Changes in carbon allocation and expression of carbon transporter genes in Betula pendula Roth. colonized by the ectomycorrhizal fungus Paxillus involutus (Batsch) Fr.

Authors


Correspondence: DrDerekWright. Fax: 0114 222 0002; E-mail: d.p.wright@sheffield.ac.uk

ABSTRACT

Comparative analyses of aspects of the carbon (C) physiology and the expression of C transporter genes in birch (Betula pendula Roth.) colonized by the ectomycorrhizal fungus Paxillus involutus (Batsch) Fr. were performed using mycorrhizal (M) and non-mycorrhizal (NM) plants of similar foliar nutrient status. After six months of growth, the biomass of M plants was significantly lower than that of NM plants. Diurnal C budgets of both sets of plants revealed that M plants exhibited higher rates of photosynthesis and root respiration expressed per unit dry weight. However, the diurnal net C gain of M and NM plants remained similar. Ectomycorrhizal roots contained higher soluble carbohydrate pools and increased activity of cell wall invertase, suggesting that additional C was allocated to these roots and their ectomycorrhizal fungi consistent with an increased sink demand for C due to the presence of the mycobiont. In M roots, the expression of two hexose and one sucrose transporter genes of birch were reduced to less than one-third of the expression level observed in NM roots. Analysis using a probe against the birch ribosomal internal transcribed spacer region revealed that M roots contained 22% less plant RNA than NM roots. As the expression of birch hexose and sucrose transporter genes was reduced to a much greater extent, this suggests that these specific genes were down-regulated in response to alterations in C metabolism within M roots.

INTRODUCTION

Most trees of the temperate and boreal forests form ectomycorrhizas. Central to the success of these symbioses is the exchange of nutrients between the plant and fungal partners. The fungus gains carbon (C) from the plant while plant nutrient uptake is mediated via the fungus ( Smith & Read 1997). Fungal colonization of the root system can substantially alter the pattern of carbon allocation within the plant partner ( Reid, Kidd & Ekwebelam 1983; Cairney, Ashford & Allaway 1989; Durall, Jones & Tinker 1994; Rygiewicz & Andersen 1994). It has been hypothesized that the presence of a fungal sink may stimulate the rate of C assimilation so that there is little or no ‘cost’ of being mycorrhizal imposed on the C economy of the plant ( Fitter 1991; Tinker, Durall & Jones 1994). However, evidence in support of this hypothesis in nutritionally matched plants remains limited.

In studies where nutritionally matched mycorrhizal (M) and non-mycorrhizal (NM) plants were used, some researchers ( Dosskey, Linderman & Boersma 1990; Rousseau & Reid 1990) have reported a stimulation in the rate of photosynthesis following ectomycorrhizal colonization, whilst others ( Eltrop & Marschner 1996a) have observed no effect. The effects, when observed, were generally smaller than those seen in arbuscular mycorrhizal associations ( Johnson 1984; Brown & Bethlenfalvay 1988; Wright, Scholes & Read 1998a).

Recently, we showed that increased allocation of C to the roots of clover plants colonized by arbuscular mycorrhizal fungi was associated with a stimulation in the activities of the sucrolytic enzymes invertase and sucrose synthase ( Wright, Read & Scholes 1998b). We suggested that such increases in enzyme activities may provide a mechanism enabling increased allocation of C both to the M root system and the fungal symbiont. However, no increases in the activities of these enzymes in response to ectomycorrhizal colonization have been observed ( Salzer & Hager 1991, 1993a, b; Schaeffer et al. 1995 ).

The form of carbon transferred to the fungus from the plant and the molecular mechanisms which mediate this exchange remain largely unknown. However, current conceptual models favour the transfer of hexoses, preferentially glucose, formed as a result of the action of acid invertases upon sucrose exported from the root cortical cells into the apoplast ( Smith & Smith 1990). In ectomycorrhizas, it is assumed that bi-directional exchange of C and mineral nutrients occurs at the same plant/fungus interface within the Hartig net ( Smith & Smith 1990). With the exception of the recent characterization of a monosaccharide transporter from the fungus Amanita muscaria ( Nehls et al. 1998 ), there are no reports of sucrose or hexose transporters of plant or fungal origin playing what must be a fundamental role in the movement of nutrients between the symbionts in ectomycorrhizal associations. An understanding of the relationship between the gross movement of C, and the mechanisms which may mediate this, within the plant partner and the expression of the genes controlling the transfer of nutrients between the symbionts is of fundamental importance to our further understanding of the association.

This study aims to integrate our knowledge of the alterations in C allocation within the plant partner with the molecular mechanisms underlying C transport between the symbionts in ectomycorrhizas. Birch (Betula pendula Roth.) and the ectomycorrhizal fungus Paxillus involutus (Batsch) Fr. were chosen as the model system. Using nutritionally matched M and NM birch plants, we have investigated the hypothesis that the presence of the fungal symbiont fundamentally influences the pattern of assimilation and allocation of C in the plant and the expression of sucrose and hexose transporter genes within ectomycorrhizal roots. Gas exchange techniques were used to determine diurnal C budgets of M and NM plants. The amounts of soluble carbohydrates and starch in the leaves and roots were measured and the roles of invertase and sucrose synthase in regulating the sink strength of M compared to NM roots were investigated. Gene fragments of sucrose and hexose transporters of birch were isolated by PCR and their expression within M and NM root systems compared.

