A novel atherothrombotic model of ischemic stroke induced by injection of collagen into the cerebral vasculature

Abstract

Background: Most ischemic strokes in humans are caused by ruptured arterial atheroma, which activate platelets and produce thrombi that occlude cerebral vessels.

Methods: To simulate these events, we threaded a catheter through the internal carotid artery toward the middle cerebral artery (MCA) orifice and injected collagen directly into the cerebral circulation of male C57Bl/6 mice and Wistar rats.

Results: Laser-Doppler flowmetry demonstrated reductions in cerebral blood flow (CBF) of ∼80% in mice and ∼60% in rats. CBF spontaneously increased but remained depressed after catheter withdrawal. Magnetic resonance imaging showed that ipsilateral CBF was reduced at 3h after collagen injection and markedly improved at 48 h. Micro-computed tomography revealed reduced blood vessel density in the ipsilateral MCA territory at 3 h. Gross examination of excised brains revealed thrombi within ipsilateral cerebral arteries at 3 h, but not 24 h, after collagen injection. Immunofluorescence microscopy confirmed that platelets and fibrinogen/fibrin were major components of these thrombi at both macrovascular and microvascular levels. Cerebral infarcts comprising ∼30% of hemispheric volume and neurobehavioral deficits were observed 48 h after ischemic injury in both mice and rats.

Comparison with existing methods: Collagen injection caused brain injury that was similar in magnitude and variability to mechanical MCA occlusion or injection of a pre-formed clot; however, alterations in CBF and the mechanism of vascular occlusion were more consistent with clinical ischemic stroke.

Conclusion: This novel rodent model of ischemic stroke has pathophysiologic characteristics consistent with clinical atherothrombotic stroke, is technically feasible, and creates reproducible brain injury.

Keywords: Atherothrombosis; Collagen; Ischemic stroke; Mouse; Platelet; Rat.

Fig. 1

Injection of collagen into the cerebral vasculature reduces cerebral blood flow (CBF) in the mouse. Laser-Doppler flowmetry was used to monitor CBF in the territory of the middle cerebral artery before, during, and after transient occlusion of the MCA with a nylon filament or injection of collagen through a catheter threaded into the distal internal carotid artery. “Base” represents stabilized CBF at baseline expressed as percent; “start” represents insertion of the filament or first collagen injection (total of 6 injections every 5 min); “reperfusion” represents removal of the filament or catheter and release of the ipsilateral common carotid ligature. N = 7 filament occlusion; N = 10 collagen injection.

Fig. 2

Cerebral perfusion and architecture are disrupted after injection of collagen in the mouse. Focal cerebral ischemia was induced by injecting six 1-μg boluses of collagen into the distal internal carotid artery of the mouse. (A) An 11.7 T magnetic resonance scanner was used to acquire T2-weighted (T2), apparent diffusion coefficient (ADC), and perfusion images on live mice 3 and 48 h after collagen injection. Regions with hyperintense T2 signals and hypointense ADC values were detected in the ipsilateral cortex (red arrow), hippocampus (yellow arrow), and thalamus (white arrow) at 3 h, accompanied by marked reductions in regional cerebral blood flow (CBF). At 48 h, T2 signals, ADC hypointensities, and CBF were improved, but abnormalities remained. (B) Micro-CT was performed on brains excised from mice 3 h after collagen injection. A coronal brain section shows disruption of vascular network in the middle cerebral artery territory and reduced vascular density ipsilateral to collagen injection (red arrow) compared to that in the contralateral hemisphere (blue arrow). (C) Micro-CT was used to construct cerebral heat maps that display average vessel density in the ischemic (red square) and contralateral brain (yellow square). Lighter colors represent higher vessel densities. (D) Right sagittal view from a micro-CT scan of the neurovasculature shows a poorly defined vascular network ipsilateral to collagen injection. (E) Left sagittal view of the neurovasculature contralateral to collagen injection. Images are representative of three separate experiments. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 3

Collagen injection causes formation of platelet- and fibrin-rich thrombi in cerebral macro- and microvessels. Focal cerebral ischemia was induced by injecting six 1-μg boluses of collagen into the distal right internal carotid artery (ICA) of the mouse. Three hours after collagen injection, brains were excised for histologic and immunofluorescent examination of cerebral vasculature. (A) Ventral (top panel) and dorsal (bottom panel) gross images show thrombus in internal carotid, middle cerebral, and anterior cerebral arteries and their branches. Black box in ventral view is magnified in (B) top panel. (B) Middle and lower panels show platelet- and fibrin-rich thrombi in cross sections of the ICA. Middle panel is labeled with anti-laminin antibody (red), anti-CD41 platelet antibody (green), and anti-nuclear stain (blue). Bottom panel is labeled with anti-laminin antibody (red), anti-fibrinogen/fibrin antibody (Fbn; green), and anti-nuclear stain (blue). (C) Top panel: triphenyltetrazolium chloride-stained coronal brain section of brain excised 3 h after collagen injection. Black box denotes area from which the images in the two lower panels were taken. Middle panel labeled with anti-laminin antibody (red), anti-CD41 platelet antibody (green), and anti-nuclear stain (blue). Bottom panel labeled with anti-laminin antibody (red), anti-fibrinogen antibody (green), and anti-nuclear stain (blue). Images are representative of six separate experiments. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 4

