Movement of Organelles Along Filaments Dissociated from the Axoplasm of the Squid Giant Axon

Home , Axoplasm

Cell, Vol. 40, 449-454, February 1985, Copyright 0 1985 by MIT 0092-8674/85/020449-06 $02.0010 Movement of Organelles Along Filaments Dissociated from the Axoplasm of the Squid Giant Axon

Ronald D. Vale,* Bruce J. Schnapp,t grade movement of cytoskeletal and soluble proteins at Thomas S. Reeset and Michael P. Sheetz* rates 100 times slower. Only materials transported by fast * Department of Neurobiology axonal transport utilize the mechanism of directed or- Stanford University Medical School ganelle movement (Grafstein and Forman, 1980). Stanford, California 94305 The recent application of video processing to polariza- TLaboratory of Neurobiology, NIH, NINCDS tion-based microscopy (Allen et al., 1981; Inoue, 1981) am- at the Marine Biological Laboratory plifies image contrast to the extent that structures below Woods Hole, Massachusetts 02543 the resolution limit of the light microscope can be made *Department of Physiology visible. Using video microscopy to visualize the squid gi- University of Connecticut Health Center ant axon, it was shown that organelles less than 200 nm Farmington, Connecticut 06032 in diameter move parallel to linear elements that could be single cytoplasmic filaments (Allen et al., 1982). In contrast to the saltatory movements of larger organelles, Summary these small organelles moved continuously at rates consistent with those reported for fast anterograde axonal Cytoplasmic filaments, separated from the axoplasm transport. These organelle movements continue in axo- of the squid giant axon and visualized by video- plasm that is extruded from the squid giant axon (Brady enhanced differential interference contrast micros- et al., 1982). copy, support the directed movement of organelles in Single filamentous elements associated with moving or- the presence of ATP. All organelles, regardless of size, ganelles occasionally separate from the edge of the iso- move continuously along isolated transport filaments lated squid axoplasm (Allen et al., 1983; Brady and Lasek, at 2.2 rt 0.2 pm/set. In the intact axoplasm, however, unpublished observations). Here we report the results of movements of the larger organelles are slow and salta- dissociating axoplasm to produce extensive fields of iso- tory. These movements may reflect a resistance to lated filaments where virtually all filaments support the movement imposed by the intact axoplasm. The uni- transport of organelles for many hours. We have used form rate of all organelles along isolated transport fila- video microscopy to examine the characteristics of or- ments suggests that a single type of molecular motor ganelle movements along these single filaments. A major powers fast axonal transport. Organelles can attach to advantage of this system is that organelle movement can and move along more than one filament at a time, sug- be examined in the absence of any impeding interactions gesting that organelles have multiple binding sites for exerted on organelles by an organized cytoplasmic gel, so this motor. characteristics of organelle movement should reflect directly the underlying molecular processes. Introduction Results Directed movement of vesicular organelles and mitochondria occurs in almost all types of cells (Schliwa, 1984), yet Organelle Movement in Intact Axoplasm the cellular machinery for generating this movement has We examined intact, extruded axoplasm from the squid gi- not been identified. Since the function of organelle move- ant axon in order to compare organelle movements in in- ment is to transport macromolecules from one compart- tact axoplasm (Allen et al., 1982; Brady et al., 1982) with ment of a cell to another, it is not surprising that this pro- organelle movements along isolated filaments in dis- cess is particularly well developed in neurons where sociated axoplasm. Organelles in the axoplasm moved in transport occurs bidirectionally in axons as long as a me- both directions at rates that varied, depending on the size ter or more (Grafstein and Forman, 1980). of the organelle, from approximately 2 pmlsec for the Rapid transport of material in axons has been examined smallest vesicles to less than 0.5 pm/set for the largest using radioactive labels to follow the fate of proteins syn- mitochondria(see Figure 2). Small organelles moved con- thesized in the cell body (Lasek, 1968); these experiments tinuously in one direction; larger organelles frequently revealed that certain proteins are transported in the an- stopped (saltatory motion) and sometimes moved back- terograde direction of rates of up to 400 mm/day. Other wards for a short distance. These movements in a proteins are transported in the retrograde direction at reversed direction have been interpreted as recoils (Allen rates of up to 200 mm/day (Kristennson and Olsson, 1971; et al., 1982); otherwise each organelle moved in only one Hendry et al., 1974). Materials moving in both the antero- direction. grade and retrograde directions at the fast rate are associated with membrane-limited organelles (Lorenz and Directed Organelle Movement in Dissociated Willard, 1978; Smith, 1980; Tsukita and Ishikawa, 1980). Axoplasm Collectively, these transport processes have been termed In dissociated preparations, l-30 discrete and well- “fast axonal transport” to distinguish them from “slow ax- separated filaments per viewing field of 400 ,um* adhered onal transport” (Lasek, 1982), which refers to a net antero- to the coverslip around the edge of the bulk axoplasm Cell 450

