Structure-Based Molecular Networking for the Discovery of Anti-HBV Compounds from Saussurea lappa (Decne.) C.B Clarke

Abstract

It is a crucial to find target compounds in natural product research. This study presents a concept of structure-guided isolation to find candidate active molecules from herbs. We establish a process of anti-viral sesquiterpene networking. An analysis of the networking suggested that new anti-HBV sesquiterpene may be attributable to eudesmane-, guaiane-, cadinane-, germacane- and bisabolane-type sesquiterpenes. In order to evaluate the efficiency of the structure-based molecular networking, ethanol extract of Saussurea lappa (Decne.) C.B Clarke was investigated, which led to the isolation of two guaiane-type (1 and 14), ten eudesmane-type (2–5 and 8–13), two chain (6 and 7) and one germacrane-type (15) sesquiterpenes, including seven new ones, lappaterpenes A–G (1–7), which are reported on herein. The absolute configurations of the new compounds were established by coupling constants, calculated ECD and ROESY correlations, as well as comparisons of optical rotation values with those of known compounds. The absolute configuration of compound 2 was further confirmed by X-ray diffraction. Compounds 1–15 were evaluated for their potency against hepatitis B virus. Compounds 4, 6, 7 and 9 showed effect on HBsAg with inhibition ratios of more than 40% at 30 μM concentrations. Compounds 14 and 15 inhibited HBsAg secretion with the values of IC₅₀ 0.73 ± 0.18 and 1.43 ± 0.54 μM, respectively. Structure-based molecular networking inspired the discovery of target compounds.

1. Introduction

Natural product structures play a significant role in drug discovery and development [1,2]. Bioactive natural products offer opportunities to discover novel targets and mechanisms for treating human diseases [3]. However, there is a lack of effective methods to find target compounds. Previously, ethnopharmacological knowledge or screening of extract for bioactivity and bioassay-guided isolation have inspired the discovery of active natural products [4,5]. In recent years, bioactivity-based molecular networking has significantly increased the efficiency with which active natural products as potential drug leads have been discovered [6,7].

As part of an ongoing effort to find anti-viral sesquiterpenes in herbs in Yunnan province in China [8,9,10], herein we establish a new strategy, i.e., structure-based molecular networking, to investigate antiviral sesquiterpenes. Firstly, we collected antiviral sesquiterpenes from ZINC and CHEMBL. Secondly, based on the skeletons of the sesquiterpenes, we calculated the degrees of similarity of those sesquiterpenes by ChemmineR and ChemmineBO [11]. Lastly, we constructed a structure-based molecular network and divided the sesquiterpenes into communities by cluster_louvain from igraph [12], which led to the identification of key nodes (i.e., representative sesquiterpenes) in sesquiterpene communities. A cluster analysis of the sesquiterpene network suggested that new anti-HBV sesquiterpenes may be attributable to eudesmane- (degree = 27), guaiane- (degree = 21), cadinane- (degree = 20), germacane- (degree = 18) and bisabolane-type (degree = 16) sesquiterpenes (Table S1). Previous studies showed that bisabolane-type sesquiterpenes have good anti-HBV activity [13].

In order to evaluate the efficiency of the structure-based molecular networking, ethanol extract of Saussurea lappa (Decne.) C.B Clarke was investigated. S. lappa belongs to Saussurea, a large genus of the Asteraceae family including more than 400 species distributed worldwide [2], more than 40 of which have been used in traditional and alternative medicine [14]. Nowadays, the plant mainly is cultivated in Yunnan province in China. The roots of S. lappa are recorded in Chinese Pharmacopoeia (2020 Edition) [15]. It is also a common Tibetan medicine and has been used to treat stomach pain and blood disorders. In Mongolian medicine, the roots of S. lappa have been used to treat lung abscesses and phlegm [16]. Some eudesmane-, guaiane- and germacane-type sesquiterpenes have been isolated from S. lappa. These compounds showed potential effects on chronic superficial gastritis, ulcer, cancer, bacterial, fungal and viral diseases [16,17,18].

Herein we report seven new sesquiterpenes from the roots of S. lappa, namely, lappaterpenes A–G (1–7) (Figure 1), as well as eight known sesquiterpenes, 11-hydroxy-4α-methoxyselinane (8) [19], ilicic aldehyde (9) [20], 4α-hydroxy-4β-methyldihydrocostol (10) [21], arbusculin E methyl ester (11) [22], 5α-hydroxy-β-costol (12) [23], arbusculin A (13) [24,25], dehydrocostus lactone (14) [24], and costunolide (15) [26]. Through X-ray diffraction and calculated ECD, the absolute configurations of Compounds 1−5 were confirmed. The absolute configurations of other two new compounds (6 and 7) were established by comparing the optical rotation values with those of known compounds, as well as coupling constants and ROESY correlations. All compounds were tested for anti-HBV activities. Compounds 4, 6, 7 and 9 showed effect on HBsAg, with inhibition ratios of more than 40% at 30 μM concentrations. Compounds 14 and 15 inhibited HBsAg secretion at IC₅₀ values of 0.73 ± 0.18 and 1.43 ± 0.54 μM, respectively. Compounds 14 and 15 also showed potential cytotoxic activity in HepG2, while Compounds 4, 6, 7, and 9 showed no cytotoxic activity at 30 μM. Structure-based molecular networking inspired the discovery of active natural products.

2. Results

2.1. Development of Structure-Based Molecular Networking

ChemmineR and ChemmineOB were applied to calculate similarities among sesquiterpenes. When a threshold value of similarity is 90%, similar sesquiterpenes can be identified and clustered by cluster_louvain to categorize the compounds into different communities. Key nodes with the highest degree values, and other similar nodes, are displayed as square nodes (Figure 1A). The top ten square nodes are listed in Figure 1A and Table S1. In the molecular networking, anti-HBV sesquiterpenes (Table S2) were extracted from Figure 1A; in Figure 1B, nodes 21 (Guaiane, degree = 21), 45 (Eudesmane, degree = 27) and 54 (Germacrane, degree = 18) are key nodes. Node 21 represents guaiane, located in the green community in Figure 1A; its community and related sesquiterpenes are illustrated in Figure 1C. Node 45 represents eudesmane, located in the red community in Figure 1A; its community and related sesquiterpenes are illustrated in Figure 1D. Node 54 represents germacrene, located in the yellow community in Figure 1A; related sesquiterpenes are illustrated in Figure 1E. The sources of germacrene-, guaiane- and eudesmane-type sesquiterpenes were searched in our database, which showed that these three kinds of sesquiterpene skeletons were simultaneously enriched in Saussurea spp. (Table S2).

