YOD1 promotes ferroptosis in acute lung injury by deubiquitination of NCOA4

Abstract

Acute lung injury (ALI) is a prevalent critical condition for which effective treatments remain elusive. Ferroptosis plays a significant role in the pathophysiology of ALI. The deubiquitinating enzyme YOD1 is implicated in the regulation of infectious diseases; however, its specific role in ALI and ferroptosis is not yet fully understood. In this study, we observed that YOD1 expression was notably elevated in the lung tissue and primary alveolar type II (ATII) cells of mice subjected to lipopolysaccharide (LPS)-induced ALI. Moreover, YOD1 deficiency significantly mitigated ferroptosis and damage to the alveolar epithelial barrier. Mechanistically, YOD1 interacts directly with NCOA4 through its OTU domain, inhibiting K48-linked ubiquitination at the K343/K353 lysis residue of NCOA4, thus facilitating NCOA4-mediated autophagic degradation of FTH1 and promoting ferroptosis. Overall, our findings indicate that YOD1 regulates NCOA4 via deubiquitination, suggesting it may serve as a potential therapeutic target for ALI.

This study reveals that the deubiquitinase YOD1 is upregulated in LPS-induced acute lung injury (ALI) and exacerbates ferroptosis by deubiquitinating and stabilizing NCOA4, thereby promoting NCOA4-mediated ferritinophagy. YOD1 deficiency alleviated lung injury, suggesting YOD1 as a potential therapeutic target for ALI.

Introduction

Acute lung injury (ALI) /acute respiratory distress syndrome (ARDS) is a severe and diffuse lung injury resulting from various factors, such as pulmonary or non-pulmonary infection, trauma, aspiration, or shock¹. Approximately 10% of critically ill patients develop ARDS, with a mortality rate ranging from 30% to 50%². Several drugs, including statins^3,4, corticosteroids⁵, acetaminophen⁶, IL-1 receptor antagonists⁷, and anti-IL-6 therapies⁸, have been investigated in clinical trials for ARDS; however, none have proven universally beneficial. The pathophysiology of ARDS involves the disruption of the alveolar-capillary epithelial barrier, impaired alveolar fluid clearance, dysregulated immune responses, and oxidative and mechanical stress. Alveolar epithelial damage exacerbates the inflammatory and pro-coagulation cascades, disrupts the alveolar epithelial barrier, depletes ATII cells, and exacerbates pulmonary edema and atelectasis¹. Thus, understanding the mechanisms underlying alveolar epithelial injury and strategies to restore an effective alveolar epithelial barrier is of critical importance.

Ferroptosis is a regulated form of cell death linking lipid metabolism, iron homeostasis disruption, and redox biology⁹. Studies have shown that ferroptosis plays a significant role in ALI¹⁰. LPS causes cell injury, increases ROS production, iron and MDA concentrations, reduces GSH levels, and the expression of SLC7A11, GPX4, and FTH1, both in vivo and in vitro^11,12. Inhibition of ferroptosis in mouse models was found to alleviate LPS- and sepsis-induced ALI¹². Ferritinophagy, mediated by nuclear receptor coactivator 4 (NCOA4), is a newly recognized autophagic cargo receptor¹³. Ferritin, composed of heavy (FTH1) and light (FTL) isoforms, acts as a cytosolic iron storage complex, chelating up to 4500 iron atoms. By sequestering redox-active iron, ferritin plays an important antioxidant role in cells¹⁴. Recent studies indicate that ferritin is a key regulator of ferroptosis, affecting susceptibility in both in vitro and in vivo models^15–17. In ferritinophagy, NCOA4 binds FTH1 and transports iron-bound ferritin into autophagosomes for lysosomal degradation and iron release. Research shows that NCOA4 knockdown inhibits ferritinophagy, reduces lipid peroxidation, and blocks ferroptosis by lowering bioavailable intracellular labile iron pools^18,19. It is worth noting that NCOA4-mediated ferritinophagy played an important role in septic ARDS²⁰, which suggests that the NCOA4–FTH1 axis and ferritinophagy may be a potential therapeutic target for ALI.

YOD1, also known as OTUD2, is a member of the ovarian-tumor-related deubiquitinating enzyme family (OTUDs). These enzymes reverse protein ubiquitination by removing ubiquitin (Ub) residues²¹. Research indicates that YOD1 regulates critical cellular processes, including proliferation, apoptosis, migration, and invasion²². Aberrant YOD1 expression is strongly associated with the pathogenesis of multiple malignancies, notably breast²³, cervical²⁴, and ovarian cancers²⁵. Furthermore, YOD1 serves a critical function in non-malignant contexts, contributing to vascular endothelial injury and inhibition of antiviral innate immune response in conditions such as vascular disorders and inflammation^26,27. However, YOD1 has also been shown to inhibit inflammation and alleviate DIC caused by sepsis²⁸, one of the most common causes of ALI. Until now, whether YOD1 plays a role in ALI is still unclear.

In this study, we examined the involvement of YOD1 in the ferroptosis of alveolar epithelial cells during LPS-induced ALI. We found that YOD1 promotes LPS-induced ALI by directly binding to NCOA4 and inhibiting its ubiquitination, thereby facilitating ferroptosis in alveolar epithelial cells. These findings provide insights into the role and mechanism of YOD1 in ALI, highlighting its potential as a therapeutic target for mitigating lung injury.

Results

YOD1 is involved in LPS-induced acute lung injury

Initially, we utilized the publicly available GEO database to investigate ubiquitination and deubiquitinating enzymes associated with acute lung injury. We analyzed the datasets GSE65682 and GSE216943, identifying 82 shared differentially expressed genes (DEGs) between the control (Con) and ARDS/ALI groups. After intersecting these with ubiquitinating and deubiquitinating genes, we identified 10 genes potentially involved in acute lung injury (Figs. 1A, B and S1A). Among these, the change of the YOD1 gene was the most significantly enriched (Figs. 1B and S1A). In further study, YOD1 expression was found to increase in both lung tissues and primary ATII cells of mice in a time-dependent manner following LPS exposure (Figs. 1C, D, and S1B). These findings suggest that YOD1 may be involved in LPS-induced acute lung injury.

Fig. 1. YOD1 is involved in LPS-induced acute lung injury.

A A Venn diagram illustrating the common differentially expressed genes (DEGs) in lung tissues of normal and LPS-damaged mice (data from GSE65682 and GSE216943) was generated, and the intersecting genes related to ubiquitination were analyzed to identify those that may play a role in ALI. B A heatmap of co-expressed ubiquitination-related genes is presented. C C57BL/6 mice were exposed to LPS for 6, 12, 18, and 24 h to assess YOD1 expression in lung tissues (n = 3 mice/group). **P < 0.01, unpaired t-test. D Primary alveolar type II (ATII) cells were extracted from mice exposed to LPS for 6, 12, 18, and 24 h to evaluate YOD1 expression in these primary cells (n = 3 mice/group). **P < 0.01, unpaired t-test. E Adeno-associated virus AAV6 carrying an SPC-specific promoter was utilized to achieve alveolar type II (ATII)-specific knockdown of the YOD1 gene in 6-week-old C57BL/6 mice. The efficiency of the knockdown was validated using a Western blot assay (n = 3 mice/group). **P < 0.01, unpaired t-test. F Immunofluorescence was employed to confirm ATII-specific YOD1 gene knockdown (n = 3). G Representative images of HE staining of mouse lung tissues are presented alongside the lung injury score, with a scale bar of 100 μm (n = 6 mice/group). ***p  <  0.001, two-way ANOVA followed by Tukey’s multiple comparison test. H The partial pressure of arterial oxygen in the mice was measured (n = 6 mice/group). **P < 0.01, ***p  <  0.001, two-way ANOVA followed by Tukey’s multiple comparison test. I The wet-to-dry weight ratio (W/D) of the lungs was calculated (n = 6 mice/group). *P < 0.05, ***p  <  0.001, two-way ANOVA followed by Tukey’s multiple comparison test. J The protein content in BALF was determined (n = 6 mice/group). ***p  <  0.001, two-way ANOVA followed by Tukey’s multiple comparison test. K Neutrophil cell counts in BALF were recorded (n = 6 mice/group). ***p  <  0.001, two-way ANOVA followed by Tukey’s multiple comparison test. L The protein expression levels of Claudin 3, E-cadherin, Occludin, and ZO-1 were evaluated using Western blot analysis (n = 3 mice/group). All data were presented as mean ± SEM.

