Untitled

Molecular Ecology (2008) 17, 1238–1247
Recombination and genetic differentiation among natural
populations of the ectomycorrhizal mushroom Tricholoma
matsutake from southwestern China
J I A N P I N G X U ,*†‡ TA O S H A ,† YA N - C H U N L I ,* Z H I - W E I Z H A O † and Z H U L . YA N G **Key Laboratory of Biodiversity and Biogeography, Kunming Institute of Botany, Chinese Academy of Sciences, Kunming, Yunnan 650204, China, †Laboratory for the Conservation and Utilization of Biological Resources, Yunnan University, Kunming, Yunnan 650091, China, ‡Center for Environmental Genomics, Department of Biology, McMaster University, Hamilton, Ontario, Canada L8S 4K1 Abstract
Effective conservation and utilization strategies for natural biological resources require a
clear understanding of the natural populations of the target organisms. Tricholoma matsutake
is an ectomycorrhizal mushroom that forms symbiotic associations with plants and plays
an important ecological role in natural forest ecosystems in many parts of the world. It is
also an economically very important gourmet mushroom. Because no artificial cultivation
is available, natural populations of this species are under increasing threats, primarily from
habitat disturbance and destruction. Despite its economical and ecological importance, little
is known about its genetics and population biology. Here, using 14 polymerase chain
reaction–restriction fragment length polymorphism markers, we analysed 154 strains from
17 geographical locations in southwestern China, a region where over 25% of the global
T. matsutake harvest comes from. Our results revealed abundant genetic variation within
individual populations. The analyses of gene and genotype frequencies within populations
indicated that most loci did not deviate from Hardy–Weinberg equilibrium in most
populations and that alleles among loci were in linkage equilibrium in the majority of the
local populations. These results are consistent with the hypothesis that sexual reproduction
and recombination play an important role in natural populations of this species. Our
analyses indicated low but significant genetic differentiation among the geographical
populations, with a significant positive correlation between genetic distance and geographical
distance. We discuss the implications of our results to the ecology and resource management
of this species.
Keywords: biodiversity, conservation, fungi, matsutake mushroom, PCR-RFLP, SNP Received 2 September 2007; revision received 3 November 2007; accepted 2 December 2007 around the roots help plants obtaining essential phosphate Introduction
and minerals from the soil. The mycorrhizae not only help Fungi play pivotal roles in human and animal health, plants grow but also contribute to plants’ disease resistance agriculture, biotechnology, and forestry. They are essential and drought tolerance (Brundrett 2004). Based on their components of all natural terrestrial ecosystems. In forest structural associations, mycorrhizae are divided into two ecosystems, fungi are commonly found as decomposers on groups, endomycorrhizae and ectomycorrhizae. Almost forest floors, as commensal endophytes or pathogens on all tree species form ectomycorrhizal associations with plant stems, roots, leaves or fruits, and in or on plant roots fungi that belong to divisions Zygomycota, Ascomycota, as symbionts. The symbiotic association between fungi and or Basidiomycota (Brundrett 2004). Among species in plant roots is called mycorrhizae. The fungal mycelia Basidiomycota that form ectomycorrhizal associationswith plant roots, many produce mushrooms that are collected Correspondence: Jianping Xu, Fax: 1-905-522-6066; as sources of exotic and highly prized food for humans Journal compilation 2008 Blackwell Publishing Ltd P O P U L AT I O N S T R U C T U R E O F T. M AT S U TA K E 1239
Tricholoma matsutake is an ectomycorrhizal basidiomycete differentiation among geographical populations, individual that produces economically important edible mushrooms populations may need to be managed and conserved commonly known as ‘matsutake’. It is predominantly separately. At present, evidence for sexual reproduction in associated with pine forests in the Northern Hemisphere a local population of T. matsutake has been reported in Japan but has been found to be associated with oaks and other tree (Lian et al. 2006). However, the prevalence of sexual species in southwestern China (Nakayama & Nakanishi reproduction in other populations of T. matsutake is unknown.
2004). Though morphologically indistinguishable, the Furthermore, very little is known about the relationships difference in their associated dominant host tree species between geographical populations, either in Japan or has been used as an ecological parameter to separate the elsewhere. Only one study has examined the relationship matsutake mushrooms from southwestern China into between genetic distance and geographical distances in two closely related species, T. matsutake and Tricholoma T. matsutake (Chapela & Garbelotto 2004). In this study, seven quercicola (Zang 1990) (synonym T. zangii Z.M. Cao et al.
strains of T. matsutake, three from China and one each (2003), corresponding to forest habitats dominated by from Korea, Japan, France, and Morocco, were examined.
pines and alpine oaks, respectively. Closely related species A significant positive correlation was found between their that produce fruiting bodies similar to T. matsutake exist in genetic dissimilarity based on genotypes identified using many parts of North Africa, North America, Central America, amplified fragment length polymorphisms (AFLP) and their South America, as well as Europe and Asia (Redhead 1997).
