Ginkgolic acid (15:1) (purity > 99.9%) was purchased from MedChemExpress (MCE, Shanghai, China). GA was soluble in the dimethyl sulfoxide (DMSO, Solarbio, Beijing, China) for the final concentration of 10 nM for cellular experiments, stored at −80°C. Alternatively, GA was dissolved in 10% DMSO and 90% corn oil for the final concentration of 20 mg/mL for animal experiments.

Male C57BL/6N mice (8 weeks old, 22–25 g) were obtained from Shanghai Slack Laboratory Animal Center [license: SCXK (HU) 2012-0002]. Mice were raised in a particular environment with specific pathogen-free conditions under 12 h light/dark cycle, at 22 ± 2°C and appropriate humidity, and had free access to standard chow and water. All experimental procedures and animal welfare protocols complied with the guidelines of the National Institutes of Health Guide for the Care and Use of Laboratory Animals. The methods were reviewed and approved by the Institutional Animal Ethics Committee of Wenzhou Medical University.

Induction of cecal ligation and puncture (CLP) was performed as described previously (26). In short, mice were anesthetized with 1% pentobarbital sodium (50 mg/kg) (TOPNUO, Taizhou, China) injected intraperitoneally. The abdominal skin of the mice was removed, and laparotomy was performed along the mid-abdominal line to separate the cecum, which was ligated at 0.5 cm distal to the cecum tip with a 4-0 non-absorbable suture. The cecum on the ligated side was penetrated with an 18-gauge needle and squeezed with ophthalmic forceps to expel some cecum contents. Then the cecum and feces were returned to the abdominal cavity. The peritoneum and skin were sutured layer by layer and then disinfected with 75% alcohol. Mice were injected subcutaneously with 1 mL of sterile pre-warmed saline behind the neck for fluid resuscitation, and then put back into their cages until recovery.

Mice were randomly divided into five groups: (1) Sham group, (2) CLP group, (3) CLP + GA (10 mg/kg) group, (4) CLP + GA (20 mg/kg) group, (5) CLP + GA (40 mg/kg) group. A single dose of GA (10, 20, or 40 mg/kg body weight) or vehicle (corn oil) was given to mice by gavage 2 h preceding CLP operation. The Sham group was operated as the CLP group, but only the cecum was evacuated and retrieved without ligation or puncture. Mice were then anesthetized and executed at 12, 24, and 48 h after the surgery. The serum and tissue samples were collected for subsequent analyses. For survival analysis, 50 mice were randomized into the Sham, CLP, CLP + GA (10 mg/kg), CLP + GA (20 mg/kg), and CLP + GA (40 mg/kg) groups. Mice were monitored hourly, and their survival was analyzed after 72 h.

Mice were sacrificed by carbon dioxide inhalation at 12, 24, and 48 h post-Sham or post-CLP operations. Lung, liver and kidney tissue samples were fixed in 4% paraformaldehyde (Solarbio, Beijing, China) for 48 h at room temperature. After dehydration, clearing, waxing and embedding, the tissue samples were cut into 4 μm-thick sections. According to the manufacturer’s instructions, the paraffin sections were deparaffinized with xylene and stained with hematoxylin and eosin (H&E) (Solarbio, Beijing, China) and followed by observation under an optical microscope.

Lung injury scores were performed as previously described in a double-blinded manner according to neutrophil migration infiltration, hemorrhage and alveolar cavity congestion, and interstitial lung edema (27). The sum of individual scores graded from 0 (normal) to 3 (severe), ranging from 0 to 12. Liver injury score criteria: hepatocyte degeneration or necrosis, hemorrhage or congestion of live tissue, and infiltration of interstitial inflammatory cells. The severity of liver injury was classified as 0, no injury; 1, mild injury; 2, moderate injury; and 3, severe injury (28). The total liver score ranged from 0 to 4. Kidney dysfunction was assessed by measuring tubular morphology. Renal tubular injury was defined as tubular epithelial swelling, brush border detachment, vacuolization, cell necrosis, tubular formation and desquamation. The degree of renal injury was determined depending on the percentage of damaged tubules with the following criteria (29): 0, no injury; 1, < 10% injury; 2, 10–25% injury; 3, 25–50% injury; 4, 50–75% injury; and 5, > 75% injury. The sum of kidney scores ranged from 0 to 5.

