- Email: [email protected]
On the cytology of Jirovecia caudata (Léger and Hesse, 1916) (microspora, bacillidiidae)
European Journal of
Europ.j.Protistol. 25, 321-330 (1990) June 29, 1990
PROTISTOLOGY
On the Cytology of Jirovecia caudata (Leger and Hesse, 1916) (Microspora, Bacillidiidae) J. I. Ronny Larsson Department of Zoology, University of Lund, Sweden
Summary The cytology of a microsporidium identified as Jirovecia (Mrazekia) caudata (Leger & Hesse, 1916), the type species of the genus Jirovecia Weiser, 1977, is described with emphasis on the ultrastructure of the mature spore. The hosts were Limnodrilus hoffmeisteri and an unidentified species of Tubificidae (Oligochaeta), collected from a brook in southern Sweden, not Tubifex tubifex, the type host. The dimensions of the spore and the caudal spore projection varied more than the sizerange reported in the description of the species.The exospore is characteristic, with a surface layer resembling a unit membrane and a wide, dense, basal layer. The surface layer remains attached also to mature spores. The caudal projection is formed only from exospore material, and it is ornamented with longitudinal ridges. The polaroplast has an anterior region with closely packed lamellae and a posterior section with wider lamellae or tubules. The polar filament has a wide, straight, anterior part, a zone with successively reduced diameter, and a posterior, narrow section, arranged as an incomplete coil. The filament is easily extruded, and the length and size difference between the wide and narrow parts is apparent also using light microscopy. Membrane-lined compartments and aggregates of granular material are present close to the posterior pole. The species status of the microsporidium and the cytological characteristics of Bacillidium-Jirovecia species are discussed.
Abbreviations A AC AP CP D E EN F G FN FW HN NL P PA PM PP PS R S UL
= anchoring apparatus = collar-like part of the anchoring apparatus = pad-like part of the anchoring apparatus
= = = =
= = = = = =
=
centriolar plaque diplokaryon exospore endospore polar filament granules narrow part of the polar filament wide part of the polar filament host cell nucleus nucleolus polaroplast anterior part of the polaroplast plasma membrane posterior part of the polaroplast posterosome ribosomes polar sac unit membrane-like layer
© 1990 by Gustav Fischer Verlag, Stuttgart
Introduction The genus ]irovecia Weiser, 1977, was created for microsporidia with rod-shaped, tail-like prolonged spores, and Mrazekia caudata Leger and Hesse, 1916 was selected as type species [15]. This species was described in a manner typical for the time. The description contains information about the dimensions of the spore, the host, and the locality in France where it was collected, and it is only illustrated by two line drawings, one showing the mature spore and another of an infected host cell [9]. Microsporidia of this type were first observed by Mrazek, from material collected in Czechoslovakia. The two spores drawn in Fig. 21 in his paper on Myxocystis [11] are so similar to spores of Mrazekia caudata that Leger and Hesse identified them to belong to this species [9]. To my knowledge there is only one further publication dealing with this species, by Lorn [10], and based on material collected in Czechoslovakia. This paper describes the life cycle and the light microscopic cytology, especially for the 0932-4739/90/0025-0321$3.50/0
322 .
J. 1. R. Larsson
presporal stages, and it is illustrated by a series of line drawings. A microsporidium with spores of similar shape and size was isolated from two species of oligochaetes, collected from a brook in southern Sweden in 1988 and 1989. There are some questions about the identification, but more speaks for the species being ]irovecia (Mrazekia) caudata than being a species new to science. It was investigated using light and electron microscopy. As no modern publications have treated this species, the cytology is briefly described here, and some traits on the cytology of the microsporidia of the family Bacillidiidae Larsson, 1986 and the problems with the identification of the species are discussed.
Material and Methods The microsporidium was present in two samples of oligochaetes collected from a small brook, communicating with the river Kavlingean, at Gardsranga, in Scania in the south of Sweden. In the first sample, collected in April 1988 and investigated using both light and electron microscopic methods, the microsporidium was present in Limnodrilus hoffmeisteri Claparede, 1862 (Tubificidae) and in an unidentified species of Tubificidae. In the second sample, from July 1989 and investigated only using light microscopy, the host was L. hoffmeisteri. Fresh squash preparations were made according to the method of Hostounsky and Zizka [3], and studied using phase contrast microscopy and dark field illumination. Permanent squash preparations were lightly air-dried and fixed in Bouin-DuboscqBrasil solution for at least one hour. A number of spores ejected the polar filament spontaneously by the squashing. For paraffin sectioning infected segments were fixed in the same fixative overnight, washed and dehydrated in an ascending series of ethanols, cleared in butanol and embedded in paraplast. Sections were cut longitudinally at 10 urn. Squash peparations and sections were stained using Heidenhain's iron haematoxylin and Giemsa solution. For details on the histological techniques used see Romeis [13]. Permanent preparations were mounted in DePeX. Measurements were made with an eye-piece micrometer at x 1000. For transmission electron microscopy infected segments were excised and fixed in 2.5 % (v/v) glutaraldehyde in 0.2 M sodium cacodylate buffer (pH 7.2) at 4°C for 22 and 27 hours. After washing in cacodylate buffer and post fixation in 2 % (w/v) osmium tetroxide in cacodylate buffer for one hour at 4°C, the pieces were washed and dehydrated in an ascending series of buffer-acetone solutions to absolute acetone, and embedded in epon. Sections were stained with uranyl acetate and lead citrate. For scanning electron microscopy spores were smeared on circular cover glasses, lightly air-dried, and fixed in 2.5 % glutaraldehyde in cacodylate buffer at 4°C for 22 hours. After
washing in buffer and critical point drying, the smears were covered with metallic gold and palladium. Smears and sections numbered 880418-(D-G) RL and 890724C RL are in the collection of the author.
Results
Prevalence and Pathology The microsporidium was not common. In the sample collected in 1988 it was present in four out of 24 Tubificidae, and in hosts of two different species. In the sample of 1989 only one of the 27 Tubificidae investigated was infected. Infected oligochaetes were paler than uninfected ones. The microsporidium developed in the coelomocytes, which became hypertrophic (Fig. 1). The cell border had distinct finger-like protrusions. As infected cells exhibited fragments of nuclei in several positions, especially close to the cell border, infection probably induces the host nucleus to divide. Immature stages and mature spores were mingled in the cell, although younger developmental stages were more common near the periphery (Fig. 1). Mature spores were assembled in regular groups, but they were not arranged in radial manner in the cell, like it has been described for some microsporidia of Bacillidiidae.
Presporal Stages All stages observed had nuclei coupled as diplokarya. The most immature cells were aggregated in pluricellular groups, where still a part of the cells were joined by cytoplasmic bridges. Sectioned cells of this type measured up to 5.5 urn in diameter, and the diplokarya up to 3.2 urn, The cell border was a c. 8 nm thick unit membrane. The cytoplasm was granular with numerous, mostly free ribosomes (Fig. 2). The nucleoplasm was uniform, but in some nuclei a nucleolus was present (Fig. 3). The greatest sectioned nucleolus measured 404 nm in diameter. Up to 319 nm wide, electron dense centriolar plaques were visible in invaginations of the envelope of the nuclei (Fig. 1). These cells are interpreted as merozoites of the last merogony. The number of merozoites per meront is unknown, but the pluricellular groups of associated merozoites indicate that numerous daughter cells are formed. When the merozoites matured to sporonts, the cytoplasm was reorganized in the normal way, in that the amount of membrane-associated ribosomes increased to
Figs. 1-4. Presporal stages of ]irovecia caudata. - Fig. 1. Periphery of a coelomocyte with finger-like peripheral projections ~ (arrow-heads), Merozoites of the last generation, sporonts, sporoblasts and mature spores are mixed. - Fig. 2. Diplokaryotic merozoite of the last generation. - Fig. 3. Detail of a merozoite nucleus showing a nucleolus. - Fig. 4. A sporoblast and part of a mature spore. The polar filament and polaroplast are organized in close connection with a posterosome-like structure. The layers of the developing filament are numbered identically to the mature filament of Fig. 25. The arrow indicates the unit membrane component of the posterosome, polaroplast and polar filament. (Scale bars: Fig. 1 = 1 urn; Figs. 2-4 = 100 nm).
