**3.2 The rise of fungal molecular ecological studies**

The last decade witnessed a substantial increase in studies focused on fungal community ecology. Conducting fungal surveys can be a tedious long-term undertaking and for a long time mycologists relied on fruit body occurrence or culturing of fungal isolates to document species occurrence and site-specific fungal diversity. Although such methods can provide important information, they tend to supply incomplete community descriptions for the reasons described in preceding sections.

The development of molecular tools to describe diversity allowed a much more straightforward, practical and rapid approach to the study of cryptic organisms such as fungi. These tools permit unveiling the communities colonizing soil (or other rich and dynamic substrates). Not only do they provide DNA-based information for identifying taxa, they also facilitate testing of ecological hypotheses, contributing for a better understanding of the structure and functioning of ecosystems. The vast majority of recent studies targeting the description of fungal communities are based on sequence data (Taylor, 2008).

In general, these molecular microbial studies target one specific short DNA region and rely on the identification of operational taxonomic units (OTUs): sequence similarity based surrogates for taxa (Sharpton et al., 2011). Although OTUs are difficult to define, they are the foundation for estimates of richness, frequency, abundance, and distributions. Most fungal environmental DNA-based diversity studies make use of the internal transcribed spacer

(ITS), a nuclear ribosomal repeat unit composed of three parts, the rapidly evolving ITS1, the very conserved 5.8S, and the moderately rapid ITS2 (Horton & Bruns, 2001, Bridge et al, 2005; fig. 4).

Fig. 4. Structure of the internal transcribed spacer (ITS), the nuclear ribosomal repetitive unit used to describe fungi to the species level. It is composed by the ITS1, 5.8S, and ITS2 regions, and flanked by SSU (ribosomal small subunit) and LSU (ribosomal large subunit).

ITS is used for identifying fungi at the species level. While it is far from being perfect, it offers several advantages that make it a popular that will likely be used for a long time. Genomes include numerous ribosomal DNA encoding genes distributed in tandem arrays along the same or different chromosomes (Rooney & Ward, 2005) and these copies are assumed to be extremely similar (Li, 1997). These coupled with the fact that ITS is easily amplified from low-quality samples (as opposed to single- or low-copy regions) makes it a fast and easy way to describe fungal diversity (Nilsson et al., 2008). However, there are several problems associated with using ITS to define fungal species. On the one hand, there are inherent biases associated with the use of DNA to document diversity, in particular problems with DNA extraction and amplification steps that might lead to distorted

Fungal Diversity – An Overview 219

ecology has been why are there so many fungal species in a given forest? What are they doing and how do they co-exist? Several explanations have been suggested, such as niche differentiation (Bruns, 1995). These could include vertical niche partitioning, where species have distinct microhabitat preferences that are distributed across a soil vertical gradient (Dickie et al., 2002), or temporal partitioning of ECM fungal communities, where species are active at different times of the year, promoting coexistence by reducing intraspecific competition (Koide et al., 2007). Although the majority of ECM fungal diversity studies are based on root tip data, fungal mycelia also live freely in soil and the community descriptions based on roots and mycelia provide different results (Koide et al., 2005), which adds another layer of complexity to the matter. Host-specificity, where different plant species associate with distinct assemblages of mycorrhizal fungi, has also been suggested as an explanation for the high ECM fungal diversity levels. In general, ECM fungi are known for not having high fidelity to their plant partners and tend to associate with a wide array of plant species. However, there is host preference, which seems to be an important factor in shaping local diversity (Dickie, 2007, Ishida et al., 2007). Inter-specific competition has been another topic of particular interest, given the high numbers of co-existing species. ECM fungi compete for access to the host, more specifically for carbon, as well as soil nutrients, and competition has recently been documented as a major player in ECM community structure (Kennedy, 2010).

Fig. 5. Typical ECM fungal species accumulation curve (top) and species rank-frequency plot (bottom). (Adapted from Branco & Ree, 2010). As more samples are described, new species are discovered at a consistently rate. This indicates that the vast majority of species

in the community are rare (see text for details).

community descriptions (Avis et al., 2009). On the other hand, it is known that there is within species variability in ITS, as the different copies within a genome are not exactly identical. Furthermore, intraspecific variation differs considerably across fungal groups (Karen et al., 1997, O'Donnell & Cigelnik, 1997, Glen et al., 2001, Horton, 2002, Rooney & Ward, 2005, Pawlowska & Taylor, 2004, Avis et al., 2006, Nilsson et al., 2008). These pose challenges in determining meaningful sequence similarity cut-offs (O'Brien et al., 2005). For the most part, OTUs are defined using a 95-97% similarity cut-off with the underlying assumption that resulting units are somewhat equivalent to fungal species. However, different fungal species have been reported to have ITS similarity as high as 99% (Dettman et al., 2001, Johannesson & Stenlid, 2003), while interspecific similarity of 90% or less has been found in other species (Kuniaga et al., 1997, O'Donnell, 2000). Despite these limitations and as mentioned above, ITS is the marker of choice for fungal diversity studies and is likely to remain so in the near future.
