Increased oxidative stress in cirrhotic rat livers: A potential mechanism contributing to reduced nitric oxide bioavailability

Authors

  • Jorge Gracia-Sancho,

    1. Hepatic Hemodynamic Laboratory, Liver Unit, Institut de Malalties Digestives i Metabòliques (IMDiM), Hospital Clínic, Institut d'Investigacions Biomèdiques August Pi i Sunyer and Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas, University of Barcelona, Spain
    Search for more papers by this author
    • These authors contributed equally to this study.

  • Bàrbara Laviña,

    1. Hepatic Hemodynamic Laboratory, Liver Unit, Institut de Malalties Digestives i Metabòliques (IMDiM), Hospital Clínic, Institut d'Investigacions Biomèdiques August Pi i Sunyer and Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas, University of Barcelona, Spain
    Search for more papers by this author
    • These authors contributed equally to this study.

  • Aina Rodríguez-Vilarrupla,

    1. Hepatic Hemodynamic Laboratory, Liver Unit, Institut de Malalties Digestives i Metabòliques (IMDiM), Hospital Clínic, Institut d'Investigacions Biomèdiques August Pi i Sunyer and Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas, University of Barcelona, Spain
    Search for more papers by this author
  • Héctor García-Calderó,

    1. Hepatic Hemodynamic Laboratory, Liver Unit, Institut de Malalties Digestives i Metabòliques (IMDiM), Hospital Clínic, Institut d'Investigacions Biomèdiques August Pi i Sunyer and Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas, University of Barcelona, Spain
    Search for more papers by this author
  • Mercedes Fernández,

    1. Hepatic Hemodynamic Laboratory, Liver Unit, Institut de Malalties Digestives i Metabòliques (IMDiM), Hospital Clínic, Institut d'Investigacions Biomèdiques August Pi i Sunyer and Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas, University of Barcelona, Spain
    Search for more papers by this author
  • Jaume Bosch,

    1. Hepatic Hemodynamic Laboratory, Liver Unit, Institut de Malalties Digestives i Metabòliques (IMDiM), Hospital Clínic, Institut d'Investigacions Biomèdiques August Pi i Sunyer and Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas, University of Barcelona, Spain
    Search for more papers by this author
  • Joan-Carles García-Pagán

    Corresponding author
    1. Hepatic Hemodynamic Laboratory, Liver Unit, Institut de Malalties Digestives i Metabòliques (IMDiM), Hospital Clínic, Institut d'Investigacions Biomèdiques August Pi i Sunyer and Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas, University of Barcelona, Spain
    • Hepatic Hemodynamic Laboratory, Liver Unit, Hospital Clínic, Villarroel 170, 08036 Barcelona, Spain
    Search for more papers by this author
    • fax: (34) 93-2279856.


  • Potential conflict of interest: Nothing to report.

Abstract

In cirrhotic livers, decreased nitric oxide (NO) bioavailability is a major factor increasing intrahepatic vascular tone. In several vascular disorders, an increase in superoxide (O2) has been shown to contribute to reduced NO bioavailability through its reaction with NO to form peroxynitrite. This study was aimed to test the hypothesis that, in cirrhotic livers, increased O2, by reacting with NO, reduces NO bioavailability. In control and cirrhotic rat livers, NO bioavailability was evaluated by the measurement of cyclic guanosine monophosphate in liver tissue and by 4-amino-5-methylamino-2′,7′-difluorofluorescein diacetate (DAF-FM-DA) fluorescence in isolated sinusoidal endothelial cells (SEC); the O2 content was determined by dihydroethidium staining in fresh liver sections. In addition, the role of endothelial nitric oxide synthase (eNOS), xanthine oxidase (XO), and cyclooxygenase (COX) as possible sources of O2 and the role of superoxide dismutase (SOD) enzymatic activity as an O2 scavenger were determined in liver homogenates. Protein-nitrotyrosination, a marker of the NO-O2 reaction, was evaluated in liver homogenates. Furthermore, in control SEC and bovine aortic endothelial cells, NO modulation by O2 was evaluated. Cirrhotic livers exhibited increased O2 levels. This was due, at least in part, to increased production by COX and XO but not eNOS and to reduced scavenging by SOD. Increased O2 was associated with a significant reduction in NO bioavailability and increased nitrotyrosinated proteins. In endothelial cells, an inverse relationship between O2 levels and NO bioavailability was observed. Conclusion: Our data show that oxidative stress may contribute to reduced NO bioavailability in cirrhotic livers, supporting the evaluation of O2 reduction as a potential mechanism to restore NO content. (HEPATOLOGY 2008.)

