Biology of the Cell 94 (2002) 109–115 www.elsevier.com/locate/biocell
Original article
Mass production of basal bodies in paraspermiogenesis of Tubificinae (Annelida, Oligochaeta) Marco Ferraguti *, Umberto Fascio, Silvia Boi Dipartimento di Biologia, Università di Milano, 26, via Celoria, 20133 Milano, Italy Received 5 April 2001; accepted 22 October 2001
Abstract The oligochaete annelids, belonging to the subfamily Tubificinae, produce two types of spermatozoa: eusperm (the fertilising ones) and parasperm, protecting and carrying the eusperm. The pathway for the production of the two types is common until the onset of meiosis, but then a regular meiosis produces eusperm, whereas parasperm is generated through a peculiar mechanism of nuclear fragmentation giving rise to an irregular, but very high, number of paraspermatozoa. Since every parasperm has its own flagellum, this entails the necessity of producing a very high number of basal bodies. In the present paper, we describe how basal bodies are generated through a mechanism similar to that producing the basal bodies in ciliated epithelia, but never observed up to now during the genesis of a uniflagellated cell. Basal bodies form in close proximity to a precursor structure called deuterosome, which originates de novo in the cytoplasm from fibrogranular material. The various stages of centriologenesis are positive to anti-centrin antibodies and, observed by electron microscopy, correspond closely to the ones described for ciliated epithelia. However, once formed, the basal bodies migrate to their final position and produce the parasperm flagellum. © 2002 Éditions scientifiques et médicales Elsevier SAS. All rights reserved. Keywords: Centrioles; Basal bodies; Spermatogenesis; Parasperm; Annelids
1. Introduction
assemblate in the cytoplasm (‘de novo’ formation typical of some ciliates: Grimes, 1973).
One of the major problems in modern cell biology is understanding the characteristics of the centrosome and its role in the cellular cycle. When and how centrioles are replicated is an important part of this problem. In somatic cells, centrioles duplicate, during interphase, in a semi-conservative way called the ‘kinetosomal mode’ (Pitelka, 1974) and the two daughter cells inherit, after cell division, a couple of centrioles each. Another model of centriole replication is typical of the ciliogenesis process in ciliated epithelia (Dirksen, 1991) and can produce the high number of basal bodies required (‘deuterosomal mode’, Pitelka, 1974). This process can either require the presence of a parental centrosome or be independent from this organelle and begin with an intermediate structure called deuterosome (Anderson and Brenner, 1971) or simply
In male germ cells, centrioles are involved in the flagellar basal body formation and many possible patterns have been described for their reproduction (for a review see Krioutchkova and Onishchenko, 1999). According to the principle of correlation between the nuclear ploidy and the number of centrioles, spermatocyte I should contain two centrosomes and spermatocyte II one centrosome, to give spermatozoa with one centriole to provide the embryo with a half centrosome. This rule is respected in many animal species but some have two centrioles in their mature spermatozoa. In other cases, the centrosome is duplicated in spermatocyte II but one or both of them are eliminated to give mature sperms with only one or no centriole. Moreover, in some species with internal fertilisation, centrioles lose their capability to form an axoneme. As a rule, in the germ line, centrioles are replicated following the semi-conservative model but some multiflagellate spermatozoa, as those of the fern Marsilea (Mizukami and Gall, 1966), of the termite Mastotermes darwiniensis (Baccetti and Dallai, 1978) and the paraspermatozoa of certain gastropod mollusc (Healy
* Corresponding author. Tel.: +39-02-5835-4782; fax: +39-02-58354781. E-mail address:
[email protected] (M. Ferraguti). © 2002 Éditions scientifiques et médicales Elsevier SAS. All rights reserved. PII: S 0 2 4 8 - 4 9 0 0 ( 0 2 ) 0 1 1 8 5 - 1
110
M. Ferraguti et al. / Biology of the Cell 94 (2002) 109–115
