Microthrombosis after experimental subarachnoid hemorrhage: time course and effect of red blood cell-bound thrombin-activated pro-urokinase and clazosentan

Abstract

Delayed cerebral ischemia (DCI) is a significant cause of morbidity and mortality for patients surviving the rupture of an intracranial aneurysm. Despite an association between vasospasm and DCI, thrombosis and thromboembolism may also contribute to DCI. In this study we investigate the time course of intravascular microclot formation after experimental subarachnoid hemorrhage (SAH) and assess the effects of the following two drugs on microclot burden: mutant thrombin-activated urokinase-type plasminogen activator (scFv/uPA-T), which is bound to red blood cells for use as a thromboprophylactic agent, and clazosentan, an endothelin antagonist. In the first study, adult male C57BL/6 mice were sacrificed at 24 (n=5), 48 (n=6), 72 (n=8), and 96 (n=3) hours after SAH induced by filament perforation of the anterior cerebral artery. Sham animals (n=5) underwent filament insertion without puncture. In the second study, animals received scFv/uPA-T (n=5) 3 hours after hemorrhage, clazosentan (n=5) by bolus and subcutaneous pump after SAH just prior to skin closure, or a combination of scFv/uPA-T and clazosentan (n=4). Control (n=6) and sham (n=5) animals received saline alone. All animals were sacrificed at 48 hours and underwent intra-cardiac perfusion with 4% paraformaldehyde. The brains were then extracted and sliced coronally on a cryostat and processed for immunohistochemistry. An antibody recognizing thrombin-anti-thrombin complexes was used to detect microclots on coronal slices. Microclot burden was calculated for each animal and compared among groups. Following SAH, positive anti-thrombin staining was detected bilaterally in the following brain regions, in order of decreasing frequency: cortex; hippocampus; hypothalamus; basal ganglia. Few microclots were found in the shams. Microclot burden peaked at 48 hours and then decreased gradually. Animals receiving scFv/uPA-T and scFv/uPA-T+clazosentan had a lower microclot burden than controls, whereas animals receiving clazosentan alone had a higher microclot burden (p<0.005). The overall mortality rate in the time course study was 40%; mortality was highest among control animals in the second study. Intravascular microclots form in a delayed fashion after experimental SAH. Microclots may be safely reduced using a novel form of thromboprophylaxis provided by RBC-targeted scFv/uPA-T and represent a potential target for therapeutic intervention in the treatment of DCI.

Fig. 1

Ventral surface of the brain of a sham animal (A) and a control animal sacrificed 48 hours after hemorrhage. The arrow shows residual blood. Bar in lower left corner=1 cm.

Fig. 2

H&E of coronal brain slice at 48 hours after SAH. Residual SAH is evident at the base (arrow) with extension into both Sylvian fissures. Original magnification 2×.

Fig. 3

Photomicrographs showing immunofluorescence staining of representative brain slices from time course study. Microclots of various sizes are shown in small cortical vessels (A) and larger penetrating arteries and veins (B). Original magnification 20×. Bar in lower right corner=50 microns.

Fig. 4

Microclot burden (#/mm³) for each group. Microclot burden reached a maximum of 27 microclots/mm³ at 48 hours. Mean values for 48-hour mice are significantly higher (p<0.001) than for all other groups. Vertical bars represent standard errors for each group.

Fig. 5

Microclot burden (#/mm³) for each experimental group. Compared to controls (23 microclots/mm³), microclot burden was lower in the scFv/uPA-T and scFv/uPA-T+clazosentan groups (each approximately 5 microclots/mm³ and higher in the clazosentan group (33 microclots/mm³). Significant differences were observed between all groups except scFv/uPA-T and scFv/uPA-T+clazosentan. On the x-axis, uPA refers to scFv/uPA-T.