Skip to main content
ACS Omega logoLink to ACS Omega
. 2021 Feb 16;6(8):5898–5909. doi: 10.1021/acsomega.0c06349

Linderaggrenolides A–N, Oxygen-Conjugated Sesquiterpenoid Dimers from the Roots of Lindera aggregata

Xin Liu , Jing Fu , Ji Yang , Ancheng C Huang , Run-Feng Li , Li-Ping Bai , Liang Liu , Zhi-Hong Jiang †,*, Guo-Yuan Zhu †,*
PMCID: PMC7931382  PMID: 33681628

Abstract

graphic file with name ao0c06349_0008.jpg

Linderaggrenolides A–N (1–14), 14 new lindenane sesquiterpenoid dimers with oxygen bridges were isolated from the roots of Lindera aggregata. Their structures were elucidated on the basis of comprehensive spectroscopic data analysis, with the absolute configurations established by empirical approaches, electronic circular dichroism calculations, and X-ray crystallography. Compounds 8 and 9 were found to exhibit significant transforming growth factor-β inhibitory activity, with IC50 values of 25.91 and 21.52 μM, respectively.

1. Introduction

Sesquiterpenoid dimers, plausibly biosynthesized via the coupling of two identical or different sesquiterpenoid molecules, are a class of naturally occurring metabolites with C30 cores and possess a variety of biological activities including anti-inflammatory, antimalarial, antitumor, and anti-HIV activities.1 These metabolites can be found in some higher plant families including the Chloranthaceae, particularly the genera Artemisia and Ligularia.1 The genus Lindera belongs to the family Lauraceae, comprising approximately 100 species. Among them, Lindera aggregata has been used for thousands of years as a traditional Chinese medicine for the treatment of diseases such as rheumatism, chest and abdominal pain, regurgitation, and frequent urination.2 Phytochemical investigations into this species have revealed the presence of miscellaneous types of natural products such as sesquiterpenoids,3,4 alkaloids,5 flavonoids,6 lignans,7 condensed tannins,8 and cyclopentenediones.9 Our previous phytochemical studies on this plant lead to the isolation of two methine- or methylene-bridged sesquiterpenoid trimers possessing a unique C46 skeleton, four carbon-bridged disesquiterpenoids with a C33 or C31 skeleton, and four disesquiterpenoid–geranylbenzofuranone conjugates.10,11 The preliminary results showed that the ethanol extract from the roots of L. aggregata exhibited the transforming growth factor (TGF)-β inhibitory activity. Motivated by these findings, we performed a thorough phytochemical investigation into the ethanol extract of the roots of L. aggregata, resulting in the isolation and identification of 14 new lindenane sesquiterpenoid dimers (1–14) featuring an oxygen bridge (Figure 1). Herein, we presented the isolation, structural characterization, and potential bioactivities of these new sesquiterpenoid dimers.

Figure 1.

Figure 1

Chemical structures of compounds 114.

2. Results and Discussion

Linderaggrenolide A (1) was obtained as colorless crystals. Its molecular formula was established as C31H38O8 based on the 13C NMR data (Table 2) and high-resolution electrospray ionization mass spectrometry (HRESIMS) at m/z 556.2931 ([M + NH4]+, calcd for 556.2905), indicating 13 indices of hydrogen deficiency. The IR spectrum revealed the presence of hydroxyl (3401 cm–1), carbonyl (1774 cm–1), and olefinic (1641 cm–1) functional groups. The 1H NMR data of 1 (Table 1) displayed the typical resonances for four singlet methyl groups (δH 0.55, 0.80, 1.69, and 2.02), three oxygenated methine protons (δH 4.31, 4.29, and 4.92), a methoxy group (δH 3.28), and four olefinic protons (δH 5.09, 5.09, 5.26, and 5.28). Its 13C NMR and distortionless enhancement by polarization transfer (DEPT) data (Table 2) showed 31 carbon signals corresponding to an ester carbonyl, six olefinic carbons, six quaternary carbons (four oxygenated), nine methines (two oxymethines), four methylenes, a methoxy, and four methyl groups.

Table 2. 13C NMR (150 MHz) Spectroscopic Data of 17 in CDCl3 (δ in ppm).

position 1 2 3 4 5 6 7
1 29.5 29.5 28.0 29.5 28.8 29.4 28.0
2 17.2 16.9 16.9 16.9 16.9 17.0 16.7
3 23.7 24.2 23.5 23.8 23.8 23.7 23.7
4 149.1 149.1 148.2 148.9 148.4 148.9 147.5
5 62.9 62.5 70.6 60.7 60.4 60.0 67.1
6 63.2 63.2 68.4 63.4 63.5 63.0 69.6
7 157.2 156.4 157.2 153.3 153.7 152.7 155.3
8 107.3 107.7 108.5 107.1 106.2 107.5 108.2
9 49.9 49.9 47.6 51.2 50.6 47.6 47.3
10 38.3 38.1 37.9 37.5 37.5 38.0 38.6
11 132.6 133.0 128.0 134.0 132.8 132.1 125.7
12 171.1 170.8 170.8 171.2 171.0 171.3 170.5
13 9.0 8.9 9.7 9.9 9.6 9.5 8.8
14 22.1 22.2 18.6 21.3 20.9 22.0 18.6
15 108.7 108.7 109.5 107.8 108.0 107.7 109.3
1′ 28.9 29.9 30.1 29.4 29.2 29.7 29.5
2′ 16.8 17.1 17.0 16.7 16.5 16.7 16.9
3′ 23.7 23.8 23.8 23.7 23.5 23.6 23.7
4′ 148.4 148.9 149.4 149.8 149.2 149.9 149.9
5′ 67.3 66.3 65.9 60.3 60.9 60.7 60.2
6′ 64.2 68.4 69.3 63.5 62.8 63.4 63.5
7′ 71.6 71.1 71.1 140.3 140.8 140.9 139.5
8′ 108.7 109.5 107.0 113.9 117.4 115.6 114.3
9′ 46.8 38.5 39.4 52.7 47.9 52.6 52.3
10′ 38.1 38.9 38.9 38.1 37.9 37.9 38.1
11′ 66.4 64.3 63.9 132.5 131.3 133.1 133.7
12′ 96.2 99.2 101.1 100.6 100.7 104.5 102.8
13′ 11.7 11.0 12.1 10.5 10.6 10.4 10.7
14′ 17.7 22.1 21.9 21.6 20.7 21.0 21.5
15′ 109.7 107.3 107.0 106.9 107.2 104.5 106.9
  8′-OCH3 8′-OCH2CH3 8′-OCH3 6-OAc 6-OAc 6-OAc 6-OAc
  51.6 57.9, 15.4 49.4 171.1 170.2 170.3 170.6
        20.9 21.3 23.6 20.8
        6′-OAc 6′-OAc 6′-OAc 6′-OAc
        170.5 170.8 170.5 170.6
        21.2 21.2 21.2 21.2

Table 1. 1H NMR (600 MHz) Spectroscopic Data of 17 in CDCl3 (δ in ppm, J in Hz).

