Microbiology (2008), 154, 3460–3468
DOI 10.1099/mic.0.2008/020495-0
Genetic diversity of the endemic gourmet
mushroom Thelephora ganbajun from southwestern China
Tao Sha,1 Jianping Xu,2 Malliya Gounder Palanichamy,1 Han-Bo Zhang,1
Tao Li,1 Zhi-Wei Zhao1 and Ya-Ping Zhang1
Correspondence
Jianping Xu
[email protected]
1
Laboratory for Conservation and Utilization of Bio-Resources, Yunnan University, Kunming
650091, PR China
2
Center for Environmental Genomics, Department of Biology, McMaster University, Hamilton, ON
L8S 4K1, Canada
Ya-Ping Zhang
[email protected]
Received 12 May 2008
Revised
9 August 2008
Accepted 12 August 2008
The ectomycorrhizal fungus Thelephora ganbajun is an endemic gourmet mushroom in Yunnan
province, south-western China. However, despite its widespread consumer appeal, nutritional
value and potential ecological role in natural forests, very little is known about its genetics,
diversity and ecology. In this study, we investigated DNA sequence variation at the internal
transcribed spacer (ITS) regions among 156 specimens collected from 23 sites of nine regions in
Yunnan Province. Our analysis identified a total of 34 ITS haplotypes and these haplotypes were
clustered into five distinct phylogenetic groups. The evolutionary divergences among these clades
are similar to or greater than many known sister species pairs within the genus Thelephora and the
closely related genus Tomentella. Among the 34 ITS haplotypes, 22 were represented by one
specimen each and the remaining 12 were each shared by two or more specimens. The most
common haplotype contained 68 specimens distributed in 21 of the 23 sites, a result consistent
with gene flow among geographical populations. However, analysis of molecular variance
(AMOVA) revealed low but significant genetic differentiation among local and regional
populations. Interestingly, the Mantel test identified that the extent of genetic differentiation was
not significantly correlated with geographical distance. Our study revealed significant genetic
divergence within Th. ganbajun and limited but detectable gene flow among geographical
populations of this endemic ectomycorrhizal gourmet mushroom.
INTRODUCTION
Fungi are important constituents of natural ecosystems.
They contribute both positively and negatively to the
health of plants, animals and humans. In forest and
grassland ecosystems, many fungi can be found and some
are collected for human consumption. Many of these fungi
form symbiotic associations with plant roots, establishing
structures called mycorrhizae. In such associations, the
fungal mycelia around the roots help plants obtain essential
minerals and water from the soil and can contribute to the
plants’ disease resistance and drought tolerance (Brundrett,
2004). Their importance is reflected by the fact that over
80 % of land plants form mycorrhizal associations with
fungi. The most noticeable mycorrhizae in natural forests
belong to the Basidiomycota because many of them
produce conspicuous mushrooms and some are collected
as a source of exotic and highly prized food for humans.
Thelephora ganbajun Zang is a gourmet mushroom that
forms ectomycorrhizae with pine trees, predominantly Pinus
yunnanensis (Zang, 1986, 1987). The mushroom, called ganba-jun by the locals, and its host trees are highly endemic to
Yunnan Province in south-western China. The mature
mushroom produces a unique and attractive aroma. As a
result, there has been a consistent consumer demand and the
mushroom is of increasing economic importance in many
regions in Yunnan. At present, freshly collected gan-ba-jun
are sold for about RMB Yuan 200 per kilogram. In unit
price, gan-ba-jun is comparable to or greater than other
well-known gourmet mushrooms such as the da-hong-jun
(or big red mushroom), the matsutake mushroom, and the
morels and boletus in that region. Because of consumer
demand and our inability to cultivate this mushroom
artificially, there has been noticeable overexploitation and
consequent decline of local populations of this species.
Abbreviations: AMOVA, analysis of molecular variance; ITS, internal
transcribed spacer.
For sustainable utilization and effective conservation of this
mushroom, we need to understand its population biology.
3460
Downloaded from www.microbiologyresearch.org by
2008/020495 G 2008 SGM
IP: 88.99.165.207
On: Wed, 14 Jun 2017 20:03:10
Printed in Great Britain
Biodiversity of the gourmet mushroom Thelephora ganbajun
However, despite its economic and potential ecological
importance, almost nothing is known about its ecology and
genetics. Our objective in this study was to examine the
patterns of genetic variation of this species from within and
between geographical areas. At present, there is no DNA
sequence information about this organism in any public
database. Therefore, we focused our attention on sequence
variation within the internal transcribed spacer (ITS)
regions of the nuclear rRNA gene cluster. The ITS regions
were chosen because there are universal fungal PCR
primers for this region that we could use for amplification
and sequencing. In addition, the ITS is among the most
commonly used gene regions for analysing relationships
among strains within as well as between closely related
fungal species (e.g. Chillali et al., 1998; Köljalg et al., 2001;
Horton, 2002; Lan & Xu, 2006; Hiremath et al., 2008). We
were specifically interested in addressing the following
questions using the ITS sequence data. First, how much
variation is there among strains of this species from the
same geographical areas? Second, what are the patterns of
distribution of specific ITS haplotypes among geographical
regions? Third, how are geographical populations related
to each other? Is there significant genetic differentiation
among geographical populations? If so, is the level of
genetic differentiation positively correlated with geographical distances? And fourth, do all specimens of gan-ba-jun
from different areas in Yunnan form a single tight cluster
within a phylogeny when compared to its closely related
species?
