We used the following reagents during this study: ciprofol (Cat# H20200013, Liaoning Haisi Ke Pharmaceutical, Shenyang, Liaoning, China), 2,3,5-triphenyltetrazolium chloride (TTC) (Cat# T109275, Aladdin, Shanghai, China), pentobarbital sodium (Cat# BC076, Solarbio Science & Technology Company, Beijing, China), 4% paraformaldehyde fixative solution (Cat# P395744, Aladdin), formic acid (chromatography grade) (Cat# F433212, Aladdin), acetonitrile (chromatography grade) (Cat# A104440, Aladdin), glycerol (Cat# G116214, Aladdin), sucrose (Cat# S112226, Aladdin), HEPES-Tris (Cat# H109407, Aladdin), KCl (Cat# P122134, Aladdin), EGTA (Cat# E104434, Aladdin), NADH (Cat# N196977, Aladdin), BSA (Cat# D259432, Aladdin), protoporphyrin IX (Cat# F2420066, Aladdin), decylubiquinone (Cat#C463488, Aladdin), hexaammineruthenium(II) chloride (Cat# H282727, Aladdin), FMN ( Cat# R106430, Aladdin), SF6847 (Cat# T129439, Aladdin), glutamic acid (Cat# G121400, Aladdin), succinic acid (Cat# S431423, Aladdin), trichloroacetic acid (Cat# T104257, Aladdin), mannitol (Cat# M108829, Aladdin), gentamicin (Cat# A132913, Aladdin) and potassium ferricyanide (Cat# P111567, Aladdin).

We used the following equipment and software during this study: analytical balance (No.ZCK201708115, Mettler-Toledo, Columbus, OH, USA), silicon rubber plug (No. 40-3756, Doccol, Sharon, MA, USA), micropipettes (QH05656, Thermo Fisher Scientific, Waltham, MA, USA), benchtop centrifuge (No.10895, Sigma-Aldrich, St. Louis, MO, USA), high-speed refrigerated centrifuge (No. LC-2019-002, Sigma-Aldrich), ultra-pure water system (No. 201100078, Millipore, Billerica, MA, USA), 4 °C refrigerator, –20 °C freezer, –80 °C ultra-low temperature freezer (No. BH03G00940OTXL8CDGCD, Haier, Qingdao, Shandong, China), disposable vacuum blood collection tube ( No. WHB-30-1, Shanghai WoHong Biological Technology Co., LTD, Shanghai, China), ultra-high-performance liquid chromatography-quadrupole time-of-flight mass spectrometry (UPLC-Q/TOF-MS) (No. LC-MS110502, Waters, Milford, MA, USA), mass spectrometer TSQ8000 (No. LC-MS100702, Thermo Fisher Scientific), microinjection pump (No. 6132301008519, Shenzhen Maiketian Biomedical Technology Co., Ltd., Shenzhen, Guangdong, China), MassLynx software (version 4.1, Waters), Image-Pro Plus software (version 5.0, Media Cybernetics, Rockville, MD, USA), SPSS software (version 20.0, International Business Machines Corporation, Chicago, IL, USA), SIMCA-P software (version 11.5, Sartorius, Gottingen, Germany), vortex mixer (Serial No.5426YN331111, Eppendorf, Hamburg, Germany), and scanner (Serial No.X6LW004066, Epson, Suwa, Japan).

Specific pathogen-free male Sprague–Dawley rats, weighing 210–230 g, were purchased from Changchun Yisi Experimental Animal Technology (Changchun, Jilin, China). The experimental animals were housed in the Central Laboratory of the Fourth Affiliated Hospital of Harbin Medical University throughout the experiment, with a temperature of 22 $\pm$ 2 °C, a relative humidity of 55% $\pm$ 5%, and a light-dark cycle of 12 h each day. All animals were fed standard food and allowed free access to water. The animals were acclimated to the rat facility for 7 days before the experiment. Adult male rats were randomly divided into a ciprofol group, a middle cerebral artery occlusion (MCAO) group, and a Sham group. During the experiment, all animals were provided with proper feeding and care according to relevant standards, and this experiment was approved by the Animal Ethics Committee. All experiments were conducted in accordance with relevant guidelines and regulations.

