- Email: [email protected]
Microtubule-membrane interactions in ctenophore swimming plate cilia
0040-8166j8l/00170197$02.00
TISSUE & CELL 1981 13 (2) 197-208 % 1981 Longman Group Ltd
WILLIAM
MICROTUBULE-MEMBRANE CTENOPHORE SWIMMING
L. DENTLER
INTERACTIONS PLATE CILIA
IN
ABSTRACT. The cilia in ctenophore swimming plates are organized into long rows and the cilia within each of the rows are connected to one another by interciliary bridges. The interciliary bridges form a type of intracellular junction and are periodically spaced at 15 nm intervals along the long axis of a cilium. The bridges bind adjacent cilia together even after dissolution of the ciliary membrane by nonionic detergent. Interciliary bridges are attached to the compartmenting lamellae, which are paracrystalline structures composed of spherical particles which are periodically attached to the outer doublet microtubules at the sites to which the microtubulemembrane bridges are bound. It is proposed that the compartmenting lamellae are modifications of the ciliary microtubule-membrane bridge found in other eukaryotic cilia and that it is associated with a junctional complex that binds adjacent cilia together in swimming plates.
Introduction CILIARY and flagellar doublet microtubles are interesting structures in part because they contain specific and well-defined sites to which various structures are attached. The best known of these structures include the two dynein arms (Allen, 1968; Gibbons, 1965) and the radial spokes (Dentler and Cunningham, 1977; Warner and Satir, 1974). Less well understood are the sites to which the peripheral links (also called nexin or interdoublet links) (Stephens, 1970; Warner, 1976) and the microtubule-membrane bridges (Dentler et al., 1980; Sattler and Staehelin, 1974) are attached. The site for the microtubule-membrane bridge is particularly important because a variety of different structures that are interposed between the outer doublet microtubules and the ciliary membranes are attached at this site. These structures include the microtubulemembrane bridges in protozoan cilia (Dentler rj al., 1980; Sattler and Staehelin, 1974), mastigonemes (Bouck et al., 1978; Markey
Marine Biological Laboratory, Woods Hole, MAO2543 and Department of Physiology and Cell Biology, McCollum Laboratories, University of Kansas, Lawrence, KS 66045. Received 2 January 13
1981.
and Bouck, 1977; Piccinni et al., 1973, paraflagellar rods (Vickerman, 1969), the r-link in eel sperm (Bacetti et al., 1979), as well as the compartmenting lamellae in ctenophores, first described by Afzelius (1961). The cilia of ctenophore swimming plates are well suited for studies of microtubulemembrane associations. In Mnemiopsis, the swimming plates are organized into eight meridional rows on the body surface. Each row contains approximately 20 plates and each plate contains more than 100,000 individual cilia which are tightly packed together (Afzelius, 1961). All of the cilia in a single plate beat simultaneously and in the same direction. Studies by Tamm (1980) have shown that the coordination of ciliary beating within a plate is due to the cilia being mechanically coupled to one another. Although Afzelius (1961) and Tamm (1980) have suggested that the compartmenting lamellae are involved with the coupling of cilia to one another, there has been no detailed description of the structures associated with the lamellae. In this report, evidence is presented showing that the cilia are joined together by filamentous interciliary bridges that are periodically spaced along the long axis of the 197
198
DENTLER
cilia. The bridges are attached, across the ciliary membrane, to the compartmenting lamellae which are, in turn, linked to the outer doublet microtubules at the site of the microtubule-membrane bridge (Dentler et a/., 1980). Materials and Methods The ctenophores Mnemiopsis leidyi were collected during the summer at Woods Hole, MA. Meridional rows of swimming plates were dissected from the organism and were fixed in 2% glutaraldehyde in 100 mM phosphate buffer, ph 7.0, for 1 hr at room temperature. The rows were rinsed in phosphate buffer and were post-fixed in 1% 0~04 in phosphate buffer for 1 hr at 4°C were rinsed in distilled water, and were stained in 1 % uranyl acetate for several hours at room temperature. In some experiments, filtered sea water was substituted for phosphate buffer; in others, post-fixation with 0~04 was omitted. To dissolve the ciliary membranes, swimming plates were dissected and were placed in 0.05% Nonidet P-40 (Particle Data Laboratories, Elmhurst, Ill.) in filtered sea water buffered with 10 mM Tris-Cl, pH 8.0, and 0.1 mM EDTA, for 0.5-S min at room temperature. Both the extracted cilia and the controls were fixed in glutaraldehyde and post-fixed in 0~04 as described above. All of the tissue was dehydrated in acetone and embedded in Epon-Araldite. Thin sections were cut with a diamond knife, were stained with methanolic uranyl acetate and lead citrate, and were examined and photographed using a Philips EM 300. Results A cross-section of a Mnemiopsis swimming plate is shown in Fig. 1. An entire swimming
plate measures approximately 2 mm wide by 20 pm thick and is composed of greater than 100,000 individual cilia (Afzelius, 1961). Each cilium is composed of the familiar ‘9+ 2’ array of microtubules and contains two filaments, called compartmenting lamellae (Afzelius, 1961), which extend from the base to the tip of each cilium. The lamellae are attached to doublet microtubules number 3 and 8, which are on opposite sides of the axoneme (Figs. 1, 2). A plane that passes through the lamellae also passes through the central pair microtubules. All of the cilia in a swimming plate are oriented in the same direction and the plane of the lamellae and of the central microtubules is perpendicular to the direction of ciliary beat. Each of the compartmenting lamellae in Mnemiopsis is comprised of a paracrystalline array of spherical particles, 15 nm in diameter (Figs. 2-7). Throughout the major portion of a cilium the lamellae are 5-7 particles wide but near the base and tip of the cilium, the lamellae gradually taper and are only 1-2 particles wide (Fig. 3). The lamellae are attached to the doublet microtubules by thin filaments that extend from the lamellar particles to the wall of the B-microtubule at a site close to the attachment of the B-microtubule to the A-microtubule wall (Figs. 2, 3, 6, 7). In longitudinal sections, the bridges that link the lamellar particles to the microtubules can be observed to occur at 15 nm intervals (Figs. 3-7). The compartmenting lamellae are attached to the ciliary membrane both at their edges and along their lateral axis (Figs. 2, 3). Filamentous bridges are often observed to connect the ciliary membrane directly to the particles that comprise the lamellae (Fig. 2). This association, however, appears to be somewhat weak and is not always preserved
Fig. 1. Cross-section of a portion of a Mnemiopsis swimming plate. Adjacent x 23,000. are connected to one another by interciliary bridges (small arrows).
