p/CIP, also known as steroid receptor coactivator 3 (SRC-3)/Nuclear Receptor Coactivator 3 (NCoA3), is a transcriptional coactivator that binds liganded nuclear hormone receptors, as well as other transcription factors, and facilitates transcription through direct recruitment of accessory factors. We have found that p/CIP is highly expressed in undifferentiated mouse embryonic stem cells (mESCs) and is downregulated during differentiation. siRNA-mediated knockdown of p/CIP decreased transcript levels of Nanog, but not Oct4 or Sox2. Microarray expression analysis showed that Klf4, Tbx3, and Dax-1 are significantly downregulated in mESCs when p/CIP is knocked down. Subsequent chromatin immunoprecipitation (ChIP) analysis demonstrated that Tbx3, Klf4, and Dax-1 are direct transcriptional targets of p/CIP. Using the piggyBac transposition system, a mouse ESC line that expresses Flag-p/CIP in a doxycycline-dependent manner was generated. p/CIP overexpression increased the level of target genes and promoted the formation of undifferentiated colonies. Collectively, these results indicate that p/CIP contributes to the maintenance of ESC pluripotency through direct regulation of essential pluripotency genes. To better understand the mechanism by which p/CIP functions in ESC pluripotency, we integrated our ChIP and transcriptome data with published protein-protein interaction and promoter occupancy data to draft a p/CIP gene regulatory network. The p/CIP gene regulatory network identifies various feed-forward modules including one in which p/CIP activates members of the extended pluripotency network, demonstrating that p/CIP is a component of this extended network. Stem Cells2014;32:204–215
Embryonic stem cells (ESCs) are pluripotent cells isolated from the inner cell mass of the preimplantation embryo that have the capacity to give rise to any cell type derived from the three primary germ layers [1, 2]. A key hallmark of ESCs is their ability to self-renew, or continuously divide without undergoing differentiation . The self-renewal and pluripotency of ESCs are governed by complex transcriptional regulatory networks involving various transcription factors, coregulators, and epigenetic modifiers, which integrate key signaling pathways activated by extrinsic factors, such as leukemia inhibitory factor (LIF) and BMP [4, 5]. Oct4, Sox2, and Nanog represent the core transcription factors that are essential for the maintenance of ESC self-renewal and pluripotency [6-11]. They regulate their own as well as each other's expression, and function in a cooperative manner to activate the expression of genes involved in self-renewal and pluripotency, and repress the expression of developmental and differentiation genes . High-throughput, genome-wide studies have identified additional factors that intimately associate with the core factors to promote pluripotency and suppress differentiation such as Tbx3, Klf4, Dax-1 (Nr0b1), c-Myc, Esrrb, Sall4, Tcf3, and Zfx [13-17] forming an extended pluripotency network.
p/CIP, also known as steroid receptor coactivator 3 (SRC-3), is a member of the p160 SRC family of nuclear receptor coactivators . p/CIP is a 160 kDa protein comprised of three structural domains which are highly conserved among its family members, SRC-1 and SRC-2, and facilitate its adapter function . p/CIP binds ligand-activated nuclear hormone receptors such as the estrogen receptor, progesterone receptor, and retinoic acid (RA) receptor, as well as other classes of transcription factors such as NF-κB, E2F1 , and STATs . In addition, p/CIP recruits additional coregulators, such as CARM1  and p300 , to remodel nearby chromatin thereby promoting assembly of the general transcription machinery at target gene promoters for activation of transcription. Studies have shown that p/CIP is required for normal somatic growth as knockout mice demonstrate retarded growth, delayed puberty, reduced female reproductive function, blunted mammary gland development, and impaired IGF-1 signaling [25, 26]. Furthermore, p/CIP has been identified as an oncogene based on its ability to initiate tumor formation in mice when overexpressed . Amplification and overexpression of p/CIP are found in various cancers, including breast and ovarian, and its overexpression is generally associated with larger tumor size, higher tumor grade, and poor disease-free survival . Through efforts aimed at uncovering novel regulators of self-renewal and pluripotency, p/CIP has been more recently identified as a candidate self-renewal gene in mouse ESCs. Microarray expression analyses have demonstrated that p/CIP is highly expressed in embryonic tissues and its expression is downregulated during differentiation [29, 30]. Interestingly, global chromatin immunoprecipitation (ChIP) analysis has shown that the promoters of p/CIP and SRC-1 are targeted by Nanog and Oct4/Sox2, respectively [14, 17, 30]. Furthermore, a genome-wide RNAi screen demonstrated that depletion of p/CIP resulted in differentiation of mESCs . These results suggest that p/CIP may play a role in the regulation of ESC identity.
