Chemical constituents of Lycium barbarum leaves and their anti-rheumatoid arthritis activity in vitro

Abstract

Two new together with 32 known compounds were isolated from the leaves of Lycium barbarum. Their structures were elucidated using 1D and 2D NMR, HRESIMS, and ECD spectroscopic techniques. Compounds 1–34 were evaluated for their anti-rheumatoid arthritis activities in a lipopolysaccharide (LPS)-induced MH7A cells inflammatory model. As a result, compounds 1–3, 6, 8, 10, 14, 17–19, 29 and 31 inhibited the activity of lactate dehydrogenase (LDH) and nitric oxide (NO) at concentrations 20 μM. Among them, compound 1 showed the best effectiveness, with inhibition rates of 46.7% for NO and 32.8% for LDH.

Graphical Abstract

Supplementary Information

The online version contains supplementary material available at 10.1007/s13659-025-00516-9.

Introduction

Lycium barbarum L., a deciduous shrub from the Solanaceae family, is a renowned traditional plant for both medicine and food [1]. The leaves of L. barbarum were proven to possess various biological effects such as boosting immunity, reducing heat, alleviating rheumatic pain, quenching thirst, promoting saliva secretion, and improving eyesight [2]. In Asian countries, the leaves served as functional vegetables, commonly used in soup making, stir-frying, and as herbal teas [3].

Rheumatoid arthritis (RA), characterized by persistent inflammation and abnormal proliferation of fibroblast-like synoviocytes (FLS), was a chronic inflammatory condition that could result in joint destruction and disability [4]. The treatment of RA included glucocorticoids, disease-modifying antirheumatic drugs, nonsteroidal anti-inflammatory drugs and biologics [5]. However, these medications were expensive and came with serious side effects [6]. Then, it was essential to explore natural products for anti-RA drugs that were effective and had low toxicity. Folk applications suggested the therapeutic effects of L. barbarum leaves on rheumatoid arthritis, but its bioactive compounds remained unknown.

Results and discussion

Structural elucidation of compounds

A total of 34 compounds were isolated from leaves of L. barbarum, including one alkaloid (1), 10 terpenoids (2–11), 8 lignans (12−19), 11 phenolic acids (20−30), and four other compounds (31−34) with compounds (1 and 2) being newly discovered (Fig. 1). Compounds 4 and 8–19 were first isolated from L. barbarum leaves.

Fig. 1.

Structures of 1–34

The molecular formula of lycibarin A (1) was assumed to be C₁₆H₁₇NO₄ based on HRESIMS peak at m/z 288.1221 [M + H]⁺(calcd for C₁₆H₁₇NO₄⁺, 288.1230) and the ¹³C NMR spectrum, requiring 9 degrees unsaturation. Its IR spectrum displayed characteristic bands assignable to the carbonyl group (1766 cm⁻¹), and olefinic (1652 cm⁻¹). The 1D and 2D NMR spectroscopic data suggested that 1 was structurally similar to 6,7,8,8a-tetrahydro-2-phenyl-4H-pyrrolo[2, 1-b][1, 3]oxazin-4-one [7] with the notable difference being the presence of an additional ethyl acetate fragment in 1. Meanwhile, the COSY correlation of δ_H 1.23 (H-2'') / 4.23 (H-1''α), and the HMBC correlation between δ_H 4.23 (H-1''α) and δ_C 170.5 (C-9) supported the presence of ethyl acetate group, while the ethyl acetate substitution at C-6 was indicated by the HMBC correlations of δ_H 2.55 (H-7) with δ_C 94.8 (C-6) and 170.5 (C-9). Then the structure of 1 was elucidated as illustrated in Fig. 2, and its chiral carbon was deduced to be 6R by comparing the experimental and calculated electronic circular dichroism (ECD) spectra (Fig. 3).

Fig. 2.

The key ¹H−¹H COSY and HMBC correlations of compounds 1 and 2

Fig. 3.

