Activation of Sea Urchin Sperm Flagellar Dynein ATPase Activity by Salt-Extracted Axonemes

Transcription

1 /. Blochem. 102, (1987) Activation of Sea Urchin Sperm Flagellar Dynein ATPase Activity by Salt-Extracted Axonemes Etsuo YOKOTA,***' 1 Issei MABUCHI,*'*** and Hidemi SATO** Department of Cell Biology, National Institute for Basic Biology, Okazaki, Aichi 444; **Sugashima Marine Biological Laboratory, Nagoya University, Toba, Mie 517; and '"Department of Biology, College of Arts and Sciences, The University of Tokyo, Meguro-ku, Tokyo 153 Received for publication, March 23, 1987 When 21S dynein ATPase [EC ] from sea urchin sperm flagellar axonemes was mixed with the salt-extracted axonemes, the ATPase activity was much higher than the sum of ATPase activities in the two fractions, as reported previously (Gibbons, I.R. & Fronk, E. (1979) /. Biol. Chem. 254, ). This high ATPase level was for the first time demonstrated to be due to the activation of the 21S dynein ATPase activity by the axonemes. The mode of the activation was studied to get an insight into the mechanism of dynein-microtubule interaction. The salt-extracted axonemes caused a 7- to 8-fold activation of the 21S dynein ATPase activity at an axoneme : dynein weight ratio of about 14 : 1. The activation was maximal at a low ionic strength (no KC1) at ph Under these conditions, 21S dynein rebound to the salt-extracted axonemes. The maximal binding ratio of 21S dynein to the axonemes was the same as that observed in the maximal activation of 21S dynein ATPase. The sliding between the outer doublet microtubules in the trypsintreated 21S dynein-rebound axonemes took place upon the addition of mm ATP in the absence of KC1. During the sliding, the rate of ATP hydrolysis was at the same level as that of the 21S dynein activated by the salt-extracted axonemes. However, it decreased to the level of 21S dynein alone after the sliding. These results suggested that an interaction of the axoneme-rebound 21S dynein with B-subfibers of the adjacent outer doublet microtubules in the axoneme causes the activation of the ATPase activity. The bending waves of cilia and flagella are induced interaction of the dynein arms on each A-subby the sliding movement between the outer doublet fiber of the doublet microtubule with B-subfibcr microtubules of the axonemes (/, 2). The sliding of the adjacent doublet, which is coupled with the movement is proposed to be generated by the hydrolysis of ATP (for reviews, see Refs. 3-5). 1 Present address: Department of Biology, College of Arts and Sciences, The University of Tokyo, Meguro-ku, Tokyo 153. Abbreviations: DTT, dithiothreithol; MES, 2-{A r -morpholino)ethanesulfonic acid; MOPS, 3-(A r -morpholino)propanesulfonic acid; PMSF, phenylmethylsulfonyl fluoride; SDS, sodium dodecyl sulfate. Vol. 102, No. 1,

2 32 E. YOKOTA, I. MABUCHI, and H. SATO Dynein ATPase is extracted from echinoderm sperm flagella with a low ionic strength medium as a particle possessing a sedimentation coefficient of 10-14S (6). However, the functional form of dynein can be extracted with a high-salt medium as a 21-22S particle. This dynein particle is able to rebind to the salt-extracted axonemes as the outer arm on the outer doublet microtubules (7, 8). Its morphology has recently been demonstrated to be similar to that of the arm in situ (9). It can restore both the beat frequency of high-saltextracted Triton-extracted sperm model (10) and the rate of sliding of outer doublet microtubules in high-salt-extracted axonemes (11). It has been described that the ATPase activity of dynein is activated upon rebinding to the axonemes in the