To further investigate the role of YOD1 in LPS-induced ALI, we performed ATII-specific YOD1 gene knockdown using an adeno-associated virus (AAV6) carrying an SPC-specific promoter (Fig. 1E, F). Our results demonstrated that YOD1 knockdown significantly mitigated LPS-induced lung tissue damage and improved lung injury scores in mice (Fig. 1G). Furthermore, YOD1 knockdown attenuated the reduction of oxygen partial pressure, the increase of the lung wet–dry ratio, and the alveolar protein exudation induced by LPS (Fig. 1H–J). Additionally, YOD1 knockdown reduced LPS-induced inflammatory cell infiltration and the release of cytokines IL-1β, IL-6, and TNF-α (Figs. 1K, and S1C, D). We also observed a significant decrease in the expression of Claudin 3, E-cadherin, Occludin, and ZO-1 after LPS injury, indicating epithelial barrier dysfunction. However, YOD1 knockdown significantly restored the expression of these cell adhesion proteins and repaired the epithelial barrier (Fig. 1L). These findings suggest that YOD1 knockdown alleviates LPS-induced ALI and protects the alveolar epithelial barrier, highlighting the important role of YOD1 in ALI.

YOD1 promotes LPS-induced alveolar epithelial cell injury

To further investigate the role of YOD1 in LPS-induced alveolar epithelial cell injury, YOD1 was knocked down in HPAEpiC and MLE-12 (Fig. S2A and B). Our findings indicate that YOD1 knockdown significantly reduced the release of cytokines IL-1β, IL-6, and TNF-α in LPS-stimulated alveolar epithelial cells (Fig. 2A, B). Additionally, LPS-induced alveolar epithelial cell death and LDH release occurred in a time-dependent manner, which was obviously alleviated by YOD1 knockdown (Fig. 2C–F).

Fig. 2. YOD1 promotes LPS-induced alveolar epithelial cell injury.

A, B An ELISA assay was conducted to measure the concentrations of IL-1β, IL-6, and TNF-α in the supernatants of HPAEpiC and MLE-12 cells (n = 6 per group). ***p  <  0.001, two-way ANOVA followed by Tukey’s multiple comparison test. C, D The CCK-8 assay was employed to evaluate the viability of HPAEpiC and MLE-12 cells (n = 6 per group). *p  <  0.05, ***p  <  0.001, two-way ANOVA followed by Tukey’s multiple comparison test. E, F LDH release from HPAEpiC and MLE-12 cells into the medium was assessed (n = 6 per group). *p  <  0.05, **p  <  0.01, two-way ANOVA followed by Tukey’s multiple comparison test. G, H The transfection efficiency of YOD1 overexpression by lentiviral assay in HPAEpiC and MLE-12 cells was validated using Western blot analysis (n = 3). I, J The concentrations of IL-1β, IL-6, and TNF-α in BALF from YOD1-overexpressing HPAEpiC and MLE-12 cells (n = 6 per group). **p  <  0.01, ***p  <  0.001, two-way ANOVA followed by Tukey’s multiple comparison test. K, L The cell viability of YOD1-overexpressing HPAEpiC and MLE-12 cells (n = 6 per group). **p  <  0.01, ***p  <  0.001, two-way ANOVA followed by Tukey’s multiple comparison test. M, N LDH release was evaluated in YOD1-overexpressing HPAEpiC and MLE-12 cells (n = 6 per group). ***p  <  0.001, two-way ANOVA followed by Tukey’s multiple comparison test. O YOD1-overexpressing HPAEpiC and MLE-12 cells were treated with inhibitors of ferroptosis (Fer-1 and Lip-1), necroptosis (NSA), and apoptosis (Z-VAD) for 48 hours, after which cell viability was measured using the CCK-8 assay 24 hours post-LPS administration (n = 6 per group). **p  <  0.01, ***p  <  0.001, two-way ANOVA followed by Tukey’s multiple comparison test. P YOD1-overexpressing HPAEpiC and MLE-12 cells were treated with ferroptosis inhibitors (Fer-1 and Lip-1) for 48 h, and LDH release into the medium was detected 24 h after LPS administration (n = 6 per group). ***p  <  0.001, two-way ANOVA followed by Tukey’s multiple comparison test. All data were presented as mean ± SEM.

Programmed cell death, including apoptosis, pyroptosis, and ferroptosis, participated in ALI^29–31. To elucidate the cell death pathway associated with YOD1, we overexpressed YOD1 in alveolar epithelial cells (Fig. 2G, H). The overexpression of YOD1 exacerbated the LPS-induced increasement of cytokine levels, cell death, and LDH release in alveolar epithelial cells (Fig. 2I–N). Furthermore, when apoptosis inhibitors (Z-VAD), pyroptosis inhibitors (NSA), and ferroptosis inhibitors (Fer-1 and Lip-1) were administered to YOD1-overexpressing alveolar epithelial cells, only the ferroptosis inhibitors significantly reversed YOD1-induced cell death and LDH release in HPAEpiC and MLE-12 (Fig. 2O, P). These results indicate that YOD1 promotes LPS-induced injury in alveolar epithelial cells, which was largely associated with ferroptosis.

YOD1 knockdown alleviates LPS-induced alveolar epithelial cell injury through regulation of ferroptosis

To determine whether ferroptosis is a key mechanism associated with YOD1, we further investigated indicators of ferroptosis in alveolar epithelial cells. Our findings revealed that the levels of lipid peroxidation markers, including LPO and MDA, were significantly increased following LPS injury, along with elevated levels of H₂O₂. Notably, YOD1 knockdown effectively reversed these increases in both HPAEpiC and MLE-12 cells (Fig. S3A and B). The benefits of YOD1 knockdown on ferroptosis were further confirmed in ferroptosis agonist (Erastin or FINO2) -induced injury. Consistent with previous studies, Erastin and FINO2 led to elevated levels of lipid peroxidation markers, including LPO, MDA, and H₂O₂, all of which were significantly suppressed by YOD1 knockdown (Fig. S3C and D). Additionally, YOD1-deficient alleviated Erastin and FINO₂-induced epithelial barrier damage (Fig. S3E and F). GPX4 and ACSL4 are key proteins involved in GSH-related and lipid peroxidation-related ferroptosis. YOD1 overexpression reduced the expression of GPX4 and increased ACSL4, at both gene and protein levels in LPS-treated cells (Fig. S3G–J). On the contrary, YOD1 knockdown increased GPX4 expression and suppressed ACSL4 expression in LPS-treated mice (Fig. S3K).

Iron overload is another feature of ferroptosis. We found Fe²⁺ levels and total intracellular iron were significantly elevated in alveolar epithelial cells after LPS injury, which was mitigated by YOD1 knockdown (Fig. 3A–D). Similarly, the increases in Fe²⁺ levels and total iron induced by both Erastin and FINO2 were effectively prevented by YOD1 knockdown (Fig. 3E–H). These results indicate YOD1 was an effective target for the treatment of LPS-induced ALI by suppressing iron overload in alveolar epithelial cells. To elucidate how YOD1 promotes intracellular iron accumulation, the genes and proteins that regulate iron storage and transport were further examined. Transferrin (TF) and transferrin receptor (TfR)-mediated iron uptake, and ferroportin 1 (FPN1)-mediated iron export; however, the gene and protein expression of which were not affected by LPS and YOD1 overexpression (Fig. 3I–L). Ferritin is essential for intracellular iron storage. The gene expression level of ferritin heavy chain 1 (FTH1), ferritin light chain (FTL), and NCOA4 was unchanged regardless of YOD1 overexpression or knockdown (Fig. 3I, J); however, significant decreases in protein expression level of FTH1 and FTL, alongside elevated protein level of NCOA4, were observed after YOD1 overexpression (Fig. 3K, L). These results suggest that YOD1 may influence the expression levels of FTH1 and NCOA4 at the protein level.

Fig. 3. YOD1 knockdown alleviates LPS-induced ferroptosis in alveolar epithelial cells.

