- Email: [email protected]
Altered endothelin receptor expression in prehepatic portal hypertension predisposes the liver to microcirculatory dysfunction in rats
Journal of Hepatology 35 (2001) 29±36
www.elsevier.com/locate/jhep
Altered endothelin receptor expression in prehepatic portal hypertension predisposes the liver to microcirculatory dysfunction in rats Yukihiro Yokoyama, Andrew Wawrzyniak, Rajiv Baveja, Natalie Sonin, Mark G. Clemens, Jian X. Zhang* Biology Department, University of North Carolina at Charlotte, 9201 University City Boulevard, Charlotte, NC 28223, USA
Background/Aims: Endothelin (ET) is one of the most active vascular regulators in the liver. It is unknown how partial portal vein ligation (PPVL) induced prehepatic portal hypertension in¯uences the response of the liver to ET and its agonists. Therefore, this study was conducted to determine the expression of ET receptors and its functional signi®cance after PPVL. Methods: Competitive receptor binding study and semi-quantitative reverse transcription-polymerase chain reaction (RT-PCR) were performed using liver homogenates after 2 weeks of PPVL or sham operation in rats. Hepatic microcirculation was evaluated in vivo using intravital microscopy. Results: Although there was no signi®cant difference in dissociation constant (Kd) and total amount of receptors (Bmax) between sham and PPVL, the proportion of ETB receptor was signi®cantly increased in PPVL. RT-PCR analysis con®rmed the up-regulation of ETB receptors demonstrated by the competitive receptor binding assay. In the functional study, infusion of ETB agonist (IRL 1620) in a low dosage did not change the hepatic microcirculation in sham but strongly constricted the sinusoids leading to a reduction of sinusoidal perfusion in PPVL. Conclusions: These results suggest that prehepatic portal hypertension may predispose the hepatic microcirculation to dysregulation in stress conditions where ET is upregulated. q 2001 European Association for the Study of the Liver. Published by Elsevier Science B.V. All rights reserved. Keywords: Endothelin; Prehepatic portal hypertension; Partial portal vein ligation; ETA receptor; ETB receptor; Receptor binding study; In vivo microscopy; IRL 1620 receptor agonist 1. Introduction Prehepatic portal hypertension, resulting from extrahepatic portal vein obstruction, is seen most commonly in children [1]. Partial portal vein ligation (PPVL) in rats has been widely used as an excellent animal model to study prehepatic portal hypertension [2]. Although no signi®cant parenchymal damage is evident, reduced liver function has been reported in prehepatic portal hypertension [3]. We recently showed extensive vascular remodeling with formation of neovessels from the hepatic arterial system following 2±6
Received 23 November 2000; received in revised form 26 February 2001; accepted 4 March 2001 * Corresponding author. Tel.: 11-704-547-2315; fax: 11-704-547-3128. E-mail address: [email protected] (J.X. Zhang).
weeks of PPVL in the rat liver [4]. It is likely that the vascular remodeling seen in PPVL is also accompanied by changes in the microvascular response to the locally released vasoactive mediators such as endothelin. Endothelin (ET) causes a marked constriction in the hepatic portal circulation at both sinusoidal and extrasinusoidal levels [5±7]. Two types of receptors, ETA and ETB, mediate the actions of ET. Our previous studies showed that ET induces the constriction of sinusoids by activation of hepatic stellate cells (HSC) through ETA receptors [6]. In contrast, a speci®c ETB receptor agonist (IRL 1620) produces a significant increase in total portal vascular resistance without changing sinusoidal diameters, suggesting the constrictive effect of ETB on extrasinusoidal vessels [8]. In addition, several papers have shown that stimulation of ETB receptor elicits both vasoconstriction and vasodilation depending on
0168-8278/01/$20.00 q 2001 European Association for the Study of the Liver. Published by Elsevier Science B.V. All rights reserved. PII: S 0168-827 8(01)00076-9
30
Y. Yokoyama et al. / Journal of Hepatology 35 (2001) 29±36
activation of the receptor subtypes, i.e. ETB1 or ETB2 [9,10]. ETB1 receptors located on the endothelial cells mediate vasodilatation through activation of endothelial nitric oxide synthase (eNOS). On the other hand, ETB2 receptors distributed on smooth muscle cells and HSC mediate vasoconstriction through an increase in intracellular Ca 21 [11]. Our recent data showed that pre-treatment with NOS inhibitor, Nitro (G)-l-arginine methyl ester, enhanced sinusoidal constriction, in response to IRL 1620, which colocalizes with HSC, suggesting that ETB1-mediated eNOS activation counteracts the ETB2 receptor-mediated constriction elicited by ET and its receptor agonists [8]. Altered expression of ETA and ETB receptors have been observed under different hepatic stresses. Levias et al. [12] reported the up-regulation of ETA and ETB receptor transcripts in human cirrhotic livers, and that there was a direct correlation between the increased ETA and ETB receptor gene expression and the elevated portal pressure [12]. Our recent studies showed that the proportion of ETB receptors in the liver increased in lipopolysaccharide (LPS)-induced endotoxemia and ischemia/reperfusion (I/R) injury [8,13]. An increase in ETB receptors over ETA receptors in the two different stress conditions resulted in different portal vascular responses to the ETB receptor agonist IRL 1620. A moderate dose of LPS treatment attenuated the portal pressure increase and sinusoidal constriction to IRL 1620 suggesting a predominantly ETB1 receptor-mediated eNOS activation contributing to hepatocellular protection while on the other hand, I/R in the liver showed an increased portal pressure and hepatocellular injury in response to IRL 1620 [13]. The latter result suggests that the up-regulated ETB receptors may be predominantly the ETB2 type, activation of which causes a greater increase in hepatic microvascular resistance and subsequently hepatocellular injury. Therefore, changes in endothelin receptor types are critical in determining the outcome of microcirculatory dysfunction and hepatocellular injury in disease conditions. Chronic portal ¯ow obstruction not only induces vascular remodeling but may also change the behavior of the hepatic vasculature in response to vasoactive mediators. Therefore, the purposes of the present study were (1) to determine whether PPVL changes the expression of ET receptors in the liver, especially to compare the expression of ETA with that of ETB following PPVL; (2) if the changes were found, to determine whether the changes in ET receptors translate to alteration in the response of the hepatic microcirculation to ET receptor agonists.
2. Materials and methods 2.1. Chemicals [ 125I]-ET-1 (speci®c activity 2200 Ci/mmol) was purchased from NEN Life Science Products (Boston, MA). Human ET-1, BQ-610 and IRL 1620 were purchased from American Peptide Company (Sunnyvale, CA). The oligonucleotides used for polymerase chain reaction (PCR) were from Gibco BRL (Gaithersburg, MD). All other chemicals were purchased from Sigma Chemical Co. (St. Louis, MO).
2.2. Partial portal vein ligation The experiments were performed using Sprague±Dawley rats weighing 280±350 g. All procedures were performed in accordance with National Institutes of Health guidelines under a protocol approved by the Institutional Animal Care and Use Committee at the University of North Carolina at Charlotte. Partial portal vein ligation (PPVL) was performed as described by Vorbioff et al [2]. After anesthesia with metofane, the portal vein was exposed and ligated with 20-gauge needle. Subsequent removal of the needle resulted in the constant calibrated stenosis of the portal vein. Sham operation was performed in the same way except for the ligation of the portal vein.
2.3. Tissue preparation and endothelin receptor binding assay Two weeks after the surgery, the density of ET receptors in the liver was analyzed with competitive receptor binding studies as previously described [13]. Binding studies were carried out in incubation buffer (50 mM Tris, 5 mM MgCl2, 0.01% Bacitracin, 0.05% Bovine serum albumin, 0.2 mM phenylmethanesulfonyl ¯uoride (PMSF) (pH, 7.4) total volume 0.1 ml) containing 100 mg protein of crude homogenate and 100 pM [ 125I]-ET-1 with varying concentration of unlabeled ET-1 as the competitor. To determine the proportion of receptor subtypes (ETA/ETB), selective ETA antagonist BQ-610 (1mM) and selective ETB agonist IRL-1620 (100 nM) were used instead of unlabeled ET-1. Dissociation constant (Kd), maximum number of binding site per mg of protein sample (Bmax), and both ETA and ETB receptor proportion were calculated from Scatchard analysis.
