Cell and Tissue Research
Cell Tiss. Res. 191, 75-82 (1978)
© by Springer-Verlag 1978
Induction of Paracrystalline Arrays by Vincristine in the Synaptic Formations of the Teleost Retina Hartwig Wolburg and Gertrud Kurz-Isler* Division of Submicroscopic Pathology and Neuropathology, Federal Republic of Germany.
University of Tiibingen, Tfibingen,
Summary. After injection of 10 Ixl 10- 3 M vincristine into the vitreous body of the eye of the rainbow trout ( S a l m o g a i r d n e r i , Richardson), paracrystalline a r r a y s in cell bodies, cell processes and presynaptic formations of the retina cells were observed. The structures resemble the paracrystalline lattice identified by several authors as microtubular protein. The paracrystals in b o t h microtubulerich cell processes and in s y n a p t i c areas, which show only a few or no microtubules, appear to be equivalent. The s y n a p t i c paracrystals are suggested to arise from b o t h soluble tubulin and synaptic vesicles, indicating a functional role of tubulin in synaptic transmission.
Key words: Paracrystalline a r r a y s - microtubules - Synapses - Retina, rainbow trout - Vincristine.
Introduction Transmission across chemical synapses is dependent u p o n the constituents of the presynaptic terminal. It is well known that membrane components, transmitters, enzymes involved in neurotransmission and microtubules are transported from the perikaryon via axoplasmic flow to the nerve ending. Investigations on the protein p a t t e r n of synaptosomes have shown a significant participation of the microtubular protein tubulin in the composition of synapses by incorporation of tubulins into synaptic membranes (Feit et al., 1971). The concept of s y n a p t i c vesicle recycling (Heuser and Reese, 1973; Fried and Blaustein, 1976) must consequently include the possibility of s y n a p t i c vesicles consisting partly of tubulin or tubulin derivatives. Send offprint requests to: Dr. HartwigWolburg and Dr. Gertrud Kurz-Isler, Division of Submicroscopic, Pathologyand Neuropathology, University of Tfibingen, LiebermeisterstraBe 8, D-7400 Tfibingen, Federal Republic of Germany * Acknowledgements. This work was supported by grant Wo 215/4 from the Deutsche Forschungsgemeinscliaft. The authors wish to thank Prof. W. Schlote for critical reading of the manuscript. Thanksare due to Mrs. B. Sabrowski for secretarial aid and to Dr. B. Boschek (University of Giessen) for correctingthe English text
0302-766X/78/0191/0075/$01.60
76
H. Wolburg and G. Kurz-Isler
The interactions between presynaptic microtubules, soluble tubulins, membraneb o u n d tubulins and other associated proteins and enzymes are of g r e a tinterest for neurotransmission; however, t h e s e processes are still largely unknown (for discussion see Nicklas et al., 1973). Based on the fact that the antimitotic drug vincristine reacts in a specific manner with microtubules inducing the formation of paracrystalline arrays, the present investigation was undertaken to elucidate the reaction of synaptic formations in fish retina after intraocular injection of vincristine.
Materials and Methods 10 gl 10-3M vincristine (Eli Lilly), dissolved in 0.65% NaCl-solution, were injected into the vitreous body ofone eye of 2 adult goldfishand 15juvenile rainbow trout. The contralateral eye injected with 10gl0.65 %NaC1 served as control. Theanimalswere decapitated 2,4, 15,24and48hafterthe injection. The retina was dissected out and fixed by immersion in 2% glutaraldehyde, buffered with 0.1 M cacodylate buffer, for 40min. After washing in buffer containing0.2M sucrose, the specimens were postfixed in 1% bufferedOsO4 for 1h, rinsed, dehydrated, stained overnightin 70%alcoholsaturated with uranyl acetate,and embedded via propylene oxide in Araldite (Ciba). Semi-and ultrathin sections were cut on aReichert OMU3 ultramicrotome. Semithin sections were stained with toluidineblue. The ultrathin sections wereplacedon copper slot grids,stainedwith leadcitrateand observed in a Siemens Elmiskop 102 electron microscope. For tilting the sections the Siemens DTL-(Double-tilt-lift-) attachment was used.
