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, Gerard BARROSOa,c,*
a
INRA (Institut National de la Recherche Agronomique) Bordeaux-Aquitaine, UR1264 MycSA (Mycologie et Securite des Aliments), BP81,
33883 Villenave d’Ornon Cedex, France
b
Laboratoire de Toxicologie et Hygiene Appliquee, UFR des Sciences Pharmaceutiques, France
c
Universite Bordeaux Segalen, 146, rue Leo 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 coldry & 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 Republique 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 Republique 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 Medecine 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.