Potential conflict of interest: Nothing to report.
Supported by grants from the Ministerio de Economía y Competitividad (SAF 2010/17043 and ACI2009-0938) and from Instituto de Salud Carlos III (FIS PS09/01261 and FIS PI11/00235). Ciberehd is funded by Instituto de Salud Carlos III. Jorge Gracia-Sancho has a Ramón y Cajal contract from the Ministerio de Economía y Competitividad.
Address reprint requests to: Juan Carlos García-Pagán, Hepatic Hemodynamic Laboratory, Liver Unit, Hospital Clínic, Villarroel 170, 08036 Barcelona, Spain. E-mail: firstname.lastname@example.org; fax: +34-93-2279856.
Increased production of vasoconstrictive prostanoids, such as thromboxane A2 (TXA2), contributes to endothelial dysfunction and increased hepatic vascular tone in cirrhosis. TXA2 induces vasoconstriction by way of activation of the thromboxane-A2/prostaglandin-endoperoxide (TP) receptor. This study investigated whether terutroban, a specific TP receptor blocker, decreases hepatic vascular tone and portal pressure in rats with cirrhosis due to carbon tetrachloride (CCl4) or bile duct ligation (BDL). Hepatic and systemic hemodynamics, endothelial dysfunction, liver fibrosis, hepatic Rho-kinase activity (a marker of hepatic stellate cell contraction), and the endothelial nitric oxide synthase (eNOS) signaling pathway were measured in CCl4 and BDL cirrhotic rats treated with terutroban (30 mg/kg/day) or its vehicle for 2 weeks. Terutroban reduced portal pressure in both models without producing significant changes in portal blood flow, suggesting a reduction in hepatic vascular resistance. Terutroban did not significantly change arterial pressure in CCl4-cirrhotic rats but decreased it significantly in BDL-cirrhotic rats. In livers from CCl4 and BDL-cirrhotic terutroban-treated rats, endothelial dysfunction was improved and Rho-kinase activity was significantly reduced. In CCl4-cirrhotic rats, terutroban reduced liver fibrosis and decreased alpha smooth muscle actin (α-SMA), collagen-I, and transforming growth factor beta messenger RNA (mRNA) expression without significant changes in the eNOS pathway. In contrast, no change in liver fibrosis was observed in BDL-cirrhotic rats but an increase in the eNOS pathway. Conclusion: Our data indicate that TP-receptor blockade with terutroban decreases portal pressure in cirrhosis. This effect is due to decreased hepatic resistance, which in CCl4-cirrhotic rats was linked to decreased hepatic fibrosis, but not in BDL rats, in which the main mediator appeared to be an enhanced eNOS-dependent vasodilatation, which was not liver-selective, as it was associated with decreased arterial pressure. The potential use of terutroban for portal hypertension requires further investigation. (Hepatology 2013;58:1424–1435)
In cirrhotic livers, increased resistance to portal blood flow resulting from architectural alterations of the liver parenchyma as well as from increased hepatic vascular tone is the primary factor in the pathophysiology of portal hypertension.[1, 2] Increased hepatic vascular tone is partly due to an increased production of cyclooxygenase-1 (COX-1)-derived vasoconstrictive prostanoids, such as thromboxane (TXA2)[3, 4] together with an insufficient intrahepatic availability of the vasodilator nitric oxide (NO).[5, 6]
We have previously demonstrated that, in isolated perfused cirrhotic livers, the blockade of the TXA2/PGH2 (TP) receptor with SQ29548 corrected the hyperresponse to methoxamine and improved endothelial dysfunction of the hepatic vascular bed. Moreover, sinusoidal endothelial cells (SEC) isolated from cirrhotic rats overexpress COX-1 and thromboxane synthase (TXAS), which represent an important source of vasoconstrictor prostanoids, such as TXA2. Importantly, COX inhibition not only reduces the exaggerated TXA2 production of cirrhotic SEC but also restores, at least in part, its decreased NO bioavailability.
