Anti-inflammatory constituents from the root of Litsea cubeba in LPS-induced RAW 264.7 macrophages

Abstract

Context Litsea cubeba (Lour.) Pers. (Lauraceae) has long been used as a folk remedy in Traditional Chinese Medicine (TCM) for the treatment of rheumatic diseases. Previous studies from our laboratory indicated that L. cubeba extract showed anti-arthritic activity in rats.

Objective To study L. cubeba chemically and biologically and to find the potential constituents responsible for its anti-arthritic effect.

Materials and methods The compounds were isolated from the root of L. cubeba by column chromatography which eluted with PE:EtOAc gradient system, and the structures were elucidated by detailed spectroscopic data analysis; the anti-inflammatory activity of the isolated compounds was evaluated by lipopolysaccharide (LPS)-induced RAW 264.7 cells and the TNF-α and NO level were measured by ELISA (commercial kit); The iNOS and COX-2 mRNA expression were measured by RT-PCR and the phosphorylation of IκBα, IKKβ, P38 and Akt were determined by western blots.

Results A novel 9-fluorenone, 1-ethoxy-3,7-dihydroxy-4,6-dimethoxy-9-fluorenone (1), together with 4 known compounds, namely pinoresinol (2), syringaresinol (3), 9,9′-O-di-(E)-feruloyl-meso-5,5′-dimethoxysecoisolariciresinol (4) and lyoniresinol (5) were isolated from the root of L. cubeba for the first time. The IC₅₀ for NO inhibition on compounds 1 and 4 were 56.1 ± 1.2 and 32.8 ± 2.3 μM, respectively. The IC₅₀ for TNF-α inhibition were 28.2 ± 0.9 and 15.0 ± 1.0 μM, respectively. Both 1 and 4 suppress mRNA expression of iNOS, COX-2 and protein phosphorylation of IκBα, IKKβ in LPS-induced RAW 264.7 cells.

Discussion and conclusion Compounds 1 and 4 isolated from L. cubeba exhibited potent anti-inflammatory activity through the NF-κB signal pathway.

Keywords:

• Anti-arthritic
• IκBα
• IKKβ
• rheumatoid arthritis
• RT-PCR
• Traditional Chinese Medicine

Introduction

Rheumatoid arthritis (RA) is a common, immune-mediated disease characterized by chronic progressive inflammation and destruction of joints and associated structures, as well as systemic symptoms (Tarner & Müller-Ladner 2008; Praveen & Suchita 2013). It is generally considered as a disease caused by genetic and external factors involved in predisposing genes, lifestyle factors such as smoking, infectious agents, occupational exposures and exposure to pets (Yang et al. 2010). Currently, the pathogenesis of RA has not been completely understood, however, it is clear that inflammatory cytokines such as TNF-α, IL-1β, IL-6 and IL-17A as well as inflammatory mediators such as cyclooxygenases (COX) and NO play an important role (Annunziato et al. 2009). The glucocorticoids, non-steroidal anti-inflammatory drugs (NSAIDs), immunosuppressant and biologic are the main drugs employed for the treatment of RA in clinic. However, the use of these drugs always associated with long-term side-effects, toxicity, high cost and incurs hypersensitivity to medications and infections (Feldmann et al. 2001). In fact, herbal medicine is being widely used virtually around the world for the treatment of rheumatic and arthritic diseases, in recent decades, considerable advances have been made in both clinical and basic research on the treatment of RA (Soeken et al. 2003; Lin et al. 2013).

Litsea cubeba (Lour.) Pers. (Lauraceae) has long been used as a folk remedy in Traditional Chinese Medicine (TCM) for the treatment of rheumatic diseases (Feng et al. 2009). Many pharmacological activities of L. cubeba extract and its compounds have been reported such as anticancer, antibacterial, antioxidant and anti-inflammatory immunological activities. Our previous study also proved that L. cubeba exhibited remarkable anti-arthritic activity on Freund’s adjuvant induced arthritis in rats (Lin et al. 2013). So far, the anti-arthritic active constituent has not been discovered yet. As part of our ongoing effort to discover the bioactive compounds from this medical plant, the chemical constituents in the root of L. cubeba were investigated in the present study. Macrophages play a central role in organizing the release of inflammatory mediators, including nitric oxide (NO) and tumor necrosis factor-alpha (TNF-α). Due to their highly reproducible response to lipopolysaccharide (LPS), the RAW 264.7 mouse macrophage cell line is widely used for inflammation studies (Ji et al. 2012). In the present study, the anti-inflammatory activities were evaluated in vitro employing LPS-induced RAW 264.7 cells. Furthermore, the probable mechanism was examined by real-time PCR and western blot analysis.

