Advertisement

Malignant Germ Cell–Like Tumors, Expressing Ki-1 Antigen (CD30), Are Revealed during in Vivo Differentiation of Partially Reprogrammed Human-Induced Pluripotent Stem Cells

      Because many of the genes used to produce induced pluripotent stem cells (iPSCs) from somatic cells are either outright established oncogenes, such as c-myc and Klf4, or potentially related to tumorigenesis in various cancers, both the safety and the risks of tumorigenesis linked to iPSC generation require evaluation. In this work, we generated, by lentivirus-mediated gene transfer of Oct4, Sox2, Nanog, and Lin28, two types of iPSCs from human mesenchymal stem cells and human amniotic fluid–derived cells: fully reprogrammed iPSCs with silencing of the four transgenes and partially reprogrammed iPSCs that still express one or several transgenes. We assessed the behavior of these cells during both their differentiation and proliferation using in vivo teratoma assays in nonobese diabetic mice with severe combined immunodeficiency. In contrast to fully reprogrammed iPSCs, 43% of partially reprogrammed iPSC cases (6 of 14 teratomas) generated major dysplasia and malignant tumors, with yolk sac tumors and embryonal carcinomas positive for α-fetoprotein, cytokeratin AE1/AE3, and CD30. This correlated with the expression of one or several transgenes used for the reprogramming, down-regulation of CDK 1A mRNA (p21/CDKN1A), and up-regulation of antiapoptotic Bcl-2 mRNA. Therefore, the oncogenicity of therapeutically valuable patient-specific iPSC-derived cells should be scrupulously evaluated before they are used for any clinical applications.
      Induced pluripotent stem cells (iPSCs) derived after somatic cell direct reprogramming by specific transcription factors represent powerful tools for research and many applications, such as patient-specific disease modeling, drug screening, or drug toxicity, or are a potential source for replacement therapies.
      Reprogramming of human somatic cells to iPSCs has been achieved by the expression of a combination of embryonic genes, such as Oct4, Sox2 with Klf4, MYC (alias c-myc),
      • Takahashi K.
      • Tanabe K.
      • Ohnuki M.
      • Narita M.
      • Ichisaka T.
      • Tomoda K.
      • Yamanaka S.
      Induction of pluripotent stem cells from adult human fibroblast by defined factors.
      or Oct4, Sox2, Lin28, and Nanog.
      • Yu J.
      • Vodyanik M.A.
      • Smuga-Otto K.
      • Antosiewicz-Bourget J.
      • Frane J.L.
      • Tian S.
      • Nie J.
      • Jonsdottir G.A.
      • Ruotti V.
      • Stewart R.
      • Slukvin I.I.
      • Thomson J.A.
      Induced pluripotent stem cell lines derived from human somatic cells.
      These genes either are outright established oncogenes, such as c-myc and Klf4, or may potentially be linked to tumorigenesis because they are expressed in various human cancer types and associated with poorly differentiated aggressive human tumors.
      • Ben-Porath I.
      • Thomson M.W.
      • Carey V.J.
      • Ge R.
      • Bell G.W.
      • Regev A.
      • Weinberg R.A.
      An embryonic stem cell–like gene expression signature in poorly differentiated aggressive human tumors.
      • Schoenhals M.
      • Kassambara A.
      • De Vos J.
      • Hose D.
      • Moreaux J.
      • Klein B.
      Embryonic stem cell markers expression in cancers.
      Klf4, directly repressing p53,
      • Rowland B.D.
      • Bernards R.
      • Peeper D.S.
      The KLF4 tumor suppressor is a transcriptional repressor of p53 that acts as a context-dependent oncogene.
      was overexpressed in human laryngeal squamous cell carcinoma,
      • Foster K.W.
      • Ren S.
      • Louro I.D.
      • Lobo-Ruppert S.M.
      • McKie-Bell P.
      • Grizzle W.
      • Hayes M.R.
      • Broker T.R.
      • Chow L.T.
      • Ruppert J.M.
      Oncogene expression cloning by retroviral transduction of adenovirus E1A-immortalized rat kidney RK3E cells: transformation of a host with epithelial features by c-MYC and the zinc finger protein GKLF.
      breast cancer,
      • Foster K.W.
      • Frost A.R.
      • McKie-Bell P.
      • Lin C.Y.
      • Engler J.A.
      • Grizzle W.E.
      • Ruppert J.M.
      Increase of GKLF messenger RNA and protein expression during progression of breast cancer.
      and gastrointestinal cancer.
      • Wei D.
      • Kanai M.
      • Huang S.
      • Xie K.
      Emerging role of KLF4 in human gastrointestinal cancer.
      The Myc protein is also a well-known oncogene that enhances proliferation and transformation
      • Adhikary S.
      • Eilers M.
      Transcriptional regulation and transformation by Myc proteins.
      and that has up to 25,000 binding sites within the mammalian genome,
      • Cawley S.
      • Bekiranov S.
      • Ng H.H.
      • Kapranov P.
      • Sekinger E.A.
      • Kampa D.
      • Piccolboni A.
      • Sementchenko V.
      • Cheng J.
      • Williams A.J.
      • Wheeler R.
      • Wong B.
      • Drenkow J.
      • Yamanaka M.
      • Patel S.
      • Brubaker S.
      • Tammana H.
      • Helt G.
      • Struhl K.
      • Gingeras T.R.
      Unbiased mapping of transcription factor binding sites along human chromosomes 21 and 22 points to widespread regulation of noncoding RNAs.
      playing, certainly for many of them, important roles in the generation of iPSCs.
      • Knoepfler P.S.
      Why myc? an unexpected ingredient in the stem cell cocktail.
      Sox2 was present in glioblastoma,
      • Gangemi R.M.
      • Griffero F.
      • Marubbi D.
      • Perera M.
      • Capra M.C.
      SOX2 silencing in glioblastoma tumor-initiating cells causes stop of proliferation and loss of tumorigenicity.
      embryonal carcinoma (EC),
      • Boer B.
      • Kopp J.
      • Mallanna S.
      • Desler M.
      • Chakravarthy H.
      • Wilder P.J.
      • Bernadt C.
      • Rizzino A.
      Elevating the levels of Sox2 in embryonal carcinoma cells and embryonic stem cells inhibits the expression of Sox2: Oct-3/4 target genes.
      and breast carcinoma,
      • Chen Y.
      • Shi L.
      • Zhang L.
      • Li R.
      • Liang J.
      • Yu W.
      • Sun L.
      • Yang X.
      • Wang Y.
      • Zhang Y.
      • Shang Y.
      The molecular mechanism governing the oncogenic potential of SOX2 in breast cancer.
      and Oct4 was expressed in various somatic breast and colon cancer cell lines, as well as in rare human lung and kidney cancer samples
      • Looijenga L.H.
      • Stoop H.
      • de Leeuw H.P.
      • de Gouveia Brazao C.A.
      • Gillis A.J.
      • van Roozendaal K.E.
      • van Zoelen E.J.
      • Weber R.F.
      • Wolffenbuttel K.P.
      • van Dekken H.
      • Honecker F.
      • Bokemeyer C.
      • Perlman E.J.
      • Schneider D.T.
      • Kononen J.
      • Sauter G.
      • Oosterhuis J.W.
      POU5F1 (OCT3/4) identifies cells with pluripotent potential in human germ cell tumors.
      and in squamous cell carcinomas.
      • Chiou S.H.
      • Yu C.C.
      • Huang C.Y.
      • Lin S.C.
      • Liu C.J.
      Positive correlations of Oct-4 and Nanog in oral cancer stem-like cells and high-grade oral squamous cell carcinoma.
      Nanog was present in carcinomas of breast,
      • Ezeh U.I.
      • Turek P.J.
      • Reijo R.A.
      • Clark A.T.
      Human embryonic stem cell genes OCT4, NANOG, STELLAR, and GDF3 are expressed in both seminoma and breast carcinoma.
      ovary,
      • Zhang S.
      • Balch C.
      • Chan M.W.
      • Lai H.C.
      • Matei D.
      • Schilder J.M.
      • Yan P.S.
      • Huang T.H.
      • Nephew K.P.
      Identification and characterization of ovarian cancer-initiating cells from primary human tumors.
      prostate,
      • Jeter C.R.
      • Badeaux M.
      • Choy G.
      • Chandra D.
      • Patrawala L.
      • Liu C.
      • Calhoun-Davis T.
      • Zaehres H.
      • Daley G.Q.
      • Tang D.G.
      Functional evidence that the self-renewal gene NANOG regulates human tumor development.
      and kidney,
      • Bussolati B.
      • Bruno S.
      • Grange C.
      • Ferrando U.
      • Camussi G.
      Identification of a tumor-initiating stem cell population in human renal carcinomas.
      whereas Lin28 was recently overexpressed in primary human tumors
      • Viswanathan S.R.
      • Powers J.T.
      • Einhorn W.
      • Hoshida Y.
      • Ng T.L.
      • Toffanin S.
      • O'Sullivan M.
      • Lu J.
      • Phillips L.A.
      • Lockhart V.L.
      • Shah S.P.
      • Tanwar P.S.
      • Mermel C.H.
      • Beroukhim R.
      • Azam M.
      • Teixeira J.
      • Meyerson M.
      • Hughes T.P.
      • Llovet J.M.
      • Radich J.
      • Mullighan C.G.
      • Golub T.R.
      • Sorensen P.H.
      • Daley G.Q.
      Lin28 promotes transformation and is associated with advanced human malignancies.
      and to enhance metastasis in an MDA-MB-231 breast tumor model.
      • Dangi-Garimella S.
      • Yun J.
      • Eves E.M.
      • Newman M.
      • Erkeland S.J.
      • Hammond S.M.
      • Minn A.J.
      • Rosner M.R.
      Raf kinase inhibitory protein suppresses a metastasis signalling cascade involving LIN28 and let-7.
      In addition, a growing number of solid tumors have contained a subpopulation of cells possessing tumor-initiating capability associated with stem cell properties that include expression of embryonic stem cell markers.
      • Rossi D.J.
      • Jamieson C.H.
      • Weissman I.L.
      Stem cells and the pathways to aging and cancer.
      Thus, it is conceivable that transformation of primary differentiated or lineage-committed cells may incur similar reprogramming, and that some degree of reprogramming may even be inherent to transformation itself. In support of these views, loss of the retinoblastoma gene that occurs in a wide variety of malignancies has caused primary mouse embryonic fibroblasts (MEFs) to undergo reprogramming and acquire cancer stem cell properties.
      • Liu Y.
      • Clem B.
      • Zuba-Surma E.K.
      • El-Naggar S.
      • Telang S.
      • Jenson A.B.
      • Wang Y.
      • Shao H.
      • Ratajczak M.Z.
      • Chesney J.
      • Dean D.C.
      Mouse fibroblasts lacking RB1 function form spheres and undergo reprogramming to a cancer stem cell phenotype.
      Little is known about the potential risk of malignancy of reprogrammed iPSCs during differentiation in vivo, an issue that needs to be scrupulously explored before future clinical applications using iPSCs in regenerative medicine therapies.
      In this study, we analyzed, using the in vivo teratoma assay in nonobese diabetic (NOD) mice with severe combined immunodeficiency (SCID), a series of iPSCs generated by the combination of Oct4, Sox2, Lin28, and Nanog, and showed that the expression of Ki-1 antigen (CD30) during differentiation is highly predictive of human iPSC-derived malignancies.

