The authors have reported that antioxidative effects play a crucial role in the volatile anesthetic-induced neuroprotection. Accumulated evidence shows that endogenous antioxidation could be up-regulated by nuclear factor-E2–related factor 2 through multiple pathways. However, whether nuclear factor-E2–related factor 2 activation is modulated by sevoflurane preconditioning and, if so, what is the signaling cascade underlying upstream of this activation are still unknown.
Sevoflurane preconditioning in mice was performed with sevoflurane (2.5%) 1 h per day for five consecutive days. Focal cerebral ischemia/reperfusion injury was induced by middle cerebral artery occlusion. Expression of nuclear factor-E2–related factor 2, kelch-like ECH-associated protein 1, manganese superoxide dismutase, thioredoxin-1, and nicotinamide adenine dinucleotide phosphate quinolone oxidoreductase-1 was detected (n = 6). The antioxidant activities and oxidative product expression were also examined. To determine the role of kelch-like ECH-associated protein 1 inhibition-dependent nuclear factor-E2–related factor 2 activation in sevoflurane preconditioning-induced neuroprotection, the kelch-like ECH–associated protein 1-nuclear factor-E2–related factor 2 signal was modulated by nuclear factor-E2–related factor 2 knockout, kelch-like ECH-associated protein 1 overexpression lentivirus, and kelch-like ECH-associated protein 1 deficiency small interfering RNA (n = 8). The infarct volume, neurologic scores, and cellular apoptosis were assessed.
Sevoflurane preconditioning elicited neuroprotection and increased nuclear factor-E2–related factor 2 nuclear translocation, which in turn up-regulated endogenous antioxidation and reduced oxidative injury. Sevoflurane preconditioning reduced kelch-like ECH-associated protein 1 expression. Nuclear factor-E2–related factor 2 ablation abolished neuroprotection and reversed sevoflurane preconditioning by mediating the up-regulation of antioxidants. Kelch-like ECH-associated protein 1 overexpression reversed nuclear factor-E2–related factor 2 up-regulation and abolished the neuroprotection induced by sevoflurane preconditioning. Kelch-like ECH-associated protein 1 small interfering RNA administration improved nuclear factor-E2–related factor 2 expression and the outcome of mice subjected to ischemia/reperfusion injury.
Kelch-like ECH-associated protein 1 down-regulation–dependent nuclear factor-E2–related factor 2 activation underlies the ability of sevoflurane preconditioning to activate the endogenous antioxidant response, which elicits its neuroprotection.
Ischemic stroke remains a major concern in the perioperative period and a leading cause of morbidity and mortality with few efficacious therapies
Further understanding of the brain endogenous neuroprotective pathways such as preconditioning induced by brief periods of ischemia or volatile anesthetics has potential to advance therapy
In a preclinical model of middle cerebral artery occlusion in mice, the authors show that sevoflurane preconditioning works in part through enhancing nuclear translocation of a transcription factor known as nuclear factor-E2–related factor 2, which is a regulator of the antioxidant responses of the body
This preclinical research suggests new elements to our understanding of volatile anesthetic-induced neuroprotection that may lead to novel therapeutic targets for ischemic stroke
ISCHEMIC stroke is one of the leading causes of morbidity and mortality and remains a vexing public health problem worldwide.1 Unfortunately, few therapies have been developed to protect the brain against ischemic injury since the efficacy of tissue plasminogen activator was discovered in 1996. Owing to its narrow therapeutic window (less than 4.5 h) and safety concerns, only less than 5% of stroke patients can be treated with tissue plasminogen activator. Therefore, more therapeutic measures need to be developed to improve stroke therapy.2
The brain has endogenous neuroprotective capacities.3 Learning how to mimic or engage the mechanism of endogenous neuroprotective response may reveal new neuroprotective avenues. In fact, despite activating damaging brain processes, ischemia also triggers a coordinated self-protective mechanism to counteract these deleterious events. These endogenous neuroprotective responses can also be induced by certain preconditioning approaches such as inhalation anesthetics, inflammatory stimuli, and pharmacologic compounds.4 Considered as an steerable and less invasive protective strategy, sevoflurane preconditioning (SPC) provides effective protection against cerebral ischemia/reperfusion (I/R) injury in both in vitro and in vivo models.5,6 However, the underlying mechanisms responsible for this protective effect remain poorly elucidated.
Oxidative stress plays a major role in the pathogenesis of ischemic stroke.7 However, accumulating studies support the dual role of reactive oxygen species (ROS) in the process of cerebral I/R. The protective antioxidant mechanisms against I/R injury are induced by brief ischemic episodes and rely on the generation of ROS.8 Preconditioning approaches, including inhalation anesthetics, also exert protective properties via similar signaling cascades.9,10 But the exact mechanism by which SPC mediates endogenous antioxidant activation still needs to be clarified.
