Glomeraceae and Gigasporaceae differ in their ability to form hyphal networks


  • Liesbeth Voets,

    1. Université catholique de Louvain, Unité de Microbiologie;
    Search for more papers by this author
  • Ivan Enrique De La Providencia,

    1. Université catholique de Louvain, Unité de Microbiologie;
    2. Instituto Nacional de Ciencias Agricolas (INCA), Km 31/2 Carretera de Tapaste, Gaveta Postal 1, San José de Las Lajas, La Habana, Cuba;
    Search for more papers by this author
  • Stéphane Declerck

    Corresponding author
    1. Mycothèque de l’Université catholique de Louvain (part of the Belgian Coordinated Collections of Micro-organisms), Unité de Microbiologie, Croix du Sud 3, 1348 Louvain-la-Neuve, Belgium
      (*Author for correspondence: email
    Search for more papers by this author

(*Author for correspondence: email


Recent studies have revealed that the Glomeraceae and Gigasporaceae differ in their capacity to link hyphae in the same mycorrhizal network (de la Providencia et al., 2005), while interconnection of different networks of the same isolate was demonstrated only with an isolate of Glomus mosseae (Giovannetti et al., 2004). As these results may have major implications for our understanding of how these obligate symbionts explore their environment, the question asked here is whether species of the Gigasporaceae also have the ability to interconnect different networks of the same isolate.

Arbuscular mycorrhizal (AM) fungi can colonize a variety of plant species, and their lack of host specificity suggests that they can interconnect plants by means of a common mycelial network (Read, 1998; van der Heijden et al., 1998). These belowground networks have broad implications for resource allocation within the fungal colony, as well as for soil-derived nutrients and plant-derived carbon redistribution in ecosystems (Smith & Read, 1997). To date, it had been assumed that the principal route by which plants are interconnected was through extraradical mycelium spreading from mycorrhizal plants into the soil and colonizing the roots of plants with which they come into contact. Recently, Giovannetti et al. (2004) demonstrated that different mycorrhizal networks of the same isolate of G. mosseae may also become interconnected by anastomoses: ‘a mechanism by which different branches of the same or different hyphae fuse to constitute a mycelial network’ (Kirk et al., 2001). Different studies have illustrated that the capacity for anastomosis formation differs between species of the Glomeraceae and the Gigasporaceae (de Souza & Declerck, 2003; de la Providencia et al., 2005). In the Glomeraceae anastomoses were observed between different hyphae within the same mycelial network, while no anastomoses were detected within the same hypha (de la Providencia et al., 2005). This characteristic provides this AM fungal family with good plasticity to extend the extraradical mycelial network. In contrast, in the Gigasporaceae anastomoses were observed mainly within the same hypha (de Souza & Declerck, 2003; de la Providencia et al., 2005), generally related to a hyphal healing mechanism, while anastomoses between different hyphae of the same colony were far less abundant. This observation suggests a strategy based on the establishment of a large-diameter mycelium by means of the individual spreading of single hyphae forming the backbone of the colony (de Souza & Declerck, 2003). However, the question as to whether hyphae of Gigasporaceae species are able to interconnect different networks of the same species, in a similar way to G. mosseae (Giovannetti et al., 2004), remains unresolved and may have major implications for the capacity of AM fungi for rapid colony increase, exploration and exploitation of new environments and substrates, and creation of large networks.

Here we report on the capacity of AM fungal isolates belonging to the Gigasporaceae and Glomeraceae to form anastomoses within and between extraradical mycelial networks of the same isolate. For this study, the monoxenic culture system was used, providing a nondestructive, three-dimensional visualization of the development pattern of the extraradical mycelium.

Experimental set-up

Two organisms from the genus Glomus, one from Scutellospora and one from Gigaspora, were used for the experiment: isolate MUCL 43194 in the Glomus intraradices clade and isolate MUCL 41827 of Glomus proliferum; Scutellospora reticulata culture no. EMBRAPA CNPAB 1 and Gigaspora margarita BEG 34.

