Noble 3,4-Seco-triterpenoid Glycosides from the Fruits of Acanthopanax sessiliflorus and Their Anti-Neuroinflammatory Effects

Acanthopanax sessiliflorus (Araliaceae) have been reported to exhibit many pharmacological activities. Our preliminary study suggested that A. sessiliflorus fruits include many bioactive 3,4-seco-triterpenoids. A. sessiliflorus fruits were extracted in aqueous EtOH and fractionated into EtOAc, n-BuOH, and H2O fractions. Repeated column chromatographies for the organic fractions led to the isolation of 3,4-seco-triterpenoid glycosides, including new compounds. Ultra-high-performance liquid chromatography (UPLC) mass spectrometry (MS) systems were used for quantitation and quantification. BV2 and RAW264.7 cells were induced by LPS, and the levels of pro-inflammatory cytokines and mediators and their underlying mechanisms were measured by ELISA and Western blotting. NMR, IR, and HR-MS analyses revealed the chemical structures of the nine noble 3,4-seco-triterpenoid glycosides, acanthosessilioside G–O, and two known ones. The amounts of the compounds were 0.01–2.806 mg/g, respectively. Acanthosessilioside K, L, and M were the most effective in inhibiting NO, PGE2, TNF-α, IL-1β, and IL-6 production and reducing iNOS and COX-2 expression. In addition, it had inhibitory effects on the LPS-induced p38 and ERK MAPK phosphorylation in both BV2 and RAW264.7 cells. Nine noble 3,4-seco-triterpenoid glycosides were isolated from A. sessiliflorus fruits, and acanthosessilioside K, L, and M showed high anti-inflammatory and anti-neuroinflammatory effects.


Introduction
Oxidative stress in the human body is caused by excess reactive oxygen or nitrogen species (ROS or RNS), including superoxide (O 2 − ), hydrogen peroxide (H 2 O 2 ), and nitric oxide (NO), and it triggers certain types of apoptotic cell death, such as neuronal cell death and macrophage immune injury. If ROS or RNS are not appropriately removed, oxidative stress can cause the lipid peroxidation of the cellular membrane, leading to inordinate cell death [1].
Recently, the number of patients with degenerative brain disease has increased due to the aging society [2]. Therefore, the investigation of the cause and drug development for degenerative brain disease is rapidly progressing. The microglia in the central nerve system (CNS) plays an important role in the immunity, degeneration, and inflammation of the CNS, so it can prevent degenerative brain diseases by regulating the inflammatory

UPLC-QTOF/MS and UPLC-MS/MS Analyses
UPLC was performed using a Waters ACUITY I-CLASS UPLC (Waters Corp., Milford, MA, USA) with a Thermo Hypersil GOLD column (2.1 mm × 100 mm; 1.9 µm). The mobile phases were composed with Solvent A (water and 0.1% formic acid (v/v)) and Solvent B (acetonitrile and 0.1% formic acid (v/v)). The flow rate was 450 µL/min and the injection volume was 2 µL. The elution conditions were as follows: 0-4 min, B 10-30%; 4-15 min, B 30-60%; 15-16 min, B 60-100%; 16-18 min, B 100-10. The column oven and sample tray were maintained at 40°C and 4°C, respectively. MS analysis was performed using a Waters Xevo G2-S QTOF MS (Waters Corp., Milford, MA, USA) operating in negative ion mode. Accurate mass measurements were obtained by an automated calibration delivery system that contained an internal reference (Leucine-enkephalin, m/z 554.262 (ESI-)). The operating parameters were set by modifying those from the preceding research [20]. UPLC with tandem mass spectrometry (MS/MS) was carried out for the quantitative analysis of compounds in the A. sessiliflorus fruits. Mass spectrometric detection was carried out using 3200 QTRAP mass spectrometer (AB SCIEX, Framingham, MA, USA) in multiple reaction monitoring (MRM) mode. Precursor ions of the compounds were selected. They and product ions were generated by applying collision energies to selected precursor ions, and product ions were used for quantification. The optimal operating parameters were set as shown in Table 3.

