A MUTANT OF CHLAMYDOMONAS REINHARDTII THAT LACKS THE FLAGELLAR OUTER DYNEIN ARM BUT CAN SWIM

Transcription

1 J. CellSci. 74, (1985) 181 Printed in Great Britain Company ofbiobgists Limited 1985 A MUTANT OF CHLAMYDOMONAS REINHARDTII THAT LACKS THE FLAGELLAR OUTER DYNEIN ARM BUT CAN SWIM RITSU KAMIYA AND MITSUMASA OKAMOTO Institute of Molecular Biology, Faculty of Science, Nagoya University, Nagoya, Japan 464 SUMMARY A new type of Chlamydomonas mutant, which lacks the outer dynein arm but can swim, was isolated. Sodium dodecyl sulphate/polyacrylamide gel electrophoresis showed that four of the ten high-molecular-weight bands of dynein present in the wild-type axoneme are missing or diminished in the mutant axoneme. The mutant has a swimming rate of about 35 /an/s and a flagellar beat frequency of about 25 Hz, both of which are about 1/2'5 to 1/3 of those of the wild type. The mutant flagella beat with an asymmetric, cilia-type pattern, similar to the forward-swimming mode of the flagellar beating pattern of the wild type. However, unlike wild-type flagella, the mutant flagella never beat with a symmetrical waveform: when the cells were stimulated by intense light, the mutant transiently stopped beating its flagella, whereas the wild-type cell transiently swam backwards with the two flagella beating with a symmetrical waveform. Both wild-type and mutant cells could be demembranated by Nonidet P40 and their swimming reactivated by addition of Mg-ATP in the virtual absence of Ca 2 " 1 ". Double reciprocal plots of the beat frequency against ATP concentrations showed a linear relationship for both strains, yielding maximal frequencies of 44 Hz (wild type) and 23 Hz (mutant). The mutant axonemes can be reactivated only when the Ca 2+ concentration is lower than 10~ 6 M : at pca 4, the wild-type axonemes beat with a symmetrical waveform, but the mutant axonemes showed no movement. These findings indicate that the outer dynein arm is dispensable for flagellar beating of the asymmetric waveform (forward-swimming mode), but not for beating of the symmetrical waveform (backward-swimming mode), and thus suggest the importance of the outer dynein arm in the switching of flagellar waveforms. INTRODUCTION In most eukaryotic flagella the nine outer doublet microtubules bear two rows of projections, the inner and the outer dynein arms, which are the key molecular complexes for converting the chemical energy of ATP into the mechanical work of flagellar beating. These two kinds of arms have entirely different protein compositions (Huang, Piperno & Luck, 1979), but their functional differentiation has not yet been clarified. Several lines of evidence have indicated that the outer dynein arm is not absolutely necessary for the beating of a flagellar axoneme. Gibbons & Gibbons (1973) have shown that sea-urchin sperm axonemes, with the outer dynein arm selectively extracted by 0-5M-KC1, still beat in the presence of Mg-ATP: the maximal beat frequency is about half the original one, but the waveform is essentially unchanged. They have also shown that the outer arm can be reattached to the extracted Key words: Chlamydomonas, dynein, flagella.

