Identity and Specificity of Rhizoctonia-Like Fungi from Different Populations of Liparis japonica (Orchidaceae) in Northeast China

Mycorrhizal association is known to be important to orchid species, and a complete understanding of the fungi that form mycorrhizas is required for orchid ecology and conservation. Liparis japonica (Orchidaceae) is a widespread terrestrial photosynthetic orchid in Northeast China. Previously, we found the genetic diversity of this species has been reduced recent years due to habitat destruction and fragmentation, but little was known about the relationship between this orchid species and the mycorrhizal fungi. The Rhizoctonia-like fungi are the commonly accepted mycorrhizal fungi associated with orchids. In this study, the distribution, diversity and specificity of culturable Rhizoctonia-like fungi associated with L. japonica species were investigated from seven populations in Northeast China. Among the 201 endophytic fungal isolates obtained, 86 Rhizoctonia-like fungi were identified based on morphological characters and molecular methods, and the ITS sequences and phylogenetic analysis revealed that all these Rhizoctonia-like fungi fell in the same main clade and were closely related to those of Tulasnella calospora species group. These findings indicated the high mycorrhizal specificity existed in L. japonica species regardless of habitats at least in Northeast China. Our results also supported the wide distribution of this fungal partner, and implied that the decline of L. japonica in Northeast China did not result from high mycorrhizal specificity. Using culture-dependent technology, these mycorrhizal fungal isolates might be important sources for the further utilizing in orchids conservation.


Introduction
The mycorrhizal association is ubiquitous but very important symbiosis in nature, which plays an essential role in the maintenance of most terrestrial ecosystems [1]. Over 90% of all plant species can form mycorrhizas with different kinds of fungi, and the existence of mycorrhizal fungi can confer to their hosts many adaptive advantages via improved water and nutrient/ minerals uptake from the soil [2,3,4], enhanced plant growth [5,6], reduced toxic element accumulation [1,7], and increased resistence to pathogen damage [8]. Nowadays, based on their essential role in plant species and the whole ecosystems, more attention has been given to the potential of exploiting fungi for effective use [9,10]. Although a large number of mycorrhizal fungi existed in plants, only a fraction has been described and explored to date, and their function in ecological systems remains indistinct.
The Orchidaceae, which is one of the largest and most diverse plant families, is distributed worldwide [11]. However, many orchid species have suffered dramatic declines in distribution and some species have become rare and endangered in recent decades [12]. Mycorrhizal association is known to be important to orchids because they depend on the presence of suitable fungal partners for seed germination and seedling development [13,14]. Therefore, a complete understanding of the mycorrhizal fungi of the many threatened orchid species is required for conservation action plans [15].
Although molecular taxonomic identification of the endophytic fungi of orchid species has now revealed that the diversity of orchid associates is much complex [16], the study of the earliestdiverging orchid lineages and distribution of mycorrhizal fungal associates across orchid phylogeny supported that the ancestral state is an association to the Rhizoctonia-like fungi lineages [17], and orchid mycorrhizas are predominantly represented by associations between photosynthetic plants and Rhizoctonia-like fungi [15]. The Rhizoctonia-like fungi includes members of the Ceratobasidiaceae, Sebacinales and Tulasnellaceae [15,17]. However, the taxonomy, diversity and distribution patterns of mycorrhizal fungi may vary considerably among different orchid species or different populations within the same species, ranging from very narrow specificity to little specificity [18][19][20][21][22][23][24][25][26]. Therefore, knowing the identity of the fungi that form mycorrhizas with orchids and the specificity of the relationships are of crucial importance for orchid ecology and conservation [27].
Liparis japonica, which is a terrestrial photosynthetic orchid species, has a wide geographic distribution in the tropical and subtropical areas such as China, Korea, Japan and Russia [28]. Historical record indicated that this species had a relatively wide distribution in Northeast China. Unfortunately, the number of L. japonica populations has declined sharply during the last several decades and many populations recorded previously has disappeared [29]. In a previous study, we assessed the genetic diversity and population genetic structure of L. japonica in Northeast China, and found potential restricted gene flow existed among populations due to habitat destruction and fragmentation [29]. To conserve surviving wild populations and reintroduce plants into declining populations requires a more complete understanding on the mycorrhizal fungal partners of L. japonica species.
In this study, we investigated in detail the taxonomy and diversity of culturable mycorrhizal fungi associated with L. japonica in Northeast China using morphological and ITS phylogenetic analysis. Seven different populations of L. japonica species were examined in order to evaluate how fungi diversity varied within and between habitats. Using a culture-dependent method, the fungal isolates obtained will be important sources for deep study and further utilizing in orchid conservation.  Table 1 and Fig. 1. Plants were excavated with a clod of their surrounding soil and transferred to the laboratory. The field studies did not involve endangered or protected species, and no specific permissions were required for these locations.