Monoxenic cultures of the four organisms were established in association with Ri T-DNA-transformed carrot (Daucus carota) roots on the modified Strullu–Romand (MSR) medium (Declerck et al., 1998 modified from Strullu & Romand, 1986), solidified with 3 g l−1 GelGro (ICN Biomedicals Inc., Irvine, CA, USA) in quadri-compartmental culture plates 12.4 × 8.5 cm (each compartment 3.1 × 8.5 cm) (quadriPERM, Greiner Bio-One, Kremsmünster Austria) (Fig. 1). However, only three compartments were used, left (LC); central (CC) and right (RC), separated by plastic walls. The three compartments were connected through an opening (1 cm wide), cut from bottom to top of the plastic walls, separating CC from LC and RC. The two openings were cut at opposite sides. MSR medium (35 ml) was added to the culture plates, filling the three compartments. A transformed carrot root approx. 8 cm long was placed in the LC and RC and inoculated with AM fungal spores of the same organism. Inoculation of the roots with the four species was achieved following the methods described by Cranenbrouck et al. (2005). Culture plates were incubated in an inverted position at 27°C in the dark. After profuse colonization of the root in the LC and RC, hyphae passed through the openings and developed in the CC. Roots that passed through the openings or over the partition walls, as well as hyphae crossing the partition walls, were trimmed.

Figure 1.

Schematic representation of the quadri-compartmental culture system (plate 12.4 × 8.5 cm; each compartment 3.1 × 8.5 cm). Transformed carrot roots associated to an arbuscular mycorrhizal (AM) fungus developed in the left (LC) and right (RC) compartments on the MSR medium. After colonization of the host root and production of an extensive extraradical mycelium, hyphae passed through an opening in the partition walls and grew in the central compartment (CC). The anastomosis formation was assessed in the CC.

Six replicates were considered for each species that consisted of a culture plate with two side compartments (LC and RC) each containing a transformed carrot root inoculated with the same AM fungal species, and a CC where the extraradical mycelia were allowed to develop and contact each other.

The culture plates used for the experiment were selected following three criteria: (1) the presence in the CC of the two mycelia intersecting each other; (2) approximately equal development of the two mycelial networks developing in the CC; and (3) total hyphal density within the CC of approx. 25 cm cm−3. Optimal culture plates (matching these three criteria) were obtained within a period of 4–8 wk according to the AM fungal species.

Different parameters were recorded in the CC: total hyphal length, hyphal density, and number and type of anastomoses. Total hyphal length and hyphal density were measured following the method detailed by Declerck et al. (1998). Three types of anastomosis were considered: (1) within the same hypha (hyphal bridge); (2) between different hyphae within the same mycelial network, originating from the same colony (from compartment LC or RC); and (3) between hyphae of the two intersecting mycelial networks developing from the LC and RC. To determine the presence and type of anastomosis, culture plates were observed under a stereomicroscope and a bright-field light microscope at ×10 to ×40, and ×50, ×125 or ×250 magnification, respectively.

Gigasporaceae do not interconnect different mycelia belonging to the same isolate

Our data demonstrate for the first time that Glomeraceae and Gigasporaceae differ in their ability to interconnect different mycelial networks belonging to the same isolate. While hyphal fusions (anastomoses) were observed between hyphae from different mycelia of the same isolate with G. proliferum and G. intraradices, this mechanism was never observed with S. reticulata and G. margarita. The results on the abundance and type of anastomoses formed within the same mycelial network of both families further confirmed earlier work by de la Providencia et al. (2005).

In cultures belonging to the Glomeraceae, 12.5–20.1% of anastomoses detected in the CC were formed between hyphae from the two different mycelial networks (Table 1; Fig. 2a–c). Anastomoses were always characterized by complete fusion of hyphal walls allowing bidirectional cytoplasmic/protoplasmic flow via the fusion bridge. These results corroborate the observations of Giovannetti et al. (2004), who observed interconnections between mycorrhizal networks of the same isolate of G. mosseae spreading from different host plants. In contrast, the Gigasporaceae species studied (S. reticulata and G. margarita) never formed anastomoses between hyphae of different mycelial networks of the same isolate (Table 1). This fundamental observation complements the earlier results obtained by Giovannetti et al. (1999) with hyphae growing from germinated spores. These authors observed that Gigaspora rosea and Scutellospora castanea were unable to form anastomoses between hyphae of germinated spores belonging to the same isolate, while such a mechanism was observed in species belonging to the Glomeraceae, G. intraradices, Glomus caledonium and G. mosseae.

Table 1.  Growth parameters and percentage anastomosis formation of different nonperturbed networks of Glomus proliferum, Glomus intraradices, Scutellospora reticulata and Gigaspora margarita
SpeciesGrowth parameters*Percentage anastomoses
Hyphal length (cm)Total no. anastomoses per hyphal length (cm)In the same hyphaIn the same networkBetween networks
  • Values represent means of six replicates (± SE for growth parameters).

