全 文 ：外生菌根菌的分子生态学 : 分子标记基因、
遗传个体及生态意义
?
李墨婵 , 徐建平??
( 加拿大麦克马斯特大学生物系 , 安大略 L8S 4K1 , 加拿大 )
摘要 : 外生菌根真菌与很多植物形成互利共生关系 , 在营养物质交换和碳循环等方面起着关键性的作用 ,
是森林生态系统的重要组成部分。近期生物技术的发展使得人们对外生菌根菌的群体遗传学和分子生态学
有了更加深入的认识。本文介绍了一些常用的鉴定外生菌根菌的分子标记 , 并对每种分子标记的特点及其
适用范围进行了讨论。文中总结了几种常用的鉴定未知外生菌根菌的方法 , 指出了一些在研究外生菌根菌
过程中需要克服的内在困难 , 其中之一就是很多外生菌根菌不可以人工培养 , 所以人们缺少对其地下部分
分布规律和动态变化的了解。在寄主专一性、物种多样性和丰富度、遗传个体大小、繁殖方式等方面 , 近
期对外生菌根菌的分子生物学研究已经获得了很多重要的结果。作者讨论了这些研究成果对于今后开展外
生菌根菌研究的重要意义以及在森林生态系统保育方面的潜在应用价值。
关键词 : 核糖体 RNA 基因 ; ITS; 菌根 ; 蘑菇 ; 繁殖方式 ; 生态相互作用
中图分类号 : Q 948.12 文献标识码 : A 文章编号 : 0253 - 2700 ( 2009) 03 - 193 - 17
Molecular Ecology of Ectomycorrhizal Fungi: Molecular
Markers, Genets and Ecological Importance
LI Mo-Chan, XU J ian-Ping* *
( Department of Biology, McMaster University, Hamilton, Ontario, L8S 4K1 , Canada)
Abstract: Ectomycorrhizal ( EcM) fungi formmutualistic symbioseswith many tree species and are regardedas key organ-
isms involved in nutrient and carbon cycling in forest ecosystems . Recent technological advances have contributed signifi-
cantly toour understanding of the population genetics and molecular ecology of EcM . In this review, we first present the
commonly used molecular markers for characterizing individual EcM fungi . The properties of different types of molecular
markers and their general utilities are discussed . Wethen summarizethe common approaches for identifying unknown EcM
fungi and point out the intrinsic difficulties associated with conducting EcM fungal research . One major deficiency is our
lack of understandingof the below-ground distribution and dynamics of EcM fungi dueto the non-cultivablenature of many
EcM fungi . Recent molecular investigations of the EcM fungi have provided a variety of important data with regard to their
host specificities, species diversity and abundance, genet size, and their reproductive strategies . We discuss the relevance
of thesefindings to further functional investigations of EcM fungi and to potential implications in theconservation andman-
agement of forest ecosystems .
Key words: rRNA genes; ITS; Mycorrhizae; Mushroom; Mode of Reproduction; Ecological Interaction
Many of theknown fungi are found with close as-
sociations of the rootsof plant species, formingmutual-
ly beneficial symbiotic relationships . These fungi can
colonize the plant roots and derive nutrients such as
云 南 植 物 研 究 2009 , 31 (3) : 193～209
Acta Botanica Yunnanica DOI : 10 .3724?SP. J . 1143 .2009.09086
?
?? ?Author for correspondence; E-mail : jpxu@ mcmaster . ca
Received date: 2009 - 04 - 27 , Accepted date: 2009 - 05 - 04
作者简介 : 李墨婵 , 女 , 在读硕士研究生 , 从事菌根真菌多样性及分子生态学研究。 ?
Foundation items: supported by Genome Canada and McMaster University
soluble carbohydrates, amino acids and vitamins from
the plants . In return, plant hosts use the expanded
surface of fungal mycelia to absorb water and minerals
fromsoil ( Read, 1991; Simard et al. , 1997 , 2002;
Table 1) . The fungi that colonize plant roots and form
mutualistic relationships with their plant hosts are
called mycorrhizae, a name derived from Greek that
means“ fungus roots”. In a mycorrhizal association,
the fungus may colonize host roots either intracellularly
or extracellularly, corresponding respectively to two
groups of mycorrhizae: endomycorrhzae and ectomycor-
rhizae . In endomycorrhizae, fungal hyphae penetrate
plant cell walls and form vesicles or arbuscules inside
host plant cells . In contrast, in ectomycorrhizae, fun-
gal hyphae grow extracellularly in the inter-cellular
spaces and formsheaths around plant roots .
Table 1 Contrast between hosts and EcM fungi in terms of
living benefits from each other
Plant hosts benefit fromEcM
fungi
EcM fungi benefit from its host
Supplying phosphorus, ni-
trogen, water
Supplying soluble carbohydrates
Protecting against pathogens Providing niche space
Creating strong soil structure Supplying amino acids
Facilitating nutrient transfer
among plants
Ensuring ecological stability and
evolutionary selectivity in nature
Enhancing cooperation and
lowering competition among
plants
Associating and interacting with
other symbiotic microbes such as
nitrogen fixers
The symbiotic association between plants and their
mycorrhizae can be traced back to 400 - 460 million
years ago, when the first plant appeared on land ( Re-
my et al. , 1994) . Through the long history, both the
plants and mycorrhizal fungi have adapted to benefit
each other from the symbiosis . For example, many
studies have demonstrated that mycorrhizal fungi play
important roles for maintaining the normal growth of
their plant partners ( Jeffries et al. , 2003 ) . Without
mycorrhizal colonization, plant hosts showed slower
growth rate than those with EcM fungi ( Smith and
Read, 1997) . Recent research suggests that the reason
for the poor performance of mycorrhizae-free plants
could be due to their inadequate water andmineral up-
take ( Richard et al. , 2005; Selosse et al. , 2006 ) .
Similarly, it has been demonstrated that plants with
mycorrhizae are often more resistant to draught and to
infectious diseases, such as those caused by microbial
soil-borne pathogens ( Bledsoe et al. , 1982) . Howev-
er, for many plant-mycorrhizal associations, it still re-
mains to be empirically determined whether this rela-
tionship is necessary and to what extent plants rely on
fungi to maintain their normal growth in natural environ-
ments .
One significant issue in mycorrhizal research is
the specificity of plant-mycorrhizal association . Some
plants can form mycorrhizae with many different fungi
whileotherswithonly afew . Similarly, somefungi can
form mycorrhizae on many plant hosts whileothers with
only one or a few . For example, trees such as oak,
beech and birch form mycorrhizal relationships with
only one or a few fungal partner (s) . Lactarius delici-
osus is typically associated with Pinus pinea while
Suillus granulatus and Russula emetica with Pinus pi-
naster (Gardes and Bruns, 1996a; Taylor and Bruns,
1999) . However, many photosynthetic plants seem to
be able to formsymbiosis with multiple unrelated EcM
fungal species and many EcM fungi seemto be able to
formsymbiotic associations with multiple unrelated pla-
nts . Such broad associations provide significant poten-
tials for mutual support among plants and fungi innatu-
ral ecosystems (Horton and Bruns, 2001) .
An estimated> 95% of all terrestrial plant species
form mycorrhizal associations ( Trappe, 1987 ) . The
mycorrhizal fungi are broadly distributed across differ-
ent phylogenetic groups and ecological niches . Such
broad distributions are indicative of the importance of
mycorrhizae in plant communities . As a result, mycor-
rhizal research has attracted a lot of attention fromsci-
entists in many disciplines such as mycology, plant bi-
ology, population biology, and ecology ( Trappe,
1987) . The focus of this review will be on ectomycor-
rhizae .
There are several notable features about EcM fun-
gi . First, EcM fungi are composed of many phylogene-
tically diverse species . More than 5 , 000 fungal spe-
cies havebeen estimated toformectomycorrhizaeworld-
wide (Amaranthus, 1998 ) . They include species from
many genera, family, class and order in the phylaBa-
491 云 南 植 物 研 究 31 卷
sidiomycetes, Ascomycetes, and Zygomycetes ( LePage
et al. , 1997 ) . Secondly, EcM fungi have board geo-
graphic ranges and arefound in most regionsof theglo-
bal ecosystem . Themainmycorrhizal centersof diversi-
ty and significant research are in temperate and tropical
forests, such as those in southwestern China, north-
eastern China, the northwest pacific coast of North
America, and northern Europe . Thirdly, EcM fungi
contain many of the high-valued gourmet mushrooms
such as truffles, matsutake, and chanterelles . There-
fore, understanding their basic biology, systematics,
population structures, andmajor reproductive strategies
for EcM fungi can aid us develop better strategies to
maintain their biodiversity and growth in natural eco-
systems .
In contrast to ectomycorrhizal fungi , endomycor-
rhizal fungi fall into a single phyletic group, the
Glomeromycota . The contrasting pattern and the
polyphyletic natureof EcM fungi have raisedmany fun-
damental questions of EcM fungi themselves ( Horton
and Bruns, 2001) , including ( i) how diverse areEcM
fungi in typical ecosystems ? ( ii ) Howmany species are
there and what is the most abundant species in EcM
community ? ( iii ) How specific are mycorrhizal fungi-
plant symbiosis ? ( iv) What effects do EcM fungal pop-
ulations exert on their local ecosystem? And ( v) what
is the historyof EcM fungi ?Howhave theyevolved and
diversified ?