position 1 2 3 4 5 6 7
1 1.42 m 1.40 m 1.34 m 1.28 m 1.27 m 1.48 m 1.37 m
2 0.86 m, 0.76 m 0.85 m, 0.76 m 0.84 m, 0.78 m 0.85 m, 0.76 m 0.86 m, 0.70 m 0.84 m, 0.71 m 1.02 m, 0.83 m
3 2.02 overlapped 1.98 m 2.00 m 2.00 m 1.96 m 1.97 overlapped 2.00 m
5 3.33 dt (11.0, 2.5) 3.34 overlapped 2.75 brd (11.6) 3.47 overlapped 3.38 d (11.2) 3.57 brd (10.8) 2.89 brd (12.4)
6 4.31 d (11.0) 4.32 t (11.6) 4.61 brd (11.6) 5.55 (10.6) 5.78 d (11.2) 5.79 d (10.8) 5.54 dd (12.4, 1.4)
9 2.59 d (14.4) 2.61 d (14.2) 2.67 d (13.6) 2.76 (14.4) 2.61 d (14.5) 3.03 d (14.3) 2.75 d (13.6)
  2.22 d (14.4) 2.19 d (14.2) 1.83 d (13.6) 2.47 (14.4) 2.37 d (14.5) 2.15 d (14.3) 1.87 d (13.6)
13 2.02 s 2.03 s 2.14 d (1.6) 2.08 s 2/04 s 1.98 s 1.93 d (1.4)
14 0.55 s 0.57 s 0.89 s 0.56 s 0.58 s 0.59 s 0.95 s
15 5.26 brs 5.25 brs 5.20 brs 5.04 brs 5.02 brs 5.03 brs 5.08 brs
  5.09 overlapped 5.09 brs 5.14 brs 4.77 brs 4.72 brs 4.72 brs 4.83 brs
1′ 1.28 m 1.31 m 1.34 m 1.37 m 1.38 m 1.35 m 1.26 m
2′ 1.02 m, 0.86 m 0.86 m, 0.76 m 1.01 m, 0.86 m 0.78 m, 0.77 m 0.79 m, 0.76 m 0.85 m, 0.78 m 0.76 m, 0.69 m
3′ 2.02 m 2.00 m 2.00 m 1.94 m 2.01 m 2.00 overlapped 2.00 overlapped
5′ 2.88 brd (11.5) 3.31 overlapped 3.39 brd (12.0) 3.44 overlapped 3.33 brd (10.6) 3.42 brd (10.5) 3.43 brd (12.0)
6′ 4.29 d (11.5) 3.84 t (9.8) 3.81 brd (12.0) 5.62 d (10.6) 5.67 d (10.6) 5.64 d (10.5) 5.61 d (12.0)
9′ 2.45 d (13.4) 2.47 d (14.6) 2.36 d (14.0) 2.28 d (13.6) 2.12 d (14.5) 2.37 d (13.6) 2.36 d (13.7)
  2.00 d (13.4) 1.73 d (14.6) 1.70 d (14.0) 1.97 d (13.6) 1.99 d (14.5) 2.32 d (13.6) 2.04 d (13.7)
12′ 4.92 s 4.86 s 5.09 s 5.15 s 5.41 s 5.98 s 5.48 s
13′ 1.69 s 1.68 s 1.62 s 1.86 s 1.89 s 1.83 s 1.84 s
14′ 0.80 s 0.83 s 0.82 s 0.53 s 0.56 s 0.66 s 0.53 s
15′ 5.28 brs 5.07 brs 5.05 brs 4.96 brs 4.99 brs 4.97 brs 4.95 brs
  5.09 overlapped 5.05 brs 5.03 brs 4.69 brs 4.72 brs 4.71 brs 4.68 brs
  3.28 s (8′-OCH3) 3.77 m, 3.62 m 3.43 (8′-OCH3) 3.24 s   3.06 s 3.21 s
    1.29 t (7.0) (8′-OCH2CH3)   (8′-OCH3)   (8′-OCH3) (8′-OCH3)
        2.12 s (6-OAc) 2.23 s (6-OAc) 2.00 s (6-OAc) 2.20 s (6-OAc)
        2.08 s (6′-OAc) 2.12 s (6′-OAc) 2.06 s (6′-OAc) 2.06 s (6′-OAc)

Inspection of the 1D and 2D NMR data of 1 [including 1H–1H COSY, heteronuclear single quantum coherence (HSQC), and heteronuclear multiple bond coherence (HMBC)] suggested that compound 1 should be a lindenane sesquiterpenoid dimer. The 1H–1H COSY spectrum showed two sets of proton spin systems of a characteristic 1,2-distributed cyclopropane ring for two lindenane sesquiterpenoids (Figure 2). The HMBC correlations from H-13 to C-7, C-11, and C-12, from H-14 to C-1, C-5, C-9, and C-10, and from the protons of the characteristic exocyclic double bond (H-15) to C-3, C-4, and C-5 indicated a sesquiterpenoid unit in 1 to be strychnistenolide, which was confirmed by comparison with NMR data previously reported in the literature.12 The remaining NMR data for another sesquiterpenoid unit are similar to those of strychnistenolide, except for an epoxy group at C-7′ and C-11′, as indicated by the typical oxygen-bearing carbon resonances at δC 71.6 (C-7′) and 66.4 (C-11′), the HMBC from H-13′ to C-7′, C-11′, and C-12′, and from H-12′ to C-7′ and C-11′, a methoxy group at C-8′ as suggested by HMBC of −OCH3/C-8′, and the absence of the carbonyl group at C-12′. The connections between two sesquiterpenoid units were resolved based on the key HMBC of H-12′/C-8. Hence, the planar structure of 1 was defined as shown.

Figure 2.

Figure 2

Key 1H–1H COSY, HMBC, and NOESY correlations of compounds 1, 8, 10, and 14.

The relative configuration of 1 was assigned by nuclear Overhauser effect spectroscopy (NOESY) correlations, as shown in Figure 2. The NOESY correlations of H-14/H-2, H-14/H-6, H-14′/H-2′, and H-14′/H-6′ indicated that these protons are cofacial and thus assigned as β-oriented according to those of natural lindenanes.12 Further examination of the NOESY cross peaks of H-1/H-3/H-5 and H-1′/H-3′/H-5′ revealed the α-orientation for these two sets of protons. The key NOESY correlations of H-12′/OCH3-8′ and OCH3-8′/H-14′ were observed, suggesting that OCH3-8′ and H-12′ were β-oriented. The resonance of the methyl protons at δH 0.80 (H-14′) was found to be considerably deshielded relative to that of H-14 at δH 0.55 due to the anisotropic effect exhibited by the β–OCH3-8′ and α–O-8 groups, consistent with what was observed in previous studies.12 Recrystallization of 1 from CHCl3–MeOH afforded single crystals suitable for X-ray diffraction (XRD) with Cu Kα radiation. These XRD results confirmed the assigned planar structure and further supported the assigned absolute configuration (Figure 3).

Figure 3.

Figure 3

ORTEP drawing of compound 1.

The configuration of C-8 in natural lindenane- and eudesmane-type sesquiterpene lactones is difficult to be deduced because of the absence of NOE correlations and crystal data. Kouno I. et al. reported the cofacial orientation of C-8 substituent group can be assigned by the chemical shift of the C-14 methyl protons in lindenane- and eudesmane-type sesquiterpene lactones.12 In order to confirm the relationship between the proton chemical shift of 14-CH3 and the configuration of C-8 in natural lindenane- and eudesmane-type sesquiterpene lactones, we reviewed those data of 31 known and 12 new lindenane- or eudesmane-type sesquiterpene lactones (Table 5). As shown in Table 5, it was observed that the α orientation of 8-R(S) in 22 compounds corresponded with the proton chemical shift of 14-CH3 in a range of 0.42–0.67 ppm (medium, 0.56 ppm), while a range of 0.82–1.47 ppm (medium, 1.06 ppm) of 14-CH3 corresponded with the β orientation of 8-S(R). These data supported the approach for discriminating the relative configurations of C8 in lindenane- or eudesmane-type sesquiterpene lactones by their proton chemical shift of 14-CH3.

Table 5. Relationship between the Proton Chemical Shift of 14-CH3 and the Configuration of C-8 in Natural Lindenane- and Eudesmane-Type Sesquiterpene Lactones.

