Transcriptional coactivator PGC-1α contributes to decidualization by forming a histone-modifying complex with C/EBPβ and p300

Abstract

We previously reported that CCAAT/enhancer-binding protein beta (C/EBPβ) is the pioneer factor inducing transcription enhancer mark H3K27 acetylation (H3K27ac) in the promoter and enhancer regions of genes encoding insulin-like growth factor-binding protein-1 (IGFBP-1) and prolactin (PRL) and that this contributes to decidualization of human endometrial stromal cells (ESCs). Peroxisome proliferator-activated receptor gamma coactivator 1-alpha (PGC-1α; PPARGC1A) is a transcriptional coactivator known to regulate H3K27ac. However, although PGC-1α is expressed in ESCs, the potential role of PGC-1α in mediating decidualization is unclear. Here, we investigated the involvement of PGC-1α in the regulation of decidualization. We incubated ESCs with cAMP to induce decidualization and knocked down PPARGC1A to inhibit cAMP-induced expression of IGFBP-1 and PRL. We found cAMP increased the recruitment of PGC-1α and p300 to C/EBPβ-binding sites in the promoter and enhancer regions of IGFBP-1 and PRL, corresponding with increases in H3K27ac. Moreover, PGC-1α knockdown inhibited these increases, suggesting PGC-1α forms a histone-modifying complex with C/EBPβ and p300 at these regions. To further investigate the regulation of PGC-1α, we focused on C/EBPβ upstream of PGC-1α. We found cAMP increased C/EBPβ recruitment to the novel enhancer regions of PPARGC1A. Deletion of these enhancers decreased PGC-1α expression, indicating that C/EBPβ upregulates PGC-1α expression by binding to novel enhancer regions. In conclusion, PGC-1α is upregulated by C/EBPβ recruitment to novel enhancers and contributes to decidualization by forming a histone-modifying complex with C/EBPβ and p300, thereby inducing epigenomic changes in the promoters and enhancers of IGFBP-1 and PRL.

Keywords: CCAAT/enhancer-binding protein beta (C/EBPβ); E1A-binding protein p300 (p300); decidualization; endometrial stromal cell; epigenetics; genome editing; histone acetylation; peroxisome proliferator–activated receptor gamma coactivator 1-alpha (PGC-1α).

Figure 1

PGC-1α expression in human endometrium and ESCs and the effect of cAMP on PGC-1α expression.A, immunohistochemical expression of PGC-1α in the late proliferative and secretory phase endometrium and decidua of early pregnancy. The photographs in the lower row are negative controls in each sample. Scale bars, 50 μm. B, ESCs were treated with or without cAMP (0.5 mM) for 4 days. Cells treated without cAMP were used as the control. mRNA expression was analyzed by quantitative real-time RT-PCR. Values of PGC-1α, IGFBP-1, and PRL were normalized to those of MRPL19 and expressed as a ratio of the control sample. SDs of control samples are shown in Table S4. Values are mean ±SD of three different incubations. A, p < 0.01 versus control. Whole-cell lysates were prepared and subjected to Western blotting to examine PGC-1α protein expression. β-Tubulin was used as an internal control. The immunoblot is a representative of three different incubations. ESC, endometrial stromal cell; IGFBP-1, insulin-like growth factor–binding protein-1; PGC-1α, peroxisome proliferator-activated receptor gamma coactivator 1 alpha; PRL, prolactin.

Figure 2

Involvement of PGC-1α in cAMP-induced gene expression of IGFBP-1 and PRL.A, ESCs were transfected with a siRNA targeted against PGC-1α or with a nontargeting siRNA as a control. Forty-eight hours after siRNA transfection, ESCs were treated with or without cAMP for 4 days. Whole-cell lysates were prepared and subjected to Western blotting to examine the knockdown of PGC-1α protein expression. β-Tubulin was used as an internal control. The immunoblot is a representative of three different incubations. B, mRNA expression was analyzed by quantitative real-time RT-PCR. Values of IGFBP-1 and PRL were normalized to those of MRPL19 and expressed as a ratio of the control sample. SDs of control samples are shown in Table S4. Values are mean ±SD of three different incubations. A, p < 0.01 versus control treatment in the control siRNA; B, p < 0.01 versus cAMP treatment in the control siRNA. ESC, endometrial stromal cell; IGFBP-1, insulin-like growth factor–binding protein-1; PGC-1α, peroxisome proliferator–activated receptor gamma coactivator 1 alpha; PRL, prolactin.

