Berberine modulates hyper-inflammation in mouse macrophages stimulated with polyinosinic-polycytidylic acid via calcium-CHOP/STAT pathway

Berberine is a well-known quaternary ammonium salt that is usually found in the roots of such plants as Phellodendron amurense and Coptis chinensis. However, the effects of berberine on double-stranded RNA (dsRNA)-induced macrophages have not been fully reported. In this study, we examined the anti-inflammatory effects of berberine on dsRNA [polyinosinic-polycytidylic acid; poly I:C]-induced macrophages. Levels of nitric oxide (NO), Prostaglandin E2 (PGE2), first apoptosis signal receptor (Fas; CD95), cytokines, intracellular calcium, phosphorylated I-kappa-B-alpha (IkB-α), phosphorylated p38 mitogen-activated protein kinase (MAPK), phosphorylated ERK1/2, phosphorylated signal transducer and activated transcription 3 (STAT3), and mRNA expression of inflammatory genes in poly I:C-induced RAW 264.7 mouse macrophages were evaluated. Berberine significantly inhibited the production of NO, PGE2, Fas, GM-CSF, LIF, LIX, RANTES, and MIP-2 as well as calcium release in poly I:C-induced RAW 264.7 cells at concentrations of up to 50 μM. Berberine also significantly inhibited the phosphorylation of p38 MAPK, ERK1/2, IkB-α, and STAT3 in poly I:C-induced RAW 264.7 cells. Additionally, berberine significantly decreased the mRNA expressions of Chop (GADD153), Stat1, Stat3, and Fas in poly I:C-induced RAW 264.7 cells. Taken together, berberine has anti-inflammatory properties related to its inhibition of NO, PGE2, Fas, GM-CSF, LIF, LIX, RANTES, and MIP-2 in dsRNA-induced macrophages via the endoplasmic reticulum stress-related calcium-CHOP/STAT pathway.

In the previous reports, we showed that bioactive compounds, such as wogonin, oroxylin A, quercetin, and baicalein, exert anti-inflammatory effects on RAW 264.7 cells induced by poly I:C 4-6 . However, the inhibitory effect of berberine on macrophages induced by viral infection has not yet been reported. In the current study, we found that berberine has anti-inflammatory effects in poly I:C-induced RAW 264.7 cells via the endoplasmic reticulum (ER) stress-related calcium-CHOP/STAT pathway.

Results
Effects of berberine on cell viability. In this study, the cell viability in RAW 264.7 incubated with poly I:C (50 µg/mL) for 24 h was 39.37 ± 7.82% of the normal value. In the addition, cell viabilities in poly I:Cinduced RAW 264.7 incubated with berberine at concentrations of 10, 25, and 50 µM were 164.17 ± 35.34%, 171.2 ± 42.97%, and 183.99 ± 41.35% of the control (poly I:C only) value ( Fig. 2A). Data mean that berberine alleviates the cytotoxic effects of poly I:C on RAW 264.7. With this result, we chose berberine concentrations of up to 50 μM for subsequent experiments.
Effects of berberine on NO production. Berberine significantly inhibited excessive production of NO in poly I:C-induced RAW 264.7 cells (Fig. 2B). The productions of NO in poly I:C-induced RAW 264.7 cells incubated with berberine at concentrations of 10, 25, and 50 µM for 24 h were 58.55 ± 1.37%, 33.49 ± 1.58%, and 46.9 ± 1.8% of the control (poly I:C only) value, respectively.  Effect of berberine on cell viability, NO production, and calcium in poly I:C-induced RAW 264.7 mouse macrophages. After 24 h incubation, cell viability (A) was evaluated by a modified MTT assay, and NO production (B) was measured by the Griess reaction assay. Calcium release (C) was measured with Fluo-4 calcium assay after 18 h incubation. Normal group (Nor) was treated with media only. Control group (Con) was treated with poly I:C (50 µg/mL) alone. Ba25 denote baicalein (25 µM). Values are the mean ± SD of three independent experiments. ### p < 0.001 vs. Nor; **p < 0.01 vs. Con; ***p < 0.001 vs. Con. Effect of berberine on cytokine production. Berberine decreased cytokine production in poly I:Cinduced RAW 264.7 cells (Fig. 3) significantly. We found the following:

Scientific
GM-CSF production. In RAW 264.7 incubated with media only, poly I:C only, berberine (10 µM) plus poly I:C, berberine ( These data indicated that berberine might alleviate hyper-inflammatory reaction leading to excessive production of chemokines and growth factors in macrophages stimulated by poly I:C. Figure 4 shows the data on the effect of berberine on mRNA expression of Stat1, Stat3, Chop, and Fas. These results mean that berberine inhibits the inflammatory reaction in poly I:C-induced macrophages via the ER stress-related CHOP pathway. The effects were as follows: the mRNA expression of Stat1 in poly I:C-induced RAW 264.7 incubated with berberine at concentrations of 10, 25, and 50 µM for 18 h was 67.74 ± 10.6%, 62.61 ± 7.5%, and 56.98 ± 4.81% of the group treated with poly I:C only, respectively; the mRNA expression of Stat3 was 11.05 ± 0.99%, 12.41 ± 1.1%, and 7.06 ± 0.7, respectively; the mRNA expression of Chop was 32.96 ± 5.56%, 57.71 ± 6.24%, and 51.46 ± 5.9, respectively; the mRNA expression of Fas was 48.44 ± 36.12%, 34.87 ± 5.11%, and 15.95 ± 7.42, respectively.

Effect of berberine on levels of Fas and PGE2. Data represented that berberine significantly decreases
the levels of Fas and PGE2 in poly I:C-stimulated RAW 264.7 cells (Fig. 6). Concretely, the level of Fas in poly I:Cstimulated RAW 264.7 cells incubated with berberine at concentrations of 10, 25, and 50 µM was 12.8 ± 3.39%, 10.92 ± 0.42%, and 10.6 ± 0.35% of the control group treated with poly I:C (50 µg/mL) only, respectively; the level of PGE2 was 17.58 ± 4.2%, 16.69 ± 3.11%, and 12.41 ± 1.52%. These data mean that berberine modulates the inflammatory reaction in poly I:C-stimulated mouse macrophages via inhibiting the production of PGE2 and Fas.

