Effects of trypsin-digested outer-arm dynein fragments on the velocity of microtubule sliding in elastase-digested flagellar axonemes

Flagellar movement is caused by the coordinated activity of outer and inner dynein arms, which induces sliding between doublet microtubules. In trypsin-treated flagellar axonemes, microtubule sliding induced by ATP is faster in the presence than in the absence of the outer arms. To elucidate the mechanism by which the outer arms regulate microtubule sliding, we studied the effect of trypsin-digested outer-arm fragments on the velocity of microtubule sliding in elastase-treated axonemes of sea urchin sperm flagella. We found that microtubule sliding was significantly slower in elastase-treated axonemes than in trypsin-treated axonemes, and that this difference disappeared after the complete removal of the outer arms. After about 95% of the outer arms were removed, however, the velocity of sliding induced by elastase and ATP increased significantly by adding outer arms that had been treated with trypsin in the presence of ATP. The increase in sliding velocity did not occur in the elastase-treated axonemes from which the outer arms had been completely removed. Among the outer arm fragments obtained by trypsin treatment, a polypeptide of about 350 kDa was found to be possibly involved in the regulation of sliding velocity. These results suggest that the velocity of sliding in the axonemes with only inner arms is similar to that in the axonemes with both inner and outer arms, and that the 350 kDa fragment, probably of the alpha heavy chains, increases the sliding activity of the intact outer and inner arms on the doublet microtubules.     

Structural and functional reconstitution of inner dynein arms in Chlamydomonas flagellar axonemes

The inner row of dynein arms contains three dynein subforms. Each is distinct in composition and location in flagellar axonemes. To begin investigating the specificity of inner dynein arm assembly, we assessed the capability of isolated inner arm dynein subforms to rebind to their appropriate positions on axonemal doublet microtubules by recombining them with either mutant or extracted axonemes missing some or all dyneins. Densitometry of Coomassie blue-stained polyacrylamide gels revealed that for each inner dynein arm subform, binding to axonemes was saturable and stoichiometric. Using structural markers of position and polarity, electron microscopy confirmed that subforms bound to the correct inner arm position. Inner arms did not bind to outer arm or inappropriate inner arm positions despite the availability of sites. These and previous observations implicate specialized tubulin isoforms or nontubulin proteins in designation of specific inner dynein arm binding sites. Further, microtubule sliding velocities were restored to dynein-depleted axonemes upon rebinding of the missing inner arm subtypes as evaluated by an ATP-induced microtubule sliding disintegration assay. Therefore, not only were the inner arm dynein subforms able to identify and bind to the correct location on doublet microtubules but they bound in a functionally active conformation.     

Structural and Functional Characterization of Paramecium Dynein: Initial Studies

ABSTRACT Dynein arms and isolated dynein from Paramecium tetraurelia ciliary axonemes are comparable in structure, direction of force generation, and microtubule translocation ability to other dyneins. In situ arms have dimensions and substructure similar to those of Tetrahymena. Based on spoke arrangement in intact axonemes, arms translocate axonemal microtubules in sliding such that active dynein arms are (-) end directed motors and the doublet to which the body and cape of the arms binds (N) translocates the adjacent doublet (N+1) upward. After salt extraction, based on ATPase activity, paramecium dynein is found as a 22S and a 14S species. the 22S dynein is a three-headed molecule that has unfolded from the in situ dimensions; the 14S dynein is single headed. Both dyneins can be photocleaved by UV light (350 nm) in the presence of Mg^2-, ATP and vanadate; the photocleavage pattern of 22S dynein differs from that seen with Tetrahymena. Both isolated dyneins translocate taxol-stabilized, bovine brain microtubules in vitro. Under standard conditions, 22S dynein, like comparable dyneins from other organisms, translocates at velocities that are about three times faster than 14S dynein.     

Structural and functional characterization of paramecium dynein: initial studies.

