The fate of [14C]glutamate and [14C]malate in birch roots is strongly modified under inoculation with Paxillus involutus

Authors

  • D. Blaudez,

    1. Université Henri Poincaré, Nancy I, Faculté des Sciences, Laboratoire de Biologie Forestière, associé INRA, BP 239, 54500 Vandœuvre-les-Nancy Cedex, France
    Search for more papers by this author
  • B. Botton,

    1. Université Henri Poincaré, Nancy I, Faculté des Sciences, Laboratoire de Biologie Forestière, associé INRA, BP 239, 54500 Vandœuvre-les-Nancy Cedex, France
    Search for more papers by this author
  • P. Dizengremel,

    1. Université Henri Poincaré, Nancy I, Faculté des Sciences, Laboratoire de Biologie Forestière, associé INRA, BP 239, 54500 Vandœuvre-les-Nancy Cedex, France
    Search for more papers by this author
  • M. Chalot

    1. Université Henri Poincaré, Nancy I, Faculté des Sciences, Laboratoire de Biologie Forestière, associé INRA, BP 239, 54500 Vandœuvre-les-Nancy Cedex, France
    Search for more papers by this author

Correspondence: M.Chalot. Fax: +33 3 83 91 22 43; e-mail: Michel.Chalot@scbiol.uhp-nancy.fr

ABSTRACT

The impact of inoculation with Paxillus involutus on the utilization of organic carbon compounds by birch roots was studied by feeding [14C]Glu or [14C]malate to the partners of the symbiosis, separately or in association, and by monitoring the subsequent distribution of 14C. Inoculation increased [14C]Glu and [14C]malate absorption capacities by up to eight and 17 times, respectively. Six- and 15-d-old mycorrhizal roots showed about four-fold higher [14C]Glu and [14C]malate absorption capacities compared with 60-d-old mycorrhizal roots, suggesting that the early stages of mycorrhiza formation induced higher requirements for C skeletons. Moreover, the results demonstrated that inoculation strongly modified the fate of [14C]Glu and [14C]malate. It was demonstrated that exogenously supplied Glu and malate might serve as C skeletons for amino acid synthesis in mycorrhizal birch roots and in the free-living fungus. Gln was the major 14C-sink in mycorrhizal roots and in the free-living P. involutus. In contrast, citrulline and insoluble compounds were the major 14C sinks in non-mycorrhizal roots, whatever the 14C source. It was concluded that mycorrhiza formation leads to a profound alteration of the metabolic fate of exogenously supplied C compounds. The ecological significance of amino acid and organic acid utilization by mycorrhizal plants is further discussed.

INTRODUCTION

Heterotrophic carbon assimilation by mycorrhizal birch has been estimated using 14C labelled proteins (Abuzinadah & Read 1986). The authors calculated that 9% of plant C may be derived from proteins. However, the utilization of C derived from low molecular weight (LMW) organic compounds (organic acids and amino acids) by mycorrhizal plants has not yet been investigated. Most of the work on ectomycorrhizal fungi or ectomycorrhizas has focused on the carbon supplies from carbohydrates to ectomycorrhizal fungi or on the transfer of C from carbon dioxide or glucose to amino acids (Smith & Read 1997; Martin, Boiffin & Pfeffer 1998). Moreover, most of the knowledge concerning the use of organic acids is based on studies with bacteria and yeasts, with few data available for filamentous fungi, and there is no information for ectomycorrhizal species (Jones 1998). However, filamentous fungi live in very different habitats and it was recently emphasized by Burgstaller (1997) that investigations with fungi that represent distinct habitats (mycorrhizal fungi, wood-decomposing fungi and the Achlya spp.) are urgently needed. Indeed, mycorrhizal fungi have a unique feature in comparison with most of the fungi studied so far in that they are able to form a symbiotic structure with plant roots.

