 |
INTRODUCTION |
MARKED HEMODYNAMIC CHANGES occur in humans and
experimental animals with cirrhotic liver disease. A hyperdynamic
circulation develops that is characterized by increased cardiac output,
increased splanchnic blood flow, low total systemic vascular
resistance, mild tachycardia, low or normal blood pressure, an
increased blood volume, portal hypertension, and reduced responsiveness
to vasoconstrictors. These changes can occur whether the disorder is
the result of cirrhosis (27, 32, 49), prehepatic portal obstruction (9, 19, 27, 35, 59), or portocaval shunting (18, 33, 58).
Ventricular dilatation, especially of the right heart, and hypertrophy,
especially in early cirrhosis, occur in human hearts (43). In animal
studies (20), the heart weight in CCl4-induced cirrhotic
rats is greater than that in controls, presumably because of chronic
ventricular overload. There are no cardiac histological abnormalities
in these animals (20). Similarly, Ma and colleagues (45) reported no
light microscopic changes in the hearts of bile duct ligation-induced
cirrhotic rats. However, these observations do not duplicate what is
seen in human cirrhotic patients, most likely because of the long
duration of cirrhosis in patients compared with the relatively brief
time the experimental animals are cirrhotic. In the portal vein
stenosis model, no change in either right or left heart size occurs
during the brief interval of 10-12 days postsurgery (Battarbee and
Zavecz, unpublished observations), suggesting that either hypertrophy
does not occur or impairment is of insufficient duration to result in
measurable hypertrophy.
In addition to altered basal hemodynamics, liver disease also disrupts
dynamic function. Tilt tests, lower body negative pressure, and other
hypotensigenic techniques indicate that both cardiac and peripheral
resistance reflex responses are impaired in hepatic disease patients
(11, 12, 41, 42) and in experimental portal hypertension (6).
Furthermore, cirrhotic patients (1, 8, 25, 34, 38, 46, 54),
experimental models of cirrhosis (13, 36), prehepatic portal
hypertensive models (6, 7), and cholestatic animals (13-15)
exhibit reduced cardiac responses to exogenous and endogenous
catecholamines. Studies (26, 43) both in human nonalcoholic cirrhosis
and in animal models of nonalcoholic cirrhosis have demonstrated
impairment of cardiac contractility in response to various stressors.
This development of high-output heart failure with systemic
vasodilation has been termed "cirrhotic cardiomyopathy." Although
the ventricular dysfunction coexisting with a high cardiac output
secondary to reduced peripheral vascular resistance is common in
cirrhotic patients (38, 43), the symptoms are usually latent, only
appearing under conditions that stress the myocardium, including liver
transplantation, surgical portosystemic shunting, transjugular
intrahepatic portosystemic stent shunts, mental stress,
physical exercise, and pharmacological stimulation (26, 43).
Interestingly, portocaval shunting, not hepatocellular disease, is the
common factor in the hepatic models used to study the cardiovascular
effects of liver disease. Thus it appears that the hyperdynamic
circulation results from the shunting of visceral venous blood into the
systemic circulation (for an in-depth discussion of cirrhotic
cardiomyopathy, see Ref. 43).
In our previous studies (7, 65) with the portal vein-stenosed rat, we
dissociated hepatocellular dysfunction from the effect of portal
hypertension using the chronic portal vein-stenosed rat, a model of
chronic liver disease in which a hyperdynamic circulation is present
without the hepatocellular damage of cirrhosis. We have
observed depressed contractile function in isolated right and left
ventricular tissue from chronic portal vein-stenosed rats as well as
decreased isoproterenol-induced positive inotropism. 
-Adrenoceptors
have been shown not to be downregulated in this model, but a greater
fraction of cardiac -adrenoceptors must be occupied to produce
equivalent absolute increases in dT/dt, the maximum
rate of tension development (65), suggesting that the cause of the
decreased response to isoproterenol is postreceptor. On the other hand,
it is conceivable that the altered responsiveness to -adrenergic
receptor activation results from a mechanism not directly involving
-adrenergic signal transduction. In the present study, we
have extended our research to include the effects of portal
hypertension with portosystemic shunting on cardiac Gs
and Gi expression, -adrenoceptor-Gs
coupling, and excitation-contraction coupling (ECC).
| |
METHODS |
Contractile experiments.
Ten to twelve days after portal vein stenosis or sham operation, each
animal was decapitated and the heart was immediately excised and
transferred to a preparative tissue bath containing Krebs-Henseleit
solution (KH) equilibrated with 95% O2-5%
CO2. The buffer contained (in mM) 118 NaCl, 5.8 KCl, 27.2 NaHCO3, 1.0 NaH2PO4, 1.2 MgSO4, 2.5 CaCl2, and 11.1 glucose. The
temperature of the buffer was maintained at 37°C and the pH at 7.4. A small strip of right ventricle was removed, and a left ventricular
papillary muscle was dissected free. The remainder of the heart was
immediately frozen in liquid nitrogen and stored at 
70°C for
use in the binding experiments (see below). One end of each muscle was
attached to a rigid support and then placed in an organ bath, where the
other end of the muscle was attached to a Grass FT03C
force-displacement transducer via a length of surgical silk. Muscles
were field stimulated at 2 Hz, with a voltage 50% greater than
threshold and a pulse duration of 4 ms. Resting length was set to the
peak of each muscle's length-tension curve. Isometric tension was
recorded by means of the force-displacement transducer connected to a
Grass model 7D recorder. dT/dt and the maximum rate of
relaxation (dT/dt) were obtained by
differentiating the output of the channel measuring isometric force. At
the end of each experiment, a graticule was used to measure each
muscle. Tissues were dried and their weights recorded. These
measurements were used to normalize the data for differences in tissue
weight and dimensions.
In the experiments in which the effect of portal vein stenosis on the
response of the myocardium to -adrenoceptor activation with
isoproterenol was studied, tissues were equilibrated in KH, and when
contractions stabilized, isoproterenol was added to the tissue bath. On
attainment of a stable positive inotropic effect, isoproterenol was
washed out of the tissue. When contractions returned to the control
level, the next concentration of isoproterenol was added. Exposure to
the different concentrations of isoproterenol was randomized.
In experiments examining the effect of portal vein stenosis on the
relationship between contractile strength and the extracellular Ca2+ concentration, the tissues were equilibrated with KH
containing 2.5 mM Ca2+. After force stabilized, the bathing
solution was quickly changed to KH with 0.3125 mM Ca2+.
When force stabilized, which required 2-3 min, an appropriate amount of Ca2+ was added to produce the next
Ca2+ concentration. Concentrations of Ca2+
>7.5 mM were not investigated because of the potential for
Ca2+ precipitation with the buffer system and because the
maximal response had already been achieved. The brief period of
exposure to low Ca2+ did not affect subsequent muscle
performance, as indicated by the similarity of force measured at the
initial 2.5 mM Ca2+ concentration and that measured at 2.5 mM Ca2+ during the concentration-response experiment (sham:
1,079 ± 117 and 1,147 ± 94 mN/cm2, respectively;
stenosed: 634 ± 130 and 660 ± 34 mN/cm2, respectively,
in papillary muscle).
