Endophytic unicellular chlorophytes: a review of Chlorochytrium and Scotinosphaera

D.E. Wujek and R.H. Thompson. 2005. Endophytic unicellular chlorophytes: a review of Chlorochytrium and Scotinosphaera. Phycologia 44: 254–260. The genera Nautococcopsis and Ectogeron are reduced to synonymy with Chlorochytrium. The inclusion of Scotinosphaera and Kentrosphaera within the genus Chlorochytrium is considered now untenable. Scotinosphaera and Kentrosphaera are considered synonymous, with the name Scotinosphaera having priority. Redefinitions are given for Chlorochytrium and Scotinosphaera based on seven opposed attributes. Chlorochytrium, in germination of the zoospore, secretes an attachment disc; has fundamentally a parietal chloroplast that develops many radiate lobes, few to many pyrenoids; and the vegetative cell contains few to many minute contractile vacuoles. Its first two to four divisions are vegetative. Further division of each protoplast results in biflagellate zoospores that are walled, compressed and have a flagellar papilla. Scotinosphaera lacks an attaching disc; is strictly unicellular with no vegetative divisions; contains an axile chloroplast with many radiate arms; has characteristically one large central pyrenoid; lacks contractile vacuoles; and division results in numerous biflagellate, terete, spindle-shaped zoospores. Old cells may develop one or more localised lamellated, wart-like thickenings of the wall, externally or internally. Sexual fusion has been observed for Chlorochytrium but not for Scotinosphaera. The relationships of Chlorokoryne, Eremotyle, Nautococcus and Excentrosphaera are reviewed.


INTRODUCTION
Throughout the past half-century, the authors have had opportunities to study specimens of Chlorochytrium Cohn, Chlorokoryne Pascher, Nautococcus Korshikov, Ectogeron Dangeard, Scotinosphaera Klebs, Kentrosphaera Borzi and Excentrosphaera Moore. The taxa described in these genera are seldom reported in the literature, some only from their initial descriptions.
The genus Nautococcus was erected by Korshikov (1926). The fifth and last species he described at that time was N. constrictus Korshikov. It differed from the other four species in having a profusely radially lobed, parietal chloroplast that contained few to many pyrenoids and in having a central nucleus. The other four species were characterised by a massive axile chloroplast with a central pyrenoid and an acentric nucleus. Because of these differences, Geitler (1943) removed N. constrictus and erected the genus Nautococcopsis to hold it. Korshikov (1953), apparently unaware of Geitler's treatment of N. constrictus, also erected the genus Nautococcopsis based on the same species. Müller, also in 1953, reported on N. constrictus from Hamburg, Germany, and redescribed the species. In 1947, Dangeard erected the genus Ectogeron with the type and only species E. elodeae Dangeard. In 1952, Geitler erected the genus Eremotyle on the type and only species, E. affixa Geitler. This species was placed in synonymy with E. elodeae by Bourrelly (1966).
Except for Korshikov's description of the neustonic phase of N. constrictus, the descriptions of the species in these genera are identical. Furthermore, the alga Chlorochytrium knyanum Szymanski (1878), the life history was reported by Klebs (1881), is also considered identical with these. This report, however, identifies features distinctive for the genus Chlorochytrium, reviews its relationship to each genus and substantiates its recognition as a distinct genus. We also make comparisons with the closely allied genera Chlorokoryne, Nautococcus, Ectogeron, Scotinosphaera, Kentrosphaera and Excentrosphaera. No molecular data on this group of chlorophytes indicating their phylogenetic positions among the Chlorophyceae have yet been published.
Observations were made both from freshly collected material and from short-term cultures grown in soil-water extract or Bold's Basal Medium (Bold 1967), with additional soil water extract. Cultures no longer survive.

RESULTS AND DISCUSSION
The alga Nautococcus constrictus or Nautococcopsis constricta (Korshikov) Geitler has been collected frequently by R.H.T. and had been carried in culture. It exhibited all the characteristics of morphology and growth habit given by Korshikov (1953), Dangeard (1947) andGeitler (1943) for their genera. In the free-living condition. it develops two morphologically distinct forms depending on whether a zoospore germinates at the surface film or on a solid substratum.
