f u n g a l b i o l o g y 1 1 7 ( 2 0 1 3 ) 1 4 5 e1 5 5 journal homepage: www.elsevier.com/locate/funbio A molecular contribution to the assessment of the Tricholoma equestre species complex b,c
, Erika BEROARDc, Jacques GUINBERTEAUa, Serge MOUKHAa,b,c, Cyril FERANDON Beno^ıt CASTANDETd, Philippe CALLACa, Edmond CREPPYb,c, G
erard BARROSOa,c,* a INRA (Institut National de la Recherche Agronomique) Bordeaux-Aquitaine, UR1264 MycSA (Mycologie et S
ecurit
e des Aliments), BP81, 33883 Villenave d’Ornon Cedex, France b Laboratoire de Toxicologie et Hygiene Appliqu
ee, UFR des Sciences Pharmaceutiques, France c Universit
e Bordeaux Segalen, 146, rue L
eo Saignat, 33076 Bordeaux Cedex, France d Boyce Thompson Institute for Plant Research, Tower Rd, Ithaca, NY, USA article info abstract Article history: In recent years, interest in the Tricholoma equestre species complex has increased because of Received 24 February 2012 several cases of severe and sometimes fatal rhabdomyolysis reported in France and Poland. Received in revised form These occurred after repeated consumption of large portions of T. equestre sporophores 31 July 2012 during consecutive meals, despite the fact that this species is renowned as a tasty edible Accepted 4 January 2013 wild mushroom. The T. equestre species complex includes three ectomycorrhizal species Available online 12 January 2013 Tricholoma flavovirens (Pers.) S. Lundell, Tricholoma auratum (Paulet) Gillet, and T. equestre Corresponding Editor: (L.) P. Kummer. All these species produce sporophores with intense yellow gills but are dif- Kentaro Hosaka ficult to distinguish by morphological analyses at both the macroscopic and microscopic levels. In T. equestre, two additional varieties are recognized: T. equestre var. populinum Keywords: (Christensen & Noordeloos) associated with Populus sp. and/or Betula sp. trees and some- Nuclear ribosomal RNA unit times recognized as Tricholoma frondosae (Kalamees & Shchukin) and T. equestre var. pallid- Species complex ifolia characterized by pale to white gills, frequently recognized as Tricholoma joachimii (Bon Tricholoma auratum & Riva). To explore the taxonomic (species delimitation), ecological, and geographical ex- Tricholoma equestre tent and limits of the T. equestre species complex, we have carried out a molecular compar- Tricholoma flavovirens ison of worldwide strains belonging to this complex by using sequences of two molecular markers: the internal transcript spacer (ITS)1/5.8S/ITS2 region of the nuclear ribosomal unit and the 50 part of the mitochondrial cox1 gene. Phylogenetic analyses support the placement of European T. equestre, T. flavovirens, and T. auratum strains as representatives of a single species. This species appears associated with various conifers trees, depending on the geographic origin (Pinus pinaster for T. auratum, Pinus sylvestris or Abies alba for T. equestre and T. flavovirens). However, in the context of a single T. equestre species, the geographical location could lead to the characterization of sub-species or varieties, as suggested by the gathering of the four Asian (Japanese) T. auratum strains in a strongly supported distinct phylogenetic clade. Moreover, our analysis strongly argues for considering T. joachimii and the synonymised T. equestre var. pallidifolia as two representatives of a different species not belonging to the T. equestre group. This species would be phylogenetically related to the Tricholoma columbetta species with which they share white gills. Similarly, the phylogenetic analysis of the molecular data and the lack of gene flow between * Corresponding author. Tel.: þ33 5 57 12 25 95. E-mail address: gerard.barroso@u-bordeaux2.fr (G. Barroso). 1878-6146/$ e see front matter ª 2013 The British Mycological Society. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.funbio.2013.01.003 146 S. Moukha et al. the strains associated with broad-leaved trees and those of the T. equestre complex, rather argues for two distinct species depending on the ecological niche: T. frondosae under broadleaved trees and T. equestre under conifers. ª 2013 The British Mycological Society. Published by Elsevier Ltd. All rights reserved. Introduction Since the medieval time, Tricholoma equestre has been appreciated as a tasty edible wild mushroom and, until now, sporophores have been harvested in large quantities from forests
dry & Gromb 2009). In recent years, a new type of mush(Be rooms poisoning involving T. equestre was described in France and Poland. Several cases of severe and sometimes fatal rhabdomyolysis were reported after repeated consumption of large portions of T. equestre and/or Tricholoma flavovirens during several consecutive meals (Bedry et al. 2001; Chodorowski et al. 2002). In Japan, such symptoms of rhabdomyolysis were associated with the consumption of Russula subnigricans, a phylogenetically distant mushroom belonging to a different order, Russulales. In this species, the rhabdomyolitic molecule was recently identified as the cycloprop-2-ene carboxylic acid (Matsuura et al. 2009). The association of rhabdomyolysis with T. equestre and/or T. flavovirens consumption was confirmed in mice (Bedry et al. 2001) by demonstrating the increase of the plasmatic creatine kinase (CK), Alanine aminotransferase (ALAT), and Aspartate aminotransferase (ASAT) activities. In this animal model, consumption of T. equestre and/or T. flavovirens is also responsible for chronic intoxication triggering a muscle disease (Bedry et al. 2001; Chodorowski et al. 2002; Nieminen et al. 2005). Specifically, a repeated consumption of 9 g (dry weight/kg body mass) T. flavovirens sporophores for 5 d or prolonged consumption of 12 g freshly frozen mushroom/kg body mass trigger myo-, cardio-, and hepatotoxic effects in mice (Nieminen et al. 2005; Nieminen et al. 2008). It is reasonable to posit, however, that the human poisoning and death cases occurring in southwest France and attributed to T. equestre synonymised with T. flavovirens in Bedry et al. (2001), would be in fact due to an excessive consumption of Tricholoma auratum which is the most abundant Tricholoma species in the pine forests (Pinus pinaster) of sea-side sand dunes of southwest France where toxic sporophores were col