geographical distances. A similar correlation was found in These are all collectively traded under the name matsutake.
two species closely related to T. matsutake, Tricholoma caligatum Among the global markets, Japan is by far the most and Tricholoma magnivelare, using five and six strains from important in terms of consumption. However, in the past diverse geographical locations, respectively (Chapela & century, because of deforestation and infestation by the pinewood nematode (Bursaphelenchus xylophilus), the host Since the 1980s, molecular markers have become plant populations of T. matsutake, Pinus densiflora, declined increasingly important tools for studying a variety of rapidly in Japan (Gill et al. 2000; Wang & Hall 2004). As a biological properties and processes such as recombination result, the annual harvest of T. matsutake in Japan has been and population structuring (Xu 2006a). Among the many much lower than it used to be in the early 20th century. In types of markers that have been developed, single nucleotide Europe, significant declines in ectomycorrhizal mushrooms, polymorphisms (SNPs) are among the fastest developing including species in the genus Tricholoma such as T. matsutake, category in biomedical and biological research. Though have also been observed in many natural forest ecosystems with its own caveats such as ascertainment bias (Morin et al.
(e.g. Arnolds 1991). Because artificial cultivation has not been 2004), SNPs are the most frequently observed differences developed for any of the matsutake mushrooms, including between DNA sequences obtained from different individuals T. matsutake, to satisfy its domestic demand, Japan imports or between alleles from within the same individual in diploid about 3000 tons of matsutake annually, mostly from Pacific or higher ploidy organisms. In addition, SNPs have several North America, Korea and China. The commercial demands properties such as a relatively low mutation rate and the from Japan have subsequently placed serious pressure on ease of scoring and sharing data that make them highly natural populations of this species elsewhere, especially in desirable for a variety of biological analyses (Brumfield et al.
southwestern China where up to a third of the Japanese 2003; Xu 2006a). Recently, we reported the identification of imports come from. Indeed, the regions around Kunming, 178 SNPs for T. matsutake (Xu et al. 2007). Some of these the provincial capital of Yunnan Province in southwestern SNPs were further distinguished using simple polymerase China, once produced significant amounts of the matsutake chain reaction (PCR) followed by digestions using specific mushroom over two decades ago but are now producing restriction enzymes to generate restriction fragment length very little or none. As a result, understanding and conserving polymorphisms (RFLP). Here, using 14 PCR–RFLP markers the natural populations of this mushroom have attracted developed previously based on the analysis of two strains, significant attention in recent years from both government we genotyped 154 strains from 17 geographical regions agencies and nongovernment conservation organizations.
from southwestern China. Our results indicate that sexual Effective conservation and management strategies for any reproduction and recombination are prevalent in these natural biological resources require a clear understanding populations and that there is limited but significant genetic of how the organisms reproduce in nature and how popu- differentiation among these geographical populations.
lations from different regions are related. For example, if However, we identified little genetic difference between sexual reproduction and recombination are common and samples obtained from the two different forest ecosystems, important in a natural population, maintaining a set of one dominated by pines and the other by alpine oaks and mature mushrooms and their associated sexual spores shrubs. The implications of these results in the taxonomy, will be important for its subsequent reproduction and conservation and sustainable utilization of T. matsutake survival in nature. In addition, if there is significant genetic 2008 The AuthorsJournal compilation 2008 Blackwell Publishing Ltd 1240 J . - P. X U E T A L .
Table 1 Populations of Tricholoma matsutake
China and their physical geographicalinformation While the detailed spatial maps of the fruiting bodies at Materials and methods
the sampled geographical areas were not recorded, we havenoted that none of the mushrooms we collected were clustered within 2 m of each other and that within each The isolates used in this study were collected in Yunnan and sampled population, the mushrooms were typically dis- Sichuan Provinces in southwestern China. A total of 154 tributed from about 15 m of each other to as far as about mushrooms were collected from 17 geographical locations.
2 km from each other. The sampling distance was not One location was in Baiyu County in western Sichuan subjectively imposed but was rather due to the rarity of Province that borders Tibet. The remaining 16 populations fruiting of this species at any given time and the difficulty were from various regions in Yunnan Province, from Luliang in finding the mushrooms in their native habitats. Indeed, in the east to Deqin in the northwest. The sample size and within each of our sampled areas, all matsutake mushrooms geographical coordinates for each population are presented we identified during our foray were collected. Because of in Table 1. The altitudes for all 17 populations are also given their habit of producing fruiting bodies buried in the soil or in Table 1. Our sampled regions spanned an area about the litter layer, matsutake mushrooms are difficult to find.
