|Home | About | Journals | Submit | Contact Us | Français|
Maryna umbrellata (Gelei, 1950) Foissner, 1993 is a colpodid ciliate common in ephemeral water bodies. Pure cultures were established and the resting cyst studied by light and electron microscopy, protease digestion of thin sections, and various cytochemical reactions shown by colour micrographs. The cyst of M. umbrellata belongs to the kinetosome-resorbing (KR) type and has a conspicuous glass cover described by Foissner et al. (2009). It is 100 μm across and the about 13 μm thick wall, which amounts for half of the total cyst volume, consists of four distinct layers: (i) the about 6 μm thick pericyst, which is composed of glass granules embedded in mucoproteins, and of a basal layer consisting of glycogen tubules about 20 nm across; (ii) the about 140 nm thick, electron-dense ectocyst, which is not digested by protease; (iii) the about 6 μm thick mesocyst, which is distinctly laminated and made of proteins; and (iv) the thin, about 500 nm thick endocyst, which is structureless and consists of glycoproteins. The meso- and endocyst are highly elastic, reducing their diameter by 50% when the encysted ciliate is removed. The cyst contents is dominated by up to 7 μm-sized “spongy globules” consisting of an electron-dense, proteinaceous matrix burrowed by electron-lucent strands of glycogen, providing the cyst with a curious, white-spotted appearance in the electron microscope. The small lipid droplets have a proteinaceous centre. The cyst plasm contains proteins and unstructured mucosubstances, which stain with alcian blue, and thus possibly originate from the decomposed mucocysts. This study suggests that ciliate resting cysts are much more diverse than indicated by literature data.
The ability to form a dormant stage, usually called cyst or spore, provides micro- organisms with an endless life, while specific properties of the cyst wall are very likely important for the biogeographic range a species or population can occur (Foissner 1987, 2006, Gutiérrez et al. 2001). Specific cyst properties must also strongly influence the occurrence or absence of a certain species under certain environmental conditions. It is thus surprising that current protistology shows only meagre interest in cyst research, although some progress has been made by ecologists (e.g., Jonsson 1994, Kim and Taniguchi 1997, Müller and Wünsch 1999, Müller et al. 2002, Endo and Taniguchi 2006, Weisse et al. 2008), biochemists (e.g., Nakamura and Matsusaka 1992, Watoh et al. 2003, Tsutsumi et al. 2004, Yamasaki et al. 2004, Akematsu and Matsuoka 2005, Oda and Matsusaka 2005), and molecular biologists (e.g., Gutiérrez et al. 2001, Gallejas and Gutiérrez 2003, Sugimoto and Endoh 2008).
The species investigated in the present study belongs to the order Colpodida, a group of ciliates where cysts have been extensively studied because some species of the genus Colpoda are easily cultivated and very common in terrestrial habitats, such as mosses, grassland puddles, and soil (for reviews see, Wagtendonk 1955, Corliss and Esser 1974, Meier-Tackmann 1982, Foissner 1993, Gutiérrez and Martin-González 2002, Gutiérrez et al. 2003). However, much of the morphological work is rather incomplete, for instance, a clear semischematic presentation of the cyst wall and its genesis is not available for any Colpoda species.
We have shown in Meseres corlissi, an oligotrichous ciliate, which has chitin in the cyst wall, highly complex organic scales (lepidosomes) on the cyst surface, and five types of cyst wall precursors, that the knowledge on ciliate resting cysts is very limited (Foissner 2005; Foissner et al. 2005, 2006; Foissner and Pichler 2006). This is sustained by studies on the cysts of Opisthonecta (Calvo et al. 2003) and Maryna umbrellata, the species investigated in the present study: it produces silicon (glass) granules in the trophic state and deposit them on the surface of the cyst (Foissner et al. 2009), the meso- and endocyst show a high elasticity, the pericyst contains a layer of glycogen tubules, and the cyst contents is studded with curious storage bodies providing the cyst with a unique, white-spotted appearance in the electron microscope. Further peculiarities occur during cyst wall genesis, which will be described in a forthcoming paper. For instance, the ectocyst precursors are released via the parasomal sacs of the kinetids.
Maryna umbrellata is a planktonic, rapidly swimming ciliate (Figs 4, 5), which reproduces in division cysts formed on the bottom of the cultivation dish. Inspection of the food vacuoles and crude experiments with potential food sources (heterotrophic and autotrophic flagellates, small ciliates) showed that M. umbrellata fed mainly on bacteria, while flagellates were taken sparsely; ciliates were rejected. Thus, about 20 specimens were isolated to set up a pure culture with French table water (Eau de Volvic) and some crushed wheat kernels to stimulate growth of the natural bacterial flora. Cultures were grown at room temperature and were renewed each a week by transferring about 20 cells and a decaying wheat kernel into fresh medium.
