Advertisement
Cancer Genes and Networks

Somatic Hypermutation of the YAP Oncogene in a Human Cutaneous Melanoma

Xiaomeng Zhang, Jian Zhong Tang, Ismael A. Vergara, Youfang Zhang, Pacman Szeto, Lie Yang, Christopher Mintoff, Andrew Colebatch, Lachlan McIntosh, Katrina A. Mitchell, Evangeline Shaw, Helen Rizos, Georgina V. Long, Nicholas Hayward, Grant A. McArthur, Anthony T. Papenfuss, Kieran F. Harvey and Mark Shackleton
DOI: 10.1158/1541-7786.MCR-18-0407 Published July 2019

Abstract

Melanoma is usually driven by mutations in BRAF or NRAS, which trigger hyperactivation of MAPK signaling. However, MAPK-targeted therapies are not sustainably effective in most patients. Accordingly, characterizing mechanisms that co-operatively drive melanoma progression is key to improving patient outcomes. One possible mechanism is the Hippo signaling pathway, which regulates cancer progression via its central oncoproteins YAP and TAZ, although is thought to be only rarely affected by direct mutation. As YAP hyperactivation occurs in uveal melanoma, we investigated this oncogene in cutaneous melanoma. YAP protein expression was elevated in most benign nevi and primary cutaneous melanomas but present at only very low levels in normal melanocytes. In patient-derived xenografts and melanoma cell lines, we observed variable reliance of cell viability on Hippo pathway signaling that was independent of TAZ activity and also of classical melanoma driver mutations such as BRAF and NRAS. Finally, in genotyping studies of melanoma, we observed the first ever hyperactivating YAP mutations in a human cancer, manifest as seven distinct missense point mutations that caused serine to alanine transpositions. Strikingly, these mutate four serine residues known to be targeted by the Hippo pathway and we show that they lead to hyperactivation of YAP.

Implications: Our studies highlight the YAP oncoprotein as a potential therapeutic target in select subgroups of melanoma patients, although successful treatment with anti-YAP therapies will depend on identification of biomarkers additional to YAP protein expression.

Introduction

Recent spectacular advances in treatment of melanoma, a common and deadly form of skin cancer, have leveraged discoveries of mechanisms of disease initiation and progression. A paradigm-changing example was the observation that most oncogenic melanoma-initiating mutations, classically in BRAF or NRAS, increase MAPK signaling (1). This led to BRAF- and MAPK kinase (MEK)-targeted therapies that dramatically changed treatment (2, 3). Unfortunately, although most BRAF-mutant melanomas respond to combination BRAF/MEK inhibition, resistance usually develops (4). For patients with non-BRAF-mutant disease, BRAF targeting is usually ineffective, although some respond to MEK targeting (5). Even T-cell checkpoint targeting with immunotherapy, increasingly and successfully applied to a wide variety of cancers including melanoma (6, 7), is not efficacious or sustained in many patients, highlighting the importance of identifying novel drivers of disease progression and therapy resistance. For instance, the Hippo signaling pathway was found to mediate resistance to MAPK pathway targeting in cancers such as melanoma (8–10), spurring efforts to develop Hippo-targeted therapies (11) that might overcome such resistance.

The Hippo pathway, which is a critical regulator of organ size, was discovered in Drosophila screens for regulators of tissue growth (12). Subsequently shown to regulate growth in mice (13, 14), Hippo pathway proteins regulate organ size by modulating nuclear access of the transcription co-activators YAP and TAZ (also known as WWTR1). Hippo signaling is regulated by properties such as cell polarity and adhesion and has been linked to G-protein–coupled receptor signaling. These signals are typically conveyed by a core kinase cassette, which is composed of Sterile 20-like kinases (e.g., MST1 and MST2), NDR kinases (e.g., LATS1 and LATS2) and two sets of adaptor proteins (Salvador and Mob family proteins). LATS1/2 repress YAP and TAZ by phosphorylation of serine residues (at least five in YAP), which promotes binding to 14-3-3 proteins and cytoplasmic sequestration. YAP and TAZ regulate gene expression with transcription factors, particularly TEAD family proteins (15).

Hippo proteins regulate many hallmarks of cancer and pathway activity is deregulated in many cancers (16). Despite this, mutations directly linked to alterations in Hippo pathway activity appear surprisingly uncommon in human cancer. Exceptions are the upstream pathway gene neurofibromin 2 (NF2), which is frequently mutated in mesothelioma (17) and meningioma (18), and the G protein genes GNAQ and GNA11, which collectively are mutated in 85% of uveal melanomas and hyperactivate YAP (19–21). Putative gain-of-function mutations of YAP are apparently very rare, with only one reported, a YAP1–TFE3 fusion in epithelioid hemangioendothelioma (22). In melanoma, Menzel and colleagues found copy number (CN) gains directly affecting YAP in 4% to 10% of patients and, overall, 62% of melanomas had CN alterations (CNA) affecting known Hippo pathway genes (23).

