Positive cardiac inotropic effect of albumin infusion in rodents with cirrhosis and ascites: molecular mechanisms

Authors


  • Potential conflict of interest: Nothing to report.

Abstract

The aim of this study was to evaluate the effect and molecular mechanism of albumin infusion on cardiac contractility in experimental cirrhosis with ascites. Cardiac contractility was recorded ex vivo in rats with cirrhosis and ascites and in control rats after the injection in the caudal vein of albumin, saline, or hydroxyethyl starch (HES). Gene and protein expression of β-receptors and pathways involved in their intracellular signaling such as Gαi2 protein (Gαi2), adenylate cyclase 3 (Adcy3), protein expression of tumor necrosis factor alpha (TNF-α) and inducible nitric oxide synthase (iNOS), were evaluated in cardiac tissue in both groups. Phosphorylation and membrane-translocation of the cytosolic components of nicotinamide adenine dinucleotide phosphate (NAD(P)H)-oxidase and translocation of nuclear factor kappa B (NF-κB) were also evaluated. After saline intravenous injection, cardiac contractility was significantly reduced in rats with cirrhosis as compared to control rats (P < 0.01). This was associated with: (1) increased expression of protein Gαi2 (P < 0.05), TNF-α (P < 0.05), iNOS (P < 0.05); (2) increased NAD(P)H-oxidase activity (P < 0.05); (3) increased nuclear translocation of NF-κB (P < 0.05); and (4) lower expression of Adcy 3 (P < 0.05) in cardiac tissue of rats with cirrhosis. After albumin injection cardiac contractility (P < 0.01), protein expression of TNF-α, iNOS, Gαi2, and Adcy3, NAD(P)H-oxidase activity and nuclear translocation of NF-κB in cardiac tissue of rats with cirrhosis were reversed to control levels (P < 0.05). HES injection did not modify cardiac contractility and nuclear translocation of NF-κB in cardiac tissue of rats with cirrhosis. Conclusion: Albumin exerts a positive cardiac inotropic effect in rats with cirrhosis and ascites counteracting the negative effects of oxidative stress- and TNF-α-induced activation of NF-κB-iNOS pathway and oxidative stress-induced alteration of β-receptor signaling. (HEPATOLOGY 2013)

Cirrhosis is associated with an impairment of cardiovascular function which includes (1) hyperdynamic systemic circulation (i.e., increased heart rate, cardiac output and plasma volume, reduced peripheral vascular resistance and arterial hypotension), and (2) cardiac dysfunction,1 which has been termed “cirrhotic cardiomyopathy.” It is characterized by the following: (i) blunted contractile responsiveness to stress and to pharmacological stimulation and/or (ii) altered diastolic relaxation and/or (iii) electrophysiological abnormalities in the absence of any other known cardiac disease.2

Several factors are involved in the development of the impairment of cardiac contractility including: altered β-adrenergic receptor signal transduction, abnormal plasma membrane fluidity, impaired cardiac excitation-contraction coupling, and conductance abnormalities.1, 2 More recently it has been observed that in experimental cirrhosis an increased level of proinflammatory cytokines such as tumor necrosis factor (TNF)-α can contribute, together with oxidative stress, to an overexpression and an overactivity of the inducible isoform of nitric oxide synthase (iNOS) in cardiac tissue.3, 4 The consequent overproduction of nitric oxide (NO) can exert a negative inotropic effect through different mechanisms including an inhibitory effect on protein kinase A (PKA) which can counteract the stimulatory effect of the β-adrenergic signaling on this enzyme.5-7 This pathway has been shown to operate in the pathophysiology of the cardiac contractile dysfunction that characterizes sepsis. The overexpression and overactivity of iNOS in the experimental model of sepsis can be effectively decreased by the infusion of albumin, which has been proven to reduce iNOS expression through an inhibitory effect on a crucial gene transcription mechanism such as the nuclear translocation of nuclear factor kappa B (NF-κB).8

Albumin infusion has been used for many years in the management of patients with cirrhosis and ascites. In particular, one of the indications for its use in patients with cirrhosis is the prevention of hepatorenal syndrome (HRS) during an episode of spontaneous bacterial peritonitis (SBP). The rationale behind the use of albumin in this clinical condition is improvement of the effective circulating volume through the capacity of albumin to prevent the fall in cardiac output due to the bacterial infection.9, 10 Although the effect of albumin on cardiac output is simply attributed, in current opinion, to its ability to increase cardiac preload, the action of albumin in this situation can be far more complex. First, albumin binds many substances such as NO, reactive oxygen species (ROS), and proinflammatory cytokines,11-14 which may be involved in the pathogenesis of both the peripheral arterial vasodilatation and the cardiac dysfunction in cirrhosis and ascites. In addition, it can be hypothesized that in cirrhosis, as in sepsis, albumin can exert a positive inotropic effect in the cardiac tissue through an inhibitory effect on the expression and activity of iNOS.15

The aim of our study was to verify in an animal model of cirrhosis with ascites if albumin infusion can improve cardiac contractility through a mechanism that is independent of the increase of the preload, and to define its possible molecular basis.