LMW organic compounds are produced during decomposition of organic material in soils, for example litter and dead roots, by fungi and bacteria or released during root exudation and micro-organism activity (Fox & Comerford 1990; Johnson 1994; Grayston, Vaughan & Jones 1996; Jones 1998). In particular, Grayston et al. (1996) have recently reviewed the importance of root exudation on microbial activity and nutrient availability for trees and have highlighted the need for further studies on the assimilation of exudate compounds to estimate more accurately the root C flow. LMW organic acids exhibit strong complexing ability with Al and Fe and it has been suggested that they are important in the podzolization process in forest soils (Lundström, van Breemen & Jongmans 1995). They also have a high carboxyl content, which may be important for the buffering capacity of soil solutions. LMW organic compounds are readily degraded by micro-organisms (Abuzinadah & Read 1989; Hadas et al. 1992; Lundström et al. 1995; Grayston et al. 1996). Because ectomycorrhizal fungi are symbiotically associated with tree roots, they may play a crucial role in the N and C cycle of forest soils (Chalot & Brun 1998). Studies on the free-living ectomycorrhizal fungus Paxillus involutus (Batsch) Fr. have demonstrated that it is capable of actively metabolizing amino acids and the pathways involved have been studied in detail (Chalot et al. 1994a,b; Chalot & Brun 1998). However, the possibility that exogenous organic acids serve as a C source for ectomycorrhizal fungi and their host has not been explored.

These considerations are also important in terms of C translocation between trees interconnected by a common mycelial network. Recent studies have quantified the extent of interspecific C transfer between paper birch and Douglas fir and have unequivocally demonstrated net C transfer from paper birch to Douglas fir (Simard et al. 1997). Although the role of the ectomycorrhizal mycelium remained uncertain to the authors, it was suggested that C could be translocated between the two plants in the form of amino acids.

Glutamate and malate are amongst the most common LMW compounds in soils (Smith 1976; Abuarghub & Read 1988; van Hees, Andersson & Lundström 1996; Jones 1998), and they are two major components of the C metabolic pool in ectomycorrhizal fungi and birch roots (Blaudez et al. 1998). For these reasons, they were chosen to study the effect of inoculation on the utilization of LMW organic compounds by birch roots at two different stages of mycorrhiza formation and at a mature stage. Indeed, ectomycorrhizal development studies have shown that structural and functional integration of root and fungus can take place very rapidly (Brun et al. 1995).

We report here the utilization of [14C]Glu and [14C]malate by studying the distribution of 14C in the insoluble and soluble carbon pools of mycorrhizal and non-mycorrhizal birch roots and in the free-living fungus P. involutus. The experimental methods used were analogous to those of Brun et al. (1995), Blaudez et al. (1998) and Chalot et al. (1994a,b) to produce mycorrhizal plants at different stages of infection and to allow for comparison of 14C distribution with the free-living partners.

MATERIALS AND METHODS

Organisms and media

The ectomycorrhizal fungus used was an isolate of Paxillus involutus (Batsch) Fr. (ATCC 200175) which was originally isolated from a fruiting body growing under Betula pendula (Roth.) on coal waste in Midlothian, Scotland. It was maintained in Petri dishes by successive transfer on agar-modified Melin–Norkrans (MMN) medium from which malt extract was omitted. The MMN medium contained 10 g glucose L−1 and (mg L−1): KH2PO4 (500), (NH4)2HPO4 (250), CaCl2 (50), NaCl (25), MgSO4·7H2O (150), thiamine hydrochloride (0·1), FeCl3·6H2O (1). For the aseptic synthesis of ectomycorrhizas, P. involutus was grown on cellophane-covered agar MMN medium as described by Brun et al. (1995). Birch (Betula pendula Roth.) seeds were surface-sterilized with 3·5% calcium hypochlorite for 30 min, rinsed in several changes of sterilized distilled water, germinated aseptically on water agar for 1 week and placed on top of 10-d-old colonies of P. involutus growing on cellophane-covered agar medium (see above), in large round Petri dishes (150 × 15 mm). The plates were incubated in a growth chamber maintained under 16 h photoperiod (150 μmol m−2 s−1), temperature (day/night) and relative humidity were 22/18 °C and 85/65%, respectively. The roots of these seedlings were harvested 6 and 15 d after contact between the two partners. Control treatments including non-mycorrhizal birch seedlings and free-living mycelium were grown and treated in the same conditions. For production of older seedlings, mycorrhizal and non-mycorrhizal birch seedlings were further transferred after 4 weeks on peat in Perspex chambers. Roots of seedlings that were mycorrhizal by P. involutus and the root tips of the non-mycorrhizal seedlings were collected and used for 14C-feeding experiments after 2 months of growth on peat. Only newly formed roots (either mycorrhizal or non-mycorrhizal) after transfer on peat were used: they were about 60-d-old at harvest.

