The DEAD-box RNA-binding protein DDX6 regulates parental RNA decay for cellular reprogramming to pluripotency

Abstract

Cellular transitions and differentiation processes require mRNAs supporting the new phenotype but also the clearance of existing mRNAs for the parental phenotype. Cellular reprogramming from fibroblasts to induced pluripotent stem cells (iPSCs) occurs at the early stage of mesenchymal epithelial transition (MET) and involves drastic morphological changes. We examined the molecular mechanism for MET, focusing on RNA metabolism. DDX6, an RNA helicase, was indispensable for iPSC formation, in addition to RO60 and RNY1, a non-coding RNA, which form complexes involved in intracellular nucleotide sensing. RO60/RNY1/DDX6 complexes formed prior to processing body formation, which is central to RNA metabolism. The abrogation of DDX6 expression inhibited iPSC generation, which was mediated by RNA decay targeting parental mRNAs supporting mesenchymal phenotypes, along with microRNAs, such as miR-302b-3p. These results show that parental mRNA clearance is a prerequisite for cellular reprogramming and that DDX6 plays a central role in this process.

Fig 1. Analysis of non-coding RNAs during the early stage of iPS reprogramming.

(A) Heat map showing non-coding RNA expression at various time points during iPS reprogramming obtained using the LncProfiler qPCR Array Kit. Graph showing RNY1 gene expression during iPS reprogramming. Statistical differences were assessed by t-test and standard Bonferroni correction between Day 0 and Days 3 and 6. (B) Statistical analysis of non-coding RNA expression during iPS reprogramming. (C) Secondary structure of RNY1 from computational prediction. (D) Structure of human RO60 from computational modeling. (E) RNA expression of RNY1 in the cytoplasm and nucleus of OSKM-treated TIG-1 fibroblasts. Individual RNA expression levels were normalized to GAPDH expression levels. Data are presented as the mean ± SEM. (F) Schematic representation of the early iPS reprogramming analysis using the transient knockdown method on Day 3 for (G) to (J). Med: medium. OSKM: OCT4, SOX2, KLF4, and c-MYC. (G) Phase contrast micrograph images of siRNA for Negative control, RNY1#1-, RNY1#2-, and OSKM-treated TIG-1 fibroblasts after 3 days. The white bar indicates 200 μm. (H) Cell numbers for OSKM- and siRNA-treated TIG-1 fibroblasts on Day 3. (I) RNA expression in OSKM- and siRNA-treated TIG-1 fibroblasts on Day 3. Individual RNA expression levels were normalized to GAPDH expression levels. Data are presented as the mean ± SEM. (J) ChIP-qPCR analysis of iPS reprogramming on Day 3. (K) Schematic representation of the iPSC reprogramming analysis using the transient knockdown method on Days 9 and 24. AP: alkaline phosphatase. (L) Efficacy of iPS reprogramming with siRNAs by flow cytometry using SSEA4 and TRA-1-60 antibodies, AP staining on Day 9, and AP-positive colony counting of iPSCs on Day 24. (M) Cell numbers and RNY1 expression levels of RNY1-overexpressing TIG-1 fibroblasts 3 days after transfection. OE: Overexpression. (N) Efficacy of iPSC reprogramming with an overexpression plasmid by flow cytometry using SSEA4 and TRA-1-60 antibodies, AP staining on Day 9, and colony counting of iPSC colonies using AP staining on Day 24. OE: overexpression. (O) Phase contrast micrograph images of RNY1-overexpressing TIG-1 fibroblasts after 3 days. The white bar indicates 200 μm.

Fig 2. Identification of the functions of RO60 and DDX6 during iPS reprogramming.

