Mongolian pine forest decline by the combinatory effect of European woodwasp and plant pathogenic fungi

Interactions between the decline of Mongolian pine woodlands and fungal communities and invasive pests in northeastern China are poorly understood. In this study, we investigated the fungal communities occurring in three tree samples: the woodwasp Sirex noctilio infested, healthy uninfested and unhealthy uninfested Mongolian pine trees. We analyzed the relationships of the Mongolian pine decline with fungal infection and woodwasp infestation. Twenty-six fungal species were identified from the sampled trees. Each tree sample harbored a fungal endophyte community with a unique structure. Pathogenic fungi richness was four times higher in infested and unhealthy un-infested trees compared to that in healthy uninfested trees. Sphaeropsis sapinea was the most dominant pathogenic fungus in the sampled Mongolian pine trees. The number of S. noctilio was higher than native bark beetles in the declining Mongolian pine trees. The invasion of the woodwasp appeared to be promoted by the fungal infection in the Mongolian pine trees. The incidence of S. noctilio infestation was higher in the fungi infected trees (83.22%) than those without infection (38.72%). S. sapinea population exhibited positive associations with within-tree colonization of S. noctilio and bark beetle. Collectively, these data indicate that the fungal disease may have caused as the initial reason the decline of the Mongolian pine trees, and also provided convenient conditions for the successful colonization of the woodwasp. The woodwasps attack the Mongolian pine trees infected by fungi and accelerated its decline.


Scientific Reports
| (2021) 11:19643 | https://doi.org/10.1038/s41598-021-98795-y www.nature.com/scientificreports/ this insect is not considered a pest in Europe 6,8 , but in several countries of the southern hemisphere and North America, the woodwasps has attracted considerable attention due to its high invasive ability and ability to kill a variety of pine species 9,10,11 . In August 2013, the woodwasps were first detected as a pest of Mongolian pine in the Duerbote Mongolian Autonomous County, Heilongjiang Province, China. To date, Mongolian pine plantations are considered to be in danger of the infestation by the woodwasp over 22 cities in northeast China 12,13 . Sirex noctilio damages pine trees by depositing an obligate symbiotic fungus, Amylostereum areolatum (Fr.) Boidin and a phytotoxic mucus in the trees during oviposition. The toxic mucus affects tree defenses and assists the colonization of fungus in the host. The symbiotic fungus acts as an external gut of the woodwasp larvae for the digestion of recalcitrant lignocellulosic compounds 14,15 . Thus, insects, toxins, and fungi act together to damage host trees. Mongolian pine tree decline is commonly considered as a multifactorial disease, in which many interacting abiotic and biotic factors such as drought, frost, insect pests and pathogens are involved. Among the biotic factors involved in the onset of Mongolian pine tree decline, pathogenic fungi play a primary role. So far, more than 20 fungi diseases of Mongolian pine trees have been reported 4,5 . In particular, many independent surveys have demonstrated the involvement of some leaf blight agents such as Lophodermium seditiosum Minter Stalay and Millar 16 , Coleosporium phellodendri Kom, Lophodermella sulcigenaa (Link) Tubeuf 17 , Septoria pinipumilae Sawada 18 , and trunk parts agents such as Cronartium quercuum 19 (Berk.) Miyabe ex Shirai and Cronartium flaccidum (Alb. et Schw) Winter 20 , and root rot agents such as Rhizoctonia solani Kühn 21 , in the Mongolian pine decline processes. However, in recent years, it is shown that the colonization of some important invasive pests may contribute to accelerating the Mongolian pine tree decline. It is unknown whether the decline of P. sylvestris var. mongolica forest invaded by S. noctilio is related to the fungal communities in host trees.
Previous studies have shown that the declining trees were preferentially infested by the woodwasps, however, when the population density was high, they also infested the healthy trees 10 . A number of pathogenic fungi have been recognized as having a prominent role in Mongolian pine tree decline and mortality 5 . The most common method of the investigation is to cut down the declining pine trees to control the woodwasp and has achieved great results. One drawback of many investigations, however, was from single, separate disciplines (e.g., climatologists, plant pathologists, entomologists, etc.), and led to broach only one possible cause at a time, without a comprehensive, holistic approach to the problem 22 . The result was in many instances a disjointed and often incomplete, which made it impossible to determine the real causes of tree declines.
