Screening of potential key ferroptosis-related genes in sepsis

Introduction

Sepsis is defined as a systemic inflammatory response syndrome (SIRS) that occurs when a dysregulated host response to bacterial, viral or fungal infection leads to unacceptably severe tissue damage and organ dysfunction (Keeley, Hine & Nsutebu, 2017). More specifically, sepsis can be viewed as a fierce competition between pathogens and the host immune response, in which pathogens impair all aspects of host immune defense (Hotchkiss, Monneret & Payen, 2013). Once the host loses control of local infection, the activated innate immune system induces a state of robust inflammation characterized by an exorbitant release of inflammatory cytokines, which is also known as a “cytokine storm” and plays an important role in initiating organ dysfunction (Delano & Ward, 2016). Although some septic patients survive this hyperinflammatory phase, they are still at an increased risk for the development of secondary persistent inflammation due to sepsis-induced long-term immunosuppression (Delano & Ward, 2016; Nolt et al., 2018). This immunosuppression is characterized by the deprivation of delayed hypersensitivity to positive control antigens, a failure to eliminate the primary infection and the development of new opportunistic infections (Hotchkiss et al., 2009). With the deepening of research, the extensive loss of immune cells has been considered a major manifestation of sepsis-induced immunosuppression, which is characterized by programmed cell death induction (Cao, Yu & Chai, 2019). Over the past decade, a newly discovered necrotic-form of programmed cell death named ferroptosis has been proposed. The latest study reported that ferroptosis could release danger-associated molecular patterns (DAMPs) that irreversibly trigger stress-exposed cells into a proinflammatory state (Liu et al., 2022). Thus, we hypothesized that ferroptosis may play important roles in the development of inflammation. However, the mechanism of ferroptosis in sepsis-induced systemic inflammation remains unclear.

Ferroptosis is a novel type of cell death characterized by increased lipid reactive oxygen species (ROS), iron-dependent lipid peroxidation and plasma membrane damage (Dixon et al., 2012). Ferroptosis is well known for its specific physiological and morphological characteristics such as iron accumulation, decreased cystine uptake, intact nuclei and shrunken mitochondria, which are different from the characteristics of apoptosis and pyroptosis (Han et al., 2020). In accordance with its name, intracellular iron homeostasis plays a significant role in regulating ferroptosis. Iron metabolism-related genes, such as transferring (TF), transferrin receptor (TFR) and ferroportin (FPN), are considered key mediators of ferroptosis (Mou et al., 2019). For example, the overexpression of heat shock protein beta-1 (HSPB1) inhibits ferroptosis by reducing intracellular iron accumulation by inhibiting TRF1 expression in tumor cells (Li et al., 2020a). Although HSPB1 has been widely considered as a molecular chaperone, upon phosphorylation, HSPB1 dissociates into monomers that act as negative regulators of ferroptosis through reducing the uptake of cellular iron and the production of lipid ROS (Sun et al., 2015). In addition, excessive redox active divalent iron (Fe²⁺) can provide electrons to hyperoxide to form harmful ROS triggering ferroptosis (Battaglia et al., 2020). It has been confirmed that the activation of Ras-mitogen-activated protein kinase (MEK) signaling sensitizes tumor cell lines to ferroptosis by promoting ROS accumulation by inhibiting cystine and mitochondrial voltage-dependent anion channel 2/3 (VDAC 2/3) (Dixon et al., 2012; Han et al., 2020). Despite the evidence emphasizing the significant roles of ferroptosis in cancer, few studies have focused on its function in other diseases. Recently, the roles of ferroptosis in inflammation-related diseases have attracted widespread attention due to its strong association with immunogenicity (Sun et al., 2020). The latest studies reported that the pharmacological suppression of ferroptosis protects against sepsis-induced cardiac injury and liver injury in mice (Li et al., 2020b; Wei et al., 2020). Furthermore, Gong et al. (2022) have identified ferroptosis-related biomarkers of septic cardiomyopathy via bioinformatics analysis. Thus, we have reason to believe that ferroptosis plays key roles in sepsis-induced multiple organ injury. However, the specific mechanism and genes associated with ferroptosis in sepsis, especially in sepsis-induced systemic inflammation, are still unknown. It is necessary to further understand the biological functions of ferroptosis-related genes in sepsis.

