PRC2 Is Required to Maintain Expression of the Maternal Gtl2-Rian-Mirg Locus by Preventing De Novo DNA Methylation in Mouse Embryonic Stem Cells

Abstract

Polycomb Repressive Complex 2 (PRC2) function and DNA methylation (DNAme) are typically correlated with gene repression. Here, we show that PRC2 is required to maintain expression of maternal microRNAs (miRNAs) and long non-coding RNAs (lncRNAs) from the Gtl2-Rian-Mirg locus, which is essential for full pluripotency of iPSCs. In the absence of PRC2, the entire locus becomes transcriptionally repressed due to gain of DNAme at the intergenic differentially methylated regions (IG-DMRs). Furthermore, we demonstrate that the IG-DMR serves as an enhancer of the maternal Gtl2-Rian-Mirg locus. Further analysis reveals that PRC2 interacts physically with Dnmt3 methyltransferases and reduces recruitment to and subsequent DNAme at the IG-DMR, thereby allowing for proper expression of the maternal Gtl2-Rian-Mirg locus. Our observations are consistent with a mechanism through which PRC2 counteracts the action of Dnmt3 methyltransferases at an imprinted locus required for full pluripotency.

Figure 1. PRC2 is required to maintain expression of maternal miRNAs and lncRNAs at the Gtl2-Rian-Mirg locus

(A) Small RNA-seq demonstrates log-fold changes of miRNA expression in Ezh2−/− mESCs compared to wild-type. Significantly reduced expression of a cluster of miRNAs is observed at the Gtl2-Rian-Mirg locus of chromosome 12 in Ezh2−/− mESCs compared to wild-type. (B) Schematic representation of the Dlk1-Dio3 imprinted gene cluster. lncRNAs genes- Gtl2, Rian and Mirg, miRNAs and snoRNAs are expressed from “maternally” inherited chromosome, whereas, protein coding genes, Dlk1, Dio3 and Rtl1 are expressed from “paternally” inherited chromosome. Empty boxes represent genes that are repressed. Imprinting is regulated by IG-DMR, which is methylated in paternally inherited chromosome, but unmethylated in maternally inherited chromosome. Therefore, by default all lncRNAs, miRNAs and snoRNAs from paternally inherited chromosome are repressed due to hypermethylation at IG-DMR, and only maternal ones are expressed. (C) Quantitative RT-PCR (RT-qPCR) confirms dramatically reduced expression of maternal miRNAs from the Gtl2-Rian-Mirg locus in Ezh2−/− mESCs. miR-130a is shown as a control. miRNAs expression represented as mean +/− SEM (n=3); p-values were calculated using a 2-way ANOVA; ***p <0.0001, *p <0.01. (D) RT-qPCR shows dramatic reduction of maternal Gtl2, Rian and Mirg lncRNAs expression in Ezh2−/− mESCs as compared to wild-type. Dlk1 and Dio3 mRNA expressions are unaltered in Ezh2−/− mESCs. Transcript levels were normalized to Gapdh. Data are represented as mean +/− SEM (n=3); p-values were calculated using a 2-way ANOVA; ***p <0.0001, ns (non-significant). (E) ChIP-seq analysis of RNA Pol II demonstrates log-fold changes of RNA Pol II occupancy in Ezh2−/− mESCs compared to wild-type. RNA Pol II occupancy is significantly reduced at the entire Gtl2-Rian-Mirg locus (~220kb) in Ezh2−/− mESCs compared to wild-type. RNA Pol II co-occupancy with H3K36me3 and H3K79me2 (elongation marks) suggests that the maternal Gtl2-Rian-Mirg locus acts as a single transcriptional unit. See also Figure S1.

Figure 2. Methylation of the Gtl2-Rian-Mirg locus in the absence of PRC2

(A–B) Several independent Ezh2 rescue clones express different levels of exogenous Ezh2 (Figures S2A & S2C). Rescue clones with lower level of Ezh2 expression fail to rescue the expression of maternal lncRNAs and miRNAs. Ezh2 rescue clones, A5 and B6 that express at a near endogenous level of Ezh2 (Figures S2A & S2C) also fail to restore the expression of maternal Gtl2, Rian and Mirg lncRNAs as well as miRNAs from the Gtl2-Rian-Mirg locus. mRNA transcript levels were normalized to Gapdh. Both mRNA and miRNA expressions are shown as mean +/− SEM (n=3); p-values were calculated using a one-way ANOVA; ***p <0.0001, *p <0.01, ns (non-significant). (C) Analysis of 29 CpGs at the IG-DMR shows gain of DNAme (%) in Ezh2−/− mESCs compared to wild-type. Ezh2 rescue clones (A5 and B6) that expresses similar level of endogenous Ezh2, retains hypermethylation at IG-DMR, indicating stable establishment of DNAme at the IG-DMR in absence of Ezh2. DNAme at Nanog proximal promoter was used as a control. Data are represented as mean +/− SEM (n=3); p-values were calculated using a 2-way ANOVA; ***p <0.0001, ns (non-significant). See also Figure S2.

