The protective action of ketanserin against lipopolysaccharide-inducedshock in mice is mediated by inhibiting inducible NO synthaseexpression via the MEK/ERK pathway

Chong Liu n,1, Xin Zhang 1, Jv-Xiang Zhou, Wei Wei, Dian-Hua Liu, Ping Ke, Gu-Fang Zhang,Guo-Jun Cai n, Ding-Feng Su n

Department of Pharmacology, Second Military Medical University, Shanghai 200433, China

a r t i c l e i n f o

Article history:Received 24 October 2012Received in revised form26 July 2013Accepted 31 July 2013Available online 13 August 2013

Keywords:KetanserinEndotoxic shockInducible nitric oxide synthaseNitric oxideFree radicals

a b s t r a c t

Nitric oxide (NO) plays an important role in the pathogenesis of endotoxic shock. This work tested thehypothesis that ketanserin could attenuate endotoxic shock by inhibiting the expression of inducible NOsynthase (iNOS). The results demonstrated that ketanserin could inhibit iNOS expression in the heart,lungs, liver, and kidneys and nitrate production in the serum upon endotoxic shock in mice. In RAW264.7cells, ketanserin significantly inhibited the expression of iNOS and decreased the production of NO, TNFα,IL-6, and reactive oxygen species upon lipopolysaccharide (LPS) challenge. Ketanserin also increased thelevel of ATP and mitochondrial membrane potential in RAW264.7 cells upon LPS exposure. LPS-inducediNOS expression was inhibited by the 5-HT2A receptor antagonist ritanserin and not the α1 receptor anta-gonist prazosin. Knockdown of 5-HT2A receptor by siRNA abolished the inhibitory effect of ketanserin onthe expression of iNOS. These results indicated that the inhibitory effect of ketanserin on the expressionof iNOS is mediated by blocking the 5-HT2A receptor. Furthermore, ketanserin significantly inhibited theactivation of ERK1/2 and NF-κB signal. Pretreatment with PD184352, a specific inhibitor of ERK1/2,blocked the inhibitory effect of ketanserin on the expression of iNOS and NO production, indicating acritical role for the MEK/ERK1/2 signaling pathway. Collectively, our findings indicate that inhibition ofthe expression of iNOS via the MEK/ERK pathway mediates the protective effects of ketanserin againstLPS-induced shock in mice.

Published by Elsevier Inc.

Ketanserin is a selective 5-hydroxytryptamine-2A (5-HT2A) recep-tor antagonist with minor activity on the α1 adrenergic receptor [1].Previous studies from this laboratory demonstrated that ketanserincould increase baroreflex sensitivity in a variety of animal modelswith baroreflex dysfunction [2]. Intact arterial baroreflex function iscritical for surviving septic shock induced by either lipopolysacchar-ide (LPS) or cecal ligation and puncture [3,4]. We demonstrated thatketanserin could reduce mortality in mice from LPS-induced shock,partially by enhancing the baroreflex function [5]. In vitro studies,however, suggested that ketanserin could produce anti-inflammatoryaction independent of baroreflex [6,7]. In a preliminary study, wefound decreased expression of inducible nitric oxide (NO) synthase(iNOS) upon ketanserin treatment in endotoxic shock mice.

NO is synthesized by NOS [8,9]. Inducible NOS can be inducedby a variety of stimuli, including LPS, interleukin 1 (IL-1), tumornecrosis factor (TNF), and interferon-γ [10–12]. The formationof NO in activated macrophages plays a key role in their

antimicrobial activity [13,14]. Enhanced formation of NO afterthe induction of iNOS also contributes to the circulatory failurein circulatory shock [15–17]. Treatment with iNOS inhibitors orknockout of the iNOS gene increased microvascular reactivity andreduced the mortality in mice with septic shock [18–21]. In thisstudy, we tested the hypothesis that ketanserin could attenuateendotoxic shock by inhibiting the expression of iNOS.

Materials and methods

Agents

Ketanserin, LPS (Escherichia coli 0111:B4), ritanserin (a selectiveantagonist of the 5-HT2A receptor), prazosin (an antagonist of theα1 receptor), and PD184352 (inhibitor of MEK1) were purchasedfrom Sigma–Aldrich Chemical (St. Louis, MO, USA). Antibody foriNOS was purchased from Abcam (Cambridge, UK); antibodies forp38, p-p38, ERK1/2, p-ERK1/2, JNK, and p-JNK were from CellSignaling Technology (Danvers, MA, USA). Lipofectamine 2000 wasobtained from Roche (Mannheim, Germany). ELISA kits for IL-6and TNFα were obtained from R&D Systems (Minneapolis, MN,

Contents lists available at ScienceDirect

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

Free Radical Biology and Medicine

0891-5849/$ - see front matter Published by Elsevier Inc.http://dx.doi.org/10.1016/j.freeradbiomed.2013.07.045

n Corresponding author.E-mail addresses: wanlc2004@aliyun.com (C. Liu),

13386277625@163.com (G.-J. Cai), dfsu2008@gmail.com (D.-F. Su).1 These authors contributed equally to this work.

Free Radical Biology and Medicine 65 (2013) 658–666

USA). The nitrate/nitrite ratio, NO, reactive oxygen species (ROS),ATP, and mitochondrial membrane potential (ψm) were examinedwith assay kits from Beyotime (Haimen, China).