In this study, we examine the role of p/CIP in mouse ESC self-renewal. We show that p/CIP is highly expressed in undifferentiated mouse ESCs and that it is downregulated during ESC differentiation induced by both RA-treatment and LIF withdrawal. Microarray expression analysis identified 46 genes that demonstrated significant changes in expression following p/CIP knockdown. These genes include the essential self-renewal genes Klf4, Tbx3, and Dax-1. ChIP analysis revealed p/CIP occupancy within the 1 kb proximal promoter regions of each of these genes, suggesting they are direct transcriptional targets of p/CIP. Conditional overexpression of p/CIP using the piggyBac transposition system resulted in increased expression of Klf4, Tbx3, and Dax1. Importantly, in a clonogenic assay, ESCs overexpressing p/CIP demonstrated increased efficiency in the formation of undifferentiated colonies. These results suggest that p/CIP contributes to the maintenance of self-renewal and pluripotency in mouse ESCs by facilitating the transcriptional activation of essential genes.
Materials and Methods
ESC Culture and Reagents
R1 and E14 ESCs were cultured on a layer of mitomycin C-inactivated mouse embryonic fibroblasts in Dulbecco's modified Eagle's medium (Wisent), supplemented with 15% fetal bovine serum (ESC qualified, Wisent St-Bruno, Quebec, http://www.wisent.ca), 100 µM β-mercaptoethanol (Sigma), 2 mM l-glutamine, 100 µM nonessential amino acids, 1 mM sodium pyruvate, penicillin/streptomycin (all from Gibco Burlington, Canada, http://www.lifetechnologies.com), and 1,000 units/ml of LIF (ESGRO, from Millipore, Billerica, Massachusetts, USA, www.millipore.com). ESCs were maintained at 37°C and 5% CO2, and passaged every second day at a ratio of 1:5 by washing with PBS (Wisent), dissociating with 0.05% trypsin (Wisent) for 5 minutes at 37°C, and resuspending in ESC media. To induce differentiation, cells were cultured in LIF-deficient ESC media, or ESC media supplemented with 1 µM all-trans RA (Sigma-Aldrich, Oakville, Canada, http://www.sigmaaldrich.com). For doxycycline-induction, cells were cultures in ESC media containing 2 µg/ml doxycycline (Sigma-Aldrich).
R1 ESCs were plated on six-well tissue culture-treated dishes coated with 0.1% gelatin at a density of 0.5 × 106 cells per well. After 6–8 hours, cells were transfected with 50 pmol siRNA using Transfectin (BioRad Mississauga, Canada, www.bio-rad.com) according to manufacturer's recommendations. Media were changed after 24 hours and cells were harvested 48 hours post-transfection for RNA extraction and Western Blotting. Two individual siRNAs targeting distinct regions of the p/CIP transcript were purchased from Dharmacon: p/CIP siRNA1, 5′ CCGGAAAGGUUGUCAAUAU; p/CIP siRNA2, 5′ GGAACAAGGUCCUCACGGG. ON-TARGET Control pool siRNA was purchased from Dharmacon Waltham, Massachusetts, http://www.thermoscientificbio.com.
Generation of Doxycycline-Inducible Flag-p/CIP-IRES-βgeo Clones
Doxycycline-inducible Flag-p/CIP clones were generated using the piggyBac transposition assay as previously described . Following PCR amplification of Flag-p/CIP using attB-modified primer pairs, the product was cloned into pDONR221 using BP clonase (Invitrogen) according to manufacturer's instructions. The resulting entry vector was then used to deliver Flag-p/CIP into the doxycycline-inducible vector PB-TET (Addgene) using LR clonase (Invitrogen). PB-TET-Flag-p/CIP, PB-CA-rtTA (Addgene), and PBase pCyL43 (Sanger) were then cotransfected into E14 mouse ESCs by electroporation and selected with blasticidin.
