Female C57BL/6 mice were originally purchased from Charles River Laboratories (Montreal) and bred in house. Studies were carried out on mice aged six weeks, weighing 17–22 g independently of the estrous cycle. Cyclic changes in estrogen levels may explain the relatively large variability in the number of apoptotic neurons detected in the brain following MCAo. Mice were kept in group cages and allowed free access to food and water throughout the duration of the study. All procedures were in accordance with the Canadian Council for Animal Care and the University Council on Animal Care guidelines for research, and all protocols involving animals, behavioural testing, surgeries and animal maintenance were approved in accordance with the above guidelines.

The tip of the 5-0 monofilament suture to be used in the MCAo was first blunted by heating over a flame and then the distal 11 mm coated with poly-L-lysine (Sigma) (0.1% in deionized water) and dried in a 60°C oven for 1 hour. A point 11 mm from the blunt end was marked with silver impregnation marker to render it visible through the arterial wall.

Mice were placed in an induction chamber supplied with oxygen at a rate of 1 L/min and isofluorane at 4%. Injectable chemical anaesthesia was avoided in order to accelerate post-surgical awakening for immediate behavioural observation. All mice were maintained on 1 L/min oxygen supplemented with 1.5% isofluorane throughout the duration of surgery using a small mask. Using an operating microscope, the left common carotid artery (CCA) was exposed through a midline neck incision and dissected free from surrounding tissue from 1 cm proximal to its bifurcation to the base of the skull. The external carotid artery (ECA) was dissected and coagulated near the origin of the lingual and maxillary artery branches. The internal carotid artery (ICA) was isolated and separated from the adjacent vagus nerve. A 5-0 vicryl suture was tied around the CCA 1 cm proximal to the bifurcation. The 2 cm poly-L-lysine-coated suture was then inserted via a small incision in the CCA distal to where it was tied, but proximal to the bifurcation, and advanced 11 mm from the CCA bifurcation to occlude the middle cerebral artery (MCA). A 5-0 vicryl suture was placed around the ICA and intraluminal nylon suture to prevent bleeding and movement of the nylon suture. The neck skin was closed with 5-0 vicryl sutures.

The mice were placed in separate cages and allowed to awaken briefly from anaesthesia during the 30 minutes of occlusion to permit neurobehavioural assessment. Following assessment, the mice were re-anaesthetized and after 30 minutes of occlusion, the neck skin was re-opened, the nylon occlusion suture carefully removed, and the neck skin closed with 5-0 vicryl sutures.

In this method, and in contrast to the procedure reported by Belayev and colleagues 17, the occipital artery branches of the external carotid artery and the pterygopalentine artery were not coagulated or ligated. Kitagawa and colleagues 15 showed that differences in cranial vasculature, specifically the posterior communicating artery, could affect the degree of ischemia and should be assessed during occlusion by measuring cortical microperfusion. Because we correlated neurobehavioural testing during the occlusion with the ischemic lesion volume, this more technically challenging evaluation could be avoided.

The mice were allowed to recover in separate cages and given antibiotics (0.01 ml ampicillin, 50 mg/kg) and analgesics (0.04 – 0.05 ml buprenorphine, 0.09 mg/kg) post-surgery. Animals were also given up to 4 mL daily subcutaneous fluids throughout the recovery period and provided with a high peptide liquid nutritional supplement (Peptamen) and rice pabulum on the cage floor for the duration of post-operative study. The criterion for early euthanasia was adhered to in the event of excessive weight loss.

Sham-operated mice were prepared for surgery as above. The left CCA was exposed through a midline neck incision, dissected free from surrounding tissue, and ligated with a 5-0 vicryl suture 1 cm proximal to the bifurcation. The neck skin was closed with 5-0 vicryl sutures.

Neurobehavioural evaluations consisted of two testing batteries and were performed during occlusion (20 minutes after suture insertion), following recovery from reperfusion, 24 hours, and 72 hours post-reperfusion. The first test battery consisted of 2 test items each scored on a scale from 0–12 (0 = normal, 12 = severely impaired) 17: the postural reflex test to observe the animal's upper body posture while being suspended by the tail 24; and the forelimb placing test to examine psychomotor responses to tactile stimuli 25. For the second battery, we attempted to assess the motor function of each mouse without handling. Each mouse was assigned a score of 0 – 4: 0 = no observable neurological deficit; 1 = failure to extend the right forepaw; 2 = circling to the right; 3 = falling to the right; and 4 = cannot walk spontaneously 26.

Based on the rat MCAo studies 21, we examined the spinal cord 24 hours after reperfusion for the extent microglial proliferation or activation and effects on lower motor neurons. Mice were anaesthetized 24 hours (n = 8) or 72 hours (n = 8) after reperfusion of the MCA or 24 hours following sham operation (n = 5) via intraperitoneal injection of ketamine-xylazine mixture (0.03 ml/10 g) and perfused transcardially with heparinized saline until the outflow ran clear. Fixation was achieved with 25 ml/min intracardial injection of 50 ml/100 g body weight of fixative (4% paraformaldehyde in PBS). Brains and spinal cords were dissected out and placed in the same fixative overnight at 4°C. The frontal brain and lumbar spinal cord from L3-S1 was processed for paraffin-sectioning. Five μm thick serial sections were cut in the coronal plane and mounted on glass slides. Every 5^th section was stained with haematoxylin and eosin (H&E), or immunoreacted using antibodies to detect macrophages (ionized calcium-binding adaptor molecule-1 (IBA-1), Wako), astrocytes (glial fibrillary acid protein (GFAP), BD Pharmingen) or apoptosis (active caspase 3, BD Pharmingen).