dry & Gromb 2009). lected (Be In 2003, the French Agency for Food Safety (Afssa, today Anses) recommended a revision of the current classification of T. equestre complex species, to ban T. equestre, T. flavovirens, and T. auratum from consumption (Afssa-Saisine N 2002-SA0285) and to refrain from harvesting all morphologically closely related species. In 2004, the import and sale on the market of T. equestre (T. auratum, T. equestre and T. flavovirens) ^ te
du 16 juin 2004). Since was stopped for a period of 1 y (Arre 2005, their import and sale is permanently prohibited
cret n 2005-1184). (De From a taxonomic point of view, the T. equestre entity appears as a species complex that includes T. equestre (L.) P. Kummer and at least two additional species, namely T. flavovirens (Pers.) S. Lundell and T. auratum (Paulet) Gillet. In fact, the sporophores of these three Tricholoma species are difficult to discriminate even by experienced mycologists, because they share similar morphologies mainly characterized by yellow tints and gills of intense yellow (Fig 1). Tricholoma flavovirens and T. equestre have recently been grouped as a single species named T. equestre (Deng & Yao 2005). This species shows a wide distribution extending from Europe to North America. Furthermore, T. auratum (also called ‘Bidaou’ in Southwest France) specifically grows under conifers (P. pinaster). Tricholoma auratum is particularly abundant in young pine forests of the sea-side sand dunes of the Atlantic coast of South Europe (Iberian Peninsula and France), where it behaves as a typically early stage mycorrhizal species. Additionally, T. auratum has been recently reported in Japan (Kikuchi et al. 2007). In this work, we use two molecular markers to investigate the relationship between the three Tricholoma species characterized by intense yellow gills from various locations. Occasionally sporophores phenotypically related to T. equestre were found associated with Betula sp. and/or Populus sp. trees; these species are considered as a variety of T. equestre: T. equestre var. populinum (Christensen & Noordeloos) or as a different species named Tricholoma frondosae (Kalamees & Shchukin). Two collection vouchers corresponding to this species or variety, collected under Betula and/or Populus sp. trees, were added to our analysis of the species complex (Table 1 and Fig 1). In the same way we added to the analysis of two strains collected in France, phenotypically close to T. equestre complex, but bearing pale to white gills (Table 1 and Fig 1). This would allow the assignment of these strains as a variety of T. equestre var. pallidifolia or, alternatively, as a different species known as Tricholoma joachimii (Bon & Riva). The first molecular marker used to identify and/or to discriminate these Tricholoma species was the conventional nuclear marker, internal transcript spacer (ITS)1/5.8 rDNA/ITS2 region of the ribosomal unit (Gardes & Bruns 1993) recognized as the most efficient tool for fungal DNA barcoding. An additional molecular marker corresponding to a mitochondrial sequence, was added: approximately 600 bp of the 50 region of the cox1 gene, which encodes subunit 1 of the cytochrome c oxidase. This marker was recently successfully used in a fungal metagenomic study (Damon et al. 2010). Materials and methods Tricholoma species sampling and morphological determination Sporophores representative of French strains of Tricholoma species were collected in various geographical locations, i. e. A molecular contribution to the assessment of the T. equestre species complex 147 Fig 1 e Morphology of sporophores from French representatives of T. auratum AuFr3 (A), T. equestre (B), T. equestre var. pallidifolia EqFrW (C) T. equestre var. populinum EqFrPop (D) strains. Photographies were from Jacques Guinberteau (A) and Laurent Deparis (B, C, and D). near the Atlantic coast (sea-side sand dunes) of the Southwest France for Tricholoma auratum and continental central regions for Tricholoma equestre and related Tricholoma species and varieties (Table 1). The strains were collected and identified at both macroscopic and microscopic levels by mycologists listed in Table 1. Sequences obtained from the collected strains were compared to sequences available in the GenBank with the help of the BLAST algorithm (Altschul et al. 1990). The analyses were extended to species belonging to the T. equestre complex with other geographical origins (Table 1), especially sequences from T. equestre, T. auratum, and Tricholoma flavovirens strains from various countries and continents (North America, Japan, and European countries). Close species such as Tricholoma portentosum (same section equestre) or Tricholoma columbetta (with white gills and belonging to the albata section) from North America (USA or Canada) or Europe (Portugal) were included in the analysis. Two Tricholoma pardinum sequences from Canada and two Tricholoma matsutake sequences from Asia (China and Korea) were added because they represent largely studied and phylogenetically distant Tricholoma species. rDNA ITS1/5.8/ITS2 region (White et al. 1990), while CoxU1/ CoxR (50 - TCTACTAATGCTAAAGATATTGG -30 /50 - CACCGGCT AATACAGGTAA -30 ) was used for amplification of the 50 region of the mitochondrial cox1 gene (Damon et al. 2010). Reaction mixes contained 10e100 ng of fungal DNA, 1 mM of both primers, 200 mM of each dNTP, 1 unit of Go Taq DNA polymerase (Promega Corp., Madison, WI, USA), in a final volume of 50 ml enzyme buffer. The conditions for thermocycling were as follow: 30 cycles of denaturation at 95  C for 30 s; hybridization at two degrees below the lowest Tm of both oligonucleotides for 30 s and elongation at 72  C for 1 min. A final 5 min incubation at 72  C was performed. The PCR products were ligated to the pGEM-T Easy Vector from Promega Corp. (Madison, WI, USA) and used to transform Escherichia. coli XL1-blue (Bullock et al. 1987) according to the Hanahan (1985) procedure. Recombinant plasmid DNAs were purified by using the Wizard Plus SV Minipreps DNA Purification System as recommended by the supplier (Promega Corp.). DNA sequencing DNA extraction, PCR amplification, and molecular cloning Total DNA was extracted from 5 g of frozen carpophores or 0.2 g of dried carpophores after grinding in liquid nitrogen. Nucleic acids were extracted according to the N-cethyl-NNNtrimethyl ammonium bromide (CTAB) procedure adapted to small quantities of basidiomycete mycelia as described by Barroso et al. (1995). PCR amplifications were carried out using primer pairs synthesized by Eurofins MWG Operon Germany. The pair ITS4/ITS5 (50 - GCATATCAATAAGCGGAGGA -30 /50 - GGAAGTAA AAGTCGTAACAAGG -30 ) was used for amplification of the Plasmid inserts were sequenced, according to the supplier recommendations, using the Big Dye Terminator v1.1 Cycle Sequencing kit (Applied Biosystems, Courtaboeuf, France) with the conventional SP6 or T7 primers that hybridize at both sides of the multicloning site. DNA sequencing was performed at the Genotyping and Sequencing Facility of University Bor
gional d’Aquitaine deaux Segalen (grants from the Conseil Re n 20030304002FA and 20040305003FA, and from the European Union, FEDER n 2003227). Sequences profiles were edited using BioEdit sequence alignment editor v7.0.9 (Ibis Biosciences Carlsbad, CA, USA). 148 S. Moukha et al. Table 1 e Origin and characteristics of the strains and sequences used in this study. Species Voucher name Geographical origin Phenotype, habitat Lacanau, Atlantic coast (33) South France Mont-de-Marsan (40) South France Tricholoma auratum AuFr1 Tricholoma auratum AuFr2 Tricholoma auratum AuFr3 Carcans, Atlantic coast (33) South France Tricholoma auratum AuJa1 Japan Yellow gills, sand dunes under Pinus pinaster Yellow gills, under Pinus pinaster Yellow gills, sand dunes under Pinus pinaster Conifers Tricholoma auratum AuJa2 Japan Conifers Tricholoma auratum AuJa3 Japan Conifers Tricholoma auratum AuJa4 Japan Conifers Tricholoma equestre EqFr1 Haute-loire (43) France Tricholoma equestre EqFr2 Aube (10) France Tricholoma equestre EqFr3 Puy-de-Dome (63) France Tricholoma equestre Tricholoma equestre Tricholoma equestre var. pallidifolia EqNo EqPo (AP58) EqFrW Norway Portugal Haute-Savoie (74) France Tricholoma joachimii JoFr Haute-Savoie (74) France Tricholoma equestre var. populinum EqFrPop Haut-Rhin (68) France Tricholoma frondosae FrFr France Tricholoma flavovirens Tricholoma flavovirens FlJa FlCar Japan Canada Tricholoma flavovirens Tricholoma flavovirens Tricholoma flavovirens FlUs1 (trh545) FlUs2 (trh546) FlUs3 USA USA USA Yellow gills, hill forest under Pinus sylvestris Yellow gills, hill forest under Abies alba and Quercus pubescens Yellow gills, hill forest, under conifers nd nd Yellow pale to white gills, hill forest under Quercus pubescens White gills, hill forest under Quercus pubescens Yellow gills, hill forest under Betula pendula and Populus sp. Yellow gills, hill forest under Populus tremula Conifers Conifers, boreal forest Conifers Conifers Conifers Tricholoma flavovirens FlUs4 USA Conifers Tricholoma flavovirens Tricholoma flavovirens Tricholoma flavovirens Tricholoma flavovirens Tricholoma columbetta FlPo1 (AP25) FlPo2 (AP21) FlPo3 (AP40) FlPo4 (AP33) T. columbetta US Portugal Portugal Portugal Portugal USA Conifers Conifers Conifers Conifers Broad-leaved trees Author or reference Collection date (identifiers and collectors)a ITS4/ITS5 sequence (GenBank Acc. N ) This work 11/2009 (GB) HM590867 This work 2002 (SM) HM590868 This work 04/12/2008 (GB, SM, JG) HM590869 Kikuchi et al. 2007 Kikuchi et al. 2007 Kikuchi et al. 2007 Kikuchi et al. 2007 This work AB289659 AB289660 AB289663 AB289662 14/11/2002 (SM) HM590870 This work 20/10/2007 (JV, PB) HM590871 This work 2002 (SM) HM590872 Hoiland, K. Portugal, A. This work 10/10/2008 (LD) AJ236081 EU186278 HM590874 This work 05/10/2008 (LD) HM590876 This work 31/10/99 (LD) HM590875 This work 28/09/2010 (OR) JF896232 Murata, H. Kranabetter et al. 2009 Horton, T. R. Horton, T. R. Bidartondo & Bruns 2002 Bidartondo & Bruns 2001 Portugal, A. Portugal, A. Portugal, A. Portugal, A. Bidartondo & Bruns 2002 AB036895 HQ650740 AF458449 AF458452 AF377181 AF349689 EU186297 EU186294 EU186310 EU186304 AF349693 A molecular contribution to the assessment of the T. equestre species complex 149 Table 1 e (continued ) Species Voucher name Geographical origin Phenotype, habitat Tricholoma columbetta T. columbetta Po Portugal Tricholoma portentosum T. portentosum Ca1 Canada Broad-leaved trees Conifers Tricholoma portentosum Tricholoma pardinum T. portentosum Ca2 T. pardinum Us Canada USA Conifers Conifers Tricholoma pardinum Tricholoma matsutake Tricholoma matsutake T. pardinum Ca T. matsutake Ch T. matsutake Ko Canada China Korea Conifers Conifers Conifers Author or reference Collection date (identifiers and collectors)a ITS4/ITS5 sequence (GenBank Acc. N ) Portugal, A. EU186277 Kranabetter et al. 2009 Denis, M.W. Bidartondo & Bruns 2002 Guichon, S.H.A. Wang, Y. Matsushita et al. 2005 HQ650742 EU486444 AF377228 JF899575 EU552801 AB188533
