Department of Pharmacology (Z.Z., G.E.A., H.D.C., P.S.,
R.G.T.), Curriculum in Toxicology (G.E.A., J.A.R., R.G.T.), Department
of Radiation Oncology (S.-C.C., J.A.R.), and Department of Cell Biology
and Anatomy (J.J.L.), University of North Carolina at Chapel Hill,
Chapel Hill; and Laboratory of Pharmacology and Chemistry, National
Institute of Environmental Health Sciences (R.P.M.), Research Triangle
Park, North Carolina
Disturbances in hepatic microcirculation increase graft injury and
failure; therefore, this study evaluates the effects of ethanol on
microcirculation after liver transplantation. Donor rats were given one
dose of ethanol (5 g/kg) by gavage 20 h before explantation, and
grafts were stored in University of Wisconsin solution for
24 h before implantation. Acute ethanol treatment decreased 7-day
survival of grafts from about 90 to 30%, increased transaminase
release nearly 4-fold, and decreased bile production by 60%. Moreover,
portal pressure increased significantly and liver surface oxygen
tension decreased about 50%, indicating that ethanol disturbs hepatic
microcirculation. Pimonidazole, a 2-nitroimidazole hypoxia marker, was
given i.v. to recipients 30 min after implantation, and grafts were
harvested 1 h later. Ethanol increased hepatic pimonidazole
binding about 3-fold, indicating that ethanol led to hypoxia in fatty
grafts. Ethanol also significantly increased free radicals in bile.
Catechin (30 mg/kg i.v. upon reperfusion), a free radical scavenger,
and Carolina Rinse solution, which contains several agents that inhibit
free radical formation, minimized disturbances in microcirculation and
prevented pimonidazole adduct formation significantly. These treatments
also blunted increases in transaminase release and improved survival of
fatty grafts. Destruction of Kupffer cells with GdCl3 (20 mg/kg i.v. 24 h before explantation) or inhibition of formation of
leukotrienes with MK-886 (50 µM in University of Wisconsin or rinse
solution) also minimized hypoxia and improved survival after
transplantation. Taken together, these results demonstrate that ethanol
disturbs hepatic microcirculation, leading to graft hypoxia after
transplantation, most likely by activating Kupffer cells and increasing
free radical production.
 |
Introduction |
Organ
donors are often victims of accidents involving either chronic or acute
consumption of ethanol. For example, in 1993, more than 40,000 fatalities due to traffic accidents occurred in the United States;
approximately 44% of them were alcohol-associated (Anonymous, 1994
).
Moreover, in a study at the University of Pittsburgh in 1991, 48% of
donors had a blood ethanol level between 4 and 40 mg/dl (Hassanein et
al., 1991). Therefore, ethanol consumption is common in organ donors.
It is well known that ethanol exposure causes fatty infiltration,
inflammation, degeneration, and necrosis in the liver, and hepatic
lipids are elevated after exposure to even a single, inebriating dose
of ethanol (Ylikahri et al., 1972). Previous studies have shown that
both acute and chronic ethanol treatments significantly increase
primary nonfunction after liver transplantation in rats (Gao et al.,
1995; Zhong et al., 1996). Moreover, grafts from victims with
elevated blood alcohol levels have high rates of primary graft failure
(Hassanein et al., 1991). Therefore, the mechanism(s) of
alcohol-induced fatty graft failure must be understood if the pool of
usable donor organs is to be expanded.
Previous studies have shown that leukocyte margination and platelet
adhesion are increased and blood flow is reduced after liver
transplantation (Takei et al., 1996). Moreover, disturbances in
hepatic microcirculation increase graft injury and failure (Marzi et
al., 1990). A recent study has shown that Kupffer cells (KCs), the
major source of vasoactive and chemotactic mediators such as
thromboxanes, leukotrienes, platelet-activating factor, and tumor
necrosis factor-
in the liver, are activated by acute ethanol
treatment (Enomoto et al., 1998). Moreover, ethanol increases oxidative
stress, which activates nuclear factor-
B and phospholipase A2, thus increasing production of vasoactive and
chemotactic cytokines, platelet-activating factor, and eicosanoids.
Therefore, it is possible that ethanol disturbs hepatic
microcirculation, leading to hypoxia and graft injury.
Organ donation is often associated with accidents that frequently
involve binge drinking. Acute ethanol treatment mimics binge drinking
and provides an inexpensive and convenient model. Accordingly, the
purpose of this study was to evaluate the effects of ethanol exposure
on microcirculation and hypoxia after liver transplantation following
acute treatment with one large dose of ethanol to mimic binge drinking.
| |
Materials and Methods |
Animals and Liver Transplantation.
