CSIRO PUBLISHING
Functional Plant Biology, 2007, 34, 457–464
www.publish.csiro.au/journals/fpb
Relative amounts of soluble and insoluble forms of phosphorus and other elements in intraradical hyphae and arbuscules of arbuscular mycorrhizas Megan H. RyanA,D , Margaret E. McCullyB and Cheng X. HuangC A
School of Plant Biology M081, University of Western Australia, 35 Stirling Hwy, Crawley, WA 6009, Australia. CSIRO Plant Industry, GPO Box 1600, Canberra, ACT 2601, Australia. C Electron Microscopy Unit, Research School of Biological Sciences, Australian National University, Canberra, ACT 0200, Australia. D Corresponding author. Email:
[email protected] B
Abstract. Transport of phosphorus (P) into host plants and its release to root cells is an important function of arbuscular mycorrhizal fungi (AMF). However, relatively little is known about the forms and water solubilities of P compounds in specific locations in the intraradical fungal structures. We determined concentrations and solubility of P components in these structures in white clover (Trifolium repens L.). Plants were grown in the field (colonised by indigenous AMF) or in the glasshouse (inoculated with Glomus intraradices). Mycorrhizas were cryo-fixed in liquid nitrogen immediately (control) or after treatments designed to destroy cell membranes and extract solubles. Thirty to 70% of total P in hyphae and 100% in arbuscules was not extracted. The unextracted proportion of P was higher in the inoculated plants suggesting an environmental effect. It is proposed that the large component of non-extractable P in the arbuscules is involved in the tight regulation of inorganic P release to the host cells. In control roots magnesium, potassium and P were present in hyphae in molar ratios 1 : 2 : 4, further evidence that this relationship may be universal for AMF, and that other P-balancing cations are present but undetectable by the analytical technique. Additional keywords: Glomus intraradices, phosphorus transfer at fungal/host interface, mycorrhizal roots, quantitative analytical cryo-SEM, Trifolium repens, polyphosphate, white clover.
Introduction Significant advances have been made in our understanding of how arbuscular mycorrhizal fungi (AMF) mediate the transfer of phosphorus (P) from the soil to host plant roots. It seems that inorganic P (Pi ) is absorbed by the extraradical mycelium of AMF with at least some of the resulting excess P in the fungal cytoplasm being very rapidly transferred into vacuoles and stored as polyphosphate (polyP) (Ezawa et al. 2003). PolyPs are salts of the linear polyphosphatic acids and may contain tens or many hundreds of phosphate residues linked by high-energy bonds. PolyP is transported from the external fungal mycelium to the intraradical fungal structures, possibly by a system of tubular vacuoles (Uetake et al. 2002) and, perhaps, acidic spherical vesicles (Saito et al. 2004) before being hydrolysed. The Pi released from polyP is transferred to the host plant root cells (Ohtomo and Saito 2005) from hyphae and arbuscules within the periplasmic space of host root cells (Smith et al. 2001) or, in part, from intercellular hyphae (Ryan et al. 2003). However, much remains to be elucidated about these processes, particularly the role of polyP or other insoluble P compounds in the controlled release of Pi to the root cells. It has long been assumed that AMF, like other fungi (e.g. Doonan et al. 1979), use polyP for P storage (Cox et al. 1974; © CSIRO 2007
Ling-Lee et al. 1975; Callow et al. 1978). Use of various techniques to detect polyP has supported this view (e.g. Ling-Lee et al. 1975; Rasmussen et al. 2000; Ezawa et al. 2003; Ohtomo and Saito 2005). Large molecular weight polyP may reduce the osmotic burden of P storage, allowing vacuoles to remain small and move easily through hyphae during cytoplasmic streaming (Cramer and Davis 1984). Nuclear magnetic resonance (NMR) studies suggest that at least some of the polyP in mycorrhizal hyphae is present as soluble, relatively short chains (around 15 Pi subunits) (Solaiman et al. 1999; Rasmussen et al. 2000). However, while NMR can distinguish polyP from other P-containing compounds it cannot detect the insoluble or immobile polyP known to be present in AMF (Solaiman et al. 1999; Ohtomo and Saito 2005). Presumably, large molecular weight polyP could function as a storage form of P within the fungal structures. Indeed, polyP is often observed to be granular (i.e. insoluble) (B¨ucking and Heyser 1999; Solaiman et al. 1999). Overall, we know little about the proportion of total fungal P that is stored as short chain, long chain or granular polyP and whether this differs between intraradical hyphae and arbuscules. It is also not known whether forms of P other than polyP are common in intraradical mycorrhizal
