- Email: [email protected]
Serotonin-induced vasoconstriction is mediated by thromboxane release and action in the human fetal-placental circulation
Placenta
(1997),
18, 197-204
Serotonin-induced Release
and Action
M. A. Cru9,
V. Gallardo”,
a Department
of
b Department
of Obstetrics,
Paper
Vasoconstriction
accepted
Physiology,
27
July
in the Human
P. Miguelb, Faculty Faculty
is Mediated
of of
Biological Medicine,
Fetal-Placental
G. Carrascoa Sciences, University
and University
of Concepckjn,
by Thromboxane Circulation
C. GonzSleza’c of Concepcibn, Concepcibn,
Concepckk,
Chile
Chile
1996
The possibility that prostanoids mediate the contractile response of serotonin on placental vessels was investigated. Rings of chorionic plate arteries and veins with and without endothelium were suspended in an organ bath for recording isometric mechanically activity. Serotonin caused dose-dependent contractions that were significantly attenuated by indomethacin (cycle-oxygenase inhibitor, 10 ,UM) and SQ29,548 (thromboxane receptor antagonist, 1 ,UM). Pretreatment of placental venous and arterial rings with indomethacin decreased sensitivity (EC,,) to serotonin of 2.3- and 1.9-fold, respectively. Pretreatment with SQJ9,548 decreased sensitivity to serotonin of twofold in veins and 2.1-fold in arteries. In the endothelium-denuded placental arteries and veins, pretreatment with indomethacin and SQ29,548 reduced the serotonin-induced contraction in a similar way to that obtained in the endothelium-intact vessels. In isolated perfused cotyledon through the fetal circulation, serotonin caused a significant increase in perfusion pressure and stimulated thromboxane release 1.9-fold compared with basal values. Therefore, serotonin-induced vasoconstriction in the human fetoplacental circulation appears to be mediated in part by thromboxane release or action. This effect is not dependent on mediators released from the endothelium. The present study provides evidence for the participation of thromboxane A, in the contractile response to serotonin in the human placental circulation. The ability of serotonin to release thromboxane A, which is also a potent vasoconstrictor agent, may be important in increase fetoplacental resistance, one of the features of pre-eclampsia. 0 1997 W. B. Saunders Company Ltd Placenta (1997), 18, 197-204
INTRODUCTION
The umbilical-placental circulation is very important for normal fetal growth and viability because it is responsible for the supply of nutrients and oxygen to the fetus. Impairment of this circulation could increase the possibility of growth retardation, fetal distress, and intrauterine fetal death. Fetoplacental blood flow is predominantly controlled by circulating or locally produced autacoids because the placental vasculature lacks innervation (Fox and Khong, 1990). These autacoids appear to interact in many organs to regulate vascular tone, tissue blood flow, and vascular resistance (Boura and Walters, 1991). Previous studies have demonstrated that serotonin produces potent vasoconstrictor responses in placental vesselsby working at specific serotonin receptor sites (Gonzalez et al., 1990; Reviriego, Fernandez-Alfonso and Marin, 1990) and has been implicated in the genesis of pre-eclamptic hypertension (Weiner, 1987; Cruz et al., 1993; Middelkoop et al., 1993). Its role is mainly due to the amplification, through activation of 5-hydroxy-tryptamine serotoninergic receptors, of the action c To whom correspondenceshould be addressedat: Department of Physiology, by) of Biological Sciences,University of Concepci6q P.P. Box 152-C, Concepcibn, Chile. 0143-4004/97/020197+08
$12.00/O
of other agonists on vascular smooth muscle cells and platelets (Cruz et al., 1991; Gonzalez et al., 1993; De Clerck and de Chaffoy de Courcelles, 1989). Serotoninergic blockade has been used with some success in hypertensive disorders of pregnancy (Weiner, Socol and Vaisrub, 1984; Spitz et al., 1993). Serotonin can work either directly or by interacting with vasoconstrictor eicosanoids. Thus, vasoconstrictor products of the cycle-oxygenase pathway have been proposed to modulate serotonin-induced contractions in some peripheral blood vessels (Bjoro, 1986; Howarth, Valance and Wilson, 1993). The human placenta is known to produce large amounts of prostaglandins that may have local actions on the fetoplacental circulation (Benedetto et al., 1987). The eicosanoids, including thromboxane and prostaglandins, have been strongly implicated in the control of fetoplacental circulation under both physiological and pathological conditions (Meagher and Fitzgerald, 1993; Lie1 et al., 1993). Others studies have demonstrated thromboxane B, release into media perfusing the fetoplacental vasculature and vasoconstriction of the fetoplacental vasculature by thromboxane analogues (Bjoro, 1986; Hedberg et al., 1989; Le et al., 1993). To evaluate the potential role of altered serotonin responsiveness and altered production of thromboxane A, in pre-eclampsia, it is important to determine whether these 0
1997 W. B. Saunders
Company
Ltd
Placenta
198
blood flow controllers interact in the normal placental vasculature. In the current study we performed a series of experiments to test the hypothesis that the release of vasoconstrictor products of the cycle-oxygenase pathway modulates serotonininduced contractions of the chorionic plate vesselson the fetal side of the human placenta.
MATERIALS
AND
METHODS
Placentae of uncomplicated pregnancies were collected immediately after vaginal delivery or caesarean section at term with approval of the institutional review boards for the G. Grant Benavente Hospital. Parturients ranged in aged from 20 to 40 years, had no history of chronic medical illness and had normal, term pregnancies and labour. In this study, we found no significant differences (P>O.5) in vascular reactivity between tissues obtained at caesarean section and those at vaginal delivery. The placentae were placed in a normal physiological salt solution (PSS) composition 119.0 mM NaCl, 4.5 rnM KC1 1.2 mM KH,PO,, 1.2 mM MgSO,, 24.9 mM NaHCO,, 1.6 mM CaCl,, 11.5 mM glucose and transported to the Research Laboratory within 30 min of collection.
Chorionsic
ring vessels
Under an illuminated operating magnifiers chorionic plate, arteries and veins of external diameter between 2 and 3 mm were isolated from the placentae, freed of adhering connective tissue, and cut into 2-mm ring segments. Two stainless steel wires were inserted through the vessel ring, one connected to a Grass FT03 force displacement transducer for isometric recording of the contractions on a Grass pen recorder (Model 79 D), and the other was used to adjust the resting tension of the vessel to 1.5 g in veins and 2.0 g in arteries (Gonzalez et al., 1990). The bath content was maintained thermostatically at 37°C pH 7.4, and was constantly bubbled with a gas mixture containing 95 per cent O,, 5 per cent CO,. Endothelium removal was achieved mechanically by gently rubbing the intimal surface of the rings. Because the precontracted intact rings did not relax with either acetylcholine or histamine, the effectiveness of the procedure for the removal of the endothelium was confirmed histologically (Helliwell, Smith and Heath, 1982) at the end of each experiment. After the optimal resting tension was obtained, the tissues were allowed to equilibrate for 1 h. At the beginning and end of each experiment, the maximal contractile response to 75 mM KC1 solution was obtained. Preliminary experiments have shown that 75 mM KC1 evoke the maximal contractile response in these vessels. The maximum response to KC1 was used as standard for comparing responses elicited by other agonists because it is independent of receptors and because the great variation in contractile force generating among isolated vessels. Only if the contractile responses to KC1 were similar in magnitude (with variation less than 10 per cent), were the
(1997,
Vol.
18
data from this particular experiment included in the analysis. In the control group isometric tensions were recorded at increasing cumulative doses of serotonin (0.01-l PM). Rings in the experimental groups were treated with either indomethacin (cycle-oxygenase inhibitor, 10 ELM) or SQ29,548 (thromboxane-receptor antagonist, 1 PM) for 30 min before addition of serotonin. To determine the effect of endothelium removal on the response to serotonin, parallel experiments were performed for two rings with endothelium and for two without endothelium, all from the same placenta. The contraction induced by each concentration of serotonin was expressed as a percentage of the contraction due to KC1 and was used for making the concentration-response curves. The concentration of serotonin eliciting 50 per cent of its own maximum response (EC,,) was determinated graphically for each curve by linear interpolation.
