Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Silymarin is a flavonoid isolated from the fruits and seeds of the milk thistle, Silybum marianum. Silymarin has a variety of biological effects, including an anticarcinogenic effect (Bhatia et al., 1999), an antihepatotoxic effect, attributed to its stabilizing effect on the plasma membrane and its inhibition of lipid peroxidation (Letteron et al., 1990; Muriel and Mourelle, 1990), an antiulcer effect due to the inhibition of lipoxygenase activity (Alarcon de la Lastra et al., 1992), and an antioxidative effect by the scavenging of reactive oxygen species (Dehmlow et al., 1996). It has also been reported that silymarin exerts an anti-inflammatory and antiarthritic effects by inhibiting the lipoxygenase pathway (Gupta et al., 2000).

Nitric oxide (NO) is a short-lived free radical and intercellular messenger that mediates a variety of biological functions, including vascular homeostasis, neurotransmission, antimicrobial defense, and antitumor activities (Nathan, 1992). NO is known to be synthesized from L-arginine by nitric-oxide synthase (NOS) (Palmer et al., 1988). Three isoforms of NOS have been identified and are classified into two major categories, namely, constitutive and inducible NOS. Neuronal and endothelial NOSs, which are constitutively expressed, are activated by calcium and calmodulin and are called constitutive NOSs (Nathan, 1992). Of the three NO synthases, inducible NO synthase (iNOS), the high-output isoform, is the most widely expressed in various cell types after its transcriptional activation (Xie et al., 1992). Most importantly, iNOS is highly expressed in LPS-activated macrophages, and this contributes to the pathogenesis of septic shock (Petros et al., 1991). In some cases, the induction of iNOS by other stimuli leads to organ destruction in some inflammatory (McCartney-Francis et al., 1993) and autoimmune diseases (Kleemann et al., 1993). Thus, the inhibition of NO production by blocking iNOS expression may present a useful strategy for the treatment of various inflammatory diseases.

The iNOS gene expression is regulated mainly at the transcriptional level in macrophages (Xie et al., 1993), and the major transcriptional regulators of iNOS gene are the NF-B/Rel family of transcription factors that is also a key regulator of a variety of genes involved in immune and inflammatory responses (Xie et al., 1994). The murine iNOS gene promoter contains two NF-B/Rel binding sites, located at 55 and 971 base pairs upstream of the TATA box. Moreover, it has been reported that protein binding to both of these B sites is necessary for the full induction of the iNOS gene by LPS (Lowenstein et al., 1993). In unstimulated cells, NF-B/Rel exists in an inactive state, in the cytoplasm, complexed with the an inhibitory protein, called IB. Upon activation, IB undergoes phosphorylation and degradation, and the NF-B/Rel heterodimer is translocated into the nucleus, where it binds to DNA and activates transcription (Rice and Ernst, 1993). A couple of groups have demonstrated previously the inhibitory effect of silymarin on NF-B/Rel binding activity. Saliou et al. (1998) reported that silymarin blocked the activation of NF-B/Rel induced by okadaic acid and LPS, but not that induced by TNF- in HepG2, a human hepatoblastoma-derived cell line. In contrast, TNF--induced NF-B/Rel binding was inhibited by silymarin in U937, a human histiocytic lymphoma (Manna et al., 1999), thus demonstrating pathway-dependent and cell type-specific inhibitory effect of silymarin.

