A scanning electron-microscopic study of the local degeneration of cilia during sexual reproduction in Paramecium

In Paramecium and other ciliates, interesting studies of inheritance and morphogenesis of cortical pattern have been possible because of local differentiation in structure and function of the cell surface (Sonneborn, 1963, 1970a). One notable case is the set of events in conjugation restricted to the ventral surface of the cell. In the conjugation process of Paramecium, cell contact is established in 3 steps: mating reaction, holdfast union, and paroral union. In the mating reaction, cells of complementary mating type stick together by ‘mating reactive’ cilia located on the ventral surface of the cell (Hiwatashi, 1961). As the mating reaction proceeds, cilia and trichocysts at the anterior tip and on the ventral surface just behind the tip disappear, and pairs of cells unite at the antero-ventral surface in a ‘holdfast union’. Cilia continue to disappear on the ventral surface, and the cells unite more firmly, especially in a region just posterior to the mouth, in a ‘paroral union’. Details of the holdfast and paroral unions may be found in Wichterman (1946), Hiwatashi (1955 b), Vivier & André (1961) and Miyake (1966). Hiwatashi (1961) suggested that there is some relation between the contact region of the mating reaction and that of the 56 holdfast union. Moreover, the works of Hiwatashi (1955b) and Miyake (1966, 1968 a, b) strongly indicate that the loss of cilia during conjugation is essential not only for the formation of holdfast and paroral union but also for nuclear activation.

Thus the loss of cilia is an important phenomenon, not only from the viewpoint of morphogenetic problems but also from that of fertilization (Miyake, 1974). However, the exact details and controlling mechanisms of the degeneration of cilia have never been reported. To clarify these points, scanning electron-microscopic observations were made on cells undergoing conjugation and autogamy.

Stocks Kt, dKKi4a, 27aG3, dNi4a and d12 –3–4 of Paramecium caudatum, syngen 3, were used throughout the study. For conjugation via mating type, the cells were cultured with lettuce juice medium (Hiwatashi, 1968) in which 1 vol. of fresh lettuce juice was diluted with 40 vol. of Dryl’s solution (Dryl, 1959) and bacterized with Klebsiella aerogenes. For chemical induction of autogamy, the cells were cultured with a Ca-poor medium in which lettuce juice was diluted in 2 rπM sodium phosphate buffer (pH 7·0) instead of Dryl’s solution. For comparison, conjugating cells of P. tetraurelia (stock 51), P. multimicronucleatum (CH-312 and CH-313? P. bursaria (TK-1 and TK-3) and P. trichium (PC-2 and PC-5) were used. Cells of P. tetraurelia, stock 51, were also used for studies of natural autogamy.

Mating-reactive cells of complementary mating types were mixed in depression slides. The cells agglutinated soon after mixing, and holdfast pairs were formed about 60 min after the beginning of the mating reaction. Although the exact time of the onset of paroral union was uncertain, firmly united pairs which had undergone paroral union were present after 90 min. Such pairs were identified when drawing them into and expelling them from a pipette failed to separate the 2 cells. Cells of P. caudatum were fixed at various times from 15 min to 5 h after onset of the mating reaction. The cells of other Paramecium species were fixed at the time of holdfast pair formation.

Strength of mating reactivity was assessed by noting if many clumps are formed quickly after mixing mating types (‘strong’ or ‘high’ reactivity) or if few cells form clumps soon after mixing (‘weak’ or ‘low’ reactivity). No method is currently available for determining the amount of mating substance on each cell.

Chemical induction of autogamy was performed by a modification (Tsukii, in preparation) of the method of Miyake (1968a, b). The cells of a single mating type of P. caudatum were cultured in Ca-poor medium and washed once in a modified Miyake’s (1958) physiological balanced solution called K-PBS-II (1·5 MM NaCl, 1·8mMKCl, 0·1mM MgCl₂ 0·01mM CaCl₂, 1·8 MM KH₂PO₄ and 0·2 MM K₂HPO₄, pH 6·0). Then the cells were treated with the autogamy-inducing medium (6 rπM KC1, 50 MM methyl urea, and 40–80 μg/ml ficin or 5–10 μg/ml papain in K-PBS-II). Partially purified ficin was prepared from crude ficin (Wako Pure Chemicals Co., Ltd.) according to the method of Hammond & Gutfreund (1959). Papain used was a crystalline preparation (Sigma, 2 × crystallized). When the cells were treated with the autogamy-inducing medium, neither agglutination nor pair formation was observed. At 20 h after the beginning of induction of autogamy, the occurrence of autogamy was ascertained by looking for macronuclear fragmentation, a characteristic of sexual reproduction. Details of the chemical induction of autogamy in P. caudatum will be described elsewhere (Tsukii, in preparation).

