Sepsis is a systemic disease resulting from pathogen infection [1]. It is one of the most deadly diseases in the world and is usually associated with organ dysfunction induced by host anti-infective defense disorders [2]. The lungs are particularly susceptible to sepsis [3]. As a result of inflammation and oxidative stress, sepsis can cause acute respiratory distress syndrome or acute lung injury (ALI) [3]. Ferroptosis is a programmed cell death strictly regulated by many factors [4,5]. The accumulation of lipid peroxidation has a significant impact on ferroptosis, thereby emphasizing the crucial role of lipid metabolism in this process. Polyunsaturated fatty acids (PUFAs) are linear fatty acids with 18-22 carbon atoms and two or more double bonds, which make them more vulnerable to reactive oxygen species (ROS) and produce lipid ROS [6]. Arachidonic acid (ARA) is a metabolically essential PUFA and a significant constituent of biofilms. ARA serves as the primary substrate for several enzymes, including cyclooxygenase/prostaglandin endoperoxide synthase (Cox1/Cox2), arachidonic acid lipoxygenase (lox/Alox), and cytochrome P450 (Cyp450) [7]. These enzymes are responsible for the production of various pro-inflammatory and pro-thrombotic eicosanoids, including prostaglandin, leukotriene, and thromboxane, which subsequently initiate additional inflammatory cascades [7].

Rhodiola rosea L., a traditional Chinese medicine belonging to the Rosaceae Crassulaceae family, is predominantly found in high altitude regions such as Gansu and Tibet in China. It is recognized as an authentic medicinal substance in Gansu. Salidroside (SAL), a tyrosol glycoside derived from Rhodiola rosea L., is widely regarded as one of its most potent medicinal components [8]. There are various pharmacological effects of SAL, including anti-inflammatory, anti-aging, anti-fatigue, and anti-oxidant properties [9,10]. Nevertheless, the understanding of SAL's mechanism in sepsis-induced ALI remains limited. Increasingly, multi-omics analysis technology is being used to identify potential biomarkers and elucidate disease pathogenesis. In comparison to single omics data, multi-omics analysis offers a more comprehensive understanding of potential pathogenic alterations [11].

In this study, ALI was induced in C57BL/6J mice through cecal ligation and puncture (CLP) technique. Transcriptome analysis was conducted using RNA-seq data to elucidate alterations in gene expression and biological processes in response to ferroptosis. Differential metabolites were identified through non-targeted metabolomics analysis employing the LC-MS/MS method, which unveiled the metabolic pathways associated with ferroptosis in mice with sepsis-induced ALI. The findings from omics approaches were integrated for comprehensive analysis, with the objective of identifying the fundamental biological process responsible for the advantageous effects of SAL on lung injury in septic mice. Subsequently, the regulatory effect of this biological process on septic lung injury mice was verified by regulating the ferroptosis process in vivo.

Animals

36 male C57BL/6J mice (6–8 weeks, 18–22 g) provided by Beijing Vital River Laboratory Animal Technology Co., Ltd. and placed in cages. The illumination time followed the normal circadian rhythm. The ambient temperature was consistently maintained at 24°C ± 1°C, while the humidity levels were maintained within the range of 40%–60%. In addition to food and water, mice were provided with ad libitum access to both. All animal experiments were approved by the Laboratory Animal Committee of the Second Hospital of Lanzhou University (No. D2023-333).

SAL (HY-N0109), Erastin (HY-15763), and Ferrostatin-1 (Fer-1, HY-100579) were all obtained from MedChem Express. The mice were allocated into 6 groups using randomization (N = 6): Sham, CLP, CLP + SAL 50, CLP + SAL 100, CLP + SAL 100 + Erastin, CLP + SAL 100 + Fer-1 groups. Sham group: mice were only treated with opening suture. CLP group: CLP surgery was used to induce a mouse sepsis model. The cecum of the mice was exposed after a midline laparotomy was performed under anesthesia (75 mg/kg 1% pelltobarbitalum natricum intraperitoneal). By pushing feces into the cecum, a 5-0 propylene line was used for fixing them. Several punctures were made in the cecum in an effort to exclude any intestinal obstruction. A small amount of feces leaked during each puncture. The cecum was then placed in the abdominal cavity and the wound was sutured. The mice in CLP + SAL 50 and CLP + SAL 100 groups were given 50 mg/kg or 100 mg/kg SAL by gavage for 24 h before CLP surgery. CLP + SAL 100 + Erastin/Fer-1 group: 24 h before CLP, mice were treated with 100 mg/kg SAL by gavage and 40 mg/kg Erastin or 5 mg/kg Fer-1 by intraperitoneal injection. One week after the operation, the mice were euthanized with excessive CO₂, and the lungs of the mice were lavaged three times by tracheal intubation to obtain bronchoalveolar lavage fluid (BALF) with a total volume of 0.5 mL. A lung tissue sample was harvested and fixed in 4.0% paraformaldehyde (PFA) for subsequent experiments.

