•Extraradical mycelia from different ectomycorrhizal (ECM) roots coexist and interact under the forest floor. We investigated structural connections of conspecific mycelia and translocation of carbon and phosphorus between the same or different genets.
•Paired ECM Pinus thunbergii seedlings colonized by the same or different Pisolithus isolates were grown side by side in a rhizobox as their mycelia contacted each other. 14CO2 or 33P-phosphoric acid was fed to leaves or a spot on the mycelium in one of the paired seedlings. Time-course distributions of 14C and 33P were visualized using a digital autoradiographic technique with imaging plates.
•Hyphal connections were observed between mycelia of the same Pisolithus isolate near the contact site, but hyphae did not connect between different isolates. 14C and 33P were translocated between mycelia of the same isolate. In 33P-fed mycelia, accumulation of 33P from the feeding spot toward the host ECM roots was observed. No 14C and 33P translocation occurred between mycelia of different isolates.
•These results provide direct evidence that contact and hyphal connection between mycelia of the same ECM isolate can cause nutrient translocation. The ecological significance of contact between extraradical mycelia is discussed.
Ectomycorrhizal (ECM) fungi form a mutualistic symbiosis with the roots of various tree species. Fungi receive photosynthetically fixed carbon from the host tree and in return supply nutrients from the soil. This nutritional exchange results in a variety of beneficial outcomes such as promotion of host growth and survival. This exchange largely depends on the functions of extraradical mycelium, which is poorly understood and requires more research for a deeper understanding of ECM symbiosis.
An ECM root tip projects many hyphae into the soil and generates an extraradical mycelium. Using host-derived carbon, the fungal genet of each ECM root tip develops its own extraradical mycelium in the soil. Thus, each mycelium is provided with fixed carbon from the host independently of other mycelia and develops as an independent structural unit. When an extraradical hypha encounters another fine root of either the host or neighbouring tree, it develops into a new ECM root tip. Repeating such an infection process, the mycelium may sometimes connect several host trees by its hyphae under the forest floor, resulting in a large common mycelial network (CMN) that has been hypothesized to allow net transfer of C and nutrients between host plants (Newman, 1988; Simard et al., 1997; Leake et al., 2004; Simard & Durall, 2004; Teste et al., 2010).
In some ECM fungal species, the ECM root tip and mycelium appear fragile under the forest floor. Zhou et al. (2001) found that Suillus grevillei mycelia and ectomycorrhizas beneath the sporocarps diminish within 1 yr. Guidot et al. (2001, 2003) also observed that the disappearance of Hebeloma cylindrosporum sporocarps is associated with the disappearance of the corresponding ectomycorrhizas within 1 yr. Koide et al. (2007) reported that the mycelium of some ECM fungi changed temporally in a Pinus resinosa plantation during a 13 month period. Pickles et al. (2010) revealed that Scots pine root tips colonized by some ECM fungal species in the forest are distributed as patches and that the size and location of the patches change dynamically over time. Enlargement, fragmentation and extinction of mycelia may frequently occur in extraradical mycelia. Fragmentation of an extraradical mycelium would produce several offspring mycelia of the same genet as the original mycelium. These fragmented mycelia can coexist and grow in the same area and later become reassociated. Thus, mycelia of the same genet may repeatedly be fragmented and reassociated under the forest floor.
The diversity of coexisting ECM species and conspecific genets in an ECM fungal community is highly dynamic. Several studies have demonstrated rapid spatiotemporal turnover of the ECM fungal community (Redecker et al., 2001; Zhou & Hogetsu, 2002; Izzo et al., 2005; Pickles et al., 2010). This turnover may reflect dynamic changes in extraradical mycelia and fine root turnover, as well as mycelial enlargement, fragmentation and extinction.
Changes in the extraradical mycelia may result in underground contact between different kinds of extraradical mycelium, that is, between the same genet, conspecific genets and different species. Both compatible and incompatible interactions among fungal mycelia generally occur by contact. Somatic compatibility, a phenomenon that occurs in filamentous fungi, may accompany hyphal fusion and expand structural continuity to neighbouring mycelia (Olsson, 1999; Sen et al., 1999; Glass et al., 2004; Fleissner et al., 2008). In ECM fungi such as Hydnum rufescens, Laccaria amethystina, Sarcodon imbricatus and Tuber borchii, somatic hyphal fusion has been observed during self-anastomosis in pure culture (Raidl & Agerer, 1992; Sbrana et al., 2007). If such somatic fusion occurs among neighbouring mycelia in the forest floor, it would structurally expand the mycelial network. Somatic incompatibility has also been observed between mycelia of different fungal species and conspecific fungal genets (Worrall, 1997; Dahlberg, 1999).
