Male C57BL/6J mice (7 weeks old) were obtained from Orient Bio (Seongnam, Korea), and acclimated for a week before experiments. Animals were housed under specific pathogen-free conditions in an air conditioned room at 23±2℃. Food and water were supplied ad libitum. All animal procedures were performed in accordance with the Guide for the Care and Use of Laboratory Animals published by the US National Institute of Health (NIH Publications, No 8523, revised 2011), and were approved by the Animal Care and Use Committee of Gachon University.

Dulbecco's modified Eagle's medium (DMEM), fetal bovine serum (FBS), penicillin, and streptomycin were obtained from GIBCO (Grand Island, NY). LPS (Escherichia Coli 055:B5) and berberine (purity >98%) were from Sigma (St Louis, MO). TNF-α, IL-1β, and IL-6 ELISA kits were from Enzo Life Sciences (New York, NY). Antibodies against phospho-AMPK-α-Thr¹⁷², AMPK, phospho-JNK (Thr¹⁸³/Tyr¹⁸⁵), and phospho-p38 MAPK (Thr¹⁸⁰/Tyr¹⁸²) were from Cell Signaling Technology (Danvers, MA). Antibodies against IgG, β-actin, ATF-3, phospho-ERK1/2 (Tyr²⁰⁴), JNK, p38 and ERK1/2 were from Santa Cruz Biotechnology (Santa Cruz, CA).

RAW264.7 murine macrophage cell lines were obtained from the Korean Cell Line Bank and seeded in 12-well plates at a density of 5×10⁵ cells/well in DMEM supplemented with 2 mM L-glutamine, 100 U/ml penicillin, 100 µg/ml streptomycin, and 10% (v/v) heat-inactivated fetal bovine serum in a humidified 5% CO₂/95% air at 37℃. Cells were treated with berberine at the indicated concentrations for 20 hr and then with LPS (10 ng/ml) for 4 hr (ELISA and RT-PCR) and for 30 min (western blot of phosphoprotein). For time dependency of berberine action, cells were treated with 20 µM berberine at the indicated times (0~20 hr), and then with LPS (10 ng/ml) for 4 hr. Final DMSO concentration was 0.5%, which itself had little effect.

Total RNA was isolated from cells using the easy-BLUE Total RNA extraction kit (iNtRON Inc. Korea). Reverse transcription of total RNA (1 µg) was performed using AccuPower RT PreMix (Bioneer Inc. Korea). The sequences of the PCR primers used to amplify TNF-α, IL-6, ATF-3, and GAPDH (the internal standard) were: TNF-α (F)-ATG AGC ACA GAA AGC ATG ATC, TNF-α (R)-TAC AGG CTT GTC ACT CGA ATT; IL-6 (F)-GAG GAT ACC ACT CCC AAC AGA CC, IL-6 (R)-GAG GAT ACC ACT CCC AAC AGA CC; IL-1β (F)-CAG GAT CAG GAC ATG AGC ACC, IL-1β (R)-CTC TGC AGA CTC AAA CTC CAC; ATF-3 (F)-TTG CTA ACC TGA CAC CCT TTG, ATF-3 (R)-CGG TGC AGG TTG AGC ATG TA; GAPDH (F)-TTC ACC ACC ATG GAG AAG GC, GAPDH (R)-GGC ATG GAC TGT GGT CAT GA. Reverse transcription-PCR was performed over 35 amplification cycles (denaturation at 94℃ for 1 min, annealing at 60℃ for 1 min, and extension at 72℃ for 1 min) followed by a 10 min extension at 72℃. PCR reaction mixtures were electrophoresed on 1.5% agarose gel and visualized using Gel red (Elpis Biotech, Seoul) staining under UV light. Relative mRNA abundances were normalized vs. GAPDH.