The experimental and calculated ECD spectra of compounds 1 and 2

Compound 2 possessed a molecular formula of C₁₈H₂₆O₄ by its ( +)-HRESIMS data at m/z 329.1698 [M + Na]⁺ (calcd for C₁₈H₂₆O₄Na, 329.1723). Detailed analysis of the ¹H and ¹³C NMR spectral data (Table 1) of 2 showed high similarity to those of 5,8-endoperoxy-2,3-dihydro-β-apocarotene-13-one [8]. The obvious differences between the two compounds were 2 with oxymethine signal at C-3 (δ_C 68.1) instead of methylene signal (δ_C 18.6) in 5,8-endoperoxy-2,3-dihydro-β-apocarotene-13-one, which assumed a hydroxylated derivative of 5,8-endoperoxy-2,3-dihydro-β-apocarotene-13-one for 2. Meanwhile, the COSY correlation (Fig. 2) of δ_H 4.14 (H-3) / 1.78 (H₂-2α) supported the hydroxyl substitution at C-3. The large coupling constant (J_H-11/H-12 = 15.4 Hz) suggested E configuration for ∆^11,12, and the E configuration of ∆^9,10 was supported by the NOE correlation of δ_H 5.20 (H-6) / 6.31 (H-10) in its ROESY spectrum. Additionally, NOE correlations of δ_H 5.20 (H-6) / 1.60 (H-16) and δ_H 1.60 (H-16) / 1.33 (H-15) indicated their syn orientation and temporarily assigned the α-orientation, while the β-configuration of H-3 was elucidated by the NOE correlation between δ_H 4.14 (H-3) / 1.18 (H₃-14) (Fig. 2). Moreover, the calculated ECD spectrum of (3R*, 5S*, 6S*) configuration was matched the experimental spectrum of 2, which assigned its absolute configuration (Fig. 3).

Table 1.

The ¹H NMR (600 MHz) and ¹³C NMR (150 MHz) spectral data of compounds 1 and 2 (δ in ppm and J in Hz)

NO	1a		2b
	δH (mult, J)	δC	δH (mult, J)	δC
1				34.9, s
2α		161.6, s	1.78, dd (14.0, 7.4)	48.7, t
2β			1.48, dd (14.0, 3.5)
3	5.85, s	98.5, d	4.14, m	68.1, d
4α		163.7, s	2.10, dd (13.5, 3.8)	48.9, t
4β			1.95, dd (13.5, 4.4)
5				89.5, s
6		94.8, s	5.20, br s	88.6, d
7	2.55, m	36.8, t	5.35, br s	120.2, d
8	2.10, m	21.2, t		156.3, s
8aα	3.93, dt (11.0, 7.0)	45.2, t
8aβ	3.73, dt (11.0, 7.0)	-
9	-	170.5, s		151.2 s
10			6.31, d (11.4)	125.3, d
11			7.57, dd (15.4, 11.4)	141.2, d
12			6.19, d (15.4)	131.5, d
13				201.8, s
14			1.18, s	31.9, q
15			1.33, s	29.7, q
16			1.60 s	29.6, q
17			1.86, s	13.5, q
18			2.30, s	27.4, q
1'	-	131.5, s
2'/6'	7.78, d (7.6)	127.2, d
3'/5'	7.44, t (7.6)	128.6, d
4'	7.48, t (7.6)	131.4, d
1''α	4.23, dq (10.8, 7.2)	62.4, t
1''β	4.18, dq (10.8, 7.2)
2''	1.23, t (7.2)	14.0, q

a: in CDCl₃

b: in CD₃OD

The 32 known compounds (3−34) were identified as (−)-(3S,4S)-eucomehastigmane B (3) [9], cis-3,6-dihydroxy-α-ionone (4) [10], ( +)-dehydrovomifoliol (5) [11], vomifoliol (6) [12], trans, trans-abscisic acid (7) [13], (3S,5R,6S,7E)-5,6-epoxy-3-hydroxy-7-megastigmen-9-one (8) [14], 3-hydroxy-7,8-dehydro-β-ionol (9) [15], phaseic acid (10) [16], loliolida (11) [17], ( +)-syringaresinol (12) [18], medioresinol (13) [19], (-)-pinoresinol (14) [20], (1R*,2R*,5R*,6S*)-6-(4-hydroxy-3-methoxyphenyl)-3,7-dioxabicyclo[3.3.0]octan-2-ol (15) [21], ( +)-lariciresinol (16) [22], ( +)-isolariciresinol (17) [23], meliasendanin B (18) [24], evofolin B (19) [25], vanillic acid (20) [26], 4-hydroxybenzoic acid (21) [27], syringic acid (22) [28], 4-hydroxyacetophenone (23) [29], ferulic acid (24) [30], p-coumaric acid (25) [31], caffeic acid methyl ester (26) [32], tyrosol (27) [33], ethyl 3,4-dihydroxybenzoate (28) [34], β-hydroxy propiovanillone (29) [35], 6,7-dimethoxy-4-hydroxy-1-naphthoic acid (30) [36], chrysoeriol (31) [37], scopoletin (32) [38], 5-hydroxy-4-phenyl-5H-furan-2-one (33) [39], 4,4-dimethylheptanedioic acid (34) [40].