Here, we first investigated ferroptosis-related differentially expressed genes (DEGs) in public datasets of adult and pediatric septic patients. Then, a functional analysis of ferroptosis-related DEGs and an analysis of the relationships among proteins were performed to understand the biological mechanisms. Furthermore, we explored crucial microRNAs (miRNAs) and long noncoding RNAs (lncRNAs) associated with biomarker genes. Finally, we examined the screened mRNAs, miRNAs and lncRNAs of interest using cecal ligation and puncture (CLP)-induced sepsis in mice. Our results may provide useful information regarding the mechanism of ferroptosis in sepsis, which could help us discover new targets for the clinical diagnosis of sepsis.

Materials and Methods

Results

Identification of ferroptosis-related DEGs

First, the GSE13904 (from children) and GSE28750 (from adults) datasets were downloaded from GEO. The Uniform Manifold Approximation and Projection (UMAP) and the principal component analysis (PCA) were used to check and visualize the grouped data. The results of UMAP showed that the gene transcriptional profiles of septic patients and controls were distinct from each other in both datasets (Figs. S1A and S1B). Similarly, the PCA plots showed that the two principal components contained 44.0% of the variance in GSE13904 and 59.1% of the variance in GSE28750 (Figs. S1C and S1D). In addition, the normalized signal intensity for each sample was analyzed in two datasets, respectively. The boxplots showed that the signal intensity of each sample was almost at the same median level (Figs. 1E and 1F). Thus, good normalization and repeatability were exhibited in the samples in these two datasets, which was a prerequisite for the subsequent analyses. Then, the ferroptosis-related DEGs were confirmed with a threshold of |log2 (fold-change)| > 0.5 and adjusted p value < 0.05. In GSE13904, a total of 53 ferroptosis-related DEGs (38 upregulated and 15 downregulated genes) were identified between the septic children and the control group. In GSE28750, a total of 75 ferroptosis-related DEGs (50 upregulated and 25 downregulated genes) were identified between the septic adults and the control group. The results are shown in volcano plots in Figs. 1A and 1B. The detailed DEG expression files are shown in Tables S3 and S4. In addition, a total of 34 genes including 26 upregulated and 8 downregulated genes showed apparent differences between the septic children and adults compared with those in the control groups in the two datasets (Figs. 2A, 2B and Table 1). And these 34 genes were marked in heatmaps that showed ferroptosis-related DEGs from GSE13904 (Fig. 2C) and GSE28750 (Fig. 2D).

Table 1:

Summary of 34 ferroptosis-related DEGs in sepsis.

	Driver	Marker	Suppressor
UP	PGD, MAPK14, WIPI1, G6PD, TLR4, ACSL4, MAPK1, ACVR1B, IDH1, SAT1, FLT3, CYBB, EPAS1, GABARAPL2	SLC2A3, ALOX5, JDP2, SLC40A1, MAFG, CAPG, MAP3K5, SLC7A5, DUSP1, CBS	LAMP2, CD44
DOWN	ZEB1, ATM, DPP4, LPIN1, PEBP1, SLC38A1, MAPK8		AKR1C3

DOI: 10.7717/peerj.13983/table-1

Functional and pathway enrichment analysis of ferroptosis-related DEGs

To explore the pathophysiological functions of the 34 coexpressed ferroptosis-related DEGs, GO and KEGG pathway analyses were performed. We found that peptidyl-serine phosphorylation, oxidation–reduction process, activation of mitogen- activated protein kinase (MAPK) activity and positive regulation of apoptotic process were significantly enriched among BPs. In addition, the statistically significant CC terms mainly included cytosol, extracellular exosome, integral component of plasma membrane, nucleus and intracellular. In the MF analysis, the enriched terms included MAP kinase activity, protein serine/threonine kinase activity, enzyme binding, adenosine triphosphate (ATP) binding and protein homodimerization activity. These results are shown in Fig. 3A. Subsequently, the KEGG pathway analysis revealed that the main enriched pathways were involved in the inflammatory response, including the forkhead box O (FoxO) signaling pathway, Toll-like receptor signaling pathway, tumor necrosis factor (TNF) signaling pathway and MAPK signaling pathway (Fig. 3B). In addition, a chord plot revealed that more than half of the coexpressed ferroptosis-related DEGs (19 of 34) were involved in the regulation of the top ten enriched KEGG pathways (Fig. 3D). Specific information regarding the statistically significant terms (BP, CC, MF and KEGG pathway analyses) with the number of enriched genes are listed in Table S5. Moreover, the biological function enrichment analysis of these 34 ferroptosis-related DEGs by Metascape indicated similar results and revealed the same terms such as central carbon metabolism in cancer, MAPK cascade and apoptotic signaling pathway (Fig. 3C). Overall, the MAPK signaling pathway seemed to be the key factor that modulated ferroptosis in sepsis.

