Mycol. Res. 108 (8): 885–896 (August 2004).

f The British Mycological Society 885

DOI: 10.1017/S0953756204000620 Printed in the United Kingdom.

Molecular phylogeny and biogeography of the widely

distributed Amanita species, A. muscaria and A. pantherina

Takashi ODA*, Chihiro TANAKA and Mitsuya TSUDA

Laboratory of Environmental Mycoscience, Graduate School of Agriculture, Kyoto University, Kyoto, 606-8502, Japan.
E-mail : takashi@kais.kyoto-u.ac.jp

Received 19 March 2003; accepted 15 May 2004.

The molecular phylogeny and biogeography of two widely distributed Amanita species, A. muscaria and A. pantherina,
were studied based on specimens from diverse localities. Analyses of both a partial sequence of the ITS region of nuclear
DNA and a partial sequence of the b-tubulin gene were able to resolve specimens of each species. Analyses revealed a
greater divergence of the b-tubulin region than the ITS region. Based on molecular phylogeny of the combination of the
ITS and b-tubulin regions, A. muscaria could be separated into at least three groups (Eurasian, Eurasian subalpine, and
North American), and A. pantherina could be separated into at least two groups (North American and Eurasian). We
hypothesize that the speciation of A. muscaria occurred in Eurasia with subsequent migration to North America via
land bridges. However, it is impossible to determine whether A. pantherina moved from Eurasia to North America or
vice versa. For both A. muscaria and A. pantherina, the intracontinental relationships of both Eurasia and North
America were closer than the relationships between eastern Asia and eastern North America.

INTRODUCTION agaric, ’ is widely distributed in Europe, Asia, Africa,

Australia, New Zealand, and North, Central, and
The genus Amanita has a global distribution and is one
South America (e.g. Gilbert 1941, Jenkins 1977, 1986,
of the most well-known genera of macrofungi ; most
Reid 1980, Imazeki & Hongo 1987, Pérez-Silva
species are ectomycorrhizal. Recent molecular phylo-
& Herrera 1991, Reid & Eicker 1991, Ridley 1991,
genetic studies of this genus have contributed to the
Tulloss, Ovrebo & Halling 1992, Adhikari & Parajuli
re-examination of its traditional classifications based
1994). A. pantherina, ‘ panther cap’, is also a well-
on morphology (e.g. Weiß, Yang & Oberwinkler 1998,
known poisonous fungus and is characterized by a
Drehmel, Moncalvo & Vilgalys 1999, Oda, Tanaka &
brownish pileus with many volval warts, a bulbose stipe
Tsuda 1999, González et al. 2002). However, in these
base, and the absence of clamps. This species is dis-
studies, little attention was given to genetic variation
tributed in Europe, Asia, Africa, and North and Cen-
within species or to the relationships between the mol-
tral America (e.g. Gilbert 1941, Jenkins 1977, 1986,
ecular phylogeny and the biogeography of each species
Imazeki & Hongo 1987, Pérez-Silva & Herrera 1991,
or of the genus Amanita as a whole. Recently, molecu-
Reid & Eicker 1991, Adhikari & Parajuli 1994, Yang
lar phylogenetic studies of macrofungi with regard
1997). We conducted a molecular phylogenetic study of
to biogeography have attracted more attention (e.g.
specimens of both species from diverse localities based
Vilgalys & Sun 1994, Wu et al. 2000, Hibbett 2001,
on a partial sequence of the ITS region of nuclear DNA
Ko et al. 2001, Mueller et al. 2001).
and a partial sequence of the b-tubulin gene. Based on
In this study, we examined the molecular phylogeny
our results, we discuss the biogeography of these two
and biogeography of two widely distributed Amanita
Amanita species.
species, A. muscaria and A. pantherina, which belong to
the subgenus Amanita section Amanita. A. muscaria is
a large mushroom that is characterized by a red pileus
MATERIALS AND METHODS
with many volval warts, a bulbose stipe base, and the
presence of clamps. This famous toxic species, the ‘fly Specimens examined
The 49 specimens that were examined in this study
* Corresponding author. are listed in Table 1. With the exception of Amanita
Molecular phylogeny and biogeography of Amanita 886

Table 1. Details and nucleotide sequence length (bp) of the ITS and b-tubulin regions of 49 specimens of Amanita species studied.

ITS b-tubulin

Specimen No.a Geographic origin Accession no.b,c Length(bp) Accesion no.b,c Length(bp)