14C feeding experiments

Discs of mycelium were cut from the actively growing edge of 6- and 15-d-old colonies using a 25 mm diameter cork borer. Mycorrhizal and non-mycorrhizal roots (about 10 mg of fresh weight, root length <1·5 cm) were cut from birch seedlings.

Material was incubated for various lengths of time in 1 mL MMN medium (pH 5·5) at 25 °C either supplemented with 0·05 μCi L-[U-14C]Glu (250 μCi μmol−1), or with 0·2 μCi L-[U-14C]malate (51 μCi μmol−1). These concentrations of 14C-substrates (0·2 μM Glu and 4 μM malate) with high specific radioactivity were used to ensure detection of radioactivity in individual compounds. Higher quantities of [14C]malate were used to obtain detectable amounts of radioactivity in tissues and in individual compounds. Therefore, no attempt was made in the ‘Results’ and ‘Discussion’ sections to compare these two compounds with regard to their utilization rates. In a separate experiment, the 14CO2 produced during [14C]Glu or [14C]malate feeding was collected by circulating air continuously through the vial and trapping the outflowing gas by bubbling through a vial containing methanol/ethanolamine (70/30, v/v) and the radioactivity was measured by scintillation spectroscopy. At the end of the uptake period, the mycelium and the roots were washed for 5 min with 0·1 mM CaSO4 and freeze-dried prior to analysis. Radioactivity was determined in an aliquot of the incubation medium after addition of scintillation solution (Ultima Gold; Packard Instrument Co., Meriden, USA) using a Kontron Betamatic V (Kontron Instruments, Montigny-le-Bretonneux, France) counter.

The uptake rates of [14C]Glu and [14C]malate were calculated by measuring the disappearance of the radioactivity in the incubation medium during the linear phase of uptake, which occurred for at least 120 min (data not shown). The depletion of radioactivity at the end of the incubation period varied between 10 and 35% relative to the start.

Determination of 14C distribution

Distribution of the 14C was measured in the different fractions using the following procedures. Mycelium and root tissues were ground in 70% cold methanol. The extracts were centrifuged for 20 min at 14 000 g and the pellet washed with additional 70% cold methanol, recentrifuged and the supernatants were combined. Aliquots of the samples were removed for liquid scintillation counting of the radioactivity in the soluble fraction and the remainder used for further purification of amino acids and organic acids, according to the following procedure. The soluble fraction was applied to disposable columns of the cation-exchange resin AG 50 W-X8, H+ form (Sigma Chemical Co, St Louis, MO, USA) pre-equilibrated with water. The resin was washed with 6 mL of distilled water and the free amino acids (FAA) were eluted with 6 mL of 4·5 M ammonium hydroxide. The water eluate was applied to disposable columns of the anion-exchange resin AG1-X2, Cl form (Sigma Chemical Co). The resin was washed with 6 mL of distilled water and the FOA (free organic acids) were eluted with 6 mL of 4 M hydrochloric acid. Aliquots of ammonium and hydrochloric acid fractions were used to measure the radioactivity incorporated in the amino acids and the organic acids, respectively. Radioactivity of the neutral fraction (mainly soluble sugars) was determined in aliquots of the water eluate from the anion-exchange resin. Free amino acids were analysed by reverse-phase high performance liquid chromatography (HPLC) after derivatization with phenylisothiocyanate (PITC) reagent as described earlier (Blaudez et al. 1998). The radioactivity incorporated into the amino acids was measured by liquid scintillation counting of separate fractions corresponding to each peak in the HPLC-eluent collected at the outlet of the spectrophotometric detector. The radioactivity recovered in the neutral and FOA fractions was too low to detect the incorporation in individual compounds. The radioactivity was also determined in an aliquot of the methanol-insoluble pellet after tissue solubilization with Soluene 350 (Packard Instrument Co) and counted as described above.

