Effects of endophytic fungi diversity in different coniferous species on the colonization of Sirex noctilio (Hymenoptera: Siricidae)

Diversity of endophyte communities of the host tree affects the oviposition behavior of Sirex noctilio and the growth of its symbiotic fungus Amylostereum areolatum. In this study, we evaluated the structure and distribution of endophyte communities in the host tree (Pinus sylvestris var. mongolica) of S. noctilio and eight potential host tree species in China. Overall, 1626 fungal strains were identified by using internal transcribed spacer sequencing and morphological features. Each tree species harbored a fungal endophyte community with a unique structure, with the genus Trichoderma common to different communities. The isolation and colonization rate of endophytes from Pinus tabulaeformis, followed by P. sylvestris var. mongolica, were lower than those of other species. The proportion of endophytic fungi that strongly inhibited S. noctilio and symbiotic fungus growth was significantly lower in P. tabulaeformis, P. sylvestris var. mongolica and P. yunnanensis. Further, the diversity of the endophyte communities appeared to be predominantly influenced by tree species and the region, and, to a lesser extent, by the trunk height. Collectively, the data indicated that P. tabulaeformis might be at a higher risk of invasion and colonization by S. noctilio than other trees.

The structure of fungal endophyte communities. The isolated 1626 endophytic fungi were assigned to 61 species and 34 genera (Table 1). Fifty-one species (83.6%) were identified based on ITS sequencing and the other ten species (16.4%) were identified based on morphological features. In HG, the endophyte richness of Pc. koraiensis (fifteen genera) was the highest, and that of P. koraiensis was the lowest with seven genera. The prevalence of Fusarium (2.9%) and Trichoderma (7.7%) of P. sylvestris var. mongolica was considerably lower than the other three tree species, while Aspergillus was more prevalent in P. sylvestris var. mongolica compared to the other three species. The endophyte richness was similar in JDZ (P. taeda, five genera; P. elliottii, six genera; and P. massoniana, six genera). The most prevalent were Penicillium and Trichoderma, which were detected in all tree species in JDZ. The endophytic fungi of P. tabuliformis in TL were identified into eight genera, and P. yunnanensis in DL were identified into ten genera ( Table 1). The particularly common genera were Trichoderma, Penicillium, Aspergillus, and Fusarium. Trichoderma isolated from nine tree species. However, 16 genera only were shared from different tree species, accounting for 47% of all the genera.
Vertical distribution of the endophytic fungi. The richness of endophytic fungi isolated from the base trunk segment was higher than that from the central and upper trunk segments in HG trees (Fig. 3). Considering the tree species, the structures of endophyte communities at different trunk heights of Pc. koraiensis and L. gmelinii were more complex than those from P. sylvestris var. mongolica and P. koraiensis. The richness of endophytic fungi at different trunk heights in trees from TL and DL was more homogeneous. In JDZ trees, the richness of endophytic fungi was lower in the base trunk segment than in the central and upper segments. The most prevalent fungi at different trunk heights were Fusarium tricinctum and T. atroviride in HG; Aspergillus versicolor in TL; A. niger and Penicillium chrysogenum in DL; and T. citrinoviride in JDZ ( Fig. 3; Table S1, Supporting Information). The same fungal species has not been found in all the trees in the current study.
The diversities and similarities of endophytic fungal communities from four conifers invaded by S. noctilio in HG. The diversity indexes of endophytic fungal communities isolated in the mixed coniferous forest invaded by S. noctilio in HG showed significant differences (Shannon's diversity index: F = 5.77, p < 0.05; Evenness index: F = 94.20, p < 0.05; Richness index: F = 48.25, p < 0.05) ( Table 2). Shannon's index and Richness index of endophytic fungal were the highest in Pc. koraiensis, and those of P. koraiensis had the lowest value ( Table 2). The similarity of the endophytic fungal communities of four tree species in HG was examined by NMDS using the Jaccard's index (Fig. 4a) and the Bray-Curtis distance matrix (Fig. 4b). The results showed that the similarity value of endophytic fungal communities of four tree species was low ( Fig. 4; Table S2 Supporting  Information). Altogether, only four fungi species (T. viride, T. atroviride, A. niger and F. tricinctum) was shared by the four trees in HG (Fig. 5).
