Mutations in the neurofibromatosis type 2 (NF2) gene that limit or abrogate expression of functional Merlin are common in malignant mesothelioma. Merlin activates the Hippo pathway to suppress nuclear translocation of YAP and TAZ, the major effectors of the pathway that associate with the TEAD transcription factors in the nucleus and promote expression of genes involved in cell proliferation and survival. In this article, we describe the discovery of compounds that selectively inhibit YAP/TAZ-TEAD promoted gene transcription, block TEAD auto-palmitoylation, and disrupt interaction between YAP/TAZ and TEAD. Optimization led to potent analogs with excellent oral bioavailability and pharmacokinetics that selectively inhibit NF2-deficient mesothelioma cell proliferation in vitro and growth of subcutaneous tumor xenografts in vivo. These highly potent and selective TEAD inhibitors provide a way to target the Hippo-YAP pathway, which thus far has been undruggable and is dysregulated frequently in malignant mesothelioma and in other YAP-driven cancers and diseases.

This article is featured in Highlights of This Issue, p. 959

The Hippo pathway is an evolutionarily conserved signaling pathway that plays an essential role in controlling organ size and maintaining tissue homeostasis. It integrates environmental cues to regulate fundamental cellular processes such as cell proliferation, cell survival, cell migration, and cell fate decision (1–4). The core of the pathway consists of tightly regulated complexes of kinases (MST1/2, MAP4K1–7, and LATS1/2) and their binding partners (SAV1 and MOB1; ref. 5). Downstream from this kinase core are the YAP and TAZ transcriptional co-activators. When YAP and TAZ are phosphorylated by the activated LATS1/2 kinases, they remain in the cytoplasm, sequestered by the 14–3–3 proteins or degraded by ubiquitin-mediated proteasome proteolysis. When the Hippo kinase core is inactivated, YAP and TAZ become dephosphorylated and translocate into the nucleus, where they interact predominantly with the TEAD transcription factors, resulting in the activation of transcriptional programs important for cell proliferation, survival, and migration (5). There are four conserved homologues of TEAD (TEAD1–4) in mammals, and they share high degree of sequence homology and highly similar structural domain architectures (6).

One of the key upstream regulators of the Hippo kinase core is the Merlin protein, encoded by the neurofibromatosis type 2 (NF2) gene. Merlin has been shown to suppress YAP/TAZ nuclear translocation via its positive regulation of the Hippo pathway. Through its interaction with components of tight junction and adherence junction, Merlin is essential in the recruitment and assembly of the Hippo core kinases and binding partners to the plasma membrane, leading ultimately to the activation of the LATS1/2 kinases (7). Unsurprisingly for a signaling pathway essential for the normal function of a cell, dysregulation of YAP/TAZ activity is associated with pathogenesis. Genetic alterations of the Hippo pathway have been documented in a variety of human malignancies (8, 9). Germline loss-of-function mutations or deletion of the NF2 gene cause Neurofibromatosis type 2, an inheritable disorder characterized by the development of bilateral vestibular schwannomas (10). Somatic NF2 inactivating mutations also occur in spontaneous schwannomas, meningiomas, and other solid tumors, such as mesothelioma and renal cell carcinoma (10). In the TCGA PanCan 2018 Combined Study of 32 cancer indications (total 10,953 patients), malignant mesothelioma has the highest reported frequency of somatic NF2 mutations (32.18% of the 87 mesothelioma patients in the study have mutations in the NF2 gene; refs. 11, 12). Not included in this PanCan Combined Study is the meningioma indication. Roughly 50% of meningiomas have allelic losses in the chromosome region encoding the NF2 gene, and 54% to 78% of sporadic meningiomas have deletions in this region (13). Therefore, agents that block YAP/TAZ function have a great potential in treating NF2-deficient mesothelioma and other cancers with altered Hippo signaling.

Malignant pleural mesothelioma (MPM), typically caused by exposure to mineral fibers is an aggressive cancer with limited treatment options (14–16). The majority of patients with MPM are diagnosed at an advanced stage. For these patients, surgical resection is no longer an option, and the current standard of care is the combination chemotherapy of cisplatin and pemetrexed, which, along with its unwanted side effects, extends the median overall survival for only a few months (14, 17). As many targeted therapies have been investigated and tested in clinical trials for this aggressive cancer without success, MPM remains a deadly disease with unmet medical needs. Thus, there is need for new therapeutics that will effectively stop the progression of this aggressive cancer and significantly improve the median survival for patients.

YAP and TAZ are transcription co-activators with no known catalytic activity, rendering them poor candidates for direct inhibition with small molecules. Potentially druggable targets lie upstream of MST and LATS, including certain GPCRs, RASSF, ROCK, and FAK (18–21). However, inhibiting an upstream target that feeds into multiple signal transduction pathways presents a hazard of poor selectivity. Small molecules that block TEAD function have been the subject of considerable research. As the four TEAD homologues require auto-palmitoylation on the sulfhydryl of a conserved cysteine to become functional (22, 23), covalent and noncovalent inhibitors of TEAD auto-palmitoylation have been explored and reported (24–28). However, none has been shown to have in vivo activity at low doses perhaps due to poor pharmacokinetics properties. Herein, we report that a target-agnostic cellular screen for inhibitors of YAP/TAZ-TEAD-dependent transcription revealed compounds that block auto-palmitoylation of TEAD. Optimization of these screen hits provided new compounds with greatly increased potency and excellent oral pharmacokinetics. Importantly, these optimized compounds selectively inhibit proliferation of NF2-deficient mesothelioma cell lines in culture and stop the growth of subcutaneous xenografts in mice at clinically relevant oral doses.

Cell lines

The list of cell lines, their corresponding sources, and culture growth media can be found in Supplementary Table S3. 8XTBD Nos. 1–3 is an engineered HEK293T cell line harboring 8XTBD-driven firefly luciferase gene and thymidine kinase promoter-driven Renilla luciferase gene. NCI-H2373-Tu-P2 was derived from NCI-H2373 serially passaged in mice. Cell lines were authenticated by STR and tested mycoplasma free prior to cell proliferation assays and in vivo studies.

YAP reporter assay

8XTBD-driven firefly luciferase and control TK-driven Renilla luciferase activities in cells treated with compounds in dose titration starting from 3 μmol/L for 24 hours were measured using Promega Dual-Luciferase Reporter Assay System.

RNA extraction and qPCR

RNA was extracted using RNeasy Plus Mini Kit (QIAGEN, 74136) and reverse-transcribed using iScript reverse transcriptase (Bio-Rad, 1708891). qPCR was performed using KAPA SYBR FAST qPCR Kit (Kapa Biosystems, KK4605) and the 7300 real-time PCR system (Applied Biosystems). qPCR Primer sequences were previously described (18) and listed in Supplementary Table S3.

Protein expression and purification

His-TEV-TEAD1(209–426), His-TEV-TEAD2(217–447), His-TEV- TEAD3(216–435), and His-TEV-TEAD4(217–434) constructs were expressed in Escherichia coli BL21-Codonplus and purified by Ni-NTA columns followed by ion exchange chromatography and size exclusion chromatography using standard purification protocols for his-tagged proteins. The N-terminal His tag was removed by TEV protease during the purification.

Crystallography

The YAP-binding domain of TEAD3 was expressed, purified, and crystallized as described (23) with some modifications. The TEAD3 protein was S-deacylated by treatment with hydroxylamine and TEV cleaved to remove the His tag. Soaking these crystals with VT105 (2 mmol/L, 2% DMSO) provided a 2.6A X-ray crystal structure (PDB 7CNL).

Cell-free TEAD palmitoylation assay

Purified recombinant TEAD1–YBD was first incubated with compounds and then with 2 μmol/L alkyne-palmitoyl-CoA (APCoA; Cayman Chemical, item No. 15968). The reaction was quenched with 1% SDS followed by click chemistry reaction with biotin-azide as described previously (29). In some experiments, APCoA was added at different concentrations and in different sequence. Palmitoylated TEAD and total TEAD proteins were detected by streptavidin HRP (Life Technologies) and anti-TEAD1 antibody (Abcam) immunoblotting, respectively.