METHODS
Sampling. Natural samples of Th. ganbajun were collected from pine
forests around Yunnan province in south-western China. These
specimens were identified based on their morphological features
(Zang, 1986, 1987). A total of 156 fruiting bodies were collected,
processed and stored at 270 uC. These fruiting bodies were collected
from 23 sites in nine regions of Yunnan. These regions stretched
about 600 km from east to west and about 300 km from south to
north. The names of the counties, communities and regional
municipalities where the sampling sites were located are shown in
Table 1. Because there were two geographical levels and some of them
share the same name, to distinguish them, we denoted the names for
the regional populations with capitalization for only the first letter of
the entire name and the names for the local populations with
capitalization for the first letter in each Chinese word. The
geographical coordinates and the sample sizes from each of the sites
are also presented in Table 1.
Table 1. Geographical distributions and genetic diversity of samples of Th. ganbajun analysed in this study from Yunnan province,
south-western China
Region*/district
CountyD/
community
Baoshan
ChangLing
BaoShan
CiXiong
NanHua
WuDin
LuFeng
MiDu
WenXian
DiZi
FengYi
ShiPing
JingLing
AnLing
YuLiang
SongMing
LuQuan
XunDian
PuEr
MaLong
GuangNan
TongHai
E’Shan
YiMeng
Cixiong
Dali
Honghe
Kunming
Pu’er
Qujing
Wenshan
Yuxi
Lat. (N)
Long. (E)
Sample size
ITS haplotype (no. of isolates
in each haplotype)
24.82
25.12
25.01
25.21
25.55
25.15
25.34
25.69
25.67
25.35
23.73
24.68
24.95
24.9
25.35
25.58
25.56
22.79
25.41
24.05
24.09
24.16
24.67
99.61
99.18
101.54
101.26
102.36
102.08
100.52
100.19
100.18
100.18
102.48
102.58
102.44
103.12
103.03
102.45
103.25
101
103.61
105.09
102.75
102.38
102.15
4
9
8
10
8
6
5
4
4
5
4
1
1
5
9
6
9
20
2
4
4
11
17
1(2); 2(1); 3(1)
1(8); 3(1)
1(4); 7(1); (22(1); 24(1)
1(3); 7(3); 13(1); 17(1); 20(1); 23(1)
1(2); 7(4); 28(1); 31(1)
1(4); 21(1); 31(1)
1(2); 25(2); 31(1)
1(4)
2(1); 7(2); 31(1)
1(2); 13(1); 7(1); 26(1)
1(1); 5(1); 6(1); 7(1)
1(1)
1(1)
1(1); 8(1); 9(1); 10(1); 11(1)
1(6); 15(1); 25(1); 31(1)
1(2); 7(2); 20(1); 25(1)
1(2); 12(1); 14(3); 16(1); 25(2)
8(13); 18(1); 19(2); 34(4)
1(1); 2(1)
1(2); 4(1); 7(1)
1(3); 7(1)
1(5); 2(4); 25(1); 32(1)
1(12); 25(1); 27(1); 30(1); 31(1); 33(1)
Haplotype
diversity
0.625
0.198
0.688
0.780
0.656
0.500
0.640
0.000
0.625
0.720
0.750
NA
NA
0.800
0.519
0.722
0.765
0.525
0.500
0.625
0.375
0.645
0.484
NA, Not applicable.
*Names for the regional populations are capitalized for only the first letter.
DNames for the local populations are capitalized for the first letter in each Chinese word to distinguish them from the regional populations.
http://mic.sgmjournals.org
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 14 Jun 2017 20:03:10
3461
T. Sha and others
Isolation of genomic DNA and ITS sequencing. The total fungal
DNA was isolated from each Th. ganbajun fruiting body using the
CTAB method slightly modified for higher fungi (Xu et al., 1994).
Briefly, the fruiting bodies were thoroughly ground with micropestles and sand in Eppendorf tubes in CTAB buffer. After incubation
at 65 uC for 45 min, the samples were extracted twice with an equal
volume of chloroform and centrifuged for 10 min at 10 000 g after
both extractions. The DNA was precipitated with 2 vols ethanol,
stored for at least 8 h at –20 uC, and then precipitated by
centrifugation at 10 000 g for 15 min. The pellet was washed with
70 % ethanol, dried and redissolved in 50 ml TE buffer (10 mM Tris/
HCl, pH 8, 1 mM EDTA). An aliquot (2 ml) was run on a 1 % (w/v)
agarose gel stained with ethidium bromide for checking the quantity
of genomic DNA.
