HO-1 pathways

Fitoterapia 115 (2016) 37–45

Contents lists available at ScienceDirect

Fitoterapia journal homepage: www.elsevier.com/locate/fitote

A new steroidal saponin, furotrilliumoside from Trillium tschonoskii inhibits lipopolysaccharide-induced inflammation in Raw264.7 cells by targeting PI3K/Akt, MARK and Nrf2/HO-1 pathways Ting Yan a,b, Xiangyong Yu a, Xianduo Sun c, Dali Meng a,⁎, Jing-ming Jia a,⁎ a b c

School of Traditional Chinese Materia Medica, Shenyang Pharmaceutical University, Shenyang 110016, China School of Pharmacy, Jinzhou Medical University, Jinzhou 121001, China School of Traditional Chinese Medicines, Guangdong Pharmaceutical University, Guangzhou 510006, China

a r t i c l e

i n f o

Article history: Received 29 July 2016 Received in revised form 23 September 2016 Accepted 26 September 2016 Available online 28 September 2016 Keywords: Trillium tschonoskii Furotrilliumoside Anti-inflammatory RAW264.7 PI3K/Akt Nrf2/HO-1

a b s t r a c t A new steroidal saponin, furotrilliumoside (FT) was isolated from the roots and rhizomes of Trillium tschonoskii Maxim. Its structure was elucidated on the basis of 1D- and 2D-NMR spectroscopic data as well as HR-ESI-MS analysis. FT showed superior activity of inhibiting NO production of RAW264.7 cells induced by lipopolysaccharide (LPS) in the preliminary biological screening. In order to develop novel therapeutic drug for acute and chronic inflammatory disorders, the anti-inflammatory activity and underlying mechanism of FT were investigated in LPS-induced RAW264.7 cells. The results showed that FT could reduce LPS-induced expression of inducible nitric oxide synthase (iNOS) and then resulted in the decrement of NO production. More meaningful, FT could downregulate the expression of cyclooxygenase-2 (COX-2) and decrease the expressions of pro-inflammatory cytokines, tumor necrosis factor-α (TNF-α), interleukin-6 (IL-6) and interleukin-1β (IL-1β), in both gene and protein levels. In mechanism study, FT blocked the LPS-induced upregulation of phosphorylated phosphoinositide-3-kinase and Akt (PI3K/Akt). Furthermore, FT inhibited the translocation of nuclear factor-kappa B (NF-κB) through the prevention of inhibitory factor kappa B alpha (IκBα) phosphorylation and degradation and also suppressed the mitogen-activated protein kinases (MAPK) signaling pathway in LPS-stimulated RAW264.7 macrophages. In addition, FT upregulated heme oxygenase-1 (HO-1) expression via nuclear translocation of nuclear factor E2-related factor 2 (Nrf2). Taken together, FT might act as a natural agent to treat some inflammatory diseases by targeting PI3K/Akt, MARK and Nrf2/HO-1 pathways. © 2016 Published by Elsevier B.V.

1. Introduction Trillium tschonoskii Maxim. (Liliaceae) is a herbaceous plant in midwestern China, known as ‘Yan Ling Cao’ locally [1]. Its dried roots and rhizomes were used as a folk medicine for the treatment of neurasthenia, headache, cancer, and ameliorating pains, and various inflammatory diseases, such as sore, ulcer, fever and rheumatic pain [2]. The antiinflammatory activity of the extract from T.tschonoskii had been reported [3,4], but its active principle had not been identified yet and the detailed molecular mechanisms remained largely unclear. Previous phytochemical and pharmacological investigations had shown that the main active components of T. tschonoskii were steroidal saponins [5–7]. Moreover, some steroidal saponins exhibited obvious anti-inflammatory Abbreviations: FT, furotrilliumoside; LPS, lipopolysaccharide; NO, nitric oxide; iNOS, inducible nitric oxide synthase; COX-2, cyclooxygenase-2; TNF-α, tumor necrosis factorα; IL-6, interleukin-6; IL-1β, interleukin-1β; PI3K, phosphoinositide-3-kinase; MARK, mitogen-activated protein kinases; HO-1, heme oxygenase-1; Nrf2, nuclear factor E2related factor 2; NF-κB, nuclear factor-κB; IκBα, inhibitory factor kappa B alpha. ⁎ Corresponding authors. E-mail addresses: [email protected] (D. Meng), [email protected] (J. Jia).

http://dx.doi.org/10.1016/j.fitote.2016.09.012 0367-326X/© 2016 Published by Elsevier B.V.

activity. For example, dioscin could inhibit the inflammatory reaction of collagen arthritis rats [8], and trillenoside A displayed remarkable inhibitory action towards COX-2 production in macrophagocytes of the mouse abdominal cavity stimulated by LPS at 10 mg/mL [9], which provided scientific foundation to our research. In the anti-inflammatory compound screening, it was found furotrilliumoside (FT, Fig. 1), a new compound isolated from T.tschonoskii, showed the strongest activity. In order to reveal the antiinflammatory potential of FT, the inhibitory activity on LPS-induced inflammatory mediator production in RAW264.7 cells and underlying mechanism were investigated in this study. 2. Experimental 2.1. General experiment procedure Optical rotations were determined using a MCP-200 (Anton Paar, Germany). UV spectra were recorded on a Shimadzu-2201 (Kyoto, Japan). The IR spectrum was obtained from a Bruker IFS-55 spectrophotometer (Karlsruhe, Germany) using KBr pellet. HR-ESI-MS data were

38

T. Yan et al. / Fitoterapia 115 (2016) 37–45

HO HO 4 HO

26

O

O

G II

OH

1

21

20

18

19 1

O HO

4

R III O

HO

1

OH

4

R II

HO

O

O 17

H

22

16

14

HO

10

O

4

O

11 9

27 25

3

O

GI

HO

H

8

H

5

1

1

O

OH HO

4

RI

HO

O

1

OH

Fig. 1. The structure of FT.

