SEARCH

SEARCH BY CITATION

Keywords:

  • Embryonic stem cells;
  • MicroRNAs;
  • Cell differentiation;
  • Retinoic acid

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosure of Potential Conflicts of Interest
  8. Acknowledgements
  9. References
  10. Supporting Information

Hundreds of microRNAs (miRNAs) are expressed in mammalian cells, where they aid in modulating gene expression by mediating mRNA transcript cleavage and/or regulation of translation rate. Functional studies to date have demonstrated that several of these miRNAs are important during development. However, the role of miRNAs in the regulation of stem cell growth and differentiation is not well understood. We show herein that microRNA (miR)-134 levels are maximally elevated at day 4 after retinoic acid-induced differentiation or day 2 after N2B27-induced differentiation of mouse embryonic stem cells (mESCs), but this change is not observed during embryoid body differentiation. The elevation of miR-134 levels alone in mESCs enhances differentiation toward ectodermal lineages, an effect that is blocked by a miR-134 antagonist. The promotion of mESC differentiation by miR-134 is due, in part, to its direct translational attenuation of Nanog and LRH1, both of which are known positive regulators of Oct4/POU5F1 and mESC growth. Together, the data demonstrate that miR-134 alone can enhance the differentiation of mESCs to ectodermal lineages and establish a functional role for miR-134 in modulating mESC differentiation through its potential to target and regulate multiple mRNAs.

Disclosure of potential conflicts of interest is found at the end of this article.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosure of Potential Conflicts of Interest
  8. Acknowledgements
  9. References
  10. Supporting Information

ESCs exhibit the capacity for unlimited self-renewal and the ability to differentiate into multiple cell lineages [1, 2]. Although significant progress has been made in elucidating the major roles and intricate interplay of key transcription factors such as Oct4, Sox2, and Nanog in regulation of self-renewal and differentiation of ESCs, the precise post-transcriptional mechanisms regulating these developmental choices in ESCs remain poorly understood. There is emerging evidence that noncoding RNAs, including microRNAs (miRNAs), that form part of the post-transcriptional machinery have an important role in modulating mRNA decay and translation rates in eukaryotes [3]. In the context of mouse ESCs (mESCs), it has been shown that DGCR8, an RNA-binding protein involved in miRNA processing, is essential for miRNA biogenesis and silencing of ESC self-renewal [4]. This highlights the importance of regulated miRNA expression in controlling ESC growth and differentiation. ESC-specific miRNAs have been identified in mouse and human ESCs [5, 6]; however, their functional significance has not yet been evaluated. Elucidation of miRNA function in the context of ESCs will likely facilitate attempts to manipulate stem cell growth or direct lineage specification and maturation.

Retinoic acid (RA) promotes differentiation of ESCs to ectodermal lineages [7, [8]9]. RA is a morphogen that is required during development of the central nervous system, lung, and kidneys and during proximodistal patterning and limb generation [10, [11], [12]13]. Dysfunction in RA synthesis or RA receptor activation can lead to embryonic lethality [14]. Neuroectodermal differentiation of mESCs can also be induced under conventional serum-free culture conditions with N2 and B27 supplements (N2B27) [15].

Recent studies in invertebrate model systems have identified lsy-6, the first miRNA found to play a role in neuronal patterning [16], and microRNA (miR)-9a, which ensures the generation of the precise number of neuronal precursor cells during development [17]. Related work has also indicated key roles for miRNAs during neural differentiation in vitro [18, 19] and in vertebrate central nervous system development [20, [21]22]. In particular, miR-134 has recently been identified as a potential regulator of dendritic spine volume and synapse formation in mature rat hippocampal neurons in vitro through the localized repression of a protein kinase, LimK1 [23]. The mouse homologue of miR-134, which demonstrates conservation across rodents and primates, was originally identified by cloning from the mouse cortex [24] and is located in a large imprinted miRNA gene cluster at the mouse Dlk1-Gtl2 domain [3].

Here, we report on our findings that implicate miR-134 in a previously uncharacterized role in enhancing differentiation of mESCs to ectodermal lineages. Beginning with microarray analysis, miRNAs whose expression changed significantly during mESC differentiation were identified. Among the miRNAs that were altered during mESC differentiation, miR-134 was upregulated by RA and N2B27 treatment but not during embryoid body differentiation. Interestingly, elevated levels of miR-134 enhanced mRNA levels of Sox1, Nestin, and Neurogenin-2, markers associated with differentiation toward ectoderm [15, 25]. This effect was blocked using an miR-134 antagonist. To investigate potential mechanisms by which miR-134 affects mESCs, the miRNA target prediction algorithm, rna22 [26], was used to identify putative mRNA targets of miR-134. Endogenous protein levels of four of these target genes tested, including Nanog and LRH1, were reduced by elevating miR-134 levels without concomitant decreases of the target mRNA levels. Together, these data provide evidence that miR-134 enhances differentiation of mESCs to ectodermal lineages and establish a functional role for miR-134 in modulating mESC differentiation through its activity against genes that include Nanog and LRH1, both of which are known to be positive regulators of Oct4 and of mESC growth.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosure of Potential Conflicts of Interest
  8. Acknowledgements
  9. References
  10. Supporting Information

Cell Culture and Transfection

All cell culture reagents and culture plastics were obtained from Gibco (Grand Island, NY, http://www.invitrogen.com) and BD Biosciences (San Diego, http://www.bdbiosciences.com), respectively, unless otherwise specified. Cell cultures were maintained at 37°C with 5% CO2. mESC lines E14 (CRL-1821; American Type Culture Collection, Manassas, VA, http://www.atcc.org) and D3 (CRL-1934; American Type Culture Collection) were cultured feeder-free on 0.1% gelatin-coated (Sigma-Aldrich, St. Louis, http://www.sigmaaldrich.com) plates in ESC medium (Dulbecco's modified Eagle's medium, 15% heat-inactivated embryonic stem-standard fetal bovine serum, 100 μM nonessential amino acids, 2 mM l-glutamine, 55 nM β-mercaptoethanol, 1% [vol/vol] penicillin/streptomycin, and mouse leukemia inhibitory factor [mLIF; 103 U/ml; Chemicon, Temecula, CA, http://www.chemicon.com]). HEK 293T/17 (CRL-11268; American Type Culture Collection) cells were maintained in Dulbecco's modified Eagle's medium supplemented with 10% heat-inactivated fetal bovine serum and penicillin/streptomycin. To induce mESC differentiation, mESCs were cultured in leukemia inhibitory factor (LIF)-deficient ESC medium with all-trans RA (100 nM) or cultured in N2B27 medium as described previously [15]. To form embryoid bodies, mESCs were trypsinized to a single-cell suspension and subsequently cultured on uncoated Petri dishes in ESC medium without mLIF. Media were changed every 2 days for all mESC differentiation conditions.

