3,5-Dimethylorsellinic Acid Derived Meroterpenoids from Penicillium

Sep 29, 2017 - Eight new chrysogenolides (A–H (1–8)) and seven known (9–15) 3,5-dimethylorsellinic acid derived meroterpenoids were isolated fro...
0 downloads 0 Views 4MB Size
Download PDF
Article Cite This: J. Nat. Prod. 2017, 80, 2699-2707

pubs.acs.org/jnp

3,5-Dimethylorsellinic Acid Derived Meroterpenoids from Penicillium chrysogenum MT-12, an Endophytic Fungus Isolated from Huperzia serrata Bowen Qi,† Xiao Liu,† Ting Mo,† Zhixiang Zhu,† Jun Li,† Juan Wang,† Xiaoping Shi,† Kewu Zeng,‡ Xiaohui Wang,† Pengfei Tu,† Ikuro Abe,§ and Shepo Shi*,† †

Modern Research Center for Traditional Chinese Medicine, School of Chinese Materia Medica, Beijing University of Chinese Medicine, Beijing 100029, People’s Republic of China ‡ State Key Laboratory of Natural and Biomimetic Drugs, Peking University, Beijing 100191, People’s Republic of China § Graduate School of Pharmaceutical Sciences, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan S Supporting Information *

ABSTRACT: Eight new chrysogenolides (A−H (1−8)) and seven known (9−15) 3,5-dimethylorsellinic acid derived meroterpenoids were isolated from the solid substrate fermentation cultures of a Huperzia serrata endophytic fungus, Penicillium chrysogenum MT-12. The structures of the new compounds were elucidated by interpretation of spectroscopic and spectrometric data (1D and 2D NMR, IR, and HRESIMS). The absolute configurations of 1−4 were determined by single-crystal X-ray crystallographic analysis, and those of 5−8 were assigned on the basis of experimental and calculated electronic circular dichroism spectra. Compounds 3, 4, 6, 11, and 12 showed inhibition of nitric oxide production in lipopolysaccharide-activated RAW 264.7 macrophage cells with IC50 values in the range of 4.3−78.2 μM (positive control, indomethacin, IC50 = 33.6 ± 1.4 μM).

I

cells has become a common approach for evaluating the potential anti-inflammatory activities of compounds.8−10 3,5-Dimethylorsellinic acid derived meroterpenoids comprise a large family of natural products that are biogenetically formed via a common intermediate produced by the hybridization of a polyketide intermediate, 3,5-dimethylorsellinic acid, and the terpenoid precursor farnesyl diphosphate (FPP).11−13 Undergoing a sequential cyclization and thereafter followed by complex oxidative ring cleavage, migration, and recyclization, the common intermediate generates a vast array of structurally diverse and chemically complex meroterpenoids, displaying a broad spectrum of biological activities.14−18 Accordingly, meroterpenoids have drawn great interest, and a number of structurally intriguing and biologically important meroterpenoids have been isolated from the genera Aspergillus14−16 and Penicillium.17−19 As an ongoing search for potential antiinflammatory natural products, the metabolites produced by

nflammation is a complex biological response of a host to injury or bacterial infection, contributing to the elimination of harmful factors and the healing process of damaged tissues. However, inappropriate inflammatory responses may cause the excessive secretion of varied inflammatory factors such as proinflammatory cytokines and chemokines, and these inflammatory factors may induce the up-regulated expression of inducible nitric oxide synthase (iNOS), consequently resulting in the sustained synthesis of nitric oxide (NO) in host cells.1−3 NO is recognized as an important mediator and regulator of inflammatory responses, and the overproduction of NO may cause diverse inflammatory disorders, playing a crucial role in the pathogenesis of asthmatic, arthritic, cardiovascular, and neurodegenerative diseases.4−6 iNOS inhibitors, e.g., 2substituted 1,2-dihydro-4-quinazolinamine (AR-C102222), could suppress the production of NO in acute and chronic models of inflammatory arthritis and thus completely abolish the inflammatory development in mice.7 Therefore, iNOS is a promising target for anti-inflammatory therapeutics, and monitoring the NO level in lipopolysaccharide (LPS)-activated © 2017 American Chemical Society and American Society of Pharmacognosy

