Exploring interactions between saprotrophic microbes and ectomycorrhizal fungi using a protein–tannin complex as an N source by red pine (Pinus resinosa)

Authors

  • Tiehang Wu,

    1. Department of Horticulture, The Pennsylvania State University, University Park, PA 16802, USA
    Search for more papers by this author
  • Jori N. Sharda,

    1. Department of Horticulture, The Pennsylvania State University, University Park, PA 16802, USA
    Search for more papers by this author
  • Roger T. Koide

    Corresponding author
    1. Department of Horticulture, The Pennsylvania State University, University Park, PA 16802, USA
      Author for correspondence: Roger T. Koide Tel: +1 (814) 863 0710 Fax: +1 (814) 863 6139 Email: rkoide@psu.edu
    Search for more papers by this author

Author for correspondence: Roger T. Koide Tel: +1 (814) 863 0710 Fax: +1 (814) 863 6139 Email: rkoide@psu.edu

Summary

  • • Recent studies suggest that some plants may circumvent N mineralization carried out by saprotrophs because their ectomycorrhizal fungi have the capacity to hydrolyse protein. When complexed by tannins, however, proteins may be unavailable to some ectomycorrhizal fungi.
  • • Here we tested the hypothesis that when protein–tannin complex is the N source, Pisolithus tinctorius will promote N uptake into red pine (Pinus resinosa) only in the presence of saprotrophs.
  • • The model protein–tannin complex was stable at field pH. P. tinctorius could not obtain N from it, but saprotrophs could. Pre-treatment of the complex by saprotrophs did make its N available to ectomycorrhizal fungi. However, when the protein–tannin complex was the major N source, P. tinctorius increased shoot P but not N content, even in the presence of saprotrophs.
  • • Interactions between saprotrophs and ectomycorrhizal fungi may be different for N and P because of immobilization of N by ectomycorrhizal fungi, or by the more rapid diffusion of ammonium than phosphate, rendering the absorptive surface area of ectomycorrhizal fungi superfluous for uptake of N but not for P.

Introduction

Comparatively little of the total nitrogen (N) pool in the floor of a temperate or boreal forest is readily available to plants as ammonium or nitrate. Most of it consists of organic forms, including proteinaceous materials, which often comprise the largest fraction, poorly characterized heterocyclic N compounds, and a much smaller fraction composed of amino sugars (Schulten & Schnitzer, 1998). Because many forest plants do not themselves have direct access to this large organic pool, it is generally believed that they rely on saprotrophic microorganisms to convert it into mineral forms. Recent studies have shown that it may be possible for some plants to circumvent the saprotrophs because some of their ectomycorrhizal fungi may possess the ability to hydrolyse protein (summarized in Chalot & Brun, 1998), absorb peptides or amino acids from the environment (Abuzinadah & Read, 1988; Chalot et al., 1995; Keller, 1996; Wallenda & Read, 1999) and transfer peptides, amino acids or their metabolites to host plants (Melin & Nilsson, 1953; Abuzinadah et al., 1986; Abuzinadah & Read, 1989a, 1989b, 1989c; Finlay et al., 1992; Turnbull et al., 1995). That each of these steps can occur is unequivocal. The extent to which each occurs in nature, however, remains unclear.

In particular, there is controversy concerning the ability of ectomycorrhizal fungi to hydrolyse proteins in the environment. In the soils of many ecosystems large concentrations of polyphenolic compounds exist, many of which are capable of hydrogen-bonding to proteins to form protein-polyphenol complexes (Qualls et al., 1991; Northup et al., 1995; Bending & Read, 1996). If the polyphenol can precipitate protein, it is considered a tannin (Hagerman, 2002). The formation of insoluble protein–tannin complexes may serve to reduce N leaching from the forest floor (Northup et al., 1995). However, the formation of such complexes may also render the proteins unavailable to various microorganisms. Indeed, many ectomycorrhizal fungi apparently have little or no enzymatic capacity to hydrolyse protein–tannin complexes (Bending & Read, 1996; Colpaert & Van Laere, 1996). Therefore, such ectomycorrhizal fungi do not apparently possess a strong ability to degrade plant residues and directly acquire N from them (Bending & Read, 1996; Colpaert & Van Laere, 1996; Colpaert & Van Tichelen, 1996).

