Notice: Wiley Online Library will be unavailable on Saturday 27th February from 09:00-14:00 GMT / 04:00-09:00 EST / 17:00-22:00 SGT for essential maintenance. Apologies for the inconvenience.
• To test the response of arbuscular mycorrhizal (AM) fungi to a difference in soil pH, the extraradical mycelium of Scutellospora calospora or Glomus intraradices, in association with Plantago lanceolata, was exposed to two different pH treatments, while the root substrate pH was left unchanged.
• Seedlings of P. lanceolata, colonized by one or other of the fungal symbionts, and nonmycorrhizal controls, were grown in mesh bags placed in pots containing pH-buffered sand (pH around 5 or 6). The systems were harvested at approximately 2-wk intervals between 20 and 80 d.
• Both fungi formed more extraradical mycelium at the higher pH. Glomus intraradices formed almost no detectable extraradical mycelium at lower pH. The extraradical mycelium of S. calospora had higher acid phosphatase activity than that of G. intraradices. Total AM root colonization decreased for both fungi at the higher pH, and high pH also reduced arbuscule and vesicle formation in G. intraradices.
• In conclusion, soil pH influences AM root colonization as well as the growth and phosphatase activities of extraradical mycelium, although the two fungi responded differently.
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
In nature, there is a large variation in plant growth and phosphorus uptake in response to colonization by different arbuscular mycorrhizal (AM) fungi (Ravnskov & Jakobsen, 1995; van der Heijden et al., 1998a,b). The AM fungal isolates differ in their phosphorus uptake efficiency (Jakobsen et al., 1992a; Joner & Jakobsen, 1994), and in their supply of phosphorus to their host (Jakobsen et al., 1992b; Pearson & Jakobsen, 1993). Most AM fungi increase the phosphorus content of plants, although some, such as Scutellospora calospora, are often rather inefficient in phosphorus transport to the host plant, and in some plants no, or only a small, growth response can be seen (Jakobsen et al., 1992a). These variations may be the result of differences in fungal root colonization or physiological characteristics, but the ability of different AM fungal isolates to produce different amounts of extraradical mycelium may also be of importance (Jakobsen et al., 1992a; Boddington & Dodd, 1998). The positive effects of AM fungi on plant growth have often been related to an increased uptake of less mobile nutrients, especially phosphorus (Bolan, 1991), from soil regions not accessible to the root hairs. Whether AM fungal phosphatase activities contribute significantly to the phosphorus nutrition of AM plants is still unclear (Joner et al., 2000).
The observed differences in plant response to AM colonization may not only result from the use of different plant hosts or fungal isolates, but may also result from variations in environmental conditions. For example, three Glomus isolates exhibited similar efficiencies in mineral acquisition when grown in acidic soils, but their efficiencies differed when they were grown in alkaline soil (Clark & Zeto, 1996b). Soil pH is known to have considerable influence on plant growth in particular because it influences the mobilization, and thus the availability, of nutrients (Haynes, 1990; Marschner, 1995). Also, root colonization and the ability to form extraradical mycelium were changed when AM fungi were grown with their host plants in soils of different pH (Abbott & Robson, 1985; Porter et al., 1987). However, in these studies, the AM fungal extraradical mycelium and the plant roots were both exposed to the pH treatment.
To test the response of AM fungi specifically to a difference in soil pH, we allowed the hyphae of S. calospora and Glomus intraradices, in association with Plantago lanceolata, to grow out into root-free compartments at different pH values. We studied not only the amount of extraradical mycelium formed at different pH values, but also its phosphatase activities. Since effects of a pH treatment may also be caused indirectly by the interactions with saprotrophic fungi and bacteria (Olsson et al., 1998; Ravnskov & Jakobsen, 1999), we also estimated their biomass. Furthermore, we studied the possible influence of the different pH treatments on total AM colonization of the roots and the distribution of internal fungal structures.
