Archives of Biochemistry and Biophysics 477 (2008) 253–258

Contents lists available at ScienceDirect

Archives of Biochemistry and Biophysics journal homepage: www.elsevier.com/locate/yabbi

Mangiferin inhibits cyclooxygenase-2 expression and prostaglandin E2 production in activated rat microglial cells Harsharan S. Bhatia a,1, Eduardo Candelario-Jalil a,b,*,1, Antonio C. Pinheiro de Oliveira a, Olumayokun A. Olajide a, Gregorio Martínez-Sánchez c, Bernd L. Fiebich a,d,* a

Neurochemistry Research Group, Department of Psychiatry, University of Freiburg Medical School, Hauptstrasse 5, D-79104 Freiburg, Germany Department of Neurology, University of New Mexico Health Sciences Center, 915 Camino de Salud NE, MSC10 5620, Albuquerque, NM 87131, USA c Department of Pharmacology (CEIEB-IFAL), University of Havana, Cuba d VivaCell Biotechnology GmbH, Ferdinand-Porsche-Street 5, D-79211 Denzlingen, Germany b

a r t i c l e

i n f o

Article history: Received 1 April 2008 and in revised form 13 June 2008 Available online 28 June 2008 Keywords: Mangiferin COX-2 Prostaglandin E2 Neuroinflammation Microglia Oxidative stress

a b s t r a c t Mangiferin, a naturally occurring glucosylxanthone, has potent antioxidant and anti-inflammatory properties, as demonstrated in several reports. However, very limited information is available on the effects of this natural polyphenol on microglial activation. Thus, the aim of this study was to examine whether mangiferin is able to reduce prostaglandin E2 (PGE2) and 8-iso-prostaglandin F2a (8-iso-PGF2a) production by lipopolysaccharide (LPS)-activated primary rat microglia. Microglial cells were stimulated with 10 ng/ ml of LPS in the presence or absence of different concentrations of mangiferin (1–50 lM). After 24 h incubation, culture media were collected to measure the production of PGE2 and 8-iso-PGF2a using enzyme immunoassays. Protein levels of cyclooxygenase (COX)-1 and COX-2 were studied by immunoblotting after 24 h of incubation with LPS. Mangiferin potently reduced LPS-induced PGE2 synthesis and the formation of 8-iso-PGF2a. Interestingly, mangiferin dose-dependently reduced LPS-induced COX-2 protein synthesis without modifying COX-2 transcription. This was due to a decrease in COX-2 transcript stability. However, mangiferin did not modify LPS-mediated phosphorylation of p38 mitogen-activated protein kinase (p38 MAPK), a key factor involved in enhancing COX-2 mRNA stability and COX-2 translation in primary microglia. Mangiferin had no effects on LPS-induced expression of inducible nitric oxide synthase (iNOS) or TNF-a production. Taken together, results from the present study indicate that mangiferin is able to limit microglial activation, in terms of attenuation of PGE2 production, free radical formation and reduction in COX-2 synthesis induced by LPS. These data suggest that modulation of microglial activation might contribute to the mechanism of cerebral protection by mangiferin. Ó 2008 Elsevier Inc. All rights reserved.

Mangiferin is a naturally occurring xanthone glucoside (2-Cb-D-gluco-pyranosyl-1,3,6,7-tetrahydroxy-xanthone) (Fig. 1), and it is particularly abundant in the bark of Mangifera indica L. (mango tree) [1,2]. This natural polyphenol exhibits potent anti-inflammatory [3–6], immunomodulatory [7,8], and antioxidant properties [6,9–13]. The pharmacology of mangiferin has recently gained great attention owing to its protective function against oxidative injury to various tissues, including the brain [4–6,14–16]. Very recent studies, both in vitro and in vivo, have found that mangiferin confers significant protection against excitotoxic insults, cerebral ischemia [17], and neurotoxicity associated to

* Corresponding authors. Fax: +1 505 272 6692 (E. Candelario-Jalil), +49 761 270 6917) (B.L. Fiebich). E-mail addresses: [email protected] (E. Candelario-Jalil), bernd.fi[email protected] (B.L. Fiebich). 1 These authors contributed equally to this study. 0003-9861/$ - see front matter Ó 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.abb.2008.06.017

1-methyl-4-phenyl-pyridine ion (MPP+) [18], the active metabolite of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP, an agent that induces Parkinson’s syndrome in humans and animals) [18]. Microglial cells constitute the resident macrophage population of the central nervous system, and are actively involved in many physiological and neuropathological processes [19,20]. Microglial activation is a key pathological factor in brain injury during neurodegenerative processes, and following trauma, ischemia and neurotoxicity [21–23]. Very limited information is available on the effects of mangiferin on microglial activation. However, several studies have been published in recent years showing the efficacy of mangiferin in reducing the production of pro-inflammatory mediators by peripheral macrophages [5,6]. The present study aims to investigate the effects of the natural polyphenol mangiferin on measures of microglial activation in a well-standardized in vitro model of primary rat microglia exposed to LPS. Our findings indicate for the first time that mangiferin is a

254

H.S. Bhatia et al. / Archives of Biochemistry and Biophysics 477 (2008) 253–258

the microglial culture was >98% as previously determined by immunofluorescence and cytochemical analysis [31]. PGE2 enzyme immunoassay Fig. 1. Chemical structure of mangiferin (2-C-b-D-gluco-pyranosyl-1,3,6,7-tetrahydroxy-xanthone).

potent inhibitor of PGE2 production and free radical formation through a mechanism that involves, in part, reduction in the synthesis of COX-2 protein in LPS-activated microglia.