Received: May 19, 2017 Published: September 29, 2017 2699

DOI: 10.1021/acs.jnatprod.7b00438 J. Nat. Prod. 2017, 80, 2699−2707

Journal of Natural Products

Article

(1H, dd, J = 12.0, 1.5 Hz), 4.71 (1H, q, J = 7.0 Hz), 4.65 (1H, d, J = 3.0 Hz), 3.20 (1H, d, J = 6.0 Hz), 6.30 (1H, d, J = 6.0 Hz)], and two olefinic protons [δH 6.05 (1H, s), 6.06 (1H, d, J = 3.0 Hz)]. The 13C NMR data (Table 2) of 1 exhibited 25 carbon resonances comprising five methyl (δC 14.6, 19.2, 25.9, 26.4, 26.5), two methlene (δC 28.6, 56.6), five sp3 methine (δC 37.6, 53.2, 84.8, 90.3, 109.8), two sp2 methine (δC 115.4, 124.5), three sp3 quaternary (δC 43.8, 47.8, 68.1), two sp3 oxygenated tertiary (δC 58.4, 82.7), one ketal (δC 121.3), two disubstituted olefinic (δC 154.1, 134.3), and three ester carbonyl (δC 163.0, 174.3, 177.9) carbons. Given that the two double bonds and three ester carbonyl groups previously assigned in 1 account for five IHD, the occurrence of eight rings in 1 was consequently confirmed, which was proposed to satisfy the 13 IHD required by the molecular formula C25H26O9. The aforementioned data suggested that compound 1 is a polyketide−terpenoid hybrid meroterpenoid. Comparison of the 1H and 13C NMR spectroscopic data of 1 with those of the known compound purpurogenolide D,20 a 3,5dimethylorsellinic acid derived meroterpenoid previously isolated from P. purpurogenum MHz 111, revealed the absence of a methoxy group and the occurrence of one more ring in 1. In addition, significant differences of the signals due to the 3,5dimethylorsellinic acid moiety were observed between 1 and purpurogenolide D, suggesting that 3,5-dimethylorsellinic acid was incorporated into the meroterpenoid backbone in a different mode. The significantly deshielded chemical shifts of H-9 (δH 4.71; ΔδH +1.36) and C-9 (δC 84.8; ΔδC +14.2) indicated that the C-9 hydroxy group was acylated and culminated in the formation of a γ-lactone ring between C-9 and C-20. The HMBC correlations between H-23 and C-10/C11 indicated the formation of an unprecedented five-membered ether ring between C-10 and C-23. The planar structure of 1 was further confirmed by HMBC correlations between H-2 and C-1/C-4/C-15; H-5 and C-4/C-6/C-7/C-12/C-13; H-13 and C-5/C-15; H-14 and C-3/C-12; H3-17 and C-18/C-16; H3-18 and C-15/C-17; H-22 and C-8/C-11/C-12/C-20; H-23 and C8/C-10/C-11; H3-24 and C-6/C-7/C-8/C-22; and H2-25 and C-3/C-5 (Figure 1). The relative configuration of 1 was determined as shown in Figure 1 on the basis of NOESY spectrum, which showed NOE correlations of H-6α/H3-19, H319/H-13, H-13/H3-21; H-9/H-23, H-22/H-23, H-22/H3-24; and H-5/H-6β, H-6β/H-25β. The absolute configuration of 1 was unequivocally defined as (4R, 5R, 7R, 9R, 10R, 11S, 12S, 13S, 22R, 23R) on the basis of single-crystal X-ray diffraction analysis (Figure 2) using Cu Kα radiation with Flack and Hooft parameters of 0.10(9) and 0.11(9), respectively. Chrysogenolide B (2) was obtained as colorless crystals (in MeOH), with a molecular formula of C26H30O10, which was elucidated on the basis of the positive-ion HRESIMS and 13C NMR data, indicating 12 IHD. The 1H and 13C spectroscopic data (Tables 1 and 2) are comparable to those of berkeleyacetal B,21 a 3,5-dimethylorsellinic acid derived meroterpenoid previously isolated from P. rubrum, except for the significantly upfield carbon chemical shifts C-14 (δC 59.5; ΔδC −70.7) and C-15 (δC 61.3; ΔδC −76.5), suggesting that the Δ14,15 double bond in berkeleyacetal B is replaced by a trisubstituted epoxide, which was also confirmed by the HMBC correlations between H-14 and C-12/C-13; H2-13 and C-14/C-15; and H-2 and C15 (Figure 1). Determination of the planar structure of 2 was achieved by unambiguous assignments of all protons and carbons by HSQC, 1H−1H COSY, and HMBC experiments. The relative configuration of 2 was determined by the NOE

