Validated high-throughput screening of drug-like small molecules for inhibitors of ErbB2 transcription.

A whole cell high-throughput screening assay was developed and tested against > 2,000 structurally and functionally diverse drug-like small molecules to identify lead compounds capable of cell permeability and selective silencing of ErbB2 transcription. Screening employed reporter sublines clonally selected from ErbB2-negative MCF7 breast cancer cells after stable genomic integration of the ErbB2 proximal promoter driving a luciferase reporter; anti-ErbB2 activities (50% inhibitory concentration values) were compared to inhibition of control MCF7 sublines bearing integrated reporters driven by either a mutated ErbB2 promoter or the cyclin D1 promoter. Of the seven resulting lead compounds, four emerged from the National Cancer Institute (NCI)/ Developmental Therapeutics Program (DTP) Structural Diversity Set (NSC-131547, NSC-176328, NSC-259968, and NSC-321237); three others emerged from a panel of anticancer compounds with known mechanistic actions and included a minor groove DNA-binding antibiotic (NSC-58514, chromomycin A3), a hydroxamic acid inhibitor of histone deacetylases (NSC-709238, trichostatin A), and a tripeptide aldehyde proteasome inhibitor (MG-132). For optimization, 58 scaffold analogs of the four NCI/DTP structural leads and nine functional analogs of the mechanistic leads were secondarily screened to identify seven compounds with comparable or superior activity relative to the leads, including an approved anticancer drug, PS-341 (bortezomib). PS-341 activity was validated against cultured ErbB2-positive breast cancer cell lines (SKBr3 and BT474) and a trastuzumab-resistant ErbB2-positive breast cancer xenograft model (B585), in which PS-341 antitumor activity correlated with selective down-regulation of ErbB2 mRNA and protein levels, confirming the ErbB2- silencing potential of proteasome inhibitors.