Male C57BL/6 mice (age, 8–10 weeks; weight, 20–25 g) were obtained from GemPharmatech Co., Ltd. All mice were housed in a cage under a 12-h dark/light cycle at a room temperature of 23±1°C and 55% humidity; and all mice had free access to food and water. Following the entrustment of the management and operation of the animal facility to China Technology Industry Holdings (Shenzhen) Co., Ltd., all experiments were approved by the Committee on the Animal Research Ethics of China Technology Industry Holdings (Shenzhen) Co., Ltd. (Shenzhen, China; approval no. 202200105) The present study was conducted in accordance with the ARRIVE guidelines (33). A total of 148 C57BL/6 male mice were used in the present study and were randomly divided into the following three experimental groups: Sham group (no ligation and puncture; n=38), cecal ligation and puncture (CLP) group (n=55) and CLP combined with capsaicin group (n=55). Capsaicin (cat. no. HY-10448; MedChemExpress) was dissolved in a solution containing 10% DMSO (v/v), 40% PEG300 (v/v), 5% Tween-80 (v/v) and 45% saline (v/v). A subcutaneous (s.c.) injection of capsaicin at 1 mg/kg was administered once 1 h before CLP surgery or three times at different time points (1 h, 1 day and 2 days before CLP surgery). Mice in the sham group were injected with an equal amount of saline vehicle. The total duration of the animal study was 19 days. Mice were then sacrificed after the Morris water maze (MWM) experiments, and hippocampal tissue was removed for western blotting and immunofluorescence analyses. The health of the mice was monitored daily throughout the experiment, and mice were euthanized if they reached any of the following humane endpoints: Weight loss >20%, persistent refusal to eat or drink for >24 h, and signs of respiratory distress. The mice were confirmed dead by an absence of pain responses, complete cessation of cardiac and respiratory activity, and pupil dilation.

Mice were subjected to CLP according to a previously described method (34). Briefly, following anesthesia of the mice with isoflurane (2% induction, 1.5% maintenance) fur was removed from the abdomen and it was sterilized with 75% alcohol. An incision of ~1.5 cm was then made along the midline of the abdomen, where the cecum was isolated and removed using blunt dissecting forceps, leaving the remaining small and large intestines in the abdominal cavity. The procedure was conducted with the utmost care to avoid any disruption or damage to the mesenteric vessels. Subsequently, the cecum was ligated at a point 50% of its length from the tip, where a 21G needle was used to puncture the cecum at a point midway between the ligation and the tip. The puncturing of blood vessels was avoided. Following the removal of the needle, a minimal quantity of feces was expressed to confirm patency. It was imperative that the cecum be repositioned within the abdominal cavity in such a way that the feces did not spread from the cecum to the edge of the abdominal wall. The abdominal cavity was then closed layer by layer using sutures. To compensate for the loss of body fluids, saline (5 ml/100 g body weight; s.c.) was injected. Additionally, buprenorphine HCl (0.05 mg/kg body weight; s.c.) was administered for postoperative analgesia after completion of the surgery. Thereafter, the animal was placed back into the cage. If a mouse exhibited prolonged lethargy and/or did not eat or drink for 24 consecutive hours, they were evaluated and if it was determined that they would die of sepsis, they were euthanized. Mice were euthanized by deep anesthesia with sodium pentobarbital [120 mg/kg, intraperitoneal (i.p.)] followed by cervical dislocation. The researchers made every effort to minimize the number and suffering of the animals for all experiments. Survival rates were 100% in the sham group and ~53% in the sepsis group, which is consistent with previous studies (16,35).

