Author for correspondence: M. Saito Tel: +81 287 37 7227Fax: +81 287 36 6629Email: email@example.com
• A tubular vacuolar system is reported here for the first time in living hyphae of Gigaspora margarita, an arbuscular mycorrhizal fungus, during various phases in the development of a symbiotic relationship with onion (Allium cepa) seedlings.
• Germ tubes, extraradical hyphae and intercellular hyphae were labeled with Oregon Green 488 carboxylic acid diacetate and observed by laser scanning confocal microscopy. Emphasis was placed on the relationship between the shape of vacuoles and the presence of cytoplasmic streaming.
• In germ tubes, labeled vacuoles showed a variety of profiles, including spherical and tubular (< 0.5 µm diameter), with various compositions of these shapes along the length of the germ tubes. The tubular vacuoles rarely interconnected with spherical vacuoles and often formed longitudinally oriented, elongated bundles. The tubular vacuolar system appeared to be associated with cytoplasmic streaming, whereas spherical vacuoles were not. Tubular vacuoles were observed in all regions of the germ tubes and were also observed in both extraradical and intercellular hyphae.
• The results question the hypothesis that discrete vacuoles may be involved in the translocation of polyphosphate along hyphae of arbuscular mycorrhizal fungi.
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
Arbuscular mycorrhizal fungi are obligate symbionts, depending exclusively on their hosts for carbon. In return, they transfer mineral nutrients, most importantly phosphate, to the host. The site of nutrient exchange is within the root, and extraradical hyphae distributed in soil extend the absorbing surfaces of roots. Therefore, long-distance translocation occurs within extraradical hyphae, where phosphate absorbed from the soil solution is directed to the roots and carbohydrates to hyphal tips and developing spores. In both arbuscular and ectomycorrhizal fungi, polyphosphate has been demonstrated in hyphae by biochemical methods, including nuclear magnetic resonance (NMR) spectroscopy (Ashford et al., 1994; Shachar-Hill et al., 1995; Solaiman et al., 1999), and in vacuoles by histochemical methods (Ashford et al., 1975; Cox et al., 1975; Ling-Lee et al., 1975; Ezawa et al., 2001) and X-ray microanalysis (Ashford et al., 1986; Peterson & Howarth, 1991; Frey et al., 1997). Since bidirectional cytoplasmic streaming has been observed in both germ tubes and extraradical hyphae of arbuscular mycorrhizal fungi, one hypothesis is that polyphosphate is in discrete compartments, namely vacuoles, and that these vacuoles are moved by cytoplasmic streaming (Cox et al., 1980; Harley & Smith, 1983; Smith & Read, 1997).
Recently, a new approach to investigating fungal vacuoles has been established using fluorescent probes such as 5-(and-6)-carboxy-2′,7′-dichlorofluorescein diacetate (carboxy-DCFDA) and Oregon Green 488 carboxylic acid diacetate (carboxy-DFFDA) (Ashford, 1998). Nonfluorescent carboxy-DFFDA permeates the plasma membrane and is hydrolysed by nonspecific esterase in the cytoplasm to release the fluorescent product. This product (Oregon Green 488 carboxylic acid) then accumulates across the tonoplast by an anion transporter and remains in the lumen of the vacuole (Cole et al., 1997; Cole et al., 1998). Using fluorescent probes, a complex vacuole system has been demonstrated in living hyphae of the ectomycorrhizal basidiomycete fungus, Pisolithus tinctorius (Shepherd et al., 1993; Allaway & Ashford, 2001). The vacuolar system was motile and composed of both spherical and tubular forms. The difference in the vacuoles’ appearance between those studies and previous transmission electron microscopy (TEM) observations was explained by the fact that conventional protocols involving chemical fixation and ethanol dehydration did not retain the tubular structure (Ashford, 1998). The presence of tubular vacuoles was also observed in isolates of Oomycota, Basidiomycota, Ascomycota, and Zygomycota (Rees et al., 1994). In that study, three isolates of Mucorales (Zygomycota) were used. Arbuscular mycorrhizal fungi in the order Glomales also belong to the Zygomycota. Smith and Read (1997) suggested the existence of tubules in germ tubes of Gigaspora margarita, based on the unpublished observations of S. Dickson & A. E. Ashford. No details of the vacuolar system labeled with fluorescent probes in living hyphae of members of Glomales have been reported. Therefore, the objective of this study was to observe the vacuolar system in living germ tubes as well as in intra- and extra-radical hyphae of G. margarita, while examining the relationship between vacuolar shape and cytoplasmic streaming.
