Phosphorus Distribution in Red Pine Roots and the ... - NCBI

Received for publication March 26, 1992 Accepted June 3, 1992

Plant Physiol. (1992) 100, 713-717 0032-0889/92/100/071 3/05/$01 .00/0

Phosphorus Distribution in Red Pine Roots and the Ectomycorrhizal Fungus Hebeloma arenosa1 Janet S. MacFall2*, Steven A. Slack3, and Suzanne Wehrli4 Department of Plant Pathology, 1630 Linden Drive, University of Wisconsin-Madison, Madison, Wisconsin 53706 U.S.M., S.A.S.); and Department of Chemistry, University of Wisconsin-Milwaukee, P.O. Box 413, Milwaukee, Wisconsin 53201 (S.W.) ABSTRACT

tissue of an ectomycorrhizal root (13). The polyP5 content of these granules has been suggested on the basis of studies by electron probe analysis (27-29). Other workers have found that the numbers of these granules increased in excised Fagus and other mycorrhizae when incubated in increasing concentrations of P (3, 7, 8). They concluded that these granules were the sites of P accumulation observed with radioactive tracer experiments. Accumulations of polyP have been observed also in mycorrhizal fungi grown in culture (17, 18, 26). Later work has shown that uptake of 32P by mycorrhizae is first into a Pi pool, followed by accumulation as polyP (7). As has been proposed for VAM fungi (27), the polyP granules may serve as a pool of reserve phosphate, available for hydrolysis with subsequent transfer to the host plant as Pi following demand by the host (3, 9). This model would predict that under conditions in which P is limited to the plant, there would be little or no polyP synthesis and accumulation by the fungus. The use of in vivo NMR spectroscopy has been shown to be an invaluable technique in the study of biological compounds that are labile or compartmentalized within the cell (5, 20). The usefulness of 31P NMR to study metabolism in intact plant tissues has also been demonstrated (10, 21-25). Cytoplasmic and vacuolar Pi can be distinguished, as well as labile compounds such as ATP (14). Use of 31P NMR to study phosphate distribution in mycorrhizal fungi and mycorrhizae can potentially eliminate difficulties such as nonspecific staining of polymeric metabolites and hydrolysis of polyP (12). The ectomycorrhizal fungus Hebeloma arenosa Burdsall, MacFall, and Albers has been shown to increase growth of red pine (Pinus resinosa Ait.) in P-limiting soil by increasing P uptake. However, growth enhancement and mycorrhizal infection decreased with increasing additions of P to soil (16). The purpose of the present study was to examine the distribution of P-containing compounds in the mycorrhizae formed by the symbiosis of H. arenosa and red pine when grown in soil of varying P levels. More specifically, we have used in vivo 31P NMR spectroscopy to test the hypothesis that polyP is synthesized when adequate to excess levels of soil P are available to the host, but that polyP synthesis by mycorrhizal plants does not occur under P-limiting conditions.

Red pines (Pinus resinosa Ait.) were grown in a pasteurized sandy loam either unamended with phosphate or fertilized with one of two levels of phosphate (34 or 136 mg/kg) as superphosphate, and with and without addition of Hebeloma arenosa inoculum. Shoot and total dry weights of mycorrhizal seedlings grown in soil unamended with P were greater than those for nonmycorrhizal seedlings grown in the same soil, but less than the dry weights of seedlings grown in soil amended with middle to high levels of P. Mycorrhizal infection was inhibited at the highest level of P amendment. 31P nuclear magnetic resonance spectra of intact mycorrhizal roots showed the presence of two dominant peaks, orthophosphate (Pi) and polyphosphate (polyP). The polyP peak was absent in spectra of nonmycorrhizal roots. The ratio for areas under the two peaks, Pi/polyP, was 1.8 for mycorrhizal roots grown in both unamended soil and soil that had received middle levels of superphosphate. Apparently, the fungus strongly mediates the supply of phosphate to the tree through the production of polyP, even at growth-limiting levels of soil P, and regulates compartmentalization of P in the mycorrhizal roots.