Protein Extraction and Western Blot Analysis
Protein extracts were obtained using RIPA lysis buffer consisting of 20 mM Tris (pH = 7.9), 300 mM KCl, 0.1% Nonidet P40, 10% glycerol, 0.1 mM dithiothreitol, 0.5 mM EDTA, 0.5 mM EGTA, and protease inhibitor cocktail. Lysates were cleared by centrifugation at 15,000 rpm for 10 minutes at 4°C and protein concentrations were determined using the Bradford reagent assay (Bio-Rad). Protein samples were separated by sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis, transferred to polyvinylidene fluoride membrane, and blocked overnight in PBS containing 0.1% Tween 20 and 5% nonfat dried milk. Membranes were probed with specific primary antibodies for 2 hours at room temperature and incubated with appropriate horseradish peroxidase-conjugated secondary antibodies for 1 hour. Signals were detected using enhanced chemiluminescence according to the manufacturer's recommendations (Amersham). Affinity purified anti-p/CIP antibody was generated as previously described . Antibodies against Flag (M2) and αTubulin were purchased from Sigma. Other antibodies purchased are as follows: Oct4 (611202; BD Transduction Laboratories), Sox2 (L1D6A2; Cell Signaling Technologies Danvers, Massachusetts, www.cellsignal.com), and Nanog (560259; BD Pharmingen).
RNA Isolation and Quantitative Real-Time PCR
Total cellular RNA was isolated using the RNeasy Mini kit (Qiagen Toronto, Ontario, www.qiagen.com) according to the manufacturer's instructions. For quantitative real-time PCR (qRT-PCR) analysis, 2 µg of RNA was reverse transcribed to cDNA using the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems Foster City, California, www.appliedbiosystems.com). Amplification was detected using predesigned and quality tested 5′ nuclease Taqman probes (Applied Biosystems). Reactions were performed in duplicate in a 96-well format according to manufacturer's recommendations (Applied Biosystems) using an Mx3000P real-time instrument (Stratagene Mississauga, Canada, http://www.genomics.agilent.com). Gene expression levels were determined based on the cycle threshold (Ct) value for each reaction and normalized to GAPDH or 18S.
ESCs were cross-linked with 1% formaldehyde for 10 minutes at room temperature and immediately washed twice with ice-cold PBS (containing 0.5 mM phenylmethanesulfonylfluoride) to terminate cross-linking. Cell pellets were resuspended in lysis buffer (50 mM Tris-HCl µM [pH 8.1], 10 mM EDTA, 1% SDS, and protease inhibitors) and incubated on ice for 10 minutes. Lysates were sonicated to yield chromatin fragments approximately 1 kb in length, and ChIP experiments were performed as previously described , using p/CIP or rabbit IgG antibodies. Immunoprecipitated DNA was purified using PCR purification spin columns (Qiagen). qRT-PCR was also performed as previously described , using Brilliant Sybr green master mix (Applied Biosystems) on an Mx3000P real-time instrument (Stratagene). The signal obtained using rabbit anti-IgG antibody was negligible and for each experiment this value was subtracted from the signal obtained using the specific antibody. The following primer pairs were used: Klf4, forward-5′ GCCGCTCTCTTTCATAGCAG, reverse-5′ ATTATCCGCGTGACTCATCC; Tbx3, forward-5′ ACGTCTGCCACGATAAGTCC, reverse-5′ GGGGTGTGGGTGTAGAGAGA; Dax-1, forward-5′ GGCATTTATTTCTGCCTCCA, reverse-5′GGCTCTGTTCCAACTCTTGC; Chr3, forward-5′ATAGGTACACCAAGGACAGTTAGGA, reverse-5′ AGTTATCACATTTTCAGAGCCCA.
Inducible, Flag-p/CIP overexpressing cells grown in the presence or absence of doxycycline for 48 hours were rinsed once in 0.1 M phosphate buffer pH 7.3 at room temperature. Cells were then fixed in fix solution consisting of 0.2% glutaraldehyde, 5 mM EGTA pH 7.3, 20 mM magnesium chloride, 0.1 M sodium phosphate pH 7.3 for 5 minutes at room temperature and then washed with wash buffer (20 mM magnesium chloride, 0.01% deoxycholate, 0.02% Nonidet P40, 0.1 M sodium phosphate pH 7.3) three times for 5 minutes at room temperature. X-gal stain (1 mg/ml X-gal in dimethylformamide, 5 mM potassium ferrocyanide, 5 mM potassium ferricyanide in wash buffer) was added to cells and plates were incubated overnight at 37°C. Stain was removed, plates were washed once with wash solution, and then stored in wash buffer at 4°C. Cells were viewed on an Olympus IX70 inverted microscope and images were obtained using Image-Pro Plus 6.2 (Media Cybernetics Inc., Bethesda, MD).
