Vol. 284, Issue 3, H867-H875, March 2003
Role of serotonin in thromboxane A2-induced
coronary chemoreflex
M. J.
Wacker,
H. L.
Wilhelm,
S. E.
Gomez,
E.
Floor, and
J. A.
Orr
Department of Molecular Biosciences, University of Kansas,
Lawrence, Kansas 66045
 |
ABSTRACT |
We reported previously that the
thromboxane A2 (TxA2) mimetic U-46619
stimulates cardiac vagal afferent nerves, eliciting a reflex decrease
in heart rate (HR) and arterial blood pressure (ABP). The present
experiments were designed to test the hypothesis that TxA2
evokes these changes via the release of serotonin
[5-hydroxytryptamine (5-HT)] and activation of the 5-HT3
receptor. Injections of the 5-HT3 antagonist tropisetron (1 mg of 3-tropanyl-indole-3-carboxylate or ICS-205-930) attenuated the
decreases in HR and ABP induced by left atrial injections of U-46619
(20 µg). Tropisetron administration also eliminated the
U-46619-induced increase in impulse frequency in a majority of cardiac,
vagal afferent units tested. Measurement of serum 5-HT levels revealed
an elevation in serum 5-HT levels after U-46619 injection in those
rabbits that displayed a significant HR change following injection of
U-46619. These results indicate that although other factors may also
contribute to these reflex responses, the release of 5-HT and
stimulation of the 5-HT3 receptor plays a significant role
in coronary reflexes induced by TxA2.
U-46619; tropisetron; pheylbiguanide; vagal afferent units; anesthetized rabbit
| |
INTRODUCTION |
WE HAVE PREVIOUSLY
REPORTED that left atrial injections of a thromboxane
A2 (TxA2) mimetic, U-46619, decreases heart
rate (HR) and arterial blood pressure (ABP) in the anesthetized rabbit (45). These reflex responses were mediated by stimulation
of cardiac, vagal afferent nerves. U-46619 stimulation of these
afferent nerves may be mediated directly by the action of
TxA2 on cardiac, vagal afferent nerves or indirectly via
the release of other substances.
It is well established that TxA2 receptors are present on
platelets, and stimulation of these receptors by TxA2
receptor agonists activates platelets leading to the release of
serotonin [5-hydroxytryptamine (5-HT)] (2, 20, 23, 28, 34,
42). Because 5-HT has been shown to stimulate the vagus nerve
and evoke the coronary chemoreflex via stimulation of 5-HT3
receptors (1, 10, 11, 13, 26, 33), there is a strong
rationale to suspect that 5-HT release and stimulation of the
5-HT3 receptor mediates at least part of the
TxA2 responses that we have previously reported. We
therefore chose to investigate the hypothesis that injection of U-46619
stimulates these reflex effects via the release of 5-HT and the
subsequent actions of 5-HT on cardiac, vagal afferent neurons.
| |
METHODS |
Animal preparation.
For all experiments, male New Zealand White rabbits (mean weight = 4 kg) were initially tranquilized with an intramuscular injection of
xylazine (Rompun, 2.5 mg/kg) and then anesthetized with an
intramuscular injection of ketamine (35 mg/kg). Catheters were inserted
into the right femoral vein and artery. Thirty to fifty minutes after
the ketamine injection, 3-4 ml of a solution of 2% 
-chloralose
(~15 mg/kg) and 10% urethane (~75 mg/kg) dissolved in a mixture of
2% borax and 98% water were given intravenously. Two milliliters of
the -chloralose-urethane solution were infused at ~40-min
intervals to maintain anesthesia. ABP was continually monitored via the
arterial catheter. The trachea was exposed, a plastic tube was
inserted, and the rabbit was ventilated (volume 25 ml; rate 30 breaths/min). End-tidal CO2 levels were monitored to ensure
proper ventilation.
Four series of experiments were carried out in a total of 59 animals.
The initial preparation of the animals was the same for all series of
experiments. An incision was made between the third and fourth ribs on
the left side of the thorax, and the ribs were spread to expose the
left surface of the heart. Cotton gauze, moistened with 0.9% saline,
was used to move the upper lobe of the left lung away from the heart.
The pericardial sac was opened around an area on the left ventricle and
the heart exposed. A catheter (polyethylene-90) was inserted into the
left atrium and secured with sutures that were tied around the catheter and through the walls of the left atrium. This catheter was then used
to administer drugs into the left atrium.
For series 3, in addition to the initial animal preparation
described above, the left cervical vagus nerve was isolated. The nerve
was laid across a small dissecting platform, and the area was immersed
in mineral oil to prevent drying of the nerve. The nerve sheath of the
vagus was removed, and small slips of the vagus nerve were dissected
and placed on bipolar recording electrodes. Recording electrodes were
connected to a high-impedence probe (model RPS 107B, Grass Instruments;
Quincy, MA), and the signal was amplified (Grass P511 amplifier). These
signals were displayed on an oscilliscope (Tektronix RM561A;
Beavertown, OR) with low and high filters set at 10 Hz and 1 kHz,
respectively, and input was fed into the PowerLab Chart program with
the Spike Sorter extension (ADInstruments), which was installed on a G4
Macintosh for recording, analysis, and discrimination of action potentials.
For series 4, an additional catheter was inserted into the
left femoral artery and advanced 3-4 inches. Just before the
U-46619 injection and at selected times after injection of U-46619,
blood was sampled from the arterial catheter. For several reasons, we chose to measure 5-HT levels in blood sampled from the femoral artery.
First, 5-HT concentrations in the femoral artery should reflect 5-HT
concentration in blood destined for the coronary circulation. Second,
the femoral artery is much more accessible than either the coronary
artery or coronary sinus, thereby simplifying surgical procedures and
reducing trauma to the animal.
All experimental protocols and procedures involving the use of animals
in this investigation were reviewed and approved by the Institutional
Animal Care and Use Committee (IACUC no. 42-02).