Mycorrhizal fungi can take up, store, and translocate P while forming mycorrhizal associations with the host plant (6, 9). Autoradiographic studies have shown uptake and translocation of P both toward the host plant and toward the hyphal tips through the fungal mycelium (4, 11), as well as accumulations within the mycorrhizae. Harley and Smith (9) summarized studies on the uptake of P by mycorrhizae and suggested that phosphate uptake by the host tissue is through the fungal symbiont. There appear to be at least two pools of phosphate within the mycorrhizal structure, a small pool of phosphate that is readily translocated from the fungus to the host, and a larger, accumulated pool of relatively inactive, inorganic phosphate. Light microscopic studies have shown the presence of inorganic phosphate as metachromatic staining granules within the fungal ' This research was supported by the University of WisconsinMadison College of Agricultural and Life Sciences and the Department of Natural Resources grant No. 133-L692. 2 Present address: Department of Radiology, Duke University Medical Center, Durham, NC 27710. 3Present address: Cornell University, Department of Plant Pathology, Ithaca, NY. 4Present address: Children's Hospital, Philadelphia, PA.

5 Abbreviations: polyP, polyphosphate; Pi, orthophosphate; VAM, vesicular-arbuscular mycorrhizae; Ti, spin-lattice relaxation time.

713

Plant Physiol. Vol. 100, 1992

MACFALL ET AL.

714

MATERIALS AND METHODS

Red pines were grown in 500-mL styrofoam cups filled with a sandy, pasteurized soil (Sparta loamy sand) either unamended with P or in soil supplemented with 34 or 136 mg P/kg. Unamended soil had a pH of 5.2 and contained 12 mg P/kg P, 18 mg exchangable K/kg, 179 mg exchangable Ca/kg, 46 mg exchangable Mg/kg, 0.5% N (w/w), and 1.3% organic matter. P content of the unamended soil would be considered extremely low for pine production (30). Soil either received inoculum of Hebeloma arenosa in leached vermiculite-peat substrate or leached nonfungal vermiculite-peat at a rate of 12% v/v. Inoculum production and soil nutrient content following amendment were described previously (16). N and K were applied at 2-week intervals, each at a rate of 60 mg/kg in the form of ammonium nitrate and K sulfate, respectively. Trees were grown in a greenhouse for 9 weeks and then were transferred to a cold frame outdoors for 11 additional weeks of growth. Immediately before analysis by NMR, lateral roots containing short roots were excised from each tree. Due to the small root biomass, nonmycorrhizal trees from soil without P amendments were sampled in groups of four, giving a total dry weight of 0.4 g. Roots from individual trees grown in soil amended with 34 (middle) or 136 (high) mg P/kg were examined as separate samples. Only roots with a diameter less than 1 mm were used. For inoculated seedlings grown in unamended soil and with middle levels of P amendments, second-order lateral roots were selected with comparable numbers of mycorrhizal short roots, excised, and then placed into the NMR tube. Approximately 80% of the roots analyzed by NMR were colonized by the fungus in both treatments. No mycorrhizal short roots were formed on seedlings grown at the highest level of P amendment, so only nonmycorrhizal roots were examined. Roots were cut into pieces and used to fill the lower portion of a 15-mm outer diameter NMR tube, and 0.01 M Tris buffer (pH 7.2) with 0.01 M EDTA was added to just cover the root material. A capillary tube containing D20 for spectrometer lock was inserted into the center of the NMR tube. Spectra were acquired from three replicate root sets for inoculated seedlings grown in unamended soil and soil with mid-P amendments. Spectra were acquired on one nonmycorrhizal seedling from each fertility level and one inoculated seedling grown at the highest level of P amendment, making a total of four spectra acquired from nonmycorrhizal seedlings. Spectra were acquired on a Bruker WM250 NMR spectrometer, with internal lock, without spin, and without perfusion. Flip angle was 300 with a 0.4-s acquisition time. The number of scans varied from 75,000 to 198,000, depending on the sample. Trees from the 136 mg/kg P amendment generally required a shortened scanning time because of greater signal from the relatively high P content of the roots. Tls were measured by the inversion recovery method, and the Ti values were obtained by a nonlinear three-parameter regression. For confirmation of NMR peak assignments, com was germinated at room temperature on moistened germination paper. Root tips were excised and spectra produced as for the pine roots. Peak assignments were made based on previously

determined assignments for corn roots with methylene diphosphonic acid as reference (22-24). Similar individual trees not used for NMR were also harvested, dried, weighed, and analyzed for P content by the Northeastern Forest Experiment Research Laboratory (Berea, KY). Ten replicate trees from each fertilizer/inoculation combination were measured for dry weights. Five trees from each inoculation/fertilization treatment combination were analyzed individually for nutrient content. Tissue analysis was as previously described (2). Complete tissue analyses have been previously reported (16). RESULTS