To address the abovequestions, both intrinsic and
extrinsic difficulties need to be overcome with EcM
community studies . Intrinsic hurdles come from EcM
fungi themselves . First, as EcM fungi do not grow nor-
mally without their hosts, it is extremely difficult to
simulate their growth in laboratory settings . Secondly,
large numbers of underground EcM fungal species are
still not describedor identified . In early mycorrhizal re-
search, only fruit bodies were used for study . This is
because the vegetative structures of these fungi ( mycor-
rhizae and mycelia) lying underneath the ground are
small andmorphologically inconspicuous, making it hard
to obtain and distinguish . In addition, some EcM fungi
do not producefruitingbodies and they are rarely inves-
tigated ( Horton and Bruns, 2001 ) . Indeed, modern
taxonomy of fungi was constructed mostly based on the
analyses of fruiting structures, without genetic markers,
matching species names to underground structures is
problematic, and in certain instances, close to impossi-
ble . Themultiple stages of EcM development also com-
plicate research . Often, different stages and aspects of
EcM fungi are investigated by different methods, and
sometimes by different scientific disciplines . Some of
the methods differentially used by different groups of
scientists include morphological versus molecular meth-
ods in terms of identifying species, choices of different
molecular markers, sensitivity in different molecular
methods, dilemma in phylogenetic analyzing, etc .
These challenges are discussed in the following text .
1 Characters and techniques used for
studying EcM fungal communities
1 .1 Morphology-visible characters for sorting
EcM fungi
Morphological features aregenerally the first piec-
es of information weusefor macro-fungi identifications .
These features can bevery easy to apply to and require
little equipment and investment . Analysis of different
morphotypes in the root tips can also be used to sort
different groups of EcM fungi . This is because certain
mycorrhizae contain signature attributes shared by a
specific group (s) of fungi . However, more often than
not, morphological features of mycorrizhae are ambigu-
ous, especially among closely related species, making
it hard to resolve unknowns to different species or even
genera .
One common method to identify EcM fungi is to
analyze morphological features of the fruiting bodies,
relying on traditional fungal taxonomy that arebasedon
characteristics of sexual reproductive structures . Com-
monly seen macroscopic characteristics of mycorrhizal
fruiting bodies include size, shape, color of the cap,
stem, flesh, gill , and concentric rings, etc; micros-
copic features include spore size, shape, and hyphae
type, as well as the arrangements of spores and hyphae
within the fruitingbodies etc . Consistent differences in
these featureshavebeen used to identify unknowns, of-
ten down to thegenus level . However, identificationof
5913 期 LI and XU: Molecular Ecology of Ectomycorrhizal Fungi : Molecular Markers, Genets and Ecological . . .
individual species usually requires more effort, since
many morphological features are undistinguishable be-
tween closely related species . The presence of juices
upon breaking, bruising reactions, spore prints are
considered as additional methods to sort species as well
as identify unknowns .
When conducting population level analyses, mac-
roscopic morphological features are extremely useful for
sorting distinctively different specimens . However,
sometimes microscopic morphological sorting could be-
come impractical when analyzing a large number of
samples . In addition, if morphological analysis takes
too long, the DNA in the samples may becomedegrad-
ed, resulting in failure in the subsequent molecular
analyses . There are drawbacks in using morphological
features alone for EcM fungi identification . Specifical-
ly, while macroscopic morphology can be used to sort
fungi into discretegroups rapidly, it is often not suffi-
cient enough to differentiate closely related species .
Often, molecular analyses reveal multiple reproductive-
ly isolated cryptic species within morphological species
(Bidochka et al. , 2005; Dettman et al. , 2003; Geml
et al. , 2003; Kauserud et al. , 2006 , 2007; Taylor et
al. , 2000) . Furthermore, convergent evolution of cer-
tain trait features amongunrelated species could lead to
misidentification, when only morphological features are
used .
Early mycologists developed many valuable mor-
phological identification keys for fungal taxonomy .
Someof thesefeatureswere later compiled and analyzed
for their usefulness . For example, in the analyses by
Luoma et al. , 1997 , 200 morphological types of ecto-
mycorrhizal truffles and mushrooms from 189 soil cores
were provided that included detailed descriptions for
eachmorphotype ( Luoma et al. , 1997 ) . Unfortunate-
ly , unknown samples’ classification remains unclear
due to a lack of report of their specific morphologies .
1 . 2 Molecular methods-fine scale sorting of EcM
fungi
Due to the limitations in morphological analyses,
molecular analyses have become increasingly common
in fungal taxonomic and ecological research, including
research into EcM fungi . In certain cases, morphoty-
ping may be skipped entirely . The emerging field of
metagenomics analyzes DNA samples directly from the
environment, including both cultivable and un-cultiva-
ble ones ( for a review in this area, see Xu 2006 ) .
Thesedevelopments are allowing us unprecedented ac-
cess to microorganisms in nature . Below we describe
and discuss some of the common molecular methods in
EcM research .
1.2.1 Restriction FragmentLengthPolymorphisms (RFLP)
RFLP is one of the most popular molecular meth-
ods to discriminate species or strainsof fungi . It usual-
ly involves using restriction enzymes to digest genomic
DNA , and analyzing the resulting patterns, with or
without specific probes . Although sequencing can char-
acterize DNA more thoroughly, RFLP analysis has been
very popular, especially during the early years of ap-
plyingmolecular markers to fungal studies . Depending
on the specific DNA fragment and restriction enzyme
combination, RFLP can cluster unknown specimens in-
to differentgroups, sometimes to the species level . The
early success of RFLP includes low cost, fast and effi-
cient . In whole genome digestion, RFLP cannot be
used to detect polymorphisms for low copy number ge-
netic elements . However, polymorphisms within high
copy number genetic elements, for instance ribosomal
DNA gene and mitochondrial DNA , can be detected
using RFLP (Xu, 2005) . However, there are disad-
vantages associated with RFLP . For example, when the
number of DNA bands is high, it is hard to distinguish
two bands with similar migration abilitieson thegel . In
order to identify individual bands, specific labeled
probes need to be used to recognize unknown bands
through DNA-DNA hybridization (Xu, 2005 ) . Alter-
natively, specific DNA fragments can be amplified
using highly selective primers through PCR and the
PCR products can be then digested, and the resulting
patterns compared side-by-side . Because of the highly
conserved nature of the ribosomal RNA gene clusters,
there is generally little or no difference among strains
within the same species for this gene region . As a re-
sult, different restriction banding patterns for this gene
region are usually indicative that the analyzed strains
areof different species .
691 云 南 植 物 研 究 31 卷
The main shortcoming of the RFLP typing is that
when it is used alone, some samples cannot be success-
fully distinguished . In addition, there areseveral issues
associated with RFLP patternmatching . Firstly, theda-
tabases are almost exclusively constructed using sporo-
carp samples but rarely nonsporocarp samples ( i . e . in
our case theEcM mycelia) . The second problemis ac-
curacy of RFLP pattern matching . For example, in the
casewherewe could match a specific RFLP pattern to a
species in theRFLP database, theremight be somemi-
nor variations in fragment sizes that we fail to detect .
Factors such as the choiceof primers, the types of res-
triction enzymes, homogeneity of the gel , variation in
electrical current can also affect the fragments’migra-
tions through the gel matrix .Thirdly, RFLP is sensitive
to intraspecific genetic variation due tosingle nucleotide
polymorphism (SNP) . While such polymorphisms can
beuseful for strain typing, they can also confound spe-
cies identification in the absence of a robust database .
Lastly, typical RFLP databases are limited in scope,
often by the investigator′s personal interests . Large im-
provement inmany aspects such as increasing thesample
size, standardizingthe useof restriction enzymes, prim-
ers and intrinsic conditions of gel matrix to allow cross-
lab comparisons . Oneof themost commonly used RFLP
typinggenomic region is the internal transcribed spacer
(ITS) region of the ribosomal RNA gene cluster . Both
species and strain-specific ITS-RFLP patterns may be
identified . As morphotyping and RFLP typing are be-
coming more inclusive, the integration of these two
types of information could significantly enhance our
ability for species and strain identifications .
1.2.2 PCR based molecular method———fast and accurate
To complement the shortcomings of the morpholo-
gy-based approach and RFLP method, rapid PCR-
basedmolecular analyses are often adopted and imple-
mented in many fungal genetics research . PCR allows
amplification of specific DNA fragments with either
specific or non-specific primers that recognizes defined
regions of the genomes (Mullis and Faloona, 1987 ) .
Because it requires very little genetic material and is
typically gene specific in its amplification, the PCR
methodology provides tremendous advantages over tradi-
tional molecular typing techniques .Thelarge amountof
products obtained in a typical PCR reaction allows sci-
entists to conduct further analyses, including obtaining
sequence information of the target genes . With se-
quence information inhand, many analyses can be per-
formed including database searching and unknown-tar-
get mapping, sequence comparison and phylogenetic
analysis . In EcM fungal community research, PCR has
been used primarily for identification . However, other
types of information, including the relative abundance
of individual species are also possible from analyzing
PCR products . For example, the ITS regions can be
amplified through PCR from an ectomycorrhizal fungal
community . The PCR products would contain amixture
of ITS sequences from different fungal species in the
community . The PCR products can be directly se-
quenced using the pyrosequencing technique or cloned
first into ahost bacteriumand then sequenced individu-
ally through conventional sequencing techniques . The
obtained sequences can be analyzed that allow calcula-
tions of relative abundances of individual ITS sequence
types (Xu, 2006) .
