fax: (34) 93-2279856
Liver Failure and Liver Disease
The eNOS cofactor tetrahydrobiopterin improves endothelial dysfunction in livers of rats with CCl4 cirrhosis†
Article first published online: 23 JUN 2006
Copyright © 2006 American Association for the Study of Liver Diseases
Volume 44, Issue 1, pages 44–52, July 2006
How to Cite
Matei, V., Rodríguez-Vilarrupla, A., Deulofeu, R., Colomer, D., Fernández, M., Bosch, J. and Garcia-Pagán, J.-C. (2006), The eNOS cofactor tetrahydrobiopterin improves endothelial dysfunction in livers of rats with CCl4 cirrhosis. Hepatology, 44: 44–52. doi: 10.1002/hep.21228
Potential conflict of interest: Nothing to report.
- Issue published online: 23 JUN 2006
- Article first published online: 23 JUN 2006
- Manuscript Accepted: 24 MAR 2006
- Manuscript Received: 24 JAN 2006
- Ministerio de Educacion y Ciencia. Grant Number: SAF 04/04783
- Fondo de Investigaciones Sanitarias. Grant Number: FIS 06/0623
- Instituto de Salud Carlos III. Grant Number: CO3/02
In cirrhosis, intrahepatic endothelial dysfunction is one of the mechanisms involved in the increased resistance to portal blood flow and therefore in the development of portal hypertension. Endothelial nitric oxide synthase (eNOS) uncoupling due to deficiency of tetrahydrobiopterin (BH4) results in decreased production of NO and plays a major role in endothelial dysfunction in other conditions. We examined whether eNOS uncoupling is involved in the pathogenesis of endothelial dysfunction of livers with cirrhosis. Basal levels of tetrahydrobiopterin and guanosine triphosphate (GTP)-cyclohydrolase (BH4 rate-limiting enzyme) expression and activity were determined in liver homogenates of control and rats with CCl4 cirrhosis. Thereafter, rats were treated with tetrahydrobiopterin, and eNOS activity, NO bioavailability, assessed with a functional assay, and the vasodilator response to acetylcholine (endothelial function) were evaluated. Livers with cirrhosis showed reduced BH4 levels and decreased GTP-cyclohydrolase activity and expression, which were associated with impaired vasorelaxation to acetylcholine. Tetrahydrobiopterin supplementation increased BH4hepatic levels and eNOS activity and significantly improved the vasodilator response to acetylcholine in rats with cirrhosis. In conclusion, the impaired response to acetylcholine of livers with cirrhosis is modulated by a reduced availability of the eNOS cofactor, tetrahydrobiopterin. Tetrahydrobiopterin supplementation improved the endothelial dysfunction of cirrhotic livers. (HEPATOLOGY 2006;44:44–52.)
In livers with cirrhosis, increased resistance to portal blood inflow is the primary factor in the pathophysiology of portal hypertension. Later, an increased portal blood inflow contributes to the maintenance and worsening of portal hypertension.1 The increased resistance of livers with cirrhosis is due to structural changes of the liver architecture but also to an increase in hepatic vascular tone. Endothelial dysfunction, characterized by an impairment in the endothelium-dependent response to vasodilators, is considered one of the mechanisms leading to the increased vascular tone of livers with cirrhosis2 and has been attributed to reduced NO bioavailability3 and to increased release of COX-1–derived vasoconstrictive prostanoids.4
The unconjugated pterin cofactor (6R)-L-erythro-5, 6, 7, 8-tetrahydrobiopterin (BH4) plays a crucial role in the regulation of eNOS activity. In the absence of BH4, NOS cannot catalyze L-arginine oxidation5, 6; rather, it receives electrons from NADPH and donates them to O2, resulting in the formation of superoxide anion (O2−) instead of NO, leading to reduced NO bioavailability. This situation is known as NOS “uncoupling.”7 In addition, O2− may interact with NO, leading to the formation of peroxinitrite, further decreasing NO bioavailability.8, 9
eNOS uncoupling, has been demonstrated to modulate endothelial dysfunction in other vascular disorders such as atherosclerosis,10 diabetes,11, 12 hypertension,13 hypercholesterolemia,14 or smoking,15 and supplementation with BH4 was able to reverse the reduced NO bioavailability and enhanced O2− production.14, 16–18
This study investigates whether BH4 deficiency and the ensuing eNOS uncoupling are involved in the reduced NO bioavailability and in the endothelial dysfunction of livers with cirrhosis.
