Contribution of hyphae and roots to uranium uptake and translocation by arbuscular mycorrhizal carrot roots under root-organ culture conditions

Authors

  • Gervais Rufyikiri,

    1. Belgian Nuclear Research Centre (SCK.CEN), Radiation Protection Research Department, Radioecology Section, Boeretang 200, 2400 Mol, Belgium;
    Search for more papers by this author
  • Yves Thiry,

    1. Belgian Nuclear Research Centre (SCK.CEN), Radiation Protection Research Department, Radioecology Section, Boeretang 200, 2400 Mol, Belgium;
    Search for more papers by this author
  • Stéphane Declerck

    1. Université catholique de Louvain, Mycothèque de l’Université catholique de Louvain (MUCL, Part of the Belgian Coordinated Collections of Micro-organisms (BCCM)), Unité de microbiologie, 3 Place Croix du Sud, 1348 Louvain-la-Neuve, Belgium
    Search for more papers by this author

Author for correspondence: Gervais Rufyikiri Tel: +32 (0) 14 33 21 16 Fax: +32 (0) 14 32 10 56 Email: grufyiki@sckcen.be

Summary

  • • Here, the respective contributions of the arbuscular mycorrhizal (AM) fungus Glomus intraradices and carrot (Daucus carota) roots to the uptake and translocation of uranium (U) were quantified and compared.
  • • The U absorption by the AM fungus and roots was observed by growing mycorrhizal and nonmycorrhizal roots in two-compartment Petri plates. The central compartment allowed growth of roots and extraradical fungal hyphae. The external compartment (EC), which was labelled with 0.1 µm233U, allowed growth of: hyphae only (hyphal compartment, HC), both mycorrhizal roots and hyphae (root hyphal compartment, RHC), or nonmycorrhizal roots (root compartment, RC).
  • • The U concentration was 5.5 and 9.6 times higher for hyphae than for the mycorrhizal and nonmycorrhizal roots, respectively, both developing in the EC’s. Translocation of U was similar for the RHC and the HC systems, and was 8 times higher for these two systems than for the RC system.
  • • These results indicate that the U flux rate was higher in fungal hyphae than in roots, while the intraradical hyphae may significantly contribute to the U immobilization by mycorrhizal roots.

Introduction

Uranium (U) contamination of surface soils and groundwater can result in environmental problems because it poses health risks to both humans and animals. There are large U-contaminated sites in the world, mostly linked with improper storage of wastes produced by mining and processing of U ores, requiring remedial actions to optimize their socio-economical value. Different techniques including chemical and physical methods (Abdelouas et al., 1999; Vandenhove et al., 2000) have been suggested for the management and restoration of areas contaminated by U and its fission by-products. However, these techniques are often expensive, with highly variable results (Shahandeh et al., 2001). Beside these techniques, phytoremediation consisting in the use of plants and associated microbiota is an option harmless for the environment and nowadays considered with interest for the long-term management of U-contaminated sites (Huang et al., 1998; Shahandeh et al., 2001). The efficiency of this technique is largely based on the capacity of the different partners involved in the association, that is plants and microorganisms, to contribute in the U uptake by plants. Therefore, understanding the function of both partners and their intimate interactions are essential to envisage any phytoremediation option.