performed on a Xevo G2 QTof (Waters MS Technologies, Manchester, UK). NMR spectra were run on a Bruker AV-400 spectrometer (Karlsruhe, Germany). Semi- preparative HPLC was performed on a YMC-Pack ODS-A column(10 × 250 mm, 5 μm; YMC-Pack, Japan), equipped with a LC-3000 A pump (Beijing Chuang Xin Tong Heng Science and Technology Co., Ltd., China) and a 3000 UV–vis detector (Beijing Chuang Xin Tong Heng Science and Technology Co., Ltd., China). Sugars analytical HPLC was carried out on a Jasco PU-4180 pump (Kyoto, Japan) and a OR-4090 detector (Kyoto, Japan). HPLC was performed with an Asahipak NH2P-50 4E column (4.6 mm × 250 mm, 5 μm, Shodex, Japan).

80:20 and 100:0, v/v) as eluent to give ten subfractions J1–J10. Subfraction J8 was followed by semi-preparative RP-HPLC with the mobile phase of MeOH-H2O (69:21, v/v, 3.0 mL/min) to afford FT (27.4 mg, tR 33.4 min). Furotrilliumoside (FT): white amorphous power; [α]20D – 73.8 (c 0.60, C5H5N); UV (MeOH) λmax 225 nm; IR (KBr) νmax 3395, 2924, 2852, 2127, 1646, 1570, 1456, 1415, 1384, 1262, 1130, 1097, 1042, 983, 912, 884, 804 and 634 cm−1; HRESIMS m/z 1175.5809[M + H]+ (calcd. for C57H91O25, 1175.5849). 1H NMR and 13C NMR (C5H5N) data see Table 2. 2.4. Acid hydrolysis of FT and determination of absolute configuration of monosaccharides

2.2. Plant material and reagents The dried roots and rhizomes were collected in July 2014 from Shennongjia, Hubei Province, China, and were identified by associate professor Jia Lingyun (School of Traditional Chinese Materia Medica, Shenyang Pharmaceutical University, Shenyang, China). The voucher sample (JTZ-20140712) had been deposited in the Department of Natural Products Chemistry, Shenyang Pharmaceutical University, Shenyang, China. LPS (Escherichia coli serotype 055: B5) and 3-(4, 5-dimethylthiazol2-yl)-2, 5-diphenyltetrazolium bromide (MTT) were purchased from Sigma-Aldrich (St. Louis, MO, USA). Dulbecco's modified Eagle's medium (DMEM), fetal bovine serum (FBS), penicillin and streptomycin were bought from Gibco BRL (Gaithersburg, MD, USA). All antibodies, except for COX-2, JNK and lamin B (Proteintech Group, Inc., Chicago, USA) and PI3K, iNOS (Abcam, Cambridge, MA, USA) were obtained from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA, USA). Primers for Real-time PCR were synthesized by Sangon Biotech (Shanghai) Co., Ltd. (Shanghai, China). All other chemicals were obtained from SigmaAldrich unless otherwise stated. 2.3. Extraction an isolation The dried roots and rhizomes (1.7 kg) of T. tschonoskii were powered and extracted three times with 70% EtOH (each 2 h) and the combined solution evaporated to dryness by a vacuum rotary evaporator to afford a syrup (600 g). The crude extract was suspended with H2O (1.0 L) and successively partitioned with petroleum ether, ethyl acetate and n-butanol to yield three layers of extracts. The n-butanol extract (210 g) was fractionated by silica gel column chromatography eluting with CH2Cl2-MeOH (100:0–0:100, v/v) to obtain thirteen fractions (Fr. A– M) based on TLC analyses. Fr. J was subjected to ODS CC eluted with MeOH-H2O (10:90, 20:80, 30:70, 35:65, 40:60, 50:50, 60:30, 70:30,

FT (2.0 mg) was hydrolysed with 2 M HCl (2 mL) at 90 °C for 3 h. The reaction mixture was extracted with CHCl3 (2 mL × 3). The aqueous layer was neutralized with 1 M NaOH and then concentrated to give a residue [10,11]. The residue was analyzed by HPLC under the following conditions: column, Asahipak NH2P-50 4E column (4.6 mm × 250 mm, 5 μm, Shodex); flow rate, 1.0 mL/min; solvent, 75% MeCN-H2O; detection, OR (Jasco OR-4090) detector. Identification of D-glucose and Lrhamnose was carried out by a comparison of the retention times and polarities with those of authentic samples. D-glucose (tR 9.0 min, positive polarity), L-rhamnose (tR 5.7 min, negative polarity). 2.5. Cell culture and cell viability for assay RAW264.7, a mouse macrophage cell line, was purchased from the Cell Bank of the Shanghai Institute of Cell Biology and Biochemistry, Chinese Academy of Sciences (Shanghai, China) and cultured in high glucose DMEM supplemented with 10% heat-inactivated FBS, 100 U/mL penicillin and 100 μg/mL streptomycin in a 37 °C humidified incubator containing 5% CO2. FT was dissolved in dimethyl sulfoxide (DMSO) to a concentration of 100 mM and further diluted in cell culture media so that the final DMSO concentration was below 0.1% v/v. MTT assay was used to evaluate the effect of FT on cell viability [12]. In Brief, RAW264.7 cells were seeded in 96-well plates (Corning Inc., Corning, NY, USA) at a density of 5 × 104 cells/well. After overnight growth, cells were treated with various concentrations of FT (1– 160 μM) for 1 h, followed in the presence or absence of LPS (100 ng/ mL) for the next 24 h. 20 μL of MTT solution (5 mg/mL) was added and the cells were further cultured for 4 h. After that, the supernatant was carefully removed and then the resulting formazan crystals were dissolved in 100 μL DMSO with horizontal shaking. The absorbance at 570 nm (ref 630 nm) was measured with a microplate reader (Molecular Devices, California, USA).

