Blocking the MDM2-p53 protein-protein interaction has long been considered to offer a broad cancer therapeutic strategy, despite the potential risks of selecting tumors harboring p53 mutations that escape MDM2 control. In this study, we report a novel small molecule inhibitor of the MDM2-p53 interaction, SAR405838 (MI-77301) that has been advanced into Phase I clinical trials. SAR405838 binds to MDM2 with Ki = 0.88 nM and has high specificity over other proteins. A co-crystal structure of the SAR405838:MDM2 complex shows that in addition to mimicking three key p53 amino acid residues, the inhibitor captures additional interactions not observed in the p53-MDM2 complex and induces refolding of the short, unstructured MDM2 N-terminal region to achieve its high affinity. SAR405838 effectively activates wild-type p53 in vitro and in xenograft tumor tissue of leukemia and solid tumors, leading to p53-dependent cell cycle arrest and/or apoptosis. At well-tolerated dose schedules, SAR405838 achieves either durable tumor regression or complete tumor growth inhibition in mouse xenograft models of SJSA-1 osteosarcoma, RS4;11 acute leukemia, LNCaP prostate cancer and HCT-116 colon cancer. Remarkably, a single oral dose of SAR405838 is sufficient to achieve complete tumor regression in the SJSA-1 model. Mechanistically, robust transcriptional up-regulation of PUMA induced by SAR405838 results in strong apoptosis in tumor tissue, leading to complete tumor regression. Our findings provide a preclinical basis upon which to evaluate SAR405838 as a therapeutic agent in patients whose tumors retain wild-type p53.

Purpose Two clinical-stage anticancer drugs, the Bcl-2 inhibitor ABT-263 and the MDM2 inhibitor SAR405838 achieve complete tumor regression in animal models of leukemia but also induce acquired resistance. Elucidation of acquired resistance mechanisms and development of strategies to overcome the resistance are critical for their successful clinical development. Experimental Design We employed RS4;11 and MV4;11 cell lines, two acute leukemia models, to investigate acquired resistance mechanisms for both drugs in vitro and in vivo and evaluated several treatment regimens in xenograft mouse models to improve long-term, complete tumor regression. Results Resistance to either SAR405838 or ABT-263 (or its analogue ABT-737) develops in acute leukemia models in vitro and in vivo. RS4;11 and MV4;11 tumors treated with SAR405838 acquire resistance to the drug by mutation of the p53 gene or compromise of p53 function. RS4;11 tumors treated with either ABT-263 or ABT-737 acquire resistance primarily through down-regulation of BAX but not BAK. When acute leukemia cells become highly resistant to the MDM2 inhibitor, they retain their sensitivity to the Bcl-2 inhibitors, or vice versa. Certain sequential or combination treatment of SAR405838 and ABT-263 can achieve longer-term tumor regression than treatment with either agent alone. Conclusion Our study provides new insights into the mechanisms of acquired resistance of Bcl-2 and MDM2 inhibitors in acute leukemia models and suggests that certain sequential or combination treatment of these two distinct classes of apoptosis-inducing agents should be tested as new treatment strategies for acute leukemia in the clinic.

Two manganese nitrosyls, namely [Mn(SBPy3)(NO)](ClO4)2 (1) and [Mn(SBPy2Q)(NO)](ClO4)2 (2) have been synthesized by using designed pentadentate Schiff base ligands N,N-bis(2-pyridylmethyl)amine-N-ethyl-2-pyridine-2-aldimine (SBPy3) and N,N-bis(2-pyridyl methyl)amine-N-ethyl-2-quinoline-2-aldimine (SBPy2Q). Reaction of NO(g) with [Mn(SBPy3)(MeOH)](ClO4)2 and [Mn(SBPy2Q)(EtOH)](ClO4)2 in MeCN affords 1 and 2 respectively in good yields. Narrow-width peaks in the 1H NMR spectra and strong νNO at 1773 cm-1 (of 1) and 1759 cm-1 (of 2) confirm a strongly-coupled {low-spin Mn(II)-NO• }formulation for both these {Mn-NO}6 nitrosyls. In MeCN, 1 exhibits two strong absorption bands with λmax at 500 and 720 nm. These bands red shifts to 550 and 785 nm in case of 2 due to substitution of the pyridyl-imine moiety of SBPy3 with quinolyl-imine moiety in the SBPy2Q ligand frame. Exposure of solutions 1 and 2 to near-infrared (NIR) light (780 nm, 5 mW) results in rapid bleaching of the orange and fuchsia solutions and free NO is detected in the solutions by an NO-sensitive electrode. The high quantum yield values (Φ) of 1 (0.580 ± 0.010, λirr = 550 nm, MeCN) and 2 (0.434 ± 0.010, λirr = 550 nm, MeCN) and in particular their sensitivity to NIR light of 800-950 nm range strongly suggest that these designed manganese nitrosyls could be used as NIR light-triggered NO donors.

