The studies involving animal participants were reviewed and approved by the Ethics Committee of Anhui Medical University. Seven-day-old pathogen-free Sprague-Dawley (DS) rats (male, weighing 250–300 g) were sourced from the Experimental Animal Center of Anhui Medical University (license no. SCXK-2018–031). The rats were housed in standard laboratory cages under controlled conditions, maintaining a 12-h light/dark cycle at 22°C ± 2°C, with free access to food and water.

Sixteen rats were randomly divided into four groups (4 rats/group): a saline control group, a group exposed to LPS (#L2630-100 MG; Sigma Aldrich, United States), a group exposed to both LPS and Dex (#MB4091; Meilunbio Co., Ltd.; 25 μg/kg for rats according to previous reports (Sun et al., 2020; Wang et al., 2018)), and a group treated with LPS and Dex and 3-YTP, SIRT3 inhibitor (#HY-108331; Med chem express, Shanghai, China; 50 mg/kg, ip). The rats were anesthetized through an intraperitoneal injection of pentobarbital sodium at a dose of 40 mg/kg (Merck KGaA; cat. No. 1063180500). Tracheal intubation was performed using a micro-atomizer; the control group received 300 µL saline, whereas the LPS group was given 5 mg/kg LPS. Nebulized solution was given at an oxygen flow rate of 4 L per minute for 25 min. Animals were then fed normally for 30 min without intubation following every 6-hour period. Animals were euthanized by decapitation 12 h later. Lung tissues or bronchoalveolar lavage (BAL) were collected for evaluation, followed by a thoracotomy. The right lung tissue samples were isolated for further experiments.

For the morphological evaluation of the lungs, freshly collected samples were first weighed to record their wet mass. The samples were then dried overnight at 75°C to assess their dry mass (Herriges and Morrisey, 2014). Hematoxylin/eosin (HE) staining was performed to examine the histological structure of the lungs and assess the inflammatory response after LPS exposure. At designated time points, lung tissues were harvested and fixed in 4% paraformaldehyde, then dehydrated and embedded in paraffin wax. The samples were subsequently sectioned into 5 µm slices. The sections were analyzed using a fluorescence microscope (Olympus IX50; Olympus Corporation) in conjunction with analyzing software (NIS-Elements F3.2). Airspace volume density was determined by dividing the total airspace area by the total area (Plosa et al., 2014). At each time-point, at least 3 randomly selected images from 5 samples per group were analyzed.

The levels of SOD, CAT, and MDA were measured in both serum and lung tissue samples, using purchased kits (Beyotime, Shanghai, China). Serum concentrations of tumor necrosis factor-α (TNF-α, ab46070; Abcam), interleukin-1β (IL-1β, # BMS630; Thermal Fisher Scientific, Waltham, MA, United States) and IL-10 (#ERA23RB; Thermal Fisher Scientific), IL-8 (#EK720269; AFG Scientific, MA, United States) and IL-12 (#EK720274; AFG Scientific), IL-18 (#RAB1147; Merck, United States) and IL-6 (#RAB0311; Merck, United States), IL-5 (#ab267811; Abcam) and IL-17 (#ab119536; Abcam) were quantified using ELISA kits accordingly. A hydrogen peroxide assay kit (#S0051; Beyotime) was applied to measure the levels of hydrogen peroxide (H2O2) and the lucigenin chemiluminescence method was applied to determine the superoxide anion (O2-) levels in lung tissue samples. Briefly, lung tissue samples were homogenized and supernatant after centrifugation was incubated with 5 μM lucigenin in buffer, and then samples were recorded using a Tecan Infinite 200 by light emission. Addition incubation into the medium of 350 U/ml SOD were used to confirm the specificity for O2-. Protein concentrations were determined using a BCA Protein Concentration Assay Kit (Enhanced; #P0010; Beyotime).

Following treatment, animals were sacrificed, and BALF was obtained by washing the lungs three times with PBS via a tracheal cannula. The samples were centrifuged at 3,000 rpm for 10 min at 4°C, and the pellet of cells was resuspended in PBS for total cell count analysis using a hemocytometer. Cytospin preparations were made for differential cell count, which involved staining with Wright-Giemsa method. The proportions of macrophages, neutrophils, and lymphocytes in the BALF samples were determined by measuring leukocytes under a light microscope.

Serum samples were collected from administrated animals, and then were subjected to ELISA assay, for the antibodies: TNF-alpha, IL-6, IL-5, IL-1beta, IL-18 and IL-17.

