To study the interactions between motor domains and tubulin, three-dimensional density maps have been calculated from EM images of intact microtubules or opened-out tubulin sheets decorated with bacterially expressed single-headed and double-headed Kinesin-1, and Ncd constructs. In our work, we have used natural microtubules with 16 protofilaments. Others have used tubulin sheets or 10-15-protofilament microtubules reassembled from purified brain tubulin. In all cases, the motor domains bind to the protofilaments at intervals of 8 nm, the axial spacing of the tubulin dimer.
Opened-out tubulin sheets decorated with Kinesin-1 or Ncd motor domains and then negatively stained have bands of dark, light and intermediate density. It is possible to relate the orientation of the pattern to the plus and minus ends of microtubules by looking at brain tubulin sheets growing from the ends of microtubules seeded by flagellar axonemes or by centrosomes isolated from tissue culture cells. Tubulin sheets decorated with Kinesin-1 or Ncd show the same polarity. Furthermore, images of tubulin sheet ends have shown that both motors bind in the same position relative to these ends and thus to identical sites on tubulin dimers. Thus, it is possible to discount the idea that Kinesin-1 and Ncd move in opposite directions by binding to tubulin in opposite orientations or to equivalent sites on different monomers of the tubulin heterodimer. 3-D maps of motor-tubulin complexes have also shown very little structural difference between Kinesin-1 or Ncd motor domains or the way in which single heads attach to tubulin.
Hoenger et al. have reported significant rearrangements of the density in the microtubule, especially within the beta-tubulin subunit, as a result of motor binding but we found only small changes. Both groups agree that the tubulin lattice parameters are unchanged by motor binding. This is consistent with a report that even glutaraldehyde-fixed microtubules can support Kinesin-1 driven motility.
We have also investigated the structure of single-headed Kinesin-1 bound to tubulin under different nucleotide conditions; that is, in the absence of nucleotides or in the presence of AMP.PNP (a non-hydrolyzable ATP analogue), when the motor domains bind strongly to microtubules, or in the presence of ADP, when they bind weakly. When the stain patterns on Kinesin-1 decorated sheets were studied, no shift in the positions of Kinesin-1 heads relative to the tubulin dimers was detected between any of these stationary states. 3-D maps from negatively stained images of Kinesin-1-decorated microtubules confirmed that the relationship between bound heads and tubulin was essentially unchanged but there appeared to be a structural change in the angle of the most distal part of the Kinesin-1 motor domain. This part was perpendicular to the microtubule axis in the presence of ADP, but pointed towards the plus end in the absence of nucleotides or in the presence of AMP-PNP. This result implies that part of a Kinesin-1 head may tilt towards the microtubule plus end when ADP is released and is consistent with a model in which structural changes in the motor domain result in directional movement.
3-D maps from cryo-EM images showed that, in the presence of AMP-PNP and under saturating binding conditions, both double-headed motors bind to microtubules via one head, with the second heads apparently tethered in fairly fixed positions. The attached heads of Kinesin-1 and Ncd dimers are very similar to single Kinesin-1 heads and bind to tubulin in the same way. In both cases, the tethered heads appear to be connected to the top right region of the first heads when the plus end of the microtubule is oriented upwards. The unattached second heads seem to emanate from slightly different points on the first heads and tilt in different directions (see Figure 1); this is consistent with pairs of Kinesin-1 motor domains being associated at their C-termini while Ncd motors are joined at their N-termini. Since the tethered head of Kinesin-1 is oriented so that it could more easily reach the next site in the plus direction on the same protofilament and the second Ncd head points towards the next site in the minus direction, this structural difference is likely to facilitate their opposite directions of movement.
Contributed by Keiko Hirose, Linda Amos & Rob Cross
View more images of motor-decorated microtubules (Rob Cross Motors Page)
View list of available fluorescent ATP analogues (Rob Cross Motors Page)
Go to the Kinesin Crystal Structure page.
Go to the Domain Organization & Conformational Changes page.
Return to the Kinesin Home Page.
If you use material from the reviews at the Kinesin site, we ask that you cite either the home page and author, or the appropriate source publication in your work.
Copyright 1997-2005. All rights reserved.
Created 21 January 1997 12:00 GMT
Modified 14 July 2006 18:40 GMT