Drug preparation.
TxA2 degrades to the inactive metabolite TxB2
under physiological conditions (half-life ~30 s). Therefore, the
stable TxA2 mimetic U-46619 (Caymon Chemical) was used to
stimulate TxA2 receptors (7). A stock solution
was made by dissolving 10 mg of U-46619 in 1 ml of 100% ethanol. A
working stock solution of 100 µg/ml was made by removing 0.1 ml from
the stock solution and adding 9.9 ml of 0.9% saline. The
5-HT3 receptor agonist phenylbiguanide (PBG) (Sigma) was
prepared by making a concentration of 1 mg/ml deionized water. 5-HT was
purchased as 5-HT creatine sulfate complex (Sigma) and prepared at a
concentration of 100 µg/ml in 5% HCl-95% deionized water.
Injections into the left atrium were made via the left atrial catheter
and a syringe filled with the appropriate amount of drug solution and
then diluted with 0.1 ml of saline. The drug was infused in 2-3 s
and was followed by a 1-ml injection of saline. The 5-HT3
receptor antagonist tropisetron (3-tropanyl-indole-3-carboxylate or
ICS-205-930; RBI) was prepared at a concentration of 2.5 mg/ml deionized water.
Protocol.
In the first series of experiments, designated as series 1a
and series 1b, the vehicle for U-46619, U-46619 (20 µg),
and PBG (400 µg) was injected into the left atrium, and HR and ABP
were recorded. ABP and HR were allowed to return to baseline levels before the subsequent drug was administered (~5-10 min). In
series 1a, PBG was given followed by U-46619 and then 1 mg
of the 5-HT3 receptor antagonist tropisetron was injected
into the left atrium. After injection of the 5-HT3
antagonist, 15 min elapsed before the injections of PBG and U-46619
were repeated. In a different group of rabbits (series 1b),
the effect of the vehicle for tropisetron was tested by administering
PBG and U-46619 before the vehicle and again 15 min following injection
of the vehicle.
In series 2, HR and ABP were recorded following a protocol
that was similar to series 1a except that 5-HT was injected
instead of U-46619. PBG (400 µg) was injected and followed by 5-HT
(100 µg). These injections were repeated 15 min following
administration of tropisetron (1 mg).
In series 3, slips of nerve fibers were teased from the
cervical vagus nerve and tested for identification of cardiac
chemosensitive units. Each fiber was cut, and the peripheral end of the
nerve was laid on recording electrodes to measure only afferent
signals. Afferent units were chosen based on their pattern of
discharge. Typically, nerve units that respond to chemical stimuli have
a low random frequency of firing under baseline conditions (8, 9,
19, 27). The origin of the afferent unit was determined by
lightly probing the heart with a cotton tip (8, 9, 19, 27). PBG (20-30 µg) was then injected into the left
atrium to verify the chemosensitive nature of the unit. If an afferent
unit responded to both probing of the heart and injection of PBG, then the response of the unit to U-46619 (5-10 µg) was tested.
Injections of PBG and U-46619 were repeated before injection of
tropisetron to verify the reproducibility of responses from repeated
injections. Five minutes were allowed between each drug injection.
Tropisetron (1 mg) was then injected into the left atrium, and 15 min
were allowed before subsequent injection of PBG and U-46619. Low doses of U-46619 and PBG were selected for these experiments to reduce nerve
sensitization and to reduce dramatic ABP changes. The vehicle for
U-46619 was administered after U-46619 and PBG injections to ensure
that neither the volume of fluid injected nor the solvent for the drug
stimulated the receptor unit.
In series 4, blood was sampled 1 min before injection of
U-46619 blood and then sampled again between 15 and 45 s after
injection of U-46619. Blood was also sampled before and after injection of the vehicle for U-46619 during the same time course in a separate group of rabbits. The following protocol was modified from Wright and
Angus (46). Blood (2 ml) was slowly withdrawn with a 10-ml syringe and then emptied into a test tube containing 25 mg ascorbic acid (Sigma). Blood was allowed to clot for 45 min and then centrifuged at 5,000 rpm (3,000 g). The serum was then removed and
frozen at 
20°C for storage until measurement with HPLC.
Measurements and data analysis.
The arterial catheter was connected to a pressure transducer to monitor
systemic ABP. All data were collected with a commercial software
package (Powerlab, ADInstruments). HR was measured from a tachometer
trace of ABP. Baseline or preinjection measurements of HR and ABP were
taken over a 10-s time period before injection and then compared with
postinjection values taken during the period of greatest change
following administration of the test drug. To examine the mechanism of
the U-46619-induced decrease in HR, only animals that responded to
U-46619 with a decrease in HR >5% were included in the data
collection. Data from individual experiments were averaged and the
means ± SE calculated. For ease of comparison between pre- and
postinjection values, selected data are reported as the percent change
from the baseline levels. A paired t-test was used to
determine whether there were significant differences in the HR and ABP
elicited by U-46619, PBG, and the vehicle. For determination of
significant differences among three means (e.g., baseline ABP,
hypotension, and hypertension), a two-factor ANOVA was performed
followed by a post hoc test (least significant difference) if
significant F-values were obtained. For all statistical
tests, significant differences were accepted at the P < 0.05 value.
For nerve recordings, preinjection values were averaged for a 5-s
period of maximal activity from the 30-s period just before drug
injection. Postinjection values were averaged for a 5-s period of
maximal activity. When more than one unit was active, individual units
were counted with the aid of a window discriminator and spike sorter
program (PowerLab), which discriminated different units based on both
amplitude and width. Multifiber recordings were not used if different
units were of similar amplitude or if there were more than three to
four active units. Data from individual experiments were averaged, the
means ± SE were calculated, a paired t-test was
conducted, and a significance level of P < 0.05 was
chosen to establish significant differences.