Most functions of ECM symbiosis are derived from nutrient acquisition involving the extraradical mycelium. Therefore, if structural alteration of the extraradical mycelium caused by somatic compatibility directly influences translocation of photosynthate and nutrients, interactions between the extraradical mycelia can alter the ECM symbiosis not only structurally but also functionally. For example, because a fungal genet can be separated into several physiological fragments in the soil (Cairney, 2005), somatic hyphal fusion among fragmented mycelia would restore functional activities of the original mycelium as well as structural integrity. However, the relationship between mycelial connection and nutrient translocation is unknown.
Nutrient exchange between the host and ECM fungus has been investigated using tracer experiments. Using autoradiography of 14C and 32P, Finlay & Read (1986a,b) clearly demonstrated that 14C-photosynthate fixed in host pine leaves and 32P-phosphate absorbed by mycelia were transferred through ECM root tips to the mycelia of colonizing Suillus spp. and to the host, respectively. Wu et al. (2001, 2002) applied another autoradiographic technique with imaging plates to a Pinus densiflora–Pisolithus symbiotic system and demonstrated changes in the time course of 14C-photosynthate translocation. This technique allowed quantification of changes in translocation over time for radioactively labelled substances such as 14C-photosynthate and 33P-phosphate.
This study investigates whether contact between the extraradical mycelia of the same genet or conspecific different genets causes structural connection and how the structural connection influences translocation within the symbiotic system. Using time-course autoradiography with imaging plates, we describe tracer experiments with 14C and 33P, and structural and functional interactions between mycelia of P. thunbergii seedlings colonized by Pisolithus isolates.
Materials and Methods
Preparation of fungal inocula
Two Pisolithus isolates from the Fukushima and Shiga provinces in Japan (named PF and PS, respectively) were provided by Dr Kazuhide Nara of the University of Tokyo (Japan). Shiga and Fukushima are > 500 km apart, and the isolates were considered to be from different populations. For phylogenetic identification of the isolates, the internal transcribed spacer (ITS) region of nuclear ribosomal DNA was sequenced for both isolates. Both sequences were aligned with 31 Pisolithus sequences used in Martin et al. (2002) by Clustal X2 (Larkin et al., 2007), and a phylogenetic dendrogram was constructed by the neighbour-joining method. Polymorphism analysis of both isolates was performed using a simple sequence repeat (SSR) marker P3 from Hitchcock et al. (2003).
Fungal inocula of PF and PS were prepared by culturing on modified Melin–Norkrans agar plates (Marx, 1969) at 25°C for 1 month.
Mycorrhiza synthesis and preparation of observation rhizoboxes
Pinus thunbergii Parlatore seedlings were grown in a plastic pot filled with an autoclaved mixture (1 : 1, v/v) of coppice soil obtained from the Koishikawa Arboretum of the University of Tokyo (black sandy loam, pH 5.3) and Shibanome soil (volcanic sand, pH 5.8–6.0; Setogahara, Gunma, Japan), both of which were autoclaved at 121°C for 90 min. Pine seedlings grown for 1 month were transferred to a rectangular flat rhizobox (140 × 205 × 15 mm) filled with the autoclaved soil mixture. The pine seedlings were inoculated with each fungal isolate by placing agar fragments (which include hyphae) onto the seedling roots. The inoculated seedlings were cultivated for 1–2 months in a growth chamber under a daily cycle of light (350–500 μmol m−2 s−1 of photosynthetically active radiation (PAR)) for 16 h at 25°C and darkness for 8 h at 23°C. Each seedling with ECM root tips was transplanted to a new flat rhizobox with two nonECM seedlings to allow for continuous ECM seedling production. ECM seedlings used in experiments were prepared by cocultivation of two nonECM seedlings with the ECM seedling in a rhizobox for 4 wk.
Each ECM seedling was transferred to the flat surface of a floral foam plate (Smithers-Oasis, Tokyo, Japan) fitted into the rhizobox. The foam plate was treated with a 1000-fold-diluted nutrient solution (6-10-5; Hyponex Japan, Osaka, Japan). After the mycelium had developed sufficiently, the extraradical mycelium was partially removed by vertically cutting the foam plate, as shown in Fig. 1. Two foam plates, each of which carried an ECM seedling with partially cut mycelium, were placed in a new rhizobox side by side. The cut surfaces of both plates were placed in direct contact. After 12 d, the surface of the rhizobox was photographed with a scanner (Offirio ES-10000G; Epson, Tokyo, Japan), and the rhizobox was used for a tracer experiment with either 14CO2 or 33P-phosphoric acid. Rhizoboxes were prepared for four ECM seedling combinations, designated PF–PF, PS–PS, PF–PS and PF–PS (the fed seedling or mycelium is underlined), each with three to four replicates.
