Cell Motility and the Cytoskeleton 19:275-281 (1991)
Microtubule Sliding in Flagellar Axonemes of Chlamydomonas Mutants Missing Inner- or Outer-Arm Dynein: Velocity Measurements on New Types of Mutants by an Improved Method Eiji Kurimoto and Ritsu Kamiya Department of Molecular Biology, School of Science, Nagoya University, Nagoya, Japan To help understand the functional properties of inner and outer dynein arms in axonemal motility, sliding velocities of outer doublets were measured in disintegrating axonemes of Chlumydomonus mutants lacking either of the arms. Measurements under improved solution conditions yielded significantly higher sliding velocities than those observed in a previous study [Okagaki and Kamiya, 1986, J . Cell Biol. 103:1895-19021. As in the previous study, it was found that the velocities in axonemes of wild type (wt)and a mutant (odul) missing the outer arm differ greatly: 18.5 2 4.1 p d s e c for wt and 4.4 & 2.3 p d s e c for odul at 0.5 mM Mg-ATP. In contrast, axonemes of two types of mutants (idu2 and idu4) that lacked different sets of two inner-arm heavy chains displayed velocities almost identical with the wild-type velocity. Moreover, axonemes of a non-motile double mutant idu2 x idu4 underwent sliding disintegration at a similar high velocity, although less frequently than in axonemes of single mutants. These observations support the hypothesis that the inner and outer dynein arms in disintegrating axonemes drive microtubules at different speeds and it is the faster outer arm that determines the overall speed when both arms are present. The inner arm may be important for the initiation of sliding. The axoneme thus appears to be equipped with two (or more) types of motors with different intrinsic speeds. Key words: sliding velocity, mechanochemistry, dynein mutant, axonemal motility, cilia
INTRODUCTION
Recent biochemical and ultrastructural studies have indicated that the inner and outer dynein arms in cilia and flagella differ significantly in subunit composition and arrangement within the axoneme [Huang et al., 1979; Goodenough and Heuser, 1985; Goodenough et al., 1987; Sale et al., 1989; Luck and Piperno, 1989; Piperno et al., 19901. It is not established, however, whether the two types of arms differ in function. The finding that removal of the outer dynein arm from sea urchin sperm axonemes results in two-fold reduction in reactivated motility has suggested that the two have almost identical functions [Gibbons and Gibbons, 1973; Yano and MikiNoumura, 19811. Studies with Chlumydomonas mutants, 0 1991 Wiley-Liss, Inc.
on the other hand, have suggested that a significant functional difference exists between the two types of arms [Brokaw and Kamiya, 1987; Kamiya et al., 19891; for example, mutants lacking a large part of the inner-arm dynein are non-motile, while those missing the entire outer arms are motile. Hence, functions specific to each type of arm remain to be elucidated. Okagaki and Kamiya [ 19861 attempted to compare
Received November 27, 1990; accepted April 3, 1991 Address reprint requests to Ritsu Kamiya, Department of Molecular Biology, School of Science, Nagoya University, Nagoya 464-01, Japan.
276
Kurimoto and Kamiya
the functional properties of the inner and outer arms by measuring the sliding velocity of microtubules in wildtype (wt) and mutant Chlumydomonus axonemes. Using the method of Summers and Gibbons [ 19711 to induce axoneme disintegration, they found that axonemes of a mutant (oda1; previously called oda38) lacking the outer arm had a maximal sliding rate of as low as 2.0 pm/sec, whereas wt axonemes had a rate of 13.2 pm/sec. They also reported that the axoneme of a non-motile mutant lacking a significant part of the inner arms (pf23) had a difficulty in undergoing sliding disintegration; its axonemes exhibited disintegration only very rarely and at an unexpectedly low rate of 1-2 pm/sec. However, it was not certain whether the difficulty and low rate were general features of axonemes missing the inner arm or were features limited to this particular mutant. Hence studies with other inner-arm mutants have been wanted. We previously reported on isolation and characterization of outer-arm-missing Chlumydomonus mutants (odus) that retained reduced motility [Kamiya and Okamoto, 1985; Kamiya, 19881. More recently, we isolated two types of motile mutants [idaA and idaB types (previously named idu and idb)] that lacked different sets of inner-arm heavy chains [Kamiya et al., 1989, 19911; of the currently identified three inner-arm subspecies 11,12, and I3 [Piperno et al., 19901, idaA type lacks I1 and idaB type 12. In the present study, we measured the microtubule sliding velocity in these new mutants under improved solution conditions. The measurements yielded significantly higher velocities than those measured in the previous study. We found that, whereas the maximal sliding velocity in odal axoneme was about one-fifth of the wild-type velocity as observed in the previous study, the velocities in axonemes of idaA, idaB, and a double mutant between idaA and idaB were almost identical with that of the wt axonemes. These results raise the possibility that the sliding movements caused by the inner and outer arms significantly differ in velocity, and thus pose a puzzling question of how the two arms can work efficiently within a normal axoneme. MATERIALS AND METHODS Strains
Chlumydomonus reinhurdtii 137c (wild type (wt); and -), a mutant missing the entire outer mating type dynein arm (odul), and two types of inner-arm mutants, idu2 (idaA type) and idu4 (idaB type), were used. Isolation and characterization of these strains have been described [Kamiya and Okamoto, 1985; Kamiya, 1988; Kamiya et al., 19911. Culture of Cells and Isolation of Axonemes
+
For isolation of flagella, cells were grown in 600 ml of Tris-acetate-phosphate (TAP) medium [Gorman
and Levine, 19651 with aeration over a 12 h/12 h light/ dark cycle. Double mutants odul X idu2, odul x idu4, and idu2 X idu4 tended to grow only short flagella in the TAP medium. To induce better flagellation, cells of these double mutants were transferred to 200 ml of 10 mM Hepes (pH 7.4) by centrifugation and resuspension when the growth in the TAP medium appeared complete [Whitman, 19861. The cells grew flagella after being kept aerated for 2-4 h, although the flagella were still shorter than normal. Flagella were isolated by the method of Witman et al. [ 19781, using dibucaine-HC1 to detach flagella from the cell blodies. The isolated flagella were suspended in HMDS solution consisting of 10 mM Hepes (pH 7.4), 5 mM MgSO,, 1 mM DTT, and 4% sucrose. Induction of Sliding Disintegration
