Activation of caspase-activated deoxyribonuclease and neuroprotective effect of caspase-3 inhibitor after focal cerebral ischemia-reperfusion injury

Caspase-3 is a key enzyme to execute apoptosis and may be cause internucleosomal DNA fragmentation in ischemic neurons. However, whether caspase-3 inhibitor can directly inhibit caspase3–dependent deoxyribonuclease activity and prevent neuronal apoptosis following cerebral ischemic is unknown. In this study, we detected the caspase-3 and CAD protein, as well as the frequencies of neuronal apoptosis in both model control group (DMSO injection) and treatment group (Ac-DEVD-CHO injection) after focal cerebral ischemia-reperfusion in rats. Our data showed that caspase-3 and CAD protein were detectable, and apoptotic-like neuronal death occurred following cerebral ischemic in both groups. However, these results obtained were inhibited by Ac-DEVD-CHO in treatment group as compared with model control group. Taken together, these data further support that the pathway of caspase-3–dependent CAD activity and neuronal apoptosis is an important mechanism in ischemic neuronal injury, and Ac-DEVD-CHO has the neuroprotective effect to a certain degree.

Previous studies have suggested that caspase-3 dependent CAD activity play an essential role during neuronal apoptosis after ischemia in the brain (Chen et al., 1998;Tsukada et al., 2001;Luo et al., 2002;Abas et al., 2010;Yan et al., 2010).Hence, antiapoptotic mechanisms through caspase inhibition may play neuroprotective role in the brain after focal ischemic injury.Le et al. (2002) has reported that caspase-3 gene knockout mice were more resistant to ischemic injury than Wistar mice; however, the study using caspase-3 gene knockout mice was limited by the lack of practicability.Therefore, in this study we aimed to demonstrate the neuroprotective effect of caspase-3 inhibitor that will be more likely utilized in future, in a rat model of focal cerebral ischemia and reperfusion.

Animal grouping
Sixty-four adult male Wistar rats, weighing 200 -250 g were randomly divided into the model control group (n = 32) and treatment group (n = 32).They were observed at 6, 12, 24, 48 and 72 h of ischemia, respectively.One more rat was included in the 48 and 72 h groups for electron microscopic observation.This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health.Animal care and experimental protocols were approved by the animal ethics committee of 2nd Affiliated Hospital of Harbin Medical University.

Model preparation
Focal ischemia was induced by occlusion to the right middle cerebral artery (MCAO) as described previously (Nagasawa and Kogure., 1989).After the anesthesia with 10% chloral hydrate (300 mg/kg) in rats, a nylon thread (0.17 mm in diameter and 0.24 mm in distal cylinder) was inserted into the lumen of the internal carotid artery and advanced to the origin of the MCAO.Then, rats were subjected to intra-cerebral ventricular infusion.The model control group was injected with 4 μL of 0.3% dimethyl sulfoxide (DMSO) for controls, whereas treatment group was injected with 5 μg of Ac-DEVD-CHO (Promega, USA) containing 4 μL 0.3% DMSO.The thread was removed 1 h later to allow for reperfusion.The rats were enrolled according to neurological findings, including left forelimb flexion or circling 1 h after MCAO.

Sample preparation
One rat was sacrificed at 48 and 72 h in both groups.The temporoparietal lobe cortex tissue of about 1 mm was taken from the right cerebral hemisphere.It was then fixed with glutaraldehyde for electron microscopic observation (n = 4).Other rats were decapitated under anesthesia at different time points (n = 60).After 12 h external fixation, coronary brain slices (2 mm anterior and posterior to the optic chiasma) were cut and embedded with paraffin for section preparation.

Hematoxylin and Eosin (HE) and immunohistochemistry staining
Tissue sections were performed with routine HE staining.Immunohistochemistry was performed according to the instructions indicated in the kit (PV-6001, Zhongshan, China).Cells with a buffystained nucleus or cytoplasma were defined positive.Tissue sections were deparaffinized, hydrated and then incubated with 3% H2O2 deionized water for 5 -10 min.They were incubated with at 37°C rabbit anti-mouse caspase-3 and CAD polyclonal antibodies for 1 -2 h.After the incubation with horseradish peroxidase (HRP)conjugated mouse anti-rabbit IgG antibody at room temperature (37°C) for 20 -30 min, the sections were DAB-stained.The number of caspase-3 and CAD positive cells was recorded in 5 randomly chosen areas on the temporo-parietal cortex tissue under the microscopy (40X objective), and then mean values were calculated.