For microscopic observations of interactions between mycelia, an additional set of rhizoboxes (PF–PF, PS–PS and PF–PS) was prepared. Mycelia at the contact line were observed using a stereomicroscope (MZ16; Leica, Houston, TX, USA). In addition, control rhizoboxes were prepared by placing a foam plate with an ECM seedling next to a plate without a seedling to estimate the distance to which the mycelium of the ECM seedling grew past the contact line.
14C feeding was performed in a draft chamber (23–25°C, 150–200 μmol m−2 s−1 PAR), as described by Teramoto et al. (2012). Briefly, 14CO2 was produced by adding 10% lactic acid with a syringe to 36.39 μg NaH14CO3 containing 925 kBq 14C in a silicon-plugged microtube. The microtube was then glued to the inside of a transparent polyethylene bag. The shoot in one of the paired seedlings in the rhizobox was enclosed in the polyethylene bag, and the silicon plug of the microtube was removed to release 14CO2 into the bag. After 2 h, unused 14CO2 was removed using a trap containing 1 N NaOH, and the bag was removed from the shoot. These seedlings were cultivated in the draft chamber under a 16h light : 8h dark photoperiod. To prevent photosynthetic re-fixation of 14CO2 by the unfed seedling, the shoot of the unfed seedling was enclosed in another polyethylene bag connected to an air pump and provided with 14CO2-free air throughout the experimental period.
33P-phosphoric acid feeding
To prepare seedlings for 33P-phosphoric acid feeding, a plastic dish (3.5 cm diameter, 1 cm deep) filled with the floral foam disk was buried in front of the mycelium in the foam plate (Fig. 1b). After more than half the dish was covered with the mycelium, circles of 2.0% agar (2.7 cm diameter, 2 mm thick) and then wet filter paper (3.0 cm diameter) were overlaid on the surface of the dish. H333PO4 was fed to the mycelium by adding 250 μl of its aqueous solution (0.16 ng, 925 kBq) to the surface of the filter paper.
14C and 33P autoradiography and radioactivity counting
After removing each rhizobox lid and covering its surface with a sheet of wrap film (Riken Vinyl, Tokyo, Japan) to prevent radioactive contamination, it was overlaid with an imaging plate (BASSR2040; Fuji Film, Tokyo, Japan). The surface of each rhizobox was exposed to the imaging plate for 90 and 60 min for 14C- and 33P-autoradiography, respectively. Standard filter paper circles (5 mm diameter) containing 1.24, 7.4 and 37 kBq [14C(U)]–sucrose or 1.85, 7.4 and 37 kBq H333PO4 were simultaneously exposed to each 14C- or 33P-labelled sample. An image analyser (FLA-2000; Fuji Film) visualized radioactivity distribution recorded on each exposed imaging plate. Each rhizobox was repeatedly autoradiographed up to 14 and 21 d after 14C and 33P feeding, respectively. Areas above and below the upper side of the floral foam plate were defined as above-ground and below-ground parts, respectively, and photostimulated luminescence (PSL) in both parts of each seedling was quantified from the autoradiograms using Multi Gauge V 3.1 software (Fuji Film) and converted to an absolute unit (Bq) based on the PSLs of standards. The area near the dish rim (5 mm wide) was excluded from the below-ground 33P radioactivity. Because all needles could not be flattened upon exposure, radiation absorption by air in the gap between needles and the imaging plate might cause some unavoidable underestimation of 14C or 33P radioactivity in the above-ground part.
Phylogenetic identification of Pisolithus isolates
Of the 673 base pairs in the ITS sequences of both PF and PS isolates, only one base pair differed: a ‘G’ in PF was replaced by a ‘C’ in PS at the 144th base pair of the ITS2 region (GenBank accession numbers AB683450 and AB683451). Phylogenetic analysis of both sequences along with 31 isolates collected worldwide from 11 Pisolithus spp. (Martin et al., 2002) showed that both isolates belonged to the species 5 clade (Martin et al., 2002; data not shown). In addition, polymorphic analysis showed that PF and PS had different SSR genotypes: ‘290/290’ and ‘289/289’, respectively (Supporting Information, Fig. S1). Thus, we concluded that PF and PS belong to the same Pisolithus species but to different genets.