Material and Methods
Animals and Induction of Cirrhosis.
Male Wistar rats weighting 175 to 200 g were induced to cirrhosis by inhalation of CCl4adding phenobarbital (0.3 g/L) to the drinking water as previously described.4, 19 When the rats with cirrhosis developed ascites (14-18 weeks of CCl4 inhalation), administration of CCl4 and phenobarbital was stopped, and the liver perfusion 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 (IDIBAPS). All experiments were performed according to the criteria of the Committee for the Care and Use of Laboratory Animals in the Hospital Clinic and IDIBAPS.
Measurement of BH4 Concentration in Control Rat Livers and Rat Livers With Cirrhosis.
Levels of biopterin were determined in homogenized livers of control and rats with cirrhosis treated with vehicle (n = 15) or BH4 (n = 15) using a high-performance liquid chromatography (HPLC) system, coupled to a fluorescence detector. Total biopterin represents the sum of BH4, its partially oxidized form 7,8-dihydrobiopterin (BH2), and the fully oxidized form biopterin (B). In acid conditions, BH4 and 7,8-dihydrobiopterin are oxidized to biopterin, whereas basic oxidation transforms only 7,8-dihydrobiopterin in biopterin, and BH4 undergoes side-chain cleavage to form pterin. Accordingly, BH4 levels were calculated from the difference of biopterin obtained after acidic and basic iodine oxidation, using the method of Fukushima and Nixon.20 Samples were separated by HPLC using a Sferisorb ODS-2 Column (4.6 × 250 mm, 5-μm particle Ø, Waters) using 5% methanol/95% water as mobile phase. Fluorescence was measured with a fluorescence detector (Waters 474, Barcelona, Spain) (350 nm excitation, 450 nm emission). Results were expressed as nmoles/mg protein. The protein content of each sample was determined by the Lowry method, with bovine serum albumin as the standard.
Measurement of GTP Cyclohydrolase I Activity in Control Livers and Livers With Cirrhosis.
GTP Cyclohydrolase I (GTPCH) activity, the critical enzyme for pterin synthesis, was assayed by incubating liver homogenate of control rats and rats with cirrhosis with GTP 500 μmol/L at 37°C for 1 hour, in the dark, as previously described.21 The reaction product was oxidized to neopterin triphosphate by acidic iodine. After reduction of excessive iodine by ascorbic acid, incubation with alkaline phosphatase for 45 minutes at 37°C dephosphorylated the neopterin phosphate, yielding neopterin, which was determined by HPLC, according to the previously described protocol for pterin measurement. GTPCH activity was expressed as pmol neopterin/mg protein/min. The protein content of each sample was determined by the Lowry method with bovine serum albumin as a standard.
Evaluation of Endothelial Function in Perfused Rat Livers: Effect of BH4 Supplementation.
Control rats and rats with cirrhosis were anesthetized with inactin (150 mg/kg intraperitoneally) and maintained at constant temperature of 37 ± 0.5°C (continuously monitored during the experiment). A tracheotomy and cannulation with a PE-240 catheter; (Portex, Kent, UK) was performed to maintain an adequate respiration during the anesthesia. Through a median laparotomy the ileocolic vein was cannulated with polyethylene tubing (PE-10; Portex, Kent, UK). Thereafter, the laparotomy was closed to minimize dehydration. After 15 minutes of stabilization, either BH4 (8 mg/kg/h) or vehicle was administered for 30 minutes through the ileocolic catheter; the liver was quickly isolated and perfused as previously described19 in a recirculating fashion with a total volume of 100 mL at a constant flow rate of 35 mL/min. An ultrasonic flow probe (T201, Transonic System, Ithaca, NY) and a pressure transducer were placed on line, immediately ahead of the portal inlet cannula, to continuously monitor portal flow and perfusion pressure. The flow probe and the pressure transducers were connected to a Powerlab (4SP) linked to a computer using the Chart v5.0.1 for Windows software (ADInstruments, Mountain View, LA).
