Post-transcriptional gene expression regulation is crucial for many diverse cellular processes, such as development, metabolism and cancer. The fate of hundreds of thousands of mRNA molecules in eukaryotic cells is likely to be coordinated and regulated by hundreds of RNA binding proteins and noncoding RNAs (for example, microRNAs). To shed light on the importance and roles of individual RNA binding proteins, it is necessary to identify the spectrum of targets recognized and associated with these RNA binding proteins. Genome-wide unbiased methods have begun to reveal the plethora of targets and diverse rules by which the post-transcriptional regulatory networks are controlled26
Here we have identified the splicing factor FOX2 as being highly expressed in the nuclei of pluripotent hESCs. hESCs constitute an excellent in vitro
model for survival, self-renewal, differentiation and development. Using a modified CLIP-seq technology and computational analyses that accounted for gene-specific variation in RNA abundance, we have uncovered thousands of FOX2 RNA targets representing ~7% of the human genes in hESCs. Confirming and extending previous computational analyses of human intronic regions14–16,18
, we observed that FOX2 was preferentially bound near alternative splice sites, and the binding sites were located within regions of higher evolutionarily conservation. Experimental validation of targets revealed that FOX2 represses exon usage when bound upstream and enhances exon inclusion when located downstream of the alternative exon, revealing an RNA map for the FOX2-mediated alternative splicing program in hESCs. Our study presenting in vivo
targets of FOX2 in a biological system strengthens computational predictions from otherwise indistinguishable conserved FOX1 and FOX2 sites27
, as both FOX1 and FOX2 recognize the same RNA element6,11,28
. The fact that FOX2 is also expressed in differentiated neural progenitors from hESCs and fetal neural stem cells but was not shown to regulate alternative splicing the same way in hESCs, despite having conserved binding sites in the same transcribed pre-mRNA, underscores the importance of experimentally identifying in vivo
targets in the appropriate cell and tissue context.
The finding that many FOX2 targets are themselves splicing regulators leads to the provocative possibility that FOX2 may function as an upstream regulator of many general and tissue-specific splicing regulators. In addition, we identified FOX2 binding within the FOX2 pre-mRNA itself and, combined with RT-PCR data, demonstrated direct evidence for autoregulation of the FOX2 gene. The alternative splicing of the FOX2 pre-mRNA may result in unique target pre-mRNA splicing regulation; this possibility deserves further attention in the future.
Last, our preliminary results indicate that FOX2 has an important role in maintaining the viability of hESCs, as depletion of FOX2 led to rapid cell death. Given the many genes controlled by FOX2 in hESCs, it is presently unclear which gene(s) or alternative splicing event(s) is responsible for the lethal phenotype. It is possible that the phenotype is a result of the combined effect of multiple affected genes. Given our observation that FOX2 may function as a master regulator of the alternative splicing program in hESCs and signaling pathways, it may be likely that many events contribute to the phenotype, that is, it may be unrealistic to think that the complex cellular mortality phenotype could be due to a single altered gene product. Nevertheless, the phenotype is remarkably specific to hESCs, and not other cell lines such as 293T or 3T3. More notably, neither neural progenitors derived from hESCs nor primary human fetal neural stem cells were sensitive to FOX2 depletion, suggesting that FOX2 has a different set of targets and, hence, a dissimilar RNA map in other cell types. Our study provides a starting point for the future characterization of the varying target repertoire of the same splicing factor in different biological systems, embracing a need to understand the uniqueness of factor-target relationships throughout biology.