Quantitative tissue proteome profile reveals neutrophil degranulation and remodeling of extracellular matrix proteins in early stage gallbladder cancer

Abstract

Gallbladder cancer (GBC) is an aggressive malignancy of the gastrointestinal tract with a poor prognosis. It is important to understand the molecular processes associated with the pathogenesis of early stage GBC and identify proteins useful for diagnostic and therapeutic strategies. Here, we have carried out an iTRAQ-based quantitative proteomic analysis of tumor tissues from early stage GBC cases (stage I, n=7 and stage II, n=5) and non-tumor controls (n=6) from gallstone disease (GSD). We identified 357 differentially expressed proteins (DEPs) based on ≥ 2 unique peptides and ≥ 2 fold change with p value < 0.05. Pathway analysis using the STRING database showed, 'neutrophil degranulation' to be the major upregulated pathway that includes proteins such as MPO, PRTN3, S100A8, MMP9, DEFA1, AZU, and 'ECM organization' to be the major downregulated pathway that includes proteins such as COL14A1, COL1A2, COL6A1, COL6A2, COL6A3, BGN, DCN. Western blot and/or IHC analysis confirmed the elevated expression of MPO, PRTN3 and S100A8 in early stage of the disease. Based on the above results, we hypothesize that there is an increased neutrophil infiltration in tumor tissue and neutrophil degranulation leading to degradation of extracellular matrix (ECM) proteins promoting cancer cell invasion in the early stage GBC. Some of the proteins (MPO, MMP9, DEFA1) associated with 'neutrophil degranulation' showed the presence of 'signal sequence' suggesting their potential as circulatory markers for early detection of GBC. Overall, the study presents a protein dataset associated with early stage GBC.

Keywords: early stage; gallbladder cancer; iTRAQ; neutrophil degranulation; tissue proteomics.

Figure 1

Overall workflow of the study. GSD, Gallstone disease; GBC, Gallbladder cancer; DEPs, Differentially expressed proteins.

Figure 2

PCA Plot showing the correlation of the individual patients with GBC stage I and II. (A) includes seven individual samples from GBC stage I and one pooled GSD control while (B) includes five individual samples from GBC stage II along with one pooled GSD control. Four patients, two from stage I (GBC-1 and 6) (A) and two from stage II (GBC-8 and 9) (B) showed similar profile as GSD (non-tumor control). The technical replicates showed a significant correlation. Replicates R1, R2 and R3 are shown in red, green and blue color. The PCA plot is derived using the iTRAQ reporter intensity from the quantitative proteomics data.

Figure 3

Venn diagram showing DEPs in early stage GBC. A total of 83 proteins are common to both stage I and II, while 101 proteins are specific to stage I and 173 proteins are specific to stage II. The details of all the proteins are shown in Supplementary Table S4 .

Figure 4

Gene ontology of 357 DEPs in early stage GBC. (A) Localization of (B) Molecular functions (C) Reactome pathways using upregulated proteins and (D) downregulated proteins as observed using STRING database.

Figure 5

Protein-protein-interaction (PPI) network of 29 deregulated proteins. PPI analysis showed four clusters including majorly the proteins associated with neutrophil degranulation (MPO, DEFA1, S100A8, PRTN3, AOC3) (marked in red), ECM proteins (COL6A1, BGN, DCN, LUM, PRELP) (green), cytoskeletal or intermediate filament (DES, MYL6, MYL9, TPM2) (blue). The subset of 29 proteins showed differential expression in ≥ 50% of early stage GBC (i.e. ≥ 6 patients).

Figure 6

Hierarchical clustering using non-redundant list of 308 DE proteins in 12 early stage GBC patients. (A) Hierarchical clustering showed two clusters with cluster A majorly including stage I samples and cluster B majorly including stage II samples. (B) We observed cytokeratins KRT7, KRT8, KRT18 and KRT19 showing upregulation in Cluster A and downregulation in Cluster B. Log2 (fold change) values for 308 proteins were used for the analysis. Red- Upregulated, Green- Downregulated.

Figure 7

Altered levels of functionally relevant proteins in early stage GBC as observed in quantitative proteomics data. The plot showing the levels of MPO, PRTN3 and S100A8 in individual patients, GBC stage-I (n=7) and stage-II (n=5).

Figure 8

Western blot images showing expression of MPO, PRTN3, S100A8 in the individual tissue samples from early stage GBC and GSD cases. A significant overexpression of MPO, PRTN3, S100A8 was found in 66.7% (n=8/12), 66.7% (n=8/12) and 83.3% (10/12) early stage GBC cases respectively.

Figure 9

IHC analysis to study the expression of MPO and S100A8 in controls and GBC cases. (A) Representative IHC images showing the expression of MPO and S100A8 in controls and GBC cases. IHC was performed on formalin-fixed paraffin-embedded (FFPE) individual tissue sections of 10 controls (GSD cases with no dysplasia), 10 early stage GBC (stage I and II) cases and 10 advanced stage GBC cases (stage III and IV). The IHC results showed that the number of MPO positive neutrophils was found to be ‘positive’ in 50% of early stage GBC and 30% of advanced stage GBC cases. All GSD cases showed ‘negative’ expression. The expression of S100A8 was found to be ‘positive’ in 10% GSD cases, 60% early and 50% advanced stage GBC. (B) The statistical analysis between cases and controls showed a significant difference of MPO positive neutrophils in early stage GBC vs controls and all GBC vs controls while a significant difference of S100A8 was observed in all GBC vs controls. The controls (≥ 90%) showed ‘Negative’ expression levels.

Figure 10

Hypothesis showing molecular events in early stage GBC. We observed overexpression of neutrophil degranulation pathway proteins and downregulation of ECM proteins. We hypothesize that there is neutrophil infiltration and degranulation in GBC tissue resulting in release of proteases which possibly degrades ECM proteins promoting cancer cell invasion.