Figure 3

cAMP-induced PGC-1α and p300 recruitment to the promoter and enhancer regions of IGFBP-1 and the promoter region of PRL.A, locations of the C/EBPβ-binding sites in the promoter and enhancer regions of IGFBP-1 and the promoter region of PRL. For the ChIP assay, three regions were amplified by PCR: -49 to -212 bp for the IGFBP-1 promoter, -6159 to -6298 bp for the IGFBP-1 enhancer, -124 to -363 bp for the PRL promoter. B, co-immunoprecipitation of PGC-1α, C/EBPβ, and p300. ESCs were treated with or without cAMP for 4 days. Whole-cell lysates were prepared and subjected to immunoprecipitation (IP) with a specific PGC-1α, C/EBPβ, and p300 antibody. Normal rabbit IgG was used as a negative control for IP. Immunoprecipitated proteins were separated on SDS-polyacrylamide gels together with samples from the input (10% amount used for IP). Immunoblot (IB) was done with anti-PGC-1α or C/EBPβ antibody. β-Tubulin was used as an internal control. C and D, recruitment levels of PGC-1α and p300 to the promoter and enhancer regions of IGFBP-1 and the promoter region of PRL were analyzed by ChIP assay. Primers were designed surrounding the C/EBPβ-binding sites. Normal rabbit IgG was used as a negative control. The relative recruitment levels were analyzed by real-time PCR. Values were expressed as a ratio of control sample. SDs of control samples are shown in Table S4. Data are mean ±SD of three different incubations. A, p < 0.01 versus control. B, p < 0.01 versus control. C/EBPβ, CCAAT/enhancer-binding protein beta; ESC, endometrial stromal cell; IGFBP-1, insulin-like growth factor–binding protein-1; PGC-1α, peroxisome proliferator–activated receptor gamma coactivator 1 alpha; PRL, prolactin.

Figure 4

Involvement of PGC-1α in the increase of H3K27ac levels in the promoter and enhancer regions of IGFBP-1 and the promoter region PRL. ESCs were transfected with a siRNA targeted against PGC-1α, C/EBPβ, or with a nontargeting siRNA as a control. Forty-eight days after siRNA transfection, ESCs were treated with or without cAMP for 4 days. The levels of p300 recruitment (A and D), H3K27ac (B), PGC-1α recruitment (C), and C/EBPβ recruitment (E) in the promoter and enhancer regions of IGFBP-1 and the promoter regions of PRL were analyzed by ChIP assay. The relative recruitment levels were analyzed by real-time PCR. Values were expressed as a ratio of cAMP treatment in the control siRNA sample. SDs of cAMP treatment in the control siRNA sample are shown in Table S4. Data are mean ±SD of three different incubations. A, p < 0.01 versus control treatment in the control siRNA; B, p < 0.01 versus cAMP treatment in the control siRNA. C, p < 0.05 versus control treatment in the control siRNA; D, p < 0.05 versus cAMP treatment in the control siRNA. F, whole-cell lysates were prepared and subjected to Western blotting to examine the effect of PGC-1α knockdown on C/EBPβ protein expression. β-Tubulin was used as an internal control. The immunoblot is a representative of three different incubations. C/EBPβ, CCAAT/enhancer-binding protein beta; ChIP, chromatin immunoprecipitation; ESC, endometrial stromal cell; H3K27ac, acetylation of histone-H3 lysine-27; IGFBP-1, insulin-like growth factor–binding protein-1; PGC-1α, peroxisome proliferator–activated receptor gamma coactivator 1 alpha; PRL, prolactin.

