Highly oxygenated polyketides produced by Trichoderma koningiopsis QA-3, an endophytic fungus obtained from the fresh roots of the medicinal plant Artemisia argyi

Bioorganic Chemistry xxx (xxxx) xxxx

Contents lists available at ScienceDirect

Bioorganic Chemistry journal homepage: www.elsevier.com/locate/bioorg

Highly oxygenated polyketides produced by Trichoderma koningiopsis QA-3, an endophytic fungus obtained from the fresh roots of the medicinal plant Artemisia argyi Xiao-Shan Shia,b,c, Hong-Lei Lia,c,d, Xiao-Ming Lia,c,d, Dun-Jia Wangb, Xin Lia,c,d, ⁎ ⁎ Ling-Hong Menga,c,d, Xing-Wang Zhoub, , Bin-Gui Wanga,c,d, a

Key Laboratory of Experimental Marine Biology, Institute of Oceanology, Chinese Academy of Sciences, and Laboratory of Marine Biology and Biotechnology, Qingdao National Laboratory for Marine Science and Technology, Nanhai Road 7, Qingdao 266071, People’s Republic of China College of Chemistry and Chemical Engineering, Hubei Normal University, Cihu Road 11, Huangshi 435002, People’s Republic of China c University of Chinese Academy of Sciences, Yuquan Road 19A, Beijing 100049, People’s Republic of China d Center for Ocean Mega-Science, Chinese Academy of Sciences, Nanhai Road 7, Qingdao 266071, People’s Republic of China b

ARTICLE INFO

ABSTRACT

Keywords: Artemisia argyi Endophytic fungus Trichoderma koningiopsis Secondary metabolites Polyketides Antimicrobial activities

Eight new highly oxygenated fungal polyketides, namely, 15-hydroxy-1,4,5,6-tetra-epi-koninginin G (1), 14hydroxykoninginin E (2), koninginin U (3), 4′-hydroxykoninginin U (4), koninginin V (5), 14-ketokoninginin B (6), 14-hydroxykoninginin B (7), and 7-O-methylkoninginin B (8), together with six known related analogues (9–14), were isolated from Trichoderma koningiopsis QA-3, a fungus obtained from the inner root tissue of the well known medicinal plant Artemisia argyi. All these compounds are bicyclic polyketides, with compound 1 contains unusual hemiketal moiety at C-5 and compounds 2–14 having ketone group at C-1 and double bond at C-5(6). The structures and absolute configurations of the new compounds were established by spectroscopic analysis, X-ray crystal diffraction, modified Mosher’s method, and ECD calculation. The absolute configurations of the known compounds 9, 10, and 12 were determined by X-ray crystal diffractions for the first time. The antimicrobial activities against human pathogen, marine-derived aquatic bacteria, and plant-pathogenic fungi of compounds 1–14 were evaluated, and compound 1 showed remarkable activity against aquatic pathogen Vibrio alginolyticus with MIC value 1 µg/mL, which is as active as that of the positive control.

1. Introduction Polyketides with a hemiketal-containing bicyclic pyran skeleton, such as koninginins, are mainly produced by the species in the fungal genus Trichoderma, which exhibited plant growth regulatory [1–4] and antimicrobial activities [5–8]. However, the absolute configurations of these compounds were merely assigned by a biogenetic perspective, or by comparing the specific rotations and chemical shifts with that of the known analogues [1,9–13], and some of them were even not assigned [8,14]. In continuation of our efforts to identify new bioactive secondary metabolites from plant-derived fungi, five tricyclic and two bicyclic polyketides were previously identified from the culture extract of an endophytic fungus Trichoderma koningiopsis QA-3, which was isolated from the inner root tissue of the medicinal plant Artemisia argyi [15].

Further work on the polar fractions of the fungus resulted in the characterization of 14 bicyclic polyketides (1–14), with eight new koninginin derivatives (1–8) and six known related analogues (9–14). Among them, compound 1 contains unusual hemiketal moiety at C-5, while compounds 2–14 having ketone group and double bond at C-1 and C-5(6), respectively. The structures of compounds 1–8 were established by detailed analysis of their spectroscopic data and the absolute configurations were confirmed by X-ray crystal diffraction, modified Mosher’s method, ECD calculation, and specific optical rotations. Details of the isolation, structure elucidation, and bioactivities of compounds 1–14, as well as determination of the absolute configuration of known koninginin analogues 9, 10, and 12 by X-ray crystallographic analysis are discussed for the first time herein.

Corresponding authors at: Key Laboratory of Experimental Marine Biology, Institute of Oceanology, Chinese Academy of Sciences, and Laboratory of Marine Biology and Biotechnology, Qingdao National Laboratory for Marine Science and Technology, Nanhai Road 7, Qingdao 266071, People’s Republic of China (B.-G. Wang). E-mail addresses: [email protected] (X.-W. Zhou), [email protected] (B.-G. Wang). ⁎

https://doi.org/10.1016/j.bioorg.2019.103448 Received 4 August 2019; Received in revised form 2 November 2019; Accepted 13 November 2019 0045-2068/ © 2019 Elsevier Inc. All rights reserved.

Please cite this article as: Xiao-Shan Shi, et al., Bioorganic Chemistry, https://doi.org/10.1016/j.bioorg.2019.103448

Bioorganic Chemistry xxx (xxxx) xxxx

X.-S. Shi, et al.

Table 1 1 H NMR (500 MHz) data for compounds 1–5. No.

1a

1 2α2β 3α3β 4β 6 7α7β 8α8β 9 10 11 12 13 14 15 16 1′

3.58, 1.28, 1.76, 3.37, 1.60, 1.49, 1.78, 4.12, 3.16, 1.49, 1.28, 1.28, 1.28, 3.58, 1.02,

2′ 3′ 4′/8′ 5′/7′ 6′ 1-OH 4-OH 5-OH 10-OH 14-OH 15-OH 3′-OH 6′-OH a b

4.35, 4.66, 5.44, 4.30,

m (overlap) m (overlap)1.60, m (overlap)1.49, m m (overlap) m (overlap)1.85, m (overlap)1.28, m m m (overlap) m (overlap) m (overlap) m (overlap) m (overlap) d (6.1)

2a

3b

4a

5a

m (overlap) m (overlap)

2.46, m2.14, dt (16.5, 5.3) 1.80, m2.01, m 4.17, br s

2.65, m2.28, m 2.03, m2.22, m 4.41, br s (overlap)

2.47, m2.15, m 1.80, m2.02, m 4.20, m

2.52, m2.18, m 1.82, m2.02, m 4.21, br s

m m (overlap)