Materials and Methods
Plantago lanceolata L. was grown in soil-filled mesh bags (root compartment) placed centrally in a pot and surrounded with quartz sand (hyphal compartment). Four hundred grams of acid-washed sand, at a pH adjusted to 5 or 6, were used for each pot. To establish the lower pH of the sand, it was flushed with 1.0 l of a 1 mm 2-(N-morpholino)ethanesulphonic acid (MES) buffer adjusted to pH 4 over a 24-h period. For the higher pH, a MES buffer adjusted to pH 7 was used. The sand was air-dried to a water-holding capacity of 60%. The growth medium used for the root compartment was a mixture of a moraine clay loam (Jakobsen & Nielsen, 1983) and river sand (1 : 1), which had been irradiated (10 kGy) to eliminate indigenous AM fungi. Inoculum (soil–root mixture) of either S. calospora (Nicol. & Gerd.) Walker & Sanders (BEG 43) or G. intraradices Schenck & Smith (BEG 87) was mixed into the growth medium in the proportions 1 : 14. Scutellospora calospora was originally isolated from a Mediterranean grassland with a pH of 4.4 in Western Australia and G. intraradices from a temperate grassland with a pH of 6.0 in Denmark. One-hundred and fifty grams of the soil/inoculum mixture were added to the mesh bags with a mesh size of 20 µm. For the non-mycorrhizal controls, 150 g of the soil without inoculum were used. Nutrients were added to the growth medium as described by Olsson et al. (1996). Bacterial inoculum was added as a soil extract from pots with nonmycorrhizal Cucumis sativus L. plants. The mesh bags were moistened with distilled water to a water-holding capacity of 60%.
Seeds of P. lanceolata L. were surface-sterilized for 15 min in 5% calcium hypochlorite solution, washed with water and transferred to Petri dishes containing moistened filter paper. The seeds were pregerminated at room temperature for 7 d in the dark and subsequently for 4 d in daylight. After germination, four seedlings were transplanted into each mesh bag, and the surface of the pots was covered with plastic pellets to reduce evaporation. The plants were grown in a greenhouse at an average temperature of 28°C during the day, and 20°C at night (16-h day, photosynthetic photon flux density of 150 µmol m−2 s−1 and a relative humidity of 60%). The system was maintained at a water-holding capacity of 60% by watering the root compartment with distilled water, and the hyphal compartment with a 0.5 mm MES buffer, set at either pH 4 or pH 7, which maintained the pH of the sand at 5 or 6, respectively (Table 1). Sets of four pots of each fungal (nonmycorrhizal, S. calospora or G. intraradices) at each pH (low or high) treatment were randomly distributed over plastic trays with transparent covers to prevent contamination between treatments. These trays, containing in total 90 pots, were distributed at random in the glasshouse. Pots and growth propagators were regularly redistributed.
Table 1. The pH of the hyphal and root compartments at the different harvesting times
pH (in H2O; n = 3)
Means marked with an asterisk are significantly different (P < 0.05) for low- and high pH treatment (two-tailed t-test with unequal variances). Low, low pH treatment; High, high pH treatment.
At 20, 32, 53, 67 and 80 d after transplantation, three randomly taken pots from each treatment were harvested. Some subsamples of the sand and soil were frozen in order to be able to determine the fatty acid content later. Other subsamples were immediately used to determine the pH (in H2O; 25 g fresh weight substrate in 50 ml distilled water, shaken at 200 rotations min−1 for 2 h, followed by 15 min of sedimentation). A third set of subsamples was used to determine the acid phosphatase (ACP) and alkaline phosphatase (ALP) activities. The remaining sand was used to collect the extraradical mycelium by wet sieving (van Aarle et al., 2001). If no hyphae could be detected by eye, the decanted liquid was checked under a dissecting microscope for hyphae, and these were collected with forceps. The fresh weight of the hyphae was measured, after which it was used to study the ACP and ALP activity. Plants were collected and washed clean of soil. Dry weights were determined for shoots and for subsamples of roots. The total root dry weight was estimated by calculations based on the fresh weight proportion. Another sample of the roots was used to study the possible ALP activity, while a third sample was used to assess the mycorrhizal colonization.
Phosphatase activity measurements
Alkaline and acid phosphatase activities of the extraradical mycelium and phosphatase activity of mycorrhizal roots were visualized with enzyme-labelled fluorescence substrate (ELF, Molecular Probes, Eugene, USA), which is a fluorogenic phosphatase substrate (van Aarle et al., 2001). The overall phosphatase activity of the mycorrhizal mycelium was assessed with a light microscope according to a relative scale ranging from no activity (denoted 0) to high activity (denoted 5) in all of the mycorrhizal hyphae. At each harvest, roots of one replicate of each fungal treatment and each pH were hand-sectioned and stained with alkaline ELF buffer solution to visualize phosphatase-active intraradical hyphae.
