N-acetylcysteine induces cell cycle arrest in hepatic stellate cells through its reducing activity.

Activation of hepatic stellate cells (HSC) has been identified as a critical step in hepatic fibrogenesis and is regulated by several factors including cytokines and oxidative stress. However, the molecular mechanism for HSC inactivation is not well understood. We investigated an N-acetyl-L-cysteine (NAC)-mediated signaling pathway involved in HSC inactivation. NAC, which acting through its reducing activity, induced cell arrest at G1 via the mitogen-activated protein kinase (MAPK) kinase (MEK)/MAPK pathway in a Ras-independent manner. The sustained activation of this extracellular signal-regulated kinase induced the expression of p21(Cip1/WAF1), a cell cycle-dependent kinase inhibitor, and mediated cell growth arrest through the Sp1 transcription activator-dependent mechanism. These effects of NAC were all reversed by treatment of HSC with MEK inhibitor PD98059 followed by culturing HSC on type I collagen-coated flasks. The collagen-mediated suppression of NAC-induced arrest may be due to an overriding of the cell cycle arrest through an acceleration of integrin-induced cell growth. NAC action is actually dependent on modulating the redox states of cysteine residues of target proteins such as Raf-1, MEK, and ERK. In conclusion, an understanding of the NAC signaling pathway in HSC should provide the theoretical basis for clinical approaches using antioxidant therapies in liver fibrosis.

Hepatic stellate cells (HSC) 1 play a key role in the pathogenesis of hepatic fibrosis (1)(2)(3)(4). The activation of HSC is regulated by several factors including extracellular matrix, growth factors, cytokines, chemokines, oxidative stress, and other soluble factors (5,6). Recently, many reports have shown that the molecular mechanism for activation of HSC was involved in the intracellular signal cascade and transcriptional regulation of certain genes. For example, transforming growth factor-␤1 in-duces activation of Ras, Raf-1, MEK, and MAPK in HSC (7). The mitogenic effect of platelet-derived growth factor in HSC requires phosphatidylinositol 3-kinase (8), calcium influx (9), and activation of the ERK pathway followed by increased expression of c-fos (10). In addition to these cytokines, reactive oxygen intermediates (ROI), and the products of lipid peroxidation have been shown to modulate gene expression and cell proliferation through involvement of c-myb and NF-B and interaction of JNK in HSC (11,12). Furthermore, HSC activation by culture on plastic is associated with the activation of Sp1 and AP-1, which regulate the expression of collagen genes (13,14). On the other hand, some studies have focused on factors that may inhibit the proliferation of HSC and suppress activation of HSC. A nitric oxide-dependent increase in cellular cGMP levels has been shown to mediate the inhibitory effect of lipopolysaccharides and interferon-␥ on smooth muscle ␣-actin (SMA) expression in HSC (15). The proliferation of HSC is also inhibited by stimulation of cAMP-protein kinase and Ca 2ϩ / calmodulin kinase-II through the induction of CREB phosphorylation on Ser 133 (16).
Oxidative stress has been implicated in hepatic fibrosis (6,11,17). Some suggest that oxidants or their by-products act directly upon HSC to stimulate collagen synthesis (18,19). In contrast, an antioxidant such as ␣-tocopherol blocks the activation of HSC induced by collagen type I matrix or transforming growth factor-␣ (11) and prevents iron-induced hepatic cirrhosis (20). In addition, effects of natural phenolic compounds such as resveratrol and quercetin on the functions of HSC were examined in a recently published study (21). These reports suggest that oxidative stress may enhance the activation of HSC, whereas antioxidants may suppress this process. However, the exact mechanism whereby antioxidants suppress the activation of HSC is still obscure.