2.4. Reverse transcription-polymerase chain reaction for ET receptor The expression of each ET receptor subtype transcripts was analyzed by semi-quantitative reverse transcription-polymerase chain reaction (RTPCR) as described before [14]. Pieces of the liver were preserved in RNA-Stat e reagent (TEL-TEST Inc., Friendswood, TX). After puri®cation, RNA samples were standardized by the intensity of ribosomal RNA bands. Reverse transcription of 1 mg of total RNA was performed using the GenAmp RNA PCR Core Kit (Perkin Elmer, Branchburg, NJ). cDNA was ampli®ed with primers (Table 1) in a total volume of 25 ml (the ®nal concentrations: 1 mM sense and antisense primer; 50 mM KCl; 10 mM Tris (pH 9.0); 1.5 mM MgCl2; 200 mM dNTPs; 0.5 U Taq DNA polymerase) [14]. The ampli®cation was carried out in 30 cycles at 948C for 45 s, 658C for 45 s and 738C for 1 min, followed by an additional 7 min of extension at
Table 1 PCR primers used in the study Name
Sense
Antisense
Product length (bp)
ETA ETB Ribosomal protein S9
AGTGCTAATCTAAGCAGCCAC AGCTGGTGCCCTTCATACAGAAGGC GCTGAGACTGACTTGCGAATTG
CAGGAAGCCACTGCTCTGTAC TGCACACCTTTCCGCAAGCACG TGGCATTCTTCCTCTTCACTCG
491 919 304
Y. Yokoyama et al. / Journal of Hepatology 35 (2001) 29±36 728C. The PCR products were electrophoresed on 1.5% agarose gel stained with ethidium bromide. The level of each PCR product was evaluated densitometrically.
2.5. Surgical preparation for intravital microscopy Rats were fasted overnight but allowed free access to water. After induction of anesthesia (sodium pentobarbital 50 mg/kg body weight intraperitoneally), a tracheotomy was performed to prevent airway obstruction. The right carotid artery was cannulated with PE-50 tubing for the assessment of hemodynamics. After midline laparotomy, a double-lumen catheter (outer catheter PE-90, inner catheter PE-10) was inserted from the splenic vein into the portal vein [5,15]. The animal was transferred to the stage of an Olympus IX70 inverted microscope. The left lobe of the liver was gently exteriorized and positioned onto a window, which was covered with a Corning no. 1 micro-cover glass. To compensate for ¯uid loss during surgery and intravital microscopy, 100 ml/kg b.w./min of 0.9% NaCl was continuously infused via the inner lumen of the portal catheter.
2.6. In vivo microscopy The liver surface was epi-illuminated with a 100 W mercury lamp with 460±500 nm excitation and 515±560 nm emission band-pass ®lters. The images were recorded and analyzed off-line during video playback using digitized frame-by-frame analysis with Image-Pro (Media Cybernetics, Silver Spring, MD). A total magni®cation (specimen to monitor) of £ 760 was used. Measurements of sinusoidal diameters (Ds) and perfused sinusoid density (Ps; de®ned as number of perfused sinusoids per 150mm) were made directly from video playback as previously described [15]. 0.2 ml of ¯uorescein isothiocyanate (FITC)-labeled red blood cells (RBC), prepared as originally described by Zimmerhackl et al. [16], were injected through the carotid artery for measurement of the RBC velocity (VRBC) in sinusoids. Volumetric ¯ow (VF) and perfusion index (PI) was calculated from the Ds, Ps, and VRBC by the following equation [17]: VF VRBC *p* Ds=22 ; PI Ps*VF. Mean arterial pressure (MAP), heart rate (HR) and portal venous pressure (PVP) were recorded simultaneously. For the functional studies, IRL 1620 (selective ETB agonist) was infused through the portal vein. Ten ®elds of images were taken randomly for the hemodynamic analysis before and after the IRL 1620 infusion.
31
proportion of ETA receptor was slightly higher than that of ETB receptor (ETA, 49.1 ^ 2.3 %; ETB, 41.9 ^ 1.9 %). In contrast, after PPVL, the proportion of ETA receptors was signi®cantly decreased, whereas that of ETB receptors was signi®cantly increased compared to sham (ETA, 32.7 ^ 2.3%; ETB, 63.1 ^ 2.5%; P , 0:05). 3.2. Semi-quantitative RT-PCR for ET receptors The change of ET receptor subtype expression was con®rmed at the mRNA level by semi-quantitative RTPCR (Fig. 2). Densitometric analysis of RT-PCR bands revealed a signi®cant decrease of ETA receptor mRNA and an increase of ETB receptor mRNA following PPVL. ETA receptor mRNA decreased by 48% and ETB receptor mRNA increased by 63% after PPVL. These data suggest that receptor remodeling after PPVL occur both in mRNA and functional protein levels. 3.3. Microhemodynamic evaluation by intravital microscopy MAP and HR were not signi®cantly different between sham and PPVL (Fig. 3A,B). The baseline portal pressure was signi®cantly higher in PPVL compared to sham con®rming prehepatic portal hypertension in PPVL animals (Fig. 3C). In spite of the increased PVP in PPVL, the baseline microhemodynamics did not show a signi®cant difference from sham (Fig. 4). These results were consistent with the data observed with 6 weeks of PPVL reported in our previous study [4].
2.7. Statistical analysis Statistical differences between the two groups were analyzed by t-test. When criteria for parametric testing were violated the appropriate nonparametric (Mann±Whitney U-test) was used. Differences within each group over time were analyzed by means of repeated measures ANOVA. P , 0:05 was considered signi®cant. All results are presented as means ^ SEM.
3. Results 3.1. Endothelin receptor binding assay Competitive receptor binding assay did not show a significant difference between sham and PPVL (Fig. 1). Neither Kd nor Bmax were found signi®cantly different by Scatchard plots analysis between the two groups (Kd: sham 70.7 ^ 12.8 vs. PPVL 78.9 ^ 25.1 pM; Bmax: sham 34.3 ^ 2.5 vs. PPVL 28.1 ^ 2.9 fmol/mg of proteins). These results suggest that there are no difference in either the receptor-ligand af®nity or the total number of ET receptors between sham and PPVL. However, PPVL signi®cantly changed the number of ET receptor subtypes. In sham, the
Fig. 1. Displacement curves of competitive receptor binding assay in sham and PPVL group. Assay was performed using crude homogenates of the livers. Samples were incubated with 100 pM [ 125I]-ET-1 with varying concentrations of cold ET-1. After 4 h of incubation, the sample was harvested on glass ®ber ®lters and the radioactivity was measured by gamma counting. Non-speci®c binding was determined in the presence of an excess amount of cold ET-1 (200 nM). Data are the means ^ SEM of ®ve separate experiments in sham and PPVL.
32
Y. Yokoyama et al. / Journal of Hepatology 35 (2001) 29±36
Fig. 2. mRNA expression of ETA and ETB receptors in the liver tissue analyzed by semi-quantitative RT-PCR. Samples were harvested after 2 weeks of PPVL or sham operation. The expressions of ETA (A) and ETB (B) mRNA levels were analyzed by RT-PCR using speci®c primers. Representative gels and ribosomal protein S9 as house keeping gene are shown on top. Average intensities of the bands analyzed by densitometry are shown in the bar graphs. Data are the means ^ SEM of four separate experiments in sham and PPVL. *P , 0.05 vs. sham.
