Isoflurane pretreatment lowers portal venous resistance by increasing hepatic heme oxygenase activity in the rat liver in vivo.

Abstract

The heme oxygenase (HO) system contributes to the maintenance of hepatic perfusion and integrity. It was the objective of this study to determine the influence of isoflurane (ISO) on hepatic HO-1 induction and its impact on hepatic hemodynamics.
Rats were pretreated with or without ISO for 5h. After hemodynamic measurements by pressure-, laser doppler-, and ultrasound based techniques, the liver was harvested. HO-1 was analyzed by an HO activity assay, Northern- and Western blotting.
ISO pretreatment induced hepatic HO-1 mRNA and protein resulting in an increase of HO activity. No effect on hsp-27, hsp-70 and hsp-90 mRNA could be observed. ISO lowered portal resistance. HO inhibition by tin protoporphyrine IX increased portal resistance in ISO pretreated animals up to control levels. This was associated with an increase in portal pressure and a reduction of portal flow. Microvascular flux was also impaired after HO blockade and ISO. However, hepatic arterial and systemic hemodynamics remained unchanged, indicating a specific effect within the portal vascular bed.
ISO pretreatment induces hepatic HO-1 mRNA and protein followed by an increase in HO activity, thereby reducing portal resistance. These findings indicate a beneficial effect of ISO on hepatic hemodynamics in vivo.

Full text

Page 1

Isoflurane pretreatment lowers portal venous resistance by increasing
hepatic heme oxygenase activity in the rat liver in vivo
Rene Schmidt1, Alexander Hoetzel1, Tilo Baechle1, Torsten Loop1, Matjaz Humar1,
Michael Bauer2, Heike L. Pahl1, Klaus K. Geiger1, Benedikt H.J. Pannen1,*
1Department of Anesthesiology and Critical Care Medicine, University Hospital Freiburg, Hugstetterstrasse 55, D-79106 Freiburg, Germany
2Department of Anesthesiology and Critical Care Medicine, University of the Saarland, D-66421 Homburg, Germany
Background/Aims: The heme oxygenase (HO) system contributes to the maintenance of hepatic perfusion and
integrity. It was the objective of this study to determine the influence of isoflurane (ISO) on hepatic HO-1 induction and
its impact on hepatic hemodynamics.
Methods: Rats were pretreated with or without ISO for 5 h. After hemodynamic measurements by pressure-, laser
doppler-, and ultrasound based techniques, the liver was harvested. HO-1 was analyzed by an HO activity assay,
Northern- and Western blotting.
Results: ISO pretreatment induced hepatic HO-1 mRNA and protein resulting in an increase of HO activity. No effect
on hsp-27, hsp-70 and hsp-90 mRNA could be observed. ISO lowered portal resistance. HO inhibition by tin
protoporphyrine IX increased portal resistance in ISO pretreated animals up to control levels. This was associated with
an increase in portal pressure and a reduction of portal flow. Microvascular flux was also impaired after HO blockade
and ISO. However, hepatic arterial and systemic hemodynamics remained unchanged, indicating aspecific effect within
the portal vascular bed.
Conclusions: ISO pretreatment induces hepatic HO-1 mRNA and protein followed by an increase in HO activity,
thereby reducing portal resistance. These findings indicate a beneficial effect of ISO on hepatic hemodynamics in vivo.
q 2004 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved.
Keywords: HO-1; Liver blood flow; Portal blood flow; Microcirculation; Isoflurane
1. Introduction
The liver plays a central role in maintaining metabolic
and immunologic homeostasis of the body. These functions
can be detrimentally altered by acute or chronic disturb-
ances of hepatic macro- or microvascular blood flow [1,2].
A consequence of liver injury that merits emphasis is that
loss of liver functions can lead to aberrations in other organ
systems and to death [3]. Accumulating evidence suggests
that heat shock proteins may protect the liver against
stressors such as hypoxia, ischemia or oxidative stress [4].
HSP-32, better known as heme oxygenase-1 (HO-1), is an
important member of this class. It catalyzes the conversion
of heme to biliverdin IXa, to free iron, and to carbon
monoxide (CO) [5]. To date, three members of the HO
family have been identified [6]. HO-1 represents the
inducible isoform of the enzyme system and is upregulated
in response to many different clinically relevant pathologi-
cal stimuli, including endotoxemia, hemorrhagic shock, or
ischemia/reperfusion [7,8]. HO-2 and HO-3 are constitu-
tively expressed isoforms. Endogenously generated
HO-derived CO reduces sinusoidal tone and is responsible
for maintaining perfusion in the normal liver [9,10]. Hepatic
upregulation of the HO system by administration of
the pharmacologic HO-1 inducer hemin reduces hepatic
vascular resistance and initiates vascular hyporeactivity to
Journal of Hepatology 41 (2004) 706–713
www.elsevier.com/locate/jhep
0168-8278/$30.00 q 2004 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved.
doi:10.1016/j.jhep.2004.07.004
Received 6 February 2004; received in revised form 2 June 2004; accepted
2 July 2004; available online 21 July 2004
* Corresponding author. Tel.: C49-761-2702306; fax: C49-761-
2702396.
E-mail address: pannen@ana1.ukl.uni-freiburg.de (B.H.J. Pannen).