2.

no. compounds skeleton 14-CH3H in ppm) 8-R the orientation of 8-R the configuration of C-8 refs
1 strychnistenolide A A 0.61 (s) OH α R (12)
2 strychnistenolide A acetate A 0.58 (s) OH α R (12)
3 linderolide T A 0.55 (s) OH α R (15)
4 heterogorgiolide A 0.52 (s) OCH3 α R (16)
5 lindenanolide F A 0.42 (s) R α R (13)
6 aggreganoid A A 0.48 (s) R α S (10)
7 decorone A A 0.56 (s) R α S (17)
8 8α-linderagalactone B 0.67 (s) OH α R (4)
9 biepiasterolide B 0.49 (s) R α R (18)
10 biatractylenolide II B 0.51 (s) R α R (19)
11 8α-methoxy-epiasterolid B 0.65 (s) OCH3 α R (20)
12 1 A 0.55 (s) OR α R  
13 2 A 0.57 (s) OR α R  
14 4 A 0.56 (s), OR α R  
      0.53 (s, H-14′)        
15 5 A 0.58 (s), OR α R  
      0.56 (s, H-14′)        
16 6 A 0.59 (s), OR α R  
      0.66 (s, H-14′)        
17 7 A 0.53 (s, H-14′) OR α R  
18 8 A 0.56 (s) OR α R  
19 9 A 0.56 (s) OR α R  
20 10 A 0.56 (s) OR α R  
21 11 A 0.58 (s) OR α R  
22 12 A 0.60 (s) OR α R  
23 strychnistenolide B A 1.23 (d, J = 1.8 Hz) OH β S (12)
24 strychnistenolide B acetate A 0.99 (s) OH β S (12)
25 8-O-β-d-glucopyranoside A 0.93 (s) OR β R (21)
26 atractylenolide III B 1.03 (s) OH β S (22)
27 atractylenoide IV B 1.06 (s) OH β S (22)
28 atractylenother B 1.05 (s) OH β S (23)
29 4α,8β-dihydroxy-5α(H)-eudesm-7(11)-en-8,12-olide B 1.06 (s) OH β S (24)
30 4α-hydroxy-5α(H)-8β-methoxy-eudesm-7(11)-en-8,12-olide. B 1.06 (s) OCH3 β S (24)
31 atractylenolidone B 1.06 (s) OH β S (25)
32 (3R)-3-hydroxyatractylenoide III B 1.32 (s) OH β S (26)
33 8β-hydroxy-1-oxoeudesma-3,7(11)-dien-12,8α-olide B 1.20 (s) OH β S (26)
34 (3S)-3-hydroxyatractylenolide III 3-O-β-d-Glucopyranoside B 1.25 (s) OH β S (27)
35 hydroxylindestenolide B 1.06 (s) OH β S (28)
36 linderagalactone D B 1.30 (s) OH β S (4)
37 8β-linderagalactone B 1.06 (s) OH β S (4)
38 linderolide I B 1.43 (s) OH β S (29)
39 linderolide A B 1.47 (s) OH β S (30)
40 linderolide B B 1.44 (s) OH β S (30)
41 biatractylolide B 1.07 (s) R β S (31)
42 bilindestenolide B 1.21 (s) R β S (19)
11 1 A 0.80 (s, H-14′) OR β S  
12 2 A 0.83 (s, H-14′) OR β S  
43 3 A 0.89 (s) OR, β S  
      0.82 (s, H-14′) OR      

Compound 2 was obtained as a white, amorphous powder. Via positive-mode HRESIMS at m/z 570.3086 for [M + NH4]+ (calcd for C32H44NO8, 570.3061), its molecular formula was determined as C32H40O8, a molecular weight that exceeded that of compound 1 by a methylene group. Comparison of the 1H and 13C NMR data of 1 and 2 (Tables 1 and 2) indicated that 2 differed from 1 in the presence of an ethoxy group (δH 3.77, 3.62, and 1.26; δC 57.9 and 15.4) instead of an methoxy group in 1. The HMBC between the methylene signals (δH 3.77 and 3.62) and C-8′ indicated that the ethoxy group is at the C-8′ position. The NOESY correlations of 2 suggested the same relative configuration as 1. The ECD spectrum of 2 was consistent with that of 1, indicating that 2 has the same absolute configuration as 1 (Figure S19). Consequently, the structure of linderaggrenolide B (2) was determined as depicted.

Compound 3 shares the same molecular formula (C31H38O8) as that of 1. The NMR data of 3 (Tables 1 and 2) suggested a 2D structure identical to that of 1. However, the downfield-shifted resonance of H-14 at δH 0.89 was indicative of an 8β oxygen atom (Table 5) due to the anisotropic effect.12 In addition, the downfield shifts of C-12′ (ΔδC 4.90 ppm) and H-12′ (ΔδH 0.17 ppm) suggested that the configuration of C-12′ in 3 was opposite of that of 1. As a consequence, the configuration at C-8 and C-12′ were defined as (8S, 12′R) for 3, which was supported by the opposite cotton effects at 220 nm in ECD spectra of 3 (Figure S29) and 1 (Figure S9). Therefore, the structure of linderaggrenolide C (3) was assigned as shown.

Linderaggrenolide D (4), a white amorphous powder, was assigned a molecular formula of C35H42O9, as deduced from its sodium adduct ions in the HRESIMS at m/z 629.2720 ([M + Na]+, calcd for C35H42O9Na, 629.2721). The 1H and 13C NMR spectroscopic data (Tables 1 and 2) suggested that compound 4 shared the same lindenane sesquiterpenoid dimer skeleton as 1. The main difference between the two compounds was that 7′,11′-epoxy group in 1 was replaced with a double bond (δC 113.9 and 140.3). This finding was confirmed by HMBC from H-13′ to C-7′ and C-11′, and from H-12′ to C-7′ and C-11′. The HMBC from H-6 to δC 171.1 and H-6′ to δC 170.5 indicated that two acetoxyl groups were attached to C-6 and C-6′, respectively. In addition, the H-14′ methyl signal at δH 0.53 of 4 was considerably upfield relative to that of 1 at δH 0.80, supporting the antirelationship of the C-14′ methyl group and the 8′-OCH3 group in the case of 4 (Table 5). The NOESY correlation of H-12′/H-14′ suggested that H-12′ was β-oriented. Furthermore, the absolute configuration of 4 was determined as depicted in Figure 4 by comparison of its experimental and calculated ECD spectra. The calculated ECD spectrum of (1R,3S,5S,6R,8R,10S,1′R,3′S,5′S,6′R,8′R,10′S,12′R)-4 agreed well with the experimental curve (Figure 4). Therefore, the structure of 4 was defined as shown.

Figure 4.

Figure 4

Experimental and calculated ECD spectra of compounds 4, 8, 9, and 11.

Compound 5 was isolated as an amorphous powder with a molecular formula of C34H40O9 as determined from the ion peak at m/z 610.3040 ([M + NH4]+, calcd for C34H46O9N, 610.3011). The 1H and13 C NMR data of 5 (Tables 1 and 2) were very similar to those of 4, with the only difference being the absence of signals corresponding to 8′-OCH3 and the presence of an additional hydroxyl group. The HMBC from H-6′and H-9′ to C-8′ and the deshielded signal of C-8′ at δC 117.4 confirmed that the 8′-OCH3 group in 4 was replaced by a hydroxyl group in 5. Analysis of the corresponding NOESY spectrum confirmed that the relative configuration of 5 was the same as that of 4. The experimental ECD spectrum of 5 was also consistent with that of 4, suggesting that both feature the same absolute configurations (Figure S49). Taken together, the structure of 5 was elucidated as shown and named linderaggrenolide E.

Linderaggrenolide F (6) and G (7) was found to have the same molecular formula as 4 based on their corresponding HRESIMS data. Analysis of their NMR data (Tables 1 and 2) of 6 and 7 indicated that they have an identical planar structure as 4. The major NMR differences between 6 and 4 were the chemical shifts of C-12′ (ΔδC 3.90 ppm) and H-12′(ΔδC 0.83 ppm), suggesting that the configuration of C-12′ in 6 was opposite of that in 4. Therefore, the configuration of 12′S was assigned in 6. The 1H and 13C-NMR data of 7 (Tables 1 and 2) were very similar to those of 4 and 6, except for the deshielded H-14 methyl signal at δH 0.95 in 7H 0.56 and 0.59 in 4 and 6, respectively), suggesting that the 8-oxygen atom in 7 is β-oriented (Table 5) and the 8S absolute configuration for linderaggrenolide G (7).