Colony-Forming Assay and Alkaline Phosphatase Staining
Inducible ESCs were seeded in a single-cell suspension at a density of 500 cells/plate onto 10 cm gelatin-coated tissue culture plates in ESC media containing LIF as previously described [30, 34, 35]. Colonies were allowed to form over 7 days at 37°C. Plates were washed once with ice-cold phosphate buffered saline and fixed in 10% cold neutral formalin buffer (NFB: 10 ml formalin in 100 ml PBS) for 45 minutes at room temperature. NFB was removed and cells were washed three times with cold PBS. Alkaline phosphatase (ALP) stain was made by dissolving 0.01 g Naphthol AS MX-PO4 (N4875, Sigma) in 400 µl of N,N-dimethylformamide (Sigma), 25 ml 0.2 M Tris-HCl (pH = 8.3) and 0.06 g red violet lysogeny broth salt in 25 ml ddH2O and filtered through Whatman's No. 1 filter paper. ALP stain was added to the fixed ESCs and incubated for 45 minutes at room temperature. Stained cells were washed three times with PBS and imaged with the microscope and software described above. ESC colonies were categorized as undifferentiated if the colony stained positive for ALP expression and had the rounded, smooth-edged morphology of undifferentiated ESCs, partially undifferentiated if the colony was stained positive for ALP expression in the center with flattened edges of differentiated ESCs, and fully differentiated if the colony was large and flattened with little or no ALP expression.
Microarray Hybridization and Analysis
Total RNA was isolated from R1 ESCs transfected with p/CIP siRNA1 or 2, or nonspecific RNA as a control, as described above. Duplicate experiments were performed. All sample labeling and GeneChip processing were performed at the London Regional Genomic Centre (Robarts Research Institute, London, ON, Canada; http://www.lrgc.ca). RNA quality was assessed using the Agilent 2100 Bioanalyzer (Agilent Technologies Inc., Palo Alto, CA) and the RNA 6000 Nano kit (Caliper Life Sciences, Mountain View, CA). cDNA was synthesized from each sample, end labeled, and hybridized to Mouse Gene 1.0 ST arrays. GeneChips were scanned with the GeneChip Scanner 3000 7G (Affymetrix, Santa Clara, CA). Probe level (.CEL file) data were generated using Affymetrix Comman Console v1.1 and summarized to gene level data in Partek Genomics Suite v6.5 (Partek, St. Louis, MO) using the Robust Multi-Array algorithm adjusted for GC content . Partek was used to determine gene level ANOVA p-values and fold changes using a χ2 test. A list of genes exhibiting a greater than 1.5-fold change in expression with a p-value <.05 in both replicates for each siRNA compared to control was generated (accession number: GSE50017). Those genes common to both siRNAs were further identified. These genes were imported into Ingenuity Pathway Systems (http://www.ingenuity.com) to group statistically significant genes based on function and only those found in the system database were used for further analysis. Comparison of the number of genes that participated in a given function relative to the total number of occurrences of those genes in all functional annotations stored in the Ingenuity Pathways database was used to assign a significance value associated with a particular function.
To generate the p/CIP network in mouse ESCs, the following data were integrated and visualized using Cytoscape as we have done previously [30, 34, 37, 38]. Differentially expressed genes in mESCs upon p/CIP knockdown (Table 1—this manuscript); known protein-protein interactors of p/CIP, including p300 , STAT6 , E2F1 , NFκβ , CBP , and CARM-1 ; promoter occupancy data [14, 39] and Figure 4 this manuscript. Color of the nodes represents differential gene expression in mESCs upon knockdown of p/CIP. Square nodes and purple connections represent protein-protein interactions from published data. Blue edges represent promoter occupancy data [14, 39].
Table 1. Genes significantly altered upon p/CIP knockdown.
Abbreviation: ER, estrogen receptor.
Calcium binding protein
Histocompatibility 2 (Ig)
Lymphocyte antigen 6 complex
Nuclear dot protein
Protein arginine methyltransferase 8
Porcupine ER protein
Junction adhesion molecule 2
Secreted frizzled-related protein 1
Zing finger transcription factor
Integrin alpha 9
T-box 3 transcription factor
Muscle and microspikes RAS
Nuclear hormone receptor
Tet oncogene family member 2
Keratinocyte differentiation-associated protein
SAM domain-containing protein 9
Kruppel-like transcription factor 4
Adenylate kinase 3-like 1
Cell adhesion molecule
Putative uncharacterized protein gene
Prostate androgen-regulated mucin-like protein
Zinc finger homeobox 2
G protein-coupled receptor 83
Solute carrier family 25
RIKEN cDNA gene
Glutathione peroxidase 2
NIMA-related expressed kinase 7
Ahnak nuclear protein 2
p/CIP Expression is Downregulated During mESC Differentiation
To examine p/CIP levels during mouse ESC differentiation, changes in p/CIP mRNA and protein expression were assessed in the R1 mouse ESC line following induction of differentiation by either treatment with 1 µM RA or LIF withdrawal. qRT-PCR analysis demonstrated downregulation of p/CIP mRNA following both RA-treatment and LIF withdrawal, to approximately 50% and 25% of control levels, respectively, after 96 hours (Fig. 1A). Expression levels of key pluripotency factors (Oct4, Nanog, and Sox2) were also reduced, thereby confirming induction of differentiation. Western blot analysis also showed a decline in p/CIP expression at the protein level during differentiation, which followed that observed at the mRNA level (Fig. 1B), with p/CIP being downregulated more rapidly in response to LIF withdrawal. These results support the hypothesis that p/CIP is highly expressed in undifferentiated mouse ESCs and that its expression is downregulated during differentiation [29, 30].