1.94, 1.84, 3.75, 3.53, 1.43, 1.38, 1.32, 3.32, 1.28, 0.84,

4.41, br s (overlap) 1.83, d (14.1)1.44, m (overlap) 4.01, m 3.56, m 1.44 m (overlap) 1.29, m (overlap) 1.29, m (overlap) 1.29, m (overlap) 1.29, m (overlap) 0.90, t (6.5) a 3.84, m b 3.72, dt (16.7, 7.4) 2.85, m

4.25, 1.83, 3.87, 3.47, 1.40, 1.27, 1.27, 1.27, 1.27, 0.87, 3.56,

4.39, br s 1.92, d (14.3)1.46, m (overlap) 3.98, dd(12.4, 2.2) 3.58, m 1.46 m (overlap) 1.28, m (overlap) 1.28, m (overlap) 1.28, m (overlap) 1.28, m (overlap) 0.87, t (6.5) 0.99, d (6.3)

7.20, d (7.1) 7.26, dd (7.1, 7.2) 7.17, d (7.2)

6.98, d (8.2) 6.64, d (8.3)

d (7.8) br s br s m (overlap)

4.30, m (overlap)

m2.24, dd (16.4,4.2) m1.50, m (overlap) m m m m m m m t (7.4)

br s m1.27, m (overlap) m m m m (overlap) m (overlap) m (overlap) m (overlap) t (6.9) m

2.63, t (7.0)

3.48, dq (11.3, 6.4) 3.54, dq (11.4, 5.3) 0.96, d (6.3)

5.37, br s

5.47, d (5.0)

5.54, br s

4.73, br s 3.75, br s

4.73, d (5.6)

4.83, br s

9.13, s

4.48, d (7.1)

Measured in DMSO‑d6. Measured in CDCl3.

2. Materials and methods

2.3. Fermentation, extraction and isolation

2.1. General experimental procedures

For chemical investigations, the fermentation was statically carried out on a rice solid medium (contained 70 g rice, 0.1 g corn flour, 0.3 g peptone, and 100 mL distilled water) in 1 L conical flasks for 30 days at room temperature. The fermented medium (90 flasks) was exhaustively extracted three times with EtOAc, which was evaporated under reduced pressure to afford an extract (76.7 g). The extract was fractionated by Si gel vacuum liquid chromatography (VLC) using different solvents of increasing polarity from petroleum ether (PE) to MeOH to yield 10 fractions (Frs. 1–10) based on HPLC and TLC analysis. Purification of Fr. 5 (4.3 g) by reversed-phase column chromatography (CC) over Lobar LiChroprep RP-18 with a MeOH-H2O gradient (from 10:90 to 100:0) yielded eight subfractions (Frs. 5.1–5.8). Fr. 5.8 (280 mg) was purified by CC on Sephadex LH-20 (MeOH) and then by recrystallization to obtain compound 9 (116.5 mg). Fr.7 (10.2 g) was separated by CC on Lobar LiChroprep C18 eluting with MeOH-H2O gradient to give 15 sub-fractions (Frs.7.1–7.15). Further purification of Fr.7.4 (1.9 g) by CC on Si gel eluting with a CH2Cl2-MeOH gradient (from 100:1 to 10:1) and then by prep. TLC (plate: 20 × 20 cm, developing solvents: PEEtOAc, 1:1) yielded compounds 1 (8.4 mg) and 6 (4.2 mg). Fr. 7.5 (2.8 g) was separated by CC on Si gel eluting with a CH2Cl2-EtOAc gradient (from 20:1 to 2:1) and then by prep. TLC (plate: 20 × 20 cm, developing solvents: PE-EtOAc, 1:1) yielded compound 13 (83.6 mg), and the residue was further purified by CC on Sephadex LH-20 (MeOH) to obtain compound 14 (21.4 mg). Fr. 7.6 (357 mg) was subjected to CC on Si gel eluting with a CH2Cl2-EtOAc gradient (from 100:1 to 10:1) and then by prep. TLC (plate: 20 × 20 cm, developing solvents: CH2Cl2MeOH, 20:1) yielded compound 11 (9.0 mg), and the residue was further purified by prep. TLC (plate: 20 × 20 cm, developing solvents: PE-Acetone, 2:1) yielded compounds 10 (6.4 mg) and 12 (13.3 mg). Fr. 7.9 (500 mg) was subjected to CC on Sephadex LH-20 (MeOH) and then further purified by prep. TLC (plate: 20 × 20 cm, developing solvents:

Optical rotations were measured on an Optical Activity AA-55 polarimeter. UV spectra were recorded on a Gold Spectrumlab 54 UV-Vis spectrophotometer. ECD spectra were acquired on a Chirascan spectropolarimeter. NMR spectra were recorded on a Bruker Avance 500 spectrometer (500 MHz for 1H and 125 MHz for 13C). Mass spectra were recorded on an API QSTAR Pulsar 1 Orbitrap mass spectrometer. HPLC analysis was performed on a Dionex HPLC system equipped with a P680 pump, an ASI-100 automated sample injector, a TCC-100 column oven, a UV-DAD 340U detector, and a Dionex Acclaim ODS column (4.6 × 250 mm, 5 μm) or a CHIRALPAK AD-H column (4.6 × 250 mm, 5 μm). Si gel (100–200 mesh and 200–300 mesh) for column chromatography and preparative thin-layer chromatography were produced by Qingdao Haiyang Chemical Group Corporation. RP18 reverse-phase Si gel (40–63 μm) and Sephadex LH-20 were purchased from the Merck Corporation. Solvents were distilled prior to use for extraction and purification procedures. 2.2. Plant and fungal materials The endophytic fungus T. koningiopsis QA-3 was isolated from the inner tissue of the medicinal plant A. argyi collected at Qichun of the Hubei Province in the central China, in July 2014, using a protocol as described in our previous report [16]. Fungal identification was performed by analysis of its ITS region of the rDNA as described previously [16]. The resulting sequence, which has been deposited in GenBank with accession no. MF616361, was same (100%) to that of T. koningiopsis DMC 795b (compared with EU 718083.1). The strain is preserved at the Key Laboratory of Experimental Marine Biology, Institute of Oceanology of the Chinese Academy of Sciences (IOCAS). 2

Bioorganic Chemistry xxx (xxxx) xxxx

X.-S. Shi, et al.

Table 2 13 C NMR (125 MHz) data for compounds 1–5. No.

1a

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 1′ 2′ 3′ 4′/8′ 5′/7′ 6′

71.2, 29.8, 25.5, 68.7, 96.5, 43.0, 20.9, 26.7, 74.7, 72.9, 32.4, 25.2, 25.1, 39.5, 65.8, 23.6,

a b

CH CH2 CH2 CH C CH CH2 CH2 CH CH CH2 CH2 CH2 CH2 CH CH3

Table 3 1 H NMR (500 MHz) data for compounds 6–8 in DMSO‑d6.