A, C Fluorescent staining of Fe²⁺ (red) and quantitative analysis were conducted in YOD1-silenced HPAEpiC and MLE-12 cells (n = 6 per group). ***p  <  0.001, two-way ANOVA followed by Tukey’s multiple comparison test. Scale bar = 10 μm. B, D The levels of total iron were measured in YOD1-silenced HPAEpiC and MLE-12 cells (n = 6 per group). ***p  <  0.001, two-way ANOVA followed by Tukey’s multiple comparison test. E, F YOD1-deficient HPAEpiC and MLE-12 cells, as well as control cells, were stimulated with Erastin or FINO2 for 24 h. Subsequently, fluorescent staining of Fe²⁺ (red) in YOD1-deficient alveolar epithelial cells stimulated with Erastin or FINO2 (n = 6 per group). ***p  <  0.001, two-way ANOVA followed by Tukey’s multiple comparison test. Scale bar = 10 μm. G, H The levels of total iron in HPAEpiC and MLE-12 cells were assessed (n = 6 per group). ***p  <  0.001, two-way ANOVA followed by Tukey’s multiple comparison test. I, J The mRNA expression levels of FPN1, FTH1, FTL, NCOA4, IRP2, and TfR in YOD1-silenced and overexpressing cells were measured (n = 6 per group). ns= non-significant. Two-way ANOVA followed by Tukey’s multiple comparison test. K, L The protein expressions of TfR, FPN1, IRP2, FTH1, FTL, and NCOA4 in YOD1-overexpressing HPAEpiC and MLE-12 cells were evaluated using Western blot analysis (n = 3). M, N Co-IP analysis of the interaction between NCOA4 and FTH1 in HPAEpiC cells was performed (n = 3). O The protein expressions of FTH1 and NCOA4 in YOD1-silenced HPAEpiC and MLE-12 cells were assessed via Western blot analysis (n = 3). P YOD1-overexpressing HPAEpiC and MLE-12 cells were treated with 3-MA or CQ to inhibit autophagy and subsequently subjected to Western blot analysis to detect FTH1 protein levels (n = 3). All data were presented as mean ± SEM.

YOD1 promotes ferroptosis and cell injury through NCOA4-FTH1-ferritinophagy in alveolar epithelial cells

Studies have shown that NCOA4 can directly bind to FTH1 and promote ferritinophagy for iron release³². Therefore, we further investigated the effects of YOD1 on the binding of NCOA4 and FTH1 and the autophagy process. Silencing YOD1 significantly inhibited the LPS-induced binding of NCOA4 and FTH1 (Fig. 3M and N). In addition, silencing YOD1 elevated FTH1 and decreased NCOA4, both in vivo and in vitro (Figs. 3O and S3K). Furthermore, autophagy inhibitors 3-MA or CQ inhibited the decrease of FTH1 induced by LPS and YOD1 overexpression (Fig. 3P). These results suggest that YOD1 plays a role in regulating the expression of FTH1 and NCOA4 and ferritinophagy.

To further clarify the role of NCOA4-mediated ferritinophagy in LPS-induced ALI, NCOA4 was knocked down in HPAEpiCs and MLE-12 cells (Fig. S4A and B). As it suggested, knockdown of NCOA4 significantly inhibited YOD1 overexpression-induced FTH1 degradation and iron overload in LPS-stimulated alveolar epithelial cells (Fig. 4A–F), suggesting that NCOA4 may be a key molecule regulating the autophagy of FTH1 in these cells. In addition, NCOA4 silencing alleviated mitochondrial injury and lipid peroxidation induced by LPS and YOD1 overexpression (Fig. 4G–L). As anticipated, we found that silencing NCOA4 reversed YOD1 overexpression-induced cell injury, epithelial barrier dysfunction and cytokine release in LPS-stimulated alveolar epithelial cells (Figs. 4M–R and S4C, D). These results demonstrate that NCOA4 is essential in YOD1 overexpression-induced ferritinophagy and acute lung injury.

Fig. 4. YOD1 promotes ferroptosis and cell injury through upregulation of NCOA4 in alveolar epithelial cells.

A, B The protein expression of FTH1 in YOD1-overexpressing cells with or without NCOA4 silencing was analyzed (n = 3). C, D Fluorescent staining of Fe²⁺ (red) and quantitative analysis in YOD1-overexpressing cells with or without NCOA4 silencing were performed (n = 6 per group). ***p  <  0.001, two-way ANOVA followed by Tukey’s multiple comparison test. Scale bar = 10 μm. E, F The total iron in YOD1-overexpressing cells with or without NCOA4 silencing was determined (n = 6 per group). ***p  <  0.001, two-way ANOVA followed by Tukey’s multiple comparison test. G, H Transmission electron microscope in YOD1-overexpressing cells with or without NCOA4 silencing were performed (n = 3). Scale bar = 1 μm. I, J C11-BODIPY 581/591 staining and quantitative analysis in YOD1-overexpressing cells with or without NCOA4 silencing were performed (n = 6 per group). *P < 0.05, ***p  <  0.001, two-way ANOVA followed by Tukey’s multiple comparison test. Scale bar = 20 μm. K, L The levels of MDA and H₂O₂ in YOD1-overexpressing cells with or without NCOA4 silencing were determined (n = 6 per group). ***p  <  0.001, two-way ANOVA followed by Tukey’s multiple comparison test. M, N Cell viability of YOD1-overexpressing cells with or without NCOA4 silencing was assessed (n = 6 per group). ***p  <  0.001, two-way ANOVA followed by Tukey’s multiple comparison test. O, P LDH release in YOD1-overexpressing cells with or without NCOA4 silencing was measured (n = 6 per group). ***p  <  0.001, two-way ANOVA followed by Tukey’s multiple comparison test. Q, R The protein expressions of Claudin 3, E-cadherin, Occludin, and ZO-1 in YOD1-overexpressing cells with or without NCOA4 silencing were evaluated using Western blot analysis (n = 3). All data were presented as mean ± SEM.

YOD1 directly binds to NCOA4 via the OTU domain to prevent its proteasomal degradation

To elucidate the mechanism by which YOD1 regulates NCOA4, we conducted co-immunoprecipitation experiments to confirm the interaction between these two proteins. As shown in Fig. 5A and B, Flag-YOD1 and HA-NCOA4 were transfected into HPAEpiC cells as indicated; YOD1 and NCOA4 directly bind to each other under LPS stimulation. We subsequently examined the time effect of LPS on this interaction. Co-immunoprecipitation experiments demonstrated that YOD1 binds to NCOA4 in a time-dependent manner after LPS stimulation Fig. 5C and D.

Fig. 5. YOD1 directly binds to NCOA4 via the OTU domain.

A Co-immunoprecipitation (Co-IP) analysis was conducted to investigate the interaction between YOD1 and NCOA4 in HPAEpiC cells, following immunoprecipitation with anti-Flag-YOD1 (n = 3). B Co-IP analysis of the interaction between YOD1 and NCOA4 in HPAEpiC cells was performed using anti-HA-NCOA4 for immunoprecipitation (n = 3). C HPAEpiC cells were exposed to either IgG or LPS for 0, 6, 12, 18, and 24 h, followed by Co-IP analysis of the interaction between YOD1 and NCOA4, utilizing anti-Flag for immunoprecipitation (n = 3). D HPAEpiC cells were exposed to IgG or LPS for 0, 6, 12, 18, and 24 h, after which Co-IP analysis was performed to assess the interaction between YOD1 and NCOA4, employing anti-HA for immunoprecipitation (n = 3). E HPAEpiC cells were co-transfected with HA-NCOA4 and Flag-YOD1 and incubated with cycloheximide (CHX) for the specified durations. The expression of NCOA4 was detected using HA-tag antibodies (n = 3). F Following co-transfection of HPAEpiC cells with HA-NCOA4 and Flag-YOD1, the cells were incubated with MG132 or CQ for 24 h. Western blot analyses were then performed using the indicated antibodies (n = 3). G A schematic diagram illustrating the mutation of YOD1 segments is presented. H Co-IP analysis of YOD1 and NCOA4 was performed in HPAEpiC cells co-transfected with YOD1-related plasmids and the HA-NCOA4 plasmid. Wild-type (WT) or mutated YOD1 was immunoprecipitated using anti-Flag antibody (n = 3). I Total iron and cell viability in HPAEpiCs overexpressing YOD1(WT) or YOD1(ΔOUT) (n = 6 per group). ***p  <  0.001, one-way ANOVA followed by Tukey’s multiple comparison test. All data were presented as mean ± SEM.

