This study identifies five cuproptosis- and ferroptosis-related biomarkers in ulcerative colitis through integrated machine learning, single-cell RNA sequencing, and experimental validation, providing insights into clinical diagnosis and targeted immunotherapy.
Research Article
Targeting Ferroptosis and Cuproptosis in Ulcerative Colitis: Findings from Comprehensive Bioinformatics and Experimental Validation
In This Article
Summary
Abstract
Ulcerative colitis (UC) is a persistent intestinal disorder with an incompletely defined pathogenesis and increasing prevalence and hospitalization rates in newly industrialized countries. In UC, excessive intestinal epithelial cell death disrupts the mucosal barrier and triggers inflammatory responses. Ferroptosis and cuproptosis are two recently described forms of regulated cell death. Most studies of UC have examined these processes separately; however, their combined roles in UC progression remain insufficiently characterized. To investigate their synergistic roles in UC pathogenesis, we used a stepwise approach involving large-scale transcriptomic profiling to identify candidate targets, followed by validation in in vitro models. Differentially expressed genes (DEGs) were intersected with ferroptosis-related genes (FRGs) and cuproptosis-related genes (CRGs). The overlapping targets were prioritized using a consensus of machine learning algorithms and weighted gene co-expression network analysis (WGCNA). Five biomarkers—LCN2, IDO1, CXCL2, NOS2, and CD274—were significantly upregulated in UC. Single-cell analysis characterized their expression across different cell types, with LCN2 and NOS2 primarily enriched in epithelial cells. The mechanistic relevance of these markers was further evaluated through in vitro assays. Treatment with ferroptosis or cuproptosis inhibitors alleviated UC-associated inflammation and modulated biomarker expression in Caco-2 cell models. These findings identify five biomarkers associated with UC progression and provide experimental evidence supporting their potential application in clinical diagnosis and therapeutic intervention.
Introduction
Ulcerative colitis (UC) is a relapsing inflammatory bowel disease associated with a growing global burden and substantial long-term treatment costs1,2. Although aminosalicylates, corticosteroids, biologics, and small-molecule therapies have expanded treatment options, durable remission remains difficult to achieve for many patients3,4. This unmet clinical need highlights the importance of identifying reproducible molecular markers and tractable mechanisms that may support earlier diagnosis and mechanism-guided therapy5.
Ferroptosis is an iron-dependent form of regulated cell death driven by lipid peroxidation, glutathione depletion, and impaired GPX4 activity6. Increased iron and malondialdehyde levels in UC are consistent with ferroptotic epithelial injury, while experimental activation of the SLC7A11–GSH–GPX4 axis or direct inhibition of ferroptosis can protect the intestinal barrier7,8.
Cuproptosis, described in 2022, is triggered when copper binds to lipoylated mitochondrial proteins, resulting in protein aggregation and proteotoxic stress9. Because the intestine plays a central role in copper handling, disruption of this pathway may amplify oxidative and inflammatory epithelial injury in UC10.
Ferroptosis and cuproptosis may converge through metal-ion imbalance, mitochondrial metabolism, glutathione depletion, and oxidative stress11. However, whether this crosstalk produces a reproducible molecular signature in UC and whether such a signature responds to pathway-specific inhibition remain unresolved.
The study hypothesis was that dysregulated ferroptosis–cuproptosis crosstalk in UC converges on a reproducible biomarker signature that can be prioritized across bulk and single-cell transcriptomic datasets and attenuated by pathway-specific inhibition in an intestinal epithelial injury model. Accordingly, multicohort bioinformatics, machine learning, single-cell analysis, and targeted Caco-2 experiments were integrated to identify and functionally evaluate candidate biomarkers (Figure 1).
Figure 1: Study workflow. Integrated workflow for multicohort transcriptomic screening, machine-learning and network-based biomarker prioritization, single-cell localization, and inhibitor-based validation in Caco-2 cells. UC: ulcerative colitis; GEO: Gene Expression Omnibus; WGCNA: weighted gene co-expression network analysis; PPI: protein–protein interaction; ROC: receiver operating characteristic; GSEA: gene set enrichment analysis. Please click here to view a larger version of this figure.
Protocol
This study used publicly available datasets and commercially available human cell lines. No human participants, newly collected human tissue, or vertebrate animals were involved; therefore, institutional ethics committee approval was not required. Research tools, software, and online resources used in the protocol are listed in the Table of Materials.
1. Data processing for differential expression analysis
To establish the transcriptomic discovery and validation cohorts, the Gene Expression Omnibus (GEO) database was searched using the term “ulcerative colitis.” Four datasets—GSE87466, GSE92415, GSE107499, and GSE75214 were used as the training set, whereas GSE47908 was reserved for external validation. The training datasets were normalized and batch-corrected using ComBat in Sangerbox 3.0, whereas the validation dataset was processed independently.
A total of 583 ferroptosis-related genes (FRGs) were obtained from FerrDB and supplemented with genes reported in the literature to reduce single-source bias12. A set of 96 cuproptosis-related genes (CRGs) was compiled from published studies12,13,14. Differential expression was analyzed using limma with thresholds of |log₂ fold change| > 1 and Benjamini–Hochberg-adjusted P < 0.05. Diagnostic plots of normalization, batch correction, differential expression, and clustering are provided in Supplementary Figure 1. Sample composition, platforms, and cohort assignments are provided in Supplementary Table 1.
2. Identification of CF-DEGs in UC
Pearson correlation analysis was performed using thresholds of |r| > 0.5 and P < 0.05 to identify co-expressed FRGs and CRGs. The resulting gene set was intersected with the differentially expressed genes (DEGs) to identify cuproptosis–ferroptosis co-expressed differentially expressed genes (CF-DEGs).
3. Enrichment analysis
Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses were performed using clusterProfiler and the Bioinformatics.com.cn platform. A Benjamini–Hochberg-adjusted P < 0.05 was used as the significance threshold. Enrichment plots are provided in Supplementary Figure 2.
4. Weighted gene co-expression network analysis (WGCNA)
The modular structure underlying the gene expression profiles was analyzed using weighted gene co-expression network analysis (WGCNA). Abnormal samples were identified and excluded through hierarchical clustering combined with the goodSamplesGenes function. A suitable soft-thresholding power (β) was subsequently determined to satisfy the scale-free topology criterion for network construction.