By contrast, many saprotrophic and ericoid mycorrhizal fungi can break down protein–tannin complexes using polyphenol oxidase, peroxidase or tannin carboxyl esterase (Bending & Read, 1996; Colpaert & Van Laere, 1996; Gramss et al., 1998) and therefore can obtain N from a protein–tannin complex. Bending and Read (1996) proposed therefore that recalcitrant plant residues may first have to be metabolized by such fungi before certain ectomycorrhizal fungi can access them as N sources. Metabolism of organic N by saprotrophic microorganisms may eventually result in its mineralization. The mineralization of N by saprotrophs, of course, would directly enhance plant N status by providing readily available inorganic N to the roots. Nitrogen mineralization could also indirectly enhance plant N status by making N available to ectomycorrhizal fungi. We therefore hypothesize that some ectomycorrhizal fungi do not circumvent the saprotrophs when protein–tannin is the major N source. Instead, these ectomycorrhizal fungi should promote N uptake into their hosts only in the presence of saprotrophic microorganisms. We tested this hypothesis in a series of experiments reported here.

Materials and Methods

Experiment 1. Effects of pH on the stability of the protein–tannin complex

Bovine serum albumin–tannic acid complex was used as a model protein–tannin complex in previous studies (Bending & Read, 1996). To be consistent with those studies, we continue here to use the same complex. Soil pH is an important environmental variable that can strongly influence the stability of hydrogen bonding between tannins and proteins, thus requiring that the relationship between complexation and pH be investigated. In order to do this, protein–tannin complex was produced by mixing solutions of tannic acid (TA, Sigma Chemical Co., #T-0125) and bovine serum albumin (BSA, Sigma #A-7906), each made up in citrate–Na2HPO4 buffers at pH 2.6, 3.6, 4.6, 5.6, 6.6, and 7.6 (Dawson et al., 1986). The final TA and BSA concentrations were each 250 mg l−1. There were four replicate test tubes at each pH. After 24 h, the suspensions in the tubes were filtered using coarse nylon mesh followed by paper, then 0.45 µm membrane and finally 0.2 µm membrane. Filtrate NH4+ concentrations were measured using the salicylate method of the Hach Company (Hach, 1997). Total dissolved N was measured by digestion of the filtrate in a mixture of concentrated H2SO4 and 30% H2O2 at 400°C for 1 h followed by colorimetric determination of NH4+ by the Nessler method (Jensen, 1962). Preliminary tests indicated that there was no measurable nitrate and nearly no measurable NH4+ in the filtrate. Dissolved organic N was calculated as the difference between total dissolved N and NH4+.

Experiment 2. Ability of Pisolithus tinctorius to use protein–tannin complex as a N source

An isolate of P. tinctorius (Pers.) Coker & Couch was obtained from Dr Michael Kiernan (Plant Health Care, Inc., Pittsburgh, PA, USA). Replicate colonies were subjected to four different N treatments on the gelled nutrient medium listed in Experiment 4 (to be described later) but without NH4Cl, malt or yeast extracts. Also, the NaFeEDTA was replaced with FeCl3 because of the possible inhibitory effects of EDTA on the activity of certain proteases (Rodier et al., 2001). The four different N treatments were: control (7 mg N l−1 as arginine, Sigma Chemical Co., #A-5131), protein (7 mg N l−1 as arginine and 80 mg N l−1 as bovine serum albumin, Sigma, #A-7906), protein–tannin complex (7 mg N l−1 as arginine and 80 mg N l−1 as bovine serum albumin, equivalent to 500 mg bovine serum albumin l−1, and 500 mg tannic acid l−1) and tannin control (7 mg N l−1 as arginine with 180 mg tannic acid l−1, Sigma, #T-0125 or #T-8406). A 180-mg l−1 tannic acid concentration was approximately equivalent to the free tannic acid concentration in the suspension of the protein–tannin complex in water at room temperature. The protein and tannin were added to previously autoclaved nutrient solutions as filter-sterilized aqueous solutions at the appropriate concentrations. There were eight replicate 100 mm Petri dishes for each treatment. The mycelium in each dish was harvested 45 d after the start of the experiment by melting the agar in boiling water. It was then rinsed in distilled water and oven dried at 65°C. The N contents of the dried mycelia were determined by digestion in a mixture of concentrated H2SO4 and 30% H2O2 at 400°C for 1 h followed by colorimetric determination of NH4+ by the Nessler method (Jensen, 1962).

