MCF-7 and MCF-7-ERα36 cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and penicillin/streptomycin. TAMR-MCF-7 cells were cultured in DMEM supplemented with 10% charcoal stripped FBS (Gemini Bio Product, CA, USA), penicillin/streptomycin and 3 μM (Z)-4-hydroxytamoxifen (Tocris Bioscience, Bristol, United Kingdom). TAMR-MCF-7 cells were established as previously reported (Park et al., 2019).

Verteporfin (#SML0534) and antibodies recognizing β-actin (#a2228), FLAG^® M2 (#F1804), glyceraldeyde-3-phosphate dehydrogenase (GAPDH, #CB1001) and 17-β-estradiol (#E1024) were purchased from Merck (Billerica, MA, USA). Antibodies recognizing YAP/TAZ (#8418), p-YAP (Ser127) (#4911), zinc finger E-box binding homeobox1 (ZEB1, #3396), Src (#2108), p-Src (Tyr416) (#2101), horseradish peroxidase-conjugated donkey anti-rabbit IgG (#7074), and horseradish peroxidase-conjugated donkey anti-mouse IgG (#7076) were purchased from Cell Signaling Technology (Danvers, MA, USA). Antibodies recognizing E-cadherin (#610181) and N-cadherin (#610920) were purchased from BD Biosciences (San Jose, CA, USA). The antibody recognizing p-YAP (Tyr357) (ab62751) was purchased from Abcam (Cambridge, United Kingdom). Antibodies recognizing Vimentin (sc-32322) was purchased from Santa Cruz biotechnology (Dallas, TX, USA). V5 Tag (#A190-120A), Alexa Fluor 488 donkey anti-mouse IgG and Alexa Fluor 568 goat anti-rabbit IgG were purchased from Thermo Fisher Scientific (Cleveland, OH, USA). Saracatinib (#S1006) and PP2 (#S7008) were purchased from Selleckchem (Houston, TX, USA). G418 was purchased from Biosesang (Gyeonggi-do, South Korea).

Total cell lysates were prepared and immunoblotting was performed as previously reported (Kim et al., 2019).

TRIzol reagent (Invitrogen, Carlsbad, CA, USA) was used to extract total RNA in accordance with the manufacturer’s instructions. RNA-sequencing was performed by Macrogen, Inc. (Seoul, South Korea). Hippo signaling-related genes used for producing a heatmap include Gene Ontology biological process (GOBP) Hippo signaling gene set (https://www.gsea-msigdb.org/gsea) and a YAP/TAZ transcriptional target signature of 22 genes (Wang Y. et al., 2018). One microgram of total RNA was reverse-transcribed to cDNA by Maxime RT-PreMix Kit (Intron biotechnology, Seongnam, South Korea). Real-time PCR was performed with the MiniOpticon real-time PCR analysis instrument (Bio-Rad, Hercules, CA, USA) by using IQ Sybr Green SuperMix (Bio-Rad). The 18S mRNA value was used to normalize the target gene’s mRNA level. The following primer sets were used for the amplification of targets: Human ERα36 (forward: 5′-GAC​AGG​AAC​CAG​GGA​AAA-3′, reverse: 5′- TCT​ACA​TGT​GAG​ATA​CCA​GA-3′), human YAP (forward: 5′-ACG​TTC​ATC​TGG​GAC​AGC​AT-3′, reverse: 5′-GTT​GGG​AGA​TGG​CAA​AGA​CA-3′), human CTGF (forward: 5′-CCA​ATG​ACA​ACG​CCT​CCT​G-3′, reverse: 5′-TGG​TGC​AGC​CAG​AAA​GCT​C-3′), human CYR61 (forward: 5′- AGC​CTC​GCA​TCC​TAT​ACA​ACC-3′, reverse: 5′-TTC​TTT​CAC​AAG​GCG​GCA​CTC-3′), human 18S rRNA (forward: 5′-GTA​ACC​CGT​TGA​ACC​CCA​TT -3′, reverse: 5′- CCA​TCC​AAT​CGG​TAG​TAG​CG -3′).

