Elevating the levels of Sox2 in embryonal carcinoma cells and embryonic stem cells inhibits the expression of Sox2:Oct-3/4 target genes

Abstract

Recent studies have identified large sets of genes in embryonic stem and embryonal carcinoma cells that are associated with the transcription factors Sox2 and Oct-3/4. Other studies have shown that Sox2 and Oct-3/4 work together cooperatively to stimulate the transcription of their own genes as well as a network of genes required for embryogenesis. Moreover, small changes in the levels of Sox2:Oct-3/4 target genes alter the fate of stem cells. Although positive feedforward and feedback loops have been proposed to explain the activation of these genes, little is known about the mechanisms that prevent their overexpression. Here, we demonstrate that elevating Sox2 levels inhibits the endogenous expression of five Sox2:Oct-3/4 target genes. In addition, we show that Sox2 repression is dependent on the binding sites for Sox2 and Oct-3/4. We also demonstrate that inhibition is dependent on the C-terminus of Sox2, which contains its transactivation domain. Finally, our studies argue that overexpression of neither Oct-3/4 nor Nanog broadly inhibits Sox2:Oct-3/4 target genes. Collectively, these studies provide new insights into the diversity of mechanisms that control Sox2:Oct-3/4 target genes and argue that Sox2 functions as a molecular rheostat for the control of a key transcriptional regulatory network.

Figure 1.

Effect of Sox2 Oct-3/4 and Sox11 overexpression on Sox2 promoter activity. Schematic of the Sox2 promoter/reporter construct is shown. The gray boxes indicate critical enhancers, the basal Sox2 promoter is represented by a black box adjacent to the CAT reporter gene, and the relative HMG/POU cassette is indicated with an asterisk. F9 EC cells were transiently transfected and assayed as described in the Materials and methods section. The cells were transfected with 12 µg of the Sox2 promoter/reporter gene construct (Sox2 + US/DSE) plus 1 µg of the CMV-β-gal expression vector. In (A), 1 or 3 µg of either the Sox2 or Oct-3/4 expression vector was co-transfected. In (B), 3 µg of the indicated expression vector (Sox2, Oct-3/4 or Sox11) was co-transfected. Reporter activity was normalized to that of β-galactosidase and the activity of Sox2 + US/DSE (with no overexpression) was set to 1. The data shown represents the mean and standard deviation of duplicate samples from a representative experiment. Each experiment was repeated twice and similar results were obtained.

Figure 2.

Effect of Sox2 overexpression on multiple Sox2:Oct-3/4 target gene promoters. Schematics of each of the promoter/reporter constructs are shown. The gray boxes indicate critical enhancers, the basal promoter is represented by a black box adjacent to the CAT reporter gene, and the relative HMG/POU cassette is indicated with an asterisk. F9 EC cells were transfected with 12 µg of the indicated promoter/reporter gene construct plus 1 µg of the CMV-β-gal expression vector. Where indicated, 1 or 3 µg of the Sox2 expression vector was co-transfected. Reporter activity was normalized to that of β-galactosidase and the activity of each individual promoter/reporter construct (with no Sox2 overexpression) was set to 1. Overexpression values are shown relative to each particular gene construct. The data shown represents the mean and standard deviation of duplicate samples from a representative experiment. This experiment was repeated twice and similar results were obtained.

Figure 3.

Effect of Sox2 plus Oct-3/4 overexpression on the Sox2 and FGF-4 promoters. Schematics of each of the promoter/reporter constructs are shown. The gray boxes indicate critical enhancers, the basal promoter is represented by a black box adjacent to the CAT reporter gene, and the relative HMG/POU cassette is indicated with an asterisk. F9 EC cells were transfected with 12 µg of either (A) Sox2 + US/DSE or (B) FGF-4 + E plus 1 µg of the CMV-β-gal expression vector. Where indicated, 1, 2 or 3 µg of the Sox2 and/or the Oct-3/4 expression vector was co-transfected. Reporter activity was normalized to that of β-galactosidase and the activity of the reporter construct (with no overexpression) was set to 1. Overexpression values are shown relative to the given promoter/reporter construct. The data shown represents the mean and standard deviation of duplicate samples from representative experiments. Each experiment was repeated twice and similar results were obtained.

