Advertisement
Regular article Tumorigenesis and neoplastic progression| Volume 181, ISSUE 5, P1782-1795, November 2012

Oncostatin M Is a Growth Factor for Ewing Sarcoma

  • Emmanuelle David
    Affiliations
    INSERM, UMR 957, Ligue Team 2012, Nantes, France

    Université de Nantes, Nantes Atlantique Universités, Pathophysiology of Bone Resorption Laboratory and Therapy of Primary Bone Tumors, Medicine Faculty, Nantes, France
    Search for articles by this author
  • Franck Tirode
    Affiliations
    Curie Institute, Paris, France

    Genetics and Biology of Cancers Unit, INSERM, UMR 830, Paris, France
    Search for articles by this author
  • Marc Baud'huin
    Affiliations
    INSERM, UMR 957, Ligue Team 2012, Nantes, France

    Université de Nantes, Nantes Atlantique Universités, Pathophysiology of Bone Resorption Laboratory and Therapy of Primary Bone Tumors, Medicine Faculty, Nantes, France

    University Hospital, Nantes, France
    Search for articles by this author
  • Pierre Guihard
    Affiliations
    INSERM, UMR 957, Ligue Team 2012, Nantes, France

    Université de Nantes, Nantes Atlantique Universités, Pathophysiology of Bone Resorption Laboratory and Therapy of Primary Bone Tumors, Medicine Faculty, Nantes, France
    Search for articles by this author
  • Karine Laud
    Affiliations
    Curie Institute, Paris, France

    Genetics and Biology of Cancers Unit, INSERM, UMR 830, Paris, France
    Search for articles by this author
  • Olivier Delattre
    Affiliations
    Curie Institute, Paris, France

    Genetics and Biology of Cancers Unit, INSERM, UMR 830, Paris, France
    Search for articles by this author
  • Marie F. Heymann
    Affiliations
    INSERM, UMR 957, Ligue Team 2012, Nantes, France

    University Hospital, Nantes, France
    Search for articles by this author
  • Dominique Heymann
    Affiliations
    INSERM, UMR 957, Ligue Team 2012, Nantes, France

    Université de Nantes, Nantes Atlantique Universités, Pathophysiology of Bone Resorption Laboratory and Therapy of Primary Bone Tumors, Medicine Faculty, Nantes, France

    University Hospital, Nantes, France
    Search for articles by this author
  • Françoise Redini
    Affiliations
    INSERM, UMR 957, Ligue Team 2012, Nantes, France

    Université de Nantes, Nantes Atlantique Universités, Pathophysiology of Bone Resorption Laboratory and Therapy of Primary Bone Tumors, Medicine Faculty, Nantes, France

    University Hospital, Nantes, France
    Search for articles by this author
  • Frédéric Blanchard
    Correspondence
    Address reprint requests to Frédéric Blanchard, Ph.D., French National Institute of Health and Medical Research (INSERM), unit 957, Medicine faculty, 1 rue Gaston Veil, F-44035 Nantes, France
    Affiliations
    INSERM, UMR 957, Ligue Team 2012, Nantes, France

    Université de Nantes, Nantes Atlantique Universités, Pathophysiology of Bone Resorption Laboratory and Therapy of Primary Bone Tumors, Medicine Faculty, Nantes, France
    Search for articles by this author
Published:September 17, 2012DOI:https://doi.org/10.1016/j.ajpath.2012.07.023
      Primary bone tumors, osteosarcomas and chondrosarcomas, derive from mesenchymal stem cells committed into osteoblasts and chondrocytes; in Ewing sarcomas (ESs), the oncogenic fusion protein EWS-FLI1 prevents mesenchymal differentiation and induces neuroectodermic features. Oncostatin M (OSM) is a cytokine from the IL-6 family that modulates proliferation and differentiation in numerous cells. The basis for inhibition versus induction of proliferation by this cytokine is obscure, although MYC was described as a potent molecular switch in OSM signaling. We show herein that, in contrast to osteosarcomas and chondrosarcomas, for which OSM was cytostatic, OSM induced proliferation of ES cell lines. Knockdown experiments demonstrated that growth induction by OSM depends on both types I [leukemia inhibitory factor receptor (LIFR)] and II [OSM receptor (OSMR)] receptors, high STAT3 activation, and induction of MYC to a high expression level. Indeed, ES cell lines, mice xenografts, and patient biopsy specimens poorly expressed LIF, precluding LIFR lysosomal degradation and OSMR transcriptional induction, thus leading to a high LIFR/OSMR ratio. Because other neuroectodermic tumors (ie, glioma, medulloblastoma, and neuroblastoma) had a similar expression profile, the main role of EWS-FLI1 could be through maintenance of stemness and neuroectodermic features, characterized by a low LIF, a high LIFR/OSMR ratio, and high MYC expression. Thus, this study on rare bone malignancies gives valuable insights on more common cancer regulatory mechanisms and could provide new therapeutic opportunities.
      Primary bone tumors are rare malignancies, with an incidence of 10 new cases per year per million inhabitants in the United States. Osteosarcomas (OSs), chondrosarcomas (CSs), and Ewing sarcomas (ESs) represent 35%, 25%, and 20% of these tumors, respectively. OS occurs mainly in adolescents, with a peak incidence at the age of 18 years, and is characterized by bone formation and osteolytic lesions.
      • Picci P.
      Osteosarcoma (osteogenic sarcoma).
      CS occurs only in adults, with a peak incidence at the age of 45 years, and is characterized by production of cartilage matrix, high bone remodeling, and poor vascularity. ES occurs mostly in children and adolescents, with a peak incidence at the age of 15 years, and is characterized by marked osteolysis with a periosteal reaction and frequent soft tissue invasion.
      • Bernstein M.
      • Kovar H.
      • Paulussen M.
      • Randall R.L.
      • Schuck A.
      • Teot L.A.
      • Juergens H.
      Ewing's sarcoma family of tumors: current management.
      Despite their common origin, these three bone tumors have a distinct differentiation status. OS and CS presumably arise from the same multipotent mesenchymal stem cell (MSC), which is able to differentiate into the osteoblast or chondrocyte lineage, respectively.
      • Mohseny A.B.
      • Szuhai K.
      • Romeo S.
      • Buddingh E.P.
      • Briaire-de Bruijn I.
      • de Jong D.
      • van Pel M.
      • Cleton-Jansen A.M.
      • Hogendoorn P.C.
      Osteosarcoma originates from mesenchymal stem cells in consequence of aneuploidization and genomic loss of Cdkn2.
      • Boeuf S.
      • Kunz P.
      • Hennig T.
      • Lehner B.
      • Hogendoorn P.
      • Bovee J.
      • Richter W.
      A chondrogenic gene expression signature in mesenchymal stem cells is a classifier of conventional central chondrosarcoma.
      Even if the molecular origin of OS and CS is not clearly defined, loss of function of p53 is observed in >50% of these tumors.
      • Wadayama B.
      • Toguchida J.
      • Yamaguchi T.
      • Sasaki M.S.
      • Yamamuro T.
      p53 Expression and its relationship to DNA alterations in bone and soft tissue sarcomas.
      ES shares a typical chromosomal translocation in >90% of cases linking the EWS gene on chromosome 22q12 to a member of the ETS gene family,
      • Delattre O.
      • Zucman J.
      • Plougastel B.
      • Desmaze C.
      • Melot T.
      • Peter M.
      • Kovar H.
      • Joubert I.
      • de Jong P.
      • Rouleau G.
      • Aurias A.
      • Thomas G.
      Gene fusion with an ETS DNA-binding domain caused by chromosome translocation in human tumours.
      most commonly to FLI1 on 11q24, leading to an aberrant transcription factor that promotes tumorigenicity
      • May W.A.
      • Gishizky M.L.
      • Lessnick S.L.
      • Lunsford L.B.
      • Lewis B.C.
      • Delattre O.
      • Zucman J.
      • Thomas G.
      • Denny C.T.
      Ewing sarcoma 11;22 translocation produces a chimeric transcription factor that requires the DNA-binding domain encoded by FLI1 for transformation.
      and inhibits mesenchymal differentiation.
      • Tirode F.
      • Laud-Duval K.
      • Prieur A.
      • Delorme B.
      • Charbord P.
      • Delattre O.
      Mesenchymal stem cell features of Ewing tumors.
      In opposition to OSs and CSs, ESs are poorly differentiated tumors, originally classified in the primitive neuroectodermal tumors. In fact, ESs presumably also derive from MSCs, in which EWS-FLI1 induces expression of various neuronal markers.
      • Tirode F.
      • Laud-Duval K.
      • Prieur A.
      • Delorme B.
      • Charbord P.
      • Delattre O.
      Mesenchymal stem cell features of Ewing tumors.
      • Riggi N.
      • Suva M.L.
      • De Vito C.
      • Provero P.
      • Stehle J.C.
      • Baumer K.
      • Cironi L.
      • Janiszewska M.
      • Petricevic T.
      • Suva D.
      • Tercier S.
      • Joseph J.M.
      • Guillou L.
      • Stamenkovic I.
      EWS-FLI-1 modulates miRNA145 and SOX2 expression to initiate mesenchymal stem cell reprogramming toward Ewing sarcoma cancer stem cells.
      • Rocchi A.
      • Manara M.C.
      • Sciandra M.
      • Zambelli D.
      • Nardi F.
      • Nicoletti G.
      • Garofalo C.
      • Meschini S.
      • Astolfi A.
      • Colombo M.P.
      • Lessnick S.L.
      • Picci P.
      • Scotlandi K.
      CD99 inhibits neural differentiation of human Ewing sarcoma cells and thereby contributes to oncogenesis.
      Current therapeutic protocols in OS and ES consist of neoadjuvant chemotherapy and local surgical resection, followed by adjuvant chemotherapy. These treatments lead to a 70% overall survival for localized disease but can decrease to 15% in case of metastasis. With CS being resistant to conventional chemotherapy and radiotherapy, the therapeutic surgical approach remains the only available treatment, with a 10-year survival between 30% and 80%, depending on the grade.
      • Lee F.Y.
      • Mankin H.J.
      • Fondren G.
      • Gebhardt M.C.
      • Springfield D.S.
      • Rosenberg A.E.
      • Jennings L.C.
      Chondrosarcoma of bone: an assessment of outcome.
      Because the survival rates have not changed in the past 30 years,
      • Cotterill S.J.
      • Ahrens S.
      • Paulussen M.
      • Jurgens H.F.
      • Voute P.A.
      • Gadner H.
      • Craft A.W.
      Prognostic factors in Ewing's tumor of bone: analysis of 975 patients from the European Intergroup Cooperative Ewing's Sarcoma Study Group.
      new therapeutic approaches are needed, and some studies described the potential of immunotherapy, including cytokine-based therapy, in combination with conventional treatment to eradicate primary bone tumors.
      • Mori K.
      • Rédini F.
      • Gouin F.
      • Cherrier B.
      • Heymann D.
      Osteosarcoma: current status of immunotherapy and future trends (review).
      The IL-6 family of cytokines actually comprises nine members, such as leukemia inhibitory factor (LIF) and oncostatin M (OSM). All IL-6–type cytokines share the signal-transducing receptor subunit gp130 (or gp130 like), and receptor specificity is provided by additional receptor chains. IL-6 first binds to the IL-6 receptor subunit (IL-6R), which is either membrane associated or soluble, and then recruits gp130.
      • Heinrich P.C.
      • Behrmann I.
      • Haan S.
      • Hermanns H.M.
      • Muller-Newen G.
      • Schaper F.
      Principles of interleukin (IL)-6-type cytokine signalling and its regulation.
      OSM and LIF are functionally and structurally related, and both cytokines can bind to the type I receptor composed of gp130 and the LIF receptor (LIFR). OSM can also transduce a signal through the type II receptor, composed of gp130 and oncostatin M receptor (OSMR).
      • Heinrich P.C.
      • Behrmann I.
      • Haan S.
      • Hermanns H.M.
      • Muller-Newen G.
      • Schaper F.
      Principles of interleukin (IL)-6-type cytokine signalling and its regulation.
      Ligand-induced oligomerization of receptor subunits allows activation of Janus kinases and, subsequently, signal transducer and activator of transcription (mainly STAT3) and mitogen-activated protein kinase (mainly extracellular signal–regulated kinase 1/2).
      • Heinrich P.C.
      • Behrmann I.
      • Haan S.
      • Hermanns H.M.
      • Muller-Newen G.
      • Schaper F.
      Principles of interleukin (IL)-6-type cytokine signalling and its regulation.
      OSM and LIF are pleiotropic cytokines that play a crucial role in the systemic inflammatory response, hematopoiesis, and liver and neuronal regeneration; in addition, they activate endothelial cells.
      • Chen S.H.
      • Benveniste E.N.
      Oncostatin M: a pleiotropic cytokine in the central nervous system.
      • Blanchard F.
      • Duplomb L.
      • Baud'huin M.
      • Brounais B.
      The dual role of IL-6-type cytokines on bone remodeling and bone tumors.
      OSM is mainly produced by macrophages, neutrophils, and T lymphocytes, whereas LIF could also be produced by tumor cells.
      • Blanchard F.
      • Duplomb L.
      • Baud'huin M.
      • Brounais B.
      The dual role of IL-6-type cytokines on bone remodeling and bone tumors.
      • Grenier A.
      • Dehoux M.
      • Boutten A.
      • Arce-Vicioso M.
      • Durand G.
      • Gougerot-Pocidalo M.A.
      • Chollet-Martin S.
      Oncostatin M production and regulation by human polymorphonuclear neutrophils.
      Interestingly, OSM was first identified by its capacity to inhibit melanoma cell growth in vitro,
      • Zarling J.M.
      • Shoyab M.
      • Marquardt H.
      • Hanson M.B.
      • Lioubin M.N.
      • Todaro G.J.
      Oncostatin M: a growth regulator produced by differentiated histiocytic lymphoma cells.
      and this property was then enlarged to breast cancer, lung cancer, or hepatoma.
      • Blanchard F.
      • Duplomb L.
      • Baud'huin M.
      • Brounais B.
      The dual role of IL-6-type cytokines on bone remodeling and bone tumors.
      LIF also inhibits proliferation of thyroid cancer and cervical carcinoma.
      • Arthan D.
      • Hong S.K.
      • Park J.I.
      Leukemia inhibitory factor can mediate Ras/Raf/MEK/ERK-induced growth inhibitory signaling in medullary thyroid cancer cells.
      Within primary bone tumors, OSM has anticancer effects on osteosarcoma through inhibition of proliferation, sensitization to p53-dependent apoptosis, and induction of differentiation into the osteoblastic pathway.
      • Brounais B.
      • Chipoy C.
      • Mori K.
      • Charrier C.
      • Battaglia S.
      • Pilet P.
      • Richards C.D.
      • Heymann D.
      • Redini F.
      • Blanchard F.
      Oncostatin M induces bone loss and sensitizes rat osteosarcoma to the antitumor effect of Midostaurin in vivo.
      • Chipoy C.
      • Brounais B.
      • Trichet V.
      • Battaglia S.
      • Berreur M.
      • Oliver L.
      • Juin P.
      • Redini F.
      • Heymann D.
      • Blanchard F.
      Sensitization of osteosarcoma cells to apoptosis by oncostatin M depends on STAT5 and p53.
      • Brounais B.
      • David E.
      • Chipoy C.
      • Trichet V.
      • Ferre V.
      • Charrier C.
      • Duplomb L.
      • Berreur M.
      • Redini F.
      • Heymann D.
      • Blanchard F.
      Long term oncostatin M treatment induces an osteocyte-like differentiation on osteosarcoma and calvaria cells.
      Recently, we also demonstrated that OSM inhibits proliferation of chondrosarcoma in vitro and in vivo.
      • David E.
      • Guihard P.
      • Brounais B.
      • Riet A.
      • Charrier C.
      • Battaglia S.
      • Gouin F.
      • Ponsolle S.
      • Bot R.L.
      • Richards C.D.
      • Heymann D.
      • Redini F.
      • Blanchard F.
      Direct anti-cancer effect of oncostatin M on chondrosarcoma.
      In contrast, OSM enhances proliferation of fibroblasts, myeloma, Kaposi's sarcoma, and cervical carcinoma.
      • Langdon C.
      • Kerr C.
      • Hassen M.
      • Hara T.
      • Arsenault A.L.
      • Richards C.D.
      Murine oncostatin M stimulates mouse synovial fibroblasts in vitro and induces inflammation and destruction in mouse joints in vivo.
      • Lundquist A.
      • Barre B.
      • Bienvenu F.
      • Hermann J.
      • Avril S.
      • Coqueret O.
      Kaposi sarcoma-associated viral cyclin K overrides cell growth inhibition mediated by oncostatin M through STAT3 inhibition.
      Primary melanoma cells are growth inhibited by OSM,
      • Lacreusette A.
      • Lartigue A.
      • Nguyen J.M.
      • Barbieux I.
      • Pandolfino M.C.
      • Paris F.
      • Khammari A.
      • Dréno B.
      • Jacques Y.
      • Blanchard F.
      • Godard A.
      Relationship between responsiveness of cancer cells to Oncostatin M and/or IL-6 and survival of stage III melanoma patients treated with tumour-infiltrating lymphocytes.
      but during the metastatic process, these cells become resistant or even growth stimulated and lose the OSM receptor.
      • Lacreusette A.
      • Nguyen J.M.
      • Pandolfino M.C.
      • Khammari A.
      • Dreno B.
      • Jacques Y.
      • Godard A.
      • Blanchard F.
      Loss of oncostatin M receptor beta in metastatic melanoma cells.
      LIF also induces proliferation of embryonic stem cells
      • Hirai H.
      • Karian P.
      • Kikyo N.
      Regulation of embryonic stem cell self-renewal and pluripotency by leukaemia inhibitory factor.
      and could participate in progression of breast cancer
      • Shin J.E.
      • Park S.H.
      • Jang Y.K.
      Epigenetic up-regulation of leukemia inhibitory factor (LIF) gene during the progression to breast cancer.
      or melanoma.
      • Maruta S.
      • Takiguchi S.
      • Ueyama M.
      • Kataoka Y.
      • Oda Y.
      • Tsuneyoshi M.
      • Iguchi H.
      A role for leukemia inhibitory factor in melanoma-induced bone metastasis.
      The cytostatic effect of OSM or LIF occurs through STAT3 activation and induction of cell cycle inhibitors, such as p21WAF1 and/or p27KIP1,
      • Kortylewski M.
      • Heinrich P.C.
      • Mackiewicz A.
      • Schniertshauer U.
      • Klingmuller U.
      • Nakajima K.
      • Hirano T.
      • Horn F.
      • Behrmann I.
      Interleukin-6 and oncostatin M-induced growth inhibition of human A375 melanoma cells is STAT-dependent and involves upregulation of the cyclin-dependent kinase inhibitor p27/Kip1.
      • Bellido T.
      • O'Brien C.A.
      • Roberson P.K.
      • Manolagas S.C.
      Transcriptional activation of the p21(WAF1,CIP1,SDI1) gene by interleukin-6 type cytokines: a prerequisite for their pro-differentiating and anti-apoptotic effects on human osteoblastic cells.
      or transcription factors linked to the differentiation process, such as C/EBPβ or δ.
      • Hutt J.A.
      • DeWille J.W.
      Oncostatin M induces growth arrest of mammary epithelium via a CCAAT/enhancer-binding protein δ-dependent pathway.
      However, STAT3 has also been activated in numerous cancers, where it behaves as an oncogene. Through collaboration with other transcription factors, such as NF-κB or hypoxia-inducible factor-1, STAT3 regulates the expression of genes that mediate survival (SURVIVIN, BCLXL, and MCL1), proliferation (FOS, MYC, and cyclin D1), invasion (matrix metalloproteinase-2), and angiogenesis (vascular endothelial growth factor).
      • Blanchard F.
      • Duplomb L.
      • Baud'huin M.
      • Brounais B.
      The dual role of IL-6-type cytokines on bone remodeling and bone tumors.
      We demonstrate herein that OSM enhances proliferation of ES cells, whereas it is cytostatic for OS and CS cells. These opposite effects on tumor cell proliferation appeared to be linked to the differentiation status, the LIFR/OSMR ratio, activation of STAT3, and induction of MYC, and could be generalized to other types of tumors.