950 km from south to north and 650 km from east to west As a result, it took us 6 years (2000–2006) of persistent effort with an altitude span of almost 2000 m, from 1540 m above to assemble this collection of 154 mushrooms from the 17 sea level in Longling to 3400 m in Deqin. All population geographical regions. In several regions (Longling, Luliang, samples were collected from mixed forests consisting of Luquan, and Yongping), aside from some initial successes, pines, alpine oak, and a variety of shrubs. The sampled we failed to obtain any additional mushrooms despite forests in northwestern Yunnan (Deqin, Weixi, Shangri-La, repeated field trips. As reported by the local residents, these and Lijiang) and Baiyu (Sichuan) were dominated by alpine areas now produce very little or no matsutake mushrooms.
oaks and shrubs. In contrast, those in other sampled regions While only two and three isolates were obtained from in Yunnan (Lanping, Longling, Yongping, Jianchua, Lincang, Longling and Luliang, they were the only ones we were Nanhua, Lufeng, Chuxiong, Luquan, Yimen, Luliang, and able to obtain and were thus included here for analyses Ailaoshan) were dominated by pine trees. The sampled area contributes over 25% of the Tricholoma matsutake mushroomsharvested globally each year. Isolates of T. matsutake were identified based on their macro- and micromorphologicalcharacteristics and confirmed based on their sequences at For each of the 154 strains, about 0.01 g tissues from the cap the internal transcribed spacer (ITS) region of the ribosomal of the dried mushroom were ground into a fine powder using a blue tip in a 1.5-mL microcentrifuge tube. The remaining Journal compilation 2008 Blackwell Publishing Ltd P O P U L AT I O N S T R U C T U R E O F T. M AT S U TA K E 1241
steps essentially followed those for extracting DNA from at 37 °C or according to manufacturer’s recommendations.
live yeasts or from dried mushroom specimens (Xu et al.
PCR products and restriction digests were electrophoresed 1994, 2000), with slight modifications. The modifications in 1.2% agarose in 1× TAE, stained with ethidium bromide, were in the last steps. Briefly, after the DNA was washed viewed by ultraviolet transillumination, and scored as with 70% ethanol and air-dried, they were re-suspended in codominant markers (Xu et al. 1999).
100 μL of Tris-EDTA (TE) buffer. The DNA was then cleanedusing the commercial GeneClean III kit BIO101 (Qbiogene) following the supplier’s instructions. The final DNA wassuspended in 50 μL TE buffer and stored at –20 °C. By Our genetic analyses were performed for both within this method, 0.01 g of dried mushroom tissues yielded individual populations as well as among populations. In addition, the genetic distances among populations werecompared to physical distances among geographical regionsas well as their altitudinal differences to examine the potential relationships between genetic distances and geographical Before the genotyping experiment using the restriction parameters. The specific analyses conducted here are site polymorphic markers based on single nucleotide polymorphisms identified earlier in Xu et al. (2007), wesequenced a random set of strains at their nuclear ITS of the Analysis of genetic variation within populations. For the analysis ribosomal RNA gene cluster using conserved primers ITS1 of the within-population genetic variation, we calculated and ITS4 following a protocol described previously (Wang the genotype diversity and analysed the associations of et al. 2007). The ITS sequencing was carried out both to alleles within the same locus as well as between loci for each confirm their species identification and to reveal potential of the 17 populations. Here, genotype diversity refers to the sequence variation among isolates at this genomic region, probability that any two individuals drawn randomly from especially strains from different forest types and geographical the population will have a different multilocus genotype regions. In total, 34 isolates were analysed, with two isolates from each of the 17 geographical populations.
To examine whether there is evidence for recombination The genotypes of the isolates were obtained based on in individual geographical populations, we used two protocols described in Xu et al. (1999, 2007). A total of 14 population genetic measures, Hardy–Weinberg equilibrium PCR–RFLP markers located on seven random DNA (HWE) test and the index of association. These two measures fragments were screened for the entire 154 isolates. The 14 assessed the association of alleles within or between loci.
PCR–RFLP markers used here were three more than the The rationale for the inferences of recombination in these original 11 presented in the Xu et al. (2007). The three tests is that in a population with significant sexual repro- additional markers were PCR fragment TmRC14 digested duction and recombination, we should observe random by the restriction enzyme HindIII, and PCR fragment associations between alleles at the same locus or between TmRC18 digested by restriction enzymes HinfI and BanII, loci. This is because genes from different individuals are respectively. Information about obtaining both fragments mixed every time sexual reproduction occurs, generating was described in Xu et al. (2007). These PCR–RFLP markers random associations between alleles at the same locus and were recently identified based on SNPs discovered through between different loci (Xu 2006b). In diploid organisms with the analyses of a shotgun genomic library and the sequence codominant genotype data, the HWE test has been used to comparisons among four alleles at 20 different loci in two examine the associations of alleles within each locus.