3.4. The effects of ETB receptor stimulation by IRL 1620 In order to determine the functional importance of the increased ETB receptors, the selective ETB agonist, IRL 1620, was infused through the portal vein. IRL 1620, Suc[Glu9, Ala11, 15] Endothelin-1 (8-21), is a potent and selective ETB agonist with Ki of 16 nM. Microcirculatory perfusion was monitored and recorded with intravital microscopy during the infusion. IRL 1620 was infused either with low dose (1 pmol/min/100 g body weight) for 10 min or high dose (5 pmol/min per 100 g body weight) for 20 min. Portal
infusion of IRL 1620 did not change the MAP or HR in the low or high dose group (data not shown). The increases in PVP were minimal and almost identical in sham and PPVL with low dose of IRL 1620 (Fig. 5A). Although, the high dose caused a substantially higher increase in PVP in sham, the peak response in PPVL was not different between the high dose and the low dose (Fig. 5B). Interestingly, the signi®cantly lower increase in PVP in PPVL than sham was followed by a sustained decrease. Different responses were observed in the hepatic microcirculation following the low dose and the high dose of IRL 1620 (Fig. 6). With the
Fig. 3. Baseline hemodynamic data of sham (opened bar) and PPVL (hatched bar). Mean arterial pressure (MAP, A), heart rate (HR, B), and portal venous pressure (PVP, C) were measured at baseline. Data are the means ^SEM of six separate experiments in sham and PPVL. *P , 0.05 vs. sham.
Y. Yokoyama et al. / Journal of Hepatology 35 (2001) 29±36
33
Fig. 4. Comparison of hepatic microhemodynamics between sham (opened bar) and PPVL (hatched bar). 0.2 ml of ¯uorescein isothiocyanate (FITC)labeled RBCs was infused from the carotid artery. Number of perfused sinusoids (Ps, A) in 150 mm were counted on the screen during the video playback. Sinusoidal diameter (Ds, B) and red blood cell velocity (VRBC, C) were measured by of¯ine of recorded images using Image-Pro. Volumetric ¯ow (VF, D) and perfusion index (Pi, E) were calculated by the following equation: VF VRBC*p *(Ds/2) 2 (picoliter/s); PI Ps*VF (picoliter/s per 150 mm). Data are the means ^SEM of six separate experiments in sham and PPVL.
low dose, no signi®cant changes in sinusoidal hemodynamics were found in sham following infusion of IRL 1620 (Fig. 6A). However, the same low dose caused a
severe sinusoidal constriction, decreases of VRBC and Ps, and as a result, a signi®cant reduction in perfusion index in PPVL. Approximately 50% of sinusoids were shut down
Fig. 5. PVP changes following infusion of ETB agonist (IRL 1620) with low dose (1 pmol/min per 100 g for 10 min, A) and high dose (5 pmol/min per 100 g for 20 min, B). PVP was monitored through the double lumen catheter inserted into the portal vein. Data are the means ^ SEM of six separate experiments in sham and PPVL. *P , 0.05 versus baseline PVP. #P , 0.05 vs. PPVL.
34
Y. Yokoyama et al. / Journal of Hepatology 35 (2001) 29±36
4. Discussion
Fig. 6. Effects of portal infusion of ETB receptor agonist IRL 1620 on hepatic microhemodynamics. IRL 1620 was infused either with low dose (1 pmol/min per 100 g for 10 min, A) or high dose (5 pmol/min per 100 g for 20 min, B) in 100 ml/min per kg of body weight. Data are shown as means ^ SEM of % changes from the baseline. *P , 0.05 vs. sham. Ps, number of perfused sinusoids/150 mm; Ds, sinusoidal diameter; VRBC, red blood cell velocity; VF, volumetric ¯ow; PI, perfusion index.
and more than a 70% decrease in PI was observed at the end of the infusion. These results strongly suggest that the ETB2mediated constriction was predominant at the sinusoidal level of the PPVL liver in response to the ETB agonist. With the high dose of IRL 1620, hepatic microcirculatory perfusion was substantially reduced in both sham and PPVL as re¯ected by the signi®cant decreases in VF and PI (Fig. 6B, 70 and 85% decreases in PI, respectively). Both groups demonstrated signi®cant decreases in Ps and VRBC in response to the high dose, but greater percentage decreases in Ps and PI were seen in PPVL than sham. Interestingly, no change in Ds was observed in PPVL with the high dose, which is contrary to what was observed with the low dose, suggesting an activation of ETB1 receptors and/or extrasinusoidal sites of action by IRL 1620.
In the present study, we investigated the effect of prehepatic portal hypertension on the changes in ET receptors in the liver and the hepatic microcirculatory response to the speci®c ETB agonist IRL 1620 using PPVL in rats. It was demonstrated that expression of ETB receptors was increased both at functional protein and mRNA levels after PPVL. Altered hepatic expression of ET receptor subtypes in PPVL resulted in a substantial change in the response of the hepatic microcirculation to ETB receptor agonists. Prehepatic portal vein stenosis not only exerts direct effects on the splanchnic circulation, but also on the hepatic circulation by substantially decreasing portal blood ¯ow. Our recent study showed that neovascularization from the hepatic arterial system occurs in the liver following PPVL suggesting vascular remodeling, possibly induced by the decreased portal blood ¯ow [4]. It is conceivable that the restriction of the portal ¯ow by PPVL and the subsequent vascular remodeling in the liver would also give rise to changes in regulation of the hepatic microcirculation. Since ET is one of the most active vasoregulators in the hepatic microcirculation, we focused on the effect of PPVL on expression of ETA and ETB receptors in the liver. PPVL did not change the total number of ET receptors in the liver but signi®cantly altered the ratio of ETA and ETB receptors. Decreased ETA receptors and increased ETB receptors were not only shown after PPVL at the functional protein level but also con®rmed at the mRNA level by RT-PCR. The mechanisms by which the ratio of ETA and ETB receptors were altered in PPVL induced portal hypertension are presently unclear. However, previous studies have shown an increased proportion of ETB receptors in the liver under a number of stress conditions such as endotoxemia and I/R injury [8]. It appears that upregulation of ETB receptors is a common response in the liver to stress conditions. In order to determine how the proportional changes in ET receptor subtypes translates to the change in functional response of the hepatic microcirculation, we determined the microhemodynamics of the hepatic portal circulation using in vivo microscopy. Low dose of IRL 1620 (1 pmol/ min per 100 g for 10 min) induced a slight increase in PVP in both sham and PPVL and no visible changes in the sinusoidal hemodynamics in sham (Fig. 6A). However, in PPVL, it caused a substantial reduction in sinusoidal perfusion as a result of sinusoidal constriction and signi®cant decreases in Ps and VRBC. These results strongly suggest that the increased ETB receptors following PPVL is in part due to the increased ETB2 receptors in the sinusoids, most likely on HSC, activation of which results in sinusoidal constriction. Nonetheless, the question remains: how does the severe reduction of sinusoidal perfusion occur after IRL 1620 without signi®cantly changing PVP? The answer probably lies in ¯ow shunting. It was observed that the ¯ow was shut down in almost a half of the sinusoids, accompanied by
Y. Yokoyama et al. / Journal of Hepatology 35 (2001) 29±36
sinusoidal constrictions and a 70% reduction of PI in PPVL. It is possible that blood ¯ow shunted through the collateral vessels under such a condition. Interestingly, high dose of IRL 1620 induced different responses. PVP in sham increased substantially in response to the ETB receptor agonist whereas only a minimal increase in PVP was observed in PPVL (Fig. 5B). The increased PVP in sham was accompanied by a signi®cant reduction of the sinusoidal perfusion (70% decrease in PI) indicating that the concentration of IRL 1620 used was overwhelming and elicited primarily an extrasinusoidal effect since the sinusoidal diameter did not change signi®cantly (Fig. 6B). In contrast to sham rats, there was a minimal increase in PVP in PPVL followed by a sustained decreased phase. However, there was a dramatic sinusoidal ¯ow shutdown with an 85% decrease in PI in PPVL following the high dose. The mechanisms underlying the different responses at the portal and sinusoidal levels are not clear. However, we speculate that the mixed actions mediated by both ETB1 and ETB2 receptors and portal collateral ¯ow shunting are responsible for the different responses [18,19]. The blunted PVP increase and the subsequent decrease are likely due to the portal-collateral ¯ow shunting and the ETB1 receptormediated action. The drastic effect of the high dose of IRL 1620 on the sinusoidal ¯ow is caused by the ETB2 receptor-mediated constriction, acting on the sinusoids and the extrasinusoidal sites especially the post-sinusoidal sites. Prehepatic portal hypertension often produces no signi®cant parenchymal damage. Indeed, no visible changes were observed in the hepatic microhemodynamics in PPVL compared to sham in spite of portal hypertension (Fig. 4). However, the present study suggests that patients with this condition may be susceptible to secondary stresses where hepatic ET level is elevated such as LPS-induced endotoxemia [20±22], hemorrhagic resuscitation [23] and I/R [24,25]. Under these stress conditions, the hepatic microcirculatory dysfunction is prone to occur as a consequence of the acute stress-induced elevated ET acting upon the increased ETB2 receptors in the sinusoids. In summary, the results of our ET receptor binding assay and RT-PCR showed a signi®cant up-regulation of ETB receptors and a decrease in ETA receptors in the liver following 2 weeks of PPVL. These changes were ultimately responsible for the increased sensitivity of the hepatic microcirculation to the ETB receptor agonist IRL 1620 in causing microcirculatory ¯ow disruption in PPVL. These results suggest that prehepatic portal hypertension may be susceptible to the hepatic microcirculatory dysfunction following secondary stress conditions. Acknowledgements The study was supported by the grant from National Institutes of Health (DK38201) and faculty grants of University of North Carolina at Charlotte.
