Introduction

Mechanical ventilation (MV) has become a life-saving treatment method for patients with acute respiratory failure, and its application in various medical fields related to respiratory failure is gradually widespread1. However, improper use of MV can also lead to new lung tissue damage or aggravate of existing lung damage, known as ventilators induced lung injury (VILI), which leading to a significant increase in mortality among severely ill patients2,3. However, the pathogenesis of VILI has not been fully elucidated. Therefore, to further explore the pathogenesis of VILI and search for new therapeutic targets have become the main contradiction that needs to be solved urgently.

MicroRNA is a unique small non-coding RNA, consisting mainly of single-stranded molecules of 21–23 nucleotides, which can negatively regulate target genes at the transcriptional penology level by binding to the 3′UTR region of mRNA4,5. Previous studies have confirmed that miRNA is involved in the regulation of various cellular functions, and abnormal expression of miRNA can lead to the occurrence and development of various diseases6, for example, heart diseases7,8, immune system diseases9,10, infectious respiratory diseases11 etc. Researchers also revealed that, miR-125b-5p mitigates acute lung injury (ALI) caused by sepsis12. In addition, the overexpression of miR-125b-5p can improve inflammation and impaired lung barrier function associated with lung injury induced by sepsis13. Notably, Vaporidi et al. detected the differential expression of miR-125b-5p in the lung tissue of VILI mice14, these may indicate the correlation between miR-125b-5p and VILI. However, the mechanism of action of miR-125b-5p in VILI has not been revealed.

Ferroptosis, a unique form of iron-dependent programmed cell death, is primarily caused by large accumulations of lipid peroxides15,16. It was regulated by multiple metabolic pathways, including glutathione metabolism, iron ion metabolism, lipid oxidative metabolism etc.17,18. Multiple studies have reported that ferroptosis plays a critical role in the development of various diseases, such as the nervous system, cardiovascular system, metabolic liver injury, and acute lung injury19,20,21,22. According to previous research, it has been confirmed that inhibiting ferroptosis can relieve lung tissue inflammation and damage in VILI mouse23. However, the specific mechanism of ferroptosis in VILI has not been elucidated. Glutathione peroxidase 4 (GPX4), is a key negative regulator factor in the ferroptosis regulation pathway, which plays a central enzymatic role in ferroptosis by reducing the level of lipid peroxidation24. Seibt et al.’s study pointed out that knockdown or degradation of GPX4 leads to accumulation of lipid peroxides, leading to ferroptosis, while up-regulation of GPX4 inhibits ferroptosis25. Nuclear factor Erythrocyte 2-associated factor 2 (Nrf2) is an endogenous transcription factor26. Keap1 (Kelch-like ECH-associated protein 1) was a sensor protein of the oxidative stress response, plays an essential role in maintaining redox homeostasis in vivo27. Under homeostatic conditions, Nrf2 is continuously degraded by ubiquitination, tightly bound to Keap1 and inactivated in the cytoplasm28. When oxidative stress occurs, the structure of Keap1 changes, causing Nrf2 to be unable to be degraded by ubiquitination, meanwhile, Nrf2 was translocated to the nucleus and binds to antioxidant response elements, thereby activating downstream antioxidant gene GPX4, reducing ROS, lipid peroxidation, and iron ion levels29. Therefore, activating the Keap1/Nrf2/GPX4 pathway can reduce lipid peroxidation, ROS production, and iron ion accumulation, thereby inhibiting ferroptosis, alleviating inflammation, and tissue damage. Studies indicated that in the myocardial ischemia/reperfusion (MI/R) injury and chronic kidney disease (CKD) mouse model, excessive accumulation of iron ions and lipid peroxides exacerbates tissue damage, while activation of the Keap1/Nrf2/GPX4 pathway can inhibit ferroptosis and alleviate tissue damage30,31. Further, it was reported that, in CLP generated sepsis model, miR-125b-5p can alleviate inflammation induced ferroptosis and protect lung injury from sepsis via Keap1/Nrf2/GPX4 axis32. However, in the VILI rat models, whether miR-125b-5p can play a protective role by regulating ferroptosis through the Keap1/Nrf2/GPX4 axis has not been reported.

In this study, we constructed VILI rat models in vivo and in vitro, exploring the role and mechanism of miR-125b-5p in regulating ferroptosis in VILI. Our results show that miR-125b-5p can inhibit ferroptosis and alleviate lung injury in VILI rats by targeting the Keap1/Nrf2/GPX4 pathway, playing a protective role in in vivo and in vitro VILI models.

Results

Ferroptosis is involved in the progression of VILI in rats

To verify whether ferroptosis is involved in the progression of VILI, a VILI rat model was established. Compared with the control group, visual observation of lung tissue morphology, HE staining, lung tissue injury score, W/D ratio, and BALF protein levels showed significant pathological damage to lung tissue in the HV group (Fig. 1A–D).

Fig. 1
figure 1

Ferroptosis is involved in the progression of VILI in rats. (A) Observation of the lung tissue with the naked eye. (B) Wet/dry weight ratio of the lung tissue. (C) Total protein content in BALF. (D) Lung tissue HE staining (200 × and 400 ×) based on observation with a light microscope and lung injury score. (E) Protein expression of GPX4, SLC7A11 and ACSL4 were detected by western blotting, with β-actin as the internal control. (F) DAB-enhanced Perls’ Prussian blue staining was used to evaluate the content of lung tissue ferric ions. Detection of biochemical indicators of ferroptosis: (G) GSH, (H) MDA, (I) Tissue iron. (J) Immunofluorescence microscopy analysis of the distribution of fluorescence labeled GPX4 protein. (***P < 0.001. ****P < 0.0001.) VILI, ventilator-induced lung injury; BALF, bronchoalveolar lavage fluid; HE, Hematoxylin and eosin; ACSL4, Acyl-CoA synthetase long-chain family member 4; GPX4, Glutathione peroxidase-4; SLC7A11, Solute carrier family 7 member 11; GSH, Glutathione; MDA, malondialdehyde.

