![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
Vol. 280, Issue 2, 761-769, 1997
Centre de Recherche Pierre Fabre,
 |
Abstract |
|---|
 Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
Experiments performed in 226 pentobarbitone-anesthetized rabbits were
designed to investigate the involvement of thromboxane/prostanoid and
5-hydroxytryptamine (5-HT)2A/C receptors during arterial
thrombus formation in distinct low- and high-shear rate thrombosis
models. Antithrombotic activities of the thromboxane/prostanoid
receptor antagonist SQ 29,548 and two chemically distinct
5-HT2A/C receptor antagonists, ritanserin and ketanserin,
were assessed first in low-shear rate (~600 sec
1)
arterial thrombosis, produced by insertion of a silk thread as
thrombogenic substrate into the central section of an extracorporeal arteriovenous shunt established between the left carotid artery and the
right jugular vein (n = 77), and second in
high-shear rate (~40,000 sec1) arterial thrombosis,
produced by critical stenosis and local endothelial injury of a carotid
artery, characterized by cyclic flow reductions (CFRs) due to recurrent
platelet aggregation and subsequent dislodgement of the thrombus
(n = 149). Under low shear rate, SQ 29,548 (10-2500 µg/kg plus 10-2500 µg/kg/hr i.v.), but not ritanserin or
ketanserin (both at 2500 µg/kg i.v.), dose-dependently inhibited
thrombus formation. In contrast, under high shear rate, SQ 29,548 (10-160 µg/kg plus 10-160 µg/kg/hr i.v.) and both ritanserin and
ketanserin (both at 10-2500 µg/kg i.v.) dose-dependently reduced CFR
frequency, with ID50 values of 35 µg/kg (95% confidence
limits, 24-58 µg/kg), 77 µg/kg (95% confidence limits, 40-132
µg/kg) and 89 µg/kg (95% confidence limits, 36-285 µg/kg) i.v.,
respectively. Furthermore, local infusion of the stable thromboxane
A2 analog U-46619 (0.63 µg/kg/min) or 5-HT (20.8 µg/kg/min) proximal to the site of injury and stenosis in rabbits
pretreated with either SQ 29,548 (40 µg/kg plus 40 µg/kg/hr i.v.)
or ritanserin (160 µg/kg i.v.), respectively, restored CFR frequency
to vehicle group levels in animals whose CFR frequency was previously
reduced. The inhibitory activity of ketanserin and ritanserin on CFRs
could not be attributed to 5-HT1B/D or
alpha-1 adrenoceptor antagonist properties or to any
hypotensive activity. These results provide firm evidence that
thromboxane/prostanoid receptors are involved in arterial thrombosis in
rabbits independently of the shear rate, whereas 5-HT2A/C
receptors play a major role only in high-shear rate thrombus formation.
| |
Introduction |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
Platelet aggregation is
influenced by shear forces (Ruggeri, 1994
). In particular, GP IIb/IIIa,
the final common pathway of platelet aggregation, interacts only with
fibrinogen in a low-shear rate environment, whereas it interacts mainly
with vWf in a high-shear rate environment (Ruggeri, 1994). Platelet
activation plays an important role in arterial thrombosis (Badimon
et al., 1992) because, early in the formation of the
hemostatic plug, platelet aggregates are formed at the site of vessel
injury, bifurcations or stenoses, which present local increases in
shear rates (Goldsmith and Turitto, 1986; Strony et al.,
1993). To mimic pathophysiological conditions, Folts developed a model
of CFRs in critically stenotic canine coronary artery with endothelial
damage, thereby producing high-shear rate thrombosis (Folts et
al., 1976; Folts, 1995). Ashton et al. (1987, 1989) and
Golino et al. (1989, 1990) showed that both TxA2 and 5-HT mediated CFRs in this canine model, via
5-HT2 and TP receptor activation, respectively.
Furthermore, elevated blood levels and tissue concentrations of
TxA2 and 5-HT have been detected around the stenosis and in
the distal canine coronary arterial blood (Schmitz et al.,
1985; Ashton et al., 1986). However, the roles of
5-HT2A/C and TP receptors in promoting thrombosis under low-shear rate situations are less well documented (Maffrand et al., 1988; Ruggeri, 1994).