To measure ACP and ALP activity of the sand and soil, a modified procedure based on that of Tabatabai & Bremner (1969) was used. Phosphatase activities were determined spectrophotometrically using p-nitrophenyl phosphate as substrate (Sigma Chemical Co., St. Louis, MO, USA). Five grams of substrate were added to 40 ml of distilled water and shaken at 200 revolutions min−1 for 15 min. Aliquots of the suspension, 20 µl for ALP and 40 µl for ACP activity determination, were incubated in 100 µl p-nitrophenyl phosphate solution (4 mg p-nitrophenyl phosphate ml−1 in deionized water) and 100 µl buffer solution. For the ALP determination alkaline buffer solution was used (1.5 mol 2-amino-2-methyl-1-propanol l−1, pH 10.3; Sigma), and for the ACP determination citrate buffer (90 mM citrate and 10 mM chloride, pH 4.8; Sigma). Adding 2 ml 0.05 m NaOH to the ALP solution or 1 ml 0.1 m NaOH to the ACP solution after 2 h stopped the reaction. Substrate suspensions of each of the fungal treatments were deactivated by heating to boiling three times in a microwave oven and these were used as substrate blanks to determine background levels of the substrate. The addition of distilled water instead of the substrate solution to the buffer solution produced blanks to check the background of the p-nitrophenyl phosphate substrate. The absorbance was measured at 420 nm after filtration through a 0.45-µm hydrophilic PTFE filter (Lida, Werner-Glas & Instrument AB, Kungsängen, Sweden). Enzyme activities (expressed as U g−1 dry wt of substrate) were calculated after correction using the blanks. One unit of phosphatase activity, under the specified conditions, was defined as the amount of enzyme activity that liberated 1 µmol of p-nitrophenol in 1 h.
Fatty acid analysis as a biomass indicator
Lipids were extracted from sand (10 g fresh weight) and soil (5 g fresh weight) samples by vortexing for 15 s with a one-phase chloroform–methanol–citrate buffer (Olsson et al., 1997). Lipid extracts were separated from soil pellets after centrifugation at 3000 g. The lipids were fractionated into neutral lipids, glycolipids and phospholipids on prepacked silica columns (100 mg sorbent mass, Varian, Harbor City, USA) by eluting with 1.5 ml chloroform, 6 ml acetone and 1.5 ml methanol, respectively. The fatty acid residues of the neutral lipids and phospholipids were transformed into free fatty acid methyl esters, which were identified and quantified using gas chromatography, according to the method described by Frostegård et al. (1993).
The nomenclature of the fatty acids follows that used by Tunlid & White (1992). Signature phospholipid fatty acids indicate the biomass of specific groups of microorganisms (Tunlid & White, 1992). Phospholipid fatty acids 16:1ω5, 18:1ω7c, 20:4 and 20:5 were analysed as possible indicators of AM fungal biomass (Olsson & Johansen, 2000). These phospholipid fatty acids also occur in varying amounts in other soil microorganisms (Olsson, 1999). Neutral lipid fatty acids 16:1ω7c, 16:1ω5, 18:1ω7c, 20:4 and 20:5 were also considered since they are common in storage lipids of AM fungi (Graham et al., 1995; Olsson & Johansen, 2000) and provide sensitive indicators of AM fungi (Olsson, 1999). The neutral lipid fatty acid 16:1ω7c may constitute a large part of the neutral lipid fatty acids of G. intraradices, while the reason for not including phospholipid fatty acid 16:1ω7c is that it has been shown not to be an important part of AM fungal phospholipid fatty acids (Johansen et al., 1996). Phospholipid and neutral lipid fatty acid 18:2ω6,9 was used as an indicator of the biomass of saprotrophic fungi, and 10 bacteria-specific phospholipid fatty acids (i15:0, a15:0, i16:0, 10Me16:0, i17:0, a17:0, cy17:0, 10Me17:0, 10Me18:0 and cy19:0) were used as indicators of bacterial biomass (Frostegård & Bååth, 1996).