Mycorrhizas are the most widespread associations between microorganisms and higher plants. They concern 80–90% of all seed plant species in ecosystems throughout the world (Harrison, 1997; Smith & Read, 1997). Their extraradical mycelium forms an active continuum from soil to plant able to take up, translocate and transfer nutrients to their host (Smith & Read, 1997). Despite their agreed effect on mineral nutrition of plants and heavy metals sequestration/transport (Colpaert & Vandenkoornhuyse, 2001), little information is available on their effect on the uptake of radionuclides. Some authors, comparing mycorrhizal and nonmycorrhizal plants, observed an influence of arbuscular mycorrhizal (AM) fungi on 137Cs (Entry et al., 1999; Berreck & Haselwandter, 2001), 134Cs (Strandberg & Johansson, 1998) and 90Sr (Entry et al., 1999) acquisition. However, as suggested for P (Pearson & Jakobsen, 1993) such an approach may be inappropriate to estimate the contribution of the mycosymbiont to the content of these radionuclides in the mycorrhizal plant. Beside direct contribution of external hyphae to the transport reported for some elements (Cooper & Tinker, 1978; Jakobsen et al., 1992), various indirect effects such as the differences in the architecture of mycorrhizal and nonmycorrhizal root systems (Hetrick, 1991) could markedly influence the acquisition of elements by mycorrhizal plants. Moreover, because the experiments were generally conducted under in vivo conditions one may not exclude direct or/and indirect influences of microorganisms other than the inoculated AM fungi.

Recently Rufyikiri et al. (2002b) demonstrated, in a two-compartment root-organ culture (ROC) system (St-Arnaud et al., 1996), initially used for labelled P transport studies (Joner et al., 2000b; Koide & Kabir, 2000; Nielsen et al., 2002), that the extraradical mycelium network of an AM fungus G. intraradices could take up and translocate U to the roots. Both these processes appeared highly influenced by the pH, whereas translocation was correlated with the number of hyphae linking both compartments. Despite this major finding, no distinction could be made between the contribution of hyphae, roots and both in combination, on the U uptake. However, the quantification of the relative importance of the fungus and the host root to U uptake may be useful to identify their respective role in the U accumulation within plants and to delineate some possible options for phytomanagement of U-contaminated sites.

The objective of this study was to distinguish between U uptake and translocation by the AM fungal hyphae and plant roots, under ROC conditions. The contribution of both would reflect the U uptake capacity of a mycorrhizal, that is mycorrhizal root vs isolated hyphae, and a nonmycorrhizal root under strict controlled conditions.

Materials and Methods

Biological material

Glomus intraradices Schenck and Smith (MUCL 41833) was used for the experiment. Root-organ cultures were established in association with Agrobacterium rhizogenes (Ri T-DNA)-transformed carrot (Daucus carota L.) roots (Declerck et al., 1998). Routine maintenance of roots and AM fungi was made on the modified Strullu-Romand (MSR) medium (Declerck et al., 1998) modified from Strullu and Romand (Strullu & Romand, 1986) but gelled with 3 g/l Gel Gro™ (ICN, Biomedicals, Inc., Irvine, CA, USA) instead of 8 g/l Bacto agar. After 3 months, several thousand spores were obtained in each Petri plate and used for the experiment.

Experimental design

A two-compartment system, consisting of a 50-mm Petri plate cover glued in a 90-mm Petri plate, thus separating a central compartment (CC) from a surrounding external compartment (EC) was used (Rufyikiri et al., 2002b). This system was, however, slightly improved by filling the CC with the MSR medium in a layer of gel extending 2 mm above the edge of the 50-mm Petri plate. This modification was aimed to facilitate hyphae and/or roots to cross the partition between the two compartments (Fig. 1.1), therefore increasing the biomass in contact with the radionuclide introduced in the EC. The CC allowed the growth of nonmycorrhizal roots or mycorrhizal roots and extraradical fungal hyphae. For the EC, three scenarios were tested on the basis of the type of biological material (hyphae or mycorrhizal roots and hyphae or nonmycorrhizal roots) making contact between the two compartments: a hyphal compartment (HC) where only hyphae were allowed to grow; a root hyphal compartment (RHC) where both mycorrhizal roots and hyphae were allowed to grow; and a root compartment (RC) where nonmycorrhizal roots were allowed to grow (Fig. 1.2).

Figure 1.

