Mushroom Biology and Mushroom Products. Sánchez et al. (eds). 2002
UAEM. ISBN 968-878-105-3
MOLECULAR PHYLOGENETIC ANALYSIS OF GRIFOLA FRONDOSA (MAITAKE)
REVEALS CRYPTIC NORTH AMERICAN AND ASIAN SPECIES
Q. Shen, D. M. Geiser and D. J. Royse
Department of Plant Pathology, The Pennsylvania State University, University Park, PA 16802
<djr4@psu.edu>
ABSTRACT
A phylogenetic analysis was performed on 51 isolates of the commercially valuable Grifola
frondosa (maitake) using sequences from the Internal Transcribed Spacers and 5.8S region of the
nuclear ribosomal DNA (rDNA) transcriptional unit and a portion of the -tubulin gene. The betatubulin gene provided more than twice as much phylogenetic information as the ITS/5.8S regions.
The isolates analyzed comprised 21 from North America, 27 from Asia, one from Europe, and two
of unknown geographic origin, one of which was the major US commercial production strain in use.
G. sordulenta was used as an outgroup. Combined and separate analysis of both genes showed a
partition separating Asian versus North American isolates. Bootstrap analysis showed strong
support for these clades in the beta-tubulin data alone and in the combined data. The major
commercial isolate of unknown geographic origins apparently of Asian decent based on grouping
within the Asian clade. The single European isolate analyzed was distinct from both the North
American and Asian clades. These results indicate strong support for a cryptic species partition
separating North American and Asian isolates of G. frondosa, despite previous studies indicating no
morphological distinction between them.
INTRODUCTION
Grifola frondosa (Dickson: Fr.) S.F.Gray is a white rot fungus widely distributed in Asia, North
America and Europe. It occurs on a variety of hardwoods, particularly oak and chestnut.
Commonly called maitake or hen-of-the-woods, it is considered a choice edible mushroom with
unique culinary and medicinal qualities. Maitake is marketed throughout Asia and, because of
increased consumer demand, its commercial production has grown dramatically in Asia and the
United States (Royse 1997, Chang 1999).
Traditional classification of G. frondosa was based solely on morphological characters. Grifola
frondosa was first named Boletus frondosus by Dickson (1785) and later, Fries (1821) changed the
name to Polyporus frondosus Dicks.: Fr. Even now, P. frondosus, the synonym of G. frondosa, is
widely used. The genus Grifola S.F.Gray was first applied by Gray (1821) and described as a
polypore with large compound basidiomes. Previous taxonomic investigations by Gilbertson and
Ryvarden (1986) and Zhao and Zhang (1992) described similar morphological characters shared
between North American and Asian isolates, although these studies separately analyzed North
American and mainland China isolates, respectively. Both studies recognized G. frondosa
(Dicks.:Fr.) S.F.Gray as the only species in the genus Grifola. However, another Grifola species,
G. sordulenta, was identified by Singer in 1969.
Biogeographic phylogenetic structure is known to exist in various polypores and agarics, such as
Pleurotus spp (Vilgalys and Sun 1994), Lentinula spp (Thon and Royse 1999, Hibbett et al. 1998).
A better understanding of the biogeographic phylogenetic structure of G. frondosa may facilitate
genetic selection and improvement of cultivated isolates in commercial mushroom production.
Molecular approaches have been used extensively for examining phylogenetic relationships in other
73
edible fungi (for example see Hibbett et al. 1995, Thon and Royse 1999 and Vilgalys and Sun
1994). Recent studies by Hibbett et al. (2000) based on mitochondrial and nuclear small and large
subunit ribosomal RNA sequences showed that G. frondosa is related to other polypores including
Laetiporus and Ganoderma.
To analyze biogeographic structure within G. frondosa, partial regions of rDNA and -tubulin
genes were analyzed and phylogenetic relationships were inferred among isolates from North
America and Asia. Results showed a strongly supported partition separating isolates from North
America and Asia, suggesting that these represent separate species.
