Cerebral Cavernous Malformations (CCM) are vascular disorders of central nervous system that arise from weakened blood capillaries causing blood leakage. Three genes have been shown to cause familial CCM in humans, CCM1, CCM2 and CCM3, and two are conserved in the worm (kri-1/CCM1 and ccm-3/CCM3). Patients with mutations in CCM3 have a earlier disease onset and a more severe prognosis than CCM1 or CCM2 patients. CCM1-3 form a ternary complex but CCM-3 protein resides predominantly in the STriatin Interacting Phosphatase and Kinase (STRIPAK) complex in cells. We previously showed that kri-1 is required in the soma to promote DNA damage-induced germline apoptosis (Ito et al., 2010) whereas ccm-3 is required for excretory canal extension (Lant et al., 2015), demonstrating distinct cellular functions for these genes. Using the germline as a model we show that CCM-3/STRIPAK is also required for rachis development by regulating both endocytic recycling and the anctinomysin cytoskeleton.
During development germ cells undergo incomplete cytokinesis to create openings into a common lumen (rachis), which requires the coordinated activity of cytoskeletal proteins and receptor-mediated signalling pathways, such as Notch and Ras/MAPK. ccm-3 mutants have fewer mitotically proliferating germ cells and reduced GLP-1/Notch. Since we previously showed that ccm-3 mutants are defective in endocytic recycling we asked if mitotic cell proliferation could be restored by inhibiting sel-9, which negatively regulates transport of GLP-1 to the membrane surface. Indeed, sel-9(RNAi) restored mitotic cell proliferation in ccm-3 mutants to wild type levels. We also observed diminished Ras/MAPK signalling and failure to present the vitellogenin receptor RME-2 on oocytes, suggesting a general defect in endocytic recycling. Consistent with this, we found that CCM-3 colocalizes with the endosome recycling protein RAB-11.
Loss of anillin proteins results in germline defects that are strikingly similar to ccm-3 mutants, suggesting potential interactions between these scaffolding proteins in promoting rachis development. CCM-3 and its binding partner GCK-1 are required for cortical localization of ANI-1 and the non-muscle myosin NMY-2. Knockdown of ccm-3 also results in defective mitotic division and loss of polarity proteins in the germline and embryo. Based on these results we propose that CCM-3 functions to couple vesicle trafficking and acto-myosin organization in order to fine-tune membrane patterning and polarity in the germline and embryo. Using C. elegans to understand the CCM-3 regulatory network should help identify non-surgical therapies for treating CCM patients, which we are pursuing also through small molecule screens.