Biol Cell (1992) 76, 291-301

291

© Elsevier, Paris

Review

Directed movements of ciliary and flagellar membrane components: A review Robert

A Bloodgood

Department o f Anatomy and Cell Biology, Box 439, and the Cancer Center, University of Virginia School of Medicine, Charlottesville, Virginia 22903, USA (Received I I September 1992; accepted 8 January 1993)

Summary - The ability to rapidly translocate polystyrene microspheres attached to the surface of a plasma membrane domain reflects a unique form of cellular force transduction occurring in association with the plasma membrane of microtubule based cell extensions. This unusual form of cell motility can be utilized by protistan organisms for whole cell locomotion, the early events in mating, and transport of food organisms along the cell surface, and possibly intracellular transport of certain organelles. Since surface motility is observed in association with cilia and flagella of algae, sea urchin embryos and cultured mammalian cells, it is likely that it serves an additional role beyond those already cited; this is likely to be the transport of precursors for the assembly and turnover of ciliary and flagellar membranes and axonemes. In the case of the Chlamydomonas flagellum, where surface motility has been most extensively studied, it appears that cross-linking of flagellar surface exposed proteins induces a transmembrane signaling pathway that activates machinery for moving flagellar membrane proteins in the plane of the flagellar membrane. This signaling pathway in vegetative Chlamydomonas reinhardtii appears to involve an influx of calcium, a rise in intraflagellar free calcium concentration and a change in the level of phosphorylation of specific membrane-matrix proteins. It is hypothesized that flagellar surface contact with a solid substrate (during gliding), a polystyrene microsphere or another flagellum (during mating) will all activate a signaling pathway similar to the one artificially activated by the use of monoclonal antibodies to flagellar membrane glycoproteins. A somewhat different signaling pathway, involving a transient rise in intracellular cAMP level, may be associated with the mating of Chlamydomonas gametes, which is initiated by flagellum-flagellum contact. The hypothesis that the widespread observation of microsphere movements on various ciliary and flagellar surfaces may reflect a mechanism normally utilized to transport axonemal and membrane subunits along the internal surface of the organelle membrane presents a paradox in that one would expect this to be a constitutive mechanism, not one necessarily activated by a signaling pathway. cilia / flagella / surface motility / axopodia / reticulopodia

Rapid plasma membrane dynamics restricted to specific plasma membrane domains associated with a microtubule cytoskeleton It is clear that cell surfaces perform work and that most of the dynamic features of cell surfaces result from the functional interaction of the plasma membrane with the cytoskeleton [24, 42, 92]. Within vertebrate cells, most movements of cytoplasmic membranous organelles are thought to be d r i v e n b y motors associated with the microtubular cytoskeleton while most aspects o f plasma membrane associated force (cytokinesis, capping of receptors, endocytosis, whole cell locomotion) are thought to be driven by motors associated with the actin cytoskeleton [72]. Genetic analysis is demonstrating that these generalizations may also apply to invertebrate systems such as Dictyostelium [66, 93] and Caenorhabditis [44]. However, not all plasma m e m b r a n e domains are associated with the actin cytoskeleton. There is a class of eukaryotic cell surface extensions that are' organized around the microtubular cytoskeleton. In metazoan cells, cilia and flagella represent the major examples of this type of plasma m e m b r a n e domain; in this case, the associated stable microtubular cytoskeleton has a specialized name, the axoneme. However, among protistan cells, in addition to cilia and flagella, there are a number of other examples of cell surface plasma membrane extensions that are

organized around microtubule arrays. Examples include the haptonema of Prymnesiophyte algae [55], the axopodia of Heliozoan protozoa [84] and the reticulopodia of Foraminiferan protozoa [88]. A unique class of cell surface associated force transduction has been visualized in association with the plasma membrane of cilia, flagella, haptonema, axopodia and r e t i c u l o p o d i a by the m o v e m e n t o f p o l y s t y r e n e microspheres and other small inert marker objects bound to the external surface of the cell (table I; figs 1, 2) [14]. All of the examples of rapid cell surface motility listed in table I occur in association with a long asymmetric extension of the plasma membrane containing microtubules that are closely apposed to the plasma membrane. Even taking in consideration only the microsphere movements associated with cilia and flagella, the use of Chlamydomonas mutants and the observations on primary cilia of tissue culture cells rule out the involvement of outer dynein arms, inner dynein arms, radial spokes, and the central pair complex in the production of force at the surface of the plasma membrane. The observations on haptonema, axopodia and reticulopodia, which contain only singlet ° microtubules, rule out the necessity for doublet microtubules. In general, the microsphere movements described in table I are rapid ( 1 - 8 tzm/s), linear, bidirectional, and saltatory (microspheres stop, start and reverse direction). The

292

RA BIoodgood

Table I. Rapid cell surface motility visualized by microsphere movements. Flagella Flagella Flagella Flagella Flagella Primary cilia Motile cilia Immotile cilia

Chlamydomonas Chlamydomonas Chlamydomonas Chlamydomonas Chlamydomonas Rat kangaroo Sea urchin blastulae Sea urchin blastulae

Gametes [38, 49, 83] V.egetative cells [8, 21] Dynein outer arm mutants [15] Radial spoke mutants [15] Central pair mutants [15] Cultured PtK I cells (Bowser, unpublished results) Somatic cells [10] Apical tuft cells [10]

Haptonema Axopodia Axopodia Reticulopodia Reticulopodia

Chrysochromulina Echinosphaerium Heterophrys Allogromia Astrammina

Prymnesiophyte algae [55] Heliozoan protozoan [9, 54, 84] Centrohelidian protozoan [90] Foraminiferan protozoan [27, 30] Foraminiferan protozoan [26]

Fig 1. Series of frames from a 16-mm movie taken with phase contrast optics showing the transport of 0.852 tzm diameter polystyrene microspheres outward along the surface of an axopodium of Echinosphaerium nucleofilunl. 3.3 s between adjacent micrographs. 2200 x . Reproduced from Bloodgood [9] Cell Biol Int Rep 2, 171-176 with permission.

r r: Fig 2. Selected frames from a phase contrast video recording of 0.45/zm diameter polystyrene microspheres (fat arrowheads) moving along the surface of a flattened reticulopodium of Allogromia. Two individual microspheres (solid fat arrow and open fat arrow) are moving in opposite directions along the pathway defined by a microtubule bundle. The narrow arrows are pointing to cytoplasmic membranous organelles moving along the surface of the same microtubule, x 4000. Reproduced from Bowser and Rieder [27] Can J Biochem Cell Biol 63,608-620 with permission.