TP receptor ligands include TXA2, PGH2, and isoprostanes.[10, 11] TXA2 acts through its G-protein-coupled receptor leading to vasoconstriction by activating the RhoA/Rho-kinase pathway, and by increasing calcium levels in hepatic stellate cells (HSC). Terutroban is an orally active, specific antagonist of the TP-receptor that improves endothelial-dependent vasodilation, reduces inflammation, attenuates oxidative stress, and exerts antifibrotic effects[16, 17] in different vascular disorders. In addition, terutroban has been shown to reduce RhoA/Rho-kinase-dependent signaling and restore NO bioavailability in endothelial cells.[18, 19]
The current study aimed at evaluating the long-term effects of the in vivo blockade of TP receptor with terutroban in two experimental rat models of cirrhosis, carbon tetrachloride (CCl4) and bile duct ligation (BDL).
Materials and Methods
Induction of Cirrhosis by CCl4
Male Wistar rats weighing 50 to 75 g underwent inhalation exposure to CCl4 three times a week as described. A high yield of micronodular cirrhosis was obtained after ∼12 to 15 weeks of CCl4 inhalation. When the cirrhotic rats developed ascites, administration of CCl4 was stopped.
Induction of Cirrhosis by Bile Duct Ligation (BDL)
Secondary biliary cirrhosis was induced in male Sprague-Dawley rats (200 to 225 g) by BDL as described.
Cirrhotic rats were randomized to receive terutroban (30 mg/kg; n = 13 in CCl4; n = 14 in BDL; kindly supplied by Servier, Courbevoie Cedex, France) or its vehicle (1% hydroxyethylcellulose; n = 8 in CCl4; n = 16 in BDL; Sigma-Aldrich, Tres Cantos, Madrid, Spain), administered orally by gavage, once a day for 2 weeks. Treatment started 1 week after development of ascites and stopping CCl4 administration in a setting of advanced cirrhosis or after 2 weeks of BDL, in a precirrhotic stage. Experiments were performed 1 hour after the last dose of terutroban or vehicle. Treatments were prepared by a third person and experimental studies were realized blindly. The code was kept sealed until the final analysis of the results. The dose of terutroban used has been previously shown to have antivasoconstricting and antiatherosclerotic properties.[16, 22, 23]
The animals were kept in environmentally controlled animal facilities at the Institut d'Investigacions Biomèdiques August Pi i Sunyer. All procedures were approved by the Laboratory Animal Care and Use Committee of the University of Barcelona and were conducted in accordance with European Community guidelines for the protection of animals used for experimental and other scientific purposes (EEC Directive 86/609).
In Vivo Hemodynamic Studies
Cirrhotic rats were anesthetized with intraperitoneal ketamine hydrochloride (100 mg/kg; Merial Laboratories, Barcelona, Spain) plus midazolam (5 mg/kg intraperitoneally; Laboratorios Reig Jofré, Barcelona, Spain). The femoral artery and the ileocolic vein were cannulated with PE-50 catheters to measure mean arterial pressure (MAP; mmHg) and portal pressure (PP; mmHg), respectively. Perivascular ultrasonic transit-time flow probes connected to a flow meter (Transonic Systems, Ithaca, NY) were placed around the portal vein, as close as possible to the liver to measure portal blood flow perfusing the liver (PBF; mL/min/g liver) and around the superior mesenteric artery, in BDL cirrhotic rats, to measure superior mesenteric artery blood flow (SMABF, mL/min/100g body weight). Hepatic vascular resistance (HVR, mmHg/mL/min/g liver) was calculated as: PP/PBF; and superior mesenteric artery resistance (SMAR, mmHg/mL/min/100g body weight) was calculated as (MAP-PP)/SMABF. Blood pressures and flows were registered on a multichannel computer-based recorder (PowerLab; AD Instruments, Colorado Springs, CO). The temperature of the animals was maintained at 37 ± 0.5°C. Hemodynamic data were collected after a 20-minute stabilization period.
Efficacy of TP Receptor Blockade
To determine if terutroban correctly blocked the TP receptor in a subgroup of CCl4 and BDL cirrhotic rats (n = 3) treated with terutroban (30 mg/kg) or vehicle for 2 weeks, measurements of MAP and PP were performed before and after the intravenous infusion of 10 μg/kg U46619. U46619 (9,11-dideoxy-9α,11α-methanoepoxy-prosta-5Z,13E-dien-1-oic acid; Cayman Chemical, Tallin, Estonia) is a synthetic TXA2 analog that specifically activates the TP-receptor.
