By continuing to browse this site you agree to us using cookies as described in About Cookies
Notice: Wiley Online Library will be unavailable on Saturday 7th Oct from 03.00 EDT / 08:00 BST / 12:30 IST / 15.00 SGT to 08.00 EDT / 13.00 BST / 17:30 IST / 20.00 SGT and Sunday 8th Oct from 03.00 EDT / 08:00 BST / 12:30 IST / 15.00 SGT to 06.00 EDT / 11.00 BST / 15:30 IST / 18.00 SGT for essential maintenance. Apologies for the inconvenience.
Potential conflict of interest: Nothing to report.
The proliferation of hepatic stellate cells (HSCs) is a critical step in hepatic fibrogenesis. Platelet-derived growth factor (PDGF) is the most potent mitogen for HSCs. We investigated the role of nonphagocytic NAD(P)H oxidase–derived reactive oxygen species (ROS) in PDGF-induced HSC proliferation. The human HSC line, LI-90 cells, murine primary-cultured HSCs, and PDGF-BB were used in this study. We examined the mechanism of PDGF-BB-induced HSC proliferation in relation to the role of a ROS scavenger and diphenylene iodonium, an inhibitor of NAD(P)H oxidase. We also measured ROS production with the aid of chemiluminescence. We showed that PDGF-BB induced proliferation of HSCs through the intracellular production of ROS. We also demonstrated that HSCs expressed key components of nonphagocytic NAD(P)H oxidase (p22phox, gp91phox, p47phox, and p67phox) at both the messenger RNA and protein levels. Diphenylene iodonium suppressed PDGF-BB–induced ROS production and HSC proliferation. Coincubation of H2O2 and PDGF-BB restored the proliferation of HSCs that was inhibited by diphenylene iodonium pretreatment. Phosphorylation of the mitogen-activated protein kinase (MAPK) family constitutes a signal transduction pathway of cell proliferation. Our data demonstrate that NAD(P)H oxidase–derived ROS induce HSC proliferation mainly through the phosphorylation of p38 MAPK. Moreover, an in vivo hepatic fibrosis model also supported the critical role of NAD(P)H oxidase in the activation and proliferation of HSCs. In conclusion, NAD(P)H oxidase is expressed in HSCs and produces ROS via activation of NAD(P)H oxidase in response to PDGF-BB. ROS further induce HSC proliferation through the phosphorylation of p38 MAPK. (HEPATOLOGY 2005;41:1272–1281.)
In chronic liver diseases, regardless of their nature (viral infection, alcohol abuse, metal overload), the progression of hepatic fibrosis is associated with the development of portal hypertension and the occurrence of hepatocellular carcinoma. Hepatic stellate cells (HSCs) are increasingly being recognized as the key mediators of the progression of hepatic fibrosis.1 In chronic liver diseases, HSCs undergo a process of activation, developing a myofibroblast-like phenotype that is associated with increased proliferation, chemotaxis, and collagen synthesis.2 Of these, the degree of fibrogenesis that occurs in liver diseases is most likely to be affected by an increased number of HSCs, which results from their proliferation.3 The results of in vitro and in vivo studies suggest that platelet-derived growth factor (PDGF) is the most potent mitogen of HSCs and is therefore likely to be an important mediator of the increased proliferation of HSCs during hepatic fibrogenesis in chronic liver diseases.4, 5 Immunohistochemistry and in situ hybridization studies have revealed that PDGF and PDGF receptors are overexpressed at both the messenger RNA (mRNA) and protein levels in liver tissue from patients with chronic hepatitis or cirrhosis, and are positively correlated with the severity of histological lesions and collagen deposition.6, 7 These reports suggest strongly that PDGF facilitates the progression of hepatic fibrosis in human chronic liver diseases. Therefore, clarification of the cellular and molecular mechanisms underlying the PDGF-induced proliferation of HSCs will aid in the development of a new antifibrotic therapy for chronic liver diseases.
