INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Ectomycorrhizae (EM) are complex interactions between fungi and plant roots, are formed mainly by basidiomycete fungi (39, 47), and are the dominant nutrient-gathering organ in temperate ecosystems (45). These structures provide plants with nitrogen and phosphorus and protect plants from disease (18) and heavy metal contamination (34, 52). Different fungal species, and even isolates of the same species, can vary in their tolerance of harsh conditions (33, 50) and in the ability to help plants grow in extreme environments, such as acidic mine tailings (51) and coal spoils (24, 28). Because of these abilities, EM fungi are used to aid tree growth in programs designed to reclaim habitats altered by factors such as mining, nutrient deposition, and acid rain (11, 17). Thus, it has become increasingly important to determine in situ reactions of EM fungal communities to soil modifications that could inhibit plant growth.

Soil pH and temperature can affect EM fungal growth, fruiting body production and distribution, and plant growth and productivity (2, 25, 35, 36, 37, 43, 46). While EM typically form in acid soils, this process can be sensitive to pH values below 3.3 (16). pH and temperature growth optima (19, 29, 46, 51) for EM fungi can vary among species, even within a single genus (29, 30, 50). Similarly, increased soil temperature can adversely affect sclerotium formation in many EM fungal species, thus influencing inoculation potential (40), and can inhibit EM formation in both disturbed and undisturbed forest soils (43). Thus, pH tolerance and temperature tolerance are important criteria that should be considered when EM fungi are selected for soil reclamation and inoculation.

Our objectives were to determine conditions in thermal soils in a Pinus contorta forest in Yellowstone National Park, Wyoming, and to test the hypothesis that conditions in these soils significantly affect EM fungal infection, species richness, and EM fungal community structure. Because our results represent correlations between conditions and effects, additional studies involving manipulations may be required to fully test this hypothesis. To this end, soil cores were obtained from soil associated with an acidic thermal spring and from adjacent unmodified soils in a neighboring forest stand. These soils are acidic (pH 2 to 4), acid leached, and often less than 10 cm deep. Furthermore, the temperature in soils associated with Yellowstone's thermal features can increase with proximity to active hot springs to more than 60°C. Thus, these conditions are some of the harshest conditions for plant survival and EM fungal growth (42). Furthermore, additional soil cores targeted at individuals of the EM fungal species Pisolithus tinctorius were obtained to specifically determine the frequency of EM of this species. P. tinctorius is the EM fungal species most commonly used in remediation of acidic soils (56), and we wanted to determine whether this species was the dominant EM former in this eco-system.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

The study site was located in the Norris Geyser Basin of Yellowstone National Park, Wyoming. Soil cores were obtained both from soils associated with a hot spring and from nonthermal soils in an adjacent forest. The soil temperatures were 37 ± 2.6°C in all coring plots in the thermal soils and 17 ± 1.5°C in the forest soils. Temperature measurements were taken on 2 days in July, at noon on both days, in the shade and under the litter of at least three trees per plot to minimize the effects of solar heating. The pH values in the plots of thermal and forest soils were 3.3 ± 0.1 and 4.02 ± 0.01, respectively. There were no understory plants present at either location.

For each soil type, cores that were 18.8 cm in diameter and 10 cm deep were taken; three cores were obtained from each of six replicate paired (thermal-nonthermal) plots situated along two parallel 100-m transects that were separated by less than 30 m. A total of 36 soil cores were taken. The coring depth was based on the depth of roots in the soils and was chosen to provide equal volumes of soil for statistical analyses. The cores were sifted to remove the soil, and mycorrhizae were separated on the basis of color (1) and then stored dry at 20°C within 10 days of collection. The EM tips of each morphotype in each core were pooled for quantification (14, 15, 27) and then freeze-dried for long-term storage. No samples were stored alive for deposition in international culture collections, although all morphotypes from all cores were archived for future analyses.

