Silymarin Protects Pancreatic ?-Cells against Cytokine-Mediated Toxicity: Implication of c-Jun NH2-Terminal Kinase and Janus Kinase/Signal T
http://www.100md.com
内分泌学杂志 2005年第1期
Southern California Islet Cell Resources Center (T.M., K.F., I.T., C.V.S., F.K., Y.M.), Department of Diabetes, Endocrinology and Metabolism, City of Hope National Medical Center/Beckman Research Institute, Duarte, California 91010; Department of Surgery (T.M., C.V.S., Y.M.), University of California at Los Angeles, Los Angeles, California 90095; and Department of Gastroenterological Surgery (Y.K.), Graduate School of Medical Sciences, Kobe University, Kobe 650-0017, Japan
Address all correspondence and requests for reprints to: Yoko Mullen, M.D., Ph.D., Southern California Islet Cell Resources Center, Department of Diabetes, Endocrinology and Metabolism, City of Hope National Medical Center/Beckman Research Institute, 1500 East Duarte Road, Duarte, California 91010. E-mail: ymullen@coh.org.
Abstract
Silymarin is a polyphenolic flavonoid that has a strong antioxidant activity and exhibits anticarcinogenic, antiinflammatory, and cytoprotective effects. Although its hepatoprotective effect has been well documented, the effect of silymarin on pancreatic ?-cells is largely unknown. In this study, the effect of silymarin on IL-1? and/or interferon (IFN)--induced ?-cell damage was investigated using RINm5F cells and human islets. IL-1? and/or IFN- induced cell death in a time-dependent manner in RINm5F cells. The time-dependent increase in cytokine-induced cell death appeared to correlate with the time-dependent nitric oxide (NO) production. Silymarin dose-dependently inhibited both cytokine-induced NO production and cell death in RINm5F cells. Treatment of human islets with a combination of IL-1? and IFN- (IL-1?+IFN-), for 48 h and 5 d, resulted in an increase of NO production and the impairment of glucose-stimulated insulin secretion, respectively. Silymarin prevented IL-1?+IFN--induced NO production and ?-cell dysfunction in human islets. These cytoprotective effects of silymarin appeared to be mediated through the suppression of c-Jun NH2-terminal kinase and Janus kinase/signal transducer and activator of transcription pathways. Our data show a direct cytoprotective effect of silymarin in pancreatic ?-cells and suggest that silymarin may be therapeutically beneficial for type 1 diabetes.
Introduction
THE PATHOGENESIS OF type 1 diabetes is characterized by an inflammatory reaction that is caused, at least in part, by inflammatory cytokines produced by infiltrating T-lymphocytes and/or macrophages in and around islets (1, 2). Inflammatory cytokines, such as IL-1? and interferon (IFN)- produced by these cells, initiate a variety of signal cascades in ?-cells that lead to ?-cell dysfunction and destruction. The generation of oxygen free radicals is well known to be one of the main mechanisms of such cytokine-mediated ?-cell damage (3, 4). In addition, nitric oxide (NO) produced through the activation of inducible NO synthase (iNOS) also appears to participate in cytokine-mediated toxicity (5, 6). Besides its direct toxicity, NO reacts with superoxide to form peroxynitrite, which has a much stronger oxidant activity and mediates ?-cell destruction in type 1 diabetes (7, 8). In fact, iNOS inhibitors, such as NG-monomethyl-L-arginine (L-NMMA), attenuate cytokine-induced ?-cell dysfunction and islet degeneration (5, 6). Therefore, not only free radical scavenging actions, but also inhibition of iNOS activation, would be beneficial to prevent or delay ?-cell destruction in the development of type 1 diabetes.
Silymarin is a polyphenolic flavonoid extracted from the milk thistle that has a strong antioxidant activity and exhibits cytoprotective, antiinflammatory, and anticarcinogenic effects (9, 10). In addition to its free radical scavenging properties, silymarin increases antioxidant enzymes, such as superoxide dismutase (SOD) and catalase, and inhibits lipid peroxidation (10, 11, 12). It inhibits lipopolysaccharide (LPS)-induced iNOS expression and NO production in microglia and macrophages by blocking nuclear factor B (NF-B) activation (13, 14). A modulatory effect of silymarin on cytokine-induced activation of MAPKs has also been reported (15). Silymarin has been clinically used for the treatment of acute and chronic liver disease because of its hepatoprotective effects. In the pancreas, Soto et al. (16) showed a preventive effect of silymarin against alloxan-induced diabetes in rats. However, its effect on cytokine-induced toxicity in pancreatic ?-cells and the mechanisms involved are largely unknown.
IL-1? activates various cytotoxic signaling pathways, such as NF-B and MAPKs, in ?-cells. Inhibition of NF-B suppresses IL-1?-induced apoptosis, at least in part, through the suppression of iNOS expression in human islets (17). Activation of MAPKs, including c-Jun NH2-terminal kinase (JNK), p38 kinase (p38), and ERK, also mediates IL-1?-induced iNOS expression and/or apoptosis in ?-cells (18, 19). IFN- alone does not stimulate iNOS expression in rat islets, but it primes for IL-1?-induced iNOS expression (20). Treatment with IFN- reduces insulin mRNA levels, glucose-stimulated insulin secretion, and viability in rodent pancreatic ?-cells (21, 22). These deleterious effects of IFN- in ?-cells are thought to be mediated through the activation of the Janus kinase (JAK)/signal transducer and activator of transcription (STAT) pathway. Upon binding of IFN- to its receptor, the receptor-associated JAK isoforms (p38 kinase1 and JAK2) are activated and phosphorylate themselves and the IFN- receptor, which serve as the binding sites for STAT proteins. Phosphorylated STAT proteins form dimers, translocate to the nucleus, and activate the transcription of target genes, including iNOS (23).
In this study, we examined the effect of silymarin on IL-1? and/or IFN--induced cytotoxicity using a rat insulinoma cell line, RINm5F cells, and human islets. Our data demonstrate the cytoprotective effect of silymarin against cytokine-induced toxicity in ?-cells. Silymarin also suppresses cytokine-induced iNOS expression. JNK and JAK/STAT, but not NF-B, pathways appear to be involved in the cytoprotective action of silymarin.
Materials and Methods
Cell culture and materials
The rat RINm5F insulinoma cells were cultured in RPMI 1640 supplemented with 10% fetal bovine serum (FBS) and antibiotic-antimycotic. Silymarin was obtained from Sigma (Milwaukee, WI) and dissolved in dimethylsulfoxide (DMSO) at 100 mM. Human recombinant IL-1? and IFN- were purchased from R&D systems Inc. (Minneapolis, MN). Rat recombinant IFN- was from BD Biosciences (San Jose, CA). SP600125 (JNK inhibitor), SB203580 (p38 inhibitor), and L-NMMA (iNOS inhibitor) were from Calbiochem (San Diego, CA). U0126 (MAPK kinase/ERK inhibitor) was from Cell Signaling Technology (Beverly, MA). Rabbit polyclonal antibodies to total and phosphorylated JNK, p38, and ERK and antibodies to phosphorylated STAT1, STAT3, and STAT5 were from Cell Signaling Technology. Rabbit polyclonal antibodies to STAT1, STAT3, STAT5, IB, and ?-actin were from Santa Cruz Biotechnology (Santa Cruz, CA). Antibodies to total and phosphorylated STAT1 react with STAT1 (91 kDa) and STAT1? (84 kDa). Antitotal STAT3 antibody recognizes STAT3 (92 kDa), and antiphosphorylated STAT3 antibody recognizes both STAT3 (92 kDa) and ? (83 kDa). Antitotal and phosphorylated STAT5 antibodies recognize STAT5a (94 kDa) and STAT5b (92 kDa). Rabbit polyclonal antibody to iNOS was purchased from BD Biosciences.
Human islets
Human islets were provided for research use by the Southern California Islet Cell Resources Center, City of Hope (Duarte, CA). Islets were isolated by the two-step digestion method (24) and cultured in serum-free CMRL1066 medium (Mediatech, Inc., Holly Hill, FL) for 2 d before being released for research use. These islets were then cultured in Ham’s F12 medium supplemented with 10 mM HEPES, antibiotic antimycotic, and 10% FBS for an additional 2–3 d before being used in experiments. The islet numbers were expressed as the number equivalent to islets 150 μm in diameter (IEQ). Only islets with greater than 80% purity, as determined by dithizone staining, were used.
Cell treatment
RINm5F cells were treated with silymarin or vehicle (DMSO) at the indicated concentration for 2 h before stimulation. In some experiments, RINm5F cells were treated with SP600125, SB203580, or U0126 at 10 μM or with L-NMMA at 0.5 mM for 30 min before stimulation. Then, cells were stimulated with human IL-1? (50 U/ml) and/or rat IFN- (100 U/ml) for the indicated period. Human islets were incubated with silymarin or DMSO at the indicated concentration for 2 h and then stimulated with a combination of human IL-1? (75 U/ml) and human IFN- (750 U/ml) for the indicated period.
Assay for NF-B binding activity
After 30-min stimulation, RINm5F cells were washed with ice-cold PBS, and nuclear extracts were prepared using NE-PER Nuclear and Cytoplasmic Extract (PIERCE, Rockford, IL). The protein concentration of each sample was measured using the Bio-Rad Protein Assay (Bio-Rad, Hercules, CA). Equal amounts of nuclear extracts from different samples were used. NF-B DNA binding activity was measured using TransAM NF-B Kit (Active Motif, Carlsbad, CA). Results were expressed as the OD.
Assessment of cell viability
The viability of RINm5F cells was determined by 3-(4,5-dimethylthiazolyl-2) 2,5-diphenyltetrazolium bromide (MTT) assay, as described previously (25), and propidium iodide (PI) staining. Cells were plated in a 96-well plate in triplicate (20,000 cells/well) and cultured overnight before cytokine stimulation. The viability was assessed after 48- and 72-h culture after cytokine stimulation. In MTT assay, viability was expressed as a percentage of the MTT absorbance of the cells without cytokine stimulation. In PI staining, cells were incubated with PI (2.5 μg/ml) for 10 min and then examined under a fluorescence microscope. The percentage of dead cells was determined by counting all of the PI-positive cells divided by all of the cells in the bright-field pictures. A minimum of 500 cells was counted in at least three random fields for each experiment.
Nitrite determination
RINm5F cells were seeded at a density of 20,000 cells/well in 100 μl culture medium in triplicate, and incubated with cytokines for 48 or 72 h. Human islets (600 IEQ/600 μl culture medium) were cultured in duplicate for 48 h after stimulation. Nitrite production in the supernatant was measured using a Griess Reagent Kit (Molecular Probes, Eugene, OR).
Semiquantitative PCR
RINm5F cells and human islets were collected 12 and 8 h, respectively, after stimulation. Total RNA was extracted from cells using TRI REAGENT (Molecular Research Center, Inc., Cincinnati, OH). Two micrograms of RNA from each sample were then reverse-transcribed into first-strand cDNA in 20 μl solution using Superscript RNase H– reverse transcriptase (Invitrogen, Carlsbad, CA). The final cDNA products were then diluted by adding 80 μl H2O. PCR was performed using the FailSafe PCR System (EPICENTRE, Madison, WI). A standard 25-μl PCR solution contained 1.5 μl cDNA, 30 pmol each of forward and reverse primers, 12.5 μl Premix D, and 0.25 μl Enzyme Mix. The following primers were used: rat iNOS (297 bp) forward 5'-CCA ACC GGA GAA GGG GAC GAA CT-3', reverse 5'-GGA GGG TGG TGC GGC TGG AC-3'; rat ?-actin (764 bp) forward 5'-TTG TAA CCA ACT GGG ACG ATA TGG-3', reverse 5'-GAT CTT GAT CTT CAT GGT GCT AGG-3'; human iNOS (236 bp) forward 5'-ACA TTG ATC AGA AGC TGT CCC AC-3', reverse 5'-CAA AGG CTG TGA GTC CTG CAC-3'; human glyceraldehyde-3-phosphate dehydrogenase (600 bp) forward 5'-CCA CCC ATG GCA AAT TCC ATG GCA-3', reverse 5'-TCT AGA CGG CAG GTC AGG TCC ACC-3'. The cycle number was determined to be in the linear range of amplification for each primer pair. Electrophoresis of PCR products were performed in 1.5% agarose gel containing 0.5 mg/ml ethidium bromide. The band intensities were measured and quantitated with an Alpha Innotech densitometer (San Leandro, CA).