Supernatants were harvested, centrifuged at 10,000g for 10 min and levels of PGE2 in the media were measured by enzyme immunoassay (EIA) (Assay Designs Inc., Ann Arbor, MI, USA; distributed by Biotrend, Cologne, Germany) according to the manufacturer’s instructions. Standards from 39 to 2500 pg/ml were used. The sensitivity of this assay was 36.2 pg/ml. Immunoblotting

Methods Reagents and antibodies Mangiferin (from Mangifera indica L. bark) and LPS (from Salmonella typhimurium) were obtained from Sigma–Aldrich (Taufkirchen, Germany). LPS was resuspended in sterile phosphate buffered saline (PBS2; Cell Concepts, Umkirch, Germany) as 5 mg/ ml stock, and was used at a final concentration of 10 ng/ml in the culture medium. Mangiferin was dissolved in dimethyl sulfoxide (DMSO). Solvent concentration in the culture media was maintained at less than 0.1%. Mangiferin, used at the given concentrations, does not affect the viability of the cells as observed through a luminescent kit (Promega, Mannheim, Germany), which measures metabolic ATP levels (data not shown). Primary antibodies against COX-1 (M-20) and COX-2 (M-19) were from Santa Cruz Biotechnology (Heidelberg, Germany). Rabbit polyclonal antibodies against phospho-specific p38 MAPK (Thr 180/Tyr 182) and total p38 MAPK were obtained from Cell Signaling (supplied by New England Biolabs, Frankfurt, Germany). Rabbit polyclonal antibody against inducible nitric oxide synthase (iNOS) was purchased from Biomol GmbH (Hamburg, Germany). The rabbit antibody against actin was from Sigma (Taufkirchen, Germany). Preparation of primary cultures of rat microglia Primary mixed glial cell cultures were established from cerebral cortices of one-day neonatal Sprague–Dawley rats as described in details in our previous reports [24–30]. Briefly, forebrains were minced and gently dissociated by repeated pipetting in PBS and filtered through a 70-lm cell strainer (Falcon). Cells were collected by centrifugation (1000g, 10 min), resuspended in Dulbecco’s modified Eagle’s medium (DMEM) containing 10% fetal calf serum (Biochrom AG, Berlin, Germany) and antibiotics (40 U/ml penicillin and 40 lg/ml streptomycin, both from PAA Laboratories, Coelbe, Germany), and cultured on 10-cm cell culture dishes (5  105 cells/ plate, Falcon) in 5% CO2 at 37 °C. Medium was prepared taking extreme care to avoid LPS contamination [31]. Floating microglia were harvested every week (between 2 and 7 week) and re-seeded into 75 cm2 culture flask (for Western blots) or 24-well plates (for 8-iso-PGF2a, PGE2 and TNF-a estimation) to give pure microglial cultures. The following day, cultures were washed to remove non-adherent cells, and fresh medium was added. The purity of

For COX-1, COX-2 and iNOS immunoblotting, microglial cells were left untreated or treated with LPS (10 ng/ml) in the presence or absence of mangiferin (1–50 lM) for 24 h. For phospho-p38 MAPK and total p38 MAPK Western blots, cells were pre-treated for 30 min with mangiferin and then stimulated for 30 min with LPS (10 ng/ml). We had previously found that phosphorylation of p38 MAPK is maximal after 30 min [28]. At the end of each experiment, cells were washed with phosphate-buffered saline (PBS) and lysed in 1.3 sodium dodecyl sulfate (SDS)-containing sample buffer without 1,4-dithio-DL-threitol (DTT) or bromophenol blue containing 100 lM orthovanadate [32]. Lysates were homogenized by repeated passage through a 26-gauge needle. Protein contents were measured using the bicinchoninic acid method (BCA protein determination kit from Pierce, distributed by KFC Chemikalien, Munich, Germany) according to the manufacturer’s instructions. Bovine serum albumin (BSA, Sigma) was used as a standard. Before electrophoresis, bromophenol blue and DTT (final concentration, 10 mM) were added to the samples. For Western blotting, 60 lg of total protein from each sample were subjected to SDS–PAGE (polyacrylamide gel electrophoresis) under reducing conditions. Proteins were then transferred onto polyvinylidene fluoride (PVDF) membranes (Millipore, Bedford, MA, USA) by semi-dry blotting. The membranes were blocked overnight at 4 °C using Rotiblock (Roth, Karlsruhe, Germany) and for another hour at room temperature before incubation with the antibodies. Primary antibodies were goat anti-COX-2 (1:500), goat anti-COX-1 (1:500), rabbit anti-phospho p38 MAPK (1:500), rabbit anti-p38 MAPK (1:500), rabbit anti-iNOS (1:3000) and rabbit anti-actin. Primary antibodies were diluted in Tris-buffered saline (TBS) containing 0.1% Tween 20 (TBST) and 1% BSA. Membranes were incubated with the corresponding primary antibody for 2 h at room temperature. After extensive washing (three times for 15 min each in TBST), proteins were detected with horseradish peroxidase-coupled rabbit anti-goat IgG (Santa Cruz, 1:100,000 dilution) or goat anti-rabbit IgG (Amersham, 1:25,000 dilution) using chemiluminescence (ECL) reagents (Amersham Pharmacia Biotech, Freiburg, Germany). Quantification of the Western blots was performed using ScanPack 3.0 software (Biometra, Göttingen, Germany). Equal protein loading and transfer were assessed by subjection of each sample to a Western blot for actin (rabbit anti-actin IgG, diluted 1:5000). All Western blot experiments were carried out at least three times. Measurement of 8-iso-PGF2a