Expression of results

Each experiment included three replicates. Data of 14C distribution in amino acids, expressed as a percentage of the total radioactivity in the FAA fraction, were arcsin-transformed prior to calculations of standard error (SE) .

RESULTS

[14C]Glu and [14C]malate fate in the free-living fungus P. involutus

When 6- and 15-d-old P. involutus colonies were fed with either [14C]Glu or [14C]malate, most of the 14C was recovered in the methanol-soluble fraction whereas only 3–11% was recovered in the methanol-insoluble fraction of mycelium (Fig. 1). The mycelium-associated radioactivity in the FAA fraction represented as much as 78 to 98% of the radioactivity in the soluble fraction in 6-d-old colonies (Fig. 1a & c). This proportion decreased slightly as the mycelium aged with a concomitant increase of the radioactivity in the neutral and FOA fractions (Fig. 1b & d). The radioactivity recovered in the FOA fraction represented only 2–5% (Fig. 1a & b) and 4–17% (Fig. 1c & d) of that of the soluble fraction under Glu and malate feeding, respectively. It is worth noting that very little incorporation was detected in the neutral fraction (0·5–3·5%). The major labelled amino acids when mycelium was exposed to [14C]Glu or [14C]malate were Gln, Glu, Asp, Ala and Arg (Tables 1 & 2). Gln was the major 14C sink in 6-d-old mycelium where it incorporated 58·7 and 37·7% of the radioactivity of the FAA fraction under Glu (Table 1) and malate (Table 2) feeding, respectively. These proportions increased as the mycelium aged. The proportion of [14C]Ala and [14C]Asp was higher under malate feeding and decreased strongly in 15-d-old mycelium, compared with 6-d-old mycelium. The radioactivity recovered in the neutral and FOA fractions was too low to detect the incorporation in individual compounds.

Figure 1.

Distribution of radioactivity derived from metabolism of [14C]Glu and [14C]malate in the free-living fungus P. involutus. Six-day-old (a and c) and 15-d-old (b and d) fungal discs were incubated in a solution containing MMN medium supplemented with 0·05 μCi L-[U-14C]Glu (specific activity 250 μCi μmol−1) (a, b) or with 0·2 μCi L-[U-14C]malate (specific activity 51 μCi μmol−1) (c and d) for 30 and 120 min. Results are expressed as radioactivity in the insoluble fraction (black bars), in the FAA fraction (white bars), in the FOA fraction (light grey bars) and in the neutral compound fraction (dark grey bars). Each value is the mean of three replicates. Vertical bars indicate SE.

Table 1.  Distribution of radioactivity in amino acids derived from metabolism of [14C]Glu in mycorrhizal or non-mycorrhizal birch roots and in the free-living fungus P. involutus. Roots and fungal discs were incubated in a solution containing MMN medium supplemented with 0·05 μCi L-[U-14C]Glu (specific activity 250 μCi μmol−1) for 120 min. Results are expressed as the percentage of total label in the FAA fraction. Each value is the mean ± SE of three replicates
 P. involutusMycorrhizal rootsNon-mycorrhizal roots
 6-d-old15-d-old6-d-old15-d-old60-d-old6-d-old15-d-old60-d-old
% total 14C
Ala 2·0 ± 1·2<2 2·5 ± 0·8<2 9·9 ± 0·3 9·7 ± 5·3 2·4 ± 4·9 3·0 ± 1·2
Arg 2·3 ± 1·40<20<2 4·6 ± 2·2 4·9 ± 1·6 4·9 ± 3·9
Asp 8·1 ± 2·4<2 2·3 ± 0·1<2<2 5·5 ± 1·0 9·1 ± 1·8 8·2 ± 0·7
Cit00 4·8 ± 0·4<2 8·8 ± 1·333·2 ± 7·837·1 ± 0·827·0 ± 8·9
Gln58·7 ± 5·978·9 ± 0·460·5 ± 1·077·9 ± 0·156·1 ± 2·116·6 ± 3·022·4 ± 5·212·2 ± 4·4
Glu27·4 ± 4·217·8 ± 0·419·2 ± 1·416·9 ± 0·720·2 ± 2·812·8 ± 0·919·9 ± 1·641·6 ± 1·3
Table 2.  Distribution of radioactivity in amino acids derived from metabolism of [14C]malate in mycorrhizal or non-mycorrhizal birch roots and in the free-living fungus P. involutus. Roots and fungal discs were incubated in a solution containing MMN medium supplemented with 0·2 μCi L-[U-14C]malate (specific activity 51 μCi μmol−1) for 120 min. Results are expressed as the percentage of total label in the FAA fraction. Each value is the mean ± SE of three replicates
 P. involutusMycorrhizal rootsNon-mycorrhizal roots
 6-d-old15-d-old6-d-old15-d-old60-d-old6-d-old15-d-old60-d-old
  1. ND, not determined.