ERα36-MCF-7 cells were constructed as previously described by Wang and others (Wang et al., 2005; Wang et al., 2006), and hER-α36 expression vector was kindly provided by Dr. Zhao-Yi Wang (Creighton University Medical School, USA). MCF-7 cells were plated at a density of 1 × 105 cells/well in a 24-well plate overnight and the cells were transfected with either a hER-α36 expression vector or empty pcDNA3.1 vector as control vector driven by the CMV promotor using Lipofectamine 2000 (Thermo Fisher Scientific). Transfected cells were replated 48 h after transfection and selected for several weeks using G418 (800 μg/ml).

For immunoprecipitation assay, ERα36-V5-His overexpressing MCF-7 cells were also established by retrovirus infection system: MCF-7 cells were exposed to the viral soup obtained by introducing MSCV-ERα36-V5-His vector to Phoenix-AMPHO cells with 4 μg/ml polybrene (Santa Cruz Biotechnology, Dallas, TX, USA). Infected cells were isolated by sorting GFP-labeled cells with FACSAria II cell sorter (BD Biosciences).

YAP knockout MCF-7-ERα36 cells were constructed by CRISPR-Cas9 gene editing as previously reported (Park et al., 2021). The plasmid DNA U6-gRNA/CMV-Cas9-RFP plasmid for YAP (HS0000121498) and CRISPR universal negative control plasmid were purchased from Sigma (San Luis, MO, USA).

Using Lipofectamine 2000 (Thermo Fisher Scientific), the indicated cells were transfected with the 8×GTIIC-luciferase vector (#34615, Addgene) and the Renilla luciferase-encoding pRL-TK plasmid. The firefly and Renilla luciferase activities were determined by the Dual-Luciferase Reporter Assay System (Promega, Madison, WA, USA) using a luminometer (Centro LB 960, Berthold Technologies, Bad Wildbad, Germany). The promoter-driven firefly luciferase activity was normalized to the pRL-TK (Renilla) luciferase activity to determine the relative luciferase activities.

Immunocytochemistry was performed as previously described (Park et al., 2021). Briefly, MCF-7 and MCF-7-ERα36 cells were cultured overnight on the coverslips, fixed with 4% paraformaldehyde for 20 min at room temperature, and incubated with 0.1% Triton X-100 for 15 min. The fixed cells were blocked with 10% horse serum for 1 h and incubated with the indicated primary antibody (1:200), followed by the incubation with fluorophore-conjugated secondary antibody (1:1000). Finally, the coverslips were washed with PBS and then mounted with ProLong Gold Antifade reagent with 4’,6-diamidino-2-phenylindole (DAPI; Invitrogen, Carlsbad, CA, USA). Images were obtained using confocal microscope (Leica TCS SP8 MP, Leica Microsystems, Wetzlar, Germany).

Cells were plated at a density of 3 × 103 in 96-well plates, and the phase confluence of the cells was monitored every 4 h using an IncuCyte S3 Live Cell Analysis System (Sartorius, Ann Arbor, MI, USA). For 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay, MCF-7-CTRL and MCF-7-ERα36 cell lines were cultured overnight on a 96-well plate at 4 × 103 cells/well, and then verteporfin was treated in a concentration-dependent manner. After 24 h, cells were exposed to 1 mg/ml MTT solution for 2 h, and 200 μL dimethyl sulfoxide (DMSO) was added after removing the MTT solution. The absorbance was measured at a wavelength of 540 nm using a SpectraMax i3x Multi-mode microplate reader (Molecular Devices, CA, USA).

Cells were dispensed into an Ultra-low attachment (ULA) plate (#7007, Corning Incorporated, Corning, NY, USA) at 2 × 103 cells/well and centrifuged at 300 g for 5 min. The tumor sphere images were captured using the Incucyte S3 Live Cell Analysis System at 168 h and the diameters were measured. The volume was calculated using the formula (length×width2)×π/6.

