A zinc-adapted fungus protects pines from zinc stress


  • Kristin Adriaensen,

    1. Centre of Environmental Sciences, Environmental Biology Group, Limburgs Universitair Centrum, Universitaire Campus, B-3590 Diepenbeek, Belgium;
    Search for more papers by this author
  • Daniël Van Der Lelie,

    1. Centre of Environmental Sciences, Environmental Biology Group, Limburgs Universitair Centrum, Universitaire Campus, B-3590 Diepenbeek, Belgium;
    2. Brookhaven National Laboratory, Biology Department, Building 463, Upton, New York 11973-5000, USA;
    Search for more papers by this author
  • André Van Laere,

    1. Laboratory of Developmental Biology, Institute of Botany and Microbiology, Katholieke Universiteit Leuven, Kasteelpark Arenberg 31, B-3001 Heverlee, Belgium
    Search for more papers by this author
  • Jaco Vangronsveld,

    1. Centre of Environmental Sciences, Environmental Biology Group, Limburgs Universitair Centrum, Universitaire Campus, B-3590 Diepenbeek, Belgium;
    Search for more papers by this author
  • Jan V. Colpaert

    Corresponding author
    1. Centre of Environmental Sciences, Environmental Biology Group, Limburgs Universitair Centrum, Universitaire Campus, B-3590 Diepenbeek, Belgium;
      Jan V. Colpaert Tel: +32 11 268304 Fax: +32 11 268301 Email: jan.colpaert@luc.ac.be
    Search for more papers by this author

Author for correspondence: Jan V. Colpaert Tel: +32 11 268304 Fax: +32 11 268301 Email: jan.colpaert@luc.ac.be


  • • Here we investigated zinc tolerance of ectomycorrhizal Scots pine (Pinus sylvestris) seedlings. An ectomycorrhizal genotype of Suillus bovinus, collected from a Zn-contaminated site and showing adaptive Zn tolerance in vitro, was compared with a nonadapted isolate from a nonpolluted area.
  • • A dose–response experiment was performed. Dynamics of plant and fungal development, and phosphate and ammonium uptake capacity, were assessed under increasing Zn stress. Effects of Zn on transpiration, nutrient content and Zn accumulation were analysed.
  • • Significant Zn–inoculation interaction effects were observed for several responses measured, including uptake rates of phosphate and ammonium; phosphorus, iron and Zn content in shoots; transpiration; biomass of external mycelia; and fungal biomass in roots.
  • • The Zn-tolerant S. bovinus genotype was particularly efficient in protecting pines from Zn stress. The growth of a Zn-sensitive genotype from a normal wild-type population was inhibited at high Zn concentrations, and this isolate could not sustain the pines’ acquisition of nutrients. This study shows that well adapted microbial root symbionts are a major component of the survival strategy of trees that colonize contaminated soils.


Plant communities colonizing soils polluted by heavy metals are characterized by low diversity, low productivity, slow ecological succession, and an increased incidence of heavy metal-tolerant genotypes (Ernst, 1990; Reeves & Baker, 2000). However, at the population level there is no strong evidence for parallel evolution to metal-contaminated environments in trees and their associated ectomycorrhizal fungi (Meharg & Cairney, 2000). Trees are thought to colonize extreme environments through large phenotypic plasticity (Dickinson et al., 1991; Watmough & Hutchinson, 1998). Rhizosphere microorganisms may assist trees in colonizing such sites (Wilkinson & Dickinson, 1995; Meharg & Cairney, 2000). In low-fertility ecosystems, pines and most other tree species depend strongly on ectomycorrhizal fungi for mineral nutrition. Their nutrient-absorbing roots are ensheathed by a mantle of fungal hyphae forming the ectomycorrhizas. An extensive mycelial network extends from these symbiotic roots into the soil, forming the main interface between soil solution and mycorrhizal plant (Smith & Read, 1997). Consequently, ectomycorrhizal fungi largely control the pathway of nutrients, whether in excess or at trace level, from the soil to the host plant.

