, consistent with the presumed mechanism of action of P61A6. Since our previous P61A6 studies have focused on pancreatic cancer, this paper provides the first evidence to suggest that P61A6 may suppress tumorigenecity of NSCLC.
Another important contribution of this paper concerns the mechanism of action of P61-A6 on NSCLC cells, by providing evidence that RhoA plays critical roles in the effects of P61A6 on H358 cells. First, we have demonstrated that P61A6 inhibits geranylgeranylation as well as membrane association of RhoA, which is known to be geranylgeranylation-dependent. Consistent with this result, activation of RhoA - examined by determining the serum response to serum-starved cells - was blocked by the treatment with P61A6. In addition, we have shown that expression of a mutant form of RhoA (Rho-F) that can bypass the geranylgeranylation requirement abrogates the inhibition of RhoA membrane assocation and the inhibition of proliferation by P61A6. While other proteins such as Rac, Ral and RhoB have previously been suggested to play a role in GGTI effects in other cell lines [27–29], our study suggests that the effects of P61A6 on H358 lung cancer cells are largely mediated by RhoA.
Further characterization provided an overall view of the action of P61A6. We found that P61A6 induces accumulation of G1 phase cells, one of the hallmarks of GGTI effects , and that the level of cyclin D1/2 was decreased by P61A6 treatment. The significance of cyclin D1 in tumor growth and metastasis of NSCLC cells has been shown by the use of cyclin D1-targeted siRNA . In addition, RhoA has been shown to play critical roles in cyclin D1 expression, cell cycle, and proliferation of lung cells [25, 26]. Together with our demonstration that RhoA plays a major role in the effects of P61A6, the general scheme for the action of P61A6 on H358 may be summarized in the following way: P61A6 inhibits RhoA, leading to a decrease in cyclin D1/2, which results in G1 cell cycle arrest and inhibition of proliferation. There could, however, be variations to this general idea. In H358 cells, we have shown that P61A6 affects cyclin D1/2, while the levels of Cdk inhibitors p21CIP1/WAF1 and p27Kip1 are not significantly affected. In other cell lines, such Panc-1, however, we have observed increased p21CIP1/WAF1 levels after GGTI treatment [12, 14]. The differences might be attributable to divergence in the levels of these cell cycle regulators in different cell lines. In fact, we noted that, in contrast to cyclin D1/2, the levels of p21CIP1/WAF1 and p27Kip1 are quite high in H358 even before treatment, which may have contributed to P61A6 having a more pronounced effect on cyclin D1/2 than on p21CIP1/WAF1 or p27Kip1.
One issue that requires further investigation concerns effects of GGTI on RhoA activation. In our experiment, we showed that the activation of RhoA in response to serum stimulation is blocked by GGTI in lung cancer cells. This is consistent with other studies in endothelial and breast cancer cells. In endothelial cells, GGTI-286 blocked increase of RhoA-GTP induced by monocyte adhesion . GGTI-286 also blocked GTP-loading of RhoA induced by thrombin in endothelial cells . In breast cancer cells, RhoA activity as detected by RhoA-GTP was inhibited by GGTI-298 . However, Khan et al. [35, 36] reported that GGTase-I deficiency in macrophage resulted in the accumulation of RhoA-GTP. Further studies are needed to examine how GGTase-I deficiency influences RhoA activation in different cellular contexts.
Down-regulation and inactivation of DLC1 expression through genetic and epigenetic alterations in multiple malignancies may represent the most frequent mechanism for aberrant activation of Rho GTPases in human oncogenesis . Activity of Rho GTPases is elevated in many human cancers and their metastases, and the oncosuppressive effect of DLC1 requires RhoGAP activity, which negatively regulates Rho GTPases, most commonly RhoA [5, 38]. The observation that down-regulation of DLC1 in NSCLC is associated with a poor clinical outcome  implies that targeting pro-oncogenic pathways activated by this down-regulation could be especially useful therapeutically, and inhibition of the RhoA pathway and Rho kinase, a downstream effector of Rho, are promising options for therapeutic interventions.