Similar to other
integrin-targeting strategies,
disintegrins have previously shown good efficacy in animal
cancer models with favorable pharmacological attributes and translational potential. Nonetheless, these
polypeptides are notoriously difficult to produce recombinantly due to their particular structure requiring the correct pairing of multiple
disulfide bonds for
biological activity. Here, we show that a sequence-engineered
disintegrin (called
vicrostatin or VCN) can be reliably produced in large scale amounts directly in the oxidative cytoplasm of Origami B E. coli. Through multiple
integrin ligation (i.e., alphavbeta3, alphavbeta5, and alpha5beta1), VCN targets both endothelial and
cancer cells significantly inhibiting their motility through a reconstituted basement membrane. Interestingly, in a manner distinct from other
integrin ligands but reminiscent of some ECM-derived endogenous anti-angiogenic fragments previously described in the literature, VCN profoundly disrupts the actin cytoskeleton of endothelial cells (EC) inducing a rapid disassembly of stress fibers and actin reorganization, ultimately interfering with EC's ability to invade and form tubes (tubulogenesis). Moreover, here we show for the first time that the addition of a
disintegrin to tubulogenic EC sandwiched in vitro between two
Matrigel layers negatively impacts their survival despite the presence of abundant haptotactic cues. A liposomal formulation of VCN (LVCN) was further evaluated in vivo in two animal
cancer models with different growth characteristics. Our data demonstrate that LVCN is well tolerated while exerting a significant delay in
tumor growth and an increase in the survival of treated animals. These results can be partially explained by potent
tumor anti-angiogenic and pro-apoptotic effects induced by LVCN.