N-Acetyl-L-cysteine (NAC), an aminothiol and synthetic precursor of intracellular cysteine and GSH, has been used therapeutically in several disorders related to oxidative stress (22). The chemical properties of the cysteinyl thiol of NAC include its nucleophilicity and redox interactions, which provide antioxidant properties and thiol-disulfide exchange reactions. Recently, NAC has been found to have survival-promoting actions in PC12 cells (23) and other cell systems (24). The NAC-promoted survival of PC12 cells has been reported to be independent of its antioxidant/radical scavenger property (25). However, it has generally been assumed that the action of NAC is due to its antioxidant/radical scavenger properties or its thiol-disulfide exchange activity as a reductant.
The present study was designed to investigate the molecular mechanisms of NAC action on the functions of cultured HSC. This report states that NAC acts as a reductant in cells due to its direct reducing activity, induces the sustained activation of ERK, up-regulates p21 Cip1/WAF1 expression, and mediates cell cycle arrest through an Sp1-dependent mechanism.
Cell Culture and Treatment-HSC were prepared from male Sprague-Dawley rats (400 -500 g), which had free access to a standard laboratory chow diet. Nonparenchymal liver cells were isolated by the Pronase-collagenase method (26), and HSC were purified from the nonparenchymal cell suspension by a single step density gradient centrifugation with Nycodenz, as reported in detail by Schafer et al. (27). HSC were identified by their typical light microscopy appearance and immunofluorescence staining for desmin (28). The mean purity of freshly isolated cells as analyzed with a fluorescence-activated cell sorter (FACSCalibur; Becton Dickinson, San Jose, CA) was 85 Ϯ 5%, cell viability as checked with Trypan blue exclusion was 90 Ϯ 5%, and the yield ranged from 12 to 20 ϫ 10 6 cells/liver. Cells were seeded at a density of 1 ϫ 10 5 cells/cm 2 and maintained with DMEM containing 4 mM L-glutamine, penicillin (100 IU/ml), streptomycin (100 g/ml), and 5% fetal bovine serum, and cultured in a humidified atmosphere of 5% CO 2 , 95% air. The first change of the medium was made 16 h after seeding, after which the purity of the HSC was greater than 95%. The medium was changed every 24 h. Experimental manipulations were performed with cells at passages 5-12. Cultures were initiated at a cell density of 1 ϫ 10 5 cells/ml, and NAC was added to the culture medium 24 h after starting the culture, unless otherwise mentioned.
GSH Assay and Reactive Oxygen Intermediate Measurement-The concentration of intracellular GSH was measured using a colorimetric assay kit (Bioxytech TM GSH-400; OXIS International, Portland, OR) according to the manufacturer's instructions. Briefly, 1 ϫ 10 6 cells were homogenized in 500 l of ice-cold 5% metaphosphoric acid and centrifuged at 3000 ϫ g for 10 min. Supernatants were used to assay GSH using 4-chloro-1-methyl-7-trifluromethylquinolinium methylsulfate and 30% NaOH at 400 nm. Intracellular ROI production was measured by the method of Bae et al. (29). Briefly, dishes treated with or without testing agents were washed with DMEM without phenol red and incubated in the dark for 5 min in the presence of 5 mM dichlorodihydrofluorescein diacetate (Molecular Probes, Inc., Eugene, OR). When dichlorodihydrofluorescein diacetate is oxidized within the cell, it becomes a highly fluorescent 2Ј,7Ј-dichlorofluorescein and can be detected (excitation, 485 nm; emission, 530 nm).
