Extracellular complexation of Cd in the Hartig net and cytosolic Zn sequestration in the fungal mantle of Picea abies – Hebeloma crustuliniforme ectomycorrhizas

Authors


Correspondence: B.Frey. E-mail: beat.frey@wsl.ch

ABSTRACT

Compartmentation of heavy metals on or within mycorrhizal fungi may serve as a protective function for the roots of forest trees growing in soils containing elevated concentrations of metals such as Cd and Zn. In this paper we present the first quantitative measurements by X-ray microanalysis of heavy metals in high-pressure frozen and cryosectioned ectomycorrhizal fungal hyphae. We used this technique to analyse the main sites of Cd and Zn in fungal cells of mantle and Hartig net hyphae and in cortical root cells of symbiotic Picea abies – Hebeloma crustuliniforme associations to gain new insights into the mechanisms of detoxification of these two metals in Norway spruce seedlings. The mycorrhizal seedlings were exposed in growth pouches to either 1 mM Cd or 2 mM Zn for 5 weeks. The microanalytical data revealed that two distinct Cd- and Zn-binding mechanisms are involved in cellular compartmentation of Cd and Zn in the mycobiont. Whereas extracellular complexation of Cd occurred predominantly in the Hartig net hyphae, both extracellular complexation and cytosolic sequestration of Zn occurred in the fungal tissue. The vacuoles were presumed not to be a significant pool for Cd and Zn storage. Cadmium was almost exclusively localized in the cell walls of the Hartig net (up to 161 mmol kg1 DW) compared with significantly lower concentrations in the cell walls of mantle hyphae (22 mmol kg1 DW) and in the cell walls of cortical cells (15 mmol kg1 DW). This suggests that the apoplast of the Hartig net is a primary accumulation site for Cd. Zinc accumulated mainly in the cell walls of the mantle hyphae (111 mmol kg1 DW), the Hartig net hyphae (130 mmol kg1 DW) and the cortical cells (152 mmol kg1 DW). In addition, Zn occurred in high concentrations in the cytoplasm of the fungal mantle hyphae (up to 164 mmol kg1 DW) suggesting that both the cell walls and the cytoplasm of fungal tissue are the main accumulation sites for Zn in P. abies resulting in decreased Zn transfer from the fungus to the root.

INTRODUCTION

Many ectomycorrhizal fungi are associated with coniferous forests growing on acidic, nutrient-poor soils. Acidic conditions in these soils increase the solubility of toxic ions, particularly Al and heavy metals. In addition, increased deposition together with intensified desorption of heavy metals may lead to their accumulation in the soils of forest ecosystems (Zöttl 1985). Toxic concentrations of trace metals in the soil may negatively affect tree root growth, leading to inhibition of root elongation and root hair formation, to a reduction in the number of root tips and the suppression of mycorrhizal colonization (Godbold 1991; Kahle 1993; Jentschke, Winter & Godbold 1999). Metal exposure studies on tree seedlings have shown that enhanced plant fitness may occur when trees are in association with mycorrhizal fungi (Leyval, Turnau & Haselwandter 1997; Godbold et al. 1998). However, it is still under debate to what extent ectomycorrhizas can ameliorate toxic effects of heavy metals. Inoculation with mycorrhizal fungi does not reduce metal uptake or improve tolerance to metals in any case. Amelioration may be dependent upon the species and strain of the ectomycorrhizal fungus and the metal under consideration (Godbold et al. 1998).

