iTRAQ-Based Proteomics Reveals Novel Biomarkers for Idiopathic Pulmonary Fibrosis

Abstract

Idiopathic pulmonary fibrosis (IPF) is a gradual lung disease with a survival of less than 5 years post-diagnosis for most patients. Poor molecular description of IPF has led to unsatisfactory interpretation of the pathogenesis of this disease, resulting in the lack of successful treatments. The objective of this study was to discover novel noninvasive biomarkers for the diagnosis of IPF. We employed a coupled isobaric tag for relative and absolute quantitation (iTRAQ)-liquid chromatography-tandem mass spectrometry (LC-MS/MS) approach to examine protein expression in patients with IPF. A total of 97 differentially expressed proteins (38 upregulated proteins and 59 downregulated proteins) were identified in the serum of IPF patients. Using String software, a regulatory network containing 87 nodes and 244 edges was built, and the functional enrichment showed that differentially expressed proteins were predominantly involved in protein activation cascade, regulation of response to wounding and extracellular components. A set of three most significantly upregulated proteins (HBB, CRP and SERPINA1) and four most significantly downregulated proteins (APOA2, AHSG, KNG1 and AMBP) were selected for validation in an independent cohort of IPF and other lung diseases using ELISA test. The results confirmed the iTRAQ profiling results and AHSG, AMBP, CRP and KNG1 were found as specific IPF biomarkers. ROC analysis indicated the diagnosis potential of the validated biomarkers. The findings of this study will contribute in understanding the pathogenesis of IPF and facilitate the development of therapeutic targets.

Fig 1. Volcano plot showing log2 fold change plotted against log10 adjusted P value for IPF samples versus healthy control samples.

Data points in the upper right (ratio > 1.5) and upper left (ratio < 0.67) sections with P>0.05 represent proteins that are significantly dysregulated in IPF patients according to the Protein Pilot analysis of the six-plex iTRAQ-labelled serum samples (3 groups of healthy individuals labelled 113, 115 and 117 and 3 groups of IPF patients labelled 114, 116 and 119). The Volcano plot was generated using GraphPad Prism software version 6.01 for windows.

Fig 2. Protein-protein interaction regulatory network of proteins differentially expressed between IPF and healthy controls.

Differentially expressed proteins were combined for building a regulatory network using String software. The network contained 87 nodes and 243 edges with an average node degree of 5.01 and a clustering coefficient of 0.655. The PPI enrichment p-value was equal to 0.

Fig 3. GO enrichment of differentially expressed proteins.

The functional enrichment of proteins in the constructed interaction network was carried out online in the STRING database. Only the 10 most significantly enriched GO terms in each GO category (Biological Process, Cellular Component and Molecular function) with their p-values were presented. A GO term was considered significant at p-value < 0.05.

Fig 4. Box plots of levels of proteins selected for the validation experiment.

Levels of these selected biomarkers were determined in serum of healthy controls (n = 20), sarcoidosis (n = 20), hypersensitivity pneumonitis (HP) (n = 20), and IPF patients (n = 20) using ELISA, n, number of subjects. P-values were calculated with one-way ANOVA test. *p<0.05, **p<0.01, ***p<0.001 and ****p<0.0001 compared to IPF group.

Fig 5. ROC curve analysis of validated differentially expressed proteins.

Accuracy of selected candidate biomarkers in discriminating IPF from healthy controls, sarcoidosis and HP was evaluated using receiver operating characteristic (ROC) curves. The area under the ROC curve AUC > 0.7 was considered in deciding if a given biomarker was informative in discriminating compared groups from each other.