Anti-Inflammatory Effects and Mechanisms of Fatsia polycarpa Hayata and Its Constituents

Fatsia polycarpa, a plant endemic to Taiwan, is an herbal medicine known for treating several inflammation-related diseases, but its biological function needs scientific support. Thus, the anti-inflammatory effects and mechanisms of the methanolic crude extract (MCE) of F. polycarpa and its feature constituents, that is, brassicasterol (a phytosterol), triterpenoids 3α-hydroxyolean-11,13(18)-dien-28-oic acid (HODA), 3α-hydroxyolean-11-en-28,13β-olide (HOEO), fatsicarpain D, and fatsicarpain F, were investigated. MCE and HOEO, but not brassicasterol, dose-dependently inhibited lipopolysaccharide- (LPS-)induced expression of inducible nitric oxide synthase and cyclooxygenase-2 in RAW 264.7 macrophage line, whereas HODA, fatsicarpain D and fatsicarpain F were toxic to RAW cells. Additionally, MCE and HOEO suppressed LPS-induced production of nitric oxide, prostaglandin E2, and interleukin-1β and interfered with LPS-promoted activation of the inhibitor kappa B kinase (IKK)/nuclear factor-κB (NF-κB) pathway, and that of the mitogen-activated protein kinases (MAPKs) extracellular signal regulated kinase (ERK), c-Jun N-terminal kinase (JNK), and p38. In animal tests, MCE and HOEO effectively ameliorated 12-O-tetradecanoylphorobol-13 acetate- (TPA-)induced ear edema of mice. Thus, MCE of F. polycarpa exhibited an obvious anti-inflammatory activity in vivo and in vitro that likely involved the inhibition of the IKK/NF-κB pathway and the MAPKs, which may be attributed by triterpenoids such as HOEO.


Introduction
There are three species in the genus of Fatsia (belonging to the Araliaceae family), that is, Fatsia polycarpa Hayata, an evergreen shrub endemic to Taiwan, Fatsia japonica, and Fatsia oligocarpela, originated from Japan and Bonin Islands, respectively [1]. F. japonica and F. polycarpa have been used as an herbal medicine in Japan and in Taiwan in treating diseases, such as coughing, ankylosing spondylitis, osteoarthritis, rheumatism, rheumatoid arthritis and tendinitis, and in blood circulation improvement [1,2]. Nonetheless, the therapeutic activity of F. polycarpa has seldom been scientifically investigated and the bioactive constituents from these plants are not clear.

Isolation and Identification of MCE, Brassicasterol,
HODA, HOEO, Fatsicarpain D, and Fatsicarpain F. The airdried and powdered leaves and twigs of F. polycarpa (7.1 kg) were extracted with methanol (MeOH) for three days at room temperature (three times) and the combined extracts were concentrated in vacuum (under 35 ∘ C). The resulting dark green gum, that is, MCE, was suspended in H 2 O and extracted sequentially with CH 2 Cl 2 , ethyl acetate (EtOAc), and n-butanol (n-BuOH) (saturated with H 2 O). The CH 2 Cl 2 extract (100 g) was subjected to column chromatography on silica gel using n-hexane, n-hexane-EtOAc, and EtOAc-MeOH mixtures of increasing polarity for elution to furnish 40 fractions. Fraction 20 (7.0 g) eluted with n-hexane-EtOAc (1 : 10) was fractionated over Sephadex LH-20 (100% acetone) to afford a major phytosterol (580 mg), which was grown by slow evaporation of the mixture CH 2 Cl 2 -MeOH (1 : 1) solution at room temperature to give a suitable colorless crystal. The chemical structure of the crystal was confirmed by comparison of its spectroscopic data with those of brassicasterol [6,7], which was reported specifically in only one alga, an unidentified species of the order Sarcinochrysidales (Chrysophyceae).
MCE was dissolved in DMSO (dimethylsulfoxide) in a concentration of 50 mg/mL as a stock solution. Brassicasterol was dissolved in 100% ethanol in a concentration of 10 mM as a stock solution. HODA, HOEO, fatsicarpain D, and fatsicarpain F were dissolved in DMSO in a concentration of 10 mM as the stock solutions.

