Proportional regulation of body form and cortical organelle pattern in the ciliate Dileptus

The terms: form, pattern and regulation are applicable to various processes and at different levels of biological organization. To avoid possible misunderstanding, we want to make precise the sense in which the terms will be used in this study. Form stands for the shape of Dileptus, and pattern we use to indicate the pattern of ciliary organelles on the cell surface. Regulation is the process of reorganization of damaged or deficient cells into entities of correct proportions on the 2-dimensional level (pattern) and 3-dimensional level (form). Both form and pattern are species-specific in ciliates; they may change within the cell cycle, with environmental conditions, and may also be changed experimentally. Our study deals with the regulation of pattern and form that follows transection of Dileptus.

In other ciliates, when a portion of the cell is cut off, the correct shape may be regained in either of 2 ways. Epimorphic regrowth of the lost part has been found in Paramecium (Tartar, 1954) and Chilodonella (Kaczanowska 1975). Shape regulation of the whole fragment occurs in Stentor, resulting in a smaller but correctly proportioned cell. This morphallactic formation of the lost part may proceed in the absence of growth and was reported in Stentor as early as 1901 by Morgan. Similar ability to regulate shape was first observed in Dileptus by Sokoloff (1924).

Regulation of ciliary pattern is best known in some Hypotricha. When a portion of the cell is cut off, ciliary primordia form and a complete new pattern develops, composed of a smaller number of elements (Jerka-Dziadosz, 1976). The regulation of pattern (change in number and size of its elements) is possible only while it develops, and the remnant of the old pattern is unable to regulate. The ciliary pattern of Stentor is, within limits, able to regulate, at least its oral portion. When the intact oral structures occupy an area that is too large in comparison to the somatic area of the cortex (i.e. when a large somatic portion of the cell is cut off), the oral parts diminish in situ (Schwartz, 1935; Tartar, 1959, 1961; De Terra, 1969), but the mode of diminution is completely unknown. A too-small oral area of Stentor is, however, unable to enlarge and a new set of oral structures has to be produced. Regulation of the ciliary pattern of the somatic area remains obscure.

Dileptus is a perfect object in which to study the regulation of form and pattern. Unlike most ciliates it has 2 distinct body regions: proboscis and trunk, which do not, however, coincide with 2 distinct areas of ciliary pattern: oral and somatic. Moreover, measurements are quite easy to perform because of rather poor contractility of the cell.

The regulative capabilities of Dileptus are exceptional. Correct proportion of body parts is generally restored in 24 h after transection, even in tiny fragments. Shape regulation involves translocation of the cell contents and is accompanied by local adjustments (resorption and proliferation) of somatic ciliature. Within the mouthparts a localized area of oral kinetosome resorption is found. This is situated at the opposite side of the mouth to the previously reported area of kinetosome proliferation (Kink, 1975). The existence of these 2 areas makes the oral structures of Dileptus able to regulate their size at other times than during stomatogenesis.

Dileptus anser O.F.M. was used in all experiments. Stock cultures were kept in beakers and fed every other day with Colpidium colpoda. Suspension of egg yolk was given to Colpidium cultures once a week. The culture medium for both organisms was Pringsheim solution.

A day before an experiment some of the Dileptus stock was transferred to Petri dishes. There the operations were performed, by hand, using a microscalpel. Cells were always sectioned transversely, and anterior fragments of a desired size were isolated into depression slides. No food was given to these fragments. Observations were performed on anterior fragments of different size, including the proboscis alone, and also on intact, isolated cells. For in vivo observations single cells were isolated on each slide. For observations with the light and electron microscope several dozen cells were cut into fragments during a 30 to 60-min interval. Suitable fragments were isolated into one depression slide; and then fixed at a known time after the end of the operation.

The length of the proboscis from its apex to its base (Fig. 1, p. 13), and the length of the rest of the body from the base of the proboscis to the most posterior tip of the trunk, were the dimensions taken into consideration. Measurements were performed on single, isolated cells. Each cell was photographed 3–4 times every 30 min during several hours following isolation. The negatives were projected at a constant magnification on to paper sheets and cell contours were outlined. The length of proboscis and trunk was then measured on the outlines.