Materials and methods

General experimental procedures

NMR spectra were measured with Bruker Avance 600 spectrometer (Billerica, MA) with solvent peak as references. IR Spectra were recorded in KBr pellets on a Bruker Vector 22 spectrophotometer (Billerica, MA). ESI-MS was taken on a Waters Q-TOF MS (Milford, MA). UV Spectra were obtained in MeOH on a Varian Cary Eclipse 300 spectrophotometer (Palo Alto, CA). Column chromatography was performed on silica gel (100–200 or 200–300; Qingdao Marine Chemical Factory) and Sephadex LH-20 (Amersham Biosciences, Uppsala, Sweden). TLC was carried out on silica-gel plates (Yan-Tai Institute of Chemical Technology, 0.2 mm thickness, 5 cm × 10 cm), then visualised under UV light and by spraying with 10% H₂SO₄, followed by heating.

Plant material

The roots of L. cubeba (dry slices) were purchased from an Herbal Medicinal Materials Company of Anton Pharmaceutical Co., Ltd (Bozhou, Anhui Province, China), and identified by Prof. B.K. Huang (The Second Military Medical University, Republic of China). A voucher specimen has been deposited in School of Pharmacy, Second Military Medical University, Shanghai, China (No. 20110820).

Extraction and purification

Sliced dried roots of L. cubeba (50 kg) were extracted three times (3 h each time) with 75% ethanol. The extract was then combined and evaporated to dryness under reduced pressure, which yielded 650 g residues. The residues were then suspended in water (5 L) and partitioned with petroleum ether (PE), CH₂Cl₂, EtOAc and n-BuOH successively, to obtain respective fractions (PE fraction 50 g, CH₂Cl₂ fraction 200 g, EtOAc fraction 56 g, n-BuOH fraction 200 g). The CH₂Cl₂ fraction (50 g) was subjected to silica gel column chromatography (1 kg) and eluted with PE:EtOAc gradient system (25:1, 20:1, 15:1, 10:1, 5:1, 2.5:1, 1:1, successively) to give seven fractions (Frs.C1-C7). Fraction C5 was fractionated on silica gel column chromatography eluted with PE:EtOAc (7:1) and purified by Sephadex LH-20 with MeOH to afford compounds 1 (70 mg), 2 (50 mg) and 3 (200 mg). Fraction C4 was fractionated on silica gel column chromatography eluted with PE:EtOAc (20:1) and further purified by preparative HPLC to give 4 (35 mg) and 5 (22 mg).

Cell culture

The RAW 264.7 cell was grown in Dulbecco’s modified Eagle’s medium (Hyclone) supplemented with 10% fetal bovine serum (Gibco) and 100 units/mL penicillin/streptomycin sulfate. The cells were incubated in a humidified 5% CO₂ atmosphere at 37 °C.

Determination of NO and TNF-α production

The cells were plated at a density of 5 × 10⁵ cells/mL in 96-well plates for 24 h and then stimulated with LPS (1 μg/mL) in the presence or absence of different concentrations of compounds 1–5 for 24 h. The supernatants were collected for the NO and TNF-α assays. NO was determined by measuring the amount of nitrite with Griess reagent (a mixture of equal amounts of reagents A and B, A: 1% sulfanilamide in 5% H₃PO₄, B: 0.1% naphthylethylene diamine dihydrochloride, Sigma-Aldrich). NO was measured by the detection of its stable oxidative metabolite, nitrite, as described previously (Zhao et al. 2013). The TNF-α concentrations were measured using radioimmunoassay according to the manufacturer’s protocol (R&D Systems, Minneapolis, MN).

RNA extraction and real-time polymerase chain reaction (real-time PCR) analysis

Real-time quantitative polymerase chain reaction (RT-PCR) analysis of iNOS, COX-2 and GAPDH mRNA were performed as described previously (Xu et al. 2010). For PCR amplification, the following mouse-specific sense and antisense primers were used: iNOS, 5′-TGGAGTCACAGAAGGAGTGGCTAAC-3′ (forward) and 5′-TCTGACCACAGTGAGGAATGTCCAC-3′ (reverse); COX-2, 5′-GGAGAGACTAAGATAGTGATC-3′ (forward) and 5′-ATGGTCAGTAGACCTTTACAGCTC-3′ (reverse); GAPDH, 5′-GGGGAGCCAAAAGGGTCATC-3′ (forward) and 5′-GACGCCTGCTTCACCACCTTCTTG-3′ (reverse).

