Author for correspondence: Frans J. M. Maathuis Tel: +44 (0)1904 434399/432854 Fax: +44 (0)1904 434317 Email: firstname.lastname@example.org
• Ericoid mycorrhizas are believed to improve N nutrition of many ericaceous plant species that typically occur in habitats with impoverished nutrient status, by releasing amino acids from organic N forms. Despite the ubiquity of mycorrhizal formation the mechanisms and regulation of nutrient transport in mycorrhizal associations are poorly understood.
• We used an electrophysiological approach to study how amino acid transport characteristics of Calluna vulgaris were affected by colonization with the ericoid mycorrhiza fungus Hymenoscyphus ericae .
• Both the Vmax and Km parameters of amino acid uptake were affected by fungal colonization in a manner consistent with an increased availability of amino acid to the plant.
• The ecophysiological significance of altered amino acid transport in colonized root cells of C. vulgaris is discussed.
Many plant species form mycorrhizal associations, with species from the Ericaceae being obligately mycorrhizal (Smith & Read, 1997). Ericoid mycorrhizas constitute an important component of a hydrologically and altitudinally diverse range of heathland and some open forest communities world-wide (Read, 1991). Such habitats are dominated by plants from the Ericaceae family and include most humus heathlands in the northern hemisphere, Mediterranean woodlands, tropical cloud forests and the dry sand plains of Australia (Read, 1991; Straker, 1996; Bergero et al., 2000). These habitats contain soils that are generally of extremely low nutrient status (Read, 1991). The success of Ericaceae in these stressful habitats is thought to lie largely in the abilities of their ericoid mycorrhizal fungal endophytes to withstand and adapt to environmental stresses, and to enhance the fitness of their plant hosts by improving N acquisition, particularly from organic soil reserves.
Mycorrhiza-derived improvements in host N nutrition occur for several reasons. First, the fungal mycelium is able to explore a much larger volume of soil and therefore extends the host plant's rhizosphere. Second, release of extracellular proteases by ericoid mycorrhizal fungi catalyses breakdown of proteins, a dominant N pool in heathland soils (Smith & Read, 1997). In addition, fungi are able to mobilize N from chitin and other cell wall components (Kerley & Read, 1997). Mobilized amino acids from organic forms of N can then be assimilated by plant roots, either directly or via the fungal intermediary. However, there may be a large degree of variation among and within fungal taxa with regard to their capacity to use diverse N sources. Cairney et al. (2000) showed that strains of the ericoid mycorrhizal fungus Hymenoscyphus ericae differed greatly in their capacity to use N sources such as ammonium, nitrate, arginine and bovine serum albumin. Further crucial questions are concerned with how fungi vary in their supply of N to their hosts, the proportion of N supplied by the fungal symbiont that reaches the host via the endosymbiont and via the external rhizosphere compartment, and how the host physiology adapts to the established symbiosis in terms of N transport.
Single-cell measurements are ideally suited to investigate nutrient transport and its regulation. We therefore used an electrophysiological approach to study the amino acid substrate specificity of C. vulgaris hair root cells and to test how amino acid transport in these cells is affected by colonization with the ericoid mycorrhizal fungus H. ericae.
Materials and Methods
Plant and fungal material
Calluna vulgaris (L.) Hull, variety ‘Alexander’, plants were cultivated in pots filled with compost soil. Alternatively, C. vulgaris seeds were sterilized by exposure to 96% ethanol for 6 min followed by 1.5% sodium hypochlorite for 6 min and four washes in sterilized water. Sterilized seeds were placed on square Petri dishes containing 0.8% agar and 25% of Murashige and Skoog medium (Sigma, Poole, UK). Both pot-grown and Petri dish-grown plants were cultivated in a controlled environment room with a 16 h/8 h light/dark cycle, at 23/20°C and a light intensity of 150 µmol m −2 s −1 .
