A new p-terphenyl derivative from the insect-derived fungus Aspergillus candidus Bdf-2 and the synergistic effects of terphenyllin

General experimental procedures

The UV spectra were scanned on a UV-2600 instrument (Shimadzu, Kyoto, Japan). IR spectra were recorded on a Shimadzu Affinity-1 instrument (Shimadzu, Kyoto, Japan). HRESIMS spectra were obtained on a Q-TOF mass spectrometer from Bruker maXis, with an electrospray ionization (ESI) interface (Bruker, Fremont, CA, USA). Standard 1D and 2D NMR spectra were recorded on a Bruker Avance-600 NMR spectrometer (¹H at 600 MHz and ¹³C at 151 MHz) (Bruker, Fremont, CA, USA). Silica gel (200–300 mesh, Qingdao Marine Chemical Inc., Qingdao, China), and Sephadex LH-20 (GE Healthcare, Uppsala, Sweden) were used for column chromatography. Semi-preparative HPLC separations were carried out on a semi-preparative HPLC instrument equipped with a UC-3281 pump and a UC-3292S UV detector, using an XB-C₁₈ column (250 mm × 10 mm, 10 µm, Welch, Shanghai, China). Precoated silica gel GF-254 plates (Qingdao Marine Chemical Inc., Qingdao, China) were used for analytical TLC and Spots were visualized under UV light (254 or 365 nm).

Fungal material

The fungus Aspergillus candidus Bdf-2 (Fig. S1) was isolated from the abdomen of Blaptica dubia which were purchased from Lianwanjia Dubia exclusive store in Qingdao on November 30, 2016. The isolate was identified as Aspergillus candidus by analysis of its morphological characteristics and the rDNA gene internal transcribed spacer (ITS) sequence (GenBank accession number MH681592) (Shan et al., 2019), as well as beta-tubulin (tub2) gene sequence (GenBank accession number MN533958), RNA polymerase II second largest subunit (RPB2) gene sequence (GenBank accession number MN533959) and large subunit ribosomal RNA gene sequence (GenBank accession number MN533960) (Figs. S2–S5). The fungus was stored on potato dextrose agar (PDA) slants at 4 °C in the College of Forestry and Landscape Architecture, South China Agricultural University, Guangzhou, China. The culture was also stored at China Center for Type Culture Collection (CCTCC) and the strain preservation number was CCTCC AF 2019013.

Fermentation, extraction, and isolation

The fungus Aspergillus candidus Bdf-2 was cultured on potato dextrose agar (PDA) (potato 200 g/L, dextrose 20 g/L, and agar 20 g/L) medium in Petri dishes at 28 °C for 5 days. Then, three agar plugs (0.5 × 0.5 cm) were inoculated in a 500 mL Erlenmeyer flask containing 200 mL of potato dextrose broth (PDB) (potato 200 g/L and dextrose 20 g/L) medium and incubated on a rotary shaker at 150 rpm and 28 °C for 5 days. The obtained liquid seeds were added into the sterilized rice culture medium (6.0 kg) in the incubators at 28 °C for 60 days before harvest. The leavening was extracted with ethyl acetate (EtOAc). Then, the EtOAc extracts were dissolved in 1 L methanol (MeOH) and extracted with petroleum ether at room temperature. The petroleum ether and MeOH extracts were dried using vacuum concentration to give crude extracts of 20.7 g and 113.5 g, respectively.

