Reactivation of euglenoid movement and flagellar beating in detergent-extracted cells of Astasia longa: Different mechanisms of force generation are involved

Euglenoid movement or ‘metaboly’ is a unique and characteristic motility of the euglenoid flagellates, and was defined to include all the kinds of cell body movements or shape changes that they exhibit (Pringsheim, 1956; Mikolajczyk & Kuźnicki, 1981). In spite of many attempts to elucidate the mechanism of euglenoid movement, little is known about it and even the site and the structures generating the active force are still open questions (Bovee, 1982). Although the involvement of an actomyosin system (Pringsheim, 1948, 1956; Leedale, 1964, 1966; Leedale, Meeuse & Pringsheim, 1965) or a microtubule-dynein system (Hofmann & Bouck, 1976; Gallo & Schrével, 1982) has been suggested, both of these hypotheses lack supporting evidence. We demonstrated recently that pellicular strips slide relative to each other during euglenoid movement in Euglena fusca, and we proposed a hypothesis that envisages an active sliding mechanism between adjacent pellicular strips (Suzaki & Williamson, 1985).

Techniques that permeabilize cells and allow their motile system to be manipulated by the experimenter have often been useful. Such ‘cell model’ techniques were successfully applied to various euglenoid flagellates by Murray (1981) who used Ca²⁺ to induce rounding of elongated cells treated with a divalent cation ionophore and/or detergent. Cell model experiments are most informative when the cell is highly permeable and lacks its own energy supply and Ca²⁺-sequestering system. The situation with the Astasia cell models is not completely clear since, while not requiring exogenous ATP for rounding, their sensitivity to cyanide and dinitrophenol (Murray, 1981) suggests that ATP generation was still occurring and was needed for rounding up. In such a situation, an intracellular membrane system sequestering Ca²⁺ (Murray, 1981) might also still be functional and capable of maintaining an intracellular free Ca²⁺ concentration that differed from that in the external solution.

We have therefore prepared more thoroughly extracted cell models lacking the organelles involved in ATP generation and Ca²⁺ sequestration. Such cells rounded up at much lower concentrations of free Ca²⁺ than the cells in Murray’s (1981) study and are stimulated by exogenous ATP. In addition, because flagellar beating was reactivated concurrently, we investigated the degree of similarity between the microtubule—dynein system powering the flagellum and the largely uncharacterized mechanism generating the force for cell rounding.

Astasia longa (strain no. 512) was purchased from the Culture Collection of Algae at the University of Texas at Austin, U.S.A., and cultured as previously described (Suzaki & Williamson, 1983). Cells were washed and kept for 2-4 days in 100mM-N-2-hydroxy-ethylpiperazine-N′-2-ethanesulphonic acid (Hepes) at pH7·0.

For each experiment, 10 μl of cell suspension in 100mM-Hepes-KOH (pH7·0) was mixed with 40μl of an extraction solution (ES) consisting of 100 mM-Hepes-KOH, 20 mM-piperazine-N, N ′-bis(2-ethanesulphonic acid) (Pipes), 10mM-ethyleneglycol-bis(β-aminoethyl ether)-N,N,N′,N′- tetraacetic acid (EGTA), 50mM-sucrose, 1 mM-dithiothreitol (DTT), 7·5% (v/v) glycerol, 4% (v/v) Triton X-100, and 4% (v/v) Nonidet P-40 at pH7·0 adjusted with KOH. Extraction of the cells was carried out at room temperature (20 ± 3 deg. C) in a small centrifuge tube. At 2-5 min after the addition of ES, the above mixture was then added to 30μl of a reactivation solution (RS) consisting of 100 mM-Tris· HC1, 20mM-Pipes-KOH, 10mM-EGTA, 50mM-sucrose, 1 mM-DTT, and various concentrations of MgCl₂, CaCl₂ and ATP (pH7·0). Reactivation was carried out at room temperature. The final concentrations were: Hepes, 62·5mM; Pipes, 17·5mM; Tris, 37·5mM; EGTA, 8·75mM; sucrose, 43·75mM; DTT, 0·88mM; glycerol, 3·75%; Triton X-100, 2%; and Nonidet P-40, 2%. (A pH change of <0·l pH unit occurred on mixing ES and RS to achieve a final free Ca²⁺ concentration of 5×10⁻⁷M, sufficient to saturate the response.) The percentage of the total volume occupied by the cell bodies was less than 0·1% of the final mixture. Cells were then fixed by adding about 200 μl of a fixative (1:1 (v/v) mixture of methanol and a commercial 37% formaldehyde solution). Cells were fixed very rapidly without changes in cell shape. After 5 min, the cells were spun down by a gentle centrifugation (about 100 g) for 1 min, and numbers of elongated and rounded cells were counted using a light microscope. Cell shape categories were determined using an egg shape (length: width = 3:2) as a critical standard. About 2000 cells were examined in each experiment.