Wet-dry weight ratio

As a measure of wet weight (W), the pulmonary organs of mice were extracted and weighed. Subsequently, A 48-h drying process was performed on the lungs at a temperature of 60°C, after which they were weighed to determine their dry weight (D). The W/D ratio was assessed as a means of evaluating the presence of pulmonary edema.

Hematoxylin-eosin (H&E) staining

The morphology of lung tissue was examined by H&E staining following the standard protocol. Lung tissue was fixed overnight with 4% PFA. After tissues had been fixed, they were embedded in paraffin and sectioned into 5 μM. Sections were dehydrated with different concentrations of ethanol and xylene, and then briefly washed with 5% hematoxylin solution and stained with nuclei for 10 min. After rinsing in distilled water for 5 min, the dyed samples were incubated in 0.1% HCl-ethanol for 30 sec. Then the samples were re-stained with eosin staining solution for 2 min. After washing and dehydration, H&E-stained sections are mounted on a microscope (IX-73; Olympus) for imaging.

Terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL)

Apoptosis was evaluated using TUNEL and 4',6-diamino-2-phenylindole (DAPI) staining (Thermo Fisher Scientific). DAPI was employed to stain the nucleus, enabling the differentiation of apoptotic cells from non-apoptotic cells. The quantification of apoptotic cells was performed through TUNEL-DAPI double staining. Following fixation with 4% PFA for 0.5 h at room temperature, tissue sections (5 μm) were incubated with 0.5% Triton X-100 for 10 min. Subsequently, the sections were exposed to a TUNEL reaction mixture and co-cultured at 37°C for 60 min. A TUNEL assay was conducted on each slide after observing the samples under an Olympus IX-73 microscope.

Enzyme-linked immunosorbent assay (ELISA)

BALF inflammation levels were quantified by employing ELISA kits in accordance with the guidelines provided by the manufacturer: interleukin (IL)-1β (MLB00C; R&D System), IL-6 (M6000B), tumor necrosis factor-α (TNF-α, MTA00B) and IL-10 (M1000B). The optical density value was assessed using a microplate reader (1410101, Thermo Fisher Scientific) at 450 nm.

Transcriptome analysis

Total RNA was extracted from the lung tissue of mice. The purity of RNA was determined by NanoDrop (Thermo Fisher Scientific), and the A260/A280 ratio was 1.8–2.0. RNA was converted into complementary DNA (cDNA) through the process of reverse transcription, utilizing a specialized cDNA reverse transcription kit (RT Master Mix for qPCR II; MedChem Express) according to the instructions, and the quality of the cDNA library was tested using NovaSeq 6000 (PE150; Illumina). DESeq2 platform was used to determine the differentially expressed genes (DEGs) by filtering the original sequence. Genes with differential expression were screened based on a Log2 |(fold change)| > 1 and false discovery rate < 0.05. Pathways from GeneOntology and Kyoto Encyclopedia of Genes and Genomes (KEGG) were analyzed for enrichment.

Metabolite extraction

100 mg of lung tissue samples were introduced into a 500 ml solution consisting of 80% methanol, contained within an eppendorf tube. Following the occurrence of vortex oscillation, they were allowed to stand for five min in an ice bath before extraction. Subsequently, it was centrifuged at 15,000 × g for 20 min at 4°C. The supernatant was collected and diluted with water to a methanol content of 53%, and then centrifuged (15,000 × g, 20 min, 4°C). Using liquid chromatography-mass spectrometry (LC-MS), the metabolite levels of the supernatant were determined [12].

