Colonization and digestion of nematodes by the endoparasitic nematophagous fungus Drechmeria coniospora

Mycol. Res, 95 (7): 873-878 (I991)

873

Printed in Great Britain

Colonization and digestion of nematodes by the endoparasitic nematophagous fungus Drechmeria coniospora

JAN DIJKSTERHUIS/ WIM HARDER/ URS WYSS 3 AND MARTEN VEENHUIS 1 Laboratory of Electron Microscopy and 2 Department of Microbiology, Biological Centre, University of Groningen, Kerklaan 30, 9751 NN Haren, The Netherlands 3 Institute of Phytopathology, Christian Albrecht University, Hermann-Rodewaldstrasse 9, 0-2300 Kiel 1, Germany 1

Infection of nematodes by conidia of Drechmeria coniospora was studied by optical- and electron microscopy. After penetration, trophiC hyphae invaded solely via the pseudocoel; penetration of internal organs was never observed. The invading hyphae had a characteristic wave-like appearance, this pattern possibly having the advantage of preventing rupture of hyphae due to host movements. After death of the host 30-48 h after infection, trophic hyphae contained numerous lipid droplets often associated with microbodies (peroxisomes) which were characterized by the presence of catalase and the l3-oxidation enzyme thiolase. Conidiophores developed from tips of trophiC hyphae; the process of outgrowth being similar to that for initial penetration in that it occurred via enzymic action and mechanical force. Intimate association between fungal cell wall and nematode cuticle suggested that leakage of nematode contents was prevented after outgrowth. Conidiophores possessed an electron dark layer at the outside of their wall which was not continuous with developing conidia. Numerous spores were successively formed from each individual peg present on the conidiophore. Pegs were ordered at regular distances, located in close proximity to septa and on the apical end of the conidiophore. Invariably immature spores are formed; development of the adhesive knob occurred after release from the peg. Approximately 5000-10000 spores were formed at the expense of a single nematode.

The obligate nematophagous parasite Drechmeria coniospora (Drechs.) Cams & Jansson (Cams & Jansson, 1985) is able to infect nematodes by free, mature conidia (Barron, 1977) which adhere to the cuticle by means of an adhesive knob (Jansson & Nordbring-Hertz, 1983, 1984). Previous infection studies conducted with the nematode Panagrellus redivivus Coodey showed that conidia adhere predominantly to the anterior and posterior end of the body in the vicinity of the sensory organs (Jansson & Nordbring-Hertz, 1983; Dijksterhuis, Veenhuis & Harder, 1990). Prior to penetration, an appressorium develops on the adhesive knob (Saikawa, 1982 a) which plays a key role in the penetration process (Dijksterhuis et al., 1990). Penetration is thought to be the result of a simultaneous enzymic weakening of the cuticle and exertion of mechanical force (Dijksterhuis et al., 1990); an identical mechanism has been proposed for penetration by the predatory fungus Arthrobotrys oligospora Fres. (Veenhuis, Nordbring-Hertz & Harder, 1985). Little is known of either subsequent infection stages and digestion of nematodes or of conidiogenesis (Jansson, von Hofsten & von Mecklenburg, 1984). In the course of cur studies on the ecophysiology of D. coniospora, we have investigated these aspects in detail by means of a combined optical- and electron microscopical study, the results of which are presented in this paper.

MATERIALS AND METHODS Organisms and growth conditions Drechmeria coniospora was maintained on 2 % com meal agar in Petri dishes at 20 DC as detailed earlier (Dijksterhuis el al., 1990). In the interaction experiments the nematode Panagrellus redivivus, maintained axenically on soya peptone-liver extract medium (Nordbring-Hertz, 1977), was used.

Interaction experiments Nematodes were harvested, washed once in distilled water, and added directly to fungal cultures (100-500 nematodes/ plate). The process of attachment and infection was followed by optical microscopy. Samples were taken at regular time intervals and processed for video-enhanced optical microscopy or electron microscopy by the methods described previously (Dijksterhuis et al., 1990).

