Bioactive components from Sinomenium acutum exhibit anti-inflammatory activity and sensitize cervical cancer cells to paclitaxel

Abstract

Six previously undescribed compounds, together with twelve known compounds, were isolated from Sinomenium acutum. The structures of the unreported compounds were determined by comprehensive spectroscopic data analysis, ¹³C NMR calculations followed by DP4 + analysis, ECD calculations, and single-crystal X-ray diffraction analysis. Nine compounds exhibited more potent anti-inflammatory activities than the positive control indomethacin. The structure–activity relationship of the isolates with anti-inflammatory effects was elucidated, providing some useful clues for designing and developing structurally novel anti-inflammatory drugs. Furthermore, five compounds reversed the paclitaxel resistance in paclitaxel-resistant cervical cancer cells HeLa/PTX. The combination of compound 4 and paclitaxel promoted cancer cell apoptosis.

Graphical Abstract

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1007/s13659-026-00652-w.

Introduction

Sinomenium acutum (Menispermaceae family) is mainly distributed in China [1]. The stems of S. acutum are documented in the Chinese Pharmacopoeia for their notable efficacy in treating rheumatic pain, joint swelling, paralysis and cardiovascular disease [2]. Previous research has revealed that S. acutum contains different types of compounds, such as alkaloids, triterpenes, and lignans [3]. Sinomenine, the major active alkaloid in S. acutum, is the effective principle that has been developed into an important anti-inflammatory drug [4].

Inflammation, a protective mechanism against tissue injury or infection, can contribute to the development of cancer, atherosclerosis, and rheumatoid arthritis when it fails to resolve [5]. Developing safer and more effective drugs remains a challenge due to the complex pathogenesis of inflammation. Natural products, as a critical source of clinical drugs, play a pivotal role in the discovery of anti-inflammatory agents.

Cancer poses a rising threat to global health and has emerged as one of the primary contributors to the global burden of disease [6, 7]. In the past decade, some exceptional breakthroughs have been achieved to combat cancer [8]. However, cancer drug resistance limits therapeutic options and is recognized as a major determinant of treatment failure [9]. Drug resistance arises through diverse mechanisms, involving genetic mutations, tumor microenvironment adaptations, and activation of pro-survival pathways [8, 10]. Cancer drug resistance represents a critical obstacle in clinical therapy. To explore novel strategies to overcome drug resistance, combination chemotherapy has become a promising approach due to multiple drugs with distinct mechanisms of action [9]. Natural products demonstrate significant potential in antitumor therapy [11, 12]. In seeking to identify new drug combinations for targeting drug-resistant cancer cells, we found that several sesquiterpenes synergistically enhance cisplatin efficacy in ovarian cancer cells [13], and some alkaloids isolated from Piper longum and Piper nigrum sensitize cervical cancer cells to chemotherapeutic agents [14, 15]. Natural medicinal compounds exhibit promising potential for developing innovative strategies to overcome chemotherapy resistance in cancer [16].

Although some studies have shown that S. acutum contains a variety of chemical constituents, the bioactive components of this plant have not yet been systematically elucidated. In this work, 18 compounds (Fig. 1), including six new ones, were isolated and characterized, and their anti-inflammatory and chemosensitizing activities to paclitaxel were evaluated. The structure–activity relationship of the isolates was analyzed for their anti-inflammatory effects. This work will promote the further exploration and application of S. acutum.

Fig. 1.

Chemical structures of six unreported compounds (1–3, 5, 6, and 13) and twelve known ones (4, 7–12, 14–18)

Results and discussion

Structural elucidation

Compound 1 possessed a molecular formula of C₃₀H₄₈O₃, which was supported by the HRESIMS ion peak at m/z 479.3488 [M + Na]⁺ (calcd. for C₃₀H₄₈O₃Na⁺, 479.3501). The ¹H NMR data (Table 1) of 1 displayed three olefinic protons [δ_H 6.77 (1H, H-22), 6.74 (1H, H-23), and 6.28 (1H, d, J = 14.3 Hz, H-24)], an aldehyde proton [δ_H 9.38 (1H, s, H-21)], and seven methyl signals. The ¹³C NMR data (Table 1) and HSQC correlations (Fig. S3) of 1 revealed 30 carbon resonances, including an aldehyde (δC 195.5), four olefinic carbons (δ_C 144.9, 149.7, 150.8, and 122.0), seven methyls, eight methylenes, five methines (one oxygenated, δ_C 79.1), and five quaternary carbons (one oxygenated, δ_C 71.3). The NMR data (Table 1) of 1 were similar to those of coggygrnoid C [17], suggesting they have the same tetracyclic triterpenoid skeleton. The main difference was that the carbonyl at C-3 in coggygrnoid C was substituted by a hydroxyl in 1, which was also supported by a 2 Da molecular weight difference observed in mass spectrometry. The HMBC correlations (Fig. 2) from Me-28 to C-3 and C-4 corroborated this hypothesis. The NOESY correlations (Fig. 3) of H-3/H-5, H-5/H-9, H-9/Me-30, and Me-30/H-17 indicated that these hydrogens were cofacial and assigned as α-oriented, while the correlations of H-13/Me-18, Me-18/Me-19, and Me-19/Me-29 suggested that they were β-oriented. Furthermore, both H-5 and Me-28 exhibited NOESY correlations with the same proton (δ_H 1.57) on C-2 (Fig. 3), thereby supporting the α-orientation of Me-28. The NOESY correlation (Fig. 3) of CHO-21/H-22 along with coupling constant (³J_23,24 = 14.3 Hz) suggested that the Δ²⁰⁽²²⁾ and Δ²³⁽²⁴⁾ double bonds were both E configurations, which confirmed that 1 contained the same side chain as that in coggygrnoid C. Furthermore, the calculated ECD spectrum of (3S, 5R, 8R, 9R, 10R, 13R, 14R, 17S)−1 showed a satisfactory match with the experimental ECD spectrum of 1 (Fig. 4), and its absolute configuration was determined. Finally, it was named as sinomenoid A.