p/CIP Knockdown Results in the Downregulation of Genes Required for the Maintenance of Pluripotency
To assess whether the expression levels of Oct4, Nanog, and Sox2 are dependent on p/CIP, knockdown experiments in R1 ESCs were conducted. Although qRT-PCR analysis did not detect significant variation for transcripts encoding Oct4 or Sox2, a significant decrease in mRNA expression levels of Nanog (30%–40%) was observed for both siRNAs, where p/CIP expression was effectively knocked down by at least 70% (Fig. 2A). However, Western blot analysis using antibodies against Oct4, Nanog, and Sox2 showed no significant changes in protein levels of these pluripotency markers at this time point (Fig. 2B). To determine if p/CIP occupies the promoter regions of each of these essential self-renewal genes, ChIP analysis was also performed. Cells were grown to the appropriate confluence fixed with formaldehyde and lysed. The chromatin was sheared by sonication to yield approximately 1 kb fragments followed by immunoprecipitation with anti-p/CIP antibody. The purified DNA was then analyzed by qPCR using specific primers flanking the promoter regions of Oct4, Sox2, and Nanog. Surprisingly, p/CIP was specifically detected on all three promoters to varying degrees of abundance despite the observation that no change in expression levels was observed (Fig. 2C) following p/CIP knockdown.
To identify more global changes in gene expression associated with p/CIP knockdown, microarray expression analysis of R1 mouse ESCs silenced for p/CIP expression was conducted. Western blotting confirmed efficient knockdown of p/CIP after 48 hours (Fig. 3A). Initial analysis identified genes that displayed statistically significant changes in gene expression (>1.5-fold, p < .05) in both replicates for each siRNA. From this preliminary list, those genes that were common to both siRNAs were identified. By this approach, 42 downregulated genes and 5 upregulated genes were identified (Fig. 3B) and are listed in Table 1. Ingenuity functional analysis was used to analyze the nature of the genes identified from the microarray screen (Fig. 3C). This analysis identified genes involved in various aspects of cancer including cellular growth and proliferation, cell cycle, and cell death. Additionally, genes related to tissue development and morphology, embryonic development, and gene expression were also identified. Consistent with the qRT-PCR data, we observed no significant changes in Oct4 or Sox2 levels following p/CIP knockdown. However, although it did not make the final list of genes due to the stringency used in analysis, there was an approximate 1.25-fold downregulation of Nanog expression that was statistically significant (p < .05) for both siRNAs (data not shown). Interestingly, the list of downregulated genes contained a number of genes encoding transcription factors that have been identified as essential regulators of self-renewal and pluripotency, including Klf4, Tbx3, and Dax-1(Nr0b1) [11, 40, 41].
p/CIP Directly Regulates Klf4, Tbx3, and Dax1 in mESCs
Kruppel-like factor 4 (Klf4), an essential transcription factor for somatic cell reprogramming , co-operates with Oct4 and Sox2 in ESCs to activate a particular set of target genes , and has been shown to function upstream of Nanog in ESC self-renewal and in preventing differentiation . Meanwhile, transcription factor box 3 (Tbx3) plays an essential role in the maintenance of pluripotency by blocking differentiation into epiblast-derived lineages . Similar to Klf4, Tbx3 is a fast responding mediator of LIF signaling that sustains the core pluripotency network through the direct activation of Nanog . The orphan nuclear receptor Dax-1, which has been shown to be directly regulated by Nanog, Oct4, and Stat3 [31, 45], functions as part of the extended core gene regulatory network that governs ESC fate where it interacts with other essential transcription factors to regulate a common set of genes involved in self-renewal and pluripotency [17, 45, 46]. Dax-1 has also demonstrated independent repressor function at differentiation genes . Due to the requirement of these critical factors for the maintenance of ESC self-renewal and pluripotency, of the genes identified as being downregulated following p/CIP depletion, Klf4, Tbx3, and Dax1 were selected for further analysis to provide insight into p/CIP function in ESCs. To validate the results of the microarray, qRT-PCR analysis was performed to assess changes in mRNA expression levels of Klf4, Tbx3, and Dax1, following p/CIP knockdown. In accordance with the expression data, the expression levels of each of these genes in p/CIP knockdown cells were significantly downregulated by approximately 50%, 60%, and 65%, respectively, compared to control cells (Fig. 4A).
