Cell Motility and the Cytoskeleton 33:175-182 (1996) Improved Preparation and Swimming Behavior of Triton-Extracted Models of Didinium nasutum Yoshiaki lwadate and Hiroshi Asai Department of Physics, Waseda University, Okubo, Shinjuku-ku, Tokyo, Japan An improved method for the preparation of Triton X-100 extracted models of Didiniim nasutum was established. Didinium models prepared by treatment with a Triton X- 100 solution, containing specific proteolysis inhibitors and dimethyl sulfoxide, maintained an intact shape at 25°C for a longer time than models prepared by treatment with a Triton X-100 solution not containing the proteolysis inhibitors and dimethyl sulfoxide. The improved Didinium models were reactivated so as to swim forward in the presence of Mg2+ and ATP or ADP. They did not swim backward in response to C a 2 + , in contrast to well-known Paramecium models. However, the improved Didiniiim models showed circular swimming and spinning in response to M or higher concentrations of C a 2 + . This observation suggests that the quick directional change as well as the spinning, which are characteristic of live Didinium, are due to an increase in the endoplasmic Ca2+ concentration around the ciliary system. The response of the Didinium ciliary system to Ca2+ may differ from the response of the Paramecium ciliary system to Ca2'. 0 1996 Wiley-Liss, Inc. Key words: ciliary girdles, circular movement, Didinium, Didinium models, spinning, Triton models INTRODUCTION Triton X-100 extracted models of Paramecium caudatum have been reported [Naitoh and Kaneko, 1972, 1973; Nakaoka and Toyotama, 19791. Naitoh and Kaneko prepared Paramecium models in an extracting solution not containing Mg2+. The swimming velocity of the Paramecium models depended on the concentrations of Mg2+ and ATP as well as the pH [Naitoh and Kaneko, 19731. The swimming direction of the models depended on the Ca2+ concentration [Naitoh and Kaneko, 19721. Nakaoka and Toyotama [ 19791 reported an extraction method utilizing a Triton X-100 extracting solution containing Mg2+, suggesting the importance of this ion in the process of preparing of Triton X-100 Paramecium models. The Paramecium models extracted by the method of Nakaoka and Toyotama changed swimming velocity in the presence of Ca2+, and this response was very sensitive to the Ca2+ concentration. Didinium nasutum, a Gymnostomatida ciliate, has two ciliary girdles. It was reported that Didinium swam both forward and backward along a right-handed helical 0 1996 Wiley-Liss, Inc. path [Machemer, 19741. Furthermore, we often observed that Didinium showed a characteristic swimming behavior, quick directional change and spinning, caused by the beating of two ciliary girdles. In the case of live didinia, little study has been devoted t o this characteristic swimming behvior. Moreover, no studies on the characteristic swimming behavior using Didinium models have yet been conducted. We previously found that Didinium models were obtained by treatment with a Triton X-100 extracting solution containing Mg2+ [Iwadate and Asai, 19951, using a preparation method similar to that used for Paramecium models by Nakaoka and Toyotama [ 19791. It was also found that Mg2+ is indispensable for the preparation of Triton X-100 extracted models of Didinium nasutum [Iwadate and Asai, 19951. Received September 12, 1995; accepted November 20, 1995 Address reprint requests to Yoshiaki Iwadate, Asai Laboratory (Room 51-706), Department of Physics, Waseda University, 3-4-1 Okubo, Shinjuku-ku, Tokyo 169, Japan. 