Increased resistance to portal blood flow is the primary factor in the pathophysiology of portal hypertension, the main complication of cirrhosis.1, 2 Architectural alterations of the liver parenchyma as well as a dynamic increase in the hepatic vascular tone contribute to the increased resistance to portal blood flow of cirrhotic livers. The increase in hepatic vascular tone is partly due to increased release of cyclooxygenase-1 (COX-1)–derived vasoconstrictive prostanoids3, 4 and to reduced bioavailability within the liver of the potent vasodilator nitric oxide (NO),5–7 both mechanisms favoring the contraction of different elements within the cirrhotic liver. Indeed, activated hepatic stellate cells have been shown to contract or relax in response to vasoconstrictive prostanoids or NO, respectively.8–10

Reduced NO bioavailability has been attributed to decreased endothelial nitric oxide synthase (eNOS) activity11 secondary to several disturbances in the posttranslational regulation of the enzyme.12–15

In several vascular disorders, the potential of reactive oxygen species to bind proteins, break DNA, and promote cell damage by reacting with several cellular components has been involved in the development of necrosis, inflammation, and apoptosis.16–18 In addition, an increase in the reactive oxygen specie superoxide (O2), by a rapid reaction with NO,19 promotes a marked reduction in NO bioavailability followed by an increase in vascular tone.20–23

An increase in O2 levels due to increased production by xanthine oxidase (XO), nicotinamide adenine dinucleotide phosphate (NADPH) oxidase, or COX, among other enzymatic systems, and reduced O2 scavenging by superoxide dismutase (SOD) has been suggested to play a pathophysiological role in different liver disorders such as alcoholic and nonalcoholic liver disease.24–28

We hypothesize that in cirrhotic livers, increased O2, by reacting with NO, would contribute to reduced NO bioavailability. As a result, removing O2 from cirrhotic livers could be a new therapeutic strategy to improve intrahepatic NO bioavailability. The present study was designed to be a proof of concept of this hypothesis.

Abbreviations

3-NT, nitrotyrosine; Ad, adenoviral; Allop, allopurinol; βgal, β-galactosidase; BAEC, bovine aortic endothelial cells; CCl4, carbon tetrachloride; cGMP, cyclic guanosine monophosphate; CH, cirrhotic; COX, cyclooxygenase; CT, control; CuZnSOD, cytoplasmatic superoxide dismutase; DAF-FM-DA, 4-amino-5-methylamino-2′,7′-difluorofluorescein diacetate; DDC, diethyldithiocarbamate; DHE, dihydroethidium; EcSOD, extracellular superoxide dismutase; eNOS, endothelial nitric oxide synthase; Indo, indomethacin; L-NAME, N-nitro-L-arginine methyl ester; MnSOD, mitochondrial superoxide dismutase; NADPH, nicotinamide adenine dinucleotide phosphate; NBT, nitro blue tetrazolium; NO, nitric oxide; NOS, nitric oxide synthase; O2, superoxide; PQ, paraquat; SEC, sinusoidal endothelial cells; SOD, superoxide dismutase; Veh, vehicle; XO, xanthine oxidase.