absence of KC1 (12, 13). However, the appropriate conditions for this activation have not been investigated. Moreover, in the absence of KC1, Triton X-100-extracted sperm model shows only poor motility (14). Therefore, the relationship among the rebinding, activation, and the recovery of motility is not clear. In the present study, we investigated the conditions for the rebinding of the sea urchin flagellar 21S dynein to the salt-extracted axonemes and activation of its ATPase activity. We found that both the rebinding and activation were maximal at ph in the absence of KCI. Moreover, it was suggested that the activation is caused by the interaction between bound dynein and adjacent outer doublet microtubules. MATERIALS AND METHODS Materials The following biochemicals were purchased: ATP from Yamasa Shoyu Co., Tokyo; phosphoenolpyruvate from Boehringer Mannheim Corp., Tokyo; trypsin (Type IH) from Sigma Chem. Co., St Louis, Mo., U.S.A.; soybean trypsin inhibitor from Worthington Biochemicals, Freehold, N.J., U.S.A.; and pyruvate kinase from Oriental Yeast Co., Ltd., Tokyo. Preparation of 21S Dynein and Salt-Extracted Axonemes Spermatozoa were obtained from sea urchin, Pseudocentrotus depressus, by the injection of 0.5 M KCI into the body cavity. Axonemes were isolated from sperm tails according to the method of Tang et al. (15), which avoids the use of Triton X-100, since this detergent is capable of activation of the latent ATPase activity of dynein (12). Crude 21S dynein and the salt-extracted axonemes were prepared by the method of Gibbons and Fronk (12) with modifications. Axonemes were resuspended in 0.6 M NaCl, 4 mm MgSO 4, 0.2 mm EDTA, 1 mm DTT, 0.1 mm PMSF, and 10 mm MOPS-NaOH (ph 7.2) and extracted for 10 min on ice. The suspension was then centrifuged at 12,000 x g for 10 min. The pellet fraction was called S-axonemes. S-Axonemes were further extracted with the 0.6 M NaCl solution containing 0.1% Triton X-100 and 10 mm Tris- HC1 buffer (ph 8.0) at 10 C for 20 min. After centrifugation, the pellet was washed four times with a solution of 0.1 M NaCl, 4mM MgSO 4, 0.2 mm EDTA, 0.5 mm DTT, 0.1 mm PMSF, and 10 mm MOPS-NaOH (ph 7.2), and resuspended in the same solution. This was called ST-axonemes fraction. PMSF was not included in the solution used for the analysis of ATPase activity during the sliding. The NaCl-extract was further centrifuged at 100,000 x g for 30 min, and 21S dynein was purified from the supernatant by a 5 to 20% (w/v) sucrose density gradient centrifugation at 26,000 rpm (Hitachi 55P ultracentrifuge, RPS 27-2 rotor) at 2 C for 23 h. The sucrose solution was made up in 0.1 M NaCl, 4 mm MgSO 4, 0.2 mm EDTA, 0.5 mm DTT, and 10 mm MOPS-NaOH (ph 7.2). After fractionation, the peak fraction of ATPase activity was used as 21S dynein (Fig. 1A). Measurement of A TPase Activity of21s Dynein Activated by the Salt-Extracted Axonemes Purified 21S dynein fraction was mixed with the saltextracted axonemes and kept standing on ice for 10 min. As controls, the salt-extracted axonemes and the 21S dynein fraction were separately treated in the same manner. A portion of each sample was assayed for ATPase activity at 20 C. The assay medium contained 4 mm MgSO^, 0.5 mm EDTA, 1 mm ATP, and either 20 mm Tris-HQ (ph 8.0) or 20 mm imidazoie-hcl (ph 7.0). The 21S dynein ATPase activity activated by axonemes was calculated by subtracting the ATPase activity of the salt-extracted axonemes alone from that of the mixture of 21S dynein and the saltextracted axonemes. The amount of Pi liberated was determined by the method of Fiske and SubbaRow (16). Co-Sedimentation Analysis A mixture of 21S /. Biochem.