A rat MCAO model was established with reference to the modified Zea Longa method to simulate CIRI [14]. Rats were anesthetized by intraperitoneal injection of 2% pentobarbital sodium (50 mg/kg) and fixed in the supine position. A silicon rubber plug was inserted from the proximal end of the external carotid artery (ECA) to occlude the origin of the middle cerebral artery (reperfusion was confirmed by laser Doppler). Ninety minutes later, the plug was gently removed, the ECA was completely ligated. The sham surgery group only underwent incision and suturing of the neck without any other procedures.

After removing abdominal hair and disinfecting with iodine, the femoral vein was separated on one side of the rat, a puncture was made with a disposable intravenous puncture needle, and a catheter was placed and connected to an extension tube. Ciprofol was injected into the femoral vein using a microinjection pump at a rate of 9 mg/kg/h starting after successful placement of the catheter in the rat vein and lasting for 2 h until the time of removing the plug. Rats in other groups were infused with physiological saline at a rate of 9 mg/kg/h [15].

After 24 h, a neurobehavioral evaluation was performed. Using the modified Garcia test, the following six parameters were tested and recorded: reperfusion spontaneous activity, symmetry of limb movements, forelimb extension, climbing, body proprioception, and response to whisker touch. The neurobehavioral scores of each group of rats were recorded according to an 18-point neurological scoring criteria [16].

After 24 h of reperfusion, the rats were euthanized by decapitation after intraperitoneal injection of 2% pentobarbital sodium and whole brain tissues were removed. Brain tissues were frozen at –20 °C for approximately 20 min. Then, the brain was sliced into six coronal sections on an ice block, with each slice being approximately 2-mm thick, and immersed in a 2% TTC solution at 37 °C for 15 min. The volumes of the white areas of brain slices were calculated using Image-Pro Plus software, with the results representing the volume of cerebral infarction [17].

Differential centrifugation was used to separate mitochondrial membranes from frozen brain hemispheres. After a frozen intact brain was divided into two hemispheres along the sagittal plane, the brain cortex with the most damage was separated from the posterior to the anterior, including the cortical and striatal areas. The brain tissue was then stored in a –80 °C freezer. Frozen brain tissue was added to a homogenizer containing 10 mL of isolation buffer mannitol/sucrose/ethyleneglycol-bis (b-aminoethyl ether)-N,N,N,N^′-tetraacetic acid (EGTA) medium (MSE) (225 mmol/L mannitol, 75 mmol/L sucrose, 20 mmol/L 4-(2-Hydroxyethyl)-1-piperazineethanesulfonic acid-Tris (HEPES-Tris), 1 mmol/L Ethylene glycol-bis(2-aminoethylether)-N,N,N^′,N^′-tetraacetic acid (EGTA), 1 mg/mL Bovine Serum Albumin (BSA), pH 7.4) with 80 µg/mL of penicillin and homogenized 60 times to release low molecular weight metabolites. After centrifugation at 1500 g for 5 min at 4 °C to remove tissue fragments, the supernatant was centrifuged at 20,000 g for 15 min at 4 °C and washed twice with MSE buffer without BSA. The final pellets are suspended in 60 µL of the same buffer and stored at –80 °C until use [18].

Mitochondrial respiratory chain complex I (MRCC-I) activity was determined by spectrophotometry at 25 °C using a microplate reader in 0.2 mL of buffer (125 mmol/L KCl, 20 mmol/L HEPES-Tris, 0.02 mmol/L EGTA, pH 7.6) [19]. The activity of MRCC-I at 0.15 mM of NADH was measured at 340 nm in a buffer containing 40 µg/mL of gentamicin and 1 mM of potassium ferricyanide (NADH culture medium). NADH: Q reductase was measured in a NADH culture medium containing 2 mg/mL BSA, 60 µmmol/L decylubiquinone, and mitochondrial protein. NADH: HAR reductase was detected in the NADH culture medium containing 1 mM HAR and mitochondrial proteins.

All data were analyzed using SPSS software. The Shapiro-Wilk test was used to check whether the data were normally distributed. Neurobehavioral test scores were compared using the Wilcoxon matched pair signed rank test. In addition, one-way analysis of variance was performed for comparisons of more than two groups, followed by Tukey’s test. For the comparison of mitochondrial complex I enzyme activity between different groups, multiple t-tests with false discovery rate (FDR) correction for multiple comparisons were used. The normally distributed data are presented as the mean $\pm$ standard error of the mean, and neurobehavioral scores are shown as the median, 25th percentile, 75th percentile, and minimum and maximum values in the box plot. Statistical significance was set at p 0.05, with one, two, and three asterisks representing significance levels below 0.05, 0.01, and 0.001, respectively.