The contiguous ITS1, 5.8S and ITS2 regions were amplified by PCR
using primers ITS4 and ITS5 described by White et al. (1990). The
conditions of PCR followed those described by Lan & Xu (2006). The
resulting PCR products were examined on a 1 % (w/v) agarose gel
stained with ethidium bromide and fragments of approximately
700 bp were cut out and purified with a Gel Extraction Mini kit
(Watson BioTechnologies). For sequencing, an ABI PRISM 3.1
BigDye Terminator kit (Perkin Elmer) was used and the electrophoresis was carried out on an ABI PRISM 3700 Genetic Analyzer
according to the manufacturer’s instructions. Sequencing was carried
out for both strands using the forward and reverse primers.
Data analysis. The complete ITS sequence was assembled for each
strain using sequences obtained from both the forward and reverse
primers. The sequences were aligned using the CLUSTALW2 multiple
alignment program (http://www.ebi.ac.uk/Tools/clustalw2). The
aligned sequences were exported to PAUP*4.0b10 (Swofford, 2002),
visually inspected and adjusted by hand when necessary. The ITS
unique haplotypes were identified using all aligned nucleotide sites,
including the insertions/deletions, using the neighbour-joining
method. Strains with the same ITS sequences were identified as
belonging to the same ITS haplotype. The haplotype information was
then used for the following two types of analyses.
In the first, the ITS haplotype information for each local and regional
population was analysed using the computer program GenAlEx6
package (Peakall & Smouse, 2006) to identify (i) the haplotype
diversity within each population; (ii) the relationships among local
populations; (iii) the relationship between the level of genetic
differentiation and geographical separation; and (iv) the contributions of local and regional geographical separations to the overall ITS
haplotype variation. Specifically, haplotype diversity and the pairwise
population FST values (a common measure of genetic differentiations;
Wright, 1943) were calculated for each individual population and
each pair of local populations respectively. The Mantel test (Mantel,
1967) was conducted to examine the relationship between the levels of
genetic differentiation and geographical distances. Geographical
distances between pairs of local populations were obtained using
the geographical latitudinal and longitudinal coordinates. Finally,
analysis of molecular variance (AMOVA; Excoffier et al., 1992) was
conducted to estimate the relative contributions of local and regional
geographical separations to the overall ITS haplotype distribution in
Yunnan.
In the second type of test, each unique ITS haplotype sequence was
first used as a query to retrieve closely related sequences from
GenBank. Our sequences and the retrieved sequences were then
aligned using CLUSTALW2. The aligned sequences were visually
inspected, adjusted and then analysed using PAUP. In this analysis,
because of the inclusion of divergent sequences and the differences in
length among the aligned DNA sequences, a large number of gaps (i.e.
insertions and deletions) were introduced in the aligned dataset. As a
result, we treated the gaps as missing data. A neighbour-joining tree
3462
was constructed using the Kimura’s two-parameter model. Bootstrap
confidence values for individual branches were obtained using 1000
resampled datasets.
RESULTS
ITS sequence variation and haplotype distribution
Using the universal fungal primers ITS4 and ITS5, we
successfully amplified the ITS regions (including the entire
5.8S rRNA gene) from all 156 specimens. The ITS sequence
for each specimen was confirmed using the forward and
reverse primers. The 156 aligned sequences were 671
nucleotides long and contained 138 variable sites. Each of
the variable sites contained only two alleles. These variable
sites included 35 insertions/deletions, 85 transitional
substitutions and 18 transversional substitutions. Among
these variable sites, eight were phylogenetically uninformative (i.e. only one strain has a nucleotide different from the
remaining 155 strains that shared another nucleotide) and
the remaining were phylogenetically informative.
The analyses of the aligned ITS sequences of all 156
specimens identified a total of 34 unique sequence types
(or haplotypes). Twelve of the ITS sequence types
representing 24 strains in total were found to contain
one to three nucleotide positions each that showed
evidence of heterozygosity in the sequencing chromatograms. At each of these positions, two alternative bases
were found and all were transitional mutations. We found
no insertion/deletion mutations or transversional mutations within any of the strains. These heterozygous sites
were coded R (for A/G) or Y (for C/T) in our data. Overall,
the difference between the 34 ITS sequence types ranged
between 1 and 138 bases (out of 671 aligned nucleotide
sites). The distributions of these ITS sequence types among
the 23 local populations are presented in Table 1.
Among the 34 ITS sequence types, 22 were represented by
one specimen each and the remaining 12 were each shared
by two or more specimens. The most common type,
haplotype 1, contained 68 specimens that were distributed
in 21 of the 23 sites (Table 1). Similarly, haplotype 7 was
also widely distributed – it contained 16 specimens
collected from eight local populations (Table 1).
However, the other major shared halpotype, haplotype 8,
was more restricted in its geographical distribution.
Among the 14 strains with the haplotype 8 ITS sequence,
13 were from PuEr and one was from YuLiang, Kunming.
