Overexpression of Trophoblast Stem Cell-Enriched ... - Cell Press

Download PDF
19 downloads 0 Views 3MB Size Report
Comment
May 9, 2017 - embryonic stem cells: Sall1, Sall4, Ccnd1, Ccdn2, and Lin28. (Figure 1B). miRNA expression data were confirmed by qPCR. (Figure 1C) and ...
Report

Overexpression of Trophoblast Stem Cell-Enriched MicroRNAs Promotes Trophoblast Fate in Embryonic Stem Cells Graphical Abstract

Authors Ursula Nosi, Fredrik Lanner, Tsu Huang, Brian Cox

Correspondence [email protected]

In Brief Nosi et al. identify microRNAs sufficient to drive embryonic to extraembryonic lineage conversion by employing stem cell models. Trophoblast-enriched microRNAs downregulate pluripotencyassociated genes in ESCs and drive the acquisition of a preimplantation mural trophectoderm phenotype. This work suggests the involvement of microRNAs in networks regulating preimplantation development.

Highlights d

Data mining strategy accurately identifies trophoblast fatespecifying miRNAs

d

ESCs with ectopic expression of miR-15b trans-differentiate into the trophoblast lineage

d

Genome-wide expression analysis of iTCs reveals a mural TE signature

d

iTCs contribute to the TE and Reichert’s membrane in vivo

Nosi et al., 2017, Cell Reports 19, 1101–1109 May 9, 2017 ª 2017 The Author(s). http://dx.doi.org/10.1016/j.celrep.2017.04.040

Accession Numbers GSE79599 GSE15519 GSE29102 GSE15519 GSE25255

Cell Reports

Report Overexpression of Trophoblast Stem Cell-Enriched MicroRNAs Promotes Trophoblast Fate in Embryonic Stem Cells Ursula Nosi,1 Fredrik Lanner,2,3 Tsu Huang,1 and Brian Cox1,4,5,* 1Department

of Physiology, University of Toronto, Toronto, ON M5S 1A8, Canada of Clinical Science, Intervention and Technology, Karolinska Institute, 171 77 Solna, Sweden 3Division of Obstetrics and Gynecology, Karolinska Universitetssjukhuset, 141 86 Stockholm, Sweden 4Department of Obstetrics and Gynecology, University of Toronto, Toronto, ON M5S 1A8, Canada 5Lead Contact *Correspondence: [email protected] http://dx.doi.org/10.1016/j.celrep.2017.04.040 2Department

SUMMARY

The first cell fate choice of the preimplantation embryo generates the extraembryonic trophoblast and embryonic epiblast lineages. Embryonic stem cells (ESCs) and trophoblast stem cells (TSCs) can be utilized to investigate molecular mechanisms of this first cell fate decision. It has been established that ESCs can be induced to acquire trophoblast lineage characteristics upon manipulation of lineage-determining transcription factors. Here, we have interrogated the potential of microRNAs (miRNAs) to drive transdifferentiation of ESCs into the trophoblast lineage. Analysis of gene expression data identified a network of TSC-enriched miRNAs that were predicted to target mRNAs enriched in ESCs. Ectopic expression of these miRNAs in ESCs resulted in a stable trophoblast phenotype, supported by gene expression changes and in vivo contribution potential. This process is highly miRNA-specific and dependent on Hdac2 inhibition. Our experimental evidence suggests that these miRNAs promote a mural trophectoderm (TE)-like cell fate with physiological properties that differentiate them from the polar TE. INTRODUCTION During preimplantation development, totipotent cells of the embryo are specified into the spatially segregated, distinct cell lineages of the blastocyst. These comprise the outer epithelial trophectoderm (TE) monolayer and the inner cell mass (ICM), which will establish the trophoblast and epiblast lineages, respectively. The trophoblast lineage gives rise to progeny that mediate implantation and generate the functional placenta, while the epiblast lineage generates the somatic and germ cell types of the embryo proper (Rossant and Tam, 2009). The window of in vivo lineage plasticity is limited to development prior to implantation, after which trophoblast and epiblast cells become restricted.

Trophoblast stem cells (TSCs) and embryonic stem cells (ESCs) can be derived from early mouse embryos and when transplanted back in to the blastocyst contribute to development in a lineage-restricted manner, indicating they are molecularly and phenotypically similar to their in vivo counterparts, trophoblast, and epiblast, respectively (Nagy et al., 1993; Ralston et al., 2008; Rugg-Gunn et al., 2012; Tanaka et al., 1998). In vitro, ESCs can be genetically manipulated into trophoblast fates using single factors. The first example of this approach revealed that expression of TE-specific transcription factor Cdx2 or downregulation of the pluripotency factor Oct4 (Pou5f1) in ESCs induces a trophoblast fate. Transplantation of these induced trophoblast cells into blastocysts revealed contribution exclusively to the trophoblast lineage in vivo, indicating that they were blocked from accessing epiblast cell fates (Niwa et al., 2005). Since then, we have shown that ectopic expression of the TSC-enriched transcription factor Gata3 in ESCs also induces trophoblast fate (Ralston et al., 2008). These studies highlight the power of using inter-conversion of stem cells to identify and characterize novel regulators of the first cell fate specification and development. The role of microRNAs (miRNAs) in lineage specification and maintenance of the trophoblast remains unresolved, as analyses have yielded conflicting results (Suh et al., 2010; Viswanathan et al., 2009). miRNAs are 19- to 23-nt RNA molecules that function as post-transcriptional regulators of gene expression (reviewed in Winter et al., 2009). The early embryo expresses many lineagespecific miRNAs, and their expression is conserved in the stem cell derivatives used to model the embryo in vitro (Spruce et al., 2010; Viswanathan et al., 2009). Surprisingly, studies investigating miRNA function in the preimplantation embryo (Suh et al., 2010) indicate that although they can be detected (Viswanathan et al., 2009), miRNA activity is suppressed during the late preimplantation stage, indicating that miRNAs may not be regulators of cell fate specification in the blastocyst. Despite this, miRNAs and small synthetic mimetic RNAs have been shown to guide differentiation into key lineages and greatly improve reprogramming to induced pluripotent stem cells (iPSCs) (Anokye-Danso et al., 2011). Here, we use an ESC trans-differentiation model to investigate whether miRNAs can induce extraembryonic trophoblast fates.

