ent-Pimarane diterpenoids from the aerial parts of Sigesbeckia pubescens and their myocardial protective activity

Abstract

Eight new ent-pimarane diterpenoids (1–8), along with three known compounds (9–11), were isolated from the aerial parts of Sigesbeckia pubescens. The structures of the new ent-pimarane diterpenoids were established by extensive spectroscopic techniques, X-ray diffraction crystallography, ECD calculations and Mo₂(OAc)₄-induced ECD. The isolated compounds were evaluated for their myocardial protective activities in an H9c2 cell hypoxia/reoxygenation (H/R) induced myocardial ischemia-reperfusion injury model. Bioassay results showed that pubescens B (2), pubescens H (8) and darutigenol (9) enhanced the cell viability at concentrations of 1, 10, and 50 µM and molecular docking explored their binding mode with H/R connected protein Ubiquitin C-terminal hydrolase L5 (UCHL5).

---

Introduction

Herba Siegesbeckiae (HS, the aerial parts of Sigesbeckia orientalis L., S. glabrescens Makino, and S. pubescens Makino from the genus Sigesbeckia of the Asteraceae family), first recorded in Xin Xiu Ben Cao in the Tang dynasty, has the effects of detoxifying, clearing meridians, and dispelling wind and damp, and has been traditionally used for the therapy of rheumatoid arthritis, hypertension, snakebites, and malaria.¹ Also, Lin Zheng Zhi Nan Yi An mentioned that HS has been effectively applied to treat cardio-cerebrovascular diseases. For example, Xixian Tongshuan Capsule and Xinshuning Tablet with HS as the monarch herb have been listed in the Chinese Pharmacopoeia (2025 edition) for the treatment of coronary heart disease and stroke.^2,3 The pharmacological research demonstrated that the traditional use of HS for the above diseases was attributed to its anti-inflammatory,^4,5 antiallergic,⁶ angiogenesis,⁷ myocardial protective^8–10 and antioxidant activities.¹¹ Previous chemical studies showed that a total of 291 compounds have been isolated and characterized from the three Sigesbeckia species, including ent-kaurane and ent-pimarane diterpenoids, sesquiterpenoids, sterols, and flavonoids.^12–14 Among them, ent-kaurane and ent-pimarane diterpenoids are the primary bioactive constituents of HS.

With the development of mechanism and medication research to prevent myocardial H/R injury, increasingly preclinical evidence has supported the therapeutic effect of traditional Chinese medicine.^15–17 Our previous study has also found that UCHL5 can affect the NLRP3 inflammasome during myocardial H/R, and the fraction of HS can inhibit the UCHL5/NLRP3 pathway to alleviated myocardial H/R injury. Furthermore, a chemical and molecular docking experiment showed that an ent-pimarane diterpenoid, kirenol, might be the core active constituent.^18,19 In order to discover more myocardial protective compounds from HS and as part of our ongoing screening program toward new lead compound discovery from Chinese medicinal plants, further chemical investigation on the aerial parts of S. pubescens led to the isolation of eight new ent-pimarane diterpenoids (1–8) named pubescens A–H and three known analogues (9–11) (Fig. 1). We described herein the isolation and structural elucidation of 1–11. Furthermore, all isolated compounds were tested for myocardial protective activity by H9c2 cardiomyocytes hypoxia/reoxygenation (H/R) induced myocardial ischemia-reperfusion injury model. We identified that pubescens B (2) and pubescens H (8) had good myocardial protective function, and discussed the interaction between the active compounds and UCHL5.

Fig. 1 Structures of compounds 1–11.

Results and discussion

The 95% EtOH extracts of the aerial parts of S. pubescens (15 kg) were systematically separated by ODS column, Sephadex LH-20, and preparative HPLC to yield compounds 1–11.

Compound 1 was obtained as a white powder with [α] +52.0. Its molecular formula was determined to be C₂₀H₃₀O₃ by (+)-HRESIMS data at m/z 319.2276 [M + H]⁺ (calcd for C₂₀H₃₁O₃, 319.2267), indicating 6 degrees of unsaturation. The IR analysis presented the signals of hydroxy (3387 cm⁻¹) and double-bond (1682 cm⁻¹). The ¹H NMR spectrum (Table 1) showed characteristic signals for three oxygenated methine protons [δ_H 3.16 (1H, dd, J = 11.2, 5.6 Hz), 3.83 (1H, dd, J = 1.4, 5.6 Hz), 3.90 (1H, s)], an oxygenated methylene protons [δ_H 3.59 (1H, dd, J = 11.9, 1.4 Hz) and 4.30 (1H, dd, J = 11.9, 4.9 Hz)] and four methyl protons [δ_H 0.89 (3H, s), 0.92 (3H, s), 0.98 (6H, s)]. In the ¹³C NMR and HSQC spectra (Table 1 and Fig. S4), twenty carbon signals were observed, assigned to four methyl, five methylene (including one oxygenated carbon, δ_C 75.3), six methine (including two olefinic carbons, δ_C 115.6 and 131.9 and three oxygenated carbons, δ_C 79.5, 79.6 and 84.7), and five quaternary carbons (including two olefinic carbons, δ_C 129.6 and 144.6). Considering the molecular formula, the degree of unsaturation, and biological source, 1 was considered as an ent-pimarene diterpenoid with an oxolane ring group.²⁰ The ¹H–¹H COSY correlations of H-15/H₂-16 and the HMBC correlations (Fig. S5 and S6) from the protons at δ_H 0.92 (3H, s, H-17) to the carbons at δ_C 34.0 (C-12), δ_C 45.2 (C-13), δ_C 84.7 (C-14), and δ_C 79.5 (C-15) ascertained the location of the oxolane ring between C-14 and C-16. The HMBC correlations from δ_H 5.87 (1H, d, H-7) to δ_C 50.0 (C-5), δ_C 25.0 (C-6), δ_C 144.6 (C-9), δ_C 84.7 (C-14) and from δ_H 5.33 (1H, t, H-11) to δ_C 129.6 (C-8), δ_C 37.6 (C-10), δ_C 34.0 (C-12), and δ_C 45.2 (C-13) confirmed a conjugated diene at Δ⁷⁽⁸⁾ and Δ⁹⁽¹¹⁾ (Fig. 2 and S5).