The number of ITS haplotypes for each local population
ranged between one and seven, with the highest number
found in YiMeng, Yuxi. Aside from the two local
populations (JingLing and AnLing, both in Kunming)
where only one specimen each was available for analysis, 20
of the remaining 21 local populations had more than one
ITS haplotype. Of these 21 local populations, 13 were
found to contain unique ITS haplotypes not found in other
populations. The local populations and the specific
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 14 Jun 2017 20:03:10
Microbiology 154
Biodiversity of the gourmet mushroom Thelephora ganbajun
haplotypes that were unique to these local populations
were: CiXiong (haplotypes 22, 24 and 29), NanHua
(haplotypes 17 and 23), LuFeng (haplotype 21), FengYi
(haplotype 26), ShiPing (haplotypes 5 and 6), YuLiang
(haplotypes 9, 10 and 11), SongMing (haplotype 15),
XunDian (haplotypes 12, 14 and 16), PuEr (haplotypes 18,
19 and 34), GuangNan (haplotype 4), EShan (haplotype
32) and YiMeng (haplotypes 27, 30 and 33) (Table 1). The
haplotype diversity in each local population ranged
between 0 (for WenXian) and 0.8 (for YuLiang) (Table 1).
Population analyses of the ITS haplotype
distribution
With the exception of two local populations (JingLing and
AnLing) where only one strain from each location was
available and which were not analysed in our population
genetics comparison, the remaining 21 local populations
containing 154 strains were analysed by the population
genetic analysis program GenAlEx 6. Our analyses
identified a range of FST values between pairs of local
populations (Table 2). The lowest value (0.021) was found
between LuQuan, Kunming, and NanHua, Cixiong, while
the highest (0.584) was found between WenXian, Dali, and
PuEr city, Pu’er.
Of the total observed haplotype allele frequency variation,
the majority was found from within local populations
(82.5 %; Table 3). Of the remaining 17.5 %, 7.4 % could be
attributed to among local populations within regions and
10.1 % could be attributed to among regional populations.
The contributions from each of the three sources were
significantly greater than 0, indicating statistically significant genetic differentiations among local and regional
populations (P,0.01 for PhiRT, PhiPR and PhiPT in the
AMOVA tests, Table 3). The Mantel test indicated that
while there was a slight positive correlation between
geographical distances and FST values (a common measure
of genetic differentiation), this correlation was statistically
insignificant (P50.09; Fig. 1).
Phylogenetic analyses
Based on BLAST searches against the GenBank database, we
retrieved a total of 83 ITS sequences with high levels of
sequence identity (¢90 %) to our sequences. These 83
retrieved sequences all have associated species identification in the GenBank database, belonging to either the
genus Thelephora or Tomentella. It should be pointed out
that there were many other ITS sequences (.100) in
GenBank with similar levels of sequence identities (~90 %)
to ours. However, those sequences were from direct
environmental DNA sampling and sequencing and there
was no species designation associated with them. Therefore
they were not included in the analyses here. Based on an
initial phylogenetic analysis of all 83 ITS sequences from
GenBank and comparison with our own 34 ITS haplotype
sequences, we selected 27 representative GenBank
http://mic.sgmjournals.org
sequences. These 27 sequences were chosen based on their
comparable sequence lengths to ours, their phylogenetic
positions, and in several cases the availability of two or
more strains for the same species. The selections of
multiple strains from the same species were to compare
the potential divergence within other species to that within
Th. ganbajun. In total, the 27 ITS sequences from GenBank
represented nine species in the genus Tomentella (T.
atramentaria, T. badia, T. pilosa, T. bryophila, T. galzinii, T.
viridula, T. subtestacea, T. coerulea and T. sublilacina) and
five species in the genus Thelephora (Th. americana, Th.
caryophyllea, Th. terrestris, Th. palmata and Th. regularis).
Their respective GenBank accession numbers for ITS
sequences are shown in Fig. 2.
Our analyses of the above samples identified five
phylogenetically distinct clades for the samples from
Yunnan, with bootstrap support all greater than 90 %.
Three of these clades (clades 1, 2 and 3 in Fig. 2) contained
multiple ITS haplotypes each while the remaining two
clades contained only one haplotype each. Specifically,
clade 1 contained eight ITS haplotypes represented by 22
strains; clade 2 contained six haplotypes represented by 20
strains; and clade 3 contained 18 haplotypes represented by
106 strains. Each of the three clades had a bootstrap
support of over 90 %. In addition to these three clades,
there were two additional divergent Th. ganbajun haplotypes, haplotype 3 (represented by two strains) and
haplotype 34 (represented by four strains). Our analyses
revealed that strains in clades 1, 2 and 3, as well as
haplotype 3, showed closer evolutionary relationships to
representative species from the genus Tomentella than to
those from the genus Thelephora. In contrast, haplotype 34
was more similar to ITS sequences of species in the genus
Thelephora, specifically Th. palmata and Th. regularis.
DISCUSSION
In this study, we analysed the ITS sequences of 156 strains
of the endemic gourmet mushroom Th. ganbajun from
Yunnan, south-western China. The strains were collected
from 23 sites distributed in nine geographical regions that
spanned an area of over 600 km from east to west and
~300 km from south to north. Our analyses identified a
large number of ITS sequence types. Both local and
regional geographical separations contributed significantly
to the observed ITS haplotype variation. The phylogenetic
results suggested that the analysed samples probably
contained five distinct cryptic species.
