Author for correspondence:M. Giovannetti Tel: +39-050-9719324 Fax: +39-050-571562 Email: firstname.lastname@example.org
•The ability of arbuscular mycorrhizal (AM) networks originating from plants of different species, genera and families to become interconnected by means of hyphal anastomoses was assessed.
• An in vivo two-dimensional experimental model system was used to reveal the occurrence of linkages between contiguous mycorrhizal networks spreading from Allium porrum root systems and those originating from Daucus carota, Gossypium hirsutum, Lactuca sativa, Solanum melongena, colonized by Glomus mosseae.
•Percentages of hyphal contacts leading to anastomosis between extraradical networks originating from different plant species ranged from 44% in the pairing A. porrum–S. melongena to 49% in A. porrum–G. hirsutum. DAPI and Sytox stainings detected nuclei in the middle of fusion bridges connecting different mycorrhizal networks.
•Present data suggest that, by means of anastomoses, AM fungal mycelium would potentially create an indefinitely large network interconnecting different plants in a community, and that, in the absence of sexual recombination, the intermingling of nuclei in extraradical mycelium may provide endless opportunities for the exchange of genetic material.
Some studies provided qualitative information on the architecture and developmental dynamics of AM extraradical mycelium, by using root observation chambers and in vitro dual systems (Friese & Allen, 1991; Bago et al., 1998). Recent investigations utilized an in vivo two-dimensional experimental model system to visualize and quantify intact AM extraradical networks originating from individual plant root systems, and showed their extremely high interconnectedness due to the widespread occurrence of anastomoses (Giovannetti et al., 2001). Here, we applied the same experimental system to assess the ability of mycorrhizal networks originating from plants of different species, genera and families, to become interconnected. This system allowed us to reveal the occurrence and frequency of anastomosis between contiguous mycorrhizal networks spreading from Allium porrum (leek) root systems and those originating from Daucus carota (carrot), Gossypium hirsutum (cotton), Lactuca sativa (lettuce), Solanum melongena (aubergine), after inoculation with the AM symbiont Glomus mosseae. We also monitored anastomosis formation via a combination of video-enhanced light and epifluorescence microscopy, with the aim of detecting cytoplasmic continuity and nuclear migration within hyphal bridges.
Materials and Methods
The AM fungal species used was G. mosseae (Nicol. & Gerd.) Gerdemann & Trappe (Rothamsted isolate = local code IMA1), obtained from pot-cultures maintained in the collection of the Department of Chemistry and Agricultural Biotechnology, University of Pisa, Italy. Voucher specimens of this isolate are deposited in the Herbarium of the Department of Botanical Sciences, University of Pisa, Herbarium Horti Botanici Pisani (PI), as PI-HMZ4. Sporocarps of G. mosseae were extracted from pot-culture soil by wet-sieving and decanting, down to a mesh size of 100 µm, flushed into Petri dishes and manually collected with forceps under a Wild dissecting microscope (Leica, Milano, Italy). They were washed by vortexing in sterile distilled water (SDW) for 20 s, rinsed three times in SDW and germinated in the dark at 24°C between two 47-mm cellulose nitrate Millipore™ membranes (0.45-µm diameter pore) placed on moist sterile quartz grit in 14-cm diameter Petri dishes. A total of 15 sporocarps were used for each membrane sandwich.
Surface-sterilised seeds of A. porrum L., D. carota L., G. hirsutum L., L. sativa L. and S. melongena L. were germinated in moist sterile grit. After 15 d the root system of each seedling was sandwiched between the Millipore™ membranes containing the germinated sporocarps. Plants were individually placed into 10-cm diameter pots and the sandwiched root systems were buried in sterile quartz grit. Pots were closed in transparent Sun Bags (Sigma Aldrich s.r.l., Milan, Italy) and maintained in a growth chamber with 24°C day and 21°C night temperature and 16 h/8 h light/dark cycle.