At the surface film it develops the form typical of Nautococcus (Fig. 1). The compressed zoospore comes to rest with its broad side against the film. It secretes a heavier wall and produces a thin flange of wall material in contact with the surface film. This remains as a cap on the cell and functions as a flotation device as the cell body develops either above or beneath the surface. As the cell grows, it becomes pear-shaped to nearly spherical. The chloroplast in the zoospore is parietal with shallow, blunt, marginal lobes, but as it grows, it becomes thicker and begins to split into radial lobes. With enlargement of the cell, the chloroplast increases in size, the lobes elongate and their apices expand against the cell wall. This forces the inner surface of the plastid toward the centre of the cell. At the same time, the pyrenoid divides until there are few to many smaller pyrenoids scattered throughout, located mostly in the lobes. Between the lobes are few to many small, contractile vacuoles. During growth, there are vegetative divisions of the protoplast into 2-8 cells, each of which produces its own wall. Further divisions of each of these cells results in 16-32 zoosporocytes in each. At maturity, the outermost wall gelatinises and bursts to release the inner wall as a rapidly expanding vesicle containing the mass of metamorphosing zoosporocytes. These become very active within the vesicle and disperse when it bursts.
When a germinating zoospore comes to rest with its flat side against a submerged substratum, it secretes an additional wall and attaching flange. All growth of the cell is to one side. The original wall of the germling remains unchanged as a small knob at that point on the cell (Fig. 2). From this knob, the cell grows out in a narrow to broad fan-wise fashion and may be even or quite irregular with narrow marginal sinuses. As it grows, there is a spreading in the plane of the substratum, a growth that lifts the attaching flange of wall material. At the same time, a new, wide flange is also produced. Repetition of such growth results in the dorsal wall bearing the older flanges in a concentric contoured arrangement (Fig. 3). Vegetative division may occur during growth or it may not begin until the cell is near mature size. The surface of the chloroplast becomes rugose with closely compacted, radiate processes with numerous pyrenoids. In between the lobes, particularly visible at the margin, are few to many minute contractile vacuoles. By the time the zoosporocytes are fully delimited, the colour of the cell contents has changed from bright green to olive-brown to yellow-brown. The mass of zoosporocytes is extruded in a vesicle through a circular pore developed in the free, dorsal surface of the cell. The vesicle is usually lobulate and at first contains discrete masses separated into individual zoospores that soon are completely intermingled through their own activity (Fig. 4).
Zoospores are biflagellate, strongly compressed, flat on one side and convex on the opposite. They contain a parietal, en-tire to bluntly and shallowly lobed plastid with one pyrenoid. The plastid bears an anterio-lateral eyespot on the narrow side. The flagella emerge from opposite sides of a minute papilla and was described and illustrated by Dangeard (1947) for his Ectogeron elodeae and by Geitler (1952) for Eremotyle affixa. The zoospore has a thin wall and maintains its shape when it comes to rest or to germinate. Zoospores measure 5.0-6.4 m wide, 6.3-8.4 m long and 3.0-3.8 m thick. The germination knob or holdfast cap on the cell remains more or less the same size as the zoospore that produced it. Under conditions unfavourable for zoospore discharge, the zoosporocytes become aplanospores that may be dispersed later or may germinate in situ.
This sessile, flattened form of Nautococcopsis constricta is prevalent on any submerged surface whether of water plants, floating sticks and leaves or plastic sheets. Characteristically, it is present on such objects that are near the water surface, in the upper 30 cm.
A third growth form of this alga develops when a zoospore settles and germinates at a stoma or cell opening of such aquatic plants as Lemna, Wolffia, Potamogeton and Sphagnum. The germination is as described earlier with the holdfast cap being formed and remaining at the point of attachment at or in the stomatal orifice. From the germling, the cell grows through the stoma or dead cell and enlarges within the substomatal chamber or dead cell (Fig. 5). Between the germination cap and the cell body, a short or a long neck is produced that often becomes filled with lamellated material (Fig.  6). The cell aspect throughout growth and at maturity is not only that of Nautococcopsis but also exactly like that described for Chlorochytrium Cohn (1872) (Fig. 7). Upon finding such a cell within a Lemna frond, one can only determine it as C. lemnae Cohn, the type species of the genus. Morphologically then, in the neuston, the epiphytic and the endophytic habitats, N. constricta, Ectogeron elodeae and C. lemnae exhibit the morphological plasticity of one organism.