3 33 FLAGELLAR DYNEIN ATPase ACTIVITY B M -A 66.2K4 5 OK-I 31.0K-" Fraction number K-" 14.4K-" F F Fig. 1. Sucrose density gradient centrifugation and SDS-gel electrophoresis of 21S dynein. A: High-salt extract of sea urchin sperm flagellar axonemes was centrifuged on 34 ml of a 5 to 20% (w/v) sucrose density gradient. Proteins sedimented from right to left., ATPase activity measured at ph 8.0 as described under "MATERIALS AND METHODS." O, A1M. B: Fraction 7 of sucrose density gradient centrifugation was electrophoresed on a 3 to 15% (w/v) acrylamide gradient gel (a) or on a 3 to 6% (w/v) acrylamide gradient gel containing 6 M urea (b). M, molecular weight markers (phosphorylase b, bovine serum albumin, ovajbumin, carbonic anhydrase, soybean trypsin inhibitor, lysozyme). Subunit polypeptides of dynein and molecular weight of the markers are indicated. dynein and the salt-extracted axonemes was centrifuged at 12,000 x g. Portions of the supernatant and the pellet were assayed for ATPase activity in the absence or in the presence of 0.1 M K.C1. The second portions were analyzed by gel electrophoresis. The third portion of the pellet was resuspended in the ATPase assay medium containing 4 mm MgSO4, 0.5 mm EDTA, and 20 mm Tris-HCl (ph 8.0). Immediately after the addition of ATP at the final concentration of 2 mm, it was centrifuged as above. Both the supernatant and the pellet were analyzed by gel electrophoresis. Measurement of the ATPase Activity during Sliding The 21S dynein was mixed with the saltextracted axonemes at an axoneme : dynein weight ratio of 1 5 : 1. The final protein concentration of axonemes was adjusted to 2.0 mg/ml. Trypsin was added to one portion of the mixture at a final concentration of 5.0 /*g/ml, and the mixture was kept standing for 4 min on ice. The digestion Vol. 102, No. 1, 1987 was terminated by the addition of soybean trypsin inhibitor at a final concentration of 70 /*g/ml. As a control, soybean trypsin inhibitor was added to a portion of the mixture first and then trypsin was added. Each sample was added to the assay solution containing 0 or M K.CI, 4 mm MgSO4, 0.5 mm EDTA, and 20 mm Tris-HCl (ph 8.0). As a reference for the ATPase activity, the salt-extracted axonemes alone were treated in the same manner. The sliding movement was started by the addition of 50 or 100/tM ATP at 20 C, and stopped by the addition of trichjoroacetic acid at a final concentration of 2.5 % (w/v). The amount of Pi liberated was measured by the method of Anner and Moosmayer (77). In the assay for longer than 1 min, 5 mm KC1, 0.1 mg/ ml pyruvate kinase, and 1 mm phosphoenolpyruvate were further included in the assay solution, and pyruvate liberated was measured according to the method of Raynard et at. (18). The Pi liberated by 21S dynein ATPase was calculated by sub- 92 5K--

4 34 E. YOKOTA, I. MABUCHI, and H. SATO trading the ATPase activity of the salt-extracted axonemes alone from that of the mixture of 21S dynein and the salt-extracted axonemes. For monitoring of the sliding movement, the turbidity of the solutions was measured at 350 nm using a Shimadzu UV-240 spectrophotometer (Shimadzu Seisakusho, Ltd., Kyoto). Determination of Protein Concentration Protein concentration was determined by the method of Lowry et al. {19), or by the biuret method (20), using bovine serum albumin as a standard. Gel Electrophoresis Polyacrylamide gel electrophoresis in the presence of SDS was performed according to the method of Laemmli (21). In order to separate A a and A f heavy chains, 6 M urea was included in both the stacking and separating gels in some cases. Gels were scanned with a dual-wavelength Chromatoscanner CS-910 (Shimadzu Seisakusho, Ltd.). Electron Microscopy Sample solution was applied to a Formvar-carbon-stabilized copper Axoneme/21S dynein (mg/mg) grid, and stained negatively with 2% (w/v) uranyl acetate. Specimens were examined on a JEOL JEM 1200EX(JEOL, Ltd., Tokyo) electron microscope operating at 80 kv. RESULTS Composition of Pseudocentrotus 21S Dynein The sedimentation pattern of protein and ATPase activity on a 5 to 20% (w/v) sucrose density gradient is shown in Fig. 1A. The main peak of protein and that of ATPase activity coincided with each other, corresponding to 21S dynein. SDSpolyacrylamide gel electrophoresis of 21S dynein showed a band of A chains, three bands of intermediate chains 11, 12, and 13, and several light chain bands (Fig. IB). The A band was further separated into two bands, A. and A f chains, on the SDS-urea gel. Activation of 21S Dynein ATPase by the Salt- Extracted A xonemes ATPase activities of the ST S dynein/axoneme (mg/mg) Fig. 2. ATPase activity of the mixture of 21S dynein and the salt-extracted axonemes. ATPase activity was assayed as described under " MATERIALS AND METHODS " at ph8.0. A: The amount of 21S dynein was constant (4.6//g/ml) while that of the STor S-axonemes was varied. Closed symbols show the total ATPase activity of the mixture (, S-axonemes;, ST-axonemes). Open symbols show the ATPase activity of the mixture from which the axonemal ATPase activity was subtracted (Q, S-axonemes; O, ST-axonemes). B: The amount of axonemes was constant (50//g/ml each) and that of dynein was varied. Closed symbols show the total ATPase activity of the mixture (, S-axonemes;, ST-axonemes). Open symbols show the ATPase activity of the mixture from which the 2IS dynein ATPase activity was subtracted (, S-axonemes; O, STaxonemes). J. Biochem.

