Elucidating shared genes and pathways in programmed cell death with necrotizing enterocolitis: insights into novel therapeutic targets and glutathione

Introduction

Necrotizing enterocolitis (NEC) is the most common acute intestinal necrosis in neonates, characterized by high morbidity and mortality. The overall incidence of NEC is approximately 7%, with a rising trend observed in infants of lower gestational age and lower birth weight (1). In very low-birth-weight infants, NEC is estimated to affect 10% to 15% (2). However, the clinical manifestations of NEC are nonspecific, and reliable indicators for early diagnosis remain limited. Common clinical features include feeding intolerance, abdominal distension, and hematochezia, while severe cases may present with respiratory and circulatory failure as well as intestinal necrosis. Despite advances in neonatal care over the past decades, the incidence and mortality of NEC have not decreased (3). With improvements in the survival of preterm infants, the baseline incidence of NEC is expected to increase further. Therefore, a further understanding of its pathogenesis may help reduce the socioeconomic burden and improve patient outcomes.

Preventing intestinal necrosis remains an urgent clinical challenge in the management of NEC. However, no effective pharmacological treatment is currently available. Early management typically includes bowel rest, gastrointestinal decompression, and antibiotic therapy. Surgical resection of necrotic intestinal tissue is often required, but postoperative complications such as wound infection, intestinal stricture, and short bowel syndrome are common and contribute to poor prognosis (4). There is an urgent need to develop safe and effective therapeutic agents for early intervention to prevent the progression of intestinal necrosis in NEC. Growing evidence indicates that oxidative stress and inflammation are key drivers of intestinal epithelial injury and disease progression in NEC, and that these processes mutually amplify each other, forming a vicious cycle (5-7). In this context, glutathione (GSH), a small molecule with multiple biological activities, has been successfully applied in the clinical treatment of various diseases (8,9). Importantly, GSH is one of the most abundant intracellular thiol antioxidants and a central determinant of redox homeostasis. It not only directly scavenges reactive oxygen or nitrogen species but also serves as an essential reducing substrate for GSH peroxidases, thereby limiting lipid peroxidation and attenuating inflammation exacerbated by redox imbalance, which processes closely linked to regulated cell death (10). Notably, redox imbalance-driven epithelial injury associated with lipid peroxidation (e.g., ferroptosis) has been increasingly implicated in NEC pathogenesis (11,12). From a translational perspective, the relatively well-characterized safety profile and practical feasibility of GSH supplementation make it an attractive candidate for early intervention in NEC (13-15). Based on these considerations, in our pilot experiments, we found that GSH levels were decreased in the peripheral blood of NEC patients. However, the role and underlying mechanisms of GSH in NEC remain unclear.

It has been shown that programmed cell death is an important pathogenic mechanism in NEC (16). However, the key molecules involved remain to be further elucidated. To identify common biomarkers and pathways of programmed cell death in NEC, gene sets from microarray data and GeneCards were first analyzed. Various bioinformatics analyses were performed to reveal critical biomarkers and biological functions, providing valuable insights into potential novel therapeutic targets. Subsequently, the differential expression of Toll-like receptor 4 (TLR4), NLR family pyrin domain containing 3 (NLRP3), and nuclear factor kappa B subunit 1 (NFKB1) was validated in NEC mouse models, cell models, and NEC patient samples. Furthermore, molecular docking analysis was performed to explore potential interactions between GSH and hub genes encoded proteins, providing structural clues supporting GSH-based therapeutic development for NEC. We present this article in accordance with the ARRIVE and MDAR reporting checklists (available at https://tp.amegroups.com/article/view/10.21037/tp-2025-1-860/rc).

Methods

Data collection

The Gene Expression Omnibus (GEO) (http://www.ncbi.nlm.nih.gov/geo) is a public functional genomics database that includes high-throughput gene expression data, chips, and microarrays. In this study, the GSE46619 microarray dataset, comprising 5 NEC samples and 9 control samples, was downloaded from the GEO database (http://www.ncbi.nlm.nih.gov/geo). Gene sets related to ferroptosis, pyroptosis, necroptosis, and autophagy were obtained from the GeneCards database.

Identification of differentially expressed genes (DEGs)

DEGs in the dataset were identified using the “limma” package in R version 4.2.1. The Benjamini & Hochberg method was applied to control the false positive rate, and the threshold for DEGs selection was set at an adjusted P value (P_adj)<0.01. Subsequently, the DEG set was intersected with gene sets associated with different forms of cell death to obtain a subset of potential cell death-related genes in NEC. A Venn diagram was constructed using the “ggplot2” (version 3.3.6) and “VennDiagram” packages in R 4.2.1 to visualize the overlap. A volcano plot and a heatmap of the DEGs were generated using the “ggplot2” (version 3.3.6) and “ComplexHeatmap” (version 2.13.1) packages, respectively.

Functional annotation and pathway enrichment of DEGs

Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis was performed on the DEGs and associated genes using the “ggplot2” (version 3.3.6) package in R version 4.2.1. Gene Ontology (GO) enrichment analysis provides a standardized system for the functional annotation of genes and proteins, including biological process (BP), molecular function (MF), and cellular component (CC) categories. KEGG enrichment analysis offers insights into the identified pathways and their interactions. Enriched GO terms and KEGG pathways with an P_adj<0.05 were considered statistically significant.

Protein-protein interaction (PPI) and identification of hub genes

PPI network of the DEGs was constructed using the Search Tool for the Retrieval of Interacting Genes (STRING) database (http://string-db.org/), with a confidence score ≥0.4 set as the threshold for interactions. Subsequently, the PPI network was visualized using Cytoscape software. The CytoHubba plugin in Cytoscape was applied to identify hub gene modules from the PPI network based on higher topological scores.

Evaluating the immune and stromal cell infiltration

We first performed Spearman correlation analysis to investigate the association between key genes and the inflammatory factors IL1B and IL6. To estimate the immune cell composition in NEC intestinal tissues, the single-sample gene set enrichment analysis (ssGSEA) algorithm in the “GSVA” package was applied in R, using immune cell markers referenced from a previous study (17). Spearman correlation analysis was further conducted between key genes and various immune cell types using the “ggplot2” package in R version 4.2.1, and the results were visualized in a heatmap.

Construction of transcription factors (TFs) network

NetworkAnalyst 3.0 (18), a powerful visual analytics tool for conducting in-depth gene expression profiling and meta-analysis, was employed to establish co-expression networks of TFs with the hub genes and expression-related genes, respectively. The TFs-gene interaction network was constructed based on the JASPAR database (http://jaspar.genereg.net).

Molecular docking validation of GSH

In this study, molecular docking experiments were conducted to investigate the interaction between GSH and the core target proteins. The corresponding protein structures of the core targets were obtained from the Protein Data Bank (PDB), and the molecular structure of GSH was downloaded from PubChem (https://ncbi.nlm.nih.gov/). AutoDock Tools was used to preprocess the core proteins and GSH, including removal of water molecules, addition of hydrogen atoms, and charge assignment.

To simulate the binding mode of GSH with the target proteins, molecular docking was performed using AutoDock Vina, and the binding affinity was subsequently calculated to evaluate the binding efficiency of the ligand to the receptor. Lower binding energy values indicated stronger binding affinity. A binding energy <−4 kcal/mol was considered indicative of a stable interaction with the target proteins (19). The docking results were further processed and visualized using PyMOL and LigPlot+.