2 182 R. Kamiya and M. Okamoto axoneme and the original beat frequency restored. In a similar experiment using trypsinized axonemes, Yano & Miki-Noumura (1981) directly demonstrated that the sliding velocity between the outer-doublet microtubules is halved after extraction of the outer arm. These results suggest that the inner and outer dynein arms are equivalent in generating force and function in an additive manner. The idea that the outer dynein arm is dispensable for axonemal beating is further supported by the natural occurrence of variant axonemes that lack the outer arm but are motile; e.g., the sperm of eel (Baccetti, Burrini, Dallai & Pallini, 1979; Gibbons, Gibbons & Baccetti, 1983) and the sperm of non-fertile man (Jouannet, Escalier, Serres & David, 1983). However, there is an exception. In the green alga Chlamydomonas, mutants known to lack the outer dynein arm (pfj3,pf22) are all non-motile (Huang et al. 1979). Therefore, the outer dynein arm seems to be indispensable forflagellarbeating in Chlamydomonas. However, we suspected that the mutants examined may have been a subset lacking the outer arm, and that the lack of motility may not be the direct consequence of the absence of the outer arm. Thus we tried to isolate mutants with reduced swimming rates, expecting that there would be mutants that could swim without the inner or the outer dynein arm. This paper describes the existence of a mutant that lacks the outer arm but can swim slowly. Quite unexpectedly, the mutantflagellawere found to beat only with an asymmetric, cilia-type pattern, and not with a symmetrical, flagella-type pattern. Owing to this feature, the mutant could swim forwards but not backwards. MATERIALS AND METHODS Isolation of mutants The method of Lewin (1954) was used to obtain mutants with impaired motility, with minor modifications. Wild-type Chlamydomonas reinhardtii 137c + was grown in Medium I of Sager & Granick (1953) to a cell density of about 1 X lt^/ml. A 15 ml cell suspension in a 9 cm Petri dish was irradiated with an ultraviolet lamp (Toshiba GL-15) at 40 cm above the sample for 5 min. The sample was then divided into 1 ml portions and each was inoculated into 5 ml of medium in a test tube of 1 cm diameter. After incubation in the dark for 24 h, the test tubes were kept under constant illumination for 3 days, and the cells grown at the bottom were carefully transferred with a Pasteur pipette to a test tube containing 5 ml of fresh medium. This process was repeated three to four times over a period of days. Cells from the bottom of the last test tube were streaked on 1-5% agar plates. Mutants with paralysed flagella or poor motility produced heaped colonies, whereas wild-type cells formed flattened ones. Since our purpose was to isolate motile mutants, we picked up colonies similar to those of the wild-type cells. The cells from those colonies were cultured in 0-5 ml of medium in a test tube without agitation. Mutants with aberrant motility, such as backward-swimming mutants (Nakamura, 1981) or uni-flagella mutants (Huang, Ramanis, Dutcher & Luck, 1982), grew only at the bottom, whereas the wild-type cells tended to gather at the upper surface of the medium. Mutants with reduced swimming speeds often showed intermediate distribution in a test tube: some cells gathered at the surface while others remained on the bottom. We saved cultures that showed unusual distribution patterns for observation with a dark-field microscope. The flagella of cells of those colonies that were judged as being slow-swimming mutants were examined by sodium dodecyl sulphate/polyacrylamide gel electrophoresis (SDS/PAGE) and electron microscopy. Demembranation and reactivation Demembranation and reactivation of cells of wild-type and mutant strains were carried out by the method described previously (Kamiya & Witman, 1984). Cells grown in the liquid medium were collected by centrifugation and demembranated in a solution containing 30mM-HEPES (ph 7-3),

3 Dynein outer armless Chlamydomonas 183 5mM-MgSO4, 1 mm-dithiothreitol (DTT), 1 mm-egta, 2SmM-KCl, 0-5% polyethylene glycol (PEG) and 0-1% Nonidet P40. Reactivation was carried out by suspending the demembranated cells in solutions containing 1 mm-atp and 2mM-EGTA. When the effect of Ca 2+ was examined, 2 mm-egta was replaced by a calcium-buffer system (Bessen, Fay & Witman, 1980). In experiments where the ATP concentration was changed, the reactivation solutions contained 30 mm- HEPES (ph7-3), 5mM-MgCl 2, 1 mm-dtt, 25mM-KCl, 2% PEG, 2mM-EGTA, 4m«- phosphocreatine, 0'5mg/ml (about 70 units/ml) of creatine phosphokinase (Sigma) and various concentrations of ATP. The flagellar beat frequency was measured with a stroboscope equipped with a frequency counter. To simplify measurement, flagella on the cell models rotating in a small area were chosen and at least 20 samples were measured at each ATP concentration. All experiments were performed at room temperature. Electron microscopy Flagella were isolated and demembranated by the method of Witman, Plummer & Sander (1978). Axonemes were fixed with 2% glutaraldehyde in the presence of 1% tannic acid, post-fixed with 1% OsCU, dehydrated through a series of ethanol solutions, and embedded in Epon 812. Sections were double-stained with aqueous solutions of 7-5% uranyl acetate and 0-4% lead citrate. A JEM100C microscope was used for observations. Other methods SDS/PAGE, to separate the high-molecular-weight (HMW) components of dynein, was carried out by the method of Jarvik & Rosenbaum (1980). The gel was composed of a 3% to 5% acrylamide gradient and 3 M to 8M-urea gradient. A silver staining method was used (Merril, Goldman, Sedman & Ebert, 1981). Dark-field micrographs of beating axonemes were taken with stroboscopic illumination powered by a Chadwick-Helmuth Strobex 236 power supply. The swimming rate of the mutant was obtained from records of swimming tracks taken with stroboscopic flashes of a constant frequency (usually 2Hz). RESULTS Isolation of an outer dynein arm-deficient mutant (oda) Over several months we isolated 60 mutants that swam considerably more slowly than the wild-type cells. We then examined the flagella from these strains by SDS/ PAGE for abnormal protein compositions in the range of the dynein HMW components. About half of them had some abnormality. Electron microscopy showed that 16 strains almost completely lackd the outer dynein arms, while other strains had either a complete number of inner and outer arms, or a complete number of inner arms and an incomplete number of outer arms. Although the 16 mutants without outer arms were isolated independently, they were almost identical in all aspects of the phenotype so far examined. We report here the results for one of those strains, oda38. Electron microscopy and SDS/PAGE Fig. 1 shows electron micrographs of cross-sections of wild-type and oda axonemes isolated from cultures at the late-log phase of growth. The outer dynein arm is completely missing from the oda axoneme. In a survey of cross-sections of 50 axonemes (i.e. 450 outer doublets), only six outer doublets of different axonemes (1-3% of the total) were found to bear the outer dynein arm, whereas essentially 100% of the doublets had the inner arm. The occurrence of the outer dynein arm in this