Section Amanita
A. muscaria FB-30961 (CBM) Aomori-shi, Aomori, Japan AB080980 695 AB095892* 489
FB-30962 (CBM) Kitakoma-gun, Yamanashi, Japan AB080981 695 AB095893* 489
FB-30963 (CBM) Kitakoma-gun, Yamanashi, Japan AB080982 695 AB095894* 489
FB-30976 (CBM) Kiso-gun, Nagano, Japan AB081294 695 AB095895* 489
FB-30977 (CBM) Ohno-gun, Gifu, Japan AB081295 695 AB095896* 489
FB-30985 (CBM) Ohno-gun, Gifu, Japan AB096048* 695 AB095897* 489
FB-30978 (CBM) Chino-shi, Nagano, Japan AB081296 697 AB095858* 487
FB-30981 (CBM) Chino-shi, Nagano, Japan AB096049* 697 AB095859* 487
FB-30982 (CBM) Chino-shi, Nagano, Japan AB096050* 697 AB095860* 487
FB-30964 (CBM) Gdynia, Poland AB080983 696 AB095899* 486
FB-30965 (CBM) Gdansk, Poland AB080984 695 AB095900* 486
M-31452 (K) Hampshire, England, UK AB080777 695 AB095901* 486
M-31445 (K) Surrey, England, UK AB080778 695 AB095902* 486
M-80048 (K) Surrey, England, UK AB080779 696 AB095903* 486
FB-30987 (CBM) Queenstown, Otago, New Zealand AB096052* 696 AB095904* 489
45843 (NY) Hampshire, Massachusetts, USA AB080788* 697 AB095884* 471
45785 (NY) Hampshire, Massachusetts, USA AB080789* 697 AB095885* 471
45840 (NY) Lawrence, Massachusetts, USA AB080791* 697 AB095887* 471
45820 (NY) Bronx, New York, USA AB080790* 698 AB095886* 471
45863 (NY) Mendocino, California, USA AB080787* 703 AB095883* 471
var. formosa 45883 (NY) Piscataquis, Massachusetts, USA AB080792* 697 AB095888* 471
HDT45060 (SFSU) Amador, California, USA AB080795* 702 AB095891* 471
HDT44761 (SFSU) Alpine, California, USA AB080794* 703 AB095890* 471
var. alba HDT49100 (SFSU) Cascade, Idaho, USA AB080793* 701 AB095889* 471
var. regalis 506 (O) Dovre, Oppland, Norway AB080780 697 AB095855* 487
1539 (O) Gjøvik, Oppland, Norway AB080781 697 AB095856* 487
4220 (O) Nordre-Land, Oppland, Norway AB080782 697 AB095857* 487
A. pantherina FB-15998 (CBM) Chiba-shi, Chiba, Japan AB080978 704 AB095867* 496
FB-30956 (CBM) Sendai-shi, Miyagi, Japan AB080973 702 AB095870* 496
FB-30957 (CBM) Ohno-gun, Gifu, Japan AB080974 703 AB095871* 495
FB-30958 (CBM) Ohtsu-shi, Shiga, Japan AB080975 703 AB095872* 495
FB-30959 (CBM) Kyoto-shi, Kyoto, Japan AB080977 703 AB095868* 496
FB-30960 (CBM) Kyoto-shi, Kyoto, Japan AB080976 703 AB095869* 496
TNS-F-4490 (TNS) Nagarkot, Kathmandu, Nepal AB096043* 704 AB095876* 496
TNS-F-4491 (TNS) Phulchowki, Kathmandu, Nepal AB096044* 703 AB095877* 496
TNS-F-4492 (TNS) Phulchowki, Kathmandu, Nepal AB096045* 704 AB095878* 496
M-31408 (K) Devon, England, UK AB080774 705 AB095873* 497
M-61495 (K) Surrey, England, UK AB096046* 706 AB095875* 497
M-79855 (K) West Suffolk, England, UK AB080775 706 AB095874* 496
45927 (NY) Santa Barbara, California, USA AB080785* 708 AB095881* 494
45929 (NY) San Francisco, California, USA AB080784* 708 AB095880* 492
FB-30984 (CBM) Loma Mar, California, USA AB096047* 708 AB095879* 492
A. aff. muscaria FB-30986 (CBM) Aomori-shi, Aomori, Japan AB096051* 698 AB095898* 488
A. concentrica FB-24901 (CBM) Awa-gun, Chiba, Japan AB080783 730 AB095849* 489
A. ibotengutake FB-30969 (CBM) Kyoto-shi, Kyoto, Japan AB080988 713 AB095848* 491
A. melleiceps FB-30953 (CBM) Ohita-shi, Ohita, Japan AB015688 706 AB095854* 491
A. rubrovolvata FB-30954 (CBM) Ohno-gun, Gifu, Japan AB015689 721 AB095850* 495
A. sinensis FB-30955 (CBM) Itoigawa-shi, Niigata, Japan AB080979 721 AB095861* 493
Outgroup
Section Phalloideae
A. pseudoporphyria FB-30951 (CBM) Kyoto-shi, Kyoto, Japan AB015702 636 AB095905* 485
a
Specimen no. in italic represents a type specimen.
b
The nucleotide squence data appear in the DDBJ/EMBL/GenBank nucleotide sequence databases.
c
Accession no. with an * means that sequence was determined in this study. Others were previously published in Oda et al. (1999) and Oda
et al. (2002).

pseudoporphyria, which was used as the outgroup, all Zealand, Norway, Poland, the UK, and the USA ; and
specimens belonged to subgenus Amanita section A. pantherina from Japan, Nepal, the UK, and the
Amanita. We examined specimens of different geo- USA. These specimens are deposited in the following
graphical origins: A. muscaria from Japan, New collections : Natural History Museum and Institute,
T. Oda, C. Tanaka and M. Tsuda 887

Chiba (CBM); Herbarium of Cryptogams, Kunming according to the manufacturer’s recommendations.