To observe the binding of V5-labeled ERα36 to FLAG-YAP (kindly donated from Dr. Kwang Youl Lee, Chonnam National University, Gwangju, South Korea), 2 μg of anti-FLAG antibody was added to 500 μg of cell lysates and reacted at 4°C for 16 h. Antigen-Antibody conjugate was captured with protein G agarose beads (Sigma) at 4°C for 1 h, then precipitated and boiled in a heat block for 5 min. The extracted protein was observed by immunoblotting.

The values were presented as means ± SD. Unpaired Student t test or one-way analysis of variance (ANOVA) was used to assess statistical significance. Results were considered significant when p < 0.05 (*, p < 0.05; **, p < 0.01; ***, p < 0.005; ****, p < 0.001).

To investigate the function of ERα36, we established a cell line that stably overexpresses ERα36 in MCF-7 cells, an ER-positive luminal A subtype among human breast cancer cells (Figure 1A). Overexpression of ERα36 in the MCF-7 cells significantly increased cell proliferation under the 10% FBS condition, as is consistent with the notion that ERα36 can lead to proliferation in several breast cancer cell lines (Mahboobifard et al., 2021) (Figure 1B). Next, we compared the 3D-sphere-forming ability on the seventh day after cells were dispensed onto a ULA plate. The volume of 3D spheres of the MCF-7-ERα36 cells was prominently elevated relative to the MCF-7-CTRL cells (Figure 1C). In addition, the MCF-7-ERα36 cells became, in their morphology, more like mesenchymal cell types than did the MCF-7-CTRL cells. As shown in Figure 1D, the protein expression of several EMT markers including E-cadherin, ZEB1 and Vimentin was significantly changed by overexpression of ERα36. Intriguingly, the estrogen dependency for proliferation was significantly reduced by overexpression of ERα36 in MCF-7 cells under the 10% charcoal-stripped FBS condition (Figure 1E). These data suggest that ERα36 overexpression could induce aggressive phenotypic changes in MCF-7 cells.

Higher ERα36 expression in breast cancer patients receiving tamoxifen treatment has been linked to worse survival in cohort studies (Shi et al., 2009). Treatment with tamoxifen also increases ERα36-positive breast cancer cell populations (Shi et al., 2009; Wang Q. et al., 2018). In order to identify novel genes associated with ERα36-related tamoxifen resistance, we analyzed the transcriptomes in MCF-7, tamoxifen-resistant MCF-7 (TAMR-MCF-7), and ERα36-MCF-7 cells. The data obtained revealed that the expression pattern of genes in the Hippo signaling pathway is very similar between TAMR-MCF-7 and MCF-7-ERα36 cells (Figure 2A). Because YAP, a downstream target gene of the Hippo pathway, is known to be related to tamoxifen sensitivity (Kim et al., 2022), we hypothesized that ERα36 might be involved in YAP activity in driving proliferation and promoting EMT phenotypes in MCF-7 cells.

Whereas the mRNA level of YAP in the MCF-7-ERα36 cells was comparable to that in the MCF-7 cells (Figure 2B), its protein level was highly increased in the MCF-7-ERα36 cells (Figure 2C). These results raise the possibility that YAP protein expression is post-translationally regulated by ERα36. Indeed, overexpression of ERα36 prolonged the half-life of YAP after treatment with cycloheximide, a protein synthesis inhibitor (Figure 2D). We next determined whether upregulated protein expression of YAP can lead to its increased activities in MCF-7-ERα36 cells. As shown in Figure 2E, a TEAD-reporter luciferase assay confirmed that YAP/TAZ-dependent transcription activity was highly increased in the MCF-7-ERα36 cells. Moreover, the mRNA levels of CYR61 and CTGF, representative downstream target genes of YAP, also were elevated in the same cells (Figure 2F). Immunocytochemistry results revealed that ERα36 overexpression induced nuclear localization of YAP (Figure 2G). These data demonstrate that ERα36 overexpression in MCF-7 cells can activate YAP by upregulation of its protein stability.