4 184 R. Kamiya and M. Okamoto 1A Fig. 1. Cross-sections of axonemes of oda (A,B) and wild-type strains (c). A, X 57000; B,C, X mutant is lower than in mutants without outer arms reported previously (Huanget al. 1979). In a control survey of cross-sections of 50 wild-type axonemes, the inner and the outer arms were found on about 100% and 90% of the outer doublets, respectively. This finding agrees with those of other reports (Huanget al. 1979; Hoops &Witman, 1983). Fig. 2 shows a portion of the SDS/PAGE pattern of the demembranated flagella in the region of the HMW peptides of dynein. The molecular weight of these components has been estimated to be to (Piperno & Luck, 1979). Of the 10 bands resolved in the pattern of the wild-type axoneme, two bands (2,5) were missing and two other bands (1,8) appeared to be weakened in theoda axoneme. Our preliminary results further suggested that the intensity of band 1 of the mutant axoneme varied from one sample to another, and that band 8 of the wild-type actually contained two overlapping bands, one of which was missing in the oda axoneme. Therefore, the oda axoneme probably lacks bands 2,5, one component of band 8, and part of band 1. The mobility of the HMW peptides changed greatly with changes in the electrophoresis conditions, such as the batch of SDS. This made it difficult to compare the pattern with those presented by others. However, the result that theoda axoneme lacks three or four HMW bands agrees with previous reports that the outer dynein arm of Chlamydomonas appears to have three or four HMW components (Huang et al. 1979; Piperno & Luck, 1979, 1981 ;Witman, Johnson, Pfister& Wall, 1983; Pfister, Fay &Witman, 1982). General behaviour of the oda mutant The oda mutant had twoflagellaof normal length and almost all cells swam during the daytime at the late-log phase of growth in a 12h/12h, light/dark, synchronous

5 Dynein outer armless Chlamydomonas 185 A B 8 - Fig. 2. Portions of SDS/PAGE patterns of wild-type (A) and oda axonemes (B), showing the HMW bands of dynein (indicated by bars). Bands that are missing or diminished in the oda axonemes are numbered. culture. The average swimming velocity was 35-0 ± 6-2 uin/s for 35 cells and the flagellar beat frequency 25-4± 3-2Hz for 32 flagella in one sample. Both of these values are 1/2-5 to 1/3 of those of the wild-type (Racey, Hallett & Nickel, 1981; Brokaw & Luck, 1983). The flagellar beating pattern in live oda cells appeared to be typical of cilia, similar to that of the wild-typeflagella (Fig. 3). However, the maximal bend angle of the principal bend in the oda axoneme (2-85 ± 0-35 rad for 13 flagella) appeared to be smaller than that in the wild-type flagella (3-30 ±0-40 rad for 15 flagella; see also Brokaw & Luck, 1983). This difference was more pronounced in reactivated axonemes (see below). Striking behaviour of the oda mutant was found when the cell was stimulated by intense light. The response of the wild-type Chlamydomonas to an intense light Fig. 3. Beating flagella of a live oda cell photographed with three flashes. This cell was tethered between the glass slide and coverslip. X Fig. 4. Stopping reaction of oda cells. Swimming oda cells were first photographed with a red filter (cut-off wavelength, 630 nm) placed between the light source and specimen, with an exposure of about 3 s. The filter was then quickly removed while the shutter of the camera remained open for an additional 0-5 s. These two cells stopped swimming upon removal of the red filter and did not move during the additional exposure. After about 1 s these cells resumed swimming. X 500.

