Zn pollution counteracts Cd toxicity in metal-tolerant ectomycorrhizal fungi and their host plant, Pinus sylvestris


  • Erik Krznaric,

    1. Hasselt University, Centre for Environmental Sciences, Environmental Biology Group, Agoralaan, Gebouw D, 3590 Diepenbeek, Belgium.
    Search for more papers by this author
  • Jan H. L. Wevers,

    1. Hasselt University, Centre for Environmental Sciences, Environmental Biology Group, Agoralaan, Gebouw D, 3590 Diepenbeek, Belgium.
    Search for more papers by this author
  • Christophe Cloquet,

    1. Ghent University, Department of Analytical Chemistry, Krijgslaan 281 – S12, 9000 Ghent, Belgium.
    Search for more papers by this author
    • Present address: CRPG, Centre de Recherches Pétrologiques et Géochimiques, 15 rue Notre Dame des Pauvres, 54501 Vandoeuvre lès Nancy, France.

  • Jaco Vangronsveld,

    1. Hasselt University, Centre for Environmental Sciences, Environmental Biology Group, Agoralaan, Gebouw D, 3590 Diepenbeek, Belgium.
    Search for more papers by this author
  • Frank Vanhaecke,

    1. Ghent University, Department of Analytical Chemistry, Krijgslaan 281 – S12, 9000 Ghent, Belgium.
    Search for more papers by this author
  • Jan V. Colpaert

    1. Hasselt University, Centre for Environmental Sciences, Environmental Biology Group, Agoralaan, Gebouw D, 3590 Diepenbeek, Belgium.
    Search for more papers by this author

E-mail jan.colpaert@uhasselt.be; Tel. (+32) 11 268304; Fax (+32) 11 268301.


Adaptive Zn and Cd tolerance have evolved in populations of the ectomycorrhizal fungus Suillus luteus. When exposed to high concentrations of both metals in vitro, a one-sided antagonism was apparent in the Zn- and Cd-tolerant isolates. Addition of high Zn concentrations restored growth of Cd-stressed isolates, but not vice versa. The antagonistic effect was not detected in a S. luteus isolate from non-contaminated land and in Paxillus involutus. The fungi were inoculated on pine seedlings and subsequently exposed to ecologically relevant Zn and Cd concentrations in single and mixed treatments. The applied doses severely reduced nutrient acquisition of non-mycorrhizal pines and pines inoculated with metal-sensitive S. luteus. Highest translocation of Zn and Cd to shoots occurred in the same plants. Seedlings inoculated with fungi collected from the polluted site reduced metal transfer to their host and maintained nutrient acquisition under high metal exposure. The isolate showing highest tolerance in vitro also offered best protection in symbiosis. The antagonistic effect of high Zn on Cd toxicity was confirmed in the plant experiment. The results indicate that a Zn- and Cd-polluted soil has selected ectomycorrhizal fungi that are able to survive and protect their phytobiont from nutrient starvation and excessive metal uptake.


Heavy metal pollution of terrestrial environments is seldom due to a single metal pollutant. Multiple metal contamination can be caused by mining and processing of ores containing several metal elements or by the use of different ores in smelting facilities. The toxicity of these metal cocktails to specific soil biota depends on a number of soil parameters which determine metal mobility, speciation and eventually bioavailability. Also indirect interactions among metals in soils may affect their toxicity to vegetation and soil organisms. Zn/Cd ratios have a major impact on Cd toxicity in plants because both metals may compete for binding sites in proteins, including Zn transporters (Hart et al., 2002; Assunção et al., 2008; Ernst et al., 2008).

In environments contaminated with a mixture of metals it is often hard to identify which metal(s) really impose selection pressure on a particular species. In order to detect past events of specific adaptations, tolerant ecotypes can be traced in laboratory dose–response experiments with single metal additions. In most organisms heavy metal adaptation, if it occurs, is very metal-specific and only active towards those metals that cause toxicity within wild-type populations. In the fungal kingdom, few examples have been reported of heavy metal-adapted populations in polluted environments.