Results 2 h after vincristine injection paracrystalline a r r a y s are found in all layers of the retina. The o p t i c axons of the fiber layer are partly filled with the paracrystalline material. The ganglion cell bodies and the cell bodies of the inner nuclear layer ( I N L ) contain only a very few small aggregates. On the other hand, the horizontal cell axons, which are rich in microtubules, contain more numerous aggregates. The cell processes in the synaptic a r e a s of b o t h plexiform layers show clearly the paracrystalline lattice. In the ribbon synapses of the outer plexiform layer (OPL) small aggregates develop marginally between the pedicule or spherule membrane and the vesicle clusters. In receptor cells, the axonal portion between the cell nucleus and the s y n a p t i c terminal and also the inner segment contain microtubules, but are free of paracrystalline aggregates. 4 h after vincristine injection the paracrystals in the horizontal cell axons, in the processes of the inner plexiform layer (IPL) and the ribbon synapses of the OPL have become larger a n d / o r more numerous (Figs. 1, 2). The amount of paracrystals in the perikarya of the I N L and the ganglion cell layer has also increased. A few small paracrystalline formations are now found in the inner segments of the receptor cells. Differences between rods and cones concerning the development of the paracrystalline lattice were not observed. 15 and 24 h after vincristine injection into the eye-bulb, crystal formation has increased in the horizontal cellprocesses and in the ribbon synapses of the OPL. The horizontal cell axons, which are densely filled with microtubules u n d e r normal conditions, contain l a r g e paracrystals; the space between t h e s e paracrystals is free
Abbreviations: OPL outer plexiform layer; 1PL inner plexiform layer; p.i. after intraocular injection of 10~tl 10- 3M vincristine; arrows: possible participation of synaptic vesicles in paracrystal structure Fig. 1. Conventional synapse in the IPL, 4 h p.i. Paracrystal in hexagonal order. Tubule diameter about 30nm x 63,000 Fig. 2. Ribbon synapse in the IPL, 4 h p.i. Development of ladder-like structure; edge to edge distance between the rungs about 21 nm. × 44,000
Fig. 3. Paracrystals in OPL ribbon synapses, 24 h p.i. filling a considerable portion of the synaptic ending; the number of synaptic vesiclesis reduced. Average distance between the paracrystal elements about 34nm. × 56,000
Synaptic Tubulin Paracrystals (Teleost Retina)
79
Fig. 4. Hexagonal paracrystals in a ribbon synapse of theOPL with preserved ribbon structures, 15 h p.i. Tubule diameter about 38 nm. x 36,000 Figs.5--7. Tilting analysis o f a vincristine-induced paracrystal 4 h p.i. x 45,000 Fig.5. Tilting angles: ~ = + 15°, fl = 0°; distance between the lattice planes about 27nm Fig. 6. No tilting, horizontal position of the section in the electron microscope; distance between the lattice planes about 28 rim Fig. 7. Tilting angles: ~t= 0 °, fl = - 3 0 ° ; hexagonal pattern of the paracrystal. Tubule diameter about 31 nm
80
H. Wolburg and G. Kurz-Isler
o f microtubules. The synaptic terminals of the receptor cells are often occupied by paracrystalline lattice, and only a few synaptic vesicles are found in this region (Fig. 3). The structure of the synaptic ribbon does not show alterations (Fig. 4). In the IPL an enlargement and discharge of synaptic profiles can be observed. Both paracrystalline arrays and synaptic vesicles have decreased in number and many presynaptic endings are nearly empty. This may depend on a general degenerative vacuolization (Koniszewski et al., 1976) and not on a specific tubulin-vincristin reaction. The fine structure of the vincristine-induced aggregates is uniform in all cell types and layers of the retina, corresponding to the images published by Bensch and Malawista (1969), Schlaepfer (1971), Stebbings (1975), Tomlinson and Bennett (1976) and Chan and Bunt (1978). The differences in the periodicity of the paracrystalline structures result from a differing projection of the same lattice, as can be shown by tilting analysis (Figs. 5-7). It varies between 24 nm and 32 nm and can be explained in the same manner as demonstrated schematically by Bensch and Malawista (1969). Sometimes crystal formations o f ladder-like appearance are found; the distance between the rungs is approximately 21 nm. In some presynaptic terminals synaptic vesicles seem to be involved in the formation o f the paracrystals (Figs. 2, 3), whereas in the cell bodies or cell processes the microtubules are the only source o f vincristine-induced paracrystals. In any case, fully developed paracrystalline aggregates in synapses and cytoplasm display similar fine structure.