Measurements of serum 5-HT levels were made by using HPLC analysis with
electrochemical detection. The analysis was performed by using a 100 × 3.2-mm phase II ODS column (BioAnalytical Systems; West Lafayette, IN)
in 10% acetonitrile-90% 0.15 M chloroacetate buffer, pH 3.0, containing 0.7 mM EDTA and 0.86 mM octane sulfonate (32)
with electrochemical detection at 0.6 V, using an LC-4C Amperometric
Detector (BioAnalytical Systems). The 5-HT concentration (in pmol/µl)
was calculated with reference to HPLC injections of 10 and 20 pmol of
5-HT standards.
| |
RESULTS |
Series 1.
Figure 1 presents the ABP and HR response
following injection of PBG and U-46619 into an animal before
tropisetron (A and C, respectively) and after
treatment with tropisetron (1.0 mg iv) (B and D,
respectively). Typically, left atrial injections of PBG elicited a
decrease in ABP and HR with an average onset equal to 6 ± 1 s following the injection. U-46619 injections usually elicited a
decrease in ABP followed by an increase in ABP. The decrease in HR
usually occurred during the decrease in ABP. The average time of onset
of this bradycardia was 9 s. However, some animals also displayed
another more latent decrease in HR, which occurred during the elevated
ABP. Because this HR change was likely due to the baroreceptor reflex,
only the decrease in HR that occurred before the increase in ABP and
during the decrease in ABP was measured. As shown in Fig. 1, the HR
change induced by U-46619 was eliminated following injection of
tropisetron, whereas the blood pressure change was attenuated but not
eliminated. The average changes in HR and ABP after injection of PBG
and U-46619 in series 1a are shown in Fig.
2 (n = 8). In previous
studies, it has been documented that HR does not decrease in all
rabbits following injections of U-46619 (45). Because the
goal of this study was to determine the mechanism of the
U-46619-induced decrease in HR, only those rabbits that exhibited a HR
change >5% (i.e., "responders") were included in the analysis of
data. In series 1a and 1b using U-46619, 15 of 21 rabbits exhibited a HR decrease >5%. PBG was also tested in the six
"nonresponders." The average decrease in HR following injection of
PBG was 8 ± 1% in the nonresponders compared with an average
change of 17 ± 3% in the responders. Neither HR nor ABP changed
significantly (P > 0.05) following injection of the
vehicle for U-46619.
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 1.
Arterial blood pressure (ABP) and heart rate (HR)
response to left atrial injections of 400 µg phenylbiguanide (PBG,
A) and 20 µg U-46619 (U4, C). ABP and HR
responses to PBG and U-46619 were recorded again 15 min after injection
of tropisetron (1 mg) (B and D, respectively).
BPM, beats per minute.
|
|
View larger version (20K):
[in this window]
[in a new window]
|
Fig. 2.
Average changes in HR (A) and mean ABP (MABP)
(B) induced by left atrial injections of 400 µg PBG and 20 µg U-46619 before and after tropisetron (Trop, n = 8). Note that PBG usually only caused a decrease in ABP, whereas
U-46619 elicited a decrease in ABP followed by an increase in ABP.
* Significant difference at P < 0.05 comparing
baseline and postinjection values; +significant difference at
P < 0.05 comparing hypotension and hypertension
values.
|
|
To verify that repeated injections of PBG and U-46619 would yield
consistent changes in HR and ABP, injections of PBG and U-46619 were
administered to a series of animals before and after the tropisetron
vehicle (n = 7). The results for series
1b are displayed in Table 1.
Series 2.
In this series, the HR and ABP responses to 5-HT and PBG were tested.
Similar to U-46619, 5-HT elicited a decrease in ABP followed by an
increase in ABP. There was also a decrease in HR that occurred during
the decrease in ABP with an average time of onset of 5 ± 1 s. The average time of onset for PBG was 6 ± 1 s. The
average changes in HR and ABP elicited by PBG and 5-HT before and after
injection of tropisetron are shown in Fig.
3 (n = 9). Nine of eleven
rabbits responded to 5-HT with a decrease in HR >5%.
View larger version (20K):
[in this window]
[in a new window]
|
Fig. 3.
Average changes in HR (A) and MABP
(B) induced by left atrial injections of 400 µg PBG and
100 µg 5-hydroxytryptamine (5-HT) before and after injection of
tropisetron (n = 9). Note that B shows that
PBG usually caused a decrease in ABP, whereas 5-HT elicited a decrease
in ABP followed by an increase in ABP. * Significant difference at
P < 0.05 comparing baseline and postinjection values;
+significant difference at P < 0.05 comparing
hypotension and hypertension values.
|
|
Series 3.
Afferent nerve recordings were made in a group of 12 cardiac vagal
units. We were able to test four units from our previous study
(45) as well as eight new units. Figure
4 displays a multiunit recording from the
previous study as well as a new unit. Both nerve recordings in Fig. 4
display multiunit activity, but the unit with the large amplitude
action potential in each recording was easily discriminated from other
action potentials and analyzed. Both units responded to probing of the
left ventricle and to injections of both PBG and U-46619 (Fig. 4,
A and C, respectively). The units did not show an
increase in impulse frequency to either PBG or U-46619 after injection
of tropisetron (Fig. 4, B and D, respectively).
View larger version (30K):
[in this window]
[in a new window]
|
Fig. 4.
Multiunit recording from cardiac vagal afferent units from two
rabbits. Large amplitude action potentials in both units 1 and 2 are units that were located within the left ventricle
of the heart. In both units, A and C represent
the increase in impulse frequency after left atrial injection of 30 µg PBG and 5 µg U-46619, respectively. B and
D represent the impulse activity to PBG and U-46619,
respectively, after injection of tropisetron. Baseline response of
unit 1 to U-46619 before tropisetron was reported previously
(45).
|
|
The average changes in impulse frequency in all the units tested (Fig.
5) support the responses that are shown
for individual animals. Injections of PBG and U-46619 were repeated
before injection of the 5-HT3 antagonist to verify that
repeated injections of PBG and U-46619 would yield consistent results.
