ROCK and Rho will never Dia

In This Issue In This Issue ROCK and Rho will never Dia irectional migration of a cell requires both lamellipodia formation at the leading edge and actomyosin contraction within the tail. New results from Tsuji et al. (page 819) indicate that Rho may mediate both tasks, and that these two arms of Rho signaling are antagonistic. Rho signaling is known to mediate the formation of stress fibers, structures similar to actin bundles that are required for tail retraction. This function of Rho requires the effector kinase ROCK. The new results demonstrate that Rho can also signal through a different effector, mDia1, and cause membrane ruffling. The mDia1 pathway activated another Rho family GTPase member, Rac. In neuronal cells, Rho inhibits Rac, but this is the first demonstration that Rho can also activate Rac. Tsuji et al. uncovered mDia1-mediated membrane ruffling by selectively inhibiting ROCK signaling. Thus, the ROCK pathway normally antagonizes the mDia1/Rac pathway. Basal levels of GTP-Rho may be sufficient to activate mDia1, whereas high concentrations of active Rho are probably required to mediate ROCK-dependent signaling. In motile cells, actomyosin contraction and lamellipodium formation probably require differentially controlled local concentrations of active Rho. ᭿ D Actin (white) forms membrane ruffles when ROCK is blocked (right). poptotic cell death is executed by the caspase family of cysteine proteases. Apoptosis of injured neuronal cells can be inhibited by blocking caspases, but these cells still die eventually, via a second, nonapoptotic, mechanism that is not well understood. On page 771, Chang and Johnson demonstrate that this process can be prevented by blocking A loss of mitochondrial membrane potential. Using growth factor–deprived sympathetic neurons from rats, the authors found that cyclosporin A (CsA) provided protection to rat neurons from caspase-independent cell death. The drug probably acts by blocking the opening of the mitochondrial permeability transition pore (PTP), and thus preventing depolarization of the mitochondrial membrane. When caspases were active, CsA had no effect, suggesting that PTP opening is not critical during apoptosis. Unlike rat neurons, mice sympathetic neurons were not protected by CsA. Neuron size may account for this difference, as smaller cells may be unable to survive until CsA arrives and mitochondria can reestablish their ion gradient. Johnson is optimistic that combinations of caspase inhibitors and CsA will have neuroprotective benefits in patients who have suffered a heart attack or stroke, although CsA may not be able to be given at a …