Loss of non-coding RNA expression from the DLK1-DIO3 imprinted locus correlates with reduced neural differentiation potential in human embryonic stem cell lines
© Mo et al.; licensee BioMed Central. 2015
Received: 28 April 2014
Accepted: 15 December 2014
Published: 5 January 2015
Pluripotent stem cells are increasingly used to build therapeutic models, including the transplantation of neural progenitors derived from human embryonic stem cells (hESCs). Recently, long non-coding RNAs (lncRNAs), including delta-like homolog 1 gene and the type III iodothyronine deiodinase gene (DLK1-DIO3) imprinted locus-derived maternally expressed gene 3 (MEG3), were found to be expressed during neural development. The deregulation of these lncRNAs is associated with various neurological diseases. The imprinted locus DLK1-DIO3 encodes abundant non-coding RNAs (ncRNAs) that are regulated by differential methylation of the locus. We aim to study the correlation between the DLK1-DIO3-derived ncRNAs and the capacity of hESCs to differentiate into neural lineages.
We classified hESC sublines into MEG3-ON and MEG3-OFF based on the expression levels of MEG3 and its downstream microRNAs as detected by quantitative reverse transcription-polymerase chain reaction (qRT-PCR). A cDNA microarray was used to analyze the gene expression profiles of hESCs. To investigate the capacity of neural differentiation in MEG3-ON and MEG3-OFF hESCs, we performed neural lineage differentiation followed by neural lineage marker expression and neurite formation analyses via qRT-PCR and immunocytochemistry, respectively. MEG3-knockdown via small interfering RNA (siRNA) and small hairpin RNA (shRNA) was used to investigate the potential causative effect of MEG3 in regulating neural lineage-related gene expression.
DLK1-DIO3-derived ncRNAs were repressed in MEG3-OFF hESCs compared with those in the MEG3-ON hESCs. The transcriptome profile indicated that many genes related to nervous system development and neural-type tumors were differentially expressed in MEG3-OFF hESCs. Three independent MEG3-knockdown assays using different siRNA and shRNA constructs consistently resulted in downregulation of some neural lineage genes. Lower expression levels of stage-specific neural lineage markers and reduced neurite formation were observed in neural lineage-like cells derived from MEG3-OFF-associated hESCs compared with those in the MEG3-ON groups at the same time points after differentiation.
Repression of ncRNAs derived from the DLK1-DIO3 imprinted locus is associated with reduced neural lineage differentiation potential in hESCs.
Pluripotent stem cells are increasingly used for therapeutic models, including the transplantation of neural progenitors derived from human embryonic stem cells (hESCs) . Pluripotent stem cells represent a potential therapy for neurological disorders . Improving the effectiveness of hESC-derived neural precursor transplantation therapies has been widely studied . However, teratoma or tumor formation after transplantation remains a major safety concern . Long non-coding RNAs (lncRNAs), defined as longer than 200 nucleotides in length, are suggested to be involved in neural developmental events [5, 6] and in neurological degenerative diseases [7, 8]. Thus, lncRNAs may play important roles in deriving neural lineage cells from hESCs.
Recently, abundant expression of an lncRNA, maternally expressed gene 3 (Meg3), which is also called gene trap locus 2 (Gtl2), was detected in the mouse nervous system. Meg3 is expressed in the forebrain in both developing and adult mice  as well as in developing corticospinal neurons . This lncRNA is derived from the delta-like homolog 1 gene and the type III iodothyronine deiodinase gene (Dlk1-Dio3) imprinted locus, which expresses many non-coding RNAs (ncRNAs) from the maternal chromosome, including the lncRNAs Meg3/MEG3, Rian/MEG8, and Mirg; one C/D box small nucleolar RNA gene cluster; and microRNA (miRNA) clusters. In contrast, three protein-coding genes, Dlk1, Rtl1, and Dio3, are expressed from the paternal chromosome. The correct expression dosages of these ncRNAs and genes are regulated by the two differentially methylated regions of the locus, including the intergenic differentially methylated region (IG-DMR) and the Meg3/MEG3 DMR [11–13]. Differential methylation of this imprinted locus is required for maintaining the full developmental potential of mouse-induced pluripotent stem cells (miPSCs) [14–16].
Humans possessing imprinting defects in the DLK1-DIO3 locus suffer from skeletal malformations, developmental delay/mental retardation, tumor development, and even postnatal death [17–24]. Moreover, ncRNAs derived from the DLK1-DIO3 locus are associated with neurodevelopmental and neurodegenerative disorders. For example, more than 85% of DLK1-DIO3 locus-derived miRNAs are downregulated in patients with schizophrenia , and decreased lncRNA MEG3 expression occurs in patients with Huntington’s disease . Recently, DLK1-DIO3 locus-derived lncRNAs were suggested to associate with polycomb repressive complex 2 (PRC2) and to affect genome-wide PRC2 targets in trans in mouse embryonic stem cells (mESCs) and human induced pluripotent stem cells (hiPSCs) . PRC2 introduces the specific repressive histone marker H3K27me3 to target regions and suppresses gene expression [27, 28]. Most PRC2 target genes are known for their important roles in developmental processes .
To study the correlations between the DLK1-DIO3-derived ncRNAs and the differentiation capacity of hESCs toward neural lineages, we classified hESCs into MEG3-ON and MEG3-OFF on the basis of the expression of MEG3 and its downstream miRNAs. We found that using hESCs with repressed DLK1-DIO3-derived ncRNA expression for neural lineage differentiation produced reduced neural lineage marker expression at different stages and reduced neurite formation.
Materials and methods
This study included the use of human pluripotent stem cells. The derivation of these cell lines and their use in this particular study were approved by the ethics committee of National Taiwan University Hospital and the Internal Research Board of Academia Sinica. Written informed consent was obtained from all subjects involved in this study.
