Abstract
Acute Myeloid leukemia (AML) has a poor prognosis in part due to lack of elimination of leukemic cells and acquisition of drug resistance. In particular mutations in the activation loop of KIT and internal tandem duplication (ITD) in Flt3 receptor are associated with poor prognosis in AML. Interestingly, more than 90% patients with aggressive systemic mastocytosis (ASM) carry the KITD816V mutation which is highly resistant to conventional therapeutics and invariably associated with poor survival. Since the efforts to target the kinase activity of FLT3 and KIT have met with little success; the focus has shifted to targeting critical downstream signaling regulators. We have elucidated the role of a novel downstream regulator of FLT3-ITD and KITD816V mediated leukemic transformation - LIM kinase and evaluated the efficacy of a novel inhibitor of LIM kinase (LIMK), Liminib. LIMK, a serine/threonine kinase has 2 isoforms; LIMK1 and LIMK2. They are substrates of the Rho associated kinase (ROCK) which we have shown to play a critical role in leukemic transformation (Cancer Cell, 2011). They in turn act on cofilin and phosphorylation by LIMK inactivates cofilin and thereby modulates F-actin reorganization. Western blot analysis showed marked increase in phospho cofilin levels in cells expressing FLT3-ITD or KITD814V oncogenes as compared to their WT counterparts. Dependence of cofilin phosphorylation on activation of ROCK and LIMK was confirmed by treating the cells with an inhibitor of ROCK, H1152 and Liminib, inhibitor of LIMK. Both the inhibitors effectively reduced oncogene-induced phospho-cofilin levels. Liminib, the LIMK inhibitor also inhibited cytokine independent proliferation driven by FLT3-ITD or KITD814V (>90% at 100 nM). It also showed concentration dependent inhibition of proliferation in mutant oncogene bearing patient derived human cell lines such as MV411, Kasumi and HMC. FLT3-ITD expressing leukemic cells become resistant to Flt3 kinase inhibitor AC220 by acquiring additional point mutations. Interestingly, the cells expressing AC220 resistant FLT3-ITDD835F or FLT3-ITDF691L mutation showed similar sensitivity to Liminib as cells with FLT3-ITD. The growth inhibitory effect of Liminib was also validated in primary blasts from AML and SM patients. To further elucidate the mechanism of growth inhibition by Liminib in leukemic cells, we assayed for apoptosis induction in the treated cells. Cells expressing FLT3-ITD were significantly more susceptible (~ 73%) to induction of apoptosis as compared the WT-FLT3 cells (~33%) on treatment with Liminib. The extent of apoptosis induction by Liminib was similar in FLT3-ITD and AC220 resistant FLT3-ITDD835F or FLT3-ITDF691L mutation bearing cells. In addition to the effects of Liminib on F-actin polymerization; experimental evidence indicates involvement of mitochondrial pathway in the induction of apoptosis by Liminib. There was increase in accumulation of cofilin in the mitochondrial fraction in Liminib treated oncogene expressing cells as compared to the untreated cells. Phosphorylation of cofilin on Ser3 suppresses its mitochondrial translocation and treatment with LIMK inhibitor would inhibit cofilin phosphorylation leading to increased mitochondrial transport. Mitochondrial transport of cofilin has been observed with other inducers of apoptosis such as staurosporine and it possibly contributes to mitochondrial dysfunction during induction of apoptosis. To validate our pharmacologic findings using Liminib genetically and to assess which of the 2 isoforms of LIMK contributes to leukemogenesis, we generated LIMK1 and LIMK2 single and double knockout mice. 3 H-thymidine incorporation assay showed that deletion of either of the LIMK significantly reduced KIT D814V induced growth factor independent proliferation by approximately 50%. The extent of inhibition was more pronounced in Limk1-/- Limk2+/- or Limk1+/- Limk2-/- bone marrow cells (~80%) as compared to cells with deletion of a single isoforms of LIMK. Thus, there is some functional redundancy between the 2 isoforms and inhibition of both is likely to have better therapeutic value. Taken together we have identified a novel signaling pathway involving LIMK and cofilin that appear to be selectively activated in regulating oncogenic signaling in AML and MPN but not in normal hematopoiesis.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.
This feature is available to Subscribers Only
Sign In or Create an Account Close Modal