ble at ScienceDirect

Journal of Pharmacological Sciences 127 (2015) 203e210

Contents lists availa

HOSTED BY

Journal of Pharmacological Sciences

journal homepage: www.elsevier .com/locate/ jphs

Full paper

Salvianolic acid B protects against acetaminophen hepatotoxicity byinducing Nrf2 and phase II detoxification gene expression viaactivation of the PI3K and PKC signaling pathways

Musen Lin a, b, 1, Xiaohan Zhai a, c, 1, Guangzhi Wang d, Xiaofeng Tian d, Dongyan Gao a,Lei Shi a, Hang Wu a, Qing Fan b, Jinyong Peng a, Kexin Liu a, Jihong Yao a, e, *

a Department of Pharmacology, Dalian Medical University, Dalian 116044, Chinab Department of Pharmacy, Second Affiliated Hospital of Dalian Medical University, Dalian 116023, Chinac Department of Pharmacy, First Affiliated Hospital of Dalian Medical University, Dalian 116023, Chinad Department of General Surgery, Second Affiliated Hospital of Dalian Medical University, Dalian 116023, Chinae Research Institute of Integrated Traditional and Western Medicine of Dalian Medical University, Dalian 116044, China

a r t i c l e i n f o

Article history:Received 22 July 2014Received in revised form8 December 2014Accepted 17 December 2014Available online 24 December 2014

Keywords:Acetaminophen hepatotoxicityNrf2Phase II enzymePI3KPKCSalvianolic acid B (SalB)

Abbreviations: ALT, Alanine aminotransferase;antioxidant response elements; AST, aspartate aminosulfoximine; ERK, extracellular signal-related proteincysteine ligase catalytic subunit; H&E, hematoxylingenase-1; JNK, c-Jun N-terminal kinase; Keap1, Kelch-lMAPKs, mitogen-activated protein kinases; MEM, MNAPQI, N-acetyl-p-benzoquinoneimine; Nrf2, nucleafactor-2; PI3K, phosphatidylinositol-3-kinase; PKC, proxygen species; SalB, Salvianolic acid B; Znpp, zinc p* Corresponding author. Department of Pharmacolo

Dalian 116044, China. Tel.: þ86 (0) 411 86110410; faxE-mail address: yaojihong65@hotmail.com (J. Yao)

Peer review under responsibility of Japanese Pharm1 The first two authors contributed equally to the w

http://dx.doi.org/10.1016/j.jphs.2014.12.0101347-8613/© 2015 Japanese Pharmacological Society.

a b s t r a c t

Acetaminophen (APAP) is used drugs worldwide for treating pain and fever. However, APAP overdose isthe principal cause of acute liver failure in Western countries. Salvianolic acid B (SalB), a major water-soluble compound extracted from Radix Salvia miltiorrhiza, has well-known antioxidant and anti-inflammatory actions. We aimed to evaluate the ability of SalB to protect against APAP-induced acutehepatotoxicity by inducing nuclear factor-erythroid-2-related factor 2 (Nrf2) expression. SalB pretreat-ment ameliorated acute liver injury caused by APAP, as indicated by blood aspartate transaminase levelsand histological findings. Moreover, SalB pretreatment increased the expression of Nrf2, Hemeoxygenase-1 (HO-1) and glutamate-L-cysteine ligase catalytic subunit (GCLC). Furthermore, the HO-1inhibitor zinc protoporphyrin and the GCLC inhibitor buthionine sulfoximine reversed the protectiveeffect of SalB. Additionally, siRNA-mediated depletion of Nrf2 reduced the induction of HO-1 and GCLC bySalB, and SalB pretreatment activated the phosphatidylinositol-3-kinase (PI3K) and protein kinase C(PKC) signaling pathways. Both inhibitors (PI3K and PKC) blocked the protective effect of SalB againstAPAP-induced cell death, abolishing the SalB-induced Nrf2 activation and decreasing HO-1 and GCLCexpression. These results indicated that SalB induces Nrf2, HO-1 and GCLC expression via activation ofthe PI3K and PKC pathways, thereby protecting against APAP-induced liver injury.© 2015 Japanese Pharmacological Society. Production and hosting by Elsevier B.V. All rights reserved.

APAP, acetaminophen; ARE,transferase; BSO, Buthioninekinases; GCLC, glutamate-L-

and eosin; HO-1, Heme oxy-ike ECH-associated protein 1;inimum Essential Medium;r factor erythroid-2-relatedotein kinase C; ROS, reactiverotoporphyrin.gy, Dalian Medical University,: þ86 411 86110408..acological Society.ork.