      Materials and Methods

      Generation of MSCs from hESCs

      The human embryonic stem cell (hESC)–derived mesenchymal stem cells (MSCs) were obtained as follows: the H9 cells (Wicell, Madison, WI) were maintained in Dulbecco's modified Eagle's medium (DMEM)/F12 (Invitrogen, Saint Quentin en Yvelines, France), supplemented with 10% heat-inactivated fetal bovine serum (Hyclone, Brebières, France), 1 ng/mL basic fibroblast growth factor (bFGF), 0.1 mmol/L nonessential amino acids, 1 mmol/L l-glutamine, 0.1 mmol/L β-mercaptoethanol, and one times penicillin-streptomycin (all from Invitrogen). Adherent cells obtained were then characterized with the human mesenchymal stem cell functional identification kit (R&D Systems, Lille, France) for their ability to differentiate into the following mesenchymal lineages: adipocytes, chondrocytes, and osteocytes. Briefly, cells were stained by immunohistochemistry (IHC) using polyclonal goat anti-mouse fatty acid–binding protein-4, monoclonal goat anti-human aggrecan, and mouse anti-human osteocalcin antibodies. Cell lines positive for mesenchymal differentiation potential were subsequently termed MSCs.

      Derivation and Characterization of iPSCs

      The MSCs from passage 7 were reprogrammed in one experiment to generate iPSCs (IPS-MSCs) using lentiviral vectors carrying the following transgenes under the control of the elongation factor-1 α promoter, using published methods
      • Yu J.
      • Vodyanik M.A.
      • Smuga-Otto K.
      • Antosiewicz-Bourget J.
      • Frane J.L.
      • Tian S.
      • Nie J.
      • Jonsdottir G.A.
      • Ruotti V.
      • Stewart R.
      • Slukvin I.I.
      • Thomson J.A.
      Induced pluripotent stem cell lines derived from human somatic cells.
      : Oct4, Sox2, Nanog, and Lin28. The H9 cells were infected, using the same procedure, with a lentivirus carrying the green fluorescent protein (GFP) reporter gene. Briefly, 6 days after the infection, 100,000 infected cells were plated on inactivated MEFs in daily changed DMEM/F12 culture medium, supplemented with 20% KnockOut Serum Replacement, 10 ng/mL bFGF, 0.1 mmol/L nonessential amino acids, 1 mmol/L l-glutamine, 0.1 mmol/L β-mercaptoethanol, and one times penicillin-streptomycin (all from Invitrogen, Saint Quentin en Yvelines, France). Nontransduced MSCs were used as a control. After 15 to 25 days of culture, nearly 100 individual ES-like colonies appeared from transduced MSCs that were then pooled for expansion and characterization by flow cytometry and RT-PCR analysis. Expansion was performed through manual picking during the four first passages and then cell lines were enzymatically passaged using 1 mg/mL of collagenase IV in DMEM/F12. For each passage, 25% of the cell suspension was plated on new mitomycin C–treated MEFs.
      We also produced seven different iPS human amniotic fluid–derived cells (hAFCs) from four different hAFCs via ectopic expression of the four human factors (Oct4, Sox2, Nanog, and Lin28) using human virus–based lentiviruses. The seven human iPS-hAFCs were expanded as iPS-MSCs and characterized by flow cytometry and RT-PCR analysis.

      Generation of EBs from H9 Cells and iPS-MSC Lines

      For the formation of embryoid bodies (EBs), H9 and iPS-MSC colonies were separated from the MEF feeder cells by 2-hour incubation in 1 mg/mL of collagenase IV in DMEM/F12 (Invitrogen) and seeded in six-well Ultra Low Attachment Plates (Corning, Chorges, France) in a final volume of 2 mL of IMDM, supplemented with 15% fetal bovine serum (Hyclone, Brebières, France), 2 mmol/L l-glutamine (Invitrogen), 1% penicillin-streptomycin (Invitrogen), 1% nonessential amino acids (Invitrogen), 100 μmol/L 2-mercaptoethanol (Invitrogen), and 10 ng/mL bFGF (Miltenyi Biotech, Paris, France). After 24 hours, bFGF was removed and the cells were incubated at 37°C in 5% CO2 until day 4, 8, 13, or 14.

      Teratoma Formation and IHC

      The teratoma assay was systematically performed with ESCs (H9) and iPSCs (iPS-MSCs and iPS-hAFCs) by i.m. injection of 0.5 × 106 to 3 × 106 cells into 6-week-old NOD/SCID mice (Charles River Laboratories, Lyon, France). We injected 5 × 106 MSCs from passages 7 and 10 and hAFCs from passage 12 as a control. After 5 to 10 weeks, teratomas were dissected and fixed in 4% paraformaldehyde and samples were embedded in paraffin and stained with H&E in association with IHC, to assess the presence of ectodermic, endodermic, and mesodermic tissues. The IHC was performed as requested with a Benchmark XT apparatus (Ventana Medical System, Illkirch, France) with prediluted primary antibodies raised against AE1/AE3 (BD Biosciences, Le Pont de Claix, France), α-fetoprotein (AFP; Dako, Trappes, France), CD30 (Ventana), Glial Fibrilary Acidic Protein (GFAP, Dako), placental alkaline phosphatase (Dako), and neurofilament (Dako). To evaluate the intensity of CD30-positive areas, an image of the whole teratoma was taken with a digital camera (PCO, Kelheim, Germany) and analyzed by the pathologists (P.D. and P.O.). For each histological image, CD30-positive areas and the total surface of the teratoma were manually selected under Adobe Photoshop software (Paris, France). The CD30 expression was analyzed in extracts of 11 different teratomas by using Western blot analysis, as previously described.
      • Galaup A.
      • Magnon C.
      • Rouffiac V.
      • Opolon P.
      • Opolon D.
      • Lassau N.
      • Tursz T.
      • Perricaudet M.
      • Griscelli F.
      Full kringles of plasminogen (aa 1–566) mediate complete regression of human MDA-MB-231 breast tumor xenografted in nude mice.
      Briefly, 40 μg of protein was loaded in each well, and the nitrocellulose membrane was probed with goat anti-CD30 antibody (1:200 Santa cruz-1737, Le Parray en Yvelines, France). Goat anti-rabbit peroxidase-linked antibody (1:10,000; Promega, Charbonnieres, France) was used as a secondary antibody. Actin was detected with monoclonal anti–β-actin peroxidase–conjugated antibody (1:25,000; Sigma-Aldrich, Lyon, France).

      Human Stem Cell Pluripotency Low-Density Array

      Gene expression levels of H9 and iPS-MSC cells and their derived EB cells were investigated using Human Stem Cell Pluripotency TaqMan Low-Density Array (TLDA; Applied Biosystems, Villebon sur-Yvette, France). The microfluidic card (Applied Biosystems, Foster City, CA) consisted of four identical 96-gene sets (90 target genes and 6 endogenous controls). Target genes included genes expressed in undifferentiated cells, genes involved in the maintenance of pluripotency, stemness-related genes, and differentiation marker genes, as previously described.
      • Lensch M.W.
      • Daley G.Q.
      Human embryonic stem cells flock together.
      Quantitative PCR was performed with the ABI PRISM 7900HT Sequence Detection System (Applied Biosystems, France). Clustering gene expression patterns were determined using hierarchical algorithms of StatMiner software (Applied Biosystems, France). In the present study, euclidean distance and complete linkage methods were applied, although clustering results were identical whatever the linkage method and distance measure used.