Quantitative estimation of conidium production Nematodes in the final stages of infection were used. These were obtained as follows: newly infected nematodes were transferred to 2 % water agar while still alive, and incubated for 3-4 d at 20 0 until digestion was completed. Subsequently individual organisms were carefully removed with a minor amount of adhering water agar with a microspatula, and

Infection of nematodes by Drechmeria coniospora

874

Figs 1, 2. Development of hyphae of D. coniospora inside a nematode visualized by video-enhanced light microscopy. Fig. 1. Median view of an infected nematode showing hypha! growth through the pseudocoel (arrow). Fig. 2. The same nematode, focused in a higher plane showing the wavy growth pattern of the trophic hyphae (arrow). Fig. 3. Ultrathin cross-section through an infected nematode; hyphae do not penetrate the oesophagus. Abbreviations: C Cuticle; Cw, Cell wall; Co, Conidiophore; Cp, Conidiiferous peg; D, Electron dense layer; Er, Endoplasmic reticulum; G, Glycogen; L, Lipid droplets; M, Mitochondrion; 0, Oesophagus; V, Vacuole. Bar = 0'5 IJm, unless otherwise indicated. resuspended in 300 III water. The number of conidia produced was determined using 50 III of the water suspension, with a heamocytometer. Nematodes of different sizes (namely eleven larger and fifteen small(er) ones) were counted.

Optical microscopy Optical microscopical studies were conducted using an enhanced contrast video system (Wyss & Zuncke, 1986). Before examination, infected nematodes were fixed as described previously (Dijksterhuis et al., 1990). In addition, living infected nematodes were embedded in 8 % gelatine and studied without further treatment.

Cytochemistry and immunocytochemistry Cytochemical staining for localization of acid phosphatase and catalase activities were performed by methods described earlier (van Dijken et aI., 1975; Veenhuis, van Dijken & Harder, 1980; Nordbring-Hertz, Veenhuis & Harder, 1984). Polysaccharides were stained by the method of Thiery (1967) on ultrathin sections of Lowicryl-embedded samples. The subcellular localization of 3-oxo-acyICoA thiolase protein was studied by immunogold labelling according to the method of Slot & Geuze (1984). For these studies antibodies raised against the enzyme from Neurospora crassa (obtained from Professor Dr W. H. Kunau, University of Bochum, Germany) were used.

Electron microscopy Infected nematodes were fixed, embedded, sectioned and examined as described in a previous paper (Dijksterhuis et al. 1990) with the exception that where nematodes in the final stages of infection were used, these were embedded in 2 % water agar prior to fixation in order to prevent removal of conidia from their pegs during subsequent steps of the fixation and embedding procedure. For SEM, infected nematodes were fixed in a mixture of 1'5% (w/v) formaldehyde+2% (v/v) glutaraldehyde in 0'1 M sodium cacodylate buffer (pH 7'2) and mounted on small agar cubes as described before (Dijksterhuis et al., 1990) and examined in a lSI DS 130 scanning electron microscope. For low-temperature SEM, infected moribund nematodes upon which conidiophores had not yet formed were transferred to 4-5 % water-agar blocks (approximately 5 x 5 mm). At different stages during the development of conidiophores and conidia, the blocks were frozen in nitrogen slush and examined in a Jeol 840 A scanning electron microscope fitted with a Hexland CT 1000 cryotransfer system.

RESULTS Invasion was confined to the pseudocoel (Figs 1, 2); penetration of the oesophagus never being observed. Invading hyphae exhibited a characteristic wavy appearance (Fig. 2). Observations on living, infected nematodes by videoenhanced contrast microscopy clearly demonstrated that

J. Dijksterhuis and others

875

O·51lffi Figs 4-7. Synthesis of storage polymers and microbodies inside trophic hyphae. Fig. 4. Optical micrograph showing accumulation of lipid droplets approximately 50 h after infection. Fig. 5. Electron micrograph of a trophic hypha showing lipid droplets, glycogen and microbodies (arrows). Fig. 6. Immunocytochemical demonstration of the presence of thiolase protein in the matrix of microbodies (arrow); after incubation of thin sections of Lowicryl-embedded cells with ex-thiolase and protein AI gold specific labelling is confined to microbodies. Fig. 7. The presence of catalase activity was demonstrated after incubation of intact cells with 3-3' diaminobenzidine (DAB) and H 2 0 2 .