35
References [1] Kralik J. Long-term prognosis of prehepatic portal hypertension. Acta Univ Palacki Olomuc Fac Med 1992;133:63±66. [2] Vorobioff J, Bredfeldt JE, Groszmann RJ. Hyperdynamic circulation in portal-hypertensive rat model: a primary factor for maintenance of chronic portal hypertension. Am J Physiol 1983;244:G52±G57. [3] Murakami S, Ohno T, Bernardo JF, Pfeifer CA, Li T, Zhang Y, Dubey RK, Branch RA, Sabra R. Reduced liver function is the trigger for renal sodium retention following portal vein ligation in the rat. J Gastroenterol Hepatol 1996;11:850±856. [4] Yokoyama Y, Baveja R, Sonin N, Clemens MG, Zhang JX. Hepatic neovascularization after partial portal vein ligation: novel mechanism of chronic regulation of blood ¯ow. Am J Physiol Gastrointest Liver Physiol 2001;280(1):G21±G31. [5] Bauer M, Zhang JX, Bauer I, Clemens MG. Endothelin-1 as a regulator of hepatic microcirculation: sublobular distribution of effects and impact on hepatocellular secretory function. Shock 1994;1:457± 465. [6] Zhang JX, Bauer M, Clemens MG. Vessel and target cell speci®c actions of endothelin-1 and endothelin-3 in rat liver. Am J Physiol 1995;269:G269±G277. [7] Rockey D. The cellular pathogenesis of portal hypertension: stellate cell contractility, endothelin, and nitric oxide. Hepatology 1997;25:2± 5. [8] Bauer M, Bauer I, Sonin NV, Kresge N, Baveja R, Yokoyama Y, Harding D, Zhang JX, Clemens MG. Functional signi®cance of endothelin B receptors in mediating sinusoidal and extrasinusoidal effects of endothelins in the intact rat liver. Hepatology 2000;31(4):937±947. [9] Mallat A, Lotersztajn S. Multiple hepatic functions of endothelin-1: physiopathological relevance. J. Hepatol. 1996;25:405±413. [10] Douglas SA, Meek TD, Ohlstein EH. Novel receptor antagonists welcome a new era in endothelin biology. Trends Pharmacol Sci 1994;15:313±316. [11] Pollock DM, Keith TL, Highsmith RF. Endothelin receptors and calcium signaling. FASEB J 1995;9:1196±1204. [12] Leivas A, Jimenez W, Bruix J, Boix L, Bosch J, Arroyo V, Rivera F, Rodes J. Gene expression of endothelin-1 and ET(A) and ET(B) receptors in human cirrhosis: relationship with hepatic hemodynamics. J Vasc Res 1998;35:186±193. [13] Yokoyama Y, Baveja R, Sonin N, Nakanishi K, Zhang JX, Clemens MG. Altered endothelin receptor subtype expression in hepatic injury after ischemia/reperfusion. Shock 2000;13(1):72±78. [14] Sonin NV, Garcia-Pagan JC, Nakanishi K, Zhang JX, Clemens MG. Patterns of vasoregulatory gene expression in the liver response to ischemia/reperfusion and endotoxemia. Shock 1999;11:175±179. [15] Clemens MG, McDonagh PF, Chaudry IH, Baue AE. Hepatic microcirculatory failure following ischemia and reperfusion: improvement with ATP-MgCl2 treatment. Am J Physiol. 1985;248:H804±H811. [16] Zimmerhackl B, Parekh N, Brinkhus H, Steinhausen M. The use of ¯uorescent labeled erythrocytes for intravital investigation of ¯ow and local hematocrit in glomerular capillaries in the rat. Int J Microcirc Clin Exp 1983;2:119±129. [17] Bauer I, Bauer M, Pannen BHJ, Leinwand MJ, Zhang JX, Clemens MG. Chronic ethanol consumption exacerbates liver injury following hemorrhagic shock: role of sinusoidal perfusion failure. Shock 1995;4:324±331. [18] Halvorsen JF, Myking AO, Tvete S. Porto-hepatic bypass by organohepatopexy in rats with prehepatic portal hypertension. Formation of new venous porto-hepatic collaterals following omento-, jejuno-, and splenohepatopexy, and their effect on portal vein pressure. Eur Surg Res 1979;11:254±266. [19] Ingles AC, Legare DJ, Lautt WW. Development of portacaval shunts in portal-stenotic cats. Can J Physiol Pharmacol 1993; 71:671±674.
36
Y. Yokoyama et al. / Journal of Hepatology 35 (2001) 29±36
[20] Eakes AT, Howard KM, Miller JE, Olson MS. Endothelin-1 production by hepatic endothelial cells: characterization and augmentation by endotoxin exposure. Am J Physiol 1997;272:G605±G611. [21] Ohuchi T, Tada K, Akamatsu K. Endogenous ET-1 contributes to liver injury induced by galactosamine and endotoxin in isolated perfused rat liver. Am J Physiol 1995;268:G997±G1003. [22] Ros J, Leivas A, Jimenez W, Morales M, Bosch-Marce M, Arroyo V, Rivera F, Rodes J. Effect of bacterial lipopolysaccharide on endothelin-1 production in human vascular endothelial cells. J Hepatol 1997;26:81±87. [23] Pannen BH, Bauer M, Noldge-Schomburg GF, Zhang JX, Robotham
JL, Clemens MG, Geiger KK. Regulation of hepatic blood ¯ow during resuscitation from hemorrhagic shock: role of NO and endothelins. Am J Physiol 1997;272:H2736±H2745. [24] Nakamura S, Nishiyama R, Serizawa A, Yokoi Y, Suzuki S, Konno H, Baba S, Muro H. Hepatic release of endothelin-1 after warm ischemia. Reperfusion injury and its hemodynamic effect. Transplantation 1995;59:679±684. [25] Ota T, Hirai R, Urakami A, Soga H, Nawa S, Shimizu N. Endothelin1 levels in portal venous blood in relation to hepatic tissue microcirculation disturbance and hepatic cell injury after ischemia/reperfusion. Surg Today 1997;27:313±320.