2.2. Evaluation of Workflow Procedure

To evaluate the efficiency of the structure-based molecular networking method, ethanol extract of S. lappa was investigated, which led to the isolation of seven new sesquiterpenes, as well as eight known sesquiterpenes (shown in Figure 2). The known compounds were identified as 4α-methoxyselinane (8) [19], ilicic aldehyde (9) [20], 4α-hydroxy-4β-methyldihydrocostol (10) [21], arbusculin E methyl ester (11) [22], 5α-hydroxy-β-costol (12) [23], arbusculin A (13) [25], dehydrocostus lactone (14) [24], and costunolide (15) [26], respectively. By X-ray diffraction and calculated ECD, the absolute configurations of Compounds 1–5 were confirmed, while those of Compounds 6 and 7 were established by comparing optical rotation values with those of known compounds, as well as coupling constants and ROESY correlations.

Compound 1, a pale-yellow oily substance, ${[\mathsf{\alpha }]}_{\mathrm{D}}^{23.1}$ −30.1 ± 0.12 (c 0.3, MeOH), had the molecular formula of C₁₄H₂₀O₂ (∆mmu 0.13), as deduced from the (+) HRESIMS (Calcd. for C₁₄H₂₀O₂Na, m/z 243.2699), ¹³C NMR and HSQC data. The spectrum for ¹H NMR (MeOH-d₄) (shown in

Table 1. ¹H and ¹³C NMR data for Compounds 1 and 2 (MeOH-d₄), 600 MHz, J in Hz).

	1		2
Position	δC	δH (J in Hz)	δC	δH (J in Hz)
1a	47.1 (CH)	2.86 (ddd, 8.5, 6.2, 2.1)	41.8 (CH2)	1.36 a
1b	-	-	-	1.09 b
2a	28.5 (CH2)	1.78 (ddd, 12.9, 9.8, 8.5)	17.9 (CH2)	1.36 a
2b	-	2.07 (ddd, 12.9, 6.2, 3.6)	-	1.88 (m)
3a	30.3 (CH2)	2.40 (ddd, 17.9, 9.8, 3.6)	40.9 (CH2)	1.66 (brd, 15.9)
3b		2.43 b	-	1.36 a
4	151.7 (C)		71.0 (C)	-
5	54.7 (CH)	2.71 a	52.1 (CH)	1.09 b
6a	69.5 (CH)	3.55 (t, 9.6)	26.3 (CH2)	1.44 (d, 13.1)
6b	-	-	-	1.74 (ddd, 13.1, 4.1, 2.0)
7	61.4 (CH)	2.71 a	42.3 (CH)	2.02 (dd, 12.2, 4.1)
8a	31.1 (CH2)	1.94 (dddd, 13.6, 5.9, 4.5, 3.1)	27.4 (CH2)	1.51 (brdd, 12.7, 3.8)
8b		1.51 (dtd, 13.6, 11.9, 3.4)	-	1.89 (m)
9a	34.3 (CH2)	2.12 (ddd, 12.3, 4.5, 3.4)	43.9 (CH2)	1.34 (dd, 12.8, 3.8)
9b	-	2.43 b	-	1.18 (dd, 12.8, 4.2)
10	151.9 (C)	-	33.6 (C)	-
11	212.9 (C)	-	154.2 (C)	-
12	-	-	63.8 (CH2)	4.06 (brd, 1.3)
13a	27.8 (CH3)	2.18 (s)	106.6 (CH2)	5.02 (d, 2.0)
13b	-	-	-	4.92 (d, 2.0)
14a	109.5 (CH2)	5.04 (d, 2.6)	17.9 (CH3)	1.13 (s)
14b	-	5.07 (d, 2.6)	-
15a	110.0 (CH2)	4.88 (d, 1.8)	29.1 (CH3)	1.07 (s)
15b		4.77 (d, 1.8)

J-values are in parentheses and are reported in Hz; chemical shifts are given in ppm; assignments were confirmed by COSY, 1D HSQC, and HMBC experiments. ^a Overlapped with each other; ^b Overlapped with each other.

) revealed one singlet methyl (δ_H 2.15) and two terminal olefinic protons (δ_H 5.01, 5.04, 4.77 and 4.86). The ¹³C NMR (MeOH-d₄) and HSQC spectra (showed in Table 1) suggested 14 carbons including one methyl (δ_C 27.8), six methylenes including two alkenyls (δ_C 109.5 and 110.0), four methines (δ_C 47.1, 54.7, 61.4, and 69.5) and three quaternary carbons including a ketone carbonyl (δ_C 212.9) and two alkenyls (δ_C 151.7 and 151.9). The aforementioned NMR data of 1 were similar to those of guaiane-type sesquiterpene watsonol A [27]. However, they presented additional signals of methyl ketone. Furthermore, HMBC correlations (Figure 3) of the methyl (δ_H 2.18, Me-13) to the additional carbonyl (δ_C 212.9, C-11) and methine (δ_C 61.4, C-7) confirmed the additional methyl ketone linked to C-7 of 1, as well as HMBC correlations of alkenyl (δ_H 5.04, 5.07, H-14) to C-1/C-9/C-10, alkenyl (δ_H 4.88, H-15) to C-3/C-4/C-5 (shown in Figure 3). Thus, the planar structure of 1 was assigned as shown in Figure 2. The relative configuration was further determined by ROESY correlations. The ROESY correlations of H-7/H-1 (shown in Figure 3) indicated that H-1, H-5 and H-7 were α-oriented, while the ROESY correlations of H-6/Me-13 (shown in Figure 3) indicated that HO-6 were α-oriented. In order to further clarify the absolute configuration of 1, the low energy conformers of the compound were optimized by applying the DFT method at the B3LYP/6-31G(d) level (in MeOH). The results showed that the calculated ECD spectrum (shown in Figure 4) of (4R, 5R, 6R, 10R)-1 agreed well with experimental data, which indicated that the absolute configuration of 1 was 4R, 5R, 6R, 10R. As a result, the structure of 1 was assigned and the compound was named lappaterpene A.

