Isoflurane Enhances Autophagy by Activating AMPK/ULK1, Inhibits NLRP3, and Reduces Cognitive Impairment After Cerebral Ischemia–Reperfusion Injury in Rats

Cerebral ischemic stroke (CIS) has become the second leading cause of death worldwide, which is largely related to cerebral ischemia reperfusion injury (CIRI). Surgical intervention is a reliable treatment for CIS, which predictably causes cerebral reperfusion. Therefore, the choice of anesthetic drugs has important clinical significance. Isoflurane (ISO), one of the most used anesthetics, attenuates cognitive impairment and has brain protective effects. However, the role of isoflurane in regulating autophagy and its regulatory mechanism on inflammation in CIRI are still unclear. The middle cerebral artery occlusion (MCAO) method was used to establish a rat model of CIRI. After 24 h of reperfusion, all rats were evaluated by mNSS scoring and dark avoidance experiment. Western blotting and immunofluorescence were used to examine the expression of key proteins. Compared with the sham group, the MCAO group showed increased neurobehavioral scores and decreased cognitive memory function (P < 0.05). As for the ISO-treated MCAO rats, the neurobehavioral score was significantly decreased, the expression of AMPK, ULK1, Beclin1, and LC3B was significantly increased, and the cognitive and memory functions were also significantly improved (P < 0.05). After inhibition of autophagy pathway or key protein AMPK in autophagy, neurobehavioral scores and protein expression of NLRP3, IL-1β, and IL-18 were significantly increased (P < 0.05). Isoflurane post-treatment may enhance autophagy by activating the AMPK/ULK1 signaling pathway and further inhibit the release of inflammatory factors from NLRP3 inflammasomes, thereby ameliorating neurological function and cognitive impairment and exerting a protective effect on the brain in CIRI rats.


Introduction
Cerebral ischemic stroke (CIS) has the characteristics of high morbidity, high disability and high mortality and has become the second leading cause of death in the world . Despite the basic research and clinical diagnosis and treatment of CIS has been carried out for many years, CIS is still one of the most important public health problems (Lyden et al. 2021). Timely opening of blood vessels and restoration of blood perfusion are the main treatment Jingwen Zhai and Nian Li have contributed equally to this work. principles of CIS . However, when blood flow is restored to ischemic brain tissue, the ischemic damage is further aggravated. This phenomenon is known as cerebral ischemia-reperfusion injury (CIRI), which often leads to cellular inflammation, necrosis, and apoptosis (Nasoohi et al. 2019), ultimately resulting in neurological deficits in cognition, memory and learning. During CIRI, stress can activate autophagy, and neurons are particularly sensitive to changes in autophagy activity. Previous studies have shown that the activation and upregulation of autophagy reverses neurological deficits  and plays an important role in neuroprotection (Wen et al. 2019). Autophagy is an important catabolic program that processes various macromolecular cellular contents to maintain homeostasis, while AMP-activated protein kinase (AMPK) is a key metabolic regulator that simultaneously inhibits anabolic pathways and activates catabolic pathways to achieve energy homeostasis. Recruitment of AMPK to lysosomes may be critical for the integration of nutrient signaling and autophagy. Numerous of studies have shown that autophagy is the downstream target of AMPK, and AMPKmediated activation of autophagy may be a potential protective mechanism against brain injury (Inoki et al. 2012;Jiang et al. 2014Jiang et al. , 2015. In addition, unc-51-like kinase 1 (ULK1) is considered to be an important initiator of autophagy (Lee et al. 2010). In mammals, the ULK1 complex can initiate the phagosome membrane structure and recruit subsequent ATG proteins to form autophagosomes. Therefore, when cellular energy is depleted, AMPK can directly activate the ULK1 complex to induce autophagy (Hurley et al. 2017). Meanwhile, the NLRP3 inflammasome also plays a critical role in brain injury induced by ischemia-reperfusion (Palomino-Antolin et al. 2021). Some studies have also reported that the NLRP3 inflammasome-mediated inflammatory response may be regulated by autophagy (Lei et al. 2021, Yu et al. 2021. Therefore, exploring the activation of autophagy after nerve injury and its influence on the further regulation of inflammation will guide us in the treatment of CIRI. Preliminary studies by our research group have confirmed that the inhaled anesthetic isoflurane (ISO), which is widely used in clinical anesthesia, plays an important protective role in central nervous system injury. It may be related to multiple signaling pathways and oxidative stress, and cell inflammation is closely related to apoptosis (Peng et al. 2019a, b, Yin, et al. 2020. Through our research group's previous studies, we have found that ISO can improve CIRI by reducing cell edema, promoting angiogenesis, and reducing cell inflammation, apoptosis, and pyrolysis, which has a protective effect on the brain Peng et al. 2019a, b;Yang et al. 2020;Yuan et al. 2018;Zhang et al. 2021). However, it is still unclear whether isoflurane can activate the AMPK/ULK1 pathway after CIRI and whether autophagy mediated by this pathway can exert a protective effect on the brain in CIRI. This study aims to investigate the role of autophagy and the AMPK/ ULK1 signaling pathway in isoflurane-induced neuroprotection and the relationship between autophagy and the NLRP3 inflammasome in neuroprotection (Fig. 1).