Light micrographs were taken with a Zeiss photomicroscope equipped with Nomarski differential interference optics. For electron microscopy, the cells were fixed either before or after treatments for reactivation by mixing with an equal volume of a fixative (6% (v/v) glutaraldehyde, 3% (w/v) paraformaldehyde, 2mM-MgSO₄, and 50 mM-potassium phosphate buffer at pH7·0). After 20 min at room temperature, the cells were centrifuged, rinsed with 25 mM-phosphate buffer, and then post-fixed with 1% (w/v) OsO₄ in 25 mM-phosphate buffer (pH7·0) for 2h at room temperature. The fixed cells were rinsed, dehydrated in a graded acetone series, and embedded in Spurr’s resin (Spurr, 1969). Thin sections were stained with 10% (w/v) uranyl acetate in ethanol for 10 min and lead citrate (Reynolds, 1963) for 5 min, and examined in a Hitachi H600 electron microscope.

Hepes, Pipes, Tris, EGTA and cytochalasin B were purchased from Sigma Chemical Co. (St Louis, U.S.A.). Sodium metavanadate (NaVO₃) was obtained from BDH Chemicals Ltd (Poole, U.K.), and recrystallized from methanol/water (Gibbons et al. 1978) before use. Apparent association constants for EGTA to calcium and magnesium were calculated from the absolute association constants (Martell & Smith, 1974). The values obtained at pH7·0 (2·66×10⁶ for calcium and 29·5 for magnesium) are in accordance with values used by many other investigators (Marban, Rink, Tsien & Tβien, 1980; Murray, 1981; Blinks, Wier, Hess & Prendergast, 1982; Keller, Jemiolo, Burgess & Rebhun, 1982). Concentrations of free calcium and magnesium ions were then calculated by an iterative procedure (Walz, 1982) using a computer.

Within 2 min of treatment with ES, all cells stopped flagellar movement and became completely elongated (Fig. 1). The elongated cells showed no responses to blue light or mechanical stimulation, which usually cause rounding-up movement of living Astasia cells (Suzaki & Williamson, 1983). Almost 100% of the cells rounded up (Fig. 2) following the subsequent addition of RS containing 5mM-CaCl₂ and 5mM-ATP (final concentrations were: l× 10⁻⁷M free Ca²⁺, 2mM-ATP). Figs 3—5 show detergent-treated cell models before (Fig. 3) and after addition of RS (Fig. 4, partially rounded cell during movement; Fig. 5, completely rounded one). Flagella did not detach from the cell body (arrows). It took about 5s to complete the cell shape change after the cell started rounding-up (Figs 3—5), and all parts of the cell increased in diameter simultaneously (Fig. 4). The cells completely rounded up into ball shapes and no surface infoldings were observed on the rounded cells. Treatment with ES removed the plasma membrane from the groove region of the cell body surface (compare Figs 6 and 7) and from the flagellum (compare Figs 8 and 9). After the treatment with ES, membranous structures such as endoplasmic reticulum and mitochondria were not observed inside the cell body (Fig. 7). Microtubules, pellicular strips and associated projections (p in Fig. 7) remained intact, and new electron-dense structures were found (arrowhead in Fig. 7), which were associated with both the microtubules and the inner surface of the pellicular strips. Traversing fibres connecting adjacent grooves (Mikolajczyk & Kuźnicki, 1981) could not be observed in either the intact (Fig. 6) or the model cell (Fig. 7). Permeability of the cell models was tested by using dyes of different molecular weight (Table 1). None of the dyes used in this experiment entered the living cells, but the dyes neutral red (M_r 289) and fluorescein (M_· 376) entered the extracted models. Almost half of the extracted cells were stained with eosin Y (M_r 692), while erythrosine B (M_r 880) failed to enter either living cells or cell models.