Fungi that formed individual EM were identified by PCR-based methods. To ensure that genetic variability within morphotypes was not missed because of undersampling, individual mycorrhizae were selected for DNA analysis by using the following scheme: one tip per core was analyzed if a morphotype was represented by one to three individual mycorrhizae per core, three tips per core were analyzed if a morphotype was represented by three to five mycorrhizae per core, and five tips per core were analyzed if a morphotype was represented by more than five individual mycorrhizae per core. DNA was amplified from root tips and from fruiting bodies by using primers and ITS1F and ITS4B, which were specific for the internal transcribed spacer region of the nuclear rRNA repeat unit of basidiomycete fungi (22). To identify ascomycete EM fungi, the universal ITS1F-ITS4 fungal primer set was used (22). We used the following parameters for PCR (12): initial denaturation at 95°C for 1 min 35 s, followed by 13 cycles of denaturation for 35 s at 94°C, primer annealing for 55 s at 55°C, and polymerization for 45 s at 72°C, nine additional cycles in which the polymerization time was extended to 2 min, nine cycles consisting of 3 min of extension, and a final 10-min polymerization step at 72°C.

Amplified DNA was digested with restriction enzymes AluI and HinfI, and the band patterns obtained from EM were compared to those obtained from fruiting bodies. Restriction fragment length polymorphism (RFLP) patterns that were identical for EM and fruiting bodies were considered a match; the utility and accuracy of this method have been demonstrated previously (23), and this method was used by us previously in Yellowstone National Park studies (8, 14, 15). When fungal species comprising more than 4% of the EM fungal community could not be identified by using fruiting bodies, the fungi were identified as members of family level monophyletic groups by amplifying a portion of the mitochondrial large rRNA subunit with PCR primers ML5 and ML6 and subsequently comparing the DNA sequences amplified from EM root tips to sequences in a previously published database (7). It is often the case in EM systems that fungal species that are common below ground do not form fruiting bodies; hence, identification beyond a family level monophyletic group is often impossible (23).

Fruiting bodies (mushrooms) of EM species were collected throughout the growing season (June to September) in a 10- by 10-m area around each collection plot in both the thermal and nonthermal soils. Very few fungal species fruited in the collection plots in thermal soils. Thus, fruiting bodies that we collected for other studies in nearby stands (8, 14, 15) were used to identify EM fungi in the thermal soils. Only fruiting bodies collected in collection plots were used to compare fruiting patterns in the two soil types.

Soil chemistry (pH and organic matter, total nitrogen, ammonium, nitrate, phosphorus, potassium, calcium, and aluminum contents) of both the thermal and forest soils was analyzed by the DANR Analytical Laboratory, University of California at Davis, Davis, Calif. As the instructions recommended, soils were collected, oven dried at 55 to 60°C, sieved through a 2-mm mesh, and approximately 300-g portions of soil were collected for analysis. The pH was determined with a pH meter after aqueous extraction (55). The organic matter content was determined by potassium dichromate reduction of organic carbon and subsequent spectrophotometric measurement (modified Walkley-Black method) (41). The soluble calcium and magnesium contents in a saturated paste extract were determined by inductively coupled plasma atomic emission spectrometry (33). The aluminum content was determined by acid dissolution (49). The P Olsen content was determined by alkaline extraction with 0.5 N NaHCO₃. The P Bray content was determined by extraction for acid soils (pH less than 7.0) by using a dilute acid-fluoride extractant (26). The soluble potassium content in the saturated paste extract was determined by emission spectrometry (33). The total nitrogen content was determined by combustion (44, 45), and the nitrate and ammonium contents were determined by extraction with potassium chloride and subsequent measurement with a diffusion-conductivity analyzer (31).

Differences in species richness (number of EM fungal species per core) and level of EM infection (number of individual EM per core) in the two soil types were examined by a Mann-Whitney U test. Differences in overall EM fungal community composition, both above ground based on fruiting body comparisons and below ground based on internal transcribed spacer-RFLP analysis, were examined by using a contingency table and a chi-square test; in the case of below-ground EM fungal diversity, this test was done by using both cores in which each fungal species was detected and the number of EM in each core. Differences in soil chemistry were examined by using Student's t test.