Cell Reports 19, 1101–1109, May 9, 2017 ª 2017 The Author(s). 1101 This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

RESULTS Bioinformatic Identification of miRNA-mRNA Networks We reasoned that miRNAs with a capacity to specify or determine trophoblast cell fate would form a network of TSC-enriched miRNAs that target ESC-enriched mRNA. Furthermore, we postulated that ectopic expression of these miRNAs in ESCs would transform them into trophoblast cells. miRNA-mRNA networks were identified from biological quadruplicate gene expression data of TSCs and ESCs representing 300 miRNAs and >20,000 transcripts (Figure 1A) and the miRanda database (Betel et al., 2010). Network analysis of the linked miRNAs and mRNAs identified a highly interconnected set of three miRNAs (miR-15b, miR-322, and miR-467 g) and 56 ESC-enriched mRNA targets (Figure 1B). The network contained TSC-enriched miR-15b, miR-322, and miR-467 g that were previously identified in deep-sequencing datasets on placental development in mouse and human placentas and mouse blastocysts (Gu et al., 2013; Viswanathan et al., 2009). Among the predicted 56 ESC-mRNA targets, 5 have been described in literature as being essential to ESC self-renewal and conserved in human embryonic stem cells: Sall1, Sall4, Ccnd1, Ccdn2, and Lin28 (Figure 1B). miRNA expression data were confirmed by qPCR (Figure 1C) and NanoString (data not shown), and we found miRNA-322 to be highly enriched to TSCs relative to ESCs, while miR-15b and miR-467 g were also enriched, but to a lesser degree. Of note, due to rapid biogenesis and turnover, quantification of miRNAs remains a challenge in the field. Using a similar strategy, we generated the opposite network of enriched ESC miRNAs that could target enriched TSC mRNAs (Figure S1A, related to Figure 1). Significantly, we observed the miR-302/367 cluster as centered in this network (Figure S1, related to Figure 1). Expression of these miRNAs was previously reported to generate iPSCs from mouse embryonic fibroblasts (MEFs) (Anokye-Danso et al., 2011). This observation was supportive of our data-mining strategy and encouraged further investigation into the function of the trophoblast miRNAs. Overexpression of TSC miRNAs in ESCs Induces a Trophoblast-like Phenotype To test the concept that TSC-enriched miRNAs could suppress embryonic fate and consequently drive trophoblast fate, we co-transfected mouse R1 ESCs with vectors constitutively expressing a GFP-puromycin resistance gene linked to miR-15b, miR-322 and miR-467 g transgenes. Cells were then cultured in standard TSC media supplemented with puromycin and valproic acid (VPA). VPA’s function as a histone deacetylase 2 (Hdac2) inhibitor has been proposed to be required for miRNA function in ectopic expression experiments (Anokye-Danso et al., 2011). Remarkably, emergent colonies were morphologically similar to TSCs (Figures 1D–1F), and immunocytochemical staining revealed expression of the TSC markers Cdx2 and Gata3 and loss of the ESC marker Pou5f1 (Figure 1G). We refer to these cells as induced trophoblast cells (iTCs). As a stringent test, we used florescence-activated cell sorting (FACS) against Pecam1 and Cdcp1, previously published markers of TSCs (Cdcp1+/Pecam1) and ESCs (Cdcp1/ Pecam1+) (Rugg-Gunn et al., 2012), followed by single-cell

1102 Cell Reports 19, 1101–1109, May 9, 2017

qPCR gene expression analysis. FACS revealed that the level of Cdcp1 expression was similar in both TSCs and iTCs (Figure S1B, related to Figure 1). Analysis of single-cell gene expression data of TSCs, iTCs, ESCs, and mouse embryonic feeder cells (MEFs) showed a clear co-clustering of iTCs and TSCs and separation from ESCs or MEFs (Figure 1H) and was driven by genes known to mark the trophoblast (Figure 1I). Together, analysis of cell morphology, immunofluorescence, cell surface markers, and single-cell mRNA expression all demonstrate that ectopic expression of TSC-enriched miRNAs in mouse ESCs leads to cell lineage conversions toward trophoblast. To investigate the trophoblast induction process in more detail, we generated two or three independent transgenic ESC lines with doxycycline-inducible expression of miR-15b, miR-322, or miR-467 g linked to a GFP reporter (Figure S2A, related to Figure 2). These cells showed uniform induction of GFP expression (Figures S2B and S2C, related to Figure 2) that corresponded to increased expression of the miRNA (Figure S2D, related to Figure 2). Induced overexpression of miR-15b, miR-322, or miR-467 g under standard TSC culture conditions supplemented with VPA could independently induce morphological changes after 2–3 days. After 5 days, colonies acquired an epithelial cobblestone-like TSC morphology (Figures 2Ai–2iv) that was maintained over a 9-day miRNA induction period (Figure 2Aiv). As anticipated, non-induced cell lines retained their ESC morphology in both ESC and TSC culture conditions (Figures 2Ai0 –2iv0 ). These results indicate that miR-15b, miR-322, and miR-467 g have independent abilities to induce trophoblast-like morphological changes in ESCs. Next, we utilized these inducible ESC lines to conduct timecourse analyses of mRNA expression in response to miRNA induction. We observed a reduction of the predicted target genes Sall1 and Sall4 mRNA as early as 4 hr post-induction and a trend for lower mRNA expression of Pou5f1 and Nanog (Figure 2B). As miRNAs primarily cause gene expression alterations post-transcriptionally, we also assessed protein levels of Sall4 and Pou5f1 by western blot and noted a stronger decrease in the protein levels as compared to mRNA (Figures 2C and D). Significantly, analysis of multiple miRNA-induced cell lines for the TSC marker genes Cdx2 and Elf5 showed a steady upregulation of mRNA throughout the induction timeline (Figure 2E). Control experiments identified miRNA induction, VPA, and MEFs as essential to the generation of iTCs (Figures S2E– S2N, related to Figure 2). Critically, over expression of cardiac-specific miR-1 in ESCs (Izarra et al., 2014) under trophoblast culture conditions did not generate iTCs (Figures 2F–2H) and led to expression of the cardiac markers Gata4 and Nkx25 (Figure 2H). These results indicate a dependence on specific miRNA expression and HDAC2 to both positively and negatively regulate gene expression. miRNA-iTCs Are Stable We determined that 5–6 days of miRNA overexpression were required to induce trophoblast cells with a stable morphology (Figures 3A–3C) and gene expression (Figure 3D), which could be passaged. Four to six days after passaging, emergent colonies maintained an epithelial morphology (Figure 3E)

