International Immunopharmacology 38 (2016) 357–366

Contents lists available at ScienceDirect

International Immunopharmacology

j ourna l homepage: www.e lsev ie r .com/ locate / in t imp

Platelet-activating factor and hydrogen peroxide exert a dualmodulatoryeffect on the transcription of LXRα and its target genes inhuman neutrophils

María E. Reyes-Quiroz a,1, Gonzalo Alba a,1, Javier Sáenz a, Isabel Geniz c, Juan Jiménez a, José Martín-Nieto d,Consuelo Santa-María b, Francisco Sobrino a,⁎a Department of Medical Biochemistry, Molecular Biology and Immunology, University of Seville, Seville, Spainb Department of Biochemistry and Molecular Biology, University of Seville, Seville, Spainc Andalucia Health Services, Seville, Spaind Department of Physiology, Genetic and Microbiology, University of Alicante, Alicante, Spain

Abbreviations: LXRα, liver X receptor; PAF, platelet-acperoxide; ABCA1, ATP-binding cassette transporter A1transporter G1; SREBP1c, sterol regulatory element-bindintor κB; ROS, reactive oxygen species; ERK1/2, extracellula2; GAPDH, glyceraldehyde 3-phosphate dehyddichlorodihydrofluorescein diacetate; DFP, diisopropyl fl1,2,4-triazole; aPAF, 1-O-hexadecyl-2-acetyl-strimethyl)hexanolamine; HO-1, hemoxygenase 1.⁎ Corresponding author at: Department ofMedical Bioch

Immunology, University of Seville, Av. Sánchez Pizjuán 4,E-mail address: fsobrino@us.es (F. Sobrino).

1 These authors contributed equally to this work.

http://dx.doi.org/10.1016/j.intimp.2016.05.0011567-5769/© 2016 Published by Elsevier B.V.

a b s t r a c t

a r t i c l e i n f o

Article history:Received 21 October 2015Received in revised form 29 April 2016Accepted 3 May 2016Available online 25 June 2016

Liver X receptors (LXRs) are ligand-activated nuclear receptors involved mainly in the regulation of cholesterolmetabolism in many organs, including liver and intestine, as well as in macrophages and neutrophils. Besides,both anti-inflammatory and pro-inflammatory properties have been ascribed to LXRs. The effect of the inflamma-tory condition on the expression of LXRα and its target genes has not been previously addressed in human neu-trophils. We have described that platelet-activating factor (PAF) and hydrogen peroxide (H2O2) are potent pro-inflammatorymediators that link the haemostatic and innate immune systems. In this workwe report that H2O2

at low doses (1 pM-1 μM) exerts an inhibitory effect on TO901317-inducedmRNA expression of LXRα and of itstarget genes encoding the ATP-binding cassette (ABC) transporters ABCA1 and ABCG1, and the sterol regulatoryelement-binding protein 1c (SREBP1c). However, an opposite behaviour, i.e., a transcription-enhancing effect,was found at higher H2O2 doses (100–500 μM) on most of these genes. A similar dual effect was observedwhen the pro-inflammatory molecule PAF was used. Interestingly, H2O2 production separately elicited by10 nM PAF or 1 μM H2O2 was similarly low, and analogously, H2O2 production levels elicited by 5 μM PAF or100 μM H2O2 were similarly high when they were compared. On the other hand, low doses of PAF or H2O2 in-duced phosphorylation of extracellular signal-regulated kinases 1 and 2 (ERK 1/2) and NF-κB activation, Howev-er, PAF or H2O2 at high doses did not produce changes in NF-κB activation levels. In summary, our results showthat H2O2, either exogenous or PAF-induced, exerts a dual regulation onmRNA expression of LXRα and its targetgenes.

© 2016 Published by Elsevier B.V.

Keywords:Liver X receptorHuman neutrophils, PAFNF-κBERK1 and 2

1. Introduction

Liver X receptors (LXRs) are nuclear receptors activated byoxysterols which regulate cholesterol homeostasis by stimulating theexpression of a set of genes involved in virtually all aspects of

tivating factor; H2O2, hydrogen; ABCG1, ATP-binding cassettegprotein 1c; NF-κB, nuclear fac-r signal-regulated kinases 1 androgenase; DCFH-DA, 2′,7′-uorophosphate; ATZ, 3-amino-n-glycerol-3-phospho-(N,N,N-

emistry,Molecular Biology and41009, Seville, Spain.

cholesterol transport and metabolism, such as those encoding theATP-binding cassette (ABC) transporters ABCA1 and ABCG1, and thesterol regulatory element-binding protein 1c (SREBP1c) [1,2]. Thesegenes are expressed in multiple organs, including liver and intestine,as well as in pro-inflammatory cells such as macrophages [3] and neu-trophils [4]. Moreover, they are involved in the modulation of other bi-ological processes, such as fatty acid synthesis and metabolism, glucosehomeostasis, steroidogenesis and neuronal homeostasis [5].

LXRs may function to integrate metabolic and immune signals [3],and are associated with the modulation of the inflammatory responseand innate immunity [6]. LXRα activation has also been related to a pos-itive regulation of inflammation in human macrophages [7] and miceneutrophils [8]. However, other authors have reported that LXRα in-hibits the transcription of pro-inflammatory genes by antagonizingthe positive actions of transcription factors such as nuclear factor κB(NF-κB) [9,10], and of inducible nitric oxide synthase, cyclooxygenase2 and interleukin 6 in response to bacterial infection or

358 M.E. Reyes-Quiroz et al. / International Immunopharmacology 38 (2016) 357–366

lipopolysaccharide stimulation [11]. A novel role of LXRs in the oxida-tive stress response has been recently proposed, because of positive reg-ulation of the expression of antioxidant enzymes by these receptorsrecently found [11].

Platelet-activating factor (PAF) is a powerful pro-inflammatoryagent that links the haemostatic and innate immune systems. This fac-tor is modestly expressed under normal physiological conditions, andsome immune-system cell types such as neutrophils andmonocytes re-lease significant amounts of PAF under particular conditions, such as ox-idative stress [12]. PAF contributes to increasing permeability of thevascular endothelium, induces the release of reactive oxygen species(ROS) and contributes to low-density lipoprotein oxidation [13]. Inthis context, PAF has been proposed as a key factor and initial triggerin atherosclerosis. PAF is extracellularly released almost immediatelyin response to pro-inflammatory stimuli and induces the activation ofNF-κB, a transcription factor that acts as a key modulator of the levelsof many proteins involved in the inflammatory process, such as recep-tors, enzymes and cytokines [13]. Many of the genes whose expressionis inhibited by LXRs are established targets of NF-κB signalling [6], andactivation by PAF of this transcription factor is exerted through the gen-eration of ROS [12]. Previous results from our group indicate that PAF atlow doses acts as an NF-κB activator in a manner dependent of phos-phorylation of the MAP kinase, ERK1/2 [14].