Based on western blotting, the expression levels of ferroptosis related indicators GPX4 and SCLC7A11 were significantly reduced in the HV group, compared to the control group, whereas the expression of ACSL4 was significantly increased (Fig. 1E). In addition, as Fig. 1F–I illustrates that, tissue iron detection, MDA detection and DAB-enhanced Perls’ Prussian blue staining showed that iron accumulation and MDA content were significantly increased, while GSH content was significantly decreased compared with the control group in VILI rats. Immunofluorescence result affirmed same regulative effect of HV group on GPX4 expression (Fig. 1J).

Subsequently, an inhibitor of ferroptosis Fer-1 was applied in our study. Significantly, we found that co-treatment with Fer-1 reduced lung edema and congestion (Fig. 2A), as well as the degree of pathological injury in the lung tissue, as shown by H&E staining and lung injury scores (Fig. 2D). The W/D ratio (Fig. 2B) and total protein level also confirmed the protective effect of Fer-1 (Fig. 2C). In addition, western blot results demonstrated that the expression levels of SLC7A11 and GPX4 proteins were upregulated in HV + Fer-1 group when compared with the HV + DMSO group, whereas the expression of ACSL4 was downregulated (Fig. 2E). Immunofluorescence analysis further confirmed that Fer-1 can inhibit ferroptosis (Fig. 2J). Furthermore, the amounts of MDA, tissue iron and iron deposition in the Perls’ Prussian blue staining in the HV + fe-1 group were lower than those in the HV + MDSO group, while the amounts of GSH in the HV + Fer-1 group were increased (Fig. 2F–I). Overall, those results suggested that ferroptosis is involved in the progression of VILI in rats.

Fig. 2
figure 2

Ferroptosis is involved in the progression of VILI in rats. (A) Observation of the lung tissue with the naked eye. (B) Wet/dry weight ratio of the lung tissue. (C) Total protein content in BALF. (D) Lung tissue HE staining (200 × and 400 ×) based on observation with a light microscope and lung injury score. (E) Protein expression of GPX4, SLC7A11 and ACSL4 were detected by western blotting, with β-actin as the internal control. (F) DAB-enhanced Perls’ Prussian blue staining was used to evaluate the content of lung tissue ferric ions. Detection of biochemical indicators of ferroptosis: (G) GSH, (H) MDA, (I) Tissue iron. (J) Immunofluorescence microscopy analysis of the distribution of fluorescence labeled GPX4 protein. (*P < 0.05. **P < 0.01. ***P < 0.001. ****P < 0.0001.) VILI, ventilator-induced lung injury; BALF, bronchoalveolar lavage fluid; HE, Hematoxylin and eosin; ACSL4, Acyl-CoA synthetase long-chain family member 4; GPX4, Glutathione peroxidase-4; SLC7A11, Solute carrier family 7 member 11; GSH, Glutathione; MDA, malondialdehyde.

Overexpression of miR-125b-5p alleviates ferroptosis and lung tissue injury in VILI rats

Real-time qPCR result indicated that the expression of miR-125b-5p was decreased in the HV group (Fig. 3A). To further clarify the function of miR-125b-5p in VILI, rats were pretreatmented with miR-125b-5p agomir and agomir-NC. Subsequently, the expression of miR-125b-5p in VILI rats was significantly increased by qRT-PCR (Fig. 3B), indicating successful transfection. In addition, after pretreatment of rats with miR-125b-5p antagonists and antagomir-NC, the expression of miR-125b-5p in lung tissues was significantly reduced by qRT-PCR (Supplementary Fig. S1), indicating successful transfection.

Fig. 3
figure 3

Overexpression of miR-125b-5p alleviates ferroptosis and lung tissue injury in VILI rats. (A,B) Expression of miR-125b-5p in the lung tissue of rats based on qRT-PCR. (C) Observation of the lung tissue with the naked eye. (D) Wet/dry weight ratio of the lung tissue. (E) Total protein content in BALF. (F) Lung tissue HE staining (200 × and 400 ×) based on observation with a light microscope and lung injury score. (**P < 0.01.***P < 0.001.****P < 0.0001). VILI, ventilator-induced lung injury; BALF, bronchoalveolar lavage fluid; HE, Hematoxylin and eosin; qRT-PCR, real-time quantitative polymerase chain reaction.

By observing the appearance of lung tissue with the naked eye, it was found that lung injury was significantly alleviated in VILI rats, treated with miR-125b-5p agomir. As shown in Fig. 3C, the lungs of these rats have a rosy appearance, mild edema, and a small amount of blood clots on the surface. Besides, pretreatment with miR-125b-5p agomir, resulting in a significantly decreased in the degree of pathological injury to lung tissue, compared to the HV group (Fig. 3D–F).

Furthermore, through western blotting, the expression levels of SLC7A11 and GPX4 were increased significantly, while the expression of ACSL4 decreased compared to the HV group (Fig. 4A). In fact, immunofluorescence result affirmed same regulative effect of HV + agomir group on GPX4 expression (Fig. 4F). Of course, pretreatment with miR-125b-5p agomir resulted in a significant decrease in MDA and iron ion content, compared to the HV group (Fig. 4B,D,E), while the amounts of GSH were significantly elevated (Fig. 4C). However the opposite results were observed in miR-125b-5p antagomir group (Supplementary Fig. S1). To sum up, overexpression of miR-125b-5p significantly improved the pathological morphology of lung tissue and reduced ferroptosis, thereby alleviating VILI in rats.