cilia
Fig. 2. Cross-section of swimming plate cilia showing the interciliary bridge (IC), and the attachment site of a compartmenting lamella (CL) to an outer doublet microtubule (M, arrow). Bridges between individual particles of a lamella and the membrane are also visible (small arrowheads). x I 11,000.
200
DENTLER
Fig. 3. Longitudinal section near the base of a swimming plate cilium showing a compartmenting lamella (reduced in size at the ciliary base) and its periodic attachment to the membrane and to the doublet microtubule (arrows). x 78,000. Fig. 4. Longitudinal section showing two cilia, a portion of the compartmenting lamellae (CL) in each cilium and the periodically spaced interciliary bridges (arrows). x 100,000. Fig. 5.-Longitudinal cross-section of the compartmenting the beaded lamellae structure (arrows). x 55,000.
lamellae (CL) showing
MICROTUBULE-MEMBRANE
INTERACTIONS
IN CTENOPHORE
Figs. 6,7. Longitudinal sections of swimming plate cilia showing the beaded structure of the compartmenring lamellae (CL). the periodic attachment of the lamellae to the doublet microtubule (small arrows) and the periodic interciliary bridges (large arrows). The central pair microtubules are labelled CP in Fig. 7. Fig. 7 is a linear translation of Fig. 6 in which the print was shifted one time for a distance equivalent to IS nm. x 55,000.
201
202
DENTLER
Fig. 8. Cross-section of a Mnemiopsu swimming plate that was fixed in glutaraldehyde and not post-fixed with osmium tetroxide. lnterciliary bridges (arrows) are indicated. x 55,000.
swimming plates before (Fig. 9) and Figs. 9, IO. Cross-sections of Mnemiopsis after (Fig. 10) treatment with non-ionic detergent. Although the ciliary membrane was dissolved by the detergent, the cilia remained attached to one another due to rhe interciliary bridges connecting the compartmenting lamellae of adjacent cilia. x 55,000.
MICROTUBULE-MEMBRANE
INTERACTIONS
by the fixation methods that have been employed. In some cilia, such as those shown in Fig. 9, the membrane appears to have been detached from the lamellae and is somewhat swollen in comparasion with the cilia shown in Figs. 1 and 2. Even in these swollen cilia portions of the lamellae appear to be attached to the membrane by long and thin filaments, which suggests that the swelling of the ciliary membrane during preparation for electron microscopy pulled the membrane away from the lamellae as well as from the outer doublet microtubules. To further examine the structure of the compartmenting lamellae and their association with the ciliary membrane, a variety of fixation conditions were tested. Although hypotonic, hypertonic, and isotonic fixatives were tested, we were not able to preserve the lateral connections between the lamellae and the membrane in every cilium. The connections were most frequently preserved, however, when the cilia were fixed with glutaraldehyde in either sea water or in phosphate buffer but were not post-fixed with osmium tetroxide. The omission of osmium from the fixation regimen resulted in cilia that frequently had their membranes tightly wrapped around the axoneme and the compartmenting lamellae (Fig. 8). There were, however, significantly more membranes that appeared to be broken in various places around the axoneme in the cilia fixed without osmium than there were in cilia fixed with osmium post-fixation. The matrix within the cilia that was not postfixed with osmium was generally very dense and it was not possible to resolve bridges between either the outer doublet microtubules and the membrane or the compartmenting lamellae and the membrane. Although the associations between the ciliary membrane and the lateral axis of the compartmenting lamellae were labile and difficult to routinely preserve, the edges of the lamellae, where adjacent cilia meet, were always closely apposed to the membrane, regardless of the fixation methods that were employed. This result probably indicates that the bridges between the edges of the lamellae and the membrane are considerably stronger than are those between the lateral axis of the lamellae and the membrane. Individual cilia in a swimming plate are
IN
CTENOPHORE
205
arranged in rows that are parallel to the plane of the compartmenting lamellae. The cilia within these rows are closely apposed to one another where the edges of the compartmenting lamellae join with the membrane (Figs. 1, 2, 8, 9). Thin filaments, called interciliary bridges, extend from one cilium to another and appear to be extensions of the bridges that link the particles of the compartmenting lamellae to the membrane (Fig. 2). In longitudinal sections (Figs. 3-6) the interciliary bridges can be seen to be periodically spaced along the cilia, binding the adjacent cilia together at intervals of approximately I5 nm. The periodic bridges extend for most of the length of the cilia but are absent at the tips and bases of the cilia, where