Adult male C57BL/6 mice (aged 8–10 weeks, weighing 22–25 g) were sourced from Hunan SJA Laboratory Animal Co., Ltd. (Changsha, Hunan, China). Before the experiment, all mice were given at least one week to habituate to the colony room. During the study, the mice were kept under standardized conditions, with constant temperature and humidity control, a 12–12 h alternating light–dark cycle, and free access to standard rodent chow and water.

Mice were randomly assigned to four experimental groups (n = 13/group) using a computer-generated randomization sequence to minimize selection bias: (1) Sham + vehicle; (2) Sham + FPS; (3) CA/CPR + vehicle; (4) CA/CPR + FPS. Randomization was performed by an investigator (QL) who was not involved in subsequent data analysis.

One designated investigator (QL) was responsible for drug preparation and administration. All other investigators involved in surgical procedures, neurological function assessments, data collection, and statistical analysis remained blind to the group allocation throughout the entire experiment to minimize potential bias.

The CA model was generated in accordance with the protocol described previously [33, 37]. In brief (Fig. 1), mice were weighed to confirm adherence to standard weight parameters. Anesthesia was induced using 5% isoflurane (v/v; vol%; R510-22-10, RWD Life Science Co., Ltd., Shenzhen, Guangdong, China), followed by tracheal intubation. Anesthesia was maintained by mechanical ventilation delivering 1.0%–1.5% isoflurane (v/v; vol%); a gas evacuation apparatus (V101, RWD Life Science Co., Ltd.) was used to remove anesthetic exhaust. Mice were ventilated through the tracheal tube at a tidal volume ranging from 0.3 to 0.5 mL. Electrocardiogram (ECG) was continuously monitored using subcutaneously placed needle electrodes and a PowerLab 26T system (PL26T04; ADInstruments Pty Ltd., Bella Vista, New South Wales, Australia). Body temperature was kept at 37.0 $\pm$ 0.2 °C with the assistance of a heating pad and lamp, with regulation monitored by a rectal probe. For drug administration, a PE10 catheter (424701, Becton, Dickinson and Company, Franklin Lakes, NJ, USA) was placed into the right jugular vein. After a 50 µL flush with heparinized saline (9041-08-1, 0.9%, Sinopharm Chemical Reagent Co., Ltd., Shanghai, China), CA was triggered by injecting 0.5 M potassium chloride (7447-40-7, [KCl], Sinopharm Chemical Reagent Co., Ltd.), 30 µL at 4 °C, followed by cessation of mechanical ventilation (MV). CA was verified by observing asystole on the ECG, cessation of spontaneous breathing, and the occurrence of urinary incontinence. Resuscitation was initiated 9.5 min after CA induction, and consisted of the resumption of mechanical ventilation with 100% oxygen, cessation of the warming protocol, and administration of 100 µL of epinephrine (61-76-7, 32 µg/mL, Sinopharm Chemical Reagent Co., Ltd.). Chest compressions were carried out at 300 beats/min. The return of spontaneous circulation (ROSC) was determined by the presence of a sustained sinus rhythm on the ECG. CPR was terminated if ROSC did not occur within 4 min. Upon the return of spontaneous respiration, the inspired gas mixture was gradually adjusted from 100% oxygen to a 1:1 oxygen-nitrogen ratio. After stabilization, independent respiration was achieved, and surgical sites were sutured. The tracheal tube was then removed, and the mice were transferred to a thermal incubator set at 33.0 °C for a 2-h recovery period before being returned to their standard housing. Sham mice underwent identical anesthesia and surgical exposure of the jugular vein, but did not receive CA and CPR.

Fig. 1.

