Metformin Inhibits STAT3 and MAPK Signaling through AMPK Activation in CSF3RT618I Cells

Abstract

Introduction: In the light of the discovery of the CSF3R mutation in chronic neutrophilic leukemia (CNL) patients, the T618I mutation demonstrated to be a molecular marker and a possible target for new therapies due to the JAK/STAT signaling pathway activation [N Engl J Med (2013) 368(19):1781-90]. Since no curative treatment options are available for patients diagnosed with CNL, it is of interest the study new therapeutical approaches for these patients. In this regard, the treatment of JAK2V617F murine models with metformin, an FDA-approved drug for type 2 diabetes, reduced splenomegaly, extramedullary hematopoiesis, and colony formation of HSPC [Cell Death Dis. 2018;9(3):311]. Hence, we characterized the preclinical cellular and molecular effects of metformin in CSF3RT618I models. Methods: The in vitro experiments were performed in Ba/F3 cells transduced retrovirally with CSF3RT618I overexpression (OE) model. Cells were treated with metformin (1-20 mM) to access cell viability and survival. Protein expression was evaluated by western blot. For in vivo evaluation, eight-week-old female NSG mice were transplanted with 2x106 Ba/F3 CSF3RT618I cells, resuspended in Matrigel: PBS (1:1 ratio). After the tumor formation (2 days), mice were treated for 10 days by intraperitoneal injection of vehicle (PBS) or metformin (125 mg/kg, n=10/group). Tumor volume was measured using the formula (V = W2 × L × 0.52), where W and L are the smallest and largest diameters, respectively. Ex vivo primary cells from a patient diagnosed with CNL harboring the CSF3RT618I mutation were used to assess the effects of metformin treatment. The primary bone marrow mononuclear cells were evaluated regarding cell survival and clonogenic capacity upon metformin treatment. All experiments were approved by the Animal and Human Ethics Committee of the Institution. Statistical analyses were performed by Mann-Whitney or ANOVA test and Bonferroni post-test. A p-value < 0.05 was considered. Results: After 24 and 48 hours, Ba/F3 CSF3RT618I treated with 1, 5, 10 and 20 mM of metformin presented a reduction in cell viability in a dose- and time-dependent manner (p < 0.01). The IC50 for metformin treatment for 24, 48, and 72 hours was 17.44, 1.3, and 3.2 mM, respectively. Increased frequency of apoptotic cells was observed at 5, 10, and 20 mM of metformin at times of 24, 48, and 72 hours (p < 0.01). Molecularly, CSF3RT618I cells displayed reduced STAT3 and ERK phosphorylation and increased AKT/mTOR activation, upon metformin 5 mM treatment for 24 hours. AMPK activation was observed in cells treated with 1-20 mM of metformin compared to vehicle. In vivo, metformin treatment decreased the tumor burden of CSF3RT618I transplanted mice after 9 days of treatment (vehicle versus metformin mean ± SEM: 731.9±90.14 vs. 481.6±38.14; p = 0.035). At the protein level, we observed inhibition in Cyclin D1 and the presence of caspase 3 cleavage in the isolated tumor from the metformin-treated mice. Ex vivo metformin treatment of the primary cells isolated from the CNL patient harboring CSF3RT618I resulted in increased apoptosis, with 51.94% (p < 0.05), 52.40% (p < 0.05), and 63.1% (p < 0.001), of apoptosis induction upon treatment with 5, 10 and 20 mM. Metformin ex vivo treatment strongly reduced hematopoietic progenitor cells colony formation on primary CNL CSF3RT618I peripheral blood (83.5% at 10 mM and 100% at 20 mM) and bone marrow mononuclear cells (100% at 10 mM and 98% at 20 mM). Conclusions: We provide evidence that metformin exerts anticancer activity in CSF3RT816I cells, by downregulation of STAT3 and MAPK signaling through AMPK activation. These results support the idea that metformin can be repurposed in the CNL clinics, as a therapeutical option.