In order to determine whether p/CIP regulates these genes directly, ChIP experiments were conducted using an antibody specific to p/CIP. qPCR analysis of immunoprecipitated DNA using specific primers flanking the promoter regions of selected pluripotency genes within 1 kb upstream of their transcriptional start sites (TSS) demonstrates that p/CIP binds to each of these gene promoters (Fig. 4B). Together, these results suggest that p/CIP directly regulates transcription of Klf4, Tbx3, and Dax-1 in a positive manner. Given the requirement of these genes for the maintenance of pluripotency in mESCs, p/CIP may contribute to the maintenance of pluripotency through the direct activation of these essential transcription factors that function cooperatively with the core pluripotency network.
To further assess the effects of p/CIP on ESC pluripotency and self-renewal, we generated a PB transposon vector containing reporter-linked, Flag-tagged p/CIP overexpression vector under the transcriptional control of the tetO2 doxycycline-inducible promoter (Fig. 5A), using the piggyBac (PB) transposon system [34, 47] which was stably integrated into mouse ESCs. Time course Western blot analysis demonstrated induction of Flag-p/CIP expression within 24 hours, with maximal expression at 48 hours (Fig. 5B). Two clones expressing the transgene most highly, based on β-galactosidase expression and Flag-p/CIP protein levels following 48-hour Dox-stimulation (Fig. 5C, 5D), were chosen for further analysis and used in subsequent experiments.
Since we identified p/CIP as a positive regulator of Tbx3, Klf4, and Dax1, based on knockdown and ChIP experiments, we examined the effects of p/CIP overexpression on the expression levels of these genes. qRT-PCR analysis showed a general increase in the transcript levels of these pluripotency genes in cells treated with doxycycline for 48 hours where p/CIP was overexpressed compared to untreated control cells, with the greatest effect being observed for Tbx3 (25%–50%) (Fig. 5E). Wild-type E14 cells were used as a control to show that the effects were due to p/CIP overexpression and not to doxycycline treatment alone. Furthermore, the level of transcriptional upregulation of these genes is proportional to the amount of p/CIP overexpression since the F2 clone overexpressing p/CIP more highly (approximately threefold) displayed greater increases in transcript levels of these genes compared to the clone overexpressing p/CIP by only 1.7-fold (B10).
Overexpression of p/CIP Enhances the Formation of Undifferentiated ESC Colonies
Since p/CIP overexpression in ESCs led to increased expression levels of genes essential for maintaining pluripotency, we wanted to determine whether overexpression of p/CIP would heighten ESC self-renewal. To do so, a clonogenic assay [34, 35] was used in which two individual Dox-inducible p/CIP-overexpressing clones were plated in a single-cell suspension at an extremely low density in ESC media containing LIF, in either the presence or absence of doxycycline. Wild-type E14 cells were used as a control. After 7 days, colonies were stained for ALP, a marker of undifferentiated ESCs. Individual colonies were strictly categorized as undifferentiated, partially differentiated, or fully differentiated based on ALP expression and morphology, and the distribution of colonies on each plate was determined (Fig. 6). While untreated clones displayed a similar distribution to the wild-type E14 cells with approximately 45% of the total colonial population being maintained in an undifferentiated state, clones treated with Dox, and thus overexpressing p/CIP, appeared to form undifferentiated colonies with greater efficiency (55%–65%). Once again, a greater effect was observed for the clone demonstrating higher levels of p/CIP overexpression. These results suggest that ESCs expressing elevated levels of p/CIP are more resistant to differentiation signals than wild-type ESCs.
ESC pluripotency is maintained, in part, by a complex gene regulatory network that promotes the activation of genes required for self-renewal and pluripotency, and the repression of developmental and differentiation genes. This network involves a wide variety of transcription factors, coregulators, and chromatin modifiers. In this study, we examined the role of p/CIP, a member of the SRC family of transcriptional coactivators, in mouse ESCs. We have shown that p/CIP is downregulated during differentiation, induced by RA-treatment or LIF withdrawal, and have identified genes directly regulated by p/CIP in mouse ESCs that are critical to the maintenance of the undifferentiated state. These results, and those of others [29, 30], suggest that p/CIP is required for the maintenance of self-renewal and pluripotency in ESCs.