A preliminary molecular analysis (six cores) failed to detect P. tinctorius EM in the thermal soils. Therefore, in addition to the 18 cores taken from the thermal soils for assessment of the EM community, six additional cores were taken directly under fruiting bodies of this species in order to determine the frequency of occurrence of P. tinctorius EM in the vicinity of fruiting bodies of this species.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

The analysis of soil chemistry (Table 1) indicated that the thermal soils were significantly more acidic, contained less organic matter, and had lower levels of phosphorus, potassium, aluminum, and total nitrogen. The ammonium and magnesium contents were significantly higher in the thermal soils, as was the calcium content. The nitrate contents of the soils did not differ significantly.

Statistical analyses of individual EM root tips indicated that both species richness (4.0 ± 0.5 species/core in nonthermal soils, 1.2 ± 0.2 species/core in thermal soils) and the level of EM infection (130 ± 22 EM root tips/core in nonthermal soils, 68 ± 22 EM root tips/core in thermal soils) were lower in the thermal soils (P < 0.005, df = 34).

Molecular analysis indicated that the EM fungal communities of the forest soils contained a diverse array of fungi, including basidiomycetes belonging to both the Agaricales and the Boletales and the ascomycete Cenococcum (Table 2); contingency table and chi-square analysis indicated that there was a significant difference between the EM fungal communities of the thermal soils and the EM fungal communities of the unmodified forest soils (P < 0.001). Furthermore, while the forest soils contained a few dominant EM fungal species that were accompanied by several rare species, the thermal soils contained three dominant fungal species that accounted for 93% of the community and only four rare taxa (Table 2).

The ascomycete Cenococcum was common in the forest soils. The remaining dominant species were members of several basidiomycete families (members of the orders Agaricales and Boletales), including the Suilloid group, the Cantharellaceae, and the Cortinariaceae. In contrast, the EM fungal community of the thermal soils was dominated by fungi in the Cortinariaceae and by a basidiomycete species that failed to form fruiting bodies in either type of soil and could not be identified by sequence as a member of any family level monophyletic group due to the presence of a large intron. The dominant EM fungal species of the thermal soils were Dermocybe phoeniceus and a species of Inocybe designated Inocybe H22. Together, these two species comprised 71% of the EM fungal community. Only one species, Inocybe H22, was present in both the thermal soils and the forest soils. This species was present in the forest soils but at a low level, comprising less than 1% of the EM fungal community. Unfortunately, Inocybe H22 was cryptic and could not be definitively identified to the species level, but it was a member of subsection Marginatae (defined by nodulose or angular spores), section Inocybe (defined by the presence of caulocystidia along the entire length of the stipe).

The levels of fruiting body diversity in the sampling plots for the two soil types were significantly different (P < 0.001). In forest soils, genera belonging to 14 families were represented; these families included the Amanitaceae (Amanita), the Boletaceae (Boletus, Leccinum, Rhizopogon, and Suillus), the Cantharellaceae (Cantharellus), the Clavariaceae (Ramaria), the Cortinariaceae (Cortinarius, Dermocybe, Hebeloma, and Inocybe), the Entolomataceae (Entoloma), the Gomphidiaceae (Chroogomphus and Gomphidius), the Hygrophoraceae (Hygrophorus), the Lycoperdaceae (Lycoperdon), the Polyporaceae (Albatrellis), the Russulaceae (Lactarius, including Lactarius rufus, and Russula), the Strobilomycetaceae (Gautiera), the Thelephoraceae (Thelephora and Sarcodon), and the Tricholomataceae (Tricholoma, Tricholomopsis, Tyromyces, and Xeromphalina). The fruiting body diversity in the thermal soils was greatly reduced, although fruiting bodies of most species detected below ground were collected. The fungi fruiting in the thermal soils included members of the Cortinariaceae (Inocybe H22), the Pisolithaceae (P. tinctorius), the Russulaceae (L. rufus), and the Thelephoraceae (Sarcodon imbricatus).

