Comprehensive proteomic meta-analysis identifies novel proteomics alterations in proliferative diabetic retinopathy

Proliferative diabetic retinopathy (PDR) is a major cause of vision impairment. Despite clinical advances, the underlying molecular mechanisms remain unclear. Proteomics, specifically mass spectrometry, has been instrumental for the unbiased exploration of pathogenic disease alterations; however, current studies have limitations, including a lack of specific focus on PDR, the oversight of potential biomarkers, and insufficient experimental validation. In this study, we conducted a meta-analysis of proteomics studies in patients with PDR. Initially, seven studies with 127 samples were included, but later one was excluded due to the use of tear sample and another was excluded due to the deviations observed through principal component analysis (PCA). The remaining five datasets were analyzed using a random effects model. PCA revealed no clear separation by diagnosis, dataset, labeling, instrument, or ocular region. We identified 681 differentially expressed proteins, and the top three upregulated proteins were identified in only one dataset. Enrichment analyses revealed significant pathways or Gene Ontology terms, such as complement and coagulation cascades and extracellular matrix. Novel pathways involving the cytoskeleton of muscle cells were identified. We constructed a PPI network and identified APP, FN1, and CNTN1 as hub proteins with high degree centrality. Consistent with meta-analysis results, Western blot validation in OIR mice and STZ induced diabetic mice showed significantly downregulated APP and CNTN1. This study presents a novel comprehensive meta-analysis of proteomics studies on patients with PDR, offering novel insights into proteomics alterations in PDR and highlighting the potential areas for further research and therapeutic interventions.