Linderaggrenolide H (8) was isolated as an amorphous powder. Its molecular formula, C30H37ClO7, was determined by the HRESIMS ion peak in the positive mode at m/z 562.2582 ([M + NH4]+, calcd for C30H41ClO7N, 562.2566). Analysis of the corresponding NMR data (Tables 3 and 4) revealed the presence of two α,β-unsaturated-γ-lactone groups in 8. In addition, six methyls, four methylenes, nine methines (three oxygenated), and eleven quaternary carbons were also distinguished with the aid of DEPT experiments. The aforementioned data suggested that 8 was likely a lindenane-type sesquiterpenoid dimer. However, in-depth NMR data analysis revealed some differences between 8 and the reported common lindenane sesquiterpenoid dimer, lindenanolide F.13 First, the absence of a methylene (δH 2.73, 2.18 and δC 45.2) and the presence of a methyl group (δH 1.11 and δC 33.1) suggested the occurrence of a ring-opening event, and the HMBC of H-9′/C-1′, C-5′, and C-10′ confirmed that the six-membered ring in lindenane sesquiterpenoid was open at the C-9′ position (Figure 2). Second, the key HMBC of H-8′/C-8 and the deshielded C-8 (δC 108.4) and C-8′ (δC 98.1) suggested that the two sesquiterpenoid units were connected via a bridge of C-8-O-C-8′. Third, analysis of the 1H–1H COSY spectrum indicated the presence of only one 1,2-distributed cyclopropane moiety, suggesting that the other one cyclopropane ring was open. Considering the presence of a chlorine atom, one deshielded methine group (δH 2.45 and δC 57.3) suggested that the chlorine atom was attached at C-1′. A pair of terminal bond proton signals were replaced by a singlet methyl group (δH 1.66 and δC 17.4) and the HMBC of H-15′/C-3′, C-4′, and C-5′, suggesting that the double bond has migrated to C-3′-C-4′. The planar structure of 8 could therefore be established as shown.

Table 3. 1H NMR (600 MHz) Spectroscopic Data of 814 in CDCl3 (δ in ppm, J in Hz).

position 8 9 10 11 12 13 14
1 1.44 m 1.34 m 1.36 m 1.37 m 1.46 m 1.32 m 1.40 m
2 0.86 m, 0.77 m 0.83 m, 0.76 m 0.83 m, 0.76 m 0.84 m, 0.78 m 0.86 m, 0.77 m 1.03 m, 0.82 m 0.88 m, 0.81 m
3 2.07 m 2.03 m 2.03 m 2.04 m 2.07 m 1.97 m 2.02 m
5 3.25 d (10.6) 3.19 dt (11.0, 2.5) 3.20 dt (11.3, 2.3) 3.42 d (11.0) 3.44 (11.0) 2.87 brd (11.8) 3.31 d (10.8)
6 4.49 d (10.6) 4.36 t (11.0) 4.36 brs 5.66 d (11,0) 5.83 d (11.0) 5.89 d (11.8) 5.72 d (10.8)
9 2.80 d (14.5) 2.52 d (14.5) 2.53 d (14.8) 2.58 d (14.4) 2.82 d (14.5) 2.64 d (13.5) 2.65 m
  2.25 d (14.5) 2.26 d (14.5) 2.26 d (14.8) 2.37 d (14.4) 2.29 d (14.5) 1.89 d (13.5)  
13 2.01 s 2.07 s 2.07 s 2.07 s 2.02 s 1.96 s 1.96 s
14 0.56 s 0.56 s 0.56 s 0.58 s 0.60 s 0.83 s 0.80 s
15 5.13 brs 5.29 brs 5.30 brs 5.04 brs 5.05 brs 5.06 brs 5.07 brs
    5.12 brs 5.12 brs 4.75 brs 4.73 brs 4.81 brs 5.03 brs
1′ 2.45 m 2.48 m 3.52 dd (9.8, 2.8) 1.35 m 1.36 m 1.34 m 1.55 m
      3.47 dd (9.8, 2.4)        
2′ 3.61 dd (10.4, 6.0) 3.61 dd (10.4, 6.0) 2.21 brs 0.92 m, 0.64 m 0.85 m, 0.64 m 0.82 m, 0.64 m 0.71 m, 0.07 m
  3.54 dd (10.4, 9.2) 3.52 dd (10.4, 9.2)          
3′ 5.67 brs 5.70 brs 5.41 brs 1.85 m 1.85 m 1.84 m 1.75 m
5′ 2.42 brs 2.45 brs 2.53 d 2.89 d (8.0) 3.00 d (9.0) 3.00 d (9.2) 4.99 brs
6′ 4.61 brs 4.62 d (8.3) 4.76 brs 5.49 d (8.0) 5.50 d (9.0) 5.47 d (9.2)  
8′ 5.92 s 5.41 brs 5.32 brs 5.44 s 5.78 s 5.43 s  
9′ 1.11 s 1.11 s 1.10 s 1.10 s 1.12 s 1.08 s 2.64 overlapped
13′ 2.08 s 2.12 s 2.23 s 2.05 s 2.05 s 2.03 s 2.02 s
14′ 1.05 s 1.04 s 1.17 s 0.93 s 0.94 s 0.90 s 1.21 s
15′ 1.66 brs 1.68 s 1.70 s 5.10 brs 5.09 brs 5.07 brs 4.72 dd (12.8, 1.2)
        4.80 brs 4.78 brs 4.77 brs 4.65 brd (12.8)
      3.33 s (1′-OCH3) 2.20 s (6-OAc) 2.05 s (6-OAc) 2.00 s (6-OAc) 2.07 s (6-OAc)
        2.07 s (6′-OAc) 2.02 s (6′-OAc) 2.20 s (6′-OAc)  

Table 4. 13C NMR (150 MHz) Spectroscopic Data of 814 in CDCl3 (δ in ppm).

position 8 9 10 11 12 13 14
1 29.4 29.4 29.4 29.4 29.4 27.8 28.2
2 17.0 16.9 16.9 17.0 17.0 16.8 17.3
3 24.0 24.0 24.0 23.8 23.8 23.7 23.7
4 149.4 149.0 149.2 148.6 148.5 147.0 149.1
5 63.2 62.9 62.9 60.3 60.3 67.8 63.5
6 62.9 63.5 63.5 62.9 62.9 69.7 67.2
7 155.4 157.3 157.3 153.3 151.4 157.3 144.9
8 108.4 107.8 107.7 107.0 107.9 197.9 199.5
9 48.2 49.7 49.8 50.7 48.4 48.1 53.2
10 39.1 38.0 38.0 37.5 37.9 38.3 37.1
11 131.4 132.9 132.8 134.1 133.8 126.0 134.9
12 170.8 170.7 170.8 170.5 169.8 170.4 170.6
13 8.9 9.1 9.1 9.9 9.7 9.1 17.8
14 21.8 22.1 22.1 21.4 21.7 18.2 20.0
15 108.2 109.3 109.2 108.1 108.1 109.5 108.7
1′ 57.3 57.4 73.0 33.8 34.0 33.9 24.7
2′ 46.8 46.7 56.0 10.8 11.0 11.1 15.3
3′ 131.0 131.4 127.9 24.7 24.6 24.6 23.6
4′ 138.4 138.2 139.6 152.3 152.4 152.2 144.4
5′ 63.4 63.4 62.7 51.9 51.8 51.6 132.0
6′ 67.1 67.0 69.0 67.8 67.6 67.6  
7′ 155.3 155.2 155.1 158.6 158.1 158.0 142.8
8′ 98.1 95.1 95.3 94.8 97.3 95.4 167.1
9′ 33.1 33.0 35.5 32.7 32.6 32.6 37.0
10′ 43.9 44.0 41.3 42.0 41.8 41.8 51.5
11′ 131.0 130.9 131.2 128.7 128.9 129.8 141.8
12′ 170.7 169.3 169.3 169.2 168.9 168.4 166.4
13′ 12.5 12.1 12.4 12.2 12.3 12.1 10.9
14′ 19.6 19.5 20.4 23.0 23.1 23.1 24.2
15′ 17.4 17.4 17.1 109.9 110.0 110.1 62.9
      1′-OCH3 6-OAc 6-OAc 6-OAc 6-OAc
      58.7 171.3 170.1 169.8 170.4
        21.0 21.0 21.1 21.2
        6′-OAc 6′-OAc 6′-OAc  
        169.7 170.0 169.5  
        21.2 21.2 20.8  

In the NOESY spectrum, the cross peaks of CH3-9′/H-1′ and H-5′ allowed for the initial assignment of α-orientation for H-1′, H-5′, and H-9′. The NOESY correlations of H-6′/CH3-14′ revealed that they were β-orientated (Figure 2). The absolute configuration of compound 8 was defined as (1R,3S,5S,6R,8R,10S,1′R,5′S,6′R,8′S)-8 by comparison of its experimental and calculated ECD data (Figure 4). Therefore, the structure of compound 8 could be defined as shown.