Reactive oxygen species (ROS) such as superoxide anion, hydrogen peroxide (H2O2), and hydroxyl radical can stimulate HSC proliferation and collagen synthesis.8, 9 Although both PDGF and ROS have been implicated in the initiation and progression of hepatic fibrogenesis, the relationship between PDGF and ROS in HSC proliferation has yet to be clarified. Recently, it has been shown that nonphagocytic nicotinamide-adenine dinucleotide phosphate [NAD(P)H] oxidase exists in vascular smooth muscle cells and endothelial cells and is involved in the proliferation of these types of cells through the continuous, low-level intracellular production of ROS.10–12 A previous in vivo study has shown that p22phox and gp91phox, which are anchored in the cell membrane, are expressed in HSCs at both the mRNA and protein levels.13 However, it is presently uncertain whether PDGF causes HSC proliferation through NAD(P)H oxidase–derived ROS. In the present study, we examined whether key components of NAD(P)H oxidase—p22phox, gp91phox, p47phox, and p67phox—are expressed in HSCs and investigated whether PDGF enhances the production of NAD(P)H-derived ROS. Moreover, we clarified the role of NAD(P)H oxidase–derived ROS in the process of PDGF-induced HSC proliferation. Here we have shown that NAD(P)H oxidase–derived ROS play a critical role in PDGF-mediated HSC proliferation and the in vivo hepatic fibrosis model.
Human recombinant PDGF-BB was purchased from Austral Biologicals (San Ramon, CA). Manganese (III) tetrakis (benzoic acid) porphyrin chloride (Mn-TBAP) was purchased from Cayman Chemical Company (Ann Arbor, MI). Diphenylene iodonium (DPI) and 10-methyl-9-(phenoxycarbamoyl)acridinium fluorosulfonate (PMAC) were purchased from Dojindo Laboratories (Kumamoto, Japan). Allopurinol, indomethacin, apocynin, PD98059, and SB203580 were obtained from Sigma Chemical Company (St. Louis, MO). All other chemicals and reagents were of analytical grade and, if not stated, were purchased from Sigma or from Wako Pure Chemical Industries (Osaka, Japan).
Rabbit anti-human p22phox, gp91phox, p47phox, and p67phox antibodies were kindly provided by Dr. S. Imajoh-Ohmi.14 Rabbit anti-phospho–extracellular signal-related kinase (ERK) and anti-phospho–p38 mitogen-activated protein kinase (p38 MAPK) antibodies were obtained from Cell Signaling Technology, (Beverly, MA). Mouse anti–α-smooth muscle actin and anti-proliferating cell nuclear antigen (PCNA) were obtained from DAKO (Kyoto, Japan).
The HSC line LI-90, which exhibits characteristics compatible with those of human HSCs, was used in this study.15 LI-90 was kindly provided by Human Science Cell Bank (Saitama, Japan). The experiments were performed on cells between the third and tenth serial passages (1:3 split ratio) using originally supplied LI-90.
Primary-cultured HSCs were isolated from female BALB/c mice (6 weeks old) via in situ collagenase perfusion and differential centrifugation on Nycodenz (Pharma AS, Oslo, Norway) density gradients, as previously described.16 Cultured HSCs at second passage were used in this study. Both LI-90 cells and primary-cultured HSCs were cultured in a 5% CO2 humidified incubator at 37°C. Dulbecco's Modified Eagle Medium (DMEM) (GIBCO BRL, Rockville, MD) containing 10% fetal bovine serum (FBS) (Filton, Brooklyn, Australia) was used as the growth medium. After cells became subconfluent (at 70%-80% confluence), the cells were cultured with DMEM without containing phenol red and FBS for 24 hours (serum starvation) before the start of all experiments.
Cell Proliferation Assay.
LI-90 cells or primary-cultured HSCs were plated in 96-well microplates (Sumitomo Bakelite Co., Tokyo, Japan) at a density of 7 × 103 cells/well in complete culture medium. The cells proliferation assay was performed using the Prex WST-1 cell proliferation assay system (Takara, Osaka, Japan).17 The number of viable cells was estimated at a wavelength of 450 nm on an enzyme-linked immunosobent assay plate reader 1 hour after the addition of WST-1. To confirm that the cell proliferation assay reflected the increase in cell number, the number of cells was counted by the trypan blue exclusion method after trypsinization.
DNA synthesis was also measured using a Biotrack cell proliferation ELISA system (Amersham, Little Chalfont, UK).18 During the last 4 hours, the cells were labeled with bromodeoxyuridine. After removing the culture medium, the cells were fixed and the incorporated bromodeoxyuridine was detected according to the manufacturer's recommended protocol.
Reverse-Transcription Polymerase Chain Reaction.
