Tat-HA-NR2B9c was designed and synthesized in our laboratory (11). It was dissolved in normal saline and intravenous (1.12 mg/kg/day, i.v.) injected. The injection volume was 3 ml/kg. Sham and vehicle groups were treated with corresponding volume of solvent.

The Institutional Animal Ethical Committee of Nanjing University (Approval No., GY20160108) authorized the experimental protocol, and all procedures involving animals were performed in accordance with National Institute of Health guidelines for the care and use of laboratory animals. In this study, adult male Sprague-Dawley rats (250–300 g; B&K Universal Group Limited, Shanghai) were used. An experimenter labeled all animals before allocation. Experiments were performed by investigators who were blinded to group allocation.

Transient global ischemia was carried out using 4-VO model (1). On the first day, rats were anesthetized by 10% chloral hydrate (300 mg/kg, i.p.) and then fixed by stereotaxic ear bars. The head was placed at about 30° to the horizontal. After disinfection using 75% alcoholic, both the vertebral arteries were occluded permanently by electrocauterization with a 0.5-mm diameter needle through the foramen of the first cervical vertebra. The rats were kept recovering for 24 h. The next day, animals were reanesthetized and both common carotid arteries (CCAs) were occluded for 20 min using microvascular clips to induce cerebral ischemia. Then, both clips were softly removed in order to reperfusion. During the whole operation and in the following 6 h, body temperature was maintained at 37 ± 0.5°C with a thermostatically controlled infrared lamp. Any rat returned to righting reflexes during the 20 min occlusion or epileptic seizure after ischemia was excluded. Animals in sham group were treated with anesthetics two times and for the same period of time as the surgical animals without occluding CCAs.

The protocol of Morris water maze task for rats has been detailedly described in our previous study (18). During the visible platform trials, rats were placed in the opaque water of the circular swimming pool (Jiliang Neuroscience Inc., Shanghai) measuring 180 cm in diameter. Four starting points around the edge of the pool were designed as N, E, S, and W, which divided the pool into four quadrants. A 10 cm diameter was located in a contrast position in the middle of one quadrant. The location of the visible platform varied for each trial. Four trials were administered. The latency to reach the visible platform and swimming speed were measured. Then, rats were trained to locate the hidden platform 1.2 cm under the surface of the water. The task for the rats was to escape from the water by locating the hidden platform. In the training to find the hidden platform, rats were allowed to swim for maximum of 60 s in the pool for each trail. One block of four trials per day was given for five consecutive days. For each trail, the rats were placed in the water facing the wall of the pool. Each trial was videotaped via a ceiling-mounted video camera and the animal’s movement was tracked using Ethovision software (Noldus Information Technology), which allows the calculation a series of parameters such as latency. Next day, rats were given one 60-s retention probe test in which the platform was removed from the pool. During retention, the time spent in the target quadrant and the number of crossing of the platform location was measured.

Golgi-Cox staining was performed with FD Rapid Golgi Stain TM Kit (FD NeuroTechnologies) as we previously reported (18). The rats were deep anesthetized by 10% chloral hydrate and killed on day 28 after TGI. The fresh brains were sliced into 10 mm thickness and placed in impregnation solution at room temperature. After 2 weeks, brains were transferred into Solution C and stored at 4°C in the dark for at least 48 h. Then, brains were cut into 100 µm coronal sections by vibratome (World Precision Instruments) and stained. For morphological analysis, 10 neurons randomly for each sample were measured and the average was regarded as the final value of one sample.