Fig. 4
figure 4

Overexpression of miR-125b-5p alleviates ferroptosis and lung tissue injury in VILI rats. (A) Protein expression of GPX4, SLC7A11 and ACSL4 were detected by western blotting, with β-actin as the internal control. (B) DAB-enhanced Perls’ Prussian blue staining was used to evaluate the content of lung tissue ferric ions. Levels of GSH (C), MDA (D) and total iron (E) in the lung tissues. (F) GPX4 immunofluorescent staining results in lung tissue of miR-125b-5p treated VILI rats. (**P < 0.01.***P < 0.001.****P < 0.0001). VILI, ventilator-induced lung injury; ACSL4, Acyl-CoA synthetase long-chain family member 4; GPX4, Glutathione peroxidase-4; SLC7A11, Solute carrier family 7 member 11; GSH, Glutathione; MDA, malondialdehyde.

Ferroptosis was present in MS-induced injury cell models

To further confirm the existence of ferroptosis in the MS induced injury cell model, we constructed a mechanical stretching (MS) model in ATII cells. Through western blotting, compared with the control group, the expression of ferroptosis related biomarkers SLC7A11 and GPX4 were significantly downregulated, while the expression of ACSL4 was significantly upregulated in the MS group (Fig. 5A). In addition, immunofluorescence results showed the ROS accumulation increased significantly in the MS-induced cell model (Fig. 5B). Similarly, the content of MDA and Cell iron ions were increased while the expression of GSH was downregulated in the MS group (Fig. 5C–E).

Fig. 5
figure 5

Ferroptosis was present in MS-induced injury cell models. (A) Protein expression of SLC7A11, GPX4 and ACSL4 in ATII cells were detected by western blotting, with β-actin as the internal control. (B) ROS was detected by immunofluorescence. Green: ROS. Magnification: 40x. Content of GSH (C), MDA (D) and cell iron (E) in MS-induced injury cell model. (**P < 0.01. ***P < 0.001. ****P < 0.0001). MS, mechanical stretch; SLC7A11, Solute carrier family 7 member 11; GPX4, Glutathione peroxidase-4; ACSL4, Acyl-CoA synthetase long-chain family member 4; ROS, Reactive oxygen species; GSH, Glutathione; MDA, malondialdehyde.

In addition, we applied the ferroptosis inhibitor Fer-1 in our study. As the results showed that the protein levels of both SLC7A11 and GPX4 was higher in the MS + Fer-1 group than the MS + DMSO group, while the protein levels of ACSL4 was lower in the MS + Fer-1 group (Fig. 6A). As Fig. 6B displayed, results indicated that ROS levels were remarkably elevated in MS + DMSO group, while Fer-1 could reduce the accumulation. Furthermore, the amounts of MDA and cell iron in MS + DMSO group were much higher than that of MS + Fer-1 group (Fig. 6D,E). Meanwhile, the expression of antioxidant GSH was significantly decreased after treated with MS (Fig. 6C), while Fer-1 could restore their contents. Therefore, these findings indicate the presence of ferroptosis in the VILI rat models.

Fig. 6
figure 6

Ferroptosis was present in MS-induced injury cell models. (A) Protein expression of SLC7A11, GPX4 and ACSL4 in ATII cells were detected by western blotting, with β-actin as the internal control. (B) ROS was detected by immunofluorescence. Green: ROS. Magnification: 40×. Content of GSH (C), MDA (D) and cell iron (E) in MS-induced injury cell model. (**P < 0.01. ***P < 0.001. ****P < 0.0001). MS, mechanical stretch; SLC7A11, Solute carrier family 7 member 11; GPX4, Glutathione peroxidase-4; ACSL4, Acyl-CoA synthetase long-chain family member 4; ROS, Reactive oxygen species; GSH, Glutathione; MDA, malondialdehyde.

Overexpression of miR-125b-5p alleviates MS induced injury to ATII cells by inhibiting ferroptosis

To further verify the function of miR-125b-5p in vitro, a VILI cell model was constructed by mechanical stretching. Based on qRT-PCR, the expression of miR-125b-5p in the MS induced injury to ATII cells was significantly downregulated (Fig. 7A). Subsequently, miR-125b-5p mimics and inhibitors was used to pre-intervene MS-induced injury cell, respectively. In parallel, miR-125b-5p was successfully overexpressed by qRT-PCR (Fig. 7B), similarly, miR-125b-5p was successfully knocked down by qRT-PCR (Supplementary Fig. S2).

Fig. 7
figure 7

Overexpression of miR-125b-5p alleviates MS induced injury to ATII cells by inhibiting ferroptosis. (A,B) mRNA levels of miR-125b-5p in MS-induced injury cell was detected by qRT-qPCR. (C) Protein expression levels of SLC7A11, GPX4 and ACSL4 in MS-induced injury cell were measured by western blotting, with β-actin as the internal control. (D) ROS detected by immunofluorescence in MS-induced injury cell model. Green: ROS. Magnification: 40×. Detection of biochemical indicators of GSH (E), MDA (F) and total iron (G) in MS-induced injury cell models. *P < 0.05. **P < 0.01. ***P < 0.001. ****P < 0.0001. MS, mechanical stretch; ATII, Type II alveolar epithelial cell; qRT-PCR, real-time quantitative polymerase chain reaction. SLC7A11, Solute carrier family 7 member 11; GPX4, Glutathione peroxidase-4; ACSL4, Acyl-CoA synthetase long-chain family member 4; ROS, Reactive oxygen species; GSH, Glutathione; MDA, malondialdehyde.

Notably, the protein expression of SLC7A11, GPX4 were significantly increased in MS + miR-125b-5p mimic group, compared with the mimic-NC, while the expression of ACSL4 was significantly decreased (Fig. 7C). Moreover, the immunofluorescence results showed that the accumulation of ROS significantly elevated after mechanical stretching on ATII cells, while the application of miR-125b-5p mimic reduced the accumulation of ROS (Fig. 7D). Furthermore, the MDA activity and cellular iron content in the miR-125b-5p mimic group were significantly lower than those in the mimic NC group, while GSH showed the opposite trend (Fig. 7E–G). However, the application of inhibitors of miR-125b-5p reversed those changes (Supplementary Fig. S2), which was consistent with the results of in vivo experiments. Collectively, these results indicated that miR-125b-5p can alleviate MS-induced cell injury by inhibiting ferroptosis.

