Latrunculin alters the actin-monomer subunit interface to prevent polymerization

atrunculin-A is a drug that is capable of rapidly, reversibly and specifically disrupting the actin cytoskeleton. The efficacy of its action has made it a compound of choice in many cell-biology laboratories, supplanting the classic actin-depolymerizing drug cytochalasin-D. One reason for this is that the mode of action of latrunculin seems to be less complex than that of cytochalasin. Whereas the latter affects the kinetics of actin-filament polymerization at both the barbed and pointed ends, latrunculin-A seems to associate only with actin monomers, thereby preventing them from repolymerizing into filaments. The association of latrunculin with monomeric, rather than filamentous, actin gave us the opportunity to further our understanding of this interaction by detailed structural analysis of actin monomers using crystallographic techniques. Here we show the first high-resolution structure of an actin-disrupting drug in association with actin and discuss how its interactions with actin, and the conformational changes that its binding causes, may explain its mode of action within the cell. Latrunculin (Fig. 1a) is purified from Latrunculia magnificans, a Red Sea sponge that exudes a noxious, red fluid that kills fish within minutes. Two related compounds, latrunculin-A and latrunculinB, isolated from the fluid were shown to depolymerize actin structures both in vitro and in vivo. The in vitro studies showed that latrunculin binds only to the actin monomer and that the kinetics of this interaction are consistent with the complex being unable to polymerize. Unlike cytochalasin, latrunculin can disrupt the actin cytoskeleton in yeast cells. This has enabled genetic studies to be carried out that have facilitated the identification of point mutations in the actin gene that cause cells to become resistant to the effects of the drug (Fig. 1b). The mutations that give rise to latrunculin resistance were found to be clustered around a distinct site, close to the nucleotide-binding site, which indicated that they might identify a potential binding site for latrunculin. However, as this site is not close to recognized subunit contacts in the filament, or to known binding sites for other proteins that associate with actin, the mechanism by which latrunculin exerts its effects has remained unclear. Actin has never been known to crystallize in the absence of a binding protein that keeps it in a monodispersed state. Of the three known examples of such binding proteins, profilin is inappropriate as it promotes nucleotide exchange, whereas deoxyribonuclease1 binds to domains that have been implicated, in studies of yeast genetics, in latrunculin binding. In contrast, gelsolin domain 1 in complex with actin leaves these domains free and also reduces nucleotide exchange, as does latrunculin. We therefore soaked latrunculin-A L