Genomic profiling of tumor initiating prostatospheres
© Duhagon et al; licensee BioMed Central Ltd. 2010
Received: 22 October 2009
Accepted: 25 May 2010
Published: 25 May 2010
The cancer stem cell (CSC) hypothesis proposes that a population of tumor cells bearing stem cell properties is responsible for the origin and maintenance of tumors. Normal and cancer stem cells possess the ability to grow in vitro as self-renewing spheres, but the molecular basis of this phenotype remains largely unknown. We intended to establish a comprehensive culture system to grow prostatospheres (PSs) from both cancer cell lines and patient tumors. We then used gene expression microarrays to gain insight on the molecular pathways that sustain the PS tumor initiating cell (TIC) phenotype.
Traditional stem cell medium (SCM) supplemented with Knockout™SR (KO) allows the propagation of monoclonal PSs from cell lines and primary cells. PSs display gene expression and tumorigenicity hallmarks of TICs. Gene expression analysis defined a gene signature composed of 66 genes that characterize LNCaP and patient PSs. This set includes novel prostate TIC growth factors (NRP1, GDF1, JAG1), proteins implicated in cell adhesion and cytoskeletal maintenance, transcriptional regulators (MYCBP, MYBL1, ID1, ID3, FOS, ELF3, ELF4, KLF2, KLF5) and factors involved in protein biosynthesis and metabolism. Meta-analysis in Oncomine reveals that some of these genes correlate with prostate cancer status and/or progression. Reporter genes and inhibitors indicate that the Notch pathway contributes to prostatosphere growth.
We have developed a model for the culture of PSs, and provide a genomic profile that support CSCs identity. This signature identifies novel markers and pathways that are predicted to correlate with prostate cancer evolution.
There is overwhelming evidence supporting the concept that only a specific group of cells, among the cellular heterogeneity of a tumor, possesses self-renewal and multilineage differentiation potential and is, therefore, responsible for tumor development. These cells, so called "tumor initiating cells" (TICs) or "cancer stem cells" (CSCs), have been documented in most circulating and solid tumors as well as in numerous established cancer cell lines. The expression of adult stem cell surface markers (e.g. CD133, CD44, ESA) as well as the expression of specific embryonic stem cell genes (e.g. OCT3/4, NANOG, SOX2) is one of the hallmarks of the TIC . TICs also display increased potential for anchorage-independent growth, capacity to form spheroids in vitro, and propensity to undergo epithelial-mesenchymal transition. Nevertheless, in the current paradigm, the gold standard property of a TIC is its ability to initiate and generate a tumor in immunodeficient mice. Due to their properties, tumor initiating cells are thought to be responsible for cancer chemo-resistance and relapse, and thus they represent a significant concern for cancer prognosis and therapy.
The isolation of TICs is based on the expression of specific cell surface markers, the ability to pump out Hoechst dye (referred as "side population" or the cells that do not retain the dye), the high aldehyde dehydrogenase 1 activity or the ability to grow in vitro as unattached spheroids in an appropriate medium . Furthermore, we and others have recently proposed that invasion ability could be also used for the enrichment of TICs [3, 4]. The isolation of neural stem cells and the propagation of "neurospheres" was the first breakout in the culture of adult stem cells . Neurospheres are relatively undifferentiated stem cell clones and mounting data have validated their use as a self-renewal model. Using neurosphere culture strategies, unattached clusters of cells with TIC properties, called 'prostatospheres' (PS), have been grown from both malignant and non-malignant prostate tissue . The isolation and culture of TICs as PSs represents a convenient model for their study, since we currently lack universal TIC surface markers, and because it allows the propagation of TICs in their undifferentiated state. However, it remains to be determined to what extent sphere culture selectively enriches for TICs. Likewise, the molecular and cellular basis of PS growth, as well as it relevance for TIC biology, have not been analyzed in depth.
Various signaling pathways, including Wnt, Notch, Hh and PI3K/AKT, have been associated with TICs . Most of these pathways have previously been found to be active in both embryonic and cancer cells. In particular, canonical Notch pathway activation has been reported in normal adult stem cells [8, 9] and malignancies [10, 11], and has recently been connected to self-renewal of TICs . Notch signaling starts with a direct interaction between cell surface ligands and Notch membrane receptor, followed by cleavage of the Notch intracellular domain (NICD) and its translocation to the nucleus where it interacts with CSL/RBPJ transcription factor and triggers the transcription of Notch target genes . In addition, Notch signaling is relevant in prostate gland development, neoplasia and possible for prostate TICs .
Here we show that stem cell medium (SCM) supplemented with Knockout™SR serum replacement (KO) promotes the growth and serial passage of clonally derived PSs. This culture method is suitable not only for established cell lines but also for prostate cancer stem cell primary cultures (PCSC-1, PCSC-2 and PCSC-3, referred to as group as "PCSCs"), providing a valuable model for the integrated study of patient and cell line material. The TIC nature of PSs is supported by their increased tumorigenicity in mouse xenograft studies, as well as by their increased expression of several established stem cell markers.
In the search for the molecular basis of PS growth we performed genome expression studies using microarrays. We found that the promoters of the genes modulated in PSs are enriched in binding sites for known TIC transcription factors, such as c-Myc and TCF3, thereby reinforcing their TIC nature. We also identified 66 PS regulated genes that are common to LNCaP and PCSCs, and highlighted interesting molecules that have recognized roles in developmental processes, tissue remodeling, cell invasion and stem cell niche. Meta-analysis in Oncomine prostate datasets indicates that 32 of the 66 genes in this signature correlate with prostate cancer status. These genes could, therefore, represent prostate CSC markers that have still uninvestigated roles in TIC biology.
Finally, the overexpression of Jag1 in the PS signature, lead us to study the contribution of the Notch pathway to PS biology. We uncovered an overexpression of several Notch signaling genes, an increase in CSL/RBPJ activity in PSs, as well as a reduction of sphere formation as a result of gamma secretase inhibition. This evidence supports a role for the Notch pathway in PS maintenance. In conclusion, our findings support the use of PS cultures in SCM-1%KO as a model for TIC propagation and isolation that unifies cell lines and primary culture systems, and exposes new molecular actors that might be relevant for TIC biology and prostate cancer progression.