Because silymarin has been described as a flavonoid antioxidant with anti-inflammatory activity, we investigated the effect of silymarin on the LPS-mediated induction of NO production and iNOS gene expression. Our results demonstrate that silymarin inhibits LPS-induced increase of NO production and activation of iNOS gene expression and suggest that this might be responsible for the anti-inflammatory action of silymarin.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Chemicals, Animals, and Cell Culture. All reagents were purchased from Sigma-Aldrich (St. Louis, MO) unless otherwise stated. Silymarin was dissolved in dimethyl sulfoxide and freshly diluted in culture media for all in vitro experiments. Virus-free female BALB/c mice were purchased from Dae Han Laboratory Animal Research Center Co., Ltd. (Chungbuk, Korea) and cared for as described previously (Lee et al., 1996). For in vivo administration, silymarin was dissolved in a water-based dosing solution containing 0.9% sodium chloride (w/v), 3% ethanol (v/v), 1% Tween-80 (v/v), and 6.6 mM sodium hydroxide as described previously (Zhao and Agarwal, 1999). Silymarin was administrated orally 2 and 0 h before LPS (200 µg/kg i.p.) treatment. The peritoneal macrophages and RAW 264.7 cells (ATCC TIB71) were grown in Dulbecco's modified Eagle's medium (Invitrogen, Carlsbad, CA) supplemented with 10% fetal bovine serum, 2 mM L-glutamine, 100 U/ml penicillin, and 100 mg/ml streptomycin at 37°C in 5% CO₂ humidified air. Peritoneal cells were harvested by sterile peritoneal lavage using phosphate-buffered saline, washed, resuspended in culture medium, and plated at 2 × 10 ⁶ cells/ml. Nonadherent cells were removed by repeated washing after 2-h incubation at 37°C.

Nitrite Quantification. NO accumulation was used as an indicator of NO production in the medium as described previously (Green et al., 1982). Mouse peritoneal macrophages and RAW 264.7 cells were plated at 2 × 10 ⁶ and 5 × 10⁵ cells/ml, respectively, and stimulated with LPS (200 ng/ml) in the presence or absence of silymarin (6.25, 12.5, 25, or 50 mg/ml) for 24 h. The isolated supernatants were mixed with an equal volume of Griess reagent (1% sulfanilamide, 0.1% naphthylethylenediamine dihydrochloride, and 2% phosphoric acid) and incubated at room temperature for 10 min. NaNO₂ was used to generate a standard curve, and nitrite production was determined by measuring optical density at 550 nm.

Quantitative RT-PCR. Competitive RT-PCR was performed as described previously (Jeon et al., 2000) with slight modifications. Briefly, total RNA was isolated using Tri Reagent (Molecular Research Center, Cincinnati, OH) as described previously (Chomczynski and Mackey, 1995). Primers used for iNOS quantitation were as described previously (Jeon et al., 2000). Equal amounts of RNA were reverse transcribed into cDNA using oligo(dT)₁₅ primers. For competitive RT-PCR, an iNOS cDNA fragment with a central 80-base pair deletion was used as an internal standard. Samples were heated to 94°C for 5 min and cycled 40 times at 94°C for 30 s, 59°C for 30 s, and 72°C for 45 s, and this was followed by an additional extension step at 72°C for 5 min. PCR products were electrophoresed in 2% agarose gel and stained with ethidium bromide. Bands were visualized, photographed, and densitometrically quantified using ImageQuant software (Molecular Dynamics, Sunnyvale, CA).

Western Immunoblot Analysis. Whole-cell lysate [20 µg, for iNOS, Erk1/2, SAPK/c-Jun N-terminal kinase (JNK), and p38 MAP kinase] or 20 µg of cytosolic extract (for IB) were separated by 10% SDS-polyacrylamide gel electrophoresis, and electrotransferred to nitrocellulose membranes (Amersham Biosciences UK, Ltd., Little Chalfont, Buckinghamshire, UK). The membranes were preincubated for 1 h at room temperature in Tris-buffered saline, pH 7.6, containing 0.05% Tween 20 and 3% fatty acid-free bovine serum albumin. The nitrocellulose membranes were then incubated with specific antibodies against iNOS (Upstate Biotechnology, Waltham, MA), IB (Santa Cruz Biotechnology, Santa Cruz, CA), or the phosphorylated forms of Erk1/2, SAPK/JNK, or p38 MAP kinase (Cell Signaling Technology, Beverly, MA). Immunoreactive bands were then detected by incubating with conjugates of anti-rabbit (for iNOS, Erk1/2, SAPK/JNK, and p38 MAP kinase) or anti-mouse (for IB) IgG with horseradish peroxidase and enhanced chemiluminescence reagents (Amersham Biosciences UK, Ltd.).

Transient Transfection and CAT Reporter Gene Assay. p(NF-B)₃CAT plasmid has been described previously (Jeon et al., 1999). Transient transfection was performed using the DEAE-dextran method with slight modifications (Xie et al., 1993). After transfection, cells were plated at 5 × 10 ⁵ cells/ml and incubated for 24 h. The transfectants were treated with silymarin 1 h before the treatment of LPS (200 ng/ml), harvested 24 h after LPS treatment, and lysed. The CAT enzyme expression levels were determined using a CAT enzyme-linked immunosorbent assay kit according to the manufacturer's instructions (Roche Applied Science, Mannheim, Germany).

