The Hippo effector YAP1/TEAD1 regulates EPHA3 expression to control cell contact and motility

Abstract

The EPHA3 protein tyrosine kinase, a member of the ephrin receptor family, regulates cell fate, cell motility, and cell-cell interaction. These cellular events are critical for tissue development, immunological responses, and the processes of tumorigenesis. Earlier studies revealed that signaling via the STK4-encoded MST1 serine-threonine protein kinase, a core component of the Hippo pathway, attenuated EPHA3 expression. Here, we investigated the mechanism by which MST1 regulates EPHA3. Our findings have revealed that the transcriptional regulators YAP1 and TEAD1 are crucial activators of EPHA3 transcription. Silencing YAP1 and TEAD1 suppressed the EPHA3 protein and mRNA levels. In addition, we identified putative TEAD enhancers in the distal EPHA3 promoter, where YAP1 and TEAD1 bind and promote EPHA3 expression. Furthermore, EPHA3 knockout by CRISPR/Cas9 technology reduced cell-cell interaction and cell motility. These findings demonstrate that EPHA3 is transcriptionally regulated by YAP1/TEAD1 of the Hippo pathway, suggesting that it is sensitive to cell contact-dependent interactions.

Figure 1

Expression of Ephrin A and B receptors and their ligands in the cell lines. (a, b) Transcripts of EphA and EphB receptors and their ligands in LNCaP, C4-2, 22Rv1, and PC3 prostate cancer cell lines. NS: nonspecific bands (c) 18S RNA was included as an internal control in PCR reactions. The PCR product was resolved on a 1.5% agarose gel and imaged using the DNA gel documentation system. (d) Quantitative PCR analysis of EPHA3 transcripts, *P < 0.01. (e) Western blot (WB) analysis of EPHA3 protein in whole-cell lysates. The β-actin protein blot was included as an internal control in the WB. (f) Immunofluorescence imaging of EPHA3 protein in fixed cells. Cells were grown under steady-state conditions. Micrographs represent at least three independent experiments—size bars: 20 µm.

Figure 2

Regulation of EPHA3 expression by MST1-YAP1-TEAD1 signaling. (a) Induction of ectopic HA-tagged MST1 protein, as assessed by immunofluorescence imaging. (b) RT-PCR analysis of EPHA3 transcripts. 18S RNA was used as an internal control in the PCR reaction. (c) WB analysis of EPHA3 and HA-tagged MST1/STK4 protein. Total RNA and proteins were isolated from C4-2/TetON/HA-MST1 cells after treatment with (+) or without (−) doxycycline (Dox, 2 µg/ml) in TetON-approved serum-fed condition. (d-f) EPHA3 protein levels in LNCaP or C4-2 cell lines with or without YAP1 or TEAD1 silencing. Membrane probed with the protein-specific antibody. The β-actin or GAPDH protein blot was used as an internal control in WB. Size bars: 20 µm. Data are the representation of three independent experiments.

Figure 3

Presence of TEAD1 binding sides located in the EPHA3 promoter. (a) Schematic representation and location of the TEAD responsive elements (TREs) within the -5 kb DNA region of the EPHA3 promoter relative to the transcriptional start site (TSS). The JASPAR and EPD bioinformatics tools were employed to identify TREs. (b-e) ChIP-qPCR analysis of protein-crosslinked genomic DNA fragments bound by IgG or ChIP-grade TEAD1, YAP1, or H3K27Ac antibodies; *P < 0.001. Protein-crosslinked and sonicated DNA fragments were isolated from LNCaP or C4-2 cells that were grown in serum-fed conditions. The bound and eluted DNA fragments were analyzed by qPCR using a primer set amplifying the indicated TRE regions, and the results were normalized to the IgG control. Data are the representation of three independent experiments.