Inbred female Lewis rats
(200-230 g) were used in liver transplantation experiments to prevent
immunological interference. Donor rats were treated with ethanol (5 g/kg body weight, dissolved in normal saline at a concentration of
20%) or an equal volume of saline vehicle by gavage 20 h before
explantation. Previous studies showed that this dose of ethanol
dramatically increased hepatic triglyceride content (Ylikahri et al.,
1972) and decreased graft survival after transplantation (Zhong et al.,
1996). This dose of ethanol resulted in a peak blood ethanol
concentration of about 370 mg/dl in about 2 h, which declined to
undetectable levels in 8 to10 h (Wendell and Thurman, 1979). No
respiratory suppression was observed in animals receiving this
treatment; therefore, it provides a convenient model mimicking fatty
grafts caused by binge drinking. Gadolinium chloride
(GdCl3; 20 mg/kg) dissolved in acidic saline (pH
3.0-3.5) was given to some donors i.v. 24 h before surgery to
destroy KCs (Hardonk et al., 1992). Rats were anesthetized with
Metofane (Schering-Plough Animal Health, Union, NJ) by
inhalation in a breathing cone; animals usually woke up within 30 min
after surgery. Liver transplantation was performed using a technique
described elsewhere (Zimmermann et al., 1979). Briefly, heparin (200 IU) in 0.5 ml of lactated Ringer's solution was injected into the
subhepatic vena cava, and a 4-mm-long stent prepared from polyethylene
tubing (PE 50) was inserted into the common bile duct and secured with
a 6-0 suture. Livers were flushed in situ with 5 ml of lactated
Ringer's solution (0-4°C) followed by 3 ml of University of
Wisconsin (UW) cold storage solution. In some experiments, MK-886, an
inhibitor of leukotriene biosynthesis (Gillard et al., 1989), was
dissolved in UW solution at a final concentration of 50 µM. Venous
cuffs prepared from 14-gauge i.v. catheters were placed in the
subhepatic vena cava and the portal vein, and grafts were stored in UW
solution at 0-4°C for 24 h. Just before implantation, grafts
were rinsed with 5 ml of lactated Ringer's solution (0-4°C) or
Carolina Rinse solution, which contains several compounds that inhibit
free radical formation. These components include desferrioxamine (1.0 mM), which chelates iron, thus inhibiting the Fenton reaction;
allopurinol (1.0 mM), which inhibits superoxide formation by xanthine
oxidase and traps radicals; and glutathione (3.0 mM), which inactivates
radicals (Gao et al., 1991). In some experiments, MK-886 (50 µM) or
catechin (400 µM), a flavonoid that scavenges free radicals and
singlet oxygen (Slater, 1981), was added to lactated Ringer's
solution. For implantation, the liver of the recipient was removed
after clamping the suprahepatic vena cava, portal vein, and subhepatic vena cava, and implanted by connecting the suprahepatic vena cava with
a running suture, then inserting cuffs into the appropriate vessels and
securing them with a 6-0 silk suture. The bile duct was anastomosed
with an intraluminal stent. Implantation surgery required less than 50 min; during this time the portal vein was clamped for 18 to 20 min.
Catechin was injected (30 mg/kg i.v.) into some recipients upon opening
of vascular clamps. Survival was monitored for 7 days after surgery.
All animals in this study received humane care in compliance with
institutional guidelines as outlined in the Guide for the Care and Use
of Laboratory Animals published by the National Institutes of Health.
All experiments were performed in accordance with protocols approved by
the Animal Use Committee of the University of North Carolina.
Measurement of Serum Aspartate Aminotransferase (AST) and
Bile Production.
Blood samples were collected from the inferior
vena cava for the first 3 h after implantation to assess AST
release. Serum was obtained by centrifugation and stored at 
20°C,
and AST activity was determined using a commercially available
analytical kit from Sigma Chemical Co. (St. Louis, MO). The
common bile duct was cannulated with polyethylene tubing (PE 50, Clay
Adams Brand, Becton-Dickinson, Sparks, MD) in some experiments,
and bile was collected for 3 h after organ implantation. Rates of
bile production were calculated from time of bile collection, the
volume of bile, and the liver wet weight.
Assessment of Arterial Blood Pressure, Microcirculation, and
Hypoxia.
After implantation of the liver, the right carotid artery
was cannulated with polyethylene tubing (PE 50), and blood pressure was
measured using a low pressure analyzer (LPA-200, Digi-Med, Louisville, KY).
To assess hepatic microcirculation after transplantation, a needle
connected to a polyethylene tube (PE 90) was inserted into the portal
vein. Portal pressure was measured by changes in the height of a water
column 3 h after implantation. Oxygen tension on the liver
surface, another indicator of hepatic microcirculation (Adachi et al.,
1995), was measured with a Teflon-shielded, Clark-type oxygen
electrode, which was placed gently on the liver surface with the aid of
a micromanipulator at the times indicated in the figures. Oxygen
tension was measured at three to five points on the liver surface at
several times, and averages were calculated. Metofane was always
removed at least 15 min before measuring surface oxygen tension (i.e.,
the rat was breathing air when surface oxygen tension was measured).
