1. Endothelin receptor blockers reduce shunting and
angiogenesis in cirrhotic rats
Shao-Jung Hsu*,†,‡
, Te-Yueh Lin§
, Sun-Sang Wang*,¶
, Chiao-Lin Chuang*,§
, Fa-Yauh Lee*,‡
, Hui-Chun Huang*,‡,§
,
I-Fang Hsin†,**
, Jing-Yi Lee†
, Han-Chieh Lin*,‡
and Shou-Dong Lee*,††
*
Faculty of Medicine, National Yang-Ming University School of Medicine, Taipei, Taiwan, †
Institute of Pharmacology,
National Yang-Ming University School of Medicine, Taipei, Taiwan, ‡
Division of Gastroenterologyand Hepatology, Taipei
Veterans General Hospital, Taipei, Taiwan, §
Division of General Medicine, Department of Medicine, Taipei Veterans General
Hospital, Taipei, Taiwan, ¶
Department of Medical Affair and Planning, Taipei Veterans General Hospital, Taipei, Taiwan,
**
Endoscopy Center for Diagnosis and Treatment, Taipei Veterans General Hospital, Taipei, Taiwan, ††
Division of
Gastroenterology, Department of Medicine, Cheng Hsin General Hospital, Taipei, Taiwan
ABSTRACT
Background Angiogenesis plays a pivotal role in splanchnic hyperaemia and portosystemic collateral formation
in cirrhosis. Endothelin-1 (ET-1), an endothelium-derived vasoconstrictor, has also been implicated in the
pathogenesis of cirrhosis and portal hypertension.
Design This study aimed to survey the influences of ET-1 in cirrhosis-related angiogenesis. Common bile duct
ligation was performed on Spraque–Dawley rats to induce cirrhosis. Since the 14th day after the operation, rats
randomly received distilled water (DW, control), bosentan [a nonselective ET receptor (ETR) blocker] or
ambrisentan (a selective ETAR blocker) for 4 weeks. On the 43rd day, portal and systemic haemodynamics, liver
biochemistry, portosystemic shunting degree, mesenteric vascular density, mRNA and/or protein expressions
of relevant angiogenic factors were evaluated.
Results In cirrhotic rats, bosentan significantly reduced portal pressure. Ambrisentan did not influence
haemodynamics and liver biochemistry. Both of them significantly improved the severity of portosystemic
collaterals and decreased the mesenteric vascular density. Compared with the DW-treated cirrhotic rats,
splenorenal shunt and mesenteric inducible nitric oxide synthase (iNOS), cyclooxygenase 2 (COX2), vascular
endothelial growth factor mRNA expressions and mesenteric iNOS, COX2, VEGF, phospho-VEGF receptor 2,
Akt and phospho-Akt protein expressions were down-regulated in both groups.
Conclusions In rats with liver cirrhosis, both nonselective and selective ETAR blockade ameliorate the severity
of portosystemic shunting and mesenteric angiogenesis via the down-regulation of VEGF pathway and relevant
angiogenic factors. ET receptors may be targeted to control the severity of portosystemic collaterals and
associated complications in cirrhosis.
Keywords Angiogenesis, endothelin-1, liver cirrhosis, portosystemic collaterals, vascular endothelial growth
factor.
Eur J Clin Invest 2016; 46 (6): 572–580
Introduction
Liver cirrhosis increases intrahepatic resistance through
fibrosis and vasoconstriction. Taken the hyperdynamic circu-
lation with increased portal blood flow together, portal
hypertension and portosystemic collaterals develop gradually
[1]. The immature collateral vessels are prone to rupture with
high morbidity and mortality, such as gastroesophageal
varices. Traditionally, the increase in splanchnic inflow was
considered a consequence of vascular dilatation in which
nitric oxide (NO) plays a role [1]. Nevertheless, recent studies
have proposed that angiogenesis, the generation of new blood
vessels, is involved in the process. Inhibition of pathological
angiogenesis effectively ameliorated severity of shunting [2].
Furthermore, even vascular contractility was unaffected,
alleviation of angiogenesis still decreased shunting degree by
suppression of vascular endothelial growth factor (VEGF), a
potent angiogenic factor [3]. Therefore, angiogenic factor
modulation can be a feasible strategy to alleviate mesenteric
angiogenesis and portosystemic collaterals.
Endothelin (ET) participates in pathogenesis of portal
hypertension. The ET family consists of three 21-amino acid
572 ª 2016 Stichting European Society for Clinical Investigation Journal Foundation
DOI: 10.1111/eci.12636
ORIGINAL ARTICLE

2. peptides (ET-1, ET-2 and ET-3) and two G-protein-coupled
receptors (ETAR and ETBR). Endothelin is up-regulated after
acute liver injury [4], which results in stellate cell activation,
fibrogenesis and portal hypertension. Furthermore, ET-1 mod-
ulates neovascularization. Elevated ET-1 expression is associ-
ated with VEGF/VEGFR over-expression and
neovascularization in ovarian carcinoma [5]. Our previous
study revealed that endothelin induced ETAR-mediated vaso-
constriction of collaterals in portal hypertensive animals [6],
suggesting the dominant role of ETAR in this vascular bed.