Animals

Male Kunming mice (25–30 g) were purchased from Sino-British SIPPR/BK Laboratory Animal Ltd (Shanghai, China) andkept at 22 1C under a 12-h light/dark cycle with unlimited accessto water and standard rodent diet. All animals received humanecare in accordance with institutional animal care guidelines andthe Guide for the Care and Use of Laboratory Animals published bythe U.S. National Institutes of Health.

Cell line

RAW264.7 cells were obtained from the American Type CultureCollection (Rockville, MD, USA) and cultured in Dulbecco'smodified Eagle's medium (Hyclone, Logan, UT, USA) supplementedwith 10% fetal bovine serum (Gibco BRL, Grand Island, NY, USA)at 37 1C in a humidified incubator with 5% CO2.

Small interfering RNA (siRNA) transfection

SiRNAs were synthesized by Genepharm Biotech (Shanghai, China).Several siRNAs targeting the 5-HT2A receptor (Gene ID: 15558) weresynthesized: siRNA1, GGAUGCUAACACUUCCGAATT; siRNA2, GACAA-CUGUCGUGAUUAUUTT; and siRNA3, GCCGUCAUCUGCAAAGAAUTT.UUCUCCGAACGUGUCACGUTT was used as a negative control. AllsiRNAs consisted of 21 nucleotides and contained symmetric 3′ over-hangs of two deoxythymidines. RAW264.7 cells were transfected withsiRNAs as described before [22]. After transfection, the cells wereincubated with ketanserin and then stimulated with LPS.

Western blotting

Proteins were extracted from the tissue or cultured cells usinga standard extraction reagent supplemented with protease inhibitors(Kangchen, Shanghai, China). Protein concentration was determinedusing a bicinchoninic acid method (Beyotime). The proteins wereseparated using SDS–PAGE and electrotransferred to nitrocellulosemembranes and then incubated with a primary antibody for 8–12 h at4 1C. Samples were then incubated with an IRDye800CW-conjugatedsecondary antibody (Rockland, Gilbertsville, PA, USA) for 1 h at 25 1C.The image was acquired with an Odyssey infrared imaging system(Li-Cor Bioscience, Lincoln, NE, USA). All immunoblotting experimentswere repeated at least three times.

Immunohistochemistry

Tissue blocks were immersed in 10% paraformaldehyde inphosphate-buffered saline (PBS) for 24 h. The samples were thenwashed, dehydrated in a graded series of alcohol, embedded inparaffin, and cut into 2-μm-thick slices. Peroxidase was inactivatedby incubation with hydrogen peroxide at room temperature for20 min. Antigens were retrieved by heating in a microwave for20 min. Sections were incubated with 20% bovine serum albuminfor 30 min at 25 1C and then with a rabbit anti-iNOS (1:100) for 3 h at37 1C, followed by treatment with EnVision reagent (Dako, Glostrup,Denmark) for rabbit for 30 min at 25 1C. The bands were visualizedwith 3,3′-diaminobenzidine treatment (Sigma) for 10 min, stainedwith hematoxylin, and examined under a light microscope [23,24].

Nitrate/nitrite assay

Nitrate/nitrite was quantified by measuring the accumulationof nitrate. Blood samples were allowed to clot for 2 h at roomtemperature before centrifugation for 20 min at 3000g. Nitrateproduction in the serum was measured using a total NO assay kit.

Enzyme-linked immunosorbent assay (ELISA)

TNFα and IL-6 concentration in culture medium was deter-mined using a commercial ELISA kit as described before [25,26].

Flow cytometry

Cultured cells were centrifuged at 3000g for 5 min and rinsedtwice with PBS. NO, ROS, and mitochondrial ψm were determinedusing commercial assay kits as described before [27–29]. Fluores-cence was detected using a flow cytometer (Becton–DickinsonFACSCalibur, Franklin, NJ, USA).

Determination of ATP levels

Cells were treated with taurine (0.1 mM) and then centrifugedat 3000g for 5 min. Supernatant was mixed with the same volumeof ATP detection working dilution. Luminance was measured bya monochromator microplate reader (Safire II, Tecan, Männedorf,Switzerland) [30].

Electrophoretic mobility shift assay (EMSA)

Cell nuclear extract was prepared using a nuclear protein extrac-tion kit (Pierce, Rockford, IL, USA). Samples containing equal amountsof nuclear extract protein (10 μg) were incubated with 1.0 μg/μl poly(dI–dC) and 500 fmol biotin-labeled double-stranded nuclear factorκB (NF-κB) binding consensus oligonucleotide 5′-AGTTGAGGG-GACTTTCCAGGC-3′ using an EMSA kit (Roche). The binding reactionproceeded for 15 min at 25 1C. The DNA–protein complexes wereelectrophoresed on a 6.5% nondenaturing polyacrylamide gel andelectrotransferred for detection.

Experimental protocols

Experiment 1: effects of ketanserin on the expression of iNOS andnitrate production in mice challenged with LPS

Mice were injected with LPS (30 mg/kg, ip), followed by ketan-serin (3.0 or 10.0 mg/kg, ip) or vehicle. The heart, lung, liver, andkidneys were excised at 12 h after LPS injection for determination ofiNOS expression using Western blotting. Blood samples werecollected at 12 h after LPS injection for nitrate/nitrite production.