      Materials and Methods

      Tumor Material

      Twenty-two human primary bone tumor cell lines were used, including 4 CSs (SW1353, CAL78, OUMS27, and BCSCH03), 8 OSs (MG63, SaOS2, U2OS, MNNG-HOS, CAL72, G-292, SJSA-1, and 143-B), and 10 ESs (TC71, SKNMC, EW24, A673, TC32, SKES1, RDES, STAET1, EW7, and BRZ). Cell origin, characteristics, and culture medium are presented in Supplemental Table S1 (available at http://ajp.amjpathol.org).
      For xenografts, 4-week-old Rj:NMRI-nude mice (Elevages Janvier, Le Genest-Saint-Isle, France) were housed in accordance with the institutional guidelines of the French Ethical Committee (CEEA Pays de la Loire n°06) and under the supervision of authorized investigators (F.R. and F.B.). Mice received an injection of 2 × 106 MNNG-HOS, SW1353, A673, STAET1, TC71, or SKES1 cells in paratibial muscle. When tumors reached 3000 mm3, they were sacrificed and tumor fragments were collected.
      Patient tumor biopsy specimens were collected at Nantes University Hospital (Nantes, France). Samples were obtained with patient informed consent, after ethical approval by the Nantes University Hospital Ethics Committee.

      Cytokines, Inhibitors, and Expression Vectors

      Human LIF (obtained from Dr. Anne Godard, INSERM, Unit 892, Nantes, France) was used at 100 ng/mL. Other human cytokines or receptors of the IL-6 family were obtained at R&D Systems (Minneapolis, MN) and used at 50 ng/mL, unless otherwise stated. IL-1β (25 ng/mL), tumor necrosis factor-α (10 ng/mL), neutralizing antibody targeting gp130, and irrelevant IgG (10 μg/mL) were from R&D Systems, and chloroquine (100 μmol/L) was from Sigma Aldrich (Saint-Quentin Fallavier, France). Expression vectors for EWS-FLI1 (in Δ EB-78),
      • Guillon N.
      • Tirode F.
      • Boeva V.
      • Zynovyev A.
      • Barillot E.
      • Delattre O.
      The oncogenic EWS-FLI1 protein binds in vivo GGAA microsatellite sequences with potential transcriptional activation function.
      human (h)OSMR (in pcDNA3.1-hygro),
      • Lacreusette A.
      • Nguyen J.M.
      • Pandolfino M.C.
      • Khammari A.
      • Dreno B.
      • Jacques Y.
      • Godard A.
      • Blanchard F.
      Loss of oncostatin M receptor beta in metastatic melanoma cells.
      and hLIFR (in pLXSP-puro)
      • Blanchard F.
      • Wang Y.
      • Kinzie E.
      • Duplomb L.
      • Godard A.
      • Baumann H.
      Oncostatin M regulates the synthesis and turnover of gp130, leukemia inhibitory factor receptor α, and oncostatin M receptor β by distinct mechanisms.
      were previously described. All transfections with these plasmids were performed using Jet-PEI, according to manufacturer's recommendations (Polyplus-Transfection, Illkirch, France).

      Viable Cell Quantification

      The number of viable cells was quantified using the Vialight plus kit (Lonza, Basel, Switzerland). Depending on the cell line, cells were plated into 96-well plates at an initial density of 3000 to 6000 cells per well (for the highly and slowly growing cell lines, respectively) and cultured 72 hours with cytokine in 1% fetal bovine serum. The kit reagent was then added, and the luminescence was read in a Tristar Luminometer (Berthold Technologies, Wildbad, Germany).

      RNA Interference

      A673 Ewing sarcoma cell line, stably modified to express a doxycycline-inducible short hairpin RNA targeting EWS-FLI1 (shA673-1C), was previously described
      • Tirode F.
      • Laud-Duval K.
      • Prieur A.
      • Delorme B.
      • Charbord P.
      • Delattre O.
      Mesenchymal stem cell features of Ewing tumors.
      and maintained in a selective medium with blasticidin (10 μg/mL; Merck, Darmstadt, Germany) and phleomycin D1 (Zeocin, 100 μg/mL; Invitrogen Life Technologies, Cergy-Pontoise, France). Inhibition of EWS-FLI1 was induced by treatment with doxycycline (1 μg/mL; Sigma Aldrich).
      For small-interfering RNA (siRNA), cells were transfected with interferin (Polyplus-Transfection) and annealed siRNA (10 nmol/L; Ambion, Applied Biosystems, Courtaboeuf, France), according to the manufacturer's recommendations. The siRNA references were LIFR (s8170), OSMR (s17542), STAT3 (s743), LIF (s8168), and MYC (s9129).

      Flow Cytometry

      Cell Cycle

      Cells were seeded at 10,000 cells/cm2 at day 0 and either treated with hOSM (50 ng/mL) or left untreated at day 1 in 1% fetal bovine serum. Subconfluent cultures of adherent and nonadherent treated cells were released at day 2 and incubated with Ki-67–fluorescein isothiocyanate antibody (10 μg/mL; BD Pharmingen, San Diego, CA) and propidium iodide (50 μg/mL; Sigma Aldrich), as previously described.
      • David E.
      • Guihard P.
      • Brounais B.
      • Riet A.
      • Charrier C.
      • Battaglia S.
      • Gouin F.
      • Ponsolle S.
      • Bot R.L.
      • Richards C.D.
      • Heymann D.
      • Redini F.
      • Blanchard F.
      Direct anti-cancer effect of oncostatin M on chondrosarcoma.
      Cell cycle distribution was studied by flow cytometry (Cytomics FC500; Beckman Coulter, Villepinte, France) and analyzed with MultiCycle AV Software, Windows version (Phoenix Flow System, San Diego) and CXP Analysis software version 2.2. (Beckman Coulter).

      Receptors

      Cells were stained with 10 μg/mL mouse antibody targeting human LIFR or OSMR (AN-E1 and AN-V2; H. Gascan, Angers, France) or irrelevant IgG (R&D Systems). Receptors levels are expressed as follows: (LIFR or OSMR FL2 Mean) – (Isotype FL2 Mean).