rez-De-Gregorio, OR Olivier Roblot, GB Ge
rard Barroso, SM Serge Moukha, JG Jacques Guina Mycologists: LD Laurent Deparis, MP Miquel A. Pe
a. berteau, PB Philippe Bineau, JV Jean Rove Sequence analyses Comparisons with sequences of the GenBank and EMBL databases were performed with the search algorithm BLAST (Altschul et al. 1990). Multi-alignments of nucleic acids were performed with ClustalW algorithm (Thompson et al. 1994). For phylogenetic analyses, alignments were manually checked for accuracy and submitted to the PhyML program (Guindon & Gascuel 2003) followed by TreeDyn for tree drawing (Chevenet et al. 2006). These programs are available on line at: http://www.phylogeny.fr/ and described by Dereeper et al. (2008, 2010). The GTR (DNA/RNA) substitution model was selected: the number of substitution rate categories was four, the gamma distribution parameter and the proportion of invariable sites were not fixed but estimated during the analysis. For trees of Fig 3, branch supports were statistically estimated by the Approximate likelihood-ratio test (aLRT) algorithm (Anisimova & Gascuel 2006). For the ML phylogram of Fig 4, reliability of internal branches was also assessed by a bootstrap analysis (Felsenstein 1985). One thousand Bootstrap replicates were employed to determine confidence in the branches order. This bootstrap analysis was performed on line at the ATGC South of France Bioinformatics platform (http:// www.atgc-montpellier.fr/phyml/). The phylogenetic analyses used the PhyML 3.0 algorithms as described by Guindon et al. 2010. Results and discussion Analysis of the rDNA ITS1/5.8S/ITS2 region of the collected French Tricholoma species Fig 2 e Percentages of nucleotide identity observed between the sequences of the ITS1/5.8S/ITS2 region of the ten studied French Tricholoma sp. Strains. The sequences of the strains located in the same box possess more than 99 % of nt identity. Sporophores of French yellow Tricholoma strains were collected under Pinus pinaster forests in South France for Tricholoma auratum (strains AuFr1, 2, and 3) and under Pinus sylvestris or Abies alba for Tricholoma equestre (strains EqFr1, 2, and 3) in hill forests of various continental regions of France. However, it is to be noticed that several small Quercus pubescens trees were also present in the parcel of A. alba trees were the EqFr2 sporophores were collected (Table 1). Two additional strains with pale yellow to white gills were collected under Q. pubescens trees. One was referred as T. equestre var pallidifolia (strain EqFrW) and the second described as Tricholoma joachimii (strain JoFr). T. equestre var. populinum (strain EqFrPop) and Tricholoma frondosae (strain FrFr) were isolated under Betula pendula and Populus sp. trees for the EqFrPop and under Populus tremula trees for FrFr. These four latter strains were from hill forests of continental French regions (Table 1). The sequences of the ITS1/5.8S/ITS2 region of these ten strains were obtained and aligned as described in the ‘Materials and methods’ section. The three T. auratum strains (AuFr1, 2, and 3) from pine forests (P. pinaster) of South France and the three typical (i. e. with intense yellow gills) T. equestre 150 S. Moukha et al. Fig 3 e Most likely ML Phylogenetic trees constructed for the collected French Tricholoma strains based on the nuclear ribosomal region sequences (A) or the 50 region of the mitochondrial cox1 gene (B). The Maximum Likelihood method was used to construct the trees. Branch supports were computed by the aLRT statistical test algorithm. strains (EqFr1, 2, and 3) from French hill forests of conifers shared highly similar ITS region (Supplementary materials Fig S1 and Table S1) with a size ranging from 723 nt (EqFr2) to 749 nt (AuFr1). This size variation was mainly due to the presence in the ITS1 region of a microsatellite sequence consisting of repeat A nucleotides, from four in EqFr2 to 35 in AuFr3 (Supplementary materials Fig S1 and Table S1). In addition to this microsatellite region, up to 31 polymorphic sites corresponding to point mutations were revealed (Supplementary materials Fig S1). An additional deletion of 11 nt (beginning at position 215) was found in the ITS1 sequence of AuFr3. A consensus sequence was deduced from the alignment of the sequences of the six French T. auratum and T. equestre ribosomal regions (Supplementary Materials Fig S1). This consensus sequence was obtained by choosing for each position (column) of the alignment, the most frequently found nucleotide. This consensus sequence was found to be strictly identical to that of the EqFr1 strain. The most divergent sequence from the consensus sequence (represented by T. equestre EqFr1) is T. equestre EqFr2 harbouring 16 point mutations (seven in ITS1 and nine in ITS2) leading to 97.9 % sequence identity (Supplementary materials Fig S1 and Fig 2). Subsequently, sequences of T. equestre var. populinum EqFrPop and T. frondosae FrFr, collected under B. pendula or Populus sp. trees, were compared revealing up to 99 % identity (Fig 2). The sequences differ by one nt in ITS2 region and one nt in the 5.8 S rDNA; hence, the difference between these two species may be considered as trivial. When compared with the three French T. equestre strains (EqFr1, 2, and 3), T. frondosae, and T. equestre var. populinum appear closely related to EqFr2 (Supplementary