Given that YOD1 overexpression or knockdown did not affect the mRNA expression of NCOA4 (Fig. 3I), we speculated that YOD1 may prevent the degradation of NCOA4. Therefore, we introduced cycloheximide (CHX) for the inhibition of protein synthesis and observed a significant reduction in the degradation rate of NCOA4 protein upon YOD1 addition (Fig. 5E). To thoroughly investigate the mechanism of NCOA4 degradation, we employed a proteasome inhibitor (MG-132) and an autophagy inhibitor (CQ) in YOD1-knockdown HPAEpiCs. Although YOD1 knockdown reduced the protein level of NCOA4, this downregulation was reversed by MG-132 treatment, while CQ treatment did not reverse it (Fig. 5F). This suggests that the decrease in NCOA4 levels induced by YOD1 knockdown is dependent on the proteasomal pathway rather than the autophagy pathway.

YOD1 comprises three distinct domains: a UBL (ubiquitin-linked) domain, an OTU (ubiquitin-specific protease) domain, and a zinc finger domain (Fig. 5G). To validate the specific region of YOD1 that interacts with NCOA4, we generated three plasmids containing different segmental mutations of YOD1. Our findings indicated that the deletion of the OTU domain in YOD1 abolished its interaction with NCOA4, whereas the deletion of the other domains did not disrupt YOD1’s ability to bind to NCOA4 (Fig. 5H). To clarify whether the OUT domain of YOD1 is essential in LPS-induced ferroptosis and cell injury, we overexpressed YOD1 (WT) and YOD1 (ΔOUT) in HPAEpiCs. Compared with the wild-type YOD1, YOD1 with the OTU domain deletion no longer aggravated the intracellular iron accumulation and cell damage caused by LPS (Fig. 5I). These results suggest that YOD1 can directly bind to and inhibit the degradation of NCOA4 via its OTU domain.

YOD1 inhibited K48-linked ubiquitination chains on K343 residue of NCOA4

As a deubiquitinating enzyme, YOD1 was found to inhibit the degradation of NCOA4 rather than GPX4 or ACSL4 via the ubiquitin-proteasome system (Figs. 6A and S5A, B). Since different types of polyubiquitin linkages mediate distinct biological functions, we expressed HA-tagged NCOA4 in HPAEpiC cells alongside various ubiquitin mutants (K6, K11, K27, K29, K33, K48, and K63), each containing a single designated lysine available for poly-linkage. Notably, overexpression of YOD1 primarily reduced K48-linked ubiquitination of NCOA4 (Fig. 6B).

Fig. 6. YOD1 inhibited K48-linked ubiquitination chains onto NCOA4 via 239–382 fragment of NCOA4.

A Co-immunoprecipitation (Co-IP) of NCOA4 was conducted in HPAEpiC cells co-transfected with Flag-YOD1, HA-NCOA4, and His-Ub plasmids. Ubiquitin (Ub) was immunoprecipitated using an anti-His antibody, while ubiquitinated NCOA4 was isolated with an anti-HA antibody (n = 3). B Immunoblot analysis was performed on lysates from HPAEpiC cells co-transfected with His-tagged ubiquitin (His-Ub), His-tagged K6-linked ubiquitin (His-K6-Ub), His-tagged K11-linked ubiquitin (His-K11-Ub), His-tagged K27-linked ubiquitin (His-K27-Ub), His-tagged K29-linked ubiquitin (His-K29-Ub), His-tagged K33-linked ubiquitin (His-K33-Ub), His-tagged K48-linked ubiquitin (His-K48-Ub), or His-tagged K63-linked ubiquitin (His-K63-Ub), in the presence or absence of HA-NCOA4 and Flag-YOD1. Immunoprecipitation was subsequently performed using anti-HA, followed by probing with anti-His antibodies (n = 3). C A schematic diagram illustrating the NCOA4 fragment is presented. D Co-IP of YOD1 and NCOA4 was performed in HPAEpiC cells co-transfected with various NCOA4 fragment plasmids and the Flag-YOD1 plasmid (n = 3). E The conserved lysine residue during different species. F Immunoblot analysis was performed on lysates from HPAEpiC cells co-transfected with His-tagged ubiquitin (His-Ub), Flag-YOD1, and NCOA4 mutant plasmid. IP was subsequently performed using anti-HA, followed by probing with anti-His antibodies (n = 3).

There are two subtypes of the human NCOA4 protein: a 614-residue α isoform and a 287-residue β isoform (Fig. 6C)³³. To identify the fragments of NCOA4 that are ubiquitinated, we expressed full-length HA-tagged NCOA4α in vitro and designed additional truncation constructs based on the secondary structure of NCOA4³⁴ (Fig. 6C). We then expressed various truncation constructs of NCOA4 in HPAEpiC cells. Notably, YOD1 can be pulled down only in NCOA4-WT and NCOA4 239–382, suggesting that YOD1 may bind with NCOA4 through 239–382 domain (Fig. 6D). Based on the analysis of three bioinformatic tools (UbiSite, BDM-PUB, and UbPred), a total of 8 lysines (K255, K263, K266, K271, K343, K356, K367, K377) were predicted as potential ubiquitination sites in 239–382 domain, and 4 lysine residues (K266, K271, K343, and K367) were conserved among various species (Fig. 6E). To determine the specific sites of NCOA4 protein that are deubiquitinated by YOD1, we mutated the 4 lysine residues of NCOA4. A ubiquitination assay indicated that K343 was the key site on NCOA4 deubiquitinated by YOD1 (Fig. 6F). Collectively, these results reveal the functional K48 ubiquitin chains and K343 lysine residue of NCOA4 involved in YOD1-mediated deubiquitination.

Deubiquitination of NCOA4 is essential in LPS-induced cell injury and ferroptosis in alveolar epithelial cells

Subsequent experiments were designed to confirm whether deubiquitination of NCOA4 plays a role in LPS-induced cell injury and ferroptosis. K343 of human NCOA4 is identical to K353 of mouse NCOA4 (Fig. 6E). Therefore, the NCOA4-K343R and NCOA4-K353R knock-in were constructed with CRISPR-Cas9 in HPAEpiC and MLE-12. LPS and YOD1 overexpression significantly increased levels of Fe²⁺, total iron, LPO, MDA, H₂O₂, and decreased protein expression level of FTH1 in NCOA4-WT alveolar epithelial cells, while NCOA4-K343R and NCOA4-K353R knock-in rendered alveolar epithelial cells more resistant to the above ferroptosis injuries (Fig. 7A–H).

Fig. 7. Deubiquitination of NCOA4 is essential in LPS-induced cell injury and ferroptosis in alveolar epithelial cells.

A, B Fluorescent staining of Fe²⁺ (red) was performed, followed by quantitative analysis in NCOA4-WT, NCOA4-K343R, or NCOA4-K353R alveolar epithelial cells (n = 6 per group). ***p  <  0.001, two-way ANOVA followed by Tukey’s multiple comparison test. Scale bar = 10 μm. C, D The contents of total iron were measured in NCOA4-WT, NCOA4-K343R, or NCOA4-K353R alveolar epithelial cells (n = 6 per group). ***p  <  0.001, two-way ANOVA followed by Tukey’s multiple comparison test. E, F The protein expression levels of FTH1 were assessed in NCOA4-WT, NCOA4-K343R, or NCOA4-K353R alveolar epithelial cells (n = 3). G, H The content of LPO, MDA, and H₂O₂ was measured in NCOA4-WT, NCOA4-K343R, or NCOA4-K353R alveolar epithelial cells (n = 6 per group). ***p  <  0.001, two-way ANOVA followed by Tukey’s multiple comparison test. I, J Cell viability was detected in NCOA4-WT, NCOA4-K343R, or NCOA4-K353R alveolar epithelial cells (n = 6 per group). ***p  <  0.001, two-way ANOVA followed by Tukey’s multiple comparison test. K, L The LDH release was detected in NCOA4-WT, NCOA4-K343R or NCOA4-K353R alveolar epithelial cells (n = 6 per group). ***p  <  0.001, two-way ANOVA followed by Tukey’s multiple comparison test. M, N The concentrations of IL-1β, IL-6, and TNF-α in NCOA4-WT, NCOA4-K343R, or NCOA4-K353R alveolar epithelial cells (n = 6 per group). ***p  <  0.001, two-way ANOVA followed by Tukey’s multiple comparison test. O, P The protein expression levels of Claudin 3, E-cadherin, Occludin, and ZO-1 in NCOA4-WT, NCOA4-K343R, or NCOA4-K353R alveolar epithelial cells (n = 3). All data were presented as mean ± SEM.