TUNLE staining
The staining procedures were performed according to the instructions indicated in the kit (Roche, USA).Sections adjacent to those for immunohistochemistry were collected, processed with hydrogen peroxide-methanol and citrate sodium, and then incubated with TUNLE reaction mixture at 37°C for 60 min.They were incubated with peroxidase-labeled anti-fluorescence antibodies at 37°C for 30 min.The substrates were DAB-stained at room temperature away from light.The shade of staining was controlled under a microscope.The sections were dehydrated, cleared, and then mounted regularly.Neurons with buffy-stained granules in the nucleus were defined positive.The number of apoptotic neurons was recorded in 5 randomly chosen areas on the temporo-parietal cortex tissue under the microscopy (40x objective), and then mean values were calculated.

Electron microscopy
The processed sections were observed under a JEM-1220 transmission electron microscope, and photographs were taken.

Statistical analysis
All data were presented as mean ± SD.Comparison between paired and unpaired groups was based on Bartlett's test for homogeneity of variances, and analysis of variance (ANOVA) or nonparametric ANOVA test was chosen by SPSS 13.0 software.P < 0.05 was considered statistically significant.

Pathologic features after ischemia
Using HE staining, we observed no obvious morphological change at 6 h after reperfusion in the model control group (Figure 1).At 12 h, karyopyknosis was found in a few nerve cells and moderate edema was found around the neurons.With the prolongation of time, the cytoplasm acidophily increased gradually, and numerous apoptotic nerve cells were observed at 48 h (Figure 1).In the treatment group, cellular edema was relieved at the early stage compared to the model control group; however, no obvious difference in apoptosis was found at the late stage.
By electron microscopy, intra-neuronal chromatolysis and formation of vacuole-like structure was found in both groups at 24 and 48 h (Figure 2).The structure of the organelles in the cytoplasm changed, and lipofuscin granules, blurred mitochondrial crista and decreased metrical density could be seen.The vascular nerves showed obviously broadened interspaces, the fibers were lysed, and the lumens were narrowed.Some cell membranes were ruptured, and necrosis and apoptosis were present.However, no typical apoptotic bodies were found.

Caspase-3 protein alteration after ischemia
In the model control group, a few caspase-3 positive cells were observed at 6 h after reperfusion, and they were significantly increased from 12 to 72 h, and reached the peak at 24 h.Caspase-3 protein was mainly located in cytoplasm at 12 h, and was shown in cytoplasm and nucleus at 24, then transferred from cytoplasm into the nucleus at 48 h.In the treatment group, the changes showed a similar trend to that in the model control group, but caspase-3 positive cells were significantly decreased 12 to 48 h (P < 0.05) (Figure 1).The data were shown in Figure 3 and Table 1.

CAD protein alteration after ischemia
In the model control group, CAD positive cells were rarely detected at 6 h.At 12 h, CAD protein began to concentrate around the nucleus in a few cells.At 48 h, it spread all over the nucleus, and the frequency of CAD positive cells reached the peak (Figure 1).In the treatment group, CAD positive cells also reached the peak at 48 h, but significant decrease were found from 24 to 72 h (P < 0.05) and not at 6 h and 12 h (P > 0.05)    (Figure 1).The data are shown in Figure 4.

Neuroprotective effect of Ac-DEVD-CHO
In situ apoptosis was detected using TUNLE staining.In the model control group, few positive cells were found at 6 h.With the prolongation of time, the number of apoptotic cells increased, and reached the peak at 72 h.In the treatment group, the number of apoptotic cells at 6 h and 12 h showed no significant difference (P > 0.05), while at 24 to 72 h, the results displayed significantly decreased as compared with model control group (P < 0.05), which was especially obvious at 72 h.The alteration suggests that Ac-DEVD-CHO inhibit the induction of apoptosis after an ischemic insult.Representative TUNLE staining is presented in Figure 5, and the data are shown in Figure 6 and Table 1.