% total 14C
Ala11·6 ± 0·7 2·7 ± 0·9 4·4 ± 1·0 3·7 ± 0·546·2 ± 0·611·3 ± 5·87·8 ± 3·7ND
Arg<2<2<20<211·9 ± 5·66·5 ± 2·7ND
Asp22·6 ± 1·9 4·2 ± 1·1 3·8 ± 1·2<2 2·8 ± 1·510·9 ± 3·212·2 ± 2·2ND
Cit00 7·0 ± 1·4<2 7·2 ± 0·625·9 ± 3·836·3 ± 0·4ND
Gln37·7 ± 2·173·0 ± 1·058·3 ± 2·473·8 ± 2·725·0 ± 2·015·1 ± 4·417·7 ± 2·6ND
Glu22·0 ± 2·118·2 ± 0·718·2 ± 2·317·1 ± 0·518·4 ± 2·110·0 ± 1·515·3 ± 3·0ND

[14C]Glu and [14C]malate fate in non-mycorrhizal birch roots

The proportion of 14C recovered in the insoluble fraction represented 15 and 17% of the total radioactivity when 6-d-old non-mycorrhizal birch roots were fed with [14C]Glu and [14C]malate for 30 min, respectively (Fig. 2a & d). This proportion reached 26% when birch roots were fed with [14C]Glu for 120 min. It also increased to 56 and 60% when 60-d-old birch roots were incubated for 120 min, with [14C]Glu and [14C]malate, respectively (Fig. 2c & f). The highest proportion of 14C was recovered in the FAA fraction when non-mycorrhizal birch roots were fed for 30 min with [14C]Glu. This fraction accounted for 99 (Fig. 2a), 81 (Fig. 2b) and 83% (Fig. 2c) of the soluble 14C fraction in 6-, 15- and 60-d-old roots, respectively. These proportions decreased to approximately 70% when roots were incubated for 120 min. Higher proportions of 14C were recovered in the FOA fraction when malate was the 14C source, varying with the incubation time from 38 to 50% in 6-d-old roots (Fig. 2d) and from 47 to 50% in 15-d-old roots (Fig. 2e). These proportions slightly decreased as the roots aged (Fig. 2f). In all cases, the neutral fraction represented less than 8% of the soluble radioactivity. Whatever the 14C source, Cit was the most highly labelled amino acid in the time course experiment (Tables 1 & 2), except in 60-d-old roots where Glu was the major radioactive amino acid under Glu feeding. As already found for the free-living fungus, the proportion of [14C]Ala and [14C]Asp was higher under malate feeding. Arg was also labelled with higher rates under [14C]malate feeding.

Figure 2.

Distribution of radioactivity derived from metabolism of [14C]Glu and [14C]malate in non-mycorrhizal birch roots. Six-day-old (a and d), 15-d-old (b and e) and 60-d-old (c and f) roots were incubated in a solution containing MMN medium supplemented with 0·05 μCi L-[U-14C]Glu (specific activity 250 μCi μmol−1) (a–c) or with 0·2 μCi L-[U-14C]malate (specific activity 51 μCi μmol−1) (d–f) for 30 and 120 min. Results are expressed as radioactivity in the insoluble fraction (black bars), in the FAA fraction (white bars), in the FOA fraction (light grey bars) and in the neutral compound fraction (dark grey bars). Each value is the mean of three replicates. Vertical bars indicate SE.