176 Iwadate and Asai However, even with the method of Nakaoka and Toyotama [ 19791, most of these Didinium models maintained an intact shape for less than 5 rnin at 25°C. It is thus important to investigate the preparation method used to extract Didinium models. In order to prepare longer lasting reactivated Didinium models, we investigated an improved method for the preparation of Triton X-100 extracted models of Didinium nasutum. The Didinium models, prepared by treatment with a Triton X- 100 extracting solution containing 1 mM p-APMSF, 0.1 mM leupeptin, 0.1 mM pepstatin, and 1% (vh) dimethyl sulfoxide (DMSO), remained intact morphologically at 25°C for a longer time. Didinium models are useful for studying characteristic swimming behavior. Utilizing our improved Didinium models, we were able to confirm characteristic swimming behavior. This report describes an improved preparation method of preparing Didinium models and the swimming behavior of these models. MATERIALS AND METHODS Cell Culture The ciliate used in this report was a clone harvested from a local pond and identified as Didinium nasutum. The stock cultures of the encysted didinia were excysted in a hay infusion. They were then fed Paramecium caudatum, cultured in a bacterized hay infusion at room temperature. Didinia were washed in standard saline solution containing 0.25 mM CaCI,, 7.5 mM KCl, 2 mM Tris-HC1 at pH 7.2 for 30 rnin at 25°C before all experiments. Previous Method of Didinium nasutum Extraction The didinia were gently centrifuged to make a loose pellet. The pellet was suspended in an ice-cold (0-1°C) extracting solution which contained 10 mM KCl, 10 mM MgCI,, 5 mM ethylene glycol bis(P-aminoethyl ether)-N,N,N’,N’-tetraacetic acid (EGTA), 0.008% (v/v) Triton X-100, and 10 mM Tris-HC1, at pH 7.2, for 60 min. The extracted Didinium cells were washed in an ice-cold (0-1°C) solution containing 20 mM KCl, 5 mM MgCI,, 3 mM EGTA, and 10 mM Tris-HC1, at pH 7.2, for 15 min to remove the Triton X-100. The Didinium cells were then further washed in an ice-cold (0-1°C) washing solution of 20 mM KC1 and 10 mM Tris-HC1, at pH 7.2, for 15 rnin to remove the EGTA. The models prepared utilizing this procedure are termed type A. Improved Method of Didinium nasutum Extraction The didinia were gently centrifuged to make a loose pellet. The pellet was suspended in an ice-cold (0-1°C) extracting solution which contained 20 mM KC1, 5 mM MgCI,, 5 mM EGTA, 0.01% (v/v) Triton X-100, 1 mM p-APMSF, 0.1 mM leupeptin, 0.1 mM pepstatin, 1% (v/v) DMSO, and 10 mM Tris-HC1, at pH 7.2, for 45 min. The extracted Didinium cells were washed in an ice-cold (0-1°C) solution containing 20 mM KC1, 3 mM EDTA, and 10 mM Tris-HC1, at pH 7.2, for 15 rnin to remove the Triton X- 100 and Mg2 . The Didinium cells were then washed again in an ice-cold (0-1°C) washing solution containing 20 mM KCl and 10 mM Tris-HC1, at pH 7.2, for 15 rnin to remove the EDTA. The models prepared utilizing this procedure are termed type B. + Comparison of Durability Between Type A and Type B Models Models kept in an ice bath during 15 min were used for the experiment. Twenty microliters of model suspension, containing 50-100 models, was pipetted onto a depression slide. The model suspension on the depression slide was kept at 25°C. Those models which maintained two intact ciliary girdles and showed a clear proboscis outline were defined as morphologically intact. The numbers of such intact models were counted every other minute for 20 min. The percentage of intact models remaining at each time point was calculated. Separate measurements of type A and B models were repeated three times. Swimming Behavior of Type B Models A 20 ~1 type B model suspension, containing 50100 models, was mixed with 20 pl of each reactivation solution at 25°C. The Mg2+ reactivation solutions contained various concentrations of MgCI,, 8 mM ATP, 6 mM EDTA, 20 mM KCI, and 10 mM Tris-HC1, at pH 7.2. The ATP reactivation solution contained various concentrations of ATP, 16 mM MgCl,, 6 mM EGTA, 20 mM KCl, and 10 mM Tris-HC1, at pH 7.2. The ADP reactivation solution contained various concentrations of ADP, 16 mM MgCl,, 6 mM EGTA, 20 mM KC1, and 10 mM Tris-HC1, at pH 7.2. The Ca2+ reactivation solution contained various concentrations of CaCl,, adjusted with 2 mM EGTA, 8 mM ATP, 16 mM MgCI,, 20 mM KCI , and 10 mM Tris-HC1, at pH 7.2. Swimming behavior was observed under dark-field microscopy and recorded on videotapes. The video image was photographed for a pre-determined exposure time. The average swimming velocity was calculated from 10-50 traces of the swimming models obtained with each measurement. RESULTS Difference in Durability Between Type A and Type B Models The time courses of the percentages of intact type A and type