Penicillium chrysogenum MT-12, an endophytic fungus isolated from Chinese club moss Huperzia serrata, were investigated. The crude EtOAc extract was found to inhibit NO production in LPS-activated RAW 264.7 macrophage cells, and subsequent separation and purification of the crude extract resulted in eight new 3,5-dimethylorsellinic acid derived meroterpenoids, chrysogenolides A−H (1−8), together with seven known ones (9−15). Herein, the isolation and structural elucidation of the isolated compounds as well as an evaluation of their inhibitory effects on NO production in LPS-activated RAW 264.7 cells are described.

■

RESULTS AND DISCUSSION The EtOAc extract of cultures of P. chrysogenum MT-12 was subjected to repeated silica gel, Sephadex LH-20, and RP-C18 gel column chromatography, followed by semipreparative RPHPLC, to afford eight new (1−8) and seven known meroterpenoids (9−15). Chrysogenolide A (1) was obtained as colorless plates via crystallization from MeOH. Its molecular formula was assigned as C25H26O9 by positive-ion HRESIMS and 13C NMR spectroscopic data, indicating 13 indices of hydrogen deficiency (IHD). The IR spectrum showed absorption bands at 1799, 1713, 1690, and 1631 cm−1, indicating the presence of carbonyl and olefinic functionalities. The 1H NMR data (Table 1) of 1 showed the resonances for five methyls [δH 1.32 (3H, s), 1.41 (3H, s), 1.48 (3H, s), 1.57 (3H, d, J = 7.0 Hz), 1.66 (3H, s)], two methlenes [δH 1.59 (1H, m), 1.52 (1H, m), 3.22 (1H, d, J = 5.0 Hz), 2.40 (1H, d, J = 5.0 Hz)], five methines [δH 2.49 2700

DOI: 10.1021/acs.jnatprod.7b00438 J. Nat. Prod. 2017, 80, 2699−2707

Journal of Natural Products

Article

Table 1. 1H NMR (500 MHz) Data of Compounds 1−8 (δ in ppm, J in Hz) in CDCl3 no.

1

2

3

4

5

6

7

8

2a 2b 5

6.05, s

6.38, s

5.89, s

5.97, s

6.01, s

6.12, s

1.52, m

2.05, dd (14.0, 3.5) 2.30, dd (14.0, 3.5) 1.25, m

2.51, dd (12.0, 3.5) 1.71, dd (14.0, 2.0) 1.55, m

2.55, dd (12.0, 4.5) 1.69, m

6b

1.68, dd (8.0, 4.0) 2.33, dd (14.0, 4.0) 1.40, m

2.02, dd (13.0, 3.5) 1.80, dd (14.0, 13.0)

2.46, dd (13.5, 2.5) 1.38, br d (13.5)

9 11 13a

4.71, q (7.0)

4.26, q (7.0)

4.30, q (7.0)

4.36, q (6.5)

1.63, dd (14.5, 12.0) 4.43, q (6.5)