The study of the elemental composition of ectomycorrhizas after heavy metal treatment has been of continuing interest, because accumulation of trace metals on or within mycorrhizal fungi may serve as a protective function for roots of forest trees. Detoxification mechanisms in the mycobiont are probably achieved via several physiological processes rather than through a single mechanism (Hartley, Cairney & Meharg 1997). The mechanisms for Cd and Zn in mycorrhizal fungi are poorly understood and the results of microanalytical studies on the subcellular compartmentation of Cd and Zn are conflicting. The reliability of electron probe microanalytical studies was considered questionable because of the migration of ions during specimen preparation, particularly where these procedures prior to microanalysis have involved aqueous buffer systems (Orlovich & Ashford 1993). Room-temperature methods are therefore considered to be of only restricted value in microanalytical studies (Zierold & Steinbrecht 1987). In the light of the potential role of ectomycorrhizal fungi in the afforestation of contaminated sites, there is clearly a need to elucidate cellular compartmentation of heavy metals in ectomycorrhizal roots with accurate preparation techniques for metal distribution at the electron microscope level.

It is now commonly accepted that the retention of ions in the cells using low-temperature microanalysis is near to reality (Van Steveninck & Van Steveninck 1991; Robards 1991). Energy dispersive X-ray microanalysis (EDXMA) in the cryo-scanning electron microscope offers a reliable method for detecting and localizing elements in complex biological systems such as ectomycorrhizas (Scheidegger & Brunner 1999). Brunner & Frey (2000) recently showed the potential usefulness of EDXMA combined with cryo-scanning electron microscopy as a tool for the detection and localization of metals in freeze-fractures of naturally occurring fine roots or ectomycorrhizas. However, EDXMA of freeze-fractured ectomycorrhizal roots gives only a qualitative estimation of the elemental distribution with limited spatial resolution. The use of cryosections instead of bulk frozen material gives better lateral resolution for analysing structures smaller than 1 μm in diameter (e.g. cell walls) by EDXMA, and offers the possibility of obtaining quantitative estimations of elements at the subcellular level (Zierold 1997). Obtaining ultra-thin cryosections (100 nm thick) of botanical tissues for electron microscopy and microanalysis has always been problematic. The structural heterogeneity and the formation of ice crystals commonly observed in frozen samples prepared by conventional freezing may hinder successful cryosectioning. High-pressure freezing is suggested as an alternative to the more conventional freezing technique because this technique allows botanical samples with a thickness up to 400 μm to be frozen without the formation of ice crystals (Michel, Hillmann & Müller 1991; Studer, Hennecke & Müller 1992). This technique of preparing ultra-thin cryosections for EDXMA has been used for animal tissue (Zierold 1997) and for root tissues (Frey et al. 1997). To our knowledge, EDXMA of ultra-thin cryosections has never been applied to fungal hyphae.

The present paper focuses on EDXMA of high-pressure frozen and cryo-sectioned ectomycorrhizas. Using these techniques, we analysed the main sites of Cd and Zn localization in symbiotic associations of Picea abies – Hebeloma crustuliniforme separately in the fungal tissue (mantle hyphae and Hartig net hyphae) and in the cortical root cells in order to gain new insights into the detoxification mechanisms of these trace metals in Norway spruce.

MATERIALS AND METHODS

Growth of mycorrhizal seedlings

The experiment was conducted in growth pouch systems as described previously (Brunner, Frey & Riesen 1996; Brunner & Frey 2000). Growth pouches (Mega International, Minneapolis, USA), 13 × 16 cm, were divided longitudinally into two chambers with a welding seam and coated inside with the two filter papers supplied with the pouches. The top of the front side of the filter paper was coated with an additional activated charcoal filter paper (4 × 3 cm) to support root growth after germination of the seeds. After autoclaving, the pouches were moistened with 5 mL of modified Melin–Norkrans (MMN) solution per chamber. Three surface-sterilized Norway spruce seeds [P. abies (L.) Karst.] were inserted into each chamber and thinned out to one seedling after germination. After 4 weeks, three to four inoculum discs with fungal mycelia of H. crustuliniforme (Bull. St. Amans) Quél. (WSL no. 6·2) were placed within 3 mm of the lateral roots and another 5 mL of MMN were added. A strip of foam (1 × 1 × 3 cm) was placed beside each root to provide air space. Synthesis experiments were carried out in a growth chamber with a 16 h day photoperiod (PAR: 100 μmol m−2 s−1) at 20 °C and 70% relative humidity. The seedlings were irrigated with sterile distilled water as required.