Cytology of jirovecia caudata . 323
-
EN E
F
324 .
J. I. R. Larsson
A
D
D
FW
...
Figs. 5-16. Unsectioned mature spores of Jirovecia caudata. - Figs. 5-7. Spores stained to show the diplokaryon, the anchoring apparatus and the unextruded polar filament. The unstained area at the posterior pole (arrows) probably corresponds to the membrane-lined posterior compartments and the darkly stained area (arrow head) to the accumulation of granular material (cf. Fig. 26).- Figs. 8-10. Unfixed spores exhibiting the diplokaryon, a posterior vacuole-like area (arrow), and the irregular texture of the tails. - Figs. 11-13. Scanning electron microscopy reveals a close connection between the tailand the surface of the spore. Arrows indicate the longitudinal ridges. The granular material is contamination. - Figs. 14-16. The extruded polar filament has a wide part, slightly longer than the spore, and a short narrow section. Arrows indicate the drop-like material at the tip of the filament, the arrow-head the darkly stained material at the posterior pole of the spore. (Figs. 5-7. Heidenhain's haematoxylin; Fig. 14. Giemsa. Scale bars: Figs. 5-10, 14-16 = 10 urn; Fig. 11 = 5 urn; Figs. 12-13 = 1 urn. Figs. 5-7 with a common bar on 5; Figs. 8-10 and 15-16 with a common bar on 16).
Cytology of ]irovecia caudata . 325
R
G) AR
326 . ~
J. 1. R. Larsson
Figs. 17-21. Theultrastructure of thespore. - Fig. 17. Thewide partof thepolar filament tapers posteriorly to thediplokaryon. The spore isconstricted at the posterior end(arrow).- Fig. 18. Longitudinallysectioned tailofexospore material, which also forms bridges across the tail (arrows). - Fig. 19. Anterior endof a mature spore, exhibitingthe twopartsof the polaroplast. Inset shows a greater magnifieddetail oftheanterior polaroplast,- Fig. 20. Theanchoring apparatus andtheattachment ofthe polar filament. Thelayers of the filament are numbered identically to Fig. 25. The translucent layers of the filament and the collar-like part of the anchoring apparatusarecontinuous (arrow-heads). Thearrows indicate thetwo layers ofthecollar-likepart.- Fig. 21. Section ofan infected host cell showingtransverselysectioned tailswith three andfourridges (arrows).(Scalebars: Figs. 17-18,21 = 1 urn; Figs. 19-20 andinset on 19 = 100 nm).
surround the diplokaryon in concentricallayers. The thick sporont wall was formed by patchily deposition of electron-dense material externally to the plasma membrane, successively increasing to a continuous dense layer. The number of sporoblasts per sporont is unknown, but probabl y the sporogony is disporoblastic, like previously observed in the genera jirovecia and Bacillidium. There were no signs of meiotic division. Young sporoblasts were cells of irregular shape. They had originally a c. 47 nm thick cellwall, with a continuous layer of electron-dense material externally to the plasma membrane. Successively the shape changed to become more elongate and regular. Simultaneously the dense surface material was stratified into a wide, uniform, basal layer, and a surface layer resembling a c. 7 nm thick unit membrane, and at the same time the first signs of the translucent endospore layer appeared (Fig. 4). The polar filament and the polaroplast were initiated from a posterosome-like structure, a system of vesicles filled with electron-dense material (Fig. 4). The internal organization of the developing polar filament is treated in comparison with the filament of the mature spore. Sporoblasts and immature spores had no exospore projections.
The Spore Mature spores were cylindrical with blunt ends (Figs. 5-16). The posterior end was prolonged in a tail-like manner. The tail was not stiff, and in squashed spores it was often curved or angular (Figs. 5-7). A part of the spores fixed for transmission electron microscopy were posteriorly distorted by shrinkage (Fig. 17). Unfixed mature spores (series 890724-C) measured 1.6-2.1 X 15.4-17.5 urn (x = 1.8 X 16.5; n = 12). The tail was 10.5-1 7.5 urn long (x = 13.1; n = 12). It was widest close to the attachment, where it measured c. 1 urn in diameter, and it tapered regularly towards the blunt tip (Figs. 8, 11, 13). While stained spores of the samples from 1988 and
1989 (both series from Limnodrilus hoffmeisteri) had approximately the same size, there was a remarkable differencein tail length. In the series 890724-C the spores measured 1.0-1.8 X 11.2-15.5 urn (x = 1.1 X 12.6; n = 25), and the tails 9.0-15.5 urn (x 12.1; n = 25). Spores of series 880418-D measured 1.1-2.0 X 12.5-15.6!lm (x = 1.2 X 13.0; n = 25), and the tails were 11.0-19.6 urn long (x = 15.1; n = 25). Alltails were measured on spores with almost straight tails. In both series the tail width close to the attachment was c. 0.9 urn, The 106-170 nm thick spore wall exhibited the normal three layers (Fig. 28): an internal c. 8 nm thick unit membrane, a translucent endospore of fairly uniform thickness, although thinner at the anterior pole (Fig. 16), and an electron-dense exospore composed of two layers. The c. 7 nm thick surface layer resembled a unit membrane with more electron-dense internal lining, and the 40-45 nm thick layer below was uniformly electron-dense (Figs. A, 28). The tail was hollow, with a wall formed from exospore material alone, approximately of the same dimensions as in the real exospore (Fig. 18). The unit membrane-like layer was present, obviously on both sides, and the internal dense layer formed bridges across the tail, dividing it into compartments (Figs. 18, 27). The tails looked rather irregular under the light microscope (Figs. 8-10), but scanning electron microscopy revealed regular tails, ornamented with prominent longitudinal ridges (Figs. 11-13 ). Tails with three and four ridges were approximately equally abundant (Fig. 21). The polar filament had a straight, 326- 404 nm wide, anterior part, reaching to about 1/4 from the posterior end of the spore, an intermediate section with successively reduced diameter, and a posterior narrow portion, 117-18 7 nm in diameter, forming an incomplete coil (Fig. 17). The filamentpassed through the centre of the spore for about 1/4 of the length, and, at the level of the nuclei, turned to the side and followed the spore wall to the coil (Fig. 17). The filament was easily extruded by squashing,
Fig. 22. Mature spores transverselysectioned at thepolaroplast and diplokaryon levels. Theblack arrow indicates thevariableshapeof thecentre of thefilament, the white arrowsthefibrous component of layer 5. - Fig. 23. Thepolar filamentof an immature spore. The internal partoflayer4 isless electron-dense (arrow). - Fig. 24. Transversely sectioned narrow partof thefilament ofa mature spore.-:Fig. 25. Transversely sectioned wide filam ent of a mature spore with the consecutive layers numbered in direction inwards. Arrows indicate the fibrous component of layer 5. - Fig. 26. Posterior end of a mature spore, showing the membrane-lined compartments (arrows)andtheposterior accumulation ofgranular material. - Fig. 27. Thewall ofthetail.Arrows indicatetheinternal bridges. - Fig. 28. The wall of a mature spore. The unit membrane-like surface layer remains attached to the exospore. (Scale bars = 100 nm).
