Suppression of Skp2 contributes to sepsis-induced acute lung injury by enhancing ferroptosis through the ubiquitination of SLC3A2

Sepsis is a life-threatening organ dysfunction caused by a dysregulated host response to infection. The inflammatory cytokine storm causes systemic organ damage, especially acute lung injury in sepsis. In this study, we found that the expression of S-phase kinase-associated protein 2 (Skp2) was significantly decreased in sepsis-induced acute lung injury (ALI). Sepsis activated the MEK/ERK pathway and inhibited Skp2 expression in the pulmonary epithelium, resulting in a reduction of K48 ubiquitination of solute carrier family 3 member 2 (SLC3A2), thereby impairing its membrane localization and cystine/glutamate exchange function. Consequently, the dysregulated intracellular redox reactions induced ferroptosis in pulmonary epithelial cells, leading to lung injury. Finally, we demonstrated that intravenous administration of Skp2 mRNA-encapsulating lipid nanoparticles (LNPs) inhibited ferroptosis in the pulmonary epithelium and alleviated lung injury in septic mice. Taken together, these data provide an innovative understanding of the underlying mechanisms of sepsis-induced ALI and a promising therapeutic strategy for sepsis. Supplementary Information The online version contains supplementary material available at 10.1007/s00018-024-05348-3.


Introduction
Sepsis is a life-threatening disease related to a dysregulated host immune response to infection.The activation of innate immune cells including neutrophils and macrophages, plays a crucial role in the excessive secretion of cytokines and hence the development of cytokine storms during sepsis [1].Cytokine storms trigger the onset of multiorgan damage or dysfunction syndrome, which is the major cause of sepsisrelated death.Sepsis-induced acute lung injury (ALI) is the most common complication and has a high mortality rate if not treated in a timely manner [2].Inflammatory cytokine storms are known to induce programmed cell death and interfere with gas exchange in lung epithelial cells [3], leading to lung injury.However, the essential regulatory components that are responsible for inducing cell death in the alveolar epithelium and the regulatory mechanisms involved in sepsis-induced ALI remain to be determined.
Ferroptosis is an iron-dependent form of programmed cell death caused by the accumulation of lipid peroxides [4], and is induced when the oxidative load in a cell surpasses its internal antioxidant system.The canonical ferroptosis axis is initiated by blocking the import of cystine Zhaoyuan Chen, Jie Zhang and Shenjia Gao contributed equally to this study.via the cystine-glutamate antiporter system Xc − (composed of SLC7A11 and SLC3A2), followed by decreased biosynthesis of reduced glutathione (GSH).A decrease in the levels of GSH, a powerful reductant and a cofactor for glutathione peroxidase 4 (GPX4), impairs GPX4-mediated regulation of lipid homeostasis, thereby disrupting the cellular redox balance and consequently ferroptosis [5].Ferroptosis actively participates in many pathological conditions, such as tumors, neurological diseases, acute kidney injury, and ischemia/ reperfusion [6].Our previous study revealed the occurrence of ferroptosis in alveolar epithelial cells during sepsis and revealed its contribution to the pathogenesis of sepsisinduced ALI [7], suggesting that targeting ferroptosis could be a potential strategy for the treatment of sepsis-induced ALI.However, the underlying mechanisms regulating ferroptosis during sepsis-induced ALI remain unclear.
Varying levels of posttranslational modifications, including ferroptosis, are key to the dynamic regulation of programmed cell death.Ubiquitination is a dynamic and reversible posttranslational modification that could be a promising therapeutic target for treating inflammatory diseases including lung injury [8].It plays an active role in the activation of pro-and anti-inflammatory signaling pathways [1] and modulates the initiation and implementation of cell death pathways [9].A previous study revealed that the E3 ubiquitin ligase MARCHF6 mediates the degradation of key ferroptosis effectors and therefore downregulates ferroptosis, indicating that ubiquitination is intricately linked to the regulation of ferroptosis [10].S-phase kinase-associated protein 2 (Skp2) is an E3 ubiquitin ligase that mediates the targeting of substrates for proteasomal degradation via ubiquitination [11].Furthermore, Skp2 can interact with polyubiquitin chains and thus regulate protein localization, downstream transcriptional events, and pathway activation through its nonproteolytic function [12].However, whether Skp2 plays a role in inducing ferroptosis during sepsis-induced ALI remains to be explored.
In the present study, we utilized patient samples and a septic mouse model to investigate the impact of the absence of Skp2 on sepsis-induced acute lung injury (ALI) through the promotion of ferroptosis.Mechanistically, Skp2 promoted the ubiquitination of SLC3A2, an upstream ferroptosis protein, via a K48-linked nondegradative mechanism and thereby regulated the normal transport of cystine and glutamate.In sepsis, inflammatory cytokine storms suppress Skp2 expression, leading to reduced SLC3A2 ubiquitination and disruption of the translocation of cysteine and glutamate.An increased oxidative stress response and ferroptosis are ultimately triggered in lung epithelial cells.
To further explore the therapeutic potential of Skp2, lipid nanoparticle (LNP) delivery systems [13] were developed to enable the targeted delivery of Skp2 mRNA to murine lungs.The overexpression of the Skp2 protein in the lungs significantly reduced mortality and ameliorated lung injury in septic mice.These results indicate that overexpressing Skp2 through the LNP system could be a potential therapeutic strategy for sepsis-induced ALI.