materials Fig S1). Indeed, those three strains share 15 polymorphic sites not found in the T. equestre consensus sequence EqFr1 (97.9 % nt identity). This shows a close relationship between T. frondosae, T. equestre var. populinum, both mycorrhizal species of broad-leaved trees such as Betula or Populus species and the T. equestre Eqfr2 strain, collected under conifers (A. alba) but in a parcel with also several Q. pubescens trees. The sequences of the French species with pale to white gills (T. equestre var pallidifolia, EqFrW, and T. joachimii JoFr) are very close and differ by only four polymorphic sites, one nt indel and three nt substitutions. However, these two sequences share 66 polymorphic sites when compared with EqFr1 (eight indels from one to three nt and 58 nt substitutions). From these results, both species with pale to white gills possess 99.5 % of nt id. but only 91.2 % of nt id. with the consensus sequence of EqFr1 (Fig 2). This suggests that both species with pale to white gills have to be considered as a distinct species not closely related to the species of the T. equestre (¼Tricholoma flavovirens) complex carrying yellow gills. To confirm this conclusion, another molecular marker was added to our study. For the ten strains, a 595 nt sequence A molecular contribution to the assessment of the T. equestre species complex 151 Fig 4 e Un-rooted most likely ML Phylogenetic tree of the T. equestre complex and distant or related Tricholoma species based on the ITS sequences. The Maximum Likelihood method was used to construct the tree. The gamma shape parameter was estimated directly from the data and found equal to 1.078, the proportion of invariant sites was 0.332. Branch supports, statistically estimated by the aLRT are indicated above the branches. One thousand Bootstrap replicates were also employed to determine confidence in the branches order. These bootstrap values are indicated between brackets. Branches with branch for broad-leaved trees) is support values lower than 60 % are collapsed. The mycorrhizal association ( for conifers or indicated after the strain name. corresponding to the 50 region of the mitochondrial cox1 gene was amplified, cloned, and sequenced. The T. equestre var pallidifolia and T. joachimii cox1 sequences were identical, confirming that the two strains can be considered as representatives of a single species named T. joachimii. Conversely, the T. joachimii cox1 sequences differed by 14 point mutations (97.6 % of nt identity) when compared with the sequences of the six T. auratum and T. equestre French strains. When strains with yellow gills were compared, no more than four point mutations were found (>99.3 % identity). This confirms that, from a molecular point of view, T. equestre var. pallidifolia and T. joachimii have to be considered as two representatives of a single species differing from T. auratum and T. equestre. Phylogenetic relationships of the French Tricholoma strains Two phylogenetic trees of the ten French strains were constructed using the PhyML program. The trees were obtained from ClustalW alignment of the nuclear ribosomal unit sequences (Fig 3A) and the 50 region of the mitochondrial cox1 gene (Fig 3B). Two major clades are clearly separated in both trees with branch support values equal to 1 (100 %). The first clade is composed of the strains harbouring pale to white gills (Tricholoma equestre var pallidifolia and Tricholoma joachimii); the second one is composed of all the other French strains with intense yellow gills, namely the Tricholoma auratum strains (AuFr1, 2, and 3), the T. equestre strains (EqFr1, 2, and 3) and the Tricholoma frondosae strain (FrFr). Hence, the phylogenetic trees confirm that the three T. auratum strains of the Atlantic coast of South France and the continental T. equestre strains from the centre of France are closely related, giving a molecular basis to the argument that the continental T. equestre species and the T. auratum species associated with the Pinus pinaster forest of the Atlantic coast belong to the same T. equestre species. In contrast, the clade composed by the strains harbouring pale to white gills (T. equestre var pallidifolia and T. joachimii) seems to represent another species (T. joachimii) not belonging to the T. equestre species complex. Phylogenetic analysis of worldwide Tricholoma species (section equestre) In order to extend the analysis to species belonging or not to the Tricholoma equestre complex and with various geographical 152 origins, an un-rooted ML phylogenetic tree was built (Fig 4) from the alignment, as previously described, and submitted to bootstrap analysis. In the resulting tree, representatives of the Tricholoma matsutake and Tricholoma pardinum species are grouped in two different clades, distant from the T. equestre complex. Despite the tree was constructed as an un-rooted tree, both species behave as outgroup to the T. equestre complex. In the same way and, despite the fact that they belong to the same section equestre, both Tricholoma portentosum representatives are grouped in a clearly separated clade from the analysed T. equestre, Tricholoma flavovirens, and Tricholoma auratum strains. Interestingly, both French strains with white or pale gills, i. e. T. equestre var. pallidifolia and Tricholoma joachimii (section parafucata) are located in a sister group to the T. equestre complex (i. e. T. equestre sensus stricto, the synonymised T. flavovirens and the T. auratum strains). Moreover, it will be noted that both studied Tricholoma columbetta strains (section albata) having also white gills appears closer, according to the branch length, to the T. equestre complex