Primary Effusion Lymphoma: Secretome Analysis Reveals Novel Candidate Biomarkers with Potential Pathogenetic Significance

Abstract

Primary effusion lymphoma (PEL) is a rare B-cell neoplasm in which tumor cells are consistently infected by Kaposi’s sarcoma–associated herpesvirus and usually grow in body cavities without tumor mass formation. To detect new proteins related to pathogenesis, four established cell lines from PEL (CRO-AP2, CRO-AP3, CRO-AP5, and CRO-AP6) were characterized by proteomics analysis of the secretome. The secretomes were analyzed using two complementary mass spectrometry platforms: liquid chromatography–mass spectrometry and matrix-assisted laser desorption/ionization time-of-flight–based approaches. Among 266 proteins identified from the proteomics analysis, 139 were considered as predicted secreted. Twenty proteins were specifically secreted by PEL cell lines after comparison with secretomes of human cell lines representative of diverse solid tumors and leukemias. More important, 27 additional proteins were shared by all CRO-AP PEL cell lines. The presence of these proteins was confirmed by IHC in CRO-AP cell lines and in six other PEL cell lines, four PEL clinical samples, and three extracavitary Kaposi’s sarcoma–associated herpesvirus–positive solid lymphomas included for comparative analysis. Functional classification showed that PEL cell secretomes were enriched in proteins specifically involved in inflammation/immune response, growth/cell cycle, and mRNA processing, in addition to structural/matrix proteins and proteins with enzymatic activity. Primary effusion lymphoma (PEL) is a rare B-cell neoplasm in which tumor cells are consistently infected by Kaposi’s sarcoma–associated herpesvirus and usually grow in body cavities without tumor mass formation. To detect new proteins related to pathogenesis, four established cell lines from PEL (CRO-AP2, CRO-AP3, CRO-AP5, and CRO-AP6) were characterized by proteomics analysis of the secretome. The secretomes were analyzed using two complementary mass spectrometry platforms: liquid chromatography–mass spectrometry and matrix-assisted laser desorption/ionization time-of-flight–based approaches. Among 266 proteins identified from the proteomics analysis, 139 were considered as predicted secreted. Twenty proteins were specifically secreted by PEL cell lines after comparison with secretomes of human cell lines representative of diverse solid tumors and leukemias. More important, 27 additional proteins were shared by all CRO-AP PEL cell lines. The presence of these proteins was confirmed by IHC in CRO-AP cell lines and in six other PEL cell lines, four PEL clinical samples, and three extracavitary Kaposi’s sarcoma–associated herpesvirus–positive solid lymphomas included for comparative analysis. Functional classification showed that PEL cell secretomes were enriched in proteins specifically involved in inflammation/immune response, growth/cell cycle, and mRNA processing, in addition to structural/matrix proteins and proteins with enzymatic activity. CME Accreditation Statement: This activity (“ASIP 2014 AJP CME Program in Pathogenesis”) has been planned and implemented in accordance with the Essential Areas and policies of the Accreditation Council for Continuing Medical Education (ACCME) through the joint sponsorship of the American Society for Clinical Pathology (ASCP) and the American Society for Investigative Pathology (ASIP). ASCP is accredited by the ACCME to provide continuing medical education for physicians.The ASCP designates this journal-based CME activity (“ASIP 2014 AJP CME Program in Pathogenesis”) for a maximum of 48 AMA PRA Category 1 Credit(s)™. Physicians should only claim credit commensurate