Assays of HSC Activation and Cell Cycle-Cell density in experimental cultures was measured by repeated counts with a hemacytometer. Viability was assessed by the exclusion of 0.2% trypan blue dye. A measurement of HSC activation was determined with the expression of SMA. At the end of cell culture, each plate was washed with phosphatebuffered saline (PBS) and fixed with methanol. Specimens were incubated in PBS containing 1% albumin for 2 h, washed, and incubated with anti-SMA antibody in PBS containing 1% albumin at 4°C overnight. Cells were washed and incubated with horseradish peroxidaseconjugated second antibody for 2 h. After being washed, samples were incubated with peroxidase substrate (o-phenylenediamine dihydrochloride) and stopped with H 2 SO 4 , and the absorbance was read at 490 nm. For cell cycle analysis, HSC were synchronized by serum starvation in a medium containing 0.1% serum for 24 h and induced to reenter the cell cycle by an exchange of 5% fetal bovine serum with/without 5 mM NAC. The distribution of cells in the cell cycle was determined by flow cytometry using propidium iodide-stained nuclei. A total of 1 ϫ 10 6 cells were harvested at each indicated time and resuspended in 0.5 ml of propidium iodide solution (50 g/ml propidium iodide, 1 mg/ml sodium citrate, 100 g/ml RNase I, and 0.1% Triton X-100). Flow cytometric analysis was done with a fluorescence-activated cell sorter. Forward light scatter characteristics were used to exclude cell debris from the analysis. The G 0 /G 1 , S, and G 2 /M phase of the cell cycle were analyzed by diploid staining profiles and ModFit software programs (version 1.00 for Mac; ModFit Verity Software House, Topsham, ME).
Kinase Assay-Following the normalization of protein content, endogenous ERK or JNK was immunoprecipitated from cell lysates using monoclonal antibodies against ERK, JNK, or p38, as described previously (30,31). Kinase activity was assayed for 20 min at 30°C in the presence of 5 g of substrate, 50 M ATP, and 3 Ci of [␥-32 P]ATP in 40 l of kinase buffer (20 mM MOPS, pH 7.2, 2 mM EGTA, 20 mM MgCl 2 , 1 mM DTT, 1 mM Na 3 VO 4 , and 25 mM ␤-glycerol phosphate). Myelin basic protein, glutathione S-transferase-c-Jun, and ATF-2 were used for assaying ERK, JNK, and p38 activity, respectively. After completion of kinase assays, the proteins were resolved by SDS-PAGE and subjected to autoradiography. For direct ERK kinase assay of Sp1 immunoprecipitates, Sp1 and ERK were immunoprecipitated from nuclear extracts using anti-Sp1 and ERK antibody, respectively, and then incubated with Protein A-Sepharose. After the beads were washed three times with washing buffer (20 mM MOPS, pH 7.2, 1 mM Na 2 VO 3 , 10 mM NaF, 1 mM benzamidine), Sp1 precipitates were mixed with ERK immunoprecipitates, and then a kinase reaction was carried out in the kinase buffer containing 5 Ci of [␥-32 P]ATP of for 30 min at 30°C. Phosphorylated proteins were analyzed by SDS-PAGE and autoradiography.
Preparation of Ligand-coated Flasks and Collagen Binding Assay-Tissue culture flasks were incubated with 20 g/ml type I collagen (rat tail), type IV collagen (human placenta), fibronectin (human plasma), and laminin (mouse sarcoma) at 4°C overnight. After blocking with 1% bovine serum albumin in PBS for 1 h at 37°C, cells were incubated for 24 h and then treated with NAC for 24 h at 37°C. Collagen-Sepharose conjugate was prepared from rat tail type I collagen and CNBr-activated Sepharose 4B, according to the manufacturer's instructions (Amersham Pharmacia Biotech). Cells were treated with NAC for 6 h before solubilization by lysis buffer, and cell lysates were incubated with 50 l of collagen-Sepharose conjugate and 20 l of 1 M MgCl 2 for 12 h at 4°C The beads were washed twice with 1 ml of lysis buffer and three times with 1 ml of PBS. The samples were subjected to SDS-PAGE, and the proteins were transferred to nitrocellulose membranes. The membranes were probed with anti-integrin ␣ 1 , ␣ 2 , and ␤ 1 antibodies and developed with ECL reagents.