The following operations were performed under aseptic conditions. The same volume of whole blood was taken into 100 μL of sodium heparin. The peritoneal lavage fluids (PLFs) were obtained by flushing the peritoneal cavity with 5 mL saline. The lung, liver, and kidney homogenates were derived by grinding in saline. Serial dilutions of blood, PLFs, and tissue homogenates of the lung, liver, and kidney were plated on Luria-Bertani agar medium at 37°C for 24 h. Colony-forming unit (CFU) counts were then counted and calculated.

Peritoneal macrophages (PMs) were isolated from mice as previously described (30). Briefly, a 1.5% soluble starch solution (3 mL/mice, Solarbio, Beijing, China) was injected intraperitoneally into the mice. After 48–72 h, the mice were lavaged intraperitoneally with 10 mL of pre-chilled Roswell Park Memorial Institute (RPMI) 1640 medium (Gibco, NY, USA), divided into three instillations of 3 mL. The recovered medium was centrifugated at 500 g for 5 min at 4°C. The suspension with PMs was obtained by removing the red blood cells with RBC lysis buffer (Solarbio, Beijing, China), and washed twice. The precipitate was resuspended in 3 mL of RPMI 1640 medium supplemented with 10% fetal bovine serum (FBS, Gibco, NY, USA), 100 U/mL penicillin, and 100 mg/mL streptomycin. The cells were plated in 6-well plated (2 × 10⁶ cells/well), and cultured at 37°C under a 5% CO₂ humidified atmosphere for 2 h. The supernatant was removed, and the adherent cells were the PMs.

Raw 264.7 cells were purchased from the cell Bank of Shanghai Institutes for Biological Science (Shanghai, China) and cultured in high-glucose Dulbecco’s Modified Eagle’s Medium (DMEM, Gibco, USA) supplemented with 10% fetal bovine serum (FBS, Gibco, USA) at 37°C under a 5% CO₂ humidified atmosphere. Cells were incubated in six-well plates at a density of 2 × 10⁶ cells/mL and cultured to 80% area. Cells were randomly divided into four groups: Ctrl, GA, LPS, and LPS + GA groups. Then the cells were pretreated with GA (10 nM) for 30 min before lipopolysaccharide (LPS) (1 μg/mL) stimulation for 24 h to establish an inflammation model. The Ctrl group was given the same volume of vehicle (DMOS). The cells and supernatant were collected for the following analysis.

Total RNA in PMs and RAW 264.7 cells was extracted using RNA simple total RNA kit (Tiangen, Beijing, China) according to the manufacturer’s instructions. cDNA was synthesized from quantified RNA by using ReverAid RT kit (Thermo Fisher, MA, USA). The target mRNAs were quantified by Q-PCR using SYBR Green Master Mix (Bio-rad, CA, USA) on a BioRad CFX96 (Bio-rad, CA, USA). The amplification curves and lysis curves were confirmed to be normal in the results after the reaction, and the relative quantitative expression of the target gene was calculated with 2^–ΔΔCT values using GAPDH as the internal reference gene. Mouse primer sequences are shown in Table 1.

Mice were anesthetized, and blood was obtained aseptically. After agglutinating for 60 min at room temperature, serum was collected by centrifugation at 1,000 g for 10 min at 4°C. The serum was dispensed and stored at −80°C for storage. The culture supernatants of RAW 264.7 cell were collected after stimulation with 1 μg/mL LPS for 24 h. The levels of Interleukin (IL)-1β, IL-6, and tumor necrosis factors (TNF)-α in the serum and the culture supernatants were detected using ELISA kits (Multisciences, Hangzhou, China).