Nevertheless, the influences of ET receptors blockade on
mesenteric angiogenesis and portosystemic shunting in cir-
rhosis have not been surveyed.
Considering the divergent interactions among ET-1, angio-
genic factors and portal hypertension, we investigated the rel-
evant effects and mechanism of nonselective and selective
ETAR blockade in common bile duct ligation (BDL)-induced
cirrhotic rats.
Material and methods
The detailed description is shown in the Supporting Informa-
tion with the Materials and Methods as the followings: real-
time quantitative RT-PCR; and Western analysis [7].
Animal model
Biliary cirrhosis was induced 4 weeks after BDL in male
Sprague–Dawley rats [8]. This study was approved by Taipei
Veterans General Hospital Animal Committee. The principle of
laboratory animal care [Guide for the Care and Use of Labo-
ratory Animals (1985), DHEW Publication no. (NIH) 85–23:
Office of Science and Health Reports; DRR/NIH, Bethesda,
MD, USA] was followed.
Systemic and portal haemodynamics
Haemodynamic parameters were recorded by a multi-channel
recorder (model RS 3400; Gould Inc., Cupertino, CA, USA) with
catheterization method. The femoral artery was cannulated
with a PE-50 catheter, and continuous recordings of mean
arterial pressure and heart rate (HR) were made. The abdomen
was opened with a mid-line incision, and the mesenteric vein
was cannulated with a PE-50 catheter for portal pressure (PP)
measurements. The external zero reference was placed at the
mid-portion of the rat [6].
Measurements of plasma levels of biochemistry
parameters
Plasma levels of aspartate transaminase (AST), alanine
transaminase (ALT) and total bilirubin were determined by
VITROS DT60II and DTSC II modules (Ortho-Clinical Diag-
nostic Inc., Raritan, NJ, USA).
Plasma VEGF Level determination
The plasma levels of VEGF were measured using commercially
available enzyme-linked immunoabsorbent assay kits (R&D
Systems Inc., Minneapolis, MN, USA) according to the manu-
facturer’s instructions. The intensity of the colour was mea-
sured to the absorbance of 450–600 nm with a Bio-kinetics
Reader (Bio-Tek Instruments Inc., Winooski, VT, USA).
Portosystemic shunting
Portosystemic shunting was determined as previously [9] and
substituting colour for radioactive microspheres. Shunting was
calculated as lung/(liver+lung) microspheres. Colour micro-
spheres provide similar results as radioactive microspheres
[10].
Mesenteric vascular density
Mesenteric angiogenesis was quantified by CD31-labelled
microvascular networks in rat mesenteric connective tissue
windows according to the previous study [2]. In brief, two
mesenteric windows (wedge-shaped regions of connective tis-
sue bordered by the intestinal wall and the ileal blood vessel
pairs) were dissected free from each rat, washed in PBS, dried
on gelatin slides and fixed in 100% MeOH (À20 °C for 30 min).
Slides were then incubated overnight at 4 °C with the primary
antibody mouse anti-rat CD31-biotin [1 : 200; AbD Serotec,
Oxford, UK]. Then, secondary antibody [CY2-conjugated
streptavidin, 1 : 1000; Jackson ImmunoResearch, West Grove,
PA, USA] was applied for 1 h at room temperature. At least
four sets of data were obtained for each mesenteric window.
(9100)-magnification immunofluorescent images were assessed
using an upright fluorescent microscope (AX80, Olympus,
Japan) with charge-couple device (QICAM, High-performance
IEEE 1394 FireWireTM Digital CCD Camera, Q IMAGING;
Surrey, BC, Canada) and thresholded by IMAGEJ software
[available for download from the National Institutes of Health
(http://rsb.info.nih.gov/ij/)]. The vascular length was manu-
ally measured with the pencil tool and the vascular area auto-
matically with histogram function, respectively. The unit of
vascular length per unit area of mesenteric window would be
lm/lm2
= per lm, and the vascular area per unit area of
mesenteric window, actually, could be pixel/pixel without
being converted to lm2
/lm2
.
Hepatic fibrosis determination with Sirius red
staining
Liver paraffin section was stained with Sirius red staining kit
(Polysciences Inc., Warrington, PA, USA). IMAGEJ was used to
measure the percentage of Sirius red-stained area [2]. Briefly,
greyscale image was used; then, the red-stained collagen was
isolated using thresholding function. After that, the
European Journal of Clinical Investigation Vol 46 573
ENDOTHELIN AND SHUNTING IN CIRRHOSIS

3. thresholded area was measured and shown as the percentage
of thresholded area per image.
