You are here

| Pluripotent Stem Cells

Conversion of Sox17 into a pluripotency reprogramming factor by reengineering its association with Oct4 on DNA --



From the June 2011 Issue of Stem Cells
By Carla Mellough

The selective dimerization of transcription factors with binding partners can allow those factors lacking site-specific DNA binding capability to elicit transcriptional control. This type of synergistic action between the Sox and POU (Oct) family members of transcription factors is well known. Co-operation between various members facilitates numerous key regulatory roles, for example the well documented interaction between Sox2 and Oct4 in stem cells and more recently the cooperation of Sox17 with Oct4 for mesodermal development.1 Interestingly, both Sox2 and Sox17 interact with Oct4 during early development, yet both yield different effects and Sox17 cannot replace Sox2 as a pluripotency or reprogramming factor. An article published in the June issue of Stem Cells by Jauch et al.2, now begins to unravel the mechanisms underlying the competitive interactions between Sox2 and Sox17 for binding to Oct4.

From x-ray crystallography analysis the authors had previously observed that the electrostatic interface of the Oct4 interaction surface of the DNA binding domain differed between Sox2 and Sox17.3 Given that the early inner mass cells co-express Oct4/Sox2/Sox17, this suggests that Sox2 and Sox17 might compete for Oct4, forming complexes on specific genomic regions. Their current work set out to identify distinct motif configurations in genomic regions binding Sox2 and Oct4 in mouse embryonic stem cells (mESC).

The authors identified different sox/oct motif configurations by screening regions of the mouse genome shown from chromatin immunoprecipitation (ChIP)-seq data to be occupied by Oct4/Sox2 in mESCs. They then constructed variants of the sox/oct motif configuration by inserting base pairs, making reverse compliment versions or altering spacer base pair sequences. They found the canonical motif to be most enriched, followed by a novel ‘compressed’ motif, which differed from the canonical by a single nucleotide deletion only. These motifs were most commonly found at sites specific for co-bound transcription factors. Screening the heterodimerization potential of Sox2 and Oct4 on different sox/oct motifs revealed a strong preference for Sox2 and Oct4 interaction on the canonical element and this was diminished on constructed variants of this motif. The compressed motif did not allow Sox2/Oct4 heterodimer formation, but instead enabled the co-operative binding of Sox17 and Oct4. Competitive binding experiments confirmed these results and revealed that the heterodimerization of each of the Sox transcription factors with Oct4 at their preferred sox/oct element was not abrogated in the presence of the other (i.e. Sox17 outcompetes Sox2 at the compressed element). Jauch et al.2 propose that these binding differences might be one way in which different Sox transcription factors elicit their diverse roles during development.

The authors then investigated the basis for the distinct Oct4 interaction surfaces on Sox2 (a B group Sox protein) and Sox17 (an F group Sox protein). A basic lysine to acidic glutamate substitution at the Oct4 contact interface exists in F group members. Site directed mutagenesis by amino acid substitutions which acted to replace lysine with glutamate or vice versa, however, swapped the binding preferences for Sox2/Oct4 and Sox17/Oct4 on sox/oct motifs. Additionally, ChIP and luciferase reporter assays revealed that these point mutations also affected transcription factor recruitment and gene expression.

But can this glutamate to lysine substitution of Sox17 protein allow it to go one step further, enabling it to induce pluripotency like its family member Sox2? To investigate this, Jauch et al.2 set out to generate induced pluripotent stem cells (iPSC) by transfecting mouse fibroblasts with Oct4, c-Myc and Klf4 (OCK) either with wild type Sox or Sox variants. They demonstrate that modification of Sox17 to contain a lysine at the Oct4 interaction surface enabled the generation of iPSCs with three fold greater efficiency when compared with wild type Sox2, while modified Sox2 (where lysine was replaced by glutamate) lost its ability to reprogram. Furthermore, the authors generated Sox variant expressing cell lines to show that the lysine to glutamate substitution in modified Sox2 not only abrogated its usual reprogramming capacity, but caused it to gain the ability to induce endoderm differentiation in ESCs, just like native Sox17.

The fact that the single point mutation created within this study allowed a complete role reversal to occur between Sox2 and Sox17 is a fascinating result. This work provides new information on transcription factor heterodimerization and binding dynamics, in this case for two widely acting families of transcription factors, shedding light on their mechanism of action during development. A greater understanding of transcription factor dynamics and function will no doubt unravel their complex and diverse roles during development and enable improved control of gene expression for innumerable applications.



1 Stefanovic S, Abboud N, Desilets S et al. Interplay of Oct4 with Sox2 and Sox17: A molecular switch from stem cell pluripotency to specifying a cardiac fate. J Cell Biol. 2009;186:665–673.

2 Jauch R, Aksoy I, Hutchins AP, Ng CK, Tian XF, Chen J, Palasingam P, Robson P, Stanton LW, Kolatkar PR. Conversion of sox17 into a pluripotency reprogramming factor by reengineering its association with oct4 on DNA. Stem Cells. 2011;29(6):940-51. doi: 10.1002/stem.639.

3 Palasingam P, Jauch R, Ng CK et al. The structure of Sox17 bound to DNA reveals a conserved bending topology but selective protein interaction platforms. J Mol Biol. 2009;388:619–630.