Institute of Botany, Academia Sinica, Kunming The sequence primers used were M13-20 and M13-RV
(HKAS) ; Royal Botanic Gardens, Kew (K) ; New (Pharmacia Biotech, Piscataway). Terminated samples
York Botanical Garden, Bronx, New York (NY) ; were electrophoresed on a CEQTM2000 DNA Analysis
Botanical Museum, Oslo (O) ; Harry D. Thiers Her- System (Beckman Coulter), and sequence data were
barium, San Francisco State University, San Francisco generated. For sequence alignments, the data of the
(SFSU) ; and the Botany Department, National Science ITS region were used with primers ITS4 and ITS5,
Museum, Tsukuba (TNS). whereas the data of the b-tubulin region were used
without primers B-TUB-1F and B-TUB-1R. The
nucleotide sequence data are deposited in the DDBJ/
DNA preparations
EMBL/GenBank nucleotide sequence databases. For
A minute slice of dried fruit body (ca 10 mg) was sus- the ITS region, some of the data had been previously
pended in 500 ml of extraction buffer (50 mM Tris–HCl published in Oda et al. (1999) and Oda et al. (2002)
pH 8.0, 125 mM EDTA, 100 mM NaCl, 2% (w/v) (Table 1).
sodium N-dodecanoylsarcosinate, 1% (v/v) 2-mercap-
toethanol). DNA was extracted by the method of
Sequence alignment and parsimony analysis
Nakada et al. (1994).
The sequences of each region were aligned using the
CLUSTAL W multiple alignment program ver. 1.8.
PCR amplification
(Thompson, Higgins & Gibson 1994). Parsimony
The ITS region was amplified with the primers ITS4 analyses of the ITS region, the b-tubulin region, and
and ITS5 (White et al. 1990). This region consists of a the combination of the two regions were performed
portion of 18S rDNA, ITS1, 5.8S rDNA, ITS2, and a with PAUP* version 4.0b9 (Swofford 2002). A heuristic
portion of 28S rDNA (van Nues et al. 1994). The re- search was conducted with the following conditions:
action mixture contained 10 pmol of each primer, 1.5 U alignment gapmode was newstate, the starting tree
of KOD Dash (Toyobo, Osaka), and about 10 ng of was obtained via stepwise addition ; addition sequence
template DNA in a volume of 50 ml. A thermal cycler was simple (reference taxon=FB-30951(CBM)); the
(iCycler, Bio-Rad, Hercules) was programmed as branching-swapping algorithm was tree-bisection-
follows : initial denaturation, 1 min at 95 xC ; 30 cycles reconnection (TBR), and the ‘MulTrees ’ option was
of 0.5 min at 95 x, 0.03 min at 49 x, and 0.5 min at in effect. Bootstrap analyses were performed using
72 x. For the partial sequence of the b-tubulin gene, we ‘full heuristic ’ with 500 replicates. The sequence align-
used the newly designed PCR primers B-TUB-1F (5k- ment has been submitted to TreeBASE, a relational
YMGNCCNGAYAAYTTYGTNTTYG-3k), B-TUB- database of phylogenetic information (URL : http://
1R (5k-TANARNGCYTCRTTRTCDATRCARAA- www.treebase.org/treebase/index.html).
3k), which were based on data of the b-tubulin gene of
basidiomycetes available in the DDBJ/EMBL/Gen-
Bank nucleotide sequence databases. The region RESULTS
amplified with this primer set consisted of three exons
Analysis of the ITS region
and two introns. The reaction mixture contained
10 pmol of each primer, 1.5 U of Taq DNA polymerase The sequences of the ITS region ranged in length from
(Takara, Otsu), and about 10 ng of template DNA in a 636 to 730 bp among the 49 specimens (Table 1). As a
volume of 50 ml. The thermal cycler (iCycler, Bio-Rad) result of the alignment of these sequences, 114 charac-
was programmed as follows : initial denaturation, ters were parsimony-informative, 215 variable charac-
4 min at 95 x ; then 30 cycles of 1 min at 95 x, 1 min at ters were parsimony-uninformative, and 463 characters
48 x, and 1.5 min at 72 x ; and final extension at 72 x for were constant out of 792 total characters. Based on the
2 min. aligned sequences, heuristic searches revealed 12 most
parsimonious trees with a length of 552, a consistency
index (CI) of 0.7663, a retention index (RI) of 0.8522,
Cloning and DNA sequencing
and a homoplasy index (HI) of 0.2337. One of the most
After electrophoresis in a low-melt agarose gel (Sea parsimonious trees is shown in Fig. 1. For both Ama-
Plaque GTG Agarose, FMC), the amplified products nita muscaria and A. pantherina, each parsimonious
were excised from the gel. The DNA fragments were tree shows the same topology, except the placements
cloned into pZErOTM-2 (Invitrogen, Carlsbad). To within the A. muscaria clade of the UK, Poland, Japan
avoid artifact DNA sequences caused by errors in and New Zealand. The placements of the other related
DNA polymerization, at least three recombinants species were unstable.
were picked up from a batch of transformants, and Specimens of A. muscaria were separated into three
the homogeneity or majority of DNA sequences clades, i.e., the clade consisting of the UK, Poland,
was confirmed. DNA was sequenced with CEQTM Japan, and New Zealand (bootstrap value=98 %) ; the
DTCS Quick Start Kit (Beckman Coulter, Fullerton) clade of Norway and Japan (bootstrap value=99 %) ;
Molecular phylogeny and biogeography of Amanita 888

Fig. 1. One of the 12 most parsimonious trees based on the sequences of the ITS regions. Amanita pseudoporphyria was
used as the outgroup. Bootstrap values over 50 % are indicated at the base of the corresponding clade. For A. muscaria and
A. pantherina, each parsimonious tree shows the same topology, with the exception of the placements within the A. muscaria
clade of the UK, Poland, Japan, and New Zealand. Placements of the other related species were unstable.

and the clade of the USA (bootstrap value=99 %). the clade of Norway and Japan had no geographically
Amanita aff. muscaria fell on a branch basal to the three separated subclades. Specimens of A. pantherina were
clades of A. muscaria. The USA clade included the sub- separated into two clades, the USA clade (bootstrap
clade of the western USA (bootstrap value=98 %), value=100 %) and the clade composed of the UK,
which contained 45863 (NY), HDT44761 (SFSU), Nepal, and Japan (bootstrap value=93 %). The
HDT49100 (SFSU), and HDT45060 (SFSU). The connection of the two clades received 72% bootstrap
clade of the UK, Poland, Japan, and New Zealand and support.
T. Oda, C. Tanaka and M. Tsuda 889

Fig. 2. One of eight most parsimonious trees based on the sequences of the b-tubulin region. Amanita pseudoporphyria was used
as the outgroup. Bootstrap values over 50 % are indicated at the base of the corresponding clade. With the exception of the
placements within the A. pantherina clade of the UK, Nepal, and Japan, each parsimonious tree shows the same topology.

