DMEM and FBS were purchased from Gibco (Thermo Fisher Scientific, Inc.). Chlorogenic acid, catechin, quercetin, quercetin-3-O-glucoside, formic acid, diclofenac, and LPS (Escherichia coli O55:B5) were purchased from Sigma-Aldrich (Merck KGaA). Antibodies against inducible nitric oxide synthase (iNOS; cat. no. 610431; BD Biosciences), cyclooxygenase-2 (COX-2; 1:2,000; cat. no. sc-376861; Santa Cruz Biotechnology, Inc.), ERK [1:2,000; cat. no. 9102S; Cell Signaling Technology, Inc. (CST)], phosphorylated (p)-ERK (1:1,000; cat. no. 9101S; CST), JNK (1:2,000; cat. no. 9258S; CST), p-JNK (1:1,000; cat. no. 9251S; CST), p38 (1:2,000; cat. no. 9212S; CST), p-p38 (1:1,000; cat. no. 9211S; CST), inhibitor of (I)κBα (1:2,000; cat. no. 9242S; CST), p-IκBα (1:1,000; cat. no. 2859S; CST), and β-actin (1:5,000; cat. no. sc-47778; Santa Cruz Biotechnology, Inc.) were used for western blot analysis. Horseradish peroxidase (HRP)-conjugated secondary antibodies for western blots, goat anti-mouse IgG (1:5,000; cat. no. 1706516; Bio-Rad Laboratories, Inc.) and goat anti-rabbit IgG (1:5,000; cat. no. 1706515; Bio-Rad Laboratories, Inc.) were used. For immunocytochemistry, p65 (1:500; cat. no. 8242S; CST Biological Reagents Co., Ltd.) and glial fibrillary acidic protein (GFAP; 1:500; cat. no. MAB3402; Merck KGaA) were used as primary antibodies, and Alexa Fluor 488 donkey anti-mouse IgG (1:500; cat. no. ab150105; Abcam) and Alexa Fluor 594 donkey anti-rabbit IgG (1:500; cat. no. ab150076; Abcam) were used as secondary antibodies.

In 2008, P. persica was obtained from various regions in Vietnam (Bằng Lũng, Chợ Đồn, and Bắc Kạn). The samples were authenticated by the chief executive of the Institute of Ecology and Biological Resources. It is characterized by 4–7 m tall, dark green, deciduous leaves and flowers in single, semi-double and double form in colors from white to deep red. A voucher specimen (KRIB0018669, accession number of KRIBB) was deposited in the herbarium of the Korea Research Institute of Bioscience and Biotechnology. P. persica dried and refined aerial parts (leaves, fruits and twigs; total weight, 63 g) were extracted with 600 ml of 99.9% (v/v) methanol and were repeatedly sonicated with 1,500 W input and 40 kHz frequency for 15 min at 2-h intervals for 3 days. All procedures were performed at 45°C. The product was filtered through non-fluorescent cotton and concentrated by rotary evaporation using an N-1000SWD device (EYELA Co., Ltd.) under reduced pressure at 45°C. Finally, 8.05 g of PPB was obtained by freeze-drying and then dissolved in DMSO (Sigma-Aldrich; Merck KGaA).