The diversities and similarities of endophytic fungal communities from seven species of pine trees. The diversities of endophytic fungal communities from seven species of pine trees were significant Lowercase letters indicate a significant difference between the isolation rates in different tree species at p < 0.05. Uppercase letters indicate a significant difference between the colonization rates in different tree species at p < 0.01. www.nature.com/scientificreports www.nature.com/scientificreports/ difference (Shannon's diversity index: F = 5.52, p < 0.05; Evenness index: F = 20.19, p < 0.05; Richness index: F = 21.51, p < 0.05) ( Table 3). The diversity index value (2.2552 ± 0.13) of P. sylvestris var. mongolica was the highest in seven tree species; the three trees species in JDZ showed the lowest Richness index (P. taeda: 1.5647 ± 0.1; P. elliottii: 1.9557 ± 0.26; P. massoniana: 1.3854 ± 0.07) and Shannon's diversity index (P. taeda: 1.8930 ± 0.08; P. elliottii: 1.8065 ± 0.07; P. massoniana:1.7163 ± 0.12). Pinus taeda had the highest Evenness index value (0.9103 ± 0.023). No significant differences in the three indices of endophytic fungi diversity were noted among P. koraiensis, P. tabuliformis, and P. yunnanensis. According to the result of NMDS, no clustering of the endophytic fungal communities from seven species of pine trees was apparent. (Fig. 4a,b). The endophyte communities from three trees in JDZ were similar. The similarity of endophyte communities from seven pine trees was more apparent from the Jaccard's index coefficient than the Bray-Curtis (Fig. 4).

Discussion
Many studies have revealed that rich endophytic fungal communities have been isolated from a range of plants, which include beneficial species that have a negative impact on pests 29 . However, there is much less information on the relationship between endophyte communities and pest invasion. In this study, we analyzed the community structure and diversity of fungal endophytes from the host tree and other potential host trees of S. noctilio in China. We found that the different tree species did not share the most abundant and prevalent fungal species ( Table 1). The number of endophytic fungal taxa species was higher in HG. However, the endophytes were most numerous in trees from JDZ. The analysis also revealed that the specificity of endophytic fungi for different coniferous trees varied (Table 3). These observations were similar to ones of a previous study in which the diversity and differences of endophytic fungal communities were significantly different among different tree species 30 . However, the endophytic fungal communities are different in different tree species, resulting in differences in the resistance of trees to pests 31 .
The diversity of endophytic fungal community in the host greatly affects the selection behavior of S. noctilio and the growth of its fungal symbiont 24,25 . A. areolatum is essential for egg eclosion (by creating a suitable environment) and larval nutrition, and contributes to increased adult insect size and reproductive success 31 Pestalotiopsis www.nature.com/scientificreports www.nature.com/scientificreports/ female woodwasp probes the sapwood by shallow drilling into the host phloem using an ovipositor to find a suitable growth environment for the development of offspring and symbiotic fungus 33,34 . However, A. areolatum grows slowly and its ability to occupy an environmental niche is lower than that of other fungi. If various endophytic fungi colonize the host, interfering with the growth of the symbiont, the female will give up ovipositing and will only inject phytotoxic mucus, causing weakening of the host 35,36 . The CR and IR values of endophytic fungi of P. sylvestris var. mongolica was the lowest in mixed forests invaded by S. noctilio in HG. Among the pine species examined in the current study, the CR and IR were lower in P. tabuliformis and P. sylvestris var. mongolica than those of the other species (Fig. 1). Both of these tree species might constitute an advantageous environment for the adult female woodwasp to spawn and also benefit the growth of the woodwasp larvae.