      Western Blot Analysis

      Cell lysates were analyzed by using Western blot analysis, as previously described.
      • Chipoy C.
      • Brounais B.
      • Trichet V.
      • Battaglia S.
      • Berreur M.
      • Oliver L.
      • Juin P.
      • Redini F.
      • Heymann D.
      • Blanchard F.
      Sensitization of osteosarcoma cells to apoptosis by oncostatin M depends on STAT5 and p53.
      The membranes were blotted with antibodies to LIFR (C-19; Santa Cruz Biotechnologies, Santa Cruz, CA), STAT3 (BD Bioscience, San Diego), phosphorylated STAT3 (Tyr705 or Ser727; Cell Signaling Technologies, Beverly, CA), MYC (Cell Signaling Technologies), or actin (Sigma Aldrich).

      Semiquantitative and Real-Time PCR

      From cell lines, total RNA was extracted using NucleoSpin RNA II (Macherey-Nagel, Hoerd, France). From human tumors or xenografts, total RNA was extracted using TRIzol reagent (Invitrogen Life Technologies) after mechanical grinding with Turrax (IKA, Staufen, Switzerland). First-strand cDNA was synthesized from 1 μg of total RNA, using the ThermoScript RT-PCR System (Invitrogen Life Technologies). PCR was performed in SYBR Green buffer (Bio-Rad, Marnes la Coquette, France) with primers previously described,
      • Tirode F.
      • Laud-Duval K.
      • Prieur A.
      • Delorme B.
      • Charbord P.
      • Delattre O.
      Mesenchymal stem cell features of Ewing tumors.
      • Brounais B.
      • David E.
      • Chipoy C.
      • Trichet V.
      • Ferre V.
      • Charrier C.
      • Duplomb L.
      • Berreur M.
      • Redini F.
      • Heymann D.
      • Blanchard F.
      Long term oncostatin M treatment induces an osteocyte-like differentiation on osteosarcoma and calvaria cells.
      • David E.
      • Guihard P.
      • Brounais B.
      • Riet A.
      • Charrier C.
      • Battaglia S.
      • Gouin F.
      • Ponsolle S.
      • Bot R.L.
      • Richards C.D.
      • Heymann D.
      • Redini F.
      • Blanchard F.
      Direct anti-cancer effect of oncostatin M on chondrosarcoma.
      • Guihard P.
      • Danger Y.
      • Brounais B.
      • David E.
      • Brion R.
      • Delecrin J.
      • Richards C.D.
      • Chevalier S.
      • Redini F.
      • Heymann D.
      • Gascan H.
      • Blanchard F.
      Induction of osteogenesis in mesenchymal stem cells by activated monocytes/macrophages depends on oncostatin M signaling.
      except as follows: SOX2, 5′-GTATCAGGAGTTGTCAAGGCAGAG-3′ (forward) and 5′-TCCTAGTCTTAAAGAGGCAGCAAAC-3′ (reverse); Nanog, 5′-ATGCCTCACACGGAGACTGT-3′ (forward) and 5′-AAGTGGGTTGTTTGCCTTTG-3′ (reverse); and c-myc, 5′-CACCAGCAGCGACTCTGA-3′ (forward) and 5′-GATCCAGACTCTGACCTTTTGC-3′ (reverse).

      cDNA Array

      Available public microarray data for gastrointestinal stromal tumor (GEO accession number GSE20710), renal cell carcinomas (GSE23629), lymphomas (GSE12453), leukemias (GSE14062), lung cancers (GSE18842), colorectal cancers (GSE20916), OSs (GSE14827), ESs (GSE12102), rhabdomyosarcoma,
      • Williamson D.
      • Missiaglia E.
      • de Reynies A.
      • Pierron G.
      • Thuille B.
      • Palenzuela G.
      • Thway K.
      • Orbach D.
      • Lae M.
      • Freneaux P.
      • Pritchard-Jones K.
      • Oberlin O.
      • Shipley J.
      • Delattre O.
      Fusion gene-negative alveolar rhabdomyosarcoma is clinically and molecularly indistinguishable from embryonal rhabdomyosarcoma.
      synovial sarcomas (GSE20196), adult sarcomas (GSE21050), neuroblastomas (GSE12460), medulloblastomas (GSE12992), normal tissues (GSE3526), MSCs (GSE10315), and gliomas (EBI array express accession number E-MEXP-1507) were used. All microarray data were simultaneously normalized using the gcrma package, version 2.22.0, and Brainarray Entrez gene CDF, version 14.1,
      • Dai M.
      • Wang P.
      • Boyd A.D.
      • Kostov G.
      • Athey B.
      • Jones E.G.
      • Bunney W.E.
      • Myers R.M.
      • Speed T.P.
      • Akil H.
      • Watson S.J.
      • Meng F.
      Evolving gene/transcript definitions significantly alter the interpretation of GeneChip data.
      in the R 2.12.0 environment (http://www.r-project.org); quality assessment was based on relative log expression and normalized unscaled SEs.

      Luciferase Reporter Assay

      For OSMR promoter activity, cells were plated in 24-well plates and transiently transfected using Jet-PEI (Polyplus-Transfection) with empty or OSMR promoter-driven pGL3-Basic, which encodes Firefly luciferase as a reporter (500 ng), together with a Renilla luciferase vector (pRL-TK) as an internal control (150 ng).
      • Lacreusette A.
      • Nguyen J.M.
      • Pandolfino M.C.
      • Khammari A.
      • Dreno B.
      • Jacques Y.
      • Godard A.
      • Blanchard F.
      Loss of oncostatin M receptor beta in metastatic melanoma cells.
      Luciferase activity was measured 48 hours after cell transfection with the Dual Luciferase reporter assay system (Promega, Madison, WI) in a Tristar Luminometer (Berthold Technologies). OSMR Promoter activity was calculated as follow: (Firefly Luciferase Activity/Renilla Luciferase Activity on pGL3OSMR Sample)/(Firefly Luciferase Activity/Renilla Luciferase Activity on Empty pGL3 Sample).
      For STAT3 activity, cells were plated at day 0 and transfected with indicated siRNA at day 1. At day 2, they were transiently transfected using Jet-PEI with pSiem (plasmid that encodes Firefly luciferase, under the control of a promoter containing multiple binding sites for STAT3) as a reporter (500 ng), together with pRL-TK (150 ng) and either treated with OSM or left untreated at day 3. STAT3 activity was measured at day 4 by luciferase assay, and induction of STAT3 activity by OSM was calculated as follows: (Firefly Luciferase Activity/Renilla Luciferase Activity on Samples Treated with OSM)/(Firefly Luciferase Activity/Renilla Luciferase Activity on Samples without OSM).

      Quantitative ChIP Assay

      A chromatin immunoprecipitation (ChIP) assay was performed, as previously described,
      • Lacreusette A.
      • Nguyen J.M.
      • Pandolfino M.C.
      • Khammari A.
      • Dreno B.
      • Jacques Y.
      • Godard A.
      • Blanchard F.
      Loss of oncostatin M receptor beta in metastatic melanoma cells.
      by the Magnify ChIP system (Invitrogen Life Technologies) using antibody to FLI1 (C-19; Santa Cruz Biotechnologies) or control rabbit IgG (R&D Systems) and real-time PCR (Bio-Rad).

      IHC and Immunocytochemistry

      Cells were fixed on a glass slide by centrifugation in Cytospin (Microm Microtech, Francheville, France). Tumor samples were fixed, decalcified by 10% nitric acid, embedded in paraffin, and sectioned (5 μm thick). Slides were incubated for 1 hour at 37°C with primary antibody targeting OSM (goat anti-mouse OSM; R&D Systems, 5 μg/mL), LIF (rabbit polyclonal anti-human LIF from Dr. Anne Godard, INSERM, Unit 892, Nantes, France, 125 μg/mL), or irrelevant IgG. Immunodetection was performed using amino ethyl-carbazole (AEC), and sections were counterstained with Mayer's hematoxylin, as described.
      • David E.
      • Guihard P.
      • Brounais B.
      • Riet A.
      • Charrier C.
      • Battaglia S.
      • Gouin F.
      • Ponsolle S.
      • Bot R.L.
      • Richards C.D.
      • Heymann D.
      • Redini F.
      • Blanchard F.
      Direct anti-cancer effect of oncostatin M on chondrosarcoma.

      Statistical Analysis

      Results were analyzed with an unpaired t-test or the U-test using GraphPad InStat version 3.02 or Prism3 software version 3.03 (La Jolla, CA). Results are given as the mean ± SD or SEM and were considered significant when P < 0.05. Correlation analyses were performed with Pearson's product-moment correlation coefficient r and a two-tailed test.

      Results

      OSM Enhances ES Cell Proliferation But Decreases OS and CS Cell Proliferation

      Different human primary bone tumor cell lines were treated with the nine cytokines of the IL-6 family [IL-6, IL-11, IL-27, IL-31, OSM, LIF, ciliary neurotrophic factor, cardiotrophin-1 (CT-1), and cardiotrophin-like cytokine] for 3 days, and the number of viable cells was determined. OSM reduced the number of viable MG63 OS cells, followed by IL-6 combined with its soluble receptor (sIL-6R), IL-27, and IL-6 alone, with the other cytokines being inactive (Figure 1A). OSM induced opposite effects on RDES, SKES1, or TC32 ES cell lines by increasing the number of viable cells, followed by IL-6 + sIL-6R, LIF, and CT-1. IL-27 significantly reduced the number of viable SKES1 and TC32 cells but had no effect on RDES cells (Figure 1A; see also Supplemental Figure S1A at http://ajp.amjpathol.org). The effect of OSM was then studied on a panel of 10 ES cell lines and compared with IL-6, LIF, and CT-1, for example. OSM significantly enhanced the number of viable cells on nine ES cell lines (Figure 1B), with a mean induction of 30% (Figure 1C). In contrast, OSM reduced the number of viable OS and CS cells, with a mean inhibition of 18% (Figure 1C; 12 cells lines were tested). IL-6 + sIL-6R, LIF, and CT-1 also significantly induced proliferation of most of our ES cell lines (n = 10), but, on average, they appeared less active than OSM (Figure 1C). In MG63 and RDES cells, the maximal induction or reduction of viable cells was reached with 10 ng/mL of OSM, with an inhibitory concentration of 50% or an EC50 between 0.1 and 1 ng/mL. On a 3-day culture, OSM treatment during the first 24 hours was sufficient to induce the number of viable ES cells (see Supplemental Figure S1, B and C, at http://ajp.amjpathol.org).
      Figure thumbnail gr1
      Figure 1OSM enhances proliferation of ES cells. A: MG63, RDES, and SKES1 cells are treated, as indicated, with different IL-6–type cytokines (50 ng/mL, except LIF, at 100 ng/mL) for 3 days, and the number of viable cells is assessed using the Vialight kit. B: The 10 Ewing sarcoma cell lines are treated for 3 days with OSM, and cell viability is assessed. C: A total of 22 cell lines are treated for 3 days with the indicated cytokines, and the number of viable cells is assessed. Cell lines are divided into two groups: OS + CS (n = 12) and ES (n = 10). D: The percentages of cells in the different phases of the cell cycle for MG63, RDES, and SKES1 cells treated for 1 day with OSM are shown using propidium iodide (PI)/Ki-67 double staining. E: The indicated cell lines are treated with OSM for 0.5 to 24 hours. mRNA expression of MYC is assessed by real-time PCR (n = 3 for each condition). F: The indicated cell lines were treated with OSM for 0.25 to 6 hours, and cell lysates are analyzed by using Western blot analysis for MYC and actin, as indicated. Unless otherwise stated, all assays are performed three times (n = 5). The results are expressed as the mean ± SD, except in C, where they are expressed as the mean ± SEM. Statistical analysis is performed using a t-test. *P < 0.05, **P < 0.01, and ***P < 0.0001 versus the control without cytokine.
      To determine whether the induction of viable ES cells with OSM was linked to induced proliferation or reduced cell death, ES cells were treated with OSM for 15 days, with regular counting of the living and dead cells. We never observed modification of cell death with OSM, whereas the increase of viable cell number reached 150% and 120% for RDES and SKES1, respectively, at day 15. In contrast, OSM reduced by 80% the number of viable MG63 cells at day 15. LIF had no effect on MG63 cells, and induction of viable RDES and SKES1 cells reached only 50% and 80%, respectively (see Supplemental Figure S1, D and E, at http://ajp.amjpathol.org). Consistently, 1 day of OSM treatment significantly decreased the number of RDES cells in the G0 phase of the cell cycle and increased the number of cells in the S phase. Similar results were observed in OSM-treated SKES1 cells (ie, fewer cells in G0 and more cells in the S-G2/M phase, whereas MG63 cells accumulated in the G0/G1 phase) (Figure 1D; see also Supplemental Figure S1F at http://ajp.amjpathol.org).
      To identify genes potentially implicated in this opposite cell cycle regulation by OSM, we first analyzed the mRNA expression of several transcription factors linked to proliferation/differentiation. OS and CS cell lines expressed higher levels of the mesenchymal differentiation transcription factors, CBFA1, SOX9, C\EBPδ, and peroxisome proliferator-activated receptor-γ, whereas ES cell lines expressed stemness transcription factors, such as ID2, SOX2, NANOG, and MYC (see Supplemental Figure S1G at http://ajp.amjpathol.org). On OSM treatment, only MYC expression was induced in ES cell lines (peak at 1 hour; Figure 1E), with other stemness markers being unchanged (ID2 and SOX2) or reduced (NANOG) and differentiation markers being unchanged. In contrast, OSM induced expression of several differentiation markers in OS or CS cell lines (C/EBPβ, C/EBPδ, and CBFA1), leaving the stemness transcription factors at a low level of expression, except for MYC (see Supplemental Figure S1H at http://ajp.amjpathol.org; data not shown). Indeed, OSM transiently induced mRNA expression of MYC in OS cells, with the expression level remaining, however, lower than in ES cells (Figure 1E). At the protein level, we confirmed the induced MYC expression by OSM in ES cell lines (SKES1, RDES, and TC71), which is severalfold higher than in OS (MG63 and SaOS2) or CS (CAL78) cell lines (Figure 1F).

