Sestrin2 Silencing Exacerbates Cerebral Ischemia/Reperfusion Injury by Decreasing Mitochondrial Biogenesis through the AMPK/PGC-1α Pathway in Rats

Sestrin2 (Sesn2) exerts neuroprotective properties in some neurodegenerative diseases. However, the role of Sesn2 in stroke is unclear. The AMP-activated protein kinase/peroxisome proliferator-activated receptor γ coactivator-1α (AMPK/PGC-1α) pathway plays an important role in regulating mitochondrial biogenesis, which helps prevent cerebral ischemia/reperfusion (I/R) injury. Here, we aimed to determine whether Sesn2 alleviated I/R damage by regulating mitochondrial biogenesis through the AMPK/PGC-1α signaling pathway. To be able to test this, Sprague-Dawley rats were subjected to middle cerebral artery occlusion (MCAO) for 1 h with Sesn2 silencing. At 24 h after reperfusion, we found that neurological deficits were exacerbated, infarct volume was enlarged, and oxidative stress and neuronal damage were greater in the Sesn2 siRNA group than in the MCAO group. To explore protective mechanisms, an AMPK activator was used. Expression levels of Sesn2, p-AMPK, PGC-1α, NRF-1, TFAM, SOD2, and UCP2 were significantly increased following cerebral I/R. However, upregulation of these proteins was prevented by Sesn2 small interfering RNA (siRNA). In contrast, activation of AMPK with 5′-aminoimidazole-4-carboxamide riboside weakened the effects of Sesn2 siRNA. These results suggest that Sesn2 silencing may suppress mitochondrial biogenesis, reduce mitochondrial biological activity, and finally aggravate cerebral I/R injury through inhibiting the AMPK/PGC-1α pathway.

Sesn2 siRNA (The sense primer 5-GCGAGAUCAACAAAUUACUTT-3 and antisense primer 5-AGUAAUUUGUUGAUCUCGCTT-3) was designed and chemically synthesized by GenePharma Corporation, Shanghai, China. Scramble siRNA, which has the same nucleotide composition of the target gene siRNA with no sequence homology to any known rat genes was used as the control. Rats were anesthetized with 3.5% chloral hydrate (350 mg/kg, ip) and placed in a stereotaxic apparatus (Taimeng Software, Chengdu, China). AICAR (Acadesine) (S1802), an activator of AMPK was purchased from Selleck, USA. 5 μl AICAR (10 μg/μl) or siRNA (2 μg/μl) was slowly injected into the left lateral ventricle over a 20-min duration using a Hamilton microsyringe with the coordinates of 1.0 mm posterior to the bregma, 2.0 mm lateral to the midline, and 3.5 mm ventral to the surface of the skull under the guidance of a stereotaxic instrument. The injection rate was 0.5 μl/min. After injection, the needle was held in place for 5 min and then removed slowly over 2 min. 24 h after the injection of siRNA, MCAO model was performed and AICAR was injected 30 min before MCAO model performed.

Neurological deficit scores were evaluated by an examiner blinded to the experimental groups after 24 h of reperfusion. The deficits were scored on a modified scoring system developed by Longa et al. 15 , as follows: 0, no neurological deficits; 1, failure to extend right forepaw fully; 2, circling to right; 3, falling to right; 4, did not walk spontaneously and has depressed levels of consciousness. The higher the neurological deficit score, the more severe the impairment is of motor motion injury. Brains from these rats were analyzed for infarct volume, histological assessment, oxidative stress analysis, western blot, and real-time qPCR.

After neurological examination, rats were anesthetized with 3.5% chloral hydrate (350 mg/kg) 24 h after reperfusion, and brains were perfused transcardially with 0.9% sodium chloride followed by 4% paraformaldehyde. After decapitation, brains were removed and immersed in paraformaldehyde for 24 h. After that, brains were dehydrated with automatic tissue hydroextractor and embedded in paraffin. Then, 5-μm coronal sections 1.20 mm anterior to and 3.6 mm posterior to the bregma were stained with 0.1% cresyl violet (Nissl staining) or hematoxylin–eosin (HE staining) according to standard protocols and prepared for subsequent microscopic mounting. Intact neurons were counted in tenfields for each section under a light microscope (400×). The cell count in the border zone around the ischemic core were obtained from same group, averaged and expressed as cell number/mm 2 16 .

