In the present study, we investigated the anticancer effects of the mitochondrial inhibitors, metaiodobenzylguanidine (MIBG), metformin and phenformin. 131I-MIBG has been used for scintigraphic detection and the targeted radiotherapy of neuroblastoma (NB), a pediatric malignancy. Non-radiolabeled MIBG has been reported to be cytotoxic to NB cells in vitro and in vivo. However, the mechanisms behind its growth suppressive effects have not yet been fully elucidated. Metformin and phenformin are diabetes medications that are being considered in anticancer therapeutics. We investigated the anticancer mechanisms of action of MIBG and metformin in NB. Our data revealed that both drugs suppressed NB cell growth and that the combination drug treatment was more potent. MIBG reduced MYCN and MYC expression in MYCN-amplified and non-MYCN-amplified NB cells in a dose- and time-dependent manner. Metformin was less effective than MIBG in destabilizing MYC/MYCN. The treatment of NB cells with metformin or MIBG resulted in an increased expression of genes encoding biomarkers for favorable outcome in NB [(ephrin (EFN)B2, EFNB3, EPH receptor B6 (EPHB6), neurotrophic tyrosine kinase, receptor, type 1 (NTRK1), CD44 and Myc-interacting zinc finger protein (MIZ-1)] and tumor suppressor genes [(early growth response 1 (EGR1), EPH receptor A2 (EPHA2), growth arrest and DNA-damage-inducible, beta (GADD45B), neuregulin 1 (NRG1), TP53 apoptosis effector (PERP) and sel-1 suppressor of lin-12-like (C. elegans) (SEL1L)]. Accordingly, metformin and MIBG augmented histone H3 acetylation in these cells. Phenformin also exhibited histone modification and was more effective than metformin in destabilizing MYC/MYCN in NB cells. Our data suggest that the destabilization of MYC/MYCN by MIBG, metformin and phenformin and their effects on histone modification are important mechanisms underlying their anticancer effects.

OSU-03012, a 3-phosphoinositide-dependent kinase-1 (PDK1) inhibitor, destabilizes MYCN and MYC proteins in neuroblastoma cells. However, AKT phosphorylation is barely detectable in neuroblastoma cells under normal culture conditions whether treated with OSU-03012 or not. This observation suggests that PDK1 is not the main target of OSU-03012 to destabilize MYC and MYCN in neuroblastoma cells. In the present study, we explored one of the possible mechanisms by which OSU-03012 destabilizes MYC and MYCN. Since Aurora kinase A is reported to phosphorylate GSK3β, leading to its inactivation, we hypothesized that one of the targets of OSU-03012 is Aurora kinase A. Comparative analysis of OSU-03012 and VX-680, a potent and specific inhibitor of Aurora kinases, showed that both inhibitors destabilized MYC and MYCN and were significantly growth suppressive to neuroblastoma cell lines. In silico molecular docking analysis further showed that the calculated interaction energy between Aurora kinase A and OSU-03012 was −109.901 kcal/mol, which was lower than that (−89.273 kcal/mol) between Aurora kinase A and FXG, an Aurora kinase-specific inhibitor. Finally, an in vitro Aurora kinase A inhibition assay using a recombinant Aurora kinase A showed that OSU-03012 significantly inhibited Aurora kinase A, although it was weaker in potency than that of VX-680. Thus, OSU-03012 has a likelihood of binding to and inhibiting Aurora kinase A in vivo. These results suggest that OSU-03012 affects multiple cellular targets, including Aurora kinase A, to exhibit its growth suppressive and MYC and MYCN-destabilizing effects on neuroblastoma and other cancer cells.