with the extent of their participation in the activity.CME Disclosures: The authors of this article and the planning committee members and staff have no relevant financial relationships with commercial interests to disclose. CME Accreditation Statement: This activity (“ASIP 2014 AJP CME Program in Pathogenesis”) has been planned and implemented in accordance with the Essential Areas and policies of the Accreditation Council for Continuing Medical Education (ACCME) through the joint sponsorship of the American Society for Clinical Pathology (ASCP) and the American Society for Investigative Pathology (ASIP). ASCP is accredited by the ACCME to provide continuing medical education for physicians. The ASCP designates this journal-based CME activity (“ASIP 2014 AJP CME Program in Pathogenesis”) for a maximum of 48 AMA PRA Category 1 Credit(s)™. Physicians should only claim credit commensurate with the extent of their participation in the activity. CME Disclosures: The authors of this article and the planning committee members and staff have no relevant financial relationships with commercial interests to disclose. Primary effusion lymphoma (PEL) is a rare B-cell neoplasm characterized by Kaposi’s sarcoma–associated herpesvirus (KSHV) infection of the tumor clone and by liquid growth in fluid-filled body spaces.1Cesarman E. Chang Y. Moore P.S. Said J.W. Knowles D.M. Kaposi’s sarcoma-associated herpesvirus-like DNA sequences in AIDS-related body-cavity-based lymphomas.N Engl J Med. 1995; 332: 1186-1191Crossref PubMed Scopus (2499) Google Scholar During its entire clinical course, the lymphoma tends to remain localized to the serous body cavities without mass formation.1Cesarman E. Chang Y. Moore P.S. Said J.W. Knowles D.M. Kaposi’s sarcoma-associated herpesvirus-like DNA sequences in AIDS-related body-cavity-based lymphomas.N Engl J Med. 1995; 332: 1186-1191Crossref PubMed Scopus (2499) Google Scholar, 2Said J. Cesarman E. Primary effusion lymphoma.in: Swerdlow S.H. Campo E. Harris N.L. Jaffe E.S. Pileri S.A. Stein H. Thiele J. Vardiman J.W. World Health Organization Classification of Tumours, Pathology and Genetics of Tumours of Haematopoietic and Lymphoid Tissues. IARC Press, Lyon, France2008: 260-261Google Scholar Pathologically, the basic features of these lymphomas are net predilections for diffuse spreading along the serous membranes without infiltrative or destructive growth patterns.3Carbone A. Gaidano G. HHV-8-positive body-cavity-based lymphoma: a novel lymphoma entity.Br J Haematol. 1997; 97: 515-522Crossref PubMed Scopus (109) Google Scholar, 4Carbone A. Gloghini A. KSHV/HHV8-associated lymphomas.Br J Haematol. 2008; 140: 13-24PubMed Google Scholar As seen at autopsy or incidentally revealed by computed tomography scan, PEL presents as multiple small tumor foci involving the serous membranes, which appear irregularly thickened.5Morassut S. Vaccher E. Balestreri L. Gloghini A. Gaidano G. Volpe R. Tirelli U. Carbone A. HIV-associated human herpesvirus 8-positive primary lymphomatous effusions: radiologic findings in six patients.Radiology. 1997; 205: 459-463Crossref PubMed Scopus (56) Google Scholar PEL is composed of B cells that bridge immunoblastic and anaplastic features and typically display a non-B, non-T phenotype consistent with late stages of B-cell differentiation.1Cesarman E. Chang Y. Moore P.S. Said J.W. Knowles D.M. Kaposi’s sarcoma-associated herpesvirus-like DNA sequences in AIDS-related body-cavity-based lymphomas.N Engl J Med. 1995; 332: 1186-1191Crossref PubMed Scopus (2499) Google Scholar, 2Said J. Cesarman E. Primary effusion lymphoma.in: Swerdlow S.H. Campo E. Harris N.L. Jaffe E.S. Pileri S.A. Stein H. Thiele J. Vardiman J.W. World Health Organization Classification of Tumours, Pathology and Genetics of Tumours of Haematopoietic and Lymphoid Tissues. IARC Press, Lyon, France2008: 260-261Google Scholar, 4Carbone A. Gloghini A. KSHV/HHV8-associated lymphomas.Br J Haematol. 