Watching molecular motors at work
by video-enhanced light microscopy

INTRODUCTION

Movement and shape changes of single cells and whole organisms, as well as intracellular transport are fundamental aspects of life. This biological motility results from the specific interaction of motor proteins (e.g., myosins, kinesins or dyneins) with filamentous cytoskeletal substrates: e.g., myosins interact with actin filaments, whereas kinesins and dyneins interact with microtubules . Motor proteins generate force in molecular steps and are fueled by ATP hydrolysis. In vitro motility assay systems have been developed that require only a motor protein, its substrate, and ATP to reconstitute motility [Kron et al., 1991]. Through them, our current view of the biochemistry and biophysics of cyclic chemo-mechanical force generation has been greatly advanced.

Figure 1. Motor proteins - structure and interaction with their substrate . (a) Electron micrograph of glycerol-sprayed/rotary metal-shadowed muscle myosin molecules which consist of two globular heads and a long, 2-stranded a-helical coiled-coil tail. (b) Schematic representation of the myosin molecule which is composed of two copies each of a heavy chain, and a regulatory and essential light chain [c.f. Rayment and Holden,1994]. For simplicity, only the two heavy chains are shown here. Myosin polymerization into thick filaments as they are found in the muscle sarcomere proceeds by lateral association of the myosin tails. (c) Interaction of myosin with actin filaments. In this cartoon, two highly schematic myosin heads depicted in light grey illustrate that one head interacts with two adjacent actin subunits [Rayment et al., 1993; Bremer et al., 1994]. (d) Schematic representation of the myosin ATPase/cross-bridge cycle.

All motor proteins, both the actin-based myosins (see Fig. 1) as well as the microtubule-based kinesins and dyneins, share a characteristic two-domain organization [c.f. Rayment & Holden, 1994; Kull et al., 1996]. As illustrated in Fig. 1b, a globular head harbors the motor and the ATPase activity, while a more or less extended tail domain interacts with the motor's cargo (e.g., vesicles, organelles or chromosomes), or with itself to form supramolecular assemblies (e.g., myosin thick filaments). As depicted in Fig. 1c, one myosin head binds to two adjacent actin subunits [Rayment et al., 1993].

Motor proteins transduce the free energy change accompanying the enzymatic hydrolysis of the terminal phosphate of bound ATP into directed movement. For example, myosin is an actin-activated ATPase that rapidly hydrolyzes bound ATP (Fig. 1d: transition from top to right) to yield a ternary complex consisting of myosin and ADP+Pi . This complex binds with high affinity to actin filaments (Fig. 1d: transition from right to bottom). Subsequent release of Pi (Fig. 1d: transition from bottom to left) triggers the "power stroke", i.e., a large conformational change of the myosin head that provides the force for its movement relative to the actin filament. Exchange of bound ADP with ATP (Fig. 1d: transition from left to top) completes the "ATPase/cross-bridge cycle" by lowering the affinity of the myosin head for actin filaments so that they are subsequently released.

PRINCIPLE

The light microscope (LM) allows dynamic biological processes to be imaged in their native (i.e., aqueous) environment with relatively high temporal resolution (i.e., on a subsecond time scale). However, the diffraction-limited resolution of the LM is relatively low, i.e., of the order of 250nm. Although the ~9-nm diameter of actin filaments is more than one order of magnitude below this theoretical resolution limit, single fluorescently labeled actin filaments are readily visualised in the LM (Fig. 2). The reason for this apparent discrepancy is that the LM is used to probe whether the area that is imaged contains fluorescent molecules or not, and bright fluorescence can be detected even if the fluorescent object is significantly smaller than the resolution limit. Nevertheless, since the LM cannot overcome the limits set by the laws of diffraction -limited optics, the apparent width of a fluorescent actin filament is defined by and equal to the resolution limit and cannot therefore be interpreted in a straightforward way. How ever, filament lengths and/or velocities can be determined accurately since these spacings are large compared to the resolution limit.

When working at or beyond the diffraction-limited resolution of the LM, a disadvantage of fluorescence imaging is the relatively low signal-to-noise (S/N) ratio of the images. However, this can be increased significantly by video and computer technology. A computer-aided , video-enhanced LM system such as shown schematically in Fig. 3, serves 3 different functions: imaging, image process ing, and image storage [for a review, see Inoué, 1986]. Images are recorded by a video camera system (i.e., a low-light level CCD or SIT camera for fluorescence LM) and generate sig nals that can be enhanced in various ways by analog and digital image processing. The camera control unit is an "analog enhancer" that allows grey level selection and gain manipu lation to optimize the image. Digital image processors convert the analog to a digital signal and can store images for (i) background subtraction (e.g., to compensate for uneven illumination), and (ii) to compute moving averages over 2, 4, 8, 16, etc. video frames (e.g., to increase the S/N ratio). To optimize image contrast, most commercially available image processors also allow histogram equalization. Finally, the processed digital images are re-converted into an analog signal that can be displayed on a TV monitor and/or recorded on a video tape.