Figure 1. ESCs and TSCs Have Reciprocal miRNA and mRNA Networks (A) Venn diagram of enriched TSC miRNAs and ESC mRNAs. (B) Highly connected network of three TSC-enriched miRNAs targeting 54 ESC-enriched mRNAs. Known pluripotency genes have green outer ring, with conserved expression in humans in red. (C) qRT-PCR TaqMan assay analysis of miR-15b, miR-322, or miR-467 g in ESCs and TSCs; sno142 and 202 were used as reference genes. SE is calculated between technical triplicates. (D) Bright-field image of ESCs 5 days post-transfection with vectors constitutively expressing the three candidate miRNAs. (E) Transfected cells can be passaged and maintain trophoblast-like morphology. (F) Control culture of TSCs. (G) Immunocytochemical analysis of transfected ESC colonies for the TSC markers Cdx2 and Gata3 and the ESC marker Pou5f1. (H) Principle-component analysis of single-cell gene expression using Made4 (R library) revealed that iTCs cluster closely with blastocyst-derived TSCs and away from both ESCs and MEFs. (I) Loadings plot of (H) highlighting genes driving sample clusters. See also Figure S1.

and expressed trophoblast markers Cdx2 and Elf5 at levels similar to embryo-derived TSCs (Figure 3F). Curiously, we observed that expression of estrogen-related receptor beta (Essrb) and Sry-box 2 (Sox2) genes was lower than in TSCs (Fig-

ure 3G). These genes are expressed and essential to both ESCs and TSCs. To determine if iTCs could also generate trophoblast giant cells (TGCs), we induced miRNA expression for 6 days and

Cell Reports 19, 1101–1109, May 9, 2017 1103

then cultured for 6 more days in the absence of Fgf4, heparin, and MEFs and compared iTCs to similarly treated TSCs. After the differentiation period, iTC colony morphology appeared unchanged, and flow cytometric analysis of DNA content indicated the cells were still diploid (Figures S3A and S3B, related to Figure 3). Surprisingly, this suggests that our iTCs cannot differentiate under standard TSC differentiation-promoting conditions. Global Expression Profiles of iTCs Reveal Properties of Preimplantation TE To test the extent of trophoblast cell fate induction on iTCs, we assessed the global gene expression of two cell lines for each tested miRNA using microarrays. As controls, we compared iTCs with TSCs, ESCs, and ESCs induced with Cdx2 activity (Cdx2-iTCs) (Niwa et al., 2005). A heatmap of unsupervised clustering of the Pearson correlation coefficient between the arrays revealed a strong difference between the TSC replicates and ESCs (Figure 4A). The miR-iTCs formed a distinct group, but with significant correlation to TSC gene expression (Figure 4A). Cdx2-iTCs co-clustered with TSCs but have a cross-correlation to miRNA lines (Figure 4A). Using linear models microarray analyses (Ritchie et al., 2015), we made a three-way contrast model of TSCs, miRiTCs, and Cdx2-iTCs relative to ESCs. A Venn diagram of increased differential gene expression relative to ESCs revealed a strong overlap of 561 genes in TSCs, miR-iTCs, and the Cdx2-iTCs and 1,034 genes enriched to only miR-iTCs and TSC (Figure 4B). The intersection of upregulated genes contained numerous known trophoblast genes (Table S1). ESC-associated pluripotency genes were downregulated, as were markers of mesoderm, neurectoderm, and primitive and definitive endoderm. To better characterize the cell’s lineage and developmental potential, we applied gene ontology overrepresentation analysis to the common upregulated genes and those in only TSCs and iTCs. We found that both sets of upregulated genes had significant enrichment of ontologies of Anatomy, Gene and Mutant Phenotype related to trophoblast and placenta (Figure 4C; Table S2). Strikingly, a greater enrichment of genes associated

Figure 2. Activity of Induced Expression of TSC-Enriched miRNAs in ESCs (A) Bright-field images of cell lines induced to express miR-15b, miR-322, or miR-467 g for 5 days in the presence of valproic acid (i–iii). Scale bar, 20 pixels. Bright-field image of induced cells after 9 days of induction culture (iv). Brightfield images of the same cell lines cultured in TSC media without the miRNA transgene-inducing agent (doxycycline) (i0 –iii0 ). Non-induced cells cultured for 6 days in ESC media (iv0 ). (B) qPCR analysis of miRNA predicted targets Sall1 and Sall4 and pluripotency genes Pou5f1 and Nanog. Shown are the average expression of three miR-467 g lines, two miR-15b lines, and two miR-322 lines. All lines show a downward trend beginning as early as 4 hr.