The average time of onset for the increase in activity was 6 ± 1 s for PBG and 9 ± 1 s for U-46619. After tropisetron
injection, there was no statistically significant increase in average
afferent unit impulse frequency to PBG and U-46619. Eight of twelve
afferent units displayed no increase in impulse frequency after U-46619
injection in the presence of tropisetron; however, four units still
elicited an increase in impulse frequency. Three of these four units
displayed a percent increase in impulse frequency to U-46619 injection
equal to that before tropisetron injection.
View larger version (15K):
[in this window]
[in a new window]
|
Fig. 5.
Average changes in impulse frequency of 12 cardiac, vagal
afferent units. PBG (20-30 µg) and U-46619 (5-10 µg) were
injected twice before injection of tropisetron to verify that repeated
injections would yield similar increases in impulse frequency.
* Significant difference at P < 0.05.
|
|
Series 4.
Serum 5-HT levels before and after injection of saline or U-46619 are
shown in Fig. 6. Rabbits were grouped
based on treatment and the presence or absence of a response to the
treatment. A total of 15 animals were used with the rabbits either
being given injections of the U-46619 vehicle (n = 4)
or U-46619 (n = 11). Rabbits given U-46619 were grouped
according to their HR response to the drug (i.e., responders vs.
nonresponders). The responders (n = 6) displayed a
significant and consistent increase in 5-HT levels, whereas the
nonresponders (n = 5) elicited variable changes in 5-HT
levels that were not statistically significant (P > 0.05). The average HR for the responders was 242 ± 23 beats/min
before U-46619 injection and 214 ± 20 beats/min after injection.
The average HR for the nonresponders was 225 ± 15 beats/min
before U-46619 and 219 ± 15 beats/min after U-46619.
View larger version (13K):
[in this window]
[in a new window]
|
Fig. 6.
Average serum levels of 5-HT in rabbits following
injection of vehicle (n = 4) or U-46619 (20 µg).
Responders (Resp, n = 6) are those animals that
displayed a decrease in HR to U-46619 that was >5% from baseline
measurements, whereas nonresponders (Nonresp, n = 5)
did not show a decrease >5%. * Significant difference at
P < 0.05.
|
|
| |
DISCUSSION |
Coronary chemoreflex.
Previous work (45) from our lab has shown that left atrial
injections of the TxA2 mimetic U-46619 stimulate cardiac
afferent units from the vagus nerve and elicit a decrease in HR and
ABP. The purpose of this study was to determine the role of the
5-HT3 receptor in the observed U-46619-induced responses.
We hypothesized that injections of U-46619 would induce the release of
5-HT, which then stimulates the 5-HT3 receptor and mediates
the neural and cardiovascular responses elicited by the
TxA2 mimetic.
Similar to our previous work, left atrial injections of U-46619 led to
a decrease in HR and a biphasic change in ABP. After tropisetron
administration, the decrease in HR was significantly reduced, whereas
the decrease in ABP was attenuated. The more latent increase in ABP
that is induced by U-46619 is likely due to systemic vasoconstriction.
This increase in ABP persisted following 5-HT3 receptor
blockade. Because it could be argued that repeated injections of the
TxA2 mimetic U-46619 could induce tachyphylaxis resulting
in a decrease in the HR and ABP response, repeated U-46619 injections
were carried out. We observed that the decreases in HR and ABP were the
same following repeated injections of the TxA2 mimetic. We
conclude from these data that a significant portion of the
U-46619-induced coronary chemoreflex that we have observed in these
rabbits is mediated by the stimulation of the 5-HT3 receptor.
We also compared U-46619 responses to those observed following left
atrial injection of 5-HT. The results with 5-HT mimic the responses to
U-46619, further supporting our hypothesis that TxA2 may
induce reflex changes via the release of 5-HT. Injections of 5-HT
elicited a decrease in HR and a biphasic change in ABP, and these
average decreases in HR and ABP were reduced after tropisetron administration. The increase in ABP was not reduced but rather augmented after tropisetron administration. The increase in ABP is
hypothesized to be due to systemic vasoconstriction either elicited
directly by 5-HT or via the release of TxA2 or another vasoactive agent.
It is interesting to note that in these reflex studies there was a
variation in the degree of HR change in these rabbits, similar to our
previous studies (45). In the current report (series
1a and 1b), 6 of 21 rabbits did not display a decrease in HR >5% in response to injection of the TxA2 mimetic.
In comparison, these same six nonresponder rabbits also did not respond
with as large an average decrease in HR to injections of PBG as the responders. We also observed that 2 of 11 rabbits did not evoke a
decrease in HR >5% to injection of 5-HT. These nonresponders may not
show a strong response to these agents for a variety of reasons,
including a lower number of 5-HT or TxA2 receptors,
differential dampening of the neural response in some animals due to
the anesthetic drugs, or variability of the vagal response among
animals. There is also a strong possibility that in some animals
U-46619 and 5-HT injections did not evoke a large release of 5-HT from
platelets or other sources, and therefore the response was dampened.
The presence of nonresponder rabbits has been reported by other groups.
Buzzard et al. (5) have shown that 25% of pulmonary or
aortic segments in rabbits did not contract in response to TxA2 mimetics. They found that the nonresponders had a
significant decrease in the number of vascular TxA2
receptors. However, in these nonresponders there was no decrease in
TxA2 receptor binding sites in platelets and no change in
platelet aggregation to U-46619. In a study more similar to ours,
Wright and Angus (46) observed that injections of the
5-HT1-like agonist, 5-carboxamidotryptamine, in the
rabbit elicited the Bezold-Jarish-like reflex, and this effect was
blocked with a 5-HT3 receptor antagonist. They concluded that 5-carboxamidotryptamine elicited the release of 5-HT from platelets and subsequently evoked the reflex via stimulation of 5-HT3 receptors. They also observed a longer delay in the
Bezold-Jarish-like response to 5-carboxamidotryptamine compared with
the response observed to injections of 5-HT or PBG, which is similar to
our findings with U-46619. Similarly, they observed that 30% of the rabbits studied did not elicit a reflex response to
5-carboxamidotryptamine. They attributed these observations to the fact
that 5-carboxamidotryptamine did not elicit a strong enough release
of 5-HT. These results are similar to our findings in that the
responders had a significant increase in serum 5-HT levels whereas the
nonresponders did not. When the findings of Buzzard et al.