Directed movements of ciliary and flagellar membrane components long saltations at a constant, rapid velocity rule out a diffusion driven mechanism. Where it has been examined, the velocity of microsphere movement appears to be load independent over a wide range of microsphere sizes [8, 13, 30]. Local force transduction occurring at all positions on the plasma membrane domain is suggested by the fact that different microspheres on the same cell extension move independently of each other and can pass each other moving in opposite directions [8, 30]. This observation effectively rules out mechanisms based on bulk membrane flow [47], membrane lipid flow [33] or surf-riding [25, 48]. The observation of rapid, local movement of polystyrene microspheres and other inert objects in contact with the external surface of the plasma membrane domain associated with certain microtubule-containing cellular extensions leads to the following assumptions: 1)the movements of the microspheres reflect the movements of plasma membrane components (presumably proteins); and 2) the movement of the microsphere and its associated plasma membrane components is driven by the interaction of the cytoskeleton with the cytoplasmic surface of the plasma membrane, and presumably requires the activity of a cytoskeleton-associated motor protein. An analogy can be made between the surface motility visualized by polystyrene microspheres and the movement of cytoplasmic membranous organelles along cytoplasmic microtubules driven by cytoplasmic dynein and kinesin [32]. If a small patch of the plasma membrane were wrapped up around the polystyrene microsphere, one would then essentially have a membranous organelle moving along a cytoplasmic microtubule. However, the formal possibility exists that the microtubules common to all o f the cellular extensions exhibiting polystyrene microsphere movements are necessary primarily for the maintenance of the cell extension and that the force for the microsphere movements is associated with a nonmicrotubular cytoskeleton, such as actin. Actin is present within reticulopodia of Allogromia [89] and in at least some cilia and flagella [36, 59, 71, 77]. In axopodia and reticulopodia, it has been assumed by some workers that the movement of cytoplasmic organelles and of surface associated microspheres represents two faces of the same mechanism [I, 90]. However, Suzaki and Shigenaka [84] present data showing very different velocity profiles and very different responses to a variety of inhibitors for organelles moving within the axopodia of the Heliozoan Echinosphaerium versus polystyrene microspheres moving along the axopodial surface. Two major differences exist between the ciliary and flagellar systems that express surface motility and the axopodia and reticulopodial systems that express surface motility: l) cilia and flagella have very stable microtubules while axopodia and reticulopodia have labile microtubules that are sensitive to anti-microtubule drugs; and 2) cilia and flagella do not contain motile membranous organelles while axopodia and reticulopodia contain a large population of membranous organelles that actively move in a bidirectional manner. Considerable effort has been directed toward the question of whether the motility of intracellular organelles and of polystyrene microspheres'on the surface of axopodia and reticulopodia are dependent upon microtubules. Edds [39] used a glass needle to create an artificial axopodium in the Heliozoan Echinosphaerium nucleofilum and demonstrated continued movement of intracellular organelles, even in the presence of 2.5 mM colchicine. Suzaki and Shigenaka [84] reported that 20 mM colchicine inhibit-

293

ed the movement of polystyrene microspheres on the axopodial surface of Echinosphaerium akamae but did not inhibit organelle movements within the axopodium. In both of these studies, no transmission electron microscopy of drug-treated cells was performed in order to determine that there was a complete loss of microtubules. Bowser and Rieder [27] showed that 1 mM colchicine, I00 tzM nocodazole and 4°C temperatures all reversibly depolymerized microtubules within Allogromia reticulopodia and also reversibly inhibited both cytoplasmic organelle movements and surface polystyrene microsphere movements. Bowser and Rieder [27] demonstrated in dramatic video sequences that polystyrene microspheres on the surface of a flattened Allogromia reticulopodial network move precisely along the path of underlying microtubules; in fact, these workers visualized intracellular organelles and surface microspheres moving along the same microtubule bundle at the same time (fig 2). These observations suggest that, in reticulopodia of Foraminifera, the movement of intracellular organelles and surface associated polystyrene microspheres are related and that they both are dependent upon microtubules. In contrast, for axopodia of Heliozoa, the evidence suggests the possibility of different mechanisms for the movement o f intracellular organelles and surface-associated microspheres in this system; the movement of intraaxopodial organelles may not be dependent upon microtubules. For none of the examples of plasma membrane forcetransduction visualized by polystyrene microsphere movement is the identity of the motor molecules definitively known. It appears that axonemal dynein is not the motor responsible for rapid cell surface motility in cilia and flagella since primary cilia are lacking dynein arms and yet exhibit microsphere movements (Bowser, personal communication) and Chlamydomonas mutants lacking the outer or inner dynein arms associated with the outer doublet microtubules exhibit normal flagellar surface motility [15]. It has not been clearly demonstrated that microtubules, much less microtubule-associated motors, are responsible for all of the cases of rapid cell surface motility cited in table I. The evidence for microtubule involvement is strongest in reticulopodia of Allogromia [27]. Bowser et al [28] reported that 0 . 5 - 1 . 5 mM E H N A (an inhibitor of axonemal dynein) inhibited both organelle transport within and microsphere transport on the reticulopodia of Allogromia. Suzaki and Shigenaka [84] reported that 1 mM vanadate (also an inhibitor of dynein) inhibited organelle transport within axopodia but not microsphere movements on the surface of axopodia of Echinosphaerium akamae. However, interpretation of experiments using these inhibitors is very complicated, not the least because these inhibitors are not specific for dynein ATPases at these concentrations [35]. What are the physiological roles played by the forceproducing system visualized by the movement of polystyrene microspheres ? Among the examples shown in table I, there are represented four classes of physiological functions that may be dependent on the force-transducing mechanism being visualized by the movement of polystyrene microspheres and other inert marker objects.

Class 1: whole cell locomotion Whole cell gliding motility in Chlamydomonas [11, 15, 62] is driven by the activity of the flagellar surface [15]; non-

294

RA Bloodgood

gliding mutant cell strains of Chlamydomonas moewusii are also totally deficient in microsphere movement [63, 73]. In algae of the class Prymnesiophyceae, cells possess two motile flagella plus a third organelle, called a haptonema, which contains a small group of singlet microtubules. Although the haptonema probably functions primarily in food capture and transport to the cell body [55] (see also below), cells can also glide along a solid substrate by the action of the surface of the haptonema [60], much as is the case for the Chlamydomonas flagellum [15]. Cachon and Cachon [34] postulated that the axopodial surface movements of certain Actinopod protozoa are responsible for both movement of food organisms along the surface of the axopodia and in the gliding locomotion of the organism along a substrate.

podia and reticulopodia) function to capture food organisms and transport them towards the cell body to locations where they can be endocytosed. Joseph Leidy [61] noted that in the Heliozoan Actinophrys sol: "the smallest infusorians or algae brought into contact with the rays glide slowly along them to their base". Subsequently, a number of workers have reported on the movement of food organisms along the surface of Heliozoan axopodia [56, 85] and Foraminiferan reticulopodia [4, 31, 53]. A somewhat different example of whole organisms being transported along the surface ofAllogromia reticulopodia was described by Bowser et al [29]; these authors showed that juvenile offspring resulting from multiple fission were transported bidirectionally along the surface of reticulopodia.

