The aim of the first part of our work was to develop a set of sensitive enzyme assays for functional profiling related to enzymatic activities of individual ECM tips. For each single assay, the optimal incubation time, substrate concentration and pH were derived from existing protocols. The detailed results are given in the following section.
Substrate saturation has already been discussed for chitinase, phosphatase and β-glucosidase by Pritsch et al. (2004). Saturation was reached at 500 µm of substrate per well for chitinase and β-glucosidase, whereas it was above 800 µm for acid phosphatase.
Cellobiohydrolase activity (Fig. 1a) reached an optimum at a substrate concentration of 400 µm per well and did not increase further at higher concentrations. For glucuronidase activity (Fig. 1b), the saturating substrate concentration was between 400 µm and 500 µm per well. With a concentration greater than 500 µm, we observed a small decrease of the activity. The four ECM morphotypes had similar xylosidase activities and reached saturation between 400 µm and 600 µm per well (Fig. 1c). Above 600 µm per well, the activity decreased. Leucine aminopeptidase activity showed an optimum at a substrate concentration of 300–400 µm for Tomentella sp. 1, L. quietus and Rhizopogon sp., and 400–500 µm for L. subdulcis (Fig. 1d). The activity decreased for all four ECM morphotypes at concentrations greater than 500 µm per well. For laccase, the substrate concentration was found to be optimal at 2 mm for L. quietus, Cortinarius anomalus (Fr. Fr.) Fr. and Tomentella sp. (data not shown).
Figure 1. Enzyme activities at increasing substrate concentrations (µmol MU or AMC per well) expressed as 4-methylumbelliferone (MU) or 7-amino-4-methylcoumarine (AMC) released per unit of time and per ectomycorrhizal (ECM) tip projected area (pmol mm−2 min−1) for individual excised ECM root tips. Bars represent SE (n = 7). Closed circles on solid line, Lactarius quietus; open circles on solid line, Lactarius subdulcis; closed circles on broken line, Tomentella sp. 1; open circles on broken line, Rhizopogon sp.
Download figure to PowerPoint
Optimal pH (4.5) has been discussed for chitinase, phosphatase and β-glucosidase activity by Pritsch et al. (2004).
The highest cellobiohydrolase activity was measured at acidic pH (< 4.5) for L. subdulcis (45 pmol mm−2 min−1) before a sharp decrease of activity until pH 7 (Fig. 2a). For the three other species, the activity was constant from pH 2 to pH 4 and highest from pH 4.5 to pH 5 (30 pmol mm−2 min−1 for L. quietus and Rhizopogon sp.; 15 pmol mm−2 min−1 for Tomentella sp. 1). Above pH 5, the activity was slightly reduced. The optimal pH was 4.5 for all species.
Figure 2. Enzyme activities at increasing pH expressed as 4-methylumbelliferone (MU) or 7-amino-4-methylcoumarin (AMC) released per unit of time per ectomycorrhizal (ECM) tip projected area (pmol mm−2 min−1) for individual excised ECM root tips. Bars represent SE (n = 7). Closed circles on solid line, Lactarius quietus; open circles on solid line, Lactarius subdulcis; closed circles on broken line, Tomentella sp. 1; open circles on broken line, Rhizopogon sp.
Download figure to PowerPoint
For glucuronidase activity, the optimal pH-range differed depending on the species: pH 3–4.5 for L. subdulcis and Rhizopogon sp., pH 4.5–5.5 for L. quietus and pH 3.5–5 for Tomentella sp. 1 (Fig. 2b).
Lactarius subdulcis showed highest xylosidase activity around pH 4–4.5, whereas it was around pH 4.5–5 for L. quietus (Fig. 2c). The pH optima were pH 4–5 for Rhizopogon sp. and pH 4.5–5.5 for Tomentella sp. 1.