H2O2 is also produced by pro-inflammatory and vascular cells andinduces oxidative stress, which may contribute to atherosclerosis andendothelial dysfunction. Although its bactericidal function during in-flammation is well established, a possible regulatory role of H2O2, eitherby promoting or inhibiting inflammation, has been proposed [15,16]. Atlow concentrations, H2O2 diffuses into a cell and reacts rapidly and spe-cifically with a limited number of molecular targets, before it is reducedby anti-oxidant defencemechanisms. Its diffusibility and relative stabil-ity make H2O2 a candidate to constitute an intercellular signalling mol-ecule, analogously to nitric oxide [17]. In contrast, H2O2 at a sufficientlyhigh concentration may act as an oxidant able to alter the overall redoxpotential of the cell. It has been proposed that H2O2 exerts a fine-tuningregulatory role, this involving both a pro-inflammatory control loopthat increases pathogen removal, and an anti-inflammatory controlloop that prevents an exacerbated, harmful inflammatory response [18].

Neutrophils constitute the principal circulating cells involved in in-nate immunity, and increasing evidence suggests an implication ofthis cell type in atherosclerosis [19,20]. In this work we have foundthat PAF and H2O2 tested at different doses exert a dual modulatoryrole on the expression of LXRα and its target genes in human neutro-phils. We have addressed as well whether these effects are mediatedby ERK1/2 phosphorylation and/or NF-κB activation.

2. Materials and methods

2.1. Materials

TO901317 and PAF were purchased from Cayman Chemical (AnnArbor, MI, USA), and RPMI 1640 medium from Biomedia (Boussens,France). Dextran T-500 and lymphocyte separation medium (Ficoll-Paque) were obtained from GE Healthcare (Barcelona, Spain), andpolyvinylidene difluoride (PVDF) membranes from Pall (Madrid,Spain). Rabbit polyclonal antibodies against phosphorylated (Thr202/Tyr204) ERK1/2 were obtained from New England Biolabs (Beverly,MA), and IκBα from Santa Cruz Biotechnology. Mouse monoclonal anti-body to glyceraldehyde 3-phosphate dehydrogenase (GAPDH)was pur-chased from Chemicon International (Madrid, Spain). Mousemonoclonal antibody to phospho-p65 was purchased from Cell Signal-ing Technology, Inc. (Danvers, Massachusetts, USA). Horseradish perox-idase (HRP)-conjugated goat anti-rabbit and anti-mouse IgG wereproducts of Promega (Madison,WI, USA). Human IL-8 ELISAKitwas pur-chased from Raybiotech, Inc. (Norcross, Georgia, USA). 22-R-OH-choles-terol, protease inhibitor cocktail, H2O2, 2′,7′-dichlorodihydrofluorescein

diacetate (DCFH-DA), diisopropyl fluorophosphate (DFP) and 3-amino-1,2,4-triazole (ATZ) were obtained from Sigma-Aldrich (Madrid,Spain). NF-κB SN50 peptide was purchased from Enzo Life Sciences(Ann Arbor, MI, USA). Fetal calf serum, L-glutamine, streptomycin, peni-cillin and amphotericin B were obtained from BioWhittaker (Basel,Switzerland).

2.2. Ethics statement

Peripheral venous bloodwas drawn fromhealthy volunteers follow-ing standardized protocols approved by the Research Ethics Committeeof the Hospital Virgen Macarena, Universidad de Sevilla. This investiga-tion was designed and conducted according to the ethical principles formedical research stated in the Declaration of Helsinki.

2.3. Isolation and culture of human neutrophils

Human peripheral blood neutrophils were isolated from fresh hepa-rinized blood of human donors by Dextran T-500 sedimentation,followed by Ficoll-Paque gradient centrifugation and hypotonic lysis ofresidual erythrocytes [21]. For experiments, 2 × 107 neutrophils werecultured in RPMI 1640 completemedium from Life Technologies (Rock-ville, MD, USA) containing 10% (v/v) heat-inactivated bovine serum,2 mM L-glutamine, 100 μg/mL streptomycin, 100 IU/mL penicillin and250 ng/mL amphotericin B. When indicated, 20 mM ATZ was includedin the assays. Before all stimulations, neutrophil suspensions were pre-incubated at room temperature with 1 mM DFP (to minimize proteoly-sis) for 5 min [22].

2.4. RT-PCR quantitation of mRNA levels

Real-time PCR analysis was performed in the presence of SYBRGreen using the ABI Prism 7300 sequence detection system from Ap-plied Biosystems (Foster City, CA, USA) under the specific thermo-cycler conditions recommended by the manufacturer for the primersused. PCR reactions were performed in triplicate. Each sample wasalso analyzed for β-actin transcript levels to normalize for RNA inputamounts. For the relative quantification of gene expression, the compar-ative threshold cycle method was used as described in the ABI Prism7700 User Bulletin 2 [23]. Primers were designed with Primer Express(Applied Biosystems) and synthesized by Roche Diagnostics (Barcelona,Spain), and their sequenceswere as follows: LXRα: forward, 5′-AAGCCCTGCATGCCTACGT-3′; reverse, 5′-TGCAGACGCAGTGCAAACA-3′; ABCA1:forward, 5′-CCCTGTGGAATGTACCTATGTG-3′;reverse, 5′-GAGGTGTCCCAAAGATGCAA-3′; ABCG1: forward, 5′-CAGTCGCTCCTTAGCACCA-3′; re-verse, 5’TCCATGCTCGGACTCTCTG-3′; SREBP1c: forward, 5′-CATGTCTTCGATGTCGGTCAG-3′; reverse, 5′-TCCTGTTGCCCATATGAAATCA-3′; β-actin: forward, 5’CCAGCTCACCATGGATGATG-3′; reverse, 5′-TGCCGGAGCCGTTGTC.

2.5. Measurement of ROS production

The level of intracellular peroxides, taken as an indicator of the intra-cellular redox state, was determined by treating cells with DCFH-DA.This is a nonpolar compound that is converted into a membrane-impermeable non-fluorescent polar derivative (DCFH) by cellular ester-ase after its incorporation into cells. The trapped DCFH is rapidly oxi-dized to fluorescent 2′,7′-dichlorofluorescein (DCF) by intracellularH2O2 [24]. 106 cells were resuspended in 1 mL PBS 1×, supplementedwith 5 mM glucose and 20 mM ATZ, and preincubated with 10 μMDCFH-DA at 37 °C for 45 min. Then, cells were washed twice and resus-pended in 1 mL of PBS for flow cytometry analysis. Further, neutrophilswere incubatedwith orwithout the indicated concentrations of H2O2 orPAF for 3 min. Emission of the trapped, oxidized DCF by these cells(events), pretreated with or without stimulus, was assayed using aflow cytometer (fluorescence-activated cell sorting, FACS, Becton-

Fig. 1. PAF exerts opposite concentration-dependent effects on TO901317-inducedmRNAexpression of LXRα and its target genes in human neutrophils. Cells were incubated at37 °C with or without PAF at the indicated concentrations for 1 h. Then, 1 μM TO901317was added and cells were incubated at 37 °C for further 5 h (A). In panel B the cellswere treated as in A, except that they were preincubated at 37 °C with or without10 nM aPAF for 30 min prior to PAF addition. In panel C the cells were incubated at37 °C with 20 mM ATZ for 30 min, and then with or without PAF at the indicatedconcentrations for 1 h. Thereafter, 1 μM TO901317 was added and cells were incubatedat 37 °C for further 5 h. Control cells were cultured for the same times as treated cells,but without any additions. Finally, the mRNA levels of genes indicated in each panelwere quantitated by RT-PCR, corrected for differences in β-actin mRNA levels andexpressed as fold induction. Values are plotted as the mean ± SEM, n = 3. Statisticaldata: *p b 0.05 for PAF- and TO901317-treated versus PAF-untreated; **p b 0.05 foraPAF, PAF- and TO901317-treated versus aPAF-untreated. ◊p b 0.05 for ATZ-, andTO901317- with or without PAF-treated (panel A) versus ATZ-untreated (panel C).