Hypoxia and inflammation model in Caco-2 cells

Caco-2 cells provided by the Guangdong Provincial Key Laboratory of Research in Structural Birth Defect Disease, Guangzhou Women and Children’s Medical Center, Guangzhou Medical University, were recovered from liquid nitrogen and rapidly thawed in a 37 ℃ water bath until fully liquefied (within 2 minutes). The cells were transferred into a centrifuge tube containing 2 mL Dulbecco’s Modified Eagle’s Medium (DMEM), centrifuged at 1,000 rpm for 5 minutes, and resuspended in 4 mL DMEM supplemented with 10% fetal bovine serum (FBS). The suspension was seeded into a T25 culture flask and incubated at 37 ℃ with 5% CO₂. The medium was first replaced 48 hours after recovery and subsequently every 72 hours. When the cells reached 80–90% confluency, they were passaged by washing once with 5 mL 1× phosphate-buffered saline (PBS), digesting with 1 mL trypsin at 37 ℃ for approximately 5 minutes, and collecting the detached cells into 3 mL DMEM. After centrifugation at 1,000 rpm for 5 minutes, the cell pellet was resuspended in 12 mL DMEM with 10% FBS and evenly distributed into three T25 flasks for expansion. For experimental stimulation, cells were seeded into 6-well plates at a density of 2×10⁵ cells per well in 2 mL DMEM containing 10% FBS. After 48 hours, cell morphology and density were assessed. Upon reaching 70–80% confluency, lipopolysaccharide (LPS) was added at a final concentration of 10 µg/mL, and the plates were transferred into a hypoxic chamber. After 24 hours of stimulation, cell morphology, density, and floating status were recorded. Cells were then lysed with 1 mL TRIzol reagent per well and stored at −80 ℃ for further analysis.

NEC mouse model

The NEC model was established based on previously described methods and was simply improved (20-22). Briefly, 4-day-old C57BL/6 mice received adult mouse cecum-derived bacteria [3.5×10⁷ colony-forming unit (CFU)/mouse] via gavage. From the next experimental day, mice were stressed by a hypoxia-cold shock cycle twice per day for 4 days. During the experimental days, NEC mice were fed 40 µL formula every 4 hours with Esbilac canine milk replacer (PetAg) at a ratio of 2:1. Breast-fed control animals were kept with the dam until the time of euthanasia. For the GSH treatment group, GSH was administered by intraperitoneal injection (i.p.) at a dose of 200 mg/kg/day starting from postnatal day 5 until the end of the experiment. Small intestines were placed in buffered formalin for hematoxylin and eosin (H&E) staining to estimate the H&E scores graded 0–4 (23). The H&E scores of 2–4 are defined as the occurrence of NEC. All animal experiments were performed at the Experimental Animal Center of Guangzhou Women and Children’s Medical Center, Guangzhou Medical University, Guangzhou, China. Experiments were performed under a project license (No. RSDW-2023-01296) granted by the Institutional Animal Care and Use Committee of Guangzhou Medical University, in compliance with national guidelines for the care and use of animals. A protocol was prepared before the study without registration.

Human study participants

Human study participants with NEC and controls had their intestinal tissues collected from the Guangzhou Women and Children’s Medical Center. All NEC cases were classified as Bell stage II or III and the ileum tissue from the afflicted area adjacent to necrosis was collected (24). The control group included mecmecial ileus, congenital intestinal atresia, intestinal duplication, Meckel’s diverticulum and other non-inflammatory diseases (24) (Table 1). Blood samples from Preterm, Term and NEC were collected from the Guangzhou Women and Children’s Medical Center (Table 2). The study was conducted in accordance with Helsinki and its subsequent amendments. The study was approved by the Human Review Board of Guangzhou Women and Children’s Medical Center (Human Investigation Committee No. 2021-174A01). Informed consent was obtained from the parents of the infant.

Reverse transcriptase-quantitative polymerase chain reaction

Intestinal tissues were homogenized in 500 µL of TRIzol reagent, followed by the addition of 100 µL of chloroform. After thorough mixing and standing at room temperature for 10 minutes, samples were centrifuged at 12,000 rpm for 15 minutes at 4 ℃. The upper aqueous phase was transferred to a new tube, mixed with 200 µL of isopropanol, and incubated at room temperature for 10 minutes. RNA was pelleted by centrifugation at 12,000 rpm for 10 minutes at 4 ℃, washed with 75% ethanol, and centrifuged again at 12,000 rpm for 5 minutes at 4 ℃. After air-drying, RNA was dissolved in 40 µL of DEPC-treated water. Reverse transcription was performed using the HiScript III RT SuperMix for qPCR (+ gDNA wiper) kit (Vazyme, Nanjing, China; R323-01) with 1 µg of RNA in a 16 µL reaction system, according to the manufacturer’s instructions. cDNA synthesis involved incubation at 42 ℃ for 2 minutes with gDNA wiper, followed by 37 ℃ for 10 minutes and 85 ℃ for 5 seconds after adding HiScript III RT SuperMix. qPCR was conducted using Universal SYBR qPCR Master Mix (Vazyme) on a Thermo Real-Time PCR System. The 10 µL reaction mixture consisted of 5 µL SYBR MM, 0.2 µL forward primer (10 µM), 0.2 µL reverse primer (10 µM), 3 µL cDNA, and 1.6 µL nuclease-free water. Amplification conditions were 95 ℃ for 30 seconds (initial denaturation), followed by 40 cycles of 95 ℃ for 10 seconds and 60 ℃ for 30 seconds, with subsequent melting curve analysis. Primer sequences were designed based on GenBank data. Relative gene expression was calculated using the 2^−ΔΔCt method. The Primer sequences are listed in Table 3.

Immunofluorescence staining

Paraffin-embedded tissue sections were deparaffinized, rehydrated, and subjected to antigen retrieval using either EDTA or citric acid buffer. After permeabilization, sections were treated with 50 µL of 3% hydrogen peroxide (H₂O₂) at room temperature for 30 minutes to block endogenous peroxidase activity, followed by three washes with 1× PBS. Sections were then blocked with 20% goat serum for 1 hour at room temperature and incubated with primary antibodies overnight at 4 ℃. After three washes with 1× PBS, sections were incubated with either Goat Anti-Mouse IgG (1:500, E032410-01, Earthox, Burlingame, USA) or Goat Anti-Rabbit IgG (1:500, E032220-01, Earthox) for 1 hour at room temperature. Nuclei were counterstained with 4',6-diamidino-2-phenylindole (DAPI) (G1401, Servicebio, Wuhan, China), and fluorescence images were acquired using a fluorescence microscope. The following primary antibodies were used: TLR4 antibody (1:350 dilution, AF7017, Affinity, Changzhou, China); NLRP3 antibody (1:350 dilution, AF4620, Affinity); NFKB1 antibody (1:350 dilution; 66992-1-Ig, Proteintech, Wuhan, China). All samples were acquired on the same microscope platform using identical imaging parameters (laser power, exposure time, gain and resolution), and signal saturation was avoided. The representative images were also displayed using the same lookup table/contrast limits to minimize visual bias. Quantification was performed in ImageJ. A tissue mask was first generated automatically based on the DAPI channel to exclude the luminal empty space and background, yielding a continuous tissue region of interest. This mask was then applied to the target-protein channel, followed by background subtraction, and the mean fluorescence intensity was calculated. For each section, three regions were randomly selected for statistical analysis, and the averaged value was taken as the mean fluorescence intensity for that section. Statistical analyses were conducted using GraphPad Prism (version 9), and all procedures were performed under double-blinded conditions.

Statistical analysis

All data are presented as mean ± standard error of the mean (SEM). Statistical analyses were performed using GraphPad Prism (version 9.3). For comparisons between two groups, an unpaired Student’s t-test was used. P<0.05 was considered statistically significant.