Table 1 ¹H and ¹³C NMR spectroscopic data of compounds 1 and 2 in CD₃OD

Position	1		2
	δH (J in Hz)^a	δC^b, type	δH (J in Hz)^a	δC^b, type
a ^1H NMR data (d) were measured in 600 MHz NMR instrument. Proton coupling constants (J) in Hz are given in parentheses.b ^13C data were recorded at 150 MHz. The assignments were based on ^1H–^1H COSY, HSQC, HMBC, and NOESY experiments.
1a	1.88, dt (15.4, 4.2)	36.5, CH2	1.83, dt (13.2, 3.6)	35.5, CH2
1b	1.49, td (15.4, 4.8)		1.20 td (13.2, 5.4)
2a	1.72 m	28.3, CH2	1.67, m	28.2, CH2
2b	1.70, m		1.65, m
3	3.16, dd (11.2, 5.6)	79.6, CH	3.18, dd (10.8, 5.4)	79.5, CH
4		40.1, C		39.9, C
5	1.24, dd (11.9, 4.9)	50.0, CH	1.11, dd (12.6, 2.4)	52.0, CH
6a	2.27, dt (18.2, 4.9)	25.0, CH2	1.77, dd (12.6, 7.2)	19.6, CH2
6b	2.21, dd (18.2, 11.9)		1.53, ddd (18.6, 12.6, 6.0)
7a	5.87, d (5.6)	131.9, CH	2.39, td (17.4, 5.4)	31.8, CH2
7b			1.92, m
8		129.6, C		125.2, C
9		144.6, C		143.1, C
10		37.6, C		38.8, C
11a	5.33, t (4.2)	115.6, CH	2.05, m	21.8, CH2
11b			2.03, m
12a	1.87, m	34.0, CH2	1.46, m	29.9, CH2
12b	1.86, m		1.35, td (12.0, 6.6)
13		45.2, C		43.1, C
14	3.90, s	84.7, CH	3.56, s	84.5, CH
15	3.83, dd (5.6, 1.4)	79.5, CH	4.89, dd (5.4, 2.4)	82.7, CH
16a	4.30, dd (11.9, 5.6)	75.3, CH2	4.26, dd (10.8, 5.4)	72.3, CH2
16b	3.59, dd (11.9, 1.4)		3.62, dd (10.8, 2.4)
17	0.92, s	16.3, CH3	0.90, s	14.6, CH3
18	0.98, s	28.4, CH3	1.01, s	28.6, CH3
19	0.89, s	16.0, CH3	0.82, s	16.3, CH3
20	0.98, s	21.3, CH3	1.02, s	19.7, CH3
OAc-15			2.08, s	172.3, C
				20.8, CH3

---

Fig. 2 The key ¹H–¹H COSY and HMBC (from ¹H to ¹³C) correlations of compounds 1–8.

Furthermore, the ¹H–¹H COSY correlations (Fig. S6) of H₂-1/H₂-2/H-3, H-5/H₂-6/H-7 and H-11/H₂-12, combined with the correlations from H₂-12 to C-9, from H-14 to C-7, C-8, C-9, from H₃-18 to C-3, C-4, C-19, and from H₃-20 to C-1, C-5, C-10 in the HMBC spectrum (Fig. S5) completed the planar structure of 1. Therefore, the gross structure of 1 was unambiguously determined to be ent-14,16-oxolane-3,15-dihydroxypimar-7,9(10)-diene. The NOESY correlations (Fig. S7) of H-3/H₃-18, H-5/H₃-18, H-3/H-5, H-6β/H₃-18, and H-14/H₃-17 indicated that H-3, H-5, H-14, H₃-17, and H₃-18 were located at the β-orientation (Fig. 3). Whereas, the NOESY correlations (Fig. S7) of H₃-19/H₃-20, H₃-20/H-12α, and H-12α/H-15 suggested the α-orientation of H₃-19, H₃-20, and H-15, respectively. Accordingly, the configurations of C-3, C-10, and C-13 were assigned as R*, R*, and R*, respectively. Thus, 3R,5S,10R,13R,14R,15R-1 (1a) and 3S,5R,10S,13S,14S,15S-1 (1b) were proposed to be the model compounds according to the relative configuration established for 1. Comparing the experimental and theoretical ECD spectra predicted by the time-dependent density functional theory (TDDFT) at the B3LYP/6-311G (d, p) level, the overall pattern of calculated ECD spectrum for 1a stereoisomer was in good agreement with experimental data of 1 (Fig. 4). Finally, the 3R,5S,10R,13R,14R,15R configuration of 1 was confirmed by X-ray diffraction analysis of crystals obtained via recrystallization from MeOH (Fig. 5). Thus, 1 was established as (3R,5S,10R,13R,14R,15R)-ent-14,16-oxolane-3,15-dihydroxypimar-7,9(10)-diene, and it was named as pubescens A.

Fig. 3 Energy-minimized conformation with key NOESY correlations for compounds 1–8.

Fig. 4 Calculated or experimental ECD spectra of 1–8 in MeOH.

Fig. 5 X-ray ORTEP drawing of compound 1, 5 and 8.

Compound 2 and compound 3 had the same molecular formula of C₂₂H₃₄O₄, as indicated by their ¹³C NMR data and positive HRESIMS ions at m/z 385.2349 [M + Na]⁺ (calcd 385.2349) for 2 and at m/z 363.2541 [M + H]⁺ (calcd 363.2530) for 3. The ¹H and ¹³C NMR data of these compounds (Tables 1 and 2) were similar to those of ent-14β,16-oxolane-8-pimarene-3β,15α-diol²¹ except for some differences of substituents. 2 had an acetyl group at C-15, which was supported by the HMBC correlations (Fig. S15) from the protons at δ_H 4.89 (1H, dd, H-15) to the carbons at δ_C 84.5 (C-14), 34.0 (C-12) and the ester carbonyl at δ_C 172.3. Whereas, 3 had an acetyl group at C-3, which was confirmed by the HMBC correlations (Fig. S25) from δ_H 4.48 (1H, dd, H-3) to δ_C 52.0 (C-5), 35.1 (C-1), 28.5 (C-18), and the ester carbonyl at δ_C 172.8. The NOESY correlations (Fig. S17) of H-15/H-12α, H-12α/H₃-20, and H₃-20/H₃-19 revealed the α-orientation of H₃-19, H₃-20, and H-15 of 2. And the NOESY correlations (Fig. S27) of H-3/H₃-18, H₃-18/H-5, H-5/H-7β, H-7β/H-14, and H-14/H₃-17 indicated the β-orientation of H-3, H-5, H-14, H₃-17, and H₃-18 of 3. Consequently, the structure of 2 and 3 were elucidated as (3R,5S,10R,13R,14R,15R)-ent-15-acetoxy-14,16-oxolane-3-hydroxypimarene-8-ene and (3R,5S,10R,13R,14R,15R)-ent-3-acetoxy-14,16-oxolane-15-hydroxypimarene-8-ene, respectively, by comparison of their experimental ECD with calculated data. Compounds 2 and 3 were named as pubescens B and pubescens C (Fig. 4).