Production and hosting by Elsevie

1. Introduction

Acute liver failure is a global disease that can be attributed tovarious liver lesions. The major causes of acute liver failure are viralinfection, drug exposure, accidental food poisoning and radiationdamage (1,2). Acetaminophen (APAP) is widely used as an analgesicand antipyretic agent; it is also known to induce liver injury, ac-counting for the most common form of acute liver failure (3).Toxicity is initiated by cytochrome P450-mediated metabolism ofAPAP to the reactive metabolite N-acetyl-p-benzoquinoneimine(NAPQI). The binding of NAPQI to cellular macromolecules inducescellular oxidant stress and DNA damage, leading to severe cen-trilobular hepatotoxicity and acute liver failure (4,5). Therefore,designing therapies that inhibit oxidative stress and scavengeelectrophiles is expected to be one of the most important strategiesfor treating APAP-induced liver injury.

r B.V. All rights reserved.

M. Lin et al. / Journal of Pharmacological Sciences 127 (2015) 203e210204

Nuclear factor-erythroid-2-related factor 2 (Nrf2) plays a centralrole in the induction of antioxidant enzymes via binding to anti-oxidant response elements (AREs). This regulation can be groupedinto several categories according to the induction of phase IIdetoxifying enzymes, antioxidant genes, scavenger receptor andtransporters (6). Furthermore, among the antioxidant and detoxi-fication enzymes, Heme oxygenase-1 (HO-1) and glutamate-L-cysteine ligase catalytic subunit (GCLC) are the most important.Recent studies have shown that activation of Nrf2 plays animportant role in acute liver injury both in vitro and in vivo (7,8). Todate, multiple signaling kinases have been reported to regulateNrf2, including mitogen-activated protein kinase (MAPK), proteinkinase C (PKC) and phosphatidylinositol-3-kinase (PI3K) (9,10).

Salvianolic acid B (SalB) is the most abundant and bioactivecomponent of salvianolic acid extracted from Radix Salvia miltior-rhiza, which is officially listed in the Chinese Pharmacopoeia (11).Previous research has demonstrated that SalB inhibits lipid per-oxidation and scavenges superoxide anions and hydroxyl radicalsboth in vitro and in vivo (12,13). There is evidence that treatmentwith SalB reduces brain damage and improves motor function aftercerebral ischemia in rats (14,15). In addition, SalB has been reportedto protect against liver fibrosis in patients and animals, as well asinhibit CCl4-induced hepatic fibrosis; these beneficial effects areassociated with the regulation of NF-kB/I(k)B(a) signaling (16,17).

The above observations regarding the biological activities ofSalB compelled us to investigate whether SalB can attenuate APAP-induced liver injury. The goals of this study were to investigate theexpression of Nrf2, HO-1 and GCLC in APAP-induced liver damage,determine whether SalB up-regulates the expression of Nrf2, HO-1and GCLC, and further elucidate whether PI3K and PKC are involvedin the regulation of Nrf2.

2. Materials and methods

2.1. Chemicals

SalB with a purity >98% was obtained from the Chinese NationalInstitute for the Control of Pharmaceutical and Biological Products(Beijing, China). Acetaminophen and dimethyl sulfoxide werepurchased from SigmaeAldrich (St. Louis, MO, USA). Eagle's mini-mum essential medium (MEM) and trypsin were purchased fromGibco (Life Technologies, Carlsbad, CA, USA). RNase inhibitor andDNA oligo were purchased from TaKaRa (Shiga, Japan). Inhibitors ofPI3K/AKT, ERK1/2, p38, JNK, PKC, PKCd and GCLC (LY294002,PD98059, SB203580, SP600125 and GF109203X), as well asbuthionine sulfoximine (BSO), were acquired from Santa CruzBiotechnology (Santa Cruz, CA, USA). Zinc protoporphyrin (ZnPP)was purchased from Alfa Aesar (Ward Hill, MA, USA).

2.2. Animals

Male Kunming mice, weighing 18e22 g, were provided by theAnimal Center of Dalian Medical University (Dalian, China). Theanimals were fed standard chow and water, housed in plasticcages and maintained at 22 ± 2 �C with 50e60% relative humidityand a 12-h lightedark cycle throughout the experiment. Allprocedures were conducted according to the institutional animalcare guidelines and approved by the Institutional EthicsCommittee.

Animals were divided into five groups of ten animals per group.Group 1 received vehicle (double-distilled water) only and servedas the normal control. Group 2 only received 50 mg/kg SalB (dis-solved in double-distilled water) once a day for three days. Group 3animals were intragastrically administered a single dose of APAP at300 mg/kg (dissolved in double-distilled water at 37 �C) and then

kept as the experimental model. Groups 4 and 5 received SalB at 25and 50 mg/kg, respectively, once a day for three consecutive days.Two hours after the last SalB pretreatment, the animals received anintragastric dose of APAP. The experimental animals were sacri-ficed, and blood samples were collected from the abdominal aorta24 h after APAP administration. The serum was separated bycentrifugation at 3000 rpm/min for 30 min at 4 �C. The liver wascollected from each mouse for biochemical, Western blot, andhistopathological analysis.

