Vestigial-like family member 3 (VGLL3), a cofactor for TEAD transcription factors, promotes cancer cell proliferation by activating the Hippo pathway

Vestigial-like 3 (VGLL3) is a member of the VGLL family, whose members serve as cofactors for TEA domain–containing transcription factors (TEADs). TEADs promote tissue and tumor development together with the cofactors Yes-associated protein (YAP) and transcriptional coactivator with PDZ-binding motif (TAZ). Although VGLL3 is involved in tumor cell proliferation, its relationship with TEADs and YAP/TAZ remains largely unknown. To close this research gap, here we established tumor cells stably expressing VGLL3 and found that they exhibit enhanced proliferation. Notably, YAP and TAZ were inactivated in the VGLL3-expressing cells, coinciding with activation of the Hippo pathway, which suppresses YAP/TAZ activities. VGLL3 in combination with TEADs promoted expression of the Hippo pathway components large tumor suppressor kinase (LATS2) and angiomotin-like 2 (AMOTL2). VGLL3 was highly expressed in malignant breast tumor cells and osteosarcoma cells, and VGLL3 knockdown increased nuclear localization of YAP and TAZ. Knockdown of LATS2 or AMOTL2, as well as VGLL3 knockdown, repressed proliferation of breast tumor cells. Together, these results suggest that VGLL3 together with TEADs promotes cell proliferation by activating the Hippo pathway through LATS2 and AMOTL2, leading to YAP/TAZ inactivation.


Edited by Alex Toker
Vestigial-like 3 (VGLL3) is a member of the VGLL family, whose members serve as cofactors for TEA domain-containing transcription factors (TEADs). TEADs promote tissue and tumor development together with the cofactors Yes-associated protein (YAP) and transcriptional coactivator with PDZ-binding motif (TAZ). Although VGLL3 is involved in tumor cell proliferation, its relationship with TEADs and YAP/TAZ remains largely unknown. To close this research gap, here we established tumor cells stably expressing VGLL3 and found that they exhibit enhanced proliferation. Notably, YAP and TAZ were inactivated in the VGLL3-expressing cells, coinciding with activation of the Hippo pathway, which suppresses YAP/TAZ activities. VGLL3 in combination with TEADs promoted expression of the Hippo pathway components large tumor suppressor kinase (LATS2) and angiomotin-like 2 (AMOTL2). VGLL3 was highly expressed in malignant breast tumor cells and osteosarcoma cells, and VGLL3 knockdown increased nuclear localization of YAP and TAZ. Knockdown of LATS2 or AMOTL2, as well as VGLL3 knockdown, repressed proliferation of breast tumor cells. Together, these results suggest that VGLL3 together with TEADs promotes cell proliferation by activating the Hippo pathway through LATS2 and AMOTL2, leading to YAP/TAZ inactivation.
VGLL3 is one of the four members of the vestigial-like (VGLL) family of proteins. VGLL3 is thought to bind to TEA domain-containing transcription factors (TEADs) through the TONDU domain and function as a transcriptional cofactor for TEADs (1,2). Although the physiological role of VGLL3 remains largely unknown, there is evidence suggesting that VGLL3 is involved in tumor development. For example, gene amplification and overexpression of VGLL3 were observed in several types of human sarcomas, such as myxoinflammatory fibroblastic sarcoma and soft-tissue sarcoma (3)(4)(5). In a cell line derived from a soft-tissue sarcoma patient, VGLL3 was shown to be required for proliferation (4). VGLL3 was also overexpressed in gastric cancer patients and was associated with a poor prognosis (6).
Mammalian cells have four TEADs, TEAD1-TEAD4, which are broadly expressed and involved in tissue development and tumorigenesis (7). TEAD activity is controlled by the Hippo pathway. The Hippo pathway primarily targets the transcriptional cofactor YAP (Yes-associated protein) and its paralog TAZ (transcriptional coactivator with PDZ-binding motif) (7)(8)(9)(10)(11). YAP/TAZ bind to TEADs in the nucleus and drive expression of target genes that are crucial for cell proliferation and survival. Hippo pathway activation induces phosphorylation of YAP/TAZ through the serine/threonine kinases LATS1/2 (large tumor suppressor kinase 1/2), promoting cytoplasmic retention and protein degradation of YAP/TAZ (12). The Hippo pathway is the key negative regulator of TEADs, and dysregulation of this pathway causes tissue overgrowth and tumor development (13).
Given the involvement of VGLL3 in tumor development, it is possible that VGLL3 promotes cell growth by regulating TEADs and YAP/TAZ. However, the relationship between these proteins is not clear. Therefore, this study was conducted to examine the interaction between VGLL3, YAP/ TAZ, TEADs, and the Hippo pathway.

