1 The mechanism underlying the foetal toxicity induced by captopril is not well understood. Since bradykinin and angiotensin II appear to be important in the.
Br J clin Pharmac 1995; 39: 497-501
Effects of captopril on the human foetal placental circulation: an interaction with bradykinin and angiotensin I R. SOARES DE MOURA & M. A. CERQUEIRA LOPES Departamento de Farmacologia e Psicobiologia, IB-HUPE, Centro Biomedico, Universidade do Estado do Rio de Janeiro, Rio de Janeiro, Brasil
1 The mechanism underlying the foetal toxicity induced by captopril is not well understood. Since bradykinin and angiotensin II appear to be important in the regulation of the placental circulation, experiments were performed to assess the effects of captopril on the vascular actions of these peptides on the human foetal placental circulation. 2 Full-term human placentas, obtained from normal pregnancy, were perfused with a modified Tyrode solution bubbled with 02 using a pulsatile pump. The placental perfusion pressure was measured with a Statham pressure transducer and recorded continuously on a Hewlett-Packard polygraph. 3 Bradykinin (0.1, 0.3 and 1.0 nmol) injected into the placental arterial circulation produced an increase in placental perfusion pressure in all experiments. This effect of bradykinin was significantly inhibited by indomethacin (3 x 10-7 M). 4 Captopril (10-7 M) significantly potentiated the pressor effect of bradykinin on the human placental circulation (n = 6). This effect of captopril was reversed by indomethacin (3 x 10-7 M). 5 Angiotensin I (n = 6) and angiotensin II (n = 6), injected into the placental arterial circulation, both produced dose-dependent increases in placental perfusion pressure. The dose-response curves to angiotensin I (n = 6) were significantly displaced to the right by captopril in a concentration-dependent manner. 6 We suggest that the toxic effects of captopril on the foetus, rather than reflecting an inhibition of angiotensin II formation, may instead be related to a potentiation of the vasoconstrictor effect of bradykinin on the foetal placental circulation, thereby reducing blood flow and causing foetal damage. The reasons for this are discussed. 7 In addition, our results also suggest that the vasoconstrictor effects of bradykinin may be due to a release of vasoconstrictor prostanoid substances.
Keywords human foetal placental circulation captopril indomethacin
bradykinin
angiotensin I
angiotensin II
Introduction
Administration of captopril during pregnancy in experimental animals is associated with an increase in stillbirths and a high incidence of intrauterine death [1, 2]. Currently captopril is contraindicated in human pregnancy since many reports in the literature have shown adverse toxic effects on the foetus [3]. For example, intrauterine growth retardation, respira-
tory and circulatory abnormalities and patent ductus arteriosus have been reported to occur in foetuses born from mothers treated with captopril during pregnancy [4-7]. The mechanism underlying the toxic effect of captopril on the foetus is not yet completely established, however, a decreased maternal and foetal placental
Correspondence: Professor R. Soares de Moura, Departamento de Farmacologia, Centro Biomedico, Universidade do Estado do Rio de Janeiro, Av. 28 de Setembro, 87, 20.551-030 Rio de Janeiro, Brasil © 1995 Blackwell Science Ltd
497
498
R. Soares de Moura & M. A. Cerqueira Lopes
blood flow appears to be important [8, 9]. Bradykinin and angiotensin II are endogenous vasoactive peptides that exert an important modulatory influence on the placental circulation. In order to assess whether the toxic effects of captopril during pregnancy might be related to an interaction with these peptides, we decided to investigate the effect of captopril on the vascular effects of bradykinin and angiotensins I and II in the human isolated foetal circulation.