Materials and Methods
Oregon Green 488 carboxylic acid diacetate (carboxy-DFFDA, Lot 3961-5; Molecular Probes Inc., Eugene OR, USA) was dissolved in dimethyl sulfoxide (DMSO) at a concentration of 20 mm for a stock solution. The solution was divided into microtubes holding 5 µl each and stored in a freezer at −20°C. A working solution was made by adding 2.5 ml of 50 mm citrate buffer (pH 5) to each 5 µl portion of the stock solution (final concentration, 40 µm).
The isolate of G. margarita (MAFF 520054; Ministry of Agriculture, Forestry and Fisheries Genebank, Tsukuba, Japan) used in this study had been propagated by pot-culture with white clover (Trifolium repens L.). The potting medium (soil–sand, 1 : 1) was kept for more than 1 month at 4°C. Before use, a portion of the soil was incubated at 42°C for 3 d to enhance germination of spores (Saito & Marumoto, 2002). Spores were collected from the soil by wet sieving and sugar gradient centrifuge (400 g for 30 s) and cleaned ultrasonically to remove debris from the surface.
Autoclaved a soil, sand and a commercial horticulture medium were mixed together (Saito, 1995) and embedded in autoclaved 1% agar in a plastic Petri dish. Pieces of cellulose-acetate membrane (10 × 15 mm, 0.8 µm pore size) were arranged on the surface, and three or four spores were placed on each membrane. The plates were inverted and incubated in the dark at 23°C for 8 h followed by 25°C for 16 h daily for more than 10 d.
The membrane with germ tubes was transferred in a drop of DFFDA on a glass slide. Under a dissecting microscope, drops of fluorescent probe were further added to the surface, and the germ tubes were carefully removed from the membrane using needles. The slide was incubated for 30 min at 25°C and then rinsed with the buffer by adding it from one side of the cover slip and removing it with a piece of filter paper from the other side. The slide was left for more than 1 h. After sealing the edge of the cover slip with nail polish, the slide was observed by a laser scanning confocal microscope (LSM510, Carl Zeiss, Jena, Germany) configured for fluorescein isothiocyanate (FITC) (Ar., 488 nm; LP505) and under differential interference contrast (DIC) optics.
Extra- and intra-radical hyphae
An onion seedling was inoculated with G. margarita in a plastic Petri dish (1.5 cm deep, 9 cm diameter, Fig. 1). The same soil mixture used for germ tube production was added to the bottom, and on top of this soil was laid a paper filter and then a cellulose acetate membrane (45 mm diameter, 0.8 µm pore size). An onion seedling and 20 spores were placed on the membrane, and these were covered by the lid. The dish was covered for shading and incubated in a growth cabinet for 4–6 wks with a 23°C, 8 h dark/25°C, 16 h light cycle using a photosynthetic photon flux density (PPFD) of 465 µmol m−2 s−1.
Extraradical hyphae growing on the cellulose acetate membrane and in the soil were used for labeling. These hyphae were collected from the soil by the following method. The lid of a Petri dish was removed and water was added from soil edge, causing some of the soil to flow into another Petri dish. Fine forceps were dragged slowly through the soil in the water under a binocular microscope, so that extraradical hyphae were caught by the tips of the forceps. The forceps were then used to carefully remove soil particles from the hyphae. The hyphae were kept in water. Loading the hyphae with the probe and observation were the same as for the germ tubes.
For intraradical hyphae, onion roots from the pot were hand-sectioned in water. The sections were immediately transferred to the probe, incubated for 30 min at 25°C, and then rinsed with buffer. More than 1 h after rinsing, the root pieces were transferred onto a glass slide, mounted in buffer and covered by a cover slip for observation.
Spores started to germinate from about 1 wk after incubation, and germ tubes grew sparsely on the membrane. Most germ tubes did not penetrate the membrane pores. By dropping the probe solution onto the membrane, the germ tubes often became separated from the membrane filter and could be placed on a glass slide using fine needles. A few contaminating bacteria were seen around hyphae under a light microscope, but few other fungi grew within the incubation period.
