Systemic infusion of angiotensin II exacerbates liver fibrosis in bile duct–ligated rats

Authors


  • Potential conflict of interest: Nothing to report.

Abstract

Recent evidence indicates that the renin–angiotensin system (RAS) plays a major role in liver fibrosis. Here, we investigate whether the circulatory RAS, which is frequently activated in patients with chronic liver disease, contributes to fibrosis progression. To test this hypothesis, we increased circulatory angiotensin II (Ang II) levels in rats undergoing biliary fibrosis. Saline or Ang II (25 ng/kg/h) were infused into bile duct–ligated rats for 2 weeks through a subcutaneous pump. Ang II infusion increased serum levels of Ang II and augmented bile duct ligation–induced liver injury, as assessed by elevated liver serum enzymes. Moreover, it increased the hepatic concentration of inflammatory proteins (tumor necrosis factor α and interleukin 1β) and the infiltration of CD43-positive inflammatory cells. Ang II infusion also favored the development of vascular thrombosis and increased the procoagulant activity of tissue factor in the liver. Livers from bile duct–ligated rats infused with Ang II showed increased transforming growth factor β1 content, collagen deposition, accumulation of smooth muscle α-actin–positive cells, and lipid peroxidation products. Moreover, Ang II infusion stimulated phosphorylation of c-Jun and p42/44 mitogen-activated protein kinase and increased proliferation of bile duct cells. In cultured rat hepatic stellate cells (HSCs), Ang II (10−8 mol/L) increased intracellular calcium and stimulated reactive oxygen species formation, cellular proliferation and secretion of proinflammatory cytokines. Moreover, Ang II stimulated the procoagulant activity of HSCs, a newly described biological function for these cells. In conclusion, increased systemic Ang II augments hepatic fibrosis and promotes inflammation, oxidative stress, and thrombogenic events. (HEPATOLOGY 2005;41:1046–1055.)

The renin–angiotensin system (RAS) plays an important role in diseases characterized by chronic inflammation and tissue remodeling.1 Angiotensin II (Ang II), the main effector of the RAS, regulates key steps in the tissue remodeling process through angiotensin type 1 (AT1) receptors.2 In target cells, Ang II induces free radical formation and oxidative stress, stimulating redox-sensitive intracellular pathways.3 Ang II–induced biological effects include cell contraction, cell growth/hypertrophy, and secretion of inflammatory cytokines.4, 5 Moreover, Ang II stimulates collagen deposition.6 This latter effect is mediated by the induction of transforming growth factor β1 (TGF-β1), a powerful fibrogenic cytokine.7 Finally, Ang II induces angiogenesis and favors thrombosis development.8, 9

There is increasing evidence that Ang II may play a role in hepatic fibrosis. The systemic RAS is activated in patients with cirrhosis, and a local RAS is induced in fibrotic livers and activated hepatic stellate cells (HSCs).10–13 The blockade of Ang II synthesis or its binding to AT1 receptors markedly ameliorates hepatic fibrosis in rats.14–19 Moreover, patients with chronic hepatitis C and a genetic polymorphism associated with increased Ang II synthesis develop more severe hepatic fibrosis.20 The mechanisms through which Ang II is profibrogenic in the liver are unknown. Ang II is secreted and binds to AT1 receptors to induce contraction and proliferation in human HSCs.21 Recently, we showed that prolonged systemic infusion of Ang II into normal rats causes HSC activation and hepatic inflammation.22 However, Ang II infusion did not induce hepatic fibrosis. These results suggest that increased systemic Ang II may theoretically augment fibrosis in livers undergoing tissue remodeling.

In the current study, we increased systemic levels of Ang II in rats undergoing bile duct ligation. Ang II infusion into bile duct–ligated rats resulted in increased liver injury, inflammation, and fibrosis. Moreover, Ang II infusion induced oxidative stress and vascular thrombosis. At the cellular level, Ang II induced similar effects in primary rat HSCs. Based on these results, we propose that activation of systemic RAS may contribute to the progression of chronic liver disease.

Abbreviations

RAS, renin–angiotensin system; Ang II, angiotensin II; HSC, hepatic stellate cell; AT1, angiotensin type 1; TGF-β1, transforming growth factor β1; HIF, hypoxia-inducible factor; ERK, extracellular-regulated kinase; PCNA, proliferating cell nuclear antigen.

Materials and Methods

Experimental Design.