Pre-miR miRNA precursors (Pre-miRs), Anti-miR miRNA inhibitors (Anti-miRs), and scrambled RNA oligomer (Scr; AGACUAGCGGUAUCUUUAUCCC) were purchased from Ambion. These were transfected into mESCs at a final concentration of 100 nM using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, http://www.invitrogen.com) as per the manufacturer's instructions. For transcript and protein analysis, 1.5 μg of gene-specific short hairpin RNA (shRNA) plasmids, nonsilencing shRNA plasmid, or empty vector were transfected into mESCs in 12-well plates, and drug selection was performed with 1 μg/ml of puromycin (Sigma-Aldrich) 24 hours post-transfection for a period of 2 days.

RNA Extraction and Microarray Analysis

For microarray, Northern blot, and reverse transcription (RT)-polymerase chain reaction (PCR) analyses, total RNA was extracted from cells using Trizol reagent as per the manufacturer's instructions and subsequently column-purified with RNeasy kits (Qiagen, Hilden, Germany, http://www1.qiagen.com). miRNA microarray was performed using a service provider (LC Sciences, Houston, TX, http://www.lcsciences.com). A brief description is given in the supplemental online data. The total array profile has Minimum Information About a Microarray Experiment (MIAME) accession number E-MEXP-977 (http://www.ebi.ac.uk/arrayexpress). Arrays were performed in duplicate, and 134 microRNAs were selected as significant (≥twofold change). Expression profiling of coding genes was carried out using Illumina MouseRef-8 BeadArrays as per the manufacturer's instructions (Illumina Inc., San Diego, http://www.illumina.com). Total chip data have been deposited for public access with GEO repository accession number GSE4522. Arrays were performed in duplicate, and 1,204 genes were selected as significant (≥twofold change).

Northern Blot Analysis

miRNAs were probed using α32P uridine triphosphate (UTP)-labeled RNA probes transcribed in vitro from a T7 promoter driving DNA oligomer templates. 5S ribosomal RNA (rRNA) was probed using γ32P-end-labeled DNA probe. The specific activity of each probe was routinely 107 cpm/μg RNA/DNA. Ten micrograms of total RNA was size-fractionated by 15% Tris-borate-EDTA-urea polyacrylamide gel electrophoresis (PAGE) and electroblotted onto MagnaProbe nylon membrane (GE Osmonics Labstore, Minnetonka, MN, http://www.osmolabstore.com). The membrane was probed overnight at 42°C in 6× standard saline citrate (SSC), 0.2% SDS, 5× Denhardt's solution with 32P-labeled RNA probes (1 × 106 cpm/ml) and washed three times at 30°C in 6× SSC, 0.2% SDS. Complexes were detected by Phosphoimager (GE Healthcare, Little Chalfont, Buckinghamshire, U.K., http://www.gehealthcare.com).

RT-PCR

cDNA synthesis for miRNAs was performed with 50 ng of total RNA according to the manufacturer's instructions (Applied Biosystems, Foster City, CA, http://www.appliedbiosystems.com). Endogenous miRNA levels were measured with inventoried TaqMan probes, and PCR Master Mix (Applied Biosystems). miR-16 was used as an internal control [27]. cDNA synthesis for coding genes was performed with 1 μg of total RNA according to the manufacturer's instructions (Applied Biosystems). Endogenous mRNA levels of pluripotency and differentiation markers were measured with inventoried TaqMan probes and PCR Master Mix. Endogenous mRNA levels of LRH1, FADD, and Gαo were measured using SYBR Green PCR Master Mix (Applied Biosystems). Primer sequences are given in the supplemental online data. β-Actin was used as an internal control. All amplicons were analyzed using ABI Prism 7900HT Sequence Detection System 2.2 software (Applied Biosystems).

Protein Extraction and Western Blot Analysis

Cell pellets were washed in chilled phosphate-buffered saline (PBS) and incubated for 20 minutes in ice cold lysis buffer containing freshly added protease inhibitors. Lysates were cleared by centrifugation at 4°C for 10 minutes at 12,100g, and protein concentrations were determined using Bradford dye (Bio-Rad, Hercules, CA, http://www.bio-rad.com). For Western blot analysis, 10 μg of total protein was size-fractionated by SDS-PAGE on 10% bis-Tris acrylamide NuPAGE gels and transferred to Hybond-P polyvinylidene difluoride (PVDF) membrane (GE Healthcare) in 1× NuPAGE transfer buffer (Invitrogen) with 10% methanol. The membrane was probed with specific primary antibodies (Oct4, sc8628; Sox2, sc17320; bone morphogenetic protein [BMP] 4, sc6896; Gata4, sc1237; Nestin, sc21248; Gαo, sc387; FADD, sc6036; Santa Cruz Biotechnology Inc., Santa Cruz, CA, http://www.scbt.com; LRH-1, F4100–44; United States Biological, Swampscott, Massachusetts, http://www.usbio.net; Nanog, AB5731; β-actin, MAB1501; Chemicon) and secondary horseradish peroxidase-conjugated antibodies (anti-goat horseradish peroxidase [HRP], sc2768; anti-rabbit HRP, sc2030; anti-mouse, sc2005; Santa Cruz Biotechnology). Antibody-protein complexes were identified by ECL-Plus (Amersham Biosciences, Piscataway, NJ, http://www.amersham.com) and film.

Colony Formation Assay

E14 cells mESCs were transfected with Pre-miRs (100 nM) and/or Anti-miRs (100 nM), scrambled RNA oligomer (100 nM), or Lipofectamine only (Mock Transfection) in 12-well culture plates at a density of 5 × 105 cells per well. These cells were subsequently trypsinized at 48 hours and resuspended in ESC medium. Various numbers of cells (200 and 400) were plated onto mouse feeder layers in six-well culture plates to form secondary ESC colonies. After 7 days, emerging colonies were stained using Wright-Giemsa stain (Sigma-Aldrich) and counted. Colony morphology and number provide an indication of the number of colony-forming undifferentiated mESCs present in a population of cells [28].

Immunostaining and Detection of Alkaline Phosphatase

Cells were fixed in 4% paraformaldehyde, washed twice with PBS, and then incubated for 5 minutes at −20°C in 95% ethanol (vol/vol in PBS). Cells were then washed three times with PBS, blocked for 1 hour in 5% normal goat serum in PBS with 0.1 × Triton X-100, and incubated overnight with anti-class III β-tubulin (Ab18660; Abcam, Cambridge, U.K., http://www.abcam.com) primary antibody at 4°C. Cells were next washed twice with PBS and incubated for 1 hour with the corresponding secondary antibody (goat anti-rabbit IgG 568; Molecular Probes, Eugene, OR, http://probes.invitrogen.com). Cells were washed and mounted. Immunofluorescence was observed using a Leica microscope (Heerbrugg, Switzerland, http://www.leica.com), and image analysis was conducted using IM50 software (Leica Microsystems, http://www.leica-microsystems.com/). The Alkaline Phosphatase Detection Kit (Chemicon) was used to determine alkaline phosphatase activity, according to the manufacturer's instructions.