T. Yan et al. / Fitoterapia 115 (2016) 37–45

2.6. Measurement of NO, TNF-α and IL-6 release RAW264.7 cells were seeded into 96-well plates at a density of 2.5 × 105 cells/mL and cultured overnight. After pretreatment with FT of various concentrations for 1 h, the cells were stimulated with LPS (100 ng/ mL) for 24 h. The concentration of NO in the conditioned culture medium was examined with the Nitric Oxide Assay Kit (Beyotime institute of Biotechnology, Jiangsu, China) and the release of TNF-α and IL-6 in the cell supernatants were assayed using ELISA kits (Shanghai Joyee Biotechnics Co., Ltd., Shanghai, China) according to the manufacturer's instructions [13]. The concentrations were calculated from the standard curves. Nitrite accumulation, an indicator of NO synthesis, was measured in the culture medium by Griess reaction (Beyotime, China). Briefly, 50 μL of cell culture medium was mixed with 50 μL of Griess reagent [equal volumes of 1% (w/v) sulfanilamide in 5% (v/v) phosphoric acid and 0.1% (w/v) naphtylethylenediamine-HCl] and incubated at room temperature for 10 min [14]. The absorbance at 540 nm was then measured using an automated microplate spectrophotometer (Bio-Tek Instruments, Winooski, VT). Fresh culture medium was used as a blank in all experiments. The amount of nitrite in the test samples was calculated from a sodium nitrite standard curve. 2.7. Quantitative real-time polymerase chain reaction (qRT-PCR) RAW264.7 cells (2.5 × 105 cells in a 24-well plate) were treated with FT for 1 h and then were stimulated with LPS for 4 h. Total RNA was extracted with TRIzol Reagent and 1 μg total RNA was reverse-transcribed using a HiScript II Q RT SuperMix for qPCR according to the manufacturer's instructions [15]. Aliquots of diluted cDNA (1:5) were amplified with TransStart Top Green qPCR SuperMix in a final volume of 20 μL. QRT-PCR was carried out using the Thermo Scientific PikoReal (Thermo Fisher Scientific, MA, USA). PCR cycles consisted of initial denaturation at 94 °C for 30 s, followed by 40 cycles of 94 °C for 5 s, 60 °C for 15 s, and 72 °C for 10 s. The Δ cycle threshold method was used for the calculation of relative differences in mRNA abundance with the Thermo Scientific PikoReal. Data were normalized to the expression of β-actin. The results were expressed as fold-changes. The normalized value of the target mRNA of the LPS control group was arbitrarily presented as 1. The sequences of primers used were listed in Table 1. 2.8. Protein samples preparation RAW264.7 cells (2.5 × 105 cells/mL) were pretreated or left untreated with FT (10, 20 and 40 μM) for 1 h and stimulated with LPS for indicated time. Then cells were rinsed, scraped off and collected in ice-PBS. After centrifugation, the PBS was pipetted completely and Cell lysis buffer for Western and IP (Beyotime institute of Biotechnology, Jiangsu, China) was added in for total proteins extraction. While nuclear and cytosolic proteins were prepared using Nuclear and Cytoplasmic Extraction Reagents (Active Motif, Carlsbad, CA, USA) according to the manufacturer's protocol. Protein concentration was determined using Table 1 Primers used for the qRT-PCR study. Gene

Sequence (5′ to 3′)

iNOS

F: AGCCAAGCCCTCACCTACTT R: GCCTCCAATCTCTGCCTATC F: CCAGCACTTCACCCATCAGT R: GGGATACACCTCTCCACCAA F: CAGACCCTCACACTCAGATCATCTT R: CAGAGCAATGACTCCAAAGTAGACCT F: CACGGCCTTCCCTACTTCAC R: TGCAAGTGCATCATCGTTGT F: GTTGACGGACCCCAAAAGAT R: CCTCATCCTGGAAGGTCCAC F: ATGTGGATCAGCAAGCAGGA R: AAGGGTGTAAAACGCAGCTCA

COX-2 TNF-α IL-6 IL-1β β-actin

39

the Enhanced BCA Protein Assay Kit (Beyotime institute of Biotechnology, Jiangsu, China). Aliquots of protein samples were mixed in loading buffer, boiled for 5 min and then stored at ‐80 °C [16]. 2.9. Western blot analysis Aliquots of the protein samples were separated on acrylamide gel by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and electrophoretically transferred into a polyvinylidene fluoride (PVDF) membranes (Millipore, Billerica, MA, USA). After blocking with 5% skimmed milk the membranes were incubated with specific primary antibodies overnight at 4 °C. After rinsing, the membranes were incubated with a HRP-labelled secondary antibody containing a blocking solution for 1 h at room temperature. Immunodetection was performed using an enhanced chemiluminescence EasySee Western Blot Kit (TransBionovo Co., Ltd., Beijing, China). The immunosignals were captured using the Gel DOC™ XR+ system (BioRad Laboratories, Hercules, CA, USA) and densitometric data were studied following normalization to the house-keeping loading control. 2.10. Statistical analysis All experiments were performed at least three times unless otherwise stated. Statistical significance was determined using GraphPad Prism 5 Software (GraphPad Software, San Diego, CA, USA). The results were analyzed using one-way ANOVA with Tukey multiple comparison test. The data were given as the means ± SD. P b 0.05 was considered as significant. 3. Results and discussion 3.1. Structure elucidation FT (Fig. 1) was obtained as amorphous powder with molecular formula of C57H90O25 determined by its HRESI-MS at m/z 1175.5809 [M + H]+ (calc. for 1175.5849) and 13C NMR data (Table 2). The 1H Table 2 1 H and 13C NMR spectroscopic data for FT in pyridine-d5. No.