The establishment of lipopolysaccharide (LPS)-induced MH7A inflammation model.

Previous studies showed that RA-FLS were stimulated by chemotactic factors, including IL-1β, TNF-α, and LPS [41, 42]. Additionally, LPS-induced RA-FLS displayed biological traits that were associated with RA, such as aberrant proliferation and resistance to cell death, which were similar to those of tumor cells [43]. Therefore, the immortalized RA-FLS MH7A cell line induced by LPS was utilized for in vitro studies to investigate the therapeutic effects of isolated compounds of L. barbarum leaves on RA. Following a 24-hour exposure to varying doses of LPS (1, 5, 10, 25, 50 and 100 μg/mL), the MTT test was used to assess the viability of the cells. As depicted in Fig. 4A, a peak cell viability of 120.8% was reached with an LPS concentration of 10 μg/mL, showing a significant difference compared to the control group (p < 0.01). As a result, for the subsequent stimulation, 10 μg/mL of LPS was selected.

Fig. 4.

A Cell viability of MH7A cells under stimulation of different concentrations of LPS. B-C The cell viability of compounds on MH7A cells. Compared with the control group, ** p < 0.01

Inhibitory effect of isolated compounds on proliferation, nitric oxide (NO) lactate dehydrogenase (LDH) production in LPS-induced MH7A cells.

Compounds were subjected to cytotoxicity testing, revealing no inhibition at a concentration of 20 μM for all (Fig. 4B–C). Furthermore, compounds 1–4, 6, 8–14, 16–21, 23–24, and 29–33 demonstrated substantial inhibitory effects on the proliferation bioactivity of LPS-induced MH7A cells (p < 0.05/0.01, Table S1). It was worth mentioning that the cell viabilities of compounds 1 and 2 were 112.91% and 116.47%, respectively (Fig. 5A–B). Synovial cells were considered the primary producers of NO in RA. Synovial fibroblasts were induced by pro-inflammatory cytokines to produce NO, thereby enhancing the production of inflammation [44]. LDH, a cytoplasmatic enzyme, present in essentially all organ systems is thought to be released only after cell death and local inflammation of cells may be a potential source of elevation of LDH [45]. As cells underwent proptosis, pores formed within them, resulting in an elevation of LDH levels [46]. Stimulation with LPS on MH7A cells led to elevated levels of NO and LDH production by MH7A cells compared to the control group, whereas compounds 1–3, 6, 8, 10–12, 14, 16–19, 21, 29 and 31 displayed inhibitory effects on NO (p < 0.05/0.01, Table S2), with compounds 1–3, 6, 8–10, 14, 17–19, 29 and 31–32 also showing inhibitory effects on LDH release (p < 0.05/0.01, Table S3). The compounds 1–3, 6, 8, 10, 14, 17–19, 29 and 31 exhibited effectiveness on LDH and NO indicators (p < 0.05/0.01, Table S4). Among them, compounds 1 and 2 exhibited good inhibitory activity on the two indicators, NO inhibition rates were 46.7% and 26.8% respectively, and LDH inhibition rates were 32.8% and 22.8% respectively (Fig. 5C–F).

Fig. 5.

The cell viabilities of compounds 1 and 2 on MH7A by LPS induction (A-B). The effect of compounds 1 and 2 on the production of NO (C-D) and the release of LDH (E–F) in LPS-induced MH7A. Compared with the control group, ## p < 0.01; compared with the model group, ** p < 0.01, respectively

Experimental

General experimental procedures

NMR, HRESIMS, UV, and IR spectra were obtained following the previously described methods [47].

Plant material

The origin of the plant material is detailed in the Plant Material section of Supplementary Material.

Extraction and purification

The detailed procedures for extraction and isolation can be found in the Extraction and purification section of Supplementary Material.

Lycibarin A (1). white powder; ${[\alpha ]}_D^{22}$−72.9 (0.073, MeOH); UV (MeOH) λ_max (log ε): 292 (3.80) nm; ECD (MeOH) λ_max (Δε) 210 (+ 14.60), 245 (–8.78) nm; IR v_max 1746, 1652, 1433, 1047, 693 cm^‒1; ¹H NMR and ¹³C NMR data see Table 1. HRESIMS m/z 288.1221[M + H]⁺ (calcd. for C₁₆H₁₈NO₄, 288.1230).