Sliding disintegration of axonemes was induced by the method of Okagaki and Kamiya [1986] with modifications. The suspension of flagella (conc.: about 0.2 mg/ml) was sonicated in a Kubota 200M sonicator (Kubota Co., Tokyo) at a setting of 20 W for about 60 sec. The fragmented axonemes were collected by centrifugation at 10,OOOg for 25 min and demembranated by suspending the pellet in an ice-chilled HMDEKP solution [30 mM Hepes (pH 7.4), 5 mM MgSO,, 50 mM K-acetate, 1 mM dithiothreitol (DTT), 1 mM EGTA, 0.5% polyethylene glycol (PEG 20,000 mol wt)] containing 0.4% Nonidet P-40. A small volume of the suspension of fragmented axonemes was mixed with 10 vol of a reactivation solution consisting of HMDEKP and an appropriate concentration of ATP. A drop of the sample was applied into a space between a glass slide and a coverslip, of which two opposite sides were sealed with Vaseline. To wash out axonemes that were not attached to the glass surface, this specimen was first perfused with the same reactivation solution by placing a drop of the solution and a filter paper tip by the two unsealed sides of the coverslip. Next, it was perfused with the reactivation solution containing 1.O-2.0 pg/ml nagarse (type VII bacterial protease; Sigma Chemical Co., St. Louis, MO). Sliding disintegration of the axonemes usually took place within a minute after the perfusion was started and was finished within a few minutes. The disintegration process was observed with a dark-field microscope equipped with an Olympus 40 X Apo objective (N.A. l.O), an Olympus DC condenser (N.A. 1.20-1.33), and a light source of a 100 W highpressure mercury arc lamp (Ushio, Tokyo, Japan). The process was recorded on videotape with an SIT camera (model C2400-08: Hamamatsu Photonics Co., Hamamatsu, Japan) and an S-VHS videotape recorder (Panasonic AG3800: Matsushita Electronics Co., Osaka, 1°C. Japan). All observations were carried out at 23
*
Microtubule Sliding in Dynein-Deficient Axonemes
Measurement of Sliding Velocity
Videotaped images of the sliding microtubules were superimposed on a monitor screen of a personal computer. An end of a moving microtubule was tracked with a “mouse cursor” and its coordinate was analyzed as a function of time. The average sliding velocity in one sliding event was determined by using a home-made program. Maximal sliding velocity (Vmax) and apparent Michaelis constant (Km) for Mg-ATP were calculated from the linear regression line of double reciprocal plots of data at different concentrations of Mg-ATP. Estimation of Percentage of Disintegrating Axonemes The percentage of axonemes that underwent sliding disintegration under a given set of conditions was estimated by counting the numbers of disintegrated and intact axonemes about 5 min after the onset of perfusion with ATP and protease. An axoneme fragment apparently disintegrated through sliding was counted as the one that underwent sliding, irrespective of how many microtubules underwent sliding in the particular fragment. More than 200 axonemes were counted for a single measurement. It is important to note that we did not observe the axonemes until the disintegration process was finished. This is because the occurrence of sliding disintegration was sensitive to the illumination of the microscope for unknown reasons; when the disintegration process was continually monitored under the microscope, far fewer axonemes usually underwent the disintegration. RESULTS Microtubule Sliding Rate Measured Under Improved Solution Conditions
Following Okagaki and Kamiya [ 19861, we induced microtubule sliding by perfusing fragmented axoneme specimens with ATP and nagarse, a protease with broad substrate specificity. Disintegrating axonemes were observed with a dark-field microscope and recorded on videotape by means of an SIT camera and a videotape recorder. The sliding, once started, proceeded fairly constantly; as observed in previous studies, the velocity did not depend on the overlap length between microtubules [Takahashi et al., 19821. The previous data on sliding velocity [Okagaki and Kamiya, 19861 were unsatisfactory in that the measured velocity was lower than the velocity calculated from the waveform of beating axonemes. Disagreement was particularly evident with odal axonemes; maximal sliding velocities in disintegrating wt and odal axonemes were 13.2 p d s e c and 2.0 p d s e c , whereas the values calcu-
277
lated from their in vivo waveforms were 15-20 pm/sec and 4.5-7 p d s e c , respectively (calculated from Brokaw and Kamiya [ 19871; see Discussion). In an effort to improve the experimental conditions, we changed the buffer solution from the one with low ionic strength used by Okagaki and Kamiya to the one used routinely for reactivating demembranated cell models [Kamiya and Okamoto, 19861. Thus we used 30 mM Hepes instead of 10 mM Hepes and included 50 mM K-acetate; K-acetate has been shown to have an effect to improve motility in reactivated sperm cell models [Gibbons et al., 19851, and the potassium concentration of 50 mM has been found to be optimal for reactivation of Chlamydomonas cell models [Kagami and Kamiya, 19901. We found that, although the low-ionic-strength buffer [Summers and Gibbons, 19711 seemed to favor the occurrence of sliding disintegration, it resulted in lower sliding velocities, as described below. Figure 1 shows histograms of the measured sliding velocity in mutant axonemes in the presence of 0.5 mM Mg-ATP. The mean sliding velocities (k standard deviations) in wt and odal were 18.5 (k4.1) p d s e c and 4.4 ( k 2 . 3 ) p d s e c , respectively. These rates were about twice as high as those measured in the previous study at 1 mM Mg-ATP: 9.2 (k1.0) p d s e c (wt)and 1.8 (k0.5) p d s e c (odal (Table 2 of Okagaki and Kamiya [1986]). The higher speed appeared to be mostly, if not entirely, due to the solution condition of the perfusion medium; our own measurement under the previous condition yielded a velocity of 13.7 ? 4.1 p d s e c at 0.5 mM Mg-ATP for wt axoneme (number of measurements = 75). The sliding velocity varied with the Mg-ATP concentration in a manner consistent with Michaelis-Menten kinetics. From the double reciprocal plots of sliding velocities at various Mg-ATP concentrations (Fig. 2), the maximal sliding velocity and the apparent Michaelis constant were determined to be 25.6 p d s e c and 177 p M (wt)and 5.1 p d s e c and 65 p M (odal). These maximal sliding rates were also about twice as high as the previous values (see above), but the Km values were similar to those previously obtained [158 pM (wt)and 64 p M (odul)]. The five- to six-fold difference between the maximal sliding rates in wt and oda 1 axonemes observed in the previous study was also observed in this study, despite the two-fold increase in the measured velocity for each kind of axoneme. Sliding Velocities in Inner Arm-Deficient Axonemes
We next measured the sliding rates in two types of mutants, ida2 (idaA type) and idu4 (idaB type), which lack different sets of two inner-arm heavy chains. As described in the Introduction, ida2 lacks I1 and ida4 I2
Kurimoto and Kamiya
wt
20 10
0 20 10 Fig. 2. Mg-ATP dependence of sliding velocities in wt ( 0 ) and odal ( 0 ) axonemes. Double reciprocal plot of the sliding velocities against the Mg-ATP concentration. Maximal sliding velocities (Vmax) and apparent Michaelis constants (Km) for Mg-ATP calculated from the linear regression lines were 25.6 p d s e c and 177 pM for wt,and 5.1 p d s e c and 64.7 pM for odul, respectively.
0
ida 2
20
measured to be 78 p d s e c and 54 Hz (idu2), 102 p d s e c and 62 Hz (idu4), and 155 p d s e c and 63 Hz (wt)[Kamiya et al., 19911. The sliding velocities in ida2 and ida4 axonemes were almost identical with the wild-type velocity; at 0.5 mM ATP, the average velocities measured were 19.0 (24.4) p d s e c for idu2 and 17.7 (k3.8) p d s e c for idu4, both being comparable with 18.5 p d s e c of wt (Fig. 1). The dependence of the sliding velocity on the Mg-ATP concentrations was also similar among these three strains (Fig. 3); reciprocal plots yielded the maximal velocity of 29.6 p d s e c and Km of 214 p M for idu2 and 22.2 p d s e c and 150 p M for idu4.
10
0
ida 4
20 10
0
ida2xida4
1
5
C
10
30
20
Sliding Velocity (,m/s) Fig. 1. Histograms of sliding velocities in axonemes of wt and dynein arm mutants at 0.5 mM Mg-ATP. Mean velocity *standard deviation = 18.5 ? 4.1 p d s e c (wt); 4.4 2.3 p d s e c (odul); 19.0 4.4 p d s e c (idu2); 17.7 ? 3.8 p d s e c (idu4); 17.1 2.9 pmisec (idu2 x idu4). Temperature: 23°C.