The present work also demonstrated that LPS and YOD1 overexpression significantly decreased cell viability, elevated LDH and cytokine release in NCOA4-WT alveolar epithelial cells, while NCOA4-K343R and NCOA4-K353R showed lighter damage to this injury (Fig. 7I–N). As anticipated, NCOA4-K343R and NCOA4-K353R knock-in attenuated LPS or YOD1 overexpression-induced reduction of intercellular adhesion proteins, compared to NCOA4-WT cell lines (Fig. 7O and P). The above results suggest that K343 and K353 are the lysine residues for deubiquitination of NCOA4 in alveolar epithelial cells and act as a potential target for YOD1.

To further validate the regulatory effects of NCOA4, PROTAC NCOA4 degrader-1 (V3), a specific NCOA4 degrader through prompting its ubiquitination, was used³⁵. V3 significantly alleviated LPS-induced elevation of LPO, MDA, H₂O₂, Fe²⁺, total iron, and decrease of FTH1 expression (Fig. S6A–H). Furthermore, V3 partly abolished LPS-induced cell injury, inflammation, and epithelial barrier damage (Fig. S6I–P). These results indicate that NCOA4 is deubiquitinated in ALI and contributes to ferroptosis in alveolar epithelial cells due to enhanced ferritinophagy.

Deubiquitination of NCOA4 at K353 residue is required in LPS-induced ALI and ferroptosis in vivo

To further define the functional target site of NCOA4 in vivo, we constructed an AAV6 carrying SPC-specific promoters to achieve alveolar endothelial-specific overexpression of NCOA4-WT or NCOA4-K353R in C57BL/6 mice. Overexpression of NCOA4-WT resulted in elevated levels of total iron, MDA, and LPO, an increase in ACSL4, and decreased expression levels of GPX4, FTH1, and FTL, which were more serious after LPS administration; however, the K353 mutation did not accentuate LPS-induced ALI (Fig. 8A and B). Furthermore, overexpression of NCOA4-WT induced and aggravated lung tissue injury in both normal and LPS-damaged mice, whereas K353 mutation showed less damage than that of NCOA4-WT (Fig. 8C). Additionally, NCOA4-WT overexpression induced hypoxia, pulmonary edema, extravasation of alveolar proteins, inflammatory cell infiltration, and increased cytokine release, whereas K353 mutation rendered resistance to this damage (Fig. 8D–F). Moreover, the K353 mutation also exhibited less damage to the alveolar epithelial barrier (Fig. 8G). These findings suggest that deubiquitination of NCOA4 at K353 residue is required in LPS-induced ferroptosis and ALI in vivo.

Fig. 8. Deubiquitination of NCOA4 at K353 residue is required in LPS-induced ferroptosis and ALI in vivo.

A The content of total iron, MDA, and LPO was measured in mice with overexpression of NCOA4-WT or NCOA4-353R (n = 6 mice/group). ***p  <  0.001, two-way ANOVA followed by Tukey’s multiple comparison test. B The protein expression level of ferroptosis-related proteins was assessed in mice with overexpression of NCOA4-WT or NCOA4-353R (n = 3). C An AAV6 carrying an SPC-specific promoter was utilized to achieve ATII-specific overexpression of NCOA4-WT or NCOA4-K353R in C57BL/6 mice. Representative images of HE staining of mouse lung tissues and the corresponding lung injury scores are presented, with a scale bar of 100 μm (n = 6 mice/group). ***p  <  0.001, two-way ANOVA followed by Tukey’s multiple comparison test. D The partial pressure of arterial oxygen, lung wet-to-dry weight ratio (W/D), and protein content were measured in mice with overexpression of NCOA4-WT or NCOA4-353R (n = 6 mice/group). *p < 0.05, **p < 0.01, ***p  <  0.001, two-way ANOVA followed by Tukey’s multiple comparison test. E Total cell counts and neutrophil counts in BALF were evaluated in mice with overexpression of NCOA4-WT or NCOA4-353R (n = 6 mice/group). ***p  <  0.001, two-way ANOVA followed by Tukey’s multiple comparison test. F The concentrations of IL-1β, IL-6, and TNF-α in BALF were quantified in mice with overexpression of NCOA4-WT or NCOA4-353R (n = 6 mice/group). ***p  <  0.001, two-way ANOVA followed by Tukey’s multiple comparison test. G The protein expression levels of Claudin 3, E-cadherin, Occludin, and ZO-1 were analyzed in mice with overexpression of NCOA4-WT or NCOA4-353R (n = 3). H, I The content of YOD1 and NCOA4 in BALF of control patients (n = 15) and ALI patients (n = 22) was detected by ELISA. ***p < 0.001, unpaired t-test. J The relationship between YOD1 content of BALF and PaO₂/FiO₂ in ALI patients (n = 22). p < 0.0001, Pearson correlation test, r = −0.7930. K The relationship between YOD1 content of BALF and APACHE-II score in ALI patients (n = 22). p = 0.0001, Pearson correlation test, r = 0.7239. L The relationship between NCOA4 content of BALF and PaO₂/FiO₂ in ALI patients (n = 22). p < 0.0001, Pearson correlation test, r = −0.8512. M The relationship between NCOA4 content of BALF and APACHE-II score in ALI patients (n = 22). p = 0.0041, Pearson correlation test, r = 0.5872. N YOD1 directly binds with and reduces K48-linked ubiquitination of NCOA4, thereby promoting FTH1 ferritinophagy and ferroptosis, which break the tight connections between alveolar epithelial cells and modulate LPS-induced ALI. All data were presented as mean ± SEM.

Further studies were designed to investigate the role of NCOA4 in vivo. Consistent with the in vitro study, V3 significantly alleviated LPS-induced ferroptosis (Fig. S7A and B). Furthermore, V3 attenuated LPS-induced lung injury, manifested as improvement in hypoxia, pulmonary edema, extravasation of alveolar proteins, inflammatory cell infiltration, cytokine release, and alveolar epithelial barrier damage (Fig. S7C–I). These results indicate that NCOA4 is a critical regulator in LPS-induced ferroptosis and ALI, which may be a potential target for the treatment of ALI.

In order to further understand the expression of YOD1 and NCOA4 in patients with acute lung injury, we enrolled 15 patients without pulmonary infections and 22 patients with ALI. The sex and age were comparable between the two groups. All the patients underwent bronchoscopy, and BALF was collected. The content of YOD1 and NCOA4 in BALF was measured by ELISA. The results showed that the levels of YOD1 and NCOA4 in the BALF of patients with ALI were significantly increased (Fig. 8H and I). Moreover, the level of YOD1 and NCOA4 in the BALF of patients with ALI was significantly negatively correlated with the PaO₂/FiO₂ of the patients, and significantly positively correlated with the APACHE-II score (Fig. 8J–M). These results indicate that YOD1 and NCOA4 are correlated with the severity of ALI, making it potential biomarkers for the prediction of prognosis in ALI patients.

In summary, our findings reveal that YOD1 directly binds with and reduces K48-linked ubiquitination of NCOA4, thereby promoting FTH1 ferritinophagy and ferroptosis, which break the tight connections between alveolar epithelial cells and modulate LPS-induced ALI (Fig. 8N).

Discussion

ALI is characterized by acute inflammation, loss of alveolar-capillary membrane integrity, excessive transepithelial neutrophil migration, and release of pro-inflammatory, cytotoxic mediators³⁶. Although numerous biomarkers have been identified to predict morbidity and mortality in ALI, clinical trials indicate that many pharmacological strategies have failed to reduce mortality³⁷. In this study, we found that YOD1 expression was increased in LPS-stimulated alveolar epithelial cells and mouse lung tissues. Silencing YOD1 alleviated LPS-induced acute lung injury and ferroptosis in alveolar epithelial cells. Mechanistically, we demonstrated that YOD1 directly binds to and deubiquitinates NCOA4 via K48-linked chains, subsequently enhancing NCOA4 expression. This process facilitates the degradation of FTH1, increases intracellular iron levels, and promotes ferroptosis. Our findings reveal a critical role for YOD1 in regulating NCOA4 expression, suggesting it as a potential therapeutic target for the treatment of ALI.