The aim of the present study was to investigate further the involvement
of TP and 5-HT2A/C receptors during arterial thrombus formation in distinct low- and high-shear rate thrombosis models in
anesthetized rabbits, 1) in an extracorporeal arteriovenous shunt model
and 2) in arterial thrombosis produced by critical stenosis and local
endothelial injury of a carotid artery. To do so, we used the selective
and neutral TP receptor antagonist SQ 29,548 (Ogletree et
al., 1985; Bertolino et al., 1995) and the nonselective
5-HT2A/C receptor antagonists ritanserin and ketanserin
(Baxter et al., 1995). Because both ketanserin and ritanserin also have affinities for alpha-1 adrenoceptors
(Leysen et al., 1981, 1985) and/or 5-HT1B/D
receptors (Weinshank et al., 1991; Pauwels et
al., 1995), the role of these receptors was investigated using the
alpha-1 adrenoceptor antagonist prazosin and the novel and
highly selective 5-HT1B/D receptor antagonist GR 127935 (Clitherow et al., 1994; Skingle et al., 1994).
The nonselective COX inhibitor aspirin was also studied.
| |
Materials and Methods |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
General procedure. Experiments were carried out in accordance with French law and local ethics committee guidelines for animal research. Animals were housed in climate-controlled conditions (21°C and 55% relative humidity, with a 12-hr light/dark cycle) and provided standard chow and water ad libitum. On the day of the experiment, male New-Zealand White rabbits (2.2-3.1 kg; Elevage Scientifique. Des Dombes, Chatillon Sur Chalaronne, France) were anesthetized with an injection of sodium pentobarbital (30 mg/kg; Sanofi Laboratories, Libourne, France), administered through the marginal ear vein, and were then placed on a table under an homeothermic blanket to maintain rectal temperature at 39.5 ± 0.5°C. Through a median incision of the neck, animals underwent tracheotomy and were mechanically ventilated (Harvard Apparatus, South Natick, MA). Polyethylene catheters (I.D., 0.58 mm; Biotrol-Merck, Paris, France) were inserted into a femoral artery and vein for respectively, infusing fluids and drugs, sampling blood and continuously measuring arterial pressure via a Statham P10EZ pressure transducer (Viggo-Spectramed, Oxnard, CA) connected to a Gould amplifier (model 13-4615-50; Gould Instruments, Longjumeau, France). The analog arterial pressure signal was digitized (model MP 100; Biopac Systems, Goleta, CA) and simultaneously recorded by means of data acquisition software (AcqKnowledge 881 version 3.1.1; Biopac Systems).
Extracorporeal arteriovenous shunt model (low shear rate).
Animals were prepared according to the method described for rats by
Umetsu and Sanai (1978) and Shand et al. (1984) and modified by Freund et al. (1993). Briefly, the right jugular vein and
left carotid artery were exposed and carefully isolated from
surrounding tissues. The shunt (30 cm in length) was constructed with
polyethylene catheters (Biotrol-Merck, France) as follows: the
sections, which were inserted into a rabbit carotid artery and jugular
vein, consisted of 12.5-cm-long catheters (I.D., 1.14 mm). They were
connected to the central part of the shunt via a 6-cm-long
catheter (I.D., 2 mm). A silk thread (Gutermann Laska, Paris, France),
placed in the central portion of the shunt, was used as the
thrombogenic substrate when exposed to the circulation by unclamping of
the shunt. The polyethylene tubing used was coated with silicone
(Silisonde, Vygon, Ecouen, France). The shunt was filled with a 0.1 ml/kg heparin solution (50 IU/ml; Choay Laboratories, Paris, France). After clamping, one extremity of the shunt was inserted into the left
carotid artery and then the other was inserted into the right jugular
vein. A thermal microprobe (type IT-23; Physitemp Instruments Inc.,
Clifton, NJ) was secured onto the central part of the shunt. Blood flow
was then established through the shunt by unclamping, thereby rapidly
raising the shunt temperature to values slightly lower than the rectal
temperature of the animal (fig. 1). Shunt temperature
reached a plateau and then fell rapidly, coinciding with increasing
thrombotic obstruction of blood flow across the shunt. Occlusion time
of the shunt was defined as the elapsed time between the start of blood
flow and the time at which the shunt temperature was 1°C higher than
the base-line temperature (i.e., before blood flow start)
and corresponded to the formation of an occlusive thrombus and complete
interruption of blood flow (Freund et al., 1993).
|
Influence of drugs on shunt occlusion. Five minutes before shunt blood flow was established, animals received an i.v. injection of either 1) vehicle (0.9% NaCl, n = 5; 0.9% NaCl plus ethyl alcohol, 9:1, v/v, n = 6; or 2 mM Na2CO3, n = 16), 2) SQ 29,548 (10, 40, 160, 630 or 2500 µg/kg, n = 4-8 rabbits/group) administered over 2 min as a 1 ml/kg solution, followed by a constant infusion of 10 to 2500 µg/kg/hr (rate, 40 µl/min), 3) ketanserin (n = 7 rabbits) or 4) ritanserin (n = 8 rabbits).
Carotid stenosis and endothelial injury model (high shear
rate).