NORT is a tool used to assess the learning and memory abilities of experimental animals (16). To evaluate the exploration drive of mice towards novel objects, the animals were placed in boxes (50×50×50 cm) within an open field experimental setup. The boxes were made of white polyethylene walls and a white polyethylene floor. Before and after each trial, the boxes and objects were cleaned with ethanol to remove any residual odors or substances; it was important to ensure that no traces of feces or odor remained from the previous mouse. Over the course of 2 consecutive days, the mice were allowed to acclimate to the empty boxes for 10 min each day. Subsequently, the mice were placed in boxes containing two different objects (A and B) for 5 min each. The number of times each object was explored was recorded, representing the sample phase. This process was repeated every 2 h. During this period, the mice were returned to their cages while the objects that had been explored the fewest times were replaced with a different object (C). If a mouse stayed in the corner and did not explore objects, it was removed, 3 mice in CLP group and 4 mice in CLP + capsaicin group were removed. After 2 h, the mice were released into the boxes and allowed to explore freely for 5 min, before the number of times each object was explored was recorded, representing the acquisition phase. The discrimination index was calculated as the number of times a novel object was explored divided by the total number of explorations.

This task was performed according to a previously described method (36). The MWM consisted of a circular pool (diameter, 100 cm; height, 38 cm) and a removable circular platform (diameter, 6 cm); the pool was divided into four quadrants with the circular platform fixed in the third quadrant. Each quadrant had a clear pattern on the wall of the pool and the temperature of the pool was controlled at 21±1°C. The entire process included a training and testing phase, the training phase was further divided into the platform visibility and platform hiding period.

The initial stage was the platform visibility period, which lasted for 1 day. During this stage, the platform was situated at a height of 1 cm above the water surface. The mice were trained four times per day, with each training session lasting for 1 min. The mice were gently guided into the water from the midpoint of the four quadrants facing the pool wall and allowed to explore freely for 1 min. The mice were permitted to locate the platform and remain there for 15 sec. In the event that the mice were unable to find the platform within 1 min, they were assisted in locating it and were allowed to remain on it for 15 sec. At the conclusion of the training period, the mice were dried with paper towels and transferred to the pre-prepared warm and dry cages.

The second phase was the platform-hiding period, which lasted for 4 days. During this period, the water surface was maintained at a height of 1 cm above the platform, with the remaining steps consistent with those employed during the platform visibility period.

The third stage was the test phase, which lasted for 1 day. During this stage, the platform was removed and the mice were placed at the midpoint of the edge of the opposite quadrant of the pool and allowed to explore freely for 1 min. The number of times the mice crossed the platform and the residence time in the quadrant where the platform was located were analyzed using the software (EthoVision XT 17.5; Noldus Information Technology BV).

BV2 cells were obtained from Shenzhen University (Shenzhen, China) and were incubated in DMEM (Gibco; Thermo Fisher Scientific, Inc.) containing 2% FBS (Gibco; Thermo Fisher Scientific, Inc.) and 1% penicillin/streptomycin, (100 U/ml; Gibco; Thermo Fisher Scientific, Inc.). Lipopolysaccharide (LPS) was purchased from Sigma-Aldrich; Merck KGaA. Capsaicin was dissolved in DMSO. LPS was dissolved in PBS. BV2 cells were categorized into three groups: Ctrl, LPS and capsaicin + LPS. The cells were cultured in the presence or absence of LPS (1 µg/ml) for 12 h at 37°C, then further treated with capsaicin (10 µM) for 24 h at 37°C.

For brain section staining, three groups (Sham, CLP + Veh and CLP + capsaicin) of mice (6 mice per group) were anesthetized with sodium pentobarbital (65 mg/kg; i.p.), and perfused transcardially with cold 4% paraformaldehyde. Next, the whole brain was extracted, and post-fixed with 4% formaldehyde overnight at 4°C, before being dehydrated with 30% sucrose solution. The brain tissues were sectioned coronally at 30 µm with a cryostat (Leica Microsystems GmbH). Subsequently, the sections were permeabilized with 0.1% Triton X-100 in PBS and blocked with freshly prepared blocking solution [3% donkey serum (cat. no. SL050; Beijing Solarbio Science & Technology Co., Ltd.) and 0.2% Triton X-100 in PBS] for 1 h at room temperature, followed by incubation with the primary antibody, anti-ionized calcium-binding adapter molecule 1 [Iba1; cat. no. ab5076; research resource identifier (RRID), AB_2224402; 1:500; Abcam], overnight at 4°C. After washing with PBS, secondary antibody conjugated with Alexa Flour 488 (cat. no. A-11055; 1:500; Thermo Fisher Scientific, Inc.) was added for incubation for 2 h at room temperature. Slices were counterstained with DAPI (cat. no. C1005; Beyotime Institute of Biotechnology) for 5 min at room temperature and mounted with Vectashield Antifade mounting medium (cat. no. H1000-10; Vector Laboratories, Inc.). Immunofluorescence images were captured by confocal microscopy (LSM 800; Carl Zeiss AG) and the experiments were repeated three times. The numbers and area of soma on microglia were quantified using ImageJ 1.54g software (National Institutes of Health).