Isolated
perfused
placental
cotyledons
Perfused isolated placental cotyledons were prepared as previously described (Mak et al., 1984). In brief, individual cotyledons supplied by a single chorionic artery and vein were selected for study and the fetal vessels cannulated with polyethylene catheters. The arterial catheter was attached to a T-connector; one limb was attached to a Viggo-Spectramed transducer (model P23XL) for continuous recording of pressure on a Grass polygraph (Grass Instruments) and the other limb was connected to a Masterflex pump (Cole Palmer). A 2-cm piece of rubber tubing was inserted into the arterial perfusion line to permit injection of test agents through syringes mounted in infusion pumps delivering at a constant flow rate. The preparation was placed in a temperaturecontrolled incubator maintained at 37°C throughout the experiment. The fetal compartment was perfused at a rate of 1.2 ml/min with a perfusion medium containing 118 mM NaCl, 4.7 mM KCl, 2.53 mM CaCl,, 1.06 mM .MgSO, 7 H,O, 1.18 mM KH,P04, 25 rnM NaHCO,, 5.55 mM glucose. The perfusate was gassed by bubbling carbogen (95 per cent O,, 5 per cent CO,) through the medium in the reservoirs. Vascular responses were studied for a 60-min stabilization period. The effect of serotonin (0.5 PM) on thromhoxane release was determined. Experiments were performed after a 60-min stabilization. During the experiment, four 15-min intervals perfusate samples were collected, the first correspond to baseline period, the second, at the onset of perfusion pressure increase induced by serotonin infusion (phase 1) the third, when maximum response to serotonin on perfusion pressure was observed (phase 2) and the last one, during the recovery period after the discontinuation of serotonin infusion. Perfusion pressure was monitored through the experiments, and venous effluent was collected on ice into test tubes containing indomethacin (100 ~1 of 10 mM) for the determination of thromboxane B, concentrations. Quantification of thromboxane formation was made by determining the level of thromboxane B, a stable hydrolysed
Cruz et al.: Serotonin and Thromboxane in Placental Circulation
product of unstable thromboxane A, which is derived from prostaglandin H, (PGH, Thromboxane B, was measured by an enzyme linked immunosorbant assay (ELISA) test (Elisa Technologies) in the venous effluent (without extraction). The intra- and interassay coefficients of variation were 3.10 and 5.83 per cent respectively.
Drugs
and
statistics
The different concentrations of drugs used were obtained from stock solutions (1 mM) in physiological saline solution containing 0.01 per cent (w/v) ascorbic acid. All stock solutions were kept at 4°C. Drugs used were 5-hydroxytryptamine creatinine sulphate, indomethacin (Sigma Chemical Co). SQ29,548 (RBI Research Biochemicals International). Concentrations are expressed as final molar concentration (mol/l) in the organ bath. EC,, values (concentration of drugs producing 50 per cent maximal response) were estimated from the concentration-response curves. Data are expressed as means * SEM. Statistical comparisons were performed using Student’s t-test for unpaired and paired observations as appropriate or by one-way analysis of variance and Tukey’s test. A P value of less than 0.05 was considered statistically significant
199
ment with indomethacin. As shown in Figures l(a) and l(b), the responses of the endothelium-denuded rings were similar to the responses of the endothelium-intact rings to serotonin in the presence of indomethacin. Pretreatment of the denuded vesselswith indomethacin (10 pM increased the EC,, for serotonin from 60.2 f 11.0 to 114.8 f 11.0 nM (PcO.05) in placental veins and from 26.3 * 8.0 to 63.9 & 6.0 nM (P
A, in modulating the to serotonin in placental
Efect of SQ29,548
RESULTS Role of prostanoids response to serotonin
in modulating in placental
the pressor vessels
Serotonin induced concentration-dependent contractions in segments of human chorionic arteries and veins. To test the hypothesis that the serotonin-induced increase in tension was due at least in part to stimulation of prostaglandin production, we measured contractile responses before and during the addition of indomethacin to the bathing medium. Exposure of placental veins and arteries with intact endothelium to indomethacin, decreased the responsiveness of the vessels to the vasoconstrictor serotonin. Pretreatment with 10 PM indomethacin increased the EC,, for serotonin from 58.3 f 5.7 to 137.0 f 15.0 nM (PcO.01) in placental veins and from 47.8 * 9.0 to 91.0 rt 4.0 nM (PcO.05) in placental arteries. These changes represent decreases in sensitivity of 2.3- and 1.9-fold, respectively. Similar to the decrease in sensitivity, indomethacin pretreatment decreased the maximal force elicited by serotonin (1 PM). Serotonin-induced maximum contractions of intact-endothelium placental veins and arteries pretreated with indomethacin decreased from 105.0 & 8.0 to 65.0 & 9.5 (PcO.01) and from 85.0 f 8.0 to 5.2 i 6.2 per cent of maximal KCl-induced contractions, respectively. Additional experiments were performed to determine whether the presence of endothelium moderates the contractile response of the placental vessels to serotonin after pretreat-
on serotonin contractile responses. The role of thromboxane A, in modifying the in vitro contractile response to serotonin was examined by studying the effect of a highly specific thromboxane receptor blocker, SQ_29,548 which is a potent and long lasting receptor blocker. The presence of SQ?9,548 (1 FM), significantly attenuated the sensitivity (PcO.05) and the maximum contractile response (PcO.05) to serotonin in chorionic veins and arteries with intact endothelium. In the endothelium-denuded placental arteries and veins, pretreatment with SQ29,548 reduced the serotonininduced contraction in a similar way to that obtained in the endothelium-intact vessels [Figure 2(a) and (b)]. In arteries and veins, the potency and the contractile efficacy of serotonin in the presence of SQ29,548 with respect to EC,, and Em,, values were similar. These results suggest that the modulation of serotonin contractile responses in the placental vessels is mediated in part by thromboxane A,, and this effect is not dependent on mediators released from the endothelium. Sensitivity and maximum contractile responses of chorionic arteries and veins to serotonin, in the presence and absence of SQ29,548 and in the presence and absence of endothelium are summarized in Table 1. thromboxane release. In these assays we used isolated perfused placenta because this method maintains physiological conditions more effectively than the ring experiments. The relationship between serotonin-induced increases in perfusion pressure and thromboxane release was studied. We measured the concentration of thromboxane B,, a stable metabolite, in venous effluent of the perfused human placenta
Serotonin-stimulated
Placenta
200
(1997),
Vol. 18
(a) 120 /
II
100
50
0
Log Figure
concentration
of serotonin
(M)
1. Effect
endothelium. experiments.