Abbreviations

Adcy3, adenylate cyclase 3; β-AR, beta-adrenergic receptor; BDL, bile duct-ligated; BSA, bovine serum albumin; CCl4, carbon tetrachloride; DTT, dithiothreitol; EGTA, ethylene glycol tetraacetic acid; ELISA, enzyme-linked immunosorbent assay; Gαi2, Gαi2 protein; Gαs, Gαs protein; GAPDH, glyceraldehyde 3-phosphate dehydrogenase; HES, hydroxyethyl starch; HPRT, hypoxanthine guanine phosphoribosyl transferase; HRS, hepatorenal syndrome; iNOS, inducible nitric oxide synthase; LVDP, left ventricular developed pressure; NAD(P)H, nicotinamide adenine dinucleotide phosphate; NF-κB, nuclear factor-κB; NO, nitric oxide; PKA, protein kinase A; PMSF, phenylmethylsulfonylfluoride; PRA, plasma renin activity; ROS, reactive oxygen species; SBP, spontaneous bacterial peritonitis; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; TNF-α, tumor necrosis factor α.

Materials and Methods

The study was performed in conscious, male, adult Wistar-Kyoto rats with cirrhosis and ascites, and in conscious, male, adult Wistar-Kyoto control rats. The study was conducted in accordance with the principles and procedures outlined in the National Institutes of Health Guide for the Care and Use of Laboratory Animals and was approved by the Italian Ministry of Health (approval on September 8th 2006 by the Italian Ministry of Health according to legislative decree no. 116/92).

Animal Model of Cirrhosis.

Cirrhosis was induced in adult (200-225 g) male Wistar-Kyoto rats (Charles River, Calco, Italy) by exposing the animals to the inhalation of carbon tetrachloride (CCl4) twice a week up to ascites appearance, as described.16 Thirty rats with cirrhosis and ascites and 30 control rats were housed in environmentally controlled facilities and allowed free access to chow and distilled water containing phenobarbital (Luminal 0.3 g/L, Bracco, Milan, Italy). All the experiments were performed 1 week after the last CCl4 dose. The presence of ascites was confirmed at the time the rats were killed following laparotomy by abdominal fluid drainage. Diagnosis of cirrhosis was confirmed postmortem by microscopic examination of hematoxylin-stained liver sections.

Albumin Infusion.

A 43% albumin (Sigma Aldrich, Milan, Italy) solution in 0.9% saline was injected into the caudal vein of 15 rats with cirrhosis and ascites and 15 control rats. Two albumin doses were administered: the first dose (1.5 g/kg) was infused 3 days before sacrifice of the animals, and the second dose (1 g/kg) was infused the day before sacrifice. The doses were chosen in order to simulate those which are commonly given to patients with cirrhosis and ascites when a spontaneous bacterial peritonitis (SBP) is diagnosed in order to prevent the development of renal failure.10, 11

Saline Infusion.

Fifteen control animals and 15 rats with cirrhosis and ascites were treated with an equivalent volume of 0.9% of saline.

Starch Infusion.

A synthetic plasma expander, hydroxyethyl starch (HES) (10% hydroxyethyl starch HES 200/0.5 in isotonic sodium chloride solution, Fresenius Kabi Deutschland, Friedberg, Germany) was injected into the caudal vein of 15 rats with cirrhosis and ascites and 15 control rats. HES was administered at the same doses and with the same schedule of albumin.

Monitoring the Effective Circulating Volume.

To monitor the effects of albumin, saline, or HES on the effective circulating volume a blood sample was obtained in rats with cirrhosis and ascites just before and after administration of the two doses of each plasma expander for the measurement of plasma renin activity (PRA). PRA was measured by means of RIA (Radim, Pomezia, Italy).

Measurement of Cardiac Contractility.