Preparations for light microscopy were stained with Protargol (Dragesco, 1962). Material for electron microscopy was prepared using conventional methods, except for fixation. The fixative used was a mixture of 2 parts of 2 % osmium tetroxide (prepared with cacodylate buffer, pH 7·2) and 1 part of 6 % glutaraldehyde, freshly prepared and maintained at 0°C during fixation (1 h). The sections were examined with a JEM 100B electron microscope.

The morphology of Dileptus has been presented in detail elsewhere (Grain & Golinska, 1969; Kink, 1975; Golinska & Kink, 1976), but the essential features will be outlined here. The shape is that of a cylinder, tapering at both ends, the anterior end forming a proboscis, the posterior forming a tail (Fig. 1). The whole body is covered by longitudinal somatic kineties. Oral structures are located at the base of the proboscis and all along its ventral side. At the base of the proboscis there is a circular oral field with a cytostome in the middle. Along the ventral side of the proboscis there is an extension of the oral field, the so-called ventral band. The oral parts are encircled by the oral kinety. This kinety at both margins of the ventral band is composed of kinetosomal pairs (Fig. 12, p. 23). The segment of the oral kinety that surrounds the cytostome is built up out of single non-ciliated kinetosomes, each bearing a nemadesma (Figs. 2, 4). Nemadesmata connected with the oral kinety form the palisade of the pharyngeal basket, which extends deeply into the endoplasm. The ectoplasm and endoplasm of Dileptus are separated by a microfibrillar layer situated at the level of the proximal ends of kinetosomes (Fig. 13, p. 24). This layer envelopes the whole body and in the oral areas is differentiated into a complicated network. The nuclear apparatus, which in Dileptus anser consists of several hundred pieces, is disposed uniformly throughout the endoplasm, and is absent only in the narrow distal portion of the proboscis and in the pharyngeal region.

The anterior fragments used for experiments were of different sizes. They were designated, according to size, as A, B, C and D fragments (Fig. 5). Type A contain the entire oral apparatus; in some of them only the deepest portion of the nemadesmata were removed. Transection just posteriorly to the oral field resulted in type B fragments, with an intact oral field but with the nemadesmal basket severely damaged. Type C were cut through the oral field, thus disrupting the oral kineties and nema-desmal basket. Type D fragments were deprived of the whole circular part of oral field; in some of them, however, a few elements of the basket may persist.

Fragments of all types were able to reorganize into correctly proportioned cells, without food intake, and thus without any increase in their volume. Our observations concern 2 aspects of the changes occurring during reestablishment of the correct proportions of the cell parts, namely, the regulation of form and the adjustments of the cortical structures, especially those of the oral apparatus.

The ratio of proboscis length versus length of the trunk was chosen as a measure of cell proportions in A, B and c fragments. Type D fragments, where the formation of a new trunk occurs, will be described separately.

In controls and in operated animals the length of proboscis and the length of trunk show some variability with time. It seems that this variability is not due just to experimental error, but reflects actual transient changes of length. The lengths of proboscis and trunk vary independently, so that a graph showing proboscisμrunk ratio against time for a single cell has the form of a zigzag line (Fig. 6). Since short-term observations are thus not reliable, observations were always prolonged up to 24 h after isolation.

Isolated trophic animals of an unknown stage of the cell cycle served as a control. Measurements were performed on 10 individuals. The proboscisμrunk ratio for these is 0·85–0·95 immediately after isolation and drops slightly after 24 h of isolation (Table 1, Figs. 6, 7).

The proboscisμrunk ratio was measured in 10 anterior fragments: 7A, 2B, and 1C. In anterior fragments the total length, i.e. length of proboscis and trunk together, does not change very much during 24 h after operation (Table 1, Fig. 7). Within this period the normal proportions of the cell are almost completely restored; the value of the proboscisμrunk ratio at the end of this period (0·80–1·10) is only slightly above the value found for normal cells (Fig. 6).