Animals were prepared according to a modification of the
method described by Golino et al. (1992). A common carotid
artery was exposed and carefully isolated from the surrounding tissues. The cranial thyroid artery, a small carotid artery collateral, was
cautiously exposed and a polyethylene catheter was inserted until the
ostium of the carotid artery was reached, thus allowing local infusion
of drugs and vehicles. Saline was continuously infused through the
catheter at the rate of 40 µl/min, to maintain patency. Carotid blood
flow velocity was continuously measured with a pulsed Doppler flow
probe (20 MHz, model HVPD-20; Crystal Biotech, Hopkinton, MA) placed
proximally to the cranial thyroid artery. Thereafter, a segment of the
exposed carotid artery, distal to the cranial thyroid artery, was
deendothelialized by gentle squeezing of the artery between a pair of
forceps. An external silicone cylinder (7-mm wide; I.D., 3 mm) was
placed around it, and critical stenosis was achieved by graded
inflation of an angioplasty balloon (model 3F; Solo, USCI-Bard
Laboratories, Paris, France) placed between the cylinder and the
carotid artery. Critical stenosis was confirmed by the absence of
hyperemia after a temporary (20-sec) complete occlusion of the carotid
artery (fig. 1). Once induced, CFRs were observed for 15 min (base
line) and for two consecutive 30-min periods. CFR frequencies were
quantified for each period and expressed per hour.
Influence of drugs on CFR frequency. After 15 min of CFRs, animals received either 1) vehicle (0.9% NaCl, n = 6; 0.9% NaCl plus ethyl alcohol, 9:1, v/v, n = 4; or 2 mM Na2CO3, n = 5), 2) SQ 29,548 (10, 20, 40 or 160 µg/kg; n = 6-10 rabbits/group) administered over 2 min as a 1 ml/kg solution, followed by a constant infusion of 10 to 160 µg/kg/hr, 3) ritanserin (10, 20, 40, 160, 630 or 2500 µg/kg, n = 5-13 rabbits/group) or 4) ketanserin (10, 40, 160, 630 or 2500 µg/kg, n = 4-6 rabbits/group). In two additional groups of rabbits the effects of local infusion, through the cranial thyroid artery, of exogenous 5-HT (20.8 µg/kg/min, n = 11) or the stable TxA2 analog U-46619 (0.63 µg/kg/min, n = 6) on CFR frequency were assessed in rabbits pretreated with 160 µg/kg ritanserin or 40 µg/kg plus 40 µg/kg/hr SQ 29,548. 5-HT and U-46619 were perfused during the second 30-min period at a rate of 40 µl/min. 5-HT1B/D and alpha-1 adrenoceptor blockade and nonselective COX inhibition were achieved by i.v. administration of GR 127935 (630 µg/kg over 1 hr at a rate of 40 µl/min, n = 9), prazosin (160 µg/kg, n = 5) or aspirin (2,500 µg/kg, n = 3, and 10,000 µg/kg, n = 5), respectively. After drug administration, hemodynamic variables and carotid blood flow velocity were continuously measured during two consecutive periods of 30 min.
Shear rate estimations and statistical evaluation.
In
arteriovenous shunt experiments, blood flow was determined in a
separate group of six rabbits by two consecutive measurements of blood
loss over 5 sec, after sectioning of the central part of the shunt.
Shear rates were estimated from the Poiseuille law for laminar flow
(Goldsmith and Turitto, 1986; Strony et al., 1993),
i.e., 
(R) = 4Q/
R3, where is the shear rate
(seconds1), Q is the flow rate (milliliters
per second) and R is the radius of the catheter
(centimeters).
): Q = 1.25 × d2 × 
f, where Q is
the flow rate (milliliters per minute), d is the diameter of
the carotid artery (millimeters), estimated to be equal to the diameter
of the Doppler probe, and f is the Doppler frequency
shift (kilohertz). Shear rates were estimated proximally to the
endothelial site of injury, and critical stenosis was determined according to the above formula.
Data are expressed as means ± S.E.M. Analysis of variance, with
or without repeated measures, followed by Dunnett's test was used to
assess significance among and between groups, respectively (StatView;
Abacus Concepts Inc., Berkeley, CA). Dose-response curves were fitted,
using an operational sigmoid model (Marquardt, 1963), from relative
inhibition of CFR determined 30 min after drug administration (first
30-min period) (Origin; Microcal Software Inc., Northampton, MA);
ID50 refers to the mean geometric antagonist dose (with
95% confidence intervals in parentheses) inhibiting a response by
50%. P = .05 was considered the minimum level of significance.