6 186 R. Kamiya and M. Okamoto stimulus is transiently to swim backwards with the two flagella beating with a symmetrical, flagella-type pattern (Ringo, 1967; Schmidt & Eckert, 1976; Hyams & Borisy, 1978). When the oda cells were stimulated by quickly removing the red filter from the light source of the microscope, they did not swim backwards but stopped for 0-2-ls. During this interval, the two flagella on a cell were always arrested in a characteristic 'hands-up' position, pointing forwards with a slight curvature (Fig. 4). When observed under a microscope, the oda cells occasionally gathered around or swam away from thefieldof observation, as is usually the case with the wild-type cells. Hence the mutant cells appear to display more or less normal phototactic behaviour. Demembranation and reactivation Like the wild-type cells, the oda mutants could be demembranated by Nonidet P40 and their swimming reactivated by the addition of 1 mm-mg-atp at low calcium concentrations (less than 10~ 8 M). The slow swimming rate of the mutant was maintained in the cell models. When detached from the cell body the oda axoneme beat by itself, circling around in the medium (Fig. 5A), as observed with the wild-type axoneme (Fig. 5B; Bessenef al. 1980). Occasionally, the detached axonemes stuck to the glass surface by the proximal end and rotated at Hz (oda) or 0*5-2 Hz (wild type). Fig. 5c and D are images of such rotating axonemes taken with multiple flashes, Fig. 5. Movement of the reactivated axonemes of oda (A,C) and wild-type strains (B,D) photographed with multiple flashes. All had become spontaneously detached from the cell body. A,B. Axonemes circling against the glass surface; C,D, axonemes rotating in the counterclockwise direction with the proximal end sticking to the glass surface. The flash rate was chosen to give a clear picture of the sequence of the beating pattern. Axonemes of both strains usually beat 6-20 times during one complete rotation. X 1450.

7 Dynein outer armless Chlamydomonas SO l/[atp](mm-') Fig. 6. Double reciprocal plots of the beat frequency against the ATP concentration for wild-type ( ) and oda axonemes (o). showing the beating patterns of the two kinds of axonemes. The beating pattern of the reactivated oda axoneme was asymmetric, but the maximal bend angle of the principal bend (6pj 2*14 ± 0-24 rad for 18 axonemes) was smaller than that in the wild-type axoneme (3-30 ± 0-27 rad for 13 axonemes), while the angle of the reverse bend was almost the same (about 02 rad). Using an ATP-regenerating system (see Materials and Methods), we measured the flagellar beat frequency at various ATP concentrations (Fig. 6). Both the wild-type and the oda axonemes showed a linear relationship in the double reciprocal plot of the beat frequency against the ATP concentration, as in the case of the sea-urchin sperm axoneme (Brokaw, 1967; Gibbons & Gibbons, 1972). The maximal beat frequency (/max) and the apparent Michaelis constant (K m ) of the Chlamydomonas axonemes in our experimental system appeared to be 44 Hz and 100 fm for the wild type, and 23 Hz and 60 ^M for the oda strain, although these values varied by as much as ± 20% depending on the culture from which the models were prepared. An approximately twofold difference in /max and a 1-5-fold difference in K m have also been observed in the sea-urchin sperm axonemes before and after extraction of the outer dynein arm (Gibbons & Gibbons, 1973). However, a comparison of/mi* between the two kinds of Chlamydomonas axonemes would be meaningful only when the difference in waveform is taken into account. The fraction of beat period required for the propagation of the principal bend may be the same for both mutant and wild-type axonemes. If so, the twofold difference in /max and the 2/3-fold difference in 6 P would mean that the maximal sliding velocity of the outer doublet microtubules in the principal bend of the oda axoneme is about 1/3 of that in the wild-type axoneme. Effect ofca 2+ on the reactivated oda axoneme Chlamydomonas axonemes are known to display several kinds of Ca 2+ -dependent behaviour. First, the beating pattern of each axoneme changes from an asymmetric