Ethics Statement
All experimental animal protocols complied with the regulations of the People's Republic of China on the Administration Fig. 1 Graphic abstract. When the body encounters external stimuli or stress such as CIRI, changes in AMP, calcium channels and ROS can cause AMPK to be activated by its upstream LKB1, TAK1, or CaMKK. Activated AMPK can affect autophagy by negatively regulating mTOR, or it can directly activate ULK1 to activate autophagy. Activation of autophagy may further inhibit the secretion and release of inflammatory factors by the NLRP3 inflammasome, thereby exerting a protective effect 1 3 of Laboratory Animals. The experimental procedures were approved by the Animal Research Committee of Shihezi University (Protocol A2019-025-01).

Animals and Grouping
Healthy and clean male Sprague-Dawley (SD) rats weighing 200-250 g, all provided by the Laboratory Animal Center of Shihezi University School of Medicine, were selected. The temperature of the breeding room is between 21 and 24 °C, the relative humidity is 40-70%, and the supply of water and food is sufficient. There were 4 rats in each cage, and the cages were cleaned once a week. A total of 87 rats were selected, after excluding the dead rats, the remaining 80 rats were randomly divided into 5 groups with 16 rats in each group: Sham operation group (Sham), middle cerebral artery occlusion model group (MCAO), isoflurane post-treatment group (M + ISO), isoflurane post-treatment rats treated with autophagy inhibitor Baf-A1 group (M + I + B), and isoflurane post-treatment rats treated with AMPK inhibitor compound C group (M + I + C). All experimental rats were treated with 3% pentobarbital sodium by intraperitoneal injection at 5 mg/100 g for base anesthesia.

Establishment of MCAO Model
The anesthetized rat was fixed in the supine position on a 37 °C constant temperature operating table, and the MCAO model was established according to the modified Longa method (Tian et al. 2022). After disinfecting of the neck skin, a median incision was made, and the right common carotid artery (CCA), external carotid artery (ECA), and internal carotid artery (ICA) were separated and exposed. A small oblique incision was made near the bifurcation of the internal and external carotid arteries of the CCA, and a 3-0 nylon threaded plug was inserted at a depth of 18.0 ± 1.0 mm from the bifurcation. In the Sham group, only the CCA, ECA, and ICA were divided on the right side, and the nylon thread bolt was not inserted. The other procedures were the same as those in the model group. After 90 min of embolization, the nylon screw was carefully removed to create a reperfusion model, and the reperfusion time was 24 h. The isoflurane post-treatment group was post-treated with 1.5% isoflurane for 60 min immediately after reperfusion. The inhibitor group was injected with Baf-A1 (10 µl/piece, 0.1 ug/piece, S1min413, Selleck) and compound C (10 µl/ piece, 20 ug/piece, S7306, Selleck) 30 min before the model was established, respectively. The modeled rats were placed on the thermal insulation and resuscitation blanket. Upon awakening, they were returned to the cage and maintained in solitary confinement, with free food and water.