UHPLC-MS/MS analysis

Analyses were conducted in Gene Denovo Co., Ltd. using an Orbitrap Q Exactive HF-X mass spectrometer (Thermo Fisher) coupled to a Vanquish UHPLC system (Thermo Fisher). At a flow rate of 0.2 ml/min, injections were made onto a Hypesil Gold column (100 × 2.1 mm, 1.9 m). A 0.1% formic acid solution in water and a methanol solution were used for the positive polarity mode. The eluents for the negative polarity mode were eluent A (5 mM ammonium acetate, pH 9.0) and eluent B (methanol). The solvent gradient was set as follows: 2% B, 1.5 min; 2%–100% B, 12.0 min; 100% B, 14.0 min; 100%–2% B, 14.1 min; 2% B, 17 min. 3.2 kVspray voltage, 320°C capillary temperature, 40 arb sheath gas flow rate, 10 aux gas flow rate were used in Q ExactiveTM HF-X mass spectrometer operation in positive/negative polarity mode.

Metabolite identification

The data from the offline sources (.raw) were imported into the CD search software, and parameters like retention time and mass-to-charge ratio were tested. 0.2 min retention time deviation and 5 ppm mass deviation were used as parameters for peak alignment. Afterwards, the peak extraction was carried out by 5 ppm mass deviation, 30% signal intensity deviation, 3 signal-to-noise ratio, additive ions and other information. A simultaneous calculation of peak area and integration of target ions were performed. After that, the molecular formula was predicted and subsequently compared with databases such as mzCloud, mzVault, and Masslist, utilizing molecular ion peaks and fragment ions as the basis for analysis. To identify the data and obtain quantitative results, we removed the background ions with blank samples and classified the quantitative results. Metabolites were analyzed using PCA clustering and supervised orthogonal partial least squares discriminant analysis (OPLS-DA). The OPLS-DA multivariate statistical analysis was used to identify differential metabolites, whereas t-test values (VIP ≥ 1, p < 0.05) were used to identify differential metabolites between us in the univariate analysis [13].

ROS

ROS level was determined in lung tissue by using 2,7-dichlorofluorescein diacetate (HY-D0940; MedChem Express). After 48 h of PFA fixation, the lung tissues were dehydrated using a gradient sucrose dehydration approach. Following this, the samples were embedded in the optimal cutting temperature compound (OCT gel) and sliced into sections measuring 8 μm. The frozen sections were then subjected to incubation with 50 μM dihydroethidium at a temperature of 37°C in a dark environment for a duration of 30 min. Subsequently, the sections were further incubated with 1 mg/ml DAPI for a period of 10 min. The acquisition of images was accomplished using a microscope (IX-73; Olympus) with a wavelength (Ex/Em) of 525 nm/610 nm.

Detection of ferroptosis index

The contents of ferrous (Fe²⁺), malondialdehyde (MDA) and reduced glutathione (GSH) in lung tissue lysate were detected by Fe²⁺ kit (MAK025; Sigma-Aldrich), MDA kit (S0131; Beyotime) and GSH kit (70-18-8; Merck Millipore) according to the protocol of relevant manufacturers.

Data analysis

Transcriptome and metabolome data analysis and visualization were conducted utilizing the Omicsmart platform (https://www.omicsmart.com/). GraphPad Prism (Version 8, GraphPad Software) was utilized for data manipulation. The Kolmogorov–Smirnov test confirmed the normal distribution of the data, with the results presented as mean ± standard deviation. Independent sample t-tests were used to compare the two groups. In the case of multiple groups, we used one-way analysis of variance with single factor. An analysis of the post-hoc results was conducted using Tukey's multiple comparison test. p was a two-sided test, and p < 0.05 was considered statistically significant.

SAL ameliorates sepsis-induced ALI

To investigate the potential therapeutic impact of SAL on septic ALI, C57BL/6J mice were subjected to CLP induction to induce septic lung injury. Initially, H&E staining was employed to examine the histopathological alterations in the relevant tissues of mice with sepsis-induced lung injury (Fig. 1A). Compared with the Sham group, the CLP group showed signs of inflammatory cell infiltration, pulmonary edema, and alveolar wall thickening. However, administration of SAL via gavage at doses of 50 mg/kg and 100 mg/kg ameliorated these pathological manifestations (Fig. 1A). Moreover, compared with the Sham group, CLP surgery significantly increased the lung W/D ratio (Fig. 1B, p < 0.001). Compared with the CLP group, 50 mg/kg and 100 mg/kg SAL administration significantly reduced the lung W/D ratio (Fig. 1B, p < 0.01). When compared with the Sham group, CLP mice had a higher number of TUNEL-positive cells in their lung tissue, while the use of SAL significantly reduced the number of TUNEL positive cells (Fig. 1C, all p < 0.001). Compared with sham group, CLP mice had higher concentration of TNF-α, IL-1β, and IL-6 in their BALF, whereas IL-10 concentration was lower. IL-10 considerably increased in BALF of CLP mice after SAL treatment, while TNF-α, IL-1β, and IL-6 levels were significantly inhibited (Fig. 1D, all p < 0.05).