Western blotting
After cytokine stimulation, RINm5F cells or human islets were harvested at the indicated time and washed twice with ice-cold PBS. Nuclear and cytoplasmic extracts were prepared using NE-PER Nuclear and Cytoplasmic Extract. For whole-cell lysate, cells were lysed with lysis buffer containing 50 mM Tris-HCl (pH 7.4), 0.5% (vol/vol) NP-40, 150 mM NaCl, 5 mM EDTA, 50 mM NaF, 1 mM Na3VO4, 1 mM phenylmethylsulfonylfluoride, 10 μg/ml leupeptin, and 10 μg/ml aprotinin for 30 min at 4 C. After centrifugation for 15 min at 12,000 rpm, the supernatants were recovered and eluted with Laemmli sample buffer. The protein concentration of each sample was measured as described above, and equal amounts of protein were used. Western blotting was performed as described previously (26). The band intensities were measured and quantitated with an Alpha Innotech densitometer.
Glucose-stimulated insulin secretion
Handpicked islets were treated with silymarin or vehicle (DMSO) for 2 h, then stimulated with IL-1? and IFN-, and cultured for 5 d. Islets were then washed three times with basal media (RPMI 1640 medium containing 3.3 mM glucose, 10 mM HEPES, antibiotic-antimycotic, 1% FBS) and preincubated with the same basal media for 1 h. After wash, islets were incubated with basal media for 1 h, followed by a second incubation with high-glucose media (RPMI 1640 medium containing 19.5 mM glucose, 10 mM HEPES, antibiotic-antimycotic, 1% FBS) for 1 h. The insulin released into the medium was measured using a human insulin ELISA kit (ALPCO Diagnostics, Windham, NH). Each experiment was performed in duplicate. In total, three different experiments using three separate human islet preparations were performed. The fold increase in the secreted insulin levels from 3.3 to 19.5 mM glucose between groups was compared for statistical analysis.
Statistical analysis
Data are expressed as mean ± SE. Statistical differences between groups were analyzed by ANOVA followed by Bonferroni’s test. P < 0.05 was considered significant.
Results
Silymarin protects RINm5F cells against cytokine-induced cell death
First, we wanted to determine whether silymarin could protect pancreatic ?-cells from cytokine-induced cell death. RINm5F insulinoma cells incubated with IL-1? alone for 72 h resulted in a 20–30% decrease in cell viability compared with the untreated control cells, although no significant decrease in viability was observed after 48 h incubation (Fig. 1A). In contrast, incubation with IFN- alone for 48 and 72 h led to 50% and 60% decrease in viability, respectively. Incubation with a combination of IL-1? and IFN- (IL-1?+IFN-) for 72 h resulted in further decrease to 20–30% viability. Pretreatment with silymarin protected against cytokine-induced cell death in a dose-dependent manner (Fig. 1B). The maximum effect was achieved at the concentration of 100 μM silymarin. Similar results were obtained by PI staining, confirming that the observed changes in MTT assay resulted from the changes in cell death but not in cell proliferation or mitochondrial metabolic activity (Fig. 1C). These results indicate that silymarin has a cytoprotective effect against cytokine-mediated cytotoxicity in RINm5F cells.
FIG. 1. Silymarin protects against cytokine-induced cell death of RINm5F cells. A, RINm5F cells were stimulated with IL-1? (50 U/ml), or rat IFN- (100 U/ml), or IL-1? (50 U/ml) + rat IFN- (100 U/ml). Viability was assessed by MTT assay at 48 h and 72 h after stimulation, and results were expressed as a percentage of the MTT absorbance of the control unstimulated cells. Data are means ± SE from four to five individual experiments. *, P < 0.05 vs. control cells without cytokine(s). B, RINm5F cells were treated with silymarin or DMSO for 2 h at the indicated concentrations, followed by stimulation with IL-1? (50 U/ml), or rat IFN- (100 U/ml), or IL-1? (50 U/ml) + rat IFN- (100 U/ml). Viability was assessed by MTT assay at 72 h after stimulation. Results were expressed as a percentage of the MTT absorbance of the control unstimulated cells treated with the same concentration of silymarin. C, RINm5F cells were treated with 100 μM silymarin or DMSO for 2 h, followed by stimulation with IL-1? (50 U/ml), or rat IFN- (100 U/ml), or IL-1? (50 U/ml) + rat IFN- (100 U/ml). Viability was assessed by PI staining at 72 h after stimulation. Results were expressed as a percentage of the dead (PI positive) cells. Data are means ± SE from four to five individual experiments. *, P < 0.05; **, P < 0.01 vs. cytokine-stimulated cells without silymarin.
Silymarin inhibits cytokine-induced iNOS expression and NO production in RINm5F cells
Based on previous studies, we hypothesized that the observed cytoprotective effect of silymarin might be attributed to inhibition of NO production. Therefore, we examined the effect of silymarin on cytokine-induced NO production. As previously shown (27), IL-1? alone induced time-dependent nitrite production in RINm5F cells (Fig. 2A). However, IFN- induced higher levels of nitrite production than IL-1? (Fig. 2A), especially at 48 h after stimulation, when the levels of nitrite production were approximately three times higher than IL-1? alone. Treatment with IL-1?+IFN- resulted in slightly increased nitrite production compared with IFN- alone. This higher rate of NO production induced by IFN- or IL-1?+IFN- in the early incubation period (48 h) appeared to correlate with higher and earlier induction of cell death in RINm5F cells compared with IL-1? alone. Treatment with silymarin inhibited cytokine-induced nitrite production in a dose-dependent manner, with maximum inhibition at 100 μM (Fig. 2B). As shown in Fig. 2, C and D, silymarin also inhibited the cytokine-induced expression of both iNOS mRNA and iNOS protein. Addition of iNOS inhibitor, L-NMMA (0.5 mM), in the culture medium with cytokine almost completely blocked cytokine induced-NO production but only partially prevented cell death (Fig. 3, A and B), suggesting that cytokine-induced cell death was only partially mediated by NO production in RINm5F cells. These results suggest that silymarin protects RINm5F cells against cytokine-induced cell death, at least in part, through the suppression of NO production by down-regulation of iNOS expression at the transcriptional level.
FIG. 2. Silymarin inhibits cytokine-induced iNOS expression and NO production in RINm5F cells. A, RINm5F cells (20,000 cells/well) in a 100-μl medium were stimulated with IL-1? (50 U/ml), rat IFN- (100 U/ml), or IL-1? (50 U/ml) + rat IFN- (100 U/ml). Nitrite production in the medium was determined using Griess reagent at 48 h and 72 h after stimulation. B, RINm5F cells were treated with silymarin or DMSO for 2 h at the indicated concentrations, followed by stimulation with IL-1? (50 U/ml), or rat IFN- (100 U/ml), or IL-1? (50 U/ml) + rat IFN- (100 U/ml). Nitrite production in the medium was determined at 72 h after stimulation. C and D, RINm5F cells were treated with 100 μM silymarin or DMSO for 2 h, followed by stimulation with IL-1? (50 U/ml), or rat IFN- (100 U/ml), or IL-1? (50 U/ml) + rat IFN- (100 U/ml). Total RNA was isolated at 12 h after isolation, and iNOS mRNA expression was determined by RT-PCR (C). Whole-cell lysates were recovered at 24 h after stimulation, and iNOS protein expression was determined by Western blotting analysis (D). The relative values of iNOS/actin mRNA or protein levels were expressed as means in bar graphs. Nitrite data are means ± SE from four to five individual experiments. RT-PCR and Western blotting data are representative from three individual experiments. *, P < 0.05; **, P < 0.01 vs. cytokine-stimulated cells without silymarin.
FIG. 3. Effect of the iNOS inhibitor, L-NMMA, on cytokine-induced NO production (A) and cell death (B) in RINm5F cells. RINm5F cells (20,000 cells/well) in a 100-μl medium were stimulated with IL-1? (50 U/ml), or rat IFN- (100 U/ml), or IL-1? (50 U/ml) + rat IFN- (100 U/ml) in the presence or absence of 0.5 mM L-NMMA. A, Nitrite production in the medium was determined using Griess reagent at 72 h after stimulation. B, Viability was assessed by MTT assay at 72 h after stimulation, and results were expressed as a percentage of the MTT absorbance of the control unstimulated cells. Data are means ± SE from three individual experiments. *, P < 0.001; **, P < 0.05 vs. the same cytokine-stimulated cells without L-NMMA.
No effect of silymarin on IL-1?-induced NF-B activation in RINm5F cells
The NF-B pathway plays a crucial role in IL-1?-induced iNOS expression and cell death in ?-cells. Because silymarin has been reported to suppress NF-B activation induced by various stimuli in many types of cells (15), the effect of silymarin on cytokine-induced NF-B activation in RINm5F cells was explored. Stimulation with IL-1? for 30 min induced an approximately 4-fold increase in NF-B activation compared with the untreated control cells (Fig. 4A). Pretreatment with silymarin had no effect on IL-1?-induced NF-B activation. This result was also confirmed by Western blotting analysis showing that IL-1?-induced degradation of IB, which precedes the translocation of NF-B to the nucleus, was not affected by silymarin (Fig. 4B). IFN- alone did not induce NF-B activation, and it did not affect IL-1?-induced NF-B activation when used in combination (Fig. 4A).
FIG. 4. Effect of silymarin on IL-?-induced NF-B activation in RINm5F cells. A, RINm5F cells were treated with 100 μM silymarin or DMSO for 2 h, followed by stimulation with IL-1? (50 U/ml), or rat IFN- (100 U/ml), or IL-1? (50 U/ml) + rat IFN- (100 U/ml). Nuclear extracts were recovered after 30 min stimulation, and NF-B activity was determined as described in Materials and Methods. Results were expressed as the OD. NF-B activity data are means ± SE from three individual experiments. B, RINm5F cells were treated with 100 μM silymarin or DMSO for 2 h, followed by stimulation with IL-1? (50 U/ml). Whole-cell lysates were recovered at the indicated time after stimulation, and IB expression was determined by Western blotting analysis. Western blotting data are representative from three individual experiments.
Silymarin modulates IL-1?-induced MAPKs activation
Besides the NF-B pathway, MAPK pathways are also known to be involved in the regulation of IL-1?-induced iNOS expression and cell death in ?-cells (18). Therefore, the effect of silymarin on IL-1?-induced MAPKs activation was investigated. IL-1? activated all three MAPK pathways, including JNK, p38, and ERK (Fig. 5A). In all cases, the maximum level of phosphorylation was observed at 30 min after stimulation. Pretreatment with silymarin attenuated IL-1?-induced phosphorylation of JNK, although it did not completely abolish JNK phosphorylation. On the other hand, IL-1?-induced phosphorylation of ERK was further augmented by silymarin pretreatment. Phosphorylation of p38 was not affected. Next, we compared the effect of specific pharmacological inhibitors of MAPKs with silymarin on the suppression of IL-1?-induced nitrite production and cell death. Pretreatment of RINm5F cells with 10 μM SP60025 (JNK inhibitor) or 10 μM SB203580 (p38 inhibitor) partially inhibited IL-1?-induced nitrite production (Fig. 5B). Furthermore, IL-1?-induced cell death was significantly reversed by pretreatment with SP600125 or SB203580 (Fig. 5C). In contrast, the MAPK kinase/ERK inhibitor, U0126, had no effect on either nitrite production or cell viability. These results suggest that silymarin may, in part, inhibit IL-1?-induced NO production and cell death through the suppression of JNK activation.
FIG. 5. Modulation of IL-1?-induced MAPKs activation by silymarin in RINm5F cells. A, RINm5F cells were treated with 100 μM silymarin or DMSO for 2 h, followed by stimulation with IL-1? (50 U/ml). Whole-cell lysates were recovered at the indicated time after stimulation, and phosphorylation of MAPKs was detected by Western blotting analysis. The relative values of p-JNK/JNK, p-p38/p38, and p-ERK/ERK were expressed as means in line graphs. B and C, RINm5F cells were treated with 10 μM SP600125, SB203580, U0126, or DMSO for 30 min, or with 100 μM silymarin for 2 h. Cells were then stimulated with IL-1? (50 U/ml) and cultured for 72 h. Nitrite production in the medium was determined using Griess reagent (B). Viability was determined by MTT assay, and results were expressed as a percentage of the MTT absorbance of the control unstimulated cells (C). Nitrite and viability data are means ± SE from three individual experiments. D, RINm5F cells were treated with 100 μM silymarin or DMSO for 2 h, followed by stimulation with IL-1? (50 U/ml), or rat IFN- (100 U/ml), or IL-1? (50 U/ml) + rat IFN- (100 U/ml). Whole-cell lysates were recovered 20 min after stimulation, and phosphorylation of MAPKs was detected by Western blotting analysis. The relative values of p-JNK/actin, p-p38/actin, and p-ERK/actin were expressed as means in bar graphs. Western blotting data are representative from three individual experiments. *, P < 0.05 vs. cytokine-stimulated cells treated with DMSO.
IFN- alone did neither activate JNK or p38, nor augment IL-1?-induced phosphorylation of JNK or p38, but slightly activated ERK (Fig. 5D). Silymarin pretreatment augmented the phosphorylation levels of ERK and, to a lesser extent, of p38.