Sphaeropsis sapinea is an important latent pathogen of Pinus spp. and widely distributed in P. radiata plantations in northern Spain. It was recognized as the most widespread necrotrophic ascomycete pathogen responsible for dramatic losses of pine trees across the continents 23,24 . Sphaeropsis sapinea has emerged as an aggressive fungal pathogen all over the world and could directly invade the young shoots of pine trees [25][26][27] . In the greenhouse experiment, Stanosz and Flowers proved that S. sapinea strains isolated from healthy and diseased pine tissues had high pathogenic potential 28,29 . Usually, numerous pycnidia of S. sapinea are present in forest stands occurring on twigs, needles branches and stems of pine trees. A high infection rate may pose a high risk to forests when there are disease-triggering factors, e.g., hail or insect feeding or extreme weather conditions such as heat and drought, as in the years 2018 und 2019 in German 24 .
The frequency of pine shoot blight on pine trees has significantly increased over the past decades in southeast China, especially in mature pine forests 5 , and the loss caused by S. sapinea is no less than that by Bursaphelenchus xylophilus (the most serious invasive species of conifer trees in China). In northeast China, the infection with S. sapinea in Mongolian pine trees was also reported 30 . However, the reasons that cause the decline of the Mongolian pine forests invaded by S. noctilio are unknown.
In this study, we hypothesize that the fungal disease acts as the initial reason for the decline of the Mongolian pine forest in northeast China, the woodwasps attack the infected and stressed trees and accelerate the decline. We investigated the fungal species occurring in three tree samples: S. noctilio infested trees, healthy uninfested trees, unhealthy uninfested. We analyzed the relationship between the decline of pine trees and the occurrence of pathogenic fungi and the woodwasp infestation. In addition, we also studied the pathogenicity of S. sapinea to healthy P. sylvestris var. mongolica trees. The findings obtained in this study allow us to characterize, for the first time, the relationship between the decline of Mongolian pine trees and fungal communities and invasive pest in northeast China pine plantations.

Number of S. noctilio and other borers.
In total 162 S. noctilio individuals, including 141 adults and 21 larvae (16 dead and 5 alive) from tree bolts, were observed from the infested trees, but no S. noctilio individual was found in healthy uninfested and unhealthy uninfested trees (Fig. 1).
Apart from the woodwasp which specifically attacks P. sylvestris var. mongolica, other wood-boring pests (all of them were bark beetles) were eventually found in sampled trees. The number of bark beetles collected was 12 and 15 in unhealthy uninfested and infested trees, respectively (Fig. 1). No insects were found in healthy uninfested trees. Native bark beetles (Ips sexdentatus) were collected not only in S. noctilio infested trees, but also in unhealthy uninfested trees. The number of S. noctilio had a significant positive correlation with the number of native bark beetles ( Table 1).
Top-eight most prevalent fungi species (genera) accounted for 90% of all the isolates, ranging from 86.1 to 92.3% (Fig. 3). The relative frequency of S. sapinea in healthy uninfested trees was lower than those of S. noctilio infested and unhealthy uninfested trees. The relative frequency of Trichoderma spp. was slightly higher in S. noctilio infested (15.3%) and unhealthy uninfested trees (14%) than that of healthy uninfested trees (11.4%), and Aspergillus spp. and Fusarium spp. in healthy uninfested trees were higher than the other two tree samples.
Four fungal species, namely Aspergillus niger, A. alternata, S. sapinea, and T. atroviride were isolated from three tree samples, and common in three tree samples accounting for 15.3% of all the species. The highest overlap (Jc = 0.381) was observed for the fungal communities between S. noctilio infested and unhealthy uninfested trees (Fig. 4). Some fungal species only existed in a single tree sample (healthy uninfested: 5 species; unhealthy uninfested: 4 species; infested: 7 species). The species Leptographium lundbergii and O. minus were isolated from S. noctilio infested and unhealthy uninfested trees, whereas C. globosum was only species isolated from healthy uninfested trees (Table 2; Fig. 3).