Sterile plants were grown for 7–8 wk and inoculated with ericoid mycorrhizal fungus Hymenoscyphus ericae (Read) Korf and Kernan, that was isolated from a lowland heath in Devon, UK (Sharples et al., 2000a), by placing plugs of fungal mycelium in the vicinity of actively growing plant roots. To promote plant infection, both N and P were lowered to 50 µm in the Murashige and Skoog medium. The H. ericae cultures were axenically maintained on plates containing modified Melin–Norkrans medium (Sharples et al., 2000a) consisting of 3.79 mm (NH4)2PO4, 2.21 mm KH2PO4, 0.57 mm MgSO4, 0.23 mm CaCl2, 0.43 mm NaCl, 34 µm Fe-ethylenediaminetetraacetic acid (EDTA), 0.3 µm thiamin, 0.23 µm l-glutamic acid, 55.5 mm d-glucose and 2% agar. Plates with cultured fungus were maintained in the dark in an incubator at 25°C. For experimentation, seven pot-grown plants were used and eight, 7- to 12-wk-old Petri dish-grown plants were used. Both sets of plants showed similar differences in the characteristics for infected and uninfected cells and hence data of both sets of plants were pooled. The infection status of cells was determined microscopically, by assessing the presence of hyphal structures inside epidermal cells, as described in Read (1996). If such structures were present, such cells were assumed to be colonized, whereas cells lacking hyphal structures were deemed uncolonized. All chemicals used were purchased from Sigma UK.
Membrane potential measurements
Young roots of plants were briefly washed with distilled water and cut 2–3 cm from the tip. Roots were fixed in a measuring chamber using low melting point agarose and acclimatized for 20–30 min before impalement. Root tissue was perfused continuously at a rate of 10 ml min−1 with a standard buffer solution containing 1 mm 2-(N-morpholino) ethanesulfonic acid (MES)–Tris pH 6.0 and 1 mm CaCl2.
The transmembrane potential difference (Em) of uncolonized and colonized cortical root cells was measured using standard techniques, as described previously (Maathuis & Sanders, 1993). The effect of different amino acids on Em was determined by supplementing the standard buffer with either 0.5 mm or a range of 0.001–5 mm amino acid. Before terminating each recording, a control depolarization was carried out by perfusing the root with standard buffer supplemented with 150 mm KCl. Membrane depolarizations induced by KCl were used to normalize overall membrane conductance in different plants and did not fluctuate by more than 20%. Amino acid-induced effects on membrane potentials were recorded for infected and uninfected cells in the same root axis, the same plant and between plants, and were found to depend on infection status of cells, irrespective of tissue location. Therefore, data derived from of all colonized and all non-colonized cells were pooled.
Two-electrode voltage clamp
A voltage clamp on axenically grown fungal hyphae was essentially performed as described by Meharg and Blatt (1995). Since H. ericae hyphae contain septa, no cable correction was necessary.
Fungal N uptake capacity
Ericoid mycorrhizal fungi are crucial in recycling organic forms of N. These fungi have shown to exhibit a broad substrate range for amino acid uptake and we tested the uptake capacity of axenically grown mycelium by applying two-electrode voltage clamp measurements. Figure 1a shows representative I/V recordings in standard buffer before, during and after washout of substrate in the form of a mixture of amino acids each at 0.1 mm. In the presence of substrate, the fungal membrane depolarized by around 70 mV and the inward current was significantly increased. The difference between control and substrate I/V graph, which is depicted in Fig. 1b, reflects the amino acid-associated current. The latter was around 60 pA and, assuming that one amino acid is associated with the entering of one charge (derived from the cotransported H+), equates to 0.6 fmol s−1 for the inter-septa hyphal compartment. The fungal inter-septa compartment averages 6 × 10 µm, constituting a volume of approximately 360 µm3 or 360 fl. Thus, on a cell-water basis, the measured flux would equate to around 100 mmol kg−1 min−1, a flux that clearly far exceeds any nutritional requirement.
Amino acid specificity in colonized and uncolonized C. vulgaris hair roots
We recorded membrane potentials in hair root cells to measure the depolarizations that occur after the addition of various amino acids to the bathing medium. Figure 2 shows that in uncolonized hair root cells, most of the amino acids tested provoked a depolarization at a concentration of 0.5 mm and an ambient pH of 6.0. This indicates the influx of positive charge during amino acid uptake, presumably via symport with protons. Such depolarizations at saturating amino acid concentrations (0.5 mm) can be assumed to reflect the apparent Vmax for the amino acid permease (Sanders et al., 1983), showing relatively low uptake capacity for asparagine, proline and histidine and relatively high rates of uptake for arginine and glutamic acid. No depolarization was observed in uncolonized cells after addition of α-aminoisobutyric acid, a nonmetabolized amino acid, suggesting a lack of transport capacity for this compound.
However, in hair roots that are colonized by H. ericae, this picture changes dramatically. Generally, recorded depolarizations were increased in colonized cells (Fig. 3) for all amino acids tested with three- and four-fold larger depolarizations for histidine and asparagine, respectively. Only in the case of proline and glutamic acid did colonization not have a significant effect on the depolarization. Interestingly, and in contrast to uncolonized cells, α-aminoisobutyric acid evoked a large depolarization in colonized cells.