2a

3b

4a

5a

No.

6

7

8

196.3, C 32.2, CH2 29.3, CH2 64.3, CH 170.2, C 110.1, C 17.3, CH2 21.6, CH2 80.0, CH 71.0, CH 31.8, CH2 23.5, CH2 36.6, CH2 65.7, CH 29.7, CH2 9.9, CH3

196.4, C 33.8, CH2 29.4, CH2 66.5, CH 171.4, C 112.3, C 63.7, CH 28.9, CH2 76.8, CH 73.2, CH 33.2, CH2 25.4, CH2 29.0, CH2 31.9, CH2 22.8, CH2 14.2, CH3 70.3, CH2 36.7, CH2 139.5, C 129.1, CH 128.4, CH 126.2, CH

195.6, C 32.9, CH2 29.4, CH2 64.8, CH 172.3, C 111.1, C 63.3, CH 27.5, CH2 76.0, CH 70.8, CH 32.1, CH2 25.2, CH2 28.7, CH2 31.3, CH2 22.0, CH2 13.9, CH3 69.4, CH2 35.1, CH2 129.1, C 129.5, CH 114.9, CH 155.5, C

197.1, C 32.7, CH2 29.2, CH2 64.6, CH 172.9, C 111.5, C 62.9, CH 27.7, CH2 76.1, CH 70.7, CH 31.3, CH2 25.2, CH2 28.7, CH2 31.9, CH2 22.0, CH2 13.9, CH3 17.1, CH3 76.4, CH 68.5, CH 18.1, CH3

2α

3.92, m

3.98, dd (11.4, 4.1)

3α3β 4α4β 7α7β 8α8β

2.06, m1.64, m (overlap) 1.42, m (overlap)1.50, m (overlap) 1.64, m (overlap)1.23, m 1.84, m1.50, m (overlap)

3.91, dd (12.4, 5.0) 2.07, m1.67, m 2.36, m2.52, m

9 10 11 12 13a13b 14 15 16 2-OH 7-OMe 10-OH 14-OH

3.75, 3.50, 1.42, 1.50, 2.34, – 2.40, 0.91, 4.93, – 4.82, –

m br s m (overlap) m (overlap) m1.94, m m t (7.1) br s br s

2.32, m1.94, m 1.85, m1.53, m 3.75, 3.50, 1.42, 1.31, 1.31, 3.30, 1.31, 0.83, 4.92, – 4.76, 3.85,

d (10.9) m m m (overlap) m (overlap) m m (overlap) t (7.4) br s br s m

2.08, m1.66, m 2.36, m2.63, m 4.20, 1.95, m 3.92, 3.53, 1.48, 1.25, 1.25, 1.25, 1.25, 0.86, 4.93, 3.21, 4.77, –

s d (14.4)1.48, d (12.9) m m (overlap) m (overlap) m (overlap) m (overlap) m (overlap) t (6.6) d (2.9) s d (5.9)

λmax (Δε) 212 (–37.57), 254 (+110.81), 289 (–14.43) nm; 1H and 13C NMR data, Tables 1 and 2; ESIMS m/z 371.24 [M+H]+; HRESIMS m/z 371.2426 [M+H]+ (calcd for C20H35O6+, 371.2428). 14-Ketokoninginin B (6): colorless oil; [ ]25 D + 80.7 (c 1.15, MeOH); UV (MeOH) λmax (log ε) 261 (2.87) nm; ECD (2.74 mM, MeOH) λmax (Δε) 260 (+7.27), 290 (–0.41) nm; 1H and 13C NMR data, Tables 3 and 4; ESIMS m/z 297.17 [M+H]+, 319.15 [M+Na]+; HRESIMS m/z 297.1693 [M+H]+ (calcd for C16H25O5+, 297.1697). 14-Hydroxykoninginin B (7): colorless oil; [ ]25 D + 162.1 (c 1.45, MeOH); UV (MeOH) λmax (log ε) 260 (3.13) nm; ECD (4.58 mM, MeOH) λmax (Δε) 259 (+46.52), 293 (–3.25) nm; 1H and 13C NMR data, Tables 3 and 4; ESIMS m/z 299.18 [M+H]+, 321.17 [M+Na]+; HRESIMS m/ z 321.1665 [M+Na]+ (calcd for C16H26O5Na+, 321.1672). 7-O-Methylkoninginin B (8): colorless, amorphous powder; [ ]25 D + 86.1 (c 0.73, MeOH); UV (MeOH) λmax (log ε) 252 (3.56) nm; ECD (4.92 mM, MeOH) λmax (Δε) 203 (–19.98), 252 (+47.93), 287 (–2.58) nm; 1H and 13C NMR data, Tables 3 and 4; ESIMS m/z 313.20 [M+H]+; HRESIMS m/z 313.2009 [M+H]+ (calcd for C17H29O5+, 313.2010).

Measured in DMSO‑d6. Measured in CDCl3.