The toxicity of individual metals to a wide range of plants, mycorrhizal fungi, rhizosphere and soil organisms has been investigated in many studies. Although not in unison, most investigations focusing on mycorrhizal fungi indicate that ectomycorrhizal fungi can help tree seedlings to survive on metal-polluted soils, both by sustaining mineral nutrition and by reducing metal transfer to the host (Jentschke and Goldbold, 2000; Colpaert, 2008). The adaptive potential of tree species with their long reproductive cycles might be too limited to quickly evolve metal tolerance (Schützendübel and Polle, 2002), and therefore symbiotic microorganisms, that potentially can adapt more quickly, come on the forefront. However, few experimental studies with ectomycorrhizal trees considered toxicity caused by multiple metal contamination, despite the fact that it is the rule under field conditions (Hartley et al., 1997).

For about one century, metal smelters in northern Limburg (Belgium) have processed zinc-lead ores, sphalerite and galena, which also contained significant amounts of cadmium. These three metals have all accumulated in soils in a wide area around the factories. Bioavailable zinc and cadmium were high enough to cause severe toxicity and thus selection pressure on the primary producers around the smelters. During previous decades, metal pollution resulted in the establishment of species-poor plant and microbial communities in the vicinity of these non-ferro smelters. These communities are characterized by the occurrence of Zn- and Cd-tolerant ecotypes of grasses (mainly Agrostis capillaris) (Vangronsveld et al., 1996). Previous screening experiments with a number of ectomycorrhizal fungal species exposed to single metal (Zn, Cd, Cu, Ni, Pb) gradients have illustrated the existence of an increased Zn and Cd tolerance in Suillus populations thriving in these metal-polluted soils, Zn tolerance being stronger and more widespread than Cd tolerance (Colpaert et al., 2004; Krznaric et al., 2009). A similar phenomenon was also found in culturable prokaryotes isolated from the same soil (Diels and Mergeay, 1990). Specific metal exclusion mechanisms help these Suillus fungi to survive when exposed to a high dose of these metals and offer protection to their host plant (Adriaensen et al., 2004; 2006; Krznaric et al., 2009).

Plants from metalliferous soils often exhibit combined tolerance to different heavy metals. Such tolerance could rely either on a combination of different metal-specific tolerance mechanisms (‘multiple tolerance’), or on less specific mechanisms that pleiotropically confer tolerance to different metals (‘cotolerance’) (Schat and Vooijs, 1997; Ernst et al., 2008). Suilloid populations from heavy metal polluted soils are unlikely to have evolved cotolerance mechanisms (Colpaert et al., 2000; Adriaensen et al., 2005).

In the present study, interactions between toxic concentrations of Cd2+ and Zn2+ ions were investigated in ectomycorrhizal fungi. Four fungal isolates were exposed both in vitro and in symbiosis to sublethal concentrations of Zn or Cd, and to the combination of both metals. Selected fungi exhibited varying degrees of Zn and Cd tolerance under single metal exposure. One metal-sensitive Suillus luteus originated from a non-contaminated area. Two S. luteus isolates originated from the polluted area both with high Zn tolerance but intermediate or high Cd tolerance and one Paxillus involutus isolate was collected from the polluted area. The Suillus isolates were chosen because of their widespread occurrence in the area and their specific evolutionary metal adaptation, Paxillus was included because it also occurs on the polluted soils – though it does not seem to develop adaptive metal tolerance – and because it is also a model species for metal studies in ectomycorrhizal fungi. The experiments were set up to investigate the growth and functioning of the different fungi under single and multiple metal stress and to detect possible Zn–Cd interactions in different pine–fungus combinations. Non-mycorrhizal plants were included for reference.


