Erk and Stat5 Feedback Control Enables Pre-B Cell Transformation and Represents a Therapeutic Target in Acute Lymphoblastic Leukemia

Background and Hypothesis: Targeted therapy of cancer typically focuses on the development of agents that will inactivate a transforming oncogene. In this study, we tested the concept that besides the oncogene itself, factors that enable permissiveness of a normal cell to oncogenic signaling represent a novel class of therapeutic targets. This hypothesis was based on three findings. First, acute activation of oncogenes in normal pre-B cells typically caused immediate cell death, unless pre-B cells were capable of adapting quickly enough to a high level of signaling output. Second, few surviving pre-B cell clones achieved permissiveness to oncogenic signaling by strong activation of negative feedback control of Erk and Stat5. Third, robust feedback control of Erk and Stat5 distinguishes normal pre-B cells from fully transformed pre-B acute lymphoblastic leukemia (ALL) cells.

Results: To test the significance of strong feedback control of Erk and Stat5 signaling in pre-B ALL cells, we developed genetic loss-of function models for six central molecules in Erk (DUSP6, SPRY2, ETV5) and Stat5 (Cish, SOCS2, SOCS3) feedback control. Genetic deletion of the sprouty family Ras inhibitor Spry2, the Erk dual specificity phosphatase Dusp6 and their transcriptional activator Etv5, decreased robustness of Erk feedback control and compromised oncogenic transformation in mouse models for pre-B ALL. Likewise, ablation of Stat5 feedback control through deletion of the suppressors of cytokine signaling (SOCS) family molecules Cish, Socs2 and Socs3 reversed permissiveness of pre-B cells. Studying deletion of Spry2 (Erk) and Cish (Stat5) in an in vivo transplant model using inducible, Cre-mediated deletion of Spry2 and Cish in pre-B ALL cells confirmed that Erk and Stat5 feedback control are essential for malignant transformation and development of lethal leukemia. Genetic deletion of Erk (Dusp6, Spry2, Etv5) and Stat5 (Cish, Socs2, Socs3) feedback control impairs leukemic transformation of pre-B cells.

Searching for factors that restrict permissiveness to oncogene signaling, we identified the pre-B cell tumor suppressor IKZF1, which is deleted in a large fraction of pre-B ALL cases. IKZF1 directly bound to and transcriptionally repressed multiple promoters of Erk and Stat5 feedback control and IKZF1 deletion raised the limit of maximum allowable oncogene signaling strength in pre-B ALL cells. We propose that the pre-B cell tumor suppressor IKZF1 functions as transcriptional repressor of Erk and Stat5 feedback control and thereby retains pre-B cells in a Non-permissive state.

Clinical relevance: To assess potential usefulness of this finding for the development of future treatment strategies, we tested the effect of a specific small molecule inhibitor of DUSP6, E-2-benzylidene-3-(cyclohexylamino)-2,3-dihydro-1H-inden-1-one (BCI), which was designed as an allosteric inhibitor of the interaction between DUSP6 and phospho-ERK1/2. Interestingly, BCI acutely subverted Erk feedback control and selectively induced cell death in pre-B ALL cells. Small molecule inhibition of DUSP6 was sufficient to overcome conventional mechanisms of drug-resistance in pre-B ALL and selectively killed patient-derived pre-B ALL cells in a leukemia transplant model. BCI treatment, similar to Dusp6-deletion in our leukemia mouse model, led to the accumulation of P53 and ARF in patient-derived pre-B ALL cells. In addition, small molecule inhibition of DUSP6 had strong selective activity on drug-resistant patient-derived pre-B ALL cells that were injected into NOD/SCID transplant recipient mice. These findings identify permissive negative feedback control of oncogenic signaling as a previously unrecognized vulnerability of pre-B ALL cells and a new class of potential therapeutic targets.

Conclusion: Targeting negative feedback regulation of both Erk and Stat5 signaling for the treatment of pre-B ALL seems counter-intuitive because it represents effectively the opposite of current efforts of targeted inhibition of oncogenic signaling. Our results, however, demonstrate that a robust negative feedback regulation is required for the leukemic transformation and development of fatal leukemia in pre-B ALL. We demonstrate that feedback control of Erk and Stat5 signaling represents a previously unrecognized vulnerability and, potentially, a novel class of therapeutic targets in human pre-B ALL.