Hippocampal tissues (~30 mg) were homogenized in RIPA buffer (Beyotime Institute of Biotechnology) containing protease and phosphatase inhibitors (Roche Diagnostics GmbH) on ice and stored at −80°C until use. The protein concentrations were quantified using a BCA kit (cat. no. 23225; Thermo Fisher Scientific, Inc.) (16). Subsequently, 30 µg protein was separated on 7.5, 10 and 15% gels by SDS-PAGE and transferred onto PVDF membranes (Merck KGaA). The membranes were blocked with 5% non-fat milk for 1 h at room temperature, before being incubated with primary antibodies overnight at 4°C. After washing with TBS-0.1% Tween-20, the membranes were incubated with the corresponding HRP-conjugated secondary antibodies (anti-rabbit; cat. no. 7074S; 1:1,000; anti-mouse; cat. no. 7076S; 1:1,000; Cell Signaling Technology, Inc.) at room temperature for 1 h. The bands were detected with a chemiluminescent reagent (MilliporeSigma). The experiments were repeated three times. The following primary antibodies were used: TRPV1 (cat. no. NB100-1617; RRID: AB_10002124; 1:1,000; Bio-Techne), LC3B (cat. no. T55992; RRID: AB_2929010; 1:500; Abmart Pharmaceutical Technology Co., Ltd.), p62 (cat. no. 23214; RRID: AB_2798858; 1:500; Cell Signaling Technology, Inc.), Parkin (cat. no. A11172; RRID: AB_2758446; 1:500; ABclonal Biotech Co., Ltd.), PINK1 (cat. no. A7131; RRID: AB_2767686; 1:500; ABclonal Biotech Co., Ltd.), pro-caspase 3 (cat. no. ET1608-64; RRID: AB_3069820; 1:500; HUABIO), cleaved caspase 3 (cat. no. BF0711; RRID: AB_2846190; 1:500; Affinity Biosciences), nucleotide-binding oligomerization domain, leucine rich repeat and CARD domain-containing 4 (NLRC4; cat. no. A7382; RRID: AB_2767914; 1:500; ABclonal Biotech Co., Ltd.), BNIP3 (cat. no. 68091-1-Ig; RRID: AB_2918828; 1:2,000; Proteintech Group, Inc.), NIX (cat. no. 12986-1-AP; RRID: AB_2877901; 1:1,000; Proteintech Group, Inc.), β-actin (cat. no. 4970S; RRID: AB_2223172; 1:3,000, Cell Signaling Technology, Inc.) and β-tubulin (cat. no. 2148S; RRID: AB_823664; 1:3,000, Cell Signaling Technology, Inc.).

The hippocampal tissues were collected and stored at −80°C until needed. The levels of TNF-α were measured using an ELISA kit (cat. no. E-EL-M3063; Elabscience Bionovation Inc.) according to the manufacturer's instructions.

A total of six mice per group were anesthetized with sodium pentobarbital (65 mg/kg; i.p.), and perfused transcardially with 4% paraformaldehyde in PBS resulting in mortality. Brain tissue was extracted and post-fixed with 4% paraformaldehyde for 24 h at 4°C and embedded in paraffin. The brain tissues were then sliced into 8-µm sections. The Hematoxylin-Eosin Stain kit (cat. no. G1076; Wuhan Servicebio Technology Co., Ltd.) was used for H&E staining according to the manufacturer's protocol. All images were captured using an ECLIPSE E100 microscope (Nikon Corporation). Hippocampal neuronal damage was evaluated by a researcher blinded to each group, based on the following observations (37): Grade 0, no damage to any hippocampal subregion; grade 1, scattered neurons were damaged in the CA1 subregion; grade 2, moderate numbers of damaged neurons in the CA1 subregion; grade 3, severe damage to pyramidal cells in the CA1 subregion; and grade 4, extensive cell damage in all hippocampal regions.