Figure 4.

Effect of Sox2 overexpression on endogenous Sox2:Oct-3/4 target gene expression. (A) F9 EC cells were transfected with either GFP or GFP-Sox2, and GFP positive cells were collected and processed as described in the Materials and methods section. Between these two cell populations, the levels of expression of five genes were compared after normalizing the values to the expression of GAPDH. Target gene expression in GFP-Sox2-transfected cells is reported as a fraction of the corresponding expression in control cells (GFP-transfected). The values shown represent the mean and standard deviation after averaging values from two separate isolates of GFP positive cells and at least two qPCR assays of each gene. (B) Western blot analysis of Sox2 protein in F9 EC cells transfected with the GFP-Sox2 expression plasmid and collected by flow cytometry. GFP-Sox2 and endogenous Sox2 are indicated.

Figure 5.

Effect of Sox2 overexpression on Sox2:Oct-3/4 target gene constructs containing mutant HMG/POU cassettes. Schematics of each of the promoter/reporter constructs are shown. The gray boxes indicate critical enhancers, the basal promoter is represented by a black box adjacent to the CAT reporter gene, and the relative HMG/POU cassette (or mutant) is indicated with an asterisk. (A) In each promoter/reporter construct, the HMG/POU cassette was destroyed. (B) Where indicated, the POU motif in the HMG/POU cassette was mutated. F9 EC cells were transfected with 12 µg of the indicated promoter/reporter gene construct plus 1 µg of the CMV-β-gal expression vector. Where indicated, 1 or 3 µg of the Sox2 expression vector was co-transfected. Reporter activity was normalized to that of β-galactosidase and the activity of each individual promoter/reporter construct (with no overexpression) was set to 1. Overexpression values are shown relative to each particular gene construct. The data shown represents the mean and standard deviation of duplicate samples from representative experiments. This experiment was repeated twice and similar results were obtained.

Figure 6.

Effect of Sox2 overexpression on the HSV TK promoter. Schematics of each of the promoter/reporter constructs are shown. The gray box indicates the FGF-4 enhancer (HMG/POU cassette is indicated with an asterisk) and the basal TK promoter is represented by a black box adjacent to the CAT reporter gene. F9 EC cells were transfected with 12 µg of the indicated promoter/reporter gene construct plus 1 µg of the CMV-β-gal expression vector. Where indicated, 3 µg of the Sox2 expression vector was co-transfected. Reporter activity was normalized to that of β-galactosidase and the activity of the TK promoter/reporter construct (with no enhancer and no overexpression) was set to 1. All other values are shown relative to that particular gene construct (TK). The data shown represents the mean and standard deviation of duplicate samples from a representative experiment. This experiment was repeated twice and similar results were obtained.

Figure 7.

Effect of chimeric Sox2/11 overexpression on Sox2 and Nanog promoter activity. (A) Schematics of chimeric Sox constructs. Sox2-2-11 contains the Sox2 DBD coupled with the Sox11 TAD whereas Sox11-11-2 contains the Sox11 DBD coupled with the Sox2 TAD. (B) Effect of chimeric Sox overexpression on Sox2 promoter activity. (C) Effect of chimeric Sox overexpression on Nanog promoter activity. In both (B) and (C), a schematic of the promoter/reporter construct is shown—the gray boxes indicate critical enhancers, the basal promoter is represented by a black box adjacent to the CAT reporter gene, and the relative HMG/POU cassette is indicated with an asterisk. F9 EC cells were transfected with 12 µg of the indicated promoter/reporter gene construct plus 1 µg of the CMV-β-gal expression vector. Where indicated, 3 µg of either the Sox2, Sox2-2-11 or Sox11-11-2 expression vector was co-transfected. Reporter activity was normalized to that of β-galactosidase and the activity of each individual promoter/reporter construct (with no overexpression) was set to 1. Overexpression values are shown relative to each particular gene construct. The data shown represents the mean and standard deviation of duplicate samples from a representative experiment. Each experiment was repeated twice and similar results were obtained.