Silymarin inhibits IFN--induced activation and nuclear translocation of STAT1, STAT3, and STAT5
IFN--elicited JAK/STAT signaling pathways have been reported to be involved in cytokine-induced inflammatory responses, including iNOS expression (28). STAT1 activation has been especially implicated in IFN--mediated cytotoxicity in ?-cells (20, 21, 29). Therefore, we investigated the effect of silymarin on IFN--induced activation of STAT proteins in RINm5F cells. Phosphorylation of STAT1/?, STAT3/?, and STAT5a/b was observed 20 min after IFN- stimulation (Fig. 6A). IL-1? alone had no effect, nor did the addition of IL-1? to IFN- increase IFN--induced phosphorylation of STAT proteins. Pretreatment with silymarin attenuated IFN--induced phosphorylation of all three STAT proteins, though the effect on STAT1 was smaller than that of STAT3 or STAT5. However, the IFN--induced translocation of STAT1, STAT3, and STAT5 proteins was similarly attenuated by silymarin pretreatment (Fig. 6B). These results imply that inhibition by silymarin of IFN--induced iNOS expression and subsequent cell death is due to suppression of STAT proteins activation.
FIG. 6. Silymarin inhibits IFN--induced activation and nuclear translocation of STAT1, STAT3, and STAT5 in RINm5F cells. A, RINm5F cells were treated with 100 μM silymarin or DMSO for 2 h, followed by stimulation with IL-1? (50 U/ml), or rat IFN- (100 U/ml), or IL-1? (50 U/ml) + rat IFN- (100 U/ml). Whole-cell lysates were recovered 20 min after stimulation, and phosphorylation of STAT proteins was detected by Western blotting analysis. The relative values of p-STAT1/STAT1, p-STAT3/STAT3, and p-STAT5/STAT5 were expressed as means in bar graphs. B, RINm5F cells were treated with 100 μM silymarin or DMSO for 2 h, followed by stimulation with rat IFN- (100 U/ml). Nuclear (N.) and cytoplasmic (C.) extracts were recovered at 30 min after stimulation, and translocation of STAT proteins to the nucleus was detected by Western blotting analysis. The densitometric values of nuclear STAT1, STAT3, and STAT5 proteins were expressed as means in bar graphs. Western blotting data are representative from three individual experiments.
Silymarin inhibits cytokine-induced iNOS expression, NO production, and activation of JNK and STAT proteins in human islets
Our results consistently showed the suppressive effect of silymarin on IL-1? and/or IFN--induced iNOS expression and NO production in RINm5F cells. To extend these studies to human cells, we next determined whether silymarin could also suppress IL-1?+IFN--induced iNOS expression and NO production in human islets. We found that, similar to results with RINm5F cells, silymarin pretreatment decreased cytokine-induced nitrite production by human islets in a dose-dependent manner (Fig. 7A). Semiquantitative RT-PCR results showed that silymarin also suppressed IL-1?+IFN--induced iNOS mRNA expression (Fig. 7B). Stimulation with IL-1?+IFN- resulted in activation of all three MAPKs and STAT1, 3, and 5a proteins within 30 min (Fig. 7C), but silymarin only attenuated JNK phosphorylation, whereas no effect was observed on phosphorylation of p38 or ERK. On the other hand, phosphorylation of all three STAT proteins was partially inhibited by silymarin pretreatment in human islets (Fig. 7C). Still, silymarin pretreatment did not completely abolish phosphorylation of JNK or STAT proteins in human islets.
FIG. 7. Silymarin inhibits cytokine-induced iNOS expression, NO production, and activation of JNK and STAT proteins in human islets. A, Human islets (600 IEQ/600 μl medium) were treated with silymarin for 2 h at the indicated concentrations, followed by stimulation with IL-1? (75 U/ml) + IFN- (750 U/ml). Nitrite production in the medium was determined using Griess reagent at 48 h after stimulation. Nitrite data are mean ± SE from three individual experiments using three different islet preparations. B and C, Human islets were treated with 100 μM silymarin or DMSO for 2 h, followed by stimulation with IL-1? (75 U/ml) + IFN- (750 U/ml). Total RNA was recovered at 12 h after stimulation, and iNOS mRNA expression was determined by RT-PCR (B). Whole-cell lysates were recovered at 30 min after stimulation, and phosphorylation of MAPKs and STAT proteins was detected by Western blotting analysis (C). The relative values of iNOS/glyceraldehyde-3-phosphate dehydrogenase, p-JNK/actin, p-p38/actin, p-ERK/actin, p-STAT1/actin, p-STAT3/actin, and p-STAT5/actin were expressed as means in bar graphs. RT-PCR and Western blotting data are representative from three individual experiments using three different islet preparations. *, P < 0.05; **, P < 0.001 vs. cytokine-stimulated islets without silymarin.
Protective effect of silymarin against cytokine-induced impairment of glucose-stimulated insulin secretion by human islets
Finally, we examined the protective effect of silymarin on cytokine-induced ?-cell damage in human islets. Treatment with IL-1?+IFN- and/or silymarin did not induce a significant increase in cell death in human islets, as assessed by MTT assay and PI staining. However, treatment of human islets with IL-1?+IFN- for 5 d resulted in the impairment of insulin secretion in response to high glucose (Fig. 8). Silymarin pretreatment prevented IL-1?+IFN--induced impairment of glucose-stimulated insulin secretion. Islets treated with silymarin alone released higher levels of insulin than the control islets treated with vehicle (DMSO) with a normal insulin secretory response to glucose.
FIG. 8. Silymarin protects human islets against cytokine-induced impairment of glucose-stimulated insulin secretion. Handpicked human islets were treated with 100 μM silymarin or DMSO for 2 h, then stimulated with IL-1? (75 U/ml) + IFN- (750 U/ml), and subsequently cultured for 5 d. After wash, islets were incubated with basal medium (containing 3.3 mM glucose) for 1 h, followed by a second incubation with high-glucose medium (containing 15.5 mM glucose) for 1 h. Insulin secreted into the medium was measured by ELISA. Data are means ± SE from three individual experiments using three different islet preparations. *, P < 0.05 vs. cytokine-stimulated islets without silymarin.
Discussion
In this study, we have demonstrated that silymarin prevents cytokine-induced cell death in RINm5F cells and cytokine-induced impairment of glucose-stimulated insulin secretion by human islets. In various types of cells, silymarin strongly blocks cytokine- or LPS-induced NF-B activation, which is crucial for iNOS expression in ?-cells (13, 30, 31). Unexpectedly, silymarin had no effect on IL-1?-induced NF-B activation in RINm5F cells. However, it partially inhibits IL-1?-induced JNK activation. JNK is known to be involved in the regulation of expression of many inflammatory genes, such as iNOS, IL-1?, and cyclooxygenase (COX)-2 (32, 33, 34). Although JNK’s roles in the induction of cell death by various stimuli are not firmly established, many evidences have recently accumulated to show that activation of JNK leads to cell death in pancreatic ?-cells and islets (19, 35, 36, 37). Furthermore, Kaneto et al. (38) showed that suppression of JNK activity by adenovirus-mediated overexpression of dominant-negative JNK prevented oxidative stress-induced inhibition of insulin gene expression and secretion in rat islets. Therefore, the observed cytoprotective effects of silymarin, including prevention of cytokine-induced impairment of glucose-stimulated insulin secretion, could be, at least in part, explained by its ability to suppress JNK activity.
Silymarin also suppressed IFN--induced phosphorylation and nuclear translocation of STAT1, STAT3, and STAT5 proteins. JAK/STAT pathways, especially JAK/STAT1, have been shown to be involved in various inflammatory actions (28). However, the roles of IFN- and IFN--elicited JAK/STAT pathways in the induction of ?-cell damage and its recovery process are still elusive. Treatment with IFN- reduces insulin mRNA levels and glucose-stimulated insulin secretion in mouse ?-cell line, CDM3D cells, without increasing NO production (21). IFN- induces cell death and a decrease of insulin gene expression in rat INS-1 cells, with no effect on NO production (22). Contrary to these previous reports, in our study, IFN- alone induced iNOS expression and NO production, which partially mediated cell death in RINm5F cells. This discrepancy may be due to not only species-specific but also cell type-specific regulation of IFN- signaling in ?-cells. However, these previously reported deleterious effects of IFN- on ?-cells, even in the absence of an increase in NO production, are thought to be mediated by STAT1 activation. Although IFN- alone does not stimulate iNOS expression in rodent and human islets, IFN- reduces the concentration of IL-1? required to induce iNOS expression in rat islets (20), and a combination of IL-1? and IFN- is required to induce iNOS expression and ?-cell dysfunction in mouse and human islets (5, 6). This priming or synergistic action of IFN- for IL-1?-induced iNOS expression is also associated with STAT1 activation. After homodimerization and translocation to the nucleus, phosphorylated STAT1 binds to the IFN--activated site (GAS), which is found in the promoters of various inflammatory genes, including iNOS and COX-2 (23, 39). It also induces the expression of IFN regulatory factor-1, which, through binding to IFN-stimulated response elements and interaction with NF-B sites in the iNOS promoter, would augment IL-1?-induced iNOS expression (39, 40). Therefore, the suppressive effect of silymarin on activation and nuclear translocation of STAT1 probably contributes to the inhibition of IFN-- and/or IL-1?+IFN--induced NO production, ?-cell death, and dysfunction in RINm5F cells and human islets.
On the other hand, the role of STAT3 or STAT5 in the regulation of iNOS expression and ?-cell dysfunction remains unclear. IL-6-induced iNOS expression is mediated by JAK2/STAT3 activation in isolated rat myocytes (41). Yoshida et al. (42) reported that NF-B p65 and non-tyrosine-phosphorylated STAT3 physically and functionally interact and that this interaction results in the inability of STAT3 to bind to and activate GAS site-dependent genes. This interaction also inhibits NF-B-dependent iNOS expression (43). Based on these reports, in addition to STAT1, inhibition of STAT3 activation by silymarin might also participate in suppression of IFN-- and/or IL-1?+IFN--induced iNOS expression and/or ?-cell dysfunction in RINm5F cells and human islets. Unlike STAT3, the evidence for the involvement of STAT5 in iNOS expression has not been reported so far. However, Yamaoka et al. (44) showed that STAT5 activation was involved in LPS-induced COX-2 expression in monocytes, suggesting that silymarin might suppress cytokine-induced COX-2 expression through the inhibition of STAT5 activation. On the contrary, STAT5b-mediated inhibition of NFB signaling to IFN regulatory factor-1 promoters was also reported (45). Therefore, further studies are necessary to elucidate whether and how STAT3 or STAT5 proteins are involved in cytokine-induced iNOS expression and dysfunction in ?-cells.
Although silymarin suppressed cytokine-induced NO production by only 50–60%, its cytoprotective effect was comparable with that of the iNOS inhibitor, L-NMMA. Silymarin’s antioxidant properties, based on its ability to scavenge free radicals and inhibit lipid peroxidation, have been well documented. Free radical production and lipid peroxidation are known to mediate cytotoxic action of cytokines on ?-cells (3). Thus, its cytoprotective effect in pancreatic ?-cells would also be attributed to its antioxidant activity in addition to the suppression of NO production. How silymarin suppresses the activation of JNK or STAT proteins is not clear. However, some antioxidants, such as N-acetyl-L-cysteine (NAC) and vitamin E, were shown to inhibit the activation of MAPKs and STAT proteins (46, 47). Furthermore, Matsubara et al. (48) showed that IL-1? and/or IFN- induced superoxide release in the culture medium within 15 min in endothelial cells, suggesting that superoxide produced intracellularly within minutes after cytokine stimulation might serve as signal transduction elements for JNK or STAT activation. Therefore, the suppressive effect of silymarin on the activation of JNK and STAT proteins might also be due to its antioxidant activity.
Another possible mechanism underlying the cytoprotective effect of silymarin could be to increase antioxidant enzymes such as SOD, glutathione peroxidase (Gpx), and catalase, as previously described by Soto et al. (11). However, this is not likely because a previous report has shown that overexpression of catalase, Gpx, or Cu/Zn SOD in RINm5F cells increased IL-1?-induced iNOS expression, though they still protected against cytokine-mediated toxicity through inactivation of reactive oxygen species generated in the signal cascades of cytokines (27). Another report has shown that overexpression of MnSOD, but not catalase, Gpx, or Cu/ZnSOD, in RINm5F cells reduced IL-1?-induced iNOS promoter activation and iNOS expression, which was mediated through the inhibition of NF-B activation (49). We confirmed no alteration in the levels of MnSOD expression in RINm5F cells treated with silymarin for up to 24 h, by Western blotting analysis (data not shown).