SCM-1%KO supports monoclonal PS growth from CD44+CD24-TICs and the total cell line
When the bulk LNCaP cell line was cultured in SCM-1%KO medium, we observed the growth of spheres morphologically similar to those originating from the CD44+CD24-subpopulation (Figure 1A, "total"). After day 7, the cultures displayed a mixed population of adherent cells similar to those grown in RPMI-1640-10% FBS, and PSs (approximately 30%) (Figure 1B). Indeed, PSs could be serially passaged without a decrease in the percentage of sphere forming cells, which denotes the preservation of their self-renewal potential. qRT-PCR analysis of some stem cell genes (CD44, Oct3/4 and Nanog) indicated maintenance of their expression along at least eight generations (data not shown). Therefore, this culture condition seems to allow for the enrichment in TIC spheres from the total LNCaP cell line.
In order to distinguish between cell aggregation and clonal origin of these PSs, we carried out cell tracing experiments. Single cell suspensions were stained with two different dyes in separate reactions and then mixed in equivalent amounts and cultured in SCM-1%KO for 7 days. Following cell up-take, these two dyes become covalently linked to thiol groups in the cytosol and segregate with the cytoplasm. At day 7, the vast majority of the spheres have only one type of fluorescence, indicating that they have a single cell origin (Figure 1C). Two-colored spheres probably arose from peripheral adhesion of small strongly-stained dead cells. This experiment demonstrates that our culture conditions promote the clonal growth of prostatospheres.
SCM-1%KO supports monoclonal PS growth from CD133+TICs isolated from primary cultures
Subsequently, we performed a similar study with three primary prostate cancer stem cells isolated from recurrent prostate tumors of 55-, 56- and 70-year old patients that underwent chemotherapy and radiation therapy prior to tumor extraction. Thereafter, the tumor cells were disaggregated and TIC were sorted for expression of the adult stem cell marker PROMININ1 (CD133) (personal communication, Jay Sharma). All three PCSCs grow as epithelial monolayers in the commercial media, express the TIC markers CD44, CD133, NANOG, OCT3/4, BMI1 and exhibit ALDH activity (Additional file 1A) and anchorage independent growth (Additional file 1B). Additionally, they are highly tumorigenic in NOD/SCID mice, with 100 cells producing tumors in all animals in less than eight weeks (Additional file 1C). These properties are consistent with PCSCs containing a TIC subpopulation.
PSs are more tumorigenic than the parental cell lines
In summary, the mice xenograft experiments demonstrate that PS culture enriches in TICs both in LNCaP as well as in PCSCs.
Gene expression analysis of prostatospheres and differentiated cells
The discovery of the molecular pathways that are required for the maintenance of self-renewing TICs may ultimately lead to targeted therapies against this unique population of tumor cells. In order to identify the molecular basis of PS biology, we carried out whole genome gene expression analysis using microarrays. Microarray data were deposited in GEO (LNCaP expression data in GSE19704 and PCSCs expression data in GSE19713). Gene expression profiles of prostatospheres growing in SCM-1%KO for 7 days were compared to the parental cell line growing in the commercial medium containing 10% FBS.
List of 66 genes modulated in PSs classified by their function.
major histocompatibility complex, class II, DM alpha
cadherin 3, type 1, P-cadherin (placental)
HLA complex group 4
activated leukocyte cell adhesion molecule
Kallmann syndrome 1 sequence
collagen, type VI, alpha 1
MYC induced nuclear antigen
tropomyosin 1 (alpha)
gelsolin (amyloidosis, Finnish type)
Src homology 3
growth differentiation factor 15
UDP-N-acteylglucosamine pyrophosphorylase 1
RAD18 homolog (S. cerevisiae)
methionine sulfoxide reductase B3
lectin galactoside-binding 8 galectins
mannosidase, alpha, class 1C, member 1
STAM binding protein-like 1
heat shock 60 kDa protein 1 (chaperonin)
YOD1 OTU deubiquinating enzyme 1 homolog
tissue factor pathway inhibitor
neuron navigator 3
prostaglandin E receptor 4 (subtype EP4)
sema, immunoglobulin (Ig) and short basic domain
dual specificity phosphatase 6
jagged 1 (Alagille syndrome)
amyloid beta (A4) precursor-like protein 2
c-mer proto-oncogene tyrosine kinase
RIO kinase 3 (yeast)
c-myc binding protein
v-myb myeloblastosis viral oncogene homolog (avian)-like 1
inhibitor of DNA binding 3
inhibitor of DNA binding 1
E74-like factor 3
v-fos FBJ murine osteosarcoma viral oncogene homolog
Kruppel-like factor 2 (lung)
CCAAT/enhancer binding protein (C/EBP), gamma
E74-like factor 4 (ets domain transcription factor)
zinc finger and BTB domain containing 38
Kruppel-like factor 5 (intestinal)
quaking homolog, KH domain RNA binding (mouse)
eukaryotic translation initiation factor 2, subunit 1 alpha
Transmembrane channel-like 4
solute carrier family 39 (zinc transporter), member 10
solute carrier family 25, member 30
solute carrier family 7, member 5
Hbc647 mRNA sequence
FLJ35024, C3orf59, CCDC113, C5orf35, CCDC50, CCDC76, LOC440731
In order to predict which transcription factors might be responsible for the maintenance of the spheroid state, we used the Molecular Signatures Database (MSigDB, http://www.broadinstitute.org/gsea/msigdb/index.jsp) to search overrepresentation of transcription factor binding sites in the promoters of the PS genes modulated ≥2-fold at p ≤ 0.01 (473 genes of PCSCs and 443 in LNCaP). Of the five common TF predictions (Additional file 4), MYC and TCF3/E2A, MYOD and YY1 are well known somatic stem cell transcription factors, whereas the function of REPIN1 is still unknown. Canonical Wnt pathway transcriptional regulators (as Tcf1, Lef1, Tcf8) and other MYC associated TFs (as MAZ) were also found to be overrepresented in the promoter regions of the specific PS datasets (Additional file 4).
Meta-analysis for Correlations of PS genes and prostate cancer status and progression
Notch pathway is upregulated in PSs and may contribute to PS maintenance
The notion that progenitors as well as TICs can grow in vitro as spheroids has been extensively documented for several tissues types. However, the lack of consensus culturing conditions has hampered the elucidation of the molecular pathways that sustain this phenotype. In an attempt to find a comprehensive medium suitable for the selection and propagation of prostate TICs, we tested various embryonic stem cell media. We found that SCM supplemented with Knockout™SR (Gibco-Invitrogen) was able to select TICs, sustaining the growth of PSs not only from normal and cancer cell lines, but also from early passage prostate tumor cell cultures. We demonstrated that these PSs have a monoclonal origin, a relevant aspect that is seldom addressed in the cancer stem cell field. The stem cell identity of these prostatospheres was supported by their extensive in vitro self-renewal potential in the "neurosphere assay", the expression of known prostate TIC markers and the increased tumorigenicity compared to the parental cell line. Indeed, SCM-1%KO culture seems to select in favor of TICs along passages, as was previously observed for normal mouse prostate progenitor cells . It is therefore possible that this culture system provides a means for the long-term enrichment of TICs.