The changes in proportions of A, B and C fragments are due both to shortening of the proboscis and to elongation of the trunk (Fig. 7). The process of shortening of the proboscis is gradual, lasting about 24 h. The proboscis in the anterior fragment 24 h after operation is about half of its initial length (Table 1). The shortening is obscured by small short-term fluctuations in proboscis length, observed from the moment of isolation. A somewhat more distinct decrease in proboscis length was noted immediately after the operation. The same was found, however, in controls. This shortening may be a reaction to handling during the isolation, rather than to transection.

When the wound has healed after operation, the hind end of the fragment appears rounded up (15 min after operation). At that time the elongation and narrowing of the trunk beings. Narrowing of the rear end of the fragment proceeds especially quickly, until a tail is formed (about 1 h after operation). The highest rate of trunk elongation was found during the hour following operation and was 60 μm/h. The slope of proboscisμrunk curves that follows the operation (Fig. 6) is almost entirely due to trunk elongation, and not to shortening of the proboscis. Later on the rate of trunk elongation decreases, and it takes over 10 h to finish.

Both the shortening of proboscis and the elongation of the trunk, occurring in a given fragment of approximately constant total length and volume, are obviously based on translocations of the cell contents. Comparison of different outlines of the same cell (Fig. 8A) shows that, even when the proboscis does not shorten, material for trunk elongation may be partially withdrawn from the proboscis, this being indicated by narrowing of the proboscis. When outlines taken immediately after operation and 24 h after operation are superimposed (Fig. 8, D²⁴) it becomes clear that shape regulation involves rearrangement of the whole cytoplasmic contents.

The proboscis, isolated from the remainder of the cell survives only if sectioned at its base. Such fragments contain the whole ventral band (extension of oral field), usually a little of nemadesmata, and in the thick basal portion several somatic kineties and numerous pieces of nuclear apparatus.

Measurements of length were performed on 10 fragments, with results summarized in Table 1 and Fig. 7. As for A, B and C fragments, the length of a type D fragment immediately after isolation is almost equal to its length 24 h later. In D fragments proportions are regained not by shortening of proboscis and elongation of trunk but by formation of new trunk.

Trunk formation begins with translocation of the content of the fragment. In 30–60 min after operation there is a very distinct change of shape in D fragments (Fig. 8D ¹). A narrow anterior portion (future proboscis) and a much thicker posterior portion tapering at its hind end (future trunk) are formed. Tapering of the hind end of type D fragments occurs very early during shape regulation. However, formation of a proper tail is much delayed in comparison to other types of anterior fragments; it takes 3–6 h. It is noteworthy that thin and thick portions of D fragments are not initially in the correct final proportion one to the other; shortening of future proboscis and elongation of the future trunk are still necessary to produce the normal shape.

Twenty-four hours after the operation, D fragments regain the proportions of the normal cell (Table 1, Fig. 7). The proboscis is not only shortened but also very narrow. Its length decreases during 24 b from 205·5 to 94·5 μm (mean values for 10 specimens). Its diameter, measured in the middle of the length of the proboscis, decreases (for the same 10 specimens) from 18·9 (S.D. ± 1·2) to 11·5 (S.D. ± 1·9) μm.

Regulation of the shape of a fragment is accompanied by changes in cortical pattern. In this paper only rearrangements of oral structures and some observations concerning the distribution of ciliary resorption are given. A detailed study of pattern regulation will be published separately.

Rearrangement of oral structures consists of shortening of the ventral band and adjustment of size of the oral field. On Protargol-impregnated preparations oral kineties on the actually shortening proboscis do not differ in density of kinetosomes from kineties on controls. It is obvious that shortening of the proboscis involves decrease in number of kinetosomal pairs within oral kineties on the proboscis.

