Diversity of Ophiostomatoid Fungi Associated with Dendroctonus armandi Infesting Pinus armandii in Western China

Pinus armandii (P. armandii) is extensively abundant in western China and, as a pioneer tree, and prominently influences local ecology. However, pine forests in this region have been significantly damaged by Dendroctonus armandi (D. armandi) infestations, in close association with ophiostomatoid fungi. This study aimed to identify the diversity of ophiostomatoid fungi associated with D. armandi infesting P. armandii in western China. A total of 695 ophiostomatoid fungal strains were isolated from 1040 tissue pieces from D. armandi galleries and 89 adult beetles at four sites. In this study, based on multiloci DNA sequence data, as well as morphological and physiological characteristics, seven species belonging to five genera were identified including three known species, Esteyea vermicola, Graphium pseudormiticum and L. wushanense, two novel taxa, Graphilbum parakesiyea and Ophiostoma shennongense, and an unidentified Ophiostoma sp. 1. A neotype of Leptographium qinlingense. Ophiostoma shennongense was the dominant taxon (78.99%) in the ophiostomatoid community. This study provides a valuable scientific theoretical basis for the occurrence and management of D. armandi in the future.

Pinus armandii (P. armandii) is a native and pioneer coniferous species on the Qinling Mountains in China [14] that has a significant role in local economy and ecology. However, P. armandii has been infested with Dendroctonus armandi (D. armandi) since it was first reported in 1932 [15]. D. armandi is an endemic beetle species that gregariously attacks 20-to 50-year-old healthy, P. armandii mainly in western China [16,17]. To date, in an area spanning more than 4000 ha, approximately half a million P. armandii trees have been decimated by this beetle [18,19].
Research on the diversity of ophiostomatoid fungi associated with D. armandi on P. armandii in China remains limited. There are no systematic studies, with only a few sporadic ophiostomatoid fungi reports [23][24][25]. Furthermore, the status of L. qinlingensis has been challenged, due to the absent type specimen, molecular analysis and limited morphological characteristics to prove that L. qinlingensis was a new species [24]. Therefore, the validity of L. qinlingensis should be considered if similar material is obtained from the same vector and host [29].
In this study, we explored the ophiostomatoid communities associated with D. armandi infecting P. armandii ecosystems in western China. Integrated morphological observations and multilocus DNA sequence data were used to analyze these communities. Our results provide insights into the communities of ophiostomatoid fungi associated with D. armandi in western China, which is a basic assignment for the subsequent study on the occurrence and management of D. armandi.

Sample Collection and Fungi Isolation
Samples including D. armandi adults and their breeding galleries were collected from infected P. armandii trees at four sites (Table 1) in western China from July to August 2018 and May to July 2019 ( Figure 1). All four sites are pure forests of P. armandii with tree ages of approximately 40 years old and diameters of approximately 40 to 60 cm. The trees used in this study showed signs of being dead or dying. The beetles were individually placed in sterilized Eppendorf tubes using tweezers, while their galleries were placed in sterile envelopes using a sterilized knife. Beetles and galleries were returned to the laboratory and stored at 4 • C for isolation within one week.
Beetles were crushed directly without superficial disinfection and transferred to a 2% malt extract agar (MEA: 20 g Biolab malt extract, 20 g Biolab agar, and 1000 mL deionized water). Galleries were cut into smaller tissue sections (5 × 5 mm), disinfected with 1.5% sodium hypochlorite (NaClO) for 60 s, rinsed with sterile water three times, and placed in 9 cm petri dishes, as described by Seifert et al. [30]. All strains were purified using mycelium apex, and cultures were grown in the dark at 25 • C. According to the preliminary analysis of culture characteristics, representative strains of each morphotype were selected for further morphological and molecular studies. All fungal strains obtained in this study were maintained in the culture collection of the Chinese Academy of Forestry (CXY), and representative strains were maintained in the China Forestry Culture Collection Center (CFCC, part of the National Infrastructure of Microbial Resources) ( Table 2).

Morphological and Physiological Characteristics
Pure cultures were incubated in the dark at 25 • C, culture morphology and growth status were observed daily, and the microstructures of reproduction forms were performed on 2% MEA media and incubated for 7 to 30 days. Microscope slides were prepared to observe the length and width of reproductive structures (such as conidiogenous apparatus, stipes cylindrical, conidiophore, and conidia) per strain using a BX51 OLYMPUS microscope with differential interference contrast. In total, 30 measurements were repeated for each morphological feature, and the statistics were presented as (min-) (mean-SD) -(mean + SD) (-max) (mean, average; SD, standard deviation; min, minimum; max, maximum).  For growth rate studies, representative strains were cultured in 90 mm diameter plates in the dark. A total of five replicate plates were included for each strain incubated at 5 • C intervals (5 to 35 • C) for two weeks. The diameter of each colony was measured daily until the mycelium reached the edge of the MEA medium. Colony colors were described according to Rayner's color chart [31].