Stable expression of VGLL3 enhances cell proliferation
Because VGLL3 is thought to bind to TEADs, we first confirmed association between VGLL3 and TEADs using TEAD4 as a representative member of TEADs. Co-expression and a subsequent immunoprecipitation assay showed that TEAD4 co-immunoprecipitated with VGLL3, supporting the notion that VGLL3 binds to TEADs (Fig. 1A). To examine the pro-proliferative effect of VGLL3, we established cell pools stably expressing VGLL3 in human lung tumor A549 cells and breast tumor MDA-MB-231 cells (Fig. 1B). Stably expressed VGLL3 bound endogenous TEAD4 proteins (Fig. 1C). Stable expression of VGLL3 caused a spindle-like morphological change in A549 cells (Fig. 1D). Notably, VGLL3 expression significantly enhanced cell proliferation with activation of mitogenic extracellular signal-regulated kinases (ERKs) (Fig. 1, E and F). Consistent with previous data (4), these results showed that VGLL3 has a pro-proliferative effect in tumor cells probably through activation of ERK signaling.  binding of TEADs to the GTIIC motif using reporter assays. YAP expression stimulated GTIIC-driven reporter activity (EGFP expression), whereas VGLL3 expression repressed, rather than induced, reporter activity ( Fig. 2A). We also examined expression of CTGF (connective tissue growth factor), CYR61 (cysteine-rich angiogenic inducer 61), and ANKRD1 (ankyrin repeat domain 1), well-known targets of TEADs in association with YAP/TAZ (17)(18)(19). CTGF and ANKRD1 expression was almost completely repressed, and only slight induction of CYR61 expression was observed in VGLL3expressing cells (Fig. 2B). These results suggest that the VGLL3-TEAD complex does not bind to the GTIIC motif and that target genes of this complex are distinct from those of YAP/TAZ-TEAD.

VGLL3 inactivates Yap/TAZ via hippo pathway activation
To investigate the involvement of YAP/TAZ in VGLL3-dependent cell proliferation, we examined activation of the Hippo pathway and subcellular localization of YAP/TAZ in VGLL3expressing A549 cells. Notably, immunofluorescence revealed that YAP/TAZ levels were significantly reduced in these cells, and residual staining was detected in the cytoplasm but not in the nucleus (Fig. 3A). Western blotting showed that TAZ is the main component of YAP/TAZ proteins in A549 cells. In VGLL3-expressing A549 cells, the band of TAZ was weakened and showed increased molecular weight, likely caused by phosphorylation (Fig. 3B). LATS2, the key kinase in the Hippo pathway, showed elevated expression and phosphorylation levels following VGLL3 expression (Fig. 3, B and C). Expression of the angiomotin (AMOT) family, which inhibits nuclear translocation of YAP/TAZ (20), was also examined. Among this family, AMOTL2 expression was elevated in VGLL3-expressing cells (Fig. 3, B and C). Meanwhile, AMOTL1 expression was reduced, and AMOT expression was not detectable (data not shown). Similar phenomena were observed in VGLL3-expressing MDA-MB-231 cells (Fig. 3, D and E). These results suggest that VGLL3 inactivates YAP/TAZ via LATS2-and AMOTL2mediated activation of the Hippo pathway.