2 Abbreviations used: LPS, lipopolysaccharide; PGE2, prostaglandin E2; COX, cyclooxygenase; EIA, enzyme immunoassay; DMEM, Dulbecco’s modified Eagle’s medium; DTT, 1,4-dithio-DL-threitol; BSA, bovine serum albumin; ANOVA, analysis of variance; 8-iso-PGF2a, 8-iso-prostaglandin F2a; MAPK, mitogen-activated protein kinase; MPP+, 1-methyl-4-phenyl-pyridine ion; MPTP, 1-methyl 4-phenyl 1,2,3,6-tetrahydropyridine; PBS, phosphate-buffered saline; PVDF, polyvinylidene fluoride; SDS, sodium dodecyl sulfate; TBS, Tris-buffered saline; TBST, Tris-buffered saline containing 0.1% Tween 20; iNOS, inducible nitric oxide synthase; TNF-a, tumor necrosis factor-a.

Microglial cells were pre-treated for 30 min with different concentrations of mangiferin (see Results and Figures). After 30 min pre-stimulation, cells were incubated with 10 ng/ml LPS for 24 h. Control experiments consisted of cells treated only with solvent without LPS. Supernatants were harvested and the levels of 8-isoPGF2a (IUPAC nomenclature: 15-F2t-IsoP) were measured by an enzyme immunoassay according to the manufacturer’s instructions

H.S. Bhatia et al. / Archives of Biochemistry and Biophysics 477 (2008) 253–258

(Cayman Chemicals, Ann Arbor, MI, USA). The standards were used in the interval of 3.9–500 pg/ml (detection limit of 5 pg/ml). RNA isolation and reverse transcription–polymerase chain reaction (RT–PCR) Total RNA was extracted using the guanidine isothiocyanate method [33]. For RT-PCR, 2 lg of total RNA was reverse transcribed using Moloney Murine Leukemia Virus (M-MLV) reverse transcriptase (Promega, Mannheim, Germany), RNase Inhibitor rRNasinÒ (Promega), dNTP master mix (Invitek, Berlin, Germany) and random hexamer primers (Promega). PCR was carried out using Taq DNA polymerase (Promega), dNTP master mix (Invitek, Berlin, Germany) and the following primers for rat COX-2 (forward, 50 -TGC GAT GCT CTT CCG AGC TGT GCT-30 and reverse, 50 -TCA GGA AGT TCC TTA TTT CCT TTC-30 , annealing temperature 55 °C, 35 cycles, amplicon size: 479 bp). Equal equilibration was determined using rat b-actin primers (forward: 50 -ATG GAT GAC GAT ATC GCT-30 , reverse: 50 -ATG AGG TAG TCT GTC AGG T-30 , 48 °C, 30 cycles, product length: 569 bp) or primers for S12 from rat (forward: 50 -ACG TCA ACA CTG CTC TAC A-30 , reverse: 50 -CTT TGC CAT AGT CCT TAA C30 , 56 °C, 30 cycles, product length: 312 bp). PCR products were separated electrophoretically on a 2% agarose gel. Potential contamination by genomic DNA was controlled by omitting reverse transcriptase and using primers for the housekeeping genes (b-actin or S12) in the subsequent PCR amplification. Only RNA samples showing no bands after this procedure were used for further investigation. Primers were designed using the Primer3 software developed by the Whitehead Institute for Biomedical Research (http:// frodo.wi.mit.edu/primer3/input.htm), and synthesized through an in-house facility (Dr. Gabor Igloi, Institute for Biology III, Freiburg, Germany). PCR analysis was performed after 4 h of stimulation with LPS. COX-2 transcript stability analysis The stability of COX-2 mRNA in mangiferin-treated and control cells was measured by incubation with actinomycin D (1 lg/ml) to block transcription. Cells were pre-stimulated for 30 min with mangiferin (50 lM). LPS was then added for 2 h. After LPS stimulation, actinomycin D was added for 30 min, media were changed to fresh media containing LPS in the presence of either mangiferin or the p38 MAPK inhibitor SB202190 (10 lM; used as a positive control). After 1 h, total RNA was isolated, and the disappearance of COX-2 mRNA abundance was determined by RT-PCR analysis, as mentioned before.