The kinetics for the reactivation of rounding-up movement are shown in Fig. 10. Treatment with ES induced a transient rounding-up (arrow in Fig. 10), but within 2 min every cell became elongated again and flagellar movement stopped. When RS containing Ca²⁺ and ATP was added at 2 min, almost all of the cells became rounded within 5 min. There was a variation in the time required before individual cells started the movement (10s to 5min), but once movement was initiated, it was completed within a short time (about 5s). Conditions were not found in which the cell shape was reversible. Re-extension did not take place when the rounded cells were washed and resuspended in ES (open squares in Fig. 10). Rounded cells did not re-elongate even when ES contained Mg²⁺ and/or ATP (data not shown). When RS was added later than 2 min, fewer cells changed in shape (open triangles in Fig. 10). Rounding-up movement of the cell body required free calcium (⩾ 10⁻⁷M for maximum effect; Fig. 11). In the absence of ATP, fewer cells were reactivated than with ATP, and the reactivation was enhanced by increasing concentrations of ATP (Fig. 12). ATP itself did not induce rounding-up, but it enhanced the movement in the presence of calcium. Of different divalent cations tested, calcium was the mosteffective in inducing the rounding-up movement, while magnesium was the least effective. Rounding-up movement of the cell body was not inhibited by cytochalasin B (100μM).

Flagellar movement could also be reactivated in the detergent-extracted cell models. Since flagella remained on the cell body, some of the extracted cell models swam in a manner similar to living cells when flagellar movement was reactivated. Reactivation of rounding-up was therefore compared with that of flagellar movement. Addition of Mg²⁺ was not required for the calcium-induced rounding-up movement, and high concentrations of magnesium (⩾ 10⁻² M) completely inhibited the movement (Fig. 13A). On the other hand, flagellar movement required magnesium and ATP (Fig. 13B). Flagellar reactivation took place even at a high magnesium concentration and calcium could not substitute for magnesium. Vanadate, which is an inhibitor of dynein ATPase (Gibbons et al. 1978), inhibited flagellar reactivation at 10⁻⁵ M (Fig. 14). Cell body rounding-up was also inhibited by vanadate, but only at a higher concentration (10⁻³ M).

A cell model has been prepared in which euglenoid cell shape changes and flagellar beating can be reactivated and their force-generation mechanisms compared.

Rounding-up movement of living Euglena cells always starts from either the anterior tip (Mackinnon & Hawes, 1961; Suzaki & Williamson, 1983) or the posterior tip (Kamiya, 1939; Mikolajczyk, 1972; Mikolajczyk & Kuźnicki, 1981; Gallo & Schrével, 1982) and propagates to the whole cell body. This anterior-posterior gradient was abolished in the cell models, and the shape change took place over all parts of the cell surface at the same time. The cell models also showed no photoresponse. This suggests that, while activation may be localized in vivo, the sensory transduction system was probably destroyed in the model cells and the motile systems directly and uniformly stimulated by the added calcium ions. Sub-pellicular endoplasmic reticulum is a Ca²⁺-sequestering structure in A. longa (Murray, 1981) and was found to be broken down in the model cells. This observation supports the earlier hypothesis that the endoplasmic reticulum is involved in the sensory transduction systems (Murray, 1981).

Another difference between cell models and living cells was found in the shape of the rounded cells. Reactivated rounding-up movement was not accompanied by the formation of infoldings on the cell surface of the sort usually observed in living cells (Mikolajczyk & Kuźnicki, 1975; Suzaki, 1984). The absence of surface infoldingsmay indicate that the volume of the cell model was able to increase when the cells rounded up, and suggests that the surface infoldings on living cells were probably passively formed to keep the cell volume constant. The cell model’s ability to expand would result from the reduction in resistance to water inflow caused by detergent solubilization of the plasma membrane.