The processed expression data were used to construct a topological overlap matrix (TOM) for gene co-expression analysis. Using the dynamic tree-cutting approach, genes with highly consistent expression patterns were organized into gene modules, each containing at least 100 genes. Associations between module eigengenes (MEs) and phenotypes were analyzed to identify modules associated with the disease process and extract the corresponding gene sets. WGCNA screening diagnostics are provided in Supplementary Figure 3.
5. Selection of biomarkers
To prioritize candidates for experimental follow-up, three machine-learning algorithms were applied in parallel15. Least absolute shrinkage and selection operator (LASSO) regression was used for feature selection, random forest with ntree = 500 was used to rank gene importance, and support vector machine–recursive feature elimination (SVM-RFE) was used to identify the feature subset with the lowest cross-validation error. Full screening diagnostics are provided in Supplementary Figure 5.
Gene importance was evaluated according to the contribution of individual genes to classification accuracy, and highly ranked genes were collected to generate a preliminary gene set16. The SVM-RFE algorithm performed feature screening and model optimization through iterative feature elimination, allowing identification of the optimal feature subset for classification17. Genes shared among the three algorithms were considered candidate biomarkers identified through machine learning.
The CF-DEGs were used to construct a protein–protein interaction (PPI) network, and CytoHubba was used to rank core nodes. The intersection of key WGCNA modules, the three machine-learning outputs, and the PPI core set defined five integrated candidate biomarkers: LCN2, IDO1, CXCL2, NOS2, and CD274. These genes were classified as UC-associated diagnostic candidates. Subsequent inhibitor experiments further classified LCN2, IDO1, CXCL2, and NOS2 as ferroptosis-inhibition-responsive markers and CD274 as a cuproptosis-inhibition-responsive marker. These genes were not considered proven upstream regulators of either pathway.
6. Predictive model development and multi-method evaluation
To evaluate the clinical translational potential of the identified gene signature, a nomogram was constructed using the rms package to assess diagnostic performance. Model performance was evaluated using a calibration curve, the Hosmer–Lemeshow (HL) test, and Harrell’s concordance index (C-index).
Model calibration was assessed using the HL test and mean absolute error (MAE). A P value > 0.05 and MAE < 0.1 were established as thresholds for acceptable model fit and accuracy, respectively. Receiver operating characteristic (ROC) curves were generated, and the area under the curve (AUC) was calculated to evaluate the nomogram's predictive performance for UC. An independent dataset was used to assess model generalizability and stability.
7. Gene set enrichment analysis (GSEA)
In the training set, Spearman correlation analysis was performed between each biomarker and all other genes. The correlation results were ranked to generate an ordered gene set. Gene set enrichment analysis (GSEA) was conducted using the c2.cp.kegg_medicus.v2025.1.Hs.symbols gene set from the Molecular Signatures Database (MSigDB) as the reference. Biological pathways with P < 0.05 were considered significantly enriched.
8. Immune landscape analysis
The CIBERSORT tool and LM22 signature matrix were used to perform deconvolution analysis of the expression data and estimate the proportions of 22 immune cell subtypes. Spearman correlation analysis was subsequently performed to assess associations between the five biomarkers and immune cell subtypes.
9. Regulatory network analysis
Information on gene-related transcription factor (TF) and microRNA (miRNA) regulation was obtained from the integrated ChEA3 and TarBase 9.0 databases through the NetworkAnalyst platform. MiRNAs associated with at least two nodes were included in the regulatory network.
10. Single-cell RNA-seq
The UC single-cell datasets GSE116222, comprising three healthy and three UC samples, and GSE214695, comprising six healthy and six UC samples, were integrated. Cells were retained when they contained 200–5,000 detected genes, >1,000 unique molecular identifiers, < 30% mitochondrial transcripts, < 3% hemoglobin transcripts, and < 50% ribosomal transcripts. The NormalizeData function was used for normalization, and Harmony was used for batch correction before clustering and annotation.
After filtering, the data were normalized using the NormalizeData function. Harmony was used to minimize batch effects among samples. Uniform manifold approximation and projection (UMAP) and t-distributed stochastic neighbor embedding (t-SNE) were used for dimensionality reduction and cell clustering to characterize local and global structures within the cell population. A total of 31,712 high-quality cells were retained for further analysis. Cell types were annotated according to canonical cell marker genes.
Pseudotime analysis was performed using Monocle 2 to investigate changes in the states of epithelial and myeloid cells from UC tissues. Cell developmental trajectories and between-group differences were evaluated, and biomarker expression patterns along the pseudotime trajectory were visualized.
CellChat was used to evaluate cell–cell communication between epithelial and myeloid cell subsets. Subpopulation-specific expression patterns were integrated to infer interaction probabilities and key signaling pathways between epithelial and myeloid cells and to characterize their potential intercellular regulatory networks.
11. Cell experiments
Human Caco-2 colon adenocarcinoma cells were cultured in Dulbecco’s modified Eagle medium containing 10% fetal bovine serum, 100 U/mL penicillin, and 100 µg/mL streptomycin at 37 °C in 5% CO₂. Caco-2 cells were selected as a reproducible epithelial monolayer model for intestinal barrier and inflammatory-injury assays; however, their malignant origin limits direct generalization to nonmalignant colonic epithelium. Passage information was unavailable in the archived experimental records.
An in vitro inflammatory-injury model was established by exposing cells to 1 µg/mL lipopolysaccharide (LPS) for 24 h. Cells were assigned to a dimethyl sulfoxide vehicle-control group, an LPS-only group, an LPS + 4 µM ferrostatin-1 group, or an LPS + 40 µM tetrathiomolybdate group, with matched final solvent concentrations. The LPS-only group served as the positive injury/model control for the inhibitor-rescue comparisons.