In this study, we showed the direct cytoprotective effect of silymarin on pancreatic ?-cells. Importantly, silymarin also has an ability to inhibit the production of inflammatory cytokines, such as IL-1?, IFN-, and TNF-, from macrophages and/or T-lymphocytes (30, 50, 51), which probably initiate the destruction of ?-cells in the development of type 1 diabetes. Taken together, these data indicate that silymarin may be useful as a therapeutic agent for type 1 diabetes.
Acknowledgments
The authors acknowledge the editorial services provided by Elizabeth Stein, Ph.D.
References
Bottazzo GF, Dean BM, McNally JM, MacKay EH, Swift PG, Gamble DR 1985 In situ characterization of autoimmune phenomena and expression of HLA molecules in the pancreas in diabetic insulitis. N Engl J Med 313:353–360
Lacy PE 1994 The intraislet macrophage and type I diabetes. Mt Sinai J Med 61:170–174
Rabinovitch A, Suarez WL, Thomas PD, Strynadka K, Simpson I 1992 Cytotoxic effects of cytokines on rat islets: evidence for involvement of free radicals and lipid peroxidation. Diabetologia 35:409–413
Rabinovitch A, Suarez-Pinzon WL, Strynadka K, Lakey JR, Rajotte RV 1996 Human pancreatic islet ?-cell destruction by cytokines involves oxygen free radicals and aldehyde production. J Clin Endocrinol Metab 81:3197–3202
Corbett JA, Sweetland MA, Wang JL, Lancaster Jr JR, McDaniel ML 1993 Nitric oxide mediates cytokine-induced inhibition of insulin secretion by human islets of Langerhans. Proc Natl Acad Sci USA 90:1731–1735
Thomas HE, Darwiche R, Corbett JA, Kay TW 2002 Interleukin-1 plus -interferon-induced pancreatic ?-cell dysfunction is mediated by ?-cell nitric oxide production. Diabetes 51:311–316
Beckman JS, Koppenol WH 1996 Nitric oxide, superoxide, and peroxynitrite: the good, the bad, and ugly. Am J Physiol 271:C1424–C1437
Suarez-Pinzon WL, Szabo C, Rabinovitch A 1997 Development of autoimmune diabetes in NOD mice is associated with the formation of peroxynitrite in pancreatic islet ?-cells. Diabetes 46:907–911
Katiyar SK, Korman NJ, Mukhtar H, Agarwal R 1997 Protective effects of silymarin against photocarcinogenesis in a mouse skin model. J Natl Cancer Inst 89:556–566
Zhao J, Lahiri-Chatterjee M, Sharma Y, Agarwal R 2000 Inhibitory effect of a flavonoid antioxidant silymarin on benzoyl peroxide-induced tumor promotion, oxidative stress and inflammatory responses in SENCAR mouse skin. Carcinogenesis 21:811–816
Soto C, Recoba R, Barron H, Alvarez C, Favari L 2003 Silymarin increases antioxidant enzymes in alloxan-induced diabetes in rat pancreas. Comp Biochem Physiol C Toxicol Pharmacol 136:205–212
Bosisio E, Benelli C, Pirola O 1992 Effect of the flavanolignans of Silybum marianum L. on lipid peroxidation in rat liver microsomes and freshly isolated hepatocytes. Pharmacol Res 25:147–154
Wang MJ, Lin WW, Chen HL, Chang YH, Ou HC, Kuo JS, Hong JS, Jeng KC 2002 Silymarin protects dopaminergic neurons against lipopolysaccharide-induced neurotoxicity by inhibiting microglia activation. Eur J Neurosci 16:2103–2112
Raso GM, Meli R, Di Carlo G, Pacilio M, Di Carlo R 2001 Inhibition of inducible nitric oxide synthase and cyclooxygenase-2 expression by flavonoids in macrophage J774A.1. Life Sci 68:921–931
Manna SK, Mukhopadhyay A, Van NT, Aggarwal BB 1999 Silymarin suppresses TNF-induced activation of NF-B, c-Jun N-terminal kinase, and apoptosis. J Immunol 163:6800–6809
Soto CP, Perez BL, Favari LP, Reyes JL 1998 Prevention of alloxan-induced diabetes mellitus in the rat by silymarin. Comp Biochem Physiol C Pharmacol Toxicol Endocrinol 119:125–129
Giannoukakis N, Rudert WA, Trucco M, Robbins PD 2000 Protection of human islets from the effects of interleukin-1? by adenoviral gene transfer of an I B repressor. J Biol Chem 275:36509–36513
Larsen CM, Wadt KA, Juhl LF, Andersen HU, Karlsen AE, Su MS, Seedorf K, Shapiro L, Dinarello CA, Mandrup-Poulsen T 1998 Interleukin-1?-induced rat pancreatic islet nitric oxide synthesis requires both the p38 and extracellular signal-regulated kinase 1/2 mitogen-activated protein kinases. J Biol Chem 273:15294–15300
Ammendrup A, Maillard A, Nielsen K, Aabenhus Andersen N, Serup P, Dragsbaek Madsen O, Mandrup-Poulsen T, Bonny C 2000 The c-Jun amino-terminal kinase pathway is preferentially activated by interleukin-1 and controls apoptosis in differentiating pancreatic ?-cells. Diabetes 49:1468–1476
Heitmeier MR, Scarim AL, Corbett JA 1999 Prolonged STAT1 activation is associated with interferon- priming for interleukin-1-induced inducible nitric-oxide synthase expression by islets of Langerhans. J Biol Chem 274:29266–29273
Cottet S, Dupraz P, Hamburger F, Dolci W, Jaquet M, Thorens B 2001 SOCS-1 protein prevents Janus kinase/STAT-dependent inhibition of ? cell insulin gene transcription and secretion in response to interferon-. J Biol Chem 276:25862–25870
Karlsen AE, Ronn SG, Lindberg K, Johannesen J, Galsgaard ED, Pociot F, Nielsen JH, Mandrup-Poulsen T, Nerup J, Billestrup N 2001 Suppressor of cytokine signaling 3 (SOCS-3) protects ?-cells against interleukin-1?- and interferon--mediated toxicity. Proc Natl Acad Sci USA 98:12191–12196
Gao J, Morrison DC, Parmely TJ, Russell SW, Murphy WJ 1997 An interferon--activated site (GAS) is necessary for full expression of the mouse iNOS gene in response to interferon- and lipopolysaccharide. J Biol Chem 272:1226–1230
Mullen Y, Arita S, Kenmochi T, Une S, Smith CV 1997 A two-step digestion process and LAP-I cold preservation solution for human islet isolation. Ann Transplant 2:40–45
Mosmann T 1983 Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays. J Immunol Methods 65:55–63
Matsuda T, Suzuki H, Oishi I, Kani S, Kuroda Y, Komori T, Sasaki A, Watanabe K, Minami Y 2003 The receptor tyrosine kinase Ror2 associates with the melanoma-associated antigen (MAGE) family protein Dlxin-1 and regulates its intracellular distribution. J Biol Chem 278:29057–29064
Lortz S, Tiedge M, Nachtwey T, Karlsen AE, Nerup J, Lenzen S 2000 Protection of insulin-producing RINm5F cells against cytokine-mediated toxicity through overexpression of antioxidant enzymes. Diabetes 49:1123–1130
Igaz P, Toth S, Falus A 2001 Biological and clinical significance of the JAK-STAT pathway; lessons from knockout mice. Inflamm Res 50:435–441
Flodstrom-Tullberg M, Yadav D, Hagerkvist R, Tsai D, Secrest P, Stotland A, Sarvetnick N 2003 Target cell expression of suppressor of cytokine signaling-1 prevents diabetes in the NOD mouse. Diabetes 52:2696–2700
Kang JS, Jeon YJ, Park SK, Yang KH, Kim HM 2004 Protection against lipopolysaccharide-induced sepsis and inhibition of interleukin-1? and prostaglandin E2 synthesis by silymarin. Biochem Pharmacol 67:175–181
Kang JS, Jeon YJ, Kim HM, Han SH, Yang KH 2002 Inhibition of inducible nitric-oxide synthase expression by silymarin in lipopolysaccharide-stimulated macrophages. J Pharmacol Exp Ther 302:138–144
Lahti A, Jalonen U, Kankaanranta H, Moilanen E 2003 c-Jun NH2-terminal kinase inhibitor anthra(1,9-cd)pyrazol-6(2H)-one reduces inducible nitric-oxide synthase expression by destabilizing mRNA in activated macrophages. Mol Pharmacol 64:308–315
Holzberg D, Knight CG, Dittrich-Breiholz O, Schneider H, Dorrie A, Hoffmann E, Resch K, Kracht M 2003 Disruption of the c-JUN-JNK complex by a cell-permeable peptide containing the c-JUN domain induces apoptosis and affects a distinct set of interleukin-1-induced inflammatory genes. J Biol Chem 278:40213–40223
Bennett BL, Sasaki DT, Murray BW, O’Leary EC, Sakata ST, Xu W, Leisten JC, Motiwala A, Pierce S, Satoh Y, Bhagwat SS, Manning AM, Anderson DW 2001 SP600125, an anthrapyrazolone inhibitor of Jun N-terminal kinase. Proc Natl Acad Sci USA 98:13681–13686
Bonny C, Oberson A, Negri S, Sauser C, Schorderet DF 2001 Cell-permeable peptide inhibitors of JNK: novel blockers of ?-cell death. Diabetes 50:77–82
Haefliger JA, Tawadros T, Meylan L, Gurun SL, Roehrich ME, Martin D, Thorens B, Waeber G 2003 The scaffold protein IB1/JIP-1 is a critical mediator of cytokine-induced apoptosis in pancreatic ? cells. J Cell Sci 116:1463–1469
Nikulina MA, Sandhu N, Shamim Z, Andersen NA, Oberson A, Dupraz P, Thorens B, Karlsen AE, Bonny C, Mandrup-Poulsen T 2003 The JNK binding domain of islet-brain 1 inhibits IL-1 induced JNK activity and apoptosis but not the transcription of key proapoptotic or protective genes in insulin-secreting cell lines. Cytokine 24:13–24
Kaneto H, Xu G, Fujii N, Kim S, Bonner-Weir S, Weir GC 2002 Involvement of c-Jun N-terminal kinase in oxidative stress-mediated suppression of insulin gene expression. J Biol Chem 277:30010–30018
Darville MI, Eizirik DL 1998 Regulation by cytokines of the inducible nitric oxide synthase promoter in insulin-producing cells. Diabetologia 41:1101–1108
Paludan SR, Malmgaard L, Ellermann-Eriksen S, Bosca L, Mogensen SC 2001 Interferon (IFN)-gamma and herpes simplex virus/tumor necrosis factor- synergistically induce nitric oxide synthase 2 in macrophages through cooperative action of nuclear factor- B and IFN regulatory factor-1. Eur Cytokine Netw 12:297–308
Yu X, Kennedy RH, Liu SJ 2003 JAK2/STAT3, not ERK1/2, mediates interleukin-6-induced activation of inducible nitric-oxide synthase and decrease in contractility of adult ventricular myocytes. J Biol Chem 278:16304–16309
Yoshida Y, Kumar A, Koyama Y, Peng H, Arman A, Boch JA, Auron PE 2004 Interleukin 1 activates STAT3/nuclear factor-B cross-talk via a unique TRAF6- and p65-dependent mechanism. J Biol Chem 279:1768–1776
Yu Z, Zhang W, Kone BC 2002 Signal transducers and activators of transcription 3 (STAT3) inhibits transcription of the inducible nitric oxide synthase gene by interacting with nuclear factor B. Biochem J 367:97–105
Ymamoka K, Otsuka T, Niiro H, Arinobu Y, Niho Y, Hamasaki N, Izuhara K 1998 Activation of STAT5 by lipopolysaccharide through granulocyte-macrophage colony-stimulating factor production in human monocytes. J Immunol 160:838–845
Luo G, Yu-Lee L 2000 Stat5b inhibits NFB-mediated signaling. Mol Endocrinol 14:114–123
Zafarullah M, Li WQ, Sylvester J, Ahmad M 2003 Molecular mechanisms of N-acetylcysteine actions. Cell Mol Life Sci 60:6–20
Maziere C, Alimardani G, Dantin F, Dubois F, Conte MA, Maziere JC 1999 Oxidized LDL activates STAT1 and STAT3 transcription factors: possible involvement of reactive oxygen species. FEBS Lett 448:49–52
Matsubara T, Ziff M 1986 Increased superoxide anion release from human endothelial cells in response to cytokines. J Immunol 137:3295–3298
Azevedo-Martins AK, Lortz S, Lenzen S, Curi R, Eizirik DL, Tiedge M 2003 Improvement of the mitochondrial antioxidant defense status prevents cytokine-induced nuclear factor-B activation in insulin-producing cells. Diabetes 52:93–101
Cho JY, Kim PS, Park J, Yoo ES, Baik KU, Kim YK, Park MH 2000 Inhibitor of tumor necrosis factor- production in lipopolysaccharide-stimulated RAW264.7 cells from Amorpha fruticosa. J Ethnopharmacol 70:127–133
Schumann J, Prockl J, Kiemer AK, Vollmar AM, Bang R, Tiegs G 2003 Silibinin protects mice from T cell-dependent liver injury. J Hepatol 39:333–340(Takeru Matsuda, Kevin Fer)
Address all correspondence and requests for reprints to: Yoko Mullen, M.D., Ph.D., Southern California Islet Cell Resources Center, Department of Diabetes, Endocrinology and Metabolism, City of Hope National Medical Center/Beckman Research Institute, 1500 East Duarte Road, Duarte, California 91010. E-mail: ymullen@coh.org.