Compound 9 was isolated as an amorphous powder. Its molecular formula was determined as C30H37ClO7 based on its HRESIMS data, which was the same as that of 8. Analysis of the NMR data (Tables 3 and 4) showed that 9 shared the same 2D structure with 8. The major differences between the two compounds were upfield shifts at C-8′ (ΔδC 3.0 ppm) and H-8′ (ΔδH 0.51 ppm) due to different configuration of C-8′. The (1R,3S,5S,6R,8R,10S,1′R,5′S,6′R,8′R) absolute configuration of compound 9 was defined by comparing its experimental and calculated ECD spectra (Figure 4). Therefore, the structure of linderaggrenolide I (9) was assigned as shown.

Compound 10 was obtained as an amorphous powder. Its molecular formula, C31H40O8, was determined by a HRESIMS pseudo-molecular ion peak in the positive mode at m/z 563.2626 [M + Na]+ (calcd for 563.2615). The spectroscopic data (Tables 3 and 4) of 10 displayed high similarity to those of 9, apart from the replacement of the chloride group by a methoxyl group in 10. In addition, the spin system, CH2(C-2′)–CH(C-1′)–CH(C-3′), could be established by analysis of the corresponding 1H–1H COSY spectrum. The HMBC of −OCH3/C-2′ provided evidence for the methoxyl group being positioned at C-2′.

The NOESY correlations of CH3-9′/H-1′ and H-5′, and H-6′/CH3-14′ were indicative of 9′-CH3α, 5′-Hα, 1′-Hα, 6′-Hβ, and 14′-CH3β in 10, equally to those in 9. The resonances for H-8′ and C-8′ agree well with those of 8′R in compound 9, suggesting that 10 exhibits the same configuration as 9. In addition, the highly similar ECD spectra of 9 and 10 also supported the above notion (Figure S99). Thus, the structure of linderaggrenolide J (10) could be assigned as shown.

Linderaggrenolide K (11) had a molecular formula C34H40O9 based on its HRESIMS (m/z, 610.3021 [M + NH4]+). The NMR data of 11 (Tables 3 and 4) suggested that compound 11 is structurally closely related to 10. The only differences were the absence of the methoxyl group, the presence of a 1,2-distributed cyclopropane, and the migration of the C3′=C4′ double bond to C4′=C15′. These changes were confirmed by the HMBC of H-15′/C-3′, C-4′, and C-5′, and 1H–1H COSY correlations of H-1′/H-2′/H-3′. In addition, the HMBC of H-6 and H-6′ with acetyl carbonyl (δC 171.3 and δC 169.5) indicated that the additional two acetoxyl groups were attached to C-6 and C-6′, respectively. Analysis of NOESY correlations indicated that 11 adopted the same relative configuration as 10. The absolute configuration of 11 was defined as (1R,3S,5S,6R,8R,10S,1′R,3′S,5′R,6′R,8′R)-11 by comparison of its experimental and calculated ECD spectra (Figure 4).

The same molecular formula, C34H40O9, for both compounds 12 and 13 was established based on the corresponding HRESIMS data at m/z 610.3027 [M + NH4]+ and 610.3014 [M + NH4]+, respectively. Analysis of the NMR data (Tables 3 and 4) suggested that both 12 and 13 shared the same planar structure as 11, except for the configurations of C-8 and C-8′. The chemical shifts of H-8′ (δH 5.78) and C-8′(δC 97.3) of 12 were deshielded in comparison to those of 11C 94.8 and δH 5.44), suggesting that the configuration of C-8′ in 12 was opposite of that of 11. Meanwhile, the H-14 methyl signal of 13 at δH 0.83 was found to be deshielded relative to that of 11 at δH 0.58 due to the anisotropic effect of the 8β oxygen atom, suggesting that the configuration of C-8 in 13 was opposite to that of 11. The ECD spectra of 12 and 8 were very similar, suggesting the (8′S) absolute configuration of 12 as shown. It is worth mentioning that the configuration of C-8′ has a profound influence on the ECD curve, while the configuration of C-8 does not. Therefore, the structures for linderaggrenolide L (12) and linderaggrenolide N (13) were proposed as shown.

Compound 14 was assigned the molecular formula of C31H34O8 on the basis of the quasimolecular ion peak at m/z 557.2151 [M + Na]+ in the positive-mode HRESIMS spectrum. The 13C NMR and DEPT data (Tables 3 and 4) showed 31 signals corresponding to five methyl, six methylene, seven methine, and thirteen quaternary carbons. The 1H–1H COSY spectrum showed the presence of two cyclopropane rings. Furthermore, compound 14 contained an acetyl group due to the presence of resonance signals at δC 170.4, 21.1 and δH 2.07. The presence of two lindenane sesquiterpenoid (units A and B) in 14 was elucidated by comprehensive analysis of the 2D NMR spectra. In unit A, the HMBC from H-6 and H-9 to C-8 suggested that the C-8 position of unit A was a carbonyl carbon (Figure 2). Moreover, the HMBC cross peaks of H-13/C-7, C-11, and C-12 confirmed the furan lactone ring in unit A was open. The NMR data of unit B were very similar to those of lindenanolide E.13 The one significant difference was the absence of an aldehyde group. NOESY correlations were observed among the following proton signals: H-2α to H-1/H-3; H-2β to H-14, which confirmed that unit B had the same relative configuration as lindenanolide E. Finally, the two units were proposed to be linked via an oxygen atom between C-12 and C-15′ based on the HMBC networks of H-15′/C-12, C-3′, C-4′, and C-5′. Thus, the structure of linderaggrenolide N (14) was identified as shown.

Although sesquiterpenoids were reported as the characteristic constituents of L. aggregata, sesquiterpenoid dimers were rarely found in this plant. To date, most sesquiterpenoid dimers have been reported from the Chloranthaceae family.1 The disesquiterpenoids found in this study may represent a group of metabolites with taxonomic significance for L. aggregata. From a structural perspective, linderaggrenolides A–N (114), belonging to the pseudo-disesquiterpenoid family, could be easily distinguished since two sesquiterpenoids were found to be connected via an oxygen bridge. According to the results reported previously, lindenane sesquiterpenoid dimers are mainly formed by Diels–Alder or hetero-Diels–Alder reactions, and the two units are generally linked directly by one or two C–C bonds.14 However, lindenane sesquiterpenoid dimers containing an oxygen bridge have been rarely reported from nature. Therefore, the linkage feature of these compounds also underscores the chemical diversity of sesquiterpenoid dimers. Compounds 814 represent a new dimerization pattern of the lindenane dimer class, which is proposed to be a rare 8,9-seco lindenane and a lindenane sesquiterpenoid. In addition, a chlorine-containing sesquiterpenoid has been isolated from L. aggregata,4 and no chlorine-containing solvents were used during the process of extraction and isolation. Therefore, it is likely that both compounds 8 and 9 are natural chlorine-containing substances.

All isolated compounds were tested for their TGF-β inhibitory activity. As shown in Figure 5, compounds 8 and 9 significantly downregulated the p-smad2 expression in a concentration-dependent manner without any impact on the expression of smad2 and β-actin protein, with the IC50 values of 25.91 and 21.52 μM, respectively. These data suggested that the chlorines in these compounds should be important for their TGF-β inhibitory activity.

Figure 5.

Figure 5

Compounds 8 and 9 blocked TGF-β, inducing smad2 phosphorylation in a dose-dependent manner.

3. Experimental Section

3.1. General Experimental Procedures

Melting point was obtained on a BA-350 (BenAng) melting point apparatus. Optical rotations were measured on a Rudolph Research Autopol I automatic polarimeter. UV and ECD spectra were recorded on a JASCO High Performance J-1500 CD spectrometer. NMR spectra were obtained at 600 MHz for 1H NMR and 150 MHz for 13C NMR, respectively, using a Bruker Ascend 600 spectrometer. HRESIMS data were acquired on an Agilent Technologies 6230 Accurate Mass Q-TOF UHPLC/MS spectrometer.

3.2. Plant Material

The dried roots of L. aggregata were collected from the Zhejiang Province, China, in March 2017, and identified by Dr. G. Y. Zhu. The voucher specimen (LA-201703) was deposited at the State Key Laboratory of Quality Research in Chinese Medicines, Macau University of Science and Technology.

