Hepatic stellate cells (HSC) play a pivotal role in liver fibrosis, and the clearance of activated HSC by apoptosis is associated with the resolution of liver fibrosis. The development of strategies that promote this process in a selective way is therefore important. We evaluated the effects of indole-3-carbinol (I3C), a nutritional component derived from vegetables from the Brassica family, on liver fibrosis and HSC apoptosis. The in vivo therapeutic effects of I3C were monitored in three rat models of liver fibrosis induced by porcine serum, bile duct ligation, or multiple hepatotoxic factors, and its proapoptotic effect and molecular mechanism were studied in vitro in HSC-T6, a rat HSC line. The results showed that I3C treatment significantly reduced the number of activated HSC in the livers of rats with liver fibrosis. In histopathology, I3C reduced hepatocyte degeneration and necrosis, accelerated collagen degradation, and promoted the reversal of liver fibrosis. I3C prescribed to HSC-T6 resulted in morphologic alterations typical of apoptosis and DNA cleavage to a nucleosomal ladder. Moreover, I3C significantly increased the HSC-T6 apoptosis rate and the expression ratio of Bax to Bcl-2. High-throughput protein array analysis indicated that the tumor necrosis factor-α/nuclear factor-κB (NF-κB) signal pathway participated in I3C-induced HSC-T6 apoptosis. Western blot and electrophoretic mobility-shift assay confirmed that I3C inhibited the phosphorylation of inhibitor of κB kinase α and inhibitor of κB-α and NF-κB DNA binding activity. In conclusion, I3C could promote the reverse process of liver fibrosis in vivo and induce apoptosis of activated HSC in vitro, which indicates the use of I3C as a potential therapeutic agent in liver fibrosis treatment.
Liver fibrosis is a common consequence to chronic hepatic damage mediated by a variety of etiological factors including xenobiotic damage (e.g., by drugs), viral infection (e.g., hepatitis B and C), and certain genetic diseases (e.g., hepatic hemochromatosis) (Bataller and Brenner, 2005). There is clear evidence that persistent fibrosis can lead to the development of liver cirrhosis or even hepatocellular carcinoma (Gutierrez-Reyes et al., 2007; Friedman, 2008). Therefore, interrupting and/or reversing liver fibrosis is important in preventing its progression with increasing severity. Liver fibrosis is characterized by an accumulation of extracellular matrix (ECM) components that impair normal hepatic function. Hepatic stellate cells (HSC) are a major source of ECM in the fibrotic liver and play a pivotal role in the development and resolution of liver fibrosis. HSC are normally quiescent and responsible mainly for the uptake, storage, and delivery of retinoids. Once liver is injured these cells become active and transdifferentiate into α-smooth muscle actin (α-SMA)-positive myofibroblast-like cells (Fowell and Iredale, 2006; Moreira, 2007).
For many years liver fibrosis and cirrhosis were considered irreversible, and liver transplantation was deemed the only therapeutic option (Bonis et al., 2001). However, there has been a paradigm shift in the field with increasing clinical and experimental evidence for reversibility. Studies from patients treated with pegylated interferon and ribavirin have shown consistent evidence of reversibility of fibrosis (Arthur, 2002). Iredale et al. (1998) showed that rats with liver fibrosis were able to recover to virtually normal histology after cessation of the injury (Issa et al., 2001). More importantly, they demonstrated that apoptosis of HSC was the initiating factor for this process by removing the cell type responsible for both the production of the ECM and the protection of matrix metalloproteinases through the production of tissue inhibitor of metalloproteinase (Elsharkawy et al., 2005; Kisseleva and Brenner, 2006). Lee et al. (2003) also showed that recovery from established experimental fibrosis can occur through the apoptosis of HSC and is associated with reductions in liver collagen. Apoptosis is the controlled mechanism by which cells are eliminated from tissues without eliciting an inflammatory response (Vaux and Korsmeyer, 1999). HSC apoptosis may therefore lead to remodeling of the hepatic ECM to a near-normal state. Therefore, strategic promotions of HSC apoptosis points out a potential direction for finding a way to treat live fibrosis.
Indole-3-carbinol (I3C) is a compound found abundantly in vegetables in the Brassica family, including broccoli, cauliflowers, Brussels sprouts, and cabbages. A broad spectrum of biological effects of I3C has been reported, such as anti-inflammation, antineoplasm, antimutagenesis, and antioxidation (Donald et al., 2004; Tsai et al., 2010). As a nutritional supplement, I3C has received attention over the years as a promising preventive and treatment agent for breast cancer and other types of cancers (Hsu et al., 2006; Acharya et al., 2010). In a previous study, we used micro-Raman spectroscopy to show the protective effect of I3C on acute alcoholic liver injury in vivo (Shen et al., 2008). We also found that I3C can remarkably protect liver slices from acetaldehyde-induced HSC activation, which shows its antioxidative ability and promotion of ECM degradation (Guo et al., 2010). Recent studies showed that, even at a concentration lower than that of inhibiting HSC proliferation (Ping et al., 2011), I3C could significantly induce HSC apoptosis. In this study, we show that in vivo administration of I3C reduces the extent of fibrotic collagen and promotes the resolution of liver fibrosis in three models of rats. Furthermore, we have demonstrated that I3C promotes the apoptosis of activated HSC in vitro, which was caused by its inhibitory action on the inhibitor of κB kinase α (IKKα)/inhibitor of κB-α (IκB-α)/nuclear factor-κB (NF-κB) signal pathway.