A total of 11 pathogenic species were identified from three tree samples, including 2 pathogenic species from healthy uninfested trees, 8 pathogenic species from unhealthy uninfested trees and 8 pathogenic species from S. noctilio infested trees ( Table 2; Supporting information Figure S1). The pathogenic fungi richness was four times higher in infested and unhealthy uninfested trees than in healthy uninfested trees. Some pathogenic fungi found in unhealthy trees were also isolated from healthy trees uninfested by S. noctilio. For example, S. sapinea (pathogen of pine shoot blight) was isolated from all three samples and the isolation rate was significantly higher compared to other fungi. S. sapinea exhibited positive interspecific associations with the within-tree colonization S. noctilio and bark beetle ( Table 1).

Diversities of the fungal community.
The diversity indexes of endophytic fungal communities showed significant differences among the three tree samples (Shannon diversity index: F = 6.72, df = 2, p < 0.05; Simpson dominance index: F = 43.47, df = 2, p < 0.05; Richness index: F = 21.25, df = 2, p < 0.05). The Shannon diversity index was higher, and the Richness index was lower for the fungal community from healthy uninfested trees than those from infested and unhealthy uninfested trees ( Table 3). The Richness index was the highest in S. noctilio infested trees compared to other two tree samples. The Simpson dominance indexes of infested and unhealthy uninfested trees were higher than that of healthy uninfested trees, demonstrating that fungal communities under these two conditions had a high concentration compared to healthy trees. In addition, the Simpson dominance index was slightly higher in unhealthy uninfested trees than that in the infested trees community (F = 1.78, df = 1, p > 0.05).
Infection ability of S. sapinea to healthy P. sylvestris var. mongolica trees. The incidence rates of S. sapinea infection to the needles of healthy P. sylvestris var. mongolica trees were significantly different in the two treatment groups (wounded + spore and nonwounded + spore) compared with that in the two negative control groups (wounded + water and nonwounded + water) (F = 318.74, df = 3, p < 0.01) ( Table 4). S. sapinea showed strong pathogenicity to the wounded P. sylvestris var. mongolica needles (incidence, 83.22%), which eventually caused the needles to wither. However, S. sapinea could also penetrate P. sylvestris var. mongolica without wounding but with a lower incidence of 38.72% (F = 122.99, df = 1, p < 0.01). In addition, pathogenic fungi re-isolated from diseased needles were the same as in the inoculum used for the healthy needles previously (Table 4). In the two negative control groups, the needles of healthy P. sylvestris var. mongolica trees could hardly be infected, regardless of whether they were wounded or not.

Discussion
Recently, the woodwasps have been found in declining Mongolian pine woodlands in northeast China 12 . The current study revealed a positive association between S. noctilio and native bark beetles (Ips sexdentatus) as reported previously (Table 1) 31 . However, the population number of the bark beetles was lower and not considered as a pest in northeast China over the past several years 30 . The current study also found 162 woodwasps only from P. sylvestris var. mongolica trees with signs of wasp egg laying (Fig. 1) and in unhealthy and S. noctilio infested P. sylvestris var. mongolica trees, as reported previously that this insect preferred to damage declining pine species [32][33][34] . However, the Mongolian pine woodlands had been declining before the invasion of S. noctilio in northeast China. This declining may be due to the fungal community in the Mongolian pine woodlands, which provides convenience for the invasion of S. noctilio 12,30 . The woodwasps attack stressed trees, particularly disease-stressed ones, which are their preferred hosts 35 . This accelerates the decline and even death of Mongolian pine trees.
The association of Mongolian pine decline with fungal communities has been shown previously 4 . In this study, a total of 26 fungal species was isolated from three tree samples. Pathogenic fungi richness was four times higher in the S. noctilio infested and unhealthy uninfested trees than that in the healthy uninfested trees (      Figure S1). Some of common pathogens of pine needles constitute a danger to weakened pine stands 36 . Ophiostoma minus was the second most common fungus in this study. It was reported that the woods colonized by O. minus dries more quickly 37 . Leptographium lundbergii and O. minus are considered as blue stain fungus of different pine trees worldwide, which is introduced to pine trees by bark beetles 38,39 . In contrast, C. globosum was the most frequent fungal isolates only found in the healthy uninfested trees ( Table 2). It is a biocontrol fungus as it produces various secondary metabolites and enzymes capable of inhibiting the mycelia growth of pathogenic fungi 40,41 . Recent research showed that C. globosum completely inhibited the mycelial growth of Amylostereum areolatum 42 .