Total cellular RNA was extracted from the cells, using Isogen (Nippon Gene, Tokyo, Japan) as described in the product protocol. cDNA was generated from 1 μg of total RNA, using random hexanucleotide primers, and reverse-transcription polymerase chain reaction was performed. The specific primers were the same as previously reported.11
Immunoblots and Immunocytochemistry.
Western blotting (10% SDS-PAGE) was performed as described elsewhere.19 The bands were visualized using a Lumiglo Substrate Kit (Kirkegaard & Perry Laboratories, Gaithersburg, MD).
Immunocytochemistry was performed using antibodies recognizing p22phox (1:100 dilution), gp91phox (1:100 dilution), p47phox (1:100 dilution), or p67phox (1:100 dilution). Immunostaining of the cells was accomplished using a commercial kit (Vector Laboratories, Burlingame, CA) according to the manufacturer's protocol.
Measurement of ROS Production in HSCs.
To measure intracellular ROS production in LI-90 cells and primary-cultured HSCs, we used a modification of a previously described method using PMAC-enhanced chemiluminescence.20 This method is based on the reaction between reduced PMAC and ROS, which results in the emission of light. The chemiluminescence images were acquired with a high-performance intensified CCD camera (PI-MAX 512RB Princeton Instruments, Nippon Roper, Chiba, Japan) that was placed 30 cm above a black polystyrene 96-well plate containing HCSs set in a dedicated black box.21 Chemiluminescence was measured every 2 minutes for 0 to 16 minutes after the addition of phosphate-buffered saline (PBS) containing PDGF-BB and PMAC.
Animals and Treatment.
The in vivo experiments were performed in accordance with the Yamagata University School of Medicine Guidelines for the Care and Use of Laboratory Animals. Specific pathogen-free BALB/c female mice (6 weeks old) were used. Hepatic fibrosis was induced by intraperitoneal (IP) injection of 10 mg/kg dimethylnitrosamine (DMN) in saline three times a week for the indicated periods for 6 weeks. Daily IP administration of Mn-TBAP (5 mg/kg) or DPI (1 mg/kg) was begun simultaneously with DMN administration. Animals were divided into the following groups: (1) control (daily IP administration of 0.1 mL PBS [n = 6]); (2) DMN (daily IP administration of 0.1 mL PBS [n = 6]); (3) DMN + Mn-TBAP (daily IP administration of 0.1 mL Mn-TBAP solution in PBS [1 mg/mL] [n = 6]); and (4) DMN + DPI (daily IP administration of 0.1 mL DPI solution in PBS [0.2 mg/mL] [n = 6]).
Histological and Immunohistochemical Examination.
Liver specimens were routinely fixed in 4% formaldehyde in PBS and embedded in paraffin. Tissue sections (4 μm thick) were either stained with Sirius red or subjected to immuno-histostaining with antibodies against PCNA. Specific staining was visualized using a commercial kit (Vector Laboratories).
Data are expressed as the mean ± SD. Statistical comparisons were made using the Student t test for unpaired samples. For studies involving more than two groups, a two-way ANOVA was determined using Scheffe's test. The level of statistical significance was set at P < .05 for all cases.
ROS Scavenger Suppressed the PDGF-BB–Induced Proliferation of HSCs.
PDGF-BB induced a dose-dependent increase in LI-90 proliferation, as measured via WST-1 assay (Fig. 1A). PDGF-BB at a concentration of 5 ng/mL, 10 ng/mL, and 20 ng/mL significantly increased cell proliferation after 24 hours of incubation compared with the unstimulated control cells. To assess whether the PDGF-BB–induced increase in formazan formation measured via WST-1 assay was really associated with cell growth, cell count experiments were also performed with the aid of trypsinization. As shown in Fig. 1B, the addition of PDGF-BB (20 ng/mL) in FBS-free DMEM caused a significant increase in cell number.
We pretreated LI-90 cells with Mn-TBAP, a cell-permeable SOD/catalase mimetic, as intracellular ROS scavenger for 30 minutes. As shown in Fig. 1C, Mn-TBAP dose-dependently suppressed the PDGF-BB–induced proliferation of cells. Particularly, Mn-TBAP at a concentration of 100 nmol/L significantly suppressed the PDGF-BB–induced proliferation of cells, and this same dose did not exert any toxic effects on cells after incubation times as long as 36 hours (data not shown). Bromodeoxyuridine analysis showed that PDGF-BB at a concentration of 20 ng/mL significantly stimulated DNA synthesis. Mn-TBAP dose-dependently inhibited the DNA synthesis induced by PDGF-BB (Fig. 1D).