Page 2

vasoconstrictor stimuli by an overproduction of CO [11].
Moreover, CO is an important mediator in the protection of
hepatocellular integrity and maintenance of hepatic micro-
circulation under pathological conditions [12,13]. It is now
widely accepted that upregulation of the HO system is one
of the most important cytoprotective mechanism initiated in
response to cellular stress [14].
Experiments related to the protective effect of HO-1
enzyme induction are hampered by the fact that most
inducers of HO-1 by themselves perturb organ function.
Hence, the substances used for pharmacologic induction of
HO-1 in many promising experiments published over the
last years are not available for use in patients because of
their undesirable or unknown side effects.
We have previously shown that isoflurane (ISO) leads to
an expression of HO-1 in the rat liver in vivo [15]. ISO
belongs to the halogenated volatile anesthetics and is used in
daily clinical practice for induction and maintenance of
general anesthesia. Therefore, we designed the present
study to evaluate the CO-mediated vasorelaxing properties
in innervated livers in vivo by inducing HO-1 with the non-
toxic and clinically available substance ISO. It was thus the
aim of this study (i) to characterize the effects of ISO on
hepatic HO-1 mRNA levels, HO-1 protein expression and
HO enzyme activity; (ii) to determine the specificity of ISO
on HO-1 induction in comparison to other ‘vasoactive’
(HSP-27, HSP-90) and ‘non-vasoactive’ (HSP-70) heat
shock proteins and finally (iii) to characterize whether an
ISO induced HO-1 upregulation exerts any hemodynami-
cally relevant effects on the macro- and microvascular blood
flow of the intact rat liver in vivo.
2. Material and methods
2.1. Reagents
Isoflurane was purchased from Abbott (Wiesbaden, Germany),
pentobarbital (PEN) from Alvetra (Neumuenster, Germany), pancuronium
from Organon(BH Oss, Netherlands)and tin protoporphyrineIX (SnPPIX)
was obtained from Frontier Scientific (Carnforth, UK). All other reagents
used were purchased from Sigma-Aldrich (Deisenhofen, Germany) if not
specified otherwise.
2.2. Animals
Male Sprague-Dawley rats (Charles River, Sulzfeld, Germany),
weighing 342G46 g, were used for all experiments. The experimental
protocol was approved by the local animal care and use committee, and all
animals received human care according to the criteria outlined in the Guide
for the Care and Use of Laboratory Animals [16].
2.3. Experimental protocol
After inhalational induction of anesthesia a tail vein was cannulated and
the animals were randomized into four groups with nZ6 animals per group
(Fig. 1): group 1: PEN (40 mg/kg/h i.v.)Cvehicle (0.5 ml NaHCO38.4%,
i.v.); group 2: ISO (3.5 vol%)Cvehicle; group 3: PENCSnPP IX (HO-
inhibitor, 50 mmol/kg i.v.); group 4: ISOCSnPP IX. A tracheotomy was
performed and after relaxation (pancuronium 1 mg/kg i.v.) all animals were
mechanically ventilated (Rodent Ventilator UB 7025-10, Harvard Appar-
atus, March-Hugstetten, Germany). For compensation of evaporative losses
4 ml/kg/h Jonosteril (Fresenius, Bad Homburg, Germany) were continu-
ously infused. Cannulation of the left femoral artery with PE-50 tubing was
performed for blood pressure monitoring. The body temperature was
maintained in a normothermic range. For time course experiments (nZ2
animals per group), rats were exposed to ISO for 0.5, 1, 2, 3, 4 and 5 h
before organ harvest.
2.4. Animal preparation for measurement of systemic
and regional hemodynamics
Five hours after onset of anesthesia the ISO administration in groups 2
and 4 was stopped and all animals received PEN to avoid any direct
hemodynamic actions of ISO at the time of data acquisition. For
measurement of cardiac output a thermistor tip catheter (9490E, Columbus
Instruments, Columbus, OH) was inserted into the aortic arch through the
left carotid artery. To registrate central venous pressure and for injection of
saline at 4 8C to obtain estimates of cardiac output, a PE-50 catheter was
positioned into the right atrium through the right jugular vein. Ultrasound
flow probes were placed around the common hepatic artery and the portal
vein (T206, small animal flowmeter, Transonic, Ithaca, NY) which was
additionally cannulated for pressure measurement. After a stabilization
period, 6 h after onset, baseline measurements were obtained and either
vehicle or SnPP IX was applied. After 5 and 10 min, all measurements were
repeated, and the liver was harvested for molecular biological analyses.
2.5. Determination of microvascular blood flow
Hepatic microvascular blood flow (flux) was determined by the laser
doppler technique (Moor DRT-4, Lawrenz, Bad Soden, Germany). Data
was acquired at baseline (6 h after onset) and 5 and 10 min after
Fig. 1. Experimental protocol. After inhalational induction of
anesthesia the animals were randomized into four groups. Groups 1
and 3 received pentobarbital sodium (PEN; 40 mg/kg/h) during the
entire experiment. Groups 2 and 4 were pretreated with isoflurane
(ISO; 3.5 vol%) for 5 h followed by 1 h of PEN anesthesia to avoid
direct hemodynamic effects of ISO at the times of data acquisition. Six
hours after onset, baseline hemodynamic measurements were obtained
and vehicle (0.5 ml sodium bicarbonate 8.4%; groups 1 and 2) or SnPP
IX (50 mmol/kg; groups 3 and 4) was applied. Five and ten minutes
after the intervention additional data was collected and the livers were
subsequently harvested for molecular analyses.
R. Schmidt et al. / Journal of Hepatology 41 (2004) 706–713 707