mushrooms. For the genotyping, the DNA fragments that Specifically, loci with genotype frequencies not significantly contained SNPs detectable by restriction enzyme digests different from those expected under the assumption of were first amplified by PCR. Each PCR contained about random mating are determined to be in HWE. The pre- 10 ng of DNA, 0.5 U Taq DNA polymerase, 1 μm each dominance of loci in HWE indicates that recombination primer, and 200 μm of each of the four deoxyribonucleotide plays a significant role in the assayed population (Xu 2006b).
triphosphates in a total volume of 15 μL. The following PCR The test for HWE was performed using the computer conditions were used for all amplifications: 4 min at 95 °C, program genalex version 6 (Peakall & Smouse 2006) for followed by 35 cycles of 30 s at 95 °C, 30 s at 50 °C, 45 s at each of the 14 loci in 15 of the 17 populations where sample 72 °C, and finally 7 min at 72 °C. After confirmation of the sizes are sufficiently large. The exceptions were Luliang and PCR products by agarose gel electrophoresis, the fragments were digested using the specific restriction enzyme. Typical In the second test, we calculated the standard, most restriction enzyme reactions consisted of 7-μL PCR product, commonly used multilocus linkage disequilibrium called 1 U restriction enzyme, 1.5 μL 10× reaction buffer, and H O the index of association I using the computer program to a total volume of 15 μL. Reactions were incubated 2–3 h multilocus (Agapow & Burt 2001). In this test, the observed 2008 The AuthorsJournal compilation 2008 Blackwell Publishing Ltd 1242 J . - P. X U E T A L .
Table 2 Patterns of genetic variation within geographical populations of Tricholoma matsutake
*Statistically different from the null hypothesis of random association at P < 0.05; † not analysed because of small sample size. data were compared against the null hypothesis that alleles Tests for genetic isolation by geographical distance or altitudinal (or genotypes) from different loci were randomly associating differences. The Mantel test was used to examine whether with each other. If no or little sexual reproduction occurred, there was evidence for genetic isolation by geographical there would be significant association between alleles at distance or by altitudes among our populations of T.
different loci due to clonal reproduction. In contrast, random matsutake. Because detailed sporocarp–sporocarp distances associations between alleles at different loci suggest recom- were not collected for most of the isolates within each bination (Xu 2006b). The formulae and inferences of statistical geographical area, the potential genetic isolation by geo- significance for this test can be found on the multilocus graphical distance within individual populations could program homepage (Agapow & Burt 2001).
not be assessed. Instead, our analyses focused on therelationships between populations. To this end, we conducted Analyses of genetic variation between populations. The genetic two separate Mantel tests. In the first Mantel test, the differences between samples were quantified using the pairwise Nei’s population genetic distances were calculated phi-statistic through the analysis of molecular variance based on gene frequency differences between populations (amova). Phi-statistics is a modified version of Wright’s F and were then compared to geographical distances between that refers to the relative contributions of between-population sampled populations. Our second Mantel test compared separation to the overall genetic variation in the whole Nei’s genetic distances with altitudinal differences between sample. The greater the F values are, the greater the populations. Both Mantel tests were conducted using the differences between populations. The relative contributions program genalex (Peakall & Smouse 2006).
of within-population genetic variation phiPT, between localpopulations within regions phiPR, and between regions phiRT were calculated using the computer program genalex(Peakall & Smouse 2006). Here, the dominant host tree Our ITS sequence results identified that all 34 sequenced species were used to define two regional-level populations, isolates had ITS sequences identical to each other and to the one dominated by alpine oaks and shrubs and the other by GenBank ITS sequence of the typical Tricholoma matsutake pines. Within the regional population dominated by oaks isolates from Japan, Korea, and China. In contrast, their ITS and shrubs are five populations from Baiyu, Deqin, Weixi, sequences were distinctly different from those in other Shangri-La, and Lijiang. The other regional group includes matsutake mushrooms such as Tricholoma bakamatsutake and Tricholoma magnivelare. Unfortunately, no sequence Journal compilation 2008 Blackwell Publishing Ltd P O P U L AT I O N S T R U C T U R E O F T. M AT S U TA K E 1243
information, including ITS sequence, is available in GenBankfor strains of Tricholoma zangii, including the type specimen,for comparison. Despite the absence of such information,the ITS results we obtained here are consistent with thehypothesis that the mushrooms analysed here belonged toone species, T. matsutake, and not its allied species. Inaddition, we saw no difference in ITS sequences betweenthose isolated from pine-dominated forests and those fromalpine oak-dominated forests. The GenBank Accessionnumbers for the 34 ITS sequences are EU294269–EU294302.
Because the ITS sequences for the 34 T. matsutake isolatesfrom the same or different regions in southwestern Chinawere all identical, we did not conduct any furtherpopulation genetic analysis of the ITS sequence data. Below,we present the analyses of the genotype information basedon SNP markers.