Experiments with the dihydropyridine L-type Ca2+ channel
activator BAY K 8644 were performed to assess further Ca2+
channel functionality. After contractions stabilized, BAY K 8644 (10
8-3 × 106 M) was
added cumulatively. Succeeding concentrations were added when the
response to the previous concentration stabilized.
Experiments in permeabilized muscle fibers.
To determine the effect of portal vein stenosis on the Ca2+
sensitivity of the myofilaments and Ca2+ uptake and release
from the sarcoplasmic reticulum (SR), left ventricular papillary
muscles were chemically skinned with either saponin or Triton X-100 to
permeabilize the sarcolemma while leaving the SR intact or to
permeabilize both the sarcolemma and the SR membrane, respectively. In
this way, the "intracellular" cytosolic concentration of
Ca2+ can be controlled and manipulated. The solutions used
in these experiments were prepared according to a computer program that takes into account the binding constants of all of the constituents (51). Papillary muscles were removed from left ventricles, and 100- to
150-µm-diameter fiber bundles, 1-2 mm in length, were dissected
free in relaxing solution containing 5 mM MgATP, 1 mM Mg2+,
5 mM EGTA, 20 mM imidazole, 15 mM creatine phosphate, and potassium methanesulfonate to yield an ionic strength of 200 mM with pCa > 8.5 adjusted to pH 7.0. One end of each bundle was tied with a human hair
to a force transducer (Kent Scientific model TRN001) and the other to a
support that positioned the fibers horizontally in a 3-ml tissue bath.
The fibers were stretched to a point at which passive tension was just
measurable (~0.05 mN). The output of the transducer was digitized and
stored on disk for subsequent analysis. All experiments were performed
at room temperature. To assess the effect of portal vein stenosis on
the sensitivity of the myofilaments to Ca2+, the fiber
bundles were permeabilized by superfusion (2 ml/min) with oxygenated
relaxing solution containing 0.5% Triton X-100 for 1 h. This treatment
permeabilizes both the sarcoplasmic reticular and sarcolemmal membranes
(24, 66). Fibers were superfused with relaxing solution for an
additional 60 min before exposure to Ca2+. In the
experiments that examined the effect of portal vein stenosis on
Ca2+ uptake and release from the SR, fibers were chemically
permeabilized by superfusion with 50 µg/ml saponin for 30 min to
permeabilize the sarcolemma only (24, 66). Saponin was washed out of
the tissues, which were then superfused for an additional 30 min in relaxing solution before the start of the experiment.
Myofilament Ca2+
sensitivity protocol.
After permeabilization with Triton X-100, fibers were checked for the
presence of functional SR by exposure to 50 mM caffeine. Preparations
that responded to caffeine were discarded. pCa-force curves were
generated by sequential application of solutions containing 1 mM MgATP,
1 mM Mg2+, 5 mM EGTA, and 20 mM imidazole, potassium
methanesulfonate to yield an ionic strength of 200 mM, and
Ca2+ (as CaCl2) to achieve the desired pCa. pH
was adjusted to 7.0 at room temperature. The data from each group were
fit to the Hill equation, and the half-maximum pCa and Hill coefficient
(nH) were determined as indexes of myofilament
sensitivity and cooperativity, respectively (17).
SR Ca2+ uptake and release.
The effect of portal vein stenosis on the uptake and release of
Ca2+ by the SR was investigated in permeabilized left
ventricular papillary muscle fiber bundles using the ability of high
concentrations of caffeine to induce Ca2+ release through
the SR Ca2+-release channels. The magnitude of the
contracture induced by 30-50 mM caffeine was the index of SR
Ca2+ content and release.
In addition to relaxing solution (ionic strength, 150 mM), the
following solutions were utilized with the permeabilized fibers: 1) loading solution that contained relaxing solution with the addition of pCa 6.0; and 2) low-EGTA solution made up of
relaxing solution with 0.1 mM Mg2+ (lowered from 1 mM to
increase myofilament Ca2+ sensitivity and to reduce
Mg2+ inhibition of the SR Ca2+ release
channels) and 0.05 mM EGTA. After skinning for 20-30 min with 50 µg/ml saponin in relaxing solution to remove the sarcolemma and leave
the SR intact, any residual Ca2+ in the SR was released by
the application of 50 mM caffeine (in low-EGTA solution). After 1 min,
the tissues were washed with relaxing solution for 2 min. The bath
solution was changed to the Ca2+ loading solution, and the
SR was permitted to take up Ca2+ (pCa 6.0) for 3 min, after
which the tissues were washed with low-EGTA solution for 1 min. The
Ca2+ content of the SR was then estimated by exposing the
fibers to 50 mM caffeine in low-EGTA solution and measuring the
consequent contraction. The tissues were washed with low-EGTA solution
until force returned to baseline, when the tissue was exposed to
caffeine a second time. This second caffeine exposure was used to
confirm that all of the caffeine-sensitive Ca2+ had been
emptied from the SR (see Fig. 7A). These procedures ensured
that loading the SR with Ca2+ always occurred under the
same conditions, i.e., the experiments would not be subject to vagaries
associated with partially loaded SR before the Ca2+ loading
period, and that differences in the response to caffeine between
sham-operated and portal vein-stenosed rats were not the result of a
defect in uptake rather than release. Loading and subsequent
Ca2+ release by caffeine was performed three times in each
muscle with identical results each time, indicating that
Ca2+ uptake by the SR under these conditions was consistent
and reproducible (not shown).
Caffeine-induced Ca2+ release from the SR was also examined
in intact fibers, i.e., unpermeabilized fibers. Tissues were removed from the animal, placed into tissue baths, and paced as described above
with the exception that the stimulation frequency was 0.5 Hz. On
stabilization of contractions, electrical pacing was terminated, and 5 min later tissues were exposed to 30 mM caffeine. Once the contracture
began to relax, the KH with caffeine was washed out with normal KH, and
the tissues were allowed to rest for 30 min with frequent exchanges of
KH. The bath solution was then changed to Ca2+-free KH
buffer containing 2 mM EGTA (Ca2+-free solution with normal
Na+) for 5 min and then exposed to 30 mM caffeine.
Ca2+-free solution with normal Na+ stimulates
Na+/Ca2+ exchange at rest (4), and in this
manner the changes in the response to caffeine in normal KH were
examined to determine whether this results from changes in
Na+/Ca2+ exchange induced by portal vein
stenosis or increased Ca2+ leak from the SR during rest.
Determination of dihydropyridine receptor density.
The effect of portal vein stenosis on the density of dihydropyridine
receptors located on L-type Ca2+ channels was assessed
using equilibrium binding of the dihydropyridine antagonist
[3H]isradipine. Rat cardiac membranes were
prepared according to the procedure described by Ebersole et al. (23).
