Five species of bacteria and 24 species of fungi and Oomycetes were used in this study and their origins were listed in Table S1. Among these microbial species, Streptomyces sp. 3–10 was used for production of antifungal metabolites. It was isolated from a culture containing spores of S. platensis F-1 jointly treated by UV-irradiation and LiCl (Che, 2011). Aspergillus niger A-1 (a soil saprophyte) was used as indicator in bioassays to detect the antifungal activity of the metabolites of strain 3–10 (Shakeel et al., 2016). The remaining two species of Oomycetes, four species of bacteria, and 21 species of fungi were used to test the antimicrobial spectrum of the metabolites of Streptomyces sp. 3–10.

A total of 22 cultural media (BM, CDM, GA, GS-1, ISM, ISP-1 to ISP7, ISP-9, KB, NA, NB, OCD, PDA, PDB, SDM, TDM, and TYM) were used in this study. The full names of these media and their ingredient compositions were listed in Table S2.

Spore suspension (1 × 10⁸ spores/ml) of Streptomyces sp. 3–10 was inoculated in 250-ml Erlenmeyer flasks each containing 100 ml potato dextrose broth (PDB), 1 ml spore suspension per flask. The flasks were incubated on a rotary shaker (150 rpm) at 28°C for 3 days for production of antifungal metabolites. The culture supernatant was collected by centrifugation at 6,000 × g for 10 min, and then extracted with ethyl acetate. The ethyl acetate phase was then dried by vacuum using R-250 Rotavapor® (BUCHI Corporation, New Castle, USA) at 37°C and, a kind of brownish colloidal substance was obtained. It was treated as the crude extract of the antifungal metabolites of Streptomyces sp. 3–10.

Inhibition of fungi and Oomycetes was done using the PDA-amendment method. The crude extract (CE) dissolved in methanol at 12.5 mg/ml (w/v) was used as stock solution, which was then amended to PDA in Petri dishes (9 cm diameter, 20 ml per dish) to the final concentrations of 1 or 5 μg/ml. PDA amended with methanol alone (0.2%, v/v) was treated as control. Mycelial agar plugs (5 mm diameter) were removed from the margin area of the actively-growing colonies of the target organisms, and inoculated on the PDA amended with either the crude extract or methanol alone, one mycelial plug per dish, and three dishes (as three replicates) for each treatment. The resulting cultures were incubated at 20°C, 25°C, or 28°C for 1–7 days depending on the mycelial growth rates of the target organisms (Table 1). Diameter of the colony in each dish was measured. Inhibition of mycelial growth (IMG) was calculated using the formula: IMG (%) = (D_CK − D_CE)/D_CK × 100, where the D_CK and D_CE represent the colony diameters in the treatments of control (CK) and CE, respectively.

Inhibition of the bacteria by the CE of Streptomyces sp. 3–10 was done using the agar diffusion method (Bonev et al., 2008). The four investigated bacteria (Table S1) were separately shake-incubated at 28°C (150 rpm) in nutrient broth for 48 h. Then, aliquots (100 μl) of the liquid culture of each bacterium were evenly spread on the nutrient agar in Petri dishes (9 cm diameter, 20 ml medium per dish). Sterilized stainless-steel Oxford cups (10 × 6 × 8 mm, height × inner diameter × outer diameter) were placed in the dishes, three cups per dish and three dishes for each bacterium. The three concentrations (10, 50, and 100 μg/ml) of the CE of Streptomyces sp. 3–10 were pipetted into the three Oxford cups in each dish, respectively, 200 μl per cup. For the control treatment, three Oxford cups in a bacteria-inoculated Petri dish were loaded with 1% methanol (v/v), 200 μl per cup. The cultures were incubated at 28°C for 48 h and formation of the clear zones around the cups was used as an indicator of antibacterial activity.

