Mechanism of Arsenate Resistance in the Ericoid Mycorrhizal Fungus Hymenoscyphus ericae

Abstract

Arsenate resistance is exhibited by the ericoid mycorrhizal fungus Hymenoscyphus ericae collected from As-contaminated mine soils. To investigate the mechanism of arsenate resistance, uptake kinetics for arsenate (H₂AsO₄⁻), arsenite (H₃AsO₃), and phosphate (H₂PO₄⁻) were determined in both arsenate-resistant and -non-resistant H. ericae. The uptake kinetics of H₂AsO₄⁻, H₃AsO₃, and H₂PO₄⁻ in both resistant and non-resistant isolates were similar. The presence of 5.0 μm H₂PO₄⁻ repressed uptake of H₂AsO₄⁻ and exposure to 0.75 mm H₂AsO₄⁻ repressed H₂PO₄⁻ uptake in both H. ericae. Mine site H. ericae demonstrated an enhanced As efflux mechanism in comparison with non-resistant H. ericae and lost approximately 90% of preloaded cellular As (1-h uptake of 0.22 μmol g⁻¹ dry weight h⁻¹ H₂AsO₄⁻) over a 5-h period in comparison with non-resistant H. ericae, which lost 40% of their total absorbed H₂AsO₄⁻. As lost from the fungal tissue was in the form of H₃AsO₃. The results of the present study demonstrate an enhanced H₃AsO₃ efflux system operating in mine site H. ericae as a mechanism for H₂AsO₄⁻ resistance. The ecological significance of this mechanism of arsenate resistance is discussed.

As is ubiquitous in nature with As levels being elevated by mining, industrial, and agricultural activities (Meharg et al., 1994). In the southwest of England, the mining and processing of As ore has led to highly contaminated mine spoil soils. As may be present in these spoil soils at 35 mmol kg⁻¹ with arsenate being the dominant form of available soil As (Colbourn et al., 1975). Within the pH range 2.0 to 6.5 arsenate exists predominantly as H₂AsO₄⁻ (Fergusson, 1990). The dominant vegetative cover on these mines is Calluna vulgaris, which is present in ericoid mycorrhizal association with the ascomycete fungus Hymenoscyphus ericae (Sharples et al., 2000). Ericoid mycorrhizal symbiosis is considered to be critical to the survival of plants in the order Ericales on natural heathland sites and sites contaminated with toxic metals. The principle benefit conferred upon plants by ericoid mycorrhizal association is fungus-mediated access to otherwise unavailable sources of organic nitrogen and phosphorus, whereas the fungus may also alleviate toxic metal stress under some circumstances (Smith and Read, 1997).

H₂AsO₄⁻ resistance is exhibited by the ericoid mycorrhizal fungus, H. ericae (Sharples et al., 1999, 2000) and selected angiosperms, including Holcus lanatus (Meharg and Macnair, 1990), Agrostis capillaris, and Deschampsia cespitosa (Meharg and Macnair, 1991, 1992), collected from As-contaminated mine soils. H₂AsO₄⁻ is a H₂PO₄⁻ analog and competes with H₂PO₄⁻ as a substrate for the H₂PO₄⁻ uptake system in angiosperms (Asher and Reay, 1979; Ullrich-Eberius et al., 1989), fungi (Rothstein and Donovan, 1963; Jung and Rothstein, 1965; Beever and Burns, 1980), mosses (Wells and Richardson, 1985), lichens (Nieboer et al., 1984), and bacteria (Silver and Misra, 1988). Resistance mechanisms to H₂AsO₄⁻ in the bacteria, Staphylococcus aureous and Eschericha coli, involve reducing cellular concentrations of As and rapidly effluxing them via a plasmid-encoded arsenical pump (Rosen, 1986; Silver and Misra, 1988). Meharg and Macnair (1990) demonstrated that H₂AsO₄⁻ resistance in the grass H. lanatus was due to suppression of the high affinity H₂PO₄⁻uptake system in H₂AsO₄⁻ tolerant plants, which led to reduced uptake of both H₂PO₄⁻ and H₂AsO₄⁻.