DNA Extraction and Sequencing
Before DNA extraction, the strains were grown on 2% MEA for 1-2 weeks at 25 • C in the dark. The mycelia of purified strains were picked from the 60 mm diameter plates and placed into 2 mL sterile Eppendorf tubes. DNA extraction and purification were performed using the Plant Genomic DNA Kit (Invisorb Spin Plant Mini Kit, DP305, Tiangen, Beijing, China), following the manufacturer's protocol.
A total of five DNA regions were amplified for sequencing and phylogenetic analyses. The internal transcribed spacer regions (ITS1 and ITS2, including the 5.8S gene) were amplified using the ITS1/ITS4 primer pair [32]; the nuclear ribosomal large subunit region (LSU) was amplified using the LROR/LR5 primer pair [33]; ITS2 and part of the ribosomal large subunit 28S (ITS2-LSU) were amplified using the ITS/LR3 primer pairs [32]; the β-tubulin (TUB2) gene was amplified using the BT2a/BT2b primer pair [34]; and the elongation factor1-α (EF1-α) gene was amplified using the EF1F/EF2R primer pair [35]. PCR reactions were conducted in 25 µL volumes (2.5 mM MgCl 2 , 1× PCR buffer, 0.2 mM dNTP, 0.2 mM of each primer, and 2.5 U Taq-polymerase enzyme), and PCR amplification was conducted using a thermocycler (Applied Biosystems, Foster City, CA, USA). The reaction conditions for the five DNA regions were similar to those described in the references for primer design. PCR products were cleaned with an MSB Spin PCR Apace Kit (250) following the manufacturer's instructions. All nucleotides were sequenced in both directions using a CEQ 2000 XL capillary automated sequencer (Beckman Coulter), and MEGA5.0 was used for splicing.

Phylogenetic Analyses
Preliminary identification of the obtained sequences based on ITS DNA fragments was performed using BLAST searches in the NCBI GenBank database. The related authentic sequences were downloaded for further phylogenetic analyses. Sequence alignment was performed online using MAFFT (http://mafft.cbrc.jp/alignment/server/ accessed on 13 December 2021), implementing the iterative refinement method (FFT-NS-i setting) [36], and edited with MEGA5.0. The gaps were treated as the fifth base. Maximum likelihood (ML), maximum parsimony (MP), and Bayesian inference (BI) were used to assess these aligned sequences in phylogenetic analysis. ML analyses were conducted using RAxML v. 7.0.3, [37] under the GTR-GAMMA model. Supports for the nodes were estimated from 1000 bootstrap replicates. MP analyses were performed using PAUP*version 4.0b10 [38]. A bootstrap analysis (1000 replicates using the neighbor-joining option) was performed to determine the support levels of the nodes. BI analyses were conducted using MrBayes v. 3.1.2 [39]. A total of four Markov chain Monte Carlo (MCMC) chains were run simultaneously from a random starting tree for five million generations, and samples were taken per 100 generations, resulting in 50,000 trees. Moreover, the first 25% of trees sampled were discarded during burn-in. Posterior probabilities were calculated by the retained trees. The topology of the resulting files was subsequently visualized using Figtree v.1.4.2 and Adobe Illustrator CS6.

Fungal Isolation
In this study, 695 strains of ophiostomatoid fungi were isolated from 1040 tissue pieces from D. armandi galleries and 89 adult beetles at four sites (Tables 2 and 3). A total of 274 strains (39.42%) and 421 strains (60.58%) were isolated from the beetles and their galleries, respectively. Moreover, 280, 69, and 346 strains were obtained from the Hubei, Gansu, and Shaanxi provinces, respectively (Table 3). Growth rates, colony characteristics, and ITS sequence BLAST results were used for preliminary sorting and identification. These strains were distributed across five genera (Esteya, Graphilbum, Leptographium, Ophiostoma in the Ophiostomatales, and Graphium in the Microascales) and seven tentative species/groups (Taxon 1 to 7; Table 2). A total of 30 representative strains were selected for subsequent morphological and phylogenetic analyses.

