In total, 50 male specific-pathogen-free Sprague-Dawley rats aged 15 months and weighing 260–280 g were purchased from the Animal Laboratory Center of China Medical University (License No. SCXK(Liao)-2013-0001; License No. SYXK(Liao)-2013-0007). The project was approved by the China Medical University Institutional Animal Care and Use Committee (IACUC No. 2017018R). All animals were maintained in individual cages under controlled ambient temperature (24±2°C) and 40–70% humidity in a 12-h light/dark cycle. Standard pelleted chow and drinking water were available ad libitum. The rats were allowed to acclimate to these conditions for at least 1 week. The health and behavior of rats were observed daily by animal husbandry staff at our facility. To reduce the suffering and distress, some animal tools and nesting material were provided to the rats.

The rats were randomly assigned into five groups: i) 10 rats were placed in the blank control group (control group); ii) 10 rats were placed in the sevoflurane group (SEV group; the rats in SEV group were anesthetized for 5 h in 2% sevoflurane); iii) 10 rats were placed in the rapamycin pretreatment group [RAP group; the rats were intraperitoneally injected with rapamycin (20 mg/kg) 2 days before anesthesia for 5 h in 2% sevoflurane (14)]; iv) 10 rats were placed in the TLR4 inhibitor group [TLR group; the rats were intraperitoneally injected with rapamycin (20 mg/kg) 2 days before anesthesia for 5 h in 2% sevoflurane, and TLR4 inhibitor TAK242 (0.5 mg/kg) was intracerebrally injected into rats at 1 h before rapamycin treatment (15)]; v) 10 rats were placed in the sevoflurane+ rapamycin+3-Methyladenine (3MA) autophagy inhibitor group [3MA group; the rats in the 3MA group were intraperitoneally injected with rapamycin (20 mg/kg) 2 days before anesthesia for 5 h in 2% sevoflurane, and 21 mg/kg 3MA autophagy inhibitor was intracerebrally injected at 1.5 h before rapamycin treatment. After 24 h, physiological data were recorded, including heart rate, mean arterial pressure, arterial carbon dioxide tension and arterial oxygen tension. After the Morris water maze (MWM) experiment, blood samples of rats were collected by saphenous vein puncture, centrifuged at 12,000 × g at 4°C to obtain serum and stored at −20°C. Then the rats were sacrificed by overdose anesthesia, 800 mg/kg pentobarbital, intraperitoneally (6,16). The hippocampus of rats was collected, one half was fixed in 4% paraformaldehyde and another half was stored in liquid nitrogen.

The MWM experiment, consisting of an acquisition test and a special probe test, is a well-established method for assessing learning and memory abilities in rodents. The acquisition test lasted 4 days, twice daily, with 15-minute interval between tests. On day 5, the spatial probe test was conducted. The acquisition test was performed to assess learning and memory abilities in rats. Rats were randomly placed into the water in a random order at a fixed entry point of each quadrant. The duration time to search and climb the platform in the water was recorded as the escape latency, and the rats stayed on the platform for at least 3 sec. If a rat failed to climb onto the platform within 120 sec, it was manually guided onto the platform with a stick and made to stay there for 30 sec. It was then returned to the cage for the next test. The spatial probe test was used to assess the spatial memory of rats. After the platform was removed from the pool, the rats were placed into the water from the opposite side of the original target quadrant, and allowed to search for the platform. The number of platform crossings within 120 sec was recorded and the proportion of time spent in the original target quadrant to the total time was calculated.

The hippocampus fixed in paraformaldehyde was immerged in different concentrations of alcohol (70, 80, 90, 95, and 100%), permeabilized in xylene, embedded in wax, and sliced into 4-µm thick sections using a microtome. The sections were dewaxed, stained with hematoxylin for 5 min, washed with PBS, differentiated with 1% alcoholic hydrochloric acid, stained with eosin for 30 sec, followed by dehydration through a graded alcohol series and washed off. Subsequently, these sections were mounted with neutral resin. The pathological changes in the hippocampus were observed using a light microscope.

Cell apoptosis was assessed in accordance with the kit instruction (Roche Diagnostics). Briefly, 5-µm paraffin sections were dewaxed, permeabilized and sealed. After treatment with 50 µl TUNEL reaction solution, the sections were incubated in a wet dark box at 37°C for 60 min. The sections were then incubated with a NEUN antibody (1:300; cat. no. ab177487; Abcam) at 37°C for 1 h. Following this, 50 µl streptavidin-HRP solution and goat anti-rabbit IgG H&L (Cy3^®; 1:1,000; cat. no. ab6939; Abcam) was added followed by incubation in the dark box for 30 min. Nuclei were stained with DAPI and observed under a fluorescence microscope. After obtaining images, the number of total nuclei and TUNEL-positive nuclei were counted. Apoptosis ratio= (the number of TUNEL-positive nuclei)/the total number of nuclei × 100%.