The regulation of size and shape of the circular portion of oral fields involves a reduction of size in A and B fragments, shape regulation and sometimes increase of size in c fragments, and formation and increase in size in D fragments (Fig. 5).

In type A fragments the process of reduction of the oral field is very slow and proceeds gradually. Measurements of the diameter of the circular portion of the oral field were performed on Protargol-impregnated preparations. The mean value (for 10 specimens) in controls was 20·4 μm. In type A fragments 3 h after operation the diameter of the oral field was 15·5 μm; 24 b after operation it was 11· 0μm. These data do not reflect the actual dimensions of oral fields, because there is shrinkage of cells in preparation. They indicate, however, that the diameter of the oral field of type A fragments 24 h after operation is about half of its initial value.

Oral fields of type B fragments undergo a striking change of shape (Fig. 5), leading to a much smaller mouth. The oral field, instead of being circular, becomes an elongated band, appearing as a prolongation of the ventral band. This was observed in Protargol preparations made 1–2 h after the operation. In preparations made 2–3 h after the operation a small oral field of normal appearance is situated at the base of the ventral band, somewhat irregular in width, that we designate as a transforming ventral band. It seems that the oral field is reduced in size through transformation of some of its anterior part into the ventral band.

Type c fragments 30–60 min after operation have oral fields which appear very small but normal in appearance. The closing of disrupted baskets, therefore, must occur immediately after the operation. We never saw an interrupted basket. In some fragments bundles of nemadesmata are seen in the endoplasm; these are possibly the remnants of the lost portion of basket.

The ventral band, the only oral structure left on D fragments, immediately after operation reaches the posterior pole of the fragment. Later on the ventral band becomes gradually shorter and oral structures are formed at its posterior end. At 2–3 h after the operation some pharyngeal parts appear there, at about the middle of the thick portion of the fragment. The ventral band is thus still longer than the thin part of the fragment. At 5–6 h after operation the length of ventral band becomes equal to the length of the thin part of the cell, so a real proboscis is formed. At this time, however, normal proportions are still not regained, the proboscis being much too long for the rest of the cell. The nemadesmal basket is clearly visible at the base of the proboscis. At 24 h after operation oral structures were checked to see whether they are able to function: tiny Dileptus cells swallowed Tetrahymena killed or immobilized by high temperature.

The fine structure of oral fields that undergo these changes shows surprisingly little modifications when compared to the normal ones. The only change in structure of oral fields in A and B fragments was the appearance of a microfibrillar ring on the inner margin of the nemadesmal basket (Fig. 4). This ring was situated at the level of the proximal ends of oral kinetosomes and equipped with smooth spherical vesicles. After a lapse of 24 h the microfibrillar ring was no longer visible (Fig. 2). Type c fragments when fixed during the 30 minutes following the operation show an unusually thick microfibrillar layer accompanied by smooth vesicles adjacent to the nemadesmal basket (Figs. 3, 9). This microfibrillar layer is identical to and continuous with the layer closing the lest of the wound (Fig. 10). Both are continuous with microfibrillar material that separates ectoplasm and endoplasm in Dileptus. In some c fragments fixed in the period 1–2 h after operation the proliferation of oral kinetosomes was observed next to the nemadesmal basket. This probably reflects the enlargement of the too-small oral fields.

Resorption of oral ciliature was found at the apex of the proboscis in all types of fragments (Fig. 11). In D fragments in addition to resorption at the apex of the proboscis, resorption at the proximal end of the ventral band was also found (Fig. 12). This is probably a resorption of ciliary elements damaged by cutting. Resorption of the oral ciliature was never observed within the region of the nemadesmal basket.

Electron-microscope observations also revealed that numerous somatic cilia are resorbed during shape regulation at the apex of the proboscis (Fig. 11) in all types of fragments. Localization of resorption of oral and somatic kinetosomes at the apex of the proboscis indicates that shortening of the ventral band and the whole proboscis is due to withdrawal of material from the apex. Resorption of somatic kinetosomes was also found in all areas that undergo narrowing; intense resorption occurs close to the posterior pole during wound healing (Figs. 10, 14), and later on during tail formation (Fig. 15). Some resorption was observed in the whole somatic area of the proboscis and in the region at the level of the cytostomal fields (Fig. 3).