1 . 2 .3 DNA sequencing—the finest scale taxonomic
identification
Sequencing one or several DNA markers is the
most sensitiveand robustway to identify species . There
are many advantages of using DNA sequence based
analysis for species and strain identifications . First,
DNA sequences are unambiguous and some regions are
highly conservedwithinspecies and can beused as mo-
lecular markers to characterize unknowns (Xu, 2005) .
Second, compared to morphological features, which
represent individual phenotypes, DNA sequences make
up genotypes at avery finescale . Compared to pattern-
based typing ( RFLP, PCR fingerprinting) , DNA se-
quence based typing can givemore accurate and robust
results (Xu, 2005 ) . Third, DNA sequences can be
stored in and retrieved frompublic databases . Suchda-
tabases can serve as valuable resources for searching
and editing ( Xu, 2005 ) . It also allows us to do fine
scale genome typing, for example, intraspecific strain
typing . Indeed, DNA-sequence based approaches are
the future of EcM research .
7913 期 LI and XU: Molecular Ecology of Ectomycorrhizal Fungi : Molecular Markers, Genets and Ecological . . .
1 .2 .4 ITS—an excellentmolecular marker to identify
species
The ITS region is asectionof DNA located within
the nuclear ribosomal RNA gene cluster ( Fig. 1) . Each
unit of the gene cluster comprises one small subunit
18S rRNA , a large subunit 28S rRNA , and a 5 .8 S
rRNA flanked by two non-coding DNA sequences
known as ITS1 and ITS2 on either side . The combined
length of ITS regions ( including the 5 .8S rRNA ) in
fungi is typically between 650 - 900 bp . Within an in-
dividual cell , theremay be 50 - 200 copies of this unit
cluster, linked by an intergenic sequence ( IGS) that
can be highly variable in length and sequence composi-
tion among species .
During the past two decades, the ITS region has
become the most commonly used molecular marker in
fungal ecology and systematic research ( Egger et al. ,
1995; Chambers et al. , 1998; McKendrick et al. ,
2000) . Its utility and popularity are due to the follow-
ing reasons . First, it is present within the high copy
number ribosomal RNA gene cluster as tandemrepeats .
As a result, a small number of cells may be sufficient
for PCR and subsequent analyses . Second, it is highly
conserved within species but can behighly variablebe-
tween closely related species . Third, the ITS regions
are relatively short and flanked by highly conserved se-
quences . Consequently, conserved primers applicable
to broad groups of species can easily be designed and
used to amplify the ITS regions by PCR . Two primers,
ITS1 and ITS4 , have been used widely in fungal sys-
tematics and population genetic studies . These two
primers were originally designed from plant sequences
and are used as universal primers across all fungi
(White et al. , 1990 , Gardes et al. , 1991) .
Aside from ITS1 and ITS4 , other taxa-specific
primers have also been developed . For example, ITS1f
and ITS4 are fungal specific; ITS1f and ITS4b are ba-
sidiomycete specific (Gardes andBruns, 1993) . These
group specific primers are desirable in terms of enhan-
cing specificity of onegroup and discriminating against
other fungal groups . For instance, primer ITS4b has
been designed to amplify sequences from all known
Fig . 1 A combined approach to identify unknown EcM fungi using ribosomal RNA genes to different taxonomic levels
891 云 南 植 物 研 究 31 卷
basidiomycetes, whileexcluding fungi outsidethegroup
(Taylor et al. , 1999 ) . Nevertheless, even though
ITS1f?ITS4b are specifically designed ITS primers for
basidiomycetes, we do not expect that theywill amplify
every species within this group . This is because muta-
tions could be present at the primer annealing sites for
some potentially unknown basidiomycetes in nature .
Indeed, non-amplified individuals with the above prim-
ers have been identified in previous experiments using
other techniques . Rhizoctonia is one of the groups .
Some species in the Rhizoctonia group cannot be am-
plified by the ITS1f?ITS4b primer pair, even though it
belongs to basidiomycetes ( Robinson et al. , 2009 ) .
Indeed, due to their unknown nature inour databases,
group-specific primers based on known taxa in that
group will alwayshave its limitations when workingwith
unknown samples . One way to overcome the difficulty
faced by group-specific primers is to use universal
primers such as ITS1?ITS4 to reanalyze non-amplified
species and to confirmthe results . Bearing these cave-
ats in mind, it′s important to realize that care should
be taken when interpreting the results .
The ITS region is currently the most popular DNA
marker used in RFLP, especially for studying EcM fun-
gal diversity . Matching the polymorphisms at the ITS re-
gions has successfully separated many EcM fungal spe-
cies ( Egger et al. , 1995, K?rén et al. , 1997; Pritsch
et al. , 1997) . Indeed, ITS-RFLP has turned out to be
themost convenient molecular method tosort fungal spe-
cies with minimal technical requirements . As mentioned
above, thereareseveral advantages associatedwith ITS .
In combination with PCR and RFLP, ITS PCR-RFLP
has allowedgenerating species-specific banding patterns
and making it possible to classify EcM fungi at thespe-
cies level . This is because the ITS region usually con-
tains restriction sites where common restriction enzymes
could target . Commonly at least 2 to 3 frequent cutting
restriction enzymes areneeded for species identification .
Since sequence differences between species can be due
to both insertions?deletions and nucleotide substitutions,
different restriction enzymes may produce very different
patterns . If several ITS PCR-RFLP patterns are associ-
atedwith a single morphotype, additional samples with
the same morphortype need to be selected and verified
by RFLP . If such patterns are confirmed, taxonomic re-
visionsmay be necessary .
Nowadays, sequencing ITS regions has been rou-
tinely used to identify fungal species, and to study ge-
netic diversity among different strains within one spe-
cies . The ITS region is now perhaps the most widely
sequenced DNA region in fungi ( Bruns et al. , 1998;
Pritsch et al. , 2000; Sha et al. , 2008; Xu, 2005 ) .
A large number of ITS sequences have been stored in
GenBank . It has become avaluable resource for identi-
fying unknown fungal species and for investigating the
evolutionary relationships among fungal species ( Bruns
et al. , 1998; See also below for the UNITE data-
base) . Because of these and the previously mentioned
features, the ITS regions have been adopted by the in-
ternational mycological community as the barcode region
for fungal identifications .
Other than the ITS sequences, additional molecu-
lar markers such as LSU rRNA , mtLSU , SSU rRNA ,
5 . 8S nuclear rRNA are alsowidely used to discriminate
fungal groups (Table 2) . Due to the difference in the
degreeof resolution, unknowns can be placed into dif-
ferent taxonomic levels using different rRNA gene frag-
ments (Table 2 ) . The choice of which marker to use
depends on the experimental goal , feasibility of PCR
amplification of each region, and the availability of the
sequence databaseof that particular marker for compar-
ison and analyses . In many investigations, fine scale
species and strain identifications are the final objec-
tives . However, sometimes tracing the specimens down
tofamily levels canprovidesufficient information to an-
swer certainphylogenetic questions . Species andstrain-
level identificationsoften require large databases ( large
taxa sets and multiple genes) to compare . Sometimes,
even with large databases, theremight be no matching
with those in databases . Indeed, it is not uncommon
that the ITS-based molecular identification can only
reach the genus level ( Gardes and Bruns, 1996b;
Ciardo et al. , 2006) .
When considering markers in addressing specific
research questions, we need to realize that there is al-
ways a trade-off between cost and effectiveness in the
9913 期 LI and XU: Molecular Ecology of Ectomycorrhizal Fungi : Molecular Markers, Genets and Ecological . . .
Table 2 Comparison of molecular markers within the ribosomal RNA gene cluster used for identification of unknown EcM fungi
Advantage Disadvantage Popularity
SSU rRNA
Most conserved, excellent for high lev-
el taxonomic investigations
Too conserved for low taxonomic level
investigations
Heavily used for identifying bacterial
and archaeal diversity, but rarely for
studying EcM fungi
LSU rRNA
More variable than SSU rRNA , can
place to generic level
Database is limited
Few EcM studies used this gene frag-
ment
5 ?. 8S rRNA
Very conserved, but can help resolve
to phylum level
Sequences too short Usually co-analyzedwith ITS sequences
mtLSU rRNA
Somewhat variable, can help place to
family level
Can have introns Some studies used this gene fragment
ITS
Highly variable, extensive database,
can place sample to species level
Too variable for high level taxonomic
analyses and sometimes sequence align-
ment is difficult
Extensively used and a large database
already available
type and number of markers selected for investigating
certain issues . Relatively sensitivemarkers such as the
ITS region works well for identification to the species
level . However, for genotype-specific fine-scale identi-
fications, other types of markers ( e.g . single copy
genes or inter-genic regions) are needed . With the in-
creasing availability of candidate molecular markers,
the best-fit candidate marker ( s) may require prior
screening for individual species before we can deter-
mine their appropriateness . Conversely, a less sensitive
marker (e. g . the SSU rRNA gene) does a cruder sort-
ing, butmapping and identification steps are easier and
more accurate, since fewer samples may be needed to
take into account . Another shortcoming for the ITS re-
gion is its high copy number, which makes it hard to
distinguish whether heterozygosity is resulted by differ-
ent copiesof ITS onone chromosomeoron two different
chromosomes ( in diploid or dikaryotic individuals) .