Materials and Methods

Induction of Cirrhosis by Carbon Tetrachloride (CCl4)

Male Wistar rats weighing 175 to 200 g underwent inhalation exposure to CCl4. Phenobarbital (0.3 g/L) was added to the drinking water as previously described.3 A high yield of micronodular cirrhosis was obtained after approximately 12 to 15 weeks of CCl4 inhalation. When the cirrhotic rats developed ascites, administration of phenobarbital was stopped, and the subsequent experiments were performed 1 week later. Control animals received only phenobarbital. The animals were kept in environmentally controlled animal facilities at the Institut d'Investigacions Biomèdiques August Pi i Sunyer. All experiments were approved by the Laboratory Animal Care and Use Committee of the University of Barcelona and were conducted in accordance with Guide for the Care and Use of Laboratory Animals (National Institutes of Health, NIH Publication 86-23, revised 1985).

Isolation and Culture of Liver Sinusoidal Endothelial Cells (SEC)

SEC were isolated from control and cirrhotic rat livers as described previously.29, 30 Briefly, after collagenase perfusion of the livers and isopycnic sedimentation of the resulting dispersed cells through a two-step density gradient of Percoll, pure monolayer cultures of SEC were established by selective attachment on a substrate of fibronectin. Afterwards, cells were cultured for 12 hours (37°C, 5% CO2) in Roswell Park Memorial Institute (RPMI) 1640 as previously described.30 Almost 93% of these cells were SEC, as assessed by specific immunocytochemical marking using rat endothelial cell antigen-131 (Fig. 1), and they had a viability of 95% (by trypan blue exclusion). All studies were performed on cells from the first passage, 12 hours after SEC isolation, to preserve their typical phenotype.32

Figure 1.

Representative immunocytochemical image of primary cultured rat sinusoidal endothelial cells (SEC). High-purity SEC (>93%) were calculated as the number of cells expressing rat endothelial cell antigen-1–positive staining (red) divided by the total number of cells (stained with nuclear Hoescht dye in blue). Original magnification, 20×.

Evaluation of NO Bioavailability in Liver Tissue and SEC of Cirrhotic and Control Rats

Measurement of NO Levels in SEC.

In situ NO levels in SEC were assessed with 4-amino-5-methylamino-2′,7′-difluorofluorescein diacetate (DAF-FM-DA; Molecular Probes, Inc., Eugene, OR) as described.33, 34 DAF-FM-DA is a cell-permeable, NO-sensitive dye that is virtually nonfluorescent until it reacts with NO to form benzotrizole. Isolated SEC were washed in RPMI 1640 without phenol red and loaded with DAF-FM-DA (10 μM final concentration, 20 minutes, 37°C). Then, SEC were rinsed three times with phosphate-buffered saline, kept in the dark, and maintained at 37°C with a warm stage on a laser scanning confocal microscope (model TCS-SL DMIRE2, Leica, Wetzlar, Germany), and images were obtained with a 488-nm (excitation) and 505- to 530-nm (emission) filter set for DAF-FM-DA, with a 40 × 1.3 oil objective. Quantitative analysis was obtained by averaging of the peak relative fluorescent intensity (optical density arbitrary units) of each confocal microscope image (Image J 1.33u software, National Institutes of Health)35 and normalization of the fluorescent result by the total number of cultured cells counted from each corresponding digitalized phase contrast microscope image. In some experiments, the nitric oxide synthase (NOS) inhibitor N-nitro-L-arginine methyl ester (L-NAME; 1.5 mM) as a negative control or the NO donor diethylenetriamene NONOate (DETANONOate) (500 μM) and the eNOS inductor vascular endothelial growth factor as positive controls were added 20 minutes before the loading of DAF-FM-DA.

Cyclic Guanosine Monophosphate (cGMP) Levels.

Measurements of cGMP, a marker of NO bioavailability, were performed in liver homogenates.36 Briefly, equal amounts of liver tissue (200 mg) were dropped into 10 volumes of 5% trichloroacetic acid and homogenized at 4°C. The precipitate was removed by centrifugation at 2000g for 15 minutes at 4°C. The supernatant was transferred to a clean test tube, washed with water-saturated diethyl ether three times, and lyophilized. The dried extract was dissolved in assay buffer, and cGMP levels were determined by enzyme immunoassay (Cayman Chemical Co., Ann Arbor, MI). Results were expressed as picomoles per milliliter.

eNOS Protein Detection.

