Wnt5a Induces ROR1 Dependent Tyrosine Phosphorylation of DOCK2, and Enhanced Activation of ERK to Promote Proliferation of CLL Cells

ROR1 (receptor tyrosine kinase-like orphan receptor 1) is a conserved, oncoembryonic surface antigen expressed in chronic lymphocytic leukemia (CLL) B cells, but not on most normal post-partem tissues. Patients with high-level leukemia-cell expression of ROR1 have been found to have a shorter median time from diagnosis to therapy and shorter median survival than patients with CLL cells that express low-to-negligible amounts of ROR1 (Cui B. et al., Blood, 128:2931, 2016). In prior studies we found that Wnt5a could induce ROR1 to recruit DOCK2 at proline 808 and activate Rac1/2, leading to enhanced leukemia-cell proliferation (Hasan M.K. et al., Blood, 132:170, 2018). Moreover, Wnt5a-enhanced proliferation could be inhibited by cirmtuzumab, a first-in-class humanized mAb specific for a functional epitope of the ROR1; this mAb is undergoing phase II clinical testing in patients with CLL (Choi MY, et al, Cell Stem Cell22:951, 2018). However, the molecular mechanism(s) of ROR1/DOCK2 signaling in the activation of Rac1/2, or downstream signaling molecules such as ERK, or AKT remains elusive. Here, we report that Wnt5a can induce tyrosine phosphorylation of DOCK2 in serum-starved primary CLL cells. The capacity of Wnt5a to induce DOCK2 phosphorylation could be inhibited by cirmtuzumab, but not by ibrutinib, even when used at concentrations that exceeded those required to inhibit B-cell receptor signaling. Phosphorylation of DOCK2 was associated with activation of ERK1/2. Furthermore, siRNA mediated knockdown of DOCK2 could inhibit Wnt5a-induced activation of ERK1/2, as did treatment of the CLL cells with cirmtuzumab. We corroborated these findings in CLL-cell line MEC1, which expresses endogenous Wnt5a but lacks expression of ROR1. We stably transfected ROR1 into MEC1 cells to generate MEC1-ROR1 cells. We found higher-levels phosphorylated DOCK2 and activation of ERK1/2 in MEC1-ROR1 cells than in MEC1 cells lacking ROR1. Treatment of MEC1-ROR1 cells with neutralizing antibodies to Wnt5a or cirmtuzumab suppresed DOCK2 phosphorylation and activation of ERK1/2. Moreover, siRNA mediated knockdown of DOCK2 reduced the level of activation of ERK1/2 in MEC1-ROR1 cells. We examined whether the proline-rich-domain (PRD) of ROR1, and SH3-binding motifs with the PRD that were necessary for these effects. We found that PRD and more specifically proline at 808 position of ROR1 was required to enhance phosphorylation of DOCK2 and activation of ERK1/2 over that seen in MEC1 cells lacking ROR1. We also found that inhibition of Rac using a small molecule inhibitor of Rac1/2 could suppress the capacity of Wnt5a to induce activation of ERK, suggesting that Wnt5a induced activation of Rac was required for activation of ERK. Finally, Wnt5a could not enhance the proliferation of CLL cells when treated with an ERK inhibitor or citmuzumab. Collectively, our results support that Wnt5a/ROR1/DOCK2 signaling axis contributes enhanced activation of ERK to promote proliferation of CLL cells independently of BTK.