Results

Identifying co‑expressed genes module associated with NEC and programmed cell death

A total of 2,784 DEGs, including 1,252 upregulated and 1,532 downregulated genes, were identified using the limma package in R software. A total of 1,839 ferroptosis-related genes, 874 pyroptosis-related genes, 871 necroptosis-related genes, and 9,950 autophagy-related genes were retrieved from the GeneCards database (available online: https://cdn.amegroups.cn/static/public/tp-2025-1-860-1.xlsx). Subsequently, we intersected the DEGs with the programmed cell death-related gene sets. The overlapping genes were visualized using a Venn diagram, volcano plot, and heatmap (Figure 1A-1C). Ultimately, 11 genes shared between programmed cell death and NEC were identified: CASP6, RBMX, BNIP3L, S100A4, VIM, TLR4, NFKB1, NLRP3, ANXA1, PANX1, and NFKBIA.

Figure 1 Identification of differentially expressed genes related to programmed cell death in NEC. (A) Venn diagram of DEGs; (B) volcano plot of DEGs in GSE46619; (C) the heatmaps of differentially DEGs; (D) GO biological process enrichment analysis; (E) GO cell component and molecular function enrichment results; (F) KEGG pathway enrichment results. An adjusted P<0.05 was identified as significantly changed GOs and KEGG. The bubble size represents the number of genes associated with each term. The color of each bubble represents the adjusted p value: the redder the color, the higher the enrichment. BP, biological process; CC, cellular component; ColAUTO, column of autophagy-related genes; ColFER, column of ferroptosis-related genes; ColNER, column of necroptosis-related genes; ColPYRO, the column of pyroptosis-related genes; DEGs, differentially expressed genes; GO, Gene Ontology; KEGG, Kyoto Encyclopedia of Genes and Genomes; MF, molecular function; NEC, necrotizing enterocolitis.

Next, GO and KEGG pathway enrichment analyses were performed to explore the potential mechanisms of these 11 shared genes. The GO enrichment results revealed that the BPs were mainly enriched in cellular response to LPS, cellular response to molecules of bacterial origin, regulation of interleukin-1 production, regulation of inflammatory response, positive regulation of interleukin-1 beta production, regulation of T-helper 2 cell differentiation, and regulation of NIK/NF-kappaB signaling (Figure 1D). CCs and MFs were primarily enriched in phagocytic cup, scaffold protein binding, calcium-dependent protein binding, RAGE receptor binding, phospholipase inhibitor activity, intermediate filament binding, cysteine-type endopeptidase activity involved in apoptotic process, lamin binding, NAD(P)+ nucleosidase activity, cyclic ADP-ribose generating, lipase inhibitor activity, peptidoglycan binding, cadherin binding involved in cell-cell adhesion, leak channel activity, and narrow pore channel activity (Figure 1E). KEGG pathway enrichment analysis showed significant enrichment in NOD-like receptor signaling pathway, lipid and atherosclerosis, Yersinia infection, influenza A, legionellosis, pathogenic Escherichia coli infection, shigellosis, Salmonella infection, programmed cell death ligand 1 (PD-L1) expression and PD-1 checkpoint pathway in cancer, NF-κB signaling pathway, and Toll-like receptor signaling pathway (Figure 1F).

Construction of PPI network and identification of hub genes

PPI network of the DEGs was constructed using the STRING database (Figure 2A) to analyze the interactions among the DEGs. The results showed that BNIP3L and RBMX were isolated from the interaction network. The remaining nine DEGs were subsequently imported into Cytoscape software for further analysis. To identify hub genes within the PPI network, the CytoHubba plugin in Cytoscape was applied. According to CytoHubba, the top six genes were selected as potential hub genes, including TLR4, NLRP3, NFKB1, NFKBIA, VIM and ANXA1 (Figure 2B).

Figure 2 PPI network analysis and hub genes identification. (A) The PPI among common differentially expressed genes; (B) PPI network for the top 6 hub genes by cytoHubba; (C) the expression level of TLR4 in the GSE46619 dataset; (D) the expression level of NLRP3 in the GSE46619 dataset; (E) the expression level of VIM in the GSE46619 dataset; (F) the expression level of NFKB1 in the GSE46619 dataset; (G) the expression level of NFKBIA in the GSE46619 dataset; (H) the expression level of ANXA1 in the GSE46619 dataset. ***, P<0.001. PPI, protein-protein interaction.

Furthermore, violin plots were used to analyze the expression levels of these potential hub genes in NEC disease and control datasets. The results demonstrated that TLR4, NLRP3, NFKB1, NFKBIA, VIMand ANXA1 were significantly upregulated in the intestinal tissues of children with NEC disease compared with controls (Figure 2C-2H).

Relationship between hub genes and inflammation

Immune cell infiltration and inflammatory storm are recognized as key mechanisms driving the progression of NEC. In this study, we first analyzed the infiltration of 24 immune cell types in intestinal tissues from NEC patients and healthy controls (Figure 3A). The results revealed a significant increase in natural killer (NK) CD56dim cells, NK cells, effector memory T cells (Tem), macrophages, neutrophils, plasmacytoid dendritic cells (pDC) and Th1 cells in NEC tissues, while mast cells, central memory T cells (Tcm), and follicular helper T cells (TFH) were significantly decreased.

Figure 3 The hub genes exacerbate inflammatory responses. (A) Analysis of inflammatory infiltration in NEC dataset; (B) expression profile of the hub genes across different immune cell types in NEC dataset; (C) IL1B expression correlated with TLR4 in the NEC database; (D) IL6 expression correlated with TLR4 in the NEC database; (E) IL1B expression correlated with NLRP3 in the NEC database; (F) IL6 expression correlated with NLRP3 in the NEC database; (G) IL1B expression correlated with VIM in the NEC database; (H) IL6 expression correlated with VIM in the NEC database; (I) IL1B expression correlated with NFKB1 in the NEC database; (J) IL6 expression correlated with NFKB1 in the NEC database; (K) IL1B expression correlated with NFKBIA in the NEC database; (L) IL6 expression correlated with NFKBIA in the NEC database; (M) IL1B expression correlated with ANXA1 in the NEC database; (N) IL6 expression correlated with ANXA1 in the NEC database. NEC, necrotizing enterocolitis.

Subsequently, correlation analysis between immune cell infiltration and the hub genes—TLR4, NLRP3, NFKB1, NFKB inhibitor alpha (NFKBIA), vimentin (VIM) and annexin A1 (ANXA1)—was performed (Figure 3B). The results showed that all six hub genes were positively correlated with macrophages, neutrophils, NK CD56dim cells, NK cells, pDC, Tem, and Th1 cells.

Furthermore, we assessed the relationship between hub gene expression and pro-inflammatory cytokines interleukin 1 beta (IL1β) and interleukin 6 (IL6) (Figure 3C-3N). The analysis demonstrated that hub gene expression was positively correlated with IL1B and IL6 expression, suggesting that these hub genes may contribute to the upregulation of inflammatory cytokines and the amplification of inflammatory responses in NEC.

PPI network of hub genes to predict genes and pathways

Based on the six hub genes identified, we utilized the STRING database to predict the top 10 closely interacting proteins for each gene and subsequently constructed PPI networks (Figure 4A-4F). Furthermore, we performed GO and KEGG enrichment analyses to investigate the signaling pathways associated with each molecule. The results consistently indicated that the hub genes were closely related to inflammation, cell death, and antimicrobial responses.

Figure 4 PPI network and pathway enrichment of the hub genes. (A) PPI network of TLR4; (B) PPI network of NLRP3; (C) PPI network of VIM; (D) PPI network of NFKB1; (E) PPI network of NFKBIA; (F) PPI network of ANXA1; (G) GO and KEGG pathway enrichment of TLR4; (H) GO and KEGG pathway enrichment of NLRP3; (I) GO and KEGG pathway enrichment of VIM; (J) GO and KEGG pathway enrichment of NFKB1; (K) GO and KEGG pathway enrichment of NFKBIA; (L) GO and KEGG pathway enrichment of ANXA1. GO, Gene Ontology; KEGG, Kyoto Encyclopedia of Genes and Genomes; PPI, protein-protein interaction.