DNA Sequencing and Phylogenetic Analyses
Phylogenetic analyses of the ITS, ITS2-LSU, LSU, TUB2, and TEF1-α gene regions were used to identify the genera/species and the genetic diversity within ophiostomatoid fungi [29,40,41]. Overall, one to eight strains were selected for each tentative species (Taxa 1-7) to construct the phylogenetic trees. The topologies generated by the ML, MP, and BI analyses were highly concordant, and phylograms obtained by ML were presented for all individual datasets, with branch supports obtained from MP and BI analyses.
The remaining 28 representative strains were distributed in three major clades by phylogenetic inferences based on the ITS and ITS-LSU data sets, including Ophiostoma s. str, Leptographium s. l, and Graphilbum s. str. Taxon 2 was represented by three (Table 2) of the 25 strains (Table 3). It formed a wellsupported clade in both ITS and TUB2 based phylogenies, closely related to Graphilbum kesiyea but distinct (Figures 2 and 3); hence its recognition as a distinct species.
Taxon 4 included 62 strains (Table 3), while Taxon 5 included 34 strains (Table 3), with 11 representative strains ( Table 2) belonging to the Leptographium lundbergii complex in the ITS2-LSU phylogenetic analyses (Figure 4). Additionally, eight strains (Table 2) of Taxon 4 formed a distinct and well-supported clade based on the TUB2 dataset ( Figure 5). However, ITS2-LSU and TEF1-α based phylogenetic inferences revealed that these eight strains clustered together with L. qinlingensis with high support (Figures 4 and 6). There was no TUB2 sequence for L. qinlingensis prior to this. L. qinlingensis is a nom. inval., because of the lack of type specimen.
Taxon 5 grouped with Taxon 4 in the L. lundbergii complex and three representative strains of Taxon 5 grouped with L. wushanense, as defined by Pan et al. [42], based on ITS2-LSU and TUB2 phylogenetic analyses (Figures 4 and 5).
Taxon 6 included 549 strains ( Table 3), seven of which ( Table 2) were included in the phylogenetic analyses and clustered in the Ophiostoma clavatum complex. A total of seven ( Table 2) of the 22 strains (Table 3) of Taxon 7 resided outside any recognized species complexes ( Figure 2). The analysis of ITS and TUB2 sequences (Figures 2 and 7) showed that Taxon 6 formed a well-supported clade close to, yet distinct from, the ex-type sequences of O. clavatum. Meanwhile, Taxon 7, nested in the vicinity of O. aggregatum, forming a clade related to, yet distinct in the ITS and TUB2 phylogenetic trees (Figures 2 and 8). Hence, these strains in Taxon 6 and Taxon 7 were interpreted as belonging to two distinct, undescribed Ophiostoma species.

Morphology and Taxonomy
A total of three of the seven taxa identified in the present study were interpreted as undescribed species. These included one species of Graphilbum (Taxon 2) and two species of Ophiostoma (Taxa 6 and 7). However, no reproductive structures were observed for Taxon 7 on the different media used in this study; thus, we have elected to refrain from describing this taxon at this time. L. qinlingense was recollected during this study, and the name is revalidated by designation of a Neotype.
Leptographium qinlingense (M. Tang) T. Wang & Q. Lu comb. nov. Figure 10.    Culture characteristics: Colonies grew rapidly on 2% MEA, attaining a 90 mm diameter after five days at 25 • C in the dark, accounting for a daily growth rate up to 20 mm. Colonies have a smooth margin, radial hyphae, curved shape, initially hyaline, discoloration progressing to olivaceous from the center of the colonies to the margin. The optimum growth temperature is 25 • C; however, growth can occur from 5 • C to 35 • C.
Known substrate and hosts: Galleries and adults of Dendroctonus armandi in Pinus armandii.
Known insect vectors: Dendroctonus armandi. Notes: Leptographium qinlingense was first isolated from D. armandi which infects P. armandii in China [24]. In this study, the strain CFCC53941 was isolated from exactly the same vector and host as the first reported of L. qinlingense. Thus, it was designated as the neotype herein. No differences were observed in the cultures or morphological characteristics between the recently collected neotype and that in the first reported article. Measurements of the asexual structures were consistent with previous descriptions of L. qinlingense ( Figure 10).  Leptographium qinlingense is characterized by a leptographium-like and hyalorhinocladiellalike asexual morphs. It is closely related to G. koreana, L. pinicola, and L. truncatum based on ITS2-LSU, TUB2, and TEF1-α genetic phylogeny (Figures 5 and 6). Within the L. lundbergii complex, L. qinlingense is the sole species containing two asexual morphs ( Figure 10). Grosmannia koreana is the sole species with a sexual morph [44]. Moreover, the four species differ in terms of the leptographium-like morph spore size. Specifically, the conidia of L. qinlingense (4.5-7.7 µm) are longer than that of L. pinicola, (3-5 µm) and either longer than G. koreana (3-10 µm), yet shorter than L. truncatum (7-11 µm) [44][45][46]. While L. qinlingense is native and endemic to China's mainland, the remaining three species of the L. lundbergii complex are widely distributed in the North temperate hemisphere (Canada, China, England, Korea, and the USA) and the southern hemisphere (New Zealand and South Africa). Furthermore, these four species are associated with bark beetle infestation of conifers [45][46][47].   Asexual morph: hyalorhinocladiella-like. Conidiophore arising directly from mycelium, simple branched, macronematous or semi-macronematous, mononematous, the ultimate branched bearing numerous conidiogenous cells; conidiogenous cells hyaline, smooth, thin-walled, aseptate, rounded apex, variable in length, ( Culture characteristics: Colonies grown on 2% MEA, attaining a 70 mm diameter after eight days at 25 • C in the dark, while, with appressed hyphae, colonies are white with a smooth margin, discoloration progresses to pale olivaceous from the center of the colonies to the margin. The optimal temperature is 30 • C; however, growth can begin from 5 • C to 35 • C.
Known substrate and hosts: Galleries and adults of Dendroctonus armandi in Pinus armandii.