Bile duct ligation or sham operation was performed in male Sprague-Dawley rats (330–400 g). After midline laparotomy, the common bile duct was doubly ligated with 4-0 silk and transected between the two ligations. The sham operation was performed similarly, with the exception of ligating and transecting the bile duct. Immediately after surgery, both sham-operated and bile duct–ligated rats were randomized to receive Ang II (25 ng/kg/min) or saline through an osmotic minipump (Alza, Palo Alto, CA). Minipumps designed to deliver their content in 2 weeks were used. At sacrifice, only minimum amounts of saline or Ang II were observed in the pumps. Ten rats were included in each experimental group. Two bile duct–ligated rats (one receiving saline and one receiving Ang II) died. Animals were sacrificed 2 weeks after surgery. For arterial blood pressure measurement, animals were anesthesized with pentobarbital (30 mg/kg intraperitoneally) and the right carotid artery was cannulated (PE-90; Transonics Systems, Ithaca, NY). After a stabilization period of 5 minutes, the mean arterial blood pressure was recorded using a pressure analyzer (LPA-200; Digi-Med, Louisville, KY) for 10 minutes. The blood pressure value for each group was calculated as the average of four separate animals. Animal procedures were performed in accordance with the Institutional Animal Care and Use Committee of the University of North Carolina.

Cell Culture.

HSCs were isolated from male Sprague-Dawley rats (400 g) via in situ infusion with collagenase and pronase, followed by arabinogalactan gradient ultracentrifugation, as previously described.23 HSC purity was above 95%. Cells were seeded on uncoated plastic tissue culture dishes and cultured in Dulbecco's Modified Eagle Medium (Gibco BRL, Grand Island, NY) supplemented with 10% fetal calf serum, 2 mmol/L L-glutamine, and standard antibiotics.

Serum Biochemical Measurements.

Serum alanine aminotransferase, aspartate aminotransferase, and bilirubin levels were measured using standard enzymatic procedures by the Pathology Department at the University of North Carolina. Serum Ang II levels were measured via sandwich ELISA (R & D Systems, Minneapolis, MN).

Measurement of Serum Endotoxin Levels.

Blood samples were collected in endotoxin-free vials and centrifuged at 400g for 15 minutes at 4°C. Samples were then diluted 1:10 in pyrogen-free water and heated to 75°C for 30 minutes to remove inhibitors of endotoxin from plasma. The limulus amoebocyte lysate test (Kinetic-QLC; Whittaker Bioproducts, Walkerville, MD) was used for measurements of endotoxin. Samples were read spectrophotometrically at 410 nm.

Immunohistochemical Studies.

Five micrometer sections of paraffin-embedded tissues were stained with hematoxylin & eosin and Masson-trichrome. For immunohistochemical analysis, sections were stained using the DAKO Envision system (DAKO, Carpinteria, CA). Sections were incubated with anti-4-hydroxy-nonenal (1:1000 dilution; Alpha Diagnostic, San Antonio, TX), anti–smooth muscle α-actin (1:1000 dilution; DAKO), anti-CD43 (1:1000 dilution; Serotec, Raleigh, NC), anti–hypoxia-inducible factor (HIF) 2α (1:1000 dilution; Novus Biologicals, Littleton, CO), anti–TGF-β1 (1:1000 dilution; Santa Cruz Biotechnology, Santa Cruz, CA), anti-phospho-c-Jun (1:1000 dilution; Santa Cruz Biotechnology), and anti-phospho-p42/p44 extracellular-regulated kinase (ERK) (1:1000 dilution; Cell Signaling, Beverly, MA). All specimens were incubated with an isotype-matched control antibody under identical conditions. Proliferating cell nuclear antigen (PCNA) expression was detected by immunostaining using monoclonal anti-PCNA primary antibody (DAKO) at a concentration of 50 μg/mL for 10 minutes. The area of positive staining was measured, as well as the number of PCNA positive cells, using a Macintosh-based morphometric analysis system (Apple Computer, Brea, CA) with MetaMorph software (Universal Imaging, Downingtown, PA).

Quantification of Hepatic Collagen Content.

Collagen content was assessed both by morphometrical analysis of Sirius red staining of liver sections and by hydroxyproline concentration. The area of positive Sirius red staining was measured using MetaMorph software. Hydroxyproline content was quantified colorimetrically from 0.2-g liver samples as previously described in detail.24 The results were expressed as micrograms of hydroxyproline per grams of liver.