δH

δC

No.

δH

δC

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

0.96 (m), 1.71 (m) 1.85 (m), 2.06 (m) 3.87 (m) 2.72 (m), 2.81 (m) ― 5.32 (br d, 3.2) 1.54 (m), 1.84 (m) 1.65 (m) 1.03 (m) ― 1.46–1.59 (m) 1.56 (m), 2.01 (m) ― 1.89 (m) 2.28 (dd, 13.9, 11.2) 2.47 (dd, 13.9, 6.3) ― ― 0.89 (s) 1.08 (s) ― 1.97 (s) ― 2.59–2.74 (m) 1.55 (m), 1.93 (m) 1.94 (m) 3.57 (dd, 9.4, 5.5) 3.94 (m) 0.98 (d, 6.3) 4.94 (d, 7.8) 4.20 (m)

37.4 (t) 30.2 (t) 78.1 (d) 39.1 (t) 141.2 (s) 121.7 (d) 32.0 (t) 30.5 (d) 51.1 (d) 37.4 (s) 20.7 (t) 35.7 (t) 41.3 (s) 60.9 (d) 27.0 (t)

G I-3 G I-4 G I-5 G I-6 R I-1 R I-2 R I-3 R I-4 R I-5 R I-6 R II-1 R II-2 R II-3 R II-4 R II-5

3.60 (m) 4.41 (m) 4.22 (m) 4.04 (m), 4.19 (m) 6.39 (br s) 4.85 (br s) 4.64 (dd, 9.4, 3.0) 4.36 (m) 4.95 (m) 1.76 (d, 6.1) 5.83 (br s) 4.55 (m) 4.50 (dd, 9.4, 2.9) 4.43 (t, 9.4) 4.94 (m)

77.1 (d) 77.8 (d) 77.9 (d) 61.3 (t) 102.2 (d) 72.6 (d) 72.9 (d) 74.2 (d) 69.6 (d) 18.7 (q) 102.3 (d) 73.3 (d) 72.9 (d) 80.5 (d) 68.4 (d)

155.3 (s) 137.2 (s) 18.3 (q) 19.4 (q) 112.0 (s) 9.0 (q) 153.8 (s) 24.4 (t) 33.2 (t) 33.6 (d) 75.0 (t)

R II-6 R III-1 R III-2 R III-3 R III-4 R III-5 R III-6 G II-1 G II-2 G II-3 G II-4

1.58 (d, 5.8) 6.28 (br s) 4.90 (br s) 4.54 (m) 4.29 (t, 9.4) 4.34 (m) 1.59 (d, 5.8) 4.81 (d, 7.8) 4.02 (m) 3.95 (m) 4.24 (m)

18.5 (q) 103.3 (d) 72.7 (d) 72.9 (d) 74.0 (d) 70.5 (d) 18.9 (q) 104.9 (d) 75.3 (d) 78.6 (d) 71.8 (d)

17.3 (q) 100.4 (d) 78.1 (d)

G II-5 G II-6

4.25 (m) 4.40 (m), 4.56 (m)

78.7 (d) 62.9 (t)

16 17 18 19 20 21 22 23 24 25 26 27 G I-1 G I-2

40

T. Yan et al. / Fitoterapia 115 (2016) 37–45

Fig. 2. Significant HMBC correlations of FT.

NMR spectrum of FT showed signals for three tertiary methyls at δH 0.89, 1.08 and 1.97 (each 3H, s); four secondary methyls at δH 0.98 (3H, d, J = 6.3 Hz), 1.58 (3H, d, J = 5.8 Hz), 1.59 (3H, d, J = 5.8 Hz), 1.76 (3H, d, J = 6.1 Hz); five anomeric protons at δH 4.81 (1H, d, J = 7.8 Hz), 4.94 (1H, d, J = 7.8 Hz), 5.83 (1H, br s), 6.28 (1H, br s) and 6.39 (1H, br s); and an olefinic proton at δH 5.32 (1H, br d, J = 3.2 Hz). The characteristic signals at δH 1.58, 1.59 and 1.76 were due to the methyl group of rhamnose. The 13C NMR spectrum of FT showed a total of 57 resonance lines, and 30 of which were attributed to five monosaccharide units. This implied a molecular formula C27H40O3 for the aglycone moiety, which suggested the aglycone to be a steroid with three oxygen atoms on the skeleton. The 1H NMR signals (Table 2) at δH 3.87 (1H, m), δH 3.57 (1H, dd, J = 9.4, 5.5 Hz) and 3.94 (1H, m) and six olefinic carbon signals (Table 2) at δC 112.0 (s), 121.7 (d), 137.2 (s), 141.2 (s), 153.8 (s) and 155.3 (s) in its 13C NMR spectrum suggested the aglycone of FT to be a Δ5,16,20(22)-furostatriene-3,26-diol. The structure of the aglycone was further confirmed by HMBC correlations (Fig. 2) from δH 1.08 (3H, s, Me-19) to C-1 (δC 37.4), C-5 (δC 141.2) and C-9 (δC 51.1), from δH 5.32 (1H, br d, J = 3.2 Hz, H-6) to C-4 (δC 39.1), C-8 (δC 30.5) and C-10 (δC 37.4), from δH 0.89 (3H, s, Me-18) to C-12 (δC 35.7), C-14 (δC 60.9) and C-17 (δC 137.2), from δH 1.97 (3H, s, Me-21) to C-17 (δC 137.2), C20 (δC 112.0) and C-22 (δC 153.8), and from δH 0.98 (3H, d, J = 6.3 Hz, Me-27) to C-24 (δC 33.2), C-25 (δC 33.6), and C-26 (δC 75.0). The β-orientation of OH-3 was deduced from the typical peak shape of H-3 and 13 C NMR data of ring A. The H-atom resonances of CH2 (26) showed a Δ (δH26b − δH26a) = 0.37 ppm which is b 0.48, allowed us to assign the 25R configuration [17]. Thus, the aglycone of FT was elucidated as (25R)-Δ5,16,20(22)-furostatriene-3β,26-diol.