The species was determined by live observation (bright field and interference contrast), silver impregnation, and scanning electron microscopy, as described by Foissner (1991) and shown in Figures 4 and 5.
Maryna umbrellata was insufficiently known when Foissner (1993) revised the genus. Thus, Foissner et al. (2002) redescribed M. umbrellata from Costa Rican and Australian populations. However, later investigations (this paper) showed that the resting cysts of the Austrian and Australian populations were rather different, suggesting the latter as an undescribed species. This matter will be treated in a separate publication. For the present study it is sufficient to know that the Austrian population represents the “true” M. umbrellata, as described by Gelei (1950) and reviewed in Foissner (1993).
After the logarithmic growth phase, when few dividing cells occurred, most specimens encysted rapidly, possibly due to the accumulation of excretory products produced by both the ciliates and the bacteria. Cysts were morphologically fully developed and thus possibly mature after one to two days. To be at the safe side, all investigations were performed on cysts older than 10 days.
Cysts were studied in vivo and various preparations, using bright field and interference contrast microscopy. In vivo measurements were conducted at magnifications of × 100–1000. Nine cysts were sectioned for transmission electron microscopy, and four of them were studied in detail. The images were digitalized and improved electronically.
For scanning electron microscopy (SEM), cysts were fixed in cacodylate-buffered osmium tetroxide (2% w/v) at pH 6 for 30 min., rinsed and cleaned with tap-water, and then transferred to a special preparation chamber, where they were dehydrated with a graded ethanol series and critical-point dried (Foissner 1991). Finally, specimens were mounted on a SEM-stub and gold sputtered. Observations were made with a Cambridge Stereoscan 250 (Cambridge Instruments, Cambridge, United Kingdom) operated at 20 kV. For transmission electron microscopy (TEM), resting cysts were fixed for 30 min. in a mixture of 10 ml glutaraldehyde (25% v/v), 6 ml aqueous osmium tetroxide (2% w/v) and 10 ml aqueous, saturated mercuric chloride. After three washes in tap-water, the specimens were transferred to glycid ether 100 (Epon 812, Serva, Heidelberg, Germany) via a graded ethanol series and propylene oxide. Flat embedding in aluminium weighing pans allowed specimens to be investigated light microscopically (up to × 400) to select well-pre-served individuals. Ultrathin sections were cut with a diamond knife mounted on a Reichert ultracut. Sections were viewed in a Zeiss EM 910 (Oberkochen, Germany), either unstained or, as usual, after staining with uranyl acetate followed by lead citrate (UL staining). Further, a series of sections was stained with bismuth subnitrate, as described by Hayat (1989). This method, which reveals polysaccharides, is referred to “bismuth stain” throughout the paper. One micrometer thick semithin sections were mounted on microscope slides by heat (120°C for 30 min.).
The cytochemical investigations were performed on unfixed cysts and/or cysts fixed with ethanol (95% w/v) or formalin (10% w/v), using the protocols and controls described in the literature cited below. Except for reactions on semithin sections and protease digestion of ultrathin sections, all tests were made on “free,” complete cysts pipetted on a microscope slide or handled with the centrifuge (2000 r/min. for 30 s; 500 g). Stained, centrifuged cysts were used either in wet condition and/or prepared with the ordinary histological technic to permanent slides (embedding in albumen-glycerol → 100% ethanol → xylol → resin).
As the cysts were about 100 μm in size and the wall was at least 10 μm thick, most investigations were performed on wet cysts slightly to strongly squeezed and flattened between microscope slide and coverslip. This reveals the cyst structures very well and fosters the photographic documentation.
Unfixed cysts were tested for chitin with Van Wisselingh's method, as described by Foissner et al. (2005), applying KOH-treatment 15 min. and 30 min.; fly wings served as control. The iodine tests were performed on unfixed cysts on microscope slides, using Lugol's solution and chlorinated-zinc-iodine (Romeis 1968). The classic Periodic Acid-Schiff reaction (PAS after Mc Manus) and the Bauer technique, both described in Romeis (1968), were performed on cysts fixed with 10% formalin and 95% ethanol, respectively, and on semithin sections made from cysts prepared for transmission electron microscopy. All methods gave similar results; Bauer's technique worked better when treatment with chromic acid was increased from 60 min. to 100 min. Aqueous, filtered alcian blue (1%, pH 5) was applied to trophic specimens and unfixed cysts on microscope slides. At the ultrastructural level, polysaccharids were revealed by bismuth staining, as described in Hayat (1989).