Figure 1. Typical Field of Transport Filaments and Associated Organelles A few small and medium-sized vesicular organelles are indicated by black triangles. A small mitochondrion (black and white arrowhead) contacts two filaments; subsequent observation showed that it was attached to both. Some stationary vesicular organelles are attached to the glass surface (open triangles); vesicles suspended in the medium are outside the plane of focus. Magnification 11,000x. Inset shows transport filaments (black triangles) settled on the glass surface at the edge of the bulk axoplasm (A).

(Figure 1). Filaments were oriented in all directions so the electron microscopy (Schnapp et al., 1985). Also, a variety original orientation in the intact axoplasm was no longer of evidence suggests that it is organelles that move at apparent. rapid rates through axons (Smith, 1980; Tsukita and Most of the filaments were distinguished by their sharp Ishikawa, 1980; Breuer et al., 1975; Lorenz and Willard, contrast with the substrate, their smooth, unkinked con- 1978). tours, and their uniform apparent diameter of approxi- A 1 :I dilution of buffer X with water in the presence of mately 0.2 ,um (Figure 1); these filaments were readily dis- 2 mM ATP (motility buffer) provided the best dissociations tinguished, especially by their lower contrast, from other, of transport filaments from the bulk axoplasm and sup- less common filaments. The true diameter of these fila- ported long-lasting and reproducible organelle movements could not be determined from the light microscopic ments. Preparations made in full-strength buffer X yielded image alone, because structures smaller than 0.2 pm fewer isolated filaments and more unattached organelles. generate an inflated image as a result of their diffraction In a buffer of lower ionic strength (20 mM KCI, 4 mM of light. Nevertheless, these filaments belonged to a MgCI,, 4 mM EGTA, 10 mM imidazole, pH 7.1) organelles recognizable class that we refer to as transport filaments were bound to transport filaments but rarely moved. No because of their obvious involvement in the organelle movements occurred in the absence of ATP with 1 mM so- movements described below. dium azide and 10 mM 2-deoxyglucose present to block Virtually all transport filaments supported directed endogenous production of ATP. Under these conditions movements of small particles, and all particle movements organelles remained bound to filaments. The transport were along these filaments. These particles are regarded filaments in motility buffer were not static but many were as organelles rather than macromolecular complexes be- bending and changing their shape. It was not clear cause they resemble organelles in the intact cytoplasm whether the movements of transport filaments could be (Allen et al., 1982), and because particles on isolated entirely explained by the movements of the associated transport filaments are membrane-limited as seen by organelles. Organelle Transport along Filaments 451

small particles medium particles 3 mitochondria

AR D A R D AR D Figure 2. Veloci ties of Small Organelles, Medium -Sized Organelles, and Mitochondria in Intact, Extruded Axoplasm (Anterograde, A, and Retrograde, R, Directions) and along Isolated Transport Filaments from Dissociated Preparations (D) Organelle velocity decreases with increasing size in intact axoplasm, but all classes of organelles move at the same velocity (2.2 f 0.2 rm/sec) along isolated filaments.