255

Results and discussion The ability of mangiferin to reduce the production of PGE2 was initially investigated using primary microglial cells exposed to LPS. Activation of microglial cells produced a dramatic increase in the production of PGE2 after 24 h of incubation with LPS (Fig. 2). Pretreatment with different doses of mangiferin led to a significant reduction in the formation of PGE2, as measured in the supernatants 24 h following LPS stimulation (Fig. 2). The inhibitory effect of mangiferin on PGE2 production was dependent on its dose, as shown in Fig. 2. Having found that micromolar concentrations of mangiferin significantly attenuated LPS-induced PGE2 formation, we expanded our study to investigate the effects of mangiferin on protein levels of both COX isoforms (COX-1 and COX-2) in rat microglia. We have demonstrated the important contribution of COX-2 to the formation of PGE2 by microglia [25,28], and have reported that its expression is absent under normal conditions, but dramatically induced by LPS [25,28,29]. Since mangiferin has been previously shown to reduce the expression of several inflammatory mediators, including COX-2 expression in peripheral macrophages [3,4,7], we next studied the effects of this natural compound on LPS-mediated COX-2 protein levels in microglia using immunoblotting analysis. Findings from three independent experiments consistently showed that mangiferin produced a dose-dependent significant reduction in COX-2 protein immunoreactivity induced by LPS (Fig. 3A and B). However, mangiferin did not significantly alter levels of COX-1 (Fig. 3A). Given that mangiferin possess antioxidant properties [6,9–11], we asked whether attenuation of free radical formation could contribute to its effects on microglial activation. Levels of 8-iso-PGF2a were measured using a specific enzyme immunoassay in the conditioned media 24 h after stimulation with LPS, in the presence or absence of different doses of mangiferin (1–50 lM). As shown in Fig. 4, mangiferin significantly reduced the formation of 8-isoPGF2a. However, this effect seems not to be dependent on the dose, which is in contrast to what was found for PGE2 levels (Fig. 2). Next, we investigated the effect of different concentrations of mangiferin on LPS-induced COX-2 mRNA expression. Cells were pre-treated for 30 min with mangiferin (1–50 lM) and LPS was added for 4 h. The expression of COX-2 mRNA was analyzed using RT-PCR analysis. Our results indicated that mangiferin did not significantly reduce LPS-mediated COX-2 transcription in activated

Determination of tumor necrosis factor-a (TNF-a) in microglial supernatants Primary microglial cells were pre-treated for 30 min with mangiferin. After 30 min pre-stimulation, cells were incubated with 10 ng/ml LPS for 24 h. The concentrations of TNF-a released into the media were measured by a specific TNF-a ELISA kit following manufacturer’s recommendations (eBioscience, distributed by NatuTec GmbH, Frankfurt, Germany). The standards were used in the interval of 16–2000 pg/ml (detection limit of 16 pg/ml). Data analysis Data from at least three independent experiments were used for analysis. Original data were converted into percentage-values of LPS control and mean ± SEM were calculated. Values were compared using t-test (two groups) or one-way ANOVA with post-hoc Student–Newman–Keuls test (multiple comparisons). The level of significance was set at p < 0.05.

Fig. 2. Mangiferin dose-dependently inhibits LPS-induced PGE2 production in primary rat microglial cells. The amounts of PGE2 in the culture medium were determined using an enzyme immunoassay after 24 h incubation with LPS. Mangiferin was added 30 min before stimulating the cells with LPS. Each column and error bar represents the mean ± SEM of four independent experiments. Asterisks indicate significant difference between the treatments. p < 0.05,  p < 0.01 and p < 0.001 with respect to LPS control (one-way ANOVA followed by the Student–Newman–Keuls post-hoc test).

256

H.S. Bhatia et al. / Archives of Biochemistry and Biophysics 477 (2008) 253–258

Fig. 3. (A) Immunoblot analysis of protein levels of COX-2, COX-1 and actin in LPSactivated microglia treated with different concentrations of mangiferin (1–50 lM). Cell extracts were prepared after 24 h of incubation with LPS, and subjected to immunoblot analysis using specific antibodies for each protein. (B) Quantitative densitometric analysis of COX-2 protein expression normalized to actin loading control. Mangiferin potently reduced COX-2 protein expression induced by LPS.  p < 0.05, p < 0.01 and p < 0.001 with respect to LPS control. Bars represent mean ± SEM of three independent experiments. Statistical analysis was performed using one-way ANOVA followed by Student–Newman–Keuls post-hoc test.