      Real-Time RT-PCR

      The theoretical and practical aspects of real-time quantitative RT-PCR were determined using the ABI PRISM 7900 Sequence Detection System. The RNA extraction method, cDNA synthesis, and PCR conditions were performed as previously described.
      • Bieche I.
      • Parfait B.
      • Le Doussal V.
      • Olivi M.
      • Rio M.C.
      • Lidereau R.
      • Vidaud M.
      Identification of CGA as a novel oestrogen receptor-responsive gene in breast cancer: an outstanding candidate marker to predict the response to endocrine therapy.
      Target gene expression was normalized relative to an endogenous RNA control (TBP, encoding TATA box-binding protein). Results expressed as N-fold differences in target gene expression relative to the TBP gene and termed NTarget were determined as follows: NTarget = 2ΔCT sample, where the ΔCT value of the sample was determined by subtracting the average CT value of the target gene from the average CT value of the TBP gene. The NTarget values of the samples were subsequently normalized to the smallest amount of target gene mRNA quantifiable (target gene CT, 35). Samples with low levels of Target gene mRNA, not quantifiable by the real-time quantitative RT-PCR assay (CT, <35), were considered as nonexpressed (NTarget, 0). The upstream and downstream primer sequences for TBP were as follows: 5′-TGCACAGGAGCCAAGAGTGAA-3′ and 5′-CACATCACAGCTCCCCACCA-3′, respectively. Primer sequences located within the cDNA of Nanog, Lin28, Sox2, and Oct4 were used to amplify endogenous and exogenous (transgene) mRNAs. The upstream and downstream primer sequences were as follows: 5′-CATCCTGAACCTCAGCTACAAACA-3′ and 5′-GTGCTGAGGCCTTCTGCGT-3′, respectively, for Nanog; 5′-TGGATGTCTTTGTGCACCAGAGTAA-3′ and 5′-CGCCTCTCACTCCCAATACAGAAT-3′, respectively, for Lin28; 5′-AAGTGGGTGGAGGAAGCTGACA-3′ and 5′-GGTTTCTGCTTTGCATATCTCCTGA-3′, respectively, for Oct4; and 5′-GCCCCCAGCAGACTTCACAT-3′ and 5′-AGGGGCAGTGTGCCGTTAAT-3′, respectively, for Sox2. For the quantification of the transgene, the upstream primers were located within the sequence of endogenous mRNA and the downstream primers were located within the sequences of the lentiviral vectors. The upstream and downstream primer sequences were as follows: 5′-GTGCACAGGGAAAGCCAACCT-3′ and 5′-ATGCATGCGGATCCTTCGAA-3′, respectively, for Lin28; 5′-ACATGCAACCTGAAGACGTGTGA-3′ and 5′-GCATGCGGATCCTTCGAACTA-3′, respectively, for Nanog; 5′-CCATGCATTCAAACTGAGGTGC-3′ and 5′-TTCGAACTAGCGCTACCGGACT-3′, respectively, for Oct4; and 5′-ATTAACGGCACACTGCCCCT-3′ and 5′-CCTAGATGCATGCGGATCCTT-3′, respectively, for Sox2.

      Statistical Analysis

      All analyses were performed in duplicate. Statistical significance was evaluated using the nonparametric Mann-Whitney or the Kruskal-Wallis test.

      Results

      Production and Characterization of iPSCs

      To determine the tumor-induction potential of human iPSCs, we used a strategy, as previously published by our group,
      • Giuliani M.
      • Oudrhiri N.
      • Noman Z.M.
      • Vernochet A.
      • Chouaib S.
      • Azzarone B.
      • Durrbach A.
      • Bennaceur-Griscelli A.
      Human mesenchymal stem cells derived from induced pluripotent stem cells down-regulate NK-cell cytolytic machinery.
      that consisted of MSCs generated from human H9 embryonic stem cells (H9) to further generate iPSCs (iPS-MSCs) by lentivirus-mediated gene transfer of Oct4, Sox2, Nanog, and Lin28,
      • Yu J.
      • Vodyanik M.A.
      • Smuga-Otto K.
      • Antosiewicz-Bourget J.
      • Frane J.L.
      • Tian S.
      • Nie J.
      • Jonsdottir G.A.
      • Ruotti V.
      • Stewart R.
      • Slukvin I.I.
      • Thomson J.A.
      Induced pluripotent stem cell lines derived from human somatic cells.
      with an efficiency of 0.1%. This strategy had the unique advantage of comparing the tumor-inducing potential of iPSCs with their genuine parental ESCs.
      We first characterized the MSCs before the reprogramming. In contrast to H9 cells, they expressed a bone marrow–MSC–like surface antigen profile (ie, positive for CD105, CD73, CD146, and CD90, and negative for Nanog, Oct4, Lin28, Sox2, CD45, CD80, HLA-DR, and CD34)
      • Giuliani M.
      • Oudrhiri N.
      • Noman Z.M.
      • Vernochet A.
      • Chouaib S.
      • Azzarone B.
      • Durrbach A.
      • Bennaceur-Griscelli A.
      Human mesenchymal stem cells derived from induced pluripotent stem cells down-regulate NK-cell cytolytic machinery.
      and differentiated toward adipogenesis, chondrogenesis, and osteogenesis (see Supplemental Figure S1 at http://ajp.amjpathol.org). In contrast to H9 and iPS-MSCs, the MSCs did not express hTERT, CD133, and ABCG2 mRNA (see Supplemental Table S1 at http://ajp.amjpathol.org) and, as expected, failed to form teratomas in NOD/SCID mice after injection of 5 × 106 cells. In addition, these MSCs failed to generate ES-like colonies when plated on MEFs in hES medium, confirming the absence of residual H9 cell contamination in the MSC suspension before the reprogramming process.
      The molecular profiling of MSCs was then performed compared with H9 and iPS-MSCs. In contrast to MSCs, iPS-MSCs have lost a mesenchymal expression profile (see Supplemental Table S1 at http://ajp.amjpathol.org), as evidenced by the drastic decrease of vimentin mRNA and the up-regulation of E-cadherin, with an increase of the ratio of the mRNA levels of vimentin and E-cadherin (1.1 versus 37,743). Furthermore, and as anticipated,
      • Mani S.A.
      • Guo W.
      • Liao M.J.
      • Eaton E.
      • Ayyanan A.
      • Zhou A.Y.
      • Brooks M.
      • Reinhard F.
      • Zhang C.C.
      • Shipitsin M.
      • Campbell L.L.
      • Polyak K.
      • Brisken C.
      • Yang J.
      • Weinberg R.A.
      The epithelial-mesenchymal transition generates cells with properties of stem cells.
      we found that as H9 and iPS-MSCs down-regulated the expression of mesenchymal mRNAs encoding CD44 and up-regulated mRNAs encoding epithelial markers, such as CD24, showing a drastic decrease of the ratio of CD44/CD24 for iPS-MSCs (ratio, 0.0008) and H9 (ratio, 0.02), compared with MSCs (ratio, 49) that are CD44 high and CD24 low. In contrast to MSCs, transforming growth factors β1 and β2, mediating epithelial-to-mesenchymal transdifferentiation through the transforming growth factor β/Smad pathway, were highly decreased in iPS-MSCs, as were specific mesenchymal transcriptional factors, such as SLUG (approximately 49-fold), ZEB1 (approximately 180-fold), FOXC2 (approximately 118-fold), and TWIST (approximately 20-fold).
      All iPS-MSC colonies were selected by morphological features, picked at different passages, and characterized. iPS-MSCs shared many properties with their H9 counterparts. Although MSCs have acquired fibroblast-like, mesenchymal appearances, the iPS-MSCs (Figure 1A) remain indistinguishable after the fourth passage from the parental H9 cells. At this time, the pooled iPS-MSC colonies highly expressed tumor rejection antigen 1-60 with the same level as hES H9 cell lines. Indeed, we have scrupulously followed the expression of TRA-1-60 by fluorescence-activated cell sorter analysis in expanded iPS-MSC cell lines at several passages (between passages 4 and 64), compared with H9 cells, and found a constant and comparable expression of TRA-1-60 for both cell lines (see Supplemental Figure S2A at http://ajp.amjpathol.org).
      Figure thumbnail gr1
      Figure 1Derivation and characterization of the iPSCs. A: Morphological characteristics of iPS-MSCs from passage 4 and MSCs from passage 7. B: Morphological characteristics of iPS-hAFCs (PB03) from passage 20 and hAFC3 from passage 2. C: Co-expression of SSEA4 and human early stem cell antigen-1 (HESCA-1) stem cell antigen (top panels) and co-expression of CD30 and TRA-1-60 stem cell antigen (bottom panels) by flow cytometric analysis of MSCs, H9 cells, and iPS-MSCs. Briefly, 105 cells were stained with a combination of fluorescein isothiocyanate (FITC)–conjugated mouse anti-human HESCA-1 (clone 051007-4A5; Millipore, Molsheim, France) and phosphatidylethanolamine (PE)–conjugated mouse anti-human SSEA-4 (clone MC-813-70; Millipore) and with a combination of FITC-conjugated mouse anti-human CD30 (clone BerH2; Dako, Trappes, France) and PE-conjugated mouse anti-human TRA-1-60 (clone TRA-1-60; BD Biosciences, Le Pont de Chaix, France). Cells were analyzed with an MACSQuant flow cytometer (Miltenyi Biotech, Paris, France) using MACSQuantify software. D: Co-expression of SSEA4 and HESCA-1 stem cell antigen (top panels) and co-expression of CD30 and TRA-1-60 stem cell antigen (bottom panels) by flow cytometric analysis of three different hAFCs (hAFC3, hAFC4, and hAFC6). E: Co-expression of SSEA4 and HESCA-1 stem cell antigens by flow cytometric analysis of three different iPS-hAFCs (PB03, PB04, and PB06).