Materials and Methods
I3C, curcumin, penicillin, streptomycin, dimethyl sulfoxide (DMSO), and l-glutamine were purchased from Sigma (St. Louis, MO). Dulbecco's modified Eagle's medium (DMEM), fetal bovine serum (FBS), and TRIzol reagent were purchased from Invitrogen (Carlsbad, CA). Sterile porcine serum (PS) was obtained from Ben Biology Engineering Co. (Zhengzhou, China). Tumor necrosis factor-α (TNFα) was purchased from PeproTech (London, UK). Glyceraldehyde-3-phosphate dehydrogenase (GAPDH), mouse polyclonal antiphospho-IKKα (p-IKKα), p-IκB-α, rabbit polyclonal anti-IKKα, and IκB-α were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). Moloney murine leukemia virus reverse transcriptase and secondary antibody conjugated with horseradish peroxidase were purchased from Promega (Madison, W). 3,3′-Diindolylmethane (DIM) was purchased from LKT Labs (St. Paul, MN). Oligonucleotide primers were custom-synthesized by Sangon Biological Engineering Technology Co. (Shanghai, China). All other chemicals and reagents were of analytical grade.
In Vivo Study
Male Wistar rats weighing ∼220 to 250 g were provided by the Experimental Center of Medical Scientific Academy of Hubei (Hubei, China). All animals received humane care in compliance with the Chinese Animal Protection Act, which is in accordance with National Research Council criteria. Animals were housed for 1 week of acclimatization before experiments.
Porcine Serum Model of Liver Fibrosis.
Fifty six rats were randomly divided into seven groups: 1) normal control; 2) fibrotic model control; 3) I3C control (12 mg/kg); 4) fibrotic + I3C (3 mg/kg); 5) fibrotic + I3C (6 mg/kg); 6) fibrotic + I3C (12 mg/kg); and 7) fibrotic + curcumin (200 mg/kg). Curcumin was used as the positive drug in this study. Other than normal control and I3C control rats, all animals were intraperitoneally injected with 5 ml/kg of sterile porcine serum twice weekly for up to 6 weeks. Normal control and I3C control rats received the same volume of saline. After the liver fibrosis models were successfully established, the rats of three I3C-treated groups received intraperitoneal injection of I3C, dissolved in 4‰ DMSO as a vehicle, at doses of 3, 6, and 12 mg/kg daily for 17 days. The rats of the curcumin-treated group were intragastrically treated with 200 mg/kg curcumin, and control animals received 4‰ DMSO alone. At the end of the experiment, all of the rats were sacrificed, and livers were rapidly removed and rinsed in cold saline.
Bile Duct Ligation Model of Liver Fibrosis.
Rats were randomly divided into six groups: 1) sham-operated control; 2) fibrotic model control; 3) I3C control (12 mg/kg); 4) fibrotic + I3C (3 mg/kg); 5) fibrotic + I3C (6 mg/kg), and 6) fibrotic + I3C (12 mg/kg). Other than sham-operated control and I3C control rats, bile duct ligation (BDL) was performed on all animals under sterile conditions as described previously (Zhang et al., 2006). Animals of the sham-operated control and I3C control groups underwent the same laparotomy except bile duct ligation and section. Animals were studied 12 days after BDL or sham surgery. The rats of the I3C-treated groups received intraperitoneal injection of I3C, dissolved in 4‰ DMSO as a vehicle, at doses of 3, 6, and 12 mg/kg daily for 14 days, and control animals received 4‰ DMSO alone. Then rats were sacrificed and livers were rapidly removed.
Multiple Hepatotoxic Model of Liver Fibrosis.