Zn–Cd interaction in vitro

The results from the in vitro study illustrate the existence of Zn and Cd tolerance in both S. luteus isolates from Lommel-Maatheide (Fig. 1). The Cd tolerance of UH-Slu-Lm2 is higher than the Cd tolerance of UH-Slu-Lm8. The S. luteus isolate from the non-contaminated site (UH-Slu-P13) is sensitive to both metals. Despite its Lommel-Maatheide origin, the P. involutus isolate is also sensitive to Zn and Cd, at least in vitro (Fig. 1). The Cd-EC50 values (the Cd doses that reduce biomass production by 50%) were affected by the Zn concentration in the medium (Table 1). The reference value (100%) is the treatment in which most biomass (dry weight) was produced. The Cd-EC50 values increased in the two S. luteus isolates from Lommel-Maatheide, when more Zn was added to the culture medium, indicating a strong antagonistic effect. The toxicity of Cd was significantly reduced by elevated Zn in these fungi. For example, at 0.2 mM Cd, biomass production of UH-Slu-Lm2 increased from 11.4 ± 1.2 mg at 0.02 mM Zn to 48.2 ± 1.3 at 5 mM Zn. In the metal-sensitive S. luteus isolate, adding more Zn had no ameliorating effect on Cd toxicity; an increase of the Cd-EC50 value was not observed at higher Zn exposure (Table 1). In the Paxillus isolate adding more Zn increased Cd toxicity, suggesting a rather synergistic effect, illustrated by a lower Cd-EC50 in the presence of 1.25 mM Zn. In none of the isolates Zn toxicity could be reduced by adding Cd to the media. In those treatments where Cd reduced growth, secretion of yellow-brown pigments into the culture medium was observed, in particular by Paxillus. Zn toxicity did not cause any secretion of pigments.

Figure 1.

Tolerance indices of Suillus luteus UH-Slu-P13, UH-Slu-Lm8 and UH-Slu-Lm2 and Paxillus involutus (A, B, C, D, respectively) exposed to binary Zn/Cd treatments in vitro. Tolerance indices were calculated as the percentage of biomass retained on the different media as compared with the highest biomass produced by each isolate. Colour code from red to purple represents values from 100% to 0% in steps of 5%.

Table 1.  Cd-EC50 values (mM Cd2+) of Suillus luteus UH-Slu-P13, UH-Slu-Lm8 and UH-Slu-Lm2 and Paxillus involutus UH-Pi-Lm23 as affected by Zn concentration when exposed in vitro to binary Zn/Cd treatments.
Zn2+ (mM)UH-Slu-P13UH-Slu-Lm8UH-Slu-Lm2UH-Pi-Lm23
1.250.0210.126> 0.20.015
2.50.0010.164> 0.2
5.00.163> 0.2

Plant and fungal growth

In the plant experiment, shoot and root biomass of the mycorrhizal and non-mycorrhizal seedlings was not significantly reduced by the metal treatments, although roots in some high metal treatments were darker and more brittle at the time of harvest. Fungal biomass was differentially affected by the metal treatments (Fig. 2).

Figure 2.

Total fungal biomass (mg d. wt.) of Suillus luteus UH-Slu-P13, UH-Slu-Lm8 and UH-Slu-Lm2 (white, light grey and dark grey bars). Paxillus involutus is shown in black. Biomass was calculated from ergosterol analyses at the harvest of the pine trees. Bars represent the mean ± SE, n = 5 (F12,80 = 6.32; Pinteraction < 0.0001).

In the control treatment, root systems and perlite were always fully colonized with healthy mycelium, the three S. luteus isolates produced similar quantities of active biomass, P. involutus produced more mycelium.

In the 150 μM Zn treatment, all mycorrhizal plants were well-colonized but at harvest external mycelia of UH-Slu-P13 were sparse and dark. The active fungal biomass of this isolate was reduced to less than 20% as compared with the control condition. The biomass of the other fungi was not significantly reduced by 150 μM Zn.

The 10 μM Cd treatment caused the strongest reductions in fungal biomass. Visual inspection of the root systems showed that UH-Slu-P13 continued to grow within the first weeks of Cd exposure; however, during the last month of exposure its mycelia were clearly deteriorating, eventually resulting in an active biomass of less than 10% as compared with the control condition. Active biomass of the Paxillus isolate was reduced to 44% of that measured under control condition, and yellow-brown pigments were excreted in the nutrient solution. Growth of UH-Slu-Lm2 was hardly affected by 10 μM Cd, whereas active biomass of UH-Slu-Lm8 was reduced to 58%.

In the 150 μM Zn/10 μM Cd metal treatment, active fungal biomass of all three Suillus isolates was similar to that observed in the 150 μM Zn treatment. The Paxillus isolate produced less biomass in the combined treatment than in the high Zn alone treatment; nevertheless its biomass remained higher than in the 0.1 μM Zn/10 μM Cd treatment. An ameliorating effect of 150 μM Zn on Cd toxicity was also observed with UH-Slu-Lm8 and UH-Slu-P13, although mycelial growth remained very poor with the latter fungus.