MATERIALS AND METHODS

Biological material, media and growth conditions

Stocks of Paxillus involutus (Batsch) Fr. were maintained aseptically on agar (15 kg m−3) containing modified Melin–Norkrans (MMN) medium. The complete MMN medium contained: (all as mol m−3) glucose 5·55, KH2PO4 3·67, (NH4)2HPO4 1·89, MgSO4·7H2O 0·61, NaCl 0·43, CaCl2 0·34, FeCl3·6H2O 0·025, and 0·3 mmol m−3 thiamine hydrochloride. The medium was adjusted to pH 5·7. Aseptic liquid cultures of P. involutus were also grown in MMN medium but with the glucose concentration increased to 55 mol m−3 and pH adjusted to 4·8. These cultures were maintained in darkness at 20 °C in an orbital shaker (Gallenkamp, UK).

An ectomycorrhizal association between birch (Betula pendula Roth.) and P. involutus was synthesized according to the method of Brun et al. (1995) . Seeds of birch were surface-sterilized in H2O2 (30 vol, Fisons Scientific, Loughborough, UK) for 25 min, washed with one dm3 of sterile distilled water then transferred aseptically to water agar plates (8 kg m−3) until germination occurred. Seedlings were transferred to 14-day-old colonies of P. involutus growing on sheets of cellophane (BCL Cellophane, Bridgewater, UK) placed over MMN agar (8 kg m−3) containing one-tenth of the standard concentration (see above) of both N and P. To produce NM plants, seedlings were transferred to a parallel series of cellophane-covered MMN agar plates without P. involutus. The plates were maintained in a Fisons Fitotron growth chamber at 20°C with a 16 h photoperiod (300 μmol m−2 s−1 irradiance). After 6 weeks, by which time inoculated plates supported heavily mycorrhizal seedlings, both M and NM plants were transferred aseptically to conical flasks containing 10 g acid-washed perlite and 35 cm−3 of 40% Rorisons nutrient solution ( Hewitt 1966). After a further 8 weeks of growth, when an extensive extra-matrical mycelium had developed in the perlite surrounding M birch roots, the roots of M and NM plants were harvested, their roots frozen in liquid N2 and stored at – 80°C prior to RNA extraction. The leaves of M and NM plants were combined, frozen in liquid N2 and stored at – 80°C until use for DNA extractions.

At this time, M and NM birch plants were also transferred to cylindrical pots (20 cm high, 3·5 cm in diameter) filled with acid-washed perlite. Both sets of plants received 40% Rorisons solution with the aim of producing M and NM plants of similar biomass and foliar N and P content. Plants were otherwise irrigated with distilled water. After a further 11 weeks growth, diurnal CO2 exchange was determined from the shoots and from the root environment for both sets of plants. The plants were then harvested, 27 weeks after germination, and an analysis of growth, N and P content, carbohydrate content, root invertase and sucrose synthase activities and fungal colonization of M roots was conducted.

Diurnal CO2 exchange from the shoots and roots of M and NM birch plants

To measure diurnal CO2 exchange from the shoot and from the root environment, the plants were enclosed in sealed plant chambers similar to those described by Wright et al. (1998b) . The plants were grown in cylindrical plastic pots (see above) which were placed into root compartments (20 cm high, 7·5 cm in diameter) and the shoot was enclosed in a clear perspex compartment (20 cm high, 10 cm in diameter). Diurnal CO2 exchange from the shoot and from the root environments of four M and four NM plants was measured simultaneously using the apparatus described by Wright et al. (1998b) . The CO2 concentration of the air supplied to the plant chambers was 380 cm3 m−3. Plants were sealed into the plant chambers and allowed to equilibrate in the growth cabinet for at least 6 h prior to initiation of data recording, and measurements were taken over the subsequent 36 h period. Conditions in the growth cabinet were as described above. The rates of CO2 exchange of whole shoot and of the root environment were calculated as μmol CO2 g−1 DW s−1. Using these data, the daily carbon gain (μmol CO2 plant−1 day−1 or μmol CO2 g−1 DW day−1) of M and NM plants was determined.

Analyses of biomass accumulation, soluble sugars, starch and the activity of sucrolytic enzymes

After the diurnal CO2 measurements had been completed, plants were removed from the CO2 exchange chambers and allowed to equilibrate in the growth cabinet. In the growth cabinet, two discs (0·78 cm2) were cut from the upper fully expanded leaves of each M and NM plant and rapidly frozen in liquid N2 for foliar carbohydrate and starch analyses. The remaining leaf material was harvested, its area determined using a Δ-T Mk 2 leaf area meter (Delta-T Devices, Cambridge, UK) and the leaf and stem material dried separately at 70 °C until constant weight was achieved. The root systems of M and NM plants were carefully removed from the cylindrical pots, separated from the perlite, and the total fresh weight determined. Sub-samples for the analysis of carbohydrates and starch, invertase and sucrose synthase activities and chitin were rapidly weighed and stored in liquid N2. The remaining root material was dried at 70 °C until constant weight was achieved. The perlite which supported the growth of each plant was stored at – 80 °C for quantification of the extra-matrical mycelium. The dry weight of the leaf discs reserved for carbohydrate analysis was calculated using the dry weight : leaf area ratio of each replicate, thus allowing the total leaf dry weight of each plant to be determined. Similarly, the dry weight of the root sub-samples was calculated using the fresh weight : dry weight ratio of each plant and the total root dry weight determined.

Leaf, root and mycelial extracts required for the analysis of soluble sugars and starch as well as root and mycelial extracts for the analysis of invertase and sucrose synthase activities were obtained from M and NM birch plants and pure cultures of P. involutus using the protocols described in Wright et al. (1998b) . The concentrations of P and N in the leaves, stems and roots of M and NM birch plants were determined using the method described by Wright et al. (1998a) .