YOD1 is a deubiquitinase belonging to the OTU protein family, which comprises three domains: a UBL domain, an OTU domain, and a Znf domain. The OTU catalytic core has been shown to be both necessary and sufficient for YOD1’s deubiquitinating activity in vitro, while the UBL and Znf domains position the OTU domain appropriately²⁸. YOD1 has been implicated in the onset and progression of viral infectious diseases and various cancers through its ability to remove ubiquitin chains and regulate protein degradation^25,38–40. Sun et al.‘s research demonstrated that YOD1 participates in Aβ-induced Alzheimer’s disease in mice by promoting microglial cell inflammation⁴¹. Liu et al.‘s study found that YOD1 promotes inflammatory responses caused by Sendai virus infection by inhibiting the innate antiviral immune response²⁷. Additionally, Zhao et al.‘s research indicated that YOD1 participates in 6-OHDA-induced Parkinson’s disease by promoting ROS release⁴². The above studies suggest that YOD1 has pro-inflammatory and pro-oxidative stress effects, and oxidative stress can induce iron overload and ferroptosis⁴³. Therefore, we speculated that YOD1 may promote oxidative stress damage and ferroptosis in alveolar epithelial cells, which has been verified in this study. However, on the other hand, the role of YOD1 in infectious diseases is still controversial. Liu et al.‘s research found that YOD1 is upregulated in MRSA infection-induced sepsis. YOD1 inhibits NLRP3 inflammasome activation through deubiquitination, thereby improving DIC caused by sepsis²⁸. This indicates that YOD1 is not merely an anti-inflammatory or pro-inflammatory effect, but has an environment-dependent and complex regulatory effect, which varies depending on the infection site and pathogen. The dual role of YOD1 in inflammatory diseases and its activation conditions deserves further research exploration in the future.

Ferroptosis is a form of non-apoptotic cell death characterized by the accumulation of iron-dependent lipid peroxides, playing a significant role in acute lung injury and various other diseases^44–48. Concurrently with ferroptosis, autophagy is activated, leading to the degradation of ferritin via the binding of NCOA4 and FTH1⁴⁹, which are also important in iron homeostasis and regulation of ferroptosis. Xu et al. reported that melatonin mitigates macrophage ferroptosis by inhibiting NCOA4-mediated ferritinophagy²⁰. Additionally, Zhang et al. found that YAP1 could suppress the autophagic degradation of FTH1 mediated by NCOA4, which is crucial for protecting pulmonary epithelial cells from sepsis-induced ferroptosis⁵⁰. Our results showed a similar role of NCOA4 and FTH1 in LPS-induced ferroptosis in alveolar epithelial cells. Furthermore, plasma ferritin is a clinically useful biomarker in ARDS and is associated with worse patient outcomes⁵¹. These suggest that NCOA4-mediated ferritinophagy may be a potential target for ALI.

The human NCOA4 protein comprises two transcription variants: NCOA4α, which consists of 614 amino acids, and NCOA4β, which consists of 287 amino acids. Although NCOA4 orthologs are present across the entire metazoan lineage, the homology within the proteome is limited. Research on the structural characteristics of NCOA4 is scarce. Mancias et al. demonstrated that NCOA4 binds to and regulates the autophagic degradation of FTH1 through the W497 and I489 residues. Furthermore, E3 ubiquitin ligase HERC binds to and ubiquitinates NCOA4 at the 383-522 region of the protein, excluding the W497 and I489 sites³⁴. Additionally, Wu et al. reported that ATM (ataxia telangiectasia mutated) phosphorylates NCOA4 at the S550 site, thereby promoting the interaction between NCOA4 and FTH1, which subsequently facilitates ferroptosis⁵². In the present study, we found that YOD1 deubiquitinates NCOA4 at the K343 residue in HPAEpiC and the K353 residue in MLE-12 and mouse, thereby regulating the expression levels of NCOA4.

However, targeting deubiquitination in clinical research still has many challenges. Numerous clinical trials have demonstrated that small-molecule deubiquitinase (DUB) inhibitors play an active role in inflammatory and oncologic diseases in preclinical studies⁵³. However, VLX1570, the only DUB inhibitor to have entered clinical research, was discontinued due to pulmonary toxicity⁵³. Given the unique mechanism of action, further efforts are needed to identify highly selective DUB inhibitors with improved therapeutic indices. Future DUB inhibitor development requires high-resolution structural mapping (cryo-EM/crystallography) of catalytic and ligand-binding sites to guide design, accommodating inherent protein flexibility. Screening platforms should utilize natural substrates in physiomimetic conditions to enhance physiological relevance and translatability. Molecular dynamics simulations must predict protein conformational impacts on inhibitor binding. Complementary SAR studies are essential for fine-tuning functional group affinity and minimizing off-target effects. Finally, multitiered validation—spanning in vitro assays, cellular models, and in vivo studies—is crucial to comprehensively assess compound efficacy, therapeutic potential, and safety profiles²².

There are some limitations in this study. Although we found that the concentration of YOD1 and NCOA4 in BALF of ALI patients was significantly elevated and correlated with the severity of the disease, these changes were not be verified in lung tissues of patients due to the difficulty in obtaining lung tissue from ALI patients. In addition, although studies have highlighted the significant roles of YOD1 and NCOA4 in various diseases, only one study has demonstrated that YOD1 inhibitors can inhibit the proliferation and metastasis of cancer in vivo and in vitro²³. No related drugs have advanced to clinical research, so further exploration of both mechanisms and clinical studies is necessary.

In conclusion, our findings demonstrate that YOD1 directly binds to and inhibits the degradation of NCOA4 through deubiquitination, which promotes the ferritinophagy of FTH1 and ferroptosis, subsequently altering LPS-induced ALI. Therefore, our results suggest that YOD1 may serve as a potential therapeutic target for the treatment of ALI.

Methods

Patients

From January 1, 2025 to August 20, 2025, 15 patients without pulmonary infections(9 male and 6 female, 69.8 ± 13.6 years old) and 22 patients with ALI (13 male and 9 female, 73.8 ± 7.5 years old) were recruited for the study. The diagnostic criteria for acute lung injury refer to the latest definition of 2023⁵⁴. We conducted bronchoscopy on these subjects and collected bronchoalveolar lavage fluid (BALF). After centrifugation, the supernatant was frozen at −80 °C in a refrigerator. And record the PaO₂/FiO₂ ratio and APACHE-II score of the patients within one day before or after the bronchoscopy. The human research has been approved by the Ethics Committee of Suzhou Hospital Affiliated to Nanjing Medical University (K-2025-095), and written informed consent was obtained from all subjects. All ethical regulations relevant to human research participants were followed.

Inclusion criteria for subjects in the ALI group:

• A.Age between 18 and 85 years old.
• B.Precipitated by an acute predisposing risk factor. Pulmonary edema is not exclusively or primarily attributable to cardiogenic pulmonary edema/fluid overload, and hypoxemia/gas exchange abnormalities are not primarily attributable to atelectasis.
• C.Acute onset or worsening of hypoxemic respiratory failure within 1 week of the estimated onset of the predisposing risk factor or new or worsening respiratory symptoms.
• D.PaO₂/FiO₂ ≤ 300 mmHg(1 mmHg=0.133 kPa).
• E.Bilateral opacities on chest radiography and computed tomography.

Inclusion criteria for subjects in the control group:

• A.Age between 18 and 85 years old.
• B.Without acute respiratory infection or respiratory system tumors.
• C.Patients who require bronchoscopy due to respiratory foreign bodies or other reasons.

Exclusion criteria for control and ALI groups:

• A.Patients have contraindications for bronchoscopy.
• B.Patients who cannot cooperate with bronchoscopy.

Detection of YOD1 and NCOA4 in BALF of patients

YOD1 and NCOA4 ELISA Kit were purchased from Antibodies (Cambridge, UK). 100 μL of BALF supernatant or the standard sample was added to the wells. Following 2 h of incubation at room temperature, the wells are washed and then incubated with biotinylated antibody for 1 h. Following washing, an HRP-Streptavidin conjugate is added to the wells, and the microtiter plate is incubated. Following incubation and washing, the TMB substrate solution is then used to visualize the HRP enzymatic reaction by catalysis to produce a blue-colored product that changes to yellow after the addition of acidic stop solution. The concentration of YOD1 or NCOA4 can then be calculated by reading the O.D. absorbance at 450 nm in a microplate reader and referring to the standard curve.

Animals

The animal study protocols were approved by the Animal Ethics Committee of Soochow University (Ethics number 202402A0195), and we have complied with all relevant ethical regulations for animal use. 6–8 weeks old C57BL/6 male mice were housed under standard conditions, with controlled temperature and humidity, maintained on a 12-h light–dark cycle with artificial lighting, and provided unrestricted access to tap water and a pelleted mouse diet. Eight-week-old male mice were used to induce the ALI model through intratracheal administration of lipopolysaccharide (LPS, Sigma-Aldrich, St. Louis, MO, USA) at a dosage of 5 mg/kg over 12 h^46,55,56. The control group received the vehicle solution via the same route and volume.