2008; 140: 13-24PubMed Google Scholar PEL pathogenesis is poorly understood, although KSHV is thought to play a major role via expression of several viral latent genes that have the potential to affect B-cell growth.6Carbone A. Cesarman E. Spina M. Gloghini A. Schulz T.F. HIV-associated lymphomas and gamma-herpesviruses.Blood. 2009; 113: 1213-1224Crossref PubMed Scopus (262) Google Scholar Other factors involved in PEL pathogenesis include deregulation of cytokine and growth factor autocrine loops, molecular alterations of the tumor DNA, cell cycle abnormalities, stimulation and selection by antigen, and infection by Epstein-Barr virus, which occurs in 70% of PEL cases.6Carbone A. Cesarman E. Spina M. Gloghini A. Schulz T.F. HIV-associated lymphomas and gamma-herpesviruses.Blood. 2009; 113: 1213-1224Crossref PubMed Scopus (262) Google Scholar, 7Carbone A. Cesarman E. Gloghini A. Drexler H.G. Understanding pathogenetic aspects and clinical presentation of primary effusion lymphoma through its derived cell lines.AIDS. 2010; 24: 479-490Crossref PubMed Scopus (40) Google Scholar KSHV viral infection alone is not sufficient for tumor development, and lesions of cellular genes may be required for PEL lymphomagenesis. In fact, recent studies on pathogenesis of PEL using gene expression profiling, comparative genomic hybridization, and proteomic analyses have shown extensive genomic gains/amplifications and losses and overexpression of proteins, which are critical for cell migration and chemotaxis, and for cell motility.8Jenner R.G. Maillard K. Cattini N. Weiss R.A. Boshoff C. Wooster R. Kellam P. Kaposi’s sarcoma-associated herpesvirus-infected primary effusion lymphoma has a plasma cell gene expression profile.Proc Natl Acad Sci U S A. 2003; 100: 10399-10404Crossref PubMed Scopus (178) Google Scholar, 9Klein U. Gloghini A. Gaidano G. Chadburn A. Cesarman E. Dalla-Favera R. Carbone A. Gene expression profile analysis of AIDS-related primary effusion lymphoma (PEL) suggests a plasmablastic derivation and identifies PEL-specific transcripts.Blood. 2003; 101: 4115-4121Crossref PubMed Scopus (223) Google Scholar, 10Yanagisawa Y. Sato Y. Asahi-Ozaki Y. Ito E. Honma R. Imai J. Kanno T. Kano M. Akiyama H. Sata T. Shinkai-Ouchi F. Yamakawa Y. Watanabe S. Katano H. Effusion and solid lymphomas have distinctive gene and protein expression profiles in an animal model of primary effusion lymphoma.J Pathol. 2006; 209: 464-473Crossref PubMed Scopus (23) Google Scholar, 11Luan S.L. Boulanger E. Ye H. Chanudet E. Johnson N. Hamoudi R.A. Bacon C.M. Liu H. Huang Y. Said J. Chu P. Clemen C.S. Cesarman E. Chadburn A. Isaacson P.G. Du M.Q. Primary effusion lymphoma: genomic profiling revealed amplification of SELPLG and CORO1C encoding for proteins important for cell migration.J Pathol. 2010; 222: 166-179Crossref PubMed Scopus (49) Google Scholar To detect new proteins related to pathogenesis of PEL, this work performed an in-depth analysis of the secretome (conditioned cell media) of PEL cell lines. A recent strategy for discovery of cancer biomarkers involved in pathogenesis is based on the identification of proteins in cancer tissue–proximal fluids and in the conditioned media of cell lines (secretomes); because these fluids are close to the tumor cells, they may be enriched with secreted or shed proteins. 