movements of newly infected nematodes, especially at their anterior end, could result in full stretching of these sinusoidal hyphae. However, rupture of them as a result of these movements was never observed. Therefore, this growth pattern may have the advantage that it prevents damaging, e.g. rupture (as, for instance, has been observed for linear growing trophic hyphae of Nematoetonus leiospora; Barron, 1977) during their initial development in the living host. That only the pseudocoel was penetrated was confirmed by electron microscopy (Fig. 3). The oesophagus was never penetrated by trophic hyphae, even during later stages of infection. In the course of infection, nematode movement gradually slowed down and 40-48 h after the onset of infection the majority of nematodes were moribund or dead. Almost simultaneously, development of conidiophores through the cuticle became apparent; this occurring initially at the anterior and/or posterior end of the body. Video-enhanced contrast microscopy revealed that at this stage the body mid-section (and in the case of females also the tail) was not yet invaded by trophic hyphae and appeared to be in a virtually unaffected state. Outgrowth of conidiophores started at those parts of the nematode which were filled with hyphae, at the time when colonization of the nematode was far from complete.

Conidiophore formation was associated with the accumulation, inside trophic hyphae, of storage materials in the form of lipid droplets together with glycogen (Figs 4-7). Lipid droplets were abundantly present during the later stages of infection (50 h of infection) (Fig. 4) and were often observed in close association with strands of endoplasmic reticulum and clusters of small microbodies (Fig. 5). The latter organelles were characterized by the presence of catalase activity and the j3-oxidation pathway enzyme thiolase (Figs. 6, 7). Conidiophores developed from the tips of trophic hyphae. For this purpose these hyphae became bent and grew towards the inner layer of the cuticle. Prior to outgrowth, hyphal tips became swollen into a bulb appressed perpendicularly against the cuticle (Fig. 9, inset). This process of bulb formation generally took approximately 2 h before the initial development of a very thin outgrowing hyphae was observed. Ultrathin sections of this stage showed that the bulb was intimately associated with the inner layer of the cuticle (Fig. 8). The wall of the hyphal tube which actually penetrated the cuticle originated from underneath the cell wall of the bulb. The originally slender conidiophore pierced the cuticle, its cell wall staying in close contact with the cuticle during and after this process (Fig. 9). Acid phosphatase activity was apparent at the outgrowth site (Fig. 10). After penetration of the cuticle,

876

Infection of nematodes by Drechmeria coniospora

Figs 8, 9. Conidiophore development. Fig. 8. Initial stage of conidiophore formation, showing the developing outgrowing hypha shortly prior to penetration of the cuticle. Fig. 9. Conidiophore grown through the cuticle. Note the electron dense layer surrounding the conidiophore and the intimate association between the nematode cuticle and the fungal cell wall at the site of outgrowth (arrow). The inset in Fig. 9. shows a video-enhanced contrast optical micrograph of an early stage of conidiophore development. Note the bulb formed by the trophic hypha (arrow). Bars = 0'5 IJm.

conidiophores rapidly developed; their walls being covered with an electron-dense layer, which was invariably absent from the trophic hyphae from which it developed (Fig. 9). In scanning electron micrographs this layer showed a rough surface (Fig. II). Combined observations suggest that the hole in the cuticle at the outgrowth site is probably effectively sealed. Additional evidence that conidiophore development is not associated with cuticle degradation is suggested by the observation that larvae could not escape from infected females during the different stages of conidiophore development (not shown). At later stages of development conidiophores became septated and pegs were formed near the septa and at the hyphal tip (Figs 12, 14). Transmission electron micrographs of two different stages of conidium development are shown in Figs 12 and 13. The dark outer layer present on the conidiophore wall and pegs was absent on newly developing conidia. The fully developed spore (albeit lacking an adhesive knob) was separated from the peg by septum formation and

Fig. 10. Demonstration of acid phosphatase activity at the base of the conidiophore. Bar = 0'5 IJm. Fig. 11. Scanning electron micrograph showing the rough surface of developing conidiophores.