Sharples et al. (2000) recently isolated populations of arsenate-resistant H. ericae from roots of C. vulgaris on As-contaminated mine spoil soils in southwestern England. Populations of the mycorrhizal fungus have evolved resistance to arsenate contamination in parallel with the host plant, and it seems likely that the presence of the mycorrhizal fungus in roots of C. vulgaris is essential its establishment and persistence on As-contaminated sites (Sharples et al., 2000). The present study investigated the mechanism of H₂AsO₄⁻ resistance in an isolate of H. ericae from an As-contaminated mine site.

RESULTS

Effect of H₂AsO₄⁻ and H₃AsO₃ on Biomass Production

The heathland H. ericae isolate demonstrated significantly greater sensitivity to H₂AsO₄⁻ and H₃AsO₃ than the mine site isolate (Fig. 1). Growth of the heathland isolate was almost completely inhibited at 1.33 mm H₂AsO₄⁻ and above, whereas growth of the mine site isolate was inhibited by only 40% at the highest concentration tested (13.3 mm) (Fig. 1A). Biomass yield of the heathland isolate was 15 μg h⁻¹ in the presence of 13.3 mm H₃AsO₃, whereas the mine isolate produced a mean biomass yield of approximately 45 μg h⁻¹ at the same H3AsO3 concentration (Fig. 1B). Growth of both isolates was more severely affected by the presence of H₂AsO₄⁻ than H₃AsO₃.

Figure 1.

A, Growth of mine site (▾) and heathland (●) H. ericae over a range of H₂AsO₄⁻ concentrations. Bars = mean ± se (n = 3). B, Growth of mine site (▾) and heathland (●) H. ericae over a range of H₃AsO₃ concentrations. Bars = mean ± se (n = 3).

Kinetics of High Affinity H₂AsO₄⁻ and H₂PO₄⁻ Uptake

In H₂PO₄⁻-deficient tissue the rate of H₂AsO₄⁻ and H₂PO₄⁻ uptake was dependent on concentration, and the uptake of both ions displayed saturation kinetics (Figs. 2 and 3). Single Michaelis-Menten functions were fitted to the data, representing the high affinity uptake carrier, which would predominate at low-substrate concentrations used here (Meharg and Macnair, 1990). Kinetics of both H₂AsO₄⁻ and H₂PO₄⁻ uptake were similar for the two H. ericae isolates (Table I). There was no significant difference in the uptake of H₂AsO₄⁻ between the heathland and mine site H. ericae isolates, as determined by ANOVA (data not shown). Growth in the presence of 5 μm H₂PO₄⁻ prior to uptake suppressed H₂AsO₄⁻ uptake in both the mine site and heathland isolates (Fig. 2). At 0.75 mm H₂AsO₄⁻ in inorganic phosphate-sufficient tissue, the rate of H₂AsO₄⁻ uptake was approximately 2 to 3 times lower than the rate of uptake in the absence of H₂PO₄⁻ (Fig. 2). Michaelis-Menten kinetic parameters could not be determined for isolates precultured on high concentration H₂PO₄⁻ media, and a linear regression was fitted to the data (Fig. 2).

Figure 2.

H₂AsO₄⁻ influx in the absence of H₂PO₄⁻ for mine site (▾, dashed line) and heathland (●, solid line) H. ericae. H₂AsO₄⁻ influx after growth in the presence of 5 mmol⁻³ H₂PO₄⁻ for mine site H. ericae (▿, dashed line) and heathland H. ericae (○, solid line). Bars = mean ± se (n = 3).

Figure 3.

H₂PO₄⁻ influx for mine site (▾, dashed line) and heathland (●, solid line) H. ericae. Bars = mean ± se (n = 3).

Table I.

