Identification and mechanistic investigation of novel biomarkers for diabetic retinopathy through hypoxia- and cuproptosis-associated gene analysis
- 1Health Examination Center, Guizhou Provincial People’s Hospital, Guiyang, Guizhou, China
- 2Gastroenterology Department, Guizhou Provincial People’s Hospital, Guiyang, Guizhou, China
- 3Endocrinology Department, Guizhou Provincial People’s Hospital, Guiyang, Guizhou, China
Abstract
Background: Emerging evidence indicates that hypoxia- and cuproptosis-related molecular alterations may contribute to diabetic retinopathy (DR). This study aimed to identify biomarkers associated with these processes and investigate their biological relevance in DR. Methods: The DR-related datasets GSE189005 and GSE221521 were sourced from public databases, while hypoxia-related and cuproptosis-related genes were extracted from published studies. Genes consistently differentially expressed across both datasets were intersected with those associated with hypoxia- and cuproptosis-related expression scores to derive candidate genes. Selection of candidate biomarkers was performed using machine-learning algorithms and expression consistency analysis, with diagnostic performance assessed via receiver operating characteristic (ROC) curves. Functional enrichment, immune cell infiltration, regulatory network analysis, candidate compound prediction, and reverse transcription quantitative PCR (RT-qPCR) were also conducted. Results: Intersection screening yielded 13 candidate genes. Machine-learning analysis identified three candidate biomarkers. Notably, MED19 and CA11 were consistently downregulated in DR, exhibiting area under the curve (AUC) values exceeding 0.70 in both datasets, thus qualifying as biomarkers. Functional enrichment analysis suggested MED19's involvement in “proteasome” and “spliceosome” pathways, while CA11 was linked to “regulation of autophagy” and “basal cell carcinoma” pathways. Differential infiltration of three immune cell populations (eosinophils, M2 macrophages, activated natural killer cells) was noted between DR and control groups in GSE189005, with M2 macrophages demonstrating a significant negative correlation with both biomarkers. Regulatory network analysis highlighted several candidate transcriptional regulators for MED19 and CA11, including ELF1–MED19 and KDM5B–CA11. Thirty candidate compounds/interventions were identified, including schizandrin B and pirinixic acid. RT-qPCR confirmed the significant downregulation of MED19 and CA11 in DR samples, providing preliminary support. Conclusion: MED19 and CA11 were identified as biomarkers, presenting a potential framework for therapeutic intervention in DR.
Related
- Identification of potential biomarkers for ulcerative colitis based on multidimensional data analysis and investigation of the mechanisms of therapeutic effects
- Ultrasound-activated phase-change nanodroplets for root canal biofilm eradication: mechanistic Investigation and experimental validation
- Mechanistic Investigation of Cryptotanshinone and Curcumin Combination in Attenuating BVDV-1-Induced Injury in Bovine Ovarian Granulosa Cells.
- Identification of CLOCK and MARK2 as Candidate Blood-Based Biomarkers in Type 2 Diabetes: An Integrated Bulk and Single-Cell Transcriptomic Study
- Identification of meat spoilage gene biomarkers in Pseudomonas putida using gene profiling
BibTeX
@article{feng2026identification,
title = {{Identification and mechanistic investigation of novel biomarkers for diabetic retinopathy through hypoxia- and cuproptosis-associated gene analysis}},
author = {Rong Feng and Xia Wang and Xiaodong Li and Zhifang Zhao and Bendan Long and Hui Zhang and Fengjuan Yue},
journal = {PLOS One},
year = {2026},
month = sep,
doi = {10.1371/journal.pone.0358182},
url = {https://doi.org/10.1371/journal.pone.0358182},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Identification and mechanistic investigation of novel biomarkers for diabetic retinopathy through hypoxia- and cuproptosis-associated gene analysis
AU - Rong Feng
AU - Xia Wang
AU - Xiaodong Li
AU - Zhifang Zhao
AU - Bendan Long
AU - Hui Zhang
AU - Fengjuan Yue
PY - 2026
DA - 2026/09/25
JO - PLOS One
DO - 10.1371/journal.pone.0358182
UR - https://doi.org/10.1371/journal.pone.0358182
AB - Background: Emerging evidence indicates that hypoxia- and cuproptosis-related molecular alterations may contribute to diabetic retinopathy (DR). This study aimed to identify biomarkers associated with these processes and investigate their biological relevance in DR. Methods: The DR-related datasets GSE189005 and GSE221521 were sourced from public databases, while hypoxia-related and cuproptosis-related genes were extracted from published studies. Genes consistently differentially expressed across both datasets were intersected with those associated with hypoxia- and cuproptosis-related expression scores to derive candidate genes. Selection of candidate biomarkers was performed using machine-learning algorithms and expression consistency analysis, with diagnostic performance assessed via receiver operating characteristic (ROC) curves. Functional enrichment, immune cell infiltration, regulatory network analysis, candidate compound prediction, and reverse transcription quantitative PCR (RT-qPCR) were also conducted. Results: Intersection screening yielded 13 candidate genes. Machine-learning analysis identified three candidate biomarkers. Notably, MED19 and CA11 were consistently downregulated in DR, exhibiting area under the curve (AUC) values exceeding 0.70 in both datasets, thus qualifying as biomarkers. Functional enrichment analysis suggested MED19's involvement in “proteasome” and “spliceosome” pathways, while CA11 was linked to “regulation of autophagy” and “basal cell carcinoma” pathways. Differential infiltration of three immune cell populations (eosinophils, M2 macrophages, activated natural killer cells) was noted between DR and control groups in GSE189005, with M2 macrophages demonstrating a significant negative correlation with both biomarkers. Regulatory network analysis highlighted several candidate transcriptional regulators for MED19 and CA11, including ELF1–MED19 and KDM5B–CA11. Thirty candidate compounds/interventions were identified, including schizandrin B and pirinixic acid. RT-qPCR confirmed the significant downregulation of MED19 and CA11 in DR samples, providing preliminary support. Conclusion: MED19 and CA11 were identified as biomarkers, presenting a potential framework for therapeutic intervention in DR.
SN - 1932-6203
ER -