~
Cytology of Jirovecia caudata . 327
-
~.
-
328 .
J. I. R. Larsson
and the great difference in diameter and length between the two regions was clearly visible in the ejected filament (Figs. 14-16). The wide part of the extruded filament measured c. 0.9 X 14.7-18.9 urn in unfixed smears, and the narrow tip was 2.1-2.8 urn long, which means that both length and width increases during extrusion. A small drop was visible at the tip of newly ejected filaments (Figs. 14-15), and it increased in size rapidly (Fig. 16). Ejected sporoplasms were not observed. The mature polar filament exhibited a prominent stratification of the normal type for microsporidia, in transverse sections visible as concentricallayers with different electron density and thickness (Figs.22, 25). It was covered by a c. 5 nm thick unit membrane-like layer, and in direction inwards followed a c. 7 nm thick uniform, moderately dense layer, a distinct c. 5 nm thick fairly lucent band, and the widest layer of the filament, c. 1/4 of the filament radius, with a characteristic mottled appearance. The amount of dense material in this layer increased in direction outwards. The following layer was the most electron dense, and it was slightly thinner than the neighbouring layer. The thickness varied in different regions of the filament, making the centre of a longitudinally sectioned filament look irregular (Fig. 20). Actually the layer was composed of material of two different electron densities. The least dense material had the same density as the most dense material of the mottled layer, and it was arranged like fibrils radiating from the centre (Figs. 22,25). The centre, slightly less than 1/3 of the diameter of the filament, was moderately dense with diffuse concentrical banding and a spot in the very centre. In transverse sections it was either perfectly circular of lobed (Fig. 22). Transverse sections through the narrow coil were rarely seen, but wide and narrow parts appeared to have approximately the same structure (Fig. 24). However, the stratification was less distinct in the narrow coil, and the mottled and most dense layers were reduced in size. In the sporoblast the developing filament was organized from posterosome-like material, which obviously also gave rise to the polaroplast lamellae (Fig. 4). The diffusely stratified centre of the filament was apparently differentiated first. In transverse sections it had reached perfectly circular shape ata time when the external part of the developing filament still was more or less irregular. The developing filament at this stage had an external thin unit membrane, and in direction inwards followed a narrow, irregular layer of the same density as the material of the posterosome, a diffuse translucent strand, and a welldefined lens-shaped area of granular material. In the immature spore, the transversely sectioned filament was still not circular, but the granular material had differentiated into an internal dense ring and an external mottled layer. The material of the mottled layer was reorganized, and the transparency increased at the border to the dense ring (Fig. 23). The fibril-like structure of the most dense layer was not visible until the spore was mature (Fig. 25). The polar filament was attached to an anchoring
apparatus with two components: a central, up to 287 nm wide, pad-like structure, and a peripheral collar-like,
convex part, up to 468 nm in diameter (Fig. 20). In longitudinal sections laterally to the central axis of the spore, the pad appeared to be composed of material of two different electron densities in regular layers. The collar had an anterior part of moderately dense material and a posterior translucent layer of somewhat fibrous appearance. The mottled layer of the filament attached to the translucent material of the collar, and the external lucent material of the mottled layer and the collar united. The centre, from the most dense layer and inwards, united with the pad. The anchoring apparatus was visible also in stained squash preparations (Figs. 6-7). The fairly short polaroplast was composed of two parts, with compartments lined with c. 5 nm thick unit membranes of the same type as surrounding the polar filament (Fig. 19). The anterior polaroplast region extended for about 1/10 of the spore length. It had closely and regularly arranged lamellae, which appeared so closely packed that they seemed to lack a lumen (Fig. 19). The periodicity was 7-8 nm. The posterior part, with up to 34 nm wide lamellar or tubular compartments (Fig. 19), ended about 1/5 from the anterior pole of the spore, anteriorly to the position of the nuclei. The polar sac had a unit membrane lining identical to the polar filament and the polaroplast. The sac enclosed the anchoring apparatus and continued backwards in an umbrella-like manner, surrounding the anterior polaroplast more or less completely (Fig. 20). The elongated diplokaryon occupied the central half of the spore (Figs. 5, 17). In stained spores the diplokaryon measured 7.1-8.9 urn long. The side facing the polar filament was concave (Fig. 22). The cytoplasm was electron-dense with helically arranged, membrane-associated ribosomes in layers parallel to the spore wall (Figs. 19, 28). The membrane layers were 43-50 nm apart. Both in unfixed and stained spores a lucent area was visible in the posterior fifth of the spore (Figs.5-6, 8-9). In transmission electron microscopy this area often appeared shrunken (Fig. 17). Membrane-lined compartments with a uniform dense content were present here, together with aggregates of 12-17 nm wide granules (Fig. 26). The dark spot visible in stained spores corresponded obviously to the granular aggregates (Fig. 6).