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 as the result of vesiculation, cell lysis, apoptosis, and necrosis. Furthermore, the secretome can accurately reflect the functional state of secreting cells at specific time points.12Pavlou M.P. Diamandis E.P. The cancer cell secretome: a good source for discovering biomarkers?.J Proteomics. 2010; 73: 1896-1906Crossref PubMed Scopus (152) Google Scholar, 13Caccia D. Zanetti Domingues L. Micciche F. De Bortoli M. Carniti C. Mondellini P. Bongarzone I. Secretome compartment is a valuable source of biomarkers for cancer-relevant pathways.J Proteome Res. 2011; 10: 4196-4207Crossref PubMed Scopus (45) Google Scholar, 14Caccia D. Dugo M. Callari M. Bongarzone I. Bioinformatics tools for secretome analysis.Biochim Biophys Acta. 2013; 1834: 2442-2453Crossref PubMed Scopus (74) Google Scholar It is commonly known that proteins secreted by cancer cells play important roles in cell interaction, adhesion, and invasion. To specifically study the secreted proteins, serum-free culture conditions have been applied successfully in some studies on cancer cell lines, including solid tumors and Hodgkin lymphoma.15Ma Y. Visser L. Roelofsen H. de Vries M. Diepstra A. van Imhoff G. van der Wal T. Luinge M. Alvarez-Llamas G. Vos H. Poppema S. Vonk R. van den Berg A. Proteomics analysis of Hodgkin lymphoma: identification of new players involved in the cross-talk between HRS cells and infiltrating lymphocytes.Blood. 2008; 111: 2339-2346Crossref PubMed Scopus (101) Google Scholar By secretome analysis of PEL cell lines, we identified specific proteins involved in growth, development, and maintenance of the tumor microenvironment (angiogenesis and inflammation), providing an important resource to study PEL biological characteristics and/or facilitate the discovery of novel PEL biomarkers with potential pathogenetic significance. Four PEL cell lines, named CRO-AP2,16Carbone A. Cilia A.M. Gloghini A. Canzonieri V. Pastore C. Todesco M. Cozzi M. Perin T. Volpe R. Pinto A. Gaidano G. Establishment of HHV-8-positive and HHV-8-negative lymphoma cell lines from primary lymphomatous effusions.Int J Cancer. 1997; 73: 562-569Crossref PubMed Scopus (60) Google Scholar CRO-AP3,17Carbone A. Cilia A.M. Gloghini A. Capello D. Todesco M. Quattrone S. Volpe R. Gaidano G. Establishment and characterization of EBV-positive and EBV-negative primary effusion lymphoma cell lines harbouring human herpesvirus type-8.Br J Haematol. 1998; 102: 1081-1089Crossref PubMed Scopus (73) Google Scholar CRO-AP5,17Carbone A. Cilia A.M. Gloghini A. Capello D. Todesco M. Quattrone S. Volpe R. Gaidano G. Establishment and characterization of EBV-positive and EBV-negative primary effusion lymphoma cell lines harbouring human herpesvirus type-8.Br J Haematol. 1998; 102: 1081-1089Crossref PubMed Scopus (73) Google Scholar and CRO-AP6,18Carbone A. Cilia A.M. Gloghini A. Capello D. Fassone L. Perin T. Rossi D. Canzonieri V. De Paoli P. Vaccher E. Tirelli U. Volpe R. Gaidano G. Characterization of a novel HHV-8-positive cell line reveals implications for the pathogenesis and cell cycle control of primary effusion lymphoma.Leukemia. 2000; 14: 1301-1309Crossref PubMed Scopus (32) Google Scholar were the basis of this study. All cell lines have been authenticated19Drexler H.G. Uphoff C.C. Gaidano G. Carbone A. Lymphoma cell lines: in vitro models for the study of HHV-8+ primary effusion lymphomas (body cavity-based lymphomas).Leukemia. 1998; 12: 1507-1517Crossref PubMed Scopus (102) Google Scholar, 20Uphoff C.C. Carbone A. Gaidano G. Drexler H.G. HHV-8 infection is specific for cell lines derived from primary effusion (body cavity-based) lymphomas.Leukemia. 