Therefore it is hard to identify genotypes for individual
alleles when the unknown is heterozygous .
1 . 2 .5 Single-gene based phylogeny may not truly re-
veal organism phylogeny
The main drawback of single-gene typing is that it
may not truly represent organism′s phylogeny (Doyle,
1992; Maddison, 1997) . For example, using a single
gene as molecular marker to study organism phylogeny
can behighly biased, since thegene chosen for analysis
may evolve fast in some groups but relatively slow in
other groups . Two species mapped together by a single
marker could be actually very different from each other
if analyzed with other genes . Simply speaking, each
singlegene has its intrinsic evolutionary bias . Oneway
to minimize the bias is to usemultiplemolecular mar-
kers . Usingmultiplegenes to study phylogeny is always
more reliable than using only one gene since it is more
representative for the entire genome ( Xu et al. ,
2006) .
2 Common approaches for examining spe-
cies diversity within and among EcM com-
munities
If the research focus was to examine the diversity
of EcM fungi within a selectedgeographical area, com-
bining morphology and PCR-RFLP of the ITS region
should initially provide a rough grouping among a large
number of fungal isolates . For individuals with nearly
identical PCR-RFLP patterns, they can be further
characterized by sequencing the ITS or other particular
DNA markers .
If the research aimwas to study one type of EcM
fungi within a selected area, morphology or PCR-RFLP
of the ITS region can be used as an initial step to iden-
tify the target EcM fungi . Further identification and
confirmation can be achieved by PCR amplification and
sequencing of marker gene fragments at other loci . Af-
terobtaining thesequenceinformation, blast search can
be carried out in order to determine the sequence iden-
tity with known specimens in databases . If the un-
known specimens have high sequence similarities (us-
ually > 97% for ITS regions) to sequences of known
species deposited in the GenBank, the identity of the
002 云 南 植 物 研 究 31 卷
unknowns can be inferred . However, if a high degree
of sequence identity is not found, individual sequences
are then used as a query to retrieve closely related se-
quences with comparable length from the GenBank or
other publicly available databases . Representative se-
quences for each closely related species fromGenBank
were then included to compare potential intra-specific
variation within different phylogenetic groups to known
species . These sequences and all the retrieved se-
quences were then aligned using appropriate computer
softwares . Subsequently, phylogenetic analyses apply-
ing different algorisms are used to reveal the phyloge-
netic relationship amongthe species . Usually, themost
closely related known species are used as outgroups for
references . Sometimes unknown fungi cannot be easily
identified by examining a single gene . Sequence infor-
mation of additional molecular markers couldbehelpful
to identify unknowns . Since different genes evolve at
different rates, multiple gene genealogy can limit the
biases created by singlegenes and reveal the true phy-
logenetic relationships among species . In many cases,
the relationships among strains and populations within
species can also be revealedusing sequences frommul-
tiplegenes . Such sequence information can be used to
determine thepotential mode of reproduction and geog-
raphic patterns of molecular variations ( e.g . Lan and
Xu, 2006) .
While GenBank, DNA Data Bank of Japan (DD-
BJ ) , and the European Molecular Biology Library
( EMBL) databases contain almost all publicly deposit-
ed DNA sequences that are cross-referenced with each
other through unique accession numbers, the ectomy-
corrhizal research community has been benefited tre-
mendously by another database called UNITE ( http:??
unite. ut. ee?) . UNITE is an rDNA sequence database
focused specifically on ectomycorrhizal fungi in two
phyla, the ascomycetes and the basidiomycetes . The
database was established because of widespread se-
quence misidentifications and inaccurate reporting in
GenBank, DDBJ and EMBL . To establish an accurate
and reliable database, sequences in the UNITE data-
base aregenerated fromfruit bodies collected and iden-
tified by specialists and deposited in public herbaria
that can beaccessed by others . In addition, type spe-
cimens are used whenever possible . As of April 2009 ,
the UNITE database contains 112363 fungal ITS se-
quences . Among these, 2736 were barcoding sequenc-
es from1202 species in 128 genera . Aside fromprovi-
ding a robust database for identification of sequences
from curated specimens, UNITE also has other features
that facilitate the identification of fungal DNA directly
fromenvironmental sources ( i . e . DNA sequences with-
out specific specimens attached) . This search tool is
becoming increasingly important because direct me-
tagenomic analysis of environmental DNA is becoming
increasingly common in EcM research .
2 . 1 Bioinformatic analyses of EcM fungi
Based on the aligned sequences, the relationships
among sequences, strains, and?or species can be re-
vealed . Themost common formof presentation for such
relationships is through phylogenetic trees . Phylogene-
tic trees can be constructed usingvarious algorisms im-
plemented indifferent phylogenetic softwares . Themost
commonly used ones are maximum parsimony, neigh-
bor-joining, maximum likelihood and Bayesian ap-
proaches . The trees generated by applyingdifferent al-
gorisms areusually consistent with each other . Howev-
er , in certain cases, they can be different from each
other . The different patterns can be generated for a
couple of reasons . One is the weighting scheme of
polymorphic nucleotides . For example, the weights of
transitional substitutions and transversional ones can
have a significant effect on the final outcome of the
analyses . The same can be said about insertions and
deletions . Indeed, dozens of weighting methods have
been developed to try to reflect the relative importance
of different types of mutations during the evolution of
specific lineages .
The second reason for different tree topologies
generated by different methods using the same dataset
relate to the differences in algorithms among the phylo-
genetic tree-construction methods . Different algorisms
process data in different ways . The neighbor-joining
method measures the genetic distance between each
pair of taxa and joins taxa with the shortest distance
first, followed by progressively more distantly-related
1023 期 LI and XU: Molecular Ecology of Ectomycorrhizal Fungi : Molecular Markers, Genets and Ecological . . .
taxa ( Saitou and Nei , 1987 ) ; maximum parsimony
produces the most preferred phylogenetic tree invoking
the fewest number of evolutionary changes ( Felsen-
stein, 1978) ; maximumlikelihoodmethod searches for
the tree with the highest probability or likelihood that
matches the data (Fisher, 1978) . Sometimes tree to-
pologies are divergent at a great extent, making inter-
pretation of phylogeny challenging . When such cases
occur, the preliminary data can be used to generate
specific hypotheses . Targeted additional sequence in-
formation can be then collected to test the hypotheses .
In general , the greater the sample sizes and the more
genes analyzed, the more robust the inferred evolutio-
nary relationships among strains and species will be .
In conclusion, a combinedmorphological and mo-
lecular approach is generally used to address issues re-
lated tospecies diversity in EcM research . Morphologi-
cal grouping is typically carried out at the initial stage
of identification . Usually EcM fungi from each soil
sample are classified into as many groups as possible
based on their morphological characteristics . The mor-
phological characterization can be then followed by ITS
PCR-RFLP banding pattern matching with an estab-
lished reference database . However, for fungi remain-
ing unknown after comparison to PCR-RFLP database,
DNA sequencing and phylogenetic analysis of ITS and
various other molecular markers are needed . Conse-
quently, the additional data can then beused to update
the ITS-RFLP database . If more than oneRFLP pattern
is found associated with asinglemorphotype, addition-
al samples with the same morphotype are further ana-
lyzed to exclude the possibilities of contamination and
potential heterozygosity within individuals for the ITS
regions ( Horton and Bruns, 2001) .
3 Common approaches for examining in-
traspecific variation among EcM fungi
Studies involving the examination of genetic varia-
tion within EcM fungal populations typically rely on
multilocus polymorphisms using several types of mar-
kers, including single-copy gene based RFLP, PCR
fingerprinting, DNA sequencing, and microsatellite
DNA . These markers have been described in detail in
an earlier review ( e.g . Xu, 2005) . Here, we briefly
mention microsatellite markers . Microsatellite markers
are those with variation in the number of simple se-
quence repeats within a DNA fragment that can be
found among individuals within species . Because of
their repetitivenature, they typically mutatemuch fast-
er than singlenucleotide substitutions, due to high fre-
quency slippage during DNA replication . The high
variability of microsatellite DNA makes them excellent
molecular makers for strain typing and population ana-
lysis, especially for recently evolved populations . To-
gether with the increasingly popular single nucleotide
polymorphism, microsatellite DNA are helping scien-
tists working on EcM fungi to determine species boun-
dary, population structure, and reproductive strategy
etc . Belowwereview recent studies on one specific is-
sue, that of genet size of EcM fungi , in natural envi-
ronments .
3 . 1 Genet-size and distribution
A genet is a group of individuals ( fruiting bodies
and?or underground mycelia for EcM fungi ) produced
from onemating event and that occupy the same geog-
raphic area (Dahlberg andStenlid, 1994; Xu, 2005) .
