Animals

One hundred and eighty C57BL/6 male mice, aged 10–12 weeks, were purchased from the Animal Center of the Nanjing Drum Tower Hospital, Affiliated Hospital of Medical School, Nanjing University, China. All animal procedures were performed according to the Guidelines for the Care and Use of Laboratory Animals from the National Institutes of Health and with approval of the Committee of the Nanjing Drum Tower Hospital, Affiliated Hospital of Medical School, Nanjing University. Mice were housed in groups of four individuals per cage with a 12 h light/dark cycle (lights on at 8 a.m. and off at 8 p.m.) at a room temperature of 22 ± 1 °C and a humidity of 50–60% and with ad libitum access to standard food and water. All mice had acclimated to the environment for at least one week prior to experiments.

Cecal ligation and puncture (CLP) model

CLP was performed after anesthesia using 2% sodium pentobarbital (40 mg/kg, Sigma, St. Louis, MO, USA) intraperitoneally (i.p.) as described in our previous studies [10]. Briefly, after intraperitoneal anesthesia, a midline incision (1.5 cm) was made in the lower abdomen. The caecum was carefully isolated under aseptic conditions and ligated using 4.0 silk suture below the ileocecal junction, approximately 1 cm from the distal end. Subsequently, the caecum was perforated with a sterile 22-gauge needle, and fecal contents were extruded through the puncture pore by gently squeezing the caecum. Finally, the intestinal tract was returned to the peritoneal cavity, and the abdomen was sutured with 4.0 silk sutures. In sham operative mice, the caecum was exposed in the same way as in CLP, but it was neither ligated nor punctured. Mice were immediately revived with a subcutaneous injection of saline solution (30 ml/kg) after surgery and returned to their cages.

Experimental design

In experiment 1, mice were randomly assigned to the following groups: sham + vehicle (n = 10), sham + NSA (n = 10), sham + Mdivi-1 (n = 10), CLP + vehicle (n = 20), CLP + NSA (n = 20), and CLP + Mdivi-1 (n = 20). Surviving mice underwent behavioral tests and hippocampal tissues were collected for histological analysis. The experimental protocol is presented in Fig. 1A.

Fig. 1

NSA or Mdivi-1 improved learning and memory in the septic mice after CLP. A Schematic timeline of the experimental procedure. B Survival rate within 10 days (n = 10 mice/sham subgroup; n = 20 mice/CLP subgroup). C Total ambulatory distance in open field tests. D Time spent in the center in open field tests. E Discrimination ratio in the novel object recognition tests. F Spontaneous alterations in Y maze tests. Data are presented as the mean ± SEM (n = 10–14 mice/group). ***P < 0.001 versus the indicated groups

In experiment 2, mice were randomly assigned to sham + vehicle (n = 8), CLP + vehicle (n = 16), CLP + NSA (n = 16), and CLP + Mdivi-1 (n = 16). Surviving mice underwent biochemical assays on postoperative day 7. In addition, 6 mice were used for co-immunoprecipitation.

In experiment 3, mice were randomly assigned to sham + vehicle (n = 4), CLP + vehicle (n = 8), CLP + NSA (n = 8), and CLP + Mdivi-1 (n = 8). Surviving mice underwent in vivo electrophysiology. The experimental protocol is presented in Fig. 7A. The number of animals used in each experiment is based on our previous studies [7, 10] and relevant researches [11, 30, 31] in the field.

Mice in the pharmacological intervention groups were administered either vehicle 1% dimethyl sulphoxide (DMSO), NSA (5 mg/kg, no. HY-100,573; MedChemExpress, Monmouth Junction, NJ, USA) [27] or Mdivi-1 (10 mg/kg, no. T1907; Topscience Co., Ltd., Shanghai, China) by intraperitoneal injection 1 h before surgery and then once daily for the subsequent 6 consecutive days [32]. Both compounds are capable of crossing the BBB [32, 33].