Table 1.

¹H (400 MHz) and ¹³C (100 MHz) NMR spectroscopic data of 1–3 measured in CDCl₃

No	1		2		3
	δH (J, Hz)	δC	δH (J, Hz)	δC	δH (J, Hz)	δC
1	0.94, m, 1H	39.3	1.38, m, 1H	33.7	1.26, m, 1H	33.8
	1.67, m, 1H		1.26, m, 1H		1.41, m, 1H
2	1.57, m, 1H	27.5	1.92, m, 1H	25.5	1.93, m, 2H	25.5
	1.62, m, 1H		1.54, m, 1H
3	3.20, dd (11.6, 4.8), 1H	79.1	3.38, m, 1H	76.4	3.38, m, 1H	76.4
4		39.1		37.7		37.7
5	0.73, dd (11.9, 2.3), 1H	56.0	1.23, m, 1H	49.6	1.24, m, 1H	49.7
6	1.44, m, 1H	18.4	1.40, m, 2H	18.3	1.39, m, 2H	18.3
	1.54, m, 1H
7	1.25, m, 1H	35.7	1.23, m, 1H	35.2	1.22, m, 1H	35.4
	1.59, m, 1H		1.57, m, 1H		1.60, m, 1H
8		40.7		40.5		40.6
9	1.30, m, 1H	51.0	1.41, m, 1H	50.6	1.41, m, 1H	51.0
10		37.4		37.4		37.4
11	1.10, m, 1H	21.5	1.14, m, 1H	21.3	1.14, m, 1H	21.4
	1.47, m, 1H		1.48, m, 1H		1.53, m, 1H
12	1.07, m, 1H	25.2	1.09, m, 1H	26.5	1.10, m, 1H	25.4
	1.32, m, 1H		1.73, m, 1H		1.58, m, 1H
13	2.03, m, 1H	46.5	1.42, m, 1H	45.8	1.45, m, 1H	44.6
14		50.3		50.2		49.5
15	1.20, m, 1H	32.2	1.04, m, 1H	30.9	1.06, m, 1H	31.4
	1.94, m, 1H		1.48, m, 1H		1.43, m, 1H
16	1.76, m, 1H	27.7	1.30, m, 1H	27.6	1.34, m, 2H	24.2
	1.88, m, 1H		1.81, m, 1H
17	2.84, m, 1H	37.9	1.51, m, 1H	41.9	1.63, m, 1H	41.2
18	1.01, s, 3H	15.6	0.93, s, 3H	15.7	0.89, s, 3H	15.7
19	0.82, s, 3H	16.4	0.84, s, 3H	16.3	0.84, s, 3H	16.2
20		144.9	2.11, m, 1H	52.7	2.33, m, 1H	48.7
21	9.38, s, 1H	195.5	3.29, m, 1H 3.64, dd, (12.5, 4.1), 1H	64.8	3.43, m, 1H 3.57, dd (12.8, 5.3), 1H	65.6
22	6.77a, 1H	149.7	5.30, dd (15.1, 9.6), 1H	132.4	5.35, dd (15.1, 9.3), 1H	129.3
23	6.74a, 1H	122.0	6.29, dd (15.1, 10.6), 1H	129.5	6.32, dd (15.1, 10.5), 1H	130.9
24	6.28, d (14.3), 1H	150.8	5.81, d (10.6), 1H	124.8	5.83, d (10.5), 1H	125.0
25		71.3		134.5		134.4
26	1.40, s, 3H	29.8	1.77, s, 3H	26.1	1.77, s, 3H	26.1
27	1.40, s, 3H	29.9	1.75, s, 3H	18.5	1.75, s, 3H	18.5
28	0.77, s, 3H	15.5	0.93, s, 3H	28.5	0.93, s, 3H	28.5
29	0.98, s, 3H	28.2	0.83, s, 3H	22.2	0.83, s, 3H	22.3
30	0.92, s, 3H	15.5	0.84, s, 3H	16.1	0.84, s, 3H	16.3

^aSignals were overlapped

Fig. 2.

Key ¹H − ¹H COSY and HMBC correlations of compounds 1–3, 5, 6 and 13

Fig. 3.

Key NOESY correlations of compounds 1–3, 5, 6 and 13

Fig. 4.