MATERIALS AND METHODS
Cultures. A total of 51 isolates of G. frondosa and one isolate of G. sordulenta were used in this
study (Table 1). The isolates of G. frondosa included all available isolates from the Pennsylvania
State University Mushroom Culture Collection (PSUMCC) and the American Type Culture
Collection (ATCC), as well as an isolate of G. sordulenta (Argentina) used as an outgroup. Isolates
represented various geographic origins including Asia (27), United States (21), Europe (1), and
unknown origin (2). All cultures were maintained on potato dextrose agar supplemented with
1.5g/L of yeast extract (PDYA).
Table 1. List of species, isolate code, source, geographic origin, substrate and locality of Grifola
frondosa used for this study.
Source a
Geographic origin
Host/Substrate
Locality
G. frondosa
Isolate
code
WC248
L.C. Schisler
PSU, PA
N/A b
N/A
G. frondosa
G. frondosa
G. frondosa
G. frondosa
G. frondosa
G. frondosa
G. frondosa
G. frondosa
G. frondosa
G. frondosa
G. frondosa
G. frondosa
G. frondosa
G. frondosa
G. frondosa
WC364
WC367
WC483
WC493
WC555
WC556
WC557
WC581
WC582
WC583
WC659
WC685
WC808
WC828
WC834
L.C. Schisler
Jodon
ATCC 11936
ATCC 48141
Y.H.Park
Y.H.Park
Y.H.Park
Y.H.Park
Y.H.Park
Y.H.Park
Y.H.Park
B.W.Yoo
Bill Shanley
D.J.Royse
NGF 001
N/A
N/A
Oak stump
Quercus robur
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
White Oak
Commercial isolate
Castanopsis spp.
N/A
Hort. Woods
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
Lowlands
G. frondosa
WC835
Hokken M-1
PSU, PA
PSU, PA
Maryland
Norway
Korea
Korea
Korea
Korea
Korea
Korea
Korea
N/A
Tidioute, PA
N/A
Nara Prefecture,
Japan
Japan
Highlands
G. frondosa
WC836
Mori 51
Japan
G. frondosa
M001
USDA FP-101988-T
Cooksville (Rock),
WI
G. frondosa
M002
USDA FP-102464-Sp
Madison, WI
G. frondosa
M003
USDA FP-102464-T
Madison(Dane), WI
G. frondosa
M004
USDA FP-103424-T
Athens(Georgia),
GA
Commercial isolate
on oak
Commercial isolate
on oak
Soil near downed
hardwood log
(Quercus?)
Quercus, soil at base
of dead stump.
Quercus, soil at base
of dead stump.
Quercus nigra, at
base
Species
74
Highlands
Edge of Old Mill
Pond at Hwy 59,
Fire #638
N/A
Picnic Point, UW
campus
N/A
Source a
Geographic origin
Host/Substrate
Locality
G. frondosa
Isolate
code
M005
USDA FP-105867-Sp
Beltsville (Prince
George), MD
G. frondosa
G. frondosa
M006
M007
USDA FP-134675-Sp
USDA FP-47462
Madison (Dane), WI
WV
Quercus coccinea
(scarlet oak) living,
at base
Quercus, underneath
Quercus alba (white
oak)
G. frondosa
G. frondosa
M008
M009
USDA RLG-6889-Sp
USDA LOO-14980-T
Syracuse, NY
LA
Forest Disease Lab
Station, Agr. Res.