Chlorogenic acid, catechin, quercetin and quercetin-3-O-glucoside were selected as marker substances in PPB based on a previous study (35). The extract contents were measured using high-performance liquid chromatography coupled with tandem mass spectrometry (HPLC-MS/MS). Briefly, to identify the product ions of each substance, the marker substances (100 ng/ml) were individually infused into the mass spectrometer at a flow rate of 10 µl/min. Precursor ions and fragmentation patterns were monitored in a negative-ion mode. Using an API 4000 LC/MS/MS system (AB SCIEX Analytical Instrument Trading Co.) equipped with an electrospray ionization interface, major peaks in the MS/MS scan were used to quantify the substances. The compounds were separated on a reversed-phase column (Kinetex^® C18, 2.6-µm particle size, 100×2.1 mm internal diameter; Phenomenex Ltd.) in the mobile phase of a water:acetonitrile mixture at a 3:7 (v/v) ratio including 0.1% formic acid. The column was heated at 30°C, and the mobile phase was delivered at a flow rate of 0.2 ml/min using an HP 1100 series pump (Agilent Technologies, Inc.). The injection volume was 5 µl. The turbo ion spray interface was operated at −4,200 V at 450°C. Nitrogen was used as a curtain gas at 35 psi, and air dried was nebulized as a collision gas at 4 psi and ion source gases 1 (45 psi) and 2 (55 psi). The ion transitions of the precursor to the product ion were monitored in negative mode as deprotonated ions [M-H]⁻ at m/z 353.0 → 190.8 [declustering potential (DP), −60 eV; collision energy (CE), −22 eV] for chlorogenic acid, 289.0 → 244.8 (DP, −90 eV; CE, −20 eV) for catechin, 301.0 → 151.0 (DP, −100 eV, CE, −32 eV) for quercetin, 463.1 → 300.0 (DP, −120 eV, CE, −36 eV) for quercetin-3-O-glucoside, and 296.1 → 251.7 (DP, −100 eV, CE, −32 eV) for diclofenac (internal standard). Quantification was performed by selective reaction monitoring of deprotonated precursor ions and related product ions using the ratio of the area under the peak for each solution. All analytical data were processed using Analyst software (version 1.5.2; Applied Biosystems; Thermo Fisher Scientific, Inc.). Standard marker substance solutions were serially diluted with methanol including the internal standard (IS, 100 ng/ml) to obtain 10–1,000 ng/ml concentrations. Using linear regression, two calibration graphs were derived from the ratios between the areas under the peaks for each substance and those of the IS. PPB (20 mg/vial) was dissolved in 1 ml of methanol and was serially diluted with methanol including the IS by 1,000-fold. The concentrations of each marker substance were predicted using the calibration graph, and their contents within the extracts were calculated.

The BV2 cells (CLC catalog code, ATL03001) were provided by Dr Dong-Kug Choi (Konkuk University, Chungju-si, Chungcheongbuk-do, Republic of Korea) and were maintained in DMEM containing 5% FBS, 100 U/ml penicillin, and 100 µg/ml streptomycin (Welgene, Inc.) in a humidified incubator with air containing 5% CO₂. In all of the experiments, 500,000 cells were plated and treated with PPB (20, 100 and 200 µg/ml) for 1 h prior to LPS treatment (200 ng/ml, cat. no. L2880; Sigma-Aldrich; Merck KGaA), and the vehicle was a DMSO only treatment (0.1%). BV2 cells were treated with LPS for 6 and 24 h for reverse transcriptase polymerase chain reaction (RT-PCR) and western blots, respectively.

Hydrogen peroxide (H₂O₂, 250 µM) was applied to BV2 cells for 24 h to measure cytotoxicity.

Primary astrocytes were prepared from the cortices of postnatal day 2 Sprague-Dawley rats (7–9 g), as previously described by McCarthy and de Villis (36). Postnatal day 1 Sprague-Dawley rats were purchased (Young Bio) and were maintained for 24 h in controlled temperature (20–25°C), humidity (50–55%) and 12-h light/dark cycle with their mother. A total of 6 animals (3 males and 3 females) were used for this study. All animals were handled in accordance with the Principle of Laboratory Animal Care (37) and the study was approved by the Institutional Animal Care and Use Committee of Chung-Ang University (2017-00093).

Briefly, the astrocytes were plated in poly-d-lysine (PDL)-coated T-75-cm² flasks in DMEM/F12 (Gibco; Thermo Fisher Scientific, Inc.) supplemented with 10% FBS, 100 U/ml penicillin, and 100 µg/ml streptomycin. Microglia were removed at 7 days in vitro (DIV), and subcultured astrocytes at 15 DIV were used for further experiments. Primary astrocytes were treated with PPB (20, 100 and 200 µg/ml) for 1 h prior to LPS treatment (10 ng/ml), and the control group was treated with vehicle (DMSO; 0.1%).

Cell viability was determined using the MTT assay. Cells were treated with PPB for 24 h and incubated with MTT (100 µg/ml) for 2 h. DMSO was added to dissolve the insoluble formazan crystals that had formed within viable cells, and the absorbance was measured at 540 nm.