Cell-based TEAD palmitoylation assays

Myc-TEAD expression plasmid transfected HEK293T cells were treated with DMSO or 100 μmol/L alkyne palmitate + DMSO/compound for 20 hours. Myc-TEAD protein was immunoprecipitated with anti-Myc antibody and subjected to click chemistry. Palmitoylated TEAD was detected by streptavidin immunoblotting. The Acyl-PEGyl Exchange Gel-Shift Assay was performed as described previously (30).

Cell proliferation assay

Cells treated for various time periods with compounds in dose titration starting from 3 μmol/L were assayed by CellTiter-Glo Luminescent Cell Viability Assay Kit (Promega) according to the manufacturers' protocol. The IC50 and maximum inhibition % were calculated using dose response curves.

Immunofluorescence

After fixation with 4% paraformaldehyde (Electron Microscopy Sciences) for 10 to 15 minutes and permeabilization with 0.1% Triton X-100 in PBS, cells were blocked in 3% BSA in PBS for 1 to 2 hours at room temperature, stained with primary antibodies overnight at 4°C, and then with Alexa fluor-conjugated secondary antibodies for 2 to 3 hours at RT. Slides were mounted with prolong gold antifade reagent with DAPI (Invitrogen). Images were captured with a Nikon Eclipse Ti confocal microscope.

Immunoprecipitation

Cells were washed with PBS and lysed [50 mmol/L Tris pH 7.5, 150 mmol/L NaCl, 1% Triton-X100, 50 mmol/L NaF, 1 mmol/L PMSF, protease inhibitor cocktail (Biotool), phosphatase inhibitor (Biotool)]. After sonication and centrifugation, supernatant was collected and incubated with anti-TEAD, anti-YAP, or control antibodies, precipitated by Protein A/G beads, and analyzed by immunoblotting (see antibody information in Supplementary Table S3) using standard protocols.

Syntheses of VT103, VT104, VT105, VT106, and VT107

The optimized analogs were prepared by published procedures: VT103 (WO2019040380A1, page 223); VT104 (WO2020097389A1, page 196); VT105 (WO2020097389A1, page 207); VT106 (WO2020097389A1, page 228); VT107 (WO2020097389A1, page 228).

Mouse pharmacokinetics

VT103, VT104, and VT107, formulated in 5% DMSO + 10% Solutol + 85% D5W, were dosed intravenously or orally at 7 or 10 mg/kg. Blood was drawn from the saphenous vein at indicated timepoints. Compounds were quantified by LC/MS-MS using a QTRAP 6500. Data were analyzed using Phoenix WinNonlin 6.3, and intravenously noncompartmental model 201, and orally noncompartmental model 200. The calculation method was linear/log trapezoidal.

In vivo pharmacodynamic and efficacy studies

All the procedures related to animal handling, care, and the treatment were performed according to the guidelines approved by the Institutional Animal Care and Use Committee (IACUC) of WuXi AppTec or Crown Bioscience, Inc., following the guidance of the Association for Assessment and Accreditation of Laboratory Animal Care (AAALAC). The testing article formulated in dosing solution (5% DMSO + 10% solutol + 85% D5W; D5W = 5% glucose) was orally administrated daily at the indicated doses. Tumor volume and animal weights were monitored twice weekly. More detailed procedures are shown in the Supplementary Materials and Methods.

Screening identified inhibitors of TEAD1 auto-palmitoylation

A stable cell line was constructed to allow simultaneous measurement of effects on gene transcription promoted by YAP/TAZ–TEAD, and on control gene transcription not modulated by Hippo signaling. In this cell line, eight tandem repeats of a TEAD binding element upstream of a firefly luciferase reporter and a thymidine kinase promoter-driven Renilla luciferase reporter are stably integrated into the genome. Endogenously expressed YAP/TAZ and TEAD associate with the TEAD binding elements to drive the firefly luciferase signal. YAP/TAZ siRNA depleted the firefly luciferase signal severalfold, compared with control scrambled siRNA, and had no effect on the Renilla luciferase signal (Supplementary Fig. S1). Hereafter, we will refer to this assay as “the YAP reporter assay.” The YAP reporter assay was used in a high-throughput screen (HTS) of a library of 160K diverse compounds at 5 μmol/L, and we identified 153 compounds that achieved >80% inhibition of the firefly reporter signal, with satisfactory dose response, and no inhibition of the Renilla reporter signal. Pure, solid samples of more than half of the 153 hits were obtained, and these were tested in the validation assays described later. Most compounds afforded undesirable data in one or more of the validation assays, but two compounds, VT101 and VT102, succeeded in all assays, and we showcase data for these two compounds here.

VT101 and VT102 inhibited the firefly luciferase reporter dose-dependently and had no effect on the Renilla luciferase reporter (Fig. 1A). CTGF is one of the best characterized target genes of YAP/TAZ. To determine whether the screen hits specifically inhibit YAP/TAZ–TEAD transcriptional activity, the compounds were tested in wildtype CTGF-promoter and mutant CTGF-promoter driven reporter assays. The mutant CTGF promoter harbors a disrupted TEAD-binding element, which does not respond to TEAD. We set aside hits that were active in the mutant CTGF-promoter reporter assay, such as the one shown in Supplementary Fig. S2A. VT101 and VT102 were not active in the mutant CTGF promoter reporter assay (Supplementary Fig. S2B). Furthermore, inhibition on YAP/TAZ–TEAD transcriptional activity was confirmed by qPCR analysis of pathway target genes (31), and we set aside hits that failed to reduce the transcript levels of these genes. VT101 and VT102 significantly downregulated the expression of CYR61 and AMOTL2 in the MCF10A cell line (Fig. 1B). In addition, VT102 downregulated the expression of CYR61 in NCI-H2373, a human mesothelioma cell line that harbors homozygous deletion of the NF2 gene (32), not only at 3 μmol/L but also at 0.3 μmol/L (Fig. 1C).

Figure 1.

Results of testing VT101 and VT102 in validation assays. A, Dose–response curves from single instances of the YAP reporter assay. Each open circle represents a well of cells treated with the compound at the indicated concentration. Each concentration was tested in quadruplicate. B, Effects on transcription of CYR61 and AMOLT2 in MCF10A cells measured by qPCR. Cells were treated with compounds at 3 μmol/L for 4 hours. In each MCF10A qPCR experiment, all samples were assayed in triplicates (technical replicates). The VT101 MCF10A qPCR result consisted of data from two independent experiments performed on different days (biologic replicates), whereas the VT102 qPCR result was obtained from one experiment in which the samples were assays in triplicates. C, Effects on transcription of CYR61 in NCI-H2373 cells measured by qPCR. Cells were treated with VT102 at 0.3 or 3 μmol/L for 4 hours. All samples were assayed in triplicates. D and E, Effects on immunofluorescent staining of YAP/TAZ and TEAD in NCI-H28 cells after treatment with DMSO or compound (3 μmol/L) for 24 hours. F, Effects on co-immunoprecipitation of YAP and TEAD from lysate of NCI-H2373 cells after treatment with DMSO or compound (3 μmol/L) for 4 hours. G and H, Effects on auto-palmitoylation of the purified recombinant YAP-binding domain of TEAD1 in the presence of palmitoyl-CoA with DMSO or compound at the indicated concentrations. The anti-streptavidin immunoblot detects palmitoylated TEAD1.

Figure 1.

Results of testing VT101 and VT102 in validation assays. A, Dose–response curves from single instances of the YAP reporter assay. Each open circle represents a well of cells treated with the compound at the indicated concentration. Each concentration was tested in quadruplicate. B, Effects on transcription of CYR61 and AMOLT2 in MCF10A cells measured by qPCR. Cells were treated with compounds at 3 μmol/L for 4 hours. In each MCF10A qPCR experiment, all samples were assayed in triplicates (technical replicates). The VT101 MCF10A qPCR result consisted of data from two independent experiments performed on different days (biologic replicates), whereas the VT102 qPCR result was obtained from one experiment in which the samples were assays in triplicates. C, Effects on transcription of CYR61 in NCI-H2373 cells measured by qPCR. Cells were treated with VT102 at 0.3 or 3 μmol/L for 4 hours. All samples were assayed in triplicates. D and E, Effects on immunofluorescent staining of YAP/TAZ and TEAD in NCI-H28 cells after treatment with DMSO or compound (3 μmol/L) for 24 hours. F, Effects on co-immunoprecipitation of YAP and TEAD from lysate of NCI-H2373 cells after treatment with DMSO or compound (3 μmol/L) for 4 hours. G and H, Effects on auto-palmitoylation of the purified recombinant YAP-binding domain of TEAD1 in the presence of palmitoyl-CoA with DMSO or compound at the indicated concentrations. The anti-streptavidin immunoblot detects palmitoylated TEAD1.