Forty eight rats were randomly divided into six groups: 1) normal control; 2) fibrotic model control; 3) I3C control (12 mg/kg); 4) fibrotic + I3C (3 mg/kg); 5) fibrotic + I3C (6 mg/kg); and 6) fibrotic + I3C (12 mg/kg). Other than normal control and I3C control rats, all animals were fed a modified high-fat diet (89.5% corn flour, 10% lard, and 0.5% cholesterol) and 10% ethanol to drink for 4 weeks. On day 8 after a high-fat feed, each rat received a subcutaneous injection of an initial dose of 40% (v/v) carbon tetrachloride (CCl4) in olive oil at 5 ml/kg and then at 3 ml/kg twice a week for 3 weeks (Wang et al., 2006). Normal control and I3C control rats received the same volume of olive oil with a normal diet and free access to water. After the liver fibrosis models were successfully established, the rats of three I3C-treated groups received intraperitoneal injection of I3C, dissolved in 4‰ DMSO as a vehicle, at doses of 3, 6, and 12 mg/kg daily for 12 days. Control animals received 4‰ DMSO alone. Then rats were sacrificed and livers were excised.
For examination in the light microscope, liver specimens were fixed overnight in 0.4 mg/ml paraformaldehyde in phosphate-buffered saline (PBS), processed by the paraffin slice technique. Sections approximately 5 μm thick were stained with hematoxylin and eosin (HE) for routine histology and Masson's trichrome staining for collagen. The content of collagen (stained light green by Masson's trichrome) was quantified using a Medical Color Image Analysis System (HMIAS-2000; Wuhan Champion Image Technology Co., Ltd., Guangzhou, China). In brief, a field containing a portal vein approximately 100 μm in diameter in its center was selected at a magnification of 200×. Digitalized images of the field and five random fields of the same size were captured for computer analysis by a digital camera. Collagen was delineated in the images, and its area was measured. The average of the five fields was calculated for assessment of the degree of fibrosis in each case.
Immunohistochemical stainings for α-SMA were assessed in paraformaldehyde-fixed rat livers by the routine immunohistochemistry streptavidin-peroxidase method, using the antibody of α-SMA (diluted 1:200). An interstitial buffy stellated structure was marked as a positive expression of α-SMA. At least five random fields of each section were examined at a magnification of 400× and analyzed using the Medical Color Image Analysis System.
In Vitro Study
HSC-T6, an immortalized rat liver stellate cell line, was kindly provided by Prof. S. L. Friedman of the Mount Sinai School of Medicine (New York, NY). HSC-T6 cells exhibit characteristics compatible with those of activated HSC in culture and are a useful cell model for the study of HSC in vitro (Vogel et al., 2000). Cells were cultured in a 5% CO2 humidified incubator at 37°C. DMEM containing 10% FBS (v/v), 4 mM l-glutamine, and 100 U/ml penicillin-streptomycin were used as the growth medium. After HSC-T6 cells became subconfluent (at 70–80% confluence), they were cultured with DMEM without containing FBS for 24 h before the start of all experiments. Then cells were treated with 25, 50, and 100 μM I3C or DIM dissolved in DMSO vehicle (added to medium from a 2000-fold molar concentrated stock). Control cells received DMSO vehicle only (0.05%, v/v).
Flow Cytometric Analysis of Annexin V-FITC/Propidium Iodide-Stained Cells.
Apoptotic cells were quantified by measuring the externalized phosphatidylserine residues by using an annexin V-FITC/propidium iodide kit (BD Biosciences, San Jose, CA). After 24 h of I3C or DIM treatment, cells were collected, washed with ice-cold PBS, and suspended in a binding buffer. Then, cells were incubated for 15 min with FITC-conjugated annexin V and propidium iodide and analyzed with a FACSort flow cytometer (BD Biosciences) using CellQuest software (BD Biosciences). Annexin V-positive/propidium iodide-negative cells (lower right quadrant) were considered to be early apoptotic, whereas the lower left quadrant contained the vital (double negative) cell population.
Examination of Low-Molecular-Weight DNA Fragmentation.
HSC-T6 cells cultured in 35-mm diameter dishes were exposed to I3C for 24 h, and cellular DNA was prepared by using the genomic DNA purification kit (Promega) according to the manufacturer's protocol. DNA fragmentation was determined by gel electrophoresis of low-molecular-weight DNA.
RNA Extraction and Real-Time Reverse Transcription-PCR.