the lamellae are reduced in width (Fig. 3). The bridges connecting the lamellae to the membrane are easily observed even in these narrow lamellae (Fig. 3). The periodicity of the interciliary bridges is the same as that of the bridges between the lamellar particles and the doublet microtubules and is approximately that of the diameter of the particles that comprise the lamellae. We propose that the interciliary bridges as well as the lamellar-microtubule bridges are similar to the bridges that link the individual particles of the lamellae to the membrane. The interciliary bridges that extend from the lamellae to the membrane may be attached to the lamellae in an adjacent cilium by passing through the ciliary membrane and into the ciliary membrane of the adjacent cilium. Alternatively, the lamellae-membrane bridges may attach to membraneassociated proteins that bind the membranes of adjacent cilia together. The interciliary bridges would, therefore, serve as an intercellular junction. If the interciliary bridges form a junction that holds adjacent cilia together, then it might be expected that the cilia would remain attached to one another even after the membrane is removed. To test this expectation, swimming plates were dissected from Mnemiopsis and were treated with non-ionic detergent prior to fixation. Examples of the swimming plates fixed before and after detergent treatment are shown in Figs. 9 and 10. The cilia were clearly held together even in the absence of the ciliary membrane. These results indicate that the interciliary bridges are tightly associated with the
206 compartmenting connected cilia.
DENTLER lamellae
within each of the
Discussion Filamentous structures, or bridges, between the membrane and outer doublet microtubules are present in most cilia and flagella. The bridges attach to the outer doublets at a site near the junction of the A- and B-microtubule walls that is distinct from the sites on the outer doublets to which the two dynein arms, peripheral (nexin) links, and the radial spokes are attached (Dentler, 1981). The walls of the doublet microtubles, therefore, have at least five separate sites to which specific axonemal structures are attached. To date, there is little that is known about these sites although it is possible that they are defined either by the structure of the microtubule wall or by proteins that are attached along the microtubule. One possibility for the origin of different sites along the microtubules is that differences in the tubulin molecules that comprise the walls may exist and these differences may provide particular domains in the wall to which the structures can bind. Although a number of different tubulin molecules have been found in cilia and flagella (Stephens, 1978), there has been no evidence to correlate these tubulins with specific binding domains on the microtubule. Although microtubule-membrane bridges are present in most cilia and flagella, the structure of the bridges is highly variable. Perhaps the simplest example of a bridge is that found in Tetvahymena cilia, which appears as a single filament interposed between the doublet microtubule and the ciliary membrane (Dentler et. al., 1980; Sattler and Staehelin, 1974). The bridge in Ochromonas and Euglena flagella appears somewhat thicker and is connected, through the flagellar membrane, to extraflagellar hairs, or mastigonemes (Markey and Bouck, 1977; Bouck et al., 1978; Picinni et al., 1975). The paraflagellar rod in trypanosomes (Vickerman, 1969; Anderson and Ellis, 1965) and in Euglena flagella (Bouck et al., 1978) is another example of a complex paracrystalline structure that is attached to the outer doublet microtubules at the microtubule-membrane bridge site and that extends to the flagellar membrane. The compartmenting lamellae in ctenophore
swimming plate cilia are similar to both the paraflagellar rod, in that they are paracrystalline structures attached to the outer doublet microtubules, and the mastigonemes, in that extensions of the lamellae not only bind to the ciliary membrane but are also attached to structures on the external membrane surface which, in turn, attach adjacent cilia together. Since each of these modifications of the microtubule-membrane bridge appear to attach to the same site on the outer doublet microtubules, it is likely that the proteins responsible for the attachments are quite similar to one another in different cilia and flagella. At this time, however, little is known about the proteins that form the bridges. The best understood microtubule-membrane bridge is found in Tetrahymena and Aequipecten cilia in which one of the proteins that compose the bridge has been proposed to be a high molecular weight dynein-like protein (Dentier, 1977; Dentler et al., 1980). Whether the dynein-like protein is tightly attached to the microtubule and reaches out to domains on the ciliary membrane or whether