Analysis of the b-tubulin region length of 376, a CI of 0.7420, an RI of 0.9062, and an

HI of 0.2580. One of the most parsimonious trees is
The b-tubulin region ranged in length from 471 to shown in Fig. 2. Except for the placements within the
497 bp (Table 1). As a result of the alignment of these Amanita pantherina clade of the UK, Nepal, and Japan,
sequences, 117 characters were parsimony-informative, each parsimonious tree showed the same topology.
90 variable characters were parsimony-uninformative, Specimens of A. muscaria were separated into four
and 299 characters were constant out of 506 total clades, i.e., the clade of Norway and Japan (bootstrap
characters. Based on these aligned sequences, heuristic value=74 %), the clade of Japan and New Zealand
searches revealed eight most parsimonious trees with a (bootstrap value=66%), the clade of the UK and
Molecular phylogeny and biogeography of Amanita 890

Poland (bootstrap value=88%), and the clade of the received 100 % bootstrap support. The clade of the
USA (bootstrap value=100%). Specimens that were in UK, Nepal, and Japan consisted of three geo-
the clade of the UK, Poland, Japan, and New Zealand graphically separated subclades : the UK (bootstrap
based on the ITS tree (Fig. 1) were separated into the value=99 %), Nepal (bootstrap value=93 %), and
clade of the UK and Poland and the clade of Japan and Japan (bootstrap value=80%).
New Zealand in the b-tubulin region tree. The clade of The combined data analysis yielded results similar to
the UK and Poland was placed next to the clade of the those based on the ITS region analysis, showing three
USA ; however, this connection was not strongly sup- distinct clades of A. muscaria and two distinct clades of
ported. In addition, A. aff. muscaria was placed among A. pantherina. Moreover, in the combined data tree, the
the A. muscaria clades, although this connection did node of each clade of the two species received much
not receive >50 % support. The clade of Norway and greater bootstrap support. The combined data analysis
Japan clearly showed a geographical separation into a supported most of the results produced by the b-tubu-
Norway sub-clade (bootstrap value=73 %) and a lin analysis, except for the placements of the A. mus-
Japan sub-clade (bootstrap value=90 %). Specimens caria clade of the UK and Poland and A. aff. muscaria ;
of A. pantherina were separated into two clades, the in contrast to the b-tubulin region tree, the combined
USA clade (bootstrap value=100 %) and the clade of data tree placed the A. muscaria clade of the UK and
the UK, Nepal, and Japan (bootstrap value=99 %). Poland as a sister clade of the clade of Japan and NZ
The connection of the two clades received 99 % boot- with 93 % bootstrap support, and A. aff. muscaria
strap support. was placed on a branch basal to the three clades of
A. muscaria with 99 % bootstrap support.
Analysis of the combination of the ITS region and the
b-tubulin region
DISCUSSION
The combination of the ITS and b-tubulin regions
ranged in length from 1121 to 1219 bp. As a result of Both the ITS region and the b-tubulin region were able
the alignments of these sequences, 231 characters were to resolve specimens of each of the two species. In
parsimony-informative, 305 variable characters were particular, b-tubulin region analysis showed greater
parsimony-uninformative, and 762 characters were con- divergence than did ITS region analysis. For example,
stant out of 1298 total characters. Based on the aligned Amanita muscaria specimens from Norway and Japan
sequences, heuristic searches revealed 108 most parsi- (506 (O), 1539 (O), 4220 (O), FB-30978 (CBM), FB-
monious trees with a length of 945, a CI of 0.7429, an 30981 (CBM), FB-30982 (CBM)) that were placed in
RI of 0.8726, and an HI of 0.2571. One of the most a single clade of the ITS region tree were clearly sep-
parsimonious trees is shown in Fig. 3. Except for the arated into the subclades of Norway and of Japan in
placements within the A. pantherina subclade of Japan, the b-tubulin region tree. Despite the differences among
the A. muscaria subclade of the UK and Poland, and the three analysis trees in the placements of A. muscaria
the A. muscaria subclade of the western USA, each from the UK and Poland and A. aff. muscaria, the
parsimonious tree shows the same topology. combined data analysis revealed distinct clades and
Specimens of A. muscaria were separated into three subclades of the two species with much higher boot-
clades : (1) the UK, Poland, Japan, and New Zealand strap values. Thus, the results provided by the com-
collections (bootstrap value=93 %) ; (2) Norway and bined data are considered reliable. We therefore discuss
Japan (bootstrap value=100 %) ; and (3) the USA the molecular phylogeny and biogeography based on
(bootstrap value=100 %). Amanita aff. muscaria fell on the combined data analysis. In Fig. 3, the three
a branch basal to the three clades of A. muscaria A. muscaria clades are labelled M-I, M-II, and M-III;
(bootstrap value=99%). The clade of the UK, Poland, and the two A. pantherina clades are P-I and P-II.
Japan, and New Zealand consisted of two subclades : Tables 2–3 detail the variable sites in the alignments of
one of the UK and Poland (bootstrap value=98 %), the ITS and b-tubulin regions among the clades and
and the other of Japanese and New Zealand collections subclades of each species.
(bootstrap value=88 %). The clade of Norway and
Japan consisted of two subclades, one for Norway
Amanita muscaria
(bootstrap value=75 %) and one for Japan (bootstrap
value=91 %). In the USA clade, four specimens, We termed clade M-III, which consisted of specimens
HDT49100 (SFSU), HDT44761 (SFSU), HDT45060 from the western and eastern USA, the ‘North Amer-
(SFSU), and 45863 (NY), made up the subclade of the ican Group ’. Jenkins (1977, 1986) examined A. mus-
western USA (bootstrap value=98 %) and were sep- caria from the USA morphologically and delineated six
arate from the other five specimens from the eastern varieties (var. muscaria, var. alba, var. flavivolvata, var.
USA. Specimens of A. pantherina were separated into formosa, var. persicina, and var. regalis) which differ
two clades, the USA clade (bootstrap value=100%) mainly in the colour of the pileus and volval remnants.
and the clade of the UK, Nepal, and Japan (bootstrap We examined three specimens of A. muscaria var.
value=100 %). The connection between the two clades formosa, which has a yellow pileus, 45883 (NY),
T. Oda, C. Tanaka and M. Tsuda 891

Fig. 3. One of the 108 most parsimonious trees based on the sequences of the ITS and b-tubulin regions. Amanita
pseudoporphyria was used as the outgroup. Bootstrap values over 50 % are indicated at the base of the corresponding clade.
With the exception of the placements within the A. pantherina subclade of Japan, the A. muscaria subclade of the UK and
Poland, and the A. muscaria subclade of the western USA, each parsimonious tree shows the same topology.

