Pan-Selectin Inhibitor GMI-1070 Blocks Cell Adhesion of Acute Myelogenous Leukemia Cell Lines

Most patients with acute myelogenous leukemia (AML) initially respond to chemotherapy but later relapse and die from the disease (50–70%). One mechanism by which AML cells resist treatment with anti-proliferative drugs is by cell adhesion in protective microenvironments in areas such as the bone marrow. Here, we investigate the effects of GMI-1070 on the mechanism of adhesion of an AML cell line (MV-4-11) derived from biphenotypic myelomonocytic cells. GMI-1070 is a small molecule rationally designed pan-selectin antagonist with particularly strong activity for E-selectin (IC50 = 3.4 μM). Over 95% of these AML cells express ligands for E and P-selectins as determined by fluorescence-activated cell sorter analysis using E or P-selectin/hIg chimeras. Virtually all of these cells strongly express the specific E-selectin ligands CD65 and FH-6 using flow cytometry. The ability of these AML cells to roll and adhere to E or P-selectin expressed on monolayers of human endothelial cells (HUVECs) under the shear forces of normal blood flow was determined in vitro using flow chambers. The data was captured by videomicroscopy and processed using digital imaging. AML cells rolled and adhered to endothelium expressing E or P-selectin and both forms of adhesion are inhibited by GMI-1070. In contrast, much less rolling and adhesion was observed by multiple myeloma cell line U-266 only on endothelial cells expressing P-selectin although these specific interactions are also inhibited by GMI-1070. To analyze the adhesion, homing and infiltration of this AML cell line in vivo, recently developed NOD-scid IL2rg^null mice were used as hosts for in vivo experimentation. These mice lack adaptive immune function, are completely deficient in host natural killer cells and support long term growth of primary human leukemias (AML, CML, and ALL). Mice were injected i.v. with MV-4-11 AML cells (1 × 10⁶) and blood samples were taken from the retro-orbital plexus at 1 and 2 weeks post injection. Blood counts were performed using an ADVIA Hematology Analyzer. At intervals post injection, the presence of AML cells in the bone marrow (BM), spleen, and peripheral blood was determined by flow cytometry of cells co-expressing human CD45 and human CD33. At 3 weeks, AML cells represented 26% of cells in the bone marrow and 2.3% of cells in the spleen. At 6 weeks, high percentages of AML cells were found in both tissues as well as in the blood (bone marrow, 83%; spleen, 69%; blood, 36%). Histological analyses showed accumulations of AML cells in the BM, spleen, lungs, liver, kidneys, and ovaries. At these later stages, hematopoietic cells in the BM cavity were largely replaced by sheets of leukemic cells. This model proved to be effective in studying the homing of an AML cell line to protective environments in the bone marrow and dissemination of disease to other organs and is being used to determine the effects on cell adhesion-mediated drug resistance (CAM-DR) by selectin-mediated inhibition of adhesion using small molecule antagonist, GMI-1070. GMI-1070 is currently in Phase I clinical trials.