In our population samples, a few haplotypes, predominantly haplotype 1, were found distributed in multiple
geographical areas (Table 1). This result is consistent with a
certain degree of gene flow between geographical populations of this species in nature. Using the classical measure
of gene flow Nm [Nm5(12FST)/4FST] as suggested by
Wright (1943), the mean gene flow among regional
populations of Th. ganbajun in Yunnan is estimated to
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 14 Jun 2017 20:03:10
3463
T. Sha and others
3464
Table 2. Pairwise FST values between geographical populations of Th. ganbajun from Yunnan province, south-western China
Abbreviations of geographical populations: C-B, ChangLing-Baoshan; B-B, BaoShan-Baoshan; C-C, CiXiong-Cixiong; N-C, NanHua-Cixiong; W-C, WuDin-Cixiong; L-C, LuFeng-Cixiong; M-D,
MiDu-Dali; W-D, WenXian-Dali; D-D, DiZi-Dali; F-D, FengYi-Dali; S-H, ShiPing-Honghe; Y-K, YuLiang-Kunming; S-K, SongMing-Kunming; L-K, LuQuan-Kunming; X-K, XunDian-Kunming;
P-P, PuEr-PuEr; M-Q, MaLong-Qujing; G-W, GuangNan-Wenshan; T-Y, TongHai-Yuxi; E-Y, Eshan-Yuxi; Y-Y, YiMeng-Yuxi.
B-B
C-C
N-C
W-C
L-C
M-D
W-D
D-D
F-D
S-H
Y-K
S-M
L-K
X-K
P-P
M-Q
G-W
T-Y
E-Y
Y-Y
C-B
B-B
C-C
N-C
W-C
L-C
M-D
W-D
D-D
F-D
S-H
Y-K
S-M
L-K
X-K
P-P
M-Q
G-W
T-Y
E-Y
0.124
0.067
0.095
0.155
0.085
0.117
0.231
0.200
0.087
0.120
0.116
0.077
0.106
0.122
0.270
0.053
0.091
0.111
0.036
0.077
0.113
0.200
0.291
0.078
0.212
0.059
0.417
0.168
0.243
0.245
0.065
0.210
0.250
0.469
0.229
0.149
0.076
0.172
0.044
0.051
0.095
0.058
0.093
0.185
0.176
0.048
0.080
0.095
0.050
0.058
0.101
0.245
0.116
0.045
0.056
0.074
0.050
0.038
0.111
0.107
0.284
0.095
0.032
0.053
0.087
0.104
0.021
0.094
0.210
0.141
0.049
0.096
0.096
0.110
0.169
0.149
0.391
0.057
0.075
0.072
0.132
0.165
0.042
0.141
0.257
0.204
0.079
0.143
0.154
0.177
0.102
0.143
0.260
0.092
0.143
0.143
0.027
0.120
0.148
0.322
0.143
0.085
0.067
0.098
0.027
0.304
0.201
0.105
0.129
0.122
0.070
0.080
0.078
0.264
0.168
0.117
0.159
0.098
0.097
0.524
0.250
0.333
0.333
0.125
0.297
0.340
0.584
0.333
0.231
0.143
0.257
0.097
0.145
0.120
0.168
0.259
0.106
0.180
0.270
0.217
0.167
0.273
0.178
0.280
0.073
0.095
0.084
0.052
0.102
0.233
0.135
0.054
0.086
0.091
0.087
0.101
0.136
0.062
0.110
0.221
0.167
0.083
0.143
0.119
0.143
0.136
0.102
0.099
0.135
0.161
0.116
0.183
0.114
0.144
0.101
0.126
0.314
0.134
0.077
0.056
0.083
0.015
0.089
0.232
0.154
0.054
0.097
0.099
0.112
0.216
0.168
0.122
0.187
0.110
0.141
0.322
0.270
0.379
0.262
0.329
0.143
0.176
0.016
0.136
0.059
0.098
0.077
0.128
0.045
0.088
Microbiology 154
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 14 Jun 2017 20:03:10
Biodiversity of the gourmet mushroom Thelephora ganbajun
Table 3. Summary results of AMOVA within and among populations of Th. ganbajun from Yunnan, south-western China
d.f., Degrees of freedom; SS, sum of squared observations; MS, mean of squared observations; Est. var., estimated variance; % Var., percentage of
total variance; PhiRT, proportion of the total genetic variance between regions; PhiPR, proportion of the total genetic variance among populations
within a region; PhiPT, proportion of the total genetic variance among individuals within populations.
Source
Among regions
Among local populations within
regions
Within local populations
Total
d.f.
SS
MS
Est. var.
% Var.
AMOVA
statistics
Value
P
8
12
9.814
6.465
1.227
0.539
0.041
0.030
10.1 %
7.4 %
PhiRT
PhiPR
0.102
0.082
0.010
0.010
133
153
44.421
60.701
0.334
2.100
0.334
0.405
82.5 %
PhiPT
0.175
0.010
be 2.225 and that among local populations within regions
is 3.128. However, the gene flow has not obscured a
statistically significant genetic differentiation among local
and regional populations. Similar evidence for gene flow
has been found in many basidiomycete mushrooms (e.g.