Human embryonic stem cell culture
The NTU1, NTU3, and H9 hESC lines [29–31] and hiPSC lines [32, 33] were used in this study. These pluripotent stem cells were maintained on murine embryonic fibroblast (MEF) feeders by using serum-free medium (ReproCELL primate and human ESC culture medium; ReproCELL, Kanagawa, Japan). The hESCs were passaged once per week by using a 30-gauge insulin needle or using dispase and collagenase type IV (Gibco, part of Invitrogen, Carlsbad, CA, USA). At 48 hours after passaging, hESC colonies were observed. The medium was refreshed every day. For embryoid body (EB) formation, the hESCs were cultured in suspension in Petri dishes for 5 or 12 days, in accordance with a previously published protocol . On days 5 and 12, EBs were collected for quantitative reverse transcription-polymerase chain reaction (qRT-PCR). The diameters of EBs were quantified by using QCapture Pro 6.0 software (QImaging, Surrey, BC, Canada).
Quantitative reverse transcription-polymerase chain reaction for long non-coding RNA and mRNA
qRT-PCR was used to quantify relative mRNA levels in the undifferentiated hESCs, EBs, and neural lineage-like cells. Briefly, at a predetermined time point, cells were collected and treated with TRIzol® (Invitrogen) to extract RNA, and samples of cDNA were obtained by using random hexamer primers following the Superscript III kit protocol (Invitrogen). We detected the expression levels of MEG3 and MEG8 in hESCs by using the SYBR Green PCR master mix (Kapa Biosystems, Wilmington, MA, USA). PCR was performed in a thermal cycler (LightCycler® 480 II Instrument; Roche, Basel, Switzerland) by using the following program: 50°C for 2 minutes, 95°C for 10 minutes, and 45 cycles of denaturation at 95°C for 15 seconds and annealing and extension at 60°C for 45 seconds. The quantitation of the endoderm, mesoderm, and ectoderm layer-specific transcripts, SOX17, HAND1, and PAX6, respectively, were measured during EB formation by qRT-PCR. The expression levels of neural lineage differentiation-associated genes, which included PAX6, RTN1, DLK1, SOX11, beta-III TUBULIN, and MAP2, were detected in MEG3-ON and MEG3-OFF hESCs and their differentiated neural lineage-like cells by using the SYBR Green PCR master mix (Kapa Biosystems). The housekeeping gene GAPDH was used as a normalization control. The 2−ΔΔCp method was used to quantify the qRT-PCR results. The primer sequences are listed in Additional file 1: Table S1. qRT-PCR results demonstrating consistently detectable and specific signals were presented in the bar charts and subject to statistical analysis. N.D. (not detectable) indicates that the expression of a particular gene in certain samples, if any, was below the sensitivity threshold of qRT-PCR analysis.
Quantitative reverse transcription-polymerase chain reaction for microRNA
The UPL probe system (Roche) was used to detect the expression of miRNAs, including miR-127-3p, miR-376c, miR-494, miR-495, miR-496, and miR-154. The miRNA detection protocol was based on a previous protocol . PCR was performed in a thermal cycler (LightCycler® 480 II Instrument; Roche). Briefly, a sequence-specific RT primer was used for cDNA synthesis of a specific miRNA in the RT step. Next, a sequence-specific forward primer, a universal reverse primer, and UPL probe 21 were used for amplifying the cDNA. RNU48 was used as an internal control to normalize the miRNA expression levels. The 2−ΔΔCp method was used to quantify the qRT-PCR results. The primer sequences are listed in Additional file 1: Table S1.
We collected and classified NTU1 hESCs into four MEG3-ON sublines (NTU 1–1, 1–2, 1–3 and 1–4) and five MEG3-OFF sublines (NTU 1–5, 1–6, 1–7, 1–8, and 1–9) on the basis of the MEG3 expression levels determined by qRT-PCR. We also validated the methylation patterns of the DLK1-DIO3 locus by bisulfite sequencing. H9 hESCs and EBs were used as positive and negative controls of pluripotency, respectively, for the PluriTest assay but were not used for identifying genes with differential expression between the MEG3-ON and MEG3-OFF groups.
RNA isolation and hybridization
Total RNA was extracted by using TRIzol® reagent (Invitrogen) and purified by using RNeasy® mini kits (Qiagen, Venlo, Limburg, The Netherlands) to obtain 2 μg of RNA. Concentration and quality were assessed by using an electrophoresis bioanalyzer instrument (Agilent Technologies, Santa Clara, CA, USA). The HumanHT-12 version 4.0 Expression BeadChip (Illumina, San Diego, CA, USA) was used on the basis of its up-to-date content derived from the National Center for Biotechnology Information Reference Sequence (NCBI RefSeq) database (Build 36.2, Release 22). Biotin-labeled cRNA prepared from 1.5 μg of RNA from each sample was fragmented, hybridized to individual HT-12 version 4.0 Expression BeadChips by using the direct hybridization assay protocol, and scanned by using a BeadArray Reader (Illumina).
Identification of differentially expressed genes
Raw idat microarray files were read into Partek Genomics Suite version 6.5 (Partek, St. Louis, MO, USA) to generate log transformation and quantile normalization. Quality assessment for each array data was also validated by using Partek or R through data distribution and principal component analysis of samples. In total, 47,319 of the 48,206 probe sets were mapped to 21,777 genes, and 14,235 genes had more than one representative probe set. For each redundancy, the probe set with the greatest average expression across all samples was chosen to represent each gene. To identify the most significantly differentially expressed genes (DEGs) between MEG3-ON and MEG3-OFF hESCs, we selected only the DEGs identified by the following criterion: (i) genes with more than 1.5 times fold change between MEG3-ON and MEG3-OFF and with significant differential expression value (P <0.05) based on Welch’s t test (307 genes). (ii) We used the same package in the Bioconductor software package  to perform significance analysis of microarray (SAM)  analysis for comparisons between two groups (MEG3-ON and MEG3-OFF) with 1,000 permutations. In total, 314 significant genes with false discovery rate (FDR) of less than 0.05 and d of 1.68 were selected.
On the basis of these criteria, among the two gene lists from Welch’s t test (307 genes) and SAM (314 genes), we obtained 114 intersection DEGs for further biological and functional analysis. A hierarchical clustering heatmap of the 114 genes was generated by using ggplot2  packages in the R platform and Euclidean distance and average linkage.