359M.E. Reyes-Quiroz et al. / International Immunopharmacology 38 (2016) 357–366

Dickinson, Mountain View, CA, USA). Data were analyzed using theFlowJo software (http://www.flowjo.com).

2.6. Western blotting analysis of phosphorylated ERK1/2, IκB and phospho-p65

107 neutrophils/mL were rinsed once with ice-cold PBS, resuspend-ed in a lysis solution containing 50mMTris-HCl (pH 7.4), 10mMEDTA,50 mM NaF, 10% glycerol, 1% Triton X-100, protease inhibitor cocktail,and kept on ice for 30 min. Then the cells were disrupted by sonicationon ice and, after centrifugation at 12,000 ×g for 5 min at 4 °C, proteinconcentration in the supernatant was determined by the Bradfordmethod [25], using BSA as a standard. Proteins were boiled in Laemmliloading buffer, resolved by SDS-PAGE (10% polyacrylamide) and trans-ferred to PVDF membranes as previously described [26]. The blotswere probed, without need of prior blocking, with rabbit polyclonalanti-phospho-ERK1/2, at a 1:1000 dilution or rabbit polyclonal anti-IκB, at a 1:2000 dilution or mouse monoclonal anti-phospho-p65, at a1:1000 dilution in PBS plus 0.5% BSA and 0.02% Tween-20 [27]. Thereaf-ter, HRP-conjugated antibodies to rabbit or mouse IgG were used at a1:5000 or 1:20,000 dilution, respectively, in PBS plus 0.5% casein, anddetectionwas carried out by enhanced chemiluminescence [21]. To ver-ify even protein loading, the blots were subsequently stripped and re-probedwithmousemonoclonal antibodies against GAPDH at a 5000 di-lution. Band intensities were measured by scanning densitometry anal-ysis using the Scion Image software (Frederick, MD).

2.7. ELISA quantitation of IL-8 release

Cell culture supernatants were collected after treatments, and thelevels of secreted IL-8 were quantified using the Human IL-8 ELISA Kit(Raybiotech, Norcross, GA). Plates were read on a Wallac 1420 Victor2spectrofluorometer (Perkin Elmer, Madrid, Spain).

2.8. Statistical analysis

mRNA levelsmeasured by real-time PCR are expressed as fold induc-tion relative to untreated cells (mean ± SEM from a minimum of 3 in-dependent experiments performed with similar results). Protein levelsmeasured from Western blots were normalized to GAPDH levels andare expressed in arbitrary units. The results were statistically analyzedusing the Statgraphics Plus 5.0 software (Manugistic Inc., Rockville,MD) by means of ANOVA and the paired Student's t-test.

3. Results

3.1. PAF exerts opposite concentration-dependent effects on TO901317-induced mRNA expression of LXRα and its target genes in humanneutrophils

Initial experiments were designed to analyze the PAF dose effect onTO901317-induced mRNA expression of LXRα and its target genes inhuman neutrophils. Relative quantification of LXRα and its targetgenes mRNA levels in human neutrophils treated with TO901317 (asynthetic LXRα agonist) by means of real time RT-PCR confirmed thatthis agonist induced these genes expression (Fig. 1A). Treatment of neu-trophils with low PAF concentrations (10 nM) inhibited the TO901317-induced LXRα and its related genes mRNA expression (Fig. 1A). At PAF100 nM this effect was only observed in LXRαmRNA expression. Thesedata confirmed previous results reported by our group [14]. Interesting-ly, a new and opposite effect was observed when neutrophils were pre-incubated with high doses of PAF (1–5 μM), which clearly enhancedTO901317-induced LXRα and its target genes mRNA expression(Fig. 1A). Similar results were observedwhen 22R-OH-cholesterol (nat-ural LXRα agonist) was used instead of TO901317 (data not shown). Noeffect was observed with PAF alone (data not shown).

In order to assess whether this effect wasmediated through the PAFreceptor (PAFR), the antagonist 1-O-hexadecyl-2-acetyl-sn-glycerol-3-phospho-(N,N,N-trimethyl)hexanolamine (aPAF) was used [28]. Pre-treatment of neutrophils with this compound prior to PAF addition de-creased the enhancing effect of high PAF concentration (5 μM) onTO901317-induced mRNA expression of LXRα and its target genes inhuman neutrophils (Fig. 1B). As our group has previously reported,aPAFwas also able to reverse the inhibitory effect of low PAF concentra-tion [14].

3.2. Effect of 3-amino-1,2,4-triazole (ATZ) onmRNA expression of LXRα andits target genes in PAF-treated human neutrophils

Recently, our group has described a relationship between the inhib-itory effect of PAF at low concentration on the LXRα expression and therelated genes and cellular redox status in human neutrophils [14].Moreover, PAF-mediated H2O2 production has been reported in otherimmune-system cells, such asmacrophages [29].We thus set to addressthe H2O2 effect on PAF-induced inhibition of LXRα and its target genes

360 M.E. Reyes-Quiroz et al. / International Immunopharmacology 38 (2016) 357–366

in human neutrophils. To this effect, we promoted an increase of endog-enous H2O2 levels by using a catalase inhibitor, 3-amino-1,2,4-triazole(ATZ), and its effect was analyzed on TO901317-inducedmRNA expres-sion of LXRα and its target genes which is represented in Fig. 1C. Whencomparing these results with those shown in Fig. 1A we noted: i) ATZitself exerted a significant positive effect on TO901317-induced ABCA1and ABCG1 mRNA expression, ii) in the presence of ATZ, PAF at a lowdose (10 nM) exerted a greater inhibition (although not statistically sig-nificant) on TO901317-induced mRNA expression of the four genes,than that obtained without ATZ treatment iii) however, the inhibitoryeffect of 100 nM PAF on mRNA expression of LXRα, shown in Fig. 1A,was prevented and a lesser inhibitory effect was observed with thistreatment on ABCG1 mRNA expression, iv) ATZ clearly enhanced thehigh-PAF concentration (1–5 μM) effect on mRNA expression of thefour genes, although, this increase was not statistically significant in1 μM PAF treatment on SREBP1c mRNA expression. These results sug-gest that the H2O2 may play an important role on PAF effects of LXRαand its target genes in human neutrophils.