Experimental and the calculated ECD spectra of compounds 1–3, 5, 6, and 13

Compound 2 was determined to possess a molecular formula of C₃₀H₅₀O₂, as evidenced by HRESIMS ion at m/z 443.3878 [M + H]⁺ (calcd. for C₃₀H₅₁O₂⁺, 443.3889). The ¹H NMR data (Table 1) indicated that there were three olefinic signals [δ_H 5.30 (1H, dd, J = 15.1, 9.6 Hz, H-22), δ_H 6.29 (1H, dd, J = 15.1, 10.6 Hz, H-23), and δ_H 5.81 (1H, d, J = 10.6 Hz, H-24)], seven methyls, one oxygenated methine [δ_H 3.38 (1H, m, H-3)], and one oxygenated methylene [δ_H 3.29/3.64 (each 1H, H₂−21)]. The ¹³C NMR data (Table 1) and HSQC spectra (Fig. S12) of 2 displayed 30 carbon signals, containing four olefinic carbons, seven methyls, nine methylenes (one oxygenated, δC 64.8), six methines (one oxygenated, δC 76.4), and four quaternary carbons. The NMR data of 2 are similar to those of (3R, 20S)−3,20-dihydroxydammar-24-ene [18], except for the presence of a hydroxymethyl and the absence of a methyl and a hydroxyl in 2. Additionally, compound 2 generated a Δ²²⁽²³⁾ double bond. These changes were also supported by the mass spectrometry-derived molecular weight difference of 2 Da. ¹H–¹H COSY correlations (Fig. S14) of H-20/H-22/H-23/H-24 and HMBC correlations from H₂−21 to C-17, C-20, and C-22, from H-22 to C-17, and C-20, and from H-23 to C-20 support the above conclusions. The ¹H–¹H COSY correlation (Fig. S14) of H-20 and H-17, along with the HMBC correlation (Fig. S13) from H-20 to C-13, indicate that the side chain is connected to C-17.

The NOESY correlation (Fig. 3) of H-13/Me-18 indicated that these hydrogens have the same orientation and were assigned as β-configured. In contrast, the cross-peaks of H-5/H-9, H-5/Me-28, H-9/Me-30, and Me-30/H-17 implied the α-orientation of those hydrogens. To define the orientation of 20-H in 2, NMR calculations with DP4 + analysis were performed. As a result, (20S*)−2 matched well with the experimental NMR data, with a DP4 + probability of 99.99% (Fig. 5). The coupling constant (³J_{22, 23} = 15.1 Hz) and NOESY correlation (Fig. 3) of H-22/H-24 revealed the E configuration of the Δ²²⁽²³⁾ double bond. Furthermore, comparison of the experimental ECD spectrum with the computed ECD curves (Fig. 4) established the absolute configuration of 2 as 3R, 5R, 8R, 9R, 10R, 13R, 14R, 17S, 20S. Finally, it was named as sinomenoid B.

Fig. 5.

¹³C NMR calculation results of 2, 3, 5, and 6. A The linear correlation curve between the measured and calculated.¹³C NMR chemical shift values of the two potential isomers of compound 2. B The linear correlation curve between the measured and calculated ¹³ C NMR chemical shift values of the two potential isomers of compound 3. C The linear correlation curve between the measured and calculated ¹³ C NMR chemical shift values of the two potential isomers of compound 5. D The linear correlation curve between the measured and calculated ¹³ C NMR chemical shift values of the two potential isomers of compound 6

The molecular formula of compound 3 was identified as C₃₀H₅₀O₂ via HRESIMS ion at m/z [M + H]⁺ 443.3878 (calcd. for C₃₀H₅₁O₂⁺, 443.3889). The ¹³C NMR data of 3 (Table 1) closely resembled those of 2, except that upfield shifts of 3.4, 4.0, and 3.1 ppm were observed for C-16, C-20, and C-22, respectively, in 3 with respect to 2.

The presence of the NOESY correlations (Fig. 3) of H-5/H-9, H-5/Me-28, H-9/Me-30, and Me-30/H-17 revealed the α-orientation of these hydrogens, while the NOESY correlation of H-13/Me-18 exhibited that they are β-configured. NMR calculations combined with DP4 + analysis revealed a 98.93% probability for (20R*)−3 (Fig. 5), which was used to determine the orientation of 20-H in compound 3. The NOESY correlation between H-22 and H-24 (Fig. 3), together with the coupling constant (³J_{22, 23} = 15.1 Hz), revealed the E configuration of the Δ²²⁽²³⁾ double bond. The experimental ECD spectrum of 3 was consistent with the calculated one (3R,5R,8R,9R,10R,13R,14R,17S,20R)−3 (Fig. 4). Therefore, the absolute configuration of 3 was defined as 3R, 5R, 8R, 9R, 10R, 13R, 14R, 17S, 20R. Finally, 3 was named as sinomenoid C.