Figure 1. SAL ameliorates sepsis-induced acute lung injury. (A) H&E staining was used to analyze the morphological changes of lung tissue in mice (x20 and x200). Yellow arrows point to alveoli, red arrows to interstitial tissue, and black arrows to inflammatory infiltrates. (B) Pulmonary edema was assessed by analyzing the lung W/D ratio. (C) TUNEL staining was used to evaluate apoptosis (x200). (D) The levels of inflammatory cytokines (TNF-α, IL-1β, IL-6 and IL-10) in BALF of mice were measured by ELISA. N = 6. Values are presented as mean ± SD. SAL, salidroside; W/D ratio, wet weight/dry weight ratio; TUNEL, terminal deoxynucleotidyl transferase dUTP nick end labeling; TNF-α, tumor necrosis factor-α; IL, interleukin; BALF, bronchoalveolar lavage fluid; CLP, cecal ligation and puncture. *p < 0.05, **p < 0.01, ***p < 0.001.

Effects of SAL on ferroptosis pathway and metabolites in septic mice

According to the above results, 100 mg/kg SAL was more effective at protecting mice from sepsis-induced ALI, so 100 mg/kg SAL was used in subsequent studies. The results of transcriptome data of lung tissue in septic mice showed that there were 1942 DEGs between the Sham group and the CLP group, with 965 genes up-regulated and 977 genes down-regulated. There were 296 DEGs in CLP group and CLP + SAL 100 group, of which 150 genes up-regulated and 146 genes down-regulated. There were 1,806 DEGs in the Sham group and the CLP + SAL 100 group, of which 904 genes up-regulated and 902 down-regulated (Fig. 2A). The results of gene set enrichment analysis (GSEA)-KEGG enrichment analysis indicated that compared with the Sham group, the ferroptosis-related pathway process was up-regulated in the CLP group, and the ferroptosis-related pathway enrichment was decreased after SAL treatment (Fig. 2B). OPLS-DA results showed that the top 10 metabolites that contributed the most to distinguish the Sham group and the CLP group were Dl-arginine, Triethylenetetramine, Oleamide, 1-oleoyl-2-myristoyl-sn-glycero-3-phosphocholine, Palmitic acidpalmitoleoyl ethanolamide, 2-oleoyl-1-palmitoyl-sn-glycero-3-phosphocholine, Glycerophosphocholine, alpha-ergocryptine and Oleic acid (Fig. 2C). Compared with the CLP group, the CLP + SAL 100 group mice Oleic acid, 1-oleoyl-2-myristoyl-sn-glyero-3-phosphocholine, ARA (peroxide free), Glycerophosphocholine, 16-hydroxy-4z, 7z, 10z, 13z, 17e. The levels of 19z-docosahexaenoic acid, 1-hexadecanoyl-2-octadecadienoyl-sn-glycero-3-phosphpcholine, alpha-ergocryptine and Zinniol were significantly different (Fig. 2C).

Figure 2. Effects of SAL on ferroptosis pathway and metabolites in septic mice. Transcriptome and metabolomics analysis of lung tissue in Sham group, CLP group and salidroside (100 mg/kg) treatment group. (A) Log2(FC) was used as the ordinate, and the grouping name was used as the abscissa to display the differentially expressed genes between groups. Green represented down-regulated genes, red represented up-regulated genes. (B) GSEA analysis was performed on all genes in the Sham group and the CLP group and the CLP and CLP + SAL 100 groups using the KEGG database. The positive and negative ES values indicated the correlation between the pathway and the comparison group. The positive ES value indicated that the pathway was positively correlated with the comparison group compared with the control group, and the negative ES value indicated that the pathway was negatively correlated with the comparison group compared with the control group. (C) OPLS-DA was used to analyze the loading map to find out the metabolites that contributed the most to the change of metabolite patterns between the Sham group and the CLP group, the CLP group and the CLP + SAL 100 group. The farther away from the origin in the abscissa direction, the greater the contribution of the variables to the distinction between the two groups of samples. SAL, salidroside; CLP, cecal ligation and puncture; GSEA, gene set enrichment analysis; KEGG, Kyoto Encyclopedia of Genes and Genomes; ES, enrichment score; OPLS-DA, orthogonal partial least squares discriminant analysis.