Gene Ontology analysis, gene set enrichment assay, and ‘PluriTest’
We used MetaCore™ software (version 6.18, build 65505) from GeneGo (part of Thomson Reuters, New York, NY, USA) to investigate whether the genes showing differential expression between MEG3-ON and MEG3-OFF hESCs are involved in particular biological processes or diseases. The MetaCore Gene Ontology (GO) biological process or GO disease analysis used the hypergeometric test to select GO biological process gene sets or disease biomarkers enriched in genes that displayed significant differential expression. The P value was calculated by using the right-tailed Fisher’s exact test based on the hypergeometric distribution. The corrected P values were obtained by FDR calculation based on the Benjamini-Hochberg method  of accounting for multiple testing. A background gene list was collected from all 21,777 mapped microarray genes.
Gene Set Enrichment Analysis (GSEA) [39, 40] was performed to determine whether the gene expression levels annotated by the GO term (GO:0007399) associated with nervous system development were specifically enriched in the MEG3-ON (NTU 1–1, 1–2, 1–3, and 1–4) or MEG3-OFF (NTU 1–5, 1–6, 1–7, 1–8, and 1–9) hESCs; gene sets were identified in which these genes were enriched, together with an associated P value calculated by permutation testing. Partial genes from the gene sets were listed in the results. FDR and familywise-error rate (FWER) statistical calculations were performed as described by Subramanian et al.. Moreover, the hESC gene sets  were adopted to test whether the gene sets associated with ESCs could be enriched in the differential expression profiles of the MEG3-ON and MEG3-OFF groups.
Furthermore, a separate ‘PluriTest’  was applied to evaluate whether the samples possessed or lost pluripotent features by using machine-learning methods. The pluripotency score is a logistic regression model that enables a probability-based choice between the pluripotent and non-pluripotent phenotypic classes based on microarray data. Overall, 12 microarray chips were analyzed in this test—four MEG3-ON hESCs (NTU 1–1, 1–2, 1–3, and 1–4), seven MEG3-OFF hESCs (NTU 1–5, 1–6, 1–7, 1–8, and 1–9), and two H9 hESCs as the pluripotent controls—and one EB was the pluripotency negative control for the PluriTest. The data were deposited in the NCBI Gene Expression Omnibus (GEO) and are accessible through GEO Series accession number GSE58809.
Neural lineage differentiation
hESC maintenance and EB formation were described above in the ‘Human embryonic stem cell culture’ section. For neuroectodermal sphere (NES) formation, 4-day-old EBs were cultured in N2 medium containing Dulbecco’s modified Eagle’s medium/F12, NEAA (1X), L-glutamine (2 mM), N2 supplement (1X), sodium pyruvate (1 mM), and basic fibroblast growth factor (20 ng/mL) (Gibco) for 3 days, followed by Matrigel attachment for 3 or 18 days. Half of the N2 medium was refreshed every 48 hours.
Cell differentiation was performed in eight-well chamber slides (Millicell EZ slide; Millipore, Billerica, MA, USA). Cells were fixed with 4% paraformaldehyde for 20 minutes and blocked with 2% bovine serum albumin (BSA)/0.1% Triton X-100 in 1X phosphate-buffered saline (PBS) for 5 to 10 minutes. The cells were incubated with primary antibodies at 4°C for 16 hours in PBS containing 2% BSA. The cells were washed and then incubated with fluorescent secondary antibodies (715-485-150 and 711-485-152; Jackson ImmunoResearch, West Grove, PA, USA) for 1 hour. Immunocytochemistry was also performed by using anti-β-III tubulin (MAB1637; Millipore), anti-MAP2 (GTX111679; GeneTex, Irvine, CA, USA), and anti-Ki67 (ab15580; Abcam, Cambridge, UK) antibodies.
Bisulfite modification was performed by using an EZ DNA Methylation Kit (Zymo Research, Irvine, CA, USA) with genomic DNA extracted from hESCs of different passage numbers. PCR amplification was initiated at 95°C for 5 minutes, followed by 40 cycles of 95°C for 30 seconds, 54°C for 45 seconds, and 72°C for 30 seconds. The following primer sequences were adopted from previous reports: H19 and KvDMR [43, 44], IG-DMR and MEG3 DMR , and PEG10 DMR . The amplified products were cloned into the pGEM-T Easy vector (Promega, Madison, WI, USA), and 20 clones from each genomic sample were picked for sequencing. The sequences were analyzed by BiQ Analyzer software , and only non-clonal sequences are presented.
Infection of human embryonic stem cells with small interfering RNA and small hairpin RNA lentiviruses
Two knockdown construct types were used in this study. (i) Two MEG3-siRNA plasmids with different target sequences and scramble siRNA-GFP plasmids (Abcam Company). The MEG3-siRNA plasmids and their target sequences for MEG3 are as follows: MEG3-451 siRNA-GFP: tgtgttcacctgctagcaaactggagtgt; MEG3-512-siRNA-GFP: actgactctgtcatcacccttatgatgtc. (ii) MEG3-shRNA plasmid with the other target sequence and the scrambled shRNA plasmid (OriGene TR30021). The shRNA plasmid and its target sequences for MEG3 are as follows: TL 320132C MEG3: gagaggttgtttcactggtatctattgca.
Packaging, envelope, and siRNA or shRNA plasmids were transfected into 293 T cells to produce lentiviral particles. Harvested media containing lentivirus were concentrated and used to infect hESCs. The following two infection methods were used in these experiments: (i) Harvested media with lentivirus-containing siRNA plasmid were used to infect hESC clumps (50 to 250 cells per clump). (ii) Harvested media with lentivirus-containing shRNA plasmid were used for single-cell infection.
The infected hESCs were cultured on feeder cells for 4 days and switched to ESC medium containing puromycin (1 μg/mL) for selection. Antibiotic-selected and GFP-positive hESCs were amplified and harvested for qRT-PCR detection.
In most of the experiments using the NTU1 and NTU3 hESC lines, the error bars represent the standard error of the mean (SEM) generated from three biological samples with three technical repeats each. In the experiments using the H9 hESC and hiPSC lines, the error bars represent the SEM generated from one biological sample with three technical repeats. In the MEG3-shRNA knockdown experiments, the error bars represent the SEM generated from one biological sample, with three technical repeats each. Student’s t test was used for calculating P values. *P <0.05 or **P <0.01 was considered to be significant. In the MEG3-siRNA knockdown experiments, the error bars represent the SEM generated from one biological sample with three technical repeats. For multiple-group analysis in the MEG3-siRNA knockdown experiments, we used one-way analysis of variance followed by Dunnett’s multiple comparisons test and statistics via GraphPad Prism 6 software (GraphPad Software, Inc., La Jolla, CA, USA). Significance was assigned at *P <0.05 and **P <0.01. The data analysis for the arrays is described in the ‘Microarray analysis’ subsection.
