Male Sprague-Dawley rats were housed under diurnal lighting conditions and allowed food and tap water ad libitum. All animal studies were carried out in strict accordance with the Guide for the Care and Use of Laboratory Animals published by the National Institute of Health and complied with ARRIVE guidelines (http://www.nc3rs.org/ARRIVE). The animal protocol used in this study was reviewed and approved by the INHA University-Institutional Animal Care and Use Committee (INHA-IACUC) with respect to ethicality (Approval Number INHA-120410-137). MCAO conducted as previously described [11]. In brief, male Sprague-Dawley rats (250~300 g) were anesthetized with 5% isoflurane in a 30% oxygen/70% nitrous oxide mixture and anesthesia was maintained during procedures using 0.5% isoflurane in the same gas mixture. Occlusion of the right middle carotid artery was induced for 1 h by advancing a nylon suture (4-0; AILEE, Busan, South Korea) with a heat-induced bulb at its tip (~0.3 mm in diameter) along the internal carotid artery for 20~22 mm from the bifurcation of the external carotid artery and this was followed by reperfusion for up to 14 days. The left femoral artery was cannulated for blood sampling to determine pH, PaO₂, PaCO₂, and blood glucose concentrations (I-STAT; Sensor Devices, Waukesha, WI). Regional cerebral blood flow (rCBF) was monitored using a laser Doppler flowmeter (Periflux System 5000; Perimed, Jarfalla, Sweden). A thermoregulated heating pad and a heating lamp were used to maintain a rectal temperature of 37.0±0.5℃ during MCAO. Animals were randomly allocated to two groups, that is MCAO group (n=27) or a sham control group, animals in sham group were operated upon in an identical manner but the right MCA was not occluded (n=3). In general, mortality was not observed during surgery, but mortality after surgery was 11.1% (3/27).

Rats were decapitated at 2 days post-MCAO, and whole brains were dissected coronally into 2-mm slices using a metallic brain matrix (RBM-40000, ASI, Springville, UT). Slices were immediately stained by immersing them in 2% 2,3,5-triphenyl tetrazolium chloride (TTC) at 37℃ for 15 min and then fixed in 4% paraformaldehyde. The TTC-stained slices were visualized using a Nikon D40 DSLR (Nikon, Tokyo, Japan).

Animals were sacrificed at prescribed times after MCAO and brains were fixed in 4% paraformaldehyde by transcardiac perfusion and then post-fixed in the same solution overnight at 4℃. Brain sections of 1 and 4 days post-MCAO (20 µm) were obtained using a vibratome. Brain of 12 days post-MCAO were embedded in paraffine, and 10 µm-thick sections were prepared. The sections were deparaffinized in zylene and rehydrated through graded series of ethanol. Immunological staining was performed as previously described [12]. Primary antibodies were diluted 1:300 for anti-ionized calcium binding adaptor molecule-1 (Iba-1) (Wako Pure Chemicals, Osaka, Japan), 1:200 for anti-myeloperoxidase-1 (MPO-1) (Abcam, Cambridge, UK), 1:200 for anti-neuronal nuclei (NeuN) (BD Transduction Laboratories, San Jose, CA), 1:200 for anti-glial fibrillary acidic protein (GFAP) (BD Transduction Laboratories, San Jose, CA), 1:200 for anti-endothelial Cell (RECA) (AbDSerotec, Oxford, UK), 1:200 for anti-CD45 (BD Transduction Laboratories, San Jose, CA), 1:200 for mouse monoclonal anti-nerve injury-induced protein 1 (Ninjurin-1) (Merck Millipore Corporation, Darmstadt, Germany), and 1:200 for rabbit polyclonal anti-Ninjurin1 (Abcam, Cambridge, UK) antibodies. After washing with PBS three times, sections were incubated with 1:200 of FITC-conjugated anti-mouse or -rabbit IgG (Jackson ImmunoRes Lab, West Grove, PA) for anti–ninjurin-1, 1:200 of rhodamine-conjugated anti-rabbit IgG (Jackson ImmunoRes Lab, West Grove, PA) for anti-Iba-1, MPO-1, NeuN, and 1:200 of rhodamine-conjugated anti-mouse IgG (Jackson ImmunoRes Lab, West Grove, PA) for anti-GFAP, RECA, and CD45 in PBS for 30 min at room temperature and examined under Zeiss LSM 510 META microscope (Carl-Zeiss-Strasse, Oberkochen. Germany). Experiments were repeated at least three times and representative images are presented.

Brain sections were prepared as described above for immunohistochemistry. Following incubation with 0.2% Triton X-100 in PBS for 5 min, sections were incubated with DAPI (1 ug/mL; 4',6-diamidino-2-phenylindole) for 15 min at room temperature, and viewed under the microscope mentioned above.

Brain samples were prepared from cortical core, cortical penumbra, striatal core, striatal penumbra as indicated regions in Fig. 1A from MCAO and sham groups, and proteins (10 µg) were purified, separated in 10% sodium dodecyl sulfate-polyacrylamide gels, and transfer to membranes. After blocking membranes with 5% non-fat milk for 1 hr, they were incubated with primary antibodies for anti-Ninjurin-1 (1:2000, BD Transduction Laboratories, San Jose, CA) and anti-α-tubulin (1:5000, Merck Millipore Corporation, Darmstadt, Germany) overnight at 4℃. The next day, blots were detected using anti-mouse HRP-conjugated secondary antibody (1:2000; Santa Cruz Biotechnology) and a chemiluminescence kit (Roche, Basel, Switzerland).