Quantitative Analysis of Immunofluorescence

Image acquisition and subsequent neurite outgrowth measurements were performed using the Cellomics ArrayScan II High Content Screening platform (Swallowfield, U.K., http://www.cellomics.com). Two separate fluorescent filters at 20× objective magnification were used; channel 1 illuminated Hoechst 33342-labeled nuclei, allowing automated focusing upon the cells, and channel 2 excited objects stained with Alexa Fluor 568-labeled anti-class III β-tubulin. Both images were acquired through a 535 nm × 35 nm bandwidth dichroic emission filter. Exposure times for each wavelength were determined empirically. All images were then analyzed using Cellomics neurite outgrowth software, and data analysis for the number of neurons per total nucleus count in each field and average intensity of neuronal cell bodies and neurites are presented. Data shown are an average of values from 16 random fields from three replicate wells in two independent experiments.

Vector Construction

pOct4-Luciferase and pNanog-Luciferase were as described previously [29]. pLuc-microRNA response element (MRE) constructs were generated as described previously [26]. The pLuc-Nanog3′-untranslated region (UTR) construct was generated by cloning the entire Nanog 3′UTR into psiCHECK-2 (Promega, Madison, WI, http://www.promega.com) at the NotI restriction site. To generate the Nanog 3′UTR mutant construct, PCR-directed mutagenesis was performed with a pair of primers containing the mutant MRE sequence (GAGGACGTGTTAACTAGTTTCC). The PCR product was subsequently used as a template for full-length PCR. The Nanog 3′UTR mutant was also cloned into psiCHECK-2 at the NotI restriction site. For shRNA constructs, 19-base pair gene-specific regions for RNA interference (RNAi) were designed and cloned into pSUPER.puro (BglII and HindIII sites) as described previously [29]. All vector and oligonucleotide sequences are available on request.

Luciferase Assays

The pOct4/pNanog-Luciferase assays were performed as described previously [29]. A brief description is given in the supplemental online data. miRNA target validation assays were performed as described previously [26]. A brief description is given in the supplemental online data.

In Situ Hybridization

Whole-mount and sectioned in situ hybridization were performed as previously described [30, 31].

Statistical Analysis

Unless otherwise stated, the unpaired Student's t test was used to determine statistical significance. For microarray data, the Pearson's correlation was used for the similarity test. Dendrograms were drawn using Eisenplots (clustering and treeview tools), with arm lengths proportional to the respective correlation coefficients.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosure of Potential Conflicts of Interest
  8. Acknowledgements
  9. References
  10. Supporting Information

Identification of miRNAs Modulated During mESC Differentiation

Microarray analysis of total RNA isolated from E14 mESCs ± RA (100 nM) at 0, 2, 4, and 6 days post-treatment interrogated the relative changes in miRNA expression during mESC differentiation. One hundred thirty-four miRNAs were differentially expressed following RA treatment, and expression of a representative sample was confirmed by Northern blot analysis (supplemental online Fig. 1A–1D). The expression pattern of several miRNAs increased 4 days after RA treatment (supplemental online Fig. 1A, 1C).

In particular, miR-134 expression increased after 4 days of RA treatment, with subsidence by day 6, but still at elevated levels above control (Fig. 1A–1C; supplemental online Fig. 1E). When an alternative neuroectodermal lineage differentiation promoting regime was used, the levels of endogenous miR-134 increased significantly within 2 days with N2B27 treatment (three times greater induction compared with RA), and again the levels of miR-134 subsided over the time course but remained elevated above control (Fig. 1B). Interestingly, miR-134 levels exhibited a small, but significant, decrease during EB formation (Fig. 1B).

thumbnail image

Figure Figure 1.. miR-134, an RA- and N2B27-responsive miR, reduces the promoter activity of pluripotency markers in mESCs. (A): miRNA microarray analyses demonstrating upregulation of miR-134 in mESCs treated with 100 nM RA for 4 days (compared with untreated mESCs). Asterisk (∗) denotes significance of difference from day 0 at p < .0001. (B): miRNA quantitative PCR analyses showing that, relative to expression in untreated E14 mESCs, miR-134 expression increases during both RA and N2B27 treatment of E14 mESCs and not EB differentiation. Experiments were performed in duplicate three times (n = 6), where error bars denote SE, and asterisk (∗) denotes significance of difference from day 0 at p < .0001. (C): Northern blot analyses demonstrating upregulation of miR-134 in mESCs treated with 100 nM RA for 4 days (compared with untreated mESCs), confirming the data obtained from the miRNA microarray and quantitative PCR analyses. Experiments were performed in triplicate twice (n = 6), and a representative blot is shown. (D,E): Relative to the Scr control transfection, PmiR-134 transfection induced a significant downregulation in Oct4 promoter (D) and Nanog promoter (E) activities in both E14 and D3 mESCs. Experiments were performed in triplicate three times (n = 9), where error bars denote SE, and asterisk (∗) denotes significance of difference from scrambled RNA oligomer control at p < .0001. (F): Colony formation assay results demonstrating that relative to Scr control, transfection of PmiR-134 into E14 mESCs significantly reduced the proportion of undifferentiated, colony-forming mESCs in the cell population. This effect was effectively blocked by AmiR-134. The data are expressed as mean number of colony-forming units ± SE. Experiments were performed in triplicate three times (n = 9). Abbreviations: AmiR, Anti-miR microRNA inhibitor; mESC, mouse embryonic stem cell; miR, microRNA; MT, mock transfection using water; PmiR, Pre-microRNA microRNA precursor; RA, retinoic acid; RNAi, RNA interference; Scr, Scrambled RNA oligomer control.