Movements along transport filaments were smooth and continuous, even for large organelles, and recoils or saltatory movements were infrequent. Organelles continued to move along isolated filaments for many hours, over 12 hr in one instance. Since rates of organelle movement remained constant for at least 90 min and then often showed a gradual decline, rates of directed organelle movement were measured during the early phase (see below). The Figure 3. Organelle Association and Dissociation from Transport frequency of organelles moving past reference points on Filaments transport filaments also declined with time. At 0:08 set two organelles (open and filled triangles) suddenly appear on a transport filament. By 0:24 set, one of the newly attached or- Rates of Organelle Movement ganelles (open triangle) has dissociated from the filament and the other organelle has moved along it to the right. Two seconds later nei- The sizes and shapes of organelles moving through intact ther organelle is attached to this filament. A vesicular organelle, at- axoplasm were similar to the sizes and shapes of or- tached to the right end of this filament during the first four frames, has ganelles moving along isolated filaments in dissociated left it by 2:48 sec. Times (in seconds) appear in the upper right of each axoplasm (Figures 1 and 2). Vesicular organelles were video frame. Magnification 10,000x. classified by size, and mitochondria by their elongate shape. Small organelles were defined as those with an ap- at 0.4 pm/set during periods of continuous movement over parent diameter of 0.2 pm, but many were presumably several micrometers (Figure 2). For all but the small or- much smaller. Medium-sized organelles varied between ganelles, rates of movement were the same in the antero- 0.2-0.53 pm in diameter, and large organelles were those grade and retrograde directions (Figure 2). larger than 0.53 pm. Elongate cylinders with diameters of Along isolated transport filaments, small- and medium- approximately 0.25 pm and lengths between 0.7 and 3.6 sized organelles, as well as mitochondria, moved at a pm were presumed to be mitochondria, though some may mean rate of 2.15 pm/.sec (Figure 2), which was remark- have been tubular-vesicular organelles (Smith, 1980; Tsu- ably consistent from preparation to preparation (.c 0.15 kita and Ishikawa, 1980). pmlsec; 23 preparations). Large spherical organelles Small organelles in intact squid axons move continu- (greater than 0.53 pm in diameter) were very infrequent ously with a mean rate of 2.5 pmlsec, but movements of along isolated transport filaments, but a few measure- larger organelles are considerably slower and intermittent ments indicated that they move somewhat slower than the (Allen et al., 1982). In intact, extruded axoplasm (Figure 2) other organelles. small organelles moved in the anterograde direction at 2.15 pm/set, and in the retrograde direction at 1.75 Interactions of Vesicles with Filaments pm/set; medium-sized organelles moved at 1.1 pm/set; Organelles suddenly attached to filaments and, either im- and large spherical organelles and mitochondria moved mediately or after a brief pause, began to move along Cell 452

Figure 4. Vesicular Organelle (Triangle) Attached to Two Transport Filaments The organelle moves along one filament (top) while attached to a second filament (arrow). The velocity of this organelle was less than nor- mal. Magnification 10,000x. them (Figure 3). Some organelles moving along filaments suddenly dissociated from them; attached and stationary organelles also dissociated from filaments (Figure 3). Once an organelle left a filament, it rapidly disappeared into the surrounding medium. Several observations indicated that organelles have multiple attachment sites for filaments. Mitochondria Figure 5. Organelle (Triangle) Reaches Cross-over with another Fila- moved along filaments with any or all of their length at- ment at 0:48 set, Switches to, and Then Moves along This Filament tached; unattached segments of mitochondria exhibited Magnification 10,000x. Brownian motion (Figure 1). Vesicular organelles and mitochondria sometimes attached to two transport fila- a different force-generating mechanism; in fact, different ments simultaneously (see Figures 1 and 4). Organelles myosin ATPases support different velocities of bead occasionally switched from one filament to another at movement in vitro (Sheetz et al., 1984). However, along points where filaments crossed (Figure 5). At some fila- isolated transport filaments, all organelles move at a sin- ment crossings every organelle switched filaments, while gle rate. This single rate of organelle movement in dis- at others only some of them switched. Organelles typically sociated axoplasm favors the idea that movements of all moved in only one direction along the filaments to which organelles are powered by the same species of molecular they switched. When mitochondria switched filaments, motor, which might depend on a single type of ATPase. one end attached to a crossing filament while the other The velocity of organelle movement along isolated end remained attached to the original filament; the trailing transport filaments, 2.2 ~m/sec, is the same as that of the end subsequently detached from the original filament and smallest organelles in intact axoplasm. The inverse rela- reattached to the crossing filament. We infer that filament tionship between organelle size and velocity in the intact switching by vesicular organelles also requires simultane- axoplasm probably depends on impeding interactions ous binding to two filaments because organelles rapidly with the axoplasm that could become increasingly impor- diffused away into the medium once they detached from tant for larger organelles, accounting for their saltatory a transport filament (Figure 3). movement. Differences in size within the class of small organelles could account for the different rates of antero- Discussion grade and retrograde movement of these organelles in the intact axoplasm (Smith, 1980; Tsukita and Ishikawa, 1980). Comparison of Organelle Movement in the These differences between organelle movement in in- Intact Axoplasm and along Isolated tact axoplasm and along isolated transport filaments un- Transport Filaments derscore the ambiguity inherent in pharmacological ex- Differences between the characteristics of organelle periments aimed at blocking transport in intact cytoplasm movement in intact axoplasm and along isolated transport or cells. Drugs or treatments could affect organelle trans- filaments provide insight into the mechanism of organelle port either directly through the transport machinery or in- transport in axons. In intact axoplasm, the larger the or- directly by disrupting the cytoskeleton. ganelle, the slower and more discontinuous or saltatory It is clear that the force production for organelle move- are its movements. It could be supposed that smaller or- ment does not require the complex tertiary structure of the ganelles have a higher density of force-generating sites or axonal cytoplasm. This casts doubt upon hypotheses that Organelle Transport along Filaments 453