Leucine aminopeptidase activity was constant for all four species from pH 2 to pH 5.5 (Fig. 2d). In this pH range, the maximal activity was shown by Tomentella sp. 1 (40 pmol mm−2 min−1). The activity increased twofold from pH 5.5 to pH 6 (Tomentella sp. 1 and Rhizopogon sp.) or from pH 5.5 to pH 6.5 (L. quietus and L. subdulcis). The activity was then constant to pH 7 for all species. For laccase, a pH optimum of 4.5–5 and lower has been described by Eggert et al. (1996), Bourbonnais et al. (1997) and Kanunfre & Zancan (1998).
Optimal incubation time
The enzyme activities recorded by these methods may result not only from the ectomycorrhizal fungus but also from the whole ectomycorrhizal complex, including root tissues and other associated microorganisms such as bacteria and free-living microfungi. In the case of field sampled ECMs, all these components belong to a natural mycorrhizosphere and neither can nor should be separated. To avoid proliferation of bacteria during the assays, which would lead to an overestimate of the bacterial contribution to enzyme activities, reaction times must be kept as short as possible. The fluorescence method has already been shown to allow the detection of enzymes such as acid phosphatase, chitinase and β-glucosidase after incubation times as short as 5, 10 and 20 min, respectively (Pritsch et al., 2004). For phosphatase detection, the fluorescence method therefore allowed a six- to over ten-fold reduction of incubation times (30–60 min) compared with the previously used photometric method using pNPP (Bartlett & Lewis, 1973; Tibett et al., 1998; Buée et al., 2004). In the present study, enzymes with lower activities could also be easily detected after 40 min (cellobiohydrolase) or 60 min incubation (xylosidase, glucuronidase and leucine aminopeptidase).
Laccase activity remained constant over 2 h for all morphotypes studied: L. quietus, L. chrysorrheus, B. atrovirens, Boletus sp. and E. nitidum (data not shown). Differences in laccase activities between the five ECM morphotypes appeared even before 30 min of incubation but were more distinct after 1 h, the incubation time which was chosen for routine assays. Controls with pasteurized mycorrhizal tips (30 min, 80°C) showed that oxidation of ABTS was restricted to living mycorrhizas.
Optimized protocol for subsequent assays on the same ectomycorrhizal tips
Assaying for different activities in succession on the same root tips was studied with two ECM morphotypes (L. quietus and Tomentella sp. 1). Since it has been previously demonstrated that there is no reduction in enzyme activities during longer incubation intervals (Pritsch et al., 2004), a number of combinations and orders for incubation with different substrates were tested. All MU substrates were put on one plate, to avoid interference with the two other substrates (ABTS and AMC). For MU-substrates, no interference of substrates was observed, resulting in the following optimal procedure: 14 tips of each ECM morphotype divided into two subsamples of seven tips each were assayed; the first subsample was used for all MU-substrates (e.g. chitinase, phosphatase, β-glucosidase, glucuronidase, xylosidase and cellobiohydrolase activities, in any order) and the second subsample was used to test for leucine aminopeptidase and laccase activity (in that order).
The final design for multiple activity tests with eight different substrates on two subsamples of seven mycorrhizal tips each (incubation time, pH, type and concentration of substrate) is summarized in Table 1.
Table 1. Optimized experimental conditions (incubation time, pH and substrate concentration) for multiple enzyme activity tests with eight different substrates on two subsamples of seven mycorrhizal tips each
|Subsample||Tests in order||Substrate||Incubation time (min)||pH||Concentration (µm)|
|No. 17 ECM tips||Glucuronidase (glr)||MU-β-d-glucuronide hydrate||60||4.5||500|
|No. 27 ECM tips||Chitinase (nag)||MU-N-acetyl-β-d-glucosaminide||20||4.5||500|
|Acid phosphatase (pho)||MU-phosphate||10||4.5||800|
|Leucine aminopeptidase (leu)||l-leucine-AMC||60||6.5||400|
|Laccases (lac)||ABTS||60||4.5|| 1.105|
At this point, it is important to stress that the necessity to extract ECMs from soils causes a loss of extramatrical mycelium (emanating hyphae and strands). It is known that extramatrical mycelium plays a major role in nutrient mobilization and uptake by ECMs (Agerer, 2001; Landeweert et al., 2003) and that there might be specific differentiations of mycelia depending on species and nutrient pools in soils (Agerer, 2001). Therefore, it has to be kept in mind that the interpretation of the results relies on activities measured in the close vicinity of ECMs deprived from emanating hyphae and strands because of extraction from the soil. Whether these activities correlate with the corresponding activities of whole ECMs, including their complete extramatrical mycelium, or whether parts of mycelia remote from the roots display functions other than those close to roots, are challenging questions for further studies. Despite the acknowledged importance of the extramatrical mycelium, there is still a lack of methods allowing study of functional aspects of field-grown mycelia in a specific way (Leake et al., 2004). Such studies need major methodological developments, which were beyond the scope of this exploratory work.