Figure 4

Involvement of PGC-1α in the increase of H3K27ac levels in the promoter and enhancer regions of IGFBP-1 and the promoter region PRL. ESCs were transfected with a siRNA targeted against PGC-1α, C/EBPβ, or with a nontargeting siRNA as a control. Forty-eight days after siRNA transfection, ESCs were treated with or without cAMP for 4 days. The levels of p300 recruitment (A and D), H3K27ac (B), PGC-1α recruitment (C), and C/EBPβ recruitment (E) in the promoter and enhancer regions of IGFBP-1 and the promoter regions of PRL were analyzed by ChIP assay. The relative recruitment levels were analyzed by real-time PCR. Values were expressed as a ratio of cAMP treatment in the control siRNA sample. SDs of cAMP treatment in the control siRNA sample are shown in Table S4. Data are mean ±SD of three different incubations. A, p < 0.01 versus control treatment in the control siRNA; B, p < 0.01 versus cAMP treatment in the control siRNA. C, p < 0.05 versus control treatment in the control siRNA; D, p < 0.05 versus cAMP treatment in the control siRNA. F, whole-cell lysates were prepared and subjected to Western blotting to examine the effect of PGC-1α knockdown on C/EBPβ protein expression. β-Tubulin was used as an internal control. The immunoblot is a representative of three different incubations. C/EBPβ, CCAAT/enhancer-binding protein beta; ChIP, chromatin immunoprecipitation; ESC, endometrial stromal cell; H3K27ac, acetylation of histone-H3 lysine-27; IGFBP-1, insulin-like growth factor–binding protein-1; PGC-1α, peroxisome proliferator–activated receptor gamma coactivator 1 alpha; PRL, prolactin.

Figure 5

Involvement of C/EBPβ in the cAMP-increased gene expression of PGC-1α. A, ESCs were transfected with a siRNA targeted against C/EBPβ or with a nontargeting siRNA as a control. Forty-eight days after siRNA transfection, ESCs were treated with or without cAMP for 4 days. Whole-cell lysates were prepared and subjected to Western blotting to examine the knockdown of C/EBPβ protein expression. β-Tubulin was used as an internal control. The immunoblot is a representative of three different incubations. B, PGC-1α mRNA expression was analyzed by quantitative real-time RT-PCR. Values of PGC-1α were normalized to those of MRPL19 and expressed as a ratio of the cAMP treatment sample in the control siRNA. SD of cAMP treatment in the control siRNA sample is shown in Table S4. Values are mean ±SD of three different incubations. A, p < 0.01 versus control treatment in the control siRNA; B, p < 0.05 versus cAMP treatment in the control siRNA. Whole-cell lysates were prepared and subjected to Western blotting to examine the knockdown of PGC-1α protein expression. β-Tubulin was used as an internal control. Immunoblot of β-tubulin is a reuse of Figure 5A because PGC-1α was examined in the same immunoblot membrane with C/EBPβ. The immunoblot is a representative of three different incubations. C, comparison of PGC-1α mRNA expression between nondecidualized ESCs (control), decidualized ESCs (cAMP), and HepG2 cells. PGC-1α mRNA expression was analyzed by quantitative real-time RT-PCR. Values of PGC-1α were normalized to those of MRPL19 and expressed as a ratio of the control of ESCs. SD of control samples is shown in Table S4. Values are mean ±SD of three different incubations. A, p < 0.01 versus control in ESCs. D, locations of the potential C/EBPβ-binding site in the enhancer regions of PGC-1α. For the ChIP assay, two regions were amplified by PCR: +8669 to +8823 bp and +25,280 to +25,395 bp for the PGC-1α enhancer regions. E, C/EBPβ recruitment to each enhancer region was examined by ChIP assay. Normal rabbit IgG was used as a negative control. The relative recruitment levels were analyzed by real-time PCR. Values were expressed as a ratio of control sample. SDs of control samples are shown in Table S4. Data are mean ±SD of three different incubations. A, p < 0.01 versus control treatment. F, effect of cAMP on the enhancer activities of the PGC-1α enhancer regions in ESCs. Each PGC-1α enhancer region was subcloned into pGL4.23 (enhancer/pGL4.23). ESCs were transfected with reporter vector (pGL4.23 or enhancer 1 or 2/pGL4.23) and pRL-TK vector as a normalization control. After 5 h of transfection, cells were treated in the presence or absence of cAMP for 4 days. The firefly luciferase activity was normalized according to Renilla luciferase activities. Values of the luciferase activities were expressed as a ratio of control sample with enhancer/pGL4.23. SDs of control samples are shown in Table S4. Values are mean ±SD of three different incubations. A, p < 0.01 versus control treatment sample with enhancer/pGL4.23. C/EBPβ, CCAAT/enhancer-binding protein beta; ChIP, chromatin immunoprecipitation; ESC, endometrial stromal cell; IgG, immunoglobulin G; IGFBP-1, insulin-like growth factor–binding protein-1; PGC-1α, peroxisome proliferator–activated receptor gamma coactivator 1 alpha.

