ARTICLE pubs.acs.org/jpr
Secretome Compartment Is a Valuable Source of Biomarkers for Cancer-Relevant Pathways Dario Caccia, Laura Zanetti Domingues, Francesca Micciche, Maida De Bortoli, Cristiana Carniti, Piera Mondellini, and Italia Bongarzone* Proteomics Laboratory, Department of Experimental Oncology and Molecular Medicine, Fondazione IRCCS Istituto Nazionale Tumori, Milan, Italy
bS Supporting Information ABSTRACT: In principle, targeted therapies have optimal activity against a specific subset of tumors that depend upon the targeted molecule or pathway for growth, survival, or metastasis. Consequently, it is important in drug development and clinical practice to have predictive biomarkers that can reliably identify patients who will benefit from a given therapy. We analyzed tumor cell-line secretomes (conditioned cell media) to look for predictive biomarkers; secretomes represent a potential source for potential biomarkers that are expressed in intracellular signaling and therefore may reflect changes induced by targeted therapy. Using Gene Ontology, we classified by function the secretome proteins of 12 tumor cell lines of different histotypes. Representations and hierarchical relationships among the functional groups differed among the cell lines. Using bioinformatics tools, we identified proteins involved in intracellular signaling pathways. For example, we found that secretome proteins related to TGF-beta signaling in thyroid cancer cells, such as vasorin, CD109, and βIG-H3 (TGFBI), were sensitive to RPI-1 and dasatinib treatments, which have been previously demonstrated to be effective in blocking cell proliferation. The secretome may be a valuable source of potential biomarkers for detecting cancer and measuring the effectiveness of cancer therapies. KEYWORDS: secretome, predictive biomarkers, signaling pathways, bioinformatics, proteomics, targeted therapy, adhesion pathway, network prediction
’ INTRODUCTION There is a mostly unmet need in cancer management for biomarkers that can be used in early stage diagnosis, therapeutic strategy design, and tumor response monitoring.1 A protein that is secreted by a tumor and can be monitored in blood samples would be an optimal biomarker. Despite an intensive search over the past few decades, only a few identified cancer biomarkers have proven to be clinically useful, for example, PSA, CEA, and CA125, proteins that are usually only present at low levels in the plasma of people without cancer.2 In the future, cancer therapy may rely more heavily on targeted and personalized molecular approaches; many such agents are currently in development.3 A major challenge to translating preclinical studies into effective clinical therapies is the accurate identification of responsive subsets of patients. In this line, studies of trastuzumab have demonstrated that patient response is specifically correlated with amplification of the Her2 gene;4 recent data demonstrate that genetic alterations in other EGF receptors must be considered when deciding upon therapeutic strategies.5 A deeper understanding of biological pathways will likely improve the outcome of current targeted therapies. Currently, drugs against known molecular alterations driving oncogenesis, in particular those specific to altered proteins, have been shown to be effective, particularly against liquid tumors.6 Several studies r 2011 American Chemical Society
have indicated that targeted therapy may be of benefit to patients in presurgical therapy, as well as those with large and/or infiltrating tumors or highly metastatic disease.3 One example is dramatic clinical response to targeted kinase therapy observed in patients with metastatic melanoma who were treated with PLX 40327 and patients with metastatic clear-cell renal cell carcinoma treated with Sorafenib.8 Optimization of tumor targeting may be achieved by the clinical use of blood biomarkers to assess the efficacy of targeted therapy in metastatic diseases, which could reduce the number of patient cancer deaths due to metastases, limit side effects suffered by nonresponding patients and obviate the need for invasive biopsies.3 The aim of this study was to identify potential biomarkers that could provide information about the functional activation state of specific pathways involved in cancer. In the past few years, cancer biomarker research has focused mainly on analysis of plasma and/or serum proteins using highthroughput mass spectrometry methods. However, plasmabased biomarker discovery has important technical limitations, due to (1) the high dynamic range of the plasma protein Received: April 13, 2011 Published: July 14, 2011 4196
dx.doi.org/10.1021/pr200344n | J. Proteome Res. 2011, 10, 4196–4207