Root samples were stained with Trypan blue according to a modification of the procedure of Phillips & Hayman (1970). Total AM fungal root colonization, arbuscular and vesicular colonization were determined as the percentage of root length colonized with the magnified intersections method, as described by McGonigle et al. (1990). Root length colonized by fine endophytes (Thippayarugs et al., 1999) was quantified independently of the AM fungal colonization.
Averages, standard errors and t-tests were calculated using Microsoft Excel 2000. Data on dry weight, fatty acid contents, phosphatase activities in the substrate and percentage colonization were subjected to anova, while the visual estimates of phosphatase activity of extraradical mycelium were subjected to Paired sign and Kruskal–Wallis tests, using statview 5.0.1 (SAS Institute Inc., Cary, NC, USA). The time factor of the anova is not reported in most cases since its significance is obvious already from the graphs.
The pH of the hyphal compartment was significantly different between the pH treatments for all, except for the nonmycorrhizal plants at 53 d (Table 1). The pH of the root compartment showed only occasional significant differences between the pH treatments and, at the last harvest, no differences were found between the pH treatments. In none of the treatments could a difference in pH (two-tailed t-test with unequal variances) between the first and the last harvest be detected for the hyphal compartment, whereas the pH of the root compartment was significantly lower (two-tailed t-test with unequal variances) at the last harvest from both pH treatments. The experimental system thus seems to have worked according to our intentions.
Growth of the extraradical mycelium
The AM fungal signature neutral lipid fatty acid 16:1ω5 was detected in both the hyphal and the root compartment for both the S. calospora and G. intraradices treatments, but not the nonmycorrhizal samples (Fig. 1a–f). Neutral lipid fatty acid 16:1ω7 was detected in the hyphal compartment only for the G. intraradices treatment. These neutral lipid fatty acids increased significantly over time and in the hyphal compartment a significantly greater increase was observed at the higher pH than at the lower pH. The amounts of neutral lipid fatty acids were higher in the root compartment than in the hyphal compartment, and in the root compartment significantly more of the neutral lipid fatty acids were found with G. intraradices treatment than with S. calospora or nonmycorrhizal treatments. The nonmycorrhizal treatments did not show any changes in neutral lipid fatty acids over time. No mycelium was found by wet sieving at the first two harvests, while at the third harvest (53 d), mycelium could be collected from the sand of the hyphal compartment from both the S. calospora and G. intraradices treatments (except for the low pH treatment of G. intraradices). There was large variation in the amount of extraradical mycelium between replicates, but generally the amount of extraradical mycelium increased over time. No mycelium was recovered from the nonmycorrhizal treatment. The pH treatment did not influence the amount of extraradical mycelium formed in the root compartment while it did affect the amount in the hyphal compartment.
The content of the AM fungal signature phospholipid fatty acid (16:1ω5) in the root compartment (Fig. 1g–i) was roughly similar for S. calospora and G. intraradices treatments and slightly, but significantly, higher than in the nonmycorrhizal treatment. The background level of phospholipid fatty acid 16:1ω5 in the nonmycorrhizal treatment can be attributed to bacteria (Olsson, 1999). With S. calospora, the low pH-treated system exhibited a significantly higher content of phospholipid fatty acid 16:1ω5 than the high pH treatment. In the hyphal compartment the phospholipid fatty acid 16:1ω5 could not be detected. Other possible AM fungal signature fatty acids (18:1ω7c, 20:4 and 20:5) were not higher in S. calospora and G. intraradices treatments than in nonmycorrhizal treatment at the first four harvests. These were therefore not used as indicators of extraradical mycelium growth in this study.
Arbuscular mycorrhizal fungal root colonization
To examine the influence of pH on mycorrhizal colonization only the data from the last three harvests, where mycelium was present in the hyphal compartment, were included in the two-way anova. Total root colonization by S. calospora was significantly higher in the low pH treatments from 53 to 80 d (Fig. 2a). Root colonization by S. calospora at 80 d was low but still increasing, and at the last harvest an average of 23% and 18% of the total root length had been colonized for low- and high pH treatment, respectively. The highest root colonization by G. intraradices was at 67 d, there being 84% and 79%, respectively, for low pH and high pH treatment (Fig. 2b). Total colonization was significantly higher for the low pH treatment between 53 d and 80 d. Roots colonized by G. intraradices were also colonized by a fine endophyte (Thippayarugs et al., 1999). The fine endophyte colonization, estimated independently from the colonization by G. intraradices, appeared first after 20 d, and increased after that linearly over time. The increase was similar for both pH treatments, that is 58% and 44% of the root length for low (R2 = 0.94) and high (R2 = 0.84) pH treatment at 80 d, respectively. Neither the roots colonized by S. calospora nor nonmycorrhizal roots showed any colonization by fine endophytes.