5 FLAGELLAR DYNEIN ATPase ACTIVITY KCI (M) Fig. 3. Effect of KCI concentration on the ATPase activity of 21S dynein. The ATPase activity was assayed as described under " MATERIALS AND METHODS," except that the KCI concentration was varied. Open symbols show the ATPase activity of 21S dynein alone (O, 20 mm Tris-HCl, ph 8.0; D, 20 mm imidazole-hcl, ph 7.0). Closed symbols show the ATPase activity of 21S dynein in the presence of axonemes (, 20 mm Tris-HCl, ph 8.0;, 20 mm imidazole-hcl, ph 7.0). Concentrations of 21S dynein and axonemes were 10//g/ml and 0.29 mg/ml, respectively. Fig. 4. Effect of ph on the ATPase activity of 21S dynein. The ATPase activity was assayed as described under "MATERIALS AND METHODS," except that the ph was varied. Open symbols show the ATPase activity of 21S dynein alone (O, no KC-1; D, 0.1 M KC1). Closed symbols show the ATPase activity of 21S dynein in the presence of axonemes (#, no KCI;, 0.1 M KCI). Concentrations of 21S dynein and axonemes were 13.0/(g/ml and 0.23 mg/ml, respectively. Buffers (20 mm) used were as follows: MES (ph 6.7), MOPS (ph ), Tris (ph ) Axoneme/21S dynein (mg/mg) Fig. 5. Co-sedimentation of 21S dynein with the saltextracted axonemes at various ratios of axoneme : 21S dynein. A fixed amount of 21S dynein (94//g/ml) was mixed with axonemes at various ratios as indicated. After incubation and subsequent centrifugation, the ATPase activities of the supernatant (open symbols) and of the pellet (closed symbols) were measured as described under " MATERIALS AND METHODS." 0,«:noKCl. D,i:0.1MKCI. A B s P s P C D E S P S P S P Fig. 6. Binding of 2lS dynein to the salt-extracted axonemes. 21S dynein fraction (90//g/ml) was mixed with axonemes (1.4 mg/ml) and kept standing for 10 min on ice. After centrifugation, aliquots of the supernatant (Cs) and the pellet (Cp) were electrophoresed. Another aliquot of the pellet was resuspended in a solution of 4 mm MgSO 4, 0.5 mm EDTA, and 20 mm Tris-HCl (ph8.0), and then ATP (final concentration of 2 mm) was added (E) or not (D). After centrifugation, the supernatant (s) and the pellet (p) were electrophoresed on a 6% (w/v) polyacrylamide gel. A and B are the supernatant (s) and pellet (p) of 21S dynein and salt-extracted axonemes alone, respectively. The pictures show only the high molecular weight region. Vol. 102, No. 1, 1987