HSCs Express Four Enzymatic Subunits of NAD(P)H Oxidase.
Reverse-transcription polymerase chain reaction analysis showed that four enzymatic subunits of NAD(P)H oxidase were expressed at the mRNA level (Fig. 2A). Western blot analysis revealed that four enzymatic subunits of NAD(P)H oxidase were expressed at the protein level in LI-90 cells (Fig. 2B). Immunocytochemical analysis of cultured cells showed positive staining for p22phox, gp91phox, p47phox, and p67phox (Fig. 2C). No cultured cells showed positive immunostaining when the antibodies were replaced with a control serum (data not shown).
NAD(P)H Oxidase Is the Main Source of PDGF-BB–Induced ROS Production by HSCs.
To investigate whether LI-90 cells produced ROS in response to PDGF-BB, we incubated cells with PDGF-BB (20 ng/mL) and PMAC (500 μmol/L) and measured the PMAC-enhanced chemiluminescence with a high-performance intensified CCD camera at an interval of 2 minutes. PDGF-BB caused a marked increase in the chemiluminescence from PMAC immediately after the incubation. The chemiluminescence observed in response to PDGF-BB peaked at 6 minutes after the incubation and declined gradually thereafter (Fig. 3A). To confirm that the chemiluminescence accurately reflected the increased production of ROS, we examined whether Mn-TBAP, a cell-permeable ROS scavenger, was able to decrease the enhanced chemiluminescence by PDGF-BB. The chemiluminescence was completely abolished when cells were preincubated with Mn-TBAP for 30 minutes (Fig. 3B). However, superoxide dismutase, a cell-nonpermeable enzyme, failed to inhibit the enhanced chemiluminescence induced by PDGF-BB (data not shown). Pretreatment with DPI for 30 minutes, an inhibitor of NAD(P)H oxidase, remarkably reduced PDGF-BB–induced chemiluminescence (Fig. 3B). Pretreatment with apocynin (100 μmol/L), another type of NAD(P)H oxidase inhibitor, for 1 hour also reduced the enhanced chemiluminescence induced by PDGF-BB (data not shown).
A high concentration of DPI inhibits flavoproteins including NAD(P)H oxidase, mitochondrial oxidases, xanthine oxidase, and cyclooxygenase. To ascertain that the enhanced chemiluminescence did not originate from a mitochondrial electron transport system, we examined whether the mitochondrial poison potassium cyanide (KCN) inhibited the PDGF-BB–enhanced chemiluminescence. As shown in Fig. 3C, pretreatment with KCN (500 μmol/L) for 30 minutes had no significant effect on the enhanced chemiluminescence induced by PDGF-BB. Preincubation for 1 hour with a maximally effective concentration of allopurinol (an inhibitor of xanthine oxidase: 100 μmol/L) or indomethacin (an inhibitor of cyclooxygenase: 100 μmol/L) failed to inhibit the PDGF-BB–mediated enhanced chemiluminescence (Fig. 3D).
NAD(P)H Oxidase–Derived ROS Are Essential to the PDGF-BB–Induced Proliferation of HSCs.
To elucidate whether PDGF-BB–induced proliferation of LI-90 cells was mediated by the activation of NAD(P)H oxidase, we examined the effect of DPI on the PDGF-BB–induced proliferation of cells. Pretreatment of the cells with DPI (25 μmol/L) for 30 minutes significantly inhibited the PDGF-BB–induced proliferation of cells (Fig. 4). DPI at a concentration of 25 μmol/L did not exert any toxic effects on cells after incubation times as long as 36 hours (data not shown). Pretreatment with apocynin (100 μmol/L), another type of NAD(P)H oxidase inhibitor, for 1 hour also inhibited the PDGF-BB–induced proliferation of cells (Fig. 4). Preincubation with neither indomethacin (an inhibitor of cyclooxygenase) nor allopurinol (an inhibitor of xanthine oxidase) for 1 hour attenuated the PDGF-BB–induced proliferation of cells (Fig. 4). Coincubation of H2O2 and PDGF-BB restored the LI-90 proliferation that was inhibited by DPI pretreatment (Fig. 4).
PDGF-BB Mediates the Phosphorylation of p38 MAPK Through the Activation of NAD(P)H Oxidase.