These were revealed in formalin-fixed cysts with the acrolein-Schiff method, as described by Pearse (1968); starch grains from the culture medium served as a control. At the ultrastructural level, protease (from Streptomyces griseus; Sigma, Japan) was applied to unstained ultrathin sections, using the following protocol: Grids with sections were treated with H2O2 (5%) for 10 min. at room temperature and rinsed three times, 10 min. each, with distilled water. Then, the grids were incubated at 37°C in protease (0.3% in 0.05 M Na-cacodylate buffer) for 25 min. (“mild” digestion) or 50 min. (“strong” digestion). Finally, the grids were washed one time in distilled water for 10 min. at room temperature. The air-dried sections were studied without staining. This simple method provides valuable data both on the cyst wall and the cyst contents.
These were revealed by sudan black (1% in 70% ethanol) applied to unfixed cysts on a microscope slide and in protease-treated sections as described above.
Hydrofluoric acid (HF) treatment, scanning electron microscope assisted X-ray microanalysis (EDAX), and energy filtered transmission electron microscopy (EFTEM) revealed that the peri-cyst of M. umbrellata contains countless silicon granules (Foissner et al. 2009). In the present paper, only HF-treated ultrathin sections are shown.
The resting cyst of M. umbrellata is difficult to study in vivo and in whole mounts because of its large size and thick wall, whose components are more or less superimposed when the focal plane is in the cyst centre (Figs 6, 10, ,14).14). On careful observation, most cyst layers recognizable in the transmission electron microscope can be seen also in the light microscope, especially when the cyst is squashed.
The resting cysts of M. umbrellata are in the size of the trophic specimens (Table 1): 102 × 102 μm and 105 × 96 × 96 μm, corresponding to volumes of about 555 366 μm3 and 504 792 μm3, respectively. The cysts are spherical to globular, those from environmental specimens are slightly larger (109 μm) than those from specimens cultivated for a month (102 μm); the latter shrink to 81 μm when prepared for transmission electron microscopy. The cysts are dark at low magnification (× 40) due to their strong light refractivity, and yellowbrown at higher magnification (≥ × 100) due to the brownish pericyst. The cyst surface is smooth to rather distinctly facetted, both in environmental and cultivated specimens (Figs (Figs1,1, 6, 7a, 14, 23–25). The number of facetted cysts increases with cyst age and in partially or completely dehydrated preparations (Figs 17, 19, 23–25, 31, 33). Further, the distinctness of the facets is influenced by cultivation, i.e., in some cultures most cysts are smooth or slightly facetted, while in other cultures most are slightly to distinctly facetted.
Comparative measurements of cyst wall thickness in vivo and in TEM sections reveal a substantial difference (Table 1): 8.3 μm vs. 13.3 μm. The high coefficients of variation indicate that part of this difference is caused by sections not being near the cyst centre, i.e., the wall becomes the thicker the more the section is outside the cyst centre. A second reason might be that the measurements are not based on the same material: the cysts used for TEM were from specimens cultivated for about one month, while those used for the in vivo measurements were from specimens cultivated over one year. Very likely, wall thickness decreased due to the long cultivation.
Thus, the notions on the thickness of the individual cyst layers must be considered in the light of this problem. However, there is no doubt that the cyst wall of M. umbrellata is extraordinarily thick, viz., about 10 μm and thus makes up almost 50% of the total cyst volume (Fig. 53)!
The about 6 μm thick pericyst is one of the most conspicuous features of the cyst of M. umbrellata because it contains countless glass granules 0.5–2 μm across (Foissner et al. 2009). The pericyst is brownish and membranous because it consists of a very flexible, fibrous matrix in which the glass granules are embedded (Figs 1, 2, 6–12; Table 1). The glass granules as well as adhering environmental bacteria and mud make the cyst surface rough in the scanning electron microscope (Fig. 7). The structural density of the pericyst decreases distally, and its thickness decreases from about 5–8 μm to 3–5 μm in old cysts and after prolonged cultivation of the population (see previous chapter). The pericyst consists of three distinct structures (from distal to proximal): the mucous cover, which contains countless glass granules, and the basal layer.
(Figs 1–3, 6–12, 32, 34--36;36; Table 1). The glass granules have been investigated by Foissner et al. (2009), using hydrofluoric acid digestion (Fig. 35), SEM- assisted X-ray microanalysis, and energy filtered transmission electron microscopy. The granules, which consist of biomineralized, amorphic silicon, have an average size of 819 × 630 μm and are firmly anchored in the slime layer (Tables (Tables1,1, ,22).