VGLL3 activates the hippo pathway through TEADs
To investigate the involvement of the Hippo pathway in down-regulating YAP/TAZ in VGLL3-expressing cells, several knockdown experiments were performed. LATS1 plus LATS2 double knockdown recovered the reduction in TAZ levels, and LATS2 single knockdown increased the lower band of TAZ (unphosphorylated TAZ proteins) (Fig. 4, A and B). AMOTL2 knockdown also recovered TAZ reduction (Fig. 4, C and D), suggesting that elevated expression of LATS2 and AMOTL2 down-regulates TAZ in VGLL3-expressing cells. To determine the TEAD members that are crucial for induction of LATS2 and AMOTL2, each TEAD was knocked down. Knockdown of TEAD1 or TEAD3 reduced expression of LATS2 and AMOTL2 and recovered TAZ reduction in VGLL3-expressing cells (Fig. 4, E and F). These results suggest that VGLL3 induces LATS2 and AMOTL2 expression via TEAD1 and TEAD3, thereby activating the Hippo pathway.
The hippo signaling pathway is involved in proliferation in breast cancer cells Because the VGLL family was suggested to be involved in breast cancer development (21), we examined VGLL3 expression in human breast cancer MCF7 cells and MDA-MB-231  cells. VGLL3 was highly expressed in MDA-MB-231 cells, a malignant type of breast cancer cells. High expression of VGLL3 was also observed in human osteosarcoma Saos-2 cells (Fig. 5A). To analyze the role of endogenous VGLL3 in activation of YAP/TAZ and the Hippo pathway, we depleted VGLL3 and analyzed the subcellular localization of YAP/TAZ in MDA-MB-231 and Saos-2 cells. VGLL3 depletion significantly increased the nuclear localization of YAP/TAZ (Fig. 5, B and C). VGLL3 depletion also caused a reduction in LATS2 expression and phosphorylation (Fig. 5, D and E). VGLL3 depletion repressed AMOTL2 mRNA expression but not its protein levels. Notably, knockdown of LATS2 or AMOTL2, as well as knockdown of VGLL3, significantly repressed proliferation in these cells. AMOTL2 depletion did not repress proliferation in Saos-2 cells. These results suggest that VGLL3-dependent induction of LATS2 is critically involved in tumor cell proliferation at the endogenous level (Fig. 5, F and G).

Discussion
The Hippo pathway is an evolutionarily conserved regulator of tissue and cell growth. Involvement of the Hippo pathway in human tumors was initially recognized by the robust overgrowth of Drosophila melanogaster tissues that had loss-offunction mutations in components of this pathway. Subsequent studies revealed that Hippo pathway dysregulation promotes tumorigenesis in mammals and that several genes of this pathway have mutations and altered expression in human tumors. These observations established that the Hippo pathway is a tumor-suppressing signaling pathway (Fig. 6A) (7,11,13).
In this study, we found that stable expression of VGLL3 promotes cell proliferation. We detected association between VGLL3 and TEAD4, supporting the notion that VGLL3 is a cofactor for TEADs. Because Hippo pathway activation represses TEAD activity, the cell growth-promoting effect of VGLL3 led us to speculate that VGLL3 inhibits the Hippo pathway and in turn induces nuclear translocation and activation of YAP/TAZ. However, contrary to this speculation, components of the Hippo pathway, LATS2 and AMOTL2, were increased, and YAP/TAZ were inactivated in VGLL3expressing cells. VGLL3 was highly expressed in the malignant breast tumor cell line MDA-MB-231, and its knockdown caused a reduction in LATS2 and AMOTL2 expression and induction of YAP/TAZ nuclear localization. Notably, knockdown of LATS2 or AMOTL2, as well as knockdown of VGLL3, repressed proliferation in MDA-MB-231 cells, suggesting that VGLL3dependent up-regulation of Hippo pathway components and the following down-regulation of YAP/TAZ are required for VGLL3-dependent cell growth. Previous studies showed that VGLL1 and VGLL4 have binding sites on TEADs that overlap with YAP/TAZ-binding sites and thus compete with YAP/TAZ for TEAD binding (10,22). Therefore, VGLL3 is likely to inacti-vate YAP/TAZ through the Hippo pathway to enhance binding of itself to TEADs to facilitate gene expression for proliferation. In this context, it can be concluded that the Hippo pathway has the potential to act as a tumor-promoting signaling pathway via VGLL3 activation (Fig. 6B).
Although VGLL3 was found to have a cell growth-promoting effect, probably through activation of mitogenic ERK signaling, target genes underlying this effect have not been identified. Whereas CTGF is a well-known YAP/TAZ-dependent TEAD target involved in cell proliferation (19), CTGF expression was almost completely repressed in VGLL3-expressing cells. Reporter assays showed that VGLL3 did not activate GTIIC TEAD-binding site-dependent gene expression. Therefore, we hypothesize that the DNA-binding site and target genes of the VGLL3-TEAD complex are different from those of the YAP/ TAZ-TEAD complex. Given that VGLL1 was shown to facilitate anchorage-independent cell growth via expression of insulin-like growth factor-binding protein 5 (IGFBP5) (10), a cell proliferation-related gene, we analyzed IGFBP5 expression in VGLL3-expressing cells. However, we could not detect IGFBP5 expression in these cells, suggesting that IGFBP5 is not a target of the VGLL3-TEAD complex (data not shown). To determine the genes responsible for VGLL3-induced cell proliferation, analysis of gene expression profiles of a VGLL3-expressing cell is required.
In contrast to our findings, VGLL3 was hypothesized to have a tumor-suppressive role in ovarian tumor (23). This hypothesis was derived from the chromosome transfer experiment: transfer of a chromosome fragment containing the VGLL3 gene suppressed tumor phenotypes in the ovarian tumor cell line OV90. Although transfer of this fragment recovered VGLL3 expression (24), VGLL3 single-gene transfer failed to repress proliferation in OV90 cells (25). Therefore, we speculate that the hypothesis of a tumor-suppressive role of VGLL3 may be still disputable.
In conclusion, we found that VGLL3 promotes cell growth through activation of the Hippo pathway and subsequent inactivation of YAP/TAZ. Our findings suggest that the Hippo pathway can function as a tumor-promoting signaling pathway via VGLL3 activation. In VGLL3-expressing tumors, such as sarcomas and malignant breast tumors, the Hippo pathway may be a therapeutic target to inhibit VGLL3 activation.