Download figure to PowerPoint

Differentiation of mESCs by miRNAs

The transcription factors Oct4, Sox2, and Nanog are known to play a central role in ESC maintenance and differentiation, and the levels and activity of these transcription factors are indicators of ESCs' pluripotent or differentiation status [32]. Constructs containing either the Oct4 promoter or the Nanog promoter that contain Oct4/Sox2 binding sites driving transcription of a luciferase reporter gene were used to measure the transcription activity of these promoters and provide a sensitive method to detect loss of pluripotency and onset of differentiation upon upregulation of miRNAs in the presence of serum and LIF (supplemental online Fig. 2A, 2B). Of the 11 microRNAs tested by treating mESCs (E14 and D3) with Pre-miR miRNA precursors (Pre-miRs), only miR-134 resulted in a significant reduction in Oct4-promoter activity and Nanog-promoter activity (Fig. 1D, 1E; supplemental online Fig. 2C, 2D) relative to a scrambled oligomer control (Scr) in both E14 and D3 mESCs.

miR-134 Modulates mESC Differentiation, Even in the Presence of LIF

In addition to the Oct4- and Nanog-promoter activity assays, colony-forming unit assays demonstrated that there was an ∼30% decrease in the ability of mESCs treated for 2 days with Pre-miR-134 to form mESC colonies compared with scrambled RNA oligomer-treated controls (Fig. 1F). This suggested that miR-134 alone could modulate mESC differentiation.

To examine whether miR-134 alone could induce changes in cell phenotype as determined by changes in the mRNA and protein levels of key marker genes associated with pluripotency and differentiation, quantitative PCR of total RNA samples taken from E14 and D3 mESCs 3 days post-transfection with Pre-miR-134 was performed. This showed that the Oct4 mRNA levels were significantly reduced, consistent with the reduction of promoter activity (Fig. 2A). Nanog mRNA levels were reduced significantly in D3, but not E14, mESCs transfected with Pre-miR-134. Sox2 mRNA levels were reduced significantly in both D3 and E14 mESCs after Pre-miR-134 transfection.

thumbnail image

Figure Figure 2.. In the presence of leukemia inhibitory factor, microRNA (miR)-134 modulates transcript levels of lineage-specific biomarkers, downregulates protein levels of pluripotency markers, and induces changes in mESC morphology indicative of differentiation. (A): Relative to the Scr control transfection, transfection of PmiR-134 into E14 and D3 mESCs led to an appreciable upregulation of ectodermal markers Nestin and Fgf5 after 3 days, with a concomitant reduction in endodermal, mesodermal, and pluripotency-associated markers. Experiments were performed in triplicate three times (n = 9), where error bars denote SE, and significance of difference from scrambled RNA oligomer control at p < .001 are denoted by asterisk (∗) and hash mark (#) for E14 and D3 mESCs, respectively. (B): Western blot analyses showing that PmiR-134 induced significant downregulation of Nanog, Oct4, and Sox2 protein levels relative to Scr control in E14 and D3 mESCs. This effect was blocked by AmiR-134, more than 3 days post-transfection. Experiments were performed in triplicate twice (n = 6), and a representative blot is shown. (C): Western blot analyses showing that PmiR-134 induced significant upregulation of Nestin protein relative to Scr control in E14 mESCs. Experiments were performed in triplicate twice (n = 6), and a representative blot is shown. (D): Photographs showing the morphology (left panel) and alkaline phosphatase (AP) staining (right panel) of E14 mESCs 4 days post-transfection with PmiR-134 or Scr control, where PmiR-134, unlike Scr, induced visible phenotypic changes to a flattened, epithelial-like morphology and reduced AP activity relative to Scr control. Scale bar = 10 μm. Abbreviations: AmiR, Anti-microRNA microRNA inhibitor; mESC, mouse embryonic stem cell; PmiR, Pre-microRNA microRNA precursor; Scr, Scrambled RNA oligomer control.

Download figure to PowerPoint

Consistently, miR-134 elevation resulted in a significant downregulation of Oct4, Nanog, and Sox2 proteins in both E14 and D3 mESCs, relative to Scr control, during the second and third days following transfection with Pre-miR-134 (Fig. 2B). Importantly, this effect was abolished by cotransfection of Pre-miR-134 with Anti-miR-134, miR-134's antagonist, thus demonstrating the specificity of miR-134 in the downregulation of these proteins (Fig. 2B). No changes in protein levels with Anti-miR-134 alone were detected, and nonspecific Anti-miRs that targeted other miRNAs were unable to rescue the observed effect of Pre-miR-134 (data not shown).

Using stringent thresholds for the rna22 algorithm [26] screening entire mRNA sequences, an MRE for miR-134 was predicted in the 3′UTR of Nanog (discussed below). No predicted MREs were found in the 5′UTR, coding region, or 3′UTR of Oct4 under the stringent thresholds used in the context of this paper. However, by relaxing these thresholds, rna22 predicts that there may be several MREs present in Oct4 mRNA. This is the subject of ongoing investigation.

In addition, elevated mRNA levels of early primitive ectoderm marker (Fgf5) and neuroectoderm marker (Nestin) were observed, with no change or a downregulation of endoderm (Gata4 and Sox17) and mesoderm (Bmp4 and Nkx2.5) marker transcripts (Fig. 2A). Conversely, transfection of other miRNAs (e.g., miR-22) did not elicit any substantial change in the level of these transcripts (data not shown). These findings indicate that elevated levels of miR-134 alone in undifferentiated mESCs promotes a transcriptional expression profile that is suggestive of differentiation toward ectoderm. Concomitant with changes in mRNA levels in E14 mESCs, Pre-miR-134 transfection led to increased protein levels of Nestin and decreased protein levels of BMP4. Gata4 protein levels were not affected (Fig. 2C).

miR-134 Induces Morphological Changes Indicative of mESC Differentiation

It was next determined whether miR-134 could elicit phenotypic effects. E14 mESCs transfected with water only (mock transfection), scrambled RNA oligomer (Scr), Pre-miR-Let-7i (a miRNA that did not promote mESC differentiation), Anti-miR-134, or Anti-miR-134+Pre-miR-134 did not display morphological changes for up to 4 days post-transfection (Fig. 2D, for Scr). The cells maintained the characteristic domed colony structures of mESCs. However, transfection with Pre-miR-134 alone induced visible morphological changes in these cells within 4 days post-transfection (Fig. 2D), with the cells acquiring a flattened epithelial-like morphology typical of differentiation. These data are consistent with reduced alkaline phosphatase activity in Pre-miR-134-treated mESCs (Fig. 2D), where alkaline phosphatase is expressed at higher levels in pluripotent ESCs compared with differentiated cells.

The mRNA Expression Patterns Between RA-Treated and miR-134-Transfected mESCs Demonstrate a High Degree of Correlation

The microarray profile of mESCs treated with Pre-miR-134 was compared with that of RA treatment alone to investigate the effects of miR-134 during RA-induced differentiation of mESCs. For RA-induced differentiation, total RNA was collected from untreated mESCs, and from mESCs after 2 and 4 days of RA treatment (in the absence of LIF), whereas for Pre-miR-induced differentiation, total RNA was collected 3 days after transfection of mESCs (in the presence of LIF) with Pre-miR-134, water only (mock transfection), scrambled oligo, or Pre-miR-Let-7i. RNA samples were labeled and hybridized to Illumina microarray chips, and heat maps were generated to compare the spectrum of transcript levels altered by RA or Pre-miR-134 treatment.