The perfused rat liver preparation was allowed to stabilize for 30 minutes before the vasoactive substances were added. The gross appearance of the liver, stable perfusion pressure, bile production over 0.4 μL/min × g liver and a stable buffer pH (7.4 ± 0.1) were measured during this period.19 If any viability criteria were not satisfied, the experiment was discarded. The intrahepatic microcirculation was preconstricted by adding to the reservoir, during a 3-minute period, the α1-adrenergic agonist methoxamine to achieve the final concentration of 10-4 mol/L. Five minutes later, concentration–effect curves to cumulative doses of acetylcholine (10−8, 10−7, 10−6 mol/L) were evaluated in control livers (n = 13) and livers with cirrhosis (n = 17) treated with vehicle (n = 15) or BH4(n = 15). The concentration of acetylcholine was increased by one log unit every 1.5 minutes. Response to cumulative doses of acetylcholine was calculated as percent change in perfusion pressure as previously described.2, 19
Effect of NO Inhibition on Baseline Portal Perfusion Pressure and on the Endothelial Function of Livers With Cirrhosis Pretreated With Vehicle or With BH4.
NO biodisponibility was estimated as the increase in portal perfusion pressure after adding the NO synthase inhibitor L-NNA. In rats with cirrhosis treated with vehicle (n = 4) or BH4 (n = 4; 8 mg/kg/h, 30 minutes), the livers were isolated and perfused with Krebs solution including 20 μmol/L indomethacin. After stabilization, L-NNA (10−3mol/L) was added to the perfusate, and after 20 minutes the increase in portal pressure was recorded. In addition, to assess whether BH4 effects were NO mediated, in isolated and perfused livers from rats with cirrhosis treated with vehicle (n = 11) or BH4 (n = 10), L-NNA (10−3 mol/L) was added to the perfusion buffer, and 20 minutes later the liver was preconstricted with Mtx (10−4 mol/L) and concentration–effect curves to cumulative doses of acetylcholine (10−8, 10−7, 10−6 mol/L) were performed. Portal perfusion pressure was continuously registered during the whole experiment.
Measurement of Nitric Oxide Synthase Activity.
Nitric oxide synthase activity was measured by determining the conversion of14 C-labeled L-arginine to14 C-labeled L-citrulline, according to a previously reported method22, 23 in livers with cirrhosis treated with vehicle (n = 10) or BH4 (n = 10). Briefly, rat livers were quickly dissected, blotted, weighed, and cut into small pieces. Tissues were homogenized in ice-cold buffer (250 mg/mL, 4°C, pH 7.4) containing 10 mmol/L HEPES, 320 mmol/L sucrose, 1 mmol/L dithiothreitol, 0.1 mmol/L ethylene guanosine tetra-acetic acid, 10 μg/mL trypsin inhibitor, 10 μg/mL leupeptin, and 2 μg/mL aprotinin. Homogenates were then centrifuged (10,000g for 20 minutes at 4°C), and the supernatants were assayed in duplicate. The supernatant (40 μL) was incubated for 10 minutes at 37°C in 100 μL of reaction buffer containing 40 mmol/L KH2PO4 (pH 7.4), 8 mmol/L L-valine, 0.3 mmol/L NADPH, 1 mmol/L MgCl2, 0.2 mmol/L CaCl2, 20 μmol/L L-arginine, and L-(U-14 C) arginine monohydrochloride (specific activity 11.8 GBq/mmol; Amersham International, Amersham, Buckinghamshire, England). The reaction was arrested via the removal of the substrate L-arginine by the addition (0.5 mL) of a 1:1 v/v suspension of Dowex-50W/distilled water (Na+ form). The Na+ form of Dowex-50W was prepared by washing the H+ form of the resin with 1 mol/L NaOH 4 times, and then washing with distilled water until the pH was less than 7. The resin–incubate mixture was dispersed and diluted by the addition of 0.86 mL distilled water. After allowing the resin to settle for 30 minutes, the supernatant was removed (0.975 mL) for the estimation of the radiolabeled products by scintillation counting. Enzymatic activity was expressed as picomoles per minute per milligrams of tissue.