Amylostereum areolatum showed up only in one sample in S. noctilio infected trees although the woodwasp attack on trees was widespread in this study. The growth rate of A. areolatum is low and the ability of occupying the resources and niche is weaker than many other endophytic fungi present in the pine ecosystems. Previous research found that some endophytic fungi can inhibit the mycelial growth of A. areolatum and destroy the mutual symbiotic relationship between S. noctilio and A. areolatum in host trees 42,43 . In addition, in this experiment, the xylem tissue of P. sylvestris var. mongolica trees was selected randomly, and the spawning site of the woodwasps was not selected specifically, so the number of isolated symbiotic fungi was very small.
Previous studies have found that the species of endophytic fungi are closely related to the health level of trees 43 . In this study, Trichoderma, Aspergillus were the dominant genera of endophytes (Table 2) as reported in different host plants [44][45][46] . The Richness index showed that endophytic fungi species in the S. noctilio infested trees were the highest. The CR and IR values of endophytic fungi of the healthy uninfested trees was the lowest compared with the values of the S. noctilio infested and unhealthy uninfested trees ( Table 2; Fig. 2). The highest similarity (0.38) was observed for the fungal communities between the S. noctilio infested and unhealthy uninfested trees (Fig. 4). The results show that the fungal community structure is greatly affected by tree health conditions 43 .
On the other hand, the invasion of the woodwasps accelerated host decay and promoted the colonization of saprophyte, such as Fusarium solani 40,47 . For example, the symbiotic fungus was only isolated from the S. noctilio infested trees in this study. Furthermore, no significant differences were observed in the CR or IR values between the unhealthy uninfested and S. noctilio infested trees. The primary endophytic fungal species from unhealthy uninfested and S. noctilio infested trees were also similar. The results of Simpson dominance index showed that fungal communities had a high concentration in unhealthy uninfested trees compared to that in other two tree samples (Table 3).
In this study, S. sapinea was the most abundant species obtained from tree trunks of S. noctilio infested (37.3%) and unhealthy uninfested (47.7%) trees (Table 2) and showed a very strong pathogenicity and could penetrate P. sylvestris var. mongolica without wounding (Table 4). S. sapinea is the causal fungal agent of Diplodia tip blight disease to the coniferous trees of relevance to forestry in the world (Supporting information Table S1). The severity of pathogenicity, the length of incubation period and propagation period of the fungi are related to the host tree vigor, tissue maturity and environmental conditions 44 . Palmer reported that S. sapinea strains from China could invade pine trees without wounding, while those from the United States could not 48 . However, Blodgett found that S. sapinea strains from the United States could also invade pine trees without wounding, but the incidence was low 49 . The occurrence of S. sapinea in healthy pine trees of this study, measured in frequency of colonization, is higher than in other studies like by Zhou 50 and Maresi 51 . In addition, the occurrence of S. sapinea had positive associations with both S. noctilio and bark beetle (Table 1), which could be driven by the attraction of the unhealthy trees infested by S. sapinea to the woodwasps to oviposit during host selection.
In our opinion, Pinus sylvestris var. mongolica forests in northeast China are being damaged by S. sapinea and other fungi, and these fungal diseases was getting worse year by year, causing the trees to decline 44 . Their infection may promote convenient conditions for the successful colonization of S. noctilio. Therefore, we considered the decline of Mongolian pine forests could be the result of the combinatory effects of S. noctilio and plant pathogenic fungi.