MiR-125b-5p alleviated ferroptosis via upregulating GPX4 in in MS-induced injury cell models

GPX 4 is a key regulatory factor in the ferroptosis regulation pathway, an inhibitor of ferroptosis, which can effectively inhibit ferroptosis by protecting cells from accumulation of lipid peroxide33,34. Therefore, this study further verified whether miR-125b-5p inhibits MS-induced cell ferroptosis through GPX4 inhibition experiments. Meanwhile, in order to achieve the best inhibitory effect, we set a concentration gradient for the GPX4 inhibitor RSL3 (Fig. 8A). As is shown in Fig. 8B, western blotting results indicated that when miR-125b-5p mimic group were treated with RSL3, expression of ACSL4 increased significantly while GPX4 and SLC7A11 were significantly inhibited, compared to the HV + miR-125b-5p mimic + DMSO group. Moreover, level of ROS expression, MDA content, and total iron in cells significantly increased (Fig. 8C,E,F), while GSH content significantly decreased (Fig. 8D), when expression of GPX4 was inhibited by RSL3 compared with the HV + miR-125b-5p mimic + DMSO group. Namely, miR-125b-5p can alleviate ferroptosis via upregulating GPX4 in MS-induced injury cell models.

Fig. 8
figure 8

MiR-125b-5p alleviated ferroptosis via upregulating GPX4 in in MS-induced injury cell models. (A) concentration gradient for the GPX4 inhibitor RSL3 by western blotting. (B) Protein expression levels of SLC7A11, GPX4 and ACSL4 in MS-induced injury cell were measured by Western blotting, with β-actin as the internal control. (C) ROS detected by immunofluorescence in MS-induced injury cell model. Green: ROS. Magnification: 40×. Detection of biochemical indicators of GSH (D), MDA (E) and total iron (F) in MS-induced injury cell models. *P < 0.05. **P < 0.01. ***P < 0.001. ****P < 0.0001. MS, mechanical stretch; SLC7A11, Solute carrier family 7 member 11; GPX4, Glutathione peroxidase-4; ACSL4, Acyl-CoA synthetase long-chain family member 4; ROS, Reactive oxygen species; GSH, Glutathione; MDA, malondialdehyde.

Keap1 was a target of miR-125b-5p

Based on the key role of miR-125b-5p in VILI, we further elucidate its potential molecular mechanism in our study. Databases including miRWalk (https://mirwalk.umm.uniheidelberg.de/), TargetScan (https://www.targetscan.org/vert_72/) and miRanda (https://www.microrna. org/microrna/home.Do) were used to predict the target genes of miR-125b-5p (Fig. 9A). What is noteworthy is that Keap1 is a predict the potential target of miR-125b-5p. Subsequently, according to the dual luciferase assay report analysis, to verify the regulatory effect of miR-125b-5p on Keap1. As is shown in Fig. 9B, Overexpression of miR-125b-5p significantly impaired the activity of Keap1-3 ′ UTR wt reporter, but has no effect on the luciferase activity of Keap1-MUT. Namely, Keap1 is a regulative target of miR-125b-5p.

Fig. 9
figure 9

Keap1 was a target of miR-125b-5p. (A) Prediction of the binding site between miR-125b-5p and Keap1 using the online bioinformatics website. (B) Structure and luciferase result of dual luciferase reporter gene. (*****P < 0.0001).

In addition, we further investigated the relationship between miR-125b-5p and Keap1 in vitro and in vivo. Compared with rats transfected with antagomir-NC, rats transfected with miR-125b-5p antagomir expressed higher levels of Keap1, whereas the opposite results were observed in agomir group (Fig. 10A–C). Furthermore, both qRT-PCR and western blotting results confirmed that the expression of Keap1 was increased in the HV group (Fig. 10G–J). However, the results of the in vitro experiment are consistent with t (Fig. 10D–F,K,L). Overall, Keap1 was recognized as the target gene of miR-125b-5p, which negatively regulates the expression of Keap1.

Fig. 10
figure 10

Keap1 was a target of miR-125b-5p. (A,D) Level of miR-125b-5p based on qRT-PCR. (B,C,EJ) Protein and mRNA expression of Keap1, with β-actin as the internal control. (**P < 0.01. ***P < 0.001. ****P < 0.0001). qRT-PCR, real-time quantitative polymerase chain reaction.

Nrf2 and Keap1 signaling pathway is one of the most crucial and classical cell survival and defense pathway that inhibits ferroptosis via regulating downstream gene GXP4. Thus,in our study, to further verify whether miR-125b-5p can alleviate lung tissue injury by targeting Keap1/Nrf2/GPX4 axis in VILI rats, we constructed an overexpression Keap1 plasmid (OE-Keap1) while using miR-125b-5p agomir in rats. After the intervention of miR-125b-5p agomir, the level of miR-125b-5p agomir was significantly upregulated by qRT-PCR, which proved successful transfection, compared with the agomir-NC intervention group (Fig. 11A). Then through qRT-PCR and western blotting analysis, the expression level of Keap1 in the OE-Keap1 group significantly increased, which indicated that Keap1 was successfully overexpressed compared with the OE-NC intervention group (Fig. 11B,C). Moerover, the appearance of the lung tissue, W/D ratio, BALF protein content and HE staining showed that miR-125b-5p agomir could alleviate lung tissue jury (Fig. 11D–G).