Surprisingly, even though P. tinctorius formed several fruiting bodies in the thermal soils, no EM of P. tinctorius were detected in any of the 36 cores taken from paired plots in either soil type. Analyses of six additional cores taken directly under P. tinctorius fruiting bodies also failed to locate a single root tip that formed EM with this fungal species.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

Not only were the thermal soils hotter than the forest soils, but they were also more acidic, contained less organic matter, and had different levels of important soil ions, such as ammonium, calcium, and aluminum. This combination of conditions was accompanied by less below-ground species richness, lower levels of EM infection, and the absence of the large number of rare species that were detected in the undisturbed forest soils. In addition, the number of higher-order taxa (family level and higher) was smaller in the thermal soils, and the composition of the EM fungal community was significantly different. The thermal soils were dominated by two species belonging to the Cortinariaceae (D. phoenecius and Inocybe H22), which together accounted for more than 70% of the total EM fungal community. D. phoenecius was the dominant species, accounting for 45% of the total thermal soil EM fungal community, but this species was absent from the forest soils. Inocybe H22 (26% of the total thermal soil community) was very rare in the forest soils (<1% of the total community); no other EM fungal species were detected in both types of soil. Both of these fungi are common, have broad host ranges, and have wide geographic distributions (32, 43). Their overall flexibility probably allowed these species to thrive in the thermal soils and makes them possible alternatives for reclamation studies involving P. contorta.

While the pH was lower in the thermal soils, the effects on the EM fungal community were probably not due to pH differences alone as EM fungal species are common in acid soils and many EM fungal species grow well at a pH of 3.3 or less (3, 38, 53). Indeed, some conditions resulting from the increased acidity in thermal soils (e.g., decreased aluminum content and increased calcium content) can enhance growth and nutrient uptake by some EM fungi (6, 30). We think that the detrimental effects on the EM fungal community probably were due to increased soil temperature and/or reduced organic matter content. The growth of most EM fungal species that have been tested in culture is reduced at temperatures below the temperature measured in the thermal soils examined (37°C) (36, 50). Thus, despite the potentially positive effects of the decreased aluminum content and increased calcium content in the thermal soils, the level of EM infection and species richness were both reduced in the presence of higher soil temperatures. For example, the ascomycete fungus Cenococcum tolerates high acidity (16, 35), is present at relatively high concentrations in forest soils, can exhibit enhanced growth in response to acid rain conditions (38), is sensitive to increased temperatures, exhibits optimum growth at 16 to 27°C in culture, and does not grow at temperatures above 38°C (10). In contrast, although the levels of Sarcodon species are low (only 1% of the community), these organisms tolerate increased soil temperatures and can help host trees grow in harsh environments, such as coal spoils (28).

Similarly, although the effects are likely to be more subtle than the growth limitation observed at temperatures at or above those measured in the soils studied, reduced levels of organic matter (and hence lower levels of organic nitrogen) can also favor some species while inhibiting others. The level of organic matter was lower in the thermal soils, and species of the genera Cenococcum and Suillus (both detected only in the forest soils with higher levels of organic matter) are more productive when they are grown on organic sources of nitrogen. Other fungi (e.g., L. rufus, which was detected only in the thermal soils) are more productive when they are grown on inorganic nitrogen sources (e.g., ammonium, which was present at significantly higher levels in the thermal soils than in the forest soils) (53). L. rufus was present at a relatively low frequency in the thermal soils but was absent from the forest soils, can associate with both Pinus and Picea species, and grows well in boggy, acidic habitats (3). Thus, the low level of L. rufus in thermal soils may seem confusing. However, although L. rufus prefers inorganic sources of nitrogen, growth of this species can be inhibited at low pH by increased ammonium levels, such as those detected in the thermal soils (29). We hypothesize that the interaction between low pH and increased ammonium content in the thermal soils prevented L. rufus from becoming a dominant organism, despite its preference for acidic soils. Although there have been no studies of D. phoeniceus and Inocybe H22 in relation to harsh soils, these organisms may respond to combined organic matter-pH-soil temperature relationships in a similar manner. Targeted studies of these species in relation to these factors are needed to determine if this is the case, and data indicate that these fungi may be candidates for use in reclamation of some harsh acidic soils.