PE-acetone, 2:1) yielded compound 5 (66.3 mg). Fr. 7.10 (89 mg) was fractioned by CC on Sephadex LH-20 (MeOH) and further purified by prep. TLC (plate: 20 × 20 cm, developing solvents: CHCl2-MeOH, 20:1), afforded compounds 3 (11.5 mg) and 4 (9.0 mg). Fr.8 (8.2 g) was separated by CC on Lobar LiChroprep C18 eluting with MeOH-H2O gradient to give seven sub-fractions (Frs.8.1–8.7). Further purification of Fr.8.2 (121 mg) by CC on Sephadex LH-20 (MeOH) and then by prep. TLC (plate: 20 × 20 cm, developing solvents: PE-acetone, 1:1) yielded compounds 7 (25.0 mg) and 2 (21.7 mg). Fr. 8.3 (101.3 mg) was separated by prep. TLC (plate: 20 × 20 cm, developing solvents: PEacetone, 1:1) yielded compound 8 (7.0 mg). 2.4. Spectroscopic data 15-Hydroxy-1,4,5,6-tetra-epi-koninginin G (1): colorless crystals; mp 1 13 64–66 °C; [ ]25 C NMR data, Tables 1 D + 17.7 (c 0.34, MeOH); H and and 2; ESIMS m/z 319.2 [M+H]+; HRESIMS m/z 336.2380 [M +NH4]+ (calcd for C16H34NO6+, 336.2381). 14-Hydroxykoninginin E (2): colorless crystals; mp 167–169 °C; [ ]25 D + 119.6 (c 1.12, MeOH); UV (MeOH) λmax (log ε) 261 (3.12) nm; ECD (3.36 mM, MeOH) λmax (Δε) 258 (+18.46), 290 (–1.77) nm; 1H and 13C NMR data, Tables 1 and 2; ESIMS m/z 299.19 [M+H]+, 321.17 [M+Na]+; HRESIMS m/z 299.1853 [M+H]+ (calcd for C16H27O5+, 299.1853) and m/z 321.1670 [M+Na]+ (calcd for C16H26O5Na+, 321.1672). Koninginin U (3): colorless oil; [ ]25 D + 30.2 (c 0.43, MeOH); UV (MeOH) λmax (log ε) 254 (2.87) nm; ECD (2.49 mM, MeOH) λmax (Δε) 219 (–4.59), 255 (+7.78), 288 (–0.52) nm; 1H and 13C NMR data, Tables 1 and 2; ESIMS m/z 403.25 [M+H]+, 425.23 [M+Na]+; HRESIMS m/z 403.2475 [M+H]+ (calcd for C24H35O5+, 403.2479) and m/z 425.2294 [M+Na]+ (calcd for C24H34O5Na+, 425.2298). 4′-Hydroxykoninginin U (4): colorless, amorphous powder; [ ]25 D + 26.8 (c 0.82, MeOH); UV (MeOH) λmax (log ε) 225 (3.61) nm, 254 (3.98) nm; ECD (3.98 mM, MeOH) λmax (Δε) 219 (–18.99), 252 (+46.07), 290 (–4.52) nm; 1H and 13C NMR data, Tables 1 and 2; ESIMS m/z 419.24 [M+H]+, 441.22 [M+Na]+; HRESIMS m/z 419.2423 [M+H]+ (calcd for C24H35O6+, 419.2428) and m/z 441.2243 [M+Na]+ (calcd for C24H34O6Na+, 441.2248). Koninginin V (5): white, amorphous powder; [ ]25 D + 73.0 (c 1.15, MeOH); UV (MeOH) λmax (log ε) 255 (4.01) nm; ECD (4.76 mM, MeOH)

2.5. X-ray crystallographic analysis of compounds 1, 2, 9, 10, and 12 All crystallographic data were collected on an Agilent Xcalibur Eos Gemini CCD plate diffractometer equipped with graphiteTable 4 13 C NMR (125 MHz) data for compounds 6–8 in DMSO‑d6.

3

no.

6

7

8

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 7-OMe

197.4, C 70.3, CH 29.2, CH2 26.7, CH2 170.5, C 108.7, C 17.6, CH2 21.7, CH2 80.0, CH 70.6, CH 31.5, CH2 19.8, CH2 41.4, CH2 210.9, C 34.9, CH2 7.6, CH3 –

197.5, C 70.4, CH 29.3, CH2 26.8, CH2 170.7, C 108.8, C 17.7, CH2 21.7, CH2 80.1, CH 70.9, CH 32.3, CH2 21.9, CH2 36.6, CH2 71.0, CH 29.8, CH2 10.0, CH3 –

197.0, C 70.3, CH 29.3, CH2 26.7, CH2 172.1, C 110.1, C 65.1, CH 26.9, CH2 75.9, CH 70.4, CH 31.2, CH2 25.3, CH2 28.7, CH2 32.2, CH2 22.0, CH2 13.9, CH3 55.9, CH3

Bioorganic Chemistry xxx (xxxx) xxxx

X.-S. Shi, et al.

monochromatic Cu-Kα radiation (λ = 1.54178 Å) at 293(2) K [17]. The data were corrected for absorption by using the program SADABS [18]. The structures were solved by direct methods with the SHELXTL software package [19]. All non-hydrogen atoms were refined anisotropically. The H atoms were located by geometrical calculations, and their positions and thermal parameters were fixed during the structures refinement. The structures were refined by full-matrix least-squares techniques [20]. Crystal data for compound 1: C16H30O6·H2O, F.W. = 336.42, Monoclinic space group P2(1), unit cell dimensions a = 7.5508(7) Å, b = 5.5222(7) Å, c = 22.236(3) Å, V = 926.14(19) Å3, α = γ = 90°, β = 92.719(2)°, Z = 2, dcalcd = 1.206 mg/m3, crystal dimensions 0.14 × 0.10 × 0.03 mm3, μ = 0.775 mm−1, F(000) = 368. The 2182 measurements yielded 1559 independent reflections after equivalent data were averaged, and Lorentz and polarization corrections were applied. The final refinement gave R1 = 0.0654 and wR2 = 0.1521 [I > 2σ(I)]. The Flack parameter was 0.0(6) in the final refinement for all 2182 reflections with 1559 Friedel pairs. Crystal data for compound 2: C16H26O5·H2O, F.W. = 316.38, Monoclinic space group P1, unit cell dimensions a = 5.1217(5) Å, b = 7.0810(8) Å, c = 12.5489(11) Å, V = 421.66(7) Å3, α = 79.916(2)°, β = 82.272(2)°, γ = 70.7840(10)°, Z = 1, dcalcd = 1.246 mg/m3, crystal dimensions 0.20 × 0.15 × 0.13 mm3, μ = 0.778 mm−1, F(000) = 172. The 1707 measurements yielded 1607 independent reflections after equivalent data were averaged, and Lorentz and polarization corrections were applied. The final refinement gave R1 = 0.0684 and wR2 = 0.1843 [I > 2σ(I)]. The Flack parameter was 0.00 in the final refinement for all 1707 reflections with 1607 Friedel pairs. Crystal data for compound 9: C18H28O5·H2O, F.W. = 342.42, Monoclinic space group P2(1), unit cell dimensions a = 5.6260(6) Å, b = 7.7887(7) Å, c = 21.488(2) Å, V = 940.46(17) Å3, α = γ = 90°, β = 92.799(2)°, Z = 2, dcalcd = 1.209 mg/m3, crystal dimensions 0.22 × 0.14 × 0.04 mm3, μ = 0.736 mm−1, F(000) = 372. The 2174 measurements yielded 1231 independent reflections after equivalent data were averaged, and Lorentz and polarization corrections were applied. The final refinement gave R1 = 0.0655 and wR2 = 0.1336 [I > 2σ(I)]. The Flack parameter was 0.0(8) in the final refinement for all 2174 reflections with 1231 Friedel pairs. Crystal data for compound 10: C16H26O5, F.W. = 298.37, Monoclinic space group P2(1), unit cell dimensions a = 5.8396(2) Å, b = 7.8781(3) Å, c = 17.2282(8) Å, V = 792.58(5) Å3, α = γ = 90°, β = 90.0640(10)°, Z = 2, dcalcd = 1.250 mg/m3, crystal dimensions 0.37 × 0.11 × 0.07 mm3, μ = 0.751 mm−1, F(000) = 324. The 2650 measurements yielded 2543 independent reflections after equivalent data were averaged, and Lorentz and polarization corrections were applied. The final refinement gave R1 = 0.0373 and wR2 = 0.0982 [I > 2σ(I)]. The Flack parameter was 0.1(2) in the final refinement for all 2650 reflections with 2543 Friedel pairs. Crystal data for compound 12: C17H28O5, F.W. = 312.39, Orthorhombic space group P2(1)2(1)2(1), unit cell dimensions a = 5.0189(4) Å, b = 13.3772(9) Å, c = 25.8044(18) Å, V = 1732.5(2) Å3, α = β = γ = 90°, Z = 4, dcalcd = 1.198 mg/m3, crystal dimensions 0.23 × 0.15 × 0.07 mm3, μ = 0.708 mm−1, F (000) = 680. The 2414 measurements yielded 1710 independent reflections after equivalent data were averaged, and Lorentz and polarization corrections were applied. The final refinement gave R1 = 0.0490 and wR2 = 0.0991 [I > 2σ(I)]. The Flack parameter was 0.0(5) in the final refinement for all 2414 reflections with 1710 Friedel pairs.