Table 2 ¹H and ¹³C NMR spectroscopic data of compounds 3 and 4 in CD₃OD

Position	3		4
	δH (J in Hz)^a	δC^b, type	δH (J in Hz)^a	δC^b, type
a ^1H NMR data (d) were measured in 600 MHz NMR instrument. Proton coupling constants (J) in Hz are given in parentheses.b ^13C data were recorded at 150 MHz. The assignments were based on ^1H–^1H COSY, HSQC, HMBC, and NOESY experiments.
1a	1.86, dt (13.2, 3.6)	35.1, CH2	3.93, t (2.8)	72.5, CH
1b	1.25, m
2a	1.72, m	24.9, CH2	1.90, td (12.6, 2.8)	35.4, CH2
2b	1.70, m		1.79, m
3	4.48, dd (9.6, 7.2)	82.3, CH	3.69, dd (11.9, 4.2)	74.1, CH
4		38.8, C		39.8, C
5	1.22, dd (12.6, 2.4)	52.0, CH	1.62, dd (12.6, 2.1)	44.7, CH
6a	1.77, dd (12.6, 7.2)	19.5, CH2	1.76, m	19.3, CH2
6b	1.56, ddd (18.6, 12.6, 6.0)		1.55, ddd (17.5, 12.6, 5.6)
7a	2.41, td (17.4, 5.4)	31.7, CH2	2.36, dd (17.5, 4.9)	31.6, CH2
7b	1.93, m		1.93, m
8		125.7, C		127.0, C
9		142.7, C		141.5, C
10		38.7, C		43.4, CH
11a	2.05, m	21.9, CH2	2.32, m	21.3, CH2
11b	2.03, m		2.08, brd (17.5)
12a	1.35, m	30.4, CH2	1.35, dd (12.6, 4.9)	30.5, CH2
12b	1.33, m		1.27, td (12.6, 4.9)
13		43.9, C		44.1, C
14	3.56, s	84.1, CH	3.57, s	84.2, CH
15	3.78, dd (5.4, 2.4)	80.5, CH	3.79, dd (5.6, 2.1)	80.6, CH
16a	4.20, dd (9.6, 5.4)	74.6, CH2	4.20, dd (9.8, 5.6)	74.6, CH2
16b	3.57, dd (9.6, 5.4)		3.57, dd (9.8, 2.1)
17	0.93, s	14.6, CH3	0.96, s	14.8, CH3
18	0.92, s	28.5, CH3	1.04, s	28.6, CH3
19	0.94, s	17.1, CH3	0.83, s	16.2, CH3
20	1.05, s	19.7, CH3	1.01, s	20.7, CH3
OAc-3	2.04, s	172.8, CH
		21.1, CH3

---

Compound 4 was purified as a white powder with a molecular formula of C₂₀H₃₂O₄ requiring 5 degrees of unsaturation, which was shown by the [M + H]⁺ ion at m/z 337.2389 (calcd for C₂₀H₃₃O₄, 337.2373). Its 1D NMR data (Table 2) were similar to those of ent-14β,16-oxolane-8-pimarene-3β,15α-diol,²¹ except for an extra hydroxymethyl (δ_C 72.5) at C-1. This inference was verified by the COSY cross-peaks of H-1/H₂-2 and the HMBC cross-peaks (Fig. S35 and S36) from δ_H 3.93 (1H, t, H-1) to δ_C 82.3 (C-3), 52.0 (C-5), 141.5 (C-9), and 20.7 (C-20). H-1 appeared as a double doublet with two small coupling constants (2.8 and 2.8 Hz), together with correlations of H-1/H₃-20 in the NOESY spectrum (Fig. S37), indicating the β-orientation of OH-1, C-1 was defined as R* configured. Finally, according to the comparison of experimental data with calculated ECD data of 1R,3R,5S,10R,13R,14R,15R-4 (4a) and 1S,3S,5R,10S,13S,14S,15S-4 (4b) (Fig. 4) and biogenetic consideration, 4 was identified as (1R,3R,5S,10R,13R,14R,15R)-ent-14,16-oxolane-1,3,15-trihydroxypimarene-8-ene, and was named as pubescens D.

Compound 5, white amorphous powder, gave a [M + Na]⁺ ion in HRESIMS at m/z 361.2353 (calcd for 361.2349), consistent with the molecular formula C₂₀H₃₄O₄. A comparison of the 1D NMR data of 5 with those of darutigenol (9) (Tables 3 and S9) revealed a considerable degree of similarity except for some resonances corresponding to ring B.²² The large chemical shift difference of C-6 (from δ_C 23.0 in 9 to δ_C 68.7 in 5) indicated the presence of a OH group in C-6 of 5, which was confirmed by the COSY cross-peaks of H-5/H-6/H₂-7 and HMBC cross-peaks shown in Fig. 2, S45 and S46. The NOESY cross-peaks of H-6/H₃-19/and H-6/H₃-20, suggested the β-orientation of OH-6 (Fig. 3 and S47). Because of the structure of a vic-diol moiety in the side chain, Mo₂(OAc)₄ induced ECD was used to suggest the absolute configuration of 5. According to the rule proposed by Snatzke, the positive Cotton effect at 305 nm in the ECD spectrum of 15S,16-dihydroxy-7-oxopimar-8(9)-ene, C-15 was indicated as an S configuration.^23,24 On the contrary, the negative sign observed in ECD spectrum of 5 established a 15R configuration (Fig. 6).^25,26 Finally, according to the comparison of experimental data with calculated ECD data of 3R,5S,6R,9R,10S,13S,15R-5 (5a) and 3S,5R,6S,9S,10R,13R,15S-5 (5b) (Fig. 4), the configuration of 5 was defined as 3R, 5S, 6R, 9R, 10S, 13S and 15R. And this identification was confirmed by X-ray diffraction analysis of crystals obtained via recrystallization from MeOH as shown in Fig. 5. Therefore, the structure of 5 was elucidated as (3R,5S,6R,9R,10S,13S,15R)-ent-3,6,15,16-tetrahydroxypimar-8(14)-ene and was named as pubescens E.

Fig. 6 ECD curves of Mo₂(OAc)₄ complex of compound 5.

Compound 6 shared the same molecular formula of C₂₀H₃₄O₄ with 5 as determined by the HRESIMS at m/z 361.2347 [M + Na]⁺ (calcd for C₂₀H₃₄NaO₄, 361.2349). Its 1D NMR spectroscopic data (Table 3) were similar to those of 9, except for an additional hydroxy group at C-2. Supporting this assignment was the presence of COSY correlation (Fig. S56) of H₂-1/H-2/H-3 and HMBC correlation (Fig. S55) observed from δ_H 2.96 (1H, d, H-3) to δ_C 69.3 (C-2), 40.6 (C-4), 29.6 (C-19), and 16.2 (C-20). The NOESY correlation (Fig. S57) of H-2/H₃-20 suggested that C-2 was S* configured. Therefore, the structure of 6 was identified as (2S,3S,5S,10S,13S,15R)-ent-2,3,15,16-tetrahydroxypimar-8(14)-ene and was named as pubescens F on the basis of the comparison of experimental data with calculated ECD data of 2S,3S,5S,10S,13S,15R-6 (6a) and 2R,3R,5R,10R,13R,15S-6 (6b) (Fig. 4).

