Comparative proteomics of proliferative diabetic retinopathy in people with Type 2 diabetes highlights the role of inflammation, visual transduction, and extracellular matrix pathways

Abstract

Purpose: To explore the vitreous humor proteome from type 2 diabetes subjects with proliferative diabetic retinopathy (PDR) in the Indian population.

Methods: We performed mass spectrometry-based label-free quantitative analysis of vitreous proteome of PDR (n = 13) and idiopathic macular hole (IMH; control) subjects (n = 14). Nine samples of PDR and 10 samples of IMH were pooled as case and control, respectively, and compared. Four samples each of PDR and IMH were analyzed individually without pooling to validate the results of the pooled analysis. Comparative quantification was performed using Scaffold software which calculated the fold changes of differential expression. Bioinformatics analysis was performed using DAVID and STRING software.

Results: We identified 469 proteins in PDR and 517 proteins in IMH vitreous, with an overlap of 172 proteins. Also, 297 unique proteins were identified in PDR and 345 in IMH. In PDR vitreous, 37 proteins were upregulated (P < 0.05) and 19 proteins were downregulated compared to IMH. Protein distribution analysis clearly demonstrated a separation of protein expression in PDR and IMH. Significantly upregulated proteins included fibrinogen gamma chain, fibrinogen beta chain, and carbonic anhydrase 1 and downregulated proteins included alpha-1-antitrypsin, retinol-binding protein 3, neuroserpin, cystatin C, carboxypeptidase E and cathepsin-D.

Conclusion: Diabetic retinopathy pathogenesis involves proteins which belong to inflammation, visual transduction, and extracellular matrix pathways. Validation-based experiments using enzyme-linked immunosorbent assay (ELISA) or western blotting are needed to establish cause and effect relationships of these proteins to the disease state, to develop them as biomarkers or drug molecules.

Keywords: Biomarker; novel; proliferative diabetic retinopathy; proteomics.

Figure 1

(a) PCA of all detected proteins showing a clear distinction in the distribution of expressed proteins of the PDR samples (P1, P2, experimental replicates) and the control sample (M1, M2, experimental replicates). (b) Hierarchical clustering analysis of all proteins, with the color gradient showing the fold change of protein abundance between the two groups and a clear separation of the protein expression in PDR compared to control. (c) Significance analysis of normalized vitreous proteins in IMH versus control by volcano plot of proteins. The X-axis shows the fold change between PDR and control (in log₂ scale), and the Y-axis shows the P value (in log₁₀ scale). Each point in the graph represents a unique protein detected in LC-MS/MS, and the differentially expressed proteins are highlighted in green. IMH = idiopathic macular hole, PCA = principal component analysis, PDR = proliferative diabetic retinopathy

Figure 2

GO of integrated proteome of IMH and control vitreous showing biological processes (a), cellular components (b), and molecular functions (c) of proteins detected. GO was performed for more than 5% significant proteins from the two groups. The highest five enriched GO terms under biological processes were proteolysis, negative regulator of endopeptidases, complement activation by classical pathway, receptor-mediated endocytosis, and platelet degranulation. The highest five enriched GO terms under cellular components were extracellular exosomes, extracellular space, extracellular region, blood microparticle, and plasma membrane. The highest five enriched GO terms under molecular functions were protein binding, structural molecule activity, serine-type endopeptidase activity, serine-type endopeptidase inhibitor activity, and calcium ion binding. The top five UniProt keywords which got enriched were polymorphism signal, secreted, disulphide bond, and glycoprotein. GO = gene ontology, IMH = idiopathic macular hole, PCA = principal component analysis, PDR = proliferative diabetic retinopathy

Figure 3

Network analysis of PDR proteome. PDR = proliferative diabetic retinopathy

Figure 4

(a) Matrisome distribution of ECM proteins detected in PDR vitreous. (b) Network analysis of matrisomal proteins. ECM = extracellular matrix, PDR = proliferative diabetic retinopathy