In a previous paper, the generation of force within the pellicle to cause active sliding between adjacent pellicular strips has been suggested as the basis for cell shape changes (Suzaki & Williamson, 1985). However, there is still an alternative explanation for the mechanism of euglenoid movement, which has not been ruled out. This idea involves an actomyosin contractile system deeper inside the cytoplasm than the pellicular structures, and has been suggested by Pringsheim (1948), Gojdics (1953), Leedale (1964, 1966, 1971), Leedale et al. (1965), Gallo et al. (1982) and Bassi & Donini (1984). According to this hypothesis, the force-generating system should be similar to that of amoeboid movement, or in other words, “Inside every Euglena, there is an Amoeba trying to get out” (Leedale, 1966).

Other than pellicular microtubules and traversing filaments within the pellicular complex (Mikolajczyk & Kuźnicki, 1981), no filamentous structures have been reported inside the cytoplasm. This, however, could be simply because of the problem of poor fixation for electron microscopy. In the case of amoeba, for instance, both thin and thick filaments were first observed by using a special procedure (pretreatment with Alcian Blue and fixation with unbuffered OsO₄), and these filaments could not be preserved by conventional fixation (Nachmias, 1964, 1968). Microfilaments have been visualized also in permeabilized cell models of amoeba that could undergo cytoplasmic contraction with the addition of magnesium and ATP (Holberton & Preston, 1970). There is a variety of different types of euglenoid movement, and each type of movement involves differently localized sliding between adjacent pellicular strips (Suzaki & Williamson, 1985). Therefore, if euglenoid movement is generated by some kind of motile system deeper in the cytoplasm, the force-generating structures (probably microfilaments) should be tightly connected to the pellicular complex at many parts of the cell surface to enable the pellicular strips to slide in restricted regions.

Actin is clearly present in the euglenoid Distigmaproteus (Gallo et al. 1982). Like some other algal actins (Marano et al. 1982; Williamson, Perkin & Hurley, 1985), Distigma actin migrates just behind mammalian skeletal muscle actins when electrophoresed in gels containing sodium dodecyl sulphate. Therefore, the conclusion of Murray (1981) that Astasia lacks significant quantities of actin is unreliable, since he considered only material co-migrating with rabbit skeletal muscle actin. Fluorescent localization techniques have not yet revealed any specific structural locations for the actin (Gallo et al. 1982; Bassi & Donini, 1984) and its functions are unknown. In the present study no filamentous structures could be observed inside the detergent-extracted cell model where visibility should be particularly favourable, and cytochalasin B did not inhibit rounding-up. In addition, Murray (1984) reported partial success in inducing a vibrating movement of isolated pellicular structures of D. proteus. These observations suggest that the site for force generation is probably located within the pellicular structures themselves, not in the deeper cytoplasm.

This study ruled out another possible mechanism for euglenoid movement, an active sliding between microtubules and plasma membrane (Hofmann & Bouck, 1976; Gallo & Schrével, 1982). As shown in Fig. 7, treatment with ES removed the plasma membrane from the groove region. Since this region is the only part of the cell surface where microtubules are apposed to the plasma membrane, it is clear that direct interaction between microtubules and plasma membrane cannot be involved in the movement.

Murray (1981) demonstrated by using low concentrations of detergent and the calcium ionophore A23187 that maximum rounding-up movement of A. longa was induced when the surrounding medium contained ⩾ 10⁻⁵ M free calcium. The present study showed, however, that a much lower concentration of Ca²⁺ (⩾ 10⁻⁷ M) induced full reactivation of euglenoid movement in the detergent-extracted cell models. The differences may be because the cells in our preparation were more completely permeabilized and lacked an intracellular system for Ca²⁺ sequestration. (See Williamson, 1984, for some further discussion of models of other plant cells whose variations in Ca sensitivity may be related to their preparation methods.) This saturation concentration of free calcium is relatively low compared with those in other calcium-regulated systems (see Martonosi (1983) for animal cells, and Williamson (1984) for plant cells), and is one of the unusual characteristics of this motile system.