Nuclear factor-E2–related factor 2 (Nrf2) has emerged as a major determinant of antioxidant-response genes and phase II detoxifying enzymes that is protective against cellular stress.11 Currently, more than 200 Nrf2 antioxidant-response element (ARE)-driven genes are known to be exploited for detoxification and antioxidant defense.12 The first investigated Nrf2 regulatory mechanism is the kelch-like ECH-associated protein 1 (Keap1)-dependent Nrf2 ubiquitination pathway. Under baseline conditions, Nrf2 binds to Keap1 homodimer, locates in the cytoplasm, and is degraded by 26s proteasome-mediated ubiquitination.13 During cellular stress or electrolytic imbalance, cysteine residue modification in Keap1 induces Nrf2 release from the Nrf2–Keap1 complex, which stabilizes Nrf2 and causes its nuclear translocation.14 Endogenous Nrf2-activated antioxidants play a central protective role in alleviating oxidative stress during cerebral I/R in vivo and in vitro.15–17 By using a transient middle cerebral artery occlusion (MCAO) model, we investigated the possibility that Keap1-dependent Nrf2 activation confers the ability of SPC to activate the endogenous antioxidant response after cerebral I/R, thereby mediating the neuroprotective effect.
Materials and Methods
All experimental procedures were approved by the Ethics Committee for Animal Experimentation of the Fourth Military Medical University (Xi’an, Shannxi, China). Male Nrf2 knockout mice (Nrf2−/−) between 8 and 10 weeks old and age-matched Institute of Cancer Research littermates (wild-type [WT] mice) were purchased from the Animal Laboratory of Nanjing Military Region General Hospital, Nanjing, China. The genotypes of mice were identified by polymerase chain reaction analysis with genomic DNA extracted from mouse-tail biopsies in this laboratory. The mice were housed in a 12-h alternating light and dark cycle at 20° to 25°C and 60 to 70% humidity with freely available water and food for at least 1 week before treatment or surgery. The sample size was based on a previous study.5 The formal statistical power analysis was not used to guide sample size in this study. Four hundred mice were used in this experiment, including 80 Nrf2−/− mice and 320 WT mice. The number of animals used and their suffering were minimized in this study. All mice were randomized to different experimental groups.
First, the neuroprotective effect induced by SPC was determined. Mice were randomly assigned to sham, I/R, or SPC (sevo plus I/R) groups. All animals except sham mice received MCAO surgery at 24 h after the last preconditioning period. The neurologic score was measured at 1, 3, and 7 days, and the survival rate and infarct volume were assessed at 7 days after reperfusion.
To measure the antioxidant effect of SPC, manganese superoxide dismutase (Mn-SOD), thioredoxin-1 (Trx1), and nicotinamide adenine dinucleotide phosphate quinolone oxidoreductase-1 (NQO1) protein levels were examined by Western blot at 2, 8, and 24 h after reperfusion. Enzyme-linked immunosorbent assay (ELISA) kits were used to examine the enzymatic activities of catalase, total superoxide dismutase (T-SOD), and total antioxidant capacity (T-AOC).
To demonstrate the effect of SPC on Nrf2 activation, Nrf2 protein expression and Nrf2’s nuclear translocation level were examined by Western blot at 2, 8, and 24 h after reperfusion. Nrf2’s cellular expression, localization, and nuclear translocation were also examined by immunofluorescence staining.
To further determine Nrf2’s role in SPC-induced neuroprotection, Nrf2−/− mice and their WT littermates were used. Animals were divided into six groups: two groups with sham operation (WT-sham and Nrf2−/−-sham), two with oxygen treatment (WT-I/R and Nrf2−/−-I/R), and two with SPC (WT-sevo plus I/R and Nrf2−/−-sevo plus I/R). The neurologic outcome and infarct volume were evaluated at 72 h after reperfusion. An additional cohort of mice was used to evaluate apoptotic cell death by terminal deoxynucleotidyl transferase deoxyuridine triphosphate-biotin nick-end labeling (TUNEL) staining at 72 h after reperfusion. The abundance of Mn-SOD, Trx1, and NQO1 proteins in the Nrf2−/− groups was compared with that in the WT groups at 8 h after reperfusion. Dihydroethidium oxidation staining, ROS level, and oxidized macromolecule products (4-hydroxynonenal [4-HNE]; malondialdehyde; carbonyl protein; and 8-hydroxy-2-deoxyguanosine [8-OhdG]) were also assessed and compared at 24 h after reperfusion.
