Despite decades of investigation, the diagnosis of acute myeloid leukemia still carries with it a bleak prognosis, primarily as a result of the failure of clinical chemotherapy to target the leukemic stem cell population. In order to overcome this obstacle a library of derivatives of the small molecule parthenolide, a leukemic stem cell targeting natural product, were screened against acute myeloid leukemia (AML). Using the M9-ENL cell line as a high-throughput screening model for leukemic stem cells, over four hundred novel parthenolide derivatives were screened by flow cytometry for apoptotic activity. Using this method, more than forty compounds with improved efficacy against primary AML were identified, but the most remarkable compounds were those that were dimers of melampomagnolide B (MMB dimers), which had virtually no measurable toxicity to healthy blood stem and progenitor cells.

Despite their dramatic improvement in therapeutic index, these derivatives were only weakly associated with the leukemic stem cell targeting mechanism of the parent compound. Pharmacological investigation utilizing a chemoproteomic approach reveals that, unlike parthenolide, these bivalent MMB dimers potently and rapidly induce the depletion of nuclear galectin-1, a chemoprotective protein that is found to be overexpressed by more than two orders of magnitude in leukemic stem cells at diagnosis and relapse. In primary AML, nuclear monomeric galectin-1 is depleted within fifteen minutes of exposure to a 4 microM dose of MMB dimer JVM 3-88A. This discovery is significant both because galectin-1 has not previously been demonstrated to play a role in acute myeloid leukemia, and because to date, no clinical inhibitors of this emerging cancer target yet exist. Future investigations will seek to further understand the mechanism by which galectin-1 protects leukemic stem cells from apoptosis and how these compounds are able to alter its nuclear localization.