Class 2: gamete interactions during mating

Class 4: transportation of materials within the cellular processes

Chlamydomonas gametes initiate mating by the adhesion of their respective flagellar surfaces to one another [43]. Following adhesion, the flagellar contact points migrate to the tips of the flagella in order to put the cell bodies into the proper orientation for subsequent cell fusion [43]. Many of the same drug treatments that inhibit microsphere movements (trifluoperazine, lidocaine, H-8) also inhibit the early events in mating [38, 65, 83]. From these observations, it has been suggested that the flagellar surface motility visualized by microsphere movements is utilized normally during the early events of mating i,n Chlamydomonas [43, 49]. Class 3: feeding mechanisms In Heliozoan, Radiolarian and Foraminiferan protozoan, the long microtubule filled extensions of the cell body (axo-

The three physiological manifestations of rapid cell surface motility discussed thus far (whole cell locomotion, gamete interactions during mating, and feeding) cannot be applicable to some of the systems where polystyrene microsphere motility has been observed (such as primary cilia on cultured kidney cells and sea urchin blastulae cilia). One hypothesis for a generalized function of the forcetransducing system being visualized as polystyrene microsphere movements on the external surface of the membrane that would be applicable to all of the systems in table I is the facilitated transport of membrane a n d / o r cytoskeletal components along the length of the process (be it a cilium, flagellum, axopodium, reticulopodium) in association with the cytoplasmic surface of the membrane. There is a need for a constant supply of protein subunits for the membrane and cytoskeleton during growth of these

,D

d

C

,

4 hi

~.:.

\

~..'

!:x.~,~:.~

?'; i:~:.~--,~;

'-~.:,

Fig 3. Redistribution of Chlamydomonas reinhardtii strain PF-18 flagellar membrane glycoproteins in response to crosslinking with FITC-labeled FMG-1 mouse monoclonal antibodies that recognize the 350-kDa flagellar membrane glycoproteins, e,f,g,h. Differential interference contrast images, a,b,c,d. Corresponding epifluorescence images showing the distribution of labeled monoclonal antibody. The cell in a was labeled in the cold and then fixed in the cold to show the unperturbed distribution of the membrane glycoproteins. The cell in b shows the earliest stage of redistribution occurring shortly after the cells were warmed to 25°C. c, d show later stages in the process of flagellar redistributiori of the labeled antibodies and presumably of the flagellar membrane glycoproteins recognized by the antibodies. This process is described in detail in [22]. The fluorescent labeling of the cell body is due primarily to cross-reactivity of this anti-carbohydrate monoclonal antibody with the cell wall; unlike the flagellum, the apparent loss of fluorescent intensity of the cell wall is not due to antibody redistribution. 1000 x.

295

Directed movements of ciliary and flagellar membrane components

cellular processes but also for maintenance of the intact processes. There is known to be turnover of membrane and axonemal proteins in intact flagella [12, 74]. Another situation in which a ciliary membrane may be involved in the transport of membrane precursors is the connecting cilium of rod and cone cells [5]. The connecting cilium and its associated membrane is the only contiguous pathway between the inner and outer segments of the photoreceptor; since there is extensive loss and replacement of rhodopsin-containing membrane sacs in the rod outer segment, it is thought that the connecting cilium is the likely site for transport of membrane lipid and opsin protein into the outer segment. Defects in this putative connecting cilium pathway for providing precursors to the rod outer segment would be predicted to result in degeneration of the rod outer segments and visual defects. In Usher's Syndrome, one of the family of diseases referred to as Retinitis pigmentosa, degeneration of the photoreceptors appears to be correlated with disorganisation of microtubules of the connecting cilium [2, 52]. Disorganization of the cytoskeleton could affect transport properties of the membrane of the connecting cilium. Bear in mind, however that there has been no direct demonstration of protein transport or polystyrene microsphere motility associated with the membrane of the connecting cilium. Another specialized example of irrtracellular transport involves the directed movements of membranous organelles within the microtubule containing cellular processes that exhibit microsphere movements, often in very close proximity to the plasma membrane [1, 90]. As pointed out above, this function applies primarily to protozoan axopodia and reticulopodia but not to cilia, flagella or haptonema. Recently, Bowser ([75] and unpublished results), while studying primary cilia from cultured PtKI cells, and Kozminski et al [59], while studying the Chlamydomonas flagellum have observed rapid bidirectional transport of material within cilia/flagella at rates comparable to or exceeding that of the polystyrene microsphere movements. An important question is whether the force transduction being visualized by the use of large inert objects that are visible in the light microscope (microspheres) somehow activates the machinery for driving the movements of the microspheres (and the membrane components to which the microspheres adhere) or whether there is constitutive movement of membrane components that is only visible when a large object adheres to a membrane component or domain. In the former case, one would have to invoke a transmembrane signaling pathway that is activated by the contact of the microsphere with the membrane surface. Two examples can be cited that support the argument that contact of microspheres with the plasma membrane surface can induce alterations in the underlying cytoskeleton and the organization of membrane components. When polycationic beads attach to the external surface of the plasma membrane of Aplysia bag cell growth cones, they stimulate the assembly of f-actin in association with the cytoplasmic surface of th~ membrane underlying the bead; this actin polymerization appears to be responsible for the movement of the bead along the surface of the growth cone membrane [40]. Poly-lysine and poly-ornithine coated polystyrene microspheres, when attached to the plasma membrane of cultured Xenopus striated muscle cells, induced clustering of acetylcholine receptors to the site of contact of the membrane with the

microsphere [69]; using transmission electron microscopy, it was shown that the microsphere contact induced structural specializations reminiscent of postsynaptic structures normally induced by nerve-muscle interactions, including a meshwork of thin filaments resembling actin [67, 68]. The clustering of acetylcholine receptors is sensitive to colchicine but the maintenance of the receptor clusters is dependent on actin [7, 37]. The observation that a microsphere binding to the external surface of a plasma membrane can alter the organization of the cytoskeleton on the cytoplasmic side of the membrane suggests that the contact of the microsphere with the membrane is activating a transmembrane signaling pathway. Evidence that a transmembrane signaling pathway, induced by flagellar membrane glycoprotein cross-linking, activates the machinery for movement of glycoproteins within the Chlamydomonas flagellar membrane will be the subject of the second half of this review.