Dynein arms and isolated dynein from Paramecium tetraurelia ciliary axonemes are comparable in structure, direction of force generation, and microtubule translocation ability to other dyneins. In situ arms have dimensions and substructure similar to those of Tetrahymena. Based on spoke arrangement in intact axonemes, arms translocate axonemal microtubules in sliding such that active dynein arms are (-) end directed motors and the doublet to which the body and cape of the arms binds (N) translocates the adjacent doublet (N + 1) tipward. After salt extraction, based on ATPase activity, paramecium dynein is found as a 22S and a 14S species. The 22S dynein is a three-headed molecule that has unfolded from the in situ dimensions; the 14S dynein is single headed. Both dyneins can be photocleaved by UV light (350 nm) in the presence of Mg2+, ATP and vanadate; the photocleavage pattern of 22S dynein differs from that seen with Tetrahymena. Both isolated dyneins translocate taxol-stabilized, bovine brain microtubules in vitro. Under standard conditions, 22S dynein, like comparable dyneins from other organisms, translocates at velocities that are about three times faster than 14S dynein.     

EFFECTS OF TRYPSIN DIGESTION ON FLAGELLAR STRUCTURES AND THEIR RELATIONSHIP TO MOTILITY

Flagellar axonemes isolated from sea urchin sperm were digested with trypsin for various time periods. The course of digestion was monitored turbidimetrically and was found to take two different courses depending on the presence or absence of ATP in the digestion mixture. It was found that ATP induced active disintegration of the axonemes after slight digestion. Samples of the digested axonemes were examined with the electron microscope to determine the effects of trypsin digestion on the substructures of the axonemes. The rate at which trypsin sensitized the axonemes to ATP paralleled the rate at which it damaged the radial spokes and the nexin links, while the dynein arms were removed much more slowly. The results suggest that inactive dynein arms form cross bridges between the adjacent doublet tubules in digested axonemes, and that when activated by the addition of ATP, they induce an active shearing force between adjacent doublets. The radial spokes and the nexin links are not directly involved in the production of mechanical force, but they may participate in regulating the sliding between tubules to produce a propagated bending wave.     

A physical model of microtubule sliding in ciliary axonemes.

Ciliary movement is caused by coordinated sliding interactions between the peripheral doublet microtubules of the axoneme. In demembranated organelles treated with trypsin and ATP, this sliding can be visualized during progressive disintegration. In this paper, microtubule sliding behavior resulting from various patterns of dynein arm activity and elastic link breakage is determined using a simplified model of the axoneme. The model consists of a cylindrical array of microtubules joined, initially, by elastic links, with the possibility of dynein arm interaction between microtubules. If no elastic links are broken, sliding can produce stable distortion of the model, which finds application to straight sections of a motile cilium. If some elastic links break, the model predicts a variety of sliding patterns, some of which match, qualitatively, the observed disintegration behavior of real axonemes. Splitting of the axoneme is most likely to occur between two doublets N and N + 1 when either the arms on doublet N + 1 are active and arms on doublet N are inactive or arms on doublet N - 1 are active while arms on doublet N are inactive. The analysis suggests further experimental studies which, in conjunction with the model, will lead to a more detailed understanding of the sliding mechanism, and will allow the mechanical properties of some axonemal components to be evaluated.     