Liver surface oxygen tension measured this way was not significantly
different from the oxygen tension found in animals that were
anesthetized with pentobarbital (50 mg/kg i.p.) but were not exposed to
inhaled anesthesia. Therefore, measurement of liver surface oxygen
tension was not influenced by anesthesia.
Pimonidazole, a 2-nitroimidazole compound, is reductively activated at
low oxygen concentrations and binds to cell molecules that possess free
thiol groups (Raleigh and Koch, 1990; Arteel et al., 1995).
Pimonidazole adducts accumulate in vivo in intact, awake animals and
measure tissue hypoxia directly at the cellular level (Arteel et al.,
1995; Durand and Raleigh, 1998). To evaluate hypoxia in grafts of awake
animals, pimonidazole hydrochloride was dissolved in normal saline at a
concentration of 120 mg/ml and injected into the tail vein (120 mg/kg)
30 min after implantation. Grafts were perfused 1 h later to
remove blood, and pimonidazole binding was determined in liver
homogenates using a competitive enzyme-linked immunosorbent assay
(ELISA) procedure as described elsewhere (Arteel et al., 1995). Protein
levels in tissue homogenates were measured with the bicinchoninic acid
assay using a commercially available kit.
Protein-bound pimonidazole in liver sections was also determined
immunohistochemically. Paraffin blocks of formalin-fixed liver tissue
were sectioned at 6 µm and pimonidazole adducts were detected with a
biotin-streptavidin-peroxidase indirect immunostaining method (Arteel
et al., 1995). Sections were hydrated and treated briefly with 0.01%
protease (pronase E) and exposed to mouse monoclonal anti-pimonidazole
IgG1 antibody (3.44.6.7) in PBS-Tween for 2 h at 37°C. Rat-adsorbed horse anti-mouse antibody was then applied to
the sections for 30 min. Once the antibody-biotin-peroxidase complex
was formed, 3,3'-diaminobenzidine chromogen was added as the peroxidase
substrate. After the immunostaining procedure was completed, a
counterstain of hematoxylin was applied, followed by mounting with
crystal mount solution. Some liver sections were also stained with
osmium and examined microscopically for fatty infiltration (Luna,
1968).
Detection of Free Radical Adducts.
To assess free radical
formation by liver grafts, the spin trapping reagent -(4-pyridyl
1-oxide)-N-tert-butylnitrone (4-POBN, 1 g/kg body weight)
was dissolved in 0.5 ml of normal saline and injected slowly into the
tail vein after opening the vascular clamps. A cannula (PE 50) was
placed in the common bile duct, and bile was collected for 1 h
after implantation into 50 µl of 30 mM dipyridyl on ice to prevent ex
vivo free radical formation. Samples were stored on dry ice until
analysis. Bile samples were thawed, placed in a quartz electron spin
resonance (ESR) cell, and scanned repeatedly until the interfering
ascorbate semiquinone signal disappeared in approximately 1 h.
Free radical adducts were detected with a Bruker ESP 106 ESR
spectrometer (Brucker, Billerica, MA). Instrument conditions were as
follows: 20-mW microwave power, 1.0-G modulation amplitude, and 80-G
scan range. Spectral data were stored on an IBM-compatible computer and
were analyzed for ESR hyperfine coupling constants by computer
simulation (Duling, 1994).
Assay for Hepatic Triglycerides.
To assess triglyceride
content in the grafts, liver tissue was homogenized in an equal volume
of normal saline and extracted with a mixture of chloroform and
methanol (2:1) as described elsewhere (Brodie et al., 1961). Zeolite
(Sigma) was added to remove phospholipids in the extract. The resulting
extract was dried under nitrogen and dissolved in 2 ml of Plasmanate
(Bayer, West Haven, CT), and triglycerides were determined
enzymatically (Bucolo and David, 1973). Triglyceride content was
standardized with DNA content in the liver tissue to rule out any
possible influences of weight changes due to fat accumulation. DNA
content in the liver tissue was measured using the bisbenzimidazole
method as described elsewhere (Labarca and Paigen, 1980).
Statistical Analysis.
All groups were compared by using the
2 or ANOVA plus Student-Newman-Keuls posthoc tests as
appropriate, and differences were considered significant at the
p < .05 level.
| |
Results |
Survival after Liver Transplantation.