These data not only indicate that transport capacity for many amino acids is increased in colonized cells, but that the spectrum of possible substrates is wider.
Histidine uptake in colonized hair roots is upregulated
We undertook a more detailed analysis of membrane depolarizations as a function of the histidine concentration to study how colonization affects the kinetics of amino acid uptake, and how they compare with fungal characteristics. Figure 4 shows that perfusing uncolonized hair roots with medium containing up to 100 µm histidine hardly affected the membrane potential. By contrast, supplementing the bathing medium of a colonized hair root with a histidine with concentration of a few µmol caused a discernible membrane depolarization. Indeed, the presence of 5 mm histidine failed to depolarize the membrane of uncolonized cells by more than around 20 mV, whereas addition of 0.5 mm histidine to the bathing medium depolarized the membrane of colonized cells by more than 55 mV.
Similarly, the membrane of septal extraradical H. ericae hyphae depolarized when exposed to low micromolar concentrations of histidine. These results show that colonization of hair root cells increases the membrane conductance for histidine, which is most likely due to a drastic increase in transport capacity after cells are colonized by H. ericae.
Colonization affects both the Vmax and Km of histidine uptake
Uptake of amino acids through proton-coupled permeases causes influx of positive charge that is directly proportional to the amount of substrate being transported (Sanders et al., 1983). Plant membrane conductance is predominantly determined by the K+ gradient (Maathuis & Sanders, 1993) and, thus, it can be assumed that the overall membrane conductance remains relatively constant at varying histidine concentration. A plot of the depolarization vs the substrate concentration can therefore yield the apparent michaelian constants that describe first-order, saturating enzyme kinetics. Figure 5 shows a replot of the data presented in Fig. 4. Fitting data according to the Michaelis–Menten formalism showed a distinct difference in both the values for the apparent Vmax and Km when comparing cells from colonized and uncolonized hair roots. In uncolonized hair root cells, the maximum depolarization amounted to around 20 mV. In colonized cells this value was more than twice as high at 55 mV. More strikingly however, is the decrease in Km value after colonization, which dropped from near 160 µm to around 19 µm and points to the derepression of specific high-affinity amino acid permeases in the plasma membrane of colonized root cells. The values obtained in colonized cells very much resemble those for the axenically grown fungus which are, respectively, 48 mV and 6 µm for Vmax and Km (Fig. 5).
The major benefit of mycorrhizal formation to ericaceous plants is believed to lie in the ability of mycorrhizal fungi to supply N from sources that would otherwise be inaccessible to the plant hosts (Smith & Read, 1997). In particular, extracellular proteases released by ericoid mycorrhizal fungi catalyse breakdown of proteins, the dominant N pool in heathland soils. Mobilized amino acids can then be assimilated by plant roots, either via the soil–plant interface or via the fungal–plant interface.
No previous electrophysiological studies appear to have been performed on either symbiont of the C. vulgaris–H. ericae ericoid mycorrhiza. Calluna vulgaris displays a simple hair root anatomy containing uniform cells (Smith & Read, 1997), which is ideally suited for the application of electrophysiological techniques. Unlike, for example, radiotracer assays, electrophysiological approaches can be applied on a single-cell basis, thus allowing differentiation between (1) the fungal and plant symbiont (2) different tissues within the plant, and (3) between colonized and uncolonized plant cells.
Our electrophysiological data (Fig. 1) show that H. ericae fungi have a transport capacity to take up amino acids that far exceeds the requirements for fungal growth. This fungal property may be of fundamental importance to the successful establishment of symbiosis with its plant partners.
In plants and fungi, uptake and transport of amino acids is believed to be mediated by relatively nonspecific permeases (Sanders et al., 1983; Rentsch et al., 1998). These cotransporters are energized by coupling amino acid transfer to the transmembrane H+ gradient, leading to electrogenic transport. For example, the Arabidopsis amino acid transporter AAP5 has been shown to transport a broad range of anionic, cationic and neutral amino acids with varying affinity (Boorer et al., 1997). Typically the amino acid : H+ stoichiometry for AAP type transporters is 1 : 1 or 2 : 1, depending on the amino acid charge. Thus, the membrane depolarization that results from the influx of positive charge can be assumed to be proportional to the amount of amino acid that is transported across the membrane, and assaying the membrane potential can therefore yield useful data about transport kinetics (Maathuis & Sanders, 1994). However, the total amount of charge carried, and therefore the extent of the depolarization, depends on transport rate, amino acid species and number of cotransported H+.