Colonization of M and NM birch roots and quantification of the amount of extra-matrical mycelium

The extent of fungal colonization and the amounts of extra-matrical mycelium surrounding M and NM birch roots were determined using a modification of the chitin assay described by Vignon et al. (1986) . Samples were acid-hydrolysed in 4 cm−3 6 kmol m−3 HCl for 16 h at 80 °C. The pH of a 50 mm−3 aliquot of the acid hydrolysate was adjusted by the addition of 0·25 cm−3 1·25 kmol m−3 sodium acetate and the glucosamine residues assayed colorimetrically. Following the addition of 0·3 cm−3 50 kg m−3 KHSO4 and 0·3 cm−3 50 kg m−3 NaNO2, the reaction mix was shaken vigorously for 30 min, after which 0·3 cm−3 of 125 kg m−3 NH4SO3NH2 was added. The reaction mix was shaken for a further 5 min, then 0·3 cm−3 5 kg m−3 3-methyl-2-benzothiazolinone hydrazone hydrochloride (MBTH) was added. The mixture was boiled for 3 min, and, after cooling to room temperature, 0·3 cm−3 5 kg m−3 FeCl3·6H2O was added. The colour was allowed to develop for 30 min and the spectrum (400–800 nm, glucosamine peak 650 nm) of each sample was recorded. Samples of P. involutus (1–10 mg DW mycelium), grown on cellophane-covered MMN agar, and M and NM birch roots (20 mg DW approximately) were assayed as described above. The perlite growth medium of each plant replicate was also dried, ground to a fine powder, and three sub-samples of each (each 0·5–1 g) assayed to determine the amount of extra-matrical mycelium. As contaminating substances which interfered with the absorbance measurement (A650), were released during the extraction procedure, spectral deconvolution of each sample (400–800 nm) was carried out. The amount of glucosamine in each sample was then determined against a standard curve (0–15 μg of glucosamine).

Genomic DNA extractions

High-molecular-weight genomic DNA was prepared from the leaves of aseptically grown birch plants using a modification of the hexadecyltrimethylammonium bromide (CTAB) extraction procedure ( Reiter, Young & Scolnik 1992). Successful DNA extractions were achieved by increasing the ratio of the initial extraction buffer to plant material from 1 cm3 buffer : 1 g fresh weight tissue to 10 cm3 buffer : 1 g fresh weight tissue. The extraction buffer contained 100 mol m−3 Tris–HCl (pH 8·0), 1·4 kmol m−3 NaCl, 20 kg m−3 CTAB, 20 mol m−3 disodium-ethylenediaminotetraacetate (Na2-EDTA) and 10 kg m−3 polyvinylpyrrolidone (molecular weight 360 000). DNA was extracted from liquid cultures of P. involutus using a Nucleon Phytopure DNA extraction kit (Amersham Life Sciences, UK) following the manufacturer’s protocol. Persistent polysaccharide contamination of the extracted DNA was removed by centrifuging the DNA twice through a CsCl gradient (404 000 g, 20 h, 22 °C) according to Sambrook, Fritsch & Maniatis (1989).

PCR amplification and sequencing of genes encoding transport proteins and internal transcribed spacer (ITS) regions

Published gene sequences of sucrose transporters were obtained from GenBank, translated to give the predicted amino acid sequence and then aligned using PILEUP (Genetics Computer Group Inc.). Conserved peptide sequences were identified and degenerate oligonucleotide primers designed complementary to the appropriate DNA sequence. The two primers used to amplify birch sucrose transporters were as follows: SUC1 5′-CATAGCNGC(C/T)GGNGTNCA(A/G)TTCGG(G/T)TG-3′ and SUC2 5′-AANGG(A/G)AACCANGC(A/G)ATCCA(A/G)TT-3′ (N indicates all four bases). The primers used to amplify birch hexose transporters were ATH1 and ATH3 ( Weig et al. 1994 ). Amplification of the internal transcribed spacer (ITS) region of birch and P. involutus was carried out using the primers ITS1 and ITS4 ( White et al. 1990 ). Primer pairs were used in combination with birch or P. involutus template DNA in PCR reactions (40 cycles of 94 °C for 30 s, 50 °C for 30 s, 72 °C for 60 s; then 72 °C for 30 min; Techne PHC-3 thermal cycler, Techne (Cambridge) Ltd, Duxford, Cambridge (UK)) in a final volume of 50 mm−3. Taq DNA polymerase was added after heating the PCR mix to 94 °C to avoid non-specific amplification of gene products. Control amplifications were performed in the absence of DNA and also with DNA from Arabidopsis thaliana where the hexose and sucrose transporter sequences are known. After amplification, the samples were electrophoresed through agarose gels (10 kg m−3) and visualized using ethidium bromide. DNA fragments of the correct size were excised from the gel and isolated using a Wizard PCR preps DNA purification kit (Promega UK). Fragments were ligated into the plasmid vector pGEM-T Easy (Promega UK), transformed into Escherichia coli DH5alpha, and transformed colonies were identified using a β-galactosidase-based blue/white screen. Six colonies were selected from each transformation, plasmid DNA prepared (Wizardplus DNA miniprep kit, Promega UK) and digested with EcoRI to determine the insert size. Duplicate representatives were sequenced in both directions from the T7 and SP6 RNA polymerase binding sites of the pGEM-T Easy vector using an AmpliTaq dye terminator cycle DNA sequencing kit (ABI Prism, PE Applied Biosystems, Warrington, UK) and processed using an automated sequencer. Homology searches of sequence data with libraries of known gene sequences were carried out using the NCBI BLAST program ( Altschul et al. 1997 ).