8 188 R. Kamiya and M. Okamoto pattern to a symmetrical pattern when the Ca z+ concentration is changed from less than 10~ 6 M to 10~ 4 M, and the axonemes stop beating in the presence of more than 10~ 3 M-Ca 2+ (Hyams & Borisy, 1978; Bessen et al. 1980). Second, the balance of beating activity between the two flagella seems to depend on the Ca 2+ concentration in the 10~ 9 to 10~ 7 M range (Kamiya & Witman, 1984). Third, the axonemes become detached in demembranated cell models when the Ca 2+ concentration is raised above 10~ 4 M. This phenomenon may be related to the autotomy of flagella observed when cells are exposed to harsh environments (Lewin, Lee & Fang, 1982). The demembranated cell models of the oda mutant first swam in the medium when reactivated in the virtual absence of Ca 2+. With time, however, they began to rotate in a small area because one of the two flagella was gradually inactivated. In most of such rotating cell models, the eyespot was on the outside of the cell body, indicating that the flagellum nearer to the eyespot (the cis-eyespot flagellum) was more active than the other. This situation is the same as in the wild-type cell models (Kamiya & Witman, 1984). In the presence of 10~ 7 M-Ca 2+, the oda axonemes were reactivated only very poorly; flagellar beating was so weak that most of the axonemes stuck to the glass surface. Thus we were unable to determine which of the two flagella was more active under these conditions. At 10~ 5 to 10~ 4 M-Ca 2+, almost no movement was observed, and flagellar detachment occurred at 10~ + M-Ca 2+. The fact that the oda axonemes did not show any active beating at higher Ca 2+ concentrations is consistent with the observation that the flagella of live oda cells stopped beating when the cells were stimulated by intense light, since photostimulation is believed to induce a transient increase in the intra-flagella Ca 2+ concentration (Schmidt & Eckert, 1976; Hyams & Borisy, 1978). It is likely that the oda axoneme is intrinsically unable to beat with a symmetrical waveform. DISCUSSION This is the first report of the isolation of a Chlamydomonas mutant (oda38) that lacks the outer dynein arm but can swim. The mutant swims forwards at a rate of about 1/3 of the normal value and cannot swim backwards. Fifteen other mutants without outer arms were isolated during the course of the study and were found to be of the same phenotype. Our preliminary genetic analysis indicated that all these mutants have single-site mutations and are classified into seven different complementation groups. The fact that seven different groups have the same phenotype strongly suggests that the reduced swimming rate and the inability to swim backwards are direct consequences of the absence of the outer arm. We cannot, however, rule out the possibility that there are still other kinds of mutants that can swim backwards without the outer dynein arm. We suppose that the reason for the lack of motility in mutantspfl3 andp/22 (Huanget al. 1979) is not simply that they lack the outer arm. These mutants may have defects that inhibit both the formation of the outer arm and some functions that are essential to the flagellar beating mechanism. For example, their inner arm function may also have been impaired, although their inner arm dynein seems to have a normal degree of ATPase activity (Piperno & Luck, 1979,