Cytokine storm during sepsis inhibited Skp2 expression in lung epithelial cells
To determine the expression of Skp2 in sepsis, patient samples and a mice model of sepsis were utilized.The experimental setup and sample collection are shown in Fig. 1A.Briefly, bronchoalveolar lavage fluid (BALF) was collected from septic patients and healthy donors, and C57BL/6 mice were randomly divided into the sham or cecal ligation and puncture (CLP)-induced sepsis groups.First, a notable reduction in the expression of Skp2 mRNA and protein was observed in the BALF of septic patients, compared to that in the BALF of healthy donors (Fig. 1B, C).Epithelial cells accounted for approximately 5% of the cell population in BALF.The number of epithelial cells in BALF tends to increase in response to infection [14].This observation suggests a plausible link between the Skp2 protein and sepsis-induced ALI.Using a CLP septic mouse model, we also observed a significant reduction in Skp2 expression in the lungs of CLP mice (Fig. 1D-F, Supplemental Fig. 1A), compared to that in the lungs of control mice.As reported, the BALF contains a mass of epithelial cells [15], as does was the lung from a mouse model of human disease, and we wondered whether the reduction in Skp2 was mainly in epithelial cells.
To explore this hypothesis, MLE-12 murine lung epithelial cells were used as a cell model.During the course of lung infection, activated alveolar macrophages secrete large amounts of cytokines which further trigger an intricate inflammatory cascade, resulting in subsequent pulmonary epithelial cell death, which is directly related to sepsis-induced ALI [16].Thus, lipopolysaccharide (LPS)activated Raw264.7 cell-conditioned medium (RCM) was used to mimic the in vivo inflammatory conditions of sepsis (Fig. 1A).We first evaluated cytokine levels in RCM and found that after LPS stimulation, macrophages secreted higher levels of IL-6, TNF-α, and MCP-1, while no significant changes were observed in IFN-γ, IL-10, IL-12 or MCP-1 (Supplemental Fig. 1C-I).Since IFN-γ does not serve as the primary factor secreted by macrophages, no detectable alterations were detected.After treatment with RCM, the expression of Skp2 in MLE-12 cells was dramatically reduced, as expected (Fig. 1G-I, Supplemental Fig. 1B).Similar findings were observed in the conditioned medium of the LPS-stimulated murine MH-S alveolar macrophage line (Fig. 1J).Together, these results suggest that inflammatory cytokines may be responsible for the decrease of Skp2 in sepsis.Previous studies have showed that TNF-α [17] or IFN-γ [18] can induce ferroptosis.Although Skp2 downregulation in lung epithelial cells was achieved solely through treatment with TNF-α or IFN-γ (Supplemental Fig. 1J), subsequent experiments involving treatment with RCM were used to better mimic the in vivo impact of the cytokine storm.

Skp2 deficiency induced ferroptosis of alveolar epithelial cells during sepsis
Our previous study [7] provided evidence for ferroptosis in lung epithelial cells during sepsis.Here, we aimed to investigate the impact of Skp2 on pulmonary epithelial ferroptosis and sepsis-induced ALI.Since Skp2 −/− mice have lower body weights than their Skp2 +/− counterparts [19], often die shortly after birth and exhibit physical abnormalities [20], we used heterozygous Skp2 +/− mice in our subsequent experiments.Morphological features of ferroptosis were observed in the lungs of Skp2 +/− CLP mice (Fig. 2A) via electron microscopy, as indicated by mitochondrial shrinkage, increased density of mitochondrial membranes, and compromised membrane integrity [21].At the molecular level, ferroptosis is characterized by the accumulation of intracellular ferrous iron and lipid peroxidation, an increased ratio of GSSG (the oxidative status of GSH) to GSH and decreased expression of GPX4.Therefore, we performed a series of tests to evaluate the redox state and iron accumulation within the lungs.Specifically, we measured ferrous ion concentrations (Fig. 2B), lipid oxidation (MDA quantification) (Fig. 2C), and the GSSG/GSH ratio (Fig. 2D).Our findings revealed a notable increase in these parameters in Skp2 +/− CLP septic mice, compared to those in wildtype CLP mice.Moreover, we observed a significant reduction in the expression of GPX4, an essential enzyme of ferroptosis, in the lungs of the Skp2 +/− mice, compared to the lungs of the wildtype CLP mice (Fig. 2E, F).These data suggest that the depletion of Skp2 is associated with the increase in ferroptosis in sepsis.
To elucidate the role of Skp2 in facilitating ferroptosis of the lung epithelium, Skp2 was overexpressed or knocked down in MLE-12 cells.The Bodipy probe was utilized as an indicator of lipid peroxide levels.This probe allowed for both qualitative visualization (lipid peroxidation in green and reduction in red) and quantitative assessment via flow cytometry.Consistently, treatment with RCM resulted in the induction of ferroptosis, as evidenced by lipid accumulation (Fig. 2G, H) observed using Bodipy probes, along with the downregulation of GPX4 (Supplemental Fig. 2A).The overexpression of Skp2 conferred resistance to RCM-induced ferroptosis.Conversely, siRNA-mediated Skp2 knockdown promoted the peroxidation of intracellular lipids (Fig. 2I,  J) and facilitated ferroptosis in MLE-12 cells (Supplemental Fig. 2C).Similarly, treatment with the Skp2 inhibitor SMIP004 depleted GPX4 (Supplemental Fig. 2D) and induced lipid peroxidation (Supplemental Fig. 2F), thus exacerbating ferroptosis.Taken together, these data indicate that Skp2 deficiency is responsible for inflammatory cytokine-induced ferroptosis in lung epithelial cells during sepsis.

RCM-induced activation of the MEK-ERK pathway mediated Skp2 inhibition in alveolar epithelial cells
Changes in the internal environment during sepsis are mediated by multiple factors, including the MEK/ERK pathway, p38 pathway, and JNK pathway, which are classical pathways activated by inflammation [22].Pathogens or pathogen components activate the three pathways to modulate cellular and humoral immunity.These pathways orchestrate a complex cascade of events involving various mediators that profoundly impact humoral immunity.Stimulation of alveolar epithelial cells with RCM confirmed the activation of three pathways in an inflammatory environment (Fig. 3A).Repression of only the MEK pathway, however, affected on Skp2 expression (Fig. 3B).In previous studies, the MEK-ERK kinase pathway was shown to exert bidirectional regulatory effects on the expression, activity, and stability of E3 ubiquitin ligases [23,24], as well as deubiquitinating enzymes [25].Activation of ERK alone is sufficient to the induce downregulation of Skp2 [26,27].
Fig. 2 Skp2 deficiency induced ferroptosis of alveolar epithelial cells during sepsis.Wildtype and Skp2 +/− C57BL/6 mice were randomly assigned to the sham or CLP group (n = 4 for Western blotting, n = 6 for other analyses).A Pulmonary mitochondrial ultrastructure was visualized by electron microscopy, scale bar = 500 μm.B The Fe 2+ levels in 10 mg lung tissues were examined with iron assay kit.C Lipid peroxidation was examined by measuring the MDA content in the lungs.D The GSSG/GSH ratio in the lung was measured with a GSSG/GSH quantification kit.The relative protein expression of GPX4 and Skp2 in the lungs was measured by western blotting (E) and immunofluorescence staining (F).G, H MLE-12 lung epithelial cells were transfected with Skp2 lentivirus (Skp2-3*Flag-PGK-Puro) to overexpress Skp2 or with control lentivirus, and the transfected cells were then selected with puromycin.G Lipid peroxidation was measured by flow cytometry using Bodipy.H The fluorescent dye Bodipy was used to visualize lipid peroxidation (green) and reduction (red) under a fluorescence microscope, scale bar = 10 μm.I, J Skp2 in MLE-12 cells was knocked down by siRNA, and the cells were treated with RCM for 4 h.I Cells transfected with the lipid peroxidation probe Bodipy were analyzed by flow cytometry for the detection of positive cells.J The fluorescent dye Bodipy was used to visualize lipid peroxidation (green) and reduction (red) under a fluorescence microscope.Scale bar = 10 μm.The data are presented as the means ± standard deviations (*P < 0.05, **P < 0.01, ***P < 0.001)