Experiment 3. Ammonification of N from protein–tannin complex by forest floor microorganisms.

We tested the ability of various microorganisms from the F-layer of the forest floor of a red pine (Pinus resinosa Ait.) plantation to ammonify N from the protein–tannin complex. Nitrification rates in the forest floor of the plantation are on the order of 3% of ammonification rates (data not shown). An assemblage of forest floor microorganisms was tested using nonsterile F-layer as inoculum. The F-layer was collected from an approx. 65-year-old red pine plantation located in State College, Centre County, PA, USA. Further details of this plantation are given in Dickie et al. (2002). We also tested four individual fungal isolates (Penicillium 1, Trichoderma 1, Mucor sp. and Unknown Fungus 1) and two bacterial isolates (A and B). The fungi were originally isolated from the forest floor on Czapek-Dox agar (Davet & Rouxel, 2000), and maintained on the same medium. The bacteria were originally isolated from the forest floor on Bunt-Rovira agar without soil extract (Bunt & Rovira, 1955) and maintained on the same medium.

Protein–tannin complex was prepared by combining filter-sterilized (0.2 µm membrane) solutions of TA and BSA (see Experiment 1), yielding a final concentration of 250 mg l−1 BSA and 250 mg l−1 TA. The initial C : N ratio of the resulting suspension was 5.9 : 1. Thirty ml of the suspension were placed into each of five sterile, 75-ml test tubes per microbial treatment and inoculated with either F-layer (5 mg) or the individual fungal or bacterial isolates. The control treatment received no microbes. The test tubes were placed on an orbital shaker at room temperature. After 40 d, suspensions were collected and filtered (nylon mesh followed by paper, then 0.45 µm membrane and finally 0.2 µm membrane). Ammonium concentrations in the filtrates were measured using the salicylate method of the Hach Company (Hach, 1997).

Experiment 4. Utilization by ectomycorrhizal fungi of N mobilized from protein–tannin complex by forest floor microorganisms

The ectomycorrhizal fungal species we tested included Amanita rubescens Pers., Cenococcum geophilum Fr., P. tinctorius (Pers.) Coker & Couch, Scleroderma citrinum Pers., Suillus intermedius (Smith & Thiers) Smith & Thiers, and Tylopilus felleus (Bull. ex Fr.) Karsten. The P. tinctorius isolate was previously described in Experiment 2. The isolate of Cenococcum geophilum Fr. (A175) was obtained from Dr James Trappe (USDA Forest Service, Corvallis, OR, USA). The other four were isolated from sporocarps found in the red pine plantation (see Experiment 3).

Suspensions (800 ml) of protein–tannin complex were prepared as in Experiment 3 in each of two 3.0-l Erlenmeyer flasks. One flask was inoculated with an assemblage of forest floor microorganisms by placing 0.1 g nonsterile F-layer into it. The other, control flask received 0.1 g sterile (autoclaved) F-layer. The flasks were placed on an orbital shaker at room temperature. After 45 d, the suspension from each flask was filter-sterilized (nylon mesh followed by paper, then 0.45 µm membrane and finally 0.2 µm membrane) and 600 ml from each flask was separately mixed with 200 ml concentrated growth medium, with or without N (as NH4Cl), to yield nutrient media of the following final compositions (not including the soluble products from the protein–tannin complex): 0.5 g or 0.0 g l−1 NH4Cl, 0.5 g l−1 KH2PO4, 0.25 g l−1 MgSO4·7H2O, 0.025 g l−1 NaCl, 0.05 g l−1 CaCl2, 8 mg l−1 NaFeEDTA, 0.75 mg l−1 KI, 6 mg l−1 MnCl2·4H2O, 2.6 mg l−1 ZnSO4·7H2O, 1.5 mg l−1 H3BO4, 0.13 mg l−1 CuSO4·5H2O, 0.0024 mg l−1 Na2MoO4·2H2O, 0.5 mg l−1 nicotinic acid, 0.1 mg l−1 thiamine HCl, 0.1 mg l−1 pyridoxine HCl, 3.0 mg l−1 glycine, 2.5 g l−1 malt extract, 0.125 g l−1 yeast extract, 10 g l−1 glucose, and 12 g l−1 agar. There were therefore four different media combinations (± microorganism treatment × ±NH4+). These media were poured into 60-mm diameter Petri dishes. The six isolates of ectomycorrhizal fungi were transferred to the media. Harvesting occurred shortly before the colonies grew to the edge of the Petri dishes. Because the different species grew at different rates, they were harvested at different times. All Pisolithus cultures were harvested after 35 d. All Cenococcum cultures were harvested after 78 d. For Amanita, Scleroderma, Suillus and Tylopilus, cultures containing NH4+ were harvested after 35 d and cultures without NH4+ were harvested after 78 d. Nitrogen content of fungal biomass was determined as in Experiment 2. For the colony N content of each ectomycorrhizal fungal species, comparisons were made between microorganism treatments within NH4+ treatments. Initially there were 7 replicates per ectomycorrhizal fungus isolate/nutrient medium combination. Due to contamination or failure to grow, the final number of replicates was variable (caption, Fig. 3).