microglial cells (Fig. 5). A trend toward an increase in COX-2 mRNA was observed in microglial cells stimulated with LPS and treated with the highest dose of mangiferin (50 lM). However, this effect was not statistically significant. We then performed an additional experiment to investigate whether the reduction in COX-2 protein synthesis by mangiferin is due to alterations in COX-2 mRNA stability. We blocked transcription with actinomycin D, and cells were incubated with LPS in the presence or absence of mangiferin (50 lM) or the p38 MAPK inhibitor SB202190 (10 lM). Data shown in Fig. 6 indicate that mangiferin reduced COX-2 transcript stability. However, this effect did not reach a statistical level of significance, although we did ob-

Fig. 4. Effect of mangiferin on 8-iso-prostaglandin F2a (8-iso-PGF2a) production in response to 10 ng/ml LPS. 8-iso-PGF2a was determined in the culture medium 24 h following stimulation of microglial cells with LPS alone, or in combination with mangiferin at the given concentrations. p < 0.05 with respect to LPS alone. Histograms represent mean ± SEM of four independent experiments. Statistical analysis was performed using one-way ANOVA followed by Student–Newman– Keuls post-hoc test.

Fig. 5. (A) Representative photomicrographs showing RT-PCR products of COX-2 and b-actin mRNAs. Microglial cells were treated with LPS (10 ng/ml) in the absence or presence of different concentrations of mangiferin (1–50 lM). The mRNA expression levels were tested for each condition. COX-2 is undetectable in untreated microglial cells, while its expression is dramatically increased in the presence of LPS. Treatment with mangiferin did not significantly modify COX-2 expression. RT-PCR analysis was performed after 4 h of incubation with LPS. Mangiferin was added to the cultures 30 min before LPS. (B) Semi-quantitative analysis of the effect of mangiferin on COX-2 expression. This analysis was done with results from four different RT-PCR experiments performed independently. Relative expression of rat b-actin was used for normalization.

serve a trend toward a reduction in COX-2 mRNA stability by mangiferin treatment (p = 0.0526). The p38 MAPK inhibitor produced a substantial reduction in the stability of COX-2 mRNA (Fig. 6). This is in line with findings from previous studies indicating that p38 MAPK is a key factor involved in the stabilization of the COX-2 mRNA resulting in increased protein synthesis [34]. We further studied whether mangiferin alters LPS-mediated phosphorylation of p38 MAPK. Results from the immunoblotting analysis shown in Fig. 7 convincingly demonstrate that mangiferin did not decrease p38 MAPK phosphorylation. These data rule out the possibility that the increase in COX-2 mRNA instability by mangiferin is mediated by a p38 MAPK mechanism. In order to provide more data of the pharmacological effects of mangiferin on microglial activation, we investigated the action of this compound on LPS-induced expression of iNOS and production of TNF-a. Contrary to what was found for the COX-2/PGE2 pathway, mangiferin did not reduce iNOS expression or TNF-a production at any of the concentrations tested, as shown in Fig. 8. Our data demonstrating that mangiferin significantly reduced PGE2 production and free radical formation in microglia represents the first evidence that this natural compound can modulate microglial activation. These in vitro data could help to explain, at least in part, the potent neuroprotective ability of mangiferin against ischemic brain injury, as reported recently [17]. Of great relevance is the ability of mangiferin to reduce COX-2 protein levels in activated microglia. Several lines of evidence suggest that increased COX-2 activity is detrimental for the injured brain. Overexpression of COX-2 has been linked to a dramatic increase in PGE2 and oxidative stress following different types of brain injury including cerebral ischemia [35,36], excitotoxicity [37–39], MPTP and b-amyloid neurotoxicity [40,41]. Reduction in COX-2 protein in LPS-stimulated microglia might underlie the inhibitory effects of mangiferin on PGE2 levels and 8-iso-PGF2a formation. It has been shown that COX-2 activity is an important source of free radicals during neuroinflammation

H.S. Bhatia et al. / Archives of Biochemistry and Biophysics 477 (2008) 253–258

257

Fig. 6. (A) Mangiferin and SB202190 produced an increase in the COX-2 mRNA instability in LPS-activated microglial cells. The stability of COX-2 mRNA in mangiferin-treated and control cells was measured by incubation with actinomycin D (1 lg/ml) to block transcription. Cells were pre-stimulated for 30 min with mangiferin (50 lM). LPS was then added for 2 h. After LPS stimulation, actinomycin D was added for 30 min, media were changed to fresh media containing LPS in the presence of either mangiferin or SB202190. After 1 h, total RNA was isolated, and the disappearance of COX-2 mRNA abundance was determined by RT-PCR analysis. (B) Densitometric analysis of the COX-2 mRNA expression normalized to the housekeeping gene S12. Histograms represent mean ± SEM of 3 independent experiments. Statistical analysis was performed using unpaired t-test. p < 0.05 with respect to LPS + actinomycin D control.