Identification of key ferroptosis-related genes involved in sepsis

A PPI network based on the 34 ferroptosis-related DEGs was constructed and visualized by selecting the proteins with an interaction score ≥0.4 (Fig. 4A). The results indicated that there were 27 nodes and 47 edges that represented genes and interactions between genes respectively (Table S6). Meanwhile, a significant clustering module composed of eight key genes which included six upregulated genes (MAPK14, MAPK1, MAP3K5, toll-like receptor (TLR) 4, cytochrome B-245 beta chain (CYBB) and dual specificity phosphatase (DUSP) 1) and two downregulated genes (MAPK8 and ataxia telangiectasia-mutated (ATM)) was screened with a threshold of MCC score >4 (Fig. 4B). The MCC scores of the 27 interacting proteins are listed in Table S7. Subsequently, a biological pathway enrichment analysis was performed to identify the functions of these eight key genes. The GO and KEGG pathway results, showed that the MAPK signaling pathway was significantly enriched (adjusted p value = 3.53E−05). The top eight GO terms and KEGG pathways are illustrated in a network diagram and a bubble chart, respectively (Figs. 5A and 5B). Moreover, five of the eight identified genes (MAPK14, MAPK8, DUSP1, MAP3K5 and MAPK1) were involved in the MAPK signaling pathway, and the specific enrichment information is listed in Table S8.

Construction of the mRNA–miRNA–lncRNA network

We chose a significant clustering module composed of eight key genes as genes of interest and analyzed the mRNA–miRNA and miRNA–lncRNA interactions. We identified a total of 168 miRNAs targeting six genes (MAPK14, TLR4, CYBB, MAPK1, MAPK8 and ATM) via utilizing both the miRWalk and miRTarBase databases with thresholds of p < 0.05 and the 3′UTR as the gene-binding region (Fig. 6A, Table S9). Specifically, 10 miRNAs (hsa-miR-7113-3p, hsa-miR-6892-3p, hsa-miR-4685-3p, hsa-miR-4287, hsa-miR-199a-3p, hsa-miR-17-5p, hsa-miR-106a-5p, hsa-let-7b-5p, hsa-miR-6734-3p and hsa-miR-214-3p) were listed because they linked several genes (≥2, Table 2). LncBase was used to predict the upstream lncRNAs of these 10 miRNAs. Forty-nine lncRNAs were identified to correlate with five miRNAs (hsa-miR-199a-3p, hsa-miR-17-5p, hsa-miR-106a-5p, hsa-let-7b-5p and hsa-miR-214-3p) targeting four genes (MAPK14, TLR4, MAPK1 and MAPK8). The mRNA–miRNA–lncRNA network is shown in Fig. 6B. In addition, among these 49 lncRNAs, 10 lncRNAs (CTB-89H12.4, LINC00657, CTB-89H12.4, RP11-553L6.5, NEAT1, XLOC_013866, RP1-309I22.2, TRG-AS1, TUG1 and HOXA10-HOXA9) had the highest predicted scores (>0.9), illustrating their strong correlation with the targeting miRNAs (Table S10). Previous studies have shown that let-7b-5p and nuclear-enriched abundant transcript 1 (NEAT1) are involved in the regulation of MAPK1 and participate in ferroptosis in cancer cells (Chen, Xin & Zhang, 2019; Du et al., 2019; Dong et al., 2021; Wu & Liu, 2021). However, their roles in sepsis are still unclear. Here, we chose let-7b-5p and NEAT1 for further investigation.

Table 2:

Summary of 10 miRNAs linked several genes.

miRNA	Target genes
hsa-miR-7113-3p	CYBB	MAPK14
hsa-miR-6892-3p
hsa-miR-4685-3p
hsa-miR-4287	MAPK1	MAPK14
hsa-miR-199a-3p
hsa-miR-17-5p
hsa-miR-106a-5p
hsa-let-7b-5p	MAPK1	TLR4
hsa-miR-6734-3p	CYBB	MAPK1
hsa-miR-214-3p	MAPK1	MAPK14	MAPK8

DOI: 10.7717/peerj.13983/table-2

Expression levels of ferroptosis-related mRNAs, miRNAs and lncRNAs in septic mice during acute inflammation