Page 3

pharmacologic intervention. In each animal, the laser doppler probe was
placed at the same defined position on the surface of the right liver lobe.
Themean flux-value at therespectivetimes wascalculatedandexpressedas
a percentage of the baseline value.
2.6. Northern blot analysis
Total RNA was extracted from frozen liver tissue as previously
described [8]. Ten micrograms of RNA were loaded per lane, size
fractioned, and transferred to a nylon membrane. Hybridization was
performed with P32g-deoxycytosine triphosphate-radiolabelled HO-1,
hsp-27, hsp-70, and hsp-90 cDNA. All blots were reprobed with 18S
ribosomal RNA cDNA to verify equal loading. Autoradiographs were
analyzed by laser scanning densitometry (Personal Densitometer; Molecu-
lar Dynamics, Krefeld, Germany).
2.7. Western blot analysis
Protein was extracted from frozen liver tissue as previously described
[8]. Each lane of a 12% SDS gel contained 100 mg of total protein. After
protein separation and electroblotting, HO-1 was detected by a rabbit
polyclonal anti-HO-1 antibody (1:1000 dilution, SPA-895; Stress Gen,
Victoria, Canada), using the ECL detection kit (Amersham, Freiburg,
Germany) according to the manufacturer’s instructions.
2.8. Determination of heme oxygenase enzyme activity
HO activity assay was performed as previously described [8]. Briefly,
liver tissue was homogenized and added to a reaction mixture containing
NADPH, liver cytosol, glucose-6-phosphate, glucose-6-phosphate dehy-
drogenase and hemin. The reaction was performed at 37 8C for 1 h in the
dark and stopped by addition of chloroform. The extracted bilirubin was
calculated by the difference in absorbance between 464 and 530 nm.
2.9. Data analysis
Data is presented as meanGSEM with nZ6 animals per group, except
time course experiments, where nZ2 animals per group were used.
Statistical differences within each group were determined using a one-way
ANOVA for repeated measurements and between the different groups by
one-way ANOVA followed by the post-hoc Student-Newman-Keuls test
for pairwise comparisons. When criteria for parametric tests were not met,
Kruskal-Wallis ANOVA on ranks followed by Dunn’s test were used. Data
was considered significant when P!0.05.
3. Results
3.1. Effects of isoflurane on hepatic HO-1 mRNA expression
As shown in Fig. 2, HO-1 transcripts were barely
detectable in animals anesthetized with PEN (lanes 1C2).
In sharp contrast, HO-1 mRNA levels were significantly
higher after 5 h of ISO pretreatment (lanes 3C4).
Administration of SnPP IX 10 min before organ harvest
had no influence on HO-1 mRNA levels (lanes 5C6:
PENCSnPP IX; lanes 7C8: ISOCSnPP IX) compared
with the vehicle groups (lanes 1–4). Equal loading was
verified by 18S rRNA labeling (Fig. 2B). Quantitation of
relative HO-1 mRNA levels by densitometric analysis is
shown in Fig. 2C (P!0.05).
3.2. Effects of ISO on hepatic HO-1 protein levels and HO
enzyme activity
Western blot analysis showed an accumulation of HO-1
protein after ISO pretreatment (Fig. 3A, lanes 3C4)
compared with PEN anesthetized animals (lanes 1C2).
Injection of SnPP IX had no influence on HO-1 protein
expression levels (lanes 5C6: PENCSnPP IX; lanes 7C8:
ISOCSnPP IX). ISO pretreatment significantly increased
hepatic HO enzyme activity compared to the PEN
anesthetized control animals (Fig. 3B; P!0.05). Adminis-
tration of SnPP IX almost completely inactivated the HO
enzyme in both groups.
3.3. Time course of HO-1 gene expression after ISO
treatment
Northern- (A) and Western blot (C) analyses of liver
samples from animals exposed to ISO for 0.5 (lanes 1C2),
Fig. 2. Northern blot analysis of hepatic HO-1 messenger RNA (A) and
18S ribosomal RNA (B) expression in two representative animals
subjected to 6 h of pentobarbital anesthesia (PEN; lanes 1C2) or 5 h of
isoflurane pretreatment (ISO; lanes 3C4) followed by 1 h of PEN
anesthesia. Each group received either vehicle (lanes 1–4) or SnPP IX
(50 mmol/kg; lanes 5–8) 10 min before organ harvesting to characterize
the HO dependent effects. A profound induction of HO-1 transcripts
was observed in the isoflurane pretreated groups ((A), lanes 3C4:
ISOCvehicle; lanes 7C8: ISOCSnPP IX). (C) Quantification of
hepatic HO-1 mRNA expression by densitometric analysis for the
different groups. Relative densitometric units were calculated as
dividends of the background-corrected densitometric values of HO-
1/18S rRNA. Data is presented as meanGSEM relative densitometric
units for nZ6 animals per group. *P!0.05 versus PENCvehicle;
#P!0.05 versus PENCSnPP IX.
R. Schmidt et al. / Journal of Hepatology 41 (2004) 706–713708