(Figs 1–3, ,9,9, 14, 15, 19, 20, ,36;36; Tables Tables1,1, ,2).2). The mucous cover is about 4 μm thick (Table 1) and composed of very fine fibres (< 10 nm) embedding the glass granules (Figs 1–3, ,9,9, ,36).36). The fibres form a flocculent, weakly osmiophilic mass.
The mucous cover stains deeply with alcian blue (Figs 14, 15; Table 2) and rather distinctly with acrolein (Figs 19, 20; Table 2), suggesting that it consists of acid mucopolysaccharides and proteins (mucoproteins).
(Figs 2, 3, ,11,11, 15, 23–25, ,3131-36, 38–40, ,49;49; Tables Tables1,1, ,2).2). The basal layer is attached to the distal surface of the ectocyst and passes into the mucous cover without sharp border. It has an average thickness of 1.7 μm (Table 1), stains ordinarily with UL and bismuth, and is composed of tortuous, strongly intertwined tubules with an average diameter of 20 nm (Figs (Figs3131-36, 38–40; Table 1). The tubules are densely packed in the proximal half of the layer, often forming a fibrogranular, slightly laminated mass (Figs (Figs32,32, 36, 38).
The basal layer, which is not digested by protease (Figs 33, 34, ,49),49), stains rather intensely with PAS in semithin sections (Fig. 25; Table 2), suggesting that it consists of glycogen. In whole cyst mounts, the PAS reaction is lacking or weak, depending on the time of chromium acid treatment (Figs 23, 24).
The ectocyst has an average thickness of 140 nm and is thus not identifiable with certainty in the light microscope (Figs (Figs2,2, ,11,11, ,15);15); actually, it is the thinnest and most electron-dense layer of the cyst wall of M. umbrellata (Table 1). In the transmission electron microscope, it is structureless, even at high magnification (Figs 36, 38).
The composition of the ectocyst could be not clarified because it is too thin to be identifiable with the histochemical methods applied. Very likely, the ectocyst is not proteinaceous because it is not digested by protease (Figs (Figs34,34, ,49).49). As lipids are unlikely, it possibly consists of carbohydrates.
When focused on the cyst centre, most of the about 6 μm thick mesocyst is hidden by the thick and comparatively massive pericyst (Figs 1, 2, 6, 10, ,14;14; Table 1). The endocyst, which is only about 500 nm thick, is distinguishable from the mesocyst only in strongly squashed cysts (Figs (Figs2,2, 8, 11). Both are bright and structureless and appear solid in the light microscope. However, when the cyst is opened by pressing the coverslip slightly with a needle, the meso- and endocyst show a surprising property: they detach from the ectocyst and shrink to about half the cyst diameter (Figs (Figs8,8, ,28).28). Frequently, the mesocyst splits during this procedure: the distal half remains attached to the ecto- and pericyst, while the proximal half and the endocyst shrink as described before (Figs 8, 11, 15, 22).
The mesocyst is the thickest layer of the cyst wall (Table 1). In the transmission electron microscope, it consists of an average of 24 alternating layers with different electron density and fibro-granular fine structure. The thickness of the layers is different in the nine cysts analysed, and their distinctness increases in bismuth or hydrofluoric acid-treated sections (Figs 31, 32, 35, 37, ,4141).
The mesocyst stains with acrolein (Fig. 19) and is digested by protease (Figs 33, 34, 43, 49), showing proteins as a main component (Table 2). It does not react with iodine (Figs 18, 22), alcian blue (Fig. 15), PAS (Figs 24, 25), and sudan black (Fig. 28).
This part of the cyst wall has an average thickness of 272 nm and is thus not identifiable in the light microscope (Table 1). In the electron microscope, it is structureless and bright (Figs (Figs3,3, ,37).37). Frequently, this zone is hardly recognizable, i.e., not distinguishable from the mesocyst and endocyst, to which it might belong because the endocyst is thicker in the light than in the electron microscope (Figs (Figs11,11, 24, 25, ,32,32, ,37,37, ,41).41). The bright zone disappears in protease-treated sections because the mesocyst is digested and then appears as bright as the bright zone proper (Figs 33, 34, ,4949).