and Jamieson, 1981) are characterised by the formation of multiple centrioles through a process similar to that of ciliogenesis. Paraspermatozoa are, according to Healy and Jamieson (1981), the non-fertilising elements of a dichotomous sperm line where also fertilising sperm (eusperm) is produced. In fact, parasperm may aid, but not contribute genetically to fertilisation. A dichotomous spermatogenesis has been described in a number of invertebrate taxa (FainMaurel, 1966) among which the oligochaete annelids of the subfamily Tubificinae. Tubificine paraspermatozoa are uniflagellate cells differing from eusperm in all cytological details (Braidotti and Ferraguti, 1982), including the much lower DNA content (Ferraguti et al., 1987). Tubificine parasperm participate to the production of sperm bundles (called spermatozeugmata) which can be seen in the spermathecae after mating. In such structures, parasperm is tightly packed by a junctional complex and surrounds the parallely arranged eusperm. Only the extremities of parasperm tails are free and move forming a metachronal wave (Ferraguti et al., 1988) that carries the spermatozeugma towards the spermathecal opening. Two types of sperms are produced by different patterns of spermiohistogenesis (Boi et al., 2001). In Tubifex tubifex, the type-species of the group, cysts of four spermatogonia, released by the testes in the seminal vesicles, divide three times to give cysts with 32 primary spermatocytes. These will undergo meiotic division and finally produce 128 euspermatozoa. Paraspermatozoa, on the contrary, are produced through a peculiar mechanism of nuclear fragmentation, starting from cysts comparable to the 32-cell cysts. This irregular process gives rise to a variable, but very high number of paraspermatozoa, on average 1250 ± 900 (evaluated counting 114 paraspermatid cysts). During euspermatogenesis, the 128 basal bodies are produced by the usual semi-conservative mechanism of centriole duplication. On the contrary, in paraspermiogenesis, where a regular meiosis is suppressed, the basal body production is independent from the nuclear material division and follows a process of centriole mass replication similar to that of ciliated epithelia. This paper is devoted to the detailed description of such process. Since centrosome duplication is a subject still widely debated, we will clarify the terminology used in this paper, adopting, where not otherwise specified, the terms proposed by Dirksen (1991) in her review on ciliogenesis process. The amorphous clouds of filamentous material, reported by a large number of workers as the earliest centriole precursors in ciliogenesis, are here referred to as fibrogranular complexes. We will call condensation forms, the electrondense aggregations which are probably formed by the material of the fibrogranular complexes, and deuterosome the electron dense, roughly cylindrical aggregates which then assume the centriole structure. They are seen associated to procentrioles and centrioles.
2. Materials and methods 2.1. Electron microscopy Sexually mature individuals of T. tubifex were collected in the Lambro River in Milan and fixed in a 0.1 M cacodylate buffered paraformaldehyde–glutaraldehyde fixative (Karnovsky, 1965). Specimens of Isochaetides arenarius were collected in the sediment of the northern Lake Baikal (Russia) by Patrick Martin and fixed in a cacodylate buffered paraformaldehyde and glutaraldehyde mixture in picric acid, adjusted to freshwater (Ermak and Eakin, 1976). Tubificoides bermudae was collected in Carry Bow Cay (Belize) by Christer Erseus and fixed in a cacodylate buffered paraformaldehyde and glutaraldehyde mixture in picric acid (Ermak and Eakin, 1976). The genital segments were cut open to facilitate the penetration of the fixative. The specimens were washed overnight in 0.1 M cacodylate buffer, post-fixed in 1% osmium tetroxide in 0.1 M cacodylate buffer, washed in distilled water, prestained in 2% uranyl acetate in 70% alcohol, dehydrated in a graded ethanol series and embedded in Spurr’s resin. Ultra-thin sections, obtained with an LKB Ultrotome V, were stained in lead citrate, carbon coated, and observed under a JEOL 100 SX electron microscope operating at 80 kV. 2.2. Immunocytochemical staining Sexually mature individuals of T. tubifex were selected and placed on a slide within a drop of buffered fixative (PIPES 10 mM, KCl 50 mM, EGTA 2 mM, MgCl2 1 mM, glycerol 2 M, pH 6.9, with 2% paraformaldehyde, 0.05% TRITON-X100). The seminal vesicles of the worms were opened by piercing them with a thin needle. The worms were then removed and the slides with the content of the seminal vesicle were transferred to a moist chamber kept at 4 °C overnight. The slides were washed in phosphate buffer (PBS, pH 7.6) and incubated with bovine serum albumin (BSA 4% in PBS) for 45 min. Polyclonal antibodies antiHsCen2p in rabbit were obtained from the Institut Curie in Paris (Laoukili et al., 2000). The slides were incubated with anti-HsCen2p 1:2000 in PBS with 4% BSA, for 45 min, then incubated with anti-rabbit antibodies conjugated to TRITC (SIGMA, 1:200 in PBS) for 45 min and finally were incubated with DAPI (4,6Diamidino2phenylindole, SIGMA) 2 µg ml for 5 min and mounted in glycerol and PBS 1:1. The fluorescence images were acquired with a Leica TCS-NT confocal microscope.