B models remaining are shown in Figure 1. Swimming Behavior of Didinium Models 0 ' 0 5 10 15 177 20 Time [min 1 Fig. 1. The time courses of the percentages of type A and B models remaining intact at 25°C. The values are averages of 50-100 models from three independent measurements. Models prepared by extraction without proteolysis inhibitors are termed type A , and those with proteolysis inhibitors are termed type B. 0:type A. 0: type B. It is clear that, by 3 min, more than half of the type A models had collapsed, whereas more than 80% of type B models were still intact. At 20 minutes, most of the type A models were no longer intact, whereas about 30% of type B models were still intact. Figure 2 shows the results typically observed in this experiment. As shown in Figure 2A, after 6 min, the type A model began to collapse around its cell surface. Whereas, as shown in Figure 2B, the type B model maintained an intact shape for more than 12 min despite gradually becoming more spherical and showing contraction of the ciliary girdle near the proboscis. Effect of Mg2+ Concentration on Swimming Behavior Swimming velocity, as a function of the Mg2+ concentration, is shown in Figure 3. The swimming velocity was measured in each Mg2 t reactivation solution, at various Mg2+ concentrations (0-8 mM), in the presence of 4 mM ATP. The swimming velocity of the models increased with increasing Mg2+ concentration, up to 8 mM. It took about 1 min to reach maximum velocity, after the model suspension had been mixed with each of the Mg2+ reactivation solutions. Effect of ATP Concentration on Swimming Behavior Swimming velocity as a function of the ATP concentration is illustrated in Figure 4. The swimming velocity was measured in each ATP reactivation solution, at various ATP concentrations (0-8 mM). The higher the ATP content, the greater the swimming velocity of the models, up to 4 mM. It took about 1 min to reach the maximum velocity after the model suspension had been mixed with each ATP reactivation solution. Reactivation was also obtained with ADP and Mg2+, without ATP, but 2-3 min was required to reach maximal velocity. Effect of Ca2+ Concentration on Swimming Behavior Swimming traces of the models at pCa of 8.0 and 6.0, are shown in Figure 5. Swimming velocities as a function of the Ca2+ concentration are shown in Figure 6. The swimming velocity was measured in each Ca2+ reactivation solution, at various Ca2 concentrations ( 10-9- 10-4.5 MI. At pCa 6.5 and lower Ca2+ concentrations, the reactivated Didinium models swam forward. The models + 178 Iwadate and Asai Fig. 2. Series of photomicrographs of typical type A and B models at 25°C. A: Type A, prepared by extraction without proteolysis inhibitors. B: Type B, prepared by extraction with proteolysis inhibitors. Bar = 20 p,m. I i - - 1 100 i I - 50 C 0 L_-__ 0 1 L L L L I _ _ 2 3 5 4 Mg2' concentrat I on 6 7 L 8 [mMl Fig. 3. Effect of Mg2+ concentration on swimming velocity of type B models. Final concentrations of ATP (4 mM), EGTA (3 mM), KCl(20 mM), and Tris-HCl(l0 mM) were kept constant throughout. Data represent means and standard errors for measurements on 50-100 models. shown in Figure 5A swam along a right-handed helical path, as illustrated in Figure 5C. The spots in Figure 5A are non-reactivated models. As shown in Figure 6, the swimming velocities were essentially constant, regardless of the Ca2+ concentration (10-9-10-6.5 M). At pCa 6.0 and higher Ca2+ concentrations, reactivated models showed circular swimming not along a right-handed helical path, as illustrated in Figure 5D, or spinning, as in Figure 5E. The reactivated models did not swim backward. As shown in Figure 5B, the swimming courses at pCa 6.0 were not straight. The spots in Figure 5B, which are indicative of no migration, are non-reactivated models as well as reactivated spinning models. The discharge of pexicysts and toxicysts in response to Ca2+ was not observable in this experiment, because of the movement of the models in the presence of ATP and Mg2+. This issue is beyond the scope of the present report. dinium in response to Ca2+ has not previously been studied. In this study, owing to