Northern blot analysis of total RNA extracted from M and NM birch roots and P. involutus

Total RNA was extracted from the roots of aseptically grown M and NM birch plants and from P. involutus liquid cultures following the procedure of Loening (1969). Northern blots were prepared using standard procedures ( Sambrook et al. 1989 ). Ten micrograms of total RNA from M and NM birch roots and from P. involutus were electrophoresed through a denaturing formaldehyde agarose gel and the RNA transferred to Zetaprobe nylon membrane (BioRad, Hemel Hempstead, UK) by capillary blotting overnight. RNA was bound to the membrane by UV cross-linking.

Antisense radio-labelled riboprobes to birch sucrose and hexose transporters were synthesized from linearized plasmid vector in the presence of uridine 5-[α-32P] triphosphate (1·85 MBq at 370 MBq cm−3) from either the T7 or SP6 RNA polymerase site of the pGEM-T Easy vector (Promega, UK). After transcription, the riboprobe was incubated at 37 °C for 30 min with 3 units RQ1 RNase-free DNase to remove template DNA. The incubation was extracted by vortexing with an equal volume of 25 vol phenol : 24 vol chloroform : 1 vol isoamyl alcohol (pH 4·5), and centrifuged (14 000 g, 30 s). The aqueous phase was added to an equal volume of 24 vol chloroform : 1 vol isoamyl alcohol, vortexed and centrifuged (14 000 g, 30 s). The aqueous phase was transferred to a new Eppendorf tube, and, following the addition of 0·5 vol 7·5 kmol m−3 diethyl pyrocarbonate (DEPC)-treated ammonium acetate and 2 vol ethanol, the riboprobe was precipitated by incubation of the mixture on ice for 30 min. The riboprobe was pelleted by centrifugation (14 000 g, 10 min), the supernatant discarded and the pellet washed using 1 cm−3 7 vol ethanol : 3 vol DEPC-treated water. The ethanol was discarded and the pellet allowed to air dry before it was resuspended in 100 mm−3 DEPC-treated water. Hybridizations were performed in 1 vol formamide : 1 vol water containing 120 mol m−3 sodium phosphate (pH 7·2), 250 mol m−3 NaCl, 70 kg m−3 SDS, 100 kg m−3 PEG 8000 at 55 °C for 16 h. After hybridization, blots were washed under stringent conditions (2 × SSC, 1 kg m−3 SDS; 0·5 × SSC, 1 kg m−3 SDS; 0·2 × SSC, 1 kg m−3 SDS, each for 20 min at 22 °C) to remove non-specific binding of radio-labelled probe (2 × SSC contains 300 mol m−3 NaCl and 30 mol m−3 sodium acetate; pH 7·0). Blots were then RNase-treated (pre-block with 2 × SSC, sheared herring sperm DNA (≈ 1·5 kb, 0·4 mg cm−3), 22°C for 20 min, then 2 × SSC, herring sperm DNA (0·4 mg cm−3), RNase (1 μg cm−3), 22 °C for 30 min, then 2 × SSC, RNase (1 μg cm−3), 22 °C for 15 min, then washed in 0·2 × SSC, 1 kg m−3 SDS, 55 °C for 30 min) to ensure all non-specific binding had been removed ( Sambrook et al. 1989 ). Blots were visualized by exposure to Kodak Biomax film (Sigma UK) at – 80 °C for a maximum of 2 days.

Birch and P. involutus ITS DNA fragments were excised from the pGEM-T vector using EcoRI, electrophoresed through agarose gels (10 kg m−3) and isolated using a Wizard DNA clean-up kit (Promega UK). Excised DNA fragments were radio-labelled with deoxycytidine 5-[α-32P] triphosphate (1·85 MBq at 370 MBq cm−3) using a Megaprime DNA labelling kit (Amersham Life Sciences). Hybridizations were performed in 500 mol m−3 sodium phosphate (pH 7·2), 70 kg m−3 SDS at 65 °C for 16 h. The blots were washed at 65 °C in two changes (each 30 min) of 80 mol m−3 sodium phosphate (pH 7·2), 50 kg m−3 SDS followed by two changes of 80 mol m−3 sodium phosphate (pH 7·2), 10 kg m−3 SDS to remove non-specifically bound probe, then visualized by exposure to Kodak Biomax film for 10 min at room temperature. An InstantImager electronic autoradiography unit (Packard Instruments, Pangbourne, UK) was used to quantify the signal intensities resulting from the Northern analysis of blots of M and NM birch roots and P. involutus.

RESULTS

Mycorrhizal colonization of birch roots

Visual inspection of M root systems revealed that all root tips were colonized. However, in order to quantify the amount of fungal biomass in M roots, the amount of chitin was determined and compared to that in hyphae from pure cultures of P. involutus. The amount of chitin in the cell walls of pure cultures of P. involutus was 71·7 ± 1·5 μg glucosamine residues mg−1 DW mycelium. Using this value, the fraction of total root dry weight of M plants calculated to be of fungal origin was 0·114 ± 0·013 g ( Table 1). When expressed as a percentage of the total root dry weight of M plants, the percentage colonization of M birch roots was 43 ± 6%. Chitin was not detected in the roots of NM birch plants.

Table 1.  The total dry weight (g) and the dry weight of the leaves, stems, roots and extra-matrical mycelium (Paxillus involutus) of non-mycorrhizal birch plants and birch ectomycorrhizal with P. involutus
 Dry weight (g)
TissueNon-mycorrhizalMycorrhizal
  1. Extra-matrical mycelium

  2. 0·048 ± 0·005

  3. The proportion of the total root dry weight of mycorrhizal roots that was due to each symbiont was calculated using the percentage colonization of each replicate. Values are means ± SE (n = 4). One-way analysis of variance was performed on the data.*P < 0·05; **P < 0·01; ***P < 0·001.