Compound 5 possessed the molecular formula of C₁₉H₂₂O₆ based on its HRESIMS data at m/z 347.1485 [M + H]⁺ (calcd. for C₁₉H₂₃O₆⁺, 347.1495). The NMR data (Table 2) of 5 were similar to those of the known compound heliophenanthrone [19], except for the appearance of two methoxy groups and a hydroxyethyl group along with the absence of two methyls. The ¹H–¹H COSY correlation (Fig. S32) of H₂−1′/H₂−2′ and the HMBC correlations (Fig. S31) from OCH₃−5 to C-5, from OCH₃−6 to C-6, from H₂−1′ to C-7, and from H₂−2′ to C-8 proved this deduction. Furthermore, a methoxy located at C-2 in heliophenanthrone was shifted to C-3 in compound 5, which was supported by the HMBC correlation (Fig. S31) from OCH₃−3 to C-3. The NOESY correlation (Fig. 3) of H-1/H-3 indicated these hydrogens were cofacial. To further determine the relative configuration of C-1 and C-3, two possible isomers ((1R*, 3S*)-5 and (1R*, 3R*)−5) were subjected to NMR calculations with DP4 + analysis. The result showed (1R*, 3S*)−5 had a 100.00% probability (Fig. 5).

Table 2.

¹H (400 MHz) and ¹³C (100 MHz) NMR spectroscopic data of 5 and 6 measured in CDCl₃

No	5		6
	δH (J, Hz)	δC	δH (J, Hz)	δC
1	4.89, dd (5.4, 2.9), 1H	67.6	5.11, t (5.5), 1H	67.0
2	2.64, ddd (14.4, 5.8, 2.9), 1H	40.2	2.55, dd (6.7, 5.5), 2H	41.3
	2.71, ddd (14.4, 5.4, 3.1), 1H
3	4.21, dd (5.8, 3.1), 1H	83.1	4.54, t (6.7), 1H	80.7
4		203.5		202.1
4a		129.3		129.8
4b		126.5		126.4
5		142.6		142.4
6		150.2		150.1
7	7.24, s, 1H	117.3	7.20, s, 1H	117.3
8		131.2		131.3
8a		128.3		128.3
9	8.06, d (8.7), 1H	128.5	7.98, d (8.7), 1H	128.3
10	7.40, d (8.7), 1H	124.3	7.34, d (8.7), 1H	122.5
10a		144.0		143.6
1′	3.29, t (6.5), 2H	36.4	3.24, t (6.7), 2H	36.3
2′	3.93, t (6.5), 2H	63.3	3.87, t (6.7), 2H	63.3
3-OCH3	3.59, s, 3H	58.3	3.59, s, 3H	58.3
5-OCH3	3.98, s, 3H	59.8	3.96, s, 3H	60.0
6-OCH3	3.98, s, 3H	56.7	3.95, s, 3H	56.7

Furthermore, the absolute configuration of 5 was assigned as 1R, 3S by comparison of its calculated and experimental ECD spectra (Fig. 4). Subsequently, this assignment was unambiguously confirmed by single-crystal X-ray diffraction analysis (Fig. 6). Accordingly, it was identified as (1R, 3S)−2, 3-dihydro-1-hydroxy-3, 5, 6-trimethoxy-8-(2′-hydroxyethyl)−4(1H)-phenanthrenone.

Fig. 6.

X-ray ORTEP drawing of compound 5

Compound 6 had a molecular formula of C₁₉H₂₂O₆, as supported by HRESIMS at m/z 369.1304 [M + Na]⁺ (calcd. for C₁₉H₂₂O₆Na⁺, 369.1314). The ¹H and ¹³C NMR data (Table 2) of 6 exhibited high similarity to those of 5, except that C-3 (δ_C 80.7) was shifted upfield compared to that in 5 (δ_C 83.1, C-3), implying that compounds 5 and 6 could be stereoisomers. Furthermore, NMR calculations combined with DP4 + analysis revealed a 100.00% probability for (1R*, 3R*)−6 (Fig. 5). The absolute configuration of 6 was identified as 1R, 3R by comparing the calculated and experimental ECD spectra (Fig. 4). Finally, it was identified as (1R, 3R)−2, 3-dihydro-1-hydroxy-3, 5, 6-trimethoxy-8- (2′-hydroxyethyl)−4 (1H)-phenanthrenone.

Compound 13 possessed a molecular formula of C₁₈H₂₂ClNO₅, based on its HRESIMS peaks at m/z 368.1254 [M + H]⁺ (calcd. for C₁₈H₂₃ClNO₅⁺, 368.1265). The presence of a chlorine atom in 13 was inferred from the 3:1 ratio of [M + H]⁺ to [M + H + 2]⁺ in the mass spectrum [20]. The ¹H NMR data (Table 3) of 13 displayed three methoxy groups [δ_H 3.62 (3H, s, 6-OCH₃), 4.09 (3H, s, 7-OCH₃), and 3.88 (3H, s, 12-OCH₃)], and one olefinic proton [δ_H 5.30 (1H, s, H-13)]. The ¹³C NMR data (Table 3) and HSQC (Fig. S48) indicated 18 carbon resonances categorized to four olefinic carbons (δ_C 136.8, 163.9, 192.6, and 106.3), two carbonyl groups (δ_C 194.3 and 205.9), three methoxy groups (δ_C 60.9, 61.7, and 59.9), five methylenes, one methine, and three quaternary carbons. The NMR data (Table 3) of 13 closely resembled those of dauricumidine (14) [21], except for the absence of a hydroxyl group and the upfield shift of C-11 (δ_C 37.3) in 13. The HMBC correlations (Fig. S49) from H₂−11 to C-10, C-12, C-14, and C-15 further proved the above deduction. In the NOESY spectrum (Fig. S51), the correlation of H-9 with H-3β suggested an α-configuration for H-9. The calculated ECD spectrum of (9S, 10S, 15R, 16R)−13 was consistent with the experimental one (Fig. 4). Therefore, the absolute configuration of 13 was assigned as 9S, 10S, 15R, 16R. Finally, 13 was named as 11-dehydroxy-dauricumidine.