Arbuscular colonization by S. calospora occurred in 9% of the total root length for both of the pH treatments at 80 d (Fig. 2c). This fungus does not form any vesicles. Glomus intraradices showed significantly higher root lengths with arbuscules and vesicles for the low pH treatment (Fig. 2d,f). Root length with arbuscules and vesicles reached the maximum estimated value at 53 d. The maximum root length showing arbuscular formation by G. intraradices was 40% of total root length for both pH treatments and for vesicles the values were 20% and 16% for the low pH and high pH treatments, respectively.
The phosphatase activities of the extraradical mycelium, as visualized by the ELF substrate, were, except for a few extremes, constant for all samples within each treatment. In Table 2 the data from each fungus at each pH treatment were averaged and the minimum and maximum are also given. Both ACP (P < 0.001) and ALP (P = 0.15) activities were higher for the extraradical mycelium of S. calospora than G. intraradices. The S. calospora extraradical mycelium seemed to have higher phosphatase activity at low pH than at high pH and the ACP activity was higher than the ALP activity (P = 0.016). This contrasts with G. intraradices where ALP seemed to have the higher activity (P = 0.22). For both pH treatments the phosphatase activity of the intraradical mycelium of both species (n = 1), as determined by staining with alkaline ELF substrate solution, increased over time and reached a maximum at 67 d. Overall, the roots colonized by G. intraradices showed more phosphatase active intraradical mycelium than roots colonized by S. calospora, and maximum values estimated were 2 and 4 (according to a relative scale) respectively.
Table 2. Phosphatase activity of the extraradical mycelium of two arbuscular mycorrhizal fungi, Scutellospora calospora and Glomus intraradices
Low (n = 9)
High (n = 8)
High (n = 9)
Alkaline (ALP) and acid (ACP) phosphatase activities, from 53 to 80 d, were estimated after staining with the enzyme-labelled fluorescence substrate; 0 = no activity to 5 = high activity. Average values are given with the minimum and maximum values in parentheses. Low, low pH treatment; High, high pH treatment; nd, not determined since no hyphae were found in the hyphal compartment.
In general, the phosphatase activities of the substrates in the hyphal and root compartment were for both fungi similar to those of the nonmycorrhizal samples (results not shown). Phosphatase activities were occasionally found in the hyphal compartment, and the highest values detected were 2.2 U g−1 for ALP, and 1.4 U g−1 for ACP. In the root compartment ACP activity was normally higher than in the hyphal compartment, where the highest ACP activity detected was 12.3 U g−1, while the highest ALP activity was only 6.1 U g−1.
Growth of plants, bacteria and saprotrophic fungi
Neither the plants colonized by S. calospora or G. intraradices, nor the nonmycorrhizal plants showed any significant difference in total plant or shoot dry weight between the high pH and the low pH treatment (Table 3). The S. calospora-colonized plants had a significantly higher root dry weight at the higher pH, but Glomus intraradices-colonized roots were significantly heavier only at the last harvest in the high pH treatment. Shoot, root and total plant dry weight of G. intraradices-colonized plants were significantly lower than those of plants colonized by S. calospora or nonmycorrhizal plants (three-way anova).
Table 3. The average shoot and root dry weight of Plantago lanceolata, measured at the different harvesting times
Plant dry wt (g; n = 3)
Low, low pH treatment; High, high pH treatment.
pH × time
pH × time
The bacterial biomass, as indicated by bacteria-specific phospholipid fatty acids, increased significantly over time for all fungal treatments in the hyphal compartment (two-way anova, Fig. 3). Both G. intraradices and S. calospora seemed to depress bacterial colonization of the hyphal compartment at higher pH (Fig. 3). The pH treatment had no effect on bacterial biomass in the non-mycorrhizal treatment (Fig. 3, Table 4). For the root compartment only the S. calospora treatment showed a difference between the pH treatments: amounts of bacteria-specific phospholipid fatty acids were greater at the lower pH (two-way anova, Table 4). In the root compartment a significantly higher bacterial biomass was found for S. calospora- than for G. intraradices-treated pots. The bacterial biomass of the S. calospora treatment significantly increased in the root compartment over time. Both S. calospora and G. intraradices treatments showed significantly higher amounts of bacteria-specific phospholipid fatty acids in the root compartment than the nonmycorrhizal treatment (three-way anova), and a considerable increase was observed in the last harvest.