After intraperitoneal injection of 300 μL of heparin (5000 U/mL) animals were sacrificed by decapitation. Subsequently, the heart, cannulated through the aorta, was mounted in a Langendorff apparatus for retrograde perfusion, perfused at constant flow (10 mL/min), and electrically driven at a frequency of 6 Hz using platinum electrodes placed in the left atrium as described.17

Left ventricular developed pressure (LVDP) was measured by inserting a steel cannula into the left ventricle and connecting it to a pressure transducer (2B Instruments, Besozzo, Italy). The perfusion medium was a modified Krebs-Henselheit saline solution with the following composition: 118 mM NaCl, 4.7 mM KCl, 2.5 mM CaCl2, 1.2 mM MgSO4, 25 mM NaHCO3, 11.1 mM glucose, and 2.0 mM disodium pyruvate, bubbled with a 95%/5% O2/CO2 mixture at 37°C. Following a stabilizing period of 20 minutes, the heart was stimulated with increasing concentrations of isoproterenol (from 10−10 to 10−8 M) to obtain a concentration-response curve. The LVDP was continuously recorded and stored by a real-time digital acquisition and analysis system (model MP-100, Biopac System, Santa Barbara, CA). LVDP was calculated as the difference between systolic and diastolic values of LV pressure.

RNA Extraction and Purification.

Pieces of left ventricular tissue from 15 rats with cirrhosis and ascites and from 15 control rats were minced using a mortar in order to obtain fractions smaller than 30 mg, which were then homogenized using a steal bead beater homogenizer (Qiagen, Jesi, Italy). Total RNA extraction was performed using the RNAeasy Fibrous Tissue kit (Qiagen). RNA quantification and purity were determined by spectrophotometric measurement (260/280 nm). RNA integrity was checked with a 2100 BioAnalyzer using RNA nanolabChips (Agilent Technologies, Cernusco, Italy).

Real-Time Reverse-Transcription Polymerase Chain Reaction (RT-PCR).

First-strand cDNA was synthesized with 1 μg of RNA extracted using the iScript cDNA synthesis kit (Bio-Rad, Hercules, CA) according to the manufacturer's instructions. A quantitative real-time PCR assay was performed in a Thermal Cycler (iCycler, Bio-Rad). Briefly, 2 μL complementary DNA (cDNA) was amplified in a real-time PCR reaction containing 400 nmol of each primer and 5× SYBR Green SuperMix (Bio-Rad). All reactions were performed in 96-well plates in triplicate. A negative control containing all reagents but no cDNA template was included in all runs. Real-time PCR was performed following the thermal protocol: 95°C for 3 minutes to denature, 45 cycles each consisting of 30 seconds at 95°C to denature and 1 minute at 60°C for annealing and extension. Primers were designed from sequences derived from the GenBank database using Primer 3 (provided by the Whitehead Institute, Cambridge, MA) and Operon's Oligo software (Alameda, CA) and were purchased from Eurofins (Ebersberg, Germany). Primer sequences were the following: beta1-adrenergic receptor (β1-AR), 5′-ag agcagaaggcgctcaag-3′ (forward) and 5′-agccagcagagcgtgaac-3′ (reverse); beta-2 adrenergic receptor (β2-AR), 5′-cctcactggtcaagtattaaggataa-3′ (forward) and 5′-tccaagg gtacaggaagaaaac-3′ (reverse); Gαi2 protein (Gαi2), 5′-tcaa tgactcagccgcttac-3′ (forward) and 5′-gggatatag tcactctgtgctatgc-3′ (reverse); Gαs protein (Gαs), 5′-cagtggttggaagc agtccttgc-3′(forward) and 5′-agcaggagagccagaggag-3′(reverse); adenylate cyclase 3 (Adcy3), 5′-gccttagagaagatgcaggt-3′ (forward) and 5′-acagtcatcgagtacttgggaag-3′ (reverse); β-actin, 5′-ccgcgagtacaaccttct-3′ (forward) and 5′-cgtcatccatggcgaact-3′ (reverse), glyceraldehyde 3-phosphate dehydrogenase (GAPDH), 5′-tcaccaccatggagaaggc-3′ (forward) and 5′-gctaagcagttggtggt gca-3′ (reverse) and hypoxanthine guanine phosphoribosyl transferase (HPRT), 5′-ggtccattcctatgactgtagatttt-3′ (forward) and 5′-caatcaagacgttctttccagtt-3′ (reverse). β-Actin, GAPDH, and HPRT were used as housekeeping genes. Data analyses were performed with the iQ Optical System Software (Bio-Rad). The comparative cycle threshold method (ΔΔCt), which compares the difference in cycle threshold values between groups, was used to obtain the relative fold change in gene expression as described.18

Quantification of messenger RNA (mRNA) included normalization to HPRT level. Furthermore, we used two additional housekeeping genes for normalization: GAPDH and β-actin. No difference was found in the gene expression of these two housekeeping genes when compared to the HPRT normalization.