Electrophoretic Mobility Shift Assay. Nuclear extracts were prepared as described previously (Jeon et al., 2000). The protein content of the nuclear extracts was determined using a Bio-Rad protein assay kit according to the manufacturer's instructions (Amersham Biosciences UK, Ltd.). The oligonucleotide sequence for NF-B/Rel (Pierce et al., 1988), AP-1, and Octamer (Annweiler et al., 1993) was as follows: 5'-GATCTCAGAGGGGACTTTCCGAGAGA-3', 5'-GATCTGCATGAGTCAGACACACA-3', and 5'-GATCTTCTAGAGGATCATGCAAATGATCA-3', respectively. The double-stranded oligonucleotides were end-labeled with [-³²P]ATP. Nuclear extracts (5 µg) were incubated with 2 µg of poly(dI-dC) and a ³²P-labeled DNA probe. DNA binding activity was analyzed using 4.8% polyacrylamide gel. After electrophoresis, the gel was dried and subjected to autoradiography. The specificity of binding was examined by competition with unlabeled oligonucleotide.

2',7'-Dichlorofluorescin Diacetate (DCFH-DA) Assay. The production of reactive oxygen species (ROS) was determined using DCFH-DA, an oxidant-sensitive fluorescent probe, and by flow cytometry as described previously (Cho et al., 2000). Briefly, RAW 264.7 cells (1 × 10⁶ cells/ml) were preincubated with Hanks' balanced salt solution in the presence of silymarin (6.25, 12.5, 25, or 50 µg/ml) for 30 min at 37°C in a water bath. After being incubated for 15 min with 20 µM DCFH-DA, 1 mM H₂O₂ was treated, and incubation continued for an additional 15 min. The relative green dichlorofluorescin fluorescence within the living cells was measured using a FACS Calibur flow cytometer (BD Biosciences, Rutherford, NJ).

Statistical Analysis. The mean ± S.D. was determined for each treatment group in each experiment. The significance was determined by either Dunnett's two-tailed t test for comparison between two groups or analysis of variance, followed by Dunnett's test in the case of multiple comparisons (Dunnett, 1955).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

In Vivo and in Vitro Effect of Silymarin on Nitric Oxide Production in Macrophages. To investigate the effect of silymarin on NO production, we measured the accumulation of nitrite, the stable metabolite of NO, in the culture media using Greiss reagent. To investigate the effect of silymarin on NO production, we treated female BALB/c mice with LPS and/or silymarin and measured the NO production in isolated peritoneal macrophages. The administration of LPS (200 µg/kg i.p.) to female BALB/c mice caused a significant increase in the production of nitrite by peritoneal macrophages isolated from these mice. Treatment with silymarin (50 mg/kg, 2 and 0 h before LPS treatment, oral administration) completely blocked the induction of nitrite generation by LPS (Table 1). To further confirm, we examined the inhibitory effect of silymarin on nitrite production in isolated peritoneal macrophages and RAW 264.7 cells. As shown in Fig. 1A, LPS (200 ng/ml) alone increased nitrite production to 10 times the basal level in peritoneal macrophages, and this induction was concentration dependently suppressed by silymarin. In RAW 264.7 cells, LPS (200 ng/ml) evoked a 4.5-fold induction of nitrite production versus the naive control, and this induction was also inhibited by silymarin treatment in a dose-dependent manner (Fig. 1B). The concentration and duration of silymarin treatment used in these studies had no significant effect on the viability of isolated peritoneal macrophages and RAW 264.7 cells.

View larger version (28K): [in this window] [in a new window]   Fig. 1.   Inhibition of nitrite production by silymarin in LPS-stimulated peritoneal macrophages and RAW 264.7 cells. Peritoneal adherent cells (A) and RAW 264.7 cells (B) were pretreated with the indicated concentrations of silymarin for 1 h before being incubated with LPS (200 ng/ml) for 24 h. The culture supernatants were subsequently isolated and analyzed for nitrite production. Each column shows the mean ± S.D. of triplicate determinations. , response that is significantly different from the control group as determined by Dunnett's two-tailed t test at P < 0.05.