Experiment 2: effects of ketanserin on the expression of iNOS andNO production in RAW264.7 cells stimulated with LPS

RAW264.7 cells (1�106/ml) were preincubated with ketan-serin (10 μM) for 10 min and then stimulated with LPS (100 ng/ml)for 0–12 h. In another set of experiments, RAW264.7 cells werepreincubated with ketanserin (0–10 μM) for 10 min and then stimu-lated with LPS for 8 h. Cells were harvested and the expression ofiNOS was examined. For NO production, RAW264.7 cells werepretreated with ketanserin (1 or 10 μM) for 10 min and stimulatedwith LPS for 8 h. Cell suspension was used for fluorescence detectionusing a flow cytometer.

C. Liu et al. / Free Radical Biology and Medicine 65 (2013) 658–666 659

Experiment 3: effects of ketanserin on cytokine productionin RAW264.7 cells stimulated with LPS

RAW264.7 cells (1�106/ml) were preincubated with ketanserin(10 μM) for 10 min and then stimulated with LPS (100 ng/ml; 0–24 h). In another set of experiments, RAW264.7 cells were pre-incubated with ketanserin (0–100 μM) for 10 min and then stimu-lated with LPS. Culture supernatant was collected to determine theconcentration of TNFα and IL-6 using ELISA.

Experiment 4: effects of ketanserin on ROS production, ATP level, andmitochondrial membrane potential in RAW264.7 cells stimulatedwith LPS

RAW264.7 cells were preincubated with ketanserin (0.1–10 μM) for 10 min and then stimulated with LPS (100 ng/ml) for8 h. Cells were collected for examination of ROS production, ATPlevel, and membrane potential as previously mentioned.

Experiment 5: influence of ERK1/2 inhibitors on the effect ofketanserin on iNOS expression and NO production in RAW264.7 cellschallenged with LPS

RAW264.7 cells (1�107/ml) were pretreated with ketanserin(10 μM) for 10 min and then stimulated with LPS (100 ng/ml) for0–2 h. The phosphorylation of ERK1/2 was examined by Westernblotting. In another set of experiments, RAW264.7 cells were prein-cubated with PD184352 for 10 min before sequential treatment withketanserin and LPS, as described above. Cells were collected and theexpression of iNOS and NO production was examined.

Experiment 6: influence of ritanserin, prazosin, and 5-HT2A receptorsiRNA on the expression of iNOS in RAW264.7 cells challengedwith LPS

RAW264.7 cells (1�106/ml) were preincubated with ritanserin(0–10 μM) or prazosin (0–1 μM) for 10 min and then stimulated withLPS (100 ng/ml) for 8 h. In another set of experiments, RAW264.7cells were transfected with control siRNA or 5-HT2A receptor siRNAand then preincubated with ketanserin (0–100 μM) for 10 min beforebeing stimulated with LPS. Cells were harvested and the expressionof iNOS was examined by Western blotting.

Experiment 7: effects of serotonin on iNOS expressionand phosphorylation of ERK1/2 in LPS-stimulated RAW264.7 cells

RAW264.7 cells were preincubated with serotonin (0.1–10 μM)for 10 min and then stimulated with LPS (100 ng/ml) for 8 h. Cellswere collected and the expression of iNOS was examined byWesternblotting. In another set of experiments, RAW264.7 cells (1�107/ml)were pretreated with serotonin (10 μM) for 10 min and thenstimulated with LPS (100 ng/ml) for 0–2 h. The phosphorylation ofERK1/2 was examined.

Experiment 8: effects of ketanserin on the activity of NF-κBin RAW264.7 cells challenged with LPS

RAW264.7 cells (1�107/ml), cultured in six-well cell cultureplates, were treated with ketanserin (10 μM) for 10 min and thenstimulated with LPS (100 ng/ml). The activity of NF-κB wasassessed using EMSA.

Fig. 1. Effects of ketanserin on the expression of iNOS and nitrate production in mice challenged with LPS. Mice received LPS (30 mg/kg, ip) and then ketanserin (3 or 10 mg/kg, ip) or vehicle. The heart, lungs, liver, and kidneys were excised at 12 h after LPS challenge to determine the expression of iNOS. (A) The expression of iNOS by Westernblotting. Data are expressed as fold change relative to vehicle control (n¼6). npo0.05 vs LPS. (B) The expression of iNOS by immunohistochemistry. Brown precipitateindicates positive immunoreaction. In the control mice, iNOS was not observed in any tissue sample. In the LPS group, LPS injection resulted in a strong expression of iNOSin the heart, lungs, liver, and kidneys. In the LPS/ketanserin group, the expression of iNOS was inhibited by ketanserin (n¼6). Original magnification �200. (C and D)Ketanserin inhibited LPS-induced nitrite production in serum, as reflected by Griess reaction assay (n¼8). npo0.05 vs LPS, nnpo0.01 vs LPS.

C. Liu et al. / Free Radical Biology and Medicine 65 (2013) 658–666660

Statistical analysis

Data are expressed as means7SEM and analyzed with one-way analysis of variance (ANOVA) followed by the Tukey post hoctest for pair-wise comparison. For experiments involving twoindependent variables, data were analyzed with two-way ANOVAfollowed by Tukey post hoc analysis. Statistical significance was setat po0.05.