RNA Isolation and Reverse Transcription.
Total RNA was isolated from each frozen CH or CT liver using Trizol method (Invitrogen, Paisley, UK). RNA was treated with DNAse (Ambion, Austin TX) to eliminate contaminating DNA. For cDNA synthesis, 1 μg of total RNA was retrotranscribed using MLV reverse transcriptase and random hexamers, as described by the manufacturer (Invitrogen).
Real Time Quantitative Polymerase Chain Reaction for Guanosine Triphosphate Cyclohydrolase I and Cyclohydrolase Feedback Regulatory Protein.
cDNA templates were amplified by quantitative polymerase chain reaction (q-PCR) using the SYBR GREEN PCR reagents (Applied Biosystems, Foster City, CA) on an ABI Prism 7900 sequence Detection System (Applied Biosystems). The primers of guanosine triphosphate cyclohydrolase I (GTPCH), cyclohydrolase feedback regulatory protein (GFRP), and the endogenous control RNA 18S were designed using Primer Express software (Applied Biosystems) and were as follows:
GTPCH Forward primer, 5′ ACCGCCATGCAGTTCTTCAC 3′
GTPCH Reverse primer, 5′ AGCATCGTTCAGGACATCTGAGA 3′
GFRP Forward primer, 5′ TGGTGTCTGCACAAGGAATGA 3′
GFRP Reverse primer, 5′ TGGCGATCCCTCCAAAGA 3′
RNA 18S Forward primer, 5′ CACGGCCGGTACAGTGAAA 3′
RNA 18S Reverse primer, 5′ AGAGGAGCGAGCGACCAA 3′
Different species of GTPCH I mRNA have been shown to exist.24–26 All of them were identical at their central and 5′-regions, but diverged at their 3′-ends. This heterogeneity might be the result of alternative splicing of pre-mRNA, alternative transcription initiation sites, or multiple polyadenylation signals. Only some of these isoforms have been shown to have enzymatic activity.27 The primers used in the current study were designed to align with a zone supposed to be common in all mRNA species, suggesting therefore that we were amplifying all GTPCH mRNA species that exist in the liver.
Each experiment was performed in 20-μL reaction volume, containing 2 μL cDNA, 1× SYBR PCR buffer, 3 mmol/L MgCl2, dNTPs Mix (4 nmol dATP, 4 nmol dCTP, 4 nmol dGTP, and 8 nmol dUTP), 1 U Amplitaq Gold® DNA polymerase, 0.1 U AmpErase®, and 250 nmol/L of each primer. After initial incubation at 95°C for 10 minutes to activate Taq Polymerase, 40 cycles at 95 °C for 30 seconds, 60 °C for 1 minute, and 72 °C for 1 minute were performed. A melting curve was obtained by increasing the temperature from 55°C to 95°C with a temperature transition rate of 0.1°C/s after the end of each q-PCR reaction to determine the number of products that were present in each reaction. Analysis of melting curves of PCR products allowed us to distinguish specific amplification products from nonspecific products and primers dimerization. All reactions were run in triplicate and included no template controls for each gene. Efficiency of the q-PCR was analyzed by generation of standard curves for each gene, using 1:5, 1:10, 1:100 and 1:1,000 dilutions of a positive sample. The slopes of Ct/log dilution plots for the reaction were −3.43 for GTPCH, −3.49 for GFRP, and −3.29 for 18S. The relative quantification of each gene was done using the comparative method (ΔΔCt) using RNA 18S (18S) as endogenous control. The mean of GTPCH expression in control samples was used as a relative calibrator, and mRNA expression levels were given as arbitrary units.
Western Blot Analysis of GTP Cyclohydrolase I.
