Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 33 (8), 878-89

Fip1 Regulates mRNA Alternative Polyadenylation to Promote Stem Cell Self-Renewal


Fip1 Regulates mRNA Alternative Polyadenylation to Promote Stem Cell Self-Renewal

Brad Lackford et al. EMBO J.


mRNA alternative polyadenylation (APA) plays a critical role in post-transcriptional gene control and is highly regulated during development and disease. However, the regulatory mechanisms and functional consequences of APA remain poorly understood. Here, we show that an mRNA 3' processing factor, Fip1, is essential for embryonic stem cell (ESC) self-renewal and somatic cell reprogramming. Fip1 promotes stem cell maintenance, in part, by activating the ESC-specific APA profiles to ensure the optimal expression of a specific set of genes, including critical self-renewal factors. Fip1 expression and the Fip1-dependent APA program change during ESC differentiation and are restored to an ESC-like state during somatic reprogramming. Mechanistically, we provide evidence that the specificity of Fip1-mediated APA regulation depends on multiple factors, including Fip1-RNA interactions and the distance between APA sites. Together, our data highlight the role for post-transcriptional control in stem cell self-renewal, provide mechanistic insight on APA regulation in development, and establish an important function for APA in cell fate specification.


Figure 1
Figure 1. Fip1 is required for ESC self-renewal
A Fip1 knock-down in ESCs. Western blots of Fip1 and other CPSF subunits using cell lysates prepared from ESCs 4 days after transfection with lipids only (Mock), control siRNA (siControl), or Fip1 siRNAs (siFip1-2 and x-3). B–D Impact of Fip1 KD on ESC self-renewal. OctGiP ESCs were transfected with the indicated siRNAs, and cells were analyzed 4 days post-transfection by FACS (the percentages of GFP-negative cells are plotted as mean ± s.e.m. [n = 3; **P < 0.01)] (B), or by AP staining (C), or by RT–qPCR to determine lineage marker expression [expression values are normalized to mock and plotted as mean ± s.e.m. (n = 3)] (D). Marker genes that showed a significant different expression between siControl and siFip1-2/3 are marked by * (P < 0.05). E ESCs were cultured without LIF for 0, 2, 3, and 4 days and harvested for Western analyses of specified factors.
Figure 2
Figure 2. Fip1 regulates the alternative polyadenylation (APA) of potential self-renewal factors in ESCs

Direct RNA sequencing analysis of APA in ESCs 4 days after they were transfected with siControl or siFip1-2. Log2 (proximal/distal ratio) are plotted for ESCs transfected with control siRNA (y-axis) and siFip1-2 (x-axis). Statistically significant (P < 10−4, Fisher's exact test) changes are colored in blue (PtoD: proximal-to-distal shift) or red (DtoP: distal-to-proximal shift). The numbers of PtoD and DtoP APA changes are shown in a column graph (inset).

UCSC genome browser tracks showing the Direct RNA sequencing results for the Ahctf1 and Ncaph2 genes in ESC with control (top track) or Fip1 KD (bottom track).

The expression levels of all genes (left panel) or Fip1 PtoD APA genes (right panel) during ESC embryoid body (EB) differentiation time course (0–96 h, x-axis). Genes are ordered based on their expression levels in ESCs, and the standardized expression values are plotted as a heat map.

Gene ontology analysis of 311 PtoD Fip1 APA targets using Ingenuity © analysis. Functional categories (y-axis) and the corresponding P-values (x-axis) are shown.

A Venn diagram showing the overlap between APA changes induced by Fip1 KD and those observed during ESC differentiation into neurons. The blue circles are PtoD genes and the red DtoP. The numbers of genes in each area are marked.

Figure 3
Figure 3. Fip1-mediated alternative polyadenylation (APA) regulation modulates the expression of self-renewal factors in ESCs

Luciferase reporter assays to determine the impact of the cUTRs (in blue) or c+aUTR (in green) on gene expression. ESCs were transfected with a reporter construct containing the firefly luciferase (Fluc) coding sequence linked to the specified 3′ UTRs and a control Renilla luciferase (Rluc) construct, and the Fluc/Rluc ratio was determined 2 days post-transfection. The Fluc/Rluc ratio (relative expression, y-axis) is normalized to cUTR-containing construct and plotted as mean ± s.e.m. (n = 3). n.s.: not significant; *P < 0.05.

Western blot analyses of the Fip1 APA targets using lysates prepared from ESCs 4 days after they were transfected with the specified siRNAs.