Discussion
Identity of the Species According to the original description of]irovecia (Mrazekia) caudata, based on material collected in France, the dimensions of the spores are 1.3-1.4 X 16-18 urn, and the tails are approximately equally long [9]. Lorn, using material with provenance from Czechoslovakia, reported the dimensions 1.5-2 X 15-16.6!-tm (spores), and 16-17 urn (tails) [10]. Spores of the Swedish microsporidium measured 1.6-2.1 X 15.4-17.5 urn, the tails 10.5-17.5 urnin unfixed condition. Fixed spores measured 1.0-1.8 X 11.2-15.5 urn, fixed tails 9.0-15.0 urn. Both series are from the same host specimen. In a second series of
Cytology of [irovecia caudata . 329
measurements, from another specimen of the same host least closely related. As the description of]. caudata [9] is species and collectedat the same locality but in a different not of today's standard, minor differences between these year, the spores were slightly greater, and the tails were microsporidia should not be exaggerated. Until j. caudata has been isolated from the type host in France, described considerably longer. There is a greater sizerange in the Swedishmaterial, and adequately, and found to be different from the Swedish it contains smaller spores than the samples from France microsporidium, strong evidence for treating this as a new and Czechoslovakia. The greatest spores are of the same ]irovecia species is lacking. size, however, and the proportions between tails and spores equal them of the description. A re-evaluationof the Cytology type material, measured using identical techniques, might The cytologyof the species treated herein is similarto the show a greater variation in dimensions, and it would also allow a closer comparison of the cytology. No micro- cytology of other ]irovecia and Bacillidium species from graphs of this species have been published, and the line which details are known [1, 7, 8, 12, 14]. The straight part drawings of the description are not sufficient for such of the polar filament, often called the manubrium, has a comparison. However, the collection of L. Legerhas been wide mottled layer (Figs. 19,25), which seems to be unique destroyed (Degrange, personal communication), so for the to these microsporidia. It is further characteristic that a zone with successively reduced diameter separates the wide moment the brief description is all we have. Another complication is the different host species. The anterior part of the filament from the posterior narrow coil type host is Tubifex tubifex, while the Swedish microspo- (Fig. 17; Fig.2 in [1], Fig.9 in [14]), while in microsporidia ridium was isolated from Limnodrilus hoffmeisteri and with an anisofilar polar filament, the filament is abruptly from an unidentified species of Tubificidae. However, constricted. At the light microscopiclevel it is typical that Leger and Hesse [9] and Kudo [5] were of the opinion that material at the posterior pole of the spore takes stain using ]irovecia (Mrazekia) caudata was identical to the organism traditional Giemsa and Heidenhain techniques (Fig. 6; Mrazek had described from Limnodrilus under the name Figs. 7, 9-10 in [6]). Myxocystis [11], which means that they did not consider]. The exospore of ]irovecia and Bacillidium species caudata to be host specific. Mrazek's publication has three differentiates at the sporoblast stage into a surface layer illustrations of microsporidia of the ]irovecia-Bacillidium- similar to a unit membrane and a wide uniformly dense type (Figs. 7, 20, 21 in [11]), from hosts of the genera basal layer (Fig. 28). However, the fate of the surface layer Limnodrilus and Lumbriculus. Figure 21, which shows is different in different species. In B. filiferum the surface spores of ]. caudata-type, is from Lubriculus not from layer partially losescontact with the basal layer and forms Limnodrilus. There is no indication of the magnification, tubule-like projections [7]. The surface layer of]. involuta and the size of the spores is unknown. jirovec studied completely loses contact with the basal layer and forms a Mrazek's slides and found that spores from Lumbriculus continuous sac, similar to a sporophorous vesicle [8]. In were 7.5-9 urn long, spores from Limnodrilus 8-10 urn the microsporidium treated herein the surface layer long - approximately half the length of the spores of ]. remains unchanged from the sporoblast stage (Fig. 4) to caudata [4]. He described these microsporidia as two new the mature spore (Fig. 28). species: Mrazekia lumbriculi and M.limnodrili [4], both of The tail-like spore projection of the ]irovecia species of them now recognized as ]irovecia species [6]. Actually this paper is formed from exospore material alone (Fig. Tubifex tubifex is so far the only host from which ]. 18), likethe tail of]. involuta (Fig. 20 in [8]).In this respect caudata has been reported. they differ from]. (Mrazekia) brevicauda, where the tail is We know little about the host specificity of microspo- formed by a constriction of the entire spore, with the ridia, and nothing at all about the host range of micro- consequence that the complete spore wall, with plasma sporidia from Oligochaeta. Many microsporidia are con- membrane, endospore and exospore, is present in the tail sidered host specific, and quite a number have been (Fig. 1 in [1]). The two waysof making the tail indicate that described as new to science for the reason that they were the genus ]irovecia in its present sense is heterogeneous. Lorn interpeted the developmental stages of ]irovecia isolated from a new host, even if they could not be distinguished from microsporidia of related hosts using (Mrazekia) caudata, including the mature spore, to be uninucleate [10], but this was obviously erroneous. In the morphological criteria. As we have clear evidence that some microsporidia, like Nosemaacridiophagus, are com- description of the species nothing specific is mentioned petent to infect a number of related species, also in about the nuclear condition [9]. However, in the same differentgenera [2],we should be careful when postulating paper, dealing with a number of new Mrazekia species, Legerand Hesse discussed the characteristicsof Mrazekia, strict host specificity. The species treated herein is clearly different from selectingM. argoisi as typical for the genus. In the spore of ]irovecia (Mrazekia) lumbriculi, J. (M.) limnodrili and ]. M. argoisi is: "... le germe, globuleux et tres petit avec un (M.) ilyodrili, for all of them have spores less than 10 urn minuscule noyau gernine ... " [9]. Obviously this is a long [47]. In addition]. lumbriculi, not]' limnodrili, has description of a diplokaryon, and it seems clear that they spores of similar shape to ]. caudata and the Swedish considered the diplokaryotic condition to be characteristic microsporidium [11, 4]. ]. caudata is the only existing for all Mrazekia species, including M. caudata. Also species where both the size and the shape of the spore are jirovec was aware that all developmental stages of these similarto the Swedish microsporidium, and the hosts are at microsporidia are diplokaryotic [4], and most later inves-
330 .
J. I. R. Larsson
tigations on Bacillidium and ]irovecia species have verified the diplokaryotic nature. The diploka ryotic condition of the species treated herein does not contradict that it is identified as J. caudata.
Acknowledgements The author is greatly indebted to Mrs. Lina Hansen and Mrs. Inger Norling, both at the Department of Zoology, University of Lund, Sweden, for skilful technical assistance, and to Prof. C. Degrange, Universite Scientifique et Medicale de Grenoble, for information about the fate of Prof. L. L. Leger's collection. The investigation was financially supported be research grants from the Swedish Natural Science Research Council.
References 1 Gorz P. (1981): Homology of the manubrium of Mrazekia brevicauda and the polar filament of other microsporidia. Z. Parasitenkd., 64, 321-333. 2 Henry J. E. (1967): Nosema acridiophagus sp. n., a microsporidian isolated from grasshoppers. J. Invertebr. Pathol., 9, 331- 34 1. 3 Hostounsky Z. and Zizka Z. (1979): A modification of the "agar cushion method " for observation and recording microspor idian spores. J. Protozool. , 26, 41A-42A.
4 [Irovec O. (1936): Zur Kenntnis von in Oligochaten para sitierenden Mikrosporidien aus der Familie Mrazekidae. Arch. Protistenkd., 87, 314-344. 5 Kudo R. (1924): A biologic and taxonomic study of the microsporidia. Ill. Biol. Monogr. , 9(2/3), 1-268. 6 Larsson J. I. R. (1986): On the taxonomy of Mrazekidae: resurrection of the genus Bacillidium Janda, 1928. J. Protozool., 33, 542-546. 7 Larsson J. I. R. (1989): The light and electron microscopic cytology of Bacillidium filiferum sp. nov. (Microspora, Bacillidiidae). Arch. Protistenkd., 13 7, 345- 355. 8 Larsson J. I. R. (1989): Light and electron microscope studies on ]irovecia involuta sp. nov. (Microspora, Bacillidiidae), a new microsporidian para site of oligochaetes in Sweden. Europ. J. Protistol., 25, 172-181. 9 Leger L. et Hesse E. (1916): Mrazekia, genre nouveau de Microsporidies a spores tubuleuses. C. R. Soc. BioI., 79, 345-348. 10 LornJ. (1958): K poznanf vyvojoveho cyclu Mrdzekia caudata Leger et Hesse 1916. Cesk, Parasitol., 5, 147-152. 11 Mrazek A. (1910): Sporozoenstudien. Zur Auffassung der Myxocystiden. Arch. Protistenkd., 18,245-259. 12 Puytorac P. de (1962): Observations sur l'ultrastrueture de la Microsporidie Mrazekia lumbriculi, Jfrovec. J. Microscopie, 1,39-46. 13 Romeis B. (1968): Mikroskop ische Technik. Oldenbourg Verlag, Munchen and Wien. 14 Vavra J., Joyon L. et Puytorac P. de (1966): Observation sur I'ultra structure du filament polaire des Microsporidie s. Protistologica, 2, 109-112. 15 Weiser J. (1977) : Contribution to the classification of microsporidida. Vestn. Cs. spol. Zool., 41, 308-320.