Compound 2, a colorless needle crystal, ${[\mathsf{\alpha }]}_{\mathrm{D}}^{23.2}$ 12.2 ± 0.31 (c 0.15, MeOH), had the molecular formula of C₁₅H₂₆O₂ (∆mmu 0.20), as deduced from the HRESIMS (Calcd. for C₁₅H₂₆O₂Na, m/z 261.1826), ¹³C NMR and HSQC data. The ¹H NMR (MeOH-d₄) data for 2 (shown in Table 1) showed two singlet methyls (δ_H 1.13, 1.07) and a terminal olefinic proton (δ_H 5.02, 4.91). The ¹³C NMR (MeOH-d₄) data for 2 (shown in Table 1) suggested 15 carbons including two methyls (δ_C 29.1 and 17.9), eight methylenes including an oxymethylene (δ_C 63.8) and a terminal olefin (δ_C 106.6), two methines (δ_C 52.1 and 42.3) and three quaternary carbons including one oxy-quaternary carbon (δ_C 71.0) and one alkenyl (δ_C 154.2). The aforementioned NMR data of 2 were similar to those of 4α-hydroxy-4β-methyldihydrocostol (10) [21], an eudesmane sesquiterpene; however, they differed in that C-2, C-3, C-4 and C-5 of Compound 2 were shifted to high field by 1.8, 1.8, 0.5 and 2.3 ppm, respectively, and C-1 and C-15 were shifted to low field by 1.0, and 7.9 ppm, respectively. The above data suggest that Compound 2 may be an epimer of Compound 10. Furthermore, HMBC correlations of methyl (δ_H 1.07, Me-15) to C-3/C-4/C-5, methyl (δ_H 1.13, Me-14) to C-1/C-5/C-9/C-10 and methylene (δ_H 4.06, CH₂-13) to C-7/C-11/C-12 (shown in Figure 5), as well as other key HMBC related signals, confirmed the planar structure of Compound 2 (shown in Figure 5). The relative configuration of 2 was established using the ROESY spectrum. The ROESY correlation of H-5 with Me-15/H-7 indicated that Me-15 and H-7 were α-oriented. (shown in Figure 5). Furthermore, a crystal X-ray diffraction experiment with Cu Kα radiation further allowed the unambiguous assignment of the absolute configuration of 2 (shown in Figure 6) (deposition No. CCDC: 2061652) as 4S, 5R, 7R, 10R [the Flack parameter is 0.03 (5) and the Hooft parameter is 0.10 (4) for 2102 Bijvoet pairs] [28,29]. Compound 2 was only reported as a bioconversion product by PCT patent plant enzymes [30]. However, its absolute configuration was not determined and detailed NMR data were unavailable. Herein we report compound 2 as a new natural product which we named lappaterpene B.

Compound 3, a pale-yellow oily substance, ${[\mathsf{\alpha }]}_{\mathrm{D}}^{22.2}$ −58.4 ± 0.10 (c 0.15, MeOH), had the molecular formula of C₁₆H₂₈O₂ (∆mmu 0.01), as deduced from the HRESIMS (Calcd. for C₁₆H₂₈O₂Na, m/z 275.2012), ¹³C NMR and HSQC data. The ¹H and ¹³C NMR (MeOH-d₄) data for 3 (shown in

Table 2. ¹H and ¹³C NMR data for Compounds 3–5 (MeOH-d₄, 600 MHz, J in Hz).

	3		4		5
Position	δC	δH (J in Hz)	δC	δH (J in Hz)	δC	δH (J in Hz)
1a	40.7 (CH2)	1.38 a	40.7 (CH2)	1.50–1.36 a	40.6 (CH2)	1.44 (m)
1b	-	1.09 b	-	1.10 b	-	1.22 (brdd, 12.2, 5.3)
2a	19.3 (CH2)	1.65 c	19.4 (CH2)	1.56 c	19.3 (CH2)	1.59 b
2b	-	-	-	-	-	-
3a	36.2 (CH2)	1.77 e	36.9 (CH2)	1.77 (ddd, 12.2, 3.2, 1.5)	35.9 (CH2)	1.50–1.43 a
3b	-	1.44 d	-	1.50–1.36 a	-	1.72 (ddd, 12.2, 3.3, 1.5)
4	76.4 (C)	-	76.3 (C)	-	76.4 (C)
5	51.1 (CH)	1.38 a	51.2 (CH)	1.50–1.36 a	50.5 (CH)	1.50–1.43 e
6a	26.3 (CH2)	1.77 e	26.3 (CH2)	1.50–1.36 a	26.2 (CH2)	1.22 (m)
6b	-	1.22 (m)	-	1.21 (m)	-	1.81 (m)
7	41.9 (CH)	2.08 (ddd, 15.5, 10.5, 3.8)	41.9 (CH)	2.03 (ddd, 12.9, 4.1, 2.2)	40.3 (CH)	2.49 (m)
8a	27.2 (CH2)	1.65 c	27.3 (CH2)	1.56 c	27.2 (CH2)	1.59 b
8b	-	1.40 (brdd, 12.3, 2.6)	-	1.50–1.36 a	-	1.50–1.43 a
9a	45.0 (CH2)	1.25 (brdd, 13.2, 3.9)	45.1 (CH2)	1.26 (brdd, 13.5, 4.4)	44.9 (CH2)	1.32 (m)
9b	-	1.44 d	-	1.50–1.36 a	-	1.50–1.43 a
10	34.3 (C)	-	34.3 (C)	-	34.2 (C)	-
11	154.1 (C)	-	154.1 (C)	-	154.1 (C)	-
12	63.8 (CH2)	4.05 (brd, 1.4)	63.8 (CH2)	4.07 (brs)	169.8 (C)	-
13a	106.5 (CH2)	5.01 (d, 1.6)	106.5 (CH2)	5.04 (d, 1.6)	120.5 (CH2)	5.52 (brs)
13b	-	4.88 (d, 1.4)	-	4.91 (d, 1.6)	-	6.07 (brs)
14a	18.2 (CH3)	1.09 b	18.3 (CH3)	0.97 (s)	18.3 (CH3)	0.97 (s)
14b	-	-	-	-	-
15a	17.6 (CH3)	0.97 (s)	18.4 (CH3)	1.10 b	18.1 (CH3)	1.08 (s)
15b
OCH3	46.6	3.20 (s)	-	-	46.5	3.14 (s)
CH2			54.6 (CH2)	3.38 (m)
			-	3.43 (m)
CH3			15.2 (CH3)	1.09 (t, 8.3)