Kinetic parameters for H₂AsO₄⁻, H₃AsO₃⁻, and H₂PO₄⁻ influx in mine site and heathland H. ericae

H. ericae Isolate	Species	Regression	Y0	A	Km	Vmax	r2	P
					mol m−3	μmol g−1 dry wt h−1
Mine	H3AsO3	Hyperbola			3.5 ± 4.6	0.86 ± 0.91	0.9736	<0.0001
Mine	H3AsO3	Linear	0.0068 ± 0.0068	0.19 ± 0.014			0.9729	<0.0001
Mine	H2AsO4	Hyperbola			0.14 ± 0.06	3.1 ± 0.5	0.9432	0.0058
Mine	H2PO4−	Hyperbola			0.055 ± 0.048	64.9 ± 13.9	0.7734	0.0209
Heathland	H3AsO3	Hyperbola			0.72 ± 0.60	0.40 ± 0.17	0.9297	0.0005
Heathland	H3AsO3	Linear	0.024 ± 0.016	0.22 ± 0.033			0.9014	0.0011
Heathland	H2AsO4	Hyperbola			0.28 ± 0.07	3.67 ± 0.37	0.9887	0.0005
Heathland	H2PO4−	Hyperbola			0.051 ± 0.017	79.12 ± 6.51	0.9571	0.0007

Figures represent kinetic parameters ± se (n = 3). P is the probability of the source term not being significant.

Increasing H₂PO₄⁻ concentrations up to 0.1 mm resulted in an increase in the rate of H₂PO₄⁻ uptake in both isolates (Fig. 3). At concentrations above 0.1 mm, H₂PO₄⁻ uptake was not further enhanced. Apparent kinetic parameters for both isolates were similar (Table I). There was no significant difference between the rates of H₂PO₄⁻ uptake in the heathland or mine site isolate as determined by ANOVA (Minitab, SPSS, Chicago).

In comparison with the rate of H₂AsO₄⁻ uptake, H₂PO₄⁻ uptake was much greater in both H. ericae isolates at all concentrations. Both isolates demonstrated a much higher V_max and a lower K_m value for H₂PO₄⁻ in comparison with H₂AsO₄⁻ (Table I), indicating a much higher affinity for H₂PO₄⁻ uptake.

Repression of H₂AsO₄⁻ and H₂PO₄⁻ Uptake

Repression of H₂AsO₄⁻ uptake by H₂PO₄⁻ and H₂PO₄⁻ uptake by H₂AsO₄⁻ was investigated at fixed concentrations. Uptake of 0.75 mm H₂AsO₄⁻ from solution in heathland and mine site H. ericae isolates was reduced by pre-exposure of fungal mycelium to H₂PO₄⁻ (5.0 μm H₂PO₄⁻). The initial rate of H₂AsO₄⁻ uptake in the absence of H₂PO₄⁻ in mine site H. ericae was 0.22 μmol g⁻¹ dry weight h⁻¹, which was lower than the initial rate of uptake in the heathland isolate (Fig. 4A). After 20 min of exposure to H₂PO₄⁻, a rapid decrease in H₂AsO₄⁻ uptake was observed for both isolates with mine site and heathland isolates demonstrating uptake rates of 0.14 and 0.29 μmol g⁻¹ dry weight h⁻¹, respectively. After 2 h of H₂PO₄⁻ uptake, H₂AsO₄⁻ uptake by both isolates was almost completely suppressed (Fig. 4A).

Figure 4.

A, Uptake of 0.75 mm H₂AsO₄⁻ in mine site (▾) and heathland (●) H. ericae after periods of exposure to 5.0 μm H₂PO₄⁻. Bars = mean ± se (n = 3). Second order decay curves were fitted to the data (Sigma Plot, Jandel Scientific). B, Uptake of 0.1 mm H₂PO₄⁻ in mine site (▾) and heathland (●) H. ericae after periods of exposure to 0.75 mm H₂AsO₄⁻. Bars = mean ± se (n = 3). Second order decay curves were fitted to the data (Sigma Plot, Jandel Scientific).

Uptake of 0.1 mm H₂PO₄⁻ was repressed by pre-exposure to 0.75 mm H₂AsO₄⁻ (Fig. 4B). Initially, the rate of H₂PO₄⁻ exposure in the heathland isolate was greater than the mine site isolate, however on exposure to H₂AsO₄⁻, rates of H₂PO₄⁻ uptake decreased in both isolates (Fig. 4B). After 20 min of exposure to H₂AsO₄⁻, the rate of H₂PO₄⁻ uptake was suppressed considerably, however, after 24 h of exposure to H₂AsO₄⁻, H₂PO₄⁻ uptake in both isolates was completely inhibited (Fig. 4B). There was no difference between the effects of H₂AsO₄⁻ on H₂PO₄⁻ uptake and H₂PO₄⁻ on H₂AsO₄⁻ uptake between the heathland and mine site isolate.