Several investigators have compared mycorrhizal and nonmycorrhizal tree seedlings exposed to toxic concentrations of heavy metals. Ectomycorrhizal fungi often reduce the heavy metal sensitivity of seedlings, but this benefit can be attributed to a generally enhanced fitness as a result of the inoculation (Dixon & Buschena, 1988; Jentschke & Godbold, 2000; Meharg & Cairney, 2000; Van Tichelen et al., 2001). These soilborne fungi exhibit considerable interspecific differences in heavy metal sensitivity in vitro, and it has been suggested that only fungal symbionts with high tolerance survive in heavily contaminated soils (Hartley et al., 1997). Depending on the degree of contamination, a high selection pressure may trigger evolutionary adaptation towards greater resistance. Such genetic adaptation to toxic Zn concentrations was found in ectomycorrhizal fungal populations of Suillus luteus and Suillus bovinus, thriving at sites severely contaminated by Zn smelters in north-east Belgium (Colpaert & Van Assche, 1987; Colpaert et al., 2000). However, no study has shown that heavy metal-tolerant ectomycorrhizal genotypes outperform sensitive isolates in terms of physiology or growth of host trees exposed to phytotoxic metal concentrations (Meharg & Cairney, 2000).

Here we report on a dose–response experiment with Scots pine seedlings inoculated with either a Zn-sensitive or a Zn-tolerant genotype of S. bovinus. The mycorrhizal pine seedlings were exposed for 6 wk to a range of Zn concentrations representative for the soluble Zn fraction in soil pore waters analysed along the pollution gradient surrounding the zinc smelters. The applied Zn range was previously shown to discriminate between populations of Zn-adapted and nonadapted grass species (Al-Hiyaly et al., 1993). Excess Zn disrupts several physiological processes, causes oxidative stress (Clijsters et al., 1999), and lowers nutrient assimilation in plants and microorganisms (Marschner, 1995; Nies, 1999; Blaudez et al., 2000). Because improved nutrient acquisition is a key function of ectomycorrhizal symbiosis (Smith & Read, 1997), we assessed the progress of the nutrient uptake capacity of the plants during Zn treatments. These nondestructive analyses are also a sensitive tool to detect early symptoms of heavy-metal toxicity (Van Tichelen et al., 1999).

Materials and Methods

Fungal genotypes and collection site

Six Suillus bovinus (Fr.) isolates were collected in October 2000 in young Scots pine forests in Meeuwen-Gruitrode (Belgium) and Lommel (Belgium). Soil samples (±2 kg) were collected from beneath each sporocarp. Pore water was sucked with Rhizon soil moisture samplers (Eijkelkamp Agrisearch Equipment, Giesbeek, the Netherlands) as described by Knight et al. (1998). The pine forest in Lommel is growing on a strongly contaminated soil adjacent to the ‘zinc desert’, an industrial site polluted with nonferrous metals emitted by the abandoned Zn smelter of Maatheide. The Zn concentration in the pore water varied from 80 to 190 µM Zn, a range that induces chronic and acute Zn toxicity in plants and microorganisms (Al-Hiyaly et al., 1993; Chaudri et al., 2000). The forest in Meeuwen-Gruitrode thrives on an uncontaminated sandy soil, 22 km south-east of the nearest smelter. The average Zn concentration in the pore water was 0.7 µM. The fungi were isolated, grown and screened for Zn tolerance in vitro according to standard procedures (Colpaert et al., 2000). All isolates from Lommel showed a high degree of Zn tolerance, in contrast to the isolates from Meeuwen-Gruitrode which were all sensitive to elevated Zn. Two genotypes with a similar growth rate on basic medium, but with contrasting Zn sensitivity, were selected for the present experiment. The in vitro EC50 value for biomass production was 1.8 mm Zn on Fries medium for the sensitive S. bovinus isolate. The Zn-tolerant isolate from the Lommel population had an in vitro EC50 value of 9.0 mm Zn.

Plant growth and treatments

Surface-sterilized seeds of Pinus sylvestris L. (provenance Groenendaal, Belgium) were sown in a perlite–vermiculite mixture moistened with Ingestad nutrient solution for pine (Ingestad & Kähr, 1985). A sandwich technique was used to inoculate 4-wk-old seedlings with vigorously growing mycelia of either the Zn-tolerant or Zn-sensitive isolate of S. bovinus. After 3 d, plants were transferred to 70 ml containers filled with pure perlite (Colpaert et al., 1999). Perlite has a low nutrient-buffering capacity so that the plants are actually growing in a semihydroponic environment. Addition of nutrients to the plants was strictly controlled, and was highly favourable for strong mycorrhizal development characteristic of the normal, low-fertility habitat of both symbionts. Nutrients (weight proportions: 100 N/9 P/54 K/6 Ca/6 Mg/9 S + micronutrients) were added at a constant relative rate of 3% d−1 (Colpaert et al., 1999). The pH of the nutrient solution was adjusted to 4.5.