Neurobehavioral Experiment
The modified Neurological Severity Scores (mNSS) standard was used to evaluate the neurological function of rats after 24 h of ischemia-reperfusion from four aspects: exercise test, sensory test, balance beam test, loss of reflex, and abnormal movement. The total score is 18 points, the higher the score, the more severe the damage to nerve function.
Analyze the learning, cognitive, and memory functions of rats by avoiding the dark experiment. Adapt: The rat was placed in the light room, the door hole between the light and dark rooms was opened, the rat was allowed to move freely for 3-5 min, and then the rat was placed back in the cage. Train: The next day, the rat was placed in the light room. Since rats have the characteristics of darkening and evil light, they will quickly enter the dark room. After the rat entered the dark room, close the door and start electrical stimulation at the same time. The incubation period of the rat that entered the dark room was recorded from the light room, which can be used as an indicator to judge whether there was a difference in the behavior of each rat. Test: 24 h after training, the rats were returned to the light chamber. The time from the time the rat was placed in the light room to the first time it voluntarily entered the dark room was recorded as the step-through latency (STL), and the maximum latency was set at 300 s. And the number of times the rat entered the dark room within 300 s was recorded, which was called error times (ET). If the rat did not enter the dark room within 300 s, the test was terminated and the next test was continued. Normal rats will not voluntarily enter the dark room due to the existence of electrical stimulation memory, while rats with neurological impairment have reduced learning and memory ability and will enter the dark room in a short period of time and repeatedly enter and exit the dark room. Therefore, when the cognitive learning and memory function is impaired, the STL is significantly smaller than that of the control group, and the ET is significantly larger than that of the control group.

Laser Speckle Imaging (LSI)
Changes in cerebral blood flow (CBF) in rats were observed by using a laser speckle imaging system. The anesthetized rat was placed in the prone position, the scalp was cut along the middle, the surface of the skull was separated and exposed, and the laser speckle imaging system (Moore FLPI-2, Beijing, China), using a laser diode (785 nm; Moore FLPI-2, Beijing, China) obtained images at 23fps (exposure time t = 5 ms). Changes in CBF in rats were observed, sutured, and resuscitated after completion.

5-Triphenyltetrazolium Chloride (TTC) Staining
TTC staining measures the volume of cerebral infarction. The anesthetized rats were decapitated and killed. The brains were quickly removed and cut into five brain slices approximately 2 mm thick rostro-caudally, and quickly placed in 2% TTC solution (Servicebio, Wuhan, China) at a constant temperature of 37 °C to avoid incubation in light for 30 min, and fixed in 10% paraformaldehyde (Servicebio, Wuhan, China) after staining. Stored in a refrigerator at 4 °C for 24 h. After 24 h, it was taken out and photographed. The infarct area of the brain slice was measured using the medical image analysis software ImageJ (Stuttgart, Germany) (red indicates normal brain tissue, and pale indicates the infarct area). The infarct volume (%) was calculated as the infarct area relative to the contralateral hemisphere area in each slice.

Western Blotting (WB) Analysis
The anesthetized rats were sacrificed by decapitation and the brain was quickly removed, placed in 4 °C artificial cerebrospinal fluid to rapidly removed the hippocampal tissue, placed in a cryopreservation tube, labeled and stored in a − 80 °C refrigerator for later use. The frozen hippocampal tissue was removed, weighed, and placed in a clean EP tube, RIPA lysis solution and protease inhibitor PMSF were added, and the hippocampal tissue was disrupted using an ultrasonic oscillator, and the whole process was run on ice to ensure protein activity. The extracted supernatant was put into a pre-cooled 4 °C centrifuge, centrifuged at 12,000 rpm for 5 min, and the centrifuged supernatant was quantified by BCA method to prepare a sample with a protein content of 3 mg/ml , and loading buffer was added to the sample. The sample was then placed in boiling water and boiled for 10 min. After cooling, the electrophoresis samples were divided and stored in a − 80 °C refrigerator for later use. Samples were taken for electrophoresis, and the amount of protein is 15 µg. After electrophoresis, the protein was transferred to PVDF membrane by wet transfer method, and blocked with 5% skim milk for 2 h, and then add rabbit anti-LC3B (1:2000, Abcam, ab192890), rabbit anti-Beclin1 (1:1000, Abcam, ab62557), rabbit anti-AMPK (1:1000, Abcam, ab32047), rabbit anti-ULK1 (1:1000, Abcam, ab240916), rabbit anti-NLRP3 (1:1000, Abcam, ab263899), rabbit anti-IL-1β (1:1000, Abcam, ab216995), rabbit anti-IL-18 (1:1000, Abcam, ab207323), and rabbit anti-GAPDH (1:1000, Bioss, 0755R, Beijing, China), incubate overnight at 4 °C in a shaker. After washing with TBST buffer the next day, the PVDF membrane were hybridized with horseradish peroxidase-conjugated secondary antibody (goat anti-rabbit, 1:1500, Bioss, bs-0295G-SA, Beijing, China) for 2 h. Images were captured using a chemiluminescence imaging system (Tanon5200, Shanghai, China). ImageJ (Stuttgart, Germany) image analysis software was used to calculate the ratio of the gray value of the target protein to the internal reference protein for statistical analysis.