This assay was conducted using a one-step detection kit (#C1086; Beyotime) following the protocol. Tissue samples were deparaffinized in xylene, rehydrated by a graded series of ethanol concentrations, and then heated for antigen retrieval. Hydrogen peroxide of 3% was used to inhibit endogenous peroxidase activity. Subsequently, different dilutions of terminal deoxynucleotidyl transferase in reaction buffer, containing a constant concentration of digoxigenin-labeled nucleotides, were applied to the sections and incubated at 37°C for 1 h, followed by a 10-minute incubation in Stop/Wash buffer. After extensive washing, 50 μL TUNEL detection dilution was applied for 30 min. Then the sections were mounted, and images were captured under the laser wavelength between 450 and 500 nm. The percentage of TUNEL+ cells was quantified and normalized relative to the total number of cells for each group.

Total RNA was extracted from fresh samples using the Total RNA kit (Omega Bio-Tek, Inc.) and stored on ice, following the manufacturer’s protocol. First-strand cDNA synthesis and SYBR^® Green qPCR were conducted with the PrimeScript™ RT Reagent kit (Takara Bio, Inc.). Specific primers used in the study for SIRT3 are listed as follows: F-5′-AAGACATACGGGTGGAGCCT -3′, R-5′ GGA​CTC​AGA​GCA​AAG​GAC​CC-3’. Real-time PCR was conducted using the SYBR Green kit (TINGEN Biotech, Beijing, China) under the following conditions: denaturation at 95°C for 20 s, annealing at 58°C for 20 s, and extension at 68°C for 30 s mRNA expression levels were measured by the 2^−ΔΔCq method and normalized to β-actin level. Levels of SIRT3 in control group against β-actin were normalized to 100%. The qPCR results represent three independent experiments.

This assay was conducted using a standard protocol with polyclonal antibodies specific to SIRT3, LKB1, phosphorylated(p)-LKB1, AMPK and phosphorylated(p)-AMPK. The procedures for protein extraction and immunoblotting were previously detailed. Protein samples were extracted from lung tissue homogenates using RIPA buffer (Beyotime, Shanghai, China), supplemented with inhibitors of protease and phosphatase. Protein concentrations were then determined using the BCA assay. The proteins were separated using 10% SDS-PAGE and subsequently transferred to a PVDF membrane. Following a blocking step with 5% non-fat milk, the membrane was incubated overnight at 4°C with antibodies targeting SIRT3 (1:1,000; #ab217319; Abcam), LKB1 (1:1,000; #ab199970), phosphorylated(p)-LKB1 (1:500; #ab63473; Abcam), AMPK (1:1,000; #ab32047; Abcam) and Anti-AMPK alpha 1 (phospho T183) + AMPK alpha 2 (phospho T172) antibody (1:1,000; #ab133448; Abcam) in TBS buffer. GAPDH was used as a loading control (1:2,000; #5174; CST, MA, United States). Following incubation with secondary antibodies, either HRP goat anti-rabbit (1:1,000; # A0208; Beyotime) or anti-mouse IgG (1:1,000; #A0216; Beyotime)—the blots were developed using the BeyoECL Star ECL (P0018AM; Beyotime) and imaged. Band intensities were quantified using ImageJ (NIH, United States). Levels in control group against GAPDH were normalized to 100%. The Western blot results represent three independent experiments.

Statistical analyses and graph construction were conducted by the software package (version 10.0; GraphPad Software, Inc.). Data are expressed as mean ± standard deviation (SD) from a minimum of three independent experiments. To assess significant differences between control and treatment groups, one-way ANOVA followed by Tukey’s post hoc test or unpaired Student’s t-test were employed. A p-value of less than 0.05 was regarded as statistically significant.

We first constructed a commonly used acute lung injury model using LPS. As described in the methods, rats were treated with LPS and then samples were collected and subjected to HE staining. Histological characteristics of the sections were observed. As illustrated in Figure 1A, LPS administration significantly caused the diffuse damage in the alveoli, alveolar tubes, alveolar sacs, bronchi, and alveolar septa. The quantified data of inflammation score shown in Figure 1B revealed that numerous inflammatory cells were observed in the alveolar septa after LPS exposure. Furthermore, samples were subjected to TUNEL assay. The results in Figure 1C showed that LPS administration markedly increased the number of TUNEL-positive pulmonary cells in lung samples. The quantified data were shown in Figure 1D. Then, we conducted bronchoalveolar lavages (BAL) assay to reveal the inflammatory cell infiltration by Wright-Giemsa method. As shown in Figure 1E, LPS administration significantly resulted in the increasement of immune cells, eosinophils, macrophages, lymphocytes and neutrophils. Furthermore, in LPS treated lung samples, the BAL eotaxin level was also increased. These data indicate that LPS administration significantly induced acute lung injury.