Methods
Twenty-four full-term human placentas (430-720 g) were obtained from spontaneous deliveries at the Hospital Universitario Pedro Ernesto in Rio de Janeiro. Placentas from mothers with eclampsia, hypertension, diabetes, Rh incompatibility, or other overt disease were excluded from this study. Institutional approval to use the placentas was given. All placentas were studied on the day of delivery. Immediately after delivery each placenta was immersed in warm saline to maintain temperature and to prevent drying. A cannula was inserted into the umbilical artery (the other umbilical artery was tied) and into the umbilical vein at the placental plate, and blood was washed out of the vascular tree by gentle injection of 100 to 200 ml of warm Tyrode solution containing heparin (Liquemin Roche 0.02 mg ml-'). The placenta was then transported directly to our laboratory, placed inside an opened plastic bag full of warm Tyrode solution and immersed in a constanttemperature (370 C) bath. The arterial cannula was perfused using a pulsatile pump and a modified Tyrode solution (mM composition: 139.9 NaCl, 2.68 KCI, 2.5 CaCl2, 1.05 MgSO4, 0.43 NaH2PO4, 11.9 NaHCO3, 0.026 EDTA, 5.5 glucose) bubbled with 02 (pH 7.4; 370 C). The interval between delivery and the establishment of tissue perfusion was no longer than 1 h. The venous flow was discarded and not reperfused. The placental perfusion pressure was measured with a Statham pressure transducer (P23Db), inserted between the pump and the arterial cannula, and recorded continuously on a Hewlett-Packard polygraph. The pump frequency in all experiments was fixed at 110 strokes min-m and the stroke volume was adjusted at the beginning of the experiment to give a mean pressure of about 40 mm Hg. Bradykinin (0.1; 0.3 and 1.0 nmol), angiotensin I (0.03 to 100 nmol) and angiotensin II (0.03 to 3 nmol) were injected into the rubber tubing close to the arterial cannulae after an equilibration period of 1 h. In order to ensure a complete recovery of the tissue, a period of 45 min and 15 min was allowed between consecutive doses of bradykinin and angiotensins I and II, respectively. Injections of bradykinin were performed before, during exposure to captopril and during exposure to captopril plus indomethacin together (Group 1, n = 6). Injections of bradykinin were performed before and during exposure to indomethacin (Group 2, n = 6). Dose-responses to
angiotensin I were obtained before and during exposure to captopril (Group 3, n = 6). Dose-responses to angiotensin II were obtained before (n = 6) and during exposure to captopril (n = 3) (Group 4). Captopril and/or indomethacin were added to the reservoir of perfusion Tyrode solution. The following drugs were used: bradykinin, angiotensin I, angiotensin II, indomethacin and captopril (Sigma Chemical Co., St Louis, Missouri, USA). The data were statistically evaluated using Mann-Whitney test. Values were considered to differ significantly when the probability was less than 0.05. The vasoconstrictor effects of bradykinin, angiotensin I and angiotensin II were expressed as increases in mm Hg from basal placental perfusion pressure obtained immediately prior to injection of the drug. The results are presented as mean ± s.e. mean; n represents the number of placentas studied for a particular group.
Results
The mean placental perfusion pressure and the mean placental flow at the end of the equilibration period were respectively 35.3 ± 2.3 mm Hg and 33.8 ± 3.9 ml min-' (n = 24).
Group I Perfusion of the placental circulation with captopril (10-7 M) did not significantly change the placental perfusion pressure, however, indomethacin 3 x 10-7 M induced a reduction in the placental perfusion pressure of 8.1 ± 2.0 mm Hg. In all experiments bradykinin produced a dose-related vasoconstriction, as indicated by an increase in placental perfusion pressure, and at no time were vasodilator effects of bradykinin observed. The bradykinin-induced vasoconstriction was slowly developing and decayed in a similar manner, in contrast to the more rapid response elicited by angiotensins I and II (Figure 1). This pressor response to bradykinin was significantly potentiated (P = 0.002) in the presence of captopril 10-7 M (n = 6). In addition, indomethacin (3 x 10- M, n = 6) significantly reduced (P = 0.005) the captopril induced potentiation of the pressor effect of bradykinin (Figure 2).
Group 2 As in group 1, bradykinin (n = 6) induced only a dose-related increase in placental perfusion pressure. Indomethacin 3 x 10-7 M (n = 6) induced a reduction in the placental perfusion pressure of 9.1 ± 2.7 mm Hg, and significantly reduced (P < 0.005) the pressor effect of bradykinin (Figure 3).