RAW 264.7 cells were plated in a 6-well plate (10⁴ cells/well) containing cell crawlers, shaken and walled, and then grouped for treatment. After 6 h of cell culture, cells were fixed with 4% paraformaldehyde for 15 min and permeabilized with PBS containing 0.3% TritonX-100 for 20 min. PBS blocking solution containing 1% BSA was added to block the cells for 1 h. The primary antibody, rabbit anti-p-NF-κB p65 antibody (1:1000 dilution, CST, USA), was incubated overnight at 4°C. Washed twice, cells were incubated with the secondary antibody (1:1000 dilution, Abcam, Cambridge, UK) in the dark at room temperature for 1 h. The nuclei were stained with 4’,6-diamidino-2-phenylindole (DAPI) for 5 min. The samples were exposed to anti-fluorescence quenching sealing solution (Solarbio, Beijing, China). A laser confocal microscope was used to observe under the corresponding excitation light, and images were acquired.

Apoptosis of cells was detected with the fluorescent probe FITC-labeled Annexin V, Annexin V-FITC. Every 10⁶ cells were collected and washed twice with pro-chilled PBS, resuspended in 500 μL Binding Buffer. The cells were incubated with 5 μL Annexin V- FITC and 10 μL PI for 5 min at room temperature for flow cytometry analysis, protected from light. The results obtained were analyzed by FlowJo software (Ashland, USA).

The cells were collected and washed with PBS, lysed in radioimmunoprecipitation (RIPA) lysis buffer (Solarbio, Beijing, China) with 1% phenylmethylsulfonyl fluoride (PMSF) (Solarbio, Beijing, China) and 1% protease and phosphatase inhibitor (Solarbio, Beijing, China) on ice for 30 min. The supernatant was obtained by centrifuging at 13,000 g for 20 min at 4°C, followed by quantification of the protein concentration with a bicinchoninic acid (BCA) kit (Thermo Fisher, MA, USA). Then the protein was mixed with 5 × protein loading buffer (Thermo Fisher, MA, USA) and denatured at 100°C for 10 min. Equal amounts of denatured proteins (30 μg) were separated on 10% sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis (PAGE) (Beyotime, Shanghai, China) by western blot and transferred onto 0.45 μm polyvinylidene fluoride (PVDF) membranes (Millipore, CA, USA) for immunoblotting; Next, the membranes were incubated with 5% fat-free dry milk in Tris-buffered saline containing Tween-20 (TBST) (Solarbio, Beijing, China) for 2 h at room temperature, followed by specific primary antibodies (anti-SUMO1, anti-SUMO2/3, anti-BCL-2, anti-BAX, anti-Cleaved caspase 3, anti-NF-κB p65, anti-phospho-NF-κB p65, anti-β-actin, anti-GAPDH, both 1:1000 dilution) (Abcam, Cambridge, UK) at 4°C overnight. The membranes were washed three times and incubated with the horseradish peroxidase (HRP)-conjugated secondary antibody (1:5000 dilution) (Beyotime, Shanghai, China) for 60 min at room temperature. Finally, protein bands were visualized by using the enhanced chemiluminescence (ECL) kit (Thermo Fisher, MA, USA) and performed with the ChemiDic TM XRS + Imaging System (Hercules, CA, USA). The density of each band was analyzed using Image J software (NIH Image, Bethesda, MD).

Data were analyzed by using GraphPad Prism 9.0 (GraphPad Software, La Jolla, CA, USA). The experiment was repeated at least three times. One-way ANOVA followed by Least Significant Difference (LSD) multiple comparison tests was performed for assessing statistical significance between multiple groups. P < 0.05 were considered statistically significant.

Inflammatory response is a highly prevalent symptom of sepsis (31). To determine whether GA affects the inflammatory response, mRNA levels of pro-inflammatory factors IL-1β, IL-6, TNF-α, inducible nitric oxide synthase (iNOS), and anti-inflammatory factor IL-10 were measured in PMs of septic mice and RAW 264.7 cells. In the in vivo experiment, mRNA levels of IL-1β, IL-6, TNF-α and iNOS were increased in PMs of mice 24 h after CLP compared to the Sham group, and mRNA levels of IL-10 were also mildly increased (Figures 1A–E). GA treatment further increased the mRNA levels of inflammatory markers, while mRNA levels of IL-10 decreased (Figures 1A–E). Consistent with the in vivo experiments, in the LPS-stimulated RAW 264.7 cell inflammation model, GA promoted the mRNA levels of IL-1β, IL-6, TNF-α and iNOS and inhibited the mRNA levels of IL-10 mRNA (Figures 1I–M). In addition, we measured the protein levels of IL-1β, IL-6, and TNF-α in blood samples from septic mice and supernatants of RAW 264.7 cells. The results also showed that GA increased IL-1β, IL-6, and TNF-α levels in vitro (Figures 1F–H) and in vivo (Figures 1N–P).