Our analysis of stem cell markers evidenced that PSs preferentially express some classic embryonic stem cell (OCT3/4, Nanog) and TIC (TP63, CDH1, ITGB1, KRT14, KRT5) markers. The down-regulation of AR mRNA in LNCaP is in agreement with the reported reduction in AR expression in prostate TICs. Our data also suggest that prominin-2 could be an interesting new marker for prostate cancer TICs (Figure 2A). PROM2 is a paralogue of PROM1 that is expressed in the prostate and co-localizes with Promin1 in epithelial cells . Indeed, PROM2 protein is enriched in cells located in the basal compartment of the glandular epithelium where only few cells were found to be positive for PROM1 . Another attractive stem cell marker candidate is the HMG-box class transcription factor SOX9, which is expressed in embryonic and somatic progenitor cells . Moreover, Sox9 is required for prostate development  and its expression was found to be restricted to basal epithelium in the normal human adult . Furthermore, prostate tumors express this TF and its expression increases in relapsed hormone refractory tumors. The same report showed that Sox9 is regulated by β-catenin and affects prostate cancer cell proliferation, angiogenesis and invasion; in particular in the LNCaP cell line. Finally, our data also support the observation that NANOG is a key regulator of prostate TICs .
In order to interpret our microarray data, we carried out several analyses. IPA highlighted cell death and cellular movement as the principal gene categories regulated in PSs, which is consistent with the modulation of self-renewal and with the morphological differences between monolayer and spheroid growth. The finding of an enrichment in embryonic TFs related to MYC and TCF-3, which are indeed regulators of cancer stem cell pathways [27, 28], reinforces the TIC identity of the PSs. In fact, our group and others recently provided evidence in favor of a role for Wnt signalling in prostate TICs [13, 29, 30]. MYOD1 has been recently implicated in myoblast maintenance, and, in combination with TCF3, it inhibits myoblast differentiation leading to Rhabdomyosarcomas . YY1 is a pleiotropic transcription factor expressed in prostate glandular epithelium and basal cells, and its expression is positively correlated with prostate cancer metastasis (reviewed in ). Indeed, it is well documented to be involved in specification of mesodermal lineages . The prediction of mesenchymal transcription factor activity (MYOD1 and YY1) could be indicative of an epithelial-mesenchymal transition, which has lately been related to tumor progenitor cell plasticity [4, 33]. In summary, the four known TF predicted to be modulated in PCSCs and LNCaP PSs are known to be active in adult stem cells.
We also defined a set of 66 genes that are analogously and significantly regulated in LNCaP and PCSCs (Table 1). It includes several cell adhesion and cytoskeletal proteins, signaling molecules, transcriptional regulators and metabolic factors. The role of most of these factors in TIC biology has not yet been studied. To date, three complete gene expression datasets of putative prostate TICs have been published. Dubrovska et. al. reported the genomic profiling of PSs derived from DU145 and PC3 cell lines . Strikingly, 11 genes of our 66 gene signature (TFPI, NAV3, MYCBP, MSRB3, MERTK, LOC440731, ID3, ID1, GSN, CCDC50 and C3orf59) overlap with their dataset. Analysis of CD133+ vs. Alpha2lowprostate tumor cell array data published by Birnie et al. reveal a set of four genes common to our signature (JAG1, APLP2, SEMA1A and RIOK3). Comparison of the gene expression profiles of CD133+ and CD133- cells isolated from hormone-refractory prostate cancer biopsies, published by Shepherd et al. , shows that EIF2S1, RAD18, QKI, MAN1C1 and MERTK are significantly regulated in the direction observed in our signature. The small overlap between the studies might be explained by the differences in the methods used for the isolation and culturing of TICs.
Remarkably, 31 of the 66 genes correlated to prostate cancer in our study of Oncomine dataset. Among them, several have already been reported to be associated with prostate cancer. Neuropilin1, for instance, has been proposed as a marker of aggressiveness of prostate carcinoma, as well as other tumor types . Its ability to modulate tumor cell invasion and migration represents one interesting commonality with cancer stem cells . Likewise, GDF15 has been associated with the progression of diseaseto metastasis  and has also been proposed as a prostate cancer biomarker . Its capacity to inhibit proliferation is consistent with the quiescence of TICs. In addition, JAG1, one of the ligands for the Notch receptor, is proposed as a prostate cancer marker . Five other PS up-regulated genes (MERTK13, APLP2 , CLDN 3 , DUSP6  and TFPI ) have been found to be overexpressed in prostate cancer and many other tumor types, but their role in tumor biology is less understood. Finally, two genes are associated with non-prostatic neoplasms: ELF3 with breast cancer  and COL6A1 with astrocytoma . In addition, several of the 18 down-regulated genes have been previously shown to be decreased in prostate cancer. Kruppel-like factor 5, for example, is considered a potential tumor suppressor in prostate and breast cancer  as well as tropomiosin1 in several tumors types . Of note, the tumor suppressor candidate ELF4 facilitates entrance to the cell cycle of quiescent hematopoietic stem cells . Similarly, expression of the cell adhesion protein KAL1 was found to be inversely correlated with metastatic capacity in prostate and other tumor types . Likewise, several studies demonstrated a significant negative correlation between expression of metallothioneins and cancer/progression in prostate tumors and cell lines . Finally, MME is contained in a potential prostate cancer susceptibility locus in prostate cancer ACI rat models . Galectins , NAV3  and QKI  are also correlated to cancer types other than prostate. Lastly, many genes of this 66 gene list have not been extensively studied yet.
Contrary to the oncogenic role assigned by the literature [56, 57], our analysis in Oncomine indicates that SLC7A5 and ID1 are negatively correlated with cancer progression. The explanation for this discrepancy is unclear. In support of our findings, ID1 and ID3 downregulation has been also observed in another TIC profiling study .