Figure 2. Single fluorescently labeled actin filaments imaged in the LM.

Figure 3. A computer-aided, video-enhanced light microscopy (LM) system performing 3 different tasks: imaging, image processing, and image storage.

APPLICATION: Watching molecular motors at work

The experimental setup to perform an in vitro motility assay is relatively simple. As illustrated in Fig.4 (top), a flow cell (i.e., an open system that allows solution exchange during the experiment) is assembled from a glass slide and a nitrocellulose coated coverslip using silicon grease along the edges of the coverslip as spacer between the two surfaces. A myosin solution is then allowed to flow into the cell. Within seconds, the myosin molecules bind to the coverslip. The nitrocellulose prevents myosin denaturation on the highly charged glass surface. After a washing step to remove excess myosin, free binding sites in the flow cell are blocked to prevent actin filaments from sticking to the glass surface. Fluorescent derivatives of phalloidin (e.g., rhodamin-phalloidin) - a mushroom toxin that tightly binds to actin filaments - are used to fluorescently label actin filaments indirectly. When added to the flow cell, these fluorescent actin filaments bind to the myosin molecules, and the motility assay system is ready for use. As depicted schematically in Fig. 4 (bottom), adding "fuel" in the form of ATP produces a sliding movement of the actin filaments powered by the myosin motors attached to the coverslip (see Fig. 1d).

Figure 4. Experimental setup of an in vitro motility assay. (Top) Front (left) and side (right) view of a flow cell. Most in vitro motility assays are performed using flow cells, i.e., open systems that allow solutions to be exchanged by flowing them through the cell. In its simplest design, a flow cell consists of a coverslip that is mounted on a glass slide using silicon grease as a spacer (i.e., to achieve a separation of ~0.5-1.0mm). (Bottom) In this cartoon, myosin heads are depicted adsorbed to a nitrocellulose-coated coverslip and are in the process of moving an actin filament. The coverslip is shown upside down, i.e., the myosin molecules and the actin filaments face the slide [adapted from Warrick & Spudich, 1987].

Figure 5. Fluorescently labeled actin filaments. Rhodamine-phalloidin labeled actin filaments were imaged with a Zeiss Axiophot light microscope using a 63x/1.4 NA Planapochromat objective lens. Digital image processing steps included background subtraction, averaging over 4 video frames, and histogram equalization.

For a practical illustration, Fig. 5 reveals 3 time frames (i.e., 0, 8 & 16 sec) of an in vitro motility assay involving a myosin powered sliding movement of fluorescent actin filaments over a cover slip. In this experiment, individual actin filaments (three of which have been false colored in red, green and blue, respectively) move with an average velocity of about 3 mm/sec relative to the cover slip.

Inspired by successful studies of single ion channels by patch-clamp recordings, a novel in vitro assay employing a feedback-enhanced laser trap system has allowed direct measurement of force and displacement that results from the cyclic interaction of a single myosin molecule with a single actin filament [Finer et al., 1994]. Accordingly, discrete , stepwise movements averaging 11nanometers were depicted under conditions of low load, and single force transients averaging 3-4 piconewtons were measured under isometric tension. The magnitudes of the single motor forces and displacements are consistent with predictions of the conventional "swinging-crossbridge" model of muscle contraction [Huxley, 1996].

REFERENCES

Bremer, A., Henn, C., Goldie, K.N., Engel, A., Smith, P.R. & Aebi, U. (1994). Towards atomic interpretation of 3-D reconstructions of F-actin filaments. Journal of Molecular Biology 242, 683-700.

Finer, J.T., Simmons, R.M. & Spudich, J.A. (1994). Single myosin molecule mechanics - piconewton forces and nanometer steps. Nature 368, 113-119.

Huxley, H.E. (1996). A personal view of muscle and motility mechanisms. Annual Review of Physiology 58, 1-19.

Inoué, S. (1986). Video Microscopy. Plenum Press , New York.

Kron, S.J., Toyoshima, Y.Y., Uyeda, T.Q.R. & Spudich, J.A. (1991). Assays for actin sliding movement over myosin-coated surfaces. Methods in Enzymology 196, 399-416.

Kull, F.J., Sablin, E.P., Lau, R., Fletterich, R.J. & Vale, R.D. (1996). Crystal structure of the kinesin motor domain reveals a structural similarity to myosin. Nature 380, 550-555.

Rayment, I., Holden, H.M., Whittaker, M., Yohn, C.B., Lorenz, M., Holmes, K.C. & Milligan, R.A. (1993). Structure of the actin-myosin complex and its implications for muscle contraction. Science 261, 58-65.

Rayment, I. & Holden, H.M. (1994). The 3-dimensional structure of a molecular motor. Trends in Biochemical Sciences 19, 129-134.

Warrick, H.M. & Spudich, J.A. (1987). Myosin structure and function in cell motility. Annual Review of Cell Biology 3, 379-421.