After implantation of
grafts, survival rates were 86% in control rats but only 29% in the
alcohol-treated group. Catechin, a free radical scavenger, and Carolina
Rinse solution restored rates to 60 to 63%. Destruction of KCs with
GdCl3 elevated survival rates significantly to
70% (Fig. 1). Leukotrienes, which are
synthesized from arachidonic acid via 5-lipoxygenase, are potent
chemoattractants (LTB4) and smooth muscle
constrictors (LTC4, LTD4,
and LTE4) (Anggard, 1985). MK-886, which inhibits
activation of 5-lipoxygenase, thus blocking leukotriene synthesis
(Gillard et al., 1989), also improved survival of grafts exposed to
ethanol to 67%. (Fig. 1).
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 1.
Survival after liver transplantation. Rats were given
ethanol (5 g/kg i.g.) 20 h before explantation, and
GdCl3 (20 mg/kg i.v.) was given to some donors 4 h
before ethanol treatment. Livers were explanted and stored in UW
solution (0-4°C) for 24 h and rinsed with lactated Ringer's or
Carolina Rinse solution before implantation. Liver transplantation was
performed with the technique described in Materials and
Methods. Catechin (30 mg/kg i.v.) was injected into recipients
upon opening vascular clamps, and grafts were rinsed with 5 ml of
Ringer's solution containing 400 µM catechin just before
implantation. In some experiments, grafts were stored in UW cold
storage solution containing 50 µM MK-886 and rinsed with lactated
Ringer's solution also containing 50 µM MK-886 before implantation.
Survival was monitored for 7 days postoperatively. Values are
percentage of survival (% survival), where fractions = survivals/total. (2 analysis, n = 8-17 in each group: aP < .05 compared
with the control group; bp < .05 compared with the ethanol group).
|
|
AST Release and Bile Production.
Serum AST levels were 34 U/l
in untreated controls without transplantation. One single dose of
ethanol (5 g/kg) did not significantly change this value. However, AST
levels increased gradually to about 650 U/l over 3 h after
implantation in controls. In contrast, recipients receiving grafts from
ethanol-treated rats under similar conditions had AST levels of 2432 U/l. Treatment with the radical scavenger catechin or destruction of
KCs with GdCl3 blunted the elevation of serum
enzymes (i.e., values only rose about 40% as high; ethanol + catechin
was not significantly different from ethanol + GdCl3) and minimized injury to ethanol-exposed
grafts immediately after implantation (Fig.
2, top).
View larger version (13K):
[in this window]
[in a new window]
|
Fig. 2.
Aspartate aminotransferase release and bile
production after liver transplantation. Conditions as in Fig. 1. Blood
samples were collected postoperatively from the inferior vena cava, and
bile was collected via a cannula placed in the common bile duct for the
first 3 h after implantation. AST activity was determined by using
a commercially available analytic kit from Sigma. Bile production was
normalized by using liver wet weight. Top, AST release; bottom, bile
production. Values are means ± S.E.M. (n = 4-10 in each group; p < .01 by ANOVA).
ap < .05 compared with controls;
bp < .05 compared with the
ethanol-treated group.
|
|
In control rats not receiving liver transplantation, bile flow rates
were around 92 µl/g liver/h, and acute ethanol treatment did not
significantly change this value. After transplantation, bile was
produced at rates of about 27 µl/g liver/h in control rats but was
reduced about 60% in grafts from alcohol-treated rats. Catechin
increased rates to 41 µl/g liver/h whereas
GdCl3 treatment before transplantation returned
bile production to control values (25 µl/g liver/h; Fig. 2, bottom).
Ethanol + catechin was significantly different from ethanol + GdCl3.
Effects of Ethanol, Catechin, and Gadolinium Chloride on Hepatic
Microcirculation.
The mean arterial blood pressure was about 90 mm
Hg in controls 2 h after transplantation. Ethanol did not alter
this value. To evaluate hepatic microcirculation, portal pressure and
liver surface oxygen tension were measured. Without liver
transplantation, portal pressure was 7.5 ± 0.1 cm of
H2O in controls (n = 4), and ethanol did not significantly alter this value. Three hours after implantation, portal pressure was around 7.2 ± 0.2 cm of
H2O in controls (n = 6) and was
increased to 9.1 ± 0.4 by acute alcohol treatment
(n = 10, p < .05 by ANOVA and
Student-Newman-Keuls posthoc test; data not shown). Catechin minimized
this increase to 7.6 ± 0.5 cm of H2O
(n = 5, p < .05 compared with the
ethanol group but p > .05 compared with the controls
by ANOVA plus Student-Newman-Keuls posthoc test).
GdCl3 also blunted the increases to 7.6 ± 0.1 cm of H2O (n = 4, p < .05 compared with the ethanol group but
p > .05 compared with the controls by ANOVA plus
Student-Newman-Keuls posthoc test).