2.3. Cell culture

Well-differentiated human hepatoma cell line HepG2 has beenused to identify factors which regulate hepatic gene expressionduring the host response to tissue injury (18). And in Guerrero JAet.al study they found that using anoikis-hepatocytes from primaryculture and H2O2-treated HepG2 cells confirmed the presence of asimilar complex (19). So in this study, we used HepG2 cells as ourmodel cells to explore the mechanism of SalB in APAP-inducedinjury.

HepG2 cells were cultured inMEM supplementedwith 10% (v/v)fetal bovine serum at 37 �C in a 5% CO2 humidified incubator. Cellsseeded at a density of 1�105 cells per well were grown for 24 h andthen treated with SalB at various concentrations (0.5, 2, 8 mmol/L).Each experimental protocol is described in the correspondingfigure legends.

2.4. Biochemical index assay

Serum alanine aminotransferase (ALT) and aspartate amino-transferase (AST) represent the liver function along with diseaseand are the most commonly used biochemical markers. Mouseserum ALT and AST levels in liver tissues were measured using anassay kit (Nanjing Jiancheng Corp., Nanjing, China) according to themanufacturer's recommendations.

2.5. Histopathological evaluation

Fresh hepatic samples were fixed in 10% neutral formalin for24e48 h before being embedded in paraffin. After embedding, thesamples were cut into 5-mm sections and mounted onto slides.Duplicate sections were stained with hematoxylin and eosin (H&E)for pathological evaluation.

2.6. Western blot analysis

Cytoplasmic and nuclear proteins were extracted from HepG2cells or liver tissue using a protein extraction kit (KeyGen Biotech,Nanjing, China). Extracted proteins were separated using SDS-PAGEand immunoblotted with the indicated primary antibodies. Theblots were then incubated with secondary antibodies. The mem-branes were exposed to enhanced chemiluminescence-plus re-agents (Beyotime Institute of Biotechnology, Hangzhou, China),documented using a BioSpectrum-410 multispectral imaging sys-tem with a Chemi HR camera 410 and analyzed using the Gel-ProAnalyzer, version 4.0 (Media Cybernetics, Rockville, MD, USA).

2.7. RNA isolation and RT-PCR

Total RNAwas extracted from HepG2 cells using TRIZOL reagent(Invitrogen, Carlsbad, CA, USA). Reverse transcription was per-formed using the TaKaRa RNA PCR Kit (AMV) Version 3.0 (TaKaRa,Shiga, Japan). For PCR analysis, the primers and their sequenceswere reported in our previous study (20). The co-amplified PCRproducts were separated on a 1.5% agarose gel using

M. Lin et al. / Journal of Pharmacological Sciences 127 (2015) 203e210 205

electrophoresis. A BioSpectrum-410 multispectral imaging systemwith a Chemi HR camera 410 (UVP Inc., Upland, CA, USA) was usedto analyze the intensity of the DNA bands.

2.8. Transient transfection and RNA interference

HepG2 cells were transfected with specific siRNA at a concen-tration of 100 nM using Lipofectamine™ 2000 reagent (Invi-trogen) according to the manufacturer's instructions. Human Nrf2siRNA was obtained from Genepharma Technology (Shanghai,China). After siRNA transfection, cells were incubated with freshculture medium containing 8 mM SalB. The protein extracts madefrom the siRNA transfection experiments were analyzed todetermine the expression of HO-1, GCLC, and Nrf2 via Westernblotting.

2.9. Statistical analysis

Data are expressed as the mean ± SD. The significance of thedifferences among the mean values was evaluated using one-way

Fig. 1. Effect of SalB on APAP-induced liver injury. Mice pretreated with SalB (25 or 50 mg/kwere sacrificed 24 h after APAP administration. (A) The activity of ALT and AST from serumdouble-distilled water (DDS) were tested controls. (b) Mice treated with 50 mg/kg SalB. (c)liver injury. Values are expressed as means ± SD, n ¼ 10. **p < 0.01 vs. the control group;

analysis of variance. The Student-Newman-Keuls (SNK)/least sig-nificant difference (LSD) test was applied to make pairwise com-parisons among the means. Differences were consideredstatistically significant when the p value was less than 0.05. Allstatistical analyses were performed using the SPSS16 statisticalsoftware package (SPSS Inc., Chicago, IL, USA).