Rhizopus stolonifer and Botrytis cinerea were incubated on PDA at 20°C for 3 and 7 days, respectively. Spores of R. stolonifer (sporangiospores) and B. cinerea (conidia) were harvested from the respective cultures by washing with sterile distilled water and the spore suspensions (1 × 10⁶ spores/ml) were then prepared. Aliquots (100 μl) of the spore suspensions were pipetted to and evenly spread on the D-glucose agar (GA) medium amended with the crude extract of Streptomyces sp. 3–10 at 0 (control), 1, 5, 10, 50, and 100 μg/ml, six dishes (as six replicates) for each treatment. The cultures were incubated at 20°C in the dark for 6, 9, and 12 h. At each time point, the number of germinated spores among randomly-selected at least 100 spores in each culture were counted under a compound light microscope. Additionally, length of the germ tubes of randomly-selected 20 germinated spores was measured in each replicate culture at 12 h post-incubation.

The crude extract (10 g) of Streptomyces sp. 3–10 was dissolved in 15 ml methanol and the solution was mixed with 15 g 60-mesh silica-gel granules (Qingdao Haiyang Chemical Co., Ltd., China). The mixture was then loaded as the top layer in a chromatography column (4.5 × 80 cm, inner diameter × length) containing 400 g silica gel (200-mesh, Qingdao Haiyang Chemical Co., Ltd.). The column was eluded with gradient chloroform/methanol solutions (from 99/1 to 0/100, v/v). The resulting fractions were individually assayed for antifungal activity against A. niger using the Oxford cup-agar diffusion method described by Shakeel et al. (2016). The fractions that had antifungal activity were combined, concentrated, and loaded in a Sephadex LH-20 chromatography column (GE Healthcare), which was eluded with methanol to remove impurities. The fractions that had antifungal activity were again combined, dried, and subjected to semi-preparative high performance liquid chromatography (HPLC). Finally, two pure compounds, designated as Compound Nos. 1 and 2, were obtained.

Electrospray ionization mass spectrometric (ESI-MS) analysis and nuclear magnetic resonance spectroscopy (NMR) were used to determine the chemical structure of the two pure compounds from Streptomyces sp. 3–10. The ESI-MS analysis was done on a Waters ACQUITY UPLC H-Class system coupled to the XEVO TQ-S tandem quadrupole (Waters Cooperation, Milford, MA, USA). The compounds were separately dissolved in methanol to the final concentration of 1 μg/ml. An aliquot (1 μl) of each solution was injected into the instrument. The operating parameters were set as: capillary voltage at 1.0 kV, cone voltage at 30 kV, Z-spray source temperature at 100°C, desolvation temperature (N₂) at 400°C, desolvation gas flow at 800 L/h, mass range of m/z from 50 to 1,200. The mass spectra were collected both in the positive mode and in the negative mode, and compared with those in Chapman Combined Chemical Dictionary on CD-ROM version 6.1 (Chapman and Hall, 2003) for determination of the chemical identity of the compounds. Moreover, synthetic reveromycin A (purity > 98%) purchased from Abcam Trading (Shanghai) Co. Ltd. (Shanghai, China) was used as a reference chemical in the UV-Vis spectrum analysis on Waters ACQUITY UPLC H-Class system.

In the NMR analysis, the two pure compounds were dissolved in methanol-d₄ (CD₃OD, Sigma-Aldrich®), and were then determined in a 400 MR DD2 spectrometer (Agilent Technologies, USA) for spectra of ¹H-NMR and ¹³C-NMR. Tetramethylsilane (TMS) was used as the internal standard in the NMR analysis.

Synthetic reveromycin A (purity > 98%) purchased from Abcam Trading (Shanghai) Co. Ltd. was used as standard in quantitative determination of the content of reveromycin A in the CE from strain 3–10. Five solutions of the synthetic reveromycin A with the concentrations of 0.08, 0.16, 0.31, 0.63 and 1.25 μg/ml in methanol were prepared. An aliquot (1 μl) of each solution of the synthetic reveromycin A or the solution of the CE of strain 3–10 at 3.125 μg/ml was injected into Waters ACQUITY UPLC H-Class system. The peak area for reveromycin A in each solution sample was measured. A standard curve was plotted based on the peak areas for the solutions of the synthetic reveromycin A and the concentrations of those solutions. The content of reveromycin A in the CE of strain 3–10 was then calculated based on that standard curve. The quantitative determination was repeated for three times.