The endocyst has an average thickness of 513 nm and is thus difficult to recognize in the light microscope (Fig. 11; Table 1). It is fairly distinctly separated from the mesocyst by the bright zone, which might belong to the endocyst (see above); proximally it is separated from the cortex of the encysted cell by a very thin, electron-lucent cleft (Figs (Figs32,32, ,41).41). A definite fine structure is lacking, i.e., it appears moderately dense and very finely granular (Figs (Figs3,3, ,32,32, ,37,37, ,4141).
In semithin sections, the endocyst stains with PAS, suggesting carbohydrates as a main component (Fig. 25). In whole cyst mounts, there is also a reaction with acrolein (Fig. 19), indicating some protein (Table 2). This is sustained by the observation that it appears often considerably brighter in protease-treated sections (Figs (Figs34,34, 43, 49). Thus, the endocyst is possibly a glycoprotein.
In the light microscope, the following cyst inclusions are recognizable: the nuclear apparatus, “spongy globules” 2–7 μm across, lipid droplets ≤ 2 μm in size, and minute “crystals” distinctly sparkling under interference contrast illumination. Autophagous vacuoles were not seen neither in the light- nor in the electron microscope.
The nuclear apparatus is in or near the cyst centre (Figs (Figs1,1, ,6).6). No distinct changes are recognizable when compared with that of trophic specimens. However, it is covered with brownish “crystals” about 1 μm in size (Fig. 6). These “crystals,” which are also scattered in the cyst plasm, disappear after hydrofluoric acid treatment, suggesting that they consist of glass (Foissner et al. 2009). However, the electron microscopical investigations show an absence of silicon granules in the mature cyst plasm. Thus, the nature of the “crystals” remains obscure.
(Figs 1–3, 10, 11, 18, 19, 21, 24, 26, 27, 29, 31, 33--35,35, 42–52; Tables Tables1,1, ,2).2). These are the most conspicuous cyst inclusions because they are very numerous and up to 7 μm across. They have a moderate light refractivity and often contain some granular material (Figs 1, 2, 6, 10, 11, 15, 21, 26, 29; Table 1). The spongy structure of the globules becomes distinct in cysts stained with Bauer's PAS, where they swell due to the chromic acid treatment (Fig. 46).
The spongy globules provide the mature cyst with a unique, white-spotted appearance in the transmission electron microscope (Fig. 53). In thin sections, the globules have an average size of 4.3 × 3.8 μm and show an average of 29 bright areas (“unstained holes”) each with a size of about 600 nm (Table 1). Thus, a spongy globule has an area of about 13 μm2, of which 8 μm2 are occupied by the bright areas. The size of the globules and bright areas is highly variable, showing coefficients of variation between 24% and 36% (Table 1). Further, these globules are stable structures present also in cysts older than one month.
The complex fine structure of the spongy globules becomes recognizable when various stains and protein digestion are applied. Basically, the globules are membrane-bound vesicles consisting of an electron-dense matrix with scattered, small bright areas (Figs (Figs3,3, ,31,31, ,35,35, 42, 44). In UL-stained sections, the matrix usually appears as a homogenous, finely granular mass (Figs 42, 44), while bismuth staining reveals polygonal, membrane-bound subunits, most having a bright area in the centre (Fig. 45). The bright areas appear structureless in UL-stained sections (Fig. 44), while conspicuously foamy in bismuth stains (Figs 45, 48).
The spongy globules react to carbohydrate stains (iodine, Figs 18, 21; PAS, Figs 23–27; bismuth, Fig. 45; Table 2), protein stains (acrolein, Fig. 19) and, especially, to protease (Figs 33, 34, 43, 49–52). Protease digests the globules, except of the external and internal membranes and the unstained areas which now stand out as greyish spots from the whitish background (Figs 33, 34, 43, 49–52). These properties show that the spongy globules consist of glycoproteins in a special arrangement: the matrix is composed of proteins and burrowed by strands of glycogen (Table 2).
(Figs 2, 3, 10, 11, 28–30, ,35,35, 42, 44, 45, 47, 49, 51, 52; Tables Tables1,1, ,2).2). These inclusions, which are very numerous, have an average size of 1161 × 1049 nm and are scattered through the cyst plasm (Table 1). In the light microscope, the lipid droplets have a bright shimmer (Figs (Figs2,2, 10, 11, ,29).29). They are rather deeply stained with UL and bismuth, but are structureless in the transmission electron microscope (Figs (Figs35,35, 42, 44, 45).
In vivo, the lipid droplets stain deeply with sudan black, showing the presence of lipids (Figs 28–30; Table 2). This is in accordance with their appearing in the transmission electron microscope (see above). However, protease digestion reveals a more complex structure: the droplets are composed of proteins covered with a rather thin lipid layer (Figs 49, 51, 52; Table 2).