Fip1 APA target genes are required for ESC self-renewal. ESCs were transfected with specified siRNAs and cellular morphology was imaged 4 days post-transfection. A second siRNA was used to deplete each gene in ESC and the cell images are shown in Supplementary Fig 15B.

Figure 4
Figure 4. Fip1 is required for somatic reprogramming

RT–qPCR analyses of Fip1 levels (x-axis) at the specified time during reprogramming (y-axis). Fip1 levels are normalized to day 0 level and plotted as mean ± s.e.m. (n = 3).

Alternative polyadenylation (APA) changes of Fip1 targets during reprogramming. RT–qPCR analyses of aUTR usage in Fip1 APA genes at the specified time during reprogramming. aUTR usage values are normalized to the day 0 value and plotted as mean ± s.e.m. (n = 3).

The growth curves of ESCs (left) or MEFs transfected with control (blue line) or Fip1 si/shRNA (red line).

Impact of Fip1 silencing on somatic cell reprogramming. The number of AP+ colonies was counted 12 days after the introduction of Oct4, Sox2, Klf4, and Myc by lentivirus transduction of MEF. Top panel: the AP staining image of MEFs induced to reprogram after transfected with control or Fip1-shRNA. Bottom panel: quantification results of APA-positive colonies are plotted as mean ± s.e.m. (n = 3).

Reprogramming efficiency was determined by the percentage of Oct4GFP-positive cells on day 12 by FACS analysis. GFP-positive cells are boxed (Top panel). Bottom panel: Quantification results of GFP-positive iPSCs based on the FACS data are plotted as mean ± s.e.m. (n = 3). **P < 0.01.

Figure 5
Figure 5. Mechanisms for Fip1-mediated alternative polyadenylation (APA) regulation

Fip1 controls mRNA 3′ processing activity in vivo. Top panel: UCSC Genome Browser track showing the 3′ end of Rpl26 mRNA and the positions of the RT–qPCR primers. Bottom panel: Relative transcription read-through is measured by the ratio between uncleaved (amplified by forward and reverse 2 primers) and the total transcripts (forward and reverse 1 primers) based on RT–qPCR results, and normalized to the value in siControl and plotted as mean ± s.e.m. (n = 3).

In vitro cleavage/polyadenylation (top) and cleavage (bottom) assays using L3 RNA substrate and nuclear extract from mock or Fip1 KD cells. Pre-mRNAs, poly(A)+, and 5′ cleavage products are marked.

PAS activities from the proximal (blue) or distal (green) PAS of the specified genes (x-axis) were determined by transfecting pPASPORT constructs into ESCs and measuring the Rluc/Fluc ratio 1 day after transfection, normalized by the value of the distal PASs, and plotted as mean ± s.e.m.

Distribution of Fip1 binding sites (blue line, based on iCLIP-seq data) and cleavage sites (green line, based on DRS data) relative to the closest upstream A(A/U)UAAA. Position 0 represents the 5′ end of AAUAAA.

Fip1 iCLIP maps for the cUTR and aUTRs of PtoD (blue), DtoP (red), and non-target genes (gray). Fip1 iCLIP signals are first normalized by transcript numbers (DRS read counts). cUTRs and aUTRs are divided into 100 bins each, and the summation of iCLIP signals in each bin for all genes in the group is divided by the number of genes. The normalized iCLIP signals (y-axis) are plotted for cUTRs and aUTR regions (x-axis). The difference in the density of the iCLIP peaks between PtoD and DtoP genes is likely due to the different numbers of genes in these groups.

A box plot showing the distribution of the distances between the proximal and distal PASs (within the same exons) for PtoD, DtoP, and non-target genes. The red lines mark the median values: 987 nt for PtoD genes, 205 nt for DtoP genes, and 759 nt for non-targets.

Figure 6
Figure 6. A model for Fip1-mediated alternative polyadenylation (APA) regulation

For APA genes whose proximal and distal PASs are far from each other, there is a significant lag between the times when the alternative PASs are transcribed. Higher levels of Fip1/CPSF (such as in ESCs and iPSCs) promote the recognition of weaker proximal PASs. When Fip1/CPSF levels are low, transcription read-through the usage of stronger distal PASs increases.

For APA genes whose proximal and distal PASs are close to each other, these alternative PASs directly compete for binding to mRNA 3′ processing factors. At high levels, Fip1/CPSF binding to the region between the two PASs may block CstF binding in the same region and the recognition of the proximal PAS. At lower levels, Fip1/CPSF binding decreases, which allow CstF binding to this region, leading to better recognition of the proximal PAS. See Discussion for more details.

Similar articles

See all similar articles

Cited by 46 articles

See all "Cited by" articles

Publication types