Center
UW Arboretum
Plat #55, at 1500
feet elevation,
Devil's Hole
N/A
N/A
G. frondosa
M010
USDA OKM-4954-T
G. frondosa
M011
G. frondosa
M012
USDA OKM-6133Sp
USDA RLG-14995-T
Beltsville (Prince
George), MD
Washington(District
of Columbia), DC
Baton Rouge, LA
G. frondosa
G. frondosa
G. frondosa
M013
M014
M015
USDA L-15552-Sp
USDA RLG-6889-T
USDA TJV-93-130-T
Syracuse, NY
Syracuse, NY
Madison(Dane), WI
G. frondosa
G. frondosa
G. frondosa
G. frondosa
G. frondosa
G. frondosa
G. frondosa
G. frondosa
G. frondosa
G. frondosa
G. frondosa
G. frondosa
G. frondosa
G. frondosa
G. frondosa
G. frondosa
G. frondosa
M016
M017
M018
M019
M020
M021
M029
M030
M031
M032
M033
M034
M035
M036
M037
M038
M039
FIRDI 36283
FIRDI 36286
FIRDI 36355
FIRDI 36356
FIRDI 36357
FIRDI 36434
PSUMCC 600
PSUMCC 601
PSUMCC 602
PSUMCC 604
PSUMCC 630
PSUMCC 644
USDA RLG-6889-Sp
X.W.Chen
X. W.Chen
ATCC 60891
Tan 0206
Taiwan
Taiwan
Taiwan
Taiwan
Taiwan
Taiwan
Taiwan
Taiwan
Taiwan
Taiwan
Taiwan
Taiwan
Syracuse, NY
China
China
China
He Bei, China
Species
Quercus alba
Quercus snag, inside
of hollow (oak)
N/A
N/A
Quercus virginiana
N/A
Quercus alba
Quercus
macrocarpa, base of
live
N/A
N/A
N/A
N/A
N/A
N/A
Commercial strain
Commercial strain
Commercial strain
Commercial strain
Commercial strain
Commercial strain
Quercus alba
N/A
N/A
N/A
N/A
Ground, Beltsville
Expt Forest
Rock Creek Park
Memorial Grove,
LA State U campus
N/A
Oakwood Cemetery
Turville Pt. Woods
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
Oakwood Cemetery
N/A
N/A
N/A
N/A
G. frondosa
M40
M. Chen
China
Commercial strain
N/A
G.
G01
ATCC 200416
Argentina
Nothofagus dombeyi
N/A
sordulenta
trunk
a ATCC = American Type Culture Collection; USDA = The Unitied States Department of Agriculture; FIRDI = Food
Industry Research and Development Institute, Taiwan; PSUMCC = Pennsylvania State University Mushroom Culture
Collection.
b N/A = not available.
DNA extraction. Cultures were grown in 50 ml of potato dextrose yeast broth (PDYB) for 20 to 30
days at room temperature. Mycelium was harvested by vacuum filtration on Whatman (grade #1)
filter paper, and washed once with distilled water. Fresh mycelium (100 mg) was used to isolate
DNA following the LETS extraction procedure (Chen et al. 1999). DNA preparations were diluted
with sterile water and used as template for PCR amplification.
PCR amplification and sequencing. PCR was performed in 25 µl reactions with a 96-well PCR
cycler (PTC-100 Programmable Thermal Controller, MJ Research, Inc.), using 10ng DNA template,
one U of Taq DNA polymerase (Promega, Madison, WI), 0.2 mM of each dNTP, 2 mM MgCl 2,
0.1% Triton, and 0.5 µM of each primer. Amplification of ITS-1, ITS-2, and 5.8 S rDNA was
performed by utilizing primers ITS1AF (5'-TCCGTAGGTGAACCTGCGG-3') (White et al. 1990)
75
and ALR0 (5'-CATATGCTTAAGTTCAGCGGG-3') (RJ Vilgalys, personal communication). PCR
reactions for ITS regions were performed using the following parameters: 94C/1 min; 35 cycles of
94C/15 s, 60C/30 s, 72C/1 min; and 72C/5 min. PCR Reactions for -tubulin regions were
performed with primer BTG5F (5'-CGTTGTGCCCAGTCCTAAGGTG-3') and BTG8R (5'GTTCTTGCTCTGCACGTTCTG-3') (Figure 1) with the following parameters: 94C/2 min; 35
cycles of 94C/15 s, 57C/30 s, 72C/1 min; and 72C/7 min. Amplification products were
electrophoresed on a 1.0% agarose gel and checked to ensure that a single DNA band was produced
of the expected size (~600bp for ITS PCR products and ~680bp for -tubulin PCR products). For
sequencing, the PCR products were purified directly from reactions using the Wizard PCR Preps
System (Promega Corp., Madison, WI) and the concentration adjusted to 20ng/l. Sequencing
reactions were performed using an ABI dye-terminator kit (ABI/Perkin-Elmer) and analyzed using
an ABI Prism® Model 377 automated sequencing system (Applied Biosystems, Foster City, CA).