100-

%

Control

50.

a O100.

%

b

50

0 lOO-

% C

1 lag / ml

50

o, 100 -

%

d

2 lag / ml

50

o 100

% e

EtOH Control

5lag/ml

50

o

Fig 4. Histogram showing the in vivo dose-dependent inhibition of FMG-1 antibody induced flagellar glycoprotein redistribution in Chlamydomonas reinhardtii by the protein kinase inhibitor staurosporine in the presence of 20 t~M free calcium in the medium. The control sample had no drug or solvent added. The ethanol control (EtOH) contains the concentration of ethanol (2.5%) present in the highest drug concentration shown (5/~g/ml staurosporine). Reproduced with permission from Bloodgood and Salomonsky [20] Eur J Cell Biol 54, 85-89.

296

RA Bloodgood

Transmembrane signaling pathways associated with flagellar surface motility in vegetative Chlamydomonas reinhardtii The Chlamydomonas flagellum has proven to be an exceptionally useful experimental system in which to dissect the mechanism of rapid plasma membrane surface motility and to investigate the transmembrane signaling pathways that may be involved. Microspheres that adhere to the surface of the Chlamydomonas flagellum are transported bidirectionally along the flagellar surface at rates of 1 - 2 # m / s [8, 21, 49]. Chlamydomonascan glide along a solid or semisolid substrate [11, 62]; this whole cell locomotion results from the activity of the flageilar surface. Mutant cell lines of Chlamydomonas that are defective in gliding motility do not exhibit microsphere movements along the flagellar surface [63, 73] demonstrating that microsphere movements reflect the activity of the motor responsible for whole cell gliding motility. The dominant protein of the flagellar membrane from both vegetative and gametic cells of Chlamydomonas reinhardtii is a 350-kDa family of glycoproteins [3, 13, 82, 94]; use of an immobilized io.dination system has demonstrated that this is the principal flagellar surface protein that contacts the

substrate during gliding motility [17]. Monoclonal antibodies to repeated carbohydrate epitopes on the 350-kDa glycoproteins induce the active redistribution of this population of glycoproteins in the plane of the flagellar membrane of both vegetative and gametic ceils (fig 3) [22]; the induction of glycoprotein movements requires crosslinking of the external domains of the 350-kDa glycoproteins ([15, 51]; Bloodgood, unpublished results), as is the case in the more traditional forms of receptor 'capping' [80, 87]. In order to determine the relationship between microsphere movement and the movements of the 350-kDa flagellar membrane glycoproteins, monoclonal antibodies and concanavalin A were utilized in conjunction with fluorescence-activated cell sorting (FACS) to select a mutant cell line (designated L-23) that exhibited dramatically increased binding of concanavalin A to the flagellar m e m b r a n e glycoproteins (including the 350-kDa glycoproteins) [23]. Treatment of the L-23 mutant ceils with 100/~g/ml of concanavalin A resulted in no redistribution of flagellar glycoproteins whereas wild type cells were induced to redistribute their flagellar membrane glycoproteins by this concentration of concanavalin A. At 10/.zg/ml of concanavalin A, both wild type and L-23 mu-

In Vitro Phosphorylation Membrane - Matrix Compartment

°

~1 ~ ."

-..

O

i"

116k

•

97k

°

66k

• 0

0

45k

•

29k

•

0

b

Low Calcium

High Calcium

Fig 5. Autoradiogram of 2-D O'Farrell gels of Chlamydomonas reinhardtii flagellar membrane-matrix proteins phosphorylated in vitro with 32p-ATP in the presence of 1 mM EGTA alone (low calcium) or in the presence of 2 mM EGTA and 2.5 mM CaCI 2 (high calcium). Methods are given in Bloodgood [16].

Directed movements of ciliary and flagellar membrane components

297

a substrate during gliding, or contact with another flagellum during mating. The working assumption is that antibody induced crosslinking of the 350-kDa glycoproteins induces a transmembrane signal that activates the machinery for moving the glycoproteins in the plane of the flagellar membrane and that this machinery is also responsible for microsphere movements and gliding motility. The glycoprotein redistribution phenomenon has been exploited to study the signaling pathway whereby cross-linking of the external domains of plasma membrane proteins can result in the activation of cytoplasmic machinery (presumably associated with the cytoskeleton). Microsphere movement, gliding motility, and antibody induced glycoprotein redistribution all require micromolar concentrations of free calcium in the medium [19, 21]. Antibody induced glycoprotein redistribution is inhibited by three different types of calcium channel blockers (diltiazem, methoxyverapamil (D-600) and barium chloride) at permissive concentrations of calcium [19]. These observations suggest that flagellar membrane glycoprotein cross-linking opens up calcium

tant cell lines were induced to redistribute their flagellar glycoproteins. Under all conditions where membrane glycoproteins were able to move in the plane of the flagellar membrane, the cells exhibited polystyrene microsphere movements; where the flagellar glycoproteins were prevented from moving, no microsphere movements occurred [18]. These experiments suggested, but did not prove, that m o v e m e n t o f the 350-kDa flagellar m e m b r a n e glycoproteins within the plane of the membrane was necessary for the expression of microsphere movements and, by inference, gliding motility. As noted previously [15], the characteristics of the antibody-induced movement of the 350-kDa flagellar glycoproteins (slow, bulk movement) differ from the characteristics of microsphere movement along the flagellar surface (rapid, local). This difference may, in part, be explained by the fact that global cross-linking of the 350-kDa glycoproteins on the flagellar surface with antibodies surely differs from the local cross-linking and signaling phenomena that are postulated to result from microsphere binding, contact of the flagellar surface with

Protein Phosphorylation Flagellar Membrane - Matrix

J 9,, tI

116k • 97k • 66k •

"qD °

45k

•

29k

• 0 #,p

In vitro (+Ca)

In vivo

Fig 6. Autoradiograms of 2-D O'Farrell gels comparing Chlamydomonas reinhardtii flagellar membrane-matrix proteins phosphorylated in vitro (at high free calcium using ~2p-ATP) and in vivo (using 32P-orthophosphoric acid). There is a striking difference in the patterns of polypeptides phosphorylated in vivo and in vitro. The large spot of label at the bottom of the in vivo autoradiogram represents labeled phospholipids.