3.3. Extraction and Isolation

The dried roots of L. aggregata were ground to powder and extracted with 80% EtOH (4 × 12 L) under reflux.10,11 The extract was then concentrated in vacuo at 50 °C. The extract was successively partitioned with petroleum ether (60–90 °C), EtOAc, and n-butanol. After solvent removal, the petroleum ether fraction (145.0 g) was subjected to silica gel column chromatography (CC), eluted with a gradient of increasing EtOAc (0–100%) in petroleum ether to afford six fractions (Fr. 1–Fr. 6) and judged by TLC analysis. Fr. 2 (20.5 g) was subjected to CC on RP C-18 gel and eluted with MeCN–H2O (40:60 to 100:0) to provide seven subfractions (Fr. 2-1–Fr. 2-7). Fr. 2-5 and Fr. 2-6 were initially subjected to RP-C8 column chromatography and eluted with MeCN–H2O and further purified by semipreparative HPLC eluted with MeOH–H2O to obtain compounds 4 (3.5 mg), 5 (6.0 mg), 6 (2.5 mg), 7 (2.0 mg), 11 (8.0 mg), 12 (10.0 mg), 13 (2.5 mg), and 14 (2.0 mg). Fr. 3 (15.0 g) was separated in fashions similar to Fr. 2 to obtain compounds 1 (10. 0 mg), 2 (3.0 mg), 3 (5.0 mg), 8 (3.0 mg), 9 (6.5 mg), and 10 (2.0 mg).

3.3.1. Linderaggrenolide A (1)

Colorless square crystal; m.p. 237–238 °C; [α]D22.5 −21.2 (c 0.3, MeOH); UV (MeOH) λmax(log ε) 196 (4.32), 230 (4.10), 300 (3.46); ECD (MeOH): 200 (Δε +4.14), 220 (Δε −5.12), 254 (Δε +0.79), 295 (Δε −0.50) nm; IR, νmax: 3401, 2960, 2362, 1774, 1641, 1538, 1462, 1362, 1240, 1094, 671 cm–1; 1H and 13C NMR (CDCl3), see Tables 1 and 2; HRESIMS m/z 556.2931: [M + NH4]+ (calcd for C31H38O8NH4, 556.2905).

3.3.2. Linderaggrenolide B (2)

White powder, [α]D22.5 −2.6 (c 0.3, MeOH); UV (MeOH) λmax(log ε), 196 (4.34), 230 (3.92), 300 (2.88); ECD (MeOH): 200 (Δε +24.26), 220 (Δε −15.23), 254 (Δε +0.56) nm; IR, νmax: 3454, 2929, 2362, 1773, 1656, 1387, 1072, 1007, 920, 740 cm–1; 1H and 13C NMR (CDCl3), see Tables 1 and 2; HRESIMS m/z 570.3086: [M + NH4]+ (calcd for C32H40O8NH4, 570.3061).

3.3.3. Linderaggrenolide C (3)

White powder, [α]D22.5 +100.0 (c 0.3, MeOH); UV (MeOH) λmax(log ε) 196 (4.44), 225 (4.11), 290 (3.07); ECD (MeOH): 205 (Δε +19.56), 248 (Δε +4.92) nm; IR, νmax: 3441, 2928, 2361, 1770, 1385, 1082, 980, 737 cm–1; 1H and 13C NMR (CDCl3), see Tables 1 and 2; HRESIMS m/z 556.2917: [M + NH4]+ (calcd for C31H38O8NH4, 556.2905).

3.3.4. Linderaggrenolide D (4)

White powder, [α]D22.5 −2.8 (c 0.3, MeOH); UV (MeOH) λmax(log ε), 196 (4.28), 225 (3.75); ECD (MeOH): 200 (Δε +15.32), 218 (Δε −7.16), 240 (Δε +1.20), 260 (Δε −1.53) nm; IR, νmax: 3365, 2964, 2362, 1771, 1645, 676 cm–1; 1H and 13C NMR (CDCl3), see Tables 1 and 2; HRESIMS m/z 629.2720: [M + Na]+ (calcd for C35H42O9Na, 629.2721).

3.3.5. Linderaggrenolide E (5)

White powder, [α]D22.5 −3.1 (c 0.3, MeOH); λmax(log ε) 196 (4.48), 230 (4.00); ECD (MeOH): 200 (Δε +14.62), 220 (Δε −7.44), 237 (Δε +1.64), 260 (Δε −1.31) nm; IR, νmax: 3449, 2361, 1738, 1515, 1373, 1291, 1024, 675 cm–1; 1H and 13C NMR (CDCl3), see Tables 1 and 2; HRESIMS m/z 610.3040: [M + NH4]+ (calcd for C34H40O9 NH4, 610.3011).

3.3.6. Linderaggrenolide F (6)

White powder, [α]D22.5 +8.6 (c 0.3, MeOH); λmax(log ε), 196 (4.39), 230 (3.86); ECD (MeOH): 200 (Δε +11.43), 218 (Δε −3.87) nm; IR, νmax: 3365, 2362, 1771, 1645, 1387 cm–1; 1H and 13C NMR (CDCl3), see Tables 1 and 2; HRESIMS m/z 629.2725: [M + Na]+ (calcd for C34H40O9 Na, 629.2721).

3.3.7. Linderaggrenolide G (7)

White powder, [α]D22.5 +24.6 (c 0.3, MeOH); λmax(log ε), 196 (4.36), 225 (3.98); ECD (MeOH): 200 (Δε +6.68), 220 (Δε −1.79), 245 (Δε +1.63) nm; IR, νmax: 3447, 2360, 1632, 1455, 1208, 753 cm–1; 1H and 13C NMR (CDCl3), see Tables 1 and 2; HRESIMS m/z 629.2732: [M + Na]+ (calcd for C35H42O9Na, 629.2721).

3.3.8. Linderaggrenolide H (8)

White powder, [α]D22.5 −139.1 (c 0.3, MeOH); λmax(log ε), 200 (4.55), 228 (4.18); ECD (MeOH): 200 (Δε +5.84), 220 (Δε −10.93), 260 (Δε −2.30) nm; IR, νmax: 3447, 2961, 2361, 1779, 1385, 1082, 980, 736 cm–1; 1H and 13C NMR (CDCl3), see Tables 3 and 4; HRESIMS m/z 562.2582: [M + NH4]+ (calcd for C30H37ClO7NH4, 562.2566).

3.3.9. Linderaggrenolide I (9)

White powder, [α]D22.5 +135.9 (c 0.3, MeOH); λmax(log ε), 200 (4.48), 228 (4.12); ECD (MeOH): 200 (Δε +4.26), 216 (Δε −4.66), 235 (Δε +7.95) nm; IR, νmax: 3394, 2935, 1740, 1697, 1230, 980, 672 cm–1; 1H and 13C NMR (CDCl3), see Tables 3 and 4; HRESIMS m/z 567.2127: [M + Na]+ (calcd for C30H37ClO7Na, 567.2120).

3.3.10. Linderaggrenolide J (10)

White powder, [α]D22.5 +5.97 (c 0.3, MeOH); λmax(log ε), 200 (4.28), 228 (3.91); ECD (MeOH): 200 (Δε +6.36), 218 (Δε −4.38), 245 (Δε +2.52) nm; IR, νmax: 3734, 2929, 2359, 1772, 1540, 1456, 1107, 950, 681 cm–1; 1H and 13C NMR (CDCl3), see Tables 3 and 4; HRESIMS m/z 563.2626: [M + Na]+ (calcd for C31H40O8Na, 563.2615).

3.3.11. Linderaggrenolide K (11)

White powder, [α]D22.5 +382.3 (c 0.2, MeOH); λmax(log ε), 200 (4.56), 228 (4.19); ECD (MeOH): 200 (Δε +18.38), 218 (Δε −16.32), 235 (Δε +6.22) nm; IR, νmax: 3448, 2957, 2360, 1772, 1649, 1230, 1021 cm–1; 1H and 13C NMR (CDCl3), see Tables 3 and 4; HRESIMS m/z 610.3021: [M + NH4]+ (calcd for C34H40O9NH4, 610.3011).

3.3.12. Linderaggrenolide L (12)

White powder, [α]D22.5 −52.9 (c 0.2, MeOH); λmax(log ε), 200 (4.56), 228 (4.14); ECD (MeOH): 200 (Δε +16.36), 218 (Δε −18.94), 252 (Δε −3.76) nm; IR, νmax: 3445, 2361, 1630, 1456, 697 cm–1; 1H and 13C NMR (CDCl3), see Tables 3 and 4; HRESIMS m/z 610.3027: [M + NH4]+ (calcd for C34H40O9NH4, 610.3011).