Discussion
Formation of crystalline structures was induced by vincristine in presynaptic endings o f the fish retina, where only a few or no microtubules are seen. Vincristine and vinblastine are known to bind to microtubular proteins. The mode of formation o f the paracrystalline arrays is not absolutely clear, but in such binding reactions mostly soluble tubulins are organized into crystalline form (Bhattacharyya and Wolff, 1976a). According to the concentration o f vincristine (approximately 3" 10-4M) used in this study, a specific crystal formation from tubulins and not from other proteins can be assumed (Bhattacharyya and Wolff, 1976a).Also, the formation o f a structurally equivalent lattice in the horizontal cell axons, which are normally filled with microtubules, is an indirect argument for the tubulin nature of synaptic paracrystals. Further, the ladder-like lattice demonstrated by Bensch et al. (1969) from purified microtubular protein after treatment with vinblastine in vitro (edge to edge distance in the lattice is about 20 nm) resembles the ladder-like aggregates shown in this report (Fig. 2; edge to edge distance between the ladder rungs is about 21 nm). Nevertheless, for the definite identification of the lattice as a tubulin aggregate, biochemical and diffraction analyses are needed. Microtubules can be visualized in presynaptic terminalsby the albumin method (Gray, 1975, 1976; Bird, 1976) or by a slight modification of conventional electron microscopic methods (Glees and Spoerri, 1977). The albuminmethod is not always successful in enhancing the visualization of microtubules in presynaptic endings (Livingston, 1977). Intact microtubules in retinal synapses extending to the synaptic
Synaptic Tubulin Paracrystals (Teleost Retina)
81
ending disappear before they protrude between the synaptic vesicles (see the description of disassembly of the microtubule cytoskeleton in the presynaptic area given by Lasek and Hoffmann, 1976). On the other hand, in a very recent paper (Chan and Bunt, 1978), microtubules were shown in synaptosomes prepared from rat cerebral cortex. These microtubules are transformed after incubation o f the synaptosomes with vinblastine sulfate into the typical paracrystalline lattice, but the synaptic vesicles are not involved in this transformation. However, the large lattices found especially in the ribbon synapses of the OPL cannot be explained without involvement of the synaptic vesicles, as they grow progressively, displacing or replacing the synaptic vesicles. The number of synaptic vesicles decreases during paracrystal formation. Transmitter release or leakage induced by interactions between the vinca alkaloids and the synaptic membrane must be considered (Nicklas et al., 1973), but this process is not linked with membrane loss. Thus, from a morphological point of view (Figs. 2, 3), it can be suggested that altered or degenerated synaptic vesicles are also capable o f producing lattices, as has been suggested by Bunt (1973) for paracrystals formed in the optic terminals o f the superior colliculus in the rabbit after intraocular injection o f vinblastine. Recent results support the hypothesis that the membrane of synaptic vesicles participates in crystal formation and thus speak in favor of a functional role o f tubulin in synaptic transmission. If tubulin is a component of neuronal membranes, as has been found by Bhattacharyya and Wolff(1976 b) in the guinea pig brain, this may hold true also for the synaptic membranes including the membranes of synaptic vesicles (see also Gray, 1975; Bird, 1976; Westrum and Gray, 1976). The tubulin nature of proteins f o u n d in synaptic vesicle fractions by Morgan et al. (1973) must be proven by application of adequate methods. Phosphatidylinositol phosphodiesterase is associated with rat brain tubulin (Quinn, 1973), and on the other hand, phosphatidylinositol is a component o f synaptic vesicle membranes (Hawthorne, 1977). The last author reported on a n unpublished result o f Lagnado, who has recently found binding activity between tubulin and membranous phosphatidylinositol. It is therefore possible that vincristine may react with membranes of the synaptic vesicles, resulting in paracrystalline formations equivalent to those developed from pure or purified tubulin (Bensch et al., 1969). Chemical analysis o f paracrystals in cellular elements o f the retina and binding studies using synaptic vesicle membranes and vinca alkaloids are needed t o clarify the possible functional role of membrane-bound tubulin in retinal synaptic transmission.