Table 3.

¹H (400 MHz) and ¹³C (100 MHz) NMR spectroscopic data of 13 measured in CD₃OD

No	δH (J, Hz)	δC
2	2.89, m, 1H	44.6
	2.77, m, 1H
3	1.81, m, 1H	43.1
	1.68, m, 1H
4	2.37, d (17.8), 1H	44.8
	2.27, d (17.8), 1H
5		194.3
6		136.8
7		163.9
8	2.73, m, 1H	47.4
	2.31, m, 1H
9	4.40, dd, (12.0, 7.1), 1H	60.8
10		66.0
11	2.82, d (18.2), 1H	37.3
	2.68, d (18.2), 1H
12		192.6
13	5.30, s, 1H	106.3
14		205.9
15		53.6
16		71.7
6-OCH3	3.62, s, 3H	60.9
7-OCH3	4.09, s, 3H	61.7
12-OCH3	3.88, s, 3H	59.9

The known compounds (4, 7–12, 14, and 15–18) were identified as coggygrnoid C (4) [17], sinocutumine J (7) [2], bianfugenine (8) [22], 2, 3-dihydrodauriporphine (9) [23], menisporphine (10) [24], daurioxoisoporphine B (11) [25], sinomenine (12) [25], dauricumidine (14) [21], 7-hydroxy-6-methoxy-1-(4-hydroxybenzyl)−3,4-dihydroisoquinoline (15) [26], 1-(4-hydroxybenzoyl)−6-methoxy-1,2,3,4-tetrahydroisoquinolin-7-ol (16) [27], longifolonine (17) [28], and juzirine (18) [28] by comparing of their NMR data with those reported in the literature.

Compounds exhibited anti-inflammatory effects

NO is an important chemical transmitter involved in the regulation of numerous key physiological processes [29]. The anti-inflammatory activities of the isolates were assessed by measuring NO production inhibition via the Griess assay, using indomethacin as the positive agent. Compounds 1 and 2 demonstrated excellent anti-inflammatory effects. Nine compounds (1, 2, 4, 5, 8, 9, 11, 13, and 17) showed anti-inflammatory activities superior to that of indomethacin (Table 4).

Table 4.

IC₅₀ values of NO inhibition of compounds 1–18

Compounds	IC50 (μM)	Compounds	IC50 (μM)
1	7.18 ± 0.80	10	46.42 ± 3.97
2	5.04 ± 1.81	11	17.88 ± 1.26
3	75.58 ± 3.20	12	41.99 ± 2.88
4	14.14 ± 0.79	13	32.79 ± 4.59
5	15.38 ± 0.70	14	44.89 ± 3.04
6	> 80	15	51.04 ± 1.09
7	51.65 ± 2.81	16	> 80
8	32.88 ± 1.04	17	32.78 ± 1.72
9	25.66 ± 1.96	18	62.32 ± 2.95
Indomethacin	33.01 ± 2.46

Molecular docking and MD simulations

Compound 2 was found to be the most potential anti-inflammatory agent among the tested compounds. The binding energy of iNOS to compound 2 was −9.0 kcal/mol. As shown in Fig. 7A, within the iNOS binding pocket: (i) the C-21 hydroxyl group formed hydrogen bonds with the ASP376 and TYR367 residues; (ii) the two six-membered rings showed hydrophobic interactions with the VAL346 residue; (iii) the C-26 formed an alkyl interaction with the ARG375 residue, C-28 with the CYS194 residue, C-29 with the VAL346, MET349, and PHE363 residues, and C-30 with the VAL346 residue.

Fig. 7.

Molecular docking and MD simulation of iNOS with 2. A 2D and 3D binding patterns of compound 2 and iNOS. B RMSD values of iNOS with and without compound 2. C RMSF values of iNOS with and without compound 2

The stability of the iNOS–compound 2 complex was evaluated using molecular dynamics simulations and by calculating the RMSD and RMSF parameters. As shown in Fig. 7B, approximately 18 ns after the start of the simulation, the RMSD value changed very little, indicating that the system had reached an equilibrium state. The average RMSD value of the iNOS-compound 2 complex was lower than that of free iNOS, suggesting that iNOS became more stable after binding to 2. As shown in Fig. 7C, the flexibility of residues in several regions, including residues 180–200, 340–350, and 360–380, was noticeably reduced in the presence of compound 2.

Structure–activity relationship analysis of anti-inflammatory compounds

Analysis of the relationship between the anti-inflammatory activity and structures of 1–4, 7–11, and 15–18 revealed the following structure–activity relationships: (i) the 3-OH and 20α-H groups act as critical structural elements in triterpenoids; (ii) the 4-OCH₃ and 6-NH₂ groups in aporphine alkaloids may serve as essential functional groups; (iii) the C = N and C = O double bonds are recognized as key bioactive groups in benzyl tetrahydroisoquinoline alkaloids.