Root exudates of many plants contain amino acids that may play a role in signalling and the release of insoluble nutrients. Release of amino acids is believed to be a passive process that takes place mainly near the root tip in maize (Jones & Darrah, 1994). The potential C and N losses that such exudates constitute are limited by the capacity of the root to (re)uptake amino acids. Our results indicate that in uncolonized hair root cells of C. vulgaris, amino acid-induced membrane depolarizations are, in general, lower than in cells colonized with H. ericae. The lesser extent of membrane depolarization in uncolonized cells is not caused by a generally lower membrane conductance since control depolarizations with KCl were similar in colonized and uncolonized roots (not shown).
Transport capacity for asparagine, histidine, ornithine and lysine in particular are increased after colonization. Interestingly, α-aminobutyric acid evoked a large depolarization only in colonized cells. This implies that colonization triggers a capacity to transport a broader range of substrates, including amino acids that are not metabolized.
Recent studies comparing ectomycorrhizal with nonmycorrhizal birch roots or arbuscular mycorrhizal and nonmycorrhizal ryegrass similarly reported increases in amino acid uptake after mycorrhization (Cliquet et al., 1997; Blaudez et al., 2001). These results may reflect some of the ecophysiological implications of the mycorrhiza formation. For example, fungal activity may lead to preferential mobilization of amino acids such as asparagine, histidine and lysine. Alternatively, the fungal partner may discriminate against assimilation of asparagine, histidine and lysine thereby raising the relative ambient abundance of these particular amino acid species and make them available to the plant symbiont.
Further analysis for histidine induced depolarizations (Figs 4 and 5) shows that in parallel to a two- to three-fold increase in apparent Vmax for histidine uptake, there is a drastic reduction in the value of the apparent Km for histidine from around 160 to 19 µm after plants become colonized. These data suggest that colonization triggers the derepression of a high-affinity amino acid uptake systems in C. vulgaris hair roots that enables the effective uptake of amino acids from the soil solution. An extensive study by Wallenda and Read (1999) into the kinetics for the uptake of various amino acids in mycorrhizal roots found Km values in the range of 20–200 µm. These values are of physiological significance since it is estimated that amino acid concentrations near ectomycorrhizal roots range between 10 µm and 100 µm (Wallenda & Read, 1999). Our data indicate that H. ericae is capable of transporting amino acids with an even higher affinity, at least in the case of histidine where we observed an apparent Km value of 6 µm. Since these fungi are particularly prevalent in N-poor habitats, this low Km value may reflect adaptive benefits of H. ericae in its role as N scavenger.
The increased affinity for amino acid transport in colonized plant cells must be the result of a direct fungal effect on plant gene transcription since both colonized and uncolonized plants were grown in similar conditions using NH4 and NO3 as N sources and without added amino acid substrate. The intricate mechanism of mutual signalling during early stages of mycorrhizal formation has been shown to alter transcription levels of many plant genes (Murphy et al., 1997) that could include transporters. Alternatively, the presence of the fungus may lead to a general increase in amino acid supply to the plant, thus requiring induction of transport capacity. Even though the N source for both fungus and plant is inorganic, the fungal symbiont could deliver part of its amino acid pool to the plant sink. Similarly, the extraradical hyphae can promote breakdown of plant- and/or fungus-derived material, thereby mobilizing amino acids. The presence of low levels of amino acid substrate in the apoplastic environment could induce transcriptional upregulation of plant high-affinity amino acid transport mechanisms.
The latter scenario, in combination with our observation that amino acid uptake capacity is present at the epidermis soil boundary, suggests that at least a portion of the mobilized amino acids is delivered to the plant symplast via the soil compartment rather then through the fungal compartment.
Our study shows some obvious advantages of using an electrophysiological approach to study aspects of mycorrhizal nutrition. Transfer of most nutrients across membranes constitutes ion fluxes which (or their effects on membrane potentials) can be determined with great sensitivity and, more importantly, with a high spatial resolution. Individual cells can be assayed and therefore interpretation of the data is not obscured by potential overlap of fungal and plant membrane functions. In combination with culturing methods that allow manipulation of the growth conditions and mycorrhizal status, the application of electrophysiology techniques appears especially promising to study underlying mechanisms of nutrient transfer in mycorrhizal associations.
This work was funded by Natural Environment Research Council.