Natural autogamy was induced in P. tetraurelia by starving sufficiently old cells (Sonneborn, 1970b) at 27 °C. It is difficult to induce autogamy synchronously or to distinguish cells undergoing autogamy by external appearance. Therefore, to obtain a population of cells in various stages of autogamy including early stages, cells were fixed when 10% of the cells showed macronuclear fragmentation. In unfixed controls, the proportion of cells with fragmented macronuclei rose to 50 % 7 h after this time of fixation. In conjugation of P. aurelia, separation of conjugants and macronuclear fragmentation occur 6—7 h after the initiation of conjugation (Jurand & Selman, 1969).

Cells were fixed in Párducz solution (Párducz, 1967) for 30 min at room temperature. After washing in deionized water, the cells were dehydrated in a series of ethanol and isoamyl acetate. The cells in isoamyl acetate were put on coverglasses or aluminium disks and airdried. The specimens were coated with gold and examined with an Hitach-Akashi MSM-4 scanning electron microscope.

The cortical structure of Paramecium has been described by many investigators using the silver impregnation technique. The ventral morphology of the vegetative cell will be described briefly. The most characteristic organelle of the ventral surface is the mouth. The position of the opening of the mouth is somewhat different from species to species. In P. caudatum, the mouth is located slightly posterior to the centre of the cell. The suture runs longitudinally across the mouth. Therefore, the ventral surface of Paramecium is divided into 4 parts by the suture and the mouth: anterior-right, anterior-left, posterior-right and posterior-left (Figs. 1, 2). The anterior-left corresponds to the oral groove, which extends from the anterior end to the vestibule of the mouth and is apparent in living cells.

The location of degenerating cilia and the order of their disappearance was examined in cells of P. caudatum undergoing conjugation after the mating reaction. At 15 min after the beginning of the mating reaction, no ciliary degeneration was observed. At 30 min, when cells are still in the mating clumps, small numbers of cells show short cilia or no cilia at their anterior tips (Fig. 3). Ciliary degeneration proceeds posteriorly along the suture at about the time of holdfast pair formation. At this time, some cells have short cilia of various lengths (Figs. 4, 5). This suggests that the degeneration of cilia is due to resorption. The degeneration of cilia extends to the sides and to the posterior end during the time of formation of holdfast to paroral unions (Figs. 6, 7). At the posterior part, loss of cilia begins near the posterior-right of the vestibule of the mouth (Fig. 8), and spreads to the posterior tip before formation of tight paroral union (Fig. 9). At the anterior part of the cell, the cilia were lost on both sides of the anterior suture. On the contrary, at the posterior part, ciliary degeneration was mainly restricted to the right side of the suture (Figs. 6–9).

The locations of ciliary degeneration during the early process of conjugation via mating type were essentially identical in all species of Paramecium examined (Figs. 10–13). In chemical induction of autogamy, which will be described later, the ciliary degeneration occurred in the same order and at the same location as in the usual conjugation. Thus the order and location of ciliary degeneration during conjugation is strikingly similar among species and among methods of induction. However, the rate of degeneration of cilia was different from sample to sample. Generally, if the cells have strong mating reactivity (see Methods), ciliary degeneration spreads more rapidly. As a result, when mating reactivity is strong the formation of normal pairs is accompanied by the formation of aberrant holdfast unions consisting of more than 3 cells (Fig. 14).

The extent of ciliary degeneration was examined in firmly united conjugation pairs. At 3 h, when pairs are united firmly, no bald free surface was observed except for the last cell of multiple unions (Fig. 15), or for ventro-posterior tip of normal pairs (Figs. 16, 17). No loss of cilia was observed on the dorsal surfaces and antero-ventral part of any pair (Fig. 16), except for a few pairs which were found after a strong mating reaction (Figs. 18, 19). These observations indicate that ciliary degeneration occurs only on the ventral surface, including the region where the cells make contact. For further study of the extent of ciliary degeneration, autogamous cells were examined, because in these there is no cell contact to hinder observation. Observations were made on more than 100 cells of P. tetr aurelia which were presumed to contain more than 50 natural autogamous cells. Ciliary degeneration was not observed on any cell. Since natural autogamy does not occur in other species of Paramecium, cells of P. caudatum undergoing chemically induced autogamy were next studied. Surprisingly, cells underwent ciliary degeneration. In order to exclude the possibility that the chemicals induced the degeneration of cilia irrespective of mating ability of the cell, immature cells, which had no ability to mate, were exposed to the same induction medium. Neither ciliary degeneration nor macronuclear fragmentation occurred. When the mating reactive cells were treated with ficin (40–80 μg/ml) or papain (5–10 μg/ml) without KC1, no loss of cilia was detected. Thus ciliary loss is not due to direct enzymic action by the proteolytic enzymes.