BTG5F

5
6
7
8
 BTG8R
Figure 1. Locations of primers used for PCR-amplification of the -tubulin gene in
Grifola frondosa and G. sordulenta (G01). Numbers indicate exons.
Sequence data analysis. Sequence ends were trimmed using the SeqMan II module in the
Lasergene package (DNAStar, Inc. Madison, WI) and adjusted manually. All sequences then were
edited and initially aligned using the clustal W algorithm (Higgins et al. 1991) in the Lasergene
package (DNAStar, Inc. Madison, WI). Multiple alignment parameters used were gap penalty = 10
and gap length penalty = 10. Final alignments then were optimized visually. Intron/exon junctions
in the beta-tubulin sequences were inferred based on comparisons with the known Schizophyllum
commune sequence (Russo et al. 1992). Levels of molecular sequence divergence in each of the
data sets were compared by calculating pairwise estimates of nucleotide substitution rates by the
two-parameter method of Kimura. Phylogenetic analyses were completed using PAUP Version
4.0b4a (Swofford 2000). A neighbor-joining (NJ) tree was constructed using the Kimura 2parameter model. The stability of clades was evaluated by bootstrap tests with 1000 replications
(Felsenstein 1985, Hills and Bull 1993). A maximum parsimony (MP) analysis was performed
using heuristic searches with 1000 random addition searches. Other indices for the generated
topology, including tree length, consistency index (CI), and retention index (RI) were calculated. A
strict consensus of the minimum length MP trees was calculated. Gaps were considered missing
data for all analyses.
RESULTS
Analysis of rDNA ITS sequences
Amplification of the ITS-1, ITS-2 and 5.8S ribosomal DNA repeat yielded fragments of
approximately 600bp as estimated by agarose gel electrophoresis. Characteristics of nucleotide
variation present in these regions of G.frondosa and its allies are summarized in Table 2. Nucleotide
variation among isolates of G. frondosa was 5.4% in the rDNA region, and 14.3% between G.
frondosa and the G. sordulenta outgroup.
76
Table 2. Site variation within the ITS-1, 5.8 S, and ITS-2 gene region of Grifola frondosa and
Grifola sordulenta.
ITS-1
Isolates
Number a variable sites Total sites
G. frondosa
51
7
199
G. frondosa +
52
30
199
G. sordulenta. e
a
Number of isolates included.
5.8S
Sites Total
0
158
0
158
ITS-2
Sites Total
24
217
52
217
Total
Sites
Total
31
574
82
574
A Neighbor-Joining analysis based on rDNA ITS sequences identified two clades within G.
frondosa. The North American clade included all of the U.S. isolates, while the Asian clade
consisted of Asian isolates. The North American clade received 83% bootstrap support, but the
Asian clade did not receive >50% support. The single European isolate (WC493) fell on a branch
basal to the North American clade, but its connection to this clade was not strongly supported
(67%). Two isolates of unknown geographic origin (WC685 and WC828, the major US
commercial isolate) were placed within the Asian clade. The maximum parsimony (MP) analysis
produced 424 equally parsimonious trees (length = 113 steps, consistency index=0.796, retention
index=0.911), which were similar in topology to the neighbor-joining tree.
Analysis of -tubulin sequences
The characteristics of nucleotide variation present in partial -tubulin sequence regions of G.
frondosa and G. sordulenta are summarized in Table 3. Among isolates of G. frondosa, nucleotide
variation was 12.2% for all sites. Most of the variation occurred in introns with 14.8% for intron 5,
23.7% for intron 6, and 32.3% for intron 7. Less variation was observed in exons with 8.3% for
exon 6 and 5.5% for exon 7. The sequences of G. frondosa and G. sordulenta showed much more
variation over the total alignment (30.2%) and across the introns and exons. The most variable
region was intron 5 (56.9%) and exon 7 was the most conserved region (18.3%) within the tubulin gene sequence regions analyzed. No amino acid changes were inferred among isolates of G.
frondosa, but two amino acid changes (Glutamate and Isoleucine in G. frondosa, but Glutamine and
Valine in G. sordulenta, both in exon 6) were observed between isolates of G. frondosa and G.
sordulenta. Most nucleotide substitutions in exons were observed in the third codon positions.