Identification of Protein Containing Free Sulfhydryl Groups-HSC grown in 100-mm plates were incubated with or without 5 mM NAC for 6 h. Cells were solubilized in 1 ml of lysis buffer, and cell lysates were incubated with 50 l of EZ-Link TM PEO-iodoacetyl Biotin (10 mg/ml; Pierce) for 12 h at 4°C and then incubated with 30 l of UltraLink TM Immobilized NeutrAvidin TM (Pierce) for 4 h on a roller system at 4°C. The beads were washed twice with 1 ml of lysis buffer and three times with 1 ml of PBS. For Western blot analysis, proteins containing sulfhydryl groups were subjected to SDS-PAGE, and the proteins were transferred to nitrocellulose membranes as described above. Membranes were probed with anti-integrin ␣ 1 /␣ 2 /␤ 1 , Ras, Raf-1, MEK, and ERK antibodies and developed with ECL reagents.

RESULTS
Antioxidants Containing a Sulfhydryl Group Induce the Suppression of HSC Activation-First, we examined the role of antioxidants on the activation of HSC cultured in DMEM with 5% fetal bovine serum. HSC activation was assessed by SMA expression. As shown in Fig. 1A, NAC, GSH, DTT, ␤-mercaptoethanol, and diethyldithiocarbamic acid inhibited the activation of HSC in a dose-dependent manner, whereas other antioxidants such as Me 2 SO, mannitol, ␣-tocopherol, and trans-retinol did not affect the activation of HSC. Interestingly, most antioxidants suppressing the activation of HSC contain a sulfhydryl group. To assess whether the suppression of SMA expression in HSC by sulfhydryl antioxidants is due to growth inhibition or cell death, we performed a time course study of cell growth after treatment with each antioxidant. As shown in Fig. 1B, the cell numbers of HSC cultured in NAC-and GSH-treated media are maintained at the level of an inoculated cell, whereas the cell numbers of HSC cultured in ␤-mercaptoethanol and diethyldithiocarbamic acidtreated media decrease below starting cell numbers. These data suggest that NAC and GSH induced cell arrest, whereas ␤-mercaptoethanol and diethyldithiocarbamic acid induced cell death. HSC grown in normal medium continued to proliferate until serum-starved, whereas cells grown in the presence of NAC underwent a dramatic and complete growth arrest up to 3 days. To determine whether the action of NAC in growth arrest is related to its antioxidant/radical scavenger properties via intracellular GSH or its reducing activity, we first treated HSC with buthionine-(S,R)-sulfoximine (BSO) in the presence of NAC, followed by GSH assay. BSO inhibits ␥-glutamylcysteine synthase and causes the depletion of intracellular GSH (33). Fig. 2A shows that the NAC-mediated increase in intracellular levels of GSH was effectively inhibited by 1 mM BSO after 8 h of continuous exposure ( Fig. 2A, inset). However, the proliferation analysis data in Fig. 2A show that, despite its inhibition of GSH accumulation, the BSO did not affect the capacity of NAC to arrest HSC. This result suggests that the action of NAC is independent of its effects on intracellular GSH. Furthermore, we show that the change in the intracellular level of ROI due to H 2 O 2 did not affect the capacity of NAC to arrest cells. As shown in Fig. 2B, the treatment of the cells with NAC caused a decrease in fluores-cence intensity as compared with the control treatment. When NAC-treated cells are added to H 2 O 2 (250 M), the fluorescence intensity of intracellular ROI in the cells is restored to the level of control, but the SMA expression of the cells is unchangeable. This suggests that the action of NAC is not dependent on its antioxidant/radical scavenger properties. Considering the fact that only antioxidants containing sulfhydryl group affect cell growth, NAC's reducing activity may be responsible for its action on cell arrest.
NAC Induces Cell Cycle Arrest at the G 1 Stage in HSC-To further examine growth arrest, fluorescence-activated cell sorting analysis of DNA content was performed on proliferating and NAC-arrested cells. The cell cycle distribution of HSC was determined at various times after NAC treatment (0 -36 h). In proliferating populations (control), the percentage of DNA content in S phase was 6.5% after 24 h. Following treatment with NAC for 24 h, this decreased to 0.4% (Table I). The data suggest that NAC prevents HSC entry into the S phase by arrest at the G 1 /S interphase.