The purified compounds were separately dissolved in methanol to 10 mg/ml (w/v) as stock solutions, which were used in the following two trials, a mycelial growth trial and a spore germination trial. In the mycelial growth trial, the stock solution of each compound was amended in PDA, which was adjusted to pH 4.5, 5.5, or 7.0 with 1 mol/L HCl or 1 mol/L NaOH. The final concentrations in PDA ranged from 0.3125 to 50.0 μg/ml for Compound No. 1, and from 0.3125 to 100 μg/ml for Compound No. 2 (Table S3). Meanwhile, methanol solution (0.2%, v/v) was added to PDA at pH 4.5, 5.5, or 7.0 as controls. Mycelial agar plugs of B. cinerea, M. hiemails, R. stolonifer, and S. sclerotiorum (5 mm diameter) were individually inoculated on PDA, three dishes for each treatment (fungus × compound × concentration × pH). The cultures were incubated at 20°C in the dark for 1 day for R. stolonifer, and for 3 days for the other three fungi. The colony diameter was measured and the percentage value of inhibition of mycelial growth (IMG) under each pH was calculated using the formula mentioned above. EC₅₀-values (effective concentrations that gave 50% inhibition) for each compound under each pH were estimated based on the IMG data and the concentrations of that compound applied in PDA (Van Ewijk and Hoekstra, 1993).

In the spore germination trial, the stock solution of each compound was amended in GA with Ph-values of 4.5, 5.5, or 7.0. The final concentrations in GA ranged from 0.0625 to 50.0 μg/ml for Compound No. 1, and from 0.3125 to 100 μg/ml for Compound No. 2 (Table S3). Meanwhile, methanol solution (0.2%, v/v) was added to GA with pH 4.5, 5.5, and 7.0 as controls. Aliquots (100 μl) of the spore suspension (1 × 10⁶ spores/ml) of B. cinerea, M. hiemails, or R. stolonifer were pipetted on GA alone or on GA amended with a purified compound, three dishes for each treatment (fungus × compound × concentration × pH). The cultures were incubated at 20°C for 12 h. Spore germination in each culture was observed under light microscope. Percent inhibition of spore germination (ISG) by each compound under each pH was calculated using the formula: ISG (%) = (P_CK–P_C)/P_CK × 100, where the P_CK and P_C represent the percentages of germinated spores in the treatments of control (CK) and a pure compound, respectively. Finally, EC₅₀-values for each compound under each pH were estimated based on the ISG data and the concentrations of that compound applied in GA (Van Ewijk and Hoekstra, 1993).

Mature strawberries (Fragaria × ananassa cultivar “Jing Yu”) of similar size (3.0–3.5 × 2.0–2.5 cm, length × width, 16–18 g per berry) were collected from strawberry plants grown in a plastic tunnel. They were surface sterilized in 70% ethanol (v/v) for 1 min, followed by washing in sterile distilled water and blotting on sterilized paper towels to remove the excess water on the fruit surface. Finally, they were placed in glass Petri dishes (9 cm diameter), six berries per dish. For each pathogen, there were 10 treatments, including one negative control treatment with 1% methanol (v/v), three CE treatments with the concentrations of 10, 50, and 100 μg/ml (pHs at 5.4, 4.3, and 4.2, respectively), three treatments of reveromycin A (from strain 3–10) with the concentrations at 10, 50, and 100 μg/ml (pH at 4.5), and three fungicide treatments (positive controls) with the concentrations of an appropriate fungicide (either captan, pyrimethanil, or dimetachlone) at 10, 50, and 100 μg a.i./ml. Captan (3a, 4, 7, 7a-tetrahydro-2-[(trichloromethyl) thio]-1 H-isoindole-1,3(2H)-dione) was used as the positive control for the pathogens M. hiemails and R. stolonifer, whereas pyrimethanil (2-anilino-4,6-dimethylpyrimidine) and dimetachlone [N-(3,5-dichlorophenyl) succinimide] were used as positive controls for the pathogens B. cinerea and S. sclerotiorum, respectively. Both captan (50% wettable powder) and pyrimethanil (80% water dispersible granule) were purchased from Hebei Guan Long Agrichemical Co. Ltd. (Hengshui, China). Dimetachlone (40% wettable powder) was purchased from Zhejiang SPACE Agrichemical Co. Ltd. (Wenzhou, China).