Group 3 In all experiments angiotensin I (n = 6) produced dose-dependent increases in placental perfusion pressure with no concomitant vasodilation (Figures 1 and
C 1995 Blackwell Science Ltd, British Journal of Clinical Pharmacology, 39, 497-501
Effect of captopril on the human foetal placental circulation 45;-v
25
499
-_|
k Al 0.03 -
0.1
45
k
A
A
0.3
1.0
3.0
JX~.. i1_A_9 All
A
A
0.03
0.1
a
A 0.3
A 3.0
1.0
45
25
BK
A
A
A
0.1
1.0
0.3
4 min
Figure 1 Typical recordings showing the effects of angiotensin I, angiotensin II and bradykinin on the perfusion pressure (mm Hg) of the human isolated foetal placenta. Upper, medium and lower recordings are from three different placentas. The arrows represent the moment of injection. Doses in nmol.
20
E Ej (L15 CL
E
~20-
o_ 15-
-
00L
10 0.
0
a)
0
0 0.1 0.3 1.0
0.1 0.3 1.0
0.1 0.3 1.0
0.1
Bradykinin (nmol)
0.3
1.0
0.1
0.3
1.0
Bradykinin (nmol)
Figure 2 Mean increase in placental perfusion pressures induced by injections of three doses of bradykinin before (O), during perfusion with capto,ril 10-7 M (U) and during Serfusion of captopril 10- M plus indomethacin 3 x 10 M (0 ).
Figure 3 Mean increase in placental perfusion pressure induced by injections of three doses of bradykinin before (LI) and during perfusion with indomethacin 3 x 10M (0).
4). In the presence of captopril (4 x 10-9 and 4 x 10-7 M) the dose-response curve to angiotensin I was significantly displaced to the right (P < 0.005) in a concentration-dependent manner (Figure 4).
4). The effects of angiotensin II were not changed by captopril 4 x 10-9 M (n = 3).
Group 4
Discussion
In all experiments angiotensin II (n = 6) produced dose-dependent increases in placental perfusion pressure with no concomitant vasodilation. The pressor effect of angiotensin II was significantly greater (P < 0.005) than that of angiotensin I (Figures 1 and
The mechanism(s) underlying the toxic effects of captopril on the foetus is not currently established, however, the deleterious effects of captopril may be related to an alteration of blood flow within the maternal or foetal placental circulations. Current
© 1995 Blackwell Science Ltd, British Journal of Clinical
Pharmacology, 39, 497-501
500
R. Soares de Moura & M. A. Cerqueira Lopes 15 r
0)
I
E E
1o0
00-
cn 0 C
51-
0
0.01
0.1
1
10
100
Angiotensin l/ll (nmol) Figure 4 Mean dose-response curves (increase in placental perfusion pressure in mm Hg) for angiotensin I and angiotension II. Each point represents the mean result of 6 experiments. Vertical bars represent s.e. mean. O angiotension II; 0 angiotensin I before captopril; V angiotension I during 4 x 10-9 M captopril; V angiotensin I during 4 x 10-7 M of captopril.
evidence suggests that vasoactive peptides such as angiotensin II and bradykinin play an important modulatory role in the control of uterine and placental blood flow during pregnancy. Thus a normal placental function, that relies on the interaction between the maternal and foetal placental circulations, will be influenced by such endogenous substances. Since the therapeutic pharmacological action of captopril depends upon the inhibition of kininase II enzyme (Angiotensin Converting Enzyme; EC3.4.15.1), it is possible that its foetal toxic effect may be related to a reduction of angiotensin II formation and/or potentiation of bradykinin actions, causing an imbalance of the mechanisms controlling maternal and foetal placental blood flow. At present it is not known whether such damaging effects of captopril on the foetus are due to a vascular effect that occurs on the maternal or the foetal side, or indeed if on both. The renin-angiotensin system, which is activated during normal pregnancy, seems to play an important role in the control of maternal placental blood flow [10]. Angiotensin II may be involved in the maintenance of high uterine blood flow during pregnancy since inhibition of angiotensin-converting enzyme or blockade of angiotensin II receptors with saralasin induces a reduction of uterine blood flow in pregnant rabbits [11], whilst infusion of angiotensin II in the pregnant animal induces an increase in uterine placental blood flow [12-14]. Captopril-induced foetal toxicity could therefore be explained by a