Table 3 ¹H and ¹³C NMR spectroscopic data of compounds 5 and 6 in CD₃OD

Position	5		6
	δH (J in Hz)^a	δC^b, type	δH (J in Hz)^a	δC^b, type
a ^1H NMR data (d) were measured in 600 MHz NMR instrument. Proton coupling constants (J) in Hz are given in parentheses.b ^13C data were recorded at 150 MHz. The assignments were based on ^1H–^1H COSY, HSQC, HMBC, and NOESY experiments.
1a	1.66, m	40.8, CH2	1.96, dd (12.6, 4.2)	46.6, CH
1b	1.21, dd (13.2, 3.6)		1.13, dd (12.6, 2.4)
2a	1.70, m	28.3, CH2	3.59, ddd (9.6, 4.2, 2.4)	69.3, CH2
2b	1.69, m
3	3.13, dd (11.4, 3.0)	80.1, CH	2.96, d (9.6)	84.2, CH
4		40.9, C		40.6, C
5	1.02, brs	57.3, CH	1.16, dt (12.6, 3.0)	55.6, CH
6a	4.35, brs	68.7, CH	1.62, m	23.4, CH
6b			1.39, ddd (18.6, 12.6, 4.2)
7a	2.29, brd (14.4)	47.0, CH2	2.29, ddd (14.4, 4.8, 1.8)	36.9, CH2
7b	2.25, brd (14.4)		2.06, td (14.4, 5.4)
8		136.8, C		139.3, C
9	1.72, m	52.1, CH	1.78, t (8.4)	52.2, CH
10		40.0, C		39.9, C
11a	1.60, m	19.1, CH2	1.59, m	19.5, CH2
11b	1.58, m		1.56, m
12a	1.97, m	32.4, CH2	2.00, dt (12.6, 3.6)	33.2, CH2
12b	0.97, td (12.0, 4.8)		0.93, td (12.6, 4.8)
13		38.9, C		38.6, C
14	5.25, s	132.0, CH	5.20, d (1.8)	130.1, CH
15	3.60, brd (9.0)	78.1, CH	3.56, dd (9.0, 2.4)	77.5, CH
16a	3.77, brd (10.2)	64.2, CH2	3.68, dd (11.4, 2.4)	64.3, CH2
16b	3.47, brd (10.2)		3.46, dd (11.4, 9.0)
17	0.87, s	23.2, CH3	0.84, s	23.0, CH3
18	1.07, s	28.7, CH3	0.83, s	17.6, CH3
19	1.18, s	17.5, CH3	1.03, s	29.6, CH3
20	1.10, s	18.5, CH3	0.87, s	16.2, CH3

---

Compound 7 was isolated as a white powder. The molecular formula was determined as C₂₄H₃₈O₅, based on the HRESIMS ion at m/z 429.2626 [M + Na]⁺. Its 1D NMR spectroscopic data were similar to those of siegesbeckia Q (10) (Tables 4 and S10), except for an extra acetyl group at C-16. This was supported by the 2D NMR experiments (Fig. S64–S67). Thus, the structure of 7 was established as (3R,5S,9R,10S,13S,15R)-ent-15,16-diacetoxy-3-hydroxypimar-8(14)-ene and was named as pubescens G according to the comparison of experimental data with calculated ECD data of 3R,5S,9R,10S,13S,15R-7 (7a) and 3S,5R,9S,10R,13R,15S-7 (7b) (Fig. 4).

Compound 8 was assigned a molecular formula of C₂₀H₃₄O₃ based on the HRESIMS ion at m/z 323.2596 [M + H]⁺ (calcd for C₂₀H₃₅O₃, 323.2581). The ¹H and ¹³C NMR data of 8 (Table 4) were indicative of the presence of three methyl groups (δ_H 0.82, 0.92 and 1.01), three oxygenated carbons (δ_C 59.9, 65.2 and 65.7). A comparison of the molecular formula and the NMR data of 8 with those of kirenol (11) revealed that these two compounds were closely related, with prominent differences being the absence of OH group in C-15 in the side chain. The COSY, HMBC, and NOESY correlations (Fig. S75–S77) matched well with this observation shown in Fig. 2 and 3. According to the comparison of experimental data with calculated ECD data of 2S,4R,5S,9R,10S,13R-8 (8a) and 2R,4S,5R,9S,10R,13S-8 (8b) (Fig. 4), the configuration of 8 was determined as 2S, 4R, 5S, 9R, 10S, and 13R, and was confirmed by single-crystal X-ray diffraction analysis of crystals obtained via recrystallization from MeOH (Fig. 5). Therefore, the structure of 8 was identified as (2S,4R,5S,9R,10S,13R)-ent-2,16,19-trihydroxypimar-8(14)-ene and was named as pubescens H.

Table 4 ¹H and ¹³C NMR spectroscopic data of compounds 7 and 8 in CD₃OD

Position	7		8
	δH (J in Hz)^a	δC^b, type	δH (J in Hz)^a	δC^b, type
a ^1H NMR data (d) were measured in 600 MHz NMR instrument. Proton coupling constants (J) in Hz are given in parentheses.b ^13C data were recorded at 150 MHz. The assignments were based on ^1H–^1H COSY, HSQC, HMBC, and NOESY experiments.
1a	1.70, dt (13.2, 3.0)	38.0, CH2	2.00, m	49.4, CH2
1b	1.18, m		1.02, t (12.6)
2a	1.62, m	28.3, CH2	3.76, tt (12.6, 4.2)	65.2, CH
2b	1.61, m
3a	3.20 dd (9.0, 6.6)	79.7, CH	2.17, ddd (12.6,4.2, 2.4)	44.4, CH2
3b			0.89, td (12.6, 1.2)
4		40.1, C		41.5, C
5	1.09, dd (12.6, 3.0)	55.6, CH	1.20, dd (12.6, 2.4)	56.5, CH
6a	1.67, m	23.5, CH2	1.71, m	23.3, CH2
6b	1.42, dd (12.6, 4.8)		1.30, ddd (18.6, 12.6, 4.2)
7a	2.32, ddd (14.4, 4.8, 1.8)	37.1, CH2	2.25, ddd (14.4, 4.2, 1.8)	37.3, CH2
7b	2.06, m		2.01, m
8		141.9, C		137.1, C
9	1.73, t (8.4)	51.9, CH	1.79, t (8.4)	52.4, CH
10		39.3, C		40.6, C
11a	1.55, m	19.4, CH2	1.62, m	20.3, CH2
11b	1.53, m		1.55, m
12a	1.70, dt (12.6, 3.6)	33.4, CH2	1.58, m	36.3, CH2
12b	1.02, td (12.6, 4.8)		1.15, td (12.6, 4.8)
13		38.3, C		33.9, C
14	5.19, brs	127.4, CH	5.23, brs	132.5, CH
15a	5.16, dd (9.0, 2.4)	75.9, CH	1.63, m	45.1, CH2
15b			1.53, m
16a	4.44, dd (12.0, 2.4)	64.8, CH2	3.63 td (10.2, 6.0)	59.9, CH2
16b	4.06, dd (12.0, 9.0)		3.57, td (10.2, 6.0)
17	0.97, s	23.5, CH3	0.92, s	28.9, CH3
18	1.00, s	29.1, CH3	1.01, s	28.0, CH3
19a	0.82, s	16.5, CH3	3.68, d (10.8)	65.7, CH2
19b			3.33, d (10.8)
20	0.85, s	16.1, CH3	0.82, s	17.3, CH3
OAc-15	2.06, s	172.5, C
		20.8, CH3
OAc-16	1.98, s	172.5, C
		20.7, CH3

---

By comparing the spectroscopic data with those reported in the literature, structures of the known compounds were identified as darutigenol (9),²² siegesbeckia Q (10),²⁵ kirenol (11),²⁷ respectively.