Sections were dried overnight, deparaffinized, and rehydrated. Antigen retrieval was performed using boiling high-pH TRIS-EDTA buffer following which the sections were blocked for endogenous hydrogen peroxide (3% hydrogen peroxide, 5 min.). Non-specific antibody binding was blocked (10% normal goat serum, 5% bovine serum albumin (BSA) in PBS containing 0.3% Triton X-100; 30 min.) and the sections then incubated with the primary antibody (1 hr, room temperature). Primary antibodies consisted of either polyclonal rabbit-anti-IBA-1 (1:1000), monoclonal mouse-anti-GFAP (1:75) or polyclonal rabbit-anti-active caspase 3 (1:500). All primary antibodies were prepared in 5% BSA, 5% normal goat serum, and 0.3% Triton X-100. Following this, sections were incubated in secondary antibody (1:200 goat-anti-rabbit or horse-anti-mouse in 5% normal goat serum, 5% BSA, 0.3% Triton X-100 in PBS) for 30 minutes at room temperature. The antigen:antibody complex was localized using the avidin-biotin complex (ABC staining kit, Vector Laboratories) as per the manufacturer's instructions. Sections were washed and developed for visualization in 0.3 mg/ml 3', 3-diaminobenzidine tetrahydrochloride (DAB, Sigma) containing 0.01% hydrogen peroxide for 10 minutes and then washed thoroughly. Sections were counterstained with Harris's Haematoxylin and then mounted in Cytoseal 60 (VWR). Negative control slides consisted of the omission of the primary antibody.

All cell counts were carried out in defined areas maintained across all sections. For the quantification of the percent IBA-1-positive area (to support the observation of an increase in the number of primed microglia), images of IBA-1-immunolabelled sections were captured using Northern Eclipse imaging software. A threshold was applied to the image which allowed for calculation of the total area of immunoreaction and the percent immunoreaction in a defined area which was consistent across all sections.

One-way ANOVAs were performed with Tukey post-hoc tests for neuronal density, neuronal caspase 3 activation, neurobehavioural assessment analysis, IBA-1 immunoreactivity assessment, and primed microglial cell counts,.

Thirty minutes of MCAo resulted in diffuse penumbral damage spanning both the subcortex and the cortex in 5 of 8 mice at the 24-hour time point and 6 of 8 mice at the 72 hour time point. In these mice a necrotic focal lesion was clearly visible (Table 1, Figure 1a). Two of 5 mice that did not exhibit cortical pathology did demonstrate behavioural deficits during occlusion meeting the inclusion criteria proposed by Belayev and colleagues 17. All 5 mice were excluded from further study.

In the remaining mice, all harbouring marked infarct regions, microglial activation was prominent, extending into the motor cortex of the lesioned hemisphere and not the intact hemisphere (Figures 1b and 1c). The number of large neurons in layer V of the primary motor cortex in these mice was decreased in the lesioned hemisphere compared to the intact hemisphere (Figures 1d–f, ANOVA p < 0.01). This was accompanied by an apparent increase in active caspase 3 immunoreactivity in the same region that did not reach significance due to variability (Figures 1g–i, ANOVA p = 0.1133). We observed microglial activation (Figure 1j) and caspase 3 activation (Figure 1k) in the corticospinal tract (CST) contralateral to MCAo, further suggesting the presence of an injury upstream to ventral horn motor neurons. Mice who did not display evidence of any focal infarct region did not have any pathology (glial or neuronal) in the motor cortex in either hemisphere (Table 1).

During occlusion, all mice exhibited head rotation, eyelid drooping and abnormal tongue holding consistent with the development of a cerebral infarct. Interestingly, mice were highly vocal throughout the course of the study. Animals were examined carefully for evidence of pain and were not in apparent distress. Analgesics were administered according to guidelines. Following reperfusion, behavioural deficits were sustained with 12 of 16 mice achieving scores above the inclusion cutoffs proposed by Belayev and colleagues 17 suggesting the presence of a lesion (Figure 2b). Eleven of these 12 mice had a marked region of infarct and cortical pathology. Twenty-four hours post-reperfusion most animals remained vocal with only a moderate head tilt and a mild postural reflex (Figure 2a–b). All animals survived 24 hours post-reperfusion and maintained good body weight. Eight of 9 animals survived 72 hours post reperfusion.

There were significantly more microglia with primed morphology in the ventral horn contralateral to MCAo than in the ventral horn ipsilateral to MCAo 24 hours and 72 hours post-reperfusion (Figure 3a–c, e–f, ANOVA p < 0.05). The total area of IBA-1 immunoreaction was also greater in the ventral horn contralateral to MCAo at both time points (Figure 3d, ANOVA p < 0.01). Cell bodies appeared thicker and processes were shorter (Figure 3c and 3f). Often primed and ramified microglial processes were enveloping large motor neurons (arrows in Figure 3f, higher magnification in 3g).

Despite the observed change in microglial activation state from ramified to primed, there was no difference observed in the morphology of GFAP-immunolabelled astrocytes in the ventral horns contralateral to MCAo compared with ventral horns ipsilateral to MCAo 24 hours and 72 hours post-reperfusion (Figure 3h–i). There was no significant difference observed in the number of apoptotic motor neurons in the ventral horns contralateral to MCAo compared with ventral horns ipsilateral to MCAo 24 hours and 72 hours post-reperfusion (Figure 3j–l).