The petroleum ether crude extract (20.7 g) was subjected to column chromatography over silica gel (200–300 mesh) eluted with a gradient of petroleum ether-acetone (100:0–0:100) to give three fractions (Fr. A-C). Fraction A (5.013 g) was chromatographed over Sephadex LH-20 (eluted with CHCl₃-MeOH, 1:1) to give 24 further subfractions, A₁ to A₂₄, and subfractions A₈–A₁₂ were combined and further purified by recrystallization to afford 4, 4″-deoxyterphenyllin (3) (19.9 mg). For the conditions of recrystallizations, the impure compound was first eluted with insoluble solvents such as petroleum ether or ethyl acetate at room temperature, followed by dissolution with soluble acetone or methanol, and repeated the above steps after crystallization until the compound was purified. Fraction B (4.072 g) was also chromatographed over Sephadex LH-20 (eluted with CHCl₃-MeOH, 1:1) to give 77 further subfractions, B₁ to B₇₇. Subfractions B₄₂–B₅₆ were combined and further purified by semi-preparative HPLC (eluted with MeOH–H₂O over 35 mins, 70:30, with an isocratic 0.01% TFA modifier, 4 mL/min, 210 nm) to afford dechlorochlorflavonin (8) (9.3 mg) and 4″-deoxy-2′-methoxyterphenyllin (1) (8.6 mg).

The MeOH extracts (113.5 g) were first subjected to column chromatography over silica gel (200–300 mesh) eluted with a gradient of petroleum ether-acetone (100:0–0:100) to give four fractions (Fr. D–G). Fraction D (62.4 g) was purified by recrystallization to afford terphenyllin (2) (43.5 g) and the residue was further chromatographed over Sephadex LH-20 (eluted with CHCl₃-MeOH, 1:1) to give 18 subfractions. Subfractions D₆-D₁₂ were combined and further purified by recrystallization to afford candidusin A (7) (61.5 mg). Fraction E (435.6 mg) was first recrystallized and then further purified by semi-preparative HPLC (eluted with MeOH–H₂O over 35 mins, 70:30, with an isocratic 0.01% TFA modifier, 4 mL/min, 210 nm) to afford 4″-deoxyterphenyllin (4) (24.7 mg). Fraction G (2.1 g) was chromatographed over Sephadex LH-20 (eluted with CHCl₃-MeOH, 1:1) to give 24 further subfractions. Subfractions G₆-G₉ were combined and further recrystallized to afford fellutanine A (9) (14.5 mg). Subfractions G₁₁-G₁₅ were combined and further recrystallized to afford 3″-hydroxyterphenyllin (5) (65.4 mg). Subfractions G₁₈-G₂₀ was purified by semi-preparative HPLC (eluted with MeOH–H₂O over 35 mins, 53:47, with constant 0.01% TFA modifier, 4 mL/min, 210 nm) to obtain 3, 3″-dihydroxyterphenyllin (6) (54.8 mg).

Compound identification

4″-Deoxy-2′-methoxyterphenyllin (1): light yellow powder; UV (MeOH) λ_max 213, 266 nm; IR ν_max 3354, 2959, 2932, 1591, 1520, 1474, 1458, 1435, 1389, 1341, 1265, 1227, 1103, 1080, 1015, 997, 837, 802, 773, 702, 652 cm⁻¹; HR-ESI-MS m/z 337.1439 [M + H]⁺ (calcd for C₂₁H₂₁O₄, 337.1434), 359.1261 [M + Na]⁺ (calcd for C₂₁H₂₀O₄Na, 359.1254), ¹³C NMR and ¹H NMR see Tables 1 and 2, and the key 2D NMR data see Fig. 2.

Terphenyllin (2): white crystal; HR-ESI-MS m/z 339.1233 [M + H]⁺ (calcd for C₂₀H₁₉O₅, 339.1227), 361.1048 [M + Na]⁺ (calcd for C₂₀H₁₈O₅Na, 361.1046). ¹³C NMR and ¹H NMR see Tables 1 and 2 and the data were consistent with the literature (Kurobane et al., 1979).

4, 4″-Deoxyterphenyllin (3): white powder; HR-ESI-MS m/z 307.1335 [M + H]⁺ (calcd for C₂₀H₁₉O₃, 307.1329), 329.1152 [M + Na]⁺ (calcd for C₂₃H₂₆O₇Na, 329.1148), ¹³C NMR and ¹H NMR see Tables 1 and 2 and the structure was confirmed by comparison with literature data (Biggins et al., 2011).