Haematoxylin and eosin staining
Mesenteries were fixed in 10% formalin, embedded in paraffin,
sectioned in 5 lm and stained with haematoxylin-eosin (H&E).
Study protocol
Rats received oral gavage of vehicle (distilled water, DW),
bosentan (nonselective ETAR and ETBR blocker, 100 mg/kg/
day) or ambrisentan (selective ETAR blocker, 35 mg/kg/day,)
form the 15th to 43rd day after BDL. The doses were deter-
mined based on in vivo pharmacology studies [11]. On the 43rd
day, the followings were evaluated in three series:
1 First series:
i Haemodynamics;
ii Plasma liver biochemistry ALT, AST, total bilirubin and
VEGF concentrations;
iii Hepatic fibrosis determination with Sirius red staining;
iv Splenorenal shunt (the most prominent intraabdominal
shunt in rodents) and mesenteric RNA and protein
expressions of iNOS, eNOS, COX1, COX2, VEGF,
VEGFR1, VEGFR2, p-VEGFR2, Akt, p-Akt, ERK and
p-ERK.
2 Second series: portal-systemic shunting.
3 Third series: mesenteric angiogenesis determination with
mesenteric vascular density.
Drugs
Bosentan was purchased from Actelion Pharmaceuticals Ltd.
(Allschwil, Switzerland), and ambrisentan was kindly provided
by GlaxoSmithKline plc. (Middlesex, UK).
Data analysis
Statistical analysis was performed using the SPSS software
(version 19.0; IBM Inc., New York, NY, USA). All results are
expressed as mean Æ SEM. Statistical analyses were performed
using one-way ANOVA. Fisher’s least significant difference (LSD)
was applied for post-hoc test. Results are considered statistically
significant at a two-tailed P-value < 0Á05. Figures were gener-
ated by Prism 5 for Windows (GraphPad Software, Inc., La
Jolla, CA, USA).
Results
Haemodynamics, liver biochemistry and VEGF
Table 1 shows the body weight (BW) and haemodynamic
parameters of BDL rats after DW (control), bosentan or
ambrisentan treatments. Bosentan significantly reduced BW
(P = 0Á036 vs. DW) and PP (P = 0Á036 vs. DW; P = 0Á023 vs.
ambrisentan).
Figure 1 depicts liver biochemistry and VEGF concentra-
tions. Bosentan and ambrisentan did not influence ALT, AST
and total bilirubin levels (P > 0Á05 among groups) but
decreased VEGF concentration (DW vs. bosentan vs.
ambrisentan: 49Á6 Æ 1Á9 vs. 41Á5 Æ 1Á6 vs. 41Á4 Æ 2Á2 pg/mL,
bosentan vs. DW: P = 0Á007; ambrisentan vs. DW: P = 0Á004).
Portosystemic shunting
Figure 2(a) reveals that bosentan and ambrisentan alleviated
the severity of portosystemic shunting, which was more
prominent by bosentan (DW vs. bosentan vs. ambrisentan:
80Á9 Æ 2Á4 vs. 50Á9 Æ 2Á7 vs. 72Á4 Æ 2Á6%, bosentan vs. DW:
P < 0Á001; ambrisentan vs. DW: P = 0Á024, bosentan vs.
ambrisentan: P < 0Á001).
Angiogenesis determination with mesenteric
vascular density
Mesenteric angiogenesis and increased intrahepatic resistance
are two major factors aggravating the severity of portosystemic
shunting in cirrhosis. The impacts of bosentan and ambrisentan
on angiogenesis were therefore evaluated with mesenteric
vascular density. The representative diagrams are shown in
Fig. 2(d). Bosentan and ambrisentan significantly decreased
vascular length per unit window area (lmÀ1
) (Fig. 2b, DW vs.
bosentan vs. ambrisentan: 2Á09 Æ 0Á31 vs. 1Á29 Æ 0Á14 vs.
1Á24 Æ 2Á4, bosentan vs. DW: P = 0Á025; ambrisentan vs. DW:
P = 0Á022) and total vascular area per unit area of mesenteric
window (%) (15Á07 Æ 1Á32 vs. 9Á28 Æ 1Á16 vs. 6Á43 Æ 1Á16,
bosentan vs. DW: P = 0Á004; ambrisentan vs. DW: P < 0Á001).
