Chemical Activation of the SHIP1 Inositol Lipid Phosphatase: A Novel Therapeutic Strategy to Suppress B-Cell Receptor Signaling and CXCR4 Expression in Malignant Human B Cells

Signaling via the B-cell receptor (BCR) is a major driver of malignant B-cell proliferation/survival in chronic lymphocytic leukaemia (CLL) and non-Hodgkin's lymphoma. The role of kinases in BCR signaling is well understood and kinase inhibitors are effective therapies for these diseases. However, resistance is increasingly common and new drugs are required. SHIP1 is a PI(3,4,5)P3-specific phosphatase which suppresses PI(3,4,5)P3-dependent signaling downstream of PI3 kinase (PI3K) and imposes "inhibitory" activity via accumulation of its product PI(4,5)P2. Here, we investigated the effects of AXQ-C5, a novel chemical activator of SHIP1, on BCR signaling and expression of CXCR4, a chemokine receptor which is thought to play an important role in tissue homing and/or retention of CLL cells in vivo.

Immunoblot analysis revealed that SHIP1 was constitutively tyrosine phosphorylated in malignant cells derived from the blood of chronic lymphocytic leukemia (CLL) patients. BCR stimulation (with anti-IgM) did not substantially alter SHIP1 expression or phosphorylation, but resulted in rapid (within 30 minutes) relocalization of phospho-SHIP1 to the plasma membrane. Confocal imaging demonstrated that a fraction of this plasma membrane-associated phospho-SHIP1 co-localized with surface IgM (sIgM). Induced association of sIgM and phospho-SHIP1 following anti-IgM treatment was confirmed by co-immunoprecipitation. Pre-treatment of CLL cells with AQX-C5 resulted in significant reduction in the levels of anti-IgM-induced phosphorylation of ERK1/2 and AKT. AQX-C5 also significantly reduced induction of the MYC oncoprotein and, in longer time-course experiments, induced caspase-dependent CLL cell apoptosis. By contrast, T cells were relatively resistant to the pro-apoptotic effects of AQX-C5. AQX-C5 also induced apoptosis in diffuse large B-cell lymphoma cell lines dependent on chronic BCR signalling.

We also investigated the effects of AQX-C5 on expression of CXCR4. Interestingly, AQX-C5 alone was sufficient to cause a strong down-modulation of CXCR4 expression (~50%) in the majority of samples analyzed. This effect was relatively specific since other cell surface receptors, including the transferrin receptor (which like CXCR4 is subject to rapid endocytosis), were unaffected by AQX-C5. Interestingly, idelalisib did not result in CXCR4 down-modulation, suggesting response to AQX-C5 is a consequence of PI(4,5)P2 accumulation rather than decreased levels of PI(3,4,5)P3 per se. Moreover, AQX-C5 was much less effective in inducing CXCR4 down-modulation in normal B cells derived from healthy donors, suggesting this response to AQX-C5 may be leukemia-specific.

In conclusion, this study supports the hypothesis that chemical SHIP1 activation is sufficient to suppress BCR-mediated signaling, acting, at least in part, by decreasing PI(3,4,5)P3-mediated activation of the AKT pathway. In addition, SHIP1 activation results in strong CXCR4 down-modulation, a response not observed with direct PI3Kdelta inhibition or in normal B cells. Combined inhibitory effects on BCR signaling and CXCR4 expression make SHIP1 activators exciting new therapeutic agents for B-cell cancers, potentially including those that have acquired resistance to kinase inhibitors.