3.3.13. Linderaggrenolide M (13)

White powder, [α]D22.5 +29.3 (c 0.2, MeOH); λmax(log ε), 200 (4.58), 228 (4.24); ECD (MeOH): 202 (Δε +28.41), 228 (Δε −20.41), 256 (Δε +2.25) nm; IR, νmax: 3445, 1653, 1450 cm–1; 1H and 13C NMR (CDCl3), see Tables 3 and 4; HRESIMS m/z 610.3014: [M + NH4]+ (calcd for C34H40O9NH4, 610.3011).

3.3.14. Linderaggrenolide N (14)

White powder, [α]D22.5 −219.4 (c 0.2, MeOH); λmax(log ε), 200 (4.44), 250 (3.94); ECD (MeOH): 214 (Δε −12.53), 300 (Δε −1.21) nm; IR, νmax: 3444, 2361, 1773, 1651, 1230, 739 cm–1; 1H and 13C NMR (CDCl3), see Tables 3 and 4; HRESIMS m/z 557.2151: [M + Na]+ (calcd for C31H34O8Na, 557.2146).

3.4. Single-Crystal X-ray Diffraction Crystallographic Data of Compound 1

A colorless crystal of 1 was mounted on a glass fiber at a random orientation. The intensity data were collected on a diffractometer Rigaku Oxford Diffraction Supernova Dual Source, Cu at Zero, equipped with an AtlasS2 CCD using Cu Kα radiation (1.54178 Å) by using a w scan mode. The structures were solved by direct methods using Olex2 software and refined anisotropically with XL using a full-matrix least squares procedure. The nonhydrogen atoms were located from the trial structure, and the hydrogen atom positions were fixed geometrically at the calculated distances and allowed to ride on the parent atoms. Crystallographic data for the structures of 1 have been deposited in the Cambridge Crystallographic Data Centre database (deposition numbers CCDC 2038628). Copies of the data can be obtained free of charge from the CCDC at www.ccdc.cam.ac.uk.

3.4.1. Crystal Data of Compound 1

C31H38O8 (M =538.61 g/mol), orthorhombic, space group P212121 (no. 19), a = 8.04240(5) Å, b = 13.88872(7) Å, c = 25.03583(14) Å, V = 2796.47(3) Å3, Z = 4, T = 100.01(10) K, μ (Cu Kα) = 0.750 mm–1, Dcalc = 1.279 g/cm3, 32118 reflections measured (7.062° ≤ 2Θ ≤ 147.322°), 5585 unique (Rint = 0.0340, Rsigma = 0.0191) which were used in all calculations. The final R1 was 0.0307 (I > 2σ(I)) and wR2 was 0.0790 (all data).

3.5. ECD Calculation

Monte Carlo conformational searches were carried out by means of the Spartan’s 10 software using Merck molecular force field (MMFF). The conformers with Boltzmann population of over 5% were chosen for ECD calculations, and then, the conformers were initially optimized at the B3LYP/6-31 + g(d, p) level in MeOH using the CPCM polarizable conductor calculation model. The theoretical calculation of ECD was conducted in MeOH using TD-DFT at the B3LYP/6-311 + g(d, p) level; for all conformers of compounds, rotatory strengths for a total of 50 excited states were calculated. ECD spectra were generated using the program SpecDis 1.6 (University of Würzburg, Würzburg, Germany) and GraphPad Prism 5 (University of California San Diego, USA) from dipole-length rotational strengths by applying Gaussian band shapes with sigma = 0.3 eV.

3.6. TGF-β Inhibitory Activity Assay

3.6.1. Cell Lines and Cell Cultures

A549 cells were cultured in 1640 medium supplemented with 10% fetal bovine serum (FBS) (GIBCO, UK). The cells were incubated at 37 °C in a humidified atmosphere containing 5% CO2.

3.6.2. Western Blot Analysis

Cells were seeded onto 6-well plates with FBS-free 1640 medium and cultured for 24 h. After treatment with 1, 10, and 25 μM of 8 and 9 (Figure 5), cells were collected, and lysis buffer was added. Lysation was allowed to proceed on ice for 30 min. Protein samples were subjected to electrophoresis in 10% SDS–PAGE gel and transferred onto nitrocellulose membrane (NC membrane). The membranes were blocked in 5% BSA and incubated at 4 °C with primary antibody overnight, followed by a secondary antibody for 1 h at room temperature. Protein bands were detected using a LI-COR Odyssey imaging system (Lincoln, NE, USA). Epigallocatechin gallate (EGCG) was used as the positive control.

Primary antibodies against p-smad2 and smad2 were purchased from Cell Signaling Technology (Beverly, MA, USA). The primary antibody against β-actin was purchased from Abcam Inc. (Cambridge, MA, USA). Secondary antibodies, goat antimouse/antirabbit IgG H&L (IRDye 800CW), were purchased from Abcam Inc. (Cambridge, MA, USA).

Acknowledgments

This work was financially supported by the Macao Science and Technology Development Funds (0008/2019/A, 0075/2019/AGJ, and 0023/2019/AKP) and Guangxi Key Laboratory of Efficacy Study on Chinese Materia Medica Fund (20-065-38).

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.0c06349.

  • HRESIMS, IR, UV, CD, 1H and 13C NMR, DEPT, HSQC, HMBC, COSY, and NOESY spectra of compounds 1–14 (PDF)

  • Single-crystal X-ray diffraction data for compound 1 (CIF).

The authors declare no competing financial interest.

Supplementary Material

ao0c06349_si_001.pdf (10.9MB, pdf)
ao0c06349_si_002.cif (658.3KB, cif)