Fig. 4. Effect of FT on TNF-α (A) and IL-6 (B) production in LPS-induced RAW264.7 macrophages. RAW264.7 cells were treated with FT (10, 20 and 40 μM) and then incubated with LPS (100 ng/mL) for 24 h. The culture supernatant was subjected to ELISA kits. Data shown are the means ± SD from three independent experiments. ***p b 0.001 vs control, #p b 0.05 and ##p b 0.01 vs LPS-treated group.

Acid hydrolysis of FT with 2 M HCl obtained D-glucose and L-rhamnose as carbohydrate moieties, which were characterized by comparison of the retention times in HPLC analysis with their corresponding authentic samples. The 1H and 13C NMR spectra (Table 2) of FT showed five anomeric proton signals, three distinct methyl doublets, and five anomeric carbons at δC 100.4, 102.2, 102.3, 103.3, and 104.9, revealing the presence of two glucose and three rhamnose units. The glucopyranoside was βconfiguration on the ground of the coupling constants (3J1,2 N 7.0 Hz) of the anomeric protons. The α-configuration for the rhamnopyranoside was deduced by comparing the 13C NMR spectroscopic data for C-3 (δCRI 72.9, δC-RII 72.9 and δC-RIII 72.9) and C-5 (δC-RI 69.6, δC-RII 68.4 and δC-RIII 70.5) of rhamnopyranoside with those reported in the literature [18]. Carefully analysis of the 1D and 2D NMR spectra of FT suggested that the sequence and linkage positions of the sugars were exactly the

Fig. 3. Cell viability and effect of FT on NO production in LPS-induced RAW264.7 macrophages. (A) Cell viability. RAW264.7 cells were treated with LPS (100 ng/mL) without or with FT (1– 160 μM) for 24 h. Cell viability was determined by MTT assay. (B) Level of nitrite. RAW264.7 cells were pretreated with FT (10, 20 and 40 μM) for 1 h prior to LPS (100 ng/mL) treatment for 24 h. The concentrations of nitrite in the medium of RAW264.7 macrophages were monitored by Griess reaction. Data shown are the means ± SD from three independent experiments. ***p b 0.001 vs control, ###p b 0.001 vs LPS-treated group.

T. Yan et al. / Fitoterapia 115 (2016) 37–45

same as those of parispseudoside C [17], which were verified by detailed analysis of long-rang HMBC correlations (Fig. 2). Furthermore, in the HMBC spectrum, the correlations of δH 4.94 (1H, d, J = 7.8 Hz) with δC 78.1 (C-3) and δH 4.81 (1H, d, J = 7.8 Hz) with δ C 75.0 (C-26) were observed, which hinted that the tetrasaccharide and the βglucopyranoside were attached to C-3 and C-26 of the aglycone, respectively. Therefore, the structure of FT was elucidated as (25R)-3βO-{α-L-rhamnopyranosyl-(1 → 4)-α-L-rhamnopyranosyl-(1 → 4)-[α-Lrhamnopyranosyl-(1 → 2)]-β-D-glucopyranosyl}furosta-5,16,20(22)triene-26-O-β-D-glucopyranoside and named as furotrilliumoside.

3.2. Effects of FT on cell viability and LPS-induced nitric oxide (NO) production in RAW264.7 macrophages To evaluate the anti-inflammatory effect of FT, the effect of FT on NO production was determined by measuring the level of nitrite accumulation (the stable metabolite of NO) in culture media. LPS (100 ng/mL) induced significant nitrite production as compared with the naive control. Meanwhile, the induction of nitrite production by LPS was inhibited by a FT treatment in a dose-dependent manner with an IC50 value of 17.78 μM (Fig. 3B). An examination of the cytotoxicity of FT in RAW 264.7 macrophages by the MTT assay indicated that even 160 μM FT did not affect the viability of the RAW 264.7 cells (Fig. 3A).

41

Therefore, the inhibition of LPS-induced nitrite production by FT was not the result of a cytotoxic effect on these cells. 3.3. Effects of FT on levels of TNF-α and IL-6 in RAW264.7 macrophages To investigate whether FT could inhibit the inflammatory response, ELISA assay was used to test the levels of TNF-α and IL-6, which are considered as classical inflammatory factors, in LPS-stimulated RAW264.7 cells. RAW264.7 cells were treated with various concentrations of FT and stimulated with LPS for 24 h, the RAW264.7 cell secretion of two pro-inflammatory cytokines was strongly increased compared with the untreated control cells. As expected, FT (10–40 μM) significantly reduced the production of TNF-α and IL-6 in dose-dependent manner (Fig. 4). The inhibition following treatment with 40 μM FT was 60.90% for TNF-α and 30.91% for IL-6, respectively. However, none of these cytokine levels returned to the basal level of the untreated group. 3.4. Effects of FT on LPS-induced iNOS, COX-2 protein and mRNA expressions in RAW264.7 macrophages In view of the involvement of iNOS in the inflammatory process, the levels of iNOS protein and mRNA gene expression in the macrophages exposed to FT were monitored. As shown in Fig. 5A, the expression of

Fig. 5. Effects of FT on iNOS, COX-2 protein and gene expressions in LPS-stimulated RAW264.7 cells. RAW264.7 macrophages were pretreated with various concentrations of FT for 1 h prior to LPS (100 ng/mL) treatment. (A and B) iNOS and COX-2 protein expressions. iNOS (A) and COX-2 (B) protein expressions were monitored 24 h after treatment of cells with LPS (100 ng/mL) using β-actin expression as an internal control. (C and D) iNOS and COX-2 mRNA expressions. The mRNA expression levels of iNOS (C) and COX-2 (D) were quantitated by qRT-PCR. in RAW264.7 macrophages exposed to LPS for 4 h. Data shown are the means ± SD from three independent experiments. ***p b 0.001 vs control, ## p b 0.01 and ###p b 0.001 vs LPS-treated group.