The relationship between ARA-related DEGs and metabolites in mice with sepsis-induced ALI

In Fig. 3A, it is evident that the CLP treatment resulted in a significant up-regulation of genes associated with the ferroptosis pathway, such as Cp, Hmox1, Cybb, Slc39a14, Fth1, Scl40a1, Slc7a11, Slc39a8, Ncoa4, and Alox15, when compared to the Sham group. Additionally, the expression levels of genes related to ARA metabolism, namely Alox15, Alox12, Ggt5, Ggt1, Alox5, Alox12e, Tbxas1, Pla2g2d, and Gpx1, were all noticeably up-regulated, while t Cyp2j9 and Cyp4a12b expression levels were significantly down-regulated (Fig. 3A). Compared with the CLP group, SAL treatment up-regulated the expression levels of Cp, Homox1, Cybb, Alox12, Ggt1, Cyp2j9, Alox12e, Alox5, Tbxas1, Gpx1, Cyp4a12b genes, and down-regulated the expression levels of Slc39a14, Fth1, Slc40a1, Slc7a11, Slc39a8, Ncoa4, Alox15, Ggt5, Pla2g2d genes (Fig. 3A). The Alox15 gene was annotated in the ferroptosis pathway and ARA metabolic pathway. Therefore, the correlation between Alox15 and ARA metabolites was analyzed in this study. Alox15 was significantly negatively correlated with (5Z, 8Z, 14Z)-11,12-Dihydroxyeicosa-5,8,14-trienoic acid (11.12-DHET) and 20-hydroxyeicosatetraenoic acid (20-HETE) (Fig. 3B, r = –0.9500, p < 0.001; r = –0.9000, p = 0.002). In order to explore the transcriptional regulation of SAL on ARA metabolism in septic lung injury mice, the main ferroptosis and ARA metabolism pathway related DEGs between the two comparison groups (Sham vs. CLP, CLP vs. CLP + SAL 100) were annotated to the metabolic pathway. As shown in Fig. 3C, in the ARA metabolic pathway, compared with the Sham group, the ARA metabolites 20-HETE, (5Z, 8Z, 10E, 14Z)-12-Oxoeicosa-5,8,10,14-tetraenoic acid (12(S)-HETE), 11,12-DHET, (5Z, 11Z, 14Z)-8,9-Dihydroxyeicosa-5,11,14-trienoic acid (8,9-DHET), Leukotriene B4 and Leukotriene D4 was up-regulated by SAL. The mRNA levels of genes Cyp4a12b, Alox12 and Cyp2j9 were up-regulated by SAL. The mRNA levels of Alox15 and Alox5 were down-regulated under SAL regulation.

Figure 3. Relationship between ferroptosis-related DEGs and metabolites in mice with sepsis-induced acute lung injury. (A) The expression levels of DEGs related to ferroptosis and arachidonic acid metabolism were displayed by heat map. (B) The correlation between arachidonic acid metabolism metabolites 11,12-dihydroxy-5Z,8Z,14Z-eicosatrienoic acid (11,12-DHET), 20-HETE and Alox15. (C) Arachidonic acid metabolism related DEGs and DMs schematic diagram. The rectangle is divided into two color blocks, left: Sham vs. CLP, right: CLP vs. CLP + SAL 100. 20-HETE, 20-hydroxyeicosatetraenoic acid; 12-OxoETE, (5Z, 8Z, 10E, 14Z)-12-Oxoeicosa-5,8,10,14-tetraenoic acid; 12(S)-HETE, (5Z, 8Z, 10E, 14Z)-(12S)-12-Hydroxyeicosa-5,8,10,14-tetraenoic acid; 11,12-DHET, (5Z, 8Z, 14Z)-11,12-Dihydroxyeicosa-5,8,14-trienoic acid; 8,9-DHET, (5Z, 11Z, 14Z)-8,9-Dihydroxyeicosa-5,11,14-trienoic acid; DEGs, differentially expressed genes; SAL, salidroside; CLP, cecal ligation and puncture; KEGG, Kyoto Encyclopedia of Genes and Genomes; DMs, differential metabolites.

