Foresight Institute Logo
Image of nano

Physical and technical parameters determining
kinesin-driven microtubule motility in vitro

Roland Stracke*, a, Konrad J. Böhma, Jörg Burgoldb, Hans-Joachim Schachtb, Eberhard Ungera

aInstitute of Molecular Biotechnology,
Beutenbergstrasse 11, D - 07745 Jena, Germany

bInstitute for Microsystem Technique, Mechatronics and Mechanics, Technical University Ilmenau,
Max-Planck-Ring 12, D-98693 Ilmenau, Germany


This is a draft paper for the
Seventh Foresight Conference on Molecular Nanotechnology.
The final version has been submitted
for publication in the special Conference issue of


Kinesin is a prominent motorprotein, which is able to mimic cellular motility in vitro driving taxol-stabilized microtubules across kinesin-coated surfaces. This model system might provide a basis for future developments of nanoscaled machinery. The present study demonstrates that this system needs a minimal free working space of about 100-nm height and works up to 3 h with nearly constant velocity. Individual microtubules were observed to cover distances of at least 1 mm without being released from the surface. Microtubules gliding across kinesin-coated surfaces were shown for the first time to be unipolar aligned by application of flow fields.


Eukaryotic organisms synthesize specialized proteins to realize different intracellular transport and motility processes. These proteins, called motorproteins, are able to convert the chemical energy of adenosine triphosphate (ATP) into mechanical energy. Kinesins are prominent microtubule-associated members of the motorprotein group. Conventional kinesin from neurons forms dimers, consisting of two globular heads, a stretched about 80-nm long stalk, and a tail. The head and the tail domain contain the microtubule- and the cargo-binding site, respectively. Kinesin molecules translocate in 8-nm steps along the surface of microtubules which represent 25-nm thick and about 5 to 20-µm long hollow-cylindrical self-assembling proteinaceous rails, having a plus and a minus end. The direction of translocation depends on microtubule polarity. It was shown that neuronal kinesin moves to the plus end (figure 1).

Figure 1. Linear motoric unit of nanoactuatoric machinery based on kinesin/microtubule interaction. From technical point of view, microtubules can be considered to be the stator and the kinesin, which carries a certain cargo, the slide.

Under defined in vitro conditions, purified taxol-stabilized microtubules are able to glide across kinesin-coated glass surfaces. Gliding velocity can be regulated between 0.1 and 2.0 µm/s by different factors, including ATP and Mg2+ concentration, temperature, and kinesin density (Böhm  et al. in press). This motility system might provide a basis for future developments of nanoscaled devices, usable e.g. for a controlled and specific substance transport over small distances. Such developments require the solution of numerous methodological and technical problems, including the temporal stability of the system, the isopolar microtubule alignment for multiplication of forces generated by single driving units (actuators), the roughness of the kinesin-coated surfaces, or the minimal distance between two surfaces, which still allows kinesin functioning. The present study provides first results concerning these problems.

2. Material and methods

2.1. Protein preparation

Kinesin was purified from porcine brain homogenates by a combined procedure of ion exchange chromatography, microtubule affinity-binding, and gel filtration (Kuznetsov and Gelfand, 1986). Microtubule protein was isolated from porcine brain by two cycles of temperature-dependent disassembly/reassembly (Shelanski et al. 1973). Microtubules were formed by 20-min incubation of phosphocellulose-purified tubulin (1 mg/ml) at 37°C with 1 mM GTP and 10 µM taxol as assembly promoter.

2.2. Microtubule gliding

Taxol-stabilized microtubules (final concentration: 40 µg/ml tubulin) and kinesin (70 µg/ml) were supplemented with 0.5 mM MgATP (motility mixture). After 10 min, 10 µl of this mixture were dropped onto glass slides pretreated with 5 mg/ml bovine serum albumin (BSA), covered by a coverslip, and sealed with a mixture of Vaseline, lanolin, and paraffin (standard conditions; further details see Böhm et al. 1997).

Microscopy: The gliding microtubules were visualized by video-enhanced differential interference contrast microscopy (transmission mode, in particular cases in reflected light mode) using an Axiophot microscope (ZEISS) equipped with a Chalnicon video camera (HAMAMATSU) and the image processing system Argus 50 (HAMAMATSU). Image processing was performed following instructions of Weiss and Maile (1992). Gliding velocity was determined by measuring the distance the microtubules migrated within a defined time, using Argus-50 software.

3. Results and discussion

3.1. Isopolar alignment of gliding microtubules

It is known that the direction of gliding is determined by microtubule polarity. Under standard conditions, the microtubules move in all possible directions. Isopolar alignment of gliding microtubules seems to be a fundamental precondition for possible nanoactuatoric applications of the kinesin-microtubule force-generating system. For microtubule alignment, a buffer flow was applied to microtubules gliding across a kinesin-coated glass surface. To realize these experiments, a special chamber was constructed (figure 2). It consists of a microscopic slide as support, a channel formed by two strips of double-sided adhesive tape and a long coverslip with two holes, connecting the buffer reservoirs with the channel and the hose connections. A defined buffer flow in the channel was realized by a motor-driven syringe.

larger image: 1666x1202 pixels, 125K
Figure 2. Device for isopolar alignment of gliding microtubules in a flow field. a - side view,  b - top view; 1 - microscopic slide, 2 - coverslip with holes, 3 - buffer reservoir, 4 - hose connections, - channel (50 mm x 2 mm x 0.04 mm), 6 - holes in the coverslip.