James et al., 1999; Xu et al., 2005). For example, in natural
populations of the button mushroom Agaricus bisporus, the
sharing of certain genotypes was found for strains from
different regions, countries and even continents (Xu et al.,
1997, 1998). Because those genotypes were identical or very
similar to cultivated strains, it was suggested that humanaided dispersal was responsible for such long-distance gene
flow in A. bisporus. However, for most other microbial
species, the mechanisms of dispersal are largely unknown.
For example, in a recent study of the ectomycorrhizal
mushroom Tricholoma matsutake from south-western
China, evidence for gene flow was also found between
populations as far as 2000 km from each other (Xu et al.,
2007, 2008). In these studies, it has been commonly
hypothesized that wind-aided dispersal of sexual basidiospores was probably responsible for the observed gene flow.
A similar process could account for the wide distribution
of haplotype 1 in our samples and for the gene flow
detected among local and regional populations.
differentiation in Th. ganbajun is similar to what has been
reported in many other fungal species (e.g. James et al.,
1999; Xu et al., 2005, 2008). For example, the two species
mentioned above, A. bisporus and Tr. matsutake, had FST
values very similar to that reported here (Xu et al., 1997,
2008). For A. bisporus, two distinct elements were found in
many natural environments: one cultivar-like and the other
mostly native (Xu et al., 1997, 1998). While the cultivarlike elements showed no genetic differentiation (see above
discussion), the native elements were significantly differentiated among the regional and continental populations
(Xu et al., 1997, 1998). The significant genetic differentiation among local and regional populations of Th.
ganbajun suggests that many local populations might
contain distinct genetic elements, a conclusion supported
by the observation that a large number of unique ITS
haplotypes were found in only one of the sampled areas
(Table 1). Because our sample sizes are relatively small for
several sites, it is possible that more ITS haplotypes might
be found if more extensive sampling were conducted and
that wider distributions of some of these unique haplotypes
might be revealed. Nevertheless, our analyses of the 156
samples analysed here strongly suggest that care should be
taken to conserve individual local populations to maintain
the genetic diversity of this species in nature. If populations
are overexploited, many of the unique local genetic
elements could be permanently lost.
While evidence for dispersal has been found in Th.
ganbajun, our population genetic analyses also suggest
that, overall, gene flow was somewhat limited and there
was low but significant genetic differentiation at both the
local (~7 %) and regional (~10 %) levels (Tables 2 and 3).
Interestingly, the result of low but significant genetic
FST
0.6
Our phylogenetic analysis identified several distinct clades
within the morphological species Th. ganbajun. Based on
the amount of divergence among these clades, each clade
y = 0.0001x + 0.1164
R2 = 0.0311, P = 0.09
0.4
0.2
100
http://mic.sgmjournals.org
200
500
300
400
Geographical distance (km)
600
700
Fig. 1. Relationship between the levels of
genetic differentiation (FST values) and geographical distances between pairs of local
populations of Th. ganbajun from Yunnan,
south-western China. Although a positive
correlation was found, it was statistically
insignificant (P50.09).
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 14 Jun 2017 20:03:10
3465
T. Sha and others
Fig. 2. A neighbour-joining tree of the ITS sequences of 34 haplotypes of Th. ganbajun and representative sequences from
species of the genera Tomentella and Thelephora. For each ITS haplotype of gan-ba-jun (GBJ), the first number represents the
haplotype assignment corresponding to those in Table 1; the second number represents the total number of strains (out of 156)
belonging to the specific haplotype. The 27 reference strains are each represented by their GenBank accession number, the
genus abbreviation (T for Tomentella and Th for Thelephora), the species name and when multiple strains from the same
species were available, the strain code (1, 2, 3 or 4). Numbers across branches are bootstrap values greater than 90 %
obtained from 1000 replicates. Gaps were treated as missing data. Branch lengths are proportional to the amount of sequence
divergence. Tree length, 670; consistency index, 0.610; retention index, 0.882.
might represent a distinct cryptic species. Specifically, the
amount of ITS sequence divergence as reflected by branch
lengths among clades 1, 2 and 3 was similar to or greater
than that between many known sister species pairs found
3466
in either Tomentella or Thelephora (Fig. 2). Similarly,
haplotypes 3 and 34 might also represent different species.
Recent gene genealogical studies have shown that many
fungal species are in fact species complexes, with each
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 14 Jun 2017 20:03:10
Microbiology 154
Biodiversity of the gourmet mushroom Thelephora ganbajun
containing more than one phylogenetically distinct, cryptic
species (e.g. Taylor et al., 2000; Pujol et al., 2005; Xu et al.,
2005). To determine whether the clades and genotypes
identified here for Th. ganbajun are indeed phylogenetically
distinct species, genealogical analyses using multiple genes
from other genomic regions are needed. If congruent
genealogical patterns of strain relationships were found for
genes from different genomic regions, formal separation of
Th. ganbajun into different species would be warranted
(Taylor et al., 2000; Moncalvo, 2005).
