1.2 Polymerisation and depolymerisation of actin
Actin is a highly conserved protein. Most organisms have several genes encoding actin; in humans there are six principal isoforms, four of which are found in different types of muscle and the other two (β and γ) in all non-muscle cells. (The term ‘isoform’ describes variants of a protein. These may be produced by different genes, or by differential splicing of the mRNA, or be generated by post-translational modifications.) The β and γ cytoskeletal forms differ by just four amino acid residues at the N-terminus. The high level of conservation is probably partly due to the structural requirements of microfilament formation and partly related to the fact that actin interacts with dozens of other highly conserved molecules, so its scope for variation is limited.
Microfilaments are formed by the polymerisation of actin monomers by the formation of multiple non-covalent bonds between adjacent molecules. The monomers form protofilaments, strings of monomers linked end-to-end, which align with and wind around another string to form the filament. The helix twists once every 37 nm, and each monomer has the same orientation within the helix.
What is the advantage of having non-covalent bonds linking the elements of the protofilaments?
The advantage of assembling the protofilaments non-covalently is that they can be disassembled and reassembled relatively quickly by loss or addition of monomers to the ends. This allows greater flexibility to the cell than would be provided by a fixed (covalently linked) network.
In solution, filament assembly starts when an actin dimer forms spontaneously. As additional monomers bind to the assembling filament it becomes increasingly stable. This process is called nucleation and in solution it is relatively slow. In a cell, however, actin monomers will normally add on to an already formed filament. Alternatively, nucleation can start from sites at the plasma membrane, and attachment of microfilaments to the plasma membrane is important in maintaining cell shape and permitting movement.
The rate of assembly of an actin filament depends on the concentration of the monomers. Once a critical threshold concentration has been exceeded, assembly of the polymeric form is favoured. However, actin monomers add on to one end of a filament much faster than to the other end, and these are referred to as the plus end and minus end of the filament, respectively. The difference in the rate of growth is due to a difference in the conformational changes that occur to the subunits when they attach to the plus or minus ends of the filament. The converse is also true: when the concentration of monomers falls below a threshold, the filament tends to depolymerise, but it also depolymerises more rapidly from the plus end. Be careful not to think of the plus end as the place where monomers are always added and the minus end as the place where they are always removed. It is better to think of the plus end as the more active end and the minus end as the less active end. Given the appropriate conditions, filaments do tend to grow from the plus end and shrink from the minus end, but this is not necessarily the case.
So far, we have thought of actin assembly just as an equilibrium reaction, which is dependent on the concentrations of actin monomers and actin polymers. However, actin is an ATPase. Normally, free actin has bound ATP (Figure 2). In polymerised actin the ATP is slowly hydrolysed to ADP, so the longer an actin molecule has been in a filament, the more likely it is to be converted into actin-ADP.
Consider the plus end of a microfilament. As it grows, new actin-ATP molecules are added to the plus end faster than the ATP is hydrolysed to ADP, and consequently the tip of the filament contains a cap of actin-ATP molecules, called an ATP cap (or T form). At the minus end, the rate of ATP hydrolysis exceeds the rate at which actin-ATP monomers are added, and the microfilament consists mainly of actin containing ADP (or the D form). Hydrolysis of ATP to ADP changes the conformation of actin, causing the critical concentration for D actin to be higher than that for T actin.
Consider the situation in which monomeric actin is present at a concentration below the critical concentration of D actin (at the minus end) but above the critical concentration of T actin (at the plus end). Will the plus end grow or shrink? Similarly, will the minus end grow or shrink?
Because the actin monomer concentration exceeds the critical concentration of T actin, the plus end will grow. However, because the actin monomer concentration is below the critical concentration of D actin, the minus end will shrink.
Because of this, when the actin monomer concentration is between the critical concentrations of T actin and D actin, actin filaments appear to move forwards at their plus end and retract at their minus end. This phenomenon is called treadmilling. It requires the presence of ATP, and a similar effect is seen with microtubules. Notice, however, that although the filament appears to move, the individual monomers do not (Figure 3).
The basic behaviour of actin is greatly modified by actin-binding proteins. In particular, capping proteins attach to the plus end of filaments and prevent addition or loss of further actin monomers. In practice, at any one time, most of the microfilaments in a cell are capped. Capping proteins can also protect the minus end of microfilaments. In muscle cells, where actin fibres are extremely stable, the plus ends of the filaments are capped by CapZ and the minus ends are capped by tropomodulin.
If an actin filament is capped at the plus end and the concentration of actin-ATP is high (above the threshold for polymerisation) what will happen to the filament?
It will grow slowly from its minus end. The plus end is capped, so nothing happens there. The level of actin-ATP is high so the fibre will grow, but because it is at the minus end (low activity), it will do so only slowly.
Other proteins can attach to the sides of filaments and promote rupture of the filaments, stabilisation of the filaments, bundling of filaments, or branching by the nucleation of new filaments. An example of a filament-cleaving molecule, gelsolin, is shown in Figure 2. It is thought that gelsolin takes advantage of random thermally induced flexions of the microfilament, to insert itself between two actin molecules, thus causing the filament to rupture. Gelsolin binds to the plus end of the actin monomer, so that it also acts as a capping protein.
Molecules that cross-link microfilaments into bundles are very important in maintaining the structure of the microfilament network. Different molecules do this in different ways, so creating different types of structure. For example, α-actinin cross-links bundles of filaments in an antiparallel fashion, forming bundles that are relatively open. This enables cellular myosin to intercalate and engage with the actin, forming a stress fibre. A quite different type of cross-link is made by filamin, which links the filaments in loose three-dimensional bundles to form a gel-like array. Within a cell, the microfilament network will be organised in different ways in different parts of the cell. Some examples of cross-linking are shown in Figure 4.
Box 1 Myosins
Myosin was originally identified as a component of skeletal muscle, which interacted with actin fibres to generate muscle contraction. Subsequently, this myosin was found to be just one of a large family of molecules present in all eukaryotic cells, including plants, all of which have motor functions and ATPase activity. Skeletal muscle myosin has a long tail and two heads, which contain the catalytic sites, and the basic units become bundled together in fibres. This protein is also found widely in animal cells, and has since been called myosin II – two-headed. The designation was prompted by the discovery of a single-headed myosin in Acanthamoeba. Since then numerous single-headed and double-headed cytoplasmic myosins have been discovered. Some are confined to animals, some to plants and some to particular protoctists, but the majority of them are found in all eukaryotes, which indicates that the myosin family of molecules evolved before the divergence of these kingdoms.