To examine the expression pattern of Ninj1 in the postischemic brain, immunoblotting and triple fluorescent immunohistochemistry were conducted. For immunoblot analysis, protein samples were prepared from cortical penumbras, cortical cores, striatal penumbras, and striatal cores (Fig. 1A) at 1, 4, 6, 8, 10, and at 14 days after 60 min of MCAO. In the cortical penumbras, the induction of Ninj1 was detected at 1 day post-MCAO and its level gradually increased until day 8 and then decreased thereafter (Fig. 1B). In cortical cores, Ninj1 protein was induced at 1 day post-MCAO, and this level was maintained until day 10 (Fig. 1C). In striatal cores and penumbras, Ninj1 was inducted in a manner similar to that observed in cortex, although levels of induction were slightly higher (Fig. 1D and E). These results describe the dynamic nature of Ninj1 expression in the postischemic brain, particularly, in the penumbra regions of ipsilateral hemispheres.

The observed dynamic expression of Ninj1 protein, in cortices and striata prompted us to identify the cell types expressing Ninj1 in different regions at different times. Since cortex and striatum revealed similar temporal expression profiles, an immunohistochemical study was conducted with focus on cortices (Fig. 1A). To identify the cell types expressing Ninj1 at 1 day post-MCAO, triple immunofluorescence staining was performed using anti-Ninj1 antibody and antibodies specific for neurons (NeuN), astrocytes (GFAP), microglia (Iba-1), neutrophils (MPO-1), and endothelial cells (RECA-1), and using DAPI. In sham controls, Ninj1⁺ cells were rarely detected in any brain region (data not shown). However, at 1 day post-MCAO, Ninj1 immunoreactivity was markedly increased in cortical penumbras (Fig. 2A). Most of Ninj1⁺ cells were merged with MPO-1⁺ (Fig. 2B) or RECA-1⁺ cells (Fig. 2C), indicating that Ninj1 protein was induced in neutrophils and endothelial cells in cortical penumbras. The majority of Ninj1⁺/MPO-1⁺ cells were observed in brain parenchyma (Fig. 2B), and some were detected within blood vessels attached to vessel walls (data not shown). Ninj1 induction was also detected in Iba-1⁺ cells, but Ninj-1 levels were low and few merged cells were observed (Fig. 2D). However, they were rarely merged with NeuN⁺ or GFAP⁺ cells (Fig. 2E and F). Similar observations were made in cortical cores, but numbers of Ninj1⁺/Iba1⁺ were higher (Fig. 2J) and those of Ninj1⁺/MPO-1⁺ (Fig. 2H) and Ninj1⁺/RECA⁺ (Fig. 2I) cells were fewer than in cortical penumbras (Fig. 2B~D). These results indicate that at 1 day post-MCAO, Ninj1 was induced mainly in neutrophils and endothelial cells in penumbras and in neutrophils and microglia in cores of ischemic hemispheres.

At 4 days post-MCAO, Ninj1 immunoreactivity was markedly increased in infarct penumbras (Fig. 3A) and cores (Fig. 3G). At this time, the numbers of both CD45⁺ (pan-leukocyte marker) and Iba-1⁺ (a marker of cells of myeloid origin) cells were remarkably increased and these cells were accumulated in infarction cores and penumbras. Iba-1⁺ cells were larger than those observed at 1 day post-MCAO (Fig. 2D and J) and were amoeboid in shape, indicating an activated state (Fig. 3C and I). Ninj1 immunoreactivity was found to co-localize with these Iba-1⁺ and CD45⁺ cells in both cortical penumbras (Fig. 3B and C) and cores (Fig. 3H and I), indicating that Ninj1 was mainly induced in activated microglia/macrophages at 4 days post-MCAO. The Ninj1⁺ cells were also merged with RECA-1⁺ cells (Fig. 3D) but Ninj1⁺/MPO-1⁺ cells were rarely detected (data not shown). Similarly, Ninj1 immunoreactivity was hardly detected in NeuN⁺ and in GFAP⁺ cells (Fig. 3E and F). Similar results were obtained at 8 days post-MCAO (data not shown).

Ninj1 levels were gradually decreased in cortical penumbras from 8 days to 14 days post-MCAO, but the levels were maintained in cortical cores at 10 days post-MCAO and decreased thereafter (Fig. 1B and C). To determine whether there are changes in cellular distribution of Ninj1 in cortex at 12 days post-MCAO, we conveyed immunohistochemical studies. In cortical penumbras, Ninj1 immunoreactivity was mainly detected in a few CD45⁺ cells but hardly detected in Iba-1⁺ cells at 12 days post-MCAO (Fig. 4B and C). Since, it has been known that in contrast to macrophages, microglia expresses low or intermediate levels of the leukocytes marker CD45, Ninj1 was expressed in reactive macrophages but not in resident microglia in cortical penumbras (Fig. 4B and C). On the contrary, in cortical cores, Ninj1 immunoreactivity was detected both in CD45⁺ and Iba-1⁺ cells, which might be reactive macrophages, and the numbers of merged cells were significantly higher than in cortical penumbra (Fig. 4H and I). Importantly, Ninj1^-/Iba-1⁺ cells displayed ramified morphology (Fig. 4C), indicating that they were in resting state. Ninj1 expression was detected in endothelial cells, but hardly detected in neurons and astrocytes (Fig. 4D-F).