Bearing a small number of mutations, the individuals
within a genet should be all genetically identical . Be-
cause they arise vegetatively from a single mating
event, this shareddescent among the individual fruiting
bodies and mycelia implies connectivity both nutrition-
ally and?or genetically . Nutritional connectivity refers
to the nutrients absorbed by mycelia from one location
of the genet can be transported to another parts of the
genet . Similarly, genetic connectivitymeans that muta-
tions and horizontally transferred genes obtained in one
areaof the genet can be spread to other parts of the
genet . At present, though neither connectivity has
been demonstrated conclusively innature, the existence
of genets of various sizes suggests their likely impor-
tance in nature .
Significant research activities have been devoted
to characterize the sizeof genets in EcM fungi . A vari-
ety of molecular markers mentioned above have been
used to determine whether individual fruiting bodies or
myceliaof one species from a defined geographic area
202 云 南 植 物 研 究 31 卷
are genetically identical with each other but different
from other individualswithin thegeneral region . Dueto
their high mutation rates, the polymorphisms at multi-
ple microsatellitemarkers have been used extensively to
robustly identify the sizeof individual genets . Typical-
ly, a single genet is identified by a unique microsatel-
lite polymorphic type for individuals sampled at a de-
fined geographic area . Similarly, genetic identity in-
ferred usingother typesof markers such as randomamp-
lified polymorphic DNA ( RAPD) is also widely used,
especially during the early phase of ectomycorrhizal re-
search .
The current data suggest that the size of genets
varywidely among fungi . Small genets in species of the
ectomycorrhizal genus Russula can be less than half a
meter . In contrast, large genets have been found
among plant pathogenic basidiomycete genus Armillar-
ia . For example, in A. gallica, somegenets canoccu-
py up to 2 , 200 acres (15 ha) . For many fungi , spo-
rocarps belonging to one genet usually are spatially ar-
ranged as a ring, known as a fairy ring . The fairy ring
structure is generated from a single mating event . As
mycelia grow outwards, the genet expands . Further-
more, if the spatial environmental conditions are rela-
tively homogeneous, the genet expands at similar rates
in all directions across the terrain . At certain times of
theyear when conditions favorable for mushroomfruit-
ing appear, fruitingbodies areproduced along the edge
of the mycelial growth, forming a ring structure . Be-
causemycelial growth depletes nutrients in the centerof
the ring, the size of the ring also expands over time .
One striking featureof EcM fairy rings is the low spe-
cies richness along the rings, dominated by the ring-
forming species . However, fungal diversity can behigh
on both sidesof the rings, with similar species diversi-
ties and compositions ( Lian et al. , 2006; Hirose et
al. , 2004 ) . This fact suggests that EcM communities
are capable of recovering soon after genet passage .
However, questions such as how recovery is established
soon after passage and how the dominance is achieved
by the ring-forming species are still unclear .
Current molecular ecological investigations suggest
significant variation among EcM fungi in their genet si-
zes . Table 3 summarizes the genet sizes of recently in-
vestigated EcM basidiomycete fungi . Genet size is typi-
cally measured in the largest distance between sporo-
carps that areproduced by onemating event . As shown
in Table 3 , genet sizes vary greatly among species .
Laccaria spp . and Russula brevipes have so far the
smallest identified genets of less than 1 m in diameter
( Baar et al. , 1994; Bergemann and Miller, 2002;
Selosse et al. , 1999) . Genetsof Tricholoma matsutake
and Suillus grevillei are less than 3 m in diameter
(Lian et al. , 2006; Zhou et al. , 2001) . Genet sizes
Table 3 Comparison of genet properties among EcM species . Genet size is measured based on the
above-ground fruit body sampling .“ - ”indicates missing data
EcM species Genet size
Estimated expansion
rates (cm?year)
Early or
late stage ?
References
Laccaria amethystina < 1 fm2 60 1- 110 Early Gherbi et al ?. , 1999; Selosse et al. , 1999
Laccaria bicolor < 12 ?. 5 m2 220 D- 100 Early Baar et al. , 1994 ; Selosse et al \. , 1999
Pisolithus tinctorius < 40 ?m - Early Anderson et al 0. , 1998
Hebeloma cylindrosporum < 3 f. 5 m 52 ^. 5 Early Gryta et al ?. , 1997, 2000; Guidot et al., 2001
Suillus pictus 3 ?. 4 - 21 m 50 Both Hirose et al. , 2004
Suillus variegatus 29 9m Both Dahlberg, 1997 S
Suillus pungens 40 +m (maximum 300 m2 ) 50 ?Both Bonello et al. , 1998
Suillus grevillei 3 ?m - Both Zhou et al. , 1999
Suillus bovinu 3 ?- 30 m - Both Dahlberg and Stenlid, 1990
Russula brevipes < 3 fm - Late Bergemann and Miller, 2002 ?
Russula cremoricolor 0 ?. 38 - 1 .27 m - Late Redecker et al 1. , 2001
Russula vinosa < 1 fm - Late Liang et al ?. , 2004b
Amanita francheti 1 ?. 5 m2 - Late Redecker et al 1. , 2001
Cortinarius rotundisporus 30 9m - Late Sawyer et al ?. , 1999
Cantharellus formosus 2 ?- 13 m - Late Dunham et al. , 2003
Lactarius xanthogalactus 9 ?. 3 m2 - Late Redecker et al 1. , 2001
Tricholoma matsutake 2 ?m 10 ^. 3 Late Lian et al. , 2006
3023 期 LI and XU: Molecular Ecology of Ectomycorrhizal Fungi : Molecular Markers, Genets and Ecological . . .
greater than 20 m are found in several species of the
genus Suillus (Hirose et al. , 2004) . To estimate gen-
et expansion rate, spatial distribution of sporocarps
within the same genet is typically monitored through
several years . The results from previous studies show
that genet expansion rates also vary amongspecies . For
example, Tricholoma matsutake expands relatively slowly
with an average rate of 10 .3 cm?yr, while Hebeloma
cylindrosporumhas amuch higher rate of 45 - 60 cm?yr
(Gryta et al. , 2000; Lian et al. , 2006 ) . Some of
Laccaria spp . genets have extremely high expansion
rates, around 100 cm?yr (Gryta et al. , 1997 , 2000;
Bonello et al. , 1998; Guidot et al. , 2001) .Taken to-
gether, these estimates suggest that EcM genets can
live up to about 30 - 40 years in natural ecosystems .
3 . 2 Early-stage versus late-stage EcM fungi
Typical natural ecological communities, including
theEcM communities, are established through a series
of changes of biological compositions and abiotic fac-
tors . The process of ecological change involving a se-
ries of natural communities that are established and re-
placed over time is called a succession . There are two
kinds of ecological successions, primary succession and
secondary succession . Primary succession occurs in an
environment in which new substrates, devoid of any
living organism and usually lacking soil , is deposited
(for example a lava flow) and allows the establishment
of an ecological community . In primary succession, pi-
oneer species likemosses, lichen, algae and fungi first
colonize the substrate . Together with abiotic factors
such aswind, water and theheat?cold cycle, thesepio-
neer species changethe substrates, making them sui-
table for subsequent growth of plants . The plants then
dominate but often replaced successively by plants bet-
ter adapted to less austere conditions . Examplesof pri-
mary succession can be found on a new lava flow, an
area left from retreated glacier, or abandoned strip
mine .
In contrast, secondary succession is a response to
a disturbance, for example, forest fire, tsunami , hur-
ricane, flood, or an abandoned field . While primary
succession takesplace in an areathat isoriginallywith-
out any livingorganism, secondary succession occurs in
an area where lifeonce existed but has sincebeen des-
troyed or disturbed, for exampleby fire, tornado, har-
vesting or other human activities such as agriculture,
that reduces an already established ecosystem ( e.g . a
forest or a wheat field) to a smaller population of spe-
cies . As such, secondary successionoccurson preexis-
ting soil , unlike primary succession that usually occurs
in a place lacking soil .
EcM fungi can be found associated with both pri-
mary and secondary successions . EcM fungi associated
with primary succession plants are called early-stage
EcM fungi . Sometypical earlycolonizingfungal species
are Hebeloma cylindrosporum ( Gryta et al. , 1997 ,
2000) , Laccaria bicolor ( Baar et al. , 1994 ) and
Pisolithus tinctorius ( Anderson et al. , 1998 ) . EcM
fungi found in secondary successions are called late-
stage EcM fungi . Typical EcM fungi in late-stage suc-
cession are those in families Russulaceae, Cortinariaceae
and Amanitaceae . Early colonizing fungal species typi-
cally colonize by spores . Because primary succession
niches are typically poor in nutrients these early EcM
fungi are expected to have relatively small and non-per-
sistent genets (Deacon et al. , 1983; Fox 1983) . Ear-
ly-stage fungi are considered as R-selectedspecies, be-
cause they produce many offspring and are capable of
filling available niches in an environment very quickly
(Deacon and Fleming, 1992 ) . In contrast, late-stage
EcM fungi may also colonize initially by spores or by
dormant mycelia underground, but because of nutrient
availability from the soil , they are able to spread by
mycelial growth and expand . As a result, their genets
are expected to be large and temporally persistent
(Dahlberg and Stenlid, 1990 ) . Late-stage EcM fungi
may be considered K-selected species, more capableof
dealing with environmental stresses ( Cooke and
Rayner, 1984; Grime et al. , 1979 ) . While the typi-
cal late-stageEcM fungi in genera Russula, Amanita,
Lactarius, are expected to propagate by mycelial
growth, researchers have found that their genet sizes
and age vary significantly, suggesting that proliferation
by sexual spores in theselate-stage fungi might bemore
important than previously expected ( Bergemann and
Miller, 2002) . Even inmature plant communitieswith
402 云 南 植 物 研 究 31 卷
relatively homogeneous ecological conditions, thegenet
sizes can vary significantly, suggesting reproduction by
spores is a prominent feature of late-stage EcM fungi .
