Temperature-induced modifications in size and pattern of microtubular organelles in a ciliate, Dileptus: I. supernumerary microtubules in axonemes of sensory cilia

Microtubule pattern in locomotor cilia is highly conservative, but may be altered by expression of some genic mutations (Forest, 1983), by the action of taxol (Herth, 1983) or by cold treatment (Szôllôsi, 1976). Usually some microtubules are missing from the modified axoneme. When supernumerary microtubules are present (Herth, 1983) the ciliary axoneme is complete, and the site of attachment of these additional microtubules must be some place other than the basal body.

Cilia that have no locomotor function, such as the primary cilia in embryonic tissue (Dalen, 1981), sensory cilia of many receptors (reviewed by Barber, 1974; Altner & Prillinger, 1980), or those called the sensory or clavate cilia in ciliates (Grain & Golinska, 1969; Holt et al. 1973; Golinska, 1982, 1983), usually have a modified microtubule pattern. When the pattern is changed, the basal body usually bears fewer microtubules than the locomotor cilium. The supernumerary microtubules, if present, are not structurally linked to the basal body, but are anchored to clusters of dense material (Phillips, 1979; Erler, 1983).

In this study supernumerary microtubules were found in axonemes of the so-called sensory cilia in Dileptus. These microtubules are not connected to the basal body, but to some dense material within the ciliary shaft. The frequency of appearance of these microtubules, and their number per cross-section, were found to increase with increasing temperature. Data were analysed to determine the possible ways of formation of the supernumerary microtubules.

The material used in this study was Dileptus margaritifer, in previous publications referred to as Dileptus anser (see revision by Wimsberger et al. 1984). Stock cultures were kept at room temperature and fed every other day with Colpidium spp. Details of culture methods have been published (Golinska & Jerka-Dziadosz, 1973).

In this study posterior fragments (opimers) of Dileptus cells were used. Transections were made in the middle of the trunk, then the posterior fragments were isolated into depression slides and placed in a thermostat adjusted to 10°C, or 32°C. Control fragments were left at room temperature. At 24 h after the operation all cells were fixed and prepared for electron microscopy. Since the sensory cilia are situated in the anterior region of the cell, all the sensory cilia observed were those formed by opimers exposed to cold, heat, or room temperature during the 24 h following the operation. Data given for‘normal’ cells were obtained from micrographs made during previous studies on untreated cells taken from growing cultures. Fig. 12 is part of a published micrograph (see Golinska, 1983, fig. 22).

Fixation for electron microscopy was performed by mixing equal amounts of OsO₄ (4 %), glutaraldehyde (6 %), and 0·05 M-cacodylate buffer, pH 6·8. The mixture was prepared immediately before fixation. During dehydration, 30% ethanol containing 0·1 % tannic acid was applied for 30 min. Further processing of the samples was standard. Preparations were examined in a JEM 100 B transmission electron microscope.

The cell of Dileptus consists of two main parts: the trunk and its anteriorly located slender elongation, the so-called proboscis. A cytostome is situated at the base of the proboscis. The whole body except for the oral region is covered with longitudinal rows of somatic cilia. Several dorsally located rows contain locomotory cilia in their posterior portion, while in their anterior portion the pairs of sensory cilia can be seen situated on the proboscis (Fig. 1). The fine structure of these cilia, as well as the pathways of their formation and transformation, have been described (Grain & Golinska, 1969; Golinska, 1982, 1983). The cilia are termed‘sensory’ not because of their function, which is unknown, but because of their structural simplification, which resembles that of cilia found in sensory cells in chemo- and mechanoreceptors of higher organisms (reviewed by Gaffal & Bassemir, 1974; Altner & Prillinger, 1980). The sensory cilia in Dileptus are shorter than locomotor ones (Fig. 1), have no B tubules (i.e. the 9 outer microtubules are single) (Figs 4, 6, 7), and contain no intertubular links. This is confirmed by the irregular distribution of micro-tubules within the sensory shafts (Figs 4, 6, 7, 10). The sensory cilia of each pair are very much alike, differing only in that the basal body of the posterior cilium is equipped with additional root fibres (Fig. 2).