After 19 weeks of growth, all inoculated seedlings grown in P unamended soil and in soil that had received 34 mg P/ kg had formed abundant mycorrhiza and had greater root and shoot dry weights than noninoculated seedlings (Fig. 1). Mycorrhizal infection was slightly greater in seedlings grown in the P unamended soil, with about 70% of the root system colonized by H. arenosa. Mycorrhizal colonization was slightly less, about 65%, with mid-P amendments. At the highest level of P fertilization, however, mycorrhizal infection was completely inhibited on the inoculated trees; thus, both inoculated and noninoculated trees were nonmycorrhizal. There were no significant differences in growth between inoculated and noninoculated trees at this level of fertility. Root and shoot P concentrations were greater in inoculated seedlings when grown in P unamended soil, but not when P was applied. At harvest, mycorrhizal short roots showed two dominant peaks on the NMR spectra (Fig. 2). Because the spectra for corn root tips (spectra not shown) obtained in the present work were nearly identical to previously published spectra for corn root tips (23, 24), vacuolar Pi was used as an internal reference for chemical shift assignments. In the mycorrhizal

2

-4l

~

I

f

f

i

I

a

uW 1.5 :0

1 -1

/ 1-0

(-e

_

--

o noc. 0.5 _~ ~~~~e l/ *-~~~-Non-lnoc.

:

0

20

40

60 80 Added P

100 120 140

Figure 1. Seedling dry weights of inoculated and noninoculated red pine seedlings grown over a range of P amendments to soil. Inoculated seedlings failed to form mycorrhizae at the highest level of applied P. Values are the mean of 10 replicate seedlings. Error bars represent the pooled SE.

PHOSPHORUS IN MYCORRHIZAL AND NONMYCORRHIZAL ROOTS

A

B

715

the spectral acquisition period, 2-h NMR acquisitions of a single mycorrhizal root from a container-grown red pine were repeatedly obtained over 18 h. When the sequentially acquired spectra were overlayed, peak size or distribution did not change, indicating that P speciation or concentration did not change during the spectral acquisition time period. Nonmycorrhizal plants were shown to have one dominant peak, vacuolar Pi (Fig. 2). PolyP was not observed in any spectra acquired from nonmycorrhizal plants. In roots from all plants grown in soil amended with 136 mg P/kg, no polyP was present because all inoculated plants failed to form mycorrhizae. Areas under the Pi and polyP peaks were measured and used to indicate relative concentrations of P within these two phosphate groups (5, 22-24). For mycorrhizal trees, the ratio of Pi/polyP was 1.86 and 1.81 when grown in P unamended soil and 34 mg P/kg, respectively. Analysis of variance on the ratios from three replicate root sets at the two fertility levels indicated that the ratios were not significantly different between trees grown in the two soils (P > 0.1).

b

DISCUSSION

C

L

0

I

-10

-20

-30

X

J

-'40

-50

-60

Figure 2. A, 31P NMR spectrum of nonmycorrhizal roots of red pine seedlings. The dominant peak is vacuolar orthophosphate. Note the absence of the polyphosphate peak seen in spectra B and C. B, 3P NMR spectrum of roots from mycorrhizal red pine seedlings grown in P unamended soil. The dominant peak (a) is vacuolar Pi, and the peak seen to the right (b) is polyP produced by the fungal symbiont. C, 3P NMR spectrum of roots from mycorrhizal red pine seedlings grown in soil amended with 34 mg/kg P. The dominant peak (a) is vacuolar Pi and the peak seen to the right (b) is polyP produced by the fungal symbiont.

pine roots, a large peak was observed resonating at -17 ppm and was identified as vacuolar Pi. The second peak at -40 ppm was identified as polyP, based on previously determined peak assignments (15, 17, 18) for linear polyPs. A third peak was also observed on the shoulder of the vacuolar Pi peak (-15 ppm) in some of the spectra. It was identified as cytoplasmic Pi (23). The presence of two main peaks for Pi and polyP is consistent with previous reports for H. crustiliniforme grown in culture (17, 18). Peak identifications were confirmed by the Ti. The calculated Ti values for vacuolar Pi and polyP were found to be similar to previous reports for H. crustiliniforme grown in culture (Table I) (17, 18). Two small, unidentified peaks at -24 and -28 ppm were present in some samples. To confirm the stability of the vacuolar Pi and polyP over

