DEVELOPMENT OF NOVEL SMALL MOLECULES THAT TARGET GLIOBLASTOMA STEM CELLS

2014 
BACKGROUND: Transcription factors (TFs) are a major class of protein signaling molecules that play a critical role in most cancers. OLIG2 is a bHLH TF essential for survival and expansion of the highly aggressive brain cancer, glioblastoma (GBM), and represents an attractive therapeutic target. TFs including OLIG2 are typically activated by dimerization, and TF inhibition has proved problematic owing to expansive protein–protein interfaces and the absence of hydrophobic pockets. In silico modeling is increasingly being used in attempts to design TF dimerization inhibitors, but these efforts have met with very limited success. METHODS: evious in silico based drug design approaches for TFs focused on single residues or small foci, called binding hotspots, on TF dimerization surfaces. This approach has largely failed because hotspots do not adequately represent the total active TF dimerization interface. In our modeling approach, which we used to identify candidate small molecule scaffolds for OLIG2 inhibition, we represent the dimerization surface as a comparatively extensive parental pharmacophore. This active surface is comprised of multiple, specific subregions we term daughter pharmacophores or subpharmacophores. We hypothesized that a small molecule capable of binding each subpharmacophore would sufficiently populate the active dimerization surface to interfere with OLIG2 dimerization, thereby suppressing pathway activation. RESULTS: Computational screens, guided by parameters defined by our multiple pharmacophore algorithm, identified potential OLIG2 inhibitors from comprehensive in silico compound libraries. A subset of these candidates convincingly demonstrated OLIG2 pathway inhibition and anti-GBM activity in an array of biochemical, cell-based, and reporter assays. Further, when we tested in mice a leading representative from a chemically tractable structural class we found, (1) attenuation of GBM xenograft growth, and (2) a favorable CNS pharmacokinetic profile. Both observations prompted us to nominate and pursue the representative for further structural optimization in the context of in vivo efficacy and pharmacokinetics. CONCLUSIONS: We have developed a novel computational modeling approach for designing TF inhibitors, using the concept of multiple pharmacophores. We have identified a small molecule compound with significant in vitro/vivo anti-GBM activity and favorable pharmacokinetics. The initial compound is now being pursued as a development candidate for GBM, and if successful its final derivative may ultimately represent the first truly GBM-specific drug. Moreover, our study in a broader context presents a new pharmacologic paradigm and may pave the way for the development of TF-targeted therapeutics in general. SECONDARY CATEGORY: n/a.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    0
    References
    1
    Citations
    NaN
    KQI
    []