(Figs 16, 18, 22, 27, 42, 44, 45, 48–52; Table 2). The resting cyst of M. umbrellata is studded with comparatively large glycoprotein globules (~ 5 μm) and minute, lipid-covered protein droplets (~ 1 μm). Thus, few space remains for other inclusions (e.g., mitochondria) and fluid cytoplasm, which appears bright or greyish in UL-stained sections (Figs 42, 44). However, when stained with bismuth, the cytoplasm shows a foamy structure highly similar to that of the bright areas in the spongy globules (Figs 45, 48). Thus, the fluid cytoplasm contains unstructured polysaccharids. This would agree with the light microscopic observations on squashed cysts, which release spongious material stainable with alcian blue (Fig. 16) and a fluid becoming deep orange when reacting with iodine (Figs 18, 22) and red in PAS stains (Fig. 27). Further, the plasm becomes lighter in sections treated with protease, showing the presence of proteins (Figs 49–52; Table 2).
Any discussion on ciliate resting cysts is limited by the simple fact that very few detailed studies are available on the intraspecific variability and on the interspecific diversity, i.e., the cysts of less than 40 species have been investigated in some detail (Gutiérrez et al. 2003). The intraspecific and environmental variability can be considerable, as shown by Weisse et al. (2008) in Meseres corlissi and by Foissner et al. (2007) in Halteria grandinella, a common cosmopolite which, however, very likely consists of several sibling species (Katz et al. 2005). Our overall ignorance on ciliate resting cysts is best demonstrated by the present and a former study (Foissner et al. 2009), where glass has been reported for the first time as a main component of the cyst wall.
The family Marynidae contains four genera with a total of 24 species (for reviews, see Foissner 1993 and Foissner et al. 2002). Electron microscopical investigations on their cysts are lacking, but light microscopical observations are available from five species and were reviewed by Foissner (1993) and Foissner et al. (2002). The cyst of M. galeata is dull red and covered with minute sand grains. The cyst of M. atra is possibly similar to that of M. umbrellata. The cysts of M. umbrellata from Europe (described here) and Australia (described in Foissner et al. 2002) are possibly quite different. The latter has an only 1 μm thick meso- and endocyst, while the slimy pericyst contains up to 5 μm-sized globules similar to those found in some Colpoda species (see below). Such globules cover also the cyst of Mycterothrix tuamotuensis, while the cyst of M. erlangeri is covered by discs and tri-armed structures reminiscent of spongial “tri-rays.” When the present data are added, a considerable diversity of the marynid resting cysts becomes likely. Possibly, this diversity is related to their habit, ranging from ephemeral ponds to permanent waters, from freshwater to highly saline inland waters, and from ordinary to highly saline coastal and desert soils (for reviews, see Foissner 1993 and Foissner et al. 2002).
The genus Colpoda belongs to the family Colpodidae which consists of eight genera with a total of about 50 species, of which we have light microscopical data on the resting cysts from 21 species (for reviews, see Foissner 1993, Foissner et al. 2002). These species show at least 12 (!) main cyst types. Unfortunately, electron microscopical investigations are available only from C. magna, C. minima, C. lucida, C. cucullus, C. inflata, and C. steinii.
A comparison of the resting cysts of the Marynidae and Colpodidae is appropriate because these families are closely related, both genetically (Dunthorn et al. 2008) and morphologically (Foissner 1993), differing mainly in the location of the oral apparatus: in anterior (Colpodidae) vs. posterior (Marynidae) half of body. However, such comparison is difficult because the data are often incomplete and the interpretation of number and kind of cyst layers varies considerably. For instance, previous researchers did not distinguish between pericyst and ectocyst but added the former to the latter. Further, species were sometimes misidentified and previous light microscopical data neglected in the interpretation of the electron microscopical findings. Thus, my comparison is strongly influenced by the investigations on M. umbrellata and the re-interpretation of previous data.
Comparison with C. magna, as described by Foissner (1993) and Frenkel (1994). Basically, the cyst is similar to that of M. umbrellata, except for the following features: (i) The slimy pericyst is inconspicuous and lacks a glass cover, but possibly has a basal layer. (ii) The mesocyst tends to split into several distinct sheets, especially in old cysts. Although the mesocyst of M. umbrellata has many layers of different electron density, it never splits, not even in cysts older than three months. (iii) The somatic basal bodies are retained, while resorbed in M. umbrellata. (iv) The cyst volume (448,693 μm3) is distinctly lower than the volume of the morphostatic specimens (992,868 μm3), while similar in M. umbrellata.