Table 3. Site variation within the partial -tubulin gene region of Grifola frondosa and
Grifola sordulenta.
Exon 5
Isolates
G. frondosa
G. frondosa +
G. sordulenta
number
51
52
a
variable sites
0
0
Intron 5
Total sites
4
4
Sites
8
33
Exon 7
Isolates
number
G. frondosa
51
G. frondosa +
52
G. sordulenta
a Number of isolates included.
Sites
9
30
Total
54
58
Intron 7
Sites
Total
20
62
33
62
Total
164
164
Exon 6
Sites
19
45
Total
229
229
Exon 8
Sites
Total
1
8
1
8
Intron 6
Sites
14
35
Total
59
62
Total
Sites Total
71
580
177
587
A Neighbor-Joining analysis based on partial -tubulin gene sequences gave similar results to those
based on the rDNA dataset, showing two distinct North American and Asian clades within G.
frondosa. The North American clade received 99% bootstrap support, whereas the Asian clade
received 74% support. In contrast to the rDNA tree, where the European isolate grouped basal to
77
the North American clade, the beta-tubulin tree placed the European isolate (WC493) on a branch
basal to the American clade, with 95% bootstrap support. As in the rDNA analysis, the two isolates
of unknown geographic origin grouped with the Asian isolates. The maximum parsimony (MP)
analysis produced 48,100 trees (length=251 steps, consistency index=0.793, retention index=0.932)
until PAUP was aborted because of lack of memory. The strict consensus of these trees was similar
in topology to the neighbor-joining tree. A phylogenetic analysis without intron 5 was also
performed because of the high variation in this region. No appreciable difference was found (data
not shown).
Analysis of combined rDNA ITS and partial -tubulin gene sequence data
A Neighbor-Joining analysis (Figure 2) based on combined rDNA ITS and partial -tubulin gene
sequence data supported most of the results produced by rDNA and -tubulin separately, with much
higher bootstrap support. The North American and Asian clades were strongly supported by high
bootstrap values (100% and 98%, respectively). The European isolate (WC493) was grouped basal
to the Asian clade with 89% bootstrap support, and agrees with the results from the -tubulin data
alone. The two isolates of unknown geographic origin (WC828 and WC685) grouped within the
Asian clade. The maximum parsimony (MP) analysis, based on combined dataset produced 405
equally parsimonious trees. The strict consensus tree (not shown, length=382 steps, consistency
index=0.741, retention index=0.901) retained a similar topology, for the most part, to the NJ trees.
DISCUSSION
The 51 isolates of G. frondosa analyzed in this study clearly clustered into two clades based on
North American and Asian origins in all analyses of rDNA, -tubulin, and combined sequences.
Six simplified trees were shown in Figure 3 (A-F). This result conflicts with previous taxonomic
studies indicating no appreciable morphological differences between North American and Asian
isolates (Gilbertson and Ryvarden 1986, Zhao and Zhang 1992), which included descriptions of
basidiomes, basidiospores, habitats and hyphal context systems. However, neither study compared
North American and Asian isolates side-by-side. No mating tests between the North American and
Asian isolates have been conducted to determine if they are different biological species.
Investigations of mating compatibility between North American and Asian isolates of G. frondosa
may uncover an intrinsic reproductive barrier separating these groups.
The name Grifola frondosa (Dicks.) Gray was applied to the basionym Boletus frondosus J. Dicks.,
which was described based on a specimen collected in England (Dickson 1785, Gray 1821).
Although only a single, non-type European isolate was analyzed in the current study, the
phylogenetic data suggest the possibility that a distinct European lineage may exist in G. frondosa.