Immunofluorescence (IF) Staining
The anesthetized rats were fixed in the supine position on the operating table, and the brains were removed by cardiac perfusion. Brain tissue was fixed in 10% paraformaldehyde (Servicebio, Wuhan, China) and stored in a 4 °C refrigerator. After 3-5 days, tissues were dehydrated in 20% sucrose-formalin solution for 2 days, stored at 4 °C, embedded in paraffin after removal, and sectioned with a paraffin microtome. Paraffin sections were dewaxed in xylene and alcohol of different concentrations in turn, then placed in citric acid repair solution (Servicebio, Wuhan, China) for high-temperature repair and blocked with 20% BSA for 1 h. Rabbit anti-LC3B (1:200, Abcam, ab192890), rabbit anti-Beclin1 (1:100, Abcam, ab62557), rabbit anti-AMPK (1:250, Abcam, ab32047), and rabbit anti-ULK1 (1:100, Abcam, ab240916) were added, and the sections were incubated overnight at 4 °C. The next day, fluorescence-labeled secondary antibody (goat anti-rabbit 488, 1:1000, Abcam, ab150077) was incubated for 1 h at 40 °C, stained with the nuclear fluorescent dye propidium iodide (PI, Servicebio, Wuhan, China) for 5 min, and then anti-fluorescence quenching mounting tablets (Servicebio, Wuhan, China) was added. Finally, a laser scanning confocal microscope (PerkinElmer, USA) was used to observe the expression of the target protein in the CA1 area of the rat hippocampus, to compare the fluorescence intensity and to perform statistical analysis.

Transmission Electron Microscopy (TEM)
Rat hippocampal tissue was harvested after reperfusion for 24 h, the sample size was approximately 2 mm × 2 mm, and the tissue was placed in 2.5% glutaraldehyde special fixative for electron microscopy (P1126, Solarbio, Beijing, China). After staining and dehydration, Epon resin (Brand, Germany) was added and incubated for 24 h. After slices of approximately 50-70 nm thickness were prepared with an ultramicrotome (RMC-PXL, USA), the autophagosome morphology was observed with a transmission electron microscope (FEI Tecnai G2 F30).

Statistics Analysis
GraphPad Prism 8.0.2 software (San Diego, CA, USA) was used for statistical analysis. Normality test was applied before one-way analysis of variance (ANOVA) for multiple 1 3 group comparison. Data fitting a parametric distribution were tested for significance using Student's t-test and oneway analysis of variance (ANOVA) with Kruskal-Wallis test. Non-normally distributed measurement data were tested by using non-parametric tests. P < 0.05 indicates that the difference is statistically significant.

Establishment and Validation of MCAO Rat Model
By laser speckle observation of cerebral blood flow and TTC measurement of cerebral infarction area, we can see that the bilateral cerebral blood flow of rats in the Sham group was normal, while the cerebral blood flow perfusion on the injured side of the MCAO group was significantly insufficient (Fig. 2, P < 0.0001). In addition, the mNSS behavioral score of the MCAO group at 24 h was significantly increased compared with the Sham group, indicating that the nerve function of the rats was damaged (Fig. 3B-F, P < 0.0001). The above results indicated that the MCAO model was successfully established in this experiment, and the damaging effect was stable.