Experimental conditions

When the first ectomycorrhizas had developed (usually after a further 4 weeks' growth) all nutrient solutions were removed, the filter papers were rinsed once with sterile distilled water, and 5 mL of fresh nutrient solution containing one or the other metal (Zn or Cd) was added. In this solution, KH2PO4 and (NH4)2HPO4 were replaced with 0·25 mg L−1 KCl and 0·2 mg L−1 NH4Cl to preclude metal complexation with phosphates while maintaining potassium and ammonium concentrations. The metal concentrations used were either 2 mM Zn (supplied as ZnCl2) or 1 mM Cd (supplied as CdCl2). In a preliminary experiment it was found that these concentrations were tolerated by the ectomycorrhizal tree seedlings without inhibition of mycorrhiza formation or reduction of plant growth (Brunner & Frey 2000). Control treatments contained no additional metal ions in the P-deficient nutrient solution. The pH was adjusted in all treatments to 4·0 with HCl. Four plants were used per treament.

Harvesting and analyses

Five weeks after application of the metals, samples of ectomycorrhizal lateral roots (approximately 200 μm in length and 100 μm in diameter) were excised with a razor blade. To avoid pressure-induced damage, samples were infiltrated with 1-hexadecene (as a non-penetrating agent that improves the contact between the metal of the specimen chamber and the sample) as described by Studer, Hennecke & Müller (1992). The ectomycorrhizas were then kept for several seconds under a mild vacuum to replace intercellular gases and then mounted in aluminium support platelets which were high-pressure frozen in a HPF 010 (BAL-TEC, Balzers, Liechtenstein). Approximately 100 nm thick freeze-dried cryosections were prepared and analysed in the scanning transmission electron microscope equipped with an energy-dispersive X-ray microanalysis system as described previously (Frey et al. 1997). The concentrations of P, S, Cl, K, Ca, Zn and Cd were measured in a spot mode or in a scanning mode over the area of interest: cell wall, cytoplasm and vacuole of fungal tissue (mantle and Hartig net hyphae) and plant tissue (cortical root cells). Elemental concentrations in terms of mmol kg−1 DW were calculated from the measured X-ray spectra according to the peak-to-continuum method. Elemental concentrations in terms of mmol kg−1 water were determined by use of dark-field measurements in the scanning transmission electron microscope (Zierold 1988). The detection limit for zinc and cadmium was approximately 5–10 mmol kg−1 dry mass. Elemental maps were obtained by scanning the electron beam in 10 frames of 128 × 128 pixels with a dwell time of 10 ms and collecting X-rays in element specific energy windows. Differences between concentrations of cellular compartments within one element were estimated by ANOVA. A resulting error probability of P < 0·05 was considered to represent a statistically significant difference.

RESULTS

Structural preservation of ultra-thin cryosections

Surveys of freeze-dried and unstained ultra-thin cryosections of ectomycorrhizas as used for EDXM-analytical purposes are shown in Figs 1 and 2. Fine granularity of cellular structures seen in the cryosections of fungal or root tissue indicates that ice crystal formation during cryofixation cannot be completely excluded (Figs 1 & 2). However, the ultra-structure of the cells was fairly well preserved and the contrast of the cryosections in the dark-field STEM was sufficient to allow discrimination between subcellular compartments. It was possible to identify unambiguously the cell wall, cytoplasm and vacuole in the cryosections of fungal mantle hyphae (Fig. 1a). The interior of the vacuoles was filled with more or less regularly dispersed material and many glycogen granules in the cytoplasm were identified without difficulty in the uncontrasted cryosections of the fungal mantle hyphae (Fig. 1b). Figure 2(a) shows a cryosection with a well-developed Hartig net between cortical cells. In the Hartig net hyphae an electron-opaque cell wall and a uniform matrix of the vacuoles were observed (Fig. 2b). The cortical cells had a large vacuole and a small rim of cytoplasm. However, in many cases the cytoplasm was not recognizable in the cortical cells (Fig. 2a). The good orientation of subcellular compartments within the cryosections allowed EDXMA measurements of root and fungal cell compartments separately and the cryosections were thin enough (approximately 100 nm) for quantitative measurements of the elements.