HDT44761 (SFSU), and HDT45060 (SFSU) ; this var- subclade and were separate from the other five spe-
iety did not appear to be monophyletic. Four speci- cimens from the eastern USA. Therefore, it appears
mens from the western USA, i.e. HDT49100 (SFSU) that A. muscaria var. formosa does not reflect its mol-
from Idaho and 45863 (NY), HDT44761 (SFSU), and ecular phylogeny and that instead the molecular phy-
HDT45060 (SFSU) from California, represented one logeny coincides with the geographical distribution.
 Molecular phylogeny and biogeography of Amanita
Table 2. Variable sites in the alignment of the ITS and b-tubulin regions among groups of Amanita muscaria. The position numbers of the nucleotides are given
(read from top to bottom).
ITS β-tubulin
1111 1111 222 22222 2222222 444 444 4455 55555 555 666 777
999 1111 3333 000 22222 3444444 666 777 9900 44444 777 777 222 666667777777777888888888
Group Specimen No. 345 1234 5678 345 34567 9012345 456 678 7890 01234 678 234 123 567890123456789012345678
M-I FB-30964(CBM) -AT CTC- CTGT CCT GCT-G TATTT-A TGT TG- T--- T---- ATA -TC T-G CG---TGTTT---ATTCTTGTGTC
(UK & Poland) FB-30965(CBM) -AT CTC- CTGT CCT GCT-G TATTT-A TGT TG- T--- T---- ATA -TC T-G CG---TGTTT---ATTCTTGTGTC
M-31452(K) -AT CTC- CTGT CCT GCT-G TATTT-A TGT TG- T--- T---- ATA -TC T-G CG---TGTTT---ATTCTTGTGTC
M-31445(K) -AT CTC- CTGT CCT GCT-G TATTT-A TGT TG- T--- T---- ATA -TC T-G CG---TGTTT---ATTCTTGTGTC
M-80048(K) -AT CTC- CTGT CCT GCT-G TATTT-A TGT TG- T--- T---- ATA -TC T-G CG---TGTTT---ATTCTTGTGTC
(Japan & NZ) FB-30961(CBM) -AT CTC- CTGT CCT GCT-G TATTT-A TGT TG- T--- T---- ATA -TC T-G CGCCCTGTTT---ATTCCTGTGCC
FB-30962(CBM) -AT CTC- CTGT CCT GCT-G TATTT-A TGT TG- T--- T---- ATA -TC T-G CGCCCTGTTT---ATTCCTGTGCC
FB-30963(CBM) -AT CTC- CTGT CCT GCT-G TATTT-A TGT TG- T--- T---- ATA -TC T-G CGCCCTGTTT---ATTCCTGTGCC
FB-30976(CBM) -AT CTC- CTGT CCT GCT-G TATTT-A TGT TG- T--- T---- ATA -TC T-G CGCCCTGTTT---ATTCCTGTGCC
FB-30977(CBM) -AT CTC- CTGT CCT GCT-G TATTT-A TGT TG- T--- T---- ATA -TC T-G CGCCCTGTTT---ATTCCTGTGCC
FB-30985(CBM) -AT CTC- CTGT CCT GCT-G TATTT-A TGT TG- T--- T---- ATA -TC T-G CGCCCTGTTT---ATTCCTGTGCC
FB-30987(CBM) -AT CTC- CTGT CCT GCT-G TATTT-A TGT TG- T--- T---- ATA -TC T-G CGCCCTGTTT---ATTCCTGTGCC
M-II 506(O) -GT CCC- CTGT CTT GTT-G TCTTTTA TGT TA- T--- T---- AGA -GC TTG CGCCCTGTTT---ATTCTT--GTC
(Norway) 1539(O) -GT CCC- CTGT CTT GTT-G TCTTTTA TGT TA- T--- T---- AGA -GC TTG CGCCCTGTTT---ATTCTT--GTC
4220(O) -GT CCC- CTGT CTT GTT-G TCTTTTA TGT TA- T--- T---- AGA -GC TTG CGCCCTGTTT---ATTCTT--GTC
(Japan) FB-30978(CBM) -GT CCC- CTGT CTT GTT-G TCTTTTA TGT TA- T--- T---- AGA -GC TTG CGCCCTGTTT---ATTCTT--GTC
FB-30981(CBM) -GT CCC- CTGT CTT GTT-G TCTTTTA TGT TA- T--- T---- AGA -GC TTG CGCCCTGTTT---ATTCTT--GTC
FB-30982(CBM) -GT CCC- CTGT CTT GTT-G TCTTTTA TGT TA- T--- T---- AGA -GC TTG CGCCCTGTTT---ATTCTT--GTC
M-III 45843(NY) -GT CCT- CCGT CCT GTT-G TCTTT-A TCT TA- TGT- T---- AGA -GC T-G C---------------------TC
(Eastern USA) 45785(NY) -GT CCT- CCGT CCT GTT-G TCTTT-A TCT TA- TGT- T---- AGA -GC T-G C---------------------TC
45820(NY) -GT CCT- CCGT CCT GTT-G TCTTT-A TCT TA- TGT- T---- AGA -GC T-G C---------------------TC
45840(NY) -GT CCT- CCGT CCT GTT-G TCTTT-A TCT TA- TGT- T---- AGA -GC T-G C---------------------TC
45883(NY) -GT CCT- CCGT CCT GTT-G TCTTT-A TCT TA- TGT- T---- AGA -GC T-G C---------------------TC
(Western USA) 45863(NY) -GT CCT- CCGT CCT GTTTG TCTTT-A TCT TA- TGT- TTAT- AGA -GC T-G C---------------------TC
HDT49100(SFSU) -GT CCT- CCAT CCT GTTTG TCTTT-A TCT TA- TGT- TTAT- AGA -GC T-G C---------------------TC
HDT44761(SFSU) -GT CCT- CCGT CCT GTTTG TCTTT-A TCT TA- TGT- TTAT- AGA -GC T-G C---------------------TC
HDT45060(SFSU) -GT CCT- CCGT CCT GTTTG TCTTT-A TCT TA- TGT- TTAT- AGA -GC T-G C---------------------TC

892
 T. Oda, C. Tanaka and M. Tsuda
Table 3. Variable sites in the alignment of the ITS and b-tubulin regions among groups of Amanita pantherina. The position numbers of the nucleotides are given (read from top to bottom).