*
*
*
inner-arm subspecies. [Kamiya et al., 19911. Both mutants swim at reduced speed; the average swimming velocities and flagellar beat frequencies at 25°C have been
Sliding Disintegration in Non-Motile Double Mutants
We previously reported that the double mutant between idu2 and ida4 is non-motile although each of the single mutants is motile [Kamiya et al., 1989, 19911. We found that the axoneme of the double mutant idu2 X idu4 slid apart when ATP and protease were perfused, although the sliding was observed much less frequently than with axonemes of wt or of single mutants. The sliding velocity at 0.5 mM Mg-ATP was 17.1 (k2.9) p d s e c (Fig. 1). The estimated maximal velocity and Km were 24.8 p d s e c and 167 pM, respectively, i.e., almost identical with the wild-type values (Fig. 3). The axoneme of another non-motile double-mutant odul x idu4 also underwent sliding disintegration on rare occasions. The sliding velocity at 0.5 mM Mg-ATP observed in seven cases was 1.5 (k0.9) p d s e c . This observation suggests that a subset of the inner dynein arms can cause microtubule sliding in an axoneme, albeit very
Microtubule Sliding in Dynein-Deficient Axonemes
0
Mg-ATP
(mM)
Fig. 3. Mg-ATP dependence of sliding velocities in inner-am-deficient mutant axonemes; the maximal sliding velocities (Vmax) and the apparent Michaelis constants (Km) for Mg-ATP from the linear regression lines of double reciprocal plots were 29.6 pmlsec and 214 pM for idu2 (A), 22.2 p d s e c and 150 pM for idu4 (O), and 24.8 pmlsec and 167 pM for a double mutant idu2 X idu4 (a).0 ; data for Wt.
slowly. However, the axoneme of another double mutant odul x ida2 did not display sliding disintegration. As mentioned before, the non-motile axonemes of double mutants idu2 x idu4 underwent sliding disintegration much less frequently than motile axonemes. To evaluate the readiness of sliding disintegration in the mutant axoneme, we measured the percentage of axonemes that underwent sliding at different Mg-ATP concentrations (see Materials and Methods for methods). The results (Fig. 4) indicate that the occurrence of sliding in the ida2 x ida4 axoneme was less frequent than in motile axonemes, and strongly depended on the Mg-ATP concentration; unlike sliding in all other mutants, the sliding in this double mutant was severely inhibited at high MgATP concentrations. DISCUSSION Large Difference in Sliding Velocity Between wt and oda We have measured the microtubule sliding velocity in disintegrating axonemes of wild-type and mutant Chlamydomonas by using an improved method. The use of solution conditions optimized for axoneme reactivation resulted in significantly higher sliding velocities than those obtained in a previous study. The new data indicate that mutant axonemes missing outer-arm dynein have a much lower sliding velocity than wild-type axonemes, but that those missing subspecies of inner-arm dynein (I1 and 12) have velocities almost identical with the wild-type velocity.
6.5
Mg-ATP
279
1.0
(mM)
Fig. 4. Sliding probability at different concentrations of Mg-ATP in axonemes of w6 ( O ) , odul (H), ida2 (A), idu4 (O), and a double mutant idu2 X idu4 ( 0 ) . The ordinate denotes the percentage of the axoneme fragments that apparently underwent sliding disintegration after a 5 min perfusion with protease and a concentration of Mg-ATP specified on the abscissa. An axoneme fragment was judged to have undergone disintegration even if only one outer doublet was extruded from it.
The maximal sliding velocity in wt axonemes , 25.6 pmhec, estimated by extrapolation, is above the range of estimate for the velocity in beating axonemes in vivo; according to Brokaw and Luck [ 19831, the mean sliding velocity in wt axonemes is 340 rad/sec, which corresponds to 15-20 p d s e c if we assume the distance between adjacent outer-doublet microtubules to be 4.5 -6 nm. (A correlated observation of sliding displacement of outer doublets and bend angle in sea urchin sperm axonemes has recently indicated that the functional interdoublet distance is 4.5 nm [Brokaw, 19891.) The maximal velocity in odal axonemes, 5.1 pmhec, is within the range of estimate of the in vivo velocity, which has been determined to be 100-120 rad/sec [Brokaw and Kamiya, 19871, i.e., 4.5-7 pm/sec. The maximal sliding velocities estimated from the measurements in vitro are thus higher than, or similar to, the velocities estimated for axonemes in vivo; this suggests that the in vitro velocities reflect well the properties of inner- and outer-arm dynein in vivo, although we cannot rule out the possibility that the sonication and protease treatment, used to induce effective axoneme disintegration might have impaired function of some dynein arms. The difference in sliding velocity between wt and odul strains is five-fold in vitro whereas it is three-fold in vivo; a similar large difference between the two strains in vitro has been observed by Okagaki and Kamiya [ 19861. Why the velocity difference between the two strains is consistently more pronounced in disintegrating axonemes than in beating axonemes may be due to the difference in nature of the two kinds of dynein arms as
280
Kurimoto and Kamiya
mechano-transducers; the microtubule sliding caused by inner arms may be less sensitive to load than that caused by outer arms (see below, for further discussion). Effects of Inner-Arm Depletion on the Sliding Velocity
The sliding velocities in ida2, ida4, and a double mutant ida2 X ida4 were almost identical with the wt velocity, in contrast with the low velocity in odal. Thus, the absence of the inner-arm subspecies does not appear to affect the sliding velocity of outer doublets. This result differs strikingly from the previous observation that the sliding velocity in a non-motile inner-arm mutant, pf23, was as low as one-sixth of that in wt. However, the data on pj23 axonemes were not reliable, because these axonemes displayed sliding disintegration only very rarely and the data were based on just a few measurements. A similar difficulty in undergoing sliding was observed with idu2 X ida4 axonemes; it appears that the lack of the inner arm suppresses the initiation of sliding. The result that the sliding velocities in inner-arm mutants were almost identical with the wt velocity is rather unexpected since, in beating flagella, the sliding velocities in these slowly swimming mutants must be lower than in wt; in fact, in vivo velocity in flagella of idal (an idaA-type mutant) has been determined to be 21 1 rad/sec, i.e., about two-thirds of the wild-type value [Brokaw and Kamiya, 19871. These results suggest that the outer-doublet sliding in idu2 and ida4 axonemes is more susceptible to the load present in beating axonemes than the sliding in wt axoneme; the absence of inner-arm components appears to decrease the force generated, but not the velocity of sliding in disintegrating axonemes. It would be interesting to measure the sliding force directly in these mutant axonemes, as has been performed with disintegrating axonemes of sea urchin sperm by using flexible glass needles [Kamimura and Takahashi, 1981; Oiwa and Takahashi, 19881. Implications of the Difference in Sliding Velocity Between Inner-Arm Mutants and Outer-Arm Mutants
If we assume that only little load is imposed upon microtubules in disintegrating axonemes, the effects of depletion of inner or outer arms on the sliding velocity as observed above are consistent with the hypothesis that the inner and outer arms drive microtubules at significantly different rates under load-free conditions. In addition, since the lack of the subspecies of the slower arm (the inner arm) neither decelerates nor accelerates the movement, the sliding rate should be mostly determined by the faster arm (the outer arm) when both arms are present within the same axoneme. A possible mechanism for such behavior would consist of inner arms not func-
tioning effectively in normal axonemes under disintegrating conditions; the slow inner arms may be unable to interact with the adjacent doublet microtubule. An alternative interpretation is that the inner arms do function together with outer arms in disintegrating axonemes, but do so with an accelerated cycling rate that matches that of the outer arm. The acceleration may be brought about through some dynamic interaction between the two arms; since the two arms are attached to the same microtubules, their movement must be somehow coupled. It remains, however, a challenging problem to understand how a system containing two (or more) arms with apparently different intrinsic speeds can work efficiently. Role of Inner Arms in Axonemal Motility
The axoneme of idu2xida4 mutant underwent sliding disintegration only very rarely, especially at high Mg-ATP concentrations, although the velocity was almost normal as long as sliding took place. This difficulty in induction of sliding was similar to the one experienced with pj23 axonemes [Okagaki and Kamiya, 19861, which lack the same inner-arm subspecies (I1 and 12) as those missing in ida2 X ida4; we also tried to measure the sliding velocity in this mutant during the course of this study, but the sliding events were too rare for a reliable measurement. These observations suggest that the lack of inner arms has an effect to suppress the initiation of sliding event. Lower concentrations of Mg-ATP have been shown to favor sliding disintegration of ciliary axonemes of Tetrahymena [Warner and Zanetti, 1980; Tanaka and Miki-Noumura, 19881, and in wild-type axonemes of Chlamydomonas [Okagaki and Kamiya, 19861; in these axonemes, the lower the Mg-ATP concentration , the more outer doublets were observed to undergo sliding. The present finding that the sliding disintegration in ida2 X ida4 axonemes has a strong dependence on MgATP concentration, whereas that in odal missing the outer arm does not have such a dependence (Fig. 4 ) , suggests that it is the outer arm that confers the ATP sensitivity on the axoneme. In other words, Chlumydomonas outer-arm dynein may need to be helped by inner-arm subspecies for its function at high Mg-ATP concentrations. The loss of flagellar beating in ida2X ida4 and pf23 mutants may have somehow come from the difficulty in the initiation of sliding of outer doublets. Another, though not independent, possibility is that the motility loss may have been caused by the disruption of the mechanism that organizes microtubule sliding into axonemal bending motion, in which the inner arm may be playing a crucial role. In fact, waveform analyses of beating axonemes revealed that lack of inner-arm subsets results in a reduction of shear amplitude [Brokaw and
Microtubule Sliding in Dynein-Deficient Axonemes
Kamiya, 19871. Inner-arm dynein may thus be important in developing the axonemal bend from straight movement of microtubule sliding. It will be of great interest to ask by physiological studies whether the lack of motility in ida2 X ida4 is due to defects in the formation of bend or in the propagation of bend within the axoneme.