than T. joachimii and T. equestre var. pallidifolia strains (Fig 4). Hence, T. equestre var. pallidifolia can be considered as a representative of T. joachimii species not related to the T. equestre (¼T. flavovirens) and T. auratum species complex and characterised by sporophores with pale to white gills. Unfortunately, the position of the Japanese T. flavovirens (FlJa), whose the colour of the gills is not available, was not resolved in this tree. As shown in the tree (Fig 4), T. equestre strains are mixed with T. flavovirens and T. auratum strains, in relation with the minor and variable morphological differences of their sporophores. However, it is possible to define three clades or groups containing the T. equestre, T. flavovirens, and T. auratum strains. The largest one (Fig 4, C1) groups the three French T. auratum strains (AuFr1, 2, and 3), two French T. equestre strains (EqFr1 and EqFr3), four T. flavovirens strains from Portugal (FlPo1, 2, 3, and 4), one T. equestre from Norway (EqNo) and four T. flavovirens from USA (FlUs1, 2, 3, and 4) and one T. flavovirens from Canada (FlCa). C1 appears to group species related to the T. equestre complex which share an ecological niche constituted by conifers. Moreover, the species from Southwest Europe grouped in the T. equestre complex could be representatives of a single species. However, as most (FlUs1e3) of the American T. flavovirens strains grouped in a supported clade, these strains could represent a North American sub-species. The clade C3, composed of four Japanese stains, possesses a high branch support value (99 %), in relation with the presence in the four strains of eight common heteromorphisms (three in ITS1, two in 5.8S rDNA, and three in ITS2) and of three large deletions (from 11 nt to 28 nt) located in ITS1 (Table 2B). These molecular data argues for considering these four Japanese strains as representative of a new Asian species, or at least as a sub-species of T. equestre. As reported by Kikuchi et al. (2007), these Japanese strains, were collected under conifers, which is in agreement with their position in the phylogram near members of the T. equestre complex associated with conifers. Another clade C2 (Fig 4, C2) is composed of four strains: the French T. equestre EqFr2 collected in a parcel containing conifers (Abies alba) but also several Quercus pubescens (broadleaved) trees, two French strains collected under broad- S. Moukha et al. leaved trees T. equestre var populinum EqFrPop and Tricholoma frondosae FrFr and one T. equestre strain from Portugal (EqPo) whose the ecological niche is unknown. C2 is supported by a high bootstrap value of 970. This is in accordance with the analysis of the 15 heteromorphic loci shared by strains (Table 2). Indeed, the four strains grouped in C2 and associated with broad-leaved trees differ from the European strains of C1 in all these loci (15 different alleles revealed when comparing these C2 strains on one hand and, on the other hand, all the French and Portuguese T. flavovirens, T. auratum, and T. equestre strains of C1 which have been collected under conifers). Among the 15 heteromorphic loci, the four T. auratum strains of C3 collected in Japan conserved 13 divergent loci with the strains of C2, additionally to the eight polymorphic sites and tree large deletions reported above as characterizing C3 (Table 2). In the same way, the American T. flavovirens strains (FlUs1e4 from USA and FlCa from Canada) and the T. equestre strain (EqNo) from Norway keep from 8 to 10 divergent loci with the strains of C2. Moreover, as shown in Table 2, most (eight) of the alleles present in the strains of C2 are common to the Tricholoma species not related to the T. equestre complex (T. columbetta, T. joachimii, and T. portentosum). This gives strong arguments to consider the four strains grouped in C2 as four representatives of a single T. frondosae species specifically collected under broad-leaved trees. Indeed, this species would have conserved ancestral alleles in common with other non-equestre Tricholoma species and there is no clear evidence of a gene flow between them and the strains of the T. equestre complex, despite they share the same geographical area. In conclusion, the phylogenetic analyses of T. flavovirens, T. equestre, and T. auratum species with various geographical origins (Europe, North America, and temperate Asia) suggest that most of the strains collected under conifer trees in Europe have to be considered as representative of the same species T. equestre. This is in accordance with the previous report of Deng & Yao (2005) that synonymised T. flavovirens with T. equestre. From our report, all strains described as T. auratum, T. flavovirens, and T. equestre in Europe have to be considered as representative of a single toxinogenic species T. equestre. Furthermore, from an ecological point of view, our results argue for ascribing the variation observed in the strain ecological niche (conifers or broad-leaved trees) to two different species, separated by the ITS molecular marker: T. frondosae under broad-leaved trees and T. equestre under conifers. It will be noted that the phylogram does not allow the elimination of the hypothesis that strains associated with broad-leaved trees would represent a variety (var. populinum) of T. equestre. However, the lack of gene flow (reported above and shown in Fig 2) between the strains associated with broad-leaved trees and those of the T. equestre complex, rather argues for two distinct species depending on the ecological niche: T. frondosae under broad-leaved trees and T. equestre under conifers. In the same way, our results show that T. joachimii and