Hepatic surface oxygen tension was around 82 µM in controls without
transplantation, a value not altered significantly by acute ethanol
treatment (data not shown). Fifteen minutes after implantation, hepatic
surface oxygen tension was around 80 µM in the control group, and
values decreased gradually to about 68 µM over 3 h (Fig.
3). In contrast, surface oxygen tension
reached only about 31 µM in the alcohol-treated group, reflecting
surface hypoxia due to disturbances in hepatic microcirculation.
Catechin minimized the decrease of oxygen tension caused by alcohol
treatment significantly, whereas GdCl3 reversed
the decrease almost completely.
View larger version (19K):
[in this window]
[in a new window]
|
Fig. 3.
Catechin and gadolinium chloride blunt the decrease
in liver surface oxygen tension caused by ethanol. Conditions as in
Fig. 1. Catechin (30 mg/kg i.v.) was injected into recipients when
blood flow was restored after implantation, and grafts were rinsed with
5 ml of Ringer's solution containing 400 µM catechin.
GdCl3 was given to some donors 24 h before
explantation to destroy KCs. Oxygen tension was measured by using a
Clark-type oxygen electrode placed gently on the liver surface with the
aid of a micromanipulator. Three to five points were measured on each
liver. Values are means ± S.E.M. (n = 4-12
in each group; ANOVA). ap < .05 compared with control; bp < .05 compared with the ethanol group (Student-Newman-Keuls test). Ethanol + catechin was also significantly different from Ethanol + GdCl3 (p < .05).
|
|
Graft hypoxia was also evaluated in awake animals after implantation by
using a tissue hypoxia marker, pimonidazole. Previous studies showed
that pimonidazole detects hypoxia in the liver (Arteel et al., 1995).
Figure 4 depicts representative images of
liver grafts in which pimonidazole adducts were detected
immunohistochemically. In livers from controls, pimonidazole adducts
were minimal and accumulated primarily in pericentral regions (Fig. 4,
top left); this is probably due to the natural low oxygen tension in
these regions. Treatment with ethanol increased the relative area of pimonidazole dramatically with binding extending from the pericentral regions toward periportal areas (Fig. 4, top right), reflecting much
more extensive hypoxia in fatty grafts. The area of
pimonidazole-labeled cells in grafts from ethanol-treated rats was
decreased significantly after treatment with catechin or Carolina Rinse
solution (Fig. 4, middle). Destruction of KCs with
GdCl3 or inhibition of leukotriene synthesis with
MK-886 also blunted increases in pimonidazole binding caused by ethanol
(Fig. 4, bottom).
View larger version (148K):
[in this window]
[in a new window]
|
Fig. 4.
Representative images of pimonidazole binding in
liver grafts after transplantation. Conditions as in Fig. 1.
Pimonidazole (120 mg/kg i.v.) was injected 30 min after implantation.
Grafts were perfused with Krebs-Henseleit buffer (pH. 7.4) 1 h
later to remove blood and fixed with 1% paraformaldehyde; then
sections were stained immunohistochemically (Arteel et al., 1995).
Representative images are shown: top left, control; top right, ethanol;
middle left, ethanol + catechin; middle right, ethanol + Carolina
Rinse; bottom left, ethanol + GdCl3;. bottom
right, ethanol + MK-886.
|
|
Whereas immunohistochemistry detects predominantly protein-bound
adducts, quantitation of pimonidazole binding with ELISA detects both
protein and nonprotein adducts (e.g., GSH adducts). The level of
pimonidazole binding in control liver grafts was 272 pmol/mg protein;
one dose of ethanol increased binding more than 3-fold (Fig.
5). Catechin or Carolina Rinse solution
diminished binding to about 240 pmol/mg protein, whereas destruction of
KCs (GdCl3) or inhibition of leukotriene
synthesis with MK-886 also blunted the increase in binding due to
ethanol significantly (Fig. 5).
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 5.
Quantitation of pimonidazole binding in fatty grafts.
Conditions as in Fig. 4, except that pimonidazole binding in liver
homogenates was quantified by using an ELISA. In some experiments,
MK-886 (50 µM) was added to UW solution, and Ringer's solution was
used to rinse grafts before implantation. Some grafts were rinsed with
5 ml of Carolina Rinse solution before implantation. Values are
means ± S.E.M. (ANOVA; n = 4-5 in each
group). ap < .05 compared with
control; bp < .05 compared with the
ethanol group.
|
|
Fat Content in Liver Grafts after Binge Drinking.
Disturbances
in hepatic microcirculation could be a result of fat accumulation in
the liver; therefore, liver sections were examined microscopically with
osmium staining. Fat droplets were barely detectable in parenchymal
cells in grafts from control rats (Fig.