Next, we tried to ask whether SIRT3 was involved in LPS-induced ALI. Samples from LPS treated rats were subjected to RT-qPCR and the levels of SIRT3 were measured. As shown in Figure 2A, the levels of SIRT3 were significantly reduced in LPS group, compared to control rats. Then cultured lung epithelial cells were treated with LPS of indicated time points (2, 4, 24, 48 and 72 h). The data revealed that the mRNA levels of SIRT3 were gradually decreasing along the time points (Figure 2B). Samples of LPS treated rats and cultured cells administrated with LPS were subjected to Western blotting assay. The data showed that SIRT3 protein levels were also markedly decreased upon LPS treatment (Figures 2C,D). Also, the levels of SIRT3 protein were gradually downregulated in LPS-treated lung endothelial cells, same trend as the mRNA levels (Figures 2E,F).

We have previously reported that DEX could alleviate LPS-induced acute lung injury (Sun et al., 2020). Since SIRT3 was involved in LPS-induced ALI, we asked whether DEX would function via the regulation of SIRT3 signaling pathway. LPS-administrated rats were treated with DEX, or the SIRT3 inhibitor 3-TYP. Firstly, we performed immunohistochemistry assay to reveal the histological characteristics of the sections under different treatment. As shown in Figure 3A, LPS-induced inflammation was significantly suppressed by DEX treatment, however, this process could be restored by SIRT3 inhibitor. The diffuse damage in the alveoli, alveolar tubes, alveolar sacs, bronchi, and alveolar septa showed the same trend as inflammation response, which was quantified in Figure 3B. Next, LPS administration significantly decreased the SIRT3 protein levels (Figures 3C,D), which could be rescued by the treatment of DEX. However, the addition of SIRT3 inhibitor 3-TYP further alleviated the restored level of SIRT3 protein (Figure 3B). Furthermore, when compared to the total level, the phosphorylated level of LKB1 and AMPK proteins were also shown the same trend as SIRT3 protein (Figures 3E,F). These results suggest that SIRT3 signaling pathway is involved in LPS-induced AKL, which could be modulated by DEX treatment.

To investigate whether the pro-inflammatory response evoked by LPS could be mitigated by DEX through the SIRT3 signaling pathway, we administered rats with LPS, DEX, or the SIRT3 inhibitor 3-TYP. We then assessed the levels of pro-inflammatory cytokines in their serum. As depicted in Figure 4, exposure to LPS markedly upregulated the serum levels of TNF-α (Figure 4A), IL-6 (Figure 4B), IL-5 (Figure 4C), IL-1β (Figure 4D), IL-18 (Figure 4E), and IL-17 (Figure 4F). However, the administration of DEX notably attenuated these cytokine levels, an effect that was nullified by the additional treatment with 3-TYP (Figure 4). Collectively, these findings indicate that pharmacological inhibition of the SIRT3 pathway undermines the protective role of DEX in LPS-induced inflammation in ALI.

Oxidative stress, closely related to SIRT3 signaling pathway (Zheng et al., 2023), is critical for ALI (Zhong et al., 2024). To elucidate the role of SIRT3 in the generation of ROS in LPS-induced acute lung injury, we measured the activities of SOD and CAT in serum and lung tissue samples after LPS treatment. As depicted in Figures 5A–D, LPS induction significantly decreased the levels of SOD activity (Figures 5A,B) and CAT activity (Figures 5C,D). However, DEX administration notably reversed these reductions in SOD and CAT activities, an effect that was abrogated by the addition of the SIRT3 inhibitor 3-TYP (Figures 5A–D). In contrast, the level of MDA was upregulated by LPS treatment, but this increase was attenuated by DEX and further restored with the additional treatment of 3-TYP (Figures 5E,F). Moreover, we observed that the levels of oxidants such as O2−, H2O2, and nitrate in LPS-induced lung tissue samples were elevated following LPS treatment, but DEX treatment reduced these levels (Figures 5G–I). The addition of 3-TYP significantly inhibited the mitigating effect of DEX. Token together, these findings suggest that the inhibition of the SIRT3 pathway impairs the protective function of DEX against LPS-induced oxidative stress in ALI.