Figure 3.

Mean (± 1 se) ectomycorrhizal fungus N content for colonies grown on media containing protein–tannin complex incubated with nonsterile or sterile F-layer and without (A) or with (B) N as NH4Cl. n was variable. The following n are for the following treatments, respectively, –NH4+/nonsterile, –NH4+/sterile, +NH4+/nonsterile, +NH4+/sterile: Amanita (5,6,5,7), Cenococcum (4,7,4,5), Pisolithus (5,5,5,5), Scleroderma (5,6,3,5), Suillus (7,5,6,7), Tylopilus (7,6,5,4). *indicates a significant difference between nonsterile and sterile treatments (P ≤ 0.05).

Experiment 5. Effect of forest floor microorganism – mycorrhizal fungus interactions on red pine seedling N and P economies

Red pine (Pinus resinosa Ait.) seeds were obtained from F.W. Shumacher Co. (Sandwich, MA, USA). On 28 May 2002 they were surface-sterilized in 30% H2O2 for 30 min, then rinsed twice with sterile deionized water. Seeds were transferred to Petri dishes containing a complete nutrient medium (see Experiment 4, described in a previous section). The gelling agent was 6.25 g l−1 gellan gum (Phytagel, Sigma). Each dish contained 12–15 seeds. The dishes were kept at 24°C for 9 d for germination. Dishes containing contaminating microorganisms were discarded.

Nine days after plating out seeds, sterile germinated seeds were transferred to magenta vessels (Sigma-Aldrich, St. Louis, MO, USA), each containing 130 ml of a gamma-irradiated (50 KGy = 5 Mrad) growth medium consisting of a mixture of vermiculite, perlite, and red pine plantation F-layer material in the volume ratio of 6 : 3 : 1, respectively. The mixture was moistened with 65 ml sterile water. Vessels were placed inside ‘Sun Bags’ (Sigma, #B-7026) and then in a controlled environment growth chamber providing 16 h of ‘day’ each 24-h period. ‘Day’ conditions consisted of c. 400 µmoles m−2 s−1 photosynthetically active radiation and 13°C air temperature, resulting in an air temperature of 22°C in the vessels. ‘Night’ conditions were no light, with the air temperature maintained at 18°C. There were 10 magenta vessels, each containing 8–10 seedlings. Sun Bags are clear, autoclavable plastic bags with 0.02 µm filter patches that allow gas exchange but are capable of keeping out contaminating microorganisms.

A solid medium for growth of red pine seedlings was prepared with protein–tannin complex as the major N source. The growth medium consisted of the same mixture of vermiculite, perlite, and F-layer described above. Nitrogen, in the form of a protein–tannin complex, was added. To prepare this, 8.75 g BSA (see Experiment 1, described in a previous section) was dissolved in 700 ml acetate buffer (0.2 M acetic acid, 0.17 M NaCl, adjusted to a pH of 4.9), and 1.75 g TA (Sigma, #T-8406) was dissolved in 700 ml water. For each 130 ml portion of pine seedling growth medium, the amount used in each magenta vessel, 12 ml of BSA solution and 12 ml of TA solution were mixed together and centrifuged for 10 min. The pellets that formed were washed twice with 24 ml deionized water, and finally suspended in 12 ml water. This suspension was added to 130 ml substrate. The substrate was then mixed thoroughly, dried, individually bagged, and sterilized by gamma-irradiation (as described earlier).