SAL inhibits ferroptosis and improves lung injury in septic mice

Ferroptosis activator Erastin and inhibitor Fer-1 were injected into SAL-treated lung injury mice, respectively. Fig. 4A showed that compared with the lung tissue of mice in CLP + SAL 100 group, Erastin increased lung tissue damage, wet/dry ratio and TUNEL positive cells. While Fer-1 reduced the pathological damage of lung tissue, the lung wet/dry ratio decreased significantly (Fig. 4B, p < 0.001), and the number of TUNEL positive cells decreased significantly (Fig. 4C, all p < 0.001). In addition, compared with CLP + SAL 100 group, Erastin increased the levels of TNF-α, IL-1β, IL-6 in BALF, while Fer-1 made them decreased, and the level of IL-10 was increased (Fig. 4D, all p < 0.001). The levels of ROS, Fe²⁺, GSH and MDA in lung tissue of each treatment group were detected by immunofluorescence and kit. The experimental results showed that compared with the Sham group, CLP treatment significantly increased the level of ROS in lung tissue of mice, the level of Fe²⁺ content increased significantly, the content of GSH decreased significantly, and the content of MDA increased significantly (Fig. 4E, F, all p < 0.05), indicating that CLP promoted ferroptosis in lung tissue of mice. After treatment with SAL in mice with lung injury, the ferroptosis pathway in lung tissue was inhibited. This was verified by the addition of the ferroptosis activator Erastin or the inhibitor Fer-1. Compared with CLP + SAL 100 group, the ROS level in the lung tissue increased significantly, the Fe²⁺ and MDA content increased significantly, and the GSH content decreased significantly after the treatment of Erastin. However, in comparison with CLP + SAL 100 group, Fer-1 decreased the ROS level, Fe²⁺ and MDA content markedly, and increased the GSH content markedly (Fig. 4E, F, all p < 0.05). The above experimental results suggested that SAL inhibited ferroptosis and alleviated lung injury in septic mice.

Figure 4. SAL alleviates lung injury in septic mice by inhibiting ferroptosis. (A) H&E staining was used to analyze the morphological changes of lung tissue in mice (x20 and x200). Yellow arrows point to alveoli, red arrows to interstitial tissue, and black arrows to inflammatory infiltrates. (B) Pulmonary edema was evaluated by analyzing the lung W/D ratio. (C) TUNEL staining to evaluate apoptosis (x200). (D) The levels of TNF-α, IL-1β, IL-6 and IL-10 in BALF were measured by ELISA. (E) ROS level was detected by immunofluorescence (x200). (F) The content of Fe²⁺, glutathione (GSH) and malondialdehyde (MDA) were detected by kit. N = 6. Values are presented as mean ± SD. SAL, salidroside; W/D ratio, wet weight/dry weight ratio; TUNEL, terminal deoxynucleotidyl transferase dUTP nick end labeling; TNF-α, tumor necrosis factor-α; IL, interleukin; BALF, bronchoalveolar lavage fluid; ROS, reactive oxygen species; CLP, cecal ligation and puncture; Fer-1, ferrostatin-1. *p < 0.05, **p < 0.01, ***p < 0.001.

SAL plays a protective role in the lung injury model [14-16]. In this study, we have also observed the protective effect of SAL on ALI in mice subjected to CLP. To investigate the potential regulatory mechanism underlying the treatment effects of SAL in sepsis-induced ALI without preference, a combination of in vivo experiments and multi-omics analysis was employed. Different from the prior studies, we performed a no-preference analysis of lung tissue gene expression based on transcriptomics. It was found that after SAL treatment of septic lung injury mouse model, the differential genes were significantly enriched in the ferroptosis-related pathway, while combined metabolomics analysis found that the abundance of ARA, a metabolite related to ferroptosis metabolism, changed significantly under the influence of SAL. The above results objectively indicated that the ferroptosis pathway is involved in the improvement of SAL on sepsis-induced lung injury, and revealed a significant association between ARA metabolism-related products and the DEGs involved in ferroptosis in the sepsis-induced ALI model. Moreover, this study employed a combination of SAL and a ferroptosis activator/inhibitor to perform experiments on mice with lung injury, aiming to investigate the mechanism by which SAL regulates sepsis-induced lung injury.

