Sustained endothelial dependent dilation in pial arterioles after crosslinked hemoglobin transfusion

Hemoglobin is known to bind nitric oxide (NO) with high affinity. Plasma-based hemoglobin may provide a more effective sink for NO than red cell-based hemoglobin because of a closer and consistent proximity to the endothelium. Despite the known endothelial tight junctions that exist in cerebral vessels, plasma-based hemoglobin may inhibit NO-derived vasoreactive mechanisms in brain. If so, the response to endothelial and non-endothelial dependent vasodilator substances should be affected. In pentobarbital anesthetized cats, we tested this hypothesis by measuring the pial arteriole blood vessel diameter using a cranial window before and after systemic transfusion of a human crosslinked hemoglobin compound. We then topically applied solutions of endothelial dependent or endothelial independent vasodilators and an NO synthase inhibitor to the surface of the brain within the window and remeasured the arteriole size. Topical acetylcholine (Ach) increased diameter in all arteriole sizes. The corresponding increases in diameter to Ach in time control cats (32% hematocrit) and in albumin transfused cats (18% hematocrit) were similar to those in hemoglobin transfused cats. Likewise, size-dependent dilation to SIN-1 in the hemoglobin group was similar to that in the control and albumin groups. The pial arteriole response to adenosine diphosphate (ADP) and sodium nitroprusside (SNP) also increased arteriole diameter in small, medium and large arterioles. Superfusion with L-nitroarginine to inhibit NO synthase markedly reduced the dilator response to Ach and ADP but not to SIN-1 or SNP. Thus, prior crosslinked hemoglobin transfusion does not interfere with vasodilator responses to either Ach, ADP, SIN-1 or SNP. When dilute solutions of crosslinked hemoglobin were superfused abluminally in the cranial window in anesthetized but non-transfused animals, the dilator response to Ach was unchanged at 10^-8 M hemoglobin, was attenuated at 10^-7 and 10^-6 M hemoglobin, and was completely blocked at 10^-5 M hemoglobin. This finding implies negligible permeation of the blood-brain barrier by crosslinked hemoglobin acutely after the transfusion.