42

T. Yan et al. / Fitoterapia 115 (2016) 37–45

Fig. 6. Effect of FT on TNF-α, IL-6 and IL-1β mRNA expression in LPS-induced RAW264.7 macrophages. RAW264.7 cells were treated with various concentrations (10, 20 and 40 μM) of FT and stimulated with LPS (100 ng/mL) for 4 h, the mRNA expression levels of TNF-α (A), IL-6 (B) and IL-1β (C) were quantitated by qRT-PCR. Data shown are the means ± SD from three independent experiments. ***p b 0.001 vs control, #p b 0.05, ##p b 0.01 and ###p b 0.001 vs LPS-treated group.

the iNOS protein was barely detected in the non-stimulated cells. However, the level increased markedly 24 h after the LPS treatment. FT exerted a concentration-dependent inhibition of iNOS protein expression in the LPS-stimulated RAW264.7 macrophages. To assess the effect

of FT on iNOS mRNA expression, the mRNA level was quantitated by qRT-PCR. iNOS mRNA expression was also hardly detectable in nonstimulated cells. On the other hand, the RAW264.7 macrophages expressed high levels of iNOS mRNA when stimulated with LPS for a

Fig. 7. Effects of FT on LPS-induced NF-κB activation and IκBα degradation and phosphorylation in RAW264.7 cells. The cells were treated with FT (10, 20 and 40 μM) in the presence of 100 ng/mL LPS for 1 h. (A) Effect of FT on LPS-induced NF-κB activation. Protein samples for nuclear (N) and cytosol (C) extract of RAW264.7 cells were analyzed by Western blot using anti-p65 antibody. Lamin B and β-actin were used as the internal control for normalization. (B) Effects of FT on LPS-induced IκBα degradation and phosphorylation. The effect of FT on IκBα degradation and phosphorylation was immunochemically assessed using phospho-specific anti-IκBα for cytosol extract of RAW264.7 cells. β-Actin was used as the internal control for normalization. The data represent the means ± SD from three independent experiments. ***p b 0.001 vs control, #p b 0.05, ##p b 0.01 and ###p b 0.001 vs LPS-treated group.

T. Yan et al. / Fitoterapia 115 (2016) 37–45

4 h period. Consistent with protein expression, FT inhibited the LPSstimulated expression of iNOS mRNA in a dose-dependent manner (Fig. 5C). The results suggested that FT significantly reduced the iNOS expression at concentration-dependent manner, which was in line with NO production. COX-2 was induced by a wide variety of stimuli such as LPS, certain serum factors, cytokines, and growth factors and was expressed predominantly at the sites of inflammation [19]. Therefore, there is increasing interest in the use of COX-2 inhibitors for treating inflammatory diseases. The effects of FT on the LPS-induced expression of the COX-2 mRNA gene and protein in macrophages were then evaluated in a further step. The expression of COX-2 mRNA was monitored in RAW264.7 macrophages exposed to LPS for 4 h. FT could effectively suppress the induction of COX-2 mRNA by LPS in a dose dependent manner (Fig. 5D). Consistent with mRNA expression, FT inhibited the LPS-stimulated expression of COX-2 protein in a dose-dependent manner (Fig. 5B). 3.5. Effect of FT on TNF-α, IL-6 and IL-1β mRNA expression in RAW264.7 macrophages FT efficiently inhibited LPS-induced production of proinflammatory cytokines in RAW264.7 cells. Furthermore, the effects of FT on TNF-α, IL-6 and IL-1β mRNA level were investigated. According to

43

Fig. 6, FT (10, 20 and 40 μM) had strong concentration-dependent reducing effects on the relative expressions of TNF-α, IL-6 and IL-1β mRNA in LPS-induced RAW264.7 cells. The high dosage of FT (40 μM) showed the best inhibitory effect, compared with those of other groups. The percent inhibition of 40 μM FT was 44.90% for TNF-α, 50.38% for IL-6 and 45.86% for IL-1β, respectively (Fig. 6A, B and C). Although there were high levels of inhibition of cytokine production, none of them returned to their basal level, compared to the control group. 3.6. Effect of FT on NF-κB activation and IκBα degradation in LPS-induced RAW264.7 macrophages The NF-κB pathway played a critical role in the inflammatory response by regulating the expression of inflammatory cytokines, including IL-1β, IL-6, and TNF-α [20,21]. Hence, it was necessary to investigate whether FT could suppress the phosphorylation of IκB-α and translocation of NF-κB (active subunit p65) into the nucleus (Fig. 7). It was observed that p65 was mostly distributed in the cytoplasm and hardly translocated into the nucleus in unstimulated cells; while treatment with LPS increased p65 protein level in the nucleus. Interestingly, FT suppressed the nuclear translocation of p65 in a concentration-dependent manner compared with LPS-induced cells. Furthermore, FT also significantly suppressed IκB-α phosphorylation and degradation,

Fig. 8. Effect of FT on LPS-induced phosphorylation of PI3K/Akt (A and B) and MAPKs (C) in RAW264.7 cells. RAW264.7 cells were stimulated with LPS (100 ng/mL) alone or LPS plus FT (40 μM) for the indicated time points. The whole-cell lysates were analyzed by immunoblot analysis using various antibodies against specific antibodies. The data represent the means ± SD from three independent experiments. **p b 0.01 and ***p b 0.001 vs control, ###p b 0.001 vs LPS-treated group.

