Cerebral blood flow is regulated locally to match the continuously changing metabolic demands of the active neuropil. Recent studies indicate that astrocytes may act as an intermediary stage in the transmission of vasodilator signals from neurones to cerebral microvessels. Epoxyeicosatrienoic acids (EETs), produced by breakdown of arachidonic acid in astrocytic membranes, have been shown to dilate isolated cerebral blood vessels (Alkayed et al. 1996a). Studies in vivo have indicated that tonic release of EETs may provide a level of dilator tone within the cerebral circulation (Alkayed et al. 1996b). However, this finding is equivocal (Bhardwal et al. 2000) and it is possible that anaesthesia is a confounding factor. We have therefore used a brain slice preparation in which responsiveness of individual intraparenchymal arterioles can be studied in situ in the neuropil in the absence of anaesthesia.

Coronal slices of cerebral cortex 350 µm thick were prepared from urethane-anaesthetised (1.5 g Kg^-1 20 % solution) 150-250 g male Wistar rats. Slices were maintained at 31-33 °C in artificial cerebrospinal fluid containing (mM): NaCl, 125.8; KCl, 3.1; MgSO₄, 1.3; CaCl₂, 2.4; NaHCO₃, 6.0; D-glucose 10.0, gassed with 95 % O₂-5 % CO₂ to pH 7.35-7.45. Images of cerebral arterioles 9.3 ± 0.4 µm (mean ± S.E.M.) internal diameter (ID), 30-70 µm below the surface of the slice, were captured using a CCD camera and Openlab image analysis software (Improvision Ltd). A low level of rhythmic contractile activity (vasomotion, 2.45 ± 0.7 min^-1) was observed.

Addition of miconazole, a cytochrome P450 epoxygenase inhibitor (20 µM for 5 min), reduced ID by 17.9 ± 2.9 % and increased vasomotion (3.9 ± 1.3 to 7.8 ± 1.5 contractions min^-1, n = 8). In four further vessels already preconstricted and showing vasomotion in the presence of a thromboxane A₂ agonist, U46619 (75 nM), addition of AMPA (1 µM for 5 min) to stimulate astrocytes and neurones, produced dilatation (ID increased from 11.8 ± 28 µm to 13.1 ± 2.8 µm) and a decrease in vasomotion (7.9 ± 1.4 to 2.3 ± 1.0 min^-1). In the presence of 20 µM miconazole, each vessel showed a further small constriction (decrease in ID from 11.4 ± 3.0 µm to 9.6 ± 2.9 µm) and increased vasomotion (8.7 ± 1.5 to 10.2 ± 2.0 min^-1) but AMPA still produced dilatation (ID increased to 11.7 ± 2.6 µm) and inhibited vasomotion to 0.8 ± 0.5 min^-1.

The results suggest that tonic release of cyctochrome P450 metabolites of arachidonic acid may participate in regulating basal cerebrovascular tone. However, these substances do not appear to be involved in mediating AMPA-evoked dilatation.

This work was supported by the British Heart Foundation.