There was a high degree of correlation between the expression pattern of genes altered after 4 days of RA treatment in mESCs and all Pre-miR-134-responsive genes (Fig. 3A, 3B). Prior to Pearson correlation analysis, the intensity data were log2-transformed and subtracted from the medium intensity. The mean Pearson correlation coefficient between these two expression profiles (and replicates) was ∼0.57, whereas the mean Pearson correlation coefficient between Pre-miR-134 transfection/4 days of RA treatment and other treatments (ESC/mock transfection using water [MT]/Scr/Pre-Let-7i) was ∼−0.36. When comparing the means of the above Pearson correlation coefficients between RAd4 and miR-134 to the corresponding means between miR-134/RAd4 and ESC/MT/Scr/Pre-Let-7i, all p values were <.001. Analysis of mRNAs upregulated in response to either RA or Pre-miR-134 indicated increased levels of transcripts associated with ectodermal differentiation (Fig. 3C) [33, [34], [35], [36], [37]38], supporting the evidence presented in Figure 2. In contrast, levels of nonectoderm markers were not significantly altered by either RA or Pre-miR-134 [33, [34], [35], [36], [37]38]. Transfection of Pre-Let-7i, an miRNA that neither induces mESC differentiation nor is induced by RA, resulted in a very different pattern of gene expression compared with Pre-miR-134 treatment (Fig. 3A).

thumbnail image

Figure Figure 3.. miR-134 induces a subset of genes similar to that induced by RA treatment of mouse ESCs. (A): Heat map depicting the expression profile of all genes with changes ≥2 fold after 4 days of RA treatment (left panel). The retinoic acid treatment day 4 (RAd4) dataset was ordered based on descending average intensity and the expression profile of these genes in the PmiR-134-treated dataset is shown in the same order (right panel). (B): Dendrogram obtained from hierarchical sample clustering showing the relatedness of all the samples profiled in (A). (C): Heat map depicting the expression profile of ectoderm, endoderm, and mesoderm markers from RA-treated versus PmiR-134-transfected datasets. Each of the lineage markers genes are mean-centered separately in the PmiR-134- and RA-treated datasets and ordered based on the fold change in the RA-treated dataset. Abbreviations: ESC, mouse embryonic stem cell; miR, microRNA; MT, mock transfection using water; RA, retinoic acid; Scr, Scrambled RNA oligomer control.

Download figure to PowerPoint

miR-134 Enhances RA- and N2B27-Mediated Differentiation of mESCs

In both E14 and D3 mESCs (data not shown for D3), transfection of Pre-miR-134 in RA-treated cells resulted in accelerated acquisition of the early primitive ectoderm marker FgF5 and the neuroectoderm markers Nestin and Neurogenin2 (Fig. 4A), with a concomitant increase in Sox1, a promoter of neurogenesis [15], relative to RA-treated cells transfected with a control oligomer (Scr). The enhancement of these mRNAs by RA was blocked by Anti-miR-134 (Fig. 4A). Pre-miR-134 enhanced the RA-induced decrease in the pluripotent marker Oct4; however, unlike the neuroectoderm markers, this effect was not blocked by Anti-miR-134 (Fig. 4A). There were no major changes in the transcript levels of mesendoderm (Gata4 and Bmp4) markers during these experiments (Fig. 4A). Mouse ESCs cotreated with both Pre-miR-134 and RA exhibited lower protein levels of Oct4, Sox2, and Nanog compared with treatment with Scrambled, Anti-miR-134, or Anti-miR-18 controls (Fig. 4B).

thumbnail image

Figure Figure 4.. MicroRNA (miR)-134 enhances the effect of RA and N2B27 medium. (A): Transfection of PmiR-134 into RA-treated E14 mouse embryonic stem cells (mESCs) led to an appreciable upregulation of ectodermal markers, with a concomitant reduction in the pluripotency-associated marker Oct4, above that induced by transfection of Scr control into RA-treated E14 mESCs. Transcript levels of biomarkers were normalized to internal β-actin and expressed relative to day 0 control levels. Experiments were performed in triplicate three times (n = 9), where error bars denote SE, and significance of difference from scrambled RNA oligomer control on RA day 2 at p < .001 are denoted by asterisks (∗) and hash marks (#) for PmiR-134 and AmiR-134, respectively. (B): Western blot analyses showing that PmiR-134 enhances the RA-induced downregulation of Nanog, Oct4, and Sox2 protein levels relative to Scr control in E14 mESCs. Experiments were performed in triplicate twice (n = 6), and a representative blot is shown. (C): Immunostaining of β-III-tubulin 4 and 6 days after the transfection of PmiR-134 into N2B27-treated E14 mESCs depicting that PmiR-134 transfection increases the number of β-III-tubulin-positive cells compared with the Scr control transfection. Scale bar = 50 μm. (D): Quantification of the immunostaining depicted in (C). Data shown are presented as the ratio of the respective readings for PmiR-transfected cells/Scr-transfected cells and are an average of values from 16 random fields from three replicate wells in two independent experiments. Error bars denote SE, and asterisk (∗) denotes significance of difference from scrambled RNA oligomer control at p < .0001. Abbreviations: AmiR, Anti-microRNA microRNA inhibitor; PmiR, Pre-microRNA microRNA precursor; RA, retinoic acid; Scr, Scrambled RNA oligomer control.

Download figure to PowerPoint

The effect of Pre-miR-134 in the presence of N2B27 medium [15] was also tested. Mouse ESCs were subjected to Pre-miR-134 transfection for 1 day before N2B27 treatment for 3 or 5 days. Cell cultures transfected with Pre-miR-134 contained a larger percentage of cells positive for the neuronal marker β-III-tubulin relative to scrambled control (Fig. 4C, 4D). In addition, Cellomics analysis determined that the intensity of β-III-tubulin immunostaining was significantly stronger in neuronal cell bodies and neurites in Pre-miR-134-treated cultures relative to Scr-treated control cultures (Fig. 4D). These data are indicative of accelerated neural induction under these conditions.

miR-134 Targets Nanog and LRH1, Among Other Genes

miRNAs play a role in all aspects of post-transcriptional regulation where they regulate proteins involved in splicing, translation inhibition, and mRNA degradation [39, [40]41]. To begin to elucidate the potential role of miR-134, its predicted mRNA targets were ascertained. Using rna22 [26], 2,800 potential 3′UTR miR-134 targets were identified. Of these, 158 predicted targets were selected for validation in vitro with luciferase assays. For each target transcript, a single copy of the sequence segment that rna22 identified as a putative miR-134 binding site (MRE), was cloned within the 3′UTR of Renilla luciferase reporter mRNAs. As was reported previously [26], a total of 129 of 158 tested targets of miR-134 were validated using luciferase assays, thus suggesting that a high proportion (>80%) of the rna22-predicted targets may be post-transcriptionally regulated by miR-134.