Based on our previous study, the 50% ethanol elution of HS inhibited UCHL5/NLRP3 pathway and alleviated myocardial H/R injury, kirenol might be the core active compound.¹⁹ Therefore, compounds 1–10 were tested for the myocardial protective function by H9c2 cell (National Experimental Cell Resource Sharing Platform, Beijing, China) H/R induced myocardial ischemia-reperfusion injury model (kirenol, 10 µM, as a positive control).^19,28 As a result, pubescens H (8) showed strong myocardial protective activity in 1 µM. Comparing with the cell viability of 59.80 ± 7.37% in the positive group, the cell viability can be increased to 62.19 ± 12.65%. Pubescens B (2) also exhibited good myocardial protective activity in 10 µM, the cell viability can be increased to 37.94 ± 6.28%. Additionally, the known compound darutigenol (9) also had great myocardial protective activity in 10 µM (Table 5). Other tested compounds showed no activity in this research. The cytotoxicity of Pubescens B (2) and pubescens H (8) was also shown in the Fig. S87. According to the data, all tested compounds exhibited no appreciable toxicity in normal H9c2 cell even at a high concentration of 100 µM, but the cell injury was observed when they were applied to cells subjected to hypoxia/reoxygenation (H/R). This result may be caused by the phenomenon that stressed or diseased cells exhibit heightened sensitivity.^29–31

Table 5 The activity of compounds on H9c2 cells^a

Compound	Control	Cell viability (%)
		H/R	Kirenol 10 µM	1 µM	10 µM	50 µM
a n = 6, *P < 0.05 vs. Sham, ^#P < 0.05 vs. H/R, and the time of treatment: 24 h.
2	100.00 ± 5.15	20.69 ± 4.36*	46.39 ± 7.58^#	29.37 ± 2.19	37.94 ± 6.28^#	26.92 ± 0.77
8	98.80 ± 2.95	37.90 ± 4.19*	59.80 ± 7.37^#	62.19 ± 12.65^#	57.42 ± 9.39^#	51.46 ± 9.83^#
9	103.77 ± 4.65	53.02 ± 7.07*	61.76 ± 6.83^#	63.18 ± 7.21	76.21 ± 6.22^#	66.18 ± 20.37^#

---

In order to study the binding mode of active compounds and UCHL5, molecular docking was performed using Autodock 4.2.6. As a result, the binding energy of pubescens B (2), pubescens H (8), darutigenol (9) and kirenol (11) were −7.5, −6.1, −6.3 and −6.3 kJ mol⁻¹, respectively. As shown in Fig. 7A, pubescens B (2) had a good interaction with UCHL5 through hydrophobic interactions with residues [Leu-38, Glu-205, Ile-208, Ser-37, Ile-35, Phe 218 and Trp-36]. As shown in Fig. 7B, pubescens H (8) formed three hydrogen bonds with the interacting residues Glu-113, Ala-129 and Asn-132. Ala-129 and Asn-132 were involved in hydrogen bonding interactions with the hydroxyl group at C-16. Pubescens H (8) also had hydrophobic interactions with residues [Val-135, Ser-133, Phe-114 and Phe-117]. As shown in Fig. 7C, darutigenol formed two hydrogen bonds with the interacting residues Trp-36 and Ile-35. Trp-36 and Ile-35 were involved in hydrogen bonding interactions with the hydroxyl group at C-16. Darutigenol had hydrophobic interactions with residues [Gln-209, Ser-37, Phe-218, Ile-208, Glu-39 and Glu-205]. As shown in Fig. 7D, kirenol formed four hydrogen bonds with the interacting residues Glu-113, Ala-129 and Asn-132. Asn-132 were involved in hydrogen bonding interactions with the hydroxyl group at C-15 and C-16. Kirenol also had hydrophobic interactions with residues [Val-135, Ser-133, Phe-114 and Phe-117].

Fig. 7 Molecular docking of compound 2, 8, 9 and kirenol with UCHL5.

Conclusions

Phytochemical investigation of EtOAc extract of the bark of the dried aerial parts of S.pubescens yielded the isolation and identification of eight previously undescribed diterpenoids, named pubescens A–H (1–8), and 3 known compounds. The absolute configuration of pubescens A (1), pubescens E (5) and pubescens H (8) were established by single-crystal X-ray diffraction. Additionally, these compounds were tested for the myocardial protective function by H9c2 cell H/R induced myocardial ischemia-reperfusion injury model. In conjunction with the molecular docking with UCHL5, new compounds pubescens B (2) and pubescens H (8) displayed good myocardial protective activity similar to the typical active compound kirenol (11). Pubescens B and pubescens H could be selected for further study. In conclusion, this study enriched the chemical composition of S. pubescens, preliminarily screened the myocardial protective activity of compounds, explored the binding mode between active compounds and UCHL5 and also provided a basis for further study.

Experimental section

General experimental procedures

Optical rotations were measured on a Rudolph Research Autopol III automatic polarimeter. UV spectra were measured on a Cary 300 spectrometer. ECD spectra were recorded on a JASCO J-815 spectrometer. IR spectra were obtained on a Nicolet Impact 400 FT-IR Spectrophotometer. 1D and 2D NMR spectra were recorded on a Bruker ARX-600 spectrometer with solvent peaks as references. HRESIMS data were obtained with an Agilent 1290 Infinity liquid chromatography system and an Agilent 6546 QTOF mass spectrometer. High-performance liquid chromatography (HPLC) data were recorded on a SHIMADZU LC-20AT equipped with a CBM-20A, an SPD-M20A and a MGII C₁₈ column (250 × 4.6 nm, 5 µm). Preparative HPLC was performed on SHIMADZU LC-16P with an SPD-16, a RID-20A and a MGII C₁₈ column (250 × 20 nm, 5 µm). Column chromatographic separations were carried out with silica gel (100–200 and 200–300 mesh, Qingdao Marine Chemical Group Corporation, Qingdao, China), MCI gel (CHP20/P120, MITSUBUSHI Chemical Corporation, Japan), Sephadex LH-20 (17-0090-02, GE Healthcare, Sweden) and ODS (Grace, USA). TLC was conducted with glass precoated with silica gel GF254 (Qingdao Marine Chemical Group Corporation, Qingdao, China). All solvents were purchased from Innochem.

Plant material

S. pubescens was purchased from Dongzhimen Hospital, Beijing University of Chinese Medicine and produced by Beijing Sanhe Pharmaceutical Limited Company. The plant material was identified by Researcher Zhan Zhilai at Chinese Medicine Resource Center, Institute of Traditional Chinese Medicine, Chinese Academy of Chinese Medical Sciences.

Extraction and isolation

The dried aerial parts of S. pubescens (15 kg) were extracted under reflux with 95% EtOH three times. The solvent was evaporated under reduced pressure to yield the aqueous residue, which was then extracted with EtOAc, the EtOAc extract (500 g) was subjected to silica gel column chromatography (CC) and eluted with petroleum ether–acetone (100 : 0 to 0 : 100) to yield 13 fractions (Fr.1–Fr.13).

Fr.6 and Fr.7 (66.7 g) were combined and subjected to silica gel column chromatography (CH₂Cl₂–MeOH, 100 : 0 to 0 : 100) to afford subfractions Fr.6.1, Fr.6.2, Fr.6.3, Fr.6.4 and Fr.6.5. Subsequently, Fr.6.1 was subjected to Sephadex LH-20 CC (5 × 120 cm) eluting with MeOH and further separated by preparative HPLC [ACE-C₁₈, 5 µm, 250 × 10 mm, ACN–H₂O–TFA (90 : 10 : 0.1)] to afford compound 9 (3.8 mg, t_R 5.5 min). Fr.6.2 was subjected to Middle Chromatogram Isolated Gel (MCI) eluting with MeOH–H₂O (90 : 10 to 100 : 0) and further separated by preparative HPLC [CN-C₁₈, 5 µm, 250 × 10 mm, ACN–H₂O–TFA (30 : 70 : 0.1)] to afford compounds 2 (3.3 mg, t_R 28.0 min), 3 (1.0 mg, t_R 26.5 min), 7 (1.5 mg, t_R 32.5 min) and 10 (2.0 mg, t_R 18.0 min).