NAC Induces a Sustained Increase in ERK Activity but Not JNK and p38 Activity-To investigate the role of the signal pathway in NAC-induced growth arrest, we examined the effects of specific inhibitors of phosphatidylinositol 3-kinase (wortmannin), protein kinase C (calphostin C), tyrosine kinase (herbimycin A), protein tyrosine phosphatase (Na 2 VO 3 ), guanylate cyclase (LY83583), MEK (PD98059), and p38 (SB202190) on this process. Among various inhibitors, the ad- dition of PD98059 to HSC cultured in the presence of NAC brought about a prominent reversal of NAC-induced suppression of SMA expression (Fig. 3). The data suggest that the MEK pathway plays a significant role in NAC-induced cell arrest.
Marshall previously proposed that, when cells make a decision regarding proliferation versus differentiation, they do it by differences in the duration of ERK activation (34). Therefore, we tested the kinase activity of ERK in NAC-treated HSC. Time course analyses showed that NAC affects ERK activity during the first 0.5 h, followed by a marked increase in ERK activity as measured at 3 h, and then a gradual decrease in the activity over a period of 24 h (Fig. 4A). Throughout, the level of ERK activation is maintained above control levels. In addition, activation of ERK1 (p44) and ERK2 (p42) requires the phosphorylation of specific tyrosine and threonine residues by MEK (35). Therefore, we assessed the phosphorylation of ERK1 and ERK2 in NAC-treated HSC by Western blot analysis of whole cell lysates with an anti-phosphorylated ERK1/2 antibody. In whole cell lysates, an intense phosphorylation of ERK1/2 appeared 0.5 h after an NAC treatment and then increased up to 24 h (Fig. 4B). The data suggest that the induction of phosphorylation of ERK1/2 by NAC correlated with the increase of ERK activity in NAC-treated HSC. On the other hand, the kinase activities of JNK and p38 were undetectable in NAC-treated HSC (data not shown).
ERK Activation Is Associated with Sp1 Phosphorylation in NAC-induced Cell Arrest-To elucidate whether a certain transcription factor is associated with NAC-induced HSC arrest, we analyzed the expression and activation of several transcription factors, as detected by immunoblotting and gel shift analysis. The nuclear expressions of c-Fos, CREB, Elk-1, NF-B, and Sp1 were invariable during NAC-induced growth arrest (data not shown). To show the DNA binding capacity of these transcription factors, we performed time course analyses in NACtreated HSC. The binding of HSC nuclear extracts to a Sp1 cognate oligonucleotide was low in control cells but increased significantly following an NAC treatment of HSC with optimal activation at around 6 h (Fig. 5A). However, regarding other transcription factors, the activation of these factors does not appear to be sufficient for NAC-induced cell arrest (data not shown). To determine whether the regulation of Sp1-dependent transcription by NAC is associated with its phosphorylation, Sp1 was immunoprecipitated from nuclear extracts and screened with anti-phosphoserine antibody. As seen in Fig. 5B, an induction of Sp1 phosphorylation was apparent by 1 h and was maximal during 3-12 h following NAC treatment. To determine whether ERK is associated with Sp1 phosphorylation in NAC-induced cell arrest, Sp1 was immunoprecipitated from the nuclear extracts of untreated cells (control), and ERK was immunoprecipitated from the nuclear extracts of control and NAC-treated cells. Sp1 and ERK immunoprecipitates were mixed, and the kinase activities of ERK in Sp1 phosphorylation were assessed. As shown in Fig. 5C, Sp1 is intensively phosphorylated by the ERK of NAC-treated cells. The data indicate that ERK binds Sp1 and that the activation of ERK is associated with Sp1 phosphorylation in NAC-induced HSC arrest.