Schematic representation of a two-compartment root-organ culture system allowing the spatial separation of a central compartment (CC) for the growth of mycorrhizal or nonmycorrhizal roots from a neighbouring external compartment (EC) where extraradical mycelium and/or roots were allowed to grow (1). The modified Strullu-Romand (MSR) medium was gelled in the CC, while the EC contained a liquid MSR medium lacking sucrose and vitamins but labelled with 0.1 µm233U. The CC compartment was filled with the MSR medium to reach a layer of gel extending 2 mm above the edge of the 50-mm Petri plate, in order to facilitate hyphae and/or roots to cross the partition between the two compartments. For the EC, three scenarios were tested: a hyphal compartment (HC) where only hyphae were allowed to grow; a root hyphal compartment (RHC) where both mycorrhizal roots and hyphae were allowed to grow; and a root compartment (RC) where nonmycorrhizal roots were allowed to grow (2). Roots are represented by full lines and fungal hyphae by broken lines. Adapted from Rufyikiri et al. (2002b).

Transformed carrot roots (approx. 70 mm length) were introduced in the CC containing 15 ml MSR medium and inoculated or not with approx. 150 spores of G. intraradices. The Petri plates were incubated horizontally in an inverted position at 27°C in the dark for 2 wk. Thereafter, they were set upright and the EC’s, that is HC, RHC and RC, were filled with 25 ml liquid MSR medium without sucrose and vitamins, for an additional week. Hyphae and/or roots started to cross the partition between the CC and the EC's and proliferated in the liquid medium. For the HC, roots that crossed the partition between the two compartments were trimmed to maintain the HC void of roots. At this stage, the cultures were ready for the U labelling.

Uranium labelling

Liquid MSR medium (15 ml) without sucrose and vitamins, but labelled with U at a concentration of 0.1 µm, the main isotope being 233U, was added to the EC's after removing the old medium with a pipette. The source of U was a solution of uranyl nitrate (IRMM-040a Spike Isotopic Reference Material) supplied by EC-JRC-IRMM (Mol, Belgium) with a specific activity of 352 MBq α-emission g−1 U (and < 0.009 MBq βγ emission) and an isotopic composition with the following mass fraction: 98.02%233U, 0.92%234U, 0.22%235U, 0.02%236U and 0.82%238U. The pH of the labelled liquid medium was adjusted to 5.5 with 0.01 m NaOH before sterilization at 121°C for 15 min The total U concentration was fixed at a value where no precipitate could form as previously reported (Rufyikiri et al., 2002b).

Capillary action was limited by filling the EC with a small volume of the U-labelled solution to have the level of the solution 2.5 mm below the edge of the partition between the two compartments. In addition, metabolic activity of roots and hyphae was inhibited in half of the Petri plates by formaldehyde (2% v/v) added to the solution in the EC’s, 24 h before U was supplied. This control treatment was included to determine if U uptake and translocation by hyphae and/or roots was an active process.

The hyphae and/or roots were maintained in contact with the U-labelled solution during 2 wk. Root-free compartments of nonmycorrhizal cultures were also filled with the U-labelled solution to verify the contamination of the CC caused by experimental manipulations. Each treatment was replicated six times.

Assessment of variables

At the end of the experiment the total extraradical hyphal length and the number of spores in the CC, and the total root length in the CC, RHC and RC were estimated using a 10-mm intersection grid method as previously described (Rufyikiri et al., 2002b). The number of hyphae and roots crossing the partition between the CC and the EC's as well as the diameter of roots at the zone lying on the partition between the two compartments were also assessed. The solutions in the EC's were sampled for pH and U activity measurements. These compartments were then rinsed three times with 15 ml distilled water before the hyphae or roots developing in them were collected. For the CC, roots without (nonmycorrhizal roots) or with intraradical fungal hyphae (mycorrhizal roots) and the gel without or with extraradical hyphae and spores were separately collected. The f. wts of the different samples collected from the CC and the EC's were consecutively measured. Roots of the CC were thereafter divided into two subsamples, one for the assessment of the U activity content by sequential extraction as described below and the other for the measurement of the root AM fungal colonization.