Experimental procedures
Plasmids cDNA fragments encoding VGLL3, YAP, or TEAD4 were generated by PCR from the reverse-transcribed product of A549 total RNA and subcloned into pIRESpuro3-CAG vector (26). VGLL3 and TEAD4 cDNAs were fused with an N-terminal Myc tag and Strep tag, respectively. The following primers were used for PCR: VGLL3, 59-AAA AGA ATT CAT GAG  Table S1. pCAG/TetR vector (26), which constitutively expresses tetracycline repressor (TetR), was used to assess transfection efficiency.

Western blotting and immunoprecipitation
Western blotting and immunoprecipitation were performed as described previously (26). Whole-cell lysates prepared in SDS-sample buffer were subjected to SDS-PAGE and electrotransferred onto polyvinylidene difluoride membranes. Protein bands were detected with the appropriate antibodies and analyzed with a ChemiDoc XRS1 image analyzer (Bio-Rad). The intensity of bands was measured by Quantity One software (Bio-Rad).

Quantitative real-time PCR (qPCR) analysis
Total RNA was isolated from cells with the RNAiso Plus reagent (Takara, Shiga, Japan), and cDNAs were synthesized from 0.5 mg of each RNA preparation using the ReverTra Ace qPCR RT kit (Toyobo, Osaka, Japan), following the manufacturer's instructions. The following primers were used for PCR: GAPDH, 59-GGA GCG AGA TCC CTC CAA AAT-39 (sense)  MTT (3-[4,5-dimethylthiazol-2-yl]-2,5 diphenyl tetrazolium bromide) assay Cells were plated with 5,000 cells/well in a 96-well plate and transfected with the indicated siRNAs for 2 days. After incubation, the solution in each well was replaced with Dulbecco's modified Eagle's medium containing 0.5 mg/ml MTT (Dojindo, Kumamoto, Japan) and incubated for an additional 1 h. The resulting crystalized product was dissolved in 100 ml of 100% DMSO and measured at 595 nm using a Filter Max F5 microplate reader (Molecular Devices, Sunnyvale, CA, USA).

Immunofluorescence
Images were obtained using an LSM 780 confocal laser-scanning microscope with a 363 1.40 numerical aperture oilimmersion objective (Zeiss, Jena, Germany) as described previously (27,28). One planar (xy) slice (1.0-mm thickness) was shown in all experiments. Briefly, cells were fixed in 4% paraformaldehyde for 20 min at room temperature and permeabilized in PBS containing 0.2% Triton X-100 and 3% BSA at room temperature. Cells were subsequently reacted with an appropriate primary antibody for 1 h, washed with PBS containing 0.1% saponin, and stained with an Alexa Fluor 488-or Alexa Fluor 546-conjugated secondary antibody for 1 h. DNA was stained with a mixture of 20 mg/ml propidium iodide and 200 mg/ml RNase A for 30 min Stained cells were mounted with Prolong Antifade reagent (Thermo Fisher Scientific). Intensities of nuclear localized YAP/TAZ were quantified by ZEN software (Zeiss).

Data availability
All data are contained within the article. Funding and additional information-This work was supported in part by Grant-in-aid for Challenging Research (Exploratory) 19K22482 (to Noritaka Yamaguchi) from the Japanese Ministry of Education, Culture, Sports, Science, and Technology; grants from the Takeda Science Foundation (to Noritaka Yamaguchi), the Foundation for Promotion of Cancer Research in Japan (to Noritaka Yamaguchi), and the Chiba Foundation for Health Promotion and Disease Prevention (to Naoto Yamaguchi); and a scholarship donation from the Daiichi Sankyo Co., Ltd. (to Naoto Yamaguchi).
Conflict of interest-The authors declare that they have no conflicts of interest with the contents of this article.