Close modal

To decipher mechanisms of action of the screening hits in the Hippo pathway, we examined the hits for effects on YAP phosphorylation, YAP cellular localization, YAP–TEAD protein–protein interaction, and TEAD auto-palmitoylation. Because it has been reported that active SRC promotes YAP nuclear localization, and dasatinib, a tyrosine-kinase inhibitor that blocks a number of tyrosine kinases such as Bcr-Abl and the Src kinase family, induces YAP cytoplasmic localization (33), we used dasatinib as control to validate our YAP localization and YAP phosphorylation assays (Supplementary Fig. S2C). The screen hits, VT101 and VT102, did not prevent YAP/TAZ or TEAD nuclear translocation (Fig. 1D and E). However, they appeared to partially prevent formation of YAP-TEAD heterodimers (Fig. 1F; Supplementary Fig. S2D). Furthermore, in a dose-dependent manner, VT101 and VT102 blocked recombinant TEAD1 auto-palmitoylation in the presence of alkyne-palmitoyl-CoA (APCoA) spiked in at 2 μmol/L (Fig. 1G and H). The failure to affect YAP nuclear localization and the activity in inhibiting the YAP–TEAD co-immunoprecipitation and TEAD auto-palmitoylation suggested that VT101 and VT102 might bind directly to TEAD.

Optimization rovided more useful compounds

On the basis of the promising data generated for VT101 and VT102 (chemical structures shown in Fig. 2), we prepared and tested numerous analogs of these two compounds (WO2019040380A1, WO2020097389A1). Details of the structure-activity relationship will be reported elsewhere (Konradi and colleagues, manuscript in preparation). Key new analogs (chemical structures shown in Fig. 2), for which we report data herein, are as follows. VT103 is an analog of VT101, which has improved potency and good oral pharmacokinetics in mice (Fig. 2; Supplementary Table S1). VT104 is an analog of VT102, which has improved potency and good oral pharmacokinetics in mice (Fig. 2; Supplementary Table S1). VT105 is a more soluble analog of VT104 (Fig. 2), which was useful in TEAD X-ray crystallography experiments. VT106 and VT107 are enantiomers analogous to VT104; they have quite different potencies, making them useful mutual controls in biochemical and cellular experiments (Fig. 2).

Figure 2.

Validated HTS hits and optimized analogs. Chemical structures and YAP reporter assay data of validated HTS hits and optimized analogs. Pharmacokinetics (PK) data of selected optimized analogs were obtained by administrating mice, intravenously (i.v.) or orally (p.o.), with a single dose of each compound: VT103 (7 mg/kg), VT104 (10 mg/kg), and VT107 (10 mg/kg).

Figure 2.

Validated HTS hits and optimized analogs. Chemical structures and YAP reporter assay data of validated HTS hits and optimized analogs. Pharmacokinetics (PK) data of selected optimized analogs were obtained by administrating mice, intravenously (i.v.) or orally (p.o.), with a single dose of each compound: VT103 (7 mg/kg), VT104 (10 mg/kg), and VT107 (10 mg/kg).

Close modal

Optimized compounds selectively block palmitoylation of the four TEAD homologues

The highly potent analogs were tested for inhibition of palmitoylation of the four different TEAD homologues in human cells. In these cell-based TEAD palmitoylation assays, HEK293T cells were transfected with plasmid expressing MYC-tagged full-length TEAD1, TEAD2, TEAD3, or TEAD4 protein. After overnight incubation of the indicated compound and alkyne palmitate, TEAD was immunoprecipitated by anti-MYC antibody and subjected to click chemistry and Streptavidin immunoblotting. All three compounds, VT103, VT104, and VT107, prevented palmitoylation of the TEAD1 protein (left panel in Fig. 3A). Interestingly, VT103 appears to be TEAD1-selective, as it did not block palmitoylation of TEAD2, TEAD3, or TEAD4, while VT104 and VT107 appear to be pan-TEAD palmitoylation inhibitors, with VT107 slightly more potent than VT104 on TEAD2 and TEAD4. In contrast, VT106, the ∼50-fold less active enantiomer of VT107, failed to block palmitoylation of all four TEAD proteins (Fig. 3A). A single high dose of 3 μmol/L compound concentration was used because the experiments were done with overexpression of TEAD proteins and high concentration (100 μmol/L) of spiked in alkyne palmitate, and, more importantly, because we wanted to determine the TEAD selective activity of these compounds.

Figure 3.

Optimized compounds show selective inhibition of TEAD palmitoylation in cell. A, HEK293T cells transfected with Myc-TEAD1, Myc-TEAD2, Myc-TEAD3, or Myc-TEAD4 expression plasmid were cultured in the absence of (lane 2 in each panel) or presence of 100 μmol/L alkyne-palmitate overnight with DMSO or the indicated compound at 3 μmol/L (lane 1 and lanes 3–6 in each panel). Palmitoylated TEAD was detected by Streptavidin-HRP whereas total TEAD was detected by MYC-HRP Western blotting. B, Endogenous TEAD immunoprecipitation followed by click chemistry from HEK293T cells treated with the listed conditions for 20 hours. C, Palmitoylation status of endogenous TEAD1, TEAD3, and TEAD4 in NCI-H2373 examined by APEGS Assay (30) after overnight compound (3 μmol/L VT103 or VT107) treatment. In this APEGS assay, mPEG-2k (final 20 mmol/L) was used.

Figure 3.

Optimized compounds show selective inhibition of TEAD palmitoylation in cell. A, HEK293T cells transfected with Myc-TEAD1, Myc-TEAD2, Myc-TEAD3, or Myc-TEAD4 expression plasmid were cultured in the absence of (lane 2 in each panel) or presence of 100 μmol/L alkyne-palmitate overnight with DMSO or the indicated compound at 3 μmol/L (lane 1 and lanes 3–6 in each panel). Palmitoylated TEAD was detected by Streptavidin-HRP whereas total TEAD was detected by MYC-HRP Western blotting. B, Endogenous TEAD immunoprecipitation followed by click chemistry from HEK293T cells treated with the listed conditions for 20 hours. C, Palmitoylation status of endogenous TEAD1, TEAD3, and TEAD4 in NCI-H2373 examined by APEGS Assay (30) after overnight compound (3 μmol/L VT103 or VT107) treatment. In this APEGS assay, mPEG-2k (final 20 mmol/L) was used.

Close modal

The effect of these compounds on palmitoylation of endogenous TEAD1, TEAD3, and TEAD4 proteins was examined by using specific TEAD antibodies to immunoprecipitate the respective TEAD protein in question after overnight incubation of the cells with compounds and alkyne palmitate, followed by click chemistry and Streptavidin Western blotting analysis. VT103 selectively inhibited endogenous TEAD1 palmitoylation in cells. VT104 and VT107 inhibited palmitoylation of both endogenous TEAD1 and TEAD3 proteins. VT107 was the most potent at blocking the palmitoylation of endogenous TEAD4 protein. VT106 weakly inhibited TEAD1 and TEAD3 palmitoylation and has no effect on TEAD4 palmitoylation (Fig. 3B). TEAD2 was not examined because it is expressed at very low levels in HEK293T cells and could not be sufficiently immunoprecipitated.

The compounds' effect on endogenous TEAD in the human mesothelioma cell line NCI-H2373 was assessed by Acyl-PEGyl Exchange Gel-Shift (APEGS) Assay (30), which allows detection of both palmitoylated and unpalmitoylated forms of TEAD proteins in whole cell lysates (see assay validation in Supplementary Fig. S3A). Treatment with VT103 and VT107 resulted in the disappearance of palmitoylated TEAD1 with a concomitant increase in unpalmitoylated TEAD1 (Fig. 3C, top left). TEAD-selective activity was also observed in this assay. VT107 decreased the levels of palmitoylated TEAD3 and TEAD4 and increased the levels of unpalmitoylated TEAD3 and TEAD4, while VT103 has limited effect on TEAD3 and TEAD4 palmitoylation, relative to its effect on TEAD1 palmitoylation (Fig. 3C, bottom left and top right). The compounds had no effect on RAS palmitoylation (Supplementary Fig. S3B).