- Cell viability
Cell viability was assessed using a cell viability assay according to the manufacturer’s instructions. After incubation at 37 °C for approximately 2 h, absorbance at 450 nm was measured using a microplate reader. - Measurement of MDA and Fe²⁺ levels
Intracellular malondialdehyde (MDA) and Fe2⁺ levels were measured using the corresponding assay kits according to the manufacturers’ instructions. Total protein was quantified using a bicinchoninic acid assay, and each readout was normalized to the protein content of the corresponding sample. - Copper ion fluorescent probe assay
After removal of the culture medium, a Cu2⁺-specific fluorescent probe was added at a final concentration of 1 µM and incubated for 30 min at 37 °C. Images were acquired immediately without washing using an inverted fluorescence microscope at excitation and emission wavelengths of 510 and 578 nm, respectively. Acquisition settings were held constant across groups, and the image scale bar was calibrated at 50 µm. - RT-qPCR
Total RNA was isolated using a phenol–guanidinium-based RNA extraction reagent. Complementary DNA was synthesized from 1 µg of RNA using a first-strand complementary DNA synthesis kit, and SYBR Green-based reagents were used for real-time PCR. Each reaction was performed in technical triplicate. Primer sequences are provided in Supplementary Table 2, and relative messenger RNA abundance was calculated using the 2⁻ΔΔCt method.
12. Statistical analysis
Normality and homogeneity of variance were assessed using the Shapiro–Wilk and median-centered Levene tests, respectively. Data meeting both assumptions were analyzed using an unpaired two-sided t-test or one-way analysis of variance followed by Tukey’s honestly significant difference test. When either assumption was not met, Welch’s t-test or Welch’s analysis of variance followed by Games–Howell comparisons was used.
Results are presented as mean ± standard deviation. Actual biological replicate counts, exact two-sided P values, and assumption checks are reported in Supplementary Table 3. The archived Figure 13 source sheet contains fewer control replicates for panels A (n = 4), C (n = 4), and F (n = 2), whereas all other panel/group combinations contain n = 6.
Computational analyses were performed using R and the limma, WGCNA, clusterProfiler, glmnet, randomForest, e1071, pROC, rms, Seurat v4, Harmony, Monocle 2, CellChat, and Cytoscape/CytoHubba workflows. Online resources included FerrDB, GEO, STRING, ChEA3, TarBase 9.0, NetworkAnalyst, CIBERSORT/LM22, MSigDB, Bioinformatics.com.cn, and Sangerbox 3.0.
Results
Identification and functional annotation of cuproptosis–ferroptosis co-expressed genes
Four training datasets comprising 421 UC samples and 97 healthy controls were integrated. A total of 551 differentially expressed genes, including 362 upregulated and 189 downregulated genes, were identified. Correlation analysis yielded 444 ferroptosis–cuproptosis-correlated genes, and intersection with the differentially expressed genes produced 32 cuproptosis–ferroptosis co-expressed differentially expressed genes (CF-DEGs).
The 32 CF-DEGs were enriched in responses to wounding, copper ions, fatty-acid transport, lipopolysaccharide, inflammatory bowel disease, NF-κB signaling, TNF signaling, and ferroptosis (Supplementary Table 4).
Prioritization of core biomarkers through integrated machine learning and WGCNA
Weighted gene co-expression network analysis (WGCNA) identified MEpurple, MEbrown, and MEblack as the modules most strongly associated with UC. Together, the selected modules contained 1,426 genes (Supplementary Table 5).
Least absolute shrinkage and selection operator (LASSO), support vector machine–recursive feature elimination (SVM-RFE), and random forest selected 21, 32, and 19 features, respectively. Thirteen genes were shared by all three models. Integration of the machine-learning consensus, the CytoHubba-ranked protein–protein interaction core, and the selected WGCNA modules yielded five UC-associated candidate biomarkers: LCN2, IDO1, CXCL2, NOS2, and CD274.
Diagnostic evaluation and pathway enrichment of the biomarker signature
All five candidate biomarkers were positively correlated and upregulated in UC in the training cohort (Figure 2A, B). Each marker achieved an area under the receiver operating characteristic curve > 0.80 in the training cohort and > 0.75 in the independent GSE47908 validation cohort (Figure 2C–E).
The five-gene nomogram showed favorable calibration, with a Hosmer–Lemeshow P > 0.05 and mean absolute error < 0.1, and discrimination in the analyzed cohort, with an area under the curve and concordance index of 0.945 and a 95% confidence interval for the concordance index of 0.922–0.968 (Figure 3).
Gene set enrichment analysis linked the five candidates to JAK–STAT, interferon–RIPK1/3, and Toll-like receptor–NF-κB signaling (Figure 4).
Figure 2: Candidate-biomarker expression and diagnostic performance. (A) Correlation heatmap. (B, C) Biomarker expression and receiver operating characteristic curves in the training cohort. (D, E) Biomarker expression and receiver operating characteristic curves in the GSE47908 validation cohort. ROC: receiver operating characteristic; AUC: area under the curve; UC: ulcerative colitis. Please click here to view a larger version of this figure.
Figure 3: Development and assessment of the UC nomogram. (A) Five-gene nomogram. (B) Calibration plot. (C) Receiver operating characteristic curve. ROC: receiver operating characteristic; UC: ulcerative colitis; AUC: area under the curve; C-index: concordance index. Please click here to view a larger version of this figure.
Figure 4: Gene set enrichment analysis of the five candidate biomarkers. (A–E) Gene set enrichment analysis results for LCN2, IDO1, CXCL2, NOS2, and CD274, respectively. GSEA: gene set enrichment analysis. Please click here to view a larger version of this figure.
Immune microenvironment landscape and regulatory network analysis
UC samples showed increased proportions of neutrophils, activated memory CD4⁺ T cells, M1 macrophages, and activated mast cells, with reciprocal decreases in M2 macrophages, resting mast cells, and resting dendritic cells. These patterns were reproduced in the validation cohort (Figure 5A–E).
The five candidates correlated positively with neutrophils, activated memory CD4⁺ T cells, and M1 macrophages and inversely with resting mast cells and M2 macrophages (Figure 5F–J). The gene miRNA network contained 98 nodes and 146 edges. hsa-miR-34a-5p and hsa-miR-16-5p showed the highest biomarker connectivity, whereas AR and RELA were the most connected transcription factors (Figure 6).
Figure 5: Immune-infiltration analysis. (A) Immune-cell composition. (B) Between-group differences in immune-cell proportions. (C) Immune-cell correlation heatmap. (D, E) Immune-cell infiltration in the training and validation cohorts. (F–J) Correlations between candidate biomarkers and immune-cell populations. UC: ulcerative colitis. Please click here to view a larger version of this figure.
Figure 6: Predicted miRNA and transcription-factor networks. (A) Gene–miRNA network. Circles indicate candidate biomarkers, and squares indicate miRNAs. (B) Transcription factor–gene network. Diamonds indicate candidate biomarkers, and inverted triangles indicate transcription factors. miRNA: microRNA; TF: transcription factor. Please click here to view a larger version of this figure.