Page 4

1 (lanes 3C4), 2 (lanes 5C6), 3 (lanes 7C8) 4 (lanes 9C
10) and 5 h (lanes 11C12) were shown in Fig. 4.
Upregulation of HO-1 mRNA and protein could be detected
after 4 h of ISO treatment and further increased after 5 h of
anesthesia. Equal loading of RNA samples was verified by
18S rRNA labeling (Fig. 4B).
3.4. Effects of ISO on hepatic hsp-27, hsp-70 and hsp-90
mRNA expression
Representative Northern blots labeled with hsp-27,
hsp-70 and hsp-90 cDNA are shown in Fig. 5. Neither
PEN (lanes 3C4) nor ISO treatment (lanes 5C6) showed
any upregulation of hepatic hsp-27 mRNA (Fig. 5A), hsp-70
mRNA (Fig. 5B), and hsp-90 mRNA (Fig. 5C). Adminis-
tration of SnPP IX had no influence on hepatic hsp-27,
hsp-70 or hsp-90 mRNA expression (lanes 7C8:
PENCSnPP IX; lanes 9C10: ISOCSnPP IX). The animal
which served as a positive control (lane 2) was subjected to
1 h of hemorrhagic shock and 5 h of resuscitation. The
negative control (lane 1) was obtained from an animal that
was sacrificed without any treatment. Equal loading was
verified by 18S rRNA labeling.
Fig. 3. (A) Western blot analysis of hepatic HO-1 protein in isoflurane
pretreated animals and pentobarbital anesthetized controls. Isoflurane
pretreatment led to a strong upregulation of hepatic HO-1 protein
(lanes 3C4: ISOCvehicle; lanes 7C8: ISOCSnPP IX). In contrast,
HO-1 protein in controls was only barely detectable (lanes 1C2:
PENCvehicle; lanes 5C6: PENCSnPP IX). Proteins were isolated
from the two representative rat livers shown in Fig. 2. (B)
Measurements of HO enzyme activity in rat liver tissue presented as
median (box: 25th and 75th percentiles; error bars: 10th and 90th
percentiles) for nZ6 animals per group. *P!0.05 versus PENC
vehicle; #P!0.05 versus ISOCvehicle.
Fig. 4. Time course of HO-1 mRNA (A) and HO-1 protein (C)
expression was assessed in liver samples after 0.5 h (lanes 1C2), 1 h
(lanes 3C4), 2 h (lanes 5C6), 3 h (lanes 7C8), 4 h (lanes 9C10) and 5 h
(lanes 11C12) of isoflurane anesthesia. Equal loading of RNA samples
was verified by 18S rRNA labeling (B). Isoflurane led to an induction of
HO-1 mRNA and protein starting after 4 h. A further increase of
expression could be observed after 5 h of isoflurane anesthesia.
Fig. 5. Expression of hepatic heat shock protein (hsp)-27 mRNA (A),
hsp-70 mRNA (B), or hsp-90 mRNA (C) of two representative animals
subjected to pentobarbital (PEN; lanes 3C4: PENCvehicle; lanes
7C8: PENCSnPP IX) or isoflurane (ISO; lanes 5C6: ISOCvehicle;
lanes 9C10: ISOCSnPP IX) anesthesia. The animal which served as a
positive control (lane 2) was subjected to 1 h of hemorrhagic shock and
5 h of resuscitation. The negative control (lane 1) is represented by an
animal that was sacrificed without any treatment. Equal loading was
verified by 18S rRNA labeling. No differences in hsp-27, hsp-70, or
hsp-90 mRNA expression could be seen between the groups.
R. Schmidt et al. / Journal of Hepatology 41 (2004) 706–713 709