3.3. Effect of PAF on intracellular H2O2 production

As above it has been indicated PAF treatment elicits H2O2 productionin macrophages [29]. Subsequent experiments were designed with theaim of evaluating the PAF effects on intracellular H2O2 production inneutrophils. Cells were incubated with different PAF concentrationsand intracellular H2O2 levels were analyzed by flow cytometry. We ob-served that PAF induced H2O2 production was dose-dependent inhuman neutrophils (Fig. 2A and B). Furthermore, as shown in Fig. 2C,PAF effect was exerted through acting on its receptor, the PAFR, sinceaPAF pretreatment reversed the PAF-induced H2O2 increase, shown inFig. 2A.

On the other hand, we performed a H2O2 calibration curve and themost significant doses are shown in Figs. 2D–F. Interestingly, 1 μMH2O2 (Fig. 2D) and 10 nM PAF treatment (Fig. 2A) induced a similarlow intracellular H2O2 level. Likewise, another calibration point,

Fig. 2. Effect of PAF on intracellular H2O2 production. Cells were preincubated at 37 °C with 10concentrations of PAF (A–C) or H2O2 (D–F) for 3 min. In panel C, the cells were additionaladdition. The levels of intracellularly oxidized DCF were measured by flow cytometry. The sharepresentative of three independent experiments yielding similar results.

100 μM H2O2, (Fig. 2E), induced a similar intracellular H2O2 as 5 μMPAF treatment (Fig. 2B).

3.4. Effect of catalase on mRNA expression of LXRα and its target genes inPAF-treated human neutrophils

In an opposite approach, catalase at 400 or 800 U/mL was used inorder to decrease endogenous H2O2 levels and study its effect on PAF-induced modulation of LXRα and its target genes mRNA expression. Inneutrophils pre-incubated with 800 U/mL catalase the inhibitory effectof 10 nM PAF on TO901317-induced LXRα mRNA expression was par-tially reduced (Fig. 3A) and totally reverted on the other target genes(Fig. 3B–D). However, 400 U/mL catalase had no effect on TO901317-induced mRNA expression of the four genes (Fig. 3A–D). In contrast,neither 400 nor 800 U/mL catalase had any influence on the enhancereffect of high PAF concentration (5 μM) on their mRNA expression(Fig. 3A–D).

To elucidate the non-effect of catalase in high doses of PAF, we ana-lyzed the catalase inhibitory effect over H2O2 intracellular levels pro-duced by PAF. As Fig. 4 shows, we observed that the catalasetreatment totally reverted the 10 nM PAF-induced H2O2 intracellular(Fig. 4A and B), but it was unable to inhibit the H2O2 intracellular levelsproduced by PAF at a high concentration (Fig. 4C and D). This effectmayexplain that catalase had no influence on the positive effect of high PAFconcentration (5 μM) on their mRNA expression as observed above inFig. 3A–D.

3.5. Effect of H2O2 on LXRα and its target genes mRNA expression in humanneutrophils

We next analyzed the effect of exogenous H2O2 addition onTO901317-induced expression of LXRα and its target genes in humanneutrophils. Fig. 5 shows that H2O2 also exerted opposite effects de-pending on whether cells were incubated with low or high concentra-tions, in similarity to PAF treatment. Thus, it was found that low H2O2

μM DCFH-DA and 20 mM ATZ for 30 min. Then the cells were treated with the indicatedly preincubated at 37 °C with 10 nM aPAF, a PAFR antagonist, for 30 min prior to PAFded areas correspond to untreated controls, and unshaded areas to treated cells. Data are

Fig. 3. Effect of catalase onmRNA expression of LXRα and its target genes in PAF-treated human neutrophils. Cells were cultured at 37 °C with or without catalase (Cat.) at the indicatedconcentrations for 2 h (A–D). Then, theywere treated or notwith PAF at the indicated concentrations for 1 h and thereafter 1 μMTO901317was added and neutrophils were incubated forfurther 5 h at 37 °C (A–D). Control cells were cultured for the same times as treated cells, but without any additions. Finally, the mRNA levels of genes indicated in each panel werequantitated by RT-PCR, corrected for differences in β-actin mRNA levels and expressed as fold induction. Values are plotted as the mean ± SEM, n = 3. Statistical data: *p b 0.05 forPAF- and TO901317-treated versus PAF-untreated; **p b 0.05 for PAF- and TO901317- plus catalase-treated versus catalase-untreated.

Fig. 4. Effect of catalase on intracellular H2O2 production PAF-induced. Cells were preincubated at 37 °Cwith 10 μMDCFH-DA and 20mMATZ for 30min. Then the cells were treatedwith800U/mL catalase for 2 h (B andD). Further, neutrophilswere incubatedwith 10nM(A and B) or 5 μM(CandD) PAF for 3min. The levels of intracellularly oxidizedDCFweremeasuredbyflow cytometry. The shaded areas correspond to untreated controls, and unshaded areas to treated cells. Data are representative of three independent experiments yielding similar results.

361M.E. Reyes-Quiroz et al. / International Immunopharmacology 38 (2016) 357–366

Fig. 5. Effect of H2O2 on LXRα and its target genesmRNA expression in human neutrophils. Cells were incubated at 37 °C in the absence (white bars) or presence (grey bars) of 20mMATZfor 30min. Then the cells were treatedwith the indicated concentrations of H2O2 for 1 h (A–D), and thereafter 1 μMTO901317was added and the cells were incubated at 37 °C for further5 h (A–D). Control cells were cultured for the same times as treated cells, butwithout any additions. Finally, themRNA levels of genes indicated in each panel were quantitated by RT-PCR,corrected for differences in β-actin mRNA levels and expressed as fold induction. Values are plotted as the mean ± SEM, n= 3. Statistical data: *p b 0.05 for ATZ- and TO901317-treatedversus ATZ-untreated; **p b 0.05 for H2O2- and TO901317-treated versus H2O2-untreated; ◊p b 0.05 for ATZ-, H2O2- and TO901317-treated versus H2O2-untreated.

362 M.E. Reyes-Quiroz et al. / International Immunopharmacology 38 (2016) 357–366

concentrations (1 pM–1 μM) elicited a clear inhibition of TO901317-induced mRNA expression of LXRα and its target genes (Fig. 5A–D,white bars). In contrast, when neutrophils were incubated with highH2O2 concentrations (above 100 μM), TO901317-induced LXRα andSREBP1c mRNA expression was enhanced (Fig. 5A and D, white bars),

Fig. 6. NF-κB involvement in PAF modulation of mRNA expression of LXRα and its target gene30 min (A–D). Then the cells were incubated with or without PAF at the indicated concentratiofurther 5 h at 37 °C (A–D). Control cells were cultured for the same times as treated cells, buquantitated by RT-PCR, corrected for differences in β-actin mRNA levels and expressed as folPAF- and TO901317-treated versus PAF-untreated; **p b 0.05 for SN50-, PAF- and TO901317-t

although it was unchanged in the case of ABCA1 and ABCG1 (Fig. 5Band C, white bars). However, when neutrophils were pre-incubatedwith ATZ, which should stabilize H2O2 levels, the results were clearer(Fig. 5A–D, grey bars), in the sense that low H2O2 concentrations(1 pM–1 μM) inhibited TO901317-induced expression of the four

s in human neutrophils. Cells were preincubated at 37 °C with or without 15 μM SN50 forns for 1 h, and thereafter 1 μM TO901317 was added and neutrophils were incubated fort without any additions. Finally, the mRNA levels of genes indicated in each panel wered induction. Values are plotted as the mean ± SEM, n = 3. Statistical data: *p b 0.05 forreated versus SN50-untreated.