ACKNOWLEDGMENTS
We thank Prof. Sho Asakura for discussion and encouragement throughout the present study. This work has been supported by grants-in-aid from the Ministry of Education, Science and Culture of Japan (01657001, 02239101).
REFERENCES Brokaw, C.J. (1989): Direct measurements of sliding between outer doublet microtubules in swimming sperm flagella. Science 243: 1593-1596. Brokaw, C.J., and Kamiya, R. (1987): Bending patterns of Chlamydomonas flagella. IV. Mutants with defects in inner and outer dynein arms indicate differences in dynein arm function. Cell Motil. Cytoskeleton 8:68-75. Brokaw, C.J., and Luck, D.J.L. (1983): Bending patterns of Chlamydomonas flagella. I. Wild-type bending patterns. Cell Motil. 3: I3 1-150. Gibbons, B.H., and Gibbons, I.R. (1973): The effects of partial extraction of dynein arms on the movement of reactivated seaurchin sperm. J . Cell Sci. 13:337-357. Gibbons, B.H., Tang, W.-J.Y., and Gibbons, I.R. (1985): Organic anions stabilize the reactivated motility of sperm flagella and the latency of dynein 1 ATPase activity. J. Cell Biol. 101: 1281-1287. Goodenough, U.W., and Heuser, J.E. (1985): Substructure of inner dynein arm, radial spokes, and the central pair/projection complex of cilia and flagella. J. Cell Biol. 100:2008-2018. Goodenough, U.W., Gebhart, B., Mermall, V., Mitchell, D., and Heuser, J.E. (1987): High-pressure liquid chromatography fractionation of Chlamydomonas dynein extracts and characterization of inner-arm dynein subunits. J. Mol. Biol. 194: 48 1-494. Gorman, D.S., and Levine, R.P. (1965): Cytochrome F and plastocyanin: their sequence in the photosynthetic electron transport chain of Chlamydomonas reinhardtii. Proc. Natl. Acad. Sci. USA 54:1665-1669. Huang, B., Piperno, G., and Luck, D.J.L. (1979): Paralyzed flagella mutants of Chlamydomonas reinhardtii defective of axonemal doublet microtubule arms. J. Biol. Chem. 254:3091-3099. Kagami, O., and Kamiya, R. (1990): Strikingly low ATPase activities in flagellar axonemes of a Chlamydomonas mutant missing outer dynein arms. Eur. J. Biochem. 187:44-446.
281
Kamimura, S . , and Takahashi, K. (1981): Direct measurement of the force of microtubule sliding in flagella. Nature 293:566-568. Kamiya, R. (1988): Mutations at twelve independent loci result in absence of outer dynein arms in Chlamydomonas reinhardtii. J. Cell Biol. 107:2253-2258. Kamiya, R., Kurimoto, E., and Muto, E. (1991): Two types of Chlamydomonas flagellar mutants missing different components of inner-arm dynein. J. Cell Biol. 112441-447. Kamiya, R., Kurimoto, E . , Sakakibara, H . , and Okagaki, T. (1989): A genetic approach to the function of inner and outer arm dynein. In Warner, F.D., Satir, P., and Gibbons, I.R. (eds.): “Cell Movement, Volume I . ” New York: Alan R. Liss, Inc., pp. 209-218. Kamiya, R., and Okamoto, M. (1985): A mutant of Chlamydomonas reinhardtii that lacks the flagellar outer dynein arm but can swim. J. Cell Sci. 74:181-191. Luck, R . A . , and Piperno, G. (1989): Dynein arm mutants of Chlamydomonas reinhardtii. In Warner, F.D., Satir, P., and Gibbons, I.R. (eds.): “Cell Movement, Volume 1.” New York: Alan R. Liss, Inc., pp. 49-60. Oiwa, K., and Takahashi, K. (1988): The force-velocity relationship for microtubule sliding in demembranated sperm flagella of the sea urchin. Cell Struct. Funct. 13:193-205. Okagaki, T., and Kamiya, R. (1986): Microtubule sliding in mutant Chlamydomonas devoid of outer or inner dynein arms. J. Cell Biol. 103:1895-1 902. Piperno, G., Ramanis, Z., Smith, E.F., and Sale, W.S. (1990): Three distinct inner dynein arms in Chlamydomonus flagella: Molecular composition and location in the axoneme. J . Cell Biol. 110:379-389. Sale, W.S., Fox, L.A., and Milgram, S.L. (1989): Composition and organization of the inner row of dynein arms. In Warner, F.D., Satir, P., and Gibbons, I.R. (eds.): “Cell Movement, Volume 1.” New York: Alan R. Liss, Inc., pp. 89-102. Summers, K.E., and Gibbons, I.R. (1971): Adenosine triphosphateinduced sliding of tubules in trypsin-treated flagella of sea urchin sperm. Proc. Natl. Acad. Sci. USA 68:3092-3096. Takahashi, K., Shingyoji, C., and Kamimura, S. (1982): Microtubule sliding in reactivated flagella. In Amos, W.B., and Duckett, J.G. (eds.): “Symposia of the Society for Experimental Biology XXXV, Prokaryotic and Eukaryotic Flagella.” Cambridge: Cambridge Univ. Press, pp. 159-177. Tanaka, M., and Miki-Nomura, T. (1988): Stepwise sliding disintegration of Tetrahymena ciliary axonemes at higher concentrations of ATP. Cell Motil. Cytoskeleton 9: 191-204. Warner, F.D., and Zanetti, N.C. (1980): Properties of microtubule sliding disintegration in isolated Tetrahymena cilia. J. Cell Biol. 86:436-445. Witman, G.B. (1986): Isolation of Chlamydomonas flagella and flagellar axonemes. Methods Enzymol. 134:280-290. Witman, G.B., Plummer, J., and Sander, G. (1978): Chlamydomonas flagellar mutants lacking radial spokes and central tubules. Structure and function of specific axonemal components. J. Cell Biol. 76:729-747. Yano, Y., and Miki-Noumura, T. (1981): Recovery of sliding ability in arm-depleted flagellar axonemes after recombination with extracted dynein 1. J. Cell Sci. 48:223-239.