The cation exchange capacity (CEC) and extractable and nonextractable U, Ca and Mg were measured on fresh roots (the first sample) and mycelium using a method adapted from Dahlgren et al. (1991) and Dufey & Braun (1986). Fresh roots (200 mg) and fresh extraradical mycelium (20 mg) were equilibrated for 3 h with 15 ml and 5 ml of a solution containing 10−2 M CuSO4, respectively. The elements desorbed from roots and mycelium by Cu2+ were considered as ‘Cu-extractable elements’. The roots and mycelium were then washed three times with deionised water in order to remove excess of CuSO4. The amount of Cu2+ retained on roots and extraradical mycelium was considered as a measure of the CEC. It was assessed by extracting Cu2+ with 15 ml and 5 ml 10−2 M HCl, for roots and mycelium, respectively, with 3 h for the equilibration with the acid solution. An additional extraction was performed with 10−1 M HCl as described above before the roots and mycelium were further treated for the residual elements measurement. These roots and mycelium as well as the gel, i.e. containing spores and extraradical hyphae, were placed in 20-ml glass scintillation vials and calcined at 500°C for 24 h. The ashes were then dissolved in 10−1 M HCl. A liquid scintillation cocktail (10 ml) was added to 5 ml aliquots of all solutions, and U activity determined by liquid scintillation counting with a counting efficiency of 100% and a detection limit of 0.03 Bq. Counts were corrected by substraction of background levels of 0.06 Bq. Calcium, Mg and Cu in solutions were measured by atomic absorption spectroscopy.

Roots of the second sample were dried and cut into 10 mm segments length, cleared with 10% KOH and stained with 0.1% Trypan blue for measurement of the root AM fungal colonization (Phillips & Hayman, 1970). Fifty randomly selected segments were examined under microscope. The frequency of AM fungal colonization (%F) was calculated as the percentage of root segments colonized by either hyphae or arbuscules or vesicles. In addition, the intensity of colonization (%I), that is the abundance of hyphae, arbuscules and vesicles in each mycorrhizal root segment, was determined (Declerck et al., 1996).

Statistical analysis

Statistical analysis of data was performed with the statistical software STATISTICA for Windows (StatSoft, 2001). Significant differences were considered at P= 0.05, and mean values were ranked by Scheffé's multiple-range test when more than two groups of data were compared by anova, or t-test paired method when only two groups of data were compared.

Results

PH of the solution

The pH values of the solutions measured at the end of the experiment in the EC's are shown in Fig. 2. As compared to the initial pH of 5.5, measured in the absence of roots and/or hyphae, the pH significantly decreased in the RHC and RC, while it significantly increased in the HC. When treated with formaldehyde, the pH measured in these EC's did not significantly change as compared to the initial pH. Similarly, the pH measured in the external root-free compartment of nonmycorrhizal cultures did not significantly change as compared to the initial pH (average value of 5.6 ± 0.1).

Figure 2.

pH measured in the external compartments, that is hyphal compartment (HC), root hyphal compartment (RHC) and root compartment (RC), (a) without and (b) with formaldehyde added to the liquid modified Strullu-Romand (MSR) medium. Values are averages and bars are standard deviations (n = 6). Significant differences between averages are indicated by different letters (P = 0.05).

Root and AM fungal biomass

The roots grew and ramified in the MSR medium in the CC with a root length and a root f. wt of 181 ± 3 cm and 841 ± 36 mg per Petri plate, respectively, at the end of the experiment (Table 1). Some roots, in the range of 4 to 6, crossed the partition between the two compartments and developed in the RHC and the RC. The root biomass produced in these compartments represented about 50% of that observed in the CC, on the basis of root length and f. wt. For the HC, roots attempting to cross the partition were trimmed to maintain this compartment devoid of roots.