J-values are in parentheses and reported in Hz; chemical shifts are given in ppm; assignments were confirmed by COSY, 1D HSQC, and HMBC experiments. ^a Overlapped with each other; ^b Overlapped with each other; ^c Overlapped with each other; ^d Overlapped with each other; ^e Overlapped with each other.

) were similar to those of 4α-hydroxy-4β-methyldihydrocostol (10) [21], but the difference is the molecular weight, which has an extra methoxy group. Furthermore, HMBC correlations of additional methoxyl (δ_H 3.20, OCH₃) to C-4 (δ_C 76.4) further confirmed the substituted position of the methoxyl, as well as the HBMC correlations of methyl (δ_H 0.97, H₃-15) to C-3/C-4/C-5, methyl (δ_H 1.08, H₃-14) to C-1/C-5/C-9/C-10, and methylene (δ_H 5.01, 4.88, H₂-13) to C-7/C-11/C-12 (shown in Figure 7). Thus, the planar structure of 3 was assigned as shown in Figure 2. The relative configuration was further determined by ROESY correlations of Me-1/Me-15 and H-7/H-5 (shown in Figure 7). In order to evaluate Compound 3, we obtained 4α-hydroxy-4β-methyldihydrocostol (10) crystals, a major compound of Saussurea spp. A crystal X-ray diffraction experiment with Cu Kα radiation further confirmed the absolute configuration of 10 (deposition No. CCDC: 2026706) (shown in Figure 8) as 4R, 5R, 7R, 10R [the Flack parameter is −0.01 (7); the Hooft parameter was 0.10 (4) for 2102 Bijvoet pairs] [28,29]. On this basis, the absolute configuration of 3 was shown to be 4R, 5R, 7R, 10R. In order to further clarify the absolute configuration of 3, the low energy conformers of the compound were optimized by applying the DFT method at the B3LYP/6-31G(d,p) level (in MeOH). The results showed that the calculated ECD spectrum (shown in Figure 9) of (4R, 5R, 7R, 10R)-3 agreed well with the experimental data. The aforementioned data allowed the unambiguous assignment of 3 as 4R, 5R, 7R, 10R. As a result, the structure of 3 was assigned and the compound was named lappaterpene C.

Compound 4, a pale-yellow oily substance, ${[\mathsf{\alpha }]}_{\mathrm{D}}^{22.6}$ −24.7 ± 0.07 (c 0.25, MeOH), had the molecular formula of C₁₇H₃₀O₂ (∆mmu 0.20), as deduced from the HRESIMS (Calcd. for C₁₇H₃₀O₂Na, m/z 289.3594), ¹³C NMR and HSQC data. The ¹H and ¹³C NMR (MeOH-d₄) data for 4 (shown in Table 2) were similar to those of Compound 3, but differed in molecular weight and through the presence of an additional ethoxyl group. HMBC correlations of oxy-methylene (δ_H 3.38, 3.43, CH₂) to C-4 (δ_C 76.3) and (δ_H 1.11, Me) to oxy-methylene (δ_C 54.6, CH₂) further confirmed the substituted position of the ethyoxyl, as well as the other key HBMC correlations (shown in Figure 10). Thus, the planar structure of 4 was assigned as shown in Figure 1. The relative configuration was further determined by ROESY correlations of H-5/H-7 and Me-15/Me-14 (shown in Figure 10). In order to further clarify the absolute configuration of 4, the low energy conformers of the compound were optimized by applying the DFT method at the B3LYP/6-31G(d,p) level (in MeOH). The results showed that the calculated ECD spectrum (shown in Figure 11) of (4R, 5R, 7R, 10R)-4 agreed well with the experimental one, indicating that the absolute configuration of 4 was 4R, 5R, 7R, 10R. As a result, the structure of 4 was assigned and the compound was named lappaterpene D.

Compound 5, a pale-yellow oily substance, ${[\mathsf{\alpha }]}_{\mathrm{D}}^{22.8}$ −29.4 ± 0.11 (c 0.15, MeOH), had the molecular formula of C₁₆H₂₆O₃ (∆mmu 0.03), as deduced from the HRESIMS (Calcd. for C₁₆H₂₆O₃Na, m/z 289.1780), ¹³C NMR and HSQC data. The ¹H NMR and ¹³C NMR (MeOH-d₄) data for 5 (shown in Table 2) were similar to those of Compound 3, but differed in molecular weight and through the presence of additional carbonyl group. HMBC correlations of methylene (δ_H 5.52, 6.07, C-13) to additional carbonyl (δ_C 169.8, C-12) and methine (δ_C 40.3, C-7) confirmed the additional carbonyl linked to C-11 of 5 (shown in Figure 12). The ROESY correlations of H-5 with H-7 (shown in Figure 12) revealed the relative configuration of 5. In order to further clarify the absolute configuration of 5, the low energy conformers of 5 were optimized by applying the DFT method at the B3LYP/6-31G(d,p) level (in MeOH). The results showed that the calculated ECD spectrum (shown in Figure 13) of (4R, 5R, 7R, 10R)-5 agreed well with the experimental one, indicating the absolute configuration of 5 to be 4R, 5R, 7R, 10R. As a result, the structure of 5 was assigned and the compound was named lappaterpene E.