The six strawberries in a Petri dish were immerged for 1 min either in the 1% methanol solution, or in a solution containing either the CE, reveromycin A or a fungicide. Then, they were maintained in a laminar flow hood for 15–30 min, re-placed in the dish, and inoculated with one of the four pathogens. B. cinerera, M. hiemails, and R. stolonifer were spray-inoculated with the spore suspensions (1 × 10⁶ spores/ml) amended with 0.5% D-glucose (w/v), approximately 10 ml spore suspension on the six strawberries in a Petri dish. S. sclerotiorum was inoculated with mycelial agar plugs (5 mm diameter), one mycelial agar plug on each berry. The dishes with the treated strawberries were placed in plastic boxes (80 × 60 × 50 cm, length × width × height), which were individually covered with a transparent plastic film (0.1 mm thick, Gold Mine Plastic Industry Ltd., Jiangmen, China) to maintain the high humidity condition. The boxes were placed in a growth room at 20°C under the light regime of 12-h light/12-h dark. After incubation for 3 and 5 days for the inoculation treatments with R. stolonifer and M. hiemails, respectively, and for 7 days for the inoculation treatments with B. cinerea, and S. sclerotiorum, disease severity on the strawberries was individually rated using a numeric scale from 0 (completely healthy) to 8 (completely rotten) according to the description by Huang et al. (2011). Disease severity index (DSI) was then calculated using the following formula: DSI =100 ×∑i = 0n(Sn× n)/8 × ∑i = 0n(Sn) 

Where n represents the rating scale (0–8) and S_n represents the number of strawberries corresponding to the rating scale n. This experiment was repeated two more times.

The strawberry seedlings (Fragaria × ananassa cultivar “Jing Yu”) were trans-planted in a plastic tunnel (Length × Width × Height, 29 × 6.5 × 2 m) in September 28 of 2016. The plants were carefully managed (watering, weeding) as required. The toxicity experiment was performed from December 11 of 2016 to December 25 of 2016, when most strawberry plants became bloomed. There were four treatments in this experiment: (i) control (CK); (ii), (iii), and (iv) the crude extract of strain 3–10 (CE) at 10, 50, and 100 mg/ml, respectively. Twenty strawberry plants for each treatment were randomly selected in the field and the leaves on each selected plants were individually immerged for 1 min either in 1% methanol (v/v) (CK) or in each aqueous solution of CE. At 0, 1, 3, 7, and 14 days post treatment (dpt), the treated plants were observed for the toxicosis symptoms on the leaves (yellowing, necrosis and malformation on leaves and flowers). Three plants for each treatment were randomly selected and the three leaves on each treated plant were detached as a leaf sample, one leaf from each plant. The leaves were immediately taken to laboratory for determination of the chlorophyll content (a, b and total). For chlorophyll extraction, the three-leaf sample for each treatment was homogenized in 95% ethanol. The homogenate was centrifuged at 10,000 × g and the supernatant was transferred out for determination of the absorbance values in DU730 Beckman spectrophotometer at 663 and 645 nm, respectively (Rout et al., 1998). The resulting absorbance values A₆₆₃ and A₆₄₅ were used to calculate the content of chlorophyll a (Chl a), chlorophyll b (Chl b) and the total chlorophyll (Chl T) using the following formula:

where V represents volume of 95% ethanol used for chlorophyll extraction in homogenization; W represents weight of the leaf sample; F. W. represent fresh weight of the leaf sample.

Additionally, at 0 or 14 dpt, five other plants for each treatment were randomly selected, uprooted and again taken to laboratory. They were washed under the running tap water and dried in an oven at 50°C. The upper part of that plant (the roots were trimmed) was then weighed.