Sixty-two days after plating out seeds (after the sterile seedlings had developed lateral roots), seedlings were randomly assigned to all combinations of ± P. tinctorius (PT), and ± saprotrophic (SAP) microorganisms, and transplanted into individual magenta vessels each containing 130 ml of the sterile seedling growth medium containing protein–tannin complex as the major N source. The growth medium was moistened with 65 ml sterile nutrient medium (see Experiment 4, described in a previous section) modified to contain no gelling agent, glucose, yeast or malt extract, and reduced concentrations of P (15 µg P ml−1 as KH2PO4) and N (7 µg N ml−1 as NH4Cl). Initially there were 12 replicates of –PT/–SAP (CON), 11 replicates of –PT/+ SAP (SAP), 11 replicates of +PT/–SAP (PT), and 10 replicates of +PT/+SAP (PT + SAP). The PT plants were inoculated with P. tinctorius, originally provided by Dr Michael Kiernan (Plant Health Care, Inc., Pittsburgh, PA, USA). P. tinctorius had been subcultured for several weeks with gentle shaking in liquid medium of the same nutrient composition as given in Experiment 4. Twenty-three seedlings were inoculated with P. tinctorius by sterilely placing cut pieces of the fungal colonies on the lateral roots. Another 23 were not inoculated. All seedlings were transplanted into the growth medium in the magenta vessels, placed into sterile Sun Bags and then in the growth chamber (conditions same as described earlier).

Ninety-nine days after plating out seeds, the seedlings assigned to the –PT/+SAP and +PT/+SAP treatments were given 1.0 ml (c. 0.35 g) nonsterile F-layer material that was collected on 3 September 2002 from five different areas in the red pine plantation. The F-layer samples were mixed together, and roots and woody debris were removed. The material was then air-dried overnight in the greenhouse, then it was ground with a mortar and pestle. All magenta vessels were placed back into sterile Sun Bags, which were placed back into the growth chamber.

All seedlings were harvested 157 d after plating out seeds. The roots and shoots were harvested separately, and dried at 70°C. Substrate from each vessel was placed on gelled nutrient medium (see previous section) to check for microbial growth. If replicates of either +PT/–SAP or –PT/–SAP treatments contained bacteria or rapidly growing saprotrophic fungi that replicate, was excluded from further analysis. Roots and shoots were weighed separately, and N (by Nessler method, Jensen, 1962) and P (by molybdo-phosphate method, Watanabe & Olsen, 1965) concentrations were assessed following digestion at 400°C in a mixture of concentrated H2SO4 and 30% H2O2. Two factor anova were performed using the Statgraphics programs (STSC, 1991). The factors were PT (+ and –) and SAP (+ and –).

Results

Experiment 1. Effects of pH on the stability of the protein–tannin complex

pH did influence the stability of protein–tannin complex. Values between 3.6 and 6.6 resulted in the most stable protein–tannin complex. At pH above 6.6 and below 3.6, however, dissolved organic N concentration was significantly increased (Fig. 1).

Figure 1.

Mean (± 1 se) dissolved organic N (DON) concentration of the suspension of protein–tannin complex at different pH values (n = 4). Different letters indicate means are significantly different (P ≤ 0.05).

Experiment 2. Ability of Pisolithus tinctorius to use protein–tannin complex as a N source

P. tinctorius in pure culture could utilize bovine serum albumin (a protein) as a N source (Table 1). However, when the protein was complexed by the tannic acid, the N content of the colonies was the lowest of all treatments. We conclude that complexation by the tannin rendered the protein N unavailable to P. tinctorius.

Table 1.  Mean colony N content of Pisolithus tinctorius cultures grown in four treatments including control, tannic acid, bovine serum albumin (BSA), and BSA-tannic acid complex (see the Materials and Methods section for explanation)
TreatmentN content (µg N)
  1. Different letters indicate means are significantly different (P < 0.05).

  2. Values are means, se in brackets. n = 6, 8, 8, 8, respectively.

Control103.2 (7.7) b
Tannin 84.8 (10.6) b
Protein875.2 (48.3) a
protein–tannin complex 21.4 (2.7) c

Experiment 3. Ammonification of N from protein–tannin complex by forest floor microorganisms

After 40 d, all microbes produced significantly more ammonium from the protein–tannin complex than the control, but the forest floor microorganisms from the F-layer and individual isolates of fungi and bacteria differed in their abilities to do so (Fig. 2). The fungi and the forest floor microorganisms from the F-layer were far more capable of ammonification than either of the bacterial isolates, and Trichoderma 1 and Mucor 1 were superior to Unknown Fungus 1, Penicillium 1 and the forest floor microorganisms from the F-layer. The biomass of the various microorganisms probably was not the same at the end of the experiment. Variation in biomass may reflect variation in tolerance to the conditions of growth in liquid culture. This variation may have contributed to variation in N mineralization. In any case, the results do show that the microorganisms in question were capable of mineralizing the substrate.

