A Nonpeptidyl, Hydrazone Class, Thrombopoietin Receptor Agonist (TpoRA), SB559457, Selectively Kills Primary Human Myeloid Leukemia Cells

Treatment of thrombocytopenia, whether autoimmune, secondary to marrow failure, or iatrogenic, can become problematic. Stimulation of endogenous megakaryocyte (MEG) development might be an effective strategy and it was originally hoped that this could be accomplished with thrombopoietin (Tpo), the natural ligand of the thrombopoietin receptor (TpoR) and regulator of MEG development. When formulated Tpo proved immunogenic, development of small molecule TpoR agonists (TpoRA) was undertaken. In course of investigating the biological effects, and mechanism of action, of SB559457 (a nonpeptidyl, hydrazone class, TpoRA supplied by GlaxoSmithKline (GSK), Collegeville, PA) we found that the molecule robustly stimulated human megakaryocytopoiesis in vitro. In anticipation of TpoRA use in the treatment of patients with hematologic malignancies, we also evaluated SB559457’s effect on human leukemia cell (HLC) growth in vitro since it is known that HLCs may express TpoR. Specifically, we examined the effects of SB559457 on samples obtained from 21 patients with AML, and 7 with ALL. Surprisingly, SB559457 not only failed to stimulate HLC growth, it proved toxic to primary acute myelogenous leukemia (AML). In 20 of 21 primary AML samples exposed to SB559457 (5μM) a large fraction (70–90%) of cells died between days 3 and 6 of culture, while untreated control cells survived. No significant effects on cell growth or viability were observed in the ALL patient samples. To investigate the mechanism(s) of AML cell killing, we examined signaling cascades initiated by SB559457 compared to recombinant human Tpo (rhTpo), which had no apparent effect on AML cell growth. Initially, we employed N2C-Tpo cells (a Tpo dependent megakaryoblastic cell line that express TpoR) and focused on kinases known to be phosphorylated after TpoR activation; STAT5, ERK, p70S6, and ribosomal kinase S6. When N2C-Tpo cells were stimulated with rhTpo (2.8μM, 30 min) all these kinases were highly phosphorylated. In contrast, none were phosphorylated in SB559457 (5μM, 30min) stimulated cells. Normal human CD34+ marrow cells were next examined using the same experimental conditions. In contrast to N2C-Tpo cells, ERK, and p70S6 kinases were both phosphorylated after exposure to SB559457 but STAT5 remained unphosphorylated. Since AML cells may require STAT5 activation for growth, we hypothesized that perturbation of STAT5 activation might be involved in the apoptotic mechanism. The molecular consequences of differential signaling were then pursued by Affymetrix GeneChip analysis. Remarkably, in 5 separate primary AML cell samples stimulated for 6 hours with either Tpo or SB559457 we found a statistically significant difference in expression in only 2 of 22000 genes represented on the chips: GAPDH and Redd1. Both of these genes are induced in stressed cells undergoing apoptosis. In addition, stimulation of primary AML cells with SB559457 resulted in increased phosphorylation of p70S6/S6 kinases, both downstream targets of mTOR kinase. Accordingly, we hypothesize that SB559457 mediated activation of mTOR pathway leads to a stress response in primary myeloid leukemia cells as reflected by the increased expression of Redd1 and GAPDH. Further investigation of this pathway, and the leukemic cell response to SB559457 exposure may lead to development of novel strategies for the treatment of myeloid leukemias.