9 Dynein outer armless Chlamydomonas ). It would be interesting to examine how flagellar beating is inhibited at the molecular level in such mutants. The oda axonemes beat with a cilia-type pattern in the presence of 1 mm-mg-atp and a concentration of Ca 2+ lower than 10~ 6 M, but did not show any movement at higher Ca 2+ concentrations. The wild-type flagella also display cilia-type beating at lower than 10~ 6 M-Ca 2+ but switch to flagella-type beating at KT^M-Ca 2 " 1 " (Hyams & Borisy, 1978; Bessen et al. 1980). These observations indicate that the outer dynein arm is important for theflagella-type beating. The mechanism of the Ca 2+ -dependent switching of the flagellar waveform may involve control of the activities of the inner and outer dynein arms. It is interesting to note that the ATPase activity of Tetrahymena dynein becomes Ca 2+ -sensitive when calmodulin is present (Blum, Hayes, Jamieson & Vanaman, 1980). Also, a recent study showed that different degrees of inhibition of the dynein ATPase by erythro-9-[3-(2-hydroxynonyl)]- adenine affect the switching of the flagellar waveform of Leishmania promastigotes in various ways (Alexander & Burns, 1983). All of these findings lead to the speculation that the inner and outer arms have different mechanochemical properties and that their differential activation or inactivation by Ca 2+ is the basis for the Ca z+ -induced switching of waveforms. Our results support the view that the outer dynein arm is dispensable for axonemal beating. The question then arises as to whether the inner arm is also dispensable. There have been no reported examples of flagellar axonemes that beat without the inner dynein arm, except for axonemes of rather unusual morphology (Baccetti et at. 1979). Hence, the inner arm may be more important than the outer arm and may actually control it. In our collection of slow-swimming mutants, none was found to lack the inner arm. Thus, the mutants without inner arms may be of a quite different phenotype from that of the oda mutant. One isolated inner-armless mutant (pf2j) was found to be non-motile (Huang et al. 1979). Mutants without inner or outer dynein arms, like those we have isolated, should be useful for biochemical and structural studies of dynein. The inner and outer arms are molecular assemblies made up of about 10 polypeptides each, including two highmolecular-weight ATPase peptides in each arm (Piperno & Luck, 1979, 1981; Tang, Bell, Sale & Gibbons, 1982; Bell & Gibbons, 1982; Pfister et al. 1982). This structural complexity of the similar inner and outer dynein arms makes it difficult to interpret direct biochemical studies correlating composition with function. Isolation of a wide variety of mutants with defects in dynein function would facilitate characterization of the dynein structure and the function of each component. All the mutants without outer arms that we have isolated so far appear to lack almost the entire outer arm: they may have defects in the regulation of the synthesis or assembly of the dynein subunits, or may have mutations in the components to be assembled at the basal portion of the arm. Other kinds of mutants, lacking some specific small parts of the dynein arm, probably exist. Single-arm mutants should also be useful as parent strains from which to obtain such mutants, since dynein deficiency in single-arm mutants will readily lead to mutants with paralysed flagella and hence simplify the screening procedures.

10 190 R. Kamiya and M. Okamoto This work was supported by a grant from the Ministry of Education, Science and Culture of Japan. We thank Sho Asakura (Nagoya University) and Don Caspar (Brandeis University) for encouragement and helpful discussions. R. K. is grateful to Professor Yoshihiro Tsubo of Kobe University, who kindly taught him methods of genetics. REFERENCES ALEXANDER, R. & BURNS, R. G. (1983). Differential inhibition by erythro-9-[3-(2-hydroxynonyl)] adenine of flagella-like and cilia-like movement olleishmania promastigotes. Nature, Lend. 305, BACCETTI, B., BURRINI, A. G., DALLAI, R. & PALLINI, V. (1979). The dynein electrophoretic bands in axonemes naturally lacking the inner or the outer arm. J. Cell Biol. 80, BELL, C. W. & GIBBONS, I. R. (1982). Structure of the dynein-1 outer arm in sea urchin sperm flagella. II. Analysis by proteolytic cleavage. J. biol. Chem. 257, BESSEN, M., FAY, R. B. & WITMAN, G. B. (1980). Calcium control of waveform in isolated flagellar axonemes of Chlamydomonas. J. Cell Biol. 86, BLUM, J. J., HAYES, A., JAMIESON, J. G. A. & VANAMAN, T. C. (1980). Calmodulin confers calcium sensitivity on ciliary dynein ATPase. J. Cell Biol. 87, BROKAW, C. J. (1967). Adenosine triphosphate usage by flagella. Science, N.Y. 156, BROKAW, C. J. &LUCK, D. J. L. (1983). Bending patterns of CWawydomonas flagella I. Wild-type bending patterns. Cell Motil. 3, GIBBONS, B. H. & GIBBONS, I. R. (1972). Flagellar movement and adenosine triphosphatase activity in sea urchin sperm extracted with Triton X-100. J. Cell Biol. 54, GIBBONS, B. H. & GIBBONS, I. R. (1973). The effect of partial extraction of dynein arms on the movement of reactivated sea-urchin sperm. J. Cell Set. 13, GIBBONS, B. H., GIBBONS, I. R. & BACCETTI, B. (1983). Structure and motility of the flagellum of eel spermatozoa. J. submicrosc. Cytol. 15, HOOPS, H. J. & WITMAN, G. B. (1983). Outer doublet heterogeneity reveals structural polarity related to beat direction in Chlamydomonas flagella. J. Cell Biol. 97, HUANG, B., PIPERNO, G. LUCK, D. J. L. (1979). Paralyzed flagella mutants of Chlamydomonas reinhardtu defective for axonemal doublet microtubule arms. J. biol. Chem. 254, HUANG, B., RAMANIS, Z., DUTCHER, S. K. & LUCK, D. J. L. (1982). Uniflagellar mutants of Chlamydomonas: evidence for the role of basal bodies in transmission of positional information. Cell 29, HYAMS, J. S. & BORISY, G. G. (1978). Isolated flagellar apparatus of Chlamydomonas: characterization of forward swimming and alteration of waveform and reversal of motion by calcium ions in vitro. J. Cell Sd. 33, JARVIK, J. W. & ROSENBAUM, J. L. (1980). Oversized flagellar membrane protein in paralyzed mutants of Chlamydomonas reinhardii. J. Cell Biol. 85, JOUANNET, P., ESCALIER, D., SERRES, C. & DAVID, G. (1983). Motility of human sperm without outer dynein arms. J'. submicrosc. Cytol. 15, KAMIYA, R. & WITMAN, G. B. (1984). Submicromolar levels of calcium control the balance of beating between the two flagella in demembranated models of Chlamydomonas. jf. Cell Biol. 98, LEWIN, R. A. (1954). Mutants of Chlamydomonas moetuusii with impaired motility. J. gen. Microbiol. 11, LEWIN, R. A., LEE, T.-H. & FANG, L.-S. (1982). Effects of various agents on flagellar activity, flagellar autotomy and cell viability in four species of Chlamydomonas (Chlorophyta, Volvocales). Soc. exp. Biol. Symp. 35, MERRIL, C. R., GOLDMAN, D., SEDMAN, S. A. & EBERT, M. H. (1981). Ultrasensitive stain for proteins in polyacrylamide gels shows regional variation in cerebrospinal fluid protein. Science, N.Y. 211, NAKAMURA, S. (1981). Two different backward-swimming mutants of Chlamydomonas reinhardtii. Cell Struct. Fund. 6, PFISTER, K. K., FAY, R. B. & WITMAN, G. B. (1982). Purification and polypeptide composition of dynein ATPases from Chlamydomonas flagella. Cell Motil. 2,