Due to the over-expression of the Jag1 Notch1 receptor in PSs, we addressed the involvement of Notch signaling in PS growth. Gene expression analysis of Notch signaling cascade, reporter gene studies, and pathway inhibition experiments presented here indicate a higher activity of the Notch pathway in PSs as compared to the parental cell lines. In support of our hypothesis, Bisson et al. recently reported that the LNCaP derivative, C4-2B, which displays about five times higher Notch activity than original LNCaP cells , expresses more stem cell markers and has higher sphere forming ability than LNCaP . It is also worth mentioning that some of the genes overexpressed in PSs (including NRP1 , SOX9 , ID1 and YY1 ) are thought to be regulated by Notch signaling. Altogether this data support the hypothesis that the Notch pathway is contributing to the PS phenotype.
We presented the isolation and propagation of monoclonal cancer PSs from several cell lines and early passaged tumor cells. In agreement with the TIC identity of these PSs, we demonstrated that they bear the fundamental TIC traits of marker expression, self-renewal and tumorigenicity. The culture medium used in our study could provide a new means for the standardization of prostate TIC cultures, and may possibly reduce the difficulties of the comparison between laboratory cell lines and patient material.
Furthermore, we propose new candidate prostate TIC markers (PROM2 and SOX9) as well as potential biomarkers for prostate cancer status and progression that may be related to the TIC phenotype. Further studies will be needed to understand the role of each one of them. Finally, we show evidence in favor of a role of the Notch pathway in the maintenance of PS phenotype.
LNCaP lymph node metastatic human prostate carcinoma cell line was purchased from ATCC (ATCC® Number CDL-1740™) and cultured following provider's instructions. LNCaP cells used in our study are late passages and have been cultured without addition of androgens.
PCSC-1, PCSC-2 and PCSC-3 prostate cancer stem cells (PCSCs) were purchased from Celprogen®. They were isolated as primary cells from human adult prostate cancer tissue based on the expression of the adult stem cell marker Prominin1 (CD133) (personal communication, Jay Sharma). The cells were cultured with the medium, extracellular matrix coated flasks and guidelines provided by the company.
Prostatosphere culture and assay
In order to obtain prostatospheres from either LNCaP or PCSCs, exponentially growing cultures were dissociated to single cells by standard trypsinization, washed three times with PBS and plated in SCM medium (DMEM:F12 plus 10 ng/mL bFGF, 20 ng/mL EGF, 5 mg/mL insulin, and 0.4% BSA) supplemented with 1% KO serum replacement (Invitrogen/Gibco, p/n 10828-028) at a density of 2500 cells/mL in tissue culture treated flasks. After approximately 7 days, the unattached spheres were removed from the flasks, washed in PBS, incubated in trypsin and gently pipetted until single cells were obtained, as determined by microscopic observation. Trypsin action was immediately inhibited by the addition of Trypsin neutralizing solution (Lonza Ltd., p/n CC-5002). Following centrifugation, the cells were resuspended at the same density in SCM-1%KO medium.
Seven day old spheres were counted and analyzed using a GelCount™ automatic plate scanner (Oxford Optronics) and GelCount Version 0.025.1 software (Oxford Optronics). Plates were scanned at 2400 dpi and the colony detection algorithm was customized for each cell type and culture time.
Flow cytometry of LNCaP
Flow cytometric analysis of LNCaP cells was performed as previously described  using antibodies against CD44 (Caltag human CD44 antibody -PE) and CD24 (Caltag human CD24 antibody -FITC) at 10 μl antibody per one million cells. Forward and side scatters were used to select live cells and scatter and pulse width to select single cells. Control scatters of unstained cells and isotype specific stains were used to set the gates. Since the number of CD44+24-cells obtained in each sorting is extremely low, no assessment of sorting efficiency was performed.
Gene expression analysis
Total RNA isolation was performed using miRNAeasy (Qiagen). cDNA was synthesized with SuperScript III First-Strand Synthesis System for RT-PCR (Invitrogen, p/n 18080-05) using random hexamers and following manufacturer instructions.
Analysis of gene expression by qRT-PCR were performed using TaqMan gene expression assays (Applied Biosystem) in a StepOne Real time PCR machine (Applied Biosystem) using the comparative Ct procedure. For the analysis of prostate CSC markers along the generations, the RNA isolations were carried out with Cell-to-Ct kit (Applied Biosystem) and the resulting cDNA was pre-amplifed prior to the qRT-PCR using the TaqMan PreAmp Master Mix Kit (Applied Biosystem, p/n 4384267) and a pool of stem cell TaqMan Gene Expression Assays for 10 cycles. 18S RNA was not amplified, and was used as the RNA content reference control.
For microarray analysis, RNA integrity was assessed with the Agilent RNA 6000 Nano LabChip Kit (Agilent, p/n 5067-1511) and only samples with RNA integrity numbers (RIN) > 9 were used. Ten ug of total RNA from LNCaP cells were labeled by reverse transcription with Superscript II (Invitrogen) and oligo-dT in the presence of Cy3-dUTP for Universal RNA reference control (Universal Human Reference RNA Stratagene p/n 740000) or Cy5-dUTP for the samples. Whole Human Genome Oligo Microarray 4 × 44 K format gene expression arrays (Agilent, p/n G4112F) were employed for gene expression studies. The hybridization and washes were performed using Agilent reagent and protocols (G2534-90001) and the slides were scanned on a GenePix 4000B scanner (Molecular Devices, Sunnyvale, CA, USA).
PCSC gene expression wasassessed by Affimetrix U133 arrays (Human Genome U133 Plus 2.0 Array; Affymetrix, p/n 900467). In this case, 1 ug of RNA was reverse transcribed with T7-oligo(dT) primer and labeled with biotin using Affymetrix 3' One Cycle Target Labeling kit, following manufacturer's instructions. Three replicates of each group were prepared, labeled, and hybridized to Affymetrix human U133 plus 2.0 GeneChip and scanned on Affymetrix GeneChip scanner 3000. Data were collected using Affymetrix GCOS software.
Statistical and clustering analyses were performed with Partek Genomics Suite software using RMA normalization algorithm for Affymetrix arrays and Loess for Agilent arrays (Partek, Inc.). Differentially expressed genes were identified with ANOVA analysis of the replicates. Unless specifically indicated, genes that are up-or down-regulated more than 2-fold and with a p < 0.01 were considered. Differentially expressed genes were identified with ANOVA analysis of the replicates. For statistical analysis, the three PCSCs were considered as replicates. Due to the use of different array platforms, the gene lists were independently generated for LNCaP and PCSCs. Pathways and signature analysis were also performed in an independent way for each platform. Commonalities in Gene Expression changes, modulated pathways and gene signature matches were obtained as the intersection of the two independent lists (LNCaP and PCSCs).