Western blotting analysis

Western blot analysis was performed as described previously (Wang et al. 2012). The concentrations of proteins from whole-cell lysates were determined by Bradford assay. Briefly, equal amounts of protein were separated by SDS-PAGE and transferred onto nitrocellulose membranes. The blots were then blocked overnight with 5% (m/v) nonfat dry milk, and probed with phospho-specific antibodies to IKBα, Ikkβ, p38 and akt antibodies in 5% (m/v) bovine serum albumin dissolved in TBST [20 mM Tris-HCl buffer, pH 7.6, containing 137 mM NaCl and 0.05% (v/v) Tween-20]. With the use of horseradish peroxidase-conjugated secondary anti-rabbit or anti-mouse antibody, bound antibodies were detected by enhanced chemiluminescence.

Results and discussion

Structures of isolated compounds

Compound 1, a faint yellow powder, was determined to have a molecular formula of C₁₇H₁₆O₆ by ESI-MS (m/z 317.18, [M + H]⁺; m/z 315.04, [M − H]⁻) and NMR data analysis. The IR absorption suggested the presence of hydroxyl (3467 cm⁻¹), carbonyl (1673.2 cm⁻¹) and phenyl groups (1602.2, 1504.6 cm⁻¹). The ¹H NMR spectrum (Table 1) of 1 exhibited signals characteristic for two methoxys at δ_H [3.91 (s, 3H), 3.97 (s, 3H)], one 1, 3, 4, 5, 6-penta-substituted aromatic ring at δ_H 6.44 (1H, s), and a 1, 2, 4, 5-tetra-substituted benzene ring at δ_H [7.34 (1H, s, H-5) and 6.94 (1H, s, H-8)]. In addition, the chemical shifts at δ_H 3.73 (2H, t, J = 7.2 Hz, H-10), δ_H 3.07 (2H, t, J = 7.2 Hz, H-11) suggested that compound 1 exhibited an ethoxy (–CH₂–CH₂–OH), which were confirmed by the HMBC spectrum (Figure 1) correlations from H-10 to C-11 and from H-11 to C-10.

Figure 1. Key HMBC (H→C) correlations of compound 1.

Table 1. ¹H (600 MHz) and ¹³C (150 MHz) NMR spectroscopic data of compound 1 in MeOD.

Position	δH	δc	HMBC (H-C)
1	–	137.5
2	6.44 (1H, s)	116.7	1,3, 4a, 8a, 10
3	–	157.2
4	–	141.7
4a	–	128.5
4b	–	135.1
5	7.34 (1H, s)	106.6	4b, 8b, 6, 7
6	–	152.2
7	–	146.6
8	6.94 (1H, s)	110.0	4b, 8b, 6, 7, 9
8a	–	123.3
8b	–	136.8
9	–	193.1
10	3.73 (2H, t, J = 7.2 Hz)	61.8	1, 2, 8a, 11
11	3.07 (2H, t, J = 7.2 Hz)	33.9	1, 10
12	3.91 (3H, s)	59.1	6
13	3.97 (3H, s)	55.2	4

In the ¹³C NMR and DEPT spectrum, the 12 sp²-hybridised carbons include three tertiary carbons at δ_C 116.7 (C-2), 106.6 (C-5), 110.0 (C-8), and four oxygenated quaternary carbons at δ_C 157.2 (C-3), 141.7 (C-4), 152.2 (C-6), 146.6 (C-7), as well as five further quaternary carbons at δ_C 137.5 (C-1), 128.5 (C-4a), 135.1 (C-4b), 123.3 (C-8a), 136.8 (C-8b), and their chemical shifts were indicative of a 9-fluorenone skeleton when the carbonyl group (δ_C 193.1, C-9) was taken into consideration (Yang et al. 2004). This hypothesis was supported by the HMBC spectrum, in which the correlations from the aromatic proton H-8 (δ_H 6.94) to C-8b and C-9, from H-5 (δ_H 7.34) to C-4a and C-4b were observed. The two methoxys were linked to C-4 and C-6, respectively, while the ethoxy was connected to C-1, as evidence by the HMBC (Figure 1) correlations from H-12 (δ_H 3.91) to C-6 (δ_C 152.2), from H-13 (δ_H 3.97) to C-4 (δ_C 141.7), and from H-10/11 to C-1 (137.5). All proton and carbon signals were assigned based on HSQC, HMBC and ¹H-¹H COZY spectra. Thus, based on the above evidences, compound 1 was established as 1-ethoxy-3,7-dihydroxy-4,6-dimethoxy-9-fluorenone (Figure 1).