Microtubule Sliding in Flagellar Axonemes of Chlamydomonas Mutants Missing Inner- or Outer-Arm Dynein: Velocity Measurements on New Types of Mutants by an Improved Method Eiji Kurimoto and Ritsu Kamiya Department of Molecular Biology, School of Science, Nagoya University, Nagoya, Japan To help understand the functional properties of inner and outer dynein arms in axonemal motility, sliding velocities of outer doublets were measured in disintegrating axonemes of Chlumydomonus mutants lacking either of the arms. Measurements under improved solution conditions yielded significantly higher sliding velocities than those observed in a previous study [Okagaki and Kamiya, 1986, J . Cell Biol. 103:1895-19021. As in the previous study, it was found that the velocities in axonemes of wild type (wt)and a mutant (odul) missing the outer arm differ greatly: 18.5 2 4.1 p d s e c for wt and 4.4 & 2.3 p d s e c for odul at 0.5 mM Mg-ATP. In contrast, axonemes of two types of mutants (idu2 and idu4) that lacked different sets of two inner-arm heavy chains displayed velocities almost identical with the wild-type velocity. Moreover, axonemes of a non-motile double mutant idu2 x idu4 underwent sliding disintegration at a similar high velocity, although less frequently than in axonemes of single mutants. These observations support the hypothesis that the inner and outer dynein arms in disintegrating axonemes drive microtubules at different speeds and it is the faster outer arm that determines the overall speed when both arms are present. The inner arm may be important for the initiation of sliding. The axoneme thus appears to be equipped with two (or more) types of motors with different intrinsic speeds. Key words: sliding velocity, mechanochemistry, dynein mutant, axonemal motility, cilia
INTRODUCTION
Recent biochemical and ultrastructural studies have indicated that the inner and outer dynein arms in cilia and flagella differ significantly in subunit composition and arrangement within the axoneme [Huang et al., 1979; Goodenough and Heuser, 1985; Goodenough et al., 1987; Sale et al., 1989; Luck and Piperno, 1989; Piperno et al., 19901. It is not established, however, whether the two types of arms differ in function. The finding that removal of the outer dynein arm from sea urchin sperm axonemes results in two-fold reduction in reactivated motility has suggested that the two have almost identical functions [Gibbons and Gibbons, 1973; Yano and MikiNoumura, 19811. Studies with Chlumydomonas mutants, 0 1991 Wiley-Liss, Inc.
on the other hand, have suggested that a significant functional difference exists between the two types of arms [Brokaw and Kamiya, 1987; Kamiya et al., 19891; for example, mutants lacking a large part of the inner-arm dynein are non-motile, while those missing the entire outer arms are motile. Hence, functions specific to each type of arm remain to be elucidated. Okagaki and Kamiya [ 19861 attempted to compare
Received November 27, 1990; accepted April 3, 1991 Address reprint requests to Ritsu Kamiya, Department of Molecular Biology, School of Science, Nagoya University, Nagoya 464-01, Japan.
276
Kurimoto and Kamiya
the functional properties of the inner and outer arms by measuring the sliding velocity of microtubules in wildtype (wt) and mutant Chlumydomonus axonemes. Using the method of Summers and Gibbons [ 19711 to induce axoneme disintegration, they found that axonemes of a mutant (oda1; previously called oda38) lacking the outer arm had a maximal sliding rate of as low as 2.0 pm/sec, whereas wt axonemes had a rate of 13.2 pm/sec. They also reported that the axoneme of a non-motile mutant lacking a significant part of the inner arms (pf23) had a difficulty in undergoing sliding disintegration; its axonemes exhibited disintegration only very rarely and at an unexpectedly low rate of 1-2 pm/sec. However, it was not certain whether the difficulty and low rate were general features of axonemes missing the inner arm or were features limited to this particular mutant. Hence studies with other inner-arm mutants have been wanted. We previously reported on isolation and characterization of outer-arm-missing Chlumydomonus mutants (odus) that retained reduced motility [Kamiya and Okamoto, 1985; Kamiya, 19881. More recently, we isolated two types of motile mutants [idaA and idaB types (previously named idu and idb)] that lacked different sets of inner-arm heavy chains [Kamiya et al., 1989, 19911; of the currently identified three inner-arm subspecies 11,12, and I3 [Piperno et al., 19901, idaA type lacks I1 and idaB type 12. In the present study, we measured the microtubule sliding velocity in these new mutants under improved solution conditions. The measurements yielded significantly higher velocities than those measured in the previous study. We found that, whereas the maximal sliding velocity in odal axoneme was about one-fifth of the wild-type velocity as observed in the previous study, the velocities in axonemes of idaA, idaB, and a double mutant between idaA and idaB were almost identical with that of the wt axonemes. These results raise the possibility that the sliding movements caused by the inner and outer arms significantly differ in velocity, and thus pose a puzzling question of how the two arms can work efficiently within a normal axoneme. MATERIALS AND METHODS Strains
Chlumydomonus reinhurdtii 137c (wild type (wt); and -), a mutant missing the entire outer mating type dynein arm (odul), and two types of inner-arm mutants, idu2 (idaA type) and idu4 (idaB type), were used. Isolation and characterization of these strains have been described [Kamiya and Okamoto, 1985; Kamiya, 1988; Kamiya et al., 19911. Culture of Cells and Isolation of Axonemes
+
For isolation of flagella, cells were grown in 600 ml of Tris-acetate-phosphate (TAP) medium [Gorman
and Levine, 19651 with aeration over a 12 h/12 h light/ dark cycle. Double mutants odul X idu2, odul x idu4, and idu2 X idu4 tended to grow only short flagella in the TAP medium. To induce better flagellation, cells of these double mutants were transferred to 200 ml of 10 mM Hepes (pH 7.4) by centrifugation and resuspension when the growth in the TAP medium appeared complete [Whitman, 19861. The cells grew flagella after being kept aerated for 2-4 h, although the flagella were still shorter than normal. Flagella were isolated by the method of Witman et al. [ 19781, using dibucaine-HC1 to detach flagella from the cell blodies. The isolated flagella were suspended in HMDS solution consisting of 10 mM Hepes (pH 7.4), 5 mM MgSO,, 1 mM DTT, and 4% sucrose. Induction of Sliding Disintegration
Sliding disintegration of axonemes was induced by the method of Okagaki and Kamiya [1986] with modifications. The suspension of flagella (conc.: about 0.2 mg/ml) was sonicated in a Kubota 200M sonicator (Kubota Co., Tokyo) at a setting of 20 W for about 60 sec. The fragmented axonemes were collected by centrifugation at 10,OOOg for 25 min and demembranated by suspending the pellet in an ice-chilled HMDEKP solution [30 mM Hepes (pH 7.4), 5 mM MgSO,, 50 mM K-acetate, 1 mM dithiothreitol (DTT), 1 mM EGTA, 0.5% polyethylene glycol (PEG 20,000 mol wt)] containing 0.4% Nonidet P-40. A small volume of the suspension of fragmented axonemes was mixed with 10 vol of a reactivation solution consisting of HMDEKP and an appropriate concentration of ATP. A drop of the sample was applied into a space between a glass slide and a coverslip, of which two opposite sides were sealed with Vaseline. To wash out axonemes that were not attached to the glass surface, this specimen was first perfused with the same reactivation solution by placing a drop of the solution and a filter paper tip by the two unsealed sides of the coverslip. Next, it was perfused with the reactivation solution containing 1.O-2.0 pg/ml nagarse (type VII bacterial protease; Sigma Chemical Co., St. Louis, MO). Sliding disintegration of the axonemes usually took place within a minute after the perfusion was started and was finished within a few minutes. The disintegration process was observed with a dark-field microscope equipped with an Olympus 40 X Apo objective (N.A. l.O), an Olympus DC condenser (N.A. 1.20-1.33), and a light source of a 100 W highpressure mercury arc lamp (Ushio, Tokyo, Japan). The process was recorded on videotape with an SIT camera (model C2400-08: Hamamatsu Photonics Co., Hamamatsu, Japan) and an S-VHS videotape recorder (Panasonic AG3800: Matsushita Electronics Co., Osaka, 1°C. Japan). All observations were carried out at 23