1104 Cell Reports 19, 1101–1109, May 9, 2017

(C) Western blot of Sall4 and Oct4 protein expression in induced and noninduced cell lines compared to TSCs. Blot was cropped for clarity and white space. (D) Quantification of western blot using ImageJ densitometry plug-in and actin as reference. (E) qPCR assessment of mRNA expression of TSC marker genes Elf5 and Cdx2 in the same miRNA lines and biological replicates. Observed is a steady increase of these genes as early as 24 hr, throughout a 10-day induction period. (F) Bright-field image of ESCs 3 days post-induction of miR-1 expression. (G) Bright-field image of ESCs 6 days post induction of miR-1. Scale bar, 10 pixels. (H) qPCR gene expression analysis comparing miR-15b, miR-1, and TSCs for TSC markers (Cdx2, Elf5, and Eomes), pluripotency markers (Pou5f1, Nanog, and Lin28), and cardiac lineage markers (Gata4 and Nkx2-5). qPCR data were normalized against Gapdh/Actb as reference and graphed as log2 normalized expression relative to ESCs. SE is calculated between two technical replicates of each biological replicate. See also Figure S2.

Figure 3. iTCs Are Stable and Propagate after Withdrawal of miRNA Induction (A) A schematic of the experiment. (B) Bright-field image of ESCs after 3-day miRNA expression and cultured until 9 days. (C) Bright-field image of ESCs after 6-day miRNA expression and cultured for 9 days. Scale bar, 50 pixels. (D) qPCR gene expression analysis of 3-day and 6-day induced ESC cell lines (NI is no induction of miR15b). (E) Bright-field images of 6-day induced lines serially passaged without transgene induction. (F) Gene expression of wild-type TSCs and passaged iTC lines corresponding to (E). (G) qPCR of ESC lines induced for 6 days and serially passaged. SE is calculated between technical replicates. See also Figure S3.

with the Anatomy Ontology terms ‘‘trophectoderm,’’ ‘‘polar trophectoderm,’’ and ‘‘mural trophectoderm’’ were observed to our iTCs (Figure 4C; Table S2). Genes annotated to mural TE cell fates had higher expression in the miR-iTCs than in TSCs and Cdx2-iTCs. The Na/K pumps Atp1a3, Atp1b2, and Atp1b3 all had higher expression in miRNA lines (1.2- to 2-fold) relative to TSCs. Additionally, Prkcz and Cebpa also showed higher levels of expression relative to TSCs (1.5-fold, each). In keeping with our qPCR analysis, the ExE-enriched transcription factors Esrrb and Sox2 were 8- and 22-fold lower in miR-iTCs, respectively. As the mural TE gives rise to primary TGCs, we assessed markers of this cell type and observed that the expression of Wnt5a, Ext1, Uchl1, Limk1, Gata2, Ctsd, and Ascl2 was significantly higher (1.4- to 6-fold) in miR-iTCs than in TSCs and Cdx2-iTCs. Higher levels of TE and primary TGC genes, together with lower levels of ExE genes, indicate that induction of our candidate miRNAs has a greater propensity to generate cells similar to the TE, while Cdx2-induced cells and embryo-derived TSCs are more similar to the ExE.

miR-iTCs Contribute Predominantly to the Mural TE In Vivo To resolve the cell identity of our iTCs, we tested their in vivo developmental potential by transplantation into blastocyst- and morula-stage embryos. We injected 10–15 miR-15b 6-day induced cells expressing mCherry into embryonic day 3.5 (E3.5) mouse blastocysts (Table S3, related to Figure 5). Remarkably, mCherry expression was detected exclusively within the Reichert’s membrane portion of E6.5 embryos, a tissue composed of primary TGCs derived from the mural TE (Figure 5A) (Rossant and Cross, 2001). It has been noted in the past that TSCs can contribute to the Reichert’s membrane, albeit at low frequencies (Tanaka et al., 1998). Control embryos were assessed under the red channel, and no epifluorescence was detected, indicating that the transplanted cells were responsible for the observed signal (Figure 5A). To assess preimplantation potential, four or five cells from 6 days of miR-15b induction were aggregated with non-compacted morulas (six to eight blastomeres) by injection under the zona pellucida and cultured in vitro until the blastocyst stage.

Cell Reports 19, 1101–1109, May 9, 2017 1105

Figure 4. Global Gene Expression Identifies a Significant Relationship between Induced and Embryo-Derived Trophoblast Cells (A) A heatmap of Pearson correlation coefficients of microarray data for ESCs, TSCs, and iTCs (miRNA and Cdx2). A high degree of cross-correlation between miRNA induced trophoblast and embryo trophoblast is observed. (B) Venn diagram of a three-way comparison of genes exhibiting increased expression relative to ESCs. (C) A bi-graph network diagram of the relationship of enriched gene sets related to trophoblast development and function spanning Gene Ontology (blue), Mutant Phenotypes (green), and Anatomy Ontologies (orange) to either the co-expressed genes of miR-iTCs and TSCs or miR-iTC, Cdx2-iTCs, and TSCs (purple nodes). See also Figure S4 and Tables S1 and S2.