4″-Deoxyterphenyllin (4): white crystal; HR-ESI-MS m/z 323.1274 [M + H]⁺ (calcd for C₂₀H₁₉O₄, 323.1278), 345.1096 [M + Na]⁺ (calcd for C₂₀H₁₈O₄Na, 345.1097), ¹³C NMR and ¹H NMR see Tables 1 and 2 and the data were consistent with the literature (Takahashi et al., 1976).

3″-Hydroxyterphenyllin (5): white powder; HR-ESI-MS m/z 355.1180 [M + H]⁺ (calcd for C₂₀H₁₉O₆, 355.1176), 377.0999 [M + Na]⁺ (calcd for C₂₀H₁₈O₆Na, 377.0996), ¹³C NMR and ¹H NMR see Tables 1 and 2 and the data were consistent with the literature (Ivanets et al., 2018).

3, 3″-Dihydroxyterphenyllin (6): light yellow solid; HR-ESI-MS m/z 371.1127 [M + H]⁺ (calcd for C₂₀H₁₉O₇, 371.1125), 393.0945 [M + Na]⁺ (calcd for C₂₀H₁₈O₇Na, 393.0945), ¹³C NMR and ¹H NMR see Tables 1 and 2 and the structure was confirmed by comparison with literature data (Kobayashi et al., 1985).

Candidusin A (7): gray amorphous solid; HR-ESI-MS m/z 353.1024 [M + H]⁺ (calcd for C₂₀H₁₇O₆, 353.1020), 375.0844 [M + Na]⁺ (calcd for C₂₀H₁₆O₆Na, 375.0839), ¹³C NMR and ¹H NMR see Tables 1 and 2 and the data were consistent with the literature (Liu et al., 2012).

Dechlorochlorflavonin (8): yellow powder; HR-ESI-MS m/z 353.1024 [M + H]⁺ (calcd for C₂₀H₁₇O₆, 353.1020), 375.0844 [M + Na]⁺ (calcd for C₂₀H₁₆O₆Na, 375.0839). ¹H-NMR (600 MHz, CDCl₃) δ 7.93 (1H, m, H-6′), 7.80 (1H, m, H-4′), 7.14 (1H, m, H-3′), 7.14 (1H, m, H-5′), 6.46 (1H, s, H-6), 3.96 (3H, s, 7-OMe), 3.89 (3H, s, 3-OMe), 3.88 (3H, s, 8-OMe); ¹³C-NMR (151 MHz, CDCl₃) δ 178.02 (C-4), 159.12 (C-7), 157.66 (C-5), 156.09 (C-2), 155.72 (C-2′), 149.25 (C-8a), 137.51 (C-3), 133.69 (C-4′), 129.86 (C-5′), 129.31 (C-8), 121.22 (C-6′), 119.99 (C-3′), 118.36 (C-1′), 105.42 (C-4a), 96.06 (C-6), 62.44 (3-OMe), 61.91 (8-OMe), 56.65 (7-OMe) (Watanabe et al., 2001).

Fellutanine A (9): pink solid; HR-ESI-MS m/z 373.1652 [M + H]⁺ (calcd for C₂₂H₂₁N₄O₂, 373.1659), 395.1473 [M + Na]⁺ (calcd for C₂₂H₂₁N₄O₂Na, 395.1478), ¹H-NMR (600 MHz, CDCl₃) δ 7.45 (d, 8.4, 2H, H-7, H-7′), 7.31 (d, 8.4, 2H, H-4, H-4′), 7.09 (m, 2H, H-5, H-5′), 7.01 (m, 2H, H-6, H-6′), 6.47 (s, 2H, H-2, H-2′), 4.04 (m, 2H, H-11, H-1 1′), 2.93 (m, 2H, H-10a, H-10a′), 2.18 (m, 2H, H-10b, H-10b′); ¹³C-NMR (151 MHz, CDCl₃) δ 169.85 (C-12, C-1 2′), 138.19 (C-8, C-8′), 128.75 (C-9, C-9′), 126.05 (C-2, C-2′), 122.67 (C-5, C-5′), 120.22 (C-6, C-6′), 119.82 (C-7, C-7′), 112.56 (C-4, C-4′), 109.60 (C-3, C-3′), 57.00 (C-11, C-11′), 31.52 (C-10, C-10′) (Kozlovsky et al., 2000).