ELISA for Interleukin 1β, Tumor Necrosis Factor α, and TGF β1.

Liver tissues (≈0.1 g) were homogenized as previously described.24 After centrifugation, cleared tissue lysates were collected and stored at −80°C. For in vitro studies, rat HSCs were cultured in 6-well plates at a density of 4 × 105 cells/well. Medium was removed and cells incubated in serum-free medium for 24 hours in the presence of agonists. Supernatants were collected and stored at −80°C until analysis. A sandwich ELISA for rat interleukin 1-β, tumor necrosis factor α, and TGF-β1 (R & D Systems, Minneapolis, MN) was performed.

Determination of Reactive Oxygen Species and [Ca2+]i in Cell Cultures.

Cells cultured in 24-well plates were loaded with the redox-sensitive dye 2′,7′-dichlorofluorescein diacetate or the Ca2+-sensitive dye Fluo-4 (Molecular Probes, Eugene, OR) (10 μmol/L) for 20 minutes at 37°C. Cells were then rinsed twice with Dulbecco's Modified Eagle Medium and stimulated with Ang II (10−8 mol/L) in the presence or absence of losartan (10−7 mol/L) (kindly provided by Merck, Rahway, NJ). Reactive oxygen species formation and [Ca2+]i were estimated according to changes in cell fluorescence using a multiwell fluorescence scanner (CytoFluor 2300; Millipore, Bedford, MA). Fluo-4–loaded cells were also plated onto glass-bottom microwell dishes (MatTek, Ashland, MA), and changes in cell fluorescence were visualized in a Zeiss LSM-510 confocal laser-scanning microscope at excitation and emission wavelengths of 488 nm and 520 nm, respectively (Carl Zeiss, Oberkochen, Germany).

DNA Synthesis.

DNA synthesis was estimated as the amount of [methyl-3H]thymidine (ICN Biomedicals, Irvine, CA) incorporated into trichloroacetic acid–precipitable material as previously described.20

Analysis of Tissue Factor Activity.

Tissue factor (factor III) is a key component of the extrinsic pathway of the coagulation cascade. Tissue factor activity was assessed via in vitro chromogenic assay (Actichrome TH; American Diagnostica, Greenwich, CT). Briefly, 0.1 g liver tissue or whole cell extracts were sonicated in a buffer of 30 mmol/L Tris-HCL (pH 7.4), 100 mmol/L NaCl, and 0.1% Triton X-100 for 30 minutes at 37°C and stored at −80°C until analysis. Ten micrograms of protein were incubated with factor VIIa to allow the formation of the factor VIIa/tissue factor complex. Samples were then incubated with factor X. The amount of factor Xa generated was measured by its ability to cleave spectrozyme Xa, a highly specific chromogenic substrate for factor Xa. The solution absorbance was read at 405 nm and compared with the values obtained from a standard curve generated using known amounts of active human tissue factor. Results are expressed as units per milligram of protein (liver tissue) or milliunits per well (cultured cells).

Statistical Analysis.

Results are expressed as the mean ± SD. For the determination of statistical significances, paired Student's t test was performed. Two-way ANOVA with Bonferroni's post hoc test was used for the determination of statistical significance between treatment groups, as appropriate. Differences were considered significant if the P value was less than .05.

Results

Ang II Infusion Induces Mild Arterial Hypertension and Increases Ang II Serum Levels.

Ang II infusion (25 ng/kg/min) was not associated with mortality and was well tolerated. Rats receiving Ang II showed a mild increase in arterial pressure (Fig. 1A). Changes in arterial pressure were similar in sham-operated and bile duct–ligated rats. Serum levels of Ang II were higher in bile duct–ligated rats than in sham-operated rats (Fig. 1B). Ang II infusion increased Ang II serum levels in both sham-operated and bile duct–ligated rats.

Figure 1.

(A) Effects of saline and Ang II infusion on arterial pressure in sham-operated and bile duct–ligated rats. *P < .05 versus rats infused with saline. (B) Effect of bile duct ligation and Ang II infusion on circulating Ang II levels. *P < .01 versus rats infused with saline. **P < .05 versus sham-operated rats infused with saline. (C-D) Effect of Ang II infusion on serum liver enzymes (C) and bilirubin levels (D). *P < .05 versus rats infused with saline. **P < .05 versus sham-operated rats infused with saline. (E) Effect of Ang II infusion on body/liver weight ratio. *P < .05 versus rats infused with saline. **P < .05 versus sham-operated rats infused with saline. Ang II, angiotensin II; BDL, bile duct ligation; ALT, alanine aminotransferase; AST, aspartate aminotransferase..