Protein Extraction.

Fifty mg of heart tissue was homogenized in a hypotonic lysis buffer (20 mmol HEPES, 2 mmol ethylene glycol tetraacetic acid [EGTA], 10 mmol β-glycerophosphate, 1 mmol dithiothreitol [DTT], 2 mmol vanadate, 10 μg/mL phenylmethylsulfonylfluoride [PMSF], 1 μg/mL leupeptin, 5 μmol aprotinin). The homogenate was then centrifuged at 10,000 rpm for 10 minutes at 4°C. The supernatant was frozen in liquid nitrogen and stored at −80°C until use. Protein concentration was determined using Lowry's method, using bovine serum albumin (BSA) as the standard.

Western Blot Analysis.

Protein samples (30 μg) were separated by SDS-PAGE (sodium dodecyl sulfate, polyacrylamide gel electrophoresis) using a 10% polyacrylamide gel as described.19 Proteins separated in the gel were electroblotted onto nitrocellulose membrane (Hybond ECL, Amersham Biosciences, Amersham, UK) in blotting solution containing 48 mmol/L Tris, 39 mmol/L glycine, 0.037% SDS, and 20% v/v methanol for 2 hours at 100 V at 4°C, using a Mini Protean 3 Electrophoresis System (Bio-Rad). The membranes were blocked overnight at 4°C in T-PBS containing phosphate-buffered saline (PBS), 0.05% v/v Tween, and 5% BSA. Subsequently, membranes were exposed to anti-β1-AR (1:1,000 dilution) and anti-β2-AR (1:1,000 dilution), anti-Gαi2 (1:3,000 dilution), anti-Gαs (1:1,000 dilution) anti-iNOS (1:1,000 dilution), anti-Adcy3 (1:1,000 dilution), anti TNF-α (1:1,000 dilution), or anti-GAPDH (1:5,000 dilution) primary antibody overnight at 4°C (Tebu-bio, Santa Cruz Biotechnology, Santa Cruz, CA). The membranes were washed (3 times, 10 minutes each) in T-PBS and then incubated with horseradish peroxidase-conjugated secondary antibody (1:10,000). Detection was achieved using an enhanced chemiluminescence system (Pierce Biotechnology, Rockford, IL). The blots were scanned and quantified using a chemiluminescence molecular imaging system (Versa Doc 3000, Bio-Rad). The results were expressed relative to the control(s) on the same blot, which were defined as 100%, and by the protein of interest/GAPDH densitometric ratio.

Oxidative Stress.

Oxidative stress in the cardiac tissue was evaluated by means of activation of nicotinamide adenine dinucleotide phosphate (NAD(P)H) oxidase. In short, NAD(P)H oxidase is a complex enzyme that catalyzes the production of superoxide from oxygen and NAD(P)H, consisting of two membrane-bound components and three components in the cytosol, plus Rac-1. Activation of the oxidase involves the phosphorylation of the cytosolic components (Rac-1 and p47-phox) and their translocation on the cellular membrane. The Rac-1 and p47-phox translocation to plasma membrane was evaluated as follows: 50 mg of heart tissue was homogenized in a hypotonic lysis buffer (12.5 mM Tris, 2 mM EGTA, 25 mM β-glycerophosphate, 2 mM Na3VO4, 10 μM PMSF, 1 μM leupeptin, 5 μM aprotinin). The homogenates were centrifuged at 15,000g for 20 minutes at 4°C and then the supernatant was ultracentrifuged at 100,000g for 1 hour at 4°C. The pellet was defined as the membrane fraction and the supernatant as the cytosolic fraction and subjected to western blot analysis for Rac-1 and p47-phox using polyclonal antibodies (1:1,000, Santa Cruz Biotechnology).19

TNF-α Quantification by Enzyme-Linked Immunosorbent Assay (ELISA).

Fifty μL of plasma and ascitic liquid were loaded on the plate of an ELISA kit (Bender MedSystems, San Diego, CA, distributed in Italy by Inalco, Milan, Italy). The limit of detection of rat TNF-α was defined to be 11 pg/mL by means of six independent assays. The samples were performed in triplicate.

NF-κB P-65 Extraction and Quantification by ELISA.