CLP, which is a model of polymicrobial sepsis, was used to induce acute inflammation in mice. Twenty-four hours after the CLP procedure, 7 of 10 mice survived, and multiorgan (including lung, kidney and liver) injury was observed in six mice, indicating the successful establishment of the sepsis model (Fig. S2). Then, peripheral blood leukocytes were obtained from the CLP group and sham group and the total RNA was extracted. The expression levels of let-7b-5p, NEAT1 and five ferroptosis-related genes (MAPK14, MAPK8, DUSP1, MAP3K5 and MAPK1) were detected by qRT–PCR. The results showed that MAPK14, MAPK8, MAP3K5, MAPK1 and NEAT1 were upregulated in the CLP group compared with those in the sham group, while DUSP1 and let-7b-5p were downregulated in the septic mice during acute inflammation (Fig. 7).

Discussion

Sepsis is currently regarded as profound systemic inflammation and a dysregulated immune response to invasive pathogens that results in multiple organ injury (Hotchkiss et al., 2016). Notably, disorders of innate and adaptive immunity are important factors that promote the development of sepsis (Hotchkiss et al., 2016). In recent years, a novel form of programmed cell death called ferroptosis has been defined. The immunological consequences of ferroptosis include the death of leukocyte subsets and the corresponding loss of immune function (Tang et al., 2021). Although the specific physiological function of ferroptosis has not been clearly elucidated, emerging evidence demonstrates that ferroptosis is associated with multiple human diseases, such as cancer, degenerative diseases and inflammation-related diseases (Han et al., 2020; Sun et al., 2020). In this study, we analyzed the differences in ferroptosis-related genes between septic patients and healthy controls. Importantly, 26 upregulated and 8 downregulated genes were identified in both septic children and adults. The MAPK signaling pathway seemed to be a key factor in modulating ferroptosis and was associated with these genes in sepsis. Moreover, the significantly differentially expressed noncoding RNAs let-7b-5p and NEAT1, which are involved in MAPK1 regulation, were identified in peripheral blood leukocytes from septic mice. This study may help us better understand the role of ferroptosis in sepsis.

It is essential to identify disease-associated gene sets by coexpression analysis to obtain new insights into disease biology (Li et al., 2021). Therefore, the 34 ferroptosis-related DEGs that were coexpressed in children and adult septic patients were screened out followed by a subsequent enrichment analysis. The GO analysis based on the DAVID database revealed that the oxidation–reduction process, activation of MAPK activity, positive regulation of apoptotic processes and extracellular exosomes were significantly enriched. Regarding cellular reductive/oxidative (redox)-based metabolic processes, imbalanced redox homeostasis drives ferroptosis by causing lethal lipid peroxidation (Jiang, Stockwell & Conrad, 2021). We noticed that the dysregulation of ferrous iron (Fe²⁺) was another important factor leading to ferroptosis. A prior study proved that prominin 2 participates in ferroptosis resistance in mammary epithelial and breast carcinoma cells by promoting exosome-mediated iron export (Brown et al., 2019). Moreover, adenosine 5′-monophosphate (AMP)-activated protein kinase (AMPK)-mediated phosphorylation of acetyl-CoA carboxylase (ACC) and polyunsaturated fatty acid biosynthesis have been confirmed to be involved in ferroptosis (Lee et al., 2020). Although ferroptosis is an iron-dependent form of nonapoptotic regulated cell death, PERK-eIF2α-ATF4-CHOP signaling pathway-mediated p53 upregulated modulator of apoptosis (PUMA) expression is involved in the synergistic interaction between ferroptosis and apoptosis (Lee et al., 2018). Therefore, combined with previous studies, the biological function enrichment results imply that ferroptosis plays a potential role in the development of sepsis. Prior studies have noted a close correlation between ferroptosis and inflammation, as ferroptotic cells produce excessive reactive oxygen species (ROS) (Han et al., 2020). We noticed that both the KEGG pathway analysis and Metascape GO analysis revealed significant enrichment of multiple inflammatory pathways and functions such as the Toll-like receptor (TLR) signaling pathway, tumor necrosis factor (TNF) signaling pathway, MAPK signaling pathway, response to oxidative stress and ROS metabolic process. It is well known that activated TLRs are responsible for the early activation of inflammatory genes and are involved in sepsis-associated cytokine storms (Hotchkiss et al., 2016). TLRs are also involved in iron metabolism in sepsis by transcriptionally downregulating ferroportin (FRP), the exporter of intracellular iron, indicating their roles in ferroptosis (Liu et al., 2021). The TNF signaling pathway and MAPK signaling pathway play important roles in oxidative stress, inflammatory cytokines and programmed cell death. A previous study demonstrated that TNF-α induced the generation of ROS and that extracellular vesicle (EV) release required the upregulation of glutaminase to participate in neuroinflammation (Wang et al., 2017). Because glutaminase is involved in ferroptosis, the TNF signaling pathway may participate in ferroptosis during the inflammatory response. In addition, a recent study focusing on neonatal mice reported that a member of the lipocalin family named lipocalin 2 plays important roles in lipopolysaccharide (LPS)-induced inflammation and oxidative stress by modulating ferroptosis via the MAPK/ERK pathway (Wang et al., 2022). Therefore, we have reason to believe that the MAPK signaling pathway may play potential roles in sepsis-induced ferroptosis.