Figure 8.

Effect of Sox2 overexpression on Sox2:Oct-3/4 target genes in P19 EC cells and D3 ES cells. (A) Effects of Sox2 overexpression on Sox2, FGF-4, Nanog, UTF1 and Oct-3/4 promoter/reporter gene constructs in P19 EC cells. (B) Effects of Sox2 overexpression on Sox2, FGF-4, Nanog and Oct-3/4 promoter/reporter gene constructs in D3 ES cells. Schematics of each of the promoter/reporter constructs are shown. The gray boxes indicate critical enhancers, the basal promoter is represented by a black box adjacent to the CAT reporter gene, and the relative HMG/POU cassette is indicated with an asterisk. In (A), P19 EC cells were transfected with 12 µg of the indicated promoter/reporter gene construct plus 1 µg of the CMV-β-gal expression vector. Where indicated, the cells were also transfected with 1 or 3 µg of the Sox2 expression vector. In (B), D3 ES cells were transfected with 4.5 µg of the indicated promoter/reporter gene construct plus 1 µg of the CMV-β-gal expression vector. Where indicated, the cells were also transfected with 0.3 or 1 µg of the Sox2 expression vector. In both (A) and (B), reporter activity was normalized to that of β-galactosidase and the activity of each individual promoter/reporter construct (with no Sox2 overexpression) was set to 1. Overexpression values are shown relative to each particular gene construct. The data shown represents the mean and standard deviation of duplicate samples from a representative experiment. Each experiment was repeated and similar results were obtained.

Figure 9.

Effect of Oct-3/4 and Nanog overexpression on promoter activity. (A) Effect of Oct-3/4 overexpression on Oct-3/4 promoter activity. (B) Effect of Nanog overexpression on Nanog and Sox2 promoter activity. Schematics of each of the promoter/reporter constructs are shown. The gray boxes indicate critical enhancers, the basal promoter is represented by a black box adjacent to the CAT reporter gene, and the relative HMG/POU cassette is indicated with an asterisk. F9 EC cells were transfected as indicated with 12 µg of either the Oct-3/4, Nanog or Sox2 promoter/reporter gene constructs (Oct-3/4 + E, Nanog + E or Sox2 + US/DSE) plus 1 µg of the CMV-β-gal expression vector. Where indicated, 1 or 3 µg of the Oct-3/4, Sox2 or Nanog expression vector was co-transfected. Reporter activity was normalized to that of β-galactosidase and the activity of each individual promoter/reporter construct (with no overexpression) was set to 1. Overexpression values are shown relative to each particular gene construct. The data shown represents the mean and standard deviation of duplicate samples from a representative experiment. Each experiment was repeated twice and similar results were obtained.

Figure 10.

Model of Sox2, Oct-3/4 and Nanog feedforward and feedback gene regulation. Sox2 is shown as a central node in a network of genes critical to embryogenesis as well as stem cell self-renewal and pluripotency. (A) At normal levels, Sox2 and Oct-3/4 activate transcription of all known Sox2:Oct-3/4 target genes. (B) Increasing levels of Sox2 leads to inhibition of target gene expression. (C) Increasing levels of Oct-3/4 inhibits expression of only Oct-3/4 and Nanog, and high levels of Nanog lead to inhibition of Nanog gene expression, but not other Sox2:Oct-3/4 target genes. Presumably, there are additional Sox2:Oct-3/4 target genes that are regulated in a similar fashion.