Key words: ]irovecia caudata - Mrazekia - Microsporidia - Oligochaeta - Ultrastructure Ronny Larsson, Department of Zoology, University of Lund, Helgonav. 3, S-223 62 Lund, Sweden
Europ.j.Protistol. 25, 321-330 (1990) June 29, 1990
PROTISTOLOGY
On the Cytology of Jirovecia caudata (Leger and Hesse, 1916) (Microspora, Bacillidiidae) J. I. Ronny Larsson Department of Zoology, University of Lund, Sweden
Summary The cytology of a microsporidium identified as Jirovecia (Mrazekia) caudata (Leger & Hesse, 1916), the type species of the genus Jirovecia Weiser, 1977, is described with emphasis on the ultrastructure of the mature spore. The hosts were Limnodrilus hoffmeisteri and an unidentified species of Tubificidae (Oligochaeta), collected from a brook in southern Sweden, not Tubifex tubifex, the type host. The dimensions of the spore and the caudal spore projection varied more than the sizerange reported in the description of the species.The exospore is characteristic, with a surface layer resembling a unit membrane and a wide, dense, basal layer. The surface layer remains attached also to mature spores. The caudal projection is formed only from exospore material, and it is ornamented with longitudinal ridges. The polaroplast has an anterior region with closely packed lamellae and a posterior section with wider lamellae or tubules. The polar filament has a wide, straight, anterior part, a zone with successively reduced diameter, and a posterior, narrow section, arranged as an incomplete coil. The filament is easily extruded, and the length and size difference between the wide and narrow parts is apparent also using light microscopy. Membrane-lined compartments and aggregates of granular material are present close to the posterior pole. The species status of the microsporidium and the cytological characteristics of Bacillidium-Jirovecia species are discussed.
Abbreviations A AC AP CP D E EN F G FN FW HN NL P PA PM PP PS R S UL
= anchoring apparatus = collar-like part of the anchoring apparatus = pad-like part of the anchoring apparatus
= = = =
= = = = = =
=
centriolar plaque diplokaryon exospore endospore polar filament granules narrow part of the polar filament wide part of the polar filament host cell nucleus nucleolus polaroplast anterior part of the polaroplast plasma membrane posterior part of the polaroplast posterosome ribosomes polar sac unit membrane-like layer
© 1990 by Gustav Fischer Verlag, Stuttgart
Introduction The genus ]irovecia Weiser, 1977, was created for microsporidia with rod-shaped, tail-like prolonged spores, and Mrazekia caudata Leger and Hesse, 1916 was selected as type species [15]. This species was described in a manner typical for the time. The description contains information about the dimensions of the spore, the host, and the locality in France where it was collected, and it is only illustrated by two line drawings, one showing the mature spore and another of an infected host cell [9]. Microsporidia of this type were first observed by Mrazek, from material collected in Czechoslovakia. The two spores drawn in Fig. 21 in his paper on Myxocystis [11] are so similar to spores of Mrazekia caudata that Leger and Hesse identified them to belong to this species [9]. To my knowledge there is only one further publication dealing with this species, by Lorn [10], and based on material collected in Czechoslovakia. This paper describes the life cycle and the light microscopic cytology, especially for the 0932-4739/90/0025-0321$3.50/0
322 .
J. 1. R. Larsson
presporal stages, and it is illustrated by a series of line drawings. A microsporidium with spores of similar shape and size was isolated from two species of oligochaetes, collected from a brook in southern Sweden in 1988 and 1989. There are some questions about the identification, but more speaks for the species being ]irovecia (Mrazekia) caudata than being a species new to science. It was investigated using light and electron microscopy. As no modern publications have treated this species, the cytology is briefly described here, and some traits on the cytology of the microsporidia of the family Bacillidiidae Larsson, 1986 and the problems with the identification of the species are discussed.
Material and Methods The microsporidium was present in two samples of oligochaetes collected from a small brook, communicating with the river Kavlingean, at Gardsranga, in Scania in the south of Sweden. In the first sample, collected in April 1988 and investigated using both light and electron microscopic methods, the microsporidium was present in Limnodrilus hoffmeisteri Claparede, 1862 (Tubificidae) and in an unidentified species of Tubificidae. In the second sample, from July 1989 and investigated only using light microscopy, the host was L. hoffmeisteri. Fresh squash preparations were made according to the method of Hostounsky and Zizka [3], and studied using phase contrast microscopy and dark field illumination. Permanent squash preparations were lightly air-dried and fixed in Bouin-DuboscqBrasil solution for at least one hour. A number of spores ejected the polar filament spontaneously by the squashing. For paraffin sectioning infected segments were fixed in the same fixative overnight, washed and dehydrated in an ascending series of ethanols, cleared in butanol and embedded in paraplast. Sections were cut longitudinally at 10 urn. Squash peparations and sections were stained using Heidenhain's iron haematoxylin and Giemsa solution. For details on the histological techniques used see Romeis [13]. Permanent preparations were mounted in DePeX. Measurements were made with an eye-piece micrometer at x 1000. For transmission electron microscopy infected segments were excised and fixed in 2.5 % (v/v) glutaraldehyde in 0.2 M sodium cacodylate buffer (pH 7.2) at 4°C for 22 and 27 hours. After washing in cacodylate buffer and post fixation in 2 % (w/v) osmium tetroxide in cacodylate buffer for one hour at 4°C, the pieces were washed and dehydrated in an ascending series of buffer-acetone solutions to absolute acetone, and embedded in epon. Sections were stained with uranyl acetate and lead citrate. For scanning electron microscopy spores were smeared on circular cover glasses, lightly air-dried, and fixed in 2.5 % glutaraldehyde in cacodylate buffer at 4°C for 22 hours. After
washing in buffer and critical point drying, the smears were covered with metallic gold and palladium. Smears and sections numbered 880418-(D-G) RL and 890724C RL are in the collection of the author.
Results
Prevalence and Pathology The microsporidium was not common. In the sample collected in 1988 it was present in four out of 24 Tubificidae, and in hosts of two different species. In the sample of 1989 only one of the 27 Tubificidae investigated was infected. Infected oligochaetes were paler than uninfected ones. The microsporidium developed in the coelomocytes, which became hypertrophic (Fig. 1). The cell border had distinct finger-like protrusions. As infected cells exhibited fragments of nuclei in several positions, especially close to the cell border, infection probably induces the host nucleus to divide. Immature stages and mature spores were mingled in the cell, although younger developmental stages were more common near the periphery (Fig. 1). Mature spores were assembled in regular groups, but they were not arranged in radial manner in the cell, like it has been described for some microsporidia of Bacillidiidae.
Presporal Stages All stages observed had nuclei coupled as diplokarya. The most immature cells were aggregated in pluricellular groups, where still a part of the cells were joined by cytoplasmic bridges. Sectioned cells of this type measured up to 5.5 urn in diameter, and the diplokarya up to 3.2 urn, The cell border was a c. 8 nm thick unit membrane. The cytoplasm was granular with numerous, mostly free ribosomes (Fig. 2). The nucleoplasm was uniform, but in some nuclei a nucleolus was present (Fig. 3). The greatest sectioned nucleolus measured 404 nm in diameter. Up to 319 nm wide, electron dense centriolar plaques were visible in invaginations of the envelope of the nuclei (Fig. 1). These cells are interpreted as merozoites of the last merogony. The number of merozoites per meront is unknown, but the pluricellular groups of associated merozoites indicate that numerous daughter cells are formed. When the merozoites matured to sporonts, the cytoplasm was reorganized in the normal way, in that the amount of membrane-associated ribosomes increased to
Figs. 1-4. Presporal stages of ]irovecia caudata. - Fig. 1. Periphery of a coelomocyte with finger-like peripheral projections ~ (arrow-heads), Merozoites of the last generation, sporonts, sporoblasts and mature spores are mixed. - Fig. 2. Diplokaryotic merozoite of the last generation. - Fig. 3. Detail of a merozoite nucleus showing a nucleolus. - Fig. 4. A sporoblast and part of a mature spore. The polar filament and polaroplast are organized in close connection with a posterosome-like structure. The layers of the developing filament are numbered identically to the mature filament of Fig. 25. The arrow indicates the unit membrane component of the posterosome, polaroplast and polar filament. (Scale bars: Fig. 1 = 1 urn; Figs. 2-4 = 100 nm).
