TDP2 is a regulator of estrogen-responsive oncogene expression

Abstract

With its ligand estrogen, the estrogen receptor (ER) initiates a global transcriptional program, promoting cell growth. This process involves topoisomerase 2 (TOP2), a key protein in resolving topological issues during transcription by cleaving a DNA duplex, passing another duplex through the break, and repairing the break. Recent studies revealed the involvement of various DNA repair proteins in the repair of TOP2-induced breaks, suggesting potential alternative repair pathways in cases where TOP2 is halted after cleavage. However, the contribution of these proteins in ER-induced transcriptional regulation remains unclear. We investigated the role of tyrosyl-DNA phosphodiesterase 2 (TDP2), an enzyme for the removal of halted TOP2 from the DNA ends, in the estrogen-induced transcriptome using both targeted and global transcription analyses. MYC activation by estrogen, a TOP2-dependent and transient event, became prolonged in the absence of TDP2 in both TDP2-deficient cells and mice. Bulk and single-cell RNA-seq analyses defined MYC and CCND1 as oncogenes whose estrogen response is tightly regulated by TDP2. These results suggest that TDP2 may inherently participate in the repair of estrogen-induced breaks at specific genomic loci, exerting precise control over oncogenic gene expression.

Graphical Abstract

Figure 1.

Estrogen and TOP2-dependent induction of MYC. (A) Experiment design. MCF7 and T47D cells were treated with 10 nM 17β-estradiol for 1, 2 and 6 h. RNA was extracted at each time point for 3D PCR analysis and RNA-seq at 2 h. (B) 3D PCR output from Analysis Suites. Blue dots represent wells only with the amplification of genes detected by FAM (MYC), red dots represent wells only with the amplification detected by VIC (TBP or TFRC) probe (internal control), yellow dots represent wells without amplification, green dots represent wells with the amplification of both probes. Normalized MYC expression was calculated by the ratio of FAM (MYC) to VIC (TBP or TFRC). (C) MYC induction in cells after 2 h of treatment with 10 nM E2 relative to cells with no E2 in two ER-positive breast cancer cell lines (MCF7 and T47D) and an ER-negative cell line (MD-MBA-231). Error bars represent the standard deviation from three independent measurements. (D) Dynamic responses of MYC induction. Cells were treated with 10 nM E2 for 0, 1, 2 and 6 h. MYC fold-induction peaked at 1 h and remained elevated until 6 h. Error bars represent the standard deviation from three independent measurements. (E) A TOP2 catalytic inhibitor ICRF-193 abrogated the induction of MYC mRNA. MCF7 and T47D cells were treated with increasing concentrations of ICRF-193 for 1 h then treated with 10 nM E2 for 2 h. MYC induction was progressively decreased in a concentration-dependent manner in both cell lines. Error bars represent the standard deviation from three independent measurements. * P< 0.05, ** P< 0.01.

Figure 2.

TDP2 suppresses the volatile and prolonged induction of MYC by estrogen. (A) Western blot showing MCF7-1, T47D and MCF7-2 cell lines and TDP2 KO cells for each cell line. (B) Etoposide sensitivity measured by colony formation assay for MCF7 (top) and T47D (bottom). Results from three independent experiments. Error bars represent the standard deviation from three independent measurements. * P< 0.05, **P< 0.01. (C) MYC transcriptional responses persist in TDP2-deficient cells. MCF7 and MCF7 TDP2 KO cells were treated with 10 nM E2 for 1, 2 and 6 h. After a similar level of induction at 1 h, high level of response persisted in MCF7 TDP2 KO cells persisted through 6-h time point. Error bars represent the standard deviation from two independent experiments. * P< 0.05, **P< 0.01. (D) MYC transcriptional responses at 2 h. MCF7-1, MCF7-2 and T47D, and their TDP2 KO counterparts were treated with 10 nm E2 for 2 h. Note that two independent TDP2 KO clones of MCF7 cells were tested. Error bars represent the standard deviation from two independent experiments. (E) NU7441 and etoposide increased the MYC transcriptional response. MCF7 cells, cultured with or without NU7441 or etoposide were then treated with 10 nM E2 for 2 h. Error bars represent the standard deviation from two independent experiments.

Figure 3.

The induction of MYC protein by E2 in vivo. (A) Representative images of mammary ducts from wild-type (left) and TDP2KO (right) mice immunostained with MYC (red) and CK-8/18 (green) proteins before injection (0 h) and 6 and 20 h after injection of E2. (B) BOX plots showing the fraction of MYC-positive cells in (CK8/18-positive) mammary epithelial cells. Each plot represents the distribution of the fraction of MYC-positive epithelial cells in a mammary duct. MYC-positive epithelial cells were scored in 3–12 mammary ducts in each mouse at each time point.

Figure 4.

Bulk and Single-cell RNA sequencing analysis of MYC transcriptional response by E2. (A) MYC transcriptional response to E2 (2 h) in MCF7 and MCF7 TDP2KO cells, measured by bulk RNA-seq and single-cell RNA-seq. (B) Violin plots show normalized MYC expression distributions in MCF7 and MCF7 TDP2KO cells with or without E2 (2 h), obtained from single-cell RNA-seq data. (C) Histograms showing the skewed unimodal distribution of MYC expression in MCF7 and MCF7 TDP2KO cells with or without E2 (2 h), obtained from single-cell RNA-seq data. (D) Violin plots showing the normalized expression of MYC at the single-cell level in G1, S and G2/M cell populations.

Figure 5.

Commonly upregulated genes between bulk- and single-cell RNA-seq. (A) Venn diagrams showing the differentially expressed genes by E2 treatment in MCF7 and MCF7 TDP2KO cells. 32 genes in MCF7 and 62 genes in MCF7 TDP2KO are common between the two platforms. The significances of overlap were calculated using Fisher's exact test. (B) Bar plots showing the number of common DEGs by E2 (2 h). 28 and 4 genes were commonly up- and down-regulated genes in E2 treated MCF7 versus untreated MCF7, respectively. 58 and 4 genes were commonly up- and downregulated in E2 treated MCF7 TDP2KO versus untreated MCF7 TDP2KO, respectively. (C) Venn diagram of 32 and 62 common DEGs from bulk and single-cell RNA-seq data and table of 8 overlapping genes. The significance of overlap was calculated using Fisher's exact test.