Gene expression analysis revealed that 1,051 of the genes potentially targeted by miR-134 were expressed in mESCs (supplemental online Fig. 3A, 3B). Of these, approximately 50% were upregulated, 20% remained unchanged, and 30% were downregulated relative to water only (mock transfection) and scrambled oligomer control transfections. This suggests that miR-134 can have different effects, direct or indirect, on target mRNAs, where it may affect mRNA translation, degradation, and splicing. Interestingly, the expression profile of predicted target genes after Pre-miR-134 transfection was highly correlated with their corresponding expression profile after 4 days of RA treatment (supplemental online Fig. 3A, 3B). The mean Pearson correlation coefficient between these two expression profiles was 0.51, whereas the mean Pearson correlation coefficient between predicted target gene expression in Pre-miR-134 transfection or after 4 days of RA treatment and other treatments (ESC/MT/Scr/Pre-Let-7i) was ∼−0.39.

To begin to understand the role of miR-134 in the post-transcriptional regulation of its potential direct targets, four predicted mRNA targets were selected for further characterization. Upon Pre-miR-134 treatment, LRH1, Gαo, and Nanog mRNA levels did not change, and FADD mRNA levels increased. The predicted MREs in the 3′UTRs of these four genes are depicted in supplemental online Fig. 3C. The orphan nuclear receptor liver receptor homologue 1 (LRH1, also known as NR5A2), is a transcription factor that binds to the proximal enhancer (PE) and proximal promoter (PP) regions of the Oct4 promoter (supplemental online Fig. 2A) and regulates its expression [42]; it is also known to modulate cell proliferation through its interaction with β-catenin [43]. LRH1 is an important modulator of stem cell fate and is required for maintaining Oct4 expression in the epiblast of the embryo; loss of LRH1 leads to early embryonic lethality [42]. Gαo (guanine nucleotide binding protein, α o) and FADD (Fas-associated via death domain) have been shown to play key roles in neurite extension [44], apoptosis [45], and embryo development [46].

For these four targets, miR-134 elicited a significant decrease in luciferase-MRE activity (Fig. 5A). No effect was observed with Pre-miR-124a, Pre-miR-21, or Scr control, which were not predicted to target these MREs. Secreted frizzled receptor protein 2 (Sfrp2) is an example of a predicted MRE that was not affected by Pre-miR-134 (Fig. 5A). Luciferase activity of miR-134 reverse complement (RC) was suppressed by 92% with Pre-miR-134 transfection (Fig. 5A), confirming Pre-miR-134 activity in this assay. To ensure that miR-134's effects are not due to nonspecific random targeting of any MRE, the RC sequence of another miRNA, miR-21, was cloned into pLuc-MRE to act as a negative control. No effect of Pre-miR-134 against the miR-21 RC was observed. In addition, miR-134 could specifically suppress luciferase activity when the Nanog 3′UTR was cloned within the 3′UTR of the Renilla luciferase reporter (Fig. 5B). The effect of miR-134 on Nanog's 3′UTR was abolished when its predicted MRE was mutated while the integrity of the rest of the 3′UTR was maintained (Fig. 5B, 5C). Other miRNAs that were not predicted to target Nanog had no effect.

thumbnail image

Figure Figure 5.. MicroRNA (miR)-134 reduces the protein levels of predicted pluripotency-associated targets LRH1, Gαo, and Nanog without altering their mRNA levels. (A): Graph showing the effect on luciferase activity in 293T cells of PmiR and Scr control transfections against five predicted miR-134 MREs. Relative to Scr control, PmiR-134 reduced luciferase-MRE activity of four of the five constructs, but not PmiR-21 or PmiR-124a. These unrelated PmiRs were used to detect any false positives. The PmiRs were also tested with luciferase-MRE constructs containing the perfect RCs of miR-134 (134 RC) and miR-21 (21 RC) as positive and negative controls, respectively. (B): Graph showing the effect on luciferase activity in 293T cells of PmiR and Scr transfections against the entire Nanog 3′UTR. Relative to Scr control, PmiR-134 reduced luciferase-UTR activity, but not PmiR-21 or PmiR-296. All luciferase experiments were performed in quadruplicate three times (n = 12), where error bars denote SE, and asterisk (∗) denotes significance of difference from scrambled RNA oligomer control at p < .0001. (C): Sequences and binding heteroduplexes of the predicted miR-134 MRE within Nanog's 3′UTR and its mutant counterpart. (D): Western blot analyses showing that PmiR-134 transfection of E14 mESCs, relative to Scr transfection, led to a reduction in the endogenous protein levels of LRH1, FADD, Gαo, and Nanog. Experiments were performed in triplicate twice (n = 6), and a representative blot is shown. (E): Graph showing quantification of the Western blots in (D); asterisk (∗) denotes significance of difference from scrambled RNA oligomer control at p < .001. (F): Graph showing the percentage change in mRNA levels of LRH1, FADD, Gαo, and Nanog in PmiR-134-transfected E14 mESCs ± RA relative to Scr transfection without RA treatment. Experiments were performed in triplicate three times (n = 9), where error bars denote SE, and asterisk (∗) denotes significance of difference from scrambled RNA oligomer control at p < .001. Abbreviations: CDS, coding sequence; CTL, control; Luc, luciferase; MRE, microRNA response element; Mut, mutant; nt, nucleotides; PmiR, Pre-microRNA microRNA precursor; RA, retinoic acid; RC, reverse complement; Scr, Scrambled RNA oligomer control; UTR, untranslated region; WT, wild-type.

Download figure to PowerPoint

miR-134 Downregulates the Endogenous Protein Levels of LRH1, Gαo, and Nanog

The effect of Pre-miR-134 on the endogenous protein levels of the four selected positive miR-134 targets in E14 mESCs was investigated (Fig. 5D, 5E). Transfection with Pre-miR-134 downregulated the endogenous protein levels of LRH1 (Fig. 5D, lanes 2 and 4), Gαo (lanes 10 and 12), and Nanog (lanes 14 and 16) in both the presence and the absence of RA. RA also downregulated the endogenous protein levels of LRH1 (Fig. 5D, lanes 1 and 3), Gαo (lanes 9 and 11), and Nanog (lanes 13 and 15), consistent with the induction of miR-134 by RA. Importantly, and for all three tested targets, protein downregulation was not associated with a concomitant downregulation in mRNA levels (Fig. 5F), suggesting that miR-134 acts, at least in part, by translationally inhibiting these mRNAs in E14 mESCs.