Abstract
Silymarin is a polyphenolic flavonoid that has a strong antioxidant activity and exhibits anticarcinogenic, antiinflammatory, and cytoprotective effects. Although its hepatoprotective effect has been well documented, the effect of silymarin on pancreatic ?-cells is largely unknown. In this study, the effect of silymarin on IL-1? and/or interferon (IFN)--induced ?-cell damage was investigated using RINm5F cells and human islets. IL-1? and/or IFN- induced cell death in a time-dependent manner in RINm5F cells. The time-dependent increase in cytokine-induced cell death appeared to correlate with the time-dependent nitric oxide (NO) production. Silymarin dose-dependently inhibited both cytokine-induced NO production and cell death in RINm5F cells. Treatment of human islets with a combination of IL-1? and IFN- (IL-1?+IFN-), for 48 h and 5 d, resulted in an increase of NO production and the impairment of glucose-stimulated insulin secretion, respectively. Silymarin prevented IL-1?+IFN--induced NO production and ?-cell dysfunction in human islets. These cytoprotective effects of silymarin appeared to be mediated through the suppression of c-Jun NH2-terminal kinase and Janus kinase/signal transducer and activator of transcription pathways. Our data show a direct cytoprotective effect of silymarin in pancreatic ?-cells and suggest that silymarin may be therapeutically beneficial for type 1 diabetes.
Introduction
THE PATHOGENESIS OF type 1 diabetes is characterized by an inflammatory reaction that is caused, at least in part, by inflammatory cytokines produced by infiltrating T-lymphocytes and/or macrophages in and around islets (1, 2). Inflammatory cytokines, such as IL-1? and interferon (IFN)- produced by these cells, initiate a variety of signal cascades in ?-cells that lead to ?-cell dysfunction and destruction. The generation of oxygen free radicals is well known to be one of the main mechanisms of such cytokine-mediated ?-cell damage (3, 4). In addition, nitric oxide (NO) produced through the activation of inducible NO synthase (iNOS) also appears to participate in cytokine-mediated toxicity (5, 6). Besides its direct toxicity, NO reacts with superoxide to form peroxynitrite, which has a much stronger oxidant activity and mediates ?-cell destruction in type 1 diabetes (7, 8). In fact, iNOS inhibitors, such as NG-monomethyl-L-arginine (L-NMMA), attenuate cytokine-induced ?-cell dysfunction and islet degeneration (5, 6). Therefore, not only free radical scavenging actions, but also inhibition of iNOS activation, would be beneficial to prevent or delay ?-cell destruction in the development of type 1 diabetes.
Silymarin is a polyphenolic flavonoid extracted from the milk thistle that has a strong antioxidant activity and exhibits cytoprotective, antiinflammatory, and anticarcinogenic effects (9, 10). In addition to its free radical scavenging properties, silymarin increases antioxidant enzymes, such as superoxide dismutase (SOD) and catalase, and inhibits lipid peroxidation (10, 11, 12). It inhibits lipopolysaccharide (LPS)-induced iNOS expression and NO production in microglia and macrophages by blocking nuclear factor B (NF-B) activation (13, 14). A modulatory effect of silymarin on cytokine-induced activation of MAPKs has also been reported (15). Silymarin has been clinically used for the treatment of acute and chronic liver disease because of its hepatoprotective effects. In the pancreas, Soto et al. (16) showed a preventive effect of silymarin against alloxan-induced diabetes in rats. However, its effect on cytokine-induced toxicity in pancreatic ?-cells and the mechanisms involved are largely unknown.
IL-1? activates various cytotoxic signaling pathways, such as NF-B and MAPKs, in ?-cells. Inhibition of NF-B suppresses IL-1?-induced apoptosis, at least in part, through the suppression of iNOS expression in human islets (17). Activation of MAPKs, including c-Jun NH2-terminal kinase (JNK), p38 kinase (p38), and ERK, also mediates IL-1?-induced iNOS expression and/or apoptosis in ?-cells (18, 19). IFN- alone does not stimulate iNOS expression in rat islets, but it primes for IL-1?-induced iNOS expression (20). Treatment with IFN- reduces insulin mRNA levels, glucose-stimulated insulin secretion, and viability in rodent pancreatic ?-cells (21, 22). These deleterious effects of IFN- in ?-cells are thought to be mediated through the activation of the Janus kinase (JAK)/signal transducer and activator of transcription (STAT) pathway. Upon binding of IFN- to its receptor, the receptor-associated JAK isoforms (p38 kinase1 and JAK2) are activated and phosphorylate themselves and the IFN- receptor, which serve as the binding sites for STAT proteins. Phosphorylated STAT proteins form dimers, translocate to the nucleus, and activate the transcription of target genes, including iNOS (23).
In this study, we examined the effect of silymarin on IL-1? and/or IFN--induced cytotoxicity using a rat insulinoma cell line, RINm5F cells, and human islets. Our data demonstrate the cytoprotective effect of silymarin against cytokine-induced toxicity in ?-cells. Silymarin also suppresses cytokine-induced iNOS expression. JNK and JAK/STAT, but not NF-B, pathways appear to be involved in the cytoprotective action of silymarin.
Materials and Methods
Cell culture and materials
The rat RINm5F insulinoma cells were cultured in RPMI 1640 supplemented with 10% fetal bovine serum (FBS) and antibiotic-antimycotic. Silymarin was obtained from Sigma (Milwaukee, WI) and dissolved in dimethylsulfoxide (DMSO) at 100 mM. Human recombinant IL-1? and IFN- were purchased from R&D systems Inc. (Minneapolis, MN). Rat recombinant IFN- was from BD Biosciences (San Jose, CA). SP600125 (JNK inhibitor), SB203580 (p38 inhibitor), and L-NMMA (iNOS inhibitor) were from Calbiochem (San Diego, CA). U0126 (MAPK kinase/ERK inhibitor) was from Cell Signaling Technology (Beverly, MA). Rabbit polyclonal antibodies to total and phosphorylated JNK, p38, and ERK and antibodies to phosphorylated STAT1, STAT3, and STAT5 were from Cell Signaling Technology. Rabbit polyclonal antibodies to STAT1, STAT3, STAT5, IB, and ?-actin were from Santa Cruz Biotechnology (Santa Cruz, CA). Antibodies to total and phosphorylated STAT1 react with STAT1 (91 kDa) and STAT1? (84 kDa). Antitotal STAT3 antibody recognizes STAT3 (92 kDa), and antiphosphorylated STAT3 antibody recognizes both STAT3 (92 kDa) and ? (83 kDa). Antitotal and phosphorylated STAT5 antibodies recognize STAT5a (94 kDa) and STAT5b (92 kDa). Rabbit polyclonal antibody to iNOS was purchased from BD Biosciences.
Human islets
Human islets were provided for research use by the Southern California Islet Cell Resources Center, City of Hope (Duarte, CA). Islets were isolated by the two-step digestion method (24) and cultured in serum-free CMRL1066 medium (Mediatech, Inc., Holly Hill, FL) for 2 d before being released for research use. These islets were then cultured in Ham’s F12 medium supplemented with 10 mM HEPES, antibiotic antimycotic, and 10% FBS for an additional 2–3 d before being used in experiments. The islet numbers were expressed as the number equivalent to islets 150 μm in diameter (IEQ). Only islets with greater than 80% purity, as determined by dithizone staining, were used.
Cell treatment
RINm5F cells were treated with silymarin or vehicle (DMSO) at the indicated concentration for 2 h before stimulation. In some experiments, RINm5F cells were treated with SP600125, SB203580, or U0126 at 10 μM or with L-NMMA at 0.5 mM for 30 min before stimulation. Then, cells were stimulated with human IL-1? (50 U/ml) and/or rat IFN- (100 U/ml) for the indicated period. Human islets were incubated with silymarin or DMSO at the indicated concentration for 2 h and then stimulated with a combination of human IL-1? (75 U/ml) and human IFN- (750 U/ml) for the indicated period.
Assay for NF-B binding activity
After 30-min stimulation, RINm5F cells were washed with ice-cold PBS, and nuclear extracts were prepared using NE-PER Nuclear and Cytoplasmic Extract (PIERCE, Rockford, IL). The protein concentration of each sample was measured using the Bio-Rad Protein Assay (Bio-Rad, Hercules, CA). Equal amounts of nuclear extracts from different samples were used. NF-B DNA binding activity was measured using TransAM NF-B Kit (Active Motif, Carlsbad, CA). Results were expressed as the OD.
Assessment of cell viability
The viability of RINm5F cells was determined by 3-(4,5-dimethylthiazolyl-2) 2,5-diphenyltetrazolium bromide (MTT) assay, as described previously (25), and propidium iodide (PI) staining. Cells were plated in a 96-well plate in triplicate (20,000 cells/well) and cultured overnight before cytokine stimulation. The viability was assessed after 48- and 72-h culture after cytokine stimulation. In MTT assay, viability was expressed as a percentage of the MTT absorbance of the cells without cytokine stimulation. In PI staining, cells were incubated with PI (2.5 μg/ml) for 10 min and then examined under a fluorescence microscope. The percentage of dead cells was determined by counting all of the PI-positive cells divided by all of the cells in the bright-field pictures. A minimum of 500 cells was counted in at least three random fields for each experiment.
Nitrite determination
RINm5F cells were seeded at a density of 20,000 cells/well in 100 μl culture medium in triplicate, and incubated with cytokines for 48 or 72 h. Human islets (600 IEQ/600 μl culture medium) were cultured in duplicate for 48 h after stimulation. Nitrite production in the supernatant was measured using a Griess Reagent Kit (Molecular Probes, Eugene, OR).
Semiquantitative PCR
RINm5F cells and human islets were collected 12 and 8 h, respectively, after stimulation. Total RNA was extracted from cells using TRI REAGENT (Molecular Research Center, Inc., Cincinnati, OH). Two micrograms of RNA from each sample were then reverse-transcribed into first-strand cDNA in 20 μl solution using Superscript RNase H– reverse transcriptase (Invitrogen, Carlsbad, CA). The final cDNA products were then diluted by adding 80 μl H2O. PCR was performed using the FailSafe PCR System (EPICENTRE, Madison, WI). A standard 25-μl PCR solution contained 1.5 μl cDNA, 30 pmol each of forward and reverse primers, 12.5 μl Premix D, and 0.25 μl Enzyme Mix. The following primers were used: rat iNOS (297 bp) forward 5'-CCA ACC GGA GAA GGG GAC GAA CT-3', reverse 5'-GGA GGG TGG TGC GGC TGG AC-3'; rat ?-actin (764 bp) forward 5'-TTG TAA CCA ACT GGG ACG ATA TGG-3', reverse 5'-GAT CTT GAT CTT CAT GGT GCT AGG-3'; human iNOS (236 bp) forward 5'-ACA TTG ATC AGA AGC TGT CCC AC-3', reverse 5'-CAA AGG CTG TGA GTC CTG CAC-3'; human glyceraldehyde-3-phosphate dehydrogenase (600 bp) forward 5'-CCA CCC ATG GCA AAT TCC ATG GCA-3', reverse 5'-TCT AGA CGG CAG GTC AGG TCC ACC-3'. The cycle number was determined to be in the linear range of amplification for each primer pair. Electrophoresis of PCR products were performed in 1.5% agarose gel containing 0.5 mg/ml ethidium bromide. The band intensities were measured and quantitated with an Alpha Innotech densitometer (San Leandro, CA).
Western blotting
After cytokine stimulation, RINm5F cells or human islets were harvested at the indicated time and washed twice with ice-cold PBS. Nuclear and cytoplasmic extracts were prepared using NE-PER Nuclear and Cytoplasmic Extract. For whole-cell lysate, cells were lysed with lysis buffer containing 50 mM Tris-HCl (pH 7.4), 0.5% (vol/vol) NP-40, 150 mM NaCl, 5 mM EDTA, 50 mM NaF, 1 mM Na3VO4, 1 mM phenylmethylsulfonylfluoride, 10 μg/ml leupeptin, and 10 μg/ml aprotinin for 30 min at 4 C. After centrifugation for 15 min at 12,000 rpm, the supernatants were recovered and eluted with Laemmli sample buffer. The protein concentration of each sample was measured as described above, and equal amounts of protein were used. Western blotting was performed as described previously (26). The band intensities were measured and quantitated with an Alpha Innotech densitometer.