6 E. YOKOTA, I. MABUCHI, and H. SATO B Fig. 7. A and B J. Biochem.

7 FLAGELLAR DYNEIN ATPase ACTIVITY 37 Fig. 7. Electron micrographs of 21S dynein-rebound axonemes. 21S dynein-rebound axonemes were treated with trypsin (A). Trypsin-treated axonemes were suspended in a solution containing 0 (B) or 0.15 M KCI (C), 4 mm MgSO 4, 0.5 mm EDTA, and 20 mm Tris-HCl (ph 8.0) at 20 C, and then ATP (final concentration of 100 /im) was added. After 4 s, samples were stained negatively with uranyl acetate as described under " MATERIALS AND METHODS." Final concentrations of 21S dynein and axonemes were 6.3 /<g/ml and 90.0/<g/ml, respectively. The bar represents 2 fim. axonemes and S-axonemes were usually //mol Pi/mg/min (recovery from the total ATPase activity of intact axonemes, about 15%) and /«mol Pi/mg/min (recovery, about 30%) in the absence of KCI, respectively. Upon mixing 21S dynein with each of these fractions, the total activity became much higher than the simple sum of the original ATPase activities. Figure 2A shows the increase in the ATPase activity when the amount of 21S dynein was kept constant while that of the ST- or S-axonemes was varied. There was no difference between ST- and S-axonemes in this regard. When the axonemal ATPase activity was subtracted from the total ATPase activity, the ATPase curves reached plateau levels at an axoneme : 21S dynein weight ratio of about 14 : 1. Preincubation of 21S dynein with axonemes for 10 min on ice was enough to obtain activation of the ATPase activity. Dialysis of the mixture against the recombination medium as reported by Gibbons and Fronk (12) resulted in less activation (data not shown). Figure 2B shows the reciprocal experiment where the amount of the axonemes was constant while that of dynein was varied. Again, the increase in the ATPase activity was basically similar with the two kinds of axonemes. When the 21S dynein ATPase activity was subtracted from the total ATPase activity, the ATPase curves reached plateau levels at a dynein : axoneme weight ratio of around 0.1. These results indicated that the increase is neither due to the interaction between the 21S dynein and the residual ATPase proteins nor due to the activation of the residual ATPase activities by the 21S dynein, but is due to the activation of the 21S dynein ATPase activity by the axonemes. In the following experiments, we investigated the Vol. 102, No. 1, 1987

8 38 E. YOKOTA, 1. MABUCHI, and H. SATO interaction of 21S dynein with ST-axonemes. The extent of activation of 21S dynein ATPase by ST-axonemes was largely dependent on both KG concentration and ph. Figure 3 shows the dependency on the KC1 concentration. With increasing KC1 concentration, the activity of 21S dynein alone increased gradually at ph either 7.0 or 8.0. The activity of 21S dynein activated by ST-axonemes, however, was maximal in the absence of KCI. With increasing KG concentration, it decreased markedly and reached a minimum at 0.2 M KCI, which may be close to the physiological condition. Above 0.2 M KCI, almost no activation was observed. Figure 4 shows the ph dependency of the ATPase activity, either in the absence or presence of 0.1 M KCI. At ph6.7, 21S dynein ATPase was scarcely activated by axonemes, either in the absence or in the presence of 0.1 M KCI. With increasing ph, the activities of both 21S dynein alone and with axonemes increased, and the extent of the activation was also enhanced. Rebinding of 2IS Dynein to the Salt-Extracted Axonemes To confirm the rebinding of 21S dynein to the ST-axonemes under the present experimental conditions, a co-sedimentation experiment was performed (Figs. 5 and 6). Figure 5 shows the ATPase activities in the supernatant and in the pellet, after centrifugation of the mixture of 21S dynein and axonemes. The ATPase activity of the supernatant decreased, whereas that of the pellet increased linearly with increasing concentration of the ST-axonemes both in the absence and in the presence of 0.1 M KCI. At an axoneme : 21S dynein weight ratio of about 14:1 or more, the ATPase activity of the supernatant was almost zero. Conversely, the ATPase activity in the pellet reached maximum at the same ratio. It was about 2.5 times higher in the absence of KCI than in the presence of 0.1 M KCI. Since the activity of 21S dynein alone was not affected by the presence of 0.1 M KCI whereas that with the axonemes was 2.7 times higher in its absence (Fig. 3), it was suggested that 21S dynein ATPase which rebound to axonemes was activated. It is possible that, in the presence of ATP, the rebound 21S dynein dissociates from axonemes again, because dynein possesses two binding sites for microtubules, that is, ATP-dependent and ATP-independent sites (22). This was investigated by gel electrophoresis (Fig. 6). When 21S dynein was mixed with ST-axonemes and centrifuged after standing for 10 min on ice, it precipitated with axonemes (Fig. 6C). Furthermore, when 