Total0·848 ± 0·1430·564 ± 0·023*
Leaf0·304 ± 0·0390·152 ± 0·013**
Stem0·173 ± 0·0230·099 ± 0·008*
Root
Total0·371 ± 0·0950·265 ± 0·014
Birch0·371 ± 0·0950·151 ± 0·023***
Paxillus0·114 ± 0·013

The effect of mycorrhizal colonization on the biomass production and nutrient content of birch

Mineral nutrient analyses confirmed that the concentrations of N and P in the leaves, stems and roots of M plants were not significantly different from those of NM plants ( Table 2).

Table 2.  The amount of phosphorus and nitrogen in the leaves, stems and roots of non-mycorrhizal birch plants and birch plants ectomycorrhizal with Paxillus involutus
 Nutrient concentration (mg g−1 DW tissue)
 Phosphorus Nitrogen
TissueNon-mycorrhizalMycorrhizalNon-mycorrhizalMycorrhizal
  1. Root

  2. 1·04 ± 0·07

  3. 1·23 ± 0·10

  4. 12·83 ± 1·55

  5. 15·97 ± 1·92

  6. Values are means ± SE (n = 4). One-way ANOVA was performed on each data set. There were no significant differences.

Leaf1·58 ± 0·041·67 ± 0·0419·38 ± 1·0818·26 ± 1·12
Stem0·97 ± 0·050·94 ± 0·06 8·83 ± 1·37 7·92 ± 1·04

The total dry weight of M plants was 33% lower than that of NM plants ( Table 1). The dry weights of the leaves and stems of M birch plants were significantly lower than those of NM birch plants. The leaf area of M plants was also significantly lower than that of NM birch plants (38·1 ± 3·0 versus 63·0 ± 8·8 cm2 for M and NM plants, respectively) (P < 0·05). Although the total root dry weight of M compared to NM birch plants was not significantly different, the dry weight of the plant fraction of M roots was significantly lower compared to that of NM roots (P < 0·001) ( Table 1). On average, there was a total of 0·048 ± 0·005 g DW P. involutus extra-matrical mycelium associated with the roots of M birch plants ( Table 1). Extra-matrical mycelium therefore represented 30% of the total fungal biomass associated with M root systems but only 15% of the total below-ground plant and fungal biomass. No extra-matrical mycelium was detected in the perlite growth medium surrounding roots of NM plants.

Diurnal C budgets of M and NM birch plants

The net amount of CO2 assimilated, the net C gain of the shoot and the net whole-plant C gain of M plants were not significantly different from those of NM plants expressed on a per plant basis ( Table 3). Root system respiration of M and NM plants was not significantly different, but the shoot respiration of M plants was significantly lower than that of NM plants expressed on a per plant basis (P < 0·05) ( Table 3). As the dry weights of individual organs of M and NM plants were significantly different, the data were also expressed per unit dry weight. The net amount of CO2 assimilated per unit leaf dry weight of M birch was increased by 29% on average in comparison to that assimilated by NM shoots, and this was reflected in the larger mean net C gain of M compared to NM shoots ( Table 3). The amount of C respired per unit dry weight of the M root system was significantly increased compared with that respired by NM root systems (P < 0·01). However, the net whole-plant C gain of M and NM plants, expressed per unit dry weight, was not significantly different ( Table 3). The difference in the amount of C respired by M compared to NM root systems represented 15% of the net CO2 assimilated by M shoots.

Table 3.  Diurnal carbon budgets of non-mycorrhizal (NM) birch plants and birch plants ectomycorrhizal with Paxillus involutus (M)
 CO2 exchange
 μmol CO2 plant−1 day−1μmol CO2 g−1 DW day−1
 Non-mycorrhizalMycorrhizalNon-mycorrhizalMycorrhizal
  1. Carbon gain of whole plant (NM) and plant and fungus (M)

  2. 342 ± 94

  3. 113 ± 44

  4. 1394 ± 257

  5. 1325 ± 326

  6. Results are expressed either per unit dry weight (μmol CO2 g−1 DW day−1) or on a per plant basis (μmol CO2 plant−1 day−1). Values are means ± SE (n = 4). The data for each parameter for M and NM plants were compared by one-way ANOVA. *P < 0·05, **P < 0·01.

Net CO2 assimilation690 ± 158444 ± 282233 ± 3942870 ± 437
Dark shoot respiration123 ± 1879 ± 3*355 ± 21516 ± 95
Net carbon gain of shoot567 ± 141365 ± 311878 ± 3812354 ± 342
Root system respiration225 ± 67252 ± 43593 ± 331029 ± 136**

The amount of soluble carbohydrates and starch in M and NM birch plants

The amounts of sucrose, glucose and fructose or starch in the leaves of M plants were not significantly different compared to the leaves of NM plants ( Fig. 1). The amount of glucose and fructose in the root systems of M birch plants was similar to that in the root systems of NM plants ( Fig. 1). When the amount of sucrose in the roots of M plants, expressed per unit dry weight of the whole root (plant and fungus), was compared to that in NM roots, no difference was observed ( Fig. 1). As sucrose was not detected in pure cultures of P. involutus, when the amount of sucrose in M roots was re-expressed per unit dry weight of the plant fraction of the M root system the amount of sucrose was significantly increased compared to that in NM root systems (14·5 ± 1·3 versus 32·3 ± 8·3 μmol glucose equivalents g−1 DW root in NM and M root systems, respectively) (P < 0·05). The amount of starch in the root systems of M plants, whether expressed per unit dry weight of the whole root system (plant and fungus) or expressed per unit dry weight of the plant fraction of M root systems, was significantly increased compared to that in the roots of NM plants (P < 0·01) ( Fig. 1).