Optimized compounds bind directly to TEAD in the central lipid pocket

To determine if these compounds bind directly to TEAD, thermal shift assays (TSA) were performed using purified recombinant TEAD1-4 proteins. The highly potent analogs demonstrated high affinity interaction with TEAD proteins, inducing significant shift in the melting temperatures of the TEAD proteins (Fig. 4A). Upon co-incubation with VT103, TEAD1 showed the highest increase in melting temperature—a shift of 8.3°C—compared with other members of the TEAD family. The thermal denaturation curves clearly showed two separate peaks for TEAD1 alone (red curve) and TEAD1+VT103 (blue curve), while the peaks of the ±VT103 curves remained largely overlapping for the other TEAD proteins (Fig. 4A, top). This is consistent with the finding from the functional palmitoylation assays that VT103 is a TEAD1-selective inhibitor. On the other hand, VT107, which was determined to be a pan-TEAD inhibitor by TEAD palmitoylation assays, significantly shifted the melting temperatures of all four TEAD family members (Fig. 4A). VT104 shifted the melting temperatures of all four TEAD family members, but higher shifts were observed for TEAD1 and TEAD3. VT106 only weakly shifted the melting temperatures of all four TEADs.

Figure 4.

Compounds bind TEAD in the central lipid pocket. A, TSAs of recombinant TEAD proteins alone or in the presence of the indicated inhibitors. Crystal structure of TEAD3 protein with a molecule of CH3(CH2)12CONHOH (B) or a molecule of VT105 (C) in the central hydrophobic pocket. Cysteine 368 is shown in orange, and lysine 345 is shown in purple (PDB 7CNL). D, Cell-free TEAD1 palmitoylation assays in which VT107 at increasing concentrations were incubated with TEAD1 recombinant protein and with the indicated concentration of APCoA spiked in sequentially (top) or concurrently (bottom). E, Quantification of relative levels of palmitoylated TEAD1.

Figure 4.

Compounds bind TEAD in the central lipid pocket. A, TSAs of recombinant TEAD proteins alone or in the presence of the indicated inhibitors. Crystal structure of TEAD3 protein with a molecule of CH3(CH2)12CONHOH (B) or a molecule of VT105 (C) in the central hydrophobic pocket. Cysteine 368 is shown in orange, and lysine 345 is shown in purple (PDB 7CNL). D, Cell-free TEAD1 palmitoylation assays in which VT107 at increasing concentrations were incubated with TEAD1 recombinant protein and with the indicated concentration of APCoA spiked in sequentially (top) or concurrently (bottom). E, Quantification of relative levels of palmitoylated TEAD1.

Close modal

To further demonstrate direct physical interaction of the compounds with TEAD protein, we explored expression and crystallization of various reported and novel constructs of the YAP-binding domains of human TEADs 1, 2, 3, and 4. We replicated a known crystal form of TEAD3 (PDB 5emw), which Noland and colleagues reported includes an S-palmitoyl group on cysteine 371 (labeled as cysteine 368 in our crystal structure) in each of the four units of TEAD3 in the asymmetric unit (23). However, mass spectrometry of TEADs produced in E. coli indicated both S-myristoylation (+210Da) and S-palmitoylation (+239Da) (Supplementary Figs. S4A–S4D, left). Having proven by TSA that the inhibitors bind to TEADs that had been S-deacylated by treatment with hydroxylamine (Supplementary Figs. S4A–S4D, right), crystallization of TEAD3 was performed using TEAD protein that had been S-deacylated by this method (Supplementary Fig. S4C, right) and un-tagged by TEV cleavage. The result was a nearly identical crystal form, wherein each of the four units of TEAD3 in the asymmetric unit is occupied by CH3(CH2)12CONHOH, in a noncovalent complex, as determined by X-ray crystallography (Fig. 4B). After soaking these crystals with inhibitors, X-ray crystallography revealed that all four CH3(CH2)12CONHOH in the asymmetric unit can be displaced, but two are more easily displaced. Soaking with VT105 provided a crystal structure with 2.6 Angstrom resolution, wherein two molecules of VT105 and two molecules of CH3(CH2)12CONHOH are present in the asymmetric unit. The molecules of VT105 and the molecules of CH3(CH2)12CONHOH occupy the same location in TEAD3 (Fig. 4C). The quinoline and the 4-trifluoromethylphenyl groups of VT105 fill a cavity in TEAD3 that is bounded by hydrophobic residues. The carbonyl group of VT105 accepts a hydrogen bond from the backbone NH of cysteine 368. The pyridine group of VT105 is sandwiched between the sidechains of lysine 345 and cysteine 368, and nearly protrudes beyond the surface of TEAD3.

On the basis of the crystal structure, it appears that the compounds inhibit TEAD auto-palmitoylation by blocking the palmitoyl-CoA binding site. Thus, cell-free palmitoylation assays were performed with purified TEAD1 protein incubated with VT107 in dose titration and different concentrations alkyne palmitoyl-CoA (APCoA) spiked in sequentially or concurrently (Fig. 4D and E). When APCoA was added to the reaction after VT107 was allowed to incubate first with TEAD1 for 30 minutes, increasing the concentration of alkyne palmitoyl-CoA 4-fold did not overcome inhibition by VT107 (Fig. 4D and E, top). This indicates that once bound by VT107, TEAD1 is no longer accessible to palmitate binding and modification. In contrast, when VT107 and APCoA were added to TEAD1 at the same time, higher concentration of APCoA more effectively overcame the inhibition by VT107 (Fig. 4D and E, bottom). In addition, VT107 was less efficacious at blocking TEAD1 palmitoylation in concurrent incubation than in sequential incubation with APCoA. These results, which were repeated in two independent experiments, indicate a competition for the same binding pocket consistent with the crystal structure.

Optimized compounds disrupt YAP/TAZ–TEAD protein–protein interaction

To determine whether our optimized compounds prevent YAP/TAZ–TEAD protein–protein interaction in the cell in TEAD-selective manner, we treated the NF2-mutant NCI-H2373 cells with VT103 or VT107 for 4 or 24 hours, immunoprecipitated the endogenous TEAD1 and TEAD4 protein using TEAD1-specific antibody and TEAD4-specific antibody (Fig. 5A and B), respectively, and probed the immunocomplexes with anti-YAP and anti-TAZ antibodies. Consistent with its selective binding to TEAD1 protein and selective inhibition of TEAD1 palmitoylation, VT103 reduced YAP interaction with TEAD1 but not TEAD4 after 4- and 24-hour treatment. The TAZ–TEAD1 interaction was disrupted with 4-hour treatment of VT103. In contrast, VT107 blocked YAP and TAZ interaction with both TEAD1 and TEAD4, with stronger effect at 24 than 4 hours (Fig. 5A and B). The less active enantiomer, VT106, failed to block the YAP/TAZ–TEAD4 interactions (Fig. 5B) and only weakly disrupted YAP/TAZ–TEAD1 interactions after 24-hour treatment (Fig. 5A). VT103 also selectively disrupted YAP–TEAD1 interaction in the NF2-deficient NCI-H226 cells after 4- and 24-hour treatment (Fig. 5C and D).

Figure 5.

Disruption of YAP/TAZ-TEAD protein–protein interaction by TEAD palmitoylation inhibitors. Co-immunoprecipitation of (A) YAP/TAZ-TEAD1 and (B) YAP/TAZ-TEAD4 from NCI-H2373 treated with the indicated compounds at 3 μmol/L for 4 or 24 hours. Co-immunoprecipitation of (C) YAP-TEAD1 and (D) YAP-TEAD4 from NCI-H226 treated with VT103 at 3 μmol/L for 4 or 24 hours.

Figure 5.