Relative to that in the sham group, the 72-h survival rate was markedly reduced in the CLP group, but was further decreased in the GA-treated mice (Figure 2A). Next, the effect of GA on the lung, liver and kidney was assessed by H&E staining during 12, 24, and 48 h of CLP-induced sepsis. Compared to the CLP group, the CLP + GA group exhibited more severe neutrophil migration and infiltration, hemorrhage, alveolar cavity congestion, and interstitial lung edema in a dose-dependent manner (Figures 2B, C). Meanwhile, the liver cell borders were blunted received with GA. The markedly severe edema, inflammatory cell infiltration and hepatocyte necrosis were observed (Figures 2D, E). Further, the mice with GA treatment exhibited more severe renal tubular damage (Figures 2F, G). These indicate that GA exacerbates pathological organ changes in septic mice.

To explore the bacterial burdens of mice, blood and PLFs were aseptically collected for 12, 24, and 48 h after surgery, followed by incubation for 24 h. The results showed that the bacterial levels in the blood and PLFs were significantly increased in the CLP group, which was further increased in the mice given GA in a dose-dependent effect (Figures 3A, B). In addition, we also examined the bacterial load in the lungs, liver, and kidneys, and the results were similar to those of blood and PLFs (Figures 3C–E).

The levels of SUMO1 and SUMO2/3 in macrophage during sepsis and the inhibitory effect of GA were unclear, which were detected by western blotting. Compared with the Sham group, the protein levels of conjugate-SUMO1 and conjugate-SUMO2/3 were reduced in PMs of CLP-operated mice, and the protein levels of free SUMO1 and free SUMO2/3 were increased, which showed that SUMOylation was attenuated. Treating with GA, the level of SUMOylation was significantly lower and showed a dose-dependent effect (Figures 4A–D). In addition, in the LPS-stimulated RAW 264.7 cell inflammation model, the levels of SUMO1 conjugates were also suppressed, which were exacerbated by GA. However, the levels of free SUMO1 and free SUMO2/3 were not increased by GA administration (Figures 4E–H).

We used immunofluorescence technique to detect the activation of NF-κB p65 into the nucleus of RAW 264.7 cells. As shown in Figure 5A, the nuclear translocation of phosphorylated NF-κB p65 (p-NF-κB p65) was increased after LPS treatment, and the fluorescence intensity of p-NF-κB p65 in the nucleus was increased. Next, we treated the cells with GA. The results showed that the nuclear translocation of p-NF-κB p65 was more pronounced. In addition, we used western blot to detect the protein levels of NF-κB p65, and obtained the same results as for immunofluorescence (Figures 5B, C). Total intracellular p-NF-κB p65 protein content increased after LPS stimulation, and this phenomenon was further aggravated by treatment with GA. Those suggested that GA in the LPS-stimulated model in vitro promotes NF-κB p65 phosphorylation and nuclear translocation.

We verified the effect of GA on apoptosis in RAW 264.7 cells in vitro. The results showed that the LPS group had higher proportion of apoptotic cells compared to Ctrl group. Meanwhile, GA-treated group exhibited more apoptotic cells (Figures 6A, B). Furthermore, the levels of BCL-2, BAX, and Cleaved-Caspase 3 in RAW 264.7 cells were detected by western blot. After LPS stimulation for 24 h, the level of Cleaved-Caspase 3 and BAX were elevated, the level of BCL-2 was decreased (Figures 6C–F). This phenomenon was further aggravated by administration of GA. Moreover, we verified the effect of GA on apoptosis in PMs in vivo. The results showed that apoptosis was increased in the CLP group compared to Sham group. Meanwhile, GA-treated group exhibited more severe apoptosis (Figures 6G–J).