1-Ethoxy-3,7-dihydroxy-4,6-dimethoxy-9-fluorenone (1), C₁₇H₁₆O₆, faint yellow powder; IR (KBr) ν_max 3467.0, 2937.8, 1673.2, 1602.2, 1578.7, 1504.6, 1265.9, 1135.5, 793.8 cm⁻¹; ESI-MS m/z 317.18, [M + H]⁺; m/z 315.04, [M − H]⁻; ¹H NMR (MeOD, 600 MHz), ¹³C NMR (MeOD, 150 MHz), HMBC data (Table 1).

Identification of known compounds

The structures of the following constituents, pinoresinol (2), syringaresinol (3), 9,9′-O-di-(E)-feruloyl-meso-5,5′-dimethoxysecoisolariciresinol (4), lyoniresinol (5) were identified by comparison of their spectral data with the literature.

Pinoresinol (2), C₂₀H₂₂O₆, white powder, ESI-MS m/z: 372.4 [M]⁺. ¹H NMR (600 MHz, CDCl₃, δ/ppm): 6.8 0–6.89 (6H, m, H-2,2′, 5,5′, 6,6′), 4.73 (2H, d, J = 5.6 Hz, H-7, 7′), 3.90 (6H, s, 2 × OMe), 3.8 6–4.25 (4H, m, H-9, 9′), 3.09 (2H, m, H-8, 8′). ¹³C NMR (150 MHz, CDCl₃, δ/ppm): 148.1 (C-3, 3′), 146.6 (C-4, 4′), 134.3 (C-1, 1′), 120.3 (C-6, 6′), 115.6 (C-5, 5′), 110.0 (C-2, 2′), 87.2 (C-7, 7′), 73.0 (C-9, 9′), 57.3 (3,3′-OMe), 55.5 (C-8, 8′) (Meagher et al. 1999).

Syringaresinol (3), C₂₂H₂₆O₈, white powder, ¹H NMR (600 MHz, MeOD, δ/ppm,): 6.57 (4H, s, H-2, 6, 2′, 6′), 5.49 (2H, s, 4, 4′-OH), 4.72 (2H, d, J = 4.0 Hz, H-7, 7′), 4.28 (2H, dd, J = 6.8, 9.0 Hz, H-9a, 9′a), 3.47 (2H, s, H-9b, 9′b), 3.90 (12H, s, 3, 3′, 5, 5′-OMe), 3.08 (2H, m, H-8, 8′); ¹³C NMR (150 MHz, MeOD, δ/ppm): 148.5 (C-4, 4′), 135.6 (C-3, 3′, 5, 5′), 133.5 (C-1, 1′), 104.0 (C-2, 6, 2′,6′), 87.4 (C-7, 7′), 73.1 (C-9, 9′), 57.7 (3, 3′, 5, 5′-OMe), 55.7 (C-8, 8′) (Sharp et al. 2001).

9,9′-O-Di-(E)-feruloyl-meso-5,5′-dimethoxysecoisolariciresinol (4), C₄₂H₄₆O₁₄, white powder, ESI-MS m/z: 775.12 [M]⁺. ¹H NMR (600 MHz, DMSO_d6, δ/ppm): 7.53 (1H × 2, d, J = 15.9 Hz, H-7″, 7‴), 7.00 (1H × 2, dd, J = 1.8, 8.2 Hz, H-6″,6‴), 6.95 (1H × 2, d, J = 1.8 Hz, H-2″, 2‴), 6.84 (1H × 2, d, J = 8.2 Hz, H-5″,5‴), 6.23 (1H × 2, d, J = 15.9 Hz, H-8″,8‴), 6.21 (2H × 2, s, H-2, 6, 2′,6′), 4.40 (1H × 2, dd, J = 5.5, 11.3 Hz H-9b, 9b′), 4.14 (1H × 2, dd, J = 5.8, 11.3 Hz, H-9a, 9a′), 3.86 (3H × 4, s, H-3, 5, 3′, 5′), 3.72 (3H × 2, s, H-3″,3‴), 2.67 (1H × 2, dd, J = 7.1, 14.0 Hz, H-7b,7b′), 2.63 (1H × 2, dd, J = 7.9, 14.0 Hz, H-7a,7a′), 2.10 (1H × 2, m, H-8,8′); ¹³C NMR (150 MHz, DMSO, δ/ppm): 166.6 (C-9″, 9‴), 149.3 (C-3″, 3‴), 147.9 (C-4″, 4‴), 147.7 (C-3, 3′, 5, 5′), 145.0 (C-7″, 7‴), 133.6(C-4, 4′), 129.9 (C-1, 1′), 125.4 (C-1″, 1‴), 123.1 (C-6″, 6‴), 115.4 (C-8″, 8‴), 114.3 (C-5″, 5‴), 111.1 (C-2″, 2‴), 106.1 (C-2, 6, 2′,6′), 63.8 (C-9, 9′), 55.7 (3, 5, 3′, 5′-OMe), 55.6 (3″,3‴-OMe), 40.3 (C-8, 8′), 34.3 (C-7,7′) (Kuroda et al. 2011).