ECM activity profiling in the two forests stands
Ten ECM morphotypes (Table 2), identified at the genus (3) or species (7) level, sampled in the Champenoux and Kranzberg sites in August 2004, have been subjected to the set of eight enzymatic tests (Table 1). The results confirmed the low variability of enzyme activities among the seven ECM tips within a sample, as already found when optimizing the tests (Figs 1 and 2). Exceptionally, higher variability was only observed, when absolute values were low, as in the case of glucuronidase activity (Fig. 3). By contrast, enzyme activities may vary widely among different samples of the same ECM, taken a few meters apart in the same soil horizon (intersample variation). In the example of Tomentella sp. 1 in the A1 horizon at the Champenoux site (Fig. 3), sample 2 markedly differed from the three other samples by higher leucine-aminopeptidase activity and low laccase activity. This result indicates that enzymatic activities may be strongly influenced by the specific conditions at each sampling point. Thus, in further studies, the choice of a sampling strategy will need to take into account the fact that the activity profile of a given ECM morphotype is very homogenous at the centimeter scale, but highly variable at the meter scale.
Figure 3. Enzymatic activity profiles of Tomentella sp. 1 sampled in Champenoux A1 horizon in four different places a few meters apart in August 2004. The polar graphs have been drawn from the relative activities (solid line polygon) calculated per cent of the mean value of the four samples. The two dotted line polygons are ± SE (n = 7) among the seven ectomycorrhizal (ECM) tips used in the laboratory test to measure each enzyme activity. Abbreviations for enzymes: pho, acid phosphatase; nag, N-acetyl-glucosaminidase; gls, β-glucosidase; cel, cellobiohydrolase; xyl, xylosidase; lac, laccase; glr, glucuronidase; leu, leucine aminopeptidase.
Download figure to PowerPoint
At both sites, the 10 ECM types displayed remarkably different mean activity profiles (Fig. 4a,b). Two main factors are likely to be responsible for these different activity profiles: (1) species-specific enzyme expression, as already reported by Gramss et al. (1998) for mycelia in pure culture, Agerer et al. (2000) for sporocarps and Buée et al. (2004) for ectomycorrhizas; and (2) influence of local soil conditions (e.g. horizontal and vertical soil heterogeneity).
Figure 4. Enzymatic activity profiles of the 10 ectomycorrhizal (ECM) types sampled in different soil horizons in the Champenoux and Kranzberg forest sites in August 2004. The polar graphs have been drawn from the relative activities calculated as per cent of the mean value of all ECM types studied in the same community (site). Each graph was built with the mean value of the one to four replicates (root samples taken a few meters apart) within each species–horizon combination. The number of replicates (N) is given on top of each graph. Abbreviations for enzymes: pho, acid phosphatase; nag, N-acetyl-glucosaminidase; gls, β-glucosidase; cel, cellobiohydrolase; xyl, xylosidase; lac, laccase; glr, glucuronidase; leu, leucine aminopeptidase. (a) Champenoux site with two horizons: A1, top row; A2, bottom row. (b) Kranzberg site with three horizons: Of, top row; Oh, medium row; Ah, bottom row.