PDGF-BB–induced LI-90 proliferation was significantly suppressed by preincubation for 1 hour with PD98059 (an inhibitor of MEK upstream activator of ERK: 25 μmol/L) or SB203580 (an inhibitor of p38 MAPK: 25 μmol/L). The degree of suppression was more significant with SB203580 than with PD98059 (Fig. 5A). To investigate the relationship between NAD(P)H oxidase–derived ROS and the phosphorylation of MAPKs, we first examined the effect of DPI or Mn-TBAP on the phosphorylation of ERKs and p38 MAPK in cells. PDGF-BB caused the phosphorylation of ERKs and p38 MAPK in cells, with a peak at 10 to 15 minutes (Fig. 5B). Preincubation of cells with PD98059 or SB203580 for 1 hour led to the inhibition of PDGF-induced ERKs or p38 MAPK phosphorylation, respectively (data not shown). As shown in Fig. 5B, preincubation with DPI inhibited the phosphorylation of p38 MAPK but had no effect on the phosphorylation of ERK induced by PDGF-BB. Preincubation with Mn-TBAP also inhibited the phosphorylation of p38 MAPK induced by PDGF-BB but did not affect the phosphorylation of ERKs (Fig. 5B). Coincubation of H2O2 (100 nmol/L) and PDGF-BB (20 ng/mL) restored the phosphorylation of p38 MAPK that was inhibited by DPI pretreatment (Fig. 5C).
NAD(P)H Oxidase–Derived ROS Are Essential for the PDGF-BB–Induced Proliferation of Primary Cultured HSCs.
To ascertain whether the phenomenon observed was not an artifact of the cell line used, we examined the role of NAD(P)H oxidase in PDGF-BB–induced proliferation of murine primary-cultured HSCs. Reverse-transcription polymerase chain reaction analysis showed that four enzymatic subunits of NAD(P)H oxidase were expressed at the mRNA level (Fig. 6A). Western blot analysis revealed that 4 enzymatic subunits of NAD(P)H oxidase were expressed at the protein level in the primary-cultured HSCs (data not shown). PDGF-BB caused a marked increase in the chemiluminescence from PMAC immediately after the incubation with primary-cultured HSCs, which peaked at 6 to 8 minutes (Fig. 6B). The chemiluminescence was abolished when cells were preincubated with Mn-TBAP or DPI for 30 minutes (Fig. 6C).
As shown in Fig. 6D, PDGF-BB stimulated the DNA synthesis of primary-cultured HSCs. This PDGF-BB–induced DNA synthesis was inhibited by pretreatment with Mn-TBAP, DPI for 30 minutes, or apocynin for 1 hour. Coincubation of H2O2 and PDGF-BB restored the DNA synthesis that was inhibited by DPI pretreatment for 30 minutes (Fig. 6D). Moreover, PDGF-BB–induced DNA synthesis of primary-cultured HSCs was significantly suppressed by pretreatment with SB203580 or PD98059 for 1 hour. Similar to the result for LI-90, the degree of suppression was more significant with SB203580 than with PD98059 (Fig. 6D).
Mn-TBAP and DPI Attenuate the Activation and Proliferation of HSCs in a Hepatic Fibrosis Model.
Murine chronic liver injury was induced via IP injection of DMN to assess the role of NAD(P)H oxidase in hepatic fibrosis. Mice in DMN + Mn-TBAP and DMN + DPI groups showed no evidence of connective tissue septa (Fig. 7A). Western blotting analysis indicated that α-smooth muscle actin expression was increased in DMN-induced fibrotic liver but that Mn-TBAP or DPI reduced the expression of α-smooth muscle actin in DMN-induced hepatic fibrosis (Fig. 7B). Immunostaining for PCNA, an index of HSC proliferation, also showed an increased number of S-phase HSCs in the DMN group and a reduced number of S-phase HSCs in the DMN + Mn-TBAP and DMN + DPI groups (Fig. 7C).
Because PDGF is the most potent mitogen for HSCs, clarification of the mechanisms underlying the PDGF-induced proliferation of HSCs is clinically important for the establishment of new antifibrotic therapies of chronic liver diseases. In the present study, we investigated the mechanisms underlying PDGF-BB–induced HSC proliferation, using not only the human HSC line, LI-90, but also primary-cultured HSCs. PDGF is a dimer of two polypeptide chains, A and B, forming three isoforms known as PDGF-AA, PDGF-AB, and PDGF-BB. Of these, PDGF-BB has the most potent activity.22 In this study, PDGF-BB induced proliferation of HSCs through the intracellular production of ROS. Moreover, NAD(P)H oxidase–derived ROS stimulated the phosphorylation of p38 MAPK, a stress-activated protein kinase, and formed the signaling pathway that mediated the PDGF-induced proliferation of HSCs. The selective suppression of NAD(P)H oxidase in HSCs may be relevant to the development of new antifibrotic therapy for chronic liver diseases.
