Septins promote stress fiber-mediated maturation of focal adhesions and renal epithelial motility.

Organogenesis and tumor metastasis involve the transformation of epithelia to highly motile mesenchymal-like cells. Septins are filamentous G proteins, which are overexpressed in metastatic carcinomas, but their functions in epithelial motility are unknown. Here, we show that a novel network of septin filaments underlies the organization of the transverse arc and radial (dorsal) stress fibers at the leading lamella of migrating renal epithelia. Surprisingly, septin depletion resulted in smaller and more transient and peripheral focal adhesions. This phenotype was accompanied by a highly disorganized lamellar actin network and rescued by the actin bundling protein α-actinin-1. We show that preassembled actin filaments are cross-linked directly by Septin 9 (SEPT9), whose expression is increased after induction of renal epithelial motility with the hepatocyte growth factor. Significantly, SEPT9 overexpression enhanced renal cell migration in 2D and 3D matrices, whereas SEPT9 knockdown decreased migration. These results suggest that septins promote epithelial motility by reinforcing the cross-linking of lamellar stress fibers and the stability of nascent focal adhesions.

In silico docking of forchlorfenuron (FCF) to septins suggests that FCF interferes with GTP binding.

Septins are GTP-binding proteins that form cytoskeleton-like filaments, which are essential for many functions in eukaryotic organisms. Small molecule compounds that disrupt septin filament assembly are valuable tools for dissecting septin functions with high temporal control. To date, forchlorfenuron (FCF) is the only compound known to affect septin assembly and functions. FCF dampens the dynamics of septin assembly inducing the formation of enlarged stable polymers, but the underlying mechanism of action is unknown. To investigate how FCF binds and affects septins, we performed in silico simulations of FCF docking to all available crystal structures of septins. Docking of FCF with SEPT2 and SEPT3 indicated that FCF interacts preferentially with the nucleotide-binding pockets of septins. Strikingly, FCF is predicted to form hydrogen bonds with residues involved in GDP-binding, mimicking nucleotide binding. FCF docking with the structure of SEPT2-GppNHp, a nonhydrolyzable GTP analog, and SEPT7 showed that FCF may assume two alternative non-overlapping conformations deeply into and on the outer side of the nucleotide-binding pocket. Surprisingly, FCF was predicted to interact with the P-loop Walker A motif GxxxxGKS/T, which binds the phosphates of GTP, and the GTP specificity motif AKAD, which interacts with the guanine base of GTP, and highly conserved amino acids including a threonine, which is critical for GTP hydrolysis. Thus, in silico FCF exhibits a conserved mechanism of binding, interacting with septin signature motifs and residues involved in GTP binding and hydrolysis. Taken together, our results suggest that FCF stabilizes septins by locking them into a conformation that mimics a nucleotide-bound state, preventing further GTP binding and hydrolysis. Overall, this study provides the first insight into how FCF may bind and stabilize septins, and offers a blueprint for the rational design of FCF derivatives that could target septins with higher affinity and specificity.