Results

Effects of ketanserin on the expression of iNOS and nitrate productionin mice challenged with LPS

LPS exposure significantly increased the expression of iNOS inthe heart, lungs, liver, and kidneys in mice (Figs. 1A–C). Ketanserin

inhibited the expression of iNOS induced by LPS in all organsmentioned above. In the liver, LPS-induced iNOS was completelyreversed by ketanserin. Ketanserin also decreased nitrate/nitriteproduction in the serum induced by LPS injection (Fig. 1D).

Effects of ketanserin on iNOS expression, NO level, and cytokineproduction in LPS-stimulated RAW264.7 cells

LPS exposure increased the expression of iNOS in RAW264.7cells (Figs. 2A and B). Ketanserin preincubation decreased theexpression of iNOS induced by LPS. Such an inhibitory effect ofketanserin peaked at 8 h. Ketanserin also inhibited the productionof NO induced by LPS (Fig. 2C). Ketanserin concentration- and time-dependently inhibited the production of TNFα (Figs. 3A and B) andIL-6 (Figs. 3C and D) in RAW264.7 cells upon LPS challenge.

Fig. 2. Effects of ketanserin on the expression of iNOS and NO production in RAW264.7 cells stimulated with LPS. (A) RAW264.7 cells were pretreated with ketanserin (0.1–10 μM) for 10 min and stimulated with LPS (100 ng/ml) for 8 h. Expression of iNOS was determined with Western blotting (n¼6). npo0.05 vs LPS, nnpo0.01 vs LPS.(B) RAW264.7 cells were pretreated with ketanserin (10 μM) for 10 min and stimulated with LPS (100 ng/ml) for 0–12 h. Expression of iNOS was determined with Westernblotting (n¼6). npo0.05 vs LPS, nnpo0.01 vs LPS. (C) RAW264.7 cells were pretreated with ketanserin (1 or 10 μM) and then stimulated with LPS (100 ng/ml) for 8 h. Thecells were then incubated with DAF-FM diacetate (n¼8). Fluorescence was detected using a flow cytometer. nnpo0.01 vs LPS.

C. Liu et al. / Free Radical Biology and Medicine 65 (2013) 658–666 661

Effects of ketanserin on ROS production, ATP level, and mitochondrialmembrane potential in RAW264.7 cells stimulated with LPS

LPS challenge increased intracellular ROS and decreased cellu-lar ATP content as well as ψm (Figs. 4A–C). Pretreatment withketanserin attenuated, or obliterated, the effects of LPS.

Influence of ERK1/2 inhibitors on the effect of ketanserin on iNOSexpression in LPS-stimulated RAW264.7 cells

Ketanserin inhibited the phosphorylation of ERK1/2 at both1 and 2 h after LPS challenge in RAW264.7 cells (Fig. 5A), but didnot affect the phosphorylation of p38 and JNK (SupplementaryFig. S2). Preincubation with PD184352 significantly attenuated theinhibitory effect of ketanserin on LPS-induced expression of iNOS(Fig. 5B) and NO production (Fig. 5C).

Influence of prazosin, ritanserin, and 5-HT2A receptor siRNAon the expression of iNOS in LPS-stimulated RAW264.7 cells

LPS-induced expression of iNOS was inhibited by the 5-HT2Areceptor antagonist ritanserin (Fig. 6A), but not the α1 receptorantagonist prazosin (Fig. 6B). Knockdown of the 5-HT2A receptor bysiRNA abolished the inhibitory effect of ketanserin on the expres-sion of iNOS in RAW264.7 cells induced by LPS (Figs. 6C and D).

Effects of serotonin on iNOS expression and phosphorylation ofERK1/2 in LPS-stimulated RAW264.7 cells

Preincubation with serotonin increased the expression of iNOSand phosphorylation of ERK1/2 induced by LPS (Fig. 7).

Fig. 3. Effects of ketanserin on cytokine production in RAW264.7 cells stimulated with LPS. Cells were pretreated with ketanserin (0.1–100 μM) and stimulated with LPS(100 ng/ml) for 0–24 h. TNFα and IL-6 in the culture mediumwere examined by ELISA. (A and C) Ketanserin concentration-dependently inhibited the production of TNFα andIL-6 (n¼8). npo0.05 vs LPS, nnpo0.01 vs LPS. (B and D) The temporal profile of the inhibitory effects of ketanserin on the production of TNFα and IL-6 (n¼8).npo0.05 vs LPS, nnpo0.01 vs LPS.

Fig. 4. Effects of ketanserin on ROS production, ATP level, and mitochondrial membrane potential in RAW264.7 cells stimulated with LPS. Cells were pretreated withketanserin (0.1–10 μM) and stimulated with LPS (100 ng/ml) for 8 h before determination of ROS production, ATP level, and mitochondrial membrane potential.(A) Ketanserin concentration-dependently inhibited the production of ROS (n¼8). nnpo0.01 vs LPS. (B) Ketanserin dose-dependently increased cellular ATP level (n¼8).npo0.05 vs LPS, nnpo0.01 vs LPS. (C) Ketanserin dose-dependently increased mitochondrial membrane potential (n¼8). nnpo0.01 vs LPS.

C. Liu et al. / Free Radical Biology and Medicine 65 (2013) 658–666662

Effects of ketanserin on the activity of NF-κB in LPS-stimulatedRAW264.7 cells

LPS increased NF-κB binding activity in RAW264.7 cells, asreflected by EMSA. Ketanserin attenuated the effect induced byLPS (Fig. 8).