The dominant P-containing compound in both mycorrhizal and nonmycorrhizal red pine roots was vacuolar Pi (-17 ppm). This finding is consistent with observations made by others for root tissue other than actively growing root tips (14, 22). The Pi in fungi has the same chemical shift as the vacuolar Pi for root tissue, so it was not possible to distinguish between fungal and root vacuolar Pi in these spectra (17, 18). In both sets of mycorrhizal roots, a second major peak (-40 ppm) was observed that was identified as polyP based on previously determined peak assignments for linear polyphosphates (15, 17, 18). This peak was absent in nonmycorrhizal roots of trees grown both in P unamended and amended soil (Fig. 2), indicating that the polyP is of fungal origin. The presence of two dominant peaks (Pi and polyP) is consistent with 31P NMR spectra acquired from the mycorrhizal fungi H. crustiliniforme, H. cylindrosporum, and Cennococcum graniforme grown in culture (17, 18). When grown in P-unamended soil, mycorrhizal trees had significantly greater dry weights than nonmycorrhizal trees grown in the same soil (16). Mycorrhizal seedlings were smaller, however, when grown in P unamended soil than in soil amended with 34 mg P/kg or 136 mg P/kg, indicating that the full growth potential had not yet been achieved and

Table I. Ti Times Comparison of Ti for phosphate groups found in H. arenosa-red pine mycorrhizae with times reported for H. crustiliniforme grown in culture.a Tl H. arenosa/red

Vacuolar Pi PolyP a Ref. 18.

pine mycorrhizae

H. crustiliniformea

0.4 0.06

0.15 0.03

MACFALL ET AL.

716

seedling growth was P-limited in unamended soil (Fig. 1). The fungal symbiont in P unamended soil had accumulated P as polyP, despite the evidence that shoots and total biomass of the colonized trees had not achieved maximum dry weights in the low P soil and, therefore, were a potential sink for additional P. Even though the mycorrhizal seedlings grown in the P unamended soil had improved P nutrition compared to the nonmycorrhizal seedlings, available soil P was still limiting growth. Nutrient limitations for growth were further reflected in the P tissue concentrations. Mycorrhizal plants grown in P unamended soil had greater root and shoot tissue P concentrations than nonmycorrhizal seedlings grown in similarly P unfertilized soil, but were less than for either mycorrhizal or nonmycorrhizal seedlings grown in P amended soil (Table II). Clearly, even though the fungal symbiont had increased the P concentration and seedling biomass when grown in the P unamended soil, available P in the soil was preventing the seedlings from reaching their full growth potential. Under these P-limiting conditions, however, the fungus had sequestered a portion of the available P as polyP within the fungal vacuole of the mycorrhizal roots. Maintainance of a relatively constant internal Pi concentration has been described for fungi grown in culture, with the polyP acting as a sink for removal of Pi (1). It was proposed that where excess intracellular Pi exists, polyP synthesis provides a mechanism for removal and storage of Pi. When transferred to a P-limiting medium, the stored polyP is hydrolyzed, releasing Pi and maintaining the Pi concentration within the fungal cell. A similar relationship has been observed by others when fungal mycelia were transferred from a basal medium to P-deficient medium (18, 19). In the present study, however, the shoots of seedlings grown in the P unamended soil were a potential Pi sink, as shown by the observation that maximum growth had not been achieved (Fig. 1) and by the low P concentrations of the root and shoot tissue. The Pi/polyP ratios in the mycorrhizal roots, however, were the same in both P amended and unamended soil rather than increased in trees most limited by soil P. The formation of polyP under P-limiting conditions also suggests that the fungus reduced the Pi pool available for transport to the host and changed the P compartmentaliza-

Table II. Tissue Concentrations of P P concentrations (jAg/g) in mycorrhizal and nonmycorrhizal red pine seedlings grown with varied P fertilizer applications. Data have been reprinted from ref. 16 with permission from the Canadian journal of Botany. Added P 0