893
Molecular phylogeny and biogeography of Amanita 894

The clade M-I consisted of specimens from Eurasia, Amanita pantherina

with the exception of one specimen from New Zealand.
Because all specimens in clade P-II were collected in
In New Zealand, A. muscaria has been introduced with
Eurasia, we termed this clade the ‘ Eurasian group ’. In
exotic trees from other countries of the Northern
contrast, clade P-I consisted of specimens from the
Hemisphere (Ridley 1991). Therefore, it is reasonable
western USA (California). Here, it is necessary to
to assume that this specimen from New Zealand was
mention a specimen of Amanita pantherina var. multi-
introduced from Japan or neighbouring areas ; clade
squamosa from eastern North America (Massachusetts)
M-I was therefore termed the ‘ Eurasian group ’. Clade
36442 (NY) (not listed in Table 1) ; we were unable to
M-II consisted of the subclade of A. muscaria var.
determine the sequence of the b-tubulin region of this
regalis from Norway and the subclade of A. muscaria
specimen, although we did obtain a sequence of the ITS
from Japan. Amanita muscaria var. regalis is a rare
region (accession no. AB103329 in DDBJ/EMBL/
taxon found mainly in the coniferous forests on the
GenBank). Based on the ITS region, it appears that
high mountains of northern and central Europe. This
this specimen is closer to specimens of clade P-I than
fungus differs from other varieties of A. muscaria in
to those of clade P-II. Therefore, we assumed that clade
that it bears a brownish pileus and yellowish volva ; it is
P-I probably predominates in both western and eastern
often considered a distinct species, A. regalis. Because
North America, and it is reasonable to treat clade P-I
the three specimens of A. muscaria var. regalis from
as the ‘North American group ’, rather than the ‘west-
Norway displayed the above morphological character-
ern North American group ’.
istics and were collected in Picea and Betula forests at
high altitudes, these specimens were considered to be
typical of A. muscaria var. regalis. In contrast, each of
Biogeographical implications
the three specimens of A. muscaria from Japan had a
red pileus and a white volva, which was not distinct Our molecular phylogenetic analyses suggested that
from the common Japanese A. muscaria collections there are three groups of Amanita muscaria (the Eur-
belonging to clade M-I. However, these three speci- asian, Eurasian subalpine, and North American), and
mens were found in subalpine Abies veitchii, A. mariesii, two of A. pantherina (Eurasian and North American).
and Betula ermanii forests in Chino-shi, Nagano However the area sampled for this study represents
Prefecture, at about 1900 m in the Yatsugatake moun- only part of the distribution of these two species. In
tain range, whereas specimens from clade M-I (Eur- particular, specimens from the Southern Hemisphere
asian group) were collected in Abies, Larix, Picea, have remained largely unexamined, except for A. mus-
Pinus, Betula, Fagus, and Quercus forests below the caria from New Zealand. Therefore, while further
subalpine zone. Although the habitats of clades M-I examination might result in a revised grouping for each
and M-II may overlap, clade M-II might be better species, here we consider the biogeographical impli-
adapted to the subalpine zone. Therefore, we conclude cations of our results.
that (1) A. muscaria var. regalis (or A. regalis) is a To better understand the biogeographical history of
group of A. muscaria rather than a distinct species; (2) the two species, it is necessary to recount the biogeo-
this fungus is not defined by colour characteristics, i.e. graphical history of the flora of the Northern Hemi-
a brownish pileus and yellowish volva ; (3) clade M-II sphere, because both species are ectomycorrhizal
is an Eurasian subalpine group of A. muscaria and is associated with host trees (mainly Pinaceae, Fagaceae,
defined by habitat conditions. Jenkins (1986) reported and Betulaceae). The floral biogeography of the
that A. muscaria var. regalis occurred in Alaska (USA). Northern Hemisphere has been well-investigated (e.g.
Therefore, it will be necessary to examine A. muscaria Gray 1846, Chaney 1947, Raven & Axelrod 1974,
var. regalis from other localities to determine whether Wolfe 1975, Tiffney 1985a, b, Graham 1993, Manche-
this group occurs only in Eurasia or in both Eurasia ster 1999, Wen 1999). Although the explanations for
and North America. the history of floral changes are controversial, it is
Our analysis suggests that at least three groups of generally accepted that important factors include the
A. muscaria exist in the world. At this point, we floristic migration between Eurasia and North America
consider these groups to be the Eurasian group, the via the Bering and North Atlantic land bridges during
Eurasian subalpine group, and the North American the late Cretaceous and the Tertiary Periods, and cli-
group, corresponding to geographical differences. In matic changes during the late Tertiary and Quaternary
addition, it is necessary to mention specimen FB-30986 Periods. In addition, based on a molecular phylogeny
(CBM); this collection looked like A. muscaria, but of rDNA sequences, Berbee & Taylor (1993) estimated
differed in the yellow stipe. In Japan, A. muscaria that ectomycorrhizal fungi existed during the early
commonly appears under Pinaceae and/or Betula Cretaceous. Considering these data and the molecular
species, but FB-30986 (CBM) was collected in a Fagus phylogeny in this study, we hypothesize that the fol-
crenata forest. Here, we termed it Amanita aff. lowing occurred during the late Cretaceous and the
muscaria, because we were unable to determine if it Tertiary. Initially, the ancestral group of A. muscaria
was a distinct species or if it belonged to a group of existed only in Eurasia. Following climate cooling
A. muscaria. during the Tertiary, it diversified into one group that
T. Oda, C. Tanaka and M. Tsuda 895