Figure 2.

Mean (± 1 se) NH4+ concentration mineralized from protein–tannin complex after 40 d incubation with different forest floor microbes and with nonsterile F-layer from a red pine plantation. n = 5. Fun 1, Unknown fungus isolate 1; Pen 1, Penicillium sp. isolate 1; Tri 1, Trichoderma sp. isolate 1; Muc 1, Mucor sp. isolate 1; BacA, bacteria isolate A; BacB, bacteria isolate B; CON, control (no microbes). Different letters indicate means are significantly different (P ≤ 0.05).

Experiment 4. Utilization by ectomycorrhizal fungi of N mobilized from protein–tannin complex by forest floor microorganisms

After 45 d treatment of protein–tannin complex, the nonsterile F-layer produced more NH4+ and dissolved organic N than did the control (Table 2). As there was only one flask per treatment it is impossible to analyse these results statistically. Tannic acid concentration was higher and pH was lower in the control.

Table 2.  Concentrations of various N sources from the soluble fraction of protein–tannin complex after 45 d incubation with nonsterile or sterile F-layer (a single flask for each)
Incubated withNH4+ (mg N ml−1)DON* (mg N ml−1)Tannic acid (TA) mg TA ml−1pH
  1. These solutions were used in Experiment 4 (Fig. 3). Initial C : N ratio was 5.9 : 1. The initial concentrations of TA and BSA were each 250 mg ml−1. *Dissolved organic nitrogen (see the Materials and Methods section for details).

Nonsterile F-layer23.48.6028.25.16
Sterile F-layer (control) 0.353.4770.84.42

Pre-treatment of the protein–tannin complex with the forest floor microorganisms allowed all species of ectomycorrhizal fungi to accumulate significantly more N in the absence of added NH4+ in the medium (Fig. 3a). By contrast, when NH4+ was added to the medium in the form of NH4Cl, pre-treatment of the protein–tannin complex with the forest floor microorganisms resulted in none of the species except P. tinctorius containing more N (Fig. 3b). We conclude from this that incubation of the protein–tannin complex with forest floor microorganisms resulted in a release of N available to the ectomycorrhizal fungi.

Experiment 5. Effect of forest floor microorganism–mycorrhizal fungus interactions on red pine seedling N and P economies

We harvested two seedlings that had been inoculated with PT just before inoculation with nonsterile F-layer. Colonization of the root system by PT had occurred, but there were no pronounced visible effects of PT on shoot growth.

By the end of the experiment, c. 30% of the root tips of seedlings in the PT and PT + SAP treatments were ectomycorrhizal, and none were ectomycorrhizal in the CON or SAP treatments. Ectomycorrhizas were not examined microscopically, nor were they analysed by molecular methods, but visual inspection indicated that they were all of the P. tinctorius morphotype.

Both shoot and root system dry weights were significantly increased by SAP, neither were significantly influenced by PT, and there were no significant interactions (Table 3). PT significantly increased shoot and root P concentrations (Table 4), but because the shoot weight was not significantly affected by PT, we conclude that shoot growth was not limited by P. By contrast, SAP did significantly increase shoot N concentration while it increased shoot weight, which is consistent with the hypothesis that N was probably a limiting resource for shoot growth. Both PT and SAP significantly increased root N concentration. In this experiment it was not possible to distinguish between N contained in root tissue and that in the fungal mantle. Thus, the effect of PT on root N concentration may have been due to the N contained in the fungus portion of the ectomycorrhizas. The SAP treatment significantly decreased root and shoot P concentration, presumably at least partly due to dilution because increased N uptake (Fig. 4b) in N-deficient plants resulted in greater growth (Table 3).

Table 3.  Red pine (Pinus resinosa) seedling shoot and root system dry weights
 Shoot d. wt (mg)Root d. wt (mg)
  1. Values are means, se in brackets.

