The nucleotidyl cyclase toxin ExoY is one of the virulence factors injected by the Pseudomonas aeruginosa type III secretion system into host cells. Inside cells, it is activated by an unknown eukaryotic cofactor to synthesize various cyclic nucleotide monophosphates. ExoY-like adenylate cyclases are also found in Multifunctional-Autoprocessing Repeats-in-ToXin (MARTX) toxins produced by various Gram-negative pathogens. Here we demonstrate that filamentous actin (F-actin) is the hitherto unknown cofactor of ExoY. Association with F-actin stimulates ExoY activity more than 10,000 fold in vitro and results in stabilization of actin filaments. ExoY is recruited to actin filaments in transfected cells and alters F-actin turnover. Actin also activates an ExoY-like adenylate cyclase MARTX effector domain from Vibrio nigripulchritudo. Finally, using a yeast genetic screen, we identify actin mutants that no longer activate ExoY. Our results thus reveal a new sub-group within the class II adenylyl cyclase family, namely actin-activated nucleotidyl cyclase (AA-NC) toxins.

The dynamic turnover of actin filaments (F-actin) controls cellular motility in eukaryotes and is coupled to changes in the F-actin nucleotide state1–3. It remains unclear how F-actin hydrolyses ATP and subsequently undergoes subtle conformational rearrangements that ultimately lead to filament depolymerization by actin-binding proteins. Here we present cryo-electron microscopy structures of F-actin in all nucleotide states, polymerized in the presence of Mg2+ or Ca2+ at approximately 2.2 Å resolution. The structures show that actin polymerization induces the relocation of water molecules in the nucleotide-binding pocket, activating one of them for the nucleophilic attack of ATP. Unexpectedly, the back door for the subsequent release of inorganic phosphate (Pi) is closed in all structures, indicating that Pi release occurs transiently. The small changes in the nucleotide-binding pocket after ATP hydrolysis and Pi release are sensed by a key amino acid, amplified and transmitted to the filament periphery. Furthermore, differences in the positions of water molecules in the nucleotide-binding pocket explain why Ca2+-actin shows slower polymerization rates than Mg2+-actin. Our work elucidates the solvent-driven rearrangements that govern actin filament assembly and aging and lays the foundation for the rational design of drugs and small molecules for imaging and therapeutic applications.

Several bacterial pathogens produce nucleotidyl cyclase toxins to manipulate eukaryotic host cells. Inside host cells they are activated by endogenous cofactors to produce high levels of cyclic nucleotides (cNMPs). The ExoY toxin from Pseudomonas aeruginosa (PaExoY) and the ExoY-like module (VnExoY) found in the MARTX (Multifunctional-Autoprocessing Repeats-in-ToXin) toxin of Vibrio nigripulchritudo share modest sequence similarity (~38%) but were both recently shown to be activated by actin after their delivery to the eukaryotic host cell. Here, we further characterized the ExoY-like cyclase of V. nigripulchritudo. We show that, in contrast to PaExoY that requires polymerized actin (F-actin) for maximum activation, VnExoY is selectively activated by monomeric actin (G-actin). These two enzymes also display different nucleotide substrate and divalent cation specificities. In vitro in presence of the cation Mg2+, the F-actin activated PaExoY exhibits a promiscuous nucleotidyl cyclase activity with the substrate preference GTP>ATP≥UTP>CTP, while the G-actin activated VnExoY shows a strong preference for ATP as substrate, as it is the case for the well-known calmodulin-activated adenylate cyclase toxins from Bordetella pertussis or Bacillus anthracis. These results suggest that the actin-activated nucleotidyl cyclase virulence factors despite sharing a common activator may actually display a greater variability of biological effects in infected cells than initially anticipated.

Lifeact is a short actin-binding peptide that is used to visualize filamentous actin (F-actin) structures in live eukaryotic cells using fluorescence microscopy. However, this popular probe has been shown to alter cellular morphology by affecting the structure of the cytoskeleton. The molecular basis for such artefacts is poorly understood. Here, we determined the high-resolution structure of the Lifeact–F-actin complex using electron cryo-microscopy (cryo-EM). The structure reveals that Lifeact interacts with a hydrophobic binding pocket on F-actin and stretches over 2 adjacent actin subunits, stabilizing the DNase I-binding loop (D-loop) of actin in the closed conformation. Interestingly, the hydrophobic binding site is also used by actin-binding proteins, such as cofilin and myosin and actin-binding toxins, such as a hypervariable region of TccC3 (TccC3HVR) from Photorhabdus luminescens and ExoY from Pseudomonas aeruginosa. In vitro binding assays and activity measurements demonstrate that Lifeact indeed competes with these proteins, providing an explanation for the altering effects of Lifeact on cell morphology in vivo. Finally, we demonstrate that the affinity of Lifeact to F-actin can be increased by introducing mutations into the peptide, laying the foundation for designing improved actin probes for live cell imaging.