On the other hand, no significant differences were observed in the CR or IR of P. sylvestris var. mongolica and the other trees species examined. Nevertheless, the primary endophytic fungal genera from the different tree species significantly differed. Many bioactive endophytes, including important biostimulants and bio-control agents, belong to such genera as Trichoderma, Cordyceps, Metarhizium, and Beauvaria, and nonpathogenic Fusarium species 37 . Therefore, endophytes belonging to these genera may present a strong resistance to the S. noctilio invasion. Trichoderma and Fusarium were the lowest in P. tabuliformis, P. sylvestris var. mongolica than in other tree species, while Aspergillus occupied a high percentage of P. tabulaeformis, P. sylvestris var. mongolica, and P. yunnanensis isolates (Fig. 2). Trichoderma exert a biological control effect against many pathogens and pests, and are characterized by rapid growth, antagonism, and parasitism 38,39 . Trichoderma were successfully applied for the treatment of pruning wounds on urban trees against colonization by wood decay fungi 40 . In fact, Trichoderma kill the mycelia of A. areolatum, while the inhibitory effect of Aspergillus against A. areolatum is less pronounced 25 . Therefore, P. tabulaeformis, P. sylvestris var. mongolica and P. yunnanensis might constitute more suitable hosts for the survival of the woodwasp larvae than other species. L. procerum, as an antagonistic fungus of A. areolatum, was isolated from P. tabulaeformis damaged by red turpentine beetle Dendroctonus valens 21,22,41 . But the fungi are rarely isolated from P. tabulaeformis in this study, and the reason may be due to the different health levels of P. tabulaeformis in the two experiments.
Ryan et al. 23 reported that two species, L. wingfieldii and O. minus, can impact the selection of the site of woodwasp spawning. Volatiles of some blue stain fungi exert a repellent effect on the adult female woodwasp and, hence, influence the selection of the oviposition location 24,25 . In the current study, we found that endophytic fungi that can repel the woodwasp are relatively more rarely in P. sylvestris var. mongolica and P. tabulaeformis than in other trees (Tables 1; S1, Supporting Information). Interestingly, earlier host selection experiments similarly indicated that S. noctilio shows the highest preference for P. tabulaeformis, followed by P. sylvestris var. mongolica, and can complete its life cycle on P. tabulaeformis and P. sylvestris var. mongolica.
The species diversity of endophytic fungi isolated from the leaf is higher than that of those isolated from the stem, although the frequency of isolation is lower 42 . Endophyte diversity in the stem is higher than the diversity in the corresponding trunk 30 . The data presented in the current study suggested that the vertical distribution of endophytic fungi was influenced by the tree species (Fig. 3). The base segment of the tree trunk of P. sylvestris var. mongolica harbored an endophyte community that was more species-rich than those of the central and upper segments (submitted information). According to a field survey, the emergence holes of S. noctilio adults are mainly located in the upper segment of the P. sylvestris var. mongolica trunk, with more dead larvae infected by endophytic fungi in the base trunk than in the upper trunk. This might be because of a greater presence of www.nature.com/scientificreports www.nature.com/scientificreports/ endophytic fungi that inhibit the growth of A. areolatum in the base trunk than in the upper trunk. Except for P. sylvestris var. mongolica and P. tabulaeformis, the abundance of endophytic fungi in the upper segment was higher than in the middle and base segments of other tree species (Fig. S1, Supporting Information). This might be conducive to the invasion and colonization by S. noctilio of the upper trunk of P. sylvestris var. mongolica and P. tabulaeformis. Further, the different tree species did not share endophytes, illustrating that some endophytes exhibit a tree-specific or segment-specific preference, leading to, for example, the presence of endophytes in P. sylvestris var. mongolica that only weakly inhibit S. noctilio.

Tree species Shannon-Wiener index (H′) Evenness index (J) Richness index (R)
Pc  Table 2. The diversities of endophytic fungal communities from four conifers in HG. Analysis of the similarity of endophytic fungal communities from four conifers in HG invaded by S. noctilio. The data were analyzed by one-way ANOVA followed by HSD test. The results are expressed as the mean ± SD (n = 9). The results followed by different letters are significantly different according to the HSD test (p < 0.05).
www.nature.com/scientificreports www.nature.com/scientificreports/ According to some studies, similar endophyte communities occur at close quarters [43][44][45] . In the current study, NMDS plots indicated that the diversity of the endophyte community was predominantly affected by tree species (different genera) in HG (Figs 4, 5). In addition, the endophytic fungal communities from seven species of pine trees has a large difference (Fig. 4). Concerning the distribution of S. noctilio in China, Carnegie 46 and Ireland 18 predicted that the regions from the northeast of Heilongjiang Province to the southwestern Yunnan Province (including the four areas evaluated in the current study) are climatically favorable for the establishment and persistence of S. noctilio, with all distribution records pointing to areas projected to be of moderate and high climatic suitability. Comparing the physical and chemical properties of host tree species in different areas invaded by S. noctilio, there is no obvious common characteristic 47,48 . P. taeda and P. elliottii are the host species of wasps    Table 3. The diversity indices of endophytic fungi from seven species of pine trees. Analysis of the similarity of endophytic fungal communities from seven species of pine trees. The data were analyzed by one-way ANOVA followed by HSD test. The results are expressed as the mean ± SD (n = 9). The results followed by different letters are significantly different according to the HSD test (p < 0.05).