SEC isolated from control and cirrhotic rat livers were homogenized in triton-lysis buffer as previously described.37 Aliquots from each sample containing equal amounts of protein (20 μg) were run on an 8% sodium dodecyl sulfate–polyacrylamide gel and transferred to a nitrocellulose membrane. After the transfer, the blots were subsequently blocked for 1 hour with trishydroxymethylaminomethane-buffered saline containing 0.05% (vol/vol) Tween 20 and 5% (wt/vol) nonfat dry milk and were probed with a mouse anti-eNOS (1 μg/mL) antibody (BD Biosciences, San Jose, CA) overnight at 4°C followed by incubation with rabbit anti-mouse horseradish peroxidase–conjugated secondary antibody (1:10000, 1 hour, room temperature; Stressgen, Victoria, British Columbia, Canada).

Protein expression was determined by densitometric analysis with the Science Lab Image Gauge (Fuji Photo Film GmbH, Düsseldorf, Germany). After stripping, blots were assayed for glyceraldehyde 3-phosphate dehydrogenase (GAPDH; Santa Cruz Biotechnology, Santa Cruz, CA) expression as a standardization of the sample loading. Quantitative densitometric values of all proteins were normalized to GAPDH.

Evaluation of O2

Measurement of the O2 Content in the Liver Tissue of Cirrhotic and Control Rats.

In situ O2 levels were evaluated with the oxidative fluorescent dye dihydroethidium (DHE; Molecular Probes).38 DHE specifically reacts with intracellular O2 and is converted to the red fluorescent compound ethidium bromide, which then binds irreversibly to double-stranded DNA and appears as punctuate nuclear staining.39 Ethidium bromide is excited at 488 nm with an emission spectrum of 610 nm. Liver cryosections (10 μm) were incubated with DHE (10 μmol/L) in phosphate-buffered saline. In order to demonstrate the specificity of the assay for O2, parallel incubations with SOD (200 U; Applichem, Darnstadt, Germany) were performed. Fluorescence images were obtained with a laser scanning confocal microscope (TCS-SL DMIRE2, Leica), and quantitative analysis was performed with Image J 1.33u software (National Institutes of Health).

O2 Generation.

To determine which proteins are responsible for O2 generation, cirrhotic rats (n = 3 per group) were pretreated with the specific XO inhibitor allopurinol (50 mg/kg intraperitoneally, 18 hours and 1 hour before the experiment), with the COX inhibitor indomethacin (20 mg/kg orally, 5 hours before the experiment), with the NOS inhibitor L-NAME (15 mg/kg intravenously, 15 minutes before the experiment), or with vehicle (saline), and the hepatic O2 content was evaluated in fresh liver sections by DHE staining as described previously. The efficacy of these doses inhibiting their respective enzymatic systems has been previously demonstrated.36, 40, 41

SOD Activity Assay.

Total SOD activity was measured in liver homogenates by an indirect assay previously described.42 The nitro blue tetrazolium (NBT) method is based on the competition reaction between SOD and the indicator molecule NBT. In brief, NBT is reduced to formazan by O2. The production of formazan is photometrically quantified at 560 nm. Dilutions of SOD were used to generate a standard curve. One unit of SOD activity is defined as the amount of protein required to give half-maximal inhibition of NBT reduction. SOD activity was expressed as units per milligram of total tissue.

SOD Isoform Protein Detection.

Frozen liver samples from control and cirrhotic rats were crushed to powder and subsequently homogenized in triton-lysis buffer as previously described.36 SOD isoforms were determined with antibodies against cytoplasmatic superoxide dismutase (CuZnSOD; Stressgen), mitochondrial superoxide dismutase (MnSOD; Upstate Biotechnology, Lake Placid, NY), and extracellular superoxide dismutase (EcSOD; Stressgen) by western blotting as described previously.