Hepatic fibrosis determination with Sirius red
staining
Increased intrahepatic resistance is the other main factor
ascribed for collaterals formation. Because hepatic fibrosis is the
structural factor that increases intrahepatic resistance in
Table 1 Body weight and haemodynamic parameters in BDL
rats with DW, bosentan or ambrisentan treatment
n BW (g)
MAP
(mmHg)
HR
(beats/
min)
PP
(mmHg)
DW 14 367 Æ 10 105 Æ 4 276 Æ 18 22Á1 Æ 1Á6
bosentan 9 316 Æ 11a,b
102 Æ 2 313 Æ 13 18Á1 Æ 1Á0a,b
Ambrisentan 11 373 Æ 11 101 Æ 3 308 Æ 8 21Á9 Æ 1Á0
DW, distilled water (control); BW, body weight; MAP, mean arterial pressure;
HR, heart rate; PP, portal pressure.
a
P < 0Á05 bosentan vs. DW; b
P < 0Á05 bosentan vs. ambrisentan.
574 ª 2016 Stichting European Society for Clinical Investigation Journal Foundation
S.-J. HSU ET AL. www.ejci-online.com

4. cirrhosis, it was evaluated by Sirius red staining. Figure 2(c)
revealed the statistical analysis of hepatic fibrosis ratio. Both
bosentan and ambrisentan alleviated the severity of liver
fibrosis (%) (Sirius red staining area: DW vs. bosentan vs.
ambrisentan: 15Á90 Æ 0Á57 vs. 10Á58 Æ 0Á79 vs. 9Á70 Æ 0Á31,
bosentan vs. DW: P < 0Á001; ambrisentan vs. DW: P < 0Á001).
Histology showed characteristic cirrhotic change in BDL rats,
including bile ductule proliferation, hepatocytic necrosis and
fibrotic band formation (Fig. 2e and f).
Angiogenic factors mRNA expressions over
splenorenal shunts and mesentery
To elucidate how bosentan and ambrisentan ameliorated
pathological angiogenesis, angiogenic factors mRNA expres-
sions were determined.
Splenorenal shunt is the most prominent intraabdominal
shunting vessel in rodents. Figure 3(a) depicts the splenorenal
shunt angiogenic mRNA expressions. Compared with DW,
bosentan and ambrisentan significantly decreased the iNOS
and VEGF mRNA expressions. In addition, ambrisentan
decreased the COX2 expression and bosentan decreased
VEGFR2 expression.
Figure 3(b) shows the mesenteric mRNA expressions.
Bosentan and ambrisentan decreased the iNOS, COX2 and
VEGF expressions.
Angiogenic factors protein expressions over
mesentery
To further validate the effects of bosentan and ambrisentan on
angiogenic factors, the relevant protein expressions over
mesentery were determined. Figure 4 reveals angiogenic fac-
tors protein expressions. Compared with the DW, bosentan and
ambrisentan significantly decreased Akt, iNOS, COX2, VEGF
and p-VEGFR2 protein expressions.
Discussion
In this study, both bosentan and ambrisentan, a mixed ETAR
and ETBR antagonist and a selective ETAR antagonist, amelio-
rated the severity of portosystemic shunting in rats with
BDL-induced liver cirrhosis. Furthermore, the antiangiogenic
effects acted through the down-regulation of VEGF pathway
and relevant angiogenic factors. The mechanism of this study is
illustrated in Fig. S1.
Two possible origins of collateral vessels in cirrhosis and
portal hypertension have been proposed: passive dilatation of
pre-existing venous channels and/or neovascularization [12].
As ET-1 is a potent vasoconstrictor and ET-1 receptor blockers
induce vasodilatation [13], angiogenesis is more likely the
mechanism implicated in the improvement of collaterals
exerted by bosentan and ambrisentan treatments. Actually,
angiogenesis has been found with pivotal roles in the aggra-
vation of portosystemic collaterals, and antiangiogenesis is
nowadays considered a promising treatment strategy in cir-
rhosis [14], as in cancer therapy. Regarding the relevant studies
in malignancy, it has been found that mixed ETAR and ETBR
antagonist inhibited tumour vascularization in breast cancer
[15]. Furthermore, the inhibition of human ovarian tumour
growth by ETAR antagonist was associated with reduced VEGF
and microvessel density [16]. Although the mechanisms of
angiogensis in malignancy and cirrhosis are not completely the
Figure 1 Circulating concentrations of
liver biochemistry parameters and VEGF in
BDL rats with DW (control), bosentan or
ambrisentan treatment. Bosentan and
ambrisentan did not significantly influence
alanine transaminase (ALT), aspartate
transaminase (AST) and total bilirubin.
Both bosentan and ambrisentan
significantly decreased plasma VEGF
concentration (*P < 0Á05).
European Journal of Clinical Investigation Vol 46 575
ENDOTHELIN AND SHUNTING IN CIRRHOSIS

5. same, it suggests that endothelin receptors inhibit angiogenesis
at least partly through interfering VEGF pathway. Consistently,
we found in the current study that both mixed ET receptor
antagonist and selective ETAR antagonist ameliorated angio-
genesis in cirrhotic rats via VEGF pathway and relevant
angiogenic factors down-regulation.