Isoflurane Post-treatment Can Upregulate the Autophagy Activity of Neuronal Cells After CIRI in Rats
First, by Western blotting, it can be seen that the expression of autophagy-related proteins Beclin1 and LC3B in the MCAO group was significantly higher than that in the Sham group, and immunofluorescence also has consistent results, indicating that CIRI can activate the autophagy activity of rat hippocampal neuronal cells (Fig. 4, Western blotting: Beclin1 P < 0.01, LC3B P < 0.01; immunofluorescence: Beclin1 P < 0.0001, LC3B P < 0.01). Next, the expression of Beclin1 and LC3B was further increased in the M + ISO group, and when autophagy was inhibited, the expression levels of Beclin1 and LC3B were significantly lower in the M + I + B group than in the M + ISO group, indicating that isoflurane post-treatment can further promote neuronal autophagy after injury (Fig. 4, Western blotting: Beclin1 P < 0.01 LC3B P < 0.05; Immunofluorescence: Beclin1 P < 0.001, LC3B P < 0.0001). In addition, the observation of cell morphology by transmission electron microscopy also showed that the morphology of rat hippocampal neurons before injury was normal, and a large number of synapses were visible (Fig. 5A-C). After injury, autophagolysosomes appeared, and the damaged cytoplasm of neurons shows Fig. 2 Model identification. A The laser speckle experiment observes the cerebral blood flow. a and b are the skull surface exposed after incision of the rat scalp; c and d are the grayscale images of the laser spot; e and f are the pseudocolor images after imaging, and the red is the cortical cerebral blood perfusion area. B The area of cerebral infarction was determined by TTC experiment; the area with suf-ficient cerebral blood flow was red, and the area of infarct ischemia was pale. C Statistics graph of laser speckle. The ordinate is the ratio of right hemisphere cerebral blood flow to bilateral cerebral blood flow (****P < 0.0001). D Percentage of infarct volume after TTC staining. (**P < 0.01) autophagy-like vacuoles, multiple vacuoles appear in the cytoplasm, mitochondrial morphology is abnormal, and cell synapses are also significantly reduced (Fig. 5D-F). The area of autophagolysosomes was also significantly increased in isoflurane-treated rat hippocampal tissue (Fig. 5G-I). At the same time, the performance of neurological function of rats in behavioral experiments also shows obvious changes: it can be seen from the scores of motor test, sensory test, balance beam test, loss of reflex and abnormal movement in the mNSS score. Compared with the MCAO group, the mNSS Fig. 3 Behavioral experiment. A The basic value of the mNSS score of each group of rats before modeling. B The total mNSS score of each group of rats after modeling. C Motor test scores in mNSS of each group. D Sensory test score in mNSS of each group. E Balance beam test score in mNSS of each group. F Loss of reflex and abnormal movement score in mNSS of each group. G, H Schematic diagram of the dark avoidance experiment. I, J The basic values of the incubation period (STL) and the number of errors entering the dark box (ET) of each group of rats in the dark avoidance experiment before modeling. K Each group after modeling the time for the rats to enter the incubation period. L The number of errors in which the rats of each group entered the dark box after modeling. ****P < 0.0001 score of the M + ISO group was significantly reduced and compared with the M + I group, the score of the M + I + B group was visibly increased (Fig. 3C-F, Western blotting: Beclin1 P < 0.001, LC3B P < 0.001; Immunofluorescence: Beclin1 P < 0.0001, LC3B P < 0.0001). At the same time, the dark avoidance experiment also showed consistent experimental results: the step latency (STL) of rats in the M + ISO group was significantly longer than that of the MCAO group, and the number of errors entering the dark box (ET) was also apparently reduced, whereas in the M + I + B group, STL was significantly shortened, and ET was apparently increased (Fig. 3K-L, all P < 0.0001), indicating that the cognitive and memory function of the rats was apparently improved by isoflurane post-treatment. As the autophagy activity was inhibited, the cognitive and memory function of rats after isoflurane treatment also decreased significantly. These results showed that isoflurane can reduce nerve damage after cerebral ischemia-reperfusion by upregulating autophagy and improve the cognitive and memory function in rats.

After CIRI in Rats, Isoflurane Post-treatment Upregulates Autophagy and Inhibits the Release of Inflammatory Factors from NLRP3 Inflammasome
The results of Western blotting experiment showed that, compared with the MCAO group, the expression of NLRP3 inflammasome and related inflammatory factors IL-1β and IL-18 were significantly decreased in the M + ISO group, indicating that isoflurane post-treatment can reduce the inflammatory injury caused by cerebral ischemia-reperfusion (Fig. 7, NLRP3 P < 0.01, IL-1β P < 0.05, IL-18 P < 0.001). In addition, the expression of NLRP3, IL-1β and IL-18 was visibly increased in the M + I + B group compared with the M + ISO group, indicating that after autophagy is inhibited, inflammation is further increased (Fig. 7, NLRP3 P < 0.01, IL-1β P < 0.01, IL-18 P < 0.01), that is to say, the alleviating effect of isoflurane post-treatment on inflammation may be achieved through the upregulation of autophagy.