Mouse experimental procedure. (A) Mice received CPR after 9.5 min of CA. FPS (RAGE inhibitor, 10 mg/kg) or an equivalent dose of vehicle (saline) was injected intraperitoneally on the day before surgery, and another dose again one h before CA. Impact of FPS on HMGB1/RAGE pathway modulation, as well as oxidative stress and pyroptosis-related signaling pathways induced by CA/CPR, was assessed 1 day after CA/CPR. The neuroprotective impact of FPS-ZM1 (FPS) was assessed 3 days after CA/CPR. Groups: (1) Sham + vehicle; (2) Sham + FPS; (3) CA/CPR + vehicle; (4) CA/CPR + FPS (n = 13/group). (B) The process of establishing a KCl-induced CA model in mice included the following steps: anesthesia; tracheal intubation; jugular vein catheterization; temperature and electrocardiogram monitoring; drug infusion; and postoperative targeted temperature management (TTM). (C) ECG monitoring was performed throughout the procedure, capturing representative waveform patterns. CA, cardiac arrest; CPR, cardiopulmonary resuscitation; ROSC, return of spontaneous circulation; RR, respiratory recovery; WB, Western Blot; IF, immunofluorescence; ROS, reactive oxygen species; MDA, malondialdehyde; SOD, superoxide dismutase; ELISA, enzyme-linked immunosorbent assay; NDS, neurological deficit score; H&E, Hematoxylin-eosin; FJB, Fluoro-Jade B; TUNEL, terminal deoxynucleotidyl transferase dUTP nick end labeling; KCl, potassium chloride; Adr, adrenaline; ECG, electrocardiogram; HMGB1, high mobility group box 1; RAGE, receptor for advanced glycation end products; TEM, transmission electron microscopy.

Compound FPS (RAGE inhibitor, 10 mg/kg, HY-19370, MedChemExpress LLC, Monmouth Junction, NJ, USA) was administered via i.p. injection 24 h pre-surgery, and again 1 h before the surgical procedure. The dosing protocol for the administration was based on previous research findings [38]. Fig. 1A displays the experimental protocol.

Neurological function was comprehensively assessed using two distinct scoring systems (Supplementary Tables 1,2). First, a 12-point scale [39] was used to assess the mice across six critical domains: responsiveness to stimuli; corneal reflex; respiration; righting reflex; coordination; and movement. Each domain was rated using a scale from 0 to 2, with 0 signifying severe impairment and 2 signifying normal function. Second, a 9-point grading scale [40] was used to evaluate neurological deficits based on motor performance. Mice were tested for their ability to climb a vertical grid, balance on a horizontal bar, and hang from a rope. Each task was scored from 0 to 3, where 0 represented severe disability and 3 represented normal function. Assessments were performed under standardized conditions (quiet room, consistent illumination, and minimal handling). Scoring was conducted independently by two investigators who were blind to group allocation and study hypotheses. When the discrepancy was $\ge$1 point, a third senior investigator with extensive experience in behavioral assessment re-evaluated the mouse, and the final score was determined based on the third rating and consensus.

Abnormal neuronal morphology was evaluated through H&E staining. Mice were anesthetized by induction in an anesthesia box with 5% isoflurane with the inspired gas mixture at a 7:3 oxygen-nitrogen ratio. After the mice were anesthetized, intracardiac perfusion with 0.9% NaCl was performed, followed by fixation in 4% paraformaldehyde (PFA, G1101, Wuhan Servicebio Technology Co., Ltd., Wuhan, Hubei, China). The brains were fixed overnight at 4 °C in 4% PFA, then dehydrated and embedded in paraffin (Wuhan Servicebio Technology Co., Ltd.). Serial coronal sections (4 µm) were sliced, deparaffinized, stained using H&E, and examined under an Olympus microscope (Olympus Corporation, Hachioji-shi, Tokyo, Japan). Image analysis was performed using ImageJ2 (version 2.16.0, Laboratory for Optical and Computational Instrumentation, University of Wisconsin–Madison, Madison, WI, USA).

Nissl staining was used in order to quantify the density of cortical neurons. Brain tissue was extracted from anesthetized mice after decapitation and fixed in 4% PFA. After fixation, the specimens were embedded in paraffin and sectioned coronally at a thickness of 4 µm. The slices were then dewaxed, rehydrated, and stained by immersing them in a 0.2% cresyl violet solution (G1086, Wuhan Servicebio Technology Co., Ltd.) for 10 min. Images of the brain sections were obtained with an Olympus microscope. Nissl body counting was performed using ImageJ software.