Cancer stem cells (CSCs) are plastic in nature, a characteristic that hampers cancer therapeutics. We previously reported the establishment of induced neuroblastoma (NB) stem cells (iCSC), which were stable stem cell-like NB cells in sphere culture. These cells recapitulated the in vivo histological phenotypes of Large-Cell NB (LCN), the most aggressive and deadly subset of NB, including vesicular nuclei and two to three prominent nucleoli (PNAS Vol. 110: 6097-102 2013). In addition to the LCN phenotype, high-level expression MYC or MYCN is the consistent feature of the iCSC xenografts grown in immune-compromised mice. The goal of this study is to identify lead organic compounds active against NB iCSCs by high throughput screening. By using MTS assays, the SKNAS iCSC and the parental monolayer cells were used to screen the Prestwick Chemical Library®, containing 1200 FDA-approved small molecules. Growth suppressive compounds were defined as they exhibited cell survival reduction of greater than 80% compared to DMSO control. There were unexplored compounds with significant antineoplastic activity among existing antibacterial, antifungal, antiprotozoal, and anthelmintic compounds. Among the compounds screened, we identified a dozen of small molecules that were preferentially growth suppressive to the SKNAS iCSC over the monolayer cells. The secondary MTS assay confirmed so far that Chlorhexidine (antiseptic), Benzethonium (antiseptic) and Digoxin (cardiac glycoside) were potential anti-NB iCSC agents: they were more effective in suppressing growth of the SKNAS iCSC than the monolayer cells. In addition, these compounds exhibited MYC destabilizing activity in the SKNAS iCSC. Notably, JQ1, the recently identified MYC destabilizer, was less effective than Chlorhexidine, Benzethonium and Digoxin in suppressing growth of SKNAS iCSC. The above three compounds also showed the MYCN destabilizing effect in another pair of MYCN-amplified SKNBE(2)C iCSCs and SKNBE(2)C monolayer cells. We are currently completing characterization of the small molecules identified with focus on their anti- iCSC growth and anti-MYC/MYCN effects. This study would reveal specific pathways that regulate biological activities of the CSCs. In addition, it is a promising approach that could lead to development of safe and effective anti-cancer therapeutics not only for NB but also other cancer types. Citation Format: Naohiko Ikegaki, Kiira Ratia, Ruth Hsiao, Mariko M. Limpar, Sarah Lomahan, Xao X. Tang. Identification of lead organic compounds active against stem cell-like neuroblastoma cells by high throughput screening. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 3910. doi:10.1158/1538-7445.AM2014-3910

Cancer stem cells (CSCs) are thought to be present in malignant tumors at various proportions, and their phenotype can drift over time. Based on the current view, this elusive population of cancer cells is likely responsible for distant metastases, drug resistance and recurrence. However, the process by which such cell population is generated and maintained in a tumor mass has been unclear. Based on our previous study (PNAS Vol. 110: 6097-102 2013) and others, this process may involve the elevated expression of stem cell factors (SOX2, OCT4, NANOG, LIN28, KLF4, MYC, MYCN and those with equivalent functions). If so, it is conceivable that destabilization of these proteins would facilitate the eradication of CSC compartments in tumor mass. We have previously identified several small molecule compounds that confer the anti-growth effect on the stem cell-like neuroblastoma cells via screening of the Prestwick Chemical Library®, containing 1200 FDA-approved small molecules and of other selected experimental small molecules by using MTS assays. These compounds were structurally and biologically diverse molecules, including Alexidine (Alex), Benzethonium (Benz), Ketoconazole (Keto), Flubendazole (Flu), Auranofin (Aura), Nifuroxazide (Nif), Itraconazole (Itac) and JQ1. In this study, we investigated the effects of these selected small molecules on the expression of stem cell factors using the human teratocarcinoma cell lines NT2 and NCCIT as model systems, because these cells retain the expression of the majority of stem cell factors. The teratocarcinoma cell lines were treated with the small molecules at 0.5 to 10 μM for 24 hours and subjected to Western blot analysis. The expression of stem cell factors was down-regulated in NT2 cells by the several small molecules tested. The expression of SOX2 was down-regulated by Benz (5 μM), Flu (10 μM), Nif (10 μM), Aura (1 μM) and JQ1 (5 μM); the expression of OCT4 was down-regulated by Benz, Flu, Nif, Alex and JQ1; the NANOG expression was down-regulated by the all compounds tested; the expression of LIN28 was down-regulated by Benz, Flu and Aura; the expression of MYC/MYCN was down-regulated by Benz, Flu, Nif, Alex, Keto and JQ1. Interestingly, Auranofin treatment augmented the expression of MYC/MYCN, although the compound was growth suppressive to NT2 cells. Unlike the stem cell factors, the expression of LIN28B, which is the homologue of LIN28, was not affected by any of the compounds tested. A similar observation was made for NCCIT cells, but the effect of the compounds on the expression of stem cell factors appeared to be weaker. Collectively, the data suggest that small molecules identified via the process described could become useful lead chemicals for developing drug-like compounds that can target the most malignant stem cell-like cancer cells. Citation Format: Naohiko Ikegaki, Sarah Lomahan, Xao Tang. Small molecule compounds that destabilize stem cell factors. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 4215. doi:10.1158/1538-7445.AM2015-4215