Drugs and solutions. SQ 29,548 and U-46619 (Cayman Chemical Co., Ann Arbor, MI) were dissolved in Na2CO3 (2 mM) and sterile saline (0.9%), respectively, and maintained on ice. Ritanserin and ketanserin tartrate (Research Biochemicals Inc., Natick, MA) were dissolved in ethyl alcohol plus sterile saline (0.9%; 1:9, v/v) and sterile saline (0.9%), respectively. GR 127935 was synthesized by the Division of Medicinal Chemistry IV, Centre de Recherche Pierre Fabre (Castres, France) and dissolved in sterile saline (0.9%). Prazosin hydrochloride and 5-HT creatinine sulfate (Sigma Chemical Co., St. Louis, MO) were dissolved in sterile saline (0.9%), and 5-HT was maintained on ice. Aspirin (acetylsalicylic acid lysine salt; Synthelabo Laboratories, Paris, France) was dissolved in sterile saline (0.9%). Drugs were administered (as 1 ml/kg solutions), in micrograms per kilogram base weight, except when specified otherwise.
| |
Results |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
Estimation of shear rates.
Results are presented in table
1. Estimated shear rates were markedly higher in
arterial thrombosis produced by carotid stenosis and endothelial
injury, compared with the arteriovenous shunt model (~40,000
sec1 vs. ~600 sec1).
|
Influence of drugs on shunt occlusion.
Results are presented
in table 2. In control, vehicle-infused rabbits, because
no statistically significant difference was found among occlusion times
for the three groups, data were pooled together. In vehicle-infused
rabbits, the occlusion time was 13.7 ± 1.3 min. The interruption
of blood flow was associated with a slight reduction in MAP and no
statistically significant changes in HR [mean maximal absolute
changes: MAP = 
9 ± 2 mm Hg and HR = 2 ± 4 beats/min; P < .05 and P = NS vs. base line,
respectively]. The TP receptor antagonist SQ 29,548 significantly and
dose-dependently inhibited thrombus formation, attaining a maximal
effect of 31.3 ± 7.3 min (P < .05 vs. vehicle
group) at the highest dose (2,500 µg/kg) without affecting MAP or HR,
compared with vehicle-treated animals. In contrast, neither ritanserin
nor ketanserin (both at 2,500 µg/kg) inhibited thrombus formation. A
statistically significant reduction in MAP and HR was observed in
ketanserin-treated but not ritanserin-treated animals, compared with
vehicle-infused rabbits.
|
Carotid stenosis and endothelial injury model.
As shown in
figure 2, critical stenosis at the site of endothelial
injury, achieved by graded inflation of an angioplasty balloon, led to
the development of the typical pattern of gradual reductions of blood
flow, followed by either spontaneous or induced (by gentle shaking of
the cylinder) restorations of flow to base-line levels
(i.e., postcritical stenosis). These CFRs are known to be
due to recurrent platelet aggregation at the site of the stenosis, followed by embolization of the thrombus. CFRs developed in all animals
(n = 149), with a mean frequency of 23.9 ± 0.5 cycles/hr. In control, vehicle-infused rabbits, no significant change
in CFR frequency was observed between base line and the first 30-min period of observation, whereas a slight decrease (~ 25%) in CFR frequency was noted between the first and second periods (fig. 3).
|
|
Influence of SQ 29,548 on CFR frequency. Results are presented in table 3 and figures 2 and 3. Administration of SQ 29,548, 15 min after initiation of CFRs, dose-dependently reduced the frequency of CFRs over the first 30-min period of observation. The ID50, determined 30 min after drug administration, was 35 µg/kg (95% confidence limits, 24-58 µg/kg). Significant inhibition of CFR frequency was observed with 40 µg/kg SQ 29,548. The highest dose abolished CFRs in seven of eight rabbits at the end of the 30-min observation period and produced a maximal CFR frequency reduction of 80 ± 8% (P < .05 vs. base line). Reduction of CFR frequency by SQ 29,548 occurred without significant changes in MAP or HR, compared with vehicle-infused animals. Furthermore, in an additional group of six rabbits to which SQ 29,548 (40 µg/kg plus 40 µg/kg/hr) was administered, infusion of U-46619 through the cranial thyroid artery, at the dose of 0.63 µg/kg/min, restored CFR frequency to vehicle group levels (from 22.7 ± 3.0 to 9.0 ± 3.3 and then 18.0 ± 6.6 cycles/hr, at base line, first and second period, respectively of the U-46619 infused group; P = NS for second period vs. base line and vehicle group; P < .05 for second period vs. first period; Fig. 3A). Interestingly, U-46619 failed to restore CFRs in two rabbits whose CFRs had been abolished by SQ 29,548.
|
Influence of ketanserin and ritanserin on CFR frequency.