Zn treatments were started 3 wk after inoculation. At this point, five plants from each inoculation treatment were harvested to determine the initial nutrient status and biomass of plants and fungi. Inoculated plants were divided at random into five Zn treatments to create a factorial setup with the factors Zn treatment (0.1, 38, 76, 153 or 229 µM Zn2+) and mycorrhizal inoculation. Each combined treatment had five replicates. Extra Zn was added to the nutrient solution as sulphate. The pH and Zn concentrations in the substrate solution were monitored and maintained weekly by flushing the plant containers with excess nutrient solution containing the respective Zn concentrations. The experiment was carried out in a growth chamber with 300 µmol m−2 s−1 PAR, a relative air humidity of 70%, and with a day/night rhythm of 18/6 h and a temperature of 22/15°C.


Net uptake of inorganic phosphate (Pi) and ammonium was analysed at 2, 4 and 6 wk after starting Zn treatments. These measurements did not disrupt the integrity of the root–fungus pathway. Depletion of Pi and NH4+ was determined simultaneously in a test solution circulated through the plant containers (Colpaert et al., 1999). Pi and NH4+ were assessed colorimetrically according to standard protocols. Specific net uptake rates for Pi and NH4+ were directly calculated from the slopes of the depletion curves at a defined external concentration. The transpiration of the plants was recorded over a 24 h period 3 d before harvest. Plants were harvested after 6 wk of Zn additions, 1 d after the last measurement of nutrient uptake.

Shoots were dried (70°C, 120 h), and roots and perlite (containing external mycelia) were freeze-dried. The material was milled in a ball mill to a fine powder for analyses of elements and ergosterol. Most analyses were performed in duplicate, and certified reference material was included for element analyses. Nitrogen concentration was determined with a Finnigan Delta-Plus linked to a continuous-flow Carlo Erba elemental analyser. Phosphate was analysed colorimetrically, and Zn and Fe were measured by Atomic Absorption Spectrophotometry (AAS). Ergosterol was extracted and analysed by high-performance liquid chromatography as described by Nylund & Wallander (1992). Ergosterol data were converted to fungal biomass with conversion factors of 7.1 and 7.5 mg ergosterol g−1 d. wt mycelium for the Zn-tolerant and Zn-sensitive S. bovinus, respectively. These conversion factors were calculated from ergosterol levels determined in the freeze-dried fungal mats of the in vitro Zn screening test. Data on biomass and element concentrations were analysed by two-way ANOVA after checking their normal distribution.


Nutrient uptake

In the absence of elevated Zn, plants inoculated with the Zn-tolerant or Zn-sensitive mycobiont could not be distinguished in relation to nutrient uptake capacity or nutrient content. The analyses of the seedlings’ ability to absorb Pi and NH4+ showed that the nutrient uptake capacity of the seedlings increased over 2-wk intervals as a result of the expanding mycorrhizal root system (Fig. 1). Soon after the Zn additions started, the kinetics of Pi uptake in the Zn-sensitive fungus–plant combination was adversely affected in all Zn-exposed plants. At the highest Zn concentration tested, the uptake of Pi in plants with the Zn-sensitive fungus decreased to almost zero by the end of the experiment, illustrating the ultimate failure of Pi uptake in the mycorrhizal root system (Fig. 1). In marked contrast, the Pi uptake in plants inoculated with the Zn-tolerant S. bovinus was not affected. Not surprisingly, mycorrhizals with the Zn-sensitive mycobiont captured significantly less P than plants colonized with the tolerant S. bovinus under Zn stress (Pinteraction < 0.001, F4,40 = 26.81).

Figure 1.

Phosphate uptake in zinc-treated pine (Pinus sylvestris) seedlings inoculated with the ectomycorrhizal fungus Suillus bovinus. A Zn-tolerant genotype (top) is compared with a Zn-sensitive isolate (below). Depletion of phosphate in nutrient solution was analysed over 5 h. The measurement was performed after 2 wk (closed circles); 4 wk (open circles); and 6 wk (closed squares) of Zn treatment. Each point represents the mean ± SE, n = 5.

Elevated levels of Zn not only interfered with Pi uptake in seedlings inoculated with the sensitive mycobiont, but also reduced NH4+ uptake rates, although this was less pronounced (Fig. 2a). A differential response in N transfer to the shoots was not found in this short-term experiment (data not shown).