The experimental schedule is shown in Fig. 1A. Thirty minutes before blocking the MCA, ciprofol administration was commenced at a rate of 9 mg/kg/h through a pump for a total duration of 2 h. Various indicators were tested after 24 h of cerebral reperfusion. The results showed that the ischemic stroke volume percentage of the MCAO group was 28.6 $\pm$ 4.1%, whereas it was 20.3 $\pm$ 3.0% in the ciprofol group (Fig. 1B,C), with a significant difference between the two groups (p 0.01). Compared to the Sham group, the neurological behavioral scores of rats in the MCAO group were significantly lower (p 0.001), whereas the neurological behavioral scores of rats in the ciprofol group were significantly higher than those in the MCAO group (Fig. 1D) (p 0.001). These results indicate that ciprofol treatment significantly alleviated the neurofunctional damage induced by cerebral ischemia-reperfusion.

Fig. 1.

Neuroprotective effects of ciprofol on cerebral ischemia-reperfusion. (A) Schematic diagram of the experimental timeline. (B) Representative TTC-stained images of the brain at 24 h after middle cerebral artery occlusion and reperfusion. The scale bar = 10 mm. (C) Quantitative analysis of infarct volume (n = 6, F (2,15) = 150.170, p = 0.0000000001). (D) Comparison of neurological behavioral scores at 24 h after reperfusion. Values are presented as the mean $\pm$ standard deviation (n = 12) (** p 0.01, *** p 0.001). MCAO, middle cerebral artery occlusion; TTC, 2,3,5-triphenyltetrazolium chloride; NS, normal saline.

This study used UPLC-Q/TOF-MS for metabolomic analyses of serum and brain tissue samples. To further investigate the similarities and differences between the Sham, MCAO, and ciprofol groups, we constructed a PLS-DA model that is widely used in metabolomics research. Based on the PLS-DA plots, we found clear separation trends in the serum (Fig. 2A) and brain (Fig. 2B) tissues of the three groups of rats, indicating good precision and reproducibility of the metabolomic data. The plots clearly show two main components of serum, PC1 (30.1%) and PC2 (25.1%), and two independent components of the brain, PC1 (37.7%) and PC2 (11.9%), suggesting a significant effect of ciprofol treatment. The serum samples achieved R² Y and Q² values of 0.492 and –0.984, respectively, in the PLS-DA score plots, whereas the brain samples had R² Y and Q² values of 0.331 and –1.014, indicating the good fit and reliability of the model. Additionally, this study used RPT to assess the model quality, with the green regression line of Q² points intersecting below zero on the left side of the vertical axis, further validating the effectiveness of the model (Fig. 2C,D).

Fig. 2.

Multivariate statistical analyses of serum and brain tissue metabolomic data. (A) PLS-DA of serum samples. (B) PLS-DA of brain tissue samples. (C) RPT plot of the PLS-DA model for serum samples. (D) RPT plot of the PLS-DA model for brain tissue samples. PLS-DA, partial least squares discriminant analysis; RPT, response permutation testing.

Research findings have shown that ciprofol causes significant changes in a variety of metabolites during the treatment of MCAO rats. To explore the biological significance of these metabolites in ciprofol-treated MCAO rats, we identified three groups of metabolites with significant differences. Previous studies have indicated that variables with VIP values of $>$1 have a significant impact on the separation of samples in PLS-DA and can be used as criteria for screening for differential metabolites [17, 20]. However, no study has confirmed a causal relationship between these screened differential metabolites and cerebral ischemia injury. For example: erucamide, lathosterol, and threonate, among others. Therefore, the selection criteria used in this study were a top-50 VIP value and p 0.05.