www.nature.com/scientificreports www.nature.com/scientificreports/ in Canada and Brazil, and they are also distributed in the southern China. Therefore, it is necessary to study the effects of endophytic fungi in P. taeda and P. elliottii on the growth and development of S. noctilio larvae.
Although S. noctilio can endanger many pine trees worldwide, there are few systematic studies on the resistance of different hosts to S. noctilio and the growth of its symbiont 49 . The available research suggests that the resistance of the host tree itself likely contributes to maintaining the S. noctilio population in North America below damaging levels to a greater extent than the natural enemies of the woodwasp 20 . Because of the concealment of endophyte distribution in the host trunk tissue, their impact on natural communities and biodiversity may be easily overlooked. These inconspicuous mutualistic associations can, however, exert a tangible force on insect population dynamics that is qualitatively similar to that of natural enemies in maintaining the insect population in many ecosystems [50][51][52][53] . In the study, we analyzed the diversity of the communities of endophytic fungi from the established tree host and potential tree hosts of S. noctilio. From the perspective of endophytic fungi, preliminary analysis revealed that P. tabulaeformis enabled spawning of the adult female S. noctilio and survival of S. noctilio larvae. However, compared with other hosts of S. noctilio, P. tabulaeformis and P. sylvestris var. mongolica are characterized by the lowest wood hardness (as per Shore's hardness method, below 20) and the thinnest phloem (below 6 mm), which are beneficial for the spawning of S. noctilio. Currently, S. noctilio is distributed in northeast China, the main distribution area of P. sylvestris var. mongolica, with gradual spreading to middle and southern China. The suitable hosts may encourage the spread of S. noctilio throughout China 46,48 . Pinus tabulaeformis is mainly distributed in the central regions and of China and, hence, it may be at a high risk of being attacked by S. noctilio.

Materials and Methods
Study sites and sample collection. Material (Fig. 6). S. noctilio has invaded P. sylvestris var. mongolica in the mixed coniferous forest in HG. During the years 2015 and 2016, tree samples were collected from mixed forest (HG, TL, and JDZ) and pure forest stands (DL) (Fig. 6). Overall, 27 trees representing nine species (three repetitions per species) were randomly chosen from the four regions. Fresh wood samples were collected from three segments of the trunk, as follows: the base (0.1 m above ground), central segment (2.1 m above ground), and upper segment (4.2 m above ground). A trunk disk (10 cm-thick cross-section) was cut off from each segment. A bark layer more than 1 cm thick was removed from the disk using a sterile knife. Next, a 10 × 10 × 5 cm 3 block was removed from each disk and sealed in a sterile vacuum bag. All samples were transferred to the laboratory of the Beijing Forestry University and stored at 4 °C until further analysis. www.nature.com/scientificreports www.nature.com/scientificreports/ Isolation and storage of the endophytic fungi. Endophytic fungi were isolated from the sample blocks using a surface sterilization method 54 . Each sample block was cut with a sterile pruner into 25 fragments (size: 5 mm 3 ). Small fragments were surface-sterilized by dipping in a series of solutions (70% ethanol for 1 min, 12% sodium hypochlorite for 30 s, and 70% ethanol for 1 min). The pieces were then washed three times in sterile distilled water. Five surface-sterilized fragments were placed in a petri dish (90 mm), which contained potato dextrose agar (PDA: 200 g potato, 20 g glucose, 15 g agar, and 1 L distilled water) supplemented with 100 μg mL −1 ampicillin and 50 μg mL −1 chloramphenicol. All samples were incubated at 25 ± 1 °C and 70 ± 5% relative humidity (RH) for 1~4 weeks or until the emergence of the mycelia. Agar cubes (ca. 1 mm 2 ) were removed aseptically from the edge of the colonies and transferred to fresh PDA plates. Each colony was transferred at least three more times until a visually uniform culture was obtained. For long-term preservation, the mycelia and spores were transferred to 20% glycerol in ultra-clean distilled water (v/v) and stored at −80 °C. Fungal cultures were generated on PDA slants in centrifuge tubes and stored under sterile mineral oil at 4 °C.