Figure 2 Effects of endotheline receptor antagonists on collateral shunting degree, mesenteric vascular density and hepatic
fibrosis in cirrhotic rats. (a) Compared to DW, bosentan and ambrisentan significantly decreased shunting, which was more
prominent by bosentan. (b) Bosentan and ambrisentan significantly decreased the pathological mesenteric angiogenesis as
compared with that of the DW group. (c) Bosentan and ambrisentan significantly decreased the liver fibrosis degree compared to
DW group. (d) The representative CD31 immunofluorescent staining images of mesenteric windows of BDL rats with DW (control),
bosentan, or ambrisentan treatment. (e, f) The representative figures of hepatic Sirius red staining and H & E staining. Characteristic
pathological change of cirrhosis induced by BDL, including ductular reaction of portal area and significant fibrotic bridging could be
observed in all groups. The fibrotic area significantly decreased in bosentan and ambrisentan treated groups. *P < 0Á05 compared to
DWtreated control group. **Scale bar = 50 lm.
576 ª 2016 Stichting European Society for Clinical Investigation Journal Foundation
S.-J. HSU ET AL. www.ejci-online.com

6. Bosentan and ambrisentan decreased the mRNA expressions
of iNOS, COX2 and VEGF in splenorenal shunt. Regarding the
influences of ET-1 on NO and prostaglandin synthesis, it has
been reported that bosentan down-regulated iNOS expression
[17]. On the other hand, ET-1 blockade prevented COX2
induction in rats [18]. In fact, both NO and the vasodilatory
prostanoid synthesized by COX participate in the modulation
of portosystemic collaterals in portal hypertension [12,19]. As
splenorenal shunt is the most prominent portosystemic collat-
eral vessel in cirrhotic rats, the results indicate a direct sup-
pression of angiogenic factors by ETR blockade on collateral
vessel, which results in the ameliorated severity of portosys-
temic shunting.
Bosentan and ambrisentan down-regulated VEGF and
p-VEGFR2 expressions in splenorenal shunt and mesentery. In
ovarian carcinoma cells, ET-1 induced VEGF transcription via
ETAR, which was inhibited by ETAR antagonist [20]. Consistent
result showed that ETAR blocker decreased the growth of
ovarian xenografts in mice, which was associated with
decreased VEGF production and microvessel density [21]. The
current study found that mesenteric angiogenic factors protein
expressions were significantly decreased by bosentan and
Figure 3 mRNA expression of iNOS, eNOS, COX1, COX2, VEGF, VEGFR-1 and VEGFR-2 in (a) splenorenal shunts and (b)
mesentery in BDL rats with DW (control), bosentan or ambrisentan treatment. (a) In splenorenal shunt, bosentan and ambrisentan
significantly decreased the iNOS and VEGF mRNA expressions. Ambrisentan also decreased the COX2 expression. (b) Both
bosentan and ambrisentan decreased the mesenteric iNOS, COX2 and VEGF mRNA expressions. *P < 0Á05 compared to DW-treated
control group.
European Journal of Clinical Investigation Vol 46 577
ENDOTHELIN AND SHUNTING IN CIRRHOSIS

7. ambrisentan as compared with the control group, but no sig-
nificant difference was identified between bosentan and
ambrisentan groups. This supports the notion that ETAR
blockade is mainly responsible for the mesenteric antiangio-
genic effect of bosentan. Nevertheless, the severity of por-
tosystemic shunting was significantly less in bosentan group as
compared to that in ambrisentan group, implying the partici-
pation of ETBR receptor in the process. In line with this finding,
in invasive breast cancer, the expression of ETBR was associated
with increased VEGF expression and vascularity [22]. Further-
more, ETBR antagonist suppressed neovascularisation and
VEGF expression [23,24].
In this study, bosentan and ambrisentan did not significantly
influence the systemic and portal haemodynamic parameters
except that bosentan significantly reduced the PP. Portal pres-
sure is determined by the net effect of portal inflow, intrahep-
atic resistance and portosystemic collateral vascular resistance.
Considering the current data, the reduced mesenteric angio-
genesis may have decreased portal inflow. A previous study
has also indicated that bosentan decreased PP by reducing
hepatocollateral vascular resistance in rats with cirrhosis [25].
Furthermore, it has been found that endothelin antagonism in
rats with liver fibrosis reduced hepatic stellate cells activation
and matrix production [26]. Consistently, in the current study,
Sirius red staining revealed that bosentan ameliorated liver
fibrosis. Therefore, it is conceivable that bosentan reduces PP
via ameliorated liver fibrosis, decreased portal inflow and
decreased hepatocollateral vascular resistance.