Loading with DFFDA showed that vacuoles occupied most of the cytoplasm of the germ tubes from tip to base. The vacuoles showed a variety of profiles, including spherical and tubular (Figs 2 and 3). The tubular vacuoles were often very narrow, less than 0.5 µm wide, and these were rarely interconnected with large spherical vacuoles. The distribution of the two vacuole forms in a hypha was divided into three types: A, longitudinally oriented elongated tubules only (Figs 2e,f,j and 3a–d); B, various sizes of spherical vacuoles only (Fig. 2a); C, both tubular and spherical vacuoles in various proportions (Fig. 2b–d). In long stretches of cytoplasm in a germ tube, one of these types was observed continuously, or was replaced in part by the other types. Tubular vacuoles were observed along the length of the germ tubes from tip to base. These vacuoles were extremely fragile when exposed to laser and UV irradiation, as indicated by a rapid change in shape from tubular to small spheres (Fig. 2g,h). This fragility limited the time for observation of the tubular vacuole system in each sample.
The labeling of vacuoles with the probe seemed to occur only after incubation in the rinsing buffer. Just after being rinsed with the buffer, nuclei and cytoplasmic materials in some hyphae were labeled with the probe, but this labeling decreased as time in the buffer increased. Vacuole profiles then became clear. Type C vacuoles tended to be stained more quickly and showed high fluorescent intensity, whereas type A vacuoles often needed more time to be obvious. Hyphae that showed weak signals in longitudinal arrays of vacuoles often showed cytoplasmic streaming (Fig. 3g).
Cytoplasmic streaming was characterized using DIC optics. The components of cytoplasmic streaming recognized were: small particles (about 0.3 µm diameter) showing independent movement (Fig. 3g); lipid bodies (about 0.6–1.0 µm diameter) that often formed aggregates that were stained with Nile red (data not shown) and had high contrast under DIC (Figs 2I and 3a–c,g); and nuclei (Fig. 3g and Fig. S1). Each particle moved independently at various rate (up to about 5 µm s−1). The cytoplasmic streaming occurred bidirectionally, and it was observed that organelles crossed each other in the same focal plane. The movements seemed to occur along thread-like structures under DIC (Fig. 3d, red arrows) that had no labeling (Fig. 3e, red arrows); these movements were in the same direction along each track, at least for a while. Not all hyphae on a slide had cytoplasmic streaming, and the degree of streaming varied among hyphae. Observation under UV and laser caused the cytoplasmic streaming to cease.
Tubular vacuoles were located between the tracks, thread-like structures (Fig. 3d–f). It was observed that a tubular vacuole located between two tracks was disrupted, after irradiation, into a series of small spherical vacuoles that still arrayed themselves linearly between the tracks. Tubular vacuoles also appeared to be moving, but it could not be established that these movements were independent of the cytoplasmic streaming in the hyphae, because of the abundance of vacuoles and the movement of large numbers of particles beneath. In addition, the extreme fragility of the tubular vacuolar system when irradiated did not allow observation periods long enough to determine the independence of these vacuolar movements.
The relationships between vacuolar forms and the presence or absence of cytoplasmic streaming were investigated. Germ tubes were observed first under DIC optics to determine whether or not cytoplasmic streaming was present, and then the sites were observed by fluorescence mode to observe vacuoles. In the germ tubes in which cytoplasmic streaming was present, the vacuoles were either labeled or not. The labeled vacuoles were bundles of the tubular form, or of both forms. In the germ tubes in which cytoplasmic streaming had ceased, vacuoles were either not labeled or both types of vacuole shapes were observed.
Extra- and intraradical hyphae
Fungal colonization in the roots of onion seedlings occurred 3–4 wks after inoculation using this method (Fig. 1). Extraradical hyphae grew extensively in the soil in a Petri dish and were often found to be forming bundles at the bottom. New spore production was found on the hyphae after about 5 wks’ incubation.
The extraradical hyphae looked like the germ tubes, except that, on former, lipid bodies were often more abundant were sometimes up to approximately 6 µm in diameter. Although a more destructive method was used to collect extraradical hyphae than was used for the germ tubes, extraradical hyphae labeled with the probe showed vacuolar patterns similar to those seen in the germ tubes. Tubular vacuoles were also observed in some of the hyphae (Fig. 4a–d). Cytoplasmic streaming was also observed in some of the hyphae collected from cellulose acetate membranes and from soil in Petri dishes.
Lipid bodies up to approximately 6 µm in diameter were characteristic of intercellular hyphae (Fig. 4e,h) and of trunks of arbuscules of intraradical hyphae. Loading hand-sectioned colonized roots with DFFDA showed that the vacuoles were located between lipid bodies or between lipid bodies and plasma membrane, and had pressed spherical shapes. Tubular vacuoles were also observed in intercellular hyphae, though infrequently (Fig. 4f, 4g, 4i,j).