of indomethacin (10 FM) on the serotonin-induced contractile responses in human chorionic (a) vein and (b) artery rings with and denuded Changes in tension are expressed as per cent of the contractile response to KC1 (75 IIIM. Each point represents the mean i SEM of six to eight “*P
in basal conditions and in the presence of serotonin in the perfusate. At a dose of 0.5 PM, serotonin increased the perfusion pressure from a baseline value of 18.0 + 2.1 to 62.0 f 7.0 and 82.5 + 9.6 mmHg, measured during phases 1 and 2 of serotonin infusion, respectively (Figure 3). During phase 1 of serotonin infusion, the concentration of thromboxane B, decreased, but this effect was not statistically different from the baseline value (P
DISCUSSION The mechanisms involved in maintenance of vascular tone and altered vascular responses to pressor agents during pregnancy are not known. Studies investigating the factors responsible for the modulation of fetoplacental blood flow have been limited. Results of several recent studies from our laboratory indicate that serotonin behaves as one of the most active vasopressor substance of the human fetal-placental circulation in vitro, and this response was specifically and competitively antagonized by ketanserin, a 5-HT, receptor blocker (Gonzilez et al., 1990). In addition, serotonin has the ability to potentiate the contractions evoked by histamine and prostaglandin F,, in placental veins (Cruz et al., 1991; Gonzilez et al., 1993). Also the importance of serotonin in normal and pathologic pregnancies has recently been recognized, and it has been hypothesized that pre-eclamptic hypertension resulted from the local release of
Cruz et al.: Serotonin
and Thromboxane
in Placental
Circulation
201
(a)
endothelium
0 Log concentration
of
SerOtOnin
(M)
2. Effect of SQ29,548 (1 FM) on the serotonin-induced contractile responses in human chorionic vein (a) and artery (b) rings endothelium. Changes in tension are expressed as per cent of the contractile response to KC1 (75 mu). Each point represents the mean f experiments. Significantly different from the respective control rings (*P
platelet serotonin (Zeeman and Dekker, 1992). Serotonin can work either directly or by interacting with vasoconstricting eicosanoids. A number of studies have suggested that serotonin stimulates the synthesis and release of thromboxane A,, and that endogenous prostaglandins may modulate the response of vascular smooth muscle to serotonin (Howarth, Valance and Wilson, 1993). Vasoconstrictor prostaglandins, particularly thromboxane A,, appear to play an important role in the control of vascular reactivity, platelet aggregation and uteroplacental blood flow during normal and pre-eclamptic pregnancies (Dekker and Sibai, 1993; Walsh, Wang and Jesse, 1993). It is known that serotonin-induced vasoconstriction is mediated mainly by 5-HT, receptor types in placental vessels, however, there is some component resistant to ketanserin (Weiner, 1987; Gonzilez et al., 1990). In the current study, we performed a series of experiments to test the hypothesis that the release of vasoconstrictor products of the cycle-oxygenase pathway modulates serotonin-
with and denuded of seven to nine
SEM
induced contractions of the chorionic plate vessels on the fetal side of the human placenta. After cycle-oxygenase inhibition with indomethacin, the constrictor effects of serotonin were reduced, suggesting that synthesis of vasoconstrictor eicosanoids contributes to the response to serotonin. Output of the vasodilator eicosanoids does not appear to modulate the response to any extent. We have demonstrated previously the importance of nitricoxide rather than vasodilator prostaglandins in the maintenance of vascular tone and in attenuate serotonin-induced placental contractile response (Gonzilez et al., 1995). The results of the present experiments also indicate that the responses to serotonin of the endothelium-denuded rings were similar to the responses of the endothelium-intact rings in the presence of indomethacin; to the extent that endotheliumderived factors may influence serotonin-induced contraction of vessels pretreated with indomethacin. This influence is limited, and demonstrate that at least some of the
202
Placenta
Table
1. Effect
of SQ29,548
on serotonin-induced
Endothelium SQ29,548
Present 0
Arteries Serotonin-induced” Serotonin EC,,
(nM)
Veins Serotonin-induced” Serotonin EC,,
(IIM)
a Force “P~0.05 tP
contractions
exuressed ‘“P
force
90.0 It 5.3 57.0 zt16.0
75.1 ic 6.P 17.0 zt 21.0*
maximal
force
108.2 f 6.0 63.0 + 10.0
87.7 f 5.1m* 26.0 zt 21.0"
as a ner cent of KCl-induced versus column 1 values; versus column 3 values;
maximal
I
Phase 1 tserotonin
Baseline
Phase
Phase 2 (0.5 pM)i
Recovery
1000
750 w” z-
500
2
250
0 ’
1
Phase
2
contractile
force.
with
and without
All
Absent 1 PM
110.0 zt 3.2 35.Ozt 8.0
85.1 * 7.5t 104.0 It 21.0?
data
are mean
18
endothelium
Absent 0
99.3 i 8.2 55.0 h 12.0
Vol.
78.2 zt6.6t-f 127.0 *21.0+ !I
SEM.
n=6. n=6.
Baseline
vessels
Present 1 PM
maximal
100
2fi 8@
in chorionic
(1997),
Recovery
tserotonin (0.5 PM){ Figure 3. Effects of serotonin (0.5 FM) on perfusion pressure and thromboxane B, production in isolated perfused placental cotyledons. Phase 1 and phase 2 represents two serotonin infusion periods (15 min each). After discontinuation of serotonin infusion perfusion pressure and thromboxane B, production remained elevated (recovery). Data are from six individual experiments. P-values are the differences from the baseline measurements.
indomethacin-sensitive contracting factors released by the placental vessels is produced by the smooth muscle. These results are in accord with previous observations in human chorionic plate arteries that exhibit contractile responses to
arachidonic acid suggesting that the cycle-oxygenase pathway in these vessels produces vasoconstrictor substances rather than vasodilators compounds (Sorokin et al., 1992) However, in a similar study of placental arteries, Ezimokhai et al. (1995) did not observe any difference between the contractile response to serotonin alone, and the contractile response to serotonin in the presence of indomethacin. There is no clear reason for these contradictory observations. One possibility is that the higher dose of indomethacin we used, and higher placental vascular sensitivity to serotonin in our study could affect vascular responses. It has been demonstrated that the human placenta produces and releases thromboxane A,, that specific high-affinity receptors for thromboxane A, exist in the placenta (Hedberg et al., 1989) and that thromboxane A, receptor agonists increased resistance in the fetoplacental vasculature (Gude et al., 1992). Furthermore, Sernieri et al. (1983) have demonstrated that in human arteries most of the thromboxane A, production takes place in the media of the smooth muscle. The data is compatible with a the role for thromboxane A, as an endogenous modulator of the fetoplacental circulation. ’ To test if thromboxane, specifically was responsible for the vasoconstriction induced by serotonin, experiments were conducted testing the highly specific thromboxane receptor blocker, SQ?9,548, which is a potent and long lasting receptor blocker. In placental arteries and veins, with and without endothelium, SQJ9,548 decreased the contractile response to serotonin, suggesting that the modulation of serotonin contractile responses in the placental vessels is mediated in part by thromboxane A,. The contraction to serotonin in the presence of SQ29,548 is not significantly altered by removal of the endothelium; therefore, it is independent of mediators derived from this cell type. However, an endothelium-dependent contractile response to serotonin in the rabbit basilar artery has been reported; in this tissue the contractions were susceptible to blockade by indomethacin, suggesting the involvement of a cycle-oxygenase metabolite of arachidonic acid (Seager, Clark and Garland, 1992). Similar data have been obtained that favour thromboxane A, as being the endothelium-dependent
Cruz et al.: Serotonin
and Thromboxane
in Placental
Circulation
vasoconstrictor substance released by SHT, receptor activation in the rat basilar artery (Descombes et al., 1993). To demonstrate further that thromboxane A, is produced and released into the fetal side of the placental circulation, we studied the effects of serotonin on the appearance of thromboxane B,, a stable thromboxane metabolite, in the isolated perfused placental cotyledon. Infusion of serotonin was associated with a 1.9-fold stimulation of thromboxane release in the venous perfusate. As shown in Figure 3, it is important to note that changes in thromboxane release and perfusion pressure do not follow the same pattern, suggesting that thromboxane release is associated with the action of serotonin and not with variations in perfusion pressure. The relationship between serotonin and thromboxane production agrees with recent data from Howarth, Valance and Wilson (1993) who described that
203
serotonin caused the release of thromboxane A, from the isolated perfused placenta obtained from normotensive women. Bjoro et al. (1986) also reported that serotonin increased thromboxane A, production in human umbilical arteries but depressed that of prostacyclin. Although we did not measure the concentration of other cycle-oxygenase products in our experiments, it is likely that there were increases in other classesof cycle-oxygenase derivates. In summary, the present study provides evidence for the participation of thromboxane A, in the contractile response to serotonin in the human placental circulation. The ability of serotonin to release thromboxane A,, which is also a potent vasoconstrictor agent may be important in increase fetoplacental resistance, one of the features of pre-eclampsia.
ACKNOWLEDGEMENTS This research was supported by grant FONDECYT are very grateful to the physicians and midwives cooperation in this study.