MEG3-OFF human embryonic stem cells have significantly repressed expression of DLK1-DIO3-derived long non-coding RNAs and microRNAs
Embryoid bodies derived from MEG3-OFF human embryonic stem cells showed unusual morphologies and decreased expression of the ectodermal and neural stem cell marker PAX6
Moreover, specific markers from the three embryonic germ layers were differentially expressed in 12-day-old EBs (Figure 2B). In MEG3-OFF EBs, the endoderm-specific marker SOX17 and the mesoderm-specific marker HAND1 were overexpressed, whereas the ectoderm-specific marker PAX6, which is also a neural stem cell marker, was only weakly expressed, which may reduce the neural differentiation potential or other differentiation abnormality from MEG3-OFF hESC sublines.
Undifferentiated MEG3-OFF human embryonic stem cells (hESCs) displayed a different transcriptome profile from MEG3-ON hESCs, particularly in the neural lineage-related gene subset
Intriguingly, the DLK1 gene, which is a paternally expressed gene from the same imprinted locus, also showed decreased expression in pluripotent stem cells with repressed MEG3 expression, in contrast to the expected reciprocal regulation of these two genes (Figure 4A). A DLK1 function related to neural lineage differentiation was recently observed in mouse and human ESCs . Notably, DLK1 was also downregulated in two out of three independent MEG3 knockdown lines: one si-MEG3 line and one sh-MEG3 line (Figure 4B). When the two si-MEG3-treated hESC lines were compared, the cell line associated with reduced DLK1 expression was shown to be associated with obvious further PAX6 downregulation, suggesting that DLK1 deregulation may be one of the reasons for the reduced expression of these neural lineage-promoting genes (PAX6 and RTN1) in undifferentiated hESCs.
MEG3 may be directly involved in regulating PRC2 target genes , which are often developmentally regulated. Some of the neural progenitor markers, PAX6, CXCR4, and SOX21, which were differentially expressed between MEG3-ON and MEG3-OFF NTU1 hESCs, were also PRC2 targets subject to EZH2- and SUZ12-mediated H3K27me3 modification [50, 51] (Additional file 5: Figure S3). Therefore, MEG3 might regulate these neural-progenitor-related genes through a PRC2-mediated mechanism in hESCs.
MEG3-OFF human embryonic stem cell-derived neural lineage-like cells showed lower expression levels of neural lineage markers and reduced neurite formation
Selecting high-quality hESC lines has been one of the key issues for stem cell-associated studies and therapies. Imprinting instability has long been a problem for medical applications of pluripotent stem cells, including developmental potential and the risk of cancer formation. The maintenance of differential methylation of the DMRs of imprinted loci is required for the correct dosage of imprinted gene expression, which is suggested to be crucial for ESC function [54, 55]. Accumulating evidence suggests that proper imprinting marks on the mouse Dlk1-Dio3 locus are important for the biological functions of miPSCs. Aberrant hypermethylation of the Dlk1-Dio3 locus has been related to the impaired developmental potential of miPSCs. According to the reports by Stadtfeld et al.[15, 16] and by Liu et al., Meg3 off miPSCs with an aberrantly hypermethylated Dlk1-Dio3 locus and repressed miRNAs could not achieve full developmental potential for generating full-term mice in a tetraploid complementation assay [14–16]. Ascorbic acid (vitamin C) has been shown to preserve a stable epigenetic state at the mouse Dlk1-Dio3 locus during somatic cell reprogramming for producing miPSC lines, thereby increasing the success rate of embryos reaching full-term development after tetraploid complementation . These studies indicated that ncRNAs derived from the Dlk1-Dio3 locus are required for developmental potential in mouse pluripotent stem cells. Moreover, a study by Stadtfeld et al. demonstrated that the head tissues of Meg3 off embryos derived from a tetraploid complementation assay with Meg3 off miPSCs displayed dramatic repression of three important neural lineage genes, Pax6, Mash-1, and Hes-5. Their data suggested a possible correlation between the ncRNAs derived from the DLK1-DIO3 imprinted locus and the expression levels of neural lineage genes. In the present study, we revealed a positive association between the activities of the DLK1-DIO3 imprinted locus and the differentiation capacity of hESCs for neural lineages on the basis of the following observations: (i) The MEG3-OFF hESCs had reduced potential for differentiating into normal EBs, as evidenced by impaired morphologies and by reduced expression of the ectodermal marker PAX6. (ii) Many genes involved in nervous system development were differentially expressed between MEG3-ON and MEG3-OFF undifferentiated hESCs. (iii) Neural lineage-like cells derived from MEG3-OFF hESCs showed downregulation of neural lineage markers and reduced neurite formation compared with the MEG3-ON groups.
None of the above-mentioned studies can exclude the possibility that unidentified genomic regions other than DLK1-DIO3 imprinted locus also acquire aberrant epigenetic marks during iPSC reprogramming or in vitro ESC culture, which accumulatively or independently cause the functional deficiency of these MEG3-OFF-associated pluripotent stem cells. Our MEG3 knockdown experiments induced downregulation of neural lineage markers, suggesting a certain level of potential causative effect between MEG3 expression level and the developmental potential of hESCs.