44

T. Yan et al. / Fitoterapia 115 (2016) 37–45

which can regulate NF-κB activity, compared with LPS-induced RAW264.7 cells. 3.7. Effect of FT on phosphorylation of PI3K/Akt and MAPKs pathways in LPS-induced RAW264.7 cells LPS-stimulated activation of PI3K/Akt signaling played an important part in the progress of inflammation [22]. As shown in Fig. 8B, LPS powerfully enhanced phosphorylation of Akt protein in RAW264.7 cells, and Akt phosphorylation was distinctly inhibited by FT. Similarly, the upregulation of PI3K was also reduced by FT (Fig. 8A). This result indicated that FT had an important repressor effect on the LPS-stimulated PI3K/ Akt pathway signaling. The MAPKs played an important role in regulating cell growth and differentiation as well as in the control of the cellular responses to cytokines and various stresses [23]. Moreover, they were also important for activating NF-κB. The effects of FT on the LPS-stimulated phosphorylation of Erk1/2, SAPK/JNK, and p38 MAPK in RAW264.7 cells were examined using Western blot analyses to determine whether the FT-induced inhibition of NF-κB activation was mediated by the MAPKs pathway. As shown in Fig. 8C, LPS treatment induced strong increases in the levels of phosphorylated ERK, JNK, and p38 at different indicated times. Meanwhile, FT significantly suppressed the LPS-induced phosphorylation of p38 MAPK and Erk1/2 MAPKS, and SAPK/JNK was also slightly affected by the FT treatment.

and HO-1 protein in macrophages at a dose-dependent manner (Fig. 9). In the current study, parispseudoside C, which was not presented in this paper due to its worse effect towords RAW264.7 macrophages cells (IC50 N 100 μM), also had the same skeleton as FT, excepted the existence of furan ring moiety. Therefore it was initially speculated that furan ring fragments might play an important role for anti-inflammatory activity. In conclusion, FT significantly attenuated LPS-induced production of NO, TNF-α and IL-6, and reduced their relevant gene expression in protein and mRNA levels in RAW264.7 murine macrophages. Mechanisms study further elucidated that FT might exert these effect via inhibition of PI3K/Akt and MARK signaling, and activation of Nrf2/HO-1 signaling. FT exhibited great potential to be developed into therapeutic drug for acute and chronic inflammatory disorders. Nevertheless, the therapeutic effect of FT in vivo need further studies. Conflict of interest The authors declare that there are no conflicting interest. Acknowledgements The work was supported by National Natural Science Foundation of China (NO. 81473423).

3.8. Effect of FT on Nrf2 and HO-1 protein expressions in RAW264.7 cells

Appendix A. Supplementary data

Nuclear factor erythroid 2-related factor 2 (Nrf2) was a transcription factor that binds to the promoter of the hemeoxygenase (HO)-1 gene leading to anti-inflammation and anti-oxidation responses [24,25]. FT significantly promoted the expression of Nrf2

The NMR, IR and HR-ESI-MS spectra of FT were available as supporting information in the Supplementary data. Supplementary data associated with this article can be found in the online version, at http://dx.doi.org/10.1016/j.fitote.2016.09.016.

Fig. 9. Effects of FT on LPS-induced Nrf2 and HO-1 protein expressions in RAW264.7 cells. (A) Effects of FT on Nrf2 protein expression in RAW264.7cells. The cells were treated with FT (10, 20 and 40 μM) in the presence of 100 ng/mL LPS for 1 h. Protein samples for nuclear (N) and cytosol (C) extract of RAW264.7 cells were analyzed by Western blot using anti-Nrf2 antibody. Lamin B and β-actin were used as the internal control for normalization. (B) Effects of FT on HO-1 protein expression in RAW264.7cells. The cells were treated with FT (10, 20 and 40 μM) in the presence of 100 ng/mL LPS for 24 h. HO-1 protein expression was immunochemically assessed using anti-HO-1 antibody. β-Actin was used as the internal control for normalization. The data represent the means ± SD from three independent experiments. **p b 0.01 and ***p b 0.001 vs control, ##p b 0.01 and ###p b 0.001 vs LPS-treated group.