RA Bloodgood

298

In Vivo Phosphorylation

Antibody

Control 205 •

66Li

45. 29

6

•

0

4.55.0 5.4

5.7

4.55.0

5.4 5.7

Fig 7. Autoradiograms of 2-D O'Farrell gels of Chlamydomonas reinhardtii flagellar membrane-matrix proteins that were phosphorylated in vivo in the absence or presence of antibody cross-linking of flagellar membrane glycoproteins with the FMG-I monoclonal antibody. One major phosphorylated component with a molecular mass of approximately 60000 and a pl in the range of 4.8-5.0 is observed to be dephosphorylated when the transmembrane signaling pathway is activated.

conductance channels in the flagellar membrane allowing an influx of calcium and a rise in the intraflagellar free calcium concentration. The calmodulin inhibitors trifluoperazine (TFP) and W-7 also inhibit antibody-induced redistribution of flagellar glycoproteins [19]. This observation must be interpreted cautiously as these agents can have cellular effects independent of calmodulin; for instance, they can stimulate phospholipase C (PLC) activity resulting in an increase in inositol trisphosphate (IP3) [41]. Both TFP and W-7 have also been reported to inhibit protein kinase C [78, 79, 86]. Since the calciumdependent protein phosphatase, calcineurin, is activated by calmodulin, its activity is also inhibited by T F P [57]. Calcium as a signal could be acting through a variety of pathways; 0.05°7o Nonidet P-40 extracts of Chlamydomonas reinhardtii flagella (referred to here as the membrane-matrix fraction) contain calcium activated ATPase activity, calcium activated protein kinase activity and calcium activated protein phosphatase activity [6, 16, 91]. Three different protein kinase inhibitors (H-7, H-8, staurosporine) inhibit antibody induced redistribution of flagellar membrane glycoproteins in vivo (fig 4) [20] suggesting that protein phosphorylation may be an important component in the signaling pathway and leading to the hypothesis that calcium may act through a calcium activated protein kinase (fig 8A). Using an in vitro

phosphorylation system, it has been demonstrated that the flagellar membrane-matrix fraction contains a number of proteins whose phosphorylation in vitro is dependent upon micromolar calcium [16] (fig 5). Since in vivo and in vitro patterns of protein phosphorylation can differ, we sought to compare the phosphorylation pattern of membrane matrix proteins phosphorylated in vivo using 32p-orthophosphoric acid with those phosphorylated in vitro using 32p-ATP or 35S-ATPgammaS; the in vivo pattern of protein phosphorylation was very different from that observed in vitro in the presence of either high or low calcium (fig 6). This led us to focus our attention on in vivo phosphorylation and to ask whether antibody stimulation of flagellar membrane redistribution induced a change in the pattern of phosphorylation of flagellar membrane-matrix proteins. Based on the effects of protein kinase inhibitors in vivo [20], we expected to find one or more polypeptides whose level of phosphorylation increased upon the induction of the signaling pathway. To our considerable surprise, the most dramatic change in the phosphorylation pattern was an almost complete dephosphorylation of a polypeptide with a molecular mass of about 60000 and pl of 4 . 8 - 5 . 0 (fig 7) suggesting that flagellar membrane protein cross-linking may induce an influx of calcium and a rise in intraflagellar calcium that stimulates a calciumdependent protein phosphatase (fig 8B). This is support-

Directed movements of ciliary and flagellar membrane components

=,

iii i": p

a **

A

D

"

a ÷*

B Fig 8. Two contrasting flagellar signaling pathway models. In each case, membrane-protein cross-linking induces a calcium influx. The subsequent rise in intraflagellar free calcium concentration leads to a change in the level of phosphorylation of a flagellar membrane-matrix protein resulting in: 1) coupling of the cytoskeleton to the membrane; and 2) activation of a motor responsible for moving the membrane glycoproteins within the plane of the flagellar membrane. A. Calcium is activating a protein kinase that activates the relevant protein through phosphorylation. B. Calcium is activating a protein phosphatase that activates the relevant protein through dephosphorylation.

ed by the f u r t h e r o b s e r v a t i o n ( B l o o d g o o d and Salomonsky, unpublished) that treatment of cells with 8°7o ethanol also induces a complete dephosphorylation of the same 60-kDa flagellar membrane-matrix protein. Ethanol at this concentration is known to elevate intracellular calcium levels and to activate Chlamydomonas moewusii gametes [64, 81]. The highly specific nature of the dephosphorylation event and the known involvement of calcium in flagellar signaling suggest that calcineurin (the calcium-calmodulin dependent protein phosphatase, also called PP2B) [57] may be one of the enzymes activated by the flagellar signaling pathway. This is consistent with the fact that the calmodulin antagonists trifluoperazine (TFP) and W-7 inhibit antibody-induced flagellar membrane glycoprotein redistribution [19]; however, these agents clearly have other biological effects .4[ 1]. Halpain. and Greengard [45, 46] observed that binding of hgands to a class of glutamate receptors in neurons stimulated the dephosphorylation of a microtubule-associated protein, MAP2, and argue that calcineurin is the phosphatase activated by the receptor induced signaling pathway. There appears to be an alternative flagellar membrane signaling pathway which is unique to Chlamydomonas gametes; this pathway is initiated by the interaction of ga-

299

metic flagellar membrane specific glycoproteins called sexual agglutinins and results in a transient rise in intracellular c A M P levels [43, 65, 70], the activation of a c A M P dependent protein phosphorylation event [76], the redistribution of the sexual agglutinin glycoproteins in the plane of the flagellar membrane [50], and the migration of the flagellar membrane contact sites such that the flagellar tips are put into register. It is interesting that antibodies and lectins recognizing the sexual agglutinins and the wheat germ agglutinin (WGA) binding protein thought to be the agglutinin receptor result in the elevation of intracellular c A M P and the redistribution of these m e m b r a n e glycoproteins [51, 58]. Recent observations [64, 81] suggest the involvement of calcium and IP 3 in the Chlamydomonas sexual signaling pathway. In summary, it appears that the flagellar m o t o r that drives dynamic events occurring at the surface of the vegetative Chlamydomonas reinhardtii flagellum is regulated by a signaling pathway that recognizes flagellar contact with an external substrate through the crosslinking of flagellar membrane glycoproteins. Flagellar glycoprotein cross-linking appears to induce an influx of calcium that regulates the level of phosphorylation of proteins located within the membrane-matrix compartment of the flagellum. The most dramatic change in protein phosphorylation that is stimulated by antibody crosslinking of the flagellar surface is the total dephosphorylation of a membrane-matrix protein with a molecular mass of approximately 60 kDa and a p l i n the range of 4.8-5.0.