Materials and methods
Study sites and wood sample collection. The research site was in the Jun De Forest Farm (130° 17′ 47′′ E, 47° 12′ 11′′ N) in Hei longjiang Province, China, which is comprised primarily of 25-30 years old P. sylvestris var. mongolica, P. koraiensis, Picea koraiensis, and Larix gmelinii plantations. The site was characterized by a cold climate with an average annual temperature of 3.7℃ and average annual precipitation of 600 ~ 650 mm. The Mongolian pine trees selected in this study came from the sample plot of pure P. sylvestris var. mongolica forests (with an area of 2 hectares) previously investigated (unpublished data) and has not been thinned since planting. Some trees showed signs of decline and had been damaged by the wasp S. noctilio as previously reported 13,52 . In April 2018, fifteen trees were randomly chosen from the pure P. sylvestris var. mongolica plantation and listed Table 4. Pathogenicity of S. sapinea to healthy P. sylvestris var. mongolica. Incidence followed by different letters were significantly different according to the Tukey's HSD test (p < 0.05).  Table 1, including three groups: 5 S. noctilio infested trees, 5 healthy uninfested trees and 5 unhealthy uninfested trees. The S. noctilio infestation of Mongolian pines was identified by typical oviposition symptoms (i.e., resin beads formed from each ovipositor insertion). The distance between individual trees was at least 10 m. In fact, the distance between the uninfested healthy trees and S. noctilio infested trees was 10 m, and the distance between other trees exceeded 30 m (Supporting information Figure S2). Fresh wood samples were collected from tree trunk segments of 2 m above ground 44 (Table 5). Briefly, a trunk disk (10 cm-thick cross-section) was cut off from the segment. A bark layer more than 1 cm thick was removed from the disk using a sterile knife. Next, a sample block (10 × 10 × 5 cm 3 ) was removed from each disk and sealed in a sterile vacuum bag. All sample blocks were transferred to the laboratory at Gansu Agriculture University and stored at 4 °C (up to 2 weeks) until further analyses.

Collection of pests from Sirex infested trees.
Fifteen Mongolian pine trees were cut into 1 m-long billets after the wood sample collection, excluding the bottom 1 m section, with a minimum 10 cm diameter. After sealing the cut ends with wax, sample logs were taken to the quarantine laboratory in Gansu Agriculture University. These logs from visually identified tree samples (healthy uninfested, unhealthy uninfested, infested) were individually placed in mesh cages in the rearing facility and maintained at 27 ± 3 °C temperature and 65 ± 5% relative humidity (RH) until the adult S. noctilio emerged. The adult numbers of S. noctilio and other pests were counted from the tree samples. In addition, the wood sample were split to count the number of S. noctilio larvae after the adult emergence.
Isolation and storage of endophytic fungi. Endophytic fungi were isolated from the sample blocks using a surface sterilization method 53 . Briefly, each sample block was cut with a sterile pruner into 30 fragments (size: 4 ~ 5 mm 3 ). The 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). They were then washed three times in sterile distilled water. Five surface-sterilized fragments were placed in a petri dish (90 mm) with potato dextrose agar (PDA: 200 g potato, 20 g glucose, 15 g agar, and 1L distilled water) supplemented with 100 μg/mL ampicillin and 50 μg/mL chloramphenicol. All fragments were incubated at 25 ± 1 °C and 70 ± 5% RH for 1 ~ 4 weeks or until the emergence of fungal mycelium. Agar cubes (ca. 1 mm 2 ) were removed aseptically from the edge of fungal colonies and transferred to fresh PDA plates. Each fungal colony was transferred at least three times until a well-defined uniform culture was obtained. Purified fungal isolates were sub-cultured with half-strength PDA in 60-mm Petri dishes and kept on the laboratory bench at about 20 ~ 25 °C, where they received indirect sunlight to enhance sporulation. The fungal isolates were initially grouped as representative isolates and classified by their macro-and micro-morphological features, such as colony appearance, size, and shape of spores with species descriptions available in the literature 54 .
The fungal cultures were generated on PDA slants in centrifuge tubes and stored under sterile mineral oil at 4 °C. For long-term preservation, the representative isolates of each taxon were transferred to 20% glycerol in ultra-clean distilled water (v/v) and stored at − 80 °C.

Molecular identification of isolates.