Efflux of As from Fungal Cells

In the methylation experiment, As was not present in the HgCl₂ traps for either mine site or heathland isolates. Because HgCl₂ complexes all volatile methylated arsines as well as AsH₃, this indicates that neither isolate methylated H₂AsO₄⁻ (data not shown).

Efflux of As from mine site H. ericae mycelia was more rapid than for the heathland isolate (Fig. 5). After 1 h of incubation in 0.75 mm H₂AsO₄⁻ (to load cells with As) and transfer to H₂AsO₄⁻-free media for time periods of up to 24 h, As concentration in the mine site isolate tissue decreased significantly (P < 0.001) (Fig. 5). After 5 h in H₂AsO₄⁻-free media, the mine site isolate lost 83% of its initial As concentration in comparison with the heathland isolate, which lost 13%, showing enhanced As cell efflux in the mine site H. ericae isolate. Similar trends were found after loading for 10 min, 20 min, and 4 h exposure to 0.75 mm H₂AsO₄⁻ (data not shown). Of the total As effluxed from cells, 71.6% was H₃AsO₃ with the remainder of As being lost as H₂AsO₄⁻. The majority of As lost from the heathland isolate similarly was in the form of H₃AsO₃ (71.3%). These results indicate an enhanced H₃AsO₃ efflux mechanism in the mine site H. ericae isolate.

Figure 5.

As content of mine site (▾) and heathland (●) H. ericae tissue over time as a percentage of total As content after incubation in 0.75 mm H₂AsO₄⁻. Bars = mean ± se (n = 3).

Uptake of H₂AsO₄⁻ over Time

The long-term uptake of H₂AsO₄⁻ was much greater in the heathland H. ericae isolate than the mine site isolate (Fig. 6). Accumulation of H₂AsO₄⁻ by the heathland isolate did not differ significantly from linearity with respect to time (r² = 0.973) over 2 h, however, after 2 h of exposure, decreased accumulation was observed (Fig. 6). This decrease in As accumulation after 2 h seems likely to reflect a response to H₂AsO₄⁻ toxicity. In contrast, while the mine site isolate demonstrated increased accumulation of As over 20 min there was no increase in accumulation after 2 h and this was sustained for up to 24 h.

Figure 6.

Long-term accumulation of 0.75 mm H₂AsO₄⁻ in mine site (▾) and heathland (●) H. ericae. Bars = mean ± se (n = 3).

Uptake of H₃AsO₃

The rates of H₃AsO₃ uptake by heathland and mine site H. ericae isolates were similar with uptake in both isolates increasing linearly in response to increasing H₃AsO₃ concentrations (Fig. 7). Although Michaelis-Menten functions could be fitted to the data, linear models demonstrated the best fits (Table I). Uptake of H₃AsO₃ in mine site and heathland H. ericae isolates at low H₃AsO₃ concentrations (0.01 mm H₃AsO₃) was 3- to 4-fold less than the rate of H₂AsO₄⁻ uptake at the same concentration (Figs. 2 and 6). At 0.75 mm H₃AsO₃, the rate of H₃AsO₃ uptake was 15 times less than the rate of H₂AsO₄⁻ uptake with both isolates have a lower affinity for H₃AsO₃ than H₂AsO₄⁻.

Figure 7.

H₃AsO₃ influx for mine site (▾) and heathland (●) H. ericae. Bars = mean ± se (n = 3). Linear regressions were fitted to the data (Sigma Plot, Jandel Scientific).

