Blood to Brain and Brain to Blood Passage of Native Horseradish Peroxidase, Wheat Germ Agglutinin, and Albumin: Pharmacokinetic and Morphological Assessments

Abstract: Native horseradish peroxidase (HRP) and the lectin wheat germ agglutinin (WGA) conjugated to HRP are protein probes represented in the blood‐brain barrier (BBB) literature for elucidating morphological routes of passage between blood and brain. We report the application of established pharmacokinetic methods, e.g., multiple‐time regression analysis and capillary depletion technique, to measure and compare bidirectional rates of passage between blood and brain for radioactive iodine‐labeled HRP (I‐HRP), WGA‐HRP (I‐WGA‐HRP), and the serum protein albumin (I‐ALB) following administration of the probes intravenously (i.v.) or by intracerebroventricular (i.c.v.) injection in mice. The pharmacokinetic data are supplemented with light and electron microscopic analyses of HRP and WGA‐HRP delivered i.v. or by i.c.v. injection. The rates of bidirectional movement between blood and brain are the same for coinjected I‐HRP and I‐ALB. Blood‐borne HRP, unlike WGA‐HRP, has unimpeded access to the CNS extracellularly through sites deficient in a BBB, such as the circumventricular organs and subarachnoid space/pial surface. Nevertheless, blood‐borne I‐WGA‐HRP enters the brain ˜10 times more rapidly than I‐HRP and I‐ALB. Separation of blood vessels from the neocortical parenchyma confirms the entry of blood‐borne I‐WGA‐HRP to the brain and sequestration of I‐WGA‐HRP by cerebral endothelial cells. Nearly half the I‐WGA‐HRP radioactivity associated with cortical vessels is judged to be subcellular. Light microscopic results suggest the extracellular pathways into the brain available to blood‐borne native HRP do not represent predominant routes of entry for blood‐borne WGA‐HRP. Ultrastructural analysis further suggests WGA‐HRP is likely to undergo adsorptive transcytosis through cerebral endothelia from blood to brain via specific subcellular compartments within the endothelium. Entry of blood‐borne I‐WGA‐HRP, but not of I‐ALB, is stimulated with coinjected unlabeled WGA‐HRP, suggesting the latter may enhance the adsorptive endocytosis of blood‐borne I‐WGA‐HRP. With i.c.v. coinjection of I‐WGA‐HRP and I‐ALB, I‐WGA‐HRP exits the brain more slowly than I‐ALB. The brain to blood passage of I‐WGA‐HRP is nil with inclusion of unlabeled WGA‐HRP, which does not alter the exit of I‐ALB. Adsorptive endocytosis of i.c.v. injected WGA‐HRP appears restricted largely to cells lining the ventricular cavities, e.g., ependymal and choroid plexus epithelia. In summary, the data suggest that the bidirectional rates of passage between brain and blood for native HRP are comparable to those for albumin. Blood‐borne WGA‐HRP is assessed to enter the brain more rapidly than native HRP and albumin, perhaps by the process of adsorptive transcytosis through BBB endothelia, but has difficulty leaving the CNS; the latter result may be due to avid binding and adsorptive endocytosis of WGA‐HRP by exposed CNS cells. Neither native HRP nor WGA‐HRP alters the integrity of the BBB to albumin. For this reason, both native HRP and WGA‐HRP are suitable probes for investigating the permeability of the BBB to macromolecules in vivo.