Protein expression for GTPCH in rat livers from three livers from rats with cirrhosis and three control rat livers was assessed by Western blot as follows. Livers were collected, snap frozen in liquid N2, and stored at −80°C until analyzed. Livers were minced thoroughly with mortar and pestle under liquid nitrogen. For each sample, a similar amount of obtained powder were collected in 200 μL triton-lysis buffer containing Tris/HCl (pH 7.4, 20 mmol/L), NaCl (150 mmol/L), NaF (20 mmol/L), Na4P2O7(10 mmol/L), okadaic acid (10 nmol/L), Na3VO4 (2 mmol/L), antipain (1 μg/mL), aprotinin (1 μg/mL), chymostatin (1 μg/mL), leupeptin (1 μg/mL), pepstatin (1 μg/mL), trypsin inhibitor (1 μg/mL), phenylmethylsulfonylfluoride (44 mg/mL), and Triton-X100 (1%v/v), left on ice for 10 minutes, and then centrifuged at 10,000g/10 minutes. Protein concentration was assessed by the Bradford method. Aliquots from each sample, containing equal amounts of protein (50 μg), were run on a 12% sodium dodecyl sulfate polyacrylamide gel, and transferred to a nitrocellulose membrane. The efficiency of the transfer was visualized by Ponceau staining. The blots were subsequently blocked for 2 hours with phosphate-buffered saline containing 0.1% (v/v) Tween 20, 5% (wt/vol) nonfat dry milk, and 10% (wt/vol) horse serum, and proved with rabbit anti-GTPCH antibody (a gift of Dr. Gabriele-Werner Felmeyer, Innsbruck University, Austria) diluted 1:2,500 in phosphate-buffered saline containing 0.1% (v/v) Tween 20, 5% (wt/vol) nonfat dry milk, and 1% (wt/vol) horse serum, overnight at 4°C followed by an incubation with goat anti-rabbit (1:10,000 in the same solution) horseradish peroxidase–conjugated secondary antibodies for 1 hour at room temperature. Blots were revealed by chemiluminescence. Protein expression was determined by densitometric analysis using the Science Lab Image Gauge. Images were obtained using Science Lab 2001 Image Gauge (Fuji Photo Film GMBH, Düsseldorf). After stripping, blots were assayed for glyceraldehyde-3-phosphate dehydrogenase. Quantitative densitometric values of GTPCH were normalized to glyceraldehyde-3-phosphate dehydrogenase and displayed in histograms.
Statistics were performed using the SPSS 10.0 for Windows statistical package (SPSS Inc., Chicago, IL). All results are expressed as mean ± SD values. Comparisons between groups were performed with Student's t test for unpaired data. A P value of < .05 was considered significant.
BH4 Hepatic Tissue Levels Are Reduced in Rat Livers With Cirrhosis.
Total biopterin level was significantly decreased in cirrhotic livers in comparison with control livers (15.1 ± 7.06 vs. 25.4 ± 10.8 pmol/mg protein; P = .008) (Fig. 1A). Reduction in biopterin in livers from rats with cirrhosis was completely attributable to a significant and marked reduction in BH4 (the active form of biopterin) (8.3 ± 5.96 vs. 21.5 ± 11.3 pmol/mg protein; P = .001) (Fig. 1A), whereas in livers from rats with cirrhosis, the oxidized pterins, those biologically inactive (dihydrobiopterin and biopterin) were significantly increased (6.8 ± 3.9 vs. 3.9 ± 1.7 pmol/mg protein in control; P = .038). As a consequence, the reduced/oxidized pterin ratio (BH4/(BH2+B)), a more reliable marker of biopterin status, was markedly reduced in livers from rats with cirrhosis compared with control livers (Fig. 1B)
BH4 Supplementation Increases Tissue Levels of BH4 But Not of Its Oxidized Forms in Control Rat Livers and in Rat Livers With Cirrhosis.
BH4 treatment markedly increased tissue BH4 levels both in control (to 44.8 ± 17.4 pmol/mg protein; P = .0017) and livers from rats with cirrhosis (to 74 ± 24.4 pmol/mg protein; P = .0012), without significant changes in the oxidized forms (5.1 ± 2 pmol/mg protein in controls, and 6.7 ± 0.9 pmol/mg protein in livers from rats with cirrhosis). As a consequence, after BH4 treatment the reduced/oxidized pterin ratio was markedly improved in livers from rats with cirrhosis (11.11 ± 3.79 vs. 1.76 ± 1.59 in livers from rats with cirrhosis treated with vehicle; P = .001) reading values similar to those of control rat livers (10.39 ± 6.47 after BH4supplementation vs. 6.69 ± 4.18 in control livers treated with vehicle; P = .13).