www.elsevier.com/locate/jhep
Altered endothelin receptor expression in prehepatic portal hypertension predisposes the liver to microcirculatory dysfunction in rats Yukihiro Yokoyama, Andrew Wawrzyniak, Rajiv Baveja, Natalie Sonin, Mark G. Clemens, Jian X. Zhang* Biology Department, University of North Carolina at Charlotte, 9201 University City Boulevard, Charlotte, NC 28223, USA
Background/Aims: Endothelin (ET) is one of the most active vascular regulators in the liver. It is unknown how partial portal vein ligation (PPVL) induced prehepatic portal hypertension in¯uences the response of the liver to ET and its agonists. Therefore, this study was conducted to determine the expression of ET receptors and its functional signi®cance after PPVL. Methods: Competitive receptor binding study and semi-quantitative reverse transcription-polymerase chain reaction (RT-PCR) were performed using liver homogenates after 2 weeks of PPVL or sham operation in rats. Hepatic microcirculation was evaluated in vivo using intravital microscopy. Results: Although there was no signi®cant difference in dissociation constant (Kd) and total amount of receptors (Bmax) between sham and PPVL, the proportion of ETB receptor was signi®cantly increased in PPVL. RT-PCR analysis con®rmed the up-regulation of ETB receptors demonstrated by the competitive receptor binding assay. In the functional study, infusion of ETB agonist (IRL 1620) in a low dosage did not change the hepatic microcirculation in sham but strongly constricted the sinusoids leading to a reduction of sinusoidal perfusion in PPVL. Conclusions: These results suggest that prehepatic portal hypertension may predispose the hepatic microcirculation to dysregulation in stress conditions where ET is upregulated. q 2001 European Association for the Study of the Liver. Published by Elsevier Science B.V. All rights reserved. Keywords: Endothelin; Prehepatic portal hypertension; Partial portal vein ligation; ETA receptor; ETB receptor; Receptor binding study; In vivo microscopy; IRL 1620 receptor agonist 1. Introduction Prehepatic portal hypertension, resulting from extrahepatic portal vein obstruction, is seen most commonly in children [1]. Partial portal vein ligation (PPVL) in rats has been widely used as an excellent animal model to study prehepatic portal hypertension [2]. Although no signi®cant parenchymal damage is evident, reduced liver function has been reported in prehepatic portal hypertension [3]. We recently showed extensive vascular remodeling with formation of neovessels from the hepatic arterial system following 2±6
Received 23 November 2000; received in revised form 26 February 2001; accepted 4 March 2001 * Corresponding author. Tel.: 11-704-547-2315; fax: 11-704-547-3128. E-mail address: [email protected] (J.X. Zhang).
weeks of PPVL in the rat liver [4]. It is likely that the vascular remodeling seen in PPVL is also accompanied by changes in the microvascular response to the locally released vasoactive mediators such as endothelin. Endothelin (ET) causes a marked constriction in the hepatic portal circulation at both sinusoidal and extrasinusoidal levels [5±7]. Two types of receptors, ETA and ETB, mediate the actions of ET. Our previous studies showed that ET induces the constriction of sinusoids by activation of hepatic stellate cells (HSC) through ETA receptors [6]. In contrast, a speci®c ETB receptor agonist (IRL 1620) produces a significant increase in total portal vascular resistance without changing sinusoidal diameters, suggesting the constrictive effect of ETB on extrasinusoidal vessels [8]. In addition, several papers have shown that stimulation of ETB receptor elicits both vasoconstriction and vasodilation depending on
0168-8278/01/$20.00 q 2001 European Association for the Study of the Liver. Published by Elsevier Science B.V. All rights reserved. PII: S 0168-827 8(01)00076-9
30
Y. Yokoyama et al. / Journal of Hepatology 35 (2001) 29±36
activation of the receptor subtypes, i.e. ETB1 or ETB2 [9,10]. ETB1 receptors located on the endothelial cells mediate vasodilatation through activation of endothelial nitric oxide synthase (eNOS). On the other hand, ETB2 receptors distributed on smooth muscle cells and HSC mediate vasoconstriction through an increase in intracellular Ca 21 [11]. Our recent data showed that pre-treatment with NOS inhibitor, Nitro (G)-l-arginine methyl ester, enhanced sinusoidal constriction, in response to IRL 1620, which colocalizes with HSC, suggesting that ETB1-mediated eNOS activation counteracts the ETB2 receptor-mediated constriction elicited by ET and its receptor agonists [8]. Altered expression of ETA and ETB receptors have been observed under different hepatic stresses. Levias et al. [12] reported the up-regulation of ETA and ETB receptor transcripts in human cirrhotic livers, and that there was a direct correlation between the increased ETA and ETB receptor gene expression and the elevated portal pressure [12]. Our recent studies showed that the proportion of ETB receptors in the liver increased in lipopolysaccharide (LPS)-induced endotoxemia and ischemia/reperfusion (I/R) injury [8,13]. An increase in ETB receptors over ETA receptors in the two different stress conditions resulted in different portal vascular responses to the ETB receptor agonist IRL 1620. A moderate dose of LPS treatment attenuated the portal pressure increase and sinusoidal constriction to IRL 1620 suggesting a predominantly ETB1 receptor-mediated eNOS activation contributing to hepatocellular protection while on the other hand, I/R in the liver showed an increased portal pressure and hepatocellular injury in response to IRL 1620 [13]. The latter result suggests that the up-regulated ETB receptors may be predominantly the ETB2 type, activation of which causes a greater increase in hepatic microvascular resistance and subsequently hepatocellular injury. Therefore, changes in endothelin receptor types are critical in determining the outcome of microcirculatory dysfunction and hepatocellular injury in disease conditions. Chronic portal ¯ow obstruction not only induces vascular remodeling but may also change the behavior of the hepatic vasculature in response to vasoactive mediators. Therefore, the purposes of the present study were (1) to determine whether PPVL changes the expression of ET receptors in the liver, especially to compare the expression of ETA with that of ETB following PPVL; (2) if the changes were found, to determine whether the changes in ET receptors translate to alteration in the response of the hepatic microcirculation to ET receptor agonists.
2. Materials and methods 2.1. Chemicals [ 125I]-ET-1 (speci®c activity 2200 Ci/mmol) was purchased from NEN Life Science Products (Boston, MA). Human ET-1, BQ-610 and IRL 1620 were purchased from American Peptide Company (Sunnyvale, CA). The oligonucleotides used for polymerase chain reaction (PCR) were from Gibco BRL (Gaithersburg, MD). All other chemicals were purchased from Sigma Chemical Co. (St. Louis, MO).
2.2. Partial portal vein ligation The experiments were performed using Sprague±Dawley rats weighing 280±350 g. All procedures were performed in accordance with National Institutes of Health guidelines under a protocol approved by the Institutional Animal Care and Use Committee at the University of North Carolina at Charlotte. Partial portal vein ligation (PPVL) was performed as described by Vorbioff et al [2]. After anesthesia with metofane, the portal vein was exposed and ligated with 20-gauge needle. Subsequent removal of the needle resulted in the constant calibrated stenosis of the portal vein. Sham operation was performed in the same way except for the ligation of the portal vein.
2.3. Tissue preparation and endothelin receptor binding assay Two weeks after the surgery, the density of ET receptors in the liver was analyzed with competitive receptor binding studies as previously described [13]. Binding studies were carried out in incubation buffer (50 mM Tris, 5 mM MgCl2, 0.01% Bacitracin, 0.05% Bovine serum albumin, 0.2 mM phenylmethanesulfonyl ¯uoride (PMSF) (pH, 7.4) total volume 0.1 ml) containing 100 mg protein of crude homogenate and 100 pM [ 125I]-ET-1 with varying concentration of unlabeled ET-1 as the competitor. To determine the proportion of receptor subtypes (ETA/ETB), selective ETA antagonist BQ-610 (1mM) and selective ETB agonist IRL-1620 (100 nM) were used instead of unlabeled ET-1. Dissociation constant (Kd), maximum number of binding site per mg of protein sample (Bmax), and both ETA and ETB receptor proportion were calculated from Scatchard analysis.