Compound 6, a pale-yellow oily substance, ${[\mathsf{\alpha }]}_{\mathrm{D}}^{22.5}$ +0.32 ± 0.19 (c 0.15, MeOH), had the molecular formula of C₁₅H₂₆O₂ (∆mmu 0.21), as deduced from the HRESIMS (Calcd. for C₁₅H₂₆O₂Na, m/z 261.2774), ¹³C NMR and HSQC data. The ¹H NMR (MeOH-d₄) data for 6 (shown in

Table 3. ¹H and ¹³C NMR data for Compounds 6 and 7 (in MeOH-d₄).

6				7
Position	Type	δC	δH (J in Hz)	Position	Type	δC	δH (J in Hz)
1	CH2	113.0	5.16 (dd, 10.9, 1.8)	1	CH2	65.1	3.58 (dd, 11.3, 4.1)
			5.29 (dd, 17.4, 1.8)				3.45 (dd, 11.3, 7.5)
2	CH	141.2	5.88 (dd, 17.4, 10.9)	2	C	149.0
3	C	75.4		3	CH	75.2	4.10 (dd, 7.5, 4.1)
4	CH2	36.5	2.54 (m)	4	CH2	32.0	2.11 (m)
5	CH2	21.5	2.06 (m)				2.03 (m)
			1.98 (m)	5	CH2	26.2	2.08 (m)
6	CH	124.4	5.12 (m)				1.98 (m)
7	C	134.5		6	CH	124.0	5.15 (td, 7.0, 1.3)
8	CH2	39.5	1.98 (m)	7	C	134.9
9	CH2	26.4	2.06 (m)	8	CH2	39.5	2.00 (m)
10	CH	124.0	5.08 (tdd, 5.7, 2.9, 1.5)	9	CH2	26.3	2.06 (m)
11	C	130.7					2.18 (m)
12	CH2	68.2	3.45 (dd, 10.8, 16.8)	10	CH	123.9	5.09 (td, 12.8, 2.9)
13	CH3	14.6	1.59 (s)	11	C	130.7
14	CH3	16.4	1.59 (s)	12	CH2	109.7	5.09 (s)
15	CH3	24.5	1.66 (s)				4.91 (s)
				13	CH3	14.7	1.62 (s)
				14	CH3	16.4	1.61 (s)
				15	CH3	24.5	1.68 (s)

) showed three methyls (δ_H 1.59, 1.59, 1.66) and five olefinic protons (δ_H 5.16, 5.29, 5.88, 5.12, 5.08). The ¹³C NMR (MeOH-d₄) data for 6 (shown in Table 3) suggested 15 carbons including three methyls (δ_C 14.6, 16.4, 24.5), five methylenes (δ_C 36.5, 21.5, 39.5, 26.4, and 68.2), one quaternary carbon (δ_C 75.4) and three sets of alkenyls (δ_C 113.0, 141.2, 124.4, 134.5, 124.0, and 130.7). The aforementioned NMR data combined with mass suggested that Compound 6 was a chain-like sesquiterpene derivative. Furthermore, ¹H-¹H COSY correlations of H-4/H-5/H-6 and H-8/H-9/H-10 connected two molecular fragments (shown in Figure 14), as well as HMBC correlations of methyl (δ_H 1.59, H-13) to C-6/C-7/C-8, methyl (δ_H 1.59, H-14) to C-10/C-15, methylene (δ_H 3.45, C-12) to C-3, the methyl (δ_H 1.66, C-15) to C-11, methylene (δ_H 5.16, 5.29, C-1) to C-2/C-3 and methane (δ_H 5.12, H-6) to C-4 (shown in Figure 14). The ROESY correlations of H-6 with CH₂-8 indicated that the double bond of in the molecule was a cis bond. Furthermore, since the crystal and suitable CD spectrum could not be obtained, we compared the optical rotation value of Compound 6 with that of (2R,6E)-5,9-undecadiene-1,2-diol, 2,6,10-trimethyl (${[\mathsf{\alpha }]}_{\mathrm{D}}^{23.4}$ +6) [31]. The result suggested that the absolute configuration of the C-3 was R. As a result, the structure of 6 was assigned and the compound was named lappaterpene F.

Compound 7, a pale-yellow oily substance, ${[\mathsf{\alpha }]}_{\mathrm{D}}^{23.4}$ +3.18 ± 0.09 (c 0.25, MeOH), had a molecular formula of C₁₅H₂₆O₂ (∆mmu 0.04), as deduced from the HRESIMS (Calcd. for C₁₅H₂₆O₂Na, m/z 261.3211), ¹³C NMR and HSQC data. The ¹H NMR (MeOH-d₄) data for 7 (shown in Table 3) showed three methyls (δ_H 1.62, 1.61, 1.68) and three alkenyls including a terminal olefin (δ_H 5.09, 4.91). The ¹³C NMR (MeOH-d₄) data for 7 (shown in Table 3) suggested 15 carbons including three methyls (δ_C 14.7, 16.4, 24.5), five methylenes including one oxymethylene (δ_C 65.1), one oxymethine (δ_C 75.2) and three sets of alkenyls (δ_C 109.7, 149.0, 124.0, 134.9, 123.9, 130.7). The aforementioned 1D NMR data combined with mass suggested that Compound 7 was a chain-type sesquiterpene derivative. Furthermore, ¹H-¹H COSY correlations of H-3/H-4/H-5/H-6 and H-8/H-9/H-10 to connect two molecular fragments (shown in Figure 15). In addition, HMBC correlations were determined of methylene (δ_H 5.09 and 4.91, H-12) to C-1/C-2/C-3, methine (δ_H 4.10, H-3) to C-4/C-5, methyl (δ_H 1.62, H-13) and methylene (δ_H 2.08 and 1.98, H-5) to C-6/C-7, methylene (δ_H 2.00, H-8) to C-10, methyl (δ_H 1.61, H-14) and methyl (δ_H 1.68, H-15) to C-11 (shown in Figure 15). Thus, the planar structure of 7 was assigned as shown in Figure 2. However, neither the crystal form not a suitable CD spectrum of 7 could be obtained. It is of note that Compound 7 shared same chiral centers with those of lysine [27]. The optical rotation value of Compound 7 with the value of +3.18 was similar to that of L-lysine, with the value of +10.2, but not to that of D-lysine, with the value of −12.8 [27]. As a result, the absolute configuration of the hydroxyl at the C-3 was speculated to be S. The structure of 7 was assigned and the compound was named lappaterpene G.