T. Yan et al. / Fitoterapia 115 (2016) 37–45

References [1] Q. Li, M. Xiao, L. Guo, L. Wang, L. Tang, Y. Xu, Genetic diversity and genetic structure of an endangered species, Trillium tschonoskii, Biochem. Genet. 43 (2005) 445–458. [2] L. Fu, China Plant Red Data Book, Volume 1 Rare and Endangered Plants, Science Publishing House Press, Beijing, 1992. [3] B.J. Xiao, G.D. Chen, Z.X. Tan, Studies on anti-inflammatory and immune effect of tschonoskii maxim, Anhui Med. Pharma. J. 9 (2005) 246–248. [4] L.L. Yu, K. Zou, J.Z. Wang, L.B. Zhu, Y. Zhou, J. Yang, Study on the anti-inflammatory, analgesic and thrombosis effects of extract of Trillium tschonoskii Maxim, Lishizhen Med. Mater. Med. Res. 19 (2008) 1178–1180. [5] J.C. Wei, S.L. Man, W.Y. Gao, X. Chai, W.S. Zhao, Y. Wang, S.S. Jing, Steroidal saponins from the rhizomes of Trillium tschonoskii maxim, Biochem. Syst. Ecol. 44 (2012) 112–116. [6] J. Chai, X.M. Song, X. Wang, Q.B. Mei, Z. Li, J.C. Cui, Z.S. Tang, Z.G. Yue, Two new compounds from the roots and rhizomes of Trillium tschonoskii Maxim, Phytochem. Lett. 10 (2014) 113–117. [7] X. Gao, W.J. Sun, Q. Fu, X.F. Niu, Rapid identification of steroidal saponins in Trillium Tschonoskii Maxim. by ultraperformance liquid chronmatography coupled to electrospray lonisation quadrupole time-of-flight tandem mass spectrometry, Phytochem. Anal. 26 (2015) 269–278. [8] C.M. Chu, H.Q. Zhang, P. Bu, Experimental study on the therapeutic effect of dioscin on rats with collagen-induced arthritis, Chin. Pharma. Bull. 28 (2012) 1464–1467. [9] J.Z. Wang, K. Zou, Y.M. Zhang, C. Liu, J. Wu, Y. Zhou, F.J. Dan, Y.X. Zhang, An 18norspirostanol saponin with inhibitory action against COX-2 production from the underground part of Trillium tschonoskii, Chem. Pharm. Bull. 55 (2007) 679–681. [10] P. Hu, D.H. Lia, X. Hua, S.G. Lia, C.M. Sai, X.C. Sun, T. Su, J. Bai, Z.H. Wang, Z.L. Li, H.M. Hua, Lignans and triterpenoids from Vitex negundo var. heterophylla and their biological evaluation, Fitoterapia 111 (2016) 147–153. [11] D. Zhao, B.M. Feng, S.F. Chen, G. Chen, Z.F. Li, X.J. Lu, X.N. Sang, X. An, H.F. Wang, Y.H. Pei, C21 steroidal glycosides from the roots of Cynanchum paniculatum, Fitoterapia 113 (2016) 51–57. [12] O.K. Kwon, M.Y. Lee, J.E. Yuk, S.R. Oh, Y.W. Chin, H.K. Lee, K.S. Ahn, Anti-inflammatory effects of methanol extracts of the root of Lilium lancifolium on LPS-stimulated raw264.7 cells, J. Ethnopharmacol. 130 (2010) 28–34. [13] J.M. Kim, H.A. Jung, J.S. Choi, N.G. Lee, Identification of anti-inflammatory target genes of Rhizoma coptidis extract in lipopolysaccharide-stimulated RAW264.7 murine macrophage-like cells, J. Ethnopharmacol. 130 (2010) 354–362.

45

[14] X.X. Wang, D. Hu, L.J. Zhang, G.N. Lian, S.Q. Zhao, C.M. Wang, J. Yin, C.F. Wu, J.Y. Yang, Gomisin A inhibits lipopolysaccharide-induced inflammatory responses in N9 microglia via blocking the NF-κB/MAPKs pathway, Food Chem. Toxicol. 63 (2014) 119–127. [15] J.Y. Tao, G.H. Zheng, L. Zhao, J.G. Wu, X.Y. Zhang, S.L. Zhang, Z.J. Huang, F.L. Xiong, C.M. Li, Anti-inflammatory effects of ethyl acetate fraction from Melilotus suaveolens Ledeb on LPS-stimulated RAW 264.7 cells, J. Ethnopharmacol. 123 (2009) 97–105. [16] Q.Z. Li, S.R. Yang, S.L. Yang, F.Z. Xin, M.J. Wang, Anti-inflammatory activity of phlomisoside F isolated from Phlomis younghusbandii Mukerjee, Int. Immunopharmacol. 28 (2015) 724–730. [17] C.M. Xiao, J. Huang, X.M. Zhong, X.Y. Tan, P.C. Deng, Two new homo-aro-cholestane glycosides and a new cholestane glycoside from the roots and rhizomes of Paris polyphylla var. pseudothibetica, Helv. Chim. Acta 92 (2009) 2587–2595. [18] R. Kasai, M. Okihara, J. Asakawa, K. Mizutani, O. Tanaka, 13C NMR study of α- and βanomeric pairs of D-mannopyranosides and L-rhamnopyranosides, Tetrahedron 35 (1979) 1427–1432. [19] D.J. Wadleigh, S.T. Reddy, E. Kopp, S. Ghosh, H.R. Herschman, Transcriptional activation of the cyclooxygenase-2 gene in endotoxin-treated RAW264.7 macrophages, J. Biol. Chem. 275 (2000) 6259–6266. [20] A.R. Brasier, The NF-kappaB regulatory network, Cardiovasc. Toxicol. 6 (2006) 111–130. [21] Y.S. Kim, C.B. Ahn, J.Y. Je, Anti-inflammatory action of high molecular weight Mytilus edulis hydrolysates fraction in LPS-induced RAW264.7 macrophage via NF-κB and MAPK pathways, Food Chem. 202 (2016) 9–14. [22] C. Guo, L. Yang, J. Luo, C. Zhang, Y.Z. Xia, T. Ma, L.Y. Kong, Sophoraflavanone G from Sophora alopecuroides inhibits lipopolysaccharide-induced inflammation in RAW264.7 cells by targeting PI3K/Akt, JAK/STAT and Nrf2/HO-1 pathways, Int. Immunopharmacol. 38 (2016) 349–356. [23] Z.X. Zhu, Y.F. Gu, Y.F. Zhao, Y.L. Song, J. Li, P.F. Tu, GYF-17, a chloride substituted 2(2-phenethyl)-chromone, suppresses LPS-induced inflammatory mediator production in RAW264.7 cells by inhibiting STAT1/3 and ERK1/2 signaling pathways, Int. Immunopharmacol. 35 (2016) 185–192. [24] X. Zhu, W.G. Fan, D.P. Li, H. Kung, M.C. Lin, Heme oxygenase-1 system and gastrointestinal inflammation: a short review, World J. Gastroenterol. 17 (2011) 4283–4288. [25] L. An, C.T. Liu, M.J. Yu, Z.H. Chen, X.G. Guo, W. Peng, J.F. Wang, X.Q. Fang, Y.H. Gao, S.Y. Yu, Heme oxygenase-1 system, inflammation and ventilator-induced lung injury, Eur. J. Pharmacol. 677 (2012) 1–4.