Figure 1.

STEM image of a high-pressure frozen, freeze-dried cryosection across the fungal mantle hyphae of a root of Picea abiesHebeloma crustuliniforme derived from seedlings grown in growth pouches containing 2 mM Zn in the nutrient solution. (a) Subcellular compartments such as the cell wall, the cytoplasm and the vacuole can be distinguished in the individual hyphae. Bar = 2 μm. (b) The interior of the vacuole (V) was filled with more or less regularly dispersed material. The cytoplasm of the fungal hyphae contains abundant granular rosettes (G), possibly glycogen. Bar = 1 μm.

Figure 2.

(a) STEM image of a high-pressure frozen, freeze-dried cryosection of a root of Picea abiesHebeloma crustuliniforme derived from seedlings grown in growth pouches containing 1 mM Cd in the nutrient solution. Section thickness is estimated to be somewhat higher than 80 nm. The cryosection shows a well-developed Hartig net (H) between cortical cells (C). Bar = 4 μm. (b) STEM image of the same high-pressure frozen and freeze-dried cryosection as in Fig. 2(a) but with a higher magnification. The cryosection shows Hartig net hyphae with some subcellular compartments such as the cell wall, cytoplasm and vacuole. Bar = 2 μm.

Element localization

The concentrations of macro-nutrients such as P, S, Cl, K and Ca were determined in the cytoplasm, the vacuoles and the cell wall of fungal and root tissue (Tables 1 & 2). Concentrations of Cd and Zn were below detection limits in all samples from control plants. The application of the metal (with or without) did not lead to alterations in the concentrations of macro-nutrients within the corresponding subcellular compartments in the ectomycorrhizas (data not shown) and no patterns of change in K distribution were evident after the metal treatment. This is important because efflux of K might indicate membrane damage. Comparison of the macro-nutrients within the corresponding subcellular compartment showed that the concentrations of all elements measured were of the same magnitude under both metal treatments (Tables 1 & 2). In general, Cl was distributed predominantly in the vacuoles of fungal and root cells. The vacuoles and cytoplasm contained relatively high concentrations of K and P and in the vacuoles both elements were evenly distributed throughout fungal vacuoles and no electron-dense Cd- or Zn-bearing deposits were found. High concentrations of P were found in the vacuoles of fungal tissue compared with the vacuoles of the cortical cells under both metal treatment (Tables 1 & 2). Calcium was mainly located in the cell walls and with some lower concentrations in the vacuoles of both fungal and root tissue. The vacuoles also contained S and variable amounts of Cl. Sulphur was distributed homogeneously throughout all compartments. Under the Cd treatment, the S concentrations were higher in the cytoplasm of Hartig net hyphae than in the vacuoles of the Hartig net hyphae. Similarly, S accumulated in higher concentrations in the cytoplasm of cortical cells than in the vacuoles (Table 2).

Table 1.  Mycorrhizal seedlings of Picea abies – Hebeloma crustulinforme were grown for 5 weeks with 2 mM Zn given in the form of ZnCl2 in the nutrient solution
 Cell wallCytoplasmVacuole
 MHHHCCMHHHCCMHHHCC
  1. Zn

  2. 111 ± 39a

  3. 130 ± 67a

  4. 152 ± 59a

  5. 164 ± 51a

  6. 42 ± 19b

  7. 57 ± 12b

  8. 9 ± 6c

  9. 5 ± 5c

  10. 7 ± 5c

  11. The element concentrations of mantle (MH) and Hartig net hyphae (HH) and root cortical cells (CC) in ultra-thin cryosections are given in mmol kg−1 DW and are the mean of 10 measurements ± SD. Significant differences between values of cellular compartments within one element are indicated by different letters (P = 0·05). ND, not detected.