Inhibition of iNOS Protein and mRNA Expression by Silymarin in LPS-Stimulated RAW 264.7 Cells. The effects of silymarin on iNOS protein and mRNA expression were examined by Western blot and quantitative RT-PCR, respectively. As shown in Fig. 2, the expression of iNOS protein was barely detectable in unstimulated cells, but markedly increased 24 h after LPS (200 ng/ml) treatment. Consistent with previous results, treatment with silymarin (6.25, 12.5, 25, or 50 µg/ml) showed a concentration-dependent inhibition of iNOS protein expression in LPS-stimulated RAW 264.7 cells (Fig. 2). To assess the effect of silymarin on iNOS mRNA expression, we measured the mRNA levels by quantitative RT-PCR. The expression of iNOS mRNA was hardly detectable in unstimulated cells. However, RAW 264.7 cells expressed high level of iNOS mRNA when stimulated with LPS (200 ng/ml) for 12 h. Furthermore, silymarin inhibited this LPS-stimulated iNOS mRNA production in a dose-dependent manner (Fig. 3).

View larger version (10K): [in this window] [in a new window]   Fig. 2.   Inhibition of iNOS protein expression by silymarin in LPS-stimulated RAW 264.7 cells. RAW 264.7 cells were pretreated with silymarin (6.25, 12.5, 25, or 50 µg/ml) before being incubated with LPS (200 ng/ml) for 24 h. Cell lysates were then prepared and subjected to Western immunoblotting using an antibody specific for murine iNOS. One of two representative experiments is shown.

View larger version (35K): [in this window] [in a new window]   Fig. 3.   Inhibition of iNOS mRNA expression by silymarin in LPS-stimulated RAW 264.7 cells. RAW 264.7 cells were treated with silymarin (6.25, 12.5, 25, or 50 µg/ml) in the presence of LPS (200 ng/ml) for 12 h. Total RNA was isolated and iNOS mRNA expression was determined by competitive RT-PCR. Data shown represent the mean ± S.D. of four separate RNA isolations. , response that is significantly different from the control group as determined by Dunnett's two-tailed t test at P < 0.05.

Inhibition of NF-B/Rel Binding Activity by Silymarin in LPS-Stimulated RAW 264.7 Cells. It is well known that NF-B/Rel is an important transcription factor for the inducibility of iNOS gene by LPS (Xie et al., 1994). To further investigate whether the transcription factor NF-B/Rel is an important target for the action of silymarin in RAW 264.7 cells, we performed an electrophoretic mobility shift assay. Treatment of RAW 264.7 cells with LPS (200 ng/ml) caused a significant increase in the DNA binding activity of NF-B/Rel within 30 min (Fig. 4A). In the presence of silymarin, LPS-induced NF-B/Rel binding was markedly suppressed in a concentration-dependent manner. Although NF-B/Rel is a critical transcription factor controlling iNOS gene expression, it has been known that AP-1 and Octamer are also involved in the expression of the iNOS gene. Therefore, AP-1 and Octamer bindings were also examined. Although AP-1 binding was induced by LPS (200 ng/ml), this AP-1 binding was not inhibited by silymarin at low concentrations, of up to 25 µg/ml, and was only slightly inhibited at high concentration (50 µg/ml) (Fig. 4B). Octamer binding was unaffected by either LPS or silymarin treatment (Fig. 4B).

View larger version (48K): [in this window] [in a new window]   Fig. 4.   Effect of silymarin on the binding of transcription factors in LPS-stimulated RAW 264.7 cells. RAW 264.7 cells were pretreated with the indicated concentrations of silymarin for 1 h before being incubated with LPS for 30 min. Nuclear extracts were then prepared, and NF-B/Rel (A) and AP-1 and Octamer (B) binding was determined by electrophoretic mobility shift assay. The binding specificity was determined using the unlabeled wild-type probe (100-fold in excess) to compete with the labeled oligonucleotide. The results presented are representative of three independent experiments.