Nuclear extracts were prepared with Nuclear and Cytoplasmic Extraction Reagents (Pierce, Rockford, IL). An ELISA kit for the NF-κB assay was purchased from Cayman (Temecula, CA, distributed in Italy by Prodotti Gianni, Milan, Italy). The ELISA was carried out according to the manufacturer's instructions.

Statistical Analysis.

Data are presented as the mean ± standard deviation (SD). All statistical analyses were performed using SPSS 10.0 for Windows (Chicago, IL). Given the fact that there were four groups of animals, to compare the differences between the groups data were analyzed by one-way analysis of variance (ANOVA) followed by Bonferroni-adjusted P values. P < 0.05 was accepted as statistically significant.

Results

Cardiac Contractility

Figure 1A shows the dose-response curves to isoproterenol (from 10−10 M to 10−8 M) of the left ventricular contractility of control rats treated with saline, HES, or albumin. No difference was observed between the three groups of animals.

Figure 1.

Dose-response curves of isoproterenol-stimulated LVDP in an isolated left ventricular heart. The heart was mounted in a Langendorff apparatus. LVDP was calculated as the difference between systolic and diastolic values of left ventricular pressure, and its value is expressed as a percentage of the maximal effect. (A) Differences in control rats according to treatment with albumin (n = 15) (▪), saline (n = 15) (▴), or HES (n = 15) (○). (B) Differences in rats with cirrhosis and ascites according to treatment with albumin (n = 15) (▪) saline (n = 15) (▴) or HES (n = 15) (○). *P < 0.01.

As shown in Fig. 2A, the increase of left ventricular contractility induced by isoproterenol was markedly reduced in rats with cirrhosis and ascites compared to control rats (P < 0.01). The maximal response to the drug was about 3-fold lower than that observed in control rats. Albumin significantly increased cardiac contractility from 5.6*10−10 M to 10−8 M in rats with cirrhosis and ascites (P < 0.01) (Fig. 1B), but not in control rats (Fig. 1A). As a consequence, any difference in left ventricular contractility was no longer detectable between the two groups of animals during treatment with albumin (Fig. 2B).

Figure 2.

Dose-response curves of isoproterenol-stimulated LVDP in an isolated left ventricular heart. The heart was mounted in a Langendorff apparatus. LVDP was calculated as the difference between systolic and diastolic values of left ventricular pressure and its value is expressed as percentage of the maximal effect. (A) Differences between control rats (n = 15) (▪) and rats with cirrhosis and ascites (n = 15) (□) after treatment with albumin. *P < 0.01. (B) Differences between control rats (n = 15) (▴) and rats with cirrhosis and ascites (n = 15) (•) after treatment with saline. (C) Differences between control rats (n = 15) (▴) and rats with cirrhosis and ascites (n = 15) (•) after treatment with HES.

The effect of saline on PRA seems to be less as compared to that of albumin or HES (Fig. 3), even if the difference, even in terms of percent change (data not shown), did not reach statistical significance. However, neither saline nor HES had any effect on cardiac contractility in rats with cirrhosis and ascites (Figs. 1A,B, 2B,C).

Figure 3.

PRA before and after the administration of albumin (n = 15), saline (n = 15), or HES (n = 15). *P < 0.05 in the comparison between the basaline and the post-treatment values within each group of rats; no difference was observed in the baseline values as well as in the posttreatment values of PRA between the three groups of rats.

Effects of Albumin Treatment on β-Adrenergic System

Gene Expression.

Figure 4 reports the gene expression of β1-AR, β2-AR, Gαi2, Gαs, and Adcy3 in the heart of animals after administration of saline or albumin. The administration of either saline or albumin did not induce any change in control rats, whereas albumin profoundly changed the gene expression of Gαi2 and Adcy3 in rats with cirrhosis and ascites (P < 0.05) (Fig. 4).

Figure 4.

Real-time PCR of β1-AR, β2-AR, Gαi2, Gαs, and Adcy3 in the left ventricle from control rats (n = 15) and from rats with cirrhosis and ascites (n = 15) after treatment with saline (S) and control rats (n = 15) and rats with cirrhosis and ascites (n = 15) after treatment with albumin (A). The relative quantification was achieved by normalizing the expression of the gene of interest to HPRT by the ΔΔCt method. Error bars represent SD, *P < 0.05 versus control; #P < 0.05 versus rats with cirrhosis treated with saline solution.