1998; 12: 1806-1809Crossref PubMed Scopus (33) Google Scholar and comparatively reviewed7Carbone A. Cesarman E. Gloghini A. Drexler H.G. Understanding pathogenetic aspects and clinical presentation of primary effusion lymphoma through its derived cell lines.AIDS. 2010; 24: 479-490Crossref PubMed Scopus (40) Google Scholar (Supplemental Figure S1). All cell lines were cultured in RPMI 1640 medium (GIBCO, Carlsbad, CA) supplemented with 20% heat-inactivated fetal bovine serum (FBS; GIBCO), 100 IU/mL penicillin/streptomycin (PenStrep; GIBCO), and 2 mmol/L l-glutammin (Lonza, Verviers, Belgium) under a 5% CO2 atmosphere at 37°C. Formalin-fixed, paraffin-embedded cell blocks from PEL cell lines were used for immunohistochemistry (IHC) and in situ hybridization (ISH) analyses. Before secretome preparation, the viral status of PEL cell lines was confirmed by IHC and ISH. Cell blocks obtained from other PEL-derived cell lines, BC-1,21Cesarman E. Moore P.S. Rao P.H. Inghirami G. Knowles D.M. Chang Y. In vitro establishment and characterization of two acquired immunodeficiency syndrome-related lymphoma cell lines (BC-1 and BC-2) containing Kaposi’s sarcoma-associated herpesvirus-like (KSHV) DNA sequences.Blood. 1995; 86: 2708-2714Crossref PubMed Google Scholar and its sister cell line, HBL-6,22Gaidano G. Cechova K. Chang Y. Moore P.S. Knowles D.M. Dalla-Favera R. Establishment of AIDS-related lymphoma cell lines from lymphomatous effusions.Leukemia. 1996; 10: 1237-1240PubMed Google Scholar BC-2,21Cesarman E. Moore P.S. Rao P.H. Inghirami G. Knowles D.M. Chang Y. In vitro establishment and characterization of two acquired immunodeficiency syndrome-related lymphoma cell lines (BC-1 and BC-2) containing Kaposi’s sarcoma-associated herpesvirus-like (KSHV) DNA sequences.Blood. 1995; 86: 2708-2714Crossref PubMed Google Scholar BC-3,23Arvanitakis L. Mesri E.A. Nador R.G. Said J.W. Asch A.S. Knowles D.M. Cesarman E. Establishment and characterization of a primary effusion (body cavity-based) lymphoma cell line (BC-3) harboring kaposi’s sarcoma-associated herpesvirus (KSHV/HHV-8) in the absence of Epstein-Barr virus.Blood. 1996; 88: 2648-2654PubMed Google Scholar BCBL-1,24Renne R. Zhong W. Herndier B. McGrath M. Abbey N. Kedes D. Ganem D. Lytic growth of Kaposi’s sarcoma-associated herpesvirus (human herpesvirus 8) in culture.Nat Med. 1996; 2: 342-346Crossref PubMed Scopus (914) Google Scholar and BCP-1,25Boshoff C. Gao S.J. Healy L.E. Matthews S. Thomas A.J. Coignet L. Warnke R.A. Strauchen J.A. Matutes E. Kamel O.W. Moore P.S. Weiss R.A. Chang Y. Establishing a KSHV+ cell line (BCP-1) from peripheral blood and characterizing its growth in Nod/SCID mice.Blood. 1998; 91: 1671-1679Crossref PubMed Google Scholar obtained from Prof. Gianluca Gaidano (Division of Hematology, Department of Translational Medicine, Amedeo Avogadro University of Eastern Piedmont, Novara, Italy), were also used for IHC analyses. Four PEL clinical samples and three primary extracavitary KSHV-positive solid lymphomas4Carbone A. Gloghini A. KSHV/HHV8-associated lymphomas.Br J Haematol. 2008; 140: 13-24PubMed Google Scholar were investigated for the expression of proteins identified in the secretome of PEL cell lines. The cases belonged to a single institution series observed at the Centro di Riferimento Oncologico, National Cancer Institute of Aviano, Italy. All cases were diagnosed and classified according to morphological, immunophenotypical, and virological features previously reported1Cesarman E. Chang Y. Moore P.S. Said J.W. Knowles D.M. Kaposi’s sarcoma-associated herpesvirus-like DNA sequences in AIDS-related body-cavity-based lymphomas.N Engl J Med. 1995; 332: 1186-1191Crossref PubMed Scopus (2499) Google Scholar, 4Carbone A. Gloghini A. KSHV/HHV8-associated lymphomas.Br J Haematol. 