◂
Our validation revealed that the activation of MEK/ ERK phosphorylation by inflammatory factors leads to the downregulation of Skp2 (Fig. 3A).To establish a comprehensive understanding of this relationship, we modulated the MEK-ERK kinase pathway using various inhibitors and agonists, resulting in distinct outcomes in terms of ferroptosis (Fig. 3C).MEK/ERK inhibitors [28,29] have the potential to rescue Skp2 inhibition and attenuate ferroptosis caused by inflammation (Fig. 3D).Based on the aforementioned findings, our observations suggest that inflammation leads to the inhibition of Skp2 by activating the MEK-ERK pathway, and that the absence of Skp2 intensifies ferroptosis, consequently exacerbating sepsis-induced ALI (Fig. 3E-G).This study revealed the existence of a regulatory cascade in inflammation and Skp2.

Skp2 deficiency aggravated sepsis-induced ALI via the induction of ferroptosis
Compared with the wild-type CLP mice, the Skp2 +/− CLPinduced septic mice exhibited a considerably greater mortality rate (Fig. 4A, B).We then further evaluated the degree of lung injury induced by sepsis.Lung injury is characterized by increased permeability of the alveolocapillary membrane which results in the release of immune cells and proteins into BALF, subsequently leading to edema and diffuse alveolar damage [30].Therefore, we typically assess the protein concentration and number of nucleated cells in BALF, the lung wet/dry ratio, and the extent of alveolar damage [31].As shown in Fig. 4C-E, the absence of Skp2 led to a significant increase in the protein concentration and total cell number in BALF, as well as a decrease in the wet/dry ratio, indicating that Skp2 deficiency exacerbated sepsis-induced ALI.HE staining (Fig. 4F) and TUNEL staining [32] (Fig. 4G, H) further revealed aggravated pulmonary injury and elevated

Skp2 promoted the ubiquitination of SLC3A2 via a K48-linked nondegradative mechanism
As the regulatory function of Skp2 relies on its ability to transfer ubiquitin to binding substrates, we used immunoprecipitation combined with mass spectrometry (IP-MS) to identify potential substrates of Skp2 by utilizing MLE-12 cells overexpressing Skp2 (Fig. 5A).Among all the Skp2binding proteins, SLC3A2 (also known as CD98hc or  5A), a cell-surface transmembrane protein that functions upstream of ferroptosis [33], decreased after the onset of sepsis (Fig. 5B), suggesting that the decrease in SLC3A2 during sepsis may be mediated by Skp2.Furthermore, we observed that Skp2 can ubiquitinate SLC3A2 (Fig. 5C), indicating a potential relationship between SLC3A2 ubiquitination and Skp2.To validate this association, we investigated various types of ubiquitination in cells [34,35] and employed a panel of ubiquitin mutants.Notably, the predominant poly-Ub forms were found to be K48 linked, as evidenced by the nearly abolished coimmunoprecipitation of a K48R Ub mutant with SLC3A2 compared to that of other mutants or wildtype Ub (Fig. 5D).Subsequently, we transfected cells with HA-K48-Ub, HA-K63-Ub, or wildtype Ub to validate the K48-linked ubiquitination of SLC3A2.Notably, positive results were obtained for both K48 Ub and wildtype Ub (Fig. 5E).Moreover, the ubiquitination of SLC3A2 (Fig. 5F) also occurred even in the absence of MG132, indicating that the regulatory effect of Skp2 on SLC3A2 did not rely on ubiquitin-mediated protein degradation.The K48 ubiquitination pathway did not function as a degradation mechanism in the present study [36].
Interestingly, there is crosstalk and interplay between ubiquitin and substrates, with an emphasis on substrate selectivity and regulation under cellular irritation [37].Skp2 promoted Akt ubiquitination and activation, which are required for the mitochondrial localization of Akt [38].The non-degradative mechanism employed by Skp2 regulates the ubiquitination of substrates during the cell cycle [39].In the present study, we observed that the cellular localization of SLC3A2, which colocalized with Skp2 shifted from a dispersed intracellular distribution to an aggregated state on the cell membrane after RCM treatment (Fig. 5G), suggesting that Skp2 regulated SLC3A2 translocation under inflammatory conditions [40].Therefore, we speculated that Skp2 promoted the ubiquitination of SLC3A2 via a K48-linked nondegradative mechanism and led to the translocation of SLC3A2.

Skp2 deficiency triggered ferroptosis by inhibiting SLC3A2 ubiquitination and disrupting cystine/ glutamate exchange
The distinct cellular distribution of SLC3A2 after RCM treatment suggested the potential impact of its intracellular localization on its function as a subunit of the cystine/glutamate exchanger (xCT), one of the three crucial pathways in ferroptosis [41].The xCT complex consists of SLC3A2 and SLC7A11, which allows the extrusion of glutamate and the internalization of cystine [42].Cystine then transforms into cysteine, which is used to synthesize GSH.Dysfunction of SLC3A2 results in the malfunction of xCT, consequently promoting ferroptosis.The overexpression of Skp2 was found to reduce RCM-induced SLC3A2 aggregation (Fig. 6A).Skp2-overexpressing MLE-12 cells exhibited equivalent increases in glutamate discharge and cystine uptake after treatment with RCM (Fig. 6B, C).SZL P1-41 serves as a functional inhibitor of Skp2 by effectively binding to the F-box domain of Skp2, thereby preventing its association with Skp1 and the formation of the Skp2 SCF complex [43].When cells were treated with SZL P1-41 which inhibited the interaction between Skp2 and SLC3A2, the membrane localization and exchange of SLC3A2 were significantly suppressed (Fig. 6A-C), elucidating the crucial functional significance of the ubiquitination of SLC3A2.In contrast, the distribution of SLC3A2 in Skp2-knockdown MLE-12 cells remained comparable to that observed after RCM treatment (Fig. 6D).Furthermore, the inhibition of Skp2 resulted in a significant decrease in glutamate release (Fig. 6E) and cystine uptake (Fig. 6F).These results suggested that Skp2 regulated the localization and functional capacity of SLC3A2 via ubiquitination, thereby regulating the transport of cystine and glutamate and resulting in intracellular lipid peroxidation.Therefore, the suppressed expression of Skp2 in sepsis patients hampered SLC3A2 ubiquitination and the translocation of cystine and glutamate, ultimately leading to ferroptosis.