3. Results

3.1. SalB protects against APAP-induced acute liver injury

We first determined the protective effect of SalB against APAP-induced hepatotoxicity in vivo. As shown in Fig. 1A, APAP alonedramatically increased serum ALT and AST activity. However, whenanimals also received SalB pretreatment, the serum ALT and ASTactivity were significantly decreased compared to those of the APAPcontrol. We further verified the protective effect of SalB by per-forming a histopathological evaluation of the liver. As shown inFig. 1B, the vehicle produced no apparent abnormalities, whereasAPAP-induced a disordered arrangement of hepatocytes,

g per day, i.g., for 3 days) received a single dose of APAP (300 mg/kg, i.g.). The animalssamples. (B) Hematoxylin and eosin staining of the liver (�400). (a) Mice treated withAPAP-induced liver injury. (d) SalB (50 mg/kg) pretreatment suppressed APAP-induced##p < 0.01 vs. the APAP group.

Fig. 3. Effect of selective HO-1 and GCLC inhibitors on HepG2 cell viability. HepG2 cellswere pretreated with 20 mM ZnPP or 200 mM BSO for 2 h and then treated with 8 mMSalB for 6 h. Next, cells were further exposed to 10 mM APAP for 24 h to evaluate cellviability. **p < 0.01 vs. the control group; ##p < 0.01 vs. the APAP group; &&p < 0.01 vs.the SalB þ APAP group.

M. Lin et al. / Journal of Pharmacological Sciences 127 (2015) 203e210206

degeneration in the hepatocytes and cytoplasm dissolution.Microscopic examination revealed that the severe hepatic lesionsinduced by APAP were remarkably reduced by SalB pretreatment.Liver injury was relieved and reversed with SalB pretreatment.These results indicated that SalB prevents APAP-induced injury in amouse model.

3.2. SalB induces expression of Nrf2 and phase II enzyme genes

Cells can be protected from oxidative stress either by directlyscavenging ROS (reactive oxygen species) or by fortifying the body'santioxidant defenses. Therefore, we determined whether the pro-tective effect of SalB against liver injury is associated with Nrf2 andphase II enzymes. In Fig. 2A, pretreatment of mice with SalB aloneincreased Nrf2, HO-1 and GCLC protein expression compared withthat in vehicle controls. Moreover, pretreatment with SalB for 3days before APAP administration significantly increased Nrf2, HO-1and GCLC protein expression compared with that in mice receivingAPAP. Additionally, we probed the effects of SalB on Nrf2, HO-1 andGCLC protein expression in vitro (Fig. 2B), which showed the sameresults as that in vivo. Consistent with protein expression, the cor-responding mRNA expression in vivo and in vitro displayed thesame trend (data not shown). To further investigate whether theprotective effects of SalB are associated with induction of HO-1 andGCLC, we used specific HO-1 and GCLC inhibitors (ZnPP and BSO,respectively). The inhibitors abrogated the protective effect of SalBon HepG2 cells (Fig. 3). Taken together, these experiments sug-gested that SalB protects the liver from APAP-induced injurythrough induction of Nrf2, HO-1 and GCLC.

3.3. SalB-induced phase II enzyme expression is mediated by Nrf2

Given that most genes encoding phase II detoxifying and antiox-idant enzymes contain an ARE, which has been reported as themechanism throughwhichNrf2 regulates transcription (21), we nextstudiedwhetherNrf2 up-regulated the expression ofHO-1 andGCLC.

Fig. 2. Effect of SalB on Nrf2, HO-1 and GCLC expression. (A) Effect of SalB on the protein lewith SalB (25 or 50 mg/kg per day, i.g., for 3 days) received a single dose of APAP (300 mg/kprotein levels of GCLC in HepG2 cells. Cells were pretreated with 0, 0.5, 2, or 8 mM SalB for 6vs. the control group; ##p < 0.01 vs. the APAP group; &&p < 0.01 vs. the SalB þ APAP group

We examined HO-1 and GCLC expression after siRNA knockdown ofNrf2 in HepG2 cells. Silencing Nrf2 expression with specific siRNAabolished SalB-induced up-regulation of HO-1 and GCLC, whereasNrf2 expression was not affected in the control siRNA group (Fig. 4).These results suggest that the SalB-induced expression of HO-1 andGCLC is regulated by the activation of Nrf2 signaling.

3.4. SalB enhances Nrf2-mediated phase II enzyme expression viaactivation of PI3K/AKT and PKC signaling

Several signaling pathways, including the MAPK, PI3K/AKT, andPKC pathways, are involved in the induction of Nrf2 activation(22,23). To elucidate the upstream signaling events leading to Nrf2

vels of Nrf2, HO-1 and GCLC during APAP-induced liver injury in mice. Mice pretreatedg, i.g.). The animals were killed 24 h after APAP administration. (B) Effect of SalB on theh. Values are expressed as means ± SD, n ¼ 3. *p < 0.05 vs. the control group; **p < 0.01.

Fig. 4. Involvement of Nrf2 in SalB-induced HO-1 and GCLC expression. HepG2 cells were transfected with a Nrf2 siRNA or control siRNA for 6 h, followed by treatment with 8 mMSalB for an additional 42 h. Nrf2, HO-1 and GCLC were evaluated using Western blotting. Values are expressed as means ± SD, n ¼ 3. **p < 0.01 vs. the si-control group.