Based on these screening criteria, we identified 19 differential metabolites in the serum of the Sham, MCAO, and ciprofol groups (Table 1). To comprehensively demonstrate the relationships between these three groups of samples and the differences in the expression patterns of these metabolites, we used a hierarchical clustering analysis to determine the quality and intensity of these differential compounds. The cluster heatmap showed variations in FMN metabolites between the serum sample groups, with color changes reflecting the overall changes in metabolites (Fig. 3A). Compared to those in the Sham group, the levels of lysophosphatidylethanolamine (LPE), 1-O-hexadecyl-SN-glycero-3-phosphocholine (lyso-PAF C-16), phosphatidylethanolamine (PE), sphinganine, and L-carnitine were significantly increased in the MCAO group. In contrast, the levels of these metabolites were lower in the ciprofol group than in the MCAO group. Additionally, citrate, succinate, lysophosphatidylcholine (LPC), and deoxycholic acid levels showed a downward trend in the MCAO group, whereas the levels of these metabolites were further decreased in the ciprofol group.

Fig. 3.

Three sets of rat serum and brain tissue complex I sub-base FMN and differential metabolite clustering analyses. (A) Heat map of significantly expressed metabolites clustered in serum samples. (B) Heat map of FMN and other significantly expressed metabolites clustered in brain tissue samples. (Data standardized between –3 and 3; colors from blue to red indicate metabolite expression levels from low to high). FMN, flavin mononucleotide; lyso-PAF, lysophosphatidyl platelet-activating factor; LPE, lysophosphatidylethanolamine; PE, phosphatidylethanolamine; TXB2, thromboxane B₂; LPC, lysophosphatidylcholine; FA, fatty acid; 12(R)-HEPE, 12(R)-hydroxyeicosapentaenoic acid.

The analysis of brain tissue sampled from the three groups of rats revealed differences in the presence of 31 metabolites (Table 2). The clustered heat map (Fig. 3B) illustrates the changes in metabolites in the brain tissue samples. Compared to the Sham group, the levels of aspartate, N-acetylaspartate (NAA), N-acetylaspartylglutamate (NAAG), 4-hydroxyisoleucine, FMN, and pyruvate were lower in the brain tissues of rats in the MCAO group, whereas ciprofol reversed the decreases in these metabolites in the brain tissues of rats in the MCAO group.

We then analyzed linear correlations between different metabolites in the serum and brain samples by calculating Pearson correlation coefficients and conducting statistical tests for p-values. The correlation analysis results for serum metabolites (Fig. 4A) showed that citric acid, pyruvate, and lipids were highly positively correlated with lipid substances. Correlation analysis results for different brain metabolites (Fig. 4B) indicated that pyruvate and FMN had high positive correlations with lipid and amino acid metabolites. However, the correlation between succinic acid and the other metabolites was weaker. This finding suggests that citric acid, pyruvate, FMN, lipid substances, and amino acid metabolites play important roles in the neuroprotective effects of ciprofol in MCAO rats.

Fig. 4.

Correlations between complex I cofactor FMN in the serum and brain tissue samples obtained from three groups of rats with differential metabolites. (A) Correlation map of significantly expressed metabolites in serum samples. (B) Correlation map of FMN and other significantly expressed metabolites in brain tissue samples. (The highest correlation coefficient is 1 (red), indicating complete positive correlation, while the lowest correlation coefficient is –1 (blue), indicating complete negative correlation. The blank part in the figure represents p $>$ 0.05 from the correlation statistical test, while the colored part represents p 0.05).

To explore the potential protective mechanisms of ciprofol against CIRI, differential metabolites were used to construct metabolic pathways. According to the Reactome enrichment analysis, of the 20 pathways with the smallest p-values in the three groups of rat serum samples, the key metabolic pathways for neuroprotection by ciprofol (Fig. 5A) were the tricarboxylic acid cycle, pyruvate metabolism, and mitochondrial respiratory chain. Pathways in brain tissue samples (Fig. 5B) also included steroid metabolism, as well as pathways for aspartate and glutamate metabolism. There was a clear correspondence between the enriched differential metabolites and metabolic pathways in the three groups of rats, with citric acid being involved in the tricarboxylic acid cycle, pyruvate metabolism, and mitochondrial respiratory chain; pyruvate being involved in pyruvate metabolism and tricarboxylic acid cycle processes; and FMN being associated with the tricarboxylic acid cycle and mitochondrial respiratory chain.

Fig. 5.

