Activation of the STAT1-BCL-2/MCL-1 Axis in Leukemic Cells Carrying a SPAG9-JAK2 Fusion

Abstract [Background and aim of this study] Philadelphia chromosome-like acute lymphoblastic leukemia (Ph-like ALL) is a distinct subtype of B-ALL with poor prognosis. JAK2 related fusion genes have been identified in this subtype, especially in adolescent and young adults (AYA). Previously, we identified SPAG9-JAK2 fusion gene in 14-year-old boy with Ph-like ALL (Kawamura M, et al. Genes Chromosomes Cancer. 2015). In this study, we performed functional analysis of the SPAG9-JAK2 fusion protein, and evaluated the efficacy of treatment with a JAK inhibitor against cells carrying the fusion. In addition, we assessed therapeutic options other than JAK2 inhibition following comprehensive molecular analysis. [Materials and Methods] Full length of SPAG9-JAK2 cDNA was cloned into retroviral construct with Tet-On system. Ba/F3 cells, which are IL-3 dependent murine pro B-ALL cells, were transduced with retroviral vector to establish Ba/F3 cells expressing SPAG9-JAK2 (Ba/F3-SPAG9-JAK2) under doxycycline (DOX) dependent manner. Ba/F3-SPAG9-JAK2 were analyzed whether IL-3 independent growth was achieved. Aberrant activation of JAK-STAT pathway achieved by SPAG9-JAK2 was evaluated by western blot. To clarify whether the tyrosine residues of JAK2 in this fusion protein were critical for IL-3-independent proliferation, Ba/F3 cells expressing SPAG9-JAK2 mutants (SPAG9-JAK2 mut) in which both tyrosine residues of JAK2 were replaced with phenylalanine was established. Gene expression analysis using Mouse Genome 430 2.0 Array was performed for comprehensive analysis of gene expression profile related to SPAG9-JAK2. Sensitivity of Ba/F3-SPAG9-JAK2 to ruxolitinib (a JAK inhibitor) was tested in cytotoxic assay and in xenograft model established using Ba/F3-SPAG9-JAK2 cells. [Results and discussions] The expression of SPAG9-JAK2 in Ba/F3 cells under DOX dependent manner was confirmed by western blot. Ba/F3-SPAG9-JAK2 proliferated without IL-3 in contrast to Ba/F3 cells (p<0.01), suggesting SPAG9-JAK2 had the proliferation activity. Western blot revealed that constitutive phosphorylation of tyrosine residue of SPAG9-JAK2, STAT3/STAT5, suggesting constitutive activation of JAK2-STAT3/STAT5 pathway. SPAG9-JAK2 mut abolished IL-3 independence (p<0.01), but had no influence on STAT3/STAT5 phosphorylation levels detected by western blot. Gene expression analysis revealed that Stat1 was significantly up-regulated in Ba/F3-SPAG9-JAK2 cells compared with mock Ba/F3 cells (fold change 8.04 with p < 0.01) [Fig. 1], confirmed by western blot. STAT1 was also phosphorylated in Ba/F3-SPAG9-JAK2 but not SPAG9-JAK2 mut cells detected by western blot [Fig. 2], suggesting that STAT1 is a key mediator for SPAG9-JAK2-mediated cell proliferation. Consistently, STAT1 induced expression of the anti-apoptotic proteins, BCL-2 and MCL-1, as did SPAG9-JAK2, but not SPAG9-JAK2 mut confirmed by western blot [Fig. 3]. Ruxolitinib abrogated Ba/F3-SPAG9-JAK2-mediated proliferation in vitro (p<0.01), with an 50% inhibitory concentration (IC50) value of 65.9 ± 9.8 nM, causing decrease of JAK2, STAT1/STAT3/STAT5 phospholyration in western blot and apoptosis in annexin V/PI staining. Ruxolitinib prolonged survival time of xenotransplanted mice (p = 0.0213), however, the proliferation of leukemic cells in mouse bone marrow was not suppressed by ruxolitinib. Ba/F3-SPAG9-JAK2 cells showed a dose-dependent response for venetoclax (a BCL-2 inhibitor) with IC50 2.57 ± 1.11 µM and AZD5991 (an MCL-1 inhibitor) with IC50 6.76 ± 3.31 µM. Treatment of Ba/F3-SPAG9-JAK2 cells with a combination of ruxolitinib and venetoclax or AZD5991 resulted in a significant reduction in the IC50 of ruxolitinib (p<0.01) [Fig. 4, 5], with a combined index (CI) value of 0.61 or 0.92, indicating a moderately or weak synergistic effect in vitro. [Conclusion] SPAG9-JAK2 promotes cell proliferation and that tyrosine phosphorylation of the JAK2 kinase domain is critical for IL-3-independent cell growth. Ruxolitinib shows sufficient cytotoxic effects against Ba/F3-SPAG9-JAK2 cells in vitro, but is only partially effective in vivo. Activation of the JAK2-STAT1-BCL-2/MCL-1 axis contributes to aberrant growth promotion by SPAG9-JAK2. BCL-2 or MCL-1 inhibitors in combination with ruxolitinib shows efficacy against Ph-like ALL carrying the SPAG9-JAK2 fusion in vitro. Figure 1 Figure 1. Disclosures No relevant conflicts of interest to declare.