If this is the case, then phylogenetic taxonomic revisions would require that the name G. frondosa
be applied to the European lineage, with different names for the North American and Asian clades.
Taxonomic changes must await study of the type material and an expanded sampling of isolates
from Europe and the British Isles.
78
Geographic origin
WI
M001
NY
M008
DC
M011
PA
WC808
LA
M012
NY
M014
NY
M035
LA
M009
NY
81
M013
WI
M002
PA
WC364
WI
M015
WI
M003
GA
WI
M004
MD
M006
100
PA
M005
MD
WC367
WV
WC483
MD
M007
PA
M010
81 WC248
Taiwan
Taiwan
M016
Taiwan
M017
Taiwan
M018
Taiwan
M031
Japan
M020
Taiwan
WC836
Taiwan
M029
Korea
M033
China
WC659
China
M036
N/A
M037
Japan
WC828
Korea
WC835
82
Korea
WC581
Korea
WC556 Korea
WC557
Taiwan
WC582
Taiwan
M030
Taiwan
M032
China
Korea
M034
M040
Japan
China
WC583
98
China
WC834
N/A
M038
Korea
M039
89
Taiwan
WC685
Taiwan
WC555
100
Norway
M019
M021
WC493
G. sordulenta G01
ITS+-tubulin
NJ
0.005 changes
I. U.S.
II. Asia
Europe
Figure 2. Phylogenetic analysis of 51 Grifola frondosa isolates.
Based on a combination of rDNA and partial -tubulin gene sequences using the neighbor-joining method
with distance analysis calculated by the Kimura 2-parameter model. Geographic origin is shown beside
isolate codes. Numbers on branches represent bootstrap values obtained from 1,000 replications (values
greater than 80% were shown). Sidebars represent inferred clades based on geographic origin.
79
We analyzed only a single isolate from Europe (WC493) and its phylogenetic relationship to other
isolates is not clearly resolved. G. frondosa is common in Europe (Breitenbach and Kränzlin 1986),
but this was the only culture readily available for this study. The beta-tubulin analysis and
combined analysis both showed WC493 to form a distinct basal branch related to the Asian clade.
However, because most of the phylogenetically informative sites came from the -tubulin sequence,
this inferred relationship between WC493 and the Asian clade cannot be considered strong (Table
4). Because the single European isolate analyzed did not show a strong relationship either to the
North American or Asian clades, a third, European clade may exist whose discovery awaits the
study of more isolates. Further study of Grifola spp and relatives from the Southern Hemisphere
also may uncover additional interesting biogeographic patterns.
Continental phylogeographic structure is common in basidiomycete macrofungi. Intersterility
groups in the Pleurotus ostreatus species complex correlate well with continental biogeography,
and also with an ITS rDNA phylogeny (Vilgalys and Sun 1994). Phylogenetic partitions may exist
that do not correlate with intersterility barriers. In Lentinula edodes (shiitake), strong continental
phylogeographic structure is evidenced based on rDNA phylogenies, including fairly distinct Asian
and North American clades (Hibbett et al. 1995, Hibbett et al. 1998). Our results are consistent
with the proposal that a biogeographic connection exists between basidiomycete macrofungi from
eastern Asia and temperate North America (Wu and Mueller 1997). In the genus Suillus, this
observation is supported by the inference that North American species tend to have closely related
Asian sister taxa, based on ITS rDNA data (Wu et al. 2000). The split-gill fungus Schizophyllum
commune shows worldwide interfertility and a limited degree of phylogeographic structure in its
cosmopolitan range (Raper et al. 1958, James et al. 1999, James et al. 2001).
Table 4. Summary of sequence alignments a of rDNA, -tubulin and combined datasets for
Grifola frondosa.
rDNA
Total nucleotide (nt) sites
Variable nt sites
Phylogenetically informative nt sites
a
574
82
25
-tubulin
587
177
62 (37 in
introns)
Combined
1161
259
87
Alignments including 51 sequences of G. frondosa and G. sordulenta.