11 Dynein outer armless Chlamydomonas 191 PIPERNO, G. & LUCK, D. J. L. (1979). Axonemal adenosine triphosphatases from flagella of Chlamydomonas reinhardtii: purification of two dyneina. J. biol. Chem. 252, PIPERNO, G. & LUCK, D. J. L. (1981). Inner arm dyneins from flagella of Chlamydomonas reinhardtii. Cell 27, RACEY, T. J., HALLETT, R. & NICKEL, B. (1981). A quasi-elastic light scattering and cinematographic investigation of motile Chlamydomonas reinhardtii. Biophys. J. 35, RINGO, D. L. (1967). Flagellar motion and fine structure of the flagellar apparatus in Chlamydomonas.J. Cell Biol. 33, SAGER, R. & GRANICK, S. (1953). Nutritional studies with Chlamydomonas reinhardii. Ann. N.Y. Acad. Sd. 56, SCHMIDT, J. A. & ECKERT, R. (1976). Calcium couples flagellar reversal to photostimulation in Chlamydomonas reinhardtii. Nature, Land. 262, TANG, W.-J. Y., BELL, C. W., SALE, W. S. & GIBBONS, I. R. (1982). Structure of the dynein-1 outer arm in sea urchin sperm flagella. I. Analysis by separation of subunits. J. biol. Chem. 257, WITMAN, G. B., PLUMMER, J. & SANDER, G. (1978). Chlamydomonas flagellar mutants lacking radial spokes and central tubules. Structure and function of specific axonemal components. J. Cell Biol. 76, WITMAN, G. B., JOHNSON, D. A., PFISTER, K. K. & WALL, J. S. (1983). Fine structure and molecular weight of the outer arm dyneins of Chlamydomonas.J. submicrosc. Cytol. 15, YANO, Y. & MIKJ-NOUMURA, T. (1981). Recovery of sliding ability in arm-depleted flagellar axonemes after recombination with extracted dynein \.J. Cell Sd. 48, {Received 3 July Accepted 24 September 1984) Note added in proof Mitchell & Rosenbaum (J. Cell Biol., in press) have independently isolated a motile Chlamydomonas mutant (pf28) lacking the outer dynein arm.

12