Glucose-stimulated insulin secretion
Handpicked islets were treated with silymarin or vehicle (DMSO) for 2 h, then stimulated with IL-1? and IFN-, and cultured for 5 d. Islets were then washed three times with basal media (RPMI 1640 medium containing 3.3 mM glucose, 10 mM HEPES, antibiotic-antimycotic, 1% FBS) and preincubated with the same basal media for 1 h. After wash, islets were incubated with basal media for 1 h, followed by a second incubation with high-glucose media (RPMI 1640 medium containing 19.5 mM glucose, 10 mM HEPES, antibiotic-antimycotic, 1% FBS) for 1 h. The insulin released into the medium was measured using a human insulin ELISA kit (ALPCO Diagnostics, Windham, NH). Each experiment was performed in duplicate. In total, three different experiments using three separate human islet preparations were performed. The fold increase in the secreted insulin levels from 3.3 to 19.5 mM glucose between groups was compared for statistical analysis.
Statistical analysis
Data are expressed as mean ± SE. Statistical differences between groups were analyzed by ANOVA followed by Bonferroni’s test. P < 0.05 was considered significant.
Results
Silymarin protects RINm5F cells against cytokine-induced cell death
First, we wanted to determine whether silymarin could protect pancreatic ?-cells from cytokine-induced cell death. RINm5F insulinoma cells incubated with IL-1? alone for 72 h resulted in a 20–30% decrease in cell viability compared with the untreated control cells, although no significant decrease in viability was observed after 48 h incubation (Fig. 1A). In contrast, incubation with IFN- alone for 48 and 72 h led to 50% and 60% decrease in viability, respectively. Incubation with a combination of IL-1? and IFN- (IL-1?+IFN-) for 72 h resulted in further decrease to 20–30% viability. Pretreatment with silymarin protected against cytokine-induced cell death in a dose-dependent manner (Fig. 1B). The maximum effect was achieved at the concentration of 100 μM silymarin. Similar results were obtained by PI staining, confirming that the observed changes in MTT assay resulted from the changes in cell death but not in cell proliferation or mitochondrial metabolic activity (Fig. 1C). These results indicate that silymarin has a cytoprotective effect against cytokine-mediated cytotoxicity in RINm5F cells.
FIG. 1. Silymarin protects against cytokine-induced cell death of RINm5F cells. A, RINm5F cells were stimulated with IL-1? (50 U/ml), or rat IFN- (100 U/ml), or IL-1? (50 U/ml) + rat IFN- (100 U/ml). Viability was assessed by MTT assay at 48 h and 72 h after stimulation, and results were expressed as a percentage of the MTT absorbance of the control unstimulated cells. Data are means ± SE from four to five individual experiments. *, P < 0.05 vs. control cells without cytokine(s). B, RINm5F cells were treated with silymarin or DMSO for 2 h at the indicated concentrations, followed by stimulation with IL-1? (50 U/ml), or rat IFN- (100 U/ml), or IL-1? (50 U/ml) + rat IFN- (100 U/ml). Viability was assessed by MTT assay at 72 h after stimulation. Results were expressed as a percentage of the MTT absorbance of the control unstimulated cells treated with the same concentration of silymarin. C, RINm5F cells were treated with 100 μM silymarin or DMSO for 2 h, followed by stimulation with IL-1? (50 U/ml), or rat IFN- (100 U/ml), or IL-1? (50 U/ml) + rat IFN- (100 U/ml). Viability was assessed by PI staining at 72 h after stimulation. Results were expressed as a percentage of the dead (PI positive) cells. Data are means ± SE from four to five individual experiments. *, P < 0.05; **, P < 0.01 vs. cytokine-stimulated cells without silymarin.
Silymarin inhibits cytokine-induced iNOS expression and NO production in RINm5F cells
Based on previous studies, we hypothesized that the observed cytoprotective effect of silymarin might be attributed to inhibition of NO production. Therefore, we examined the effect of silymarin on cytokine-induced NO production. As previously shown (27), IL-1? alone induced time-dependent nitrite production in RINm5F cells (Fig. 2A). However, IFN- induced higher levels of nitrite production than IL-1? (Fig. 2A), especially at 48 h after stimulation, when the levels of nitrite production were approximately three times higher than IL-1? alone. Treatment with IL-1?+IFN- resulted in slightly increased nitrite production compared with IFN- alone. This higher rate of NO production induced by IFN- or IL-1?+IFN- in the early incubation period (48 h) appeared to correlate with higher and earlier induction of cell death in RINm5F cells compared with IL-1? alone. Treatment with silymarin inhibited cytokine-induced nitrite production in a dose-dependent manner, with maximum inhibition at 100 μM (Fig. 2B). As shown in Fig. 2, C and D, silymarin also inhibited the cytokine-induced expression of both iNOS mRNA and iNOS protein. Addition of iNOS inhibitor, L-NMMA (0.5 mM), in the culture medium with cytokine almost completely blocked cytokine induced-NO production but only partially prevented cell death (Fig. 3, A and B), suggesting that cytokine-induced cell death was only partially mediated by NO production in RINm5F cells. These results suggest that silymarin protects RINm5F cells against cytokine-induced cell death, at least in part, through the suppression of NO production by down-regulation of iNOS expression at the transcriptional level.
FIG. 2. Silymarin inhibits cytokine-induced iNOS expression and NO production in RINm5F cells. A, RINm5F cells (20,000 cells/well) in a 100-μl medium were stimulated with IL-1? (50 U/ml), rat IFN- (100 U/ml), or IL-1? (50 U/ml) + rat IFN- (100 U/ml). Nitrite production in the medium was determined using Griess reagent at 48 h and 72 h after stimulation. B, RINm5F cells were treated with silymarin or DMSO for 2 h at the indicated concentrations, followed by stimulation with IL-1? (50 U/ml), or rat IFN- (100 U/ml), or IL-1? (50 U/ml) + rat IFN- (100 U/ml). Nitrite production in the medium was determined at 72 h after stimulation. C and D, RINm5F cells were treated with 100 μM silymarin or DMSO for 2 h, followed by stimulation with IL-1? (50 U/ml), or rat IFN- (100 U/ml), or IL-1? (50 U/ml) + rat IFN- (100 U/ml). Total RNA was isolated at 12 h after isolation, and iNOS mRNA expression was determined by RT-PCR (C). Whole-cell lysates were recovered at 24 h after stimulation, and iNOS protein expression was determined by Western blotting analysis (D). The relative values of iNOS/actin mRNA or protein levels were expressed as means in bar graphs. Nitrite data are means ± SE from four to five individual experiments. RT-PCR and Western blotting data are representative from three individual experiments. *, P < 0.05; **, P < 0.01 vs. cytokine-stimulated cells without silymarin.
FIG. 3. Effect of the iNOS inhibitor, L-NMMA, on cytokine-induced NO production (A) and cell death (B) in RINm5F cells. RINm5F cells (20,000 cells/well) in a 100-μl medium were stimulated with IL-1? (50 U/ml), or rat IFN- (100 U/ml), or IL-1? (50 U/ml) + rat IFN- (100 U/ml) in the presence or absence of 0.5 mM L-NMMA. A, Nitrite production in the medium was determined using Griess reagent at 72 h after stimulation. B, Viability was assessed by MTT assay at 72 h after stimulation, and results were expressed as a percentage of the MTT absorbance of the control unstimulated cells. Data are means ± SE from three individual experiments. *, P < 0.001; **, P < 0.05 vs. the same cytokine-stimulated cells without L-NMMA.
No effect of silymarin on IL-1?-induced NF-B activation in RINm5F cells
The NF-B pathway plays a crucial role in IL-1?-induced iNOS expression and cell death in ?-cells. Because silymarin has been reported to suppress NF-B activation induced by various stimuli in many types of cells (15), the effect of silymarin on cytokine-induced NF-B activation in RINm5F cells was explored. Stimulation with IL-1? for 30 min induced an approximately 4-fold increase in NF-B activation compared with the untreated control cells (Fig. 4A). Pretreatment with silymarin had no effect on IL-1?-induced NF-B activation. This result was also confirmed by Western blotting analysis showing that IL-1?-induced degradation of IB, which precedes the translocation of NF-B to the nucleus, was not affected by silymarin (Fig. 4B). IFN- alone did not induce NF-B activation, and it did not affect IL-1?-induced NF-B activation when used in combination (Fig. 4A).
FIG. 4. Effect of silymarin on IL-?-induced NF-B activation in RINm5F cells. A, RINm5F cells were treated with 100 μM silymarin or DMSO for 2 h, followed by stimulation with IL-1? (50 U/ml), or rat IFN- (100 U/ml), or IL-1? (50 U/ml) + rat IFN- (100 U/ml). Nuclear extracts were recovered after 30 min stimulation, and NF-B activity was determined as described in Materials and Methods. Results were expressed as the OD. NF-B activity data are means ± SE from three individual experiments. B, RINm5F cells were treated with 100 μM silymarin or DMSO for 2 h, followed by stimulation with IL-1? (50 U/ml). Whole-cell lysates were recovered at the indicated time after stimulation, and IB expression was determined by Western blotting analysis. Western blotting data are representative from three individual experiments.
Silymarin modulates IL-1?-induced MAPKs activation
Besides the NF-B pathway, MAPK pathways are also known to be involved in the regulation of IL-1?-induced iNOS expression and cell death in ?-cells (18). Therefore, the effect of silymarin on IL-1?-induced MAPKs activation was investigated. IL-1? activated all three MAPK pathways, including JNK, p38, and ERK (Fig. 5A). In all cases, the maximum level of phosphorylation was observed at 30 min after stimulation. Pretreatment with silymarin attenuated IL-1?-induced phosphorylation of JNK, although it did not completely abolish JNK phosphorylation. On the other hand, IL-1?-induced phosphorylation of ERK was further augmented by silymarin pretreatment. Phosphorylation of p38 was not affected. Next, we compared the effect of specific pharmacological inhibitors of MAPKs with silymarin on the suppression of IL-1?-induced nitrite production and cell death. Pretreatment of RINm5F cells with 10 μM SP60025 (JNK inhibitor) or 10 μM SB203580 (p38 inhibitor) partially inhibited IL-1?-induced nitrite production (Fig. 5B). Furthermore, IL-1?-induced cell death was significantly reversed by pretreatment with SP600125 or SB203580 (Fig. 5C). In contrast, the MAPK kinase/ERK inhibitor, U0126, had no effect on either nitrite production or cell viability. These results suggest that silymarin may, in part, inhibit IL-1?-induced NO production and cell death through the suppression of JNK activation.
FIG. 5. Modulation of IL-1?-induced MAPKs activation by silymarin in RINm5F cells. A, RINm5F cells were treated with 100 μM silymarin or DMSO for 2 h, followed by stimulation with IL-1? (50 U/ml). Whole-cell lysates were recovered at the indicated time after stimulation, and phosphorylation of MAPKs was detected by Western blotting analysis. The relative values of p-JNK/JNK, p-p38/p38, and p-ERK/ERK were expressed as means in line graphs. B and C, RINm5F cells were treated with 10 μM SP600125, SB203580, U0126, or DMSO for 30 min, or with 100 μM silymarin for 2 h. Cells were then stimulated with IL-1? (50 U/ml) and cultured for 72 h. Nitrite production in the medium was determined using Griess reagent (B). Viability was determined by MTT assay, and results were expressed as a percentage of the MTT absorbance of the control unstimulated cells (C). Nitrite and viability data are means ± SE from three individual experiments. D, RINm5F cells were treated with 100 μM silymarin or DMSO for 2 h, followed by stimulation with IL-1? (50 U/ml), or rat IFN- (100 U/ml), or IL-1? (50 U/ml) + rat IFN- (100 U/ml). Whole-cell lysates were recovered 20 min after stimulation, and phosphorylation of MAPKs was detected by Western blotting analysis. The relative values of p-JNK/actin, p-p38/actin, and p-ERK/actin were expressed as means in bar graphs. Western blotting data are representative from three individual experiments. *, P < 0.05 vs. cytokine-stimulated cells treated with DMSO.
IFN- alone did neither activate JNK or p38, nor augment IL-1?-induced phosphorylation of JNK or p38, but slightly activated ERK (Fig. 5D). Silymarin pretreatment augmented the phosphorylation levels of ERK and, to a lesser extent, of p38.