Evaluation of Endothelial Function
An additional group of cirrhotic rats were randomized to receive terutroban (30 mg/kg; n = 8 in CCl4; n = 8 in BDL) or vehicle (n = 9 in CCl4; n = 9 in BDL) for 3 days. Rats were anesthetized and livers were quickly isolated and perfused by a flow-controlled perfusion system as described. The perfused rat liver preparation was allowed to stabilize for 20 minutes before vasoactive substances were added. The intrahepatic microcirculation was preconstricted by adding the α1-adrenergic agonist methoxamine (Mtx; 10−4 mol/L; Sigma) to the reservoir. After 5 minutes concentration-response curves to cumulative doses of acetylcholine (Ach; 10−7, 10−6, and 10−5 mol/L; Sigma) were evaluated. Responses to Ach were calculated as percent change in portal perfusion pressure. The gross appearance of the liver, stable perfusion pressure, bile production over 0.4 μL/min/g of liver, and a stable buffer pH (7.4 ± 0.1) were monitored during this period. If any viability criteria were not satisfied, the experiment was discarded.
At the end of the in vivo hemodynamic study, serum samples from cirrhotic rats were collected by cardiac puncture to subsequently evaluate alanine aminotransferase (ALT), aspartate aminotransferase (AST), bilirubin, and albumin, all by standard protocols.
Assessment of Rho-kinase Activity
Hepatic samples were obtained as described. Rho-kinase activity was assessed by the phosphorylation of the endogenous Rho-kinase substrate, moesin at Thr558 normalized to the level of total moesin expression. Moesin-phosphorylation and moesin total expression were assessed by western blot using a goat antiphosphorylated moesin at Thr558 antibody (1:200; Santa Cruz Biotechnology, Santa Cruz, CA) and a mouse antibody recognizing moesin (1:200; Santa Cruz Biotechnology) overnight at 4°C followed by incubation with their associated horseradish peroxidase-conjugated secondary antibody (1:10,000, 1 hour, room temperature, Stressgen, Victoria, BC, Canada). After stripping, blots were assayed for glyceraldehyde-3-phosphate dehydrogenase (GAPDH) expression as standardization of sample loading.
Evaluation of NO Pathway
Western Blot Analysis of eNOS-Phosphorylation and eNOS Total Protein Expression
eNOS-phosphorylation (eNOS-P) and eNOS total expression were assessed in hepatic homogenates using a rabbit antiphosphorylated eNOS at Ser1176 (1:1,000; Cell Signaling Technology, Beverly, MA) and a mouse antibody recognizing eNOS (1:1,000; BD Transduction Laboratories, Lexington, KY). Quantitative densitometric values of proteins were normalized to GAPDH.
Measurements of cGMP, a secondary marker of NO bioavailability, were performed in rat liver homogenates treated with terutroban or vehicle by enzyme immunoassay (Cayman Chemical) as described.
Evaluation of Hepatic Fibrosis
Quantification of Hepatic Fibrosis
Livers from cirrhotic rats were fixed in 10% formalin, embedded in paraffin, sectioned, and stained with 0.1% Sirius red, photographed, and analyzed using a microscope equipped with a digital camera. Eight fields from each slide were randomly selected, and the red-stained area per total area was measured using AxioVision software. Values are expressed as the mean of eight fields taken from vehicle- (n = 9) and terutroban- (n = 11) CCl4-cirrhotic rats or n = 10 animals per group in the BDL model.
Protein Expression of α-Smooth Muscle Actin (α-SMA)
Hepatic protein expression of α-SMA was determined by western blot in hepatic samples using a mouse antibody against α-SMA (1:1,000, Sigma-Aldrich).
Collagen I and Transforming Growth Factor Beta (TGF-β) Gene Expression
Hepatic messenger RNA (mRNA) expression of Collagen I and TGF-β was analyzed by real-time polymerase chain reaction (PCR) using predesigned gene expression assays obtained from Applied Biosystems (AB, Foster City, CA) according to the manufacturer's protocol and reported relative to endogenous control GAPDH. All PCR reactions were performed in duplicate and using nuclease-free water as no template control.
Analysis of TP Receptor Expression in HSC
To directly study the expression of TP receptor, hepatic stellate cells (HSC) were isolated from control, CCl4-, sham-operated, and BDL-rats as described. In the CCl4 model, HSC were isolated 1 week after development of ascites and in the BDL model 2 weeks after surgery (when terutroban treatment was initiated in the in vivo studies). TP receptor protein expression was determined by western blot in hepatic samples using a mouse antibody against TP receptor (1:1,000; Cayman Chemical).