21S dynein-rebound axonemes were resuspended in the assay medium and ATP was added at the final concentration of 2mM, 21S dynein precipitated with axonemes by centrifugation (Fig. 6E). Thus it was confirmed that the rebound 21S dynein was. not released from axonemes even in the presence of ATP. ATP Hydrolysis during the Sliding Movement When 21S dynein-rebound axonemes are treated with trypsin, sliding between the outer doublet microtubules can be induced by the addition of ATP under physiological conditions (//). We studied ATP hydrolysis during the sliding. Figure 7 shows electron micrographs of negatively stained 21S dynein-rebound axonemes. Four seconds after the addition of ATP to the trypsin-treated axonemes, sliding between outer doublet microtubules was observed both in the absence (Fig. 7B) and in the presence of 0.15 M KCI (Fig. 7Q. Banana-peeling (23) of outer doublet microtubules was rarely observed. Figure 8 shows the time course of ATP hydrolysis by 21S dynein ATPase and that of the turbidity change during the sliding. The rate of ATP hydrolysis by 21S dynein-rebound axonemes, which were not trypsin-treated, was constant during the assay period. It was about 6 times higher than that of 21S dynein ATPase alone in 5 mm KCI. On the other hand, the rate of ATP hydrolysis by 21S dynein-rebound axonemes after the trypsin treatment exhibited two phases. During the sliding, the rate of ATP hydrolysis was fast and the same as that of 21S dynein ATPase activated by axonemes. After that period, it was similar to that of 21S dynein ATPase alone. A similar result was obtained in 0.15 M KCI (data not shown). In this case, however, the difference between the rates of ATP hydrolysis by 21S dyneinrebound axonemes before and after the trypsin treatment was not so large, as shown in Fig. 8. To analyze in detail the rate of ATP hydrolysis during sliding at the initial phase (0 to 25 s), Pi liberation by 21S dynein ATPase was measured at a lower ATP concentration (50 pm) in the absence of KCI (Fig. 9). There was no difference between the rates of ATP hydrolysis by the 21S J. Biochem.

9 FLAGELLAR DYNEIN ATPase ACTIVITY 39 E c o to CO Time (mln) Fig. 8. Time course of ATP hydrolysis by 21S dynein and change in turbidity of 21S dynein-rebound axonemes in 5 mm KC1. The dashed line indicates the change in the turbidity of trypsin-treated 21S dyneinrebound axonemes. Solid lines indicate the time course of Pi liberation by 21S dynein. Open circles, 21S dynein-rebound trypsin-treated axonemes; closed circles, 21S dynein-rebound non-trypsin-treated axonemes; open squares, 21S dynein alone. Fig. 9. Time course of ATP hydrolysis by 21S dynein during sliding in the absence of KC1. The ATP concentration was 50 ftm. For symbols, see the legend of Fig. 8. dynein-rebound axonemes before and after the trypsin treatment, during 6 s after the addition of ATP. Thereafter the rate became lower, being close to but a little higher than that of 21S dynein alone. DISCUSSION The ATPase activity increased when 21S dynein and the salt-extracted axonemes were recombined as reported by Gibbons and Fronk (12). Since some ATPase activity remained in the salt-extracted axonemes, it was possible that 21S dynein ATPase was activated by interaction with the remaining ATPase in the salt-extracted axonemes. However, the elevated ATPase levels were similar to each other when different axonemes were used, namely S- and ST-axonemes, in which the activities of ATPase remaining were different. Therefore, this possibility was excluded. Conversely, it is also possible that the residual ATPase in the axonemes was activated by 21S dynein. If this was the case, the increased ATPase level with S-axonemes would have been about twice that with ST-axonemes, and the weight ratio of 21S dynein : axonemes where maximal activation was observed in the case of S-axonemes would have been twice that in the case of ST-axonemes. Using these axonemes, however, the same level of ATPase activity was observed when the concentration of 21S dynein was increased, and the weight ratio for the STaxonemes was only a little higher than (or similar to) that for S-axonemes. Therefore, it is reasonable to conclude that the increase in the ATPase activity observed is due to the activation of the 21S dynein ATPase activity by the axonemal microtubules. The activation of 21S dynein ATPase by axonemes was markedly dependent on the KC1 concentration. It was maximal at a low KC1 concentration and minimal between 0.1 and 0.2 M KC1. It was also dependent on ph: very low at ph near 7.0 but high at ph 8.0. The extent of the activation was also maximal at ph 8.0. These properties are similar to those of axonemal ATPase activity before extraction (24). The 21S dynein could rebind to the saltextracted axonemes within a short incubation time, that is, 10 min on ice. Since it was not released from the axonemes even in the presence of ATP, its binding site may be the ATP-independent site on the A-subfiber as reported by Gibbons and Fronk (12). Consequently, the 21S dynein rebound to the salt-extracted axonemes might be essentially the same as that in intact axonemes as regards the Vol. 102, No. 1, 1987