Skp2 overexpression improves sepsis-induced acute lung injury
To explore the therapeutic potential of Skp2 in septic lung injury, we developed lipid nanoparticles (LNPs) that encapsulate Skp2 mRNA, including its Cap1 structure, Skp2 domain, and poly(A) tail (Fig. 7A).These LNPs were administered via tail vein injection after the mice underwent CLP surgery to overexpress Skp2.After the administration of Skp2-loaded LNPs, Skp2 was successfully expressed in the lungs (Fig. 7B, Supplemental Fig. 6B).The liver exhibited high LNP expression, which was attributed to the preferential accumulation of cationic LNPs in the lungs and liver.As the primary blood-clearing organ, the liver possesses a significant capacity for LNP uptake [44].More importantly, mice injected with LNPs exhibited reduced mortality (Fig. 7C) and ameliorated lung injury (Fig. 7D,  E).Significant improvements in lung tissue injury after the administration of LNPs were further evidenced by reduced cell counts and total protein concentrations in the BALF, as well as reductions in the lung W/D weight ratio (Fig. 7F-H).The degree of ferroptosis in pulmonary epithelial cells was also significantly reduced after LNP infusion, as indicated by the decreased ferrous ion concentration (Fig. 7I), GSSG/ GSH ratio (Fig. 7J), and MDA content (Fig. 7K).The reduction in lung injury was further supported by decreased epithelial cell death in the lungs as shown by TUNEL staining (Fig. 7L, M) and reduced mitochondrial damage (Fig. 7N).These data suggest that overexpressing Skp2 is effective in ameliorating sepsis-induced ALI and highlight the potential of LNPs carrying Skp2 mRNA for the treatment of sepsis.