      In addition, flow cytometric analysis showed a similar co-expression of human early stem cell antigen-1 (HESCA-1) and stage-specific embryonic antigen (SSEA)-4 (>90%) and a similar co-expression of TRA-1-60 and CD30 (>70%) in H9 (picked at passage 35) cells and iPS-MSCs (picked at passage 45), in contrast to the MSCs (Figure 1C).
      In addition, we showed, by flow cytometric analysis, that similar to H9 control cells, CD30 was expressed in the same levels in iPS-MSCs picked at an early or late passage (11 versus 63) with 73% and 77% of positive cells, respectively (Figure 2). These CD30 expressions were drastically decreased when H9 and iPS-MSCs, collected at early or late passages (11 and 63, respectively), were differentiated into EBs in vitro during 14 days of culture (Figure 2). Furthermore, we noted a comparable differentiation potential of iPS-MSCs into the three primary germ layers, with parental H9, both having similar hierarchical clustering of pluripotent gene expression (see Supplemental Figure S2B at http://ajp.amjpathol.org) and differentiation genes (see Supplemental Figure S2C at http://ajp.amjpathol.org) by TLDA in sequential day 4, 8, and 13 EBs (EB-4, EB-8, and EB-13, respectively). The TLDA analysis showed similar gene expression patterns between iPS-MSCs and H9 cells, both expressing ectodermal, endodermal, and mesodermal lineage markers.
      Figure thumbnail gr2
      Figure 2Expression of CD30 by flow cytometric analysis of H9 and iPSC-MSCs before and after their differentiation in vitro. The quantification of CD30 was performed on H9 cells at passage 35 and on iPSC-MSCs at early (11) and late (63) passages before (day 0) and after their differentiation into EBs after 14 days of culture. FITC, fluorescein isothiocyanate.
      We then reprogrammed human amniotic fluid–derived cells (hAFCs) obtained through amniocentesis during routine prenatal diagnosis from four different patients. After long-term culture (>10 passages), in appropriate medium, hAFCs displayed homogeneous morphological characteristics (Figure 1B) and induced iPSCs via ectopic expression of Oct4, Sox2, Nanog, and Lin28. To further characterize the hAFCs, we performed immunofluorescence staining, using antibodies against pluripotent markers. All of the hAFCs tested had uniformly positive signals for SSEA-4 and HESCA-1 (Figure 1D). We detected TRA-1-60 only in hAF3 and hAF4 cell lines (6.21% and 23%, respectively), and we were not able to detect significant expressions of CD30 in the three cell lines tested (Figure 1D). We generated seven different iPSCs from four different hAFCs (iPS-hAFCs called PB03, PB04, PB05, PB05.1, PB06, PB09, and PB10). Flow cytometric analysis showed a similar expression of HESCA-1 and SSEA-4 (82% to 99%) in three different iPS-hAFCs (Figure 1E). In addition, we performed a double immunostaining (TRA-1-60 and CD30) of four different iPS-hAFCs (PB03, PB06, PB09, and PB10) collected at different passages (early versus late). We showed that early- and late-passage iPS-hAFCs displayed a positive signal (65% to 85%) of TRA-1-60 (Figure 3A) and that significant portions (28% to 65%) of iPS-hAFCs co-expressed TRA-1-60 and CD30 (Figure 3A). In addition, we showed that the CD30-level expressions were not significantly different (P = 0.81) in early and late passages of hAFC-derived iPSCs, with 54% ± 4.43% and 57.6% ± 3.66% of CD30-positive cells, respectively (Figure 3A). These fractions of CD30-positive cells were significantly (P < 0.0002) down-regulated (2.4% ± 0.60% and 1.4% ± 0.35% for early and late passages, respectively) after their differentiation in vitro into EBs during 14 days of culture (Figure 3B).
      Figure thumbnail gr3
      Figure 3Characterization of the iPS-hAFCs and expression of CD30 by flow cytometric analysis. A: Co-expression of CD30 and TRA-1-60 stem cell antigen by flow cytometric analysis of four different iPS-hAFCs (PB03, PB06, PB09, and PB10) picked at early (top panels) and late (bottom panels) passages. B: CD30 expression of iPS-hAFC–derived EBs. Early (top panels) and late (bottom panels) passages of the four iPS-hAFCs were differentiated in vitro into EBs after 14 days of culture and tested for CD30 expression by flow cytometric analysis.
      Molecular expression of the reprogramming factors was then performed in parental and reprogrammed cells by quantitative real-time RT-PCR. Transgene expression of Oct4, Sox2, Nanog, and Lin28 was first analyzed in iPS-MSCs picked at different passages (4, 7, 9, 11, and 31) and in parental MSC cell lines using two sets of primers: one set (with both primers located within the cDNA of Nanog, Lin28, Sox2, and Oct4) to amplify the endogenous and exogenous mRNAs and one set (with one primer located within the sequence of endogenous mRNA and the other primer located within the sequences of the lentiviral vectors) to amplify only the exogenous mRNA.
      The MSCs used as control were always negative with the two sets of primers. Analysis of the gene expression of these iPS-MSCs revealed that only late passage (passage 31) allows the emergence of fully reprogrammed iPSCs expressing TRA-1-60 with a silencing of all of the transgenes (Figure 4A). In contrast, lentiviral expressions of Lin28 and Nanog, but not Oct4 and Sox2, were not totally down-regulated in early and intermediate passages (passages 4 to 11), corresponding to incompletely reprogrammed iPSCs. RT-PCR using cDNA-specific primers for the transcription factors were then performed to quantify the emergence of endogenous mRNA and exogenous mRNA after transduction. We showed that the level of endogenous transcripts of the four factors was optimal at passage 31, when the exogenous transcripts were practically undetectable (Figure 4A).
      Figure thumbnail gr4
      Figure 4Molecular characterization of iPS-MSCs and MSCs. A: RT-PCR analysis of Lin28, Nanog, Sox2, and Oct4 in MSCs and iPS-MSCs picked at different passages. We used primer sets that amplified endogenous and exogenous transcripts (total) or primers for the detection of the specific transgenes. B: Karyotype analysis of H9 cell and iPS-MSC lines.
      All H9 and iPS-MSCs, which were positive for TRA-1-60, SSEA-4, HESCA-1 and CD30, exhibited a normal diploid karyotype (Figure 4B), performed at late passages (40 and 50, respectively).
      The expression of Nanog, Lin28, Sox2, and Oct4 was then analyzed by quantitative real-time RT-PCR in H9 cells and in iPS-hAFCs, allowing the detection of total (endogenous and exogenous) and exogenous gene levels. The four exogenously transduced transcription factors were not detected in H9 control cells and were silenced in iPS-hAFCs picked at late passages: passage 32 for PB03 and PB04 (Figure 5A), passage 28 for PB05 (Figure 5B), and passage 22 for PB05.1 (Figure 5C). Interestingly, differential mRNA levels of exogenous genes were found in various early and intermediate passages of iPS-hAFCs. PB06 (passage 13) and PB10 (passage 14) showed strong-to-moderate expression of exogenous Sox2 and Oct4 (Figure 5A). When PB05 (Figure 5B) and PB05.1 (Figure 5C) were tested at different passages, we showed that both iPS-hAFCs retained strong expressions of exogenous transcription factors only at early passages, such as Nanog/Sox2/Oct4 and Nanog, respectively. Furthermore, the endogenous mRNAs were detected only in later passages, corresponding to fully reprogrammed iPSCs because the transgene mRNAs were not detected anymore.
      Figure thumbnail gr5
      Figure 5Molecular characterization of iPS-hAFCs and H9. A: RT-PCR analysis of Lin28, Nanog, Sox2, and Oct4 in H9 and iPS-hAFCs (PB03, PB04, PB06, PB09, and PB10) at different passages. B: RT-PCR analysis of Lin28, Nanog, Sox2, and Oct4 in PB05 passages 7, 9, 11, and 28. C: RT-PCR analysis of Lin28, Nanog, Sox2, and Oct4 in PB05.1 at passages 9, 10, 11, and 22. In all cases, we used primer sets that amplified endogenous and exogenous transcripts (total) or primers for the detection of the specific transgenes.
      These results revealed a varying degree of reprogramming efficiency, consistent with the conclusion that early and late passages of iPSCs corresponded to partially and fully reprogrammed iPSCs, respectively.