10 40 E. YOKOTA, I. MABUCHI, and H. SATO enzymatic properties and the binding state. It is not clear why the ATPase activity of the 21S dynein rebound to the axonemes and that of the intact axonemes were maximal at a low KG concentration, which may not be physiological. It is interesting to note that a similar result has been obtained for the actin-activated ATPase activity of myosin (25). In the absence of KC1 as well as in the presence of 0.15 M KC1, sliding between the outer doublets could be induced in the 21S dyneinrebound axonemes by the addition of ATP when the axonemes were pretreated with trypsin. The fact that the Triton-extracted sperm model is not well reactivated in the absence of KC1 (14) might be explained by assuming that the mechanism which converts the sliding of microtubules into bending of flagella does not work in the absence ofkcl. While the sliding was occurring, the rate of ATP hydrolysis by 21S dynein ATPase was high and the same as that of 21S dynein ATPase activated by the axonemes without trypsin treatment. After the sliding, when axonemes were disintegrated into the individual doublets, the rate of ATP hydrolysis by 21S dynein ATPase was low and close to that of 21S dynein ATPase alone. The rebound 21S dynein was not released from the axonemes by.atp. From these results, it was suggested that the rebinding itself on the A-subfiber is not the cause of the activation of the ATPase activity of dynein but that the interaction of dynein with the outer doublet microtubules, plausibly the adjacent B-subfiber, causes the activation. However, the ATPase activity after the sliding movement was a little higher than that of 21S dynein alone. Since a few axonemes and outer doublet microtubules did not undergo the sliding movement because of the mild trypsin treatment, 21S dynein ATPase in these axonemes would be activated continuously by the outer doublet microtubules. Trypsin treatment under other conditions enhances the activity of 21S dynein ATPase, which results in loss of the latency (26). When 21S dynein ATPase alone was trypsin-treated, its activity showed only slight enhancement (data not shown). This might also be the reason why the ATPase activity after sliding was a little higher than that of the free 21S dynein. It has been reported that calmodulin activates dynein ATPase activity in a Ca t+ -dependent manner (27, 28), and that this protein is present in axonemes (29-31). In the present study, the optimal ph of 21S dynein ATPase activated by axonemes was very different from that of 21S dynein activated by calmodulin (28), and changes in the Ca* + concentration did not affect the degree of activation by axonemes (data not shown). From these results, it is concluded that the activation of 21S dynein ATPase by axonemes was not due to calmodulin. Recently, Kamimura et al. (32) studied the ATP-hydrolyzing activity of sea urchin sperm fiagellar axonemes during sliding movement. In their experiments, motion-dependent hydrolysis of ATP in the trypsin-treated axonemes was observed immediately after the addition of ATP at a temperature lower than 15 C: The rate of the ATP hydrolysis by such axonemes was a little higher than that of the axonemes without trypsin treatment, and higher than that of the trypsin-treated axonemes after complete disintegration. However, we did not observe any difference between the ATP hydrolysis in trypsinized axonemes during sliding and steady-state hydrolysis in intact axonemes. This discrepancy may be caused by the experimental conditions, especially the temperature. The difference between the motion-dependent ATP hydrolysis and the steady-state hydrolysis decreases with increasing temperature (32). We assayed the ATPase activities at a natural sea water temperature (20 Q, at which the difference may be small. There had been no convincing investigation of the activation of dynein ATPase by microtubules until Omoto and Johnson (33) recently reported that Tetrahymena 22S dynein ATPase is activated by repolymerized brain microtubules. Under their experimental conditions, the activation took place only in the presence of a large excess of microtubules or at a low ATP concentration. In contrast, in the study by Gibbons and Fronk (12) and in the present one the maximal activation was obtained at lower weight ratio of axoneme : 21S dynein, such as 14 : 1. This ratio was the same as that of the maximal amount of 21S dynein rebound to the salt-extracted axonemes. It corresponded to the tubulin : 21S dynein weight ratio of 5.6 : 1, since the content of tubulin in the axonemes was estimated to be about 40% from the densitometry of an SDS-gel. From the result /. Biochem.