Silymarin Suppresses LPS-Induced NF-B/Rel Transcriptional Activity in RAW 264.7 Cells. To determine the effect of silymarin on LPS-stimulated NF-B/Rel-dependent reporter gene expression, we used p(NF-B)₃CAT plasmid, which was generated by inserting three spaced NF-B/Rel binding sites into pCAT-Promoter vector (Promega, Madison, WI). RAW 264.7 cells were transiently transfected with p(NF-B)₃CAT plasmid using the DEAE-dextran method and then stimulated with 200 ng/ml LPS either in the presence or absence of silymarin. A 14.5-fold increase in CAT enzyme expression was detected after stimulation with LPS for 24 h, and the treatment of silymarin significantly reduced the LPS-induced increase in NF-B/Rel-dependent CAT enzyme expression (Fig. 5).

View larger version (47K): [in this window] [in a new window]   Fig. 5.   Inhibition of NF-B-dependent reporter gene expression by silymarin in LPS-stimulated RAW 264.7 cells. Cells were transiently transfected with p(NF-B)3CAT containing three copies of the NF-B binding site, treated with the indicated concentrations of silymarin and LPS (200 ng/ml) for 24 h, and assayed for CAT expression using a CAT enzyme-linked immunosorbent assay kit. Each column shows the mean ± S.D. of six separate samples. , response that is significantly different from the control group as determined by Dunnett's two-tailed t test at P < 0.05.

Inhibition of LPS-Induced IB Protein Degradation in RAW 264.7 Cells. The nuclear translocation and DNA binding of the NF-B/Rel transcription factor is preceded by the phosphorylation and degradation of IB (Stancovski and Baltimore, 1997). To determine whether the inhibition of NF-B DNA binding by silymarin is due to an effect on IB degradation, the cytoplasmic levels of IB were examined by Western immunoblot analysis. In RAW 264.7 cells, IB protein decreased almost completely 5 min after LPS (200 ng/ml) treatment, and returned to the normal level within 1 h. Pretreatment of RAW 264.7 cells with 50 µg/ml silymarin completely blocked the LPS-induced IB degradation (Fig. 6).

View larger version (17K): [in this window] [in a new window]   Fig. 6.   Effect of silymarin on the LPS-induced degradation of IB in RAW 264.7 cells. RAW 264.7 cells were pretreated with 50 µg/ml silymarin for 1 h, incubated for the indicated times with LPS (200 ng/ml), and then assayed for IB in cytosolic fractions by Western immunoblot analysis. One of two representative experiments is shown.

Effect of Silymarin on TNF-- and Okadaic Acid-Induced Activation of NF-B/Rel Binding in RAW 264.7 Cells. A variety of agents, including TNF-, okadaic acid, and phorbol ester, are known to induce NF-B/Rel binding in various cell types (Saliou et al., 1998; Manna et al., 1999). Each of these agents induces NF-B/Rel DNA binding by a different signal transduction pathway. Therefore, we investigated the effect of silymarin on the NF-B/Rel binding activities induced by these agents. In RAW 264.7 cells, TNF- (50 ng/ml) and okadaic acid (500 nM) induced NF-B/Rel DNA binding, but the induction was weaker than that induced by LPS (200 ng/ml) (Fig. 7). Figure 7 also shows that TNF--induced NF-B/Rel DNA binding was blocked by silymarin treatment. However, silymarin had no effect on okadaic acid-induced NF-B/Rel DNA binding. Unlike the other agents, PMA (80 nM) alone did not induce NF-B/Rel DNA binding in these cells.

View larger version (69K): [in this window] [in a new window]   Fig. 7.   Effect of silymarin on the NF-B/Rel binding induced by TNF-, okadaic acid, and PMA in RAW 264.7 cells. RAW 264.7 cells were pretreated with 50 mg/ml silymarin for 1 h before being incubated with TNF-, okadaic acid, and PMA for 30 min. NF-B/Rel binding was determined as described under Materials and Methods. The binding specificity was determined using the unlabeled wild-type probe (100-fold in excess) to compete with the labeled oligonucleotide. One of two representative experiments is shown.