(A) Silver staining of immunoprecipitated proteins using the RO60 antibody. C: Control antibody; R: RO60 antibody. (B) Candidate proteins that interact with RO60. Protein localization was evaluated using GeneCards (http://www.genecards.org). (C) Immunoprecipitation (IP) of protein extracts from OSKM-transduced TIG-1 fibroblasts with magnetic beads coupled to RO60 or DDX6 antibodies or to an irrelevant isotype-matched control antibody, followed by an immunoblot analysis with RO60 or DDX6 antibodies. Graphs showing RNY1 abundances in immunoprecipitated TIG-1 proteins determined by qPCR. C: Control antibody. R: RO60 antibody. D: DDX6 antibody. (D) Immunoblotting analysis of rDDX6 and rRO60 binding assays in vitro with RO60 and DDX6 antibodies. (E) Binding assays of rDDX6 to rRO60 using the BLItz system. Whole rDDX6 binding to rRO60. Binding kinetics of rDDX6 was titratable to 193 nM. (F) Binding assay results for rDDX6 to rRO60.

Fig 3. DDX6-RO60 complexes were regulated by RNY1.

(A) Immunocytochemistry of OSKM-treated TIG-1 fibroblasts on Day 3. Graph showing the percentage of DDX6-positive cells. (B) Immunoblotting analysis of Cas9-treated TIG-1 fibroblasts. Tom: tdTomato as a mock control. (C) iPS reprogramming of Cas9-treated TIG-1 fibroblasts. AP-positive colonies were counted on Day 20. (D) Immunoblotting analysis of RO60-IP proteins from OSKM- and siRNA-treated TIG-1 fibroblasts with RO60, DDX6, and GAPDH antibodies. The raw blotting data are attached to S8 Fig. (E) Immunocytochemical analyses of RO60, DDX6, and OCT4 in OSKM- and siRNA-treated TIG-1 fibroblasts. (F) Immunocytochemical analyses of TNRC6A, DDX6, and OCT4 in OSKM- and siRNA-treated TIG-1 fibroblasts. (G) Immunocytochemical results for RO60, TNRC6A, and OCT4 in OSKM- and siRNA-treated TIG-1 fibroblasts. (H) Immunoblotting analysis of DDX6-IP proteins from OSKM- and siRNA-treated TIG-1 fibroblasts with TNRC6A, DDX6, and GAPDH antibodies. The white bar indicates 50 μm in (A), (E) to (G).

Fig 4. miRNA analysis of iPS reprogramming.

(A) Phase contrast microscopy images of OSKM- and siRNA-treated TIG-1 fibroblasts from Days 1 to 3. Immunocytochemistry analysis on Day 3 using OCT4 antibody and CytoPainter Phalloidin-iFluor 594. The white bar indicates 200 μm. (B) Immunocytochemical analyses of OCT4, DAPI, and F-actin in OSKM- and siRNA-treated TIG-1 fibroblasts. Arrowheads indicate OCT4-positive cells. The white bar indicates 50 μm. (C) Immunocytochemical analyses of DAPI and CDH1 or THY1 in OSKM- and siRNA-treated TIG-1 fibroblasts. The white bar indicates 100 μm. (D) Heat map of miRNA expression of DDX6-IP in TIG-1 fibroblasts using the nCounter system. These miRNAs were significantly expressed. Volcano plot of global miRNA expression in DDX6-IP proteins in OSKM-treated TIG-1 fibroblasts using the nCounter system. (E) Mature miRNA expression in OSKM- and siRNA-treated TIG-1 fibroblasts from Days 1 to 3. Individual RNA expression levels were normalized to the respective RNU44 expression levels. Data represent the mean ± SEM. (F) Target genes of miRNAs in DDX6-IP proteins were categorized based on KEGG pathways. Red letters indicate iPS reprogramming-related signaling pathways. (G) Heat map of mRNA expression in whole TIG-1 fibroblast lysates. (H) Luciferase activity using hsa-miR-302b-3p targeting the pmirGLO plasmid carrying the 3′ UTR of TGFβR2. d2 and d3 indicates 2 or 3 days after OSKM induction.

Fig 5. RNY1 is essential for RNA decay via the regulation of DDX6.

(A) Schematic representation of the early iPS reprogramming analysis using the BRIQ-qPCR method. (B) BrU-RNA degradation in OSKM- and siRNA-treated TIG-1 fibroblasts. Individual RNA expression levels were normalized to the Spike-In control. Data are presented as the mean ± SEM. (C) Table of T_1/2 of OSKM- and siRNA-treated TIG-1 fibroblasts. (D) Illustration of the proposed model to explain the molecular mechanism for RNA decay at the early reprogramming stage.