Streptomyces sp. 3–10 was streak-inoculated on various agar media (Tables S2, S5). The cultures were incubated at 28°C in the dark for 14 days for observation of the colony morphology (shape, size, color of the substrate mycelium and the aerial mycelium, soluble pigments in media). For morphological observation of the substrate mycelium, a sterilized glass slide (7.5 × 2.5 × 0.1 cm, length × width × thickness) was placed on a PDA culture of strain 3–10 at 28°C for 14 days. Then, the slide was removed from the culture, and the hyphae on the slide were stained with 1% methyl green and observed under a compound light microscope (Ruan and Huang, 2011). For observation of the spore morphology, a sterilized cellophane film was placed on PDA. Strain 3–10 was inoculated on the cellophane film and the culture was incubated at 28°C for 14 days. Then, the cellophane film was removed, and cut to small pieces (approximately 3 × 3 mm, length × width). The resulting film pieces with the colonies of strain 3–10 were immediately fixed in the glutaraldehyde fixative, followed by dehydration with ethanol, drying in a Critical Point Dryer (Model: 13200E-AB, SPI SUPPLIES, West Chester, PA, USA), and gold-coating in a sputter coater (Model: JFC-1600, NTC, Tokyo, Japan) using the conventional procedures. Finally, the specimens on the film pieces were observed under a scanning electron microscope (Model: JSM-6390/LV, NTC, Tokyo, Japan). To characterize physiological features, strain 3–10 was inoculated on specific media (Tables S2, S6, S7) using related procedures described by Ruan and Huang (2011).

Streptomyces sp. 3–10 was shake-incubated at 28°C for 2 days in the liquid ISP-2 medium (Table S2). The mycelia in the cultures were harvested by centrifugation at 7,000 × g for 5 min. Genomic DNA (gDNA) was extracted from the mycelium using the reagents in the Wizard® Genomic DNA Purification Kit (Promega, Madison, WI, USA). It was used as template in PCR for amplification of the 16S rDNA sequence using the universal primers 27f (5′-AGAGTTTGATCCTGGCTCAG-3′) and 1492r (5′-TACGGCTACCTTGACGACTT-3′; Zhang et al., 2016). The 25-μl PCR reaction system contained 1 μl of gDNA (approximately 50 ng), 2.5 U Taq DNA polymerase (TaKaRa Biotechnol. Co., Ltd., Dalian, China), 2.5 μl 10 × PCR buffer, 1 μl of each primer (20 μmol) and 0.5 μl of dNTPs mixture (10 mmol/L). The PCR was performed in a S1000^TM thermal cycler (Bio-Rad, USA) with the following thermal program: initial denaturation at 95°C for 5 min, followed by 35 cycles (95°C for 30 s, 56°C for 30 s, 72°C for 1.5 min), and final extension at 72°C for 5 min. The resulting PCR product was purified from the agarose gel after electrophoresis and cloned into the pMD18-T vector (TaKaRa Biotechnol Co. Ltd., Dalian, China), which was subsequently transformed into E. coli DH5α. A positive E. coli clone with the correct DNA insert size was sequenced at Sangon Biotechnol. Co. Ltd. (Shanghai, China). The sequence was submitted to GenBank at NCBI and was assigned with the accession number KX811537.

For phylogenetic analysis, a dataset was established based on the 16S rDNA sequences of Streptomyces sp. 3–10, 39 other taxa of Streptomyces species, and strain DSM44928 of Catenulispora acidiphila (Table S4). They were aligned using the Clustal W program in the MEGA 7.0 software and phylogenetic analysis was done based on the alignment using the maximum likelihood (ML) methods. All the nucleotides in the DNA sequences were treated as un-ordered and un-weighted, and the gaps were treated as the missing data. The bootstrap consensus trees were inferred from 1,000 replicates.

Data on spore germination rate, length of germ tubes, disease incidence and disease severity index in related experiments were separately analyzed for ANOVA (analysis of variance) using the PROC ANOVA procedure (SAS Institute, Cary, NC, USA, version 8.0, 1999). Treatment means in each experiment were separated using Least significance Different (LSD) test at α = 0.05. Before ANOVA, the percentage data on spore germination rate and disease incidence were transformed to numerical data by multiplication with 100. After analysis, the data were back-transformed to the percentage values. Data on content of chlorophyll a, chlorophyll b, the total chlorophyll and dry weight of strawberry plant between the treatments of control and the crude extract of Streptomyces sp. 3–10 was compared at α = 0.05 using the PROC TTEST in the SAS software.

