Nature advance online publication 17 September 2008 | doi:10.1038/nature07299; Received 10 April 2008; Accepted 30 July 2008; Published online 17 September 2008
MicroRNAs to Nanog, Oct4 and Sox2 coding regions modulate embryonic stem cell differentiation
Yvonne Tay1,4, Jinqiu Zhang1, Andrew M. Thomson1,5, Bing Lim1,2,5 & Isidore Rigoutsos3,5
1. Stem Cell and Developmental Biology, Genome Institute of Singapore, Agency for Science Technology and Research (A*STAR), #08-01, Genome, 60 Biopolis Street, Singapore 138672, Singapore
2. Beth-Israel Medical Center, Harvard Medical School, 77 Avenue Louis Pasteur, Boston, Massachusetts 02215, USA
3. Bioinformatics and Pattern Discovery Group, IBM Thomas J. Watson Research Center, Yorktown Heights, PO Box 218, New York 10598, USA
4. Present address: Experimental Therapeutics Center, Agency for Science Technology and Research (A*STAR), Nanos Level 3, 31 Biopolis Way, Singapore 138669, Singapore.
5. These authors contributed equally to this work.
MicroRNAs (miRNAs) are short RNAs that direct messenger RNA degradation or disrupt mRNA translation in a sequence-dependent manner1, 2, 3, 4, 5, 6, 7. For more than a decade, attempts to study the interaction of miRNAs with their targets were confined to the 3' untranslated regions of mRNAs1, fuelling an underlying assumption that these regions are the principal recipients of miRNA activity. Here we focus on the mouse Nanog, Oct4 (also known as Pou5f1) and Sox2 genes8, 9, 10, 11 and demonstrate the existence of many naturally occurring miRNA targets in their amino acid coding sequence (CDS). Some of the mouse targets analysed do not contain the miRNA seed, whereas others span exon–exon junctions or are not conserved in the human and rhesus genomes. miR-134, miR-296 and miR-470, upregulated on retinoic-acid-induced differentiation of mouse embryonic stem cells, target the CDS of each transcription factor in various combinations, leading to transcriptional and morphological changes characteristic of differentiating mouse embryonic stem cells, and resulting in a new phenotype. Silent mutations at the predicted targets abolish miRNA activity, prevent the downregulation of the corresponding genes and delay the induced phenotype. Our findings demonstrate the abundance of CDS-located miRNA targets, some of which can be species-specific, and support an augmented model whereby animal miRNAs exercise their control on mRNAs through targets that can reside beyond the 3' untranslated region.