NAC-induced p21 Cip1/WAF1 Induction and Morphologic Change Are Reversed by an Addition of MEK Inhibitor-Cyclindependent kinase inhibitors have been known to decrease progression through the cell cycle, and their expression can lead to cell cycle arrest (36). Especially, in PC 12 cells, p21 Cip1/WAF1 induction is associated with nerve growth factor-mediated growth arrest during differentiation (37,38). Therefore, we examined the level of p21 Cip1/WAF1 protein during NAC-induced cell arrest. As shown in Fig. 6A, NAC caused a sustained increase in levels of p21 Cip1/WAF1 protein after 6 h, and 2-3-fold increased levels of p21 Cip1/WAF1 were seen as early as 6 h following NAC addition. These results suggested that NAC induced cell arrest by the induction of p21 Cip1/WAF1 protein. Furthermore, we tested whether the MEK/MAPK pathway might regulate this effect by using MEK inhibitor. As shown in Fig. 6B, treatment of HSC with PD98059 caused a dose-dependent reversal of NAC-induced p21 Cip1/WAF1 protein and an approximate 87% reduction in the levels of p21 Cip1/WAF1 protein with 6 M PD98059. This result was confirmed by an observation of cell morphology, in which normal features of HSC are recovered from NAC-induced morphologic change by an addition of 6 M PD98059 (Fig. 6C).
HSC Plated on the Type I Collagen Resist Cell Arrest by NAC-We next investigated whether cell-matrix interactions were involved in regulating NAC-induced cell arrest. After cells were plated onto plastic coated with various ECM proteins including fibronectin, laminin, and collagens, we observed the growth of HSC treated with NAC for 24 h. Only a cell plated onto type I collagen-coated plastics overrides NAC-induced cell arrest, whereas other cells plated onto various ECM protein-

FIG. 2. HSC-arresting action of NAC is independent of intracellular GSH and ROS levels.
A, BSO inhibits the NAC-induced increase in GSH levels but has no effect on NAC-promoted cell arrest. HSC were incubated with BSO for 16 h prior to NAC treatment. Cells were treated with no additive (q), 1 mM BSO (E), 5 mM NAC (), and 1 mM BSO ϩ 5 mM NAC (ƒ). At the indicated time after NAC treatments, cell lysates were prepared, and the levels of GSH were estimated (inset). Cell numbers represent the means Ϯ S.E. of three separate experiments. B, despite a recovery of ROS production in NAC-treated cells by H 2 O 2 , NAC suppressed the SMA expression of HSC. Cells were cultured for 24 h with no additive (control), 5 mM NAC, or 5 mM NAC plus 0.25 mM H 2 O 2 . The cells were washed with DMEM without phenol red and incubated in the dark for 5 min in the presence of dichlorodihydrofluorescein diacetate. ROS productions were determined with FL500 microplate fluorescence reader (Bio-Tek). The SMA expression of HSC was measured by the enzyme-linked immunosorbent assay method. coated plastics are subject to NAC action (Fig. 7A). Since integrin ␣ 1 ␤ 1 and ␣ 2 ␤ 1 are the two major collagen receptors, a reversal of an arrested cell by collagen is due to inducing of cell proliferation by an interaction between type I collagen and integrin ␣ 1 ␤ 1 (and/or ␣ 2 ␤ 1 ). To assess whether NAC action can be ascribed to the repression of the intracellular signaling pathway through the disruption of integrins, we investigated amounts of integrin ␣ 1 ␤ 1 and ␣ 2 ␤ 1 , which bind to type I collagen in NAC-treated cells and control cells. As shown in Fig. 7B, the amounts of ␣ 1 and ␣ 2 are indistinguishable from those of the NAC-treated cells and control cells. This finding implies that NAC does not disrupt the integrity of integrin ␣ 1 and ␣ 2 . Therefore, the additional interaction between integrin ␣ 1 ␤ 1 (or ␣ 2 ␤ 1 ) and type I collagen overrides the ability of NAC to arrest HSC via integrin-mediated cell cycle progression.