From 11 embryos imaged using confocal microscopy, we observed 9 with multiple florescent cells in the outer TE layer and none in the inner cell mass (Figure 5B). This data showed that the miR-iTCs readily incorporate into the TE and have lost epiblast lineage potential. Lastly, we employed a cell physiological assay to provide further experimental evidence of the mural TE nature of our cell line. Rassoulzadegan et al. (2000) reported that the mural TE has measureable phagocytic properties using 1–3 mm fluorescent microspheres, which we could reproduce (Figure S5A). We observed that iTCs had a high degree of uptake of 1.75 mm fluorescently labeled microspheres (Figure 5C). In contrast, TSCs showed rare cells with microsphere internalization, while TGCs, a known phagocytic cell type, also readily uptake microspheres (Figure 5C). DISCUSSION The bioinformatics data-mining strategy developed and employed in this study readily identified interaction networks of reciprocally enriched miRNAs and mRNAs. The TSC-enriched miRNAs were shown to function to repress embryonic and promote trophoblast cell fates. Significantly, the inverse model redis-

1106 Cell Reports 19, 1101–1109, May 9, 2017

covered the ESC-enriched miRNA miR-302/367 cluster previously shown to induce pluripotency (Anokye-Danso et al., 2011). The unbiased approach of our network model is in contrast to previous selections of miRNA based on targets. Overexpression of miR-145 in ESCs, a miRNA that targets pluripotency genes Oct3/4, Sox2, and Klf4, induces mesoderm and ectoderm lineages, but not trophoblast (Xu et al., 2009). Our validation of miR-1-induced cardiac fate (Izarra et al., 2014) coupled with published data on miR-145 supports our hypothesis that specific networks of miRNAs and mRNA regulated cell fate in trophoblasts. Furthermore our observations strengthen findings by others that trophoblast is not a default cell fate during repression of pluripotency, as loss of Esrrb, Sall4, Nanog, or Tcfap4 alone results in continued epiblast lineage development (Festuccia et al., 2012; Nishiyama et al., 2013; Tan et al., 2013). Evidence from our current study and those of others indicates that ESCs are open to trophoblast fates, but inducing this developmental path is dependent on the inducing agents used, as other cellular fates are also available. Our bioinformatic prediction of Sall1 and Sall4 as targets is supported by the observed decrease in Sall4 mRNA and protein expression upon miRNA expression. The role of Sall4 in pluripotency is confused by conflicting reports (Zhang et al., 2006; Yuri et al., 2009). Sall4-null ESCs have been reported to selfrenew under feeder-free (Yuri et al., 2009) and 2i/LIF (leukemia inhibitory factor) culture conditions (Miller et al., 2016), which are optimal for the maintenance of pluripotency. Alternatively, Sall4-null ESCs grown in TSC media supplemented with Fgf4 are reported to be unstable and spontaneously differentiate into TSCs (Zhang et al., 2006). Similarly, we found that generation of iTCs from ESCs is dependent on extrinsic factors (fibroblast growth factor [FGF] and MEFs). Significantly, Sall4-null embryos fail to generate a stable inner cell mass, and ESCs

Figure 5. miRNA-iTCs Contribute to the Trophectoderm In Vivo and Have Physiological Properties of the Mural TE (A) Bright-field (BF) images of representative E6.5 embryos are presented on the left and fluorescence (RFP) images of corresponding embryos on the right. mCherry localization within pockets of the Reichert’s membrane was observed. Control embryos were assessed under the same microscope settings. Scale bar, 5 pixels. (B) Single optical sections of individual embryos cultured from E2.5 to E3.5, with green representing GFP-expressing iTCs, blue Hoechst-labeled nuclei, and merged channels. The approximate position of the inner cell mass is outlined by a white dashed line. Scale bar, 10 pixels. (C) Fluorescent confocal microscopy of miR-15b day-6-induced cells, TSCs, and trophoblast giant cells (TGC) exposed to 1.75 mm fluorescently labeled microspheres (red). Scale bar, 10 pixels. Cell nuclei are labeled with Hoechst (blue). See also Figure S5 and Table S3.

cannot be derived from these blastocysts (Elling et al., 2006; Zhang et al., 2006). Combining in vitro and in vivo data suggests that Sall4 is key to prevent access of the epiblast to the trophoblast lineage during cell fate change. As ESCs already have acquired epiblast fate, in the absence of trophoblast-inducing factors, the cells continue to self-renew. However, in vivo, the Sall4-null epiblast-fated cells are free to respond to trophoblast-promoting signals. The miRNA-mediated reduction in Sall4 makes ESCs permissive to trophoblast cell fates. Analysis of global gene expression in iTCs revealed a robustly expressed trophoblast gene signature, with little evidence of mesoderm, endoderm, or ectoderm being generated. Additionally, in contrast to Cdx2-iTCs, both iTCs and TSC showed enrichment of gene sets annotated to TE and primary TGC genes. This indicates that our miRNA-iTC lines may have a propensity to make preimplantation mural TE, which may explain their failure to differentiate upon FGF4 withdrawal (Figure S3) and lower Sox2 and Essrb expression (Figure 3G). Our transplantation observations support the gene expression data and indicated that iTCs are both lineage restricted to the trophoblast and the mural TE specifically. Interestingly, TSCs have also been observed to incorporate into the mural TE though at a lesser propensity (Tanaka et al., 1998). Through cell labeling studies others have shown a natural polar to mural TE flow of trophoblast cells, with the polar TE containing the TSC niche (Gardner, 2000). We observed iTCs only in the mural portion of the TE, which is likely their point of incorporation.