Fig. 11
figure 11

MiR-125b-5p inhibits ferroptosis by targeting the Keap1/Nrf2/GPX4 axis, thereby alleviating VILI. (A,B) expression of miR-125b-5p and Keap1 mRNA in the lung tissue of rats in the different groups were determined by qRT-PCR. (C) Protein expression of Keap1 was detected by western blotting, with β-actin as the internal control. (D) Observation of the lung tissue with the naked eye. (E) Wet/dry weight ratio of the lung tissue. (F) Total protein content in BALF. (G) Lung tissue HE staining (200 × and 400 ×) based on observation with a light microscope and lung injury score. (**P < 0.01. ***P < 0.001. ****P < 0.0001). VILI, ventilator-induced lung injury; BALF, bronchoalveolar lavage fluid; HE, Hematoxylin and eosin; qRT-PCR, real-time quantitative polymerase chain reaction.

Western blotting analysis showed that the expression of Keap1 and was significantly upregulated in the HV + agomir NC group, compared with the Control group, while the expression of Nrf2 and GPX4 were significantly downregulated (Fig. 12A,B). In addition, western blotting analysis also showed that the expression of Keap1 and ACSL4 were evidently lower in the miR-125b-5p agomir group, compared with the agomir-NC group, while the expression of SLC7A11 and GPX4 were significantly increased (Fig. 12A). The protein expression of Keap1 and GPX4 were also confirmed by immunofluorescence results (Fig. 12C,D). Besides, compared with the agomir NC group, the expression level of Nrf2 protein in the whole lung tissue and isolated nuclear samples of rats was significantly up-regulated in the miR-125b-5p agomir group (Fig. 12B). In addition, MDA level, total iron content and iron content showed by DAB enhanced Perls Prussian blue staining were decreased in miR-125b-5p agomir group. While the GSH content of miR-125b-5p agomir group was increased (Fig. 12E–H). In contrast, the combined action of miR-125b-5p agomir and OE-Keap1 reversed the protective effect of miR-125b-5p agomir. Collectively, these results suggest that the lung protective of overexpressed miR-125b-5p in VILI rats can be reversed by the overexpression of Keap1.

Fig. 12
figure 12

MiR-125b-5p inhibits ferroptosis by targeting the Keap1/Nrf2/GPX4 axis, thereby alleviating VILI. (A) Protein expression of Keap1, GPX4, SLC7A11 and ACSL4 were detected by western blotting, with β-actin as the internal control. (B) Protein expression level of Nrf2 was measured by Western blotting, with Histone H3 as the internal control. (C,D) Immunofluorescence microscopy analysis of the distribution of fluorescence labeled Keap1 and GPX4 protein. (E) DAB-enhanced Perls’ Prussian blue staining was used to evaluate the content of lung tissue ferric ions. Detection of biochemical indicators of ferroptosis: (F) GSH, (G) MDA, (H) Tissue iron. (**P < 0.01. ***P < 0.001. ****P < 0.0001). VILI, ventilator-induced lung injury; BALF, bronchoalveolar lavage fluid; HE, Hematoxylin and eosin; Keap1, Kelch-like epichlorohydrin-associated protein 1; GPX4, Glutathione peroxidase-4; SLC7A11, Solute carrier family 7 member 11; ACSL4, Acyl-CoA synthetase long-chain family member 4; GSH, Glutathione; MDA, malondialdehyde.

MiR-125b-5p inhibits ferroptosis by targeting the Keap1/Nrf2/GPX4 axis, thereby alleviating MS-induced injury cell

To further elucidate whether miR-125b-5p can alleviate VILI in vitro by targeting Keap1/Nrf2/GPX4 axis, we transfected ATII cells with miR-125b-5p mimic, constructed the overexpressed Keap1 plasmid (OE-Keap1), which was subjected to cyclic MS for 4 h. Subsequently, after the intervention of miR-125b-5p mimic, the level of miR-125b-5p mimic was significantly upregulated by qRT-PCR, which proved successful transfection, compared with the mmic-NC intervention group (Fig. 13A). Meanwhile, by qRT-PCR and western blotting analysis, the expression level of Keap1 in OE-Keap1 group was significantly increased, indicating that Keap1 was successfully overexpressed compared with OE-NC group (Fig. 13B,C). In addition, compared with the MS + mimic NC group, Western blot results showed that the expression of Keap1 and ACSL4 proteins were significantly decreased in the MS + miR-125b-5p mimic group, while the protein expression of SLC7A11and GPX4 were significantly increased. However, due to the combined action of miR-125b-5p mimic and OE Keap1, the expression of the above proteins was reversed (Fig. 13D). In addition, compared with the mimic NC group, the expression levels of Nrf2 protein were significantly upregulated in the total cells and isolated nuclear samples of the miR-125b-5p mimic group. In contrast, the application of OE Keap1 reversed the upregulation of miR-125b-5p mimic (Fig. 13E).

Fig. 13
figure 13

MiR-125b-5p inhibits ferroptosis by targeting the Keap1/Nrf2/GPX4 axis, thereby alleviating MS-induced injury cell. (A,B) expression of miR-125b-5p and Keap1 mRNA in the different groups of MS-induced injury cell models were determined by qRT-PCR. (C,D) Protein expression of Keap1, GPX4, SLC7A11 and ACSL4 were detected by western blotting, with β-actin as the internal control. (E) Protein expression level of Nrf2 was measured by western blotting, with Histone H3 as the internal control. (**P < 0.01. ***P < 0.001. ****P < 0.0001). MS, mechanical stretch; qRT-PCR, real-time quantitative polymerase chain reaction; Keap1, Kelch-like epichlorohydrin-associated protein 1; GPX4, Glutathione peroxidase-4; SLC7A11, Solute carrier family 7 member 11; ACSL4, Acyl-CoA synthetase long-chain family member 4; Nrf2, Nuclear factor-erythroid2-related factor 2.

Similarly, the immunofluorescence expression of ROS, MDA content, and total iron levels were significantly reduced compared with mimic-NC group, while GSH content was significantly elevated. However, ROS, MDA content, and total iron and GSH expressions were reversed by the combination of miR-125b-5p mimic and OE Keap1 (Fig. 14A–D). Together, these findings suggest that the protective effect of miR-125b-5p is to reduce ferroptosis levels by inhibiting Keap1 activation, thereby mitigating MS-induced cell injury.