ROS—which include the superoxide anion, H2O2, and hydroxyl radical—are highly reactive molecules. It is now well established that lower, more “physiological” levels of ROS can exert regulatory roles within the cells. Recently, a series of reports has documented that the intracellular redox system can also modulate protein phosphorylation and dephosphorylation as a result of the presence of redox-sensitive functional groups in the structures of both protein kinases and phosphatases.23, 24 Redox homeostasis also regulates the activation of transcription factors. Our present results show that Mn-TBAP, a cell-permeable ROS scavenger, suppresses the PDGF-BB–induced proliferation of HSCs. The effective concentration of Mn-TBAP was in the nanomole order and is clinically achievable in humans. Moreover, H2O2 significantly increased HSC proliferation. These results suggest that a low level of ROS induces the proliferation of HSCs and is essential for PDGF-BB–mediated HSC proliferation.
To clarify the mechanism of ROS production in HSCs by PDGF-BB, we detected ROS production by PMAC-enhanced chemiluminescence with a high-performance intensified CCD camera. The ROS detection system we employed in the present study can minimize sampling error and has the advantage of being able to quantify intracellular ROS production by various types of cultured cells.21 The chemiluminescence was completely abolished when cells were incubated with Mn-TBAP, a cell-permeable ROS scavenger. However, superoxide dismutase, a cell-nonpermeable enzyme, failed to inhibit the enhanced chemiluminescence induced by PDGF-BB. Taken together with the finding that Mn-TBAP significantly inhibited PDGF-induced HSC proliferation, ROS produced after incubation with PDGF-BB appear to be implicated in PDGF-BB–induced HSC proliferation.
NAD(P)H oxidase is a four-subunit enzyme consisting of two membrane components, p22phox and gp91phox, and two cytosolic components, p47phox and p67phox. A previous in vivo study showed that p22phox and gp91phox, anchored in the cell membrane, are expressed in HSCs at both the mRNA and protein levels.13 However, the role of NAD(P)H oxidase in HSCs has not been understood until recently. Bataller et al. reported that NAD(P)H oxidase–derived ROS were involved in the signal transduction pathway of angiotensin II.25 They showed p47phox, gp91phox, and Nox1, a homologue of gp91phox, were expressed at the mRNA level. In the present study, we demonstrated that p22phox, gp91phox, p47phox, and p67phox were expressed in HSCs at both the mRNA and protein levels. PDGF-BB induced ROS production through the activation of NAD(P)H oxidase, and this was confirmed by the inhibition of ROS production by NAD(P)H oxidase inhibitors, DPI, and apocynin. Because ROS production by NAD(P)H oxidase in HSCs is observed not only in angiotensin II but also in PDGF, as we showed in the present study, NAD(P)H oxidase– derived ROS may act as common intracellular signaling molecules for hepatic fibrogenesis of HSCs in human chronic liver diseases.
In the present study, we used DPI as an inhibitor of NAD(P)H oxidase. However, a high concentration of DPI has known to inhibit flavoprotein enzymes including NAD(P)H oxidase, mitochondrial oxidases, xanthine oxidase, and cyclooxygenase.26 To ensure that the effect of DPI (25 μmol/L) actually reflected NAD(P)H oxidase inhibition, we examined whether the observed chemiluminescence was sensitive to enzyme inhibitors of mitochondrial oxidases, xanthine oxidase, and cyclooxygenase. In intact HSCs, we found that PDGF-BB–induced ROS production was insensitive to KCN, an inhibitor of mitochondria oxidases, allopurinol, an inhibitor of xanthine oxidase, and indomethacin, an inhibitor of cyclooxygenase. PDGF-BB–induced HSCs proliferation was also inhibited by DPI but not by allopurinol and indomethacin. Moreover, apocynin, another type of NAD(P)H oxidase inhibitor, inhibited PDGF-BB–mediated ROS production and HSC proliferation. Therefore, the experimental results obtained from DPI at a concentration of 25 μmol/L reflected the inhibition of NAD(P)H oxidase.