Statistical analysis was performed with SPSS 18.0 for Windows (IBM, Armonk, NY). All results are expressed as mean ± SEM. Comparisons between groups were performed with the Student t test for unpaired data or with Mann-Whitney test when assumptions of normality could not be verified. Differences were considered significant at P < 0.05.
Efficacy of TP Receptor Blockade in CCl4 and BDL Cirrhotic Rats
As expected, infusion of U46619 produced a significant increase in MAP (23 ± 13%) and PP (11 ± 5%) in CCl4-cirrhotic rats treated with vehicle (Fig. 1A,B, black bars). By contrast, in CCl4-cirrhotic rats treated with terutroban, the increase of MAP (3 ± 5%) and PP (2 ± 3%) in response to TP agonist was markedly attenuated, indicating an effective blockade of the TP-receptor (Fig. 1A,B, white bars). Terutroban produced a similar blockade of the TP-receptor in BDL-cirrhotic rats, as shown by the attenuation of the increase in MAP (5 ± 3% versus 18 ± 8% in vehicle) and in PP (3 ± 3% versus 12 ± 3% in vehicle) caused by U46619 (Fig. 1D,1E).
TP Receptor Protein Expression in CCl4 and BDL Cirrhotic Rats
TP receptor expression was determined in HSC from control, CCl4- (Fig. 1C), and BDL-cirrhotic rats (Fig. 1F). Both cirrhotic models exhibited a significantly higher TP receptor expression compared to control rats.
Effects of Chronic Treatment With Terutroban in CCl4-Cirrhotic Rats
TP Receptor Blockade Lowers Portal Pressure in CCl4-Cirrhotic Rats
PP was significantly lower in CCl4-cirrhotic rats treated with terutroban (11.9 ± 2.8 mmHg) as compared with vehicle-treated rats (14.5 ± 1.4 mmHg) (mean difference −17.9%; P = 0.035). This reduction was not associated with a significant change in PBF reflecting a fall in HVR (7.9 ± 2.6 versus 10.3 ± 2.9 mmHg/mL/min/g in vehicle-treated rats) (mean decrease 25%; P = 0.047). MAP was not significantly reduced by terutroban (Fig. 2).
TP Receptor Blockade Attenuates Rho-kinase Activity in CCl4-Cirrhotic Rats
To further explore the intrahepatic molecular mechanisms behind TP receptor blockade, we evaluated moesin phosphorylation in hepatic samples, a marker of Rho-kinase activity. TP receptor blockade with terutroban reduced hepatic moesin phosphorylation indicating a reduction in Rho-kinase activity (Fig. 3B).
Since RhoA/Rho-kinase may modulate eNOS activity, we characterized eNOS expression and phosphorylation. However, terutroban administration did not modify eNOS phosphorylation at Ser1176 (Fig. 3C), total eNOS expression (Fig. 3D), or hepatic cGMP levels (18.3 ± 2.9 pmol/mL versus 19.2 ± 3.4 pmol/mL in vehicle-treated rats) (Fig. 3E).
TP Receptor Blockade Attenuates Hepatic Fibrosis in CCl4-Cirrhotic Rats
As expected, CCl4-cirrhotic rats exhibited a marked distortion of the normal liver architecture, as identified by staining of liver sections with Sirius red. Terutroban treatment produced a significant reduction in hepatic fibrosis, measured by the percentage of fibrosis area on Sirius red-stained liver sections (13.7 ± 4% versus 20.8 ± 3% in vehicle-treated rats) (Fig. 4A). This was associated with a significant reduction in collagen I mRNA expression (Fig. 4B), a marked decrease in α-SMA protein expression, a surrogate marker of HSC activation (Fig. 4C), and decreased TGF-β mRNA levels (Fig. 4D).
Effects of TP Receptor Blockade on Liver Function in CCl4-Cirrhotic Rats
There were no significant differences in transaminases or bilirubin between CCl4-cirrhotic rats treated with vehicle or terutroban. However, albumin levels were significantly increased in terutroban-treated rats. Liver, spleen, and body weight were not different between groups (Table 1A).