Angiogenesis can occur in physiological conditions such as
liver regeneration or in pathological settings such as cirrhosis
[27]. Accumulating evidences have indicated that hepatic
angiogenesis promotes fibrogenesis [28]. Furthermore, agents
that specifically inhibit angiogenesis by targeting molecules not
involved in fibrogenic pathway, such as VEGFR2, alleviate
hepatic fibrosis [29]. While these data suggest that angiogenesis
may be a requisite promoting fibrogenesis, angiogenesis may
on the contrary take place as a response to fibrosis and partic-
ipate in tissue repair. For instance, the inhibition of integrin
avb3 (avb5) decreased angiogenesis but worsened fibrosis [30].
In this study, we have found that ET receptor antagonists
exerted antiangiogenesis effect and at the same time amelio-
rated liver fibrosis. It is thus inferred that the regimen used in
Figure 4 Mesenteric protein expressions of iNOS, eNOS,
COX1, COX2, VEGF, VEGFR2 p-VEGFR2, Akt and p-Akt in BDL
rats with DW (control), bosentan or ambrisentan treatment.
Compared with the DW-treated group, bosentan and
ambrisentan significantly decreased Akt, p-Akt, iNOS, COX2,
VEGF, and p-VEGFR2 protein expressions. *P < 0Á05 compared
to DW-treated control group. D: DW-treated control group; B:
bosentan-treated group; A: ambrisentan-treated group.
578 ª 2016 Stichting European Society for Clinical Investigation Journal Foundation
S.-J. HSU ET AL. www.ejci-online.com

8. the current study inhibited pathological angiogenesis that was
beneficial for fibrosis resolution. Nevertheless, further clinical
investigation and correlation are required.
The side effects of ET antagonism must be taken into con-
sideration. Bosentan is potentially hepatotoxic. A postmarket-
ing surveillance showed that 7Á6% patients had elevated
aminotransferases and 3Á2% patients withdrew because of this
side effect [31]. As a result, bosentan is suggested to be avoided
in patients with moderate–to-severe liver cirrhosis [32]. On the
contrary, ambrisentan is well tolerated by pulmonary hyper-
tensive patients without noticeable liver toxicity [32]. A phase 2
study of ambrisentan use in patients with pulmonary hyper-
tension who had discontinued bosentan treatment due to ele-
vated liver enzyme levels indicated that only 1 of 31 patients
had transient enzyme elevation and recovered after dose
reduction [33]. In this study, we did not identify significant
differences of liver biochemistry among control, bosentan and
ambrisentan groups but the body weight was significantly
lower in rats with bosentan treatment. Actually, poor appetite
and fatigue are common side effects of bosentan, which may be
responsible for the body weight change. Taken together,
ambrisentan, compared with bosentan, exerts the same benefi-
cial effects on the severity of portosystemic shunting and
angiogenesis with a minor concern of liver toxicity and body
weight loss.
In conclusion, in rats with liver cirrhosis, the nonselective
ETR or selective ETAR blocker ameliorates the severity of por-
tosystemic shunting and mesenteric angiogenesis via suppres-
sion of iNOS, COX2, VEGF and p-VEGFR2 and down-
regulation of Akt signalling. The ET receptors and relevant
signalling pathways may be therapeutic targets in controlling
the severity of collaterals and complications in cirrhosis.
Acknowledgements
The authors thank Tzu-Hua Teng and Yi-Chou Chen for their
excellent technical support. This work was supported by the
grants from National Science Council, Taiwan (grant no. NSC
98-2314-B-075-028)
Address
Faculty of Medicine, National Yang-Ming University School of
Medicine, No.155, Sec.2, Li-Nong Street, Taipei 112, Taiwan (S.-J.
Hsu, S.-S. Wang, C.-L. Chuang, F.-Y. Lee, H.-C. Huang; H.-C.
Lin, S.-D. Lee); Institute of Pharmacology, National Yang-Ming
University School of Medicine, No.155, Sec.2, Li-Nong Street,
Taipei 112, Taiwan (S.-J. Hsu, I.-F. Hsin, J.-Y. Lee); Division of
Gastroenterology, Taipei Veterans General Hospital, No. 201,
Sec. 2, Shih-Pai Road, Taipei 112, Taiwan (S.-J. Hsu, F.-Y. Lee, H.-
C. Huang, H.-C. Lin); General Medicine, Department of Medi-
cine, Taipei Veterans General Hospital, No. 201, Sec. 2, Shih-Pai
Road, Taipei 112, Taiwan (T.-Y. Lin, C.-L. Chuang); Department
of Medical Affair and Planning, Taipei Veterans General
Hospital, No.45, Cheng-Hsin Street, Taipei 112, Taiwan (S.-S.