Table 1.  Arbuscular mycorrhizal fungal and root biomass production in the central compartment (CC) and external compartments (ECs) without or with formaldehyde added to the solution in the external compartments
 Without formaldehydeWith formaldehyde
  • 1

    Averages for the hyphal biomass from the hyphal compartment and root hyphal compartment, and

  • 2

    averages for root biomass from the root hyphal compartment and root compartment; for each variable, values were combined to present averages values ± standard deviations because no significant differences were observed between scenarios.

CC
 %F  86 ± 8  79 ± 2
 %I  27 ± 2  23 ± 4
 Mycelium length (cm)1059 ± 111 613 ± 26
 Root length (cm) 181 ± 3 117 ± 10
 Root f. wt (mg) 841 ± 36 355 ± 22
EC
 Number of hyphae crossing the barrier1 147 ± 11  70 ± 12
 Mycelium f. wt (mg)1  21 ± 11.62 ± 0.15
 Root length (cm)2  98 ± 4  22 ± 3
 Number of roots crossing the barrier   5 ± 1   5 ± 2
 Root f. wt (mg)2 386 ± 62  17 ± 4

For the CC containing roots inoculated with the AM fungus, roots were highly colonized as shown by the %F and %I (Table 1). Extraradical mycelium length within the CC reached over 1000 cm with thousands of spores produced. Numerous hyphae crossed the partition between the CC and EC's and developed in the HC and in the RHC. Once in contact with the liquid MSR medium, a good branched mycelium developed and thousands of spores were produced. The hyphal and spores f. wt in these compartments reached 21 ± 1 mg.

In the cultures with formaldehyde added to the solution, hyphae and roots were killed in the EC’s, and no further growth or development was observed: the number of hyphae or roots crossing the partition remained constant, no hyphal or root branching and no fungal spores were produced during the two weeks of the experiment.

Uranium uptake and translocation by fungal hyphae and roots

At the end of the experiment, 4–5% of the U supplied in the EC's was taken up by the AM fungal mycelium developing in the HC and RHC (Table 2). The uptake of U by the carrot roots was largely influenced by the presence or not of AM fungus. It represented 17.5% of the initial U supply in the mycorrhizal roots developing in the RHC while only 6.1% of the initial U supply were observed in the nonmycorrhizal roots developing in the RC, although the root biomass production was identical for both these compartments as mentioned in Table 1. Significant differences were also observed for the biomass-specific U content between hyphae and roots, both grown in the EC’s. The biomass-specific U content was 5.5 and 9.7 times larger for AM fungal mycelium than for mycorrhizal roots and nonmycorrhizal roots, respectively, while it was 1.8 times larger for mycorrhizal roots than for nonmycorrhizal roots.

Table 2.  Uranium activity in the different compartments in function of the external compartment system, that is hyphal compartment (HC), root-hyphal compartment (RHC), root compartment (RC), without and with formaldehyde added to the solution in these compartments
 U activity content (Bq/Petri plate)1U activity concentration (Bq g1 f. wt)
HCRHCRCHCRHCRC
  • 1

    The U supply was 125 Bq/Petri plate. Values are averages of 6 replicates, and values in parentheses indicate percentages of the U supply; nd = not detectable. Within rows, averages followed by the same small letter for U content or the same capital letter for U concentration are not significantly different (P = 0.05).

Without formaldehyde
 Solution105 (84)a80 (64)b114 (92)a   
 Mycelium6 (5)a5.4 (4)a276A265A 
 Roots in the external compartments22 (17.5)a8 (6.1)b49A28B
 Gel with fungal biomass6.0 (4.8)b9.4 (7.5)a0.5 (0.43)c0.41A0.63A0.04B
 Roots in the central compartment8.1 (6)a9.8 (8)a1.1 (0.86)b9.8A10.9A1.4B
With formaldehyde
 Solution122 (98)a121 (97)a121 (97)a   
 Mycelium0.98 (0.8)a0.9 (0.7)a658A503A
 Roots in the external compartments1.9 (1.6)a2.7 (2.1)a146A129A
 Gel with fungal biomass0.08 (0.06)a0.15 (0.12)a0.05 (0.04)a0.005A0.01A0.003B
 Roots in the central compartmentndndndndndnd