The TUNEL assay was used to identify apoptotic neurons, according to the protocol included with the In Situ Cell-Death Detection Kit (PN0017, Wuhan Pinuofei Biotechnology Co., Ltd., Wuhan, Hubei, China). Paraffin-embedded brain-tissue sections were deparaffinized and subsequently exposed to proteinase K (Wuhan Pinuofei Biotechnology Co., Ltd.) to increase cellular membrane permeability. Thereafter, the sections were incubated with the TUNEL reaction solution at room temperature for 1 h to facilitate the detection process. After DAPI (Wuhan Pinuofei Biotechnology Co., Ltd.) staining to label the nuclei, the slides were then coverslipped using an anti-fade mounting medium (Wuhan Pinuofei Biotechnology Co., Ltd.). The tissue sections were examined under an Olympus microscope, and images were captured for quantification of cortical apoptotic cells per slide.

FJB staining was used to evaluate neuronal degeneration and apoptosis. Brain sections were initially immersed in 0.06% potassium permanganate for 10 min, followed by incubation at room temperature for 20 min in 0.004% FJB solution (Merck Millipore, AG310, Merck Millipore, Darmstadt, Germany). After thorough rinsing and drying, the samples were coverslipped and analyzed using an Olympus microscope. The FJB-positive cell count within the cortical region was manually quantified using ImageJ software.

After deparaffinization, rehydration, and antigen retrieval, tissue sections were delineated and then treated with 3% bovine serum albumin (GC305010, Wuhan Servicebio Technology Co., Ltd.). Carefully prepared primary antibodies, sourced from the following suppliers, were then applied to the samples: HMGB1 (GB11103, Wuhan Servicebio Technology Co., Ltd.,1:500); RAGE (TC52809S, Abmart, Shanghai, China, 1:500); Nrf2 (#12721, Cell Signaling Technology, Danvers, MA, USA, 1:500); and caspase-1 (341030, Chengdu Zen Bioscience Co., Ltd., Chengdu, Sichuan, China, 1:500). The tissue slices were kept in a humidified chamber and incubated at 4 °C overnight to ensure adequate antibody binding. Next, the samples were exposed to the corresponding secondary antibody (#4412 and #4408, Cell Signaling Technology, 1:1000) at room temperature for 1 h. Fluorescence intensity was analyzed with ImageJ software. HMGB1/RAGE colocalization analysis was as follows. Briefly, double immunofluorescence images were analyzed using ImageJ. For each group, three mice were included; for each mouse, three sections were analyzed, and one scanning field covering the cortical region per section was quantified. Within predefined cortical regions of interest (ROIs), HMGB1 and RAGE-positive cells were identified using a consistent thresholding strategy, and HMGB1⁺RAGE⁺ double-positive cells were confirmed on individual channels and merged images. Colocalization was quantified as the percentage of HMGB1⁺RAGE⁺double-positive cells among total cells, with total cell number defined as the number of DAPI-positive nuclei within the ROI.

After cardiac perfusion with physiological saline, we immediately performed perfusion fixation using pre-chilled 2.5% glutaraldehyde fixative (G1102, Wuhan Servicebio Technology Co., Ltd.). The target cortical region was rapidly excised and sectioned into approximately 1 mm³ cubes using a sharp blade. These were transferred into EP tubes containing the aforementioned fixative solution for post-fixation at 4 °C for 4–6 h. Ultrastructural alterations in mouse cortical neurons were imaged using a TEM (HT7700, Hitachi High-Tech Corporation, Tokyo, Japan). Neurons were identified by systematic scanning from low to high magnification based on ultrastructural characteristics.

After rapid cryopreservation in liquid nitrogen, brain tissue was embedded in OCT (G6059, Wuhan Servicebio Technology Co., Ltd.) and sliced coronally. To visualize ROS, sections were treated with dihydroethidium stain (D7008, Sigma-Aldrich, St. Louis, MO, USA) for 30 min at room temperature in darkness. Images were acquired with an Olympus microscope, and the fluorescence signals were analyzed and quantified using ImageJ software.

The level of lipid peroxidation was quantified using the MDA assay, and SOD activity was measured as an indicator of cellular capacity to neutralize ROS. Brain cortex samples were carefully dissected, homogenized in saline, and then subjected to centrifugation to obtain the supernatant. The activities of SOD and MDA were then assessed using commercial assay kits (A001-3-2 and A003-1-2, Nanjing Jiancheng Bioengineering Institute, Nanjing, Jiangsu, China) following the guidelines provided by the manufacturer.