Regulation of Chlamydomonas flagellar dynein by an axonemal protein kinase

Genetic, biochemical, and structural data support a model in which axonemal radial spokes regulate dynein-driven microtubule sliding in Chlamydomonas flagella. However, the molecular mechanism by which dynein activity is regulated is unknown. We describe results from three different in vitro approaches to test the hypothesis that an axonemal protein kinase inhibits dynein in spoke-deficient axonemes from Chlamydomonas flagella. First, the velocity of dynein-driven microtubule sliding in spoke-deficient mutants (pf14, pf17) was increased to wild-type level after treatment with the kinase inhibitors HA-1004 or H-7 or by the specific peptide inhibitors of cAMP-dependent protein kinase (cAPK) PKI(6-22)amide or N alpha-acetyl-PKI(6-22)amide. In particular, the peptide inhibitors of cAPK were very potent, stimulating half-maximal velocity at 12-15 nM. In contrast, kinase inhibitors did not affect microtubule sliding in axonemes from wild- type cells. PKI treatment of axonemes from a double mutant missing both the radial spokes and the outer row of dynein arms (pf14pf28) also increased microtubule sliding to control (pf28) velocity. Second, addition of the type-II regulatory subunit of cAPK (RII) to spoke- deficient axonemes increased microtubule sliding to wild-type velocity. Addition of 10 microM cAMP to spokeless axonemes, reconstituted with RII, reversed the effect of RII. Third, our previous studies revealed that inner dynein arms from the Chlamydomonas mutants pf28 or pf14pf28 could be extracted in high salt buffer and subsequently reconstituted onto extracted axonemes restoring original microtubule sliding activity. Inner arm dyneins isolated from PKI-treated axonemes (mutant strain pf14pf28) generated fast microtubule sliding velocities when reconstituted onto both PKI-treated or control axonemes. In contrast, dynein from control axonemes generated slow microtubule sliding velocities on either PKI-treated or control axonemes. Together, the data indicate that an endogenous axonemal cAPK-type protein kinase inhibits dynein-driven microtubule sliding in spoke-deficient axonemes. The kinase is likely to reside in close association with its substrate(s), and the substrate targets are not exclusively localized to the central pair, radial spokes, dynein regulatory complex, or outer dynein arms. The results are consistent with a model in which the radial spokes regulate dynein activity through suppression of a cAMP- mediated mechanism.     

ATP-dependent structural changes of the outer dynein arm in Tetrahymena cilia: a freeze-etch replica study

With the rapid-freeze, deep-etch replica technique, the structural conformations of outer dynein arms in demembranated cilia from Tetrahymena were analyzed under two different conditions, i.e., in the absence of ATP and in the presence of ATP and vanadate. In the absence of ATP, the lateral view of axonemes was characterized by the egg- shaped outer dynein arms, which showed a slightly baseward tilt with a mean inclination of 11.1 degrees +/- 3.4 degrees SD from the perpendicular to the doublet microtubules. On the other hand, in the presence of 1 mM ATP and 100 microM vanadate, the outer arms were extended and slender and showed an increased baseward tilt with a mean inclination of 31.6 degrees +/- 4.9 degrees SD. In ATP-activated axonemes, these two types of arms coexisted, each type occurring in groups along one row of outer arms. These findings strongly suggest that the interdoublet sliding is caused by dynamic structural changes of dynein arms that follow the hydrolysis of ATP.     

Inner arm dynein ATPase fraction of sea urchin sperm flagella causes active sliding of axonemal outer doublet microtubule.

In order to clarify the role of the inner arms of the axoneme in sperm flagellar movement, we prepared an ATPase fraction (12S) from the outer arm-depleted axonemes of sea urchin sperm flagella. When both arm-depleted axonemes were incubated with the 12S ATPase, they exhibited the sliding disintegration of outer doublet microtubules. Electron microscopy revealed that the ATPase rebound to the original inner arm sites of the axoneme. Therefore, it is quite likely that the 12S ATPase is one of the components of the inner arms. We referred to it as "inner arm dynein".     

High-frequency vibration in flagellar axonemes with amplitudes reflecting the size of tubulin

Flagellar axonemes of sea urchin sperm display high-frequency (approximately 300 Hz) vibration with nanometer-scale amplitudes in the presence of ATP (Kamimura, S., and R. Kamiya. 1989. Nature (Lond.). 340:476-478). The vibration appears to represent normal mechanochemical interaction between dynein and microtubules because the dependence of the frequency on MgATP concentration is similar to that of the axonemal motility, and because it is inhibited by micromolar concentrations of vanadate. In this study a two-dimensional photo-sensor was used to characterize this phenomenon in detail. Several new features were revealed. First, the vibration was found to be due to a back-and-forth movement of the doublet microtubules along the axonemal length. Two beads attached to different parts of the same axoneme vibrated in unison, i.e., synchronized exactly in phase. This suggested that the outer doublet can be regarded as a stiff rod in vibrating axonemes. Second, evidence was obtained that the amplitude of the vibration reflected the number of active dynein arms. Third, under certain conditions, the vibration amplitude took stepwise values of 8 x N + 4 nm (N = 0, 1, 2, 3, or 4), indicating that the amplitude of microtubule sliding was limited by the size of tubulin dimer (8 nm) or monomer (4 nm). To explain this phenomenon, a model is presented based on an assumption that the force production by dynein is turned off when dynein is subjected to tensile force; i.e., dynein is assumed to be equipped with a feedback mechanism necessary for oscillation.     