ELISA was performed to assess the expression of brain injury markers S-100β, NSE and inflammatory factors TNF-α, IL-1β, IL-6, and IL-10. Standard preparation (100 µl) and diluted samples (100 µl) were added in the corresponding reaction plate wells, lightly shaken for 30 sec, and incubated at 20–25°C for 20 min. The reaction plate was washed with a washer machine. Serum sample (100 µl) was added in each well and incubated at 37°C for 2 h. After washing, 100 µl HRP-labeled secondary antibody was added in each well and incubated at 37°C for 30 min. After subsequent washing, 50 µl of substrate A and 50 µl of substrate B were added and visualized in a dark room for 15 min. The reaction was terminated by adding 50 µl of stop buffer. OD values were measured at 450 nm using a microplate reader (EXL808). The standard curve was drawn with the OD value as the ordinate and the standard concentration as the abscissa. The curve equation and R value were calculated. The corresponding concentration of the sample was obtained according to the curve equation.

Tissue sample was added in RIPA lysate (cat. no. 89900; Thermo Fisher Scientific, Inc.) containing a protease inhibitor on ice for 30 min. The supernatant was collected, and the protein concentration was measured with a BCA Protein Quantification kit (cat. no. 23225; Thermo Fisher Scientific, Inc.). After 10% SDS-PAGE (20 µg per lane), the samples were transferred onto PVDF membranes. Bax (1:1,000; cat. no. ab32503; Abcam), Bcl-2 (1:1,000; cat. no. ab59348; Abcam), caspase-3 (1:500; cat. no. ab13847; Abcam), Beclin1 (1:2,000; cat. no. ab207612; Abcam), LC3B (1:3,000; cat. no. ab51520; Abcam), TLR4 (1:500; cat. no. ab13556; Abcam), MyD88 (1:1,000; cat. no. ab133739; Abcam), NF-κB p65 (1:1,000; cat. no. ab246347; Abcam) and GAPDH (1:10,000; cat. no. ab181602; Abcam) antibodies were added and incubated at 4°C overnight. Next, the PVDF membrane was washed with PBS, a goat anti-rabbit IgG horseradish peroxidase-conjugated secondary antibody (1:10,000; cat. no. ab6721; Abcam) was added and incubated at room temperature for 2 h. The proteins were visualized with Novex™ ECL Chemiluminescent Substrate Reagent kit (Invitrogen; Thermo Fisher Scientific, Inc.) and gel imaging system (Gel Doc™ XR; Bio-Rad Laboratories, Inc). Absorbance values were analyzed using ImageJ (v1.8.0; National Institutes of Health).

Brain tissue was added in TRIzol reagent (cat. no. 15596026; Invitrogen; Thermo Fisher Scientific, Inc.). According to TRIzol reagent instructions, total RNA was extracted from tissues and cells, and reversely-transcribed into first-strand cDNA (cat. no. 4387406; Invitrogen; Thermo Fisher Scientific, Inc.). In accordance with a Real-time qPCR kit (cat. no. RR820A, Takara Biotechnology Co., Ltd.), PCR was performed with the following conditions: pre-denaturation at 95°C for 30 sec; 40 cycles at 95°C for 5 sec and 60°C for 20 sec; analysis of melting curve: 95°C for 1 sec, 65°C for 15 sec and 95°C for 5 sec. Results were calculated using 2^−ΔΔCq method (17). Primer sequences are listed in Table I.

Data were analyzed using SPSS 19.0 software (IBM Corp.), and expressed as the mean ± standard deviation. Paired comparisons were conducted using a Student's t-test. Intergroup comparisons were analyzed using one way ANOVA followed by Tukey's post hoc test. P<0.05 was considered to indicate a statistically significant difference.

The results from H&E staining revealed that neurons were arranged neatly and tightly, and the cytoplasm and nucleus were plump and distinct in the control group. Neurons were arranged disorderly and loosely, cells were smaller, and nuclear condensation and cytoplasm were decreased in the SEV group. A small number of cells exhibited morphological changes, and normal neurons were still observed, which were arranged neatly and tightly in the RAP group (Fig. 1A). By ELISA detection it was revealed that compared with the control group, serum S-100β and NSE expression were significantly higher in the SEV group (P<0.05). Compared with the SEV group, serum S-100β and NSE expression were significantly lower in the RAP group (P<0.05; Fig. 1B and C). These results indicated that rapamycin reduced sevoflurane-induced brain injury in aged rats.

Models of cognitive dysfunction induced by sevoflurane inhalation in aged rats were intervened with rapamycin for 24 h. An MWM experiment was conducted to determine the effects of rapamycin on learning and memory abilities of sevoflurane-treated rats. In the acquisition test, the escape latency of rats in each group was gradually shortened. From 2 to 4 days, compared with control group, the escape latency was significantly increased in the SEV group (P<0.05). Compared with the SEV group, the escape latency of the RAP group was significantly decreased (P<0.05; Fig. 2A). In the spatial probe test, compared with control group, the percentage of time spent, the distance covered in the target quadrant and the number of platform crossings were significantly decreased in the SEV group (P<0.05). Compared with the SEV group, the percentage of time spent, the distance covered in the target quadrant and the number of platform crossings were significantly increased in the RAP group (P<0.05; Fig. 2B). It is therefore indicated that rapamycin can alleviate sevoflurane-induced learning and memory impairments in aged rats.

