Discovery and validation of novel protein markers in mucosa of portal hypertensive gastropathy

Abstract

Background: Portal hypertension induced esophageal and gastric variceal bleeding is the main cause of death among patients of decompensated liver cirrhosis. Therefore, a standardized, biomarker-based test, to make an early-stage non-invasive risk assessment of portal hypertension, is highly desirable. However, no fit-for-purpose biomarkers have yet been identified.

Methods: We conducted a pilot study consisting of 5 portal hypertensive gastropathy (PHG) patients and 5 normal controls, sampling the gastric mucosa of normal controls and PHG patients before and after endoscopic cyanoacrylate injection, using label-free quantitative (LFQ) mass spectrometry, to identify potential biomarker candidates in gastric mucosa from PHG patients and normal controls. Then we further used parallel reaction monitoring (PRM) to verify the abundance of the targeted protein.

Results: LFQ analyses identified 423 significantly differentially expressed proteins. 17 proteins that significantly elevated in the gastric mucosa of PHG patients were further validated using PRM.

Conclusions: This is the first application of an LFQ-PRM workflow to identify and validate PHG-specific biomarkers in patient gastric mucosa samples. Our findings lay the foundation for comprehending the molecular mechanisms of PHG pathogenesis, and provide potential applications for useful biomarkers in early diagnosis and treatment. Trial registration and ethics approval: Trial registration was completed (ChiCTR2000029840) on February 25, 2020. Ethics Approvals were completed on July 17, 2017 (NYSZYYEC20180003) and February 15, 2020 (NYSZYYEC20200005).

Keywords: Biomarker; Label-free quantitative mass spectrometry; Liver cirrhosis; Parallel reaction monitoring; Portal hypertension gastropathy; Proteomics.

Fig. 1

a Workflow describing LFQ and PRM analysis. Proteins from samples were digested by trypsin and analyzed by LS–MS/MS. The differentially expressed proteins were analyzed by database researches and validated by PRM. Nor. group = healthy individuals, Ph. group = PHG patients and Inter. group = PHG patients after endoscopic cyanoacrylate injection for 1 week. b The gastroscopic photographs show normal appearance of esophagus and fundus, c severe esophageal varices and mild PHG and d esophageal varices alleviation and PHG exacerbation after endoscopic cyanoacrylate injection

Fig. 2

DEPs distribution in different comparative analysis groups between Inter versus Nor, Ph versus Nor and Inter versus Ph. a The DEPs distribution of each group is presented by bar chart. b A Venn diagram showing the distribution of DEPs. c DEPs distribution exhibits by volcano plots, which had fold change > 1.5 or < 1/1.5 and p < 0.05

Fig. 3

Heatmap and cluster analysis of 423 DEPs in Inter versus Nor, Ph versus Nor and Inter versus Ph. Each line in the heatmap represents the mean fold-change of up-regulation (red) or down-regulation (green) of a protein's expression levels

Fig. 4

Gene Ontology (GO) level 2 analysis and subcellular location of DEPs. a Functional classification of DEPs for relevant biological processes, molecular functions, and cellular components. b Subcellular location of DEPs

Fig. 5

KEGG pathway analysis and heatmap [–52] of DEPs in Inter versus Nor, Ph versus Nor and Inter versus Ph

Fig. 6

STRING protein–protein interaction (PPI) network of validated interactions between DEPs enriched in focal adhesion and vascular smooth muscle contraction. They were created using STRING v11.0 software together with a ‘Homo sapiens’ database

Fig. 7

Heatmap and box plot of 17 DEPs in PHG patients before and after endoscopic cyanoacrylate injection compared with controls by PRM analysis. Each line in the heatmap represents the mean fold-change of upregulation (red) or downregulation (green) of a protein's expression levels