In AML, Cbl Serves as Nexus for Signaling from Flt3, and Is Required for Coupling JNK1 in a Pathway of Survival and Proliferation Involving c-jun/AP-1.

Here we demonstrate by immunoprecipitation and immunoblot, cbl is among the most heavily tyrosine phosphorylated adaptor proteins in primary AML blasts with Flt3 signaling, in the context of either mutation or overexpression/autocrine mechanisms. The human leukemic cell lines MV-4-11 and THP-1 model primary AML blasts in terms of Flt3 signaling by these respective criteria and demonstrate identical coupling between Flt3 and p85, the PI-3-kinase adaptor, by coimmunoprecipitation/blot experiments. Although cbl has no direct binding site on Flt3, it binds tightly to p85 SH2 by virtue of its tyrosine phosphorylation, also demonstrated by co-IP in cell lines and primary cells. Tyrosine phosphorylated cbl is a docking site for CrkII/L SH2’s and this provides a branch point for signals from Flt3 to PI-3-kinase or JNK, respectively, because CrkII(L) is known to bind JNK1 through SH3: polyproline interaction to serve as scaffolding; and interaction of JNK1 and CrkII/L was also observed by co-IP. In a survey of primary AML cases (n=33) there was a strict relationship between expression levels of (active) Flt3 and phospho-c-jun as readout for JNK activity level (p=0.001, r=0.54). To demonstrate the functional relevance of these interactions, siRNA knockdown of components was pursued in the cell lines and in primary AML blasts. JNK1 knockdown, and, to a much lesser degree, JNK2 knockdown, led to loss of phospho-c-jun expression in MV-4-11 and THP-1. Indeed, Flt3 signaling is required for JNK signaling because knockdown of Flt3 led to total loss of p-jun and c-jun expression in MV-4-11 and patient blast. By contrast, cbl knockdown led to selective loss of JNK signaling to p-jun without significantly affecting Flt3 or its downstream activating phosphorylation of AKT. Thus, despite binding by cbl to p85, cbl is not required for PI-3-kinase signaling because of redundancy supplied by the p85-Flt3 interaction. Further, by use of LY294002 to inhibit PI-3-kinase, PI-3-kinase is also not required for JNK signaling. However, selective inhibition of JNK signaling by the small molecule approach in these cell lines and in primary AML blasts leads to loss of proliferation, induction of apoptosis, and synergistic killing with daunorubicin. These observations form the platform for a phase I trial of JNK inhibition in refractory, multidrug-resistant, and Flt3-driven AML.