Fig. 1 Permutation analysis showing the relationship between the
number of assayed loci and genotype diversity in the total
population of Tricholoma matsutake from southwestern China. The
Genetic variation within geographical populations X-axis shows the number of randomly analysed loci and the Y-axisshows the genotype diversity. Each data point shows the mean and Our analyses identified abundant genetic variations within standard deviation of genotype diversity from 100 permutations.
each of the 17 geographical populations of T. matsutake from The maximum genotype diversity is 1 when all strains in the southwestern China. The analysed loci showed a high population have different multilocus genotypes. The minimum discriminating power among individuals. Based on results genotype diversity is 0 when all analysed strains have the same from randomizations, any eight of the loci analysed here multilocus genotype. Note the tapering off of genotypic diversity were sufficient to achieve a very high discriminating power (Fig. 1), indicating a level of saturation for identifying uniquegenotypes using the 14 SNP markers. The overall genotypicdiversity for the whole sample of 154 individuals was 0.966, Genetic variation among geographical populations suggesting that over 96% of the time, two randomlydrawn individuals from the total population will have Table 3 shows the F values between pairs of geographical genotypes different in at least one of the assayed loci. There populations. Overall, the mean F value was about 0.10, is, however, a range of variation in genotypic diversity among indicating that about 10% of the gene diversity was due to the 17 geographical populations, from a low of 0.583 in Lijiang geographical separations between pairs of populations. The to a high of 0.972 in both Jianchuan and Chuxiong (Table 2).
lowest F value (0.007) was found between Weixi and The overall results from the HWE tests suggest that Lijiang while the highest (0.232) was between Weixi and most loci in most populations were in Hardy–Weinberg Luliang (Table 3). The amova results suggested that all equilibrium, a result consistent with the hypothesis that three levels contributed significantly to the overall genetic recombination plays an important role in the natural popu- variation, with the regional level contributing the least lations of this species. Interestingly, most populations (Table 4). Specifically, the host tree species (i.e. between have two to three loci in Hardy–Weinberg disequilibrium regions) contributed 9% of the total genetic variance. The (Table 2). A close examination of the genotype data indicated next level, between populations within regions, contributed that two of the 14 polymorphic restriction sites consistently 34% of the total genetic variance, while the remaining 57% showed excess heterozygosity in all of the populations. These genetic variance came from within individual populations two sites were located on the same DNA fragment, RC14 (Table 4). Permutation analyses indicated that the amova (the raw genotype data are not shown here but see Table 2 results were consistent with significant genetic differentiations in Xu et al. 2007 for the representative excess heterozygosity).
among the 17 analysed populations (P < 0.01 for phiRT, Results from the index of association analyses similarly phiPR and phiPT in the amova tests).
suggested widespread random associations among allelesat different loci in these populations. Specifically, only two The relationship between genetic distance and geographical of the geographical populations showed any evidence of significant allelic associations among loci, one populationfrom Nanhua and the other from Ailaoshan. When all the The results from the Mantel tests are shown in Fig. 2. The 154 strains were analysed together, we saw no evidence of test showed a significant positive correlation between genetic significant allelic associations among loci (Table 2).
distance and geographical distance among the analysed 2008 The AuthorsJournal compilation 2008 Blackwell Publishing Ltd 1244 J . - P. X U E T A L .
Fig. 2 Results from two Mantel tests between genetic differences
and geographical distances among populations. (a) A Mantel test between Nei’s genetic distance and the two-dimensional geographicaldistances (based on longitudinal and latitudinal coordinates) amongpopulations. (b) A Mantel test between Nei’s genetic distance and altitudinal differences between populations. In both 2a and 2b, theX-axis represents the geographical distance parameter and the Y- axis represents Nei’s genetic distances between populations.
populations (Fig. 2a; P = 0.028), with a correlation coefficientof 0.318. However, we found no correlation betweenaltitudinal differences and population genetic distances Discussion
In this study, we analysed a large number of geographicalpopulations of the ectomycorrhizal mushroom Tricholomamatsutake from southwestern China. Our results identified significant genetic variation within and between populations.