Previous studies have shown that PDGF positively regulates HSC proliferation by indirectly activating ERK via the activation of Ras in culture systems.27, 28 Our results show that ERK is involved in PDGF-mediated HSC proliferation, independently of its association with oxidative stress. However, activation of ERK alone did not sufficiently induce HSC proliferation. Our results showed that p38 MAPK, a redox-sensitive MAPK, was more closely linked to PDGF-induced HSC proliferation than was ERK. These results suggest strongly that in relation to PDGF-NAD(P)H oxidase–mediated HSC proliferation, p38 MAPK is a growth stimulator. Recently, it was proposed that p38 MAPK plays functional roles in HSC activation and proliferation.29, 30 Our results that p38 MAPK plays an important role in HSC proliferation is inconsistent with these reports.
Phosphatidylinositol 3-kinase or Na+/K+ exchanger activation have also been reported to be involved in PDGF-induced HSC proliferation.31, 32 Phosphatidylinositol 3-kinase associates with phosphotyrosine residue of PDGF receptor (src homology 2 domains) and is considered to be relevant to PDGF-mediated HSC proliferation.31 Although we did not show the data, we found that LY294002, an inhibitor of phosphatidylinositol 3-kinase, suppressed NAD(P)H oxidase activity. We therefore consider that phosphatidylinositol 3-kinase is located upstream of NAD(P)H oxidase. Oxidative stress is known to increase Na+/K+ exchanger activation, and thus NAD(P)H oxidase–derived ROS may induce HSC proliferation through Na+/K+ exchanger activation. Nuclear receptors such as peroxisome proliferator–activated receptors are also thought to be involved in PDGF-induced HSC proliferation.27, 30 The expression of peroxisome proliferator–activated receptor β was reported to be stimulated by p38 MAPK. The results of the present study reveal that the PDGF/NAD(P)H oxidase/ROS/p38 MAPK activation pathway may mediate HSC proliferation and may form a network linking the already well-established pathways that mediate PDGF-induced HSC proliferation.
Our in vitro results revealed that NAD(P)H oxidase–derived ROS play an important role in PDGF-mediated HSC proliferation. NAD(P)H oxidase–derived ROS are also thought to be involved in angiotensin II–mediated liver fibrosis.25 Furthermore, it has been reported that NAD(P)H oxidase is involved in the transforming growth factor β1 signaling pathway in various types of cells.33 To explore the possibility that the findings for HSCs can also be adapted to an in vivo model of hepatic fibrosis, we performed the in vivo experiments using mice with chronic liver injury. An experiment involving an in vivo animal experiment of hepatic fibrosis induced by DMN showed that Mn-TBAP, an intracellular ROS scavenger, and DPI, an inhibitor of NAD(P)H oxidase, significantly suppressed the activation and proliferation of HSCs. Although further study using more selective and specific drugs will be required, our preliminary in vivo findings suggest that NAD(P)H oxidase and NAD(P)H oxidase–derived ROS might constitute a pathway for activation and proliferation of HSCs, consistent with the in vitro findings. Recently, carriers for the targeting of specific drugs to activated HSCs have been under intense investigation.34 By controlling the delivery and release of drugs within activated HSCs in vivo, drugs such as antioxidants, NAD(P)H oxidase inhibitors, or small interfering RNA for enzymatic components of NAD(P)H oxidase may be successfully targeted to activated HSCs, providing a novel therapy for chronic liver diseases.
In conclusion, we have investigated the mechanisms underlying PDGF-BB–mediated HSC proliferation using LI-90 cells and primary cultured HSCs. In this study, PDGF-BB activated NAD(P)H oxidase, resulting in the generation of ROS. NAD(P)H oxidase–derived ROS then activated p38 MAPK and induced the proliferation of HSCs. An in vivo hepatic fibrosis model also supported the critical role of NAD(P)H oxidase in the activation and proliferation of HSCs. We have clarified a pathway that mediates the proliferation of HSCs: PDGF/NAD(P)H oxidase/ROS/p38 MAPK.
The authors thank Dr. S. Imajoh-Ohmi (The Institute of Medical Science, University of Tokyo) for providing anti-human p21phox, gp91phox, p47phox, and p67phox antibodies and the Human Science Cell Bank (Saitama, Japan) for providing the human hepatic stellate cell line LI-90. The authors also thank the editors and reviewers for their helpful comments and suggestions.