In this work, the researchers
determined the structure of the KIF1A core region bound to ADP (to
2.2 Å) and to a nonhydrolyzable analogue of ATP (to 2.0 Å).
Previously, all attempts to crystallize the ATP-like complex had
failed. The research team's success in crystallizing this complex
provided an excellent opportunity to compare "snapshots" of two
key moments in the kinesin stepping sequence. Although the two structures
look very similar overall, marked differences were observed in two
"switch" regions near the ATP/ADP binding site.
Comparison
of KIF1A switch regions in the ADP (yellow) and ATP-like (red)
states. Part of the bound ATP is shown in gray. |
In the switch I region,
a helix flanked by two short loops in the ADP state forms a short
b-hairpin structure in the ATP-like state. The switch II region
shows a series of structural changes, including the partial unwinding
of a helix and its rotation by roughly 20 degrees in the ATP-like
state. To visualize the effect of these changes when the kinesin
is attached to a microtubule, the researchers embedded the molecular
structures within cryo-electron microscope (cryo-EM) images of kinesin
in contact with a microtubule surface. The results indicate that
the switch II region actually remains fixed relative to the microtubule,
while the rest of the KIF1A core rotates by about 20 degrees in
the opposite direction. The authors suggest that this rotation binds
the kinesin more tightly to the microtubule and creates a directional
bias by pointing the tip of the KIF1A core in the direction of motion.
Rotation
of KIF1A core (yellow) relative to microtubule surface (gray).
Left: Kinesin molecular model embedded within a cryo-EM map
of kinesin. Right: Kinesin orientation in ADP (pink line)
and ATP-like (red line) states. |
Another
important difference between the two states was found in the "neck
linker" domain, which connects the two heads in the dimeric form
of kinesin (and which was "grafted" onto the KIF1A monomer
for this experiment). The structures show that the linker is "docked"
near the core in the ATP-like state but is undocked and disordered
in the ADP state. This finding supports the hypothesis that the neck
linker is a crucial part of the mechanism that drives kinesin. In
this view, the kinesin core is a modular base onto which different
types of neck domains serve as mechanical amplifiers and transmitters
(transmissions and drive shafts) whose exact function depends on the
kinesin variant to which the neck linker belongs.
In general, these results confirm expectationsarrived at by
analogy to previously studied, structurally similar proteinsthat
the conformational changes observed in KIF1A are modular and extend
to all kinesins. They also suggest a rationale for kinesin's tendency
to move in a given direction along a microtubule.
Research conducted by
M. Kikkawa, Y. Okada, H. Yajima, and N. Hirokawa (University of
Tokyo); E.P. Sablin and R.J. Fletterick (University of California,
San Francisco).
Research funding: Ministry
of Education, Science, Sports, and Culture of Japan; National Institutes
of Health. Operation of the ALS is supported by the U.S. Deptartment
of Energy, Office of Basic Energy Sciences.
Publication about this
research: M. Kikkawa, E.P. Sablin, Y. Okada, H. Yajima, R.J. Fletterick,
N. Hirokawa, "Switch-based mechnanism of kinesin motors,"
Nature 411, 439 (2001).
ALSNews
Vol. 198, May 8, 2002