Hearts were weighed, minced with scissors, and suspended in 2 vol of
50 mM Tris · HCl (pH 7.7 at 25°C) at
4°C. Tissues were homogenized twice with a Polytron homogenizer for
10 s at a setting of 6. The homogenate was centrifuged at 24,000 g for 10 min. The pellet was resuspended in 3 vol of
Tris · HCl and centrifuged at 24,000 g for 10 min. This process was repeated three additional times. The final pellet was suspended in 4 vol of Tris · HCl (1 g original
weight to 4 ml buffer). Membrane preparations were stored at
70°C for no longer than 3 days before use. Membranes were
thawed and diluted 1:9 with Tris · HCl buffer for use
in the binding assays. Protein content was determined as described by
Lowry et al. (39).
Binding assays were performed at 37°C for 30 min in a total volume
of 200 µl containing 50 mM Tris · HCl (pH 7.4), 100 µg of membrane protein, 1 mM CaCl2, and 30-600 pM
[3H]isradipine (85.8 Ci/mmol; DuPont NEN,
Boston, MA). Nonspecific binding, which amounted to 10-20% of
total binding, was determined by the addition of 1 µM nitrendipine to
appropriate incubation vessels. [3H]isradipine
bound was determined by filtration (65). Each assay was performed in
duplicate. Receptor density and the apparent equilibrium dissociation
constant (Kd) for
[3H]isradipine were determined using nonlinear
regression (22).
-Adrenoceptor-G protein coupling.
When Gs binds to the agonist--adrenoceptor complex,
forming an agonist--adrenoceptor-Gs complex, the agonist
is bound with higher affinity than when it is bound to the receptor
only. The fraction of -adrenoceptors in this "high-affinity"
state is indicative of the coupling of the agonist-receptor complex
with Gs. The fraction of -adrenoceptors in the
high-affinity state and the apparent dissociation constant of an
agonist for the high-affinity state were determined as described
previously (52, 53). Competition binding experiments were
performed using cardiac membranes prepared as previously described
(65). Left ventricles were minced finely and homogenized with a
Polytron PT10 homogenizer (twice at half-speed for 5 s) in 35 ml of
ice-cold buffer containing 5 mM Tris · HCl (pH 7.4),
1 mM MgCl2, 0.25 M sucrose, 1 mM EDTA, and 1 mM
dithiothreitol (DTT). The homogenate was filtered through three layers
of cheesecloth and centrifuged at 12,000 g for 10 min. The
supernatant was centrifuged at 45,500 g for 25 min,
resuspended, and recentrifuged twice. The final pellet was suspended in
2 ml of buffer (50 mM Tris · HCl, pH 7.4, 10 mM
MgCl2, 1 mM EDTA, and 1 mM DTT). A 100-µl aliquot was
removed for protein determination, and the remainder was stored in
liquid nitrogen until used. Approximately 90% of all preparations were
used within 48 h, and the remainder were used within 1 wk of freezing.
Binding assays were performed at 37°C for 30 min in a total volume
of 0.5 ml containing 50 mM Tris · HCl (pH 7.4), 5 mM
MgCl2, 1 mM ascorbic acid, 50 µg of protein, 25-50
pM [125I]iodopindolol (46 Ci/mmol; Amersham
Life Sciences, Arlington Heights, IL), and 15 different concentrations
of isoproterenol (1011-104M)
with or without the addition of 2 mM guanylylimidodiphosphate. Nonspecific binding, which amounted to 10-15% of total binding, was determined by the addition of 100 µM ()isoproterenol to
appropriate incubation vessels. Binding of
[125I]iodopindolol to -adrenoceptors was
assessed by filtration (52, 65). Assays were performed in duplicate.
The curves obtained by plotting
[125I]iodopindolol bound vs. the log of the
isoproterenol concentration were analyzed by nonlinear regression (22),
and IC50 values for isoproterenol and the percentage bound
to high- and low-affinity states were determined.
KH and KL, the apparent
Kd for isoproterenol binding to the high- and
low-affinity states, respectively, were calculated by the method of
Cheng and Prusoff (21).
G protein separation and immunoblotting.
G protein expression was assessed using the Western blot technique.
Cardiac membranes containing 50 or 100 µg of protein were mixed with
an equal volume of a buffer containing 0.125 M
Tris · HCl, pH 6.8, 4% SDS, 20% glycerol, and 10%
2-mercaptoethanol (treatment buffer), heated at 95°C for 2 min, and
subjected to SDS-PAGE, 9% acrylamide in 6 M urea (55). After
electrophoresis, proteins were transferred onto Immobilon-P membranes
(Millipore, Bedford, MA) for immunoblotting. After protein transfer,
Immobilon-P membranes were incubated in TBS (20 mM Tris, pH 7.6, and
137 mM NaCl) containing 0.1% Tween plus 5% milk for 2 h at room
temperature with gentle shaking. The membranes were then incubated
overnight with primary antisera (1:1,000 dilution in the same buffer).
Antisera used were RM/1 (directed against carboxy-terminal decapeptide sequences of Gs), AS/7 (directed against
carboxy-terminal decapeptide sequences of Gi-1 and
Gi-2), Gi-1 (directed against the
internal amino acid sequence 159-168 and does not cross react with
Gi-2), EC/2 (directed against carboxy-terminal
decapeptide sequences of Gi-3), and GC/2 (directed
against amino-terminal decapeptide sequences of Go) (60,
61). Secondary antibody coupled to horseradish peroxidase was used for
detection of proteins by enhanced chemiluminescence (ECL Western blot
system, Amersham). Immunoreactive bands on X-ray films were digitized
using the gel documentation system (GDS 7500 from UVP) and quantified
using the GelBase software program (UVP) on an IBM computer.
ADP ribosylation of G proteins.
For pertussis toxin (PTX)-catalyzed ADP-ribosylation experiments,
membrane protein (100 µg) was incubated with a mixture containing 50 mM Tris · HCl (pH 8.0), 0.25% Lubrol, 20 mM
thymidine, 1 mM ATP, 5 µM GTP, 20 mM arginine, 50 mM NaCl, 4 µM
MgCl2, 100 mM DTT, 1 µg of PTX, and 3-5 µCi
[-32P]NAD in a final volume of 0.1 ml. The
mixtures were incubated for 2 h at 30°C, after which 1 ml ice-cold
acetone was added and samples were centrifuged for 10 min at 10,000 rpm
in a Microfuge at 4°C.
For cholera toxin (CTX)-induced ADP-ribosylation experiments, CTX was
activated by preincubation in 50 mM DTT for 10 min at 37°C.