T. equestre var. pallidifolia have to be considered as a clearly different and phylogenetically distant species from T. equestre. However, is to be noted that these infrequent white gilled species are representated in our study by only one specimen each and that does not allow the knowledge of intraspecific variations of the used molecular markers. It, consequently, A Clade (characteristics) Species (Strain) rDNA location ITS1 ITS1 ITS1 ITS1 ITS1 ITS1 ITS2 ITS2 ITS2 ITS2 ITS2 ITS2 ITS2 ITS2 ITS2 Heteromorphic locus H1 (104)a H2 (108)a H3 (109)a H4 (202)a H5 (252)a H6 (282)a H7 (460)a H8 (520)a H9 (533)a H10 (541)a H11 (551)a H12 (578)a H13 (579)a H14 (614)a H15 (649)a B-l (Eu) nd (Eu) CoþB-l (Eu) B-l (Eu) G G G G G G G G A A A A G G G G T T T T T T T T A A A A del-1ntd del-1nt del-1nt del-1nt C C C C T T T T C C C C T T T T G G G G T T T T T T T T Co (Ja) Co (Ja) Co (Ja) Co (Ja) Co (Eu) Co (Eu) Co (Eu) Co (Eu) Co (Eu) Co (Eu) Co (Eu) Co (Eu) Co (Eu) Co (N. Am) Co (N. Am) Co (N. Am) Co (N. Am) Co (N. Am) nd (EU) Co (Ja) B-l (Eu) B-l (Eu) A A A A A A A A A A A A A A A A G G A G G G A A A A A A A A A A A A A A A A A A A A G G G G G G G G G G G G G G G G G G A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A T T T T T T T T T T T T T A A A A A A A A A T T T T T T T T T T T T T T T T T T T T T T A A A A A A G T T del-1nt del-1nt del-1nt del-1nt C C C C T C T T T del-1nt del-1nt del-1nt del-1nt del-1nt del-1nt del-1nt del-1nt del-1nt T T T T T T T T T T T T T T T T T T T T del-1nt del-1nt C C C C C C C C C C C C C C C C C C C C T T T T T T T T T T T T T T T T T T T T T C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C del-1nt del-1nt del-1nt del-1nt del-1nt del-1nt del-1nt del-1nt del-1nt del-1nt del-1nt del-1nt del-1nt del-1nt del-1nt del-1nt del-1nt del-1nt del-1nt del-1nt T T G G G G G G G G G G G G G T T T T T T T C C B-l (N. Am) B-l (Eu) Co (N. Am) Co (N. Am) G G G G G G G G A A A A A A A A T T T T T T T T T T T T del-1nt del-1nt del-1nt del-1nt del-1nt del-1nt del-1nt del-1nt T T C C T T T T T T C C C C del-1nt del-1nt del-1nt del-1nt del-1nt del-1nt T T T T Characteristicsb (Originc) C2 (clade T. Frondosae) C3 (Japan) C1 (complex T. equestre) 0 Clade T. joachimii Clade T. columbetta Outgroup T. frondosae FrFr T. equestre EqPo T. equestre EqFr2 T. equestre var. populinum EqFrPop T. auratum AuJa1 T. auratum AuJa2 T. auratum AuJa3 T. auratum AuJa4 T. flavovirens FlPo1 T. flavovirens FlPo2 T. flavovirens FlPo3 T. flavovirens FlPo4 T. auratum AuFr1 T. auratum AuFr2 T. auratum AuFr3 T. equestre EqFr1 T. equestre EqFr3 T. flavovirens FlUs1 T. flavovirens FlUs2 T. flavovirens FlUs3 T. flavovirens FlUs4 T. flavovirens FlCa T. equestre EqNo T. flavovirens FlJa T. joachimii JoFr T. equestre var. pallidifolia EqFrW T. columbetta Us T. columbetta Po T. portentosum Ca1 T. portentosum Ca2 A molecular contribution to the assessment of the T. equestre species complex Table 2 e A) Distribution of the shared heteromorphisms in the analysed Tricholoma strains; (B) Specific heteromorphisms carried by the Japanese T. auratum strains. (continued on next page) 153 0 C T C Acknowledgements insþ1nt(C) T del-1nt prevents the use of these sequences for identification/barcoding purpose. Interestingly, T. joachimii and the species of the T. equestre complex are phylogenetically related to other species with white gills such as T. columbetta. This species from the albata section is also renowned as an edible mushroom. However, the phylogenetic relationship observed between these species with white gills and the T. equestre complex with intense yellow gills strongly suggests that toxicologic studies are necessary to evaluate the potential danger (rhabdomyolitic effect) of consumption of T. columbetta mushrooms in high quantity. We would like to thank all mycologists who graciously provide
rezvouchers for this research: Laurent Deparis, Miquel A. Pe
a. De-Gregorio, Olivier Roblot, Philippe Bineau, and Jean Rove DNA sequencing was mainly performed at the Genotyping
Bordeaux Segalen and Sequencing Facility of Universite
gional d’Aquitaine n 20030304 (grants from the Conseil Re 002FA and 20040305003FA, and from the European Union, FEDER n 2003227). The authors also thank David Stern for helpful comments and suggestions. J10 (655) J9 (521) T J8 (317) C G insþ11nt insþ28nt del-1nt del-11nt del-28nt J2 (149) C Co/B-l (Eu, N. Am) A Co (Ja) C1 (complex T. equestre) and C2 (clade T. frondosae) T. auratum AuJa1, 2, 3 4 All strains C3 (Japan) J1 (65) Heteromorphic locus/characteristicsb (originc) Appendix A. Supplementary data Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.funbio.2013.01.003. references a Numbers indicate the position of the polymorphic nt in the consensus EqFr1 sequence. b B-l: broad-leaved trees, Co: conifers, nd: not determined. c EU: Europe, Ja: Japan, N. Am: North America. 0 ins þ11nt T insþ1nt(T) del-11 J7 (311) J6 (237) J5 (208) J4 (204) J3 (186) ITS1 ITS1 rDNA location Species (strain) Clade (characteristics) B Table 2 e (continued ) a ITS1 a ITS1 a a ITS1 a ITS1 a 5.8S a 5.8S a ITS2 a ITS2 a J11 (668)a S. Moukha et al. ITS2 154 Afssa Saisine N 2002-SA-0285. Avis de l’Agence française de
curite
sanitaire des aliments relatif a  une demande se
valuation en terme de sante
publique du risque e
ventuel lie
d’e  la consommation de Tricholome e
questre. http: a //www.afssa.fr/Documents/AUT2002sa0285.pdf. ^te
du 16 juin 2004 portant sur la suspension d’importation et Arre
du Tricholome
equestre et ordonnant son de mise sur le marche retrait. Le Journal Officiel de la R
epublique Française 0142 Le 20 juin 2004: 11099. http://www.legifrance.gouv.fr/affichTexte. do?cidTexte¼JORFTEXT000000804245&dateTexte¼.