After the administration of saline, β2-AR, and Gαi2 gene expression was significantly increased in rats with cirrhosis and ascites as compared to control rats (P < 0.05). After albumin treatment the gene expression of the Gαi2 in rats with cirrhosis and ascites was reduced to the level observed in control animals, whereas β2 adrenergic remained unchanged (P < 0.05) (Fig. 4).

Gene expression of Gαs is not influenced by the administration of saline or albumin in controls animal or in rats with cirrhosis and ascites. After saline, the gene expression of Adcy3 in rats with cirrhosis and ascites was lower than in control rats (P < 0.05). Albumin administration brought gene expression of Adcy3 back to the level observed in control animals (P < 0.05).

Protein Expression.

Figure 5 reports the protein expression of β1-AR, β2-AR, Gαi2, Gαs, and Adcy3 in the heart of animals after the administration of saline or albumin. According to gene expression, after saline β2-AR protein expression was significantly increased in rats with cirrhosis and ascites as compared with control rats (P < 0.05). Albumin administration did not modify this difference.

Figure 5.

Protein expression of β1-AR, β2-AR, Gαi2, Gαs, and Adcy3 by western blot in the heart from control rats (n = 15) and from rats with cirrhosis and ascites (n = 15) treated with saline (S) and control rats (n = 15) and rats with cirrhosis and ascites (n = 15) after treatment with albumin (A). The proteins were separated by electrophoresis and immunoblotted. The upper panels report the representative immunoblots of β1-AR, β2-AR, Gαi2, Gαs, Adcy3, and the corresponding GAPDH. The lower panels report the densitometric analysis after normalization for GAPDH. Data are expressed as β1-AR/GAPDH, β2-AR/GAPDH, Gαi2/GAPDH, Gαs/GAPDH, and Adcy3/GAPDH and normalized to control rats treated with saline. *P < 0.05 versus control; #P < 0.05 versus rats with cirrhosis treated with saline solution. CTR: control rats; CIRR: cirrhotic rats.

i2 protein expression was significantly increased in animals with cirrhosis and ascites compared with control rats after saline (P < 0.05). Albumin administration significantly reduced this difference (P < 0.05) (Fig. 5).

Administration of saline or albumin did not influence protein Gαs expression in either group of animals. After saline, the expression of Adcy3 in rats with cirrhosis and ascites was lower compared with control animals (P < 0.05). Albumin administration returned the gene expression of Adcy3 to levels similar to those of controls (P < 0.05).

Effects of Albumin on Oxidative Stress

Activation of NAD(P)H oxidase was detected through the translocation of two cytosolic subunits (Rac-1 and p47-phox) to the plasmatic membrane. Figure 6 shows that the increased p47-phox and Rac-1 membrane/cytosol ratio, which was observed after saline administration in rats with ascites as compared to control animals (P < 0.05), was almost completely normalized after albumin administration (P < 0.05).

Figure 6.

The Western blot of membrane (Mem) and cytosolic (Cyt) fraction of NAD(P)H oxidase subunits, p47-phox and RAC-1 from control rats (n = 15) and rats with cirrhosis and ascites (n = 15) treated with saline (S) and control rats (n = 15) and rats with cirrhosis and ascites (n = 15) after treatment with albumin (A). The membrane fractions were separated by ultracentrifugation at 100,000g for 1 hour at 4°C. The proteins were separated by electrophoresis and immunoblotted. The upper panels report the representative immunoblots of p47-phox and RAC-1. Data are expressed as membrane fraction/cytosolic fraction and normalized to control rats treated with saline. *P < 0.05 versus control; #P < 0.05 versus rats with cirrhosis with saline solution.

Effects of Albumin on Nuclear Translocation of NF-κB

After saline, NF-κB translocation, measured in cardiac tissue after nuclear extraction, was significantly increased in animals with cirrhosis and ascites as compared to control animals (P < 0.05) (Fig. 7A). Albumin administration decreased NF-κB activation in rats with cirrhosis and ascites to a level similar to that observed in control rats (P < 0.05).

Figure 7.

Myocardial levels of NF-κB determined by ELISA and the protein expression of iNOS by western blot in control rats (n = 15) and rats with cirrhosis and ascites (n = 15) treated with saline (S) and control rats (n = 15) and rats with cirrhosis and ascites (n = 15) after treatment with albumin (A). The proteins were separated by electrophoresis and immunoblotted. The upper panels report the representative immunoblots of iNOS. The lower panels report the densitometric analysis after normalization for GAPDH. Data are expressed as iNOS/GAPDH and normalized to control rats treated with saline. *P < 0.05 versus control; #P < 0.05 versus rats with cirrhosis treated with saline solution. CTR: control rats; CIRR: cirrhotic rats.