Figure 1.

The amounts of sucrose and starch and the sum of glucose and fructose in the leaves and roots of non-mycorrhizal birch plants (NM; □) and birch plant ectomycorrhizal with Paxillus involutus (M; ▪). Data represent the mean ± standard error (n = 5).

The activities of sucrolytic enzymes in the roots of M and NM birch plants

The activity of cell wall bound invertase, expressed per unit dry weight of the plant fraction of M root systems, was significantly higher in M roots compared to NM root systems (P < 0·01) ( Table 4). When the activity of cell wall invertase was re-expressed per unit dry weight of the whole M root system (plant and fungus), there was no significant difference between the activity in M and NM root systems, probably due to the dilution effect of the fungal biomass in M roots. The activity of soluble acid invertase in the roots of M plants, whether expressed per dry weight of the whole root system (plant and fungus) or expressed per unit dry weight of the plant fraction of M root systems, was not significantly different from the activity in NM root systems ( Table 4). No cytoplasmic invertase (pH optimum 7·5) or sucrose synthase activity could be detected in M or NM root systems. Invertase and sucrose synthase activities were not detected in the mycelium of pure cultures of P. involutus.

Table 4.  The activity (μmol glucose equivalent g−1 DW root min−1) of invertase enzymes from non-mycorrhizal birch roots and birch roots ectomycorrhizal with Paxillus involutus
 Cell wall boundSoluble acidic 
Tissue(pH 4·5)(pH 5·5)
  1. Plant and fungus

  2. 9·1 ± 1·0

  3. 1·5 ± 0·5

  4. Values for mycorrhizal roots were calculated either based on the dry weight of the plant fraction of M roots or on the total dry weight of the whole (plant and fungus) M root system. Values are means ± SE (n = 5). One-way ANOVA was performed on the data. **P < 0·01.

Non-mycorrhizal 9·1 ± 1·02·7 ± 1·1
Mycorrhizal
Plant fraction only17·0 ± 2·6**2·7 ± 0·6

Cloning birch sucrose and hexose transporters

Degenerate oligonucleotides corresponding to highly conserved sequences of known sucrose or hexose transporters were used in a polymerase chain reaction to amplify DNA fragments from high-molecular-weight birch DNA. Three DNA fragments were isolated. Using the primer pair SUC1/SUC2, a DNA fragment of 789 bp was successfully cloned and was named BpSUC1 (GenBank AF168771) ( Fig. 2). Sequence analysis indicated that BpSUC1 encoded 262 amino acids and contained no introns ( Fig. 2). BpSUC1 displayed closest homology to the sucrose carrier Scr1 from Ricinus communis ( Weig & Komor 1996; GenBank Z31561), showing 76% identity and 82% similarity at the amino acid level. Two DNA fragments were successfully isolated using the primer pair ATH1/ATH3. Both fragments were 246 bp in length, each encoding 82 amino acids and containing no introns ( Fig. 2). One fragment, named BpHEX1 (GenBank AF168772), exhibited closest homology to the hexose carrier HEX5 from Ricinus communis ( Weig et al. 1994 ; GenBank L08193), showing 78% identity and 83% similarity at the amino acid level. The second fragment, named BpHEX2 (GenBank AF168773), displayed closest homology to the hexose carrier HEX8 from Ricinus communis ( Weig et al. 1994 ; GenBank L08195), showing 85% identity and 89% similarity at the amino acid level. A BESTFIT alignment (Genetics Computer Group Inc.) of BpHEX1 against BpHEX2 showed that they were 56% identical at the nucleotide level and 53% identical at the amino acid level. No gene fragments of either sucrose or hexose transporters were isolated from DNA of cultured P. involutus using the SUC1/SUC2, ATH1/ATH3, T1/T2 ( Harrison 1996) or T1/ATH3 primer pairs, even under low-stringency PCR conditions.

Figure 2.

DNA and deduced amino acid sequences of birch sucrose (BpSUC1) and hexose (BpHEX1, BpHEX2) transporter genes. Underlined nucleotides represent the DNA sequences to which the primers hybridized during PCR amplification.

DNA fragments corresponding to the internal transcribed spacer (ITS) region between the 18S and 28S rDNA genes were successfully isolated from NM birch roots and cultured P. involutus. The fragment from NM birch roots was named BpITS1 and the fragment from P. involutus was named PiITS1.

The expression of sucrose and hexose transporter and ITS genes in M and NM birch roots and cultured P. involutus

The expressions of BpSUC1, BpHEX1 and BpHEX2 were examined by Northern blot analyses of total RNA from NM and M roots of birch and cultured P. involutus. Ethidium bromide staining of the RNA samples following electrophoresis showed two major bands corresponding to the 18S and 28S rRNAs ( Fig. 3). The intensity of the bands was similar, indicating that similar amounts of total RNA were loaded in each lane ( Fig. 3). Whereas RNA prepared from NM roots is exclusively of plant origin and that from cultured P. involutus is of fungal origin, RNA from M roots contained a mixture of the two. The ITS probes, BpITS1 (birch) and PiITS1 (P. involutus), prepared from each organism, were used to determine the relative amounts of plant and fungal RNA in the M roots. The plant and fungal ITS probes were completely specific ( Fig. 3). Quantification of BpITS1 showed that the amount of plant RNA in M roots had decreased to 78% of that observed in NM birch roots ( Table 5).

Figure 3.