      Evaluation of the Tumorigenic Potential of Human ESCs and iPSCs

      Teratoma-forming potential and tumor content by histological analysis were explored in human iPS-MSCs and iPS-hAFCs, and compared with the H9 cells. As expected, injection of many MSCs or hAFCs (5 × 106) did not generate any teratoma in NOD/SCID mice. On the other hand, all iPSCs produced teratomas after 76 to 163 days, a latency comparable to those generated from H9, demonstrating their properties of pluripotency. A histological analysis showed a differentiation into ectodermal, endodermal, and mesodermal tissues, mainly represented by glial tissues, glandular epitheliums, and large cartilaginous areas, respectively (Table 1).
      Table 1Germ Cell Layers and Malignant Components within Teratomas
      Line typePassageEctodermEndoderm: glandular epitheliumMesodermGerm cell tumorsCD30mRNA expression, Tg/T
      The expression of Oct4, Sox2, Lin28, and Nanog was analyzed by quantitative real-time PCR (as described in Materials and Methods). 0 indicates that Tg mRNA was not amplified.
      Skin and dermoid cystGlial tissueImmature neuronal tissueMuscle fibersCartilageBoneFatty tissueOct4Sox2Lin28Nanog
      MSC4++++++++++RareRareEC, YST++017.811451.56
      MSC4++++++++++RareRareEC, YST++8.234.23183.641.07
      MSC6++++Rare0000
      MSC16++0000
      MSC16++++++++001.10
      MSC31++++++Rare0000
      MSC31++++++++NDNDNDND
      PB106++++++++EC, YST+0000
      PB05.112+++++++++EC, YST+03.21477.26
      PB0612++++++++EC, YST+08.03047.96
      PB0515++EC, YST+01.4405.02
      PB0917++++++++++0000
      PB0432+++++++0000
      PB0332++++++++++0000
      hESC H938++++++++++++0000
      hESC H938++++++++++++++0000
      hESC H938+++++++++0000
      The presence of the three lineages (ectoderm, endoderm, and mesoderm) was scored as follows: +, weak (one area); ++, moderate (two to five areas); +++, high (more than five areas); −, absence of the tissue.
      ND, no data; T, total endogenous and exogenous mRNAs; Tg, exogenous mRNAs.
      low asterisk The expression of Oct4, Sox2, Lin28, and Nanog was analyzed by quantitative real-time PCR (as described in Materials and Methods). 0 indicates that Tg mRNA was not amplified.
      Of the 14 teratomas analyzed, 5 have revealed the presence of immature neuronal tissue and 6 have revealed the presence of malignant germ cell–like tumors (Table 1). The immature neuronal tissue was immune reactive for placental alkaline phosphatase, glial fibrillary acidic protein, and neurofilament protein (data not shown). This primitive neural tissue, showing strong mitotic activity, was highly suggestive of grade 3 malignant tumors (data not shown).
      In 6 of 14 teratomas, large mixed germ cell tumors were observed, exhibiting typical features of two types of tumors, yolk sac tumor (YST) and EC (Table 1). These malignant tumors were only revealed after the injection of 3 × 106 partially reprogrammed iPSCs (Table 1). In counterpart, we never observed malignant tumors with H9 cells at the same cell number and even with a higher cell number nor when fully reprogrammed iPSCs were injected. Moreover, no evidence of malignancy was observed in teratomas derived from 3 × 106 H9 cells transduced with a control lentivirus expressing GFP (data not shown).
      A histological description of YSTs showed a reticular pattern consisting of a loose meshwork of space lined by primitive-appearing cells, typically with clear cytoplasms and nuclei that are hyperchromatic and irregular (Figures 6A and 7A). Large nests of primitive epithelial cells, arranged as tubular glands and villi lined by cells with subnuclear vacuoles, were observed (Figures 6B and 7A). The YSTs showed cystic spaces lined by a flattened, endothelium-like layer of cells (Figure 6C) and eosinophilic, intracellular, and extracellular strongly AFP-positive hyaline bodies among a diffuse AFP-expressing background, which are consistently found in the YSTs (Figure 6D). The ECs contained tumor cells with dark cytoplasms and large and hyperchromatic nuclei, with one or more prominent nucleoli (Figures 6E and 7A). The YSTs and ECs were positive for AFP (Figure 6F) and cytokeratin AE1/AE3 (Figure 6G) and, as expected,
      • Dürkop H.
      • Foss H.D.
      • Eitelbach F.
      • Anagnostopoulos I.
      • Latza U.
      • Pileri S.
      • Stein H.
      Expression of the CD30 antigen in non-lymphoid tissues and cells.
      exhibited high positivity for CD30 (Figure 6H), in contrast to benign mature teratomas (Figure 6I). The CD30 immunostaining areas represented approximately 15% of the total teratoma mass in two specimens (Figure 6J) and <5% of the total teratoma mass in the four other teratomas (Figure 7A). The expression of CD30 was confirmed by using Western blot analysis, showing an expression of CD30 in all of the six teratomas exhibiting ECs and YSTs (Figure 8A), whereas all benign teratomas tested were negative for this expression. In some extracts, the immunoreactive CD30 appeared as a doublet, with molecular masses of 120 and 105 kDa, most likely reflecting a different extent of glycosylation, as previously described.
      • Nagata S.
      • Ise T.
      • Onda M.
      • Nakamura K.
      • Ho M.
      • Raubitschek A.
      • Pastan I.H.
      Cell membrane-specific epitopes on CD30: potentially superior targets for immunotherapy.
      Figure thumbnail gr6
      Figure 6Characteristics of iPS-MSC–derived teratomas with YSTs and ECs. A: YSTs present in teratoma from iPS-MSCs stained with H&E stain showing a loose meshwork of space lined by primitive-appearing cells. B: YSTs stained with HES showing primitive epithelial cells arranged as tubular glands and villi. C: YSTs stained with HES showing cystic spaces lines by a flattened, endothelium-like layer of cells. D: YSTs showing AFP-positive hyaline bodies (arrows). EH: ECs present in teratomas from iPS-MSCs stained with HES (E); additional slides show neoplastic cells positive for AFP (F), cytokeratin AE1/AE3 (G), and CD30 (H). I: Benign teratoma from hESCs stained with HES showing benign tissues negative for CD30. J: Teratoma with YSTs and ECs showing 13% of CD30-positive areas within the total teratoma tissue. Original magnification: ×400 (A and D); ×200 (B, C, and EI).
      Figure thumbnail gr7
      Figure 7Characteristics of iPS-hAFC–derived teratomas with YSTs, ECs, and osteosarcomas. A: YSTs and ECs in teratomas generated from PB06, PB05.1, and PB10; stained with HES showing typical microcystic, glandular-alveolar, and papillary formations (left and middle panels); and after staining for CD30 (right panel). B: Generation of a grade 3 osteosarcoma from an iPS-MSC–derived teratoma (left panel). A small area of mature teratoma was observed showing a benign cartilaginous structure (arrow). This osteosarcoma was devoid of immature or other malignancies (carcinoma or germ cell tumors). An osteosarcoma (middle panel) showing a fibrous stroma–producing osteoid substance (arrow) and tumor cells positive for neuron-specific enolase (NSE; right panel). Original magnification: ×200 [A (left panels) and B (middle panel)]; ×100 [A (middle and right panels) and B (right panel)].
      Figure thumbnail gr8
      Figure 8Molecular characterization of teratomas. A: CD30 expression by using Western blot analysis. Extracts from six teratomas with malignancy and five benign teratomas were tested. The signal corresponding to CD30 and β-actin (control) is indicated. B: Quantitative RT-PCR for the expression of CD30, Bcl-2, and p21/CDKN1A in teratomas from iPS-MSCs with malignant (1) and benign (2) tissues and from teratomas from H9 cells (3). *P < 0.05.
      The mechanisms underlying the different malignant teratoma–forming propensities of iPSCs are not clear. The expression of total (endogenous and exogenous) and exogenous cDNA of Nanog, Lin28, Sox2, and Oct4 was first analyzed by quantitative real-time PCR in iPSC-derived teratomas, and a ratio of the exogenous to the total gene level was obtained. Although expression of the endogenous and exogenous genes was always negative in H9-derived teratomas and in fully reprogrammed iPSC-derived teratomas, we noted a variable expression of the transgenes in partially iPSC-derived teratomas. Interestingly, we showed that exogenous Lin28, Nanog, Sox2, and/or Oct4 were overexpressed in five of six ECs and YSTs; their reactivation was probably attributable to the appearance of malignancy (Table 1).
      We then measured the expression of mRNA levels of p21/CDKN1A in H9, MSCs, iPS-MSCs, and corresponding teratomas. We showed that the transcripts of p21 were drastically repressed in iPS-MSCs after reprogramming (approximately 39-fold) compared with MSCs at a similar level as H9 cells (approximately 60-fold) (see Supplemental Table S1 at http://ajp.amjpathol.org). In the teratoma tissues, in addition to the high expression of CD30 mRNA, p21/CDKN1A, and Bcl-2, mRNAs were significantly down-regulated and up-regulated, respectively, in malignant-derived teratomas, compared with benign-derived teratomas (Figure 8B).
      In a recent series of teratoma assays, after injection of 3 × 106 iPSC-MSCs, we observed, in addition to YSTs and ECs, a new type of cancer corresponding to an aggressive osteosarcoma (Figure 7B), showing that several tumor phenotypes can be revealed by the teratoma assays. This osteosarcoma showed two components: a mass of large pleomorphic spindle cells with hyperchromatic nuclei with prominent nucleoli and numerous mitoses and focal loci of fibrous stroma occasionally producing osteoid substances (Figure 7B). The neoplastic cells were negative for CD99, CD30, AFP, and pan-cytokeratin AE1/AE3 and are immunoreactive for neuron-specific enolase (Figure 7B). This pattern led us to the diagnosis of grade 3 osteosarcoma. This osteosarcoma was devoid of immature or other malignancies (carcinoma or germ cell tumors). Nevertheless, we could observe a small area of mature teratoma (Figure 7B), preferentially showing a benign cartilaginous structure.