Open field test

The open field (OF) test was carried out in a white opaque plastic chamber (40 × 40 × 40 cm) to assess the locomotor and exploratory activities of mice. Each mouse was gently positioned at the center of the arena for 5 min, and its activity was automatically recorded using a video tracking system (XR-XZ301, Shanghai XinRuan Information Technology Co., Ltd., Shanghai, China). All tests were conducted between 8 a.m. and 5 p.m. To eliminate any olfactory cues, the experimental apparatus was cleaned with a solution of 75% ethanol after each trial.

Novel object recognition

The novel object recognition (NOR) test was performed in a square plastic apparatus (40 × 40 × 40 cm) to evaluate learning and memory ability of mice. Before the test, animals were accustomed to two identical objects in the open box and allowed to explore for 10 min. Subsequently, one of the two objects was replaced by a novel subject with a distinct shape and color. The time spent exploring the familiar and novel subjects was recorded automatically. The discrimination ratio was calculated as the time spent exploring the novel object/ (time spent exploring the novel object + time spent exploring the old object). The apparatus was thoroughly cleaned with 75% ethanol between each test.

Y maze

The Y maze, consisting of three arms at 120° angles labelled A, B, and C, was used to evaluate the spatial working memory of mice. Each animal was placed in the center of the equipment and allowed to explore freely for 8 min throughout the three different arms of the Y maze. The sequence and total number of arms entered were recorded. When the hind paws of the animal had been completely placed in the arm, arm entry was complete. For instance, a sequence of entries to the three arms ABC, ACBABACABA, would generate four “successful” alternations, ACB, CBA, BAC, and CAB. The score of alternation was calculated as the number of triads containing entries into all three arms divided by the maximum possible alternations (the total number of arms entered minus 2) × 100. Re-entry into the same arm was not counted for the analysis. The experimental equipment was thoroughly cleaned with 75% alcohol after each test.

Haematoxylin and eosin (HE) staining

Mice were deeply anaesthetized using 2% sodium pentobarbital (60 mg/kg, i.p.; Sigma, St. Louis, MO, USA). The whole brain was removed and fixed in 4% paraformaldehyde and embedded in paraffin for HE staining using a Hematoxylin-Eosin Stain kit (Nanjing Jiancheng Bioengineering Institue, Nanjing, China). The HE staining procedure included dyeing with haematoxylin for 5 min, decolorization with a 75% hydrochloric acid alcohol solution for 30 s; eosin staining for 5 min, and final decolorization with 90% ethanol for 35 s. Hippocampal neuronal damage was assessed with a standard semiquantitative scale [10, 34]. In brief, 0 indicates no lesions in the hippocampal CA1 region, 1 indicates scattered damaged neurons in the hippocampal CA1 region, 2 indicates moderate numbers of damaged neurons in the hippocampal CA1 region (< 50% neurons affected), 3 indicates severe damage in the hippocampal CA1 region (> 50% of cells damaged), and grade 4 indicates extensive neuronal damage in the hippocampal regions. Evaluation was evaluated by an investigator blinded to each group.

Immunofluorescence staining

Coronal brain sections with a thickness of 10 μm were prepared and placed onto glass slides. Slices were blocked with 1% bovine serum albumin (BSA) for 1 h at room temperature. Subsequently, sections were incubated overnight at 4 °C with rabbit anti-ionized calcium-binding adaptor molecule-1 (Iba-1; 1:200, CY7217, Abways, Shanghai, China); rabbit anti-glial fibrillary acidic protein (GFAP; 1:200, BA0056, Boster, Pleasanton, CA, USA); rabbit anti-Drp1 (1:500, BP51203, Abbkine Scientific Co., Ltd., Wuhan, China); and rabbit anti-GSDMD (1:100, A22602, ABclonal Technology Co., Ltd., Wuhan, China) overnight at 4 °C, followed by a 1 h incubation with donkey anti-rabbit IgG-FITC (1:500, Abbkine Scientific Co., Ltd., Wuhan, China) or goat anti-rabbit IgG-Cy3 (1:600, Abbkine Scientific Co., Ltd., Wuhan, China) at room temperature. For mounting and counterstaining, sections were incubated with DAPI (1:1000, Beyotime Biotechnology, Shanghai, China). Fluorescence images were captured using a confocal microscope (Leica, TCS SP2, Germany), and ImageJ software (National Institutes of Health, Bethesda, MD, USA) was utilized to detect the mean value of the immunofluorescence intensity in each section.