*
Microtubule Sliding in Dynein-Deficient Axonemes
Measurement of Sliding Velocity
Videotaped images of the sliding microtubules were superimposed on a monitor screen of a personal computer. An end of a moving microtubule was tracked with a “mouse cursor” and its coordinate was analyzed as a function of time. The average sliding velocity in one sliding event was determined by using a home-made program. Maximal sliding velocity (Vmax) and apparent Michaelis constant (Km) for Mg-ATP were calculated from the linear regression line of double reciprocal plots of data at different concentrations of Mg-ATP. Estimation of Percentage of Disintegrating Axonemes The percentage of axonemes that underwent sliding disintegration under a given set of conditions was estimated by counting the numbers of disintegrated and intact axonemes about 5 min after the onset of perfusion with ATP and protease. An axoneme fragment apparently disintegrated through sliding was counted as the one that underwent sliding, irrespective of how many microtubules underwent sliding in the particular fragment. More than 200 axonemes were counted for a single measurement. It is important to note that we did not observe the axonemes until the disintegration process was finished. This is because the occurrence of sliding disintegration was sensitive to the illumination of the microscope for unknown reasons; when the disintegration process was continually monitored under the microscope, far fewer axonemes usually underwent the disintegration. RESULTS Microtubule Sliding Rate Measured Under Improved Solution Conditions
Following Okagaki and Kamiya [ 19861, we induced microtubule sliding by perfusing fragmented axoneme specimens with ATP and nagarse, a protease with broad substrate specificity. Disintegrating axonemes were observed with a dark-field microscope and recorded on videotape by means of an SIT camera and a videotape recorder. The sliding, once started, proceeded fairly constantly; as observed in previous studies, the velocity did not depend on the overlap length between microtubules [Takahashi et al., 19821. The previous data on sliding velocity [Okagaki and Kamiya, 19861 were unsatisfactory in that the measured velocity was lower than the velocity calculated from the waveform of beating axonemes. Disagreement was particularly evident with odal axonemes; maximal sliding velocities in disintegrating wt and odal axonemes were 13.2 p d s e c and 2.0 p d s e c , whereas the values calcu-
277
lated from their in vivo waveforms were 15-20 pm/sec and 4.5-7 p d s e c , respectively (calculated from Brokaw and Kamiya [ 19871; see Discussion). In an effort to improve the experimental conditions, we changed the buffer solution from the one with low ionic strength used by Okagaki and Kamiya to the one used routinely for reactivating demembranated cell models [Kamiya and Okamoto, 19861. Thus we used 30 mM Hepes instead of 10 mM Hepes and included 50 mM K-acetate; K-acetate has been shown to have an effect to improve motility in reactivated sperm cell models [Gibbons et al., 19851, and the potassium concentration of 50 mM has been found to be optimal for reactivation of Chlamydomonas cell models [Kagami and Kamiya, 19901. We found that, although the low-ionic-strength buffer [Summers and Gibbons, 19711 seemed to favor the occurrence of sliding disintegration, it resulted in lower sliding velocities, as described below. Figure 1 shows histograms of the measured sliding velocity in mutant axonemes in the presence of 0.5 mM Mg-ATP. The mean sliding velocities (k standard deviations) in wt and odal were 18.5 (k4.1) p d s e c and 4.4 ( k 2 . 3 ) p d s e c , respectively. These rates were about twice as high as those measured in the previous study at 1 mM Mg-ATP: 9.2 (k1.0) p d s e c (wt)and 1.8 (k0.5) p d s e c (odal (Table 2 of Okagaki and Kamiya [1986]). The higher speed appeared to be mostly, if not entirely, due to the solution condition of the perfusion medium; our own measurement under the previous condition yielded a velocity of 13.7 ? 4.1 p d s e c at 0.5 mM Mg-ATP for wt axoneme (number of measurements = 75). The sliding velocity varied with the Mg-ATP concentration in a manner consistent with Michaelis-Menten kinetics. From the double reciprocal plots of sliding velocities at various Mg-ATP concentrations (Fig. 2), the maximal sliding velocity and the apparent Michaelis constant were determined to be 25.6 p d s e c and 177 p M (wt)and 5.1 p d s e c and 65 p M (odal). These maximal sliding rates were also about twice as high as the previous values (see above), but the Km values were similar to those previously obtained [158 pM (wt)and 64 p M (odul)]. The five- to six-fold difference between the maximal sliding rates in wt and oda 1 axonemes observed in the previous study was also observed in this study, despite the two-fold increase in the measured velocity for each kind of axoneme. Sliding Velocities in Inner Arm-Deficient Axonemes
We next measured the sliding rates in two types of mutants, ida2 (idaA type) and idu4 (idaB type), which lack different sets of two inner-arm heavy chains. As described in the Introduction, ida2 lacks I1 and ida4 I2
Kurimoto and Kamiya
wt
20 10
0 20 10 Fig. 2. Mg-ATP dependence of sliding velocities in wt ( 0 ) and odal ( 0 ) axonemes. Double reciprocal plot of the sliding velocities against the Mg-ATP concentration. Maximal sliding velocities (Vmax) and apparent Michaelis constants (Km) for Mg-ATP calculated from the linear regression lines were 25.6 p d s e c and 177 pM for wt,and 5.1 p d s e c and 64.7 pM for odul, respectively.
0
ida 2
20
measured to be 78 p d s e c and 54 Hz (idu2), 102 p d s e c and 62 Hz (idu4), and 155 p d s e c and 63 Hz (wt)[Kamiya et al., 19911. The sliding velocities in ida2 and ida4 axonemes were almost identical with the wild-type velocity; at 0.5 mM ATP, the average velocities measured were 19.0 (24.4) p d s e c for idu2 and 17.7 (k3.8) p d s e c for idu4, both being comparable with 18.5 p d s e c of wt (Fig. 1). The dependence of the sliding velocity on the Mg-ATP concentrations was also similar among these three strains (Fig. 3); reciprocal plots yielded the maximal velocity of 29.6 p d s e c and Km of 214 p M for idu2 and 22.2 p d s e c and 150 p M for idu4.
10
0
ida 4
20 10
0
ida2xida4
1
5
C
10
30
20
Sliding Velocity (,m/s) Fig. 1. Histograms of sliding velocities in axonemes of wt and dynein arm mutants at 0.5 mM Mg-ATP. Mean velocity *standard deviation = 18.5 ? 4.1 p d s e c (wt); 4.4 2.3 p d s e c (odul); 19.0 4.4 p d s e c (idu2); 17.7 ? 3.8 p d s e c (idu4); 17.1 2.9 pmisec (idu2 x idu4). Temperature: 23°C.