After 42 d growth, the roots were gently removed from the sandwich by immersion in water and checked for the presence of extraradical hyphae, spores and sporocarps, which were carefully plucked with forceps under the Wild dissecting microscope. The occurrence of mycorrhizal infection was confirmed by autofluorescence of intraradical fungal structures in fresh whole roots mounted in water and observed under a Reichert-Jung (Vienna, Austria) Polyvar microscope equipped with epifluorescence optics (HBO 200 mercury lamp; Osram, Münich, Germany), by using the B1 filter combination (BP 450–495, LP 520, DS 510). The roots of either the same or different plant species were paired, approx. 3 cm apart, between Millipore™ membranes, and at least seven replicate sandwiches were prepared for each pairing. Plants were then transferred into 14-cm diameter Petri dishes containing sterile quartz grit and maintained in a growth chamber as described. At intervals, the sandwiches were carefully opened, and the development of symbiotic extraradical mycelium spreading from the roots on the membranes was checked, to detect harvest time, i.e. the day when the two mycorrhizal networks came into contact.
The following plant pairings were prepared: A. porrum–A. porrum, A. porrum–D. carota, A. porrum–G. hirsutum, A. porrum–L. sativa and A. porrum–S. melongena. At least two independent replicate experiments were performed for each pairing.
Occurrence and frequency of anastomoses within and between mycorrhizal networks
The root sandwiches were opened and the extraradical mycelium growing from the roots on the membranes was stained for the presence of succinate dehydrogenase activity (SDH) (Smith & Gianinazzi-Pearson, 1990). Deposition of formazan salts in hyphae allowed the assessment of hyphal viability and protoplasmic continuity established between fusing hyphae. After SDH test, the same mycelium was stained with Trypan blue in lactic acid (0.05%) to assess hyphal density under the Wild dissecting microscope. Hyphal density within mycorrhizal networks of each plant species (hyphal length, mm−2) was estimated with the gridline intersect method (Giovannetti & Mosse, 1980) by counting the length of hyphae in five areas of 64 mm2, using a grid eyepiece. Numbers of anastomoses were counted in five areas of 16 mm2. Areas of contact between paired mycelia were cut and mounted on microscope slides and examined under the Polyvar microscope. Hyphal anastomoses were counted at magnifications of ×125–500 and verified at a magnification of ×1250. Findings are based on at least 83 hyphal contacts for anastomoses within the same individual network and on at least 99 contacts for anastomoses between mycorrhizal networks originating from different plant species.
Frequency of anastomoses was calculated by determining the percentage of hyphal contacts leading to hyphal fusions and 95% confidence intervals of the obtained values are given. The chi-square test of independence was performed to detect significant differences in anastomosis frequency between hyphae within the same mycorrhizal network and between mycorrhizal networks originating from plants of different species. Z-test was performed on data of appropriate pairings. Data on hyphal and anastomosis density were also subjected to anova and differences in treatment means were evaluated by the Duncan multiple range test. Percentage data were subjected to arcsin transformation before analysis.
Some membranes were stained with CellTracker™ Green CMFDA (Molecular Probes Europe BV, Leiden, The Netherlands) to confirm the occurrence of protoplasm continuity between interconnecting networks.
Detection of nuclei in hyphal bridges connecting different mycorrhizal networks
To visualize the occurrence and location of nuclei in anastomosed hyphae, some membranes were stained with 5 µg ml−1 diamidinophenylindole (DAPI) in a 1 : 1 (v : v) water : glycerol solution and observed under epifluorescence with the Polyvar microscope using the filter combination U1 (BP 330–380, LP 418, DS 420). Additional membranes were stained with 50 nm solution of Sytox™ Green (Molecular Probes Europe BV) and observed under epifluorescence with the Polyvar microscope using the B1 filter combination.
To detect incompatibility reactions in hyphal contacts, the presence of wall thickenings and retraction septa was investigated on some membranes stained with DAPI, which were mounted in a 0.01% (w/v) solution of Calcofluor White (Sigma-Aldrich s.r.l., Milan, Italy) and observed under epifluorescence with the filter combination U1. The microscope was equipped with a 3 CCD colour video camera connected to a video-graphic printer (UP-890 CE; Sony-Italia, Milan, Italy).