1950925 and grants DI 93.33.71-1, 94.33.74-1.3 and 94.33.77.1 1 of the University of Conception, Chile. We of the Service of Obstetrics and Gynecology at the Hospital G. Grant Benavente, Conception, Chile, for their
REFERENCES Benedetto, C., Barbero, M., Rey, L., Zonca, M., Massobrio, M., Rocco, G. & Slater, T. F. (1987) Production of prostacyclin, h-keto-PGF,, and thromboxane B, by human umbilical vessels increases from the placenta towards the fetus. Brituh 3ournal of Obstetrics and Gynaecology, 94, 1165-1169. Bjoro, K. (1986) Prostacyclin and thromboxane formation in human umbilical arteries following stimulation with vasoactive autacoids. Prostuglandins, 31, 6999714. Bjoro, K., Hovig, T., Stokke, K. T. & Stray-Pederson, S. (1986) Formation of prostanoids in human umbilical vessels perfused in vitro. Prostagandins, 31, 683-695. Boura. A. L. A. & Walters, W. A. W. (1991) Autacoids and the control of vascular tone in the human umbilical-placental circulation. Placenta, 12, 453477. Cruz, M. A., Gonz6leq C., Gallardo, V., Lagos, M. & Albornoz, J. (1993) Venous placental reactivity to serotonin in normal and pre-eclamptic gestants. Gynecologic and Obstetric Imxstigation, 36, 148-152. Cruz, M. A., Gonz&lez, C., Septilveda, W. H. & Rudolph, M. I. (1991) Effect of histamine on human placental chorionic veins: interaction with serotonin. Pharmacology, 42, 86-90. De Clerck, F. & de Chaffoy de Courcelles (1989) Amplification mechanisms in platelet activation. In Blood Cells and Arteries in Hypertension and Atherosclerosis (Ed.) Meyer, P. & Marche, P. pp. 105-140. New York: Raven Press. Dekker, G. A. & Sibai, B. M. (1993) Low-dose aspirin in the prevention of pre-eclampsia and fetal growth retardation: rationale, mechanisms, and clinical trials. American 3oumal of Obstetrics and Gynecology, 168, 214-227. Descombes, J. J., Devys, M., Laubie, M. & Verbeuren, T. J. (1993) Endothelial thromboxane production plays a role in the contraction caused by 55hydroxytryptamine in rat basilar arteries. European 3ournal of Pharmacology, 243, 193-199. Ezimokhai, M., Aloamaka, C. P., Osman, N. A., Mensah-Brown, E. P. K. & Morrison, J. (1995) The role of the vascular endothelium in the contractile responses of human chorionic plate artery in pre-eclampsia to prostaglandin FZa, 5-hydroxytryptamine, and potassium chloride. Research zn Experimental Medicine, 195, 171-182. Fox, S. B. & Kbong, T. Y. (1990) Lack of innervation of human umbilical cord. An immuno-histological and histochemical study. Placenta, 11, 59-62. Gonzklez, C., Cruz, M. A., Septilveda, W. H. & Rudolph, M. I. (1990) Effects of serotonin on vascular tone of isolated human placental chorionic veins. Gynecologic and Obstetric Investigation, 29, 88-91. Gonzblez, C., Cruz, M. A., Gallardo, V., Albornoz, J. & Bravo, I. (1993) Serotonin-induced vasoconstriction in human placental chorionic veins:
interaction with prostaglandin F2a. Gynecologic and Obstetrzc Investigation, 35, 86-90. Gonzklez, C., Cruz, M. A., Gallardo, V., Lagos, M., Varela, J., Albornoz, J. & Schulz, C. (1995) Nitric oxide and prostaglandin systems inhibition on the isolated perfused human placenta from normal and pre-eclamptic pregnancies. Gjmecologic and Obstetric Investigation, 40, 244 248. Gude, N. M., Boura, A. L. A., King, R. G., Brennecke, S. P., Jamal, 0. S., Smith, R. & Walters, W. A. W. (1992) Evidence for inhibition by endothelium-derived relaxing factor of thromboxane A, receptor-mediated vasoconstriction in the fetal vessels of the human perfused placenta. Placenta, 13, 5977605. Hedberg, A., Mento, P. F., Liu, E. C., Hollander, A. M. & Wilkes, B. M. (1989) Evidence for functional thromboxane A,-prostaglandin H, receptors in human placenta. American 3ournal of Physiology, 256, E256-E263. Helliwell, T., Smith, P. & Heath, D. (1982) Endothelial pavement patterns in human arteries. 3ournal of Pathology, 136, 227-240. Howarth, S. R., Valance, P. & Wilson, C. A. (1993) The vasoconstrictor effect of endothelin and serotonin on placental vessels is mediated in part by thromboxane A,. B&h 3021ma1 of Pharmacology, 108 (Suppl.), 137. Le. S. Q., Wasserstrum, N., Mombouli, J. V. & Vanhoutte, P. M. (1993) Contractile effect of endothelin in human placental veins: role of endothelium prostaglandins and thromboxane. American Journal of Obstetric and Gynecology, 169, 919-924. Liel, N., Nathan, I., Yermiyahu, T., Zolotov, Z., Lieberman, J. R., Dvilansky, A. & Halushka, P. V. (1993) Increased platelet thromboxane A,/prostaglandin H, receptors in patients with pregnancy induced hypertension. Thrombosis Research, 70, 205-210. Mak, K. K., Gude, N. M., Walters, W. A. & Boura, A. L. (1984) Effects of vasoactive autacoids on the human umbilical-fetal placental vasculature. British Journal of Obstetrics and Gynaecology, 91, 99-106. Meagher, E. A. & Fitzgerald, G. A. (1993) Disordered eicosanoid formation in pregnancy-induced hypertension. Cwculation, 88, 1324-1333. Middelkoop, C. M., Dekker, G. A., Kraayenbrink, A. A. & Popp-Snijders, C. (1993) Platelet-poor plasma serotonin in normal and pre-eclamptic pregnancy. Clinical Chemistry, 39(8), 167551678. Reviriego, J., Fernandez-Alfonso, M. S. & Martn, J. (1990) Actions of vasoactive drugs on human placental vascular smooth muscle. General Pharmacology, 21, 719-727. Seager, J. M., Clark, A. H. & Garland, C. J. (1992) Endotheliumdependent contractile responses to 5-hydroxytryptamine in the rabbit basilar artery. British 3ournal of Pharmacology, 105, 424428. Sernieri, N. G. G., Abbate, R., Gensini, G. F., Panetta, A., Casolo, G. C. & Carini, M. (1983) TXA, production by human arteries and veins. Prostaglandzns, 25, 753-766.
204 Spitz, B., Witters, K., Hanssens, M., Van Assche, F. A. & Keith, J. C. (1993) Ketanserin, a 5HT, serotoninergic receptor antagonist, could be useful in the Hellp Syndrome. Hypertension in Pregnancy, 12(2), 183-190. Sorokin, G., Grant, N. M., Egan, B. M. & Lombard, J. H. (1992) Cyclooxygenase products do not modulate angiotensin II-induced contractions of human chorionic plate arteries. American 3ournal of Obstetrics and GJwcolog3,, 167, 110-114. Walsh, S. W., Wang, Y. A. & Jesse, R. (1993) Peroxides induces vasoconstriction in the human placenta by stimulating thromboxane. American 3ournal of Obstetrics and Gynecology, 169, 1007-1012.
Placenta
(1997),
Vol.
18
Weiner, C. P. (1987) The role of serotonin in the genesis of hypertension in pre-eclampsia. iimericnn 3ournal of Obstetrics and Gynecology, 16, 885 888. Weiner, C. P., Socol, M. L. & Vaisrub, N. (1984) Control of pre-eclamptic hypertension by ketanserin, a new serotonin receptor antagonist. Amwican 3ournal of Obstetrics and Gynecology, 149, 496-500. Zeeman, G. G. & Dekker, G. A. (1992) Pathogenesis of pre-eclampsia: a hypothesis. Clinical Obstetrics and Gynecology, 35, 317-337.