The proliferation of somatic kinetosomes was found to occur at random all over the cortex, in all types of fragments. No particular area of intense proliferation was detected.

The process of shape regulation, leading to proportionality of cell parts, is almost completely unexplored in ciliates. For study of shape regulation Dileptus represents as suitable material among Protozoa as Hydra does among Metazoa. The possibility of conversion of trunk into proboscis (Golinska & Kink, 1976) and of proboscis into trunk (this study) makes this shape regulation of the ‘French Flag’ type, having all the features that Wolpert (1968) has formulated ‘(a)…the proportion between parts is, within limits, invariant with size; (b) part of the system can produce the whole, and…any part can become any other part; (c) the polarity of the system is maintained’.

An interesting feature of shape regulation found during this study on anterior fragments of Dileptus is that the total length of the regulating cell stays almost invariant. We suppose that shaping is executed by changes in tension forces along the main axis of the body, causing translocations of the cell contents. These translocations result in narrowing of both ends–proboscis and tail–and either narrowing or widening of the middle portion of the body. As indicated by comparison of the outlines, the changes in the shape of the body begin at the posterior end of the fragment and later on spread up to the anterior end. It must be stressed that in posterior fragments shape changes begin at the anterior end and spread down to the posterior pole (Golinska, unpublished), indicating that shape regulation begins in the region of the wound.

The structure which is most likely to be involved during body shaping in Dileptus is the microfibrillar system. This system forms a continuous layer between ectoplasm and endoplasm and is accompanied by numerous elements of rough endoplasmic reticulum, and by a variable number of smooth spherical vesicles. Within cell areas that undergo quick changes of shape the thickening of the microfilamentous layer was observed, together with the appearance of numerous smooth vesicles. This was observed during wound healing in Dileptus cygnus (Golinska & Grain, 1969), and D. anser (this study), in the fission line (Golinska, 1972), close to nemadesmal baskets in A and B fragments and as a structure closing the basket together with the rest of the wound in early c fragments (this study). It was observed, moreover, that fragments of Dileptus unable to change their shape after puromycin treatment (Golinska, 1974) have a damaged microfibrillar system. We suggest that the microfibrils, together with their vesicular elements, which appear to be responsible for shape regulation in Dileptus, may be analogous to the contractile unit in Spirostomum (Legrand & Prensier, 1976).

In other unicellular organisms the structures that are generally supposed to play an essential role in cell shape formation and maintenance are microtubules (Tilney, 1968; Tartar & Pitelka, 1969; Banerjee & Margulis, 1973; Bouck & Brown, 1973; Brown & Bouck 1973; reviewed by Porter, 1973). Microfibrillar structures are supposed, however, to maintain the cell shape in Paramecium (Sibley & Hanson, 1974), to close the wound in Amoeba (Jeon & Jeon, 1975), act in fission of Nassula (Tucker, 1971), and move some other organelles to their final positions during development (Lynn & Tucker, 1976) Microfibrils are also supposed to play a role in shaping of some metazoan cells (reviewed by Wessels et al. 1971).