Comparison with C. minima, as described by Foissner (1993) and Diaz et al. (2003). This species is difficult to distinguish from C. magna, both genetically (Dunthorn et al. 2008) and morphologically, including the light microscopical structure of the resting cyst (Foissner 1993). However, at electron microscopic level, the cysts of C. minima and C. magna appear rather different, although the data of Diaz et al. (2003) are very incomplete and possibly from young, not yet mature cysts. Compared to M. umbrellata, the cyst of C. minima (i) lacks a distinct pericyst, except of a thin slime cover; (ii) probably has a much thicker ectocyst; (iii) lacks any lamination of the mesocyst (Diaz et al. 2003), which splits into several distinct sheets in old cysts (Foissner 1993); and (iv) the cyst volume (220 781 μm3) is rather distinctly lower than the volume of the morphostatic specimens (331 488 μm3). No data are available whether or not the somatic basal bodies are resorbed.
Comparison with C. cucullus, as described by Janisch (1980) and Foissner (1993). Possibly, only Janisch (1980) identified the species correctly, while other C. cucullus populations belong either to C. lucida or C. inflata (see these species below). According to the investigations of the authors mentioned above, the cyst walls of M. umbrellata and C. cucullus differ in the following features: (i) pericyst complex vs. a simple mucous layer; (ii) mesocyst stable vs. splitting in several distinct sheets in old cysts; (iii) somatic cilia and basal bodies resorbed vs. retained (Foissner 1993, p. 155); (iv) cyst volume equal vs. slightly lower (50 939 μm3) than in morphostatic specimens (73 267 μm3).
Comparison with C. lucida, as described by Kawakami and Yagiu (1963) and Foissner (1993). The Japanese authors identified their population as C. cucullus, according to the state of art. However, C. cucullus was split into two species by Foissner (1993; and unpubl. data): C. cucullus (with inconspicuous mucocysts, without cyst lepidosomes, mesocyst split in old cysts) and C. lucida (with conspicuous mucocysts and cyst lepidosomes with hexagonal fine structure, mesocyst does not split). The cyst wall of C. lucida is similar to that of M. umbrellata but thinner, while the ectocyst is comparatively thick (see Fig. 7 in Kawakami and Yagiu 1963). Thus, the main difference is the pericyst, which contains glass granules in M. umbrellata, while lepidosomes in C. lucida. The cyst volume is conspicuously lower (38,772 μm3, 124 725 μm3 when the pericyst is included) than that of trophic cells (214 776 μm3). No solid data are available whether or not the somatic basal bodies are resorbed.
Comparison with C. inflata, as described by Foissner (1993) and Foissner (unpubl. TEM data). This species has, like C. lucida (see above), lepidosomes consisting of fine, highly intertwined tubules. Thus, the C. cucullus population studied by Chessa et al. (2002) could be a C. inflata. The population investigated by Martin-González et al. (1992a, b) lacks lepidosomes and thus cannot be C. inflata; a scanning micrograph (Fig. 1 in Martin-González et al. 1992b) indicates that it could be C. maupasi. Considering these problems, I briefly describe the cyst wall of C. inflata according to my unpublished data. The cyst wall of C. inflata is similar to that of M. umbrellata but thinner and differs in the following features: (i) pericyst with lepidosomes vs. glass granules; (ii) mesocyst without vs. with lamination; (iii) cyst volume (4846 μm3) conspicuously lower than that of trophic cells (31,250 μm3) vs. about equal. No data are available whether or not the somatic basal bodies are resorbed.
Comparison with C. steinii, as described by Tibbs (1968), Ruthmann and Kuck (1985), and Foissner (1993). This minute species seems to have only two cyst layers: a thin, dense ectocyst and a moderately dense meso- or endocyst. However, the data of Tibbs (1968, Fig. 16) and Ruthmann and Kuck (1985, Fig. 4) can be interpreted also differently: ectocyst very thin, sloughs of membranous material forming the pericyst; a thin, compact mesocyst; and a thick, less dense endocyst. According to the data of the authors mentioned above, the cyst walls of M. umbrellata and C. steinii differ in the following features: (i) pericyst with glass granules vs. membranous material; (ii) meso- or endocyst possibly lacking in C. steinii; (iii) somatic cilia and basal bodies resorbed vs. retained; (iv) cyst volume equal vs. distinctly lower (2144 μm3) than that of morphostatic specimens (6782 μm3).