To further reveal the mechanism underlying the biological processes, a PPI analysis based on the 34 ferroptosis-related DEGs was performed. A key clustering module was screened based on the CytoHubba plug-in unit in Cytoscape. The results indicated that 8 genes (MAPK14, MAPK8, TLR4, MAP3K5, CYBB, DUSP1, MAPK1 and ATM) might be closely related to the occurrence of sepsis. Among them, MAPK14, TLR4, MAPK1 and MAPK8 have been confirmed to be involved in proinflammatory cytokine responses, especially in the TLR/IL-1R pathway cascade. A previous study investigating the hyperinflammatory phase during sepsis suggested that activated autophagy degenerated an E3 ubiquitin ligase and scaffold protein called Pellino 3 in TLR4-signaling, resulting in a reduction in phosphorylated MAPK14, MAPK8 and MAPK1 and ultimately inhibiting proinflammatory IL-1β expression (Giegerich et al., 2014). Moreover, CYBB, also known as nicotinamide adenine dinucleotide phosphate (NADPH) oxidase 2 (NOX2), is not only involved in the generation of the autophagy protein LC3 during TLR signaling, but also activated by the Hippo pathway effector PDZ-binding motif (TAZ) to promote ferroptosis (Huang et al., 2009; Yang et al., 2020). Since autophagy regulates cellular iron homeostasis and is required for ferroptosis-associated ROS accumulation, the potential roles of the autophagy-related signaling pathway in ferroptosis have been suggested (Gao et al., 2016). MAP3K5, which is also known as apoptosis signal-regulating kinase 1 (ASK1), has been proven to be associated with erastin- and RAS-selective lethal 3 (RSL3)-induced ferroptosis through the p38/c-Jun N-terminal kinase (JNK) pathway in multiple cell lines (Hattori et al., 2017). A recent study also indicated that ASK1 expression was increased in septic mice and that the inhibition of ASK1 impaired JNK-mediated cytokine production during LPS-induced endothelial inflammation (Miller et al., 2021). In the p38/JUN pathway, a type of phosphatase with dual specificity for tyrosine and threonine named DUSP1 plays an important role in autophagy-related ferroptosis (Chen et al., 2021). Moreover, the ectopic expression of DUSP1 in LPS-stimulated macrophages inhibited the release of TNF-α and IL-6 (Hoppstädter & Ammit, 2019). In contrast, the depletion of DUSP1 resulted in the excessive release of inflammatory cytokines (Hoppstädter & Ammit, 2019). ROS-mediated DNA damage (DDR) is a key factor in ferroptosis. ATM is an important coordinator of the DDR, mainly because the depletion of ATM results in the accumulation of ROS and damage to cellular DNA (Huff, Yan & Clemens, 2021). A previous study indicated that ATM is essential for Ikβ phosphorylation and the translocation of nuclear factor kappa-B (NF-κB) to the nucleus and that ATM participates in the TLR-mediated inflammatory response (Huff, Yan & Clemens, 2021). Importantly, the NF-κB signaling pathway has been verified to be involved in erastin-induced ferroptosis (Oh et al., 2019). These results suggested that changes in MAPK14, TLR4, MAP3K5, CYBB, DUSP1, MAPK1, MAPK8 and ATM may be partially associated with ferroptosis in the sepsis-induced inflammatory response. Next, GO and KEGG analyses of these eight key genes were performed. The results implied that the MAPK signaling pathway, which consists of MAPK14, DUSP1, MAP3K5, MAPK1 and MAPK8 was significantly enriched. A previous study demonstrated that the excessive activation of the MAPK pathway might promote iron-related ROS generation by inhibiting cystine (Cys2) or mitochondrial VDAC2/3 and subsequently induce cell ferroptosis (Han et al., 2020). In addition to ROS generation, MAPK is associated with iron metabolism-related inflammatory signaling. For example, the administration of ferritin light chain (FTL) to macrophages suppressed ERK1/2 activation, decreased the production of NF-κB-induced proinflammatory factors and protected septic mice (Zarjou et al., 2019). The increase in glutathione peroxidase (GPX) 4 and the labile iron pool (LIP) in sepsis-induced myocardial injury indicates that ferroptosis plays a role in sepsis. The regulatory mechanism of systemic inflammation and the immune response in MAPK-related ferroptosis needs to be further studied.