The experimental system used here allowed us to detect linkages between intact extraradical hyphal networks originating from plants of different species, genera and families. Such linkages were established by means of anastomoses between contacting hyphae and were accomplished and visualized in vivo. The anastomosing mycelial networks evenly spreading around the whole root system were visible to the naked eye after SDH and Trypan blue staining (Fig. 1a–e). Successful anastomoses were characterized by complete fusion of hyphal walls and cytoplasm continuity, evidenced by formazan salt depositions (SDH activity) and CellTracker staining in hyphal bridges between fusing hyphae (Figs 1f and 2g–h). Fungal hyphae growing from A. porrum mycorrhizal roots came into contact with those of paired plants after 5–14 d incubation, depending on plant species. In this bidimensional experimental system, the density of hyphae within the same network ranged from 2.9 mm mm−2 in L. sativa to 6.8 mm mm−2 in G. hirsutum. Values obtained for G. hirsutum were significantly different from those of all the other plant species (Table 1; Fig. 1b).
Table 1. Hyphal and anastomosis densities and anastomosis frequency within individual and between paired extraradical networks produced by Glomus mosseae living in symbiosis with different plant species
Hyphal density within individual networks (mm mm−2)
Anastomosis density within individual networks (number mm−2)
Anastomosis frequency within individual networks (%)
Anastomosis frequency between networks (%)
Within the columns of hyphal and anastomosis densities, means ± SE followed by different letters are significantly different (P < 0.01, Duncan's test). Data on anastomosis frequency within mycorrhizal networks or between mycorrhizal networks originating from Allium porrum and either the same or different plant species, are followed by 95% confidence intervals.
3.49 ± 0.35ab
1.34 ± 0.12a
3.86 ± 0.18b
0.98 ± 0.07b
2.95 ± 0.11a
0.89 ± 0.08b
6.79 ± 0.37c
4.24 ± 0.08c
4.09 ± 0.19b
0.86 ± 0.11b
Viability of the mycorrhizal network formed by the different plant species was 100%, as determined by SDH activity.
Occurrence and frequency of anastomoses between mycorrhizal networks originating from plants of different species, genera and families
High percentages of anastomoses occurred between extraradical fungal networks originating from different plant species, ranging from 44% in the pairing A. porrum–S. melongena (107 fusions, 243 contacts) to 49% in A. porrum–G. hirsutum (67 fusions, 137 contacts) (Table 1), and were significantly different from those detected between mycorrhizal networks of the same species, A. porrum (62%; 90 fusions, 151 contacts) (Chi square = 13.2, P = 0.01).
Occurrence and frequency of anastomoses within mycorrhizal networks originating from plants of the same species
Fungal hyphae growing from mycorrhizal roots were highly branched and showed many anastomoses (Fig. 1b). The number of anastomoses mm−2 ranged from 0.86 in S. melongena to 4.24 in G. hirsutum (Table 1).
The frequency of anastomosis between hyphae of the same mycorrhizal network ranged from 46% in D. carota (51 fusions, 112 contacts) to 64% in L. sativa (74 fusions, 116 contacts) (Table 1). Statistical analysis gave a Chi square value of 10.7 (P = 0.03), showing a host plant effect. Significant differences between anastomosis rate were found for L. sativa and D. carota (Z-test 2.79, P = 0.006).
Nuclear migration through hyphal bridges connecting different mycorrhizal networks
The establishment of cytoplasmic continuity between fused hyphae, which represents the characteristic feature of true anastomosis, evidenced by formazan salt depositions (SDH activity) in the middle of hyphal bridges connecting different mycorrhizal networks, was consistently achieved in all the anastomoses observed (Figs 1f and 2h). Calcofluor staining failed to detect signs of hyphal incompatibility reactions such as septa development or hyphal lysis, either before or after hyphal fusions in interactions between different mycorrhizal networks (Fig. 2a–d). DAPI and Sytox stainings showed that the process of hyphal anastomosis involved the exchange of nuclei, which were detected in the middle of fusion bridges connecting networks originated from either the same or different plant species (Fig. 2b–f).
In this study we show a novel mechanism by which plants become interconnected, and evidence the wide occurrence of hyphal fusions between mycorrhizal networks originating from plant roots of different species, genera and families. To our knowledge, this is the first time that such linkages have been accomplished and visualized in vivo.
