Animal model

Procedures in animals were performed according to the ARRIVE guidelines, the “Guide for the care and use of laboratory animals”, 8th edition, published by the National Institutes of Health (NIH), and the Colombian standards (Law 84/1989 and Resolution 8430/1993), taking special care to minimize suffering and reduce their number. The study was approved by the Ethics Committee for Animal Experimentation of the University of Antioquia, Medellín, Colombia.

Animals were obtained from a specific pathogen-free colony of the vivarium at the Sede de Investigación Universitaria (SIU), University of Antioquia, Medellín, Colombia. Rats were maintained in groups of five with food and water ad libitum and on a 12:12-hour dark:light cycle. We used 40 eight-week-old male Wistar rats (Rattus novergicus albinus) of average size and an average weight of 350 g. Ten rats per experimental group were used for neurological scoring. In addition, we used 5 per experimental group for the histologic assessments.

Surgical procedure

We performed a transient middle cerebral artery occlusion (t-MCAO) and reperfusion as described previously ^12,14. Initially, animals were anaesthetized with a mixture of ketamine via IP (60 mg/kg; Holliday Scott S.A. Int. Neyer, Buenos Aires, Argentina), xylazine (5 mg/kg; Synthesis Ltda. & Cia., S.C.A, Bogotá, Colombia), subcutaneous atropine (100 mg/kg; Erma S.A., Bogotá, Colombia); they also received an inhalational mixture of 2% to 4% of isoflurane (Baxter, Deerfield, IL, USA) and 96% oxygen using an inhalation anaesthesia machine (VMC™ Table top gas anaesthesia Instrument).

The right common carotid artery (CCA) was exposed and carefully isolated from the vagus nerve (VN); the right external and internal carotid arteries (ECA and ICA) were exposed and isolated. ECA ascending pharyngeal artery, superior thyroid and occipital branches were cauterized by electrocoagulation (AARON Bipolar Cautery™, Albany, NY, USA). The ICA artery pterygopalatine first branch was isolated. The distal portion of the ECA was tied with a 6/0 nylon and then cut. A 4-0 monofilament nylon (Corpaul, Bogotá, Colombia), previously rounded at one end by flaming and coated with poly-L-lysine solution (0.1% w/v in deionized water; Sigma, St Louis, MO, USA), and inserted 17-19 mm from the ECA to the ICA to occlude the MCA and its origin. Then, we verified that the monofilament was not inside the pterygopalatine artery. The incision was sutured leaving a bulge of 1 cm. After 60 minutes, the nylon was removed allowing reperfusion and the wound was closed.

Sham animals were subjected to the same procedures, but the filament was not introduced. Half of the animals were euthanized 15 days after reperfusion and the other half, 30 days after for which all animals were anesthetized with xylazine 10 mg/kg + ketamine 90 mg/kg and immediately perfused with 4% paraformaldehyde (0.1 M phosphate buffer [PBS], pH 7.4). Brains were carefully removed and post-fixed with 4% paraformaldehyde at 4°C for 48 hours, then preserved by a sucrose gradient (7%, 25%, 30%) at 4°C; they were then placed in a cryopreservation solution at -20°C. Coronal sections were sectioned at 50 μm with a vibrating blade microtome (VT1000S™, Leica Microsystems, Nussloch, Germany).

Short hairpin RNAmiR delivery

During the t-MCA occlusion, at 30 minutes, we performed a right hippocampal stereotaxic injection with anteroposterior (AP) coordinates of -2.56 mm, mediolateral (ML) 0.8 mm and dorsoventral (DV) 4.1 mm. We used a 10 μL Hamilton syringe for injection at a rate of 0.2 μL/min of a solution containing adeno-associated viral (AAV) particles obtained from the Davidson Laboratory (University of Iowa Viral Vector Core) with the RNAi sequences for silencing CDK5 (short hairpin CDK5miR) and a control scrambled RNA sequence (shSCRmiR) ^12,15-18 for treating ischemic and control animals, respectively. Five minutes elapsed after the infusion before withdrawing the syringe. We injected a volume of 2.5 μl of AAV2.5.shSCRmiR.GFP or AAV2.5.shCDK5miR.GFP as previously described ^12,18. After the injection, we proceeded to suture the animals keeping them in observation until their awakening (figure 1 A,B).