CON 77.5 (1.5)23.2 (1.7)
PT 82.6 (5.5)20.8 (1.3)
SAP119.8 (3.9)42.8 (2.1)
PT + SAP124.9 (5.2)45.1 (2.0)
anova
PT  0.2421 0.9978
SAP  0.0001 0.0001
Interaction  0.8603 0.2324
Table 4.  Shoot and root system N and P concentrations of red pine (Pinus resinosa) seedlings
 Shoot P (mg g−1 d. wt)Root P (mg g−1 d. wt)Shoot N (mg g−1 d. wt)Root N (mg g−1 d. wt)
  1. Values are means, se in brackets. n = 12, 7, 11, 10 for control, PT, SAP and PT + SAP treatments, respectively (see text for details of treatments).

CON1.67 (0.07)2.15 (0.07)16.2 (0.7)17.9 (0.5)
PT1.89 (0.10)3.32 (0.31)16.5 (0.5)23.0 (1.0)
SAP1.34 (0.07)1.62 (0.14)18.3 (0.4)23.6 (0.8)
PT + SAP1.46 (0.04)1.96 (0.13)18.2 (0.5)25.5 (1.0)
anova
PT 0.0254 0.0001 0.8810 0.0002
SAP 0.0001 0.0001 0.0017 0.0001
Interaction 0.4787 0.1195 0.7024 0.0662
Figure 4.

(a) Mean (± 1 se) shoot and root phosphorus contents per red pine (Pinus resinosa) seedling. (b) Mean (± 1 se) shoot and root nitrogen contents per seedling. n = 12, 7, 11, 10 for control, PT, SAP and PT + SAP treatments, respectively (see text for treatment details). The dotted lines occur at the values for the CON treatment for ease of comparison.

Shoot P content (Fig. 4a) was significantly increased by both PT (P = 0.0292) and SAP (P = 0.0088). There was no significant interaction, indicating that the positive effects of PT and SAP on shoot P content were essentially additive. Root P content was significantly increased by both PT (P = 0.0002) and SAP (P = 0.0003). There was no significant interaction, indicating that the positive effects of PT and SAP on root P content were also essentially additive. As mentioned above, however, the inability to distinguish between fungal and root nutrients in the root systems confounds this result.

Shoot N content (Fig. 4b) was not significantly influenced by PT (P = 0.2591), but was significantly increased by SAP (P = 0.0001). There was no significant interaction. Root N content was significantly increased by both PT (P = 0.0269) and SAP (P = 0.0001). There was no significant interaction, indicating that the positive effects of PT and SAP on root N content were additive. However, the inability to distinguish between fungal and root nutrients in the root systems confounds this result.

Discussion

The goal of this research was to test the hypothesis that when protein–tannin complex is the N source, P. tinctorius should promote N uptake into its host only in the presence of saprotrophic microorganisms. This hypothesis was based on previous observations that N in the floor of temperate forests is largely organic (Swift et al., 1979), that much of it is protein (Schulten & Schnitzer, 1998), that the protein is complexed by polyphenolic compounds (Qualls et al., 1991; Northup et al., 1995; Bending & Read, 1996), that many ectomycorrhizal fungi do not have the capacity to hydrolyse these complexes (Bending & Read, 1996), but that at least some saprotrophic microorganisms do (Bending & Read, 1996).

We chose to use a complex of bovine serum albumin and tannic acid as the N source because it had been used previously (Bending & Read, 1996). We determined that within the range of pH from 6.6 and 3.6, it was largely insoluble, and so met with the criterion we had set that at ecologically relevant pH it would remain a stable complex. The F-layer of the forest floor of our red pine plantation has a pH of 3.7 (Shumway & Koide, unpublished results). At pH 7.6 c. 70% of the N in the protein–tannin complex became dissolved organic N, and at pH 2.6 c. 33% of the N became dissolved organic N. Osawa & Walsh (1993) also showed that both alkaline (pH > 7) and highly acidic (pH < 3) conditions result in the dissociation of bovine serum albumin from tannic acid. The exact relationship between pH and solubility is likely to vary with the nature of both the polyphenol and the protein. The protease secreted by many ectomycorrhizal fungi has an optimum pH of c. 3 (Read, 1993), and ectomycorrhizal fungi, when cultured with protein or peptides as N sources, can depress pH to < 3 (Abuzinadah & Read, 1986). Therefore, substantial release of dissolved organic N may occur simply as a consequence of local alteration of pH. Whether or not dissolved organic N is still complexed with polyphenolic molecules is not known with certainty, but this seems likely (Northup et al., 1995) because at least some forests contain a large excess of polyphenol relative to protein (Bending & Read, 1996). Whether or not the dissolved organic N is available to ectomycorrhizal fungi is not clear.