Ang II Infusion Augments Liver Injury in Bile Duct–Ligated Rats.

Ang II infusion increased aspartate aminotransferase and alanine aminotransferase serum levels in both sham-operated and bile duct–ligated rats, while bilirubin levels were not affected (Fig. 1C–D). The liver/body weight ratio increased in both sham-operated and bile duct–ligated rats receiving Ang II compared with saline (Fig. 1E). A small amount of ascitic fluid (less than 5 mL) was detected in 30% and 80% of bile duct–ligated rats receiving saline or Ang II, respectively (P < .05). Moreover, bile duct–ligated rats infused with saline showed higher serum endotoxin levels compared with sham-operated rats (12.4 ± 2.3 and 6.4 ± 1.5 pg/mL, respectively; P < .05). Ang II infusion increased serum endotoxin levels in both sham-operated and bile duct–ligated rats (11.9 ± 2.4 and 20.8 ± 4.7 pg/mL, respectively; P < .05 vs. saline).

Ang II Infusion Induces Inflammation, Vascular Thrombosis, and Oxidative Stress in Bile Duct–Ligated Rats.

Liver inflammation was assessed by identifiying CD43-positive infiltrating cells and by determining the tissue concentration of inflammatory cytokines. Ang II infusion into sham-operated rats induced a mild increase in the infiltration of CD43-positive cells around the central veins (data not shown). Ang II infusion into bile duct–ligated rats markedly increased the number of infiltrating CD43+ inflammatory cells around the proliferating bile ducts (5.6 ± 1.1 vs. 13.2 ± 2.1 positive cells per field in bile duct–ligated rats infused with saline and Ang II 25 ng/kg/min, respectively; P < .05) (Fig. 2A). Similarly, Ang II infusion increased the hepatic concentration of the inflammatory cytokines tumor necrosis factor α and interleukin 1β (Fig. 2B) and the accumulation of lipid peroxidation products (Fig. 2C). Ang II infusion increased the number of vascular thromboses in both sham-operated and bile duct–ligated rats compared with saline infusion (Fig. 3A–B). Next, to confirm the procoagulant effect of Ang II on the liver, we used an in vitro chromogenic assay to investigate the activity of tissue factor, a membrane glycoprotein that typically activates the extrinsic pathway of the coagulation cascade. Tissue factor procoagulant activity was higher in both sham-operated and bile duct–ligated livers infused with Ang II compared with rats infused with saline (Fig. 3C). Because of the vasoactive properties and vascular changes induced by Ang II, it is conceivable that Ang II infusion resulted in liver hypoxia. To address this question, the expression of the hypoxia-sensitive factor HIF-2α was assessed. Ang II infusion resulted in increased expression of HIF-2α in both sham-operated and bile duct–ligated rats (Fig. 3D). These results indicate that increased circulating Ang II induces inflammation and oxidative stress and is thrombogenic in the rat liver.

Figure 2.

(A) Ang II infusion markedly increased the number of CD43-positive cells around proliferating bile ducts (original magnification ×100). (B) Ang II infusion increased the hepatic concentration of tumor necrosis factor α and interleukin 1β in both sham-operated and bile duct–ligated rats. *P < .05 versus rats infused with saline. **P < .05 versus sham-operated rats infused with saline. (C) Detection of 4-hydroxy-nonenal protein adducts. Ang II infusion increased the areas with positive staining compared with rats infused with saline (original magnification ×100). BDL, bile duct ligation; Ang II, angiotensin II; TNFα, tumor necrosis factor α; IL1β, interleukin 1β.

Figure 3.

(A) Masson's Trichromic staining of liver specimens. Ang II infusion induced vessel wall thickening and thrombosis in the portal tracts of bile duct–ligated rats (arrow). (Original magnification ×100.) (B) Quantification of the number of small vessels with thrombosis in liver specimens. Fifteen high-power fields (×200) were studied in each liver sample. *P < .05 in rats receiving Ang II versus saline in both sham-operated and bile duct–ligated rats. **P < .05 versus sham-operated rats infused with Ang II. (C) Effects of Ang II infusion in tissue factor activity in liver tissue. *P < .05 versus rats infused with saline. **P < .05 versus sham-operated rats infused with saline. (D) Expression of HIF-2 in both sham-operated and bile duct–ligated rats, as assessed via Western blot analysis. The graph shows the relative expression of HIF-2 compared with β-actin expression of four independent experiments. *P < .05 versus bile duct–ligated rats infused with saline. BDL, bile duct ligation; Ang II, angiotensin II; HIF-2, hypoxia-induced factor 2; Sal, saline; IB, immunoblot.