Cell tracing analysis
To assess the clonal origin of each prostatosphere, dissociated single cells were stained with the fluorescent dye CellTracker green or CellTracker red (Molecular Probes/Invitrogen) at a concentration of 5 μM in separate reactions, following manufacturer instructions. Equal amounts of the two reactions were immediately mixed and plated at a density of 2500 viable cells/mL in SCM-1%KO. Pictures were taken at day 6 using an Olympus IX70 inverted microscope equipped with a triple 41001 FITC EGFP/BODIPY filter (Ex: HQ480/40, Em: HQ535, BS: Q505LP).
Mice xenograft for tumorigenicity determinations
NCI-Frederick is accredited by AAALAC International and follows the Public Health Service Policy for the Care and Use of Laboratory Animals. Animal Care was provided in accordance with the procedures outlined in the 'Guide for Care and Use of Laboratory Animals' (Institute of Laboratory Animal Resources, 1996). Cell cultures were trypsinized and resuspended in SCM-1%KO at the indicated cell density. Fifty μl of the suspensions were combined with 50 μl of BD Matrigel™ Matrix Basement Membrane (BD Bioscience, p/n 354234) on ice and subcutaneously injected into eight-week old male NOD/SCID mice (Jackson Labs). Tumor growth was monitored weekly and dimensions were measured by calipers. Animals were euthanized and tumors were dissected when they reached 1.5 sq.mm.
Soft agar assay
Soft agar assays were performed in 96 well plates using the CytoSelectTM 96-well Cell Transformation Assay (Soft Agar Colony Formation) kit (Cell Biolabs, Inc., p/n CBA-130) following manufacturer instructions. Five hundred to five thousand cells per well were used and the plates were incubated for 10 days at 37°C, 5% CO2. Cells were stained with Neutral Red (Sigma) for 1 hour, scanned in a GelCount plate scanner (Oxford Optronics) at 2400 dpi and analysed using GelCount software.
Meta-analysis for Correlations of PS genes and prostate cancer status and progression
For the assessment of correlation between PS gene set and prostate cancer status and progression we used Oncomine database analysis tool http://www.oncomine.org developed by Chinnaiyan and colleagues . The P-Value Threshold to generate the map was set at 1E-4, whereas the Outlier Rank Threshold level used to generate the Outlier column was 50. Student's t-test was used for analyzing differences between publisheddatasets in the database.
Transfection and reporter gene assay
For LNCaP reporter gene experiments, cultures growing in exponential phase were trypsinized and single cell suspensions were plated in a standard tissue culture 24 well-plates at a density of 40000 cells/well, and then cultured overnight in a 37°C incubator with 5% CO2. The pGL4.17-HES1 plasmid bears the -194 to +160 promoter fragment of the human HES1 gene in the MCS of pGL4.17[luc2/Neo] vector (Promega, p/n E6721). The fragment was excised from a pGL2 basic-HES1 plasmid  generously provided by Ndiaye Delphine (Pasteur Institute, France). pBOS-NICD, a plasmid encoding the Notch intracellular domain (NICD), was a gift from Gerry Weinmaster (UCLA, USA). pGL4-75 (hRluc/CMV) (Promega, p/n E6931) expresses Renilla luciferase gene under the control of the CMV promoter. Plasmids were transiently transfected using Fugene 6 (Roche, p/n 11 815 091 001) following manufacturer's instructions. Two hundred ng of pGL4- and 50 ng of pBOS-NICD and control pGL4-76 were used in each reaction. All assays were performed in triplicate. After 24 h of incubation the cells were lysed and the lysate was transferred to a 96 well plate. Reporter protein activity was measured in an Infinite 2000® Series plate reader (Tecan) using the Dual-Glo® Luciferase Assay (Promega, p/n E2920) following manufacturer's protocol.
Cancer Stem Cell
Ingenuity Pathway Analysis
KnockOut serum replacement
Notch intracellular domain
Prostate Cancer Stem Cell
Stem Cell Medium
Tumor Initiating Cell
We would like to acknowledge Dr. Robert H. Shoemaker and Karen Hite for kindly sharing the GelCount Scanner used for the quantization of PSs.
This publication has been funded in part with Federal funds from the National Cancer Institute, National Institutes of Health, under contract No. N01-CO-12400. This research was supported in part by the Intramural Research Program of the NIH, National Cancer Institute, Center for Cancer Research. The content of this publication does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the U.S. Government.