To demonstrate the upstream activation of Nrf2 induced by SPC, Keap1 protein expression was assessed by Western blot at 2, 8, and 24 h after reperfusion. Keap1 cellular expression was also examined by immunofluorescence staining. Coimmunoprecipitation experiment was used to examine the Keap–Nrf2 complex. Additionally, a Keap1-overexpressing lentivirus (LV-Keap1) and its scrambled control lentivirus (LVc) were used. After testing the efficiency of LV-Keap1 at 72 h after administration, 48 additional mice were randomly assigned to sham, I/R, sevo plus I/R, sevo plus I/R plus vehicle, sevo plus I/R plus Keap1-Lv, and sevo plus I/R plus Keap1-Lvc groups. Neurologic outcome and infarct volume were examined at 72 h after reperfusion. To further demonstrate Keap1’s role in Nrf2 activation induced by SPC, the small interfering RNA (siRNA) targeting Keap1 (Keap1-siRNA) and its scrambled control siRNA (siRNAc) were used. The interference efficiency was checked at 72 h after administration. Then, 48 additional mice were randomly assigned to sham, I/R, sevo, I/R plus vehicle, I/R plus siRNA, and I/R plus siRNAc groups. Neurologic outcome and infarct volume were examined at 72 h after reperfusion.
The mice in the sevo plus I/R group were treated with 97% oxygen containing 2.5% volume sevoflurane 1 h per day for five consecutive days according to a previous report.5 The I/R mice were treated for the same duration but were exposed only to 97% oxygen contained with 3% nitrogen.
Model of Transient Focal Cerebral I/R
The focal cerebral I/R injury was induced by MCAO at 24 h after the last session of sevoflurane or oxygen preconditioning as described previously.18 Mice were anesthetized with 2.5% sevoflurane via nose cone. A 6-0 nylon monofilament with a heated, rounded tip was inserted through the right internal carotid artery to occlude the right middle cerebral artery. At 60 min after occlusion, the filament was drawn out to let the blood flow recover. The animals’ temperature during the procedures was maintained at 37.0° ± 0.5°C by using a thermostatically controlled heating blanket connected to a thermometer probe in the rectum. The regional cerebral blood flow (rCBF) was quantitated with a laser Doppler flowmeter (PeriFlux 5000; Perimed AB, Sweden) placed on the right side of the skull’s dorsal surface (2 mm caudal and 5 mm lateral to bregma) before, during, and after the operation. Mice were excluded if their rCBF could not fall below 20% of baseline during occlusion or recover over 80% during reperfusion.
Arterial Blood Gas Determination
The physiologic variables were measured in separate I/R and SPC groups (n = 5 in each group). The left femoral arterial catheter was placed, and about 0.2 ml blood was taken from the femoral artery of each mouse from respective groups at the end of the last exposure, the onset of MCAO, and reperfusion; the volume of warm saline was injected simultaneously into the caudal vein. After the last sample was taken, the mice were euthanized, and the blood gas measurement was confirmed immediately by using the OMNI Modular System (Rapidlab 1260; Bayer HealthCare, United Kingdom).
Neurobehavioral Assessment and Infarct Evaluation
The survival rate was observed and presented as a percentage. A Garcia system including six tests was used to score the neurologic deficit by an observer blinded to experiment paradigm.19 For infarct volume assessment, mice were euthanized, and the brains were removed and sliced into 1-mm-thick coronal sections after the evaluation of neurologic score. Sections were stained with 1% 2,3,5-triphenyltetrazolium chloride (Sigma-Aldrich, USA). The infarct volume percentage was calculated by the following equation to correct the edema according to the Swanson method: 100 × (contralateral hemisphere volume − nonlesioned ipsilateral hemisphere volume)/contralateral hemisphere volume.20
Coimmunoprecipitation and Western Blot
Deeply anesthetized mice were decapitated, and the penumbra tissue was harvested at each time point according to a previous report.21 For the measurement of Nrf2 nuclear translocation level (at 2 and 8 h after reperfusion), nuclear-cytoplasm protein extraction was performed using an NE-PER extraction kit (Pierce Biotechnology, USA) according to the manufacturer’s instructions. Protein concentration was quantified using a bicinchoninic protein assay kit (Beyotime, China). For coimmunoprecipitation, 15 μl antibody (anti-Keap1 or anti-Nrf2; Abcam, United Kingdom) was added to 300 μg protein. The mixture was subsequently incubated with glutathione sepharose 4B beads (GE Healthcare, USA), and supernatant proteins were separated on gel, transferred to nitrocellulose membranes, and incubated with primary antibodies (anti-Nrf2 or anti-Keap1). For Western blot, equal amounts of protein samples (40 μg) were loaded and probed with antibodies as follows: Nrf2 (1:500; Abcam), Keap1 (1:1,000; Abcam), Mn-SOD (1:500; Millipore, USA), Trx1 (1:200; Abcam), NQO1 (1:500, Epitomics, USA), β-tubulin (1:10,000, CWBIO, China), glyceraldehyde-3-phosphate dehydrogenase (1:2,000; Abcam), or histone (1:500; Signal Way Antibody, USA). After incubating the membranes with appropriate secondary antibodies, the immunoblots were immersed in enhanced chemiluminescent reagent, followed by exposure to enhanced chemiluminescent-Hyperfilm to visualize specific protein bands (Amersham Biosciences, Sweden).