To achieve alveolar epithelium-specific overexpression of YOD1(WT), YOD1(ΔOTU), NCOA4-WT, or NCOA4-K353R, six-week-old C57BL/6 mice were intratracheally injected with GFP-tagged adeno-associated virus 6 containing the surfactant protein C (SPC)-specific promoter (1 × 10¹² vg/ml) once. A negative control adeno-associated virus 6 vector (AAV6-NC) was administered at the same dosage and intervals. After 3 weeks, mice were executed, and the specificity and efficiency of AAV6 transfection were verified using Western blot analysis^46,57,58. To degrade NCOA4 in vivo, mice were intraperitoneally administered with 30 mg/kg PROTAC NCOA4 degrader-1 (V3) (MedChemExpress, USA) 3 h before LPS exposure³⁵.

Arterial blood gas analysis

After anesthetizing the mice, a midline incision was made to expose the abdominal aorta, from which blood samples were collected. Subsequently, blood gas analyses were conducted using an i-STAT Analyzer (Abbott, Chicago, USA).

Lung wet/dry (W/D) weight measurement

The lung weight-to-dry weight (W/D) ratio serves as an indicator of the severity of pulmonary edema. Following euthanasia, the left lung was excised and weighed to obtain its wet weight (W). Subsequently, the wet lung tissue was desiccated in an oven at 60 °C for 48 h before reweighing to determine the dry weight (D). The W/D ratio was calculated by dividing the wet weight by the dry weight.

Bronchoalveolar lavage

After anesthetizing the mice, we exposed the chest cavity, intubated the trachea, and performed whole lung lavage using sterile PBS (1 mL). The collected lavage fluid was centrifuged at 1200 × g for 10 min at 4 °C. Cell-free supernatants were harvested and stored at −80 °C for subsequent analysis. Total protein concentration in the BALF was measured using a BCA Protein Assay Kit (Beyotime, Shanghai, China). To determine total and neutrophil cell counts, BALF cytospin preparations were stained with Wright-Giemsa and examined under a microscope.

Pathological examination of lung tissues

For the histological assessment of acute lung injury, lung tissues were fixed in a 4% paraformaldehyde solution, gradually dehydrated, paraffin-embedded, sectioned into 5 μm slices, and stained with hematoxylin and eosin (H&E). The evaluation of lung injury involved examining six randomly selected lung fields per mouse and calculating the lung injury score based on the official guidelines from the American Thoracic Society workshop report on experimental acute lung injury in animals⁵⁹.

Cell culture

Human pulmonary alveolar epithelial cells (HPAEpiC) and mouse alveolar epithelial cells (MLE-12) were utilized in the present study. HPAEpiC cells were obtained from ScienCell (Carlsbad, CA, USA), while MLE-12 cells were sourced from the American Type Culture Collection (ATCC, MD, USA). The 2nd–4th generation HPAEpiC cells and the 3rd–8th generation MLE-12 cells were used for cell experiments. Both cell types were cultured in DMEM/F12 supplemented with 10% fetal bovine serum (FBS), 100 U/mL penicillin, and 100 µg/mL streptomycin in a humidified atmosphere of 5% CO₂ at 37 °C. Upon reaching ~80% confluence, the cells were treated with LPS at a concentration of 2 µg/mL for 24 h^46,50,60.

HPAEpiC and MLE-12 were transfected with lentiviral vectors (LV) encoding YOD1, shYOD1, shNCOA4, or a negative control, following the manufacturer’s instructions. Transfection efficiencies were confirmed using Western blot analysis. Deleted, truncated, and point mutants were obtained by PCR-based amplification using the construct encoding WT proteins as templates and verified by sequencing⁶¹. HPAEpiC cells were transfected with lentiviral vectors encoding Flag-YOD1^WT, Flag-YOD1^ΔUBL, Flag-YOD1^ΔOUT, Flag-YOD1^ΔZnf, HA-NCOA4 (WT), HA-NCOA4(1-238), HA-NCOA4(239-382), HA-NCOA4(383-522), HA-NCOA4(523-614), HA-NCOA4-K266R, HA-NCOA4-K271R, HA-NCOA4-K343R, HA-NCOA4-K367R in accordance with the manufacturer’s guidelines. The wild-type (WT) and mutant plasmids of human YOD1 were co-transfected with HA-NCOA4 into HPAEpiC cells for 72 h using Lipofectamine™ 3000 Transfection Reagent. Similarly, the WT and mutant plasmids of human NCOA4 were co-transfected with Flag-YOD1 into HPAEpiC cells for 72 h using Lipofectamine™ 3000 Transfection Reagent. MLE-12 cells were transfected with lentiviral vectors encoding YOD1 in accordance with the manufacturer’s guidelines.

Genomic mutations were introduced into cells using the CRISPR–Cas9 system. Single-guide RNAs (sgRNAs) targeting genomic regions adjacent to the mutation sites at K343 or K353 were designed by QEgene (Shanghai, China). Once the cells reached 60% confluence, they were co-transfected with 0.5 μg of sgRNAs and single-stranded donor oligonucleotides (ssODNs) as templates to introduce the mutations. 24 hours post-transfection, the cells were trypsinized, diluted to obtain single cells, and plated in 96-well plates. Genomic DNA was then extracted, and the PCR products spanning the mutation sites were sequenced⁵⁸.

To inhibit ferroptosis, necroptosis, or apoptosis, YOD1-overexpressing cells were incubated with Fer-1 (1 μM), Lip-1 (0.2 μM), NSA (0.5 μM), or Z-VAD (10 μM) for 48 h. To induce ferroptosis, YOD1-deficient HPAEpiC and MLE-12 cells were stimulated with Erastin (10 μM) or FINO2 (10 μM) for 24 h. To inhibit autophagy, YOD1-overexpressing HPAEpiC or MLE-12 cells were incubated with 3-Methyladenine (3-MA, 5 mM) or chloroquine (CQ, 3 μM) for 24 hours. To degrade NCOA4 in vitro, HPAEpiC and MLE-12 were administered with 200 nM PROTAC NCOA4 degrader-1 (V3) (MedChemExpress, USA) 24 hours before LPS exposure³⁵.

Isolation of primary ATII in mice

Consistent with reported⁶², after anesthetizing and sterilizing, the mice with 75% alcohol for one minute, the thoracic cavity was opened under sterile conditions, and the lungs were carefully excised, ensuring the removal of the trachea and bronchi. The lungs were washed with phosphate-buffered saline (PBS) and then cut into pieces in Dulbecco’s modified Eagle medium (DMEM). Trypsin (with EDTA) and type I collagenase were added for digestion, followed by centrifugation to collect and suspend the cell pellet. Red blood cell lysis buffer was used to lyse the red blood cells, after which centrifugation was performed to remove the lysed cells. The resulting cell suspension was filtered through a 200-mesh filter, and the filtrate was collected and centrifuged before suspending the cells in a culture bottle. This culture bottle was placed in a 37 °C, 5% CO₂ incubator for 40 min. Non-adherent cells were then removed, and the remaining cells were centrifuged, suspended, and cultured again, with this process repeated three times. The cell suspension was passed through 38 and 19 μm filters, and the filtrate was transferred to a culture dish coated with mouse immunoglobulin G (IgG) and incubated in a 37 °C, 5% CO₂ incubator for 2 h. Following the purification process, non-adherent cells were removed, and the remaining cells were centrifuged, suspended, and cultured again. The cells obtained were identified as primary ATⅡ. Immunofluorescence staining using an SP-C antibody revealed a positive rate of 98%, confirming the successful extraction of primary ATⅡ cells from mice.

Cell viability measurement

Cell viability was assessed using a Cell Counting Kit-8 (CCK-8) according to the manufacturer’s instructions (Beyotime, Shanghai, China). Cell death was analyzed by measuring lactate dehydrogenase (LDH) release in the medium using a commercial kit (Beyotime, Shanghai, China). Briefly, cells were removed by centrifugation at 400 × g, and the cell-free supernatant was incubated with WST Substrate Mix and LDH Reaction Mix at room temperature for 30 minutes. The absorbance was measured at 450 nm.