Thus, the appearanceof aspecies insecondary succes-
sion communities cannot be used to draw conclusions
about their genet size and modes of reproduction and
colonization (Bonello et al. , 1998) .
3 . 3 Factors affecting the reproductive modes and
persistence of mycelia network
The mode of reproduction for EcM fungi is typi-
cally inferred based on the associations of alleles at the
sameor different loci . Populations with alleles random-
ly associating with each other are considered sexually
reproducingpopulations whilethosewith significant sig-
natures of non-randomassociations are considered asex-
ual populations (Xu, 2005 ) . Another indicator in EcM
fungi about the relative importanceof sexual and asexu-
al reproduction is the sizeof genets . By examining the
genet sizes of EcM fungal species in nature, the repro-
ductive strategies of individual species may be predict-
ed . It has been suggested that species forming many
small genets likely reproducemainly by spore coloniza-
tion and the species that formlarge genets mainly pro-
pagateby undergroundmycelial extension (Anderson et
al. , 1998; Bonello et al. , 1998; Dahlberg and Sten-
lid, 1990 , 1994; Dahlberg, 1997; Zhou et al. ,
1999) . Most species use a combination of both strate-
gies . For example, Suillus spp . use amixed strategy:
they form many small genets by spore colonization at
the early stage of lifecycle, while then produce large
genets by mycelia expansion at late stages ( Dahlberg
and Stenlid, 1990 ) .
Studies of Suillus spp . also revealed that the size
of genets is correlated with age of host stands (Dahl-
berg and Stenlid, 1994 ) . Small genets are mainly
found in young-aged stands, while large genets are
found in mature stands . However, exceptions have
been found in Lactariusxanthogalctusand Russula cre-
moricolor , which formmany small genets inmature tree
forests ( Redecker et al. , 2001) . The age of stands is
thus not sufficient to predict colonization strategies
across all EcM species . The spatial pattern of EcM
fungi also depends on many other factors, such as ( a)
environmental physical parameters, e.g . temperature,
moisture, and light; ( b) edaphic factors, e.g . soil
moisture, depthof organic matter, andsoil pH ( Erland
and Taylor, 2002) ; ( c) biological traits of the fungi;
and ( d) competition with other fungal species (Ander-
son et al. , 1998; Dahlberg and Stenlid, 1990 , 1994;
Dhlberg, 1997) . Guidot et al . demonstrated that He-
beloma cylindroporum forms large genets under condi-
tions with low competition with their neighbor fungi
(Guidot et al. , 2001 ) . Conversely, small genets are
observedunder high intensities of competition . The ca-
pacity to survive and expand for a longperiodof timein
the soil asgenerativemyceliaand the ability to colonize
using spores differ greatly among species . In general ,
in thelaboratory settings, fungi tend to reproducesexu-
ally when environmental conditions are unfavorable,
and reproduce asexually when environmental conditions
become favorable . The fruiting seasons for many EcM
mushrooms are associated with theonset of stressful en-
vironmental conditions, consistent with laboratory ob-
servations .
Genet expansion rates, also known asmycelial ex-
pansion ratevary amongspecies in agreat range from10
cm?yr to 110 cm?yr (Table 3) . The variation could be
due to a combination of genetic and environmental fac-
tors, such as mentioned above ( Bonello et al. , 1998;
Lian et al. , 2006) . In thestudiesof most EcM species,
genet expansion rates are relatively constant within spe-
cies through several years of growth ( Bonello et al. ,
1998; Dahlberg and Stenlid, 1994; Lian et al. , 2006;
Zhou et al. , 1999 ) . While EcM genet establishment
and growth rates are known to be affected by environ-
mental factors, however, the specific mechanisms rema-
in largely unknown . Future research should be focusing
on monitoringgenet progression invarious environments .
The ability of EcM mycelia to expand is a unique
characteristic for filamentous fungi , therefore studying
genet size can understand the ecological features of
these species in nature . The characteristics of EcM
genet can be used as indicators of succession stage of
forests, individual population boundary, and the
required area of study site in EcM research ( Liang et
al. , 2004a) . It can also help understand the roles of
5023 期 LI and XU: Molecular Ecology of Ectomycorrhizal Fungi : Molecular Markers, Genets and Ecological . . .
spore dispersal and mycelial growth in the life histories
of EcM fungi . For example, as a late-stageEcM fungi ,
species in Russula maintain reproduction by sexual
spores even in suitable environmental conditions with
minimumcompetitors, indicating that spore colonization
plays a much more important role in the life history of
Russula even in mature forests ( Redecker et al. ,
2001) .
Though still preliminary, the data in Table 3 also
suggest that phylogenetic background of species may
play a role in genet size . Specifically, if the phyloge-
netic background were important, we should expect to
see similar genet sizes among closely related species .
Based on the limited data available, species in certain
genera, e.g . Suillus, all seemed to have relatively
largegenets . However, thismay not hold inother taxo-
nomic groups . For example, distinct genet sizes and
propagation strategies areoften found among closely re-
lated taxa in other genera, e.g . between two sister
species of Rhizopogon ( Kretzer et al. , 2003; Table
3) . Further analyses of more taxa in representative en-
vironmental conditions are needed to critically test this
hypothesis .
4 Ecological importance of EcM fungi di-
versity
Aside fromgenet size differences, EcM fungi can
vary significantly in other aspects, including the sub-
strates they inhabit, the ability to uptakevarious nutri-
ents, and the tolerance to water stress and temperature
extremes . Because different groups of EcM fungi have
different adaptive features, we expect that plants asso-
ciated with a greater diversity of EcM fungi in their
roots to be fitter and more capable of surviving and re-
producing . In contrast, plants associated with a single
typeof EcM fungi may be vulnerable to environmental
stresses . However this hypothesis has not been tested
either in laboratory or field conditions, therefore whe-
ther highdiversityof colonizing fungi could enhancethe
survival of plants remains unclear .
Previous researches on EcM fungi community
structure have demonstrated that the abundance and
rarity in EcM community seemed to show reverse rela-
tionships ( Horton and Bruns, 2001 ) . This pattern
means that the EcM samples that werandomly select in
nature are likely highly biased . The high number of
rare EcM species could mask the actual abundance of
common EcM species in a selected EcM community .
The bigger thesamplesize, themore species andgeno-
types will likely be identified . Whether a sample size
wouldbe representativeof theEcM community structure
can be determined using saturation analysis . In this
analysis, when the number of samples goes up, the
number of species should alsogo up, to apoint when it
becomes stable ( McCune and Mefford, 1999 ) . The
number of samples before the curve reaches plateau is
the suggested one that can robustly represent the EcM
fungal diversity . Jackknife estimate can also be used to
predict the number of taxa needed to resolve the com-
munity structure ( Mueller-Dombois and Ellenberg,
1974; Palmer et al. , 1991 ) .
Based on studieson EcM community conducted so
far, the most dominant EcM fungi fruiting bodies are
found belonging to Russulaceae, Thelephoraceae, and
non-thelephoroid resupinates ( Gardes and Bruns,
1996a; Jonsson et al. , 1999c; Luoma et al. , 1997 ) .
At present, the reasons for their numerical and poten-
tially functional dominance are unknown . However, it
has been suggested that for most EcM fungi , the ones
that fruit themost abundant arenot necessarily theones
most abundant as mycorrhizae ( Gardes and Bruns,
1996a; Jonsson et al. , 1999 a, b; Mehmann, 1995) .
Onepossibility is that species that are abundant as my-
corrhizaeon roots but fruits infrequently likely require
more energy for sexual reproduction than for vegetative
growth . Direct analysesof metabolisms and energy con-
sumptions areneeded to test this hypothesis .
A better understanding the community structure
and the relationship between hosts and symbiotic EcM
fungi will allow us better able to develop sound strate-
gies to conserve the diversity of EcM fungi . This in
turn will help construct a suitable forest management
practices to protect forest ecosystems . In thesedevelop-
ments, two specific issues warrant consideration: ( 1)
different EcM fungi play different roles in plant growth;
and (2) different EcM communities are associatedwith
602 云 南 植 物 研 究 31 卷
different aged forests (Molina and Trappe, 1982; Syl-
via et al. , 1997 ) . Often, diverse EcM fungi are found
associated with mature trees, and this association can
benefit nearby seedlings (Sylvia et al. , 1997) . It has
been demonstrated that a greater EcM diversity exists
for seedlings located closer to mature trees than those
closer to immature or newly-planted tress (Sylvia et
al. , 1997) . This result suggests that mature trees can
be used as a reservoir to help maintain diverse EcM
fungi for the broader forest community . Furthermore,
Taylor et al . ( 1989 ) identified a high diversity of
EcM fungi in the treeswhose nearby trees arefromother
species . Therefore, planting a mixture of tree species
can also be a great way to improve EcM diversity . In
thesemixed forest ecosystems, species with broad host
ranges will likely be more competitive than those with
narrow host ranges . Therefore, during or after timber
harvesting, representative host plants should be re-
tained and?or replanted .