Multilocus phylogenetics provides a powerful tool for the recognition of species, as phylogenetic
partitions shared among different loci indicate a historical reproductive barrier between clades (=
genealogical concordance phylogenetic species recognition or GCPSR, Taylor et al. 2000). The
shared partition between Asian and North American isolates indicated by both genes indicates a
reproductive barrier between these groups, meeting the criteria of GCPSR. However, we cannot
say whether the reproductive barrier is intrinsic (i.e., reflecting intersterility) or extrinsic (i.e.,
reflecting geographic separation). Because phylogenetic partitioning can precede the evolution of
intersterility, it cannot be assumed that Asian and North American isolates of G. frondosa are
intersterile. Indeed, strong geographic and phylogenetic partitions, albeit inferred from single
genes, can be observed within intersterility groups in the genus Pleurotus (Isikhuemhen et al. 2000,
R. Vilgalys, personal communication), and, despite worldwide interfertility (Raper 1958), some
degree of continental biogeographic structure can be found in the split-gill fungus Schizophyllum
commune (James et al. 1999, James et al. 2001). It is expected that recombination occurs among
isolates within the two geographic lineages of G. frondosa, but evidence for that cannot be
extrapolated from these data. Neither the beta-tubulin nor the ITS rDNA trees show much strongly
supported structure within lineages, and the partition homogeneity test (PHT) or incongruence
length difference (ILD) test cannot be applied to these datasets because of relatively high levels of
homoplasy (Farris et al. 1995, Koufoupanou et al. 1997, Barker and Lutzoni 2000).
80
100
99
U.S.
83
U.S.
U.S.
67
98
74
WC828 Asia
WC828 Asia
WC493 Europe
95
89
WC493 Europe
WC493
Europe
G. sordulenta
A
Combined/NJ
G. sordulenta
G. sordulenta
B
-tubulin /NJ
97
97
WC828 Asia
C
rDNA/NJ
U.S.
U.S.
U.S.
52
84
WC828 Asia
74
WC828 Asia
91
61 WC828 Asia
WC493 Europe
WC493 Europe
WC493 Europe
G. sordulenta
G. sordulenta
G. sordulenta
D
Combined /MP
One of 405 MP trees
Length=382 steps
CI=0.741, RI=0.901
E
-tubulin /MP
One of 5000 MP trees
length=251 steps,
CI=0.793, RI=0.932
F
rDNA/MP
One of 424 MP
trees length = 113
steps CI=0.796,
RI=0.911
Figure 3. Phylogenetic analysis of 51 Grifola frondosa isolates.
Based on rDNA ITS (C and F), -tubulin (B and E) and combined (A and D) datasets using the neighbor-joining
method (D, E and F) with distance analysis calculated by the Kimura 2-parameter model and maximum
parsimony method (D, E and F). Numbers on branches represent bootstrap values obtained from 1,000
replications.
81
Both neighbor joining (NJ) and most parsimonious (MP) trees derived from all DNA datasets
revealed a consistent grouping of U.S. commercial cultivar WC828 in the Asian clade (Figure 3).
This suggested that WC828 has an Asian origin and is closely related to Asian commercial
cultivars. It is known that molecular data can be effectively used to select unique shiitake
genotypes for evaluation of biological efficiency, quality and average weight of mushrooms (Diehle
and Royse 1986, Levanon et al. 1993). So, a better understanding of the phylogenetic relationships
of G. frondosa may help the selection and breeding of commercial lines and help to improve
commercial cultivation of these mushrooms.
The beta-tubulin gene region provided more than twice as many phylogenetically informative
nucleotide sites as did the ITS rDNA region, in a similar-sized amplicon. This result is consistent
with findings in other fungi (e.g., O’Donnell 2000) and shows that beta-tubulin may be a more
powerful tool for analyzing the intraspecific phylogeography of basidiomycete macrofungi than ITS
rDNA. Other partial protein-coding genes such as translation elongation factor 1- may also be
extremely useful.
REFERENCES
Chang, S.T. 1999. World production of cultivated edible and medicinal mushrooms in 1997 with emphasis
on Lentinus edodes(Berk.) Sing. in China. Intenational J. med. Mushrooms 1:291-300.