Enzyme‑linked immunosorbent assays (ELISAs)

Proinflammatory cytokine concentrations, including IL-1β, IL-6, and TNF-α, in mouse BALF and cell culture supernatant were quantified using ELISA kits according to the manufacturer’s guidelines (Beyotime, Shanghai, China).

Lipid peroxidation measurements

Lipid peroxidation was assessed by measuring intracellular lipid peroxidase (LPO) levels using a commercial kit according to the manufacturer’s instructions (Abcam, Cambridge, UK). Briefly, lipid hydroperoxides were extracted from the samples into chloroform. The extracts were then incubated with Reagents 1 and 2 at room temperature for 5 min. LPO levels were detected at 500 nm.

C11-BODIPY 581/591 (BODIPY) staining (Beyotime, Shanghai, China) was also used to measure lipid peroxidation. Briefly, cells were incubated with BODIPY at a final concentration of 2 µM for 30 min at 37 °C in the dark. After washing with PBS for two times, DAPI staining (1:5000) was performed. Then, samples were examined using a laser confocal microscope (Olympus FV3000, Japan), and fluorescence intensity analysis was conducted with ImageJ software (version 1.53c, NIH, USA).

Additionally, the level of intracellular malondialdehyde (MDA) was measured using an MDA detection assay (Beyotime, Shanghai, China). Tissues or cells were lysed on ice and centrifuged at 12,000 × g for 10 min to remove insoluble materials. The supernatants were incubated with the MDA detection working solution at 100 °C for 15 min, cooled to room temperature, centrifuged at 1000×g for 10 min, and measured at 532 nm.

Furthermore, the production of hydrogen peroxide (H₂O₂) was also measured (Beyotime, Shanghai, China). Tissues or cell samples were lysed on ice and centrifuged at 12,000×g and 4 °C for 5 min to eliminate insoluble materials. The supernatants were incubated with the H₂O₂ detection working solution at room temperature for 30 minutes and measured at 560 nm.

Iron content assay

The total iron content was quantified using iron assay kits (Applygen, Shanghai, China). Lung tissues were lysed, homogenized on ice, and subsequently centrifuged to collect the supernatant. Similarly, cells were homogenized on ice and then centrifuged to obtain the supernatant. The supernatants from both tissues and cells were then sequentially incubated with buffer solution and 4.5% potassium permanganate solution at 60°C for 1 h. After cooling and centrifugation, an iron ion detector was added, and the mixture was incubated at room temperature for 30 min. After centrifugation at 12,000 rpm for 5 min, the supernatant was transferred to a 96-well plate, and the absorbance at 550 nm was measured.

The FerroOrange kit from Dojindo (Kyushu, Japan) was utilized to measure Fe²⁺ levels in cells. The cells were inoculated into the fluorescent culture dishes and incubated overnight in a 37 °C, 5% CO₂ incubator. After washing with serum-free medium three times, the cells were treated with different drugs according to the experimental groups. Then, cells were treated with a 1 μM FerroOrange working solution for 30 min and subsequently visualized using a laser confocal microscope (Olympus FV3000, Japan).

Transmission electron microscopy

The cultured cells were digested, centrifuged, and rapidly fixed to examine the ultrastructure of the cells. The specimens were then postfixed with 1% osmium tetroxide, progressively dehydrated, and embedded in LR white medium (Electron Microscopy Laboratory). Sections (60 nm) were counterstained with uranyl acetate and analyzed using a transmission electron microscope (TEM) (HT7800, Hitachi, Japan). At least six random fields were captured and evaluated.

Immunoblot analysis and immunoprecipitation

Cells were lysed in cell lysis buffer and subjected to SDS-PAGE, followed by immunoblot analysis using the specified antibodies. For immunoprecipitation assays, cells were lysed in IP lysis buffer (Beyotime, Shanghai, China). The lysates were immunoprecipitated with Ig-A/G-magnetic beads (BioLinkedIn, China) pre-conjugated with either IgG (Beyotime, Shanghai, China) or the specified antibodies. The precipitates were then washed from the beads, boiled, and separated by SDS–PAGE. The primary antibodies utilized are listed in Table S1.

Protein degradation assay

To determine whether YOD1 stabilizes NCOA4 protein, cells overexpressing YOD1 or control cells were incubated with cycloheximide (CHX, 10 μM) to inhibit protein synthesis. Cell lysates were prepared at 0, 3, 6, and 9 h for immunoblot analysis. Additionally, YOD1-deficient HPAEpiC cells were treated with MG-132 (5 μM) for 8 h or CQ (3 μM) for 24 h to suppress proteasome-mediated protein degradation and autophagy.

Ubiquitylation assay

To analyze the ubiquitination of NCOA4, GPX4, and ACSL4 in HPAEpiC cells, the cells were transfected with the specified expression plasmids using polyethyleneimine for 24 h prior to harvesting. Following transfection, the cells were lysed in lysis buffer and incubated on ice for 30 min. After a subsequent incubation for 2 h at 4 °C, the HA beads were washed three times with lysis buffer. The samples were then subjected to immunoblot analysis using anti-HA, anti-Myc, or anti-His antibodies to assess NCOA4, GPX4, and ACLS4 ubiquitination. To identify the chain linkage of ubiquitinated NCOA4, HPAEpiC cells were transfected with various ubiquitin mutants, including K6, K11, K27, K29, K33, K48, and K63, each containing a single lysine available for poly-linkage⁶³. Ubiquitination was detected as described above.

Immunofluorescence staining

Lung tissues were fixed in 4% paraformaldehyde, gradually dehydrated, paraffin-embedded, and sectioned into 5 μm slices. The lung sections were permeabilized with 0.3% Triton X-100 for 10 min, blocked with 3% BSA for 1 h, and incubated with primary antibodies overnight at 4 °C. After washing with PBS the following day, fluorescent secondary antibodies were applied for 1 h in the dark. DAPI staining (1:5000) was performed 8 min prior to fluorescence microscopy. Samples were examined using a laser confocal microscope (Olympus FV3000, Japan), and fluorescence intensity analysis was conducted with ImageJ software (version 1.53c, NIH, USA). The primary antibodies used for immunofluorescence staining are listed in Table S1.

Quantitative real-time PCR

The mRNA levels of FPN1, FTH1, FTL, NCOA4, IRP2, and TfR were quantified using real-time PCR with gene-specific primers on a dedicated real-time PCR platform. RNA was isolated from HPAEpiC and MLE-12 cells using the TRIzol extraction method (Vazyme, Nanjing, China). Subsequently, reverse transcription was performed using a cDNA synthesis kit from Takara (Japan). Quantitative real-time PCR analysis was conducted on a Bio-Rad CFX96 real-time system (USA) with SYBR Green I from Takara (Japan). Detailed primer information is provided in Table S2, and the results were analyzed using the 2^−ΔΔCt method.

Statistical analysis

Data analysis was performed using GraphPad Prism software (Version 8.0.1, USA). Results are presented as mean ± standard error of the mean (SEM). The normality of the data was assessed using Shapiro–Wilk and Kolmogorov–Smirnov tests. For two groups of continuous variables, Student’s t-test was used if the variance homogeneity was satisfied; otherwise, Welch’s correction was used. For three or more groups of continuous variables, one-way or two-way ANOVA was applied if variance homogeneity was satisfied;, otherwise, Brown-Forsythe and Welch ANOVA were used. Statistical significance was defined as two-sided P < 0.05.

Author contributions

Conceptualization: Y. C. and S. L. Methodology: X. Y., X. Z., C. Q., and X. L. Validation: L. X. and G. X. Formal analysis: X. Y. and T. R. Investigation: C. Q. and Y. C. Resources: C. H., Y. C., S. L., and C. C. Data curation: Y. C. and L. X. Visualization: X. Y. Supervision: X. Z. Project administration: X. Z., Y. C., and S. L. Funding acquisition: Y. C., S. L., X. Y., C. C., and G. X. Manuscript drafting and revision: X. Y., S. L., and Y. C.

Peer review

Peer review information

Communications Biology thanks Xiaoran Liu, Hongquan Wang and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Primary Handling Editors: Laura Rodríguez Pérez. [A peer review file is available].

Data availability

The data used and/or analyzed during the study are available on request from the corresponding author upon reasonable request. The uncropped blots are included in the Supplement information file as Supplementary Fig. 8. The source data for all bar charts are provided in Figureshare (10.6084/m9.figshare.30509027).

Competing interests

The authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at 10.1038/s42003-025-09237-8.