.-

Figs 12, 13. Electron micrographs of conidiophore development. Fig. 12. Early stage of conidium development. Note that the electron dense outer layer, typical for conidiophore and peg, is not continuous with the conidium-initial (arrow). Fig. 13. Thin section through a fully developed conidium, prior to liberation; a septum (arrow) has formed between peg and spore. Bars = 0'5 IJm.

J. Dijksterhuis and others

877

the conidia as evidenced by formation of the adhesive knob, occurred after release (Figs 16, 17). Quantitative measurements revealed that digestion of a single nematode may yield 5000-10000 conidia produced over several days.

DISCUSSION

Figs 14-17. Low-temperature scanning electron micrographs of different stages of conidiogenesis. Fig. 14. Developing conidiophore during formation of conidiiferous pegs. Note that all pegs are formed at the same time. Fig. 15. After development of the first conidium new conidia are formed on the same peg (arrows). Fully developed conidia are liberated as the result of the next spore but generally maintain attached to them, resulting in the formation of clusters of spores. Figs 16 and 17 show two subsequent stages of this cluster formation. After liberation knobs (some of these are indicated by an arrow) are formed.

As an obligate parasite of nematodes the fungus D. coniospora is dependent on the presence of nematodes for survival. All observations made in this study clearly favour the idea that the main strategy of the fungus is not the production of vegetative hyphae but instead, to produce conidia - at the expense of the host - at the highest possible rate. This view is based on the following observations: (i) invasion is highly efficient. The sole routing of trophic hyphae through the pseudocoel offers little resistance to their development and therefore may be energetically advantageous; (ii) conidiophore formation is already initiated when colonization of the nematode body is still far from complete; this way the amount of biomass, not immediately involved in conidia production, remains relatively low; (iii) the mechanisms involved in outgrowth of these hyphae are designed to prevent leakage of the nematode contents and (iv) conidiospore production is very efficient by the successive formation of numerous spores from one and the same peg until the nematode contents is exhausted. As already indicated, maturation of the conidia occurred after their release from the conidiiferous peg thus shortening the time required for their production. Subsequent maturation of the spores is dependent on environmental conditions (P. H. J. F. van der Boogert and J. Dijksterhuis, unpublished results). The mechanisms observed are not unique for D. coniospora but resemble those observed in other nematophagous fungi, e.g. Arthrobotrys oligospora, in different aspects. Main similarities include (i) the mechanisms of penetration and sealing of penetration sites; (ii) the concurrent development of trophic hyphae and vegetative mycelium outside the nematode body and (iii) the subcellular morphology of the trophic hyphae. As in other species (A. oligospora; Veenhuis et al., 1989 b, Dactylella Iysipaga; Wimble & Young, 1984, Verticillium balanoides; K. Sjollema, J. Dijksterhuis & M. Veenhuis, unpublished results) trophic hyphae of D. coniospora are characterized by the presence of numerous lipid droplets, often closely associated with several microbodies. Part of these lipid/microbody complexes are donated to each individual conidium formed. The presence of enzymes of the [3-oxidation pathway in these organelles suggests that the endogenous lipids at least partly serve as energy supply, for instance required during maturation (formation of the adhesive knob) but also to facilitate the initial penetration after capturing (Veenhuis et al., 1989a; Dijksterhuis et al., 1990).

J. D. is supported by the Netherlands Integrated Soil Research was subsequently liberated as the result of the development of a next conidium on the same peg. (Fig. 15). By this mechanism numerous conidia successively developed from the same peg often organized in large clusters (Figs 16, 17). Maturation of

Programme. We gratefully acknowledge the kind hospitality of Dr R. A. Samson and M. 1. van der Horst (Centraal Bureau voor Schimmelcultures, Baarn, the Netherlands) and Dr P. Staugaard for the expert advice and help during low temperature scanning electron microscopy.