Gaussian 09 software to give the energy-minimized conformers. Then, the optimized conformers were subjected to the calculations of ECD spectra using TDDFT at PEB0/TZVP level; solvent effects of the MeOH solution were evaluated at the same DFT level using the SCRF/PCM method [21]. 2.7. Antimicrobial assays Antimicrobial evaluation against human pathogen (Escherichia coli EMBLC-1), marine-derived aquatic pathogenetic bacteria (Edwardsiella megi QDIO-10, Ed. tarda QDIO-2, Vibrio alginolyticus QDIO-5, V. harveyi QDIO-7), and plant-pathogenic fungi (Bipolaris sorokiniana QDAU-7, Ceratobasidium cornigerum QDAU-8, and Helminthosporium maydis QDAU-18) was carried out by the microplate assay [22]. The aquatic pathogens were obtained from the Institute of Oceanology, Chinese Academy of Sciences, while the plant pathogenic fungi were provided by the Qingdao Agricultural University. Chloramphenicol and amphotericin B were used as the positive control against bacteria and fungi, respectively. 3. Results and discussion 3.1. Structural elucidation of the new compounds The EtOAc extracts from the culture of T. koningiopsis QA-3 were separated and purified by repeated column chromatography (CC) on Si gel, Lobar LiChroprep RP-18, and Sephadex LH-20 as well as by preparative TLC, to yield compounds 1–14 (Fig. 1). Compound 1 was obtained as colorless crystals and its molecular formula was determined as C16H30O6 on the basis of HRESIMS data, indicating two degrees of unsaturation. The 1H and 13C NMR data of 1 indicated the presence of one methyl, eight methylenes, six methines (with five oxygenated), and one oxygenated non-protonated carbon as well as five exchangeable protons (1-OH, 4-OH, 5-OH, 10-OH, and 15OH) (Tables 1 and 2). A comparison of the NMR data of 1 with those of koninginin G, a koninginin derivative obtained from the fungus Trichoderma aureoviride, indicated that compound 1 could be an analogue of the koninginin possessing the same bicyclic skeleton and C-9 side chain as seen in koninginin G [4]. However, resonances for the methylene group CH2-15 at δC 22.6 and δH 1.50 (2H) in the NMR spectra of koninginin G were not present in those of 1. Instead, signals for an additional oxygenated methine group at δC 65.8 and δH 3.56 (1H) were observed in the NMR spectra of 1. Moreover, the 13C NMR resonances of C-1, C-2, C-4, and C-7, compared to those of koninginin G [4], were slightly shielded (−2.3 ~ −4.9), while C-10 and C-11 were deshielded (+3.4 and +3.0, respectively) in the 13C NMR spectrum of 1. The COSY correlations from H-15 to H-14 and H-16 as well as HMBC cross peaks from the proton of 15-OH to C-15 and C-16 and from H-16 to C14 and C-15 (Fig. 2) also indicated the presence of 15-OH group in 1. The above spectroscopic features suggested that compound 1 is 15hydroxylated derivative of koninginin G, but having different steric configurations. The relative and absolute configurations of 1 were unambiguously established by a single-crystal X-ray diffraction experiment using Cu Kα radiation (Fig. 3) The Flack parameter 0.0(6) allowed for the establishment of the absolute configuration of 1 as 1S, 4R, 5R, 6S, 9S, 10S, and 15S. On the basis of the above data, the structure of 1 was determined, and was named as 15-hydroxy-1,4,5,6-tetra-epi-koninginin G. 14-Hydroxykoninginin E (2) was initially obtained as an amorphous powder and its molecular formula was determined as C16H26O5 (four degrees of unsaturation) on the basis of HRESIMS data. Comprehensive analysis of the 1H and 13C NMR spectra of 2 (Tables 1 and 2) suggested that 2 might be a hydroxylated analogue of koninginin E (14, Fig. 1), a fungal polyketide isolated from T. koningii ATCC 46314 [3]. This was evidenced that the resonance for a methylene group in the 13C NMR spectrum of koninginin E disappeared in that of 2. Instead, signals for

2.6. Computational section Conformational searches were performed via the molecular mechanics using MM+ method in HyperChem 8.0 software, and the geometries were further optimized at B3LYP/6-31G(d) level in vacuo via 4

Bioorganic Chemistry xxx (xxxx) xxxx

X.-S. Shi, et al.

Fig. 1. Structures of compounds 1–14.