Injection of collagen into the cerebral vasculature causes anatomic brain injury and behavioral deficits in the mouse. Focal cerebral ischemia was induced by transient filament-occlusion of the middle cerebral artery or by injecting six 1-μg boluses of collagen into the distal right internal carotid artery of the mouse. After 48 h, cerebral infarct volume was quantified by staining brain slices with triphenyltetrazolium chloride (TTC) and neurobehavioral deficit score was assessed. (A) Representative TTC-stained coronal brain sections from the mouse filament-occlusion model (top) and collagen injection model (bottom). (B) Cerebral infarct volumes for the filament-occlusion (solid bar) and collagen injection (open bar) models. (C) Neurobehavioral deficit scores for the filament-occlusion (solid bar) and collagen injection (open bar) models. N = 5 filament occlusion; N = 8 collagen injection.

Fig. 5

Effect of serial vs. single injection of collagen on cerebral blood flow (CBF), anatomic brain injury and behavioral deficit in the mouse. (A) Laser-Doppler flowmetry was used to monitor CBF in the territory of the MCA before, during, and after serial injections of collagen (6× 1 μg collagen) or a single injection of collagen (5 μg collagen) through a catheter threaded into the distal internal carotid artery. “Base” represents stabilized CBF at baseline expressed as percent; “start” represents first collagen injection; “reperfusion” represents removal of the catheter and release of the the ipsilateral common carotid ligature. After 48 h, cerebral infarct volume was quantified by staining brain slices with triphenyltetrazolium chloride (TTC) and neurobehavioral deficit score was assessed. (B) Cerebral infarct volumes after serial injections of collagen (6× 1 μg; solid bar) or single injection of collagen (1× 5 μg; open bar). (C) Neurobehavioral deficit scores after serial collagen injection (6× 1 μg; solid bar) or single collagen injection (1 × 5 μg; open bar). N = 5 for 5 μg group; N = 10 for 6× 1 μg group.

Fig. 6

Injection of collagen into the middle cerebral artery (MCA) reduces cerebral blood flow (CBF) in the rat. Laser-Doppler flowmetry was used to monitor CBF in the territory of the MCA before, during, and after occlusion of the MCA by injection of collagen or a preformed clot through a catheter threaded to the MCA orifice. “Base” represents stabilized CBF at baseline expressed as percent; “start” represents insertion of the clot or first collagen injection (total of 6 injections every 5 min); “reperfusion” represents removal of the catheter and release of the bilateral common carotid ligatures. N = 12 each group.

Fig. 7

Injection of collagen into the middle cerebral artery (MCA) causes formation of platelet- and fibrin-rich thrombi in cerebral macro- and microvessels. Focal cerebral ischemia was induced by administering six 10-μl boluses of collagen near the right MCA orifice of the rat. At 3 and 24 h after collagen injection, brains were excised for histologic and immunofluorescent examinations of cerebral vasculature. (A) Ventral (upper panel) and dorsal (lower panel) gross images show thrombus in the MCA and its branches at 3 h (left panel) but not at 24 h (right panel) after collagen injection. (B) Immunofluorescent images show platelet- and fibrin-rich thrombi in cross sections of the MCA at 3 h (left panel) but not 24 h (right panel) after collagen injection. Upper panels are labeled with anti-laminin antibody (red), anti-CD41 platelet antibody (green), and anti-nuclear stain (blue). Lower panels are labeled with anti-fibrinogen/fibrin antibody (Fbn; red), anti-CD41 platelet antibody (green), and anti-nuclear stain (blue). (C) Coronal brain slices display platelet and fibrin deposition in cortical cerebral microvessels at 3 h (left panel) and 24 h (right panel) after collagen injection. Upper panels are labeled with anti-laminin antibody (red), anti-CD41 platelet antibody (green), and anti-nuclear stain (blue). Lower panels are labeled with anti-fibrinogen/fibrin antibody (red), anti-CD41 platelet antibody (green), and anti-nuclear stain (blue). Images are representative of four separate experiments. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 8

Injection of collagen into the middle cerebral artery (MCA) causes anatomic brain injury and behavioral deficits in the rat. Focal cerebral ischemia was induced by injection of a preformed clot into the MCA or by injection of six 10-μl boluses of collagen near the MCA orifice of the rat. After 48 h, cerebral infarct volume was quantified by staining brain slices with triphenyltetrazolium chloride and neurobehavioral deficit score was assessed. (A) Cerebral infarct volumes for the embolic clot (solid bar) and collagen injection (open bar) models. (B) Neurobehavioral deficit scores for the embolic clot (solid bar) and collagen injection (open bar) models. N = 12 each group.