SUMMARY – PROJECT 3 Recent studies by our group and others have identified MEKK3-KLF2/4 signaling as the direct target of CCM complex function, and showed that loss of CCM function confers lesion formation through gain of MEKK3 signaling and elevated KLF2 and KLF4 expression in brain endothelial cells. Unexpectedly, we have also found that endothelial TLR4 activation by lipopolysaccharide derived from gram negative bacteria in the gut microbiome plays a central, upstream role in the activation of MEKK3-KLF2/4 signaling and CCM formation in both mice and humans. These findings have culminated in a now widely accepted model of CCM disease, but the downstream effectors by which KLF2 and KLF4 drive lesion formation remain unknown. Our preliminary studies reveal two new insights into the downstream events that drive CCM formation: 1) PI3K gain of function synergizes with CCM loss of function to drive lesion formation in mice and a majority of surgically resected human CCM lesions, and 2) ADAMTS cleavage of peri- vascular versican drives the CCM phenotype in mice. Project 3 will define the roles of PI3K signaling and versican proteolysis during CCM pathogenesis using established and newly developed mouse genetic models. These studies will strongly complement those in Project 1 and Core A that examine the PIK3CA mutations in human CCM lesions and a cell non-autonomous mechanism by which wild-type endothelial cells contribute to CCM lesions. Most importantly, we expect these studies to be rapidly translated to the clinic by providing support for the use of FDA-approved agents such as the PI3K pathway inhibitor rapamycin to treat CCM disease.