*
*
*
inner-arm subspecies. [Kamiya et al., 19911. Both mutants swim at reduced speed; the average swimming velocities and flagellar beat frequencies at 25°C have been
Sliding Disintegration in Non-Motile Double Mutants
We previously reported that the double mutant between idu2 and ida4 is non-motile although each of the single mutants is motile [Kamiya et al., 1989, 19911. We found that the axoneme of the double mutant idu2 X idu4 slid apart when ATP and protease were perfused, although the sliding was observed much less frequently than with axonemes of wt or of single mutants. The sliding velocity at 0.5 mM Mg-ATP was 17.1 (k2.9) p d s e c (Fig. 1). The estimated maximal velocity and Km were 24.8 p d s e c and 167 pM, respectively, i.e., almost identical with the wild-type values (Fig. 3). The axoneme of another non-motile double-mutant odul x idu4 also underwent sliding disintegration on rare occasions. The sliding velocity at 0.5 mM Mg-ATP observed in seven cases was 1.5 (k0.9) p d s e c . This observation suggests that a subset of the inner dynein arms can cause microtubule sliding in an axoneme, albeit very
Microtubule Sliding in Dynein-Deficient Axonemes
0
Mg-ATP
(mM)
Fig. 3. Mg-ATP dependence of sliding velocities in inner-am-deficient mutant axonemes; the maximal sliding velocities (Vmax) and the apparent Michaelis constants (Km) for Mg-ATP from the linear regression lines of double reciprocal plots were 29.6 pmlsec and 214 pM for idu2 (A), 22.2 p d s e c and 150 pM for idu4 (O), and 24.8 pmlsec and 167 pM for a double mutant idu2 X idu4 (a).0 ; data for Wt.
slowly. However, the axoneme of another double mutant odul x ida2 did not display sliding disintegration. As mentioned before, the non-motile axonemes of double mutants idu2 x idu4 underwent sliding disintegration much less frequently than motile axonemes. To evaluate the readiness of sliding disintegration in the mutant axoneme, we measured the percentage of axonemes that underwent sliding at different Mg-ATP concentrations (see Materials and Methods for methods). The results (Fig. 4) indicate that the occurrence of sliding in the ida2 x ida4 axoneme was less frequent than in motile axonemes, and strongly depended on the Mg-ATP concentration; unlike sliding in all other mutants, the sliding in this double mutant was severely inhibited at high MgATP concentrations. DISCUSSION Large Difference in Sliding Velocity Between wt and oda We have measured the microtubule sliding velocity in disintegrating axonemes of wild-type and mutant Chlamydomonas by using an improved method. The use of solution conditions optimized for axoneme reactivation resulted in significantly higher sliding velocities than those obtained in a previous study. The new data indicate that mutant axonemes missing outer-arm dynein have a much lower sliding velocity than wild-type axonemes, but that those missing subspecies of inner-arm dynein (I1 and 12) have velocities almost identical with the wild-type velocity.
6.5
Mg-ATP
279
1.0
(mM)
Fig. 4. Sliding probability at different concentrations of Mg-ATP in axonemes of w6 ( O ) , odul (H), ida2 (A), idu4 (O), and a double mutant idu2 X idu4 ( 0 ) . The ordinate denotes the percentage of the axoneme fragments that apparently underwent sliding disintegration after a 5 min perfusion with protease and a concentration of Mg-ATP specified on the abscissa. An axoneme fragment was judged to have undergone disintegration even if only one outer doublet was extruded from it.
The maximal sliding velocity in wt axonemes , 25.6 pmhec, estimated by extrapolation, is above the range of estimate for the velocity in beating axonemes in vivo; according to Brokaw and Luck [ 19831, the mean sliding velocity in wt axonemes is 340 rad/sec, which corresponds to 15-20 p d s e c if we assume the distance between adjacent outer-doublet microtubules to be 4.5 -6 nm. (A correlated observation of sliding displacement of outer doublets and bend angle in sea urchin sperm axonemes has recently indicated that the functional interdoublet distance is 4.5 nm [Brokaw, 19891.) The maximal velocity in odal axonemes, 5.1 pmhec, is within the range of estimate of the in vivo velocity, which has been determined to be 100-120 rad/sec [Brokaw and Kamiya, 19871, i.e., 4.5-7 pm/sec. The maximal sliding velocities estimated from the measurements in vitro are thus higher than, or similar to, the velocities estimated for axonemes in vivo; this suggests that the in vitro velocities reflect well the properties of inner- and outer-arm dynein in vivo, although we cannot rule out the possibility that the sonication and protease treatment, used to induce effective axoneme disintegration might have impaired function of some dynein arms. The difference in sliding velocity between wt and odul strains is five-fold in vitro whereas it is three-fold in vivo; a similar large difference between the two strains in vitro has been observed by Okagaki and Kamiya [ 19861. Why the velocity difference between the two strains is consistently more pronounced in disintegrating axonemes than in beating axonemes may be due to the difference in nature of the two kinds of dynein arms as
280
Kurimoto and Kamiya
mechano-transducers; the microtubule sliding caused by inner arms may be less sensitive to load than that caused by outer arms (see below, for further discussion). Effects of Inner-Arm Depletion on the Sliding Velocity
The sliding velocities in ida2, ida4, and a double mutant ida2 X ida4 were almost identical with the wt velocity, in contrast with the low velocity in odal. Thus, the absence of the inner-arm subspecies does not appear to affect the sliding velocity of outer doublets. This result differs strikingly from the previous observation that the sliding velocity in a non-motile inner-arm mutant, pf23, was as low as one-sixth of that in wt. However, the data on pj23 axonemes were not reliable, because these axonemes displayed sliding disintegration only very rarely and the data were based on just a few measurements. A similar difficulty in undergoing sliding was observed with idu2 X ida4 axonemes; it appears that the lack of the inner arm suppresses the initiation of sliding. The result that the sliding velocities in inner-arm mutants were almost identical with the wt velocity is rather unexpected since, in beating flagella, the sliding velocities in these slowly swimming mutants must be lower than in wt; in fact, in vivo velocity in flagella of idal (an idaA-type mutant) has been determined to be 21 1 rad/sec, i.e., about two-thirds of the wild-type value [Brokaw and Kamiya, 19871. These results suggest that the outer-doublet sliding in idu2 and ida4 axonemes is more susceptible to the load present in beating axonemes than the sliding in wt axoneme; the absence of inner-arm components appears to decrease the force generated, but not the velocity of sliding in disintegrating axonemes. It would be interesting to measure the sliding force directly in these mutant axonemes, as has been performed with disintegrating axonemes of sea urchin sperm by using flexible glass needles [Kamimura and Takahashi, 1981; Oiwa and Takahashi, 19881. Implications of the Difference in Sliding Velocity Between Inner-Arm Mutants and Outer-Arm Mutants
If we assume that only little load is imposed upon microtubules in disintegrating axonemes, the effects of depletion of inner or outer arms on the sliding velocity as observed above are consistent with the hypothesis that the inner and outer arms drive microtubules at significantly different rates under load-free conditions. In addition, since the lack of the subspecies of the slower arm (the inner arm) neither decelerates nor accelerates the movement, the sliding rate should be mostly determined by the faster arm (the outer arm) when both arms are present within the same axoneme. A possible mechanism for such behavior would consist of inner arms not func-
tioning effectively in normal axonemes under disintegrating conditions; the slow inner arms may be unable to interact with the adjacent doublet microtubule. An alternative interpretation is that the inner arms do function together with outer arms in disintegrating axonemes, but do so with an accelerated cycling rate that matches that of the outer arm. The acceleration may be brought about through some dynamic interaction between the two arms; since the two arms are attached to the same microtubules, their movement must be somehow coupled. It remains, however, a challenging problem to understand how a system containing two (or more) arms with apparently different intrinsic speeds can work efficiently. Role of Inner Arms in Axonemal Motility
The axoneme of idu2xida4 mutant underwent sliding disintegration only very rarely, especially at high Mg-ATP concentrations, although the velocity was almost normal as long as sliding took place. This difficulty in induction of sliding was similar to the one experienced with pj23 axonemes [Okagaki and Kamiya, 19861, which lack the same inner-arm subspecies (I1 and 12) as those missing in ida2 X ida4; we also tried to measure the sliding velocity in this mutant during the course of this study, but the sliding events were too rare for a reliable measurement. These observations suggest that the lack of inner arms has an effect to suppress the initiation of sliding event. Lower concentrations of Mg-ATP have been shown to favor sliding disintegration of ciliary axonemes of Tetrahymena [Warner and Zanetti, 1980; Tanaka and Miki-Noumura, 19881, and in wild-type axonemes of Chlamydomonas [Okagaki and Kamiya, 19861; in these axonemes, the lower the Mg-ATP concentration , the more outer doublets were observed to undergo sliding. The present finding that the sliding disintegration in ida2 X ida4 axonemes has a strong dependence on MgATP concentration, whereas that in odal missing the outer arm does not have such a dependence (Fig. 4 ) , suggests that it is the outer arm that confers the ATP sensitivity on the axoneme. In other words, Chlumydomonas outer-arm dynein may need to be helped by inner-arm subspecies for its function at high Mg-ATP concentrations. The loss of flagellar beating in ida2X ida4 and pf23 mutants may have somehow come from the difficulty in the initiation of sliding of outer doublets. Another, though not independent, possibility is that the motility loss may have been caused by the disruption of the mechanism that organizes microtubule sliding into axonemal bending motion, in which the inner arm may be playing a crucial role. In fact, waveform analyses of beating axonemes revealed that lack of inner-arm subsets results in a reduction of shear amplitude [Brokaw and
Microtubule Sliding in Dynein-Deficient Axonemes
Kamiya, 19871. Inner-arm dynein may thus be important in developing the axonemal bend from straight movement of microtubule sliding. It will be of great interest to ask by physiological studies whether the lack of motility in ida2 X ida4 is due to defects in the formation of bend or in the propagation of bend within the axoneme.