NO-O2 Interaction

Nitrotyrosine Protein Detection.

Protein nitrotyrosination, a marker of peroxynitrite production, was determined by western blotting with a mouse anti-nitrotyrosine (1 μg/mL) antibody (Cayman Chemical Co.) as described previously.

NO Modulation by O2.

SEC isolated from control rat livers were incubated for 6 hours with vehicle or with the CuZnSOD inhibitor diethyldithiocarbamate (DDC; 25 μM)43 alone or in association with the O2 scavenging enzyme SOD (300 U). Then, O2 and NO levels were evaluated in different subsets of SEC by DHE (10 μmol/L) and DAF-FM-DA staining, respectively.

In order to further strengthen data on NO modulation by O2, a complementary molecular approach was designed: Low-passage (n < 6) bovine aortic endothelial cells (BAEC) were seeded into p35 plates and infected in duplicate for 16 hours at a multiplicity of infection of 100 with either the adenoviral (Ad) construct codifying for the antioxidant enzyme SOD (the extracellular isoform EcSOD; constructed by Dr. Chu and provided by the Vector Core at the University of Iowa) or the control adenovirus β-galactosidase (βgal; kindly provided by Dr. C. B. Newgard of Duke University, Durham, NC) as previously described.44 These EcSOD adenoviruses are capable of infecting and expressing enzymatically active SOD in endothelial cells.45 After 48 hours, the medium was replaced, and endothelial cells were incubated for 6 hours with the oxidative stress inductor paraquat (0.01 mM).46 Then, O2 and NO levels were evaluated as described previously.

Drugs and Reagents

Mouse anti–rat endothelial cell antigen 1 monoclonal antibody was obtained from Serotec (Oxford, United Kingdom). Collagenase was from Roche Diagnostics (Mannheim, Germany). Percoll was from Amersham Biosciences (Uppsala, Sweden). Reagents for cell culture were provided by Biological Industries, Ltd. (Kibbutz Beit Haemek, Israel). L-NAME, paraquat, and other chemical reagents were purchased from Sigma (Tres Cantos, Madrid, Spain).

Statistical Analysis

Statistical analysis was performed with the SPSS 12.0 for Windows statistical package (SPSS, Inc., Chicago, IL). All results are expressed as mean ± standard error of the mean. Comparisons between groups were performed with the Student t test or Mann-Whitney t test for unpaired data when adequate. Differences were considered significant at a P value < 0.05.

Results

NO Bioavailability Is Reduced in SEC from Cirrhotic Livers

SEC isolated from cirrhotic rat livers had decreased NO bioavailability in comparison with SEC isolated from control livers as shown by the significant and marked reduction in DAF-FM-DA fluorescence (40.1% ± 8% versus 100% ± 24% in SEC from control rat livers; P < 0.01; Fig. 2). However, no differences in eNOS protein expression were observed in SEC isolated from cirrhotic or control rat livers (Fig. 2).

Figure 2.

Top: Fluorescent detection of intracellular nitric oxide (NO) in sinusoidal endothelial cells (SEC) isolated from control (CT) and cirrhotic (CH) rat livers. Representative images of 4-amino-5-methylamino-2′,7′-difluorofluorescein diacetate (DAF-FM-DA) fluorescence from CT SEC and CH SEC, including an area of the corresponding phase contrast images, were visualized and collected with confocal laser scanning microscopy. Original magnification, 40×; bar, 10 μm. The fluorescence intensity of DAF-FM-DA in arbitrary units was normalized by the total number of counted cells, indicating NO bioavailability. The data shown are from 1105 individual CT SEC and 1245 CH SEC obtained from three different experiments. The DAF-FM-DA fluorescence intensity was significantly lower in SEC isolated from CH rats (*P < 0.01 versus CT). Bottom left: Representative western blot and analysis of endothelial nitric oxide synthase (eNOS) protein expression in SEC isolated from CT (n = 4) and CH (n = 4) rat livers. Densitometry quantification in arbitrary units, normalized to glyceraldehyde 3-phosphate dehydrogenase (GAPDH), showed no differences between both groups. Bottom right: Cyclic guanosine monophosphate (cGMP) levels in liver homogenates from CT (n = 5) and CH (n = 5) rats. cGMP levels were significantly reduced in CH rat livers (*P = 0.05 versus CT).