decrease in angiotensin II levels leading to a reduction of uterine blood flow [11] and consequent foetal oxygen deprivation [8]. Similarly, the toxic effect of captopril could also be a consequence of the decreased formation of angiotensin II on the foetal side of the placental circulation. It has been shown that acute administration of captopril to pregnant ewes induces a decrease in foetal placental blood flow and an increase in foetal placental vascular resistance [9], and these observations may provide an important indication of the mechanism underlying captopril-induced foetal toxicity. For example, it appears unlikely that a reduction of angiotensin II formation in the foetal circulation would be responsible for these changes. Our results have shown that captopril, in concentrations that are similar to the plasma levels observed in patients treated with this compound (50 ng ml-') [15] inhibits the vasoconstrictor effect of angiotensin I on the human foetal circulation, an action that would be expected to reduce foetal placental resistance and to increase blood flow to the foetal placenta during pregnancy and therefore be beneficial. It is possible instead that the toxic effects of captopril on the foetus may be related to an interaction with the vascular actions of bradykinin, the other peptide influenced by the inhibition of kininase II. Interestingly, during parturition the bradykinin plasma concentration increases approximately six-fold [16] suggesting an important physiological role for this peptide during pregnancy and delivery. Our results confirm the previous findings that bradykinin induces constriction of human foetal placental vessels [17-19], although at present the type of receptor and mechanism is not known. Since the pressor responses to bradykinin in our study were slow in onset, compared with the much more rapid vasoconstrictions observed with angiotensin II, and additionally required considerable time between subsequent doses to avoid tachyphylaxis, it is likely that the effects of bradykinin in the human isolated placental circulation involve the formation and release of other vasoactive substances. The vascular effect of bradykinin observed in experimental animals may involve the participation of arachidonic acid metabolism since it is dependent on the synthesis and release of prostaglandins [20]. In the present study we found that the vasoconstrictor effect of bradykinin on the human placental circulation was significantly reduced by indomethacin, an inhibitor of cyclo-oxygenase, suggesting the involvement of prostanoids in its action. This is in agreement with the previous observations that bradykinin induces release of thromboxane B2 from human placental vessels [19]. However, the vasoconstrictor effect of bradykinin on the human placental circulation may also involve the release of PGF2a, a substance that induces vasoconstriction of isolated human placental vessels (Soares de Moura, unpublished results), and/or other prostanoids synthesized, via the cyclo-oxygenase pathway such as prostaglandin H2, a putative mediator of endotheliumdependent contraction [21].
© 1995 Blackwell Science
Ltd, British Journal of Clinical Pharmacology, 39, 497-501
Effect of captopril on the human foetal placental circulation Whereas many reports have shown that the vasodilator effect of bradykinin is potentiated by captopril [22, 23] studies concerning the effect of kininase II inhibitors on the vasoconstrictor action of bradykinin on human blood vessels are scant. Our present results have clearly demonstrated that captopril potentiates the vasoconstrictor effects of bradykinin on the placental vascular bed. Interestingly, such an action may also explain the previously reported reduction in blood flow induced by captopril in pregnant ewes [9]. It is thus probable that at least some of the deleterious effects of captopril on the foetus may be mediated via an augmentation of the constriction of foetal placental blood vessels induced by endogenous bradykinin. These effects are unlikely to be due to
501
actions of bradykinin on the maternal placental circulation since bradykinin actually increases uterine and placental blood flow [24, 25]. In conclusion, we suggest that the foetal toxic effects of captopril may at least be partly due to a reduction of foetal placental blood flow via a potentiation of the vasoconstrictor effect of endogenous bradykinin, an effect likely to be mediated via the release of prostanoids. We wish to thank Dr David N. Criddle for reviewing the manuscript and we are grateful to the physicians and nurses of the Department of Obstetrics at the Hospital Universitario Pedro Ernesto for their contribution to this project. This work was supported in part by CNPq.