Membrane proteins (100 µg) were preincubated in 200 µM
guanylylimidodiphosphate for 10 min at 37°C, and then an assay
mixture containing 20 mM thymidine, 100 mM NaCl, and 20 mM HEPES (pH
7.7) was added. The activated CTX (5 µg) and 3-5 µCi
[-32P]NAD were added to make a final volume
of 0.1 ml. The mixture was incubated for 2 h at 30°C, and then 1 ml
of ice-cold HEPES (20 mM, pH 7.7) was added and the mixture was
centrifuged for 10 min in a Microfuge at 4°C. The pellet was
resuspended in 0.5 ml of the same buffer and centrifuged again for 10 min.
The protein pellets were resuspended in treatment buffer and separated
by 6 M urea SDS-PAGE as described above for immunoblot analysis. Gels
were dried and exposed to X-ray film. Bands on the films were analyzed
by densitometry as described above. Five samples each from
sham-operated and portal vein-stenosed animals were analyzed. Each
sample was analyzed on at least two different gels.
Chemicals.
Caffeine, isoproterenol, EGTA, creatine phosphate, and creatine
phosphokinase were purchased from Sigma Chemical (St. Louis, MO).
Nitrendipine and BAY K 8644 were purchased from Research Biochemicals
International (Natick, MA). RM/1, AS/7, EC/2, and GC/2 antisera used in
the immunoblots were purchased from DuPont NEN. Gi-1
antiserum was purchased from Calbiochem (San Diego, CA). Other reagents
used for immunoblotting G proteins were the best grade available.
Statistics.
Because there were only two groups of animals, the appropriate
univariate t-test was used to evaluate the data. P < 0.05 was considered significant.
| |
RESULTS |
Basal contractile function.
Table 1 summarizes the effect of 10-12
days of portal vein stenosis on basal developed tension,
dT/dt, and dT/dt in left ventricular papillary muscles and right ventricular strips. Developed tension and dT/dt were decreased 30-46% and
28-50%, respectively. Portal vein stenosis had no significant
effect on the rate of relaxation.
-Adrenoceptors.
Although the absolute increase in contractility induced by
isoproterenol was markedly decreased by portal vein stenosis (Fig. 1A), there was no difference in
EC50 (sham, 28.8 ± 2.8 nM; stenosed, 25.0 ± 2.7 nM).
Figure 1B demonstrates that the relative change in
dT/dt was not different between the sham-operated and
portal vein-stenosed groups. Similar results were observed in both left and right ventricular tissue (right ventricle not shown). The effect of
portal vein stenosis on the enhancement of dT/dt
by isoproterenol is shown in Fig. 2. Portal
vein stenosis was without effect in either absolute (Fig. 2A)
or relative terms (Fig. 2B), and the EC50 was not
different (sham, 38.5 ± 13.8 nM; stenosed, 41.4 ± 8.4 nM). A
similar result was observed in the left ventricle (data not shown).
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 1.
Effect of portal vein stenosis on absolute (A) and relative
(B) increases in the maximal rate of tension development
(dT/dt) induced by isoproterenol in left ventricular
papillary muscle. Values are means ± SE of 4-5
experiments. EC50 did not differ between the 2 groups
irrespective of the manner in which data were presented (sham-operated
rats, 28.8 ± 2.8 nM; portal vein-stenosed rats, 25.0 ± 2.7 nM).
Effect of portal vein stenosis on contractile response to isoproterenol
was qualitatively similar in right ventricle. * P < 0.05.
|
|
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 2.
Effect of portal vein stenosis on absolute (A) and relative
(B) increases in the maximal rate of relaxation
(dT/dt) induced by isoproterenol in right
ventricle. Values are means ± SE of 5 experiments. EC50
did not differ between the 2 groups irrespective of the manner in which
data were presented (sham-operated rats, 38.5 ± 13.8 nM; portal
vein-stenosed rats, 41.4 ± 8.4 nM). Effect of portal vein stenosis on
the positive lusitropic action of isoproterenol was qualitatively
similar in left ventricle.
|
|
Although -adrenoceptors are not downregulated after 2 wk of portal
vein stenosis, and their affinity for isoproterenol is unaffected (65),
a nearly threefold increase in receptor occupancy in hearts from portal
vein-stenosed rats is required to produce the same increase in
myocardial contractility as in sham-operated rats (65). Therefore, we
examined whether a decrease in G protein expression or a defect in
-adrenoceptor-G protein coupling is responsible for the diminished
response to isoproterenol. Figures 3 and 4 illustrate
the effect of portal vein stenosis on Gs and Gi expression in the left ventricle. Cardiac membranes
from both sham-operated and portal vein-stenosed animals were incubated with specific antisera directed against selective peptide sequences in
the -subunits of different G protein subtypes. With the antiserum RM/1, which is specific for the -subunit of stimulatory G proteins (Gs), 3-4 immunopositive bands with a molecular
mass of 42-44 kDa were found to be present in cardiac membranes
from sham-operated and portal vein-stenosed rats. Representative
immunoblots are shown in Fig. 3C. The relative density of the
bands from ventricles taken from eight sham-operated and eight portal
vein-stenosed rats, all run on the same gel, was determined, and the
results are shown in Fig. 4. No difference in Gs
expression was detectable between the sham-operated and portal
vein-stenosed groups. Expression of the inhibitory G protein subtypes
Gi-1 and Gi-2 was detected using antisera
AS/7 (Gi-1 and Gi-2) and
Gi-1. EC/2 was used to detect Gi-3.
Figure 3B shows a representative autoradiograph of
immunodetectable Gi-1 and Gi-2 using
AS/7. An intense band with the same mobility as the lower band from
spinal cord membranes was detected in the cardiac membranes. This band
has been previously identified as Gi-2 (64). Two faint
immunopositive bands were also observed with AS/7 (Fig. 3B),
which could be Gi-1. Thus additional tests were
performed using antisera selective for Gi-1. Results of
these studies are shown in Fig. 3A. Spinal cord
membranes, which contain Gi-1 (64), showed two
immunopositive bands with this antisera, but no bands were observed in
the cardiac membranes. The autoradiographic intensity of the lower
bands (Gi-2) seen in Fig. 3B is shown in Fig. 4
for all eight sham-operated and all eight portal vein-stenosed rats,
demonstrating that there is no difference in expression of
Gi-2 in the sham-operated and the portal vein-stenosed
groups. No immunopositive bands for Gi-3 or
Go were detected in the rat heart (data not shown).
View larger version (52K):
[in this window]
[in a new window]
|
Fig. 3.
G protein -subunits in membranes from mouse spinal cord (SC) and rat
ventricles [sham and portal vein stenosis (pvs)].
Representative immunoblots of Gi-1 (A),
Gi-1 and Gi-2 (B), and
Gs expression in ventricles from sham-operated and
portal vein-stenosed rats. Spinal cord expression was included as an
internal control because spinal cord expresses Gi-1,
Gi-2, and Gs. A: antisera
selective for Gi-1. B: antiserum AS/7 selective
for Gi-1 and Gi-2. C: antiserum
RM/1 selective for Gs.
|
|
View larger version (28K):
[in this window]
[in a new window]
|
Fig. 4.
