Quantitative tissue proteomics of esophageal squamous cell carcinoma for novel biomarker discovery

Abstract

Esophageal squamous cell carcinoma (ESCC) is among the top ten most frequent malignancies worldwide. In this study, our objective was to identify potential biomarkers for ESCC through a quantitative proteomic approach using the isobaric tags for relative and absolute quantitation (iTRAQ) approach. We compared the protein expression profiles of ESCC tumor tissues with the corresponding adjacent normal tissue from ten patients. LC-MS/MS analysis of strong cation exchange chromatography fractions was carried out on an Accurate Mass QTOF mass spectrometer, which led to the identification of 687 proteins. In all, 257 proteins were identified as differentially expressed in ESCC as compared to normal. We found several previously known protein biomarkers to be upregulated in ESCC including thrombospondin 1 (THBS1), periostin 1 (POSTN) and heat shock 70 kDa protein 9 (HSPA9) confirming the validity of our approach. In addition, several novel proteins that had not been reported previously were identified in our screen. These novel biomarker candidates included prosaposin (PSAP), plectin 1 (PLEC1) and protein disulfide isomerase A 4 (PDIA4) that were further validated to be overexpressed by immunohistochemical labeling using tissue microarrays. The success of our study shows that this mass spectrometric strategy can be applied to cancers in general to develop a panel of candidate biomarkers, which can then be validated by other techniques.

Figures

Figure 1

Work flow for quantitative tissue proteomics using iTRAQ labeling and validation of biomarkers for esophageal squamous cell carcinoma. For iTRAQ labeling, proteins were isolated from ten tumor and adjacent normal tissues. Proteins were subjected to trypsin digestion followed by iTRAQ labeling of peptides. Post labeling of peptides the tumor and adjacent normal derived peptide mixture was pooled and fractionated using strong cation exchange (SCX) chromatography, followed by liquid chromatography tandem mass spectrometry (LC-MS/MS) on a QTOF mass spectrometer. The data was searched using Mascot and Spectrum Mill search engines. Some of the overexpressed proteins that were not previously described (e.g., PLEC1) were validated using IHC labeling using tissue microarrays.

Figure 2

Classification of proteins by gene ontology based on their cellular localization and biological process. (A) Distribution of proteins based on their cellular component using gene ontology classifier. (B) Distribution of proteins based on their biological process using gene ontology classifier.

Figure 3

MS and MS/MS spectra of known and novel upregulated proteins in ESCC tissues as compared with the adjacent normal epithelia. MS and MS/MS spectra of peptide from representative differentially expressed proteins identified in this study. (A) Periostin (POSTN); (B) Plectin 1 (PLEC1); (C) Prosaposin (PSAP); and (D) Protein disulfide isomerase A4 (PDIA4).

Figure 4

Validation of Plectin 1 using immunohistochemical labeling. Representative sections from tissue microarrays stained with anti-Plectin is shown (A) expression of Plectin 1 in representative normal esophageal squamous mucosa and (B) expression of plectin 1 in ESCC.

Figure 5

Validation of Prosaposin using immunohistochemical labeling. Representative sections from tissue microarrays stained with anti-prosaposin is shown (A) expression of prosaposin in representative normal esophageal squamous mucosa and (B) expression of prosaposin in ESCC.

Figure 6

Validation of protein disulfide isomerase A4 using immunohistochemical labeling. Representative sections from tissue microarrays stained with anti-protein disulfide isomerase A4 as shown (A) expression of protein disulfide isomerase A4 in representative normal esophageal squamous mucosa and (B) expression of protein disulfide isomerase A4 in ESCC.