Previous work took for granted that connections between different plants are exclusively established by means of extraradical mycelium spreading from mycorrhizal plants into the soil and colonizing the roots of contiguous plant species (Graves et al., 1997; Read, 1998; Van der Heijden et al., 1998). Present data extend the knowledge of the mechanisms underlying plant interconnectedness, showing that, by means of anastomoses, mycorrhizal mycelium would potentially create an indefinitely large network of continuous fungal hyphae interconnecting different plants in a community. Given the wide host range of AM fungi, the emerging picture of mycorrhizal networks is one of previously unimagined dynamism and provides further support to the view that AM fungal symbionts play a fundamental role in the distribution of resources in plant communities (Chiariello et al., 1982; Francis & Read, 1984; Grime et al., 1987; Watkins et al., 1996; Graves et al., 1997; Fitter et al., 1998).
The experimental model system allowed us to perform spatio-temporal studies and monitor the spread of the symbiotic mycelium from host roots onto the underlying membranes and the process of formation of hyphal connections between contiguous plants. Although the system is bi-dimensional and may influence the rate of development and the structure of the mycorrhizal network, it represents a basic technique, which could be implemented for further studies aimed at detecting and quantifying nutrient and carbon transfer in the ‘web’ (Robinson & Fitter, 1999).
In this work hyphal densities and anastomosis frequency between hyphae of the same mycorrhizal network were consistent with previous findings obtained with A. porrum, Prunus cerasifera and Thymus vulgaris (Giovannetti et al., 2001).
Data on anastomosis frequency between mycorrhizal hyphae originating from plants of the same species, 62% in A. porrum, were significantly different from those found for hyphal connections between networks of different plant species, which ranged from 44 to 49%, in A. porrum–S. melongena and A. porrum–G. hirsutum, respectively.
Statistical analyses of hyphal density and anastomosis rate data showed a plant species effect, suggesting that mycelial growth and interconnectedness are not only dependent on fungal symbionts, but also on host plants. In fact, some studies reported that AM fungi exhibit host specificity or ecological specialization, leading to variations in fungal species fitness when establishing symbioses with different plant species (Sanders & Fitter, 1992; Eom & Hartnett, 2000; Bever, 2002; Bidartondo et al., 2002).
The detection of SDH activity and CellTracker fluorescence in hyphal bridges evidenced the establishment of cytoplasm flow between mycorrhizal networks connecting different host plants, fundamental for the maintenance of physiological and genetic continuity (Schubert et al., 1987; Hamel et al., 1990; Jones et al., 1998; Giovannetti et al., 2001). The absence of incompatible responses, such as protoplasm retraction, wall thickenings and septa formation between contacting hyphae and the occurrence of nuclei in anastomosing hyphae confirmed the viability of hyphal fusions and the established protoplasm continuity between hyphae of different mycorrhizal networks, in agreement with SDH results.
Previous findings showed that AM fungi, during the presymbiotic stage of their life cycle are able to exchange cytoplasm and nuclei, when originating from the same spore or from different spores of the same isolate (Giovannetti et al., 1999, 2003). Other studies confirmed the occurrence of nuclear exchange also in symbiotic extraradical hyphae (Giovannetti et al., 2001), which have been shown to be morphologically, physiologically and genetically very different from those originating from germinated spores (Harrison & van Buuren, 1995; Maldonado et al., 2001).
The detection of nuclei in hyphal bridges connecting extraradical mycelia of plants of different species, genera and families, provides further support to the view that anastomoses may represent a means for the maintenance of genetic diversity in the absence of sexual recombination (Bever & Morton, 1999; Kuhn et al., 2001). Although genetically different isolates of the same AM fungal species were not able to fuse (Giovannetti et al., 2003), we should consider the possibility that anastomosing ability may be still maintained in different mycelia carrying somatic mutations which do not involve vegetative compatibility genes. Since in this work plants were colonized by the same isolate, the major challenge remains concerning the correlation between genetic diversity of interacting mycelia and their ability to anastomose and exchange nuclei.
This work was supported by funds from the University of Pisa (Italy) and by C.N.R. (National Research Council, Italy).