Penumbra samples were homogenized in ice-cold saline with a 1:10 weight-to-volume ratio and then centrifuged. The supernatant was collected for ELISA analysis. Catalase, T-SOD, T-AOC, malondialdehyde (all from Beyotime), 4-HNE, carbonyl protein, 8-OHdG (all from Cell Signaling, USA), and ROS (West-tone, China) assay kits were used to assess the production of antioxidant enzymes, the oxidative productions, and ROS according to their instructions, respectively.
Cell apoptosis was evaluated by in situ TUNEL assay kit (in situ Cell Death Detection Kit; Roche Diagnostics, Germany) at 72 h after reperfusion. Slides were stained according to the manufacturer’s instruction, and the nuclei were counterstained by hematoxylin. Positive cells were observed using a ×40 objective lens from three random areas in the ischemic penumbra, and the number of positive cells was calculated and expressed as number per 100 μm2.
Alternating sections were cut at 10-μm thickness in the coronal plane. The sections were incubated with a combination of primary mouse anti-neuron nuclei (1:500; Millipore) or anti-glial fibrillary acidic protein (1:500; Sigma-Aldrich) and a primary rabbit anti-Nrf2 (1:100; Abcam) or anti-Keap1 (1:100; Abcam) at 4°C overnight and visualized with mixed secondary antibodies including donkey anti-mouse conjugated to green-fluorescent Alexa Fluor 488 and donkey anti-rabbit conjugated to red-fluorescent Alexa Fluor 594 (1:500 for both; Vector Laboratories, USA) for 2 h at room temperature. 4',6-diamidino-2-phenylindole (1 ng/μl; Sigma-Aldrich, United Kingdom) was used to stain the nuclei. Fluorescent signals were detected by using a confocal laser scanning microscope (Olympus, Japan).
At 24 h after reperfusion, brains were cut into 20-μm-thick coronal sections. The sections were incubated with dihydroethidium working solution (dihydroethidium, 10 mM; Beyotime) at 37°C for 30 min and covered with square glass coverslips. Dihydroethidium oxidation images were captured using a confocal fluorescence microscope and expressed as positive numbers. The calculation of dihydroethidium-positive cells was done according to the report by Dugan et al.22
Construction and Transfection of Keap1-siRNA or LV-Keap1
LV-Keap1 (Ubi-MCS-3FLAG-SV40-EGFP) was constructed by Gene Chem Company (Shanghai, China). Keap1 siRNA and its siRNAc were designed and purchased from Qiagen (China). The target sequences of the Keap1-siRNA were as follows:
Target sequence: 5′-AAGGCTTATTGAGTTCGCCTA-3′;
Sense strand: 5′-GGCUUAUUGAGUUCGCCUATT-3′;
Antisense strand: 5′-UAGGCGAACUCAAUAAGCCTT-3′.
Titers of 4 × 109 U/ml (for siRNA) or 2 × 109 U/ml (for lentivirus) were routinely achieved. The transfection was confirmed by intracerebroventricular injection. The stereotaxic coordinate location of the lateral cerebral ventricle was 0.4 mm posterior to bregma, 1.0 mm lateral to the midsagittal line, and 2.0 mm deep from the cranial surface. After 72 h of injection, the reliability of siRNA and lentivirus on the Keap1 expression was examined by Western blot.
Data were analyzed with SPSS 18.0 for Windows (IBM, USA). The data for survival rates were expressed as percentages and compared using Fisher exact probability test. The neurologic score was presented as the median (interquartile range) and was analyzed using nonparametric statistics (Kruskal–Wallis test) with Bonferroni correction. Other values are presented as mean ± SD. The temporal protein changes were analyzed by two-way repeated measurement ANOVA with Bonferroni post hoc test for between-group differences. The other values were analyzed by a one-way ANOVA with Tukey post hoc test for between-group differences. Two-tailed values of P < 0.05 were considered statistically significant.
SPC Alleviates Brain Injury Induced by I/R
All animals in this study showed similar values for physiologic variables (table 1, Supplemental Digital Content 1, http://links.lww.com/ALN/B351, which shows the blood gas analysis result). The laser Doppler flowmetry results showed that rCBF was reduced equivalently in all MCAO groups during occlusion and recovered more than 80% during reperfusion. Sevoflurane exposure did not alter the rCBF change during the MCAO procedure (fig. 1, Supplemental Digital Content 2, http://links.lww.com/ALN/B352, which shows the changes of rCBF in mice before, during, and after MCAO operation).