Figure 1 Neurological recovery induced by CDK5 silencing in ischemic rats. A) Experimental design to evaluate the effect of CDK5 RNAi on neural population and neurological performance after an ischemic insult. B) Ipsilateral injection site at stereotaxic coordinates AP -2.56, ML 0.8, and DV 4.1. and evaluated hippocampal regions (g: granular; sg: subgranular; p: polimorphic zone. C) Neurological analysis of CDK5 silencing in ischemic rats. The error bars indicate the SEM. All data are presented as mean ± SEM. Sham-SCR: Control group treated with shSCRmiR; Sham-CDK5: Control group treated with shCDK5miR; Isch-SCR: Ischemic group treated with SCRmiR; Isch- CDK5: Ischemic group treated with shCDK5miR. n=10, p<0.05 = *; p<0.001 = **.  

Recovery and neurological evaluation

Animals were evaluated from day 1 to 15 after the t-MCAO (N=10). Neurological evaluation was scored on an 18-point scale. The evaluation was divided into six independent tests scored from 0 to 3 to determine the following parameters:

The highest possible score in the tests was 18, which indicated there were no neurological deficits, while the lowest was 3 for the most severe impairment. Neurological evaluation was performed every day and in the same order in all animals. To validate and reduce bias, a double-blind procedure was performed recording videos for every animal during the test.

Insertion of 5-Bromo-2-deuxyuridine

For the BrdU (Catalog Number B5002, Sigma- Aldrich) insertion, we injected rats via IP twice a day during seven days after ischemia at 50 mg/kg (0.1 M phosphate buffer [PBS], pH 7.4).

Immunohistochemistry

Brain sections were treated with methanol and hydrogen peroxide (H2O2) in 0.1M PBS (pH 7.4) to inhibit endogenous peroxidase activity for 20 minutes while free-floating with moderate shaking. Subsequently, we washed them three times with 0.3% (v/v) Triton X-100™ (Sigma-Aldrich) in 0.1 M PBS to increase tissue permeability and we then incubated for 30 minutes with Triton™ 0.5% (v / v) X-100 in 0.1 M PBS. Then, we backwashed with 0.3% (v/v) Triton X-100™ in 0.1 M PBS for three times.

For BrdU, we denaturalized brain DNA sections with HCl 2N for 20 minutes at 37°C, we washed three times with 0.3% (v/v) Triton X-100™ in 0.1 M PBS, then incubated with citrate buffer (Sodium citrate 10Mm, 0.05% tween 20, pH 6.0) for 10 minutes at room temperature and, finally, we washed three times with 0.3% (v/v) TritonX-100™ in 0.1 M PBS. Nonspecific binding sites were blocked by preincubation with BSA (1%) and 0.3% (v/v) Triton X-100™ in 0.1 M PBS for 1 hour.

Sections were incubated 24 hours while freefloating with moderate shaking and with the primary anti-body diluted in incubation solution (0.1 M PBS with 0.3% (v/v) Triton X-100™ and 0.3% BSA) at 4°C.

Anti-BrdU (1:100 mouse monoclonal B8434, Sigma- Aldrich) was used to detect bromodeoxyuridine (BrdU) specifically incorporated into the DNA. Next day, sections were washed 5 minutes with 0.1 M PBS three times and incubated with the secondary antibody (1:250 biotinylated mouse, Pierce) for 1 hour at room temperature and then washed 3 times with 0.1 M PBS, incubated with Avidin-Biotin complex (Pierce complex™; 1: 250 reagents A and B), and washed again three more times and developed with a chromogenic reaction diaminobenzidine (DBA) in 0.01% H2O2. For negative controls, we followed the same protocol for each experimental group without incubation with the respective primary antibody.