      Discussion

      All published data describing the emergence of malignant tissue from iPSCs, in vivo, have been observed in mice by a retrovirus-mediated system encoding Oct4, Sox2, Klf4, and Myc transgenes. In a first report, the four factors were introduced into MEFs containing the Nanog-GFP-IRES-Puror reporter cassette to produce Nanog-iPS cells that were injected into blastocysts to generate adult chimeras.
      • Okita K.
      • Ichisaka T.
      • Yamanaka S.
      Generation of germline-competent induced pluripotent stem cells.
      Herein, 24 (20%) of 121 F1 mice died from weakness or paralysis and developed ganglioneuroblastoma with follicular carcinoma of the thyroid gland, attributable to the reactivation of the Myc retrovirus. The same group has also documented the presence of carcinoma in teratomas originating from secondary neurospheres generated from different murine iPSCs produced by the same cocktail of retroviruses.
      • Miura K.
      • Okada Y.
      • Aoi T.
      • Okada A.
      • Takahashi K.
      • Okita K.
      • Nakagawa M.
      • Koyanagi M.
      • Tanabe K.
      • Ohnuki M.
      • Ogawa D.
      • Ikeda E.
      • Okano H.
      • Yamanaka S.
      Variation in the safety of induced pluripotent stem cell lines.
      To characterize and assess their differentiation potential in vivo, the secondary neurospheres were transplanted into the striata of NOD/SCID mice and the tumor formation was evaluated for up to 45 weeks. In this assay, 9%, 36%, and 83% of mice that underwent transplantation with iPS generated from MEFs, hepatocyte epithelial cells (adult), or tail-tip fibroblasts, respectively, died or became weak within 9 to 19 weeks after transplantation because of the emergence of teratocarcinomas showing large portions of undifferentiated cells. More recently, the degree of safety of iPSCs varied with the efficiency of the reprogramming process.
      • Chan E.M.
      • Ratanasirintrawoot S.
      • Park I.
      • Manos P.D.
      • Loh Y.
      • Huo H.
      • Miller J.D.
      • Hartung O.
      • Rho J.
      • Ince T.A.
      • Daley G.Q.
      • Schlaeger T.M.
      Live cell imaging distinguishes bona fide human iPS cells from partially reprogrammed cells.
      For this issue, the four reprogramming factors (Oct4, Sox2, Klf4, and Myc) were introduced into human fibroblasts obtained after differentiation of the hESC line (H1OGN) using defective murine retroviral backbones. They documented, in a few teratoma assays (n = 3), that partially reprogrammed iPSCs produced embryonic carcinoma-like tumors, attributable to the absence of transgene silencing and the persistence of Oct4 and Sox2 transgene expression.
      In our work, we have produced partially and fully reprogrammed human iPSCs using another strategy based on the transduction of H9-derived MSCs and amniotic fluid–derived cells by defective lentiviral vectors carrying the Oct4, Sox2, Nanog, and Lin28 transgenes. In addition, we assessed their behavior during their differentiation and proliferation using an in vivo teratoma assay in NOD/SCID mice.
      We have demonstrated that early-passage iPSCs were associated, in 42% of the teratomas analyzed, with high-grade malignancies, including aggressive and immature YSTs and ECs. All these lines were partially reprogrammed iPSCs and expressed at least one of the transgenes at mRNA levels. These malignancies were never observed in H9 cells, even in those infected by a GFP lentivirus, nor in fully reprogrammed iPSCs picked at later passages. Furthermore, we showed that the emergence of YSTs and ECs in teratomas is not dependent on the cell type origin used to generate iPSCs and that other types of tumors can be revealed, such as one case of an aggressive osteosarcoma. All YSTs and ECs express CD30, a member of the tumor necrosis factor receptor superfamily that was originally identified as a surface marker for the malignant Reed-Sternberg cells of Hodgkin's disease
      • Dürkop H.
      • Latza U.
      • Hummel M.
      • Eitelbach F.
      • Seed B.
      • Stein H.
      Molecular cloning and expression of a new member of the nerve growth factor receptor family that is characteristic for Hodgkin's disease.
      and for ECs.
      • Nagata S.
      • Ise T.
      • Onda M.
      • Nakamura K.
      • Ho M.
      • Raubitschek A.
      • Pastan I.H.
      Cell membrane-specific epitopes on CD30: potentially superior targets for immunotherapy.
      In all teratomas, CD30 expression was always restricted to undifferentiated and malignant tissues and undetectable in areas of differentiation.
      In contrast to hESCs, the cell surface marker CD30 was never documented before for human iPSCs. Concerning hESCs, there are two contradictory results. Indeed, one study
      • Herszfeld D.
      • Wolvetang E.
      • Langton-Bunker E.
      • Chung T.L.
      • Filipczyk A.A.
      • Houssami S.
      • Jamshidi P.
      • Koh K.
      • Laslett A.L.
      • Michalska A.
      • Nguyen L.
      • Reubinoff B.E.
      • Tellis I.
      • Auerbach J.M.
      • Ording C.J.
      • Looijenga L.H.
      • Pera M.F.
      CD30 is a survival factor and a biomarker for transformed human pluripotent stem cells.
      suggests that only the karyotypically abnormal hESCs express CD30, which confers resistance to apoptosis and can, thus, be used as a biomarker of genetic instability of hESCs. In contrast to this report, it has been documented that CD30 expression does not denote karyotypically abnormal cells and that its expression does not always correlate with reduced cell death.
      • Harrison N.J.
      • Barnes J.
      • Jones M.
      • Baker D.
      • Gokhale P.J.
      • Andrews P.W.
      CD30 expression reveals that culture adaptation of human embryonic stem cells can occur through differing routes.
      Furthermore, CD30 has recently been reported as a marker of undifferentiated hESCs, acting akin to, for example, SSEA4 and TRA-1-60 in recognizing the pluripotent state and was not a marker of genetic instability.
      • Lagarkova M.A.
      • Volchkov P.Y.
      • Philonenko E.S.
      • Pfannkuche K.
      • Prokhorovich M.A.
      • Zabotina T.
      • Hescheler J.
      • Kiselev S.L.
      CD 30 is a marker of undifferentiated human embryonic stem cells rather than a biomarker of transformed hESCs.
      In our experimental setting, CD30 antigen was present in H9 cells with a normal diploid karyotype and was undetectable after their differentiation to MSCs and after EB formation after 14 days of culture. Furthermore, we showed that undifferentiated iPSCs with normal genetic constitution are highly positive for CD30 and that their differentiation led to the CD30 silencing. It is still unknown whether CD30 is a signaling receptor in hESCs and iPSCs. Considering constitutively high levels of CD30 expression and a similar gene signature in different hESC lines,
      • Hoffman L.M.
      • Carpenter M.K.
      Characterization and culture of human embryonic stem cells.
      it is more likely that cells are virtually CD30 unresponsive. Furthermore, we were never able to detect the CD30 ligand by RT-PCR in H9 cells and iPSCs (data not shown). Thus, CD30 transgene overexpression in this case could probably induce activation of the transcription factor NF-κB
      • Herszfeld D.
      • Wolvetang E.
      • Langton-Bunker E.
      • Chung T.L.
      • Filipczyk A.A.
      • Houssami S.
      • Jamshidi P.
      • Koh K.
      • Laslett A.L.
      • Michalska A.
      • Nguyen L.
      • Reubinoff B.E.
      • Tellis I.
      • Auerbach J.M.
      • Ording C.J.
      • Looijenga L.H.
      • Pera M.F.
      CD30 is a survival factor and a biomarker for transformed human pluripotent stem cells.
      by ligand-independent signaling, as has been demonstrated for Hodgkin's lymphoma cells.
      • Horie R.
      • Watanabe T.
      • Morishita Y.
      • Ito K.
      • Ishida T.
      • Kanegae Y.
      • Saito I.
      • Higashihara M.
      • Mori S.
      • Kadin M.E.
      • Watanabe T.
      Ligand-independent signaling by overexpressed CD30 drives NF-kappaB activation in Hodgkin-Reed-Sternberg cells.
      Indeed, overexpression of CD30 in Hodgkin-Reed-Steinberg cells has activated NF-κB, which can promote cell survival through the up-regulation of antiapoptotic genes or the down-regulation of pro-apoptotic genes.
      • Shishodia S.
      • Aggarwal B.B.
      Nuclear factor-κB activation: a question of life or death.
      We showed herein that the presence of YSTs and ECs was correlated with the reactivation of one or several transgenes used for the reprogramming process and, in particular, a strong reactivation of Lin28 in 50% of teratomas with ECs and YSTs. The latter was recently associated with high-grade human
      • Viswanathan S.R.
      • Powers J.T.
      • Einhorn W.
      • Hoshida Y.
      • Ng T.L.
      • Toffanin S.
      • O'Sullivan M.
      • Lu J.
      • Phillips L.A.
      • Lockhart V.L.
      • Shah S.P.
      • Tanwar P.S.
      • Mermel C.H.
      • Beroukhim R.
      • Azam M.
      • Teixeira J.
      • Meyerson M.
      • Hughes T.P.
      • Llovet J.M.
      • Radich J.
      • Mullighan C.G.
      • Golub T.R.
      • Sorensen P.H.
      • Daley G.Q.
      Lin28 promotes transformation and is associated with advanced human malignancies.
      and experimental malignancies, implying the repression of the let-7 microRNA family.
      • Takamizawa J.
      • Konishi H.
      • Yanagisawa K.
      • Tomida S.
      • Osada H.
      • Endoh H.
      • Harano T.
      • Yatabe Y.
      • Nagino M.
      • Nimura Y.
      • Mitsudomi T.
      • Takahashi T.
      Reduced expression of the let-7 microRNAs in human lung cancers in association with shortened postoperative survival.
      We hypothesized that the reactivation of Lin28 could contribute to malignancy, in which levels of all microRNAs in the let-7 family are coordinately repressed, therefore promoting oncogenesis by derepressing target genes involved in tumorigenesis. The two best-characterized let-7 targets are the Ras oncogene family and high-mobility group A2. The let-7 family has regulated both N-Ras and K-Ras mRNA via let-7 binding in the untranslated regions.
      • Johnson S.M.
      • Grosshans H.
      • Shingara J.
      • Byrom M.
      • Jarvis R.
      • Cheng A.
      • Labourier E.
      • Reinert K.L.
      • Brown D.
      • Slack F.J.
      RAS is regulated by the let-7 microRNA family.
      The high-mobility group A proteins are major nonhistone chromosomal proteins involved in transcriptional regulation, controlling differentiation and proliferation. Furthermore, high-mobility group A2 is implicated in tumorigenesis via chromosomal translocations, and this chromatin remodeling protein activated pro-invasive and prometastatic genes.
      • Dangi-Garimella S.
      • Yun J.
      • Eves E.M.
      • Newman M.
      • Erkeland S.J.
      • Hammond S.M.
      • Minn A.J.
      • Rosner M.R.
      Raf kinase inhibitory protein suppresses a metastasis signalling cascade involving LIN28 and let-7.
      Recent works
      • Krizhanovsky V.
      • Lowe S.W.
      Stem cells: the promises and peril of p53.
      • Hong H.
      • Takahashi K.
      • Ichisaka T.
      • Aoi T.
      • Kanagawa O.
      • Nakagawa M.
      • Okita K.
      • Yamanaka S.
      Suppression of induced pluripotent stem cell generation by the p53-p21 pathway.
      have shown the involvement of the p21-p53 pathway during iPSC production in mice models. The tumor suppressor gene, p21/CDKN1A, is a central regulator of cell cycle inhibition that inhibits the activity of cyclin-CDK2 or cyclin-CDK4 complexes, and the expression of this gene is tightly controlled by the tumor suppressor protein p53, through which this protein mediates p53-dependent cell cycle G1 phase arrest. We observed, in our experimental settings, a drastic reduction in the levels of p21/CDKN1A transcripts in iPSCs and H9 cells, showing that the silencing of the p21-p53 pathway is important during the reprogramming, promoting cell proliferation and survival. Interestingly, we documented that malignant teratomas with YSTs and ECs correlated with the down-regulation of p21, and we hypothesize that the permanent suppression of p21 in these tissues should contribute to the loss of cell cycle control and the loss of the p53/p21-dependant G1/S checkpoint in response to DNA damage. Thus, the down-regulation of p21 and the up-regulation of Bcl-2–protecting cells from apoptosis should contribute to malignancy.
      In conclusion, we showed that the in vivo teratoma model and the CD30 immunostaining provide an accessible system for the evaluation of the oncogenicity of patient-specific iPSCs generated by integrative vectors. This model can be valuable for the discovery of novel genes involved in malignancy and in genetic diseases predisposed to cancer. Furthermore, the teratoma model was recently found to be powerful for the study of the immunogenicity of patient-specific iPSCs. Indeed, by using the teratoma model, Zhao et al
      • Zhao T.
      • Zhan Z.N.
      • Rong Z.
      • Xu Y.
      Immunogenicity of induced pluripotent stem cells.
      have studied the immunogenicity of murine iPSCs generated by retroviral and episomal approaches and have shown that most teratomas were immunogenic in the syngeneic recipients, with T-cell infiltration and apparent tissue damage and regression. In addition, it will be pertinent to explore a large series of iPSCs generated by nonintegrative approaches or recombinant proteins and plasmids. The improvement of reprogramming techniques to eliminate the use of integrating vectors is at the center of intense investigation, and integration is dispensable, by using episomal viral vectors, such adenoviruses, Epstein-Barr viruses, plasmids, or recombinant proteins.
      • Stadtfeld M.
      • Nagaya M.
      • Utikal J.
      • Weir G.
      • Hochedlinger K.
      Induced pluripotent stem cells generated without viral integration.
      • Yu J.
      • Hu K.
      • Smuga-Otto K.
      • Tian S.
      • Stewart R.
      • Slukvin I.I.
      • Thomson J.A.
      Human induced pluripotent stem cells free of vector and transgene sequences.
      • Okita K.
      • Hong H.
      • Takahashi K.
      • Yamanaka S.
      Generation of mouse-induced pluripotent stem cells with plasmid vectors.
      • Lyssiotis C.A.
      • Foreman R.K.
      • Staerk J.
      • Garcia M.
      • Mathur D.
      • Markoulaki S.
      • Hanna J.
      • Lairson L.L.
      • Charette B.D.
      • Bouchez L.C.
      • Bollong M.
      • Kunick C.
      • Brinker A.
      • Cho C.Y.
      • Schultz P.G.
      • Jaenisch R.
      Reprogramming of murine fibroblasts to induced pluripotent stem cells with chemical complementation of Klf4.
      However, these strategies are limited by technical difficulties.