Hippocampal tissues were fixed with 2.5% glutaraldehyde for 24 h at 4°C (16). After washing with 0.1 M PBS (pH 7.4) three times for 15 min each, the tissues were post-fixed with 1% OsO4 in 0.1 M PBS (pH 7.4) for 2 h at room temperature, followed by dehydration with ethanol and embedding in Acetone-EMBed 812 resin for penetration for 2–4 h at 37°C. After polymerization, the semi-thin slices were used for positioning, and then cut into 60–80-nm ultra-thin sections. The sections were then stained with 2% uranium acetate and 2.6% lead citrate for 8 min each at room temperature before being observed under TEM (HT7800; Hitachi, Ltd).

BV2 cells were plated into 24-well plates at a density of 1×10⁵ cells/well and were cultured overnight. Subsequently, the cells were infected with LC3-GFP-mCherry lentivirus (MOI, 40; OBiO Technology (Shanghai) Corp., Ltd.) for 72 h at 37°C (38). Following a change of medium, the virus-transfected cells were randomly divided into the Ctrl, LPS and LPS + capsaicin groups. LPS (1 µg/ml) was added to the LPS and LPS + capsaicin groups for 12 h. The Ctrl groups were given the same volume of saline. The cells were incubated with 10 µM capsaicin for 24 h and were then fixed with 4% paraformaldehyde for 30 min at room temperature and stained with DAPI (1 µg/ml) for 15 min at room temperature. Immunofluorescence images were captured using a confocal microscope (LSM 800; Carl Zeiss AG), with yellow fluorescence representing autophagosomes and red fluorescence representing autophagolysosomes. The experiments were repeated three times.

With the exception of the neuronal damage scores, all data are presented as the mean ± standard deviation and analyzed using GraphPad Prism 9.0 software (Dotmatics). Neuronal damage scores are presented as median values with interquartile ranges. Statistical significance was estimated using unpaired Student's t-test or one-way ANOVA followed by Bonferroni post hoc test. To assess neuronal damage score, Kruskal-Wallis test followed by Dunn's multiple comparisons test was used. The Kaplan-Meier method was used for survival analysis, and differences were analyzed using the log-rank (Mantel-Cox) test. P<0.05 was considered to indicate a statistically significant difference.

Initially, TRPV1 expression was measured in the hippocampus of mice with CLP-induced sepsis, and the expression levels of TRPV1 were significantly reduced in mice in the CLP group compared with those in the sham group (Fig. 1A), suggesting that TRPV1 expression in the hippocampus may be associated with CLP-induced SAE. Next, the possible effect of capsaicin, a TRPV1 agonist, on the learning and memory of mice with sepsis was assessed. Specifically, NORT and MWM test were conducted to assess cognitive function. The s.c. administration of capsaicin (1 mg/kg) 1 h before surgery significantly improved the survival rate of mice (Fig. 1B). According to NORT, the recognition index was significantly reduced after CLP surgery compared with that in the sham group; however, a single application of capsaicin did not ameliorate the recognition index compared with that in mice from the CLP group (Fig. 1C). Consistent with the NORT results, MWM experiments demonstrated that there was no significant difference in the time spent in the platform quadrant between the CLP group and the sham group. Moreover, mice in the capsaicin-treated group did not exhibit improvements in learning and memory (Fig. 1D).