3.10, dd (13.0, 4.5) 1.84, dd (15.0, 4.5) 1.65, m

3.44, d (21.5) 3.32, d (21.5) 2.05, br d (13.0)

6a

2.49, dd (12.0, 1.5) 1.59, m

3.38, d (21.0) 3.25, d (21.0) 1.91, br d (14.0)

4.65, d (3.0)

3.15, dd (16.0, 3.0) 2.40, dd (16.0, 8.0) 3.36, dd (8.0,3.0) 1.35, s 1.41, s 0.91, s 1.36, d (7.0) 3.94, d (6.5)

4.06, dd (13.5, 3.0) 2.18, dd (13.5, 8.5) 6.05, dd (8.5,4.0) 1.49, s 1.61, s 0.73, s 1.39, d (7.0) 3.95, d (6.5)

4.96, d (5.5)

4.96, d (2.5)

6.08, d (5.5)

6.13, d (2.5)

6.33, s

1.43, s 1.62, s 1.38, s 1.34, d (6.5)

1.46, s 1.65, s 1.30, s 1.35, d (6.5) 2.91, d (2.5)

1.60, s 1.69, s 1.316, s 1.35, d (7.0)

1.56, s 1.26, s 1.16, s 1.31, s 3.18, d (4.0)

6.08, d (6.5) 1.42, s 2.92, d (5.5) 2.76, d (5.5) 3.88, s

6.09, d (6.5) 1.38, s 1.50, s

5.62, s 1.23, s 3.15, d (5.5) 2.40, d (5.5)

6.00, d (2.5) 1.29, s 3.13, d (5.5) 2.41, d (5.5)

5.74, s 1.317, s 3.24, d (5.0) 2.60, d (5.0)

6.06, d (4.0) 1.43, s 1.71, s

13b

a

14

6.06, d (3.0)

17 18 19 21 22

1.48, s 1.66, s 1.32, s 1.57, d (7.0) 3.20, d (6.0)

23 24 25a 25b 26

6.30, d (6.0) 1.41, s 3.22, d (5.0) 2.40, d (5.0)

4.25, q (7.0) 3.81, s 3.52, dd (14.5, 8.5) 1.83, dd (14.5, 4.5) 6.17, dd (8.5, 4.5)

3.84, s

2.52, d (10.0) 2.51, dd (13.0, 8.5) 1.75, dd (13.0, 5.0) 6.38, dd (8.5, 5.0) 1.58, s 1.39, s 1.02, s 1.63, s 3.13,dd (10.0, 7.0) 6.06, d (7.0) 1.42, s 1.80, s 3.71, s

All assignments are based on HSQC, HMBC, and NOE experiments.

Table 2. 13C NMR (125 MHz) Data of Compounds 1−8a (δ in ppm) in CDCl3

a

no.

1

2

3

4

5

6

7

8

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26

163.0, C 115.4, CH 154.1, C 58.4, C 37.6, CH 28.6, CH2 43.8, C 177.9, C 84.8, CH 121.3, C 68.1, C 47.8, C 90.3, CH 124.5, CH 134.3, C 82.7, C 25.9, CH3 26.5, CH3 19.2, CH3 174.3, C 14.6, CH3 53.2, CH 109.8, CH 26.4, CH3 56.6, CH2

162.6, C 123.2, CH 153.3, C 56.5, C 43.8, CH 27.8, CH2 43.7, C 176.6, C 73.0, CH 208.1, C 63.2, C 39.6, C 34.5, CH2 59.5, CH 61.3, C 81.4, C 26.2, CH3 23.7, CH3 18.4, CH3 167.7, C 18.5, CH3 49.6, CH 97.8, CH 26.2, CH3 58.2, CH2 53.5, CH3

164.5, C 115.6, CH 157.2, C 76.8, C 50.4, CH 30.2, CH2 44.3, C 177.2, C 73.2, CH 208.6, C 63.6, C 43.1, C 37.0, CH2 131.2, CH 134.9, C 83.9, C 27.5, CH3 26.6, CH3 17.9, CH3 168.3, C 18.5, CH3 49.8, CH 98.1, CH 28.3, CH3 27.9, CH3 53.2, CH3