Our study also identified another taxonomic issue worth
further investigating. Specifically, the majority of the strains
(152 out of 156) analysed here were found to have ITS
sequences clustering with those from species in the genus
Tomentella, not with their originally designated genus
Thelephora. Only one of the 34 ITS haplotypes (i.e. ITS
haplotype 34), containing four strains, clustered more closely
with species in the genus Thelephora than with those in the
genus Tomentella (Fig. 2). All four strains were from one
geographical location, PuEr. Because no sequence (for either
ITS or other genes) is available for the type specimen of Th.
ganbajun, we cannot determine if the ITS sequence of the
original type specimen is more similar to haplotype 34 or to
the other 33 haplotypes. The type specimen of Th. ganbajun
was collected in the 1980s and only a limited amount of
material for the type specimen is left (M. Zang, personal
communication). In addition, fungal materials collected and
preserved before the mid-1990s at the Herbarium in
Kunming Institute of Botany, Chinese Academy of Sciences,
have been found to produce very poor quality DNA or no
DNA at all (Z.-L. Yang, personal communication) due to
inappropriate conditions for drying and preservation during
that time period. Therefore, we were unable to obtain the
material for DNA isolation and sequencing from the type
specimen. However, the type specimen was collected from the
same geographical region and ecological conditions as the
four strains belonging to haplotype 34, in PuEr (Zang, 1986,
1987). Therefore, the type specimen might contain the same
ITS sequence as that of haplotype 34, more similar to
sequences from the majority species in the genus Thelephora
than to those from Tomentella.
If the above hypothesis were true, that only four of the 156
specimens collected from only one geographical region
belong to the species Th. ganbajun, then the majority of the
collected and consumed ‘gan-ba-jun’ in Yunnan probably
belong to one or several new species, in a different genus,
Tomentella. Morphologically, Thelephora and Tomentella are
very similar, and difficult to distinguish. The defining
morphological criteria (established based on materials from
Europe) indicated that species of Tomentella have very short
fruiting bodies close to the substrate while Thelephora has
more pronounced fruiting bodies (Zang, 1986, 1987).
However, all the 156 specimens analysed here had similarly
sized, pronounced fruiting bodies.
That using the size and growth morphology of fruiting
bodies to separate specimens into Tomentella and
http://mic.sgmjournals.org
Thelephora is not adequate is supported by our analyses
of representative sequences from other species in these two
genera. Specifically, our preliminary analyses of the 34 ITS
haplotypes from our study and the 27 representative
sequences from GenBank indicated that neither Tomentella
nor Thelephora is a monophyletic group (Fig. 2; e.g. see
Tometella sublilacina clustering with species in Thelephora).
Similar findings have been reported in several recent
studies on Tomentella/Thelephora using ITS sequences
(Haug et al., 2005; Taylor & McCormick, 2008). More
detailed taxonomic investigations of their microscopic and
macroscopic features coupled with the analyses of multiple
additional genes from more species are needed to resolve
these taxonomic issues.
In conclusion, this study has revealed extensive ITS
sequence variation within and among populations of the
gourmet ectomycorrhizal mushroom gan-ba-jun. Our
study identified evidence for gene flow and low but
significant genetic differentiation among local and regional
populations of this species in its endemic range in southwestern China. Furthermore, several distinct clades were
revealed within this species and our analyses suggested that
these clades probably corresponded to distinct phylogenetic species. We have discussed the taxonomic and
conservation implications of our results and we suggest
that analyses of DNA sequences of more genes from
different genomic regions as well as more species are
needed in order to resolve the taxonomic issues related to
gan-ba-jun.
ACKNOWLEDGEMENTS
We thank Professors Pei-Gui Liu and Yong-Chang Zhao for
morphological identification of Thelephora ganbajun. This work was
supported by grants from the National Basic Research Program of
China (the 973 Program, 2007CB411600), the Department of Science
and Technology of Yunnan Province, China (2005C0003M;
2007PY01-24), and the Natural Science and Engineering Research
Council (NSERC) of Canada.
REFERENCES
Brundrett, M. (2004). Diversity and classification of mycorrhizal
associations. Biol Rev 79, 473–495.
Chillali, M., Wipf, D., Guillaumin, J. J., Mohammed, C. & Botton, B.
(1998). Delineation of the European Armillaria species based on the
sequences of the internal transcribed spacer (ITS) of ribosomal DNA.
New Phytol 138, 553–561.
Excoffier, L., Smouse, P. E. & Quattro, J. M. (1992). Analysis of
molecular variance inferred from metric distances among DNA
haplotypes: application to human mitochondrial DNA restriction
data. Genetics 131, 479–491.
Haug, I., Weiss, M., Homeier, J., Oberwinkler, F. & Kottke, I. (2005).
Russulaceae and Thelephoraceae form ectomycorrhizas with members
of the Nyctaginaceae (Caryophyllales) in the tropical mountain rain
forest of southern Ecuador. New Phytol 165, 923–936.