ACKNOWLEDGMENTS
We thank Prof. Sho Asakura for discussion and encouragement throughout the present study. This work has been supported by grants-in-aid from the Ministry of Education, Science and Culture of Japan (01657001, 02239101).
REFERENCES Brokaw, C.J. (1989): Direct measurements of sliding between outer doublet microtubules in swimming sperm flagella. Science 243: 1593-1596. Brokaw, C.J., and Kamiya, R. (1987): Bending patterns of Chlamydomonas flagella. IV. Mutants with defects in inner and outer dynein arms indicate differences in dynein arm function. Cell Motil. Cytoskeleton 8:68-75. Brokaw, C.J., and Luck, D.J.L. (1983): Bending patterns of Chlamydomonas flagella. I. Wild-type bending patterns. Cell Motil. 3: I3 1-150. Gibbons, B.H., and Gibbons, I.R. (1973): The effects of partial extraction of dynein arms on the movement of reactivated seaurchin sperm. J . Cell Sci. 13:337-357. Gibbons, B.H., Tang, W.-J.Y., and Gibbons, I.R. (1985): Organic anions stabilize the reactivated motility of sperm flagella and the latency of dynein 1 ATPase activity. J. Cell Biol. 101: 1281-1287. Goodenough, U.W., and Heuser, J.E. (1985): Substructure of inner dynein arm, radial spokes, and the central pair/projection complex of cilia and flagella. J. Cell Biol. 100:2008-2018. Goodenough, U.W., Gebhart, B., Mermall, V., Mitchell, D., and Heuser, J.E. (1987): High-pressure liquid chromatography fractionation of Chlamydomonas dynein extracts and characterization of inner-arm dynein subunits. J. Mol. Biol. 194: 48 1-494. Gorman, D.S., and Levine, R.P. (1965): Cytochrome F and plastocyanin: their sequence in the photosynthetic electron transport chain of Chlamydomonas reinhardtii. Proc. Natl. Acad. Sci. USA 54:1665-1669. Huang, B., Piperno, G., and Luck, D.J.L. (1979): Paralyzed flagella mutants of Chlamydomonas reinhardtii defective of axonemal doublet microtubule arms. J. Biol. Chem. 254:3091-3099. Kagami, O., and Kamiya, R. (1990): Strikingly low ATPase activities in flagellar axonemes of a Chlamydomonas mutant missing outer dynein arms. Eur. J. Biochem. 187:44-446.
281
Kamimura, S . , and Takahashi, K. (1981): Direct measurement of the force of microtubule sliding in flagella. Nature 293:566-568. Kamiya, R. (1988): Mutations at twelve independent loci result in absence of outer dynein arms in Chlamydomonas reinhardtii. J. Cell Biol. 107:2253-2258. Kamiya, R., Kurimoto, E., and Muto, E. (1991): Two types of Chlamydomonas flagellar mutants missing different components of inner-arm dynein. J. Cell Biol. 112441-447. Kamiya, R., Kurimoto, E . , Sakakibara, H . , and Okagaki, T. (1989): A genetic approach to the function of inner and outer arm dynein. In Warner, F.D., Satir, P., and Gibbons, I.R. (eds.): “Cell Movement, Volume I . ” New York: Alan R. Liss, Inc., pp. 209-218. Kamiya, R., and Okamoto, M. (1985): A mutant of Chlamydomonas reinhardtii that lacks the flagellar outer dynein arm but can swim. J. Cell Sci. 74:181-191. Luck, R . A . , and Piperno, G. (1989): Dynein arm mutants of Chlamydomonas reinhardtii. In Warner, F.D., Satir, P., and Gibbons, I.R. (eds.): “Cell Movement, Volume 1.” New York: Alan R. Liss, Inc., pp. 49-60. Oiwa, K., and Takahashi, K. (1988): The force-velocity relationship for microtubule sliding in demembranated sperm flagella of the sea urchin. Cell Struct. Funct. 13:193-205. Okagaki, T., and Kamiya, R. (1986): Microtubule sliding in mutant Chlamydomonas devoid of outer or inner dynein arms. J. Cell Biol. 103:1895-1 902. Piperno, G., Ramanis, Z., Smith, E.F., and Sale, W.S. (1990): Three distinct inner dynein arms in Chlamydomonus flagella: Molecular composition and location in the axoneme. J . Cell Biol. 110:379-389. Sale, W.S., Fox, L.A., and Milgram, S.L. (1989): Composition and organization of the inner row of dynein arms. In Warner, F.D., Satir, P., and Gibbons, I.R. (eds.): “Cell Movement, Volume 1.” New York: Alan R. Liss, Inc., pp. 89-102. Summers, K.E., and Gibbons, I.R. (1971): Adenosine triphosphateinduced sliding of tubules in trypsin-treated flagella of sea urchin sperm. Proc. Natl. Acad. Sci. USA 68:3092-3096. Takahashi, K., Shingyoji, C., and Kamimura, S. (1982): Microtubule sliding in reactivated flagella. In Amos, W.B., and Duckett, J.G. (eds.): “Symposia of the Society for Experimental Biology XXXV, Prokaryotic and Eukaryotic Flagella.” Cambridge: Cambridge Univ. Press, pp. 159-177. Tanaka, M., and Miki-Nomura, T. (1988): Stepwise sliding disintegration of Tetrahymena ciliary axonemes at higher concentrations of ATP. Cell Motil. Cytoskeleton 9: 191-204. Warner, F.D., and Zanetti, N.C. (1980): Properties of microtubule sliding disintegration in isolated Tetrahymena cilia. J. Cell Biol. 86:436-445. Witman, G.B. (1986): Isolation of Chlamydomonas flagella and flagellar axonemes. Methods Enzymol. 134:280-290. Witman, G.B., Plummer, J., and Sander, G. (1978): Chlamydomonas flagellar mutants lacking radial spokes and central tubules. Structure and function of specific axonemal components. J. Cell Biol. 76:729-747. Yano, Y., and Miki-Noumura, T. (1981): Recovery of sliding ability in arm-depleted flagellar axonemes after recombination with extracted dynein 1. J. Cell Sci. 48:223-239.
