A simple screening procedure for evaluating central nervous system tissue sections showing structural and cytochemical alterations of the blood-brain barrier

Early blood-brain barrier changes in the rat following transient complete cerebral ischemia induced by cardiac arrest

This study examined regional patterns of increased vascular permeability following transient global cerebral ischemia. Rats underwent 3.5, 5 or 10 min of cardiac vessel bundle occlusion, i.e. cardiac arrest. The animals were killed at 2, 3, 5 and 15 min, or 1, 3, 6 and 24 h after global cerebral ischemia. Thirty minutes before the end of each blood recirculation period, the electron dense protein tracer--horseradish peroxidase (HRP) was intravenously injected and rats were perfusion-fixed for light and electron microscopic analysis. Control rats showed no HRP leakage. Post-ischemic rats demonstrated random blood-brain barrier (BBB) alterations. Permeability alterations were spotty and widespread in cortical, thalamic, basal ganglia, hippocampal, brain stem regions, cerebellum and white matter. Peroxidase extravasation frequently involved arterioles, veins and venules surrounded by perivascular spaces. Routes of increased HRP permeability included endothelial cell (EC) vesiculo-canalicular profiles and diffuse leakage through damaged ECs. Barrier damage determined by HRP permeability revealed a biphasic nature. The first stage appeared immediately after ischemia at the 2nd min and involved the 1st post-insult hour. There was no HRP leakage in rats sacrificed 3 h after insult. BBB opening appeared again 6 h after ischemia and remained open 24 h after cardiac arrest. The openings of BBB did not increase in frequency with longer periods of ischemia and recirculation. These results demonstrate that cardiac arrest produces a spotty BBB disturbances at vessel bifurcations and suggest that BBB changes associated with cardiac arrest may be multifactorial in time course and location.

Abnormalities of the blood-brain barrier in global cerebral ischemia in rats due to experimental cardiac arrest

The aim of the study was to characterize the impact of global cerebral ischemia resulting from cardiac arrest on the BBB permeability. Survival time of experimental animals after 3.5, 5 or 10 min ischemia range from 3.5-10 min to 24 h. Vascular permeability was evaluated with horseradish peroxidase (HRP). BBB disturbances were of biphasic nature. In the first phase, appearing immediately after ischemia and persisting till 1st postischemic hour, HRP extravasation involved mainly venous site of microcirculation and was limited to the cerebral and cerebellar cortex and the central periventricular structures. The second phase covering the period between 6 and 24 h after resuscitation was characterized by random HRP leakage in all the CNS structures. HRP penetrated through increased microvesicular and canalicular endothelial systems, through interendothelial junctions and via disintegrated endothelial cells. Distribution and perivenous localization of BBB changes in early phase suggests their connection with venostasis resulting from cardiac arrest. The second phase seems to be pathogenetically related with the consequences of the ischemic process.

Sites of egress of inflammatory cells and horseradish peroxidase transport across the blood-brain barrier in a murine model of chronic relapsing experimental allergic encephalomyelitis

Results are reported of experiments designed to focus at attachment sites of inflammatory cells (ICs) on the luminal surface of brain endothelial cells (ECs) and on the mechanisms of horseradish peroxidase (HRP) transport across the altered blood-brain barrier (BBB) in a murine model of chronic relapsing experimental allergic encephalomyelitis. Cationized ferritin (CF) served as a marker for evaluating the electrostatic nature of brain microblood vessels (MBVs) on the plasma membranes of ICs or normal mouse peripheral white blood cells and erythrocytes. SJL/J mice demonstrating clinical illness were given HRP or CF, in vivo or in situ, respectively. Light microscopy and conventional transmission electron microscopy of cerebellum or thoracic and lumbar spinal cord regions demonstrated HRP leakage most pronounced in MBVs with perivascular infiltrates. HRP traversed across the ECs via numerous vesicles and tubular profiles located mostly in the parajunctional regions, while EC junctions appeared closed. Scanning electron microscopy demonstrated that IC attachment was primarily at parajunctional sites on the EC surface. We also observed increased microvillar projections extending from the EC surface into the lumen. CF demonstrated a patchy decoration on both the luminal EC surface and IC membranes but did not label uncoated invaginating membrane pits or tubular structures. Our data indicate that the points of attachment of the ICs on the EC surface may reflect specific receptor sites where the ICs eventually gain entrance into CNS across the BBB during brain inflammation.