G protein -subunit expression in rat cardiac ventricles. The
relative density of bands from immunoblots of ventricular protein from
8 sham and 8 stenosed rats run on the same gel is represented. Protein
derived from hearts of the same sham and stenosed rats was used to
determine expression of all G proteins examined. Separate gels were run
for each different G protein. G protein expression in the portal
vein-stenosed (PVS) group is compared with expression in the sham
group. Expression of Gs is not compared with expression
of Gi-2. For Gs, the entire
immunopositive area, containing 3-4 relatively distinct bands, was
quantified. For Gi-2, the band migrating identically to
the lower band (Gi-2) from spinal cord membranes was
quantified. SO, sham operated.
|
|
The extent of ADP ribosylation of Gi (with PTX) and
Gs (with CTX) was examined to determine whether portal vein
stenosis altered G protein function. No differences in ADP ribosylation
were observed between the sham-operated and portal vein-stenosed groups
(data not shown).
The effect of portal vein stenosis on the coupling of the
-adrenoceptor agonist-receptor complex to Gs was
determined by quantifying the percentage of -adrenoceptors in the
high-affinity state of the receptor. The results are summarized in
Table 2. Although
KH was lower in the portal vein stenosis group,
suggesting a difference in the stability of the agonist-receptor-G
protein complex, the difference between the sham-operated and portal
vein-stenosed groups was not significant (P > 0.05). Furthermore, portal vein stenosis-induced portal
hypertension and portosystemic shunting did not affect the fraction of
receptors in the high-affinity state.
Extracellular Ca2+.
The effect of portal vein stenosis on the relationship between
extracellular Ca2+ and developed tension was studied in
left ventricular papillary muscle and right ventricular strips.
Developed tension was significantly reduced before any manipulation of
the extracellular Ca2+ concentration in portal
vein-stenosed rats (Table 1) and, as Fig.
5A demonstrates for papillary
muscle, it remained consistently less over the complete range of
Ca2+ concentrations examined. Maximal force development was
observed at the same extracellular Ca2+ concentration in
the sham-operated and portal vein-stenosed groups. Using force
developed in 2.5 mM Ca2+ before the experimental protocol
was performed to normalize the data in Fig. 5A, the two groups
could be compared despite the lower developed tension in the portal
vein stenosis group. This comparison is shown in Fig. 5B and
demonstrates that the relationship between the extracellular
Ca2+ concentration and the relative magnitude of force was
similar in the sham-operated and portal vein-stenosed groups when the data were normalized to account for the lower control tension in the
stenosed group. Similar results were obtained in right ventricle (data
not shown). Furthermore, the time course of the contractile response to
stepped changes in the extracellular Ca2+ concentration was
the same in the sham-operated and portal vein-stenosed groups (data not
shown). There was no difference in the small rise in diastolic tension
between the sham-operated and portal vein-stenosed groups at 5 and 7.5 mM Ca2+.
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 5.
A: effect of extracellular Ca2+ concentration on
developed tension in left ventricular papillary muscles from
sham-operated and portal vein-stenosed rats. Values are means ± SE of
7-8 muscles. Each papillary muscle was equilibrated in normal KH
containing 2.5 mM Ca2+ (control data point) before
switching to the lowest concentration of Ca2+ (0.3125 mM).
Subsequent addition of Ca2+ to attain the next higher
Ca2+ concentration was made after force had reached a
plateau. B: relative changes in papillary muscle force
generation with changes in extracellular Ca2+
concentration. Values represent mean ± SE percentage of the
control value in 2.5 mM Ca2+. Similar results were observed
in right ventricle (data not shown). * P < 0.05, ** P < 0.01 significantly different from sham.
|
|
[3H]isradipine binding.
The density of dihydropyridine binding sites on L-type Ca2+
channels was estimated by measuring the equilibrium binding of
[3H]isradipine, a dihydropyridine antagonist.
Table 3 summarizes the effect of portal
vein stenosis. No change in the affinity of isradipine for its binding
sites was observed, but dihydropyridine receptor density decreased
63%. nH for [3H]isradipine
binding to membranes from both sham-operated and portal vein-stenosed
rats did not differ from 1, indicating that isradipine interacted with
a single site.
View this table:
[in this window]
[in a new window]
|
Table 3.
[3H]isradipine binding to L-type Ca2+
channels in cardiac membranes from left ventricles of sham-operated
and portal vein-stenosed rats
|
|
Myofibril Ca2+ sensitivity.
To determine whether the depressed contractile function observed in
cardiac tissue from portal vein-stenosed rats was related to decreased
sensitivity of the myofilaments to Ca2+, the contractile
response of permeabilized left ventricular papillary muscle fiber
bundles to Ca2+ was examined. The effect of portal vein
stenosis is shown in Fig. 6. There was no
difference in myocardial Ca2+ sensitivity (half-maximal
pCa: 5.57 and 5.55 for sham and stenosed groups, respectively), maximum
tension development (520 ± 100 and 520 ± 180 µN for sham and
stenosed groups, respectively), or cooperativity as determined from
nH (2.2 ± 0.4 and 2.7 ± 0.4 for sham and stenosed
groups, respectively) in fibers between sham-operated and portal
vein-stenosed rats.
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 6.
Effect of portal vein stenosis on pCa-force relationship in skinned
left ventricular papillary muscle fiber bundles. Values represent means ± SE %maximum response obtained in tissues from sham-operated rats;
n = 5. Data from each group were fit to Hill equation and gave
a half-maximal pCa of 5.57 and 5.55 for fibers from sham-operated and
portal vein-stenosed rats, respectively. Maximal response was 520 ± 100 and 520 ± 180 µN in sham and stenosed rats, respectively. Hill
coefficients for sham and stenosed groups were 2.2 ± 0.4 and 2.7 ± 0.4, respectively.
|
|
Effect of portal vein stenosis on
Ca2+ uptake and release by SR.
The effect of portal vein stenosis on the uptake and release of
Ca2+ by the SR was investigated in left ventricular
papillary muscles in which the sarcolemma was chemically permeabilized.
The contraction produced by high concentrations of caffeine by
promoting Ca2+ release through the SR Ca2+
release channels was used to assess the SR Ca2+ content.
Ca2+ uptake and release were not different in the
sham-operated and portal vein-stenosed groups as evidenced by the
nearly identical responses to caffeine (Fig.
7B). Loading and subsequent release by caffeine was performed three times in each muscle with identical results, indicating that Ca2+ uptake by the SR under these
conditions was consistent and reproducible (not shown). Furthermore,
there was no difference between the sham-operated and portal
vein-stenosed groups in the rate of tension developed in response to
caffeine.
View larger version (8K):
[in this window]
[in a new window]
|
Fig. 7.