Almost no DAF-FM-DA fluorescence was observed in SEC pretreated with the NOS inhibitor L-NAME (65% reduction versus those pretreated with vehicle; P < 0.01), and a strong cytoplasmic fluorescence was observed in SEC isolated from both control and cirrhotic rats pretreated with the NO donor DETANONOate (10-fold increase versus those pretreated with vehicle; P < 0.01). Similarly, BAEC incubated with vascular endothelial growth factor showed a significant increase in DAF-FM-DA fluorescence (data not shown).

Reduced NO bioavailability within the cirrhotic liver was further confirmed by the measurement of hepatic cGMP levels, a surrogate marker of NO bioavailability. Indeed, hepatic cGMP levels were significantly lower in cirrhotic rat livers than in control rat livers (2.5 ± 0.2 versus 5.0 ± 1.1 pmol/mL; P < 0.05; Fig. 2).

O2 Is Increased in Cirrhotic Livers Because of Overproduction by COX and XO and Reduced Degradation by SOD

Along with reduced NO bioavailability, cirrhotic livers exhibited an increase in O2 content. Thus, confocal microscopy showed a marked increase in DHE fluorescence in tissue sections from cirrhotic rat livers in comparison with control rat livers (Fig. 3). Cirrhotic livers preincubated with SOD, the enzyme that metabolizes O2 to H2O2, had a marked attenuation in DHE fluorescence (66.4% ± 0.6% reduction versus cirrhotic livers preincubated with vehicle), demonstrating the specificity of the assay for O2.

Figure 3.

Top: Representative confocal microscopy images of in situ detection of superoxide (O2) using the oxidative dye dihydroethidium (bar, 10 μm), in fresh liver sections from control rats (n = 3) and cirrhotic rats treated with vehicle (Veh; n = 3), with the nitric oxide synthase inhibitor N-nitro-L-arginine methyl ester (L-NAME; n = 3), with the xanthine oxidase inhibitor allopurinol (Allop; n = 3), or with the cyclooxygenase inhibitor indomethacin (Indo; n = 3). Bottom: Fluorescence intensity analysis showed increased O2 levels in cirrhotic livers in comparison with controls. A significant reduction in the cirrhotic intrahepatic O2 content was observed when xanthine oxidase or cyclooxygenase was selectively inhibited (values represent arbitrary units normalized to control livers; *P < 0.01 versus control; #P < 0.05 versus cirrhotic-Veh).

Specific inhibition of XO or COX resulted in a significant reduction in cirrhotic hepatic O2 levels in comparison with those animals receiving vehicle. However, no differences were observed when animals were treated with the NOS inhibitor L-NAME (Fig. 3).

Total SOD activity was significantly reduced in cirrhotic rat livers (n = 10) compared with control rat livers (n = 8, 2.1 ± 0.2 versus 4.2 ± 0.4 U/mg of tissue; P < 0.01; Fig. 4). This was associated with a significant reduction in CuZnSOD and MnSOD protein expressions (Fig. 4), without significant changes in EcSOD expression (Fig. 4).

Figure 4.

(A) Representative western blots and analysis of cytoplasmatic superoxide dismutase (CuZnSOD), mitochondrial superoxide dismutase (MnSOD), and extracellular superoxide dismutase (EcSOD) protein expression in liver homogenates of control (CT; n = 4) and cirrhotic (CH; n = 4) rats. Densitometry quantification in arbitrary units, normalized to glyceraldehyde 3-phosphate dehydrogenase (GAPDH), showed a decreased expression of CuZnSOD and MnSOD isoforms in CH livers, whereas no differences in EcSOD protein expression were observed (*P < 0.01 versus CT). (B) Total superoxide dismutase (SOD) activity in liver homogenates from CT (n = 8) and CH (n = 10) rats. Total SOD activity was significantly reduced in CH rat livers (*P < 0.01 versus CT).