Identification of endophytic fungi. The endophytic fungi were identified based on both morphology and internal transcribed spacer (ITS) sequencing. The endophytic fungi were first identified using ITS sequencing. DNA was extracted from fungal mycelia from fresh cultures, using the Extract-N-Amp tissue polymerase chain reaction (PCR) kit (Sigma-Aldrich Corporation, USA), following the manufacturer's instructions. The fungal ribosomal ITS1 (ITS1), 5.8S (where present), and ITS2 regions were amplified using fungal-specific ITS1 and ITS4 primers 55 . The PCR reactions were carried out in a volume of 25 μL using 23 μL Golden Medal MIX (Thermo Scientific, USA), 1 μL of each primer, and 1 μL template DNA. Amplification was conducted using the following settings: an initial denaturation step of 98 °C for 2 min; followed by 30 cycles that included denaturation at 98 °C for 10 s, annealing at 50 °C for 15 s, and polymerization at 72 °C for 15 s; and a final extension step of 5 min at 72 °C.
The PCR amplification products were separated by electrophoresis on 1% (w/v) agarose gels and stained with ethidium bromide for visual examination. The PCR products were purified using the agarose gel DNA extraction kit (Takara, Japan) and sequenced at Qinke Biotech (Beijing, China). The sequences were submitted to BLAST search in the GenBank (http://blast.ncbi.nlm.nih.gov/Blast.cgi). Sequences sharing ≥99% similarity with a partial 28S rDNA sequence (ca. 600 bp) were considered as representing identical species.
When the sequences shared <99% similarity with known species, morphological features were used to identify the endophytic fungi. The following morphological features were evaluated: mycelium shape, mycelium surface texture, colony color, production of pigments and their diffusion in the medium, spore production, and mycelium growth rate on the PDA plates. The endophytic fungi that did not sporulate on this medium were transferred to the malt extract agar (MEA, 2%) plates and to plates with xylogen extracts of the host to activate sporulation. The following characteristics were evaluated for the anamorph: the conidiomata, conidiogenous cells, conidiophores, and conidia morphology (e.g. size, color, shape, and ornamentation). The following characteristics were evaluated for the teleomorph: the sporomata and their associated structures, and spore morphology 56 . Ultimately, the species of the endophytic fungal isolates were determined. Diversity analysis. The colonization rate (CR) was calculated as the number of fragments from which one or more endophytic fungi were isolated, divided by the total number of incubated fragments 57 . The isolation rate (IR) was defined as the number of endophytic fungi isolated, divided by the total number of fragments incubated 58,59 . The CR and IR were analyzed using one-way ANOVA. The differences between mean values were evaluated using Tukey's honestly significant differences (HSD) test. A Chi-square test was applied to analyze the data for some tree species. The statistical analyses were performed using the IBM SPSS Statistics version 23.0 (Chicago, IL, USA). The differences and distribution of endophytic fungi isolated from each tree species were examined using the range diversity analysis 60 . The analysis of variance was used to test for differences in endophyte richness among tree species.
The diversity of endophytic fungi isolated from seven species of pine trees and four conifers in HG were evaluated using the Shannon-Weiner Index (H′), Evenness Index (J), and Margalef richness index (R); the differences between the indices were analyzed using one-way ANOVA.
where N is the total number of individuals; Ni refers to the number of individuals; and S indicates the total number of species. The differences in endophyte community structure identified at different trunk heights of each tree species and the four conifers in HG were analyzed respectively using Venn diagrams (GraphPad Prism 7, San Diego, CA, USA). Endophyte communities were compared by using non-metric multidimensional scaling (NMDS), using the R package VEGAN (version 2.3-0). Two NMDS plots were constructed, each based on a different calculated similarity index: the Jaccard's index, based on the presence/absence of taxa among tree species 61 ; and the Bray-Curtis coefficient, based on the incidence and abundance of taxa in the tree species 61 .
Informed consent. All experimental protocols were approved by Key Laboratory for Silviculture and Conservation of Ministry of Education, Beijing Forestry University, Beijing, China.