- Reya T, Morrison SJ, Clarke MF, Weissman IL: Stem cells, cancer, and cancer stem cells. Nature. 2001, 414: 105-111. 10.1038/35102167.PubMedView Article
- Visvader JE, Lindeman GJ: Cancer stem cells in solid tumours: accumulating evidence and unresolved questions. Nature reviews. 2008, 8: 755-768. 10.1038/nrc2499.PubMed
- Yu SC, Bian XW: Enrichment of cancer stem cells based on heterogeneity of invasiveness. Stem cell reviews and reports. 2009, 5: 66-71. 10.1007/s12015-008-9047-8.PubMedView Article
- Klarmann GJ, Hurt EM, Mathews LA, Zhang X, Duhagon MA, Mistree T, Thomas SB, Farrar WL: Invasive prostate cancer cells are tumor initiating cells that have a stem cell-like genomic signature. Clinical & experimental metastasis. 2009, 26: 433-446. 10.1007/s10585-009-9242-2.View Article
- Reynolds BA, Weiss S: Generation of neurons and astrocytes from isolated cells of the adult mammalian central nervous system. Science (New York, NY). 1992, 255: 1707-1710.View Article
- Miki J, Rhim JS: Prostate cell cultures as in vitro models for the study of normal stem cells and cancer stem cells. Prostate Cancer Prostatic Dis. 2007, 11: 32-39. 10.1038/sj.pcan.4501018.PubMedView Article
- Sell S: Cancer and stem cell signaling: a guide to preventive and therapeutic strategies for cancer stem cells. Stem cell reviews. 2007, 3: 1-6. 10.1007/s12015-007-0015-5.PubMedView Article
- Dontu G, Abdallah WM, Foley JM, Jackson KW, Clarke MF, Kawamura MJ, Wicha MS: In vitro propagation and transcriptional profiling of human mammary stem/progenitor cells. Genes & Development. 2003, 17: 1253-1270. 10.1101/gad.1061803.View Article
- Hitoshi S, Seaberg RM, Koscik C, Alexson T, Kusunoki S, Kanazawa I, Tsuji S, Kooy van der D: Primitive neural stem cells from the mammalian epiblast differentiate to definitive neural stem cells under the control of Notch signaling. Genes Dev. 2004, 18: 1806-1811. 10.1101/gad.1208404.PubMed CentralPubMedView Article
- Androutsellis-Theotokis A, Leker RR, Soldner F, Hoeppner DJ, Ravin R, Poser SW, Rueger MA, Bae S-K, Kittappa R, McKay RDG: Notch signalling regulates stem cell numbers in vitro and in vivo. Nature. 2006, 442: 823-826. 10.1038/nature04940.PubMedView Article
- Leong KG, Gao WQ: The Notch pathway in prostate development and cancer. Differentiation; research in biological diversity. 2008, 76: 699-716.PubMedView Article
- Wang Z, Li Y, Banerjee S, Sarkar FH: Emerging role of Notch in stem cells and cancer. Cancer letters. 2009, 279: 8-12. 10.1016/j.canlet.2008.09.030.PubMed CentralPubMedView Article
- Hurt EM, Kawasaki BT, Klarmann GJ, Thomas SB, Farrar WL: CD44+ CD24(-) prostate cells are early cancer progenitor/stem cells that provide a model for patients with poor prognosis. British journal of cancer. 2008, 98: 756-765. 10.1038/sj.bjc.6604242.PubMed CentralPubMedView Article
- Jaszai J, Fargeas CA, Haase M, Farkas LM, Huttner WB, Corbeil D: Robust expression of Prominin-2 all along the adult male reproductive system and urinary bladder. Histochemistry and cell biology. 2008, 130: 749-759. 10.1007/s00418-008-0445-4.PubMedView Article
- Thomsen MK, Butler CM, Shen MM, Swain A: Sox9 is required for prostate development. Developmental biology. 2008, 316: 302-311. 10.1016/j.ydbio.2008.01.030.PubMedView Article
- Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gillette MA, Paulovich A, Pomeroy SL, Golub TR, Lander ES, Mesirov JP: Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proceedings of the National Academy of Sciences of the United States of America. 2005, 102: 15545-15550. 10.1073/pnas.0506580102.PubMed CentralPubMedView Article
- Glinsky GV, Glinskii AB, Stephenson AJ, Hoffman RM, Gerald WL: Gene expression profiling predicts clinical outcome of prostate cancer. The Journal of clinical investigation. 2004, 113: 913-923.PubMed CentralPubMedView Article
- Tomlins SA, Mehra R, Rhodes DR, Cao X, Wang L, Dhanasekaran SM, Kalyana-Sundaram S, Wei JT, Rubin MA, Pienta KJ, Shah RB, Chinnaiyan AM: Integrative molecular concept modeling of prostate cancer progression. Nature genetics. 2007, 39: 41-51. 10.1038/ng1935.PubMedView Article
- Singh D, Febbo PG, Ross K, Jackson DG, Manola J, Ladd C, Tamayo P, Renshaw AA, D'Amico AV, Richie JP, Lander ES, Loda M, Kantoff PW, Golub TR, Sellers WR: Gene expression correlates of clinical prostate cancer behavior. Cancer cell. 2002, 1: 203-209. 10.1016/S1535-6108(02)00030-2.PubMedView Article
- Lapointe J, Li C, Higgins JP, Rijn van de M, Bair E, Montgomery K, Ferrari M, Egevad L, Rayford W, Bergerheim U, Ekman P, DeMarzo AM, Tibshirani R, Botstein D, Brown PO, Brooks JD, Pollack JR: Gene expression profiling identifies clinically relevant subtypes of prostate cancer. Proceedings of the National Academy of Sciences of the United States of America. 2004, 101: 811-816. 10.1073/pnas.0304146101.PubMed CentralPubMedView Article
- Dreesen O, Brivanlou AH: Signaling pathways in cancer and embryonic stem cells. Stem cell reviews. 2007, 3: 7-17. 10.1007/s12015-007-0004-8.PubMedView Article
- Lukacs RU, Lawson DA, Xin L, Zong Y, Garraway I, Goldstein AS, Memarzadeh S, Witte ON: Epithelial stem cells of the prostate and their role in cancer progression. Cold Spring Harbor symposia on quantitative biology. 2008, 73: 491-502.PubMedView Article
- Florek M, Bauer N, Janich P, Wilsch-Braeuninger M, Fargeas CA, Marzesco AM, Ehninger G, Thiele C, Huttner WB, Corbeil D: Prominin-2 is a cholesterol-binding protein associated with apical and basolateral plasmalemmal protrusions in polarized epithelial cells and released into urine. Cell and tissue research. 2007, 328: 31-47. 10.1007/s00441-006-0324-z.PubMedView Article
- Shiota M, Heike T, Haruyama M, Baba S, Tsuchiya A, Fujino H, Kobayashi H, Kato T, Umeda K, Yoshimoto M, Nakahata T: Isolation and characterization of bone marrow-derived mesenchymal progenitor cells with myogenic and neuronal properties. Experimental cell research. 2007, 313: 1008-1023. 10.1016/j.yexcr.2006.12.017.PubMedView Article
- Wang H, Leav I, Ibaragi S, Wegner M, Hu G-f, Lu ML, Balk SP, Yuan X: SOX9 Is Expressed in Human Fetal Prostate Epithelium and Enhances Prostate Cancer Invasion. Book SOX9 Is Expressed in Human Fetal Prostate Epithelium and Enhances Prostate Cancer Invasion (Editor ed.^eds.), City. 2008, 68: 1625-1630.