Most cross-sections of sensory cilia contain 11 microtubules, i.e. the two microtubules of the central pair and nine outer single microtubules (Fig. 4). The microtubules of the central pair are probably shorter than the outer ones, since there are many sections containing nine and fewer microtubules. Such sections were not taken into account. Sections showing 11 single microtubules and no central pair were frequently encountered. These were regarded as probably resulting from separation of the central pair and classified as cilia without additional microtubules, although we cannot exclude the possibility that such sections represent axonemes with two additional microtubules sectioned above the central pair. In counting microtubules no discrimination was made between outer microtubules and microtubules belonging to the central pair.

The additional microtubules were first observed in sensory cilia of cells that regenerated their probosces while exposed to elevated temperature. This led to observations being made on sensory cilia formed by opimers exposed to 32°C, 20°C and 10°C. Since my equipment was rather primitive, possible temperature deviations were ±2 deg. C. The results of microtubule counting on cross-sections of sensory shafts are summarized in Table 1. The number of cells to which the cilia belonged was always more than 20. The results show clearly that elevation of temperature is followed by an increase in the number of supernumerary microtubules, while cold treatment causes reduction in their number. At room temperature the additional microtubules are formed in intermediate quantities, similar to that found in normal cells.

The additional microtubules were never observed at the level of the axial granule, and no modifications of microtubule pattern were found in the basal bodies of sensory cilia. It was always in the portion of a sensory cilium above the axial granule that the supernumerary microtubules were found. Within shafts containing supernumerary microtubules some of the micro-tubules were usually in contact with masses of dense material, often situated beneath the ciliary membrane (Figs 6, 7). Similar clusters of dense material, but without apparent connection with microtubules or ciliary membrane, were often observed in sensory shafts (Figs 8, 10), and also in shafts of locomotor cilia (Figs 9, 11), where additional microtubules were never encountered. No counts of the cilia containing such dense material were made, but they can be found in cells exposed to both elevated and low temperatures. The dense material may serve as the site of attachment for additional microtubules in sensory cilia.

The supernumerary microtubules, when present in several cilia in one section, were always found in every other cilium in the row, i.e. in one cilium of each sensory pair (Fig. 5). In four cases grazing sections were found, showing not only cilia with supernumerary microtubules, but also the proximal parts of basal bodies in another sensory pair of the same region. This enabled a distinction to be made between the anterior and posterior cilium in a pair. In three cases the additional microtubules were situated in the anterior shaft, and in one case in the posterior sensory shaft.

This peculiar location of additional microtubules, in only one cilium of the sensory pair, indicates that the formation of these microtubules takes place during the formation of the sensory cilium. As was reported earlier (Golinska, 1983), the sensory unit is formed by transformation of a locomotor unit: through the resorption of a locomotor cilium (leaving its basal body intact), formation of a basal body for the anterior cilium, and formation of sensory shafts for both basal bodies of the pair. The additional microtubules, when situated in the posterior ciliary shaft, may represent the remnants of an old locomotor cilium, not properly resorbed.

A possible explanation of the presence of additional microtubules in the anterior ciliary shaft came from a confirmation of the earlier observation (Golinska, 1983) that there is no definite sequence of anterior basal body formation, resorption of the locomotor cilium and formation of sensory cilia. In the region where the sensory units are forming, basal body formation may precede or follow resorption of the locomotor cilium, and anterior sensory cilia may be found with posterior cilia not yet resorbed, with posterior cilia undergoing resorption, or with posterior newly formed sensory cilia (Figs 12, 13). It was in the first configuration that a longitudinal section of a developing sensory pair was found, showing a separation of the new sensory axoneme from the anterior basal body (Fig. 14), possibly in response to the resorption process beginning in the locomotor axoneme of the posterior cilium. A probable later stage is presented in Fig. 15, where sensory shafts have already formed on both basal bodies, and supernumerary microtubules in the anterior one are distally displaced.

The complete lack of additional microtubules in locomotor cilia, in spite of the presence of dense material in their shafts (Figs 9, 11), is further confirmation that the supernumerary microtubules are not nucleated at clusters of dense material (or some other structure, i.e. ciliary membrane), but are the remnants of previously existing axonemes.

Several questions concerning the supernumerary microtubules seem to be of interest, namely their sites of nucleation, their temperature sensitivity, and their bearing upon the morphogenesis of microtubular organelles.