Animal Tests.
All protocols of animal tests were approved by the Committee for Animal Experiments of National Pingtung University of Science and Technology in accordance with international guidelines. Experiments were performed using groups of 8 male ICR mice (30-35 g; purchased from BioLASCO, Taipei, Taiwan) fed with a regular laboratory rodent diet and housed under a 12-hour light-dark cycle. TPA-induced ear edema of mice was performed according to previously published methods with modifications [8,9]. In the experiment for MCE, TPA (2.5 g/ear dissolved in 20 L of acetone) was topically applied to the surfaces of both ears of mice, whereas in the group of TPA control, TPA was applied to only the right ear. Meanwhile, in the group of normal control, 20 L of acetone was applied to both ears. Four hours later, MCE (100, 300, or 500 g/ear) or indomethacin (500 g/ear), both dissolved in acetone, was applied to the surface of the right ear, and acetone to the left ear, whereas for normal control and TPA control, acetone was applied to both ears. The thickness of ears was measured before and at 4 h, 16 h, and 24 h after TPA treatment using a dial thickness gauge (Peacock, Ozaki, Tokyo, Japan). The experiment for HOEO was performed similarly using a different batch of mice, but each group contained 6 mice, the dosage of TPA was 3.0 g/ear, and HOEO (300 g/ear) or indomethacin (500 g/ear) was applied to the ears at 1 h (instead of 4 h) after TPA treatment.

Cell Culture and Western Blot
Analysis. RAW 264.7 cells (ATCC number AT TIB-71) were cultured in DMEM (Dulbecco's modified Eagle's medium) containing 10% fetal bovine serum. FL83B cells (ATCC number CRL-2390) were cultured as previously described [10]. Both cells were incubated at 37 ∘ C in a humidified incubator supplied with 5% CO 2 .
RAW 264.7 cells were seeded at a density of 1.5−2.0 × 10 6 /35-mm plate, grown for 16 h, and treated with MCE or the natural products for 3 h. LPS was then added to the medium (final concentration 100 ng/mL) for 16 h. Subsequently, cells were washed with cold PBS (phosphate-buffered saline, pH 7.4) twice and submerged in lysis buffer (1X Cell Culture Lysis Reagent (Promega, Madison, WI, USA) containing 1 mM of phenylmethylsulfonyl fluoride, 1 g/mL of pepstatin, 1 g/mL of leupeptin, and 1 g/mL of aprotinin). For the detection of phosphorylated proteins, the lysis buffer also contained 10 mM of NaF, 1 mM of sodium orthovanadate, and 10 mM of sodium pyrophosphate. Cells were scraped off the plate on ice and the suspension was centrifuged at 14,000 ×g for 15 min at 4 ∘ C. The supernatant was collected and the protein concentration was analyzed using Bradford assay reagent (Bio-Rad, Hercules, CA, USA). Equal amounts of proteins were sampled and subjected to electrophoresis and Western blotting as described previously [11]. The resulting blots were scanned by a gel documentation and image analysis system (Syngene, Frederick, MD, USA), and band intensities were analyzed using the program supplied with the system. Alternatively, after hybridization with a horseradishperoxidase-(HRP-) conjugated secondary antibody, followed by incubation with "Western HRP substrate" (Luminata Classico, Millipore, Temecula, CA, USA), immunoreactive bands were detected using a biospectrum imaging system (UVP Biospectrum, UVP, LLC, Upland, CA, USA) and band intensities were analyzed by the supplied software.