confirmed its relative and absolute configuration (Fig. 3). The final refinement of the Cu Kα data resulted in a 0.00 Flack parameter, allowing for the unambiguous assignment of the absolute configuration as 4R, 9S, 10S, and 14R. The molecular formula of compounds 3 and 4 were determined by HRESIMS to be C24H34O5 and C24H34O6, respectively, with one O-atom difference, but both having eight degrees of unsaturation. The 1H and 13 C NMR data (Tables 1 and 2) of 3 and 4 were similar to those of 7-Omethylkoninginin D (12, Fig. 1), a fungal polyketide isolated from marine mud-derived fungus T. koningii [11] and our present study as well. However, signals for the methoxy group in the NMR spectra of 12 disappeared in those of 3 and 4. Instead, resonances for phenethoxy and para-hydroxyphenethoxy groups were observed in the NMR spectra of 3 and 4, respectively (Tables 1 and 2). HMBC correlations from H-7 to C1′ and from H-1′ to C-7 evidenced that the phenethoxy moiety in 3 and para-hydroxyphenethoxy residue in 4 linked to C-7 in the molecules (Fig. 2). In the NOESY spectra of compounds 3 and 4, NOEs from H-8β to H-7 indicated the cofacial orientation of these groups, while correlation from H-8α to H-9 located these protons on the other side of the molecules (Fig. 4). However, no other diagnostic NOEs were observed and thus the relative configurations of compounds 3 and 4 could not be fully characterized by NOESY experiments. On the other hand, It is not helpful to determine the absolute configuration of 3 and 4 by comparing with those of the literature data, since their most similar compound 7-O-methylkoninginin D (12) was solely assigned from the biogenetic point of view comparing to that of koninginin D [11]. To unambiguously clarify the stereochemistry, we turned our efforts to cultivate quality crystals for compounds 3, 4, and 12. After many attempts by optimizing the condition of crystallization, the suitable crystals for 12 were obtained to make a feasible X-ray crystallographic analysis using Cu Kα radiation, while it was not successful for compounds 3 and 4. The results from X-ray experiments, with Flack parameter 0.0(5), not only confirmed the planar structure of compound 12, but also unequivocally defined its relative and absolute configurations for the first time as 4R, 7R, 9S, and 10S (Fig. 3). The very similar ECD (electronic circular dichroism) curves of 3, 4, and 12, which showed negative Cotton effects (CEs) around 219 and 288 nm and a positive CE near 255 nm, allowed the assignment of the absolute configuration of 3 and 4 same as that of 12. The absolute configurations of compounds 3 and 4 were also investigated by the time-dependent density functional (TDDFT)-ECD

Fig. 2. Key HMBC (arrows) and COSY (bold lines) correlations of compounds 1–8.

an oxygenated methine group at δH 3.32 (H-14) and δC 70.9 (C-14) were observed in the NMR spectra of 2 (Tables 1 and 2). The COSY and HMBC correlations (Fig. 2) supported this deduction. The relative configuration of 2 could not be determined by NOESY experiment since no diagnostic NOE interactions were observed. Upon slow evaporation of the solvent (MeOH:H2O = 10:1) by storing in a refrigerator, quality single crystals of compound 2 were obtained, making feasible an X-ray diffraction analysis that unequivocally 5

Bioorganic Chemistry xxx (xxxx) xxxx

X.-S. Shi, et al.

Fig. 3. X-ray crystallographic structures of compounds 1, 2, and 12.

7

7

9

1

4

9

4 3

3

4

7 9 2

10

9 10

5

4

2

6

5

9

7

2

10

8

Fig. 4. Key NOE correlations of compounds 2–8.

calculation in Gaussian 09 [21]. Geometry optimization of each possible isomer of 3 and 4 was performed to generate minimum energy conformers, and then TDDFT method was employed at the PEB0/TZVP level to get calculated ECD spectra of 3 and 4. Assessment of the experimental and calculated ECD spectra of 3 and 4 revealed excellent agreement with that calculated for (4R, 7R, 9S, and 10S)-3 and -4 (Fig. 5), which agreed with the conclusion from the ECD comparison with that of 7-O-methylkoninginin D (12). Thus, the structures of compounds 3 and 4 were fully characterized and were named as koninginin U and 4′-hydroxykoninginin U, respectively. Compound 5 was obtained as a white amorphous powder. The molecular formula was determined to be C20H34O6 (four degrees of unsaturation) by analysis of its HRESIMS data. The 1H and 13C NMR spectroscopic data (Tables 1 and 2) were also similar to those of 7-O-

Fig. 5. The experimental and calculated ECD spectrum of compounds 3–5.

methylkoninginin D (12, Fig. 1) [11]. However, signals for the methoxy group in the NMR spectra of 12 disappeared in those of 5. Instead, additional resonances were observed at δH/δC 0.99/17.1 (CH3-1′), 3.48/76.4 (CH-2′), 3.54/68.5 (CH-3′), and 0.96/18.1 (CH3-4′) (Tables 1 6

Bioorganic Chemistry xxx (xxxx) xxxx

X.-S. Shi, et al.

and 2), suggesting the presence of a 3-hydroxybutan-2-oxyl group in 5, which was strongly supported by the observed COSY cross peaks (Fig. 2). In the HMBC spectrum, correlations from H-2′ to C-7, and vice versa, demonstrated that the 3-hydroxybutan-2-oxyl group connected at C-7 (Fig. 2). In the NOESY experiments, NOE correlations from H-7 to H-8β indicated the cofacial orientation of these groups (Fig. 4), while interaction from H-9 to H-8α located these protons on the other side of the molecule. As for H-2′ and H-3′, the absence of a big J value (usually > 10 Hz) among these two protons support them in cis-relationship [1]. However, the relative configurations of C-4 and C-10 could not be established by NOESY spectrum since no diagnostic NOE correlations to assign the configurations of C-4 and C-10 were observed, but from a biosynthetic point of view, they should have same configuration as that of the co-occurring compounds 1–4. The ECD spectrum of 5 also showed negative CEs around 212 and 289 nm and positive CE near 254 nm, which were nearly identical to that of compounds 3, 4, and 12, and closely matched with that of the calculated curve for 5 as well (Fig. 5). All these data suggested that the ECD spectra in this class of compounds are largely dictated by the configurations of the 3,4,7,8-tetrahydro-2H-chromen-5(6H)-one moiety, whereas the overall shape of ECD curves is likely unaffected by the residues substituted at C-7 (3–5 vs 12, Fig. 5). To define the absolute configuration of C-2′ and C-3′ of compound 5, modified Mosher’s reaction was performed [23]. Briefly, compound 5 was separately esterified with (S)- and (R)-MTPA chloride to give the corresponding (R)and (S)- MTPA esters, 5a and 5b, respectively. The distribution of Δδ values between 5b and 5a (Fig. 6) not only deduced the 2′R, 3′R-configuration, but also confirmed the 4R, 7R, 9S, and 10S-configuration for 5. The structure of compound 5 was thus determined and the trival name koninginin V was assigned to this compound. Compounds 6–8 were obtained as colorless oil (6 and 7) or amorphous powder (8), and had the molecular formula C16H24O5, C16H26O5, and C17H28O5, respectively, as determined by HRESIMS experiments. Comprehensive analysis of their 1H and 13C NMR data (Tables 3 and 4) suggested that compounds 6–8 also belonging to fungal polyketides with structural similarity to that of koninginin B (11, Fig. 1), a fungal polyketide characterized from T. koningii (isolated from the roots and soil line of a Diffenbachia sp.) [2] and in our present study as well. However, one of the two OH groups in koninginin B was improperly assigned at C-4 in the original proposed structure [2], and was revised to be at C-2 in a synthetic approach [10]. Comparing to koninginin B (11), resonances for the methylene group at δH/δC 1.30/31.75 (CH2-14) of 11 disappeared in those of 6 and 7. Instead, signals corresponding to a ketone group at δC 210.9 (C-14) for 6 and resembling an oxymethine group at δH/δC 3.30/71.0 (CH-14) for 7 were observed in their NMR spectra (Tables 3 and 4). These data suggested that the methylene group CH2-14 in 11 was replaced by a keto and an oxymethine groups in 6 and 7, respectively. As in the case of compound 8, the main difference in the NMR spectra of 8 comparing to that of 11 was evident in the methylene group at δH 2.09/2.57 and δC 17.7 (CH2-7) of 11 were not detected in that of 8, whereas signals for a –CH–OMe moiety at δH/