Ang II Increases the Number of Bile Ducts in Bile Duct–Ligated Rats.

Bile duct–ligated rats infused with saline showed extensive bile duct proliferation, as assessed by proliferating cell nuclear antigen (PCNA) staining (Fig. 4A). Ang II infusion into bile duct–ligated rats increased the number of bile ducts as well as the number of PCNA-positive bile duct epithelial cells, suggesting that Ang II may increase cholangiocyte proliferation (Fig. 4B). Moreover, Ang II infusion stimulated phosphorylation of mitogen-activated protein kinase in bile duct epithelial cells. Increased detection of phospho-ERK and phospho-c-Jun was observed in bile duct–ligated livers infused with Ang II (Fig. 4C,E). Increased ERK phosphorylation was confirmed via Western blot analysis (Fig. 4D). Sham-operated rats infused with Ang II showed a minor increase in bile duct proliferation in 20% of portal tracts as well as a slight increase in PCNA-positive cells (not shown).

Figure 4.

Representative photomicrographs of proliferating bile ducts from saline- and Ang II–infused bile duct–ligated rats. (A) Paraffin-embedded sections were processed for PCNAs. Increased bile duct proliferation was observed in Ang II–treated bile duct–ligated rats compared with rats infused with saline. (B) Quantification of PCNA-postive cells per field in sham-operated and bile duct–ligated rats infused with saline or Ang II. *P < .05 versus rats infused with saline. **P < .05 versus sham-operated rats infused with saline. Ang II infusion stimulated the expression of (C) phospho-ERK and (E) phospho-c-jun in bile duct cells compared with rats infused with saline (original magnification ×200 for both). (D) Relative expression of phospho-ERK compared with β-actin expression of four independent Western blot experiments. *P < .05 versus bile duct–ligated rats infused with saline. PCNA, proliferating cell nuclear antigen; Ang II, angiotensin II; BDL, bile duct ligation; p-ERK, phospho–extracellular-regulated kinase; Sal, saline; IB, immunoblot.

Ang II Infusion Exacerbates Liver Fibrosis in Bile Duct–Ligated Rats.

Ang II infusion did not significantly increase collagen deposition in sham-operated rats, as assessed by Masson Trichrome and Sirius red staining and hepatic hydroxyproline content (data not shown). Only a mild increase in Sirius red staining around small vessels was observed. As expected, livers from bile duct–ligated rats infused with saline showed extensive areas of anti–smooth muscle α-actin staining, bridging fibrosis, and increased hydroxyproline content compared with sham-operated rats (Fig. 5A–C,E). Ang II infusion into bile duct–ligated rats significantly increased the extent of anti–smooth muscle α-actin–positive staining as well as collagen content in the liver. Interestingly, the profibrogenic effect of Ang II infusion was associated with an increase in TGF-β1 hepatic concentration as well as increased immunostaining in peribiliary areas (Fig. 5D,F). Collectively, these results suggest that infusion of Ang II into bile duct–ligated rats markedly increases collagen deposition.

Figure 5.

Ang II infusion increased fibrosis development in bile duct–ligated rats. Representative photomicrographs of livers from bile duct–ligated rats treated with saline and Ang II (25 ng/kg/min) are shown. Paraffin-embedded sections were processed for (A) Masson trichromic and (B) Sirius red staining. Extensive bile duct proliferation and peribiliary fibrosis with bridging fibrosis was observed (original magnifications ×100 and ×400, respectively). Ang II infusion increased the accumulation of (C) smooth muscle α-actin–positive cells and (D) TGF-β1 staining around proliferating bile ducts and fibrous septa compared with rats infused with saline (original magnifications ×40 and ×200, respectively). (E) Collagen quantification in liver tissue, as assessed by hepatic hydroxyproline content. *P < .01 versus sham-operated rats infused with saline. **P < .05 versus bile duct–ligated rats infused with saline. (F) Quantification of TGF-β1 in hepatic tissue by ELISA. *P < .01 versus sham-operated rats infused with saline. **P < .05 versus bile duct–ligated rats infused with saline. BDL, bile duct ligation; Ang II, angiotensin II; TGFβ1, transforming growth factor β1.

