13C NMR (125 MHz, DMSO-503.9 (M C H)?. 3-((1-Hydroxy-4-(naphthalene-2-sulfonamido)naphthalen-2-yl)thio)propanoic Acid (38) Synthesized using the procedure for 1 except 49ee was used as the starting (+)-Phenserine material. The title compound (23 mg, 15%) was obtained as a white solid after HPLC purification. 1, 2), Pd(PPh3)2Cl2, CuI, Et3N/THF, 60 C, 2 (+)-Phenserine h (4:1), 60 C, 2 h; (c) Fe, AcOH, 70 C, 1 h, or Pd/C, H2 30 psi, EtOH/EtOAc (6:1), rt, overnight; (d) RSO2Cl, pyridine, CH2Cl2, rt, overnight, or RCOCl, Et3N, CH2Cl2, rt, overnight, or , CH3CN, 80 C, 15 h; (e) BBr3, CH2Cl2, 0 C to rt, 1 h, or BBr3, CH2Cl2, 0 C to rt, 1 h, quench with MeOH at 0 C; (g) phenyl boronic acid, Pd(PPh3)4, Na2CO3, THF/H2O, 60 C, 2 h; (h) NH4OH, rt, 1 h; (i) NaN3, SiCl4, CH3CN, 80 C, 15 h; (j) LiOH, THF, rt, 1 h; (k) H3CCOCl, Et3N, 0 C, rt, 30 min. StructureCActivity Relationships Structure-based design of analogues based on 1 yielded a focused library of compounds leading to clear SAR for this series. The binding affinities of our Mcl-1 inhibitors were determined by using competitive fluorescence polarization (FP) and surface plasmon resonance (SPR) binding assays, which test the ability of inhibitors to disrupt interaction between Mcl-1 and two different BH3 peptides, fluorescently labeled Bid and biotin-labeled Bim, respectively. Concurrently, HSQC NMR experiments were performed to provide structural insights of protein-bound ligand and experimental validation for the modeling studies. The predicted binding model showed that the thiophene ring at R1 of 1 1 projects into the h2 pocket (Figure ?(Figure1B),1B), which is the biggest and deepest pocket among the four hydrophobic pockets of Mcl-1.56 To investigate the importance of hydrophobic interaction at this site and increase the binding affinity of 1 1, a series of analogues with variation at R1 was synthesized and evaluated (Table 1). When R1 is changed to a methyl group in 2, the binding affinity is significantly reduced, confirmed by SPR (IC50 > 100 M) and NMR experiments which showed lack of chemical shift perturbation of backbone residues in the Mcl-1 BH3 binding site after adding 2 (Supporting Information Figure S1). As was expected, isosteric replacement of the thiophene in 1 to a phenyl in 3 maintained binding affinity with test, and the number of data is shown for each tested concentration with corresponding significance: (**) < 0.01 and (***) < 0.001. To further confirm the specificity of our novel Mcl-1 inhibitors and to determine whether different prosurvival Bcl-2 proteins could suppress the apoptotic activities of novel Mcl-1 inhibitors, we used reported cell lines developed by retroviral transduction of lymphoma cells isolated from E-myc transgenic mice which differ only in their expression of prosurvival Bcl-2 family proteins.81 Lymphoma cells overexpressing Mcl-1 and Bcl-2 were treated with varying concentrations of tested compounds for 15C18 h, and then cell viability was determined by flow cytometry using a fluorescent reactive dye (LIVE/DEAD fixable violet stain kit). ABT-263 (navitoclax), a selective inhibitor of Bcl-2, Bcl-xL, and Bcl-w, was used as a positive control. As predicted, lymphoma cells overexpressing Mcl-1 Mouse Monoclonal to Human IgG were significantly sensitive to 19 and 21 as assessed by an increased percentage of cell death in a concentration-dependent manner. In contrast, 19 and 21 were ineffective against E-myc/Bcl-2 lymphomas (Figure ?(Figure7).7). Importantly, 41 did not show any activity against both cell lines overexpressing Mcl-1 or Bcl-2, consistent with our binding studies which showed that 41 does not bind to Mcl-1. As expected, lymphoma cells overexpressing Bcl-2 were sensitive to cell death induced by ABT-263, while cells overexpressing Mcl-1 were insensitive to ABT-263, consistent with its binding specificity. Collectively, these results, demonstrate that 19 and 21 specifically bind and inhibit Mcl-1 and have no (+)-Phenserine effect on Bcl-2, which is consistent with our biochemical data for their selectivity profiles. Furthermore, this supports the concept that tumor cells addicted to Mcl-1 protein will be the most sensitive target cell population for selective small-molecule Mcl-1 inhibitors. Open in a separate window Figure 7 Sensitivity of E-myc lymphoma cells overexpressing Mcl-1 and Bcl-2 antiapoptotic proteins to inhibitor-induced cell death. E-myc/Mcl-1 (+)-Phenserine and E-myc/Bcl-2 lymphomas were treated for 15C18 h with increasing concentrations of 19, 21, 41, and ABT-263. Dead cells were assessed by LIVE/DEAD fixable dead cell stain kit (ViVID). The data shown represents means SEM from 3C7 independent experiments. The (+)-Phenserine significance was calculated using unpaired test, and the number of data is shown for each tested concentration with corresponding significance: (*) is < 0.05, (**) < 0.01, and (***) < 0.001. We next evaluated our most potent compounds for their ability to inhibit cell growth in the leukemia cell lines HL-60, MV4,11, and K-562 (Figure ?(Figure8).8). It has been shown that AML-derived cell lines, HL-60 and MV4,11, are sensitive to inhibition of the antiapoptotic protein Mcl-1, while CML-derived.