Table 4. Fatty acids indicating bacterial and saprotrophic fungi (SE in parentheses) at 80 d in hyphal and root compartments
Fatty acids (nmol g−1 dry wt)
Saprotrophic fungi Neutral lipid
Low, low pH treatment; High, high pH treatment. aNo saprotrophic fungal fatty acid could be detected.
Low amounts of saprotrophic fungal biomass, as indicated by neutral lipid and phospholipid fatty acid 18:2ω6,9, were detected in the hyphal compartment. Saprotrophic fungi were only occasionally detected in the hyphal compartment by microscopic observations. In the root compartment the biomass of saprotrophic fungi increased significantly over time for all fungal treatments (see Table 4 for values at 80 d). For S. calospora, it was initially relatively constant but at the final harvest a marked increase was observed.
This study demonstrated an effect of pH on AM fungal hyphae even though the roots of the host plant were not exposed to the same pH. In earlier studies where the effect of pH on AM hyphal development, root colonization and plant growth has been examined, both symbionts were subjected to the same pH (Abbott & Robson, 1985; Clark & Zeto, 1996a). By using mesh bags we created a hyphal compartment that excluded the host roots. Here, we have shown that the pH of a substrate colonized by the AM fungal mycelium influences not only the growth of the mycelium but also the AM colonization of the host plant roots.
The increased extent of root length with vesicles and arbuscules in G. intraradices at the lower pH could indicate a stress response, especially since no hyphae were found in the hyphal compartment (Figs 1a–c and 2d,f). Siqueira et al. (1984) found no correlation between total root colonization and pH of the soil, and they suggested that the effect of the pH might be greater on the fungus than on the host plant. Other studies (Medeiros et al., 1994; Clark & Zeto, 1996a) indicate that AM fungal colonization can be strongly affected by pH per se, although the effects on different fungal isolates differed. In this study the amount of extraradical mycelium in the hyphal compartment did not appear to be related to the fungal colonization of the roots. However, a large amount of AM fungal neutral lipid fatty acids in the root compartment for G. intraradices was correlated with high AM fungal colonization, and for S. calospora both neutral lipid fatty acid content in the root compartment and colonization were low (Figs 1e,f and 2). The pH of the root compartment declined strongly during the experimental period, while only a small effect of the pH treatment on soil pH was detectable (Table 1). The decreased soil pH at the low pH treatment probably contributed to a higher root colonization and AM fungal biomass of S. calospora in the root compartment (Figs 1h and 2a). This may look like a contradiction to the result in the hyphal compartment (Fig. 1b), but the actual pH value was close to 6 both in the root compartment at the low pH treatment and in the hyphal compartment at the high pH treatment (Table 1). This indicates that S. calospora has a pH optimum of c. 6. The pH treatment influenced root growth only in AM colonized plants, indicating that this effect was mediated by the AM fungal responses to pH.
The lower pH of the growth substrate of the extraradical mycelium inhibited mycelial growth and possibly spore formation in both species (Fig. 1b,c). This agrees with Abbott & Robson (1985), who showed that the spread of extraradical mycelium by a Glomus isolate was strongly inhibited at low soil pH. It is likely that low growth of extraradical mycelium at lower pH was caused by aversion to the substrate. Acidic soils are known to have a fungistatic effect on spores of Glomus mosseae (Siqueira et al., 1984). The hyphae of S. calospora in our experiment did not appear to be severely stressed at this lower pH, since their ALP activity was not reduced (Table 2). In other studies it was shown that stress caused by simulated acid rain (pH 3–4) reduced the ALP activity of AM fungal extraradical mycelium in a range of substrates (Vosatka & Dodd, 1998; Malcováet al., 1999). Local responses of AM fungal extraradical mycelium have been found, mainly as proliferation in patches rich in organic material (Mosse, 1959; St John et al., 1983; Ravnskov et al., 1999). This stresses the importance of using experimental systems that allow separation of extraradical mycelium and roots when studying effects on extraradical mycelium, since in nature mycorrhizal hyphae can reach patches not available to plant roots.