References
Bensch, K.G., Malawista, S.E.: Microtubular crystals in mammalian cells. J.CellBiol. 40, 95-107 (1969) Bensch, K.G., Marantz, R., Wisniewski, H., Shelanski, M.: Introduction in vitro of microtubular crystals by vinca alkaloids. Science 165, 495-496 (1969) Bhattacharyya, B., Wolff, J.: Tubulin aggregation and disaggregation - mediation by 2 distinct vinblastine-binding sites. Proc. nat. Acad. Sci. (Wash.) 73, 2375-2378 (1976a) Bhattacharyya, B., Wolff, J.: Polymerisation of membrane tubulin. Nature (Lond.) 264, 576-577 (1976b)
82
H. Wolburg and G. Kurz-Isler
Bird, M.M. : Microtubule-synaptic vesicleassociations in cultured rat spinal cord neurons. Cell Tiss. Res. 168, 101-115 (1976) Bunt, A.H.: Paracrystalline inclusion in optic nerve terminals following intraocular injection of vinblastine. Brain Res. 53, 29-39 (1973) Chan, K.Y., Bunt, A.H.: An association between mitochondria and microtubules in synaptosomes and axon terminals of cerebral cortex. J. Neurocytol. 7, 137-143 (1978) Feit, H., Dutton, G.R., Barondes, S.H., Shelanski, M.L.: Microtubule protein: identification in and transport to nerve endings. J. Cell Biol. 51, 138-147 (1971) Fried, R.C., Blaustein, M.P.: Synaptic vesicles recycling in synaptosomes in vitro. Nature (Lond.) 261, 255-256 (1976) Glees, P., Spoerri, P.E." Microtubule-vesicle-ribbon associations in the monkey retina. J. Neurocytol. 6, 353-354 (1977) Gray, E.G.: Presynaptic microtubules and their association with synapticvesicles.Proc. roy. Soc. B 190, 369-372 (1975) Gray, E.G.: Microtubules in synapses of the retina. J. Neurocytol. 5, 361-370 (1976) Hawthorne, J.N.: Phospholipid metabolism in some excitable biological membranes. In: Structure of biological membranes (S. Abrahamsson, J. Pascher, eds.). New York-London: Nobel Foundation Symposium, Plenum Press 1977 Heuser, J.E., Reese,T.S.: Evidencefor recycling of synaptic vesicle membrane during transmitter release at the frog neuromuscular junction. J. Cell Biol. 57, 315-344 (1973) Koniszewski, G., Rix, G., Brunner, P.: Ahistological and electron microscopic study ofthe rabbit retina after application of vincristine. Albrecht v. Graefes Arch. klin. exp. Ophthal. 199, 147-156 (1976) Lasek, R.J., Hoffmann, P.N.: The neuronal cytoskeleton, axonal transport and axonal growth. In: Cell motility, Book C. (R. Goldman, T. Pollard, J. Rosenbaum, eds.). Cold Spring Harbor Conf. on Cell Proliferation, Vol. 3, 1021-1049 (1976) Livingston, A.: Microtubules in the neurosecretory neurones of the posterior pituitary of the rat. Cell Tiss. Res. 180, 253-261 (1977) Morgan, I.G., Zanetta, J.-P., Breckenridge, W.C., Vincendon, G., Gombos, G.: The chemical structure o f synaptic membranes. Brain Res. 62, 405-411 (1973) Nicklas, W.J., Puszkin, S., Bed, S.: Effect ofvinblastine and colchicine on uptake and release ofputative transmitters by synaptosomes and on brain actomyosin-like protein. J. Neurochem. 20, 109-122 (1973) Quinn, P.L: The association between phosphatidylinositol phosphodiesterase activity and a specific subunit o f microtubular protein in rat brain. Biochem. J. 133, 273-281 (1973) Schlaepfer, W.W.: Vincristine-induced axonal alterations in rat peripheral nerve. J. Neuropath. exp. Neurol. 30, 488-505 (1971) Stebbings, H.: The role of microtubules in the assembly of vinblastine-induced crystals. Cell Tiss. Res. 159, 141-145 (1975) Tomlinson, D.R., Bennett, T.: An ultrastructural examination of the action of vinblastine on microtubules, neurofilaments and muscle filaments in vitro. Cell Tiss. Res. 166, 41~420 (1976) Westrum, L.E., Gray, E.G.: Microtubules and membrane specializations. Brain Res. 105, 547-550 (1976)
Accepted April 14, 1978
Cell Tiss. Res. 191, 75-82 (1978)
© by Springer-Verlag 1978
Induction of Paracrystalline Arrays by Vincristine in the Synaptic Formations of the Teleost Retina Hartwig Wolburg and Gertrud Kurz-Isler* Division of Submicroscopic Pathology and Neuropathology, Federal Republic of Germany.