M. Lin et al. / Journal of Pharmacological Sciences 127 (2015) 203e210 207

activation in SalB-treated cells, the following kinase inhibitors wereused: GF109203X (for PKC), LY294002 (for PI3K), PD98059 (forextracellular signal-related protein kinases 1/2 (ERK1/2)),SB203580 (for P38), and SP600125 (for c-Jun N-terminal kinase(JNK)). The results indicated that inhibiting PI3K with LY294002 orthe PKC pathways with GF109203X blocked the protective effect ofSalB against cell death induced by APAP; however, the other in-hibitors did not show this effect (Fig. 5A).

To further elucidate the roles of PKC and PI3K signaling in SalB-induced Nrf2 activation, we examined the effects of SalB on theactivation of PKC and AKT, a PI3K downstream target. SalB signifi-cantly increased the protein phosphorylation levels of AKTand PKC,whereas no difference was found in the phosphorylation of ERK,P38 and JNK proteins (Fig. 5B). To further elucidate the molecularmechanism underlying Nrf2 activation, we attempted to determinethe link between PI3K/AKT, PKC and Nrf2 activation induced bySalB. Interestingly, specifically inhibiting PI3K/AKT and PKCapparently abolished the ability of SalB to increase Nrf2, HO-1 and

Fig. 5. Effect of selective kinase inhibitors on cell viability and protein expression in the presLY294002, 20 mM PD98059, 20 mM SB203580 or 20 mM SP600125 for 1 h and then treatedValues are expressed as means ± SD, n ¼ 10. ** p < 0.01 vs. the control group; ## p < 0.01 vs. tphosphorylation levels of AKT, PKC, ERK, P38, and JNK. Cells treated with 0, 0.5, 2, or 8 mM Sblotting. Values are expressed as means ± SD, n ¼ 3.

GCLC expression (Fig. 6A and B). These data indicated that PI3K/AKTand PKC are candidate kinases for the regulation of Nrf2, HO-1 andGCLC expression induced by SalB (Fig. 7).

4. Discussion

Radix Salvia miltiorrhiza is one of the most important traditionalherbal medicines and is widely used to treat coronary artery dis-ease, cardiovascular diseases and liver fibrosis (14,24). SalB, themost abundant and bioactive ingredient of Radix Salvia miltiorrhiza,has raised considerable interest in recent years. Previous in-vestigations have reported that SalB possesses hepatoprotectiveeffects (15,25). SalB suppresses TGF-b1 expression, decreases MAPKactivity and effectively reverses the effects of liver fibrosis duringchronic hepatitis B. In a study by Kong R et al, it was shown thatSalB pretreatment protected liver tissue against ischemia-reperfusion injury through decreased post-ischemic oxidativestress, improved energy metabolism, and reduced hepatocellular

ence of SalB. (A) Cell viability of HepG2 cells pretreated with 20 mM GF109203X, 20 mMwith 8 m M SalB for 6 h. The cells were then further exposed to 10 mM APAP for 24 h.he APAP group; && p < 0.01 vs. the SalB þ APAP group. (B) Effect of SalB on the total andalB for 6 h. Phosphorylated AKT, PKC, ERK, P38 and JNK were evaluated using Western

Fig. 6. The roles of PKC and PI3K signaling in SalB-induced Nrf2, HO-1 and GCLC expression. HepG2 cells were pretreated with 20 mM GF109203X or 20 mM LY294002 for 1 h andthen treated with 8 mM SalB for 6 h. AKT, PKC, Nrf2, HO-1 and GCLC levels were assessed using Western blotting. PKC, Nrf2, HO-1 and GCLC levels were assessed using Westernblotting. Values are expressed as means ± SD, n ¼ 3. **p < 0.01 vs. the control group.

M. Lin et al. / Journal of Pharmacological Sciences 127 (2015) 203e210208

apoptosis (26). SalB was also reported to protect against platelet-derived growth factor (PDGF)-induced cell proliferation andmigration, and these beneficial effects are associated with the in-duction of HO-1 via Nrf2 activation (27). In the present study, forthe first time, we reported that SalB clearly protects hepatocytesagainst APAP-induced damage. The protective effect is mainlyassociated with up-regulation of HO-1 and GCLC expression

Fig. 7. Scheme summarizing the inhibition of APAP-induced liver injury by SalB via theup-regulation of Nrf2, HO-1 and GCLC expression through the PI3K and PKC pathways.

through the activation of Nrf2 translocation and the stimulation ofthe PI3K/AKT and PKC pathways.