2.4. Reverse transcription-polymerase chain reaction for ET receptor The expression of each ET receptor subtype transcripts was analyzed by semi-quantitative reverse transcription-polymerase chain reaction (RTPCR) as described before [14]. Pieces of the liver were preserved in RNA-Stat e reagent (TEL-TEST Inc., Friendswood, TX). After puri®cation, RNA samples were standardized by the intensity of ribosomal RNA bands. Reverse transcription of 1 mg of total RNA was performed using the GenAmp RNA PCR Core Kit (Perkin Elmer, Branchburg, NJ). cDNA was ampli®ed with primers (Table 1) in a total volume of 25 ml (the ®nal concentrations: 1 mM sense and antisense primer; 50 mM KCl; 10 mM Tris (pH 9.0); 1.5 mM MgCl2; 200 mM dNTPs; 0.5 U Taq DNA polymerase) [14]. The ampli®cation was carried out in 30 cycles at 948C for 45 s, 658C for 45 s and 738C for 1 min, followed by an additional 7 min of extension at
Table 1 PCR primers used in the study Name
Sense
Antisense
Product length (bp)
ETA ETB Ribosomal protein S9
AGTGCTAATCTAAGCAGCCAC AGCTGGTGCCCTTCATACAGAAGGC GCTGAGACTGACTTGCGAATTG
CAGGAAGCCACTGCTCTGTAC TGCACACCTTTCCGCAAGCACG TGGCATTCTTCCTCTTCACTCG
491 919 304
Y. Yokoyama et al. / Journal of Hepatology 35 (2001) 29±36 728C. The PCR products were electrophoresed on 1.5% agarose gel stained with ethidium bromide. The level of each PCR product was evaluated densitometrically.
2.5. Surgical preparation for intravital microscopy Rats were fasted overnight but allowed free access to water. After induction of anesthesia (sodium pentobarbital 50 mg/kg body weight intraperitoneally), a tracheotomy was performed to prevent airway obstruction. The right carotid artery was cannulated with PE-50 tubing for the assessment of hemodynamics. After midline laparotomy, a double-lumen catheter (outer catheter PE-90, inner catheter PE-10) was inserted from the splenic vein into the portal vein [5,15]. The animal was transferred to the stage of an Olympus IX70 inverted microscope. The left lobe of the liver was gently exteriorized and positioned onto a window, which was covered with a Corning no. 1 micro-cover glass. To compensate for ¯uid loss during surgery and intravital microscopy, 100 ml/kg b.w./min of 0.9% NaCl was continuously infused via the inner lumen of the portal catheter.
2.6. In vivo microscopy The liver surface was epi-illuminated with a 100 W mercury lamp with 460±500 nm excitation and 515±560 nm emission band-pass ®lters. The images were recorded and analyzed off-line during video playback using digitized frame-by-frame analysis with Image-Pro (Media Cybernetics, Silver Spring, MD). A total magni®cation (specimen to monitor) of £ 760 was used. Measurements of sinusoidal diameters (Ds) and perfused sinusoid density (Ps; de®ned as number of perfused sinusoids per 150mm) were made directly from video playback as previously described [15]. 0.2 ml of ¯uorescein isothiocyanate (FITC)-labeled red blood cells (RBC), prepared as originally described by Zimmerhackl et al. [16], were injected through the carotid artery for measurement of the RBC velocity (VRBC) in sinusoids. Volumetric ¯ow (VF) and perfusion index (PI) was calculated from the Ds, Ps, and VRBC by the following equation [17]: VF VRBC *p* Ds=22 ; PI Ps*VF. Mean arterial pressure (MAP), heart rate (HR) and portal venous pressure (PVP) were recorded simultaneously. For the functional studies, IRL 1620 (selective ETB agonist) was infused through the portal vein. Ten ®elds of images were taken randomly for the hemodynamic analysis before and after the IRL 1620 infusion.
31
proportion of ETA receptor was slightly higher than that of ETB receptor (ETA, 49.1 ^ 2.3 %; ETB, 41.9 ^ 1.9 %). In contrast, after PPVL, the proportion of ETA receptors was signi®cantly decreased, whereas that of ETB receptors was signi®cantly increased compared to sham (ETA, 32.7 ^ 2.3%; ETB, 63.1 ^ 2.5%; P , 0:05). 3.2. Semi-quantitative RT-PCR for ET receptors The change of ET receptor subtype expression was con®rmed at the mRNA level by semi-quantitative RTPCR (Fig. 2). Densitometric analysis of RT-PCR bands revealed a signi®cant decrease of ETA receptor mRNA and an increase of ETB receptor mRNA following PPVL. ETA receptor mRNA decreased by 48% and ETB receptor mRNA increased by 63% after PPVL. These data suggest that receptor remodeling after PPVL occur both in mRNA and functional protein levels. 3.3. Microhemodynamic evaluation by intravital microscopy MAP and HR were not signi®cantly different between sham and PPVL (Fig. 3A,B). The baseline portal pressure was signi®cantly higher in PPVL compared to sham con®rming prehepatic portal hypertension in PPVL animals (Fig. 3C). In spite of the increased PVP in PPVL, the baseline microhemodynamics did not show a signi®cant difference from sham (Fig. 4). These results were consistent with the data observed with 6 weeks of PPVL reported in our previous study [4].
2.7. Statistical analysis Statistical differences between the two groups were analyzed by t-test. When criteria for parametric testing were violated the appropriate nonparametric (Mann±Whitney U-test) was used. Differences within each group over time were analyzed by means of repeated measures ANOVA. P , 0:05 was considered signi®cant. All results are presented as means ^ SEM.
3. Results 3.1. Endothelin receptor binding assay Competitive receptor binding assay did not show a significant difference between sham and PPVL (Fig. 1). Neither Kd nor Bmax were found signi®cantly different by Scatchard plots analysis between the two groups (Kd: sham 70.7 ^ 12.8 vs. PPVL 78.9 ^ 25.1 pM; Bmax: sham 34.3 ^ 2.5 vs. PPVL 28.1 ^ 2.9 fmol/mg of proteins). These results suggest that there are no difference in either the receptor-ligand af®nity or the total number of ET receptors between sham and PPVL. However, PPVL signi®cantly changed the number of ET receptor subtypes. In sham, the
Fig. 1. Displacement curves of competitive receptor binding assay in sham and PPVL group. Assay was performed using crude homogenates of the livers. Samples were incubated with 100 pM [ 125I]-ET-1 with varying concentrations of cold ET-1. After 4 h of incubation, the sample was harvested on glass ®ber ®lters and the radioactivity was measured by gamma counting. Non-speci®c binding was determined in the presence of an excess amount of cold ET-1 (200 nM). Data are the means ^ SEM of ®ve separate experiments in sham and PPVL.
32
Y. Yokoyama et al. / Journal of Hepatology 35 (2001) 29±36
Fig. 2. mRNA expression of ETA and ETB receptors in the liver tissue analyzed by semi-quantitative RT-PCR. Samples were harvested after 2 weeks of PPVL or sham operation. The expressions of ETA (A) and ETB (B) mRNA levels were analyzed by RT-PCR using speci®c primers. Representative gels and ribosomal protein S9 as house keeping gene are shown on top. Average intensities of the bands analyzed by densitometry are shown in the bar graphs. Data are the means ^ SEM of four separate experiments in sham and PPVL. *P , 0.05 vs. sham.