Specifically, TLR4 was enriched in pathways such as positive regulation of autophagy, necroptosis, regulation of apoptotic cell clearance, positive regulation of reactive oxygen species (ROS) biosynthetic process, positive regulation of oxidative stress-induced cell death, toll-like receptor signaling pathway, pattern recognition receptor signaling pathway, interleukin-6 production, and response to IL-1 (GO and KEGG, Figure 4G). NLRP3 was associated with regulation of autophagy, necroptosis, pyroptosis, positive regulation of cysteine-type endopeptidase activity involved in apoptotic process, pathogenic Escherichia coli infection, positive regulation of interleukin-1 beta production, regulation of inflammatory response, and interleukin-6 production (GO and KEGG, Figure 4H). NFKB1 was involved in regulation of apoptotic signaling pathway, the intrinsic apoptotic signaling pathway in response to DNA damage, apoptosis, negative regulation of lipid catabolic process, cellular response to ROS, cellular response to molecules of bacterial origin, cellular response to IL-1, cellular response to IL-6, and regulation of inflammatory response (GO and KEGG, Figure 4I). VIM was enriched in apoptosis, apoptosis-multiple species, intermediate filament-based process, Th1 and Th2 cell differentiation, TNF signaling pathway, and regulation of cell killing (GO and KEGG, Figure 4J). NFKBIA was associated with apoptosis, cellular response to ROS, cellular response to LPS, cellular response to interleukin-1, cellular response to interleukin-6, and cellular response to molecules of bacterial origin (GO and KEGG, Figure 4K). ANXA1 was enriched in neutrophil extracellular trap formation, positive regulation of lipid metabolic process, Staphylococcus aureus infection, Salmonella infection, membrane raft assembly, plasma membrane repair, macrophage activation, immune response-regulating signaling pathway, and myeloid leukocyte activation (GO and KEGG, Figure 4L).

TFs-gene interaction network

The NetworkAnalyst database was used to predict and visualize TFs-gene interactions of the hub genes. The analysis revealed that NLRP3 was associated with nine TFs, including E2F transcription factor 1 (E2F1), Homeobox A5 (HOXA5), Forkhead box L1 (FOXL1), Sp1 transcription factor (SPF), YY1 transcription factor (YY1), Jun proto-oncogene (JUN), RUNX family transcription factor 2 (RUNX2), CCAAT/enhancer binding protein beta (CEBPB), GATA binding protein 2 (GATA2), Fos proto-oncogene (FOS), and Signal transducer and activator of transcription 3 (STAT3). VIM was linked to six TFs, including STAT3, FOS, GATA2, CEBPB, PRDM1, and FOXC1. TLR4 was associated with six TFs, including PR/SET domain 1 (PRDM1), RUNX2, JUN, FOXC1, YY1, and NR3C1. ANXA1 was linked to four TFs, including nuclear receptor subfamily 3 group C member 1 (NR3C1), YY1, SPF, and nuclear factor IC (NFIC). NFKBIA was associated with five TFs, including NFIC, FOXC1, FOXL1, and HOXA5. NFKB1 was linked to two TFs, including E2F1 and YY1.

Notably, YY1 was associated with TLR4, ANXA1, NFKB1, and NLRP3, while FOXC1 was associated with TLR4, NFKBIA, and VIM (Figure 5). However, these findings require further experimental validation.

Figure 5 TFs coregulatory network of the hub genes. Network for TFs-gene interaction with 6 hub genes. TFs, transcription factors.

Expression of the hub genes in vivo and in vitro models

To further validate the expression of the identified hub genes in NEC, we established an NEC mouse model (Figure 6A). Compared with control mice, NEC mice exhibited severe intestinal pneumatosis and necrosis (Figure 6B). H&E staining showed significant villus shedding in the intestines of NEC mice (Figure 6C). Further analysis of intestinal tissues revealed significantly increased expression of Tlr4, Nlrp3, Nfkb1, Nfkbia, Vim and Anxa1 in NEC mice (Figure 6D-6I).

Figure 6 Expression of the hub genes in NEC mice and Caco-2 models. (A) Illustration of NEC mouse model construction; (B) representative images showing the gross appearance of intestines from NEC and control mice; (C) histological analysis of intestinal tissues by H&E staining in NEC and control groups; (D) expression of Anxa1 in intestinal tissues of NEC (n=7) and control mice (n=7) determined by RT-qPCR; (E) expression of Nfkb1 in intestinal tissues of NEC (n=7) and control mice (n=7) determined by RT-qPCR; (F) expression of Nfkbia in intestinal tissues of NEC (n=7) and control mice (n=7) determined by RT-qPCR; (G) expression of Nlrp3 in intestinal tissues of NEC (n=7) and control mice (n=7) determined by RT-qPCR; (H) expression of Tlr4 in intestinal tissues of NEC (n=7) and control mice (n=7) determined by RT-qPCR; (I) expression of Vim in intestinal tissues of NEC (n=7) and control mice (n=7) determined by RT-qPCR; (J) schematic diagram of the Caco-2 cell model; (K) expression of Anxa1 in NEC-mimicking (n=3) and control (n=3) Caco-2 cell determined by RT-qPCR; (L) expression of Nfkb1 in NEC-mimicking (n=3) and control (n=3) Caco-2 cell determined by RT-qPCR; (M) expression of Nfkbia in NEC-mimicking (n=3) and control (n=3) Caco-2 cell determined by RT-qPCR; (N) expression of Nlrp3 in NEC-mimicking (n=3) and control (n=3) Caco-2 cell determined by RT-qPCR; (O) expression of Tlr4 in NEC-mimicking (n=3) and control (n=3) Caco-2 cell determined by RT-qPCR; (P) expression of Vim in NEC-mimicking (n=3) and control (n=3) Caco-2 cell determined by RT-qPCR. **, P<0.01; *, P<0.05; ns, not significant. H&E, hematoxylin and eosin; NEC, necrotizing enterocolitis; RT-qPCR, reverse transcription-quantitative polymerase chain reaction.

To assess whether the molecular changes observed in NEC mice could be reproduced in an epithelial setting, we performed supportive experiments in Caco-2 cells (Figure 6J), which form a polarized intestinal epithelial monolayer and preserve key barrier and innate immune features, including TLR4 signaling (25-27). These in vitro assays were intended to complement the in vivo findings by confirming the reproducibility of the observed alterations and their epithelial relevance. Consistent with the mouse intestinal tissues, which similarly confirmed the upregulation of TLR4, NLRP3, NFKB1, NFKBIA, VIM and ANXA1 (Figure 6K-6P).

Therapeutic efficacy of GSH in NEC

GSH is a class of small-molecule compounds that has been safely applied in clinical treatment. We first measured the levels of GSH in the peripheral blood of NEC patients, as well as in full-term and preterm neonates. The results showed that the peripheral blood GSH level was significantly decreased in NEC patients (Figure 7A).