Cytotoxicity
Assay. Cytotoxicity was assayed using MTT. RAW 264.7 (3×10 5 cell/well) or FL83B cells (2×10 5 cell/well) were cultured in a 96-well plate. Cells were washed twice with PBS, treated with the compounds in the indicated concentrations in serum-free medium for 24 h (HODA) or 19 h (other compounds). Cells were then washed twice with PBS and incubated in 30 L of 5 mg/mL MTT solution at 37 ∘ C in the dark for 1 h. Subsequently, DMSO (100 L) was added to each well for 10 minutes, withdrawn to another 96well plate, and the absorbance at 570 nm determined using a microplate reader (Molecular Devices, Sunnyvale, CA, USA). Experiments were performed in triplicate. Average ratio ± standard deviation of viable cells was calculated against the control (cells treated by the solvent).

Semiquantitative RT-PCR (Reverse Transcriptase-Polymerase Chain Reaction).
Total cellular RNA was purified using Trizol (Invitrogen) as per the manufacturer's instruction. The quality of RNA was checked by agarose gel electrophoresis. Sequences of the forward and reverse PCR primers of interleukin-1 (IL-1 ) and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were as published previously [12]. RT-PCR was performed using the SuperScript One-Step RT-PCR System kit (Invitrogen) with the following formulation: 2 g of total RNA, 0.4 M of forward primer and reverse primer, respectively, 12.5 L of 2X reaction mix, and 1 L of SuperScript RT/Platinum Taq. The final volume was adjusted to 25 L with sterilized distilled water. It was then subjected to PCR amplification in a thermal cycler by the following program: 48 ∘ C, 45 min (for cDNA synthesis); one cycle of 95 ∘ C, 2 min, 35 cycles of 95 ∘ C, 45 s, 60 ∘ C, 45 s, and 72 ∘ C, 1 min, and one cycle of 72 ∘ C, 5 min. The RT-PCR products were separated electrophoretically on 1% agarose gels and visualized by ethidium bromide staining. The resulting images were recorded and band intensities were quantified as described previously [12]. Meanwhile, the RT-PCR products were purified and the sequences confirmed by DNA sequencing (Mission Biotech, Taipei, Taiwan).

Analysis of Nitric Oxide (NO) and
Prostaglandin E 2 (PGE 2 ). The concentration of NO in the culture medium of cells was analyzed using a kit of Griess reagent according to the manufacturer's protocol (Promega). That of PGE 2 was analyzed using an ELISA kit as per the protocol suggested by the manufacturer (R&D Systems, Minneapolis, MN, USA).

Protein Extraction from the Nuclear and Cytoplasmic
Fractions of Cells. RAW 264.7 cells were seeded on 10 mm dishes until 80% confluence and treated with MCE or HOEO for 3 h, followed by treating with LPS for 2 h. Cells were harvested, washed with cold PBS twice, and centrifuged at 500 ×g for 5 min. Nuclear proteins and cytoplasmic proteins of the cell pellet were then separated and extracted using a nuclear extract kit as per the manufacturer's instruction (Millipore, Temecula, CA, USA). Protein concentrations in the extracts were quantified using Bradford assay reagent before being subjected to Western blot analysis.

Statistical
Analysis. Data were analyzed by one-way analysis of variance (ANOVA) followed by Scheffe's post hoc test and significance was considered when < 0.05 and > 3.5546.

In Vivo Efficacy of MCE in TPA-Induced Mouse Ear
Edema. To assess the anti-inflammatory effect of MCE in vivo, a topical inflammation model, TPA-induced mouse ear edema, was used. Both ears of mice were stimulated with TPA to induce inflammation. The right ear was treated with MCE (100, 300, or 500 g/ear) or indomethacin (500 g/ear, as a reference drug) [13,14] at 4 h after TPA stimulation, whereas the left ear remained unmedicated for comparison. As shown in Figure 2(a), ear edema was obvious at 4 h, 16 h, and 24 h after TPA application without any medication (TPA control), while all three dosages of MCE significantly reduced the edema of the right ear at 16 h and 24 h in a time-dependent manner and in an extent comparable to the effect of indomethacin. While comparing the thickness of the right ear and that of the unmedicated left ear at 24 h after TPA stimulation, the data also revealed that all three dosages of MCE effectively ameliorated the TPA-induced edema (Figure 2(b)). These results confirmed the anti-inflammatory activity of MCE in vivo. Accordingly, the anti-inflammatory effects of MCE and its characteristic components brassicasterol, HODA, HOEO, fatsicarpain D, and fatsicarpain F were further investigated.