continued living under moderate climatic conditions in Fukiharu (Natural History Museum and Institute, Chiba), Yoko
Eurasia (Eurasian group), and another adapted to Ando, Masaki Endo, Kyoko Isoda, Susumu Ito, Keiko Kudo,
Shinichi Kudo, and Youkin-no-kai. We are also grateful to the
cooler climates. After the climate warmed again, the following institutions: The Natural History Museum and Institute,
latter diversified into the group that invaded subalpine Chiba; Herbarium of Cryptogams, Kunming Institute of Botany,
regions of Eurasia (Eurasian subalpine group) as a Academia Sinica, Kunming; Royal Botanic Gardens, Kew; New
relic, and another group that transferred to North York Botanical Garden, Bronx, New York; Botanical Museum,
America via the land bridges (North American group). Oslo, Harry D. Thiers Herbarium, San Francisco State University,
San Francisco; and Botany Department, National Science Museum,
A. pantherina is divided into at least two groups, the Tsukuba. For helping with the collection of specimens from Nepal,
North American group and the Eurasian group. Based we are grateful to Mahesh K. Adhikari (National Herbarium and
on its topology, it is impossible to determine whether A. Plant Laboratory, Department of Plant Resources, Ministry of
pantherina moved from Eurasia to North America or Forest and Soil Conservation, Kathmandu, Nepal), who guided
vice versa. Based on accounts by Tiffney (1985a, b), us around the forests of Nepal, and Yoshimichi Doi (Department of
Botany, National Science Museum, Tsukuba, Japan), who arranged
during the early Eocene to Miocene, two species could the research trip to Nepal. We also thank Naoki Taniguchi (Kyoto
have migrated with their host trees to each continent Research Center for Food Hygiene and Technology), who provided
via the North Atlantic land bridge and/or the Bering support for the DNA sequence analyses.
land bridge. The genetic variations in and between
each group of A. pantherina are greater than those of
A. muscaria (Fig. 3). Therefore, we assume that the REFERENCES
diversification and disjunction of A. muscaria are more
recent events than those of A. pantherina. Adhikari, M. K. & Parajuli, P. (1994) The genus Amanita in Nepal.
Banko Janakari 4 : 130–134.
With regard to the flora of the Northern Hemi-
Berbee, M. L. & Taylor, J. W. (1993) Dating the evolutionary radi-
sphere, the disjunction described by Gray (1846) is ations of the true fungi. Canadian Journal of Botany 71: 1114–1127.
noteworthy. This author made detailed comparisons Chaney, R. W. (1947) Tertiary centers and migration routes. Eco-
among the North Temperate flora, and concluded that logical Monographs 17: 139–148.
the flora of eastern North America and eastern Asia Drehmel, D., Moncalvo, J. M. & Vilgalys, R. (1999) Molecular
phylogeny of Amanita based on large-submit ribosomal DNA
were more similar to each other than to the flora of
sequences: implications for taxonomy and character evolution.
western North America or Europe. In terms of macro- Mycologia 91 : 610–618.
fungi, Hongo & Yokoyama (1978) suggested a close Gilbert, E. J. (1941) Amanitaceae. Iconographia Mycologica 27
affinity of mycobiotas of eastern Asia and eastern (Suppl. 1): 201–427.
North America, corresponding to Gray’s theory. González, V., Arenal, F., Platas, G., Esteve-Raventós, F. & Peláez, F.
(2002) Molecular typing of Spanish species Amanita by restriction
Moreover, Hongo & Yokoyama (1978) stated that
analysis of the ITS region of the DNA. Mycological Research 106:
specific differentiation in the fungi had not advanced as 903–910.
much in the vascular plants, which had differentiated at Graham, A. (1993) History of the vegetation: Cretaceous (Maas-
the specific or even generic levels. Wu & Mueller (1997) trichtian) – Tertiary. In Flora of North America and North Mexico
reached the same conclusions based on a literature and (Flora of North America Editional Committee, ed.) 1: 57–70.
Oxford University Press, New York.
herbarium survey. However, more recent molecular
Gray, A. (1846) Analogy between the flora of Japan and that of
phylogenetic studies suggest that the biogeographical the United States. American Journal of Science and Arts, ser. 2
relationships of macrofungi between eastern North 2: 135–136.
America and eastern Asia may not be as close as in- Hibbett, D. S. (2001) Shiitake mushrooms and molecular clocks:
dicated by morphological data, whereas the molecular historical biogeography of Lentinula. Journal of Biogeography 28:
231–241.
data do not refute the hypothesis that a relatively close
Hongo, T. & Yokoyama, K. (1978) Mycofloristic ties of Japan to the
relationship exists between them (Mueller et al. 2001). continents. Memoirs of Shiga University 28: 76–80.
In this study, we found that in A. muscaria and A. Imazeki, R. & Hongo, T. (1987) Colored illustrations of mushrooms of
pantherina, the intracontinental relationships of both Japan. Vol. 1. Hoikusha, Osaka.
Eurasia and North America were closer than the re- Jenkins, D. T. (1977) A taxonomic and nomenclatural study of the
genus Amanita section Amanita for North America. Bibliotheca
lationships between eastern Asia and eastern North
Mycologica 57: 1–126.
America. Additional molecular phylogenetic data of Jenkins, D. T. (1986) Amanita of North America. Mad River Press,
other Amanita species and specimens from diverse Eureka.
localities will provide an even better understanding of Ko, K. S., Lim, Y. W., Kim, Y. H. & Jung, H. S. (2001) Phylogeo-
the biogeographical and evolutionary history of the graphic divergences of nuclear ITS sequences in Coprinus species
sensu lato. Mycological Research 105: 1519–1526.
genus.
Manchester, S. R. (1999) Biogeographical relationships of North
American Tertiary floras. Annals of the Missouri Botanical Garden
86: 472–522.
ACKNOWLEDGEMENTS Mueller, G. M., Wu, Q. X., Huang, Y. Q., Guo, S. Y., Aldana-
Gomez, R. & Vilgalys, R. (2001) Assessing biogeographic relation-
This work was supported in part by Hokuto Foundation, and by a ships between North American and Chinese macrofungi. Journal of
Grant-in-Aid (15380104) for Scientific Research (B) from the Japan Biogeography 28 : 271–281.
Society of the Promotion of Science. For supplying specimens, we Nakada, M., Tanaka, C., Tsunewaki, K. & Tsuda, M. (1994)
thank Roman Kotlowski (Department of Food Chemistry and RFLP analysis for species separation in the genera Bipolaris and
Technology, Gdansk University of Technology, Gdansk), Toshimitsu Curvularia. Mycoscience 35 : 271–278.
Molecular phylogeny and biogeography of Amanita 896