The Reichert’s membrane is an extraembryonic membrane derived from the mural TE portion of the blastocyst and composed mainly of primary TGCs (Tam and Rossant, 2003). The primary TGCs of the Reichert’s membrane and the mural TE transfer nutrients to the embryo before the maturation of the functional placenta. This functional property was observed in our iTCs, as they readily phagocytosed 1.75-mm florescent particles, contrary to embryoderived TSCs. Of note, the Reichert’s membrane also contains scattered parietal endoderm cells (Kunath et al., 2005), but our microarray data did not identify enriched expression of visceral or parietal endoderm markers in the iTCs. Surprisingly, studies investigating miRNA function in the preimplantation embryo (Suh et al., 2010) observed that although miRNAs can be detected, their activity may be suppressed during the late preimplantation stage. This suggests that miRNAs may not be regulators of cell fate specification in the blastocyst. In contrast, our observations of miRNAs that can function to transform ESCs into early preimplantation trophoblast could indicate that these miRNAs may in fact have a function in the early embryo. Specific examples of miRNA function in mouse blastocyst development have been shown by others (Joo et al., 2014; Zhang et al., 2015). Perturbations of miR29b expression in early embryos disrupted methyltransferase Dnmt3a/3b expression and morula to blastocyst development. Knockdowns of miR-34c in zygotes caused developmental arrest at the two- to four-cell stage, likely due to dysregulated Bcl-2 expression. Our results have potential translation to human health and development. Our cell lines may model the mural TE, which mediates the embryo implantation process (James et al., 2012) and thus may have utility for investigating the implantation process using in vitro cell assays. Moreover, poor development of the placenta is associated with pregnancy pathologies and future

Cell Reports 19, 1101–1109, May 9, 2017 1107

risk for chronic health conditions (Lewis et al., 2011). miR-15b and miR-322 (miR-424 in humans) exhibit conserved expression in human placentas and are downregulated in placentas affected by preeclampsia and growth restriction (Mayor-Lynn et al., 2011; Mouillet et al., 2013). The decreased expression of these TSC miRNAs in human pathologies with small placentas points to their potential involvement in cell fate by controlling the differentiation of trophoblasts or maintaining a pool of TSCs or progenitors. EXPERIMENTAL PROCEDURES All procedures involving animals were performed in compliance with the Animals for Research Act of Ontario and the Guidelines of the Canadian Council on Animal Care. The Centre for Phenogenomics (TCP) Animal Care Committee reviewed and approved all procedures conducted on animals at TCP (protocol number: 21-0306H). Bioinformatics Analysis for miRNA Identification In silico targets of candidate miRNAs were predicted using the miRanda miRNA target prediction tool. Statistical Methods qPCR data were assessed with a t test, and a p value of < 0.05 was significant. Microarray data were assessed by linear models (limma), and a false discovery rate (FDR) corrected p value of < 0.05 was significant. Expression Vector Production miRNA oligos were (sequences obtained from miRBase database) cloned into piggybac, tetracycline-inducible expression vectors. miRNA accessions and vectors are specified in Supplemental Experimental Procedures. Cell Culture ESCs were cultured on MEF15.5-seeded dishes as described previously (Nagy et al., 2003). TSCs were cultured in TSC media as described previously (Tanaka et al., 1998). To induce differentiation of TSCs and miR-induced lines, cells were cultured as per Tanaka et al. (1998). For cell line details, refer to Supplemental Experimental Procedures. Generation of Transgenic ESCs PolyJet Transfection Complex (FroggaBio) was used to transfect 6 3 106 cells with 12 mg total DNA, composed of 45% PB TET-miRNA, 10% PBase-Integrase, and 45% PB CAG-rtTA. Transgenic cell lines were re-targeted with an mCherry vector. For details of cell line characterization and vector information, refer to Supplemental Experimental Procedures. miRNA Induction Transgenic ESCs were seeded at a density of 2 3 104 per well of a MEF15.5containing six-well plate and cultured in ESC media for the first 24 hr. To initiate induction, cells were cultured in TSC media with doxycycline (1 mg/mL). 24 hr following doxycycline induction, valproic acid (0.2 mM) was added to the media. Media was replaced daily for up to 10 days. Real-Time qPCR and Microarray Experiments RNA isolation, cDNA preparation, and real-time qPCR microarray details are specified in Supplemental Experimental Procedures. Western Blots Lysates were harvested using Laemmli buffer (Bio-Rad) diluted in betamercaptoethanol (Gibco, 1,0003) and sonicated for 10 s. 20 mg protein was loaded onto premade Mini-PROTEAN TGX gels (Bio-Rad), and a Trans-Blot Turbo Kit was used to prepare polyvinylidene fluoride (PVDF) membranes (Bio-Rad). Membranes were incubated with primary antibody overnight at 4 C and with secondary antibody for 1 hr at room temperature.