Lyoniresinol (5), C₂₂H₂₈O₈, yellow powder, ¹H NMR (600 MHz, MeOD, δ/ppm): 6.45 (1H, s, H-2), 6.34 (2H, s, H-2′, 6′), 5.36 (1H, s, 3-OH), 5.34 (1H, s, 4′-OH), 4.02 (1H, d, J = 7.8 Hz, H-7), 3.5 6–3.79 (4H, m, H-9, H-9′), 3.88 (3H, s, 3-OMe), 3.79 (6H, s, H-3′, 5′-OMe), 3.30 (3H, s, 5-OMe), 2.70 (1H, dd, J = 4.0, 16.0 Hz, H-7a), 2.58 (1H, t, J = 9.2 Hz, H-7b), 1.63 (1H, m, H-8), 1.96 (1H, dd, J = 4.4, 10.8 Hz, H-8); ¹³C NMR (150 MHz, MeOD, δ/ppm): 129.9 (C-1), 107.8 (C-2), 147.5 (C-3), 138.3 (C-4), 146.8 (C-5), 126.3 (C-6), 34.8 (C-7), 41.7 (C-8), 68.1 (C-9), 139.6 (C-1′), 106.7 (C-2′, C-6′), 148.1 (C-3′, C-5′), 134.5 (C-4′), 44.4 (C-7′), 50.8 (C-8′), 65.3 (C-9′), 57.79 (3′, 5′-OMe), 60.8 (5-OMe), 57.4 (3-OMe) (Vecchietti et al. 1979).

Anti-inflammatory activity

All the isolated compounds were evaluated for their inhibitory effects both on NO and TNF-α production on RAW 264.7 cell induced by LPS. The IC₅₀ value of each compound on NO and TNF-α production was listed in Table 2. The results showed that the new compound, 1-ethoxy-3,7-dihydroxy-4,6-dimethoxy-9-fluorenone (1) and compound 4 exhibited potent anti-inflammatory activity with IC₅₀ values of 56 and 32 μM on NO and of 28 and 15 μM on TNF-α, respectively. Compounds 2, 3 and 5 also showed modest anti-inflammatory activity with IC₅₀ values in the range of 93–221 μM. These results indicated that the synergistic anti-inflammatory effect of multiple components may be responsible for the anti-arthritic activity of L. cubeba.

Table 2. The IC₅₀ of TNF-α and NO production.

Compounds	IC50 (μM) NO	IC50 (μM) TNF-α
C1	56.1 ± 1.2	28.2 ± 0.9
C2	119.2 ± 5.3	93.5 ± 1.3
C3	117.6 ± 4.2	150.6 ± 3.8
C4	32.8 ± 2.3	15.0 ± 1.0
C5	215.0 ± 6.4	221.1 ± 6.3

Effect of C1 and C4 on iNOS and COX-2 mRNA expression in RAW 264.7 cells

During the inflammatory response, activated macrophages secrete large amounts of pro-inflammatory mediators, such as NO and prostaglandin E₂ (PGE₂), via the inducible isoforms of NO synthase (iNOS) and cyclooxygenase-2 (COX-2), respectively, as well as pro-inflammatory cytokines, such as TNF-α, IL-6 (Jung et al. 2009). Most NO are synthesised through the oxidative deamination of l-arginine by inducible NOS (iNOS), and high levels of NO cause inflammatory diseases, such as bowel disease, RA (Blantz & Munger 2002). Current results showed that compounds 1–5 exhibited inhibitory effect both on NO and TNF-α production on RAW 264.7 cell induced by LPS, and the mRNA expression of iNOS and COX-2 was also investigated by RT-PCR. As showed in Figure 2, compounds 1 and 4 strongly suppressed the mRNA expression levels of both iNOS and COX-2. These data suggested that pretranslational events were involved in L. cubeba inhibition of LPS-induced expression of iNOS and COX-2 (GAPDH as a control gene), and compounds 1 and 4 showed the most potent inhibitory effect among the isolated compounds.