Determination of Nitrite
As an indicator of nitrite oxide (NO) production in cells, the production of nitrite, a stable end-product of NO oxidation, was measured. Briefly, the concentration of nitrite in the conditioned media was determined by a method based on the Griess reaction [23]. The details of the assay have been described previously [21].

PGE 2 Assay
The concentration of PGE 2 in each sample was measured with the use of a commercially available kit from R&D Systems, Inc (Minneapolis, MN, USA), according to a method described previously [21]. Briefly, BV2 and RAW264.7 cells treated with compounds were cultured in 48-well plates, followed by a pre-incubation with different concentrations of compounds for 3 h. Subsequently, stimulation was performed for 24 h with LPS stimulation (1 µg/mL). The resulting cell culture supernatants were collected and centrifuged at 13,000× g for 2 min to remove particulate matter. At the end of the procedure, the samples were added to a 96-well plate pre-coated with polyclonal antibodies specific to PGE 2 . Enzyme-linked polyclonal antibodies were added to the wells and allowed to react for 20 h, followed by a final washing step to remove any unbound antibody-enzyme reagent. A substrate solution was added, after which the intensity of the color produced was measured at 450 nm (a correction wavelength set at 540 or 570 nm), which was proportional to the amount of PGE 2 present.

Assays for IL-1β, IL-6, and TNF-α
The culture medium was collected to determine the levels of IL-β, IL-6, and TNF-α present in each sample using ELISA kits tailored to each collection process (R&D Systems, Inc.), as per the manufacturer's instructions. Briefly, BV2 and RAW264.7 cells were seeded in 48-well culture plates at a density of 5 × 10 5 cells/well. After incubation, the supernatant was collected and used in the cytokine ELISA kits for measuring the concentrations of IL-1β, IL-6, and TNF-α.

Western Blotting Analysis
The pelleted BV2 and RAW264.7 cells were washed with PBS and then lysed in RIPA buffer. Equal amounts of proteins quantified by Protein Assay Dye Reagent Concentrate obtained from Bio-Rad Laboratories (#5000006; Hercules, CA, USA), mixed in the sample loading buffer and separated by SDS-PAGE. The separated proteins were transferred to a nitrocellulose membrane. Non-specific binding to the membrane was blocked by incubation in a solution of skimmed milk. The membrane was incubated with primary antibodies at 4 • C overnight and then reacted with a horseradish peroxidase-conjugated secondary antibody from Millipore.

Statistical Analysis
Data are presented as the mean ± standard deviation of three independent experiments. A one-way analysis of variance, followed by Dunnett's comparison tests, was used to compare the two groups. Statistical analyses were performed using GraphPad Prism software, Version 5.01 (GraphPad Software Inc., San Diego, CA, USA).

Chemical Structure Elucidation of Compounds 1-11
A 70% ethanolic extract of dried A. sessiliflorus fruits was suspended in H 2 O and extracted successively with EtOAc and n-BuOH. The EtOAc-and n-BuOH-soluble fractions were concentrated under reduced pressure to produce a residue that was subjected to multiple chromatographic steps, using Diaion HP-20, Sephadex LH-20, silica gel, and reversed-phase C18 silica gel, yielding compounds 1-11. Comparing the 1D-and 2D-NMR and QTOF/MS data with reported values allowed us to identify known compounds sessiloside (4) [18] and inermoside (6) [19]. The other nine noble compounds are newly reported here (Figure 1).