Nutrient uptake capacity

Nutrient uptake was assessed using net inorganic phosphate (Pi) and ammonium (NH4+) uptake as markers, after 4 and 8 weeks of treatment. Metal treatment effects were more pronounced after 8 weeks than after 4 weeks. Only the results from week 8 are illustrated in Fig. 3.

Figure 3.

Phosphate (A) and ammonium (B) uptake rate (μmol h−1) in pines (Pinus sylvestris) treated with Zn, Cd, or a combination. The pines are inoculated with Suillus luteus genotypes UH-Slu-P13, UH-Slu-Lm8 and UH-Slu-Lm2 (white, light grey, dark grey bars), with a Paxillus involutus isolate (black bars) or were left non-mycorrhizal (hatched bars). The measurement was performed after 8 weeks of treatment. Each bar represents the mean ± SE, n = 5 (F12,80 = 27.33; Pinteraction < 0.0001).

In the control treatment, Pi uptake is about two times higher in Suillus inoculated pines compared with non-mycorrhizal plants, but in the Paxillus-pine combination it is six times higher than in non-mycorrhizal plants. Ammonium uptake is on average 2.1 and 2.6 times higher in Suillus, respectively, Paxillus when compared with the non-inoculated seedlings.

In the 150 μM Zn treatment, Pi uptake was only weakly reduced after 4 weeks in Paxillus-inoculated plants. After 8 weeks, it further decreased to 50% of the value recorded in untreated plants. Among the Suillus isolates, only plants with UH-Slu-P13 showed a clear decrease in Pi acquisition. Also in non-mycorrhizal plants, Pi uptake was severely reduced. Elevated Zn results also in a strong reduction of the ammonium acquisition in plants inoculated with UH-Slu-P13 and in non-mycorrhizal seedlings.

The 10 μM Cd treatment caused severe reductions in both Pi and NH4+ uptake capacity in all fungal treatments, except for plants inoculated with UH-Slu-Lm2, the most Cd-tolerant fungus in vitro. After 4 weeks' treatment, plants associated with UH-Slu-Lm8 were more similar to those inoculated with UH-Slu-Lm2; however, at harvest their nutrient mobilization capacity had collapsed so that they resembled plants inoculated with UH-Slu-P13. These data confirm the intermediate Cd tolerance of UH-Slu-Lm8.

The combination of high Zn and high Cd resulted in low Pi and NH4+ uptake rates in non-mycorrhizal plants and plants inoculated with UH-Slu-P13. In seedlings with UH-Slu-Lm8 and UH-Pi-Lm23, nutrient uptake was more efficient compared with the low Zn/high Cd treatment, but less efficient compared with the high Zn treatment. Pi and NH4+ uptake were least hampered in the plants colonized with UH-Slu-Lm2.

Metal uptake into pine needles

In the control treatment, mycorrhizal colonization resulted in higher Zn needle concentrations: 26.7 mg g−1 versus 16.8 mg g−1 in non-mycorrhizal plants (Fig. 4). Seedlings exposed to 150 μM Zn showed the opposite pattern: non-mycorrhizal plants had more Zn in their needles than mycorrhizal ones. Transfer of Zn to needles was lower in plants inoculated with Zn-tolerant fungi, an effect that was further strengthened in the 150 μM Zn/10 μM Cd treatment for plants inoculated with UH-Slu-Lm2, UH-Slu-Lm8 and UH-Pi-Lm23.

Figure 4.

Zn concentration in the needles of Cd/Zn-treated mycorrhizal Pinus sylvestris. Hatched bars refer to the non-mycorrhizal state; white, light grey and dark grey bars to the Suillus luteus genotypes UH-Slu-P13, UH-Slu-Lm8 and UH-Slu-Lm2 respectively; black bars to the Paxillus involutus genotype UH-Pi-Lm23. Mean ± SE, n = 5 (F12,80 = 4.70; Pinteraction < 0.0001).