10.1071/FP06242
1445-4408/07/050457
458
Functional Plant Biology
M. H. Ryan et al.
structures, or whether the above results (generally obtained from glasshouse and laboratory studies) can be extrapolated to field conditions. For the first time we have determined the concentrations of P and other elements in situ in intraradical fungal structures and host cells of fully hydrated arbuscular mycorrhizas (Ryan et al. 2003). This study, which used quantitative cryo-analytical scanning electron microscopy (SEM), showed that concentrations of P in intraradical hyphae were unexpectedly high, up to 600 mM, and magnesium (Mg), potassium (K) and P were present in the ratio of 1 mol Mg to 2 mol K to 4 mol P. However, while this study showed total concentration of each of the elements analysed, it provided no information about the chemical forms in which the elements were present, or their solubility in aqueous solution. Presumably, as is the case in plant cells, the detected K was ionic and constituted the major part of the cationic component of the solutes which maintained turgor pressure (Epstein and Bloom 2005). For AMF, high turgor pressue would be needed for hyphae to push through intercellular spaces and root cell walls, and for arbuscules to expand in the periplasmic space of living cells. However, other elements, particularly P, cannot be assumed to be present primarily in ionic form. Knowledge of the relative amounts of soluble and insoluble P and associated elements in intraradical mycorrhizal structures is essential for development of a better understanding of the dynamics of P storage and release to host cells in arbuscular mycorrhizas. This is particularly pertinent in light of the recent advances in knowledge of gene expression patterns in relation to nutrient transfer at the fungal–host interface (e.g. Aono et al. 2004; Poulsen et al. 2005). In the present study we have extended the use of quantitative cryo-analytical SEM to compare concentrations of P, K, Mg, sulfur (S) and sodium (Na) in intraradical hyphae, arbuscules and host cells of control mycorrhizas and those pretreated to destroy membrane integrity and wash out soluble forms of these elements. Comparisons were made for white clover (Trifolium repens L.) grown both in the field (colonised by indigenous AMF) and in the glasshouse (inoculated with Glomus intraradices).
sand contained 4 mg kg−1 bicarbonate extractable P (Colwell 1963), 7 mg kg−1 mineral N, 1.5% organic carbon and had a pH of 5.0 (CaCl2 ). Nutrients were added to the Lancelin sand using three stock solutions: (1) KH2 PO4 78 mM, K2 SO4 122 mM, (2) CaCl2 192 mM, and (3) CuSO4 .5H2 O 2.6 mM, ZnSO4 .7H2 O 5.2 mM, MnSO4 .7H2 O 10.8 mM, CoSO4 .7H2 O 0.4 mM, Na2 MoO4 .2H2 O 0.25 mM and MgSO4 .7H2 O 24.2 mM. Each solution was added separately to the surface of the Lancelin sand at 10 mL 3 kg−1 and mixed. This mixture was then combined with an equal weight of the prepared river sand. The G. intraradices Schenck and Smith (DAOM 181602) isolate had been subcultured from an axenic culture on transformed roots obtained from Professor J. A. Fortin, University of Montreal, Canada. Inoculum was bulked up using leeks (Allium porrum L.) growing in the mix of river and Lancelin sand. To inoculate the pots, 300 g of leek pot culture (sand, roots, hyphae and spores) was mixed with 5.4 kg of the sand mix and placed in five 1-kg white free-draining pots with muslin in the bottom. The pots were placed in sun bags (Sigma #B7026, Sigma, St Louis, USA) and watered to 10% soil water content. White clover seeds were surface sterilised and pregerminated and five seeds planted in each pot. A suitable rhizobium inoculum was added as slurry on the sand surface 3 weeks after planting. Modified Long Ashton nutrient solution minus P was added once a fortnight after 2 weeks of growth (10 mL kg−1 soil): K2 SO4 2 mM, MgSO4 .7H2 O 1.5 mM, CaCl2 .2H2 O 3 mM, FeEDTA 0.1 mM, (NH4 )2 SO4 4 mM, NaNO3 8 mM, H3 BO3 46 µM, MnCl2 .4H2 O 9 µM, ZnSO4 .7H2 O 8 µM, CuSO4 .5H2 O 0.3 µM and Na2 MoO4 .2H2 O 0.01 µM (Cavagnaro et al. 2001). Plants were grown for 10 weeks in a glasshouse maintained at 22◦ C/18◦ C day/night. Light levels ranged from 300 to 1000 µmol m−2 s−1 PAR. At harvest a subsample of ∼2 m of roots was cleared in KOH and stained with aniline blue (Grace and Stribley 1991). The percentage of root length colonised by AMF was then assessed by the gridline intersect method at 100× magnification (Giovannetti and Mosse 1980). Colonisation averaged 49%.