To extract proteins, mouse brain cortex samples were extensively homogenized in a lysis buffer (G2002, Wuhan Servicebio Technology Co., Ltd.) using an ultrasonic cell processor (Ningbo Scientz Biotechnology Co., Ltd., Ningbo, Zhejiang, China). The extracted protein samples were combined with 5$\times$ loading buffer, heated at 95 °C for 5 min, and subjected to electrophoresis on 10–12% glycine gels (PG112-113, Epizyme Biotech Co., Ltd., Shanghai, China) for separation. The separated protein bands were transferred onto the polyvinylidene fluoride membrane (IPVH00010, Merck Millipore, Darmstadt, Germany). Membranes were blocked using either a rapid blocking buffer (PS108P, Epizyme Biotech Co., Ltd.) for 20 min or 5% skimmed milk powder in PBS for 1 h. Then they were exposed to the appropriate primary antibody and incubated overnight at 4 °C: HMGB1 (GB11103, Wuhan Servicebio Technology Co., Ltd., 1:1000); RAGE (TC52809S, Abmart, 1:1000); and caspase-1 (341030, Chengdu Zen Bioscience Co., Ltd., 1:500); gasdermin D ( ER1907-37; HUABIO, Hangzhou, Zhejiang, China, 1:1000); Nrf2 (#12721, Cell Signaling Technology, 1:500); or HO-1 (ab68477, Abcam, Cambridge, UK, 1:500). After washing, the membranes were exposed to the respective secondary antibodies (#7074 and #7076, Cell Signaling Technology, 1:10,000) and maintained at room temperature for 1 h. To facilitate the detection of proteins with varying molecular weights, membranes were horizontally sectioned as needed. After three additional washes, protein signals were detected using an enhanced chemiluminescence (ECL) kit (HY-K1005, MedChemExpress LLC) and imaged with a Bio-Rad imaging system (Bio-Rad Laboratories, Inc., Hercules, California, USA). Protein expression levels were analyzed semi-quantitatively by ImageJ software.

Brain samples and peripheral blood were harvested in sterile tubes and stored at –80 °C. Levels of cytokine (IL-1$\beta$, IL-18, HMGB1, and RAGE) were quantified with the use of ELISA kits: Mouse HMGB1 (MU30043, Wuhan Bioswamp Biotechnology Co., Ltd., Wuhan, Hubei, China); RAGE (MU31640, Wuhan Bioswamp Biotechnology Co., Ltd.); IL-1$\beta$ (JL18442, Shanghai Jianglai Industrial Co., Ltd., Shanghai, China); and IL-18 (JL20253, Shanghai Jianglai Industrial Co., Ltd.). Samples obtained after the previously described procedure were processed, with absorbance measured at 450 nm for all wells within each subgroup. The concentrations of cytokines were then determined by extrapolating values from the established standard curve.

All collected experimental results were subjected to statistical analysis using GraphPad Prism (version 9.0.0.121, GraphPad Software, LLC, Boston, MA, USA), and the findings are presented as mean $\pm$ standard error of the mean (SEM). For comparisons between two groups, an independent two-tailed Student’s t-test was used. For comparisons among multiple groups, one-way ANOVA was performed. When the ANOVA indicated an overall significant difference, Tukey’s multiple comparisons test was applied for post-hoc pairwise comparisons across groups. Statistical significance was set at p $\le$ 0.05.

To investigate alterations in RAGE and HMGB1 expression in mice after CA/CPR, cortical brain tissue and serum samples were analyzed using immunofluorescence and ELISA 24 h post-CA. Immunofluorescence results revealed a marked increase in RAGE and HMGB1 expression within the cerebral cortex of CA/CPR mice (Fig. 2A–D). Subsequent ELISA assays further confirmed that serum HMGB1 levels were significantly higher in the CA/CPR group than in the sham group (Fig. 2E,F). Similarly, serum RAGE levels were markedly higher in CA mice than in the sham group (Fig. 2G,H). These results indicated that both RAGE and HMGB1 were substantially upregulated during PCABI, suggesting activation of the HMGB1/RAGE axis and its potential involvement in the pathogenesis of PCABI.

Fig. 2.