TUNEL assay results revealed that compared with the control group, the number of positive cells was significantly higher in the SEV group (P<0.05). Compared with the SEV group, the number of positive cells in the rat hippocampus was significantly lower in the RAP group (P<0.05; Fig. 3). Western blot assay results revealed that compared with the control group, Bcl-2 expression was significantly lower, but Bax and cleaved caspase-3 expression was significantly higher in the SEV group (P<0.05). Compared with the SEV group, Bcl-2 expression was significantly higher, but Bax and cleaved caspase-3 expression was significantly lower after rapamycin pretreatment in the RAP group (P<0.05; Fig. 4B). These findings indicated that rapamycin can inhibit sevoflurane-induced neuronal apoptosis and prevent neuronal degeneration.

ELISA results demonstrated that compared with the control group, IL-1β, IL-6 and TNF-α expression was significantly higher, but IL-10 expression was significantly lower in the SEV group (P<0.05). Compared with the SEV group, IL-1β, IL-6 and TNF-α expression was significantly lower, but IL-10 expression was significantly higher after pretreatment in the RAP group (P<0.05; Fig. 4A). These results indicated that rapamycin relieves sevoflurane-induced inflammatory responses in aged rats.

Western blot assay (Fig. 4B) and RT-qPCR (Fig. 4C) were used to determine the changes in the TLR4/MyD88/NF-κB signaling pathway in the rat brain. Compared with the control group, TLR4, MyD88 and NF-κB p65 expression was significantly higher in the SEV group (P<0.05). Compared with the SEV group, TLR4, MyD88 and NF-κB p65 expression was significantly lower in the RAP group (P<0.05). These findings indicated that rapamycin improves sevoflurane-induced brain injury in aged rats possibly by inhibiting the TLR4/MyD88/NF-κB signaling pathway.

Western blot assay (Fig. 5A) results revealed that compared with the control group, LC3II/I and Beclin1 expression was significantly lower, but p62 expression was significantly higher in the SEV group (P<0.05). Compared with the SEV group, LC3II/I and Beclin1 expression was significantly higher, but p62 expression was significantly lower in the RAP group (P<0.05). The effect of rapamycin weakened after the administration of autophagy inhibitors. Compared with the RAP group, LC3II/I and Beclin1 expression was significantly lower, but p62 expression was significantly higher in the 3MA group (P<0.05). An MWM experiment was conducted to determine the effects of autophagy inhibitor treatment on learning and memory abilities. In the acquisition test, the escape latency of rats in each group was gradually shortened, and in 3–4 days, compared with the RAP group, the escape latency was significantly increased in the 3MA group (P<0.05). In the spatial probe test, compared with the RAP group, the percentage of time spent, the distance covered in the target quadrant and the number of platform crossings were significantly decreased in the 3MA group (P<0.05´Fig. 5B). These results confirmed that rapamycin activates sevoflurane-induced autophagy in aged rats.

To investigate the relationship between rapamycin-activated autophagy and the TLR4/MyD88/NF-κB signaling pathway, the aged rats with sevoflurane-induced cognitive dysfunction were treated with TLR4 inhibitor. It was revealed that compared with the RAP group, LC3II/I and Beclin1 expression was significantly lower, but p62 expression was significantly higher after suppressing the TLR4 signaling pathway (P<0.05; Fig. 6A). ELISA was further applied to detect inflammatory factors, and as a result, the inhibitory effect of rapamycin on inflammatory factors was weakened. Compared with the RAP group, IL-1β, IL-6 and TNF-α expression was significantly higher, but IL-10 expression was significantly lower in the TLR group (P<0.05; Fig. 6B). An MWM experiment was conducted to determine the effects of the TLR4/MyD88/NF-κB signaling pathway on learning and memory abilities. In the acquisition test, the escape latency of rats in each group was gradually shortened, and in 3–4 days, compared with the RAP group, the escape latency was significantly increased in the TLR group (P<0.05). In the spatial probe test, compared with the RAP group, the percentage of time, the distance covered in the target quadrant and the number of platform crossings were significantly decreased in the TLR group (P<0.05; Fig. 6C). These findings indicated that by inhibiting the TLR4 signaling pathway, the effect of rapamycin on the activation of autophagy was weakened, and the protective effect of rapamycin on sevoflurane-induced brain injury was markedly weakened in the aged rats. These results confirmed that rapamycin improves sevoflurane-induced cognitive dysfunction in aged rats by mediating autophagy via the TLR4/MyD88/NF-κB signaling pathway.