There was evidence for recombination within each of the analysed populations. Overall, these populations showed relatively low but significant genetic differentiation. Inaddition, the amount of differentiation varies betweenpopulations, with the level of differentiation correlated to some extent to geographical distances separating the populations, consistent with results from a previous study using different samples (Chapela & Garbelotto 2004). Our values between geographical populations of analyses identified no correlation between genetic distance and altitudinal differences among populations. In addition, the habitat differences as represented by the dominant host trees contributed about 9% to the total genetic variation, less Pairwise
3
than a third of that between populations within habitats (34%) and a sixth of that within individual populations Journal compilation 2008 Blackwell Publishing Ltd P O P U L AT I O N S T R U C T U R E O F T. M AT S U TA K E 1245
Table 4 Summary results of the analysis of molecular variance (amova) within and among populations of Tricholoma matsutake from
southwestern China
d.f., degree of freedom; SS, sum of squared observations; MS, mean of squared observations; PhiRT, proportion of the total genetic variance that are between regions; PhiPR, proportion of the total genetic variance that are among populations within a region; PhiPT, proportion of the total genetic variance that are among individuals within a population The abundant genetic variations found in this study It is possible that this locus was located within or adjacent and in our recent study (Xu et al. 2007) among strains and to the mating-type locus in T. matsutake. Indeed, in heteroth- populations of T. matsutake from southwestern China form allic fungi, heterozygosity at the mating type locus is required a stark contrast to previous results obtained using either the for fruiting body formation (Raper 1966). Unfortunately, repetitive element MarY1 for fingerprinting (Murata et al.
little is known at present about the mating system in T. 2005) or sequence polymorphism at the ITS regions and the matsutake, including the number of segregating loci that intergenic spacer (IGS) region of the ribosomal RNA gene control mating compatibility and the number of functional cluster (Sha et al. 2007). In the study by Sha et al. (2007), alleles at each locus. The inferred recombination from our virtually no polymorphism was found among 56 fruiting data and those of Lian et al. (2006) and of Xu et al. (2007) do bodies collected from 13 counties in Yunnan when analysed suggest, however, that T. matsutake is likely a heterothallic using a specific PCR primer that had been shown to be species. Other possibilities for excess heterozygosity include highly polymorphic for strains from Japan. Our samples (i) heterozygous advantage due to complementary gene here included all 56 strains from the 13 locations in the actions, and (ii) genomic rearrangements that inhibit study by Sha et al. (2007). Though we have been unable to recombination and maintain heterozygosity (Xu 1995).
obtain samples from Japan for comparison using our In several of our populations, the analysed sample sizes genotyping method, the differences in the MarY1 finger- were very small. This was especially true for two populations, printing patterns between the Chinese and the Japanese Longling and Luliang, where only two and three isolates, samples seemed to suggest that there might be significant respectively, were obtained and analysed in this study. The genetic differentiation between the Japanese populations of small sample sizes and the relatively long distances between T. matsutake and those from southwestern China.
specimens within each of the populations make our samples Our within-population genetic analyses suggest that of limited use to infer the sizes of genetic individuals in sexual reproduction and recombination are widespread natural populations of this species in southwestern China.
among the geographical populations from southwestern A previous study in seven different sites in Japan identified China (Table 2). This conclusion is similar to what has been that the average size of T. matsutake genets was about 2 m in found in one Japanese population of T. matsutake where the longer dimension, with the largest about 11.5 m and the Hardy–Weinberg equilibrium was observed for four rest ranged from 0 to 5.0 m (Lian et al. 2006). Interestingly, microsatellite loci in a local population (Lian et al. 2006).
some mushrooms located very close to each other and However, in our study, two polymorphic nucleotide sites thought to belong to the same ‘shiro’ (‘genet’ in Japanese) on the same DNA fragment showed consistent excess of were found to have different genotypes at the assayed heterozygosity across all populations analysed here. While microsatellite loci (Lian et al. 2006). If the genet sizes of the detailed mechanism for this excess of heterozygosity T. matsutake in southwestern China were similar to those in is unknown, there may be several possibilities. One possibility Japan, the vast majority of the strains analysed here would might be due to ascertainment bias where the analysed represent different genets or ‘shiros’. Therefore, we believe locus was highly polymorphic but might not be represent- it unlikely that the low genotype diversity observed in ative of the rest of the genome. Ascertainment biases have some populations such as Lijiang was due to repeated been found in many SNP applications in a variety of species sampling of mushrooms from unusually large genets. The (Morin et al. 2004). In fungi, DNA fragments located within small genet sizes identified in Japan in combination with the mating-type locus or close to the mating-type locus the small sample sizes in many of the populations convinced often exhibit excess heterozygosities in natural populations.
us not to do analysis using clone-corrected samples. The 2008 The AuthorsJournal compilation 2008 Blackwell Publishing Ltd 1246 J . - P. X U E T A L .
establishment of long-term experimental sites would be Elevated levels of organic debris on the forest floor and needed to allow us to critically examine the size of genetic nitrogen levels in the soil have also been observed to individuals and the dynamics of microscale genetic variation contribute to decreased production of T. matsutake. Aside from in natural populations of this species in various regions in better management decisions based on these observations, we believe our population genetic study can also contribute The F values observed between geographical popu- to improved management strategy on sustainable resource lations of T. matsutake are similar to those reported in several utilizations of T. matsutake in southwestern China. Specifically, basidiomycete species (James et al. 1999; Xu et al. 2005). For the observation that sexual recombination plays a significant example, between regional populations of the button role in all the study populations in southwestern China mushroom, Agaricus bisporus, the F values were found suggests that sexual spores are important propagules for ranging between 0.019 and 0.076 (Xu et al. 1997). In the the reproduction of this species in nature. As a result, oyster mushrooms Pleurotus ferulae and P. eryngii, 19 and 6 management plans should enforce the notion that at each regional populations from Italy showed F values of 0.45 site, a certain number of mushrooms must be allowed to and 0.10, respectively, for these two species (Urbanelli et al.