University of Tiibingen, Tfibingen,
Summary. After injection of 10 Ixl 10- 3 M vincristine into the vitreous body of the eye of the rainbow trout ( S a l m o g a i r d n e r i , Richardson), paracrystalline a r r a y s in cell bodies, cell processes and presynaptic formations of the retina cells were observed. The structures resemble the paracrystalline lattice identified by several authors as microtubular protein. The paracrystals in b o t h microtubulerich cell processes and in s y n a p t i c areas, which show only a few or no microtubules, appear to be equivalent. The s y n a p t i c paracrystals are suggested to arise from b o t h soluble tubulin and synaptic vesicles, indicating a functional role of tubulin in synaptic transmission.
Key words: Paracrystalline a r r a y s - microtubules - Synapses - Retina, rainbow trout - Vincristine.
Introduction Transmission across chemical synapses is dependent u p o n the constituents of the presynaptic terminal. It is well known that membrane components, transmitters, enzymes involved in neurotransmission and microtubules are transported from the perikaryon via axoplasmic flow to the nerve ending. Investigations on the protein p a t t e r n of synaptosomes have shown a significant participation of the microtubular protein tubulin in the composition of synapses by incorporation of tubulins into synaptic membranes (Feit et al., 1971). The concept of s y n a p t i c vesicle recycling (Heuser and Reese, 1973; Fried and Blaustein, 1976) must consequently include the possibility of s y n a p t i c vesicles consisting partly of tubulin or tubulin derivatives. Send offprint requests to: Dr. HartwigWolburg and Dr. Gertrud Kurz-Isler, Division of Submicroscopic, Pathologyand Neuropathology, University of Tfibingen, LiebermeisterstraBe 8, D-7400 Tfibingen, Federal Republic of Germany * Acknowledgements. This work was supported by grant Wo 215/4 from the Deutsche Forschungsgemeinscliaft. The authors wish to thank Prof. W. Schlote for critical reading of the manuscript. Thanksare due to Mrs. B. Sabrowski for secretarial aid and to Dr. B. Boschek (University of Giessen) for correctingthe English text
0302-766X/78/0191/0075/$01.60
76
H. Wolburg and G. Kurz-Isler
The interactions between presynaptic microtubules, soluble tubulins, membraneb o u n d tubulins and other associated proteins and enzymes are of g r e a tinterest for neurotransmission; however, t h e s e processes are still largely unknown (for discussion see Nicklas et al., 1973). Based on the fact that the antimitotic drug vincristine reacts in a specific manner with microtubules inducing the formation of paracrystalline arrays, the present investigation was undertaken to elucidate the reaction of synaptic formations in fish retina after intraocular injection of vincristine.
Materials and Methods 10 gl 10-3M vincristine (Eli Lilly), dissolved in 0.65% NaCl-solution, were injected into the vitreous body ofone eye of 2 adult goldfishand 15juvenile rainbow trout. The contralateral eye injected with 10gl0.65 %NaC1 served as control. Theanimalswere decapitated 2,4, 15,24and48hafterthe injection. The retina was dissected out and fixed by immersion in 2% glutaraldehyde, buffered with 0.1 M cacodylate buffer, for 40min. After washing in buffer containing0.2M sucrose, the specimens were postfixed in 1% bufferedOsO4 for 1h, rinsed, dehydrated, stained overnightin 70%alcoholsaturated with uranyl acetate,and embedded via propylene oxide in Araldite (Ciba). Semi-and ultrathin sections were cut on aReichert OMU3 ultramicrotome. Semithin sections were stained with toluidineblue. The ultrathin sections wereplacedon copper slot grids,stainedwith leadcitrateand observed in a Siemens Elmiskop 102 electron microscope. For tilting the sections the Siemens DTL-(Double-tilt-lift-) attachment was used.