Immunofluorescence was performed as we previously reported (18). Animals were deep anesthetized by 10% chloral hydrate and perfused transcardially with 0.05 M sodium phosphate (pH 7.4) containing 0.8% NaCl, followed by 4% paraformaldehyde in 0.05 M sodium phosphate (pH 7.4, containing 0.8% NaCl). Brains were removed and postfixed overnight in the same solution. Serial hippocampal sections (40 µm) were made on an oscillating tissue slicer in a bath of physiological saline. The sections were heated (85°C for 5 min) in antigen-unmasking solution (Vector Laboratories); incubated in 2 M HCl (37°C for 30 min); rinsed in 0.1 M boric acid, pH 8.5, for 10 min; and blocked in PBS containing 3% normal goat serum, 0.3% (w/v) Triton X-100, and 0.1% bovine serum albumin at room temperature for 1 h, followed by incubation with primary antibodies at 4°C overnight. The primary antibodies used were as follows: rat anti-BrdU (1:200; Accurate Chemical & Scientific Corporation), mouse anti-NeuN (1:500; Millipore Bioscience Research Reagents). Subsequently, the sections were incubated with secondary antibodies goat anti-rat Cy3 (1:200; Millipore Bioscience Research Reagents), goat anti-mouse dylight 488 (1:400; Jackson ImmunoResearch). Appropriate horseradish peroxidase-linked secondary antibodies were used for detection by enhanced chemiluminescence (Pierce). An experimenter coded all slides from the experiments before quantitative analysis. To determine the total number of BrdU⁺/NeuN⁺ cells, the counts from sampled five sections were averaged, and the mean values were multiplied by the total number of sections. We used a confocal laser-scanning microscope (LSM700, Zeiss) to capture images and confirm colocalization of BrdU⁺/NeuN⁺ cells.

Culture of hippocampal neurons was performed as previously described (15). Primary neurons were isolated from E18 Sprague-Dawley rat hippocampus and cultured on dishes coated with polyornithine (10 µg/ml; Sigma-Aldrich) in Neurobasal medium (Invitrogen) containing 2% B27 supplement. The planting density 1 × 10⁵ cells/cm² for biochemical detection. Cell cultures were kept in a humidified atmosphere of 95% air and 5% CO₂ at 37°C. Half of the medium was replaced with fresh medium without glutamate every 2–3 days.

A recombinant lentivirus, LV-CREB133-GFP, was generated with the plasmid pCMV-CREB133 (Clontech Laboratories, Mountain View, CA, USA) as we previous reported (14). LV-CREB133-GFP or LV-GFP (control LV) were added into cultured neurons at 5 μl/dish (diameter 3.5 cm). Twenty-four hours later, the medium was fully changed with fresh medium without lentivirus. After infected with the recombinant lentivirus for 5 days, neurons were passaged for experiments. The procedures concerning recombinant lentivirus were performed following National Institutes of Health guidelines.

Oxygen glucose deprivation was performed as previously described with slight modification (19). Briefly, the primary hippocampal neurons were washed with glucose-free DMEM (Gibco, USA). Then, after the addition of glucose-free DMEM, they were placed in an anaerobic chamber containing 5% CO₂ and 95% N₂ at 37°C. We sealed cultures inside a modular chamber (Billups-Rothenberg) flushed for 10 min with the same premixed gas and placed inside an incubator for 3 h. OGD was terminated by replacing the exposure medium with neuronal culture medium added with Tat-HA-NR2B9c 250 nM, and returning the cells to a normoxic incubator to allow reperfusion for another 24 h. The cells cultured in the plain medium with ambient oxygen served as control (no exposure to OGD).

The cell viability was evaluated by using Hoechst 33258 (Sigma-Aldrich) nuclear staining as described previously (20), imaged with a fluorescence microscope (Axio Imager, Zeiss, Oberkochen, Germany), and analyzed with Image-Pro Plus software (Media Cybernetics, Silver Spring, MD, USA).

Western bolt analysis was performed as we previously reported (18). Primary antibodies were rabbit antiphospho-CREB-ser133 (1:1,000; Cell signaling Technology); rabbit anti-CREB (1:1,000; Cell signaling Technology); rabbit anti-Bcl-2 (1:1,000; Cell Signaling Technology). Internal control was mouse anti-β-actin (1:1,000; Sigma-Aldrich). Appropriate horseradish peroxidase-linked secondary antibodies were used for detection by enhanced chemiluminescence (Pierce).

Statistical analysis for escape latency in the MWM test: repeated measured ANOVA followed by a post hoc Bonferroni multiple comparison test. Comparisons among multiple groups were made with one-way ANOVA (one factor) or two-way ANOVA (two factors) followed by Scheffe post hoc test. Data were presented as mean ± SEM and P < 0.05 was considered statistically significant. Investigators were blind to the group allocation when assessing the outcome.