Fr.8 and Fr.9 were combined and performed on reversed phase medium pressure CC with MeOH–H₂O (5 : 95 to 100 : 0), yielding 8 subfractions Fr.8.1–Fr.8.8. Separation on Fr.8.6 by Sephadex LH-20 CC (5 × 120 cm) eluting with MeOH afforded 6 subfractions Fr.8.6.1–Fr.8.6.6. Fr.8.6.2 was further separated by preparative HPLC [UG-C₁₈, 5 µm, 250 × 10 mm, ACN–H₂O (39 : 61)] to afford compounds 1 (2.9 mg, t_R 31.0 min) and 4 (2.2 mg, t_R 27.0 min). Fr.8.7 was subjected to Sephadex LH-20 CC (5 × 120 cm) eluting with CH₂Cl₂–MeOH (50 : 50) and silica gel column chromatography (CH₂Cl₂–MeOH, 100 : 0 to 10 : 90) to afford 12 subfractions Fr.8.7.1–Fr.8.7.12. Fr.8.7.4 was further separated by preparative HPLC [MGII-C₁₈, 5 µm, 250 × 10 mm, ACN–H₂O (30 : 70)] to afford compounds 5 (1.9 mg, t_R 36.0 min), 6 (1.5 mg, t_R 34.5 min) and 8 (2.3 mg, t_R 39.0 min). Fr.8.7.11 was further separated by preparative HPLC [SP-C₁₈, 5 µm, 250 × 10 mm, ACN–H₂O (42 : 58)] to afford compound 11 (10.5 mg, t_R 32.5 min).

Pubescens A (1). White powder; [α] +52.0 (c 0.01, MeOH); UV (MeOH, 0.001) λ_max (log ε) 204 (4.23), 231 (2.28), 284 (0.30) nm; FT-IR (ATR) ν_max 3387, 2932, 1682, 1456, 1375, 1208, 1140, 1094, 1038, 1004, 967, 917, 843, 802, 724, 666 cm⁻¹; ¹H NMR (700 MHz, CD₃OD) and ¹³C NMR (150 MHz, CD₃OD) data, see Tables 1 and 2; HRESIMS m/z 319.2276 [M + H]⁺ (calcd for C₂₀H₃₁O₃, 319.2267).

Pubescens B (2). White powder; [α] +1.0 (c 0.01, MeOH); UV (MeOH, 0.001) λ_max (log ε) 205 (0.31), 247 (0.07) nm; FT-IR (ATR) ν_max 3445, 2925, 2871, 1740, 1687, 1456, 1374, 1246, 1208, 1187, 1093, 1033, 979, 933, 843, 803, 725, 649, 607 cm⁻¹; ¹H NMR (600 MHz, CD₃OD) and ¹³C NMR (150 MHz, CD₃OD) data, see Tables 1 and 2; HRESIMS m/z 385.2349 [M + H]⁺ (calcd for C₂₂H₃₄NaO₄, 385.2349).

Pubescens C (3). White powder; [α] +1.0 (c 0.01, MeOH); UV (MeOH, 0.001) λ_max (log ε) 204 (0.32), 242 (0.10) nm; FT-IR (ATR) ν_max 3445, 2940, 2873, 1734, 1704, 1682, 1644, 1453, 1374, 1246, 1207, 1185, 1138, 1032, 978, 842, 802, 724 cm⁻¹; ¹H NMR (600 MHz, CD₃OD) and ¹³C NMR (150 MHz, CD₃OD) data, see Tables 1 and 2; HRESIMS m/z 363.2541 [M + H]⁺ (calcd for C₂₂H₃₅O₄, 363.2530).

Pubescens D (4). White powder; [α] −8.0 (c 0.01, MeOH); UV (MeOH, 0.001) λ_max (log ε) 204 (0.28), 283 (0.02) nm; FT-IR (ATR) ν_max 3386, 2926, 2881, 1680, 1446, 1377, 1260, 1206, 1140, 1090, 1037, 1005, 917, 842, 801, 660, 628 cm⁻¹; ¹H NMR (700 MHz, CD₃OD) and ¹³C NMR (150 MHz, CD₃OD) data, see Tables 1 and 2; HRESIMS m/z 337.2389 [M + H]⁺ (calcd for C₂₀H₃₃O₄, 337.2373).

Pubescens E (5). White powder; [α] +16.0 (c 0.01, MeOH); UV (MeOH, 0.001) λ_max (log ε) 204 (0.26), 281 (0.02) nm; FT-IR (ATR) ν_max 3381, 2930, 2852, 1682, 1446, 1379, 1208, 1140, 1054, 1015, 844, 802, 725, 683, 655, 623, 607 cm⁻¹; ¹H NMR (600 MHz, CD₃OD) and ¹³C NMR (150 MHz, CD₃OD) data, see Tables 1 and 2; HRESIMS m/z 361.2353 [M + Na]⁺ (calcd for C₂₀H₃₄NaO₄, 361.2349).

Pubescens F (6). White powder; [α] −7.0 (c 0.01, MeOH); UV (MeOH, 0.001) λ_max (log ε) 205 (0.20); FT-IR (ATR) ν_max 3388, 2940, 2875, 1679, 1455, 1434, 1261, 1203, 1140, 1055, 1032, 996, 802, 722, 655, 621 cm⁻¹; ¹H NMR (600 MHz, CD₃OD) and ¹³C NMR (150 MHz, CD₃OD) data, see Tables 1 and 2; HRESIMS m/z 361.2347 [M + Na]⁺ (calcd for C₂₀H₃₄NaO₄, 361.2349).

Pubescens G (7). White powder; [α] −18.0 (c 0.01, MeOH); UV (MeOH, 0.001) λ_max (log ε) 205 (0.63), 240 (0.17) nm; FT-IR (ATR) ν_max 3445, 2940, 2867, 1743, 1698, 1647, 1456, 1434, 1371, 1244, 1227, 1142, 1093, 1033, 968, 942, 866, 835, 803, 724, 617, 607 cm⁻¹; ¹H NMR (600 MHz, CD₃OD) and ¹³C NMR (150 MHz, CD₃OD) data, see Tables 1 and 2; HRESIMS m/z 429.2626 [M + Na]⁺ (calcd for C₂₄H₃₈NaO₅, 429.2611).

Pubescens H (8). White powder; [α] −27.0 (c 0.01, MeOH); UV (MeOH, 0.001) λ_max (log ε) 205 (0.37), 281 (0.01) nm; FT-IR (ATR) ν_max 3334, 2937, 2871, 2831, 1681, 1455, 1363, 1206, 1143, 1032, 996, 965, 655 cm⁻¹; ¹H NMR (600 MHz, CD₃OD) and ¹³C NMR (150 MHz, CD₃OD) data, see Tables 1 and 2; HRESIMS m/z 323.2596 [M + H]⁺ (calcd for C₂₀H₃₅O₃, 323.2581).

X-ray crystal structure analysis of compounds 1, 5 and 8

Crystallographic data for the structures of 1, 5 and 8 have been deposited with the Cambridge Crystallographic Data Centre (CCDC 2453116 for 1; CCDC 2453118 for 5; CCDC 2453134 for 8).