would explain organelle transport by contraction of a sue paper, and washed with 5 PI of a 1 :l dilution of buffer X containing cytoplasmic lattice (Ellisman and Porter, 1980) or cytoplas- 2 mM ATP (motility buffer). Buffer X (aspartate, 350 mM; taurine, 130 mic streaming through low-viscosity channels (Weiss and mM; betaine, 170 mM; glycine, 50 mM; Hepes, 20 mM; MgCI,, 12.9 mM; EGTA, 10.0 mM; CaCI,, 3.0 mM; glucose, 1 mM, pH 7.2; see Brady et Gross, 1982). al., 1984) is similar to the internal medium of the squid giant axon. The axon was blotted again, placed on a glass coverslip (#II/z, Clay Adams, Interactions between Organelles and Parsippany, NJ), and transected. Axoplasm was extruded by pulling Transport Filaments the axon under a Pasteur pipette covered with polyethylene tubing, or by pulling the axon between the silicone rubber covered tips of #5 Organelles bind to and dissociate from isolated transport forceps. filaments. The number of organelles bound to and moving Intact axoplasm was examined directly by sealing it between two along transport filaments at any moment must, therefore, coverslips. To retard air drying, the volume of this simple chamber was depend upon the rate constants governing association reduced by applying a length of silicone grease on either side of the and dissociation as well as the concentraUon of orga- axoplasm. The upper coverslip was then pressed down onto wax spacers until the axoplasm just made contact with its surface. nelles and filaments. Dilution of organelles by the medium To make dissociated preparations, the axoplasm was placed on 40 would favor dissociation over association and, therefore, ~1 of motility buffer on the surface of a coverslip (22 x 60 mm; #O). A the number of organelles moving along filaments would second coverslip (22 x 22 mm or 22 x 40 mm) with wax spacers was decrease as a preparation ages. A function of the microtu- then placed on top of the axoplasm, and vertical pressure was applied and released to spread out the disrupted axoplasm. The coverslips bule domains in the intact axoplasm (Schnapp and Reese, were then sealed with a mixture (1:l:l) of Vaseline-lanolin-paraffin. 1982) might be to confine organelles near transport filaments to shift the equilibrium in favor of binding. Video Microscopy The finding that separate ends of the same mitochon- Axoplasm was viewed in a Zeiss ICM inverted microscope set up for drion can move along separate transport filaments shows differential interference contrast. Bias retardation was introduced by desenarmont compensation. The contrast of single transport fila- that each mitochondrion has several sites that are not only ments and of the small organelles on the video monitor was optimal capable of attaching to transport filaments but of interact- when the bias compensation for the combined optical-video system ing with them to generate movement. Recent observa- was between l/36 and l/20 of a wavelength (how to set up the optical tions of mitochondria moving in intact axoplasm and at- conditions to match optimally the performance of video tubes is con- troversial; e.g., Inoue, 1981; Allen et al., 1981). The rationale for the taching to more than one filament also led to the parameters selected in this paper, as well as a more detailed account conclusion that these organelles contain multiple sites of the video microscopy set-up and operation will be presented else- that interact with the force-generating system (Martz et al., where (Schnapp et al., submitted). The specimen was illuminated by 1984). Vesicular organelles also appear to have multiple a 100 watt mercury arc, which was heat-filtered and made monochro- matic for the 546 nm mercury peak with an interference filter. attachment sites. The image was projected out the side port of the ICM with a 12.5x corrected eyepiece and