Fungal isolation and morphological identification
Potential mycorrhizal fungi were isolated from the orchid plants and identified from pure culture. L. japonica, unlike many other temperate, terrestrial orchids which have thick roots and often produce abundant pelotons, has few active pelotons suitable for isolation. Thus, endophytic fungi were isolated from single hyphal tips emerging from sterilized root portions as in the isolation method described by [30]. Three roots per plant were carefully cleaned from the soil under running water, surface-sterilized in 0.1% (v/v) Mercuric chloride and 75% (v/v) ethanol for 3 min and 5 s respectively, and subsequently washed three times in sterile distilled water. Root sections of 3-5 mm thickness were obtained by cutting and placed in a petri dish with potato dextrose agar (PDA). Petri dishes were incubated at 25uC in the dark and observed for fungi growing every 2 days for at least 3 weeks. The growing colonies were separated onto fresh media for purity and this process was repeated three times. Classification of the endophytic fungi was based on their growth rate and morphological characteristics, including colonial morphology, production of conidiogenous cells, conidial size and dimension on PDA medium [31], and similar isolates were grouped into one morphotype. The Rhizoctonia-like fungal endophytes were recognized by the following characteristics: hyphas hyaline with constricted branch points, 2.5-7 mm diam; submerged growth in PDA; ellipsoid, globose or irregular monilioid cells 10-20610-20 mm; colony creamy white to pale tan or orange, rubbery or leathery in appearance and texture [32][33][34].

DNA extraction, amplification and sequencing
Genomic DNA was extracted from each Rhizoctonia-like fungi isolate as described by [35]. ITS sequences of rDNA were amplified using the following primer pairs: ITS1-OF/ITS4-OF [36]. PCRs were performed in 50-mL volumes containing approximately 10 mg of fungal genomic DNA, 16PCR buffer (Sangon Biotech Co. Ltd), 1.5 mM MgCl 2 , 200 mM each of dNTP (dATP, dCTP, dGTP, and dTTP), 100 nM of each primer and 2.5 unit of Taq DNA polymerase (Sangon Biotech Co. Ltd). PCRs were carried out with a BIO-RAD T100 tm Thermalcycler with an initial incubation at 95uC for 5 min, followed by 35 cycles with denaturation at 95uC for 30 s, annealing at 60uC for 1 min and extension at 72uC for 1 min, and a final extension reaction at 72uC for 10 min. The PCR products were purified using the TaKaRa Agarose Gel DNA Purification Kit (TaKaRa, Dalian, China) and sequenced using the BigDye terminator v3.1 on an ABI 3730xl DNA Analyzer (Applied Biosystems, Forster City, California, USA). New ITS sequences from this study have been deposited in GenBank (the accession numbers is from KF537635 to KF537658).