MEG3 and DLK1 have been observed to be reciprocally regulated in mouse embryos . However, how these two genes are regulated in mouse and human pluripotent stem cells is not fully understood. Even in mESC lines in which Meg3 is highly expressed, the expression patterns of Dlk1 are extremely variable . Based on array data, both Meg3 and Dlk1 were shown to be abundantly expressed in one mESC line; however, in three other mESC lines with Meg3 expression, the Dlk1 expression level was shown to be lower . Moreover, Dlk1 expression has also been shown to be low in Meg3 off miPSC lines . In regard to the expression of DLK1 in hESC lines, the array data from another study demonstrated that it was downregulated in most of the tested MEG3-OFF hESC lines . These data imply that the cell type-specific and stage-specific developmental regulation may be the primary factor that controls the net expression level of DLK1 in undifferentiated hESCs. Kaneko et al. did not observe reciprocal correlation of MEG3 and DLK1 expression levels between undifferentiated MEG3 + and MEG3 − hiPSCs identified in their study. They also did not find the differential PRC2 binding on the DLK1 promoter when overexpressing MEG3 in hiPSCs . A DLK1 function related to neural lineage differentiation was recently observed in mESCs and hESCs. DLK1-overexpressing mESCs and hESCs have been shown to promote increased neurogenesis in mESC-derived neural progenitors and hESC-derived neural progenitors, respectively . Interestingly, in our study, the MEG3-ON hESCs with increased capacity for neural lineage differentiation displayed relatively higher DLK1 expression levels compared with the MEG3-OFF hESCs. Our study is the first to link these associations between MEG3 and DLK1 expression with the capacity for neural lineage differentiation in hESCs.
Although MEG3 and DLK1 are reciprocally regulated in E16.6 full mouse embryos by parental origin-specific methylation of their imprinting control region , in many mouse pluripotent stem cell lines and hESC lines, including the cells lines in our study, lower MEG3 expression is surprisingly associated with lower expression of DLK1. Because imprinting instability is a frequent phenomenon in in vitro culture systems of mouse and human pluripotent stem cells, this finding raises the possibility that amplification of epigenetic abnormalities might be easily observed in in vitro culture conditions. We speculate that when maternally expressed ncRNAs derived from the DLK1-DIO3 locus are silenced, the deregulation of miRNAs from the same locus might affect other downstream factors, including certain epigenetic modifiers. Consistent with this hypothesis, we previously detected a dramatic increase in H3K9me3 and a decrease in H3K4me3 in total protein levels in NTU1 MEG3-OFF hESCs (data not shown). These results suggest that the global chromatin state in this cell type may be more condensed, which may further affect the deregulation of genes related to MEG3 expression levels found in our study, including the DLK1 gene.
The lncRNA MEG3 derived from the DLK1-DIO3 locus possesses tumor suppressor properties in several cancer types, many of which are associated with the nervous system, including pituitary tumors , neuroblastomas , meningiomas , and gliomas . MEG3 expression is significantly decreased in tumor tissues compared with adjacent normal tissues. MEG3 overexpression has been shown to promote cell apoptosis in glioma cell lines and inhibit cell proliferation in glioma and meningioma cell lines [21, 58]. We observed that many genes related to various cancer types, primarily those associated with the nervous system, were deregulated in the MEG3-OFF hESCs in which MEG3 and the other downstream ncRNAs were repressed. Therefore, avoiding the use of MEG3-OFF hESCs may reduce the potential risk of tumorigenesis from stem cell-based therapies.
Notably, the RTN1 gene, which is correlated with several neural lineage-type tumors, was dramatically downregulated in MEG3-OFF hESCs and in their neural lineage-like derivatives. The RTN1 gene encodes the RTN1C protein, which is a marker of neuronal differentiation and is known to be downregulated in patients with Alzheimer’s disease and Down syndrome . RTN1C can induce cell apoptosis in neuroectodermal tumors and may be a potential molecular target for therapy . Intriguingly, we observed that RTN1 was downregulated when MEG3 was knocked down, suggesting that MEG3 might participate in the regulation of this gene. Additionally, because MEG3 and RTN1C can induce apoptosis in cancer cells, determining whether RTN1C is regulated by MEG3 in cancer models may be worth further investigation.
For basic and preclinical research, choosing hESC lines with higher expression levels of DLK1-DIO3 imprinted locus-derived ncRNAs as starting materials, characterized by the expression of the lncRNA MEG3, may benefit studies of neural lineages. Additionally, screening for MEG3-expressing hESCs may potentially lower the risk of tumorigenesis in stem cell-based therapies.
bovine serum albumin
differentially expressed gene
- DLK1-DIO3 :
delta-like homolog 1 gene and the type III iodothyronine deiodinase gene
false discovery rate
Gene Expression Omnibus
Gene Set Enrichment Analysis
- Gtl2 :
gene trap locus 2
human embryonic stem cell
human induced pluripotent stem cell
intergenic differentially methylated region
long non-coding RNA
- MEG3 :
maternally expressed gene 3
mouse embryonic stem cell
mouse-induced pluripotent stem cell
polycomb repressive complex 2
quantitative reverse transcription-polymerase chain reaction
significance analysis of microarrays
standard error of the mean
small hairpin RNA
small interfering RNA.
This work was supported by the National Science Council (NSC-100-2321-B-002-074 and NSC 102-2321-B-002-031 for SPL and NSC 100-2321-B-002-072 and NSC 102-2321-B-002-030 for HFC) and the Ministry of Science and Technology (MOST 103-2321-B-002-099 for SPL and MOST 103-2321-B-002-098 for HFC). The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The authors would like to express their sincere gratitude to Profs. Anne Ferguson-Smith, Jérôme Cavaillé, Chung-Liang Chien, Drs. Petra Hajkova, Joanna Maldonado-Youngson, Neil Youngson, Hung-Fu Liao, and Ms. Ya-Ping Yen for their useful comments during this study and manuscript preparation. We thank Prof. Li-Yu Liu and Dr. Hervé Seitz for their insightful comments regarding the statistical analysis. We thank Ms. Ming-Shan Chiang and Chih-Yun Yu for their excellent technical support. Illumina microarrays were performed by the Yong Ling Biomedical Engineering Center, which is supported by National Taiwan University (http://homepage.ntu.edu.tw/~ntuylbme/geneservice.html).