the protein is a membrane-bound protein that attaches to the microtubule is not known at this time. It is expected, however, that studies of the microtubule-membrane bridges in cilia and flagella may further our understanding of the interactions between cytoplasmic microtubules and membranes since filamentous bridges are also observed between the cytoplasmic structures. What are the functions of the interciliary bridges? Since all of the cilia in a swimming plate beat simultaneously in a direction perpendicular to the plane of the compartmenting lamellae and since the interciliary bridges are extensions of the lamellae that extend between adjacent cilia, it is likely that the interciliary bridges join the cilia together and ensure that the cilia in a swimming plate beat as a single unit. Further evidence for this role comes from the observation that the axonemes and compartmenting lamellae of adjacent cilia remained linked together by the interciliary bridges even after removal of the ciliary membranes by detergents. The interciliary bridges, therefore, form a type of intercellular junction that is associated with ciliary microtubules. Evidence for the importance of physical connections between beating ctenophore
MICROTUBULE-MEMBRANE
INTERACTIONS
swimming plate cilia has been recently presented by Tamm (1980). He separated portions of a swimming plate with a needle and showed that only the central portion of a swimming plate is stimulated to beat by the nervous system. The majority of the ciliary movement is dependent upon mechanical connections to the central portion of the swimming plate. When the cilia in the central portion of a plate were separated from the cilia at the edges, the cilia near the edges of the plate ceased to beat in register with the central cilia. The coordination of ciliary beating throughout the swimming plate depended not on the transmission of nervous impulses through the cells at the ciliary bases, which were not disrupted by the needle, but rather, upon the adhesion of cilia to one another across the swimming plate. Upon removal of the needle, the cilia in the separated portions of the plate rapidly regained their associations with the central cilia and the coordinated beating of all the
IN
CTENOPHORE
207
cilia was resumed. Whether or not the interciliary bridges were rapidly re-connected upon removal of the needle was not determined. It will be important to study the fate of the interciliary bridges before and after separation of the swimming plate cilia to determine if the bridges are absolutely required for the coordination of ciliary beating. It remains possible that the bridges may merely stabilize the interactions between adjacent cilia and are not totally required for ciliary coordination. Acknowledgements I would like to thank Dr Ray Stephens for introducing me to ctenophore cilia and to Drs Sid Tamm and S. Beroue for a number of stimulating conversations. This work was supported by Grants AM 21672 and GM 24584 from the National Institutes of Health and RR0 7037 and 3045 from the University of Kansas.
References AFZELIUS. B. A. 1961. The fine structure of the cilia from ctenophore swimming plates. J. biophys. hiochem. Cytol., 9, 383. ALLEN, R. D. 1968. A re-investigation of cross-sections of cilia. J. Cell Biol., 37, 825. ANDERSON, W. A. and ELLIS, R. A. 1965. Ultrastructure of Trypanosoma lewisi: flagellum, microtubules, and the kinetoplast. J. Protozool., 12, 483. BACETTI,B., BURRINI, A. G., DALLAI, R. and PALLINI, V. 1979. The dynein electrophoretic bands in axonemes naturally lacking the inner or the outer arm. J. CeN Bid., 80, 334. BOUCK, G. B., ROGALSKI,A. and VALAITIS,A. 1978. Surface organization and composition of Euglma. II. Flagellar mastigonemes. J. Cell Bid., 77, 805. DENTLER, W. L. 1977. Fine structural localization of phosphatases in cilia and basal bodies of Tefrahymena pyriformis.
Tissue & Cell, 9, 209.
DENTLER, W. L. 1981. Microtubule-membrane interactions in cilia and flagella. Inr. Rev. Cytol., 72, 1. DENTLER, W. L. and CUNNINGHAM, W. P. 1977. Structure and organization of radial spokes in cilia of Tetrahymena
pyriformis.
J. Morphol.,
153, 143.
DENTLER, W. L., PRATT, M. M. and STEPHENS,R. E. 1980. Microtubule-membrane interactions in cilia. II. Photochemical cross-linking of bridge structures and the identification of a membrane-associated dynein-like ATPase. J. CeN Biol., 84, 381. GIBBONS,I.R. 1965. Chemical dissection of cilia. Arch. Biol., 76, 317. MARKEY. D. R. and BOUCK, G. B. 1977. Mastigonemeattachment in 0chromonas.J. Ultrastruct. Rrs.,59, 173. PICINNI,E.,ALBERGONI, V. and COPPELLOTTI,0. 1975. ATPase activity in flagella from Euglena gracilis. Localization of the enzyme and effects of detergents. J. Protorool., 22, 33 I, SATTLER,C. A. and STAEHELIN,L. A. 1974. Ciliary membrane differentiations in Tetrahymena pyrijormis. Tetrahymena has four types of cilia. J. Cell Biol., 62, 473.