3.4. The effects of ETB receptor stimulation by IRL 1620 In order to determine the functional importance of the increased ETB receptors, the selective ETB agonist, IRL 1620, was infused through the portal vein. IRL 1620, Suc[Glu9, Ala11, 15] Endothelin-1 (8-21), is a potent and selective ETB agonist with Ki of 16 nM. Microcirculatory perfusion was monitored and recorded with intravital microscopy during the infusion. IRL 1620 was infused either with low dose (1 pmol/min/100 g body weight) for 10 min or high dose (5 pmol/min per 100 g body weight) for 20 min. Portal
infusion of IRL 1620 did not change the MAP or HR in the low or high dose group (data not shown). The increases in PVP were minimal and almost identical in sham and PPVL with low dose of IRL 1620 (Fig. 5A). Although, the high dose caused a substantially higher increase in PVP in sham, the peak response in PPVL was not different between the high dose and the low dose (Fig. 5B). Interestingly, the signi®cantly lower increase in PVP in PPVL than sham was followed by a sustained decrease. Different responses were observed in the hepatic microcirculation following the low dose and the high dose of IRL 1620 (Fig. 6). With the
Fig. 3. Baseline hemodynamic data of sham (opened bar) and PPVL (hatched bar). Mean arterial pressure (MAP, A), heart rate (HR, B), and portal venous pressure (PVP, C) were measured at baseline. Data are the means ^SEM of six separate experiments in sham and PPVL. *P , 0.05 vs. sham.
Y. Yokoyama et al. / Journal of Hepatology 35 (2001) 29±36
33
Fig. 4. Comparison of hepatic microhemodynamics between sham (opened bar) and PPVL (hatched bar). 0.2 ml of ¯uorescein isothiocyanate (FITC)labeled RBCs was infused from the carotid artery. Number of perfused sinusoids (Ps, A) in 150 mm were counted on the screen during the video playback. Sinusoidal diameter (Ds, B) and red blood cell velocity (VRBC, C) were measured by of¯ine of recorded images using Image-Pro. Volumetric ¯ow (VF, D) and perfusion index (Pi, E) were calculated by the following equation: VF VRBC*p *(Ds/2) 2 (picoliter/s); PI Ps*VF (picoliter/s per 150 mm). Data are the means ^SEM of six separate experiments in sham and PPVL.
low dose, no signi®cant changes in sinusoidal hemodynamics were found in sham following infusion of IRL 1620 (Fig. 6A). However, the same low dose caused a
severe sinusoidal constriction, decreases of VRBC and Ps, and as a result, a signi®cant reduction in perfusion index in PPVL. Approximately 50% of sinusoids were shut down
Fig. 5. PVP changes following infusion of ETB agonist (IRL 1620) with low dose (1 pmol/min per 100 g for 10 min, A) and high dose (5 pmol/min per 100 g for 20 min, B). PVP was monitored through the double lumen catheter inserted into the portal vein. Data are the means ^ SEM of six separate experiments in sham and PPVL. *P , 0.05 versus baseline PVP. #P , 0.05 vs. PPVL.
34
Y. Yokoyama et al. / Journal of Hepatology 35 (2001) 29±36
4. Discussion
Fig. 6. Effects of portal infusion of ETB receptor agonist IRL 1620 on hepatic microhemodynamics. IRL 1620 was infused either with low dose (1 pmol/min per 100 g for 10 min, A) or high dose (5 pmol/min per 100 g for 20 min, B) in 100 ml/min per kg of body weight. Data are shown as means ^ SEM of % changes from the baseline. *P , 0.05 vs. sham. Ps, number of perfused sinusoids/150 mm; Ds, sinusoidal diameter; VRBC, red blood cell velocity; VF, volumetric ¯ow; PI, perfusion index.
and more than a 70% decrease in PI was observed at the end of the infusion. These results strongly suggest that the ETB2mediated constriction was predominant at the sinusoidal level of the PPVL liver in response to the ETB agonist. With the high dose of IRL 1620, hepatic microcirculatory perfusion was substantially reduced in both sham and PPVL as re¯ected by the signi®cant decreases in VF and PI (Fig. 6B, 70 and 85% decreases in PI, respectively). Both groups demonstrated signi®cant decreases in Ps and VRBC in response to the high dose, but greater percentage decreases in Ps and PI were seen in PPVL than sham. Interestingly, no change in Ds was observed in PPVL with the high dose, which is contrary to what was observed with the low dose, suggesting an activation of ETB1 receptors and/or extrasinusoidal sites of action by IRL 1620.
In the present study, we investigated the effect of prehepatic portal hypertension on the changes in ET receptors in the liver and the hepatic microcirculatory response to the speci®c ETB agonist IRL 1620 using PPVL in rats. It was demonstrated that expression of ETB receptors was increased both at functional protein and mRNA levels after PPVL. Altered hepatic expression of ET receptor subtypes in PPVL resulted in a substantial change in the response of the hepatic microcirculation to ETB receptor agonists. Prehepatic portal vein stenosis not only exerts direct effects on the splanchnic circulation, but also on the hepatic circulation by substantially decreasing portal blood ¯ow. Our recent study showed that neovascularization from the hepatic arterial system occurs in the liver following PPVL suggesting vascular remodeling, possibly induced by the decreased portal blood ¯ow [4]. It is conceivable that the restriction of the portal ¯ow by PPVL and the subsequent vascular remodeling in the liver would also give rise to changes in regulation of the hepatic microcirculation. Since ET is one of the most active vasoregulators in the hepatic microcirculation, we focused on the effect of PPVL on expression of ETA and ETB receptors in the liver. PPVL did not change the total number of ET receptors in the liver but signi®cantly altered the ratio of ETA and ETB receptors. Decreased ETA receptors and increased ETB receptors were not only shown after PPVL at the functional protein level but also con®rmed at the mRNA level by RT-PCR. The mechanisms by which the ratio of ETA and ETB receptors were altered in PPVL induced portal hypertension are presently unclear. However, previous studies have shown an increased proportion of ETB receptors in the liver under a number of stress conditions such as endotoxemia and I/R injury [8]. It appears that upregulation of ETB receptors is a common response in the liver to stress conditions. In order to determine how the proportional changes in ET receptor subtypes translates to the change in functional response of the hepatic microcirculation, we determined the microhemodynamics of the hepatic portal circulation using in vivo microscopy. Low dose of IRL 1620 (1 pmol/ min per 100 g for 10 min) induced a slight increase in PVP in both sham and PPVL and no visible changes in the sinusoidal hemodynamics in sham (Fig. 6A). However, in PPVL, it caused a substantial reduction in sinusoidal perfusion as a result of sinusoidal constriction and signi®cant decreases in Ps and VRBC. These results strongly suggest that the increased ETB receptors following PPVL is in part due to the increased ETB2 receptors in the sinusoids, most likely on HSC, activation of which results in sinusoidal constriction. Nonetheless, the question remains: how does the severe reduction of sinusoidal perfusion occur after IRL 1620 without signi®cantly changing PVP? The answer probably lies in ¯ow shunting. It was observed that the ¯ow was shut down in almost a half of the sinusoids, accompanied by
Y. Yokoyama et al. / Journal of Hepatology 35 (2001) 29±36
sinusoidal constrictions and a 70% reduction of PI in PPVL. It is possible that blood ¯ow shunted through the collateral vessels under such a condition. Interestingly, high dose of IRL 1620 induced different responses. PVP in sham increased substantially in response to the ETB receptor agonist whereas only a minimal increase in PVP was observed in PPVL (Fig. 5B). The increased PVP in sham was accompanied by a signi®cant reduction of the sinusoidal perfusion (70% decrease in PI) indicating that the concentration of IRL 1620 used was overwhelming and elicited primarily an extrasinusoidal effect since the sinusoidal diameter did not change signi®cantly (Fig. 6B). In contrast to sham rats, there was a minimal increase in PVP in PPVL followed by a sustained decreased phase. However, there was a dramatic sinusoidal ¯ow shutdown with an 85% decrease in PI in PPVL following the high dose. The mechanisms underlying the different responses at the portal and sinusoidal levels are not clear. However, we speculate that the mixed actions mediated by both ETB1 and ETB2 receptors and portal collateral ¯ow shunting are responsible for the different responses [18,19]. The blunted PVP increase and the subsequent decrease are likely due to the portal-collateral ¯ow shunting and the ETB1 receptormediated action. The drastic effect of the high dose of IRL 1620 on the sinusoidal ¯ow is caused by the ETB2 receptor-mediated constriction, acting on the sinusoids and the extrasinusoidal sites especially the post-sinusoidal sites. Prehepatic portal hypertension often produces no signi®cant parenchymal damage. Indeed, no visible changes were observed in the hepatic microhemodynamics in PPVL compared to sham in spite of portal hypertension (Fig. 4). However, the present study suggests that patients with this condition may be susceptible to secondary stresses where hepatic ET level is elevated such as LPS-induced endotoxemia [20±22], hemorrhagic resuscitation [23] and I/R [24,25]. Under these stress conditions, the hepatic microcirculatory dysfunction is prone to occur as a consequence of the acute stress-induced elevated ET acting upon the increased ETB2 receptors in the sinusoids. In summary, the results of our ET receptor binding assay and RT-PCR showed a signi®cant up-regulation of ETB receptors and a decrease in ETA receptors in the liver following 2 weeks of PPVL. These changes were ultimately responsible for the increased sensitivity of the hepatic microcirculation to the ETB receptor agonist IRL 1620 in causing microcirculatory ¯ow disruption in PPVL. These results suggest that prehepatic portal hypertension may be susceptible to the hepatic microcirculatory dysfunction following secondary stress conditions. Acknowledgements The study was supported by the grant from National Institutes of Health (DK38201) and faculty grants of University of North Carolina at Charlotte.