Acknowledgments The author's research described in this paper was supported by research grants from the National Science Foundation (DCB-8905530 and MCB-9206535) and the Jeffress Trust (J-175). Excellent technical assistance was provided by Ms Nancy L Salomonsky. Dr Samuel S Bowser kindly provided figure 2 in addition to much valuable advice during the preparation of this review. Figure 1 is reproduced with the permission of Cell Biol Int Rep, figure 2 is reproduced with permission of the Can J Biochem and Cell Biol, and figure 4 is reproduced with the permission of the Eur J Cell Biol. This paper is based on a talk presented at the 2nd International Cell Motility Symposium held at Sophia Antipolis, France, July 19-23, 1992.

References I Bardele CF (1976) Particle movement in heliozoan axopods associated with lateral displacement of highly ordered membrane domains. Z Naturforsch C: Biochem Biophys Biol l/irol 31, 190-194 2 Barrong SD, Chaltin MH, Fliesler SJ, Possin DE, Jacobson SG, Milam AH (1992) Ultrastructure of connecting cilia in different forms of Retinitis pigmentosa. Arch Ophthalmol 110, 706-710 3 Bergman K, Goodenough UW, Goodenough DA, Jawitz J, Martin H (1975) Gametic differentiation in Chlamydomonas reinhardtii. II. Flagellar membranes and the agglutination reaction. J Cell Biol 67, 606-622 4 Bernhard JM, Bowser SS (1992) Bacterial biofilms as a trophic resource for certain benthic foraminifera. Mar Eco~ Prog Ser 83, 263-272 5 Besharse JC, Horst CJ (1990) The photoreceptor connecting cilium : A model for the transition zone. ln: Ciliary and Flagellar Membranes (Bloodgood RA, ed) Plenum Press, New York, 398-417 6 Bessen M, Fay RB, Witman GB (1980) Calcium control of

300

7 8 9 10 lI 12 13

14

15

16 17 18

19 20 21 22

23

24 25 26 27 28 29

RA Bloodgood waveform in isolated flagellar axonemes of Chlamydomonas. J Cell Biol 86, 446-455 Bloch RJ (1983) Acetylcholine receptor clustering in rat myotubes: requirement for Ca 2+ and effects of drugs which depolymerize microtubules. J Neurosci 3, 2670-2680 Bloodgood RA (1977) Rapid motility occurring in association with the Chlamydomonas flagellar membrane. J Cell Biol 75, 983-989 Bloodgood RA (1978) Unidirectional motility occurring in association with the axopodial membrane of Echinosphaerium nucleofilum. Cell Biol Int Rep 2, 171-176 Bloodgood RA (1980) Direct visualization of dynamic membrane events in cilia. J Exp Zool 213, 293-295 Bloodgood RA (1981) Flagella-dependent gliding motility in Chlamydomonas. Protoplasma 106, 183-192 Bloodgood RA (1984) Protein turnover in the Chlamydomonas flagellum. Exp Cell Res 150, 488-493 Bloodgood RA (1987) Glycoprotein dynamics in the Chlamydomonas flagellar membrane. In: Advances in Cell Biology, Vol 1 (Miller, KR, ed) JAI Press, Greenwich, 97-130 Bloodgood RA (1988) The use Of microspheres in the study of cell motility. In: Microspheres: Medical and Biological Applications (Rembaum A, Tokes Z, eds) CRC Press, Boca Raton, 165-192 Bloodgood' RA (1990) Gliding motility and flagellar glycoprotein dynamics in Chlamydomonas. In: Ciliary and Flagellar Membranes (Bloodgood RA, ed) Plenum Press, New York, 91-128 Bloodgood RA (1992) Calcium-regulated phosphorylation of proteins in the membrane-matrix compartment of the Chlamydomonas flagellum. Exp Cell Res 198, 228-236 Bloodgood RA, Workman LJ (1984) A flagellar surface glycoprotein mediating cell-substrate interaction in Chlamydomonas. Cell MotU 4, 77-87 Bloodgood RA, Salomonsky NL (1989) Use of a novel Chlamydomonas mutant to demonstrate that flagellar glycoprotein movements are necessary for the expression of gliding motility. Cell Motil Cytoskeleton 13, 1-8 Bloodgood RA, Salomonsky NL (1990) Calcium influx regulates antibody-induced glycoprotein movements within the Chlamydomonas flagellar membrane. J Cell Sci 96, 27-33 Bloodgood RA, Salomonsky NL (1991) Regulation of flagellar glycoprotein movements by protein phosphorylation. Eur J Cell Biol 54, 85-89 Bloodgood RA, Leffler EM, Bojczuk AT (1979) Reversible inhibition of Chlamydomonas flagellar surface motility. J Cell Biol 82, 664-674 Bloodgood RA, Woodward MP, Salomonsky NL (1986) Redistribution and shedding of flagellar membrane •glycoproteins visualized using an anti-carbohydrate monoclonal antibody and concanavalin A. J Cell Biol 102, 1797-1812 Bloodgood RA, Salomonsky NL, Reinhart FD (1987) Use of carbohydrate probes in conjunction with fluorescenceactivated cell sorting to select mutant cell fines of Chlamydomonas with defects in cell surface glycoproteins. Exp Cell Res 173, 572-585 Bourguignon LYW, Bourguignon GJ (1984) Capping and the cytoskeleton. Int Rev Cytol 87, 195-224 Bowser SS, Bloodgood RA (1984) Evidence against surfriding as a general mechanism for surface motility. Cell Motil 4, 305-314 Bowser SS, DeLaca TE (1985) Rapid intracellular motility and dynamic membrane events in an Antarctic foraminifer. Cell Biol Int Rep 9, 901-910 Bowser SS, Rieder CL (1985) Evidence that cell surface motility in Allogromia is mediated by cytoplasmic microtubules. Can J Biochem Cell Biol 63,608-620 Bowser SS, McGee-Russell SM, Rieder C (1983) EHNA inhibits particle movement in and on the reticulopods of AIIogromia sp Strain NF (Lee). JProtozoo130, 12A (Abstract) Bowser SS, McGee-Russell SM, Rieder CL (1984) Multiple fission in Allogromia sp, Strain NF (Foraminiferida):

