Abstract B236: A small molecule disrupts the BCL6 transcriptional repressor complex and kills lymphoma cells in vitro and in vivo

Abstract The BCL6 transcriptional repressor is the most frequently involved oncogene in diffuse large B-cell lymphoma (DLBCL) and is also implicated in solid tumors. BCL6 depletion by siRNA or BCL6 blockade using peptide inhibitors kills DLBCL cells, demonstrating that lymphoma cells are addicted to BCL6. Based on these data we hypothesized that BCL6 could serve as an excellent therapeutic target in lymphomas and other cancers. From the mechanistic standpoint BCL6 mediates transcriptional repression through recruitment of the SMRT, N-CoR and BCOR corepressors to a lateral groove motif on its BTB domain. Our crystallographic, biochemical, and functional studies suggested that a particular section of the BTB lateral groove was amenable to development of small molecule inhibitors. Since these residues are unique to BCL6 and no other BTB proteins we predicted that molecules designed to bind to this region would be specific. To identify such molecules we performed a computer-aided drug design (CADD) screen of over 1,000,000 compounds. After two rounds of docking with increasing stringency, target flexibility simulation and chemical diversity clustering, we generated a priority list of compounds likely to bind the BCL6 BTB lateral groove. A number of these compounds indeed blocked the BCL6 - corepressor interaction and inhibited the transcriptional repressor function of BCL6. Analogs of the 10 leading compounds were subjected to biological testing. One of these families, the 79 series, was selected for further study because this group contained the largest number of active compounds. X-ray crystallography of the most active 79 family compound, called 79-6, confirmed binding to the critical BTB lateral groove region. 79-6 penetrated and accumulated in intact lymphoma cells as determined by HPLC-MS. 79-6 blocked the repressor activity of BCL6 but not other BTB proteins, blocked the recruitment of corepressors to endogenous BCL6 target genes as shown by ChIP, and induced the re-expression of BCL6 target genes in BCL6-positive (but not BCL6 negative) lymphoma cells as shown by q-PCR. Most importantly, 79-6 killed BCL6-positive lymphoma cells by apoptosis induction but had no effect on BCL6-negative lymphoma cells. Extensive toxicological studies performed in C57/BL mice revealed no toxic effects (by complete serum chemistry, CBC, histopathological examination of every organ including bone marrow, etc). Pharmacokinetic analysis (by HPLC-MS/MS) in SCID mice showed that 79-6 penetrates and accumulates in tumors after intraperitoneal administration. 79-6 potently inhibited the growth of established human BCL6-positive xenografts (Ly7 and SUDHL6) but not of BCL6-negative tumors (Toledo), again without toxicity to other organs. The compound also killed primary BCL6-positive DLBCL explants from human patients. 79-6 is currently being optimized for used in humans. These results demonstrate that therapeutic chemical manipulation of transcriptional repression via modulation of protein-protein interactions is a viable option for development of potent and non-toxic anti-cancer drugs. Citation Information: Mol Cancer Ther 2009;8(12 Suppl):B236.