The survival rate in the SPC group on day 7 was significantly higher than that in the I/R group (fig. 2A, Supplemental Digital Content 3, http://links.lww.com/ALN/B353, which shows that SPC induced cerebral ischemic tolerance). Seven days after reperfusion, only nine mice in the I/R group (50%) and 14 mice in the sevo plus I/R group (74%) survived. The neurologic scores measured at 1 (9.0 [7.0 to 11.0] vs. 12.0 [10.0 to 13.5]; P < 0.001), 3 (7.5 [5.0 to 10.5] vs. 10.5 [9.5 to 12.5]; P = 0.013), and 7 (9.5 [7.5 to 12.0] vs. 12.5 [10.5 to 13.5]; P = 0.044) days after reperfusion were higher in the SPC group than those in the I/R group (fig. 2B, Supplemental Digital Content 3, http://links.lww.com/ALN/B353, which shows that SPC induced cerebral ischemic tolerance). The infarct volume in the sevoflurane-preconditioned mice was reduced as compared with that in the I/R animals (20.5 ± 5.4% vs. 14.9 ± 2.4%; P = 0.036; fig. 2, C and D, Supplemental Digital Content 3, http://links.lww.com/ALN/B353, which shows that SPC induced cerebral ischemic tolerance).
SPC Up-regulated Antioxidant Enzyme Expression and Activity
As shown in figure 1A, Mn-SOD protein level was increased in the sevoflurane-pretreated mice at 2 and 8 h compared to that in the I/R animals (1.1 ± 0.3 vs. 2.7 ± 0.9, P = 0.002; 1.5 ± 0.4 vs. 2.4 ± 0.4, P = 0.010, respectively). Trx1 and NQO1 expression was increased at 8 h in the SPC group compared with that in the I/R group (fig. 1, B and C; 1.0 ± 0.4 vs. 1.9 ± 0.5, P = 0.003; 0.9 ± 0.3 vs. 1.3 ± 0.2, P = 0.049, respectively). The activities of some antioxidant enzymes (catalase, T-AOC, and T-SOD) were also enhanced by SPC at 24 h after reperfusion (fig. 1, D to F; 0.006 ± 0.003 vs. 0.013 ± 0.005, P = 0.027; 0.24 ± 0.06 vs. 0.59 ± 0.16, P = 0.003; 214.3 ± 99.6 vs. 537.6 ± 106.8, P = 0.037, respectively).
SPC Increased Nrf2 Protein Expression and Nuclear Translocation
The sevoflurane single or 5-day exposure alone did not alter the Nrf2 expression compared with that in the control group (1.1 ± 0.2 vs. 1.0 ± 0.2, P =0.765; 1.1 ± 0.2 vs. 1.0 ± 0.1, P =0.599, respectively; fig. 3, Supplemental Digital Content 4, http://links.lww.com/ALN/B354, which shows that SPC alone did not affect the expression of Nrf2). As shown in figure 2A, SPC increased Nrf2 content compared with the I/R group at 2 and 8 h (1.6 ± 0.3 vs. 2.2 ± 0.3, P = 0.045; 1.6 ± 0.3 vs. 2.1 ± 0.3, P = 0.036, respectively) but not at 24 h after reperfusion. The immunofluorescent staining results showed that the number of Nrf2-positive neurons in the penumbra of sevoflurane-preconditioned mice was notably increased compared to that of the MCAO animals at 8 h after reperfusion (fig. 2B).
Nrf2 expression in the cytoplasm was increased in the SPC group as compared with that in the I/R group at 8 h after reperfusion (fig. 2C; 1.2 ± 0.5 vs. 2.2 ± 0.5; P = 0.027) but not at 2 h after reperfusion (1.2 ± 0.4 vs. 1.2 ± 0.5; P = 0.115). A marked increment of nucleic Nrf2 expression was found in the sevoflurane-preconditioned mice compared to the I/R mice at 2 and 8 h after reperfusion (fig. 2D; 1.4 ± 0.4 vs. 1.9 ± 0.6, P = 0.042; 1.2 ± 0.3 vs. 2.7 ± 0.4, P = 0.037, respectively). The Nrf2 nuclear translocation ratio in the SPC group was higher than that in the I/R group at 2 and 8 h after reperfusion (fig. 2E; P = 0.044 and P = 0.041, respectively).
The immunofluorescent staining results showed a weak nuclear immunoreactive staining of Nrf2 in ischemic penumbra of the I/R group. In the SPC group, Nrf2-positive expression was stronger in the nucleus, and the Nrf2 protein was mostly colocalized with nuclear marker 4',6-diamidino-2-phenylindole (blue) in neurons (fig. 2F) and astrocytes (fig. 2G). This indicated that the increased Nrf2 nuclear translocation in neurons and astrocytes was a response to SPC.
Nrf2 Ablation Exaggerated I/R Injury and Abolished the Cerebral Ischemic Tolerance Induced by SPC
When subjected to MCAO, Nrf2−/− mice had significantly larger infarct volumes (53.4 ± 12.9% vs. 27.7 ± 10.0%; P = 0.002) and poorer neurologic manifestations (8.5 [6.0 to 11.0] vs. 12.25 [11.0 to 14.0]; P = 0.018) than the WT mice (fig. 3, A to C). Moreover, SPC had no significant neuroprotective effects in Nrf2−/− mice either in terms of neurologic deficits or infarct volumes (6.0 [4.0 to 9.0] vs. 7.25 [5.0 to 8.5]; P = 0.204 and P = 0.310, respectively).