In late autumn, during and after leaf fall, immersed leaves are colonised by many sedentary algae. Germlings from Chlorochytrium zoospores released late in the season have a limited growth. Both superficial and endophytic germlings develop in much the same fashion. In limited growth, the cell enlarges away from the point of attachment and a neck is produced that is usually filled with lamellated material (Fig.  8). Growth from the point of germination and attachment is frequently irregular. These are the kind of cells that Printz (1926) made the basis for his Chlorochytrium willei Printz. Depending on the lateness in the season, such cells may become akinetes, thick-walled and green to red-orange in colour. If they have progressed to zoosporocyte production, zoospores may escape or may develop in situ. Development of germlings may be in a radiate arrangement or all the germlings may be oriented toward one pole of the mother-cell wall. In this latter case, one finds clusters of germlings stacked together with their lamellate necks erect and their broad ends against the substratum. In this condition, they resemble the tribophyte genus Chlorokoryne Pascher (1938). Furthermore the pyrenoids, expanded against the cell wall, present the appearance of discoidal plastids as in Chlorokoryne. These too may become orange-red akinetes with oil globules.
The suspicion that this depauperate condition of Chlorochytrium may have been the basis for Pascher's Chlorokoryne receives some credence when one reads in his account that reproduction was not observed, that there were few parietal chromatophores without pyrenoids and that there were red oil globules. His illustrations are perfect for the depauperate phase of Chlorochytrium.
With Chlorochytrium now involved, previous studies on this genus were reviewed. It was found that West (1904) reduced Stomatochytrium Cunningham (1887) and later (1916) reduced Scotinosphaera and Endosphaera Klebs (1881) to synonymy with Chlorochytrium. In 1917, Bristol, West's student, reduced the genus Kentrosphaera Borzi (1883) (ϭ Centrosphaera) to synonymy with Chlorochytrium. Her reasons for the reduction developed out of a detailed study of Chlorochytrium grande Bristol. She showed that the chloroplast lobes widened toward the periphery. At the wall, each expanded into a thin disc, thus presenting the false impression of a parietal plastid with an inwardly directed and attenuated process. When the discoidal expanses do not touch one another, there is the erroneous appearance of many discoidal, parietal plastids, each with an inward prolongation. This is the aspect that Klebs described for both Scotinosphaera and Endosphaera and Borzi for Kentrosphaera. It is also the same aspect and misinterpretation that led Moore (1901) to erect the genus Excentrosphaera. Moore reported each 'plastid' to have 'a minute pyrenoid'. When one reads that he based this on iodine-stained material, there is the suspicion that he mistook the scattered autochthonous starch grains for 'minute' pyrenoids.
In her next paper, Bristol (1920) reviewed the genus Chlorochytrium and tried to resolve the conflicting and inadequately described species into some order. In her summation, she listed six freshwater, four marine and three doubtful freshwater species. Except for two marine species, the descriptions afford no means of distinguishing one from another with certainty. The marine species, C. cohnii Gardner and C. moorei Gardner (Gardner 1917), are described as having quadriflagellate zoospores of two sizes. Where it is known, the other species (five) have biflagellate zoospores or gametes of the same size. Zoospore dimensions are given for three species, but descriptions of their contents and shape are virtually useless. Cell dimensions overlap completely and cell shape among the species is too variable to be of use. Nevertheless, Bristol (1917) gave the following summation of Chlorochytrium species, while three species (C. laetum Schröter, C. viride Schröter and C. rubrum Schröter) are species dubium et noncom post esse cognitum.