Page 5

3.5. Effects of ISO-induced HO-1 expression on hepatic
hemodynamics at baseline
Hepatic hemodynamic parameters at baseline (6 h after
onset) are presented in Table 1. Five hours of ISO
pretreatment slightly decreased portal pressure but had no
influence on hepatic arterial and portal flow compared with
the PEN treated controls. Likewise, hepatic arterial
resistance showed no differences between the ISO and
PEN groups. In sharp contrast, portal resistance was
significantly lower in ISO pretreated animals (Table 1;
P!0.01). At the times of baseline measurements no
significant differences in systemic hemodynamics could be
observed between the groups (Table 2).
3.6. Effects of SnPP IX on hepatic macrohemodynamics
Administration of vehicle to PEN or ISO anesthetized
animals did not cause any significant changes in local
hepatic or systemic hemodynamics (Table 2, Figs. 6 and 7).
In contrast, inhibition of the HO enzyme by SnPP IX in
ISO pretreated rats led to a profound increase of portal
pressure in coincidence with a reduction of portal flow
(Fig. 6A and B; P!0.05). These effects caused a massive
increase in portal resistance after acute HO inhibition in the
ISO pretreated group (Fig. 7A; P!0.05). However, hepatic
arterial flow and hepatic arterial vascular resistance did not
change (Figs. 6C and 7B). Application of SnPP IX
increased portal pressure and reduced portal flow in
PEN treated animals to a much lesser extent (Fig. 6A
and B; P!0.05).
3.7. Effect of SnPP IX on hepatic microhemodynamics
As shown in Fig. 8, the application of SnPP IX led to a
significant reduction of hepatic microvascular flux in ISO
pretreated rats at 5 and 10 min after pharmacological
intervention compared with the respective baseline levels
(P!0.05). No changes in flux were seen in the vehicle and
PEN groups after HO blockade.
4. Discussion
It was the aim of the present study to characterize the
functional significance of upregulation of the HO system by
ISO in the intact rat liver. Our results demonstrate that
pretreatment with ISO, a clinically used volatile anesthetic,
significantly lowered the resistance in the portal venous
vascular bed through a profound increase in hepatic HO
enzyme activity. Blockade of the enzyme in ISO pretreated
rats led to a massive increase in portal venous pressure
Table 1
Effect of isoflurane pretreatment on hepatic macro- and micro-hemodynamics in rats at baseline before pharmacological intervention