(1997),
18, 197-204
Serotonin-induced Release
and Action
M. A. Cru9,
V. Gallardo”,
a Department
of
b Department
of Obstetrics,
Paper
Vasoconstriction
accepted
Physiology,
27
July
in the Human
P. Miguelb, Faculty Faculty
is Mediated
of of
Biological Medicine,
Fetal-Placental
G. Carrascoa Sciences, University
and University
of Concepckjn,
by Thromboxane Circulation
C. GonzSleza’c of Concepcibn, Concepcibn,
Concepckk,
Chile
Chile
1996
The possibility that prostanoids mediate the contractile response of serotonin on placental vessels was investigated. Rings of chorionic plate arteries and veins with and without endothelium were suspended in an organ bath for recording isometric mechanically activity. Serotonin caused dose-dependent contractions that were significantly attenuated by indomethacin (cycle-oxygenase inhibitor, 10 ,UM) and SQ29,548 (thromboxane receptor antagonist, 1 ,UM). Pretreatment of placental venous and arterial rings with indomethacin decreased sensitivity (EC,,) to serotonin of 2.3- and 1.9-fold, respectively. Pretreatment with SQJ9,548 decreased sensitivity to serotonin of twofold in veins and 2.1-fold in arteries. In the endothelium-denuded placental arteries and veins, pretreatment with indomethacin and SQ29,548 reduced the serotonin-induced contraction in a similar way to that obtained in the endothelium-intact vessels. In isolated perfused cotyledon through the fetal circulation, serotonin caused a significant increase in perfusion pressure and stimulated thromboxane release 1.9-fold compared with basal values. Therefore, serotonin-induced vasoconstriction in the human fetoplacental circulation appears to be mediated in part by thromboxane release or action. This effect is not dependent on mediators released from the endothelium. The present study provides evidence for the participation of thromboxane A, in the contractile response to serotonin in the human placental circulation. The ability of serotonin to release thromboxane A, which is also a potent vasoconstrictor agent, may be important in increase fetoplacental resistance, one of the features of pre-eclampsia. 0 1997 W. B. Saunders Company Ltd Placenta (1997), 18, 197-204
INTRODUCTION
The umbilical-placental circulation is very important for normal fetal growth and viability because it is responsible for the supply of nutrients and oxygen to the fetus. Impairment of this circulation could increase the possibility of growth retardation, fetal distress, and intrauterine fetal death. Fetoplacental blood flow is predominantly controlled by circulating or locally produced autacoids because the placental vasculature lacks innervation (Fox and Khong, 1990). These autacoids appear to interact in many organs to regulate vascular tone, tissue blood flow, and vascular resistance (Boura and Walters, 1991). Previous studies have demonstrated that serotonin produces potent vasoconstrictor responses in placental vesselsby working at specific serotonin receptor sites (Gonzalez et al., 1990; Reviriego, Fernandez-Alfonso and Marin, 1990) and has been implicated in the genesis of pre-eclamptic hypertension (Weiner, 1987; Cruz et al., 1993; Middelkoop et al., 1993). Its role is mainly due to the amplification, through activation of 5-hydroxy-tryptamine serotoninergic receptors, of the action c To whom correspondenceshould be addressedat: Department of Physiology, by) of Biological Sciences,University of Concepci6q P.P. Box 152-C, Concepcibn, Chile. 0143-4004/97/020197+08
$12.00/O
of other agonists on vascular smooth muscle cells and platelets (Cruz et al., 1991; Gonzalez et al., 1993; De Clerck and de Chaffoy de Courcelles, 1989). Serotoninergic blockade has been used with some success in hypertensive disorders of pregnancy (Weiner, Socol and Vaisrub, 1984; Spitz et al., 1993). Serotonin can work either directly or by interacting with vasoconstrictor eicosanoids. Thus, vasoconstrictor products of the cycle-oxygenase pathway have been proposed to modulate serotonin-induced contractions in some peripheral blood vessels (Bjoro, 1986; Howarth, Valance and Wilson, 1993). The human placenta is known to produce large amounts of prostaglandins that may have local actions on the fetoplacental circulation (Benedetto et al., 1987). The eicosanoids, including thromboxane and prostaglandins, have been strongly implicated in the control of fetoplacental circulation under both physiological and pathological conditions (Meagher and Fitzgerald, 1993; Lie1 et al., 1993). Others studies have demonstrated thromboxane B, release into media perfusing the fetoplacental vasculature and vasoconstriction of the fetoplacental vasculature by thromboxane analogues (Bjoro, 1986; Hedberg et al., 1989; Le et al., 1993). To evaluate the potential role of altered serotonin responsiveness and altered production of thromboxane A, in pre-eclampsia, it is important to determine whether these 0
1997 W. B. Saunders
Company
Ltd
Placenta
198
blood flow controllers interact in the normal placental vasculature. In the current study we performed a series of experiments to test the hypothesis that the release of vasoconstrictor products of the cycle-oxygenase pathway modulates serotonininduced contractions of the chorionic plate vesselson the fetal side of the human placenta.
MATERIALS
AND
METHODS
Placentae of uncomplicated pregnancies were collected immediately after vaginal delivery or caesarean section at term with approval of the institutional review boards for the G. Grant Benavente Hospital. Parturients ranged in aged from 20 to 40 years, had no history of chronic medical illness and had normal, term pregnancies and labour. In this study, we found no significant differences (P>O.5) in vascular reactivity between tissues obtained at caesarean section and those at vaginal delivery. The placentae were placed in a normal physiological salt solution (PSS) composition 119.0 mM NaCl, 4.5 rnM KC1 1.2 mM KH,PO,, 1.2 mM MgSO,, 24.9 mM NaHCO,, 1.6 mM CaCl,, 11.5 mM glucose and transported to the Research Laboratory within 30 min of collection.
Chorionsic
ring vessels
Under an illuminated operating magnifiers chorionic plate, arteries and veins of external diameter between 2 and 3 mm were isolated from the placentae, freed of adhering connective tissue, and cut into 2-mm ring segments. Two stainless steel wires were inserted through the vessel ring, one connected to a Grass FT03 force displacement transducer for isometric recording of the contractions on a Grass pen recorder (Model 79 D), and the other was used to adjust the resting tension of the vessel to 1.5 g in veins and 2.0 g in arteries (Gonzalez et al., 1990). The bath content was maintained thermostatically at 37°C pH 7.4, and was constantly bubbled with a gas mixture containing 95 per cent O,, 5 per cent CO,. Endothelium removal was achieved mechanically by gently rubbing the intimal surface of the rings. Because the precontracted intact rings did not relax with either acetylcholine or histamine, the effectiveness of the procedure for the removal of the endothelium was confirmed histologically (Helliwell, Smith and Heath, 1982) at the end of each experiment. After the optimal resting tension was obtained, the tissues were allowed to equilibrate for 1 h. At the beginning and end of each experiment, the maximal contractile response to 75 mM KC1 solution was obtained. Preliminary experiments have shown that 75 mM KC1 evoke the maximal contractile response in these vessels. The maximum response to KC1 was used as standard for comparing responses elicited by other agonists because it is independent of receptors and because the great variation in contractile force generating among isolated vessels. Only if the contractile responses to KC1 were similar in magnitude (with variation less than 10 per cent), were the
(1997,
Vol.
18
data from this particular experiment included in the analysis. In the control group isometric tensions were recorded at increasing cumulative doses of serotonin (0.01-l PM). Rings in the experimental groups were treated with either indomethacin (cycle-oxygenase inhibitor, 10 ELM) or SQ29,548 (thromboxane-receptor antagonist, 1 PM) for 30 min before addition of serotonin. To determine the effect of endothelium removal on the response to serotonin, parallel experiments were performed for two rings with endothelium and for two without endothelium, all from the same placenta. The contraction induced by each concentration of serotonin was expressed as a percentage of the contraction due to KC1 and was used for making the concentration-response curves. The concentration of serotonin eliciting 50 per cent of its own maximum response (EC,,) was determinated graphically for each curve by linear interpolation.