Figure 7 GSH reduces intestinal necrosis and the expression of Tlr4, Nlrp3 and Nfkb1 in NEC mice. (A) Concentration of GSH in the peripheral plasma of preterm infants (n=11), full-term infants (n=10) and NEC patients (n=12); (B) representative images showing the external appearance of mice in the NEC and NEC + GSH groups; (C) representative images showing the gross appearance of intestines from NEC and NEC + GSH groups; (D) histological analysis of intestinal tissues by H&E staining in NEC and NEC + GSH groups; (E) histopathological severity scores of intestinal tissues in NEC (n=8) and NEC + GSH groups (n=8); (F) terminal deoxynucleotidyl TUNEL staining showing epithelial cell death in the intestines of NEC and NEC + GSH mice; (G) expression of Tlr4 in intestinal tissues of NEC (n=5) and NEC + GSH mice (n=5) determined by RT-qPCR; (H) expression of Nfkb1 in intestinal tissues of NEC (n=5) and NEC+GSH mice (n=5) determined by RT-qPCR; (I) expression of Nlrp3 in intestinal tissues of NEC (n=5) and NEC + GSH mice (n=5) determined by RT-qPCR; (J) expression of Nfkbia in intestinal tissues of NEC (n=5) and NEC+GSH mice (n=5) determined by RT-qPCR; (K) expression of Anxa1 in intestinal tissues of NEC (n=5) and NEC + GSH mice (n=5) determined by RT-qPCR; (L) expression of Vim in intestinal tissues of NEC (n=5) and NEC + GSH mice (n=5) determined by RT-qPCR. **, P<0.01; *, P<0.05; ns, not significant. GSH, glutathione; H&E, hematoxylin and eosin; NEC, necrotizing enterocolitis; RT-qPCR, reverse transcription-quantitative polymerase chain reaction; TUNEL, transferase dUTP nick end labeling.

Next, we treated NEC mice with GSH and observed that GSH ameliorated intestinal injury in NEC mice (Figure 7B-7E). Terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) staining further demonstrated that GSH reduced intestinal cell death in NEC mice (Figure 7F). To explore the molecular mechanism of GSH, we analyzed the expression levels of hub genes in the intestinal tissues of NEC mice following GSH treatment. The results showed significant changes in the expression of Nfkb1, Tlr4, and Nlrp3 (Figure 7G-7L). GSH reduces the mRNA expression of Tlr4, Nlrp3 and Nfkb1 in NEC mice.

Expression of TLR4, NFKB1 and NLRP3 in NEC patients

We next assessed the protein expression of TLR4, NLRP3 and NFKB1 in intestinal tissues of NEC and NEC + GSH mice. As expected, GSH treatment markedly suppressed the protein expression of TLR4, NLRP3 and NFKB1 in mice (Figure 8A-8F). Subsequently, we assessed the protein expression of TLR4, NLRP3 and NFKB1 in the intestinal tissues of NEC patients (Figure 9A-9F). Our findings revealed a significant upregulation of TLR4, NLRP3 and NFKB1 in the intestinal tissues of NEC patients, further implicating their regulatory roles in the pathogenesis of NEC and underscoring the therapeutic promise of GSH.

Figure 8 GSH reduces the expression of TLR4, NLRP3 and NFKB1 in NEC mice. (A) Immunofluorescence staining of TLR4 protein in intestinal tissues from NEC and NEC + GSH mice; (B) quantification of TLR4 fluorescence intensity in NEC (n=5) and NEC + GSH groups (n=5); (C) immunofluorescence staining of NLRP3 protein in intestinal tissues from NEC and NEC + GSH mice; (D) quantification of NLRP3 fluorescence intensity in NEC (n=5) and NEC + GSH groups (n=5); (E) immunofluorescence staining of NFKB1 protein in intestinal tissues from NEC and NEC + GSH mice; (F) quantification of NFKB1 fluorescence intensity in NEC (n=5) and NEC + GSH groups (n=5). **, P<0.01. DAPI, 4',6-diamidino-2-phenylindole; GSH, glutathione; NEC, necrotizing enterocolitis.

Figure 9 The expression of TLR4, NLRP3 and NFKB1 in NEC patients. (A) Immunofluorescence staining of TLR4 protein in intestinal tissues from Control and NEC groups; (B) quantification of TLR4 fluorescence intensity in Control (n=5) and NEC (n=5) groups; (C) immunofluorescence staining of NLRP3 protein in intestinal tissues from Control and NEC groups; (D) quantification of NLRP3 fluorescence intensity in Control (n=5) and NEC (n=5) groups; (E) immunofluorescence staining of NFKB1 protein in intestinal tissues from Control and NEC; (F) quantification of NFKB1 fluorescence intensity in Control (n=5) and NEC (n=5). **, P<0.01. DAPI, 4',6-diamidino-2-phenylindole; NEC, necrotizing enterocolitis.

Analysis of molecular docking

To provide mechanistic insight into the above observations, we performed molecular docking as a hypothesis-generating analysis. Given that GSH treatment alleviated intestine injury of NEC and modulated the expression of hub genes, we next explored the putative binding of GSH to TLR4, NFKB1 and NLRP3 proteins. Based on the regulatory effects of GSH on hub gene mRNA expression, we further investigated the binding affinity of GSH to TLR4, NFKB1, and NLRP3 proteins using molecular docking analysis. We first obtained the chemical structure and 3D conformation of GSH (Figure 10A,10B). In ligand-receptor docking, lower binding energy indicates a more stable interaction between the molecule and the protein. Docking energy lower than −4.0 kcal/mol is considered indicative of a stable binding affinity (19). The results showed that GSH could bind to the target proteins TLR4 and NFKB1 through multiple amino acid residues (Table 4), with binding energies all below −4.0 kcal/mol, indicating stable interactions. Furthermore, we visualized the three-dimensional and two-dimensional interactions of GSH with TLR4 and NFKB1 proteins (Figure 10C-10F).

Figure 10 Molecular docking results of GSH with key target proteins. (A) Chemical structure of GSH; (B) three-dimensional structure of GSH; (C) 3D docking model of GSH with TLR4; (D) two-dimensional interaction diagram of GSH docked with TLR4; (E) 3D docking model of GSH with NFKB1; (F) two-dimensional interaction diagram of GSH docked with NFKB1. 3D, three dimensional; GSH, glutathione.

Discussion

NEC is a leading cause of neonatal mortality, yet effective therapies with an acceptable safety profile remain limited. NEC pathogenesis is complex and involves multiple forms of programmed cell death, raising the question of whether these pathways converge on actionable targets. In our cohort, plasma GSH levels were significantly reduced in patients with NEC. Given GSH’s central role in redox homeostasis and its favorable clinical safety profile, we evaluated its therapeutic effects in NEC and specifically investigated whether GSH modulates multiple programmed cell-death pathways during NEC progression. Mechanistically, we integrated network pharmacology, molecular docking, and bioinformatics with in vivo/in vitro experiments and analyses of clinical blood and intestinal tissue samples. In this study, we initially identified 11 DEGs and subsequently narrowed them down to 6 hub genes. Using the NEC mouse model and Caco-2 cell experiments, we validated the significant upregulation of TLR4, NFKB1 and NLRP3. Treatment with GSH ameliorated intestinal necrosis in NEC mice and significantly reduced the expression levels of TLR4, NFKB1 and NLRP3 (Figure 11). Furthermore, network pharmacology and molecular docking analyses demonstrated the binding interactions between GSH and the TLR4, NFKB1.

Figure 11 Key regulatory mechanisms of programmed cell death in NEC and the protective role of GSH. GSH, glutathione; NEC, necrotizing enterocolitis.

TLR4 is an evolutionarily conserved innate immune receptor that recognizes pathogen-associated molecular patterns (PAMPs) and damage-associated molecular patterns (DAMPs). TLR4 is a type I transmembrane receptor comprising a ligand-recognition leucine-rich repeat (LRR) domain and a Toll/interleukin-1 receptor (TIR) domain responsible for signal transduction (28). Intestinal microbiota, LPS, and inflammatory responses can activate the expression of TLR4. The enriched signaling pathways identified in our study further supported this finding. Previous studies have shown that TLR4 expression is upregulated in preterm infants, and activation of TLR4 can promote intestinal epithelial cell injury, apoptosis, and the progression of NEC (29). Kandadi et al. reported that knockdown of TLR4 significantly reduced autophagy levels and subsequently improved cardiac function (30). Hypoxic injury can upregulate TLR4 expression, downregulate SLC7A11 and GPX4 levels, and induce mitochondrial damage. Inhibition of TLR4 has been shown to enhance cell viability and reduce ferroptosis (31). Li et al. reported that overexpression of TLR4 increased the levels of pyroptosis-related proteins IL-1β and IL-18, thereby reducing cell viability (32). Notably, studies have shown that TLR4 and NFKB mRNA expression levels are elevated in NEC, and that LPS-TLR4 can activate NF-κB signaling (33,34).