163.4, C 115.5, CH 155.3, C 57.8, C 37.6, CH 25.5, CH2 49.1, C 176.3, C 63.6, CH 154.4, C 107.5, C 47.4, C 90.9, CH 127.6, CH 133.4, C 82.8, C 26.0, CH3 25.6, CH3 21.4, CH3

163.3, C 115.7, CH 155.2, C 57.9, C 37.4, CH 27.6, CH2 46.6, C 177.6, C 63.6, CH 150.3, C 105.2, C 47.7, C 90.4, CH 128.0, CH 133.4, C 82.8, C 26.1, CH3 25.7, CH3 19.5, CH3

162.8, C 117.2, CH 153.2, C 58.9, C 32.6, CH 26.2, CH2 46.9, C 174.5, C 74.3, CH 209.9, C 60.5, CH 48.3, C 205.6, C 130.1, CH 141.3, C 82.9, C 26.6, CH3 26.0, CH3 18.9, CH3

17.4, CH3 72.7, C 102.3, CH 20.0, CH3 55.7, CH2

17.6, 40.8, 97.8, 24.2, 55.8,

17.0, CH3 80.9, C 103.7, CH 22.0, CH3 55.5, CH2

169.9, C 34.6, CH2 126.8, C 134.2, C 43.6, CH 37.5, CH2 44.2, C 206.2, C 87.3, C 198.7, C 61.1, C 58.9, C 34.4, CH2 129.2, CH 139.9, C 82.8, C 27.0, CH3 29.6, CH3 20.0, CH3 167.4, C 11.2, CH3 46.7, CH 97.0, CH 22.7, CH3 15.3, CH3

170.4, C 34.4, CH2 126.2, C 136.4, C 42.1, CH 30.3, CH2 45.7, C 207.9, C 79.7, C 170.2, C 50.3, CH 54.2, C 38.1, CH2 128.7, CH 138.9, C 82.9, C 26.4, CH3 29.4, CH3 25.2, CH3 170.5, C 25.3, CH3 43.3, CH 96.8, CH 31.1, CH3 15.4, CH3 53.3, CH3

CH3 CH CH CH3 CH2

All assignments are based on HSQC, HMBC, and NOE experiments.

correlations of H-5/H-6β, H-5/H-25β, H-6α/H3-19, H-5/H13β, H-14/H-13α, H-13α/H3-19, H3-19/H-22; and H-22/H-

23, H-23/H3-24, and H-23/H3-21 in the NOESY spectrum (Figure 1). The absolute configuration of 2 was unambiguously 2701

