Identification of unique biomarkers for proliferative diabetic retinopathy with tractional retinal detachment by proteomics profiling of vitreous humor

Abstract

Diabetic retinopathy (DR) is one of the major complications of diabetes, resulting in severe vision loss. Traction retinal detachment (TRD) is the main factor affecting the effect of proliferative diabetic retinopathy (PDR) surgery. Liquid Chromatography with tandem mass spectrometry (LC-MS/MS) was adopted to analyze the proteomes of the vitreous in the TRD, vitreous hemorrhage (VH) and macular hole (MH) groups. By employing bioinformatics tools for GO and KEGG pathway annotation, as well as conducting protein-protein interaction(PPI) network analysis, we investigated the functional enrichment of proteins in the TRD vitreous and their associated pathways. Additionally, peptide center analysis was performed on the proteomic data to identify key differentially expressed proteins based on screening results. Bioinformatics analysis showed that DEPs is mainly enriched in the complement, the coagulation cascade systems and regulation of actin cytoskeleton. The protein interaction network analysis showed that the central proteins were mainly related to sphingolipid metabolism. APOA4, CHI3L1, LTBP2 were significantly up-regulated in TRD, which were related to the complement system, coagulation cascade and platelet activation, sphingolipid metabolism and other pathways. APOA4 and CHI3L1 protein in patients with TRD group raised significantly in the vitreous humor, shows the potential biomarkers for TRD.

Keywords: APOA4; CHI3L1; Diabetic retinopathy; Liquid chromatography with tandem mass spectrometry (LC-MS/MS); Sphingolipid metabolism; Traction retinal detachment.

Fig. 1

Workflow of the VH proteome profile analysis and the validation of DEPs in different groups.

Fig. 2

Compares the Venn plots and the volcano plots of the different groups of DEPs. (A) Bar graph of the quantitative difference results in different histones. (B) Venn plots identified between all sample groups, 422 co-expressed DEPs in TRD, vitreous hemorrhage (VH) and Control, 54 unique DEPs in Control Group, 25 unique DEPs in TRD, and 13 unique DEPs in vitreous hemorrhage (VH). (C–E) Volcano plot shows the distribution of differentially expressed proteins in each comparison. The horizontal axis represents the multiple (FC) and the vertical axis represents the p value. Red dots are upregulated DEPs, blue dots are downregulated DEPs, and grey dots are proteins not significantly differentially expressed.(| log2 (FC) |> 0.58; p-value < 0.05).

Fig. 3

Results of cluster analysis and GO analysis of differentially expressed proteins between groups. The left column (A, B, C) shows the results of the cluster analysis. The horizontal axis represents sample information, while the vertical axis represents significantly differentially expressed proteins. The expression of significantly different proteins in different samples is standardized by Z-Score method and displayed in the heat map, where red representing significantly up-regulated proteins and blue representing significantly downregulated proteins. The middle column (D, E, F) shows the GO annotation statistics (Levle 2) of differentially expressed proteins, including BP, MF and CC, shown as in blue, red and orange, and the horizontal axis indicates the number of differentially expressed proteins under each group. The column on the left shows the Top 20 enriched GO terms of significantly DEPs in each comparison. Cluster analysis of differentially expressed proteins between TRD and vitreous hemorrhage (VH) groups, GO statistical analysis map (Level 2), string GO-term enrichment (A, D, G). Hierarchical clustering analysis and GO-term enrichment of differentially expressed proteins between vitreous hemorrhage (VH) and control groups (B, E, H).Hierarchical cluster analysis and string GO-term enrichment of differentially expressed proteins between TRD and control groups (C, F, I) .(|log2(FC)| > 0.58; P < 0.05).

Fig. 4

KEGG pathways enrichment for DEPs in the different comparisons. The KEGG pathway enrichment analysis is represented as bar plots and bubble plots, respectively. For the upregulated deg, the top-enriched pathways are indicated in red. For the down-regulated degs, the top-enriched pathways are shown in blue. In the bubble plot, the horizontal axis is the enrichment factor, and the vertical axis is the name of the KEGG pathway. The bubble color is the significance level of the enrichment, and more red represents the more significant enrichment. The size of the bubbles is proportional to the number of the protein. Bar graph and bubble plot of KEGG pathway enrichment for TRD and vitreous hemorrhage (VH) (A, D), VH and control group (B, E), TRD and control group (C, F).

Fig. 5

PPI network and functional cluster analysis results of DEPs in different comparisons. After the functional classification of the interaction network map, the proteins in the same network module often have similar biological functions, and the proteins within the functional module of interest can be selected as the focus of subsequent research. (A, B) Differential protein PPI network and major functional clusters in groups of TRD and control group. (C, D) differential protein PPI networks and major functional clusters in groups VH and B. (E) Network diagram of differentially expressed protein interactions between TRD and vitreous hemorrhage (VH) groups.

Fig. 6

Results of the candidate biomarkers in one-way ANOVA analysis, including the ROC curves for statistically significant comparisons between groups and respective plots. (A).Comparison of Apo-A4 protein abundance between groups measured by ELISA; (B) ROC curves depicting the classification power of Apo-A4; (C).Comparison of CHI3L1 protein abundance between groups measured by ELISA; (D).ROC curves depicting the classification power of CHI3L1. Data is presented in the plots as median ± SD, with asterisk determining the statistical significance: *p-value < 0.05, ** p-value < 0.01, ***p-value < 0.001, and ****p-value ≤ 0.0001.