Cd transfer towards needles was highest in non-mycorrhizal plants and was differentially reduced by different mycorrhizal isolates (Fig. 5). UH-Slu-P13 was least efficient in reducing Cd transfer to needles. The two S. luteus isolates from Lommel-Maatheide were most efficient in avoiding Cd-transfer to the shoots in the 0.1 μM Zn/10 μM Cd treatment. In the 150 μM Zn/10 μM Cd treatment, the presence of excess Zn could reduce the excessive Cd uptake into needles in non-mycorrhizal plants and in plants colonized with UH-Pi-Lm23 and UH-Slu-P13. On the contrary, under these conditions Cd transfer to needles increased when the pine seedlings were inoculated with UH-Slu-Lm2 or UH-Slu-Lm8.

Figure 5.

Cd concentration in the needles of Cd/Zn-treated mycorrhizal Pinus sylvestris. Hatched bars refer to the non-mycorrhizal state; white, light grey and dark grey bars to the Suillus luteus genotypes UH-Slu-P13, UH-Slu-Lm8 and UH-Slu-Lm2 respectively; black bars to the Paxillus involutus genotype UH-Pi-Lm23. Mean ± SE, n = 5 (F12,80 = 6.04; Pinteraction < 0.0001).


Metal-tolerant ecotypes

Contamination of the Lommel-Maatheide area was due to processing of zinc-lead ores, mainly sphalerite and galena, that contained significant amounts of Cd. Analyses of pore waters from the Lommel-Maatheide soil confirm an elevated availability of both Zn and Cd (Krznaric et al., 2009). Co-occurrence of these metals is a normal condition in Zn minerals, resulting in Zn/Cd molar ratios in soil solution of mine tailings that vary from 50 to 2000 (Ernst and Nelissen, 2000). Simultaneous tolerance against Zn and Cd has been repeatedly observed in prokaryote populations from metal-contaminated environments (Mergeay et al., 2003), but also in plant populations thriving on metalliferous soils (Simon, 1977; Roosens et al., 2003; Arnetoli et al., 2008). However, the co-occurrence of zinc and cadmium tolerance does not imply similar tolerance or detoxification mechanisms for both metals, evidence for cotolerance is not strong (Schat et al., 2000; Hall, 2002; Ramesh et al., 2009; Verbruggen et al., 2009). Only toxic Cd concentrations induced the secretion of pigments by our fungi both in the in vitro and in symbiosis experiment. Fungal pigments, often polyphenolic compounds, have been associated with anti-oxidative activity (Kasuga et al., 1995).

Zn–Cd interaction in vitro

The in vitro screening of the fungi shows a reduction of Cd toxicity by elevated Zn in the S. luteus isolates from Lommel-Maatheide. In the two other fungi, the antagonism was not observed, although we cannot exclude that there is some protection in the Zn range from 0.02 to 1.25 mM Zn, the lowest Zn concentration in the gradient. Hartley and colleagues (1997) report reduction of Cd toxicity by Zn in three ectomycorrhizal fungi collected from non-polluted environments, including S. luteus and P. involutus. The one-sided antagonism was observed at lower Zn/Cd concentrations: 0–0.5 mM Zn in factorial combination with 0–0.025 mM Cd. In a fourth species, Suillus granulatus, Cd toxicity rather increased when more Zn was added to the medium (Hartley et al., 1997). The antagonism in the Lommel-Maatheide Suillus isolates is one-sided. Cd toxicity is reduced by Zn but not vice versa. Although we previously showed that adaptive Zn tolerance itself does not necessarily coincide with an elevated Cd tolerance (Cd-only additions), the present experiment suggests that the Zn tolerance mechanism may further increase Cd tolerance when both metals are in excess. Further research will be necessary to elucidate whether the antagonism is caused by some competitive transport interaction at the plasma membrane, by a change in regulation of transporter expression or by a physiological detoxification response induced by elevated Zn that coincidentally confers decreased Cd sensitivity.

Investigating the effects of binary Zn/Cd exposures is of great importance, not only because of the regular co-occurrence of both metals, but also because of possible antagonistic interactions. Ernst and colleagues (2008) point to a major problem in many Cd studies wherein plants are exposed to concentrations that are ecologically and physiological irrelevant (too high) and where Zn/Cd ratios can be considerably lower than 0.1. This problem should be avoided because the toxicity of Cd can be highly modified by the Zn/Cd ratio, not only in plants (Marschner, 1995; Hart et al., 2002), but also in fungi (Colpaert and Van Assche, 1992; Hartley et al., 1997). The well-recognized inhibition of Cd uptake by Zn is mostly explained through a common use of the same transport systems (Cataldo et al., 1983; Hart et al., 2002).