Silymarin inhibits IFN--induced activation and nuclear translocation of STAT1, STAT3, and STAT5
IFN--elicited JAK/STAT signaling pathways have been reported to be involved in cytokine-induced inflammatory responses, including iNOS expression (28). STAT1 activation has been especially implicated in IFN--mediated cytotoxicity in ?-cells (20, 21, 29). Therefore, we investigated the effect of silymarin on IFN--induced activation of STAT proteins in RINm5F cells. Phosphorylation of STAT1/?, STAT3/?, and STAT5a/b was observed 20 min after IFN- stimulation (Fig. 6A). IL-1? alone had no effect, nor did the addition of IL-1? to IFN- increase IFN--induced phosphorylation of STAT proteins. Pretreatment with silymarin attenuated IFN--induced phosphorylation of all three STAT proteins, though the effect on STAT1 was smaller than that of STAT3 or STAT5. However, the IFN--induced translocation of STAT1, STAT3, and STAT5 proteins was similarly attenuated by silymarin pretreatment (Fig. 6B). These results imply that inhibition by silymarin of IFN--induced iNOS expression and subsequent cell death is due to suppression of STAT proteins activation.
FIG. 6. Silymarin inhibits IFN--induced activation and nuclear translocation of STAT1, STAT3, and STAT5 in RINm5F cells. A, RINm5F cells were treated with 100 μM silymarin or DMSO for 2 h, followed by stimulation with IL-1? (50 U/ml), or rat IFN- (100 U/ml), or IL-1? (50 U/ml) + rat IFN- (100 U/ml). Whole-cell lysates were recovered 20 min after stimulation, and phosphorylation of STAT proteins was detected by Western blotting analysis. The relative values of p-STAT1/STAT1, p-STAT3/STAT3, and p-STAT5/STAT5 were expressed as means in bar graphs. B, RINm5F cells were treated with 100 μM silymarin or DMSO for 2 h, followed by stimulation with rat IFN- (100 U/ml). Nuclear (N.) and cytoplasmic (C.) extracts were recovered at 30 min after stimulation, and translocation of STAT proteins to the nucleus was detected by Western blotting analysis. The densitometric values of nuclear STAT1, STAT3, and STAT5 proteins were expressed as means in bar graphs. Western blotting data are representative from three individual experiments.
Silymarin inhibits cytokine-induced iNOS expression, NO production, and activation of JNK and STAT proteins in human islets
Our results consistently showed the suppressive effect of silymarin on IL-1? and/or IFN--induced iNOS expression and NO production in RINm5F cells. To extend these studies to human cells, we next determined whether silymarin could also suppress IL-1?+IFN--induced iNOS expression and NO production in human islets. We found that, similar to results with RINm5F cells, silymarin pretreatment decreased cytokine-induced nitrite production by human islets in a dose-dependent manner (Fig. 7A). Semiquantitative RT-PCR results showed that silymarin also suppressed IL-1?+IFN--induced iNOS mRNA expression (Fig. 7B). Stimulation with IL-1?+IFN- resulted in activation of all three MAPKs and STAT1, 3, and 5a proteins within 30 min (Fig. 7C), but silymarin only attenuated JNK phosphorylation, whereas no effect was observed on phosphorylation of p38 or ERK. On the other hand, phosphorylation of all three STAT proteins was partially inhibited by silymarin pretreatment in human islets (Fig. 7C). Still, silymarin pretreatment did not completely abolish phosphorylation of JNK or STAT proteins in human islets.
FIG. 7. Silymarin inhibits cytokine-induced iNOS expression, NO production, and activation of JNK and STAT proteins in human islets. A, Human islets (600 IEQ/600 μl medium) were treated with silymarin for 2 h at the indicated concentrations, followed by stimulation with IL-1? (75 U/ml) + IFN- (750 U/ml). Nitrite production in the medium was determined using Griess reagent at 48 h after stimulation. Nitrite data are mean ± SE from three individual experiments using three different islet preparations. B and C, Human islets were treated with 100 μM silymarin or DMSO for 2 h, followed by stimulation with IL-1? (75 U/ml) + IFN- (750 U/ml). Total RNA was recovered at 12 h after stimulation, and iNOS mRNA expression was determined by RT-PCR (B). Whole-cell lysates were recovered at 30 min after stimulation, and phosphorylation of MAPKs and STAT proteins was detected by Western blotting analysis (C). The relative values of iNOS/glyceraldehyde-3-phosphate dehydrogenase, p-JNK/actin, p-p38/actin, p-ERK/actin, p-STAT1/actin, p-STAT3/actin, and p-STAT5/actin were expressed as means in bar graphs. RT-PCR and Western blotting data are representative from three individual experiments using three different islet preparations. *, P < 0.05; **, P < 0.001 vs. cytokine-stimulated islets without silymarin.
Protective effect of silymarin against cytokine-induced impairment of glucose-stimulated insulin secretion by human islets
Finally, we examined the protective effect of silymarin on cytokine-induced ?-cell damage in human islets. Treatment with IL-1?+IFN- and/or silymarin did not induce a significant increase in cell death in human islets, as assessed by MTT assay and PI staining. However, treatment of human islets with IL-1?+IFN- for 5 d resulted in the impairment of insulin secretion in response to high glucose (Fig. 8). Silymarin pretreatment prevented IL-1?+IFN--induced impairment of glucose-stimulated insulin secretion. Islets treated with silymarin alone released higher levels of insulin than the control islets treated with vehicle (DMSO) with a normal insulin secretory response to glucose.
FIG. 8. Silymarin protects human islets against cytokine-induced impairment of glucose-stimulated insulin secretion. Handpicked human islets were treated with 100 μM silymarin or DMSO for 2 h, then stimulated with IL-1? (75 U/ml) + IFN- (750 U/ml), and subsequently cultured for 5 d. After wash, islets were incubated with basal medium (containing 3.3 mM glucose) for 1 h, followed by a second incubation with high-glucose medium (containing 15.5 mM glucose) for 1 h. Insulin secreted into the medium was measured by ELISA. Data are means ± SE from three individual experiments using three different islet preparations. *, P < 0.05 vs. cytokine-stimulated islets without silymarin.
Discussion
In this study, we have demonstrated that silymarin prevents cytokine-induced cell death in RINm5F cells and cytokine-induced impairment of glucose-stimulated insulin secretion by human islets. In various types of cells, silymarin strongly blocks cytokine- or LPS-induced NF-B activation, which is crucial for iNOS expression in ?-cells (13, 30, 31). Unexpectedly, silymarin had no effect on IL-1?-induced NF-B activation in RINm5F cells. However, it partially inhibits IL-1?-induced JNK activation. JNK is known to be involved in the regulation of expression of many inflammatory genes, such as iNOS, IL-1?, and cyclooxygenase (COX)-2 (32, 33, 34). Although JNK’s roles in the induction of cell death by various stimuli are not firmly established, many evidences have recently accumulated to show that activation of JNK leads to cell death in pancreatic ?-cells and islets (19, 35, 36, 37). Furthermore, Kaneto et al. (38) showed that suppression of JNK activity by adenovirus-mediated overexpression of dominant-negative JNK prevented oxidative stress-induced inhibition of insulin gene expression and secretion in rat islets. Therefore, the observed cytoprotective effects of silymarin, including prevention of cytokine-induced impairment of glucose-stimulated insulin secretion, could be, at least in part, explained by its ability to suppress JNK activity.
Silymarin also suppressed IFN--induced phosphorylation and nuclear translocation of STAT1, STAT3, and STAT5 proteins. JAK/STAT pathways, especially JAK/STAT1, have been shown to be involved in various inflammatory actions (28). However, the roles of IFN- and IFN--elicited JAK/STAT pathways in the induction of ?-cell damage and its recovery process are still elusive. Treatment with IFN- reduces insulin mRNA levels and glucose-stimulated insulin secretion in mouse ?-cell line, CDM3D cells, without increasing NO production (21). IFN- induces cell death and a decrease of insulin gene expression in rat INS-1 cells, with no effect on NO production (22). Contrary to these previous reports, in our study, IFN- alone induced iNOS expression and NO production, which partially mediated cell death in RINm5F cells. This discrepancy may be due to not only species-specific but also cell type-specific regulation of IFN- signaling in ?-cells. However, these previously reported deleterious effects of IFN- on ?-cells, even in the absence of an increase in NO production, are thought to be mediated by STAT1 activation. Although IFN- alone does not stimulate iNOS expression in rodent and human islets, IFN- reduces the concentration of IL-1? required to induce iNOS expression in rat islets (20), and a combination of IL-1? and IFN- is required to induce iNOS expression and ?-cell dysfunction in mouse and human islets (5, 6). This priming or synergistic action of IFN- for IL-1?-induced iNOS expression is also associated with STAT1 activation. After homodimerization and translocation to the nucleus, phosphorylated STAT1 binds to the IFN--activated site (GAS), which is found in the promoters of various inflammatory genes, including iNOS and COX-2 (23, 39). It also induces the expression of IFN regulatory factor-1, which, through binding to IFN-stimulated response elements and interaction with NF-B sites in the iNOS promoter, would augment IL-1?-induced iNOS expression (39, 40). Therefore, the suppressive effect of silymarin on activation and nuclear translocation of STAT1 probably contributes to the inhibition of IFN-- and/or IL-1?+IFN--induced NO production, ?-cell death, and dysfunction in RINm5F cells and human islets.
On the other hand, the role of STAT3 or STAT5 in the regulation of iNOS expression and ?-cell dysfunction remains unclear. IL-6-induced iNOS expression is mediated by JAK2/STAT3 activation in isolated rat myocytes (41). Yoshida et al. (42) reported that NF-B p65 and non-tyrosine-phosphorylated STAT3 physically and functionally interact and that this interaction results in the inability of STAT3 to bind to and activate GAS site-dependent genes. This interaction also inhibits NF-B-dependent iNOS expression (43). Based on these reports, in addition to STAT1, inhibition of STAT3 activation by silymarin might also participate in suppression of IFN-- and/or IL-1?+IFN--induced iNOS expression and/or ?-cell dysfunction in RINm5F cells and human islets. Unlike STAT3, the evidence for the involvement of STAT5 in iNOS expression has not been reported so far. However, Yamaoka et al. (44) showed that STAT5 activation was involved in LPS-induced COX-2 expression in monocytes, suggesting that silymarin might suppress cytokine-induced COX-2 expression through the inhibition of STAT5 activation. On the contrary, STAT5b-mediated inhibition of NFB signaling to IFN regulatory factor-1 promoters was also reported (45). Therefore, further studies are necessary to elucidate whether and how STAT3 or STAT5 proteins are involved in cytokine-induced iNOS expression and dysfunction in ?-cells.
Although silymarin suppressed cytokine-induced NO production by only 50–60%, its cytoprotective effect was comparable with that of the iNOS inhibitor, L-NMMA. Silymarin’s antioxidant properties, based on its ability to scavenge free radicals and inhibit lipid peroxidation, have been well documented. Free radical production and lipid peroxidation are known to mediate cytotoxic action of cytokines on ?-cells (3). Thus, its cytoprotective effect in pancreatic ?-cells would also be attributed to its antioxidant activity in addition to the suppression of NO production. How silymarin suppresses the activation of JNK or STAT proteins is not clear. However, some antioxidants, such as N-acetyl-L-cysteine (NAC) and vitamin E, were shown to inhibit the activation of MAPKs and STAT proteins (46, 47). Furthermore, Matsubara et al. (48) showed that IL-1? and/or IFN- induced superoxide release in the culture medium within 15 min in endothelial cells, suggesting that superoxide produced intracellularly within minutes after cytokine stimulation might serve as signal transduction elements for JNK or STAT activation. Therefore, the suppressive effect of silymarin on the activation of JNK and STAT proteins might also be due to its antioxidant activity.
Another possible mechanism underlying the cytoprotective effect of silymarin could be to increase antioxidant enzymes such as SOD, glutathione peroxidase (Gpx), and catalase, as previously described by Soto et al. (11). However, this is not likely because a previous report has shown that overexpression of catalase, Gpx, or Cu/Zn SOD in RINm5F cells increased IL-1?-induced iNOS expression, though they still protected against cytokine-mediated toxicity through inactivation of reactive oxygen species generated in the signal cascades of cytokines (27). Another report has shown that overexpression of MnSOD, but not catalase, Gpx, or Cu/ZnSOD, in RINm5F cells reduced IL-1?-induced iNOS promoter activation and iNOS expression, which was mediated through the inhibition of NF-B activation (49). We confirmed no alteration in the levels of MnSOD expression in RINm5F cells treated with silymarin for up to 24 h, by Western blotting analysis (data not shown).
In this study, we showed the direct cytoprotective effect of silymarin on pancreatic ?-cells. Importantly, silymarin also has an ability to inhibit the production of inflammatory cytokines, such as IL-1?, IFN-, and TNF-, from macrophages and/or T-lymphocytes (30, 50, 51), which probably initiate the destruction of ?-cells in the development of type 1 diabetes. Taken together, these data indicate that silymarin may be useful as a therapeutic agent for type 1 diabetes.
Acknowledgments
The authors acknowledge the editorial services provided by Elizabeth Stein, Ph.D.