2.3. Anti-HBV Activities and SARs of Sesquiterpene Derivatives

S. lappa is a famous medicinal plant growing in the Himalayan region. It is now mainly is cultured in Yunnan and Sichuan provinces in China. The roots of S. lappa have been used to treat viral diseases in Ayurveda, Unani and Siddha as well as in traditional Chinese medicine. The plants have also been used in Tibet and other minority regions in China. Herein we reported the isolation and determination of seven new sesquiterpenes (1–7), together with eight known ones (8–15). All compounds were tested for their anti-HBV and cytotoxic activities. An anti-HBV assay suggested that known compounds dehydrocostus lactone (14) and costunolide (15) showed potent effects on HBsAg and HBeAg (displayed in Table 3), which agreed with the preliminary screening results. Active compounds (14 and 15) featured a α,β-unsaturated-lactone, and both showed inhibition toward both HBsAg and HBeAg. However, it was of note that the eudesmane sesquiterpene (13) bearing with α,β-unsaturated-lactone showed no activity against HBV at 30 µM, while new eudesmane sesquiterpenes 4 and 9, in which α,β-unsaturated-lactone was broken, showed some effect on HBsAg, with an inhibition ratio of more than 40% at 30 μM. Although further research will be required to evaluate the mechanism of HBV inhibition and the structure–activity relationships (SARs) of these compounds, the results suggest that germacrane and guaiane sesquiterpenes may be the anti-HBV active chemical constituents of S. lappa. This research is the first to report that eudesmane sesquiterpenes without α,β-unsaturated-lactone show moderate anti-HBV activity.

3. Discussion

The loss of the bioactive compounds during isolation is a common problem in natural product research. Herein we present the concept of structure-guided isolation to find candidate active molecules directly from herbs. We established a library of antiviral sesquiterpenes which included structures, skeleton type, bioactivities and network of sesquiterpenes relationships to illustrate key node (i.e., representative sesquiterpenes). By employing similarity calculations, we constructed a sesquiterpene molecular network and categorized molecular clusters. Then, we used the bioactivity characteristics to filter the molecular clusters and predict the bioactivity of the sesquiterpenes in the clusters. We applied this workflow to discover antiviral compounds from an extract of S. lappa. It can be expected that this approach, i.e., structure-based molecular networking, will lead to the discovery of active natural products. It can be based on biologically active molecular networks for the analysis of bioassay-guided separation

4. Materials and Methods

4.1. General Experimental Techniques

Optical rotations were measured with a Jasco P-1020 digital polarimeter. IR spectra were measured on a Thermo NICOLET iS10 with KBr pellets. UV spectra were recorded on a Shimadzu UV-2700 spectrophotometer. CD spectra were measured on an Applied Photophysics Chirascan instrument. X-ray diffraction was measured on a Bruker D8 Quest instrument. ESIMS and HRESI-MS were run on an Agilent 1290 UPLC spectrometer and Agilent 6500 series Q-TOF spectrometer, respectively. NMR spectra were measured in CD₃OD solution and recorded on a Bruker Avance III HD-600 or AV 800 spectrometer at 25 °C, using TMS as an internal standard. Chemical shifts were reported in units of δ (ppm), and coupling constants (J) were expressed in Hz. Column chromatography (CC) was carried out over silica gel (200–300 or 500–800 mesh, Qingdao Marine Chemical Factory), Sephadex LH-20 (25–100 μm, Pharmacia Fine Chemical Co., Ltd., Tokyo, Japan), MCI-gel CHP-20P (75–150 µm, Mitsubishi Chemical Industry, Ltd., Guangzhou, China), Rp-18 (40–63 μm, Merck, Shanghai, China). Precoated silica gel plates (Qingdao Haiyang Chemical Co., Qingdao, China) were used for TLC. Detection was done under UV light (254 nm and 365 nm) and by spraying the plates with 10% sulfuric acid followed by heating. A Waters 1525/2998 liquid chromatography machine (Waters Technologies, Wexford, Ireland) was used for HPLC. An ACE C₁₈-PFP and Waters sunfire-C₁₈ column 5 μm 143 Å column (250 mm × 10 mm) were used for semipreparative HPLC separations.

4.2. Sesquiterpene Network

A dataset of 11,741 sesquiterpenes was collected from the ChEMBL database, Binding DB, and publications (Supplementary Material). Duplicate compounds were removed. Among them, anti-HBV activity was represented according to IC₅₀ value; 152 compounds (agonists) in the data set were active (IC₅₀ < 10 µM), and were tagged with “1”. Other compounds were marked with “0” (inactive).

Skeleton type was determined for each sesquiterpene. Similarities among sesquiterpenes were calculate by the ChemmineR and ChemmineBO to afford a matrix of similarity. Based on a threshold value of similarity (90%), a network of sesquiterpenes was generated with “ggnet” in R, and nodes in the network were divided into different communities by cluster_louvain.