Plant experiment

In this study, the biomass of the plants was not yet significantly reduced by the metal treatments. Woody plants show in general a slow physiological adaptation to changing environmental conditions, e.g., deficiencies or toxic metal concentrations (Ingestad and Kähr, 1985; Adriaensen et al., 2004).

In contrast to woody plants, microorganisms respond much faster to toxic soil metal concentrations. The biomass of the fungal isolates in the present study was clearly affected by the metal treatments. Growth of the isolates varied strongly and depended on genotype and metal treatment.

The measurements of nutrient uptake rates are a sensitive method to quantify the functioning of the root systems in the different treatments (Van Tichelen et al., 1999). Plants associated with the metal-sensitive S. luteus isolate showed severely reduced uptake rates after 8 weeks of metal treatments; they were hardly different from non-mycorrhizal plants. Uptake rates of plants associated with Zn-tolerant isolates were not affected by the Zn treatments; plants associated with the most Cd-tolerant UH-Slu-Lm2 were the only ones that were able to maintain a high phosphate and ammonium uptake rate after 8 weeks of 10 μM Cd treatment.

Despite its smaller size, the plant experiment also confirmed that Cd toxicity can be moderated by the addition of a high Zn concentration. The protective effect of Zn towards Cd toxicity was in general confirmed in mycorrhizal pines when fungal biomass and nutrient uptake are considered. In the plant experiment, the antagonistic effect between Zn and Cd was also observed with Paxillus, in contrast to the in vitro screening. The Zn/Cd antagonism was not confirmed when Cd transfer to shoots was considered in plants inoculated with the S. luteus isolates from Lommel-Maatheide. Adaptive Cd tolerance in S. luteus was shown to be based on a Cd exclusion mechanism (Krznaric et al., 2009). It is possible that this exclusion mechanism is triggered more strongly in the absence of elevated Zn, explaining why the lowest Cd transfer is recorded in host plants in the 0.1 μM Zn/10 μM Cd treatment.

Zn pollution counteracts Cd toxicity

The combined Zn/Cd treatment is representative for the actual condition of the soil solution in Lommel-Maatheide (Krznaric et al., 2009). Zn and Cd concentrations in the needles of the metal-exposed seedlings were within the range that can be measured in field-collected pine needles from this site (Colpaert et al., 2004). Pines inoculated with the three fungi isolated from Lommel-Maatheide clearly have better chances for survival than non-mycorrhizal pines or pines inoculated with a metal-sensitive fungus. The former fungi can keep up growth and nutrient uptake and they reduce Zn and Cd transfer to their host. Analyses of element profiles of Zn and Cd-tolerant S. luteus showed that the tolerance mechanism is largely based on Zn and Cd exclusion mechanisms (Colpaert et al., 2005; Krznaric et al., 2009), mechanisms that may contribute to the low metal transfer to shoots. The multipolluted environment of Lommel-Maatheide has clearly selected ectomycorrhizal fungi that offer good opportunities for survival of their host plants, both in terms of a safe nutrient acquisition and a reduced transfer of metals to shoots. In soil solution, Zn and Cd reach concentrations that cause adverse effects on nutrient acquisition in non-mycorrhizal pines and pines inoculated with non-adapted fungi, at least when applied individually (Adriaensen et al., 2006; Krznaric et al., 2009). The present experiment showed that the protective effect of the metal-tolerant S. luteus was maintained under a combined Zn-Cd pollution. The evolutionary Zn and Cd adaptation in this fungus was necessary for the survival of the species in Lommel-Maatheide soil, but meanwhile also contributed to the protection of its host plant. The antagonistic effect of Zn on Cd stress may explain why adaptive Cd tolerance in Suillus populations is not as widespread as Zn tolerance (Krznaric et al., 2009). Nevertheless, elevated Cd tolerance has a selective advantage because highest protection of host plants against Zn/Cd stress was achieved in those seedlings inoculated with the most Cd-tolerant Suillus isolate. Paxillus involutus which is often studied in relation to Cd toxicity (Bellion et al., 2006; Johansson et al., 2008) was shown to be quite metal-sensitive in vitro, especially when compared with the S. luteus isolates from Lommel-Maatheide. Nevertheless, the Paxillus isolate was able to protect the pine seedlings in the present experiments, in particular when compared with non-mycorrhizal plants or plants inoculated with a non-adapted S. luteus. In the 10 μM Cd treatment, UH-Slu-Lm2 performed clearly better in terms of nutrient acquisition. In plant experiments P. involutus rapidly produces large amounts of fungal biomass and it is possible that this offers some advantage to this fungus in experiments where there is a time shift between inoculation of the plants and the start of the metal treatments (4 weeks in this experiment).