BV2 cells, which were protected from the light, were fixed with 70% ethanol at for 30 min at 20–25°C and stained using FxCycle PI/RNase staining solution (cat. no. F10797; Thermo Fisher, Inc.) for 30 min at 20–25°C. Nuclei were stained with DAPI (cat. no. D1306; Invitrogen; Thermo Fisher Scientific, Inc.) for 10 min at 20–25°C. Cell images were visualized using a confocal microscope (Zeiss GmbH).

NO release was measured from nitrite levels using a Griess reagent (0.1% naphthylethylenediamine and 1% sulfanilamide in 5% H₃PO₄). The supernatant was mixed with an equal volume of Griess reagent for 5 min at 20–25°C. Mixture absorbance was measured at 540 nm using a microplate reader (Tecan Group, Ltd.).

Total RNA was isolated from LPS-stimulated cells using TRIzol reagent (Invitrogen; Thermo Fisher Scientific, Inc.) according to the manufacturer's instructions. Reverse transcription was conducted for 5 min at 70°C and 5 min at 4°C with total RNA (1 µg) and GoScript Reverse Transcriptase (Promega Corporation). cDNA was subsequently amplified by RT-PCR with specific primers and GoTaq DNA polymerase (Promega Corporation). The PCR primer sets were as follows: iNOS (sense, 5′-GAGGTACTCAGCGTGCTCCA-3′; antisense, 5′-AGGGAGGAAAGGGAGAGAGG-3′), COX-2 (sense, 5′-TGAGTGGTAGCCAGCAAAGC-3′; antisense, 5′-CTGCAGTCCAGGTTCAATGG-3′), TNF-α (sense, 5′-AGGGAGAGTGGTCAGGTTGC-3′; antisense, 5′-CAGCCTGGTCACCAAATCAG-3′), IL-1β (sense, 5′-CAAGGAGAACCAAGCAACGA-3′; antisense, 5′-TTGGCCGAGGACTAAGGAGT-3′), IL-6 (sense, 5′-GGAGGCTTAATTACACATGTT-3′; antisense, 5′-TGATTTCAAGATGAATTGGAT-3′), and GAPDH (sense, 5′-CCAGTAGACTCCACTCACG-3′; antisense, 5′-CCTTCCACAATGCCAAAGTT-3′). Thermal cycling was performed as follows: Initial denaturation (95°C for 2 min), amplification (95°C for 1 min; 55°C for 1 min and 72°C for 1 min, 35 cycles), and final extension (72°C for 5 min). After amplification, the PCR products were resolved by agarose gel electrophoresis on 1% agarose gel containing 1X TAE (Biosesang, Inc.) and 1X Midori green advance (Nippon Genetics Europe GmbH), and visualized using a LAS-3000 device (FujiFilm). The band intensity was determined using ImageJ software (version 1.38; National Institutes of Health).

Lysates were extracted from cells using RIPA buffer (Biosesang, Inc.) and total protein was determined by BCA assay. A total amount of 30 µg of proteins were loaded. Proteins were separated using 10% SDS-PAGE and transferred to nitrocellulose membranes (Whatman; GE Healthcare Life Sciences). The membranes were blocked with 5% skimmed milk (Biosesang, Inc.) in TBS-Tween (0.1% Tween-20) for 1 h at 20–25°C. The membranes were incubated with primary antibodies for overnight at 4°C. Then, the membranes were incubated with HRP-conjugated secondary antibodies for 1 h at 20–25°C. The bands were developed using enhanced chemiluminescence solution (WEST-ZOL Plus, iNtRON Biotechnology) and visualized with the LAS-3000 device (FujiFilm). The band intensity was analyzed with ImageJ software (version 1.38; National Institutes of Health).

Cells were fixed with 4% paraformaldehyde for 15 min at 20–25°C and then permeabilized with 0.1% Triton X-100 for 10 min. Coverslips were incubated with blocking buffer (1% BSA in PBS) for 1 h at 20–25°C and then incubated with primary antibodies for overnight at 4°C. Coverslips were incubated with secondary antibodies for 1 h at 20–25°C, and images were visualized using a confocal microscopy (400× magnification; Carl Zeiss).

Data were expressed as mean ± SEM from ≥3 independent experiments. Statistical analyses were performed using GraphPad Prism 5.0 software (version 5.01; GraphPad Software, Inc.). The significance of each group was determined using one-way ANOVA followed by Turkey's post-hoc test. P<0.05 was considered to indicate a statistically significant difference.