Figure 2.

Nitrogen acquisition, iron concentration and water use of pine (Pinus sylvestris) seedlings treated with elevated zinc for 6 wk. Plants were inoculated with a Zn-tolerant (open squares/columns) or Zn-sensitive (dark squares/columns) Suillus bovinus genotype. (a) Specific NH4+ uptake rates at an external NH4+ concentration of 350 µM (Pinteraction < 0.001, F4,40 = 7.63). (b) Iron concentration in needles (Pinteraction = 0.008, F4,40 = 3.98). (c) Transpiration of plants 3 d before harvest (Pinteraction = 0.016, F4,40 = 3.45). Averages of five replicates ± SE.

Element analysis of the pine needles revealed that Fe nutrition was also hampered by Zn stress (Fig. 2b). A decrease in Fe in needles was observed in both inoculation treatments, but it became significantly greater in pine seedlings colonized by the Zn-sensitive fungus.

Zinc distribution in plants

As expected, exposure to elevated Zn concentrations also increased transfer of Zn to the needles of plants, at least in the lower range of the Zn gradient (Table 1). However, with increasing Zn toxicity, transpiration of plants diminished (Fig. 2c) and, almost simultaneously, the transfer of Zn to shoots declined. In roots, Zn concentrations increased in both inoculation treatments once plants were exposed to elevated Zn, and became considerably higher than concentrations measured in the shoots (Table 1).

Table 1.  Zinc concentration in needles and roots of ectomycorrhizal pines (Pinus sylvestris) exposed to elevated Zn for 6 wk
MycobiontZn treatment (µM)
  1. Data expressed in µg Zn g−1 (d. wt), values are means ± SE (n = 5). Pinteraction (needles) = 0.022, F4,40 = 3.19; Pinteraction (roots) = 0.52, F4,40 = 0.82.

Suillus bovinus, Zn-tolerant24 ± 1.5 92 ± 3.3 107 ± 11 189 ± 8 170 ± 22
S. bovinus, Zn-sensitive20 ± 2.5110 ± 15 189 ± 29 161 ± 21 152 ± 26
S. bovinus, Zn-tolerant68 ± 2786 ± 621339 ± 923175 ± 1524287 ± 353
S. bovinus, Zn-sensitive72 ± 6767 ± 641575 ± 1692769 ± 644257 ± 359

Plant and fungal growth

The above-ground plant biomass was not affected by the Zn treatments (F4,40 = 1.20, P = 0.32) in this short-term experiment. Root growth tended to decrease in the Zn-sensitive fungus–plant combination, so that slightly lower relative growth rates were noted at 153 and 229 µM Zn (Table 2). However, after 6 wk of treatment excess Zn significantly reduced the development of the Zn-sensitive fungus in both roots and substrate (Fig. 3). The relative growth rate of the Zn-tolerant S. bovinus was similar in all Zn treatments, whereas the relative growth rate of the sensitive isolate dropped from 3.4% d−1 in the control treatment to 1.8% d−1 in 229 µM Zn (Table 2). Fungal biomass was strongly correlated to the P content of plants (Fig. 4).

Table 2.  Relative growth rates of plants (Pinus sylvestris) and fungi exposed to elevated Zn for 6 wk
MycobiontZn treatment (µM)
  1. Relative growth rates (% d−1) were calculated for total plant biomass (Pinteraction = 0.35, F4,40 = 1.13) and for total fungal biomass (Pinteraction < 0.001, F4,40 = 10.97) determined at the start of the treatment and at harvest. Fungal biomass was calculated from ergosterol data. Values are means ± SE (n = 5).

Suillus bovinus, Zn-tolerant2.9 ± 0.22.9 ± 0.23.0 ± 0.32.7 ± 0.32.7 ± 0.4
S. bovinus, Zn-sensitive2.9 ± 0.22.8 ± 0.32.8 ± 0.42.3 ± 0.32.2 ± 0.4
S. bovinus, Zn-tolerant3.3 ± 0.23.4 ± 0.33.6 ± 0.33.3 ± 0.13.4 ± 0.3
S. bovinus, Zn-sensitive3.4 ± 0.33.4 ± 0.23.0 ± 0.62.1 ± 0.41.8 ± 0.6
Figure 3.