Data analysis
ITS sequences were analyzed with BLAST [37] against the NCBI sequence database to find the closest sequence matches in the GenBank database (http://www.ncbi.nlm.nih.gov). Genus and species of the database match were accepted whenever identity between our sequence and that of the database was greater than 97% [38]. Sequences used to construct phylogenetic trees were from organisms considered to be closely related to the mycorrhizal fungus of L. japonica based on BLAST searches, and additional orchid mycorrhizal fungi sequences (mainly based on [19]) were also chosen for phylogenetic analysis. Selected sequences were aligned using the program Clustal_X 2.1 [39]. Alignments were checked for ambiguities and adjusted manually when necessary and the alignment gaps were treated as missing information.
Phylogenetic analyses were estimated using maximum parsimony (MP), Neighbor-joining (NJ) and Bayesian inference (BI) methods in PAUP* version 4.0b10 [40], MEGA version 4 [41] and MrBayes version 3.0b4 [42], respectively. For MP analysis, the bootstrap support values were estimated by 1000 replicates with 10 random sequence additions and tree-bisection-reconnection (TBR) branch swapping. For NJ analysis, genetic distances Table 2. Distribution of 201 fungal isolates in different morphotypes. were calculated using the Kimura two-parameter model, and the bootstrap support values were estimated by 1000 replicates. Prior to the Bayesian analysis, the Akaike information criterion (AIC) in Modeltest version 3.7 [43] was used to select the best-fit model of molecular evolution for each dataset. For BI analysis, four chains of the Markov Chain Monte Carlo (MCMC) chains were run, sampling one tree every 1000 generations for 3,000,000 generations, starting with a random tree. Bayesian posterior probabilities (PP) were calculated from the majority rule consensus of the tree. The sequence of Septobasidium carestianum (GenBank accession number DQ241448) was selected as outgroup [21].

Fungal distribution and morphological diversity
After isolation and purification, a total of 201 isolates of endophytic fungi from 185 plants were obtained from seven L. japonica populations. According to their morphological characters and growth rate on PDA medium, fungal isolates were classified into nine morphotypes with four to seven morphotypes in each population ( Table 2). The isolates in morphotypes I and II were observed in all populations studied and represented the majority of the fungal community (61.7%), while isolates from other morphotypes were not common and occurred only at one or a few populations (Table 2). Morphological characters and detailed descriptions of the nine morphotypes were given in Fig. 2 and Table 3, and according to the colony and micro-morphological characters, we identified morphotype I as Rhizoctonia sp., morphotype II as Phomopsis sp., and morphotype III to IX as Verticillium sp., Fusarium sp., Chaetomium sp., Gliocladium sp., Cylindrocarpon sp., Phialophora sp., and Paecilomyces sp., respectively. Only the Rhizoctonia-like isolates in morphotype I were given codes and subjected to further collection of mycelium for DNA extraction and phylogenetic analysis.
Of the 185 individuals of L. japonica samples, each of 180 samples gave rise to one Rhizoctonia-like isolate, whereas the other five samples gave rise to two Rhizoctonia-like isolates, and no sample gave rise to three or more Rhizoctonia-like isolates.

Phylogenetic analysis
About 640-bp portions of the ITS were sequenced for the 24 representative isolates of morphotype I, with 3 to 4 isolates within each population. All sequences identified by BLAST in NCBI were from Tulasnellaceae strains. Thus, the representative Tulasnella spp. that have been identified were selected for phylogenetic analysis with the sequences obtained in this study, and a total of 54 different sequences were used.
The overall topology of the trees produced with Maximum parsimony (MP), Bayesian inference (BI) and Neighbour-joining (NJ) methods were similar based on our ITS data (Fig. 3). The MP strict consensus tree (Fig. 3) showed that fungal associates of L. japonica all belonged to the genus Tulasnella, and all isolates belonged to a single major clade. Some mycorrhizal isolates from other orchid species which have been identified as T. calospora, and a mycorrhizal isolate (Epulorhiza sp., AM040890; [44]) from L. loeselii species in Hungary also fell in this major clade.

