Like a miniature railway, conventional kinesin can transport molecular cargoes over long distances. It does this without dissociating from its microtubule rails — a property known as processivity. But how does kinesin stay on track? In the Journal of Cell Biology, Ronald Vale and colleagues describe two studies aimed at finding out.

Conventional kinesin is a dimer, with two catalytic motor domains connected through a stalk to the cargo-binding carboxy-terminal tail. Each motor head is joined to the stalk by a flexible 'neck' region consisting of two parts — a 'neck linker' that interacts with the motor and an adjacent 'neck coiled-coil'. The neck linker drives the characteristic hand-over-hand 'stepping' movement of the two kinesin heads, a mechanism that ensures both heads do not dissociate simultaneously.

This mechanism relies on a conformational change, and there are two theories for how this might occur — that the neck linker is 'unzippered' from the motor, or that there is partial unwinding of the neck coiled-coil. Tomishige and Vale have now tried to distinguish these possibilities by manipulating movement of the neck region using disulphide crosslinking of cysteine residues engineered into recombinant kinesin motors.

Using this system to immobilize the neck linker, the authors found that kinesin no longer moved in only one direction. But crosslinking to permit limited movement allowed biased unidirectional diffusion of the kinesin. This indicates that partial movement of the neck linker is enough to determine directionality, but that full movement is required for active, processive movement. In support of the argument that conformational changes in the neck linker are needed for processivity, immobilization of the neck coiled-coiled region had very little effect on motor activity.

So what does this region do? Tomishige and Vale noticed that the extent of processivity — or 'run length' — was decreased by up to 50% when the neck coiled-coil was immobilized. In the second study, Thorn et al. investigated this further by adding positive charge to the neck coiled-coil to generate 'ultra-processive' kinesin mutants. The gain in processivity was diminished by high salt concentrations or by cleaving off the negative carboxyl terminus of the microtubule protein tubulin, indicating that there might be an electrostatic interaction between this region of tubulin and the neck coiled-coil. This interaction is probably weak, however, as it was abolished by adding relatively low loads to the kinesin in an optical-trap assay.

The current train of thought, then, is that the neck coiled-coil tethers kinesin near the microtubule surface, whereas the neck linker is involved in the conformational change behind processive movement.