Search

Overexpression of the anti-apoptotic Bcl-2 family proteins occurs commonly in human cancers. Bfl-1 is highly expressed in some types of malignant cells, contributing significantly to tumor cell survival and chemoresistance. Therefore, it would be desirable to have chemical antagonists of Bfl-1. To this end, we devised a fluorescence polarization assay (FPA) using Bfl-1 protein and fluorescein-conjugated Bid BH3 peptide, which was employed for high-throughput screening of chemical libraries. Approximately 66 000 compounds were screened for the ability to inhibit BH3 peptide binding to Bfl-1, yielding 14 reproducible hits with ≥50% displacement. After dose-response analysis and confirmation using a secondary assay based on time-resolved fluorescence resonance energy transfer (TR-FRET), two groups of Bfl-1-specific inhibitors were identified, including chloromaleimide and sulfonylpyrimidine series compounds. FPAs generated for each of the six anti-apoptotic Bcl-2 proteins demonstrated selective binding of both classes of compounds to Bfl-1. Analogs of the sulfonylpyrimidine series were synthesized and compared with the original hit for Bfl-1 binding by both FPAs and TR-FRET assays. The resulting structure-activity relation analysis led to the chemical probe compound CID-2980973 (ML042). Collectively, these findings demonstrate the feasibility of using the HTS assay for discovery of selective chemical inhibitors of Bfl-1.

Nuclear receptor TR3/Nur77/NR4A1 binds several antiapoptotic Bcl-2-family proteins (Bcl-B, Bcl-2, Bfl-1) in a non-BH3-dependent manner. A 9-amino-acid peptide derived from full-length TR3 with polyarginine tail (TR3-r8) recapitulates TR3's binding specificity, displaying high affinity for Bcl-B. TR3-r8 peptide was used to screen for small molecule Bcl-B inhibitors. A fluorescence polarization assay (FPA) employing fluorescein isothiocyanate (FITC)-labeled TR3-r8 peptide (FITC-TR3-r8) and Bcl-B protein was optimized, with nonfluorescent TR3-r8 serving to demonstrate reversible, competitive binding. Approximately 50,000 compounds were screened at 3.75 mg/L, yielding 145 reproducible hits with > or =50% FITC-TR3-r8 displacement (a confirmed hit rate of 0.29%). After dose-response analyses and counterscreening with an unrelated FITC-based FPA, 6 candidate compounds remained. Nuclear magnetic resonance (NMR) showed that 2 of these compounds bound Bcl-B, but not glutathione S-transferase (GST) control protein. One Bcl-B-binding compound was unable to displace FITClabeled BH3 peptides from Bcl-B, confirming a unique binding mechanism compared with traditional antagonists of antiapoptotic Bcl-2-family proteins. This compound bound Bcl-B with Kd 1.94 +/- 0.38 microM, as determined by isothermal titration calorimetry. Experiments using Bcl-B overexpressing HeLa cells demonstrated that this compound induced Bcl-B-dependent cell death. The current FPA represents a screen that can identify noncanonical inhibitors of Bcl-2-family proteins.