- Jeter CR, Badeaux M, Choy G, Chandra D, Patrawala L, Liu C, Calhoun-Davis T, Zaehres H, Daley GQ, Tang DG: Functional evidence that the self-renewal gene NANOG regulates human tumor development. Stem cells (Dayton, Ohio). 2009, 27: 993-1005. 10.1002/stem.29.View Article
- Ben-Porath I, Thomson MW, Carey VJ, Ge R, Bell GW, Regev A, Weinberg RA: An embryonic stem cell-like gene expression signature in poorly differentiated aggressive human tumors. Nature genetics. 2008, 40: 499-507. 10.1038/ng.127.PubMed CentralPubMedView Article
- Wong DJ, Liu H, Ridky TW, Cassarino D, Segal E, Chang HY: Module map of stem cell genes guides creation of epithelial cancer stem cells. Cell Stem Cell. 2008, 2: 333-344. 10.1016/j.stem.2008.02.009.PubMed CentralPubMedView Article
- Bisson I, Prowse DM: WNT signaling regulates self-renewal and differentiation of prostate cancer cells with stem cell characteristics. Cell research. 2009, 19: 683-697. 10.1038/cr.2009.43.PubMedView Article
- Blum R, Gupta R, Burger PE, Ontiveros CS, Salm SN, Xiong X, Kamb A, Wesche H, Marshall L, Cutler G, Wang X, Zavadil J, Moscatelli D, Wilson EL: Molecular signatures of prostate stem cells reveal novel signaling pathways and provide insights into prostate cancer. PloS one. 2009, 4: e5722-10.1371/journal.pone.0005722.PubMed CentralPubMedView Article
- Yang Z, MacQuarrie KL, Analau E, Tyler AE, Dilworth FJ, Cao Y, Diede SJ, Tapscott SJ: MyoD and E-protein heterodimers switch rhabdomyosarcoma cells from an arrested myoblast phase to a differentiated state. Book MyoD and E-protein heterodimers switch rhabdomyosarcoma cells from an arrested myoblast phase to a differentiated state (Editor ed.^eds.), City. 2009, 23: 694-707.
- Castellano G, Torrisi E, Ligresti G, Malaponte G, Militello L, Russo AE, McCubrey JA, Canevari S, Libra M: The involvement of the transcription factor Yin Yang 1 in cancer development and progression. Cell cycle (Georgetown, Tex). 2009, 8: 1367-1372.View Article
- Mani SA, Guo W, Liao MJ, Eaton EN, Ayyanan A, Zhou AY, Brooks M, Reinhard F, Zhang CC, Shipitsin M, Campbell LL, Polyak K, Brisken C, Yang J, Weinberg RA: The epithelial-mesenchymal transition generates cells with properties of stem cells. Cell. 2008, 133: 704-715. 10.1016/j.cell.2008.03.027.PubMed CentralPubMedView Article
- Dubrovska A, Kim S, Salamone RJ, Walker JR, Maira SM, Garcia-Echeverria C, Schultz PG, Reddy VA: The role of PTEN/Akt/PI3K signaling in the maintenance and viability of prostate cancer stem-like cell populations. Proceedings of the National Academy of Sciences of the United States of America. 2009, 106: 268-273. 10.1073/pnas.0810956106.PubMed CentralPubMedView Article
- Birnie R, Bryce SD, Roome C, Dussupt V, Droop A, Lang SH, Berry PA, Hyde CF, Lewis JL, Stower MJ, Maitland NJ, Collins AT: Gene expression profiling of human prostate cancer stem cells reveals a pro-inflammatory phenotype and the importance of extracellular matrix interactions. Genome biology. 2008, 9: R83-10.1186/gb-2008-9-5-r83.PubMed CentralPubMedView Article
- Shepherd CJ, Rizzo S, Ledaki I, Davies M, Brewer D, Attard G, de Bono J, Hudson DL: Expression profiling of CD133+ and CD133- epithelial cells from human prostate. The Prostate. 2008, 68: 1007-1024. 10.1002/pros.20765.PubMedView Article
- Latil A, Bieche I, Pesche S, Valeri A, Fournier G, Cussenot O, Lidereau R: VEGF overexpression in clinically localized prostate tumors and neuropilin-1 overexpression in metastatic forms. International journal of cancer. 2000, 89: 167-171. 10.1002/(SICI)1097-0215(20000320)89:2<167::AID-IJC11>3.0.CO;2-9.View Article
- Herman JG, Meadows GG: Increased class 3 semaphorin expression modulates the invasive and adhesive properties of prostate cancer cells. International journal of oncology. 2007, 30: 1231-1238.PubMed
- Karan D, Holzbeierlein J, Thrasher JB: Macrophage Inhibitory Cytokine-1: Possible Bridge Molecule of Inflammation and Prostate Cancer. Book Macrophage Inhibitory Cytokine-1: Possible Bridge Molecule of Inflammation and Prostate Cancer (Editor ed.^eds.), City. 2009, 69: 2-5.
- Brown DA, Stephan C, Ward RL, Law M, Hunter M, Bauskin AR, Amin J, Jung K, Diamandis EP, Hampton GM, Russell PJ, Giles GG, Breit SN: Measurement of Serum Levels of Macrophage Inhibitory Cytokine 1 Combined with Prostate-Specific Antigen Improves Prostate Cancer Diagnosis. Book Measurement of Serum Levels of Macrophage Inhibitory Cytokine 1 Combined with Prostate-Specific Antigen Improves Prostate Cancer Diagnosis (Editor ed.^eds.), City. 2006, 12: 89-96.
