IS THE SWITCH TO AN EC TOMYCORRHIZAL STATE AN EVOLUTIONARY KEY INNOVATION IN MUSHROOM- ForFORMING Peer FUNGI? Review A CASE STUDY Only IN THE TRICHOLOMATINEAE (AGARICALES) Journal: Evolution Manuscript ID 16-0471.R2 Manuscript Type: Original Article Date Submitted by the Author: 05-Oct-2016 Complete List of Authors: Sanchez-Garcia, Marisol; University of Tennessee, Department of Ecology and Evolutionary Biology Matheny, P.; University of Tennessee, Ecology and Evolutionary Biology Basidiomycota, speciation and extinction, Macroevolution, ecological Keywords: opportunity Page 1 of 45 1 IS THE SWITCH TO AN ECTOMYCORRHIZAL STATE AN EVOLUTIONARY 2 KEY INNOVATION IN MUSHROOM-FORMING FUNGI? A CASE STUDY IN THE 3 TRICHOLOMATINEAE (AGARICALES) 4 Running title: Diversification of ectomycorrhizal fungi 5 For Peer Review Only 6 Marisol Sánchez-García*1 and P. Brandon Matheny 7 Department of Ecology and Evolutionary Biology, 569 Dabney Hall, University of 8 Tennessee, Knoxville, TN 37996-1610, U.S.A. 9 *Email: [email protected] 10 1 Present address: Biology Department, Clark University, Worcester, MA 01610, U.S.A. 11 12 KEY WORDS: Basidiomycota, speciation and extinction, macroevolution, ecological 13 opportunity 14 15 Data will be deposited in Dryad Data Repository upon acceptance 16 17 ABSTRACT 18 Although fungi are one of the most diverse groups of organisms, little is known about the 19 processes that shape their high taxonomic diversity. This study focuses on evolution of 20 ectomycorrhizal (ECM) mushroom-forming fungi, symbiotic associates of many trees 21 and shrubs, in the suborder Tricholomatineae of the Agaricales. We used the BiSSE 22 model and BAMM to test the hypothesis that the ECM habit represents an evolutionary 23 key innovation that allowed the colonization of new niches followed by an increase in 1 Page 2 of 45 24 diversification rate. Ancestral state reconstruction supports the ancestor of the 25 Tricholomatineae as non-ECM. We detected two diversification rate increases in the 26 genus Tricholoma and the Rhodopolioid clade of the genus Entoloma . However, no 27 increases in diversification were detected in the four other ECM clades of 28 Tricholomatineae.For We Peersuggest that diversification Review of Tricholoma Only was not only due to the 29 evolution of the ECM lifestyle, but to the expansion and dominance of its main hosts and 30 ability to associate with a variety of hosts. Diversification in the Rhodopolioid clade 31 could be due to the unique combination of spore morphology and ECM habit. The spore 32 morphology may represent an exaptation that aided spore dispersal and colonization. This 33 is the first study to investigate rate shifts across a phylogeny that contains both non-ECM 34 and ECM lineages. 35 36 Introduction 37 Species diversity is unevenly distributed across the tree of life (Gould et al. 1987). One 38 explanation for this heterogeneous diversity pattern is clade age, in which older clades 39 have had more time to diversify and accumulate species than younger clades (McPeek 40 and Brown 2007; Wiens 2011; Bloom et al. 2014). An alternative explanation is that 41 species-rich clades have higher net diversification rates (speciation minus extinction) than 42 species-poor clades (Mooers and Heard 1997; Rabosky et al. 2007; Mullen et al. 2011; 43 Wiens 2011; Bloom et al. 2014). Diversification rates can be affected by ecological and 44 geographical opportunities such as opening of new environments, extinction of 45 competitors, or the evolution of key innovations that permit entry into new niche spaces 46 (Simpson 1953; Heard and Hauser 1995; Schluter 2000; Yoder et al. 2010). Estimating 2 Page 3 of 45 47 diversification rates can help to the study of patterns of species richness and to 48 understand the role of biotic and abiotic factors in shaping extant diversity (Wiens 2011). 49 50 Novel analytical methods have provided insights into the evolution of fungi, and some 51 patterns of Fordiversification Peer have begun Review to be explored. Wilson Only et al. (2011) investigated 52 the effects of gasteromycetation on rates of diversification and found that the net 53 diversification rate of gasteroid forms of some ectomycorrhizal Boletales is higher than 54 that of non-gasteroid forms, suggesting that if such rates remain constant, the gasteroid 55 forms will become dominant in the clades where they have arisen. A study of the genus 56 Trichoderma , a genus of ascomycetous molds, showed that host jumps are associated 57 with an increase in speciation rate, indicating that a transition from saprotrophy to 58 mycoparasitism may be acting as a driver of species diversification (Chaverri and 59 Samuels 2013). Kraichak et al. (2015) studied two of the largest families of lichenized 60 fungi and showed that while these two families have evolved similar ecological strategies, 61 they have reached such diversity through different evolutionary pathways. Recently, 62 Gaya et al. (2015) studied an order of lichenized fungi, the Teloschistales, and found that 63 the presence of anthraquinones, a pigment that protects against ultraviolet light, promoted 64 diversification of this group by providing the opportunity to colonize habitats with 65 increased sun exposure. Nagy et al. (2012) studied the family Psathyrellaceae to identify 66 morphological traits that could represent key innovations. Their results suggest that the 67 loss of a veil (a protective tissue layer found on the fruiting bodies of some mushroom- 68 forming fungi) and the gain of hairs, which represent a transition to a new defense 69 mechanism, might be correlated with an explosive radiation within this group. 