Discussion
Despite the lack of direct evidence, there are hints suggesting that Skp2 might function as a protective protein in sepsis-induced ALI.As reported in the literature, Skp2 was significantly inhibited during LPS-induced myocardial dysfunction [45], and the upregulation of circEXOC5 during acute lung injury accelerated Skp2 mRNA decay and aggravated sepsis-induced ALI.Consistently, we confirmed the association between Skp2 depletion and the occurrence of sepsis in the present study, showing that the repression of Skp2 is implicated in the progression of sepsis through the regulation of ferroptosis.Obtaining epithelial cells from BALF, especially from healthy donors, can be challenging due to the typically low cell yield.However, BALF provides a valuable and ethically feasible source for studying lung pathology.This allows us to investigate the expression of Skp2 in lung tissue, albeit indirectly, while also addressing the practical and ethical challenges associated with direct lung tissue research.Based on the results of BALF, lung tissue, and pulmonary epithelial cell experiments, we found that Skp2 suppression during sepsis causes changes in the oxidative-reductive system in pulmonary epithelial cells, leading to ferroptosis.Although homozygous knockouts offer robust evidence for causality due to comprehensive gene elimination among individuals, it should be noted that these genetically altered animals occasionally manifest severe phenotypes.Conversely, heterozygous knockouts maintain one functional copy of the target gene, potentially enhancing mouse health and longevity without compromising the intended genetic modification [46].Notably, in our study, Skp2 heterozygous knockout mice displayed a more pronounced lung injury phenotype during sepsis than did wildtype mice.Skp2-mediated ubiquitination of the ferroptosis regulatory protein SLC3A2 could significantly influence its ability to regulate the transport of substances across the cell membrane.This function reveals the unconventional role of the ubiquitination system in the context of lung injury and ferroptosis.Enhancing our understanding of Skp2 could provide insights into the details of ubiquitination and reveal specific mechanisms that regulate protein homeostasis and signal attenuation in biological systems.More importantly, we developed an efficient way to overexpress Skp2 in the lungs of mice and provided evidence that targeting Skp2 could be a potential strategy for treating sepsis-induced ALI through ferroptosis intervention.
In addition to ferroptosis, other types of cell death are also induced by inflammatory cytokine storms in sepsis.To determine whether the decreased expression of Skp2 is related to these processes, we observed the expression of Skp2 in mice or cells undergoing ferroptosis, pyroptosis, and apoptosis (Supplemental Fig. 3A-E).However, when pyroptosis was induced by LPS + CTB [47] or apoptosis was induced by procaspase activating compound-1 (PAC-1) [48], we did not detect lipid oxidation, which is characteristic of lung epithelial cells undergoing ferroptosis induced by Erastin (Supplemental Fig. 3D).In addition, the pan-caspase inhibitor Z-VAD-FMK [49] and the RIP1K inhibitor necrostatin-1 (Nec-1) [50] (Supplemental Fig. 3E) did not prevent the Erastin-induced downregulation of Skp2.
Moreover, the treatment with Fer-1 [51] led to a noticeable increase in the expression of Skp2 in the lungs of CLP mice compared to those in the control group (Supplemental Fig. 4A).Fer-1 attenuated the inhibitory effect of Erastin on Skp2 and GPX4 (Supplemental Fig. 4B).These data suggest that Skp2 downregulation occurs during ferroptosis.Taken together, these data indicate that the inhibition of Skp2 specifically affects ferroptosis of lung epithelium.
Given the intricate mechanism underlying sepsis-induced death of pulmonary epithelial cells [52], we confirmed the specificity of decreased Skp2 expression in ferroptosis.However, it is unclear whether the decreased expression of Skp2 was the cause or the consequence of ferroptosis.The downregulation of Skp2 could be attributed to its involvement in the cascade of inflammatory reactions and cell death.A complex interplay exists between cytokines and chemokines in the context of inflammatory microenvironments [53,54].The MEK-ERK kinase pathway transmits signals from cell surface receptors to transcription factors, thereby regulating gene expression.Depending on the stimulus and cell type, this pathway can either inhibit or induce programmed cell death [55].Ferroptosis, which has been experimentally linked to MEK-dependent oxidative cell death, can be effectively attenuated by the administration of MEK inhibitors [56].The MEK-ERK kinase pathway is a commonly activated in response to infection and cell or tissue damage [57].Moreover, Erastin also facilitated the inhibition of Skp2 via activation of the MEK-ERK kinase pathway (Supplemental Fig. 4C).The association of the MEK-ERK kinase pathway with Skp2 provides mechanistic insight into the cascade of inflammation activation, alteration of target proteins (Skp2 inhibition), and ferroptosis.Through these experiments, we demonstrated the intricate association between Skp2 and ferroptosis in sepsis.However, our study focused on the complex system by which inflammatory factors trigger alveolar epithelial cell death without exploring the specific contributions of individual factors.Future studies could focus on the assessment of various combinations of cytokines to determine their respective roles.
The present study further indicated that Skp2 modulates ferroptosis via ubiquitination.To identify the ubiquitinated substrate of Skp2 and the potential targets involved in ferroptosis, IP-MS was proven to be a reliable approach [58].The Skp2-binding protein SLC3A2 contributes to the modulation of ferroptosis via its interaction with SLC7A11 to sustain the transport of cystine and glutamate [59].However, a pronounced decrease in the expression of both Skp2 and SLC3A2 was observed in the lungs of septic mice, which contradicts the classical function of ubiquitination [60].Interestingly, SLC3A2 ubiquitination was observed even in the absence of MG132-mediated proteasome inhibition, suggesting that K48-linked SLC3A2 ubiquitination is a possible mechanism of signal transduction that might not be exclusively linked to protein degradation.SZL P1-41 has been shown to impede substrate ubiquitination by interacting with Skp2.By utilizing immunofluorescence to observe the intracellular distribution of SLC3A2 and measuring material transport to verify its function, we revealed the intricate interplay between SLC3A2 and the process of ubiquitination and revealed a novel mechanism by which Skp2 and SLC3A2 function in sepsis-mediated ferroptosis.
The complexity of the ubiquitination system arises from the various topologies of ubiquitin chains on substrates [61], which confer upon this system the ability to modulate many signaling pathways that underlie cellular functions [62].The precise impact of K48 ubiquitination on the function of SLC3A2 remains unclear.This effect may occur via an unconventional ubiquitination chain structure [63] that influences the function of SLC3A2.A previous study proposed that the conformation of the K48 ubiquitin chain comprises cyclization and noncyclization.Cyclized ubiquitin chains affect substrate function due to abnormal conformational changes within binding regions [64].K48 ubiquitination in combination with other branches could either enhance or suppress its original effect [65].E3 ubiquitin ligases are capable of mediating K11/K48- [66], K29/ K48- [67], and K48/K63-branched chains [68].Branching might have an important impact on the maintenance of ubiquitin signaling [69].The significance of ubiquitination as a cellular signal that regulates changes in the cellular environment is indicated by its spatial and temporal dynamics [70].Currently, there is a lack of clarity regarding the atypical ubiquitin modification of Skp2 substrates.It is imperative to elucidate the precise regulatory effects of the ubiquitination system on SLC3A2 through topological methods in upcoming studies.
Current treatments for sepsis predominantly target immune cells [71] and inflammatory factors [72], but the efficacy of these interventions has been consistently unsatisfactory.Therefore, the delivery of mRNA to the lungs is a viable alternative strategy for mitigating the consequences of absent or defective Skp2 function and potentially alleviating the impairment of lung function [73].LNPs are an established delivery system for mRNA therapeutics, and they have strong potential for use as an efficacious treatment approach in the future [74].Upon cellular uptake in the lungs [75], the delivered Skp2 mRNA is subsequently released into the cytoplasm and efficiently translated, consequently restoring the functional integrity of membrane transport in alveolar epithelial cells.To date, the efficacy of LNP technology has been enhanced through the regulation of lipid composition and content, while transfection efficiency has improved through the implementation of novel techniques [76].These LNPs satisfy all the requisite conditions for successful delivery systems, including specific Skp2 mRNA modification patterns, proportionate lipids, and the ability to reassemble.The scalable and nonviral characteristics of LNPs present promising potential for serving as a clinically viable therapy for sepsis-induced ALI.
In conclusion, we identified a distinct mechanism by which ferroptosis is regulated by ubiquitination through an extensive investigation of sepsis-induced acute lung injury.Skp2, which functions as an E3 ubiquitin ligase, exhibited noteworthy anti-inflammatory effects on ferroptosis triggered by inflammatory cytokines, and represents a promising approach for the management of sepsis-induced ALI.

Ethics statement
This study was approved by the Ethics Committee of Zhongshan Hospital of Fudan University.The protocol adhered to the principles outlined in the Declaration of Helsinki.
Prior to participation, either the patients themselves or their respective relatives provided written informed consent (Protocol license number: B2021-182R).All animal experiments were carried out in compliance with the guidelines of the Animal Review Committee at Zhongshan Hospital of Fudan University (Protocol license number: 2020-119).

Patients
Biological samples were prospectively collected from 6 patients with sepsis in Zhongshan Hospital of Fudan University from January 2022 to March 2022.Moreover, general data and various examination data of patients were collected.Patients aged 18-80 years who met the Sepsis 3.0 guidelines [77] and ARDS Berlin criteria [78] were included.Exclusion criteria included malignancy, autoimmune disorders, acute cardiovascular and cerebrovascular ailments, a history of intensive care unit treatment, active massive hemorrhage, recent blood transfusion within two weeks, and pregnancy or lactation in females.Six healthy donors were enrolled in this study, all of whom were admitted for evaluation of solitary pulmonary nodules without any evidence of pulmonary infection.Informed consent was obtained from all individual participants included in the study.Bronchoalveolar lavage fluid (BALF) was performed and cells in BALF were collected for analysis.