363M.E. Reyes-Quiroz et al. / International Immunopharmacology 38 (2016) 357–366

genes and high H2O2 concentrations (100–500 μM) were found to en-hance it.

3.6. NF-κB involvement in PAF andH2O2modulation of mRNA expression ofLXRα and its target genes in human neutrophils

PAF is known to act as an activator of NF-κB in vivo and in vitro bypromoting the phosphorylation of NF-κB-bound IκB inhibitory proteins[14]. These then become ubiquitinated, releasing NF-κB which thentranslocates to the nucleus to activate or repress its target genes [15].In order to study whether the PAF effect on transcription of LXRα andits target genes could be mediated by NF-κB, neutrophils were pre-incubated with SN50, a specific NF-κB inhibitor. As shown in Fig. 6A–D, SN50 was able to cancel the inhibitory effect of low PAF concentra-tions (i.e., 10 nM) on TO901317-induced mRNA expression of thesegenes, except for SREBP1c (Fig. 6A–D). However, the NF-κB inhibition(by SN50 treatment) did notmodify positive effect by high PAF concen-tration (i.e., 5 μM) (Fig. 6A–D). To note, SN50 itself did not exert any sig-nificant effect on either basal levels or TO901317-induced mRNA levelsof LXRα and its target genes (Fig. 6A–D).

H2O2 has also been reported to activate NF-κB by promoting phos-phorylation of IκB [30]. This also led us to analyze whether NF-κB inhi-bition modified H2O2 effect as observed above. A similar behaviour tothat described above for PAF was found for H2O2. SN50was able to can-cel the inhibitory effect of low H2O2 concentrations (i.e., 1 μM) onTO901317-induced mRNA expression of these genes, except forSREBP1c (Fig. 7A–D). However, the high H2O2 concentration(i.e., 100 μM) positive effect on this process was not reversed by SN50treatment (Fig. 7A–D).

Fig. 7. NF-κB involvement in H2O2 modulation of mRNA expression of LXRα and its target gene30min. Then, the cells were incubated in the absence or presence of H2O2 at the indicated concincubated for further 5 h at 37 °C (A–D). Control cells were cultured for the same times as treatewere quantitated by RT-PCR, corrected for differences in β-actin mRNA levels and expressed asH2O2- and TO901317-treated versus H2O2-untreated; **p b 0.05 for SN50-, H2O2- and TO90131

In addition, the effects of low and high PAF or H2O2 concentrationson NF-κB activation were assessed by analyzing IκBα levels. Fig. 8Aand B show that low PAF (10 nM) or H2O2 (1 μM) concentrations elicit-ed IκBα degradation, but high PAF (5 μM) or H2O2 (100 μM) concentra-tions had no effect. On the other hand, the NF-κB activation was alsoanalyzed by NF-κB p65 phosphorylation levels. In this experiment weobserved that lowPAForH2O2 concentrations induced a clear p65 phos-phorylation (Fig. 8C) but the treatment with high PAF or H2O2 concen-trations were unable to induce p65 phosphorylation (Fig. 8C).Therefore, high concentrations of these two compounds do not appearto activate NF-κB in human neutrophils (Fig. 8A–C).

3.7. Possible involvement of MAP kinases in the modulation by PAF or H2O2

of mRNA expression of LXRα and its target genes in human neutrophils

LXRα is phosphorylated at Ser198 by MAP kinases [31], and ourgroup has recently described that the low-PAF negative effect onmRNA expression of LXRα and its target genes is mediated by ERK1/2activation in human neutrophils [14]. Subsequent experiments werethus addressed to analyze the ERK1/2 involvement in transcriptionalmodulation of these genes by low and high PAF or H2O2 concentration.Fig. 9A and B show that ERK1/2 phosphorylation was induced by low(10 nM) and high PAF (5 μM) or H2O2 (1 μM and 100 μM)concentrations.

3.8. Effect of PAF or H2O2 treatment on IL-8 release in human neutrophils

To study the effects of different PAF orH2O2 concentrations on a neu-trophils function, we analyzed the effect of (5 μM) PAF or (100 μM)

s in human neutrophils. Cells were preincubated at 37 °C with or without 15 μM SN50 forentrations for 1 h (A–D), and thereafter 1 μM TO901317 was added and neutrophils wered cells, but without any additions. Finally, themRNA levels of genes indicated in each panelfold induction. Values are plotted as the mean± SEM, n= 3. Statistical data: *p b 0.05 for7-treated versus SN50-untreated.

Fig. 8. Possible involvement of NF-κB in themodulation by PAF or H2O2 ofmRNA expression of LXRα and its target genes in human neutrophils. Cells were preincubatedwith 1 μg/mL LPSfor 30min (C). Then neutrophils were incubated with PAF (A, C) or H2O2 (B, C) at the indicated concentrations for 3 min. The levels of IκBα (A, B) or phosphorylated p65 (phospho-p65)(C) and GAPDH (A–C)were assessed byWestern blotting and expressed as ratio IκB or phospho-p65/GAPDH levels. Control cells were cultured for the same timeswithout any additions.Each panel is representative of a set of three experiments yielding similar results and values are plotted as the mean ± SEM. Statistical data: *p b 0.01 for PAF- or -H2O2 or -LPS treatedversus untreated cells.

364 M.E. Reyes-Quiroz et al. / International Immunopharmacology 38 (2016) 357–366

H2O2 on IL-8 release in human neutrophils stimulated with or withoutN-formyl-methionyl-leucyl-phenylalanine (fMLP), a known neutro-phils immune response stimulator [32]. Fig. 9C shows that high PAF orH2O2 concentration treatment reversed fMLP-mediated IL-8 release. Incontrast, previous results have shown that IL-8 release was enhancedby treatment with a low concentration of PAF [14] or H2O2 [33].

4. Discussion

The effects of oxidants as modulators of nuclear receptor activityhave been studied by a number of groups, often with conflicting resultsbecause of the existence of a multi-step and complex regulation to oxi-dative stress [34,35]. The present study proposes a possible signallingpathway whereby a PAF-elicited increase in H2O2 levels modulatesLXRα-mediated transcriptional regulation.

Treatment of neutrophils with low PAF concentrations (10 nM)inhibited the TO901317-induced LXRα and related genes mRNA ex-pression. An opposite effect was found in neutrophils pretreated withhigh doses of PAF (1–5 μM). These effects were mediated through thePAFR. H2O2 production of by PAF has been reported in other immune-system cells, such asmacrophages [29]. Hereinwe describe intracellularH2O2 production in neutrophils induced by PAF which this is dose-dependent. The alteration of H2O2 intracellular levels either by enhanc-ing its content by ATZ treatment (a catalase inhibitor) [36] or by directlylowering its levels with catalase, modified the effects of PAF onTO901317-induced LXRα and related genes mRNA expression.

