Biodiversity of Trichoderma Community in the Tidal Flats and Wetland of Southeastern China

To investigate the biodiversity of Trichoderma (Hypocreaceae) and their relation to sediment physical and chemical properties, we collected a total of 491 sediment samples from coastal wetlands (tidal flat and wetland) in Southeast China. Further, we applied two types of molecular approaches such as culture dependent and independent methods for identification of Trichoderma spp. A total of 254 isolates were obtained and identified to 13 species such as T. aureoviride, T. asperellum, T. harzianum, T. atroviride, T. koningiopsis, T. longibrachiatum, T. koningii. T. tawa, T. viridescens, T. virens, T. hamatum, T. viride, and T. velutinum by the culture-dependent (CD) method of these, T. tawa was newly described in China. Subsequently, the culture indepented method of 454 pyrosequencing analysis revealed a total of six species such as T. citrinoviride, T. virens, T. polysporum, T. harzianum/Hypocrea lixii and two unknown species. Notably, T. citrinoviride and T. polysporum were not found by the CD method. Therefore, this work revealed that the combination of these two methods could show the higher biodiversity of Trichoderma spp., than either of this method alone. Among the sampling sites, Hangzhou, Zhejiang Province, exhibited rich biodiversity and low in Fengxian. Correlation and Redundancy discriminant analysis (RDA) revealed that sediment properties of temperature, redox potential (Eh) and pH significantly influenced the biodiversity of Trichoderma spp.


Introduction
Trichoderma speceis are economically important fungi, occurs in various environments at different geographical locations [1][2][3][4]. Trichoderma research have been received a global interest because of their biotechnological and agricultural applications. Especially, Trichoderma strains are extensively used as biocontrol agents to reduce plants infections caused by phytopathogens [5]. Biodiversity of Trichoderma in China is richly diversified because of varied ecological conditions [6]. Several microbial ecologists have been investigated Trichoderma biodiversity in various geographical locations in China [7].
According to Zhu and Zhuang [8], a total of 91 Trichoderma spp., have been reported from various ecological samples collected from China, such as agriculture substrate [7,[9][10][11][12][13][14] rhizosphere sediment [15], and edible fungi [16][17]. In which, most of strains are derived from agriculture resources and fewer from other resources. Moreover, researchers have mostly studied the biodiversity of Trichoderma spp., by using microscopic examination followed by sequencing of the internal transcribed spacer regions (ITS) and/or partial translation elongation factor-1α (tef-1) [3,7,18], but these methods cannot provide the information about culture independent strains due to close relationship between Trichoderma sp., [19]. Moreover, the accurate identification of Trichoderma spp., based on the morphological method is difficult because of morphological similarly between the species [20,21] and that can be identified only based on the multiple gene sequences of DNA characters [22,23]. Molecular-based multiple genes sequence [rDNA and genes encoding actin, calmodulin, endochitinase, RNA polymerase II, and translation elongation factor 1-alpha (tef-α)] analysis could offer the consistent identification of Trichoderma spp., [20,24,25]. To the best of our knowledge, there is no report on the biodiversity of Trichoderma in coastal wetlands in China and also there is no comparison studies on culture-dependent and independent methods in assessment of Trichoderma biodiversity. Therefore, in this work, we studied the biodiversity of Trichoderma spp., in coastal wetland through both culture-dependent and independent methods. For traditional molecular methods (culture-dependent), we used the two-genes approach of internal transcribed spacer regions 1 and 2 (ITS1 and ITS2) followed by translation-elongation factor 1-alpha (tef-α) [7,19] and culture independent method were used the 454 pyrosequencing technique. In addition, we analyzed the relation of culture-dependent and independent Trichoderma counts on sediment properties of coastal wetlands in Southeast China.

Ethics statement
No specific permissions were required for these locations/activities for native research institutes or researchers. In addition, this study was funded by Chinese government for the enrichment of public welfare and also this study did not involve endangered or protected species.