Microscopic observations of interactions between extraradical mycelia
In the contact zone between paired mycelia of the same isolate, pre-existent hyphal cords began to elongate from the cut tips and connected with an elongating cord from the other mycelium 3 d after foam plate contact (Fig. 2a,b). The thickness of hyphal cords increased after hyphal connection, which was especially conspicuous in PS–PS rhizoboxes. Such connection occurred within several mm of the contact line; no extension of hyphal cords from the countering mycelium was observed > 1 cm from the contact line.
No hyphal cords were observed connecting mycelia at the contact zone between the mycelia of different isolates, PF–PS (Fig. 2c). In control rhizoboxes, which had only one seedling, cut hyphal cords and hyphae extended into the vacant foam plate up to 8 mm 14 d after contact (Fig. 3).
14C transfer between extraradical mycelia
Of the paired seedlings colonized by the same isolate (PF or PS), the seedling on one side was fed with 14CO2 12 d after mycelium contact. 14C was present only in the leaves of the fed seedling immediately after feeding. However, within 1 d, 14C was transported to the mycelia and mycorrhizas of both the fed and unfed seedlings. Intensive accumulation of 14C occurred in mycorrhizas and the distal mycelium (Figs 4, S2). In the unfed seedling, no 14C was found in the above-ground parts, long roots or nonmycorrhizal root tips throughout the chase duration, that is, up to 14 d after 14C feeding.
The quantity of 14C accumulation in the fed seedlings decreased in above-ground parts and increased below ground. The above-ground 14C accumulation in the fed seedlings continued to decrease up to 3 d after 14C feeding (Fig. 5a-1). The below-ground 14C concentration in the fed seedling peaked 2 or 3 d after feeding and thereafter maintained a constant or slightly decreased concentration (Fig. 5a-3). In the unfed seedling, no above-ground 14C accumulation was detected during the experiment (Fig. 5a-2), but the below-ground section increased and reached a maximum 2 or 3 d after feeding and thereafter maintained a constant or slightly decreased 14C concentration (Fig. 5a-4), as observed in the fed seedlings. In ECM fungal combinations of PF–PF and PS–PS, the below-ground 14C in the unfed seedling area was 9.8 and 39.1% of that in the fed seedling, respectively.
In the PF–PS combination, 14C accumulated in the above-ground and below-ground parts of the fed seedling in the same pattern seen in PF–PF and PS–PS. However, 14C accumulation was restricted to the fed seedling, corresponding to the demarcation zone formed between the mycelia (Figs 6, S3).
In combination with the vacant foam plate, distribution of 14C was completely superimposed on the mycelium, including the hyphae extending into the vacant foam plate (Fig. 3). Quantitative changes in 14C distribution over time in the fed seedling area resembled those seen with the paired seedlings.
33P transfer between extraradical mycelia
In PF–PF and PS–PS, when a section of the mycelium was fed with 33P 12 d after the neighbouring mycelia contacted, 33P was transferred to the mycelium and mycorrhizas within 1 d in the fed seedlings, and was also detected in the mycelium of unfed seedlings within 2–3 d. 33P was first detected in the shoots of fed seedlings 7 d after feeding. The entire mycelium and all mycorrhizas of unfed seedlings contained 33P within 14 d after feeding (Figs 7, S4). 33P in both above-ground and below-ground areas of fed and unfed seedlings continued to increase 21 d after feeding (Fig. 5b).
In PF–PS, 33P was detected only in fed seedlings and never in unfed seedlings (Figs 8, S5). The pattern of 33P translocation in fed seedlings in PF–PS and PF–PS was the same as that in PF–PF and PS–PS, respectively; 33P radioactivity continued to increase 21 d after feeding (Fig. 5b).
The present study demonstrates that neighbouring mycelia of different conspecific genets show typical somatic incompatibility reactions, such as inhibition of hyphal invasion into the countering mycelium and formation of a demarcation zone between the mycelia. Also, no translocation of 14C photosynthate and 33P phosphoric acid was observed between the demarcation zone, which suggests that C and P translocation between the mycelia could occur only through hyphae but not via any extracellular pathway involving nutrient exudation from hyphae. The fact that the edge of the radioactivity distribution completely consisted with the mycelial front in the control rhizobox may support this suggestion.
Connections by hyphal cords developed between neighbouring mycelia of the same isolate within a few days after the contact. Accompanying the structural connection, 14C and 33P were translocated from the fed to the unfed mycelium. Because connections of hyphal cords were observed only near the contact line, these may develop soon after the extending hyphal cord encounters another cord in the counter mycelium. Although it is not known how the encountered hyphal cords were connected, there are two possibilities: by somatic fusion between hyphal cells in both cords or by physical attachment without somatic cell fusion. If both mycelia connect without cell fusion, photosynthate and phosphoric acid in hyphal cells should exude from cells. As discussed earlier, however, nutrients were not observed exuding from hyphae. Thus, cell fusion between hyphae of countering mycelia may occur at the hyphal cord connections and allow photosynthate and phosphoric acid to be freely translocated through the connections.