DISCUSSION

Populations of H₂AsO₄⁻-resistant H. ericae have been isolated from As/Cu mine soils (Sharples et al., 2000), where soil solution H₂AsO₄⁻ content is 20 to 75 times higher than natural heathland soils that have not been contaminated by mining and other industrial processes. The present study demonstrates increased H₂AsO₄⁻ and H₃AsO₃ resistance in an isolate of H. ericae obtained from an As/Cu mine site in comparison with an isolate from an uncontaminated heathland site. Growth of the heathland isolate was completely inhibited at 1.33 mm H₂AsO₄⁻, whereas the mine site H. ericae isolate continued growth at the highest concentration tested (13.3 mm) (Fig. 1A). H₃AsO₃ is usually considered to be the most toxic form of As available to plants and fungi, however both isolates of H. ericae demonstrated greater sensitivity to H₂AsO₄⁻ (Fig. 1), although rates of growth of both isolates were similar in the absence of As.

H₂AsO₄⁻ behaves as a H₂PO₄⁻ analog (Meharg and Macnair, 1990) and is accumulated by the H₂PO₄⁻ transport system in a wide range of organisms (Meharg and Macnair, 1992). Inorganic phosphate transport across membranes is carrier-mediated and described by Michaelis-Menten kinetics (Beever and Burns, 1980). Adaptation of the H₂PO₄⁻ uptake system is a mechanism of H₂AsO₄⁻ resistance in the grasses H. lanatus, D. cespitosa, and A. capillaris (Meharg and Macnair, 1990, 1991, 1992, 1994). In H. lanatus, H₂AsO₄⁻ resistance is achieved by constitutive suppression of the high-affinity uptake system by carrier synthesis inhibition and is independent of plant phosphorus status (Meharg and Macnair, 1992). The present study demonstrates similar apparent K_m and V_max parameters for H₂AsO₄⁻ transport and H₂PO₄⁻ transport in H₂AsO₄⁻-resistant and -non-resistant H. ericae (Table I). Mine site and heathland H. ericae apparent K_m values (0.14 ± 0.06 mm and 0.28 ± 0.07 mm, respectively) were higher than values previously reported for H₂AsO₄⁻-resistant H. lanatus (0.074 mm) (Meharg and Macnair, 1992) and other fungi (Sharples et al., 1999). For example, Saccharomycetes cerevisiae has a K_m for H₂AsO₄⁻ of 0.004 mm (Jung and Rothstein, 1965), and Candida tropicalis has a K_m value of 0.005 mm (Beever and Burns, 1980). The apparent V_max value for H₂AsO₄⁻ uptake obtained for H. ericae (Table I) was similar to those reported for S. cerevisiae (10.2 μmol g⁻¹ dry weight h⁻¹), lower than the ectomycorrhizal fungus Hebeloma crustuliniforme (Sharples et al., 1999), yet higher than those found in H₂AsO₄⁻-resistant H. lanatus (Meharg and Macnair, 1990).

Both isolates demonstrated similar H₃AsO₃ transport kinetics, however influx was markedly lower than when compared with the rate of H₂AsO₄⁻ uptake (Figs. 2 and 7). This decrease was expected because H₃AsO₃ is not a H₂PO₄⁻ analog and is therefore not transported by H₂PO₄⁻ carriers. It is not known how H₃AsO₃ enters cells, however the present study seems to indicate passive diffusion (Fig. 7).

Pregrowth of H. ericae in the presence of 5 mm H₂PO₄⁻ significantly suppressed the uptake of H₂AsO₄⁻ for both mine site and heathland isolates (Fig. 2). Suppression of H₂AsO₄⁻ uptake by long-term exposure to H₂PO₄⁻ has also been demonstrated in the plant Hordum vulgare and Silene vulgaris (Lee, 1982; Paliouris and Hutchinson, 1991). In the case of H. vulgare, plants grown in the presence and absence of 0.5 mm H₂PO₄⁻, demonstrated H₂AsO₄⁻ uptake rates of 27.7 and 81.6 nmol g⁻¹ fresh weight h⁻¹, respectively (Lee, 1982). Meharg and Macnair (1991) suggest that H₂AsO₄⁻ resistance in H. lanatus is due to a decrease in the concentration of protein carriers in the plasma membrane rather than a change in the carrying capacity of the protein.