Description of study area and sample collection
The study area is located in the Southeast China coast of Yellow Sea (Yantai, Shandong; Lianyungang, Jiangsu), East Sea (Huangpu River, Fengxian, Chongming Shanghai; Hangzhou, Ningbo Zhejiang; Fuzhou, Fujian), South Sea (Shantou, Zhuhai, Guangdong), North Sea (Beihai Guangxi) [S1 Table]. A total of 491 sediment samples were collected from different coastal wetlands at two seasons of spring and summer, 2014. Prior to collect the sediment samples the sampling site was marked equally for 10 m x 10 m dimension (quadrate plot of 10 m 2 ) using a nylon rope and GPS reading was recorded and it was divided into four sub-sample plots (5 m x 5 m). At least one sample was taken from each sub-sample plot at 5 different sediment depths (0-20, 20-40, 40-60, 60-80 and 80-100 cm) by using a corer (1.5 m long stainless steel corer with 50 mm diameter) during low tide and made one composite sample according to the depth respectively. The composite of sediment were placed in pre-cleaned polyethylene bags and immediately transported to laboratory and stored at -20˚C for CD microbiological and pyrosequencing analysis.

In situ and laboratory sediment properties analysis
The sediment temperature, Eh, and pH were determined at the time of sampling. The temperature was measured using a thermometer with ±0.5˚C accuracy. Hydrogen ion concentration of the sediment sample was measured using a pH meter with a platinum electrode with an accuracy of ± 0.1, (pH 315i/ SET, Wissenschaftlich Technische Werkstatten, Germany) and calibrated with standard buffer solution prior to use. Porewater salinity was recorded by using a hand refractometer (Atago hand refractometer, Japan), after crushing a small amount of sediment through Whatman No.1 filter paper with a syringe. For this, a known amount of sediment sample was moisturized with double distilled water up to the moisture saturation level of the sediment.
In order to analysis the sediment properties by the laboratory experiments, the sediment samples were separately collected and immediately transferred to the laboratory in sterile polyethene bags. Plant roots, and other debris were removed from the samples and dried in an oven at 110˚C and ground to fine powder. Sediment composition of clay, silt and sand were analyzed using a hydrometer method [26]. Total organic carbon in sediment was estimated by adopting the method of El Wakeel and Riley [27].
The soil dilution method was adopted for isolation of the Trichoderma by using a Modified Potato-dextrose agar (PDAm) [28]. In detail, 1 g of sediment sample was serially diluted and one milliliter (10 4 dilution factor) of the serially diluted sample was pipetted out into sterile PDAm Petri-dishes (90 x 15 mm) and spread over the surface of the PDAm by an L-shaped glass rod and incubated at 25±2˚C for 5 to 7 days. After the incubation, the Trichoderma colonies were enumerated and calculated the Trichoderma load in the given sample using the standard formula for Colony Forming Unit (CFU) per gram of the sample. All the determinations were carried out in triplicate.

Identification of culture-dependent Trichoderma spp.
Trichoderma isolates were identified at the species level by using the combination of morphological [29][30][31][32] and microbial molecular analysis [7]. The isolates were purified by repeated sub-culturing and the pure cultures were stored by cryopreservation in 20% glycerol at -80˚C in School of Agriculture and Biology, Shanghai Jiao Tong University, P. R. China.

