Identification of pancreatic cancer invasion-related proteins by proteomic analysis

Abstract

Background: Markers of pancreatic cancer invasion were investigated in two clonal populations of the cell line, MiaPaCa-2, Clone #3 (high invasion) and Clone #8 (low invasion) using proteomic profiling of an in vitro model of pancreatic cancer.

Materials and methods: Using 2D-DIGE followed by MALDI-TOF MS, two clonal sub-populations of the pancreatic cancer cell line, MiaPaCa-2 with high and low invasive capacities were incubated on matrigel 24 hours prior to analysis to stimulate cell-ECM contact and mimic in vivo interaction with the basement membrane.

Results: Sixty proteins were identified as being differentially expressed (> 1.2 fold change and p < or = 0.05) between Clone #3 and Clone #8. Proteins found to have higher abundance levels in the highly invasive Clone #3 compared to the low invasive Clone #8 include members of the chaperone activity proteins and cytoskeleton constituents whereas metabolism-associated and catalytic proteins had lower abundance levels. Differential protein expression levels of ALDH1A1, VIM, STIP1 and KRT18 and GAPDH were confirmed by immunoblot. Using RNAi technology, STIP1 knockdown significantly reduced invasion and proliferation of the highly invasive Clone #3. Knockdown of another target, VIM by siRNA in Clone #3 cells also resulted in decreased invasion abilities of Clone #3. Elevated expression of STIP1 was observed in pancreatic tumour tissue compared to normal pancreas, whereas ALDH1A1 stained at lower levels in pancreatic tumours, as detected by immunohistochemistry.

Conclusion: Identification of targets which play a role in the highly invasive phenotype of pancreatic cancer may help to understand the biological behaviour, the rapid progression of this cancer and may be of importance in the development of new therapeutic strategies for pancreatic cancer.

Figure 1

Graph displays the total number of cells invading after 24 hr incubation on matrigel of Clone #8, MiaPaCa-2 and Clone #3. Experiments performed in triplicate.

Figure 2

Representative 2D DIGE gel image of Cy2-labelled pool of Clone #3 and Clone #8 cell lysate samples. Differentially expressed proteins that have been successfully identified by MALDI-TOF MS (p ≤ 0.05, protein fold ≥ 1.2) are represented on the gel using DeCyder software. Proteins are labelled numerically for visual clarity and are outlined in Table S1.

Figure 3

Term-ranking Gene Ontology categories. Representation of the 10 top-ranked functional categories, using GO terms that are enriched in all significantly differentially expressed proteins between Clone #3 versus Clone #8.

Figure 4

Differentially expressed proteins, A. ALDH1A1, B. VIM, C. STIP1, D. KRT18 and E. GAPDH in Clone #3 versus Clone #8 confirmed by (i) 3D spot image, (ii) PEM spot and (iii) immunoblot.

Figure 5

A. Immunoblot of STIP1 silencing in Clone #3 cells untreated control, scrambled control, STIP1-siRNA (1), STIP1-siRNA (2) and STIP1-siRNA (3) (upper) and α-tubulin as loading control (lower). B. Total number of invading cells after siRNA transfection in Clone #3 untreated control, scrambled control, STIP1-siRNA (1), STIP1-siRNA (2) and STIP1-siRNA (3). Results are displayed as the total number of invading cells, determined by counting the number of cells per field in 10 random fields, at 200× magnification. The average number of cells per field was then multiplied by a factor of 140 (growth area of membrane/field area viewed at 200× magnification (calibrated using a microscope graticule)). Experiments performed in triplicate. Total mean number of cells invading at 200× magnification (n = 3). C. Proliferation assays of Clone #3 untreated control, scrambled control, STIP1-siRNA (1), STIP1-siRNA (2) and STIP1-siRNA (3). Results displayed as percentage survival relative to untreated control. Student's t-test; p = 0.05*, 0.01**, 0.005*** (n = 3).

Figure 6

A. Immunoblot of VIM knockdown in Clone #3 untreated control, scrambled control, VIM-siRNA (1), VIM-siRNA (2) and VIM-siRNA (3) (upper) and α-tubulin as loading control (lower). B. Invasion assay of Clone #3 cells after siRNA silencing of VIM. The total number of invading cells was determined by counting the number of cells per field in 10 random fields, at 200× magnification. The average number of cells per field was then multiplied by a factor of 140 (growth area of membrane/field area viewed at 200× magnification (calibrated using a microscope graticule)). Experiments performed in triplicate. Statistics; * ≤ 0.05, ** ≤ 0.01, *** ≤ 0.005.

Figure 7

IHC detection of STIP1 (A-D) and ALDH1A1 (E-H) in pancreatic cancer and normal pancreas tissues. (A) Strong STIP1 cytoplasmic staining in PC tumour ducts. (B-C) Strong STIP1 expression in poorly differentiated PC tumours. (D) Moderate staining of normal pancreas ducts and acinar cells. (E) ALDH1A1 highly expressed in well differentiated PC tumour. (F-G) Weak ALDH1A1 staining observed in < 10% of poorly differentiated PC tumours. (H) Positive staining in epithelial cells of normal pancreas. Original magnification 200×.