Effect of Silymarin on LPS-Induced Phosphorylation of Erk1/2, SAPK/JNK, and p38 MAP Kinase in RAW 264.7 Cells. Evidence has accumulated that the mitogen-activated protein kinase pathway is important in the activation of NF-B/Rel (Nakano et al., 1998). To investigate whether the inhibition of NF-B/Rel activation by silymarin is mediated through the modulation of the mitogen-activated protein kinase pathway, we examined the effect of silymarin on the LPS-stimulated phosphorylation of Erk1/2, SAPK/JNK, and p38 MAP kinase in RAW 264.7 cells using Western immunoblot analysis. Treatment with silymarin caused a slight inhibition of SAPK/JNK phosphorylation at high concentrations, but this inhibition was not significant (Fig. 8). The phosphorylation of the Erk1/2 and p38 MAP kinase was unaffected by silymarin treatment (Fig. 8).

View larger version (36K): [in this window] [in a new window]   Fig. 8.   Effect of silymarin on LPS-induced phosphorylation of Erk1/2, JNK, and p38 MAP kinase in RAW 264.7 cells. RAW 264.7 cells were treated with the indicated concentrations (6.25, 12.5, 25, or 50 µg/ml) of silymarin before being incubated with LPS (200 ng/ml) for 15 min. The whole-cell lysates were analyzed by Western immunoblot analysis. The results presented are representative of three independent experiments.

Effect of Silymarin on ROS Production in RAW 264.7 Cells. ROS are known to be involved in the activation of NF-B/Rel (Flohe et al., 1997). To assess the mechanism responsible for the inhibitory effect of silymarin on NF-B/Rel activation, we examined the effect of silymarin on the H₂O₂-induced production of ROS in RAW 264.7 cells. Figure 9 shows that 1 mM H₂O₂ markedly increased the production of ROS in RAW 264.7 cells. Pretreatment of cells with silymarin blocked the production of ROS by H₂O₂ even at a low concentration (6.25 µg/ml), and higher concentrations of silymarin caused a dose-dependent inhibition of ROS production (Fig. 9).

View larger version (41K): [in this window] [in a new window]   Fig. 9.   Inhibition of ROS production by silymarin in RAW 264.7 cells. Cells were pretreated with the indicated concentrations (6.25, 12.5, 25, or 50 µg/ml) of silymarin for 30 min. After being stimulated with 1 mM H2O2 for 15 min, the relative mean fluorescence intensity (MFI) was measured using a FACS Calibur flow cytometer. Each column shows the mean ± S.D. of triplicate determinations. , response that is significantly different from the control group as determined by Dunnett's two-tailed t test at P < 0.05.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Silymarin is known to have anti-inflammatory and antioxidant effects. In the present study, we demonstrated that silymarin inhibits NO production and iNOS gene expression in macrophages, and that these effects are mediated through the inhibition of NF-B/Rel transcription factor. As stated earlier, NO plays an important role in the pathogenesis of various inflammatory diseases (Kleemann et al., 1993). Therefore, the inhibitory effect of silymarin on iNOS gene expression suggests one of the mechanisms responsible for the anti-inflammatory action of silymarin.

NF-B/Rel is known as a pleiotropic regulator of various genes involved in immune and inflammatory responses. Importantly, NF-B/Rel is a critical transcriptional regulator of iNOS gene expression (Xie et al., 1994). NF-B/Rel has been shown to be activated by a variety of stimuli, including LPS, TNF-, and okadaic acid, and it is believed that different signal transduction pathways seem to be involved in the activation of NF-B/Rel by these agents. To determine the specific signal transduction pathway that is involved in the inhibition of NF-B/Rel by silymarin, we examined the effects of silymarin on the activation of NF-B/Rel induced by these agents. In this study, we showed that the LPS- and TNF--induced NF-B/Rel binding activity is inhibited by silymarin, which is in agreement with the report of Manna et al. (1999). In contrast to our result, Saliou et al. (1998) reported that TNF--induced NF-B/Rel activation was unaffected by silymarin treatment. In this article, they also showed that silymarin inhibits okadaic acid- and LPS-induced NF-B activation in HepG2 cells. We also examined the effect of silymarin on okadaic acid-induced NF-B/Rel activation. Okadaic acid is a serine/threonine phosphatase inhibitor (PP1 and PP2A) and has been shown to activate NF-B/Rel by blocking PP2A-induced IB kinase inactivation. Our results show that the okadaic acid-induced activation of NF-B/Rel binding was unaffected by silymarin treatment. This result is in line with previous reports, which showed that phosphatase inhibitors induce NF-B/Rel activation via an antioxidant-insensitive pathway (Sun et al., 1995). Therefore, our results show that the PP2A-mediated pathway is not involved in the inhibitory effect of silymarin on NF-B/Rel activation.