The HMGB1/RAGE axis is activated by CA/CPR. (A) Typical immunofluorescence pictures of HMGB1. (B) Statistics of the proportion of HMGB1-positive cells across different groups. (C) Typical immunofluorescence pictures of RAGE. (D) Statistics of the proportion of RAGE-positive cells across different groups. (E,F) ELISA quantification of HMGB1 levels in serum and cortex in different groups. (G,H) ELISA quantification of RAGE levels in serum and cortex in different groups. Data are depicted as mean $\pm$ SEM. Scale bar = 50 µm (n = 3/group). ****p 0.0001, ***p 0.001, **p 0.01. SEM, standard error of the mean.

Although FPS is considered a specific RAGE inhibitor, the mechanistic basis for its suppression of RAGE activity in mouse brain tissue remains unclear. In the present study, immunofluorescence analysis revealed that, under physiological conditions, HMGB1 and RAGE exhibited only partial overlap in the mouse cerebral cortex, indicating a relatively low degree of colocalization (Fig. 3A). The CA/CPR + FPS group showed markedly less HMGB1/RAGE colocalization than did the CA/CPR + vehicle group (p = 0.0009, Fig. 3B). Consistent with the above observations, Western Blot analysis (Fig. 3C, the original Western blot image can be seen in the Supplementary Material-Western Blot) demonstrated significantly lower HMGB1 protein levels in cortex of the CA/CPR + FPS-pretreated group than in the CA/CPR + vehicle group (p = 0.0299, Fig. 3D), accompanied by a substantial decrease in RAGE protein levels (Fig. 3E). Furthermore, ELISA analysis revealed that HMGB1 and RAGE levels in both serum and cortex were significantly lower in the FPS–pretreated group than in the CA/CPR + vehicle group (Fig. 3F–I). Taken together, these findings suggested that FPS pretreatment exerted its effects, at least in part, by inhibiting CA/CPR-induced activation of the HMGB1/RAGE signaling pathway.

Fig. 3.

FPS pretreatment prevented activation of the HMGB1/RAGE axis induced by CA/CPR. (A) Typical pictures of HMGB1 and RAGE double immunofluorescence staining. (B) Statistics of the proportion of HMGB1 and RAGE double-positive cells across different groups. (C) Typical protein immunoblotting bands of HMGB1 and RAGE. (D,E) Relative protein expression semiquantitative analysis of HMGB1 and RAGE across different groups. (F,G) ELISA quantification of HMGB1 levels in serum and cortex in different groups. (H,I) ELISA quantification of RAGE levels in serum and cortex in different groups. Data are depicted as mean $\pm$ SEM. Scale bar = 50 µm (n = 3/group). ****p 0.0001, ***p 0.001, **p 0.01, *p 0.05.

Recent research has demonstrated that the HMGB1/RAGE axis is critically involved in mediating pyroptosis in neonatal hypoxic-ischemic brain injury and Kawasaki disease. Pretreatment with FPS has been shown to inhibit pyroptosis under these pathological conditions [27, 28]. Building on these findings, the present study investigated whether FPS pretreatment could similarly attenuate CA/CPR-induced pyroptosis. Immunofluorescence (Fig. 4A–D) analysis revealed that, after CA/CPR, levels of activated caspase-1 and the proportion of TUNEL-positive cells in the mouse cerebral cortex were significantly increased. By contrast, FPS pretreatment markedly reduced activated caspase-1 levels and the number of TUNEL-positive cells. Notably, we used TUNEL as an indicator of DNA fragmentation and overall cell-death burden. To further explore pyroptosis, TEM revealed prominent ultrastructural cortical neuronal damage after CA/CPR, including membrane disruption, nuclear condensation, and severe organelle injury, which were alleviated by FPS pretreatment (Fig. 4E). As expected, ELISA (Fig. 4F,G) assays demonstrated that the concentrations of the cytokines IL-1$\beta$ and IL-18 in the serum and brain tissue were significantly higher after CA/CPR. However, FPS pretreatment significantly blunted this increase. Protein analysis (Fig. 4H–L, the original Western Blot image can be seen in the Supplementary Material-Western Blot) showed that CA/CPR mice expressed higher levels of cleaved caspase-1 p20, GSDMD, and its N-terminal fragment N-GSDMD than did the sham group. It is important to note that FPS pretreatment substantially downregulated these protein levels. These results suggested that FPS pretreatment effectively suppressed CA/CPR-induced pyroptosis.

Fig. 4.