Fig. 7. Mangiferin did not modify LPS-induced phosphorylation of p38 MAPK. Primary microglial cells were treated with LPS for 30 min in the presence or absence of different concentrations of mangiferin. Immunoblotting was performed to detect p38 MAPK in its phosphorylated state. Controls for loading and transfer included actin and total p38 MAPK. Representative Western blots from three independent experiments are shown. Densitometric analysis indicated that mangiferin did not alter LPS-mediated phosphorylation of p38 MAPK (not shown).

[28,42]. In addition, the activity of COX is reduced by antioxidants [29,42–44]. In this particular scenario, mangiferin could impact the overall PGE2 synthesis through reduction in COX-2 expression and direct antioxidant properties. Several other antioxidants have been found to significantly diminish COX activity during inflammatory processes [29,42,44]. An important way of regulation of COX-2 expression in macrophages is post-transcriptional regulation through mRNA destabilization. A common feature of COX-2 mRNA is the presence of multiple copies of adenine/uridine-rich elements composed of the sequence 50 -AUUUA-30 within the 30 -untranslated region (30 UTR) [45,46]. The stability of COX-2 mRNA is regulated by the p38 MAPK signaling cascade, and it has been shown that p38 MAPK inhibition results in destabilization of COX-2 mRNA, as shown in Fig. 6, and reported by others previously [47]. Other mechanisms are also involved in COX-2 mRNA stabilization, including the proteins HuR, AUBF and AUF1 which bind to adenine/uridine-rich elements [46,48,49]. Since mangiferin did not alter p38 MAPK phosphorylation, it seems that this natural compound alter the

Fig. 8. (A) Immunoblot analysis of protein levels of iNOS and actin in LPS-activated microglia treated with different concentrations of mangiferin (1–50 lM). Cell extracts were prepared after 24 h of incubation with LPS, and subjected to immunoblot analysis using specific antibodies for each protein. (B) Quantitative densitometric analysis of iNOS protein expression normalized to actin loading control. Mangiferin had no effect on LPS-induced iNOS synthesis. (C) Lack of effect of mangiferin on the production of TNF-a induced by LPS. Bars represent mean ± SEM of three independent experiments.

COX-2 transcript stability through more complex mechanisms possibly involving interaction/modification of adenine/uridine-rich element-binding proteins. It remains to be investigated how mangiferin modulates the intracellular signaling pathways leading to microglial COX-2 protein expression during neuroinflammation. Present data are the foundation for these mechanistic studies. However, we find appropriate to confront present data with other reports on the actions of mangiferin in other cell types, including macrophages. Previous work has shown that mangiferin significantly limits the expression of COX-2 and iNOS proteins in mouse peritoneal macrophages exposed to LPS and interferon IFNc, as detected using a slot-blot assay [7]. Macrophage activation, as measured by the formation of TNF-a and nitric oxide production, is potently blocked by mangiferin in the microglial cell line N9 and in RAW264.7 mouse macrophages exposed to pro-inflammatory stimuli [5]. However, these effects were observed at very high concentrations, 50–200 lg/ml, which corresponds to approximately 100–400 lM. This is not in line with our present observations where statistically significant inhibitory effects on PGE2 production were seen at the low micromolar range (1–50 lM) of mangiferin (Fig. 2).

258

H.S. Bhatia et al. / Archives of Biochemistry and Biophysics 477 (2008) 253–258

In mouse macrophages stimulated with LPS, high concentrations of mangiferin reduced iNOS expression and TNF-a production [5,7]. In primary microglia, our data showed that mangiferin did not modify these pro-inflammatory mediators (Fig. 8). The reason(s) for these apparent discrepancies are not clear, but different cell types and concentrations of mangiferin may explain the differences. In two different macrophage cell lines (RAW264.7 and J774) stimulated with LPS + IFNc or calcium ionophore A23187, mangiferin reduced PGE2 and leukotriene B4 synthesis [3,4]. Interestingly, mangiferin attenuated the activity of the secretory phospholipase A2 [3], suggesting a multifactorial mode of action of this compound in blocking the inflammatory cascade. However, it remains to be investigated whether mangiferin reduces the release of free arachidonic acid during LPS-mediated inflammatory events in primary microglia. Experiments investigating the effects of mangiferin on the expression/activity of phospholipases and other lipases are needed to further clarify the actions of this natural compound on microglial activation. Recent emerging evidence indicates that mangiferin is an effective neuroprotectant. Nanomolar concentrations of mangiferin effectively reduced Ca2+ overload, oxidative stress and apoptosis caused by excitotoxic insults in primary cultures of rat cortical neurons [17]. In addition, mangiferin treatment rescued hippocampal CA1 neurons from delayed neuronal death in a rat model of global brain ischemia [17]. In line with these data, we had previously shown that pretreatment with Vimang, an extract from M. indica L. enriched in the polyphenol mangiferin (20%), ameliorated postischemic hippocampal neuronal death and improved neurological function in gerbils subjected to global cerebral ischemia [1]. Furthermore, mangiferin protects against MPP+ toxicity in neuronal cells [18], and against excitotoxic oligodendrocyte death [50], mainly due to reduction of oxidative stress. Oral pretreatment with plant extracts enriched in the polyphenol mangiferin reduces postischemic hippocampal neuronal death [1]. In addition, oral administration to rats of mangiferin-containing natural extracts elicited plasma mangiferin concentrations higher than 1 lg/ml (2.5 lM) [51]. These results suggest that the effects we observed in vitro could be relevant to the in vivo conditions. In summary, findings from the present investigation showed that mangiferin effectively blocks microglial activation in terms of attenuation of COX-2 protein expression and the concomitant PGE2 production, as well as reduction in free radical formation. These data suggest that modulation of microglial activation might contribute to the mechanism of protection by mangiferin in different types of brain injury. Acknowledgments The skilful technical assistance of Ulrike Götzinger-Berger and Brigitte Günter is greatly acknowledged. Eduardo Candelario-Jalil was supported by a research fellowship from the Alexander von Humboldt Foundation (Bonn, Germany) and a research grant from the American Heart Association (0720160Z, Pacific Mountain Affiliate). Antonio Carlos Pinheiro de Oliveira was supported by the CAPES Foundation (Brasília, Brazil). Olumayokun A. Olajide was sponsored by the Humboldt Foundation (Bonn, Germany). References [1] G. Martinez Sanchez, E. Candelario-Jalil, A. Giuliani, O.S. Leon, S. Sam, R. Delgado, A.J. Nunez Selles, Free Radic. Res. 35 (2001) 465–473. [2] M.M. Pinto, M.E. Sousa, M.S. Nascimento, Curr. Med. Chem. 12 (2005) 2517– 2538.