Numerous studies have shown that administering naturalcompounds to induce phase II enzymes is sufficient to promoteantioxidant and protective activities. It is generally known that HO-1 and GCLC are principal phase II enzymes. HO-1 is induced as aprotective mechanism in response to various stimuli. HO-1 over-expression has been demonstrated to protect against ethanol-induced apoptosis (28). By contrast, suppression of HO-1 activityresults in increased hepatocellular injury, apoptosis, and deathfrom infection/sepsis in mice (29). In addition, certain HO-1 in-ducers have the potential to protect hepatocytes against oxidativeinjury (30). GCLC is the rate-limiting enzyme in the synthesis ofglutathione, a major antioxidant molecule in cells. Previous studieshave shown that oleanolic acid lessens APAP-induced cellulardamage by up-regulating GCLC expression (31). Consistent withthese observations, in the present study, we found that SalB-induced HO-1 and GCLC expression in liver tissue and HepG2 cells(Fig. 2). More importantly, the protective effect of SalB against APAPinjury in HepG2 cells was attenuated by the HO-1 inhibitor ZnPPand the GCLC inhibitor BSO, suggesting that the former effect isassociated with HO-1 and GCLC (Fig. 3).

Induction of HO-1 and GCLC is mainly due to transcriptionalactivation mediated by Nrf2 through its interaction with ARE reg-ulatory DNA elements (32,33). Additionally, Nrf2 target genes havebeen shown to be involved in the detoxification of highly reactiveintermediates; therefore, the substance that activates Nrf2 mightbe a good candidate target for protecting against APAP-inducedhepatotoxicity. In the present study, we found that SalB up-

M. Lin et al. / Journal of Pharmacological Sciences 127 (2015) 203e210 209

regulated Nrf2 expression in vivo and in vitro. To determine theinvolvement of Nrf2 in SalB-induced HO-1 and GCLC expression, weused RNA interference directed against Nrf2. The results indicatedthat silencing Nrf2 with siRNA abolished SalB-induced Nrf2, HO-1and GCLC protein expression (Fig. 4), suggesting that the SalB-induced up-regulation of HO-1 and GCLC depends on increasednuclear expression of Nrf2.

It has been reported that mice administration of APAP couldincrease cytoprotective genes, including Nrf2, NADPH quinoneoxidoreductase 1 (Nqo1) and GCLC, which may provide insight intoadaptive recovery mechanisms and development of resistance ofcytotoxic xenobiotics (34). Combined with the previous results, wealso found that APAP itself increased Nrf2, HO-1 and GCLC proteinexpression, which suggested that cellular defense is able to act inconcert to resist harmful reactive intermediates formed from APAPin a certain extent. In other words, increased Nrf2, HO-1 and GCLCexpression is an adaptive response to APAP-induced hepatitis.

Several cytosolic kinases, such asMAPK, PI3K/AKT and PKC, havebeen shown to modulate Nrf2 and participate in signal trans-duction from antioxidants and xenobiotics to ARE (35,36). ERK, JNK,and p38 are the primary members of the MAPK family. We there-fore observed the effects of SalB on the expression levels of MAPKs,PI3K/AKT and PKC. Our results suggested that AKT and PKC but notp38, JNK or ERK are mainly activated in the presence of SalB(Fig. 5B). Our observations therefore support the hypothesis thatPI3K/AKT and PKC play an important role in the protective effect ofSalB against liver injury induced by APAP overdose. In the presentstudy, we demonstrated that the PI3K/AKT inhibitor LY293002 andthe PKC inhibitor GF109203X alleviated the cytoprotective effect ofSalB against APAP-induced cell death (Fig. 5A). Consequently, weconcluded that SalB triggers PI3K and PKC signaling to activate theNrf2 signaling pathway, a process that is, at least in part, respon-sible for the protective effect against APAP-induced hepatotoxicity.

In summary, we demonstrated for the first time that SalB pro-tects the liver from APAP-induced injury by up-regulating HO-1and GCLC expression via the activation of Nrf2 nuclear trans-location and that the PI3K/AKT and PKC pathways are involved inthis process. These findings provide a rationale for potential clinicalapplications of SalB for the prevention or treatment of liverintoxication.

Conflict of interest

There are no conflicts of interest to disclose for any of theauthors.

Acknowledgments

This work was supported by the grants of Chinese NationalNatural Science Foundation (No. 81173641, 30872449) and the KeyLaboratory Foundation of Liaoning Province of China (No.LS2010052). All authors read and approved the final manuscript.

The authors would like to acknowledge Dr. Qianzheng Zhu (OhioState University, USA) for helpful discussion and editing the paper.

References

(1) Svobodova AR, Galandakova A, Sianska J, Dolezal D, Ulrichova J, Vostalova J.Acute exposure to solar simulated ultraviolet radiation affects oxidativestress-related biomarkers in skin, liver and blood of hairless mice. Biol PharmBull. 2011;34(4):471e479.

(2) Davis ML, Hashemi N. Acute liver failure as a rare initial manifestation ofperipheral T-cell lymphoma. World J Hepatol. 2010;2(10):384e386.