FPS pretreatment attenuated the level of pyroptosis after CA/CPR. (A) Typical pictures of caspase-1 immunofluorescence staining. (B) Quantitative assessment of caspase-1 fluorescence intensity in different groups. (C) Typical pictures of TUNEL immunofluorescence staining. (D) Statistics of the proportion of TUNEL-positive cells in different groups. (E) Typical TEM pictures of cortical tissue (n = 3/group). Scale bar = 2 µm (low magnification) and 500 nm (high magnification). Arrows indicate pyroptotic neurons exhibiting cell membrane rupture (red) or organelle damage (orange). (F,G) ELISA quantification of the inflammatory cytokine IL-1$\beta$ and IL-18 levels in serum and cortex in different groups. (H) Typical protein immunoblotting bands: Pro-caspase-1; cleaved-caspase-1; GSDMD; and N-GSDMD. (I–L) Relative protein expression semiquantitative analysis of pro-caspase-1, cleaved-caspase-1, GSDMD, and N-GSDMD in different groups. Data are depicted as mean $\pm$ SEM. Scale bar = 50 µm (n = 3/group). ****p 0.0001, ***p 0.001, **p 0.01, *p 0.05, ns: not significant. IL-1$\beta$, interleukin-1 beta; GSDMD, gasdermin D.

To further elucidate the molecular mechanisms through which FPS mitigates PCABI, we measured oxidative stress levels induced by CA/CPR. As illustrated in Fig. 5A,B, immunofluorescence analysis demonstrated that CA/CPR significantly elevated ROS production in the cerebral cortex; whereas this effect was diminished by pretreatment with FPS. In addition, cortical SOD activity was markedly higher in the FPS-pretreated group than the CA/CPR + vehicle group (p = 0.0495, Fig. 5C), and accompanied by a substantial reduction in MDA levels (p = 0.0001, Fig. 5D). Collectively, these findings indicated that FPS pretreatment significantly alleviated CA/CPR-induced oxidative stress.

Fig. 5.

FPS pretreatment reduced oxidative stress after CA/CPR. (A) Typical pictures of ROS immunofluorescence staining. (B) Quantitative assessment of ROS fluorescence intensity in different groups. (C,D) Evaluation of SOD activity and MDA levels. Data are depicted as mean $\pm$ SEM. Scale bar = 50 µm (n = 3/group). ****p 0.0001, ***p 0.001, **p 0.01, *p 0.05.

Previous research demonstrated that FPS possesses antioxidant properties and can upregulate Nrf2 and HO-1 levels in a dose-dependent fashion, thereby providing protective effects [35]. Based on those findings, we hypothesized that FPS pretreatment attenuates CA/CPR-induced oxidative injury by modulating the Nrf2-HO1 axis. To examine changes in Nrf2 and its downstream antioxidant signaling, we performed immunofluorescence staining (Fig. 6A,B) and Western Blot analysis (Fig. 6C–E, the original Western Blot image can be seen in the Supplementary Material-Western Blot). Immunofluorescence examination revealed that Nrf2 expression exhibited a slight elevation after CA/CPR. This suggested that CA/CPR manipulation stimulated Nrf2 expression to a modest increase, although it did not produce a sufficient protective effect. Notably, FPS pretreatment markedly increased the overall Nrf2 protein signal after CA/CPR. Consistent with these findings, Western Blot analysis further confirmed that FPS pretreatment significantly upregulated total Nrf2 protein levels as well as the key downstream effector HO-1 after CA/CPR. Collectively, these results indicated that FPS pretreatment strongly enhanced Nrf2 protein expression after CA/CPR, accompanied by upregulation of the antioxidant protein HO-1. These results are consistent with the effect of FPS pretreatment in reducing oxidative stress, suggesting that the Nrf2/HO-1 signaling axis may be involved in its protective effect against oxidative stress after CA/CPR.

Fig. 6.

FPS pretreatment upregulated Nrf2 and HO-1 expression after CA/CPR. (A) Typical pictures of Nrf2 immunofluorescence staining. (B) Statistics of the proportion of Nrf2-positive cells in different groups. (C) Typical protein immunoblotting bands of Nrf2 and HO-1. (D,E) Relative protein expression semiquantitative analysis of Nrf2 and HO-1 in different groups. Data are depicted as mean $\pm$ SEM. Scale bar = 50 µm (n = 3/group). ****p 0.0001, **p 0.01, *p 0.05.