2008; 140: 13-24PubMed Google Scholar and acknowledged by the 2008 World Health Classification of Tumors of Hematopoietic and Lymphoid Tissues.2Said J. Cesarman E. Primary effusion lymphoma.in: Swerdlow S.H. Campo E. Harris N.L. Jaffe E.S. Pileri S.A. Stein H. Thiele J. Vardiman J.W. World Health Organization Classification of Tumours, Pathology and Genetics of Tumours of Haematopoietic and Lymphoid Tissues. IARC Press, Lyon, France2008: 260-261Google Scholar Cell lines were cultured in FBS-free RPMI 1640 medium, without phenol red and with 100 UI PenStrep and 2 mmol/L l-glutamine, to have at least 3 × 107 cells. After 16 to 18 hours, cells were collected and their viability was verified. If the cell death rate was <10%, the secretome samples in conditioned media were collected. The conditioned media were then centrifuged at 3116 × g for 15 minutes at 4°C to remove cells debris, and finally were concentrated and desalted with 5-kDa cutoff spin concentrators (Agilent Technologies Inc., Wilmington, DE). Protein concentrations of the secretome samples were determined by bicinchoninic acid assay (Bio-Rad Laboratories Srl, Segrate, Italy). All electrophoresis (SDS-PAGE) experiments were performed with precast 4% to 12% Bis-Tris polyacrylamide NuPAGE NOVEX gels (Invitrogen, Segrate, Italy) and stained with colloidal Coomassie Blue, according to standard procedures.26Caccia D. Micciche F. Cassinelli G. Mondellini P. Casalini P. Bongarzone I. Dasatinib reduces FAK phosphorylation increasing the effects of RPI-1 inhibition in a RET/PTC1-expressing cell line.Mol Cancer. 2010; 9: 278Crossref PubMed Scopus (37) Google Scholar Protein identification was performed using matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) and liquid chromatography coupled to high-resolution Orbitrap MS (LC-Orbitrap MS), as previously described.13Caccia D. Zanetti Domingues L. Micciche F. De Bortoli M. Carniti C. Mondellini P. Bongarzone I. Secretome compartment is a valuable source of biomarkers for cancer-relevant pathways.J Proteome Res. 2011; 10: 4196-4207Crossref PubMed Scopus (45) Google Scholar The same amount of protein (ie, 20 μg) from all cell lines was profiled for secretome analysis. For MALDI-TOF MS identification of proteins, secreted proteins were resolved by either SDS-PAGE or OFFGEL fractionator (Agilent Technologies, Milan, Italy) and visualized by silver or colloidal Coomassie Blue staining, according to standard procedures. Protein bands were excised and processed, as previously described.26Caccia D. Micciche F. Cassinelli G. Mondellini P. Casalini P. Bongarzone I. Dasatinib reduces FAK phosphorylation increasing the effects of RPI-1 inhibition in a RET/PTC1-expressing cell line.Mol Cancer. 2010; 9: 278Crossref PubMed Scopus (37) Google Scholar Briefly, proteins were in-gel digested with trypsin (13 ng/μL) for 18 hours at 37°C. The peptide mixture obtained from each band was analyzed by a MALDI-TOF Voyager-DE STR (Applied Biosystems, Framingham, MA) equipped with a nitrogen laser (337 nm). The resulting spectra were analyzed with MASCOT peptide Mass Fingerprint (http://www.matrixscience.com/cgi/search_form.pl?FORMVER=2&SEARCH=PMF, last accessed August 5, 2011). Input was searched according to the following database: SwissProt; taxonomy, Mammalia; peptide tolerance, 20 ppm; spectrometer internal error maximum, 25. Only proteins identified in at least three separate experiments were considered. For nanoscale LC-MS/MS analysis, 20 μg of reduced and carbamylated proteins from secretome samples was digested with Trypsin Gold (1:20, w/w; Promega Corporation, Fitchburg, WI). Peptide separations were performed on a 12-cm reverse-phase spraying fused silica capillary column (75 μm i.d. × 15 cm), which was packed with 3-μm ReproSil 100 Å C18 (Dr. Maisch HPLC GmbH, Ammerbuch-Entringen, Germany). The LC system was connected