4.3. Plant Material

The roots of S. lappa were collected in Lijiang county, Yunnan province, China, and identified by Associate Professor Wu Zhikun (School of Pharmacy, Guizhou University of Chinese Medicine). A voucher (KUMST-BS-0007) specimen was deposited in the Laboratory of Chemical Biology for Natural Medicines, School of Life Science and Technology, Kunming University of Science and Technology.

4.4. Extraction and Isolation

The air-dried and powdered roots of S. lappa (119.0 kg) were extracted by EtOH at room temperature. After removal of the solvent under reduced pressure, the crude extract (13.4 kg) was suspended in EtOAc (10 L) and partitioned with H₂O (5 × 10 L). The EtOAc extract (4.4 kg) was subjected to MCI-gel CHP-20P column, eluting with a MeOH–H₂O (40–100%) to give fourteen fractions, i.e., A–N. Fr.K (53.7 g) was subjected to a MCI-gel CHP-20P column using a step gradient of MeOH–H₂O (35–100%) to afford eighteen fractions. Fr.K1–Fr.K18. Fr.K11 (19.3 g) was separated by RP-18 using a step gradient of MeOH–H₂O (40–100%) to afford nine subfractions, i.e., Fr.K11a–Fr.K11i. Fr.K11d (7.3 g) was subjected to a silica gel column using a step gradient of n-hexane–EtOAc (96:4 to 80:20) to afford fourteen fractions, i.e., Fr.K11d1–Fr.K11d14. Fr.K11d4 (221.0 mg) was separated by RP-18 using a step gradient of MeOH–H₂O (30–45%) to afford seven subfractions, i.e., Fr.K11d4a–Fr.K11d4g. Fr.K11d4c (69.9 mg) was purified with p–HPLC (CH₃CN–H₂O, 30:70, v/v) to afford 1 (25.7 mg). Fr.K12 (12.5 g) was separated by RP-18 using a step gradient of MeOH–H₂O (50–100%) to afford nine subfractions, i.e., Fr.K12a–Fr.K12i. Fr.K12e (1.7 g) was subjected to a silica gel column using a step gradient of n-hexane–EtOAc (96:4 to 80:20) to afford six fractions, i.e., Fr.K12e1–Fr.K12e6. Fr.K12e3 (483.0 mg) was purified with p–HPLC (CH₃CN–H₂O, 52:48, v/v) to afford 9 (0.5 mg), 8 (5.3 mg). Fr.K12f (800.5 mg) was subjected to a silica gel column using a step gradient of n-hexane–EtOAc (96:4 to 80:20) to afford eleven fractions, i.e., Fr.K12f1–Fr.K12f11. Fr.K12f4 (23.8 mg) was purified with p–HPLC (CH₃CN–H₂O, 45:55, v/v) to afford 3 (10.6 mg), 5 (4.7 mg). Fr.K12f6 (22.1 mg) was purified with p–HPLC (CH₃CN–H₂O, 45:55, v/v) to afford 6 (12.2 mg). Fr.K12f8 (72.5 mg) was purified with p–HPLC (CH₃CN–H₂O, 45:55, v/v) to afford 7 (54.5 mg). Fr.K12g (494.1 mg) was subjected to a silica gel column using a step gradient of petroleum ether–EtOAc (97:3 to 91:9) to afford eight fractions, i.e., Fr.K12g1–Fr.K12g8. Fr.K12g4 (106.8 mg) was purified with p–HPLC (CH₃CN–H₂O, 60:40, v/v) to afford 4 (87.2 mg). Fr.I (88.7 g) was separated by RP–18 using a step gradient of MeOH–H₂O (30–100%) to afford eighteen subfractions, i.e., Fr.I1–Fr.I18. Fr.I4 (734.0 mg) was separated by sephadex LH–20 with MeOH (100%) to give fourteen subfractions, i.e., Fr.I4a–Fr.I4n. Fr.I4a (87.6 mg) was separated by RP–18 using a step gradient of MeOH–H₂O (30–90%) to afford twelve subfractions, i.e., Fr.I4a1–Fr.I4a12. Fr.I4a7 (2.1 mg) was purified with p–HPLC (CH₃CN–H₂O, 34:66, v/v) to afford 11 (0.4 mg). Fr.I7 (6.7 g) was separated by sephadex LH–20 with MeOH (100%) to give eight subfractions, i.e., Fr.I7a–Fr.I7h. Fr.I7a (498.0 mg) was separated by RP–18 using a step gradient of MeOH–H₂O (30–90%) to afford fourteen subfractions, i.e., Fr.I7a1–Fr.I7a14. Fr.I7a6 (143.1 mg) was purified by p–HPLC followed with PTLC to afford 13 (2.6 mg). Fr.I7a8 (17.2 mg) was purified by p–HPLC followed with PTLC to afford 15 (0.6 mg), 14 (1.4 mg). Fr.I10 (9.8 g) was separated by RP–18 using a step gradient of MeOH–H₂O (45–90%) to afford nine subfractions, i.e., Fr.I10a–Fr.I10i. Fr.I10g was subjected to a silica gel column using a step gradient of petroleum ether–EtOAc (8:1 to 2:1) to afford 10 (5.0 g). Fr.I10e was subjected to a silica gel column using a step gradient of petroleum ether–EtOAc (12:1 to 0:1) to afford thirteen fractions, i.e., Fr.I10e1–Fr.I10e13. Fr.I10e5 (222.0 mg) was purified with p–HPLC (CH₃CN–H₂O, 40:60, v/v) to afford 2 (180.3 mg). Fr.I10e8 (534.6 mg) was purified with p–HPLC (CH₃CN–H₂O, 39:61, v/v) to afford 12 (0.7 mg).

Lappaterpene A (1): pale yellow oily substance; ${[\alpha ]}_{\mathrm{D}}^{23.1}$ −30.1 ± 0.12 (c 0.3, MeOH); UV (MeOH) λ_max (log ε): 196 (0.469) nm; IR (KBr) ν_max: 3525, 3078, 2981, 2934, 2861 cm⁻¹; ¹H and ¹³C NMR data, see Table 1; HREIMS: m/z 243.1367 [M + Na]⁺ (calcd for C₁₄H₂₀O₂Na).