(5) and Wright and Angus (46) are compared
with our results, it is possible that the TxA2 receptor
density in platelets in nonresponders are similar to responders, but
the 5-HT levels contained in platelets is either lower or the signal
transduction mechanism for release of 5-HT is less effective.
Afferent nerve fiber recordings.
To further investigate the mechanism for the tropisetron-induced
reduction in the coronary chemoreflex, recordings were made from
cardiac afferent units within the vagus nerve. These units were
determined to originate from either the left atrium or left ventricle
based on increases in impulse frequency in response to mechanical
probing of the heart (8, 9, 19, 27). All of these units
responded to PBG. Before administration of tropisetron, PBG and U-46619
were given twice to ensure that repeated stimulation of an afferent
unit would cause reproducible increases in impulse frequency (Fig. 5).
Of the 12 nerve fibers tested, all responded with an increase in
impulse frequency to U-46619, although the magnitude of the response
varied. After tropisetron administration in 8 of 12 fibers, there was
no increase in impulse frequency in response to U-46619. Three units
still displayed a percent increase in frequency similar to that before
tropisetron administration. It is interesting to note that these three
fibers did not respond as strongly as other fibers to PBG before
tropisetron. It is possible that these three units were not as
sensitive to 5-HT and were more responsive to other stimuli, such as
direct stimulation by TxA2 or stimulation due to the
release of some other substance. It is not likely that these units were
stimulated by mechanical changes in the heart, because the onset of the
response occurred before the increase in ABP and with the same time
course as the other fibers.
Tropisetron, the antagonist utilized in our study, has been widely used
as a 5-HT3 receptor-blocking agent. The dose of tropisetron that was used in this study has been used in similar protocols to block
the 5-HT3 receptor (15, 16, 30). It has also
been reported that tropisetron is a weak antagonist for the
5-HT4 receptor (14, 40). Therefore, it is
possible that some of the actions of tropisetron in our study may
result from blockade of the 5-HT4 receptor. However,
Freeman et al. (14) report that the affinity of
tropisetron for the 5-HT3 receptor is ~1,000:1 over other
receptors. Furthermore, the actions of 5-HT in eliciting the coronary
chemoreflex would most likely be due to stimulation of the
5-HT3 receptor. It has been well documented that
5-HT3 receptor antagonists inhibit the depolarizing actions
of 5-HT on isolated vagus nerve and nodose preparations (1, 11,
26) and eliminate the 5-HT-evoked coronary chemoreflex
(10, 13, 33). The presence of 5-HT3 receptors
on sensory neurons has been established (21, 38), and
radioactively labeled tropisetron has been used to localize 5-HT3 receptors in both rabbit vagal and nodose
preparations (25). Therefore, we hypothesize that the
actions of tropisetron in this study are due to 5-HT3
receptor antagonism.
PBG was used as a control in all of these studies because it is an
accepted 5-HT3 receptor agonist, is documented to stimulate cardiac nerves, and elicits decreases in HR and ABP (3,
12). As shown in the present study, PBG elicited a decrease in
HR and ABP in all the series tested, and these cardiovascular changes were effectively eliminated after administration of tropisetron. Therefore, we are confident that we effectively blocked the
5-HT3 receptor during these studies.
5-HT release.
Even though previous authors have reported that TxA2 can
elicit the release of 5-HT from platelets (20, 28, 42), we chose to verify that there was a rise in 5-HT in our preparations using
the chosen dose of U-46619. Comparing blood samples taken from the
femoral artery before and after U-46619 injection, the responders
displayed a significant elevation of 5-HT in the serum, whereas the
average 5-HT concentration did not significantly change in the
nonresponders. The concentrations of serum 5-HT that we measured were
consistent with reports from other authors (29, 37, 46).
Rabbits contain a larger quantity of 5-HT in their platelets than other
species including humans (37), and therefore it is
possible that the release of 5-HT from sources such as platelets in
rabbits is much larger than other species, and therefore the effects of
5-HT could mask other direct or indirect actions of TxA2.
In the rat, Sun et al. (41) have shown that
TxA2 applied to the epicardial surface (where there is less
likely to be 5-HT release by platelets) stimulates cardiac vagal
afferent nerves in the anesthetized rat to a level comparable to
prostaglandin (PG)I2 and PGE2.
Other factors contributing to cardiovascular responses.
We conclude that U-46619 induces a significant portion of
these cardiovascular and neural changes via the 5-HT3
receptor in the anesthetized rabbit. However, after 5-HT3
receptor blockade, because there was a residual decrease in HR and ABP
and an increase in impulse frequency in a small group of units tested,
it is possible that other factors may also play a role in the
U-46619-induced coronary chemoreflex.
Even though it is generally accepted that the coronary chemoreflex
evoked by 5-HT is due to stimulation of the 5-HT3 receptor, it is possible that other 5-HT receptors may also be involved. Specifically, in the isolated rat vagus nerve it has been shown that
there is a residual depolarization induced by 5-HT after 5-HT3 receptor antagonism by granisetron or odansetron,
which is then eliminated following blockade of the 5-HT4
receptor (4, 6, 39). Therefore, the 5-HT4
receptor may play a minor role in the 5-HT-induced responses. Further
work will have to be conducted to more fully investigate the role of
the 5-HT4 receptor in these responses. It has also been
suggested that the 5-HT2 receptor may play a role in 5-HT
responses (47). However, other research has demonstrated a
lack of a role of 5-HT2 receptors in stimulation of the
vagus nerve by 5-HT (4, 39, 46). Previous work from our
lab supported no involvement of the 5-HT2 receptor in
U-46619-evoked reflexes because the 5-HT2 receptor
antagonist ketanserin did not block reflex responses evoked by
TxA2 in the anesthetized cat (36).