X-ray crystallographic data for 1. C₂₀H₃₂O₄ (M = 336.45): monoclinic, space group P2₁ (no. 4); a = 6.84365 (12) Å, b = 11.17471 (20) Å, c = 11.9856 (2) Å, β = 92.9687 (16)°, V = 915.38 (3) ų, Z = 2, T = 169.99 (10) K, µ (Cu Kα) = 0.663 mm⁻¹, D_calc = 1.221 g cm⁻³, 6787 reflections measured (7.386° ≤ 2Θ ≤ 148.144°), 3579 unique (R_int = 0.0210, R_sigma = 0.0247) which were used in all calculations. The final R₁ was 0.0299 (I > 2σ(I)) and wR₂ was 0.0843 (all data).

X-ray crystallographic data for 5. C₂₀H₃₄O₄ (M = 338.47): orthorhombic, space group P2₁2₁2₁ (no. 19); a = 7.3800 (4) Å, b = 12.1066 (4) Å, c = 20.3919 (10) Å, V = 1821.95 (15) ų, Z = 4, T = 149.99 (10) K, µ (Cu Kα) = 0.667 mm⁻¹, D_calc = 1.234 g cm⁻³, 20 517 reflections measured (8.494° ≤ 2Θ ≤ 148.322°), 3654 unique (R_int = 0.1125, R_sigma = 0.0672) which were used in all calculations. The final R₁ was 0.0765 (I > 2σ(I)) and wR₂ was 0.2114 (all data).

X-ray crystallographic data for 8. C₂₀H₃₄O₃ (M = 322.47): monoclinic, space group P2₁ (no. 4), a = 11.2450 (3) Å, b = 7.2402 (2) Å, c = 11.5790 (3) Å, β = 99.615 (3)°, V = 929.48 (5) ų, Z = 2, T = 170.00 (10) K, µ (Cu Kα) = 0.588 mm⁻¹, D_calc = 1.152 g cm⁻³, 7872 reflections measured (7.744° ≤ 2Θ ≤ 147.56°), 3472 unique (R_int = 0.0258, R_sigma = 0.0273) which were used in all calculations. The final R₁ was 0.0338 (I > 2σ(I)) and wR₂ was 0.0859 (all data).

Methodology for ECD analysis

All calculations were performed using Gaussian 16.1. Conformation search using molecular mechanics calculations was performed in DS (Discovery Studio) 2018 with MMFF94 s force field with 20 kcal mol⁻¹ upper energy limit at best level. The conformers performed with the DS 2018 software package were further optimized by using the TDDFT method at the B3LYP/6-31G(d, p) level, and the frequency was calculated at the same level of theory. For all optimized structures, vibrational spectra were calculated to ensure that no imaginary frequencies for energy minimum were obtained. The average values were obtained by the Boltzmann distributions, using the relative Gibbs free energies as weighting factors. The stable conformers were subjected to ECD calculation by the TDDFT method at the B3LYP/6-311G+(d,p) level with the CPCM model in MeOH. ECD spectra of different conformers were simulated using SpecDis 1.71 with a half-bandwidth of 0.3 eV, and the final calculated ECD spectra were obtained according to the Boltzmann-calculated contribution of each con-former. The calculated ECD spectra were compared with the experimental data. Additional details are provided in SI (pages 86–117).

Bioactivity assay

H9c2 cardiomyocytes (National Experimental Cell Resource Sharing Platform, Beijing, China) were cultured with glucose-free, fetal bovine serum-free DMEM and incubated in 1% O₂ and 5% CO₂ mixed with 94% N₂ for 6 h, followed by re-oxygenation for 18 h to establish the H/R model. The cells were divided into blank group, model group, positive group, low-dose group, medium-dose group and high-dose experimental group. Except for the blank group, the remaining groups established the H/R injury model. The positive group was supplemented with kirenol at 10 µM. The test for each compound was conducted individually. The experimental groups were supplemented with tested compounds at 1, 10 and 50 µM, respectively. The blank group and model group received the same volume of serum-free and glucose-free. All data were expressed as the mean ± SD values. All data were analyzed by GraphPad Prism 8.0 software and compared using one-way analysis of variance. Statistical significance was considered when the probability was <0.05.

Molecular docking

Protein and ligand preparation. The three-dimensional structure of the target protein was retrieved from the Protein Data Bank (RCSB) (https://www.rcsb.org/), and download the PDB format file. In Pymol 2.5, removed redundant ligands, then configured the protein in AutoDock as follows: remove water, replace with hydrogen, designate the protein as the receptor, and save the structure as a PDBQT protein receptor file. The drug molecular structures were download from the PubChem database (https://pubchem.ncbi.nlm.nih.gov/). Converted the SDF format to PDB format using the OpenBabel GUI software. Similarly, in AutoDock, we configured the drug as follows: remove water, add hydrogen, and set the drug as a ligand. The software automatically configured the torsion tree. We exported the ligand file in PDBQT format.

Grid box dimensions and parameters. The PDBQT structures of the receptor and ligand were imported into AutoDock to define the molecular docking range. With the target protein as the grid center, adjusted the center coordinates (center −23.9/2.075/−8.161) and box size (size 110/54/92) parameters to ensure complete coverage of the protein within the docking box.

Docking protocol and scoring function. In AutoDock Vina, molecular docking was performed by detecting protein macromolecules and inserting small drug molecules, as well as configuring operational methods and docking parameters. The PDBQT format was used to calculate the minimum binding energy. The PyMol 2.5 software converted the combined PDBQT format into PDB format. Finally, for visualization, PyMol 2.5 and Ligplot v.2.29 were used to process the composite PDB format file.

Author contributions

Wanting Li: writing – original draft, methodology, investigation, formal analysis, data curation. Guiyang Xia: writing – review and editing, validation, software, formal analysis, data curation. Jinyu Xia: formal analysis, methodology, investigation. Qiyao Liu: investigation, formal analysis, data curation. Xue-Fen Wu: investigation, data curation. Linnan Zhou: investigation, data curation. Xiaohong Wei: methodology, data curation. Huan Xia: writing – review and editing, methodology, investigation, formal analysis, data curation. Sheng Lin: supervision, project administration, funding acquisition, conceptualization.

Conflicts of interest

The authors declare no conflict of interest.

Data availability

The data supporting this article have been included as part of the supplementary information (SI). Supplementary information: HRESIMS, IR, UV, CD and NMR spectra for compounds 1–8. NMR spectra for compounds 9–11. X-ray crystallographic data for compounds 1, 5 and 8. ECD details for compounds 1–8. The cytotoxicity data for 2 and 8. The FCF data for compounds 1, 5 and 8. See DOI: https://doi.org/10.1039/d5ra09664b.

CCDC 2453116, 2453118 and 2453134 contain the supplementary crystallographic data for this paper.^32a–c

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (82574555 and 82430116); the Special Fund of Central Committee High Level Chinese Medicine Hospital (CZ015-DZMG-LJRC-0014 and CZ015-DZMG-ZJXY-23013) and the Fundamental Research Funds for the Central Universities (2024-JYB-058).