[14C]Glu and [14C]malate fate in mycorrhizal birch roots

In 6-d-old mycorrhizal roots, the proportions of 14C recovered in the insoluble and soluble fractions were very similar to those of non-mycorrhizal roots, whatever the incubation time and the 14C source (Fig. 3a & d). However, compared with non-mycorrhizal roots (Fig. 2), the proportions of 14C recovered in the insoluble fraction of mycorrhizal roots were much lower in 15- and 60-d-old roots (Fig. 3b, c, e & f) but remained higher than that found in the free-living fungus (Fig. 1). The proportion of 14C recovered in the different components of the soluble fraction remained remarkably constant over the course of the experiment. The FAA and FOA fractions, whatever the age of the roots, represented, respectively, 85 and 10% in [14C]Glu-fed roots (Fig. 3a–c) and these proportions varied only slightly with increasing incubation times, whereas significant variations were found with non-mycorrhizal roots (Fig. 2). Similar proportions were found when 6- and 60-d-old birch roots were fed with [14C]malate (Fig. 3d & f) whereas the proportion of 14C recovered in the FOA fraction was higher in 15-d-old roots (Fig. 3e). Additionally, whatever the 14C source, the incubation time and the age of the root, the 14C recovered in the neutral fraction, representing up to 8% of the soluble fraction, was higher than that in non-mycorrhizal roots. Gln was the major 14C metabolite recovered in the FAA fraction for [14C]Glu-fed mycorrhizal roots, and accounted for up to 60 and 78% of the radioactivity in the FAA fraction in 6- and 15-d-old mycorrhizal roots, respectively (Table 1). Similar values were observed for [14C]malate-fed roots (Table 2). The proportions of [14C]Cit, [14C]Ala and [14C]Asp were much lower when compared with non-mycorrhizal roots.

Figure 3.

Distribution of radioactivity derived from metabolism of [14C]Glu and [14C]malate in P. involutus mycorrhizal birch roots (for details, see legend to Fig. 2). Results are expressed as radioactivity in the insoluble fraction (black bars), in the FAA fraction (white bars), in the FOA fraction (light grey bars) and in the neutral compound fraction (dark grey bars). Each value is the mean of three replicates. Vertical bars indicate SE.

[14C] accumulation and respiration in birch roots and in P. involutus

The total amount of 14C recovered from either [14C]Glu or [14C]malate was much higher in mycorrhizal roots (Fig. 3) compared with non-mycorrhizal roots (Fig. 2), whatever the age of the tissues. However, the total amount of 14C was much lower in 60-d-old mycorrhizal roots, compared with younger roots (Fig. 3), which was not the case for non-mycorrhizal roots. The respiratory loss of carbon from [14C]Glu and [14C]malate, estimated during a 30 min incubation period, accounted, respectively, for 5·4 and 7% of the total absorbed 14C in 15-d-old mycelium (data not shown). The 14CO2 evolution from Glu in non-mycorrhizal roots (which varied according to root age from 36 to 67% of the total absorbed 14C) was much greater than that in mycorrhizal roots (which varied according to root age from 14 to 41% of the total absorbed 14C). Conversion of [14C]malate to 14CO2 was appreciable and comparable to that found when [14C]Glu was supplied for the same period (data not shown).

DISCUSSION

We have previously shown by measuring enzyme capacities and metabolite pools that mycorrhization causes a re-arrangement of the main metabolic pathways at the very early stages following contact between the two partners (Blaudez et al. 1998), which is correlated with the observed structural changes (Brun et al. 1995). The relevance and the limitations of the axenic system used has been discussed previously (Blaudez et al. 1998).

In the free living fungus, 14C from Glu appeared in amino acids very rapidly (Fig. 1), especially in Gln (Table 1). This is in good agreement with our previous results where P. involutus was fed with [14C]Ala or [14C]Glu (Chalot et al. 1994a,b). Most of the labelling was found in the FAA fraction. More surprisingly, when colonies were fed with [14C]malate, a very similar pattern was found, with an intense labelling of the FAA pool, except that the 14C recovered in the FOA fraction was slightly higher. We can speculate that this is certainly correlated with high activities of transaminases for the transfer of carbon skeletons from malate and other organic acids into amino acids. Furthermore, comparison of 14C distribution from the two 14C sources indicated a higher flow of carbon through the citric acid cycle since Ala and Asp were much more intensely labelled with [14C]malate. These results are well supported by the fact that amino acids are a major sink for 14C skeletons in the free-living symbiont (Chalot et al. 1994a,b; Chalot & Brun 1998).