The role of axonemal components in ciliary motility

1. The axoneme is the detergent-insoluble cytoskeleton of the cilium. 2. All axonemes generate movement by the same fundamental mechanism: microtubule sliding utilizing ATP hydrolysis during a mechanochemical cycling of dynein arms on the axonemal doublets. 3. Structure, fundamental biochemistry and physiology of the axoneme are conserved evolutionarily, but the phenotypes of beating movements and the responses to specific cytoplasmic signals differ greatly from organism to organism. 4. A model of asynchronous dynein arm activity--the switch point hypothesis--has been proposed to account for cyclic beating in the face of unidirectional sliding. The model suggests that the diversity of beat phenotype may be explicable by changes in the timing of switching between active and inactive states of doublet arm activity. Evidence of axonemal splitting in arrested axonemes provides new support for the hypothesis.     

The roles of noncatalytic ATP binding and ADP binding in the regulation of dynein motile activity in flagella

The regulation of dynein activity to produce microtubule sliding in flagella has not been well understood. To gain more insight into the roles of ATP and ADP in the regulation, we examined the effects of fluorescent ATP analogues and fluorescent ADP analogues on the ATPase activity and motile activity of dynein. 21S dynein purified from the outer arms of sea urchin sperm flagella hydrolyzed BODIPY(R) FL ATP (FL-ATP) at 78% of the rate for ATP hydrolysis. FL-ATP at 0.1-1 mM, however, induced neither microtubule translocation on a dynein-coated glass surface nor sliding disintegration of elastase-treated axonemes. Direct observation of single molecules of the fluorescent analogues showed that both the ATP and ADP analogues were stably bound to dynein over several minutes (dissociation rates = 0.0038-0.0082/s). When microtubule translocation on 21S dynein was induced by ATP, the initial increase of the mean velocity was accelerated by preincubation of the dynein with ADP. Similar increase was also induced by the preincubation with the ADP analogues. Even after preincubation with ADP, FL-ATP did not induce sliding disintegration of elastase-treated axonemes. After preincubation with a nonhydrolyzable ATP analogue, AMPPNP (adenosine 5'-(beta:gamma-imido)triphosphate), however, FL-ATP induced sliding disintegration in approximately 10% of the axonemes. These results indicate that both noncatalytic ATP binding and stable ADP binding, in addition to ATP hydrolysis, are involved in the regulation of the chemo-mechanical transduction in axonemal dynein.     

High speed sliding of axonemal microtubules produced by outer arm dynein

To study dynein arm activity at high temporal resolution, axonemal sliding was measured field by field for wild type and dynein arm mutants of Tetrahymena thermophila. For wt SB255 cells, when the rate of data acquisition was 60 fps, about 5x greater than previously published observations, sliding was observed to be discontinuous with very high velocity sliding (average 196 microm/sec) for a few msec (1 or 2 fields) followed by a pause of several fields. The sliding velocities measured were an order of magnitude greater than rates previously measured by video analysis. However, when the data were analyzed at 12 fps for the same axonemes, consistent with previous observations, sliding was linear as the axonemes extended several times their original length with an average velocity of approximately 10 microm/sec. The pauses or stops occurred at approximately 200 and 300% of the initial length, suggesting that dynein arms on one axonemal doublet were initially active to the limit of extension, and then the arms on the next doublet became activated. In contrast, in a mutant where OADs are missing, sliding observed at 60 fps was continuous and slow (5 microm/sec), as opposed to the discontinuous high-velocity sliding of SB255 and of the mutant at the permissive temperature where OADs are present. High-velocity step-wise sliding was also present in axonemes from an inner arm dynein mutant (KO6). These results indicate that the high-speed discontinuous pattern of sliding is produced by the mechanochemical activity of outer arm dynein. The rate of sliding is consistent with a low duty ratio of the outer arm dynein and with the operation of each arm along a doublet once per beat.     