Spatiotemporal expression dynamics at single-cell resolution
After quality control, 31,712 cells formed 22 clusters that were annotated as nine major cell populations (Figure 7A–C). Epithelial and plasma-cell proportions were higher in UC. LCN2 and NOS2 were enriched in epithelial cells, whereas IDO1, CXCL2, and CD274 were enriched in myeloid cells (Figure 7D–G).
Epithelial-cell subclustering identified 11 subsets, with expansion of inflammatory colonocytes and enrichment of LCN2 and NOS2 in this subset (Figure 8). Myeloid-cell subclustering identified seven subsets, with increased monocytes, reduced macrophages, and enrichment of IDO1, CXCL2, and CD274 in monocytes (Figure 9).
Inflammatory colonocytes accumulated late in the epithelial trajectory, with increasing LCN2 and NOS2 expression (Figure 10). Monocytes exhibited a distinct UC-associated trajectory, with dynamic expression of IDO1, CXCL2, and CD274 (Figure 11).
Inflammatory colonocytes showed the strongest outgoing signaling and prominent communication with monocytes. APP–CD74 was a leading ligand–receptor pair between these subsets (Figure 12).
Figure 7: Candidate-biomarker expression across cell populations. (A) Cell clusters. (B) Annotation markers. (C) Nine annotated cell populations. (D, E) Cell distributions and proportions in healthy and UC samples. (F, G) Candidate-biomarker expression visualized by UMAP and bubble plot. UC: ulcerative colitis; UMAP: uniform manifold approximation and projection. Please click here to view a larger version of this figure.
Figure 8: Epithelial-cell subclustering and candidate expression. (A) Initial epithelial-cell clusters. (B) Annotation markers. (C) Annotated epithelial-cell subsets. (D) Subset proportions in healthy and UC samples. (E) LCN2 and NOS2 expression across epithelial-cell subsets. t-SNE: t-distributed stochastic neighbor embedding; UC: ulcerative colitis. Please click here to view a larger version of this figure.
Figure 9: Myeloid-cell subclustering and candidate expression. (A) Initial myeloid-cell clusters. (B) Annotation markers. (C) Annotated myeloid-cell subsets. (D) Subset proportions in healthy and UC samples. (E) IDO1, CXCL2, and CD274 expression across myeloid-cell subsets. t-SNE: t-distributed stochastic neighbor embedding; UC: ulcerative colitis. Please click here to view a larger version of this figure.
Figure 10: Epithelial-cell pseudotime analysis. (A) Pseudotime trajectory and state assignments. (B) Distribution of healthy and UC epithelial cells along the trajectory. (C) LCN2 and NOS2 expression dynamics along pseudotime. UC: ulcerative colitis. Please click here to view a larger version of this figure.
Figure 11: Myeloid-cell pseudotime analysis. (A) Pseudotime trajectory and state assignments. (B) Distribution of healthy and UC myeloid cells along the trajectory. (C) IDO1, CXCL2, and CD274 expression dynamics along pseudotime. UC: ulcerative colitis. Please click here to view a larger version of this figure.
Figure 12: Epithelial–myeloid communication. (A) Interaction number and strength. (B) Outgoing and incoming signaling strengths. (C) Interactions involving inflammatory colonocytes. (D) Communication-strength heatmap. (E) Ligand–receptor pairs involving inflammatory colonocytes. Please click here to view a larger version of this figure.
In vitro experimental validation of ferroptosis and cuproptosis interventions
Lipopolysaccharide (LPS) exposure reduced Caco-2 cell viability and increased IL-6 and IL-1β expression relative to the vehicle control (Figure 13A–C and Supplementary Figure 5).
LPS increased Fe2⁺, malondialdehyde, and the messenger RNA abundance of LCN2, IDO1, CXCL2, and NOS2, whereas ferrostatin-1 reversed each LPS-associated change (Figure 13D–I). For the LPS versus LPS + ferrostatin-1 comparisons, exact two-sided P values ranged from 9.45 × 10⁻5 to 0.0027 after averaging technical replicates within each biological replicate.
LPS reduced viability and increased IL-6, IL-1β, FDX1/DLAT, CD274, and copper-sensitive fluorescence, whereas tetrathiomolybdate reversed these changes (Figure 14 A–E). For the LPS versus LPS + tetrathiomolybdate comparisons, exact two-sided P values ranged from < 1 × 10⁻15 to 0.0008.
Together, the analyses identified five UC-associated diagnostic candidates, localized their expression to epithelial and myeloid populations, and showed that LCN2, IDO1, CXCL2, and NOS2 responded to ferroptosis inhibition, whereas CD274 responded to copper chelation in Caco-2 cells.
Figure 13: Ferrostatin-1 attenuates LPS-associated inflammatory and ferroptosis-related changes in Caco-2 cells. (A) Cell viability. (B, C) IL-6 and IL-1β messenger RNA expression. (D, E) Intracellular malondialdehyde and Fe2⁺ levels. (F–I) LCN2, IDO1, CXCL2, and NOS2 messenger RNA expression. Technical triplicates were averaged within each independent biological replicate; error bars indicate standard deviation. Available biological replicate counts were Control n = 4 in A and C, Control n = 2 in F, and n = 6 for all other panel/group combinations. Exact two-sided P values and assumption checks are reported in Supplementary Table 5. Fer-1: ferrostatin-1; LPS: lipopolysaccharide; MDA: malondialdehyde. Please click here to view a larger version of this figure.
Figure 14: Tetrathiomolybdate attenuates LPS-associated inflammatory and cuproptosis-related changes in Caco-2 cells. (A) Cell viability. (B, C) IL-6 and IL-1β messenger RNA expression. (D) FDX1 and DLAT messenger RNA expression. (E) CD274 messenger RNA expression. (F) Intracellular Cu2⁺ fluorescence; scale bar = 50 µm. Technical triplicates were averaged within each of six independent biological replicates per group; error bars indicate standard deviation. Exact two-sided P values and assumption checks are reported in Supplementary Table 5. TTM: tetrathiomolybdate; LPS: lipopolysaccharide. Please click here to view a larger version of this figure.
Raw and processed data, analysis scripts, and cell-experiment source spreadsheets are publicly available at https://doi.org/10.5281/zenodo.21202720.