A: caffeine-induced Ca2+ release from sarcoplasmic
reticulum (SR) in permeabilized papillary muscle from a sham-operated
rat. After permeabilization and release of SR Ca2+ content
with caffeine, muscles were exposed to the following solutions:
a, relaxing solution for 2 min; b, Ca2+
loading (pCa 6.0) for 3 min; c, low EGTA for 1 min;
d, 50 mM caffeine; e, low EGTA; f,
caffeine; and g, relaxing solution. Rapid
upward deflection seen in tracing at arrows is an artifact from
replacement of bathing solution. B: compilation of the
effect of portal vein stenosis on caffeine-induced Ca2+
release determined as shown in A. Values are means ± SE of
response to 50 mM caffeine; n = 5-8.
|
|
SR Ca2+ content and
Na+/Ca2+
exchange in intact muscle fibers.
Figure 8 illustrates the effect of portal
vein stenosis on caffeine-induced Ca2+ release in
nonpermeabilized muscle. Figure 8A demonstrates that the
stimulation of Na+/Ca2+ exchange and the
resultant increase in Ca2+ efflux with
Ca2+-free solution with normal Na+ reduces the
SR Ca2+ content. Using this protocol, the effect of portal
vein stenosis on SR Ca2+ content and
Na+/Ca2+ exchange was investigated. Figure
8B presents a comparison between the sham-operated and portal
vein-stenosed groups in normal KH buffer solution (2.5 mM
Ca2+) and in Ca2+-free solution with normal
Na+, which reduces SR reuptake of Ca2+ leaked
during rest by stimulating Na+/Ca2+ exchange.
There was a decrease in the response to caffeine in both left
ventricular papillary muscles (not shown) and right ventricle (Fig.
8B), although the difference was significant only in the right
ventricle. As expected, stimulation of Na+/Ca2+
exchange for 5 min with Ca2+-free solution with normal
Na+ significantly decreased the response to caffeine in the
sham-operated and portal vein-stenosed groups. However, the magnitude
of the decrease was greater in the sham-operated group in absolute as well as relative terms (sham: 379.1 mN/cm2,
62%; stenosed: 155.3 mN/cm2, 39%).
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 8.
Effect of portal vein stenosis on caffeine-induced Ca2+
release from SR in nonpermeabilized right ventricles. SR
Ca2+ concentration was estimated by the magnitude of
caffeine-induced contraction. A: representative tracing in a
right ventricle from a sham-operated rat. B: effect of portal
vein stenosis on caffeine-induced Ca2+ release in intact
right ventricle from sham and stenosed rats in normal KH buffer
solution (2.5 mM Ca2+) and in Ca2+-free
solution with normal Na+ (Ca2+-free). Caffeine
was administered in normal buffer after 5 min of no stimulation. Before
switching to Ca2+-free solution with normal
Na+, muscle was stimulated with several electrical impulses
to determine that it was still viable (not shown in tracing). Bars are
means ± SE of force developed in response to addition of 30 mM
caffeine; n = 5-11.
|
|
Effect of BAY K 8644.
BAY K 8644, a dihydropyridine that prolongs the average open time of
L-type Ca2+ channels, increased the development of force as
a function of concentration between 10 nM and 1 µM. As demonstrated
by Fig. 9, BAY K 8644 produced comparable
increments in dT/dt at each concentration in tissues
from sham-operated and portal vein-stenosed animals.
View larger version (19K):
[in this window]
[in a new window]
|
Fig. 9.
Effect of BAY K 8644 on dT/dt of right ventricular
strips. Values are means ± SE of %change; n = 4-6. Data
were normalized to allow comparison of the effect of BAY K 8644 in
tissues with differing muscle stress before drug administration (see
Table 1). EC50 of BAY K 8644 was unaffected by portal vein
stenosis (sham, 0.16 ± 0.05 µM; stenosed, 0.22 ± 0.09 µM).
Similar results were obtained in left ventricle (data not shown).
|
|
| |
DISCUSSION |
Studies (7, 10, 13, 16, 29, 37, 40) in both human alcoholic and
nonalcoholic cirrhosis and in nonalcoholic animal models of cirrhosis
have shown that liver disease is associated with impaired basal cardiac
contractile function and diminished responses to -adrenergic
stimulation. In cirrhotic humans, this altered cardiac function has
been termed cirrhotic cardiomyopathy (43). In earlier
studies (7, 65), our laboratory has used the chronic portal
vein-stenosed rat, a hepatic model with extensive portosystemic
shunting without hepatocellular disease, to dissociate the cardiac
effects of portosystemic shunting and hepatocellular disease. The site
of cardiac impairment responsible for the decrease in basal contractile
function is largely unexplored in this model. In the present study, we
have extended our work to include the effect of chronic portal vein
stenosis on 1) sites in myocardial ECC coupling that could
reduce basal contractility as well as the contractile response to
-adrenoceptor activation and 2) postreceptor sites that
could alter responses to -adrenoceptor activation.
In the current study, basal myocardial contractility was reduced
30-50% in both the right and left ventricle of the chronic portal
vein-stenosed rat (Table 1), and -adrenoceptor responsiveness was
also diminished (Fig. 1), in agreement with previous reports from our
laboratory (7, 65). However, the positive lusitropic action of
isoproterenol was unaltered by portal vein stenosis (Fig. 2).
ECC.
The effect of portal vein stenosis on ECC was tested initially by
estimating sarcolemmal dihydropyridine receptor density as an
indication of L-type Ca2+ channel density. The decrease in
dihydropyridine binding sites observed in hearts from portal
vein-stenosed animals (Table 3) suggests that the density of L-type
Ca2+ channels was reduced. If the remaining channels still
gate the inward movement of Ca2+ normally, relative changes
in cardiac force in sham-operated and portal vein-stenosed rats would
be expected to be similar, and the maximum force developed in both
groups should be achieved at the same extracellular Ca2+
concentration. Muscles from the stenosed group, however, would not be
expected to generate the same absolute force as the sham group. A
between-group difference in the relative change in force development
would suggest the possibility of additional effects of portal vein
stenosis. This expectation is, of course, founded on the proviso that
the sensitivity of the myofilaments to Ca2+ and SR
Ca2+ uptake and release are unaffected by portal vein stenosis.
In actively contracting muscle, an increase or decrease in
extracellular Ca2+ concentration results in a parallel
increase or decrease in Ca2+ entry during the action
potential because of the change in its electromotive force (57). This
increase or decrease in Ca2+ influx is accompanied by
enhancement or attenuation of the intracellular Ca2+
transient (2) and increased or decreased force development (31). In the
present study, this relationship between extracellular Ca2+
concentration and force development held for myocardium from both
groups. However, at all extracellular Ca2+ concentrations,
the force developed by the stenosed group was significantly lower than
in the sham-operated group (Fig. 5A). Although this observation
could be the result of dysfunction in any of several steps in ECC, it
is consistent with decreased Ca2+ influx during muscle contraction.