to an LTQ Orbitrap MS (ThermoScientific, Bremen, Germany) equipped with a nanoelectrospray ion source (Proxeon Biosystems, Odense, Denmark). Full-scan mass spectra were acquired in the LTQ Orbitrap MS with the resolution set to 60,000. The lock-mass option was used for accurate mass measurements. The acquisition mass/charge ratio range for each sample was from 350 to 1700 Da. The six most intense doubly and triply charged ions were automatically selected and fragmented in the ion trap after accumulation to a target value of 10,000. Target ions already selected for the MS/MS were dynamically excluded for 45 seconds. All MS/MS samples were analyzed using the Mascot search engine, version 2.2.06 (Matrix Science, London, UK) and X! Tandem, version 2007.01.01.1. X! Tandem and Mascot were set up to search the SwissProt database. For LTQ-Orbitrap data, mass tolerance was set to 5 ppm and 0.6 Da for precursor and fragment ions, respectively. Scaffold, version Scaffold_2_06_00 (Proteome Software Inc., Portland, OR), was used to validate MS/MS-based peptide and protein identifications. Protein thresholds were set to a 99.0% minimum and a two-peptide minimum, whereas peptide thresholds were set to a 95% minimum. The lists of identified proteins were analyzed and classified using bioinformatics software: Signal Peptide Predictor (SignalP version 4.1; http://www.cbs.dtu.dk/services/SignalP, last accessed October 20, 2013) was used to analyze the secretion features of the identified proteins.27Bendtsen J.D. Jensen L.J. Blom N. Von Heijne G. Brunak S. Feature-based prediction of non-classical and leaderless protein secretion.Protein Eng Des Sel. 2004; 17: 349-356Crossref PubMed Scopus (938) Google Scholar, 28Nielsen H. Engelbrecht J. Brunak S. von Heijne G. Identification of prokaryotic and eukaryotic signal peptides and prediction of their cleavage sites.Protein Eng. 1997; 10: 1-6Crossref PubMed Scopus (4928) Google Scholar SignalP uses amino acid sequences to predict the existence and location of signal peptide cleavage sites. A protein is considered to be classically secreted if it receives a signal peptide probability of >0.9. To identify non-classical, or leaderless, protein secretion, SecretomeP (http://www.cbs.dtu.dk/services/SecretomeP, last accessed October 24, 2013) was used.29Bendtsen J.D. Nielsen H. von Heijne G. Brunak S. Improved prediction of signal peptides: SignalP 3.0.J Mol Biol. 2004; 340: 783-795Crossref PubMed Scopus (5634) Google Scholar SecretomeP uses a neural network that combines six protein characteristics to determine whether a protein is non-classically secreted. These characteristics include the following: number of amino acids, number of positively charged residues, presence of transmembrane helices, presence of low-complexity regions, presence of propeptides, and subcellular localization. A protein is considered non-classically secreted if it receives an NNscore of >0.5. Finally, it was also possible that proteins located on the plasma membrane were shed and released to the extracellular space. Therefore, TMHMM (http://www.cbs.dtu.dk/services/TMHMM, last accessed October 24, 2013) was used to predict transmembrane helices. Some secretory proteins, such as proteins secreted through exosome vesicles, may not pass the SignalP and SecretomeP score cutoffs.30Mathivanan S. Simpson R.J. ExoCarta: a compendium of exosomal proteins and RNA.Proteomics. 2009; 9: 4997-5000Crossref PubMed Scopus (664) Google Scholar, 31Simpson R.J. Kalra H. Mathivanan S. ExoCarta as a resource for exosomal research.J Extracell Vesicles. 