It was also demonstrated that ATP enhanced the Ca²⁺-induced response. However, the exact significance of ATP for the rounding-up movement still needs clarification, since calcium itself could induce some response. Some protozoan motile systems do not require ATP and Ca²⁺ alone induces contractile movements. These movements include contraction of spasmonemes in Vorticella and Zoothamnium (Amos, 1971; Weis-Fogh & Amos, 1972; Asai et al. 1978) and contraction of myonemes in Stentor and Spirostomum (Huang & Pitelka, 1973; Huang & Mazia, 1975; Yogosawa-Ohara & Shigenaka, 1985). However, the cell body movement of Euglena does not seem to be similar to any of these Ca²⁺-induced, ATP-independent systems, since euglenoid movement is based on sliding and not pellicular contraction (Suzaki & Williamson, 1985). Furthermore, filamentous structures like myonemes or spasmonemes could not be detected in Euglena cells. The detergent-treated cell models used in the present study were not completely permeabilized; as shown in Table 1, molecules bigger than ∼500M_r seem to have difficulty in penetrating the cell surface. This is probably because the plasma membrane was removed only at the groove region where microtubules and other electron-dense materials lie beneath the plasma membrane and might hinder passage of bigger molecules (see Fig. 7). Judging from the molecular weight, it might be reasonable to speculate that some endogenous ATP (M_r 507) might still remain after treatment with ES for 2 min and could be used in the reactivation process with the added Ca²⁺. ATP enhanced the Ca²⁺-induced reactivation (Fig. 12) and maximum reactivation could not be achieved without addition of ATP (Fig. 11). These enhancing effects of ATP can be explained if ATP is required for the mechanism. The ES used in this study apparently needs improvement, since prolonged treatment with ES reduced the degree of reactivation (Fig. 10). A more long-lasting model system would probably help to clarify further the role of ATP.

Since euglenoid cell shape change requires sliding between pellicular strips and the possible involvement of microtubules has been suggested (Hofmann & Bouck, 1976; Gallo & Schrével, 1982), a microtubule—dynein system of the type seen in flagella could be the force-generating mechanism. Dynein is a microtubule-associated ATPase (>300× 10³ in sub-unit M_r), which generates a force for sliding movement coupled to the hydrolysis of ATP (Johnson, Porter & Shimizu, 1984). The microtubule-dynein system is not restricted to cilia and flagella, but has also been strongly suggested to be present in termite flagellates as a force-generating mechanism in the bending movement of axostyles (Bloodgood & Miller, 1974; Bloodgood, 1975). The possible involvement of a microtubule-dynein system has also been suggested in food ingestion systems in suctorian ciliates (Tucker, 1974) and cytopharyngeal basket-bearing ciliates (Hausmann & Peck, 1978), and in the reelongation system of contractile ciliates such as Stentor (Huang & Mazia, 1975) and Spirostomum (Yogosawa-Ohara, Suzaki & Shigenaka, 1985). In Astasia, periodic projections are associated with the no. 1 microtubule and are located where the sliding between adjacent pellicular strips might take place (Suzaki, 1984, and unpublished data). It is therefore possible that these structures are involved in the force-generating mechanism for the active sliding movement, and the possibility that they are dynein was examined in the present study. Since flagellar movement was also reactivated in the detergent-extracted cell models of Astasia, this could be used as a good internal control for dynein-based motility. Requirements for divalent cations and sensitivity to vanadate were found to be significantly different (Figs 13, 14) for cell body and flagellar motility. These results clearly indicate that the molecular mechanism of euglenoid movement is different from that of flagellar movement and that euglenoid movement does not depend on an ATPase that shares flagellar dynein’s high sensitivity to vanadate.

Thus, euglenoid movement is concluded to be a unique form of protozoan movement that is probably controlled in vivo by Ca²⁺. A crucial problem still waiting to be solved is the mechanism of force-generation for the active sliding between pellicular strips.