There are valid reasons for disagreeing with some of the generic reductions above. The reduction that initiated the point of view that culminated in Bristol's summation was that of Scotinosphaera by West (1916). Bristol's description of C. grande and her reduction of Kentrosphaera to synonymy with Chlorochytrium were natural extensions of West's action. Her detailed study and interpretation of plastid morphology in C. grande and her interpretation of Borzi's description of K. facciolaae Borzi seem eminently reasonable. Smith (1933) rerecognized Kentrosphaera on the basis of its free-living habit and lack of sexual reproduction. In doing this, he used Bristol's (1920) description and illustrations of material that she had identified as C. paradoxum (Klebs) G.S. West (ϭ Scotinosphaera paradoxa) to erect the new species K. bristolae G.M. Smith. Kentrosphaera was described as free-living, but often associated with filamentous blue-green algae. One can collect this alga intermingled with many other algae in moist or seeping areas in the soil, and it may also be taken as surface plankton, where it may project above the surface or lie parallel to it. The cells are not always free, however, and they may be found frequently and in quantity within the tissues of dead Lemna, Potamogeton and Typha culms and the floating canes of Hibiscus militaris. The cells one finds here fulfill all the features of C. grande, C. paradoxum (sensu Bristol), S. paradoxa (sensu Klebs), K. facciolae (sensu Borzi) and Excentrosphaera (sensu Moore). They have the axile, multilobate chloroplast that, from the surface, suggests numerous parietal plastids in radial arrangement. They have one or two central pyrenoids and can exist endophytically or free. West's (1916) reduction of Scotinosphaera appears to have been based solely on its reported endophytic habit and he apparently attached no importance to Klebs' description of spindle-shaped, biflagellate zoospores. Though the germination-knob or germling 'stopper', as it was termed in Bristol's papers, had been stressed as a character of Chlorochytrium, the lack of one by Scotinosphaera was not taken into account.
In the experience of the writers, the endophytism of the forms under discussion is among the least useful of taxonomic characters. The idea that these algae, particularly Chlorochytrium, are characteristically if not wholly obligately endophytic is an accrued and erroneous belief stemming from Cohn's original publication (1872) and perpetuated by the fact that this was the habitat that one looked for, found and identified Chlorochytrium. Their presence within a 'host' means no more than that a zoospore germinated there and grew into an opening in the surface, whether a stoma or an injury. The same development may occur on any dead, water-logged, floating plant tissue. In the same Lemna frond harbouring Chlorochytrium, there may be a wide variety of other algae inhabiting substomatal chambers and the intercellular spaces. In these sites, one can find such diatoms as Navicula, Rhopalodia, Nitzschia, Epithemia and Amphora in dense masses; various coccoid blue-greens; Euglena; Gymnodinium; Glenodinium; Ankistrodesmus; Scenedesmus and various other coccoid green algae; and filamentous forms such as Plectonema, Phormidium, Oscillatoria, Nostoc, Oedogonium, Stigeoclonium, Leptosiropsis, Cylindrocapsa, Coleochaete, Chaetopeltis and Aphanochaete. Cohn's fig. 5 (1872) depicts the tissues of a Lemna frond containing several cells of C. lemnae, together with Calothrix, Nostoc, Oscillatoria and a coccoid form of some other alga. All of the above algae were described from their more prevalent free-living condition, so no one should attach any significance to their endophytism beyond their ability to get into and grow in such places.
Endophytic cells referable to the genus Chlorochytrium occur predominantly in dead tissues. One can find them in living Lemna and Potamogeton leaves, especially if through wind action the leaves have been submerged for a time. In this condition, it is only the substomatal chamber that is normally invaded because the alga cannot grow through the cuticle or force its way between the walls of healthy epidermal cells. In dead leaves of Lemna and Potamogeton, however, the zoospores not only enter the stomata but are able to grow between water-softened cells of the undersurface of the leaves. Zoospores that are released within such leaves are able to disperse widely throughout the water-filled spaces. We conclude that, for Chlorochytrium, as for many other algae, aquatic plants are merely habitats, not obligate hosts. The very fact that Chlorochytrium can be carried for generations in inorganic media argues against the idea that it is an obligate endophyte. From field observations that Chlorochytrium occurs in or on substrata in the upper 30 cm of water, there is the further inference that its presence here is because of better aeration or the amount of light relative to oxygen and carbon dioxide tension.