Among triterpenoids, compound 4 with a 3-C = O group showed weaker anti-inflammatory effect than compound 1 with 3-OH. Meanwhile, compound 2, possessing a 20α-H, exhibited stronger anti-inflammatory effect than 3 with a 20β-H. Among aporphine alkaloids, compound 7 without a 4-OCH₃ group showed weaker anti-inflammatory activity than 11, indicating that the 4-OCH₃ group contributes more to the anti-inflammatory effects. Compound 8 with a 4-OCH₃ group showed more potent anti-inflammatory activity than 10, further supporting the essential role of 4-OCH₃ in the anti-inflammatory activity. Compound 11 with a 6-NH₂ exhibited stronger anti-inflammatory activity than 8 with a 6-OCH₃, suggesting that 6-NH₂ may be more effective in enhancing anti-inflammatory activity than 6-OCH₃. Compound 8 with a Δ²⁽³⁾ double bond exhibited weaker anti-inflammatory activity than 9. Among benzyl tetrahydroisoquinoline alkaloids, compound 15 without Δ³⁽⁴⁾ double bond possessed more potent anti-inflammatory activity than 18 with a Δ³⁽⁴⁾ double bond, which further supported the Δ³⁽⁴⁾ double bond was a deactivating group to the anti-inflammatory activity. Compound 17 with a 11-C = O group exhibited more significant anti-inflammatory activity than 15, which indicated that 11- C = O group may be an activating group. Compound 16, with a C-N single bond between positions 1 and 2, exhibited weaker anti-inflammatory activity than 17 containing a C = N double bond at the same positions, demonstrating that the C = N double bond contributed more to the anti-inflammatory activity.

Compounds sensitized HeLa/PTX cells to paclitaxel

No significant reduction in cell viability was observed with increasing concentrations of paclitaxel alone, demonstrating resistance of HeLa/PTX cells to paclitaxel, as presented in Fig. 8. Notably, co-treatment with paclitaxel and compounds (1–4, 7) resulted in a significant reduction in cell viability.

Fig. 8.

Effect of combination treatment of compounds (1–4, 7) and paclitaxel on cell viability. HeLa/PTX cells were treated with paclitaxel alone or in combination with the compounds (50 μM) for 72 h. The Tukey test was used to compare the cell viability in the combination treatment with that of paclitaxel treatment alone. The asterisks indicate a significant difference (** p < 0.01)

Combination treatment inhibited cell proliferation

Compound 4 was selected for further analysis. As illustrated in Fig. 9, co-treatment with compound 4 and paclitaxel markedly inhibited colony formation.

Fig. 9.

Cell colony formation in HeLa/PTX cells treated with paclitaxel alone or in combination with 4 for 12 h. Data are shown as the mean ± SD (n = 3). The asterisks indicate a significant difference (** p < 0.01)

Combination treatment induced apoptosis in HeLa/PTX cells

We analyzed nuclear morphology to investigate whether compound 4 enhanced paclitaxel-induced apoptosis in HeLa/PTX. No significant effects were observed with treatment using compound 4 alone or paclitaxel alone in HeLa/PTX cells, but their combination clearly induced apoptotic bodies (Fig. 10A). These findings suggest that the combined treatment exerts anti-tumor effects by inducing apoptosis.

Fig. 10.

Apoptotic effect of the combination of 4 and paclitaxel on HeLa/PTX cells. The cells were treated with paclitaxel alone or in combination with 4 for 24 h. A Nuclear morphology analysis by Hoechst 33,258 staining (scale bar = 40 μm). B The expression levels of apoptosis-related genes were analyzed by the western blot assay. The asterisks indicate a significant difference (* p < 0.05 and ** p < 0.01)

Combination treatment changed the expression of apoptosis-associated proteins

PARP, a key biomarker of apoptosis, exhibited enhanced cleavage upon combined treatment with 4 and paclitaxel. A significant increase in the Bax/Bcl-2 ratio was induced by the combination of compound 4 and paclitaxel, serving as an important factor for apoptosis in HeLa/PTX cells (Fig. 10B). These findings further demonstrate that the combination treatment promoted apoptosis in HeLa/PTX cells.

Conclusion

In summary, six new compounds, including three triterpenoids (1–3), two phenanthrenes (5 and 6), and one alkaloid (13), were obtained from the stems of S. acutum. Five compounds (11, 15–18) were described from this plant for the first time. Bioactivities of all compounds (1–18) obtained were assessed. Nine compounds showed significantly stronger anti-inflammatory activities than the positive control indomethacin. Four triterpenoids and one alkaloid evidently enhance paclitaxel effect against cancer cells. Furthermore, the combination of compound 4 and paclitaxel significantly facilitated apoptosis in cervical cancer cells. These results broaden the structural diversity of S. acutum and provide critical insights for developing the novel anti-inflammatory drugs and sensitizing agents against cancer.

Materials and methods

General methods

A Nexus 470 FT-IR spectrometer was utilized to acquire the IR spectra. UV spectra were obtained on a Shimadzu UV-1700. Bruker BIOSPIN AVANCE 400 MHz and 600 MHz spectrometers were employed to record the NMR spectra. ECD data were recorded on a JASCO J-810 spectrometer. HRESIMS spectra were acquired using a Thermo Fisher LTQ Orbitrap XL spectrometer.