Uranium was translocated by hyphae and/or roots from the EC's towards the roots developing in the CC. Uranium was found in the gel and in the roots in both nonmycorrhizal and mycorrhizal cultures. The total amount of U translocated from the EC's to the CC significantly differed between the three EC’s. It was the largest with the RHC, intermediate with the HC and the lowest with the RC.

With formaldehyde added in the solution, the U concentration was 2–3 times larger in AM fungal mycelium and in roots developing in the EC's than in the absence of formaldehyde, but U was not detected in the roots developing in the CC and was at extremely low concentrations in the gel with fungal biomass. Besides, U was not detected in the CC in control cultures without hyphae or roots in contact with the solution in the external compartments.

Cation exchange capacity of roots and mycelium

Table 3 shows the values of CEC and sequential extraction of U, Ca and Mg for carrot roots and fungal hyphae. The CEC was approx. 4 times larger for the fungal mycelium than for the roots. It appears that the Ca and Mg contents in roots and mycelium were to a large extent present as Cu-extractable forms, whereas the Cu-extractable U represented only 6 and 15% of the total U contents in roots and mycelium, respectively. After the extractions with CuSO4 and HCl, the residual U in roots and in mycelium represented 47% and 67% of the total U contents, respectively, while the residual Ca and Mg were in the range 5–13% of their respective total contents.

Table 3.  Cation exchange capacity (CEC) and sequential extraction of U, Ca and Mg with 10−2 M CuSO4, 10−2 M HCl and 10−1 M HCl for carrot roots and mycelium of Glomus intraradices
VariablesRootsHyphae
  1. Values are averages of 6 replicates, and values in parentheses indicate percentages of the total biomass-specific U contents. Within rows, averages followed by the same letter are not significantly different (P = 0.05).

CEC (cmolc kg−1 d. wt)47b187a
U (Bq g−1 f. wt)
 CuSO4 extract 2.8 (6.4)b 41.0 (15.0)a
 10−2 M HCl extract 6.1 (14.0)b 33.4 (12.2)a
 10−1 M HCl extract14.4 (33.0)a 17.2 (6.3)a
 Residual20.3 (46.6)b182 (66.5)a
Ca (mg g−1 f. wt)
 CuSO4 extract 0.323 (74.1)b  1.015 (80.0)a
 10−2 M HCl extract 0.039 (8.9)b  0.066 (5.2)a
 10−1 M HCl extract 0.025 (5.7)b  0.088 (6.9)a
 Residual 0.049 (11.2b  0.100 (7.9)a
Mg (mg g−1 f. wt)
 CuSO4 extract 0.391 (59.3)a  0.276 (88.5)b
 10−2 M HCl extract 0.150 (22.8)a  0.016 (5.1)b
 10−1 M HCl extract 0.031 (4.7)a  0.006 (1.9)b
 Residual 0.087 (13.2)a  0.014 (4.5)b

Discussion

Adequacy of the root-organ culture system

The ROC system allowing the physical separation of a nonlabelled mycorrhizal root compartment from an extraradical mycelium ramifying in a neighbouring labelled root-free compartment has successfully been used in a number of experiments to study the uptake and translocation of phosphorus (Joner et al., 2000b; Nielsen et al., 2002), as well as uranium (Rufyikiri et al., 2002b). In most experiments, the hyphal biomass produced in the labelled compartment was low, < 5 mg f. wt, caused by a limited number of hyphae crossing the partition between the two compartments (Joner et al., 2000b; Rufyikiri et al., 2002b). This rather restricted hyphal development could have resulted in an underestimation of the effect of hyphae on element uptake and translocation and could have masked the effect of hyphae on the chemical characteristics of the growth medium such as pH. Indeed, Rufyikiri et al. (2002b) reported that the pH of the solution was not significantly affected by the hyphal development.