The crude extract from Streptomyces sp. 3–10 showed a wide antifungal spectrum (Table 1). It inhibited mycelial growth of two species of Pythium and 20 species of fungi. The percentages of inhibition of mycelial growth varied greatly among the target species, ranging from 16.6 to 97.2% at 1 μg/ml, and from 45.8 to 100% at 5 μg/ml. Among the 20 fungal species, M. hiemails, B. cinerea, R. stolonifer, and S. sclerotiorum are the causal agents of strawberry fruit rot. They were inhibited by 80.8, 86.0, 91.4, and 97.2%, respectively, at 1 μg/ml, and by the rates higher than 97% at 5 μg/ml (Table 1). In contrast, the crude extract weakly inhibited (<10%) mycelial growth of Fusarium graminearum (the causal agent of wheat head blight) and Pyricularia oryzae (the causal agent of rice blast) at two concentrations (Table 1).

Results from the antibacterial experiment showed that the crude extract of Streptomyces sp. 3–10 had no antibacterial activity (Figure S1). It failed even at 100 μg/ml to inhibit growth and proliferation of all the four investigated bacteria, including Acidovorax citrulli (the causal agent of watermelon bacterial fruit blotch), Curtobacterium flaccumfaciens pv. flaccumfaciens (the causal agent of bacterial wilt of beans), and Erwinia carotovora (the causal agent of potato soft rot), and Bacillus subtilis (a saprophyte).

The crude extract effectively inhibited spore germination of B. cinerea and R. stolonifer in GA at 20°C (Table 2). For B. cinerea, the inhibitory efficacy depended greatly on concentration of the crude extract. In the treatments of the crude extract at 1 and 5 μg/ml, the conidia germinated by 93–99% at 6–12 h post-incubation (hpi), not significantly different (P > 0.05) from the germination rate of 99% in the control treatment. The average values of germ-tube length reached 146.6 and 106.5 μm for the two concentrations of the crude extract, respectively. In the treatments of the crude extract at 10, 50, and 100 μg/ml, the percentages of germinated conidia at 12 hpi were reduced by 12, 90, and 99%, respectively, and the average germ tube length was reduced by 51, 89, and 92%, respectively, compared to the control treatment (Table 2). For R. stolonifer, while the control treatment had percentages of germinated sporangiospores of 0, 8.3 and 63.1% at 6, 9, and 12 hpi, respectively, and the average germ-tube length reached 67.4 μm at 12 hpi, the treatments of the crude extract at 1, 5, 10, 50, and 100 μg/ml completely inhibited sporangiospore germination at 6, 9, and 12 hpi, just as the fungicide treatment with captan at 100 μg a.i./ml (Table 2).

Two compounds (Nos. 1 and 2) were purified from the crude extract. Compound No. 1 is a white amorphous powder. ESI±MS (100 kV) of this compound showed a molecular ion peak at m/z 683 (M+Na)⁺, 659 (M−H)⁻ and HR-ESI (positive) MS at m/z 683.3394 (M+Na)⁺ (calc., 683.3047). Based on these data, the molecular formula of this compound was inferred to be C₃₆H₅₂O₁₁. The UV-Vis spectrum in methanol gave the UV maximum absorbance at 237.9 nm (Figure S2). The ¹H-NMR spectrum (CD₃OD, 400 MHz; Figure S3) showed five methyl protons at δ_H 2.26 (3H, s), 1.75 (3H, s), 1.08 (3H, d, J = 6.8 Hz), 0.85 (3H, t, J = 6.9 Hz), and 0.79 (3H, d, J = 5.6 Hz), eight olefinic protons at δ_H 6.97 (1H, dd, J = 15.7, 7.7 Hz), 6.44 (2H, m), 6.25 (1H, d, J = 15.6 Hz), 5.88 (1H, s), 5.81 (1H, d, J = 15.7 Hz), 5.59 (1H, t, J = 7.0 Hz), and 5.53 (1H, dd, J = 15.6, 7.3 Hz), and three oxymethine protons at δ_H 4.62 (1H, d, J = 7.3 Hz), 4.07 (1H, m), and 3.45 (1H, m; Table 3). The ¹³C NMR spectrum (CD₃OD, 100 MHz; Figure S4) showed 36 signals, including five methyl groups at δ_C 14.6, 15.3, 13.0, 18.0, and 14.2, 10 methylene carbons at δ_C 32.7, 28.7, 36.9, 35.2, 25.4, 34.8, 23.8, 23.4, 31.2, and 29.9, eight olefinic carbons at δ_C 122.7, 152.6, 127.9, 137.8, 129.5, 134.1, 139.3, and 121.9, two methine carbons at δ_C 44.1 and 36.3, two olefinic quaternary carbons at δ_C 135.5 and 152.2, four oxymethine carbons at δ_C 76.9, 76.3, 84.2, and 79.7, three carboxyl carbonyl carbons at δ_C 170.3, 170.4, and 176.2, one ester carbonyl carbon at δ_C 173.4, and one quaternary spiroketal carbon at δ_C 97.0 (Table 3). Compound No. 1 was identified as reveromycin A (Figure 1) by comparing the data on the spectra of ¹H NMR, ¹³C NMR, MS and UV-Vis of this compound with the related spectra of reveromycin A reported by Fremlin et al. (2011). The average content of reveromycin A in the crude extract of strain 3–10 was 37.7% (Figure S5).