It is also possible that U-46619-induced depolarization of the vagus
nerve is mediated or possibly modulated by the TxA2
receptor. It has been shown that members of the prostaglandin family
PGI2, PGE2, and TxA2 induce
increases in firing of sensory neurons (22, 41, 45), and
the receptors for PGI2 and PGE2 (IP and EP,
respectively) have been localized to sensory neurons (31).
Although the TxA2 receptor (TP) has yet to be localized to
sensory neurons, preliminary work from our lab has shown that TP mRNA
is present in the cell bodies of cultured sensory neurons
(43).
The possibility for direct (nonneuronal) actions as well as other
factors cannot be ruled out. It is not likely that U-46619, 5-HT, or
other released substances would elicit a direct effect on the heart
affecting HR because previous work showed that the HR decrease was
eliminated after vagotomy (45). However, the decrease in
ABP was not completely eliminated after vagotomy, and it is possible
that U-46619 or 5-HT evoked the release of other mediators from
platelets or elsewhere and elicited vasodilation. It is also possible
that pulmonary vasoconstriction induced by U-46619 or 5-HT would alter
the cardiac output and transiently lower ABP. Further work will be
needed to investigate these other factors; however, from our studies in
the anesthetized rabbit, it can be concluded that the 5-HT3
receptor does play a significant role in the U-46619 coronary chemoreflex.
Significance.
It is concluded from these data that U-46619 can induce changes in
afferent unit activity, HR, and ABP via release of 5-HT and subsequent
stimulation of the 5-HT3 receptor. It has been demonstrated
that TxA2 elicits platelet aggregation and can induce the
release of 5-HT from platelets (2, 20, 23, 28, 34, 42).
Therefore, the mechanism of this stimulation is likely the release of
5-HT from activated platelets. This may provide another significant
role for TxA2 in that release of TxA2 during tissue trauma may trigger a cascade of events involving platelets and
5-HT that induce significant cardiovascular changes.
The role of platelets and 5-HT in stimulating afferent nerves during
myocardial ischemia has been documented. Fu and Longhurst (17) have shown that platelet-rich plasma plus collagen
(activated platelets) can stimulate cardiac afferents, and decreasing
the circulating platelet number during myocardial ischemia can
attenuate the ischemia-induced increase in impulse activity in
cardiac spinal afferents. They have found that they can inhibit this
platelet-induced response with tropisetron (30).
Likewise, thromboxane may also be important in stimulating nerves
during ischemia. It has been shown that TxA2 is
released during myocardial ischemia (24, 35) and
that activation of cardiac vagal afferent nerves by left anterior
descending artery occlusion in the rat was attenuated by pretreatment
with indomethacin (an inhibitor of prostaglandin and TxA2
formation) (44). Recent work has now shown that
TxA2 also stimulates ischemically sensitive cardiac
afferent nerves that travel through the spinal cord in the anesthetized
cat and may activate these afferents during ischemia (18).
It is possible that the actions of TxA2, 5-HT, and
platelets may all be intertwined during myocardial ischemia and
stimulate nerves via a cascade of events, which involve activation and
amplification of platelets, TxA2, 5-HT, or other factors.
Specifically in this study, we have reported that TxA2 may
play a significant role in stimulating cardiovascular reflexes by
stimulating the release of 5-HT and subsequent stimulation of the
5-HT3 receptor.
| |
ACKNOWLEDGEMENTS |
We thank John Tyburski, Chris Adams, Chris Koster, and Chris
Stachura for useful scientific discussions that contributed to the
completion of this project.
| |
FOOTNOTES |
This work was supported by Grant-in-Aid 0051148Z from the American
Heart Association, Heartland Affiliate. S. Gomez was supported in part
by an Initiative for Minority Student Development grant from the
National Institute of General Medical Sciences, Division of Minority
Opportunities in Research (1-R25-GM-62232).
Address for reprint requests and other correspondence: M. Wacker, Dept. of Molecular Biosciences, 1200 Sunnyside Ave., Univ. of
Kansas, Lawrence, KS 66045 (E-mail:
mjwacker{at}ku.edu).
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. Section 1734 solely to indicate this fact.
First published October 31, 2002;10.1152/ajpheart.00617.2002
Received 18 July 2002; accepted in final form 30 October 2002.
| |
REFERENCES |
1.
Azami, J,
Fozard JR,
Round AA,
and
Wallis DI.
The depolarizing action of 5-hydroxytryptamine on rabbit vagal primary afferent and sympathetic neurons and its selective blockade by MDL 72222.
Naunyn Schmiedebergs Arch Pharmacol
328:
423-429,
1985[Web of Science][Medline].
2.
Baldassare, JJ,
Tarver AP,
Henderson PA,
Mackin WM,
Sahagan B,
and
Fisher GJ.
Reconstitution of thromboxane A2 receptor-stimulated phosphoinositide hydrolysis in isolated platelet membranes: involvement of phosphoinositide-specific phospholipase C-
and GTP-binding protein Gq.
Biochem J
291:
235-240,
1993[Medline].
3.
Bell, LB,
O'Hagan KP,
and
Clifford PS.
Cardiac but not pulmonary receptors mediate depressor response to IV phenyl biguanide in conscious rabbits.
Am J Physiol Regul Integr Comp Physiol
264:
R1050-R1057,
1993[Abstract/Free Full Text].
4.
Bley, KR,
Eglen RM,
and
Wong EHF
Characterization of 5-hydroxytryptamine-induced depolarizations in rat isolated vagus nerve.
Eur J Pharmacol
260:
139-147,
1994[Web of Science][Medline].
5.
Buzzard, CJ,
Pfister SL,
Halushka PV,
and
Campbell WB.