Disruption of YAP/TAZ-TEAD protein–protein interaction by TEAD palmitoylation inhibitors. Co-immunoprecipitation of (A) YAP/TAZ-TEAD1 and (B) YAP/TAZ-TEAD4 from NCI-H2373 treated with the indicated compounds at 3 μmol/L for 4 or 24 hours. Co-immunoprecipitation of (C) YAP-TEAD1 and (D) YAP-TEAD4 from NCI-H226 treated with VT103 at 3 μmol/L for 4 or 24 hours.

Close modal

TEAD palmitoylation inhibitors blocked proliferation of NF2-deficient mesothelioma in vitro

Frequent genetic alterations in the Hippo–YAP/TAZ–TEAD pathway have been documented in malignant mesothelioma, leading to constitutive YAP activation in more than 70% of malignant mesothelioma (ref. 34 and references therein). Importantly, YAP in cooperation with TEAD transcription factor has been shown to be functionally essential for malignant mesothelioma cell proliferation and anchorage-independent growth (35). The TEAD palmitoylation inhibitors, that potently inhibit the YAP reporter, were found to stop the proliferation of NF2-deficient or NF2-mutated mesothelioma cell lines. Demonstrating the selectivity of the compounds, they showed 100- to 1,000-fold less activity in NF2 wild-type mesothelioma cells (Fig. 6A).

Figure 6.

Inhibition of NF2-deficient mesothelioma cell proliferation and tumor growth by TEAD palmitoylation inhibitors. A, Dose–response curves from single instances of the cell proliferation assay. Each open circle represents a well of cells treated with the indicated compound at a particular concentration. Each concentration was tested in duplicate. The dose titration (a 10-point threefold serial dilution) started at a top concentration of 3 μmol/L. For each graph, X-axis = [Compound] Log10(μmol/L), and Y-axis = Cell proliferation (%). Antiproliferation IC50 values for the indicated compounds in the indicated mesothelioma cell lines are listed. Maximum % inhibitions less than 90% are shown in parentheses (). B, qPCR analysis of human CTGF and CYR61 gene expression in tumors harvested from vehicle or VT103-treated (once per day x 3 days at the indicated doses) NCI-H226-tumor bearing mice (n = 4 per group). C,In vivo efficacy of VT103 in NCI-H226 CDX model (n = 5 per group). D, APEGS analysis of TEAD1 palmitoylation level and state and (E) qPCR analysis of human CTGF gene expression in tumors from vehicle or VT103-treated (once per day x 3 days at 10 mg/kg) NCI-H2373-tumor bearing mice (n = 3 per group). F,In vivo efficacy of VT103 in NCI-H2373-Tu-P2 CDX model (n = 5 per group). G,In vivo efficacy of VT104 in NCI-H226 CDX model (n = 5 per group).

Figure 6.

Inhibition of NF2-deficient mesothelioma cell proliferation and tumor growth by TEAD palmitoylation inhibitors. A, Dose–response curves from single instances of the cell proliferation assay. Each open circle represents a well of cells treated with the indicated compound at a particular concentration. Each concentration was tested in duplicate. The dose titration (a 10-point threefold serial dilution) started at a top concentration of 3 μmol/L. For each graph, X-axis = [Compound] Log10(μmol/L), and Y-axis = Cell proliferation (%). Antiproliferation IC50 values for the indicated compounds in the indicated mesothelioma cell lines are listed. Maximum % inhibitions less than 90% are shown in parentheses (). B, qPCR analysis of human CTGF and CYR61 gene expression in tumors harvested from vehicle or VT103-treated (once per day x 3 days at the indicated doses) NCI-H226-tumor bearing mice (n = 4 per group). C,In vivo efficacy of VT103 in NCI-H226 CDX model (n = 5 per group). D, APEGS analysis of TEAD1 palmitoylation level and state and (E) qPCR analysis of human CTGF gene expression in tumors from vehicle or VT103-treated (once per day x 3 days at 10 mg/kg) NCI-H2373-tumor bearing mice (n = 3 per group). F,In vivo efficacy of VT103 in NCI-H2373-Tu-P2 CDX model (n = 5 per group). G,In vivo efficacy of VT104 in NCI-H226 CDX model (n = 5 per group).

Close modal

To further investigate whether TEAD inhibitors also block proliferations of malignant mesothelioma cells harboring mutations in Hippo pathway components other than Merlin, the compounds were tested in a panel of 26 mesothelioma cell lines. The inhibitors with broader-spectrum TEAD inhibition, such as VT104 and VT107, exhibited inhibitory activity in more mesothelioma cell lines than the TEAD1-selective inhibitor VT103 (Supplementary Table S2). All three compounds potently inhibited the proliferation of NF2-mutated/deficient cell lines in vitro, but only the broader-spectrum TEAD inhibitors showed antiproliferation activity in additional mesothelioma cell lines without NF2 mutations, but harboring alterations in other Hippo pathway core components (Supplementary Table S2). We confirmed literature reports of the NF2 mutations and loss of endogenous Merlin protein expression in these mesothelioma cell lines via whole exome sequencing (WES) and Western blotting analysis (ref. 36; Supplementary Table S2; Supplementary Fig. S5). We also identified and confirmed their mutational status for the Hippo pathway core components by RNA-seq. A subset of mesothelioma cell lines, which apparently did not harbor NF2 mutations detectable by WES, showed low/undetectable Merlin protein levels perhaps because of undetected mutations or epigenetic effects.

VT103 and VT104 block growth of NF2-deficient mesothelioma xenografts

With excellent oral bioavailability (≥75%) and long half-life (>12 hours) in mice (Fig. 2), VT103 and VT104 allowed us to evaluate target engagement and antitumor efficacy of TEAD auto-palmitoylation inhibition in human mesothelioma xenograft models in vivo. As shown in Fig. 6B, 4 hours after the third daily dose, VT103 significantly downregulated the expression of the Hippo pathway target genes, CTGF and CYR61, in the NF2-deficient NCI-H226 tumors in mice in a dose-dependent manner. Within the same NCI-H226-tumor bearing animals, VT103 also downregulated target gene expression in kidneys and livers in dose-dependent manner (Supplementary Fig. S6A). However, there did not seem to be any pathologic effect at the same time point as H&E images of kidneys and livers showed no difference between vehicle and drug-treated groups (Supplementary Fig. S6B). Bioanalysis showed that there was dose-dependent exposure of VT103 in circulation as well as in tumor tissues (Supplementary Fig. S6C). In addition, within the same animal, there appeared to be more compound accumulation in tumor than in circulation (Supplementary Fig. S6C). Examination of YAP and TEAD1 proteins in the VT103-treated NCI-H226 tumor tissues by IHC indicated no change in the cellular localization or levels of these proteins (Supplementary Fig. S6D).

VT103 significantly blocked NCI-H226 tumor growth in mice in dose-dependent manner (Fig. 6C). NCI-H226 tumor-bearing mice were randomized and daily oral administration of VT103 was initiated when tumors reached approximately 110 mm3. The last dose was administered on day 45. Plasma samples were collected on day 46. Bioanalysis showed the expected drug concentrations in circulation in dose-dependent manner (Supplementary Fig. S6E). VT103 exhibited strong antitumor efficacy, leading to tumor regression, when orally administered at 3 mg/kg once daily (TGI = 106.14%, P < 0.001, by day 46) in the NCI-H226 CDX model. Given at 1 and 0.3 mg/kg once daily, VT103 also significantly inhibited tumor growth (TGI = 83.79%, P < 0.001, and 47.95%, P = 0.001, respectively). No adverse effect on body weights were observed during such a long-term treatment with daily dosing of VT103 (Fig. 6C).

NCI-H2373 is a human mesothelioma cell line that harbors homozygous deletion of the NF2 gene (32). Although this cell line does not grow tumors well in mice (Supplementary Fig. S7 and described in detail below), we were able to perform a short-term drug treatment to assess VT103 target engagement and effect on pathway target gene expression in a second human mesothelioma model using this line. The parental NCI-H2373 cells were injected subcutaneously in mice, and when tumors reached approximately 300 mm3 in size, the mice were treated by oral administration of VT103 at 10 mg/kg once daily for 3 days. APEGS assay analysis of tumors harvested 4 hours after the third dose from these VT103-treated NCI-H2373 tumor bearing mice indicated that VT103 reduced the level of lipid-modified form of TEAD1 relative to total TEAD1 by 40.0±0.06% (Fig. 6D; Supplementary Fig. S8). In these NCI-H2373 tumors, 5-fold downregulation of the CTGF gene expression was observed (Fig. 6E; we did not check CYR61 expression in this model study).