Discussion

Previous studies from this laboratory demonstrated that ketan-serin could dose-dependently decrease the mortality induced bya lethal dose of LPS and prolong the survival time [5]. We alsoshowed that ketanserin could decrease serum TNFα and IL-1βin LPS-treated mice. Sinoaortic denervation could attenuate theseeffects, suggesting that the anti-shock effects of ketanserin couldbe partially attributed to increased arterial baroreflex.

In the present study, we found ketanserin could inhibit theproduction of TNFα and IL-6 in RAW264.7 cells upon LPS chal-lenge. In addition, ketanserin also inhibited ROS production andincreased the cellular ATP content and mitochondrial membranepotential, suggesting that the anti-inflammatory and antioxidativeaction is not entirely dependent on baroreflex.

NO promotes the release of TNFα and IL-6 [31,32]. This studyrevealed increased expression of iNOS in the heart, lungs, liver,and kidneys in LPS-induced-shock mice. In contrast to a previousstudy that suggested the lungs, spleen, and kidneys as the major

sites for iNOS expression in endotoxic shock [33], the increasewas also prominent in the liver and heart in the current study.Ketanserin significantly inhibited LPS-induced expression of iNOSin the heart, kidneys, liver, and lungs and decreased nitrite/nitrateproduction. Taken together with our findings in cultured cells,these results suggest that the direct anti-inflammatory effects ofketanserin are mediated by inhibiting iNOS.

In previous studies from this laboratory, we showed thatketanserin increases baroreflex in spontaneously hypertensive ratsmainly via the central 5-HT2A receptors [3,34]. In the current study,ritanserin, a more selective 5-HT2A receptor antagonist withoutnoticeable activity for the α1 receptor [35], significantly inhibitedthe expression of iNOS in RAW264.7 cells upon LPS challenge. Incontrast, the selective α1 receptor antagonist prazosin failed toinhibit the expression of iNOS. Also, our data indicated thatserotonin increased the expression of iNOS in RAW264.7 cellsstimulated with LPS. In addition to experiments with siRNA, theseresults revealed an essential role of the 5-HT2A receptor in theinhibitory effect of ketanserin on LPS induction of iNOS.

Mitogen-activated protein kinases play important roles inthe regulation of iNOS and many other inducible enzymes impli-cated in the pathophysiology of septic shock [36–38]. This studyrevealed that ketanserin could decrease LPS-induced phosphory-lation of ERK1/2 in RAW264.7 cells. Pretreatment with PD184352,a specific inhibitor of ERK1/2, significantly attenuated the inhibi-tory effect of ketanserin on LPS-induced expression of iNOSand NO production. These findings implicated the MEK/ERK1/2

Fig. 5. Influence of the ERK1/2 pathway on the effect of ketanserin on iNOS expression and NO production. RAW264.7 cells were pretreated with ketanserin (10 μM) for10 min and then stimulated with LPS (100 ng/ml) for 0–2 h before determination of ERK1/2 phosphorylation and NO production. (A) Ketanserin significantly inhibited thephosphorylation of ERK1/2 (n¼6). npo0.05 vs LPS, nnpo0.01 vs LPS. (B) PD184352 significantly attenuated the inhibitory effect of ketanserin on LPS-induced expression ofiNOS (n¼6). nnpo0.01; NS, no significance. (C) PD184352 significantly attenuated the inhibitory effect of ketanserin on LPS-induced production of NO (n¼8). nnpo0.01; NS,no significance.

C. Liu et al. / Free Radical Biology and Medicine 65 (2013) 658–666 663

Fig. 6. Influence of prazosin, ritanserin, and 5-HT2A receptor siRNA on LPS-induced expression of iNOS in RAW264.7 cells. Cells were pretreated with ritanserin (0.01–10 μM)or prazosin (0.01–1 μM) for 10 min and then stimulated with LPS (100 ng/ml) for 8 h before determination of iNOS expression using Western blotting. (A) Prazosin did notaffect the expression of iNOS (n¼6). p40.05 vs LPS. (B) Ritanserin significantly inhibited the expression of iNOS (n¼6). npo0.05 vs LPS, nnpo0.01 vs LPS. (C) The expressionof 5-HT2A receptor mRNA in RAW264.7 cells after transfection with siRNA (n¼6). nnpo0.01 vs control siRNA. (D) Knockdown of 5-HT2A receptor by siRNA abolished theinhibitory effect of ketanserin on the expression of iNOS (n¼6). nnpo0.01.

Fig. 7. Effects of serotonin on the expression of iNOS and phosphorylation of ERK1/2 in RAW264.7 cells stimulated with LPS. (A) RAW264.7 cells were pretreated withserotonin (0.1–10 μM) for 10 min and stimulated with LPS (100 ng/ml) for 8 h. Expression of iNOS was determined with Western blotting (n¼6). nnpo0.01 vs LPS.(B) RAW264.7 cells were pretreated with ketanserin (10 μM) for 10 min and then stimulated with LPS (100 ng/ml) for 0–2 h before determination of ERK1/2 phosphorylationusing Western blotting (n¼6). nnpo0.01 vs LPS.