Antibacterial assay

One Gram-positive (S. aureus ATCC29213) and two Gram-negative (E. coli ATCC25922 and R. solanacearum) bacteria were selected for antibacterial activity assay. Amikacin was used as a positive control. The minimum inhibitory concentrations (MIC) of the compounds and positive control were determined in sterile 96-well plates using the microculture tetrazolium (MTT) assay as described previously (Ouyang et al., 2018; Zhao et al., 2010). The synergistic antibacterial effects of terphenyllin (2) with different compounds were determined by the same method. The compounds in different combinations were mixed together at the same concentration and volume.

Antioxidant assay

Radical scavenging assay was determined by a microplate spectrophotometric method based on the reduction of a methanol solution of DPPH according to our previous reports (Ouyang et al., 2018). Inhibition (%) of free radical (DPPH) in percent was determined as [(A_control − A_sample)∕A_control] × 100, where A_control is the absorbance of the control reaction containing all reagents except the test sample, and A_sample is the absorbance of the test compounds. Tests were carried out in triplicate. BHT was used as a positive control. The IC₅₀ value was calculated using linear relation between the compound concentration and probability of the percentage of DPPH inhibition. The synergistic antioxidant effects of terphenyllin (2) with different compounds were determined by the same method. The compounds in different combinations were mixed together at the same concentration and volume as well as the synergistic antibacterial effects assay.

Purification and characterization

The solid fermentation product of Aspergillus candidus Bdf-2 was extracted with ethyl acetate (EtOAc) and the resulting extracts were partitioned into petroleum ether- and methanol (MeOH)-soluble fractions. The petroleum ether and MeOH fractions were further purified by conventional chromatographic techniques. One new p-terphenyl derivative (1) and eight known compounds (2–9) were structurally characterized (Fig. 1). Among them, compounds 1–7 were p-terphenyl derivatives. Interestingly, terphenyllin (2) (43.5 g/113.5 g) was the dominant compound in the EtOAc crude extracts of Aspergillus sp. Bdf-2.

Structure elucidation

The chemical structure of the new compound was elucidated by UV, IR, HR-ESI-MS and 1D and 2D NMR experiments (COSY, NOESY, HSQC and HMBC) (Figs. S6–S14). Compound 1 was obtained as a light yellow powder. Its molecular formula was established as C₂₁H₂₀O₄ (twelve degrees of unsaturation), according to the prominent pseudomolecular ion peak at m/z 337.1439 [M + H]⁺ and 359.1261 [M + Na]⁺ in the HR-ESI-MS spectrum (Fig. S8). The IR spectrum exhibited the absorption bands due to the presence of hydroxyl (3354 cm⁻¹), and aromatic (1591 and 1520 cm⁻¹) moieties. The ¹³C-NMR (Table 1) and HSQC spectra suggested 21 carbon signals, including four O-substituted aromatic quaternary carbons (δ_C 157.30, 154.40, 153.02 and 145.87), ten methines and three methoxy carbons (δ_C 61.04, 60.84 and 56.43). The ¹H-NMR spectrum (Table 2) of 1 showed the presence of one monosubstituted benzene at δ_H 7.61 (2H, t, 7.3, H-2″, 6″), 7.55 (2H, t, 7.7, H-3″, 5″) and 7.36(1H, t, 7.4, H-4″). The existence of a para-substituted benzene ring was also suggested by the ¹H NMR data of δ_H 7.22 (2H, d, 8.4, H-2, 6) and 6.89 (2H, d, 8.4, H-3, 5). Therefore, an aromatic proton of pentasubstituted benzene at δ_H 6.77 (1H, s, H-5′) was identified based on the left 4 degrees of unsaturation and the ¹³C NMR. The NMR spectral data of 1 indicated the presence of the p-terphenyl skeleton with one hydroxyl and three methoxy groups.