208 STEPHENS,R. E. 1970. Isolation
DENTLER
of nexin-the linkage protein responsible for maintenance of the nine-fold configuration of flagellar axonemes. Biol. Bull., 139, 438. STEPHENS,R. E. 1978. Primary structural differences among tubulin subunits from flagella, cilia, and the cytoplasm. Biochemistry, 17, 2882. TAMM, S. L. 1980. Biological Bulletin (in press). VICKERMAN,K. 1969. On the surface coat and flagellar adhesion in trypanosomes. J. Cell Sci., 5, 163. WARNER, F. D. 1976. Ciliary inter-microtubule bridges. J. Cell Sri., 20, 201. WARNER, F. D. and SATIR, P. 1974. The structural basis of ciliary bend formation. Radial spoke positional changes accompanying microtubule sliding. J. Cell Bid., 63, 35
TISSUE & CELL 1981 13 (2) 197-208 % 1981 Longman Group Ltd
WILLIAM
MICROTUBULE-MEMBRANE CTENOPHORE SWIMMING
L. DENTLER
INTERACTIONS PLATE CILIA
IN
ABSTRACT. The cilia in ctenophore swimming plates are organized into long rows and the cilia within each of the rows are connected to one another by interciliary bridges. The interciliary bridges form a type of intracellular junction and are periodically spaced at 15 nm intervals along the long axis of a cilium. The bridges bind adjacent cilia together even after dissolution of the ciliary membrane by nonionic detergent. Interciliary bridges are attached to the compartmenting lamellae, which are paracrystalline structures composed of spherical particles which are periodically attached to the outer doublet microtubules at the sites to which the microtubulemembrane bridges are bound. It is proposed that the compartmenting lamellae are modifications of the ciliary microtubule-membrane bridge found in other eukaryotic cilia and that it is associated with a junctional complex that binds adjacent cilia together in swimming plates.
Introduction CILIARY and flagellar doublet microtubles are interesting structures in part because they contain specific and well-defined sites to which various structures are attached. The best known of these structures include the two dynein arms (Allen, 1968; Gibbons, 1965) and the radial spokes (Dentler and Cunningham, 1977; Warner and Satir, 1974). Less well understood are the sites to which the peripheral links (also called nexin or interdoublet links) (Stephens, 1970; Warner, 1976) and the microtubule-membrane bridges (Dentler et al., 1980; Sattler and Staehelin, 1974) are attached. The site for the microtubule-membrane bridge is particularly important because a variety of different structures that are interposed between the outer doublet microtubules and the ciliary membranes are attached at this site. These structures include the microtubulemembrane bridges in protozoan cilia (Dentler rj al., 1980; Sattler and Staehelin, 1974), mastigonemes (Bouck et al., 1978; Markey
Marine Biological Laboratory, Woods Hole, MAO2543 and Department of Physiology and Cell Biology, McCollum Laboratories, University of Kansas, Lawrence, KS 66045. Received 2 January 13
1981.
and Bouck, 1977; Piccinni et al., 1973, paraflagellar rods (Vickerman, 1969), the r-link in eel sperm (Bacetti et al., 1979), as well as the compartmenting lamellae in ctenophores, first described by Afzelius (1961). The cilia of ctenophore swimming plates are well suited for studies of microtubulemembrane associations. In Mnemiopsis, the swimming plates are organized into eight meridional rows on the body surface. Each row contains approximately 20 plates and each plate contains more than 100,000 individual cilia which are tightly packed together (Afzelius, 1961). All of the cilia in a single plate beat simultaneously and in the same direction. Studies by Tamm (1980) have shown that the coordination of ciliary beating within a plate is due to the cilia being mechanically coupled to one another. Although Afzelius (1961) and Tamm (1980) have suggested that the compartmenting lamellae are involved with the coupling of cilia to one another, there has been no detailed description of the structures associated with the lamellae. In this report, evidence is presented showing that the cilia are joined together by filamentous interciliary bridges that are periodically spaced along the long axis of the 197
198
DENTLER
cilia. The bridges are attached, across the ciliary membrane, to the compartmenting lamellae which are, in turn, linked to the outer doublet microtubules at the site of the microtubule-membrane bridge (Dentler et a/., 1980). Materials and Methods The ctenophores Mnemiopsis leidyi were collected during the summer at Woods Hole, MA. Meridional rows of swimming plates were dissected from the organism and were fixed in 2% glutaraldehyde in 100 mM phosphate buffer, ph 7.0, for 1 hr at room temperature. The rows were rinsed in phosphate buffer and were post-fixed in 1% 0~04 in phosphate buffer for 1 hr at 4°C were rinsed in distilled water, and were stained in 1 % uranyl acetate for several hours at room temperature. In some experiments, filtered sea water was substituted for phosphate buffer; in others, post-fixation with 0~04 was omitted. To dissolve the ciliary membranes, swimming plates were dissected and were placed in 0.05% Nonidet P-40 (Particle Data Laboratories, Elmhurst, Ill.) in filtered sea water buffered with 10 mM Tris-Cl, pH 8.0, and 0.1 mM EDTA, for 0.5-S min at room temperature. Both the extracted cilia and the controls were fixed in glutaraldehyde and post-fixed in 0~04 as described above. All of the tissue was dehydrated in acetone and embedded in Epon-Araldite. Thin sections were cut with a diamond knife, were stained with methanolic uranyl acetate and lead citrate, and were examined and photographed using a Philips EM 300. Results A cross-section of a Mnemiopsis swimming plate is shown in Fig. 1. An entire swimming
plate measures approximately 2 mm wide by 20 pm thick and is composed of greater than 100,000 individual cilia (Afzelius, 1961). Each cilium is composed of the familiar ‘9+ 2’ array of microtubules and contains two filaments, called compartmenting lamellae (Afzelius, 1961), which extend from the base to the tip of each cilium. The lamellae are attached to doublet microtubules number 3 and 8, which are on opposite sides of the axoneme (Figs. 1, 2). A plane that passes through the lamellae also passes through the central pair microtubules. All of the cilia in a swimming plate are oriented in the same direction and the plane of the lamellae and of the central microtubules is perpendicular to the direction of ciliary beat. Each of the compartmenting lamellae in Mnemiopsis is comprised of a paracrystalline array of spherical particles, 15 nm in diameter (Figs. 2-7). Throughout the major portion of a cilium the lamellae are 5-7 particles wide but near the base and tip of the cilium, the lamellae gradually taper and are only 1-2 particles wide (Fig. 3). The lamellae are attached to the doublet microtubules by thin filaments that extend from the lamellar particles to the wall of the B-microtubule at a site close to the attachment of the B-microtubule to the A-microtubule wall (Figs. 2, 3, 6, 7). In longitudinal sections, the bridges that link the lamellar particles to the microtubules can be observed to occur at 15 nm intervals (Figs. 3-7). The compartmenting lamellae are attached to the ciliary membrane both at their edges and along their lateral axis (Figs. 2, 3). Filamentous bridges are often observed to connect the ciliary membrane directly to the particles that comprise the lamellae (Fig. 2). This association, however, appears to be somewhat weak and is not always preserved
Fig. 1. Cross-section of a portion of a Mnemiopsis swimming plate. Adjacent x 23,000. are connected to one another by interciliary bridges (small arrows).