Figure 3. IG-DMR/Enhancer1 serves as an enhancer for the Gtl2-Rian-Mirg locus

(A–D) Co-occupancy of ESC-specific TFs (e.g. Oct4, Nanog, Sox2, Klf4, Esrrb), mediator (Med1/12), cohesin (Smc1/3), Lsd1, H3K27ac and H3K4me1 at IG-DMR/Enh1 and Enh2 fulfills criteria for putative enhancer regions of Gtl2-Rian-Mirg locus. The zoomed in shaded regions show Dlk1 promoter, IG-DMR/Enh1, Gtl2 promoter and Enh2 regions, which are occupied with several factors and histones marks in Ezh2−/− and wild-type mESCs. Multiple individual Chip-seq genomic tracks of PRC2 components show weak occupancy of Ezh2, Jarid2 and no binding of Suz12 of PRC2 components at the IG-DMR/Enh1 and Enh2, and failed to observe detectable H3K27me3 deposition. (E) Luciferase reporter assays of Enh1 and Enh2 demonstrate strong enhancer activity as Nanog enhancer. Non-Enh1 and Non-Enh2 (lacks binding of any of the factors and histone marks, see Figure. S3A) both were used as controls. Data are represented as mean +/− SEM (n=3); p-values were calculated using a 2-way ANOVA; ***p <0.0001, **p <0.001. (F) Biallelic deletion of Enh2 (Enh2−/−) (~7kb) reveals no effect on the Gtl2-Rian-Mirg locus, whereas, biallelic deletion of IG-DMR/Enh1 (IG-DMR/Enh1−/−) (~7kb) abrogates expression of maternal Gtl2, Rian and Mirg. Non-Enh2−/− (~7kb) used as a control. mRNA expression of Dlk1, Dio3, Gtl2, Rian and Mirg were examined from undifferentiated wild-type, IG-DMR−/−, Enh2−/− and Non-Enh2−/− mESCs. mRNA expressions are represented as mean +/− SEM (n=3); p-values were calculated using a 2-way ANOVA; ***p <0.0001, **p <0.001, *p <0.01, ns (non-significant). See also Figure S3.

Figure 4. PRC2 physically interacts with Dnmt3a/3l in Gtl2 lncRNA-independent manner; the interaction between Gtl2 lncRNA-Ezh2 inhibits binding of Ezh2/PRC2 at the IG-DMR

(A) Scatter plot representing differentially expressed genes from Ezh2−/− mESCs compared to wild-type. Red dots represents significantly up- and down down-regulated genes in Ezh2−/− mESCs with a q-value <0.01. Genes of interest are labeled in the scatter-plot. (B) mRNA expression shows significant up-regulation of Dnmt3a, Dnmt3b and Dnmt3l, but not Dnmt1, in Ezh2−/− mESCs as compared to wild-type. Transcript levels were normalized to Gapdh. Data are represented as mean +/− SEM (n=3); p-values were calculated using a 2-way ANOVA; ***p <0.0001, *p <0.01, ns (non-significant). (C) Anti-Ezh2 antibody was used to immunoprecipitate endogenous Ezh2 from mESCs nuclear extracts, which shows a specific interaction between Ezh2 and Dnmt3a/Dnmt3l. (D) RT-qPCR shows that biallelic deletion of IG-DMR−/− causes abrogation of maternal Gtl2 and Rian lncRNAs in mESCs. Endogenous Ezh2 maintains interaction with Dnmt3a/Dnmt3l in absence of Gtl2 lncRNA. (E) RNA immunoprecipitation (RIP) demonstrates a strong interaction of Gtl2 lncRNA with Ezh2, but not with Dnmt3a. U1 RNA and Oct4 mRNA were used as controls. Data are represented as mean +/− SEM (n=3); p-values were calculated using a 2-way ANOVA; ***p <0.0001, ns (non-significant). (F) RT-qPCR shows that biallelic deletion of Gtl2 promoter (~7kb) disrupts the formation of Gtl2 lncRNA. (G) ChIP-qPCR shows increased Ezh2 occupancy at the IG-DMR in absence of Gtl2 lncRNA. Data are represented as mean +/− SEM (n=3); p-values were calculated using a 2-way ANOVA; **p <0.001, *p <0.01, ns (non-significant). (H) ChIP-qPCR shows no significant increase binding of H3K27me3 at the IG-DMR in absence of Gtl2 lncRNA. Data are represented as mean +/− SEM (n=3); p-values were calculated using a 2-way ANOVA; ns (non-significant). See also Figure S4.