To examine whether Tat-HA-NR2B9c benefits TGI outcome, we subjected rats to 4-VO and treated them with Tat-HA-NR2B9c (1.12 mg/kg i.v.) for 7 days starting at reperfusion. Spatial cognitive performance in Morris water maze was measured during days 21–27 after TGI. Ischemia significantly increased escaping latency, decreased time spent in target quadrant and target crossings in Morris water maze (Figures 1A–C), compared with sham, suggesting impaired spatial memory. Tat-HA-NR2B9c significantly ameliorated the ischemia-induced impairments of spatial memory in latency (Figure 1A); in target crossings (Figure 1B); and in total time in target quadrant (Figure 1C) compared with vehicle. The swimming speed (Figure 1D) and latency to reach platform (Figure 1E) were measured at day 21 in visible platform trials, Tat-HA-NR2B9c have no effect on cued behavior and gross motor skills of animals (Figures 1D,E).

Transient global ischemia leads to selective delayed neuronal death of pyramidal neurons in hippocampus CA1. We investigated whether Tat-HA-NR2B9c increased neurogenesis in dentate gyrus and reduced the apoptosis of pyramidal neurons in hippocampus CA1 region. Tat-HA-NR2B9c was administered for 7 days beginning immediately after reperfusion. Rats were treated with BrdU (100 mg/kg × 2 i.v.; 24-h intervals) to label proliferating cells during days 1–2 after TGI. These rats were killed at day 28 after TGI to estimate the number of BrdU⁺ and BrdU⁺/NeuN⁺ (a marker for mature neurons) cells (Figure 2A). Treatment with Tat-HA-NR2B9c significantly increased neurogenesis in hippocampus dentate gyrus (Figures 2B,C) and reversed the ischemia-induced dendrite spine loss of pyramidal neurons in hippocampus CA1 region compared with vehicle (Figures 2D,E).

What is the underlying mechanism by which Tat-HA-NR2B9c improves the outcome of TGI? CREB is involved in the regulation of dendritic spine growth, neurogenesis, and synaptic plasticity. Activation of CREB leads to expression of genes such as Bcl-2 (17). It is well accepted that nNOS-derived NO exerts a negative control on the adult neurogenesis in normal and ischemic brains in vivo (14). NO negative control CREB activation (15), we thus investigated whether Tat-HA-NR2B9c affects CREB activation and expression of Bcl-2 after ischemia. To determine CREB activity, we examined phosphorylated CREB at Ser-133 (p-CREB-S133), an activated form of CREB. We subjected rats to TGI and measured levels of p-CREB, CREB, Bcl-2 at day 1 after TGI in hippocampus CA1 region. TGI caused a significant increase in CREB activity at 3 and 6 h postischemia and restore physiological levels at 24 h postischemia (Figures 3A,B). Tat-HA-NR2B9c increased the phosphorylation of CREB (Figures 3C,D) and elevated Bcl-2 (Figures 3E,F) expression in the hippocampal CA1 region at 24 h following global ischemia.

To determine whether hippocampal CREB activity is essential for the regulation of Tat-HA-NR2B9c to Bcl-2, we incubated the cultured hippocampal neurons with LV-CREB133-GFP, a recombinant lentivirus expressing a mutant variant of CREB protein, which could not be phosphorylated at Ser133. Tat-HA-NR2B9c activated phosphorylation of CREB (Figures 4A,B) and increased Bcl-2 expression (Figures 4C,D) in primary hippocampal neurons exposed to OGD 24 h later. LV-CREB133-GFP infection abolished Tat-HA-NR2B9c’s regulating on CREB activation (Figures 4A,B), and expression of Bcl-2 (Figures 4C,D). These findings indicate that hippocampus CREB mediated Tat-HA-NR2B9c regulation on Bcl-2. We also employed for the assessment of the numbers of viable cells, CREB phosphorilation, and Bcl-2 expression immediately after OGD. We found that there are not differences in neuronal viability immediately after OGD (Figure 5). This result suggested that the differences observed in these two proteins in Figure 4 are not due to different levels of neuronal death.