Analyses of Compounds 1-11 in ASFE by UPLC-QTOF/MS and Tandem Mass Spectrometry (MS/MS)
Compounds 1-11 were analyzed using ultra-high-performance liquid chromatography coupled with quadrupole time-of-flight mass spectrometry (UPLC-QTOF/MS). The extract of A. sessiliflorus fruits (ASFEx) was also subjected to UPLC-QTOF/MS. Four compounds, namely compound 1-3 and 5, were determined in the extract but the intensities in the extract were too low. As a result of this, fractionation was carried out. A typical base peak intensity (BPI) chromatogram of the extract and the n-butanol fraction (AFB) are shown in Figure 2. After fractionation, eight compounds, 1, 3-7, 9, and 10, were detected in the UPLC chromatogram of the AFB. However, the intensities of compounds were low in the nbutanol fraction for quantitative analysis using UPLC-QTOF/MS. Hence, QTRAP ® tandem mass spectrometry (MS/MS) was used for quantifying compounds 1-11. The precursor ions of noble compounds, acanthosessilioside G-O were selected under optimized multiple reaction monitoring (MRM) conditions, and collision energy was applied to the selected precursor ions. After that, the product ions with the highest intensities were selected for quantitative analysis ( Table 4). The linear calibration curves were based on regression analysis of the values measured with various concentration of compounds 1-11. The values of the calibration plots are shown in Table 5. The result of quantitative analysis was measured in the range of 0.01-2.806 mg/g.

Effects of Compounds 1-11 on Cell Viability and Nitrite Contents in BV2 and RAW264.7 Cells
To examine whether compounds 1-11 have cell toxicity, the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assay was conducted in BV2 and RAW264.7 cells. The cells were incubated for 48 h with various concentrations of compounds 1-11. The cell viability did not alter following the addition of 40 µM concentrations of compounds 1-6, 10, and 11 for 48 h; however, compound 7 at 20 µM, Compound 8 at 40 µM, and compound 9 at 20 µM demonstrated cell toxicity in BV2 and RAW264.7 cells. Therefore, adequate concentrations of compounds 1-11 in subsequent experiments were determined based on the results of Figure 3.
To assess the anti-inflammatory and neuroinflammatory effects of compounds 1-11 in LPS-stimulated BV2 and RAW264.7 cells, nitrite concentration was measured using the Griess reagents. Over-production of iNOS-derived NO may promote the formation of RNS, aggravate the inflammatory response, and can even lead to neuronal cell death. LPS stimulates macrophage or/and microglia and advances the production of ROS [24]. Therefore, reducing ROS production in microglia and macrophage cells might be an effective strategy to protect against inflammatory damage. In other previous studies, LPS treatment significantly increased ROS production [25]. LPS-treated groups significantly increased nitrite concentration compared with the control group in both BV2 and RAW264.7 cells. In BV2 cells, the increased nitrite concentration was significantly inhibited by compounds 2, 5, 6, 7, 8, 9, and 10. In RAW264.7 cells, the increased nitrite concentration was significantly inhibited by compounds 2, 5, 6, 7, 8, and 9 ( Figure 4). Among them, compounds 7, 8, and 9 were the most effective in inhibiting nitrite production in both BV2 and RAW264.7 cells. These results showed that, among the eleven compounds, compounds 7, 8, and 9 are suitable for further experiments aimed at determining the biological mechanism. Macrophages and microglia play an important role in maintaining homeostasis of our body, they can be activated by many stimulates, such as ROS, external stimuli, and pro-inflammatory mediators. Cytokines have a complex regulatory action on inflammatory and immune responses. Pro-inflammatory cytokines, which contain both IL-6 and TNF-α, are mainly produced by various types of activated immune cells and are associated with all inflammatory properties of inflammatory diseases [26]. Therefore, we assessed the inhibitory effects of compounds 7-9 on the LPS-induced production of NO, PGE 2 , TNF-α, IL-1β, and IL-6. BV2 and RAW264.7 cells were incubated with compounds 7-9 for 3 h and then stimulated with LPS for 24 h. The production of PGE 2 , TNF-α, IL-1β, and IL-6 was significantly inhibited by compounds 7-9 in BV2 cells ( Figure 5A-D). Figure 5E-H show that compounds 7-9 also significantly repressed PGE 2 , TNF-α, IL-1β, and IL-6 in BV2 cells. These results suggest that compounds 7-9 exert anti-neuroinflammatory and anti-inflammatory effects by modulating inflammatory mediators and cytokines.    Inflammatory cytokines and mediators are rapidly increased after macrophages and microglia are activated. The expression of pro-inflammatory proteins can also be increased. In BV2 and RAW264.7 cells, LPS induces NO production through activating iNOS expression. LPS also increases COX-2 expression, which mediates the synthesis of prostaglandins and cytokines. NO or prostaglandins regulate various biological functions of immune responses [27]. Therefore, we assessed iNOS and COX-2 expression in BV2 and RAW264.7 cells that affect the production of inflammatory mediators and cytokines, respectively. Cells were pretreated for 3 h with indicated concentrations of compounds 7-9 and stimulated for 24 h with LPS. The LPS-induced expression of iNOS and COX-2 in both types of cells was significantly inhibited by compounds 7-9 ( Figure 6). In this result, compounds 7-9 significantly regulated iNOS and COX-2 protein expression, suggesting that it exhibits inhibitory effects on inflammatory mediators and cytokines. In mammalian cells and tissues, many physiological and pathological responses are mediated by the mitogen-activated protein kinase (MAPK) signaling pathway, including stress responses, inflammation, and apoptosis. ROS modify the gene expression of proinflammatory mediators by altering MAPK cascades [28]. MAPKs are activated during the process, releasing ILs, TNFs, and various inflammatory mediators such as NO, PGE 2 , histamine, and lysosome granules [29]. Therefore, we explored the effects of compounds 7-9 on the MAPK pathway activation in BV2 and RAW264.7 cells. Cells were incubated with compounds 7-9 for 3 h and stimulated with LPS for 30 min. First, compound 7 inhibited the LPS-induced phosphorylation of p38 and ERK MAPKs, but did not reduce the phosphorylation of JNK in both BV2 and RAW264.7 cells (Figure 7). Compound 8 showed an excellent inhibitory effect on the LPS-induced phosphorylation of p38 and ERK MAPK in both BV2 and RAW264.7 cells, but not the phosphorylation of JNK (Figure 7). Compound 9 showed a significant inhibitory effect on LPS-induced phosphorylation of p38 MAPK and showed a tendency to inhibit ERK MAPK phosphorylation in BV2 cells. In addition, Compound 9 significantly inhibited the phosphorylation of p38 and ERK MAPK in RAW264.7 cells. However, it did not decrease the phosphorylation of JNK in both BV2 and RAW264.7 cells (Figure 7). Taken together, in both BV2 and RAW264.7 cells, compounds 7-9 (acanthosessilioside K, L, and M) had inhibitory effects on the phosphorylation of p38 and ERK MAPKs, but not the phosphorylation of JNK. In addition, compound 8 (acanthosessilioside L) had the most effects on the inhibition of MAPK phosphorylation. These results indicate that acanthosessilioside L significantly regulated the inflammatory mediators and cytokines through the inhibiting p38 and ERK MAPK. Cell extracts were analyzed using Western blotting with antibodies specific for phosphorylated p-p38, p-JNK1/2, and p-ERK1/2. Membranes were stripped and re-probed to measure total abundance of each mitogen-activated protein kinase (MAPK) as a control measurement. Representative blots from three independent experiments are shown.