Oda, T., Tanaka, C. & Tsuda, M. (1999) Molecular phylogeny of van Nues, R. W., Rientjes, J. M. J., van der Sande, C. A. F. M., Zerp,
Japanese Amanita species based on nucleotide sequences of the S. F., Sluiter, C., Venema, J., Planta, R. J. & Raué, H. A. (1994)
internal transcribed spacer region of nuclear ribosomal DNA. Separate structural elements within internal transcribed spacer 1
Mycoscience 40: 57–64. of Saccharomyces cerevisiae precursor ribosomal RNA direct the
Oda, T., Yamazaki, T., Tanaka, C., Terashita, T., Taniguchi, N. & formation of 17S and 26S rRNA. Nucleic Acids Research 22:
Tsuda, M. (2002) Amanita ibotengutake sp. nov., a poisonous 912–919.
fungus from Japan. Mycological Progress 1: 355–365. Vilgalys, R. & Sun, B. L. (1994) Ancient and recent patterns of geo-
Pérez-Silva, E. & Herrera, T. (1991) Iconografı´a de macromicetos de graphic speciation in the oyster mushroom Pleurotus revealed by
México I Amanita. [Publicacioñes Especiales del Instituto de Bio- phylogenetic analysis of ribosomal DNA sequences. Proceedings of
logı́a No. 6.] Universidad Nacional Autónoma de México, México. the National Academy of Sciences, USA 91: 4599–4603.
Raven, P. H. & Axelrod, D. I. (1974) Angiosperm biogeography and Weiß, M., Yang, Z. L. & Oberwinkler, F. (1998) Molecular phylo-
past continental movements. Annals of the Missouri Botanical genetic studies in the genus Amanita. Canadian Journal of Botany
Garden 61 : 539–673. 76: 1170–1179.
Reid, D. A. (1980) A monograph of the Australian species of Amanita Wen, J. (1999) Evolution of eastern Asian and eastern North America
Pers. ex Hook. (Fungi). Australian Journal of Botany, suppl. ser. 8: disjunct distributions in flowering plants. Annual Review of Ecology
1–96. and Systematics 30: 421–455.
Reid, D. A. & Eicker, A. (1991) South African fungi: the genus White, T. J., Bruns, T., Lee, S. & Taylor, J. (1990) Amplification
Amanita. Mycological Research 95: 80–95. and direct sequencing of fungal ribosomal RNA genes for phylo-
Ridley, G. S. (1991) The New Zealand species of Amanita (Fungi: genetics. In PCR Protocols: a guide to methods and applications
Agaricales). Australian Systematic Botany 4: 325–354. (M. A. Innis, D. H. Gelfand, J. J. Sninsky & T. J. White, eds):
Swofford, D. L. (2002) PAUP*: phylogenetic analysis using parsi- 315–322. Academic Press, San Diego.
mony (and other methods). Version 4.0b9. Sinauer Associates, Wolfe, J. A. (1975) Some aspects of plant geography of the Northern
Sunderland, MA. Hemisphere during the late Cretaceous and Tertiary. Annals of the
Thompson, J. D., Higgins, D. G. & Gibson, T. J. (1994) CLUSTAL Missouri Botanical Garden 62: 264–279.
W: improving the sensitivity of progressive multiple sequence Wu, Q. X. & Mueller, G. M. (1997) Biogeographic relationships
alignment through sequence weighting, position-specific gap between the macrofungi of temperate eastern Asia and eastern
penalties and weight matrix choice. Nucleic Acids Research 22: North America. Canadian Journal of Botany 75: 2108–2116.
4673–4680. Wu, Q. X., Mueller, G. M., Lutzoni, F. M., Huang, Y. Q & Guo,
Tiffney, B. H. (1985a) Perspectives on the origin of the floristic simi- S. Y. (2000) Phylogenetic and biogeographic relationships of
larity between eastern Asia and eastern North America. Journal of Eastern Asian and Eastern North American disjunct Suillus species
the Arnold Arboretum 66: 73–94. (fungi) as inferred from nuclear ribosomal RNA ITS sequences.
Tiffney, B. H. (1985b) The Eocene North Atlantic land bridge: its Molecular Phylogenetics and Evolution 17: 37–47.
importance in Tertiary and modern phytogeography of the Yang, Z. L. (1997) Die Amanita-Arten von Südwestchina. Bibliotheca
Northern Hemisphere. Journal of the Arnold Arboretum 66: Mycologica 170: 1–240.
243–273.
Tulloss, R. E., Ovrebo, C. L. & Halling, R. E. (1992) Studies on
Amanita (Amanitaceae) from Andean Colombia. Memoirs of the
New York Botanical Garden 66 : 1–46. Corresponding Editor: D. S. Hibbett