A rise in theH2O2 intracellular levels exerted: i) a slight inhibition in-crease produced by PAF at low doses (10 nM) on TO901317-inducedmRNA expression of the four genes, ii) a clear enhancement increaseproduced by PAF at high doses (1–5 μM) on TO901317-induced mRNAexpression of the four genes. However, a decrease in the H2O2

intracellular levels reverted the inhibition produced by low doses ofPAF, but it had no influence on the effects of high doses of PAF. Thismight be because the catalase treatments were unable to reduce thehigh intracellular H2O2 levels produced by PAF 5 μM as was verified byflow cytometry.

Similar to PAF treatment, the exogenous H2O2 addition to humanneutrophils also exerted dual effects on TO901317-induced expressionof LXRα and its target genes depends on whether cells were incubatedwith H2O2 at low or high concentrations. Our results suggest that theLXRα signalling pathway is modulated in response to variations in in-tracellular H2O2 levels, and support studies from other groups showingthat H2O2 is able to elicit dual effects depending on its concentration indifferent cellular functions [37–39].

NF-κB is one of the first molecules to be expressed and activated byROS, and is known to be related to PAF [40] and H2O2 [30] signal trans-duction mechanisms. The NF-κB inhibition reverted the effect of bothPAF and H2O2 in the range of low concentrations. In contrast, in thepresence of high concentrations of both PAF andH2O2, theNF-κB inhibi-tion did not elicit changes on transcription of genes studied. Moreover,NF-κB activation was assessed by analyzing IκBα levels and by p65phosphorylation levels and a NF-κB inhibition at low PAF doses wasfound, confirming previouswork from our lab [14] and others [41]. Sim-ilar behaviour was found for as described by others [33]. Interestingly,neither PAF nor H2O2 at a high dose produced NF-κB stimulation. Theno NF-κB stimulation by both molecules at high doses could representa mechanism potentially able to reduce the inflammatory process, ashas been suggested by other authors [37,42].

In addition to NF-κB, our results identify ERK1/2 as a candidate MAPkinase to act as a mediator on LXRα signalling. The activation of stresskinases such as ERK1/2 is thought to play a key role as mediators ofH2O2-associated effects. In this context, it has been recently

Fig. 9. Possible involvement of MAP kinases in the modulation ofmRNA expression of LXRα and its target genes by PAF or H2O2 in human neutrophils, and effects of these on IL-8 release.Cells were incubatedwith PAF (A, C) or H2O2 (B, C) at the indicated concentrations for 3min (A, B) 5 h (C). Then cells were stimulatedwith 100 nM fMLP (C). The levels of phosphorylatedERK1/2 (pERK1/2) and GAPDH were assessed byWestern blotting and expressed as ratio pERK1/2/GAPDH levels (A, B). Levels of IL-8 released were analyzed by ELISA. (C). Control cellswere cultured for the same times without any additions. Each panel is representative of a set of three experiments yielding similar results and values are plotted as the mean ± SEM.Statistical data: *p b 0.01 for PAF- or H2O2- or fMLP- treated versus untreated cells. **p b 0.01 for PAF- or H2O2- plus fMLP-treated versus PAF- or H2O2-untreated.

365M.E. Reyes-Quiroz et al. / International Immunopharmacology 38 (2016) 357–366

demonstrated that ERK1/2 becomes activated in response to H2O2 (by20-fold), and molecules acting downstream of phosphorylated ERK1/2[43].

We propose that the dual actions of either PAF or H2O2 on intracellu-lar LXRα signalling are determined by the net balance between ERK1/2andNF-κB activation. For instance, 10 nMPAF or 1 μMH2O2 significantlyelicited phosphorylation of ERK1/2 to low levels, and this induced a sig-nificantNF-κB activation [44]. As a result of the sumof these two factors,LXRα and its target genes expression become inhibited, suggesting thatactivation of both factors ERK1/2 and NF-κB are necessary to such effect.However, phosphorylation of ERK1/2 without NF-κB activation, an ef-fect observed in this work at concentrations of 5 μM PAF or 100 μMH2O2, might be enough to increase TO901317-induced LXRα pathwayexpression in human neutrophils. Other authors have reported thathigh H2O2 levels induce ERK1/2 phosphorylation in neutrophils [45].Moreover, ERK1/2 acts as a negative regulator of NF-κB inmacrophagestreated with H2O2 [46]. The protective effect of PAF-induced high levelsH2O2 is in keeping with the enhanced expression of hemoxygenase 1(HO-1) observed by us (data not shown). Some studies have describedthat HO-1 exerts an important protective function as an inhibitor ofhypoxia-induced vasoconstrictory and pro-inflammatory pathways[47], as well as an antioxidant role [48].

On the other hand we have described that dual LXRα modulationmay affect a proinflammatory property in neutrophils, as IL-8 release.High PAF or H2O2 levels inhibited fMLP-mediated IL-8 release. However,low PAF or H2O2 doses induced IL-8 release as described by us [14] andothers [33].

In conclusion, this study suggests that intracellular H2O2 levelsmight be a modulator of the LXRα signalling pathway. To our knowl-edge, the present work is the first to demonstrate the dual effects of

H2O2 on the LXRα pathway and its target genes. Further research onthe modulation of this pathway exerted by the PAF/H2O2 interplaymay lead to the development of novel anti-oxidative therapies to selec-tively counteract oxidative stress, and thereby the symptoms of cardio-vascular diseases.

Disclosures

The authors have declared that no competing interests exist.

Authorship contributions

Participated in research design: Reyes-Quiroz, Alba, Santa-María,Jiménez, Martín-Nieto, Pintado and Sobrino. Conducted experiments:Reyes-Quiroz, Alba, Sáenz, Geniz and Santa-María. Performed data anal-ysis: Reyes-Quiroz, Alba, Sáenz, Santa-María and Sobrino.Wrote or con-tributed to the writing of the manuscript: Reyes-Quiroz, Alba, Geniz,Santa-María, Martín-Nieto, and Sobrino.

Acknowledgements

M.E.R-Q was supported by a fellowship from the Asociación VirgenMacarena, Hospital Universitario Virgen Macarena, Sevilla. G.A. wassupported by fellowships from the Ministerio de Educación y Ciencia(BFU2006-13802) and the Consejería de Innovación, Ciencia y Empresa,Junta de Andalucía (P08-CVI-03550). This work was funded by grantsfrom the latter (P06-CTS-01936 and P08-CVI-03550) to F.S., and fromthe Consejería de Salud, Junta de Andalucía (CS 0116/2007) to E.P. Weare indebted to Margarita Rodríguez Borrego for her technical assis-tance. To Remedios Ramirez in memoriam.

366 M.E. Reyes-Quiroz et al. / International Immunopharmacology 38 (2016) 357–366

References

[1] J.J. Repa, D.J. Mangelsdorf, The role of orphan nuclear receptors in the regulation ofcholesterol homeostasis, Annu. Rev. Cell Dev. Biol. 16 (2000) 459–481, http://dx.doi.org/10.1146/annurev.cellbio.16.1.459.