TP Receptor Blockade Improves Endothelial Dysfunction in CCl4-Cirrhotic Rats
Improved vasorelaxation in response to Ach was observed in 3-day terutroban-treated rats in comparison to cirrhotic rats treated with vehicle, which exhibited the expected impaired vasodilatory response to Ach (endothelial dysfunction) (Fig. 3A). After NO synthase inhibition, terutroban also improved the vasodilatory response to Ach (Ach 10−7 M: −4.3 ± 0.3%; 10−6 M: −8.0 ± 1.5%; 10−5 M: −14.3 ± 2.4).
Table 1. Effects of Terutroban on Biochemical Parameters in CCl4-Cirrhotic Rats (1A) and in BDL-Cirrhotic Rats (1B)
Effects of Chronic Treatment With Terutroban in BDL-Cirrhotic Rats
TP Receptor Blockade Lowers Portal Pressure in BDL-Cirrhotic Rats
BDL cirrhotic rats treated with terutroban also had a significantly lower PP than those treated with vehicle (15.2 ± 1.9 versus 17.3 ± 2 mmHg; P = 0.007; mean difference −12.1%). Reduction in PP was observed without significant changes in PBF, supporting a reduction in HVR (17.8 ± 5.2 versus 22.8 ± 3.8 mmHg/mL/min/g; P = 0.038; mean decrease 22%). However, BDL rats treated with terutroban exhibited a significantly lower MAP (70 ± 8 mmHg versus 91 ± 16 mmHg; P < 0.05) than those receiving vehicle. As SMABF was similar in both groups, terutroban produced a significant reduction in splanchnic arteriolar resistance (Fig. 5).
TP Receptor Blockade Attenuates Rho-kinase Activity and Improves NO Bioavailability in BDL-Cirrhotic Rats
Moesin phosphorylation was significantly decreased in livers from terutroban-treated BDL rats (Fig. 6B). Contrary to CCl4-cirrhotic rats, livers from BDL rats treated with terutroban exhibited an enhanced eNOS phosphorylation at Ser1176 (Fig. 6C) and increased total eNOS expression (Fig. 6D), together with increased hepatic cGMP levels (7.2 ± 2.7 pmol/mL versus 4.1 ± 2.5 pmol/mL in vehicle-treated rats; P < 0.05) (Fig. 6E).
TP Receptor Blockade Does Not Attenuate Hepatic Fibrosis in BDL-Cirrhotic Rats
Contrary to CCl4-cirrhotic rats, terutroban administration to BDL rats did not reduce liver fibrosis as evaluated by the percentage of Sirius staining (36.9 ± 3.7% versus 34.7 ± 7.5% in vehicle) (Fig. 7A), and did not significantly change α-SMA protein expression (Fig. 7C), Type I collagen (Fig. 7B), or TGF-β mRNA levels (Fig. 7D).
Effects of TP Receptor Blockade on Liver Function in BDL-Cirrhotic Rats
There were no significant differences in transaminases, bilirubin, or albumin between BDL cirrhotic rats treated with vehicle or terutroban. Liver, spleen, and body weight were not different between groups (Table 1B).
TP Receptor Blockade Improves Endothelial Dysfunction in BDL-Cirrhotic Rats
Livers from cirrhotic rats treated with vehicle exhibited an impaired vasodilatory response to Ach. Terutroban treatment significantly improved vasorelaxation in response to Ach (Fig. 6A).
An increase in the hepatic production of the vasoconstrictor prostanoid TXA2 has been shown to increase hepatic resistance in cirrhotic livers, contributing to increased portal pressure.[3, 20, 30-32] Up to now, in vivo efforts to reduce this increased hepatic resistance by reducing TXA2 levels have been based on treatments with nonselective COX inhibitors.[32, 33] However, the strategy to block TXA2 production by nonselective COX inhibition is not acceptable in cirrhosis due to its demonstrated deleterious effects on sodium and water retention and renal function.[34, 35] There are no previous reports of the effects of TP-receptor blockade in vivo in cirrhotic animals. We hereby report the effects of terutroban, a specific TP-receptor blocker, in cirrhotic rats. Terutroban has been extensively used in clinical trials in vascular diseases and proven to be safe.[14, 36, 37]
Our study demonstrates that in vivo chronic TP-receptor blockade with terutroban produced a similar reduction in portal pressure in two different models, CCl4 and BDL. The decrease in portal pressure was not associated with changes in portal blood flow, suggesting a reduction in hepatic vascular resistance. This beneficial effect of terutroban on hepatic resistance could be attributed, in part, to the blockade of the vasoconstrictor effect of TXA2.[3, 32] Indeed, the blunted increase in portal pressure after the infusion of the TXA2 agonist U46619 in terutroban-treated rats suggests an adequate TP-receptor blockade and inhibition of TXA2-derived vasoconstriction of HSC and/or vascular smooth muscle cells in the hepatic vasculature. We also characterized the effects of terutroban on Rho-kinase activity. Rho-kinase, which is activated among other factors by the TP-receptor, is a well-known mechanism of HSC contraction[18, 38, 39] and it has been recently shown that its inhibition reduces hepatic vascular resistance.[40, 41] Our results showing that terutroban reduces hepatic Rho-kinase activity in both experimental models suggest that this may be an additional mechanism by which terutroban decreases hepatic vascular resistance.