Decrease in vascular TxA2 receptors in a subgroup of rabbits unresponsive to the TxA2 mimetic.
Am J Physiol Heart Circ Physiol
266:
H2320-H2326,
1994[Abstract/Free Full Text].
6.
Coleman, J,
and
Rhodes KF.
Further characterization of the putative 5-HT4 receptor mediating depolarization of the rat isolated vagus nerve.
Naunyn Schmiedebergs Arch Pharmacol
352:
74-78,
1995[Web of Science][Medline].
7.
Coleman, RA,
Humphrey PPA,
Kennedy I,
Levy GP,
and
Lumley P.
Comparison of the actions of U-46619, a prostaglandin H2-analogue, with those of prostaglandin H2 and thromboxane A2 on some isolated smooth muscle preparations.
Br J Pharmacol
73:
773-778,
1981[Web of Science][Medline].
8.
Coleridge, JCG,
and
Coleridge HM.
Chemoreflex regulation of the heart.
In: Handbook of Physiology. The Cardiovascular System. The Heart. Bethesda, MD: Am Physiol Soc, 1979, sect. 2, vol. I, chapt. 18, p. 653-676.
9.
Coleridge, JCG,
Coleridge HM,
and
Kidd C.
Cardiac receptors in the dog, with particular reference to two types of afferent endings in the ventricular wall.
J Physiol
174:
323-339,
1964[Free Full Text].
10.
Csillik-Perczel, V,
Bakonyi A,
Yemane T,
Vitalis B,
Horvath E,
Solyom S,
Szekely JI,
and
Harsing LG, Jr.
GYKI-46 903, a non-competitive antagonist for 5-HT3 receptors.
Pharmacol Toxicol
79:
32-39,
1996[Web of Science][Medline].
11.
Elliot, P,
Seemungal BM,
and
Wallis DI.
Antagonism of the effects of 5-hydroxytryptamine on the rabbit isolated vagus nerve by BRL 43694 and metoclopramide.
Naunyn Schmiedebergs Arch Pharmacol
341:
503-509,
1990[Web of Science][Medline].
12.
Evans, RG,
Ludbrook J,
and
Michalicek J.
Characteristics of cardiovascular reflexes originating from 5-HT3 receptors in the heart and lungs of unanesthetized rabbits.
Clin Exp Pharmacol Physiol
17:
665-679,
1990[Web of Science][Medline].
13.
Fozard, JR.
MDL 72222: a potent and highly selective antagonist at neuronal 5-hydroxytryptamine receptors.
Naunyn Schmiedebergs Arch Pharmacol
326:
36-44,
1984[Web of Science][Medline].
14.
Freeman, AJ,
Cunningham KT,
and
Tyers MB.
Selectivity of 5-HT3 receptor antagonists and anti-emetic mechanisms of action.
Anticancer Drugs
3:
79-85,
1992[Medline].
15.
Fu, LW,
and
Longhurst JC.
Reflex pressor response to arterial phenylbiguanide: role of abdominal sympathetic visceral afferents.
Am J Physiol Heart Circ Physiol
275:
H2025-H2035,
1998[Abstract/Free Full Text].
16.
Fu, LW,
and
Longhurst JC.
Role of 5-HT3 receptors in activation of abdominal sympathetic C fibre afferents during ischemia in cats.
J Physiol
509:
729-740,
1998[Abstract/Free Full Text].
17.
Fu, LW,
and
Longhurst JC.
Role of activated platelets in excitation of cardiac afferents during myocardial ischemia in cats.
Am J Physiol Heart Circ Physiol
282:
H100-H109,
2002[Abstract/Free Full Text].
18.
Fu, LW,
and
Longhurst JC.
Thromboxane A2 stimulates ischemically sensitive cardiac afferents (Abstract).
FASEB J
16:
A830,
2002.
19.
Hainsworth, R.
Reflexes from the heart.
Physiol Rev
71:
617-658,
1991[Free Full Text].
20.
Hamberg, M,
Svensson J,
and
Samuelsson B.
Thromboxanes: a new group of biologically active compounds derived from prostaglandin endoperoxides.
Proc Natl Acad Sci USA
72:
2994-2998,
1975[Abstract/Free Full Text].
21.
Higashi, H,
and
Nishi S.
5-Hydroxytryptamine receptors of visceral primary afferent neurons on rabbit nodose ganglia.
J Physiol
323:
543-567,
1982[Abstract/Free Full Text].
22.
Hintze, TH.
Reflex regulation of the circulation after stimulation of cardiac receptors by prostaglandins.
Fed Proc
46:
73-80,
1987[Web of Science][Medline].
23.
Hirata, T,
Ushikubi F,
Kakizuka A,
Okuma M,
and
Narumiya S.
Two thromboxane A2 receptor isoforms in human platelets.
J Clin Invest
97:
949-956,
1996[Web of Science][Medline].
24.
Hirsh, PD,
Hillis LD,
Campbell WB,
Firth BG,
and
Willerson JT.
Release of prostaglandins and thromboxanes into the coronary circulation in patients with ischemic heart disease.
N Engl J Med
304:
685-691,
1981[Abstract].
25.
Hoyer, D,
Waeber C,
Karpf A,
Neijt H,
and
Palacios JM.
[3H]ICS 205-930 labels 5-HT3 recognition sites in membranes of cat and rabbit vagus nerve and superior cervical ganglion.
Naunyn Schmiedebergs Arch Pharmacol
340:
396-402,
1989[Web of Science][Medline].
26.
Ireland, SJ,
and
Tyers MB.
Pharmacological characterization of 5-hydroxytryptamine-induced depolarization of the rat isolated vagus nerve.
Br J Pharmacol
90:
229-238,
1987[Web of Science][Medline].
27.
Kaufman, MP,
Baker DG,
Coleridge HM,
and
Coleridge JCG
Stimulation by bradykinin of afferent vagal c-fibers with chemosensitive endings in the heart and aorta of the dog.