The NLRP3 protein belongs to the family of nucleotide-binding oligomerization domain-like receptors (NLRs). DAMPs or PAMPs are recognized by Toll-like receptors and activate the NF-κB signaling pathway, which subsequently activates NLRP3. In addition, LPS and mitochondrial dysfunction can also activate NLRP3 (35). Activated NLRP3 inflammasomes promote the maturation of caspase-1, leading to the conversion of pro-IL-1β and pro-IL-18 into their active cytokine forms (36). NLRP3 is a key molecule involved in inducing various forms of cell death (37). The necroptotic factor receptor-interacting serine/threonine-protein kinase 3 (RIPK3) forms a complex with mixed lineage kinase domain-like protein (MLKL), inducing mitochondrial mROS production, which in turn triggers NLRP3 inflammasome activation to protect against Streptococcus pneumoniae infection (38). Notably, studies have shown that sodium butyrate alleviates experimental A disease by inhibiting TLR4-mediated, NLRP3 inflammasome-dependent pyroptosis (39). These findings highlight the potential of NLRP3 as a therapeutic target in NEC.

NFKB1 is a key cellular TF that regulates cell growth, differentiation, and apoptosis (40-43). Gao et al. found that NF-κB1 binds to the promoter of the necroptosis-related molecule Ripk3, thereby blocking TNFα-induced Ripk3 transcription and inhibiting primary endothelial cell death in cultured human primary endothelial cells (44). AbdAllah et al. reported that NFKB1 expression is elevated in neonatal infectious diseases and is closely associated with poor prognosis (45). In addition, the NFKB1 protein, as a TF, plays a critical role in cellular immune responses and inflammatory responses (40,46). Moreover, numerous studies have shown that inhibition of the TLR4/NF-κB/NLRP3 signaling axis can alleviate inflammation and apoptosis (47-49). In our study, we found that the expression levels of TLR4, NFKB1, and NLRP3 were all upregulated in NEC. Furthermore, fluorescence co-expression analysis showed that the expression of the TF protein NFKB1 was positively correlated with the inflammatory protein NLRP3. Notably, Sampath et al. (50) previously identified the NFKB1 (g.-24519delATTG) variant in all infants with NEC, whereas it was present in only 65% of infants without NEC disease (P=0.003). After correction for multiple comparisons, the NFKB1 variant remained significantly (P<0.05). These findings highlight the potential of NFKB1 as a therapeutic target in NEC.

The TFs-gene interactions were identified based on the six hub DEGs. In the TFs-gene interaction network, we found that the TFs YY1 and FOXC1 were associated with multiple hub genes. Kumar et al. (51) reported that YY1 is essential for mitochondrial gene expression. Loss of YY1 in the late gestational endoderm resulted in impaired intestinal differentiation and severely underdeveloped villi. Interestingly, the study also found that the loss of YY1 resulted in pathway enrichment patterns similar to those observed for downregulated genes in NEC, both exhibiting abnormalities in oxidative phosphorylation. YY1 can inhibit infection-induced ferroptosis and reduce intracellular bacterial proliferation in pulmonary epithelial cells (52). Intestinal ischemia is one of the major causes of intestinal necrosis in NEC. Tan et al. reported that FOXC1 and FOXC2 are essential for intestinal regeneration by promoting paracrine CXCL12 and Wnt signaling (53). Another study showed that enhancement of SIRT6/FoxC reduces colitis-associated cell death by inhibiting the NLRP3/cleaved caspase-1/Gasdermin D pyroptosis signaling pathway (54). These studies highlight the potential research value of the newly identified TFs YY1 and FOXC1 in NEC.

GSH is a tripeptide composed of glutamate, cysteine and glycine, and serves as the primary intracellular antioxidant. It is shown that GSH levels are closely associated with the development of neonatal gastrointestinal diseases (55). Moreover, GSH has been safely applied in clinical treatment for conditions such as liver injury and infections (56-58). Our preliminary experiments further demonstrated that GSH levels in the peripheral blood of NEC patients were significantly lower compared with those in full-term neonates. Sun et al. reported that depletion of GSH can promote ferroptosis and autophagy in epithelial cells (59). The other studies showed that GSH can protect cells from death through its antioxidant effects (60,61). In our study, treatment with GSH ameliorated intestinal necrosis in NEC mice, an effect closely associated with the expression of TLR4, NFKB1, and NLRP3 genes in the cell death pathway. Molecular docking analysis further suggested that GSH can stably bind to the TLR4 and NFKB1 proteins. Specifically, GSH was predicted to form stable hydrogen bonds with TLR4 at amino acid residues Gly343 (B), Lys388 (A), Asn365 (B), Ala366 (B), and Phe345 (B), and with NFKB1 at residues Thr145 (A), Ser112 (A), Glu151 (A), and Arg156 (A). These findings suggest that GSH may be developed as an inhibitor of TLR4 and NFKB1. However, further experiments are needed to validate the binding of GSH to TLR4 and NFKB1.

There are some limitations in our study. First, part of the data used in the analysis was obtained from publicly available databases, and the quality of these data may influence the reliability and accuracy of the predictions. Second, although network pharmacology and molecular modeling provide valuable insights, these computational approaches may not fully capture the precise biological roles or underlying mechanisms. In addition, further studies are needed to elucidate the interactions between GSH and target proteins, as well as the associated molecular mechanisms, using follow-up assays such as western blotting. Nevertheless, we validate our findings across multiple dimensions, including animal experiments, cellular models and clinical samples, which strengthens the robustness of the conclusions.

Conclusions

This study identified and validated DEGs closely associated with various forms of programmed cell death in NEC. In addition, treatment with GSH alleviated intestinal necrosis in NEC. Through network pharmacology, molecular docking, and both in vivo and in vitro experiments, we identified TLR4 and NFKB1 as potential target genes of GSH, thereby enhancing the therapeutic potential of GSH in NEC. Further studies with larger sample sizes are needed to validate these findings.

Acknowledgments

The authors would like to thank the support from the grant of National Natural Science Foundation, Science and Technology Project of Guangzhou, Guangzhou Major Difficult and Rare Diseases Project, Guangzhou Science and Technology Youth Talent Cultivation Project. The authors would like to thank the support from Guangdong Provincial Key Laboratory of Research in Structural Birth Defect Disease. We would like to thank Prof. Mahmoud AL-Azab, whose research focuses on inflammation and immunology, for his assistance in the content review and English language polishing of the manuscript.

An abstract based on this manuscript was accepted after peer review for presentation at the 38th International Symposium on Pediatric Surgical Research (ISPSR 2025), to be held in Guangzhou, China, in October 2025.