The brains of rats were removed rapidly after saline perfusion and rinsed in 0.9% normal saline (4 °C) to wash away the blood and blood clot. And then parts of cortex were homogenized in RIPA buffer. Lysate was centrifuged at 12000 g at 4 °C for 20 min, and the supernatant was collected. The protein concentration was estimated by the Enhanced BCA Protein Assay Kit (Beyotime, Jiangsu, China). Equal amount of protein was separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and then transferred onto nitro-cellulose membranes (Millopore, CA, USA). The membranes were blocked in 5% non-fat milk containing 20 mM Tris-HCl, pH 7.6, 137 mM NaCl, 0.1% Tween-20 (TBS-T) and then incubated at 4 °C overnight with primary antibodies against Sesn2(1:500, Proteintech, USA), AMPK(1:1000, Proteintech, USA), phospho-AMPK (1:500, Bioworld, USA), PGC-1α (1:1000, Abcam, Cambridge, MA, USA), NRF-1(1:500, Bioworld, USA), TFAM (1:1000, Bioworld, USA), SOD2 (1:2000, Bioworld, USA), UCP2 (1: 500, Bioworld, USA), cytochrome c(1:1000, Abcam, Cambridge, MA, USA), AIF(1:1000, Santa Cruz, Dallas, TX, USA) and β-actin(1:10000, Bioworld, USA). After that, it was incubated with HRP-conjugated secondary antibody (1:5000) for one hour at room temperature. The blotted protein bands were visualized by enhanced chemiluminescence (ECL). The density of bands was quantified using NIH Image J, and the data were normalized to β-actin.

Data was expressed as mean ± standard error of the mean. Analysis was performed using GraphPad Prism software. Statistical differences between two groups were analyzed using Student’s unpaired, two-tailed t-test. Multiple comparisons (without rating scale data) were statistically analyzed with one-way analysis of variance (ANOVA) followed by Student-Newman-Keuls test. The neurological deficit scores were analyzed using the Mann-Whitney U test. A probability level of less than 0.05 was considered statistically significant.

HE and Nissl staining were used to assess morphological changes 24 h after reperfusion ( Fig. 2A ). The area examined was shown in Fig. 2B . HE staining showed that, compared with neurocyte in the sham group, neurocyte in the MCAO and scramble siRNA groups displayed a disorderly arrangement with loosened cytoplasm and karyopyknosis. However, the Sesn2 siRNA group showed diffuse vacuolization and edema in the interstitial spaces and a decrease in the number of intact neurocyte ( Fig. 2C ). Similar results were obtained with Nissl staining, which showed that there were no morphological changes of neurons in the sham group. In contrast, many atrophic neurons with shrunken cytoplasm and damaged nuclei were observed in the MCAO and scramble siRNA groups. Sesn2 siRNA increased the number of degenerated neurons, and most neurons showed loss of Nissl bodies ( p < 0.001). Oxidative stress was examined by measuring SOD activity and MDA content in ischemic cortical tissue. A significant decrease in SOD activity was observed in rats subjected to I/R ( p < 0.001, Fig. 2D ). Animals injected with Sesn2 siRNA showed a further reduction of SOD activity compared with the MCAO group (29.51 ± 2.54 versus 55.19 ± 4.85 U/mg, p = 0.0076). The results of MDA content analysis are shown in Fig. 2E . Compared with the sham group, animals in the MCAO group showed significant increases in MDA levels in the ischemic cortex (7.78 ± 0.39 versus 3.91 ± 0.24 μmol/mg, p = 0.0013). Furthermore, the MDA level was greater in the Sesn2 siRNA group than in the MCAO group (11.01 ± 0.53 versus 7.78 ± 0.39 μmol/mg, p = 0.0042).

Protein and gene expression of NRF-1, TFAM, SOD2, and UCP2 were analyzed by western blot and qPCR, respectively ( Fig. 4 ). In the MCAO group, protein levels and gene expression of NRF-1 and TFAM significantly increased ( p < 0.001), accompanied by increases in SOD2 and UCP2 expression ( p < 0.001). Sesn2 siRNA prevented this up-regulation and reduced expression of all of these factors (All p < 0.001 except TFAM: p = 0.0018). However, under Sesn2 silencing conditions, after AICAR was administered, expression of these factors increased again following MCAO (NRF-1 and TFAM: p < 0.001; SOD2: p = 0.0151; UCP2: 0.0012).

Twenty-four hours after AICAR injection with Sesn2 silencing, neurological scores were significantly improved ( p < 0.001), cerebral infarct volume was obviously diminished ( p = 0.0032), SOD activity was enhanced ( p = 0.0030) and MDA content was decreased ( p = 0.0205) compared to rats only injected with Sesn2 siRNA ( Fig. 6 ).