[2] A.C. Calkin, P. Tontonoz, Transcriptional integration of metabolism by the nuclearsterol-activated receptors LXR and FXR, Nat. Rev. Cell Biol. 13 (2012) 213–224,http://dx.doi.org/10.1038/nrm3312; 10.1038/nrm3312.

[3] S.B. Joseph, P. Tontonoz, LXRs: new therapeutic targets in atherosclerosis? Curr.Opin. Pharmacol. 3 (2) (2003) 192–197.

[4] G. Alba, M.E. Reyes, C. Santa-Maria, R. Ramirez, I. Geniz, J. Jimenez, et al., Transcrip-tion of liver X receptor is down-regulated by 15-deoxy-delta(12,14)-prostaglandinJ(2) through oxidative stress in human neutrophils, PLoS One 7 (2012), e42195,http://dx.doi.org/10.1371/journal.pone.0042195.

[5] E. Viennois, K. Mouzat, J. Dufour, L. Morel, J.M. Lobaccaro, S. Baron, Selective liver Xreceptor modulators (SLiMs): what use in human health? Mol. Cell. Endocrinol. 351(2012) 129–141, http://dx.doi.org/10.1016/j.mce.2011.08.036.

[6] M. Pascual-García, A.F. Valledor, Biological roles of liver X receptors in immune cells,Arch. Immunol. Ther. Exp. (Warsz) 60 (2012) 235–249, http://dx.doi.org/10.1007/s00005-012-0179-9.

[7] C. Fontaine, E. Rigamonti, A. Nohara, P. Gervois, E. Teissier, J.C. Fruchart, et al., Liver Xreceptor activation potentiates the lipopolysaccharide response in human macro-phages, Circ. Res. 101 (2007) 40–49, http://dx.doi.org/10.1161/CIRCRESAHA.106.135814.

[8] H. Korf, S. Vander Beken, M. Romano, K.R. Steffensen, B. Stijlemans, J.A. Gustafsson,et al., Liver X receptors contribute to the protective immune response againstMyco-bacterium tuberculosis in mice, J. Clin. Invest. 119 (2009) 1626–1637, http://dx.doi.org/10.1172/JCI35288.

[9] C.K. Glass, S. Ogawa, Combinatorial roles of nuclear receptors in inflammation andimmunity, Nat. Rev. 6 (2006) 44–55, http://dx.doi.org/10.1038/nri1748.

[10] S. Ghisletti, W. Huang, K. Jepsen, C. Benner, G. Hardiman, M.G. Rosenfeld, et al., Co-operative NCoR/SMRT interactions establish a corepressor-based strategy for inte-gration of inflammatory and anti-inflammatory signaling pathways, Genes Dev. 23(2009) 681–693, http://dx.doi.org/10.1101/gad.1773109.

[11] H. Gong, J. He, J.H. Lee, E. Mallick, X. Gao, S. Li, et al., Activation of the liver X receptorprevents lipopolysaccharide-induced lung injury, J. Biol. Chem. 284 (2009)30113–30121, http://dx.doi.org/10.1074/jbc.M109.047753.

[12] C. Penna, E. Bassino, G. Alloatti, Platelet activating factor: the good and the bad in theischemic/reperfused heart, Exp. Biol. Med. (Maywood) 236 (2011) 390–401, http://dx.doi.org/10.1258/ebm.2011.010316.

[13] A.D. Tselepis, M.J. Chapman, M. John Chapman, Inflammation, bioactive lipids andatherosclerosis: potential roles of a lipoprotein-associated phospholipase A2, plate-let activating factor-acetylhydrolase, Atheroscler. Suppl. 3 (2002) 57–68, http://dx.doi.org/10.1016/S1567-5688(02)00045-4.

[14] M.E. Reyes-Quiroz, G. Alba, C. Santa-María, J. Saenz, I. Geniz, J. Jiménez, et al.,Platelet-activating factor downregulates the expression of liver X receptor-α andits target genes in human neutrophils, FEBS J. 281 (3) (2014) 970–982.

[15] I. Jaspers, W. Zhang, A. Fraser, J.M. Samet, W. Reed, Hydrogen peroxide has opposingeffects on IKK activity and IκBα breakdown in airway epithelial cells, Am. J. Respir.Cell Mol. Biol. 24 (2001) 769–777, http://dx.doi.org/10.1165/ajrcmb.24.6.4344.

[16] S.H. Korn, E.F. Wouters, N. Vos, Y.M. Janssen-Heininger, Cytokine-induced activationof nuclear factor-kappa B is inhibited by hydrogen peroxide through oxidative inac-tivation of IkappaB kinase, J. Biol. Chem. 276 (38) (2001) 35693–35700, http://dx.doi.org/10.1074/jbc.M104321200.

[17] J.R. Stone, T. Collins, The role of hydrogen peroxide in endothelial proliferative re-sponses, Endothelium 9 (4) (2002) 231–238.

[18] V. Oliveira-Marques, L. Cyrne, H.S. Marinho, F. Antunes, A quantitative study of NF-αB activation by H2O2: relevance in inflammation and synergy with TNF-1, J.Immunol. 178 (2007) 3893–3902.

[19] S. Sela, R. Mazor, M. Amsalam, C. Yagil, Y. Yagil, B. Kristal, Primed polymorphonucle-ar leukocytes, oxidative stress, and inflammation antecede hypertension in theSabra rat, Hypertension 44 (2004) 764–769, http://dx.doi.org/10.1161/01.HYP.0000144480.10207.34.

[20] M. Bakele, M. Joos, S. Burdi, N. Allgaier, S. Pöschel, B. Fehrenbacher, et al., Localiza-tion and functionality of the inflammasome in neutrophils, J. Biol. Chem. 289 (8)(2014) 5320–5329.

[21] M. Carballo, G. Márquez, M. Conde, J. Martín-Nieto, J. Monteseirín, J. Conde, et al.,Characterization of calcineurin in human neutrophils. Inhibitory effect of hydrogenperoxide on its enzyme activity and on NF-kappaB DNA binding, J. Biol. Chem. 274(1999) 93–100.

[22] C. Gilbert, E. Rollet-Labelle, P.H. Naccache, Preservation of the pattern of tyrosinephosphorylation in human neutrophil lysates. II. A sequential lysis protocol for theanalysis of tyrosine phosphorylation-dependent signalling, J. Immunol. Methods261 (2002) 85–101.

[23] G. Alba, R. El Bekay, P. Chacon, M.E. Reyes, E. Ramos, J. Olivan, et al., Hemeoxygenase-1 expression is down-regulated by angiotensin II and under hyperten-sion in human neutrophils, J. Leukoc. Biol. 84 (2008) 397–405, http://dx.doi.org/10.1189/jlb.0108035.

[24] D.A. Bass, J.W. Parce, L.R. Dechatelet, P. Szejda, M.C. Seeds, M. Thomas, Flow cyto-metric studies of oxidative product formation by neutrophils: a graded responseto membrane stimulation, J. Immunol. 130 (4) (1983) 1910–1917.

[25] M.M. Bradford, A rapid and sensitive method for the quantitation of microgramquantities of protein utilizing the principle of protein-dye binding, Anal. Biochem.72 (1976) 248–254.