H₂AsO₄⁻ uptake was rapidly repressed on exposure to H₂PO₄⁻ in both mine site and heathland H. ericae (Fig. 4). Similarly, the presence of H₂AsO₄⁻ rapidly repressed H₂PO₄⁻ uptake (Fig. 4). Rapid repression of the high affinity H₂PO₄⁻ uptake system in plants under high plant H₂PO₄⁻ status has long been reported (Meharg and Macnair, 1992). The nature of this repression differs for different species. Repression may occur by a decrease in V_max with little or no change in the K_m, which occurs in algae (McPharlin and Bieleski, 1987), bacteria (Beever and Burns, 1980), and selected plants (Anghinoni and Barber, 1980; Lee, 1982; Cogliatti and Santa Maria, 1990; Jungk et al., 1990). Repression of the high affinity uptake system in the vesicular arbuscular mycorrhizal fungus Gigaspora margarita is achieved by increase in the apparent K_m with little change in the apparent V_max (Thompson et al., 1990), whereas repression in the plant, Solanum tuberosm, is achieved by both an increase in apparent K_m and a decrease in apparent V_max (Cogliatti and Clarkson, 1983). Changes in H₂PO₄⁻ uptake with changing H₂PO₄⁻ status may be under allosteric control (Levebvre and Glass, 1982; Schorring and Jensen, 1984) and by the synthesis and breakdown of transport sites (Jeanjean, 1973; Drew et al., 1984). Because the speed of repression is too rapid to be explained by protein turnover, under short exposure times it is likely that for H. ericae both H₂PO₄⁻ and H₂AsO₄⁻ act allosterically.

Short-term uptake of H₂AsO₄⁻ was similar between isolates, however in the longer term, accumulation of As by the heathland isolate decreased significantly in 24 h (Fig. 6). Such a decrease in the rate of As accumulation in the heathland isolate may reflect cell death in response to H₂AsO₄⁻ toxicity. Arsenate causes toxicity in fungi and plants by interfering with aerobic phosphorylation, following intracellular reduction of H₂AsO₄⁻ to H₃AsO₃, which then breaks down protein sulfydryl groups (Ullrich-Eberius et al., 1989).

The present study demonstrates the ability of an isolate of H. ericae from a mine site to efflux H₃AsO₃ from its hyphae (Fig. 5), which may provide H₂AsO₄⁻ resistance to this isolate. H₃AsO₃ efflux has been reported as a mechanism of H₂AsO₄⁻ resistance in the bacterium S. aureous (Broer et al., 1993) and the yeast S. cerevisiae (Wysocki et al., 1997) and contrasts to the mechanism of resistance reported in higher plants (Meharg and Macnair, 1990). Arsenate resistance in S. cerevisiae is mediated by an H₃AsO₃ transporter (Wysocki et al., 1997), and in the case of S. aureous, intracellular H₂AsO₄⁻ is reduced to H₃AsO₃ before being actively exported from the cells (Broer et al., 1993). The present study suggests a similar mechanism of As resistance in H. ericae at As-contaminated mine sites. The steady state of As accumulation observed in the mine site isolate after 20 min is probably not due to a suppression of the high affinity uptake system but rapid internal reduction of H₂AsO₄⁻ to H₃AsO₃, which then initiates efflux of H₃AsO₃ from the hyphae. The ability of the mine site H. ericae isolate to efflux H₃AsO₃from cells into the surrounding soil indicates a need for enhanced resistance to H₃AsO₃. Arsenate was much more toxic to H. ericae than H₃AsO₃ (Fig. 1), which supports efflux of H₃AsO₃ as the mechanism of H₂AsO₄⁻ resistance in H. ericae.

The mechanism of H₂AsO₄⁻ resistance we have described in H. ericae is likely to be of ecological importance for its host plant (C. vulgaris) on contaminated mine sites. Arsenite efflux enables the fungus to retain its ability to transport inorganic phosphate from the soil (much of which will, in turn, is transferred to the host plant), whereas effluxing absorbed H₂AsO₄⁻. The fungus may thus act as a filter to maintain low plant As levels, while maintaining an adequate supply of phosphorus to the host (Sharples et al., 2000). Efflux of H₃AsO₃ from the fungal cells into the soil also ensures that re-absorption of As from the soil is limited.