In the present study, the ROC system was improved by extending the gelled medium 2 mm above the physical separation between the CC and the EC’s. Around 147 hyphae crossed the partition between the two compartments increasing by sixfold the hyphal biomass reported in previous studies (Joner et al., 2000b; Rufyikiri et al., 2002b).

Impact of mycelium and root development on pH

The considerable mycelium development observed in this study resulted in a marked alkalinization of the solution, while roots induced an acidification under the same growth conditions. The pH changes of growth media are a common phenomena related to imbalances in the uptake of cations and anions (Marschner, 1995). In hydroponic culture conditions, it was shown that a net proton excretion resulting in an acidification occurs when excess cations were absorbed over anions, while alkalinization resulted from a net OH excretion because of an excess uptake of anions over cations (Rufyikiri et al., 2001). The root-induced pH modification of growth media is well documented for both nonmycorrhizal and mycorrhizal plants (Li et al., 1991; Rufyikiri et al., 2000; Hinsinger, 2001), but few data are available on the effects of extraradical hyphal on the pH of the growth medium. Recently, Bago et al. (1996) used a technique of pH indicator bromocresol purple and observed a similar pH increase induced by the extraradical hyphae of G. intraradices in the presence of NO3-N as source of N, but not in media lacking this N-form. It was suggested that the pH increase was a consequence of the active uptake of NO3-N involving the NO3/H+ symport or NO3/OH antiport mechanisms used by the fungus for nitrate uptake. These mechanisms would mask any other hyphal-promoted acidification, resulting in a net alkalinization. In the present study, the modification of pH was also an active process for both roots and hyphae because such an effect was not observed when their metabolic activity was inhibited by formaldehyde added to the solution. Because U speciation in aqueous systems as well as in soil is pH-dependent (Grenthe et al., 1992) and considering differences between U species in their uptake by plants (Ebbs et al., 2000), and uptake and translocation by fungal hyphae (Rufyikiri et al., 2002b), the modification of pH is a physiological process by which growing mycelium and roots can influence the U bio-availability.

Comparison of hyphae and roots on U uptake and translocation

The findings of higher CEC in fungal mycelium than in roots observed in this study corroborated previously reported values (Joner et al., 2000a). The larger U concentration in fungal mycelium than in roots suggests that the exchange sites significantly contributed to the U uptake, as shown by the differences of Cu-extractable U between the mycelium and roots. However, sequential extractions of U with CuSO4 and HCl showed that only a small fraction of U taken up was released from roots and hyphae, while these procedures allowed us to extract most of the contents in Ca and Mg. The reason might be that U fixed on exchange sites served as starting point for precipitation or complexation reactions with various anions. The formation of stable complexes or precipitates was likely the main mechanism of U accumulation in both roots and fungal hyphae in contact with U in the external compartments. This is assumed to contribute to the low translocation of U as especially observed for roots. However, the measurements carried out in this study did not allow us to identify these nonextractable U forms. Similar root accumulation in nonexchangeable dominant forms was reported for Al in roots of banana (Musa spp.) plants (Dufey et al., 2001; Rufyikiri et al., 2002a) and roots of Abies amabilis (Dahlgren et al., 1991). These authors suggested that various mechanisms may be involved including a high affinity with the root exchange sites, complexation and precipitation with organic compounds such as oxalate or with inorganic compounds such as phosphate, and polynuclear hydroxyl formation.