Discussion
Berberine is a well-known quaternary ammonium salt that is usually found in the roots of such plants as Phellodendron amurense (Amur corktree), Berberis vulgaris (barberry), Berberis aristata (tree turmeric), Mahonia aquifolium (Oregon grape), Hydrastis Canadensis (goldenseal), Xanthorhiza simplicissima (yellow root), Coptis chinensis (Chinese goldthread), Tinospora cordifolia, Argemone mexicana (prickly poppy), and Eschscholzia californica (Californian poppy) 7 . For pharmacological efficacy of berberine, Cicero and Baggioni reported that berberine has anti-inflammatory, antioxidant, neuroprotective, and cardiovascular protective effects, which might be used to manage chronic cardiometabolic disorders 7     www.nature.com/scientificreports/ and virus. Sometimes an unregulated infection might cause a syndrome of excessive cytokine production, i.e., "cytokine storm (hypercytokinemia)" [10][11][12][13][14] . Viral infection as well as bacterial infection might evoke massive production of pro-inflammatory mediators. In spite of modern developments in drugs and vaccines, there are still many intractable disorders caused by pathogens. Thus, it is reasonable to search for a material to modulate viral inflammation and virus-induced hypercytokinemia.
In these studies, poly I:C was used to bring about inflammatory reactions in mouse macrophages, which are a virus-induced cellular inflammation model. It is well known that dsRNA is produced in immune cell through viral replication and elicit cytotoxic responses 22 . In the current study, our data show that berberine alleviates the cytotoxic effects of poly I:C on RAW 264.7. Thus, berberine might be an inhibitor of apoptosis in macrophages stimulated by poly I:C.
In 2001, Alexopoulou et al. reported that mammalian toll-like receptor 3 could recognize dsRNA and gradually induce cytokine production via NF-kB and MAPK activation 23 . This study shows that berberine significantly inhibited the production of NO, PGE2, Fas, GM-CSF, LIF, LIX, RANTES, and MIP-2 as well as calcium release and phosphorylation of p38 MAPK, ERK1/2, IkB-α, and STAT3 in poly I:C-stimulated RAW 264.7 cells. These results suggest that the anti-inflammatory effect of berberine in poly I:C-stimulated RAW 264.7 cells might involve the calcium signaling pathway, including activation of MAPK, NF-kB, and STAT3.
It is not yet fully explained how viral infection causes an ER stress with intracellular calcium release and CHOP expression. In relation to CHOP signaling pathway in stressed cells, Wang and Ron reported in 1996 that CHOP is phosphorylated and activated by p38 MAPK in stressed cells under conditions of stress 24 . In 2006, Endo et al. reported that the start of LPS-induced inflammation cascade in lung might be triggered through ER stress-CHOP pathway including the activation p38 MAPK and NF-kB 25 . With CHOP-induced apoptosis in macrophages, Gotoh et al. reported in 2004 that ER stress-CHOP pathway is involved in NO-mediated apoptosis in LPS plus interferon-γ-stimulated macrophages 26 . Although NO is known to be necessary for various physiological processes, it means that NO might be a crucial key factor in ER stress-CHOP pathway in virusstimulated macrophages. Moreover, Mory M. reported meaningfully in 2007 that NO depletes ER Ca2 + , which consequently causes ER stress and leads apoptosis in macrophages 27 . An increase in cytosolic calcium is well known to activate CAMKII, which induces the apoptotic signaling cascade in macrophages including Fas induction and activation of STAT1 [28][29] .
Our experimental results show that berberine inhibits the overexpression of Chop, Stat1, Stat3, and Fas in poly I:C-stimulated RAW 264.7 cells. Hence berberine may be able to alleviate ER stress induced by dsRNA via the calcium-STAT pathway including activation of MAPK and NF-kB. But this study could not evaluate whether calcium signaling causes CAMKII activation in poly I:C-stimulated RAW 264.7 cells or not. These studies also could not verify exactly the effect of berberine on Type I interferons signaling in poly I:C-stimulated RAW 264.7 cells. It is meaningful that intracellular calcium is an important signaling molecule of ER stress and increases with Chop overexpression in poly I:C-stimulated RAW 264.7 cells. Cell viability. RAW 264.7 cells were obtained from the Korea Cell Line Bank (Seoul, Korea). Cells were cultured in DMEM supplemented with 10% FBS containing 