[3] G. Garrido, D. Gonzalez, Y. Lemus, D. Garcia, L. Lodeiro, G. Quintero, C. Delporte, A.J. Nunez-Selles, R. Delgado, Pharmacol. Res. 50 (2004) 143–149. [4] G. Garrido, D. Gonzalez, Y. Lemus, C. Delporte, R. Delgado, Phytomedicine 13 (2006) 412–418. [5] G. Garrido, R. Delgado, Y. Lemus, J. Rodriguez, D. Garcia, A.J. Nunez-Selles, Pharmacol. Res. 50 (2004) 165–172. [6] G.M. Sanchez, L. Re, A. Giuliani, A.J. Nunez-Selles, G.P. Davison, O.S. LeonFernandez, Pharmacol. Res. 42 (2000) 565–573. [7] J. Leiro, D. Garcia, J.A. Arranz, R. Delgado, M.L. Sanmartin, F. Orallo, Int. Immunopharmacol. 4 (2004) 991–1003. [8] S. Guha, S. Ghosal, U. Chattopadhyay, Chemotherapy 42 (1996) 443–451. [9] T. Sato, A. Kawamoto, A. Tamura, Y. Tatsumi, T. Fujii, Chem. Pharm. Bull. (Tokyo) 40 (1992) 721–724. [10] G.L. Pardo-Andreu, R. Delgado, A.J. Nunez-Selles, A.E. Vercesi, Pharmacol. Res. 53 (2006) 253–260. [11] G.P. Andreu, R. Delgado, J.A. Velho, C. Curti, A.E. Vercesi, Eur. J. Pharmacol. 513 (2005) 47–55. [12] G.L. Pardo-Andreu, R.A. Cavalheiro, D.J. Dorta, Z. Naal, R. Delgado, A.E. Vercesi, C. Curti, J. Pharmacol. Exp. Ther. 320 (2007) 646–653. [13] A. Dar, S. Faizi, S. Naqvi, T. Roome, S. Zikr-ur-Rehman, M. Ali, S. Firdous, S.T. Moin, Biol. Pharm. Bull. 28 (2005) 596–600. [14] S. Muruganandan, S. Gupta, M. Kataria, J. Lal, P.K. Gupta, Toxicology 176 (2002) 165–173. [15] S. Muruganandan, K. Srinivasan, S. Gupta, P.K. Gupta, J. Lal, J. Ethnopharmacol. 97 (2005) 497–501. [16] S. Muruganandan, J. Lal, P.K. Gupta, Toxicology 215 (2005) 57–68. [17] M. Gottlieb, R. Leal-Campanario, M.R. Campos-Esparza, M.V. Sanchez-Gomez, E. Alberdi, A. Arranz, J.M. gado-Garcia, A. Gruart, C. Matute, Neurobiol. Dis. 23 (2006) 374–386. [18] L. Amazzal, A. Lapotre, F. Quignon, D. Bagrel, Neurosci. Lett. 418 (2007) 159–164. [19] S.U. Kim, J. de Vellis, J. Neurosci. Res. 81 (2005) 302–313. [20] U.K. Hanisch, H. Kettenmann, Nat. Neurosci. 10 (2007) 1387–1394. [21] W.J. Streit, R.E. Mrak, W.S. Griffin, J. Neuroinflammation 1 (2004) 14. [22] M.L. Block, L. Zecca, J.S. Hong, Nat. Rev. Neurosci. 8 (2007) 57–69. [23] M.L. Block, J.S. Hong, Prog. Neurobiol. 76 (2005) 77–98. [24] B.L. Fiebich, K. Biber, K. Lieb, C.D. van, M. Berger, J. Bauer, P.J. Gebicke-Haerter, Glia 18 (1996) 152–160. [25] M.K. Bauer, K. Lieb, K. Schulze-Osthoff, M. Berger, P.J. Gebicke-Haerter, J. Bauer, B.L. Fiebich, Eur. J. Biochem. 