Supplementary Figure 1: Data preprocessing and differential-expression diagnostics. (A–F) Sample distributions before (A–C) and after (D–F) batch correction. (G) Volcano plot of differentially expressed genes. (H) Heatmap of normalized expression patterns. DEGs: differentially expressed genes; GEO: Gene Expression Omnibus.Please click here to download this file.
Supplementary Figure 2: Enrichment analysis of CF-DEGs. (A) Gene Ontology enrichment. (B) Kyoto Encyclopedia of Genes and Genomes enrichment. CF-DEGs: cuproptosis–ferroptosis co-expressed differentially expressed genes; GO: Gene Ontology; KEGG: Kyoto Encyclopedia of Genes and Genomes.Please click here to download this file.
Supplementary Figure 3: WGCNA screening diagnostics. (A) Soft-threshold selection. (B) Module-eigengene clustering. (C) Gene-module dendrogram. (D) Module–trait associations. (E–G) Gene-significance/module-membership relationships for the purple, brown, and black modules. WGCNA: weighted gene co-expression network analysis; UC: ulcerative colitis.Please click here to download this file.
Supplementary Figure 4: Multistage biomarker screening. (A, B) LASSO feature selection. (C, D) SVM-RFE feature selection. (E, F) Random-forest feature ranking. (G) Consensus among the three models. (H, I) PPI network and CytoHubba core nodes. (J) Integration of machine learning, WGCNA, and PPI results. (K) Chromosomal locations of the five candidates. LASSO: least absolute shrinkage and selection operator; SVM-RFE: support vector machine–recursive feature elimination; RF: random forest; PPI: protein–protein interaction; WGCNA: weighted gene co-expression network analysis.Please click here to download this file.
Supplementary Figure 5: LPS model validation in Caco-2 cells. (A) Cell viability. (B, C) IL-6 and IL-1β messenger RNA expression. Technical triplicates were averaged within six independent biological replicates per group; error bars indicate standard deviation. LPS: lipopolysaccharide.Please click here to download this file.
Supplementary Table 1: GEO datasets used for discovery and validation. Dataset accession numbers, platforms, sample composition, and cohort assignments for the GEO datasets included in the discovery and validation analyses. GEO: Gene Expression Omnibus.Please click here to download this file.
Supplementary Table 2: Reverse transcription quantitative PCR primer sequences. Primer sequences used for reverse transcription quantitative PCR analysis of the genes evaluated in this study.Please click here to download this file.
Supplementary Table 3: Statistical details for Figures 13, 14, and Supplementary Figure 1. Biological-replicate counts, assumption checks, statistical analysis methods, and exact P values for the indicated experimental comparisons.Please click here to download this file.
Supplementary Table 4: Cuproptosis–ferroptosis co-expressed differentially expressed genes. List of CF-DEGs identified by intersecting ferroptosis–cuproptosis-correlated genes with differentially expressed genes.Please click here to download this file.
Supplementary Table 5: Genes in the WGCNA modules selected for biomarker screening. List of genes contained in the WGCNA modules selected for subsequent biomarker screening. WGCNA: weighted gene co-expression network analysis.Please click here to download this file.
Discussion
This study integrated bulk transcriptomics, machine learning, single-cell analysis, and targeted cell experiments to examine ferroptosis–cuproptosis crosstalk in UC. Five UC-associated diagnostic candidates were reproducible across discovery and validation cohorts, localized mainly to epithelial and myeloid compartments, and showed pathway-inhibition-responsive expression in Caco-2 cells. These findings support a staged workflow in which computational prioritization guides focused biological validation while avoiding the stronger claim that the five genes are established upstream regulators of ferroptosis or cuproptosis.
The CF-DEGs were enriched in responses to lipopolysaccharide, copper handling, lipid transport, NF-κB/TNF signaling, and ferroptosis, providing a coherent link between metal-dependent oxidative stress and mucosal inflammation10,19.
The five candidates occupy complementary inflammatory and metal-stress contexts. LCN2 links iron sequestration, microbial dysbiosis, and the LCN2–ALOX15 ferroptosis axis20,21,22,23; IDO1 connects tryptophan metabolism, mucosal inflammation, and disturbed iron handling24,25,26; and CXCL2 links IL-17/IL-22 signaling, neutrophil recruitment, and lipid peroxidation27,28,29,30. NOS2 has context-dependent effects but can amplify oxidative injury and ferroptosis susceptibility when excessively induced31,32,33,34,35,36. CD274 is an immune-regulatory epithelial marker whose relationship with FDX1 and copper-ionophore sensitivity suggests cuproptosis responsiveness rather than established pathway control37,38,39. The inhibitor experiments were consistent with these classifications: ferrostatin-1 reduced LCN2, IDO1, CXCL2, and NOS2, whereas tetrathiomolybdate reduced CD274.
The immune analysis further connected the candidate signature to M1 macrophages, neutrophils, activated memory CD4⁺ T cells, and activated mast cells, all of which can sustain epithelial injury through cytokines, reactive oxygen species, and barrier-disrupting signaling40,41,42,43,44. RELA, hsa-miR-34a-5p, and hsa-miR-16-5p emerged as plausible upstream regulators, whereas evidence for a direct androgen-receptor role in UC remains limited45,46,47,48,49,50. These findings should be interpreted as network-level hypotheses for future perturbation experiments rather than as proof of direct regulation.
Single-cell analysis localized LCN2 and NOS2 to inflammatory colonocytes and IDO1, CXCL2, and CD274 to monocyte-rich myeloid populations. Inflammatory colonocytes were predicted to act as communication hubs, with APP–CD74 among the leading epithelial–myeloid ligand–receptor pairs51,52,53,54,55,56. This cell-type resolution narrows the biological context in which the five candidates should be tested and supports co-culture or organoid–immune-cell systems as the next experimental step.
