Cyclooxygenase 1-derived prostaglandin E2 and EP1 receptors are required for the cerebrovascular dysfunction induced by angiotensin II

Abstract

Prostaglandin E(2) (PGE(2)) EP1 receptors (EP1Rs) may contribute to hypertension and related end-organ damage. Because of the key role of angiotensin II (Ang II) in hypertension, we investigated the role of EP1R in the cerebrovascular alterations induced by Ang II. Mice were equipped with a cranial window, and cerebral blood flow was monitored by laser-Doppler flowmetry. The attenuation in cerebral blood flow responses to whisker stimulation (-46+/-4%) and the endothelium-dependent vasodilator acetylcholine (-40+/-4%) induced by acute administration of Ang II (250 ng/kg per minute; IV for 30 to 40 minutes) were not observed after cyclooxygenase 1 or EP1R inhibition or in cyclooxygenase 1 or EP1-null mice. In contrast, cyclooxygenase 2 inhibition or genetic inactivation did not prevent the attenuation. Ang II-induced oxidative stress was not observed after cyclooxygenase 1 or EP1R inhibition or in EP1R-null mice. Prostaglandin E(2) reinstated the Ang II-induced cerebrovascular dysfunction and oxidative stress after cyclooxygenase 1 inhibition. Brain prostaglandin E(2) levels were not increased by Ang II but were attenuated by cyclooxygenase 1 and not cyclooxygenase 2 inhibition. The cerebrovascular dysfunction induced by 14-day administration of "slow-pressor" doses of Ang II (600 ng/kg per minute) was attenuated by neocortical application of SC51089. Cyclooxygenase 1 immunoreactivity was observed in microglia and EP1R in endothelial cells. We conclude that the cerebrovascular dysfunction induced by Ang II requires activation of EP1R by constitutive production of prostaglandin E(2) derived from cyclooxygenase 1. The findings provide the first evidence that EP1Rs are involved in the deleterious cerebrovascular effects of Ang II and suggest new therapeutic approaches to counteract them.

Figure 1

Effect of AngII on the CBF increase produced by whisker stimulation (A) or acetylcholine (B) in wild type (WT) mice with neocortical superfusion of SC51089 or in EP1-null mice, with or without PGE₂. WT (+ or −) and EP1−/− (+ or −) refer to the genotype of the mice. Mean±SEM. *p<0.05 from vehicle; analysis of variance and Tukey’s test; n=5/group.

Figure 2

Effect of 14 day administration of AngII via osmotic minipumps on the increase in CBF induced by whisker stimulation (A) or acetylcholine (B) in mice with neocortical superfusion of vehicle or SC51089. *p<0.05 from vehicle; ^#p<0.05 from AngII vehicle; n=5/group.

Figure 3

Effect of AngII on the CBF increase produced by whisker stimulation (A) or acetylcholine (B) in mice with neocortical superfusion of NS398, SC560, PGE₂ and PGF_2α. *p<0.05 from vehicle; ^#p<0.05 from NS398 vehicle; n=5/group.

Figure 4

Effect of AngII on the CBF increase produced by whisker stimulation (A) or acetylcholine (B) in wild type (WT) or NOX2−/− mice with or without neocortical superfusion with PGE₂. WT (+ or −) and NOX2−/− (+ or −) refer to the genotype of the mice. *p<0.05 from vehicle; n=5/group.

Figure 5

Effect of Ang II on ROS production assessed by in situ HE microfluorography. A: Mice treated with neocortical superfusion of SC51089 and/or PGE₂. B: Mice treated with NS398, SC560 and PGE₂ or PGF_2α. *p<0.05 from vehicle; n=5/group.

Figure 6

In somatosensory cortex, COX-1 immunoreactivity (red) is co-localized with the microglial marker Iba-1 (green)(top row), but not with the endothelial marker CD31 (green) (middle row, arrows). EP1 receptor (EP1R) immunoreactivity (green) is co-localized with CD31 (red) (lower row, arrows). The cortical surface is at the top of the panels. The second, third and fourth panels in each row represent enlargements of the square in the first panel. Calibration bar: 10μm.