MATERIALS AND METHODS

Fungal Material

The arsenate-resistant Hymenoscyphus ericae genotype was isolated from roots of Calluna vulgaris obtained from the Gawton United mine (Devon, S.W. UK) whereas the non-resistant genotype was obtained from C. vulgaris roots from an uncontaminated natural heathland site at Aylesbeare Common (Devon, S.W. UK). These fungal isolates were randomly selected from approximately 25 isolates from each site and were previously identified by PCR-RFLP analysis (Sharples et al., 2000). Although the characteristics of arsenate absorption and arsenite efflux have only been studied in detail for single isolates from each population, preliminary experiments for absorption and efflux by further isolates indicate consistent patterns for absorption/efflux within the mine site or heathland populations (data not shown). The fungi were maintained on modified Melin Norkrans agar medium (MMN) containing: 3.79 mm (NH₄)₂HPO₄; 2.21 mm KH₂PO₄; 0.57 mm MgSO₄·7H₂O; 0.23 mm CaCl₂·6H₂O; 0.43 mm NaCl; 0.034 mm FeEDTA; 55.5 mm d-Glc; and 0.3 μm thiamine, adjusted to pH 5.5, and incubated at 25°C. In H₂AsO₄⁻ and H₃AsO₃ uptake experiments, mycelia were grown in liquid MMN for 17 d and transferred to H₂PO₄⁻-free MMN for 48 h prior to uptake analysis. To determine the effects of H₂PO₄⁻ on H₂AsO₄⁻ uptake, mycelia were grown in liquid MMN containing 5 μm H₂PO₄⁻ for 17 d before H₂AsO₄⁻ uptake. Mycelia for H₂PO₄⁻ uptake studies were grown on liquid MMN containing 0.01 mm H₂PO₄⁻ for 17 d at 25°C.

Effect of H₂AsO₄⁻ and H₃AsO₃ on Biomass Production

Two plugs (6-mm diameter) of each H. ericae isolate were cut from the edge of actively growing mycelia on MMN and inoculated into 9-cm-diameter Petri dishes containing 25 mL of liquid MMN. After 11 d of incubation at 25°C, fungal plugs were transferred to 25 mL of liquid MMN supplemented with either H₂AsO₄⁻ or H₃AsO₃, supplied as Na₂HAsO₄ and NaAsO₂, respectively, at concentrations of 0, 0.67, 1.33, and 4.67 mm. For all treatments, H₂PO₄⁻ concentration in the medium was adjusted to 0.01 mm. After 7 d of incubation, mycelial mats were removed from basal medium, dried overnight (80°C), and the biomass increase determined gravimetrically. All treatments were replicated three times.

Kinetics of H₂AsO₄⁻ and H₂PO₄⁻ Uptake

To determine H₂AsO₄⁻, H₃AsO₃, and H₂PO₄⁻ uptake, three replicate mycelial mats of each isolate were incubated in 25 mL of aerated test solution for 20 min (except when uptake was determined with respect to time). Test solutions contained 10 mm 2-(N-morpholino) ethanesulfonic acid (MES), 0.5 mm Ca(NO₃)₂, and different concentrations of either H₂AsO₄⁻, H₃AsO₃, or H₂PO₄⁻ in the form of Na₂HAsO₄·7H₂0, NaAsO₂, and Na₂HPO₄, respectively. In the experiments to determine the rate of H₂PO₄⁻ uptake, [³²P] (as NaH₂PO₄, supplied by Amersham) was added to give an activity of 37 kBq mL⁻¹. Using the methodology of Meharg and Macnair (1990), uptake was terminated by rinsing mycelia in 25 mL of an ice-cold solution containing 1 mm Na₂HPO₄, 10 mm MES, and 0.5 mm Ca(NO₃)₂. Mycelia were transferred to 25 mL of an aerated ice-cold solution of the same composition for 10 min to ensure desorption of H₂AsO₄⁻, H₃AsO₃, or [³²P] from the cell-free space. Mycelia were dried (80°C, 24 h) and biomass determined gravimetrically before analysis.