Experimental groupsPENCvehicleISOCvehicle PENCSnPP IXISOCSnPP IX
Qha (ml/min)
Qpv (ml/min)
Ppv (mmHg)
Microvascular flux (AU)
HAVR (mmHg min mlK1)
PVR (mmHg min mlK1)
2.9G0.4
19.9G1.1
7.7G0.7
251.8G26,0
24.62G1.95
0.27G0.04
2.8G0.4
20.6G1.5
6.5G0.4
220.8G20,0
24.35G2.92
0.16G0.01*,#
3.4G0.3
20.8G1.2
7.7G0.4
309.0G22.1
20.46G1.57
0.28G0.02
3.0G0.3
21.0G1.8
6.2G0.5
239.0G13.7
20.27G2.48
0.14G0.01*,#
Measurements were obtained 6 h after onset of anesthesia. Five hours of pretreatment with isoflurane (ISO) led to a significant decrease of portal vascular
resistance (PVR) at baseline compared to the pentobarbital (PEN) control groups (P!0.01). No significant differences were seen in hepatic arterial flow (Qha),
portal venous flow (Qpv), portal pressure (Ppv), microvascular flow (Flux), or hepatic arterial vascular resistance (HAVR). Data is presented as meanGSEM
(nZ6 animals per group). *P!0.01 versus PENCvehicle; #P!0.01 versus PENCSnPP IX.
Table 2
Systemic hemodynamic parameters at baseline and 5 or 10 min after
pharmacological intervention
Baseline5 min 10 min
MAP (mmHg)
PENCvehicle
ISOCvehicle
PENCSnPP IX
ISOCSnPP IX
Heart rate (bpm)
PENCvehicle
ISOCvehicle
PENCSnPP IX
ISOCSnPP IX
Cardiac output (ml/min)
PENCvehicle
ISOCvehicle
PENCSnPP IX
ISOCSnPP IX
SVR (mmHg min mlK1)
PENCvehicle
ISOCvehicle
PENCSnPP IX
ISOCSnPP IX
70G4.8
65G2.0
70G4.1
63G2.6
70G4.2
63G1.5
69G4.4
65G4.8
69G4.6
65G1.5
63G4.8
63G3.8
370G14.3
355G14.3
357G12.0
340G11.4
370G13.2
358G14.2
360G11.6
345G10.2
375G17.3
358G11.2
369G10.2
348G7.2
107G8.5
115G6.9
99G6.1
108G3.5
106G7.5
112G6.7
98G5.9
105G3.5
105G9.8
110G9.6
92G8.7
97G4.3
0.6527G0.08
0.5520G0.05
0.7007G0.04
0.5539G0.02
0.6528G0.07
0.5500G0.04
0.6804G0.06
0.5822G0.04
0.6676G0.08
0.5881G0.07
0.6811G0.06
0.6131G0.03
Systemic hemodynamic data was obtained 6 h after onset (baseline)
followed by an intravenous injection of either vehicle or SnPP IX (50 mmol/
kg) to block HO activity. Two additional measurements were performed 5
and 10 min after the intervention. No significant differences in MAP, heart
rate, cardiac output, or systemic vascular resistance (SVR) were observed
between the isoflurane (ISO) pretreated and pentobarbital (PEN)
anesthetized control groups at baseline. Inhibition of the HO pathway as
well as the application of vehicle did also not influence any of the systemic
hemodynamic parameters at the times measured. Values are presented as
meanGSEM (nZ6 animals per group).
R. Schmidt et al. / Journal of Hepatology 41 (2004) 706–713710