Infection of nematodes by Drechmeria coniospora We also thank Dr B. Nordbring-Hertz, Lund, Sweden and Dr P. H. J. F. van der Boogert, Dr L. Brussaard and Dr K. B. Zwart for their interest and stimulating discussions and Jan Zagers and Klaas Sjollema for skilful technical assistance.

REFERENCES Barron, G. L. (1977). The Nematode-destroying Fungi. Topics in Mycobiology. No. I. pp. 82, 83, 109. Guelph. Ontario, Canada: Canadian Biological Publications Ltd. Dijken. j. P. van, Veenhuis, M., Venneulen, C. A. & Harder, W. (1975). Cytochemical localization of catalase activity in methanol grown Hansenula polymorpha. Archives Microbiology 105, 261-267. Dijksterhuis.}.. Veenhuis. M. & Harder. W. (1990). Ultrastructural study of the adhesion and initial stages of infection of nematodes by conidiospores of Drechmeria coniospora. Mycological Research 94 (1). 1-8 Gams. W. & jansson. H. B. (1985). The nematode parasite Meria conlOspora Drechsler and its classification. Mycotaxon XXII. 33-38. jansson. H. B. & Nordbring-Hertz. B. (1983). The endoparasitic nematophagous fungus Meria coniospora infects nematodes specifically at the chemosensory organs. Journal of General Microbiology 129.1121-1126. jansson. H. B. & Nordbring-Hertz. B. (1984). Involvement of sialic acid in nematode chemotaxis and infection by an endoparasitic nematophagous fungus. Journal of General Microbiology 130. 39-43. jansson. H. B.• Hofsten. A. von & Mecklenburg, C. von (1984). Life cycle of the endoparasitic nematophagous fungus Meria coniospora: a light and electron microscopic study. Antonie van Leeuwenhoek 50. 321-327. (Received for publication 17 September 1990 and in revised form 4 January 1991)

878 Nordbring-Hertz. B. (1977). Nematode-induced morphogenesis in the predacious fungus Arthrobotrys oligospora. Nematologica 23. 443-451. Nordbring-Hertz. B.. Veenhuis. M. & Harder. W. (1984). Dialysis membrane technique for ultrastructural studies of microbial interactions. Applied Environmental Microbiology 47. 195-197. Saikawa. M. (1982). An electron microscope study of Meria coniospora. an endozoic nematophagous Hyphomycete. Canadian Journal of Botany 60. 2019-2023. Slot. j. W. & Geuze. H. j. (1984). Gold markers for single double immunolabelling of ultrathin cryosections. In Immunolabelling for Electron Microscopy (ed. j. H. Polak & I. H. Varndell), pp.129-142. Elsevier: Amsterdam. Netherlands. Thiery. j. P. (1967). Mise en evidence des polysaccharides sur coupes fines en microscopie electronique. Journal microscopie 6. 987-1018. Veenhuis. M .. Di;ken. j. P. van & Harder. W. (1980). A new method for the cytochemical demonstration of phosphatase activities in yeast based on the use of cerous ions. FEMS Microbiology Letlers 9. 285-291. Veenhuis. M .. Nordbring-Hertz. B. & Harder. W. (1985). An electronmicroscopical analysis of capture and initial stages of penetration of nematodes by Arthrobotrys oligospora. Antonie van Leeuwenhoek 51. 385-398. Veenhuis. M .. Sjollcma. K. A.. Nordbring-Hertz. B. & Harder. W. (1989a). An improved method for light- and electron microscopical studies of nematode/fungal interactions. Antonie van Leeuwenhoek 55. 361-368. Veenhuis. M .. Harder. W. & Nordbring-Hertz, B. (1989b). Occurrence and metabolic significance of microbodies in trophic hyphae of the nematophagous fungus Arthrobotrys oligospora. Antonie van Leeuwenhoek 56. 241-249. Wimble. D. B. & Young. T. W. K. (1984). Ultrastructure of the infection of nematodes by Dactylella lysipaga. Nova Hedwigia 40. 9-29. Wyss. U. & Zuncke. U. (1986). The potential of high resolution videoenhanced contrast microscopy in nematological research. Revue nematologie 9. 91-94.