an improved preparation of Didinium models, the swimming behavior of Didinium models in response to Ca2+ was observable. We demonstrated circular swimming and spinning of Didinium models in response to M or higher concentrations of Ca2+ (see Fig. 5 ) . We speculate that the circular swimming and the spinning of the Didinium models, respectively, correspond to the quick directional change and the spinning of live Didinium. The characteristic swimming behavior of live Didinium, quick directional change and spinning, may thus be caused by an increase in the endoplasmic Ca2+ concentration to lop6 M or more around the ciliary system. Paramecium extracted models have been reactivated to swim forward at Ca2+ concentrations below lop6 M, and to swim backward at Ca2+ concentrations higher than lop6 M [Naitoh and Kaneko, 1972, 19731. These results indicate that the ciliary beat direction in Paramecium is regulated by the intracellular Ca2 concentration. In this investigation, however, we did not observe backward swimming of Didinium type B models M or higher concentrations of Ca2+. in response to In the case of Paramecium, an increase in the endoplasmic Ca2+ concentration around the ciliary system simply causes ciliary reversal and backward swimming. How+ DISCUSSION In studying the ciliary beat of live didinia in response to Ca2+, each Didinium cell had to be fixed for the purpose of Ca2+ injection [Hara and Asai, 1980; Hara et al., 19851. Thus, the swimming behavior of Di- 180 Iwadate and Asai 0 U U 1 2 3 - 1 5 4 ATP concentrat ion - - L 6 7 [mMl Fig 4 Effect of ATP concentration on swimming velocity of the type B models Final concentrations of MgCI, (8 mM), EGTA (3 mM), KCI (20 mM), and Tns-HC1 (10 mM) were kept constant throughout Data represent means and standard errors for measurements on 50-100 models ever, the beat of the two ciliary girdles of Didinium is suggested that the ciliary beat of Paramecium requires apparently more complicated. The two ciliary girdles, Mg2 as co-factors for ATP-energized reactivation [Naipeculiar to Didinium nasutum, may cause the quick di- toh and Kaneko, 19731. In the present Didinium type B rectional change and the spinning, rather than the back- models as well, there was no reactivation swimming in ward swimming, in response to a M or higher con- solutions containing ATP but no Mg2+. We thus propose centration of Ca2 . that, as with Paramecium, the cyclic ciliary beating of Presumably, the two types of ciliary girdles have Didinium requires Mg2+. In other words, the substrate different roles in the characteristic swimming behavior of of dynein ATPase in cilia of Paramecium and also DiDidinium, quick directional change and spinning. The dinium is Mg2+-ATP rather than free ATP. difference between, and character of these roles of the In this experiment, we observed the reactivation of two ciliary girdles have yet to be clarified. What change ciliary beating of Didinium models in the presence of in the ciliary beat pattern of our Didinium models is ADP and a delay of about 1 min for reactivation with responsible for the circular swimming and what is re- addition of ADP instead of ATP. This delay may responsible for the spinning are questions that require fur- flect the conversion of ADP to ATP by an adenylate ther investigation. kinase. The ATP, ADP, and Mg2+ concentrations all afWe previously found that the preparation of Didifected the swimming velocity. However, when the Ca2+ nium models was possible in Triton X-100 solution conconcentration was lower than lop6 M, no effect on taining Mg2+ [Iwadate and Asai, 19951. Mg2+ was inswimming velocity was observed (see Fig. 5). These dispensable for the preparation of Didinium models. It results suggest that an increase in the endoplasmic Ca2+ was previously impossible to prepare Didinium models concentration around the ciliary system, within physio- in the absence of Mg2+, in contrast to models of Paralogical range, does not affect the beat frequency. mecium and other ciliates. Paramecium has two fibrous In the case of Paramecium models, it was reported networks spanning the entire cell surface, an outer lattice by Naitoh and