1108 Cell Reports 19, 1101–1109, May 9, 2017

Antibody specifications and imaging details are specified in Supplemental Experimental Procedures. Embryo Chimeras Cells were prepared for injections as per Nagy et al. (2003). For post-implantation analysis, cells were injected into the blastocoel cavity of E3.5 blastocysts. Blastocysts were transferred into pseudopregnant outbred ICR (CD-1) albino female recipients. Embryos were dissected at E6.5. For preimplantation analysis, cells were injected into eight-cell-stage CD-1 embryos. Embryos were cultured in vitro until the blastocyst stage and stained with Hoechst for 20 min. Culture, sedimentation, and confocal imaging are specified in Supplemental Experimental Procedures. Microsphere Internalization E3.5 embryos were exposed to 1.75 mm microspheres at a final concentration of 15 mg/mL overnight. After fixation with 4% paraformaldehyde (PFA), embryos were mounted in Vectashield with DAPI (Vector Laboratories) for imaging. Cell cultures were grown on MEF-plated coverslips and exposed to 2 3 106 beads for 24 hr. Cultures were washed with PBS and fixed with 4% PFA for 15–20 min. Microsphere, embryo culture, and microscopy details are specified in Supplemental Experimental Procedures. Flow Cytometry for Cell-Cycle Analysis Cells were fixed with ice-cold 85% ethanol, washed with PBS, resuspended in FxCycle PI/RNase staining solution (Thermo Fisher Scientific), and incubated in room temperature for 30 min (protected from light). Samples were analyzed (without washing) using 488-nm excitation and collected emission using a 585/42 bandpass filter. ACCESSION NUMBERS The accession number for the microarray data reported in this paper is GEO: GSE79599. Additional ESC samples were added from GEO: GSE15519 and GEO: GSE29102, and TSC samples were added from GEO: GSE15519 and GEO: GSE25255. SUPPLEMENTAL INFORMATION Supplemental Information includes Supplemental Experimental Procedures, five figures, and three tables and can be found with this article online at http://dx.doi.org/10.1016/j.celrep.2017.04.040. AUTHOR CONTRIBUTIONS The project was conceptualized by B.C. The manuscript was prepared by U.N. and B.C. Identification of miRNAs was performed by B.C. Transgenic line generation, transdifferentiation, and phenotype characterization were performed by U.N. Microarray experiments were performed by U.N. and B.C. miR-15b constructs were generated and validated by T.H. Triple transgenic singlecell qPCR analysis was performed by F.L. and B.C. ACKNOWLEDGMENTS These studies were supported by funding from the Canadian Institute of Health Research (CIHR) (EPS-129130) and CRC tier II chair to B.C. (EPS-129130). We greatly acknowledge the support of Jodie Garner and the Rossant laboratory at SickKids Research Institute. We wish to acknowledge the contribution of Marina Gertsenstein, Monica Pereira, Sandra Tondat, and Suzanne McMaster at the Toronto Center for Phenogenomics (TCP) for assistance with the in vivo study. We thank Hitoshi Niwa for the gift of the CdxER ESC line. Received: May 16, 2016 Revised: January 25, 2017 Accepted: April 14, 2017 Published: May 9, 2017

REFERENCES Anokye-Danso, F., Trivedi, C.M., Juhr, D., Gupta, M., Cui, Z., Tian, Y., Zhang, Y., Yang, W., Gruber, P.J., Epstein, J.A., and Morrisey, E.E. (2011). Highly efficient miRNA-mediated reprogramming of mouse and human somatic cells to pluripotency. Cell Stem Cell 8, 376–388. Betel, D., Koppal, A., Agius, P., Sander, C., and Leslie, C. (2010). Comprehensive modeling of microRNA targets predicts functional non-conserved and non-canonical sites. Genome Biol. 11, R90. Elling, U., Klasen, C., Eisenberger, T., Anlag, K., and Treier, M. (2006). Murine inner cell mass-derived lineages depend on Sall4 function. Proc. Natl. Acad. Sci. USA 103, 16319–16324. Festuccia, N., Osorno, R., Halbritter, F., Karwacki-Neisius, V., Navarro, P., Colby, D., Wong, F., Yates, A., Tomlinson, S.R., and Chambers, I. (2012). Esrrb is a direct Nanog target gene that can substitute for Nanog function in pluripotent cells. Cell Stem Cell 11, 477–490. Gardner, R.L. (2000). Flow of cells from polar to mural trophectoderm is polarized in the mouse blastocyst. Hum. Reprod. 15, 694–701. Gu, Y., Sun, J., Groome, L.J., and Wang, Y. (2013). Differential miRNA expression profiles between the first and third trimester human placentas. Am. J. Physiol. Endocrinol. Metab. 304, E836–E843. Izarra, A., Moscoso, I., Can˜o´n, S., Carreiro, C., Fondevila, D., Martı´n-Caballero, J., Blanca, V., Valiente, I., Dı´ez-Juan, A., and Bernad, A. (2014). miRNA-1 and miRNA-133a are involved in early commitment of pluripotent stem cells and demonstrate antagonistic roles in the regulation of cardiac differentiation. J. Tissue Eng. Regen. Med. 11, 787–799. James, J.L., Carter, A.M., and Chamley, L.W. (2012). Human placentation from nidation to 5 weeks of gestation. Part I: What do we know about formative placental development following implantation? Placenta 33, 327–334. Joo, J.Y., Lee, J., Ko, H.Y., Lee, Y.S., Lim, D.H., Kim, E.Y., Cho, S., Hong, K.S., Ko, J.J., Lee, S., et al. (2014). Microinjection free delivery of miRNA inhibitor into zygotes. Sci. Rep. 4, 5417. Kunath, T., Arnaud, D., Uy, G.D., Okamoto, I., Chureau, C., Yamanaka, Y., Heard, E., Gardner, R.L., Avner, P., and Rossant, J. (2005). Imprinted X-inactivation in extra-embryonic endoderm cell lines from mouse blastocytes. Development 132, 1649–1661. Lewis, R.M., Cleal, J.K., and Hanson, M.A. (2011). Review: Placenta, evolution and lifelong health. Placenta 33 (Suppl), 1–5. Mayor-Lynn, K., Toloubeydokhti, T., Cruz, A.C.A.C., and Chegini, N. (2011). Expression profile of microRNAs and mRNAs in human placentas from pregnancies complicated by preeclampsia and preterm labor. Reprod. Sci. 18, 46–56. Miller, A., Ralser, M., Kloet, S.L., Loos, R., Nishinakamura, R., Bertone, P., Vermeulen, M., and Hendrich, B. (2016). Sall4 controls differentiation of pluripotent cells independently of the Nucleosome Remodelling and Deacetylation (NuRD) complex. Development 143, 3074–3084. Mouillet, J.-F., Donker, R.B., Mishima, T., Cronqvist, T., Chu, T., and Sadovsky, Y. (2013). The unique expression and function of miR-424 in human placental trophoblasts. Biol. Reprod. 89, 25. Nagy, A., Rossant, J., Nagy, R., Abramow-Newerly, W., and Roder, J.C. (1993). Derivation of completely cell culture-derived mice from early-passage embryonic stem cells. Proc. Natl. Acad. Sci. USA 90, 8424–8428. Nagy, A., Gertsenstein, M., Vintersten, K., and Behringer, R. (2003). Manipulating the Mouse Embryo: A Laboratory Manual, Third Edition (Cold Spring Harbor Laboratory Press). Nishiyama, A., Sharov, A.A., Piao, Y., Amano, M., Amano, T., Hoang, H.G., Binder, B.Y., Tapnio, R., Bassey, U., Malinou, J.N., et al. (2013). Systematic repression of transcription factors reveals limited patterns of gene expression changes in ES cells. Sci. Rep. 3, 1390.