TUNEL staining

Neuronal cell death in the hippocampus was assessed using terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling (TUNEL) assays with One-step TUNEL Apoptosis Assay Kit (red TRITC labeled fluorescence) (KGA7063, Jiangsu KeyGen Biotech Co., Ltd., Nanjing, China) [35]. The sections were first incubated with rabbit anti-NeuN (1:400, 26975-1-AP, Proteintech, Wuhan, China), followed by TUNEL reagents according to the manufacturer’s instructions. The counts of TUNEL-positive neuron in the hippocampus CA1 region were evaluated by an investigator blinded to the intervention conditions.

Transmission electron microscopy

After anesthesia using 2% sodium pentobarbital, mice were perfused intracardially with phosphate-buffered saline (PBS). Hippocampi were removed and immediately put in petri dishes with transmission electron microscopy (TEM) fixative, cut into 1 mm3 tissue blocks, and then preserved at 4 °C. The tissues were postfixed with 1% osmium tetroxide in 0.1 M PBS (pH 7.4) for 2 h at room temperature, followed by rinsing in 0.1 M PBS 3 times for 15 min each. Dehydration was achieved through a graded ethanol series, followed by resin embedding, and cutting into 60–80 nm thin sections using an ultramicrotome. The sections were placed on 150-mesh copper grids with formvar film. Staining involved 2% uranium acetate saturated alcohol solution for 8 min, followed by rinsing in 70% ethanol and ultrapure water. Lead citrate (2.6%) was used for 8 min to avoid CO2 staining, with subsequent rinsing in ultrapure water 3 times. After drying with filter paper, the cuprum grids were placed on a grid board and dried overnight at room temperature. The cuprum grids were observed under TEM, and images were taken.

Golgi-Cox staining

On Day 16 post-surgery, mouse brains were subjected to Golgi-Cox staining using a Golgi Stain Kit (no. GP1152, Servicebio Technology Co., Ltd., Wuhan, China). In brief, following deep anesthesia with 2% sodium pentobarbital, the brains were promptly extracted and rinsed in double-distilled water. They were immersed in impregnation solutions A and B and placed in a dark, undergoing a 14-day treatment shielded from light in a dark, well-ventilated room at room temperature. Following this, the staining solution was replaced with solution C and stored for an additional 7 days. Brain tissue was sectioned into 100 microns using an oscillating microtome and mounted on gelatine slides for staining. After undergoing alcohol dehydration, the tissue sections were subsequently cleared in xylene and coverslipped. All sections were then observed using an optical light microscope (Olympus Microscope System BX51, Olympus Corporation, Tokyo, Japan), scanned using a digital slice scanner, and subsequently analyzed using Fiji software.

Co-immunoprecipitation (Co-IP) and western blotting analysis

The obtained samples were homogenized in ice-cold lysis buffer supplemented with a protease inhibitor cocktail, followed by centrifugation at 12,000 × g for 10 min at 4 °C. The supernatant was collected, and its protein concentration was determined using the Bradford assay.

For Co-IP assay, the supernatant was subjected to anti-GSDMD or IgG control or anti-Drp1 antibody immunoprecipitation for 2 h followed by overnight incubation with protein A/G-agarose beads following the manufacturer’s instruction. The beads were washed for five times with the lysis buffer, and then boiled with the SDS loading buffer for 10 min.