Cell apoptosis results are shown in fig. 3, D and E. The number of TUNEL-positive cells in the WT-sevo plus I/R group was reduced compared with that in the WT-I/R group at 72 h after reperfusion (29.4 ± 11.5 vs. 17.2 ± 8.2; P < 0.001). When pretreated with sevoflurane, Nrf2−/− mice showed more TUNEL-positive cells than WT mice (P = 0.031). No statistical difference was found between the Nrf2−/−-sevo plus I/R group and the Nrf2−/−-I/R group (47.2 ± 2.2 vs. 40.0 ± 2.8; P = 0.227).
As shown in figure 4, A to C, in the WT mice, the expressions of Mn-SOD, Trx1, and NQO1 were increased by SPC, compared to the I/R group. However, with SPC, proteins’ expression in the WT-sevo plus I/R group was increased compared with that in the Nrf2−/−-sevo plus I/R group (P = 0.004, 0.015, and 0.028, respectively). No statistical difference was found between Nrf2−/−-I/R and Nrf2−/−-sevo plus I/R groups (1.5 ± 0.6 vs. 1.5 ± 0.5, P = 0.201; 1.8 ± 0.2 vs. 1.5 ± 0.4, P = 0.284; 0.7 ± 0.2 vs. 0.8 ± 0.2, P =0.354, respectively).
As shown in figure 4, D to H, no significant difference of the oxidative products’ (the lipid oxidative products 4-HNE and malondialdehyde, protein carbonyl, 8-OHdG, and ROS) levels was detected between the Nrf2−/−-sevo plus I/R group and the Nrf2−/−-I/R group (4.8 ± 1.0 vs. 4.8 ± 1.4, P = 0.166; 9.4 ± 1.4 vs. 8.7 ± 2.3, P = 0.263; 2.7 ± 0.6 vs. 2.6 ± 0.5, P = 0.851; 19.2 ± 4.7 vs. 18.9 ± 4.9, P = 0.363; 466.2 ± 93.6 vs. 423.2 ± 81.6, P = 0.240, respectively). As the indicator of superoxidant generation, dihydroethidium staining results also showed more dihydroethidium-positive cells in the Nrf2−/−-sevo plus I/R group than in the WT-sevo plus I/R group (P = 0.032) and no significant increment of dihydroethidium-positive cell numbers in the Nrf2−/−-I/R mice compared to the Nrf2−/− mice exposed to SPC (fig. 5, A to C; P = 0.135).
SPC-induced Nrf2 Activation Was Partially Keap1 Dependent
As shown in figure 6A, SPC reduced Keap1 abundance only at 8 h after reperfusion compared to the I/R group (1.3 ± 0.4 vs. 1.0 ± 0.3; P = 0.049). Keap1-positive neurons in the SPC group were notably reduced in the ischemic penumbral area compared to those in the I/R group at 8 h (fig. 6B). To demonstrate whether SPC dissociates the Keap1-Nrf2 complex, the coimmunoprecipitation experiment was performed and is shown in figure 6C. First, cytosolic fractions were immunoprecipitated with anti-Keap1 antibody and immunoblotted with anti-Nrf2 antibody. The expression of protein corresponding to Nrf2 bound to Keap1 was more decreased in the sevo plus I/R group than in the I/R group. Next, the lysates were precipitated with anti-Nrf2 antibody and immunoblotted with anti-Keap1 antibody. The expression of protein corresponding to Keap1 that had bound to Nrf2 was significantly lower in the sevo plus I/R group than in the I/R group. These results support the concept that SPC promotes dissociation of Nrf2 from Keap1, the first step in Nrf2 nuclear translocation.
LV-Keap1 microinjection led to an approximately 2.1-fold increase of Keap1 expression in target brain regions (fig. 5A, Supplemental Digital Content 5, http://links.lww.com/ALN/B355, which shows Keap1 expression after lentivirus injection). Overexpression of Keap1 reversed the increased Nrf2 expression induced by SPC (fig. 6D; P = 0.030). The beneficial effect on neurologic outcome of SPC was blocked by LV-Keap1 administration (fig. 6E; 12.25 [7.5 to 13.5] vs. 8.0 [5.0 to 11.0]; P = 0.012). Additionally, LV-Keap1 supplement markedly reversed the reduction of infarct volume induced by SPC (fig. 6F; 27.9 ± 8.0% vs. 42.5 ± 6.6%; P = 0.019). The supplementation of vehicle or control lentivirus had no effect on the neuroprotective effect induced by SPC.