243 (1997) 726–731. [26] B.L. Fiebich, K. Lieb, M. Hull, B. Aicher, R.J. van, M. Pairet, G. Engelhardt, Neuropharmacology 39 (2000) 2205–2213. [27] K. Lieb, S. Engels, B.L. Fiebich, Neurochem. Int. 42 (2003) 131–137. [28] R.S. Akundi, E. Candelario-Jalil, S. Hess, M. Hull, K. Lieb, P.J. Gebicke-Haerter, B.L. Fiebich, Glia 51 (2005) 199–208. [29] E. Candelario-Jalil, A.C. de Oliveira, S. Graf, H.S. Bhatia, M. Hull, E. Munoz, B.L. Fiebich, J. Neuroinflammation 4 (2007) 25. [30] A.C. de Oliveira, E. Candelario-Jalil, H.S. Bhatia, K. Lieb, M. Hull, B. Fiebich, Glia 56 (2008) 844–855. [31] P.J. Gebicke-Haerter, J. Bauer, A. Schobert, H. Northoff, J. Neurosci. 9 (1989) 183–194. [32] U.K. Laemmli, Nature 227 (1970) 680–685. [33] P. Chomczynski, N. Sacchi, Anal. Biochem. 162 (1987) 156–159. [34] S.H. Ridley, J.L. Dean, S.J. Sarsfield, M. Brook, A.R. Clark, J. Saklatvala, FEBS Lett. 439 (1998) 75–80. [35] E. Candelario-Jalil, A. Gonzalez-Falcon, M. Garcia-Cabrera, D. Alvarez, S. AlDalain, G. Martinez, O.S. Leon, J.E. Springer, J. Neurochem. 86 (2003) 545–555. [36] E. Candelario-Jalil, A. Gonzalez-Falcon, M. Garcia-Cabrera, O.S. Leon, B.L. Fiebich, J. Neurochem. 100 (2007) 1108–1120. [37] S.J. Hewett, T.F. Uliasz, A.S. Vidwans, J.A. Hewett, J. Pharmacol. Exp. Ther. 293 (2000) 417–425. [38] O. Pepicelli, E. Fedele, M. Berardi, M. Raiteri, G. Levi, A. Greco, M.A. Ajmone-Cat, L. Minghetti, J. Neurochem. 93 (2005) 1561–1567. [39] O. Pepicelli, E. Fedele, G. Bonanno, M. Raiteri, M.A. Ajmone-Cat, A. Greco, G. Levi, L. Minghetti, J. Neurochem. 81 (2002) 1028–1034. [40] R. Vijitruth, M. Liu, D.Y. Choi, X.V. Nguyen, R.L. Hunter, G. Bing, J. Neuroinflammation 3 (2006) 6. [41] M. Hull, B. Muksch, R.S. Akundi, A. Waschbisch, J.J. Hoozemans, R. Veerhuis, B.L. Fiebich, Neurochem. Int. 48 (2006) 663–672. [42] E. Candelario-Jalil, R.S. Akundi, H.S. Bhatia, K. Lieb, K. Appel, E. Munoz, M. Hull, B.L. Fiebich, J. Neuroimmunol. 174 (2006) 39–51. [43] A.A. Beharka, D. Wu, M. Serafini, S.N. Meydani, Free Radic. Biol. Med. 32 (2002) 503–511. [44] Q. Jiang, I. Elson-Schwab, C. Courtemanche, B.N. Ames, Proc. Natl. Acad. Sci. USA 97 (2000) 11494–11499. [45] C.Y. Chen, A.B. Shyu, Mol. Cell Biol. 14 (1994) 8471–8482. [46] D.A. Dixon, C.D. Kaplan, T.M. McIntyre, G.A. Zimmerman, S.M. Prescott, J. Biol. Chem. 275 (2000) 11750–11757. [47] M. Lasa, K.R. Mahtani, A. Finch, G. Brewer, J. Saklatvala, A.R. Clark, Mol. Cell Biol. 20 (2000) 4265–4274. [48] S.J. Cok, S.J. Acton, A.R. Morrison, J. Biol. Chem. 278 (2003) 36157–36162. [49] A. Doller, A. Huwiler, R. Muller, H.H. Radeke, J. Pfeilschifter, W. Eberhardt, Mol. Biol. Cell 18 (2007) 2137–2148. [50] G. Ibarretxe, M.V. Sanchez-Gomez, M.R. Campos-Esparza, E. Alberdi, C. Matute, Glia 53 (2006) 201–211. [51] R. Dai, K. Li, Q. Li, K. Bi, J. Chromatogr. Sci. 42 (2004) 207–210.