Plant material

The stems of S. acutum were collected in Hubei Province, China, in 2021, and authenticated by Prof. Jianqing Yu of the School of Pharmaceutical Sciences at Wuhan University.

Extraction and isolation

The stems of S. acutum (8 kg) were immersed in 25 L of 90% ethanol at room temperature to yield 700 g of crude extract, which was subjected to a DA-201 macroporous resin column using a 40–100% methanol (MeOH) in H₂O gradient, affording four fractions (Fr.1–Fr.4). Column chromatography (CC) was performed on Fr. 2 (8 g) using a gradient system of PE, DCM, and EA (PE: DCM, 1:6; DCM: EA, 50:1–0:1), yielding Fr.2.1–Fr.2.9. Fr.2.4.1 to Fr.2.4.7 were obtained by subjecting Fr.2.4 to an ODS column (MeOH–H₂O, 30–100%). Purifying Fr.2.4.1 repeatedly by CC over silica gel (DCM:MeOH, 22:1) yielded 12 (20 mg), 14 (8 mg), and 17 (14 mg). Purification of Fr.2.4.2 by CC over silica gel, eluting with DCM–MeOH (20:1), yielded 13 (60 mg). 5 (7 mg, t_R = 15 min) and 6 (8 mg, t_R = 10.5 min) were obtained by purifying Fr.2.4.3 via semipreparative HPLC. Fr.2.5 was subjected to an ODS column (MeOH–H₂O, 30–100%) and then further isolated by silica gel CC (DCM: MeOH, 13:1) to yield 15 (37 mg) and 16 (13 mg). Silica gel CC (DCM: MeOH, 12:1) was used to isolate compound 18 (14 mg) from Fr. 2.6.7. Silica gel column chromatography was applied to purify Fr.3 (14 g) using eluents of PE: DCM (1:4) and DCM:MeOH (30:1–0:1), yielding Fr.3.1–Fr.3.9. Compound 10 (4 mg) was isolated by recrystallization from Fr.3.2.6 in a methanol. Purification of Fr.3.2.8 via silica gel CC with PE: DCM (4:1) as the eluent resulted in the isolation of compound 8 (87 mg). Fr.4 (32 g) was applied to silica gel CC (DCM: MeOH, 20:1–0:1) to yield 12 fractions (Fr.4.1–Fr.4.12). Semi-preparative HPLC was employed for the isolation of compound 9 (5 mg, t_R = 17 min) from Fr.4.7. 2 (11 mg, t_R = 12 min), 3 (9 mg, t_R = 13 min), and 4 (16 mg, t_R = 4.5 min) were obtained by isolating Fr.4.9 via semi-preparative HPLC. Fr.4.10 was purified by silica gel CC (PE: EA, 2.5:1), affording two compounds: 1 (5 mg) and 11 (6 mg). Compound 7 (4 mg) was obtained by separating Fr.4.11 using silica gel CC (PE: EA, 5:1).

Spectroscopic data

Sinomenoid A (1). Colorless oil; ${[\alpha ]}_D^{20}$ − 21 (c 0.10, MeOH); UV (MeOH) λ_max (log ε) 280 (4.17) nm; IR (KBr) ν_max 3376, 2926, 1629, 1383, 1089, 981 cm⁻¹; NMR data: Table 1; HRESIMS m/z 479.3488 [M + Na]⁺ (calcd for C₃₀H₄₈O₃Na, 479.3501).

Sinomenoid B (2). Colorless oil; ${[\alpha ]}_D^{20}$ + 25 (c 0.10, MeOH); UV (MeOH) λ_max (log ε) 239 (4.07) nm; IR (KBr) ν_max 3373, 2946, 2866, 1450, 1382, 1034, 990 cm⁻¹; NMR data: Table 1; HRESIMS m/z 443.3878 [M + H]⁺ (calcd for C₃₀H₅₁O₂, 443.3889).

Sinomenoid C (3). Colorless oil; ${[\alpha ]}_D^{20}$ + 28 (c 0.10, MeOH); UV (MeOH) λ_max (log ε) 239 (4.37) nm; IR (KBr) ν_max 3475, 2941, 1447, 1382, 1067, 988 cm⁻¹; NMR data: Table 1; HRESIMS m/z 443.3878 [M + H]⁺ (calcd for C₃₀H₅₁O₂, 443.3889).

(1R, 3S)−2, 3-Dihydro-1-hydroxy-3, 5, 6-trimethoxy-8-(2′-hydroxyethyl)−4(1H)-phenanthrenone (5). Yellow bulk crystal; mp 154.8–156.5 °C; ${[\alpha ]}_D^{20}$ + 26 (c 0.10, MeOH); UV (MeOH) λ_max (log ε) 201 (4.38), 220 (4.56), 253 (4.40) nm; IR (KBr) ν_max 3397, 2946, 1602, 1424, 1343, 1261, 1108, 1037 cm⁻¹; NMR data: Table 2; HRESIMS m/z 347.1485 [M + H]⁺ (calcd for C₁₉H₂₃O₆, 347.1495).