cilia
Fig. 2. Cross-section of swimming plate cilia showing the interciliary bridge (IC), and the attachment site of a compartmenting lamella (CL) to an outer doublet microtubule (M, arrow). Bridges between individual particles of a lamella and the membrane are also visible (small arrowheads). x I 11,000.
200
DENTLER
Fig. 3. Longitudinal section near the base of a swimming plate cilium showing a compartmenting lamella (reduced in size at the ciliary base) and its periodic attachment to the membrane and to the doublet microtubule (arrows). x 78,000. Fig. 4. Longitudinal section showing two cilia, a portion of the compartmenting lamellae (CL) in each cilium and the periodically spaced interciliary bridges (arrows). x 100,000. Fig. 5.-Longitudinal cross-section of the compartmenting the beaded lamellae structure (arrows). x 55,000.
lamellae (CL) showing
MICROTUBULE-MEMBRANE
INTERACTIONS
IN CTENOPHORE
Figs. 6,7. Longitudinal sections of swimming plate cilia showing the beaded structure of the compartmenring lamellae (CL). the periodic attachment of the lamellae to the doublet microtubule (small arrows) and the periodic interciliary bridges (large arrows). The central pair microtubules are labelled CP in Fig. 7. Fig. 7 is a linear translation of Fig. 6 in which the print was shifted one time for a distance equivalent to IS nm. x 55,000.
201
202
DENTLER
Fig. 8. Cross-section of a Mnemiopsu swimming plate that was fixed in glutaraldehyde and not post-fixed with osmium tetroxide. lnterciliary bridges (arrows) are indicated. x 55,000.
swimming plates before (Fig. 9) and Figs. 9, IO. Cross-sections of Mnemiopsis after (Fig. 10) treatment with non-ionic detergent. Although the ciliary membrane was dissolved by the detergent, the cilia remained attached to one another due to rhe interciliary bridges connecting the compartmenting lamellae of adjacent cilia. x 55,000.
MICROTUBULE-MEMBRANE
INTERACTIONS
by the fixation methods that have been employed. In some cilia, such as those shown in Fig. 9, the membrane appears to have been detached from the lamellae and is somewhat swollen in comparasion with the cilia shown in Figs. 1 and 2. Even in these swollen cilia portions of the lamellae appear to be attached to the membrane by long and thin filaments, which suggests that the swelling of the ciliary membrane during preparation for electron microscopy pulled the membrane away from the lamellae as well as from the outer doublet microtubules. To further examine the structure of the compartmenting lamellae and their association with the ciliary membrane, a variety of fixation conditions were tested. Although hypotonic, hypertonic, and isotonic fixatives were tested, we were not able to preserve the lateral connections between the lamellae and the membrane in every cilium. The connections were most frequently preserved, however, when the cilia were fixed with glutaraldehyde in either sea water or in phosphate buffer but were not post-fixed with osmium tetroxide. The omission of osmium from the fixation regimen resulted in cilia that frequently had their membranes tightly wrapped around the axoneme and the compartmenting lamellae (Fig. 8). There were, however, significantly more membranes that appeared to be broken in various places around the axoneme in the cilia fixed without osmium than there were in cilia fixed with osmium post-fixation. The matrix within the cilia that was not postfixed with osmium was generally very dense and it was not possible to resolve bridges between either the outer doublet microtubules and the membrane or the compartmenting lamellae and the membrane. Although the associations between the ciliary membrane and the lateral axis of the compartmenting lamellae were labile and difficult to routinely preserve, the edges of the lamellae, where adjacent cilia meet, were always closely apposed to the membrane, regardless of the fixation methods that were employed. This result probably indicates that the bridges between the edges of the lamellae and the membrane are considerably stronger than are those between the lateral axis of the lamellae and the membrane. Individual cilia in a swimming plate are
IN
CTENOPHORE
205
arranged in rows that are parallel to the plane of the compartmenting lamellae. The cilia within these rows are closely apposed to one another where the edges of the compartmenting lamellae join with the membrane (Figs. 1, 2, 8, 9). Thin filaments, called interciliary bridges, extend from one cilium to another and appear to be extensions of the bridges that link the particles of the compartmenting lamellae to the membrane (Fig. 2). In longitudinal sections (Figs. 3-6) the interciliary bridges can be seen to be periodically spaced along the cilia, binding the adjacent cilia together at intervals of approximately I5 nm. The periodic bridges extend for most of the length of the cilia but are absent at the tips and bases of the cilia, where the lamellae are reduced in width (Fig. 3). The bridges connecting the lamellae to the membrane are easily observed even in these narrow lamellae (Fig. 3). The periodicity of the interciliary bridges is the same as that of the bridges