Zfp206, Oct4, and Sox2 are integrated components of a transcriptional regulatory network in embryonic stem cells

Abstract

Zfp206 (recently renamed Zscan10) encodes a zinc finger transcription factor specifically expressed in human and mouse embryonic stem cells (ESC). It has been shown that Zfp206 is required to maintain ESC in an undifferentiated, pluripotent state. Presented here are data showing that Zfp206 works together with two other transcription factors, Oct4 and Sox2, which are also essential regulators of ESC pluripotency. We show that Zfp206 binds to the Oct4 promoter and directly regulates Oct4 expression. Genome-wide mapping of Zfp206-binding sites in ESC identifies more than 3000 target genes, many of which encode transcription factors that are also targeted for regulation by Oct4 and Sox2. In addition, we show that Zfp206 physically interacts with both Oct4 and Sox2. These data demonstrate that Zfp206 is a key component of the core transcriptional regulatory network and together with Oct4 and Sox2 regulates differentiation of ESC.

FIGURE 1.

Mapping of Zfp206-binding sites at the Oct4 promoter. A, two binding sites for Zfp206 were identified within the enhancer/promoter region of Oct4. Chromatin immunoprecipitations were performed in ESC using IgG, a V5 antibody, or an antibody specific to Zfp206. ChIP enrichments were measured by Q-RT-PCR and expressed as fold differences, relative to non-IP (input) chromatin. Positions of the 14 regions (probe numbers 1–14) amplified are indicated as distances (−6 to 3 kb) from the transcriptional start site (TSS) of the Oct4 gene. B, the transcriptional activity of the Oct4 promoter is dependent on Zfp206-binding sites. A luciferase reporter assay was used to measure activity of the Oct4 promoter in ESC that expressed shRNA against Zfp206, Dax1, and Rcor1. The relative luciferase activity is compared with a vector that expressed no shRNA. Knockdown of Zfp206 resulted in a significant decrease in Oct4 promoter activity (*, p < 0.01 by Student's t test). Knockdown of Dax1 or Rcor1 had no significant affect on Oct4 promoter activity.

FIGURE 2.

Validation of selected Zfp206 target genes. ChIP-PCR was performed on 22 binding sites identified by ChIP-chip. The ChIP-PCR experiments confirmed enrichments of Zfp206 binding at all 22 sites. Shown are the ChIP enrichments (fold differences relative to input DNA) using IgG, a V5 antibody, and a Zfp206-specific antibody. The primers used in this study are listed in the supplemental data.

FIGURE 3.

Consensus DNA motif for Zfp206-binding sites. A consensus binding site motif for Zfp206 was computationally determined by analysis of ChIP-chip data. A logo of the newly defined 10-base DNA-binding motif is presented. The motif was determined by the Weeder algorithm (17) using the top 500 binding regions identified by ChIP-chip.

FIGURE 4.

Frequent co-targeting of genes by Zfp206 and other pluripotency transcription factors. The Venn diagrams show the number of Zfp206 target genes that are also targets of Zfp281, Nanog, Oct4, and Sox2. The results indicate statistically significant overlap of Zfp206 target genes and genes targeted also by pluripotency transcription factors, Zfp281, Sox2, Nanog, and Oct4.

FIGURE 5.

Zfp206 binding co-localizes with Zfp281, Oct4, Sox2, and Nanog. A, using Zfp206 sites as reference, the nearest binding sites for each of the other TF were calculated. The results indicate substantial overlap of Zfp206 sites with sites for Zfp281 and Oct4 but less so with Sox2 and Nanog. The data are expressed as frequencies of co-occupancies for Zfp206 with the indicated TF at interval distances indicated from the Zfp206-binding site. B, examples of close proximities of Oct4- (blue bars), Sox2- (green bar), and Zfp206 (red bar)-binding sites at three target genes, Ssrp1, Stmn1, and Cited2.

FIGURE 6.

Zfp206 activates or represses its target genes. Luciferase reporter constructs containing the promoter and flanking Zfp206-binding site of the indicated Zfp206 target genes were transfected into Zfp206 knockdown or control ESC. For Klf4, the intronic Zfp206-binding site from intron-1 was linked to a minimal promoter to drive luciferase. Knockdown of Zfp206 expression by two shRNA constructs resulted in decreased expression of four genes, increased expression of three genes, and no change in three others. The luciferase activities are expressed as fold differences relative to an empty vector control. The mean values are shown with error bars (S.D.) from four independent assays. The statistical significance was assessed by comparing data to empty vector control, using Student's t test (*, p < 0.01).

FIGURE 7.

Zfp206 binding is necessary for activation or repression of target gene expression. The luciferase reporter constructs for Jarid1c and mir-124-a2 promoters were deleted of their Zfp206-binding sites. The activities of the truncated and full versions were compared. Removal of the Zfp206-binding sites resulted in a loss of regulation in response to Zfp206 knockdown by shRNA. An empty shRNA vector and shRNA direct against Dax-1 had no effect on Zfp206 target genes. The luciferase activities are expressed as fold differences relative to an empty vector control. The mean values are shown with error bars (S.D.) from four independent assays. Statistical significance was assessed by comparing data with empty vector control, using Student's t test (*, p < 0.01).

FIGURE 8.

Zfp206 regulates expression of target genes in somatic cells. The luciferase reporter constructs described in Fig. 6 were co-transfected into HEK293T somatic cells together with a Zfp206 expression construct. The transcriptional responses to Zfp206 overexpression were perfectly opposed to the results seen by Zfp206 knockdown by shRNA treatment shown in Fig. 6. The luciferase activities are expressed as fold differences relative to HEK293 cells co-transfected with an empty expression vector. The mean values are shown with error bars (S.D.) from four independent assays. Statistical significance was assessed by comparing data to empty vector control, using Student's t test (*, p < 0.01).

FIGURE 9.

Zfp206 physically interacts with Oct4 and Sox2 proteins. ESC were transfected to express V5 epitope tagged Oct4 (Oct4-V5) (A) or Sox2 (Sox2-V5) (B). Zfp206 antibody was used to immunoprecipitate (IP) Zfp206 and the co-immunoprecipitation products were analyzed by Western blot (WB) with a V5 antibody to test whether Zfp206 interact with Oct4 and Sox2. V5 antibody was used to immunoprecipitate Oct4-V5 (A) and Sox2-V5 (B), and co-immunoprecipitation products were analyzed by Western blot with a Zfp206 antibody. The results show co-IP of Zfp206 with Oct and co-IP of Zfp206 with Sox2. No co-IP of Oct4 or Sox2 was observed in control experiments using an antibody directed against the ESC-expressed transcription factor Rcor1.