5 Concluding remarks and future direc-
tions
In this review, we outlined a set of methods for
identifying EcM fungi and illustrated their utilities in
EcM research on genet size, modes of reproduction,
and ecology . A combination of morphological and
DNA-based molecular techniques is often required to
identify largenumbers of unknown EcM fungi innatural
ecosystems . Samples of both sporocarps and mycorrhi-
zae are needed to help obtain a clear picture of the
richness and abundance of EcM fungal species in a
particular study site . At present, thefull extentof EcM
fungal diversity remains largely unknown . Future ex-
periments should directly examine the relationships be-
tween EcM fungal diversity and function in representa-
tive ecosystems . In addition, the patterns and mecha-
nisms for the specificity and inter-connectedness among
different host plants, different mycorrhizal fungi , and
between mycorrhizal fungi and plants will be of great
interest for both basic scientific investigations and ap-
plied research . The molecular ecological analyses of
more EcM fungi should help us understand the broad
patterns of genet sizes and their modes of reproduction .
Such understandings will help develop suitable forest
management strategies .
References:
Amar ?anthus MP , 1998 . The importance and conservation of ectomycorrhi-
zal fungal diversity in forest ecosystems lessons from Europe and the
Pacific Northwest [ J ] . Pacific Northwest Research Station General
Technical Report, PNW-QTR-431
Baar ?J , Ozinga WA, Kuyper TW et al. , 1994 . Spatial distribution of
Laccaria bicolor genets reflected by sporocarps after removal of litter
and humus layers in a Pinus sylvestris forest [ J ] . Mycological Re-
search, 98 : 726—728
Berg ?emann S, Miller S, 2002 . Size, distribution, and persistenceof gen-
ets in local populations of the late-stage ectomycorrhizal basidiomyce-
te, Russula brevipes [ J ] . New Phytologist, 156: 313—320
Bled ?soe CS, Tennyson K , Lopushinsky W et al. , 1982 . Survival and
growth of mycorrhizal Douglas-fir seedlings outplanted on burned-over
sites in eastern WA [ J ] . Canadian Journal of Forest Research, 12 :
720—723
Bido ?chka MJ , Small CLN , Spironello M et al. , 2005 . Recombination
within sympatric cryptic species of the insect pathogenic fungus
Metarhizium anisopliae [ J ] . Environmental Microbiology, 7 ( 9 ) :
1361—1389
Bone ?llo P, Bruns TD, Gardes M et al. , 1998 . Genetic structureof anat-
ural population of the ectomycorrhizal fungus Suillus pungens [ J ] .
New Phytologist, 38 : 533—542
Brun ?s TD, GardesM, 1993 .Molecular tools for the identification of ecto-
mycorrhizal fungi : taxon-specific oligonucleotide probes for suilloid
fungi [ J ] . Molecular Ecology, 2 : 233—242
Brun ?s TD, Szaro TM, Gardes M et al. , 1998 . A sequence database for
the identification of ectomycorrhizal basidiomycetes by phylogenetic
analysis [ J ] . Molecular Ecology, 7 : 257—272
Ciar ?do DE, Sch?r G, B?ttger EC et al. , 2006 . Internal transcribed spac-
er sequencing versus biochemical profiling for identification of medi-
cally important yeasts [ J ] . J ournal of Clinical Microbiology, 44 (1) :
77—84
Dahl ?bergA , 1997 . Population ecology of Suillusvariegates in oldSwedish
Scots pine forests [ J ] . Mycological Research, 101 : 47—54
Dahl ?bergA , Strenlid J , 1990 . Population structure and dynamics in Suil-
lus bovinus as indicated by spatial disturbance of fungal clones [ J ] .
New Phytologist, 115: 487—493
Dahl ?berg A , Strenlid J , 1994 . Size, distribution and biomassof genets in
population of Suillus bovinus (L . Fr .) Roussel revealed by somatic
incompatibility [ J ] . New Phytologist, 12 : 225—234
Dett ?man J R, Jacobson DJ , Taylor JW et al. , 2003 . A multilocus genea-
logical approach to phylogenetic species recognition in the model eu-
karyote Neurospora [ J ] . Evolution, 57 : 2703—2720
Doyl ?e JJ , 1992 . Gene trees and species trees: molecular systematics as
one-character taxonomy [ J ] . Systematic Botany, 17 : 144—163
Dunh ?amSM , Kretzer A , Pfrender ME et al. , 2003 . Characterization of
Pacific golden chanterelle ( Cantharellus formosus) genet size using
7023 期 LI and XU: Molecular Ecology of Ectomycorrhizal Fungi : Molecular Markers, Genets and Ecological . . .
co-dominant microsatellite markers [ J ] . Molecular Ecology, 12 :
1607—1618
Egge #r KN , 1995 . Molecular analysis of ectomycorrhizal fungal communi-
ties [ J ] . Canadian J ournal of Botany, 73 : S1415—S1422
Fels ?enstein J , 1978 . Cases inwhich parsimony and compatibility methods
will be positively misleading [ J ] . Systematic Zoology, 27 : 401—410
Fish ?er J , 1978 . R . A . Fisher: The Life of a Scientist [ M] . New York:
Wiley, ISBN 0-471-09300-9
Gard #es M, WhiteTJ , Fortin JA et al. , 1991 . Identification of indigenous
and introduced symbiotic in ectomycorrhizae by amplification of the
nuclear and mitochondrial ribosomal DNA [ J ] . Canadian J ournal of
Botany, 69 : 180—190
Gard #es M, Bruns TD, 1993 . ITS primers with enhanced specificity for
basidiomycetes—application to the identification of mycorrhizae and
rusts [ J ] . Molecular Ecology, 2 : 113—118
Gard #es M, Bruns TD, 1996a . Community structure of ectomycorrhizal
fungi in a Pinus muricata forest: above- and below-ground views
[ J ] . Canadian J ournal of Botany, 74 : 1572—1583
Gard #es M , Bruns TD, 1996b . ITS-RFLP matching for identification of
fungi [ J ] . Methods in Molecular Biology, 50 : 177—186
Geml BJ , Laursen GA , O′neill K et al. , 2006 . Beringian origins and
cryptic speciation events in the fly agaric ( Amanita muscaria) [ J ] .