(3) Shi Y, Zhang L, Jiang R, Chen W, Zheng W, Chen L, et al. Protective effects ofnicotinamide against acetaminophen-induced acute liver injury. Int Immu-nopharmacol. 2012;14(4):530e537.

(4) Vad NM, Yount G, Moore D, Weidanz J, Moridani MY. Biochemical mechanismof acetaminophen (APAP) induced toxicity in melanoma cell lines. J Pharm Sci.2009;98(4):1409e1425.

(5) Bessems JG, Vermeulen NP. Paracetamol (acetaminophen)-induced toxicity:molecular and biochemical mechanisms, analogues and protective ap-proaches. Crit Rev Toxicol. 2001;31(1):55e138.

(6) A Barve, Khor TO, Nair S, Lin W, Yu S, Jain MR, et al. Pharmacogenomic profileof soy isoflavone concentrate in the prostate of Nrf2 deficient and wild-typemice. J Pharm Sci. 2008;97(10):4528e4545.

(7) Donovan EL, McCord JM, Reuland DJ, Miller BF, Hamilton KL. Phytochemicalactivation of Nrf2 protects human coronary artery endothelial cells against anoxidative challenge. Oxid Med Cell Longev. 2012;2012:132931.

(8) Liu XM, Peyton KJ, Shebib AR, Wang H, Durante W. Compound C stimulatesheme oxygenase-1 gene expression via the Nrf2-ARE pathway to preservehuman endothelial cell survival. Biochem Pharmacol. 2011;82(4):371e379.

(9) Li L, Dong H, Song E, Xu X, Liu L, Song Y. Nrf2/ARE pathway activation, HO-1and NQO1 induction by polychlorinated biphenyl quinone is associated withreactive oxygen species and PI3K/AKT signaling. Chem Biol Interact.2014;209:56e67.

(10) Hsu WH, Lee BH, Li CH, Hsu YW, Pan TM. Monascin and AITC attenuatemethylglyoxal-induced PPARgamma phosphorylation and degradationthrough inhibition of the oxidative stress/PKC pathway depending on Nrf2activation. J Agric Food Chem. 2013;61(25):5996e6006.

(11) Zhao GR, Zhang HM, Ye TX, Xiang ZJ, Yuan YJ, Guo ZX, et al. Characterization ofthe radical scavenging and antioxidant activities of danshensu and salvianolicacid B. Food Chem Toxicol. 2008;46(1):73e81.

(12) Wu XJ, Wang YP, Wang W, Sun WK, Xu YM, Xuan LJ. Free radical scavengingand inhibition of lipid peroxidation by magnesium lithospermate B. ActaPharmacol Sin. 2000;21(9):855e858.

(13) Soung DY, Rhee SH, Kim JS, Lee JY, Yang HS, Choi JS, et al. Peroxynitritescavenging activity of lithospermate B from Salvia miltiorrhiza. J PharmPharmacol. 2003;55(10):1427e1432.

(14) Lin YL, Wu CH, Luo MH, Huang YJ, Wang CN, Shiao MS, et al. In vitro protectiveeffects of salvianolic acid B on primary hepatocytes and hepatic stellate cells.J Ethnopharmacol. 2006;105(1e2):215e222.

(15) Wang Y, Xu F, Chen J, Shen X, Deng Y, Xu L, et al. Matrix metalloproteinase-9induces cardiac fibroblast migration, collagen and cytokine secretion: inhi-bition by salvianolic acid B from Salvia miltiorrhiza. Phytomedicine.2011;19(1):13e19.

(16) Liu P, Hu YY, Liu C, Zhu DY, Xue HM, Xu ZQ, et al. Clinical observation ofsalvianolic acid B in treatment of liver fibrosis in chronic hepatitis B. World JGastroenterol. 2002;8(4):679e685.

(17) Wang R, Yu XY, Guo ZY, Wang YJ, Wu Y, Yuan YF. Inhibitory effects of sal-vianolic acid B on CCl(4)-induced hepatic fibrosis through regulating NF-kappaB/IkappaBalpha signaling. J Ethnopharmacol. 2012;144(3):592e598.

(18) Perlmutter DH, Colten HR, Adams SP, May LT, Sehgal PB, Fallon RJ. Acytokine-selective defect in interleukin-1 beta-mediated acute phase gene expressionin a subclone of the human hepatoma cell line (HEPG2). J Biol Chem.1989;264(13):7669e7674.

(19) Guerrero JA, Teruel R, Martínez C, Arcas I, Martínez-Martínez I, de la Morena-Barrio ME, et al. Protective role of antithrombin in mouse models of liverinjury. J Hepatol. 2012;57(5):980e986.