Lappaterpene B (2): needle colorless crystal; ${[\alpha ]}_{\mathrm{D}}^{23.2}$ 12.2 ± 0.31 (c 0.15, MeOH); UV (MeOH) λ_max (log ε): 196 (0.260) nm; IR (KBr) ν_max: 3413, 3082, 2973, 2928, 2868 cm⁻¹; ¹H and ¹³C NMR data, see Table 1; HREIMS: m/z 261.1816 [M + Na]⁺ (calcd for C₁₅H₂₆O₂Na).

Lappaterpene C (3): pale yellow oily substance; ${[\alpha ]}_{\mathrm{D}}^{22.2}$ −58.4 ± 0.10 (c 0.15, MeOH); UV (MeOH) λ_max (log ε): 282 (0.013), 196 (0.365), 253 (0.009), 192 (0.224) nm; IR (KBr) ν_max: 3413, 3081, 2972, 2931, 2866 cm⁻¹; ¹H and ¹³C NMR data, see Table 1; HREIMS: m/z 275.2012 [M + Na]⁺ (calcd for C₁₆H₂₈O₂Na).

Lappaterpene D (4): pale yellow oily substance; ${[\alpha ]}_{\mathrm{D}}^{22.6}$ −24.7 ± 0.07 (c 0.25, MeOH); UV (MeOH) λ_max (log ε): 196 (0.260) nm; IR (KBr) ν_max: 3413, 3082, 2973, 2928, 2868 cm⁻¹; ¹H and ¹³C NMR data, see Table 1; HREIMS: m/z 289.3594 [M + Na]⁺ (calcd for C₁₇H₃₀O₂Na).

Lappaterpene E (5): pale yellow oily substance; ${[\alpha ]}_{\mathrm{D}}^{22.8}$ −29.4 ± 0.11 (c 0.15, MeOH); UV (MeOH) λ_max (log ε): 196 (0.636) nm; IR (KBr) ν_max: 3436, 3103, 2972, 2932, 2867 cm⁻¹; ¹H and ¹³C NMR data, see Table 1; HREIMS: m/z 289.1780 [M + Na]⁺ (calcd for C₁₆H₂₆O₃Na).

Lappaterpene F (6): pale yellow oily substance; ${[\alpha ]}_{\mathrm{D}}^{22.5}$ +0.32 ± 0.19 (c 0.15, MeOH); UV (MeOH) λ_max (log ε): 196 (0.413) nm; IR (KBr) ν_max: 3408, 3087, 2967, 2926, 2874 cm⁻¹; ¹H and ¹³C NMR data, see Table 2; HREIMS: m/z 261.1856 [M + Na]⁺ (calcd for C₁₅H₂₆O₂Na).

Lappaterpene G (7): pale yellow oily substance; ${[\alpha ]}_{\mathrm{D}}^{23.4}$ +3.18 ± 0.09 (c 0.25, MeOH); UV (MeOH) λ_max (log ε): 196 (0.405), 192 (0.311) nm; IR (KBr) ν_max: 3386, 3085, 2966, 2924, 2857 cm⁻¹; ¹H and ¹³C NMR data, see Table 2; HREIMS: m/z 261.1841 [M + Na]⁺ (calcd for C₁₅H₂₆O₂Na).

4.5. ECD Calculation

The aglycons of the compounds were used as the chemical models to carry out ECD calculations. A conformation analysis was carried out using molecular mechanics MMFF. The resulting conformers (<15 KJ/mol) were optimized using DFT at the B3LYP-SCRF/6-311G(d,p) level using the integral equation formalism variant of the polarizable continuum model (IEF-PCM). All the calculations were run with Gaussian 09 [32]. The free energies and vibrational frequencies were calculated at the same level to confirm their stability, and no imaginary frequencies were found. The optimized low energy conformers with energy < 2 Kcal/mol were considered for ECD calculations. The TD-DFT/B3LYP-SCRF/6-311G(d,p) method was applied to calculate the excited energies, oscillator strength and rotational strength. The excited energies and rotational strength were used to simulate ECD spectra of each conformer by introducing the Gaussian Function. The final ECD spectrum of each compound was obtained by averaging all the simulated ECD spectra of all conformers according to their excited energies and Boltzmann distribution.

4.6. Anti-HBV Activity Evaluation

HepG2.2.15 cells, a human cancer cell line, were obtained from China Center for Type Culture Collection (Wuhan, China) and maintained in supplemented with 10% fetal bovine serum (meilunbio, Dalian, China), and 380 µg/mL G418 in a humidified 5% CO₂ atmosphere at 37 °C. The inhibition to HBsAg and HBeAg was detected by ELISA.

HepG2.2.15 cells were seeded in 96-well plates and treated with compounds for 6 days. On day 3, the culture medium containing compounds was collected and replaced. The levels of HBsAg and HBeAg from cell culture supernatant were measured by HBsAg and HBeAg ELISA kits (Kehua, Shanghai, China), according to the manufacturer’s instructions. Lamivudine (Adamas, Shanghai, China) was tested as the positive control for anti HBV.

4.7. Cytotoxic Activity Evaluation

HepG2 cells were plated in 96-well plates in 100 μL medium (meilunbio, Dalian, China), to which the test samples were added at varied concentrations. After 72 h of incubation, MTT [[3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide] solution [0.5 mg/mL in phosphate buffered saline (PBS)] was added (20 μL/well) [33], and the incubation continued for another 4 h to give a formazan product. In each well, 200 μL DMSO was added after the medium had been removed. Then, the formazan product is completely dissolved by sufficient oscillation. The absorbance of the solution was measured at 490 nm using a microplate reader (Tecan, Mendov, Switzerland). MTT is reduced by dehydrogenase activities in cells to give a purple formazan dye. The amount of the formazan dye generated by dehydrogenases in cells is directly proportional to the number of living cells. Compound concentrations reducing the viability of HepG2 cells culture by 50% (CC₅₀) were calculated by regression analysis of the dose-response curves.