2012; 1: 18374https://doi.org/10.3402/jev.v1i0.18374Crossref Scopus (263) Google Scholar, 32Mathivanan S. Fahner C.J. Reid G.E. Simpson R.J. ExoCarta 2012: database of exosomal proteins, RNA and lipids.Nucleic Acids Res. 2012; 40: D1241-D1244Crossref PubMed Scopus (751) Google Scholar The list of PEL-specific proteins was accomplished by filtering the list of identified proteins against a background database of proteins found in the conditioned medium of tumor cell lines of a different histotype generated in-house.13Caccia D. Zanetti Domingues L. Micciche F. De Bortoli M. Carniti C. Mondellini P. Bongarzone I. Secretome compartment is a valuable source of biomarkers for cancer-relevant pathways.J Proteome Res. 2011; 10: 4196-4207Crossref PubMed Scopus (45) Google Scholar The database includes 1370 proteins, 717 (52%) of which are either classically or non-classically secreted, and 125 (9%) of which are transmembrane proteins; the remaining proteins are intracellular or nonconventional secreted proteins. Finally, the PEL secretomes were compared with the secretome from the B-cell lymphoma cell line, DOOH-2 [KSHV−/Epstein-Barr virus (EBV)−]. Data from DOOH-2 were obtained from an in-house database of secreted proteins. We applied a clustering method to protein expression matrix in which rows corresponded to proteins and columns to cell lines. Unsupervised clustering only organizes the data based on their expression similarity. In this study, hierarchical clustering was performed on both genes and experiments by using the average linkage method. Eisen’s CLUSTER and TREEVIEW software version 1.2 (http://rana.lbl.gov/EisenSoftware.htm, last accessed October 24, 2013) was used to perform this analysis and display the results. Functional categories of proteins found in the conditioned media were analyzed using MetaCore version 6.8 build 30387 (Thomson Reuters, St. Joseph, MI), including gene ontology (GO) and canonical pathway maps. The MetaCore program uses GeneGo’s MetaBase knowledgebase containing approximately 300,000 protein-protein and protein–small molecule interactions manually extracted from the literature by a group of experts.33Dezso Z. Nikolsky Y. Nikolskaya T. Miller J. Cherba D. Webb C. Bugrim A. Identifying disease-specific genes based on their topological significance in protein networks.BMC Syst Biol. 2009; 3: 36Crossref PubMed Scopus (94) Google Scholar The P value is essentially the probability of a particular mapping arising by chance given the number of proteins in the set relative to all proteins on maps/processes, proteins on a particular map/process, and proteins in the analyzed experiment. A pathway or network with P ≤ 0.01 was considered as significant. Version 9.05 of the Search Tool for the Analysis of Interacting Genes/Proteins (STRING; http://string-db.org, last accessed October 16, 2013) resource was used.34Szklarczyk D. Franceschini A. Kuhn M. Simonovic M. Roth A. Minguez P. Doerks T. Stark M. Muller J. Bork P. Jensen L.J. von Mering C. The STRING database in 2011: functional interaction networks of proteins, globally integrated and scored.Nucleic Acids Res. 2011; 39: D561-D568Crossref PubMed Scopus (2705) Google Scholar The STRING algorithm links genes or proteins into networks based on published functional or informatics-predicted interactions. Herein, reported interactions are predicted with a medium confidence threshold of 0.400. Overrepresented categories were identified relative to the Homo sapiens background gene set using a hypergeometric test with a Benjamini and Hochberg false-discovery rate (FDR) correction to define significance. Moreover, a map of Top Canonical Pathways that may be overrepresented in the PEL secretome was obtained using Ingenuity software (Ingenuity Pathway Analysis, version 9; IngenuitySystems, Redwood City, CA). Both IHC and ISH were performed on an automated immunosta