Effects of Albumin on iNOS Expression

Figure 7B shows that the increased protein expression of iNOS, which was observed after saline administration in rats with cirrhosis and ascites (P < 0.05), was brought back to the level observed in control animals by albumin administration (P < 0.05).

Effects of Albumin on TNF-α Concentration

Figure 8A shows that in rats with cirrhosis and ascites the plasma level of TNF-α was significantly increased 2-fold as compared to control animals (P < 0.05). Albumin reduced the plasma level of TNF-α in these animals to that observed in control rats (P < 0.05). Likewise, albumin administration caused a significant decrease of TNF-α concentration in the ascitic fluid in rats with cirrhosis (P < 0.05).

Figure 8.

(A) Plasma and ascitic fluid levels of TNF-α in control rats (n = 15) and rats with cirrhosis and ascites (n = 15) treated with saline (S) and control rats (n = 15) and rats with cirrhosis and ascites (n = 15) after treatment with albumin (A) analyzed with an ELISA kit. (B) Protein expression of TNF-α by western blot in the heart from control rats (n = 15) and from rats with cirrhosis and ascites (n = 15) treated with saline (S) and control rats (n = 15) and rats with cirrhosis and ascites (n = 15) after treatment with albumin (A). The proteins were separated by electrophoresis and immunoblotted. The upper panels report the representative immunoblots of TNF-α and the corresponding GAPDH. The lower panels report the densitometric analysis after normalization for GAPDH. Data are expressed as TNF-α/GAPDH and normalized to control rats treated with saline. *P < 0.05 versus control; #P < 0.05 versus rats treated with cirrhosis with saline solution. CTR: control rats; CIRR: cirrhotic rats.

Figure 8B reports protein expression TNF-α in the heart of animals after the administration of saline or albumin analyzed with western blot. After saline injection the cardiac levels of this cytokine is higher in rats with cirrhosis than in control rats. After albumin administration, in rats with cirrhosis and ascites TNF-α levels were brought back to levels observed in control animals (P < 0.05).

Molecular Effects of HES

HES had no effect on NF-κB translocation, membrane, and cytosol ratio of P47-phox and Rac 1, and protein expression of β1-AR, β2-AR, Gαi2, Gαs Adcy3, and iNOS in the cardiac tissue of rats with cirrhosis and ascites (data not shown).

Discussion

The main result of our study is the observation that the intravenous infusion of albumin almost normalizes the defect in cardiac contractility which can be detected in rats with cirrhosis with ascites (Figs. 1B, 2A). As this action was evaluated ex vivo, it should be considered part of the mechanism by which albumin can increase the cardiac output in cirrhosis, together with the increase in plasma volume. It seems to be mediated by two molecular pathways: (1) a blunting effect on the overexpression and an overactivity of iNOS (Fig. 7A), and (2) a blunting effect on the enhancement of β-receptors-inhibitory G-protein (Gi-protein) signaling pathway in the cardiac tissue of these animals (Figs. 4, 5). With regard to the first molecular pathway, it has been recently shown that the increased synthesis of NO in the cardiac tissue of bile duct-ligated (BDL) mice is related to an increased level of TNF-α.3 Furthermore, it has been observed that the genetic deletion, as well as the pharmacological inhibition of TNF-α, decreased NO levels in BDL mice, and this change was accompanied by the correction of cardiomyocyte contractile dysfunction.20 Although we did not perform experiments based on the inhibition of the release of TNF-α, according to these observations it can be hypothesized that the positive inotropic effect of albumin observed in our study was associated with its capacity to bind serum TNF-α and also to blunt the overexpression of TNF-α in the cardiac tissue of rats with cirrhosis. Previous studies have shown that the overexpression of TNF-α in the cardiac tissue of rats with cirrhosis can be related to two main factors (1) oxidative stress20, 21 and (2) an increased nuclear translocation of NF-κB.8 In turn, TNF-α is capable of inducing oxidative stress in adult rat cardiomyocytes and of further triggering NF-κB activity. Compatible with these observations, the membrane translocation of p47-phox and Rac-1 (Fig. 6), an index of oxidative stress, and the nuclear translocation of NF-κB (Fig. 7A) were found to be significantly increased in rats with cirrhosis as compared with control rats. Albumin infusion resulted in a significant reduction of both the membrane translocation of p47-phox and Rac-1 (Fig. 6) and the increased nuclear translocation of NF-κB (Fig. 7A) in rats with cirrhosis. Taking into account previous observations in an experimental model of sepsis,15 it may be assumed that through these two effects albumin infusion could have reduced the protein expression of iNOS (Fig. 7B) in the cardiac tissue of rats with cirrhosis and ascites