To determine the mRNA levels of Bax and Bcl-2, a quantitative real-time PCR assay was performed using a SYBR Green PCR Core Reagents kit (TaKaRa Biotechnology Co., Dalian, China). The cells were seeded into 90-mm culture dishes in DMEM containing I3C for 24 h. Total RNA was extracted from the cultured cells using the TRIzol reagent following the manufacturer's protocol. The concentration and purity of RNA were determined using a spectrophotometer (UV-1601; Shimadzu, Kyoto, Japan). Then single-strand cDNA was prepared from 2 μg of total RNA according to the protocol of the kit (TaKaRa Biotechnology Co.) and stored at −20°C until used. Primers were designed using Primer Premier 5.0 (Premier Biosoft International, Palo Alto, CA), and their sequences are as follows: Bax gene, sense 5′-TGTTTGCTGATGGCAACTTC-3′, antisense 5′-GATCAGCTCGGGCACTTTAG-3′ (annealing temperature = 60°C); Bcl-2 gene, sense 5′-TCTGTGGATGACTGAGTACCTGAA-3′, antisense, 5′-AGAGACAGCCAGGAGAAATCAAA-3′ (annealing temperature = 62°C); and GAPDH gene, sense 5′-TAAAGAACAGGCTCTTAGCAC-3′, antisense, 5′-AGTCTTGGAAATGGATTGTCTC-3′ (annealing temperature = 59°C). A relative standard curve was constructed for target genes (Bax and Bcl-2) and housekeeping gene (GAPDH) using their DNA isolated with a DNA extraction kit (TaKaRa Biotechnology Co.) with different concentrations ranging from 10 to 10,000 pg per reaction. PCR assays were performed in 36-well optical reaction plates using the RG-3000 Rotor-Gene 4 Channel Multiplexing System (Corbett Research, Mortlake, Australia) in a total volume of 25-μl reaction mixture containing 2 μl of 1 μg/μl cDNA template, 0.5 μl of 10 μM each primer, 12.5 μl of 2× Premix Ex Taq, 0.5 μl of 20× SYBR Green I, and 9 μl of diethyl pyrocarbonate-H2O. Each sample was normalized on the basis of GAPDH mRNA content. The specificity of PCR products was confirmed by agarose electrophoresis.
Protein Array Screening.
The signal transduction pathways modulated in I3C-treated HSC-T6 cells were screened by using the high-throughput protein arrays (Lab Vision, Fremont,, CA) according to the manufacturer's instructions. In brief, HSC-T6 cells were incubated with 100 μM I3C for 24 h. The whole-cell lysates were subjected to protein extraction, protein concentration determination, incubation, and fluorescent detection. Images were scanned with a Genepix 4000B (Molecular Devices, Sunnyvale, CA), and data were analyzed with Genepix Pro 6.0 software.
Western Blot Analysis.
HSC-T6 cells on 6-cm culture dishes were incubated with I3C for 24 h and then stimulated with or without TNFα (20 ng/ml) for 20 min. Then the cells were harvested and rinsed twice in PBS (4°C). Lysis buffer (50 mM HEPES, pH 7.5, 150 mM NaCl, 1.5 mM MgCl2, 5 mM EDTA, 10% glycerol, 1% Triton X-100, 0.5% nonidet P-40, 10 mM NaF, and 1 mM Na3VO4) and protease inhibitor cocktail (Beyotime Co., Beijing, China) were then added, followed by vortexing and sonication. HSC-T6 lysates were centrifuged at 14,000g for 10 min at 4°C. The supernatant was harvested, and protein concentrations were determined using a bicinchoninic acid protein assay kit (Beyotime Co.). Protein samples (80 μg/lane) were separated by 12% SDS-polyacrylamide gel electrophoresis and transferred onto nitrocellulose membranes (Millipore Corporation, Billerica, MA) using a Bio-Rad Laboratories (Hercules, CA) gel system. After blocking at room temperature for 2 h with 5% nonfat milk in 25 mM Tris-HCl, 50 mM NaCl, and 0.05% Tween 20, membranes were incubated with primary antibodies (anti-IKKα, anti-IκB-α, anti-p-IKKα, or anti-p-IκB-α; 1:200 dilution) for 1 h at room temperature and then overnight at 4°C. The membranes were then incubated with the secondary antibody conjugated with horseradish peroxidase (1:4000) at room temperature for 1 h. GAPDH was used as the internal control. Immunoreactive bands were visualized by using a diaminobenzidine kit (Santa Cruz Biotechnology Co.), and the intensities of bands were quantified with Odyssey (West Henrietta, NY) software.
Electrophoretic Mobility-Shift Assay for NF-κB DNA Binding Activity
Cells were incubated with different concentrations of I3C for 24 h and then stimulated with or without 20 ng/ml TNFα for 20 min. Then cells were washed with ice-cooled PBS and nuclear extracts were prepared. Protein concentrations were determined using a bicinchoninic acid protein assay kit. NF-κB DNA binding activity was analyzed using an EMSA kit (Viagene Biotech, Los Angeles, CA) according to the manufacturer's instruction. A double-strand oligonucleotide representing the consensus binding site for the NF-κB (5′-GGCTGGGGATTCCCCATCT-3′) served as the probe.
Before transfection, cells were plated on six-well plates and incubated in DMEM for 24 h. Cells were transiently transfected with a constitutively active form of p38 (kindly provided by Professor Christopher A. McCulloch, University of Toronto, Toronto, Canada) using Lipofectamine 2000 transfection reagent (Invitrogen) according to the manufacturer's protocols, and then treated by 100 μM I3C for 24 h. In all wells, the total amount of plasmid was adjusted with corresponding empty vector.