Reactivation of outer-arm-depleted lung axonemes: evidence for functional differences between inner and outer dynein arms in situ.

Demembranated axonemes isolated from newt lung ciliated cells show a complex beat frequency response to varying [MgATP] and temperature [Hard and Cypher, 1992, Cell Motil. Cytoskeleton 21:187-198]. The present study was undertaken to ascertain whether the beat frequency of outer-arm-depleted newt lung axonemes is controlled in a manner similar to that of intact axonemes. Populations of demembranated ciliary axonemes were isolated by Triton X-100 extraction of lungs from the newt, Taricha granulosa. Aliquots of the demembranated axonemes were further treated with solutions containing high salt (0.375 M KC1) and 1.25 mM MgATP. This treatment resulted in the selective removal of outer dynein arms and a concomitant decrease in beat frequency to a stable level, 33-35% of control values. The effects of pH, salt concentration, nucleotides, and temperature on the beat frequency of reactivated outer-arm-depleted axonemes were ascertained and compared with those of intact axonemes. Some reactivation properties, such as nucleotide specificity, the effect of pH on beat frequency and the threshold [MgATP] required for reactivation (approximately 5 microM) were similar to those observed for intact axonemes. Other properties, such as the relationship between beat frequency and varying [MgATP] or salt concentration, differed both qualitatively and quantitatively from those of control axonemes, as did their response to temperature over the range, 5 degrees-32 degrees C. The nature of the results obtained with temperature and MgATP suggests that inner and outer dynein arms are not functionally equivalent in situ.     

Dynein as a microtubule translocator in ciliary motility: current studies of arm structure and activity pattern

The dynein arms of ciliary doublet microtubules cause adjacent axonemal doublets to slide apart with fixed polarity. This suggests that there is a unique mechanochemistry to the dynein arm with unidirectional force generation in all active arms and also that not all arms are active at once during a ciliary beat. Negative stain and thin-section images of arms in axonemes treated with beta, gamma methylene adenosine triphosphate (AMP-PCP) show a consistent subunit construction where the globular head of the arm interacts with subfiber B of doublet N+1. This interpretation differs from that provided by freeze etch and STEM interpretations of in situ arm construction and has implications for the mechanochemical cycle of the arm. A computer model of the arms in relation to other axonemal structures has been constructed to test these interpretations. Attachment of the head of the arm subfiber B is directly demonstrable in splayed axonemes in AMP-PCP. About half of the doublets in an axoneme show such attachments, while half do not. This might imply that about half the doublets in an axoneme are active at any given instant and can be identified as such. This information may be useful in probing questions of how active arms differ biochemically from inactive arms and of how microtubule translocators in general become active.     

Analyses of 22S Dynein Binding to Tetrahymena Axonemes Lacking Outer Dynein Arms

ABSTRACT. Tetrahymena thermophila mutants homozygous for the oad mutation become nonmotile when grown at the restrictive temperature of 39° C. Axonemes isolated from nonmotile oad mutants ( oad 39° C axonemes) lack approximately 90% of their outer dynein arms and are deficient in 22S dynein. Here we report that oad 39° C axonemes contain 40% of the 22S dynein heavy chains that wild-type axonemes contain and that oad axonemes do not undergo ATP-induced microtubule sliding in vitro. Wild-type 22S dynein will bind to the outer arm position in oad axonemes and restore ATP-induced microtubule sliding in those axonemes. Unlike wild-type 22S dynein, oad 22S dynein does not bind to the outer arm position in oad axonemes. These data indicate that the oad mutation affects some component of the outer arm dynein itself rather than the outer arm dynein binding site. These data also indicate that oad axonemes can be used to assay outer dynein arm function.     