Chen, X., C.P. Romaine, M.D. Ospina-Giraldo and D.J. Royse. 1999. A polymerase chain reaction-based
test for the identification of Trichoderma harzianum biotypes 2 and 4, responsible for the worldwide green
mold epidemic in cultivated Agaricus bisporus. Appl. Microbiol. Biotech 52:246-250.
Dickson, J. 1785. Fasciculus Plantarum Crytogamicarum Britanniae. Londini: Nicol.
Diehle, D.A and D.J. Royse. 1986. Shiitake cultivation on sawdust: evaluation of selected genotypes for
biological efficiency and mushroom size. Mycologia 78:929-933.
Felsenstein, J. 1985. Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39:783791.
Fries, E. M. 1821. Systema Mycologicum: Sistens Fungorum Ordines, Genera et Species, Huc U.S.que
Cognitas, Quas ad Normam Methodi Naturalis Determinavit, Disposuit Atque Descripsit. Lundae : ex
officina Berlingiana. 355.
Gilbertson, R. L and L. Ryvarden. 1987. North American Polypores. Fungiflora, Oslo, Norway. 332-333.
Gray, S. F. 1821. A natural arrangement of british plants, according to their relations to each other as pointed
out by Jussieu, De Candolle, Brown, and C including those cultivated for use, with an introduction to
botany, in which the terms newly introduced are explained. Baldwin, Cradock and Joy, London. 643.
Hibbett, D. S, L.B. Gilbert and M.J. Donoghue. 2000. Evolutionary instability of ectomycorrhizal symbioses
in basidiomycetes. Nature 407:506-508.
Hibbett, D. S., Y. Fukumasa-Nakai, A. Tsuneda and M.J. Donoghue. 1995. Phylogenetic diversity in
shiitake inferred from nuclear ribosomal DNA sequences. Mycologia 87:618-638.
Higgins, D.G, A.J. Bleasby and R. Fuchs. 1991. CLUSTAL W: improved software for multiple sequence
alignment. CABIOS 8:189-191.
Hills, D.M. and J.J. Bull. 1993. An empirical test of bootstrapping as a method for assessing confidence in
phylogenetic analysis. Syst Biol 42:182-192.
Levanon, D, N. Rothschild, O. Danai and S. Masaphy. 1993. Strain selection for cultivation of shiitake
mushrooms (Lentinus edodes) on straw. Bioresource Tech 45:9-12.
Royse, D. J. 1997. Specialty mushrooms and their cultivation. Hort Rev 19:59-97.
Singer, R. 1969. Mycoflora Australis. Lehre: Cramer. 382-383.
Swofford, D. L. 2000. PAUP*. Phylogenetic analysis using parsimony (*and other methods), 4.0b4a.
Sinauer Associates, Sunderland, MA.
Taylor, J.W, D.J. Jacobson, S. Kroken, T. Kasuga, D.M. Geiser, D.S. Hibbett and M.C. Fisher. 2000.
Phylogenetic species recognition and species concepts in fungi. Fungal Gen Bio 31:21-32.
Thon, M.R. and D.J. Royse. 1999. Evidence for two independent lineages of shiitake of the Americas
(Lentinula boryana) based on rDNA and -tubulin genes sequences. Mol Phylogen Evol 13:520-524.
82
Vilgalys, R. and B.L. Sun. 1994. Ancient and recent patterns of geographic speciation in the oyster
mushroom Pleurotus revealed by phylogenetic analysis of ribosomal DNA sequences. Proc Natl Acad Sci
USA 91:4599-4603.
White, T., T. Bruns, S. Lee, and J. Taylor. 1990. Amplification and direct sequencing of fungal ribosomal
RNA genes for phylogenetics. In: Innis et al. (ed.) PCR protocols: a guide to methods and applications.
Academic Press, New York. 315-322.
Zhao, J.D.and X. Zhang . 1992. The polypores of china. Bibliotheca Mycologica. J. Cramer, Berlin. 209210.
83