Wang); Endoscopy Center for Diagnosis and Treatment, Taipei
Veterans General Hospital, No. 201, Sec. 2, Shih-Pai Road, Taipei
112, Taiwan (I.-F. Hsin); Division of Gastroenterology, Depart-
ment of Medicine, Cheng Hsin General Hospital, No.45, Cheng-
Hsin Street, Taipei 112, Taiwan (S.-D. Lee).
Correspondence to: Fa-Yauh Lee and Hui-Chun Huang, MD,
Division of Gastroenterology, Department of Medicine, Taipei
Veterans General Hospital, No. 201, Sec. 2, Shih-Pai Road,
Taipei, 11217, Taiwan. Tel.: +886-2-28712121, ext. 2049; fax:
+886-2-28739318; e-mails: fylee@vghtpe.gov.tw;
hchuang2@vghtpe.gov.tw
Received 22 June 2014; accepted 17 April 2016
References
1 Iwakiri Y, Groszmann RJ. Vascular endothelial dysfunction in
cirrhosis. J Hepatol 2007;46:927–34.
2 Huang HC, Wang SS, Hsin IF, Chang CC, Lee FY, Lin HC, et al.
Cannabinoid receptor 2 agonist ameliorates mesenteric angiogenesis
and portosystemic collaterals in cirrhotic rats. Hepatology
2012;56:248–58.
3 Hsu SJ, Lee FY, Wang SS, Hsin IF, Lin TY, Huang HC, et al. Caffeine
ameliorates hemodynamic derangements and portosystemic
collaterals in cirrhotic rats. Hepatology 2015;61:1672–84.
4 Rockey DC, Weisiger RA. Endothelin induced contractility of stellate
cells from normal and cirrhotic rat liver: implications for regulation
of portal pressure and resistance. Hepatology 1996;24:233–40.
5 Salani D, Taraboletti G, Rosano L, Di Castro V, Borsotti P, Giavazzi
R, et al. Endothelin-1 induces an angiogenic phenotype in cultured
endothelial cells and stimulates neovascularization in vivo. Am J
Pathol 2000;157:1703–11.
6 Chan CC, Wang SS, Lee FY, Chang FY, Lin HC, Chu CJ, et al.
Endothelin-l induces vasoconstriction on portal-systemic collaterals
of portal hypertensive rats. Hepatology 2001;33:816–20.
7 Bradford MM. A rapid and sensitive method for the quantitation of
microgram quantities of protein utilizing the principle of protein-
dye binding. Anal Biochem 1976;72:248–54.
8 Franco D, Gigou M, Szekely AM, Bismuth H. Portal hypertension
after bile duct obstruction. Effect of the bile diversion on portal
pressure in the rat. Arch Surg 1979;114:1064–7.
9 Chojkier M, Groszmann RJ. Measurement of portal-systemic
shunting in the rat by using c-labeled microspheres. A. J Physiol
1981;240:G371–5.
10 Hodeige D, de Pauw M, Eechaute W, Weyne J, Heyndrickx GR. On
the validity of blood flow measurement using colored microspheres.
Am J Physiol 1999;276:H1150–8.
11 Schroll S, Arzt M, Sebah D, Stoelcker B, Luchner A, Budweiser S,
et al. Effects of selective and unselective endothelin-receptor
antagonists on prostacyclin synthase gene expression in
experimental pulmonary hypertension. Scand J Clin Lab Invest
2008;68:270–6.
12 Sumanovski LT, Battegay E, Stumm M, Van Der Kooij M, Sieber CC.
Increased angiogenesis in portal hypertensive rats: role of nitric
oxide. Hepatology 1999;29:1044–9.
European Journal of Clinical Investigation Vol 46 579
ENDOTHELIN AND SHUNTING IN CIRRHOSIS

9. 13 Yanagisawa M, Kurihara H, Kimura S, Tomobe Y, Kobayashi M,
Mitsui Y, et al. A novel potent vasoconstrictor peptide produced by
vascular endothelial cells. Nature 1988;332:411–5.
14 Fernandez M. Molecular pathophysiology of portal hypertension.
Hepatology 2015;61:1406–15.
15 Dreau D, Karaa A, Culberson C, Wyan H, McKillop IH, Clemens
MG. Bosentan inhibits tumor vascularization and bone metastasis in
an immunocompetent skin-fold chamber model of breast carcinoma
cell metastasis. Clin Exp Metastasis 2006;23:41–53.
16 Spinella F, Rosano L, Di Castro V, Nicotra MR, Natali PG, Bagnato
A. Inhibition of cyclooxygenase-1 and -2 expression by targeting the
endothelin a receptor in human ovarian carcinoma cells. Clin Cancer
Res 2004;10:4670–9.