Enrichment analysis of rat serum and brain tissue samples obtained from the three groups, including the tricarboxylic acid cycle, mitochondrial respiratory chain, and other top 20 pathways. (A) Bubble plot of Reactome pathways in serum samples. (B) Bubble plot of Reactome pathways in brain tissue samples. Pathways with larger bubbles contain more different metabolites; the color of the bubbles changes from blue to red, with smaller enrichment p-values indicating greater significance. IP, inositol 1,4,5-trisphosphate; SLC, solute carrier.

To clarify the potential mechanisms, we correlated important metabolic pathways with differential metabolites, presenting a comprehensive analysis of the effects of ciprofol on metabolic processes in MCAO rats (Fig. 6). The citric acid cycle links pyruvate metabolism and mitochondrial respiratory chain pathways. Compared with the Sham group, the FMN content of the brain tissue samples of the ciprofol group were significantly downregulated, and the MCAO group showed a further significant downregulation of FMN content, indicating that ciprofol regulates FMN content in the brains of MCAO rats.

Fig. 6.

Analysis of differential metabolites in the tricarboxylic acid cycle and mitochondrial respiratory chain pathway in the three groups of rats. Results of the Kruskal–Wallis test showing significant differences in succinic acid (H(2) = 4.393, p = 0.0112), citric acid (H(2) = 8.433, p = 0.0148), and pyruvate (H(2) = 8.047, p = 0.0179). Results of the Bonferroni correction showing significant differences in succinic acid (adjusted p = 0.1317, MCAO group vs. ciprofol group), citric acid (adjusted p = 0.0128, Sham group vs. MCAO group) and pyruvate (adjusted p = 0.0148, Sham group vs. MCAO group) (* p 0.05).

MRCC-I activity in rats with cerebral ischemia-reperfusion is closely related to CIRI. To detect MRCC-I activity in the mitochondrial membrane, we examined the effects of different concentrations of ciprofol. The results showed that, after treatment with ciprofol at concentrations of 25, 50, and 100 µmol/L [21], the activities of the Q enzyme and hexaammineruthenium (HAR) reductase of MRCC-I were significantly increased (Fig. 7A,B). Within 4 h, the Q-reductase activity showed a concentration-dependent increase with increasing concentrations of ciprofol. Within 1 h, the activity of HAR reductase increased with the concentration of ciprofol; however, after 1h, ciprofol at 25-µmol/L demonstrated higher activity compared to the 50-µmol/L concentration. These findings indicate that ciprofol enhances the activity of MRCC-I after cerebral ischemia; however, the concentration dependence of HAR reductase changes after 1 h.

Fig. 7.

The effect of ciprofol on the activity of MRCC-I after cerebral ischemia-reperfusion. (A) Q reductase activity ratio. (B) HAR reductase activity ratio. C 25 µM, C 50 µM, C 100 µM represents 25, 50, and 100 µmol/L ciprofol. HAR, hexaammineruthenium.

Within a 15-min time period after pretreatment with 100 µmol/L of ciprofol and without any other treatments, FMN was added, trichloroacetic acid was used as an FMN-dissociating agent, the Epidermal Growth Factor Receptor (EGFR) inhibitor, SF6847, was applied, and the activities of mitochondrial membrane Q and HAR reductase were determined (Fig. 8A,B). The results showed that, after treatment with 10 µmol/L of FMN, the activities of both enzymes were upregulated. After removal of FMN from the mitochondrial membrane with trichloroacetic acid, the activities of both enzymes were significantly decreased in the FMN inhibitor group, which indicates that ciprofol modulates the amount of membrane-bound FMN. Under RET conditions, the activities of both enzymes were inhibited. In order to inhibit the RET state, the activity of Q reductase was higher than in the RET group, but lower than in the Control group after using the EGFR inhibitor, whereas the activity of HAR reductase was higher than in the Control group. This result indicates that inhibition of RET by ciprofol mainly affects the activity of HAR reductase.

Fig. 8.

Activities of MRCC-I under four conditions: RET, RET inhibition, FMN supplementation, and FMN dissociation. (A) Q reductase activity ratio. (B) HAR reductase activity ratio. FMN: 10 µmol/LFMN, FMN dissociation: 10% trichloroacetic acid. RET: 5 mmol/L succinic acid and 1 mmol glutamic acid. RET inhibition: 50 nmol/L SF6847. RET, reverse electron transfer.