Conclusions
Nine noble 3,4-seco-triterpenoid glycosides, acanthosessilioside G-O, along with two previously known triterpenoid glycosides were isolated from A. sessiliflorus fruits, and the chemical structures were determined without ambiguity based on the intensive analysis of 1D-NMR, 2D-NMR, UV, IR, and MS data. In this study, the LC-MS/MS MRM analysis method for quality control of the A. sessiliflorus fruit was first developed using the isolated compounds 1-11. The advantages of hybrid LC-QTOF mass spectrometry include not only accurate and sensitivity, but also fast LC-MS/MS MRM analysis, making structural elucidations easier. It can be used for the qualitative and quantitative determination of minor or novel compounds, which is helpful in improving the quality control of A. sessiliflorus fruits. Acanthosessilioside K, L, and M (7-9) were the most effective in inhibiting NO, PGE 2 , TNF-α, IL-1β, and IL-6 production and reducing iNOS and COX-2 expression. In addition, these compounds had inhibitory effects on p38 and ERK MAPK phosphorylation in both BV2 and RAW264.7 cells. Our results suggest that acanthosessilioside K, L, and M could be good candidates for the development of therapeutic agents for inflammatory and neuroinflammatory diseases.

Conflicts of Interest:
The authors declare that there is no conflict of interest.