Isoflurane Post-treatment Upregulates Autophagy After Focal CIRI in Rats Through the AMPK/ULK1 Signaling Pathway, Thereby Inhibiting the Release of Inflammatory Factors by the NLRP3 Inflammasome
From the results of Western blotting experiment, it can be seen that the expressions of NLRP3, IL-1β, and IL-18 in the M + I + C group given AMPK pathway inhibitors were significantly increased compared with the M + ISO group, indicating that the inhibitory effect of isoflurane post-treatment on NLRP3 inflammasome may be achieved by activating the AMPK/ULK1 signaling pathway and upregulating autophagy after CIRI in rats (Fig. 8, NLRP3 P < 0.001, IL-1β P < 0.01, IL-18 P < 0.01).

Discussion
Because cerebral ischemia-reperfusion injury (CIRI) can cause severe brain tissue damage and is an important factor leading to poor prognosis such as disability or death in patients with ischemic stroke  it is particularly important to explore treatments that can effectively promote patient recovery of patients. Isoflurane is a safe and reliable inhalational anesthetic gas. In recent years, it has been widely studied due to its brain protective effect (Zhang et al. 2022a, b). Our research group also found in a preliminary study that 1.5% isoflurane post-treatment showed better , and LC3B in different groups in the immunofluorescence experiment. Green fluorescence is the target protein, red fluorescence is the nucleus, and the ruler is 500 µm; J-M Quantitative analysis of AMPK, ULK1, Beclin1, and LC3B expression levels. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 brain protection (Wang et al. 2016). In this study, we verified the success of the MCAO model through by laser speckling and TTC, and behavioral experiments also showed that the damaging effect of the reperfusion model was relatively stable. Among them, in the behavioral experiments we found that after 24 h of cerebral ischemia-reperfusion in rats, the total mNSS score increased significantly. In the motor test, the rats turned around or turned to the paralyzed side when walking, and could not hold the balance beam firmly in the balance beam test and easily fell, in addition, the auricular reflex and corneal reflex were significantly weakened, and showed a clear panic response to noise. In the dark avoidance experiment, although the rats in the MCAO group had undergone adaptation and electrical stimulation training in the dark room, they still entered the dark room in a short period of time, and repeatedly entered within 300 s, it indicated that the cognitive, learning and memory abilities of the rats were significantly decreased after CIRI. In CIRI, hippocampal tissue has been proven to be more sensitive to ischemia and hypoxia . Therefore, 1.5% isoflurane post-treatment was used to discuss the changes and regulatory mechanisms of neuronal autophagy activity in the hippocampal CA1 region of CIRI rats.
Autophagy is an important cellar degradation program that can play a protective role in a variety of cells, degrading waste proteins and damaged organelles, regulating cell growth, metabolism and survival, and maintaining cell homeostasis (Boya et al. 2013;Ha et al. 2017;Zhang et al. 2022a, b). Studies have shown that CIRI can activate the autophagy in neurons (Ping et al. 2021) and the further upregulation of autophagy can significantly improve nerve function after injury and play an important role in neuroprotection Wen et al. 2019;Zhang et al. 2022a, b). Microtubule-associated protein light chain 3 (LC3), the mammalian homolog of yeast ATG8, is synthesized and cleaved into LC3-I, which activates transcription factor 4 protease, thereby activating the autophagy process. LC3-I is linked to the lipid phosphatidylethanolamine to form LC3-II, and the protein level of its membrane component is often used to measure autophagic activity. The LC3 Fig. 7 The effect of isoflurane post-treatment on NLRP3 after the use of autophagy inhibitors. A The expression of NLRP3 inflammasome and the inflammatory factors IL-1β and IL-18 released by Western blotting experiments in different groups. B-D The quantitative analysis statistics of the expression levels of various proteins in Western blotting. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 family includes three members, LC3A, LC3B, and LC3C, of which LC3B is a well-known marker of autophagy and one of the most widely studied members. Beclin1 is an important autophagy-related molecule in I/R injury and plays an important role in inducing the initiation of autophagy and promoting autophagosome maturation. Therefore, in this study, LC3B and Beclin1 were selected as the main protein molecules to observe the changes in autophagy activity. During the study, we found that the expression of the levels of autophagy-related proteins Beclin1 and LC3B were significantly increased in MCAO group rats as demonstrated by Western blotting and immunofluorescence experiments, which further supports the conclusion that CIRI can activate the autophagy activity of neural cells in injured rats. In addition, we found that isoflurane post-treatment further increased the expression of Beclin1 and LC3B. At the same time, mNSS scores and dark avoidance experiment showed that the neurobehavior and learning and memory function of rats were obviously improved, indicating that isoflurane treatment can ameliorate the neurological damage of rats by further upregulating the autophagy activity of rat hippocampal neurons after CIRI.