GTPCH Tissue Enzymatic Activity and GTPCH mRNA and Protein Expression Are Reduced in Rat Livers With Cirrhosis.
GTPCH activity was significantly decreased in livers from rats with cirrhosis compared with controls (0.04 ± 0.01 vs. 0.07 ± 0.01 pmol/mg protein/min; P = .001) (Fig. 2A). A positive correlation between GTPCH activity and total biopterin levels was found in livers from rats with cirrhosis (Fig. 3), evidencing the role of GTPCH as the limiting enzyme in biopterin synthesis. Reduced GTPCH-1 activity was associated with a 25% reduction in GTPCH mRNA expression, as evidenced by q-PCR (0.78± 0.12 arbitrary units in cirrhotic livers [n = 12] vs 1.03 ± 0.08 in controls [n = 10]; P < .05), and an even greater reduction in GTPCH protein expression (25% of the control livers) (Fig. 2B).
GFRP Expression Is Not Upregulated in Livers With Cirrhosis.
There were no significant differences in GFRP mRNA expression in livers from rats with cirrhosis and in control rat livers (0.89 ± 0.37 vs. 1.04 ± 0.29 arbitrary units; NS).
BH4 Supplementation Improves the Endothelial Dysfunction of Rat Livers With Cirrhosis.
CCl4 livers from rats with cirrhosis had a significantly greater baseline PP (7.3 ± 1.1 vs. 4.1 ± 0.7 mmHg; P < .001) and intrahepatic vascular resistance (2.4 ± 0.5 vs. 1.4 ± mmHg/g/min/mL; P < .05) than control livers. No significant difference in liver weight was observed (11.8 ± 1.1 vs. 11.5 ± 0.7 g; P = .8).
No significant differences were seen in the baseline PP in livers from rats with cirrhosis treated with BH4or vehicle or (9.1 ± 1.4 vs. 7.3±0.4 mmHg; NS). The response of PP to 10-4mol/L methoxamine after BH4 administration, although lower, was not significantly different from that observed after vehicle (12.2 ± 2.2 mmHg in BH4 vs 14.1 ± 1.4 mmHg in vehicle; NS).
Control livers exhibited an incremental vasorelaxation to cumulative doses of Ach. However, livers from rats with cirrhosis had an impaired vasodilatory response to Ach, as compared with controls: 10−8 (−7.1 ± 4 vs. −14.7 ± 6% in controls; P = .009), 10−7(−10.8 ± 7 vs. −22.5 ± 7%; P = .008), 10−6 mol/L (−7.6 ± 15 vs. −28.9 ± 8%; P = .008) (Fig. 4). BH4 treatment did not significantly modify the response to Ach of control livers (Fig. 4A). However, in livers from rats with cirrhosis, BH4 supplementation markedly and significantly improved the vasodilatory response to Ach (Fig. 4B) to the point that the response to Ach of livers from rats with cirrhosis treated with BH4 was no longer significantly different from that observed in control livers treated with BH4: 10−8 (−12 ± 6 vs. −12.9 ± 6%, P = .76), 10−7 (−19.4 ± 9 vs. −21.4 ± 7%, P = .65), 10−6 (−21.3 ± 12 vs. −25.4 ± 10%, P = .49) (Fig. 4). Equivalent results are observed if the results are expressed as absolute changes (data not shown).
Nitric Oxide Synthase Activity and NO Biodisponibility Are Increased in Rat Livers With Cirrhosis Treated With BH4.
BH4 administration increased NOS activity in livers from rats with cirrhosis (2.66 ± 0.05 vs. 2.23 ± 0.09 pmol/min/mg tissue in livers from rats with cirrhosis treated with vehicle; P < .001). Similar findings were observed when the results were expressed as pmol/min/mg protein. Most of this NOS activity may be eNOS, because although total NOS activity was measured, it has been shown that eNOS is the main NOS isoform in CCl4 livers from rats with cirrhosis.3;28–30
L-NNA administration significantly increased portal perfusion pressure in livers from rats with cirrhosis pretreated with BH4 or vehicle. This increase was significantly greater in rats treated with BH4(2.5 ± 1.3 vs. 0.37 ± 0.2; P = .016) (Fig 5), suggesting that BH4 increased NO biodisponibility.