      Acknowledgments

      We thank Ingrid Laurendeau, Olivia Bawa, Cécile Bas, Dominique Divers, and Sophie Vacher for their technical assistance and George Daley for the gifts of the MIGR–Oct4, MIGR-Sox2, MIGR-Klf4, and MIGR-Myc plasmids.

      Supplementary data

      • Supplemental Figure S1

        Characterization of the MCSs. Differentiation of MSCs toward adipogenic, chondrogenic, and osteogenic lineages using the Human Mesenchymal Stem Cell Functional Identification Kit (R&D Systems). Adipogenesis differentiation was highly efficient, with oil droplets observed in all of the cells and being immunoreactive for fatty acid–binding protein (FABP)-4 antigen. Chondrogenesis and osteogenesis were also highly efficient, with >90% of cells producing the cartilage-specific extracellular matrix proteins aggrecan and osteocalcin, respectively.

      • Supplemental Figure S2

        Characterization of the MCS, iPS-MSC, and H9 cell lines. A: TRA-1-60 expression by flow cytometric analysis of iPS-MSCs and H9 cells from different passages. The results with iPS-MSCs are not significantly different (P = 1.0) from those observed with H9 cells. B: Hierarchical clustering of H9 cells and iPS-MSC lines and their derived EBs generated after days 4, 8, and 13 of culture, using human pluripotency stem cell TLDA analysis. ESC and iPS-MSC cell lines were clustered according to their differentiation stage. C: Expression of marker genes identifying the three specific germ layers. H9 cells, iPS-MSCs, and derived EBs generated at days 4, 8, and 13 were analyzed using the human pluripotency stem cell TLDA analysis. Data show the log10 expression relative quantification values of ectoderm, endoderm, and mesoderm marker genes normalized to the expression of the H1 ESC line.