NAC Modulates the Redox State of Cysteine Residues on the Target Proteins-In previous data, we suggested that NAC action in our system is due to its reducing activity (Fig. 1A). To elucidate whether NAC performs thiodisulfide exchange reactions in NAC-induced cell arrest, we tested free sulfhydryl groups in the downstream components of NAC-mediated signaling via sulfhydryl-reactive biotinylation reagent. Screening of sulfhydryl groups against anti-integrin ␣ 1 /␣ 2 /␤ 1 , Ras, Raf-1, MEK, and ERK revealed that NAC turns the redox states of the cysteine residues of Raf-1, MEK, and ERK into reduced states, whereas the redox states of cysteine residues in Ras are not affected by NAC (Fig. 8A). In the meantime, we measured the activity of Raf-1, MEK, and ERK following immunoprecipitation and subsequent treatment with NAC or buffer control to determine whether the reduction by NAC directly affects the activity of any of these kinases. As shown in Fig. 8B, the activities of Raf-1, MEK, and ERK were increased in a dose-dependent manner. These results suggest that cysteine residues of Raf-1, MEK, and ERK play an important role in NACinduced signaling in HSC. On the other hand, free sulfhydryl TABLE I Induction of growth arrest in HSC by NAC Cells were synchronized by serum starvation in a medium containing 0.1% serum for 24 h and induced to reenter the cell cycle by an exchange of 5% FBS with/without 5 mM NAC. At the indicated time, the distribution of cells in the cell cycle was determined by flow cytometry using propidium iodide-stained nuclei.  groups in integrin ␣ 1 /␣ 2 /␤ 1 were not detected both in control and NAC-treated cells (data not shown). DISCUSSION Oxidative stress has been implicated in the pathogenesis of a variety of human diseases such as cancer, cardiovascular diseases, chronic inflammation diseases, central nervous system disorders, aging, and liver fibrosis (39). Since oxidative stress results from an imbalance between oxidants and antioxidants, a number of antioxidants are necessary to counteract harmful effects (40). In this context, recent studies demonstrated the role of antioxidants in preventing liver fibrosis (41,42). Especially, the administration of NAC has been reported to reduce mortality in patients suffering from fulminant hepatic failure and biliary obstruction (43,44). However, the cellular and Upper arrows indicate the migration of the increased DNA binding complexes, and the lowest arrow indicates the migration of free probe. B, the time course for NAC induction of Sp1 phosphorylation. HSC were cultured for 24 h prior to the addition of NAC. At the indicated times (in hours) after the addition of NAC, nuclei were extracted, and Sp1 was immunoprecipitated and subjected to SDS-PAGE. The phosphorylation of Sp1 was detected by anti-phosphoserine antibody. C, Sp1 is phosphorylated by activated ERK. The direct ERK kinase assay of Sp1 immunoprecipitates was described under "Experimental Procedures." The ERK of NAC-treated cells (NAC) phosphorylated Sp1 more intensively than did the ERK of untreated cells (Control; upper panel). In Western blotting using anti-Sp1 antibody, the amounts of Sp1 in control and NAC-treated cells showed a constant level as a loading control (lower panel). molecular mechanism of antioxidants in preventing liver fibrosis remains to be clarified.
The aim of this study was to elucidate the molecular mechanism underlying NAC action on HSC, which is known to play an important role in the development of liver fibrosis. On the basis of a time course analysis of NAC-mediated signaling, we demonstrated that NAC induces the growth arrest of HSC at the G 1 stage, in which the sustained activation of ERK leads to Sp1 activation and is followed by p21 Cip1/WAF1 induction. The NAC action actually depends on its ability to modulate the redox state of cysteine residues on target proteins such as Raf-1, MEK, and ERK.