- Erceg S, Ronaghi M, Stojkovic M: Human embryonic stem cell differentiation toward regional specific neural precursors. Stem Cells 2009, 27:78–87. 10.1634/stemcells.2008-0543View ArticlePubMed CentralPubMedGoogle Scholar
- Chicha L, Smith T, Guzman R: Stem cells for brain repair in neonatal hypoxia-ischemia. Childs Nerv Syst 2014, 30:37–46. 10.1007/s00381-013-2304-4View ArticlePubMedGoogle Scholar
- Yuan T, Liao W, Feng NH, Lou YL, Niu X, Zhang AJ, et al.: Human induced pluripotent stem cell-derived neural stem cells survive, migrate, differentiate, and improve neurologic function in a rat model of middle cerebral artery occlusion. Stem Cell Res Ther 2013, 4:73. 10.1186/scrt224View ArticlePubMed CentralPubMedGoogle Scholar
- Seminatore C, Polentes J, Ellman D, Kozubenko N, Itier V, Tine S, et al.: The postischemic environment differentially impacts teratoma or tumor formation after transplantation of human embryonic stem cell-derived neural progenitors. Stroke 2010, 41:153–9. 10.1161/STROKEAHA.109.563015View ArticlePubMedGoogle Scholar
- Mercer TR, Qureshi IA, Gokhan S, Dinger ME, Li G, Mattick JS, et al.: Long noncoding RNAs in neuronal-glial fate specification and oligodendrocyte lineage maturation. BMC Neurosci 2010, 11:14. 10.1186/1471-2202-11-14View ArticlePubMed CentralPubMedGoogle Scholar
- Tochitani S, Hayashizaki Y: Nkx2.2 antisense RNA overexpression enhanced oligodendrocytic differentiation. Biochem Biophys Res Commun 2008, 372:691–6. 10.1016/j.bbrc.2008.05.127View ArticlePubMedGoogle Scholar
- Johnson R: Long non-coding RNAs in Huntington’s disease neurodegeneration. Neurobiol Dis 2012, 46:245–54. 10.1016/j.nbd.2011.12.006View ArticlePubMedGoogle Scholar
- Airavaara M, Pletnikova O, Doyle ME, Zhang YE, Troncoso JC, Liu QR: Identification of novel GDNF isoforms and cis-antisense GDNFOS gene and their regulation in human middle temporal gyrus of Alzheimer disease. J Biol Chem 2011, 286:45093–102. 10.1074/jbc.M111.310250View ArticlePubMed CentralPubMedGoogle Scholar
- McLaughlin D, Vidaki M, Renieri E, Karagogeos D: Expression pattern of the maternally imprinted gene Gtl2 in the forebrain during embryonic development and adulthood. Gene Expr Patterns 2006, 6:394–9. 10.1016/j.modgep.2005.09.007View ArticlePubMedGoogle Scholar
- Qu C, Jiang T, Li Y, Wang X, Cao H, Xu H, et al.: Gene expression and IG-DMR hypomethylation of maternally expressed gene 3 in developing corticospinal neurons. Gene Expr Patterns 2013, 13:51–6. 10.1016/j.gep.2012.11.003View ArticlePubMedGoogle Scholar
- Lin SP, Youngson N, Takada S, Seitz H, Reik W, Paulsen M, et al.: Asymmetric regulation of imprinting on the maternal and paternal chromosomes at the Dlk1-Gtl2 imprinted cluster on mouse chromosome 12. Nat Genet 2003, 35:97–102. 10.1038/ng1233View ArticlePubMedGoogle Scholar
- Zhou Y, Cheunsuchon P, Nakayama Y, Lawlor MW, Zhong Y, Rice KA, et al.: Activation of paternally expressed genes and perinatal death caused by deletion of the Gtl2 gene. Development 2010, 137:2643–52. 10.1242/dev.045724View ArticlePubMed CentralPubMedGoogle Scholar
- Lin SP, Coan P, da Rocha ST, Seitz H, Cavaille J, Teng PW, et al.: Differential regulation of imprinting in the murine embryo and placenta by the Dlk1-Dio3 imprinting control region. Development 2007, 134:417–26. 10.1242/dev.02726View ArticlePubMedGoogle Scholar
- Liu L, Luo GZ, Yang W, Zhao X, Zheng Q, Lv Z, et al.: Activation of the imprinted Dlk1-Dio3 region correlates with pluripotency levels of mouse stem cells. J Biol Chem 2010, 285:19483–90. 10.1074/jbc.M110.131995View ArticlePubMed CentralPubMedGoogle Scholar
- Stadtfeld M, Apostolou E, Akutsu H, Fukuda A, Follett P, Natesan S, et al.: Aberrant silencing of imprinted genes on chromosome 12qF1 in mouse induced pluripotent stem cells. Nature 2010, 465:175–81. 10.1038/nature09017View ArticlePubMed CentralPubMedGoogle Scholar
- Stadtfeld M, Apostolou E, Ferrari F, Choi J, Walsh RM, Chen T, et al.: Ascorbic acid prevents loss of Dlk1-Dio3 imprinting and facilitates generation of all-iPS cell mice from terminally differentiated B cells. Nat Genet 2012, 44:398–405. S391–392 10.1038/ng.1110View ArticlePubMed CentralPubMedGoogle Scholar
- Sutton VR, Shaffer LG: Search for imprinted regions on chromosome 14: comparison of maternal and paternal UPD cases with cases of chromosome 14 deletion. Am J Med Genet 2000, 93:381–7. 10.1002/1096-8628(20000828)93:5<381::AID-AJMG7>3.0.CO;2-9View ArticlePubMedGoogle Scholar
- Kagami M, Sekita Y, Nishimura G, Irie M, Kato F, Okada M, et al.: Deletions and epimutations affecting the human 14q32.2 imprinted region in individuals with paternal and maternal upd(14)-like phenotypes. Nat Genet 2008, 40:237–42. 10.1038/ng.2007.56View ArticlePubMedGoogle Scholar
- Kagami M, O’Sullivan MJ, Green AJ, Watabe Y, Arisaka O, Masawa N, et al.: The IG-DMR and the MEG3-DMR at human chromosome 14q32.2: hierarchical interaction and distinct functional properties as imprinting control centers. PLoS Genet 2010, 6:e1000992. 10.1371/journal.pgen.1000992View ArticlePubMed CentralPubMedGoogle Scholar