Materials and methods Field-grown clover colonised by indigenous AMF Roots colonised by indigenous AMF were collected from white clover grown in a garden in Canberra (35◦ 15 S, 149◦ 8 E) in a loamy soil with 29 mg kg−1 bicarbonate-extractable P (Colwell 1963), 34 mg kg−1 mineral N, 3.2% organic carbon and a pH of 4.7 (CaCl2 ). Plants were watered regularly but received no fertiliser. Plants were harvested in early May 2004 (mid-autumn). In the preceding month the mean daily maximum and minimum temperatures were 22.3◦ C and 8.1◦ C, respectively, and 2.4 mm of rainfall were recorded (Bureau of Meteorology, Canberra Airport recording station).
Root extraction treatments The idea of extracting solubles from fresh mycorrhizal roots before cryo-fixation and quantitative X-ray microanalysis came to us as a result of earlier observations of root pieces stained by Schiff ’s reagent and subsequently observed by cryo-SEM. These tissues lacked any detectable K when analysed by X-ray microanalysis, but P was still present. It became obvious that quantitative cryo-analytical microscopy offered the possibility of comparing concentrations of elements in control and differentially extracted tissues to determine the proportion of specific elements present in soluble or insoluble form. We subsequently found that a comparable method had been used to qualitatively localise soluble and insoluble forms of Mn in hyperaccumulating plants (Memon and Yatazawa 1982). Plants from the field or glasshouse were carefully excavated so that their roots remained in situ in a ball of soil and sand and quickly brought to a nearby laboratory where they were processed as described by Ryan et al. (2003). Main roots ∼10 cm
Glasshouse-grown clover colonised by Glomus intraradices Washed river sand was autoclaved at 121◦ C, and Lancelin sand steamed at 80◦ C, each for 1 h on two successive days. Both sands were dried overnight at 100◦ C and stored. The Lancelin
Solubility of P in hyphae and arbuscules of AMF
long with attached branch roots were cut from ∼10 nodes that supported healthy leaves. We used two standard methods for extracting solutes from plant tissues (fast or slow freezing, followed by water washout), as well as treatment with Schiff ’s reagent followed by water washout. Roots were cryo-fixed in liquid N2 (LN2 ) immediately (Control), or after each of the following extraction treatments: (1) snap-frozen in LN2 and thawed at room temperature in agitated tap water for 30 min (Liquid N) (field and glasshouse roots); (2) frozen at −20◦ C for 2 h and thawed as above (Freezer) (field-grown roots only); (3) immersion in Schiff’s reagent (#191203S lot 940339844; BDH Laboratory Supplies, Poole, UK) for 0.5 or 2 h followed by immersion in agitated tap water for 30 min (Schiff ’s) (field and glasshouse roots). Roots were then cut into 2 cm lengths under LN2 and either processed immediately for cryo-SEM and microanalysis or placed in vials and stored in a cryo-store and processed later. Cryo-SEM and X-ray microanalysis For cryo-SEM and microanalysis, 2 mm lengths were cut from the frozen roots under LN2 and affixed to stubs with low-temperature Tissue Tek (Miles Inc., Elkhart, IA, USA), transferred to a cryo-microtome under LN2 and planed at −90◦ C to a smooth transverse face with glass and diamond knives. Planed specimens were transferred under LN2 to a cryo-transfer unit (Oxford Instruments, Eynsham, UK) and then to the stage of the cryo-SEM (JEOL 6400, JEOL Ltd, Tokyo, Japan). Frost was removed by brief, carefully controlled, gentle etching at −90◦ C and the specimens then cooled to −170◦ C and coated with evaporated, high purity aluminium. Images were recorded between 10 and 15 kV to a digital recorder (ImageSlave, OED Pty Ltd, Hornsby, NSW, Australia). Fungal structures (intercellular and periplasmic hyphae, and arbuscules) and host root cortical cell vacuoles were analysed with a Link eXL system (Oxford Instruments) using the Be window. Spectral data for K, P, Mg, Ca, Na and S were converted to elemental concentrations with frozen standards prepared exactly as the specimens. As the analysis does not distinguish between ions and elements in insoluble form, the concentrations of elements in all structures are expressed in mM. The lower limit of reliable quantitation of elements by the X-ray microanalysis was considered to be 10 mM and hence values below 10 mM were ascribed a notional value of 5 mM. (For further details of preparation methods for cryo-SEM and analysis see Huang et al. 1994 and McCully et al. 2000.) Experimental design and data analysis Analyses were conducted only on planed faces that included regions colonised by AMF. Analyses were done on material from 16 ‘Control’ roots, 10 ‘Schiff ’s’ roots, five ‘Liquid’ N roots and four ‘Freezer’ roots from field-grown plants, and from seven Control roots, four Schiff’s roots and four Liquid N roots from glasshouse-grown plants inoculated with G. intraradices. Three different structures/areas in each transverse face were analysed if present. (1) The contents (not including the wall) of intraradical mycorrhizal hyphae growing either intracellularly (inside the cell wall of a cortical cell, but still outside the symplast – i.e. in periplasmic space) or intercellularly (in root