3 Page 4 of 45 70 71 Symbiotic relationships are widespread and ecologically important, and they may be 72 considered evolutionarily advantageous, as they provide access to newly unexplored 73 ecological resources (Pirozynski and Malloch 1975; Margulis and Fester 1991; Leigh Jr 74 2010). EvolutionaryFor keyPeer innovations areReview defined as evolutionary Only changes that promote an 75 increase in net diversification rate (Heard and Hauser 1995). If these evolutionary 76 innovations evolve in multiple lineages, they can be recognized as being of adaptive 77 significance (Hunter 1998). The ECM symbiosis has evolved numerous times across the 78 fungal and land plant tree of life (Tedersoo and Smith 2013). This association increases 79 the efficiency in the acquisition of nitrogen and phosphorous in plants, while providing 80 carbon to the fungi (Smith and Read 2008), which can consequently allow both partners 81 to allocate more resources into growth and improve fitness, resulting in the exploitation 82 of new habitats (Pirozynski and Malloch 1975), and as a result increasing diversification 83 rates. 84 85 In this study, we investigate diversity patterns of ectomycorrhizal (ECM) fungi within the 86 suborder Tricholomatineae of the Agaricales, the largest order of mushroom-forming 87 fungi. Approximately 30 families of plants are involved in ECM symbiosis, which 88 include boreal, temperate, and tropical tree species in families that dominate terrestrial 89 landscapes such as Pinaceae, Fagaceae, Betulaceae, Salicaceae, Myrtaceae, and 90 Dipterocarpaceae (Malloch 1980; Wang and Qiu 2006; Brundrett 2009). At least 7,750 91 species of ECM fungi have been described, but it has been estimated that there are as 92 many as 25,000 ECM fungal species (Rinaldi et al. 2008; Brundrett 2009). Furthermore, 4 Page 5 of 45 93 the ECM mode has evolved from saprotrophic ancestors as many as 78–82 times with no 94 known reversals to their ancestral state (Tedersoo and Smith 2013). 95 96 Many ECM fungal clades have high species richness. The diversity of plants and fungi 97 involved inFor this association, Peer the potential Review ability to colonize newOnly niches, and the 98 ecological impact that it has on terrestrial ecosystems coupled with the multiple and 99 asynchronous origins of ECM fungi over time suggest that the ECM habit represents an 100 evolutionary key innovation that provides access to new ecological resources followed by 101 an increase in diversification rate (Pirozynski and Malloch 1975; Malloch 1980; 102 Brundrett 2009; Ryberg and Matheny 2012; Tedersoo and Smith 2013). However, this 103 hypothesis has not been explicitly tested. Here, we used two different methods to 104 estimate trait dependent and independent diversification rates in order to investigate 105 whether transitions in nutritional mode (non-ECM to ECM) are associated with an 106 increase in diversification rate. We focus our study on the suborder Tricholomatineae 107 (formerly the Tricholomatoid clade; Matheny et al. 2006), which contains roughly 30 108 genera including ECM and non-ECM groups (Sánchez-García et al. 2014). If the ECM 109 mode represents a key innovation, then we would expect to see an increase in 110 diversification rate in clades with this trait compared to those that are non-ECM. 111 112 Materials and Methods 113 DATA ASSEMBLY 114 A dataset of the Tricholomatineae was assembled using sequences from Sánchez-García 115 et al. (2014), and expanding gene sampling by incorporating newly produced rpb1 5 Page 6 of 45 116 sequences as well as sequences from GenBank (Table S1). To expand taxon sampling, 117 we searched for sequences of other members of the Tricholomatineae that were available 118 in GenBank. To avoid species redundancy, we only included taxa that were not 119 represented in our dataset and for which nLSU-rRNA sequences were available; then we 120 searched forFor more loci Peeravailable (5.8s, Review rpb1 , rpb2 , and nSSU-rRNA) Only and included them 121 only if we could confirm that such sequences were from the same collections (i.e . 122 comparing definition lines and modifiers). In addition, we obtained sequences from 123 collections of Tricholomatineae accessioned at the University of Tennessee Herbarium 124 (TENN), targeting taxa that were not already represented in our dataset. 125 126 DNA extraction, PCR, and sequencing protocols for ITS, nLSU, nSSU, and rpb2 are 127 described in Sánchez-García et al.