The critical analysis of the data shows a basic similarity of the cyst wall of M. umbrellata and Colpoda spp., possibly except of C. steinii: all have a pericyst; a very thin, dense ectocyst; a thick, less dense mesocyst which may split into several sharply contoured sheets in Colpoda spp.; and a thin, moderately dense endocyst. Most of the cyst wall diversity is associated with the pericyst, i.e., the most external layer: glass granules embedded in slime and a tubular basal layer in M .umbrellata; organic scales (lepidosomes) embedded in slime in C. lucida and C. inflata; membranous material in C. steinii; and a more or less thick, simple slime layer in C. magna, C. minima, and C. cucullus. A further main difference concerns the somatic basal bodies: they are resorbed in M. umbrellata, while retained in several (C. magna, C. cucullus, C. steinii) or even all Colpoda species. Maryna umbrellata is also curious in that the cyst volume is about the same as in morphostatic specimens. However, the same is found in, e.g., C. cucullus (see above) and two oligotrich ciliates, Meseres corlissi (Foissner et al. 2005) and Halteria grandinella (Foissner et al. 2007), when the pericyst is included in the calculations. Obviously, there is great diversity also in this feature, making it difficult to establish rules (see the “quantum” hypothesis of Ricci et al. 1985).
The data available on cytochemistry and biochemistry of ciliate cyst walls were excellently reviewed by Gutiérrez and Martin-González (2002) and Gutiérrez et al. (2003). They concluded, inter alia, that proteins, glycoproteins and carbohydrates are the main wall constituents. This is in accordance with the results from M. umbrellata: the pericyst consists of acid mucoproteins, the ectocyst possibly of carbohydrates, the mesocyst mainly of proteins, and the endocyst of glycoproteins (Figs 14–25; Table 2).
However, data on the composition of the individual cyst layers are lacking in colpodids or match the scheme mentioned above. More detailed investigations are available from stichotrichine hypotrichs, such as Oxytricha and Gastrostyla, and were collated by Delgado et al. (1987). Although there is a considerable diversity, some similarities and differences to M. umbrellata are recognizable. In both, the pericyst and ectocyst often consist of acid mucoproteins, while the mesocyst is usually composed of neutral and acid mucosubstances, rarely of proteins, as in M. umbrellata. The endocyst consists of glycoproteins in M. umbrellata (Table 2), in various Colpoda species, and in most or all stichotrichine hypotrichs (Delagado et al. 1987, Gutiérrez et al. 2003). Silicon (glass) is as yet unique to M. umbrellata (Foissner et al. 2009).
The cyst contents has been rarely investigated, except of the nuclear apparatus and the autophagosomes (Gutiérrez and Martin-González 2002, Gutiérrez et al. 2003). Usually, only (para)glycogen inclusions are mentioned. More recently, Foissner (2005) and Foissner et al. (2007) reported “curious” cyst inclusions in Meseres and Halteria, two oligotrich ciliates, and speculated that they originated from the autophagosomes. These “curious structures” resemble the “spongy globules” of M. umbrellata, which consist of proteins and polysaccharides and are thus obviously a kind of storage product. It is strange that such composite storage bodies have been not reported from other ciliates.
Other conspicuous structures never reported before from ciliate cysts are the lipid-covered protein globules and a mass of alcian blue positive substance in the cyst plasm. Very likely, the latter material originates from the mucocysts, which are very numerous in the trophic specimen (Fig. 13) and decomposed during encystment (Foissner, unpubl.).
When the cyst contents of M. umbrellata is cautiously squashed out, the pericyst and the ectocyst keep their sizes, while the diameter of the mesocyst and the endocyst shrinks by about 50%, without forming wrinkles (Figs 8, 11, ,28).28). When the mesocyst splits, only the proximal part shrinks, while the distal portion remains attached to the ectocyst showing its inelasticity (Fig. 8). The elasticity of the endocyst is well known (for a review, see Gutiérrez et al. 2003), while that of the mesocyst seems to be a speciality of M. umbrellata. I cannot ascribe this feature a definite function but it shows again the high diversity of ciliate resting cysts. This conclusion is sustained when one compares detailed studies of cysts from, e.g., colpodids (Maryna), some oligotrichs (Foissner and Pichler 2006; Foissner et al. 2005, 2006, 2007, 2009), a peritrich (Calvo et al. 2003), and some hypotrichs (Walker and Hoffman 1985, Delgado et al. 1987, Foissner and Foissner 1987).
Financial support was provided by the Austrian Science Foundation, FWF project P20360-B17. The technical assistance of Mag. Birgit Weissenbacher, Mag. Gudrun Fuss, Mag. Barbara Harl, Robert Schörghofer, and Andreas Zankl is greatly acknowledged.