Classification of liposarcoma into three biological types encompassing five subtypes, (a) well-differentiated/dedifferentiated, (b) myxoid/round cell, and (c) pleomorphic, based on morphologic features and cytogenetic aberrations, is widely accepted. However, diagnostic discordance remains even among expert sarcoma pathologists. We sought to develop a more systematic approach to liposarcoma classification based on gene expression analysis and to identify subtype-specific differentially expressed genes that may be involved in liposarcoma genesis/progression and serve as potential therapeutic targets. A classifier based on gene expression profiling was able to distinguish between liposarcoma subtypes, lipoma, and normal fat samples. A 142-gene predictor of tissue class was derived to automatically determine the class of an independent validation set of lipomatous samples and shows the feasibility of liposarcoma classification based entirely on gene expression monitoring. Differentially expressed genes for each liposarcoma subtype compared with normal fat were used to identify histology-specific candidate genes with an indepth analysis of signaling pathways important to liposarcoma pathogenesis and progression in the well-differentiated/ dedifferentiated subset. The activation of cell cycle and checkpoint pathways in well-differentiated/dedifferentiated liposarcoma provides several possible novel therapeutic strategies with MDM2 serving as a particularly promising target. We show that Nutlin-3a, an antagonist of MDM2, preferentially induces apoptosis and growth arrest in dedifferentiated liposarcoma cells compared with normal adipocytes. These results support the development of a clinical trial with MDM2 antagonists for liposarcoma subtypes which overexpress MDM2 and show the promise of using this expression dataset for new drug discovery in liposarcoma. [Cancer Res 2007;67(14):6626-36]

MDM2 is a critical negative regulator of the p53 tumor suppressor protein. Recently, small-molecule antagonists of MDM2, the Nutlins, have been developed to inhibit the p53-MDM2 interaction and activate p53 signaling. However, half of human cancers have mutated p53 and they are resistant to Nutlin treatment. Here, we report that treatment of the p53-mutant malignant peripheral nerve sheath (MPNST) and p53-null HCT116 cells with cisplatin (Cis) and Nutlin-3a induced a degree of apoptosis that was significantly greater than either drug alone. Nutlin-3a also increased the cytotoxicity of both carboplatin and doxorubicin in a series of p53-mutant human tumor cell lines. In the human dedifferentiated liposarcoma cell line (LS141) and the p53 wild-type HCT116 cells, Nutlin-3a induced downregulation of E2F1 and this effect appeared to be proteasome dependent. In contrast, in MPNST and HCTp53À/À cells, Nutlin-3a inhibited the binding of E2F1 to MDM2 and induced transcriptional activation of free E2F1 in the presence of Cis-induced DNA damage. Downregulation of E2F1 by small interfering RNA significantly decreased the level of apoptosis induced by Cis and Nutlin-3a treatment. Moreover, expression of a dominant-negative form of E2F1 rescued cells from apoptosis, whereas cells overexpressing wildtype E2F1 showed an increase in cell death. This correlated with the induction of the proapoptotic proteins p73a and Noxa, which are both regulated by E2F1. These results indicate that antagonism of MDM2 by Nutlin-3a in cells with mutant p53 enhances chemosensitivity in an E2F1-dependent manner. Nutlin-3a therefore may provide a therapeutic benefit in tumors with mutant p53 provided it is combined with chemotherapy.