Northern hybridizations showing the expression of birch sucrose (BpSUC1) and hexose (BpHEX1, BpHEX2) transporter genes (32P-labelled RNA probes) and internal transcribed spacer (ITS) regions of birch (BpITS1) and Paxillus involutus (PiITS1) rDNA genes (32P-labelled DNA probes) against total RNA extracted from non-mycorrhizal (NM) birch roots, birch roots ectomycorrhizal with P. involutus (M) and cultured P. involutus. Species-specific ITS probes were employed to determine the proportion of RNA of the plant and fungal partners present in M birch roots in comparison to NM birch roots or cultured P. involutus, respectively. Equal loading of total RNA (10 μg) in each lane of the gel was checked by ethidium bromide staining.

Table 5.  Percentage hybridization signal intensity of sucrose and hexose transporter and internal transcribed spacer (ITS) 32P-labelled probes to Northern blots of total RNA from non-mycorrhizal birch roots, birch roots ectomycorrhizal with Paxillus involutus and cultured P. involutus
 Non-mycorrhizalMycorrhizalP. involutus
  1. PiITS1

  2. 84

  3. 100

BpSUC110024
BpHEX110033
BpHEX210033
BpITS110078

Hybridization signals were obtained from riboprobes prepared from BpSUC1 (birch sucrose transporter), BpHEX1 and BpHEX2 (birch hexose transporters) ( Fig. 3). Both the sucrose and hexose transporters were expressed in NM roots and expression was much lower in M roots ( Fig. 3). Quantification of the signal intensities showed that expression of BpSUC1 in M roots was 24% of that in NM roots ( Table 5). Similarly, the expression of both BpHEX1 and BpHEX2 in M roots was 33% of that in NM roots ( Table 5). Although there was 22% less plant RNA in M compared to NM roots, the reductions in the signal intensities of BpSUC1, BpHEX1 and BpHEX2 were much greater than this, indicating that the expression of these genes had been down-regulated in M roots. In preliminary experiments using a DNA probe for BpSUC1, expression was observed in NM roots but was barely visible in M roots on autoradiographs after a 7-day exposure (data not shown). Although this represents the same pattern of expression as was observed using riboprobes, the intensity of the hybridization was much stronger using the latter, making the study of low-abundance gene expression easier. No cross-hybridization was observed between either BpSUC1, BpHEX1 or BpHEX2 and RNA prepared from P. involutus.

DISCUSSION

Analyses of diurnal C budgets revealed that the net amount of C assimilated per unit dry weight by M birch plants was approximately 29% higher than that assimilated by NM birch plants. However, more C was respired from the root systems of M than NM plants so that the net C gain of whole M plants was similar to that of nutritionally matched NM plants. Since the total biomass of the M plants was significantly lower than that of the NM plants, it can be concluded that during the growth of these plants any increases in net C assimilation of the M plant partner were not sufficient to fully meet the C requirements of both the plant and fungal partners. Under such circumstances, reduced growth of M plants would be inevitable.

Studies of nutrient-independent stimulation of C assimilation rates and growth of plants following ectomycorrhizal colonization have yielded different responses. In some plant–fungus partnerships, increases in assimilation rate relative to those seen in NM plants have been observed ( Dosskey et al. 1990 ; Rousseau & Reid 1990; Conjeaud, Scheromm & Mousain 1996), while in others no stimulation was reported ( Dosskey et al. 1990 ; Eltrop & Marschner 1996a, b). The biomass of such M plants was usually found to be the same as or less than that of their NM counterparts ( Dosskey et al. 1990 ; Conjeaud et al. 1996 ; Eltrop & Marschner 1996a, b).

The diurnal C budgets presented in this study indicate that M colonization resulted in increased allocation of C to the roots of M plants and that most of the additional C was used to support growth and metabolism of the mycobiont. Increased allocation of C to the roots of plants upon colonization by M fungi has been widely reported ( Reid et al. 1983 ; Durall et al. 1994 ; Rygiewicz & Andersen 1994; Ek 1997).

The difference in the biomass between M and NM birch plants reported here could not be accounted for simply by the biomass of the extra-matrical mycelium. The extra-matrical mycelium represented 30% of the DW of the total fungal biomass in M roots but only 15% of the total below-ground plant and fungal biomass. However, these values are in accordance with other studies which have shown that extra-matrical mycelium can account for 30–87% of total fungal biomass associated with M roots or 5–15% of the total below-ground plant and fungal biomass ( Colpaert, Van Assche & Luijtens 1992; Wallander & Nylund 1992; Ekblad et al. 1995 ).

The sizes of the pools of sucrose, starch, glucose and fructose in the leaves of M and NM plants were similar despite the fact that at the time of measurement the rate of photosynthesis of M plants was higher. This is additional evidence in support of an increased export of C to M root systems. The pools of glucose and fructose were also similar in M and NM roots. However, after a correction for the ‘dilution effect’ of the biomass of the fungal partner, the amounts of sucrose and starch were greater in the plant fraction of M roots in comparison to NM roots. In other studies using nutritionally matched M and NM plants, little effect of colonization upon C pools has been observed. For example, no differences in the soluble sugar or starch content of the needles or roots of M and NM Pinus pinaster ( Conjeaud et al. 1996 ) or Picea abies ( Eltrop & Marschner 1996a, b) were observed. Schaeffer et al. (1995) reported lower amounts of glucose and fructose in Picea abies colonized by either Amanita muscaria or Cenococcum geophilum. However, they observed no difference in the sucrose pools between M and NM plants after a correction for the ‘dilution effect’ of the fungal tissue in M roots was made.