mature and sporulate so as to allow future sexual repro- 2003). The intercontinental populations of the model duction. At present, in order to satisfy consumer demands basidiomycete Schizophyllum commune were significantly in Japan, virtually all T. matsutake mushrooms are picked differentiated when examined using isozyme markers, with prematurely before the veils of the mushrooms are ruptured an F value of 0.214 (James et al. 1999). In T. matsutake, to release spores. Leaving a few mushrooms at each site to Chapela & Garbelotto (2004) identified that among the seven mature and sporulate, in combination with careful harvesting strains they analysed, they found a positive correlation techniques that disturbs little of the underground mycor- between their pairwise genetic dissimilarity and geographical rhizae, should help populations to sustain. In addition, the distances. Those seven strains were far apart from each low but significant genetic differentiations among geo- other (up to over thousands of kilometres) and included graphical populations observed here suggest that while three from China (one northeastern China and two from long-distance gene flow is present, such a gene flow might southwestern China) and one each from France, Japan, not be strong enough to counter local genetic differentiation Korea, and Morocco. However, because of the small sample between certain populations. As a result, whenever possible, sizes (mostly only one strain from each country or geo- each site should leave a few mushrooms to mature and graphical area), the level of genetic differentiation among sporulate to ensure its continued reproduction.
geographical populations could not be assessed. At present,little is known about the genetic relationships of mostectomycorrhizal fungal populations from the 100-km ranges.
Acknowledgements
The statistically significant genetic differentiation between This study was supported by grants from the National Natural populations from forests dominated by different tree species Science Foundation of China (No. 30525002), the Knowledge seems to suggest that these populations are genetically Innovation Program of the Chinese Academy of Sciences (No.
distinct from each other. However, our analyses indicated KSCX2-YW-G-025), Genome Canada ( JX), PREA ( JX), and the that the level of differentiation between the two forest types Natural Science and Engineering Research Council (NSERC) of was low, less than one-third of the genetic differences between Canada ( JX). Members of Ailaoshan Biological Station of theChinese Academy of Sciences are acknowledged for their supports populations from within the same forest types and less than one-sixth of the genetic variation contributed from withinlocal individual populations (Table 4). The lack of sequencevariation at the ITS region between T. matsutake and T. zangii, References
and our population genetic analyses presented here suggestthat it has to be addressed by further detailed studies whether Agapow PM, Burt A (2001) Indices of multilocus linkage disequi- librium. Molecular Ecology Notes, 1, 101–102.
the taxonomic separation of T. zangii [syn. T. quercicola] Arnolds E (1991) Decline of ectomycorrhizal fungi in Europe.
(Zang 1990; Cao et al. 2003) and T. matsutake is justified.
Agriculture, Ecosystems and Environment, 35, 209–244.
While much remained unknown, many factors have Brumfield RT, Beerli P, Nickerson DA, Edwards SV (2003) The been suggested to influence the productions of natural utility of single nucleotide polymorphisms in inference of ectomycorrhizal mushrooms, including those of T. matsutake.
population history. Trends in Ecology & Evolution, 18, 249–256.
These factors include both biotic factors such as the type Brundrett M (2004) Diversity and classification of mycorrhizal and age of vegetation in the forests and abiotic factors such associations. Biological Reviews, 78, 473–495.
Cao ZM, Yao YJ, Pegler DN (2003) Tricholoma zangii, a new name for as nutrient levels, pH, and air quality (Arnolds 1991). For T. quercicola M. Zang. Mycotaxon, 85, 159–164.
example, in Europe, forests with pine trees more than Chapela IH, Garbelotto M (2004) Phylogeography and evolution in 40 years old have shown to exhibit reduced ectomycorrhizal matsutake and close allies inferred by analyses of ITS sequences mushroom production than those with younger trees.
and AFLPs. Mycologia, 96, 730–741.
Journal compilation 2008 Blackwell Publishing Ltd P O P U L AT I O N S T R U C T U R E O F T. M AT S U TA K E 1247
Gill WM, Guerin-Laguette A, Lapeyrie F, Suzuki K (2000) Matsutake Wang H, Wang Y, Chen J, Zhan Z, Li Y, Xu J (2007) Oral yeast flora — morphological evidence of ectomycorrhizal formation between and its ITS sequence diversity among a large cohort of medical Tricholoma matsutake and host roots in a pure Pinus densiflora students in Hainan, China. Mycopathologia, 164, 65–72.
forest stand. New Phytologist, 147, 381–388.