The larger U concentration in the mycorrhizal roots grown in the RHC vs the nonmycorrhizal roots grown in the RC could probably be explained by U uptake mechanisms more active in the mycorrhizal roots than in the nonmycorrhizal ones, and/or a marked contribution of the intraradical hyphae to the accumulation of U in the host roots. A high concentration of U in intraradical fungal hyphae, of an undefined AM fungal species, than in the host root tissues was previously reported (Weiersbye et al., 1999), probably as a result of particular chemical conditions prevailing in the intraradical fungal cells. Large P concentration in the intraradical parts of AM fungi was recently observed (Pfeffer et al., 2001; Nielsen et al., 2002), while intracellular pH varying between 5.6 and 7.0 was reported for hyphae of G. intraradices (Jolicoeur et al., 1998). Both high P concentration and weakly acidic to neutral pH are factors which can promote the formation of U-phosphate complexes and precipitates in the intraradical hyphae, and thus favouring the U accumulation in mycorrhizal roots.

The improvement of the culture system with a large hyphal biomass production in the external compartments resulted also in a large U translocation by the extraradical fungal hyphae in comparison with the results reported previously (Rufyikiri et al., 2002b). Whatever the mechanisms involved in the U translocation, being active or passive, increasing the number of hyphae crossing the partition between the two compartments would result in a proportionally higher U translocation (Rufyikiri et al., 2002b). However, the flux of elements in hyphae may depend, not only on absorbing hyphae, but also on other factors such as metabolic control related to the demand of the host plant and may differ from one element to another (Cooper & Tinker, 1978).

Considering hyphae and roots as cylinders with an average diameter of 11 ± 2 µm for a hyphae (Nielsen et al., 2002) and 1000 µm for a root (measured in this experiment), the total cross section area at the partition between the two compartments (A) was calculated as A= (diameter/2)2 × π × number of hyphae/roots. The A was c. 0.014 mm2 for the average 147 hyphae and 3.93 mm2 for the five roots. Although the total section area of roots was 281-fold larger than what was observed for hyphae the U translocation by roots was lower than its translocation by hyphae. This indicates that U flux rate was larger in hyphae than in roots, perhaps as a result of differences in interactions or exchange reactions between U and cell components as well as in further transport mechanisms involved.

The effect of formaldehyde added to the solution to control capillary action was similar to the experiment reported in our previous paper (Rufyikiri et al., 2002b). The comparison between data for living and formaldehyde-killed hyphae or roots indicated that the hyphal and root U concentration and translocation were influenced by the metabolic activity of fungal hyphae and roots. Killing them has resulted in increasing U concentration and in limiting U translocation from the EC to the EC. As previously reported for hyphae (Rufyikiri et al., 2002b), the U accumulation by the formaldehyde-killed hyphae and roots was caused by only passive mechanisms such as the adsorption on exchange sites of hyphae and roots, while for living hyphae, active mechanisms were also involved to control the U absorption. High capacity of U adsorption of hyphae for killed hyphae than for living ones was also reported in another study (Joner et al., 2000a). The absence of detectable amounts of U in roots developing in the CC and its presence in negligible amounts in the gel, although killed hyphae and roots in the EC remained attached to the CC, as observed at harvest, indicate that if capillary action existed, its contribution was certainly negligible.

In conclusion, these results are the first to determine the uptake and translocation of U by roots under strictly controlled ROC conditions and to compare them to those of the AM fungal hyphae. They demonstrated that the extraradical hyphae of G. intraradices have a relatively higher capacity to take up and to translocate U than carrot roots, while the intraradical hyphae may play an important role in the U immobilization by mycorrhizal roots.

Acknowledgements

This work was supported by the Belgian Nuclear Research Centre (SCK.CEN) and the EU-MYRRH project contract -CT-2000–00014 ‘Use of mycorrhizal fungi for the phytostabilization of radio-contaminated environments’. S. Declerck gratefully acknowledges the financial support from the Belgian Federal Office for Scientific, Technical and Cultural affairs (OSTC, contract BCCM C2/10/007) and thanks the director of MUCL for the facilities provided and for continual encouragement.

Ancillary