Elsevier

Journal of Proteomics

Volume 136, 16 March 2016, Pages 1-12
Journal of Proteomics

Secretome protein signature of human pancreatic cancer stem-like cells

https://doi.org/10.1016/j.jprot.2016.01.017Get rights and content

Highlights

  • iTRAQ proteomic analysis of secretome of pancreatic cancer stem cells (Panc1 CSCs)

  • Identification of 72 secreted proteins of Panc1 CSCs

  • Panc1 CSCs secretome involved in cancer differentiation, invasion and metastasis

  • Identification of pool of proteins from which cancer biomarkers can be identified

  • Ceruloplasmin as promising marker for patients negative for CA19-9

Abstract

Emerging research has demonstrated that pancreatic ductal adenocarcinoma (PDAC) contains a sub-population of cancer stem cells (CSCs) characterized by self-renewal, anchorage-independent-growth, long-term proliferation and chemoresistance. The secretome analysis of pancreatic CSCs has not yet been performed, although it may provide insight into tumour/microenvironment interactions and intracellular processes, as well as to identify potential biomarkers.

To characterize the secreted proteins of pancreatic CSCs, we performed an iTRAQ-based proteomic analysis to compare the secretomes of Panc1 cancer stem-like cells (Panc1 CSCs) and parental cell line. A total of 72 proteins were found up-/down-regulated in the conditioned medium of Panc1 CSCs. The pathway analysis revealed modulation of vital physiological pathways including glycolysis, gluconeogenesis and pentose phosphate.

Through ELISA immunoassays we analysed the presence of the three proteins most highly secreted by Panc1 CSCs (ceruloplasmin, galectin-3, and MARCKS) in sera of PDAC patient. ROC curve analysis suggests ceruloplasmin as promising marker for patients negative for CA19-9.

Overall, our study provides a systemic secretome analysis of pancreatic CSCs revealing a number of secreted proteins which participate in pathological conditions including cancer differentiation, invasion and metastasis. They may serve as a valuable pool of proteins from which biomarkers and therapeutic targets can be identified.

Biological significance

The secretome of CSCs is a rich reservoir of biomarkers of cancer progression and molecular therapeutic targets, and thus is a topic of great interest for cancer research. The secretome analysis of pancreatic CSCs has not yet been performed.

Recently, our group has demonstrated that Panc-1 CSCs isolated from parental cell line by using the CSC selective medium, represent a model of great importance to deepen the understanding of the biology of pancreatic adenocarcinoma.

To our knowledge, this is the first proteomic study of pancreatic CSC secretome. We performed an iTRAQ-based analysis to compare the secretomes of Panc1 CSCs and Panc1 parental cell line and identified a total of 43 proteins secreted at higher level by pancreatic cancer stem cells. We found modulation of different vital physiological pathways (such as glycolysis and gluconeogenesis, pentose phosphate pathway) and the involvement of CSC secreted proteins (for example 72 kDa type IV collagenase, galectin-3, alpha-actinin-4, and MARCKS) in pathological conditions including cancer differentiation, invasion and metastasis.

By ELISA verification we found that MARCKS and ceruloplasmin discriminate between controls and PDAC patients; in addition ROC curve analyses indicate that MARCKS does not have diagnostic accuracy, while ceruloplasmin could be a promising marker only for patients negative for CA19-9.

We think that the findings reported in our manuscript advance the understanding of the pathways implicated in tumourigenesis, metastasis and chemoresistance of pancreatic cancer, and also identify a pool of proteins from which novel candidate diagnostic and therapeutic biomarkers could be discovered.

Introduction

Pancreatic ductal adenocarcinoma (PDAC) is one of the most aggressive and devastating human malignancies with a death-to-incidence ratio of 0.99. Most patients have metastatic disease at the time of diagnosis. More than 75% of patients who undergo surgical resection of small pancreatic tumours with clear surgical margins and no evidence of metastasis nonetheless die from metastasis within 5 years [1], a finding that is consistent with early spread. In addition to late diagnosis, high resistance to chemotherapy and radiation seems to be responsible for the dismal outcome of PDAC. Recent studies have demonstrated that in a mouse model of PDAC cellular dissemination leading to metastasis occurs prior to the formation of an identifiable primary tumour [2]. This behaviour is associated with epithelial-to-mesenchymal transition (EMT) and with the establishment of circulating pancreatic cells which maintain a mesenchymal phenotype and express typical markers of cancer stem cells (CSCs) [3]. Evidence for the existence of CSCs has also been provided in primary human pancreatic adenocarcinomas grown in immunocompromised mice [4]. At present, PDAC CSCs may be considered a subpopulation of cells in the bulk of the tumour characterised by the exclusive ability to drive tumourigenesis and metastasis and to play a fundamental role in disease relapse. Hence, to substantially impact long-term survival of PDAC patients, the study of the biological features and of the secretome of PDAC CSCs is critical, and will inform the development of more efficient therapies and the identification of early biomarkers.