Genomic DNA sequence-based identification
Genomic DNA was extracted from mycelium of Trichoderma pure cultures grown on potato dextrose agar (PDA) for 3 d according to the method of Doohan et al [33]. The ITS 1 and ITS 2 regions of the rDNA gene were amplified using the ITS4 and ITS5 primers [34]. Followed by a fragment of the partial translation elongation factor 1-alpha (tef-α) gene was amplified using the primer EF728M [35] and tef1R [36]. The 50 μl reaction mixture consisted of 2 μl 50 ng/μl of template DNA, 25μl of 2x Taq PCR MasterMix with loading dye (Tiangen, China), 0.2 μl 100 mM of each primer, and 19 μl of MQ H 2 O. Polymerase chain reactions (PCR) were carried out under conditions described by Błaszczyk et al. [19]. Further, the PCR products were separated in 1.2% of agarose gel as described by Sun et al. [7].
Finally, 3 μl of the PCR products were purified with exonuclease I and shrimp alkaline phosphatase [37]. The sequences were analyzed by with a MegaBACE 1000 DNA automatic sequencing system (Pharmacia), with the DYEnamic ET Dye Terminator Cycle sequencing kit (Pharmacia). The obtained sequences were edited using Chromas V.1.43 (Applied Pharmacia). The successfully sequenced ITS and tef1-α genes were aligned using the CLUSTAL W Multiple Sequence Alignment Program [38].
Phylogenetic trees and molecular evolutionary analysis were made using the MEGA version 6.0 [42]. The phylogenetic tree was constructed according to the neighbor-joining, maximum parsimony, and maximum likelihood algorithms [43,44]. In order to evaluate the consensus of the branching, the bootstrap of phylogenetic tree analysis was employed with 1000 replicates of the data set. A suite of SPSS 11.5 software (IBM) was used for the statistical analysis to find the relation between culture depentent Trichoderma spp., counts (CFU) and other sediment properties (Sediment depths, sites or seasons) by ANOVA followed by post hoc test (Tukey's) and correlation analysis using Pearson's correlation method. In addition to this, PRIMER (Plymouth Routines in Multivariate Ecological Research (version 6.1.10) was used to analyze the diversity indices.
Culture independent Trichoderma spp., analysis Genomic DNA was extracted from the sediment samples using the TIANamp soil DNA kit according to manufacturer's instructions (Tiangen, China). After extraction, the purity of the DNA was tested using a UV spectrophotometer followed by 0.8% agarose gel electrophoresis at a voltage of 120 V for 20 min. The internal transcribed spacer (ITS) regions were amplified using forward primer: (5'-GGAAGTAAAAGTCGTAACAAGG-3') and reverse primer: (5'-TCCTCCGCTTATTGATATGC-3'). PCR reaction was conducted in 20 ng/μl of reaction mixture consisted of 8.75 μl of ultra pure H 2 O, 5 μl of 5x Q5 Buffer, 5 μl of 5x GC Enhancer, 2 μl of dNTP (2.5mM), 2 μl of template DNA (2ng/ μl), 1 μl each forward and reverse primer (10 μM), 0.25 μl of Q5 DNA polymerase. PCR cycling conditions were set as 4 min at 98˚C, 27 cycles (98˚C for the 30s, 47.6˚C for the 45s, and 72˚C for the 1 min), and then the final extension at 72˚C for 5 min, and the experiment was halted at 10˚C. PCR products were purified using AMpure Beads followed by the PicoGreen dsDNA assay kit used for the quantification of DNA. The mixture was pyrosequenced by using Roche 454 GS FLX (Shanghai Personalbio Co., Ltd., China). The sequenced data was analyzed by using Qiime (version 1.7.0, http:// qiime.org/) followed by version 1.31.2, http://www.mothur.org/ and further the data was analyzed according to Sun et al. [6]. In brief, the below sequences score of 25 and 200 bp length were trimmed and binned into operational taxonomic units (OTUs) using a 97% of the threshold for the bioinformatics and subsequent analysis [6]. Finally, the Redundancy discriminant analysis (RDA) of R-vegan and R-map tools for Linex were selected to find the relationship between the analyzed sediment environmental parameters and Trichoderma spp.

Biodiversity of culture-depentent Trichoderma spp.
A total of 254 isolates of Trichoderma (S2 Table) were isolated and identified based on a combination of morphological, ecological, and phylogenetic gene sequences of ITS1, ITS2 and translation-elongation factor 1-alpha (tef1); the nucleotide sequences were deposited in NCBI GenBank database (Accession numbers: KR868229-KR868570). Out of 254 isolates, a total of 13 species were identified such as T. aureoviride Rifai (12 isolates), T. asperellum  Table 1.
Trichoderma atroviride, T. harzianum and T. asperellum were abundant in sediment samples than other Trichoderma spp. According to the recent report of Jaklitsch and Voglmayr  Table 3. The number of isolates was found high in Ningbo, Zhejiang province and less in Lianyungang, Fengxian, and Beihai. Trichoderma atroviride and T. harzianum counts were high in the coastal region of Ningbo, Zhejiang province. Hence, these results revealed that the coastal wetland ecosystem of Zhejiang province was richly diverse for Trichoderma spp.

Phylogenetic analysis
The result of phylogenetic analysis based on the internal transcribed spacer regions 1 and 2 (ITS1 and ITS2) sequences of 52 Trichoderma isolates is shown in S1 Fig [3,7,46]. In the case of unambiguous identification of Trichoderma isolates by ITS1 and ITS2, the fragment of the tef gene was sequenced and phylogenetic analysis was performed (S2 Fig). As a result of this analysis, T. aureoviride, T. virens T. koningii, T. viride, T. atroviride, T. viridescens were resolved with high bootstrap support. Sixty-three strains were identified as T. harzianum, but this species known to include several ITS alleles [46]. In addition the tef1 tree clearly indicates the species complex of T. harzianum with the bootstrap support of only 17% with haplotypes and considered to be a species complex [32].

