Ninety-six strains were isolated. Genomic DNA was extracted and the 28S rDNA D1/D2 region was amplified and sequenced. The obtained sequences were compared with those in the GenBank database, and the results are shown in Table 1.

A phylogenetic tree built from the 28S rDNA sequences is shown in Figure 1. From Table 1 and Figure 1, the identified fungi included: Alternaria, Botryosphaeria, Camarosporium, Cryptosporiopsis, Diaporthe, Dictyochaeta, Penicillium, Fusarium, Nectria, Peniophora, Schizophyllum, Cladosporium and Trichoderma. The most abundant genera were Diaporthe and Alternaria with 25 and 12.5% of the total number of isolates, respectively. G49 was not identified because its sequence was significantly similar to unknown fungal sequences in the GenBank database.

Diaporthe and Altenaria were found to be the predominant genera, a finding that is different from previous studies showing that Strumella, Rhizopycnis and Entrophospora were the most abundant taxa in A. elata[11]. A possible reason for this discrepancy is that the A. elata specimens were derived from different regions [17].

Diaporthe are endophytic fungi that grow in several of plant species and have been shown to produced different secondary metabolites. For example Diaporthe sp. isolated from Espeletia sp. can inhibit the growth of Phytophthora infestans, a plant pathogen [18]. While Diaporthe sp. from Curcuma longa can convert curcumin into colorless hydroderivatives. Curcumin has a potent antioxidant effect. However, the distinct yellow color limits its use. This conversion may expand its application [19]. Diaporthe sp. P133, isolated from Pandanus amaryllifolius, can secrete benzopyranones, which inhibit a virulent strain of Mycobacterium tuberculosis[20]. Diaporthe phaseolorum isolated from mangrove forest can produce the antibacterial agent 3-hydroxypropionic acid [21]. A new varied species Alternaria alternata from the bark of 200-year-old Taxus cuspidate could produce taxoids of type III with the anti-neoplastic action [22]. Alternaria sp. isolated from Brassica juncea have demonstrated potential applications in biofuel feedback [23]. To the best of our knowledge, there are no reports concerning Diaporthe and Altenaria isolated from A. elata.

The concentration of triterpenoid saponin in the representative isolate from each group (Table 1) is shown in Table 2. The highest concentration of saponins was found in G22 (2.049 mg/mL), and this concentration is significantly higher than the level observed in P11 and P18 (p < 0.05). According to the results of phylogenetic analysis, G22 was identified as a Penicillium sp., P11 was identified as a Dictyochaeta sp., and P18 was identified as Camarosporium sp. The saponin concentrations among the strains of the same genus, such as G22 (2.049 mg/mL) and P23 (0.049 mg/mL), were significantly different (p < 0.05).

The growth-promoting factors and metabolites produced by endophytic fungi have been widely investigated and applied in both medicine and agriculture. The most notable substance produced by host endophytic fungi is taxol, a mitotic inhibitor used in cancer chemotherapy, and was originally produced by the yew tree [9]. Saponins produced by A. elata have multiple therapeutic values. The culture filtrates of endophytic fungi were analyzed to identify endophytes that produce triterpenoid saponins. It is well known that Penicillium is the source of penicillin, and recent studies show that endophytic Penicillium sp. also has the capacity to secrete anti-tumor substances [24,25] and gibberellin [26,27]. In this study, G22 (Penicillium sp.) has the high capacity to produce triterpenoid saponins.

From Table 3, G22 exhibited antibacterial activity against Gram-positive bacterium Staphylococcus aureus ACCC10499. P11 inhibited the growth of S. aureus ACCC10499, Rhizoctonia solani ACCC36233 and Fusarium sporotrichioides. At the same time, it expressed strong inhibition to Klebsiella pneumoniae ACCC10498. P18 also showed inhibition to K. pneumoniae ACCC10498.

To further analyze the composition of saponins, the culture filtrates of G22, P11, and P18, underwent HPLC. As a reference, eight ginsenoside standards were also analyzed. According to the spectra, which are shown in Figure 2, G22, P11 and P18 all produced detectable concentrations of saponins. Ginsenoside Rb2 was detected in the G22, P11, and P18 culture filtrates. More peaks were observed in the G22 culture filtrate. Ginsenoside Re was also detected in the G22 culture filtrate. These results indicate that the three strains have the capacity to produce ginsenosides, especially for G22. Additionally, G22 inhibited S. aureus growth (Table 3). These results indicate that G22 has a great potential for the further detailed study.