Figure 2. RT-PCR analysis of the expression of iNOS and COX-2 mRNA. RAW 264.7 cells were pretreated with C1 and C4 (40, 20, 10 μM) for 2 h, and then incubated with LPS for 18 h. The cells were lysed, and total RNA was subjected to RT-PCR with the primers iNOS or COX-2 with GAPDH as internal control. These experiments were repeated three times with similar results. *p < 0.05, **p < 0.01 indicate statistically significant differences from the LPS-treated group.

Effect of C1 and C4 on IκBα, IKKβ, P38 and Akt protein phosphorylation in RAW 264.7 cells

Inflammatory processes play a fundamental role in the damage of articular tissues, many in vitro and in vivo studies have examined the contribution of components of the NF-κB signaling pathways to the pathogenesis of various rheumatic diseases, in particular, of osteoarthritis (OA) and RA. The NF-κB also seems to be responsible for inducing iNOS and COX-2 in inflammatory cells (Roman Blas & Jimenez 2006). Activation of NF-κB by LPS is induced by a cascade of events leading to the activation of inhibitor κB (IκB) kinases (IKKs), which in turn phosphorylates IκB and leads to the degradation of NF-κB and its translocation to the nucleus (Griscavage et al. 1996). The phosphorylation and degradation of IκB are crucial steps for NF-κB activation in activated macrophages. In addition, the phosphorylation of NF-κB also regulated by (ERK)1/2, P38 MAPK, and PI3K/Akt pathways (Chandrasekar et al. 2004). The effect of compounds 1 and 4 on protein expression and phosphorylation of IκBα, IKKβ, P38 and Akt were evaluated by western blot assay. As shown in Figure 3, the phosphorylated forms of IκBα and IKKβ, were barely detectable in the unstimulated RAW 264.7 cells. However, the IκBα and IKKβ phosphorylation level was increased significantly after the LPS (2 μg/mL) treatment for 30 min. A pretreatment with compound 1 significantly inhibited the LPS-mediated IκBα and IKKβ phosphorylation at 40, 20, 10 μM and 40, 20 μM, respectively (Figure 3A and B). However, compound 1 had no effect on the phosphorylation level of P38 and Akt (Figure 3C and D). As showed in Figure 3(E)–(H), compound 4 significantly inhibited both phosphorylation level of IκBα and IKKβ at 40, 20 μM, and no significant effect on the phosphorylation level of P38 and Akt.

Figure 3. Effect of C1 and C4 on LPS-induced phosphorylation and degradation of IκBα, IKKβ, P38 and Akt. RAW 264.7 cells were pretreated with C1 and C4 (40, 20, 10 μM) for 2 h, and then incubated with LPS for 30 min. The cellular lysates were prepared and analysed for content of IκBα, p-IκBα, IKKβ, p-IKKβ, P38, p-P38, Akt, p-Akt and GAPDH by western blot. These experiments were repeated three times with similar results. *p < 0.05, **p < 0.01 indicate statistically significant differences from the LPS-treated group.

Conclusion

In summary, the anti-inflammatory activity of L. cubeba was characterised by the suppression mRNA expression of iNOS, COX-2, and various cytokines, such as TNF-α and NO, through the NF-κB signal pathway. These results showed that compounds 1 and 4, two main active compounds of L. cubeba, inhibit the LPS-induced expression of iNOS and COX-2 at the mRNA levels as well as the expression of TNF-α, NO transcripts in macrophage RAW 264.7 cells. These suppressive effects are mediated by inhibiting the phosphorylation of IκBα and IKKβ which are crucial steps for NF-κB activation. These results indicate that the synergistic anti-inflammatory effect of multiple components may be contributed to the anti-arthritic activity of L. cubeba.

Declaration of interest

The authors report no declarations of interest. This work was supported by the National Natural Science Foundation of China (Grant Nos. 81573696, U1203104, 81403162, 81202865), Youth Scholars Breeding Project for Medical Science of PLA (Grant No. 13QNP098) and Shanghai Municipal Committee of Science and Technology (Grant Nos. 13401900102, 14401902900).