The Anti-Inflammatory Effects of MCE and Its Feature
Constituents on Macrophage Cells. The cytotoxicity of MCE and the compounds was evaluated in RAW 264.7 macrophage cell line and in FL83B normal liver cell line. MCE did not display obvious cytotoxicity in FL83B cells in concentrations of 10-400 g/mL but did result in reduced cell survival in 400 g/mL in RAW 264.7 cells (Figure 3(a)). Brassicasterol did not show obvious cytotoxicity in both cell lines in 10-100 M (Figure 3(b)). HOEO did not have a significant toxic effect in FL83B cells in concentrations of 10-100 M but did result in about 20% inhibition for the growth of RAW 264.7 cells in 100 M (Figure 3(c)). HODA, fatsicarpain D, and fatsicarpain F on the other hand, obviously suppressed the growth of RAW 264.7 cells in concentrations of 50 M or ≧20 M, but they were not toxic to FL83B cells ( Figures  3(d), 3(e), and 3(f), resp.). Therefore, the anti-inflammatory effects of MCE, brassicasterol, and HOEO were further characterized using RAW 264.7 cells as a model. Those of HODA, fatsicarpain D, and fatsicarpain F will be analyzed using other appropriate models.
Inducible nitric oxide synthase (iNOS) and cyclooxygenase-2 (COX-2) are the major enzymes involved in the synthesis of the inflammatory mediators, NO, and PGE 2 , respectively, and are common markers of inflammation Thus, the anti-inflammatory effect of HOEO was examined in animal models. As shown in Figure 5(a), the application of 300 g/ear of HOEO on the right ear significantly reduced the extent of mouse ear edema at 16 h, 20 h, and 24 h after TPA treatment and the effect was similar with that of 500 g/ear of indomethacin. Figure 5(b) showed the comparison of the extents of edema between the right ear and the unmedicated left ear on the same group of mice at 24 h after TPA stimulation, and the findings also supported that HOEO effectively hindered TPA-induced edema. Consequently, these results confirmed the anti-inflammatory activity of HOEO in vivo.

Interference of MCE and HOEO with the Activation of the IKK/NF-B Pathway and MAPKs.
The activation of nuclear factor-B (NF-B) plays a key role in mediating inflammatory responses. LPS is known to activate IKK (I B kinase), which in turn catalyzes the phosphorylation of I B (inhibitor of NF-B), leading to I B degradation and the release of NF-B, which enters the nucleus and activates the expression of proinflammatory factors such as iNOS, COX-2, and IL-1 [17][18][19].
As revealed in Figures 6(a) and 6(b), MCE strongly inhibited LPS-induced phosphorylation of IKK (A) and I B (B) (lanes 5 and 6) as compared with cells treated with LPS alone (lanes 3 and 4). HOEO exhibited a less potent inhibitory effect on the activation of IKK (Figure 6(a),  lanes 7 and 8 versus lanes 3 and 4), but still showed a substantial suppressive effect on the phosphorylation of I B (Figure 6(b), lanes 7 and 8 versus lanes 3 and 4). Therefore, LPS-induced I B phosphorylation was obviously inhibited by both MCE and HOEO. Furthermore, Figure 6(c) showed that the nuclear translocation of p65, an NF-B subunit, was apparently increased by LPS stimulation (lanes 3 and 4) as compared to untreated cells (lanes 1 and 2), whereas MCE (lanes 5 and 6) or HOEO (lanes 7 and 8) critically suppressed this LPS-induced effect, indicating that MCE and HOEO inhibited the shift of NF-B to the nucleus. Overall, these data suggest that MCE and HOEO inhibit the LPS-mediated activation of IKK/NF-B signaling. Mitogen-activated protein kinases (MAPKs), including extracellular signal-regulated kinase (ERK)1/2, c-Jun Nterminal kinase (JNK), and p38, are also known to mediate the LPS-induced inflammatory responses [20][21][22][23]. Thus, whether MCE and HOEO could inhibit the activation of JNK, ERK, and p38 was examined. As shown in Figure 6

