The cyclo-oxygenase-dependent regulation of rabbit vein contraction: evidence for a prostaglandin E2-mediated relaxation

Abstract

1. Arachidonic acid (0.01-1 microM) induced relaxation of precontracted rings of rabbit saphenous vein, which was counteracted by contraction at concentrations higher than 1 microM. Concentrations higher than 1 microM were required to induce dose-dependent contraction of vena cava and thoracic aorta from the same animals. 2. Pretreatment with a TP receptor antagonist (GR32191B or SQ29548, 3 microM) potentiated the relaxant effect in the saphenous vein, revealed a vasorelaxant component in the vena cava response and did not affect the response of the aorta. 3. Removal of the endothelium from the venous rings, caused a 10 fold rightward shift in the concentration-relaxation curves to arachidonic acid. Whether or not the endothelium was present, the arachidonic acid-induced relaxations were prevented by indomethacin (10 microM) pretreatment. 4. In the saphenous vein, PGE2 was respectively a 50 and 100 fold more potent relaxant prostaglandin than PGI2 and PGD2. Pretreatment with the EP4 receptor antagonist, AH23848B, shifted the concentration-relaxation curves of this tissue to arachidonic acid in a dose-dependent manner. 5. In the presence of 1 microM arachidonic acid, venous rings produced 8-10 fold more PGE2 than did aorta whereas 6keto-PGF1alpha and TXB2 productions remained comparable. 6. Intact rings of saphenous vein relaxed in response to A23187. Pretreatment with L-NAME (100 microM) or indomethacin (10 microM) reduced this response by 50% whereas concomitant pretreatment totally suppressed it. After endothelium removal, the remaining relaxing response to A23187 was prevented by indomethacin but not affected by L-NAME. 7. We conclude that stimulation of the cyclo-oxygenase pathway by arachidonic acid induced endothelium-dependent, PGE2/EP4 mediated relaxation of the rabbit saphenous vein. This process might participate in the A23187-induced relaxation of the saphenous vein and account for a relaxing component in the response of the vena cava to arachidonic acid. It was not observed in thoracic aorta because of the lack of a vasodilatory receptor and/or the poorer ability of this tissue than veins to produce PGE2.

Figure 1

Effects of cumulative doses of arachidonic acid on rings of rabbit saphenous vein, vena cava and thoracic aorta submaximally contracted by external application of KCl. Time courses of representative experiments without (a) and with (b) 3 mM TP receptor antagonist (SQ29548 or GR32191B) pretreatment. (c) Summarized results obtained with untreated rings (n=4, 7, 15) and rings pretreated with the TP receptor antagonist (n=40, 21, 25 for saphenous vein, vena cava and thoracic aorta respectively). Results are expressed as means±s.e.mean of percentage of KCl precontraction.

Figure 2

Effects of endothelium removal and/or cyclo-oxygenase inhibition on the effects of arachidonic acid on rabbit saphenous vein, vena cava and thoracic aorta preincubated with 3 mM TP receptor antagonist (SQ29548 or GR32191B). Intact rings (E+) and endothelium-free rings (E−) were incubated without (Control) or with 10 mM indomethacin (Indo). Results are expressed as means±s.e.mean of 8–40 rings depending on the experimental conditions for saphenous vein, 7–21 rings for vena cava and 9–25 rings for thoracic aorta.

Figure 3

Effects of PGE₂ and EP₄ antagonist receptor, AH23848B, on the arachidonic acid response of saphenous vein. (a) Effects of cumulative doses of PGE₂, PGI₂ and PGD₂ on precontracted saphenous vein (SV, n=12–15) and cumulative doses of PGE₂ and PGI₂ on precontracted vena cava (VC, n=7–9) and thoracic aorta (TA, n=8–14). Experiments were performed on rings without endothelium in the presence of 3 μM TP receptor antagonist. Results are expressed as the means±s.e.mean. (b) Effects of pretreatment with 30 μM and 100 μM EP₄ antagonist receptor, AH23848B, or vehicle on the effects of cumulative doses of arachidonic acid on precontracted saphenous vein. Experiments were performed on rings with endothelium in the presence of 3 μM TP receptor antagonist. Results are expressed as means±s.e.mean of percentage of maximal relaxation, n=8–40 depending on the experimental conditions.

Figure 4

Measurements of PGE₂, 6-ketoPGF_1α and TXB₂ production by rings isolated from saphenous vein (SV), vena cava (VC) and thoracic aorta (TA) in basal conditions (a) and stimulated by the addition of 1 μM arachidonic acid (b). Open columns represent production by intact rings (E+) and hatched columns illustrate production by endothelium-free rings (E−). Results are expressed as means±s.e.mean of percentage of 8–35 rings depending on the experimental conditions. ^***, P<0.001.

Figure 5

Effects of cumulative doses of A23187 on KCl precontracted intact or endothelium-free rings of saphenous vein rings. Time courses of representative experiments performed without (a) and with concomitant pretreatment with NO synthase inhibitor (L-NAME, 100 μM) and cyclo-oxygenase inhibitor (indomethacin, 10 μM) (b, left) and only with cyclo-oxygenase inhibitor (indomethacin, 10 μM) (b, right). (c) Results obtained with untreated rings (n=32–43), and rings pretreated with indomethacin (n=8–16) and/or L-NAME (n=11–25).