3. Results We examined by electron microscopy ultra-thin sections of the seminal vesicles of three tubificine species: T. tubifex,
M. Ferraguti et al. / Biology of the Cell 94 (2002) 109–115
I. arenarius and T. bermudae. We also examined the seminal vesicles content of T. tubifex by fluorescence microscopy after staining of DNA and centrioles (Paoletti et al., 1996). In the eusperm line, centrosome appears to undergo regular kinetosomal replication cycles. Cysts at various spermatogonial stages have been observed by electron microscopy showing a couple of centrioles perpendicularly arranged. At spermatid stage, only one centriole is visible. In the parasperm line, cysts of the size of 32-cell stage undergo a nuclear fragmentation along with basal body mass production. The fragmentation process of the nuclear material has already been described in our previous work (Boi et al., 2001).The first recognisable event of basal body mass production is the presence in the cytoplasm of the fibrogranular complexes (Dirksen, 1991) (Fig. 1B). These are amorphous clouds of filamentous material with electrondense particles of various size; they are often but not always associated with a Golgi body (Fig. 1B) and often with endoplasmic reticulum. Fibrogranular complexes as well as the successive stages of centriologenesis are observed in proximity of the nucleus. At this stage, the cyst nuclei appear with a typical partially condensed chromatin (Fig. 1A) with a large nucleolus. Under fluorescence microscope, some cysts at 32-cell stage show one or more areas per cell decorated by anti-centrin antibodies (Fig. 3B). The diameter of such areas is roughly 1 µm, corresponding to the size of fibrogranular complexes and/or to the successive stages of basal body production size. In some cells, also smaller areas (Fig. 3B, arrowhead) which could probably be the original centrosomes are decorated with anti-centrin antibodies. In the same cysts or in cysts at a more advanced stage of nuclear fragmentation, large (with a diameter of roughly 0.25 µm) electron-dense aggregates generating from the fibrogranular complexes have been observed (Fig. 1C–E). These are called condensation forms and have been described in most ciliated epithelia as the following important stage in centriole formation (Dirksen, 1991). These structures are usually observed in the number of two per cell but in some cases up to five of them have been counted in the same cell (Fig. 1C,D). Often, when only two condensation forms are present in the same cell, the forming microtubules appear in the same section tangentially cut in one of them, while in the other are transversally cut (Fig. 1E), suggesting a right angle disposition of such structures. In the following stage, the condensation forms transform into a hollow cylinder of electron-dense material, called deuterosome, which gradually assumes the typical aspect of a centriole formed by nine triplets of microtubules and coated with electron-dense material (Fig. 1F). These structures are surrounded by less electron-dense procentrioles which successively transform into mature centrioles. The growing procentrioles are associated to the deuterosome by some fibrous material and are disposed radially in the space all around it (Fig. 1F). From the few longitudinal sections of
111
the deuterosome we were able to examine, it appears that in the early stages, when the deuterosome is still growing, no procentrioles are associated to the extremities of this structure (Fig. 1G). Later on, when the deuterosome reaches its final length, procentrioles appear associated only to one end of the structure (Fig. 1H). Once it reaches its final size, deuterosome maintains its original structure and size during all the procentriole formation process (Fig. 1H,I). The deuterosome has the shape of a recorder’s mouthpiece, already described in the basal bodies of Tubifex sperm (Ferraguti, 1984). In some cross-sections of procentrioles, often composed only by tubules A and B, a cartwheel structure (Pitelka, 1974) is visible connecting internally the forming triplets (Figs. 1F and 2A, arrowheads). Finally, the mature centrioles appear more detached from the central deuterosome (Figs. 1I and 2A). At the same time, nucleus fragments (Fig. 2B) and centrioles migrate to the periphery of cytophore towards the regions where the spermatid cell bodies will be formed (Figs. 2C,D and 3A).