The iron nitrosyl [(PaPy2Q)Fe(NO)](ClO4)2 (2), derived from the quinoline-based ligand PaPy2QH (N,N-bis(2-pyridylmethyl)amine-N-ethyl-2-quinoline-2-carboxamide, where H is dissociable proton) has been characterized by spectroscopy and X-ray diffraction techniques. The 1H NMR spectrum (S = 0 ground state) and v(NO) value of 1885 cm(-1) indicate that 2 is a [Fe-NO]6 nitrosyl. Although 2 is stable in the dark, exposure of an acetonitrile solution of 2 (lambdamax = 510 nm) to light in the visible range causes rapid release of NO and formation of the solvato species [(PaPy2Q)Fe(MeCN)](ClO4)2 (6). Quantum yield (Phi) measurements indicate that 2 is a more efficient NO donor (Phi = 0.258) than [(PaPy3)Fe(NO)](ClO4)2 (1, Phi = 0.185), a complex derived from a similar but pyridine-based ligand. Interestingly, when the photoproduct 6 is exposed to water or a small amount of base, the triply bridged diiron(III) species [(PaPy2Q)FeOFe(PaPy2Q)](ClO4)2 (3) forms in good yield. This species can be independently synthesized from aerobic oxidation of the Fe(II) species [(PaPy2Q)Fe(MeCN)](ClO4) in acetonitrile. The structure of 3 reveals a unique Fe(III)-O-Fe(III) link supported by two (eta2,mu2)mu-NCO bridges derived from the carboxamido groups of the two (PaPy2Q)Fe(III) moieties.

SAR405838 is a potent and specific MDM2 inhibitor currently being evaluated in Phase I clinical trials for the treatment of human cancer. Using the SJSA-1 osteosarcoma cell line which harbors an amplified MDM2 gene and wild-type p53, we have investigated the acquired resistance mechanisms both in vitro and in vivo to SAR405838. Treatment of SJSA-1 cells with SAR405838 in vitro leads to dose-dependent cell growth inhibition, cell cycle arrest and robust apoptosis. However, prolonged treatment of SJSA-1 cells in vitro with SAR405838 results in profound acquired resistance to the drug. Analysis of in vitro-derived resistant cell lines showed that p53 is mutated in the DNA binding domain and can no longer be activated by SAR405838. Treatment of the parental SJSA-1 xenograft tumors with SAR405838 in mice yields rapid tumor regression but the tumors eventually regrow. Culturing the regrown tumors established a number of sublines, which showed only modest (3–5 times) loss of sensitivity to SAR405838 in vitro. Sequencing of the p53 showed that it retains its wild-type status in these in vivo sublines, with the exception of one subline, which harbors a single heterozygous C176F p53 mutation. Using xenograft models of two in vivo derived sublines, which has either wild-type p53 or p53 containing a single heterozygous C176F mutation, we showed that while SAR405838 effectively achieves partial tumor regression in these models, it no longer induces complete tumor regression and tumors resume growth once the treatment is stopped. Harvesting and culturing tumors obtained from a prolonged treatment with SAR405838 in mice established additional in vivo sublines, which all contain a single heterozygous C176F mutation with no additional p53 mutation detected. Interestingly, SAR405838 can still effectively activate p53 in all sublines containing a single heterozygous C176F mutation, with a moderately reduced potency as compared to that in the parental cell line. Consistently, SAR405838 is 3–5 times less effective in all the in vivo derived sublines containing a single heterozygous C176F p53 mutation than in the SJSA-1 parental cell line in assays of cell growth and apoptosis. Computational modeling suggested that a p53 tetramer containing two wild-type p53 molecules and two C176F mutated molecules can maintain the structural stability and interactions with DNA by formation of additional hydrophobic and cation-π interactions which compensate for the loss of sulphur-zinc coordination. Our data thus show that SJSA-1 tumor cells acquire very different levels of resistance in vitro and in vivo to the MDM2 inhibitor SAR405838. Our present study may have a significant implication for the investigation of resistant mechanisms for other classes of anticancer drugs.