For western blotting, twenty micrograms of protein were separated via SDS‒PAGE and transferred to nitrocellulose membranes. After blocking with 5% nonfat dry milk for 1 h at room temperature, the membranes were subsequently exposed to primary antibodies, including rabbit anti-GSDMD (1:1000, Ab219800, Abcam, Cambridge, UK); rabbit anti-cleaved GSDMD (1:1000, #10137S, Cell Signaling Technology, Danvers, MA, USA); rabbit anti-Drp1 (1:1000, ABP51203, Abbkine Scientific Co., Ltd., Wuhan, China); rabbit anti-interleukin-1β (IL-1β; 1:1000, GB11113, Servicebio Technology Co., Ltd., Wuhan, China); rabbit anti-IL-18 (1:1000, GB115632, Servicebio Technology Co., Ltd., Wuhan, China); rabbit anti-tumor necrosis factor-α (TNF-α; 1:1000, GB11188, Servicebio Technology Co., Ltd., Wuhan, China), and rabbit anti-β-actin (1:1000, GB15003-100, Servicebio Technology Co., Ltd., Wuhan, China). After washing in TBST three times, the membranes were incubated with goat anti-rabbit IgG-horseradish peroxidase-conjugated secondary antibodies (1:3000, Servicebio Technology Co., Ltd., Wuhan, China). The protein bands were detected via enhanced chemiluminescence and quantified utilizing ImageJ software (National Institutes of Health, Bethesda, MD, USA).

ROS detection

Hippocampal tissues were harvested and homogenized in ice-cold lysis buffer supplemented with a protease inhibitor cocktail. Cellular ROS levels were detected using a ROS assay kit (no. S0033S, Beyotime Biotechnology, Shanghai, China) utilizing an oxidation-sensitive fluorescent probe (DCFH-DA), and measured in a spectrofluorometer (excitation 490 nm, emission 520 nm), following established protocols [7]. All measurements were performed in accordance with the manufacturer’s instructions.

In vivo electrophysiology

Mice were anaesthetized using 2% sodium pentobarbital (40 mg/kg, i.p.; Sigma, St. Louis, MO, USA) while maintaining their body temperature with an electric blanket. To fully expose the skull, the muscle, fascia tissue, and periosteum were removed. After balancing the skull bone, three dimensional coordinates of the CA1 region (DP, -2.0 mm; ML, ± 1.5 mm; DV, -1.5 mm) were obtained from the mouse brain atlas. A 3 × 3 mm window was drilled on the skull above the CA1 region, and the cortex was fully exposed. A 4 × 2 electrode array with 8 microwires was used. The electrode was slowly lowered through the window to the predetermined position. After closing the window with paraffin, the electrode was fixed on the skull surface with dental cement. Raw data were recorded in NOR on Day 12 after CLP. All data analyses were performed using NeuroExplorer 5 (Plexon Inc., Dallas, TX). The signals were transmitted into a digital preamplifier and digitized at a sampling rate of 40 kHz. To prevent interference, all other electronic devices were turned off during the recording session. Data processing was performed on a specific workstation running professional electrophysiological analysis software. For the local field potential (LFP) analysis, wide band recordings were downsampled at 1250 Hz. The bands were filtered as follows: delta (0.5–3 Hz), theta (4–7 Hz), alpha (8–13 Hz), beta (14–29 Hz), and gamma (30–80 Hz) oscillations.

Statistical analysis

Statistical analyses were performed using GraphPad Prism 8.0 (GraphPad Software, Inc.). Data are expressed as the mean ± S.E.M. Normal distribution of the data was assessed using the Kolmogorov‒Smirnov test. Survival rate was assessed by the Kaplan‒Meier method and compared among groups using the log-rank test. Differences among groups were analyzed using one-way analysis of variance (ANOVA) for normally distributed data, followed by Tukey’s multiple comparisons post hoc test, or for non-normally distributed data, followed by Fisher’s LSD test. A p value < 0.05 was considered statistically significant. If significant, overall effects were reported in each graph.