Consistent with the finding that p/CIP is a downstream target of Nanog in mouse ESCs, the decrease in p/CIP expression (∼50%) observed during RA-induced differentiation was preceded by the rapid downregulation of the core pluripotency factors (Oct4, Sox2, and Nanog). Interestingly, a more drastic downregulation of p/CIP expression (∼80%) was observed in response to LIF withdrawal. This result suggests that p/CIP may function in LIF-mediated signaling events that support self-renewal and pluripotency.
A loss of function approach using siRNAs to target the p/CIP transcript indicated that p/CIP knockdown results in a significant downregulation of several genes encoding transcription factors that are required for maintaining ESC self-renewal and pluripotency including Klf4, Tbx3, Dax1, and to a lesser extent Nanog [11, 40, 41, 44]. This finding could account for the previous observation by Hu et al.  that downregulation of p/CIP in mouse ESCs induced differentiation. The Oct4GiP assay used in their RNAi screen measured differentiation based on decreased Oct4-driven green fluorescent protein expression. However, significant downregulation of Oct4 or Sox2 expression following p/CIP knockdown was not observed in our study. While differences in experimental conditions may be attributable, this discrepancy is likely due to the fact that in this study changes in gene expression were assessed 48 hours post-transfection, whereas Hu et al.  performed their analysis 96 hours after p/CIP knockdown suggesting that their effects may be a consequence of secondary or tertiary events occurring following p/CIP knockdown.
The microarray expression analysis of mESCs following p/CIP knockdown identified genes encoding essential regulators of pluripotency thereby providing some insight into the role p/CIP plays in determining ESC fate. However, gene expression differences following p/CIP downregulation were surprisingly low suggesting that p/CIP may play a more prominent role during differentiation. This may be explained, in part, by the incomplete silencing of p/CIP (∼75%) following siRNA transfection. Any residual p/CIP in the cells may still be functioning at target genes to mediate transcriptional effects. Another possibility is that members of the SRC family of coactivators (SRC1, SRC2, and p/CIP) may exhibit some level of functional redundancy in mouse ESCs. Although SRC2 has not been implicated in ESCs, SRC1 was identified as a gene commonly downregulated by both RA-treatment and LIF withdrawal in mouse ESCs . Consequently, a more dramatic effect on genome-wide expression levels would be expected in ESCs where both SRC1 and p/CIP are knocked down based on the observation that the double knockout mice lacking SRC1 and p/CIP display a much more severe phenotype than those lacking either SRC member . ESC pluripotency is maintained by a very complex, yet self-sustaining core gene regulatory network. In order to maintain their viability, ESCs are likely equipped with compensatory mechanisms to deal with the loss of a single factor with a highly homologous family member capable of performing a similar function. This type of redundancy in ESCs has been observed for the Klf and Sall families of transcription factors [44, 49]. Due to the high degree of interconnectedness demonstrated by regulators of self-renewal and pluripotency, it has also been suggested that loss of function of one regulator can be compensated by adjusting the expression levels of other components of the transcriptional network governing ESC identity .
Klf4, Tbx3, and Dax1 have been implicated in ESC self-renewal and pluripotency as part of the extended core gene regulatory network that governs ESC identity. These essential factors have been shown to function upstream, downstream, or even in cooperation with the core pluripotency factors to maintain the self-renewal and pluripotency of ESCs. Therefore, of the 46 downregulated genes from the microarray study, the genes encoding these factors were the most desirable candidates for subsequent validation by qRT-PCR and ChIP analysis. The observation that p/CIP occupies the promoter regions of Klf4, Tbx3, and Dax1 within 1 kb of their TSS, analyzed in conjunction with real-time PCR data, suggests that these pluripotency-associated genes are direct regulatory targets of p/CIP-mediated transcriptional regulation in mouse ESCs. This notion is further supported by the finding that overexpression of p/CIP in mESCs results in the upregulation of these genes in a dose-dependent manner. Consequently, this upregulation of essential self-renewal genes that sustain the core pluripotency network and repress differentiation may account for the ability of p/CIP-overexpressing clones to form undifferentiated ESC colonies with greater efficiency than those in which p/CIP is not overexpressed, as observed in the clonogenic assay.