35
References [1] Kralik J. Long-term prognosis of prehepatic portal hypertension. Acta Univ Palacki Olomuc Fac Med 1992;133:63±66. [2] Vorobioff J, Bredfeldt JE, Groszmann RJ. Hyperdynamic circulation in portal-hypertensive rat model: a primary factor for maintenance of chronic portal hypertension. Am J Physiol 1983;244:G52±G57. [3] Murakami S, Ohno T, Bernardo JF, Pfeifer CA, Li T, Zhang Y, Dubey RK, Branch RA, Sabra R. Reduced liver function is the trigger for renal sodium retention following portal vein ligation in the rat. J Gastroenterol Hepatol 1996;11:850±856. [4] Yokoyama Y, Baveja R, Sonin N, Clemens MG, Zhang JX. Hepatic neovascularization after partial portal vein ligation: novel mechanism of chronic regulation of blood ¯ow. Am J Physiol Gastrointest Liver Physiol 2001;280(1):G21±G31. [5] Bauer M, Zhang JX, Bauer I, Clemens MG. Endothelin-1 as a regulator of hepatic microcirculation: sublobular distribution of effects and impact on hepatocellular secretory function. Shock 1994;1:457± 465. [6] Zhang JX, Bauer M, Clemens MG. Vessel and target cell speci®c actions of endothelin-1 and endothelin-3 in rat liver. Am J Physiol 1995;269:G269±G277. [7] Rockey D. The cellular pathogenesis of portal hypertension: stellate cell contractility, endothelin, and nitric oxide. Hepatology 1997;25:2± 5. [8] Bauer M, Bauer I, Sonin NV, Kresge N, Baveja R, Yokoyama Y, Harding D, Zhang JX, Clemens MG. Functional signi®cance of endothelin B receptors in mediating sinusoidal and extrasinusoidal effects of endothelins in the intact rat liver. Hepatology 2000;31(4):937±947. [9] Mallat A, Lotersztajn S. Multiple hepatic functions of endothelin-1: physiopathological relevance. J. Hepatol. 1996;25:405±413. [10] Douglas SA, Meek TD, Ohlstein EH. Novel receptor antagonists welcome a new era in endothelin biology. Trends Pharmacol Sci 1994;15:313±316. [11] Pollock DM, Keith TL, Highsmith RF. Endothelin receptors and calcium signaling. FASEB J 1995;9:1196±1204. [12] Leivas A, Jimenez W, Bruix J, Boix L, Bosch J, Arroyo V, Rivera F, Rodes J. Gene expression of endothelin-1 and ET(A) and ET(B) receptors in human cirrhosis: relationship with hepatic hemodynamics. J Vasc Res 1998;35:186±193. [13] Yokoyama Y, Baveja R, Sonin N, Nakanishi K, Zhang JX, Clemens MG. Altered endothelin receptor subtype expression in hepatic injury after ischemia/reperfusion. Shock 2000;13(1):72±78. [14] Sonin NV, Garcia-Pagan JC, Nakanishi K, Zhang JX, Clemens MG. Patterns of vasoregulatory gene expression in the liver response to ischemia/reperfusion and endotoxemia. Shock 1999;11:175±179. [15] Clemens MG, McDonagh PF, Chaudry IH, Baue AE. Hepatic microcirculatory failure following ischemia and reperfusion: improvement with ATP-MgCl2 treatment. Am J Physiol. 1985;248:H804±H811. [16] Zimmerhackl B, Parekh N, Brinkhus H, Steinhausen M. The use of ¯uorescent labeled erythrocytes for intravital investigation of ¯ow and local hematocrit in glomerular capillaries in the rat. Int J Microcirc Clin Exp 1983;2:119±129. [17] Bauer I, Bauer M, Pannen BHJ, Leinwand MJ, Zhang JX, Clemens MG. Chronic ethanol consumption exacerbates liver injury following hemorrhagic shock: role of sinusoidal perfusion failure. Shock 1995;4:324±331. [18] Halvorsen JF, Myking AO, Tvete S. Porto-hepatic bypass by organohepatopexy in rats with prehepatic portal hypertension. Formation of new venous porto-hepatic collaterals following omento-, jejuno-, and splenohepatopexy, and their effect on portal vein pressure. Eur Surg Res 1979;11:254±266. [19] Ingles AC, Legare DJ, Lautt WW. Development of portacaval shunts in portal-stenotic cats. Can J Physiol Pharmacol 1993; 71:671±674.
36
Y. Yokoyama et al. / Journal of Hepatology 35 (2001) 29±36
[20] Eakes AT, Howard KM, Miller JE, Olson MS. Endothelin-1 production by hepatic endothelial cells: characterization and augmentation by endotoxin exposure. Am J Physiol 1997;272:G605±G611. [21] Ohuchi T, Tada K, Akamatsu K. Endogenous ET-1 contributes to liver injury induced by galactosamine and endotoxin in isolated perfused rat liver. Am J Physiol 1995;268:G997±G1003. [22] Ros J, Leivas A, Jimenez W, Morales M, Bosch-Marce M, Arroyo V, Rivera F, Rodes J. Effect of bacterial lipopolysaccharide on endothelin-1 production in human vascular endothelial cells. J Hepatol 1997;26:81±87. [23] Pannen BH, Bauer M, Noldge-Schomburg GF, Zhang JX, Robotham
JL, Clemens MG, Geiger KK. Regulation of hepatic blood ¯ow during resuscitation from hemorrhagic shock: role of NO and endothelins. Am J Physiol 1997;272:H2736±H2745. [24] Nakamura S, Nishiyama R, Serizawa A, Yokoi Y, Suzuki S, Konno H, Baba S, Muro H. Hepatic release of endothelin-1 after warm ischemia. Reperfusion injury and its hemodynamic effect. Transplantation 1995;59:679±684. [25] Ota T, Hirai R, Urakami A, Soga H, Nawa S, Shimizu N. Endothelin1 levels in portal venous blood in relation to hepatic tissue microcirculation disturbance and hepatic cell injury after ischemia/reperfusion. Surg Today 1997;27:313±320.