Several limitations constrain interpretation. Caco-2 is a colorectal adenocarcinoma cell line and does not fully recapitulate nonmalignant colonic epithelium, patient heterogeneity, stromal–immune interactions, or chronic UC; therefore, validation in primary intestinal epithelial cells, patient-derived organoids, co-culture systems, and in vivo colitis models is needed. The archived Figure 13 source spreadsheet also lacked some control replicate records in panels A, C, and F, reducing precision for those comparisons; all available replicate counts and exact P values are disclosed in Supplementary Table 5. Finally, age, sex, treatment exposure, and other clinical covariates were not uniformly available across public cohorts. Prospective multicenter cohorts and direct perturbation of the five candidates will be required to establish diagnostic utility and causal roles. Within these limitations, the study provides a transparent, experimentally anchored set of hypotheses linking metal-dependent cell death, epithelial myeloid communication, and UC. Comparable network-pharmacology research in prostate cancer illustrates the hypothesis-generating value of computational target prioritization but does not constitute UC-specific validation57. Likewise, recent discussion of ferroptosis in osteoarthritis and bone degeneration reflects broader therapeutic interest in ferroptosis modulation, although disease-specific mechanisms cannot be directly extrapolated to UC58. External-cohort validation and inhibitor experiments were therefore used to strengthen, but not overstate, the bioinformatic associations.
Disclosures
The authors have nothing to disclose.
Acknowledgements
This work was supported by the Young Qihuang Scholar Project of the National Administration of Traditional Chinese Medicine (Grant No. 2022256). Acknowledgment is extended to all members of the research teams at Liaoning University of Traditional Chinese Medicine and the Third Affiliated Hospital of Liaoning University of Traditional Chinese Medicine for technical support and contributions to data curation and bioinformatics analysis.
OpenAI Codex (OpenAI) was used during revision solely to assist with English-language editing, document formatting, and figure-quality checks. All scientific content, analyses, citations, and the final presentation were reviewed and verified by the authors, who take full responsibility for the manuscript.
Materials
| Name | Company | Catalog Number | Comments |
|---|---|---|---|
| BCA Protein Assay Kit | Beyotime Institute of Biotechnology | P0010 | Protein quantification for normalization. |
| Bioinformatics.com.cn | Bioinformatics.com.cn | Online platform; accessed 2026 | GO/KEGG visualization and analysis platform. |
| Caco-2 cell line | ATCC | HTB-37 | Human colorectal adenocarcinoma cell line used for the epithelial injury model. |
| Cell Counting Kit-8 | GLPBio | GK10001 | Cell-viability assay measured at 450 nm. |
| CIBERSORT with LM22 signature | CIBERSORT developers | LM22 | Deconvolution of 22 immune-cell fractions. |
| ChEA3 | Ma'ayan Laboratory | Online platform; accessed 2026 | Transcription-factor enrichment and regulatory-network prediction. |
| CO2 incubator | Not recorded | N/A | Humidified incubator used for culture at 37 °C and 5% CO2. |
| Cu2+ fluorescent probe | BIOFOUNT | CAS 98907-26-7 | Intracellular Cu2+ fluorescence detection. |
| Cytoscape with CytoHubba | Cytoscape Consortium | Version not recorded | PPI visualization and core-node ranking. |
| Dulbecco's modified Eagle medium | Gibco | 11965092 | Caco-2 cell culture medium. |
| Dimethyl sulfoxide | Not recorded | N/A | Vehicle control. |
| Fetal bovine serum | Not recorded | N/A | Medium supplement at 10%. |
| Fe2+ Assay Kit | Abcam | ab83366 | Measurement of intracellular Fe2+. |
| FerrDB | FerrDB developers | Online database; accessed 2026 | Source of ferroptosis-related genes. |
| Ferrostatin-1 | Sigma-Aldrich | SML0583 | Ferroptosis inhibitor; final concentration 4 µM. |
| Harmony | Broad Institute/community | Version not recorded | Single-cell batch correction. |
| Inverted fluorescence microscope | Olympus | IX71 | Cu2+ fluorescence imaging at 510/578 nm excitation/emission. |
| Lipopolysaccharide | Not recorded | N/A | Inflammatory injury stimulus at 1 µg/mL for 24 h. |
| Malondialdehyde Assay Kit | Beyotime Institute of Biotechnology | S0131S | Measurement of lipid peroxidation. |
| Microplate reader | Not recorded | N/A | Absorbance measurement at 450 nm. |
| Monocle | Bioconductor/community | Version 2 | Single-cell pseudotime analysis. |
| Molecular Signatures Database | Broad Institute | c2.cp.kegg_medicus.v2025.1.Hs | Reference gene set for GSEA. |
| NetworkAnalyst | NetworkAnalyst developers | Online platform; accessed 2026 | Integration of transcription-factor and miRNA networks. |
| Penicillin–streptomycin | Not recorded | N/A | 100 U/mL penicillin and 100 µg/mL streptomycin. |
| R | R Foundation for Statistical Computing | Version not recorded | Bioinformatics analysis environment. |
| R package set | CRAN/Bioconductor | Package builds not recorded | limma, WGCNA, clusterProfiler, glmnet, randomForest, e1071, pROC, and rms workflows. |
| Real-time PCR system | Not recorded | N/A | SYBR Green RT-qPCR instrument. |
| RevertAid First Strand cDNA Synthesis Kit | Thermo Fisher Scientific | K1622 | First-strand cDNA synthesis from 1 µg RNA. |
| Sangerbox | Sangerbox developers | 3.0 | Normalization and ComBat batch correction. |
| Seurat | Satija Lab/community | Version 4 | Single-cell quality control, normalization, clustering, and visualization. |
| Statistical analysis software | Not recorded | N/A | Statistical analysis of the original cell experiments. |
| STRING | STRING Consortium | Online database; accessed 2026 | Protein–protein interaction network construction. |
| SYBR Green real-time PCR reagents | Takara | RR820A | Quantitative real-time PCR. |
| TarBase | DIANA Tools | 9.0 | Experimentally supported miRNA–gene interactions. |
| Tetrathiomolybdate | Sigma-Aldrich | 323446 | Copper chelator; final concentration 40 µM. |
| TRIzol reagent | Takara | 9108 | Total RNA extraction. |
References
- Fu M, Wang QW, Liu YR, Chen SJ. The role of the three major intestinal barriers in ulcerative colitis in the elderly. Ageing Res Rev. 2025;108:102752. https://doi.org/10.1016/j.arr.2025.102752.
- Le Berre C, Honap S, Peyrin-Biroulet L. Ulcerative colitis. Lancet. 2023;402:571–584.
- Wangchuk P, Yeshi K, Loukas A. Ulcerative colitis: Clinical biomarkers, therapeutic targets, and emerging treatments. Trends Pharmacol Sci. 2024;45:892–903.