The distribution of 14C from exogenously supplied Glu or malate in non-mycorrhizal roots differed from that in mycorrhizal roots (Fig. 4). The amount of 14C recovered in the insoluble fraction of non-mycorrhizal roots was much higher (Fig. 2) which indicates that the root tissues probably allocate more carbon to the synthesis of insoluble material compared with the mycorrhizal roots (Fig. 4). This is in good agreement with the data from Rygiewicz & Andersen (1994) who concluded that below-ground mycorrhizal plants shifted the allocation of carbon to pools that readily turned over. Methanol-insoluble material consists of proteins, lipids and complex sugars such as starch and glycogen. Our results are therefore well supported by the decrease in starch grains observed during the ontogenesis of P. involutusB. pendula ectomycorrhizas (Jordy et al. 1998). The higher proportion of 14C recovered in the FAA is well supported by higher glutamine synthetase (GS), malate dehydrogenase (MDH), aminotransferase activities in mycorrhizal roots, compared with non-mycorrhizal roots (Blaudez et al. 1998). Furthermore, the greater proportion of 14C respiratory loss from [14C]Glu and [14C]malate by non-mycorrhizal birch roots confirms previous results showing that the presence of the fungus reduces the rate of the basal respiration of the plant tissues on which it grows (Marshall & Perry 1987).

Figure 4.

Possible pathways for metabolism of [14C]glutamate and [14C]malate by non-mycorrhizal and mycorrhizal birch roots. AlAT, alanine aminotransferase; AAT, aspartate aminotransferase; GS, glutamine synthetase; MDH, malate dehydrogenase. 1Insoluble carbohydrates represent the major compounds of the methanol-insoluble fraction.

The distribution of the 14C within the different components of the soluble fraction changed strikingly. Cit was the most highly labelled component in non-mycorrhizal birch roots (Fig. 4), regardless of the 14C source whereas the proportion of 14C incorporated into Gln was three- to four-fold lower than in mycorrhizal roots (Tables 1 & 2). Whatever the developmental stage, we previously found that Gln was the major component of the FAA pool in mycorrhizal or non-mycorrhizal birch roots (Blaudez et al. 1998). Cit was also found to be present in birch roots in significant amounts, although at a two- to three-fold lower amount than that of Gln. These large amounts of 14C in Cit in non-mycorrhizal birch roots were correlated with a high 14C specific activity in Cit, indicating that Cit synthesis is probably a major metabolic pathway for the newly absorbed 14C organic compounds in non-mycorrhizal birch roots. On the contrary, Gln was the major sink for 14C in mycorrhizal roots, whatever the 14C source, and the proportion of [14C]Cit was much lower when compared with non-mycorrhizal roots (Fig. 4). Recently, Martin et al. (1998) found that the assimilation of [13C]glucose in Eucalypt–Pisolithus tinctorius ectomycorrhizal roots resulted in the production of large amounts of labelled polyols, trehalose, Gln and Ala, whereas non-mycorrhizal eucalypt roots mainly synthesized Suc and Gln. The mycorrhizal association used in the present work therefore showed specific features with Gln being the major 14C sink for exogenously supplied organic carbon, when birch roots were mycorrhizal with P. involutus.

A procedure using detached roots has been successfully used by Chapin, Moilanen & Kielland (1993), Kielland (1994) and Wallenda & Read (1999) to characterize uptake of nutrients by both non-mycorrhizal and mycorrhizal plants. In particular, data obtained by Kielland (1994) demonstrated that ectomycorrhizal species had higher amino acid absorption capacities than non-mycorrhizal species, which is in good agreement with the present data. Absorption capacities for Glu by mycorrhizal B. pendula roots were of the same order of magnitude as those obtained with detached mycorrhizal roots from Betula nana (Kielland 1994). Transport of amino acids was investigated in the mycorrhizal fungi P. involutus (Chalot et al. 1996) and Amanita muscaria (Nehls et al. 1999) which demonstrated their ability to take up a variety of amino acids.

