Quantitative Comparative Proteomics Reveal Biomarkers for Dengue Disease Severity

Abstract

Dengue fever (DF) could develop into dengue haemorrhagic fever (DHF) with increased mortality rate. Since the clinical characteristics and pathogen are same in DF and DHF. It's important to identify different molecular biomarkers to predict DHF patients from DF. We conducted a clinical plasma proteomics study using quantification (TMT)-based quantitative proteomics methodology to found the differential expressed protein in DF patients before they developed into DHF. In total 441 proteins were identified up or down regulated. There proteins are enriched in diverse biological processes such as proteasome pathway, Alanine, aspartate, and glutamate metabolism and arginine biosynthesis. Several proteins such as PLAT, LAMB2, and F9 were upregulated in only DF patients which developed into DHF cases, not in DF, compared with healthy-control. In another way, FGL1, MFAP4, GLUL, and VCAM1 were upregulated in both DHF and DF cases compare with healthy-control. RT-PCR and ELISA were used to validate these upregulated gene expression and protein level in 54 individuals. Results displayed the same pattern as proteomics analysis. All including PLAT, LAMB2, F9, VCAM1, FGL1, MFAP4, and GLUL could be considered as potential markers of predicting DHF since the levels of these proteins vary between DF and DHF. These new founding identified potential molecular biomarkers for future development in precision prediction of DHF in DF patients.

Keywords: TMT; biomarker; dengue fever; dengue hemorrhagic fever; proteomomics.

FIGURE 1

Workflow of TMT quantitatively proteomics analysis. A total of 54 serum samples from dengue haemorrhagic fever (DHF), Dengue fever (DF), and healthy groups were collected, numbered A, B, and C, with 18 in each group. After protein extraction, the serum proteins of every six people in the group were mixed into a mixed pool for TMT labeling. The order of labeling is: A1-126, A2-127N, A3-127C, B1-128C, B2-129N, B3-129C, C1-130N, C2-130C, C3-131. When the labeling was completed, all the proteins were mixed and rotated to dry. The 1D LC separation – high-pH reverse-phase separation was performed after resolving. The separated components are combined and evaporated to dryness and then re-dissolved. LC-MS/MS analysis was performed after that.

FIGURE 2

Differential expression proteins in DHF, DF compared with health person. (A) Clustering of proteomics data based on differential proteins. Note that the data in Group B show differences. B2 was clustered with A1, A2, and A3, while B1, C1, C2, and C3 were clustered together. (B–D) Volcano map of A/B, B/C, and A/C. (E,F) Venn diagram of down-regulated and up-regulated proteins.

FIGURE 3

KEGG pathway analysis of the differentially expressed proteins of A/C. (A) Up-regulated proteins of A/C are enriched in diverse KEGG pathways. (B) The most significantly enriched pathways corresponding to the down-regulated proteins of A/C.

FIGURE 4

KEGG pathway analysis of the differentially expressed proteins of B/C. (A) Up-regulated proteins of B/C are enriched in diverse KEGG pathways. (B) The most significantly enriched pathways corresponding to the down-regulated proteins of B/C.

FIGURE 5

Validation of differentially expressed proteins. (A) western blot analysis of iTRAQ samples using GADPH as internal reference. (B) RT-qPCR analysis of genes in individual patients’ serum, DF, and DHF samples were collected when no severe manifestations detected, after detection, the DHF group has reduced platelets less than 50 × 10⁹/L. (C) Fold change of protein levels using ELISA analysis using the individual sample in RT-qPCR assay. The sample size were 47, 41, and 22, respectively. ^∗p < 0.05. The data are shown as the mean values ± SD.