Effects of Ang II on Cultured Rat HSCs.

We investigated whether the effects of Ang II are reproduced at the cellular level. We studied primary rat HSCs, a cell type that plays a major role in hepatic wound healing. We have previously demonstrated that Ang II stimulates type I collagen and TGF-β1 secretion through cultured rat HSCs.24 We explored whether Ang II induces other biological effects in these cells. Ang II (10−8 mol/L) increased [Ca2+]i and stimulated DNA synthesis (Fig. 6A–C). The AT1 antagonist losartan (10−7 mol/L), but not the AT2 antagonist PD123319 (10−7 mol/L), blocked Ang II–induced effects, indicating that rat HSCs express functional AT1 receptors. We next explored whether Ang II induces pro-oxidant and proinflammatory actions in HSCs. Stimulation of cells with Ang II (10−8 mol/L) induced a marked increase in reactive oxygen species formation and stimulated the secretion of tumor necrosis factor α into the culture media (Figs. 6D, 7A). These effects were blocked by losartan (10−7 mol/L), while the AT2 antagonist PD123319 (10−7 mol/L) had no effect. Finally, we investigated whether Ang II stimulates the procoagulant activity of rat HSCs. Extracts from freshly isolated quiescent rat HSCs did not induce activation of factor VII or X in an in vitro assay, indicating that quiescent HSCs do not have procoagulant properties. However, following HSC activation in culture (5 d after isolation), HSCs markedly stimulated factors VII and X, indicating that they have functional tissue factor activity (Fig. 7B). Incubation of culture-activated rat HSCs with Ang II (10−8 mol/L) for 12 hours significantly increased tissue factor activity (Fig. 7C). This effect was blocked by losartan (10−8 mol/L). Collectively, these results indicate that Ang II induces pro-oxidant, proinflammatory, and procoagulant effects in rat HSCs.

Figure 6.

Biological effects of Ang II (10−8 mol/L) on rat cultured HSCs. (A) Effect of Ang II on [Ca2+]i. Ang II induced a marked [Ca2+]i increase, which consists of a Ca2+ peak followed by a sustained [Ca2+]i increase. Incubation of cells with the AT1 receptor antagonist losartan (10−7 mol/L), but not the AT2 receptor antagonist PD123319 (10−7 mol/L), prevented both phases. The graph is representative of four different experiments. (B) Confocal microscopy images showing changes in cell fluorescence in rat HSCs loaded with Fluo-4. Ang II caused a rapid increase in cell fluorescence, indicative of [Ca2+]i increase. (C) Ang II increased cell growth, as assessed by [3H]thymidine incorporation. Experiments were done in triplicate four times. *P < .01 versus buffer. **P < .05 versus Ang II. (D) Effect of Ang II on reactive oxygen species in rat HSCs. Cells were incubated with 2′,7′-dichlorofluorescein diacetate (10 μmol/L) and studied with a fluorometer at 485/535 nm. The graph is representative of 4 different experiments. Ang II, angiotensin II.

Figure 7.

(A) Effect of Ang II on cytokine secretion by rat HSCs. Cells were incubated with Ang II for 24 hours in the presence or absence of losartan and PD123319. *P < .05 versus buffer. The graph shows the mean ± SD of four independent experiments done in triplicate. (B) Effect of Ang II on tissue factor activity, as assessed via an in vitro chromogenic assay. Activated but not quiescent HSCs showed tissue factor activity. Stimulation with Ang II (10−7 mol/L) increased tissue factor activity in activated HSCs. *P < .05 versus quiescent HSCs. **P < .05 versus activated HSCs incubated with buffer. (C) Ang II increased tissue factor activity in rat HSCs in a time-dependent manner. Incubation of cells with losartan but not PD123319 prevented this effect. The figure is representative of three independent experiments. TNFα, tumor necrosis factor α; Ang II, angiotensin II.