      Expression of LIFR and OSMR Subunits in ES Cell Lines

      Because the inverse effect of OSM on ES versus OS or CS cell proliferation could rely on different OSM receptor recruitment, the expression of OSM receptor subunits was analyzed. The OSMR transcript was fivefold less abundant in ES cells compared with OS + CS cells (P < 0.0001), whereas the LIFR mRNA was slightly more expressed, although not significantly (Figure 2A). This result was in accordance with decreased OSMR promoter activity in ES cells (Figure 2B; 15 cell lines that could be efficiently transfected were used). At the protein level, cell surface receptor analysis by flow cytometry showed that ES cells expressed threefold more LIFR than OS + CS cells, whereas expression of OSMR was similar in both groups (Figure 2, C and D). By using Western blot analysis on total cell lysates, ES cells appeared to express a higher level of the mature LIFR form (form 1 in Figure 2E). Thus, ES cells have a higher LIFR/OSMR expression ratio than OS or CS cells. The OSMR appeared to be down-regulated at the transcriptional level, whereas the LIFR was up-regulated mainly at the post-transcriptional level.
      Figure thumbnail gr2
      Figure 2ES cells express low levels of OSMR and high levels of LIFR. A: mRNA expression of LIFR and OSMR in 19 primary bone tumor cell lines is assessed by real-time PCR (n = 3 for each cell line). Results are represented by box (median and the 25th and 75th percentiles) and whiskers (largest and smallest values). Statistical difference in ES versus OS + CS is performed using the U-test. B: Fifteen cell lines are transiently transfected with pGL3OSMR or empty vector. Luciferase activity is measured 48 hours after cell transfection. OSMR promoter activity is calculated as indicated in Materials and Methods. Line, mean. C: Histograms represent the protein level of LIFR and OSMR versus control, with irrelevant IgG in RDES and MG63 cells assessed by flow cytometry. D: The cell surface protein level of LIFR and OSMR in 19 primary bone tumor cell lines is assessed by flow cytometry (n = 3 for each cell line). The results and statistical analysis are presented as in A. E: The expression of LIFR protein in 13 cell lines is analyzed by using Western blot analysis. Two forms of LIFR are found: 1, mature form; 2, immature form. *P < 0.05, **P < 0.01, and ***P < 0.0001. ns, not significant.

      Induction of Proliferation by OSM in ES Cells Depends on gp130, LIFR, OSMR, STAT3, and MYC

      To identify the receptors and pathways leading to growth induction or inhibition by OSM, specific neutralizing antibody or small-interfering RNA was used. A neutralizing antibody targeting gp130, the coreceptor for LIFR and OSMR, nearly completely prevented the cytostatic effect of OSM on MG63 cells and induction of proliferation on RDES and SKES1 cells (Figure 3A). Next, LIFR and OSMR down-regulation by small-interfering (si)LIFR and siOSMR was validated by flow cytometry and real-time PCR (Figure 3B; see also Supplemental Figure S2, A and B, at http://ajp.amjpathol.org). In OS cells (MG63), siLIFR had no effect on proliferation, whereas siOSMR totally prevented the growth inhibition by OSM (Figure 3D). In ES cells (RDES and SKES1), siLIFR or siOSMR only partly prevented the induced proliferation by OSM, this effect being not even statistically significant in RDES cells. However, a combination of both siRNA totally prevented the growth-stimulatory effect of OSM in these two cell lines (Figure 3D). We also tried to re-express LIFR in OS cells and OSMR in ES cells. We succeeded in obtaining stable MG63 and SKES1 clones overexpressing LIFR and OSMR mRNA, respectively, but expression of these receptor subunits was poorly induced at the cell surface (only 1.5-fold to twofold), again suggesting the presence of strong post-transcriptional regulatory mechanisms. Consequently, the growth-inhibitory or growth-stimulatory effect of OSM was not affected (see Supplemental Figure S2C at http://ajp.amjpathol.org; data not shown).
      Figure thumbnail gr3
      Figure 3Implication of receptor subunits, STAT3, and MYC in growth control by OSM. A: MG63, RDES, and SKES1 cells are pretreated 1 hour with antibody targeting gp130 or control IgG antibody and treated for 3 days with OSM. The number of viable cells is assessed using the Vialight kit. Assays are performed three times (n = 5). B: RDES cells are transfected with siRNA targeting LIFR, OSMR, or siCT. Histograms represent LIFR and OSMR protein level versus control with irrelevant IgG by flow cytometry 48 hours after transfection. C: RDES cells are either transfected with siRNA targeting STAT3 and siCT or left untreated. The expression of STAT3 protein is analyzed by using Western blot analysis 48 hours after transfection. D: MG63, RDES, and SKES1 cells are transfected with the indicated siRNA in 24-well plates; 48 hours later, cells are released, plated in 96-well plates, and treated with OSM or left untreated for 3 days in 1% fetal bovine serum. The number of viable cells is assessed using the Vialight kit. Assays are performed three times (n = 5). E: MG63, SKES1, RDES, and TC32 cells are transfected with the indicated siRNA. At day 1, cells are transiently transfected with a STAT3 reporter (pSiem-Luc) and treated with OSM at day 3 or left untreated. STAT3 activity is measured at day 4 by luciferase assay and expressed as indicated in Materials and Methods. F: SKES1 and RDES cells are transfected with siRNA targeting STAT3, MYC, or siCT or left untreated. At 2 days later, cells are treated with OSM for 6 to 24 hours, and expression of MYC protein is analyzed by using Western blot analysis. Signal quantification of MYC on actin is assessed using Genetools software version 4.02 (Syngene, Cambridge, UK). G: The indicated cells are transfected with the si–MYC, treated with OSM, and analyzed for cell viability, as in D. The assays are performed two times (n = 5). All viability results are expressed as the mean ± SD, relative to the culture with corresponding antibody or siRNA, without OSM, except in G, where they are expressed relative to the siCT without OSM. Statistical analysis compares the effect of OSM in the presence of antibody or siRNA with the effect of OSM in the presence of irrelevant IgG or siCT. *P < 0.05, **P < 0.01, and ***P < 0.0001. Ab, antibody; ns, not significant.
      STAT3 down-regulation by siSTAT3 was next validated by using Western blot analysis (Figure 3C). As observed with siLIFR and siOSMR, siSTAT3 completely prevented growth inhibition by OSM on MG63 cells and induction of proliferation on RDES and SKES1 cells (Figure 3D). The link between LIFR/OSMR, STAT3, and cell cycle control was further established by analyzing activation of STAT3 by OSM using a Luciferase reporter driven by a STAT3-dependent promoter. siLIFR had no effect on STAT3 activity in MG63 cells, whereas siOSMR totally prevented STAT3 activation by OSM. In ES cells (SKES1, RDES, and TC32), siLIFR strikingly increased STAT3 activation by OSM, whereas siOSMR partly decreased it. As observed in the proliferation test, a combination of siLIFR and siOSMR totally prevented STAT3 activation by OSM in SKES1 and TC32 cells (Figure 3E). In RDES cells, the synergistic inhibitory effect between the two siRNAs was less pronounced, in correlation with a smaller reduction of receptor expression (see Supplemental Figure S2A at http://ajp.amjpathol.org).
      From this set of experiments, it also appeared that induction of STAT3 activity by OSM was significantly greater in ES than in OS or CS cell lines (Figure 3E; see also Supplemental Figure S2D at http://ajp.amjpathol.org). Phosphorylation at Tyr705 and Ser727 was important for full activation of this transcription factor, but we observed a similar induction of STAT3 phosphorylation by OSM in OS, CS, and ES cell lines (see Supplemental Figure S2E at http://ajp.amjpathol.org). Similarly, there was no difference in activation of STAT1, STAT5, or extracellular signal–regulated kinase 1/2, and Akt was even more activated by OSM in OS and CS than in ES cells (data not shown). From these results, we hypothesized that transcriptional co-activators increased STAT3 activity in ES cells.
      We next validated MYC siRNA (Figure 3F; see also Supplemental Figure S3A at http://ajp.amjpathol.org) and observed that MYC was also implicated in the OSM-induced proliferation in SKES1 and RDES ES cells (Figure 3G). Again, down-regulation of MYC was lower in RDES than in SKES1 cells, in relation with a lower prevention of growth stimulation by OSM. The si–MYC also significantly reduced the basal proliferation rate in these two cell lines, but only by 10% to 15% (Figure 3G), and no cell death was observed (data not shown). In sharp contrast, the si–MYC did not affect growth inhibition by OSM in MG63 cells (Figure 3G). To confirm that MYC was induced by STAT3, we again transfected the siSTAT3 in SKES1 and RDES cells and observed that induction of MYC by OSM was reduced (RDES) or totally prevented (SKES1) by the STAT3 knockdown (Figure 3F). These results suggested that OSM induced ES cell proliferation through an LIFR/OSMR–STAT3–MYC pathway.

      EWS-FLI1 Prevents Mesenchymal Differentiation and LIF Expression, Resulting in OSMR Down-Regulation and LIFR Overexpression

      The regulatory role of EWS-FLI1 on OSM signaling was evaluated in shA673-1C ES cell lines expressing a doxycycline-inducible short hairpin RNA targeting EWS-FLI1. This model was not suitable for proliferation studies because A673 cells were poorly responsive to the growth-stimulatory effect of OSM (10%/20%; Figure 1B), and knockdown of EWS-FLI1 in these cells rapidly inhibited proliferation (data not shown). Indeed, doxycycline treatment rapidly and stably decreased the EWS-FLI1 mRNA level (−60%), which allowed the cells to differentiate into osteoblasts and chondrocytes, and, thus, to express master transcription factors that control the mesenchymal differentiation, such as CBFA1, SOX9, and C/EBPδ. The stemness transcription factors, SOX2, NANOG, and MYC, were decreased (Figure 4A).
      • Tirode F.
      • Laud-Duval K.
      • Prieur A.
      • Delorme B.
      • Charbord P.
      • Delattre O.
      Mesenchymal stem cell features of Ewing tumors.
      After 10 days of doxycycline treatment, shA673-1C cells expressed twofold more OSMR and 1.6-fold more LIFR transcript (Figure 4A). At the cell surface protein level, doxycycline treatment only slightly increased OSMR (10%) but significantly decreased LIFR (−40%; Figure 4B). The OSMR promoter assay confirmed transcriptional repression because doxycycline treatment increased OSMR promoter activity in shA673-1C cells (Figure 4C), and inversely transient transfection of EWS-FLI1 in MG63 or MSCs reduced OSMR promoter activity (see Supplemental Figure S3B at http://ajp.amjpathol.org). Quantitative ChIP assays revealed that EWS-FLI1 directly binds to the OSMR promoter in a region located immediately upstream the CpG island in SKES1 (Figure 4, D and E), RDES, and A673 cells (data not shown). This region contained several GGAA potential binding sites for EWS-FLI1 and previously unidentified negative regulatory elements.
      • Lacreusette A.
      • Nguyen J.M.
      • Pandolfino M.C.
      • Khammari A.
      • Dreno B.
      • Jacques Y.
      • Godard A.
      • Blanchard F.
      Loss of oncostatin M receptor beta in metastatic melanoma cells.
      • Guillon N.
      • Tirode F.
      • Boeva V.
      • Zynovyev A.
      • Barillot E.
      • Delattre O.
      The oncogenic EWS-FLI1 protein binds in vivo GGAA microsatellite sequences with potential transcriptional activation function.
      Figure thumbnail gr4
      Figure 4OSMR and LIFR are regulated by distinct mechanisms. A: shA673-1C cells are treated for 2 to 10 days with doxycycline. mRNA expression of indicated genes is assessed by real-time PCR (n = 3). The results represent the percentage of mRNA expression versus untreated cells. B: shA673-1C cells are treated for 2 to 10 days with doxycycline. Protein expression of LIFR and OSMR is assessed by flow cytometry. The results represent the percentage of receptor expression versus untreated cells. C: shA673-1C cells are treated with doxycycline (DOX) for 72 hours or left untreated. Cells are released, plated in 24-well plates, transfected with pGL3OSMR or empty vector, and maintained with or without doxycycline treatment. Luciferase activity is measured 48 hours after cell transfection. OSMR promoter activity is calculated as indicated in Materials and Methods. D: A ChIP assay is performed on SKES1 cells using 1 or 10 μg of antibody to FLI1 or control IgG, as indicated. Five regions of the OSMR promoter are analyzed by real-time PCR and expressed as percentage of input control. E: Schematic map of the OSMR promoter (square exon 1), showing the CpG island, transcription start site (arrow), and regions analyzed by ChIP assay. F: The protein level of LIFR and OSMR in SKES1 and RDES, treated with LIF for 3 days or left untreated, is assessed by flow cytometry. The results represent the percentage of each receptor expression versus untreated cells. G: RDES and SKES1 cells are pretreated or not with chloroquine for 1 hour and treated with LIF for 2 hours. The expression of LIFR protein (1, mature form; 2, immature form) is analyzed by using Western blot analysis. The signal quantification of the LIFR mature form on actin is assessed using Genetools software. Ab, antibody.
      Because the LIFR was mainly regulated at the post-transcriptional level and LIF induced internalization and degradation of LIFR in several cell types,
      • Blanchard F.
      • Duplomb L.
      • Wang Y.
      • Robledo O.
      • Kinzie E.
      • Pitard V.
      • Godard A.
      • Jacques Y.
      • Baumann H.
      Stimulation of leukemia inhibitory factor receptor degradation by extracellular signal-regulated kinase.
      we next asked whether similar regulatory mechanisms existed in ES cells. First, in shA673-1C cells, knockdown of EWS-FLI1 with doxycycline induced a 15-fold increase in LIF mRNA level, whereas OSM was lower than the detection limit (Figure 4A and data not shown). Second, LIF treatment on RDES or SKES1 cells decreased LIFR and increased OSMR protein on the cell surface (Figure 4F). This LIFR down-regulation by LIF was also observed using Western blot analysis, with a disappearance of the mature LIFR (form 1). Pretreatment with chloroquine (a lysosome inhibitor) prevented this LIFR down-regulation by LIF (Figure 4F).
      These results suggested that ES cells expressed a low level of OSMR transcript because EWS-FLI1 directly binds to and represses the OSMR promoter. However, they also raised the possibility that LIF was poorly expressed in ES cells, also resulting in a low level of OSMR. Inversely, low LIF expression could lead to a high level of LIFR at the cell surface through altered internalization/lysosomal degradation.