Figure 4

YAP1 and TEAD1 activate the EPHA3 promoter-directed reporter gene. (a) Luciferase activity of the pGLA4.24-minP-Luc (mock) or TEAD-responsive pA3-5xTRE-Luc (test) vector in LNCaP and C4-2 cells grown in serum fed-conditions. Schematic illustrations represent a pGLA4.24-minP-Luc and pA3-5xTRE-Luc vector. (b) Luciferase activity of pA3-5xTRE-Luc vector in LNCaP and C4-2 cells transiently transfected with scramble (mock) or with YAP1 or TEAD1 siRNA, *P < 0.001. LNCaP and C4-2 cells were transiently transfected with pA3-5xTRE-Luc plasmid DNA. (c) Luciferase activity of pA3-5xTRE-Luc vector in C4-2/TetON/MST1 cells exposed to Dox (−) or Dox (+). Relative luciferase activity was assessed at 48 h post-transfection and normalized to the total protein. Data are the representation of three independent experiments in triplicates.

Figure 5

The activity of EPHA3 signals in LNCaP and C4-2 cells. (a) Analysis of phospho-Tyr779 and total EPHA3 proteins in LNCaP and C4-2 cells treated with ephrin-A5/Fc with varying doses in serum-depleted condition. Cells were serum-starved 24 h before being stimulated with Ephrin-A5/Fc, an activated Ephrin-A5 ligand. (b) Quantification of phospho-EPHA3-Tyr779 normalized to total EPHA3 and GAPDH protein levels; *P > 0.05 for C4-2 and *P < 0.01 for LNCaP cells. (c) Levels of the RhoA and EPHA3 proteins in LNCaP and C4-2 cells that were transiently transfected with mock (−), EPHA3 (+), or/and EFNA5 (+ , Ephrin-A5 ligand) siRNA. (d) Quantification of the RhoA protein blot normalized to total GAPDH protein levels; *, **, ***P < 0.01 for both C4-2 and LNCaP cells. (e) Levels of RhoA and EPHA3 proteins in C4-2 cells were transiently transfected with mock (−) or TEAD1 (+) siRNA. Whole-cell lysates were prepared from cells grown in reduced-serum and analyzed at 48 h post-transfection. Membranes were probed with the protein-specific antibody. The GAPDH was used as an internal control in WB. (f) Quantification of the RhoA protein blot normalized to total GAPDH protein levels. ImageJ software was used to quantify the phospho-EPHA3 and total RhoA, EPHA3, and GAPDH proteins; *P < 0.01. Data are the representation of at least three independent experiments.

Figure 6

EPHA3 regulates cell–cell interaction and cell motility. (a) WB analysis of EPHA3 protein in EPHA3-KO C4-2 cell clones. The β-actin protein blot was used as an internal control. (b) Schematic representation of cell–cell interaction between EPHA3-WT/WT, EPHA3-WT/KO, or EPHA3-KO/KO cells. Light intensity inversely correlates with the distance between the two cell types. (c) Cell proximity assay (CPA) was conducted using EPHA3-WT and EPHA3-KO cell lines, as illustrated in panel b. *P < 0.03 and **, ***, ****P < 0.001. Data represent three independent experiments. The CPA system utilizes beta-galactosidase (β-gal) and luciferase (Luc) enzymes expressed by β-gal and Luc vectors. Bioluminescent signals are produced upon the addition of luciferase substrate, A3: EPHA3. (d) Micrographs are the representation of wound closure in select time, h: hour. Size bars: 20 µm. (e) The graph shows quantification of wound closure as a function of time, *P < 1.14E-13. Data represent from two independent experiments with eight data points.

Figure 7

The model summarizes the results of the study. The YAP1 and TEAD1 proteins bind the TRE and transcriptionally regulate the EPHA3 expression and biology downstream of the Hippo pathway. The membrane and nuclear boundaries and double-stranded DNA images were created using the http://www.BioRender.com web page, and the remaining content was constructed using Microsoft PowerPoint.