P29 ± 12c21 ± 8c24 ± 21c121 ± 58ab157 ± 63ab177 ± 81a224 ± 31a180 ± 51a81 ± 38b
S60 ± 29a64 ± 24a46 ± 30a74 ± 29a96 ± 37a90 ± 43a76 ± 41a67 ± 32a72 ± 26a
Cl36 ± 11b5 ± 5c21 ± 6b39 ± 28b32 ± 7b39 ± 21b151 ± 43a198 ± 57a206 ± 63a
K94 ± 49b51 ± 18b47 ± 23b370 ± 178a295 ± 98a251 ± 104a374 ± 231a361 ± 179a187 ± 117a
Ca53 ± 32a72 ± 35a62 ± 27aNDNDND68 ± 20a23 ± 9ab12 ± 6b
Table 2.  Mycorrhizal seedlings of Picea abies – Hebeloma crustulinforme were grown for 5 weeks with 1 mM Cd given in the form of CdCl2 in the nutrient solution
 Cell wallCytoplasmVacuole
 MHHHCCMHHHCCMHHHCC
  1. Cd

  2. 22 ± 10b

  3. 161 ± 45a

  4. 15 ± 14b

  5. ND

  6. ND

  7. ND

  8. 10 ± 5b

  9. 12 ± 3b

  10. 16 ± 4b

  11. The element concentrations of mantle (MH) and Hartig net hyphae (HH) and root cortical cells (CC) in ultra-thin cryosections are given in mmol kg−1 DW and are the mean of 10 measurements ± SD. Significant differences between values of cellular compartments within one element are indicated by different letters (P = 0·05). ND, not detected.

P22 ± 9d15 ± 12d13 ± 6d191 ± 87ab289 ± 101a220 ± 83ab187 ± 54ab104 ± 37b49 ± 12c
S71 ± 18b96 ± 31ab75 ± 20b129 ± 26a135 ± 32a95 ± 30ab60 ± 24bc65 ± 18bc46 ± 13c
Cl29 ± 12b9 ± 3c23 ± 5b35 ± 14b29 ± 10b38 ± 17b172 ± 52a134 ± 85a221 ± 78a
K50 ± 27d54 ± 14d41 ± 32d368 ± 142ab537 ± 206a341 ± 117ab216 ± 96bc190 ± 104bc141 ± 62c
Ca47 ± 15a63 ± 21a49 ± 16aNDNDND39 ± 14a14 ± 3b8 ± 3b

Heavy metal localization

Cadmium and zinc added to the nutrient solution were found in cortical cells as well as in fungal cells of mantle and Hartig net hyphae (Tables 1 & 2). Analysis of Zn revealed relatively high concentrations in the cell walls of mantle hyphae (111 kg−1 DW), of Hartig net hyphae (130 mmol kg−1 DW), of cortical cells (152 mmol kg−1 DW) and in the cytoplasm of the fungal mantle hyphae (up to 164 mmol kg−1 DW) (Table 1). The concentrations of Zn in the cytoplasm of the mantle hyphae were higher than those in the cytoplasm of the Hartig net hyphae (42 mmol kg−1 DW) and the cortical cells (57 mmol kg−1 DW). In contrast, Zn was only just at the detection limit (5 mmol kg−1 DW) in the vacuoles of both fungal and cortical cells (Table 1).

In contrast, Cd was almost exclusively localized in the cell wall, both in the fungal and the root tissue (Table 2). The highest accumulation of Cd was found in the cell walls of the Hartig net hyphae (up to 161 mmol kg−1 DW). The concentrations of Cd in the cell walls of the Hartig net hyphae were about seven times higher than those in the cell walls of the mantle hyphae (22 mmol kg−1 DW) and about 10 times higher in the cell walls of cortical cells (15 mmol kg−1 DW). In contrast to the cell walls, Cd was below the detection limit (4 mmol kg−1 DW) in the cytoplasm and Cd was somewhat higher than the detection limit in the vacuoles of both fungal and cortical cells, ranging from 10 to 16 mmol kg−1 DW (Table 2).