      Expression of OSM, LIF, and Their Receptors in Vitro and in Vivo

      To demonstrate more convincingly that ES cells expressed only a low level of LIF, real-time PCR analyses were performed in cell lines, patient tumor biopsy specimens, and human tumors developed in nude mice (xenografts). LIF mRNA expression was 100-fold lower in ES compared with OS or CS cell lines, with similar results obtained with patient biopsy specimens or xenografts (Figure 5A). Inflammatory cytokines, such as IL-1β or tumor necrosis factor-α, enhanced LIF expression in OS cell lines, also suggesting an induced production in vivo, whereas in ES cell lines, the LIF mRNA level remained low (see Supplemental Figure S3C at http://ajp.amjpathol.org). A higher expression of LIF by OS and CS cell lines in culture, compared with ES, was confirmed at the protein level by immunostaining (Figure 5B) and enzyme-linked immunosorbent assay (see Supplemental Figure S3D at http://ajp.amjpathol.org). By immunohistochemistry (IHC) on patient tumor biopsy specimens, OS cells appeared positive for LIF, whereas in ES, LIF staining appeared lower and localized to stromal or vascular areas (Figure 5C and data not shown). In two OS cell lines that constitutively expressed LIF (MG63 and SaOS2), treatment with siRNA targeting LIF improved LIFR protein expression, but the OSMR was not altered. In contrast, LIFR expression was not induced by the siLIF in RDES ES cells that did not express LIF. In MG63 cells transfected with the siLIF, the basal proliferation rate and the growth-inhibitory effect of OSM were not altered, despite the increased expression of LIFR (see Supplemental Figure S3, E and F, at http://ajp.amjpathol.org). These results confirmed that ES cells expressed only a low level of LIF, with its receptor LIFR, therefore, being much more stable at the cell surface. Although exogenously added LIF induced OSMR expression in ES cells, knockdown of LIF did not alter OSMR expression in OS cells, indicating that other OSMR inducers were implicated. One candidate was IL-6, which was also highly expressed in OSs and CSs compared with ESs (see Supplemental Figure S3D at http://ajp.amjpathol.org).
      Figure thumbnail gr5
      Figure 5OSM and LIF expression in patient tumors, cell lines, and xenografts. A: LIF and OSM mRNA expression is assessed by real-time PCR on human biopsy specimens (n = 9), xenograft fragments of bone tumors in nude mice (n = 6), and cell lines (n = 19). Statistical analysis to compare ES with OS + CS cell lines was performed using the U-test. ***P < 0.0001. B: Immunostaining of LIF on cells after cytospin is presented in MG63 and RDES cell lines. C: Immunostaining of LIF (arrow, example of positive staining) on paraffin sections is presented in one representative OS and one ES biopsy specimen. D: hOSM and murine (m)OSM mRNA expression is assessed on human biopsy specimens (n = 9) and xenografts (n = 5) by semiquantitative PCR. GAPDH, glyceraldehyde-3-phosphate dehydrogenase. E: Immunostaining of OSM (arrow) on paraffin sections is presented in three ES biopsy specimens (one negative control is shown in a serial section for patient 3). Original magnification, ×400 (AE).
      OSM was also a known inducer of OSMR, and its major role on ES cell proliferation prompted us to study its expression in vitro and in vivo. Real-time PCR, performed on human tumor biopsy specimens, demonstrated a low, but significant, expression of OSM in three of four ES and four of five OS cells (Figure 5A). In contrast, ES, OS, or CS cell lines in culture expressed 10- to 100-fold less OSM mRNA (except one CS cell line), suggesting either an induced production in vivo or a production by nontumor cells. To discriminate between these possibilities, OSM expression was next analyzed in human tumors developed in nude mice (xenografts). Species-specific PCR allowed us to quantify human OSM (produced by tumor cells) and murine Osm (produced by host cells). Human OSM was not detected in five different ES or OS xenografts, whereas murine Osm was detected in four of five tumors (Figure 5D). By IHC on patient biopsy specimens, OSM was detected in three of three ES, three of three OS, and one of three CS cells studied. OSM staining was observed close to or inside vascular vessels, with one ES biopsy specimen (patient 1) showing positive myofibroblasts (Figure 5E). Cancer cells or macrophages were never positive.
      The expression of LIF, OSM, their receptor subunits, and MYC was also analyzed in a large panel of patient biopsy specimens using cDNA arrays. The OSMR transcript appeared to be expressed threefold less in ESs compared with OSs (P = 4 × 10−7), whereas LIFR mRNA was expressed at a similar level in these two types of tumor (Figure 6A). When compared with other tumors (pediatric or not) or normal tissues, the LIFR/OSMR analysis revealed a striking homology between ES and other tumors with (neuro)ectodermic features (glioma, medulloblastoma, or neuroblastoma) and tissues of the central nervous system (Figure 6B; see also Supplemental Figure S4 at http://ajp.amjpathol.org). OSs, most other tumors with mesodermic features (liposarcoma, leiomyosarcoma, gastrointestinal stromal tumor, and undifferentiated sarcoma), MSCs, and carcinomas (lung cancer and renal cell carcinoma) appeared to express a significantly higher level of OSMR. The expression of LIFR was more variable within each group, with the highest expression being observed in ESs, gliomas, OSs, and renal cell carcinomas. In correlation with the low OSMR expression, the expression of LIF and IL-6 appeared low in ectodermic tumors compared with mesodermic tumors or carcinomas. There was no significant difference in OSM mRNA expression between ectodermic tumors, mesodermic tumors, and carcinomas, but its expression was higher in hematopoietic cells (see Supplemental Figure S4 at http://ajp.amjpathol.org). When accounting for all samples (n = 943), we found that LIF or IL-6 expression positively correlated with OSMR expression (r = 0.491, P < 0.0001 for LIF; r = 0.469, P < 0.0001 for IL-6), but not with LIFR expression (r = −0.02, P = 0.5 for LIF; r = 0.0001, P = 0.99 for IL-6). In contrast, there was no positive correlation between OSM and OSMR or LIFR expression. For these three cytokines, we noticed an important number of outliers that could be related to the inflammatory status of these patients. For MYC, the expression was high in all tumors and there was no difference between ectodermic tumors, mesodermic tumors, carcinomas, and hematopoietic cells. However, MYC expression was significantly higher in ESs than in all other tumors or normal tissues (see Supplemental Figure S4 at http://ajp.amjpathol.org). In conclusion, ESs and other ectodermic tumors had a higher LIFR/OSMR ratio that correlated with a lower level of LIF and IL-6. ESs, but not other ectodermic tumors, also had a characteristic high level of MYC.
      Figure thumbnail gr6
      Figure 6OSMR and LIFR expression in patient tumors. A: mRNA expression of LIFR and OSMR in human biopsy specimens of OS and ES is assessed by cDNA array. Results are represented by box (median and 25th and 75th percentiles) and whiskers (largest and smallest values). Outlier values are also presented (circles). Statistical difference of expression in ES versus OS is performed using a t-test (P = 4 × 10−7). Red line, threshold under which the gene is considered not expressed. B: mRNA expression of LIFR and OSMR is assessed by cDNA array on different human tumors or normal tissues. Dot, median expression of both receptors. Red, tumors or normal tissues with an ectodermic origin; blue, those with a mesodermic origin; green, hematopoietic cells; and black, carcinomas. GIST, gastrointestinal stromal tumor; RMS, rhabdomyosarcoma; Undiff., undifferentiated. C: Proposed mechanisms of action of OSM to enhance or inhibit proliferation in OSs, CSs, and ESs, based on our study.