Fig. 14
figure 14

MiR-125b-5p inhibits ferroptosis by targeting the Keap1/Nrf2/GPX4 axis, thereby alleviating MS-induced injury cell. (A) ROS detected by immunofluorescence in MS-induced injury cell model. Green: ROS. Magnification: 40x. Detection of biochemical indicators of GSH (B), MDA (C) and total iron (D) in MS-induced injury cell models. (**P < 0.01. ****P < 0.0001). MS, mechanical stretch; ROS, Reactive oxygen species; GSH, Glutathione; MDA, malondialdehyde.

To sum up, both in vivo and in vitro experiments have confirmed that, miR-125b-5p can inhibit ferroptosis by activating the Keap1/Nrf2/GPX4 axis, thereby alleviate VILI.

Discussion

MV is a means of advanced life support for patients with acute respiratory distress syndrome (ARDS), especially in the severe situation where various viral pneumonia is prevalent worldwide. Improper use may induce or aggravate lung tissue injury and lead to VILI. Multiple studies have shown that VILI is not only a mechanical trauma, but also a biological trauma35,36. In addition, numerous researchers have confirmed that barrier dysfunction caused by inflammatory cell infiltration, oxidative stress, iron death, etc., is involved in key pathophysiological processes of VILI37,38,39. However, the specific mechanism of ferroptosis in VILI needs further investigation. Our results showed that MV significantly induces the downregulation of miR-125b-5p, and MiR-125b-5p can inhibit ferroptosis in VILI by targeting the Keap1/Nrf2/GPX4 axis.

Ferroptosis is a novel form of cell death, which is distinguished from traditional cell death methods, such as pyroptosis, necrosis, and autophagy40. The accumulation of iron ions and the occurrence of lipid peroxides are characteristic features of iron death41. In addition, the main manifestation of ferroptosis is the production of ROS and lipid peroxides, depletion of GSH, and increase in iron ions, which are closely related to the pathophysiology of tumors, acute liver injury, myocardial ischemia/reperfusion injury, and acute lung injury42,43,44. In addition, another study found that pretreatment of VILI mice with Fer-1 resulted in increased GSH levels, decreased MDA and tissue iron levels, and reduced lung tissue damage23. Notably, our research findings are consistent with this study. Similarly, in our study, the administration of Fer-1 also increased the protein levels of both SLC7A11 and GPX4 in lung cells and tissues, while the protein levels of ACSL4 was decreased. Thus suggesting, that ferroptosis is involved in the process of VILI.Studies have suggested that miR-125b-5p plays a crucial role in various tissue injuries, especially in lung injury45,46,47. Katerina et al. found that the expression of miR-125b-5p was downregulated in the lung tissue in VILI mouse model14. However, the role and mechanism of miR-125b-5p in VILI have not been further studied. Subsequently, in vitro and in vivo models of VILI rat were established in our study, the results reveal that miR-125b-5p was significantly downregulated in VILI. Further studies showed miR-125b-5p could reduce VILI by inhibiting ferroptosis. Thus, this finding may support miR-125b-5p as a potential target for VILI therapy.

To further investigate the specific mechanism of miR-125b-5p mediated ferroptosis in VILI, we used bioinformatics to predict Keap1, a potential target of miR-125b-5p. Former research found that miR-125b-5p is under-expressed, and Keap1 is confirmed the target gene by dual luciferase report analysis, in acute liver failure48. Likewise, emerging study also indicated that Keap1 has been shown to be a target gene of miR-125b-5p in CLP-induced acute lung injury, overexpression of miR-125b-5p can reduce Keap1 expression and alleviate lung injury32. Correspondingly, in our study, we first validated Keap1 as a target gene of miR-125b-5p through dual luciferase reporter gene assay, and confirmed that miR-125b-5p negatively regulates Keap1 expression. Thereafter, miR-125b-5p or both miR-125b-5p and Keap1 were overexpressed in VILI rats. We found that overexpression of miR-125b-5p improved lung tissue pathological injury, decreased the expression of target gene Keap1, and inhibited iron death. However, overexpression of Keap1 reversed this protective effect. Taken together, miR-125b-5p inhibited ferroptosis in VILI rats by targeting down-regulation of Keap1 expression to play a protective role.

GPX4 is a seleniumprotein with antioxidant activity, which is expressed in both cell membrane and cytoplasm25. GPX4 converting toxic lipid peroxides into non-toxic alcohols, exerting antioxidant protection effects49. Previously, it has been found that, inhibition of GPX4 induced ferroptosis in lung cancer50. In addition, Fang J et al. also confirmed that overexpression of GPX4 can reverse hippocampal ferroptosis and synaptic damage in traumatic brain injury (TBI)51. Hence, increasing the expression level of GPX4 activity has become an important pathway to inhibit ferroptosis. In our study, we found the expression of GPX4 was upregulated by overexpression of miR-125b-5p, whereas after the application of GPX4 inhibitor RSL3, the protective effect of miR-125b-5p on ATII cell was reversed. Furthermore, the content of ROS, MDA and ferric ion were also elevated, the amount of GSH was reduced. These data indicate that the use of GPX4 inhibitor RSL3 can reverse the protective effect of miR-125b-5p.