Isolated
perfused
placental
cotyledons
Perfused isolated placental cotyledons were prepared as previously described (Mak et al., 1984). In brief, individual cotyledons supplied by a single chorionic artery and vein were selected for study and the fetal vessels cannulated with polyethylene catheters. The arterial catheter was attached to a T-connector; one limb was attached to a Viggo-Spectramed transducer (model P23XL) for continuous recording of pressure on a Grass polygraph (Grass Instruments) and the other limb was connected to a Masterflex pump (Cole Palmer). A 2-cm piece of rubber tubing was inserted into the arterial perfusion line to permit injection of test agents through syringes mounted in infusion pumps delivering at a constant flow rate. The preparation was placed in a temperaturecontrolled incubator maintained at 37°C throughout the experiment. The fetal compartment was perfused at a rate of 1.2 ml/min with a perfusion medium containing 118 mM NaCl, 4.7 mM KCl, 2.53 mM CaCl,, 1.06 mM .MgSO, 7 H,O, 1.18 mM KH,P04, 25 rnM NaHCO,, 5.55 mM glucose. The perfusate was gassed by bubbling carbogen (95 per cent O,, 5 per cent CO,) through the medium in the reservoirs. Vascular responses were studied for a 60-min stabilization period. The effect of serotonin (0.5 PM) on thromhoxane release was determined. Experiments were performed after a 60-min stabilization. During the experiment, four 15-min intervals perfusate samples were collected, the first correspond to baseline period, the second, at the onset of perfusion pressure increase induced by serotonin infusion (phase 1) the third, when maximum response to serotonin on perfusion pressure was observed (phase 2) and the last one, during the recovery period after the discontinuation of serotonin infusion. Perfusion pressure was monitored through the experiments, and venous effluent was collected on ice into test tubes containing indomethacin (100 ~1 of 10 mM) for the determination of thromboxane B, concentrations. Quantification of thromboxane formation was made by determining the level of thromboxane B, a stable hydrolysed
Cruz et al.: Serotonin and Thromboxane in Placental Circulation
product of unstable thromboxane A, which is derived from prostaglandin H, (PGH, Thromboxane B, was measured by an enzyme linked immunosorbant assay (ELISA) test (Elisa Technologies) in the venous effluent (without extraction). The intra- and interassay coefficients of variation were 3.10 and 5.83 per cent respectively.
Drugs
and
statistics
The different concentrations of drugs used were obtained from stock solutions (1 mM) in physiological saline solution containing 0.01 per cent (w/v) ascorbic acid. All stock solutions were kept at 4°C. Drugs used were 5-hydroxytryptamine creatinine sulphate, indomethacin (Sigma Chemical Co). SQ29,548 (RBI Research Biochemicals International). Concentrations are expressed as final molar concentration (mol/l) in the organ bath. EC,, values (concentration of drugs producing 50 per cent maximal response) were estimated from the concentration-response curves. Data are expressed as means * SEM. Statistical comparisons were performed using Student’s t-test for unpaired and paired observations as appropriate or by one-way analysis of variance and Tukey’s test. A P value of less than 0.05 was considered statistically significant
199
ment with indomethacin. As shown in Figures l(a) and l(b), the responses of the endothelium-denuded rings were similar to the responses of the endothelium-intact rings to serotonin in the presence of indomethacin. Pretreatment of the denuded vesselswith indomethacin (10 pM increased the EC,, for serotonin from 60.2 f 11.0 to 114.8 f 11.0 nM (PcO.05) in placental veins and from 26.3 * 8.0 to 63.9 & 6.0 nM (P
A, in modulating the to serotonin in placental
Efect of SQ29,548
RESULTS Role of prostanoids response to serotonin
in modulating in placental
the pressor vessels
Serotonin induced concentration-dependent contractions in segments of human chorionic arteries and veins. To test the hypothesis that the serotonin-induced increase in tension was due at least in part to stimulation of prostaglandin production, we measured contractile responses before and during the addition of indomethacin to the bathing medium. Exposure of placental veins and arteries with intact endothelium to indomethacin, decreased the responsiveness of the vessels to the vasoconstrictor serotonin. Pretreatment with 10 PM indomethacin increased the EC,, for serotonin from 58.3 f 5.7 to 137.0 f 15.0 nM (PcO.01) in placental veins and from 47.8 * 9.0 to 91.0 rt 4.0 nM (PcO.05) in placental arteries. These changes represent decreases in sensitivity of 2.3- and 1.9-fold, respectively. Similar to the decrease in sensitivity, indomethacin pretreatment decreased the maximal force elicited by serotonin (1 PM). Serotonin-induced maximum contractions of intact-endothelium placental veins and arteries pretreated with indomethacin decreased from 105.0 & 8.0 to 65.0 & 9.5 (PcO.01) and from 85.0 f 8.0 to 5.2 i 6.2 per cent of maximal KCl-induced contractions, respectively. Additional experiments were performed to determine whether the presence of endothelium moderates the contractile response of the placental vessels to serotonin after pretreat-
on serotonin contractile responses. The role of thromboxane A, in modifying the in vitro contractile response to serotonin was examined by studying the effect of a highly specific thromboxane receptor blocker, SQ_29,548 which is a potent and long lasting receptor blocker. The presence of SQ?9,548 (1 FM), significantly attenuated the sensitivity (PcO.05) and the maximum contractile response (PcO.05) to serotonin in chorionic veins and arteries with intact endothelium. In the endothelium-denuded placental arteries and veins, pretreatment with SQ29,548 reduced the serotonininduced contraction in a similar way to that obtained in the endothelium-intact vessels [Figure 2(a) and (b)]. In arteries and veins, the potency and the contractile efficacy of serotonin in the presence of SQ29,548 with respect to EC,, and Em,, values were similar. These results suggest that the modulation of serotonin contractile responses in the placental vessels is mediated in part by thromboxane A,, and this effect is not dependent on mediators released from the endothelium. Sensitivity and maximum contractile responses of chorionic arteries and veins to serotonin, in the presence and absence of SQ29,548 and in the presence and absence of endothelium are summarized in Table 1. thromboxane release. In these assays we used isolated perfused placenta because this method maintains physiological conditions more effectively than the ring experiments. The relationship between serotonin-induced increases in perfusion pressure and thromboxane release was studied. We measured the concentration of thromboxane B,, a stable metabolite, in venous effluent of the perfused human placenta
Serotonin-stimulated
Placenta
200
(1997),
Vol. 18
(a) 120 /
II
100
50
0
Log Figure
concentration
of serotonin
(M)
1. Effect
endothelium. experiments.