between the lamellar particles and the doublet microtubules and is approximately that of the diameter of the particles that comprise the lamellae. We propose that the interciliary bridges as well as the lamellar-microtubule bridges are similar to the bridges that link the individual particles of the lamellae to the membrane. The interciliary bridges that extend from the lamellae to the membrane may be attached to the lamellae in an adjacent cilium by passing through the ciliary membrane and into the ciliary membrane of the adjacent cilium. Alternatively, the lamellae-membrane bridges may attach to membraneassociated proteins that bind the membranes of adjacent cilia together. The interciliary bridges would, therefore, serve as an intercellular junction. If the interciliary bridges form a junction that holds adjacent cilia together, then it might be expected that the cilia would remain attached to one another even after the membrane is removed. To test this expectation, swimming plates were dissected from Mnemiopsis and were treated with non-ionic detergent prior to fixation. Examples of the swimming plates fixed before and after detergent treatment are shown in Figs. 9 and 10. The cilia were clearly held together even in the absence of the ciliary membrane. These results indicate that the interciliary bridges are tightly associated with the
206 compartmenting connected cilia.
DENTLER lamellae
within each of the
Discussion Filamentous structures, or bridges, between the membrane and outer doublet microtubules are present in most cilia and flagella. The bridges attach to the outer doublets at a site near the junction of the A- and B-microtubule walls that is distinct from the sites on the outer doublets to which the two dynein arms, peripheral (nexin) links, and the radial spokes are attached (Dentler, 1981). The walls of the doublet microtubles, therefore, have at least five separate sites to which specific axonemal structures are attached. To date, there is little that is known about these sites although it is possible that they are defined either by the structure of the microtubule wall or by proteins that are attached along the microtubule. One possibility for the origin of different sites along the microtubules is that differences in the tubulin molecules that comprise the walls may exist and these differences may provide particular domains in the wall to which the structures can bind. Although a number of different tubulin molecules have been found in cilia and flagella (Stephens, 1978), there has been no evidence to correlate these tubulins with specific binding domains on the microtubule. Although microtubule-membrane bridges are present in most cilia and flagella, the structure of the bridges is highly variable. Perhaps the simplest example of a bridge is that found in Tetvahymena cilia, which appears as a single filament interposed between the doublet microtubule and the ciliary membrane (Dentler et. al., 1980; Sattler and Staehelin, 1974). The bridge in Ochromonas and Euglena flagella appears somewhat thicker and is connected, through the flagellar membrane, to extraflagellar hairs, or mastigonemes (Markey and Bouck, 1977; Bouck et al., 1978; Picinni et al., 1975). The paraflagellar rod in trypanosomes (Vickerman, 1969; Anderson and Ellis, 1965) and in Euglena flagella (Bouck et al., 1978) is another example of a complex paracrystalline structure that is attached to the outer doublet microtubules at the microtubule-membrane bridge site and that extends to the flagellar membrane. The compartmenting lamellae in ctenophore
swimming plate cilia are similar to both the paraflagellar rod, in that they are paracrystalline structures attached to the outer doublet microtubules, and the mastigonemes, in that extensions of the lamellae not only bind to the ciliary membrane but are also attached to structures on the external membrane surface which, in turn, attach adjacent cilia together. Since each of these modifications of the microtubule-membrane bridge appear to attach to the same site on the outer doublet microtubules, it is likely that the proteins responsible for the attachments are quite similar to one another in different cilia and flagella. At this time, however, little is known about the proteins that form the bridges. The best understood microtubule-membrane bridge is found in Tetrahymena and Aequipecten cilia in which one of the proteins that compose the bridge has been proposed to be a high molecular weight dynein-like protein (Dentier, 1977; Dentler et al., 1980). Whether the dynein-like protein is tightly attached to the microtubule and reaches out to domains on the ciliary membrane or whether the protein is a membrane-bound protein that attaches to the microtubule is not known at this time. It is expected, however, that studies of the microtubule-membrane bridges in cilia and flagella may further our understanding of the interactions between cytoplasmic microtubules and membranes since filamentous bridges are also observed between the cytoplasmic structures. What are the functions of the interciliary bridges? Since all of the cilia in a swimming plate beat simultaneously in a direction perpendicular to the plane of the compartmenting lamellae and since the interciliary bridges are extensions of the lamellae that extend between adjacent cilia, it is likely that the interciliary bridges join the cilia together and ensure that the cilia in a swimming plate beat as a single unit. Further evidence for this role comes from the observation that the axonemes and compartmenting lamellae of adjacent cilia remained linked together by the interciliary bridges even after removal of the ciliary membranes by detergents. The interciliary bridges, therefore, form a type of intercellular junction that is associated with ciliary microtubules. Evidence for the importance of physical connections between beating ctenophore