Figure 3 shows a set of elemental maps from a freeze-dried cryosection of Hartig net hyphae. The X-ray map presented confirms the data from static probe EDXMA showing the preferred accumulation of Cd in the cell wall of the Hartig net (Fig. 3). Potassium was homogeneously distributed within the cells of the Hartig net hyphae, whereas P was predominantly found in the nuclei. Identification of the nucleus in the cryosections was facilitated by single spot measurements showing a very high peak for P (2100 mmol kg−1 DW) with distinctly lower concentrations of P in the cytoplasm (200 mmol kg−1 DW). Furthermore, the map shows that Cd was almost absent from the vacuoles and was not accumulated in association with P.

Figure 3.

Elemental maps of the same high-pressure frozen and freeze-dried cryosection as in Fig. 2 but at a higher magnification. The STEM micrograph (bright/dark field image) and the corresponding element distribution maps for Cd, K and P were recorded. The colour scale indicates X-ray intensity from low (black, blue) to high (red, white).

DISCUSSION

The present study on high-pressure frozen and cryosectioned roots of P. abies in symbiotic association with H. crustuliniforme has demonstrated substantial retention of both Cd and Zn in the fungal tissue. The retention of trace metals in the mycobiont and the resulting restriction of metal movement to the plant root cells is assumed to be the main mechanism involved in tree tolerance against trace metal stress (Godbold et al. 1998). In the case of Cd, extracellular complexation involving binding to cell walls occurred predominantly in the Hartig net hyphae, whereas both extracellular complexation and cytosolic sequestration of Zn were found to occur in the mantle and Hartig net hyphae. Furthermore, the Hartig net hyphae rather than the mantle showed a filter function against Cd. The high capacity of the Hartig net hyphae to retain Cd agrees with the findings of Brunner & Frey (2000). However, using EDXMA in freeze-fractures they were unable to distinguish subcellular compartments in the Hartig net hyphae. The use of cryosections instead of bulk frozen material gives a better lateral resolution for analysing cellular compartments smaller than 1 μm in diameter (e.g. cell walls), by EDXMA. Consequently, in the present study we could show that – with respect to the detection limit – cadmium in the Hartig net was localized exclusively in the cell wall systems. The capacity of the cell walls of the Hartig net hyphae to bind Cd is assumed to be limited and to fail when the plant is exposed to high (10 mM Cd) external metal concentrations (Brunner & Frey 2000). The finding that the Hartig net hyphae accumulate large amounts of Cd may explain why mycorrhizal fungi may alleviate Cd toxicity in Norway spruce seedlings as shown by Jentschke, Winter & Godbold (1999). Galli, Meier & Brunold (1993) found that the Cd concentration in needles of P. abies colonized with Laccaria laccata was significantly lower than that in non-mycorrhizal plants. The Cd complexed in L. laccata was easily exchangable by washing with a solution containing Ni instead of Cd. The authors therefore suggested that most of the Cd in L. laccata was associated with the cell walls. Another site for the binding of Cd was proposed to be in the extramatrical mycelium of mycorrhizal Pinus sylvestris (Colpaert & Van Assche 1993). These workers showed that when the roots were surrounded by a dense extramycelium the potential for retaining Cd increased.

In contrast to Cd, which was absent in the cytoplasm of fungal cells, a high concentration of Zn was found to be sequestered in the cytoplasm of mantle hyphae besides extracellular complexation. Extra-cellular complexation of Zn compared with intracellular sequestration indicates the involvement of different Zn-binding and detoxification processes in the mantle hyphae. The high levels of Zn binding to the cell walls agree well with the findings of Denny & Wilkins (1987) who studied Zn toxicity in ectomycorrhizal Betula species in freeze-substituted samples. Their work indicated that the ameliorating effect of the mycobiont is realized by the immobilization of large amounts of Zn in the fungal cell walls or extrahyphal polysaccharide slime but not in the fungal cytoplasm. The zinc concentrations they found in the cell walls of mantle hyphae were within the range obtained in our study. Cytosolic Zn compartmentation has also been implied by Bücking & Heyser (1999) in axenic cultures of a Zn-tolerant strain of Suillus bovinus by which most of the Zn was sequestered in the fungal cytoplasm.