To investigate the role of YAP in the cell survival of ERα36-overexpressing MCF-7 cells, we used verteporfin, a small-molecule inhibitor of YAP, to block YAP-TEAD interaction (Liu-Chittenden et al., 2012). MTT assay results showed MCF-7-ERα36 cells’ higher sensitivity to verteporfin (IC₅₀ value: 8.6 μM in MCF-7 vs 3.0 μM in MCF-7-ERα36 cells), indicating that they are more dependent on YAP activity for survival than are MCF-7-CTRL cells.

We then established YAP-knockout MCF-7-ERα36 cells (sgYAP) using the CRISPR system (Figure 3B). A TEAD luciferase reporter assay confirmed that YAP-dependent transcription had been diminished in sgYAP MCF-7-ERα36 cells (Figure 3C). Interestingly, YAP knockout in MCF-7-ERα36 cells reduced cell proliferation, and this cell type was more sensitive to tamoxifen treatment (Figure 3D). In addition, the 3D spheroid volume and mesenchymal cell marker expression were decreased by YAP knockout in MCF-7-ERα36 cells (Figures 3E,F). These data suggest that YAP promotes the aggressive phenotypes acquired by ERα36 overexpression.

To clarify the detailed mechanism of how ERα36 can regulate the transcriptional activity of YAP, we first assessed whether ERα36 could directly bind to YAP. To that end, we transfected MCF-7 cells overexpressing ERα36-V5-His with FLAG-tagged YAP plasmid and performed FLAG immunoprecipitation to detect ERα36-V5. However, YAP did not bind to ERα36 (Figure 4A). We then explored whether YAP could be activated by indirect signaling pathway(s) in response to ERα36.

Note first that the protein expression and activity of YAP can be controlled by its phosphorylation. Specifically, phosphorylation of YAP at Ser127 promotes its cytoplasmic retention and degradation by 14-3-3ζ-mediated ubiquitination, thereby suppressing its transcriptional activity (Zanconato et al., 2016). In contrast, phosphorylation of Tyr357 and other tyrosine residues enhances nuclear translocation of YAP and increases its ability to stimulate transcription in the nucleus (Dasgupta and McCollum, 2019). Meanwhile, Src and Src-family kinases belong to the signal transduction pathways downstream of both ERα66 and ERα36, and are known to directly interact with ERα36 (Chu et al., 2007; Zhang et al., 2011; Pagano et al., 2020). In addition, Src can mediate the phosphorylation of YAP and promote its protein stability (Rosenbluh et al., 2012; Li et al., 2016; Byun et al., 2017; Smoot et al., 2018; Lamar et al., 2019). Src-family kinase especially increases the amounts of Tyr357-phosphorylated YAP, a representative nuclear form of YAP, regardless of LATS activity (Sugihara et al., 2018).

Surprisingly, ERα36 overexpression in MCF-7 cells promoted Tyr357 phosphorylation of YAP (an active form of YAP) (Figure 4B). Based on our result that Tyr416 phosphorylation of the Src family (an active form of Src) was also increased in MCF-7-ERα36 cells (Figure 4C), we evaluated the involvement of Src in ERα36-mediated phosphorylation of YAP at Tyr357. As expected, treatment with PP2, a Src-family kinase inhibitor, decreased the protein expression of both p-YAP (Tyr357) and CTGF (Figure 4D). We also found that saracatinib, the potent Src kinase inhibitor, reduces TEAD-reporter luciferase activity in MCF-7-ERα36 cells (Figure 4E). Taken together, our data indicate that Src activation plays a critical role in ERα36-induced YAP activation in MCF-7 cells.