However, other effects of terutroban were different according to the cirrhotic rat model. In CCl4-cirrhotic rats, terutroban ameliorated the architectural abnormalities of the liver, as shown by the reduction in liver fibrosis area on Sirius red staining. This was associated with a decrease in collagen I mRNA expression, suggesting a reduced collagen synthesis as a consequence of TP-receptor blockade, as it was not observed in vehicle-treated rats. This effect was associated, and probably linked, to a reduction of HSC activation, as suggested by the decreased α-SMA protein expression. It has been suggested that isoprostanes, a natural ligand for TP receptor, have been identified in HSC and mediate HSC proliferation and collagen production. Furthermore, terutroban significantly reduced TGF-β, which is one of the main fibrogenic cytokines that stimulates extracellular matrix deposition. These findings are in agreement with previous studies in an animal model of severe arterial hypertension showing that terutroban was able to prevent fibrosis in the aorta by reducing TGF-β gene expression. Thus, in CCl4-cirrhotic rats, both reduction in fibrosis and decreased hepatic vascular tone contribute to decrease the hepatic vascular resistance. Remarkably, the beneficial effects of terutroban on fibrosis were not observed in the BDL model. Although we do not have an explanation for this, we may speculate that this differential effect on fibrosis may be due to the fact that the BDL model is characterized by a very rapid and progressive fibrosis, while CCl4 represents a model with much slower fibrosis, susceptible of regression once CCl4 inhalation is interrupted.
Another differential effect of terutroban between the models was that observed on the NO signaling pathway. In BDL rats, terutroban promoted a significant increase of both eNOS protein expression, of its biologically active phosphorylated form, and the NO second messenger, cGMP, suggesting that in BDL rats an increase in NO bioavailability may also play a role reducing hepatic vascular resistance. By contrast, TP-receptor blockade in CCl4-cirrhotic rats did not produce significant changes in any of these parameters. At present, we do not have a clear explanation for such a differential effect of terutroban. It is remarkable that although terutroban did not change MAP in CCl4-cirrhotic rats, this was not the case in BDL rats, where a marked reduction was observed. It is possible that in the more severely ill rats with BDL cirrhosis, blocking the TXA2 vasoconstrictive systemic pathway together with an increase in NO bioavailability, probably also at the systemic level, may be responsible for such an effect decreasing MAP.
It is important to emphasize that terutroban reduces portal pressure in two different experimental settings of chronic liver disease. In the BDL model, terutroban was administered after 2 weeks of bile duct ligation when cirrhosis and the portal hypertension syndrome is not fully established and there is still an ongoing active injury. In this situation, although we cannot discard that longer periods of treatment may act on fibrosis, the main effect of terutroban was over the dynamic component of resistance. By contrast, in the CCl4 model terutroban was administered once the injury (CCl4 inhalation) was stopped in a setting of potential fibrosis reversal. Here, in addition to improving the dynamic component of the hepatic resistance, terutroban also facilitates fibrosis regression.
In conclusion, our data show that TP-receptor blockade with terutroban significantly reduces portal pressure in cirrhotic rats by decreasing hepatic vascular resistance (with a similar and comparable order of magnitude in both cirrhotic models), suggesting that terutroban may represent a useful agent in the treatment of portal hypertension in cirrhosis. However, in further translational steps of the investigation special care must be taken regarding possible effects reducing MAP.
The work was carried out at the Centre Esther Koplowitz, Barcelona. The authors thank Montse Monclús for excellent technical assistance.