      Discussion

      Herein, we show, for the first time to our knowledge, that cytokines of the IL-6 family can enhance proliferation of ES cells, whereas they are cytostatic factors for two closely related bone tumors, OS and CS. We mainly focus on OSM; when comparing the activity of the different cytokines, OSM appeared as the most efficient. The induction of ES cell proliferation by OSM is observed in 9 of 10 cell lines, indicating an extremely general and conserved phenomenon. In fact, OSM activates quiescent cells (G0 phase) to enter into the cell cycle and to replicate (S phase). In contrast, OSM induces a cell cycle blockage in G0/G1 or in S/G2/M phases on OS or CS cells.
      • Brounais B.
      • Chipoy C.
      • Mori K.
      • Charrier C.
      • Battaglia S.
      • Pilet P.
      • Richards C.D.
      • Heymann D.
      • Redini F.
      • Blanchard F.
      Oncostatin M induces bone loss and sensitizes rat osteosarcoma to the antitumor effect of Midostaurin in vivo.
      • Chipoy C.
      • Brounais B.
      • Trichet V.
      • Battaglia S.
      • Berreur M.
      • Oliver L.
      • Juin P.
      • Redini F.
      • Heymann D.
      • Blanchard F.
      Sensitization of osteosarcoma cells to apoptosis by oncostatin M depends on STAT5 and p53.
      • David E.
      • Guihard P.
      • Brounais B.
      • Riet A.
      • Charrier C.
      • Battaglia S.
      • Gouin F.
      • Ponsolle S.
      • Bot R.L.
      • Richards C.D.
      • Heymann D.
      • Redini F.
      • Blanchard F.
      Direct anti-cancer effect of oncostatin M on chondrosarcoma.
      Therefore, this study demonstrates an opposite role for OSM on three different primary bone tumors and elucidates which factors influence the tumor response to OSM.
      OSM inhibits proliferation in OSs by binding only to OSMR + gp130 (type II OSM receptor), the LIFR protein (type I receptor) being presumably not sufficiently expressed at the cell surface. In ES cells, which express high levels of LIFR mRNA and protein, knockdown of LIFR does not lead to inhibition of proliferation by OSM. Therefore, we can reject the hypothesis of an antiproliferative role of OSMR versus a proproliferative role of LIFR. Rather, OSM enhances ES cell proliferation through recruitment of both types I and II receptors. In accordance, LIF and CT-1, which recruit the LIFR but not the OSMR, induce ES proliferation, but to a lower extent than OSM. IL-6, mainly when combined with its soluble agonistic receptor, sIL-6R, can inhibit OS and CS proliferation and induce ES proliferation. IL-27 is the only IL-6–type cytokine able to inhibit proliferation in OS, CS, and some ES cell lines. To define the exact role of IL-6 and IL-27 on proliferation, additional experiments are needed.
      STAT3 was often shown as an important key in OSM or LIF signaling, inhibition of proliferation,
      • Kritikou E.A.
      • Sharkey A.
      • Abell K.
      • Came P.J.
      • Anderson E.
      • Clarkson R.W.
      • Watson C.J.
      A dual, non-redundant, role for LIF as a regulator of development and STAT3-mediated cell death in mammary gland.
      • Matsui T.
      • Kinoshita T.
      • Hirano T.
      • Yokota T.
      • Miyajima A.
      STAT3 down-regulates the expression of cyclin D during liver development.
      or induction of proliferation.
      • Li Q.
      • Zhu J.
      • Sun F.
      • Liu L.
      • Liu X.
      • Yue Y.
      Oncostatin M promotes proliferation of ovarian cancer cells through signal transducer and activator of transcription 3.
      • Daheron L.
      • Opitz S.L.
      • Zaehres H.
      • Lensch M.W.
      • Andrews P.W.
      • Itskovitz-Eldor J.
      • Daley G.Q.
      LIF/STAT3 signaling fails to maintain self-renewal of human embryonic stem cells.
      We reinforce this notion of a dual role of STAT3 because siRNA targeting STAT3 completely prevents growth inhibition in OSs and induction of proliferation in ESs. In addition, the induction of STAT3 transcriptional activity by OSM (measured using a luciferase reporter) is greater in ESs than in OSs or CSs, suggesting that STAT3 could either inhibit or enhance proliferation, depending on the intensity of its activation. Similarly, the duration of STAT3 activation dictated either growth arrest or proliferation.
      • Lu Y.
      • Fukuyama S.
      • Yoshida R.
      • Kobayashi T.
      • Saeki K.
      • Shiraishi H.
      • Yoshimura A.
      • Takaesu G.
      Loss of SOCS3 gene expression converts STAT3 function from anti-apoptotic to pro-apoptotic.
      However, the induction of STAT3 phosphorylation by OSM is similar between ES, OS, and CS cells, suggesting that transcriptional co-activators may have a role to enhance STAT3 activity and, thus, to switch from growth inhibition to growth promotion. For example, OS and CS cells express mesenchymal differentiation transcription factors, such as CBFA1, SOX9, or C\EBPδ, whereas ES cells harbor neuronal features and express stemness transcription factors, such as ID2, SOX2, NANOG, or MYC, some of which are already known to interact with and influence STAT3 transcriptional activity.
      • Ziros P.G.
      • Georgakopoulos T.
      • Habeos I.
      • Basdra E.K.
      • Papavassiliou A.G.
      Growth hormone attenuates the transcriptional activity of Runx2 by facilitating its physical association with Stat3β.
      • Foshay K.M.
      • Gallicano G.I.
      Regulation of Sox2 by STAT3 initiates commitment to the neural precursor cell fate.
      • Kidder B.L.
      • Yang J.
      • Palmer S.
      Stat3 and c-Myc genome-wide promoter occupancy in embryonic stem cells.
      Strikingly in embryonic stem cells, LIF maintains self-renewal and proliferation through STAT3 activation, its interaction with NANOG, OCT4, and SOX2, and induction of MYC.
      • Chen X.
      • Xu H.
      • Yuan P.
      • Fang F.
      • Huss M.
      • Vega V.B.
      • Wong E.
      • Orlov Y.L.
      • Zhang W.
      • Jiang J.
      • Loh Y.H.
      • Yeo H.C.
      • Yeo Z.X.
      • Narang V.
      • Govindarajan K.R.
      • Leong B.
      • Shahab A.
      • Ruan Y.
      • Bourque G.
      • Sung W.K.
      • Clarke N.D.
      • Wei C.L.
      • Ng H.H.
      Integration of external signaling pathways with the core transcriptional network in embryonic stem cells.
      Recently, MYC also appeared as a molecular switch that abrogates the STAT3-mediated growth arrest, and cooperates with OSM to promote anchorage-independent growth.
      • Kan C.E.
      • Cipriano R.
      • Jackson M.W.
      c-MYC functions as a molecular switch to alter the response of human mammary epithelial cells to oncostatin M.
      In conclusion, for OSs and presumably also for CSs, OSM binds to the OSMR/gp130 receptor and activates STAT3, which induces cell cycle inhibitors (p21WAF1 and p27KIP1)
      • Bellido T.
      • O'Brien C.A.
      • Roberson P.K.
      • Manolagas S.C.
      Transcriptional activation of the p21(WAF1,CIP1,SDI1) gene by interleukin-6 type cytokines: a prerequisite for their pro-differentiating and anti-apoptotic effects on human osteoblastic cells.
      and differentiation transcription factors (C\EBPβ and δ).
      • Hutt J.A.
      • DeWille J.W.
      Oncostatin M induces growth arrest of mammary epithelium via a CCAAT/enhancer-binding protein δ-dependent pathway.
      • Guihard P.
      • Danger Y.
      • Brounais B.
      • David E.
      • Brion R.
      • Delecrin J.
      • Richards C.D.
      • Chevalier S.
      • Redini F.
      • Heymann D.
      • Gascan H.
      • Blanchard F.
      Induction of osteogenesis in mesenchymal stem cells by activated monocytes/macrophages depends on oncostatin M signaling.
      However, the relative role and potent cross talk between these OSM-induced genes is not well defined in OS and CS cells and deserves further studies. In ES cells, OSM binds to LIFR + gp130 and OSMR + gp130 receptors, which highly induce STAT3 transcriptional activity. We propose that, in the presence of EWS-FLI1, stemness transcription factors, and/or neuronal features, STAT3 does not induce C/EBP but does induce MYC at a high level, reaching a threshold that allows G0-G1 transition and proliferation quickening. These proposed mechanisms of action are schematically presented in Figure 6C.
      In ES cells, EWS-FLI1 has a key role in the regulation of proliferation, stemness, the induction of neuronal markers, and the prevention of mesenchymal differentiation.
      • Tirode F.
      • Laud-Duval K.
      • Prieur A.
      • Delorme B.
      • Charbord P.
      • Delattre O.
      Mesenchymal stem cell features of Ewing tumors.
      • Dauphinot L.
      • De Oliveira C.
      • Melot T.
      • Sevenet N.
      • Thomas V.
      • Weissman B.E.
      • Delattre O.
      Analysis of the expression of cell cycle regulators in Ewing cell lines: EWS-FLI-1 modulates p57KIP2 and c-Myc expression.
      ChIP revealed a direct interaction of EWS-FLI1 with the promoter regions of the ID2, TGF-type II receptor, cyclin D1, DAX1/NR0B1, and MYC genes (eg, expression of these genes being either induced or repressed by EWS-FLI1).
      • Guillon N.
      • Tirode F.
      • Boeva V.
      • Zynovyev A.
      • Barillot E.
      • Delattre O.
      The oncogenic EWS-FLI1 protein binds in vivo GGAA microsatellite sequences with potential transcriptional activation function.
      • Fukuma M.
      • Okita H.
      • Hata J.
      • Umezawa A.
      Upregulation of Id2, an oncogenic helix-loop-helix protein, is mediated by the chimeric EWS/ets protein in Ewing sarcoma.
      Consequently, knockdown of EWS-FLI1 allows these cells to differentiate into osteoblasts, chondrocytes, and adipocytes and, thus, to express transcription factors that characterized OSs or CSs, such as CBFA1 or SOX9; stemness transcription factors are down-regulated, such as MYC, SOX2, and NANOG. Our results indicate that the OSMR gene is another direct target of EWS-FLI1. In that case, EWS-FLI1 behaves as a transcriptional repressor, giving a first explanation for the low OSMR transcript level in ES cells. EWS-FLI1 also prevents LIF production in ES cells, but our attempt to identify EWS-FLI1 binding to the LIF promoter using ChIP assays was unsuccessful (three LIF promoter regions tested; data not shown), suggesting an indirect regulation. Whatever the exact mechanism, the different LIF expression observed in primary bone tumors appeared as an important regulatory mechanism because this cytokine can induce internalization and lysosomal degradation of LIFR; in contrast, it induces OSMR expression.
      • Blanchard F.
      • Wang Y.
      • Kinzie E.
      • Duplomb L.
      • Godard A.
      • Baumann H.
      Oncostatin M regulates the synthesis and turnover of gp130, leukemia inhibitory factor receptor α, and oncostatin M receptor β by distinct mechanisms.
      Other factors are presumably also implicated, and we propose that IL-6, which is highly expressed in mesenchymal cells, such as osteoblasts, OS cells, and CS cells (see Supplemental Figure S3 at http://ajp.amjpathol.org),
      • Blanchard F.
      • Duplomb L.
      • Baud'huin M.
      • Brounais B.
      The dual role of IL-6-type cytokines on bone remodeling and bone tumors.
      also has a role to sustain OSMR expression.
      • Blanchard F.
      • Wang Y.
      • Kinzie E.
      • Duplomb L.
      • Godard A.
      • Baumann H.
      Oncostatin M regulates the synthesis and turnover of gp130, leukemia inhibitory factor receptor α, and oncostatin M receptor β by distinct mechanisms.
      OSM was excluded because it is not endogenously expressed by OS, CS, or ES cells.
      To translate these observations to the in vivo situation, we have analyzed multiple patient tumors, xenografts, and normal tissues. Our results indicate that, within tumors, OS cells produce more LIF than ES cells, although the tumor environment could also be a source of LIF. In contrast, OSM is not produced by tumor cells but mainly by host cells, presumably in response to inflammation, as previously suggested in canine osteosarcomas.
      • Fossey S.L.
      • Bear M.D.
      • Kisseberth W.C.
      • Pennell M.
      • London C.A.
      Oncostatin M promotes STAT3 activation, VEGF production, and invasion in osteosarcoma cell lines.
      Patient tumor immunostaining underlies a high level of OSM in vascular vessels and tumor stroma. Strikingly, this cytokine is released from neutrophils during transendothelial migration,
      • Elbjeirami W.M.
      • Donnachie E.M.
      • Burns A.R.
      • Smith C.W.
      Endothelium-derived GM-CSF influences expression of oncostatin M.
      suggesting a possible source for OSM other than tumor-associated macrophages or lymphocytes. Because we previously described that OSM overexpression reduces tumor burden in OS and CS murine models,
      • Brounais B.
      • Chipoy C.
      • Mori K.
      • Charrier C.
      • Battaglia S.
      • Pilet P.
      • Richards C.D.
      • Heymann D.
      • Redini F.
      • Blanchard F.
      Oncostatin M induces bone loss and sensitizes rat osteosarcoma to the antitumor effect of Midostaurin in vivo.
      • David E.
      • Guihard P.
      • Brounais B.
      • Riet A.
      • Charrier C.
      • Battaglia S.
      • Gouin F.
      • Ponsolle S.
      • Bot R.L.
      • Richards C.D.
      • Heymann D.
      • Redini F.
      • Blanchard F.
      Direct anti-cancer effect of oncostatin M on chondrosarcoma.
      the present results confirm that OSM could be a valuable therapeutic anticancer agent for OSs and CSs (if the adverse effects of this cytokine are adequately managed). In sharp contrast, we can predict that OSM, its receptors, or STAT3 should be neutralized in ESs. Ongoing studies will hopefully confirm the therapeutic interest of targeting OSM signaling in ES xenograft murine models.
      ES is not the only cancer type known to be growth stimulated by OSM. Because an opposite role of OSM is observed, depending on the expression of LIF, LIFR, OSMR, and MYC, it would be interesting to enlarge this statement to other kinds of tumors. Based on cDNA array analysis on patient biopsy specimens, we could differentiate two groups of tumors: tumors of neuroectodermic origin, such as gliomas, medulloblastomas, neuroblastomas, or ES tumors with low LIF and IL-6 expression and a high LIFR/OSMR ratio; and carcinomas and tumors of mesodermic origin, including liposarcomas, leiomyosarcomas, and osteosarcomas with high LIF and IL-6 expression and a lower LIFR/OSMR ratio. For MYC, the situation is more complex because it is highly expressed in ES tumors but not in other neuroectodermic tumors. However, the other proto-oncogene, MYCN, which is fundamental in the development of the peripheral and central nervous systems, is amplified or overexpressed in neuroblastomas, medulloblastomas, and gliomas.
      • Pession A.
      • Tonelli R.
      The MYCN oncogene as a specific and selective drug target for peripheral and central nervous system tumors.
      Therefore, all neuroectodermic tumors could have a similar expression of the genes that we found necessary for growth induction by OSM in ES cells. Interestingly, previous studies already described that OSM and LIF can sustain proliferation of glioma and medulloblastoma, and neutralization of IL-6 increases the survival of mice bearing glioma xenografts, supporting our view that these cytokines are important growth factors for tumors that originate from the neuroectoderm.
      • Ball S.
      • Li C.
      • Li P.K.
      • Lin J.
      The small molecule, LLL12, inhibits STAT3 phosphorylation and induces apoptosis in medulloblastoma and glioblastoma cells.
      • Liu J.
      • Li J.W.
      • Gang Y.
      • Guo L.
      • Li H.
      Expression of leukemia-inhibitory factor as an autocrinal growth factor in human medulloblastomas.
      • Wang H.
      • Lathia J.D.
      • Wu Q.
      • Wang J.
      • Li Z.
      • Heddleston J.M.
      • Eyler C.E.
      • Elderbroom J.
      • Gallagher J.
      • Schuschu J.
      • MacSwords J.
      • Cao Y.
      • McLendon R.E.
      • Wang X.F.
      • Hjelmeland A.B.
      • Rich J.N.
      Targeting interleukin 6 signaling suppresses glioma stem cell survival and tumor growth.
      Of course, this statement does not exclude that this particular cytokine-receptor-MYC signal is also implicated in the malignant transformation of certain carcinoma and mesodermic cancer cells, especially undifferentiated, metastatic, and/or cancer stem cells.
      • Blanchard F.
      • Duplomb L.
      • Baud'huin M.
      • Brounais B.
      The dual role of IL-6-type cytokines on bone remodeling and bone tumors.
      • Lacreusette A.
      • Lartigue A.
      • Nguyen J.M.
      • Barbieux I.
      • Pandolfino M.C.
      • Paris F.
      • Khammari A.
      • Dréno B.
      • Jacques Y.
      • Blanchard F.
      • Godard A.
      Relationship between responsiveness of cancer cells to Oncostatin M and/or IL-6 and survival of stage III melanoma patients treated with tumour-infiltrating lymphocytes.
      • Lacreusette A.
      • Nguyen J.M.
      • Pandolfino M.C.
      • Khammari A.
      • Dreno B.
      • Jacques Y.
      • Godard A.
      • Blanchard F.
      Loss of oncostatin M receptor beta in metastatic melanoma cells.
      • Kan C.E.
      • Cipriano R.
      • Jackson M.W.
      c-MYC functions as a molecular switch to alter the response of human mammary epithelial cells to oncostatin M.
      Our results also reinforce the current view that an important oncogenic role of EWS-FLI1 fusion protein is to induce neuronal/stemness features in ES, while preventing expression of mesenchymal transcription factors. These stemness/differentiation factors could then prevent LIF expression, induce the LIFR/OSMR ratio, STAT3 activity, and MYC expression, and, thus, alter the OSM signaling from tumor suppression to tumor promotion. Thus, a study on rare malignancies could give valuable insights on much more common cancer regulatory mechanisms.

