The role of actin in nonmuscle cell motility

Since actin, myosin, and in certain systems, tropomyosin have now been described in nonmuscle cells, it is not surprising that most of the current models of nonmuscle cell motility are simplified versions of myofibril contraction (see Huxley, 1973; Pollard and Weihing, 1974; Spudich, 1974). These models may very well be valid, but it is unfortunate to limit one's thinking by assuming that one particular specialized cell type, skeletal muscle, must necessarily reflect how the generally occurring phenomenon of cell motility in all cells is achieved. I have described in this chapter two systems in which the generation of motion is carried out in association with actin filaments alone. No myosin appears to be present. In one system motion is effected by the rapid polymerization of actin, in the other by changing the packing of actin filaments. My purpose is not only to suggest that alternative mechanisms are possible and should be looked for in the diverse types of actin associated motility present in nonmuscle cells, but also, and this is the main aim of this chapter, I am suggesting that the basic control mechanisms which control motility in nonmuscle cells may be very different from those in skeletal muscle. I have tried to demonstrate in this regard that the assembly and disassembly of the contractile apparatus in nonmuscle cells is probably the basic control mechanism. Unlike skeletal muscle, in which the contractile apparatus is stable, in nonmuscle cells it is extremely labile forming and breaking down rapidly and repeatedly during development. In fact, much of the motion may be attributed not to myosin-actin interactions, which undoubtedly occur, but to the assembly and disassembly of the apparatus itself. Thus I concentrated on control of the assembly of actin in cells, in particular on what determines the proper spatial and temporal coordinates of this assembly. I described how actin in sperm and red blood cell ghosts can be nonrandomly distributed in cells in the nonfilamentous state which, interestingly enough, is not monomeric actin. Instead much of the actin appears to be bound in the cell. I also discussed how the actin may polymerize relative to membranes, how actin filaments may associate with membranes along their lengths, how the polarity of the actin may be determined, and what might the substance be which holds the actin to the membranes.