Fine branches of arbuscules were rarely labeled with the probe; the signal was very low if present at all. The resolution of laser scanning contocal microscopy (LSCM) was not sufficient to observe the interiors of the fine branches, especially when the fluorescent intensity of labeling was low. Therefore, the possibility of staining other components of arbuscules (e.g. cytosol, fungal cell wall and interfacial matrix) could not be excluded. In some cases, host vacuoles were labeled with the probe, outlining the arbuscule in silhouette (data not shown).
Tubular vacuoles in G. margarita
In preliminary experiments using germ tubes, the structure labeled with DFFDA was also labeled with carboxy-DCFDA, dipeptidyl peptidase substrate Ala–Pro–CMAC (Molecular Probes,) and CellTracker Blue CMAC (Molecular Probes), which are all known to label vacuole lumen of yeast and filamentous fungi (Cole et al., 1998). Ultrastructural observation of germ tubes also showed that vacuoles occupied most of the volume of the cell compartments of germ tubes and that endoplasmic reticulum (ER) lumen was less than 50 nm wide (data not shown). Therefore, the organelles labeled with DFFDA in germ tubes of G. margarita were thought to be equivalent to the vacuoles demonstrated in hyphae of P. tinctorius.
Because G. margarita cannot be propagated without the establishment of a symbiotic relationship, it was more difficult to handle the hyphae and methods involved cutting and transferring hyphae, which might be rather destructive. The vacuolar shapes in G. margarita hyphae varied from spherical to tubular. The tubular form, the longitudinal arrays and the tubular bundles observed in some hyphae were considered to reflect the true nature of vacuoles in this fungus for three reasons. First, these structures were observed in germ tubes, as well as in both extraradical and intercellular hyphae. Second, the diameter of the vacuolar tube system in the germ tubes (about 0.3 µm) was consistent with those observed in other phyla (0.24–0.48 µm) (Rees et al., 1994); those diameters were remarkably constant independent of the fungal species or the diameter of the hyphae. Third, cytoplasmic streaming occurred concomitantly with the presence of tubular vacuoles in germ tubes and in extraradical hyphae. Active hyphae, as indicated by cytoplasmic streaming, appeared to resist being labeled with this probe. In these active hyphae, tubular vacuoles tended to become labeled slowly compared with the spherical, and they had either tubular vacuoles only or no labeling.
By contrast to the finding that the tubular shape indicated integrity, it was harder to assess what the spherical shape indicated, considering the sensitive nature of cytoplasmic streaming to various stimuli and to observation. It is likely that not all hyphae on a membrane originally have cytoplasmic streaming, as shown in Gigaspora rosea (Bago et al., 1998) in which vacuoles may be spherical. However, the possibility that the spherical shape observed in this study was the result of handling could not be excluded. Staining hyphae on a glass slide has the advantage of allowing observation of a great deal of detail on the inside with at all DIC, UV and laser optics. Substrates commonly used for hyphal growth, such as agar and membrane, often cause high background staining, autofluorescence and/or low transparency. A more intact system is needed to determine the variations in the shapes of the vacuoles in this fungus.
Differences with the other taxa
The tubular vacuoles formed extensive longitudinal arrays along G. margarita hyphae. In P. tinctorius, the vacuole system was composed of tubules and large spherical vacuoles, which were interconnected with the tubules (Shepherd et al., 1993). The tubular vacuoles observed in this study, however, rarely interconnected with spherical vacuoles. Another characteristic of the vacuolar system of G. margarita was the large number of tubules that occupied most of the cytoplasm of the hyphae, which were reticulate or often formed extensive parallel arrays at the periphery. Rees et al. (1994) observed vacuole systems in the fungal phyla Basidiomycota, Ascomycota and Zygomycota, as well as in Oomycota. The vacuolar system observed in three isolates of Mucorales (Zygomycota) consisted of short, branched tubules connected with small vacuoles in the hyphal tip region, as well as large, more frequent and clustered vacuoles in more basal regions. Conversely, the vacuolar system in Oomycota (although this group was categorized into Chromista) was a reticulum of fine tubules at hyphal tips and clusters of spheres in more mature subapical regions (Rees et al., 1994; Allaway et al., 1997; Bachewich & Heath, 1999). These isolates belonging to one fungal phylum and to Oomycota lack septa in hyphae, similar to arbuscular mycorrhizal fungi, but the tubular bundle system of vacuoles seen in G. margarita has not been reported in other phyla.