4. Discussion In the eusperm line, since spermatid stage, only one centriole is visible meaning either that centrosome does not replicate in spermatocyte I or that one of the centrioles degenerates. We were unable to decide between the two hypotheses, even if the second one is not supported by our observations. According to our observations, in tubificine parasperm, the centrosomes are never associated to deuterosomes generating from fibrogranular complexes or to any of the various stages of basal body replication. This feature suggests that the mass basal body production of the Tubificinae follows the acentriolar pathway (Anderson and Brenner, 1971; Tournier and Bornens, 1994). In this case new centriole/basal bodies assemble in the cytoplasm without the induction of a centriole. However, the fact that centriologenesis always occurs in a specific area, in the proximity of the nucleus and of a Golgi apparatus, suggests that continuity with centrosomal structures may be related to some components present in such area (Tournier and Bornens, 1994). The centrosomal centrioles of paraspermatocyte cells could either be eliminated or become basal bodies themselves. The large areas stained by anti-centrin2p antibodies, observed in some 32-cell cysts by immunocytochemical experiments, could correspond to fibrogranular complexes and/or following stages of centriologenesis. In fact, centrin is a protein associated with centrioles (Paoletti et al., 1996), with MTOC in protozoa (Levy et al., 1996) and with precursor of basal bodies in ciliated epithelia (Laoukili et al., 2000) and the size of such areas is similar to the area visibly involved in the various stages of centriologenesis, as observed in electron micrographs. It is interesting to point
112
M. Ferraguti et al. / Biology of the Cell 94 (2002) 109–115
Fig. 1. Centriologenesis in the parasperm line of T. tubifex (A, B, C, D, E, H, I) and I. arenarius (F). (A) A low magnification image of two cysts undergoing the first stage of fragmentation. Two fibrogranular complexes are visible near the nuclei (arrows). The one at right is enlarged in (B). Note the aspect of the nuclear chromatin (n). (B) Fibrogranular complex (f) close to a Golgi complex (g). (C) A condensation form (asterisk) arising from a fibrogranular complex. (D) A group of condensation forms (at least four) arising from the same fibrogranular complex. (E) Two neighbouring condensation forms. One (bottom) is cross-cut and shows forming microtubules (arrows). A second one (top), tangentially cut, is probably perpendicular to the first one. This is the most common arrangement. (F) A transversal section of a deuterosome (bottom) surrounded by procentrioles (arrows). On top, a cartwheel (arrowhead) is visible, probably belonging to procentrioles arising from another deuterosome. (G) A longitudinal section of a growing deuterosome surrounded by procentrioles only on the sides while both ends are free. (H) A longitudinal section of a mature deuterosome surrounded by procentrioles: only one end is free. (I) A transversal section of a deuterosome surrounded by at least six mature centrioles at the beginning of their migration towards their final destination. The bar is 0.3 µm except in (A), where it is 3 µm.
M. Ferraguti et al. / Biology of the Cell 94 (2002) 109–115
113
Fig. 2. Centriologenesis in the parasperm line of T. bermudae (A), I. arenarius (B) and T. tubifex (C, D). (A). Final stage of basal body production in T. bermudae. Basal bodies show the structure of mature centrioles and are beginning the migration towards their final destination. One of them (arrowhead) displays a cartwheel structure. (B) A fragmenting nucleus during paraspermiogenesis in I. arenarius. Two mature centrioles are visible (arrowheads), probably ready to follow the forming nuclei in the paraspermatid cellular bodies. Also, two Golgi bodies are visible (g) on the other side of the nucleus. (C) A recently formed paraspermatid of T. tubifex: (b) basal body; (g) Golgi body; (n) nucleus; (r) residual chromatin in the cytophore. (D) A paraspermatid of T. tubifex at a more advanced stage. The tail (arrowhead) is formed, the two mitochondria (m) are migrating to form the midpiece, the Golgi body (g) is still in the proximity of the midpiece, the nucleus (n) is rounded and the fibrous collar (c) is clearly visible. The bar is 1 µm.