[26] R. El Bekay, M. Alvarez, M. Carballo, J. Martin-Nieto, J. Monteseirin, E. Pintado, et al.,Activation of phagocytic cell NADPH oxidase by norfloxacin: a potential mechanismto explain its bactericidal action, J. Leukoc. Biol. 71 (2002) 255–261.

[27] M.A. Mansfield, Rapid immunodetection on polyvinylidene fluoride membraneblots without blocking, Anal. Biochem. 229 (1995) 140–143, http://dx.doi.org/10.1006/abio.1995.1391.

[28] I. Sethy-Coraci, L.W. Crock, S.C. Silverstein, PAF-receptor antagonists, lovastatin, andthe PTK inhibitor genistein inhibit H2O2 secretion by macrophages cultured onoxidized-LDL matrices, J. Leukoc. Biol. 78 (5) (2005) 1166–1174, http://dx.doi.org/10.1189/jlb.0205101.

[29] M. Rouis, F. Nigon, M.J. Chapman, Platelet activating factor is a potent stimulant ofthe production of active oxygen species by humanmonocyte-derived macrophages,Biochem. Biophys. Res. Commun. 156 (1988) 1293–1301.

[30] Y. Takada, A. Mukhopadhyay, G.C. Kundu, G.H. Mahabeleshwar, S. Singh, B.B.Aggarwal, Hydrogen peroxide activates NF-kappa B through tyrosine phosphoryla-tion of I kappa B alpha and serine phosphorylation of p65: evidence for the involve-ment of I kappa B alpha kinase and Syk protein-tyrosine kinase, J. Biol. Chem. 278(26) (2003) 24233–24241.

[31] M. Chen, M.N. Bradley, S.W. Beaven, P. Tontonoz, Phosphorylation of the liver X re-ceptors, FEBS Lett. 580 (2006) 4835–4841, http://dx.doi.org/10.1016/j.febslet.2006.07.074.

[32] M.A. Panaro, V. Mitolo, Cellular responses to FMLP challenging: a mini-review,Immunopharmacol. Immunotoxicol. 21 (1999) 397–419, http://dx.doi.org/10.3109/08923979909007117.

[33] S. Yanagisawa, A. Koarai, H. Sugiura, T. Ichikawa, M. Kanda, R. Tanaka, et al., Oxida-tive stress augments toll-like receptor 8 mediated neutrophilic responses in healthysubjects, Respir. Res. 10 (2009) 50, http://dx.doi.org/10.1186/1465-9921-10-50.

[34] Y. Wang, C. Li, K. Cheng, R. Zhang, K. Narsinh, S. Li, et al., Activation of liver X recep-tor improves viability of adipose-derived mesenchymal stem cells to attenuatemyocardial ischemia injury through TLR4/NF-κB and Keap-1/Nrf-2 signaling path-ways, Antioxid. Redox Signal. 21 (2014) 2543–2557, http://dx.doi.org/10.1089/ars.2013.5683.

[35] L. Villacorta, M.T. Garcia-Barrio, Y.E. Chen, Transcriptional regulation of peroxisomeproliferator-activated receptors and liver X receptors, Curr. Atheroscler. Rep. 9(2007) 230–237.

[36] C. Chen, G.E. Hennig, H.E. Whiteley, J.E. Manautou, Protection against acetamino-phen hepatotoxicity by clofibrate pretreatment: role of catalase induction, J.Biochem. Mol. Toxicol. 16 (5) (2002) 227–234.

[37] J.W. Zmijewski, E. Lorne, X. Zhao, Y. Tsuruta, Y. Sha, G. Liu, et al., Antiinflammatoryeffects of hydrogen peroxide in neutrophil activation and acute lung injury, Am. J.Respir. Crit. Care Med. 179 (8) (2009) 694–704, http://dx.doi.org/10.1164/rccm.200806-851OC.

[38] S. Iwakami, H. Misu, T. Takeda, M. Sugimori, S. Matsugo, S. Kaneko, et al.,Concentration-dependent dual effects of hydrogen peroxide on insulin signal trans-duction in H4IIEC hepatocytes, PLoS One 6 (2011) 2–11, http://dx.doi.org/10.1371/journal.pone.0027401.

[39] A.C. Matias, S.M. Marinho, L. Cyrne, E. Herrero, F. Antunes, Biphasic modulation offatty acid synthase by hydrogen peroxide in Saccharomyces cerevisiae, Arch.Biochem. Biophys. 515 (2011) 107–111.

[40] A. Borthakur, S. Bhattacharyya, W.A. Alrefai, J.K. Tobacman, K. Ramaswamy, P.K.Dudeja, Platelet-activating factor-induced NF-kappaB activation and IL-8 produc-tion in intestinal epithelial cells are Bcl10-dependent, Inflamm. Bowel Dis. 16(2010) 593–603, http://dx.doi.org/10.1002/ibd.21092.

[41] J.H. Choi, W.J. Chung, S.J. Han, H.B. Lee, I.W. Choi, H.K. Lee, et al., Selective involve-ment of reactive oxygen intermediates in platelet-activating factor-mediated activa-tion of NF-kappaB, Inflammation 24 (2000) 385–398.

[42] S. Mitra, E. Abraham, Participation of superoxide in neutrophil activation and cyto-kine production, Biochim. Biophys. Acta 1762 (8) (2006) 732–741.

[43] T.H. Truong, K.S. Carroll, Redox regulation of protein kinases, Crit. Rev. Biochem.Mol. Biol. 48 (4) (2013) 332–356.

[44] X.C. Bai, D. Lu, J. Bai, H. Zheng, Z.Y. Ke, X.M. Li, et al., Oxidative stress inhibits osteo-blastic differentiation of bone cells by ERK and NF-kappaB, Biochem. Biophys. Res.Commun. 314 (1) (2004) 197–207.

[45] D. Strassheim, K. Asehnoune, J.S. Park, J.Y. Kim, Q. He, D. Richter, et al., Modulation ofbone marrow-derived neutrophil signaling by H2O2: disparate effects on kinases,NF-κB, and cytokine expression, Am. J. Phys. Cell Physiol. 286 (2004) C683–C692.

[46] M. Jaramillo, M. Olivier, Hydrogen peroxide induces murine macrophage chemokinegene transcription via extracellular signal-regulated kinase- and cyclic adenosine 5′-monophosphate (cAMP)-dependent pathways: involvement of NF-kappa B, activa-tor protein 1, and cAMP response element, J. Immunol. 169 (12) (2002) 7026–7038.

[47] T. Minamino, H. Christou, C. Hsieh, L.M. Y., V. Dhawan, N.G. Abraham, et al., Targetedexpression of heme oxygenase-1 prevents the pulmonary inflammatory and vascu-lar responses to hypoxia, Proc. Natl. Acad. Sci. 98 (2001) 8798–8803.

[48] S. Gonzales, M.J. Perez, J.C. Perazzo, M.L. Tomaro, Antioxidant role of hemeoxygenase-1 in prehepatic portal hypertensive rats, World J. Gastroenterol. 12(2006) 4149–4155.