Mechanisms of vascular reactivity during cortical spreading depression (CSD) and cortical spreading ischemia (CSI)

Introduction

The cerebral blood flow response to CSD in vivo in the rat can be inverted to severe vasoconstriction (CSI) by nitric oxide synthase (NOS) inhibition and elevated potassium concentration ¹ . In addition we have recently shown that CSI in vivo can be modelled in vitro in the isolated middle cerebral artery (MCA) by application of combined extraluminal ion changes according to changes of the extracellular milieu during CSI ² . The aim of this study was to further evaluate the role of endothelial and neuronal NOS and the vascular endothelium for vasodilation and vasoconstriction during CSD/CSI induced ion changes, respectively.

Methods

Rats were anesthetized, decapitated, and the brain was rapidly removed. The MCA was carefully dissected from the brain, cannulated and pressurized. Vascular reactivity to CSD/CSI induced ion changes (mmol/l: Na⁺ 60; K⁺ 50; Ca²⁺ 0.1; Mg²⁺ 0.7; Cl- 110; glucose 5.0; pH 6.90; osmolarity ∼224) was investigated under baseline conditions and following NOS inhibition, cytochrome P450-epoxygenase inhibition or endothelium removal, respectively.

Results

In response to CSD/CSI ion cocktail the arteries significantly dilated from 107 ± 8 μm to 146 ± 15 μm. In presence of the unspecific NOS inhibitor L-NNA we found vasoconstriction from 85 ± 5 μm to 73 ± 10 μm. In contrast, in presence of the n-NOS specific inhibitor 7-NI vasodilation was only prevented. During unspecific as well as n-NOS specific NOS inhibition, application of the L-type calcium channel antagonist nimodipine not only reestablished but significantly augmented vasodilation to CSD/CSI ion-cocktail. CYP450-epoxygenase inhibition with miconazole did not change vasodilation to CSD/CSI induced ion changes. Following removal of the endothelium by transient air application a slight vasodilation was observed at baseline conditions (diameter before and after air application: 116 ± 10 μm and 136 ± 7 μm, respectively). Starting from this new baseline, CSD/CSI ion-cocktail induced a significant vasoconstriction to 98 ± 8 μm.

Conclusion

Application of CSD/CSI ion-cocktail caused vasodilation of the MCA in absence and vasoconstriction in presence of unspecific NOS inhibition or following removal of the endothelium. Specific n-NOS inhibition only prevented vasodilation. These results indicate that mediators released from the endothelium play an important role for vasodilation during CSD/CSI induced ion changes, preventing vasoconstriction during enhanced potassium concentrations. The exact nature of these mediators have to be revealed in further experiments, but it seems likely that endothelial NO plays a predominant role.