The mitogen-activated protein kinases play a critical role in the regulation of cell growth and differentiation and in the control of cellular responses to cytokines and stresses. Moreover, they are also known to be important for the activation of NF-B/Rel (Nakano et al., 1998). Manna et al. (1999) reported the inhibitory effect of silymarin on the TNF--induced activation of mitogen-activated protein kinase and JNK. Furthermore, Zi and Agarwal (1999) reported that silymarin treatment resulted in the inhibition of Erk1/2 activation at lower doses and of JNK1 at higher doses. We also investigated the effects of silymarin on the LPS-induced phosphorylation of Erk1/2, SAPK/JNK, and p38 MAP kinase in RAW 264.7 cells. However, no significant changes in the LPS-induced phosphorylation of Erk1/2, SAPK/JNK, or p38 MAP kinase after silymarin treatment were observed. This result suggests that mitogen-activated protein kinases are not involved in the inhibitory effect of silymarin on LPS-stimulated NF-B/Rel binding in RAW 264.7 cells.

Reactive oxygen species pathway might be another target of silymarin and is believed to be involved in NF-B/Rel activation (Flohe et al., 1997). Moreover, a number of articles demonstrated the radical scavenging and antioxidant activity of silymarin (Dehmlow et al., 1996). In the present study, we were also able to show that silymarin has a radical scavenging activity in RAW 264.7 cells, suggesting the possible mechanism for the inhibitory effect of silymarin on NF-B/Rel activation. It is well known that NF-B/Rel activation is regulated by the redox status of the cells (Cho et al., 2000). However, the exact molecular targets responsible for the redox regulation of NF-B/Rel activation remain unknown and need to be investigated.

Flavonoids are naturally occurring compounds and have a wide range of biological effects, which include antihepatotoxic, antiallergic, anti-inflammatory, antiosteoporotic, and antitumor activities (Di Carlo et al., 1999). Of all these effects, the anti-inflammatory effect is one of the most important properties of flavonoids that has been studied extensively. Landolfi et al. (1984) reported that many of the flavonoids are able to modify platelet function and arachidonic acid metabolism and showed that some flavonoids, such as myricetin and quercetin, block both the cyclooxygenase and lipoxygenase pathways. Nepetin, a flavonoid obtained from Nepeta hindostana, was found to have a significant effect on both proliferative and exudative phases of inflammation (Agarwal, 1982). Other studies have been carried out on the anti-inflammatory effect of quercetin, which reduced leukocyte migration in the exudate, and leukotriene B₄ synthesis in cells stimulated with ionophore A23187 (Mascolo et al., 1988). Quercetin has also been reported to inhibit NO production and iNOS expression in macrophages (Kim et al., 1999; Raso et al., 2001). However, despite its potent anti-inflammatory properties, quercetin is not absorbed in humans when orally administered (Di Carlo et al., 1999). Silymarin, on the other hand, was reported to be distributed rapidly to various tissues after oral administration to mice (Zhao and Agarwal, 1999). Furthermore, silymarin has been clinically used for a long time to treat various liver diseases due to alcohol or drug intoxication, mushroom poisoning, and viral hepatitis (Flora et al., 1998), demonstrating its bioavailability and its benefits as a therapeutic agent.

In summary, this study demonstrates that silymarin inhibits LPS-induced NO production and iNOS gene expression in macrophages and that these effects are mediated, at least in part, by blocking NF-B/Rel transcriptional activation. The fact that NF-B/Rel is negatively regulated by silymarin is important because this transcription factor plays a critical role in the regulation of a variety of genes that are involved in inflammatory responses. Our results also suggest that the inhibition of NF-B/Rel activation by silymarin is mediated by its radical scavenging activity. In view of the facts that NO plays an important role in mediating inflammatory responses and that silymarin is a nontoxic and pharmacologically active compound, the inhibitory effect of silymarin on iNOS gene expression suggests the possible application of silymarin as a useful anti-inflammatory agent.