High voltage electron microscopic studies of endothelial cell tubular structures in the mouse blood-brain barrier following brain trauma

High-voltage electron microscopy was applied to the study of endothelial cell (EC) transport of macromolecules in a murine model of blood-brain barrier injury to study the role of the EC canalicular system following brain insult. Semithick sections from mouse brains subjected to acute (2-3 h) mechanical trauma demonstrated permeation of intravenously injected horseradish peroxidase via tubular structures either (a) in the absence of lysosome-associated structures in close proximity, or (b) in association with lysosomes, dense bodies or multivesicular bodies. Our data suggest a dual-purposed system of tubules, one portion that supplies the metabolic requirements of the cell and another portion, suggested to be more limited, that opens up as a result of brain injury.

Ultrastructural observations of spinal cord lesions and blood-brain barrier changes in scrapie-infected mice

Spinal cord samples from IM or VM mice injected intracerebrally with the 87V scrapie agent were examined ultrastructurally at the clinical stage of disease for changes in blood vessel permeability and for pathological alterations. In several animals, (3 of 16), massive changes were noted in the cervical spinal cords in the subependymal area of the cortical gray matter immediately surrounding the central canal including ependymal cell changes, the presence of amyloid plaque in close association with microglial cells, extensive neuropil vacuolation, the appearance of reactive astrocytes, degenerating neurites and vacuolated neurons. In those regions showing structural damage, localized increased permeability to horseradish peroxidase across the blood-brain barrier was noticed along with the appearance of numerous vesiculo-canalicular profiles in micro-blood vessel endothelial cells with extravasation of the tracer to the neuropil. Some damaged neurons appeared flooded with this tracer. These changes were not observed in either the thoracic or lumbar spinal cord regions. The occurrence of pathological changes in the spinal cords of a small percentage of intracerebrally injected mice was probably due to a high concentration of the scrapie agent which localized in the cervical spinal cord, presumably after entering the spinal fluid via the lateral ventricle at the time of injection.

Ultracytochemical studies of vesicular and canalicular transport structures in the injured mammalian blood-brain barrier

An ultracytochemical investigation was performed to study the origin of pinocytic vesicles and canalicular structures within endothelial cells (EC) of the injured mammalian blood-brain barrier (BBB). To accomplish this goal, two electron-dense tracers, native ferritin (NF) and horseradish peroxidase (HRP), were used in conjunction with the detection of alkaline phosphatase (AP) activity, a known marker of EC plasmalemma of brain micro-blood vessels. Brain ECs from (1) mice subjected to crude leptomeningeal damage for 1, 2, or 3 days and (2) cats subjected to cold lesion injury for 1, 4, or 24 h were evaluated for tracer transport and AP activity. Fine structural analysis of leaking segments of micro-blood vessels from damaged cerebral cortex or basal ganglia demonstrated pinocytic vesicles, deep invaginations of the luminal plasmalemma and elongated, tubular profiles, all containing tracer. Because we observed in ECs from both experimental models of brain injury a positive reaction for AP activity in the luminal plasmalemma, in its deep invaginations, in delimiting membranes of pinocytic vesicles, and in tubulo-canalicular structures, we conclude that all types of transport structures derive from the same 100A thick exoplasmic plasmalemmal membranes. Further, besides the pinocytic vesicular transport system (PTS), the canalicular transport system (CTS) appears to serve as an additional important mechanism for macromolecular transport across the damaged mammalian BBB.