Results 2 h after vincristine injection paracrystalline a r r a y s are found in all layers of the retina. The o p t i c axons of the fiber layer are partly filled with the paracrystalline material. The ganglion cell bodies and the cell bodies of the inner nuclear layer ( I N L ) contain only a very few small aggregates. On the other hand, the horizontal cell axons, which are rich in microtubules, contain more numerous aggregates. The cell processes in the synaptic a r e a s of b o t h plexiform layers show clearly the paracrystalline lattice. In the ribbon synapses of the outer plexiform layer (OPL) small aggregates develop marginally between the pedicule or spherule membrane and the vesicle clusters. In receptor cells, the axonal portion between the cell nucleus and the s y n a p t i c terminal and also the inner segment contain microtubules, but are free of paracrystalline aggregates. 4 h after vincristine injection the paracrystals in the horizontal cell axons, in the processes of the inner plexiform layer (IPL) and the ribbon synapses of the OPL have become larger a n d / o r more numerous (Figs. 1, 2). The amount of paracrystals in the perikarya of the I N L and the ganglion cell layer has also increased. A few small paracrystalline formations are now found in the inner segments of the receptor cells. Differences between rods and cones concerning the development of the paracrystalline lattice were not observed. 15 and 24 h after vincristine injection into the eye-bulb, crystal formation has increased in the horizontal cellprocesses and in the ribbon synapses of the OPL. The horizontal cell axons, which are densely filled with microtubules u n d e r normal conditions, contain l a r g e paracrystals; the space between t h e s e paracrystals is free
Abbreviations: OPL outer plexiform layer; 1PL inner plexiform layer; p.i. after intraocular injection of 10~tl 10- 3M vincristine; arrows: possible participation of synaptic vesicles in paracrystal structure Fig. 1. Conventional synapse in the IPL, 4 h p.i. Paracrystal in hexagonal order. Tubule diameter about 30nm x 63,000 Fig. 2. Ribbon synapse in the IPL, 4 h p.i. Development of ladder-like structure; edge to edge distance between the rungs about 21 nm. × 44,000
Fig. 3. Paracrystals in OPL ribbon synapses, 24 h p.i. filling a considerable portion of the synaptic ending; the number of synaptic vesiclesis reduced. Average distance between the paracrystal elements about 34nm. × 56,000
Synaptic Tubulin Paracrystals (Teleost Retina)
79
Fig. 4. Hexagonal paracrystals in a ribbon synapse of theOPL with preserved ribbon structures, 15 h p.i. Tubule diameter about 38 nm. x 36,000 Figs.5--7. Tilting analysis o f a vincristine-induced paracrystal 4 h p.i. x 45,000 Fig.5. Tilting angles: ~ = + 15°, fl = 0°; distance between the lattice planes about 27nm Fig. 6. No tilting, horizontal position of the section in the electron microscope; distance between the lattice planes about 28 rim Fig. 7. Tilting angles: ~t= 0 °, fl = - 3 0 ° ; hexagonal pattern of the paracrystal. Tubule diameter about 31 nm
80
H. Wolburg and G. Kurz-Isler
o f microtubules. The synaptic terminals of the receptor cells are often occupied by paracrystalline lattice, and only a few synaptic vesicles are found in this region (Fig. 3). The structure of the synaptic ribbon does not show alterations (Fig. 4). In the IPL an enlargement and discharge of synaptic profiles can be observed. Both paracrystalline arrays and synaptic vesicles have decreased in number and many presynaptic endings are nearly empty. This may depend on a general degenerative vacuolization (Koniszewski et al., 1976) and not on a specific tubulin-vincristin reaction. The fine structure of the vincristine-induced aggregates is uniform in all cell types and layers of the retina, corresponding to the images published by Bensch and Malawista (1969), Schlaepfer (1971), Stebbings (1975), Tomlinson and Bennett (1976) and Chan and Bunt (1978). The differences in the periodicity of the paracrystalline structures result from a differing projection of the same lattice, as can be shown by tilting analysis (Figs. 5-7). It varies between 24 nm and 32 nm and can be explained in the same manner as demonstrated schematically by Bensch and Malawista (1969). Sometimes crystal formations o f ladder-like appearance are found; the distance between the rungs is approximately 21 nm. In some presynaptic terminals synaptic vesicles seem to be involved in the formation o f the paracrystals (Figs. 2, 3), whereas in the cell bodies or cell processes the microtubules are the only source o f vincristine-induced paracrystals. In any case, fully developed paracrystalline aggregates in synapses and cytoplasm display similar fine structure.