Download figure to PowerPoint
At the Champenoux site (Fig. 4a), Tomentella sp. 1 had very low activities for all enzymes, while the other ECMs appeared specialized in at least one function (Fig. 4a; B. atrovirens. laccase; C. olivaceofuscus, xylosidase; L. quietus, glucuronidase, xylosidase and leucine aminopeptidase). For the last three ECMs, shifts in their activity profiles were observed with depth. The observed increase of laccase activity of B. atrovirens in the deeper horizon (A2) points to the fact that laccase activity is involved in breaking down recalcitrant phenolic molecules, as found in humic substances, resulting in the release of entrapped nitrogen and phosphorus (Criquet et al., 1999). Therefore, higher laccase activity may indicate a possible mechanism of mobilization of these nutrients by B. atrovirens. The activity profiles of C. olivaceofuscus and L. quietus were also different in the deeper A2 horizon. While the former had higher β-glucosidase, L. quietus lost its high β-glucosidase and cellobiohydrolase relative activities compared with the A1 horizon, but had high leucine aminopeptidase values and acquired high relative activities for phosphatase and chitinase. These shifts in activities may reflect the need for a better ability to mobilize organic phosphorus and to gain access to nitrogen from proteins and chitin in the deeper horizon.
The ECM types at the Kranzberg site also showed different enzyme activity profiles within each horizon, and the same ECM expressed different enzyme profiles in different horizons (Fig. 4b). Especially high activities of the ECMs at Kranzberg were found for Tomentella sp. 2 (cellobiohydrolase and β-glucosidase), Xerocomus sp. (chitinase and phosphatase), R. nigricans and C. geophilum (laccase, glucuronidase and xylosidase), L. subdulcis (laccase and leucine aminopeptidase).
In the case of Tomentella sp. 2 (Fig. 4b), a shift in distinct activities (leucine aminopeptidase increase in Oh, and chitinase increase in Ah) was observed in different horizons, displaying two ways of N-mobilization from N-containing biopolymers (proteins and chitin). Xerocomus sp. ECMs showed increased intensity from Of to Ah horizons but not different expression patterns of potential enzyme activities (Fig. 4b). By contrast, R. fellea showed rather different profiles in the Of horizon with high leucine aminopeptidase activity and a lack of almost all activities (especially phosphatase and glucuronidase) found in the deeper horizons Oh and Ah (Fig. 4b). L. subdulcis, an abundant contact type (Agerer, 2001) in the litter layer, may use its proteolytic capacity to release amino acids from fresh litter while simultaneously profiting from its high laccase activity, which may enable this fungus to cope with high amounts of phenolic compounds in weakly degraded beech litter. In the Ah horizon, C. geophilum and R. nigricans displayed very similar profiles with high laccase, glucuronidase and xylosidase activities, following probably the same strategy in nutrient mobilization, as discussed for B. atrovirens at Champenoux.
At the stand level, the observed high functional diversity reflects a clear complementarity of the different ECM types. For the Champenoux site, the four types inhabiting the A1 horizon, or the three types in the A2 horizon, when considered together, were performing the eight enzymatic activities with values of c. 100% of the mean values in the community. Exceptions are phosphatase and chitinase activities which are slightly under-represented in the A1 horizon. At Kranzberg, similarly, the mean of all types in one horizon is around or above average of the total community with two clear exceptions: a very low glucuronidase activity in the litter layer and a total lack of leucine aminopeptidase activity in the Ah horizon. This suggests that all activities are not constitutively produced, but induced under specific soil conditions.
Although each ECM type appears to be more or less specialized in a small number of activities among the eight we measured here, activity profiles are clearly influenced by the location and also by the horizon from which the samples were taken. However, the occurrence of different ecotypes belonging to the same species but adapted to different soil horizons cannot be ruled out. Such short distance genetic and functional variability has not been reported, but Jany et al. (2002, 2003) showed that C. geophilum displayed a high diversity in ribotypes and respiration potential activity at the forest stand scale. To be elucidated, this point would necessitate more sophisticated molecular methods discriminating at the genet level.