Kaneko  that the reactivation of and an infraciliary lattice, serving as a cortical cytoskelciliary beating by ATP required Mg2+ or Mn2+. They eton [Garreau de Loubresse et a]. , 19881. Didinium does + + Swimming Behavior of Didinium Models 181 C Fig. 5 . Swimming traces of the models at pCa 8.0 and 6.0. A: Exposure time 2 sec. At pCa 8.0, the Didinium models swam along a right-handed helical path, as illustrated in C. B: Exposure time 4 sec. At pCa 6.0, some of the models showed circular swimming, as in D, and others showed spinning in one spot, as in E. The spots in A are non-reactivated models. The spots in B, which are indicative of no migration, represent non-reactivated models and spinning models. Both circular swimming and spinning, which are characteristic of the Gymnostomatida ciliate Didinium nasurum, were restored by utilizing our improved Didinium models. Bar = 200 pm. not appear to have such a network, making the prepara- Tris inevitably as a pH buffer at pH 7.2 in this experition of Didinium models relatively difficult. In this in- ment. The interesting cell motilities of Didinium, the vestigation, proteolysis inhibitors and DMSO were found effective for preparing the better Didinium models characteristic swimming behavior [Wessenberg and An(see Fig. 1). None of the Didinium models could be tipa, 19701, the recognition of Paramecium as prey prepared when we tried to use piperazine-N,N’ bis-(2- [Harumoto and Miyake, 1992; Miyake and Harumoto, ethanesulphonic acid) (Pipes) (pKa = 6.8 at 20°C) or 19931, and the discharge of extrusomes [Hara and Asai, 3-(N-morpholino)propanesulfonic acid (Mops) (pKa = 1980; Hara et a]., 19851, have been investigated utilizing 7.2 at 20°C) or N-2-hydroxyethylpiperazine-N’-2-live didinia. The application of our improved Didinium ethane-sulphic acid (Hepes) (pKa = 7.6 at 20°C) as a pH models was found to be of great value for investigating buffer instead of Tris (pKa = 8.3 at 20°C). So, we used cell motilities on a molecular basis. 182 Iwadate and Asai 600 - 9 8.5 8 7.5 I 6.5 6 5.5 5 4.5 4 pCa Fig. 6. Effect of Ca2+ concentration on swimming velocity of the type B models. Final concentrations of ATP (4 mM), MgCI, (8 mM), KCI (20 mM), and Tris-HC1 (10 mM) were kept constant throughout. The Ca2+ concentration was adjusted with Ca-EGTA buffer. Data represent means and standard errors for measurements on 50-100 models. REFERENCES Garreau, de Loubresse N., Keryer, G., Viguts, B., and Beisson, J. ( I 988): A contractile cytoskeletal network of Paramecium: The infraciliary lattice. J. Cell Sci. Y0:35 1-364. Hara, R., and Asai, H. (1980): Electrophysiological response of Didinium nasutum to Paramecium capture and mechanical stimulation. Nature 283:869-870. Hara, R.,Asai, H., and Naitoh, Y . (1985): Electrical responses of the carnivorous ciliate Didinium nasutum in relation to discharge of the extrusive organelles. J. Exp. Biol. 119:211-224. Harumoto, T., and Miyake, A. (1992): Possible participation of surface antigens of Paramecium in predator-prey interaction. J. Protozool. 39:47A. Iwadate, Y., and Asai, H. (1995): Success in preparing detergentextracted models of Didinium nasutum and extrusive organelle discharge in response to calcium ions. Cytobios 82:29-37. Machemer, H. (1974): Ciliary activity and metachronism in Protozoa. In Sleigh, M.A. (ed.): “Cilia and Flagella.” London: Academic Press, pp. 199-286. Miyake, A,, and Harumoto, T. (1993): Possible participation of surface antigens in the predator-prey interaction between Didinium and Paramecium. J. Euk. Microbiol. 40:27a. Naitoh, Y., and Kaneko, H. (1972): Reactivated Triton-extracted models of Paramecium: Modification of ciliary movement by calcium ions. Science 176:523-524. Naitoh, Y., and Kaneko, H. (1973): Control of ciliary activities by adenosine-triphosphate and divalent cations in Triton-extracted models of Paramecium caudatum. J. Exp. Biol. 58:657-676. Nakaoka, Y . , and Toyotama, H. (1979): Directional change of ciliary beat effected with Mg2+ in Paramecium. J. Cell Sci. 40:207214. Wessenberg, H., and Antipa, G . (1970): Capture and ingestion of Paramecium by Didinium nasutum. J . Protozool. 17:250-270.