Effect of sediment depth, sampling seasons, and sampling sites on culture-dependent Trichoderma biodiversity
Trichoderma counts were significantly influenced by sediment depth, sampling seasons, and sampling sites (S3 Table). Trichoderma counts were higher in the Ningbo, Zhejiang province and lower in Fengxian, Shanghai. It was 0.480±0.03 CFU.g -1 sediment in Ningbo and 0.100 ±0.05 CFU.g -1 sediment in Fengxian. Trichoderma counts varied from 0.004±0.01 to 0.685 ±0.03 CFU.g -1 sediment at different sediment depths. The count was maximum at 0-20 cm depth and minimum in 80-100 cm depth at two seasons. Trichoderma strain counts ranged from 0.17±0.02 to 0.26±0.02 CFU.g -1 sediment at two sampling stations. The count was maximum in summer and minimum in spring (S3 Table).

Effect of sediment properties on culture-dependent Trichoderma biodiversity
Trichoderma counts were influenced by physical and chemical properties of sediment and the results are shown in the S4 Table. Trichoderma counts exhibited a positive correlation with temperature, Eh, and pH. All the correlated parameters were significant between sampling sites or seasons or sediment depths (S4 Table & S3 Fig). The significant correlation of pH revealed that while the pH was less than 5 (pH <5) the colonization of Trichoderma was reduced. Sediment temperature also played a significant role in the colonization of Trichoderma in sediment. Redox potential is a measure of oxygen in the sediment which revealed that sediment oxygen could enhance Trichoderma colonization. Sediment temperature ranged from 15.57±0.32 to 25.54±0.21˚C at different stations and it was higher in the Ningbo and minimum in the Chongming Island. In case of sediment depth it varied from 21.22±1.5 to 22.96±0.21˚C at depths and it was maximum at 0-20 cm depth and minimum in 80-100 cm depth. In case of seasons, it ranged between 17.93±0.13 and 20.95±0.14˚C at different seasons and it was maximum in summer and minimum in spring (S3 Table).
Redox potential ranged from -37.47±8.3 to -154.06±14.4 mV at different sites and it was higher in the Zhuhai and lower in the Chongming Island. In case of sediment depths, it ranged from -1.06±9.4 to 44.27±9.3 mV at depths and it was high at 0-20 cm depth and low at 40-60 cm depth. In case of seasons, it ranged between -66.93±5.89 and -89.36±5.94 mV at different seasons and it was maximum in summer and minimum in spring (S3 Table).
The pH was ranged from 7.55±0.12 to 8.61±0.1 at different sites and it was higher in the Fengxian and lower in the Shantou. In case of sediment depths, it ranged from 7.80±0.07 to 8.37±0.07 at depths and it was higher at 0-20 cm depth and lower in 80-100 cm depth. In case of seasons, it was ranged from 7.95±0.5 and 8.18±0.5 at different seasons and it was maximum in spring and minimum in summer (S3 Table).
The heatmap was drawn based on the Trichoderma spp., in accordance with the sampling sites, which were classified into the four groups (Fig 4). The abundance and richness of the uncultured Trichoderma spp., were higher in wt14 (33.33%), Trichoderma sp., was higher in wt15 (100%), T. citrinoviride was higher in wt12 (44.44%), T. virens was higher in wt11 (60%), T. polysporum was higher in wt18 (14.28%), and H. lixii was higher in wt15 (72.72%). Regarding species relationship the H. lixii and unculturable Trichoderma showed significant similarity and formed as one group (Fig 4).
Effect of sediemnt properties on culture independent Trichoderma spp.
The effect of sediemnt properties on distribution of culture independent Trichoderma spp.,was determined by Redundancy Discriminant Analysis (RDA) (Fig 5; S5 Table) The eigenvalues of the sediment properties was explained for 1,142,651 with the significant explanatory power of the RDA. According to the species variation data, a score of 1 st PC contribution was 49.52% and the score of the 2 nd PC contribution rate was 25.97%. The soil properties such as temperature, Eh and pH were significant at 5% level P values based on 999 permutations. The RDA analysis revealed that all the sediment properties were moderately influenced the culture independent Trichoderma biodiversity. Satatstical analsyis of relation between the sediemnt properties and culture-dependent and independent Trichoderma counts revealed that the temperature and Eh were significant factors and highly influenced the Trichoderma biodiversity (Fig 5, S3 Fig & S6 Table).