Whether Keap1 down-regulation affected the I/R injury was clarified by using Keap1-siRNA. Western blot results showed that microinjecting Keap1-siRNA suppressed about 40% of Keap1 expression at 72 h after administration (fig. 5B, Supplemental Digital Content 5, http://links.lww.com/ALN/B355, which shows Keap1 expression after siRNA injection). Keap1-siRNA improved Nrf2 abundance, as compared with the I/R plus vehicle group (fig. 6G; P = 0.036). The down-regulation of Keap1 improved the neurologic behavioral outcome (fig. 6H; 8.5 [6.0 to 11.0] vs. 11.75 [8.5 to 13.5]; P = 0.019) and reduced the infarct volume (fig. 6I; 29.2 ± 9.7% vs. 43.7 ± 11.0%; P = 0.017) after MCAO. Additionally, no statistical difference was found between the Keap1-siRNA group and the sevo plus I/R group nor between the vehicle group and the scrambled control siRNAc group in neurologic outcome and infarct volume.
In this study, we found that SPC reduced Keap1 expression, dissociated the Keap1–Nrf2 complex, and enhanced Nrf2 protein expression and nuclear translocation. This activation of Nrf2 resulted in the increased expression of endogenous antioxidants, improved neurologic outcome, and reduced cerebral infarct volume. Additionally, the ablation of Nrf2 blunted the antioxidant activation and the neuroprotective effect induced by SPC. Moreover, the neuroprotective effect and Nrf2 activation induced by SPC were reversed by the overexpression of Keap1. Down-regulation of Keap1 increased Nrf2 expression, improved neurologic outcome, and reduced the infarct volume. Thus, we conclude that SPC enhances endogenous antioxidation and induces cerebral ischemic tolerance against experimental ischemic stroke injury via Keap1 down-regulation–dependent Nrf2 activation.
Considerable evidence implicates the disruption of ROS generation and subsequent related cellular damage as a crucial cause of neuronal injury after the I/R process. The ability of the central nervous system to maintain homeostasis during periods of oxidative stress resides in the capacity to induce or activate endogenous protective enzymes. However, during I/R, the activity of numerous endogenous antioxidant enzyme systems is compromised or even disrupted, indicating that engaging the activity of cellular antioxidant enzymes may protect cells against I/R-induced oxidative stress injury. As a widely used volatile anesthetic agent, sevoflurane triggers the antioxidative pathway by increasing the activities of SOD, catalase, and glutathione peroxidase against I/R-induced damage processes when animals are preconditioned with it before ischemic injury. A major finding of this study indicates the notion that sevoflurane activates antioxidant capacity systemically after I/R. These endogenous antioxidants also exist in living brains, play a critical protective role in alleviating the damaging process during I/R, and are detected as the crucial constitution of antioxidant defense against reperfusion-induced oxidative stress.23–25 Whether the increased antioxidants resulting from SPC attenuated oxidative stress was also addressed in the current study. Oxidants have a very short half-life, which makes them very difficult to be measured directly, especially in vivo. Alternatively, O2·− production could be measured by detecting hydroethidium oxidation, including dihydroethidium staining, which was used in this study.26 We investigated a diffuse fluorescent dihydroethidium signal in the ischemic penumbral area and found it could be suppressed by SPC. This result indicates that the oxidative stress cascade induced by reperfusion could be attenuated by SPC. Oxidative stress could also be assessed by evaluating oxidized macromolecules.8 The levels of 4-HNE, malondialdehyde, protein carbonyl, 8-OHdG, and ROS were measured in this study as the representative oxidative products. SPC effectively attenuated I/R-increased oxidized products’ expression, which is in line with the increased antioxidant activities mediated by SPC. These findings suggest that SPC effectively attenuated I/R-induced oxidative stress via up-regulation of antioxidant expression.