      References

        • Takahashi K.
        • Tanabe K.
        • Ohnuki M.
        • Narita M.
        • Ichisaka T.
        • Tomoda K.
        • Yamanaka S.
        Induction of pluripotent stem cells from adult human fibroblast by defined factors.
        Cell. 2007; 131: 1-12
        • Yu J.
        • Vodyanik M.A.
        • Smuga-Otto K.
        • Antosiewicz-Bourget J.
        • Frane J.L.
        • Tian S.
        • Nie J.
        • Jonsdottir G.A.
        • Ruotti V.
        • Stewart R.
        • Slukvin I.I.
        • Thomson J.A.
        Induced pluripotent stem cell lines derived from human somatic cells.
        Science. 2007; 318: 1917-1920
        • Ben-Porath I.
        • Thomson M.W.
        • Carey V.J.
        • Ge R.
        • Bell G.W.
        • Regev A.
        • Weinberg R.A.
        An embryonic stem cell–like gene expression signature in poorly differentiated aggressive human tumors.
        Nat Genet. 2008; 40: 499-507
        • Schoenhals M.
        • Kassambara A.
        • De Vos J.
        • Hose D.
        • Moreaux J.
        • Klein B.
        Embryonic stem cell markers expression in cancers.
        Biochem Biophys Res Commun. 2009; 29: 154-162
        • Rowland B.D.
        • Bernards R.
        • Peeper D.S.
        The KLF4 tumor suppressor is a transcriptional repressor of p53 that acts as a context-dependent oncogene.
        Nat Cell Biol. 2005; 7: 1074-1082
        • Foster K.W.
        • Ren S.
        • Louro I.D.
        • Lobo-Ruppert S.M.
        • McKie-Bell P.
        • Grizzle W.
        • Hayes M.R.
        • Broker T.R.
        • Chow L.T.
        • Ruppert J.M.
        Oncogene expression cloning by retroviral transduction of adenovirus E1A-immortalized rat kidney RK3E cells: transformation of a host with epithelial features by c-MYC and the zinc finger protein GKLF.
        Cell Growth Differ. 1999; 10: 423-434
        • Foster K.W.
        • Frost A.R.
        • McKie-Bell P.
        • Lin C.Y.
        • Engler J.A.
        • Grizzle W.E.
        • Ruppert J.M.
        Increase of GKLF messenger RNA and protein expression during progression of breast cancer.
        Cancer Res. 2000; 15: 6488-6495
        • Wei D.
        • Kanai M.
        • Huang S.
        • Xie K.
        Emerging role of KLF4 in human gastrointestinal cancer.
        Carcinogenesis. 2006; 27: 23-31
        • Adhikary S.
        • Eilers M.
        Transcriptional regulation and transformation by Myc proteins.
        Nat Rev Mol Cell Biol. 2005; 6: 635-645
        • Cawley S.
        • Bekiranov S.
        • Ng H.H.
        • Kapranov P.
        • Sekinger E.A.
        • Kampa D.
        • Piccolboni A.
        • Sementchenko V.
        • Cheng J.
        • Williams A.J.
        • Wheeler R.
        • Wong B.
        • Drenkow J.
        • Yamanaka M.
        • Patel S.
        • Brubaker S.
        • Tammana H.
        • Helt G.
        • Struhl K.
        • Gingeras T.R.
        Unbiased mapping of transcription factor binding sites along human chromosomes 21 and 22 points to widespread regulation of noncoding RNAs.
        Cell. 2004; 116: 499-509
        • Knoepfler P.S.
        Why myc?.
        Cell Stem Cell. 2008; 2: 18-21
        • Gangemi R.M.
        • Griffero F.
        • Marubbi D.
        • Perera M.
        • Capra M.C.
        SOX2 silencing in glioblastoma tumor-initiating cells causes stop of proliferation and loss of tumorigenicity.
        Stem Cells. 2009; 27: 40-48
        • Boer B.
        • Kopp J.
        • Mallanna S.
        • Desler M.
        • Chakravarthy H.
        • Wilder P.J.
        • Bernadt C.
        • Rizzino A.
        Elevating the levels of Sox2 in embryonal carcinoma cells and embryonic stem cells inhibits the expression of Sox2: Oct-3/4 target genes.
        Nucleic Acids Res. 2007; 35: 1773-1786
        • Chen Y.
        • Shi L.
        • Zhang L.
        • Li R.
        • Liang J.
        • Yu W.
        • Sun L.
        • Yang X.
        • Wang Y.
        • Zhang Y.
        • Shang Y.
        The molecular mechanism governing the oncogenic potential of SOX2 in breast cancer.
        J Biol Chem. 2008; 283: 17969-17978
        • Looijenga L.H.
        • Stoop H.
        • de Leeuw H.P.
        • de Gouveia Brazao C.A.
        • Gillis A.J.
        • van Roozendaal K.E.
        • van Zoelen E.J.
        • Weber R.F.
        • Wolffenbuttel K.P.
        • van Dekken H.
        • Honecker F.
        • Bokemeyer C.
        • Perlman E.J.
        • Schneider D.T.
        • Kononen J.
        • Sauter G.
        • Oosterhuis J.W.
        POU5F1 (OCT3/4) identifies cells with pluripotent potential in human germ cell tumors.
        Cancer Res. 2003; 1: 2244-2250
        • Chiou S.H.
        • Yu C.C.
        • Huang C.Y.
        • Lin S.C.
        • Liu C.J.
        Positive correlations of Oct-4 and Nanog in oral cancer stem-like cells and high-grade oral squamous cell carcinoma.
        Clin Cancer Res. 2008; 14: 4085-4095
        • Ezeh U.I.
        • Turek P.J.
        • Reijo R.A.
        • Clark A.T.
        Human embryonic stem cell genes OCT4, NANOG, STELLAR, and GDF3 are expressed in both seminoma and breast carcinoma.
        Cancer. 2005; 15: 2255-2265
        • Zhang S.
        • Balch C.
        • Chan M.W.
        • Lai H.C.
        • Matei D.
        • Schilder J.M.
        • Yan P.S.
        • Huang T.H.
        • Nephew K.P.
        Identification and characterization of ovarian cancer-initiating cells from primary human tumors.
        Cancer Res. 2008; 1: 4311-4320
        • Jeter C.R.
        • Badeaux M.
        • Choy G.
        • Chandra D.
        • Patrawala L.
        • Liu C.
        • Calhoun-Davis T.
        • Zaehres H.
        • Daley G.Q.
        • Tang D.G.
        Functional evidence that the self-renewal gene NANOG regulates human tumor development.
        Stem Cells. 2009; 27: 993-1005
        • Bussolati B.
        • Bruno S.
        • Grange C.
        • Ferrando U.
        • Camussi G.
        Identification of a tumor-initiating stem cell population in human renal carcinomas.
        FASEB J. 2008; 22: 3696-3705
        • Viswanathan S.R.
        • Powers J.T.
        • Einhorn W.
        • Hoshida Y.
        • Ng T.L.
        • Toffanin S.
        • O'Sullivan M.
        • Lu J.
        • Phillips L.A.
        • Lockhart V.L.
        • Shah S.P.
        • Tanwar P.S.
        • Mermel C.H.
        • Beroukhim R.
        • Azam M.
        • Teixeira J.
        • Meyerson M.
        • Hughes T.P.
        • Llovet J.M.
        • Radich J.
        • Mullighan C.G.
        • Golub T.R.
        • Sorensen P.H.
        • Daley G.Q.
        Lin28 promotes transformation and is associated with advanced human malignancies.
        Nat Genet. 2009; 41: 843-848
        • Dangi-Garimella S.
        • Yun J.
        • Eves E.M.
        • Newman M.
        • Erkeland S.J.
        • Hammond S.M.
        • Minn A.J.
        • Rosner M.R.
        Raf kinase inhibitory protein suppresses a metastasis signalling cascade involving LIN28 and let-7.
        EMBO J. 2009; 18: 347-358
        • Rossi D.J.
        • Jamieson C.H.
        • Weissman I.L.
        Stem cells and the pathways to aging and cancer.
        Cell. 2008; 22: 681-696
        • Liu Y.
        • Clem B.
        • Zuba-Surma E.K.
        • El-Naggar S.
        • Telang S.
        • Jenson A.B.
        • Wang Y.
        • Shao H.
        • Ratajczak M.Z.
        • Chesney J.
        • Dean D.C.
        Mouse fibroblasts lacking RB1 function form spheres and undergo reprogramming to a cancer stem cell phenotype.
        Cell Stem Cell. 2009; 4: 336-347
        • Galaup A.
        • Magnon C.
        • Rouffiac V.
        • Opolon P.
        • Opolon D.
        • Lassau N.
        • Tursz T.
        • Perricaudet M.
        • Griscelli F.
        Full kringles of plasminogen (aa 1–566) mediate complete regression of human MDA-MB-231 breast tumor xenografted in nude mice.
        Gene Ther. 2005; 12: 831-842
        • Lensch M.W.
        • Daley G.Q.
        Human embryonic stem cells flock together.
        Nat Biotechnol. 2007; 25: 748-750
        • Bieche I.
        • Parfait B.
        • Le Doussal V.
        • Olivi M.
        • Rio M.C.
        • Lidereau R.
        • Vidaud M.
        Identification of CGA as a novel oestrogen receptor-responsive gene in breast cancer: an outstanding candidate marker to predict the response to endocrine therapy.
        Cancer Res. 2001; 61: 1652-1658
        • Giuliani M.
        • Oudrhiri N.
        • Noman Z.M.
        • Vernochet A.
        • Chouaib S.
        • Azzarone B.
        • Durrbach A.
        • Bennaceur-Griscelli A.
        Human mesenchymal stem cells derived from induced pluripotent stem cells down-regulate NK-cell cytolytic machinery.
        Blood. 2011; 118: 3254-3262
        • Mani S.A.
        • Guo W.
        • Liao M.J.
        • Eaton E.
        • Ayyanan A.
        • Zhou A.Y.
        • Brooks M.
        • Reinhard F.
        • Zhang C.C.
        • Shipitsin M.
        • Campbell L.L.
        • Polyak K.
        • Brisken C.
        • Yang J.
        • Weinberg R.A.
        The epithelial-mesenchymal transition generates cells with properties of stem cells.
        Cell. 2008; 4: 704-715
        • Dürkop H.
        • Foss H.D.
        • Eitelbach F.
        • Anagnostopoulos I.
        • Latza U.
        • Pileri S.
        • Stein H.
        Expression of the CD30 antigen in non-lymphoid tissues and cells.
        J Pathol. 2000; 190: 613-618
        • Nagata S.
        • Ise T.
        • Onda M.
        • Nakamura K.
        • Ho M.
        • Raubitschek A.
        • Pastan I.H.
        Cell membrane-specific epitopes on CD30: potentially superior targets for immunotherapy.
        Proc Natl Acad Sci U S A. 2005; 102: 7946-7951
        • Okita K.
        • Ichisaka T.
        • Yamanaka S.
        Generation of germline-competent induced pluripotent stem cells.
        Nature. 2007; 448: 313-317
        • Miura K.
        • Okada Y.
        • Aoi T.
        • Okada A.
        • Takahashi K.
        • Okita K.
        • Nakagawa M.
        • Koyanagi M.
        • Tanabe K.
        • Ohnuki M.
        • Ogawa D.
        • Ikeda E.
        • Okano H.
        • Yamanaka S.
        Variation in the safety of induced pluripotent stem cell lines.
        Nat Biotechnol. 2009; 27: 743-745
        • Chan E.M.
        • Ratanasirintrawoot S.
        • Park I.
        • Manos P.D.
        • Loh Y.
        • Huo H.
        • Miller J.D.
        • Hartung O.
        • Rho J.
        • Ince T.A.
        • Daley G.Q.
        • Schlaeger T.M.
        Live cell imaging distinguishes bona fide human iPS cells from partially reprogrammed cells.
        Nat Biotechnol. 2009; 27: 1033-1037
        • Dürkop H.
        • Latza U.
        • Hummel M.
        • Eitelbach F.
        • Seed B.
        • Stein H.
        Molecular cloning and expression of a new member of the nerve growth factor receptor family that is characteristic for Hodgkin's disease.
        Cell. 1992; 7: 421-427
        • Herszfeld D.
        • Wolvetang E.
        • Langton-Bunker E.
        • Chung T.L.
        • Filipczyk A.A.
        • Houssami S.
        • Jamshidi P.
        • Koh K.
        • Laslett A.L.
        • Michalska A.
        • Nguyen L.
        • Reubinoff B.E.
        • Tellis I.
        • Auerbach J.M.
        • Ording C.J.
        • Looijenga L.H.
        • Pera M.F.
        CD30 is a survival factor and a biomarker for transformed human pluripotent stem cells.
        Nat Biotechnol. 2006; 24: 351-357
        • Harrison N.J.
        • Barnes J.
        • Jones M.
        • Baker D.
        • Gokhale P.J.
        • Andrews P.W.
        CD30 expression reveals that culture adaptation of human embryonic stem cells can occur through differing routes.
        Stem Cells. 2009; 27: 1057-1065
        • Lagarkova M.A.
        • Volchkov P.Y.
        • Philonenko E.S.
        • Pfannkuche K.
        • Prokhorovich M.A.
        • Zabotina T.
        • Hescheler J.
        • Kiselev S.L.
        CD 30 is a marker of undifferentiated human embryonic stem cells rather than a biomarker of transformed hESCs.
        Cell Cycle. 2008; 15: 3610-3612
        • Hoffman L.M.
        • Carpenter M.K.
        Characterization and culture of human embryonic stem cells.
        Nat Biotechnol. 2005; 23: 699-708
        • Horie R.
        • Watanabe T.
        • Morishita Y.
        • Ito K.
        • Ishida T.
        • Kanegae Y.
        • Saito I.
        • Higashihara M.
        • Mori S.
        • Kadin M.E.
        • Watanabe T.
        Ligand-independent signaling by overexpressed CD30 drives NF-kappaB activation in Hodgkin-Reed-Sternberg cells.
        Oncogene. 2002; 11: 2493-2503
        • Shishodia S.
        • Aggarwal B.B.
        Nuclear factor-κB activation: a question of life or death.
        J Biochem Mol Biol. 2002; 31: 28-40
        • Takamizawa J.
        • Konishi H.
        • Yanagisawa K.
        • Tomida S.
        • Osada H.
        • Endoh H.
        • Harano T.
        • Yatabe Y.
        • Nagino M.
        • Nimura Y.
        • Mitsudomi T.
        • Takahashi T.
        Reduced expression of the let-7 microRNAs in human lung cancers in association with shortened postoperative survival.
        Cancer Res. 2004; 64: 3753-3756
        • Johnson S.M.
        • Grosshans H.
        • Shingara J.
        • Byrom M.
        • Jarvis R.
        • Cheng A.
        • Labourier E.
        • Reinert K.L.
        • Brown D.
        • Slack F.J.
        RAS is regulated by the let-7 microRNA family.
        Cell. 2005; 11: 635-647
        • Krizhanovsky V.
        • Lowe S.W.
        Stem cells: the promises and peril of p53.
        Nature. 2009; 460: 1085-1086
        • Hong H.
        • Takahashi K.
        • Ichisaka T.
        • Aoi T.
        • Kanagawa O.
        • Nakagawa M.
        • Okita K.
        • Yamanaka S.
        Suppression of induced pluripotent stem cell generation by the p53-p21 pathway.
        Nature. 2009; 460: 1132-1135
        • Zhao T.
        • Zhan Z.N.
        • Rong Z.
        • Xu Y.
        Immunogenicity of induced pluripotent stem cells.
        Nature. 2011; 474: 212-215
        • Stadtfeld M.
        • Nagaya M.
        • Utikal J.
        • Weir G.
        • Hochedlinger K.
        Induced pluripotent stem cells generated without viral integration.
        Science. 2008; 7: 945-949
        • Yu J.
        • Hu K.
        • Smuga-Otto K.
        • Tian S.
        • Stewart R.
        • Slukvin I.I.
        • Thomson J.A.
        Human induced pluripotent stem cells free of vector and transgene sequences.
        Science. 2009; 324: 797-801
        • Okita K.
        • Hong H.
        • Takahashi K.
        • Yamanaka S.
        Generation of mouse-induced pluripotent stem cells with plasmid vectors.
        Nat Protoc. 2010; 5: 418-428
        • Lyssiotis C.A.
        • Foreman R.K.
        • Staerk J.
        • Garcia M.
        • Mathur D.
        • Markoulaki S.
        • Hanna J.
        • Lairson L.L.
        • Charette B.D.
        • Bouchez L.C.
        • Bollong M.
        • Kunick C.
        • Brinker A.
        • Cho C.Y.
        • Schultz P.G.
        • Jaenisch R.
        Reprogramming of murine fibroblasts to induced pluripotent stem cells with chemical complementation of Klf4.
        Proc Natl Acad Sci U S A. 2009; 106: 8912-8917