30

31

32 33 34 35 36 37 38 39

40 41

42 43 44 45 46

47 48 49 50

51

Release, dispersal and ultrastructure of offspring. J Protozool 31, 272-275 Bowser SS, Israel HA, McGee-Russell SM, Rieder CL (1984) Surface transport properties of reticulopodia: do intracellular and extracellular motility share a common mechanism? Cell Biol Int Rep 8, 1051-1063 Bowser SS, McGee-Russell SM, Rieder CL (1985) Digestion of prey in Foraminifera is not anomalous: a correlation of light microscopic, cytochemical, and HVEM technics to study phagotrophy in two Allogromids. Tissue Cell 17, 823-839 Brady ST (1991) Molecular motors in the nervous system. Neuron 7, 521-533 Bretscher MS (1976) Directed lipid flow in cell membranes. Nature 260, 21-23 Cachon J, Cachon MCE (1964) Les mouvements de cyclose darts les axopodes d'Acanthaires. Leur r61e lots de la nutrition et de la locomotion. Bull Inst Oceanogr Monaco 61, 1-8 Cande WZ, Wolniak SM (1978) Chromosome movement in lysed mitotic cells is inhibited by vanadate. J Cell Biol 79, 573-580 Chailley B, Bork K, Gounon P, Sandoz D (1986) Immunological detection of actin in isolated cilia from quail oviduct. Biol Cell 58, 43-52 Connolly JA (1984) Role of the cytoskeleton in the formation, stabilization, and removal of acetylcholine receptor clusters in cultured muscle cells. J Cell Biol 99, 148-154 Detmers PA, Condeelis JS (1986) Trifluoperazine and W-7 inhibit mating in Chlamydomonas at an early stage of gametic interaction. Exp Cell Res 163, 317-326 Edds KT (1975) Motility in Echinosphaerium nucleofilum. I. An analysis of particle movements in the axopodia and a direct test of the involvement of the axoneme. J Cell Biol 66, 145-155 Forscher P, Lin CH, Thompson C (1992) Novel form of growth cone motility involving site-directed actin assembly. Nature 357, 515-518 Gehm BD, Pinke RM, Laquerre S, Chafouleas JG, Schultz DA, Pepper D J, McConnell DG (1991) Activation of bovine rod outer segment phosphatidylinositol-4,5-bisphosphate phospholipase C by calmodulin antagonists does not depend on calmodulin. Biochemistry 30, 11302-11306 Gieger B (1983) Membrane-cytoskeleton interactions. Biochim Biophys Acta 737, 305-341 Goodenough UW (1991) Chlamydomonas mating interactions. In: Microbial Cell-Cell Interactions (Dworkin M, ed) Am Soc Microbiol, Washington, 71-112 Hall DH, Hedgecock EM (1991) Kinesin-related gene unc-104 is required for axonal transport of synaptic vesicles in C elegans. Cell 65, 837-847 Halpain S, Greengard P (1990) Activation of NMDA receptors induces rapid dephosphorylation of the cytoskeletal protein MAP2. Neuron 5, 237-246 Halpain S, Greengard P (1992) Protein dephosphorylation as a mediator of NMDA receptor signal transduction. In: Excitatory Amino Acids and Second Messenger Systems (Teichberg VI, Turski L, eds) Springer-Verlag, Berlin, 121-142 Harris AK (1973) Cell surface movements related to cell locomotion. Ciba Found Syrup 14, 3-20 Hewitt JA (1979) Surf-riding model for cell capping. J Theor Biol 80, 115-127 Hoffman JL, Goodenough UW (1980) Experimental dissection of flagellar surface motility in Chlamydomonas. J Cell Biol 86, 656-665 Homan WL, Sigon C, van den Briel W, Wagter R, de Nobel H, Mesland DAM, Musgrave A, van den Ende H (1987) Transport of membrane receptors and the mechanics of sexual cell fusion in Chlamydomonas eugametos. FEBS Lett 215, 323-326 Homan WL, Musgrave A, de Nobel H, Wagter R, de Wit D, Kolk A, van den Ende H (1988) Monoclonal antibodies directed against the sexual binding site of Chlam.vdomonas eugametos gametes. J Cell Biol 107, 177-189

Directed movements of ciliary and flagellar membrane components 52 Hunter DG, Fishman GA, Mehta RS, Kretzer FL (1986) Abnormal sperm and photoreceptor axonemes in Usher's Syndrome. Arch Ophthalmol 104, 385-389 53 Jahn TL, Rinaldi RA (1959) Protoplasmic movement in the foraminiferan, Allogromia laticollaris; and a theory of its mechanism. Biol Bull (Woods Hole) 117, 100-118 54 Kanno F, Ishii K (1979) Movement of the surface layer on the axopodium and its protoplasm in Actinosphaeriurn. Bull Fac Lib Arts Hosei Univ 31, 1-8 55 Kawachi M, Inouye I, Maeda O, Chihara M (1991) The haptonema as a food-capturing device: observations on Chrysochromulina hirta (Prymnesiophyceae). Phycologia 30, 563-573 56 Kitching JA (1960) Responses of the Heliozoan Actinophrys sol to prey, to mechanical stimulation, and to solutions of proteins and certain other substances. J Exp Biol 37, 407-416 57 Klee CB, Cohen P (1988) The calmodulin-regulated protein phosphatase. In: Mol Aspects Cell Reg Vol 5 (Cohen P, Klee CB, eds) Elsevier, Amsterdam, 225-248 58 Kooijman R, de Wildt P, Beumer S, van den Vliet G, Homan W, Kalshoven H, Musgrave A, van den Ende H (1989) Wheat germ agglutinin induces mating reactions in Chlamydomonas eugametos by cross-linking agglutininassociated glycoproteins in the flagellar membrane. J Cell Biol 109, 1677-1687 59 Kozminski KG, Johnson KA, Forscher P, Rosenbaum JL (1992) A new motility associated with the eukaryotic flagellum. Mol Biol Cell 3, 51a 60 Leadbeater BSC (1971) Observations by means of cine photography on the behaviour of the haptonema in plankton flagellates of the class Haptophyceae. J Marine Biol Assoc UK 51, 207-217 61 Leidy J (1879) Fresh-Water Rhizopods o f North America. US Geological Survey of the Territories, Washington, p 238 62 Lewin RA (1952) Studies of the flagella of algae, I. General observations of Chlamydomonas moewusii Gerloff. Biol Bull (Woods Hole) 103, 74-79 63 Lewin RA (1982) A new kind of motility mutant (nongliding) in Chlamydomonas. Experientia 38, 348-349 64 Musgrave A, Kuin H, Jorgen M, de Wilt P, Schuring F, Klerk H, van den Ende H (1992) Ethanol stimulates phospholipid turnover and inositol 1,4,5-trisphosphate production in Chlamydomonas eugametos gametes. Planta 186, 442-449 65 Pasquale SM, Goodenough UW (1987) Cyclic AMP functions as a primary sexual signal in gametes of Chlamydomonas reinhardtii. J Cell Biol 105, 2279-2293 66 Pasternak C, Spudich JA, Elson EL (1989) Capping of surface receptors and concomitant cortical tension are generated by conventional myosin. Nature 341, 549-551 67 Peng HB, Cheng PC (1982) Formation of postsynaptic specializations induced by latex beads in cultured muscle cells. J Neurosci 2, 1760-1774 68 Peng HB, Phelan KA (1984) Early cytoplasmic specialization at the presumptive acetylcholine receptor cluster: a meshwork of thin filaments. J Cell Biol 99, 344-349 69 Peng HB, Cheng PC, Luther PW (1981) Formation of ACh receptor clusters induced by positively charged latex beads. Nature 292, 831-834 70 Pijst HLA, van Driel R, Janssens PMW, Musgrave A, van den Ende H (1984) Cyclic AMP is involved in sexual reproduction of Chlamydomonas eugametos. FEBS Lett 174, 132-136 71 Piperno G, Luck DJL (1979) An actin-like protein is a component of axonemes from Chlamydomonas ax6nemes. J Biol Chem 254, 2187-2190 72 Preston TM, King CA, Hyams JS (1990) The Cytoskeleton and Cell Motility. Blackie, Glasgow and London, 202 p 73 Reinhart FD, Bloodgood RA (1988) Gliding defective mutant cell strains of Chlamydomonas moewusii exhibit alterations in a 240 kDa surface-exposed flagellar glycoprotein. Protoplasma 144, 110-118 74 Remillard SP, Witman GB (1982) Synthesis, transport, and

