The p53-induced lincRNA-p21 derails somatic cell reprogramming by sustaining H3K9me3 and CpG methylation at pluripotency gene promoters

Abstract

Recent studies have boosted our understanding of long noncoding RNAs (lncRNAs) in numerous biological processes, but few have examined their roles in somatic cell reprogramming. Through expression profiling and functional screening, we have identified that the large intergenic noncoding RNA p21 (lincRNA-p21) impairs reprogramming. Notably, lincRNA-p21 is induced by p53 but does not promote apoptosis or cell senescence in reprogramming. Instead, lincRNA-p21 associates with the H3K9 methyltransferase SETDB1 and the maintenance DNA methyltransferase DNMT1, which is facilitated by the RNA-binding protein HNRNPK. Consequently, lincRNA-p21 prevents reprogramming by sustaining H3K9me3 and/or CpG methylation at pluripotency gene promoters. Our results provide insight into the role of lncRNAs in reprogramming and establish a novel link between p53 and heterochromatin regulation.

Figure 1

Functional screening of lincRNAs in the preiPSC to iPSC conversion. (A) Schematic view of the screening. (B) Heatmap of 67 lincRNAs quantified by qPCR in the 4 indicated cell types. The expression level of individual lincRNAs was normalized to MEFs. Colors represent fold change above (red) or below (blue) the values in MEFs. (C) Box plot analysis of lincRNA expression levels in the indicated cell types. I-VI indicates distinct expression patterns as shown in B. PSCs = iPSCs and ESCs. (D) Upper panel, knockdown efficiency of shRNAs for the indicated lincRNAs in a selected preiPSC clone. Data are presented as mean ± SD of 3 independent experiments with triplicate items (also in the lower panel). shRNA targeting firefly luciferase was used as control (shCtrl; also hereafter in similar experiments). Lower panel, relative number of GFP+ colonies arising from the same preiPSC clone after lincRNA shRNA knockdown and treatment with Vc compared to control shRNA. ^*P < 0.05, ^**P < 0.01, ^***P < 0.001 by the Student's t-test (also hereafter). (E) Representative images of GFP+ colonies produced as in D. Bright-field (phase) is also shown. Scale bar, 100 μm. Supportive data are included in Supplementary information, Figure S1 and Table S1.

Figure 2

p53-induced lincRNA-p21 is a barrier for reprogramming. (A) Representative western blot for p53 in the 4 indicated cell types. ACTIN was the loading control. (B) Relative number of GFP+ colonies arising from 2 independent preiPSC clones cotransduced with control shRNA or 2 independent shRNAs for p53. Data are presented as mean ± SD of 3 independent experiments with triplicate items (also in D, I, J and K). (C) qPCR analysis for lincRNA-p21 in MEFs cotransduced with OSKM and either control shRNA or 2 independent shRNAs for p53. Data are presented as mean ± SD of 2 independent experiments with triplicate items (also in F). (D) Left panel, qPCR analysis for lincRNA-p21 in MEFs cotransduced with OSKM and either control shRNA or 2 independent shRNAs for lincRNA-p21 on day 6. Right panel, relative number of GFP+ colonies counted at day 15 (also in I, J and K) in similar reprogramming experiments. (E) Apoptosis in MEFs cotransduced with OSKM and either control shRNA or 2 independent shRNAs for lincRNA-p21. Samples were analyzed on day 6. shRNA for p53 was used as positive control (also in F). Data are presented as mean ± SD of a representative experiment with triplicate items. (F) Cell proliferation in MEFs cotransduced with OSKM and either control shRNA or 2 independent shRNAs for lincRNA-p21. (G) Representative immunofluorescence, phase contrast photographs and photographs of chimeric mice obtained with an iPSC clone produced with OSKM and shRNA for lincRNA-p21. Scale bar, 50 μm (left panels) and 200 μm (right panels). (H) Karyotype analysis for the same iPSC clone employed for chimeric mouse generation in G. (I) Relative number of GFP+ colonies in MEFs cotransduced with OSKM and lincRNA-p21. Empty vector was used as control (control; also hereafter in similar experiments). (J) Relative number of GFP+ colonies in MEFs cotransduced with OSKM and the indicated combinations of lincRNA-p21 and shRNAs. (K) Relative number of GFP+ colonies in MEFs co-transduced with OSKM and either control shRNA, shRNA for lincRNA-p21 or shRNA for lincRNA-p21 plus shRNA-resistant lincRNA-p21. Supportive data are included in Supplementary information, Figures S2 and S3.