Discussion
Ninety-one Trichoderma spp., are previously been reported from China [7,8]. However, the China coastal ecosystem, interface between terrestrial and marine biotopes are significant in extent in Southeast China but has not been studied the distribution and/or ecology of Trichoderma spp. Hence this work assessed the culture-dependent and independent biodiversity of Trichoderma spp., from coastal wetlands in Southeast China by traditional molecular (culturedependent) and 454 Pyrosequencing (culture independent) methods.  Here and the previous study, T. harzianum was the predominant taxon [3,20,32,[46][47][48][49]. Trichoderma harzianum is frequently recorded in the genus, occurring in diverse ecosystem [19]. The species identified in the present study were: T. koningiopsis has been often isolated from natural substrata in tropical America, East Africa, Europe and Canada, and from ascospores [24]. Trichoderma hamatum isolated from soil are wildly used for the biocontrol of plant diseases caused by Sclerotinia sclerotiorum and Rhizoctonia solani and also trigger the plant growth [50,51]. Trichoderma longibrachiatum is isolated from Cerastoderma edule (Mollusc) [52] and T. viride is isolated from the rhizosphere soil of Avicennia marina with the nematocidal activity [53]. Trichoderma aureviride is isolated from the sea sediments of South China and showed the significant lytic enzyme activity [54] and T. atroviride is isolated from sediments of the root of mangroves Ceriops tagal and sponge Psammocinia sp. [55,56]. Trichoderma harzianum is isolated from marine sponge Gelliodes fibrosa & Suberites zeteki, Tethya aurantium with the antimicrobial activity [56][57][58][59][60].
The present study reports 13 Trichoderma spp., including a new record of T. tawa, from Hangzhou. Trichoderma spp., were significantly diversified in Hangzhou (10 species) and Ningbo (4 species), both are in Zhejiang province is provided a high biodiversity of Trichoderma spp., due to the extensive ecological conditions found in the coastline of this province. Several researchers have reported that the plant's litter decomposition supported higher microbial colonization because of nutrient richness [61][62][63]. We found that T. atroviride (134 isolates) followed by H. lixii (63) were the most common species in our study sites this result was supported both by culture-dependent and independent methods. Especially, H. lixii/ Trichoderma harzianum is a significantly predominant species in China and this was also supported by 454 pyrosequencing analysis showing higher 26% OTUs of H. lixii compared with other reported species. This result is consistent with previous reports of Trichoderma biodiversity in China [3,25,45,48].
The comparison of the culture-dependent and independent Trichoderma analysis revealed that the culture-dependent Trichoderma biodiversity was 46.15% higher than the culture independent Trichoderma biodiversity, as assessed by 454 pyrosequencing. The comparison of the pyrosequencing with traditional identification (CD) methods could enable us understating the Trichoderma biodiversity. Trichoderma citrinoviride and T. polysporum were found in 454 pyrosequencing but not in traditional (CD) methods, whereas Trichoderma harzianum and T. virens were commonly found in both traditional (CD) and 454 pyrosequencing methods. These results indicated the drawback in 454 pyrosequencing analysis based on ITS sequencing in estimating Trichoderma biodiversity [64].
Redundancy discriminant analysis (RDA) biplot and correlations analysis of the environmental variables revealed that the temperature, Eh and pH significantly influenced the biodiversity of culture-dependent and independent Trichoderma spp. The ecological variance is an important factor for the colonization of the microbes in the sediment [47,64]. Temporal and ecological conditions of sediments could influence the distribution of the Trichoderma spp. [47,65]. Hence, the climatic, spatial and temporal changes can play a significant role in the distribution of Trichoderma spp., in China.

Conclusion
The present study assessed the biodiversity of the Trichoderma spp., by culture-dependent and independent methods. A total of 13 Trichoderma spp., including a new report of T. tawa was recovered by traditional method (culture-depentent). In contrast, culture independent pyrosequencing, we identified a total of six Trichoderma spp. Thus both methods could provide complementary advantages; in other words, the combination of these two methods could reveal more biodiversity of Trichoderma spp. Since 454 pyrosequencing is an advanced method to study the culture independent microbes, the proper designing of primers and availability of specific database is required according to the microbe's specificity and it could enable the obvious study of microbial diversity. The sediment properties of temperature, Eh and pH could significantly influence the biodiversity of Trichoderma spp., from coastal wetlands in Southeast China.