focused onto the Newvicon target of a Dage- Nature of the Molecular Motor MTI video camera (Michigan City, Indiana; series 65, 67, or 68) with a Organelle movement along transport filaments requires Fujicon 105 mm zoom lens and 2x telextender. An improvement in im- ATP, suggesting that an ATPase is involved in this process. age quality was achieved by replacing the zoom lens and telextender with a Zeiss 63 mm luminar lens. A 25x corrected ocular replaced the In the absence of ATP, organelles bind to filaments without 12.5x ocular to provide comparable magnification. moving along them so organelle binding to filaments may Visualization of transport filaments required a large amplification of be independent of their directed movement. The static, image contrast. The video gain was amplified by increasing the inten- bound state could be analogous to the condition of rigor sity of illumination (by aligning the optics for critical rather than Kohler illumination) and by adjusting the gain of the camera amplifier. At such that develops between myosin and actin in muscle (Adel- high gain the black level of the Dage cameras had insufficient range stein and Eisenberg, 1980), or dynein and microtubules in to keep the signal from saturating. Range was extended by a simple cilia (Gibbons and Gibbons, 1974) when ATP is absent. In modification of the camera control or by using the contrast expansion these two familiar examples, attachment and force gener- capability of the digital video processor in series with the camera. A ation are functions of a single molecule. mottled background pattern under these conditions of enhanced contrast was subtracted from the real-time video image with a digital video What cytoplasmic filamer+ are responsible for or- processor (Quantex DS-30; Quantex, Sunnyvale, CA), as previously ganelle transport and what is the molecular nature of the described with the Hamamatsu system (Allen and Allen, 1983). The ATPase? These questions have been difficult to answer analog output of the Quantex went through a video processor (#302-2 using whole axons or intact axoplasm because of the com- Colorado Video sync stripper and #604 video processor, Colorado Video Inc., Boulder, CO) to amplify slices of the gray scale. The output plexity of these systems. The dissociated filament prepa- of this analog processor was connected to a video recorder and a video ration provides an opportunity to answer these questions, monitor. Photographs of video images were taken primarily from a 16 in particular to determine the structure of the transport fila- mm movie film made directly from the video tapes by Image Transform ments (Schnapp et al., 1985). (N. Hollywood, CA), or from digitally averaged video images, using a 50-line Ronchi diffraction grating placed in front of the camera lens to Experimental Procedures minimize the contrast of the TV scan lines (Figures 1, inset, and 6). Rates of organelle movement along filaments having straight seg- Preparation of Intact and Dissociated Axoplasm ments that were at least 4 pm long were calculated from distances Squid (Loligo pealeii) were collected at the Marine Biological Labora- measured on the video monitor. tory in Woods Hole, MA. The giant axon was dissected under running sea water and ligatured just distal to the stellate ganglion. The ligatured Acknowledgments axon was then removed and placed in Ca*+-free artificial sea water at 2%, where it could be kept for several hours without any loss of We thank Raymond Lasek, Scott Brady, Susan Gilbert, and Robert Al- directed organelle movement. Just prior to use, the axon was cleaned len for many helpful discussions and for sharing unpublished observa- of surrounding connective tissue and small nerve fibers, blotted on tis- tions with us. We would also like to thank Robert Allen for use of his Cell 454