Hiremath, S. S., Chowdhary, A., Kowshik, T., Randhawa, H. S.,
Sun, S. & Xu, J. (2008). Long-distance dispersal and recombination in
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 14 Jun 2017 20:03:10
3467
T. Sha and others
environmental populations of Cryptococcus neoformans var. grubii
from India. Microbiology 154, 1513–1524.
Taylor, J. W., Jacobson, D. J., Kroken, S., Kasuga, T., Geiser, D. M.,
Hibbett, D. S. & Fisher, M. C. (2000). Phylogenetic species recognition
Horton, T. R. (2002). Molecular approaches to ectomycorrhizal
and species concepts in fungi. Fungal Genet Biol 31, 21–32.
diversity studies: variation in ITS at a local scale. Plant Soil 244,
29–39.
White, T. J., Bruns, T., Lee, S. & Taylor, J. (1990). Amplification and
James, T. Y., Porter, D., Hamrick, J. L. & Vilgalys, R. (1999). Evidence
for limited intercontinental gene flow in the cosmopolitan mushroom, Schizophyllum commune. Evolution 53, 1665–1677.
Köljalg, U., Jakucs, E., Bóka, K. & Agerer, R. (2001). Three
ectomycorrhizae with cystidia formed by different Tomentella species
as revealed by rDNA ITS sequences and anatomical characteristics.
Folia Cryptogam Est 38, 27–39.
Lan, L. & Xu, J. (2006). Multiple gene genealogical analyses suggest
divergence and recent clonal dispersal in the opportunistic human
pathogen Candida guilliermondii. Microbiology 152, 1539–1549.
Mantel, N. (1967). The detection of disease clustering and a
generalized regression approach. Cancer Res 27, 209–220.
Moncalvo, J.-M. (2005). Molecular systematics: major fungal phylogen-
etic groups and fungal species concepts. In Evolutionary Genetics of
Fungi, pp. 1–34. Edited by J. Xu. Norwich, UK: Horizon Scientific Press.
Peakall, R. & Smouse, P. E. (2006). GENALEX 6: genetic analysis in
Excel. Population genetic software for teaching and research. Mol Ecol
Notes 6, 288–295.
Pujol, C., Dodgson, A. & Soll, D. R. (2005). Population genetics of
ascomycetes pathogenic to humans and animals. In Evolutionary
Genetics of Fungi, pp. 149–188. Edited by J. Xu. Norwich, UK:
Horizon Scientific Press.
Phylogenetic analysis using parsimony
(and other methods), version 4. Sunderland, MA: Sinauer Associates.
Swofford, D. L. (2002).
PAUP*:
Taylor, D. L. & McCormick, M. K. (2008). Internal transcribed spacer
primers and sequences for improved characterization of basidiomycetous orchid mycorrhizas. New Phytol 177, 1020–1033.
3468
direct sequencing of fungal ribosomal genes for phylogenetics. In PCR
Protocols: a Guide to Methods and Applications, pp. 315–322. Edited by
M. A. Innis, D. H. Gelfand, J. J. Sninsky & T. J. White. New York:
Academic Press.
Wright, S. (1943). Isolation by distance. Genetics 28, 114–138.
Xu, J., Yoell, H. J. & Anderson, J. B. (1994). An efficient protocol for
isolating DNA from higher fungi. Trends Genet 10, 226–227.
Xu, J., Kerrigan, R. W., Callac, P., Horgen, P. A. & Anderson, J. B.
(1997). Genetic structure of natural populations of Agaricus bisporus,
the commercial button mushroom. J Hered 88, 482–488.
Xu, J., Kerrigan, R. W., Sonnenberg, A. S., Callac, P., Horgen, P. A. &
Anderson, J. B. (1998). Mitochondrial DNA variation in natural
populations of the mushroom Agaricus bisporus. Mol Ecol 7, 19–33.
Xu, J., Cheng, M., Tan, Q. & Pan, Y. (2005). Molecular population
genetics of basidiomycete fungi. In Evolutionary Genetics of Fungi, pp.
189–220. Edited by J. Xu. Norwich, UK: Horizon Scientific Press.
Xu, J., Guo, H. & Yang, Z.-L. (2007). Single nucleotide polymorphisms
in the ectomycorrhizal mushroom
Microbiology 153, 2002–2012.
Tricholoma
matsutake.
Xu, J., Sha, T., Li, Y.-C., Zhao, Z.-W. & Yang, Z.-L. (2008).
Recombination and genetic differentiation among natural populations of the ectomycorrhizal mushroom Tricholoma matsutake from
southwestern China. Mol Ecol 17, 1238–1247.
Zang, M. (1986). Criticism on Thelephora ganbajun position. Edible
Fungi 4, 1–2 (in Chinese).
Zang, M. (1987). Some new and noteworthy higher fungi from eastern
Himalayas. Acta Botanica Yunnanica 9, 81–88 (in Chinese).
Edited by: J.-R. Xu
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 14 Jun 2017 20:03:10
Microbiology 154