6, top left). However, microvesicular and
macrovesicular fat droplets (black staining) were numerous in
parenchymal cells from livers of rats treated acutely with ethanol
(Fig. 6, top right). The extent of fatty infiltration and its lobular
distribution in grafts from rats treated with ethanol receiving
catechin (Fig. 6, middle left), Carolina Rinse (Fig. 6, middle right),
GdCl3 (Fig. 6, bottom left), or MK-886 (Fig. 6,
bottom right) were similar to those of grafts from rats receiving
ethanol alone. Hepatic triglyceride content determined enzymatically
was about 0.8 mg/mg DNA in the control group, whereas values were
elevated 2.8-fold by ethanol treatment. GdCl3 did
not significantly alter triglyceride content in the liver under these
conditions (Fig. 7), confirming the
histological findings.
View larger version (150K):
[in this window]
[in a new window]
|
Fig. 6.
Fatty infiltration in liver grafts. Conditions as in
Fig. 1. After implantation (1.5 h), grafts were rinsed with normal
saline followed by infusion of 10 ml of 4% paraformaldehyde. Fixed
tissue was stained with osmium (Luna, 1968), and sections were
counterstained with eosin. Typical images (original magnification,
200×) of at least four sections in each group. Top left, control; top
right, ethanol-treated group; middle left, ethanol + catechin; middle
right, ethanol + Carolina Rinse; bottom left, ethanol + GdCl3; bottom right, ethanol + MK-886.
|
|
View larger version (13K):
[in this window]
[in a new window]
|
Fig. 7.
Effects of acute ethanol and gadolinium chloride
treatment on hepatic triglyceride content. Conditions as in Fig. 1.
Liver homogenates were extracted with a mixture of chloroform and
methanol (2:1, v/v) (Brodie et al., 1961), and hepatic
triglycerides were determined by enzymatic methods. Data are expressed
per unit DNA to rule out any possible effect of liver weight changes
due to fat accumulation. Values are means ± S.E.M.
(p < .05 by ANOVA; n = 4 in each
group). *p < .05 for comparison with controls
(Student-Newman-Keuls test).
|
|
Free Radical Production after Liver Transplantation.
Because
oxidative stress could lead to microcirculatory disturbances (Ikai et
al., 1994), free radical production after liver transplantation was
evaluated using the spin trapping reagent 4-POBN and ESR. A six-line
ESR spectrum due to a 4-POBN/radical adduct was detected in bile
samples from all livers 1 h after transplantation (4.3 ± 0.6 arbitrary units). Acute ethanol treatment of the donor rat before liver
explantation increased the signal amplitude about 2-fold (9.4 ± 1.2 arbitrary units, p < .05; Fig. 8). Computer simulation of the spectrum
was accomplished by using hyperfine coupling constants of
aN = 15.70 G and
aH = 2.66 G for a single radical
species. Such coupling constants are characteristic of a
carbon-centered 4-POBN radical adduct and match closely values
(aN = 15.63 G;
aH = 2.73 G) obtained from bile of
rats given spin trap and oxidized polyunsaturated fatty acids
(Chamulitrat et al., 1992). Based on comparison with model systems,
this type of spectrum has been assigned to the POBN/pentyl radical
adduct or a closely related species. Pentyl free radicals would be
formed in vivo upon the 
-scission of arachidonic or linoleic
acid-derived alkoxyl radicals (Kadiiska et al., 1998). Catechin blunted
free radical adduct formation with values reaching only about 20% of
those observed in grafts from ethanol-treated rats (2.0 ± 0.6 arbitrary units, Fig. 8). When fatty grafts were rinsed with Carolina
Rinse solution before implantation, free radical adduct formation was
reduced by about 50% (5.0 ± 1.5 arbitrary units). Destruction of
KCs with GdCl3 also reduced free radical
production significantly (3.0 ± 0.6 arbitrary units, Fig. 8).
View larger version (36K):
[in this window]
[in a new window]
|
Fig. 8.
ESR spectrum of free radical adducts after
transplantation. Conditions as in Fig. 1. The spin trapping reagent
4-POBN (1 g/kg body weight) was dissolved in 0.5 ml of normal saline
and injected slowly into the tail vein upon opening of vascular clamps.
Bile was collected into 50 µl of 30 mM dipyridyl on ice to prevent ex
vivo free radical formation for 1 h after injection of 4-POBN.
Free radical adducts in bile were detected with a Bruker ESP 106 EPR
spectrometer. Typical spectra (n = 4-5 in each
group). A, control; B, ethanol; C, ethanol + catechin; D, ethanol + Carolina Rinse; E, ethanol + GdCl3.
|
|
| |
Discussion |
Binge Drinking Disturbs Microcirculation in Fatty Grafts after
Tranplantation.