SPC protects brain from oxidative stress, but there is some debate regarding the nature of how SPC reduces oxidative stress. Another major investigation of this study is that SPC engages the antioxidant enzymes against oxidative stress through a signaling mechanism, as evidenced by the capacity of SPC to up-regulate cerebral cellular antioxidants in an Nrf2-dependent manner. As a member of the nuclear factor-E2 family of nuclear basic leucine zipper transcription manner, Nrf2 tightly regulates numbers of endogenous antioxidants, which execute the defense against ischemic oxidative injury. This regulation is only mediated by nuclear accumulated Nrf2 binding to the AREs, a cisacting regulatory element or enhancer sequence found in the promoter region of certain genes including Mn-SOD, Trx1, and NOQ1.27 Nrf2 mutant mice displayed an increased cerebral vulnerability to I/R injury, indicating that Nrf2 is an important endogenous protective signal against cerebral ischemic injury. Some preconditioning approaches, such as hydrogen sulfide, carbon dioxide, and hydrogen peroxide, have been demonstrated to induce Nrf2 activation in vivo or in vitro.28–30 In the current study, we proved that the sevoflurane single or 5-day exposure alone did not alter the Nrf2 expression. But sevoflurane pretreated before ischemic injury enhanced Nrf2 expression and nuclear translocation after reperfusion, which subsequently increased the protein expression of Mn-SOD, Trx1, and NQO1. Some other mechanisms may also be involved in the regulation of antioxidant expression, such as signal transducer and activator of transcription 3, nuclear factor-κB, and tumor necrosis factor-α.31,32 In an I/R model, Calvert et al.29 demonstrated hydrogen sulfide modestly increased the expression of Trx1 but failed to increase the expression of HO-1 in the hearts of Nrf2 mutant mice. To further consolidate the determinant role of Nrf2 in SPC-induced antioxidation, Nrf2 knockout mice were used in the current study. Ablation of Nrf2 inhibited the up-regulation of antioxidant enzyme contents and attenuated the expression of oxidative products induced by SPC. Additionally, lack of Nrf2 abolished the neuroprotective effect of SPC. These indicate that although Nrf2 activation may not be the only regulatory pathway, it is a crucial regulative signaling mechanism contributing to SPC-mediated cellular antioxidant activation.
One of the main mechanisms of Nrf2 activation is the inhibition of Keap1-dependent ubiquitination.11 This regulatory mechanism has been determined in considerable detail and provides an understanding of the response to cellular stress and oxidant disruption.12 This study noted that SPC reduced Keap1 expression and dissociated the Keap1–Nrf2 complex, as evidenced by Western blot and coimmunoprecipitation results, suggesting SPC activated Nrf2 through Keap1 inhibition-dependent mechanism. To further elucidate the exact role of Keap1 in SPC-mediated Nrf2 activation, the Keap1 overexpression lentivirus was used in this study. The improved neurologic outcome and the attenuated infarct volume induced by sevoflurane were absent in Keap1 overexpression lentivirus-treated mice. Additionally, Keap1 siRNA was introduced to down-regulate Keap1 expression in mice. The administration of Keap1-siRNA induced up-regulation of Nrf2 activation and mimicked the neuroprotective effect of SPC. These results indicate that the neuroprotection and Nrf2 activation induced by SPC were achieved through Keap1 down-regulation. In fact, Keap1 acts as a hub for multiple second messengers. During cellular stress, Keap1 can perceive some endogenous signaling molecules such as ROS, nitric oxide, zinc, and 4-hydroxy hexenal, and subsequently promote the dissociation of the Keap1-Nrf2 complex and increase the nuclear translocation of Nrf2.33 The generation of ROS is an important mechanism underlying the beneficial effects of anesthetic pretreatment before ischemia. We postulate that Keap1 can recognize the ROS and its peroxidation molecules and electrophiles induced by SPC before ischemia. After it perceives these stress signals, Keap1 reduces its capability to ubiquitinate and turn over Nrf2. It allows this cap 'n' collar basic leucine zipper factor to accumulate in the nucleus, initiate the transcription of ARE-regulated genes, and defend against ischemic injury. However, the precise mechanism of Keap1-related changes is still unclear.
Some limitations need to be noted in this study. In the current study, we found that SPC enhanced Nrf2 activity at 2 and 8 h after reperfusion, but the decrease of Keap1 expression was detected only at 8 h after reperfusion. The inconsistent changes of Nrf2 and Keap1 indicated that some other mechanisms may contribute to SPC-induced Nrf2 activation. Other research studies have documented that Nrf2 activation can be induced by Nrf2 phosphorylation. Multiple intracellular signaling systems, including phosphoinositide 3-kinase, extracellular signal-regulated kinase 1/2, glucose syntheses kinase-3β, and mitogen-activated protein kinase, have also been demonstrated to modulate Nrf2 activity.34,35 The above intracellular signaling systems also mediate the cytoprotective effect of sevoflurane.36–38 Whether another Keap1 inhibition-independent mechanism is involved in SPC-induced Nrf2 activation and downstream antioxidative effects still needs further exploration.
In summary, we demonstrated the down-regulation of Keap1-dependent Nrf2 activation confers the ability of SPC to activate endogenous antioxidant responses after cerebral I/R, which mediates its neuroprotective effects (Fig. 7). The clarification of this novel neuroprotective target is helpful for translating volatile anesthetic pretreatment from bench to bedside against acute or perioperative ischemic stroke.
We thank Dr. Bairen Wang, M.D., Ph.D. (Department of Anesthesiology and Perioperative Medicine, Xijing Hospital, Fourth Military Medical University, Xi’an, Shaanxi, China), for his critical revision of the manuscript.
Supported by grants from the National Natural Science Foundation of China, Beijing, China (no. 81128005 and 81371510 to Dr. Dong), and Shaanxi Province Innovation Team, Xi’an, Shaanxi, China (no. 2013-kct 30 to Dr. Dong).
The authors declare no competing interests.