The high heterogeneity of CSCs, which originates from genotypic and phenotypic plasticity, and their low presence in cancer sample tissues make their isolation and correct identification extremely difficult, strongly limiting the realization of biochemical studies. The current approach to isolate CSCs from tissue samples is mainly based on the difference in cell size or on the expression of specific antigens. However, these methods do not permit the recovery of sufficient cells to perform proteome or secretome studies. In order to obtain valid and reproducible results, the biochemical approach to CSC pathophysiology can take advantage of the observation that CSCs can be isolated and enriched from several human cancer cell lines [5]. Recently, our group has been able to isolate cancer stem-like cells from five out of nine PDAC cell lines [6]. In particular, we have demonstrated that Panc1 cancer stem-like cells (Panc1 CSCs) isolated from parental (P) cell line by using the CSC selective medium, represent a model of great importance to deepen the understanding of the biology of pancreatic adenocarcinoma. Panc1 CSCs showed the highest tumorsphere-forming ability, were more resistant to the action of the anticancer drugs, had typical surface stem cell markers, and when subcutaneously injected into nude female mice were more tumorigenic than parental cells. Thus far, proteomic approaches have been applied to pancreatic CSCs isolated from xenografted tumours in mice [7], early stage tumours [8], or established cell lines [9]. However, secretome analysis of pancreatic CSCs has not yet been reported, although the secreted proteins may serve as a valuable tool to obtain insight into interaction of the tumour with its microenvironment as well as intracellular processes, taking into account the observation that many tumour cells shed intracellular and even nuclear proteins into the extracellular space. Furthermore, these studies may allow the identification of potential PDAC biomarkers. Contrarily to PDAC, secretome approaches have been used to investigate CSCs of colon [10] and prostate cancer [11]. In general, secretome studies on cell lines grown in culture medium are limited by contamination from intracellular proteins originating from spontaneous cell autolysis. For this reason, a filtering criterion [12] must be established to select bona fide secreted proteins while avoiding the contaminants for downstream validation works. In this study, we have adopted a shotgun proteomics approach using iTRAQ 8-plex coupled with 2D-LC–MS/MS to compare the secretome of Panc1 pancreatic adenocarcinoma cell line with that of their derived stem-like cells (Panc1 CSCs). In order to identify only secreted proteins, we have compared the protein expression levels of the conditioned medium (CM) with those of the whole cell lysate, taking into account that the relative abundance of secreted proteins should be higher in CM than in cell lysate. Following this approach, we have identified 43 proteins secreted at higher level by Panc1 CSCs relative to the parental cells. In silico functional pathway analysis has demonstrated a predominant association of these proteins to glycolysis, gluconeogenesis, IGF-1 signalling, atherosclerosis signalling, pyruvate fermentation to lactate, and pentose phosphate pathway. Among the identified proteins, ceruloplasmin was the most abundant detected in CM from Panc1 CSCs and showed promise as predictors for PDAC, particularly for patients negative for CA19-9.

To our knowledge, this is the first proteomic study of pancreatic CSCs secretome. Our findings advance the understanding of the pathways implicated in tumourigenesis, metastasis and chemoresistance of pancreatic cancer, and also identify a pool of proteins from which novel candidate diagnostic and therapeutic biomarkers could be discovered.

Section snippets

Cell culture

The human PDAC cell line Panc1 was grown in RPMI 1640 supplemented with 10% FBS, 2 mM glutamine, and 50 μg/ml gentamicin sulfate (Gibco, Life Technologies). Adherent cells were maintained in standard conditions for a few passages at 37 °C with 5% CO2. Panc1 CSCs were obtained as previously described [6]. Briefly, adherent cells were cultured in CSC medium (i.e. DMEM/F-12 supplemented with glucose, B27, fungizone, penicillin/streptomycin, heparin, epidermal growth factor and fibroblast growth

Protein identification and quantification in Panc1 cell and Panc1 CSC conditioned media using iTRAQ

The iTRAQ-labeled CM protein samples of Panc1 cells and Panc1 CSCs were analyzed together with their respective whole cell lysates as shown in Fig. 1. A total of 2045 proteins with at least 93.8% confidence and an Unused ProteinPilot scores > 1.09 (equating to a global FDR of 1%) were identified, among these a total of 1157 proteins were quantified with a peptide confidence cut-off of 95% (Supplemental Table 3). Of these, 608 were identified via a single peptide with a confidence of 99% (

Discussion

The ability to identify and isolate CSCs in various tumour models has led to the possibility to study the mechanisms by which CSCs can contribute to tumour initiation as well as continued tumour progression. Up to now, a deep comprehension of CSC biology, and in particular that of pancreatic CSCs, is lacking. In the present study, a comparative secretome approach was taken to dissect the differences in protein level between pancreatic cancer and pancreatic cancer stem cells. Secreted proteins

Conflict of interest

The authors declare no conflicts of interest.