  • *

    Values of growth parameters in the same column followed by a different letter differ significantly at P < 0.05 (Tukey's HSD).

  • Values of percentage anastomoses in the same column followed by a different letter differ significantly at P < 0.05 (χ2 test for comparison of observed and expected values in contingency tables).

G. proliferum319.6 ± 26.8 a0.172 ± 0.024 a 0.3 a79.6 a20.1 a
G. intraradices390.3 ± 38.6 a0.086 ± 0.013 b 0 a87.5 a12.5 a
S. reticulata374.3 ± 49.6 a0.020 ± 0.001 c94.7 b 5.3 b 0 b
G. margarita415.5 ± 27.9 a0.009 ± 0.002 c95.5 b 4.5 b 0 b
Figure 2.

Anastomosis formation between two hyphae of different arbuscular mycorrhizal (AM) fungal networks of the same isolate of Glomus proliferum. (a) Two anastomoses were observed (square boxes). Bar, 50 µm (b,c) Detailed view of the anastomoses, formed by a tip-to-tip connection (arrows). Bars, 20 µm.

Regarding anastomosis formation within the same mycelial network, the two Glomus species mostly formed anastomoses between different hyphae. Anastomosis within the same hypha was noted only once, with G. proliferum (Table 1). Such an event was not reported by de la Providencia et al. (2005), and seldom reported in old colonies of G. intraradices following protrusion of cytoplasmic/protoplasmic material in damaged hyphae and subsequent hyphal bridging (Bago & Cano, 2005), demonstrating its rarity in Glomus species. The number of anastomosis within the same mycelial network was comparable with results obtained by de la Providencia et al. (2005) with the same accessions, but was markedly lower compared with the results of Giovannetti et al. (2004). The lower number of anastomoses found here could be related to the different AM fungal species or host plant, but more probably to the experimental system. A two-dimensional system was used; as Mosse (1959) states: ‘on cellophane, anastomoses are extremely common, possibly because all hyphae are in the same horizontal plane and must therefore meet frequently’.

The Gigasporaceae species studied developed numerous anastomoses within the same hypha, while few anastomoses were detected between different hyphae of the same mycelial network. The proportion between both types of anastomosis (Table 1) was in agreement with the results of de la Providencia et al. (2005). Both types of anastomosis were habitually a consequence of cytoplasmic flow obstruction following damage. We hypothesized that damage was presumably associated with the high pressure of the actively streaming cytoplasm/protoplasm in the hyphae. This pressure could cause microfractures in the hyphal wall, resulting in an efflux of cytoplasmic/protosplasmic material. This was corroborated by the frequent observation of cytoplasm/protoplasm protrusion from these hyphal sections, and supported by earlier observations (de la Providencia et al., 2005).


The dissimilar behaviour of the mycelial networks of the Glomeraceae and Gigasporaceae species under study suggested divergent strategies for the fungal colonies to explore and to exploit new environments and substrates and to create large networks. In Glomus species, anastomoses were observed both within and between mycelial networks of the same isolate. This behaviour has physiological as well as genetic implications, enhancing sink-regulated redistribution of resources within a single fungal colony (Rayner, 1991) and between fungal colonies of the same isolate. It may also participate in the maintenance of genetic diversity in the absence of sexual recombination by migration of nuclei via the fusion bridge between mycelia (Giovannetti et al., 2004). In contrast, development of the Scutellospora and Gigaspora species studied is oriented towards the individual spreading of thick hyphae, forming the backbone of the colony (de Souza & Declerck, 2003). These hyphae were not observed to anastomose with hyphae from other mycelial networks belonging to the same isolate; seldom anastomosed with hyphae from the same network; but mainly formed intrahyphal anastomoses (hyphal bridges). This mechanism suggests the propensity of these hyphae to colonize new substrates and new hosts by themselves, without the support of adjacent hyphae.


This work was supported by a grant of the ‘Fonds Spéciaux de Recherche’ (FSR) of the Université catholique de Louvain. I.E.P. thanks Dr José R. Martin Triana, Director of INCA, for supporting AM fungal research. S.D. gratefully acknowledges financial support from the Belgian Federal Office for Scientific, Technical and Cultural affairs (OSTC, contract BCCM C3/10/003). Thanks are also addressed to Professor E. Le Boulengé and Dr E. Kestens for statistical advice.