Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2017 Apr 17;12(1):8.
doi: 10.1186/s13062-017-0180-7.

TFAP2A Is a Component of the ZEB1/2 Network That Regulates TGFB1-induced Epithelial to Mesenchymal Transition

Affiliations
Free PMC article

TFAP2A Is a Component of the ZEB1/2 Network That Regulates TGFB1-induced Epithelial to Mesenchymal Transition

Yoana Dimitrova et al. Biol Direct. .
Free PMC article

Abstract

Background: The transition between epithelial and mesenchymal phenotypes (EMT) occurs in a variety of contexts. It is critical for mammalian development and it is also involved in tumor initiation and progression. Master transcription factor (TF) regulators of this process are conserved between mouse and human.

Methods: From a computational analysis of a variety of high-throughput sequencing data sets we initially inferred that TFAP2A is connected to the core EMT network in both species. We then analysed publicly available human breast cancer data for TFAP2A expression and also studied the expression (by mRNA sequencing), activity (by monitoring the expression of its predicted targets), and binding (by electrophoretic mobility shift assay and chromatin immunoprecipitation) of this factor in a mouse mammary gland EMT model system (NMuMG) cell line.

Results: We found that upon induction of EMT, the activity of TFAP2A, reflected in the expression level of its predicted targets, is up-regulated in a variety of systems, both murine and human, while TFAP2A's expression is increased in more "stem-like" cancers. We provide strong evidence for the direct interaction between the TFAP2A TF and the ZEB2 promoter and we demonstrate that this interaction affects ZEB2 expression. Overexpression of TFAP2A from an exogenous construct perturbs EMT, however, in a manner similar to the downregulation of endogenous TFAP2A that takes place during EMT.

Conclusions: Our study reveals that TFAP2A is a conserved component of the core network that regulates EMT, acting as a repressor of many genes, including ZEB2.

Reviewers: This article has been reviewed by Dr. Martijn Huynen and Dr. Nicola Aceto.

Keywords: EMT; Epithelial-to-mesenchymal transition; NMuMG; TFAP2A; TGFb1; Transcription regulatory network; ZEB2.