DISCUSSION
Utilizing a rat model of MCAO, the neuroprotective effects of caspase-3 inhibitor Ac-DEVD-CHO was confirmed in our study.The major results demonstrated that Ac-DEVD-CHO injection prior to reperfusion reduced up-regulation   of caspase-3 and CAD and decreased neuronal apoptosis following cerebral ischemia.Accumulating evidence indicate that the caspase-family is the promoter and implementer of apoptosis in mammalian cells, and caspase-3 is the most critical downstream apoptosis protease in the caspase cascade "waterfall" (Le et al., 2002;Cho and Toledo-Pereyra, 2008).In the present study, we found that there were significant increases in caspase-3 and CAD protein and apoptotic cells in a timedependent manner after focal ischemia and reperfusion in the brain.The up-regulated expression of caspase-3 protein first reached the peak (at 24 h, and then began to decrease), then CAD protein spread all over the nucleus and reach the peak and at 48 h.However, a large quantity of TUNEL positive cells appeared up to 72 h after ischemia-reperfusion.These results revealed a dynamic apoptotic process encompassing caspase-3 activation, caspase-3 nuclear transfer, CAD protein up-regulation, CAD translocation, DNA degradation (TUNLE positive cells).This process is an important pathway to neuronal injuries in rats after cerebral ischemia-reperfusion.The time-dependent relationship between caspase-3 and apoptosis further supports the theory that caspase-3mediated mechanism play a key role in the final execution of neuronal apoptosis in various forms of central nervous system injuries (Yakovlev et al., 1997;Namura et al., 1998;Springer et al., 1999;Zhang et al., 1999;Citron et al., 2000;Clark et al., 2000;Sharp et al., 2000;Cao et al., 2001;Graham and Chen, 2001;Tsukada et al., 2001).Furthermore, we observed that caspase-3 inhibitor Ac-DEVD-CHO diminished CAD activation and prevented endogenous DNA fragmentation after ischemia.In brief, 4.5 μg of z-DEVD-fmk (also a caspase-3 inhibitor) has the best inhibitory effect on caspse-3 activity and the optimal time window is within 6 h after ischemia in a 30 min focal ischemic model (Luo et al., 2002;Wei et al., 2004).Based on the aforementioned views, in the current study, 5 μg of Ac-DEVD-CHO was selected as the therapy dose.Despite inhibition of the caspase-3 activity by Ac-DEVD-CHO, we observed a large number of apoptotic cells in the treatment group after ischemia.Therefore, it is certain that neuronal apoptosis is a complex process in which many factors are involved, and the inhibition of CAD can only partially inhibit the process (Didenko et al., 2002;Nielsen et al., 2008).Apoptotic response originates from the release of cytochrome oxidase C, and then proceeds with the activation of caspase-3 and other multi-target genes (such as PARP, CAD and other nucleate endonucleases), while CAD is only one of the multiple target genes.This is consistent to the findings of Le et al. (2002) that genetic deletion of caspase-3 renders neurons more resistant to ischemic stress, but features of apoptotic-like neuronal death remained be observed both in vivo and in vitro, despite deletion of the caspase-3 gene.These evidences suggest that caspase-independent mechanisms could play a critical role in the death process as well.

Caspase
In summary, caspase-3 dependent CAD activity and neuronal apoptosis is induced in the brain after focal ischemia and reperfusion.Caspase-3 inhibition can suppress ischemic injury and play the role of neuroprotection.Taken together, blockage of caspase may be an effective treatment for stroke in future.

Figure 1 .
Figure 1.Representative photomicrographs (400× magnification) of HE staining (A and B) and immuno-histochemistry staining for caspase-3 (C and D) and CAD (E and F).HE staining: (A) Normal nerve cells (red arrows), (B) numerous apoptotic nerve cells (red arrows) after 48 h reperfusion.Caspase-3 staining: (C and D) positive cells were stained brown (red arrows), both in the cytoplasm and nucleus, higher caspase-3 positive cells in model control group (C) than in treatment group (D) after 48 h reperfusion.CAD staining: (E and F) positive cells were stained brown (red arrows), CAD protein stained concentrated around the nucleus in the treatment group (F) or spread all over the nucleus in the model control group (E) after 48 h reperfusion, and comparison of the treatment group with model control group showed that CAD positive cells decreased significantly.

Figure 2 .
Figure 2. Electron micrographs showing the formation of vacuolelike structure and chromatin condensed in margin (red arrows) after 48 h reperfusion.

Figure 3 .
Figure 3. Comparisons of the caspase-3 positive cells between the model control and treatment group at each time point after reperfusion.*P < 0.05 compared with model control group at the same time point.

Figure 4 .
Figure 4. Comparisons of the CAD positive cells between the model control and treatment group at each time point after reperfusion.*P < 0.05 compared with model control group at the same time point.

Figure 5 .
Figure 5. Representative photomicrographs (400× magnification) of TUNEL staining from model control group (A) and treatment group (B) after 72 h reperfusion.TUNEL positive cells showed a condensed, shrunken, or fragmented nucleus (red arrows).TUNEL positive cells were significantly reduced in treatment group as compared with model control group.

Figure 6 .
Figure 6.Comparisons of the TUNLE positive cells between the model control and treatment group at each time point after reperfusion.*P < 0.05 compared with model control group at the same time point.

Table 1 .
Caspase-3, CAD and TUNEL positive cells in the model group and treatment group at each time point after reperfusion.