δC 4.20/65.1 (CH-7) and 3.21/55.9 (OMe-7) were observed in the NMR spectra of 8 (Tables 3 and 4). These observations indicated that 8 differed from 11 by replacing of the methylene group (CH2-7) in 11 with the –CH–OMe moiety in 8. All of the required key COSY and HMBC correlations, which fully supporting the differences of compounds 6–8 with that of 11, were observed as shown in Fig. 2. In the NOESY spectra of compounds 6–8, NOEs from H-8α to H-9 indicated the cofacial orientation of these groups, while correlations from H-8β to H-7β and from H-10 to H-7β or H-8β located these protons on the other side of the molecules (Fig. 4). However, no other diagnostic NOEs could be observed and thus the relative configurations of compounds 6–8 could not be fully characterized by NOESY experiments. Efforts to obtain quality crystals were not successful for compounds 6–8, thus determining its relative and absolute configurations by X-ray crystallographic diffraction is also not applicable. On the other hand, the absolute configurations of the koningiopisin E (9) [8] and koninginin F (10) [10] were mainly assigned from the biogenetic point of view or by synthetic approach [10]. After many attempts by optimizing the condition of crystallization, quality single crystals for the known polyketides koningiopisin E (9) [8] and koninginin F (10) [10] were obtained to make a feasible X-ray diffraction analysis and further unequivocally confirmed their relative and absolute configuration (Fig. 7). The final refinement of the Cu Kα data resulted in the Flack parameters 0.0(8) for 9 and 0.1(2) for 10, allowing for the unambiguous assignment of the absolute configuration of compounds 9 and 10 for the first time as 2R, 9S, and 10S and 2R, 7R, 9S, and 10S, respectively (Fig. 7). The very similar ECD curves of 6, 7, and 9, which showed a negative CE around 290 nm and a positive CE near 260 nm (Supplementary data, Fig. S51), indicated that compounds 6 and 7 have same absolute configuration as that of 9 at C-2, C-9, and C10, while compounds 8 and 10 had similar ECD shapes with two negative CEs around 203 and 287 nm and a positive CE near 252 nm (Supplementary data, Fig. S52), revealed that compound 8 has same absolute configuration as that of 10 at C-2, C-7, C-9, and C-10. However, the absolute configuration of C-14 could not be established in 7, but from a biosynthetic point of view, it should have 14R-configuration, same as that of the co-occurring compound 2. Thus, the structures of compounds 6–8 were characterized and were named as 14-ketokoninginin B, 14-hydroxykoninginin B, and 7-O-methylkoninginin B, respectively. In addition to compounds 1–8, six known analogues including koningiopisin E (9) [8], koninginin F (10) [10], koninginin B (11) [2,10], 7-O-methylkoninginin D (12) [11], koninginin D (13) [1,10], and koninginin E (14) [3] were also isolated and identified. Their structures were elucidated by comparing their NMR data with those reported in the literature. The absolute configurations of compounds 9, 10, and 12 were previously assigned in published reports by a biogenetic perspective, or by comparing the specific rotations and chemical shifts with that of the known analogues, but in the present study the singlecrystal X-ray diffraction was used to confirm their absolute configurations. 3.2. Antimicrobial activities of the isolated compounds The six known compounds 9–14 were previously assayed for antimicrobial activities against Alternaria baumanii, A. panax, Fusarium oxysporum f. sp. ciceris, F. solani, Gaeumannomyces graminis var. tritici, Plectosphaerella cucumerina, Rhizoctonia bataticola, and/or Staphylococcus aureus in literature reports [5–8]. However, none of them were tested for the activity against marine-derived aquatic pathogenetic bacteria (MdAPB). Since the EtOAc extracts of the fungus exhibited potency against MdAPBs in our preliminary assays, the known compounds 9–14 as well as the new compounds 1–8 were evaluated for the activity against MdAPBs, and the property against some of the human pathogens and plant-pathogenic fungi were screened as well. All of these compounds showed inhibitory activity

Fig. 6. Δδ (=δS − δR) values for (S)- and (R)-MTPA esters of 5. 7

Bioorganic Chemistry xxx (xxxx) xxxx

X.-S. Shi, et al.

Fig. 7. X-ray crystallographic structures of compounds 9 and 10. Table 5 Antibacterial activity of compounds 1–14 (MIC, μg/mL). No.

1

2

3

4

5

6

7

8

9

10

11

12

13

14

Chloramphenicol

EC EM ET VAl VH

64 32 – 1 –

64 64 – 64 –

64 16 – 64 4

64 64 – 64 –

64 64 – 64 –

64 32 2 64 –

64 64 – 64 –

64 32 – 64 –

64 16 – 64 –

64 64 – 64 –

64 64 – 64 –

64 64 – 64 –

64 64 – 64 –

64 64 – 64 –

2 0.5 0.5 1 1

EC: E. coli. EM: Ed. megi. ET: Ed. tarda. VAl: V. alginolyticus. VH: V. harveyi. –: no activity (MIC > 64 µg/mL).

absolute configurations of compounds 1, 2, 9, 10, and 12 were unequivocally assigned by X-ray crystal diffraction experiments. Compound 1 showed potent inhibitory activity against aquatic bacterium V. alginolyticus with MIC value 1 µg/mL, which is comparable to that of the positive control chloramphenicol, and compounds 2, 3, and 6 showed potent activity against C. cornigerum, each with MIC value of 8 µg/mL, which is same as that of the positive control amphotericin B. These results prove that compounds 1–3 and 6 might be used as potential molecules in the development of drug leads or being modified to develop more active derivatives for the treatments of microbial infection.

Table 6 Antifungal activity of compounds 1–14 (MIC, μg mL−1). No.

