Bone sarcomas cause disproportionate morbidity and mortality and desperately need new therapies as there has been little improvement in outcomes in 20 years. Identification of critical signaling pathways, including type 1 insulin-like growth factor receptor (IGF-1R) for Ewing sarcoma and possibly osteosarcoma, and the ERBB and the Wnt signaling pathways for osteosarcoma, have emerged as receptors mediating vital signals for bone sarcoma. Akt, mammalian target of rapamycin (mTOR), phosphoinositide 3-kinases, mitogen-activated protein kinase kinase, extracellular signal-regulated kinases, and Ras pathway play key roles in at least some tumors, and inhibition of mTOR in particular will likely lead to improved survival, although clinical trials are still underway. The Notch pathway and ezrin are essential for osteosarcoma metastasis, and Fas downregulation is necessary for survival of metastases in lungs. As little is known about chondrosarcoma signaling, more preclinical work is needed. By defining vital signaling pathways in bone sarcomas, small molecule inhibitors can be applied rationally, leading to longer survival and reducing morbidity and late effects from intensive chemotherapy.

Introduction

Farnesyltransferase inhibitors (FTIs), small molecules that inhibit prenylation of multiple proteins such as Ras and Rho, have clinical promise for cancer. FTIs block malignant growth and promote cell death in a wide variety of tumor types in vitro and in vivo, but the mechanisms underlying the antitumor activity of FTIs are not completely understood. We have observed that colon cancer cell line OS187 responds to the FTI tipifarnib with a senescent phenotype, demonstrated as altered cell morphology, elevated senescence-associated β-galactosidase activity, upregulated protein and mRNA levels of senescence markers, and long-term cell cycle arrest. Our previous results also indicated that cells retaining intact p53 are more sensitive to tipifarnib than are cells with absent/inactive p53, suggesting a possible role for p53 in the senescent response to tipifarnib. To test this hypothesis, we used shRNA to reduce p53 expression in OS187 cells. Loss of p53 reversed the swollen and enlarged senescent-like cellular phenotype induced by tipifarnib. In addition, p53 knockdown cells showed less senescence-associated β-galactosidase activity than did scrambled control cells after 7-day treatment with tipifarnib. Cell viability assay suggested that p53 knockdown partially rescued the cytotoxic effects of tipifarnib at low concentrations. Western blot also demonstrated that cell cycle regulatory proteins such as p27, p21, and p15 were significantly upregulated in scrambled control cells while only a slight increase of these proteins were observed in p53 knockdown cells after tipifarnib treatment. Flow cytometry analysis showed that cells with decreased p53 level had a reduced percentage of cells in sub G1 phase when treated with a high concentration of tipifarnib, and fewer cells arresting in G2/M phase when treated with low concentration of tipifarnib compared to the scrambled control cells. This observation suggested that p53 may involve in different pathways depending on the strength of the stress signal imposed by different concentration of tipifarnib. In conclusion, these results confirm that p53 is important for inducing senescence when farnesylation is inhibited in OS187 cells. However, since p53 knockdown has very limited impact on the anti-proliferation effects of tipifarnib at high concentrations, other pathways are expected to play a role in mediating cell susceptibility to tipifarnib. Future experiments will focus on investigating the p53-independent pathways that are responsible for tipifarnib-induced senescence and apoptosis. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 2876. doi:1538-7445.AM2012-2876

Farnesyltransferase inhibitors (FTIs), small molecules that inhibit prenylation of multiple proteins such as Ras and Rho, have clinical promise for cancer. FTIs block malignant growth and promote cell death in a wide variety of tumor types in vitro and in vivo, but the mechanisms underlying the antitumor activity of FTIs are not completely understood. We have observed that osteosarcoma cell lines respond to the FTI tipifarnib with growth arrest, cellular enlargement, and flattened morphology, all changes consistent with the senescent phenotype (Hall et al., 2010). These observations led us to evaluate the ability of FTIs to induce senescence in osteosarcoma. Further, it was clear that cells retaining intact p53 responded at much lower concentrations of FTIs than did cells with absent/inactive p53, suggesting a possible role for p53 in this response. We measured endogenous β-galactosidase activity by in situ chromatography and quantified expression levels of common senescence markers including p53, p21Cip1, p16INK4a and DEC1 by densitometry analysis of western blot and qPCR in two human osteosarcoma cell lines, OS187 and Saos2. Cell growth was measured by direct counting of nuclei, and cell cycle was analyzed by PI staining and flow cytometry. Apoptosis was assessed by PARP cleavage, annexin V staining and flow cytometry. Variable responses to tipifarnib correlated with p53 status. In p53 proficient OS187 cells, 24h exposure to very low concentration of tipifarnib (10−9 M) induced upregulation of DEC1 and a 45-fold increase in p21Cip1 protein, accompanied by increased expression and phosphorylation of p53. This event also correlated with transient reduction of cyclin A, B1, and E proteins in a time-dependent manner, leading to cell cycle arrest which is comparable to tipifarnib-treated 293T cells. OS187 cells also exhibited elevated senescence-associated β-galactosidase activity in response to low-dose tipifarnib treatment after 24h. Meanwhile, Saos2 cells (p53-null) gave no response at 10−9 M tipifarnib. At higher concentrations (10−7 M), modest induction of p21Cip1 (10-fold increase) and p16INK4a (2-fold increase) proteins and mild inhibition of cell cycle progression were observed. Augmented senescence-associated β-galactosidase activity became detectable after Saos2 cells were exposed to high concentration of tipifarnib (10−6 M) for at least 48h. In addition to senescence, tipifarnib increased PARP cleavage and annexin V staining in OS187 cells, indicating that apoptotic cell death may also play a role in the anti-tumor activity of FTIs. These findings suggest that tipifarnib could efficiently promote senescence through a p53-dependent pathway at low concentrations, and via p53 independent pathways at higher concentrations. Induction of both cellular senescence and apoptosis may contribute to the growth inhibition induced by FTIs in osteosarcoma. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 2640. doi:10.1158/1538-7445.AM2011-2640

of Molecular Cancer Therapeutics (1), new data, including DNA fingerprint analysis, in vivo experiments, and protein biochemistry studies, were obtained showing that two cell lines used for this study, OS 187 and COL, are not osteosarcoma cell lines. The authors state that they are the first group to report and characterize these cell lines (2), though OS 187 was not generated by the group. The authors are correcting their publications that have used OS 187 and COL. DNA fingerprint analysis for OS 187 is identical to HCT 15, a common colon cancer cell line included in the NCI60 panel. COL has a unique DNA fingerprint, but several in vitro and in vivo analyses recently described (3) demonstrate that it is, in fact, a neuroblastoma.Although the report was written with the assumption that all lines used were osteosarcoma cell lines, it is now clear that these results cannot be interpreted to represent osteosarcoma biology specifically, and that any references specific to osteosarcoma in the body of the article should be disregarded. The title and abstract now are revised to reflect this change. The biology of Ras signaling and response to farnesyltransferase inhibitors, especially in cell lines with low endogenous Ras activity, is likely to be general and not specific to one particular disease type. Indeed, the observation of similar effects in cell lines of diverse origins speaks to the generalizability of these results. Thus, although the identity of the cell lines requires correction, the authors stand by the observations reported.