Summary: Using a closed cranial window system and in travital microscopy/videometry, we studied the rat pial arteriolar (30-60 !Lm) responses to CO2 before and fol lowing a light/dye (LID) endothelial injury or topical ap plication of the nitric oxide synthase (NOS) inhibitor, ni tro-L-arginine (L-NA) or its inactive form, D-NA. LID treatment consisted of intravenous injection of sodium fluorescein and the illumination (for 90 s) of arteriolar discrete segments on the cortical surface with light from a mercury lamp. Functional changes in pial arteriolar en dothelium were characterized by evaluating responses to topical application of acetylcholine (Ach, 5 x 10-4 M) and to intravenous (i.v.) oxotremorine (OXO, a stable blood-brain barrier permeant muscarinic agonist, 1 !Lg kg -1 min -1 ) . After the LID injury, dilation to Ach was absent whereas dilations to the NO donor, S-nitroso acetyl-penicillamine (SNAP, 10-5 M) and to CO2 (5%) were unchanged (Paco2 = 70 mm Hg). Loss of Ach re sponse but intact SNAP response confirmed functional endothelial injury and intact smooth-muscle function. The global endothelium-dependent vasodilation induced by i. v. OXO was markedly attenuated when expanding the LID injury field from 300 !Lm to 6 mm in diameter.
However, the global vasodilation induced by inhalation ofNitric oxide (NO) is a rapidly diffusible and po tent dilator of vascular smooth muscle (Moncada et ai., 1991). It is synthesized from L-arginine, in com bination with molecular oxygen, by nitric oxide synthase (NOS) (e.g., Luscher and Vanhoutte, 1990), a calcium/calmodulin and NADPH-depen- Abbreviations used: Ach, acetylcholine; aCSF, artificial CSF; ICP, intracranial pressure; eNOS, endothelial nitric oxide syn thase; iNOS, inducible nitric oxide synthase; LID, light/dye; L-NA, nitro-L-arginine; NOS, nitric oxide synthase; OXO, oxo tremorine; SNAP, S-nitr.oso-acetyl-penicillamine; TTX, tetrodo toxin.
944CO2 was still unaffected by this increase in the area of light exposure. This provides evidence that the expanded exposure was capable of impairing global vasodilation re SUlting from endothelium-dependent stimuli but not from inhalation of CO2, The intact CO2 response despite an endothelial dysfunction suggests that the reported NO de pendence of hypercapnia-induced cerebral hyperemia in rats cannot be attributed to an endothelial NO source.Topical suffusion of L-NA (1 mM) for 45-60 min in our preparation blocked the pial arteriolar response to Ach, whereas CO and SNAP responses were unaffected. An attenuation (by 50%) of the response to CO2 was achieved if suffusion of L-NA was given for ?2 h. Suffusion of D-NA, applied in the same manner, did not influence re sponses to any of the above applications. This demon strates that there is a NO-dependent component for hy percapnic cerebral vasodilation even at the pial arteriolar level. The strikingly different time-related effect of topi cal L-NA on the Ach and CO2 responses, together with the lack of effect of endothelial injury on CO2-induced dilation, strongly sugg...