34

136

mg/kg Inoculated Roots Shoots Noninoculated Roots Shoots

Plant Physiol. Vol. 100, 1992

tion within the fungal/host system. When plants were grown in P unamended soil, root P concentrations of mycorrhizal seedlings were greater than root P concentrations for the nonmycorrhizal seedlings. For the mycorrhizal seedlings, however, with a root Pi/polyP ratio of 1.8, the polyP component is 35% of the Pi+polyP pool. Although there are many P-containing compounds that cannot be detected by in vivo 31P NMR due to low mobility or low concentrations, spectra of perchloric acid extracts of plant tissue have shown that Pi is in significantly greater concentration than any other P compound (20). This indicates that although the roots of mycorrhizal seedlings grown in P unamended soil had a higher P concentration and content than nonmycorrhizal seedlings, a significant portion of the total P was in the form of fungal polyP. The production of polyP by the fungus would, therefore, make the pool of translocatable Pi in mycorrhizal and nonmycorrhizal roots nearer to being equal to each other than it would appear from tissue analysis only. When grown in soil amended with 34 mg P/kg, the total root tissue P concentration of mycorrhizal and nonmycorrhizal plants was the same (Table II) (16). This implies that when the polyP component of the Pi+polyP pool is subtracted (based on a Pi/polyP ratio of 1.8), the pool of potentially translocatable root Pi would be less in mycorrhizal plants than in nonmycorrhizal plants when grown at middle levels of applied P. This observation suggests that the fungus, through synthesis of polyP, was mediating Pi compartmentalization within the mycorrhizal roots and was regulating amounts of Pi available for transport to the shoots and for seedling growth. The greater P content in the seedlings grown at mid-P compared to seedlings grown in unamended soil can also be seen in the spectra. The signal-to-noise ratio per 100,000 pulses for vacuolar Pi was 10 ± 1 for seedlings grown in P unamended soil compared to 25 ± 2 for seedlings grown at mid-P. The greater ratio in spectra acquired of mid-P mycorrhizal roots shows that there was more vacuolar Pi in mycorrhizal seedlings grown at mid-P than when grown in unamended soil; however, it is not possible to determine if the Pi is fungal or in the root tissue. As the ratio of vacuolar Pi/ poly P was the same for seedlings grown in both soils, this would imply greater polyP synthesis by the fungus with P fertilization. This work has demonstrated that even under soil P-limited conditions for seedling growth, significant P is accumulated as polyP by the fungal symbiont. Greater vacuolar Pi and polyP content was found in mycorrhizal seedlings with P application to the soil. It is the first report relating the soil P status with seedling growth and nutrition and the impact of the ectomycorrhizal association on P partitioning. This finding suggests that the fungal partner not only improves the P nutrition of mycorrhizal seedlings when grown under Plimiting conditions, but that it also plays a role in regulating P compartmentalization and translocation.

1194 933

1581 1180

1994 1654

ACKNOWLEDGMENTS

796 759

1555 1277

2166 1928

The authors would like to thank Dr. Jim Otvoss for use of his NMR spectrometer. They would also like to thank Mark Miller and the Lambert Spawn Company of Coatesville, PA, for generous

PHOSPHORUS IN MYCORRHIZAL AND NONMYCORRHIZAL ROOTS donation of the fungal inoculum used in this study. Also, thanks to Drs. Jaya Iyer and Jennifer Parke for helpful discussion and insight in the development of this project.

17.

LITERATURE CITED 1. Beever RE, Burns DJW (1980) Phosphorus uptake, and utilization by fungi. In HW Woolhouse, ed, Advances in Botanical Research. Vol 8. Academic Press, New York, pp 127-219 2. Bowers LJ, Melhuish JH Jr (1988) Silicon content of bald cypress compared to loblolly pine and southern red oak. Trans Ky Acad Sci 49: 107 3. Chilvers GA, Harley JL (1980) Visualization of phosphate accumulation in beach mycorrhizas. New Phytol 84: 319-326 4. Finlay RD, Read DJ (1986) The structure and function of the vegetative mycelium of ectomycorrhizal plants. II. The uptake and distribution of phosphorus by mycelial strands interconnecting host plants. New Phytol 103: 157-165 5. Gadian GD (1982) Nuclear Magnetic Resonance and its Applications to Living Systems. Oxford University Press, Oxford 6. Harley JD, McCready CC (1952) The uptake of phosphate by excised mycorrhizal roots of the beech. III. The effect of the fungal sheath on the availability of phosphate to the core. New Phytol 51: 342-348 7. Harley JL, McCready CC (1981) The relationship between phosphate absorption and oxygen uptake in excised beech mycorrhizas. New Phytol 88: 675-681 8. Harley JL, McCready CC (1981) Phosphate accumulation in Fagus mycorrhizas. New Phytol 89: 75-80 9. Harley JL, Smith SE (1983) Mycorrhizal Symbiosis. Academic Press, New York 10. Kime MJ, Ratcliffe RG, Williams RJP (1982) The application of 31P nuclear magnetic resonance to higher plant tissue. I. Detection of spectra. J Exp Bot 33: 670-681 11. Kramer PJ, Wilbur KM (1949) Absorption of radioactive phosphorus by mycorrhizal roots of pine. Science 110: 8-9 12. Kulaev IS (1979) The Biochemistry of Inorganic Polyphosphates. John Wiley and Sons, New York 13. Ling-Lee M, GA Chilvers GA, Ashford, AE (1975) Polyphosphate granules in three different kinds of tree mycorrhizas. New Phytol 75: 551-555 14. Loughman BC (1987) The application of in vivo techniques in the study of metabolic aspects of ion absorption in crop plants. Plant Soil 99: 63-74 15. MacDonald JC, Mazurek M (1987) Phosphorus magnetic resonance spectra of open-chain linear polyphosphates. J Mag Res 72: 48-60 16. MacFall JS, Slack SA, Iyer J (1991) Effects of Hebeloma arenosa