100 U/mL of penicillin and 100 µg/mL of streptomycin at 37 °C in a 5% CO 2 humidified incubator. RAW 264.7 were cultured for 24 h, and cell viability was evaluated with the tetrazolium-based colorimetric assay (MTT) according to the previous study with a microplate reader (Bio-Rad) 4-6 . NO concentration. After 24 h incubation, NO concentration in culture medium was measured by the Griess reaction as in previous studies with a microplate reader (Bio-Rad) 4-6 . Intracellular calcium assay. Fluo-4 AM is a well-known fluorescent Ca 2+ indicator used for the in-cell measurement of calcium signaling. After 18 h of treatment in 96-well plates, the intracellular calcium signaling from each well containing RAW 264.7 (1 × 10 5 cells/well) was identified using Fluo-4 NW Calcium Assay Kits (Thermo Fisher Scientific) following the protocol of our previous study [4][5][6] . Specifically, after incubating the cells with poly I:C and/or berberine for 18 h at 37 °C, we measured the intracellular calcium level using Fluo-4 assay with a spectrofluorometer (Dynex, West Sussex, UK) with excitation and emission filters of 485 nm and 535 nm, respectively.
Real-Time RT-PCR assay. In 6-well plates, RAW 264.7 (3 × 10 5 cells/well) were incubated in each well with or without berberine in poly I:C for 18 h. After 18 h incubation, total RNA of each well was isolated using NucleoSpin RNA kit (Macherey-Nagel, Duren, Germany). Then, cDNA of the RNA samples was produced using iScript cDNA Synthesis kit (Bio-Rad). The mRNA expression of Stat1, Stat3, Chop, and Fas were evaluated with real-time RT-PCR using an Experion Automatic Electrophoresis System (Bio-Rad) and Bio-Rad CFX 96 Real-Time PCR Detection System (Bio-Rad) 6 . Tbp was used for RNA normalization. The sequence of each primer set is showed in Table 1.
Flow cytometric analysis. In 6-well plates, RAW 264.7 (3 × 10 5 cells/well) were incubated in each well with or without berberine in poly I:C for 24 h. After 24 h incubation, cells were harvested and washed with Flow Cytometry Staining Buffer (Thermo Fisher Scientific). Cells were stained with antibodies for phospho-p38 MAPK, phospho-ERK1/2, phospho-STAT3, phospho-IkB-α, and Fas according to the manufacture's protocol as described previously 21 . Prior to antibody staining, cells were fixed and permeabilized using Fix Buffer I (Thermo Fisher Scientific) and Perm Buffer III (Thermo Fisher Scientific), respectively. Stained cells were analyzed on the   Stat3  GTC TGC AGAGT TCA AGC ACCT  TCC TCA GTC ACG ATC AAG GAG   Chop  CCA CCA CAC CTG AAA GCA G  TCC TCA TAC CAG GCT TCC A   Fas  CGC TGT TTT CCC TTG CTG  CCT TGA GTA TGA ACT CTT AAC TGT GAG   Tbp  GGG GAG CTG TGA TGT GAA GT  CCA GGA AAT AAT TCT GGC  www.nature.com/scientificreports/ Attune NxT flow cytometer (Thermo Fisher Scientific). Data were obtained from the mean fluorescent intensities of samples. Details for startup, proper calibration and operation of the Attune can be found in the Attune User Guide (https:// assets. therm ofish er. com/ TFS-Assets/ LSG/ manua ls/ 10002 4235_ Attun eNxT_ HW_ UG. pdf). A serial gating strategy used forward scatter versus side scatter plots and the target antibody expression plots. Unstained cells were used as the negative controls for gating. Mouse IgG1 kappa Isotype Control was used to confirm the specificity of phospho-p38 MAPK Antibody and Fas Antibody. Mouse IgG2b kappa Isotype Control was used to confirm the specificity of phospho-ERK1/2 Antibody, phospho-STAT3 Antibody, and phospho-IkB-α Antibody. For analysis of raw data, we used Attune NxT software (Thermo Fisher Scientific). Baicalein (25 µM) was used as a positive control.
Prostaglandin E2 assay. In 96-well plates, RAW 264.7 (1 × 10 4 cells/well) were incubated in each well with or without berberine in poly I:C for 24 h. After 24 h incubation, the production of PGE2 in each well containing RAW 264.7 cells was evaluated using Prostaglandin E 2 parameter assay kit (R&D Systems) according to the manufacture's protocol. The absorbance of the end product was measured at a wavelength of 450 nm using a microplate reader (Bio-Rad) 30 .

Statistical analysis.
We present data using the mean ± SD of three independent experiments. Significant differences were examined using a one-way analysis of variance test, followed by Tukey's multiple-comparison test with SPSS 11.0 software (SPSS, Inc., Chicago, IL, USA). In all cases, a p < 0.05 was considered significant.