cret n 2005-1184 du 19 septembre 2005 portant interdiction de De ces sous espe ces ou varie
te
s de champignons. plusieurs espe Le Journal Officiel de la R
epublique Française 0220 le 21 septembre 2005. http://www.legifrance.gouv.fr/affichTexte.do? cidTexte¼JORFTEXT000000604247&dateTexte¼. Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ, 1990. Basic local alignment search tool. Journal of Molecular Biology 215: 403e410. Anisimova M, Gascuel O, 2006. Approximate likelihood-ratio test for branches: a fast, accurate, and powerful alternative. Systematic Biology 55: 539e552. Barroso G, Blesa S, Labarere J, 1995. Wide distribution of mitochondrial genome rearrangements in wild strains of the cultivated basidiomycete Agrocybe aegerita. Applied and Environmental Microbiology 61: 1187e1193. Bedry R, Baudrimont I, Deffieux G, Creppy EE, Pomies JP, Ragnaud JM, Dupon M, Neau D, Gabinski C, De Witte S, Chapalain JC, Godeau P, Beylot J, 2001. Wild-mushroom intoxication as a cause of rhabdomyolysis. The New England Journal of Medicine 345: 798e802. A molecular contribution to the assessment of the T. equestre species complex Bidartondo MI, Bruns TD, 2001. Extreme specificity in epiparasitic Monotropoideae (Ericaceae): widespread phylogenetic and geographical structure. Molecular Ecology 10: 2285e2295. Bidartondo MI, Bruns TD, 2002. Fine-level mycorrhizal specificity in the Monotropoideae (Ericaceae): specificity for fungal species groups. Molecular Ecology 11: 557e569.
dry R, Gromb S, 2009. Intoxications specific to the Aquitaine Be region. La Revue de M
edecine Interne 30: 640e645. Bullock WO, Fernandez JM, Stuart JM, 1987. XL1-Blue: a high efficiency plasmid transforming recA Escherichia coli strain with beta-galactosidase selection. Bio Techniques 5: 376e379. Chevenet F, Brun C, Banuls AL, Jacq B, Christen R, 2006. TreeDyn: towards dynamic graphics and annotations for analyses of trees. BMC Bioinformatics 7: 439. Chodorowski Z, Waldman W, Sein Anand J, 2002. Acute poisoning with Tricholoma equestre. Przeglad Lekarski 59: 386e387. Damon C, Barroso G, Ferandon C, Ranger J, Fraissinet-Tachet L, Marmeisse R, 2010. Performance of the COX1 gene as a marker for the study of metabolically active Pezizomycotina and Agaricomycetes fungal communities from the analysis of soil RNA. FEMS Microbiology Ecology 74: 693e705. Deng H, Yao YJ, 2005. Tricholoma equestre, the correct name for T. flavovirens (Agaricales). Mycotaxon 94: 325e329. Dereeper A, Audic S, Claverie JM, Blanc G, 2010. BLAST-EXPLORER helps you building datasets for phylogenetic analysis. BMC Evolutionary Biology 10: 8. Dereeper A, Guignon V, Blanc G, Audic S, Buffet S, Chevenet F, Dufayard JF, Guindon S, Lefort V, Lescot M, Claverie JM, Gascuel O, 2008. Phylogeny.fr: robust phylogenetic analysis for the non-specialist. Nucleic Acids Research 36: W465eW469. Felsenstein J, 1985. Confidence limits of phylogenies: an approach using the bootstrap. Evolution 39: 783e791. Gardes M, Bruns TD, 1993. ITS primers with enhanced specificity for basidiomycetes e application to the identification of mycorrhizae and rusts. Molecular Ecology 2: 113e118. Guindon S, Gascuel O, 2003. A simple, fast, and accurate algorithm to estimate large phylogenies by maximum likelihood. Systematic Biology 52: 696e704. 155 Guindon S, Dufayard JF, Lefort V, Anisimova M, Hordijk W, Gascuel O, 2010. New algorithms and methods to estimate maximum-likelihood phylogenies: assessing the performance of PhyML3.0. Systematic Biology 59: 307e321. Hanahan D, 1985. Techniques for transformation in Escherichia coli. In: Glover DM (ed.), DNA Cloning: a practical approach. IRL Press, Oxford-Washington, pp. 1e47. Kikuchi K, Matsushita N, Suzuki K, 2007. Discrimination of Tricholoma species by species-specific ITS primers. Mycoscience 48: 316e320. Kranabetter JM, Friesen J, Gamiet S, Kroeger P, 2009. Epigeous fruiting bodies of ectomycorrhizal fungi as indicators of soil fertility and associated nitrogen status of boreal forests. Mycorrhiza 19: 535e548. Matsushita N, Kikuchi K, Sasaki Y, Guerin-Laguette A, Lapeyrie F, Vaario L-M, Intini M, Suzuki K, 2005. Genetic relationship of Tricholoma matsutake and T. nauseosum from the Northern Hemisphere based on analyses of ribosomal DNA spacer regions. Mycoscience 46: 90e96. Matsuura M, Saikawa Y, Inui K, Nakae K, Igarashi M, Hashimoto K, Nakata M, 2009. Identification of the toxic trigger in mushroom poisoning. Nature Chemical Biology 5: 465e467. Nieminen P, Karja V, Mustonen AM, 2008. Indications of hepatic and cardiac toxicity caused by subchronic Tricholoma flavovirens consumption. Food and Chemical Toxicology 46: 781e786. Nieminen P, Mustonen AM, Kirsi M, 2005. Increased plasma creatine kinase activities triggered by edible wild mushrooms. Food and Chemical Toxicology 43: 133e138. Thompson JD, Higgins DG, Gibson TJ, 1994. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Research 22: 4673e4680. White TJ, Bruns TD, Lee S, Taylor J, 1990. Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. In: Innis MA, Gelfand DH, Sninsky JJ, White TJ (eds), PCR Protocols: a guide to methods and applications. Academic press, San Diego, pp. 315e322.