Repression of H₂PO₄⁻ and H₂AsO₄⁻ Uptake

To investigate the effects of H₂PO₄⁻ on H₂AsO₄⁻ uptake, three replicate mycelial mats of each isolate were pre-incubated in liquid MMN containing 5 μm H₂PO₄⁻ for 0, 20, 60, 120, 240, or 1,440 min prior to H₂AsO₄⁻ uptake (as described above). The effects of H₂AsO₄⁻ on H₂PO₄⁻ uptake were determined by pre-incubating three replicate mycelia in 0.75 μm H₂AsO₄⁻ for 0, 20, 60, 120, 240, or 1,440 min before exposure to [³²P]-uptake solution.

Methylation and Efflux of As by Fungi

Methylation of H₂AsO₄⁻ was investigated using a modified method of Gates et al. (1997), involving the chemofocussing of volatile As species on mercuric chloride. Conical flasks were inoculated with 20 mL of a solution containing 10 mm MES, 0.5 mm Ca(NO₃)₂, and 0.67 mm H₂AsO₄⁻ for mine site H. ericae mycelia and 0.27 mm H₂AsO₄⁻ for heathland H. ericae. These H₂AsO₄⁻ concentrations were the approximate H₂AsO₄⁻ EC₅₀ values of the mine and heathland population (preliminary data by Sharples et al., 2000) (data not shown). Polyurethane plugs were soaked in 0.1 m HgCl₂ and oven dried at 50°C for 12 h before being placed inside a glass condenser fitted to the conical flasks. Three replicate flasks were set up for each isolate. The presence of As was indicated by brown discoloration of the HgCl₂ plugs and after 24 h of incubation, HgCl₂ plugs were removed and analyzed for As by atomic absorption spectrometry.

To investigate the mechanism of H₂AsO₄⁻ resistance, mycelia of each isolate (n = 3) were exposed to H₂AsO₄⁻ uptake solution for 10 min, 20 min, 1 h, 4 h, or 24 h. After termination of uptake, mycelia were transferred to 25 mL of liquid MMN containing no H₂PO₄⁻ for 0, 30 min, 90 min, 5 h, or 24 h. Mycelia were then dried (80°C, 24 h) and biomass determined gravimetrically before As analysis.

Speciation of As

Three replicate mycelia of each isolate were incubated in 0.75 mm H₂AsO₄⁻ uptake solution for 1 h. Uptake was terminated and mycelia transferred to a 2-mL test tube containing 1 mL of H₂PO₄⁻-free liquid MMN. Fresh liquid MMN was continually pumped into and out of the test tube at a flow rate of 0.7 mL min⁻¹ and removed from the tube at the same rate for 5 h. After 5 h, fungal material was dried, digested, and analyzed for As. Two milliliters of MMN pumped from the test tube was also analyzed for H₂AsO₄⁻ and H₃AsO₃ using atomic absorption spectrometry (Glaubig and Goldberg, 1988).

Analysis

To determine [³²P]H₂PO₄⁻ uptake, dried fungal mycelia were placed in 20-mL glass scintillation vials, to which 10 mL of deionized water was added. [³²P]-activity was determined by Cherenkov counting using a Tri Carb 2100TR liquid scintillation counter (Packard Bell, Sacramento, CA) with data corrected for quenching.

As was determined by digesting mycelia in 2 mL of concentrated nitric acid (Aristar grade) using a block digester, for 1 h at 120°C followed by 1 h at 180°C, to evaporate the samples to dryness. The As residue was redissolved in 20 mL of a solution containing 5% (v/v) HCl (Analar grade) containing 20 mm potassium iodide. The amount of As present in the digests was determined using hydride generation interfaced with an atomic absorption spectrometer (ThermoUnicam Solaar 929, Cambridge, UK). As species were determined using pH selectivity, H₂AsO₄³⁻ reduced to arsine (AsH₃) by NaBH₄ at pH < 6, whereas H₃AsO₃ reduced to AsH₃ at pH 7.

Statistical Analysis

Data were analyzed by ANOVA using the computer package Minitab v. 11 (Minitab, State College, PA). Curve fitting was carried out using the fitting regimes within the computer package Sigma Plot (Jandel Scientific, Erkrath, Germany), which uses the Marquardt non-linear curve fitting algorithm (Marquardt, 1963).