With the above in mind and with the recognition that Chlorochytrium lemnae may be neustonic, epiphytic and endophytic and that Scotinosphaera paradoxa may be neustonic, endophytic or subaerial, the endophytic ability should no longer be considered a generic attribute.
We further believe that the separation of Scotinosphaeralike (or Kentrosphaera-like) forms from Chlorochytrium is warranted on the basis of seven distinctively contrasting features (Table 1).

Separation of Chlorochytrium and Scotinosphaera
The two genera may be redefined as follows, with Scotinosphaera having priority over Kentrosphaera.
CHLOROCHYTRIUM: Unicellular to limitedly sarcinoid, with a parietal chloroplast that becomes perforate-vacuolate and splits into many radially oriented lobes with few to many pyrenoids. The protoplast contains few to many contractile vacuoles. Cells globular, pyriform to flattened and irregular. Zoospores compressed, walled and biflagellate. The flagella issue from opposite sides of a papilla. Plastid parietal with one pyrenoid and an anterio-lateral stigma. Zoospores released in a vesicle. On germination, they secrete an attachment disc and may or may not develop a tubular 'neck' before enlarging. While naked zoospores were noted by Watanabe & Floyd (1994), we can only surmise they may have been working with contaminated or misidentified cultures. Zoosporocytes may develop into aplanospores that can germinate into new thalli. Fusion of biflagellate isogametes both within the vesicle and after release has been reported (Klebs 1881). Species may be neustonic, epiphytic or endophytic on living or dead aquatic plants or other substrata. SCOTINOSPHAERA: Unicellular, with a massive, axile chloroplast containing one or two large, central pyrenoids and divided into numerous radiate lobes that expand into thin discs at the periphery (Figs 9-11). Vegetative cells lack contractile vacuoles. Cells may become multinucleate, or nuclear divisions and cleavage into zoosporocytes may take place only upon reaching mature size. Cells linear, lanceolate or ovoid, globular or irregular. They characteristically develop localised, lamellated thickenings of the wall (Fig. 12) that may be external knobs or internal, vermiform growths (Figs 13, 14). Cells may also develop into akinetes with thick, lamellated walls. Zoospores biflagellate, spindle-shaped or lanceolate and terete in section. One parietal chloroplast with stigma and pyrenoid. Daughter cells are either extruded in a vesicle or escape singly through a pore in the wall. Asexual reproduction may be by aplanospores. Sexual reproduction unknown. Species may be neustonic, endophytic in living or dead aquatic plants or subaerial in seepage areas in soil.
Contrary to the claim presented by Punčochářová (1992) in recognising Kentrosphaera and not accepting Scotinosphaera as a valid genus, we believe that Klebs (1881), and later Bristol (1917), do present data relating to cell cycle, cell morphology and cytomorphological variability as sufficient in recognising Scotinosphaera. Care must be taken to interpret Klebs', and later Bristol's, words precisely, recognising the context in which the original work was done and the type of data that were available at the time. Even the application of the Botanical Code of Nomenclature can create problems, and commonsense may be as important as scientific strictness.
As a final note on Chlorochytrium, it should be pointed out that, if the reported observations of meiosis (Kurssanov & Schemakhanova 1927), gamete fusion and the subsequent germination of the motile zygote are correct, as well as the known production of zoospores and their germination, then Chlorochytrium has an isomorphic alternation of unicellular to sarcinoid generations and is not a diploid unicell alternating with haploid gametes, as has been stated in the literature. Also, Chlorochytrium, with its secretion of an attaching disc and its limited vegetative division, may have evolved from an attached parenchymatous ancestor.