Figures

Fig. 1
Fig. 1
The transcriptional networks inferred from different EMT systems. Motif–motif interaction networks derived from mouse (a) and human (b) datasets. An arrow was drawn from a motif A to a motif B if motif A was consistently (across datasets from the corresponding species) predicted to regulate a transcriptional regulator b that is known to bind motif B. The probability product that A targets b is reflected by the thickness of the line. For readability, only motifs with an absolute z-score > 2.0 and having at least one interaction with another such motif (with a target probability product > 0.35 for human and > 0.15 for mouse) are depicted. The color intensity of the nodes representing motifs is proportional to the significance of the motif given by its z-score. Red indicates increased and green indicates decreased activity upon EMT
Fig. 2
Fig. 2
TFAP2A expression and activity profile in the NMuMG EMT model. a-b NMuMG cells were treated with 2 ng/mL of TGFβ1 for 72 h and were stained for TFAP2A and F-Actin (a) and TFAP2A and E-cadherin (b). The merged panels represent colocalization of the imaged markers with the nucleus which was stained with DAPI and compared to controls. Scale bar represents 50 μm. c NMuMG cells were treated for 14 days with 2 ng/mL of TGFβ1. Quantitative RT-PCR of Tfap2a during the time course of this treatment indicates that Tfap2a mRNA levels are reduced upon EMT. The EMT markers E-cadherin (Cdh1), Fibronectin (Fn1), Occludin (Ocln), and Vimentin (Vim) follow the expected trend. d Two mRNA-seq samples from independent wells were prepared from a time course of NMuMG cells treated for 14 days with 2 ng/mL of TGFβ1, and the data was consequently analyzed with ISMARA [30]. The figure depicts the dynamics of TFAP2A/C transcriptional activity during the time course. The sequence logo of the TFAP2A/C binding motif is also indicated. e-f Lysates from NMuMG/E9 cells treated with 2 ng/mL of TGFβ1 for 72 h were probed for TFAP2A, GAPDH and Lamin B expression by WB and their levels compared with the expression levels of Actin and also to the Ponceau-stained membrane (e). The bar plot represents the densitometric quantification of the TFAP2A protein levels upon treatment compared to the control (f) ** indicates a p-value < 0.01 in the paired t-test (P = 0.0014)
Fig. 3
Fig. 3
TFAP2A expression and activity in breast cancers. Box plots of TFAP2A gene expression (a) and expression levels of the top 100 ISMARA-inferred TFAP2A targets (b) in a panel of breast cancer cell lines grouped in the basal A (red), basal B (grey) and luminal (blue) subgroups based on the annotation from Neve et al. [32]. c Box plot of TFAP2A gene expression for tumor samples stratified according to PAM50 subtypes [57]. All plots were generated with the GOBO online tool [33]
Fig. 4
Fig. 4
TFAP2A binds directly to the Zeb2 promoter region. a Sketch of the region around the second exon of mouse Zeb2, showing the two transcription start sites found in SwissRegulon [44]. The blue filled box indicates the non-coding untranslated region (UTR) in exon 2, while the white filled box designates the start of the coding region (CDS). The predicted TFAP2A binding sites from SwissRegulon are marked with red arrows, and the probes that were used in (b) are indicated with green lines below the gene structure. Predicted transcription start sites (TSS) are also indicated. b Radiography of TFAP2A Electrophoretic Mobility Shift Assay (EMSA) with radiolabeled oligonucleotides, each spanning one of the predicted binding sites. The presence or absence of TFAP2A protein in the assay is indicated by a + or – sign, respectively. Cold competitors were used at 200-fold excess over the radiolabelled probes. Wt corresponds to unlabeled probe; M indicates a double-stranded oligonucleotide with a mutated TFAP2A binding site. Red arrows indicate the predicted TFAP2A binding probes that behave as expected from specific binding of TFAP2A. c TFAP2A ChIP was performed in NMuMG cells stably transduced with pCLX-TFAP2A (denoted as TFAP2A-OE (blue)) or with pCLX-GFP (denoted as TFAP2A-GFP (green)) viral vectors and further treated with 2 μg/mL doxycycline. Quantitative PCR data shows the enrichment of Zeb2 promoter relative to a non-transcribed genomic region in TFAP2A-ChIP normalized to IgG control (red). Two independent experiments were performed for each condition and shown are means and standard deviations. The one-tail paired t-test indicates that TFAP2A is significantly enriched at the Zeb2 (** for p < 0.01). d ChIP-seq libraries from TFAP2A ChIP or input chromatin were generated and the coverage of the genomic region spanning the second exon of Zeb2 by reads is shown in a mouse genome browser (www.clipz.unibas.ch and [45]). The results of two independent experiments are presented. The TFAP2A ChIP-seq the Zeb2 promoter region previously assessed by qPCR is enriched with respect to the input control sample. Mapping, annotation and visualization of deep-sequencing data was done with the ClipZ server [45]
Fig. 5
Fig. 5
TFAP2A overexpression in NMuMG modulates epithelial plasticity. a Expression of either GFP or TFAP2A was induced by 72 h doxycycline treatment in NMuMG cells stably transduced with either pCLX-GFP or pCLX-TFAP2A. Morphological changes and sparse cell arrangement are visible in phase contrast microscopy upon TFAP2A expression. Scale bar: 50 μm. b Gene expression log2 fold changes of EMT markers (TFs) were calculated from mRNA-seq samples of doxycycline-induced, TGFβ1-treated (72 h, 2 ng/mL) pCLX-GFP (pCLX-GFP + TGF-beta), doxycycline-induced pCLX-TFAP2A (pCLX-TFAP2A), as well as of doxycycline-induced, TGFβ1-treated (72 h, 2 ng/mL) pCLX-TFAP2A (pCLX-TFAP2A + TGF-beta) cell lines relative to doxycycline-induced pCLX-GFP (pCLX-GFP) cell line. Shown are the mean log2 fold changes (+/- 1 standard deviation) from two experiments. TFAP2A overexpression is apparent in both TFAP2A-induced samples (dark green and dark blue) but is not induced in cells treated with TGFβ1 alone (light blue). The EMT-inducing TFs have increased expression upon TFAP2A induction. * indicates a p-value ≤ 0.05 and ** a p-value ≤ 0.01 in a two-tailed t-test. c The transcriptional activities of TFAP2{A,C} and SNAI1..3 motifs in different conditions, as inferred with ISMARA from mRNA-seq data as described in (b). The two replicates from each condition are plotted next to each other

Similar articles

See all similar articles

Cited by 3 articles

References

    1. Thiery JP, Acloque H, Huang RY, Nieto MA. Epithelial-mesenchymal transitions in development and disease. Cell. 2009;139(5):871–90. doi: 10.1016/j.cell.2009.11.007. - DOI - PubMed
    1. Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell. 2011;144(5):646–74. doi: 10.1016/j.cell.2011.02.013. - DOI - PubMed
    1. Micalizzi DS, Farabaugh SM, Ford HL. Epithelial-mesenchymal transition in cancer: parallels between normal development and tumor progression. J Mammary Gland Biol Neoplasia. 2010;15(2):117–34. doi: 10.1007/s10911-010-9178-9. - DOI - PMC - PubMed
    1. Mani SA, Guo W, Liao MJ, Eaton EN, Ayyanan A, Zhou AY, Brooks M, Reinhard F, Zhang CC, Shipitsin M, et al. The epithelial-mesenchymal transition generates cells with properties of stem cells. Cell. 2008;133(4):704–15. doi: 10.1016/j.cell.2008.03.027. - DOI - PMC - PubMed
    1. Scheel C, Eaton EN, Li SH, Chaffer CL, Reinhardt F, Kah KJ, Bell G, Guo W, Rubin J, Richardson AL, et al. Paracrine and autocrine signals induce and maintain mesenchymal and stem cell states in the breast. Cell. 2011;145(6):926–40. doi: 10.1016/j.cell.2011.04.029. - DOI - PMC - PubMed

MeSH terms

Feedback