In the control treatment, P. involutus exhibited a very high phosphate uptake capacity, a trait that was already observed in previous experiments with another Paxillus isolate (Colpaert et al., 1999). P uptake in pines colonized with P. involutus declined in the presence of elevated Zn and Cd. This is partly due to a reduced fungal growth, but in the 150 μM Zn treatment, fungal growth was still unaffected. Specific interactions between phosphate and zinc metabolism have been revealed in many organisms (Huang et al., 2000; Jensen et al., 2003; Jackson et al., 2008). Adriaensen and colleagues (2004) previously speculated that an elevated Zn status in Suillus bovinus may be responsible for a downregulation of the expression or activity of some P transporter genes or proteins in this species.

Finally, the results also suggest that soils contaminated by a mixture of Zn and Cd might not be as toxic to fungi and plants as predicted from toxicity assessment of the individual metals. We further corroborate that survival of at least some plant species on metal-contaminated sites is not only a matter of the plant genome but that it is also a matter of symbiotic microorganisms, both fungi and prokaryotes (Hall, 2002; Schützendübel and Polle, 2002; Kozdrój et al., 2007) that may evolve themselves increased heavy metal tolerance.

Experimental procedures

Fungal material

Two fungal species were selected for the experiments: the basidiomycetes S. luteus (Linnaeus: Fries) Roussel (three isolates) and one P. involutus (Batsch: Fries) genotype. Both species are mycobionts from pine on metal-polluted and unpolluted soils. The three S. luteus isolates were UH-Slu-P13, collected from an unpolluted area in Paal (B); UH-Slu-Lm8 and UH-Slu-Lm2, two isolates from Lommel-Maatheide (B). The three S. luteus isolates had different degrees of Zn and Cd tolerance (Fig. 1, Table 1). One P. involutus isolate was included, UH-Pi-Lm23 collected from Lommel-Maatheide. This isolate had Zn and Cd tolerances that fell within the range found for the S. luteus isolates. Paxillus involutus is considered to have a fairly high constitutive tolerance to heavy metals.

Zn–Cd interaction in vitro

In this experiment, the effects of interaction between Zn and Cd on biomass production was investigated. The four fungal isolates were grown on solid modified Fries medium (Colpaert et al., 2004) supplemented with 0, 0.010, 0.025, 0.050, 0.100 or 0.200 mM Cd, through addition of 3CdSO4·8H2O. The Zn concentration was set to 0.02 mM (control), 1.25, 2.5, 5 or 15 mM Zn with ZnSO4.7H2O. All combinations of metals were investigated, each treatment was replicated five times. Mycelia were incubated at 23°C in darkness and were harvested after 12 days of exposure. Mycelia were frozen at −80°C, and subsequently freeze-dried before weighing. The Zn/Cd ratio varied from 0.1 to 1500.

Inoculation of the host

For the in-symbiosis experiments, fungi were associated with 7-week-old pine seedlings. Surface-sterilized seeds were sown in perlite, moistened with a balanced nutrient solution (mass ratios: 100 N/9 P/54 K/6 Ca/6 Mg/9 S + micronutrients) for Pinus sylvestris (Ingestad and Kähr, 1985; Colpaert and Verstuyft, 1999). Uniform seedlings were inoculated using a sandwich technique. The root system of pine seedlings was brought into close contact with fresh, actively growing mycelium grown on an agar medium covered with cellophane (Van Tichelen and Colpaert, 2000). In this way, the root system was inoculated quickly and evenly. The pine seedlings were inoculated with the above-mentioned fungi. One set of plants was left non-mycorrhizal, so that there were five inoculation treatments in the experiment. After inoculation, pines were transferred into containers filled with acid-washed perlite. Containers are 150 ml syringes with a semi-transparant wall, but kept in darkness to avoid growth of algae. Plants were grown in a semi-hydroponic system in which nutrient supply is tuned to nutrient uptake by plants. Nutrients were added every other day to obtain a relative growth rate of approximately 3% day−1. The pH of the solution was adjusted to 5. After 4 weeks, roots and perlite were visually colonized with mycelia and the metal treatments were started.