Some differences exist between translocation patterns of 14C-photosynthate and 33P-phosporic acid. Although 14C is spread homogeneously over the entire mycelium, 33P showed a strand-like distribution that connected the fed disk and ECM roots. If 33P is freely translocated through hyphal anastomosis, ECM roots are suggested to have a strong attraction for 33P.
Time-course quantification of radioactivity revealed another difference. 14C translocation in the mycelium ended 2–3 d after feeding. This suggests that only recently fixed photosynthate was translocated, as shown in our previous studies (Wu et al., 2001, 2002; Teramoto et al., 2012). By contrast, translocation of 33P in mycelia continued throughout the experimental duration of 21 d. Although 14C photosynthate is rapidly transformed into unmovable forms, 33P may remain in movable forms.
The quantification also showed that although below-ground translocation of 33P differed little between PS–PS and PF–PF, the below-ground 14C in the unfed seedling area in ECM fungal combinations of PS–PS was much more than that of PF–PF. This may indicate independence in the translocation of photosynthate and phosphorus, and reflect a difference in translocation between ECM fungal clones, PF and PS.
Although 33P was detected in the shoot and root of the unfed seedling, 14C was not. Several studies state that CMNs allow net transfer of C and nutrients between host plants (Brownlee et al., 1983; Finlay and Read, 1986a, b; Newman, 1988; Simard et al., 1997; Leake et al., 2004; Simard & Durall, 2004; Teste et al., 2010). However, our experiments demonstrated that 14C photosynthate was transferred to mycorrhizas of the unfed seedling, but not to the shoots. Teste et al. (2010) pointed out that shading of the unfed seedling in our previous study (Wu et al., 2001) would prevent photosynthetic activity and induce stomatal closure, reducing transpiration flow and 14C accumulation in the shoot. However, this study in which the unfed seedling was irradiated confirms the results of our previous work. Our experimental system is quite different from theirs, and conditions under which the seedlings were cultivated might have influenced 14C-transfer between neighbouring seedlings.
Although contact between the extraradical mycelia of different conspecific genets (PS–PF) in this study failed to connect the two mycelia, anastomosis by somatic hyphal fusion is known to occur between some genetically different fungi (Croll et al., 2009). Somatic compatibility between different but genetically similar genets, such as siblings, is also known in ECM fungi. Fries (1987) reported that the demarcation zone was generally less pronounced and in some cases almost invisible in combinations of sib-related Suillus luteus heterokaryons produced by mating in the laboratory. If different genets are compatible and connect to each other in nature, they could form a kind of guild for nutrient acquisition and translocation. The presence of such an ECM guild should be investigated in the future.
The CMN in an actual forest can be fragmented (according to soil turnover) by physical disturbances, climatic factors and host physiology (Cairney, 2005). Fragments of the original CMN may again become enlarged if the host tree provides enough photosynthate and the fragments contact each other. The hyphae in associated mycelial fragments would then fuse and form an integrated, larger mycelium. Photosynthate and phosphate could then be freely translocated within the larger mycelium and used more effectively because the nutrients could be translocated from more active hosts or portions of the mycelium in the CMN to sections of mycelia or host plants that require nutrients. This effective matching of demand and supply enhances symbiotic function. Because hyphal anastomosis is usually observed in the same fungal mycelium (Rayner et al., 1999; Sbrana et al., 2007), contact between mycelia of the same genet may frequently occur and result in hyphal connection in any fungal species. Such structural and functional connection between the same mycelium has been found in arbuscular mycorrhizal symbiosis (Mikkelsen et al., 2008). Connection between extraradical mycelia of ECM fungi may have a significant ecological impact on the forest in a similar way to the case of arbuscular mycorrhizal fungi.
In conclusion, the present study demonstrates that contact between mycelia of the same Pisolithus genets can result in an expanded extraradical mycelium network and enhanced nutrient translocation. Such contact can greatly enlarge the area and amount of nutrient exchange in the mycelial network, and also have an impact on the functions and nutrient cycling in the ECM symbiosis system.
This work was supported in part by Grants-in-Aid for Scientific Research (no. 21248018 and no. 23380080) from the Ministry of Education, Culture, Sports, Science and Technology, Japan.