DOI: 10.1021/acs.jnatprod.7b00438 J. Nat. Prod. 2017, 80, 2699−2707

Journal of Natural Products

Article

oriented). Thus, the absolute configuration of 3 was unequivocally elucidated as (4R, 5R, 7R, 9R, 11S, 12S, 22R, 23R) and finally confirmed by single-crystal X-ray diffraction experiment (Figure 4). Chrysogenolide D (4) was obtained as colorless crystals (in MeOH). The positive-ion HRESIMS gave a [M + H]+ ion at m/z 443.1696, consistent with an empirical molecular formula of C24H26O8, indicating 12 IHD. Comparison of the NMR data of 4 with those of paraherquonin22 revealed that the C-25 methyl in paraherquonin was replaced by an epoxide ring at C4 in 4, which was supported by the deshielded chemical shifts of C-4 (δC 57.8; ΔδC +25.8) and C-25 (δC 55.7; ΔδC +35.5) as well as the HMBC correlations between H-2 and C-1/C-3/C15; H-5 and C-4/C-6/C-12/C-13; and H2-25 and C-3/C-5 (Figure 3). The relative stereochemistry of 4 for C-9 was established by NOE difference spectroscopy (Figure S28, Supporting Information). Irradiation of H3-21 resulted in enhanced signals for both H-13 and H-23, suggesting that the orientation of C-21 methyl in 4 (α-oriented) was different from that in paraherquonin (β-oriented). The absolute configuration of 4 was unambiguously defined as (4R, 5R, 7R, 9R, 12S, 13S, 22R, 23R) by single-crystal X-ray diffraction analysis (Figure 4). Chrysogenolide E (5) was obtained as a white powder, with a molecular formula of C24H26O7 determined by HRESIMS and 13 C NMR spectroscopic data. The 1H and 13C spectroscopic data resembled those of 4, except for the significantly upfield chemical shift of C-22 (δC 40.8, ΔδC −31.9), suggesting the absence of a hydroxy group at C-22 in 5. The deduction was confirmed by the presence of a proton resonance at δH 2.91 (1H, d, J = 2.5 Hz) and the 1H−1H COSY correlation of H-22/ H-23 as well as the HMBC correlations between H-22 and C-8, C-10, C-11, C-12, and C-23 (Figure 3). Analysis of the NOESY spectrum of 5 revealed that the relative configuration of 5 was completely the same as that of 4. To determine the absolute configuration of 5, the electronic circular dichroism (ECD) spectra of (4R,5R,7R,9R,12S,13S,22S,23R)-5 and its enantiomer (4S,5S,7S,9S,12R,13R,22R,23S)-5 were calculated using timedependent density functional theory (TDDFT) at the B3LYP/ 6-31G level with the CPCM model. As shown in Figure 6, the calculated spectrum of (4R,5R,7R,9R,12S,13S,22S,23R)-5 was in good agreement with the experimental spectrum. Therefore, the absolute configuration of 5 was defined as (4R, 5R, 7R, 9R, 12S, 13S, 22S, 23R).

Figure 1. Selected HMBC and NOE correlations of compounds 1 and 2.

elucidated as (4R, 5R, 7R, 9R, 11S, 12S, 14R, 15S, 22R, 23R) on the basis of single-crystal X-ray diffraction analysis (Figure 2) using Cu Kα radiation with Flack and Hooft parameters of −0.2(2) and −0.14(19), respectively. Chrysogenolide C (3) was obtained as colorless crystals (in MeOH). Its molecular formula was identified as C26H32O9 by positive-ion HRESIMS and 13C NMR data (Table 2), indicating 11 IHD. Comparison of the NMR data of 3 with those of the known compound berkeleyacetal B21 revealed that signals due to the C-4 oxirane moiety in berkeleyacetal B were replaced by resonances contributing to a methyl group [δH 1.50 (3H, s), δC 27.9] and an oxygenated tertiary carbon (δC 76.8), suggesting the epoxide ring in berkeleyacetal B was opened in 3, which readily decipher the 1 IHD reduction of 3. The complete determination of the planar structure and relative configuration of 3 was achieved by various 2D NMR experiments including HSQC, 1H−1H COSY, HMBC, and NOESY. Notably, the relative configuration of the C-21 methyl in 3 (α-oriented) was different from that in berkeleyacetal B (β-

Figure 2. ORTEP drawing of compounds 1 and 2. 2702

DOI: 10.1021/acs.jnatprod.7b00438 J. Nat. Prod. 2017, 80, 2699−2707

Journal of Natural Products

Article

Figure 3. Selected HMBC correlations of compounds 3−8.

Figure 4. ORTEP drawing of compounds 3 and 4.

Figure 5. Selected NOESY correlations of compounds 3−8. 2703

DOI: 10.1021/acs.jnatprod.7b00438 J. Nat. Prod. 2017, 80, 2699−2707

Journal of Natural Products

Article

Figure 6. Experimental and calculated ECD spectra of compounds 5−8 (in MeOH).