Results are presented in table 3 and figures 2 and 3. Administration of
either ketanserin or ritanserin, 15 min after initiation of CFRs,
dose-dependently reduced CFR frequency over the first 30 min of
observation, with significant inhibition from 40 and 20 µg/kg,
respectively (both P < .05), giving ID50 values of 89 µg/kg (95% confidence limits, 36-286 µg/kg) and 77 µg/kg (95% confidence limits, 40-132 µg/kg), respectively. The highest doses of
ketanserin and ritanserin (2500 µg/kg) abolished CFRs in four of five
and six of six rabbits, respectively, and produced maximal reductions
in CFR frequency of 83 ± 7% and 98 ± 2% (both P < .05 vs. base line). MAP was significantly reduced by the
high dose of ketanserin (MAP = 21 ± 7 mm Hg; P < .05 vs. vehicle group), whereas no significant reduction was
observed in ritanserin-treated, compared with vehicle-treated, animals.
HR was statistically significantly reduced by the high dose (2500 µg/kg) of ketanserin and ritanserin (HR = 36 ± 22 and
36 ± 9 beats/min respectively; P < .05 vs. vehicle group). In additional experiments, local infusion of exogenous 5-HT (20.8 µg/kg/min, n = 11) in rabbits pretreated
with 160 µg/kg ritanserin restored CFR frequency to vehicle group
levels (from 22.2 ± 1.4 to 10.9 ± 2.0 and then 17.1 ± 3.7 cycles/hr, at base line, first and second period, respectively, of
the 5-HT infused group; P = NS for second period vs.
base line and vehicle group; P < .05 for second period
vs. first period; Fig. 3B). 5-HT was also unable to restore
CFRs in three rabbits whose CFRs had been abolished by ritanserin.
Influence of 5-HT1B/D and alpha-1 adrenergic receptor blockade and COX inhibition on CFR frequency. Because both ketanserin and ritanserin have affinity for 5-HT1B/D receptors, we addressed the possibility that inhibitory activities of both compounds on CFR frequency could be mediated through 5-HT1B/D receptor blockade. For this purpose, we determined whether CFR frequency could be reduced by the novel and highly selective 5-HT1B/D receptor antagonist GR 127935. Administration of GR 127935 (630 µg/kg i.v.) did not statistically significantly reduce CFR frequency or modify MAP or HR, compared with vehicle-infused animals (table 3).
To further evaluate whether the activity of ketanserin on CFR frequency could be related to its alpha-1 adrenoceptor antagonist properties and associated systemic hypotensive effects, we explored whether CFR frequency could be reduced by the alpha-1 adrenoceptor antagonist prazosin, at a dose (160 µg/kg i.v.) producing systemic hypotension equivalent to that induced by the highest dose of ketanserin studied (MAP = 29 ± 5 vs. 21 ± 7 mm Hg; both P < .05 vs.
vehicle-treated rabbits and P = NS between groups). Under these
conditions, prazosin did not statistically significantly reduce CFR
frequency, with respect to vehicle-treated animals (table 3).
Finally, to verify the platelet dependency of CFRs under our
experimental conditions, we determined whether CFR frequency could be
reduced by the COX inhibitor aspirin. Acute i.v. administration of
aspirin, 15 min after initiation of CFRs, dose-dependently reduced CFR
frequency over the first 30 min of observation, by 59 ± 21 and
92 ± 3% at 2,500 and 10,000 µg/kg, respectively, without statistically significantly affecting MAP or HR (table 3).
| |
Discussion |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
The present studies performed in anesthetized rabbits demonstrated that the TP receptor antagonist SQ 29,548 dose-dependently inhibited thrombus formation in both low- and high-shear rate arterial thrombosis, whereas ketanserin and ritanserin were effective only in the high-shear rate model of CFRs. The damping activity of ketanserin and ritanserin on CFR frequency could not be attributed to 5-HT1B/D or alpha-1 adrenoceptor antagonist properties and associated systemic hypotensive activities but, rather, was attributed to 5-HT2A/C receptor antagonist properties. Furthermore, local infusions of either the TxA2 analog U-46619 or 5-HT to animals pretreated with SQ 29,548 or ritanserin, respectively, restored CFR frequency to vehicle-infused levels. These results strongly suggest that TP receptors are involved in arterial thrombosis in rabbits independently of the shear rate, whereas 5-HT2A/C receptors play a major role only in high-shear rate thrombus formation.
Involvement of TP and 5-HT2A/C receptors in
high-shear rate arterial thrombosis.