Consistent with our expectation, P. tinctorius could not access N from the protein–tannin complex, but a mixed assemblage of microorganisms from nonsterile F-layer from the pine plantation was able to ammonify a significant amount of N from the protein–tannin complex. Moreover, isolates of fungi, including those of Penicillium, Trichoderma, Mucor, and an unknown fungus (FUN1) that would not sporulate in culture, were individually capable of ammonifying N from the protein–tannin complex. While the two bacterial isolates also were capable of this to a very limited degree, they were much less capable than all of the fungal isolates and the assemblage from nonsterile F-layer.

As shown by the results of Experiment 4, incubation of protein–tannin complex with an assemblage of microorganisms from nonsterile F-layer did release N that was available to six isolates of ectomycorrhizal fungi including those of Amanita rubescens, Cenococcum geophilum, P. tinctorius, Scleroderma citrinum, Suillus intermedius and Tylopilus felleus. The assemblage of microorganisms released a relatively large amount of N through ammonification, but some dissolved organic N was also produced. Again, the ecological relevance of the dissolved organic N fraction to ectomycorrhizal fungi is currently unknown. It seems likely, however, that mineralization caused by saprotrophic microorganisms was an important mechanism to make available N to ectomycorrhizal fungi from protein–tannin complex. Other researchers support the hypothesis that inorganic N is commonly the form of N taken up by ectomycorrhizal fungi (Colpaert & Van Laere, 1996; Smith & Read, 1997).

The results from Experiment 5, however, were contrary to our expectations. Because P. tinctorius could not access N from BSA-tannic acid complex and because saprotrophic forest floor microorganisms could, we assumed that P. tinctorius would only positively affect host N status in the presence of the saprotrophic microorganisms. This did not occur. Instead P. tinctorius was never beneficial in terms of host shoot N content, while it was always beneficial in terms of host shoot P content. There are possibly two major reasons for this. Firstly, the availability of N in the soil solution may have been such that P. tinctorius was itself limited by N so that it had no excess N to pass to its host. In this experiment N was supplied almost entirely via the protein–tannin complex, which contains essentially no ammonium or nitrate before it is mineralized. A small amount of N as ammonium may occur in the F-layer in the medium (a total of 0.27 mg N per vessel). If mineralization rates were low enough, the amount of freely available N may not have been sufficient to meet the growth needs of P. tinctorius. If that is true, P. tinctorius could have immobilized the N and thus competed against the host for this limiting nutrient. This phenomenon has been described elsewhere (Perez-Moreno & Read, 2000; Koide & Kabir, 2001). Consistent with this, in this experiment P. tinctorius did have a slight but significant positive effect on root system N content while having no significant effect on shoot N content.

Secondly, in order for our hypothesis to be correct, there must be an advantage in having a more extensive absorptive surface area supplied by P. tinctorius. In Experiment 5, P. tinctorius had a significant positive effect on root and shoot P contents. That was also true in two previous experiments (Koide & Kabir, 2001). These results are consistent with the fact that the diffusion coefficient for phosphate in soils may vary from 1 × 10−14 to 4 × 10−9 (Fitter & Hay, 1987). By contrast, the diffusion coefficient for NH4+ in soil is very much higher, often between 4 × 10−8 and 1 × 10−6 (Fitter & Hay, 1987). This suggests that while the NH4+ produced by saprotrophic mineralization would become available to P. tinctorius, its absorptive surface area may have been superfluous if diffusion of NH4+ did not limit the uptake of N by the roots of the pine seedlings. Therefore, while saprotrophic fungi may be required for mineralization before ectomycorrhizal fungi can have access to some forms of organic N, the ectomycorrhizal fungi may not be needed by the host for N uptake as long as diffusion does not strongly limit its uptake by the roots. We conclude therefore that the interaction between saprotrophic microorganisms and ectomycorrhizal fungi on host nutrient economy may often be different for N and P.

Acknowledgements

We thank the A. W. Mellon Foundation and the Department of Horticulture, Penn State University, for financial assistance.

Ancillary