Chemistry Reference
In-Depth Information
at the N-terminus, whereas some classes of kinesin have C-terminal or central motor
domains. The C-terminal motors translate in the opposite direction, toward the slow
growing (minus) end of microtubules [132]. Gliding
filament and processivity assays
with engineered chimeras of plus-end and minus-end directed kinesin classes have
shown that the sequences of the linkers and their interaction with the motor domain
control the direction of motion [133].
Kinesins and dyneins are essential for formation of the mitotic spindle and its
motions during chromosome segregation. A tetrameric, processive plus-end directed
motor, kinesin-5, also termed Eg5, crosslinks microtubules and causes relative
microtubule sliding [134, 135]. Another mitotic motor, kinesin-13 also termed
MCAK (mitotic centromere-associated kinesin), targets the ends of microtubules
and facilitates their depolymerization, presumably helping to power chromosome
separation. MCAK oligomerizes into rings that encircle the microtubule [136]. A
single-molecule TIRF assay [137] showed that MCAK reaches the ends of the
microtubule by a one-dimensional, ATP-independent, diffusional process that is
20
-
50-fold faster than direct binding to the microtubule ends from the solution. The
interactions of many proteins with microtubules are sensitive to the ionic strength
suggesting that they are mediated by ionic bonds.
For kinesin-1, the evidence that both heads are required for processivity came
from gliding
filament assays [138, 139], optical trap mechanics [118], single-
molecule
fluorescence imaging [117] and FIONA [108]. Therefore it was a surprise
when a monomeric isoform, kinesin-3 also termed KIF1A, was shown to exhibit
processive motility [140]. Like myosin VI, whether this motor is monomeric or
dimeric in cells was called into question [141] and the motions have an extensive
diffusive component, like MCAK, as well as an ATP-driven directionality. Still, the
intriguing question remains: how can a single motor domain dissociate from
tubulin to move along the microtubule and not diffuse away? KIF1A contains an
unusual loop containing six closely spaced lysine residues (the K-loop) forming a
highly positively charged region that seems to interact with a C-terminal segment
of tubulin containing negatively charged glutamate residues (the E-hook). Delet-
ing the K-loop or proteolyzing the E-hook eliminated processive motility by single-
headed KIF1A [140]. Thus, a secondary interaction between the KIF1A motor
domain and tubulin might hold the motor near the microtubule while the ATP-
dependent tubulin binding domain is weakly bound or detached. Crystal struc-
tures of the KIF1A motor domain showed that the position of the K-loop is
dependent on the identity of the bound nucleotide [142]. This result suggests
that the KIF1A motor domain has two alternating, ATP-dependent sites of
microtubule interaction, thereby obviating the requirement for a second head.
Whether this concept applies to any other kinesins or other molecular motors is
unknown.
Many questions remain on the mechanism of kinesin motility [143]. Does kinesin
take substeps? How does kinesin track a proto
lament? Which tubulin dimers are
targeted during a step?How are the biochemical andmechanical cycles related? Is the
neck linker hypothesis compatible with the energy transduction and thermodynamic