In Figs. 20–22, presumed autogamous cells obtained by chemical induction are shown. The extent of ciliary degeneration 1·5 h after the onset of the treatment was the same as that of usual conjugation at 1·5 h (Fig. 20). Ciliary degeneration reached its maximum extent 3 h after the begninning of the induction of autogamy, because no significant additional degeneration was observed after 5 h (Figs. 21, 22). Anterior to the vestibule, the area of ciliary loss extended 7 kineties to the left of the suture and 4 kineties to the right; posterior to the vestibule, it extended only 10 or more kineties to the right (Figs. 23-25). The right wall of the vestibule was also deciliated (Fig. 24).

The localization of ciliary degeneration during sexual reproduction is an interesting morphogenetic problem in Paramecium. Mechanisms which might control ciliary degeneration at a definite place and time have not yet been clarified. However, there should be some relation between acquisition of mating reactivity and the degeneration of cilia, because ciliary degeneration occurs only in cells physiologically competent to give a mating reaction, and in cells with higher mating reactivity, ciliary degeneration extends more rapidly. The facts that cells lose the capacity to mate before undergoing natural autogamy and that they also show no degeneration of cilia during autogamy also support the above hypothesis. Significantly, P. tetraurelia does undergo ciliary degeneration when autogamy is induced chemically (T. M. Sonneborn, personal communication).

Mating reactive cilia are distributed on the ventral surface of the cell (Hiwatashi, 1961; Cohen & Siegel, 1963; Cohen, 1964). In P. caudatum, if mating reactivity is weak, cells contact mainly at the anterior portion, but if it is strong, they contact over a much wider area (Hiwatashi, 1961). This suggests that in P. caudatum, the strength of mating reactivity extends from anterior to posterior cilia. The pattern of ciliary degeneration follows that of mating reactivity. This correlation may not be significant, since in P. bursaria mating reactivity spreads from posterior to anterior (Cohen, 1964), whereas we have observed ciliary degeneration extending from anterior to posterior. Therefore, there is no direct relation between the pattern of development of mating reactivity and of ciliary degeneration in P. bursaria, though the final locations of both phenomena seem essentially the same.

The location of ciliary degeneration is identical in all 5 species of Paramecium examined in this study: both sides of the anterior suture and the right side of the posterior suture. Vivier & André (1961) cut serial sections of conjugant pairs and showed that cells unite at the right side of the suture. However, it seems likely that cells may unite on both sides of the anterior suture, because ciliary degeneration occurs on both sides of it and one never sees on conjugating cells any exposed deciliated surface on the anterior part of the cell. In our experience, it is difficult to determine the location of the suture with certainty in sectioned material.

Others have proposed that the loss of cilia during conjugation is due to their separation from the cell body (Vivier & André, 1961; Bloodgood, 1974). However, the present observations of many short cilia during degeneration suggest that cilia may be lost by resorption. Since the electron-microscopic observations on late conjugating pairs show that kinetosomes remain intact at the contact region (Vivier & André, 1961; Jurand & Selman, 1969; T. Watanabe, unpublished), degeneration of each cilium seems to stop at the region between the kinetosome and the ciliary shaft. Resorption of cilia has been reported in other ciliates: during conjugation of Oxytricha (Hammersmith, 1976) and during oral replacement and partial deciliation in Tetrahymena (Williams & Nelsen, 1973; Rannestad, 1974). As shown in Tetrahymena, active resorption mechanism(s) act at a specific position and a specific time in the cell cycle (Williams, 1975). Likewise in the conjugation of Paramecium, resorption mechanism(s) act at a specific position, the ventral side of the cell, and at a specific time, namely, when cells are activated by the ciliary mating reaction or by chemical treatment inducing conjugation or autogamy.

The extent of ciliary degeneration reached its maximum 3 h after beginning of the chemical induction of autogamy. Miyake (1968b) reported that if the cells of P. multimicronucleatum were treated with autogamy-inducing chemicals for 3 h, nuclear activation was induced. Similarly, in P. caudatum, the minimum length of the chemical treatment necessary for nuclear activation is 3 h (Y. Tsukii, unpublished). Hiwatashi (1955 a) reported that paroral union, which is formed more than 2 h after the start of the mating reaction, is essential for nuclear activation in the conjugation of P. caudatum. These results suggest that the time for nuclear activation corresponds closely to that of the maximum extension of ciliary degeneration. Studies are still in progress to discover if there is a causal role for ciliary degeneration in cell contact and nuclear activation in Paramecium.

The author is much indebted to D. L. Cronkite for his help in the preparation of the manuscript, and would like to thank Dr T. M. Sonneborn and Dr K. Hiwatashi for reading the manuscript, and Dr K. Mikami and Mr Y. Tsukii for technical assistance. This work was supported by a grant from the Ministry of Education of Japan.