Compound No. 2 was also a white amorphous powder. ESI±MS (100 kV) of this compound had a molecular ion peak at m/z 683 (M+Na)⁺, 659 (M−H)⁻, HR-ESI (positive) MS at m/z 683.3041 (M+Na)⁺ (calc., 683.3047). Based on these data, the molecular formula of this compound was inferred to be C₃₆H₅₂O₁₁. The UV spectrum in methanol gave a UV maximum absorbance at 238.9 nm (Figure S6). The ¹H-NMR spectrum (CD₃OD, 400 MHz; Figure S7) showed five methyl protons at δ_H 2.24 (3H, s), 1.74 (3H, s), 1.01 (3H, d, J = 6.9 Hz), 0.92 (3H, t, J = 6.9 Hz) and 0.89 (3H, d, J = 6.5 Hz), eight olefinic protons at δ_H 6.99 (1H, dd, J = 15.8, 7.5 Hz), 6.39 (1H, d, J = 15.7 Hz), 6.28 (1H, d, J = 16.1 Hz), 6.24 (1H, dd, J = 16.0, 3.8 Hz), 5.79 (1H, d, J = 15.6 Hz), 5.78, (1H, s), 5.77 (1H, m), 5.47 (1H, dd, J = 15.7, 7.6 Hz), and three oxymethine protons at δ_H 5.57 (1H, d, J = 3.7 Hz), 4.08 (1H, dd, J = 7.4, 5.4 Hz), 3.45, (1H, m; Table 3). The ¹³C NMR spectrum (CD₃OD, 100 MHz; Figure S8) also showed 36 signals, including five methyl groups at δ_C 14.5, 15.1, 12.7, 18.2, and 14.0, 10 methylene carbons at δ_C 32.6, 30.3, 35.3, 39.8, 32.9, 35.6, 26.6, 24.4, 30.4, and 29.8, eight olefinic carbons at δ_C 121.2, 152.9, 127.2, 138.6, 130.9, 132.6, 136.1, and 122.5, two methine carbons at δ_C 44.0 and 35.7, two olefinic quaternary carbons at δ_C 135.2 and 152.5, four oxymethine carbons at δ_C 77.1, 78.5, 88.8, and 80.4, three carboxyl carbonyl carbons at δ_C 170.4, 170.2, and 175.9, one ester carbonyl carbon at δ_C 173.1 and one quaternary spiroketal carbon at δ_C 108.6 (Table 3). Compound No. 2 was identified as reveromycin B (Figure 1) by comparing the data on the spectra of ¹H NMR, ¹³C NMR, MS, and UV-Vis of this compound with related spectra of reveromycin B reported by Fremlin et al. (2011).

Reveromycins A and B from Streptomyces sp. 3–10 effectively suppressed mycelial growth of B. cinerea, M. hiemails, R. stolonifer, and S. sclerotiorum, and spore germination of B. cinerea, M. hiemails, and R. stolonifer (Table 4). The suppressive efficacy was greatly affected by ambient pH. For reveromycin A, the EC₅₀-values ranged from 0.10 to 0.88 μg/ml at pH 4.5, from 0.42 to 1.76 μg/ml at pH 5.5, and from 14.04 to 53.35 μg/ml at pH 7.0. For reveromycin B, the EC₅₀-values ranged from 1.15 to 5.49 μg/ml at pH 4.5, from 6.12 to 35.46 μg/ml at pH 5.5, and >100 μg/ml at pH 7.0.