Fungal biomass of the zinc-tolerant (light columns) and Zn-sensitive (dark columns) Suillus bovinus genotype. Biomass was calculated from ergosterol analyses at the start (*) of the Zn treatments and at harvest. Bottom part of columns is fungal biomass in roots (Hartig net and mantle mycelia) (Pinteraction = 0.002, F4,40 = 5.10); top part of columns is biomass of external mycelium (Pinteraction = 0.001, F4,40 = 5.60). Standard errors were calculated on total fungal biomass.

Figure 4.

Relationship between amount of phosphorus in pines (Pinus sylvestris) and fungal biomass of the zinc-tolerant (open squares) and the Zn-sensitive (closed squares) Suillus bovinus isolate after 6 wk of Zn additions. Values at the start of Zn additions are also shown (triangles). Averages of Zn-sensitive fungus–plant combination were fitted by a significant regression (mg P = 0.152 + 0.016× mg fungal biomass).


Sensitivity to potentially toxic metals of a range of ectomycorrhizal fungal species and higher plant hosts has been studied in symbiosis. Whilst a wide range of responses have been reported, in most cases some degree of host amelioration of metal toxicity by mycorrhizal fungi has been demonstrated (Brown & Wilkins, 1985; Jones & Hutchinson, 1986; Colpaert & Van Assche, 1992; Marschner et al., 1996; Van Tichelen et al., 2001). Comparisons have been made between ectomycorrhizal and nonmycorrhizal plants, and between plants inoculated with different ectomycorrhizal fungal species. Results indicate that interspecific variation exists in the ability of ectomycorrhizal fungi to confer a reduction in metal sensitivity to the host plant. The majority of studies on metal sensitivity in ectomycorrhizal seedlings focused on the impact of potentially toxic metals on the plant, and reduced sensitivity conferred to the plant by the ectomycorrhizal fungus. The effects of metals on the fungal partner in symbiosis have been studied in less detail. Screening ectomycorrhizal fungi for tolerance in symbiosis is difficult because of the intimate association between fungus and plant. A significant problem is establishing the relative contributions of plant and fungus in the tolerance of the association (Hartley et al., 1997). In the present study, we have demonstrated a different response to elevated Zn concentrations of two different genotypes of the same fungus, S. bovinus, one collected from a Zn-polluted soil and one from a control soil. From this study we can conclude that a Zn-sensitive fungus has difficulty in developing in a Zn-contaminated substrate. The results demonstrate that a Zn-adapted fungal genotype survives Zn stress more successfully. The better performance of this fungus significantly improves the nutrient status of its host.

Nutrient uptake analysis is a sensitive tool for rapid detection and for monitoring symptoms of heavy metal toxicity (Van Tichelen et al., 1999). The present experiment demonstrates that relatively low Zn concentrations (38 µM Zn) disturb Pi uptake in the Zn-sensitive plant–fungus combination. The reduced Pi uptake cannot be ascribed only to reduced growth of the Zn-sensitive fungus. Currently, we can only speculate about the mechanism responsible for this effect. Zinc ions play a specific role in the regulation of genes encoding high-affinity P transporters in plant roots (Huang et al., 2000). Zinc deficiency upregulates P transporters in barley roots irrespective of the P status of plants. Our own observations suggest that high cellular Zn levels may cause downregulation of the expression or activity of P transporter genes. However, we do not know whether fungal P transporter genes are also regulated in a Zn-dependent way. Zinc ions also play a crucial role in membrane stability, affecting the activity of membrane-bound proton-pumping enzymes and proton channels, and thus disturbing the uptake of nutrients across the plasma membranes (Cakmak, 2000). Whatever the mechanism of reduced Pi uptake, sufficient uptake of Pi in the mycorrhizal pines is essential for the productivity of both symbiotic partners, and polyphosphates are thought to contribute to the cellular homeostasis of Zn in fungal cells (Bücking & Heyser, 1999).

As a result of better nutrient acquisition by the tolerant plant–fungus combination, these plants can maintain a higher nutrient status. It is surprising that this Zn–inoculation interaction with regard to the macro- and micronutrient (P and Fe) status of a host plant was not observed previously, even when mycorrhizal and nonmycorrhizal plants were compared at different Zn levels (Dixon & Buschena, 1988; Meharg & Cairney, 2000).