Circ Res
46:
476-484,
1980[Free Full Text].
28.
LeBreton, GC,
Venton DL,
Enke SE,
and
Halushka PV.
13-Azaprostanoic acid: a specific antagonist of the human blood platelet thromboxane/endoperoxide receptor.
Proc Natl Acad Sci USA
76:
4097-4101,
1979[Abstract/Free Full Text].
29.
Leone, M,
Silvestrini B,
Palmery M,
and
Cheng CY.
Identification of multiple biological factors in rabbit serum that modulate dopamine-mediated aortic constriction.
Biochem Int
6:
1079-1090,
1992.
30.
Longhurst, JC,
Tjen-ALooi SC,
and
Fu LW.
Cardiac sympathetic afferent activation provoked by myocardial ischemia and reperfusion.
Ann NY Acad Sci
940:
74-95,
2001[Web of Science][Medline].
31.
Matsumura, K,
Wtanabe YU,
Onoe H,
and
Watanabe Y.
Prostacyclin receptor in the brain and central terminals of the primary sensory neurons: an autoradiographic study using a stable prostacyclin analogue [3H]iloprost.
Neuroscience
65:
493-503,
1995[Web of Science][Medline].
32.
Mayer, GS,
and
Shoup RE.
Simultaneous multiple electrode liquid chromatographic-electrochemical assay for catecholamines, indoleamines, and metabolites in brain tissue.
J Chromatogr
255:
533-544,
1983[Web of Science][Medline].
33.
Meller, ST,
Lewis SJ,
Brody MJ,
and
Gebhart GF.
Vagal afferent-mediated inhibition of a nociceptive reflex by iv serotonin in the rat II Role of 5-HT receptor subtypes.
Brain Res
585:
71-86,
1992[Web of Science][Medline].
34.
Moncada, S,
and
Vane JR.
Pharmacology and endogenous roles of prostaglandin endoperoxides, thromboxane A2, and prostacyclin.
Pharmacol Rev
30:
293-331,
1975.
35.
Montalescot, G,
Drobinski G,
Maclouf J,
maillet F,
Salloum J,
Ankri A,
Kazatchkine M,
Eugene L,
Thomas D,
and
Grosgoeat Y.
Evaluation of thromboxane production and complement activation during myocardial ischemia in patients with angina pectoris.
Circulation
84:
2054-2062,
1991[Abstract/Free Full Text].
36.
Orr, JA,
Ernst M,
Carrithers J,
and
Shirer HW.
Cardiopulmonary responses to HCl infusion are mediated by thromboxane A2 but not by serotonin.
Respir Physiol
80:
203-218,
1990[Web of Science][Medline].
37.
Paasonen, MK.
Release of 5-hydroxytryptamine from blood platelets.
J Pharm Pharmacol
17:
681-697,
1956.
38.
Peters, JA,
Malone HM,
and
Lamert JJ.
An electrophysiological investigation of the properties of 5-HT3 receptors of rabbit nodose ganglion neurons in culture.
Br J Pharmacol
110:
665-676,
1993[Web of Science][Medline].
39.
Rhodes, KF,
Coleman J,
and
Lattimer N.
A component of 5-HT-evoked depolarization of the rat isolated vagus nerve is mediated by a putative 5-HT4 receptor.
Naunyn Schmiedebergs Arch Pharmacol
346:
496-503,
1992[Web of Science][Medline].
40.
Rizzi, CA,
Coccini T,
Onori L,
Manzo L,
and
Tonini M.
Benzimidazolone derivatives: a new class of 5-hydroxytryptamine4 receptor agonists with prokinetic and acetylcholine releasing properties in the guinea pig ileum.
J Pharmacol Exp Ther
261:
412-419,
1992[Abstract/Free Full Text].
41.
Sun, S,
Wang W,
and
Schultz HD.
Activation of cardiac afferents by arachidonic acid: relative contributions of metabolic pathways.
Am J Physiol Heart Circ Physiol
281:
H93-H104,
2001[Abstract/Free Full Text].
42.
Takahara, K,
Murray R,
Fitzgerald G,
and
Fitzgerald DJ.
The response to thromboxane A2 analogues in human platelets.
J Biol Chem
265:
6836-6844,
1989.
43.
Tyburski, JB,
Koster CJ,
Lundquist EA,
Floor E,
Wu JY,
and
Orr JA.
Thromboxane A2 receptor mRNA expression in rat sensory ganglia neurons (Abstract).
FASEB J
16:
A832,
2002.
44.
Ustinova, EE,
and
Schultz HD.
Activation of cardiac vagal afferents in ischemia and reperfusion; prostaglandins versus oxygen-derived free radicals.
Circ Res
74:
904-911,
1994[Abstract/Free Full Text].
45.
Wacker, MJ,
Tehrani RN,
Smoot RL,
and
Orr JA.
Thromboxane A2 mimetic evokes a bradycardia mediated by stimulation of cardiac vagal afferent nerves.
Am J Physiol Heart Circ Physiol
282:
H482-H490,
2002[Abstract/Free Full Text].
46.
Wright, CE,
and
Angus JA.
5-Carboxamidotryptamine elicits 5-HT2 and 5-HT3 receptor-mediated cardiovascular responses in the conscious rabbit: evidence for 5-HT release from platelets.
J Cardiovasc Pharmacol
13:
557-564,
1989[Web of Science][Medline].
47.
Yoshioka, M,
Ikeda T,
Abe M,
Togashi H,
Minami M,
and
Saito H.
Pharmacological characterization of 5-hydroxytryptamine-induced excitation of afferent cervical vagus nerve in anaesthetized rats.
Br J Pharmacol
106:
544-549,
1992[Web of Science][Medline].
Am J Physiol Heart Circ Physiol 284(3):H867-H875
0363-6135/03 $5.00
Copyright © 2003 the American Physiological Society
TOP
ABSTRACT
INTRODUCTION