Discussion

The current study demonstrates that increased circulatory Ang II levels accelerate inflammation and fibrosis in rats subjected to bile duct ligation. Among the potential underlying mechanisms, we show that Ang II increases oxidative stress, favors procoagulant events, and stimulates bile duct proliferation. Moreover, Ang II infusion stimulates the activation of mitogen-activated protein kinase in the liver. Because patients with advanced chronic liver disease are characterized by activation of the circulatory RAS,10 systemic Ang II may contribute to the progression of liver fibrosis. Our results are consistent with those of previous studies showing that increased systemic Ang II causes oxidative stress and profound inflammation in the vasculature, heart, and kidney.25–27 In a previous study, we demonstrated that prolonged systemic infusion of Ang II into normal rats induces hepatic inflammation and HSC activation.22 These effects were associated with increased oxidative stress and vascular thrombosis. Although these pathogenic effects can theoretically lead to fibrogenesis, Ang II infusion alone was not sufficient to induce parenchymal liver fibrosis in normal rats. Based on this previous study, we hypothesized that increased systemic Ang II could exacerbate fibrosis in livers undergoing tissue remodeling from other causes (i.e., biliary obstruction). The results of the current study support this assumption.

We used a model of secondary biliary fibrosis that reproduces the fibrogenic response of patients with chronic cholestatic disorders.28 Ligation of the common duct results in intense bile duct proliferation, inflammatory changes, and peribiliary fibrosis. A role of the RAS in this experimental model is suggested by studies showing that intrahepatic RAS is markedly upregulated,12 and RAS inhibition attenuates progression of fibrosis.15, 16 Besides upregulation of the intrahepatic RAS, we found that the circulatory levels of Ang II increased in bile duct–ligated rats, suggesting an activation of the systemic RAS. Although circulatory parameters were not investigated in this study, the finding that 30% of bile duct–ligated rats receiving saline developed small amounts of ascites suggests the development of some degree of circulatory dysfunction (i.e., activation of endogenous vasoconstrictor systems). To confirm that the RAS participates in the pathogenesis of liver fibrosis and to explore the mechanisms involved, we studied whether a further increase in circulatory Ang II levels modulates development of fibrosis. Circulatory Ang II levels were increased by means of a subcutaneous pump. This method has been widely used to evaluate the fibrogenic effect of systemic RAS in other organs such as the kidney.25, 27 We showed that increased Ang II circulatory levels markedly exacerbate development of liver fibrosis in bile duct–ligated rats, but not in sham-operated rats. This result suggests that systemic RAS may play a role in the progression of liver fibrosis. Both increased synthesis and decreased collagen degradation may be involved.29 Although it is conceivable that the intrahepatic RAS may play a similar role, we have not specifically addressed its role. Moreover, it is possible that a cross-communication between the systemic and intrahepatic RAS occurs. Additional studies should address the role of intrahepatic RAS in liver fibrosis.

Several mechanisms can mediate the fibrogenic effects of Ang II infusion in bile duct–ligated livers. Both systemic and local effects are potentially involved. First, Ang II infusion increased serum endotoxin levels, which can induce a systemic inflammatory response involving the liver. Ang II perfusion could also promote vasoconstriction of the hepatic vessels and decreased hepatic blood flow. Although hepatic hemodynamics were not directly assessed in this study, the hypoxia-sensitive protein HIF-2α was upregulated in the livers from rats infused with Ang II, suggesting that some hepatic hypoperfusion occurred. Moreover, Ang II infusion caused inflammation and sclerosis of small vessels, in addition to the promotion of thrombosis, which could contribute to impaired liver oxygenation. Liver hypoxia has been reported to contribute to the progression of liver fibrosis.30 Second, the procoagulant action of Ang II can induce pathogenic consequences in the liver. In our study, Ang II induced procoagulant effects in vivo and in vitro, indicating a thrombogenic role for Ang II in the liver. Besides its effect in oxygen supply, the activation of the coagulation cascade can promote inflammation and modulate extracellular matrix deposition.31 In fact, a recent study indicates that the existence of thrombotic factors favors the progression of liver fibrosis in patients with chronic hepatitis C.32 Third, oxidative stress may be implicated in the fibrogenic effect of Ang II. Increased lipid peroxidation protein adducts were found in livers from bile duct–ligated rats infused with Ang II. In other organs, oxidative stress mediates the pathogenic effect of Ang II.2 In the liver, oxidative stress stimulates collagen synthesis through HSCs, and antioxidants attenuate the development of fibrosis.33 Ang II perfusion also increased the proliferation of cholangiocytes and stimulated mitogen-activated protein kinase in these cells. It is unknown whether Ang II directly interacts with bile duct cells or if these effects are due to indirect causes (i.e., oxidative stress). In cholestasis-induced liver fibrosis, proliferating bile duct cells secrete growth factors and profibrogenic cytokines that promote the accumulation of myofibroblasts.34 Further studies should address whether Ang II modulates these actions. Finally, it is uncertain whether mild arterial hypertension induced by Ang II infusion contributed to increased bile duct ligation–induced liver injury. Ang II–induced inflammatory effects in normal rat livers do not depend on the occurrence of arterial hypertension.23 Moreover, contrary to what has been described in the kidney or the heart, there are no experimental or clinical data directly demonstrating that arterial hypertension induces liver fibrosis. On the contrary, rats with arterial hypertension do not spontaneously develop liver fibrosis.35