The accumulation of Cd in the cell walls of the Hartig net raises the question of whether the different chemical constituents in the cell walls between Hartig net and cortical cells are responsible for the high Cd concentration in the mycobiont. Cadmium and Zn generally bind to the cell walls of roots and fungi. In plant roots, metal cations can be electrostatically bound to pectins (which are characterized by a considerable cation exchange capacity and the ability to form cross-linked structures) or to charged groups of wall proteins (Marschner 1995). Fungal cell walls, however, do not contain carboxylate groups of pectins. In fungal cell walls, metal cations are bound predominantly to chitin, cellulose and melanins (Gadd 1993) which lack acidic groups. Marscher, Jentschke & Godbold (1998) determined a cation exchange capacity in the mycelium of Laccaria bicolor S238, which was increased about 30-fold compared with the cation exchange capacity of the roots of Norway spruce while that of Paxillus involutus 533 was increased about 12-fold. High-cation exchange capacity combined with a high surface area as a result of the labyrinthine mode of growth makes the Hartig net hyphae a powerful tool for binding divalent cations such as trace metals. The Hartig net is the site for nutrient exchange in ectomycorrhizal systems and is a complex system of labyrinthine branching between fungal wall and host wall (Scheidegger & Brunner 1993).

Heavy metals induce in plant and fungal cells a wide range of low molecular weight polypeptides and proteins with high cysteine contents, the so-called metallothioneins (Gadd 1993; Zenk 1996). Metallothioneins have been found to bind to heavy metals and the resulting complexes can be detoxified by transportation into vacuoles. The occurrence of metallothioneins or metallothionein-like compounds in ectomycorrhizal fungi is not clear and has been little investigated. Cu-binding proteins are known to exist in tolerant strains of L. laccata and P. involutus in response to high Cu concentrations (Howe, Evans & Ketteridge 1997). However, the high Zn concentration in the cytoplasm raises the question of how Zn is integrated in the cytoplasm. In yeast cells cytoplasmic storage of Zn is one proposed detoxification mechanism where there is some evidence for the production of metallothionein-like zinc-binding proteins, for example in the cytoplasm of Candida utilis. However, information on other fungi (including ectomycorrhizal fungi) is lacking (Ross 1994). In the case of Cd, the efficient detoxification mechanism in the cell walls suggests that the plasma membrane acts as the primary barrier. Alternatively, some Cd may have been transported to the vacuoles. Transport of Cd across the cytoplasm to the vacuole would require the production of compounds capable of making this potentially toxic metal unavailable for interaction with metabolically active cellular compartments. This compartmentation may be achieved by complexation of the metal ions in a chemically inactive form immediately after their transport into the cytoplasmic compartment. The synthesis of Cd-binding peptides, named cadystin, has been induced in fission yeast Schizosaccaromyces pombe (Hayashi & Mutoh 1994). However, from this study it is not clear whether metallothionein-like proteins are involved in the detoxification of Cd in ectomycorrhizal fungi and this awaits further investigation. The cytosolic Zn sequestration found here compared with the absence of Cd from the fungal cytoplasm requires a further comment. Zinc plays an essential metabolic role in a living organism as a component of a variety of enzymes and is required for the maintenance of integrity of biomembranes where it might be bound to phospholipid and sulfhydryl groups of membrane constituents. Cadmium, on the other hand, is not known to have any function in the cells and is assumed to be very toxic at higher concentrations (Marschner 1995). Whether the higher toxicity of Cd in the cells is partly responsible for the different detoxification mechanisms remains unclear.