In the carotid stenosis and
endothelial injury model, SQ 29,548, ritanserin and ketanserin all
dose-dependently reduced CFR frequency (ID50 values of
35, 77 and 89 µg/kg, respectively). The inhibitory activity of
ketanserin and ritanserin on CFRs could not be attributed to either
5-HT1B/D or alpha-1 adrenoceptor antagonist properties and associated systemic hypotensive activities (see below)
but, rather, was attributed to 5-HT2A/C receptor antagonist properties. Local infusion of U-46619, a stable TxA2
analog, or 5-HT at the site of the injury restored CFR frequency to
vehicle-infused levels in animals whose CFRs had been reduced, but not
abolished, by SQ 29,548 or ritanserin, respectively. These results
provided further evidence that TP and 5-HT2A/C receptors
mediated thrombus formation and maintenance in the CFR experiments, in
agreement with previous reports (Willerson et al., 1989;
Golino et al., 1990, 1992, 1993; Torr et al.,
1990; Salvati et al., 1993; Beaughard et al.,
1995). Furthermore, aspirin also dose-dependently reduced CFR
frequency, thus confirming the platelet-dependent CFR occurrence described in dogs with coronary artery stenosis and endothelial injury
(Folts, 1995).
; Hoyer et
al., 1994), ketanserin and ritanserin also have affinity for
5-HT1B/D receptors (Weinshank et al., 1991;
Pauwels et al., 1995). Therefore, we addressed the
possibility that inhibitory activities of ketanserin and ritanserin on
CFR frequency could be mediated through 5-HT1B/D receptor
antagonism. The novel and highly selective receptor antagonist GR
127935 (Clitherow et al., 1994; Skingle et al.,
1994), at a dose (630 µg/kg i.v.) that is higher than that required
to fully block 5-HT1B/D receptor-mediated carotid
vasoconstriction in anesthetized pigs (De Vries et al.,
1996), did not alter CFR frequency or hemodynamic parameters. Thus,
these results provide evidence that 5-HT1B/D receptors are
apparently not involved in mediating thrombus formation under the
present experimental conditions.
The damping activity of ketanserin on CFR frequency cannot be related
to alpha-1 adrenoceptor antagonist properties (Leysen et al., 1981) either, because the selective
alpha-1 adrenoceptor antagonist prazosin, at a dose
producing systemic hypotension equivalent to that induced by the
highest dose of ketanserin (2,500 µg/kg), did not reduce CFR
frequency. These results suggest that alpha-1 adrenoceptors
are not involved in mediating arterial thrombus formation under
high-shear rate conditions.
Both TxA2 and 5-HT are implicated in the pathogenesis of
platelet aggregation and dynamic vasoconstriction that occur at sites of endothelial injury and coronary artery stenosis. When platelets aggregate, they release (among other factors) 5-HT, which causes local
vasoconstriction and acts to amplify and further promote aggregation.
It might be expected that vasodilation (reflected by systemic
hypotension), as observed after administration of the highest dose of
ketanserin investigated, would counteract the local vasoconstriction
induced by the vasoactive agents released during aggregation and would
thus reduce further aggregation (i.e., reduce CFR
frequency). This possibility can be excluded because 1) a statistically
significant reduction in CFR frequency was observed at doses of
ketanserin that did not affect MAP (
630 µg/kg), 2) ritanserin at a
dose that reduced CFR frequency to the same extent as that produced by
the highest dose of ketanserin (83 ± 8% at 630 µg/kg
vs. 83 ± 7% at 2,500 µg/kg) did not reduce MAP and,
3) at a dose producing systemic hypotension equivalent to that evoked
by the highest dose of ketanserin, the alpha-1 adrenoceptor
antagonist prazosin did not significantly reduce CFR frequency.
Moreover both ketanserin and ritanserin, at the highest dose
investigated (2,500 µg/kg), induced bradycardia, which could reduce
cardiac output, as reported by Bolt and Saxena (1985), and possibly
carotid blood flow. In fact, ritanserin at a dose that reduced CFR
frequency to the same extent as that produced by the highest dose of
ketanserin (83 ± 8% at 630 µg/kg vs. 83 ± 7%
at 2,500 µg/kg) did not alter HR, thus excluding bradycardia as a
major antithrombotic mechanism of action of these drugs. Taken
together, these results provide evidence that the 5-HT1B/D and alpha-1 adrenergic antagonist properties of ketanserin
and ritanserin are not involved in reducing CFR frequency, thus
confirming that both 5-HT2A/C and TP receptor activation
are major mechanisms involved in CFR occurrence and maintenance in
stenotic and endothelially injured rabbit carotid arteries.
Differential involvement of TP and 5-HT2A/C
receptors in low-shear rate arterial thrombosis.
