Mechanisms of actin filament turnover in animal cells.

About half of the total actin in the cytoplasm of cultured animal cells is polymerised into filaments at any time. The filaments are further ordered into 3-dimensional patterns by their interaction with a number of actin-binding proteins (ABPs) to form the functional actin cytoskeleton. Three consistent patterns of organisation can be discerned both by their complement of ABPs, as determined by co-localisation, and by the characteristic arrangement of actin filaments: isotropic arrays, parallel bundles and anti-parallel bundles. These three patterns of organisation appear to have discrete functional properties which, together, give rise to the motile behavior of the cells. The proportions and locations of these actin filament organisations reflect, or give rise to, the particular properties of the cells in which they are found. In locomoting cells in particular, the extent and precise cellular location of these three classes of organisation are constantly changing during the process of locomotion. This constant adaptation of the actin cytoskeleton appears, from a number of approaches, to result from a continuous cycle of assembly and disassembly of filaments leading to continuous adaptation of the actin cytoskeleton. In some cells at least, net assembly occurs predominantly at the extreme leading edge of the lamellipodium, and it must be presumed that filaments assembled in the isotropic arrays here give rise to the other levels of architecture found in the cell. However, overlying this appears to be a continuous cycle of assembly and disassembly of filaments within all parts of the actin cytoskeleton. The underlying imperative for turnover of actin filaments derives from the ATPase associated with polymerisation. The implications of this assembly ATPase for both turnover and for the progressive evolution of actin filament architectures is discussed.