NO Bioavailability Can Be Modulated by O2

Decreased NO bioavailability and increased O2 content in cirrhotic livers coexist with an increase in hepatic nitrotyrosinated proteins (16.2 ± 8.1 versus 1 ± 0.6 AU in control livers; P = 0.025; Fig. 5). Nitrotyrosinated proteins are considered a fingerprint of peroxynitrite formation, the result of O2 reacting with, and therefore scavenging, NO.

Figure 5.

Top: Representative western blot of nitrotyrosine (3-NT) abundance in liver homogenates from control (CT; n = 4) and cirrhotic (CH; n = 4) rats. Bottom: Densitometry quantification in arbitrary units, normalized to glyceraldehyde 3-phosphate dehydrogenase (GAPDH), showing more than a 20-fold increase in the main nitrated protein band in CH livers (*P < 0.01 versus CT).

SEC isolated from control rat livers treated with the SOD inhibitor DDC displayed a marked and significant increase in O2 levels, as shown by DHE staining under confocal microscopy, in comparison with vehicle-treated SEC (Fig. 6, top). The increase in O2 was associated with a significant reduction in NO bioavailability, as evidenced by DAF-FM-DA staining. Coadministration of SOD attenuated both the increase in O2 levels observed with DDC and the reduction in NO bioavailability (Fig. 6, top).

Figure 6.

Top: In situ superoxide (O2) detection by dihydroethidium staining (10 μM) and nitric oxide (NO) quantification by 4-amino-5-methylamino-2′,7′-difluorofluorescein diacetate (DAF-FM-DA) staining (10 μM) in sinusoidal endothelial cells (SEC) isolated from control rat livers treated with the specific superoxide dismutase (SOD) inhibitor diethyldithiocarbamate (DDC) alone or in association with SOD (DDC + SOD). Left: DDC-treated SEC had markedly increased intracellular O2 levels, which were significantly inhibited by coincubation with SOD [*P < 0.01 versus vehicle (Veh); #P < 0.01 versus DDC]. The data shown are from 817 individual Veh-SEC, 829 DDC-SEC, and 723 DDC + SOD-SEC obtained from three different experiments. Right: NO levels were significantly reduced when intracellular oxidative stress was induced by DDC. This low NO content was significantly ameliorated when cells were coincubated with SOD (*P < 0.01 versus Veh; #P < 0.01 versus DDC). The data shown are from 1395 individual Veh-SEC, 878 DDC-SEC, and 1083 DDC + SOD-SEC obtained from three different experiments. Bottom: In situ O2 and NO detection in bovine aortic endothelial cells (BAEC) infected with the adenovirus encoding for extracellular superoxide dismutase (EcSOD) or with the control adenovirus β-galactosidase (βgal) and treated with the oxidative stress inductor paraquat (PQ) or its Veh. PQ-treated BAEC transfected with βgal-Ad had a marked increase in intracellular O2 levels and a significant reduction in NO bioavailability. EcSOD-Ad transfection attenuated the increase in O2 promoted by PQ and blunted the reduction in NO bioavailability (left: *P < 0.05 versus βgal-Veh and #P < 0.01 versus βgal-PQ; right: *P < 0.01 versus βgal-Veh and P < 0.05 versus EcSOD + PQ). The data shown are from three independent experiments.

Similarly, endothelial cells (BAEC) transfected with EcSOD-Ad overexpressing SOD had significantly lower paraquat-induced O2 generation associated with significantly higher NO bioavailability in comparison with cells transfected with βgal-Ad (Fig. 6, bottom).