Isolation of BALF cells
A total of 20 mL of BALF was collected and promptly placed on ice.The BALF was then filtered through a 40 µm nylon cell strainer to remove clumps and debris.Following filtration, the resulting supernatant was centrifuged, and the cells were resuspended in RPMI 1640.The BALF obtained from patients and healthy controls exhibited cell concentrations ranging from 0.5-15 × 10 4 /mL.The cells were subsequently counted and resuspended at a concentration of 10 6 / mL for further experimental procedures.Approximately 5% of epithelial cells were identified within these groups.

Cecal ligation and puncture (CLP) mouse model
C57BL/6 mice (22-25 g, 8-10 weeks, male) were purchased from Shanghai Lab.Animal Research Center.All the animals were housed in a specific pathogen-free facility; the room was maintained at a constant temperature of 22 ± 2 °C.Age-and sex-matched wildtype and Skp2 +/− mice were intraperitoneally injected with 2% pentobarbital sodium for anesthesia.The skin and muscle layers were meticulously separated, revealing the cecum.Using a 4-0 silk thread, ligation was performed two-thirds of the way from the end of the cecum, followed by distal penetration of the cecum using a 22 G needle.Muscles and skin were closed with suture.An initial dose of 1 mL of warmed normal saline was administered immediately after the surgical procedure.To alleviate pain and discomfort, buprenorphine (0.05 mg/kg) was administered subcutaneously for analgesia.The mice were kept on a 37 °C warming pad to sustain their body temperature until they regained consciousness.Antibiotic treatment with imipenem (25 mg/kg) was initiated 2 h after CLP.The model was established according to the Minimum Quality Threshold in Pre-Clinical Sepsis Studies (MQTiPSS) [79].Mice in Sham group underwent a sham laparotomy.Humane endpoints were utilized to determine whether the mice were in a moribund state and met the criteria for euthanasia.These criteria included the inability to stand, agonal breathing, a decrease in body temperature to less than 30 °C for more than 6 h, or excessive loss of body weight.

Survival assays
The mortality rates of the experimental mice in the groups were recorded daily until the end of the seventh day.There were 24 mice in each group.

Cell cultures
The murine macrophage line Raw264.

Lentiviral transfection
A Skp2 overexpressing lentiviral virus was purchased from Genomeditech (gene ID: NM_013787.3,vector: PGMLV-CMV-MCS-3 × FLAG-PGK-Puro).HEK293T cells were incubated overnight, resulting in a cell density of approximately 30-50%.Next, the virus was diluted to an MOI of 50 and added to the cells overnight.The culture medium was replaced with puromycin-supplemented medium at a concentration of 8 μg/mL, and the cells were subsequently maintained for 72 h.The expression of Skp2 was measured using quantitative real-time PCR and western blotting to assess the efficiency of Skp2 overexpression.Skp2 siRNA was purchased from Obiosh, and the sequence of Skp2 siRNA used were 5′-GGG CAA AGG GAG UGA CAA ATT-3′ and 5′-UUU GUC ACU CCC UUU GCC CTT-3′.The siRNA was diluted in Opti-MEM ® medium, followed by the addition of transfection reagent after proper mixing by vortexing and incubation.The mixture was transferred to MLE-12 cells after the medium was changed.The cells were incubated for 48 h, followed by quantification of the mRNA and protein levels at 72 h posttransfection.

Quantitative real-time polymerase chain reaction (qPCR)
Cells in BALF were treated with an RNA extraction kit (Yishan Biotechnology), the RNA was reverse transcribed into cDNA (TaKaRa, Japan), and the target gene was amplified by PCR (TaKaRa, Japan).

Immunoblotting
A total of 30-60 μg of protein per lane was subjected to SDS-PAGE with constant electrophoresis at 90 V until the complete separation of proteins with varying molecular weights was achieved.The proteins were subsequently transferred to PVDF membranes.The bands were developed and analyzed after incubation with primary antibodies and secondary antibodies.Further details about the antibodies used are listed in the key resources table.

Cell slide immunofluorescence
The cells were transferred to a lysine-coated coverslip and plated in a 24-well plate.Following adequate processing and stimulation, the cells were gently rinsed with PBS.The cells were treated with 4% paraformaldehyde (PFA) for 15 min at room temperature.Membrane permeabilization was performed by treatment with 0.05% Triton X-100.Subsequently, the cells were incubated with 3% bovine serum albumin (BSA) in PBS to minimize background staining.After incubation with the primary and secondary antibodies, the slides were sealed with a fluorescence quenching agent containing DAPI.The protocol for triple fluorescence staining was the same as before, up to the step of applying the primary antibody.After overnight incubation with the primary antibody at 4 °C, the corresponding HRP-conjugated secondary antibody was added and the membrane was allowed to incubate.Subsequently, 50-100 μL of TSA-488/555/594 staining working solution was applied to the tissue, ensuring complete coverage.The sample was then incubated at room temperature in the dark for 10 min, and finally sealed with an anti-fluorescence quenching mounting agent.The average fluorescence intensity (mean) was analyzed using Fiji software, where mean was calculated as the total fluorescence intensity (IntDen) divided by the area of the corresponding region (area).

HE staining
The sections underwent were dewaxed by immersion in xylene and ethanol at varying concentrations, followed by hematoxylin staining, differentiation, and washing.Subsequently, eosin dye solution was used for staining, and dehydration was performed before the sections were sealed in sequential ethanol baths with differing concentrations.

Electron microscopy
Lung tissues obtained from mice were sectioned into small pieces.The tissue was fixed in a suitable fixative solution, specifically 2.5% glutaraldehyde in 0.1 M phosphate buffer.Subsequently, the samples were incubated in an electron microscope fixation solution, and then subjected to further incubation in 1% osmic acid for 1-2 h at room temperature.The dehydrating process was carried out using varying concentrations of alcohol, with 20-min incubations at each concentration.The tissues were continuously incubated in 100% acetone, with each incubation lasting 15 min.Following embedding, the plate was incubated in a 60 °C temperature-controlled box for 48 h, after which the resin block was sliced using a microtome.Subsequently, the sections were stained with uranium acetate saturated alcohol solution and lead citrate solution to enhance contrast.After staining, the sections were carefully cleaned, allowed to dry overnight, and subsequently analyzed using a transmission electron microscope (TEM) at an appropriate magnification.