The effects of s.c. capsaicin administration for 2 days, 1 day and 1 h before CLP surgery on at 1 mg/kg were then assessed on cognitive and memory deficits by NORT and MWM experiments. Similar with single administration, 3 days of application also ameliorated cognitive deficits in mice with sepsis. Specifically, the results of NORT indicated that mice with sepsis exhibited a diminished recognition index compared with those in the sham group, whereas capsaicin application significantly reversed this in the CLP mouse model (Fig. 2A). Mice in the CLP group also had a longer escape latency compared with those in the sham group during the training phase of MWM (Fig. 2B and D). However, the mice that underwent CLP and were treated with capsaicin displayed a significantly reduced escape latency compared with that in the CLP group. In addition, during the testing phase of MWM, mice in the CLP group spent less time in the target quadrant, whereas capsaicin treatment markedly increased the time in this quadrant (Fig. 2C and D). Notably, there was no significant difference in the swim speed among the three treatment groups (Fig. 2E). Taken together, these results suggested that longer-term capsaicin application may ameliorate learning and memory deficits in mice with sepsis, and this dose (1 mg/kg; 3 times) was subsequently used in the following studies.

Subsequently, the extent of microglial activation was assessed. Immunofluorescence staining revealed a notable activation of microglia in the mouse hippocampal tissue following CLP surgery (Fig. 3A-C). The number of Iba1-positive microglia was significantly increased compared with that in the sham group, whereas this was reversed by capsaicin application. The levels of TNF-α were detected by ELISA and were revealed to be elevated in mice with sepsis compared with those in the sham group mice; however, this was significantly reversed by capsaicin treatment (Fig. 3D). In addition, the protein expression levels of NLRC4 in the hippocampus were detected and were significantly increased in mice in the CLP group compared with those in the sham-operated group; however, capsaicin treatment reversed this increase (Fig. 3E and F).

To detect the effects of capsaicin on sepsis-induced neuronal damage, H&E staining was performed. The cells in the CA1 region of the hippocampus of mice in the sham-operated group exhibited intact and regular morphology, with clearly delineated boundaries (Fig. 4A). However, the cells in the CLP mice exhibited atrophy and profound staining. Administration of capsaicin resulted in a reduction in the number of abnormal neurons and a decrease in neuronal damage scores (Fig. 4A and B). Next, the protein levels of pro-caspase 3 and cleaved caspase 3 were examined. The expression levels of pro-caspase 3 and cleaved caspase 3 were markedly reduced and enhanced in in the CLP group, respectively, when compared with those in the sham group; however these findings were markedly reserved by capsaicin application (Fig. 4C and D).

The expression levels of LC3 and p62 were next assessed by western blotting. The results showed that the ratio of LC3-II/LC3-I and p62 expression were significantly increased in the CLP group compared with those in the sham group, whereas capsaicin treatment markedly enhanced the ratio of LC3-II/LC3-I and decreased p62 expression (Fig. 5A-C). TEM was next used to further investigate the degree of mitophagy, which. showed that the number of autophagic vesicles was increased in the mitochondria of the CLP group, whereas the number of autophagic vesicles was further increased in the CLP + capsaicin group compared with that in the CLP group (Fig. 5D).

The observed increase in both p62 expression, an autophagy substrate, and LC3-II induced by CLP indicates a blockage in autophagic flux. Therefore, autophagic flux was next measured using the LC3-GFP-mCherry fluorescent reporter method in vitro. As shown in Fig. 6, the relative quantities of yellow puncta were significantly increased in the LPS group, indicating that autophagic flux was blocked; however, in the LPS + capsaicin group, the red puncta were markedly increased, suggesting that autophagic flux was activated. Taken together, these findings suggested that capsaicin application may promote mitophagy.

To provide further insights into the molecular mechanisms through which capsaicin can promote mitophagy, the expression levels of proteins associated with this process were assessed. The results revealed that there were no significant changes in the expression levels of PINK1 and Parkin (Fig. 7A-C), whereas BNIP3 and NIX expression were significantly increased after CLP surgery compared with that in the sham group (Fig. 7A, D and E). Notably, capsaicin application further enhanced the expression levels of BNIP3 and NIX induced by CLP. These findings suggested that capsaicin acted through the BNIP3/NIX signaling pathway to activate mitophagy in CLP-induced mice.