of indomethacin (10 FM) on the serotonin-induced contractile responses in human chorionic (a) vein and (b) artery rings with and denuded Changes in tension are expressed as per cent of the contractile response to KC1 (75 IIIM. Each point represents the mean i SEM of six to eight “*P
in basal conditions and in the presence of serotonin in the perfusate. At a dose of 0.5 PM, serotonin increased the perfusion pressure from a baseline value of 18.0 + 2.1 to 62.0 f 7.0 and 82.5 + 9.6 mmHg, measured during phases 1 and 2 of serotonin infusion, respectively (Figure 3). During phase 1 of serotonin infusion, the concentration of thromboxane B, decreased, but this effect was not statistically different from the baseline value (P
DISCUSSION The mechanisms involved in maintenance of vascular tone and altered vascular responses to pressor agents during pregnancy are not known. Studies investigating the factors responsible for the modulation of fetoplacental blood flow have been limited. Results of several recent studies from our laboratory indicate that serotonin behaves as one of the most active vasopressor substance of the human fetal-placental circulation in vitro, and this response was specifically and competitively antagonized by ketanserin, a 5-HT, receptor blocker (Gonzilez et al., 1990). In addition, serotonin has the ability to potentiate the contractions evoked by histamine and prostaglandin F,, in placental veins (Cruz et al., 1991; Gonzilez et al., 1993). Also the importance of serotonin in normal and pathologic pregnancies has recently been recognized, and it has been hypothesized that pre-eclamptic hypertension resulted from the local release of
Cruz et al.: Serotonin
and Thromboxane
in Placental
Circulation
201
(a)
endothelium
0 Log concentration
of
SerOtOnin
(M)
2. Effect of SQ29,548 (1 FM) on the serotonin-induced contractile responses in human chorionic vein (a) and artery (b) rings endothelium. Changes in tension are expressed as per cent of the contractile response to KC1 (75 mu). Each point represents the mean f experiments. Significantly different from the respective control rings (*P
platelet serotonin (Zeeman and Dekker, 1992). Serotonin can work either directly or by interacting with vasoconstricting eicosanoids. A number of studies have suggested that serotonin stimulates the synthesis and release of thromboxane A,, and that endogenous prostaglandins may modulate the response of vascular smooth muscle to serotonin (Howarth, Valance and Wilson, 1993). Vasoconstrictor prostaglandins, particularly thromboxane A,, appear to play an important role in the control of vascular reactivity, platelet aggregation and uteroplacental blood flow during normal and pre-eclamptic pregnancies (Dekker and Sibai, 1993; Walsh, Wang and Jesse, 1993). It is known that serotonin-induced vasoconstriction is mediated mainly by 5-HT, receptor types in placental vessels, however, there is some component resistant to ketanserin (Weiner, 1987; Gonzilez et al., 1990). In the current study, we performed a series of experiments to test the hypothesis that the release of vasoconstrictor products of the cycle-oxygenase pathway modulates serotonin-
with and denuded of seven to nine
SEM
induced contractions of the chorionic plate vessels on the fetal side of the human placenta. After cycle-oxygenase inhibition with indomethacin, the constrictor effects of serotonin were reduced, suggesting that synthesis of vasoconstrictor eicosanoids contributes to the response to serotonin. Output of the vasodilator eicosanoids does not appear to modulate the response to any extent. We have demonstrated previously the importance of nitricoxide rather than vasodilator prostaglandins in the maintenance of vascular tone and in attenuate serotonin-induced placental contractile response (Gonzilez et al., 1995). The results of the present experiments also indicate that the responses to serotonin of the endothelium-denuded rings were similar to the responses of the endothelium-intact rings in the presence of indomethacin; to the extent that endotheliumderived factors may influence serotonin-induced contraction of vessels pretreated with indomethacin. This influence is limited, and demonstrate that at least some of the
202
Placenta
Table
1. Effect
of SQ29,548
on serotonin-induced
Endothelium SQ29,548
Present 0
Arteries Serotonin-induced” Serotonin EC,,
(nM)
Veins Serotonin-induced” Serotonin EC,,
(IIM)
a Force “P~0.05 tP
contractions
exuressed ‘“P
force
90.0 It 5.3 57.0 zt16.0
75.1 ic 6.P 17.0 zt 21.0*
maximal
force
108.2 f 6.0 63.0 + 10.0
87.7 f 5.1m* 26.0 zt 21.0"
as a ner cent of KCl-induced versus column 1 values; versus column 3 values;
maximal
I
Phase 1 tserotonin
Baseline
Phase
Phase 2 (0.5 pM)i
Recovery
1000
750 w” z-
500
2
250
0 ’
1
Phase
2
contractile
force.
with
and without
All
Absent 1 PM
110.0 zt 3.2 35.Ozt 8.0
85.1 * 7.5t 104.0 It 21.0?
data
are mean
18
endothelium
Absent 0
99.3 i 8.2 55.0 h 12.0
Vol.
78.2 zt6.6t-f 127.0 *21.0+ !I
SEM.
n=6. n=6.
Baseline
vessels
Present 1 PM
maximal
100
2fi 8@
in chorionic
(1997),
Recovery
tserotonin (0.5 PM){ Figure 3. Effects of serotonin (0.5 FM) on perfusion pressure and thromboxane B, production in isolated perfused placental cotyledons. Phase 1 and phase 2 represents two serotonin infusion periods (15 min each). After discontinuation of serotonin infusion perfusion pressure and thromboxane B, production remained elevated (recovery). Data are from six individual experiments. P-values are the differences from the baseline measurements.
indomethacin-sensitive contracting factors released by the placental vessels is produced by the smooth muscle. These results are in accord with previous observations in human chorionic plate arteries that exhibit contractile responses to
arachidonic acid suggesting that the cycle-oxygenase pathway in these vessels produces vasoconstrictor substances rather than vasodilators compounds (Sorokin et al., 1992) However, in a similar study of placental arteries, Ezimokhai et al. (1995) did not observe any difference between the contractile response to serotonin alone, and the contractile response to serotonin in the presence of indomethacin. There is no clear reason for these contradictory observations. One possibility is that the higher dose of indomethacin we used, and higher placental vascular sensitivity to serotonin in our study could affect vascular responses. It has been demonstrated that the human placenta produces and releases thromboxane A,, that specific high-affinity receptors for thromboxane A, exist in the placenta (Hedberg et al., 1989) and that thromboxane A, receptor agonists increased resistance in the fetoplacental vasculature (Gude et al., 1992). Furthermore, Sernieri et al. (1983) have demonstrated that in human arteries most of the thromboxane A, production takes place in the media of the smooth muscle. The data is compatible with a the role for thromboxane A, as an endogenous modulator of the fetoplacental circulation. ’ To test if thromboxane, specifically was responsible for the vasoconstriction induced by serotonin, experiments were conducted testing the highly specific thromboxane receptor blocker, SQ?9,548, which is a potent and long lasting receptor blocker. In placental arteries and veins, with and without endothelium, SQJ9,548 decreased the contractile response to serotonin, suggesting that the modulation of serotonin contractile responses in the placental vessels is mediated in part by thromboxane A,. The contraction to serotonin in the presence of SQ29,548 is not significantly altered by removal of the endothelium; therefore, it is independent of mediators derived from this cell type. However, an endothelium-dependent contractile response to serotonin in the rabbit basilar artery has been reported; in this tissue the contractions were susceptible to blockade by indomethacin, suggesting the involvement of a cycle-oxygenase metabolite of arachidonic acid (Seager, Clark and Garland, 1992). Similar data have been obtained that favour thromboxane A, as being the endothelium-dependent
Cruz et al.: Serotonin
and Thromboxane
in Placental
Circulation
vasoconstrictor substance released by SHT, receptor activation in the rat basilar artery (Descombes et al., 1993). To demonstrate further that thromboxane A, is produced and released into the fetal side of the placental circulation, we studied the effects of serotonin on the appearance of thromboxane B,, a stable thromboxane metabolite, in the isolated perfused placental cotyledon. Infusion of serotonin was associated with a 1.9-fold stimulation of thromboxane release in the venous perfusate. As shown in Figure 3, it is important to note that changes in thromboxane release and perfusion pressure do not follow the same pattern, suggesting that thromboxane release is associated with the action of serotonin and not with variations in perfusion pressure. The relationship between serotonin and thromboxane production agrees with recent data from Howarth, Valance and Wilson (1993) who described that
203
serotonin caused the release of thromboxane A, from the isolated perfused placenta obtained from normotensive women. Bjoro et al. (1986) also reported that serotonin increased thromboxane A, production in human umbilical arteries but depressed that of prostacyclin. Although we did not measure the concentration of other cycle-oxygenase products in our experiments, it is likely that there were increases in other classesof cycle-oxygenase derivates. In summary, the present study provides evidence for the participation of thromboxane A, in the contractile response to serotonin in the human placental circulation. The ability of serotonin to release thromboxane A,, which is also a potent vasoconstrictor agent may be important in increase fetoplacental resistance, one of the features of pre-eclampsia.
ACKNOWLEDGEMENTS This research was supported by grant FONDECYT are very grateful to the physicians and midwives cooperation in this study.
