STOCKBYTE
By the early 1990s, the ability of actin filaments to organize into distinct structures including stress fibres and membrane ruffles was recognized; but there was little idea of how their formation tuned into signals received at the cell surface. It was work from the laboratory of Alan Hall in London, and in particular that of Anne Ridley, that turned attention to Rho GTPases as key components of a signalling pathway linking different growth factors to specific actin structures.
Hugh Paterson et al. had already provided evidence that Rho might regulate the actin cytoskeleton, in 1990, showing that injection of Rho into cells induced formation of stress fibres. In the first of the two milestone studies, Ridley and Hall tested whether the effects of Rho might be part of an endogenous signal-transduction pathway that regulates stress-fibre formation. They showed that these effects could be mimicked by the re-addition of serum to starved fibroblasts and that stress-fibre formation paralleled the assembly of focal adhesions. By testing a panel of peptides present in serum, they found that the activity responsible for the serum effect was a phospholipid, probably lysophosphatidic acid (LPA). They went on to show, using two inhibition strategies, that endogenous Rho was required for the effects of LPA on both stress-fibre formation and focal-adhesion assembly.
In an accompanying paper, inspired by previous observations from Dafna Bar-Sagi et al. that the GTPase Ras can induce membrane ruffling in addition to proliferation, Ridley et al. tested whether a second Rho family member, Rac, also affects the actin cytoskeleton. They found that Rac acts in a pathway distinct from that of Ras and Rho to regulate actin organization in response to certain growth factors. They showed that microinjection of activated Rac triggered membrane ruffling and lamellipodia in confluent fibroblasts, and that a dominant inhibitory Rac prevented the induction of membrane ruffling and lamellipodia by growth factors and Ras. In addition, Rac appeared to have a slower effect on stress fibres — this effect was dependent on Rho. Thus, the effects of Rac and Rho downstream of particular growth factors are interdependent: Rho can trigger stress-fibre formation by two distinct pathways, only one of which is Rac-dependent.
For the first time, the glimmerings of a signal transduction pathway to the cytoskeleton could be clearly discerned.
William Bement
Three years later, Catherine Nobes and Hall completed the trilogy, with the finding that CDC42 controls the formation of filopodia. Together, these studies showed that each member of the Rho GTPase trio controls a specific cytoskeletal response to different growth factors, but also introduced the idea that these pathways are intricately entwined. They forged a clear pathway from growth-factor signalling at the membrane to specific actin structures, and provided a potential molecular basis for specificity. Importantly, they also guided others to test the role of Rho GTPases in the regulation of microtubule structures — a finding that placed the Rho family firmly at the intersection between different filament systems.