In contrast to
ketanserin and ritanserin, SQ 29,548 significantly and dose-dependently
inhibited arteriovenous shunt occlusion without affecting hemodynamic
parameters. Inactivity of ketanserin and ritanserin in the present
experimental model cannot be explained by the use of inadequately low
doses, because 2,500 µg/kg is relatively high, compared with doses
that substantially inhibit 5-HT2A/C receptor-mediated
responses in vivo (Bolt and Saxena, 1985; Pettersson et al., 1985; Docherty, 1989; Valentin et al.,
1995). Antithrombotic inactivity of ketanserin has previously been
reported in an arteriovenous shunt model in rats (Maffrand et
al., 1988). In addition, such doses of ketanserin and ritanserin
are likely to have extensively blocked alpha-1 adrenergic
(Bolt and Saxena, 1985; Pettersson et al., 1985; Docherty,
1989) or 5-HT1B/D (De Vries et al., 1996) receptors, respectively (see above), thereby excluding any involvement of alpha-1 adrenoceptors or 5-HT1B/D receptors
in arterial thrombus formation under low-shear rate conditions.
5-HT2A/C and TP receptors clearly do not share similar
involvement in mediating arterial thrombus formation under low-shear
rate conditions. Moreover, the antithrombotic effectiveness of SQ
29,548 can be accounted for by inhibition of locally produced
TxA2/endoperoxides, which elicit platelet aggregation
(Ogletree, 1987). The extracorporeal arteriovenous shunt was previously
described in rats as a platelet-predominant thrombosis model (Umetsu
and Sanai, 1978; Shand et al., 1984), and the histological
analyses of thrombus composition we performed (data not shown) confirm
and extend these observations to rabbits. Evidence is therefore
presented that platelet activation is a key component of arterial
thrombus formation under low-shear rate conditions in the rabbit
arteriovenous shunt. Interestingly, no information is currently
available on the existence of putative platelet 5-HT1B/D
receptors. Thus, platelet activation is mediated partly through TP
receptor stimulation, whereas platelet 5-HT2A/C receptors
appear to have little or no involvement.
Thrombus formation in high- vs. low-shear rate arterial
thrombosis.
A major finding of the present study was that SQ
29,548 elicited antithrombotic activity independently of the shear
rate, whereas ketanserin and ritanserin exerted substantial
antithrombotic activity only when shear rates were high. High shear
rates, such as those found in the present study, are not physiological
(20,000-60,000 sec1) but are reached under pathological
conditions in stenotic arteries (Goldsmith and Turitto, 1986; Strony
et al., 1993). The way in which shear stress can induce
aggregation of platelets is gradually being elucidated. It is now
established that the GP IIb/IIIa receptor, the final common pathway of
platelet aggregation, interacts only with fibrinogen in a low-shear
rate environment, whereas it interacts mainly with vWf in a high-shear
rate environment (Ruggeri, 1994). The role of vWf appears to be most
significant at high shear rates, presumably as a consequence of its
unique molecular architecture. Under the effects of high shear forces,
vWf molecules take the shape of extended filaments; the repeating
subunit structure of these large multimers offers an array of
interaction sites capable of binding in a multivalent manner to
receptors on the platelet membrane, thereby increasing the number of
contact points and the strength of interaction. As a result, the
overall force linking platelets to the surface and/or to one another is
increased. This interpretation of events explains why the role of vWf
is less relevant at lower shear rates, because other adhesive molecules may provide sufficient force of interaction to withstand opposing shear
forces of lesser magnitude (Chow et al., 1992; Ikeda
et al., 1993; Ruggeri, 1994). In remarkable contrast, GP
IIb/IIIa in a low-shear rate environment shows the ability to interact only with immobilized fibrinogen (Savage and Ruggeri, 1991).
Interestingly, monoclonal antibodies directed against GP IIb/IIIa and
GP IIb/IIIa receptor antagonists have demonstrated high efficacy in
situations of both low (Ikeda et al., 1991) and high (Coller
and Scudder, 1985; Gold et al., 1988; Shebuski et
al., 1989a,b; Chow et al., 1992) shear rates,
confirming the pivotal role of GP IIb/IIIa in the process of thrombus
formation, independently of shear rate. Furthermore, Golino et
al. (1995) recently demonstrated the key role of GP IIb/IIIa in
the stenotic and endothelially injured rabbit carotid artery model
(Golino et al., 1995).
