The arbuscular mycorrhizal (AM) grass Calamagrostis epigejos and predominantly ectomycorrhizal (EcM) tree Salix caprea co-occur at post-mining sites spontaneously colonized by vegetation. During succession, AM herbaceous vegetation is replaced by predominantly EcM woody species. To better understand the interaction of AM and EcM plants during vegetation transition, we studied the reciprocal effects of these species' coexistence on their root-associated fungi (RAF). We collected root and soil samples from three different microenvironments: stand of C. epigejos, under S. caprea canopy, and contact zone where roots of the two species interacted. RAF communities and mycorrhizal colonization were determined in sampled roots, and the soil was tested for EcM and AM inoculation potentials. Although the microenvironment significantly affected composition of the RAF communities in both plant species, the effect was greater in the case of C. epigejos RAF communities than in that of S. caprea RAF communities. The presence of S. caprea also significantly decreased AM fungal abundance in soil as well as AM colonization and richness of AM fungi in C. epigejos roots. Changes observed in the abundance and community composition of AM fungi might constitute an important factor in transition from AM-dominated to EcM-dominated vegetation during succession.

Arbuscular mycorrhiza (AM) and ectomycorrhiza (EcM) are the most abundant and widespread types of mycorrhizal symbiosis, but there is little and sometimes conflicting information regarding the interaction between AM fungi (AMF) and EcM fungi (EcMF) in soils. Their competition for resources can be particularly relevant in successional ecosystems, which usually present a transition from AM-forming herbaceous vegetation to EcM-forming woody species. The aims of this study were to describe the interaction between mycorrhizal fungal communities associated with AM and EcM hosts naturally coexisting during primary succession on spoil banks and to evaluate how this interaction affects growth and mycorrhizal colonization of seedlings of both species. We conducted a greenhouse microcosm experiment with Betula pendula and Hieracium caespitosum as EcM and AM hosts, respectively. They were cultivated in three-compartment rhizoboxes. Two lateral compartments contained different combinations of both host plants as sources of fungal mycelia colonizing the middle compartment, where fungal biomass, diversity, and community composition as well as the growth of each host plant species’ seedlings were analyzed. The study’s main finding was an asymmetric outcome of the interaction between the two plant species: while H. caespitosum and associated AMF reduced the abundance of EcMF in soil, modified the composition of EcMF communities, and also tended to decrease growth and mycorrhizal colonization of B. pendula seedlings, the EcM host did not have such effects on AM plants and associated AMF. In the context of primary succession, these findings suggest that ruderal AM hosts could hinder the development of EcM tree seedlings, thus slowing the transition from AM-dominated to EcM-dominated vegetation in early successional stages.

Core Ericaceae produce delicate hair roots with inflated rhizodermal cells that host plethora of fungal symbionts. These poorly known mycobionts include various endophytes, parasites, saprobes, and the ericoid mycorrhizal (ErM) fungi (ErMF) that form the ErM symbiosis crucial for the fitness of their hosts. Using microscopy and high-throughput sequencing, we investigated their structural and molecular diversity in 14 different host × site combinations in Northern Bohemia (Central Europe) and Patagonia (South America). While we found typical ericoid mycorrhiza (=intracellular hyphal coils in the rhizodermis) in all combinations, we did not detect ectomycorrhiza (Hartig net) and arbuscular mycorrhiza (arbuscules). Superficial mantles of various thickness formed by non-clamped hyphae were observed in all combinations except Calluna vulgaris from N. Bohemia. Some samples contained frequent intercellular hyphae while others previously unreported intracellular haustoria-like structures linked with intracellular hyphal coils. The 711 detected fungal OTU were dominated by Ascomycota (563) and Basidiomycota (119), followed by four other phyla. Ascomycetes comprised Helotiales (255), Pleosporales (53), Chaetothyriales (42), and other 19 orders, while basidiomycetes Sebacinales (42), Agaricales (28), Auriculariales (7), and other 14 orders. While many dominant OTU from both Hemispheres lacked close relatives in reference databases, many were very similar to identical to unnamed sequences from around the world. On the other hand, several significant ericaceous mycobionts were absent in our dataset, incl. Cairneyella, Gamarada, Kurtia, Lachnum, and Leohumicola. Most of the detected OTU (623) could not be reliably linked to a particular trophic mode and only two (ca. 3% of all reads) could be reliably assigned to the archetypal ErMF Hyaloscypha hepaticicola. Probable ErMF comprised H. variabilis (4 OTU/0.8%) and Oidiodendron maius (1/0.5%), both detected only in N. Bohemia. Possible ErMF comprised sebacinoid fungi (42/14%) and several unnamed members of Hyaloscypha s. str. (14/0.5%). While H. hepaticicola was dominant only in C. vulgaris (1/32%), this model ErM host lacked O. maius and sebacinoid mycobionts. Hyaloscypha hepaticicola was absent in two and very rare (≤10 reads) in six combinations from Patagonia. Nine OTU represented dark septate endophytes from the Phialocephala fortinii s. lat. – Acephala applanata species complex, including the most abundant OTU (16%, the only detected in all combinations). Statistical analyses revealed marked differences between N. Bohemia and Patagonia, but also within Patagonia, due to the unique community detected in a Valdivian temperate rainforest. Ericaceous hair roots host a high diversity of mycobionts with mostly unknown functions and many novel ErMF lineages apparently await discovery. Transhemispheric differences (thousands of km) in composition of their communities can be evenly matched by local differences (scales of km, m, and less).