C. Liu et al. / Free Radical Biology and Medicine 65 (2013) 658–666664

signaling pathway in the inhibitory effect of ketanserin on theexpression of iNOS.

NF-κB plays a central role in the regulation of immune andinflammatory processes and is implicated in sepsis [39–41]. NF-κBrepresents a target for developing novel treatments for inflammatorydiseases [42–44]. In this study, ketanserin inhibited the activity ofNF-κB in LPS-stimulated RAW264.7 cells. Excessive NO production isknown to activate inflammatory cascades through the activation ofNF-κB [31,45,46]. As a result, the inhibitory effect of ketanserin onNF-κB activation may be attained by inhibiting iNOS.

This study showed that the protective effects of ketanserinagainst LPS-induced shock could be partly attributed to inhibitionof iNOS expression. The underlying mechanism may include5-HT2A receptor blockade, inhibition of ERK1/2 phosphorylation,inhibition of iNOS expression, and activation of the NF-κB signal

pathway (Fig. 9). These molecular events/pathways could influenceeach other via cross talk, but such a possibility requires furtherinvestigation.

Also based on the findings in this study, we advocate develop-ing 5-HT2A receptor antagonists as a treatment for endotoxic shockas well as similar conditions.

Acknowledgments

This work was funded by the National Basic Research Programof China (973 Program, 2009CB521901) and the Shanghai NaturalScience Foundation of China (13ZR1448400).

Appendix A. Supplementry material

Supplementary data associated with this article can be foundin the online version at http://dx.doi.org/10.1016/j.freeradbiomed.2013.07.045.

References

[1] Wang, D; Chen, T; Gao, Y; Quirion, R; Hong, Y. Inhibition of SNL-inducedupregulation of CGRP and NPY in the spinal cord and dorsal root ganglia bythe 5-HT(2A) receptor antagonist ketanserin in rats. Pharmacol. Biochem. Behav101:379–386; 2012.

[2] Fu, YJ; Shu, H; Miao, CY; Wang, MW; Su, DF. Restoration of baroreflex functionby ketanserin is not blood pressure dependent in conscious freely moving rats.J. Hypertens 22:1165–1172; 2004.

[3] Shen, FM; Wang, J; Ni, CR; Yu, JG; Wang, WZ; Su, DF. Ketanserin inducedbaroreflex enhancement in spontaneously hypertensive rats depends oncentral 5-HT(2A) receptors. Clin. Exp. Pharmacol. Physiol 34:702–707; 2007.

[4] Shi, KY; Shen, FM; Liu, AJ; Chu, ZX; Cao, YL; Su, DF. The survival time post-cecal ligation and puncture in sinoaortic denervated rats. J. Cardiovasc.Pharmacol 50:162–167; 2007.

[5] Liu, C; Zhang, GF; Song, SW; Cai, GJ; Liu, WH; Miao, CY; Su, DF. Effects ofketanserin on endotoxic shock and baroreflex function in rodents. J. Infect. Dis204:1605–1612; 2011.

[6] Ito, T; Ikeda, U; Shimpo, M; Yamamoto, K; Shimada, K. Serotonin increasesinterleukin-6 synthesis in human vascular smooth muscle cells. Circulation102:2522–2527; 2000.

[7] Lau, WK; Chan, SC; Law, AC; Ip, MS; Mak, JC. The role of MAPK and Nrf2pathways in ketanserin-elicited attenuation of cigarette smoke-induced IL-8production in human bronchial epithelial cells. Toxicol. Sci 125:569–577; 2012.

[8] Cauwels, A. Nitric oxide in shock. Kidney Int. 72:557–565; 2007.[9] Papapetropoulos, A; Rudic, RD; Sessa, WC. Molecular control of nitric oxide

synthases in the cardiovascular system. Cardiovasc. Res 43:509–520; 1999.[10] Riquelme, P; Tomiuk, S; Kammler, A; Fändrich, F; Schlitt, HJ; Geissler, EK;

Hutchinson, JA. IFN-γ-induced iNOS expression in mouse regulatory macro-phages prolongs allograft survival in fully immunocompetent recipients.Mol. Ther21:409–422; 2012.

[11] Yayeh, T; Hong, M; Jia, Q; Lee, YC; Kim, HJ; Hyun, E; Kim, TW; Rhee, MH. Pistaciachinensis inhibits NO production and upregulates HO-1 induction via PI-3K/Aktpathway in LPS stimulated macrophage cells. Am. J. Chin. Med 40:1085–1097;2012.

[12] Malik, M; Jividen, K; Padmakumar, VC; Cataisson, C; Li, L; Lee, J; Howard, OM;Yuspa, SH. Inducible NOS-induced chloride intracellular channel 4 (CLIC4)nuclear translocation regulates macrophage deactivation. Proc. Natl. Acad. Sci.USA 109:6130–6135; 2012.

[13] Lange, M; Enkhbaatar, P; Nakano, Y; Traber, DL. Role of nitric oxide in shock:the large animal perspective. Front. Biosci. 14:1979–1989; 2009.

[14] Gross, SS; Kilbourn, RG; Griffith, OW. NO in septic shock: good, bad or ugly?Learning from iNOS knockouts Trends Microbiol 4:47–49; 1996.