Improvement in the Intrahepatic Endothelial Function Caused by BH4 Administration Is Mediated by NO.
As shown in Fig. 6, after NO synthase inhibition no differences in the response to Ach were observed in livers from rats with cirrhosis treated with vehicle or BH4.Thus, NO synthase inhibition completely prevented the beneficial effect of BH4supplementation.
Endothelial dysfunction is one of the pathophysiological mechanisms contributing to the dynamic increase in intrahepatic resistance of the liver with cirrhosis. A reduced NO availability in the liver with cirrhosis has been well demonstrated2, 28 and may be accounted for by decreased eNOS activity due (among others) to decreased eNOS phosphorylation and increased caveolin expression31 and possibly to increased scavenging of NO by reactive oxygen species.8, 9 In addition, we have recently shown that an increased production of hepatic COX1-derived prostanoids is involved in the pathogenesis of endothelial dysfunction in livers from rats with cirrhosis.32 However, the effects of these compounds unmasked by COX blockade were dependent on NO availability, because these were almost entirely blunted when NO was also blocked.32 Thus, both reduced intrahepatic NO availability and increased production of COX-derived prostanoids seems to be involved in the development of the increased vascular tone in cirrhosis.
Different studies performed in other vascular disorders have shown that, in situations in which a deficit of the cofactor BH4 exist or there is a shift from BH4 to its oxidized inactive forms BH2 and B, the production of NO by NOS is shifted to the production of O2−, a phenomenon known as “eNOS uncoupling.”5–7 The reduced production of NO may be aggravated in the setting of impaired antioxidant mechanisms. The generated superoxide radical by uncoupled eNOS would rapidly interact with NO, leading to the formation of peroxynitrite. At physiological pH, peroxynitrite rapidly oxidizes BH4 to its biologically inactive form BH2.8 Dihydrobiopterin (BH2), by displacing the prebound BH4 from its binding site in eNOS, may further enhance eNOS uncoupling in a vicious cycle.33
The liver is one of the major sites where biosynthetic and recycling enzymes of BH4 are constitutively expressed. Tetrahydrobiopterin results both from de novo synthesis, in which GTP cyclohydrolase (GTPCH) is the first and rate-limiting enzyme and the only one being regulated in the pathway of BH4-synthesis, and from a regenerating pathway, the so called “salvage pathway,” in which the oxidized form q-dihydrobiopterin is reduced to BH4.34
The finding of the current study that livers from rats with cirrhosis have reduced levels of BH4and, more importantly, a decreased reduced/oxidized pterin ratio that correlates more tightly than absolute BH4 levels with superoxide/nitric oxide formation from NOS, strongly suggests that BH4 deficiency may play a role in the decreased NO availability of livers from rats with cirrhosis. In this regard, BH4supplementation to rats with cirrhosis increased eNOS activity as well as NO bioavailability, as shown by a greater increase in baseline portal perfusion pressure after NO-inhibition in livers from rats with cirrhosis pretreated with BH4compared with those pretreated with vehicle, and by a marked improvement in the vasodilator response to Ach. Altogether these findings further support that tetrahydrobiopterin deficiency is involved in the reduction of NO bioavailability and in the pathophysiology of intrahepatic endothelial dysfunction in cirrhosis.
Part of the increase in NO bioavailability may be related to the antioxidant effect of BH4.35 However, this is unlikely because BH4 supplementation enhanced hepatic BH4 levels without increasing the levels of its oxidized forms. The main mechanisms for the beneficial effects of BH4 supplementation appear to be an amelioration of eNOS uncoupling. Our finding that BH4 treatment did not correct endothelial dysfunction in rats previously treated with the NO synthase inhibitor L-NNA further supports that the beneficial effect of BH4 on endothelium-dependent relaxation of the liver with cirrhosis is NO-mediated. All these observations are in accordance with our working hypothesis that BH4/eNOS uncoupling contributes to decrease NO bioavailability and therefore worsens endothelial dysfunction in the liver with cirrhosis.