References 1 Broughton Pipkin F, Turner SR, Symonds EM. Possible risk with captopril in pregnancy. Some animal data. Lancet 1980; i: 1256. 2 Keith IM, Will JA, Weir EK. Captopril: Association with fetal death and pulmonary vascular changes in the rabbit. Proc Soc exp Biol Med 1982; 170: 378-383. 3 Shotan A, Widerhorn J, Hurst A, Elkayam U. Risks of angiotensin-converting enzyme inhibition during pregnancy: Experimental and clinical evidence, potential mechanisms, and recommendation for use. Am J Med 1994; 93: 451-456. 4 Millar JA, Morrison NWD. Management of severe hypertension in pregnancy by a combined drug regimen including captopril. Case report. NZ med J 1983; 96: 796-798. 5 Boutroy MJ, Vert P, Ligny BH, Miton A. Captopril administration in pregnancy impairs fetal angiotensin converting enzyme activity and neonatal adaptation. Lancet 1984; i: 935-936. 6 Fiochi R, Lijnen P, Fagard R, Staessen J, Amery A, Van Assche F, Spitz B, Rademaker M. Captopril during pregnancy. Lancet 1984; ii: 1153. 7 Caraman PL, Milton A, De Ligny BH, Kessler M, Boutroy MJ, Schweitzer M, Brocard 0, Ragage JP, Netter P. Grossesses sous captopril. The'rapie 1984; 39: 59-62. 8 Binder ND, Faber JJ. Effects of captoril on blood pressure, placental blood flow and uterine oxygen consumption in pregnant rabbits. J Pharmac exp Ther 1992; 260: 294-299. 9 Lumbers ER, Kingsford NM, Menzies RI, Stevens AD. Acute effects of captopril, an angiotensin-converting enzyme inhibitor, on the pregnant ewe and fetus. Am J Physiol 1992; 262: R754-R760. 10 Broughton Pipkin F, Smalers ORC. A study of factors affecting blood pressure and angiotensin II in newborn infants. J Pediatrics 1977; 91: 113-119. 11 Ferris TF, Weir EK. Effect of captopril on uterine blood flow and prostaglandin E synthesis in the pregnant rabbit. J clin Invest 1983; 71: 809-815. 12 Ferris TF, Stein JH, Kauffman J. Uterine blood flow and uterine renin secretion. J clin Invest 1972; 51: 2827-2833. 13 Albertini R, Seino M, Scicli AG, Carretero OA. Uteroplacental renin in regulation of blood pressure in the
pregnant rabbit. Am J Physiol 1980; 239: H266-H271. 14 Rosenfeld CR, Naden RP. Uterine and nonuterine responses to angiotensin II in bovine pregnancy. Am J Physiol 1988; 257: H17-H24. 15 Bennet LZ, Williams RL. Design and optimization of dosage regimens; Pharmacokinetics data. In The pharmacological basis of therapeutics, eighth edition, eds Goodman L, Gilman A, p. 1662. Pergamon Press, New York, 1990. 16 Melmon KL, Cline MH, Hughest T, Nies AS. Kinins: possible mediators of neonatal circulatory changes in man. J clin Invest 1968; 47: 1295-1302. 17 Tulenko TN. Regional sensitivity to vasoactive polypeptides in the human umbilicoplacental vasculature. Am J Obstet Gynecol 1979; 135: 629-636. 18 Soares de Moura R, Solano Vale N. Effect of captopril on bradykinin inactivation by human foetal placental circulation. Br J clin Pharmac 1986; 21: 143-148. 19 Wilks BM, Mento PF. Bradykinin-induced vasoconstriction and thromboxane release in perfused human placenta. Am J Physiol 1988; 254: E681-E686. 20 Regoli D, Barabe J. Pharmacology of bradykinin and related kinins. Pharmac Rev 1980; 32: 1-46. 21 Luscher TF, Boulanger CM, Dohi Y, Yang Z. Endothelium-derived contracting factors. Hypertension 1992; 19: 117-130. 22 Feletou M, Germaine M, Teisseire B. Converting-enzyme inhibitors potentiate bradykinin-induced relaxation in vitro. Am J Physiol 1992; 262: H839-H845. 23 Pontieri V, Lopes OU, Ferreira SH. Hypotensive effect of captopril. Role of bradykinin and prostaglandin like substances. Hypertension 1990; 15 (suppl): 155-158. 24 Still JG, Greis FC. The effect of prostaglandins and other vasoactive substances on uterine blood flow and myometrial activity. Am J Obstet Gynecol 1978; 130: 1-8. 25 Seino M, Albertini R, Scicli AG, Carretero OA. Role of kinins in the effect of converting enzyme inhibitor (SQ 20,881) on utero-placental blood flow. Fed Proc 1979; 38: 685.
C) 1995 Blackwell Science Ltd, British Journal of Clinical Pharmacology, 39, 497-501
(Received 15 September 1994, accepted 21 December 1994)