In ciliates the relations between shape regulation and ciliary pattern regulation are not clear. There exists some evidence showing independence of pattern and form, and vice versa. The correct pattern with abnormal shape (Doerder, Frankel, Jenkins & De Bault, 1975) and abnormal pattern with correct shape (Frankel, 1973) have been observed in mutants. Unfortunately both cases deal with organisms (Euplotes and Tetrahymena) that are known not to be good regulative systems. The process of shape regulation in ciliates may proceed in the absence of nuclear apparatus (reviewed by Balamuth (1940); for data concerning Condylostoma see Yagiu (1956), Blepharisma, Suzuki (1957), Stentor, Tartar (1961), and Dileptus, Golinska (1966)). On the other hand, the regulation of ciliary pattern–diminution of the too-large mouth or formation of a smaller set of cortical structures–is known to be impossible in the absence of a nucleus (Dembowska, 1926; Tartar, 1959). In Dileptus form and pattern regulation show different sensitivity to puromycin: shape regulation is inhibited completely, while formation of oral kinetosomes is still possible (Golinska, 1974). It must be emphasized, that shape regulation and pattern regulation may be quite different processes, controlled in different ways. The information necessary for positioning of cortical structures (Frankel, 1974; Jerka-Dziadosz, 1974; Kaczanowska, 1974; Lynn & Tucker, 1976) may be different from the information necessary for shaping of cell parts in proportional fashion.

We have found in this study, that during formation of 2 body regions, namely proboscis and trunk in type D fragments, shaping and pattern regulation have different timing, i.e. the length of the future proboscis (thin portion of D fragment) is shaped long before the oral structures of the proboscis shrink enough to be of the same length. We do not know, however, if a similar lapse of time exists between shaping and patterning in the somatic part of the body.

The process of pattern regulation in anterior fragments of Dileptus involves the somatic as well as oral ciliature. In both kinds of ciliature the remnants of the old pattern undergo local adjustments through resorption and/or proliferation of ciliary elements. An unusual feature of this regulation is the ability to repair injured mouthparts. In ciliates other than Dileptus the injured oral structures are either replaced by a new set of oral structures (e.g. in Stentor (Tartar, 1961), in Urostyla (Jerka-Dziadosz, 1963)) or stay incomplete (Euplotes (Wise, 1965)). It seems that this ability to repair in situ the injured mouth is due to the relative simplicity of oral ciliary pattern in Dileptus (Jerka-Dziadosz & Golinska, 1976).

The high regulatory capabilities of oral ciliary pattern in Dileptus are due to the existence of 2 special areas located within the mouthparts: one apical for kinetosomal resorption (found in this study) and one proximal for kinetosomal proliferation. The latter area was found at the outer margin of the nemadesmal basket in growing cells (Kink, 1975) and during regrowth of the proboscis (Golinska & Kink, 1976). Both areas may function simultaneously, i.e. in early c fragments resorption at the apex of the proboscis and proliferation next to the nemadesmal basket may be found in the same cell. The existence of special areas that make possible regulation of the size of the mature mouth is so far evident only in Dileptus. However, it seems probable that a similar system may be found also in other ciliates. It cannot be excluded that the oral apparatus of Stentor which can decrease in size, has its area of resorption, located at the far end of the membraneliar band. It cannot be excluded, either, that the oral primordia of ciliates during their stage of differentiation have a regulative area of kinetosomal proliferation located proximally, i.e. in the region that is the last one to differentiate.

Another problem is the difference found in regulation of oral and somatic pattern in Dileptus. The somatic ciliature undergoes local adjustments, ciliary resorption being localized within narrowing areas of the cell. Thus, areas of resorption of the somatic ciliature reflect shaping processes. This is not true for regulation of the oral ciliature. There is no resorption of the ciliature of shrinking baskets. Resorption was found to be localized in the apical part of the proboscis. We believe, that instead of resorption there is transformation of some portion of the ciliature of the basket into oral ciliature on the proboscis, resorption at the proboscis tip being sufficiently intense to allow diminution of the basket together with shortening of the proboscis. This difference in regulation of oral and somatic pattern in Dileptus is in accordance with the recently proposed idea (Jerka-Dziadosz & Golinska, 1976), that each kind of ciliary pattern has its own specific way of regulation.

This investigation was supported by Polish Academy of Science, research grant PAN 22, II, 3. Electron micrographs were made in the Laboratory of Electron Microscopy, M. Nencki Institute of Experimental Biology. We are much indebted to Dr Joseph Frankel, Dr Maria Jerka-Dziadosz and Dr Norman E. Williams for their valuable advice.