Cytology of jirovecia caudata . 323
-
EN E
F
324 .
J. I. R. Larsson
A
D
D
FW
...
Figs. 5-16. Unsectioned mature spores of Jirovecia caudata. - Figs. 5-7. Spores stained to show the diplokaryon, the anchoring apparatus and the unextruded polar filament. The unstained area at the posterior pole (arrows) probably corresponds to the membrane-lined posterior compartments and the darkly stained area (arrow head) to the accumulation of granular material (cf. Fig. 26).- Figs. 8-10. Unfixed spores exhibiting the diplokaryon, a posterior vacuole-like area (arrow), and the irregular texture of the tails. - Figs. 11-13. Scanning electron microscopy reveals a close connection between the tailand the surface of the spore. Arrows indicate the longitudinal ridges. The granular material is contamination. - Figs. 14-16. The extruded polar filament has a wide part, slightly longer than the spore, and a short narrow section. Arrows indicate the drop-like material at the tip of the filament, the arrow-head the darkly stained material at the posterior pole of the spore. (Figs. 5-7. Heidenhain's haematoxylin; Fig. 14. Giemsa. Scale bars: Figs. 5-10, 14-16 = 10 urn; Fig. 11 = 5 urn; Figs. 12-13 = 1 urn. Figs. 5-7 with a common bar on 5; Figs. 8-10 and 15-16 with a common bar on 16).
Cytology of ]irovecia caudata . 325
R
G) AR
326 . ~
J. 1. R. Larsson
Figs. 17-21. Theultrastructure of thespore. - Fig. 17. Thewide partof thepolar filament tapers posteriorly to thediplokaryon. The spore isconstricted at the posterior end(arrow).- Fig. 18. Longitudinallysectioned tailofexospore material, which also forms bridges across the tail (arrows). - Fig. 19. Anterior endof a mature spore, exhibitingthe twopartsof the polaroplast. Inset shows a greater magnifieddetail oftheanterior polaroplast,- Fig. 20. Theanchoring apparatus andtheattachment ofthe polar filament. Thelayers of the filament are numbered identically to Fig. 25. The translucent layers of the filament and the collar-like part of the anchoring apparatusarecontinuous (arrow-heads). Thearrows indicate thetwo layers ofthecollar-likepart.- Fig. 21. Section ofan infected host cell showingtransverselysectioned tailswith three andfourridges (arrows).(Scalebars: Figs. 17-18,21 = 1 urn; Figs. 19-20 andinset on 19 = 100 nm).
surround the diplokaryon in concentricallayers. The thick sporont wall was formed by patchily deposition of electron-dense material externally to the plasma membrane, successively increasing to a continuous dense layer. The number of sporoblasts per sporont is unknown, but probabl y the sporogony is disporoblastic, like previously observed in the genera jirovecia and Bacillidium. There were no signs of meiotic division. Young sporoblasts were cells of irregular shape. They had originally a c. 47 nm thick cellwall, with a continuous layer of electron-dense material externally to the plasma membrane. Successively the shape changed to become more elongate and regular. Simultaneously the dense surface material was stratified into a wide, uniform, basal layer, and a surface layer resembling a c. 7 nm thick unit membrane, and at the same time the first signs of the translucent endospore layer appeared (Fig. 4). The polar filament and the polaroplast were initiated from a posterosome-like structure, a system of vesicles filled with electron-dense material (Fig. 4). The internal organization of the developing polar filament is treated in comparison with the filament of the mature spore. Sporoblasts and immature spores had no exospore projections.
The Spore Mature spores were cylindrical with blunt ends (Figs. 5-16). The posterior end was prolonged in a tail-like manner. The tail was not stiff, and in squashed spores it was often curved or angular (Figs. 5-7). A part of the spores fixed for transmission electron microscopy were posteriorly distorted by shrinkage (Fig. 17). Unfixed mature spores (series 890724-C) measured 1.6-2.1 X 15.4-17.5 urn (x = 1.8 X 16.5; n = 12). The tail was 10.5-1 7.5 urn long (x = 13.1; n = 12). It was widest close to the attachment, where it measured c. 1 urn in diameter, and it tapered regularly towards the blunt tip (Figs. 8, 11, 13). While stained spores of the samples from 1988 and
1989 (both series from Limnodrilus hoffmeisteri) had approximately the same size, there was a remarkable differencein tail length. In the series 890724-C the spores measured 1.0-1.8 X 11.2-15.5 urn (x = 1.1 X 12.6; n = 25), and the tails 9.0-15.5 urn (x 12.1; n = 25). Spores of series 880418-D measured 1.1-2.0 X 12.5-15.6!lm (x = 1.2 X 13.0; n = 25), and the tails were 11.0-19.6 urn long (x = 15.1; n = 25). Alltails were measured on spores with almost straight tails. In both series the tail width close to the attachment was c. 0.9 urn, The 106-170 nm thick spore wall exhibited the normal three layers (Fig. 28): an internal c. 8 nm thick unit membrane, a translucent endospore of fairly uniform thickness, although thinner at the anterior pole (Fig. 16), and an electron-dense exospore composed of two layers. The c. 7 nm thick surface layer resembled a unit membrane with more electron-dense internal lining, and the 40-45 nm thick layer below was uniformly electron-dense (Figs. A, 28). The tail was hollow, with a wall formed from exospore material alone, approximately of the same dimensions as in the real exospore (Fig. 18). The unit membrane-like layer was present, obviously on both sides, and the internal dense layer formed bridges across the tail, dividing it into compartments (Figs. 18, 27). The tails looked rather irregular under the light microscope (Figs. 8-10), but scanning electron microscopy revealed regular tails, ornamented with prominent longitudinal ridges (Figs. 11-13 ). Tails with three and four ridges were approximately equally abundant (Fig. 21). The polar filament had a straight, 326- 404 nm wide, anterior part, reaching to about 1/4 from the posterior end of the spore, an intermediate section with successively reduced diameter, and a posterior narrow portion, 117-18 7 nm in diameter, forming an incomplete coil (Fig. 17). The filamentpassed through the centre of the spore for about 1/4 of the length, and, at the level of the nuclei, turned to the side and followed the spore wall to the coil (Fig. 17). The filament was easily extruded by squashing,
Fig. 22. Mature spores transverselysectioned at thepolaroplast and diplokaryon levels. Theblack arrow indicates thevariableshapeof thecentre of thefilament, the white arrowsthefibrous component of layer 5. - Fig. 23. Thepolar filamentof an immature spore. The internal partoflayer4 isless electron-dense (arrow). - Fig. 24. Transversely sectioned narrow partof thefilament ofa mature spore.-:Fig. 25. Transversely sectioned wide filam ent of a mature spore with the consecutive layers numbered in direction inwards. Arrows indicate the fibrous component of layer 5. - Fig. 26. Posterior end of a mature spore, showing the membrane-lined compartments (arrows)andtheposterior accumulation ofgranular material. - Fig. 27. Thewall ofthetail.Arrows indicatetheinternal bridges. - Fig. 28. The wall of a mature spore. The unit membrane-like surface layer remains attached to the exospore. (Scale bars = 100 nm).