These findings suggest that p/CIP integrates with the extended core gene regulatory network to maintain ESC self-renewal and pluripotency. Thus, we integrated our data with published datasets (as described in Materials and Methods) to draft a p/CIP network in mouse ESCs. Indeed, gene regulatory network analysis confirms our findings that p/CIP promotes mESC self-renewal and pluripotency by regulating the extended pluripotency network (Fig. 7A). For example, one regulatory module within the larger p/CIP network is a feed-forward module (Fig. 7B) promoting expression of Nanog (part of the pluripotency core module), Klf4, and Tbx3 (part of the extended pluripotency network). Klf4 is known to activate its own expression as well as Nanog, Tbx3, and Oct4 while Nanog activates it own expression as well as Oct4, Sox2, and Tbx3. Upon ESC differentiation, p/CIP expression decreases (Fig. 1), collapsing this feed-forward loop. Hence, the 1.25-fold decrease in Nanog expression observed in response to p/CIP knockdown appears to be an indirect effect resulting from the more robust downregulation of Klf4 and Tbx3. Since p/CIP has been previously identified as a target gene of Nanog , Nanog may subsequently activate p/CIP expression in mESCs thereby forming a positive feedback loop. Moreover, p/CIP has been well studied for its role as a coactivator of nuclear receptors. For example, Esrrb is a nuclear receptor that has been identified as a critical regulator of ESC pluripotency that cooperates with the core pluripotency network to promote pluripotency gene expression. Interestingly, Esrrb has been shown to directly regulate Tbx3 expression  as well as directly associate with p/CIP in pluripotent mouse ESCs . While this manuscript was under revision, two overlapping studies were published demonstrating that SRC3/NCoA3 contributes to pluripotency maintenance in mESCs. p/CIP/SRC3 is required for the expression of several critical pluripotency genes including Nanog, KlF4, Dax1, and Tbx1 [52, 53] and is essential for controlling Esrrb-dependant activation suggesting that SRC3/NCoA3 is a key member of the pluripotency transcriptional circuitry.
Global localization studies have uncovered complex transcriptional networks wherein critical transcription factors have been found to coexist in large complexes with various members of the core pluripotency network (up to 13 factors) for the regulation of common target genes [14, 17]. p/CIP is a transcriptional coactivator that functions as an adapter protein that localizes to target genes upon interaction with a wide variety of transcription factors and promotes the assembly of coactivator complexes by recruiting various chromatin regulators and modifying enzymes that alter the surrounding chromatin structure to facilitate transcriptional activation . Both p300/CBP and CARM1 are chromatin modifying enzymes that directly interact with p/CIP through its AD1 and AD2 domains, respectively, resulting in the assembly of such a complex. While p300 has been shown to colocalize with core pluripotency factors as part of these large protein complexes at common target genes in mouse ESCs, CARM1 has been shown to be essential for the maintenance of pluripotency as a regulator of histone H3 arginine methylation at pluripotency gene promoters . Thus, it is likely that p/CIP functions in the assembly of multisubunit regulator complexes that contain chromatin regulators such as p300 and CARM1, as well as other important pluripotency transcription factors, to activate pluripotency gene expression.
The transcriptional coactivator p/CIP is highly expressed in undifferentiated mESCs and undergoes downregulation upon differentiation. Microarray expression in conjunction with ChIP analysis identified Tbx3, Klf4 and Dax-1 as direct transcriptional targets of p/CIP. Stable overexpression of p/CIP in mESCs promoted the formation of undifferentiated colonies suggesting that p/CIP levels are important for the maintenance of pleuripotency. p/CIP is a critical component of an extended pluripotency network and contributes to the maintenance of ESC pleuripotency through both direct and indirect regulation of essential pluripotency genes. These results may have implication in many cancers where p/CIP is amplified and overexpressed.
This work was generously supported by the following organizations and funding agencies: a CCS research grant (015428) to J.T., an Ontario graduate scholarship to J.C., and a Canadian Institutes of Health Research (CIHR) doctoral fellowship to G.T. The Canadian Institutes of Health Research (CIHR) to W.L.S. for an operating grant (MOP-89910) and a CIHR Banting and Best CGS Doctoral Research Award to J.L.M.; the Heart & Stroke Foundation of Canada for a Postdoctoral Fellowship to W.Y.C.; and the Canada Research Chair program that provides support to W.L.S. through a Tier 1 Chair in Integrative Stem Cell Biology. W.Y.C. is currently affiliated with the Stem Cell Technologies, Inc., Vancouver, BC, Canada.
J.C.: collection and assembly of data, manuscript writing, and conception and design; G.T., J.L.M., W.Y.C., E.W., and M.I.: Collection and assembly of data; W.L.S. and J.T.: manuscript writing, conception and design, and financial support.
Disclosure of Potential Conflicts of Interest
The authors have no financial conflict of interests to declare.