Footnote

Reporting Checklist: The authors have completed the ARRIVE and MDAR reporting checklists. Available at https://tp.amegroups.com/article/view/10.21037/tp-2025-1-860/rc

Data Sharing Statement: Available at https://tp.amegroups.com/article/view/10.21037/tp-2025-1-860/dss

Peer Review File: Available at https://tp.amegroups.com/article/view/10.21037/tp-2025-1-860/prf

Funding: This work was supported by the National Natural Science Foundation of China (No. 82370526, to W.Z.), the Science and Technology Project of Guangzhou (No. 2024A03J1171 to W.Z., and No. 20261A031036 to P.Z.), and Guangzhou Major Difficult and Rare Diseases Project (No. 2024MDRD17 to W.Z.).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tp.amegroups.com/article/view/10.21037/tp-2025-1-860/coif). H.X. serves as an unpaid editorial board member of Translational Pediatrics from July 2025 to June 2026. The other authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Human Review Board of Guangzhou Women and Children’s Medical Center (Human Investigation Committee No.2021-174A01). Informed consent was obtained from the parents of the infant. Experiments were performed under a project license (No. RSDW-2023-01296) granted by the Institutional Animal Care and Use Committee of Guangzhou Medical University, in compliance with national guidelines for the care and use of animals. A protocol was prepared before the study without registration.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Hackam DJ, Sodhi CP. Bench to bedside - new insights into the pathogenesis of necrotizing enterocolitis. Nat Rev Gastroenterol Hepatol 2022;19:468-79. [Crossref] [PubMed]
2. Battersby C, Modi N. Challenges in Advancing Necrotizing Enterocolitis Research. Clin Perinatol 2019;46:19-27. [Crossref] [PubMed]
3. Lim JC, Golden JM, Ford HR. Pathogenesis of neonatal necrotizing enterocolitis. Pediatr Surg Int 2015;31:509-18. [Crossref] [PubMed]
4. Carr BD, Gadepalli SK. Does Surgical Management Alter Outcome in Necrotizing Enterocolitis? Clin Perinatol 2019;46:89-100. [Crossref] [PubMed]
5. Xiong X, Wu L, Liu X, et al. Comprehensive Characterization of the Oxidative Stress Profiles in Neonatal Necrotizing Enterocolitis. Int J Med Sci 2025;22:2139-54. [Crossref] [PubMed]
6. Hao Z, Han J, Yao T, et al. ALAS2 Prevents Neonatal Necrotizing Enterocolitis by Improving Ferroptosis in Intestinal Epithelial Cells Through Inhibition of Oxidative Stress. Mediators Inflamm 2026;2026:6683001. [Crossref] [PubMed]
7. Lan C, Tian B, Liu Y, et al. m6A-Dependent Upregulation of PFKFB3 Drives Macrophage-Mediated Inflammation in Necrotizing Enterocolitis. Antioxid Redox Signal 2025;43:765-81. [Crossref] [PubMed]
8. Rushworth GF, Megson IL. Existing and potential therapeutic uses for N-acetylcysteine: the need for conversion to intracellular glutathione for antioxidant benefits. Pharmacol Ther 2014;141:150-9. [Crossref] [PubMed]
9. Wu G, Fang YZ, Yang S, et al. Glutathione metabolism and its implications for health. J Nutr 2004;134:489-92. [Crossref] [PubMed]
10. Gershner GH, Calkins C, Golubkova A, et al. The Role of GPX Enzymes, Lipid Profiles, and Iron Accumulation in Necrotizing Enterocolitis. Int J Mol Sci 2025;26:6077. [Crossref] [PubMed]
11. Zhu Q, Wang Y, Zhou Y, et al. L-cystine alleviates necrotizing enterocolitis by regulating ferroptosis and Th17 cell differentiation via the IL-6/STAT3 pathway. Commun Biol 2025;9:162. [Crossref] [PubMed]
12. Lee J, Joseph S, Manohar K, et al. Ferroptosis as a novel pathway in the pathogenesis of necrotizing enterocolitis. Shock 2025;64:12-8. [Crossref] [PubMed]
13. Golubkova A, Leiva T, Snyder K, et al. Response of the Glutathione (GSH) Antioxidant Defense System to Oxidative Injury in Necrotizing Enterocolitis. Antioxidants (Basel) 2023;12:1385. [Crossref] [PubMed]
14. Lana JV, Rios A, Takeyama R, et al. Nebulized Glutathione as a Key Antioxidant for the Treatment of Oxidative Stress in Neurodegenerative Conditions. Nutrients 2024;16:2476. [Crossref] [PubMed]
15. Lin H, Wang L, Jiang X, et al. Glutathione dynamics in subcellular compartments and implications for drug development. Curr Opin Chem Biol 2024;81:102505. [Crossref] [PubMed]
16. Yang S, Wei X, Zhong Y, et al. Programmed death of intestinal epithelial cells in neonatal necrotizing enterocolitis: a mini-review. Front Pediatr 2023;11:1199878. [Crossref] [PubMed]
17. Bindea G, Mlecnik B, Tosolini M, et al. Spatiotemporal dynamics of intratumoral immune cells reveal the immune landscape in human cancer. Immunity 2013;39:782-95. [Crossref] [PubMed]
18. Zhou G, Soufan O, Ewald J, et al. NetworkAnalyst 3.0: a visual analytics platform for comprehensive gene expression profiling and meta-analysis. Nucleic Acids Res 2019;47:W234-41. [Crossref] [PubMed]
19. Wang F, Yuan C, Liu B, et al. Syringin exerts anti-breast cancer effects through PI3K-AKT and EGFR-RAS-RAF pathways. J Transl Med 2022;20:310. [Crossref] [PubMed]
20. Liu Y, Perego M, Xiao Q, et al. Lactoferrin-induced myeloid-derived suppressor cell therapy attenuates pathologic inflammatory conditions in newborn mice. J Clin Invest 2019;129:4261-75. [Crossref] [PubMed]
21. Zhou D, Yao M, Zhang L, et al. Adenosine Alleviates Necrotizing Enterocolitis by Enhancing the Immunosuppressive Function of Myeloid-Derived Suppressor Cells in Newborns. J Immunol 2022;209:401-11. [Crossref] [PubMed]
22. Zhang Y, Jing Y, He J, et al. Bile acid receptor FXR promotes intestinal epithelial ferroptosis and subsequent ILC3 dysfunction in neonatal necrotizing enterocolitis. Immunity 2025;58:683-700.e10. [Crossref] [PubMed]
23. Rager TM, Olson JK, Zhou Y, et al. Exosomes secreted from bone marrow-derived mesenchymal stem cells protect the intestines from experimental necrotizing enterocolitis. J Pediatr Surg 2016;51:942-7. [Crossref] [PubMed]
24. Cho SX, Rudloff I, Lao JC, et al. Characterization of the pathoimmunology of necrotizing enterocolitis reveals novel therapeutic opportunities. Nat Commun 2020;11:5794. [Crossref] [PubMed]
25. Xu L, Ma S, Qu M, et al. Parthanatos initiated by ROS-induced DNA damage is involved in intestinal epithelial injury during necrotizing enterocolitis. Cell Death Discov 2024;10:345. [Crossref] [PubMed]
26. Wei J, Dang D, Li Z, et al. Piezo1-mediated endoplasmic reticulum stress-dependent apoptosis exacerbates intestinal epithelial barrier disruption in necrotizing enterocolitis. Int Immunopharmacol 2025;167:115721. [Crossref] [PubMed]
27. Chu C, Wang X, Chen F, et al. Neutrophil extracellular traps aggravate intestinal epithelial necroptosis in ischaemia-reperfusion by regulating TLR4/RIPK3/FUNDC1-required mitophagy. Cell Prolif 2024;57:e13538. [Crossref] [PubMed]