An intriguing finding was that miR-134 on its own led to an increase (+50%) in FADD mRNA levels (Fig. 5F), whereas FADD protein levels decreased (−80%) (Fig. 5D, lanes 5 and 6), suggesting post-transcriptional suppression of FADD mRNA by miR-134 in the absence of RA. RA induced a significant increase in FADD mRNA (+100%) and enhanced the upregulation of FADD mRNA by Pre-miR-134 (+250%) (Fig. 5F, columns 7 and 8). However, in the presence of RA, miR-134 lost its ability to suppress FADD protein levels as both FADD mRNA and protein levels increased (Fig. 5D, lanes 7 and 8). This highlights that miR-134 on its own can affect mRNA levels and that its effect on protein levels can be context-dependent. RA- or N2B27-mediated changes on other components of the post-transcriptional machinery [47] may alter the mechanism of action by which miR-134 affects FADD translation.

Knockdown of LRH1, Gαo, and Nanog Results in Differentiation of mESCs

To examine whether modulation of individual miRNA-targeted genes could directly promote mESC differentiation, FADD, LRH1, Gαo, and Nanog levels were perturbed by RNAi. RNAi was effective in significantly reducing transcript levels of LRH1, FADD, Gαo, and Nanog (Fig. 6A). Knockdown of Nanog, LRH1, and Gαo resulted in differentiation of mESCs as determined by the reduction in Oct4 promoter activity (Fig. 6B). To compare the effects of miR-134 and target gene RNAi on mESC differentiation, transcripts associated with pluripotency (Oct4 and Sox2) and differentiation (Nestin, Otx2, Fgf5, Sox17, and Bmp4; Fig. 6C) were examined. RNAi against LRH1, Gαo, and Nanog decreased Oct4 and Sox2, with concomitant elevation of ectoderm-associated Fgf5, but had variable effects on Nestin, Otx2, Sox17, and Bmp4 (Fig. 6C).

thumbnail image

Figure Figure 6.. Knockdown of Nanog, LRH1, and Gαo results in differentiation of mESCs. (A): Graph showing that transfection of gene-specific RNAi in mESCs resulted in a reduction in target mRNAs relative to NonSil shRNA CTL. No reduction was observed in target mRNAs with vector-only transfection. The expression of each gene was normalized to internal β-actin and expressed as the relative percentage change to its respective gene in the CTL. Experiments were performed in triplicate three times (n = 9), where error bars denote SE, and asterisk (∗) denotes significance of difference from NonSil vector CTL at p < .0001. (B): Graph showing changes in Oct4 promoter activity in E14 mESCs induced by RNAi to LRH1, FADD, Gαo, Nanog, and Oct4, relative to the NonSil shRNA CTL. RNAi to LRH1, o, Nanog, and Oct4 elicited reductions in Oct4 promoter activity. To correct for transfection efficiency, firefly luciferase activity was normalized to Renilla luciferase activity and expressed as percentage change relative to the CTL. Experiments were performed in quadruplicate three times (n = 12), where error bars denote SE, and asterisk (∗) denotes significance of difference from NonSil vector CTL at p < .0001. (C): Graph showing transcript changes in pluripotency and differentiation markers induced by PmiR-134 and RNAi against LRH1, FADD, Gαo, and Nanog relative to NonSil shRNA CTL. Experiments were performed in triplicate three times (n = 9), where error bars denote SE, and asterisk (∗) denotes significance of difference from NonSil vector CTL at p ≤ .001. Abbreviations: CTL, control; NonSil, nonsilencing; PmiR, Pre-microRNA microRNA precursor; RNAi, RNA interference; shRNA, short hairpin RNA.

Download figure to PowerPoint

Expression Profiling of miR-134 in Embryos and Adult Tissues

The effect of miR-134 on mESCs, a model for studying early development in vitro, led to an investigation of its expression profile during early in vivo mouse development. Quantitative RT-PCR was used to examine the temporal and spatial expression patterns of miR-134 in mouse embryos and adult tissues (Fig. 7A). During mouse embryogenesis, miR-134 expression was detected in embryonic day (E) 7.5 whole embryos at relatively low levels. This increased between E7.5 and E15.5 and decreased toward E17.5. In the adult (postnatal day 60) animal, miR-134 expression was notably enriched in brain and spinal cord (as with miR-124a), although miR-134 was also observed in additional tissues, including the ovary and stomach.

thumbnail image

Figure Figure 7.. miR-134 expression increases during embryogenesis and is present in several adult tissues. (A): MicroRNA (miRNA) quantitative PCR analysis demonstrating that miR-134 expression increases during mouse embryogenesis and is present in various adult mouse tissues, notably the brain and spinal cord. miR-16 expression was used as a loading ctrl. (B,C): miRNA quantitative PCR analysis showing miR-134 expression in various parts of the embryonic (B) and adult (C) brain. miR-16 expression was used as a loading ctrl. (D): Whole mount and section in situ hybridization depicting miR-134, miR-124a, and β-actin expression during mouse embryogenesis. Diffuse miR-134 expression was observed throughout the E11.5 embryo, with significant enrichment in the central and peripheral nervous systems. Abbreviations: ctrl, control; Hb, hindbrain; miR, microRNA; T, ventral telencephalon/midbrain.

Download figure to PowerPoint

Spatial distribution of miR-134 at E12.5 indicated that miR-134 levels were highest in the midbrain compared with the cortex, hindbrain, and hippocampus (Fig. 7B). Levels of brain-specific miR-124a were lower than those of miR-134 in the embryonic brain. Interestingly, this pattern was reversed in the adult brain, where miR-124a levels were higher in all brain regions relative to miR-134, although the latter's levels remained comparably higher in the midbrain (Fig. 7C). miR-134 and miR-124a distribution were also examined by in situ hybridization (Fig. 7D). A diffuse miR-134 expression was observed throughout the E11.5 embryo, although there was significant enrichment in caudal and ventral brain regions. This was consistent with quantitative RT-PCR measurements showing significant miR-134 expression in hindbrain and the ventral telencephalon/midbrain relative to miR-124a at E12.5.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosure of Potential Conflicts of Interest
  8. Acknowledgements
  9. References
  10. Supporting Information

By combining miRNA identification and expression, functional analyses, and validation of miRNA-mRNA heteroduplexes, we present data that describe the role of miR-134 in enhancing embryonic stem cell differentiation in vitro to ectodermal lineages. miR-134 was identified as one of many miRNAs upregulated during RA- and N2B27-induced ectodermal differentiation of mESCs. It was not upregulated during EB differentiation. Significantly, miR-134 was the only one of those miRNAs tested that on its own modulated differentiation of mESCs under the conditions used and that enhanced RA- and N2B27-induced transition to neuroectodermal lineages. We show that these effects are brought about, at least in part, by miR-134 downregulation of translation of three of ∼1,051 mESC-expressed predicted target mRNAs, Nanog, LRH1, and Gαo, through heteroduplex formation with 3′UTR cis-elements.