Discussion
This study scientifically proved that the crude extract of F. polycarpa is a potential anti-inflammatory agent. Animal tests confirmed the anti-inflammatory effect of MCE in vivo.
Evidence-Based Complementary and Alternative Medicine 9 Topical application of MCE for 12-20 h effectively ameliorated acute skin inflammation on mice ear. While compared with indomethacin, a medicine for inflammation, MCE showed a similar or even stronger effect in reducing ear edema of mice. Meanwhile, in RAW 264.7 macrophage line, MCE obviously inhibited LPS-stimulated expression or production of iNOS, COX-2, NO, PGE 2 , and IL-1 . Overall, these results support the traditional use of F. polycarpa as an herbal medicine in treating inflammatory-related diseases.
MCE obviously reduced the phosphorylation of IKK and I B and the nuclear translocation of NF-B. These findings suggested that MCE inhibited the IKK/NF-B pathway. MAPKs are another group of signaling pathways regulating inflammation. MCE critically suppressed LPS-induced activation of ERK and p38 and had a milder inhibitory effect on that of JNK. Generally, these results suggest that MCE interferes with the IKK/NF-B pathway and the MAPK pathways, which likely underlies, at least in part, the mechanism of its anti-inflammatory activity.
MCE was rich in triterpenoids as well as in phytosterols. It is rational to speculate that these components may contribute to the anti-inflammatory function of F. polycarpa. Thus, brassicasterol and a few major or feature triterpenes of MCE were characterized. Yet, brassicasterol did not exhibit an obvious anti-inflammatory activity, implying that brassicasterol may not contribute to the anti-inflammatory effect of MCE. Consistently, brassicasterol is known to be a phytosterol commonly present in many plants and foods [24][25][26], including those unknown to possess an anti-inflammatory activity. Thus, it is reasonable to infer that brassicasterol does not have an anti-inflammatory effect.
On the other hand, HOEO displayed a significant antiinflammatory activity both in vivo and in vitro. HOEO obviously diminished TPA-induced ear edema in mice and inhibited the expression of proinflammatory markers in LPSstimulated RAW 264.7 cells, including iNOS, COX-2, IL-1 , NO, and PGE 2 . Furthermore, in accordance with the action of MCE, HOEO interfered with the activation of the IKK/NF-B pathway and that of MAPKs ERK, JNK, and p38. Hence, HOEO likely represents one of the constituents contributing to the anti-inflammatory function of F. polycarpa. It is very likely that other components in MCE also play roles in the anti-inflammatory effect of F. polycarpa and deserve to be explored.
Interestingly, HODA, fatsicarpain D, and fatsicarpain F showed obvious inhibitory effects on the survival of RAW 264.7 cells, but they were found nontoxic to the normal liver line FL83B cell. The molecular mechanism for this effect is not clear. Nonetheless, promoting the death of macrophages without hurting other somatic cells could also be an effect of anti-inflammation. Thus, the actions, immunomodulatory effects, and molecular mechanisms of these molecules should be further investigated.

Conclusions
Evidence presented in this study suggests that the crude extract from the leaves and twigs of Fatsia polycarpa is a potent anti-inflammatory agent both in vitro and in vivo.
The anti-inflammatory activity of MCE is probably not due to brassicasterol, a major phytosterol present in the extract, but is likely contributed by other components in the extract, such as triterpenoids like HOEO. The molecular mechanism underlying the effects of MCE and HOEO probably involves the inhibition of the IKK/NF-B pathway and the MAPKs. To further our understanding, more studies should be conducted to continuously explore the effective constituents in the extract of F. polycarpa.