Sections were mounted on glass slides, dehydrated with increasing concentration of alcohol (70%, 96%, 100%) and cleared with xylene. Finally, the slides were sealed using Vecta-shield mounting medium (Vector Laboratories, Burlingame, CA, USA), and then analyzed through a light microscope (Nikon Eclipse E200™). Images were captured using a Nikon digital camera (Sight DS-L1™). Quantification of immunoreactivity in the hippocampus was determined using a 10X objective. Photomicrographs were analyzed using the ImageJ software program (version 1.45, US National Institutes of Health, Bethesda, MD, USA). To quantify the relative intensity of immunostaining, we modified the images to a binary system and obtained integrated densities (relative units) substracting automatically the background for each image.

Immunofluorescence

Brain slices were initially heated (80°C) in 10mM Tris pH 6.0 to increase exposure of the antigen and then treated with ammonium chloride (NH4Cl) for 20 minutes to avoid autofluorescence. Three washes with 0.3% v/v Triton X-100™ in 0.1M PBS were made to increase tissue permeability. For BrdU, we denaturalized DNA brain sections with HCl 2N for 20 minutes at 37°C, washed them three times with 0.3% (v/v) Triton X-100™ in 0.1 M PBS, and then incubated them with citrate buffer (sodium citrate 10 Mm, 0.05% tween 20, pH 6.0) for 10 minutes at room temperature to finally wash them three times with 0.3% (v/v) Triton X-100™ in 0.1 M PBS. Nonspecific binding sites were blocked by preincubation with BSA (3%) and 0.5% (v/v) Triton X-100™ in 0.1 M PBS for 1 hour.

Sections were incubated overnight while freefloating with moderate shaking at 4°C with the respective primary anti-BrdU (mouse monoclonal 1: 100, Sigma) for the new cells produced, anti-DCX (polyclonal rabbit 1: 250, Abcam) for immature neurons, anti-GFAP (rabbit polyclonal 1: 500, Sigma) for astrocytic cells and anti-NeuN (polyclonal rabbit 1: 500, Millipore) for mature neuronal cells. After three washes, tissues were incubated with secondary antibodies: Rabbit Alexa Fluor 594™, and mouse Alexa Fluor 350™ (molecular probes, 1:1,000) for 90 minutes at room temperature. Finally, we washed with PBS and slides were coverslipped using Gel Mount™ (BioMeda, Hatfield, PA, USA) in dark conditions. For negative controls, we followed the same protocol for each experimental group without incubation with the respective primary antibodies. Tissues were observed with a motorized spinning disk confocal microscope (Disk Scan Unit) Olympus IX 81™ (Olympus, Latin America, Inc., Miami, FL, USA). Images were photographed at 10X and the analysis was performed by evaluating the fluorescence intensity (FI) using the Image Pro Plus (Media Cybernetics) software.

Statistical analysis

For the neurological analysis, we used ten (10) animals per group and for the histological analysis, five ⁵ per group. The data obtained from the neurological and tissue immunoreactivity evaluations were analyzed using a homogeneity of variance test, followed by ANOVA of repeated measurement and one-way ANOVA tests, and a Tukey’s post hoc test to define differences between groups. We used the SPSS Statistics™ software (20.0 version) and a p ≤0.05 significance. All experimental groups were processed in parallel for reducing variability. The data are presented as mean ± SEM.

CDK5 reduction promoted neurological recovery in ischemic rats

The neurological assessment showed that the Isch-SCR animals had a lower score (12.8 and 14.3 points) at days 7 and 15 post-ischemia, respectively, compared with the sham groups with and without CDK5 RNAi treatment (16.5 points). Interestingly, the CDK5 RNAi-treated ischemic group presented a significantly higher score of 14.5 on day 7 and of 16 points on day 15, recovering neurological skills similar to those of the control animals (figure 1C; p<0.05 = *, p<0.001 = **, repeated-measures in ANOVA).

BrdU incorporation in the hippocampus was promoted by CDK5 RNAi treatment in ischemic rats

We found a significantly increased number of BrdU+ cells in the Isch-SCR group compared with the sham-SCR and sham-CDK5 controls (figure 2A a; p<0.05 = *, p<0.001 = **, repeatedmeasures in ANOVA and Tukey’s test). However, the CDK5 RNAi treatment induced a significantly greater incorporation of BrdU in ischemic animals compared with the Isch-SCR group without treatment (figure 2A a). Although there was no significant incorporation of BrdU in the polymorphic area of the Isch-SCR animals compared to controls (figure 2A b; p<0.05 = *, p<0.001 = **, repeated-measures in ANOVA and Tukey’s test), CDK5 silencing significantly increased BrdU+ cells compared to Isch-SCR and controls in the cerebral zone (figure 2A b). This effect of CDK5 RNAi on ischemic rats remained for a month in the granular/ subgranular and polymorphic zones (figure 2 B b). The Isch-group had a slight tendency or significantly increased BrdU+ cells in comparison with the control groups (figure 2 B a, b).