Results were expressed as the mean ± S.D. SPSS 11.5 (SPSS Inc., Chicago, IL) was used for data analysis. For studies involving more than two groups, data were evaluated with one-way analysis of variance. The level of statistical significance was set at P < 0.05 for all cases.
In Vivo Study: I3C Accelerates the Resolution of Liver Fibrosis in Rats
I3C Decreases Pathologic Characters and Reduces Collagen Deposition of Liver Tissues.
The in vivo therapeutic effects of I3C on liver fibrosis were monitored in three rat models. The extent of liver fibrosis was evaluated by qualitative and semiquantitative histological methods (HE staining and Masson's trichrome staining), and both methods showed the same pattern. The histological analysis of the livers from normal control and I3C control rats indicated a normal architecture. Treatment of the model control animals by sterile porcine serum, bile duct ligation, or multiple hepatotoxic (MHT) factors produced generally extensive liver fibrosis, as indicated by both qualitative and semiquantitative histopathological examination. Representative photographs of liver morphology are shown in Fig. 1, A to C. In contrast to normal rat liver morphology, porcine serum/bile duct ligation/multiple hepatotoxic factors caused severe histopathological changes such as steatosis, macrophage infiltration, and myofibroblast proliferation, and liver fibrosis was evident as disruption of tissue architecture, extension of fibers, large fibrous septa formation, pseudolobe separation, and accumulation of collagen. These alterations were reduced remarkably in the liver sections of rats that received I3C (12 mg/kg) or curcumin (200 mg/kg); the intensity of liver fibrosis was reduced by treatment with I3C or curcumin, resulting in only marginal fibrosis and weak portal inflammation. Semiquantitative analysis confirmed that the resolution of liver fibrosis was significantly accelerated by I3C (Fig. 1D).
I3C Decreases the Numbers of Activated HSC in Fibrotic Livers.
Because α-SMA is expressed by activated HSC, immunostaining for this protein was used to detect and quantify the numbers of activated HSC. There were hardly any α-SMA-positive cells in the normal control. In contrast, considerable expression of α-SMA was detected in Disse's space, around the periportal fibrotic band areas, central vein, and fibrous septa in the control group of three liver fibrosis models (Fig. 2, A–C). There was a statistically significant difference between these two groups (P < 0.01). Expression of α-SMA in the groups treated with I3C at doses of 3, 6, and 12 mg/kg showed a significant reduction of α-SMA-positive regions, compared with the model control group (P < 0.01), but still higher than the normal control (P < 0.05) (Fig. 2D).
In Vitro Study: I3C Stimulates the Apoptosis of HSC-T6
To further determine whether I3C could induce apoptosis of activated HSC during recovery from liver fibrosis, we measured cell apoptosis by morphology, DNA fragmentation electrophoresis, quantitative flow cytometric analysis, and Bax/Bcl-2 mRNA expression analysis.
I3C Alters HSC-T6 Morphology.
The addition of 100 μM I3C to HSC-T6 cells resulted in morphologic alteration of shrinking within 24 h as judged by light microscopy. As demonstrated in Fig. 3A, HSC-T6 cells changed from a flattened fibroblastic phenotype to star-like configuration with thin, slender, and dendritic processes by the treatment with 100 μM I3C.
I3C Increases Bax/Bcl-2 mRNA Expression in HSC-T6 Cells.
We then investigated whether Bax and Bcl-2, the important genes in some forms of apoptosis, were altered in HSC-T6 cells by I3C. As shown in Fig. 3B, the treatment of HSC-T6 cells with 50 to 100 μM I3C enhanced the expressional ratio of Bax to Bcl-2 mRNA levels in HSC-T6 cells.
I3C Induces Annexin V-FITC/Propidium Iodide-Positive Staining in HSC-T6 Cells.
HSC-T6 apoptosis also was confirmed and quantified by flow cytometric analysis of annexin V-FITC/propidium iodide-stained cells. As shown in Fig. 3C, incubation of HSC-T6 cells with 25, 50, and 100 μM I3C stimulated 35.7, 60.7, and 76.7% apoptosis, respectively, compared with DMSO solvent vehicle control.
I3C Induces DNA Cleavage to a Nucleosomal Ladder in HSC-T6 Cells.
The fragmentation of DNA into oligonucleosomal lengths, a further feature of apoptosis, was examined by electrophoresis. The treatment of HSC-T6 cells with I3C enhanced DNA fragmentation in a concentration-related manner as demonstrated in Fig. 3D. I3C at 50 and 100 μM significantly induced an increase of DNA cleavage to a nucleosomal ladder.
Mechanism of Action of I3C on HSC-T6
High-Throughput Protein Array Analysis.