Discussion
Compared with the direct DNA amplification technique, identification of orchid mycorrhizal fungi with the pure culture technique has been criticized because of the bias toward fastgrowing and easily cultured fungi [15]. However, the culturedependent technique is still used in a vast majority of orchid mycorrhizal fungi studies, mainly because fungi partners of these plants were usually easy to isolate and no significant differences were found between the pure culture technique and direct DNA amplification technique [20,23,[45][46][47]. Unlike many other temperate, terrestrial orchids which have thick roots and often produce abundant pelotons, L. japonica has few active pelotons suitable for isolation and direct PCR identification, so we used standard culture method for fungi identification in this study, and  the results were credible by the striking observation of the same fungal clade isolated across such a wide geographical distribution of L. japonica. Moreover, although the molecular method is powerful for fungi identification, isolation and cultivation of the mycorrhizal fungi isolates are still needed for deep study and further utilizing in orchids conservation.
In the present study, various endophytic fungi were isolated from roots of L. japonica and identified by morphological and the rDNA ITS sequences, and most of the identified endophytes of L. japonica are Basidiomycetes and Ascomycetes. In general, endophytic fungi in orchid plants are very abundant, but it seems likely that only Rhizoctonia-like fungi species may form mycorrhizal associations with host [48]. In this study, 86 endophyte fungi isolates belonging to the Tulasnella genus were isolated (42.8%) based on morphological characters and phylogenetic relationships reconstructed from ITS sequence, and they existed in every L. japonica population. Although some endophytes have been shown to be able to increase the tolerance of some plant species [49], little is known about the role of endophytes for orchid performance, and the non-Rhizoctonia-like endophytic fungi also deserve additional study.
The specificity of orchid mycorrhizal associations has important implications for orchid biology, conservation and restoration of orchid populations [50]. Specific fungal partners may lead to enhanced seed germination rates, and the efficiency of nutrient exchange between partners may be heightened with specific plantfungus combinations [45]. Our results showed that all mycorrhizal fungi isolates obtained belonged to a single clade of Tulasnella, and the molecular analysis showed that they are all representatives of the same species, T. calospora, which indicates that the relationship between L. japonica and its mycorrhizal fungus is very specific regardless of habitats, at least in Northeast China. There are more and more reports showed that the degree of specificity can vary among species independent on orchid types [19,22]. There is also evidence that a preference exists among some orchids for specific fungal partners, and some photosynthetic orchids, even when sampled over a wide range, have a single dominant mycorrhizal fungus, for example, Spiranthes sinensis [33], Goodyera pubescens [51], and Pterostylis nutans [52]. Our results also supported that high specificity in widespread and photosynthetic orchids may be more common than previously thought. Moreover, the phenomenon that multiple fungal taxa existed in a single root system was detected in some photosynthetic orchids which indicated that multiple fungi in a single root system may occur more often in photosynthetic than in mycoheterotrophic orchids [26]. While our results showed that this characteristic is not universal in orchid mycorrhizal systems.
From another perspective, the mycorrhizal specificity we observed might not be accurate for the reason that our results were based on a single collection of root tissue in adult L. japonica plants. However, the diversity and identify of orchid mycorrhizal fungi may vary according to different life stages of plant. Specificity during germination and early development has been detected in orchids, whereas the adult stage of plant growth in the same species can have less specific fungal partners [13,51,53]. Accordingly, fungi required for germination and recruitment of L. japonica might be different from those associated with adult plants, and this will be the next logical steps to investigate the pattern of mycorrhizal association.
Although L. japonica had a strong preference for T. calospora species across the sampled range, the fungus did not have an equivalent specificity. T. calospora has been shown to be a very common mycorrhizal fungus of many terrestrial orchid species [15], and the fact that T. calospora stains were present in different populations of L. japonica in Northeast China in our study and many other orchids in other regions of the world indicates its wide distribution in different ecological environments. Because of this, we considered that the decline of L. japonica in Northeast China was not resulted from fungal specificity, although there exists a point that a dependence on narrowly specific interactions with fungi and pollinator may predispose many orchids to become rare [15,45,54].