Cells were lysed in lysis buffer (50 mM HEPES, pH 7.0, 250 mM NaCl, 5 mM EDTA, 0.1% Nonidet P-40, 1 mM phenylmethylsulfonyl fluoride, 0.5 mM dithiothreitol, 5 mM Na fluoride, 0.5 mM Na orthovanadate, and 5 µg/ml each of leupeptin and aprotinin) for 10 min at 4℃. Following centrifugation at 12,000 rpm and denaturation, 50 µg aliquots of total proteins were loaded into an 8% sodium dodecyl sulfatepolyacrylamide gel, and then transferred to nitrocellulose membranes (Amersham Biosciences, Piscataway, NJ). Blocked membranes were incubated with anti-ATF-3, anti-AMPK, anti-phospho-AMPKα-Thr¹⁷², anti-phospho-ERK-Tyr²⁰⁴, anti-phospho-p38 MAPK-Thr¹⁸⁰/Tyr¹⁸², anti-p38 MAPK, anti-JNK, anti-phospho-JNK-Thr¹⁸³/Tyr¹⁸⁵, and anti-actin antibodies and then blotted using secondary antibodies conjugated with horseradish peroxidase (HRP). Immunoreactive bands were visualized by enhanced chemiluminescence (Amersham Life Science, Buckinghamshire, UK), and band signals were quantified using UN-SCAN-IT gel 5.1 software (Silk Scientific, Inc., Orem, Utah). Protein concentrations were determined using the Bio-Rad protein assay reagent, according to the manufacturer's instructions.

Concentrations of TNF-α, IL-6 and IL-1β in supernatants of cultured media were measured using ELISA kits. Briefly, supernatants were incubated in coated 96-well plates at room temperature for 2 hr. Plates were washed, and then incubated with detection antibody for 2 hr. After rewashing the plates, conjugate was added for 30 min, followed by the addition of substrate solution. The reaction was stopped with 1N H₂SO₄ and optical densities were measured at 450 nm using a microplate reader (Perkin Elmer VictorX4, Waltham, MA).

Raw 264.7 cells were maintained in DMEM medium containing 10% fetal bovine serum, and 5×10⁵ cells/well in 12-well plates were transfected with α1-AMPK siRNA, ATF-3 siRNA, or negative control siRNA (Santa Cruz, CA) (100 nM/well) using 6 µl of Lipofectamine 2000 reagent (Invitrogen, Carlsbad, CA) in OPTI MEM medium for 6 hr. Medium was then replaced with DMEM containing 10% FBS and 18 hr later, cells were treated with berberine (10 and 20 µM) for 20 hr and then with 10 ng/ml LPS for 30 min (western blot of phosphoprotein) and for 4 hr (ELISA). Efficiencies of ATF-3 and AMPK knockdowns were determined by Western blotting and by RT-PCR.

A subcellular proteome extraction kit (Calbiochem, Inc. San Diego, CA) was used to extract cytosolic and nuclear fractions of RAW264.7 cells. Briefly, cells were incubated with 20 µM berberine for 20 hrs, exposed to LPS (10 ng/ml) for 4 hr, washed twice with PBS, resuspended in 150 µl of ice-cold Extraction I containing 0.75 µl of protease inhibitor mixture, and then incubated for 10 min at 4℃ with gentle agitation. The suspensions obtained were centrifuged at 1000×g for 10 min at 4℃. Supernatants were referred to as cytosolic fraction. Pellets were resuspended in 150 µl of ice-cold Extraction II (for membrane fractions) containing 0.75 µl of protease inhibitor mixture, and incubated for 30 min at 4℃. Mixtures were then centrifuged at 6000×g at 4℃, and the supernatants obtained were referred to as membrane fraction. Remaining pellets were re-suspended in ice-cold Extraction buffer IIII with 0.75 µl of protease inhibitor mixture and incubate for 10 min at 4℃ under gentle agitation. Mixtures were then centrifuged at 8800×g at 4℃ and supernatants were transferred to the tube as a nuclear fraction.

C57BL/6 mice were orally administered vehicle (saline, 200 µl/20 g, n=7) or berberine (100 or 200 mg/kg/day, n=8 each dose) for 3 days twice daily. On day 4, LPS (20 mg/kg, i.p.) was injected 3 hours after last berberine administration. One hour later, mice were anesthetized with diethyl ether and whole blood was collected by cardiac puncture. Plasma was obtained by whole blood centrifugation and stored at -20℃ until assayed. Lungs and spleens were isolated and immediately frozen, and stored in liquid nitrogen until assayed. All tissues were homogenized and then lysed using a Prep-sol (iNtRON, Korea) prior to analysis.

Results are expressed as means±SEMs. Statistical significance was determined by one-way analysis of variance (ANOVA) followed by Tukey's multiple-comparison test. Statistical significance was accepted for p values < 0.05.