- Benetatos L, Vartholomatos G, Hatzimichael E: MEG3 imprinted gene contribution in tumorigenesis. Int J Cancer 2011, 129:773–9. 10.1002/ijc.26052View ArticlePubMedGoogle Scholar
- Zhang X, Gejman R, Mahta A, Zhong Y, Rice KA, Zhou Y, et al.: Maternally expressed gene 3, an imprinted noncoding RNA gene, is associated with meningioma pathogenesis and progression. Cancer Res 2010, 70:2350–8. 10.1158/0008-5472.CAN-09-3885View ArticlePubMed CentralPubMedGoogle Scholar
- Zhao J, Dahle D, Zhou Y, Zhang X, Klibanski A: Hypermethylation of the promoter region is associated with the loss of MEG3 gene expression in human pituitary tumors. J Clin Endocrinol Metab 2005, 90:2179–86. 10.1210/jc.2004-1848View ArticlePubMedGoogle Scholar
- Haga CL, Phinney DG: MicroRNAs in the imprinted DLK1-DIO3 region repress the epithelial-to-mesenchymal transition by targeting the TWIST1 protein signaling network. J Biol Chem 2012, 287:42695–707. 10.1074/jbc.M112.387761View ArticlePubMed CentralPubMedGoogle Scholar
- Sutton VR, McAlister WH, Bertin TK, Kaffe S, Wang JC, Yano S, et al.: Skeletal defects in paternal uniparental disomy for chromosome 14 are re-capitulated in the mouse model (paternal uniparental disomy 12). Hum Genet 2003, 113:447–51. 10.1007/s00439-003-0981-xView ArticlePubMedGoogle Scholar
- Gardiner E, Beveridge NJ, Wu JQ, Carr V, Scott RJ, Tooney PA, et al.: Imprinted DLK1-DIO3 region of 14q32 defines a schizophrenia-associated miRNA signature in peripheral blood mononuclear cells. Mol Psychiatry 2012, 17:827–40. 10.1038/mp.2011.78View ArticlePubMed CentralPubMedGoogle Scholar
- Kaneko S, Bonasio R, Saldana-Meyer R, Yoshida T, Son J, Nishino K, et al.: Interactions between JARID2 and Noncoding RNAs Regulate PRC2 Recruitment to Chromatin. Mol Cell 2014, 290:300.Google Scholar
- Cao R, Wang L, Wang H, Xia L, Erdjument-Bromage H, Tempst P, et al.: Role of histone H3 lysine 27 methylation in Polycomb-group silencing. Science 2002, 298:1039–43. 10.1126/science.1076997View ArticlePubMedGoogle Scholar
- Lee TI, Jenner RG, Boyer LA, Guenther MG, Levine SS, Kumar RM, et al.: Control of developmental regulators by Polycomb in human embryonic stem cells. Cell 2006, 125:301–13. 10.1016/j.cell.2006.02.043View ArticlePubMed CentralPubMedGoogle Scholar
- Chen HF, Kuo HC, Chien CL, Shun CT, Yao YL, Ip PL, et al.: Derivation, characterization and differentiation of human embryonic stem cells: comparing serum-containing versus serum-free media and evidence of germ cell differentiation. Hum Reprod 2007, 22:567–77.View ArticlePubMedGoogle Scholar
- Huang HP, Yu CY, Chen HF, Chen PH, Chuang CY, Lin SJ, et al.: Factors from human embryonic stem cell-derived fibroblast-like cells promote topology-dependent hepatic differentiation in primate embryonic and induced pluripotent stem cells. J Biol Chem 2010, 285:33510–9. 10.1074/jbc.M110.122093View ArticlePubMed CentralPubMedGoogle Scholar
- Thomson JA, Itskovitz-Eldor J, Shapiro SS, Waknitz MA, Swiergiel JJ, Marshall VS, et al.: Embryonic stem cell lines derived from human blastocysts. Science 1998, 282:1145–7.View ArticlePubMedGoogle Scholar
- Huang HP, Chen PH, Yu CY, Chuang CY, Stone L, Hsiao WC, et al.: Epithelial cell adhesion molecule (EpCAM) complex proteins promote transcription factor-mediated pluripotency reprogramming. J Biol Chem 2011, 286:33520–32. 10.1074/jbc.M111.256164View ArticlePubMed CentralPubMedGoogle Scholar
- Chen HF, Yu CY, Chen MJ, Chou SH, Chiang MS, Chou WH, et al.: Characteristic expression of major histocompatibility complex and immune privilege genes in human pluripotent stem cells and the derivatives. Cell Transplant 2013. Published onlineGoogle Scholar
- Varkonyi-Gasic E, Wu R, Wood M, Walton EF, Hellens RP: Protocol: a highly sensitive RT-PCR method for detection and quantification of microRNAs. Plant Methods 2007, 3:12. 10.1186/1746-4811-3-12View ArticlePubMed CentralPubMedGoogle Scholar
- Gentleman RC, Carey VJ, Bates DM, Bolstad B, Dettling M, Dudoit S, et al.: Bioconductor: open software development for computational biology and bioinformatics. Genome Biol 2004, 5:R80. 10.1186/gb-2004-5-10-r80View ArticlePubMed CentralPubMedGoogle Scholar
- Tusher VG, Tibshirani R, Chu G: Significance analysis of microarrays applied to the ionizing radiation response. Proc Natl Acad Sci U S A 2001, 98:5116–21. 10.1073/pnas.091062498View ArticlePubMed CentralPubMedGoogle Scholar
- Ginestet C: ggplot2: elegant graphics for data analysis. J R Stat Soc a Stat 2011, 174:245. 10.1111/j.1467-985X.2010.00676_9.xView ArticleGoogle Scholar
- Benjamini Y, Drai D, Elmer G, Kafkafi N, Golani I: Controlling the false discovery rate in behavior genetics research. Behav Brain Res 2001, 125:279–84. 10.1016/S0166-4328(01)00297-2View ArticlePubMedGoogle Scholar
- Mootha VK, Lindgren CM, Eriksson KF, Subramanian A, Sihag S, Lehar J, et al.: PGC-1alpha-responsive genes involved in oxidative phosphorylation are coordinately downregulated in human diabetes. Nat Genet 2003, 34:267–73. 10.1038/ng1180View ArticlePubMedGoogle Scholar
- Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gillette MA, et al.: Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci U S A 2005, 102:15545–50. 10.1073/pnas.0506580102View ArticlePubMed CentralPubMedGoogle Scholar