(20) Zhai X, Lin M, Zhang F, Hu Y, Xu X, Li Y, et al. Dietary flavonoid genisteininduces Nrf2 and phase II detoxification gene expression via ERKs and PKCpathways and protects against oxidative stress in Caco-2 cells. Mol Nutr FoodRes. 2013;57(2):249e259.

(21) Wakabayashi N, Itoh K, Wakabayashi J, Motohashi H, Noda S, Takahashi S,et al. Keap1-null mutation leads to postnatal lethality due to constitutive Nrf2activation. Nat Genet. 2003;35(3):238e245.

(22) Hamdulay SS, Wang B, Birdsey GM, Ali F, Dumont O, Evans PC, et al. Celecoxibactivates PI-3K/Akt and mitochondrial redox signaling to enhance hemeoxygenase-1-mediated anti-inflammatory activity in vascular endothelium.Free Radic Biol Med. 2010;48(8):1013e1023.

(23) Weng CJ, Chen MJ, Yeh CT, Yen GC. Hepatoprotection of quercetin againstoxidative stress by induction of metallothionein expression through acti-vating MAPK and PI3K pathways and enhancing Nrf2 DNA-binding activity.N Biotechnol. 2011;28(6):767e777.

(24) Han JY, Fan JY, Horie Y, Miura S, Cui DH, Ishii H, et al. Ameliorating effects ofcompounds derived from Salvia miltiorrhiza root extract on microcirculatorydisturbance and target organ injury by ischemia and reperfusion. PharmacolTher. 2008;117(2):280e295.

(25) Lv Z, Song Y, Xue D, Zhang W, Cheng Y, Xu L. Effect of salvianolic-acid B oninhibiting MAPK signaling induced by transforming growth factor-beta1 inactivated rat hepatic stellate cells. J Ethnopharmacol. 2010;132(2):384e392.

(26) Kong R, Gao Y, Sun B, Chen H,Wang G, Wang X, et al. The strategy of combinedischemia preconditioning and salvianolic acid-B pretreatment to prevent he-patic ischemia-reperfusion injury in rats. Dig Dis Sci. 2009;54(12):2568e2576.

(27) Lee HJ, Seo M, Lee EJ. Salvianolic acid B inhibits atherogenesis of vascular cellsthrough induction of Nrf2-dependent heme oxygenase-1. Curr Med Chem.2014;21(26):3095e3106.

(28) Nepal S, Kim MJ, Subedi A, Lee ES, Yong CS, Kim JA, et al. Globular adiponectininhibits ethanol-induced apoptosis in HepG2 cells through heme oxygenase-1induction. Biochem Pharmacol. 2012;84(7):974e983.

(29) Carchman EH, Rao J, Loughran PA, Rosengart MR, Zuckerbraun BS. Hemeoxygenase-1-mediated autophagy protects against hepatocyte cell death and

M. Lin et al. / Journal of Pharmacological Sciences 127 (2015) 203e210210

hepatic injury from infection/sepsis in mice. Hepatology. 2011;53(6):2053e2062.

(30) McNally SJ, Harrison EM, Ross JA, Garden OJ, Wigmore SJ. Curcumin inducesheme oxygenase-1 in hepatocytes and is protective in simulated coldpreservation and warm reperfusion injury. Transplantation. 2006;81(4):623e626.

(31) Liu J, Wu Q, Lu YF, Pi J. New insights into generalized hepatoprotective effectsof oleanolic acid: key roles of metallothionein and Nrf2 induction. BiochemPharmacol. 2008;76(7):922e928.

(32) Ishii T, Itoh K, Takahashi S, Sato H, Yanagawa T, Katoh Y, et al. Transcriptionfactor Nrf2 coordinately regulates a group of oxidative stress-inducible genesin macrophages. J Biol Chem. 2000;275(21):16023e16029.

(33) Kaspar JW, Niture SK, Jaiswal AK. Nrf2:INrf2 (Keap1) signaling in oxidativestress. Free Radic Biol Med. 2009;47(9):1304e1309.

(34) Aleksunes LM, Slitt AL, Maher JM, Augustine LM, Goedken MJ, Chan JY, et al.Induction of Mrp3 and Mrp4 transporters during acetaminophen hepato-toxicity is dependent on Nrf2. Toxicol Appl Pharmacol. 2008;226(1):74e83.

(35) Angeloni C, Motori E, Fabbri D, Malaguti M, Leoncini E, Lorenzini A, et al. H2O2preconditioning modulates phase II enzymes through p38 MAPK and PI3K/Akt activation. Am J Physiol Heart Circ Physiol. 2011;300(6):H2196eH2205.

(36) Cheng SE, Lee IT, Lin CC, Kou YR, Yang CM. Cigarette smoke particle-phaseextract induces HO-1 expression in human tracheal smooth muscle cells:role of the c-Src/NADPH oxidase/MAPK/Nrf2 signaling pathway. Free RadicBiol Med. 2010;48(10):1410e1422.