1

2

3

4

5

6

7

8

9

10

11

12

13

14

Amphotericin B

BS CC HM

– – –

– 8 –

– 8 –

8 – 16

– – –

– 8 –

– – –

– 32 –

– 16 4

32 32 –

– 32 –

– 16 –

– 64 –

– 16 –

0.5 8 2

BS: B. sorokiniana. CC: C. cornigerum. HM: H. maydis. –: no activity (MIC > 64 µg/mL).

against human pathogen Escherichia coli and aquatic pathogens Edwardsiella megi and Vibrio alginolyticus with MIC values ranging from 1 to 64 µg mL−1 (Table 5), while compounds 3 and 6 further exhibited activity against aquatic bacteria V. harveyi and Ed. tarda with MICs 4 and 2 µg/mL, respectively. Among the compounds tested, compound 1 showed strongest activity against V. alginolyticus with MIC value 1 µg/ mL, which is comparable to that of the positive control chloramphenicol. In the assays against plant pathogenic fungi, compounds 2, 3, and 6 showed potent activity against C. cornigerum, each with MIC value of 8 µg/mL, which is same as that of the positive control amphotericin B (Table 6). But compound 1 showed no activity against the tested fungal strains at 64 µg/mL. The preliminary structure-activity relationship analysis revealed that the hemiketal moiety at C-5 (compound 1) greatly strengthened the antibacterial ability, but appears to dramatically reduced the antifungal activity.

Declaration of Competing Interest The authors declared that there is no conflict of interest. Acknowledgements This work was funded by the National Natural Science Foundation of China (NSFC grant no. 31870328). B.-G.W. appreciates the support of Taishan Scholar Program from Shandong Province of China (ts201511060). Appendix A. Supplementary material Supplementary data to this article can be found online at https:// doi.org/10.1016/j.bioorg.2019.103448.

4. Conclusion

References

In summary, eight new fungal polyketides (1–8) as well as six known analogues (9–14) were isolated and identified from the culture extract of T. koningiopsis QA-3, an endophytic fungus obtained from the inner root tissue of A. argyi that was collected from Qichun in China. These compounds belong to bicyclic polyketides, with compound 1 contained a hemiketal moiety at C-5, and compounds 2–14 possessed keto and double bond at C-1 and C-5, respectively. The structures and

[1] R.W. Dunlop, A. Simon, K. Sivasithamparam, E.L. Ghisalberti, J. Nat. Prod. 52 (1989) 67–74. [2] H.G. Cutler, D.S. Himmelsbach, B. Yagen, R.F. Arrendale, J.M. Jacyno, P.D. Cole, R.H. Cox, J. Agric. Food Chem. 39 (1991) 977–980. [3] S.R. Parker, H.G. Cutler, P.R. Schreiner, Biosci. Biotechnol. Biochem. 59 (1995) 1747–1749. [4] H.G. Cutler, S.J. Cutler, S.A. Ross, K.E. Sayed, F.M. Dugan, M.G. Bartlett, A.A. Hill, R.A. Hill, S.R. Parker, J. Nat. Prod. 62 (1999) 137–139.

8

Bioorganic Chemistry xxx (xxxx) xxxx

X.-S. Shi, et al. [5] F. Almassi, E.L. Ghisalberti, M.J. Narbey, K. Sivasithamparam, J. Nat. Prod. 54 (1991) 396–402. [6] E.L. Ghisalberti, C.Y. Rowland, J. Nat. Prod. 56 (1993) 1799–1804. [7] S.C. Dubey, A. Tripathi, P. Dureja, A. Grover, Indian. J. Agr. Sci. 81 (2011) 455–461. [8] K. Liu, Y.B. Yang, C.P. Miao, Y.K. Zheng, J.L. Chen, Y.W. Chen, L.H. Xu, H.L. Guang, Z.T. Ding, L.X. Zhao, Planta Med. 82 (2016) 371–376. [9] G. Liu, Z.Q. Wang, Chem. Commun. (1999) 1129–1130. [10] G. Liu, Z.Q. Wang, Synthesis 1 (2001) 119–127. [11] F.H. Song, H.Q. Dai, Y.J. Tong, B. Ren, C.X. Chen, N. Sun, X.Y. Liu, J. Bian, M. Liu, H. Gao, H.W. Liu, X.P. Chen, L.X. Zhang, J. Nat. Prod. 73 (2010) 806–810. [12] X.X. Zhou, J. Li, Y.H. Yang, Y.Z. Zeng, J. Pei, Phytochem. Lett. 8 (2014) 137–140. [13] K. Liu, Y.B. Yang, J.L. Chen, C.P. Miao, Q. Wang, H. Zhou, Y.W. Chen, Y.Q. Li, Z.T. Ding, L.X. Zhao, Nat. Prod. Bioprospect. 6 (2016) 49–55. [14] A.H. Tarawneh, F. Leon, M.M. Radwan, L.H. Rosa, S.J. Cutler, Nat. Prod. Commun. 8 (2013) 1285–1288. [15] X.S. Shi, D.J. Wang, X.M. Li, H.L. Li, L.H. Meng, X. Li, Y. Pi, X.W. Zhou, B.G. Wang, RSC Adv. 7 (2017) 51335–51342. [16] S. Wang, X.M. Li, F. Teuscher, D.L. Li, A. Diesel, R. Ebel, P. Proksch, B.G. Wang, J. Nat. Prod. 69 (2006) 1622–1625.

[17] Crystallographic data of compounds 1, 2, 9, 10, and 12 have been deposited in the Cambridge Crystallographic Data Centre as CCDC 1860010 (for 1), CCDC 1860034 (for 2), CCDC 1860041 (for 9), CCDC 1860042 (for 10), and CCDC CCDC 1860044 (for 12). The data can be obtained free of charge via http://www.ccdc.cam.ac.uk/ data_request/cif (or from the CCDC, 12 Union Road, Cambridge CB21EZ, U.K.; fax: +44-1223-336-033; e-mail: deposit@c from the CCDC, 12 Union Road,cdc.cam. ac. uk). [18] G.M. Sheldrick, SADABS, Software for Empirical Absorption Correction, University of Gottingen, Germany, 1996. [19] G.M. Sheldrick, SHELXTL, Structure Determination Software Programs, Bruker Analytical X-ray System Inc., Madison, WI, 1997. [20] G.M. Sheldrick, SHELXL-97 and SHELXS-97, Program for X-ray Crystal Structure Solution and Refinement, University of Gottingen, Germany, 1997. [21] M.J. Frisch, G.W. Trucks, H.B. Schlegel, Gaussian 09, Revision D.01, Gaussian, Inc., Wallingford CT, 2013. [22] C.G. Pierce, P. Uppuluri, A.R. Tristan, F.L. Wormley Jr, E. Mowat, G. Ramage, J.L. Lopez-Ribot, Nat. Protoc. 3 (2008) 1494–1500. [23] B.N. Su, E.J. Park, Z.H. Mlbwambo, B.D. Santarsiero, A.D. Mesecar, H.H.S. Fong, J.S. Pezzuto, A.D. Kinghorn, J. Nat. Prod. 65 (2002) 1278–1282.

9