**2.** *Trichoderma* **taxonomy, a tool to assess diversity**

There remain many difficulties in the morphological identification of *Trichoderma* due to the homoplasy of morphological and phenetic characters, particularly among the *Trichoderma* anamorph forms (Chaverri & Samuels, 2003; Druzhinina *et al.,* 2006). For very many years since the genus was first described by Persoon in 1794, and connected to its sexual state by Tulasne and Tulasne in 1865, the genus continued without additions and was commonly assumed to comprise a single species, *T. viride.* This concept resulted in misleading species identifications which are still evident today. The type species, *T. viride sensu stricto,* is a relatively rare species more or less restricted in its distribution to regions in Europe and North America, and yet it is frequently cited as a native biological control agent in other regions (Jacklist *et al.*, 2006). Similarly, the widely reported *T*. *aureoviride*, for example, appears to have a limited distribution in northern Europe (Lieckfeldt *et al*., 2001).

Rifai (1969) made the initial approach to understand the diversity of *Trichoderma*, introducing the concept of "species aggregates" in *Trichoderma* and featuring Nine of them, but clarifying that these aggregate species could include multiple species not distinguishable by morphological characters. Later revisions of Bissett (1984, 1991 a, b, c, 1992) and Gams & Bissett (1998), increased the number of species based on morphological distinctions and made connections between anamorph and teleomorph states to include also some species previously placed in the genus *Gliocladium*. Studies on *Hypocrea* demonstrated the overlapping morphological characteristics among species in the anamorph genus *Trichoderma* (e.g. Chaverri and Samuels, 2003; Jaklitsch, 2009), definitively showing that morphological distinctions were not reliable indicators of the degree of genetic divergence between species, and that morphological observations alone were insufficient for accurate identification of species of *Trichoderma*.

To compensate for the lack of reliable morphological characters, research in *Trichoderma* biodiversity over the past 20 years has concentrated on the development of a variety of molecular markers to differentiate species, including isozymes, RAPDs, RFLP, AFLP and, most recently, the nucleotide sequences of various gene loci. The introduction of molecular tools resulted in greatly expanding the number of species recognized in *Trichoderma*. 104 species of *Trichoderma* are listed on the website of the International Commission on the Taxonomy of Fungi subcommision on *Trichoderma* and *Hypocrea* (www.isth.info, Druzhinina & Kopchinskiy 2008), and 193 named taxa are represented to date by sequences deposited in Genbank (www.ncbi.nlm.nih.gov/Taxonomy/Browser/wwwtax.cgi?id=29859).

### **2.1 Species concepts**

304 The Dynamical Processes of Biodiversity – Case Studies of Evolution and Spatial Distribution

Many species of *Trichoderma* are closely associated with plant roots and specific strains may form endophytic associations with their plant host (Bailey et al., 2006, Evans et al., 2003, Hoyos-Carvajal et al., 2009b; Manesh et al., 2006, Sette et al., 2006, Viterbo & Chet 2006, Yedidia et al., 2000). As endophytes they are particularly effective biological control agents of fungi in the rhizosphere, producing antimicrobials, activating plant defence mechanisms, and stimulating plant growth and vigour by solubilizing minerals and providing other nutrients and growth regulating compounds (Alfano et al., 2007; Altomare et al. 1999; Sharon et al., 2001; Vinale et al., 2006, Woo et al., 2006, Yedidia et al., 2000). The multiple roles of *Trichoderma* in biotrophic decomposition, parasitism and endophytic associations are of particular importance to the sustainability of agricultural and natural ecosystems (Harman et al. 2004). However, one of the great impediments to the study of *Trichoderma* has been the incorrect and confusing application of species names, making comparisons and generalizations from many published studies unreliable (Kopchinskiy et al., 2005). In addition, different isolates of *Trichoderma* species may exhibit as much variation in metabolic activity as observed among species, making careful study of their biodiversity essential to

There remain many difficulties in the morphological identification of *Trichoderma* due to the homoplasy of morphological and phenetic characters, particularly among the *Trichoderma* anamorph forms (Chaverri & Samuels, 2003; Druzhinina *et al.,* 2006). For very many years since the genus was first described by Persoon in 1794, and connected to its sexual state by Tulasne and Tulasne in 1865, the genus continued without additions and was commonly assumed to comprise a single species, *T. viride.* This concept resulted in misleading species identifications which are still evident today. The type species, *T. viride sensu stricto,* is a relatively rare species more or less restricted in its distribution to regions in Europe and North America, and yet it is frequently cited as a native biological control agent in other regions (Jacklist *et al.*, 2006). Similarly, the widely reported *T*. *aureoviride*, for example,

Rifai (1969) made the initial approach to understand the diversity of *Trichoderma*, introducing the concept of "species aggregates" in *Trichoderma* and featuring Nine of them, but clarifying that these aggregate species could include multiple species not distinguishable by morphological characters. Later revisions of Bissett (1984, 1991 a, b, c, 1992) and Gams & Bissett (1998), increased the number of species based on morphological distinctions and made connections between anamorph and teleomorph states to include also some species previously placed in the genus *Gliocladium*. Studies on *Hypocrea* demonstrated the overlapping morphological characteristics among species in the anamorph genus *Trichoderma* (e.g. Chaverri and Samuels, 2003; Jaklitsch, 2009), definitively showing that morphological distinctions were not reliable indicators of the degree of genetic divergence between species, and that morphological observations alone were insufficient for accurate

To compensate for the lack of reliable morphological characters, research in *Trichoderma* biodiversity over the past 20 years has concentrated on the development of a variety of molecular markers to differentiate species, including isozymes, RAPDs, RFLP, AFLP and, most recently, the nucleotide sequences of various gene loci. The introduction of molecular tools resulted in greatly expanding the number of species recognized in *Trichoderma*. 104

appears to have a limited distribution in northern Europe (Lieckfeldt *et al*., 2001).

fully exploit the potential of these fungi.

identification of species of *Trichoderma*.

**2.** *Trichoderma* **taxonomy, a tool to assess diversity** 

In the past, species in *Trichoderma* were defined primarily by the application of the concept Morphological Species Recognition (MSR), sometimes in combination with other phenetic characters. However, morphological identifications are highly prone to error due to the lack of definitive morphological characteristics and variations in culture. Consequently, perhaps 50% or more of the *Trichoderma* isolates deposited in culture collections may be incorrectly named based on morphological identifications. Furthermore, *Trichoderma* strains evidently cannot consistently be crossed to apply the Biological Species Recognition (BSR) concept based on their reproductive behavior. Therefore, Genealogical Concordance Phylogenetic Species Recognition (GCPSR) (Taylor et al., 2000), based on the concordance of multiple gene phylogenies, is an attractive alternative to apply the Phylogenetic Species Concept (PSC) in recognizing species of *Trichoderma*.