To further characterize the neuropathological alterations in the brain after CA/CPR, brain samples were harvested on postoperative day 3 and subjected to H&E, Nissl, and FJB staining. The H&E (Fig. 7A,B) results showed that neurons in the sham group displayed clear, intact morphology. In contrast, the CA/CPR + vehicle group neurons were atrophied, with a reduced cytoplasmic volume, and nuclei and nucleoli were blurred. In the CA/CPR + FPS group, the majority of neurons retained relatively normal cytoplasmic and nuclear architecture, and only a small proportion of cells showed mild shrinkage, characterized by decreased cytoplasm and atypical nuclei. Nissl staining further demonstrated pronounced neuronal damage in the CA/CPR + vehicle group, characterized by pyknosis, karyorrhexis, and karyolysis, a notable loss of Nissl bodies, and a reduction in Nissl-positive neuronal counts compared with the sham group (p 0.0001, Fig. 7C,D). Notably, FPS pretreatment resulted in a markedly higher count of Nissl-positive neurons than in the CA/CPR + vehicle group (p = 0.0065, Fig. 7C,D). FJB staining revealed a lower number of FJB-positive cells in the FPS pretreatment group, indicating less neuronal degeneration and better neuronal survival than in the CA/CPR + vehicle group (p 0.0001, Fig. 7E,F). Taken together, these data showed that FPS alleviated PCABI and conferred a neuroprotective effect.

Fig. 7.

FPS pretreatment reduced neuron damage after CA/CPR. The mouse brain cortex was stained with H&E, Nissl, and FJB 3 days after CA/CPR to assess neuropathological damage. (A) Typical pictures of H&E staining. (B) Number of dead cortical neurons. (C) Typical pictures of Nissl staining. (D) Quantification of Nissl-positive cells. (E) Typical pictures of FJB staining. (F) Quantification of FJB-positive cells. Data are depicted as mean $\pm$ SEM. Scale bar = 50 µm (n = 3/group). ****p 0.0001, **p 0.01.

Neurological outcomes after CA/CPR were assessed using 9- and 12-point neurological-function scoring systems. Body weight changes were monitored on days 1 and 3 after resuscitation, and survival rates were recorded daily. Survival analysis (Fig. 8A) showed that the FPS pretreatment group showed markedly better survival rates than did the CA/CPR + vehicle group, with higher survival rates at 1 day (69.23% vs. 46.15%), 2 days (69.23% vs. 30.77%), and 3 days (53.85% vs. 23.08%). Analysis of neurological-function scores indicated pronounced functional impairment in the CA/CPR group relative to the sham group at both time points. On the 9-point scale (Fig. 8B), the mean scores in the CA/CPR group were 1.00 on day 1 and 0.33 on day 3, whereas on the 12-point scale (Fig. 8C), the corresponding scores were 6.17 on day 1 and 3.33 on day 3. In contrast, mice receiving FPS pretreatment exhibited significantly better neurological performance than did the CA/CPR + vehicle group: the 9-point scores increased to 2.22 on day 1 (p = 0.0079) and 1.86 on day 3 (p = 0.0128), and the 12-point scores rose to 7.56 on day 1 (p = 0.0005) and 6.43 on day 3 (p 0.0001). Furthermore, analysis of weight changes (Fig. 8D) revealed that mice in the FPS pretreatment group experienced significantly less weight loss than did the CA/CPR + vehicle group on both day 1 (1.46 g vs. 2.78 g, p = 0.018) and day 3 (2.53 g vs. 4.31 g, p = 0.0109). In summary, these data suggested that FPS improved neurological outcomes in mice after CA/CPR.

Fig. 8.

FPS pretreatment ameliorated neurological function and survival after CA/CPR. (A) Survival percentage of mice was measured 3 days. Survival curves were generated using the Kaplan–Meier method and compared by the log-rank (Mantel–Cox) test. (B,C) 9-point and 12-point scoring systems at 1 and 3 days. (D) Body weight reduction in mice was observed at 1 and 3 days. Data are depicted as mean $\pm$ SEM (n = 13/group). ****p 0.0001, ***p 0.001, **p 0.01, *p 0.05.