The fruiting bodies collected indicated that above-ground (apparent) EM fungal diversity was reduced in the thermal soils. P. tinctorius formed the majority of the fruiting bodies in the thermal feature. This result was not surprising; P. tinctorius grows well in culture at temperatures up to 40°C, tolerates high acidity (10, 32, 36, 37, 50), and is the fungal species most commonly used to help remediate severely altered soils (54). Yet we did not detect P. tinctorius on a single EM root tip from either soil type. Targeted assays of EM obtained directly under P. tinctorius fruiting bodies also failed to detect EM of this fungal species. Therefore, it is unlikely that we missed EM of P. tinctorius simply due to sampling error. Why this species was not detected is not clear, although culture experiments have demonstrated that P. tinctorius can be either mycorrhizal or not mycorrhizal depending on the sugar conditions in the medium or possibly on the stage of development of the host plant (21). For example, the quality and quantity of exogenous sugars can strongly influence EM development; hence, the low organic matter content of our thermal soils could have inhibited growth of this species. In addition, EM fungi demand significant portions of the sugars fixed by their host plants via photosynthesis (4, 20), and the P. contorta individuals growing in the harsh conditions studied may have been under sufficient stress that they could not provide P. tinctorius with sufficient carbon. Regardless of the mechanism behind the pattern observed here, P. tinctorius must obtain carbon to form fruiting bodies, although the source of this carbon is not known.

A similar pattern of copious fruiting and rare EM occurrence has been observed in another EM fungal genus, Suillus (23), and two hypotheses to explain this phenomenon have been advanced. The first hypothesis is that there is a very efficient mechanism to transfer carbon via coevolved recognition systems between the plant and fungal partners (23). Up to 50% of the total carbon fixed by photosynthesis is passed to the fungal associates (20), and it is possible that highly specific systems could enhance this transfer by providing a more efficient avenue of nutrient transfer (13), enabling fungal growth via very few, difficult-to-locate EM connections. However, P. tinctorius is not considered to be highly host specialized (38), which makes this explanation less likely. An alternate hypothesis is that EM fungi obtain carbon through saprophytic growth (23). Many EM fungi possess the enzymes and transport mechanisms to break down forest litter and acquire metabolizable nitrogen and carbon (9). Despite the low organic matter content of thermal soils, the lack of association of P. tinctorius with roots of P. contorta suggests that P. tinctorius may indeed be capable of enzymatically breaking down organic substrates. Targeted studies of the enzymatic capabilities of P. tinctorius in this system are required to test this hypothesis. Furthermore, further study is needed to determine the role of P. tinctorius in this ecosystem. Our results indicate that this species may not be suitable for EM-related rehabilitation projects in all low-pH-high-temperature soils.

In summary, our results indicate that chemical changes in acidic thermal areas can significantly alter EM fungal community structure. In thermal areas of Yellowstone National Park, the conditions can vary greatly over short distances (48), and as a result, pH-neutral regions can occur close to acidic hot springs. These small adjacent regions are subject to the same input of fungal inoculum from adjacent forest stands. Thus, Yellowstone's thermal features could act as natural field laboratories for studying the physiological potential of EM fungi by providing a range of chemical conditions in which the adaptive abilities of the fungi could be assessed at spatial scales which ensure that the variability in the EM fungal community is due less to spatial variation than to soil properties.