Multiphasic and Dynamic Changes in Alternative Splicing during Induction of Pluripotency Are Coordinated by Numerous RNA-Binding Proteins

Abstract

Alternative splicing (AS) plays a critical role in cell fate transitions, development, and disease. Recent studies have shown that AS also influences pluripotency and somatic cell reprogramming. We profiled transcriptome-wide AS changes that occur during reprogramming of fibroblasts to pluripotency. This analysis revealed distinct phases of AS, including a splicing program that is unique to transgene-independent induced pluripotent stem cells (iPSCs). Changes in the expression of AS factors Zcchc24, Esrp1, Mbnl1/2, and Rbm47 were demonstrated to contribute to phase-specific AS. RNA-binding motif enrichment analysis near alternatively spliced exons provided further insight into the combinatorial regulation of AS during reprogramming by different RNA-binding proteins. Ectopic expression of Esrp1 enhanced reprogramming, in part by modulating the AS of the epithelial specific transcription factor Grhl1. These data represent a comprehensive temporal analysis of the dynamic regulation of AS during the acquisition of pluripotency.

Figure 1. Genome-wide Analysis of Alternative Splicing during Induced Pluripotency Reveals Distinct Phases of Regulation

(A) Experimental design and heatmap of genome-wide alternative splicing changes; blue and red represent decreased and increased PSI, respectively, relative to the mean of each transcript across the time course. (B) Examples of clusters with the median PSI of three replicates at each time point of each exon within selected clusters are graphed in gray and the median PSI for the cluster in red. (C) Radiolabeled RT-PCR validations of phase-specific AS during reprogramming as well as in ESCs (last lane). For supplemental data, see also Figure S1.

Figure 2. Complex Regulation of AS during Reprogramming Involving Multiple Splicing Factors

(A) Heatmap of gene expression changes of 95 RBPs selected based on the criteria described in the text. Examples of RBPs addressed in subsequent figures are listed to the right. (B) RT-PCR analysis of AS events that occur during reprogramming and are induced by Mbnl1/2 knockdown in MEFs (left column) or are reverted by ectopic MBNL1 in V6.5 ESCs (right column and graph, which is the average PSI of biological duplicate with error bars representing SD from the mean [SDM]). (C) RT-PCR analysis of AS events that occur by day 4 and are induced by two independent Zcchc24 small hairpin RNAs (shRNAs) in MEFs. (D) RT-PCR analysis of AS events that change between day 20 and iPSC and are reverted by two independent siRNAs for Rbm47 in ESCs (for C and D, heatmaps of RNA-seq data are shown [upper right] and RT-PCR [left] with quantitation of triplicate RT-PCRs in the graph and error bars representing SDM). For supplemental data, see also Figures S2 and S3.

Figure 3. Esrp1 Regulates MET Phase AS and Enhances Reprogramming

(A) Heatmap of AS events that coincide with the expression changes in epithelial and mesenchymal marker genes during reprogramming. (B) RT-PCR validation of MET-phase AS events that are reverted by Esrp1/2 knockout in ESCs. (C) Quantitation of the PSI changes induced by Esrp1/2 KO in ESCs in biological duplicate; error bars are SDM. (D) Enrichment for Esrp-binding motifs near exons regulated at the MET phase (day 7). Esrp-binding motif enrichment is seen upstream of day 7 alternative exons that decrease in PSI (blue) and downstream of those that increase in PSI (red); the pre-mRNA is shown below with the alternative exon represented in green and coordinates in base pairs (bps). (E) Doxycycline-reprogrammable MEFs with an Oct4-NEO-Resistance allele were transduced with EGFP or Esrp1-emerald retroviral vectors and treated with Dox (2 μg/ml) for 7 days followed by G418 selection for 1 week and then AP stain. Average colonies per well of five biological replicates with error bars representing SDM (above) and representative images of AP-positive colonies (below) are shown. (F) Time course of Dox treatment prior to G418 selection for 1 week using the same MEF line as (E). Error bars are SDM of biological triplicate at each time point. (G) AP-positive colonies that persisted after 10 days of Dox treatment followed by its removal and 10 days of Dox-free culture, conducted in biological triplicate. For supplemental data, see Figures S3 and S4.

Figure 4. Esrp1 Promotes Expression of a Grhl1 Isoform that Enhances Reprogramming

(A) RT-PCR analysis of Grhl1 exon 5 splicing during reprogramming and effect of Esrp1 KO in ESCs. (B) Diagram of the full-length isoform of Grhl1 (Grhl1-FL) including the CP2 DNA-binding domain in blue and the truncated/short isoform (Grhl1-S) with exon 5 skipped and three out-of-frame amino acids from exon 6 that precede the PTC depicted in yellow. (C) Dox-reprogrammable MEFs were transduced with empty vector or Esrp1-FLAG and treated with Dox for the indicated time followed by RNA isolation and RT-PCR to assess Grhl1 exon 5 inclusion. (D) Dox-reprogrammable MEFs were transduced with EGFP, Grhl1-S, or Grhl1-FL and assayed for AP-positive colonies after 10 days of Dox treatment in biological quadruplicate; representative AP-positive colonies shown below; error bars are SDM. (E) Schematic of regulators of alternative splicing during induced pluripotency. For supplemental data, see Figure S4.