Discussion
Formation of crystalline structures was induced by vincristine in presynaptic endings o f the fish retina, where only a few or no microtubules are seen. Vincristine and vinblastine are known to bind to microtubular proteins. The mode of formation o f the paracrystalline arrays is not absolutely clear, but in such binding reactions mostly soluble tubulins are organized into crystalline form (Bhattacharyya and Wolff, 1976a). According to the concentration o f vincristine (approximately 3" 10-4M) used in this study, a specific crystal formation from tubulins and not from other proteins can be assumed (Bhattacharyya and Wolff, 1976a).Also, the formation o f a structurally equivalent lattice in the horizontal cell axons, which are normally filled with microtubules, is an indirect argument for the tubulin nature of synaptic paracrystals. Further, the ladder-like lattice demonstrated by Bensch et al. (1969) from purified microtubular protein after treatment with vinblastine in vitro (edge to edge distance in the lattice is about 20 nm) resembles the ladder-like aggregates shown in this report (Fig. 2; edge to edge distance between the ladder rungs is about 21 nm). Nevertheless, for the definite identification of the lattice as a tubulin aggregate, biochemical and diffraction analyses are needed. Microtubules can be visualized in presynaptic terminalsby the albumin method (Gray, 1975, 1976; Bird, 1976) or by a slight modification of conventional electron microscopic methods (Glees and Spoerri, 1977). The albuminmethod is not always successful in enhancing the visualization of microtubules in presynaptic endings (Livingston, 1977). Intact microtubules in retinal synapses extending to the synaptic
Synaptic Tubulin Paracrystals (Teleost Retina)
81
ending disappear before they protrude between the synaptic vesicles (see the description of disassembly of the microtubule cytoskeleton in the presynaptic area given by Lasek and Hoffmann, 1976). On the other hand, in a very recent paper (Chan and Bunt, 1978), microtubules were shown in synaptosomes prepared from rat cerebral cortex. These microtubules are transformed after incubation o f the synaptosomes with vinblastine sulfate into the typical paracrystalline lattice, but the synaptic vesicles are not involved in this transformation. However, the large lattices found especially in the ribbon synapses of the OPL cannot be explained without involvement of the synaptic vesicles, as they grow progressively, displacing or replacing the synaptic vesicles. The number of synaptic vesicles decreases during paracrystal formation. Transmitter release or leakage induced by interactions between the vinca alkaloids and the synaptic membrane must be considered (Nicklas et al., 1973), but this process is not linked with membrane loss. Thus, from a morphological point of view (Figs. 2, 3), it can be suggested that altered or degenerated synaptic vesicles are also capable o f producing lattices, as has been suggested by Bunt (1973) for paracrystals formed in the optic terminals o f the superior colliculus in the rabbit after intraocular injection o f vinblastine. Recent results support the hypothesis that the membrane of synaptic vesicles participates in crystal formation and thus speak in favor of a functional role o f tubulin in synaptic transmission. If tubulin is a component of neuronal membranes, as has been found by Bhattacharyya and Wolff(1976 b) in the guinea pig brain, this may hold true also for the synaptic membranes including the membranes of synaptic vesicles (see also Gray, 1975; Bird, 1976; Westrum and Gray, 1976). The tubulin nature of proteins f o u n d in synaptic vesicle fractions by Morgan et al. (1973) must be proven by application of adequate methods. Phosphatidylinositol phosphodiesterase is associated with rat brain tubulin (Quinn, 1973), and on the other hand, phosphatidylinositol is a component o f synaptic vesicle membranes (Hawthorne, 1977). The last author reported on a n unpublished result o f Lagnado, who has recently found binding activity between tubulin and membranous phosphatidylinositol. It is therefore possible that vincristine may react with membranes of the synaptic vesicles, resulting in paracrystalline formations equivalent to those developed from pure or purified tubulin (Bensch et al., 1969). Chemical analysis o f paracrystals in cellular elements o f the retina and binding studies using synaptic vesicle membranes and vinca alkaloids are needed t o clarify the possible functional role of membrane-bound tubulin in retinal synaptic transmission.