Studies have shown that SAL alleviates kidney injury by inhibiting renal inflammation and tubular epithelial cell apoptosis to prevent acute kidney injury in septic rats [17]. In addition, studies have confirmed that SAL has a protective effect on ALI in mice [18]. According to this study, SAL could significantly relieve the pathological damage of lung tissue, and play a protective role on sepsis mice. Through differential gene enrichment analysis of transcriptome data, it was observed that CLP up-regulated ferroptosis in lung tissue of mice, while SAL inhibited ferroptosis in lung tissue of septic mice. Wang et al. [19] showed that SAL activated the Nrf2/Slc7a11 signaling axis to reduce ferroptosis-mediated ischemia-reperfusion-induced lung injury. A form of programmed cell death, ferroptosis is controlled by lipid oxidation and dependent on iron [20]. Metabolomics analysis showed that ARA metabolism-related processes may play an important role in the process of SAL protecting lung injury mice. Although ARA is widely distributed in the body, it is mainly bound to hydroxyl groups on glycerophospholipids, and therefore is little free ARA [21]. Phospholipase A2 is activated when oxidative stress occurs in the body, which promotes the release of ARA [22]. ARA is mainly used for the synthesis of pro-inflammatory factors. When pro-inflammatory factors bind to inflammatory cell receptors, intracellular inflammatory signal transduction is activated, causing an increase in the release of inflammatory factors (such as TNF-α and IL), and thus promote the inflammatory response [23]. Given that ARA serves as the primary substrate for lipoxygenases, it holds significant importance as a substrate for ferroptosis [24,25]. ARA constitutes a highly organized oxygenation center, which may undergo severe peroxidation and form ROS under the iron-dependent mechanism, leading to lipid peroxidation, thereby reducing GSH levels and activating the ferroptosis process [21,26]. In this study, it was observed that CLP increased the ROS level, decreased the GSH level, increased the MDA level, and promoted the ferroptosis process in the lung tissue of mice. Alox5, Alox12, and Alox15 are lipoxygenases that mediate the enzymatic oxidation of PUFAs selectively and specifically [27]. Alox15 plays a crucial role in ferroptosis [28]. In the present study, it was observed that Alox15 is involved in the mechanism of ferroptosis and ARA metabolism in mice with lung injury treated with SAL. ARA undergoes oxidation mediated by Alox15, resulting in the formation of 13-hydroperoxyoctadecadienoic acid and 15-hydroperoxyeicosatetraenoic acid. These two compounds are considered as intermediary products in the synthesis of 4-Hydroxynonenal, which is recognized as a consequential product of ferroptosis [29]. In this study, it was observed that CLP treatment resulted in an up-regulation of Alox15 expression. Additionally, a significant negative correlation was found between the expression level of Alox15 and the ARA metabolites 11,12-DHET and 20-HETE. To further investigate the impact of ferroptosis on sepsis-induced lung injury in mice, Erastin or Fer-1 was administered to sepsis mice treated with SAL. The administration of Erastin led to severe lung tissue damage, increased expression of inflammatory factors in BALF, and facilitated ferroptosis in lung tissue. It was verified that inhibition of ferroptosis during SAL treatment of sepsis-induced lung injury in mice exerted an inhibitory effect on lung tissue injury and inflammatory factors in mice.

Compared with sepsis-induced ALI mice, SAL significantly protects lung function and alleviates pathological damage. Combined with transcriptome and metabolomics data analysis, the specific mechanism of protection is closely related to ferroptosis and ARA metabolism. SAL may mediate ARA metabolism to inhibit ferroptosis and alleviate sepsis-induced ALI. These findings offer novel insights into the therapeutic approach for sepsis-induced ALI and hold significant potential for practical applications. However, a limitation of this study is the absence of evidence indicating the potential involvement of Alox15 in the mechanism by which SAL improves ferroptosis in sepsis-induced lung injury. Furthermore, the study did not investigate the regulation of Alox15 expression in animals to validate its role. Therefore, future research should focus on conducting comprehensive evidence gathering and exploring the molecular mechanisms involved.

None.

The Second Hospital of Lanzhou University &quot;Cuiying Science and Technology Innovation&quot; Program (CY2020-MS06), Natural Science Foundation of Gansu Province (20JR10RA724), Lanzhou Science and Technology Development Program (2020-ZD-91).

The authors declare no conflicts of interest.