~
Cytology of Jirovecia caudata . 327
-
~.
-
328 .
J. I. R. Larsson
and the great difference in diameter and length between the two regions was clearly visible in the ejected filament (Figs. 14-16). The wide part of the extruded filament measured c. 0.9 X 14.7-18.9 urn in unfixed smears, and the narrow tip was 2.1-2.8 urn long, which means that both length and width increases during extrusion. A small drop was visible at the tip of newly ejected filaments (Figs. 14-15), and it increased in size rapidly (Fig. 16). Ejected sporoplasms were not observed. The mature polar filament exhibited a prominent stratification of the normal type for microsporidia, in transverse sections visible as concentricallayers with different electron density and thickness (Figs.22, 25). It was covered by a c. 5 nm thick unit membrane-like layer, and in direction inwards followed a c. 7 nm thick uniform, moderately dense layer, a distinct c. 5 nm thick fairly lucent band, and the widest layer of the filament, c. 1/4 of the filament radius, with a characteristic mottled appearance. The amount of dense material in this layer increased in direction outwards. The following layer was the most electron dense, and it was slightly thinner than the neighbouring layer. The thickness varied in different regions of the filament, making the centre of a longitudinally sectioned filament look irregular (Fig. 20). Actually the layer was composed of material of two different electron densities. The least dense material had the same density as the most dense material of the mottled layer, and it was arranged like fibrils radiating from the centre (Figs. 22,25). The centre, slightly less than 1/3 of the diameter of the filament, was moderately dense with diffuse concentrical banding and a spot in the very centre. In transverse sections it was either perfectly circular of lobed (Fig. 22). Transverse sections through the narrow coil were rarely seen, but wide and narrow parts appeared to have approximately the same structure (Fig. 24). However, the stratification was less distinct in the narrow coil, and the mottled and most dense layers were reduced in size. In the sporoblast the developing filament was organized from posterosome-like material, which obviously also gave rise to the polaroplast lamellae (Fig. 4). The diffusely stratified centre of the filament was apparently differentiated first. In transverse sections it had reached perfectly circular shape ata time when the external part of the developing filament still was more or less irregular. The developing filament at this stage had an external thin unit membrane, and in direction inwards followed a narrow, irregular layer of the same density as the material of the posterosome, a diffuse translucent strand, and a welldefined lens-shaped area of granular material. In the immature spore, the transversely sectioned filament was still not circular, but the granular material had differentiated into an internal dense ring and an external mottled layer. The material of the mottled layer was reorganized, and the transparency increased at the border to the dense ring (Fig. 23). The fibril-like structure of the most dense layer was not visible until the spore was mature (Fig. 25). The polar filament was attached to an anchoring
apparatus with two components: a central, up to 287 nm wide, pad-like structure, and a peripheral collar-like,
convex part, up to 468 nm in diameter (Fig. 20). In longitudinal sections laterally to the central axis of the spore, the pad appeared to be composed of material of two different electron densities in regular layers. The collar had an anterior part of moderately dense material and a posterior translucent layer of somewhat fibrous appearance. The mottled layer of the filament attached to the translucent material of the collar, and the external lucent material of the mottled layer and the collar united. The centre, from the most dense layer and inwards, united with the pad. The anchoring apparatus was visible also in stained squash preparations (Figs. 6-7). The fairly short polaroplast was composed of two parts, with compartments lined with c. 5 nm thick unit membranes of the same type as surrounding the polar filament (Fig. 19). The anterior polaroplast region extended for about 1/10 of the spore length. It had closely and regularly arranged lamellae, which appeared so closely packed that they seemed to lack a lumen (Fig. 19). The periodicity was 7-8 nm. The posterior part, with up to 34 nm wide lamellar or tubular compartments (Fig. 19), ended about 1/5 from the anterior pole of the spore, anteriorly to the position of the nuclei. The polar sac had a unit membrane lining identical to the polar filament and the polaroplast. The sac enclosed the anchoring apparatus and continued backwards in an umbrella-like manner, surrounding the anterior polaroplast more or less completely (Fig. 20). The elongated diplokaryon occupied the central half of the spore (Figs. 5, 17). In stained spores the diplokaryon measured 7.1-8.9 urn long. The side facing the polar filament was concave (Fig. 22). The cytoplasm was electron-dense with helically arranged, membrane-associated ribosomes in layers parallel to the spore wall (Figs. 19, 28). The membrane layers were 43-50 nm apart. Both in unfixed and stained spores a lucent area was visible in the posterior fifth of the spore (Figs.5-6, 8-9). In transmission electron microscopy this area often appeared shrunken (Fig. 17). Membrane-lined compartments with a uniform dense content were present here, together with aggregates of 12-17 nm wide granules (Fig. 26). The dark spot visible in stained spores corresponded obviously to the granular aggregates (Fig. 6).