      Supplementary data

      • Supplemental Figure S1

        A: TC32 cells are treated with different IL-6–type cytokines (50 ng/mL, except LIF at 100 ng/mL) for 3 days, and the number of viable cells is assessed using the Vialight kit. B: MG63 and RDES cells are treated 3 days with an increasing concentration of OSM, and viability is determined. C: SKES1 and RDES cells are treated with OSM (50 ng/mL) for 3 to 72 hours, washed, and cultured without OSM until day 3. Viability is determined and expressed as in A. D: MG63 and RDES cells are treated with OSM for 15 days, and living cells are quantified by trypan blue exclusion every 3 to 7 days. Assays are performed in triplicate, and results are expressed as the mean ± SD of an extrapolated number of cells. E: Indicated cells are treated as in D with OSM or LIF. Results represent the relative number of viable cells after 15 days of treatment. F: RDES cells are treated for 1 day with OSM. Histograms, cell cycle profile; dot plots, double staining for Ki67 and propidium iodide (PI). G: Sixteen cell lines are analyzed by real-time PCR for mesenchymal differentiation markers (top panel) or stemness markers (bottom panel). Results are expressed as the mean ± SEM. H: SKES1 and MG63 cells are treated with OSM for 2 to 96 hours. mRNA expression of indicated genes is assessed by real-time PCR (n = 3 for each condition). *P < 0.05, **P < 0.01, and ***P < 0.0001. ns, not significant.

      • Supplemental Figure S2

        A: MG63, RDES, and SKES1 cells are transfected with siRNA targeting LIFR, OSMR, or siCT. mRNA expression of LIFR and OSMR is assessed by real-time PCR 48 hours after transfection. Results represent the percentage of each receptor expression versus siCT transfected cells. B: SKES1 cells are transfected with indicated siRNA and analyzed by flow cytometry 48 and 72 hours later. Histograms represent LIFR and OSMR protein level as percentage of siCT transfected cells. C: SKES1 cells are transfected with pcDNA3.1-OSMR expression vector or empty vector and selected for resistance to hygromycin. Selected stable clones are picked manually, expanded, and analyzed for OSMR expression by real-time PCR and flow cytometry, as in A and B. Clones are also treated for 3 days with OSM, and the number of viable cells is assessed using the Vialight kit (right panel). D: Primary bone tumor cell lines (n = 12) are transfected with an STAT3 reporter (pSiem-Luc). At 24 hours after transfection, cells are either treated with OSM or left untreated for 24 hours. The induction of STAT3 activity by OSM is measured by luciferase assay. Black line, mean. Statistical analysis to compare ES with OS + CS is performed using the U-test. **P < 0.01. E: Indicated cell lines are treated with OSM for 15 to 360 minutes, and cell lysates are analyzed by using Western blot analysis for phosphorylated STAT3 (Tyr705 and Ser727) and STAT3, as indicated.

      • Supplemental Figure S3

        A: MG63 cells are transfected with siRNA targeting MYC (c-myc) and analyzed 3 days later for MYC expression by using Western blot analysis. B: MG63 and MSCs (obtained from the bone marrow of a 39-year-old male donor, as previously described

        • Williamson D.
        • Missiaglia E.
        • de Reynies A.
        • Pierron G.
        • Thuille B.
        • Palenzuela G.
        • Thway K.
        • Orbach D.
        • Lae M.
        • Freneaux P.
        • Pritchard-Jones K.
        • Oberlin O.
        • Shipley J.
        • Delattre O.
        Fusion gene-negative alveolar rhabdomyosarcoma is clinically and molecularly indistinguishable from embryonal rhabdomyosarcoma.
        ) are transiently transfected with an EWS-FLI1 or empty expression vector, together with pGL3OSMR or empty vector (basic). Luciferase activity is measured 48 hours after cell transfection. C: LIF mRNA expression is assessed by real-time PCR on OS (MG63 and HOS) and ES (RDES1 and SKES) cells treated with IL-1β or tumor necrosis factor-α or left untreated for 24 hours. ND, not detected. D: Sixteen different cell lines are cultured for 3 days, and the LIF or IL-6 concentration in conditioned medium is measured using a Luminex assay (Bioplex; Bio-Rad). **P < 0.01. E: MG63, SaOS2, and RDES cells are transfected with siRNA targeting LIF or siCT. At 3 days later, the protein level of LIFR and OSMR is assessed by flow cytometry. Results represent the percentage of each receptor expression versus cells transfected with siCT. F: MG63 cells are transfected with siLIF for 2 days, treated with OSM for 3 additional days, and analyzed for cell viability.

      • Supplemental Figure S4

        mRNA expression of indicated genes is assessed by cDNA array on different human tumors and normal tissues. Red, tumors or normal tissues with an ectodermic origin; blue, those with a mesodermic origin; green, hematopoietic cells; black, carcinomas. GIST, gastrointestinal stromal tumor. Results are represented by box (median and 25th and 75th percentiles) and whiskers (largest and smallest values). Outlier values are also presented (circles). Tumors or tissues that significantly expressed a higher level of LIF, IL-6, OSM, or OSMR versus ES are indicated below the boxes. *P < 0.05, **P < 0.01, and ***P < 0.0001, U-test. Tumors that significantly expressed a lower level of MYC (c-myc) or LIFR, versus ES, are similarly indicated. Statistical analysis comparing the ectodermic group with other groups (mesodermic, carcinoma, or hematopoietic) is performed using the Wilcoxon rank-sum test with continuity correction (P values are indicated above the boxes). ns, not significant.