In the present study, the pH of the hyphal compartment was buffered with a MES buffer. MES has been shown to stimulate extraradical mycelium growth, but only significantly so at concentrations 100 times higher than in our study (Vilariño et al., 1997). In our study, similar amounts of MES were added for all treatments and so it is unlikely to be the cause of differences in extraradical mycelium biomass. The addition of MES to the hyphal compartment, containing acid-washed sand largely deprived of nutrients, should not influence the nutrient uptake. It has been observed to affect nutrient uptake in plants (Medeiros et al., 1993), although at a concentration four times higher than here, and plants were grown in a nutrient solution. Since the decrease in pH of the root compartment over time was greater than the difference in the pH levels employed (Table 1), we conclude that the acidifying capacity of the roots had a stronger influence on the pH of the root compartment than the effect of diffusion of MES from the hyphal compartment to the root compartment.
Our results show that the ELF substrate can be used both to detect and estimate the extent of extraradical mycelium with active phosphatases (Table 2). Most studies on phosphatases of AM fungal extraradical mycelium have focused on ALP activity. Our results indicate that ACP activity of the extraradical mycelium contributed significantly to total phosphatase activity, in agreement with recent findings by Joner & Johansen (2000). Thus, ACP activity should also be considered in phosphatase studies, especially since recent observations have shown the vacuoles of AM fungi to be acidic (Ezawa et al., 2001). Our results (Table 2) indicate that phosphatase activity of the extraradical mycelium can be highest at a substrate pH quite different from the pH for optimal enzyme activity. In most cases no significant phosphatase activity could be detected in the substrate of the hyphal compartment. However, where phosphatase activity was detected in the substrate, the values were of the same order as those observed by Joner & Jakobsen (1995) and Joner et al. (1995). In the root compartment, the observed ACP activity in the substrate probably did not originate from the AM fungi, since similar values were found for all fungal treatments. This supports the view of Joner et al. (2000) that AM fungal phosphatases are not actively released into the soil.
Interaction studies between AM fungi and saprotrophic microorganisms have often reported conflicting results (Ames et al., 1984; Posta et al., 1994; Olsson et al., 1996). As shown in Fig. 3, temporal differences may be one factor explaining such discrepancies. In an earlier study, saprotrophic fungi, detected by phospholipid fatty acid 18:2ω6,9, were inhibited by AM fungal extraradical mycelium in root-free dune sand (Olsson et al., 1998). This could not be confirmed here since the biomass of saprotrophic fungi was too variable (Table 4).
The two fungi showed differences in their ability to colonize the roots and to produce extraradical mycelium biomass (Figs 1 and 2, respectively). These differences persisted throughout the experiment and can not be explained only by a difference in infective propagule density of the inoculum. Since AM fungal signature neutral lipid fatty acids have been shown to correlate with the number of spores formed by the extraradical mycelium (Olsson et al., 1997), the high neutral lipid fatty acid values indicate that sporulation was higher in G. intraradices. High amounts of AM fungal neutral lipid fatty acids in the extraradical mycelium indicate that high amounts of neutral lipids, mainly triacylglycerides, have been transported from the intraradical mycelium (Bago et al., 2000). This carbon cost to the plants may have caused the lower weight of G. intraradices-colonized plants compared with S. calospora-colonized and nonmycorrhizal plants. The differences in responses of the AM fungal isolates support previous reports showing the importance of studying more than one fungal isolate (van der Heijden et al., 1998a,b; Smith et al., 2000; Ezawa et al., 2001).
In conclusion, the results show that the growth of AM fungal mycelia is directly influenced by the pH of the substrate. This is further evidence that AM fungi are organisms that respond to a heterogeneous environment. We have also shown that the way in which AM fungi species respond to pH may vary between species. This study also reveals a possible connection between the response of the extraradical mycelium and the colonization strategy within the host root. Finally, the results support the view that the external phosphatases of AM fungi are of little importance.
The financial support of the Swedish Council for Forestry and Agricultural Research is gratefully acknowledged. We thank Professor Anne Ashford for critically reading the manuscript.