Figure 6

Effect of PGC-1α enhancer deletion on cAMP-increased PGC-1α mRNA expression.A, the vertical lines represent the locations of sgRNA1 and sgRNA2 flanking the C/EBPβ-binding site in the PGC-1α enhancer regions. Two PCR primers surrounding two sgRNAs were used to amplify the genomic DNAs of each clone. B, PCR amplification products from genomic DNAs of the representative clones. For region 1 (left), a 329-bp PCR product was observed in wildtype clones. Successful deletion of the 88 bp generated between the two sgRNAs generated a smaller PCR product of 241 bp in enhancer region 1–deleted clones. For region 2 (right), a 432-bp PCR product was observed in wildtype clones. Successful deletion of the 132 bp generated between the two sgRNAs generated a smaller PCR product of 300 bp in enhancer region 1–deleted clones. C, DNA sequencing results around both deletion junctions amplified from genomic DNA of enhancer-deleted clone (Mut) (left; region 1, right; region 2). The deleted 88-bp and 132-bp regions are shown in dashed lines. The sequences of each sgRNA are boxed (sgRNA1 to sgRNA4). Primer sequences for genomic PCR are underlined. D, PGC-1α mRNA expression was analyzed by quantitative real-time RT-PCR. Values of PGC-1α were normalized to those of MRPL19 and expressed as a ratio of one wildtype sample. Values of each clone and mean ±SD are shown. A, p < 0.05 versus. wild type clone. C/EBPβ, CCAAT/enhancer-binding protein beta; PGC-1α, peroxisome proliferator–activated receptor gamma coactivator 1 alpha; sgRNA, single guide RNA.

Figure 7

Validation of RNA-sequence data by real-time RT-PCR. ESCs were transfected with a siRNA targeted against PGC-1α (A), C/EBPβ (B), or with a nontargeting siRNA as a control. Forty-eight days after siRNA transfection, ESCs were treated with or without cAMP for 4 days. mRNA expression was analyzed by quantitative real-time RT-PCR. Values of PGC-1α–upregulated genes were normalized to those of MRPL19 and expressed as a ratio of the cAMP treatment sample in the control siRNA. SDs of cAMP treatment in the control siRNA sample are shown in Table S4. Values are mean ±SD of three different incubations. A, p < 0.01 versus. control treatment in the control siRNA; B, p < 0.01 versus. cAMP treatment in the control siRNA. C/EBPβ, CCAAT/enhancer-binding protein beta; ESC, endometrial stromal cell; PGC-1α, peroxisome proliferator–activated receptor gamma coactivator 1 alpha.

Figure 8

Role and the regulation of PGC1α during decidualization. PGC-1α is upregulated by the C/EBPβ recruitment to the novel enhancer regions during decidualization. PGC-1α contributes to decidualization by forming a histone-modifying complex with C/EBPβ and p300 and increasing H3K27ac in the promoter and enhancer regions of IGFBP-1 and the promoter region of PRL. C/EBPβ, CCAAT/enhancer-binding protein beta; ESC, endometrial stromal cell; H3K27ac, acetylation of histone-H3 lysine-27; IGFBP-1, insulin-like growth factor–binding protein-1; PGC-1α, peroxisome proliferator–activated receptor gamma coactivator 1 alpha; PRL, prolactin.