RAW264.7 macrophages were pretreated with various concentrations of berberine for 20 hr and then stimulated with 10 ng/ml LPS for 4 hr. Although LPS markedly increased proinflammatory cytokine production as compared with control cells, berberine (10~100 µM) concentration-dependently suppressed this production (Fig. 1A). Likewise, the mRNA expressions of proinflammatory cytokines were concentration-dependently reduced by berberine pretreatment (Fig. 1B). Suppression of LPS-stimulated TNF-α and IL-6 production by berberine (20 µM) was also time dependent, maximum effects being achieved at 20 hr incubation (Fig. 1C).

We next examined the involvement of ATF-3 in the berberine-mediated suppression of LPS-induced proinf lammatory cytokines. Along with the suppression of proinflammatory cytokine production, berberine concentration-dependently increased ATF-3 mRNA expression (2.1 to 4.9 fold vs. controls, and 1.4 to 2.3 fold vs. LPS alone). Similarly, as has been previously reported, ATF-3 protein levels were increased by LPS [1518], and were further enhanced by berberine pretreatment (3.4 to 4.1 fold vs. controls, and 2.4 to 2.9 fold vs. LPS alone) (Fig. 1D).

Since ATF-3 is known to translocate into nucleus upon stimulation, we further examined the effects of berberine on nuclear translocation of ATF-3 after subcellular fractionation. As shown in Fig. 1E, β-tubulin, a cytosolic marker was detected in the cytosolic fraction, whereas the expression of lamin B1 was restricted in the nucleus fraction, indicating that subcellular fractionation was successful. On the other hand, ATF-3 was primarily detected in nucleus fraction, and berberine (20 µM) consistently super-induced protein levels of ATF-3.

To elucidate the role of ATF-3 induction by berberine, we examined the effects of ATF-3 knockdown on the anti-inflammatory effects of berberine. Transfection of ATF-3 siRNA decreased ATF-3 mRNA and protein expressions by ~60% as compared with control siRNA treatment (Fig. 2A). ATF-3 siRNA transfection completely blocked the inhibitory effects of berberine on proinflammatory cytokine production (Fig. 2B), suggesting that ATF-3 is critical for the suppressive actions of berberine on LPS-induced proinflammatory cytokine production in macrophages.

MAPK signaling is known to participate in the inflammatory effect of LPS [20]. Thus, we examined the effects of berberine on MAPK phosphorylation and the effects of ATF-3 siRNA transfection on the effects of berberine on MAPKs. As shown in Fig. 3A, berberine suppressed the phosphorylations of JNK, ERK, and p38 MAPK, and these suppressions were completely blocked by ATF-3 knockdown. On the other hand, Thr¹⁷² phosphorylation of AMPKα was little affected by ATF-3 siRNA treatment (Fig. 3B). These results suggest berberine induces ATF-3 induction and suppresses MAPK activation, and that AMPK activation by berberine maybe an upstream mediator of ATF-3 induction.

To examine the signaling pathway of ATF-3 induction by berberine further, we checked the effects of AMPK knockdown on berberine-induced ATF-3 induction. As shown in Fig. 4A, AMPK siRNA reduced the mRNA and protein expressions of AMPK to 36% and 46% of control levels, respectively. As was expected, AMPK siRNA blunted the effects of berberine on proinflammatory cytokine production (Fig. 4B). Likewise, ATF-3 induction and MAPK deactivation by berberine were also blocked by AMPK knockdown (Fig. 4C). In view of the observation that ATF-3 knockdown had little effect on berberine-induced AMPK phosphorylation, these results suggest sequential signaling pathways in the actions of berberine, i.e., AMPK phosphorylation → ATF-3 induction → inactivation of MAPK.

The involvement of ATF-3 induction in the anti-inflammatory action of berberine was further investigated in vivo using a LPS-induced septic shock mouse model. LPS injection significantly increased the plasma levels of proinflammatory cytokines (Fig. 5A); its effects peaked at 1 hr post-injection and then declined rapidly (results not shown). However, oral administration of berberine (100 or 200 mg/kg, bid) for 3 days significantly reduced the plasma levels of proinflammatory cytokines (Fig. 5A). Consistent with our in vitro results, ATF-3 levels were elevated in lungs and spleens (both major contributors to macrophage production) of berberine-treated animals, and phosphorylated AMPK levels were elevated and the protein levels of inflammatory markers were diminished (Figs. 5B and C). Taken together, it appears that AMPK-mediated ATF-3 induction is an important component of the anti-inflammatory effects of berberine in murine macrophages.