Abstract
Through both in vitro and in vivo validation studies, such as antisense RNA and gene knock-out experiments, DYRK3 has been implicated as a negative regulator of erythropoiesis. DYRK3, a member of the dual-specificity tyrosine phosphorylation-regulated kinase family, is expressed at low levels in erythroid progenitors and plays a regulatory role in their cellular proliferation and differentiation. High-throughput screening at GSK resulted in the identification of a novel series of thiazolidinone inhibitors of DYRK3. Lead optimization efforts then led to the discovery of GSK626616 as a potent, orally bioavailable inhibitor of DYRK3. GSK626616, inhibits DYRK3 in vitro with an IC50of 0.7 nM. This low molecular weight compound (401 Da) inhibits other members of the DYRK family, e.g., DYRK1A and DYRK2, with similar potency and with an approximate 20-fold selectivity versus the next most potently inhibited kinase, casein kinase 2. GSK626616AC, the meglumine, monohydrate salt form, has high oral bioavailability in all pre-clinical species studied (e.g. canine AUC = 23.64 ± 2.84 μghr/mL for a 30 mg/kg dose presented as crystalline material packed within a capsule). In cellular assays, GSK626616 enhances the number of CFU-E stimulated by Epo from human marrow, although it has no CFU-E activity on its own, consistent with the target’s functional role as a negative regulator. GSK626616 is specific for the erythroid lineage and does not stimulate CFU-GM colonies, either alone or in the presence of G-CSF or GM-CSF. There is also no direct effect on megakaryocyte colony growth from progenitor cells. 3H-thymidine incorporation in kit+ murine marrow is also stimulated after 3 days by exposure to GSK626616 in the presence of SCF and EPO. Similarly, 3 day treatment with GSK626616 increased the percentage, as well as the absolute number, of Ter119+/CD71+ erythroid progenitors derived from kit+ mouse marrow in the presence of SCF and Epo. GSK626616 (0.0001 to 30 uM) dosed i.p., daily for 14 days had no effect on the blood counts of normal mice, either alone or in the presence of 200 or 600 U/kg Epo. This was not unexpected based upon the very low levels of DYRK3 expression in the bone marrow of normal, non-anemic mice and on results previously reported for knock-out mice. These mice have normal erythropoiesis, but demonstrate an enhanced recovery from erythropoietic stress (Wojchowski, Blood 2005). Anemic mice, treated with a daily dose of GSK626616, i.p., had a statistically significant increase in hemoglobin compared to those dosed with vehicle alone at day 15 after anemia was induced through carboplatin/radiation treatment. Platelet levels were also elevated compared to vehicle control at day 15. These data suggest that treatment with GSK626616, through inhibition of DYRK3 activity, leads to an increase in the proliferation of kit+ cells producing an increased number of Ter119+/CD71+ erythroblasts, thereby accelerating recovery from the anemic insult in a carboplatin/radiation mouse model. It is hypothesized that this mechanism may only function under anemic conditions when DYRK3 is elevated and therefore, that the effects will be self-regulated as hemoglobin and EPO levels approach the normal range.