Discussion
Identity of the Species According to the original description of]irovecia (Mrazekia) caudata, based on material collected in France, the dimensions of the spores are 1.3-1.4 X 16-18 urn, and the tails are approximately equally long [9]. Lorn, using material with provenance from Czechoslovakia, reported the dimensions 1.5-2 X 15-16.6!-tm (spores), and 16-17 urn (tails) [10]. Spores of the Swedish microsporidium measured 1.6-2.1 X 15.4-17.5 urn, the tails 10.5-17.5 urnin unfixed condition. Fixed spores measured 1.0-1.8 X 11.2-15.5 urn, fixed tails 9.0-15.0 urn. Both series are from the same host specimen. In a second series of
Cytology of [irovecia caudata . 329
measurements, from another specimen of the same host least closely related. As the description of]. caudata [9] is species and collectedat the same locality but in a different not of today's standard, minor differences between these year, the spores were slightly greater, and the tails were microsporidia should not be exaggerated. Until j. caudata has been isolated from the type host in France, described considerably longer. There is a greater sizerange in the Swedishmaterial, and adequately, and found to be different from the Swedish it contains smaller spores than the samples from France microsporidium, strong evidence for treating this as a new and Czechoslovakia. The greatest spores are of the same ]irovecia species is lacking. size, however, and the proportions between tails and spores equal them of the description. A re-evaluationof the Cytology type material, measured using identical techniques, might The cytologyof the species treated herein is similarto the show a greater variation in dimensions, and it would also allow a closer comparison of the cytology. No micro- cytology of other ]irovecia and Bacillidium species from graphs of this species have been published, and the line which details are known [1, 7, 8, 12, 14]. The straight part drawings of the description are not sufficient for such of the polar filament, often called the manubrium, has a comparison. However, the collection of L. Legerhas been wide mottled layer (Figs. 19,25), which seems to be unique destroyed (Degrange, personal communication), so for the to these microsporidia. It is further characteristic that a zone with successively reduced diameter separates the wide moment the brief description is all we have. Another complication is the different host species. The anterior part of the filament from the posterior narrow coil type host is Tubifex tubifex, while the Swedish microspo- (Fig. 17; Fig.2 in [1], Fig.9 in [14]), while in microsporidia ridium was isolated from Limnodrilus hoffmeisteri and with an anisofilar polar filament, the filament is abruptly from an unidentified species of Tubificidae. However, constricted. At the light microscopiclevel it is typical that Leger and Hesse [9] and Kudo [5] were of the opinion that material at the posterior pole of the spore takes stain using ]irovecia (Mrazekia) caudata was identical to the organism traditional Giemsa and Heidenhain techniques (Fig. 6; Mrazek had described from Limnodrilus under the name Figs. 7, 9-10 in [6]). Myxocystis [11], which means that they did not consider]. The exospore of ]irovecia and Bacillidium species caudata to be host specific. Mrazek's publication has three differentiates at the sporoblast stage into a surface layer illustrations of microsporidia of the ]irovecia-Bacillidium- similar to a unit membrane and a wide uniformly dense type (Figs. 7, 20, 21 in [11]), from hosts of the genera basal layer (Fig. 28). However, the fate of the surface layer Limnodrilus and Lumbriculus. Figure 21, which shows is different in different species. In B. filiferum the surface spores of ]. caudata-type, is from Lubriculus not from layer partially losescontact with the basal layer and forms Limnodrilus. There is no indication of the magnification, tubule-like projections [7]. The surface layer of]. involuta and the size of the spores is unknown. jirovec studied completely loses contact with the basal layer and forms a Mrazek's slides and found that spores from Lumbriculus continuous sac, similar to a sporophorous vesicle [8]. In were 7.5-9 urn long, spores from Limnodrilus 8-10 urn the microsporidium treated herein the surface layer long - approximately half the length of the spores of ]. remains unchanged from the sporoblast stage (Fig. 4) to caudata [4]. He described these microsporidia as two new the mature spore (Fig. 28). species: Mrazekia lumbriculi and M.limnodrili [4], both of The tail-like spore projection of the ]irovecia species of them now recognized as ]irovecia species [6]. Actually this paper is formed from exospore material alone (Fig. Tubifex tubifex is so far the only host from which ]. 18), likethe tail of]. involuta (Fig. 20 in [8]).In this respect caudata has been reported. they differ from]. (Mrazekia) brevicauda, where the tail is We know little about the host specificity of microspo- formed by a constriction of the entire spore, with the ridia, and nothing at all about the host range of micro- consequence that the complete spore wall, with plasma sporidia from Oligochaeta. Many microsporidia are con- membrane, endospore and exospore, is present in the tail sidered host specific, and quite a number have been (Fig. 1 in [1]). The two waysof making the tail indicate that described as new to science for the reason that they were the genus ]irovecia in its present sense is heterogeneous. Lorn interpeted the developmental stages of ]irovecia isolated from a new host, even if they could not be distinguished from microsporidia of related hosts using (Mrazekia) caudata, including the mature spore, to be uninucleate [10], but this was obviously erroneous. In the morphological criteria. As we have clear evidence that some microsporidia, like Nosemaacridiophagus, are com- description of the species nothing specific is mentioned petent to infect a number of related species, also in about the nuclear condition [9]. However, in the same differentgenera [2],we should be careful when postulating paper, dealing with a number of new Mrazekia species, Legerand Hesse discussed the characteristicsof Mrazekia, strict host specificity. The species treated herein is clearly different from selectingM. argoisi as typical for the genus. In the spore of ]irovecia (Mrazekia) lumbriculi, J. (M.) limnodrili and ]. M. argoisi is: "... le germe, globuleux et tres petit avec un (M.) ilyodrili, for all of them have spores less than 10 urn minuscule noyau gernine ... " [9]. Obviously this is a long [47]. In addition]. lumbriculi, not]' limnodrili, has description of a diplokaryon, and it seems clear that they spores of similar shape to ]. caudata and the Swedish considered the diplokaryotic condition to be characteristic microsporidium [11, 4]. ]. caudata is the only existing for all Mrazekia species, including M. caudata. Also species where both the size and the shape of the spore are jirovec was aware that all developmental stages of these similarto the Swedish microsporidium, and the hosts are at microsporidia are diplokaryotic [4], and most later inves-
330 .
J. I. R. Larsson
tigations on Bacillidium and ]irovecia species have verified the diplokaryotic nature. The diploka ryotic condition of the species treated herein does not contradict that it is identified as J. caudata.
Acknowledgements The author is greatly indebted to Mrs. Lina Hansen and Mrs. Inger Norling, both at the Department of Zoology, University of Lund, Sweden, for skilful technical assistance, and to Prof. C. Degrange, Universite Scientifique et Medicale de Grenoble, for information about the fate of Prof. L. L. Leger's collection. The investigation was financially supported be research grants from the Swedish Natural Science Research Council.
References 1 Gorz P. (1981): Homology of the manubrium of Mrazekia brevicauda and the polar filament of other microsporidia. Z. Parasitenkd., 64, 321-333. 2 Henry J. E. (1967): Nosema acridiophagus sp. n., a microsporidian isolated from grasshoppers. J. Invertebr. Pathol., 9, 331- 34 1. 3 Hostounsky Z. and Zizka Z. (1979): A modification of the "agar cushion method " for observation and recording microspor idian spores. J. Protozool. , 26, 41A-42A.
4 [Irovec O. (1936): Zur Kenntnis von in Oligochaten para sitierenden Mikrosporidien aus der Familie Mrazekidae. Arch. Protistenkd., 87, 314-344. 5 Kudo R. (1924): A biologic and taxonomic study of the microsporidia. Ill. Biol. Monogr. , 9(2/3), 1-268. 6 Larsson J. I. R. (1986): On the taxonomy of Mrazekidae: resurrection of the genus Bacillidium Janda, 1928. J. Protozool., 33, 542-546. 7 Larsson J. I. R. (1989): The light and electron microscopic cytology of Bacillidium filiferum sp. nov. (Microspora, Bacillidiidae). Arch. Protistenkd., 13 7, 345- 355. 8 Larsson J. I. R. (1989): Light and electron microscope studies on ]irovecia involuta sp. nov. (Microspora, Bacillidiidae), a new microsporidian para site of oligochaetes in Sweden. Europ. J. Protistol., 25, 172-181. 9 Leger L. et Hesse E. (1916): Mrazekia, genre nouveau de Microsporidies a spores tubuleuses. C. R. Soc. BioI., 79, 345-348. 10 LornJ. (1958): K poznanf vyvojoveho cyclu Mrdzekia caudata Leger et Hesse 1916. Cesk, Parasitol., 5, 147-152. 11 Mrazek A. (1910): Sporozoenstudien. Zur Auffassung der Myxocystiden. Arch. Protistenkd., 18,245-259. 12 Puytorac P. de (1962): Observations sur l'ultrastrueture de la Microsporidie Mrazekia lumbriculi, Jfrovec. J. Microscopie, 1,39-46. 13 Romeis B. (1968): Mikroskop ische Technik. Oldenbourg Verlag, Munchen and Wien. 14 Vavra J., Joyon L. et Puytorac P. de (1966): Observation sur I'ultra structure du filament polaire des Microsporidie s. Protistologica, 2, 109-112. 15 Weiser J. (1977) : Contribution to the classification of microsporidida. Vestn. Cs. spol. Zool., 41, 308-320.
Key words: ]irovecia caudata - Mrazekia - Microsporidia - Oligochaeta - Ultrastructure Ronny Larsson, Department of Zoology, University of Lund, Helgonav. 3, S-223 62 Lund, Sweden