Although measurable tumors were observed by day 7 after subcutaneous injection of NCI-H2373 cells in mice, these tumors grew very slowly for the following 4 weeks, with some regressing during this period (Supplementary Fig. S7). Thus, even though sufficient for short-term pharmacodynamic study, this was not a desirable model for drug efficacy study. To obtain a second working in vivo model for evaluating antitumor efficacy in long-term treatment, we serially passaged the NCI-H2373 tumors in mice and derived the NCI-H2373-Tu-P2 line, which was confirmed to be genetically related to NCI-H2373 by STR and sensitive to our TEAD inhibitors by in vitro cell proliferation assay. In this second human mesothelioma CDX model, NCI-H2373-Tu-P2, we also observed significant tumor regression when VT103 was orally administered once daily at 10, 3, and 1 mg/kg (TGI = 126.70%, 118.32%, 110.51%; P = 0.018, 0.022, and 0.030; Fig. 6F). Even at 0.3 mg/kg, VT103 blocked tumor growth in vivo, albeit not statistically significant (TGI value = 76.43%, P = 0.124). Animals in all four dose groups of VT103 had no body weight loss (Fig. 6F). Given the difficulty of growing NF2 mutant mesothelioma tumors in vivo, we were able to repeat the efficacy study twice before the model was eventually lost. The results indicated that VT103 is highly active in vivo at tolerated doses, and worthy of further evaluation.

Using the NCI-H226 CDX model, we evaluated another TEAD auto-palmitoylation inhibitor, VT104, with excellent pharmacokinetics (F = 78%; T1/2 = 24.2 hours) in mice (Fig. 2). Like VT103, VT104 demonstrated significant antitumor activity in vivo (Fig. 6G). Once daily oral administration of 10 and 3 mg/kg worked equally well and resulted in tumor regression (TGI = 103.67%, P < 0.001 and 102.49%, P < 0.001, respectively). However, animals receiving 10 mg/kg stopped gaining body weight during treatment, whereas 3 mg/kg had no effect on body weight gain. Daily 1 mg/kg dosing also significantly blocked tumor growth (TGI = 87.12%, P < 0.001; Fig. 6G). Thus, VT104 has very strong antitumor efficacy in the human mesothelioma NCI-H226 CDX model.

The importance of the Hippo–YAP/TAZ–TEAD signaling pathway in multiple cancer types creates a compelling case for evaluation as a therapeutic strategy. This report describes a target-agnostic screen for inhibitors of YAP-driven gene expression, which produced hits that were optimized into potent, selective, in vivo active TEAD palmitoylation inhibitors. These compounds bind TEAD noncovalently in the central hydrophobic pocket of TEAD, preventing TEAD autopalmitoylation and TEAD protein interaction with the YAP/TAZ transcription coactivators. The compounds (VT103 and VT104) with excellent orally bioavailability and long half-lives inhibit tumor growth and also lead to shrinkage of established tumors in preclinical models of human mesothelioma CDX models deficient in functional Merlin.

Verteporfin, the first small molecule compound identified as inhibitor of YAP–TEAD interaction, is commonly used as a tool compound in the Hippo pathway research (37). However, due to its mechanism of action as a photosensitizer, verteporfin introduces nonspecific effects into experiments designed to study YAP–TEAD function. In this study, we showed that a pair of enantiomers, VT106 and VT107, have quite different potencies in our biochemical and cellular experiments, and hence, highlight the specificity of the potent TEAD inhibitors. They serve as useful tool for investigating YAP/TAZ–TEAD function and TEAD biology.

Covalent and noncovalent TEAD palmitoylation inhibitors have been described (24–28), but most demonstrated, if any, weak in vitro cellular activity, requiring micromolar to millimolar concentrations, and no in vivo data. The two reports of in vivo TEAD inhibition data showed compounds with high clearance and short half-lives, requiring frequent and high doses to achieve in vivo response (27, 28). VT103 and VT104 reported here potently inhibit TEAD activity, selectively suppress the proliferation of NF2-deficient mesothelioma cells in vitro at nanomolar concentrations, and with excellent oral bioavailability and long half-lives, significantly block tumor growth in vivo at low doses (MED = 1–3 mg/kg, p.o., once a day).

The research described herein does not address the significance of inhibiting the auto-palmitoylation of different TEAD homologs to different degrees. The central hydrophobic pocket of TEAD where our TEAD inhibitors bind is highly homologous among the four TEAD homologues. It was therefore unexpected to find different classes of inhibitors with differential TEAD selectivity in cell-base palmitoylation and YAP/TAZ–TEAD protein–protein interaction assays. Crystal structure analysis also did not provide explanation for such selectivity. One possibility is that the TEAD homologs have different turn-over rates in their unpalmitoylated state. Further investigation to elucidate TEAD selectivity by our inhibitors, probably involving more diverse compounds and extensive kinetics studies, would be desirable. Nevertheless, the identification of TEAD isoform selective inhibitors is valuable as research tools to probe function of different TEAD members and potential anticancer agents with selectivity and less toxicity.

It has been reported that palmitoylation does not alter TEAD cellular localization but is essential for TEAD interaction with YAP/TAZ (22). With our TEAD auto-palmitoylation inhibitors, we also demonstrated that palmitoylation is not required for nuclear localization of endogenous TEAD proteins but is necessary for TEAD to interact with YAP/TAZ. However, blocking palmitoylation by our compounds did not reduce TEAD protein stability in contrast to the finding by Noland and colleagues (23). The discrepancy could result from different assay conditions, such as overexpressed TEAD versus endogenous TEAD and/or the time points at which the proteins were assayed. More quantitative approaches will be necessary to address whether palmitoylation is truly critical for TEAD protein stability and for the regulation of TEAD protein turn-over rates. Despite these apparently different mechanisms of how palmitoylation regulates TEAD function, all reports consistently have found that palmitoylation is absolutely essential for TEAD transcriptional activity.

Loss of Merlin function is observed at high frequencies in vestibular schwannomas and meningioma (ref. 10 and references therein). We hypothesize that the TEAD inhibitors in this report could also be applicable to these other Merlin-null indications. Laraba and Parkinson have shown that the TEAD palmitoylation inhibitors from the series in this report reduce tumor cell proliferation in primary human Merlin-negative meningioma and schwannoma (38). Laraba and colleagues further demonstrated that oral administration of the TEAD inhibitors reduced TEAD and CTGF expression in peripheral nerve tissues and decreased proliferation of Schwann cells and satellite glial cells in dorsal root ganglia in the Periostin-CRE-NF2fl/fl mouse model of vestibular schwannomas. Loss of NF2 has also been found at significant frequencies in renal cell carcinoma (39, 40) and cervical squamous cell carcinoma (8). Further investigation will determine the effect of the TEAD inhibitors in these NF2-deficient indications.

Because TEAD transcription factors are the main drivers for YAP/TAZ recruitment to chromatin, we propose that the TEAD inhibitors would also have applications in human malignancies characterized by abnormally elevated levels of nuclear YAP/TAZ proteins. Uncontrolled YAP/TAZ activation has been reported in many human cancers (ref. 9 and references therein). Recently, increased nuclear YAP expression and function in NF1-driven tumors have also been reported (41, 42). Nuclear YAP accumulation and functional requirement have also been linked to resistance to targeted therapies and cancer relapse in BRAF-mutant, KRAS-mutant, EGFR-mutant, ALK-rearranged non–small cell lung cancer, and RAS-driven neuroblastoma (43–46). Preliminary data have been reported with compounds related to the series in this report suggesting that combinations of TEAD palmitoylation inhibitors and signal transduction inhibitors are a promising avenue of investigation (47, 48).

Other than establishing the tolerability of our compounds in mice, the research described herein does not address any toxicity of the compounds, which could be related or unrelated to inhibition of TEAD palmitoylation. Formal toxicologic evaluation in multiple animal species will be required to characterize the safety of the small molecule compounds. If favorable, clinical evaluation of a TEAD palmitoylation inhibitor is warranted in NF2 mutant mesothelioma and cancers with activated YAP/TAZ-TEAD transcriptional activity as monotherapy or in combination with other targeted cancer therapies.