Molecular Ecology, 15 : 225—239
Guid #ot A , Debaud JC , Marmeisse R et al. , 2001 . Correspondence be-
tween genet diversity and spatial distribution of above- and below-
ground populations of the ectomycorrhizal fungus Hebeloma cylindros-
porum: mycelial persistence and inbreedin [ J ] . Heredity, 84 :
294—302
Gryt ?a H , Debaud JC , Effosse A et al., 1997 . Fine-scale structure of
populations of the ectomycorrhizal fungus Hebeloma cylindrosporum in
coastal sand dune forest ecosystems [ J ] . Molecular Ecology, 6 :
353—364
Gryt ?a H , Debaud JC , Marmeisse R et al., 2000 . Population dynamics of
the symbiotic mushroom Hebeloma cylindrosporum: mycelial persis-
tence and inbreeding [ J ] . Heredity, 84 : 294—302
Hiro ?se D, Kikuchi J , Kanzaki N et al. , 2004 . Genet distribution of spo-
rocarps and ectomycorrhizas of Suillus pictus in a Japanesewhite pine
plantation [ J ] . New Phytologist, 164 (3 ) : 527—541
Hort ?on TR , Bruns TD, 2001 . The molecular revolution in ectomycorrhizal
ecology: peeking into the black-box [ J ] . Molecular Ecology, 10
( 8) : 1855—1871
J eff ?ries P, Gianinazzi S, Perotto S et al. , 2003 . The contribution of ar-
buscular mycorrhizal fungi in sustainable maintenance of plant health
and soil fertility [ J ] . Biology and Fertility of Soils, 37 : 1—16
J ons ?son L , Dahlberg A , Nilsson MC et al. , 1999a . Continuity of ectomy-
corrhizal fungi in self-regenerating boreal Pinus sylvestris forests stud-
ied by comparing mycobiont diversity on seedlings and mature trees
[ J ] . New Phytologist, 142: 151—162
J ons ?son L , DahlbergA , Nilsson MC et al. , 1999b . Ectomycorrhizal fun-
gal communities in late-successional Swedish boreal forests, and their
composition followingwildfire [ J ] . Molecular Ecology, 8 : 205—215
J ons ?son T , Kokalj S, Finlay R et al. , 1999c . Ectomycorrhizal community
structure in a limed spruce forest [ J ] . Mycological Research, 103:
501—508
Kaus ?erud H , Stensrud?, DeCock C et al. , 2006 . Multiplegenegeneal-
ogies and AFLPs suggest cryptic speciation and long-distancedispers-
al in the basidiomycete Serpula himantioides (Boletales) [ J ] . Mo-
lecular Ecology, 15 : 421—431
Kauserud H , Svegarden IB , Decock C et al. , 2007 . Hybridization among
cryptic species of the cellar fungus Coniophora puteana ( Basidiomy-
cota) [ J ] . Molecular Ecology, 16 : 389—399
K?ré ?n O, Hogberg N , Dahlberg A et al. , 1997 . Inter- and intraspecific
variation in the ITS region of rDNA of ectomycorrhizal fungi in Fen-
noscandia as detected by endonuclease analysis [ J ] . New Phytolo-
gist, 136: 313—325
Kret ?zer AM , DunhamS, Molina R et al. , 2003 . Microsatallite markers
reveal the below ground distribution of genets in two speices of Rhizo-
pogon forming tuberculate ectomycorrhizas on Douglas fir [ J ] . New
Phytologist, 161: 313—320
Lan ?L , Xu J , 2006 . Multiple gene genealogical analyses suggest diver-
gence and recent clonal dispersal in theopportunistic human pathogen
Candida guilliermondii [ J ] . Microbiology, 152: 1539—1549
Lepa ?geOM, 1997 . He′morragie dans les poches gutturales ( 1 ) . Anato-
mie, diagnostic et e′tiologie [ J ] . Pratique Vete′rinaire Equine, 26 :
225—261
Lian ?CL , Narimatsu M, NaraK et al. , 2006 . Tricholoma matsutake in a
natural Pinus densiflora forest: correspondence between above- and
below-ground genets, association with multiple host trees and alter-
ation of existing ectomycorrhizal communities [ J ] . New Phytologist,
171 : 825—836
LiangY , Guo LD, Ma KP et al. , 2004a . Spatial pattern of the most
common late-stage ectomycorrhizal fungi in a subtropical forest inDu-
jiangyan , southwest of China [ J ] . Acta Botanica Sinica, 46 : 29—34
Lian ?g Y , Guo LD, MaKP et al. , 2004b . Genetic structureof a popula-
tion of the ectomycorrhizal fungus Russula vinosa in subtropical wood-
lands in southwest China [ J ] . Mycorrhiza, 14 : 235—240
Luom ?aDL , Eberhart JL , Amaranthus MP et al. , 1997 . Biodiversity of
ectomycorrhizal types from southwest Oregon [ A ] . In: Kaye TN ,
Liston A , Love RM et al. ed, Conservation and Management of Na-
tive Plants andFungi [M ] . Oregon: Native Plant Society of Oregon,
249—253
Madd ?ison WP, 1997 . Gene trees in species trees [ J ] . Systematic Biolo-
gy, 46 : 523—536
McCune B, Mefford MJ , 1999 . PC-ORD: multivariate analysisof ecolog-
ical data . Version 4 . User′s guide [ A ] . In: MjM Software Design
[ M] . Oregon: Gleneden Beach
Molina R , Trappe JM, 1982 . Patterns of ectomycorrhizal host specificity
and potential among pacific northwest conifers and fungi [ J ] . Forest
Science, 28 : 423—458
Muel ?ler-Dombois D, EllenbergH , 1974 . Aims and Methods of Vegetation
Ecology [M ] . New York: John Wiley and Sons
Mull ?is KB , Faloona FA et al. , 1987 . Specific synthesis of DNA in vitro
via apolymerase-catalyzed chain reaction [ J ] . Methods in Enzymlogy,
155: 335—350
802 云 南 植 物 研 究 31 卷
Palm &er MW, 1991 . Estimating species richness: the second-order jack-
knife reconsidered [ J ] . Ecology, 72 : 1512—1513
Prit ?sch K , BoyleH , Munch JC et al. , 1997 . Characterization and identi-
fication of black alder ectomycorrhizas by PCR-RFLP analyses of the
rDNA internal transcribed spacer ( ITS) [ J ] . New Phytologist, 137:
357—369
Prit ?sch K , Munch JC, Buscot F et al. , 2000 . Identification and differen-
tiation of mycorrhizal isolates of black alder by sequence analysis of the
ITS region [ J ] . Mycorrhiza, 10 : 87—93
Read ?DJ , 1991 . Mycorrhizas in ecosystems [ J ] . Experentia, 47 : 376—
391
Rede +cker D, Szaro TM, Bowman RJ et al. , 2001 . Small genets of Lac-
tarius xanthogalactus, Russula cremoricolor and Amanita francheti in
late-stage ectomycorrhizal succession [ J ] . Molecular Ecology, 10 :
1025—1034
Remy LW, Taylor TN , Hass H et al., 1994 . Four hundred-million-year-
old vesicular arbuscular mycorrhizae [ J ] . Proceedings of the Na-
tional Academy of Sciences ( USA) , 91 : 11841—11843
Rich ?ard F , Millot S, Gardes M et al. , 2005 . Diversity and specificity of
ectomycorrhizal fungi retrieved froman old-growth Mediterranean forest
dominated by Quercus ilex [ J ] . New Phytologist, 166: 1011—1023
Robi ?nson CH , Szaro TM, Izzo AD et al. , 2009 . Spatial distribution of
fungal communities in a coastal grassland soil [ J ] . Soil Biology and
Biochemistry, 41 : 414—416
Sait ?ou N , Nei M, 1987 . The neighbor- joiningmethod: anew method for
reconstructing phylogenetic trees [ J ] . Molecular Biology and Evolu-
tion, 4 : 406—425
Sawy er NA , Chambers SM , Cairney JWG et al. , 1999 . Molecular inves-
tigation of genet distribution and genetic variation of Cortinarius ro-
tundisporus in eastern Australian sclerophyll forests [ J ] . New Phy-
tologist, 142: 561—568
Selo ?sse MA , MartinF , BouchardD et al. , 1999 . Structure and dynamics
of experimentally introduced and naturallyoccurring Laccaria sp . Dis-
crete genotypes in aDouglas fir plantation [ J ] . Applied and Environ-
mental Microbiology, 65 : 2006—2014
Selo ?sse MA , Richard F , He X et al. , 2006 . Mycorrhizal networks: des
liaisons dangereuses [ J ] ? Trends in Ecology and Evolution, 21 :
621—628
Sha ?T, Xu J , Palanichamy M et al. , 2008 . Genetic diversity of the en-
demic gourmet mushroom Thelephora ganbajun fromsouthwestern Chi-
na [ J ] . Microbiology, 154: 3460—3468
Sima ?rd SW, Perry DA , Jones MD et al. , 1997 . Net transfer of carbon
between ectomycorrhizal tree species in the field [ J ] . Nature, 388:
579—582
Sima ?rdSW, Jones MD, Durall DM et al. , 2002 . Carbon and nutrient
fluxes within and between mycorrhizal plants [ J ] . Ecological Stud-
ies, 157: 33—61
Smit ?h SE, Read DJ , 1997 . Mycorrhizal Symbiosis ( 2nd ed) [M ] . Lon-
don: ISBN 0-12-652840-3
Sylvia DM , Jasrstfer AG, 1997 . Distribution of mycorrhizaeon competing
pines and weeds in a southern pine plantation [ J ] . Soil ScienceSoci-
ety of America J ournal , 61 : 139—144
Tayl ?or AFS, Alexander IJ , 1989 . Demography and population dynamics
of ectomycorrhizaof Sitkasprue fertilizedwithN [ J ] . Agriculture Ec-
osystems and environment, 28 : 493—496
Tayl ?or DL , Bruns TB, 1999 .Community structureof ectomycorrhizal fun-
gi in a pinus muricata forest: minimal overlap between themature for-
est and resistant propagule communities N [ J ] . Molecular Ecology,
8 : 1837—1850
Tayl ?or JW, Jacobson DJ , Kroken S et al. , 2000 . Phylogenetic species
recognition and species concepts in fungi [ J ] . Fungal Genetics and
Biology, 31 : 21—32
Trap ?pe, JM, 1987 . Phylogenetic and ecologic aspects of mycotrophy in
the angiosperms from an evolutionary standpoint [ A ] . In: Safir GR
ed, Ecophysiology of VA Mycorrhizal Plants [ M ] . Florida: CRC
Press
Whit ?e TJ , Bruns TD, Lee SB et al. , 1990 . Amplicaion and direct se-
quencin of fungal ribosomalRNA genes for phylogenetics [ A ] . In:
Innis MA , Gelfand DH , Sninsky JJ et al. eds, In PCR protocols: a
Guide to Methods and Applications [ M] . London: Academic Press,
315—322
Xu J , 2005 . Evolutionary Genetics of Fungi [ M ] . U .K .: Horizon Bio-
science
Xu J ?, 2006 . Microbial ecology in the ageof genomicsand metagenomics:
concepts, tools, and recent advances [ J ] . Molecular Ecology, 15 :
1713—1731
Zhou ?Z, Miwa M , Hogetsu T et al. , 1999 . Analysis of genetic structure
of a Suillus grevillei population in a Larix kaempferi stand by poly-
morphism of inter-simple sequence repeat ( ISSR) [ J ] . New Phytol-
ogist, 144: 55—63
9023 期 LI and XU: Molecular Ecology of Ectomycorrhizal Fungi : Molecular Markers, Genets and Ecological . . .