In addition, we found that [14C]malate absorption capacities of mycorrhizal roots were much greater than that of non-mycorrhizal roots, and the beneficial effect of inoculation on malate utilization was even more marked than that for Glu. There has been strong interest in the role of organic acids in plant nutrition through their capacity to improve availability of some macro- and micronutrients (Marschner 1995). Malate, together with oxalate, fumarate and citrate, are the most abundant organic acids in the root exudates of trees (Smith 1976; Jones 1998). The present work demonstrated that organic acids could be utilized by the fungal cells.

Non-mycorrhizal roots were able to utilize amino acid and organic acid, although at much lower rates. These experiments were carried out on axenic seedlings to preclude microbial degradation of the substrates, and therefore the ecological significance of amino acid and organic acid utilization by non-mycorrhizal roots is still unclear. However studies with non-mycorrhizal herbaceous plants support the possibility of direct uptake of amino acids by root cells (Schobert & Komor 1987; Chapin et al. 1993). Roots of castor bean seedlings grown under natural or axenic conditions successfully compete with micro-organisms for free amino acids in the soil (Schobert, Köckenberger & Komor 1988). In tundra ecosystems where plant uptake of organic nitrogen is observed (Kielland 1994), competition between plant and microbes has also been suggested (Kaye & Hart 1997). Such research is necessary to evaluate the potential role of the ectomycorrhizal mycelium in the acquisition of organic carbon compounds in forest ecosystems. It has already been suggested that plants can directly control levels of C within the rhizosphere (Jones & Darrah 1996) and that uptake of amino acids may also be important for the N nutrition of plants (Jones & Darrah 1994).

Whatever the 14C source, the fungal partner always had higher capacities for utilizing the organic substrates when compared with birch roots. Based on ergosterol content, it was found that the proportion of fungal biomass amounted to 25 and 45% in 6- and 15-d-old roots, respectively (Blaudez et al. 1998). Taking these two assessments into account, it was expected that 15-d-old mycorrhizal roots had higher uptake capacities than 6-d-old roots, with regards to the proportion of fungal biomass. However, this was not the case since 6- and 15-d-old mycorrhizal roots had similar absorption capacities (Fig. 3). Furthermore, 60-d-old non-mycorrhizal roots had absorption capacities that were comparable to the younger non-mycorrhizal roots, whereas 60-d-old mycorrhizal roots had lower absorption capacities than the youngest mycorrhizal roots, although still higher than the corresponding non-mycorrhizal roots. However, interpretation must be taken with caution since 60-d-old roots were grown under non-sterile conditions. These results further support the conclusion that the early stages of mycorrhiza formation induced higher requirements for C skeletons as suggested earlier (Blaudez et al. 1998). Cairney & Alexander (1992) demonstrated that the ability of ectomycorrhizas to attract photosynthates was greatest soon after their formation and that there was a progressive reduction in the amount of the photosynthates translocated to them as they aged. Spruce mycorrhizas were found to have a reduced ability to respire glucose associated with ageing (Al-Abras et al. 1988). The present data extend these findings to the exogenously supplied C from amino acids and organic acids.

Taken together with previous results (Blaudez et al. 1998; Jordy et al. 1998; Martin et al. 1998), these data indicate that mycorrhiza formation has a profound alteration of the metabolic fate of exogenously supplied organic carbon. We have demonstrated that exogenously supplied organic carbon may serve as a source of C skeletons for Gln synthesis in mycorrhizal roots, whereas Cit was the major sink in non-mycorrhizal roots. The present data unequivocally demonstrate that malate was actively metabolized by ectomycorrhizal fungi and ectomycorrhizal birch roots, and that it may serve as C skeletons for respiration and amino acid synthesis. Taken together with the relatively high concentrations of organic acids in forest soils, especially in organic horizons (van Hees et al. 1996) where most of the ectomycorrhizal roots are located (Smith & Read 1997), our results strongly suggest that organic acids may effectively participate in the C nutrition of mycorrhizas. However, whether this additional C supply will contribute to the overall C status of the whole tree awaits further investigations.

Ancillary