T.T. Tang, A.W. Konradi, and L. Post report employment with Vivace Therapeutics and have equity interest in Vivace Therapeutics. Y. Feng, former Vivace employee, has equity interest in Vivace Therapeutics. K. Guan reports other support from UCSD during the conduct of the study and has equity interest in Vivace Therapeutics. No disclosures were reported by the other authors.

The crystal structure was deposited in the PDB with ID 7CNL. An MTA is required to obtain the compounds reported in this article.

T.T. Tang: Conceptualization, resources, data curation, formal analysis, supervision, validation, investigation, methodology, writing–original draft, project administration, writing–review and editing, conceived and directed the project. Conceived the biological experiments. Designed studies and performed data collection and analysis. Wrote and edited the paper. A.W. Konradi: Conceptualization, resources, data curation, formal analysis, supervision, investigation, methodology, writing–original draft, project administration, writing–review and editing, conceived and directed the project. Conceived the chemical structures and synthesis. Y. Feng: Data curation, formal analysis, investigation, methodology, designed studies and performed data collection and analysis. X. Peng: Data curation, formal analysis, validation, investigation, visualization, methodology, designed and performed experiments. M. Ma: Investigation, methodology, performed the Co-IP experiments. J. Li: Investigation, methodology, performed the Co-IP experiments. F.-X. Yu: Conceptualization, resources, supervision, investigation, writing–review and editing, conceived the biological experiments. Designed and performed experiments. Edited the paper. K.-L. Guan: Conceptualization, resources, writing–review and editing, conceived the biological experiments. Edited the paper. L. Post: Conceptualization, supervision, writing–review and editing, conceived and directed the project. Conceived the biological experiments. Collection and analysis. Wrote and edited the paper.