- Ben-Porath I, Thomson MW, Carey VJ, Ge R, Bell GW, Regev A, et al.: An embryonic stem cell-like gene expression signature in poorly differentiated aggressive human tumors. Nat Genet 2008, 40:499–507. 10.1038/ng.127View ArticlePubMed CentralPubMedGoogle Scholar
- Müller FJ, Schuldt BM, Williams R, Mason D, Altun G, Papapetrou EP, et al.: A bioinformatic assay for pluripotency in human cells. Nat Methods 2011, 8:315–7. 10.1038/nmeth.1580View ArticlePubMed CentralPubMedGoogle Scholar
- Monk D, Arnaud P, Apostolidou S, Hills FA, Kelsey G, Stanier P, et al.: Limited evolutionary conservation of imprinting in the human placenta. Proc Natl Acad Sci U S A 2006, 103:6623–8. 10.1073/pnas.0511031103View ArticlePubMed CentralPubMedGoogle Scholar
- Judson H, Hayward BE, Sheridan E, Bonthron DT: A global disorder of imprinting in the human female germ line. Nature 2002, 416:539–42. 10.1038/416539aView ArticlePubMedGoogle Scholar
- Monk D, Wagschal A, Arnaud P, Muller PS, Parker-Katiraee L, Bourc’his D, et al.: Comparative analysis of human chromosome 7q21 and mouse proximal chromosome 6 reveals a placental-specific imprinted gene, TFPI2/Tfpi2, which requires EHMT2 and EED for allelic-silencing. Genome Res 2008, 18:1270–81. 10.1101/gr.077115.108View ArticlePubMed CentralPubMedGoogle Scholar
- Bock C, Reither S, Mikeska T, Paulsen M, Walter J, Lengauer T: BiQ Analyzer: visualization and quality control for DNA methylation data from bisulfite sequencing. Bioinformatics 2005, 21:4067–8. 10.1093/bioinformatics/bti652View ArticlePubMedGoogle Scholar
- Wyman SK, Knouf EC, Parkin RK, Fritz BR, Lin DW, Dennis LM, et al.: Post-transcriptional generation of miRNA variants by multiple nucleotidyl transferases contributes to miRNA transcriptome complexity. Genome Res 2011, 21:1450–61. 10.1101/gr.118059.110View ArticlePubMed CentralPubMedGoogle Scholar
- Kim SH, Yoo BC, Broers JL, Cairns N, Lubec G: Neuroendocrine-specific protein C, a marker of neuronal differentiation, is reduced in brain of patients with Down syndrome and Alzheimer’s disease. Biochem Biophys Res Commun 2000, 276:329–34. 10.1006/bbrc.2000.3464View ArticlePubMedGoogle Scholar
- Surmacz B, Noisa P, Risner-Janiczek JR, Hui K, Ungless M, Cui W, et al.: DLK1 promotes neurogenesis of human and mouse pluripotent stem cell-derived neural progenitors via modulating Notch and BMP signalling. Stem Cell Rev 2012, 8:459–71. 10.1007/s12015-011-9298-7View ArticlePubMedGoogle Scholar
- Ku M, Koche RP, Rheinbay E, Mendenhall EM, Endoh M, Mikkelsen TS, et al.: Genomewide analysis of PRC1 and PRC2 occupancy identifies two classes of bivalent domains. PLoS Genet 2008, 4:e1000242. 10.1371/journal.pgen.1000242View ArticlePubMed CentralPubMedGoogle Scholar
- Ram O, Goren A, Amit I, Shoresh N, Yosef N, Ernst J, et al.: Combinatorial patterning of chromatin regulators uncovered by genome-wide location analysis in human cells. Cell 2011, 147:1628–39. 10.1016/j.cell.2011.09.057View ArticlePubMed CentralPubMedGoogle Scholar
- Haslinger A, Schwarz TJ, Covic M, Lie DC: Expression of Sox11 in adult neurogenic niches suggests a stage-specific role in adult neurogenesis. Eur J Neurosci 2009, 29:2103–14.View ArticlePubMedGoogle Scholar
- Axell MZ, Zlateva S, Curtis M: A method for rapid derivation and propagation of neural progenitors from human embryonic stem cells. J Neurosci Methods 2009, 184:275–84. 10.1016/j.jneumeth.2009.08.015View ArticlePubMedGoogle Scholar
- Kim KP, Thurston A, Mummery C, Ward-van Oostwaard D, Priddle H, Allegrucci C, et al.: Gene-specific vulnerability to imprinting variability in human embryonic stem cell lines. Genome Res 2007, 17:1731–42. 10.1101/gr.6609207View ArticlePubMed CentralPubMedGoogle Scholar
- Nishino K, Toyoda M, Yamazaki-Inoue M, Fukawatase Y, Chikazawa E, Sakaguchi H, et al.: DNA methylation dynamics in human induced pluripotent stem cells over time. PLoS Genet 2011, 7:e1002085. 10.1371/journal.pgen.1002085View ArticlePubMed CentralPubMedGoogle Scholar
- Xie P, Sun Y, Ouyang Q, Hu L, Tan Y, Zhou X, et al.: Physiological oxygen prevents frequent silencing of the DLK1-DIO3 cluster during human embryonic stem cells culture. Stem Cells 2014, 391:401.Google Scholar
- Astuti D, Latif F, Wagner K, Gentle D, Cooper WN, Catchpoole D, et al.: Epigenetic alteration at the DLK1-GTL2 imprinted domain in human neoplasia: analysis of neuroblastoma, phaeochromocytoma and Wilms’ tumour. Br J Cancer 2005, 92:1574–80. 10.1038/sj.bjc.6602478View ArticlePubMed CentralPubMedGoogle Scholar
- Wang P, Ren Z, Sun P: Overexpression of the long non-coding RNA MEG3 impairs in vitro glioma cell proliferation. J Cell Biochem 2012, 113:1868–74. 10.1002/jcb.24055View ArticlePubMedGoogle Scholar
- Di Sano F, Fazi B, Tufi R, Nardacci R, Piacentini M: Reticulon-1C acts as a molecular switch between endoplasmic reticulum stress and genotoxic cell death pathway in human neuroblastoma cells. J Neurochem 2007, 102:345–53. 10.1111/j.1471-4159.2007.04479.xView ArticlePubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.