(1R, 3R)−2, 3-Dihydro-1-hydroxy-3, 5, 6-trimethoxy-8-(2′-hydroxyethyl)−4(1H)-phenanthrenone (6). Yellow oil; ${[\alpha ]}_D^{20}$ + 31 (c 0.10, MeOH); UV (MeOH) λ_max (log ε) 202 (4.19), 220 (4.38), 254 (4.21) nm; IR (KBr) ν_max 3428, 2942, 1602, 1423, 1349, 1262, 1103, 1045 cm⁻¹; NMR data: Table 2; HRESIMS m/z 369.1304 [M + Na]⁺ (calcd for C₁₉H₂₂O₆Na, 369.1314).

11-Dehydroxy-dauricumidine (13). Colorless oil; ${[\alpha ]}_D^{20}$ + 73 (c 0.10, MeOH); UV (MeOH) λmax (log ε) 246 (4.42) nm; IR (KBr) ν_max 2948, 1690, 1455, 1429, 1362, 1324, 1270, 1244, 1192, 1141, 1062, 991 cm⁻¹; NMR data: Table 3; HRESIMS m/z 368.1254 [M + H]⁺ (100) (calcd for C₁₈H₂₃NO₅Cl, 368.1265).

X-ray diffraction analysis

The structure of compound 5 (CCDC 2476411) was elucidated using SHELXT. The crystallographic data for this compound have been deposited at the CCDC and can be downloaded from https://www.ccdc.cam.ac.uk.

Crystallographic data for compound 5: C₁₉H₂₂O₆, M = 346.36, a = 10.6309(11) Å, b = 11.4283(9) Å, c = 14.1914(14) Å, α = 90°, β = 90°, γ = 90°, V = 1724.2(3)ų, T = 301.48(10) K, space group P2₁2₁2₁, Z = 4, μ (Cu K^α) = 0.823 mm⁻¹, 11,893 reflections collected, 3405 independent reflections (R_int = 0.0300, R_sigma = 0.0212). The final R₁ values were 0.0500 (I > = 2σ (I)) and wR (F²) values were 0.1394 (I > = 2σ (I)). The final R₁ values were 0.0591 (all data). Flack parameter = 0.11(11).

NMR calculations

NMR calculations with DP4 + analysis were conducted for compounds 2 and 3 using published methods [30].

ECD calculations

Preliminary geometries for six compounds (1–3, 6, and 13) were obtained using Chem3D software, and ECD calculations were carried out using established methods, as previously reported [31]. The calculated ECD spectra were obtained through SpecVis1.71 software by performing Boltzmann-weighting averaging of conformers with UV correction.

Cell culture

RAW 264.7 cells were purchased from the China Center for Type Culture Collection. HeLa/PTX cells were established in our laboratory as reported previously [32]. All cells were maintained in DMEM containing 10% FBS and 1% P/S at 37 °C with 5% CO₂.

Anti-inflammatory assay

RAW 264.7 macrophages were stimulated with LPS (1 μg/mL) in combination with different isolates for 24 h. NO production was subsequently assessed using Griess reagents as previously described [14].

Molecular docking and dynamics simulations analysis

The X-ray crystal structure of iNOS (PDB ID: 3E6T) was obtained from the Protein Data Bank (https://www.rcsb.org/). AutoDock Vina 1.5.6 was employed to perform docking calculations [33]. The docked complexes were then examined and rendered using Discovery Studio 4.5 [34] and PyMOL 3.0 software.

GROMACS 2020.6 was employed to conduct molecular dynamics (MD) simulations. In the system, TIP3P was chosen as the water model, and charge neutralization was achieved by adding Na⁺ and Cl⁻. Prior to the simulations, energy minimization was performed first, followed by pre-equilibration steps under NVT and NPT conditions. Finally, MD simulations were performed with a total simulation time of 50 ns.

Cell viability assay

HeLa/PTX cells (5 × 10³ cells/well) were treated with drugs for 72 h. Following treatment, cell viability was measured using the MTT assay.

Colony formation assay

HeLa/PTX cells were treated with drugs for 12 h, then cultured without drug for 14 days as previously described [35]. Viable colonies were stained and quantified.

Nuclear morphology assay

After drug treatment, cell staining was performed using Hoechst 33,258 as previously described [32].

Western blot assay

Samples were processed using 10% SDS-PAGE and then transferred to PVDF. After blocking, the membrane was incubated sequentially with primary antibodies overnight and with secondary antibodies for 2 h. A chemiluminescence system was used to visualize the image.

Statistical analysis

Data are shown as the mean ± SD (n = 3). Statistical analyses were performed using GraphPad Prism 8.3.0 and Origin 2024. One-way ANOVA followed by Tukey’s multiple comparisons test was employed to evaluate statistical significance (P < 0.05).

Author contributions

Xilei Wang: Conceptualization, Investigation, Data curation, Visualization, Writing-original draft. Chengjing Zhu: Investigation, Data curation. Xinyu Du: Investigation, Data curation. Qiang Wang: Formal analysis, Data curation. Jianqing Yu: Conceptualization, Funding acquisition, Writing-review & editing, Project administration.

Funding

National Natural Science Foundation of China, 32270409, Jianqing Yu

Data availability

The data generated in this study were included in the article and its supplementary materials.

Declarations

Competing interests

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.