Increased C allocation to M root systems has been attributed to an increased sink strength arising from colonization by the fungal symbiont. Rapid utilization of C and its conversion into fungal-specific compounds, e.g. trehalose and mannitol, as well as increased respiration of the M root system have been proposed as mechanisms by which sink strength may be increased ( Lewis & Harley 1965; Smith, Muscatine & Lewis 1969). Sucrolytic enzymes, invertase and sucrose synthase, have been implicated in the regulation of the sink strength of plant tissues ( Ho 1988; Sung, Xu & Black 1989; Wang et al. 1993 ; Sung et al. 1994 ; Zrenner et al. 1995 ). We could not detect invertase or sucrose synthase activity in the hyphae from pure cultures of P. involutus. Invertase activity was also not detected in mycelium of Amanita muscaria, Hebeloma crustuliniforme or Cenococcum geophilum ( Salzer & Hager 1991; Schaeffer et al. 1995 ). In the present study, a stimulation in the activity of cell wall invertase was observed after the ‘dilution effect’ of the presence of fungal biomass in M roots had been taken into account. This represents a further mechanism contributing to the increased sink strength arising from M colonization. In the only other study in which invertase activity has been measured, Schaeffer et al. (1995) showed that acid invertase was the major sucrolytic enzyme in roots of Picea abies but that its activity did not increase following infection by Amanita muscaria or Cenococcum geophilum. These findings for ectomycorrhizas contrast with the stimulation in the activities of cell wall and cytoplasmic invertases and sucrose synthase in response to arbuscular M colonization ( Wright et al. 1998b ).

A role for cell wall invertases in the provision of hexoses for uptake by the fungal partner has been postulated. The current model favours the transfer of hexoses formed as a result of the action of acid invertases upon sucrose exported from the root cortical cells into the apoplast ( Smith & Smith 1990). In ectomycorrhizas, it is assumed that bi-directional exchange of C and mineral nutrients occurs at the same plant/fungus interface within the Hartig net ( Smith & Smith 1990). However, the exact molecular form of C transferred to the fungus from the plant and the molecular mechanisms which mediate this exchange remain unknown. Hexose and phosphate transporter genes which may play a role in nutrient exchange between the plant and fungal partners of arbuscular M associations have been identified ( Harrison & van Buuren 1995; Harrison 1996; Liu et al. 1998 ), but with the exception of the recent characterization of a monosaccharide transporter from the ectomycorrhizal fungus Amanita muscaria ( Nehls et al. 1998 ), there are no reports of plant or fungal transporters playing what must be a fundamental role in the movement of nutrients between the symbionts in ectomycorrhizal associations.

In this study we have isolated fragments of birch hexose and sucrose transporter genes. The sucrose transporter gene fragment, BpSUC1, showed highest homology to the sucrose carrier Scr1 from Ricinus communis ( Weig & Komor 1996). Both hexose transporter gene fragments, BpHEX1 and BpHEX2, showed highest homology to different hexose carriers from Ricinus communis ( Weig et al. 1994 ).

It is likely that the proportions of plant and fungal biomass and hence the plant/fungal RNA ratio will be constantly changing in ectomycorrhizal roots as the symbiosis develops. To study changes in the level of expression of plant or fungal mRNA in ectomycorrhizal roots, their expression must be normalized against the level of expression of constitutive ‘housekeeping’ genes. However, the expression of commonly used standards such as actin or tubulin alters during ectomycorrhizal formation ( Timonen et al. 1993 ; Carnero Diaz, Martin & Tagu 1996). Carnero Diaz, Tagu & Martin (1997) recently demonstrated that the ribosomal ITS region can be used as a species-specific probe for estimation of the proportion of plant or fungal RNA during development of the ectomycorrhizal association between Eucalyptus globulus and Pisolithus tinctorius. In this study, we employed ITS probes specific for birch (BpITS1) or P. involutus (PiITS1) to determine the proportion of plant and fungal RNA present in birch ectomycorrhizas with that present in NM birch roots or pure cultures of the P. involutus, and to facilitate study of the expression patterns of BpSUC1, BpHEX1 and BpHEX2 in M and NM birch roots.

We have shown that the levels of expression of two hexose and one sucrose transporter genes of the plant partner Betula pendula in M roots were reduced to less than one-third of the expression level observed in NM roots. Although a comparison of the birch ITS signal intensity showed that there was only 22% less plant RNA in M compared to NM roots, the reductions in the signal intensities of BpSUC1, BpHEX1 and BpHEX2 were much greater than this, indicating that the expression of these genes had been down-regulated in M roots. This down-regulation of expression, particularly that of the sucrose transporter, appears incompatible with stimulation of the flux of C to the roots in response to increased sink strength arising from fungal colonization. Possible explanations for this apparent discrepancy are that these transporters are not directly involved in the movement of C to the symbiotic interface within M roots, that their expression may be highly localized in response to alterations in the pattern of C allocation within M roots, or that other transporters are induced. Both sucrose and hexose transporters have been shown to be encoded by multi-gene families ( Kuhn et al. 1999 ; Lalonde et al. 1999 ). Further studies, including immunolocalization of the sucrose and hexose transporter proteins, are necessary before a role in symbiosis can be fully discounted. It is noteworthy that even when a fourfold stimulation in the expression of a fungal hexose transporter, AmMst1, was observed in Picea abies/Amanita muscaria mycorrhizas, Nehls et al. (1998) speculated that expression of the gene might be tenfold higher at the symbiotic interface within the Hartig net region but that expression in the outer mantle may be much lower than the average fourfold stimulation in expression may suggest.

There is clearly a need for further analyses of the regulation and localization of sugar transport proteins in ectomycorrhizal systems.

ACKNOWLEDGMENTS

This research was funded by NERC grants GR3/9258 and GR9/3033.

Ancillary