Vigorous beating of Chlamydomonas axonemes lacking central pair/radial spoke structures in the presence of salts and organic compounds

Flagella of Chlamydomonas mutants lacking the central pair of microtubules or radial spokes do not beat; however, axonemes isolated from these mutants were found to display vigorous bending movements in the presence of ATP and various salts, sugars, alcohols, and other organic compounds. For example, about 15% of the total axonemes isolated from pf18, a mutant lacking the central pair, displayed beating in the presence of 10 mM MgSO(4) and 0.2 mM ATP at about 22 Hz, while none beat with the same concentration of ATP and < or = 5 mM or > or = 25 mM MgSO(4). The beat frequency and waveform of beating pf18 axonemes were similar to those of wild type axonemes beating under the same conditions. Similarly, 10-50% of the axonemes beat in the presence of 0.5 M sucrose, 2.0 M glycerol, or 1.7 M[10% (v/v)] ethanol. The appearance of motility did not correlate with the change in axonemal ATPase; however, these substances at those concentrations commonly increased the amplitude of nanometer-scale oscillation (hyper-oscillation) in pf18 axonemes, as well as the extent of ATP-induced sliding disintegration of protease-treated axonemes. Axonemes of double mutants lacking both the central pair and various subspecies of inner-arm dynein also beat at increased MgSO(4) concentrations, but axonemes lacking outer-arm dynein in addition to the central pair did not beat. These and other observations suggest that small molecules perturb the regulation of microtubule sliding through some change in water activity or osmotic stress. Axonemes must have an intrinsic ability to beat without the central pair/radial spokes under a variety of non-physiological solution conditions, as long as the outer dynein arms are present. Apparently, the major function of the central pair/radial spoke structures is to restore this activity under physiological conditions.     

A regulatory light chain of ciliary outer arm dynein in Tetrahymena thermophila

Ciliary beat frequency is primarily regulated by outer arm dyneins (22 S dynein). Chilcote and Johnson (Chilcote, T. J., and Johnson, K. A. (1990) J. Biol. Chem. 256, 17257-17266) previously studied isolated Tetrahymena 22 S dynein, identifying a protein p34, which showed cAMP-dependent phosphorylation. Here, we characterize the molecular biochemistry of p34 further, demonstrating that it is the functional ortholog of the 22 S dynein regulatory light chain, p29, in Paramecium. p34, thiophosphorylated in isolated axonemes in the presence of cAMP, co-purified with 22 S dynein and not with inner arm dynein (14 S dynein). Isolated 22 S dynein containing phosphorylated p34 showed approximately 70% increase in in vitro microtubule translocation velocity compared with its unphosphorylated counterpart. Extracted p34 rebound to isolated 22 S dynein from either Tetrahymena or Paramecium but not to 14 S dynein from either ciliate. Binding of radiolabeled p34 to 22 S dynein was competitive with p29. Phosphorylated p34 was not present in axonemes isolated from a mutant lacking outer arms. Two-dimensional gel electrophoresis followed by phosphorimaging revealed at least five phosphorylated p34-related spots, consistent with multiple phosphorylation sites in p34 or perhaps multiple isoforms of p34. These new features suggest that a class of outer arm dynein light chains including p34 regulates microtubule sliding velocity and consequently ciliary beat frequency through phosphorylation.     

Activation of sea urchin sperm flagellar dynein ATPase activity by salt-extracted axonemes

When 21S dynein ATPase [EC 3.6.1.3] 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) J. Biol. Chem. 254, 187-196). 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 KCl) at pH 7.9-8.3. 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 trypsin-treated 21S dynein-rebound axonemes took place upon the addition of 0.05-0.1 mM ATP in the absence of KCl. 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.     

Functional recombination of dynein 1 with demembranated sea urchin sperm partially extracted with KC1.

Demembranated sea urchin sperm were extracted with 0.5 M KCl as described earlier and reactivated in a solution containing 1 mM ATP. Their flagellar beat frequency was approximately 13 Hz, while that of standard reactivated sperm which had not been extracted with KCl was approximately 31 Hz at 23°C. Addition of soluble dynein 1 caused a gradual increase in the flagellar beat frequency to approximately 25 Hz after 10 min at room temperature. This restoration of frequency occurred in the absence or presence of ATP. Examination by electron microscopy showed that, whereas KCl-extracted sperm were lacking the majority of the outer arms on the doublet tubules, they had regained most of their outer arms following incubation with soluble dynein 1.