[15] Rim, HK; Cho, W; Sung, SH; Lee, KT. Nodakenin suppresses lipopolysaccharide-induced inflammatory responses in macrophage cells by inhibiting tumornecrosis factor receptor-associated factor 6 and nuclear factor-κB pathways andprotects mice from lethal endotoxin shock. J. Pharmacol. Exp. Ther 342:654–664;2012.

[16] Jiang, H; Huang, Y; Xu, H; Hu, R; Li, QF. Inhibition of hypoxia inducible factor-1α ameliorates lung injury induced by trauma and hemorrhagic shock in rats.Acta Pharmacol. Sin 33:635–643; 2012.

[17] Thiemermann, C. Nitric oxide and septic shock. Gen. Pharmacol. 29:159–166;1997.

[18] Szabó, C; Southan, GJ; Thiemermann, C. Beneficial effects and improvedsurvival in rodent models of septic shock with S-methylisothiourea sulfate,a potent and selective inhibitor of inducible nitric oxide synthase. Proc. Natl.Acad. Sci. USA 91:12472–12476; 1994.

Fig. 8. Effects of ketanserin on DNA-binding activity of NF-κB in LPS-stimulatedRAW264.7 cells. RAW264.7 cells were stimulated with LPS in the presence orabsence of ketanserin. Nuclear extracts from these cells were obtained and NF-κBactivation was determined by EMSA. Lanes 1 and 2, supershift; lanes 3–5, stimulatedwith LPS; lanes 6 and 7, stimulated with LPS in the presence of 10 μM ketanserin;lane 8, unstimulated.

Fig. 9. Schematic illustration of the signaling pathway for anti-shock effects ofketanserin. LPS binds to Toll-like receptor 4 (TLR4) and induces the expression ofiNOS. The resulting NO combines with superoxide to form peroxynitrite, which inturn exerts direct cytotoxicity and initiates an inflammatory cascade through theactivation of NF-κB. Ketanserin binds to the 5-HT2A receptor and inhibitsthe expression of iNOS by attenuating ERK1/2 phosphorylation. As a result, NOproduction is inhibited, and the level of inflammatory cytokines is decreased.

C. Liu et al. / Free Radical Biology and Medicine 65 (2013) 658–666 665

[19] Wu, CC; Chen, SJ; Szabó, C; Thiemermann, C; Vane, JR. Aminoguanidine attenuatesthe delayed circulatory failure and improves survival in rodent models of endotoxicshock. Br. J. Pharmacol 114:1666–1672; 1995.

[20] Hollenberg, SM; Broussard, M; Osman, J; Parrillo, JE. Increased microvascularreactivity and improved mortality in septic mice lacking inducible nitric oxidesynthase. Circ. Res 86:774–778; 2000.

[21] Macmicking, JD; Nathan, C; Hom, G; Chartrain, N; Fletcher, DS; Trumbauer, M;Stevens, K; Xie, QW; Sokol, K; Hutchinson, N, et al. Altered responses tobacterial infection and endotoxic shock in mice lacking inducible nitric oxidesynthase. Cell 81:641–650; 1995.

[22] Siew, EL; Chan, KM; Williams, GT; Ross, D; Inayat-Hussain, SH. Protection ofhydroquinone-induced apoptosis by downregulation of Fau is mediated byNQO1. Free Radic. Biol. Med 53:1616–1624; 2012.

[23] Sun, L; Zhang, GF; Zhang, X; Liu, Q; Liu, JG; Su, DF; Liu, C. Combined administrationof anisodamine and neostigmine produces anti-shock effects: involvement of α7nicotinic acetylcholine receptors. Acta Pharmacol. Sin 33:761–766; 2012.

[24] Liu, C; Shen, FM; Le, YY; Kong, Y; Liu, X; Cai, GJ; Chen, AF; Su, DF. Antishockeffect of anisodamine involves a novel pathway for activating alpha7 nicotinicacetylcholine receptor. Crit. Care Med 37:634–641; 2009.

[25] Yang, C; Yu, JG; Liu, JG; Cai, GJ; Su, DF. Role of intracranial microaneurysm inthe pathogenesis of stroke in SHR-SP. CNS Neurosci. Ther 17:787–789; 2011.

[26] Zhang, XH; Lei, H; Liu, AJ; Zou, YX; Shen, FM; Su, DF. Increased oxidative stressis responsible for severer cerebral infarction in stroke-prone spontaneouslyhypertensive rats. CNS Neurosci. Ther 17:590–598; 2011.

[27] Cortese-Krott, MM; Rodriguez-Mateos, A; Kuhnle, GG; Brown, G; Feelisch, M;Kelm, M. A multilevel analytical approach for detection and visualization ofintracellular NO production and nitrosation events using diaminofluoresceins.Free Radic. Biol. Med 53:2146–2158; 2012.

[28] Rana, T; Schultz, MA; Freeman, ML; Biswas, S. Loss of Nrf2 accelerates ionizingradiation-induced bone loss by upregulating RANKL. Free Radic. Biol. Med53:2298–2307; 2012.

[29] Cole-Ezea, P; Swan, D; Shanley, D; Hesketh, J. Glutathione peroxidase 4 has amajor role in protecting mitochondria from oxidative damage and maintain-ing oxidative phosphorylation complexes in gut epithelial cells. Free Radic.Biol. Med 53:488–497; 2012.

[30] Jeong, W; Bae, SH; Toledano, MB; Rhee, SG. Role of sulfiredoxin as a regulatorof peroxiredoxin function and regulation of its expression. Free Radic. Biol. Med53:447–456; 2012.