Additional indirect support for the hypothesis that the cardiac effect
of portal vein stenosis results from decreased density of L-type
Ca2+ channels was obtained by determining the effects of
stenosis on the sensitivity of the myofilaments to Ca2+, SR
Ca2+ uptake and release, and the positive inotropic effect
of the dihydropyridine agonist BAY K 8644. Although a decrease in
myofilament Ca2+ sensitivity and/or SR Ca2+
transport would not necessarily negate the hypothesis, the absence of
an effect of portal vein stenosis on these aspects of ECC would support
the hypothesis. As Fig. 6 illustrates, no difference in myofilament
Ca2+ sensitivity was observed (half-maximal pCa was 5.57 and 5.55 in sham-operated and portal vein-stenosed rats, respectively), and maximum tension development occurred at the same Ca2+
concentration in the two groups. With respect to SR Ca2+
uptake and release, a lack of effect of portal vein stenosis on the SR
is suggested by the absence of an effect on caffeine-induced Ca2+ release in permeabilized fibers. In permeabilized
fibers, where the Ca2+ concentration can be rigidly
controlled and uptake into the SR limited to a fixed time interval,
there was no difference in the contraction caused by caffeine between
the sham-operated and stenosed groups (Fig. 7). Two conclusions about
the effect of portal vein stenosis on SR Ca2+ handling are
suggested by the absence of a difference between the two experimental
groups: 1) there is no direct effect of portal vein stenosis on
SR uptake and release; and 2) there is no change in the number
of release channels, because the magnitude of a caffeine contraction is
dependent not only on the amount of Ca2+ in the SR but also
on the number of release channels available to interact with caffeine.
The experiments performed with the L-type Ca2+ channel
agonist BAY K 8644 are also consistent with a decrease in
Ca2+ channel density contributing to the reduced
contractile function in portal vein-stenosed rats. BAY K 8644 increases
the transsarcolemmal Ca2+ influx by lengthening the mean
open time of L-type Ca2+ channels. Because the SR
Ca2+ and myofilament sensitivity to Ca2+ were
unaffected by portal vein stenosis, if the remaining L-type Ca2+ channels function normally, BAY K 8644 should increase
force by the same relative increments in tissues from both
sham-operated and stenosed animals because the only difference between
the two groups is the number of Ca2+ channels. As can be
seen in Fig. 9, BAY K 8644 produced similar relative increments in
myocardial force in sham-operated and portal vein-stenosed rats. It
should be noted that although these experiments on ECC suggest that
Ca2+ entry into the myocytes from portal vein-stenosed rats
is decreased, actual measurement of the current carried by the L-type
Ca2+ channel is necessary to confirm unequivocally the role
of the L-type Ca2+ channel in the myocardial action of
portal vein stenosis.
Aside from the decrease in dihydropyridine binding sites, the only
other difference between the portal vein-stenosed and sham-operated groups was a decrease in the response to caffeine in nonpermeabilized muscles in normal KH buffer (Fig. 8B). Although these data do not permit a conclusion to be made concerning the cause of the decrease, possible explanations include a reduced SR Ca2+
content, a change in Na+/Ca2+ exchange, and
greater leakage of Ca2+ from the SR during rest (diastole).
As mentioned above, the caffeine-induced SR Ca2+-release
data shown for skinned fibers (Fig. 7B) indicate that SR
Ca2+ uptake and release in the portal vein-stenosed group
is unaffected when loading conditions are controlled. These data are
inconsistent with a direct effect of portal vein stenosis on SR
Ca2+ content.
Because no effect of portal vein stenosis on SR Ca2+
content was observed in permeabilized muscle, the decreased response to caffeine in intact fibers would appear to be caused by a mechanism other than SR Ca2+ uptake and release. However, in
functioning, nonpermeabilized muscle, a change in the normal
relationship between SR uptake of cytosolic Ca2+ and
transport of Ca2+ out of the cell by
Na+/Ca2+ exchange during rest (diastole) could
affect the SR Ca2+ content. In the rat heart, nearly all of
the Ca2+ released from the SR is sequestered into the SR by
SR Ca2+-ATPase (5). Na+/Ca2+
exchange plays only a minor role in the removal of Ca2+
from the cytoplasm relative to the SR (5). A substantive increase or
decrease in Na+/Ca2+ exchange activity in
portal vein-stenosed rats, however, might have an inverse effect on the
SR Ca2+ content and the amount of Ca2+
available to induce subsequent contractions. To test this hypothesis, nonpermeabilized fibers were exposed to a Ca2+-free
solution with normal Na+, which stimulates
Na+/Ca2+ exchange, and under these conditions
at rest, a decrease in SR Ca2+ content sensitive to
caffeine is observed in the rat heart (Ref. 3; compare the response to
caffeine with and without Ca2+ in the sham and stenosed
groups in Fig. 8B). Therefore, Na+/Ca2+
exchange can be assessed using caffeine to determine the relative amount of Ca2+ remaining in the SR after stimulation of
Na+/Ca2+ exchange. If portal vein stenosis was
associated with either increased or decreased
Na+/Ca2+ exchange, a change in the response to
caffeine could be expected. In our study, the decrease in the response
to caffeine after Ca2+-free solution with normal
Na+ was considerably greater in sham-operated rats in
absolute as well as relative terms (sham: 379.1
mN/cm2, 62%; stenosed: 155.3
mN/cm2, 39%). The greater decrease in the
sham-operated rats (Fig. 8B) suggests that
Na+/Ca2+ exchange activity was greater in the
sham-operated group than in the portal vein-stenosed group. Because the
response to caffeine in the presence of extracellular Ca2+
was greater in sham-operated rats than in portal vein-stenosed rats, a
difference in absolute terms might be expected, but not in relative
terms if Na+/Ca2+ exchange was not affected.
One could speculate that an apparent decrease in
Na+/Ca2+ exchange would produce a greater
response to caffeine in Ca2+-free solution with normal
Na+ in the portal vein-stenosed group because the exchanger
is less effective, and therefore, a greater amount of Ca2+
can be taken up by the SR. However, the effect of caffeine in intact
fibers suggests that there is less, not more, Ca2+ in the
SR of the portal vein-stenosed group. A possible explanation for the
reduction in Na+/Ca2+ exchange may lie in the
relationship between Ca2+ entry through L-type
Ca2+ channels and extrusion by
Na+/Ca2+ exchange. It has been suggested that,
at equilibrium, the amount of Ca2+ that is removed by
Na+/Ca2+ exchange varies directly with the
amount of Ca2+ that enters via Ca2+ channel
current (ICa) (50, 62) and that the amount of
Ca2+ entering via ICa is balanced by an
equal amount of Ca2+ leaving the cell at
equilibrium. In that case, a chronically reduced
Ca2+ entry associated with downregulation of L-type
Ca2+ channels could result in decreased
Na+/Ca2+ exchange. Although this is
circumstantial evidence, it does fit the data. A decrease in
Na+/Ca2+ exchange expression would lend
credence to this hypothesis.