MICROTUBULE-MEMBRANE
INTERACTIONS
swimming plate cilia has been recently presented by Tamm (1980). He separated portions of a swimming plate with a needle and showed that only the central portion of a swimming plate is stimulated to beat by the nervous system. The majority of the ciliary movement is dependent upon mechanical connections to the central portion of the swimming plate. When the cilia in the central portion of a plate were separated from the cilia at the edges, the cilia near the edges of the plate ceased to beat in register with the central cilia. The coordination of ciliary beating throughout the swimming plate depended not on the transmission of nervous impulses through the cells at the ciliary bases, which were not disrupted by the needle, but rather, upon the adhesion of cilia to one another across the swimming plate. Upon removal of the needle, the cilia in the separated portions of the plate rapidly regained their associations with the central cilia and the coordinated beating of all the
IN
CTENOPHORE
207
cilia was resumed. Whether or not the interciliary bridges were rapidly re-connected upon removal of the needle was not determined. It will be important to study the fate of the interciliary bridges before and after separation of the swimming plate cilia to determine if the bridges are absolutely required for the coordination of ciliary beating. It remains possible that the bridges may merely stabilize the interactions between adjacent cilia and are not totally required for ciliary coordination. Acknowledgements I would like to thank Dr Ray Stephens for introducing me to ctenophore cilia and to Drs Sid Tamm and S. Beroue for a number of stimulating conversations. This work was supported by Grants AM 21672 and GM 24584 from the National Institutes of Health and RR0 7037 and 3045 from the University of Kansas.
References AFZELIUS. B. A. 1961. The fine structure of the cilia from ctenophore swimming plates. J. biophys. hiochem. Cytol., 9, 383. ALLEN, R. D. 1968. A re-investigation of cross-sections of cilia. J. Cell Biol., 37, 825. ANDERSON, W. A. and ELLIS, R. A. 1965. Ultrastructure of Trypanosoma lewisi: flagellum, microtubules, and the kinetoplast. J. Protozool., 12, 483. BACETTI,B., BURRINI, A. G., DALLAI, R. and PALLINI, V. 1979. The dynein electrophoretic bands in axonemes naturally lacking the inner or the outer arm. J. CeN Bid., 80, 334. BOUCK, G. B., ROGALSKI,A. and VALAITIS,A. 1978. Surface organization and composition of Euglma. II. Flagellar mastigonemes. J. Cell Bid., 77, 805. DENTLER, W. L. 1977. Fine structural localization of phosphatases in cilia and basal bodies of Tefrahymena pyriformis.
Tissue & Cell, 9, 209.
DENTLER, W. L. 1981. Microtubule-membrane interactions in cilia and flagella. Inr. Rev. Cytol., 72, 1. DENTLER, W. L. and CUNNINGHAM, W. P. 1977. Structure and organization of radial spokes in cilia of Tetrahymena
pyriformis.
J. Morphol.,
153, 143.
DENTLER, W. L., PRATT, M. M. and STEPHENS,R. E. 1980. Microtubule-membrane interactions in cilia. II. Photochemical cross-linking of bridge structures and the identification of a membrane-associated dynein-like ATPase. J. CeN Biol., 84, 381. GIBBONS,I.R. 1965. Chemical dissection of cilia. Arch. Biol., 76, 317. MARKEY. D. R. and BOUCK, G. B. 1977. Mastigonemeattachment in 0chromonas.J. Ultrastruct. Rrs.,59, 173. PICINNI,E.,ALBERGONI, V. and COPPELLOTTI,0. 1975. ATPase activity in flagella from Euglena gracilis. Localization of the enzyme and effects of detergents. J. Protorool., 22, 33 I, SATTLER,C. A. and STAEHELIN,L. A. 1974. Ciliary membrane differentiations in Tetrahymena pyrijormis. Tetrahymena has four types of cilia. J. Cell Biol., 62, 473.
208 STEPHENS,R. E. 1970. Isolation
DENTLER
of nexin-the linkage protein responsible for maintenance of the nine-fold configuration of flagellar axonemes. Biol. Bull., 139, 438. STEPHENS,R. E. 1978. Primary structural differences among tubulin subunits from flagella, cilia, and the cytoplasm. Biochemistry, 17, 2882. TAMM, S. L. 1980. Biological Bulletin (in press). VICKERMAN,K. 1969. On the surface coat and flagellar adhesion in trypanosomes. J. Cell Sci., 5, 163. WARNER, F. D. 1976. Ciliary inter-microtubule bridges. J. Cell Sri., 20, 201. WARNER, F. D. and SATIR, P. 1974. The structural basis of ciliary bend formation. Radial spoke positional changes accompanying microtubule sliding. J. Cell Bid., 63, 35