      References

        • Picci P.
        Osteosarcoma (osteogenic sarcoma).
        Orphanet J Rare Dis. 2007; 2: 6
        • Bernstein M.
        • Kovar H.
        • Paulussen M.
        • Randall R.L.
        • Schuck A.
        • Teot L.A.
        • Juergens H.
        Ewing's sarcoma family of tumors: current management.
        Oncologist. 2006; 11: 503-519
        • Mohseny A.B.
        • Szuhai K.
        • Romeo S.
        • Buddingh E.P.
        • Briaire-de Bruijn I.
        • de Jong D.
        • van Pel M.
        • Cleton-Jansen A.M.
        • Hogendoorn P.C.
        Osteosarcoma originates from mesenchymal stem cells in consequence of aneuploidization and genomic loss of Cdkn2.
        J Pathol. 2009; 219: 294-305
        • Boeuf S.
        • Kunz P.
        • Hennig T.
        • Lehner B.
        • Hogendoorn P.
        • Bovee J.
        • Richter W.
        A chondrogenic gene expression signature in mesenchymal stem cells is a classifier of conventional central chondrosarcoma.
        J Pathol. 2008; 216: 158-166
        • Wadayama B.
        • Toguchida J.
        • Yamaguchi T.
        • Sasaki M.S.
        • Yamamuro T.
        p53 Expression and its relationship to DNA alterations in bone and soft tissue sarcomas.
        Br J Cancer. 1993; 68: 1134-1139
        • Delattre O.
        • Zucman J.
        • Plougastel B.
        • Desmaze C.
        • Melot T.
        • Peter M.
        • Kovar H.
        • Joubert I.
        • de Jong P.
        • Rouleau G.
        • Aurias A.
        • Thomas G.
        Gene fusion with an ETS DNA-binding domain caused by chromosome translocation in human tumours.
        Nature. 1992; 359: 162-165
        • May W.A.
        • Gishizky M.L.
        • Lessnick S.L.
        • Lunsford L.B.
        • Lewis B.C.
        • Delattre O.
        • Zucman J.
        • Thomas G.
        • Denny C.T.
        Ewing sarcoma 11;22 translocation produces a chimeric transcription factor that requires the DNA-binding domain encoded by FLI1 for transformation.
        Proc Natl Acad Sci U S A. 1993; 90: 5752-5756
        • Tirode F.
        • Laud-Duval K.
        • Prieur A.
        • Delorme B.
        • Charbord P.
        • Delattre O.
        Mesenchymal stem cell features of Ewing tumors.
        Cancer Cell. 2007; 11: 421-429
        • Riggi N.
        • Suva M.L.
        • De Vito C.
        • Provero P.
        • Stehle J.C.
        • Baumer K.
        • Cironi L.
        • Janiszewska M.
        • Petricevic T.
        • Suva D.
        • Tercier S.
        • Joseph J.M.
        • Guillou L.
        • Stamenkovic I.
        EWS-FLI-1 modulates miRNA145 and SOX2 expression to initiate mesenchymal stem cell reprogramming toward Ewing sarcoma cancer stem cells.
        Genes Dev. 2010; 24: 916-932
        • Rocchi A.
        • Manara M.C.
        • Sciandra M.
        • Zambelli D.
        • Nardi F.
        • Nicoletti G.
        • Garofalo C.
        • Meschini S.
        • Astolfi A.
        • Colombo M.P.
        • Lessnick S.L.
        • Picci P.
        • Scotlandi K.
        CD99 inhibits neural differentiation of human Ewing sarcoma cells and thereby contributes to oncogenesis.
        J Clin Invest. 2010; 120: 668-680
        • Lee F.Y.
        • Mankin H.J.
        • Fondren G.
        • Gebhardt M.C.
        • Springfield D.S.
        • Rosenberg A.E.
        • Jennings L.C.
        Chondrosarcoma of bone: an assessment of outcome.
        J Bone Joint Surg Am. 1999; 81: 326-338
        • Cotterill S.J.
        • Ahrens S.
        • Paulussen M.
        • Jurgens H.F.
        • Voute P.A.
        • Gadner H.
        • Craft A.W.
        Prognostic factors in Ewing's tumor of bone: analysis of 975 patients from the European Intergroup Cooperative Ewing's Sarcoma Study Group.
        J Clin Oncol. 2000; 18: 3108-3114
        • Mori K.
        • Rédini F.
        • Gouin F.
        • Cherrier B.
        • Heymann D.
        Osteosarcoma: current status of immunotherapy and future trends (review).
        Oncol Rep. 2006; 15: 693-700
        • Heinrich P.C.
        • Behrmann I.
        • Haan S.
        • Hermanns H.M.
        • Muller-Newen G.
        • Schaper F.
        Principles of interleukin (IL)-6-type cytokine signalling and its regulation.
        Biochem J. 2003; 374: 1-20
        • Chen S.H.
        • Benveniste E.N.
        Oncostatin M: a pleiotropic cytokine in the central nervous system.
        Cytokine Growth Factor Rev. 2004; 15: 379-391
        • Blanchard F.
        • Duplomb L.
        • Baud'huin M.
        • Brounais B.
        The dual role of IL-6-type cytokines on bone remodeling and bone tumors.
        Cytokine Growth Factor Rev. 2009; 20: 19-28
        • Grenier A.
        • Dehoux M.
        • Boutten A.
        • Arce-Vicioso M.
        • Durand G.
        • Gougerot-Pocidalo M.A.
        • Chollet-Martin S.
        Oncostatin M production and regulation by human polymorphonuclear neutrophils.
        Blood. 1999; 93: 1413-1421
        • Zarling J.M.
        • Shoyab M.
        • Marquardt H.
        • Hanson M.B.
        • Lioubin M.N.
        • Todaro G.J.
        Oncostatin M: a growth regulator produced by differentiated histiocytic lymphoma cells.
        Proc Natl Acad Sci U S A. 1986; 83: 9739-9743
        • Arthan D.
        • Hong S.K.
        • Park J.I.
        Leukemia inhibitory factor can mediate Ras/Raf/MEK/ERK-induced growth inhibitory signaling in medullary thyroid cancer cells.
        Cancer Lett. 2010; 297: 31-41
        • Brounais B.
        • Chipoy C.
        • Mori K.
        • Charrier C.
        • Battaglia S.
        • Pilet P.
        • Richards C.D.
        • Heymann D.
        • Redini F.
        • Blanchard F.
        Oncostatin M induces bone loss and sensitizes rat osteosarcoma to the antitumor effect of Midostaurin in vivo.
        Clin Cancer Res. 2008; 14: 5400-5409
        • Chipoy C.
        • Brounais B.
        • Trichet V.
        • Battaglia S.
        • Berreur M.
        • Oliver L.
        • Juin P.
        • Redini F.
        • Heymann D.
        • Blanchard F.
        Sensitization of osteosarcoma cells to apoptosis by oncostatin M depends on STAT5 and p53.
        Oncogene. 2007; 26: 6653-6664
        • Brounais B.
        • David E.
        • Chipoy C.
        • Trichet V.
        • Ferre V.
        • Charrier C.
        • Duplomb L.
        • Berreur M.
        • Redini F.
        • Heymann D.
        • Blanchard F.
        Long term oncostatin M treatment induces an osteocyte-like differentiation on osteosarcoma and calvaria cells.
        Bone. 2009; 44: 830-839
        • David E.
        • Guihard P.
        • Brounais B.
        • Riet A.
        • Charrier C.
        • Battaglia S.
        • Gouin F.
        • Ponsolle S.
        • Bot R.L.
        • Richards C.D.
        • Heymann D.
        • Redini F.
        • Blanchard F.
        Direct anti-cancer effect of oncostatin M on chondrosarcoma.
        Int J Cancer. 2011; 128: 1822-1835
        • Langdon C.
        • Kerr C.
        • Hassen M.
        • Hara T.
        • Arsenault A.L.
        • Richards C.D.
        Murine oncostatin M stimulates mouse synovial fibroblasts in vitro and induces inflammation and destruction in mouse joints in vivo.
        Am J Pathol. 2000; 157: 1187-1196
        • Lundquist A.
        • Barre B.
        • Bienvenu F.
        • Hermann J.
        • Avril S.
        • Coqueret O.
        Kaposi sarcoma-associated viral cyclin K overrides cell growth inhibition mediated by oncostatin M through STAT3 inhibition.
        Blood. 2003; 101: 4070-4077
        • Lacreusette A.
        • Lartigue A.
        • Nguyen J.M.
        • Barbieux I.
        • Pandolfino M.C.
        • Paris F.
        • Khammari A.
        • Dréno B.
        • Jacques Y.
        • Blanchard F.
        • Godard A.
        Relationship between responsiveness of cancer cells to Oncostatin M and/or IL-6 and survival of stage III melanoma patients treated with tumour-infiltrating lymphocytes.
        J Pathol. 2008; 216: 451-459
        • Lacreusette A.
        • Nguyen J.M.
        • Pandolfino M.C.
        • Khammari A.
        • Dreno B.
        • Jacques Y.
        • Godard A.
        • Blanchard F.
        Loss of oncostatin M receptor beta in metastatic melanoma cells.
        Oncogene. 2007; 26: 881-892
        • Hirai H.
        • Karian P.
        • Kikyo N.
        Regulation of embryonic stem cell self-renewal and pluripotency by leukaemia inhibitory factor.
        Biochem J. 2011; 438: 11-23
        • Shin J.E.
        • Park S.H.
        • Jang Y.K.
        Epigenetic up-regulation of leukemia inhibitory factor (LIF) gene during the progression to breast cancer.
        Mol Cells. 2011; 31: 181-189
        • Maruta S.
        • Takiguchi S.
        • Ueyama M.
        • Kataoka Y.
        • Oda Y.
        • Tsuneyoshi M.
        • Iguchi H.
        A role for leukemia inhibitory factor in melanoma-induced bone metastasis.
        Clin Exp Metastasis. 2009; 26: 133-141
        • Kortylewski M.
        • Heinrich P.C.
        • Mackiewicz A.
        • Schniertshauer U.
        • Klingmuller U.
        • Nakajima K.
        • Hirano T.
        • Horn F.
        • Behrmann I.
        Interleukin-6 and oncostatin M-induced growth inhibition of human A375 melanoma cells is STAT-dependent and involves upregulation of the cyclin-dependent kinase inhibitor p27/Kip1.
        Oncogene. 1999; 18: 3742-3753
        • Bellido T.
        • O'Brien C.A.
        • Roberson P.K.
        • Manolagas S.C.
        Transcriptional activation of the p21(WAF1,CIP1,SDI1) gene by interleukin-6 type cytokines: a prerequisite for their pro-differentiating and anti-apoptotic effects on human osteoblastic cells.
        J Biol Chem. 1998; 273: 21137-21144
        • Hutt J.A.
        • DeWille J.W.
        Oncostatin M induces growth arrest of mammary epithelium via a CCAAT/enhancer-binding protein δ-dependent pathway.
        Mol Cancer Ther. 2002; 1: 601-610
        • Guillon N.
        • Tirode F.
        • Boeva V.
        • Zynovyev A.
        • Barillot E.
        • Delattre O.
        The oncogenic EWS-FLI1 protein binds in vivo GGAA microsatellite sequences with potential transcriptional activation function.
        PLoS One. 2009; 4: e4932
        • Blanchard F.
        • Wang Y.
        • Kinzie E.
        • Duplomb L.
        • Godard A.
        • Baumann H.
        Oncostatin M regulates the synthesis and turnover of gp130, leukemia inhibitory factor receptor α, and oncostatin M receptor β by distinct mechanisms.
        J Biol Chem. 2001; 276: 47038-47045
        • Guihard P.
        • Danger Y.
        • Brounais B.
        • David E.
        • Brion R.
        • Delecrin J.
        • Richards C.D.
        • Chevalier S.
        • Redini F.
        • Heymann D.
        • Gascan H.
        • Blanchard F.
        Induction of osteogenesis in mesenchymal stem cells by activated monocytes/macrophages depends on oncostatin M signaling.
        Stem Cells. 2012; 30: 762-772
        • Williamson D.
        • Missiaglia E.
        • de Reynies A.
        • Pierron G.
        • Thuille B.
        • Palenzuela G.
        • Thway K.
        • Orbach D.
        • Lae M.
        • Freneaux P.
        • Pritchard-Jones K.
        • Oberlin O.
        • Shipley J.
        • Delattre O.
        Fusion gene-negative alveolar rhabdomyosarcoma is clinically and molecularly indistinguishable from embryonal rhabdomyosarcoma.
        J Clin Oncol. 2010; 28: 2151-2158
        • Dai M.
        • Wang P.
        • Boyd A.D.
        • Kostov G.
        • Athey B.
        • Jones E.G.
        • Bunney W.E.
        • Myers R.M.
        • Speed T.P.
        • Akil H.
        • Watson S.J.
        • Meng F.
        Evolving gene/transcript definitions significantly alter the interpretation of GeneChip data.
        Nucleic Acids Res. 2005; 33: e175
        • Blanchard F.
        • Duplomb L.
        • Wang Y.
        • Robledo O.
        • Kinzie E.
        • Pitard V.
        • Godard A.
        • Jacques Y.
        • Baumann H.
        Stimulation of leukemia inhibitory factor receptor degradation by extracellular signal-regulated kinase.
        J Biol Chem. 2000; 275: 28793-28801
        • Kritikou E.A.
        • Sharkey A.
        • Abell K.
        • Came P.J.
        • Anderson E.
        • Clarkson R.W.
        • Watson C.J.
        A dual, non-redundant, role for LIF as a regulator of development and STAT3-mediated cell death in mammary gland.
        Development. 2003; 130: 3459-3468
        • Matsui T.
        • Kinoshita T.
        • Hirano T.
        • Yokota T.
        • Miyajima A.
        STAT3 down-regulates the expression of cyclin D during liver development.
        J Biol Chem. 2002; 277: 36167-36173
        • Li Q.
        • Zhu J.
        • Sun F.
        • Liu L.
        • Liu X.
        • Yue Y.
        Oncostatin M promotes proliferation of ovarian cancer cells through signal transducer and activator of transcription 3.
        Int J Mol Med. 2011; 28: 101-108
        • Daheron L.
        • Opitz S.L.
        • Zaehres H.
        • Lensch M.W.
        • Andrews P.W.
        • Itskovitz-Eldor J.
        • Daley G.Q.
        LIF/STAT3 signaling fails to maintain self-renewal of human embryonic stem cells.
        Stem Cells. 2004; 22: 770-778
        • Lu Y.
        • Fukuyama S.
        • Yoshida R.
        • Kobayashi T.
        • Saeki K.
        • Shiraishi H.
        • Yoshimura A.
        • Takaesu G.
        Loss of SOCS3 gene expression converts STAT3 function from anti-apoptotic to pro-apoptotic.
        J Biol Chem. 2006; 281: 36683-36690
        • Ziros P.G.
        • Georgakopoulos T.
        • Habeos I.
        • Basdra E.K.
        • Papavassiliou A.G.
        Growth hormone attenuates the transcriptional activity of Runx2 by facilitating its physical association with Stat3β.
        J Bone Miner Res. 2004; 19: 1892-1904
        • Foshay K.M.
        • Gallicano G.I.
        Regulation of Sox2 by STAT3 initiates commitment to the neural precursor cell fate.
        Stem Cells Dev. 2008; 17: 269-278
        • Kidder B.L.
        • Yang J.
        • Palmer S.
        Stat3 and c-Myc genome-wide promoter occupancy in embryonic stem cells.
        PLoS One. 2008; 3: e3932
        • Chen X.
        • Xu H.
        • Yuan P.
        • Fang F.
        • Huss M.
        • Vega V.B.
        • Wong E.
        • Orlov Y.L.
        • Zhang W.
        • Jiang J.
        • Loh Y.H.
        • Yeo H.C.
        • Yeo Z.X.
        • Narang V.
        • Govindarajan K.R.
        • Leong B.
        • Shahab A.
        • Ruan Y.
        • Bourque G.
        • Sung W.K.
        • Clarke N.D.
        • Wei C.L.
        • Ng H.H.
        Integration of external signaling pathways with the core transcriptional network in embryonic stem cells.
        Cell. 2008; 133: 1106-1117
        • Kan C.E.
        • Cipriano R.
        • Jackson M.W.
        c-MYC functions as a molecular switch to alter the response of human mammary epithelial cells to oncostatin M.
        Cancer Res. 2011; 71: 6930-6939
        • Dauphinot L.
        • De Oliveira C.
        • Melot T.
        • Sevenet N.
        • Thomas V.
        • Weissman B.E.
        • Delattre O.
        Analysis of the expression of cell cycle regulators in Ewing cell lines: EWS-FLI-1 modulates p57KIP2 and c-Myc expression.
        Oncogene. 2001; 20: 3258-3265
        • Fukuma M.
        • Okita H.
        • Hata J.
        • Umezawa A.
        Upregulation of Id2, an oncogenic helix-loop-helix protein, is mediated by the chimeric EWS/ets protein in Ewing sarcoma.
        Oncogene. 2003; 22: 1-9
        • Fossey S.L.
        • Bear M.D.
        • Kisseberth W.C.
        • Pennell M.
        • London C.A.
        Oncostatin M promotes STAT3 activation, VEGF production, and invasion in osteosarcoma cell lines.
        BMC Cancer. 2011; 11: 125
        • Elbjeirami W.M.
        • Donnachie E.M.
        • Burns A.R.
        • Smith C.W.
        Endothelium-derived GM-CSF influences expression of oncostatin M.
        Am J Physiol Cell Physiol. 2011; 301: C947-C953
        • Pession A.
        • Tonelli R.
        The MYCN oncogene as a specific and selective drug target for peripheral and central nervous system tumors.
        Curr Cancer Drug Targets. 2005; 5: 273-283
        • Ball S.
        • Li C.
        • Li P.K.
        • Lin J.
        The small molecule, LLL12, inhibits STAT3 phosphorylation and induces apoptosis in medulloblastoma and glioblastoma cells.
        PLoS One. 2011; 6: e18820
        • Liu J.
        • Li J.W.
        • Gang Y.
        • Guo L.
        • Li H.
        Expression of leukemia-inhibitory factor as an autocrinal growth factor in human medulloblastomas.
        J Cancer Res Clin Oncol. 1999; 125: 475-480
        • Wang H.
        • Lathia J.D.
        • Wu Q.
        • Wang J.
        • Li Z.
        • Heddleston J.M.
        • Eyler C.E.
        • Elderbroom J.
        • Gallagher J.
        • Schuschu J.
        • MacSwords J.
        • Cao Y.
        • McLendon R.E.
        • Wang X.F.
        • Hjelmeland A.B.
        • Rich J.N.
        Targeting interleukin 6 signaling suppresses glioma stem cell survival and tumor growth.
        Stem Cells. 2009; 27: 2393-2404