BH4 functions as an essential cofactor not only of NOSs but also of aromatic amino acid hydroxylases (e.g., phenylalanine-, tryptophan-, tyrosine-) enzymes, which take part in the synthesis of dopamine, norepinephrine, or serotonin among others.34, 36 We cannot rule out that BH4 administration could have effects on multiple other intrahepatic processes as well.
Our finding of a decreased activity and expression of GTPCH, the key enzyme in BH4synthesis, indicates that this is likely to be the main mechanism of the low liver biopterin content in cirrhosis. Activity of GTPCH also may be down-regulated via the interaction, induced by its end-product BH4, with GFRP. However, this mechanism appears unlikely to explain the BH4 depletion observed in cirrhosis, because GFRP mRNA expression was not significantly increased and total BH4 levels were actually reduced.
Mechanisms regulating GTPCH expression are not well understood. Both cyclic adenosine monophosphate–dependent protein kinase and stress-activated protein kinases are able to activate GTP cyclohydrolase gene transcription in normal conditions.37 By contrast, glucocorticoids decrease GTPCH gene transcription,38, 39 and transforming growth factor beta (TGF-β) has been shown to counterbalance the cytokine activation of nitric oxide synthase in endothelial cells by down-regulating GTPCH mRNA expression.40–42 Because increased levels of TGF-β are present in cirrhosis, speculating that TGF-β could play a role in the decreased GTPCH expression found in cirrhosis is tempting. In addition, insulin has been shown to induce GTPCH I activation. Reduced insulin levels and insulin resistance may be the cause of the reduced GTPCH I activity and mRNA expression found in diabetes.12 Because insulin resistance may be present in patients with liver cirrhosis, we could speculate that this could also be another possible factor contributing to the decreased GTPCH expression that we found in cirrhosis. The possible disturbance of all these signaling pathways within the livers with cirrhosis deserves further study. Similarly, we have measured BH4 content and GTPCH-1 activity in liver homogenates, and we cannot assume which is the cell or cells mostly responsible for such a deficit.
We cannot rule out that alterations on the BH4 recycling systems also could be involved in decreasing BH4 levels. In addition, the increased oxidation of the remaining BH4 to its inactive forms (dihydrobiopterin and biopterin) further contributes to decrease the reduced/oxidized pterin ratio, a major determinant of eNOS coupling. Increased oxidation of BH4 in livers with cirrhosis might be caused by the increased reactive oxygen species (ROS) observed in livers with cirrhosis.43
Our findings of BH4/eNOS uncoupling in the liver circulation represent another instance of opposite disturbances in mechanism of circulatory homeostasis in the hepatic versus splanchnic circulation because up-regulation of GTPCH activity was reported in the mesenteric vasculature of rats with cirrhosis.44 This study, however, did not assess protein or mRNA expression of GTPCH or GFRP, and therefore whether up-regulation of GTPCH activity was due to transductional or a postransductional mechanism remained obscure. Other examples of these apparently paradoxically opposite effects in cirrhosis are the increase in NO and vasodilator prostanoids observed in the mesenteric circulation, whereas liver with cirrhosis exhibits reduced NO and increased vasoconstrictor prostanoids.45
In conclusion, this study provides evidence that the impaired vasorelaxation to acetylcholine of livers with cirrhosis is modulated by a reduced availability of the eNOS cofactor tetrahydrobiopterin, determining eNOS uncoupling and decreasing the net amount of NO in the intrahepatic circulation. In addition, the study demonstrates that BH4 administration almost completely corrects the impaired vasorelaxation of the cirrhotic liver, which suggests that BH4 supplementation may represent a novel strategy to improve liver vascular sensitivity in cirrhosis.
- 3Low NO bioavailability in CCl4 cirrhotic rat livers might result from low NO synthesis combined with decreased superoxide dismutase activity allowing superoxide-mediated NO breakdown: A comparison of two portal hypertensive rat models with healthy controls. Comp Hepatol 2003; 2(1): 2., , , , .
- 11Cellular basis of endothelial dysfunction in small mesenteric arteries from spontaneously diabetic (db/db -/-) mice: role of decreased tetrahydrobiopterin bioavailability. Br J Pharmacol 2002; 136: 255–263., , , .