To further determine the molecular mechanism of apoptosis induced by I3C, high-throughput protein array analysis was conducted to screen the main intracellular apoptosis-related signal transduction pathways in HSC-T6 cells. Many of the proteins modulated by 100 μM I3C are involved in the regulation of cell apoptosis, the cell cycle, or both. Accurate differential expression measurements were obtained by selecting genes that had been up- or down-regulated at least 2-fold in the array. Using these criteria, the expression of proteins involved in the TNFα-NF-κB signal pathway, NF-κB/p50, NF-κB/p65, and IKKα, were significantly down-regulated, compared with the control (Table 1). Other apoptosis-related proteins that were significantly regulated by I3C included E3-binding protein, transcription factor Oct-2, antiapoptotic factor bcl-xL, CD14, prostate apoptosis response protein-4, cyclin-dependent kinase 8, and prohibitin, a mitochondrial marker of cell apoptosis. In addition, many of the proteins that were slightly regulated by I3C represent genes that are involved in apoptotic signaling, such as p53, cytochrome c, apoptosis-inducing factor, caspase-8, caspase-9, TNFα and Wnt.
I3C Inhibits the TNFα-NF-κB Signal Transduction Pathway.
According to the results of protein array analysis, the effect of I3C on signaling components involved in the TNFα-NF-κB pathway were confirmed using Western blot and EMSA. As shown in Fig. 4B, at 20 min after TNFα stimulation the phosphorylated levels of IKKα and IκB-α were remarkably increased. I3C treatment can decrease the p-IKKα and p-IκB-α levels in a concentration-dependent manner. Total IKKα and IκB-α protein levels were equally expressed in all of the groups studied and not modified by I3C. EMSA for NF-κB DNA binding activity showed that I3C inhibited TNFα-induced NF-κB DNA binding activity in HSC-T6 cells in a concentration-dependent manner and almost completely at a concentration of 100 μM. (Fig. 4A).
To distinguish whether the effects of I3C on IKKα/IκB-α and NF-κB directly mediated the apoptotic response or are indirect consequences of the cells undergoing apoptosis, we reversed the effects of I3C by expressing a constitutive form of p38 that is upstream of NF-κB. The results showed that transfection of cells with active p38 vector attenuated the I3C down-regulation of NFκB activity (Fig. 4C) and attenuated the apoptotic response (Fig. 4D).
DIM Probably Accounts for Some of the Observed I3C Effects.
It is reported that I3C undergoes dimerization both in vivo and in vitro, in which DIM is the predominant active product (Bradlow and Zeligs, 2010). To determine whether the apoptotic effects of I3C could potentially be caused by the conversion of I3C into DIM we directly compared the cellular apoptotic effects of I3C and DIM. The results showed that treatment of HSC-T6 with 25 to 100 μM DIM also stimulated the cell apoptosis rates, but the proapoptotic effects of DIM is relatively weaker than that of I3C (Fig. 5).
There is now considerable interest in discovering compounds that selectively promote apoptosis of activated HSC, because proof-of-principle studies have shown that in vivo stimulation of HSC apoptosis will promote recovery from liver fibrosis (Issa et al., 2001; Lee et al., 2003; Friedman and Bansal, 2006). However, there is as yet no clearly established therapy for reversing liver fibrosis. As one of the bioactive components in cruciferous vegetables, I3C's anticarcinogenic effect and several of its mechanisms have been raised into the spotlight (Hsu et al., 2006; Shen et al., 2008; Acharya et al., 2010; Guo et al., 2010; Tsai et al., 2010). But little is known about the effect of I3C on liver fibrosis.
In this study, at first we proved the therapeutic effects of I3C on three different experimental models of liver fibrosis: immunological fibrosis induced by porcine serum, biliary fibrosis induced by bile duct ligation, and MHT fibrosis induced by carbon tetrachloride, ethanol, and high-fat/low-protein diet. These models have been widely used in the research field of liver fibrosis, which exhibits dissimilar pathogenesis. Immunological hepatic fibrosis models mimic chronic hepatic injury caused by hepatitis B virus (Zhang et al., 2010). BDL has been used as an animal model of chronic liver injury because of its resemblance to hepatocyte damage, HSC activation, and the liver fibrosis observed in human cholestatic liver disease (Lemos and Andrade, 2010). MHT is a well established model of liver fibrosis that best mimics the etiological factors in daily life (Wang et al., 2006). Observation of hepatic pathology demonstrated that we have successfully established these models of liver fibrosis in rats. The pathological characteristics of the liver tissues were markedly reduced by treatment with I3C, resulting in only marginal fibrosis and weak portal inflammation. Collagen as one type of ECM is an important component of fibrotic hepatic tissue. We observed collagen deposition through semiquantitation of the Masson's trichrome staining. The accumulation of collagen in the model control group increased remarkably compared with the normal control, but decreased, to some extent, in all I3C-treated groups. α-SMA, an intermediate filament protein that is expressed by activated HSC and is widely accepted as a marker of HSC activation, was used to identify and quantify activated HSC by immunostaining the fibrotic liver tissue. In this study, we found that the numbers of activated HSC present in the fibrotic livers were increased in the model control group, but significantly decreased by I3C treatment. All of these results indicate that administration of I3C to liver fibrosis rats promoted rapid clearance of α-SMA-positive myofibroblasts, reduced hepatic collagen deposition, and accelerated resolution of liver fibrosis. Abundant reports showed that curcumin exerts multiple biological effects on HSC and prevents the development of liver fibrosis through suppressing proliferation and inducing apoptosis in HSC (O'Connell and Rushworth, 2008). So we chose curcumin as the positive drug in our study. It is noteworthy that we gave I3C to animals after the liver fibrosis models were successfully established in the present study, which better reflect the actual clinical situation. The results showed that I3C exhibited considerable effect with curcumin in the treatment of hepatic fibrosis in rats, indicating I3C is probably capable of being a promising therapeutic agent of liver fibrosis.