Metal treatments of plants

Four metal treatments were set up: 0.1 μM Zn (control), 150 μM Zn, 10 μM Cd and 150 μM Zn + 10 μM Cd. Zn was added as ZnSO4·7H2O and Cd as 3CdSO4·8H2O; a trace of Zn (0.1 μM) was also present in the 10 μM Cd solution. The control solution is most representative for the composition of the pore water in unpolluted Paal soil; the 150 μM Zn + 10 μM Cd treatment is a realistic condition for the Lommel-Maatheide site. In 2000, Cd in the pore water at Lommel-Maatheide (Lm) varied between 1 and 10 μM, Zn between 80 and 190 μM. The Zn and Cd concentration in soil solutions in Lommel-Maatheide are within the range reported for land polluted by emissions of industries (Ernst et al., 2008).

Exposure of tree seedlings to 150 μM Zn or 10 μM Cd have been shown to cause physiological damage (Schlegel et al., 1987; Balsberg Påhlsson, 1989; Adriaensen et al., 2004). Previous dose–response experiments showed that 150 μM Zn and 10 μM Cd (single metal applications) also cause strong differential responses, both at the level of growth and nutrient acquisition, in pines mycorrhizal with metal-sensitive or metal-tolerant isolates. There were five replicates for each metal-inoculation treatment. The treatments lasted for 8 weeks. At the start of the treatments, five plants of each plant–fungus combination were harvested for the determination of the initial condition of the plants.

Nutrient uptake capacity

Nutrient uptake of all individual plants was measured 4 and 8 weeks after the start of the metal treatments. The depletion of ammonium and phosphate was determined in a test solution that circulated through the plant containers during 5 h. Such measurements are non-destructive because the integrity of the fungus-root pathway is maintained (Colpaert et al., 1999). Under standard conditions, these measurements are highly reproducible and give an integrated view of the nutrient uptake capacity of the (mycorrhizal) root system. Nutrient uptake capacity is very sensitive to metal stress and therefore is a good marker for the prediction of metal stress in plants grown under steady-state conditions (Van Tichelen et al., 1999; Adriaensen et al., 2004). The concentrations of ammonium and phosphate were determined spectrophotometrically with a flow injection analyser. Uptake rates were calculated on the basis of the individual depletion curves in the range of 20–30 μM phosphate and 200–300 μM ammonium.

Harvest and analyses

After 8 weeks of metal treatment, plants were harvested; subsamples of root and perlite were frozen in liquid N2 in order to determine fungal biomass. The ergosterol concentration was used for estimation of active fungal biomass. Ergosterol was extracted from freeze-dried roots and perlite. Quantification was performed with high-performance liquid chromatography (Nylund and Wallander, 1992). Plant material (shoot, stem, roots) was dried at 70°C, weighed and subsequently pulverized to powder. Subsamples of pulverized plant material were destructed in concentrated acid (HNO3/HCl). The concentration of Cd and Zn was determined with inductively coupled plasma mass spectrometry.

Statistical analysis

The experiment had a crossed design with as factors inoculation treatment and metal treatment. To statistically analyse the biomass of plants and fungi, the results of the uptake measurements and the determination of Zn and Cd concentrations, two-way anova calculations were carried out using the proc GLM, proc anova and proc MIXED procedures of SAS. The Tukey-Kramer test was chosen as a post hoc test. The significance level was set at 0.05.


The research was financially supported by the Fund for Scientific Research – Flanders (FWO-projects G.0669.06 and G.0046.06) and a UHasselt Methusalem project (08M03). PhD grants were assigned by the Institute for the Promotion of Innovation through Science and Technology in Flanders (IWT-Vlaanderen) to Erik Krznaric and by the Fund for Scientific Research – Flanders to Jan Wevers.