Niwa, H., Toyooka, Y., Shimosato, D., Strumpf, D., Takahashi, K., Yagi, R., and Rossant, J. (2005). Interaction between Oct3/4 and Cdx2 determines trophectoderm differentiation. Cell 123, 917–929. Ralston, A., Draper, J., Cox, B., and Rossant, J. (2008). Gata3 regulates stem cell self-renewal and differentiation in the extraembryonic lineage during mouse development. Dev. Biol. 319, 479. Rassoulzadegan, M., Rosen, B.S., Gillot, I., and Cuzin, F. (2000). Phagocytosis reveals a reversible differentiated state early in the development of the mouse embryo. EMBO J 19, 3295–3303. Ritchie, M.E., Phipson, B., Wu, D., Hu, Y., Law, C.W., Shi, W., and Smyth, G.K. (2015). limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 43, e47. Rossant, J., and Cross, J.C. (2001). Placental development: lessons from mouse mutants. Nat. Rev. Genet. 2, 538–548. Rossant, J., and Tam, P.P.L. (2009). Blastocyst lineage formation, early embryonic asymmetries and axis patterning in the mouse. Development 136, 701–713. Rugg-Gunn, P.J., Cox, B.J., Lanner, F., Sharma, P., Ignatchenko, V., McDonald, A.C.C.H.H., Garner, J., Gramolini, A.O., Rossant, J., and Kislinger, T. (2012). Cell-surface proteomics identifies lineage-specific markers of embryo-derived stem cells. Dev. Cell 22, 887–901. Spruce, T., Pernaute, B., Di-Gregorio, A., Cobb, B.S., Merkenschlager, M., Manzanares, M., and Rodriguez, T.A. (2010). An early developmental role for miRNAs in the maintenance of extraembryonic stem cells in the mouse embryo. Dev. Cell 19, 207–219. Suh, N., Baehner, L., Moltzahn, F., Melton, C., Shenoy, A., Chen, J., and Blelloch, R. (2010). MicroRNA function is globally suppressed in mouse oocytes and early embryos. Curr. Biol. 20, 271–277. Tam, P.P., and Rossant, J. (2003). Mouse embryonic chimeras: tools for studying mammalian development. Development 130, 6155–6163. Tan, M.H., Au, K.F., Leong, D.E., Foygel, K., Wong, W.H., and Yao, M.W. (2013). An Oct4-Sall4-Nanog network controls developmental progression in the pre-implantation mouse embryo. Mol. Syst. Biol. 9, 632. Tanaka, S., Kunath, T., Hadjantonakis, A.K., Nagy, A., and Rossant, J. (1998). Promotion of trophoblast stem cell proliferation by FGF4. Science 282, 2072– 2075. Viswanathan, S.R., Mermel, C.H., Lu, J., Lu, C.-W., Golub, T.R., and Daley, G.Q. (2009). microRNA expression during trophectoderm specification. PLoS ONE 4, e6143. Winter, J., Jung, S., Keller, S., Gregory, R.I., and Diederichs, S. (2009). Many roads to maturity: microRNA biogenesis pathways and their regulation. Nat. Cell Biol. 11, 228–234. Xu, N., Papagiannakopoulos, T., Pan, G., Thomson, J.A., and Kosik, K.S. (2009). MicroRNA-145 regulates OCT4, SOX2, and KLF4 and represses pluripotency in human embryonic stem cells. Cell 137, 647–658. Yuri, S., Fujimara, S., Nimura, K., Takeda, N., Toyooka, Y., Fujimura, Y., Aburatani, H., Ura, K., Koseki, H., Niwa, H., and Nishinakamura, R. (2009). Sall4 is essential for stabilization, but not for pluripotency, of embryonic stem cells by repressing aberrant trophectoderm gene expression. Stem Cells 27, 796–805. Zhang, J., Tam, W.L., Tong, G.Q., Wu, Q., Chan, H.Y., Soh, B.S., Lou, Y., Yang, J., Ma, Y., Chai, L., et al. (2006). Sall4 modulates embryonic stem cell pluripotency and early embryonic development by the transcriptional regulation of Pou5f1. Nat. Cell Biol. 8, 1114–1123. Zhang, J., Wang, Y., Liu, X., Jiang, S., Zhao, C., Shen, R., Guo, X., Ling, X., and Liu, C. (2015). Expression and potential role of microRNA-29b in mouse early embryo development. Cell. Physiol. Biochem. 35, 1178–1187.

Cell Reports 19, 1101–1109, May 9, 2017 1109