At 3–7 days post inoculation (20°C), the strawberries in the control treatment were severely diseased (Figure 2, Table 5). The percentages of diseased strawberries reached 100, 88.9, 100, and 94.4% for the control inoculations with B. cinerea, M. hiemails, R. stolonifer, and S. sclerotiorum, respectively. The disease severity index values reached 78.5, 62.5, 88.1, and 82.6 in these treatments, respectively. In contrast, most strawberries in the treatments of the crude extract and reveromycin A from strain 3–10 at 10, 50, and 100 μg/ml, as well as in the treatments of the fungicides at 10, 50, and 100 μg a.i./ml appeared healthy (Figure 2, Table 5). The percentages of diseased strawberries were lower than 40% and the disease severity index values were lower than 35 in these treatments. Reveromycin A at 50 and 100 μg/ml completely suppressed strawberry fruit rot caused by all the four fungi. Statistical analysis indicated that for each pathogen, the treatments of the CE, reveromycin A and fungicide differed significantly (P < 0.05) from the control treatment both in disease incidence and in disease severity index. Under the same concentration, the crude extract and reveromycin A from strain 3–10 did not significantly (P > 0.05) differ from the corresponding fungicide.

The treatments of the strawberry leaves with the crude extract of Streptomyces sp. 3–10 at 10, 50 and 100 μg/ml did not produce any visible toxic symptoms (yellowing, necrosis and malformation) on leaves and flowers (Figure S9). They grew normally and did not significantly differ (P > 0.05) from the control treatment (water) both in leaf chlorophyll content and in dry weight of the upper part of strawberry plants.

Streptomyces sp. 3–10 grew and sporulated on nine agar media (BM, GS-1, ISP-1, ISP-2, ISP-3, ISP-5, ISP-6, ISP-7, PDA) after incubation at 28°C for 14 days with formation of whitish to grayish colonies. They hardly grew on the ISP-4 medium (Figure S10, Table S5). In the PDA cultures, the single colonies of Streptomyces sp. 3–10 were averagely sized by 1.6 ± 0.3 mm in diameter, dome-shaped in the colony center, and the saw tooth-shaped at the colony margin (Figure S11). The substrate mycelium was pale yellowish to orange in color and the aerial mycelium was whitish to grayish in color (Figure S11). The spore chains were rectiflexible in shape with the spores being short rod in shape, 0.9 × 0.6 μm (length × width) in size, and smooth surfaced (Figure S11). The cultural and morphological characteristics of Streptomyces sp. 3–10 matched the description for Streptomyces yanglinensis 1307^T (Xu et al., 2006).

Results of the physiological determination showed that strain 3–10 was similar to S. yanglinensis 1307^T in most of the measured features, but differed from S. yanglinensis 1307^T in utilization of myo-inositol, growth response to pH 3.5, degradation of Tween 80 and sensitivity to streptomycin sulfate and sulfamethoxazole. Strain 3–10 could utilize myo-inositol, whereas strain 1307^T could not. Strain 3–10 could grow on ISP-3 at pH 3.5, whereas strain 1307^T could not. Strain 3–10 could not degrade Tween 80, whereas strain 1307^T could. Strain 3–10 showed sensitive to streptomycin sulfate (10 μg/ml), but resistant to sulfamethoxazole (25 μg/ml). In contrast, strain 1307^T showed resistant to streptomycin sulfate (10 μg/ml), but sensitive to sulfamethoxazole (25 μg/ml; Tables S6, S7).

The close relationship between Streptomyces sp. 3–10 and S. yanglinensis 1307^T was further confirmed by phylogenetic analysis of the 16S rDNA sequences. Streptomyces sp. 3–10 is distantly related to strains F-1, JCM4662^T and NRBC13818^T of S. platensis, but is closely related to S. yanglinensis 1307^T (Figure 3). Therefore, Streptomyces sp. 3–10 might represent a novel phylotype of S. yanglinensis.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.