Figure 5. PRC2 antagonizes de novo DNAme at the IG-DMR through distinct mechanism

(A–B) ChIP-qPCR shows Dnmt3a, Dnmt3b and Dnmt3l occupancy at IG-DMR is significantly increased in absence of Ezh2 and Jarid2, but occupancy of Dnmt1 remains unchanged. Data are represented as mean +/− SEM (n=3); p-values were calculated using a 2-way ANOVA; ***p <0.0001, ns (non-significant). (C) Analysis of 29 CpGs at the IG-DMR shows different DNAme (%) levels in the absence of PRC2 components. Data are represented as mean +/− SEM (n=3); p-values were calculated using a 2- way ANOVA; ***p <0.0001. (D) Global DNA methylation (DNAme) analysis from Ezh2−/− and wild-type mESCs, using reduced-representation bisulfite sequencing (RRBS), represented as heat map of genome wide methylation patterns. The genome was divided into non-overlapping 10kb windows and the fraction of methylated CpGs in each window was computed for wild type and Ezh2−/− mutants. The hue represents the number of genomic windows with a given fractional methylation in Ezh2−/− vs wild-type. Trends suggest significantly increased global DNAmethylation in Ezh2−/−. See also Figure S5.

Figure 6. PRC2 protects IG-DMR from de novo DNAme to allow proper expression of the maternal Gtl2-Rian-Mirg locus

(A) Overexpression of Ezh2 in wild-type mESCs. Protein expression of Ezh2 was checked through western blot. Actin used as an internal control. (B) mRNA expression shows no significant change of Gtl2 lncRNA expression upon overexpression of Ezh2. Transcript levels were normalized to Gapdh. Data are represented as mean +/− SEM (n=3); p-values were calculated using a 2-way ANOVA; **p <0.001, ns (non-significant). (C) Analysis of 29 CpGs at the IG-DMR shows no significant changes of DNAme (%) levels upon overexpression of Ezh2. Data are represented as mean +/− SEM (n=3); p-values were calculated using a 2- way ANOVA; ns (non-significant). (D) Dox-inducible overexpression of Dnmt3a (western blot) does not change Gtl2 lncRNA expression (RT-qPCR) (E), with slight increase of DNAme level at the IG-DMR (F). Data are represented as mean +/− SEM (n=3); p-values were calculated using a 2- way ANOVA; **p <0.001, ns (non-significant). (G) Overexpression of Ezh2 in Dnmt3a−/− mESCs (western blot) leads to no significant change in Gtl2 lncRNA expression (RT-qPCR) (H) and DNAme at the IG-DMR (I). Data are represented as mean +/− SEM (n=3); p-values were calculated using a 2- way ANOVA; ***p <0.0001, ns (non-significant). See also Figure S6.

Figure 7. The working model portrays the mechanism by which Ezh2/PRC2 protects the IG-DMR locus from de novo DNAme to allow proper expression of the maternal Gtl2-Rian-Mirg locus in mESCs

A model schematically representing our findings, where Gtl2 lncRNA binds to Ezh2 and inhibits interaction of Ezh2/PRC2 at the IG-DMR locus, and subsequent deposition of H3K27me3. The presence of Ezh2/PRC2 in association with Gtl2 lncRNA prevents Dnmt3s recruitment and subsequent de novo DNAme, and allows ESC-specific TFs, mediators and other histone modifiers to bind at the IG-DMR/Enhancer1 locus that ultimately drives expression of the maternal Gtl2-Rian-Mirg locus. In the absence Ezh2, it is unable to prevent recruitment of Dnmt3s at the IG-DMR locus. Dnmt3s is then recruited to the IG-DMR and deposits de novo DNAme, leading to transcription repression of the maternal Gtl2-Rian-Mirg locus. Significant reduction of H3K27ac and H3K4me3 occupancy at the IG-DMR and Gtl2 promoter is observed in the absence of Ezh2. For simplicity, only the maternal allele is shown.