75

76 77

78

79

80 81

82

83 84

85 86 87

88 89 90 91 92 93

94

301

utilization of specific flagellar proteins during flagellar regeneration in Chlamydomonas. J Cell Biol 93,615-631 Roth KE, Davison EA, Myers N, Rieder CL, Bowser SS (1987) Correlative light and high-voltage electron microscopy of primary (9 + 0) cilia in cultured kidney epithelial cells. Proc 45th Ann Mtg Electron Micros Soc Am 828-829 Saito T, Small L, Goodenough UW (1991) Adenylate cyclase stimulation and protein phosphorylation induced during Chlamydomonas mating. J Cell Biol 115,434a (Abstract) Sandoz D, Gounon P, Karsenti E, Sauron ME (1982) Immunocytochemical localization of tubulin, actin, and myosin in axonemes of ciliated cells from quail oviduct. Proc Natl Acad Sci USA 79, 3198-3202 Schatzman RC, Wise BC, Kuo JF (1981) Phospholipidsensitive calcium-dependent protein kinase: inhibition by anti-psychotic drugs. Biochem Biophys Res Comraun 98, 669-676 Schatzman RC, Raynor RL, Kuo JF (1983) N-(6-aminohexyl)-5-chloro-l-naphthalenesulfonamide (W-7), a calmodulin antagonist, also inhibits phospholipid-sensitive calcium-dependent protein kinase. Biochim Biophys Acta 755, 144-147 Schreiner GF, Unanue ER (1976) Membrane and cytoplasmic changes in B lymphocytes induced by ligand-surface immunoglobulin interaction. Adv Immunol 24, 37-165 Schuring F, Brederoo J, Musgrave A, van den Ende H (1990) Increase in calcium triggers mating structure activation in Chlamydomonas eugametos. FEMS Microbiol Lett 71, 237 -240 Snell WJ (1976) Mating in Chlarnydoraonas: A system for the study of specific cell adhesion. I. UItrastructural and electrophoretic analysis of flagellar surface components involved in adhesion. J Cell Biol 68, 48-69 Snell W J, Buchanan M, Clausell A (1982) Lidocaine reversibly inhibits fertilization in Chlarnydornonas: a possible role for calcium in sexual signaling. J Cell Biol 94, 607-612 Suzaki T, Shigenaka Y (1982) lntra-axopodial particle movement and axopodial surface motility in Echinosphaeriura akamae. In: Biological Functions of Microtubules and Related Structures (Sakai H, Mohri H, Borisy GG, eds) Academic Press, New York, 91-103 Suzaki T, Shigenaka Y, Watanabe S, Toyohara A (1980) Food capture and ingestion in the large heliozoan, Echinosphaerium nucleofilum. J Cell Sci 42, 61-79 Tanaka T, Ohmura T, Yamakado T, Hidaka H (1982) Two types of calcium-dependent protein phosphorylations modulated by calmodulin antagonists. Mol Pharmaco122, 408-412 Taylor RB, Duffus W, Raft M, de Petris S (1971) Redistribution and pinocytosis of lymphocyte surface immunoglobulin molecules induced by anti-immunoglobulin antibody. Nature New Biol 233, 225-229 Travis JL, Allen RD (1981) Studies on the motility of the Foraminifera. I. UItrastructure of the reticulopodial network of Allogromia laticollaris (Arnold). J Cell Bio190, 211-221 Travis JL, Bowser SS (1986) A new model of reticulopodial motility and shape: evidence for a microtubule-based motor and an actin skeleton. Cell Motil Cytoskeleton 6, 2 - 1 4 Troyer D (1975) Possible involvement of the plasma membrane in saltatory particle movement in heliozoan axopods. Nature 254, 696-698 Watanabe T, Flavin M (1976) Nucleotide-metabolizing enzymes in Chlamydomonas flagella. J Biol Chem 251, 182-192 Weatherbee JA (1981) Membranes and cell movement: interactions of membranes with the proteins of the cytoskeleton. Int Rev Cytol 12, 113-176 Wessels D, Soil DR, Knecht D, Loomis WF, De Lozanne A, Spudich J (1988) Cell motility and chemotaxis in Di~tyostelium amebae lacking myosin heavy chain. Dev Biol 128, 164-177 Witman GB, Carlson K, Berliner J, Rosenbaum JL (1972) Chlamydoraonas flagella. I. Isolation and electrophoretic analysis of microtubules, matrix, membranes and mastigonemes. J Cell Biol 54, 507-539