Using cultured HSCs, we confirmed the biological effects induced by Ang II in vivo. We investigated the proinflammatory and profibrogenic effects of Ang II on cultured rat HSCs. We have demonstrated that rat HSCs bear functional AT1 receptors that modulate cell activation, collagen synthesis, and TGF-β expression.22, 24 In the current study, we expand these observations by showing that Ang II stimulates reactive oxygen species formation and induces the secretion of proinflammatory cytokines. Moreover, Ang II increases the procoagulant function of these cells. This is consistent with reports showing that Ang II increases the procoagulant activity of cultured endothelial cells.36 We provide evidence that rat HSCs express tissue factor and are able to activate the extrinsic pathway of the coagulation cascade. This novel function for HSCs could contribute to their pathogenic role in chronic liver disease. Although it is known that HSCs bear receptors for thrombin,37 a protein involved in coagulation, this is the first study specifically addressing the procoagulant activity of cultured HSCs. Because HSCs migrate and accumulate to the areas of active tissue repair, and microthrombi are frequently found in chronically-injured livers, it is likely that HSCs could play a role in the formation of thrombosis. Whether other nonparenchymal cells (i.e., Kupffer cells, endothelial sinusoidal cells) contribute to the biological effects of Ang II in the liver is unknown and deserves additional study.

Several of the effects observed after Ang II infusion could contribute to the progression of liver disease. First, oxidative stress plays an important role in the hepatic fibrogenic response to different etiological agents.32 Increased reactive oxygen species and resulting lipid peroxidation products are commonly detected in livers from patients with alcohol abuse, hepatitis C virus infection, iron overload or chronic cholestasis, as well as in most types of experimental liver fibrogenesis.30, 38, 39 Moreover, antioxidant agents attenuate the development of hepatic fibrosis in rodents and exert beneficial effects in patients with chronic liver disease.40 Second, thrombosis of small vessels is frequently found in chronic liver disease, including alcoholic hepatitis, hepatic rejection, and cirrhosis. Thrombosis phenomena can impair liver perfusion and have been implicated in the pathogenesis of portal hypertension and progression of liver fibrosis.31 Potential mechanisms include impaired liver oxygenation and inflammatory effects of coagulation proteins. Therefore, the procoagulant effect of Ang II could induce liver damage. Finally, although we did not assess hepatic hemodynamics, increased systemic Ang II could worsen portal hypertension by increasing the intrahepatic resistance to blood flow or by stimulating the release of aldosterone. The finding that Ang II infusion stimulates ascites formation in bile duct–ligated rats supports this hypothesis. Ang II infusion increased the number of contractile myofibroblasts, and Ang II is known to induce contraction of HSCs. Moreover, oxidative stress mediates the contractile effect of Ang II in different vascular beds and may also play a role in the liver.41

Systemic RAS activation is only observed in patients with advanced cirrhosis.10, 11 In this phase of the disease, hepatic fibrosis is massive, and it is conceivable that the activation of systemic RAS does not further increase the degree of hepatic fibrosis. Instead, increased systemic Ang II may accelerate the progression of liver disease through mechanisms other than progression of fibrosis, such as promoting vascular thrombosis or increasing circulatory LPS levels.

In summary, we provide evidence that systemic infusion of Ang II in rats undergoing hepatic remodeling exacerbates inflammation and development of fibrosis. Moreover, Ang II exerts similar effects on rat primary HSCs. These results are consistent with an antifibrotic effect of RAS inhibitors in liver fibrosis.

Acknowledgements

The authors thank Charlotte Walters for ELISA measurements and Jennyfer Dulyx for help with histological studies.

Ancillary