Fig. 1 illustrates the mass spectra and proposed fragment ions of the four marker substances (left column), chromatograms of the standard solution (250 ng/ml) of each substance (middle column), and PPB solution that was diluted 1,000-fold (right column). PPB contained 0.193% chlorogenic acid, 0.097% catechin, 0.040% quercetin and 0.035% quercetin-3-O-glucoside.

To examine the anti-inflammatory role of PPB, the nitrite levels in LPS-stimulated BV2 cells was measured. LPS-induced upregulation of NO production was reduced by PPB treatment in a dose-dependent manner (Fig. 2A). The MTT assay showed that PPB did not have a significant effect on cell viability (Fig. 2B). To further examine the presence of cytotoxicity, PI staining was performed as previously described (38) and quantitatively showed that H₂O₂, but not PPB, induced cellular toxicity (Fig. 2C). These results suggested that PPB suppressed NO release in LPS-stimulated BV2 cells without inducing cellular toxicity.

LPS promoted iNOS and COX-2 protein expression, which was inhibited by PPB (Fig. 3A). PPB attenuated LPS-induced elevation of iNOS and COX-2 mRNA (Fig. 3B). Therefore, PPB inhibits iNOS and COX-2 expression in LPS-stimulated BV2 cells. To further investigate the anti-inflammatory role of PPB, TNF-α, IL-1β and IL-6 mRNA expression levels were investigated. PPB alleviated LPS-induced enhancement of proinflammatory cytokines (Fig. 3C). Notably, PPB significantly inhibited iNOS and COX-2 expression at a lower concentration (100 µg/ml) than the concentration that affected TNF-α, IL-1β, and IL-6 expression (200 µg/ml). The present results may indicate that PPB had preferential inhibitory effects on iNOS and COX-2 compared with TNF-α, IL-1β and IL-6 in LPS-stimulated BV2 cells.

NF-κB is a major transcription factor that regulates the expression of proinflammatory cytokines during inflammation. In response to inflammatory stimuli, IκB is rapidly phosphorylated and degraded, and subsequently dissociates from NF-κB, which translocates to the nucleus, leading to the transcription of proinflammatory mediators and cytokines (39,40). To clarify the underlying mechanism of the anti-inflammatory effects of PPB, the NF-κB signaling pathway was investigated. Normally, the level of IκBα expression is high, whereas that of p-IκBα expression is low. In the present study, LPS exposure decreased IκBα expression and enhanced p-IκBα expression (Fig. 4A). PPB ameliorated the LPS-induced alteration of IκBα and p-IκBα expression. To further examine the effect of PPB on p65 translocation, the expression of p65 was investigated by immunostaining. As expected, PPB suppressed LPS-induced p65 translocation into the nucleus (Fig. 4B). The present results indicated that PPB regulates proinflammatory cytokine expression by suppressing the NF-κB signaling pathway.

In microglia, LPS exposure enhances iNOS and TNF-α expression, which is accompanied by the activation of MAPK signaling pathways, including the ERK, JNK and p38 pathways (16). In addition, MAPK is involved in COX-2 and iNOS production in LPS-stimulated microglial cells (41). To further examine the underlying mechanism of PPB, MAPK activation was investigated. PPB repressed LPS-induced phosphorylation of ERK, JN and p38 (Fig. 5). Notably, PPB significantly suppressed the JNK phosphorylation at a lower concentration (20 µg/ml) than that for ERK or p38 (100 µg/ml). The present results indicated that PPB preferentially inhibited JNK in LPS-stimulated BV2 cells. Taken together, the present results demonstrated that PPB exerts an anti-inflammatory effect in LPS-stimulated BV2 cells through the inhibition of NF-κB and MAPK pathway activation.

To further evaluate the suppressive effect of PPB on neuroinflammation, NO production in primary cultured astrocytes was investigated. Consistent with the BV2 cell results, PPB also suppressed LPS-induced NO production without inducing cellular toxicity (Fig. 6A and B). In addition, immunostaining results indicated that PPB successfully alleviated LPS-induced p65 translocation into the nucleus (Fig. 6C). The present results indicated the anti-inflammatory role of PPB in glial cell activation by repressing repressed NO production and p65 translocation in primary astrocytes (Fig. S1).