- Kobayashi T, et al. Ulcerative colitis (primer). Nat Rev Dis Primers. 2020;6:74. https://doi.org/10.1038/s41572-020-0205-x.
- Huang J, et al. Comprehensive analysis of cuproptosis-related genes in immune infiltration and diagnosis in ulcerative colitis. Front Immunol. 2022;13:1008146. https://doi.org/10.3389/fimmu.2022.1008146.
- Liu Y, et al. Development and experimental validation of machine learning-based disulfidptosis-related ferroptosis biomarkers in inflammatory bowel disease. Genes. 2025;16:496. https://doi.org/10.3390/genes16050496.
- Hu W, et al. HIF-1α alleviates ferroptosis in ulcerative colitis by regulation of GPX4. Cell Death Dis. 2025;16:542. https://doi.org/10.1038/s41419-025-07883-8.
- Zhao C, et al. FXR acts as a therapeutic target for ulcerative colitis via suppressing ferroptosis. Mol Med. 2025;31:258. https://doi.org/10.1186/s10020-025-01305-3.
- Tsvetkov P, et al. Copper induces cell death by targeting lipoylated TCA cycle proteins. Science. 2022;375:1254–1261.
- Liu J, et al. Unveiling the cuproptosis in colitis and colitis-related carcinogenesis: A multifaceted player and immune moderator. Research (Wash D C). 2025;8:0698. https://doi.org/10.34133/research.0698.
- Liu N, Chen M. Crosstalk between ferroptosis and cuproptosis: From mechanism to potential clinical application. Biomed Pharmacother. 2024;171:116115. https://doi.org/10.1016/j.biopha.2023.116115.
- Hao S, et al. Development of prognostic model incorporating a ferroptosis/cuproptosis-related signature and mutational landscape analysis in muscle-invasive bladder cancer. BMC Cancer. 2024;24:958. https://doi.org/10.1186/s12885-024-12716-1.
- Wang Z, et al. Exosomes secreted by macrophages upon copper ion stimulation can promote angiogenesis. Mater Sci Eng C Mater Biol Appl. 2021;123:111981. https://doi.org/10.1016/j.msec.2021.111981.
- Zou M, Zhang W, Zhu Y, Xu Y. Identification of 6 cuproptosis-related genes for active ulcerative colitis with both diagnostic and therapeutic values. Medicine (Baltimore). 2023;102:e35503. https://doi.org/10.1097/MD.0000000000035503.
- Chen W, Miao J, Chen J, Chen J. Development of machine learning models for diagnostic biomarker identification and immune cell infiltration analysis in PCOS. J Ovarian Res. 2025;18:1. https://doi.org/10.1186/s13048-024-01582-8.
- Lv JH, et al. WGCNA combined with machine learning to find potential biomarkers of liver cancer. Medicine (Baltimore). 2023;102:e36536. https://doi.org/10.1097/MD.0000000000036536.
- Wu Q, Yang C, Huang C, Lin Z. Screening key genes for intracranial aneurysm rupture using LASSO regression and the SVM-RFE algorithm. Front Med. 2025;11:1487224. https://doi.org/10.3389/fmed.2024.1487224.
- Tang D, Pu B, Liu S, Li H. Identification of cuproptosis-associated subtypes and signature genes for diagnosis and risk prediction of ulcerative colitis based on machine learning. Front Immunol. 2023;14:1142215. https://doi.org/10.3389/fimmu.2023.1142215.
- Yang C, et al. Identification of cuproptosis hub genes contributing to the immune microenvironment in ulcerative colitis using bioinformatic analysis and experimental verification. Front Immunol. 2023;14:1113385. https://doi.org/10.3389/fimmu.2023.1113385.
- Yadav SK, et al. Fecal Lcn-2 level is a sensitive biological indicator for gut dysbiosis and intestinal inflammation in multiple sclerosis. Front Immunol. 2022;13:1015372. https://doi.org/10.3389/fimmu.2022.1015372.
- Luo Q, et al. Bioinformatics analysis reveals potential crosstalk genes and molecular mechanisms between ulcerative colitis and psoriasis. Arch Dermatol Res. 2024;317:118. https://doi.org/10.1007/s00403-024-02856-1.
- Nielsen OH, et al. Rectal dialysate and fecal concentrations of neutrophil gelatinase-associated lipocalin, interleukin-8, and tumor necrosis factor-alpha in ulcerative colitis. Am J Gastroenterol. 1999;94:2923–2928.
- Sun Y, et al. Natural polyphenol rutin inhibits ferroptosis in ulcerative colitis by regulating lipid metabolism via blocking the LCN2-ALOX15 axis. Food Funct. 2025;16:7169–7186.
- Sofia MA, et al. Tryptophan metabolism through the kynurenine pathway is associated with endoscopic inflammation in ulcerative colitis. Inflamm Bowel Dis. 2018;24:1471–1480.
- Shon WJ, et al. Severity of DSS-induced colitis is reduced in Ido1-deficient mice with down-regulation of TLR-MyD88-NF-kB transcriptional networks. Sci Rep. 2015;5:17305. https://doi.org/10.1038/srep17305.
- Zeng T, et al. Indoleamine 2, 3-dioxygenase 1 enhances hepatocytes ferroptosis in acute immune hepatitis associated with excess nitrative stress. Free Radic Biol Med. 2020;152:668–679.
- Pavlidis P, et al. Interleukin-22 regulates neutrophil recruitment in ulcerative colitis and is associated with resistance to ustekinumab therapy. Nat Commun. 2022;13:5820. https://doi.org/10.1038/s41467-022-33333-3.
- Urzua U, et al. The ovarian transcriptome of reproductively aged multiparous mice: Candidate genes for ovarian cancer protection. Biomolecules. 2020;10:113. https://doi.org/10.3390/biom10010113.
- Zahn A, et al. Transcript levels of different cytokines and chemokines correlate with clinical and endoscopic activity in ulcerative colitis. BMC Gastroenterol. 2009;9:13. https://doi.org/10.1186/1471-230X-9-13.
- Yi Q, et al. Application of genomic selection and experimental techniques to predict cell death and immunotherapeutic efficacy of ferroptosis-related CXCL2 in hepatocellular carcinoma. Front Oncol. 2022;12:998736. https://doi.org/10.3389/fonc.2022.998736.
- Kolios G, Valatas V, Ward SG. Nitric oxide in inflammatory bowel disease: A universal messenger in an unsolved puzzle. Immunology. 2004;113:427–437.