We thank Sheng Deng for scientific advice and discussions, Chiho Li, Guoqiang Ma, Lu Zhang, Xiaoling Lan, Honghua Shang, Wenjun Li, Xiaoyan Wu, Zhe Ma, Xiaojing Chi, Yingjie Zhu, Guozhen Zhu, Qunsheng Ji, Zhixiang Zhang, Chen Chen, and Shimei Cao at WuXi AppTec; Zhonghua Yan, Chang Liu, and Qiang Wu at HD Biosciences; Xin Tang at Crown Biosciences; Yingjia Zhang, Linfeng Xu, Yuanyuan Ma, and Ying Chen at ChemPartner; and Yu Xia at Viva Biotech for their services in in vitro assays, in vivo studies, cell line screening, protein purification, protein crystallization, and structure determination. This work was supported by Vivace Therapeutics' internal funding.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Udan
RS
,
Kango-Singh
M
,
Nolo
R
,
Tao
C
,
Halder
G
. 
Hippo promotes proliferation arrest and apoptosis in the Salvador/Warts pathway
.
Nat Cell Biol
2003
;
5
:
914
20
.
2.
Wu
S
,
Huang
J
,
Dong
J
,
Pan
D
. 
Hippo encodes a Ste-20 family protein kinase that restricts cell proliferation and promotes apoptosis in conjunction with salvador and warts
.
Cell
2003
;
114
:
445
56
.
3.
Dong
J
,
Feldmann
G
,
Huang
J
,
Wu
S
,
Zhang
N
,
Comerford
SA
, et al
Elucidation of a universal size-control mechanism in Drosophila and mammals
.
Cell
2007
;
130
:
1120
33
.
4.
Harvey
KF
,
Zhang
X
,
Thomas
DM
. 
The Hippo pathway and human cancer
.
Nat Rev Cancer
2013
;
13
:
246
57
.
5.
Ma
S
,
Meng
Z
,
Chen
R
,
Guan
KL
. 
The Hippo pathway: biology and pathophysiology
.
Annu Rev Biochem
2019
;
88
:
577
604
.
6.
Huh
HD
,
Kim
DH
,
Jeong
HS
,
Park
HW
. 
Regulation of TEAD transcription factors in cancer biology
.
Cells
2019
;
8
:
600
.
7.
Yin
F
,
Yu
J
,
Zheng
Y
,
Chen
Q
,
Zhang
N
,
Pan
D
. 
Spatial organization of Hippo signaling at the plasma membrane mediated by the tumor suppressor Merlin/NF2
.
Cell
2013
;
154
:
1342
55
.
8.
Wang
Y
,
Xu
X
,
Maglic
D
,
Dill
MT
,
Mojumdar
K
,
Ng
PK
, et al
Comprehensive molecular characterization of the Hippo signaling pathway in cancer
.
Cell Rep
2018
;
25
:
1304
17
.
9.
Zheng
Y
,
Pan
D
. 
The Hippo signaling pathway in development and disease
.
Dev Cell
2019
;
50
:
264
82
.
10.
Petrilli
AM
,
Fernández-Valle
C
. 
Role of Merlin/NF2 inactivation in tumor biology
.
Oncogene
2016
;
35
:
537
48
.
11.
Cerami
E
,
Gao
J
,
Dogrusoz
U
,
Gross
BE
,
Sumer
SO
,
Aksoy
BA
, et al
The cBio cancer genomics portal: an open platform for exploring multidimensional cancer genomics data
.
Cancer Discov
2012
;
2
:
401
4
.
12.
Gao
J
,
Aksoy
BA
,
Dogrusoz
U
,
Dresdner
G
,
Gross
B
,
Sumer
SO
, et al
Integrative analysis of complex cancer genomics and clinical profiles using the cBioPortal
.
Sci Signal
2013
;
6
:
pl1
.
13.
Choy
W
,
Kim
W
,
Nagasawa
D
,
Stramotas
S
,
Yew
A
,
Gopen
Q
, et al
The molecular genetics and tumor pathogenesis of meningiomas and the future directions of meningioma treatments
.
Neurosurg Focus
2011
;
30
:
E6
.
14.
Christoph
DC
,
Eberhardt
WE
. 
Systemic treatment of malignant pleural mesothelioma: new agents in clinical trials raise hope of relevant improvements
.
Curr Opin Oncol
2014
;
26
:
171
81
.
15.
Remon
J
,
Reguart
N
,
Corral
J
,
Lianes
P
. 
Malignant pleural mesothelioma: new hope in the horizon with novel therapeutic strategies
.
Cancer Treat Rev
2015
;
41
:
27
34
.
16.
Carbone
M
,
Ly
BH
,
Dodson
RF
,
Pagano
I
,
Morris
PT
,
Dogan
UA
, et al
Malignant mesothelioma: facts, myths, and hypotheses
.
J Cell Physiol
2012
;
227
:
44
58
.
17.
Damhuis
RA
,
Schroten
C
,
Burgers
JA
. 
Population-based survival for malignant mesothelioma after introduction of novel chemotherapy
.
Eur Respir J
2012
;
40
:
185
9
.
18.
Yu
FX
,
Zhao
B
,
Panupinthu
N
,
Jewell
JL
,
Lian
I
,
Wang
LH
, et al
Regulation of the Hippo-YAP pathway by G-protein-coupled receptor signaling
.
Cell
2012
;
150
:
780
91
.
19.
Kim
NG
,
Gumbiner
BM
. 
Adhesion to fibronectin regulates Hippo signaling via the FAK-Src-PI3K pathway
.
J Cell Biol
2015
;
210
:
503
15
.
20.
Woodard
GA
,
Yang
YL
,
You
L
,
Jablons
DM
. 
Drug development against the hippo pathway in mesothelioma
.
Transl Lung Cancer Res
2017
;
6
:
335
42
.
21.
Dhanaraman
T
,
Singh
S
,
Killoran
RC
,
Singh
A
,
Xu
X
,
Shifman
JM
, et al
RASSF effectors couple diverse RAS subfamily GTPases to the Hippo pathway
.
Sci Signal
2020
;
13
:
eabb4778
.
22.
Chan
P
,
Han
X
,
Zheng
B
,
DeRan
M
,
Yu
J
,
Jarugumilli
GK
, et al
Autopalmitoylation of TEAD proteins regulates transcriptional output of the Hippo pathway
.
Nat Chem Biol
2016
;
12
:
282
9
.
23.
Noland
CL
,
Gierke
S
,
Schnier
PD
,
Murray
J
,
Sandoval
WN
,
Sagolla
M
, et al
Palmitoylation of TEAD transcription factors is required for their stability and function in Hippo pathway signaling
.
Structure
2016
;
24
:
179
86
.
24.
Pobbati
AV
,
Han
X
,
Hung
AW
,
Weiguang
S
,
Huda
N
,
Chen
GY
, et al
Targeting the central pocket in human transcription factor TEAD as a potential cancer therapeutic strategy
.
Structure
2015
;
23
:
2076
86
.
25.
Lu
W
,
Wang
J
,
Li
Y
,
Tao
H
,
Xiong
H
,
Lian
F
, et al
Discovery and biological evaluation of vinylsulfonamide derivatives as highly potent, covalent TEAD autopalmitoylation inhibitors
.
Eur J Med Chem
2019
;
184
:
111767
.
26.
Bum-Erdene
K
,
Zhou
D
,
Gonzalez-Gutierrez
G
,
Ghozayel
MK
,
Si
Y
,
Xu
D
, et al
Small-molecule covalent modification of conserved cysteine leads to allosteric inhibition of the TEAD-YAP protein-protein interaction
.
Cell Chem Biol
2019
;
26
:
378
89
.
27.
Li
Q
,
Sun
Y
,
Jarugumilli
GK
,
Liu
S
,
Dang
K
,
Cotton
JL
, et al
Lats1/2 sustain intestinal stem cells and Wnt activation through TEAD-dependent and independent transcription
.
Cell Stem Cell
2020
;
26
:
675
92
.
28.
Holden
JK
,
Crawford
JJ
,
Noland
CL
,
Schmidt
S
,
Zbieg
JR
,
Lacap
JA
, et al
Small molecule dysregulation of TEAD lipidation induces a dominant-negative inhibition of Hippo pathway signaling
.
Cell Rep
2020
;
31
:
107809
.
29.
Zheng
B
,
DeRan
M
,
Li
X
,
Liao
X
,
Fukata
M
,
Wu
X
. 
2-Bromopalmitate analogues as activity-based probes to explore palmitoyl acyltransferases
.
J Am Chem Soc
2013
;
135
:
7082
5
.
30.
Kanadome
T
,
Yokoi
N
,
Fukata
Y
,
Fukata
M
. 
Systematic screening of depalmitoylating enzymes and evaluation of their activities by the Acyl-PEGyl exchange gel-shift (APEGS) assay
.
Methods Mol Biol
2019
;
2009
:
83
98
.
31.
Galli
GG
,
Carrara
M
,
Yuan
WC
,
Valdes-Quezada
C
,
Gurung
B
,
Pepe-Mooney
B
, et al
YAP drives growth by controlling transcriptional pause release from dynamic enhancers
.
Mol Cell
2015
;
60
:
328
37
.
32.
Sekido
Y
,
Pass
HI
,
Bader
S
,
Mew
DJ
,
Christman
MF
,
Gazdar
AF
, et al
Neurofibromatosis type 2 (NF2) gene is somatically mutated in mesothelioma but not in lung cancer
.
Cancer Res
1995
;
55
:
1227
31
.
33.
Li
P
,
Silvis
MR
,
Honaker
Y
,
Lien
WH
,
Arron
ST
,
Vasioukhin
V
. 
αE-catenin inhibits a Src-YAP1 oncogenic module that couples tyrosine kinases and the effector of Hippo signaling pathway
.
Genes Dev
2016
;
30
:
798
811
.
34.
Sato
T
,
Sekido
Y
. 
NF2/Merlin inactivation and potential therapeutic targets in mesothelioma
.
Int J Mol Sci
2018
;
19
:
988
.
35.
Mizuno
T
,
Murakami
H
,
Fujii
M
,
Ishiguro
F
,
Tanaka
I
,
Kondo
Y
, et al
YAP induces malignant mesothelioma cell proliferation by upregulating transcription of cell cycle-promoting genes
.
Oncogene
2012
;
31
:
5117
22
.
36.
Miyanaga
A
,
Masuda
M
,
Tsuta
K
,
Kawasaki
K
,
Nakamura
Y
,
Sakuma
T
, et al
Hippo pathway gene mutations in malignant mesothelioma: revealed by RNA and targeted exon sequencing
.
J Thorac Oncol
2015
;
10
:
844
51
.
37.
Liu-Chittenden
Y
,
Huang
B
,
Shim
JS
,
Chen
Q
,
Lee
SJ
,
Anders
RA
, et al
Genetic and pharmacological disruption of the TEAD-YAP complex suppresses the oncogenic activity of YAP
.
Genes Dev
2012
;
26
:
1300
5
.
38.
Laraba
L
,
Ercolano
E
,
Tang
TT
,
Post
L
,
Parkinson
DB
. 
e-poster presented at American Association for Cancer Research Virtual Annual Meeting II, 22 to 24 June 2020
.
Abstract nr 6384
.
39.
Sourbier
C
,
Liao
PJ
,
Ricketts
CJ
,
Wei
D
,
Yang
Y
,
Baranes
SM
, et al
Targeting loss of the Hippo signaling pathway in
.
Oncotarget
2018
;
9
:
10723
33
.
40.
Pal
SK
,
Ali
SM
,
Yakirevich
E
,
Geynisman
DM
,
Karam
JA
,
Elvin
JA
, et al
Characterization of clinical cases of advanced papillary renal cell carcinoma via comprehensive genomic profiling
.
Eur Urol
2018
;
73
:
71
8
.
41.
Wu
LMN
,
Deng
Y
,
Wang
J
,
Zhao
C
,
Rao
R
,
Xu
L
, et al
Programming of schwann cells by Lats1/2-TAZ/YAP signaling drives malignant peripheral nerve sheath tumorigenesis
.
Cancer Cell
2018
;
33
:
292
308
.
42.
Chen
Z
,
Mo
J
,
Brosseau
JP
,
Shipman
T
,
Wang
Y
,
Liao
CP
, et al
Spatiotemporal loss of NF1 in Schwann cell lineage leads to different types of cutaneous neurofibroma susceptible to modification by the Hippo pathway
.
Cancer Discov
2019
;
9
:
114
29
.
43.
Coggins
GE
,
Farrel
A
,
Rathi
KS
,
Hayes
CM
,
Scolaro
L
,
Rokita
JL
, et al
YAP1 mediates resistance to MEK1/2 inhibition in neuroblastomas with hyperactivated RAS signaling
.
Cancer Res
2019
;
79
:
6204
14
.
44.
Lin
L
,
Sabnis
AJ
,
Chan
E
,
Olivas
V
,
Cade
L
,
Pazarentzos
E
, et al
The Hippo effector YAP promotes resistance to RAF- and MEK-targeted cancer therapies
.
Nat Genet
2015
;
47
:
250
6
.
45.
Tsuji
T
,
Ozasa
H
,
Aoki
W
,
Aburaya
S
,
Yamamoto Funazo
T
,
Furugaki
K
, et al
YAP1 mediates survival of ALK-rearranged lung cancer cells treated with alectinib via pro-apoptotic protein regulation
.
Nat Commun
2020
;
11
:
74
.
46.
Nilsson
MB
,
Sun
H
,
Robichaux
J
,
Pfeifer
M
,
McDermott
U
,
Travers
J
, et al
A YAP/FOXM1 axis mediates EMT-associated EGFR inhibitor resistance and increased expression of spindle assembly checkpoint components
.
Sci Transl Med
2020
;
12
:
eaaz4589
.
47.
Coggins
GE
,
Farrel
A
,
Rathi
K
,
Hayes
CM
,
Tang
TT
,
Post
L
, et al
Poster presented at American Association for Cancer Research Annual Meeting 2019, Atlanta, GA, 29 March to 3 April 2019, in Proceedings of the American Association for Cancer Research Annual Meeting 2019
,
Cancer Res
2019
;
79
(13 Suppl):Abstract nr 2886
.
48.
Yang
Y
,
White
SM
,
Tang
TT
,
Post
L
,
Yi
C
,
Field
J
. 
Poster presented at the 2019 Global NF Conference, San Francisco, CA, 21 to 24 September 2019, sponsored by Children's Tumor Foundation
.