Hepatic microcirculation is pivotal for graft
survival; however, cold storage/reperfusion causes critical injury to
sinusoidal lining cells (Marzi et al., 1989). Disturbance of hepatic
microcirculation increases graft injury and failure after
transplantation (Marzi et al., 1990), whereas rinsing grafts with warm
lactated Ringer's to improve microcirculation increases survival
(Takei et al., 1991a). Previous studies have shown that the velocity of
polymorphonuclear neutrophils decreases and leukocyte adhesion
increases after liver transplantation (Marzi et al., 1990). Moreover,
platelet adhesion is increased and blood flow is reduced (Takei et al.,
1991b). It is known that infusion of ethanol disturbs microcirculation, which leads to liver injury (Hijioka et al., 1991). Therefore, it is
possible that ethanol increases graft injury by disturbing the
microcirculation, which leads to graft hypoxia and injury after
transplantation. Indeed, liver surface oxygen tension decreased dramatically and portal pressure increased in fatty grafts from ethanol-treated rats (Fig. 3), confirming that ethanol disturbs the microcirculation.
Most of the methods for the assessment of microcirculation, such as in
vivo microscopy or detection of hemoglobin absorbance on the liver
surface, are conducted in anesthetized animals, which could influence
microcirculation. To avoid this potential problem, pimonidazole, a
2-nitroimidazole compound, was used in this study to assess graft
hypoxia in vivo after transplantation in the absence of anesthesia.
This compound is reductively activated at low oxygen concentrations and
binds to cellular macromolecules (Durand and Raleigh, 1998). It has
several distinct advantages. It accumulates in vivo in intact, awake
animals and measures tissue hypoxia directly at the cellular level,
unlike other methods. Pimonidazole binding was increased massively
after transplantation in ethanol-treated grafts (Figs. 4 and 5),
indicating that significant tissue hypoxia was associated with
increased graft injury (Fig. 2, top), inhibited liver function (Fig. 2,
bottom), and decreased survival (Fig. 1). Importantly, the use
of free radical scavengers, destruction of KCs, and inhibition
of biosynthesis of leukotrienes with MK-886, which improved graft
microcirculation (Figs. 3 and 4), significantly decreased injury (Fig.
2), and reduced graft failure (Fig. 1). Taken together, these data are
consistent with the hypothesis that ethanol causes graft injury by
disturbing the hepatic microcirculation, leading to hypoxia.
Fat Accumulation Is Not Responsible for Graft Hypoxia.
How
ethanol disturbs graft microcirculation leading to hypoxia after
transplantation remains unknown. One attractive hypothesis is that it
is due to fatty infiltration caused by ethanol. It is well known that
alcohol induces fatty liver: even one single large dose of alcohol
significantly increases hepatic lipid content (Ylikahri et al., 1972).
Severely steatotic grafts transplant poorly, often necessitating
replacement surgery resulting in high mortality (Todo et al., 1989);
therefore, biopsies are performed at many transplant units, and livers
with highly elevated fat infiltration are often not utilized. Fat
droplets in parenchymal cells could theoretically cause obstruction of
the hepatic microcirculation mechanically, and rupture of parenchymal
cells in a liver with severe steatosis could result in release of fat
globules into the hepatic microcirculation (Todo et al., 1989).
However, livers from rats fed ethanol-containing high-fat or low-fat
diets chronically have similar cell injury after hypoxia/reoxygenation,
although only the former exhibit fatty infiltration (Zhong et al.,
1998). In this study, one large dose of ethanol caused massive fat
accumulation in hepatocytes (Fig. 6); however, free radical scavengers,
removal of KCs, and inhibition of leukotriene biosynthesis all
dramatically improved hepatic microcirculation reflected by higher
liver surface oxygen tension, lower pimonidazole binding, lower portal
pressure, and higher bile flow [Figs. 2 (bottom), 3, and 4]
without affecting hepatic fat content (Fig. 6). Therefore, it is
concluded that accumulation of fat droplets in parenchymal cells is not
responsible for disturbances in hepatic microcirculation in the binge
drinking model studied here.
Oxidative Stress Plays an Important Role in Ethanol-Induced
Disturbances in Hepatic Microcirculation after Transplantation.
An
important finding of this study is that catechin, a free radical
scavenger, and Carolina Rinse significantly improved hepatic microcirculation and minimized ethanol-induced hypoxia after
transplantation (Figs. 3-5). These results suggest that ethanol
disturbs hepatic microcirculation, most likely by increasing oxidative
stress (Fig. 9). Indeed, free radical
production increased 2-fold in grafts from ethanol-treated rats, an
effect that was blocked by catechin and Carolina Rinse (Fig. 8).