), is
also in agreement with different mechanisms mediating thrombus formation at high vs. low shear rates and strongly suggests
that 5-HT plays a major role in thrombus growth only under high-shear rate conditions. A role for 5-HT in mediating the formation of arterial
thrombi under low-shear rate conditions cannot, however, be excluded,
because the indoleamine has been reported to mediate platelet
aggregation in vitro, albeit weakly, and to contribute to
aggregate growth in vivo (Menys, 1993). The basis for a
major role of 5-HT in high-shear rate thrombus formation, compared with a minor role with low shear rates, is unclear at present but could involve enhanced 5-HT release from platelet dense granules with high
shear rates. It is well established that, in vitro, high shear stresses (>50 dyn/cm2) activate spontaneous or
agonist-induced aggregation by the release of platelet dense granule
contents (Brown et al., 1975). This would be compatible with
the increased transcardiac 5-HT concentrations that have been observed
in patients with coronary stenoses (Van den Berg et al.,
1989). Factors other than shear rate may have influenced the
differential responsiveness of 5-HT2A/C receptor antagonists in the two models. Differences in 1) thrombogenic substrate
between the two models (i.e., silk thread vs.
exposed subendothelial collagen) and 2) the size of the thrombogenic
surface may also be involved in the differential antithrombotic
effectiveness of ketanserin and ritanserin, which would also lend
support to the hypothesis that the mechanisms of thrombus formation, in
particular platelet activation, are different under high-
vs. low-shear rate conditions. In contrast to the
5-HT2A/C receptor, TP receptor activation plays a major
role in both high- and low-shear rate arterial thrombosis. A possible
explanation is that arachidonic acid metabolism in platelet membranes,
leading to TxA2/endoperoxide production and consequent TP
receptor activation, occurs independently of the mechanism of platelet
activation in vivo (Badimon et al., 1992; Reilly
and FitzGerald, 1993). Although 5-HT2A/C and TP receptor antagonists have demonstrated efficacy in preventing thrombus formation
in high-shear rate situations, the relative physiological and
pathological importance of these mechanisms is probably moderate, compared with functional antagonism of GP IIb/IIIa.
In conclusion, our results indicate that TP receptors are involved in
arterial thrombosis in rabbits, independently of the shear rate,
whereas 5-HT2A/C receptors play a role only in high-shear rate thrombus formation. The precise mechanism underlying the differential role of 5-HT2A/C receptors in high-
vs. low-shear rate arterial thrombosis deserves further
study.
| |
Acknowledgments |
|---|
The authors are very grateful to Drs. S. Halazy and C. Jorand, Centre de Recherche Pierre Fabre, for the synthesis of GR 127935.
| |
Footnotes |
|---|
Accepted for publication October 11, 1996.
Received for publication July 23, 1996.
1 Part of this work was presented at the 68th Scientific Sessions of the American Heart Association, Anaheim, CA, November 13-16, 1995, and was published in abstract form in Circulation 92: suppl. I, A2979, 1995.
Send reprint requests to: Gareth W. John, Ph.D., Centre de Recherche Pierre Fabre, Division of Cardiovascular Diseases, 17 avenue Jean Moulin, 81106 Castres cédex, France.
| |
Abbreviations |
|---|
CFR, cyclic flow reduction;
COX, cyclooxygenase;
GP, glycoprotein;
HR, heart rate;
5-HT, 5-hydroxytryptamine;
I.D., internal diameter;
MAP, mean arterial
pressure;
NS, not significant;
SQ 29, 548,
[1S-[1
,2(5Z),3,4]]-7-[3-[[2-(phenylaminocarbonyl)hydrazino]methyl]-7-oxabicyclo[2.2.1]hept-2-yl]-5-heptenoic
acid ;
TP receptor, thromboxane A2/prostanoid receptor;
TxA2, thromboxane A2;
U-46619, 9,11-dideoxy-9,11-methanoepoxy-prostaglandin F2
;
vWf, von Willebrand factor.
| |
References |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
)2 fragments of a monoclonal antibody to the platelet glycoprotein IIb/IIIa receptor.
Blood
66: 1456-1459, 1985 receptors and cloned human 5-HT1D
, sheep and bovine caudate nucleus 5-HT1D receptor sites.
Cell. Pharmacol.
2: 183-191, 1995.This article has been cited by other articles:
|
|
 |
|
  B. Aktas, M. Pozgajova, W. Bergmeier, S. Sunnarborg, S. Offermanns, D. Lee, D. D. Wagner, and B. Nieswandt Aspirin Induces Platelet Receptor Shedding via ADAM17 (TACE) J. Biol. Chem., December 2, 2005; 280(48): 39716 - 39722. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||
) or SQ 29,548 (40 µg/kg plus 40 µg/kg/hr)
alone (n = 10) (
) or followed by local infusion
of U-46619 (0.63 µg/kg/min; n = 6) (
). B,
animals received either the vehicle (
) or followed by local infusion
of 5-HT (20.8 µg/kg/min) (
). Values are mean ± S.E.M.
*P < .05 vs. base line; #P < .05 for SQ
29,548 or ritanserin alone vs. vehicle-infused groups;
§P < .05 for SQ 29,548 or ritanserin alone or followed by
U-46619 or 5-HT, respectively. CFRs initiated at