With regard to the second pathway, the important role of oxidative stress should be taken into account in attenuating the beta-adrenoceptor-linked signal transduction through its effect on G-proteins, either stimulatory (Gs-proteins) or inhibitory (Gi-proteins)22 and/or on adenilate cyclase enzymes, which have been observed either in chronic heart failure or in cardiac ischemia-reperfusion injury.23 The results of our study, although confirming a reduced gene and protein expression of Adcy 3 in the cardiac tissue of rats with cirrhosis,24 showed that the reduced expression of Adcy 3 was corrected by the administration of albumin. The interpretation Adcy3 expression before and after administration of albumin cannot be separated from that for Gαi2 or for Gαs. In fact, the increased RNA and protein expression of Gαi2 that has been observed in baseline conditions in the cardiac tissue of rats with cirrhosis surely contributes to the inhibition of Adcy 3. Gi-proteins, including Gαi2, is coupled to β2-AR, which, when stimulated, can also induce an enhancement of the β-receptors-Gi-protein signaling pathway.24, 25 In our study an increased expression of both gene and protein expression of β2-AR was detected in the cardiac tissue of rats with cirrhosis as compared to control rats (Figs. 4, 5). Therefore, an overexpression of Gi-proteins due to exaggerated β2-AR signaling can be hypothesized as contributing to the reduced contractility in rats with cirrhosis, as in an experimental model of a decompensated failing heart.26, 27 Nevertheless, the expression of β2-AR was not significantly changed by albumin infusion, whereas, as discussed previously, the increased expression of Gαi2 observed in baseline conditions in rats with cirrhosis was almost normalized (Figs. 4, 5). Consequently, our results suggest that the effect of albumin on β-AR signal transduction is not related to a change in the expression of β1-AR and/or β2-AR, but to a blunting effect on the expression of Gαi2, probably mediated by the effect of albumin on oxidative stress.

Taken together, these results led us to confirm and consolidate the hypothesis that albumin improves cardiac contractility (1) by reducing the negative inotropic effect of the NF-κB-iNOS-NO pathway and (2) by blunting the oxidative stress-mediated overexpression of Gi-proteins and down-expression of Adcy3.

The last two points are the most critical ones because we may wonder what the origin of the increased systemic availability of TNF-α in rats with cirrhosis may be, and how albumin exerts its effects in the cardiomyocytes of rats with cirrhosis. The answers to these questions are closely interlinked. First, as previously observed, in an experimental model of cirrhosis bacterial translocation may induce a systemic inflammatory response with levels of TNF-α that are similar to those observed in animals with an infection of the ascitic fluid.29 Furthermore, it has already been observed that by lowering the systemic availability of TNF-α by means of the intraperitoneal administration of a specific antibody it is possible to counteract oxidative stress, as well as the overexpression of iNOS triggered by this cytokine in the cardiac tissue of BDL mice.20 Therefore, the effects of albumin infusion in the cardiomyocytes of rats with cirrhosis may be related to its capacity to bind TNF-α in the systemic circulation,28 blunting the effects of this cytokine in the cardiac tissue. To confirm this, in our study albumin significantly reduced the levels of TNF-α in plasma and ascites in rats with cirrhosis. In addition, if the trend towards a lower efficacy of saline with plasma expander than albumin (Fig. 3) still leaves the possibility to assume that the effect of albumin on cardiac contractility was also indirectly linked to its potential larger effect on plasma volume, then the data obtained with HES on cardiac contractility seem to exclude this hypothesis definitively.

In conclusion, the results of the study support the view that albumin exerts a positive inotropic effect in rats with cirrhosis and this effect is independent of the effect of albumin on plasma volume. The modality of action of albumin in the cardiac tissue of cirrhotic rats is complex and involves its capacity to counteract the negative effects of both TNF-α and oxidative stress on cardiac contractility.

Acknowledgements

The authors thank Mrs. Daniela Cinquemani for technical assistance.

Author Contributions: A.B., acquisition of data and drafting of the article; G.C., analysis and interpretation of data; E.G., S.B., S.F., acquisition of data; A.S., technical support; F.M., statistical analysis; S.P., critical revision of the article; S.R., technical support and art work; A.G., study supervision; P.A., study concept and design and writing the article.

Ancillary