Figure 3

LincRNA-p21 and HNRNPK associate with SETDB1 and DNMT1 in reprogramming. (A) Representative analysis of lincRNA-p21 distribution by cellular fractionation of MEFs reprogrammed with OSKM on day 6. U6 RNA and Gapdh mRNA served as controls for nuclear and cytoplasmic RNAs, respectively. (B) RIP analysis of DNMT1, DNMT3A, DNMT3B, SUV39H1 and SETDB1 in MEFs reprogrammed with OSKM on day 12. Data are presented as mean ± SD of 2 independent experiments with triplicate items. (C) Representative immunoprecipitation (IP) of SETDB1 (left) or DNMT1 (right) in MEFs reprogrammed with OSKM on day 12. The membranes were immunoblotted (IB) for HNRNPK, SETDB1 or DNMT1. Specific bands are marked with an arrow (also in D). (D) Representative IP of HNRNPK in MEFs reprogrammed with OSKM on day 12. The membranes were immunoblotted for HNRNPK, SETDB1 or DNMT1. (E) In vitro RIP analysis of SETDB1 using purified lincRNA-p21, SETDB1-Flag and HNRNPK-Flag at different concentrations. The extracted RNA was then analyzed by qPCR (upper) and semi-qPCR (lower). Data are presented as mean ± SD of 2 independent experiments with triplicate items (upper) and a representative experiment (lower). Supportive data are included in Supplementary information, Figure S4.

Figure 4

HNRNPK regulates reprogramming. (A, B) RIP analysis of SETDB1 (A) or DNMT1 (B) in MEFs cotransduced with OSKM and either control shRNA or an shRNA for Hnrnpk on day 12. The concentration of lincRNA-p21 (measured by qPCR) in the different input samples (shRNA for Hnrnpk or control) employed for RIP was comparable. Data are presented as mean ± SD of 3 independent experiments with triplicate items (also in C, D, E and F). (C) Relative number of GFP+ colonies counted on day 15 (also in D and E) in MEFs cotransduced with OSKM and either control shRNA or 2 independent shRNAs for Hnrnpk.(D) Relative number of GFP+ colonies in MEFs reprogrammed with OSKM and overexpressed HNRNPK. (E) Relative number of GFP+ colonies in MEFs cotransduced with OSKM and the indicated shRNAs. NS = not significant. (F) Relative number of GFP+ colonies arising from 2 independent preiPSC clones transduced with control shRNA or 2 independent shRNAs for Hnrnpk. Supportive data are included in Supplementary information, Figure S4.

Figure 5

The function of LincRNA-p21 in reprogramming is mainly mediated by HNRNPK. (A) Venn diagrams show the overlap of genes significantly differentially upregulated (> 1.5-fold) by lincRNA-p21 or Hnrnpk knockdown in MEFs cotransduced with OSKM. (B) Heatmap shows all genes significantly differentially regulated (q-value < 0.05) by lincRNA-p21 and Hnrnpk knockdown. Colors represent fold-change levels above (red) or below (blue) the control. Selected genes are indicated. (C) Principal component analysis of the variance between samples. D = day. (D) qPCR analysis for the indicated pluripotency genes in MEFs cotransduced with OSKM and either control shRNA or 2 independent shRNAs for lincRNA-p21 on day 12. Data are presented as mean ± SD of 3 independent experiments with triplicate items. Supportive data are included in Supplementary information, Figures S5, S6 and Table S2.

Figure 6

LincRNA-p21 regulates H3K9me3 and CpG methylation at pluripotency genes. (A-C) ChIP analysis of MEFs cotransduced with OSKM and control shRNA or shRNA for lincRNA-p21 on day 12. Binding of HNRNPK (A), SETDB1 (B) and DNMT1 (C) at the promoters of the indicated pluripotency genes was quantified by qPCR and shown as relative enrichment compared with IgG. Gapdh promoter served as control. Data are presented as mean ± SD of 3 (A) or 2 (B, C and D) independent experiments with triplicate items. (D) ChIP analysis of MEFs cotransduced with OSKM and the indicated combinations on day 12. H3K9me3 enrichment at the promoters of the indicated pluripotency genes were quantified by qPCR and then normalized to total histone H3. Gapdh promoter served as control. NC = non-targeted control, siSetdb1 and siDnmt1 refer to siRNA targeting Setdb1 or Dnmt1, respectively (E). (E) Representative bisulfite sequencing analysis of the Nanog promoter in MEFs cotransduced with OSKM and the indicated combinations on day 15. (F) Model for lincRNA-p21 function in reprogramming. Supportive data are included in Supplementary information, Figure S7.