Discussion

Reduced NO bioavailability within the liver plays a major role in increasing hepatic vascular tone in cirrhosis. Up to now, it has been attributed to decreased NO production due to reduced eNOS activity in the liver.12 Our present results confirm diminished NO bioavailability within cirrhotic livers as shown by the observed reduced hepatic cGMP content. In addition, by using DAF staining and confocal microscopy, we were able to directly demonstrate reduced NO bioavailability in SEC from cirrhotic livers. This is the first time that NO content has been directly quantified in SEC only 12 hours after their isolation while they still preserve their typical phenotype.

Reduced NO content was associated with increased O2 levels. On the basis of indirect data, such as the increase in plasma and tissue levels of lipid peroxidation markers and the observation of reduced hepatic and plasma antioxidant content, it has been previously suggested that there is an increase in oxidative stress in several liver disorders.25, 47 However, this is the first study that, by using DHE staining and confocal microscopy, specifically demonstrates a marked increase in O2 levels in livers of rats induced to cirrhosis by CCl4 administration.

In agreement with a previous report,6 we found reduced SOD activity, the enzyme dismutating O2 to H2O2, as a possible mechanism underlying the observed increase in O2 in cirrhotic livers. Furthermore, our study clarifies that reduced SOD activity is, at least in part, due to decreased protein expression of the CuZnSOD and MnSOD isoforms but not of the EcSOD isoform.

In addition to reduced metabolization, the increase in O2 content may also be due to enhanced generation. In a previous study, we demonstrated that eNOS uncoupling, secondary to tetrahydrobiopterin deficiency, contributed to the reduced NO bioavailability of the cirrhotic liver.15 Uncoupled eNOS may also produce O2; however, in our study, no reduction in O2 was observed after NOS inhibition, and this suggests that eNOS is not a significant contributor of the increased O2 content of cirrhotic livers.

By contrast, our results showing that COX or XO inhibition markedly reduced intrahepatic O2 levels point out, for the first time, that these enzymatic systems are potential sources of O2 in cirrhosis, which provides the rationale for further investigations of potential conceptual and therapeutic relevance. The possible role of another potential source of O2 in cirrhosis, the NADPH oxidase system, has been recently discarded.37

The pathophysiological role of increased O2 reducing NO bioavailability has been extensively demonstrated in several vascular disorders.48–50 Our finding of an increase in nitrotyrosinated proteins, a well-recognized marker of the reaction of O2 with NO, strongly supports that this mechanism of reduction in NO bioavailability also occurs in the cirrhotic liver. The relationship between NO bioavailability and O2 content in the liver is further supported by our experiments in SEC demonstrating that NO bioavailability is modulated by O2. Indeed, increasing O2 content in SEC by incubation with the SOD inhibitor DDC was associated with a prominent reduction in NO bioavailability. Furthermore, abolition of the increase in O2 with SOD supplementation was followed by partial restoration of NO bioavailability. O2-NO interaction in endothelial cells is further emphasized by our complementary molecular experiments showing that, when endothelial cells are transfected with EcSOD and the increase in O2 caused by paraquat is prevented, NO bioavailability is maintained.

Altogether, these findings strongly support the concept that NO scavenging by O2 may be an important determinant of decreased NO bioavailability, endothelial dysfunction, and increasing hepatic vascular tone in cirrhosis.

The role of increased O2 in impairing NO biology within cirrhotic livers probably goes beyond its direct reaction with NO. Indeed, O2 could oxidize and therefore inactivate the NO synthase cofactor tetrahydrobiopterin51 or adjust interactions of eNOS with other inactivating or activating proteins.52 These considerations further emphasize that antioxidant therapy, by removing O2 from the cirrhotic livers, could be a new therapeutic strategy to improve intrahepatic NO bioavailability and to ameliorate hepatic vascular tone in cirrhosis and encourage further studies elucidating the mechanism of the O2-NO interaction and testing antioxidants as adjunctive therapy in the medical treatment of portal hypertension.

Acknowledgements

The authors are in debt to Montse Monclús for technical assistance.

Ancillary