Journal of Proteome Research concentration, in which a few high-abundance proteins (e.g., albumin and immunoglobulins) dominate the protein content, making the detection of low-abundance proteins extremely difficult, and (2) the low relative abundance of many diseasespecific biomarkers in the large volume of blood into which the putative biomarker is secreted.9 A recent strategy for cancer biomarker discovery proposed focusing on identifying biomarkers in cancer tissue-proximal fluids and the conditioned media of cell lines (secretomes); since these fluids are close to the tumor cells, they may be enriched with secreted or shed proteins.2 The cell secretome is a simplified system that does not fully replicate the complexity of the tumor microenvironment. However, its limited complexity, compared to serum and proximal fluids, increases the likelihood of identifying proteins directly affected by intracellular tumor signaling. Furthermore, the secretome can be used to verify and characterize the effects of specific drugs on potential biomarkers.10,11 Cell secretomes (cell-conditioned media) comprise proteins that are secreted or shed from the cell surface, as well as intracellular proteins released into the supernatant due to vesiculation, cell lysis, apoptosis, and necrosis. The secretome can accurately reflect the functional state of secreting cells at specific time points.12 In vivo, tumor proteins may be secreted or shed into the bloodstream and could be used as noninvasive biomarkers, with the added analytical advantage of relatively reduced complexity compared to tissue, plasma, or serum analysis. Therefore, there are several ongoing studies to identify biomarkers in secretomes and then determine their potential clinical use.13 15 In this work, we determined the secretome composition of 12 different tumor cell lines and identified proteins that were potentially specific to tumor types. Using bioinformatics tools, we made predictions from each profile about intracellular signaling; these predictions were supported in subsequent signaling studies. Finally, we showed that treatment of a thyroid cancer cell line with two tyrosine kinase inhibitors, RPI-1 and dasatinib, which were previously demonstrated to inhibit signaling-related pathways, changed the functional categorization of the secretome composition. Collectively, our data support the concept that targeted drugs that act on specific and relevant targets may potentially modify the functional state of cancer cells, for example, by reducing vesiculation, modulating protein synthesis and protein shedding. This functional state modification can cause changes in the composition of the secretome compartment; these changes could point to new potential predictive biomarkers.
’ MATERIALS AND METHODS Cell Culture and Secretome Collection
TPC-1 cells were grown in DMEM (GIBCO, Carlsbad, CA) supplemented with 10% fetal bovine serum (FBS) (HyClone Laboratories, Logan, UT) and 1 mM sodium pyruvate (Lonza Group, Basel, Switzerland). ACHN cells were grown in E-MEM (GIBCO) supplemented with 10% FBS. MICOL cells were grown in Ham F-12 (GIBCO) supplemented with 10% FBS and 1 mM HEPES (Lonza Group). SW620, SW403, 786-O, 796P, A549, and SKOV-3 cells were grown in RPMI (Lonza Group) supplemented with 10% FBS. PC3 and LNCaP cells were grown in RPMI supplemented with 10% FBS and 1 mM HEPES. JVM13 cells were grown in Advanced RPMI 1640 (GIBCO) supplemented with 2% FBS.
ARTICLE
Cell lysates were obtained as follows: cells were washed five times with cold phosphate-buffered saline (PBS) containing 0.1 mM sodium orthovanadate and then harvested by scraping into another 0.5 mL of cold PBS before centrifugation at 2000 g for 5 min at 4 °C. The supernatants were discarded and the cell pellets were solubilized in cold lysis buffer containing 50 mM HEPES (pH 7.6), 150 mM NaCl, 10% glycerol, 1% Triton X-100, 1.5 mM MgCl2, 1 mM EGTA, 10 mM Na4P2O7, 100 mM NaF, and 1 mM sodium orthovanadate in the presence of protease inhibitors (Sigma Aldrich, St. Louis, MO). After 20 min of gentle rocking incubation at 4 °C, the lysates were cleared by centrifugation. To collect the secretomes, cells at 70% of confluence were washed five times with sterile PBS and then exposed to FBS-free medium for 18 24 h. The conditioned medium was collected and centrifuged at 2600 g for 15 min at 4 °C to remove cell debris and then concentrated and desalted 16 000 g at 4 °C) with 5 kDa cutoff spin concentrators (Agilent Technologies, Wilmington, DE). Protein concentrations were determined by BCA assay (Bio-Rad Laboratories, Milan, Italy). Drug Treatment of TPC-1 Cells
TPC-1 cells were treated for 24 h with RPI-1 (8 mM in 100% dimethylsulfoxide [DMSO]) and dasatinib (0.2 mM in 100% DMSO). Drugs were directly diluted in cell culture medium to achieve a final concentration of 40 μM for RPI-1 and 100 nM for dasatinib. The final solvent concentration was