References

1. Q. Wang, Y. Y. Liang, K. W. Li, Y. Li, F. J. Niu, S. J. Zhou, H. C. Wei and C. Z. Zhou, J. Ethnopharmacol., 2021, 275, 114117,  DOI:10.1016/j.jep.2021.114117.
2. Chinese Pharmacopoeia Commission, Pharmacopoeia of the People's Republic of China, China Medical Science, 2020, vol. 1.
3. Z. Xia, Y. Yin, C. Xia, Y. Zhang, W. Zhang, D. Chen, H. Zhao and Y. Huang, Chin. J. Geriatr. Heart, Brain Vessel Dis., 2018, 20, 1016–1018.
4. H. Guo, Y. Zhang, B. Cheng, X. Fu, P. Zhu, J. Chen, Y. Chan, C. Yin, Y. Wang, M. Hossen, A. Amin, A. Tse and Z. Yu, BioSci. Trends, 2018, 12, 330–337,  DOI:10.5582/bst.2018.01103.
5. C. Chen, Z. L. Feng, J. Petrović, M. Soković, Y. Ye and L. G. Lin, Acta Mater. Med., 2023, 2, 357–376,  DOI:10.15212/AMM-2023-0026.
6. X. H. Liu, D. U. Yajun, G. X. Liu, G. M. Dan, X. Tong and J. Xiao, J. South. Med. Univ., 2019, 39, 1387–1392.
7. B. Xiao, J. Li, Z. Qiao, S. Yang, H. Kwan, T. Jiang, M. Zhang, Q. Xia, Z. Liu and T. Su, J. Ethnopharmacol., 2023, 5, 116852,  DOI:10.1016/j.jep.2023.116852.
8. B. Wu, X. Y. Huang, L. Li, X. H. Fan, P. C. Li, C. Q. Huang, J. Xiao, R. Gui and S. Wang, J. Cell. Mol. Med., 2019, 23, 7651–7663,  DOI:10.1111/jcmm.14638.
9. J. Liu, Z. Ge, X. Jiang, J. Zhang, J. Sun and X. Mao, Chin. J. Nat. Med., 2023, 21, 499–515,  DOI:10.1016/S1875-5364(23)60410-8.
10. P. Yanpiset, C. Maneechote, S. Sriwichaiin, N. Siri-Angkul, S. C. Chattipakorn and N. Chattipakorn, Acta Pharm. Sin. B, 2023, 13, 29–53,  DOI:10.1016/j.apsb.2022.08.007.
11. C. Zhang, S. J. Cai, W. Y. Cheng, N. M. Deng, Y. H. Han and C. Chen, J. South. Agric., 2014, 45, 1060–1064.
12. D. Wang, X. Dong, Y. Nie, W. Yang and C. Li, Rec. Nat. Prod., 2022, 6, 515–537,  DOI:10.25135/rnp.317.2201.2332.
13. T. Xue, R. Liu, H. Chen, M. E. Hussien Ibrahim, J. Dong, W. Zhao, L. Chen, H. Fu and J. Wang, Nat. Prod. Res., 2024, 1–15,  DOI:10.1080/14786419.2024.2426203.
14. M. Liu, T. Luo, R. Li, W. Yin, F. Yang, D. Ge and N. Liu, Chin. J. Nat. Med., 2025, 23, 234–244,  DOI:10.1016/S1875-5364(25)60827-2.
15. Y. Chen, R. Tian, J. Xue, J. Yu and H. Wang, MedComm: Future Med., 2022, 1, e21,  DOI:10.1002/mef2.21.
16. C. Zhang, X. Chang, D. Zhao, Y. He, G. Dong and L. Gao, J. Pharm. Anal., 2024, 15, 101051,  DOI:10.1016/j.jpha.2024.101051.
17. X. Wei, Q. Wang, W. Wen, L. Yang, H. Chen, G. Xu, Y. Zhou, J. Yang and Z. Duan, MedComm: Future Med., 2025, 4, e70016,  DOI:10.1002/mef2.70016.
18. X. Wei, Y. Wu, H. Pan, Q. Zhang, K. He, G. Xia, H. Xia, S. Lin and H. Shang, Front. Cardiovasc. Med., 2022, 6, 895797,  DOI:10.3389/fcvm.2022.895797.
19. X. Wu, H. Xia, W. Li, J. Chen, L. Zhou, Q. Zhang, H. Shang, G. Xia, X. Wei and S. Lin, Acta Mater. Med., 2024, 3, 365–382,  DOI:10.15212/AMM-2024-0030.
20. Y. Xiang, H. Zhang, C. Q. Fan and J. M. Yue, J. Nat. Prod., 2004, 67, 1517–1521,  DOI:10.1021/np0400407.
21. F. Wang, X. Cheng, Y. Li, S. Shi and J. Liu, J. Nat. Prod., 2009, 72, 2005–2008,  DOI:10.1021/np900449r.
22. J. Wang, H. Duan, Y. Wang, B. Pan, C. Gao, C. Gai, Q. Wu and H. Fu, J. Nat. Prod., 2017, 80, 19–29,  DOI:10.1021/acs.jnatprod.6b00150.
23. M. Politi, N. De Tommasi, G. Pescitelli, L. Di Bari, I. Morelli and A. Braca, J. Nat. Prod., 2002, 65, 1742–1745,  DOI:10.1021/np020260p.
24. M. Górecki, E. Jabłońska, A. Kruszewska, A. Suszczyńska, Z. Urbańczyk-Lipkowska, M. Gerards, J. W. Morzycki, W. J. Szczepek and J. Frelek, J. Org. Chem., 2007, 72, 2906–2916,  DOI:10.1021/jo062445x.
25. Y. Zheng, Z. Guo, H. Chen, T. Bao, X. Gao, A. Wang and J. Jia, Phytochemistry, 2023, 205, 113503,  DOI:10.1016/j.phytochem.2022.113503.
26. K. Fan, L. Zhang, B. Tan, G. S. S. Njateng, M. Qin, R. Guo, X. Huang, C. F. Ding, W. Gao, R. Zhang and H. Yu, Chin. J. Nat. Med., 2023, 21, 146–153,  DOI:10.1016/S1875-5364(23)60393-0.
27. K. Liu and E. Röder, Planta Med., 1991, 57, 395–396,  DOI:10.1055/s-2006-960129.
28. J. Y. Jia, E. H. Zang, L. J. Lv, Q. Y. Li, C. H. Zhang, Y. Xia, L. Zhang, L. S. Dang and M. H. Li, Chin. Herb. Med., 2020, 13, 49–63,  DOI:10.1016/j.chmed.2020.09.002.
29. E. Murphy and C. Steenbergen, Physiol. Rev., 2008, 88, 581–609,  DOI:10.1152/physrev.00024.2007.
30. A. B. Lentsch, A. Kato, H. Yoshidome, K. M. McMasters and M. J. Edwards, Hepatology, 2000, 32, 169–173,  DOI:10.1053/jhep.2000.9323.
31. J. V. Bonventre and L. Yang, J. Clin. Invest., 2011, 121, 4210–4221,  DOI:10.1172/JCI45161.
32. (a) CCDC 2453116: Experimental Crystal Structure Determination, 2026,  DOI:10.5517/ccdc.csd.cc2nbnsg; (b) CCDC 2453118: Experimental Crystal Structure Determination, 2026,  DOI:10.5517/ccdc.csd.cc2nbnvj; (c) CCDC 2453134: Experimental Crystal Structure Determination, 2026,  DOI:10.5517/ccdc.csd.cc2nbpc2.

---

Footnote

† These authors contributed equally to this work.

---