Figure 2 BrdU incorporation in the hippocampus of CDK5 RNAi-treated ischemic rats. A) Representative images at bregma -2.56 mm of BrdU immunoreactivity in the granular-subgranular (a) and polymorphic (b) zones at 15 days and B) 28 days post-ischemia in treated and untreated rats. The error bars indicate the SEM. n= 5, p<0.05 = *; p<0.001 = **. All data are presented as mean ± SEM. 10X. Scale bar=100 µm. 

CDK5 RNAi exerted neuroprotection associated with the induction of immature neurons and down-regulation of the astroglial population in ischemic rats

Our findings showed that CDK5 RNAi treatment induced proliferation of immature neurons in the granular/subgranular zones. This is supported by the significant increase of BrdU+ and DCX+ cells in comparison with ischemic and control groups at 15 days post-ischemia (figure 3A, C; p<0.05 = *, p<0.001 = **, repeated-measures in ANOVA and Tukey’s test). Although the sham-CDK5 and Isch- SCR groups presented a significant increase of BrdU in the granular/subgranular zones, this was not confirmed by the variations in the amount of DCX+ or GFAP+ cells (figure 3A, C). On the other side, the polymorphic zone presented variability between groups and treatments (figure 3B, c; p < 0.05 = *, p<0.001 = **, repeated-measures in ANOVA and Tukey’s test), however, there was a significant increase of BrdU+ and GFAP+ cells in the ischemic-SCR group 15 days post-ischemia, which was significantly down-regulated by the CDK5 RNAi treatment in ischemic rats (figure 3B, C). The green fluorescent protein (GFP) was used to confirm the expression of AAV2.5.shSCRmiR. GFP and AAV2.5.shCDK5miR.GFP in each group, respectively.

Figure 3 CDK5 RNAi induced immature neurons and down-regulation of astroglial population in ischemic rats. A) BrdU and DCX immunoreactivities in the granular-subgranular and polymorphic zones at 15 days post-ischemia in treated and untreated rats (A, C). BrdU and GFAP in the granular-subgranular and polymorphic zones (B, C) at 15 days post-ischemia in treated and untreated rats. The error bars indicate the SEM. n=5, p<0.05 = *; p<0.001 = **. All data are presented as mean ± SEM. Representative images at bregma -2.56 mm. 10X. Scale bar = 100 µm. 

CDK5 RNAi preserved the down-regulation of astrogliosis at one month post-ischemia

Interestingly, when we evaluated the neural population at 30 days post-ischemia, we did not find changes in the mature neuron immunoreactivity neither in the granular/subgranular nor in the polymorphic zones in the untreated and treated rats, because there was a significant increase in the BrdU+ cells but without changes in the NeuN immunoreactivity (figure 4A, B; p<0.05 = *, p<0.001 = **, repeated measures in ANOVA and Tukey’s test). The GFAP+ population did not present changes in the granular/subgranular zone either (figure 4A, B), but stayed increased in the polymorphic zone at one-month post-ischemia in untreated rats, which was reversed by the CDK5 RNAi treatment as in the control groups (figure 4A, C, p<0.05 = *, p<0.001 = **, repeated measures in ANOVA and Tukey’s test).

Figure 4 CDK5 RNAi prevents astrogliosis at one-month post-ischemia. A) NeuN, BrdU and GFAP immunoreactivities in the granular-subgranular (A, B) and polymorphic (A, C) zones at 30 days post-ischemia in treated and untreated rats. The error bars indicate the SEM. All data are presented as mean+/-SEM. Representative images at bregma -2.56 mm. n=5, p<0.05 = *; p<0.001 =**. 10X. Scale bar = 100 µm