As expected, exposure to elevated Zn concentrations also increased the transfer of Zn to the needles of plants, at least in the lower range of the Zn gradient. The transfer and accumulation of excess heavy metals in leaves are complex processes, strongly controlled by the transpiration stream. For this reason, metal concentrations in leaf tissues must be interpreted with caution (Jentschke & Godbold, 2000). In the present experiment, the transpiration stream slowed down most dramatically in plants inoculated with the Zn-sensitive fungus (Fig. 2c). The accumulation pattern of Zn in needles was different between both inoculation treatments (Table 1), and confirms that the benefit to pines conferred by the Zn-tolerant genotype surpassed that of the sensitive isolate.

Although we saw a clear difference in nutrient uptake capacity and nutrient status of the plants, we found no significant difference in plant growth. That also means that differences in plant biomass cannot be responsible for the higher nutrient acquisition and transpiration observed in pines inoculated with the Zn-tolerant fungus. The absence of a significant differential growth response is caused by the slow physiological adaptation of growth rates of woody plants to changing environmental conditions, including nutrient deficiencies or toxicities (Ingestad & Kähr, 1985; Van Tichelen et al., 2001). Microorganisms respond much more quickly to changing environmental conditions. In this short-term experiment we could clearly see that excess Zn significantly reduced the growth of the Zn-sensitive fungus.

The importance of a well established ectomycorrhizal mycelium for nutrient acquisition is illustrated by the close relationship between fungal biomass and P capture in the plants inoculated with the Zn-sensitive fungus (Fig. 4). The mycorrhizal fungus contributes considerably to the P assimilation of the plants, but when the fungal symbiont suffers Zn stress, the plant roots cannot themselves compensate for P acquisition.

The experiment shows that a genetic adaptation for increased tolerance to Zn is required for normal development of S. bovinus itself, and also to maintain a normal level of nutrient acquisition in mycorrhizal pines exposed to high Zn concentrations. It is important to remember that the mycorrhizal plants were grown at Zn concentrations found in the pore water of toxic soils. On these Zn-contaminated soils, tree seedlings that are not rapidly colonized by a well adapted fungus risk severe nutrient deficiencies. It is often assumed that the toxicity of metal-polluted soils, combined with the necessity of the ectomycorrhizal symbiosis, imposes strong selection pressure for heavy metal adaptation in both symbionts. Such a coevolution towards increased resistance against a pollutant was previously described for populations of the ericoid mycorrhizal fungus Hymenoscyphus ericae and its host plant Calluna vulgaris growing on arsenic spoils (Sharples et al., 2000). However, the host pines of the Zn-adapted Suillus populations apparently have not developed additional Zn tolerance, in contrast to local grasses and bacteria (Diels & Mergeay, 1990). The anthropogenic pollution of soils has caused an unprecedented, rapid change in environmental conditions which is likely to override the adaptive potential of most plants, especially that of tree species with their long reproductive cycles (Schützendübel & Polle, 2002). As plants are sessile organisms and have only limited mechanisms for stress avoidance, they need flexible means for acclimation to changing environmental conditions. We therefore speculate that these trees can resist this extreme environment through their association with a small guild of well adapted ectomycorrhizal fungi. Plant adaptation to selective pressures is often considered to be regulated by the plant genome, but it is evident that mutualistic fungi can contribute to plant adaptation. Recently, it was shown that a fungal endophyte from the genus Curvularia can improve the thermotolerance of the hot-spring panic grass, Dichanthelium lanuginosum (Redman et al., 2002).

The specific heavy metal-adapted SuillusPinus combination is most suited for land reclamation of metalliferous and industrial sites that have little plant cover because of their phytotoxicity. Phytostabilization of the vast areas of Zn-contaminated soils or wastes on mining and smelting sites is urgently needed to reduce dispersion of the contaminants. Phytostabilization, combined with in situ immobilization of heavy metals, is one low-cost option currently available to deal with this problem (Van der Lelie et al., 2001). Reforestation would be especially attractive for industrial and mining sites, as forests reduce erosion, restrict the dispersal of metals, and are sinks for atmospheric carbon dioxide.


We thank B. Willems, C. Put and A. Wijgaerts for experimental assistance. For the nitrogen analyses we are grateful to E.A. Hobbie at the University of New Hampshire. This research was supported in part by the FWO-Vlaanderen and by the EU (MYCOREM project no. QLK3-CT-1999-00097). K.A. is supported by an IWT grant, and D.v.d.L. is supported by Laboratory Directed Research and Development funds at the Brookhaven National Laboratory under contract with the US Department of Energy.