- Santagata S, Demichelis F, Riva A, Varambally S, Hofer MD, Kutok JL, Kim R, Tang J, Montie JE, Chinnaiyan AM, Rubin MA, Aster JC: JAGGED1 expression is associated with prostate cancer metastasis and recurrence. Cancer research. 2004, 64: 6854-6857. 10.1158/0008-5472.CAN-04-2500.PubMedView Article
- Tuli A, Sharma M, Wang X, Simone LC, Capek HL, Cate S, Hildebrand WH, Naslavsky N, Caplan S, Solheim JC: Amyloid precursor-like protein 2 association with HLA class I molecules. Cancer Immunol Immunother. 2009, 58: 1419-1431. 10.1007/s00262-009-0657-z.PubMed CentralPubMedView Article
- Kominsky SL: Claudins: emerging targets for cancer therapy. Expert reviews in molecular medicine. 2006, 8: 1-11. 10.1017/S1462399406000056.PubMedView Article
- Kniazev Iu P, Cheburkin Iu V, Spikermann K, Peter S, Jenster G, Bangma KH, Karelin MI, Shkol'nik MI, Urbanskii AI, Evtushenko VI, Ullrich A, Kniazev PG: Gene expression profiles of protein kinases and phosphatases obtained by hybridization with cDNA arrays: molecular portrait of human prostate carcinoma. Molekuliarnaia biologiia. 2003, 37: 97-111.PubMed
- Kurer MA: Protein and mRNA expression of tissue factor pathway inhibitor-1 (TFPI-1) in breast, pancreatic and colorectal cancer cells. Molecular biology reports. 2007, 34: 221-224. 10.1007/s11033-006-9036-4.PubMedView Article
- He J, Pan Y, Hu J, Albarracin C, Wu Y, Dai JL: Profile of Ets gene expression in human breast carcinoma. Cancer biology & therapy. 2007, 6: 76-82.View Article
- Fujita A, Sato JR, Festa F, Gomes LR, Oba-Shinjo SM, Marie SK, Ferreira CE, Sogayar MC: Identification of COL6A1 as a differentially expressed gene in human astrocytomas. Genet Mol Res. 2008, 7: 371-378. 10.4238/vol7-2gmr432.PubMedView Article
- Chen C, Bhalala HV, Vessella RL, Dong JT: KLF5 is frequently deleted and down-regulated but rarely mutated in prostate cancer. The Prostate. 2003, 55: 81-88. 10.1002/pros.10205.PubMedView Article
- Raval GN, Bharadwaj S, Levine EA, Willingham MC, Geary RL, Kute T, Prasad GL: Loss of expression of tropomyosin-1, a novel class II tumor suppressor that induces anoikis, in primary breast tumors. Oncogene. 2003, 22: 6194-6203. 10.1038/sj.onc.1206719.PubMedView Article
- Lacorazza HD, Yamada T, Liu Y, Miyata Y, Sivina M, Nunes J, Nimer SD: The transcription factor MEF/ELF4 regulates the quiescence of primitive hematopoietic cells. Cancer cell. 2006, 9: 175-187. 10.1016/j.ccr.2006.02.017.PubMedView Article
- Kauffman EC, Barocas DA, Chen YT, Yang XJ, Scherr DS, Tu JJ: Differential expression of KAI1 metastasis suppressor protein in renal cell tumor histological subtypes. The Journal of urology. 2009, 181: 2305-2311. 10.1016/j.juro.2009.01.003.PubMedView Article
- Yamashita S, Suzuki S, Nomoto T, Kondo Y, Wakazono K, Tsujino Y, Sugimura T, Shirai T, Homma Y, Ushijima T: Linkage and microarray analyses of susceptibility genes in ACI/Seg rats: a model for prostate cancers in the aged. Cancer research. 2005, 65: 2610-2616. 10.1158/0008-5472.CAN-04-2932.PubMedView Article
- Bidon-Wagner N, Le Pennec JP: Human galectin-8 isoforms and cancer. Glycoconjugate journal. 2004, 19: 557-563. 10.1023/B:GLYC.0000014086.38343.98.PubMedView Article
- Karenko L, Hahtola S, Paivinen S, Karhu R, Syrja S, Kahkonen M, Nedoszytko B, Kytola S, Zhou Y, Blazevic V, Pesonen M, Nevala H, Nupponen N, Sihto H, Krebs I, Poustka A, Roszkiewicz J, Saksela K, Peterson P, Visakorpi T, Ranki A: Primary cutaneous T-cell lymphomas show a deletion or translocation affecting NAV3, the human UNC-53 homologue. Cancer research. 2005, 65: 8101-8110. 10.1158/0008-5472.CAN-04-0366.PubMedView Article
- Yin D, Ogawa S, Kawamata N, Tunici P, Finocchiaro G, Eoli M, Ruckert C, Huynh T, Liu G, Kato M, Sanada M, Jauch A, Dugas M, Black KL, Koeffler HP: High-resolution genomic copy number profiling of glioblastoma multiforme by single nucleotide polymorphism DNA microarray. Mol Cancer Res. 2009, 7: 665-677. 10.1158/1541-7786.MCR-08-0270.PubMedView Article
- Sakata T, Ferdous G, Tsuruta T, Satoh T, Baba S, Muto T, Ueno A, Kanai Y, Endou H, Okayasu I: L-type amino-acid transporter 1 as a novel biomarker for high-grade malignancy in prostate cancer. Pathology international. 2009, 59: 7-18. 10.1111/j.1440-1827.2008.02319.x.PubMedView Article
- Coppe JP, Itahana Y, Moore DH, Bennington JL, Desprez PY: Id-1 and Id-2 proteins as molecular markers for human prostate cancer progression. Clin Cancer Res. 2004, 10: 2044-2051. 10.1158/1078-0432.CCR-03-0933.PubMedView Article
- Farnie G, Clarke RB: Mammary stem cells and breast cancer--role of Notch signalling. Stem cell reviews. 2007, 3: 169-175. 10.1007/s12015-007-0023-5.PubMedView Article
- Sorensen I, Adams RH, Gossler A: DLL1-mediated Notch activation regulates endothelial identity in mouse fetal arteries. Blood. 2009, 113: 5680-5688. 10.1182/blood-2008-08-174508.PubMedView Article
- Muto A, Iida A, Satoh S, Watanabe S: The group E Sox genes Sox8 and Sox9 are regulated by Notch signaling and are required for Muller glial cell development in mouse retina. Experimental eye research. 2009, 89 (4): 549-58. 10.1016/j.exer.2009.05.006.PubMedView Article
- Wang HC, Perry SS, Sun XH: Id1 attenuates Notch signaling and impairs T cell commitment by elevating Deltex1 expression. Molecular and cellular biology. 2009, 29 (17): 4640-52. 10.1128/MCB.00119-09.PubMed CentralPubMedView Article
- Yeh TS, Lin YM, Hsieh RH, Tseng MJ: Association of transcription factor YY1 with the high molecular weight Notch complex suppresses the transactivation activity of Notch. The Journal of biological chemistry. 2003, 278: 41963-41969. 10.1074/jbc.M304353200.PubMedView Article
- Rhodes DR, Kalyana-Sundaram S, Mahavisno V, Varambally R, Yu J, Briggs BB, Barrette TR, Anstet MJ, Kincead-Beal C, Kulkarni P, Varambally S, Ghosh D, Chinnaiyan AM: Oncomine 3.0: genes, pathways, and networks in a collection of 18,000 cancer gene expression profiles. Neoplasia (New York, NY). 2007, 9: 166-180.View Article
- Jarriault S, Brou C, Logeat F, Schroeter EH, Kopan R, Israel A: Signalling downstream of activated mammalian Notch. Nature. 1995, 377: 355-358. 10.1038/377355a0.PubMedView Article
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.