Nrf2, is a transcription factor that maintains intracellular oxidative balance and is activated upon oxidative stress stimulation52,53. Under normal conditions, Nrf2 and its negative regulator Keap1, are rapidly degraded by ubiquitination, and bound in the cytoplasm. When cells are subjected to excessive stress response, Nrf2 is dissociated from Keap1 and transported to the nucleus, further activating the downstream antioxidant gene GPX454,55. Recent study reported that miR-125b-5p also can alleviate inflammation and oxidative stress, thereby inhibit ferroptosis, and alleviate sepsis induced acute lung injury by targeting the Keap1/Nrf2/GPX4 pathway32. This conclusion is consistent with our research findings. Firstly, our research results indicate that miR-125b-5p expression was significantly downregulated in VILI in vitro and in vivo models, while the expression of Keap1 was upregulated. Secondly, we found that overexpression of miR-125b-5p decreased the expression of ferroptosis promoting protein ACSL4, while increased the expression of ferroptosis inhibition protein SLC7A11and GPX4, thereby alleviates VILI. In addition, after the application of GPX4 inhibitor RSL3, the expression of GPX4 was significantly downregulated, the protective effect of miR-125b-5p is counteracted. Finally, in rescue experiment, our data indicates that overexpression of Keap1 leads to increased pathological injury of lung tissue, while the expression of ACSL4 were up-regulated, the expression of SLC7A11, GPX4 and Nrf2 were down-regulated. In addition, the amount of ROS, MDA and iron ion were significantly increased, the level of GSH was significantly decreased, the protective effect of miR-125b-5p is also reversed. Certainly, the results of cell experiment are consistent with those of animal. To sum up, above in vitro ex periments revealed that miR-125b-5p could alleviate MV induced ROS accumulation, oxid oxidative injury and ferroptosis via regulating Keap1/Nrf2/GPX4 axis.

Overall, our results confirm that miR-125b-5p can improve VILI by inhibiting ferroptosis through activation of the Keap1/Nrf2/GPX4 pathway, in vivo and in vitro in rats. However, this study had some limitations. For one thing, the mechanism of miR-125b-5p inhibiting ferroptosis in VILI, by targeting the Keap1/Nrf2/GPX4 pathway has not been confirmed clinically, and further clinical research is necessary. For another, as the pathogenesis of VILI is complex and can be associated with multiple signaling pathways, whether miR-125b-5p can alleviate VILI by mediating ferroptosis in other signaling pathways needs to be further studied.

Conclusion

Collectively, our results suggest that ferroptosis is involved in the occurrence and development of VILI. MiR-125b-5p can activate the Keap1/Nrf2/GPX4 signaling pathway, reduces iron accumulation in vivo and in vitro in VILI rats, enhances antioxidant capacity, and reduces lipid peroxidation, thereby alleviating VILI.

Materials and methods

Animal experiment

60 male Sprague–Dawley (SD) rats (6–8 weeks, 220–280 g) were purchased from Guangdong Weitong Lihua Laboratory Animal Technology Co. ltd. (Guangdong, China). All animals were raised in animal laboratories free of specific pathogens.

60 rats were randomly assigned to 10 groups(n = 6): (1) Control (Spontaneous breathing), (2) HV (High tidal volume mechanical ventilation group), (3) HV + agomir-NC (Negative control), (4) HV + antagomir-NC (Negative control), (5) HV + miR-125b-5p agomir, (6) HV + miR-125b-5p antagomir, (7) HV + DMSO (Dimethyl sulfoxide), (8) HV + Fer-1 (Ferrostain-1), (9) HV + miR-125b-5p agomir + OE-NC (Empty vector transfection control group), (10) HV + miR-125b-5p agomir + OE-Keap1(Overexpression plasmid group). The rats in the intervention group were injected with miR-9a-5p agomir or miR-9a-5-p antagomir, agomir-NC or antagomir-NC through the tail vein as a negative control. Subsequently, a mixture of EntransterTM-in vivo (Engreen biosystem Co, Ltd, Beijing, China) and Keap1 (Shanghai Jikai Gene Technology Co., Ltd, Shanghai, China) plasmid was immediately prepared and administered via the airway, meanwhile, the OE-NC was used as a negative control. The rats were tracheotomized and intubated, and kept breathing autonomously as the control group. The other rats were intubated by tracheotomy and connected to animal ventilators. Rats were intraperitoneally injected with pentobarbital (50 mg/kg, Narcoren, Merial, Germany) plus fentanyl (0.05 mg/kg, Janssen-Cilag, Neuss, Germany) to induce anesthesia; anesthesia is supplemented every hour: pentobarbital (5–10 mg/kg per hour) and fentanyl (2.5–5 μg /kg per hour). After full anesthesia, the rats were mechanically ventilated for 4 h and were euthanized by overanesthesia. The ventilation conditions were set to: TV = 40 ml/kg, breathing frequency = 60 beats/minute. All animal experiments were approved by the Animal Experiment Ethics Committee of Guizhou Medical University (approval number: 20230002, Guizhou, China), and conducted at the Clinical Research Center of Guizhou Medical University. Experiments were conducted with the National Institutes of Health’s (NIH) Guidelines for the Care and Use of Experimental Animals, and followed the Guidelines for ARRIVE.

Cell experiment

The rat ATII cell line was purchased from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). High-sugar Dulbecco Modified Eagle medium (DMEM, Thermo Fisher Scientific, U.S.A.) was added with 10% fetal bovine serum and 1% penicillin/streptomycin respectively, and the cells were cultured in this mixed solution, and continue to culture in a 37 °C incubator containing 5% CO2. Experimental grouping is as follows: (1) Control group (without cyclic stretching group), (2) mechanical stretch group (MS), (3) MS + mimic-NC (Negative control), (4) MS + inhibitor-NC (Negative control), (5) MS + miR-125b-5p mimic, (6) MS + miR-125b-5p inhibitor, (7) MS + DMSO (Dimethyl sulfoxide), (8) MS + Fer-1 (Ferrostain-1), (9) MS + miR-125b-5p mimic + OE-NC (Empty vector transfection control group), (10) MS + miR-125b-5p mimic + OE-Keap1(Overexpression plasmid group), (11) MS + miR-125b-5p mimic + DMSO (Dimethyl sulfoxide), (12) MS + miR-125b-5p mimic + RSL3(Inhibitor of GPX4). The small interfering RNA and plasmid were transfected into ATII cells using Lipofectamine 2000 reagent (Thermo Fisher Scientific, USA), and continue cultivating for 72 h. In addition, 24 h before cyclic stretching, cells were treated with 5 μm of RSL3 (HY-100218A, MCE, USA) in addition to transfection with small interfering RNA. The ATII cells were cyclically stretched with Flexcell PluFX-4000 ™ (Flexcell International Corporation, Burlington, U.S.A.) for 4 h and collected for follow-up experiments.