Discussion
This study was undertaken to determine the diversity of ophiostomatoid fungi associated with D. armandi infesting P. armandii in the Qinling Mountains of western China. A total of 695 strains of ophiostomatoid fungi were identified from seven species in five genera comprising of four known species, E. vermicola, Gra. pseudormiticum, L. wushanense, and L. qinlingense, as well as a novel neotype strain, as assessed by its type, locality, and combination insect/host (CFCC53941); two novel taxa, Gra. parakesiyea, O. shennongense; and an unidentified Ophiostoma sp. 1.
Among the seven species of ophiostomatoid fungi, O. shennongense was the most frequently isolated species, accounting for an abundance of 78.99%, representing the predominant component of the community associated with D. armandi-P. armandii (Table 3), compared to the second most abundant species, L. qinlingense (8.92%). Leptographium qinlingense was the first ophiostomatoid species reported to be associated with D. armandi, and is currently isolated only from China [24]. This species has previously been shown to exhibit pathogenicity with high virulence [18,25,50]. Due to its common occurrence associated with D. armandi infesting P. armandii, L. qinlingense may have a significant role in the damage observed in D. armandi-infected P. armandii in China [22,[50][51][52].
This study is the second report showing that L. wushanense associates with P. armandii; however, P. armandii is infected by D. armandi in Shaanxi Province rather than by Tomicus armandii in Yunnan Province [42]. Before this study, only one strain of L. wushanense has been reported from T. armandii, which showed occasional association with the beetle and pine. In the present study, L. wushanense was also isolated only at Huoditang Forest Farm out of four investigated sites, showing a limited occurrence ( Table 3). The species, therefore, are sporadically located throughout southwestern China and loosely associated with the beetle.
Esteya vermicola and Gra. pseudormiticum were each represented by a single strain. Esteya is a unique genus of Ophiostomataceae, with two species: E. vermicola and E. floridanum. Both species exhibit high infectivity toward the pinewood nematode (Bursaphelenchus xylophilus) by their lunate conidia, and are potentially biocontrol agents against this epidemic pine disease [53][54][55][56][57][58]. Esteya vermicola was first isolated from Japanese black pine in Taiwan in 1999, and is associated with the pinewood nematode [53]. Since then, eight strains have been recorded worldwide [53][54][55][56][57][58]. However, although the species appear to be widely distributed, only a single strain was recorded in each of these previous studies. Graphium pseudormiticum was first reported in South Africa and was associated with Orthotomicus erosus [59], subsequently reported in Sweden as associated with Ips typographus, in Austria associated with Tomicus minor [60], and in China associated with Pissodes sp., a mite of Ips acuminatus [43,61].
Although an association between fungi and bark beetles has been observed, the classic theory of reciprocal symbiosis between bark beetles and fungi has been challenged due to the lack of in-depth explanation of the symbiosis mechanism or the existence of contradictory research cases. The present study expanded our knowledge of D. armandi and its associated fungi; however, the symbiosis mechanism between ophiostomatoid fungi and D. armandi warrant further investigations.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/ 10.3390/jof8030214/s1, Figure S1: ML tree of Esteya and related taxa (Taxon 1) generated from the combined (LSU+TUB2) sequence data. Novel sequences obtained in this study are presented in bold typeface. Bold branches indicate posterior probability values ≥ 0.9. Bootstrap values ≥ 70% for ML and MP are indicated above branches; Figure S2: ML tree of Graphium (Taxon 3) generated from the combined (ITS+tEF1-α) sequence data. Novel sequences obtained in this study are presented in bold typeface. Bold branches indicate posterior probability values ≥ 0.9. Bootstrap values ≥ 70% for ML and MP are indicated above branches.

Institutional Review Board Statement:
Not applicable for studies involving humans or animals.

Informed Consent Statement:
Not applicable for studies involving humans.
Data Availability Statement: All sequence data are available in NCBI GenBank following the accession numbers in the manuscript.