Lycibarin B (2). white power; ${[\alpha ]}_D^{22}$−57.2 (0.056, MeOH); UV (MeOH) λ_max (log ε): 288 (3.54) nm; ECD (MeOH) λ_max (Δε) 209 (-1.08), 252 (+ 0.32) nm; IR v_max 3434, 1713, 1630, 1459, 1384, 1149, 1081 cm^‒1; ¹H and ¹³C NMR data see Table 1; HRESIMS m/z 329.1698[M + Na]⁺(calcd. for C₁₈H₂₆O₄Na, 329.1723).

Cell culture

MH7A (RA-FLS) cell line was obtained from Jennio Biotech Co., Ltd. (Guangzhou, China) and cultured at 37℃ with 5% CO₂ in DMEM (Gbico, USA) supplemented with 10% FBS (Procell, China), 100 U/mL penicillin, and 100 μg/mL streptomycin.

LPS-induced MH7A cells inflammation model

The MTT test was used to assess the cell viability of LPS on MH7A cells [48]. Each well was initially seeded with 1.5 × 10⁴ cells, which were then spilt into control and LPS groups. The cells were incubated for 24 h before being treated to different doses of LPS (1, 5, 10, 25, 50 and 100 μg/mL) for another 24 h. Subsequently, 100 μL of MTT (Aladdin, Shanghai, China) was added at 0.5 mg/mL concentration and continued for 4 h. The supernatant was sucked out, and 100 μL DMSO was added to each well, followed by a 10-min incubation. The absorbance cell medium was measured at 490 nm using microplate reader (Molecular Devices, Shanghai, China).

Cell viability assay

In short, 100 μL of MH7A cell suspension (1.5 × 10⁵ cells/mL) was added to a 96-well plate. Upon reaching 70%-80% confluence, the cells were exposed to the compounds at a 20 μM concentration for 24 h.

Inhibition of compounds on LPS-induced MH7A cells proliferation

According to the results of viability, MH7A cells were seeded in 96-cell plates for 24 h incubation and divided into control, LPS (model group), LPS and different compounds (20 μM). All groups, except for the control, were stimulated with LPS for 24 h. The steps that followed for evaluating cell viability were in accordance with the instructions outlined in previous section [49].

NO assay

After being planted in 96-well plates at a density of 1.5 × 10⁴ cells per cell, MH7A cells were incubated for 24 h. Following this, the cells were stimulated with LPS (10 μg/mL) and treated with the tested compounds at a concentration of 20 μM. A microplate reader (Shanghai, Molecular Devices, China) was used to quantify NO at 540 nm after 50 μL of the supernatant from each well as moved to a new 96-well plate following a 24-hour incubation period [50].

LDH measurement

MH7A cells were seeded in 6-well plates and incubated with LPS and the tested compounds for 24 h. Removed the cell culture supernatant, collected the cells, and then performed the measurement according to the LDH assay kit [49].

Author contributions

Xiao-Dong Luo conceptualized the projects, revised manuscript and provided financial support. Yun-Li Zhao supervised the experiment, revised the manuscript. Zi-Jiao Wang, Bang-Yin Tan, Yun Zhao, Chang-Bin Wang finished the isolation and identification studies. Zi-Jiao Wang implemented the anti-rheumatoid arthritis studies. Zi-Jiao Wang wrote the original draft. All authors approved the final version of the manuscript.

Funding

This research received financial backing from the High-level Talent Promotion and Training Project of Kunming (grant number 2022SCP003), Project of Yunnan Characteristic Plant Screening and R&D Service CXO Platform (grant numbers 2022YKZY001, 2023YKZY001, 2023YKZY004), Yunnan Provincial Science and Technology Project at Southwest United Graduate School (No. 202302AP370006), the Major Science and Technology Project of Yunnan Province (202302AA310013), and the Guiding Funds of the Central Government for Supporting the Development of Local Science and Technology (202407AD110002). Special thanks are extended to the Analysis and Measurement Center of Kunming Institute of Botany for their technical assistance.

Data availability

The data supporting the results of this study can be obtained from the corresponding author on reasonable request.

Declarations

Competing interests

According to policy as well as my moral obligation, the corresponding author Xiao-Dong Luo is the Executive Editor-in-Chief of Natural Products and Bioprospecting. This submission represents genuine academic work and is not influenced by editorial role of Prof. Luo. To ensure impartiality and adherence to publishing ethics, the following measures have been taken: (1) Full Recusal from Editorial Handling. (2) Transparency in Peer Review.