11 FLAGELLAR DYNEIN ATPase ACTIVITY 41 that 21S dynein rebound to the axonemes was not released from the A-subfiber even in the presence of ATP, the local concentration of tubulin near dynein in the axoneme must be very large and 21S dynein would interact effectively with adjacent B-subfiber microtubules. We thank Drs. S. Hisanaga, S. Kamimura (The University of Tokyo) and H. Hosoya (National Institute for Basic Biology) for valuable discussions. REFERENCES 1. Satir, P. (1968) /. Cell Biol. 39, Summers, K.E. & Gibbons, I.R. (1971) Proc. Natl. Acad. Set. U.S. 68, Summers, K. (1975) Biochim. Biophys. Ada 416, Mabuchi, I. (1978) in Horizons in Biochemistry and Biophysics Vol. 5 (Quagliariello, E., Palmieri, F., & Singer, T.P., eds.) pp. 1-36, Addison-Wesley, Reading, Massachusetts 5. Gibbons, I.R. (1981) /. Cell Biol. 91, 107s-124s 6. Ogawa, K. & Mohri, H. (1972) Biochim. Biophys. Ada 256, Mabuchi, I., Shimizu, T, & Mabuchi, Y. (1976) Arch. Biochem. Biophys. 176, Gibbons, B.H. & Gibbons, I.R. (1976) Biochem. Biophys. Res. Commun. 73, Sale, W.S., Goodenough, U.W., & Heuser, J.E. (1985) J. Cell Biol. 101, Gibbons, B.H. & Gibbons, I.R. (1979) /. Biol. Chem. 254, Yano, Y. & Miki-Noumura, Y. (1981) /. Cell Sci. 48, Gibbons, I.R. & Fronk, E. (1979) /. Biol. Chem. 254, Shimizu, T. (1975) J. Biochem. 78, Gibbons, B.H. & Gibbons, I.R. (1972) /. Cell Biol. 54, Tang, W.Y., Bell, C.K., Sale, W.S., & Gibbons, I.R. (1982) /. Biol. Chem. 257, Fiske, C.H. & SubbaRow, T. (1925) /. Biol. Chem. 66, Anner, B. & Moosmayer, M. (1975) Anal. Biochem. 65, Raynard, A.M., Hass, L.F., Jacobsen, D.D., & Boyer, P.D. (1961) /. Biol. Chem. 236, Lowry, O.H., Rosebrough, N.J., Fan, A.L., & Randall, R.J. (1951) /. Biol. Chem. 193, Gornall, A.G., Bardwill, C.J., & David, M.M. (1949) /. Biol. Chem. 177, Laemmli, U.K. (1970) Nature 111, Takahashi, M. & Tonomura, Y. (1978) /. Biochem. 84, Sale, W.S. & Gibbons, I.R. (1979) /. Cell Biol. 82, Gibbons, I.R. & Fronk, E. (1972) /. Cell Biol. 54, Maruyama, K. & Ishikawa, Y. (1964) /. Biochem. 55, Bell, C.W. & Gibbons, I.R. (1982) /. Biol. Chem. 257, Blum, J.J., Hayes, A., Jamieson, G.A., & Vanaman, T.C. (1980) J. Cell Biol. 87, Hisanaga, S. & Pratt, M.M. (1984) Biochemistry 23, Gitelman, S.E. & Witman, G.B. (1980) /. Cell Biol. 87, Maihle, N.J., Dcdman, J.R., Means, A.R., Chafouleas, J.G., & Satir, B.H. (1981) /. Cell Biol. 89, Ohnishi, K., Suzuki, Y., & Watanabe, Y. (1982) Exp. Cell Res. 137, Kamimura, S., Yano, M., & Shimizu, H. (1985) J. Biochem. 97, Omoto, C.K. & Johnson, K.A. (1986) Biochemistry IS, Vol. 102, No. 1, 1987