AMP-activated protein kinase (AMPK) is an evolutionarily conserved serine/threonine protein kinase that can inhibit cellular energy expenditure pathways while activating compensatory energy production processes to achieve energy homeostasis (Sheng et al. 2014). Studies have shown that AMPK-mediated autophagy activation is a potential protective mechanism in the early stages of brain injury (Inoki et al. 2012;Jiang et al. 2014Jiang et al. , 2015. In the absence of energy, AMPK can directly activate the ULK1 complex to induce autophagy (Hurley et al. 2017). In this study, we found that when isoflurane was used, the expression of AMPK and ULK1 was significantly increased compared to the model group, and the expression of the autophagy-related proteins Beclin1 and LC3B was also apparently increased. In addition, behavioral experiments also showed that the neurological damage in the rats was significantly ameliorated. However, when AMPK was inhibited, the expressions of Beclin1 and LC3B were significantly decreased, and behavioral experiments also showed that the neurological function of rats was severely damaged, indicating that isoflurane posttreatment may upregulate autophagy of neuronal cells after CIRI by activating the AMPK/ULK1 signaling pathway, thereby exerting brain protection.
In addition, studies have shown that the NLRP3 inflammasome plays a key role in the development of CIRI, it can recruit and activate procaspase-1 to generate active caspase-1, and then convert cytokine precursors pro-IL-1β and pro-IL-18 to mature and biologically active IL-1β and IL-18, respectively, which once activated, it triggers a series of inflammatory responses (Palomino-Antolin et al. 2021). Therefore, targeted control of the activation of NLRP3 inflammation is of great importance in the treatment of CIRI. Some studies have also reported that the activation and upregulation of autophagy can well inhibit the inflammatory response mediated by NLRP3 inflammasome , Yu et al. 2021. In this study, we found that when CIRI rats were post-treated with isoflurane, the expression of the NLRP3 inflammasome and its mediated IL-1β and IL-18 were significantly reduced. When specific autophagy inhibitors Baf-A1 and the AMPK inhibitor compound C were used, the expression of NLRP3, IL-1β, and IL-18 were obviously increased, indicating that isoflurane post-treatment can inhibit the release of inflammatory factors from NLRP3 inflammasome, and this inhibitory effect may be achieved by isoflurane through the up-regulation of autophagy by the AMPK/ULK1 signaling pathway. At the same time, behavioral experimental results also show that isoflurane improves the neurobehavior and cognitive memory learning function of CIRI rats by regulating autophagy and inhibiting NLRP3-mediated inflammation. In summary, the results of this study may provide effective evidence for the treatment of cerebral ischemia-reperfusion injury-related diseases in terms of targeted regulation of autophagy and inhibition of inflammation.
There are still some limitations associated with this research. First, in this study, we only test the influence of downregulation of autophagy in CIRI rats by autophagy inhibitor and AMPK inhibitor, and the possible effect of overexpression of autophagy was not investigated. If the experiment can increase the group that promotes autophagy, the experimental conclusion can be more fully verified. Second, due to the complex of the changes in the autophagic induced by CIRI, the relationship between the brain protective properties of ISO post-treatment and the autophagy will be explored in the future research. Finally, although isoflurane has been used in the clinic for many years, the possible side effects of isoflurane should not be ignored, such as depressing the respiratory center, lowering blood pressure, causing arrhythmia, and so on. Meanwhile, the length of time that isoflurane is administered, prolonged exposure may result in neurotoxicity (Burchell et al. 2013).

Conclusions
Our results suggest that autophagy is activated in hippocampal neurons after CIRI in rats. Isoflurane post-treatment can further upregulate autophagy by activating the AMPK/ ULK1 signaling pathway, and the upregulated autophagy can further inhibit NLRP3 inflammation, which releases IL-1β, IL-18, and other inflammatory factors, thereby reducing the damage of hippocampal neurons in CIRI rats and improving the neurobehavioral function and cognitive learning function of rats. In general, the results of this study may provide effective clues for the treatment of CIRI-related diseases in terms of targeted regulation of autophagy and inhibition of inflammation.