References
Bensch, K.G., Malawista, S.E.: Microtubular crystals in mammalian cells. J.CellBiol. 40, 95-107 (1969) Bensch, K.G., Marantz, R., Wisniewski, H., Shelanski, M.: Introduction in vitro of microtubular crystals by vinca alkaloids. Science 165, 495-496 (1969) Bhattacharyya, B., Wolff, J.: Tubulin aggregation and disaggregation - mediation by 2 distinct vinblastine-binding sites. Proc. nat. Acad. Sci. (Wash.) 73, 2375-2378 (1976a) Bhattacharyya, B., Wolff, J.: Polymerisation of membrane tubulin. Nature (Lond.) 264, 576-577 (1976b)
82
H. Wolburg and G. Kurz-Isler
Bird, M.M. : Microtubule-synaptic vesicleassociations in cultured rat spinal cord neurons. Cell Tiss. Res. 168, 101-115 (1976) Bunt, A.H.: Paracrystalline inclusion in optic nerve terminals following intraocular injection of vinblastine. Brain Res. 53, 29-39 (1973) Chan, K.Y., Bunt, A.H.: An association between mitochondria and microtubules in synaptosomes and axon terminals of cerebral cortex. J. Neurocytol. 7, 137-143 (1978) Feit, H., Dutton, G.R., Barondes, S.H., Shelanski, M.L.: Microtubule protein: identification in and transport to nerve endings. J. Cell Biol. 51, 138-147 (1971) Fried, R.C., Blaustein, M.P.: Synaptic vesicles recycling in synaptosomes in vitro. Nature (Lond.) 261, 255-256 (1976) Glees, P., Spoerri, P.E." Microtubule-vesicle-ribbon associations in the monkey retina. J. Neurocytol. 6, 353-354 (1977) Gray, E.G.: Presynaptic microtubules and their association with synapticvesicles.Proc. roy. Soc. B 190, 369-372 (1975) Gray, E.G.: Microtubules in synapses of the retina. J. Neurocytol. 5, 361-370 (1976) Hawthorne, J.N.: Phospholipid metabolism in some excitable biological membranes. In: Structure of biological membranes (S. Abrahamsson, J. Pascher, eds.). New York-London: Nobel Foundation Symposium, Plenum Press 1977 Heuser, J.E., Reese,T.S.: Evidencefor recycling of synaptic vesicle membrane during transmitter release at the frog neuromuscular junction. J. Cell Biol. 57, 315-344 (1973) Koniszewski, G., Rix, G., Brunner, P.: Ahistological and electron microscopic study ofthe rabbit retina after application of vincristine. Albrecht v. Graefes Arch. klin. exp. Ophthal. 199, 147-156 (1976) Lasek, R.J., Hoffmann, P.N.: The neuronal cytoskeleton, axonal transport and axonal growth. In: Cell motility, Book C. (R. Goldman, T. Pollard, J. Rosenbaum, eds.). Cold Spring Harbor Conf. on Cell Proliferation, Vol. 3, 1021-1049 (1976) Livingston, A.: Microtubules in the neurosecretory neurones of the posterior pituitary of the rat. Cell Tiss. Res. 180, 253-261 (1977) Morgan, I.G., Zanetta, J.-P., Breckenridge, W.C., Vincendon, G., Gombos, G.: The chemical structure o f synaptic membranes. Brain Res. 62, 405-411 (1973) Nicklas, W.J., Puszkin, S., Bed, S.: Effect ofvinblastine and colchicine on uptake and release ofputative transmitters by synaptosomes and on brain actomyosin-like protein. J. Neurochem. 20, 109-122 (1973) Quinn, P.L: The association between phosphatidylinositol phosphodiesterase activity and a specific subunit o f microtubular protein in rat brain. Biochem. J. 133, 273-281 (1973) Schlaepfer, W.W.: Vincristine-induced axonal alterations in rat peripheral nerve. J. Neuropath. exp. Neurol. 30, 488-505 (1971) Stebbings, H.: The role of microtubules in the assembly of vinblastine-induced crystals. Cell Tiss. Res. 159, 141-145 (1975) Tomlinson, D.R., Bennett, T.: An ultrastructural examination of the action of vinblastine on microtubules, neurofilaments and muscle filaments in vitro. Cell Tiss. Res. 166, 41~420 (1976) Westrum, L.E., Gray, E.G.: Microtubules and membrane specializations. Brain Res. 105, 547-550 (1976)
Accepted April 14, 1978