Apoptosis has been known for mediating HSC loss during recovery from fibrosis, and control of apoptosis may be a key for regulating fibrosis (Lee et al., 2003; Elsharkawy et al., 2005; Kisseleva and Brenner, 2006). In our previous study, I3C has been shown to inhibit cell cycle progression at the G1 checkpoint in HSC-T6 cells, which resulted in the significant reduction of cell proliferation (Ping et al., 2011). Studies from other laboratories have demonstrated that I3C can induce apoptosis in several types of cancer cells, including myeloid, leukemia, and breast cancer cells (Takada et al., 2005; Ahmad et al., 2010). It is noteworthy that our pilot experiment showed that I3C can induce the HSC apoptosis at a concentration lower than that of inhibiting HSC proliferation. So we hypothesized that I3C promotes the reverse process of liver fibrosis by inducing the apoptosis of activated HSC. Similar results were observed when we evaluated the effect of I3C on HSC apoptosis by four methods: cell morphology, DNA fragmentation electrophoresis, quantitative flow cytometric analysis after annexin V binding and propidium iodide staining, and the ratio of Bax to Bcl-2 gene expression. These results supported our hypothesis that I3C could significantly induce the apoptosis of activated HSC.
As our next step we examined the molecular mechanism of HSC apoptosis induced by I3C. High-throughput protein array analysis was conducted in HSC. We focused on the observation of proteins involved in the HSC apoptosis-related pathways, death receptor-mediated pathway, and mitochondrial-dependent pathway. Among the proteins in these pathways, proteins involved in the TNFα-NF-κB signal pathway (e.g., NF-κB/p50, NF-κB/p65, and IKKα) were significantly differentially expressed by I3C treatment. TNF-α has been shown to activate the proapoptosis pathway as well as the antiapoptosis pathway. Experiments using a super repressor of NF-κB to HSC demonstrated that NF-κB inhibition sensitize the cells to TNF-α apoptosis (Lang et al., 2000; Elsharkawy et al., 2005). NF-κB is a transcription factor that has been implicated in the suppression of apoptosis, cell survival, and proliferation in addition to immune and inflammatory responses. Normally NF-κB retained in the cytoplasm by IκB is an inactive form (Lang et al., 2000). In response to extracellular stimuli, such as TNFα, IKK is activated and then phosphorylates IκB, leading to IκB proteosomal degradation and subsequent NF-κB release. Liberated NF-κB undergoes nuclear localization and induces the transcription of TNF-α regulated antiapoptosis genes (Wahl et al., 1998; Oakley et al., 2005). Potent TNF-α-NF-κB signal antagonists could thus be of intrinsic clinical use. It has been reported that nerve growth factor-induced apoptosis of HSC is associated with the inhibition of NF-κB DNA binding as well as suppression of NF-κB transcriptional activity (Trim et al., 2000). Likewise, curcumin induction of HSC apoptosis is also associated with the inhibition of NF-κB DNA binding, which mediated through the activation of peroxisome proliferator-activated receptor-γ (Zheng and Chen, 2004; O'Connell and Rushworth, 2008). Some studies also showed the I3C regulation of NF-κB activity in other cell lines (Aronchik et al., 2010). In our present study, the activated components in TNFα-induced antiapoptotic pathway, phosphorylated IKKα and IκB-α, and NF-κB DNA binding activity were significantly increased after TNF-α stimulation. Treatment of HSC by I3C can inhibit these components compared with untreated cells, indicating that the IKKα/IκB-α/NF-κB pathway participated in the proapoptotic effect of I3C on HSC. But whether the effects of I3C on IKKα/IκB-α and NF-κB directly mediated the apoptotic response or are indirect consequences of the cells undergoing apoptosis is unknown. TNFα also can activate the phosphorylation of p38 and then enhance the antiapoptotic effect of NF-κB (Saha et al., 2007). Our experiment showed that I3C can reduce the phosphorylation level of p38 protein (Ping et al., 2011), suggesting that the p38/NF-κB pathway participated in the proapoptotic effect of I3C on HSC. Thus, we reversed the effects of I3C by expressing the constitutive form of p38. The results showed that transfection of cells with active p38 vector attenuated the I3C down-regulation of NF-κB activity and attenuated the apoptotic response. These results indicated that the I3C effects on the NF-κB signal pathway directly mediated the apoptotic response. Furthermore, the inhibition of NF-κB pathway could strengthen the TNFα-induced proapoptotic signal pathway, which sensitizes HSC to apoptosis (Lang et al., 2000; Elsharkawy et al., 2005).