For the determination at the species level, the representative isolates of each taxon identified by the morphological features above were grown on PDA and incubated at 25℃ in the dark using InstaGene Matrix (BioRad Laboratories, Hercules, CA, USA). Genomic DNAs were extracted from 5-day-old cultures. The primers ITS1 and ITS4 55 were used to amplify the internal transcribed spacer (ITS) regions by PCR. 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 (10umol/L), and 1 μL template DNA (50 μg/mL). The PCR amplification was conducted using the following conditions: an initial denaturation step of 98 °C for 2 min; followed by 30 cycles of denaturation at 98 °C for 10 s, annealing at 50 °C for 15 s, and polymerization at 72 °C for 15 s; and then a final extension step of 5 min at 72 °C.
The PCR products were separated by electrophoresis on 1% (w/v) agarose gels, stained with ethidium bromide for visual examination, and purified using the agarose gel DNA extraction kit (Takara, Japan) and sequenced at Qinke Biotech (Beijing, China). The sequences were submitted for BLAST search in the GenBank (http:// blast. ncbi. nlm. nih. gov/ Blast. cgi). The representative isolates were assigned to a species when their sequences were at least 99% identical to the sequence of a known species. Besides, morphological features of the representative isolates were also used an important role to confirm the classification by the DNA sequence comparison. 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.  37 , and then the spores were washed with sterile water and 50 spores were collected under a low power microscope (Wincom, China) to make into fungus suspension. Then, the inoculation experiment was conducted on the needles of 3-year-old healthy seedlings of P. sylvestris var. mongolica in the laboratory 30 . First, the needles of P. sylvestris var. mongolica seedlings were stabbed with a sterile knife at the base of the needles, with one wound per needle. The uninjured needles were used as a control treatment. Then, the fungus suspension, prepared as above, was smeared on the stabbed needles of P. sylvestris var. mongolica with a brush and bound with self-adhesive plastic film for 10 days. The uninjured and stabbed needles smeared with sterile water served as negative controls. In this experiment, the fungus was inoculated twice (once more after 10 days). Ten independents healthy seedlings of P. sylvestris var. mongolica were used for each of the four treatments (stabbed and uninjured needles smeared with fungus spores and with water) with 19 ~ 54 needles each seedling. The incidence rates of S. sapinea were investigated after 3 months post-inoculation. After the incidence rate of needle infections was determined, S. sapinea was re-isolated from 20 diseased needles randomly selected from each treatment group.
Data analysis. The colonization rate (CR) was calculated as the number of tree fragments from which one or more endophytic fungi were isolated, divided by the total number of incubated trees fragments 56 . The isolation rate (IR) was defined as the number of endophytic fungi isolated, divided by the total number of tree fragments incubated 57 . The incidence rate was calculated as the number of diseased needles, divided by the total number of inoculation needles. The CR and IR of endophytic fungi and the incidence of S. sapinea to healthy P. sylvestris var. mongolica were analyzed using one-way ANOVA. The differences between mean values were evaluated using Tukey's honestly significant differences (HSD) test. Pearson's chi-square test was applied to analyze the differences between pathogenic fungi and other fungi (remaining fungi except for pathogenic fungi) from each tree sample. We analyzed within-tree correlations of presence/absence of S. sapinea and pests using phi (φ) coefficients. The statistical analyses were performed using the IBM SPSS Statistics version 23.0 (Chicago, IL, USA). The relative frequency of the common fungi isolated from each tree sample was examined using the range diversity analysis 58,59 . The diversity of endophytic fungi isolated from each tree sample was evaluated using the Shannon-Weiner Index (H′), Simpson dominance index (D), and Margalef richness index (R) 60 .
where N is the total number of individuals; Ni refers to the number of individuals; and S indicates the total number of species. In addition, the similarity of fungal communities was evaluated using the Jaccard similarity coefficient (Jc) 61 . The similarities in fungal taxonomic richness between communities were summarized in Venn diagrams using GeneVenn software (http:// genev enn. sourc eforge. net/).

Informed consent. All experimental protocols were approved by Biocontrol Engineering Laboratory of
Crop Diseases and Pests of Gansu Province, Gansu Agricultural University, Lanzhou, China. All the methods were carried out in accordance with the relevant guidelines and regulations.