1950925 and grants DI 93.33.71-1, 94.33.74-1.3 and 94.33.77.1 1 of the University of Conception, Chile. We of the Service of Obstetrics and Gynecology at the Hospital G. Grant Benavente, Conception, Chile, for their
REFERENCES Benedetto, C., Barbero, M., Rey, L., Zonca, M., Massobrio, M., Rocco, G. & Slater, T. F. (1987) Production of prostacyclin, h-keto-PGF,, and thromboxane B, by human umbilical vessels increases from the placenta towards the fetus. Brituh 3ournal of Obstetrics and Gynaecology, 94, 1165-1169. Bjoro, K. (1986) Prostacyclin and thromboxane formation in human umbilical arteries following stimulation with vasoactive autacoids. Prostuglandins, 31, 6999714. Bjoro, K., Hovig, T., Stokke, K. T. & Stray-Pederson, S. (1986) Formation of prostanoids in human umbilical vessels perfused in vitro. Prostagandins, 31, 683-695. Boura. A. L. A. & Walters, W. A. W. (1991) Autacoids and the control of vascular tone in the human umbilical-placental circulation. Placenta, 12, 453477. Cruz, M. A., Gonz6leq C., Gallardo, V., Lagos, M. & Albornoz, J. (1993) Venous placental reactivity to serotonin in normal and pre-eclamptic gestants. Gynecologic and Obstetric Imxstigation, 36, 148-152. Cruz, M. A., Gonz&lez, C., Septilveda, W. H. & Rudolph, M. I. (1991) Effect of histamine on human placental chorionic veins: interaction with serotonin. Pharmacology, 42, 86-90. De Clerck, F. & de Chaffoy de Courcelles (1989) Amplification mechanisms in platelet activation. In Blood Cells and Arteries in Hypertension and Atherosclerosis (Ed.) Meyer, P. & Marche, P. pp. 105-140. New York: Raven Press. Dekker, G. A. & Sibai, B. M. (1993) Low-dose aspirin in the prevention of pre-eclampsia and fetal growth retardation: rationale, mechanisms, and clinical trials. American 3oumal of Obstetrics and Gynecology, 168, 214-227. Descombes, J. J., Devys, M., Laubie, M. & Verbeuren, T. J. (1993) Endothelial thromboxane production plays a role in the contraction caused by 55hydroxytryptamine in rat basilar arteries. European 3ournal of Pharmacology, 243, 193-199. Ezimokhai, M., Aloamaka, C. P., Osman, N. A., Mensah-Brown, E. P. K. & Morrison, J. (1995) The role of the vascular endothelium in the contractile responses of human chorionic plate artery in pre-eclampsia to prostaglandin FZa, 5-hydroxytryptamine, and potassium chloride. Research zn Experimental Medicine, 195, 171-182. Fox, S. B. & Kbong, T. Y. (1990) Lack of innervation of human umbilical cord. An immuno-histological and histochemical study. Placenta, 11, 59-62. Gonzklez, C., Cruz, M. A., Septilveda, W. H. & Rudolph, M. I. (1990) Effects of serotonin on vascular tone of isolated human placental chorionic veins. Gynecologic and Obstetric Investigation, 29, 88-91. Gonzblez, C., Cruz, M. A., Gallardo, V., Albornoz, J. & Bravo, I. (1993) Serotonin-induced vasoconstriction in human placental chorionic veins:
interaction with prostaglandin F2a. Gynecologic and Obstetrzc Investigation, 35, 86-90. Gonzklez, C., Cruz, M. A., Gallardo, V., Lagos, M., Varela, J., Albornoz, J. & Schulz, C. (1995) Nitric oxide and prostaglandin systems inhibition on the isolated perfused human placenta from normal and pre-eclamptic pregnancies. Gjmecologic and Obstetric Investigation, 40, 244 248. Gude, N. M., Boura, A. L. A., King, R. G., Brennecke, S. P., Jamal, 0. S., Smith, R. & Walters, W. A. W. (1992) Evidence for inhibition by endothelium-derived relaxing factor of thromboxane A, receptor-mediated vasoconstriction in the fetal vessels of the human perfused placenta. Placenta, 13, 5977605. Hedberg, A., Mento, P. F., Liu, E. C., Hollander, A. M. & Wilkes, B. M. (1989) Evidence for functional thromboxane A,-prostaglandin H, receptors in human placenta. American 3ournal of Physiology, 256, E256-E263. Helliwell, T., Smith, P. & Heath, D. (1982) Endothelial pavement patterns in human arteries. 3ournal of Pathology, 136, 227-240. Howarth, S. R., Valance, P. & Wilson, C. A. (1993) The vasoconstrictor effect of endothelin and serotonin on placental vessels is mediated in part by thromboxane A,. B&h 3021ma1 of Pharmacology, 108 (Suppl.), 137. Le. S. Q., Wasserstrum, N., Mombouli, J. V. & Vanhoutte, P. M. (1993) Contractile effect of endothelin in human placental veins: role of endothelium prostaglandins and thromboxane. American Journal of Obstetric and Gynecology, 169, 919-924. Liel, N., Nathan, I., Yermiyahu, T., Zolotov, Z., Lieberman, J. R., Dvilansky, A. & Halushka, P. V. (1993) Increased platelet thromboxane A,/prostaglandin H, receptors in patients with pregnancy induced hypertension. Thrombosis Research, 70, 205-210. Mak, K. K., Gude, N. M., Walters, W. A. & Boura, A. L. (1984) Effects of vasoactive autacoids on the human umbilical-fetal placental vasculature. British Journal of Obstetrics and Gynaecology, 91, 99-106. Meagher, E. A. & Fitzgerald, G. A. (1993) Disordered eicosanoid formation in pregnancy-induced hypertension. Cwculation, 88, 1324-1333. Middelkoop, C. M., Dekker, G. A., Kraayenbrink, A. A. & Popp-Snijders, C. (1993) Platelet-poor plasma serotonin in normal and pre-eclamptic pregnancy. Clinical Chemistry, 39(8), 167551678. Reviriego, J., Fernandez-Alfonso, M. S. & Martn, J. (1990) Actions of vasoactive drugs on human placental vascular smooth muscle. General Pharmacology, 21, 719-727. Seager, J. M., Clark, A. H. & Garland, C. J. (1992) Endotheliumdependent contractile responses to 5-hydroxytryptamine in the rabbit basilar artery. British 3ournal of Pharmacology, 105, 424428. Sernieri, N. G. G., Abbate, R., Gensini, G. F., Panetta, A., Casolo, G. C. & Carini, M. (1983) TXA, production by human arteries and veins. Prostaglandzns, 25, 753-766.
204 Spitz, B., Witters, K., Hanssens, M., Van Assche, F. A. & Keith, J. C. (1993) Ketanserin, a 5HT, serotoninergic receptor antagonist, could be useful in the Hellp Syndrome. Hypertension in Pregnancy, 12(2), 183-190. Sorokin, G., Grant, N. M., Egan, B. M. & Lombard, J. H. (1992) Cyclooxygenase products do not modulate angiotensin II-induced contractions of human chorionic plate arteries. American 3ournal of Obstetrics and GJwcolog3,, 167, 110-114. Walsh, S. W., Wang, Y. A. & Jesse, R. (1993) Peroxides induces vasoconstriction in the human placenta by stimulating thromboxane. American 3ournal of Obstetrics and Gynecology, 169, 1007-1012.
Placenta
(1997),
Vol.
18
Weiner, C. P. (1987) The role of serotonin in the genesis of hypertension in pre-eclampsia. iimericnn 3ournal of Obstetrics and Gynecology, 16, 885 888. Weiner, C. P., Socol, M. L. & Vaisrub, N. (1984) Control of pre-eclamptic hypertension by ketanserin, a new serotonin receptor antagonist. Amwican 3ournal of Obstetrics and Gynecology, 149, 496-500. Zeeman, G. G. & Dekker, G. A. (1992) Pathogenesis of pre-eclampsia: a hypothesis. Clinical Obstetrics and Gynecology, 35, 317-337.