- Cheng F, et al. Identification of differential intestinal mucosa transcriptomic biomarkers for ulcerative colitis by bioinformatics analysis. Dis Markers. 2020;2020:8876565. https://doi.org/10.1155/2020/8876565.
- Tratensek A, et al. Oxidative stress-related biomarkers as promising indicators of inflammatory bowel disease activity: A systematic review and meta-analysis. Redox Biol. 2024;77:103380. https://doi.org/10.1016/j.redox.2024.103380.
- Zhang JL, et al. Jatrorrhizine alleviates ulcerative colitis via regulating gut microbiota and NOS2 expression. Gut Pathog. 2022;14:41. https://doi.org/10.1186/s13099-022-00516-7.
- Li H, et al. Development of a prognostic model based on ferroptosis-related genes for colorectal cancer patients and exploration of the biological functions of NOS2 in vivo and in vitro. Front Oncol. 2023;13:1133946. https://doi.org/10.3389/fonc.2023.1133946.
- Seibt TM, Proneth B, Conrad M. Role of GPX4 in ferroptosis and its pharmacological implication. Free Radic Biol Med. 2019;133:144–152.
- Holland T, Mattner J. Intestinal eosinophils: Characterization of elusive granulocytes as anti-bacterial and immunomodulatory effector cells in colitis. Signal Transduct Target Ther. 2023;8:321. https://doi.org/10.1038/s41392-023-01588-4.
- Nguyen J, et al. Overexpression of programmed death ligand 1 in refractory inflammatory bowel disease. Hum Pathol. 2022;126:19–27.
- Shen Z, et al. Exploration of a screening model for intrahepatic cholangiocarcinoma patients prone to cuproptosis and mechanisms of the susceptibility of CD274-knockdown intrahepatic cholangiocarcinoma cells to cuproptosis. Cancer Gene Ther. 2023;30:1663–1678.
- Niu Y, et al. Platelet activation stimulates macrophages to enhance ulcerative colitis through PF4/CXCR3 signaling. Int J Mol Med. 2025;55:78. https://doi.org/10.3892/ijmm.2025.5499.
- Qu F, et al. The role of macrophage polarization in ulcerative colitis and its treatment. Microb Pathog. 2025;199:107227. https://doi.org/10.1016/j.micpath.2025.107227.
- Danne C, Skerniskyte J, Marteyn B, Sokol H. Neutrophils: From IBD to the gut microbiota. Nat Rev Gastroenterol Hepatol. 2024;21:184–197.
- Zhao Q, et al. CD4+ T cell activation and concomitant mTOR metabolic inhibition can ablate microbiota-specific memory cells and prevent colitis. Sci Immunol. 2020;5:eabc6373. https://doi.org/10.1126/sciimmunol.abc6373.
- Chen E, et al. Inflamed ulcerative colitis regions associated with MRGPRX2-mediated mast cell degranulation and cell activation modules, defining a new therapeutic target. Gastroenterology. 2021;160:1709–1724.
- Xu H, Wu C, Wang D, Wang H. Alleviating effect of Nexrutine on mucosal inflammation in mice with ulcerative colitis: Involvement of the RELA suppression. Immun Inflamm Dis. 2024;12:e1147. https://doi.org/10.1002/iid3.1147.
- Luo Y, et al. Rab27A promotes cellular apoptosis and ROS production by regulating the miRNA-124-3p/STAT3/RelA signalling pathway in ulcerative colitis. J Cell Mol Med. 2020;24:11330–11342.
- Pace F, Watnick PI. The interplay of sex steroids, the immune response, and the intestinal microbiota. Trends Microbiol. 2021;29:849–859.
- Rodriguez J, Olivares M, Delzenne NM. Implication of the gut microbiota in metabolic inflammation associated with nutritional disorders and obesity. Mol Nutr Food Res. 2021;65:e1900481. https://doi.org/10.1002/mnfr.201900481.
- Ojha R, et al. Emerging role of circulating microRNA in the diagnosis of human infectious diseases. J Cell Physiol. 2019;234:1030–1043.
- Tian Y, et al. LncRNA CDKN2B-AS1 relieved inflammation of ulcerative colitis via sponging miR-16 and miR-195. Int Immunopharmacol. 2020;88:106970. https://doi.org/10.1016/j.intimp.2020.106970.
- Kulkarni N, Pathak M, Lal G. Role of chemokine receptors and intestinal epithelial cells in the mucosal inflammation and tolerance. J Leukoc Biol. 2017;101:377–394.
- Stadnyk AW. Intestinal epithelial cells as a source of inflammatory cytokines and chemokines. Can J Gastroenterol. 2002;16:241–246.
- Xu Y, Chen N. DUOX2-mediated innate immune hyperactivation drives macrophage M1 polarization via the APP-CD74 axis in inflammatory bowel disease. Presented at: 19th Congress of the European Crohn’s and Colitis Organisation (ECCO 2026); Stockholm, Sweden; 2026. https://academic.oup.com/ecco-jcc/article/20/Supplement_1/jjaf231.232/8432779.
- Shi CL, et al. Single-nucleus transcriptome unveils the role of ferroptosis in ischemic stroke. Heliyon. 2024;10:e32727. https://doi.org/10.1016/j.heliyon.2024.e32727.
- Tu H, et al. Integrated single-cell analysis with experimental validation reveals ANXA2 as a therapeutic target for ferroptosis in inflammatory bowel disease. Am J Transl Res. 2026;18:91–111.
- Yan C, et al. Exploring the anti-ferroptosis mechanism of Kai-Xin-San against Alzheimer's disease through integrating network pharmacology, bioinformatics, and experimental validation strategy in vivo and in vitro. J Ethnopharmacol. 2024;326:117915. https://doi.org/10.1016/j.jep.2024.117915.
- Shaik JP, Ahmad A. Pharmacological mechanism of cinnamic acid in the treatment of prostate cancer by network pharmacology. Archives of Medical Reports. 2024;1(2):1–6. https://doi.org/10.26524/amr.2024.2.1.
- Nimmagadda S, Patel S, MR Vishwathika. Targeting ferroptosis: A novel therapeutic avenue in osteoarthritis and bone degeneration. Archives of Medical Reports. 2025;2(5):6–9. https://doi.org/10.26524/amr.2025.5.2.