Competing financial interests

The authors declare no competing financial interests.

Acknowledgements

We thank Dr. Salvagno Gianluca for the technical assistance with the CA19-9 ELISA assays. This work was supported by AIRC-Fondazione CariPaRo, Padova, Italy; AIRC 5 per mille grant n. 12182, Italy and the NIHR Liverpool Pancreas Biomedical Research Unit. JB received a travel Fellowship from the European Pancreatic Club.

References (65)

  • J. Lu et al.

    14-3-3zeta Cooperates with ErbB2 to promote ductal carcinoma in situ progression to invasive breast cancer by inducing epithelial–mesenchymal transition

    Cancer Cell

    (2009)
  • T. Ohmori et al.

    Vinculin is indispensable for repopulation by hematopoietic stem cells, independent of integrin function

    J. Biol. Chem.

    (2010)
  • R. Barderas et al.

    In-depth characterization of the secretome of colorectal cancer metastatic cells identifies key proteins in cell adhesion, migration, and invasion

    Mol. Cell. Proteomics

    (2013)
  • X. Chen et al.

    PhosphoMARCKS drives motility of mouse melanoma cells

    Cell. Signal.

    (2010)
  • J. Gardner-Thorpe et al.

    Differential display of expressed genes in pancreatic cancer cells

    Biochem. Biophys. Res. Commun.

    (2002)
  • M.W. Saif

    Pancreatic neoplasm in 2011: an update

    JOP

    (2011)
  • C.J. Lee et al.

    Pancreatic cancer stem cells

    J. Clin. Oncol.

    (2008)
  • C. Li et al.

    Identification of pancreatic cancer stem cells

    Cancer Res.

    (2007)
  • T. Kondo

    Stem cell-like cancer cells in cancer cell lines

    Cancer Biomark.

    (2007)
  • E. Dalla Pozza et al.

    Pancreatic ductal adenocarcinoma cell lines display a plastic ability to bidirectionally convert into cancer stem cells

    Int. J. Oncol.

    (2015)
  • L. Dai et al.

    Quantitative proteomic profiling studies of pancreatic cancer stem cells

    J. Proteome Res.

    (2010)
  • J. Zhu et al.

    Target proteomic profiling of frozen pancreatic CD24 + adenocarcinoma tissues by immuno-laser capture microdissection and nano-LC–MS/MS

    J. Proteome Res.

    (2013)
  • J. Zhu et al.

    Identification of glycoprotein markers for pancreatic cancer CD24 + CD44 + stem-like cells using nano-LC–MS/MS and tissue microarray

    J. Proteome Res.

    (2012)
  • F. Mateo et al.

    SPARC mediates metastatic cooperation between CSC and non-CSC prostate cancer cell subpopulations

    Mol. Cancer

    (2014)
  • H. Loei et al.

    Mining the gastric cancer secretome: identification of GRN as a potential diagnostic marker for early gastric cancer

    J. Proteome Res.

    (2012)
  • W.H. Tang et al.

    Nonlinear fitting method for determining local false discovery rates from decoy database searches

    J. Proteome Res.

    (2008)
  • J.D. Bendtsen et al.

    Feature-based prediction of non-classical and leaderless protein secretion

    Protein Eng. Des. Sel.

    (2004)
  • R.J. Simpson et al.

    ExoCarta as a resource for exosomal research

    J. Extracellular Vesicles

    (2012)
  • H. Kalra et al.

    Vesiclepedia: a compendium for extracellular vesicles with continuous community annotation

    PLoS Biol.

    (2012)
  • M. Quan et al.

    The roles of FOXM1 in pancreatic stem cells and carcinogenesis

    Mol. Cancer

    (2013)
  • M. Capello et al.

    alpha-Enolase: a promising therapeutic and diagnostic tumor target

    FEBS J.

    (2011)
  • E. Perales-Clemente et al.

    Metabolic regulation of redox status in stem cells

    Antioxid. Redox Signal.

    (2014)
  • Cited by (52)

    • Advances and applications of stable isotope labeling-based methods for proteome relative quantitation

      2020, TrAC - Trends in Analytical Chemistry
      Citation Excerpt :

      To distinguish the secreted/EV proteins from interfering proteins in the culture medium, stable isotope metabolic labeling-based methods have the unique advantage of labeling the cell-derived proteins and quantifying them accurately. To distinguish between secreted proteins and apoptosis-induced intracellular proteins, Brandi et al. labeled intracellular proteins and proteins in cell culture medium respectively using iTRAQ reagents and calculated their ratio [123]. Because the abundance of secreted protein in cell culture medium should be higher than that in cell cytoplasm, the secreted proteins were confirmed with the ratio of cell culture medium/cytoplasm>1, or >1.5.

    View all citing articles on Scopus
    View full text