17 Keller S, Karaa A, Paxian M, Clemens MG, Zhang JX. Inhibition of
endothelin-1-mediated up-regulation of iNOS by bosentan
ameliorates endotoxin-induced liver injury in cirrhosis. Shock
2006;25:306–13.
18 Jiang J, Tran L, Vasudevan H, Xia Z, Yuen VG, McNeill JH.
Endothelin-1 blockade prevents COX2 induction and TXA2
production in the fructose hypertensive rat. Can J Physiol Pharmacol
2007;85:422–9.
19 Campbell KA, Wu YP, Sitzmann JV. Modulation of portosystemic
shunting by prostacyclin in portal hypertension. Curr Surg
1990;47:17–9.
20 Spinella F, Rosano L, Di Castro V, Natali PG, Bagnato A. Endothelin-
1 induces vascular endothelial growth factor by increasing hypoxia-
inducible factor-1alpha in ovarian carcinoma cells. J Biol Chem
2002;277:27850–5.
21 Rosano L, Spinella F, Salani D, Di Castro V, Venuti A, Nicotra MR,
et al. Therapeutic targeting of the endothelin a receptor in human
ovarian carcinoma. Cancer Res 2003;63:2447–53.
22 W€ulfing P, Kersting C, Tio J, Fischer RJ, W€ulfing C, Poremba C, et al.
Endothelin-1-, endothelin-A-, and endothelin-B-receptor expression
is correlated with vascular endothelial growth factor expression and
angiogenesis in breast cancer. Clin Cancer Res 2004;10:2393–400.
23 Bagnato A, Spinella F. Emerging role of endothelin-1 in tumor
angiogenesis. Trends Endocrinol Metab 2002;14:44–50.
24 Grimshaw MJ. Endothelins and hypoxia-inducible factor in cancer.
Endocr Relat Cancer 2007;14:233–44.
25 Sogni P, Moreau R, Gomola A, Gadano A, Cailmail S, Calmus Y,
et al. Beneficial hemodynamic effects of bosentan, a mixed ET(A)
and ET(B) receptor antagonist, in portal hypertensive rats.
Hepatology 1998;28:655–9.
26 Rockey DC, Chung JJ. Endothelin antagonism in experimental
hepatic fibrosis. Implications for endothelin in the pathogenesis of
wound healing. J Clin Invest 1996;98:1381–8.
27 Thabut D, Shah V. Intrahepatic angiogenesis and sinusoidal
remodeling in chronic liver disease: new targets for the treatment of
portal hypertension? J Hepatol 2010;53:976–80.
28 Tugues S, Fernandez-Varo G, Munoz-Luque J, Ros J, Arroyo V,
Rodes J, et al. Antiangiogenic treatment with sunitinib ameliorates
inflammatory infiltrate, fibrosis, and portal pressure in cirrhotic rats.
Hepatology 2007;46:1919–26.
29 Yoshiji H, Kuriyama S, Yoshii J, Ikenaka Y, Noguchi R, Hicklin DJ,
et al. Vascular endothelial growth factor and receptor interaction is a
prerequisite for murine hepatic fibrogenesis. Gut 2003;52:1347–54.
30 Patsenker E, Popov Y, Stickel F, Schneider V, Ledermann M,
S€agesser H, et al. Pharmacological inhibition of integrin alphavbeta3
aggravates experimental liver fibrosis and suppresses hepatic
angiogenesis. Hepatology 2009;50:1501–11.
31 Rubin LJ, Badesch DB, Barst RJ. Bosentan therapy for pulmonary
arterial hypertension. N Engl J Med 2003;346:896–903.
32 Galie N, Olschewski H, Oudiz RJ, Torres F, Frost A, Ghofrani HA,
et al. Ambrisentan for the treatment of pulmonary arterial
hypertension: results of the ambrisentan in pulmonary arterial
hypertension, randomized, double-blind, placebo controlled,
multicenter, efficacy (ARIES) study 1 and 2. Circulation
2008;117:3010–9.
33 McGoon MD, Frost AE, Oudiz RJ. Ambrisentan therapy in patients
with pulmonary arterial hypertension who discontinued bosentan
or sitaxsentan due to liver function test abnormalities. Chest
2009;135:122–9.
Supporting Information
Additional Supporting Information may be found online in the
supporting information tab for this article:
Figure S1. Effects of endothelin receptor blockers in cirrhotic
rats. Endothelin receptor blockers reduce the severity of por-
tosystemic in cirrhotic rats by ameliorate hepatic fibrosis and
pathological angiogenesis. The anti-angiogenic effects acted
through the down-regulation of VEGF pathway and relevant
angiogenic factors.
Data S1. Materials and Methods.
580 ª 2016 Stichting European Society for Clinical Investigation Journal Foundation
S.-J. HSU ET AL. www.ejci-online.com