Xu J (1995) Analysis of inbreeding depression in Agaricus bisporus.
James TY, Porter D, Hamrick JL, Vilgalys R (1999) Evidence for Genetics, 141, 137–145.
limited intercontinental gene flow in the cosmopolitan mushroom, Xu J (2006a) Microbial ecology in the age of genomics and metage- Schizophyllum commune. Evolution, 53, 1665–1677.
nomics: concepts, tools, and recent advances. Molecular Ecology, Lian CL, Narimatsu M, Nara K, Hogetsu T (2006) Tricholoma 15, 1713–1731.
matsutake in a natural Pinus densiflora forest: correspondence Xu J (2006b) Fundamentals of fungal molecular population genetic between above- and below–ground genets, association with analyses. Current Issues in Molecular Biology, 8, 75–89.
multiple host trees and alteration of existing ectomycorrhizal Xu J, Yoell HJ, Anderson JB (1994) An efficient protocol for isolating communities. New Phytologist, 171, 825–836.
DNA from higher fungi. Trends in Genetics, 10, 26–27.
Morin PA, Luikart G, Wayne RK and the SNP Working Group Xu J, Kerrigan RW, Callac P, Horgen PA, Anderson JB (1997) The (2004) SNPs in ecology, evolution and conservation. Trends in genetic structure of natural populations of Agaricus bisporus, the Ecology & Evolution, 19, 208–216.
commercial button mushroom. Journal of Heredity, 88, 482–488.
Murata H, Babasaki K, Yamada A (2005) Highly polymorphic DNA Xu J, Mitchell TG, Vilgalys R (1999) PCR-restriction fragment markers to specify strains of the ectomycorrhizal basidiomycete length polymorphism (RFLP) analyses reveal both extensive Tricholoma matsutake based on ómarY1, the long terminal repeat clonality and local genetic differences in Candida albicans.
of gypsy-type retroelement marY1. Mycorrhiza, 15, 179–186.
Molecular Ecology, 8, 59–73.
Nakayama K, Nakanishi J (2004) Tricholoma matsutake in Xichang, Xu J, Ramos AR, Vilgalys R, Mitchell TG (2000) Clonal and Sichuan and Chuxioug, Yunnan, China. Abstract of the 48th spontaneous origins of fluconazole resistance in Candida albicans.
Annual Meeting of the Mycological Society of Japan, p. 45.
Journal of Clinical Microbiology, 38, 1214–1220.
Peakall R, Smouse PE (2006) genalex 6: genetic analysis in Excel.
Xu J, Cheng M, Tan Q, Pan Y (2005) Molecular population genetics Population genetic software for teaching and research. Molecular of basidiomycete fungi. In: Evolutionary Genetics of Fungi (ed. Xu Ecology Notes, 6, 288–295.
J), pp. 221–252. Horizon Biosciences, Norfolk, UK.
Raper JR (1966) Genetics of Sexuality in Higher Fungi. Ronald Press, Xu J, Guo H, Yang ZL (2007) Single nucleotide polymorphisms in the ectomycorrhizal mushroom Tricholoma matsutake. Microbiology, Redhead SA (1997) The pine mushroom industry in Canada 153, 2002–2012.
and the United States: why it exists and where it is going. In: Zang M (1990) A taxonomic and geographic study on the songrong Mycology in Sustainable Development (eds Palm MA, Chapela IH), (matsutake) group and its allied species. Acta Mycologia Sinica, 9,
pp. 15–54. Parkway Publishers, Boone, North Carolina.
Sha T, Zhang HB, Ding HS et al. (2007) Genetic diversity of Tricholoma matsutake in Yunnan Province. Chinese Science Bulletin, 52,
1212–1216.
The research interests of Tao Sha and Zhi-Wei Zhao are in the Urbanelli S, Della Rosa V, Fanelli C, Fabbri AA, Reverberi M (2003) conservation and utilization of biological resources, with a focus Genetic diversity and population structure of the Italian fungi on microbial and plant resources from southwestern China. Zhu- belonging to the taxa Pleurotus eryngii (DC. : Fr.) Quel and P. Liang Yang and Yan-Chun Li work on the taxonomy and ferulae (DC. : Fr.) Quel. Heredity, 90, 253–259.
biogeography of fungi in China, focusing on basidiomycetes in Wang Y, Hall IR (2004) Edible ectomycorrhizal mushrooms: southwestern China. Jianping Xu’s research interests are in the challenges and achievements. Canadian Journal of Botany, 82,
general areas of microbial ecology and evolutionary genetics.
2008 The AuthorsJournal compilation 2008 Blackwell Publishing Ltd

Source: http://www.kib.cas.cn/yjzwz/sdsys/zjytxyjz/tnh_fblw/201201/P020120116390221008342.pdf