It has been reported that noncoding RNAs (including miRNAs and lncRNAs) play significant roles in the diagnosis, severity, and prognosis of septic patients (Wu et al., 2020). First, we identified 10 miRNAs that targeted more than one gene. Furthermore, 5 of these 10 miRNAs (hsa-miR-199a-3p, hsa-miR-17-5p, hsa-miR-106a-5p, hsa-let-7b-5p and hsa-miR-214-3p) predicted upstream lncRNAs. In previous studies, miR-199a-3p has been identified as a potential biomarker of neonatal sepsis (Abdelaleem et al., 2022). miR-17-5p and miR-106a-5p participate in sepsis-associated cardiomyopathy or acute kidney injury (AKI) by regulating oxidative stress via high mobility group box (HMBG) 1 (Qiu et al., 2021; Xu, Ma & Yang, 2021). miR-214-3p attenuates sepsis-induced myocardial dysfunction by inhibiting autophagy (Sang et al., 2020). Let-7b-5p was significantly downregulated in renal tubular epithelial cells during sepsis-induced AKI (Gao et al., 2020). In terms of upstream lncRNAs, taurine-upregulated gene (TUG) one was downregulated and NEAT1 was upregulated in septic patients (Gao et al., 2020; Wang et al., 2020). Our results suggest the roles of the mRNA–miRNA–lncRNA network in sepsis. Previous studies have demonstrated that the noncoding RNAs let-7b-5p and NEAT1 participate in ferroptosis by regulating the MAPK signaling pathway (Chen, Xin & Zhang, 2019; Du et al., 2019). Combined with our gene enrichment analysis results, the MAPK-related mRNA–miRNA–lncRNA network may regulate ferroptosis during sepsis. The latest study found that the ferroptosis-related gene LPIN1, which is associated with the immune status, could become a reliable biomarker of patient survival in sepsis (Dai et al., 2022). To verify the changes in the above screened ferroptosis-related key mRNAs, miRNAs and lncRNAs in the immune system, we next detected the expression of MAPK14, DUSP1, MAP3K5, MAPK1, MAPK8, let-7b-5p and NEAT1 in peripheral blood leukocytes from septic mice during the acute inflammatory phase. The upregulation of MAPK14, MAP3K5, MAPK1, MAPK8 and NETA1 indicated their important roles in promoting acute systemic inflammation. In contrast, the downregulation of DUSP1 and let-7b-5p demonstrated their protective roles during acute inflammation. Notably, the mRNA expression of DUSP1 and MAPK8 is contradictory between the human datasets and the CLP mouse model. One possible reason is that the types of samples detected are different. In the human datasets, the array-based gene expression profiles were obtained from whole blood, including plasma, blood cells, platelets, etc. In the CLP mouse model, we extracted the total RNA from peripheral blood leukocytes in mice, which included lymphocytes, neutrophils, monocytes, etc. Thus, it is possible that DUSP1 and MAPK8 play different biological roles in immune cells and whole blood, and further studies are needed. Despite the regulatory network of MAPK molecules involved in ferroptosis has been shown to be changed in sepsis, the specific mechanisms of MAPK- related ferroptosis in sepsis-induced immune dysregulation and systemic inflammation need to be further studied.

There are still several limitations in the present study. First, the data we analyzed were downloaded from GEO datasets, and further prospective clinical studies are necessary to validate the observations. Second, the potential mRNA–miRNA–lncRNA network we indicated needs further experimental validation. In addition, with an in-depth study of ferroptosis, basic experimental studies are needed to explore the relationship between ferroptosis and sepsis.

Conclusion

The MAPK signaling pathway may play a key role in regulating ferroptosis during sepsis. This study provides a valuable resource for future studies investigating the mechanism of MAPK-related ferroptosis in sepsis. More comprehensive and in-depth studies should be conducted to explore the associations between ferroptosis-related genes and sepsis-induced immune dysregulation and systemic inflammation.

Supplemental Information