Interestingly, our data showed that miR-134 was not upregulated during differentiation of EBs. As EB formation results in a random differentiation of ESCs to endodermal, mesodermal, and ectodermal lineages, the heterogeneity of cell types present in EBs may explain why we do not see an overall increase in miR-134 levels, even though its expression may increase in the population of cells differentiating toward ectoderm. The data suggest that upregulation of miR-134 and its effects on target mRNAs are not absolute requirements for mESC differentiation per se but highlight its role in promoting ectodermal lineage differentiation that we observed (Figs. 2, 4). Indeed, Anti-miR-134 was able to attenuate the RA-induced changes in Sox1, Nestin, and Neurogenin2 mRNA levels but could not attenuate the Oct4 mRNA downregulation. Again, this suggests that miR-134 elevation by RA and N2B27 is not an absolute requirement for differentiation, but it enhances the RA- and N2B27-induced differentiation and aids the specification to ectodermal lineages.

Expression profiling demonstrated that miR-134 levels were notably higher in the adult brain and spinal cord relative to other tissues. The peak of miR-134 expression during embryogenesis between E13.5 and E17.5 correlates with the progressive expansion and stratification of the neuroepithelium (a result of neuronal proliferation, migration and differentiation) and a transition from neuro- to gliogenesis. We also observed a dichotomy between the expression of miR-124a and miR-134 during nervous system development, where miR-134 is higher relative to miR-124a in embryonic, but not adult, tissues. miR-124a expression is correlated with the transition from non-nervous system to nervous system-specific alternative splicing patterns by inhibiting the splicing suppressor PTBP1 [39]. PTBP1 controls the splicing of the pre-mRNA of its homolog PTBP2 and vice versa [39]. Interestingly, PTBP2 is a predicted target of miR-134 (supplemental online Fig. 3C). One might speculate that the relative levels of miR-124a and miR-134 expression may alter specific splicing patterns in a developmental stage-specific manner. In any case, these data indicate a developmental role for miR-134 prior to its role in dendritic spine formation in terminally differentiated neurons in vitro [23].

It has been proposed that miRNAs function in vertebrates by subtly modulating cell types [48], and most of our augmentation and reduction of miRNA expression experiments did not show any significant phenotype change in mESCs, apart from experiments with miR-134. This may be linked to the number of potential mRNAs with which an miRNA can form heteroduplexes. The algorithm for calculating miRNA-mRNA interactions, rna22, predicted 2,800 mRNA 3′UTR cis-elements in the mouse genome as targets for forming heteroduplexes with miR-134, of which we confirmed 129 of 158 as positive (>80%) [26]. More than 1,000 genes expressed in mESCs comprised miR-134 potential targets. This implies that the mechanism by which miR-134 post-transcriptionally regulates mRNA to enhance mESC differentiation is complex. Interestingly, RNAi of any single target of miR-134 was not sufficient to replicate the Pre-miR-134 induced changes in mESC transcript levels, again suggestive of a wide-ranging miR-134 target network. Together, these data indicate that RA-induced elevation of miR-134, or exogenously elevated levels of miR-134 alone, may promote mESC differentiation through the coordinate regulation of a potentially large target-gene pool, which includes the pluripotency-associated target genes, such as Nanog, LRH1, and o. Interestingly, the sequences of the predicted miR-134 target sites in LRH1 and Gαo are conserved in mouse, rat, and human and in mouse, rat, human, dog, and chicken, respectively.

As stated previously, miR-134 suppresses the protein levels of Nanog and LRH1. Nanog and LRH1 are known regulators of mESC self-renewal, where their interaction with specific promoter regions is integral to the transcriptional regulation of many genes, including Oct4 [32]. miR-134 overexpression resulted in the reduction of three other endogenous proteins tested to date, in part, through translation repression, an effect that was selectively blocked by Anti-miR-134. These genes, Wnt15, Gαo, and FADD, have all been shown to play key roles in development [49, [50]51]. Although Nanog mRNA levels remained unchanged in E14 mESCs during Pre-miR-134 treatment over 3 days but decreased in D3 mESCs, this may reflect subtle differences across cell types or that D3 mESCs may be more sensitive to elevated levels of Pre-miR-134 than E14 ESCs. Intriguingly, we observed that miR-134 on its own led to an increase in FADD mRNA levels, whereas FADD protein levels decreased. RA induced a significant increase in FADD mRNA and enhanced the effect of Pre-miR miRNA precursor 134, with concomitant increase in FADD protein. This suggests that the suppressive effect of miR-134 on FADD mRNA translation is lost in the presence of RA, and this highlights that the effect of an miRNA on a target or nontarget mRNA may be context-dependent. Our array profile data show that there was upregulation, no change, or downregulation of target and nontarget mRNAs, which again highlights that miR-134 may act through multiple gene-regulatory mechanisms.

In conclusion, elevated levels of miR-134 alone can enhance differentiation of mESCs, where this promiscuous miRNA exerts its effects through post-transcriptional regulation of multiple mRNAs, including translation attenuation of Nanog and LRH1. Elevation of miR-134 enhances the ectodermal differentiation of mESCs treated with RA or N2B27, where it augments the RA- and N2B27-induced elevation of ectodermal markers (β-III-tubulin, Nestin, Neurogenin2, and Sox1). These findings will aid in our understanding of miRNA function in embryonic stem cells and facilitate our ability to inhibit or enhance the differentiation of ectodermal lineages.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosure of Potential Conflicts of Interest
  8. Acknowledgements
  9. References
  10. Supporting Information

We thank Lee Yin Loon and Karrie Ko for technical and administrative support, and we are grateful to Wayne Mitchell, Koo Sumin, and Yeap Leng Siew for useful discussions. This work was supported by the Agency for Science, Technology and Research, Singapore. Y.M.-S.T., W.-L.T., and Y.-S.A. contributed equally to this work. I.R., A.M.T., and B.L. are co-senior authors. R.J.P. is currently affiliated with the Mercer University School of Medicine, Savannah, GA.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosure of Potential Conflicts of Interest
  8. Acknowledgements
  9. References
  10. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosure of Potential Conflicts of Interest
  8. Acknowledgements
  9. References
  10. Supporting Information
FilenameFormatSizeDescription
07-0295_SFig1.pdf199KSupplemental Figure 1
07-0295_SFig2.pdf143KSupplemental Figure 2
07-0295_SFig3.pdf149KSupplemental Figure 3
07-0295_Stable1.pdf83KSupplemental Table
07-0295_Supp_meth.pdf111KSupplemental Methods
07-0295Supp_Leg.pdf20KSupplemental Legends

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.