3.18 (1H, d, J = 4.0 Hz)], one oxygenated methine [δH 6.06 (1H, d, J = 4.0 Hz)], and one olefinic proton [δH 6.17 (1H, dd, J = 8.5, 4.5 Hz)]. The 13C NMR data of 7 (Table 2) exhibited the presence of six methyl [δC 11.2 (C-21), 15.3 (C-25), 20.0 (C-19), 22.7 (C-24), 27.0 (C-17), 29.6 (C-18) ], three sp3 methylene [δC 34.4 (C-13), 34.6 (C-2), 37.5 (C-6)], two sp3 methine [δC 43.6 (C-5), 46.7 (C-22)], one oxygenated sp3 methine [δC 97.0 (C-23)], one sp2 methine [δC 129.2 (C-14)], three sp3 quaternary [δC 44.2 (C-7), 58.9 (C-12), 61.1 (C-11)], two oxygenated sp3 tertiary [δC 82.8 (C-16), 87.3 (C-9)], three disubstituted olefinic [δC 129.2 (C-3), 134.2 (C-4), 139.9 (C15)], two ester carbonyl [δC 167.4 (C-20), 169.9 (C-1)], and two ketocarbonyl [δC 198.7 (C-10), 206.2 (C-7)] carbons. The four carbonyl carbons together with the two double bonds aforementioned in 7 contribute to 6 IHD, suggesting the occurrence of six rings in 7 to satisfy the 12 IHD required by the molecular formula of C25H28O7. Comparison of the NMR data with those of berkeleydione23 revealed that the Δ22,23 double bond and methoxy group were absent in 7, which were supported by the significant upfield chemical shifts of C-22 (δC 46.7; ΔδC −98.6) and C-23 (δC 97.0; ΔδC −16.1) as well as the 1 H−1H COSY correlation between H-22 and H-23. In the HMBC spectrum of 7 (Figure 3), the correlations between H23 and C-11, C-20, and C-22 revealed the formation of a γlactone ring between C-20 and C-23. In addition, the HMBC

Chrysogenolide F (6) was obtained as a white powder, with a molecular formula of C24H26O9 deduced by the HRESIMS data and 13C NMR spectroscopic data, indicating 12 IHD. The NMR data of 6 were similar to those of berkeleyacetal C,21 except for the significantly deshielded chemical shift of C-22 (δC 80.9, ΔδC +35.1), indicating the presence of a hydroxyl group at C-22 in 6. The deduction was confirmed by the absence of a proton signal due to the resonance of H-22 in the 1 H NMR spectrum of berkeleyacetal C. The planar structure of 6 was established by 2D NMR experiments including HSQC, 1 H−1H COSY, and HMBC. The relative configuration of 6 was determined by NOE correlations of H-6β/H3-19, H-6β/H3-24, H-11/H3-21, H-11/H3-19, H-23/H3-24, H-5/H-6α, and H-5/ H-25α, H-6α/H-25α (Figure 5). The experimental and calculated ECD spectra of (4S,5S,7S,9S,11R,12R,22S,23S)-6 are superimposed. Accordingly, the absolute configuration of 6 was assigned as shown in Figure 6. Chrysogenolide G (7) was obtained as a white powder. Its molecular formula was assigned as C25H28O7 by HRESIMS. The 1H NMR data of 7 (Table 1) showed the resonances for six methyls (δH 1.16, 1.26, 1.31, 1.43, 1.56, 1.71, each 3H, s), three methylenes [δH 3.52 (1H, dd, J = 14.5, 8.5 Hz), 1.83 (1H, dd, J = 14.5, 4.5); 3.38 (1H, d, J = 21.0 Hz), 3.25 (1H, d, J = 21.0 Hz); 2.02 (1H, dd, J = 13.0, 3.5 Hz), 1.80 (1H, dd, J = 14.0, 13.0 Hz)], two methines [δH 1.91 (1H, br d, J = 14.5 Hz); 2704

DOI: 10.1021/acs.jnatprod.7b00438 J. Nat. Prod. 2017, 80, 2699−2707

Journal of Natural Products

Article

2, 5, 7−10, and 13−15 were inactive (