18.

19.

20. 21. 22.

23. 24.

25. 26.

27.

28.

29.

30.

71 7

and phosphorus fertility on growth of red pine (Pinus resinosa) seedlings. Can J Bot 69: 372-379 Martin F, Canet D, Rolin D, Marchal JP, Larher F (1983) Phosphorus-31 nuclear magnetic resonance study of polyphosphate metabolism in intact ectomycorrhizal fungi. Plant Soil 71: 469-476 Martin F, Marchal JP, Tyminska A, Canet D (1985) The metabolism and physical state of polyphosphates in ectomycorrhizal fungi. A 31P nuclear magnetic resonance study. New Phytol 101: 275-290 Mousain D, Salsac L (1985) Growth and accumulation of phosphorus in the ectomycorrhizal fungus Pisolithus tinctorius. In V Gianinazai-Pearson, S Gianinazzi, eds, Mycorrhizae: Physiology and Genetics. Institut National de la Recherge Agronomique, Paris, France, pp 357-361 Pfeffer P (1989) Magnetic Resonance in Agriculture. CRC Press, Boca Raton, FL Pfeffer P, Tu S, Gerasimowicz W, Boswell RT (1987) Effects of aluminum on the release and-or immobilization of soluble phosphate in corn root tissue. Planta 172: 200-208 Roberts JKM (1984) Study of plant metabolism in vivo using NMR spectroscopy. Annu Rev Plant Physiol 35: 375-386 Roberts JKM, Jardetsky 0 (1981) Monitoring of cellular metabolism by NMR. Biochim Biophys Acta 639: 53-76 Roberts JKM, Ray PM, Wade-Jardetsky N, Jardetsky 0 (1980) Estimation of cytoplasmic and vacuolar pH in higher plant cells by 31P NMR. Nature 283: 869-872 Roby C, Bligny R, Douce R, Tu S, Pfeffer P (1988) Facilitated transport of Mn2+ in sycamore (Acer pseudoplantanus) cells and excised maize root tips. Biochem J 252: 401-408 Rolin D, le Tacon F, Larher F (1984) Characterization of the different forms of phosphorus in the mycelium of the ectomycorrhizal fungus Hebeloma cylindrosporum, cultivated in pure culture. New Phytol 98: 335-343 Strullu DG, Gourret JP, Garrec JP (1981) Microanalyse des granules vacuolaires des ectomycorhizes, endomycorhizes et endomycothalles. Physiol Veget 19: 367-378 Strullu DG, Grellier B, Romand C, Garrec JP (1985) Sur le modele unifie de fonctionnement des mycorhizes: Synthese in vitro de granules metachromatiques vacuolaires. In V Gianinazzi-Pearson, S Gianinazzi, eds, Mycorrhizae: Physiology and Genetics. Institute National de la Recherche Agronomique, Paris, France, pp 441-444 Strullu DG, Harley JL, Gourret JP, Garrec JP (1983) A note on the relative phosphorus and calcium contents of metachromatic granules in Fagus mycorrhiza. New Phytol 94: 89-94 Wilde SA, Corey RB, Iyer JG, Voigt GK (1979) Soil and plant analysis for tree culture. Ed 5. Oxford and IBH Publishing Co., New Delhi, India