Proteomic Differences and Linkages between Chemoresistance and Metastasis of Pancreatic Cancer Using Knowledge-Based Pathway Analysis

Introduction

Pancreatic cancer (PC) has been recognized as one of the most life-threatening diseases because of the lack of diagnostic methods in the early stage as well as its rapid progression . Among the well-known key barriers for the effective cure of PC are chemoresistance and metastasis , which complicate therapeutic strategies. A thorough understanding of the specific cellular and molecular mechanisms of PC development and progression is required for early detection strategies and effective therapy .

Aggressive growth behavior makes PC resistant to chemotherapy, radiotherapy, and immunotherapy . Gemcitabine (GEM) has been recognized as a primary chemotherapeutic agent, which may increase relative survival rates , although it showed limited improvement. The poor prognosis of PC is derived from the unpredictable and uncontrollable metastatic pattern, which comes along with the development of chemoresistance. Recently, some transcription factors have been shown to play a key role in epithelial-to-mesenchymal transition (EMT). One of these transcription factors, Snail, has been reported to be one of the markers related to resistance against chemotherapy . Metastasis is a multistep process including the loss of cell-to-cell adhesion, which promotes cell motility and migration–invasion into surrounding tissues, as well as transport through the blood stream. Chemoresistance has been considered to be an integrated process with metastatic progression, although the identified markers are not sufficient so far to hypothesize the interconnection between the two biological networks.

This chapter introduces knowledgebase pathway analysis. This analysis has been performed using comparative data sets generated by the subcellular proteomics of matched pairs of PC cell lines by phenotypic and genotypic grouping. Specifically, Su8686 and BXPC-3 were utilized as representatives for GEM-sensitive cell lines, MiaPaCa-2 and Panc-1 for GEM-resistant cell lines, BXPC-3 and Capan-2 for primary cell lines, and Su8686 and Capan-1 for metastatic cell lines. Identified proteins with upregulation from the GEM-resistant and metastasis group compared with the baseline (GEM sensitive or primary) were processed using MetaCore™ (Thomson Reuters, NY, USA) analysis, which provided biological information through the generation of pathway maps with high data relevancy. During the course of analysis, lists of proteins with signal-to-noise (STN) were uploaded into the web-based application, which clustered relevant proteins into specific biological networks. MetaCore™ analysis provided the most relevant biological processes, from which plausible linkages between development of chemoresistance and metastatic progression were identified. This data was further supported using gene knockdown and GEM treatment experiments. The knowledgebase pathway analysis has demonstrated that it may provide useful systemic information regarding correlation of protein data sets in a high-throughput data mining approach that generates disease marker candidates for therapeutic targeting with a minimal time investment.

Proteomic Analysis at Subcellular Level

A brief work flow for the proteomic analysis followed by pathway analysis is introduced in Figure 10.1 A . Proteomic analysis was performed using six cell lines—namely, Panc-1, BXPC-3, MiaPaCa-2, Su8686, Capan-1, and Capan-2, for which phenotypic and genotypic characteristics are depicted in Figure 10.1 B.

Capan-1, Su8686, and MiaPaCa-2 cells are classified as metastatic cell lines generated from PC metastatic lesions of the patients while BXPC-3, Capan-2, Panc-1, and MiaPaCa-2 as primary cell lines, derived from primary pancreatic tumor tissue . The cell lines have been well known to show different sensitivity to chemotherapeutics, such as GEM . The six cell lines were classified as either drug sensitive or resistant by their sensitivity to GEM treatment as demonstrated by viability assays. Capan-1, Capan-2, Su8686, and BXPC-3 cells are classified as chemosensitive, while Panc-1 and MiaPaCa-2 cell lines are considered chemoresistant.

Cells were obtained from the American Type Culture Collection (ATCC; Manassas, VA, USA), and cultured in ATCC-recommended media with 10% fetal bovine serum. Cells were maintained at 37 °C under humidified 5% CO ₂ and grown to 80% confluence in culture dishes and used for the experiments. Trypsinized cells were washed with phosphate-buffered saline (PBS) three times, and then subcellular fractionation was performed using the Proteo Extract Subcellular Proteome Extraction Kit (Calbiochem, CA, USA) according to the manufacturer’s protocol. Four fractions were generated: fraction 1 for cytosolic proteins, fraction 2 for membrane proteins, fraction 3 for nuclear proteins, and fraction 4 for cytoskeletal proteins. Then, 30 μg of denatured protein from the subcellular fractions from each cell line were separated on sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) followed by in-gel digestion by the method reported previously. Peptides were extracted from the gel matrix by adding 100 μL 50% (v/v) acetonitrile (ACN) containing 5% (v/v) formic acid and incubated at room temperature for 30 min three times. The extracts were dried under vacuum and then were suspended in 10% (v/v) ACN containing 3% (v/v) formic acid to be subjected to liquid chromatography with tandem mass spectrometry (LC MS/MS).

Peptide samples were separated on a Nano-Acquity ultra performance liquid chromatography system with a Nano-Acquity C ₁₈ trap column (5 μm, 180 μm × 20 mm) and a Nano-Acquity BEH130 C ₁₈ analytical column (1.7 μm, 75 μm × 150 mm) (Waters Corporation, Milford, MA, USA) and then analyzed by a high-resolution linear ion trap mass spectrometer (Linear Trap Quadrupole/Orbitrap-XL, Thermo Finnigan, San Jose, CA, USA) equipped with a nano-scale electrospray source (Thermo Finnigan, San Jose, CA, USA). A linear gradient system that consisted of binary mobile phases (A, water with 0.1% formic acid; B, ACN with 0.1% formic acid) flowing from 5% mobile phase B to 45% mobile phase B for 75 min at the flow rate of 0.35 μL/min was employed for separation of peptides. Tandem MS scan was conducted by a data dependent scan for the top-10 intense ions acquired from each full MS scan using the dynamic exclusion option.

Protein Identification and Data Compiling

MS spectra were searched in the human International Protein Index (IPI) database v3.72 FASTA database (86,392 entries) using the SEQUEST search algorithm (SRF v.5) of the Bioworks software v3.3.1sp1 (Thermo Fisher Scientific, San Jose, CA, USA) with the following parameters: parent mass tolerance of 10 ppm, fragment tolerance of 0.5 Da (monoisotopic), variable modification on methionine of 16 Da (oxidation), and maximum missed cleavage of two sites assuming the digestion enzyme trypsin. Data were compiled with Scaffold software (v3_06_03, Proteome Software, Portland, OR, USA) for comparison of spectral counts with filtering criteria of two peptides minimum; XCorr scores of greater than 1.9, 2.3, and 3.4 for singly, doubly, and triply charged peptides; and deltaCn scores of greater than 0.10.