that the number of such areas per cell is extremely variable and sometimes greater than two, which would be the expected number if centriologenesis would start from each of the two centrioles of a centrosome. Deuterosomes have often been described as spherical structures, but in some cases also, a cylindrical shape has been observed (for a review see Lemullois et al., 1988). The mature deuterosomes we observed are similar in any detail to a true centriole made of particularly electron-dense material and coated with fibrous material. Moreover, they maintain their original structure during all the procentriole formation process (Fig. 1F–I); contrary to the depletion described in some ciliated epithelia (Pitelka, 1974; Dirksen 1991). The semi-conservative centriole replication process (‘kinetosomal mode’, Pitelka, 1974) is strictly associated with cellular division, whereas centriole replication by the ‘deuterosomal mode’ (Pitelka, 1974) is typical of post-mitotic cells and is unrelated to cellular division. The tubificine
parasperm line is characterised by a non-meiotic mode of nuclear division (fragmentation in Boi et al., 2001) which does not imply spindle formation. The number of nuclei formed by such fragmentation process is 10–20 times higher than in a regular meiosis (Boi et al., 2001), thus requiring a much higher number of basal bodies. The deuterosomal mode of centriole reproduction is typical of ciliated epithelia and has been observed in few cases in pluriflagellate germ cells. In fact, in the termite M. darwiniensis (Baccetti and Dallai, 1978) and in the parasperm line of some gastropod molluscs (Gall, 1961; Healy and Jamieson, 1981; Buckland-Nicks 1982; Buckland-Nicks, 1998) the basal bodies of the multiflagellate spermatozoon are produced by a similar process. In the vegetal kingdom, the male germ cells of the fern Marsilea and of the cycad Zamia are pluriflagellate spermatozoa whose basal bodies are produced by a mechanism which has some resemblance with that described in ciliated epithelia (Mizukami and Gall, 1966).
114
M. Ferraguti et al. / Biology of the Cell 94 (2002) 109–115
arenarius. Michel Bornens kindly went through the manuscript. We also thank Giulia Mazzoleni and Samuele Cerea for technical support. This work has been supported by a grant from MURST, Rome.
References
Fig. 3. Fluorescence images of paraspermiogenesis in T. tubifex. Two cysts are shown with the same magnification, stained for DNA (with DAPI, in blue) and for basal bodies (anti-centrin antibodies, in red). (A) A cyst at the end of the fragmentation process. Lumps of chromatin are visible, each associated with a basal body. (B) A cyst at the beginning of the fragmentation process: the 32 nuclei show a regular aspect, and the spots decorated with anti-centrin antibodies probably correspond to areas where basal body production is in progress. Some of the spots are visibly smaller (arrowheads). Note the difference in size between the centrin-positive areas in (B) and in (A). The bar is 10 µm.
In all these cases, however, uncoupling between centriole reproduction and nuclear division is due to the exceeding number of basal bodies required by each cell; on the contrary, in the Tubificinae parasperm line only one basal body per cell is required. In fact, in the latter case, the uncoupling is due to the non-meiotic nuclear division method adopted which does not depend from spindle formation and, presumably, from centrioles activity (Boi et al., 2001). Such a method has been selected to provide with a basal body the thousand of germ cells formed by a cyst of the parasperm line. It is interesting to notice that in T. tubifex after copulation the paraspermatozoa gather to form the cortex of the spermatozeugmata which will behave as a ciliated tissue (Ferraguti et al., 1988).