TUNEL staining
The frozen sections were thawed at 37 °C to eliminate moisture before being fixed with 4% paraformaldehyde for 30 min.To facilitate antigen repair, protease K was added, and the samples were incubated.Subsequently, a membrane permeabilization working solution was added and incubated.A mixture of the TDT enzyme, dUTP, and buffer was then prepared according to the instructions and added to the tissues.DAPI was used to stain the nuclei in the dark.The sections were sealed and imaged under a microscope.

Lung wet-to-dry ratio
The procedure involved dissecting and harvesting the right lung, followed by measuring its weight (wet weight).Subsequently, the lung was dried overnight at 60 °C to obtain the dry weight.The wet weight/dry weight ratio was calculated by dividing the wet weight by the dry weight.

Cell count and protein concentration in BALF
A small incision was made in the neck to expose the mouse trachea.A cannula was inserted into the trachea and securely fixed in place.A gentle instillation of 1 mL of sterile saline into the lung was performed through the cannula.The fluid was then aspirated back into the syringe, ensuring complete recovery.The BALF was transferred to a sterile tube and centrifuged to pellet the cells.The supernatant was removed for protein concentration analysis using the BCA assay.The cells were resuspended in PBS, and cell counting was conducted using a hemocytometer.

Quantification of cellular GSSG/GSH levels
Ten milligrams of mouse lung tissue was weighed and washed in precooled PBS.A GSSG/GSH Quantification Kit II (G263) was used according to the instructions of Dojindo.The absorbance was measured with a microplate system, and the GSH and GSSG concentrations were calculated.

Malondialdehyde (MDA) quantification
First, 10 mg of mouse lung tissue was weighed and washed with precooled PBS before preparing the lysis solution.Ultrasonic cell disruption may be utilized for lysis, followed by subsequent centrifugation at a rate of 13,000g/10 min to glean the resulting supernatant.A standard lipid curve was then prepared, along with the necessary sample formulations, following the recommended protocol for Abcam lipid peroxidation (MDA) assay (ab118970).

Concentration of ferrous ions
The content of ferrous iron was determined using an Abcam iron assay kit (ab83366).Ten milligrams of lung tissue was weighed and washed with cold PBS.Iron assay buffer was added, and the tissue was homogenized at 16,000 g for 10 min.Gradient concentrations of standard and test samples were then added to a 96-well plate, and assay buffer and probe were added to each well.The absorbance was measured at OD593 nm, and concentration of ferrous ions in the test samples was calculated.

Analysis of lipid peroxidation
For flow cytometry analysis, a Bodipy probe (Invitrogen™ D3861) was added to the cell medium at a final concentration of 2.5 µM.The cells were then collected and digested with pancreatic enzymes, and confirmed by the flow cytometry.
Lung epithelial cells were treated with confocal glass bottom dishes for immunofluorescence using fluorescence staining.Bodipy probe was added at a final concentration of 5 μM.Immunoprecipitation and mass spectrometry were then performed, with fluorescence imaging carried out using 488 nm and 594 nm microscopic wavelengths, after washing with PBS.

Apoptosis analysis
Treated MLE-12 cells were dissociated using EDTA-free pancreatic enzymes.Thereafter, 300 μL of binding buffer was added to tubes, followed by the addition of 5 μL of Annexin V-FITC.The cells were thoroughly mixed and incubated for 15 min.Finally, 5 μL of PI was added for staining, followed by a 5-min suspension prior to machine analysis.

Immunoprecipitation mass spectrometry
IP-MS cell lysis buffer containing PMSF was added to a Petri dish, mixed thoroughly, and subsequently incubated on ice for a period of 20 min.The supernatant was collected, and an anti-Flag antibody was added to the supernatant overnight.Then, protein A/G magnetic beads were introduced into EP tubes containing the aforementioned antibodies and subsequently incubated overnight at 4 °C.IP-MS wash buffer was used to wash the magnetic beads.Subsequently, loading buffer was added to the EP tube and incubated in a metal bath at 100 °C for 10 min.A magnetic rack enabled the separation of the magnetic beads and subsequent absorption of the supernatant liquid containing the target protein.
After electrophoresis, the proteins were silver stained, the protein bands were excised, and the samples were subjected to mass spectrometry analysis.

Ubiquitination experiment
The following plasmids were purchased from Addgene: HA-Ub (17608), HA-Ub-K48 (17605), HA-Ub-K48R (17604), and HA-Ub-K63 (17606).Myc-SLC3A2, Flag-Skp2 and other ubiquitin mutants were obtained from Zhizhang Wang (Department of Pathology, Nanfang Hospital, Southern Medical University, Guangzhou, Guangdong, China).In accordance with the experimental aim, 20 μM MG-132 was administered or not to the cells.Prechilled lysis buffer containing protease inhibitor was used for cell lysis and collection of the supernatant.A portion of the wholecell lysate was retained as a control, while the rest of the protein lysate was incubated overnight with 20 µL of beads.Then, 100 μL of loading buffer was added, and the samples were incubated at 100 °C and frozen.The Western blot protocol was followed.The concentration of the separation gel that was used corresponded to the molecular weight of the target protein.A corresponding labeled antibody was used to detect the target protein, and an anti-HA antibody was used to detect ubiquitination of the target protein.

Cystine absorption experiment
Cystine absorption experiment was determined using Dojindo Cystine Uptake Assay Kit (UP05).The black 96-well plates were seeded with cells and incubated overnight.The cystinefree medium was replaced with medium containing Cystine Analog, and the cells were incubated at 37 °C for 30 min.Then, 50 μL methanol was added to each well, 200 μL of the prepared working solution was added, and the plates were sealed to prevent evaporation.The plates were incubated for 30 min.A microplate reader with fluorescence detection capabilities was utilized to measure the fluorescence intensity of the labeled cells.The fluorescence intensity was assessed at an excitation wavelength of 490 nm and an emission wavelength of 535 nm.To determine the specific fluorescence intensity resulting from the cellular uptake of the cystine analog, the values measured in the sample wells were subtracted from the values detected in the blank wells.