Reactive oxygen species could cause damage to endothelial cells, thus
disturbing the microcirculation. Alternatively, previous studies have
shown that reactive oxygen species induce vasoconstriction in isolated
perfused rat livers (Ikai et al., 1994), and ischemia/reperfusion
stimulates secretion of vasoactive and chemotactic mediators (Fig. 9)
such as platelet-activating factor (Nishiyama et al., 1993) and
leukotrienes (Clavien et al., 1993), thus causing vasoconstriction,
leukocyte plugging, and platelet adhesion (Fig. 9). Oxidative stress
activates nuclear factor-B, a transcription factor
responsible for the expression of proinflammatory cytokines and cell
adhesion molecules (Lin et al., 1995). In addition, reactive oxygen
species activate phospholipase A2 (Goldman et
al., 1997), which increases production of lipid-derived vasoactive and
chemotactic mediators such as eicosanoids and platelet-activating factor (Fig. 9). All of these pathophysiological alterations caused by
oxidative stress could contribute to disturbances of graft microcirculation and hypoxia. Indeed, MK-886, which inhibits activation of 5-lipoxygenase thus blocking biosynthesis of leukotrienes (Gillard et al., 1989; Fig. 9), significantly minimized hepatic hypoxia (Figs. 3
and 4) and improved survival (Fig. 1) after transplantation of fatty
grafts from ethanol-treated rats. It is also possible that MK-886 works
by diminishing leukotriene production with a resultant increase in
other eicosanoids, which dilate vessels. These results support the
hypothesis that ethanol increases oxidative stress after
transplantation, which in turn stimulates production of vasoactive and
chemotactic mediators and disturbs the microcirculation of fatty grafts
(Fig. 9). Why catechin eliminated free radicals yet did not totally
reverse changes in survival and surface oxygen tension to control
levels is unclear; it is possible that other factors are involved.
View larger version (51K):
[in this window]
[in a new window]
|
Fig. 9.
Diagram depicting the working hypothesis by which
binge drinking causes hypoxia in fatty grafts after transplantation. It
is proposed that acute ethanol treatment ("binge drinking") causes
graft hypoxia after transplantation by the following mechanisms:
ethanol activates KCs and increases formation of free radicals via
NADPH oxidase and xanthine oxidase (XO). Reactive oxygen species
stimulate production of chemotactic and vasoactive mediators such as
eicosanoids, platelet-activating factor, and cytokines, which causes
vasoconstriction and increases white blood cell and platelet adhesion,
leading to microcirculatory disturbances and hypoxia. GdCl3
destroys KC, catechin scavenges free radicals, and MK-886 inhibits
synthesis of leukotrienes, thereby minimizing ethanol-induced graft
hypoxia, liver injury, and mortality. SC, stellate cell; EC,
endothelial cell; EtOH, ethanol; O2.-,
superoxide radicals; Xan, xanthine; HX, hypoxanthine; PLA2,
phospholipase A2.
|
|
Ethanol increases free radical formation after transplantation, most
likely by activating KCs (Fig. 9). It is well known that macrophages
and neutrophils in previously ischemic tissue produce superoxide
radicals upon activation. Moreover, ethanol treatment increases gut
permeability, which could increase absorption of endotoxin (Bode et
al., 1987), a known activator of KCs. LPS-induced elevation in
intracellular calcium in KCs is stimulated significantly in 24 h
after one large dose of ethanol (Enomoto et al., 1998), the time point
when the donor liver was explanted here. Previous studies also showed
that destruction of KCs diminishes free radical formation in a blood
element-free liver perfusion model (Bremer et al., 1994), indicating
that activated KCs alone represent an important source of free
radicals. Moreover, KCs are a major source of chemotactic mediators,
such as platelet-activating factor, leukotrienes, thromboxanes, and
toxic cytokines in the liver (Decker, 1990; Fig. 9). Indeed,
destruction of KCs with GdCl3 largely blocked free radical production and significantly minimized microcirculatory disturbances caused by ethanol in this study (Figs. 4 and 8). Thus, it is concluded that acute ethanol treatment ("binge
drinking") causes graft hypoxia after transplantation most likely by
activating KCs and increasing formation of free radicals. Reactive
oxygen species activate production of chemotactic and vasoactive
mediators, which leads to microcirculatory disturbances and hypoxia.
GdCl3 destroys KCs, catechin scavenges free
radicals, and MK-886 inhibits synthesis of leukotrienes, thereby
minimizing ethanol-induced graft hypoxia, liver injury, and mortality
(Fig. 9).
KC, Kupffer cell;
ELISA, enzyme-linked
immunosorbent assay;
UW, University of Wisconsin;
GdCl3, gadolinium chloride;
AST, aspartate aminotransferase;
4-POBN, -(4-pyridyl
1-oxide)-N-tert-butylnitrone;
ESR, electron spin resonance.
![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
Top
Abstract
Introduction
United States, 1982-1993.
Morb Mortal Wkly Rep
43:
861-867[Medline].