28. Lin C, Wang H, Zhang M, et al. TLR4 biased small molecule modulators. Pharmacol Ther 2021;228:107918. [Crossref] [PubMed]
29. Blum L, Vincent D, Boettcher M, et al. Immunological aspects of necrotizing enterocolitis models: a review. Front Immunol 2024;15:1434281. [Crossref] [PubMed]
30. Kandadi MR, Frankel AE, Ren J. Toll-like receptor 4 knockout protects against anthrax lethal toxin-induced cardiac contractile dysfunction: role of autophagy. Br J Pharmacol 2012;167:612-26. [Crossref] [PubMed]
31. Zhu K, Zhu X, Sun S, et al. Inhibition of TLR4 prevents hippocampal hypoxic-ischemic injury by regulating ferroptosis in neonatal rats. Exp Neurol 2021;345:113828. [Crossref] [PubMed]
32. Li YX, Lv L, Li SL, et al. Gentianine alleviates dextran sulfate sodium-induced ulcerative colitis via inhibition of TLR4/NLRP3-mediated pyroptosis. Int Immunopharmacol 2024;126:111214. [Crossref] [PubMed]
33. Hunter CJ, De Plaen IG. Inflammatory signaling in NEC: Role of NF-kappaB, cytokines and other inflammatory mediators. Pathophysiology 2014;21:55-65. [Crossref] [PubMed]
34. Yin Y, Liu F, Li Y, et al. mRNA expression of TLR4, TLR9 and NF-kappaB in a neonatal murine model of necrotizing enterocolitis. Mol Med Rep 2016;14:1953-6. [Crossref] [PubMed]
35. Chen Y, Ye X, Escames G, et al. The NLRP3 inflammasome: contributions to inflammation-related diseases. Cell Mol Biol Lett 2023;28:51. [Crossref] [PubMed]
36. Brydges SD, Broderick L, McGeough MD, et al. Divergence of IL-1, IL-18, and cell death in NLRP3 inflammasomopathies. J Clin Invest 2013;123:4695-705. [Crossref] [PubMed]
37. Huang Y, Xu W, Zhou R. NLRP3 inflammasome activation and cell death. Cell Mol Immunol 2021;18:2114-27. [Crossref] [PubMed]
38. Chen KW, Demarco B, Heilig R, et al. Extrinsic and intrinsic apoptosis activate pannexin-1 to drive NLRP3 inflammasome assembly. EMBO J 2019;38:e101638. [Crossref] [PubMed]
39. Gao Y, Yang L, Wu H, et al. Sodium butyrate attenuates experimental neonatal necrotizing enterocolitis by suppressing TLR4-mediated NLRP3 inflammasome-dependent pyroptosis. Food Funct 2025;16:3508-24. [Crossref] [PubMed]
40. Cartwright T, Perkins ND. L Wilson C. NFKB1: a suppressor of inflammation, ageing and cancer. FEBS J 2016;283:1812-22. [Crossref] [PubMed]
41. Li L, Zhang ZT. Genetic Association between NFKBIA and NFKB1 Gene Polymorphisms and the Susceptibility to Head and Neck Cancer: A Meta-Analysis. Dis Markers 2019;2019:6523837. [Crossref] [PubMed]
42. Deng Q, Ma L, Chen T, et al. NF-kappaB 1-induced LINC00665 regulates inflammation and apoptosis of neurons caused by spinal cord injury by targeting miR-34a-5p. Neurol Res 2021;43:418-27. [Crossref] [PubMed]
43. Zhong L, Jia J, Ye G. Rian/miR-210-3p/Nfkb1 Feedback Loop Promotes Hypoxia-Induced Cell Apoptosis in Myocardial Infarction Through Deactivating the PI3K/Akt Signaling Pathway. J Cardiovasc Pharmacol 2020;76:207-15. [Crossref] [PubMed]
44. Gao S, Menendez M, Kurylowicz K, et al. Genomic locus proteomic screening identifies the NF-kappaB signaling pathway components NFkappaB1 and IKBKG as transcriptional regulators of Ripk3 in endothelial cells. PLoS One 2021;16:e0253519. [Crossref] [PubMed]
45. AbdAllah NB. MYD88, NFKB1, and IL6 transcripts overexpression are associated with poor outcomes and short survival in neonatal sepsis. Sci Rep 2021;11:13374. [Crossref] [PubMed]
46. Lemche E. Early Life Stress and Epigenetics in Late-onset Alzheimer’s Dementia: A Systematic Review. Curr Genomics 2018;19:522-602. [Crossref] [PubMed]
47. Zhang B, Chen L, Kang M, et al. Gegen Qinlian Decoction improves Alzheimer’s disease through TLR4/NF-kappaB/NLRP3 pathway. Tissue Cell 2025;95:102818. [Crossref] [PubMed]
48. Jiang K, Zhang F, Chen Y, et al. Fosfenopril Attenuates Inflammatory Response in Diabetic Dry Eye Models by Inhibiting the TLR4/NF-kappaB/NLRP3 Signaling Pathway. Invest Ophthalmol Vis Sci 2024;65:2. [Crossref] [PubMed]
49. Luo M, Yan D, Sun Q, et al. Ginsenoside Rg1 attenuates cardiomyocyte apoptosis and inflammation via the TLR4/NF-kB/NLRP3 pathway. J Cell Biochem 2020;121:2994-3004. [Crossref] [PubMed]
50. Sampath V, Le M, Lane L, et al. The NFKB1 (g.-24519delATTG) variant is associated with necrotizing enterocolitis (NEC) in premature infants. J Surg Res 2011;169:e51-7. [Crossref] [PubMed]
51. Kumar N, Srivillibhuthur M, Joshi S, et al. A YY1-dependent increase in aerobic metabolism is indispensable for intestinal organogenesis. Development 2016;143:3711-22. [Crossref] [PubMed]
52. Fei Q, Zhang J, Chen L, et al. Modulating ferroptosis and mycobactericidal activity in lung epithelial cells via YY1/iNOS pathway. Life Sci 2024;358:123131. [Crossref] [PubMed]
53. Tan C, Norden PR, Yu W, et al. Endothelial FOXC1 and FOXC2 promote intestinal regeneration after ischemia-reperfusion injury. EMBO Rep 2023;24:e56030. [Crossref] [PubMed]
54. Oraby MA, Abdel Mageed SS, Amr Raouf A, et al. Remdesivir ameliorates ulcerative colitis-propelled cell inflammation and pyroptosis in acetic acid rats by restoring SIRT6/FoxC1 pathway. Int Immunopharmacol 2024;137:112465. [Crossref] [PubMed]
55. Zhu JJ, Yang YF, Dong R, et al. Biliatresone: progress in biliary atresia study. World J Pediatr 2023;19:417-24. [Crossref] [PubMed]
56. Santacroce G, Gentile A, Soriano S, et al. Glutathione: Pharmacological aspects and implications for clinical use in non-alcoholic fatty liver disease. Front Med (Lausanne) 2023;10:1124275. [Crossref] [PubMed]
57. Hsiao YF, Huang SC, Cheng SB, et al. Glutathione and Selenium Supplementation Attenuates Liver Injury in Diethylnitrosamine-Induced Hepatocarcinogenic Mice by Enhancing Glutathione-Related Antioxidant Capacities. Int J Mol Sci 2024;25:11339. [Crossref] [PubMed]
58. Ma Z, Wang H, Shi Z, et al. Inhalable GSH-Triggered Nanoparticles to Treat Commensal Bacterial Infection in In Situ Lung Tumors. ACS Nano 2023;17:5740-56. [Crossref] [PubMed]
59. Sun Y, Zheng Y, Wang C, et al. Glutathione depletion induces ferroptosis, autophagy, and premature cell senescence in retinal pigment epithelial cells. Cell Death Dis 2018;9:753. [Crossref] [PubMed]
60. Brown IAM, Gulbransen BD. The antioxidant glutathione protects against enteric neuron death in situ, but its depletion is protective during colitis. Am J Physiol Gastrointest Liver Physiol 2018;314:G39-52. [Crossref] [PubMed]
61. Moine L, Díaz de Barboza G, Pérez A, et al. Glutamine protects intestinal calcium absorption against oxidative stress and apoptosis. Comp Biochem Physiol A Mol Integr Physiol 2017;212:64-71. [Crossref] [PubMed]