As a nutritional supplement, I3C is only for oral use because of its liposolubility. But our previous work indicated that I3C showed less powerful protective effect on liver injury via oral administration (Shen et al., 2008) than via intraperitoneal injection. Thus, we treated the liver fibrosis rats with I3C via intraperitoneal injection in the present study. Studies showed that I3C after ingestion is rapidly converted to a series of oligomeric products that are presumably responsible for the in vivo effects of I3C (Aggarwal and Ichikawa, 2005). For example, the predominant active product, DIM, has been reported to alter NF-κB transcriptional activity and induce an apoptotic response in breast cancer cells (Rahman et al., 2007). However, other research suggested that, the parent compound would have pharmacological activity, too (Anderton et al., 2004; Acharya et al., 2010; Tsai et al., 2010). Combining the abovementioned with our previous findings, we hypothesized that I3C itself has a more potent protective effect on liver injury than its metabolites. In this study, we also directly compared the cellular effects of I3C and DIM. The results showed that DIM can mimic the proapoptotic effects of I3C, but the effects of DIM are relatively weaker than that of I3C. DIM can spontaneously form from I3C during cell culture (Bradlow and Zeligs, 2010). Our results indicated that the proapoptotic effects of I3C in vitro could partly be caused by the conversion of I3C into its natural self-condensation product DIM. It was reported that I3C has been well tolerated by individuals with a daily dose ranges between 400 and 800 mg (Reed et al., 2005). Tissue concentrations over 1 mM (much higher than seen at therapeutic doses) have been safely achieved in animal studies (Stresser et al., 1995). Our previous studies and other laboratories' research data have shown the protective effects of I3C on hepatocytes in vivo (Shertzer et al., 1987; Shen et al., 2008) and in vitro (Guo et al., 2010). Our study also demonstrated that lactate dehydrogenase in the culture medium that released from HSC did not increase after treatment with 25 to 200 μM I3C (data not shown). All of these data indicate that I3C is a relatively safe chemical and probably promotes the apoptosis of activated HSC in a selective manner.
In conclusion, we offer direct evidence that, as a safe nutritional component derived from cruciferous vegetables, I3C could promote the reverse process of liver fibrosis in vivo and induce apoptosis of activated HSC in vitro. Furthermore, we demonstrated that I3C exerted its effects via blocking the IKKα/IκB-α/NF-κB signal transduction pathway and suggested that the natural dimerization of I3C in cells into DIM probably accounts for some of the observed I3C effects. Our current data, therefore, provide cogent animal, cellular, and molecular evidence that I3C is capable of being a promising therapeutic agent in liver fibrosis. Our findings further point to HSC apoptosis as a potential direction for providing strategic therapeutic systems on liver fibrosis.
Participated in research design: Ping and Wang.
Conducted experiments: Ping, Gao, Qin, Wei, Liu, X. Li, R. Li, and Ao.
Performed data analysis: Ping, Gao, Qin, and Wang.
Wrote or contributed to the writing of the manuscript: Ping, Bai, and Wang.
This work was supported by the National Nature Science Foundation of China [Grants 30800931, 30371666] and the Natural Science Foundation of Hubei Province, China [Grant 2008CDB117].
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.
- extracellular matrix
- hepatic stellate cells
- α-smooth muscle actin
- inhibitor of κB kinase α
- inhibitor of κB-α
- nuclear factor-κB
- dimethyl sulfoxide
- Dulbecco's modified Eagle's medium
- fetal bovine serum
- tumor necrosis factor-α
- porcine serum
- bile duct ligation
- multiple hepatotoxic
- phosphate-buffered saline
- hematoxylin and eosin
- glyceraldehyde-3-phosphate dehydrogenase
- electrophoretic mobility-shift assay
- fluorescein isothiocyanate
- polymerase chain reaction.
- Received January 24, 2011.
- Accepted August 19, 2011.
- Copyright © 2011 by The American Society for Pharmacology and Experimental Therapeutics