High-resolution magic-angle-spinning (HR-MASMagic-angle-spinning (MAS) of cell and tissue samples increases the sensitivity and resolution of metabolites by removing multiple line-broadening mechanisms such as proton dipolar coupling and susceptibility effects arising within a tissue or cell sample (1-4). This enhances our ability to study and dissect interactions of water with tissue and cell components compared to NMR studies performed under static conditions. Most importantly, MAS allows a direct detection of magnetization transfer (MT) between water and the membrane phospholipids and proteins that cannot be observed in the static state. Understanding the mechanism of magnetization exchange between water, phospholipids, protein, and other metabolites in tissue and cells is essential in many respects. This transfer of magnetization is the principal determinant of water relaxation behavior in tissue and cells (5,6). MT induced by the interaction between water and cellular metabolites provides a useful contrast that can be used in MRI (7-9). A complete understanding of MT between water and cellular components in cell and tissue samples is critical for optimal pulse sequence design in NMR experiments in these complex biological systems, and for interpretation of the HR-MAS NMR data acquired with different pulse sequence techniques. One such example is the selection of the most appropriate water suppression approach to acquire the NMR spectra without the loss of useful signals from magnetization exchange between water and these cellular metabolites and components (10,11).The interaction between water and metabolites involves a multistep process of MT that includes direct magnetization mixing between water and metabolites, and concurrent MT within metabolites and to other metabolites. These processes have equal importance in elucidating water relaxation and MT in vivo. Direct water interaction with metabolites or generally the intermolecular spin magnetization exchange is brought about through two distinct mechanisms: chemical exchange and dipolar cross relaxation nuclear overhauser effect (NOE). The most notable chemical-exchange effect in tissue and cells occurs between water and protein amide protons, while the much faster exchange between water and protein hydroxyl protons is beyond the time scale of direct NMR detection (12). The intermolecular NOE between water and metabolites in tissue and cells is rather complicated because the time scales of motion for different metabolites are significantly different. Two proton groups display observable NOE if they are within a reasonable distance from each other. Water molecules trapped inside or at the surface of the proteins and membrane phospholipids involve NOE with the surrounding protons of proteins or lipids, as well as chemical exchange with the protein amide protons (13,14). The NOE between water and protein, and the NOE between water and model membrane lipids have been examined in pure samples in the static state (14). The NOE in a model membrane was also exte...

Malignant peripheral nerve sheath tumors (MPNST) are soft-tissue tumors with a very poor prognosis and largely resistant to chemotherapy. MPNSTs are characterized by activation of the Ras pathway by loss of tumor suppressor neurofibromatosis type 1. In view of this, MPNST may be susceptible to inhibition of the activated Ras/Raf/mitogenactivated protein kinase pathway by the B-Raf inhibitor sorafenib. MPNST (MPNST and ST8814) and dedifferentiated liposarcoma (LS141 and DDLS) human tumor cell lines were characterized for Ras activation and B-Raf expression. Tumor cells were treated with sorafenib and examined for growth inhibition, inhibition of phospho-MEK, phospho-ERK, cell cycle arrest, and changes in cyclin D1 and pRb expression. MPNSTs were sensitive to sorafenib at nanomolar concentrations. This appeared to be due to inhibition of phospho-MEK, phospho-ERK, suppression of cyclin D1, and hypophosphorylation of pRb at the CDK4-specific sites, resulting in a G 1 cell cycle arrest. These effects were not seen in the liposarcoma cells, which either did not express B-Raf or showed decreased Ras activation. Small interfering RNA -mediated depletion of B-Raf in MPNSTs also induced a G 1 cell cycle arrest in these cells, with a marked inhibition of cyclin D1 expression and Rb phosphorylation, whereas depletion of C-Raf did not affect either. With growth inhibition at the low nanomolar range, sorafenib, by inhibiting the mitogenactivated protein kinase pathway, may prove to be a novel therapy for patients with MPNST. [Mol Cancer Ther 2008; 7(4):890 -6]

Gastrointestinal stromal tumors (GIST) are characterized by activating mutations in the c-KIT gene which confers ligandindependent activation of the KIT receptor. Imatinib mesylate has been shown to effectively block constitutively active KIT and delay tumor growth. However, resistance to imatinib mesylate is emerging as a major clinical problem and novel therapies are needed. We report that treatment of GIST cells with the transcriptional inhibitor flavopiridol, initially downregulates the antiapoptotic proteins bcl-2, mcl-1, and X-linked inhibitor of apoptosis protein which occurs as early as 4 hours after exposure. This is followed at 24 hours by the transcriptional suppression of KIT resulting in poly(ADP-ribose) polymerase cleavage and apoptosis. To separate the apoptotic effect of KIT suppression relative to the down-regulation of antiapoptotic proteins, we used small interfering RNAdirected knockdown of KIT. Results show that focused suppression of KIT alone is sufficient to induce apoptosis in GIST cells, but not to the same extent as flavopiridol. In contrast, imatinib mesylate, which inhibits KIT kinase activity but does not suppress total KIT expression, fails to cause apoptosis. We also show that flavopiridol suppresses KIT mRNA expression through positive transcriptional elongation factor inhibition and decreases KIT promoter activity. This causes a global decrease in the level of functionally mature KIT at the cell surface, resulting in a decrease in autophosphorylation at tyrosine residues 703 and 721, which characterizes activated KIT. Our results indicate that targeting KIT expression and these antiapoptotic proteins with flavopiridol represents a novel means to disrupt GIST cell dependence on KIT signaling and collectively renders these cells sensitive to apoptosis. (Cancer Res 2006; 66(11): 5858-66)