Besides the fungal cell walls and the fungal cytoplasm other possible mechanisms for detoxification of metal cations in ectomycorrhizas may occur. The capacity of polyphosphate granules to bind toxic cations has been suggested in ectomycorrhizas (Turnau, Kottke & Oberwinkler 1993). Using electron energy loss spectroscopy in chemically fixed samples, they reported an intracellular accumulation of Cd within P containing granules of the fungal vacuoles in P. sylvestrisP. involutus roots. In contrast, Turnau, Kottke & Dexheimer (1996) detected Cd in the cytoplasm of mantle hyphae in chemically fixed Rhizopogon mycorrhizas (Turnau, Kottke & Dexheimer 1996). With EDXMA on Spurr's epoxy resin-embedded samples of an axenic culture of a Zn-tolerant strain of Suillus bovinus, Bücking & Heyser (1999) found that Zn in the mycelium was sometimes associated with granules, which they thought were polyphosphate, in the fungal vacuoles. In the present study, the vacuoles were not assumed to be a significant storage pool for Zn and Zn did not appear to be associated with polyphosphate granules in fungal vacuoles. Here, relatively high P concentrations were found in the fungal vacuoles compared with the concentrations in the vacuoles of cortical cells in both metal treatments. The high P concentrations might be the result of the storage of P as polyphosphate within fungal vacuoles as demonstrated by Ashford, Ryde & Barrow (1994). Vacuoles appear to be filled with more or less uniformly dispersed material, which is in good accordance with the findings of Ashford et al. (1999) for fungal vacuoles with Pisolithus tinctorius in symbiosis with Eucalyptus pilularis suggesting that poly P occurs solely in a soluble form. The formation of P-containing granules within fungal vacuoles, however, is still under debate (Ashford et al. 1999). We found no P-containing granules in the fungal vacuoles of H. crustuliniforme. The presence or absence of P-containing granules in fungal vacuoles may depend on the fungal species, the age of the culture, the amount of P, the accompanying cations present in the growth medium and the preparation technique used for observation in the electron microscope.

In the present study, we have demonstrated a suitable approach for localizing Cd and Zn in ectomycorrhizal roots using a low-temperature procedure which is known not to (or only minimally to) involve the redistribution of mobile ions. This is of major significance as most studies on metal localization have been undertaken with conventional tissue preparation methods (Turnau, Kottke & Oberwinkler 1993). Our present work is, as far as we know, the first study on the quantitative measurement of EDXMA of heavy metals in ultra-thin cryosections of ectomycorrhizal hyphae. Cryoultramicrotomy of high-pressure frozen ectomycorrhizal hyphae and roots now offers new opportunities for studying trace metals within the mycobiont. This study has given a new insight into the Zn and Cd toxifying mechanisms of Norway spruce. From our results we conclude that ectomycorrhizal roots of Norway spruce have different mechanisms for retaining Cd and Zn in the fungal tissue. An efficient detoxification mechanism of the highly toxic Cd within the cell walls of the Hartig net was detected. Both the cell walls and the cytoplasm were involved in the Zn tolerance mechanisms for protecting metabolically active cellular compartments from toxic Zn concentrations. In the light of the increasing deposition of heavy metals in forest soils and biogeochemical weathering of heavy metals due to increasing soil acidification, the full elucidation of the mechanisms for the detoxification of heavy metals in the fungal cells awaits further investigation.

ACKNOWLEDGMENTS

This work was partly supported by a grant from the Swiss Federal Institute of Technology (ETH) in Zurich (CRICEPF). We thank S. Dongard, Dortmund for excellent technical assistance in preparing the ultra-thin cryosections for X-ray microanalysis and P. Walther, ETH Zurich for help in using the high-pressure freezer. We are grateful to P. Christie of The Queen's University of Belfast and M. J. Sieber for helpful comments on the manuscript.

Ancillary

Advertisement