At sufficiently high flow rates (some mm/s), the microtubules turned within 10 to 15 s and moved approximately in flow direction (figures 3 and 4).

larger image: 931x293 pixels, 111K
Figure 3. Flow-induced orientation of gliding microtubules, demonstrated by video-enhanced DIC microscopy. a - control without flow, b, c - microtubules aligned in flow direction (from left to right), 5-s interval between both images.


larger image: 1321x802 pixels, 122K
Figure 4. Diagram of flow-induced orientation of gliding microtubules. The relative number (n/N) of gliding microtubules is shown with respect to flow direction.


The supposed mechanism of alignment is that the weakly bound leading end (minus end) of the microtubules turns in flow direction, guiding the whole microtubule in flow direction (figure 5). This method provides the possibility to align large numbers of microtubules enabling them to a common action in a desired direction.

larger image: 1440x943 pixels, 112K
Figure 5. Mechanism of microtubule alignment in a flow field.

3.2. Determination of the tolerable surface roughness

For functioning of the motor system, the quality of the surface to which the kinesin binds seems to be of great importance. BSA-pretreated polished silicon wafers with etched cavities or steps of defined height and edge steepness were used (figure 6) to check how the surface profile or impurities in form of small particles might affect gliding. Gliding was observed by video-enhanced DIC microscopy in reflected light mode. The maximum step height the microtubules could overcome upwards was found to be about 280 nm at an edge angle of 60-70°.


larger image: 1228x460 pixels, 32K
Figure 6. Scheme of microtubule movement across a silicon wafer with etched steps of defined height.

3.3. Microtubule gliding between glass surfaces of defined distance

A possible task in development of nanoscaled machinery is the relative movement of two thin plates. In this context, it is necessary to determine the minimal distance between two surfaces, which still allows kinesin-dependent force generation. Therefore, we studied microtubule gliding in clefts of variable and defined height.

To produce such clefts, a slightly curved coverslip (obtained by thermal deformation under one-sided pressure) was approximated (convex side down) to the microscopic slide to which the gliding microtubules were attached. Approximation was realized using a small vacuum chamber. Around the attachment point of both glasses, Newton fringes appear giving the possibility of interferometric determination of the distance between the inner surfaces. We found that the gliding velocity of microtubules depends on the gap height between both glasses. Motility was observed down to a distance of about 100 nm, which corresponds nearly to the length of the kinesin molecule. In a single case, a microtubule migrated into a gap of about 30 nm, where its leading edge stuck and the other parts showed a fishtailing-like movement.

3.4. Determination of the maximal time of function

For possible technical applications, it is important to know how long the system is able to work. In a sealed chamber, microtubules moved over a period of 3 hours with nearly constant velocity of 0.6 - 0.8 µm/s. That means that a microtubule can theoretically migrate over a distance of 6.5 - 8.6 mm within this time interval. To study whether the microtubules were attached to the surface all the time, the path of individual microtubules was followed by displacement of the microscope stage, to keep them in the observation field. It was observed that single microtubules migrated over distances up to 1 mm without detaching from the surface. Thereafter, they were lost either by inaccurate stage displacement or by impaired imaging.


Our results demonstrate that kinesin-driven gliding microtubules, which can be regarded as biomolecular nanomotors, are able to work continuously up to several hours on uneven surfaces with height differences up to 280 nm and in narrow chambers down to about 100 nm heights. It is possible to force microtubules to glide across kinesin-coated surfaces in a desired direction by application of flow fields. For technical usability of the described kinesin-based motility system, further studies are necessary concerning e.g. the determination of maximal loads to be transported and the stabilization of the protein components involved in force generation.


  • Böhm KJ, Steinmetzer P, Daniel A, Vater W, Baum M and Unger E 1997 Kinesin-driven microtubule motility in the presence of alkaline-earth metal ions. Indication for calcium ion-dependent motility. Cell Motility and Cytoskeleton 37 226-231
  • Böhm KJ, Stracke R and Unger E 2000 Speeding up kinesin-driven microtubule gliding in vitro by variation of cofactor composition and physicochemical parameters. Cell Biol. Int. (in press)
  • Kuznetsov SA and Gelfand VI 1986. Bovine brain kinesin is a microtubule-activated ATPase. Proc Natl Acad Sci USA 83 8530-8534
  • Shelanski ML, Gaskin F and Cantor CR 1973. Microtubule assembly in the absence of added nucleotides. Proc Natl Acad Sci USA 70 765-768
  • Weiss DG and Maile W 1992 in: Principles, practice, and applications of video-enhanced contrast microscopy. Electronic Light  Microscopy (Shotton DM, ed.) New York: Wiley-Liss, pp 105-140


Foresight Programs


Home About Foresight Blog News & Events Roadmap About Nanotechnology Resources Facebook Contact Privacy Policy

Foresight materials on the Web are ©1986–2017 Foresight Institute. All rights reserved. Legal Notices.

Web site developed by Stephan Spencer and Netconcepts; maintained by James B. Lewis Enterprises.