Systematic placement
The systematic placement of Scotinosphaera remains within the Chlorococcales. The family placement of Chlorochytrium, however, presents some difficulty. Fritsch (1935) placed it in a tribe or section, Chlorochytreae, in the Chlorococcaceae of the Chlorococcales, while Smith (1950) placed it in the Endosphaeraceae within the Chlorococcales. Bourrelly (1966) placed it in the Chlorococcaceae, into which he merged the Endosphaeraceae. Lewin (1984) transferred it to the order Chaetophorales, family Chlorosarcinaceae. Melkonian (1990) suggested placing it in a new family, the Chlorochytriaceae within the Chlorococcales, but did not formally describe it. However, the family had already been recognised (see Komárek & Fott 1983). More recently, Watanabe & Floyd (1994) suggest that Chlorochytrium and Scotinosphaera (as Kentrosphaera) should be classified in the Chlorophyceae and Pleurastrophyceae, respectively. We, however, believe both are best retained within the Chlorophyceae and not split between the two classes, as also proposed in the classification by Mattox & Stewart (1984), until molecular and further ultrastructural studies their resolve taxonomic placement at the class level (also see arguments below).
The order Chlorococcales is an unwieldy, artificial conglomeration of unicellular and colonial forms that exhibit nearly every conceivable kind of diversity in cell morphology and colonial agglomeration.
Chlorochytrium, with its limited sarcinoid development (in that each cell secretes its own wall and functions independently in production of zoospores) has no place in the Chlorococcaceae. It likewise has no place in the Chlorosarcinaceae (ϭ Chlorosphaeraceae), a family of packet-forming green algae that has lost nearly all plant body organisation. Each cell divides in three planes to produce a cubical or an irregular packet of cells. With repetition of this growth, a larger mass is produced that then fragments into the smaller packets. There is zoosporic reproduction by biflagellate or quadriflagellate zoospores, depending on the genus. In all cases known, these are naked, and when biflagellate, of the Protosiphon type. None are flattened and none are walled. Furthermore the vegetative cells of algae in this family lack contractile vacuoles.
It is thus clear that Chlorochytrium has none of the attributes of the Chlorococcaceae or of the Chlorosarcinaceae, and we therefore propose that it be placed in a family of its own, the Chlorochytriaceae (G.S. West) Setchell & Gardner.
Where to intercalate this family is a new problem. As suggested earlier, it conceivably is a form derived by reduction from an attached, parenchymatous ancestor. As such, it might be placed among the Ulvales or Ulotrichales. On the other hand, when one examines the nature of the vegetative Chlorochytrium cell, one finds that its unique characteristics of a gelatinous wall (instead of a rigid, cellulosic one) and incised or fissured chloroplast with numerous small contractile vacuoles along the fissures compare favourably with the similar characteristics of Characiochloris and Characiosiphon. The zoospore of Chlorochytrium are markedly different, however. They are rigid walled, flattened, and bear a fairly large stigma. Those of Characiochloris and Characiosiphon are obconic, biflagellate, with a distinct or a minute flagellar papilla and a conspicuous elongate-oval stigma; their wall is plastic. The zoospore plastid in all three genera is parietal, shallowly lobed, and contains a single pyrenoid.
The vegetative plant body of all three becomes multicellular and, as each grows in cellularity, the original wall inherited from the zoospore is added to and expanded. The individual cells have either a plasmalemma only or also a delicate external sheath. In Chlorochytrium, the sheaths of the individual cells are thicker but gelatinous in texture. This genus likewise differs in that the outer zoosporic wall is not added to and expanded but functions either in attachment or as a flotation cap when neustonic. At germination, the zoospore secretes an additional sheath that expands as the cell grows beneath or away from the zoosporic sheath. With further growth, this second sheath is stretched and ruptured as a new one is secreted within it. Thus, Chlorochytrium, if it does have a relative affinity with Characiochloris and Characiosiphon, exhibits greater reduction in size and cellularity but has, to a greater degree, a residual retention of an ancestral wall formation.
If we accept these similarities as significant, then the Chlorochytriaceae may well be placed with the Characiosiphonaceae in the Characiosiphonales, an order of unknown affinity within the Chlorophyta. However, Buchheim et al. (2002), using molecular techniques, suggest that the Characiosiphonaceae form 'a clade near the base of the ''Dunaliella'' group within the chlamydomonad lineage'. Further molecular and ultrastructural studies may resolve this taxonomic problem.