NAC has been known to act as an antioxidant/free radical scavenger or reducing agent (22). In this study, we showed that BSO treatment blocks an NAC-induced increase in intracellular GSH but has no effect on the capacity of NAC to induce cell arrest. Thus, GSH levels enhanced by NAC cannot account for the actions of NAC. We have also considered the direct action of NAC as an antioxidant or free radical scavenger. Several other antioxidants/free radical scavengers such as Me 2 SO and mannitol were unable to mimic NAC cell-arresting action in our experimental conditions, despite the reduction of intracellular ROI (data not shown). Furthermore, an increase of ROI due to H 2 O 2 did not affect the capacity of NAC to arrest cell growth (Fig. 2B). An additional mechanistic possibility we considered is that the effect of NAC might be due to its reducing activity. To assess this action, we tested several other reducing agents in our systems. Like NAC, GSH was also found to induce the growth arrest of HSC. Diethyldithiocarbamic acid and ␤-mercaptoethanol, in contrast, cause cell death rather than cell arrest. Therefore, NAC may provide thiol-disulfide exchange reactions with other thiol redox couples and provoke specialized cellular functions.
The mechanism by which NAC as a reducing agent alters cellular functions to arrest cell cycle is unclear, but NAC may act directly on the sulfhydryl group of cellular components without receptor-mediated signaling. In this case, target proteins modulated by NAC might contain reactive cysteine residues that participate in a thiol-disulfide reaction through a redox status. Recently, redox-sensitive cellular signal transduction components have been reported. These include Ras, Raf-1, and transcription factors such as AP-1 and NF-B. Direct evidence of the redox regulation of Ras by binding nitric oxide to a cysteine residue (Cys-118) has been presented; this process triggers its guanine nucleotide exchange and downstream signaling (45). In the redox regulation of Raf, the importance of cysteine residues in the zinc finger motif of Raf has been demonstrated by the fact that cysteine mutation results in the suppression of the Ras-Raf interaction and activation of the Raf kinase (46). In cases of transcription factors, AP-1 and NF-B are subjected to redox regulation through their conserved cysteine residue (47,48). In this report, we demonstrated that the NAC signaling pathway is subjected to a reducing activity in which reactive cysteine residues of several proteins containing Raf-1, MEK, and ERK turn to a reduced state by NAC (Fig. 8). However, it remains to be determined whether the redox status of Raf-1, MEK, and ERK are directly associated with their kinase activities or indirectly associated with a redox-regulatory protein (e.g. thioredoxin) thought to have cell arrest effects.
Cellular responses to the activation of the MEK/MAPK pathway such as cell cycle arrest and cellular proliferation depend on the strength and duration of the MAPK signal; transient activation may contribute to cell cycle progression, whereas sustained high levels of activity may result in cell cycle arrest via the induction of p21 Cip1/Waf1 expression and the inhibition FIG. 7. Cell cycle progression by integrin ␣ 1 (␣ 2 )␤ 1 -type I collagen interaction overrides an NAC-induced cell arrest. A, morphological changes of HSC in various ECM protein-coated plastics in the presence (NAC) or absence (control) of 5 mM NAC. ECM-coated flasks were prepared as described under "Experimental Procedures." B, NAC does not change the amounts and integrity of integrin ␣ 1 and ␣ 2 . The collagen binding assay was performed as described under "Experimental Procedures." FIG. 8. NAC modulates the redox states of cysteine residues and the activities of the target proteins. A, HSC were cultured in the absence (control) or presence (NAC) of 5 mM NAC for 6 h and solubilized in lysis buffer. Cell lysates were incubated with EZ-Link TM PEO-iodoacetyl Biotin for 12 h at 4°C and then incubated with Ultra-Link TM immobilized NeutrAvidin TM for 4 h at 4°C. Proteins containing sulfhydryl groups were subjected to SDS-PAGE and transferred to nitrocellulose membranes. Signaling components were probed with the appropriate primary antibodies and developed with ECL reagents. B, the dose-dependent activation of the immunoprecipitated kinases by NAC. Raf-1, MEK, and ERK were partially purified from HSC cell lysates by using immunoprecipitation kinase assay kits (Upstate Biotechnology). Each immunoprecipitated kinase was incubated with various concentrations of NAC for 1 h. The activity of each kinase was measured using the appropriate kits as outlined by the manufacturer.