video microscopy system during the initial phase of this project, Dr. Schnapp, B. J., and Reese, T. S. (1982). Cytoplasmicstructure in rapid- Hideshi Fujiwakafor assistance in using it, and John Murphy for prepa- frozen axons. J. Cell Biol. 94, 667-679. ration of the figures. R. D. V. is a trainee in the National Institutes of Schnapp, B. J., Vale, R. D., Sheet& M. R, and Reese, T S. (1985). Sin- Health Medical Scientist Training Program (GM 07365). This work was gle microtubules from squid axoplasm support bidirectional move- supported in part by Hoechst-Rousselle Pharmaceuticals and NIH ments of organelles. (GM 33351; M. P S.). M. P. S. is an established investigator of the Amer- Smith, R. S. (1980). The short term accumulation of axonally trans- ican Heart Association. ported organelles in the region of localized lesions of single myelinated The costs of publication of this article were defrayed in part by the axons. J. Neurocytol. 9, 39-65. payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 Sheet& M. P, Chasan, R., and Spudich, J. A. (1984). ATP-dependent movement of myosin in vitro: characterization of a quantitative assay. solely to indicate this fact. J. Cell Biol. 99, 1867-1872. Received September 10, 1984; revised December 10, 1984 Tsukita, S., and Ishikawa. H. (1980). The movement of membranous organelles in axons. Electron microscopic identification of anterogradely and retrogradely transported organelles. J. Cell Biol. 84, 513-530. ‘Weiss, D. G., and Gross, G. W. (1982). The microstream hypothesis of axoplasmic transport: characteristics, predictions and compatibility Adelstein, R. S., and Eisenberg, E. (1980). Regulation and kinetics of with data. In Axoplasmic Transport, D. G. Weiss, ed. (Berlin: Springer- the actin-myosin-ATP interaction. Ann. Rev. Biochem. 49, 921-956. Verlag), pp. 330-341. Allen, R. D., and Allen, N. S. (1983). Video-enhanced microscopy with a computer frame memory. J. Microsc. 729, 3-17. Allen, Ft. D., Allen, N. S., and Travis, J. L. (1981). Video-enhanced differential interference contrast (AVEC-DIC) microscopy: a new method capable of analyzing microtubule-related movement in the reticulopodial network of Allogromia laticollaris. Cell Motil. 7, 291-302. Allen, R. D., Metuzals, J., Tasaki, I., Brady, S. T., and Gilbert, S. P. (1982). Fast axonal transport in squid giant axon. Science 278, 1127-1128. Allen, R. D., Brown, D. T, Gilbert, S. P, and Fujiwaka, H. (1983). Trans- port of vesicles along filaments dissociated from squid axoplasm. Biol. Bull. 765, 523. Brady, S. T, Lasek, R. J., and Allen, R. D. (1982). Fast axonal transport in extruded axoplasm from squid giant axon. Science 278, 1129-1131. Brady, S. T, Lasek, R. J., Allen, R. D., Yin, H. L., and Stossel, T. P (1984). Gelsolin inhibition of fast axonal transport indicates a require- ment for actin microfilaments. Nature 370, 56-58. Breuer, A. C., Christian, C. N., Her&art, M., and Nelson, l? G. (1975). Computer analysis of organelle translocation in primary neuronal cul- tures and continuous cell lines. J. Cell Biol. 65, 562-576. Ellisman, M. H., and Porter, K. R. (1980). Microtrabecular structure of the axoplasmic matrix: visualization of cross-linking structures and their distribution. J. Cell Biol. 87, 464-479. Gibbons, B. H., and Gibbons, I. R. (1974). Properties of flagellar “rigor waves” formed by abrupt removal of adenosine triphosphate from ac- tively swimming sea urchin sperm. J. Cell Biol. 63, 970-985. Grafstein, B., and Forman, D. S. (1980). Intracellular transport in neurons. Physiol. Rev. 60, 1167-1283. Hendry, I. A., Stockel, K., Thoenen, H., and Iverson, L. L. (1974). The retrograde transport of nerve growth factor. Brain Res. 68, 103-121. Inoue, S. (1981). Video image processing greatly enhances contrast, quality and speed in polarization-based microscopy. J. Cell Biol. 89, 346-356. Kristennson, K., and Olsson, Y. (1971). Retrograde axonal transport of protein. Brain Res. 29, 363-365. Lasek, R. J. (1968). Axoplasmic transport in cat dorsal root ganglion cells as studied with 3H-leucine. Brain Res. 7, 360377. Lasek, R. J. (1982). Translocation of the neurbnal cytoskeleton and axonal locomotion. Phil. Trans. R. Sot. Lond. 229, 313-327. Lorenz, T, and Willard, M. (1978). Subcellular fractionation of intra- axonally transported polypeptides in the rabbit visual system. Proc. Natl. Acad. Sci. USA 75, 505-509. Martz, D., Lasek, R. J., Brady, S. T, and Allen, R. D. (1984). Mitochon- drial movement in axons: membranous organelles may interact with the force generating system through multiple surface binding sites. Cell Motil. 4, 89-101. Schliwa, M. (1984). Mechanisms of intracellular organelle transport. In Cell and Muscle Motility, Vol. 5, J. W. Shaw, ed. (New York: Plenum Publishing Co.), pp. l-82.