References

  1. Zhan Z.-J.; Ying Y.-M.; Ma L.-F.; Shan W.-G. Natural disesquiterpenoids. Nat. Prod. Rep. 2011, 28, 594–629. 10.1039/c0np00050g. [DOI] [PubMed] [Google Scholar]
  2. Wang N.; Minatoguchi S.; Arai M.; Uno Y.; Hashimoto K.; Chen X.-H.; Fukuda K.; Akao S.; Takemura G.; Fujiwara H. Lindera strychnifolia is protective against post-ischemic myocardial dysfunction through scavenging hydroxyl radicals and opening the mitochondrial K ATP channels in isolated rat hearts. Am. J. Chin. Med. 2004, 32, 587–598. 10.1142/s0192415x04002223. [DOI] [PubMed] [Google Scholar]
  3. Cheng X.-L.; Ma S.-C.; Wei F.; Wang G. L.; Xiao X. Y.; Lin R. C. A new sesquiterpene isolated from Lindera aggregata (SIMS) KOSTERM. Chem. Pharm. Bull. 2007, 55, 1390–1392. 10.1248/cpb.55.1390. [DOI] [PubMed] [Google Scholar]
  4. Gan L.-S.; Zheng Y.-L.; Mo J.-X.; Liu X.; Li X.-H.; Zhou C.-X. Sesquiterpene lactones from the root tubers of Lindera aggregata. J. Nat. Prod. 2009, 72, 1497–1501. 10.1021/np900354q. [DOI] [PubMed] [Google Scholar]
  5. Gan L.-S.; Yao W.; Mo J.-X.; Zhou C.-X. Alkaloids from Lindera aggregata. Nat. Prod. Commun. 2009, 4, 1934578X0900400. 10.1177/1934578x0900400111. [DOI] [PubMed] [Google Scholar]
  6. Zhang C.; Sun Q.; Zhao Y.; Wang Z. Studies on flavonoids from leaves of Lindera aggregata (Sims) Kosterm. Chin. J. Med. Chem. 2001, 11, 274–276. [PubMed] [Google Scholar]
  7. Zhang C. F.; Sun Q. S.; Zhao Y. Y.; Wang Z. T. One new type proanthocyanidin trimer from Lindera aggregata (Sims) Kosterm. Chin. Chem. Lett. 2003, 14, 1033–1036. [Google Scholar]
  8. Zhang C. F.; Chou G. X.; Sun Q. S.; Wang Z. T.; Hattori M. Tannins from the stems of Lindera aggregata. Chin. J. Nat. Med. 2003, 1, 204–206. [Google Scholar]
  9. Wang F.; Gao Y.; Zhang L.; Bai B.; Hu Y. N.; Dong Z. J.; Zhai Q. W.; Zhu H. J.; Liu J. K. A pair of windmill-shaped enantiomers from Lindera aggregata with activity toward improvement of insulin sensitivity. Org. Lett. 2010, 12, 3196–3199. 10.1021/ol1011289. [DOI] [PubMed] [Google Scholar]
  10. Liu X.; Yang J.; Fu J.; Yao X.-J.; Wang J.-R.; Liu L.; Jiang Z.-H.; Zhu G.-Y. Aggreganoids A–F, carbon-bridged sesquiterpenoid dimers and trimers from Lindera aggregata. Org. Lett. 2019, 21, 5753–5756. 10.1021/acs.orglett.9b02166. [DOI] [PubMed] [Google Scholar]
  11. Liu X.; Yang J.; Yao X.-J.; Yang X.; Fu J.; Bai L.-P.; Liu L.; Jiang Z.-H.; Zhu G.-Y. Linderalides A-D, disesquiterpenoid–geranylbenzofuranone conjugates from Lindera aggregata. J. Org. Chem. 2019, 84, 8242–8247. 10.1021/acs.joc.9b00522. [DOI] [PubMed] [Google Scholar]
  12. Kouno I.; Hirai A.; Fukushige A.; Jiang Z.-H.; Tanaka T. New eudesmane sesquiterpenes from the root of Lindera strychnifolia. J. Nat. Prod. 2001, 64, 286–288. 10.1021/np000154s. [DOI] [PubMed] [Google Scholar]
  13. Zhang C.-f.; Nakamura N.; Tewtrakul S.; Hattori M.; Sun Q.-s.; Wang Z.-t.; Fujiwara T. Sesquiterpenes and alkaloids from Lindera chunii and their inhibitory activities against HIV-1 integrase. Chem. Pharm. Bull. 2002, 50, 1195. 10.1248/cpb.50.1195. [DOI] [PubMed] [Google Scholar]
  14. Li C.; Lei X. Strategies toward the biomimetic syntheses of oligomeric sesquiterpenoids. J. Org. Chem. 2014, 79, 3289–3295. 10.1021/jo5002092. [DOI] [PubMed] [Google Scholar]
  15. Liu Q.; Jo Y. H.; Kim S. B.; Jin Q.; Hwang B. Y.; Lee M. K. Sesquiterpenes from the roots of Lindera strychnifolia with inhibitory effects on nitric oxide production in RAW 264.7 cells. Bioorg. Med. Chem. Lett. 2016, 26, 4950–4954. 10.1016/j.bmcl.2016.09.012. [DOI] [PubMed] [Google Scholar]
  16. Maia L. F.; Epifanio R. D. A.; Eve T.; Fenical W. New fish feeding deterrents, including a novel sesquiterpenoid heterogorgiolide, from the Brazilian gorgonian Heterogorgia uatumani (Octocorallia, Gorgonacea). J. Nat. Prod. 1999, 62, 1322–1324. 10.1021/np990138z. [DOI] [PubMed] [Google Scholar]
  17. Cao Q.; Dai W.; Li B.; Zhang M. Sesquiterpenoids from the stems and leaves of Gochnatia decora. Phytochem. Lett. 2019, 30, 6–9. 10.1016/j.phytol.2019.01.003. [DOI] [Google Scholar]
  18. Wang B. D.; Yu Y. H.; Teng N. N.; Jiang S. H.; Zhu D. Y. Structural elucidation of biepiasterolid. Acta Chim. Sin. 1999, 57, 1022–1025. [Google Scholar]
  19. Li Y.; Dai M.; Peng D. New bisesquiterpenoid lactone from the wild rhizome of Atractylodes macrocephala Koidz grown in Qimen. Nat. Prod. Res. 2017, 31, 2381–2386. 10.1080/14786419.2017.1309531. [DOI] [PubMed] [Google Scholar]
  20. Wang P.; Xie Z.-S.; Song J.-Y.; Zeng H.-H.; Dai L.-P.; Ye Z.-P.; Gao S.; Xu J.-Y.; Zhang Z.-Q. Four new sesquiterpene lactones from Atractylodes macrocephala and their CREB agonistic activities. Fitoterapia 2020, 147, 104730. 10.1016/j.fitote.2020.104730. [DOI] [PubMed] [Google Scholar]
  21. Yaermaimaiti S.; Wang P.; Luo J.; Li R.-J.; Kong L.-Y. Sesquiterpenoids from the seeds of Sarcandra glabra and the potential anti-inflammatory effects. Fitoterapia 2016, 111, 7–11. 10.1016/j.fitote.2016.03.020. [DOI] [PubMed] [Google Scholar]
  22. Duan J.-a.; Wang L.; Qian S.; Su S.; Tang Y. A new cytotoxic prenylated dihydrobenzofuran derivative and other chemical constituents from the rhizomes of Atracylodes lancea DC. Arch Pharm. Res. 2008, 31, 965–969. 10.1007/s12272-001-1252-z. [DOI] [PubMed] [Google Scholar]
  23. Li Y.; Yang X.-W. New eudesmane-type sesquiterpenoids from the processed rhizomes of Atractylodes macrocephala. J. Asian Nat. Prod. Res. 2014, 16, 123–128. 10.1080/10286020.2013.840295. [DOI] [PubMed] [Google Scholar]
  24. Xiao Z.-Y.; Wang X.-C.; Zhang G.-P.; Huang Z.-L.; Hu L.-H. Terpenoids from roots of Chloranthus spicatus. Helv. Chim. Acta 2010, 93, 803–810. 10.1002/hlca.200900299. [DOI] [Google Scholar]
  25. Lin S.; Li Y. C.; Guo Y. Y.; Guo S. M.; Que H. Q.; Qi Y. P. Chemical constituents of Melastoma dodecandrum (II). Chin. Tradit. Herb. Drugs 1994, 40, 1192–1195. [Google Scholar]
  26. Xu Y.-J.; Tang C.-P.; Tan M.-J.; Ke C.-Q.; Wu T.; Ye Y. Sesquiterpenoids and diterpenoids from Chloranthus anhuiensis. Chem. Biodivers. 2010, 7, 151–157. 10.1002/cbdv.200800300. [DOI] [PubMed] [Google Scholar]
  27. Kitajima J.; Kamoshita A.; Ishikawa T.; Takano A.; Fukuda T.; Isoda S.; Ida Y. Glycosides of Atractylodes japonica. Chem. Pharm. Bull. 2003, 51, 152–157. 10.1248/cpb.51.152. [DOI] [PubMed] [Google Scholar]
  28. Zhao J.-b.; Lu X.-h.; Xu X.-d.; Luo L.; Fu H.; Li J.; Sun L.; Zhang L. Chemical constituents of leaves of Lindera aggregata (Sims) Kosterm. Chin. J. Pharm. 2012, 47, 1702–1705. [Google Scholar]
  29. Liu Q.; Ahn J. H.; Kim S. B.; Lee C.; Hwang B. Y.; Lee M. K. Sesquiterpene lactones from the roots of Lindera strychnifolia. Phytochemistry 2013, 87, 112–118. 10.1016/j.phytochem.2012.11.004. [DOI] [PubMed] [Google Scholar]
  30. Sumioka H.; Harinantenaina L.; Matsunami K.; Otsuka H.; Kawahata M.; Yamaguchi K. Linderolides A–F, eudesmane-type sesquiterpene lactones and linderoline, a germacrane-type sesquiterpene from the roots of Lindera strychnifolia and their inhibitory activity on NO production in RAW 264.7 cells in vitro. Phytochemistry 2011, 72, 2165–2171. 10.1016/j.phytochem.2011.08.004. [DOI] [PubMed] [Google Scholar]
  31. Rosquete C.; Olmo E. D.; Sanz F.; Feliciano A. S. The crystal structure of biatractylolide, an 8, 8′(C–C) linked dimeric 12, 8-eudesmanolide from the resin of Trattinickia rhoifolia WILLD. Chem. Pharm. Bull. 2002, 50, 964–965. 10.1248/cpb.50.964. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

ao0c06349_si_001.pdf (10.9MB, pdf)
ao0c06349_si_002.cif (658.3KB, cif)

Articles from ACS Omega are provided here courtesy of American Chemical Society

RESOURCES