Acknowledgements We thank Nicole Bordes and Michel Bornens for useful advices and for providing us with the anti-centrin antibodies, Christer Erséus and Patrick Martin for providing us with the specimens of Tubificoides bermudae and Isochaetides
Anderson, R.G.W., Brenner, R.M., 1971. The formation of basal body (centrioles) in the rhesus monkey oviduct. J. Cell Biol. 50, 10–34. Baccetti, B., Dallai, R., 1978. The spermatozoon in the termite Mastotermes darwiniensis. J. Cell Biol. 76, 569–576. Boi, S., Fascio, U., Ferraguti, M., 2001. A nuclear fragmentation characterises paraspermiogenesis in Tubifex tubifex (Annelida, Oligochaeta). Mol. Repr. Dev. 59, 442–450. Braidotti, P., Ferraguti, M., 1982. Two sperm types in the spermatozeugmata of Tubifex tubifex (Annelida, Oligochaeta). J. Morphol. 171, 123–136. Buckland-Nicks, J., 1998. Prosobranch parasperm: sterile germ cells that promote paternity? Micron 29, 267–280. Buckland-Nicks, J., Williams, D., Chia, F., Fontaine, A., 1982. Studies on the polymorphic spermatozoa of a marine snail. I. Genesis of the apyrene sperm. Biol. Cell 44, 305–314. Dirksen, E.R., 1991. Centriole and basal body formation during ciliogenesis revisited. Biol. Cell 72, 31–38. Ermak, T.H., Eakin, R.M., 1976. Fine structure of the cerebral and pygidial ocelli in Chone ecaudata (Polychaeta; Sabellidae). J. Ultrastruct. Res. 54, 243–260. Fain-Maurel, M.A., 1966. Acquisitions récentes sur les spermatogénèses atypiques. Ann. Biol. 5, 513–564. Ferraguti, M., 1984. Slanted centriole and transient anchoring apparatus during the spermiogenesis of an oligochaete. Biol. Cell 52, 175–180. Ferraguti, M., Bernardini, G., Melone, G., Dallai, R., 1988. Structure and function of the metachronal wave in Tubifex tubifex spermatozeugmata (Annelida, Oligochaeta). J. Ultrastruct. Mol. Struct. Res. 99, 79–95. Ferraguti, M., Redi, C.A., Garagna, S., 1987. DNA amount in the double spermatogenic line of Tubifex tubifex. In: Mohri, H (Ed.), New Horizons in Sperm Cell Research. Japan Scientific Societies Press, Tokyo, pp. 357–363. Gall, J.G., 1961. A study of spermatogenesis in the snail Viviparus. J. Biophys. Biochem. Cytol. 10, 163–193. Grimes, G.W., 1973. Origin and development of kinetosomes in Oxytricha fallax. J. Cell Sci. 13, 43–53. Healy, J.M., Jamieson, B.G.M., 1981. An ultrastructural examination of developing and mature spermatozoa in Pyrazus ebeninus (Mollusca, Gastropoda, Potamididae). Zoomorphology 98, 101–119. Karnovsky, M.J., 1965. A formaldehyde–glutaraldehyde fixative of high osmolality for use in electron microscopy. J. Cell Biol. 27, 137A. Krioutchkova, M.M., Onishchenko, G.E., 1999. Structural and functional characteristic of the centrosome in gametogenesis and early embriogenesis of animals. Int. Rev. Cytol. 185, 107–156. Laoukili, J., Perret, E., Middendorp, S., Houcine, O., Guennou, C., Marano, F., Bornens, M., Tournier, F., 2000. Differential expression and cellular distribution of centrin isoforms during human ciliated cell differentiation in vitro. J. Cell Sci. 113, 1355–1364. Lemullois, M., Boisvieux, E., Laine, M., Chailley, B., Sandoz, D., 1988. Development and functions of the cytoskeleton during ciliogenesis in metazoa. Biol. Cell 63, 125–208. Levy, Y.Y., Lai, E.Y., Remillard, S.P., Heintzelman, M.B., Fulton, C., 1996. Centrin is a conserved protein that forms diverse associations with centrioles and MTOCs in Negleria and other organisms. Cell Motil. Cytoskel. 33, 298–323. Mizukami, I., Gall, J., 1966. Centriole replication II. Sperm formation in the fern. Marsilea, and in the cycad Zamia. J. Cell Biol. 29, 97–111.
M. Ferraguti et al. / Biology of the Cell 94 (2002) 109–115 Paoletti, A., Moudjou, M., Paintrand, M., Salisbury, J., Bornens, M., 1996. Most of centrin in animal cells is not centrosome-associated and centrosomal centrin is confined to the distal lumen of centrioles. J. Cell Sci. 109, 3089–3102. Pitelka, D.R., 1974. Basal bodies and root structures. In: Sleigh, M.A. (Ed.), Cilia and Flagella. Academic Press, London, pp. 437–470.
115
Tournier, F., Bornens, M., 1994. Cell cycle regulation of centrosome function. In: Hyams, J.S., Lloyd, C.W. (Eds.), Microtubules. Wiley– Liss, New York, pp. 303–324.