The site of nucleation of supernumerary microtubules may be represented either by the basal body, or some other structure where the microtubules are anchored, i.e. dense material or ciliary membrane. The supernumerary microtubules in sensory cilia of Dileptus are, in my opinion, nucleated at basal bodies, being the remnants of an axoneme separated from the basal body and later replaced by a set of microtubules forming a new axoneme (Fig. 16). The presence of additional microtubules in only one cilium of the pair indicates that their formation takes place during the formation of a sensory unit. The separation of the locomotor axoneme from the posterior basal body is a normal event in ciliogenesis of sensory cilia (Golinska, 1983). Separation of the sensory axoneme from the anterior basal body may be induced when the resorption of a posterior locomotor cilium coincides with the presence of an already developed anterior sensory cilium (Fig. 16). Since the formation of additional microtubules is rather uncommon (in the posterior cilium as well as in the anterior one), their occurrence in both components of a sensory pair may be extremely rare, giving the impression that additional microtubules can be found only in one cilium of a pair. This pattern of occurrence of additional microtubules would be inexplicable if they were nucleated at some structure other than a basal body.

This speculation concerning the sites of nucleation of supernumerary microtubules is further supported by several observations. The total number of microtubules per sensory shaft was always less than two sets of microtubules; in fact, the highest number of additional microtubules was five. This, again, may be easily explained only when the basal body is accepted as the site of nucleation of all microtubules in the ciliary shaft, nucleation at any other structure being theoretically able to generate an unlimited number of microtubules. Also a complete lack of supernumerary microtubules in locomotor cilia (well equipped with structures representing possible nucleating sites) speaks in favour of the basal body as the site of nucleation of additional microtubules.

There is a possibility that the formation of additional microtubules by separation of the axoneme from the basal body, may not be found in Dileptus only. The observations of Herth (1983) on taxol-induced formation of supernumerary microtubules in flagella has revealed that besides single microtubules, supernumerary doublets and central pairs were also formed. It seems very unlikely that such structures would be nucleated at some structure other than a basal body. In the same study the taxol-induced resorption of flagella was reported as being frequent (Herth, 1983).

In this study temperature was found to influence both the number and the frequency of appearance of additional microtubules. This means that either there is an alteration in the number of cilia containing the supernumerary microtubules, or a change in length of these microtubules, or both. Whichever is the case, the additional microtubules show higher sensitivity to temperature than the microtubules of the sensory shaft. This may be directly related to the stability of microtubules, which is well known from in vitro studies to be temperature-sensitive (review by Scheele & Borisy, 1979). In vivo the microtubules show differences in their sensitivity to temperature, even if anchored at the same microtubule-organizing centre (MTOC) (Jones & Tucker, 1981; Schultze & Kirschner, 1986). Different sensitivity for microtubules anchored at different structures, like those of the sensory shaft and supernumerary microtubules in Dileptus, seems highly likely. Another possibility is that the temperature sensitivity of supernumerary microtubules is a consequence of a temperature effect upon the process of ciliary resorption (which is known to be temperature-dependent: see Hinrichson, 1981), or upon the process of capping the microtubules with dense material or some other structure.

The idea that microtubules can be nucleated at one site, and afterwards anchored at another site, has some interesting consequences. For microtubular organelles in ciliates it is generally accepted that a MTOC is both nucleating and capping the microtubules (Tucker, 1982). A second function for a MTOC, that of anchoring and thereby stabilizing the microtubules nucleated elsewhere, was proposed by Kirschner (1980). Basal bodies in sensory cilia of Dileptus operate as a MTOC of the first kind upon microtubules of the sensory axoneme, while the function of the MTOC of the second kind is fulfilled by some other structure, i.e. by clusters of dense material in the sensory axoneme. The basal bodies of sensory cilia, in their formation of consecutive generations of microtubules, resemble the dynamic model of the centrosome presented by Schultze & Kirschner (1986), but differ from the model in the formation of separate populations of microtubules from these microtubule generations.

This work was supported by grant no. C.P.B.P. 04.01 from the Polish Academy of Sciences. I thank Dr A. V. Grimstone and Dr Maria Jerka-Dziadosz for critical reading of the manuscript, and Mrs Lidia Wiernicka for expert technical assistance.