Glutamate concentration measurement
Glutamate concentration was measured using Dojindo Glutamate Assay Kit (G269).The 96-well plates were seeded with cells and incubated overnight.Serum-free medium without cystine was warmed to 37 °C and added to the wells.Next, cystine medium was added at 37 °C for 30 min, followed by the addition of three washes with cold PBS.Methanol (50 μL) was added to each well, followed by 200 μL of the prepared working liquid.The wells were sealed with sealing plate film to prevent evaporation of the liquid and incubated for 30 min.The fluorescence intensity of each well was measured using a fluorescence spectrophotometer at wavelengths of 490 nm and 535 nm, and the results were subsequently calculated.
The components of the lipid nanoparticles were dissolved in 100% ethanol.The molar ratio of the LNP components ionizable lipids/DSPC/cholesterol/DMG-PEG2000 was 50/10/38.5/1.5.The mRNA was dissolved in a 50 mM acetic acid buffer (pH 4.5), and the aqueous phase (mRNA) and organic phase (LNP component) were mixed by Precision Nanosystems nanoassembly at a 3:1 ratio.The LNPs were collected and diluted in PBS (pH 7.2).A 10 kDa Slide a-Lyzer G2 Dialysis Cassette was used to dialyze the LNPs overnight at 4 °C.After filtering through a 0.2 mm sterile filter, the final RNA load concentration was 598 ng/μL.The LNPs were stored at 2 °C to 8 °C.The particle size and polydispersity of LNPS were measured by a Malvern Zetasizer dynamic light scattering (DLS) instrument.RNA packaging efficiency was measured with a Quant-it RiboGreen RNA assay kit.

Fig. 1
Fig. 1 Cytokine storm during sepsis inhibited Skp2 expression in lung epithelial cells.A Experimental setup and sample collection (generated by Biorender).BALF samples were collected from septic patients and healthy donors (n = 6).C57BL/6 mice were randomly divided into the sham or CLP sepsis groups (n = 4).The murine monocyte/macrophage line Raw264.7 was stimulated with LPS and the culture medium was collected after 12 h (from Raw264.7 cellconditioned medium, RCM).MLE-12 murine lung epithelial cells were treated with RCM or LPS for 4 h.B, C The mRNA and protein expression of Skp2 in the BALF of septic patients and healthy donors were measured by qPCR (B) and western blotting (C).D-F Lung tissue of mice was collected 24 h after CLP surgery, and the protein expression of Skp2 in the lungs was measured by western blotting (D) or immunofluorescence staining (E, F).Scale bar = 20 μm.G-I The

Fig. 3
Fig. 3 RCM-induced activation of the MEK-ERK pathway mediated Skp2 inhibition in alveolar epithelial cells.A MLE-12 cells were treated with RCM, and changes in Skp2 and the MEK/ERK pathway, the p38 pathway, and the JNK pathway were measured by western blotting.B MLE-12 cells were subjected to treatment with RCM and specific inhibitors targeting the MEK, p38, and JNK signaling pathways.Alterations in protein expression levels in MLE-12 cells were assessed using western blot analysis.C MLE-12 cells were treated with RCM and a specific activator targeting MEK/ERK (C16-PAF and (rel)-AR234960), and the protein levels were measured by western blotting.D MLE-12 cells were treated with RCM and/or a MEK/

Fig. 4
Fig.4Skp2 deficiency aggravated sepsis-induced ALI via the induction of ferroptosis.A Experimental setup for C57BL/6 (generated by Biorender).Wildtype and Skp2 +/− C57BL/6 mice were randomly divided into the sham, CLP and CLP + Fer-1 groups, C57BL/6 mice were intravenously injected with Fer-1 (3 mg/kg).n = 24 for survival analysis, n = 6 for other analyses.Survival curve analysis was conducted to evaluate the 7 day survival rate of the mice, while the remaining measurements were carried out at the 24-h time point.B

Fig. 5
Fig. 5 Skp2 promoted the ubiquitination of SLC3A2 via a K48linked nondegradative mechanism.A MLE-12 cells overexpressing Skp2 were subjected to immunoprecipitation-mass spectrometry (IP-MS).The identified proteins were annotated with GO terms.B The expression levels of Skp2, GPX4 and SLC3A2 in the lungs of wildtype/Skp2 +/− Sham and CLP mice were measured by western blotting.C HEK293T cells were transfected with Flag-Skp2, Myc-SLC3A2, or HA-Ub and treated with MG132, and the IP samples and whole cell lysates were analyzed by immunoblotting.D HEK293T

Fig. 6 ◂ Page 12 of 22 Fig. 7
Fig. 6 Skp2 deficiency triggered ferroptosis by inhibiting SLC3A2 ubiquitination and disrupting cystine/glutamate exchange.A-C MLE-12 mouse lung epithelial cells were transfected with Skp2-3*Flag-PGK-Puro to overexpress Skp2, and Puro Lentivirus transfected MLE-12 cells were used as controls.A EpCAM functions as an indicator of epithelial cells and was used to indicate the interplay between SLC3A2 and the alveolar epithelium.The distributions of Skp2, SLC3A2 and EpCAM were detected by immunofluorescence staining.Scale bar = 20 μm; B, C transfected MLE-12 cells were stimulated with RCM or Skp2 inhibitor SZL P1-41.B The glutamate content in the cell culture supernatant was analyzed with a glutamate assay kit.C Cystine uptake by MLE-12 cells was analyzed with a cystine uptake kit.D-F MLE-12 cells were transfected with Skp2 siRNA to knock down Skp2.Scrambled siRNA was used as a control.D The distributions of Skp2 and SLC3A2 were detected by immunofluorescence staining in MLE-12 cells transfected with Skp2 siRNA, scale bar = 20 μm; E, F transfected cells were stimulated with RCM or control medium (Con.).E The glutamate content in the medium of MLE-12 cells transfected with Skp2 siRNA and stimulated with RCM was analyzed by a glutamate assay kit; F the cystine uptake of MLE-12 cells transfected with Skp2 siRNA and stimulated with RCM was determined with a cystine uptake detection kit.The data are presented as the means ± standard deviations (ns P > 0.05, *P < 0.05, **P < 0.01, ***P < 0.001) ◂

Fig. 8
Fig. 8 Skp2 regulates ferroptosis by modulating the distribution of SLC3A2 and affecting cystine and glutamate transport (schematic diagram by Biorender).Skp2 mediates K48-linked ubiquitination but not SLC3A2 degradation, which is essential for the normal transportation of cystine and glutamate antiporters through the cell membrane.In sepsis, inflammatory cytokine storms suppress Skp2 expression, leading to reduced SLC3A2 ubiquitination and hence hindering