[31] Cuzzocrea, S; Mazzon, E; Di Paola, R; Esposito, E; Macarthur, H; Matuschak,GM; Salvemini, D. A role for nitric oxide-mediated peroxynitrite formation ina model of endotoxin-induced shock. J. Pharmacol. Exp. Ther 319:73–81; 2006.

[32] Bultinck, J; Sips, P; Vakaet, L; Brouckaert, P; Cauwels, A. Systemic NO productionduring (septic) shock depends on parenchymal and not on hematopoietic cells:in vivo iNOS expression pattern in (septic) shock. FASEB J 20:2363–2365; 2006.

[33] Kan, W; Zhao, KS; Jiang, Y; Yan, W; Huang, Q; Wang, J; Qin, Q; Huang, X; Lung,Wang S. spleen, and kidney are the major places for inducible nitric oxidesynthase expression in endotoxic shock: role of p38 mitogen-activated proteinkinase in signal transduction of inducible nitric oxide synthase expression. Shock21:281–287; 2004.

[34] Miao, CY; Xie, HH; Yu, H; Chu, ZX; Su, DF. Ketanserin stabilizes blood pressurein conscious spontaneously hypertensive rats. Clin. Exp. Pharmacol. Physiol.30:189–193; 2003.

[35] Sonino, N; Fava, GA; Fallo, F; Franceschetto, A; Belluardo, P; Boscaro, M. Effectof the serotonin antagonists ritanserin and ketanserin in Cushing's disease.Pituitary 3:55–59; 2000.

[36] Choi, Y; Lee, MK; Lim, SY; Sung, SH; Kim, YC. Inhibition of inducible NOsynthase, cyclooxygenase-2 and interleukin-1beta by torilin is mediated bymitogen-activated protein kinases in microglial BV2 cells. Br. J. Pharmacol156:933–940; 2009.

[37] Dai, JN; Zong, Y; Zhong, LM; Li, YM; Zhang, W; Bian, LG; Ai, QL; Liu, YD; Sun, J;Lu, D. Gastrodin inhibits expression of inducible NO synthase, cyclooxygenase-2 and proinflammatory cytokines in cultured LPS-stimulated microglia viaMAPK pathways. PLoS One 6:e21891; 2011.

[38] Liu, HT; Huang, P; Ma, P; Liu, QS; Yu, C; Du, YG. Chitosan oligosaccharidessuppress LPS-induced IL-8 expression in human umbilical vein endothelialcells through blockade of p38 and Akt protein kinases. Acta Pharmacol. Sin32:478–486; 2011.

[39] Zingarelli, B; Sheehan, M; Wong, HR. Nuclear factor-κB as a therapeutic targetin critical care medicine. Crit. Care Med 31:S105–111; 2003.

[40] Ling, M; Li, Y; Xu, Y; Pang, Y; Shen, L; Jiang, R; Zhao, Y; Yang, X; Zhang, J; Zhou, J;Wang, X; Liu, Q. Regulation of miRNA-21 by reactive oxygen species-activatedERK/NF-κB in arsenite-induced cell transformation. Free Radic. Biol. Med52:1508–1518; 2012.

[41] Han, DW; Lee, MH; Kim, HH; Hyon, SH; Park, JC. Epigallocatechin-3-gallateregulates cell growth, cell cycle and phosphorylated nuclear factor-κB inhuman dermal fibroblasts. Acta Pharmacol. Sin 32:637–646; 2011.

[42] Meng, HT; Zhu, L; Ni, WM; You, LS; Jin, J; Qian, WB. Triptolide inhibits theproliferation of cells from lymphocytic leukemic cell lines in association withdownregulation of NF-κB activity and miR-16-1*. Acta Pharmacol. Sin 32:503–-511; 2011.

[43] Fang, JP; Liu, Y; Li, J; Liao, WF; Hu, YH; Ding, K. A novel small molecule, HK-156, inhibits lipopolysaccharide-induced activation of NF-κB signaling andimproves survival in mouse models of sepsis. Acta Pharmacol. Sin 33:1204–-1216; 2012.

[44] Checker, R; Patwardhan, RS; Sharma, D; Menon, J; Thoh, M; Bhilwade, HN;Konishi, T; Sandur, SK.; Schisandrin, B exhibits anti-inflammatory activitythrough modulation of the redox-sensitive transcription factors Nrf2 and NF-κB. Free Radic. Biol. Med 53:1421–1430; 2012.

[45] Gilad, E; Wong, HR; Zingarelli, B; Virág, L; O'Connor, M; Salzman, AL; Szabó, C.Melatonin inhibits expression of the inducible isoform of nitric oxide synthasein murine macrophages: role of inhibition of NFkappaB activation. FASEB J12:685–693; 1998.

[46] Aznavour, N; Cendres-Bozzi, C; Lemoine, L; Buda, C; Sastre, JP; Mincheva, Z;Zimmer, L; Lin, JS. MPTP animal model of parkinsonism: dopamine cell deathor only tyrosine hydroxylase impairment? A study using PET imaging,autoradiography, and immunohistochemistry in the cat CNS Neurosci. Ther18:934–941; 2012.

C. Liu et al. / Free Radical Biology and Medicine 65 (2013) 658–666666