Wild A. elata plants (5 years old) were sampled from the Xiaoxing’anling area in Northeast China. The A. elata roots were immediately placed in sterile plastic bags and stored at 4 °C. The endophytes were isolated within 48 h of collection. Before disinfection, the plant samples were thoroughly washed under running tap water for 10 h. The roots were surface-disinfected with 70% (v/v) ethanol for 0.5–1 min, 5% NaOCl for 5–10 min, 70% (v/v) ethanol for 0.5–1 min and burning for 10–30 s. The samples were subsequently rinsed with sterile water, and the outer tissue was removed with a sterile scalpel. Small pieces (0.5 × 0.5 cm) of A. elata were placed in Petri dishes containing malt extract agar (Oxoid-Unipath Ltd., Hampshire, UK), Czapeck agar (Oxoid-Unipath Ltd., Hampshire, UK), or potato dextrose agar (Oxoid-Unipath Ltd., Hampshire, UK), and were incubated at 28 °C for seven days. Following the incubation, single colonies of distinctive morphotypes were isolated on the basis of their morphological characteristics and appearance. The colonies were subsequently re-isolated by plating on PDA and incubating at 28 °C for 24–48 h to obtain pure cultures. All of the isolates were vacuum freeze-dried and deposited in the collection of the College of Life Sciences, Northeast Forestry University.

F ungal genomic DNA was extracted using the EZNA Fungal DNA Mini Kit (OMEGA, USA) according to the manufacturer’s instructions. The 50 μL PCR mixtures contained 15 ng of template DNA, 1× PCR buffer (Mg²⁺ free), 0.16 mM of each dNTP, 1.5 mM MgCl₂, 0.45 μM of each primer, and 1 U of Takara rTaq DNA polymerase (Takara, Japan). The primers employed for the amplification of the D1/D2 region of the fungal 28S rRNA gene were NL1 (5′-GCATATCAATAAGCGGAGGAAAAG-3′) and NL4 (5′-GGTCCGTGTTTCAAGACGG-3′) [28]. The thermocycling program consisted of initial DNA denaturation at 95 °C for 5 min followed by 30 cycles of denaturation at 95 °C for 1 min, annealing at 52 °C for 45 s, and elongation at 72 °C for 1 min 30 s, ending with a final elongation step at 72 °C for 6 min [29].

The PCR amplification products were separated by electrophoresis through 1% (w/v) agarose gels, stained with ethidium bromide and visually examined under UV light. The PCR products were purified using the Agarose Gel DNA Extraction Kit (Takara, Japan) and sequenced by Sangon Biotech (Shanghai, China).

The sequences generated in this study were compared with those in GenBank [30]; those sequences with ≥99% similarity to the 28S rDNA D1/D2 regions (approximately 600 bp) were considered to belong to identical genera and were included in the alignment. A neighbor-joining tree was constructed using MEGA 5.0 software [31]. The number of bootstrap replications was 1000. The sequences were deposited in GenBank under the accession numbers listed in Table 1.

Each isolate was inoculated into 100 mL of PDA liquid medium (250 mL flask) and stirred at 150 rpm at 28 °C for two weeks. After ultrasonication, the supernatant was separated from the cell debris by centrifugation at 4000× g for 20 min. A 20 mL aliquot of the supernatant was poured into a 50 mL centrifuge tube (Corning Inc., Corning, NY, USA), and 20 mL of ethyl acetate was added to the same tube. After mixing, ultrasonication and incubation for 5 min, 5 mL of the supernatant was evaporated to dryness under a vacuum at 50 °C. The residue was dissolved in 2 mL of methanol. The methanol solutions were centrifuged at 4000× g for 10 min, and the supernatants were used for subsequent analysis of the total saponin content.

The measurement of the total extracted saponins was based on a color reaction of the acid-hydrolysis products of the saponins (i.e., sapogenins) with vanillin. In total, 5 mL of the supernatant was added to a test tube and evaporated at 60 °C in a water bath. The residue was dissolved in 0.2 mL of 5% vanillin, mixed with 0.8 mL of perchloric acid, incubated in a 60 °C water bath for 15 min and quickly cooled in ice water. The concentration (mg/mL) of saponins in the reaction sample was determined using a spectrophotometer at 560 nm and comparing the readings against a calibration curve established with an oleanolic acid standard (National Institute for the Control of Pharmaceutical and Biological Products, Beijing, China) [32].

The antimicrobial activity of typical strains after 14-day cultivation against 8 microorganisms (listed in Table 3) was assessed by the agar diffusion method [33]. Three 6 mm wells were made in each disk. With the exception of Fusarium sporotrichioides (isolated in our lab), the strains were purchased from the Agricultural Culture Collection of China (ACCC). Streptomycin sulfate (5 mg/well), amoxicillin (5 mg/well) and itraconazole (4.4 mg/well) were used as positive antimicrobial controls. The activity of the extracts was estimated from the diameter (mm) of the zone of inhibition.

One hundred milliliters of ethyl acetate was added to 100 mL liquid culture. Following 30 min of agitation at 160 rpm and ultrasonication at 50 °C, the supernatant was separated from the cell debris by centrifugation at 4000× g for 30 min. After evaporation, the pellet was dissolved in 5 mL of methanol, followed by filtration through a SepPak C-18 Cartridge (Waters, Milford, MA, USA). The following water/acetonitrile gradient system was employed during HPLC analysis: 0 min, 18% acetonitrile and 82% water; 40 min, 18% acetonitrile and 82% water; 50 min, 22% acetonitrile and 78% water; 70 min, 28% acetonitrile and 72% water; 100 min, 38% acetonitrile and 62% water; and 110 min, 18% acetonitrile and 82% water.