Ectomycorrhizal lifestyle in fungi: global diversity, distribution, and evolution of phylogenetic lineages.

The ectomycorrhizal (EcM) symbiosis involves a large number of plant and fungal taxa worldwide. During studies on EcM diversity, numerous misidentifications, and contradictory reports on EcM status have been published. This review aims to: (1) critically assess the current knowledge of the fungi involved in the EcM by integrating data from axenic synthesis trials, anatomical, molecular, and isotope studies; (2) group these taxa into monophyletic lineages based on molecular sequence data and published phylogenies; (3) investigate the trophic status of sister taxa to EcM lineages; (4) highlight other potentially EcM taxa that lack both information on EcM status and DNA sequence data; (5) recover the main distribution patterns of the EcM fungal lineages in the world. Based on critically examining original reports, EcM lifestyle is proven in 162 fungal genera that are supplemented by two genera based on isotopic evidence and 52 genera based on phylogenetic data. Additionally, 33 genera are highlighted as potentially EcM based on habitat, although their EcM records and DNA sequence data are lacking. Molecular phylogenetic and identification studies suggest that EcM symbiosis has arisen independently and persisted at least 66 times in fungi, in the Basidiomycota, Ascomycota, and Zygomycota. The orders Pezizales, Agaricales, Helotiales, Boletales, and Cantharellales include the largest number of EcM fungal lineages. Regular updates of the EcM lineages and genera therein can be found at the UNITE homepage http://unite.ut.ee/EcM_lineages . The vast majority of EcM fungi evolved from humus and wood saprotrophic ancestors without any obvious reversals. Herbarium records from 11 major biogeographic regions revealed three main patterns in distribution of EcM lineages: (1) Austral; (2) Panglobal; (3) Holarctic (with or without some reports from the Austral or tropical realms). The holarctic regions host the largest number of EcM lineages; none are restricted to a tropical distribution with Dipterocarpaceae and Caesalpiniaceae hosts. We caution that EcM-dominated habitats and hosts in South America, Southeast Asia, Africa, and Australia remain undersampled relative to the north temperate regions. In conclusion, EcM fungi are phylogenetically highly diverse, and molecular surveys particularly in tropical and south temperate habitats are likely to supplement to the present figures. Due to great risk of contamination, future reports on EcM status of previously unstudied taxa should integrate molecular identification tools with axenic synthesis experiments, detailed morphological descriptions, and/or stable isotope investigations. We believe that the introduced lineage concept facilitates design of biogeographical studies and improves our understanding about phylogenetic structure of EcM fungal communities.