Motility of the tubular vacuoles and cytoplasmic streaming
In P. tinctorius, the tubular vacuoles were motile, showing extension and retraction, whereas the spherical vacuoles were more stable (Shepherd et al., 1993). Tubular vacuoles have been proposed to play a role in intracellular translocation: transport via tubules permits the bulk flow of contents for relatively long distances without concomitant transport of the membrane (Ashford, 1998; Cole et al., 1998). Because of the difficulty of observing the tubular vacuole system in G. margarita hyphae, we could not be certain whether that system was autonomously motile or not.
Involvement of cytoplasmic streaming in the translocation of phosphate in arbuscular mycorrhizal fungi has been suggested based on calculations of the energy required for the high flux rates of phosphorus in hyphae (Tinker, 1975; Cox et al., 1980; Harley & Smith, 1983) and inhibition, etc. (Cooper & Tinker, 1981). However, the importance of the bidirectional cytoplasmic streaming in hyphae of arbuscular mycorrhizal fungi in translocation has yet to be clarified (Smith & Read, 1997). Spherical vacuoles labeled with DFFDA observed in the present study did not move longitudinally together with visible components of cytoplasmic streaming.
The behaviors of lipid bodies (Bago et al., 2002) and nuclei (Bago et al., 1998; Bago et al., 1999) have been reported in G. rosea and Glomus intraradices labeled with Nile red and 4,6-diamidino-2-phenylindole (DAPI), respectively. The visible components of cytoplasmic streaming in germ tubes of G. margarita observed with light microscopy were lipid bodies, nuclei and small vesicles stained positively with Neutral Red (Fig. S1). Each of these components seemed to move linearly along undefined tracks at independent rates and beneath regions of tubular vacuoles. The identity of the tracks in G. margarita is unknown, but involvement of the cytoskeleton is probable because of the roles of the cytoskeleton in the positioning and movement of cellular components. Cryofixation and freeze substitution of Pisolithus hyphae showed a close spatial association between tubular vacuoles and microtubules (Shepherd et al., 1993; Hyde et al., 1999). Longitudinal arrays of microtubules have been observed in arbuscular mycorrhizal fungi (Åström et al., 1994; Timonen et al., 2001). It would therefore be worth investigating the role of cytoskeletons in the characteristic bidirectional cytoplasmic streaming and in the tubular vacuolar system in G. margarita.
The present study demonstrated for the first time the vacuole system in living hyphae, mostly in germ tubes, of an arbuscular mycorrhizal fungus. One of the unique features revealed in this study is the extensive tubular vacuolar system that often occurred in bundles and extended for the length of the germ tubes. During the establishment of symbiosis with an onion seedling, the tubular vacuoles were also observed in both extraradical hyphae and intercellular hyphae, although the number of observations was limited by difficulties in staining and visualizing vacuolar systems inside hyphae that were deep inside the root tissues. The relationship between tubular-form vacuoles and cytoplasmic streaming implies the importance of the form in transport inside the hyphae. Further investigations are necessary to observe vacuolar systems in other arbuscular mycorrhizal fungi. In such studies, when the involvement of vacuoles in the translocation of phosphate is discussed, it will be important to verify the presence or absence of phosphorus compounds in these structures. Many questions have arisen about on the tubular vacuole in arbuscular mycorrhizal fungi: whether it is motile, whether the contents translocate inside it, whether a concentration gradient is present in the contents and whether cytoplasmic streaming is involved in the translocation of phosphate. These answers may lead us to picture the mechanisms of long-distance translocation of phosphate in arbuscular mycorrhizal fungi.
This work was supported in part by Promotion of Basic Research Activities for Innovative Biosciences (PROBRAIN), Bio-oriented Technology Research Advancement Institution. We thank Dr W. G. Allaway for discussion, and Dr R. L. Peterson for reviewing the manuscript.
Fig. 1 Germ tubes of Gigaspora margarita labelled with Oregon Green® 488 carboxylic acid diacetate (DFFDA). Time series of superimposed differential interference contrast (DIC) and DFFDA of a single optical slice showing vesicle movements between tubular vacuoles. Images are taken at 0.8 sec intervals for 5.6 sec. Aggregates of lipid bodies having high contrast move from right to left. Nuclei move from left to right beside a stationary nucleus. The hypha is not exactly parallel to the optical section. The surface areas are shown on the right of the picture, and deeper areas on the left.