Alternatively, the reduced response of intact fibers to caffeine in
normal KH buffer solution in the portal vein-stenosed group could be
the result of a greater leakage of Ca2+ from the SR during
rest. In the rat, however, this is complicated by the fact that nearly
all Ca2+ leaked from the SR is taken back up into the SR
and a decline in the response to caffeine is not observed (5). However,
a decrease in the response to caffeine can be observed if
Na+/Ca2+ exchange is stimulated during the
period of rest (5). The data from permeabilized tissues suggest that
the SR from the stenosed group takes up the same amount of
Ca2+ as the sham group, at least under controlled loading
conditions (Fig. 7B). A larger Ca2+ leak after
portal vein stenosis would be expected to produce greater
Ca2+ efflux when Na+/Ca2+ exchange
is stimulated by Ca2+-free solution with normal
Na+, and therefore, a reduced response to caffeine.
However, our data showed smaller absolute and relative decrements in
the response to caffeine in the portal vein-stenosed group than in the
sham-operated group after a 5-min period in Ca2+-free
solution with normal Na+ (sham: 379.1
mN/cm2, 62%; stenosed: 155.3
mN/cm2, 39%) (Fig. 8B). This smaller
decrement suggests that Na+/Ca2+ exchange was
less in the portal vein-stenosed group and is not indicative of
increased leakage of Ca2+ from the SR during rest.
One question that can be asked is, How do the experiments performed in
permeabilized and nonpermeabilized quiescent fibers apply to the
observations of depressed contractile function in paced right
ventricles and left ventricular papillary muscles (Table 1 and Fig. 1)?
The experiments in permeabilized tissues had as their goal the
determination of whether the sensitivity of the myofilaments to
Ca2+ and SR Ca2+ uptake and release were
directly impaired by portal vein stenosis. The results from these
experiments suggest that these processes were not directly affected by
portal vein stenosis. If these processes are altered in intact
contracting muscle, the effect would have to be an indirect one that is
removed by permeabilizing the sarcolemma. The data from
caffeine-induced contraction in quiescent, intact muscle are probably
more indicative of the situation in contracting muscle. It has been
shown that the effect of a high concentration of caffeine in intact
muscle is similar to the effect of rapid administration of caffeine
during a contraction cycle in rat ventricular myocytes, i.e., there is
an increase in the Ca2+ transient, a decrease in SR
Ca2+ content, and increased extrusion of Ca2+
via Na+/Ca2+ exchange in response to the
increase in Ca2+ release (50, 62).
-Adrenoceptor responsiveness.
Although we had previously demonstrated that hearts from stenosed rats
require a threefold greater -adrenoceptor occupancy to produce the
same absolute increase in force in response to isoproterenol (65), no
alteration in myocardial -adrenoceptor density and affinity occurred
in hearts from portal vein-stenosed rats (44, 65). In the present
study, we investigated whether a change in G protein expression or
coupling of -adrenoceptors to Gs
could explain the
effect of portal vein stenosis on -adrenoceptor activation. As
demonstrated by Figs. 3 and 4, portal vein stenosis did not alter G
protein expression.
Failure of the coupling of the -adrenoceptor with Gs
could cause the effect of portal vein stenosis on
-adrenoceptor-mediated responses in the heart. It is well known that
the -adrenoceptor has two affinity states for a bound agonist (28)
and that only -adrenoceptors coupled to Gs are in the
high-affinity state. Fewer -adrenoceptors in the high-affinity state
would be expected to decrease the efficacy of GTP because there is less
agonist-receptor-G protein complex with which GTP can interact.
-Adrenoceptor-Gs coupling has been shown to be
significantly decreased in some models of cardiac hypertrophy and
failure, even though -adrenoceptor density, determined by antagonist
binding, was not reduced (63). This suggests that for any given number
of -adrenoceptors to which an agonist has bound, there will be a
reduced positive inotropic effect (63). Because we (65) had previously
found that hearts from portal vein-stenosed rats required a greater
-adrenoceptor fractional receptor occupancy to produce the same
positive inotropic response to isoproterenol as sham-operated rats, the
effect of portal vein stenosis on the high- and low-affinity state of
the -adrenoceptor was investigated. The unchanged fraction of
receptors in the high-affinity state (Table 2) clearly shows that
portal vein stenosis was without effect on -adrenoceptor-G protein
coupling. Therefore, -adrenoceptor signaling does not appear to be
influenced by portal vein stenosis. Indeed, one could argue
that the lack of a relative change in the response to -adrenoceptor
activation in the presence of a decrease in basal cardiac force
generation is indicative of no effect on -adrenoceptor
responsiveness at all. If this is so, then another process must be
responsible. Although the process of ECC has not been studied in
animals subjected to portal vein stenosis until now, there is evidence
from several models of hypertrophy and heart failure in different
species, as well as in human congestive heart failure, demonstrating,
even in mild pathological states where basal L-type Ca2+
channel peak current density or dihydropyridine binding was unaltered, reduced -adrenoceptor responsiveness that coincides with a decrease in the -adrenoceptor-mediated enhancement of L-type Ca2+
channel current density (30, 47, 48, 56).
In summary, myocardial contractility was depressed by the induction of
portal hypertension by chronic ligation of the prehepatic portal vein.
Various aspects of -adrenoceptor signaling and ECC were examined in
rat right ventricular strips and left ventricular papillary muscles. In
permeabilized muscle, myofilament Ca2+ sensitivity and SR
Ca2+ uptake and release did not differ between the sham and
the stenosed groups. The number of dihydropyridine binding sites was
reduced, and in intact muscles, Na+/Ca2+
exchange and the SR Ca2+ content were reduced in the portal
vein-stenosed group. No effect of portal vein stenosis on
-adrenoceptor-G protein coupling and G protein expression or
function was observed. The data suggest that in the rat heart, portal
vein stenosis-induced myocardial dysfunction is associated with
alterations in ECC but not -adrenoceptor signaling.
We thank Stephanie R. Edwards for technical assistance and Dr.
Robert E. Godt for providing the program to prepare the solutions for
the permeabilized fiber experiments.
This study was supported by grants from the American Heart Association,
Louisiana Affiliate, the Stiles Foundation, National Institute of
Mental Health Grants MH-01231 and MH-40694 (J. M. O'Donnell) and
National Institute of Drug Abuse Grant DA-07972 (S. C. Roerig).
The costs of publication of this
article were defrayed in part by the
payment of page charges. The article
must therefore be hereby marked
"advertisement"
in accordance with 18 U.S.C. §1734 solely to indicate this fact.
Address for reprint requests and other correspondence: J. H. Zavecz,
Dept. of Pharmacology, Louisiana State Univ. Health Sciences Center, PO
Box 33932, Shreveport, LA 71130-3932 (E-mail: jzavec{at}lsumc.edu.
TOP
ABSTRACT
INTRODUCTION