Histological analysis

After the successful construction of the model, the lung tissue was fixed in 4% formalin solution for 24 h, preserved at 4°, then dehydrated with gradient ethanol and embedded in paraffin. The wax blocks were cut into 4 μm thin slices, stained with hematoxylin–eosin (H&E) (Solarbio, China), the alveolar morphology was observed under a microscope, and the lung tissue pathological injury score was performed according to the Smith score system56.

The accumulation of iron ions in rat lung tissue was detected by DAB enhanced Perls Prussian blue staining. Slices were placed in Prussian blue solution (Servicebio, Wuhan, China) and incubated at room temperature for 30 min. Subsequently, DAB staining solution (Servicebio, Wuhan, China) was added to enhance the staining effect, followed by rinsing, dehydration, and sealing. Finally, digital imaging analysis was performed on the stained sections using an optical microscope.

Lung wet/dry (W/D) weight ratio detection

The upper lobe of the right lung of rats was wet weighted, and then dried in an oven at 65 °C for 72 h. The dry weight was weighed and the dry–wet ratio was calculated.

BALF analysis

The left lung was irrigated with PBS, bronchoalveolar lavage fluid was collected, centrifuged at 1000 g and 4 °C for 10 min, supernatant was obtained, and total protein content was measured with BCA kit (Solarbio, China).

RNA quantification

TRIzol reagent (Invitrogen, San Diego, CA), was selected to extract total RNA from animal tissues and cells, and the extracted RNA was reverse-transcribed into cDNA using the First Strand cDNA Synthesis kit (FSK-101, TOYOBO, OSAKA, Japan) , and Hieff UNICON ®Universal blue qPCR SYBR Green Master mixture (Yeasen, Shanghai, China) was used for fluorescence quantitative polymerase chain reaction and detected in Applied Biosystems (Thermo Fisher, USA). U6 serves as an internal control for miR-125b-5p, while β—actin serves as an internal control for Keap1. All primers (Table 1). were designed and synthesized by Shenggong Biological Engineering Co., LTD. (Shanghai, China). The relative expression of each gene was statistically analyzed by 2−ΔΔCT.

Table 1 Primer sequence for quantitative reverse transcription polymerase chain reaction.

Detection of GSH, MDA and iron content

GSH content was detected by the reduced glutathione assay kit (A006-2-1, Nanjing Jiancheng Biotechnology, China), MDA content of lipid peroxide was detected by the MDA assay kit (BC0025, Solarbio, Beijing, China), and iron ion content was measured by the iron assay kit (BC5135, Solarbio, Beijing, China), a and strictly follow the instructions for testing.

Detection of ROS

Intracellular ROS levels were measured using a 2′,7′-dichlorofluorescein diacetate fluorescent probe (DCFH-DA, G1706, Servicebio, Wuhan, China). After the intervention, the cells were washed with PBS, the DCFH-DA working liquid was added according to the corresponding volume, and the cells were incubated in a CO2 incubator at 37 °C for 30 min in the dark. Subsequently, the DCFH-DA working liquid was removed, washed again with PBS buffer, PBS covered the cells, and finally the images were obtained by fluorescence microscopy.

Dual-luciferase reporter gene assay

Bioinformatics prediction website TargetScan was used to predict the target gene of miR-125 b-5p, and Keap1 was verified as the target gene of miR-125 b-5p by dual luciferase report analysis. The 3′ UTR of synthetic Keap1 was inserted into pSICheck 2 (Han Biotechnology, Shanghai, China) (KEap1-wt), and the mutated form of the possible miR-125 b-5 p binding site was also inserted into pSICheck 2 (Keap1-MUT). The cells were then co-transfected with luciferase vectors using liposome transfection reagent (Han Biotechnology, Shanghai, China). 48 h after transfection, luciferase activity was measured using the dual luciferase reporter Assay System (Beyotime).

Western blotting

The total protein was extracted by adding RIPA lysate (Solarbio, Beijing, China) into the prepared tissues and cells, and the protein concentration was detected by BCA assay kit (Solarbio, Beijing, China). The proteins were then isolated by SDS-PAGE and transferred to polyvinylidene fluoride (PVDF) membrane, which was glued together with 5% skim milk and incubated at room temperature for 1 h, and then incubated with specific primary antibody Keap1(1:4000, Proteintech, Chicago, USA) or Nrf2(1:4000, Proteintech, hicago, SA) or SLC7A11(1:2000, Proteintech, Chicago, USA) or GPX4(1:2000, Proteintech, Chicago, USA) or ACSL4(1:2000, Proteintech, Chicago, USA) overnight in a refrigerator at 4 °C. Then incubate the membrane with a secondary antibody (IgG, 1:10,000, Proteintech, Chicago, USA) at 25 °C for 1 h. Finally, protein bands were visualized using the CLiNX imaging system (Shanghai, China).

Immunofluorescence

Immunofluorescence was employed to evaluate the expression of GPX4 and Keap1 proteins. Paraffin Sections (4 μm) were dried at 60 °C for 60 min, deparaffinized, and hydrated. The slides were immersed in goat serum at 37 °C for half an hour to block non-specific binding sites, followed by overnight incubation with primary antibodies against GPX4 (1:200) and Keap1 (1:200). Subsequently, the sections were warmed to 37 °C for half an hour and reacted with fluorescent-labeled secondary antibodies (1:500) in a humidified chamber for 1 h. After adding DAPI under light-protected conditions, the sections were counterstained for 10 min. Finally, the sections were covered and visualized with a fluorescence microscope.