The HMBC correlations (Fig. 2) of 2′-OCH₃ with C-2′, 3′-OCH₃ with C-3′, 6′-OCH₃ with C-6′, and H-5′ with C-1′, C-3′, C-4′, and C-6′ and the NOE between 3′-OCH₃ and 4′-OCH₃, H-5′ to 6′-OCH₃ suggested that the structure of the pentasubstituted benzene ring was 2′, 3′ and 6′ three methoxyl substituted. So the hydroxyl group was set at the para-substituted benzene ring unit. The HMBC signals of H-5′ with C-1″, H-2″ with C-4′, H-2 with C-1′ established the structure of compound 1, and which was named 4″-dehydroxy-2′-methoxyterphenyllin.

The known compounds were identified according to their spectroscopic data (¹H NMR, ¹³C NMR and HR-ESI-MS) as well as comparison with literature data. The clear HR-ESI-MS and NMR spectra of known compounds were also provided in Supplementary Materials to further confirm the structure (Figs. S15–S38).

Antibacterial assay

All isolated compounds were evaluated for the antibacterial activities to a Gram-positive bacterium (S. aureus ATCC29213) and two Gram-negative bacteria (E. coli ATCC25922 and R. solanacearum), and the results were shown in Table 3. Amikacin was used as a positive control for antibacterial activity and the MIC values were all 1 µg/mL. This was the first time that compounds 1–9 were studied for antibacterial activities against R. solanacearum. Compounds 5, 6 and 7 showed antibacterial activity against S. aureus ATCC29213 and R. solanacearum. Among them, 3, 3″-dihydroxyterphenillin (6) displayed the best antibacterial activity with MIC values of 32 µg/mL against S. aureus ATCC29213 and R. solanacearum, respectively. Compounds 5 and 7 demonstrated the equal inhibitory activity against S. aureus ATCC29213 and R. solanacearum with MIC values of 64 µg/mL. However, none of the nine compounds showed inhibitory activity against E. coli ATCC25922 when applied individually. Surprisingly, when E. coli ATCC25922 was treated with the combinations of 2 + 5 and 2 + 7, significant inhibition activity was observed, and the MIC values were 128 µg/mL and 256 µg/mL, respectively. Compounds 5 and 6 also showed a synergistic effect against E. coli ATCC25922 when combined and the MIC value was 256 µg/mL. The combination of 2 + 6 displayed the strongest inhibitory activity against S. aureus ATCC29213 among the tested samples and the MIC value was 4 µg/mL, which demonstrated an obvious synergistic antibacterial effect. Combinations of 2 + 7, 5 + 6, 2 + 5, 2 + 1 and 2 + 4 also showed synergistic antibacterial effect against S. aureus ATCC29213, which was significantly stronger than any single compound. For R. solanacearum, only combinations of 2 + 1 and 2 + 4 showed synergistic antibacterial effect.

Antioxidant assay

Compounds 1–9 were also subjected to screening for possible antioxidant activity by the DPPH free radical scavenging assay (Table 4). As a result, compound 6 showed the best antioxidant activity with an IC₅₀ value of 17.62 µg/mL. Compounds 5 and 2 also displayed stronger antioxidant activity than BHT with the IC₅₀ value of 25.93 µg/mL and 39.03 µg/mL, respectively. Compound 4 displayed relatively weak antioxidant activity with an IC₅₀ value of 130.68 µg/mL. Other compounds indicated inactivity and the IC₅₀ values were larger than 200 µg/mL. Combinations of 5 + 6, 2 + 4 + 5 and 2 + 4 + 5 + 6 displayed synergistic effect and the antioxidant activities were better than the single compound. While combinations of 2 + 1, 2 + 3 and 2 + 4 showed minus effect and the IC₅₀ values were all larger than 200 µg/mL.