In the present study, we demonstrated that expression of Sesn2 was elevated in the cortex after cerebral I/R. In contrast, Sesn2 knockdown effectively eliminated I/R-induced Sesn2 elevation in brain tissues. Sesn2 silencing exacerbated neurological deficits, increased cerebral infarction, and increased oxidative stress and neuron damage. To investigate the mechanism of Sesn2 in this experiment, AICAR, an AMPK activator, was administered after Sesn2 down-regulation. Sesn2 silencing was accompanied by decreases in p-AMPK/AMPK, mitochondrial biogenesis proteins, PGC-1α, NRF-1, and TFAM, as well as ROS-detoxifying proteins, UCP2 and SOD2, and increases in mitochondria-dependent apoptotic factors, cytochrome c and AIF. However, these effects were prevented by AICAR. These results suggest that Sesn2-induced AMPK-PGC-1α activation may be one underlying mechanism of the neuroprotective roles of Sesn2 against brain injury following cerebral I/R. Oxidative and nitrosative stress are powerful mediators of ischemic injury. After cerebral ischemia, the balance between ROS production and clearance are compromised, which may result in oxidative stress-induced signaling and cell injury 17 . Sesn2 is an inducible protein that protects cells against exaggerated ROS levels. One study reported that Sesn2 was induced in human glioblastoma U87 cells following ionizing radiation (IR). Sesn2 silencing not only increased oxidative stress but also sensitized U87 cells to IR 18 . Another study suggested that silencing Sesn2 in renal proximal tubule cells increased hyperoxidized peroxiredoxins and ROS production 3 . Recent studies showed that Sesn2 played a pivotal role in age-dependent neurodegenerative diseases. Sesn2 expression was elevated in primary rat cortical neurons upon Aβ exposure and in the cortices of a mouse model of Alzheimer’s disease 19 . Nevertheless, the potential involvement of Sesn2 in ischemia has not been well studied. In our experiment, Sesn2 expression was elevated approximately 2.6-fold after cerebral I/R. However, Sesn2 knockdown exacerbated neurologic deficits and increased cerebral infarction. Moreover, Sesn2 elimination decreased SOD activity and increased levels of MDA product after MCAO. These results demonstrated the neuroprotective role of Sesn2. The mechanisms Sesn2 exerts its protective effects against oxidative damage in cerebral ischemia remain unclear. It may promote the activity of other oxidoreductases, such as peroxiredoxin (Prdx), sulfiredoxin (Srx), that also regenerate Prdx 20 21 . However, another study showed that Sesn2 is not a reductase for cysteine sulfinic acid of peroxiredoxins 20 . Till now, the antioxidative mechanism of Sesn2 is not sure. It was accepted that ROS were generated principally by mitochondria during ischemia. Mitochondria are also the sources of energy and metabolism 6 . Several recent studies have shown that ROS could mediate mitochondrial biogenesis 22 . Another study has shown that excessive ROS may cause oxidative damage to mitochondria and elicit cell apoptosis. Cytochrome c and AIF are 2 key mitochondria-related apoptosis genes 23 . Therefore, maintaining the biogenesis of mitochondria and respiratory functions is pretty important 24 . Mitochondrial biogenesis is a highly regulated process and occurs on a regular basis in healthy cells, where it is controlled by the nuclear genome. It has been demonstrated that several protein factors encoded by nuclear genes are involved in the biogenesis of mitochondrial and respiratory functions, including NRF-1, TFAM, and PGC-1α 25 26 . It has been shown that PGC-1α may be the initiating factor of mitochondrial biogenesis 27 28 . Similar to Sesn2, PGC-1α is activated under oxidative stress as well. It has been reported that PGC-1α is expressed in the mouse cerebral subcortex under hypobaric hypoxia 29 30 . Silencing PGC-1α expression in mice exacerbates the neurodegenerative effects of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) and oxidative stressors in the substantia nigra and hippocampus. In contrast, increasing PGC-1α levels dramatically protect neural cells in culture from oxidative stressor-mediated death 10 31 . It was reported that PGC-1α is required for induction of many ROS-detoxifying proteins, including UCP2 and SOD2 32 . PGC-1α activity is enhanced by interacting with NRF-1 through the NRF-1 DNA binding domain. In neuronal cells, PGC-1α activation or overexpression promotes expression and activation of NRF-1, which regulates expression of downstream TFAM and other nuclear-e ncoded genes associated with oxidative phosphorylation to promote mitochondrial biogenesis 33 . It has also been reported that reperfusion increased mitochondrial biogenesis by increasing PGC-1α, NRF-1 and TFAM following focal cerebral ischemia 22 . In our experiment, after cerebral I/R, we observed an increase in expression of PGC-1α followed by increased expression of NRF-1, TFAM, SOD2 and UCP2. Several signaling kinases have been

How to cite this article : Li, L. et al. Sestrin2 Silencing Exacerbates Cerebral Ischemia/Reperfusion Injury by Decreasing Mitochondrial Biogenesis through the AMPK/PGC-1α Pathway in Rats. Sci. Rep. 6 , 30272; doi: 10.1038/srep30272 (2016).