References
Bottazzo GF, Dean BM, McNally JM, MacKay EH, Swift PG, Gamble DR 1985 In situ characterization of autoimmune phenomena and expression of HLA molecules in the pancreas in diabetic insulitis. N Engl J Med 313:353–360
Lacy PE 1994 The intraislet macrophage and type I diabetes. Mt Sinai J Med 61:170–174
Rabinovitch A, Suarez WL, Thomas PD, Strynadka K, Simpson I 1992 Cytotoxic effects of cytokines on rat islets: evidence for involvement of free radicals and lipid peroxidation. Diabetologia 35:409–413
Rabinovitch A, Suarez-Pinzon WL, Strynadka K, Lakey JR, Rajotte RV 1996 Human pancreatic islet ?-cell destruction by cytokines involves oxygen free radicals and aldehyde production. J Clin Endocrinol Metab 81:3197–3202
Corbett JA, Sweetland MA, Wang JL, Lancaster Jr JR, McDaniel ML 1993 Nitric oxide mediates cytokine-induced inhibition of insulin secretion by human islets of Langerhans. Proc Natl Acad Sci USA 90:1731–1735
Thomas HE, Darwiche R, Corbett JA, Kay TW 2002 Interleukin-1 plus -interferon-induced pancreatic ?-cell dysfunction is mediated by ?-cell nitric oxide production. Diabetes 51:311–316
Beckman JS, Koppenol WH 1996 Nitric oxide, superoxide, and peroxynitrite: the good, the bad, and ugly. Am J Physiol 271:C1424–C1437
Suarez-Pinzon WL, Szabo C, Rabinovitch A 1997 Development of autoimmune diabetes in NOD mice is associated with the formation of peroxynitrite in pancreatic islet ?-cells. Diabetes 46:907–911
Katiyar SK, Korman NJ, Mukhtar H, Agarwal R 1997 Protective effects of silymarin against photocarcinogenesis in a mouse skin model. J Natl Cancer Inst 89:556–566
Zhao J, Lahiri-Chatterjee M, Sharma Y, Agarwal R 2000 Inhibitory effect of a flavonoid antioxidant silymarin on benzoyl peroxide-induced tumor promotion, oxidative stress and inflammatory responses in SENCAR mouse skin. Carcinogenesis 21:811–816
Soto C, Recoba R, Barron H, Alvarez C, Favari L 2003 Silymarin increases antioxidant enzymes in alloxan-induced diabetes in rat pancreas. Comp Biochem Physiol C Toxicol Pharmacol 136:205–212
Bosisio E, Benelli C, Pirola O 1992 Effect of the flavanolignans of Silybum marianum L. on lipid peroxidation in rat liver microsomes and freshly isolated hepatocytes. Pharmacol Res 25:147–154
Wang MJ, Lin WW, Chen HL, Chang YH, Ou HC, Kuo JS, Hong JS, Jeng KC 2002 Silymarin protects dopaminergic neurons against lipopolysaccharide-induced neurotoxicity by inhibiting microglia activation. Eur J Neurosci 16:2103–2112
Raso GM, Meli R, Di Carlo G, Pacilio M, Di Carlo R 2001 Inhibition of inducible nitric oxide synthase and cyclooxygenase-2 expression by flavonoids in macrophage J774A.1. Life Sci 68:921–931
Manna SK, Mukhopadhyay A, Van NT, Aggarwal BB 1999 Silymarin suppresses TNF-induced activation of NF-B, c-Jun N-terminal kinase, and apoptosis. J Immunol 163:6800–6809
Soto CP, Perez BL, Favari LP, Reyes JL 1998 Prevention of alloxan-induced diabetes mellitus in the rat by silymarin. Comp Biochem Physiol C Pharmacol Toxicol Endocrinol 119:125–129
Giannoukakis N, Rudert WA, Trucco M, Robbins PD 2000 Protection of human islets from the effects of interleukin-1? by adenoviral gene transfer of an I B repressor. J Biol Chem 275:36509–36513
Larsen CM, Wadt KA, Juhl LF, Andersen HU, Karlsen AE, Su MS, Seedorf K, Shapiro L, Dinarello CA, Mandrup-Poulsen T 1998 Interleukin-1?-induced rat pancreatic islet nitric oxide synthesis requires both the p38 and extracellular signal-regulated kinase 1/2 mitogen-activated protein kinases. J Biol Chem 273:15294–15300
Ammendrup A, Maillard A, Nielsen K, Aabenhus Andersen N, Serup P, Dragsbaek Madsen O, Mandrup-Poulsen T, Bonny C 2000 The c-Jun amino-terminal kinase pathway is preferentially activated by interleukin-1 and controls apoptosis in differentiating pancreatic ?-cells. Diabetes 49:1468–1476
Heitmeier MR, Scarim AL, Corbett JA 1999 Prolonged STAT1 activation is associated with interferon- priming for interleukin-1-induced inducible nitric-oxide synthase expression by islets of Langerhans. J Biol Chem 274:29266–29273
Cottet S, Dupraz P, Hamburger F, Dolci W, Jaquet M, Thorens B 2001 SOCS-1 protein prevents Janus kinase/STAT-dependent inhibition of ? cell insulin gene transcription and secretion in response to interferon-. J Biol Chem 276:25862–25870
Karlsen AE, Ronn SG, Lindberg K, Johannesen J, Galsgaard ED, Pociot F, Nielsen JH, Mandrup-Poulsen T, Nerup J, Billestrup N 2001 Suppressor of cytokine signaling 3 (SOCS-3) protects ?-cells against interleukin-1?- and interferon--mediated toxicity. Proc Natl Acad Sci USA 98:12191–12196
Gao J, Morrison DC, Parmely TJ, Russell SW, Murphy WJ 1997 An interferon--activated site (GAS) is necessary for full expression of the mouse iNOS gene in response to interferon- and lipopolysaccharide. J Biol Chem 272:1226–1230
Mullen Y, Arita S, Kenmochi T, Une S, Smith CV 1997 A two-step digestion process and LAP-I cold preservation solution for human islet isolation. Ann Transplant 2:40–45
Mosmann T 1983 Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays. J Immunol Methods 65:55–63
Matsuda T, Suzuki H, Oishi I, Kani S, Kuroda Y, Komori T, Sasaki A, Watanabe K, Minami Y 2003 The receptor tyrosine kinase Ror2 associates with the melanoma-associated antigen (MAGE) family protein Dlxin-1 and regulates its intracellular distribution. J Biol Chem 278:29057–29064
Lortz S, Tiedge M, Nachtwey T, Karlsen AE, Nerup J, Lenzen S 2000 Protection of insulin-producing RINm5F cells against cytokine-mediated toxicity through overexpression of antioxidant enzymes. Diabetes 49:1123–1130
Igaz P, Toth S, Falus A 2001 Biological and clinical significance of the JAK-STAT pathway; lessons from knockout mice. Inflamm Res 50:435–441
Flodstrom-Tullberg M, Yadav D, Hagerkvist R, Tsai D, Secrest P, Stotland A, Sarvetnick N 2003 Target cell expression of suppressor of cytokine signaling-1 prevents diabetes in the NOD mouse. Diabetes 52:2696–2700
Kang JS, Jeon YJ, Park SK, Yang KH, Kim HM 2004 Protection against lipopolysaccharide-induced sepsis and inhibition of interleukin-1? and prostaglandin E2 synthesis by silymarin. Biochem Pharmacol 67:175–181
Kang JS, Jeon YJ, Kim HM, Han SH, Yang KH 2002 Inhibition of inducible nitric-oxide synthase expression by silymarin in lipopolysaccharide-stimulated macrophages. J Pharmacol Exp Ther 302:138–144
Lahti A, Jalonen U, Kankaanranta H, Moilanen E 2003 c-Jun NH2-terminal kinase inhibitor anthra(1,9-cd)pyrazol-6(2H)-one reduces inducible nitric-oxide synthase expression by destabilizing mRNA in activated macrophages. Mol Pharmacol 64:308–315
Holzberg D, Knight CG, Dittrich-Breiholz O, Schneider H, Dorrie A, Hoffmann E, Resch K, Kracht M 2003 Disruption of the c-JUN-JNK complex by a cell-permeable peptide containing the c-JUN domain induces apoptosis and affects a distinct set of interleukin-1-induced inflammatory genes. J Biol Chem 278:40213–40223
Bennett BL, Sasaki DT, Murray BW, O’Leary EC, Sakata ST, Xu W, Leisten JC, Motiwala A, Pierce S, Satoh Y, Bhagwat SS, Manning AM, Anderson DW 2001 SP600125, an anthrapyrazolone inhibitor of Jun N-terminal kinase. Proc Natl Acad Sci USA 98:13681–13686
Bonny C, Oberson A, Negri S, Sauser C, Schorderet DF 2001 Cell-permeable peptide inhibitors of JNK: novel blockers of ?-cell death. Diabetes 50:77–82
Haefliger JA, Tawadros T, Meylan L, Gurun SL, Roehrich ME, Martin D, Thorens B, Waeber G 2003 The scaffold protein IB1/JIP-1 is a critical mediator of cytokine-induced apoptosis in pancreatic ? cells. J Cell Sci 116:1463–1469
Nikulina MA, Sandhu N, Shamim Z, Andersen NA, Oberson A, Dupraz P, Thorens B, Karlsen AE, Bonny C, Mandrup-Poulsen T 2003 The JNK binding domain of islet-brain 1 inhibits IL-1 induced JNK activity and apoptosis but not the transcription of key proapoptotic or protective genes in insulin-secreting cell lines. Cytokine 24:13–24
Kaneto H, Xu G, Fujii N, Kim S, Bonner-Weir S, Weir GC 2002 Involvement of c-Jun N-terminal kinase in oxidative stress-mediated suppression of insulin gene expression. J Biol Chem 277:30010–30018
Darville MI, Eizirik DL 1998 Regulation by cytokines of the inducible nitric oxide synthase promoter in insulin-producing cells. Diabetologia 41:1101–1108
Paludan SR, Malmgaard L, Ellermann-Eriksen S, Bosca L, Mogensen SC 2001 Interferon (IFN)-gamma and herpes simplex virus/tumor necrosis factor- synergistically induce nitric oxide synthase 2 in macrophages through cooperative action of nuclear factor- B and IFN regulatory factor-1. Eur Cytokine Netw 12:297–308
Yu X, Kennedy RH, Liu SJ 2003 JAK2/STAT3, not ERK1/2, mediates interleukin-6-induced activation of inducible nitric-oxide synthase and decrease in contractility of adult ventricular myocytes. J Biol Chem 278:16304–16309
Yoshida Y, Kumar A, Koyama Y, Peng H, Arman A, Boch JA, Auron PE 2004 Interleukin 1 activates STAT3/nuclear factor-B cross-talk via a unique TRAF6- and p65-dependent mechanism. J Biol Chem 279:1768–1776
Yu Z, Zhang W, Kone BC 2002 Signal transducers and activators of transcription 3 (STAT3) inhibits transcription of the inducible nitric oxide synthase gene by interacting with nuclear factor B. Biochem J 367:97–105
Ymamoka K, Otsuka T, Niiro H, Arinobu Y, Niho Y, Hamasaki N, Izuhara K 1998 Activation of STAT5 by lipopolysaccharide through granulocyte-macrophage colony-stimulating factor production in human monocytes. J Immunol 160:838–845
Luo G, Yu-Lee L 2000 Stat5b inhibits NFB-mediated signaling. Mol Endocrinol 14:114–123
Zafarullah M, Li WQ, Sylvester J, Ahmad M 2003 Molecular mechanisms of N-acetylcysteine actions. Cell Mol Life Sci 60:6–20
Maziere C, Alimardani G, Dantin F, Dubois F, Conte MA, Maziere JC 1999 Oxidized LDL activates STAT1 and STAT3 transcription factors: possible involvement of reactive oxygen species. FEBS Lett 448:49–52
Matsubara T, Ziff M 1986 Increased superoxide anion release from human endothelial cells in response to cytokines. J Immunol 137:3295–3298
Azevedo-Martins AK, Lortz S, Lenzen S, Curi R, Eizirik DL, Tiedge M 2003 Improvement of the mitochondrial antioxidant defense status prevents cytokine-induced nuclear factor-B activation in insulin-producing cells. Diabetes 52:93–101
Cho JY, Kim PS, Park J, Yoo ES, Baik KU, Kim YK, Park MH 2000 Inhibitor of tumor necrosis factor- production in lipopolysaccharide-stimulated RAW264.7 cells from Amorpha fruticosa. J Ethnopharmacol 70:127–133
Schumann J, Prockl J, Kiemer AK, Vollmar AM, Bang R, Tiegs G 2003 Silibinin protects mice from T cell-dependent liver injury. J Hepatol 39:333–340(Takeru Matsuda, Kevin Fer)