Abstract
The abl tyrosine kinase inhibitor (TKI) imatinib mesylate has revolutionized treatment of Chronic Myelogenous Leukemia (CML). Nonetheless, for the small percentage of chronic phase patients in whom resistance to imatinib develops, and for patients in accelerated phase/blast crisis, disease management is problematic. Second generation TKI may address the resistance issue, but serious cardiotoxicity might be a concern for all TKIs. We are therefore investigating other candidates for rationally directed CML therapy. Herein we report that targeting tubulin with computationally designed small molecules may prove useful for the treatment of TKI resistant CML cells. Tubulin inhibitors were obtained from Locus Pharmaceuticals, Blue Bell, PA. They were designed using a method that computes virtual inhibitor molecule binding efficiency after integrating free energy calculations from all chemically possible combinations of molecule fragment poses. Molecules that perform well in in silico screens are then synthesized for biological testing. We evaluated 3 such molecules (LP-261, LOC-011294, and LOC-011423), and 2 control compounds (LOC-007708, LP-590), on K562 human leukemia cells, murine BaF3 cells expressing wild type bcr-abl, or the Y253F, T315I, E255K, H296P and M351T kinase domain mutations (gift from B. Druker, Portland, Oregon), and on consenting donors of normal, and CML, bone marrow cells. LP-261, LOC-011294, and LOC-011423 all had significant activity in K562 cells and BaF3 cells expressing wild type bcr-abl. Inhibition of cell growth in these lines was ~90% when employed at concentrations of > 100nM. More importantly in BaF3 cells expressing each of the mutant abl kinases, including T315I, growth inhibition was also ~90%. Moreover, LP-261, and LOC-011294 were also highly effective against primary cells obtained from patients with chronic phase and blast crisis CML. Treatment of primary CML cells with LP-261 resulted in >80% inhibition of proliferation in all five CML patient samples when compared to control cells. LOC-011294 inhibited cell proliferation by >80% in 3 out of 5 primary patient samples, and by ~50% in one other. Additional testing revealed that LP-261 was not a substrate for the p-glycoprotein multi-drug resistance porter, and that it is orally bioavailable. Neither LOC-007708, (a selective inhibitor of p38 kinase), nor LP-590 (an inhibitor of p38, Flt-3 and tie-2 kinases) inhibited proliferation of the cells expressing mutated bcr-abl. Since these kinases are not thought to influence bcr-abl driven cell growth, these results were expected. In contrast, MOLM14 cells, which overexpress Flt-3, were profoundly inhibited by LP-590. Disease specificity is suggested by the fact that none of these compounds had any effect on growth of acute lymphoid leukemia (ALL) patient samples. To be clinically useful, it was important to demonstrate that normal CD34+ were less sensitive to the growth inhibitory effects of these compounds. For this purpose, normal CD34+ cells were exposed to LP-261, LOC-011294, and LOC-011423 for 24 hours at concentrations shown to inhibit CML cell growth and then plated in methylcellulose with cytokines. Under these conditions, no significant inhibition of CFU-GM, CFU-E, or BFU-E in comparison to control cells was shown. These results suggests that rationally designed anti-tubulin small molecules, alone, or in combination with other active agents, may prove quite useful for treating kinase inhibitor resistant, as well as de novo, CML. This hypothesis may be tested in the near future as an IND has been filed for LP-261.
Growth of pancreatic cancers with hemizygous chromosomal 17p loss of MYBBP1A can be preferentially targeted by PARP inhibitors