Functional Plant Biology
459
intercellular spaces). (2) Areas that contained a dense mass of the small hyphae that formed arbuscules. In contrast to the analyses of intraradical hyphae, cell walls of the small hyphae that constituted the arbuscules were analysed along with their contents, as individual hyphae could not always be distinguished in the extracted roots. Arbuscules that appeared to be active (see Ryan et al. 2003) were chosen for analyses, although in the extracted roots these were often difficult to distinguish confidently from arbuscules that were beginning to deteriorate. (3) The vacuoles of uncolonised cortical cells. Results are presented from only metabolically active root cortical tissues, arbitrarily defined as those with cell vacuole [K] >20 mM. As this could not be measured in the extracted roots, a subjective judgement was made on the basis of whether roots had an intact cortex. All data analysis was conducted with GENSTAT version 7.2 (VSN International, Hemel Hempstead, UK) (Lawes Agricultural Trust) and graphs generated using SigmaPlot version 7.101 (SPSS, Chicago, IL, USA). No outliers were omitted from datasets unless stated. It is important to note that the very time-consuming processes of sample preparation coupled with the relatively low probability of a planed face containing the desired structures for analysis resulted in low sample sizes. Results Structural features The appearance of the cryo-planed faces of root pieces was similar for both the field-grown and inoculated roots. Control roots contained well preserved AM fungal structures, including inter- and intracellular hyphae and arbuscules, all consistent with those described by Ryan et al. (2003) (Fig. 1A, B). The profiles of root cells were not distorted and the vacuoles of healthy uncolonised cells, as well as colonised cells, showed the lines and dots indicative of solutes sequestered by ice crystal formation during freezing (Fig. 1A, B). In the extracted roots, sequestered solutes were very much less evident (Fig. 1C, D). The intercellular and periplasmic hyphae did not differ in diameter between the Control and extracted roots and they retained their original shape in the extracted roots, probably as a result of their thick walls (Fig. 1). Thus, in profile, triangular intercellular spaces containing round hyphae could be observed in both the Control and extracted roots. In Control roots, the many small hyphae that made up the arbuscules could be clearly observed as discrete transverse or tangential profiles; where these arbuscules did not fill the whole cell, dense dots and lines of solutes in the host cortical cell vacuole were evident (Fig. 1A, B). In extracted roots, the hyphae of the arbuscules were broken up and distorted, but the overall form of the arbuscules was retained, and although the fungal material had often dispersed throughout the entire cell, there was no difficulty in distinguishing these cells from those that were not colonised (Fig. 1C, D). Elemental analyses Control roots Average [P] in both field-grown and inoculated roots was between 150 and 175 mM in hyphae, 50 and 60 mM in arbuscules,
460
Functional Plant Biology
M. H. Ryan et al.
A
C
a
a a
a
B
D
Fig. 1. Cryo-planed transverse faces of white clover roots colonised by arbuscular mycorrhizal fungi (AMF) imaged by cryo-scanning electron microscopy. A, B are Control roots, cryo-fixed while alive. C, D are similar roots to A and B from which soluble components were extracted by Schiff ’s reagent (C) or by thawing following freezing in liquid N (D) before cryo-fixation (see text). Mycorrhizal structures can be seen in the mid and inner cortex of all roots including hyphae of AMF (indicated by arrows) and arbuscules (arrowheads in A and C, labelled ‘a’ in B and D). *, xylem vessels. Some small unsublimed frost deposits remain on the surface of the face in D. Bar = 20 µm in A and C, 10 µm in B and D.
but below the level of reliable detection (i.e.