Unlike uveal melanoma, the role of Hippo signaling in the far more prevalent cutaneous melanoma is poorly defined, particularly in therapy-naïve contexts. Expression of YAP in cutaneous melanoma was described, with one study (23) but not another (24) reporting elevated expression with increasing disease stage. Nallet-Staub and colleagues noted that TAZ expression followed YAP expression, and that benign nevi displayed YAP/TAZ levels similar to melanomas (although this was not reproduced; refs. 23, 24). In functional studies and consistent with other data (16), YAP, TAZ, or TEAD were found to promote malignant behaviors in melanoma cell lines (23, 24). Interestingly, whereas YAP and TAZ inhibited invasion in 1205Lu and SKMEL28 cells (24), heterogeneity in clonogenic growth amongst lines was noted upon TAZ knockdown. This raises the possibility that these oncoproteins might regulate malignant behaviors in only some cancers. Further, it is unknown whether YAP and TAZ, which are paralogous proteins, can compensate for each other's inhibition. Both these possibilities have implications for development of inhibitors of YAP- or TAZ-mediated transcription.

To address these issues, we used melanoma patient samples, patient-derived xenografts (PDX), and cell lines to investigate systematically the Hippo pathway in melanoma. We found Hippo pathway deregulation across a broad spectrum of melanocytic neoplasia, from benign nevi to melanomas. In molecular studies, we identified a patient melanoma with multiple YAP mutations that led to its hyperactivation. In PDX and cell lines, we found that melanoma cells rely on YAP for survival only variably, that this is independent of mutations in BRAF and NRAS, and that TAZ does not consistently compensate for YAP depletion in maintaining melanoma cell viability. We thus report the first obvious gain-of-function missense mutations in the YAP oncogene in human cancer and implicate the Hippo pathway as a potential therapeutic target in a subset of melanomas.

Materials and Methods

Human tissues and cell lines

Human tissues were obtained from the Melbourne Melanoma Project, the Victorian Cancer Biobank, TissuPath, and the CASCADE Rapid Autopsy programme under Peter MacCallum Cancer Centre Human Research Ethics Committee protocol #10/02. Melanoma cell lines and RPE cells were cultured with RPMI1640+HEPES 20 mmol/L containing 10% FBS and 1% pen/strep at 37°C in humidified 5% CO2. The NIH-3T3 cell line was cultured with DMEM (Invitrogen) containing 10% FBS and 1% pen/strep at 37°C in humidified 5% CO2.

Immunofluorescence microscopy and analysis

Following dewaxing, formalin-fixed and paraffin-embedded (FFPE) sections underwent antigen retrieval in pH 6 buffer by heating at 120°C for 3 minutes in a pressure cooker. After treatment with H2O2, sections were blocked in TNB blocking buffer provided by tyramide signal amplification (TSA) indirect kit (Perkin Elmer). Following overnight incubation at 4°C with anti-YAP antibody (Santa Cruz; clone H-125), sections were washed by 0.1% Tween20 in TBS and then sequentially probed with anti-rabbit HRP and biotin-conjugated tyramide. Tyramide amplified-signals were visualized by streptavidin-Alexa Fluor 488. Co-staining of melan-A or S100 was performed to label melanocytic cells. 4′,6-Diamidino-2-phenylindole was used to identify nuclei. Sections were evaluated and imaged on an Olympus BX51 microscope. YAP signal intensity was quantified in the cytoplasm and nucleus of melan-A/S100 positive cells by MetaMorph Microscopy Automation & Image Analysis Software (Molecular Devices). To internally control for differences in YAP signal that were due to differences in background, YAP signals in each melanocytic cell were normalized to the average YAP signals in normal basal keratinocytes located in the same section but distantly from the area of melanocytic cells being evaluated.

Western blotting

Total cell lysates were prepared in RIPA buffer for cultured cells, or in Gordan buffer for tumor pieces, with protease and phosphatase inhibitor cocktails (Roche). Lysates were fractionated using SDS-PAGE and proteins transferred onto polyvinylidene difluoride (PVDF) membranes (Millipore). Membranes were probed with antibodies against YAP (Cell Signaling Technology; 4912), phosphor-YAP/Ser-127 (Cell Signaling Technology; 4911), TEAD1 (BD Biosciences; 610922), TAZ (Cell Signaling Technology; 4883), Actin (Cell Signaling Technology; 4967), or GAPDH (Abcam; ab9484), followed by secondary detection and chemiluminescent visualization.

DNA extraction and whole exome sequencing

FFPE tissue was sectioned and stained with hematoxylin and eosin and labeled with melan-A and S100 antibodies to identify regions of melanoma. Tissue was manually dissected from corresponding regions in adjacent sections. Genomic DNA from FFPE sections (or buffy coats as matching normal tissue) was extracted using standard kits (Qiagen). DNA was quantified using a Qubit fluorometer (Invitrogen) and a Quant-iT dsDNA HS Kit (Invitrogen). Input DNA was fragmented using a Covaris S2 focal acoustic device (Covaris) to an average fragment size range of 180–220 bp. Libraries were prepared using the KAPA Hyper Prep Kit (Kapa Biosystems) with SureSelect XT adaptors and primers (Agilent Technologies). Hybridization capture was performed with SureSelect Human All Exon V5 (Agilent Technologies). Three indexed libraries were run per lane on an Illumina HiSeq2500 platform (paired-end 100 bp) according to standard protocols.

Variant calling and CN estimation

Sequenced reads were processed with cutadapt v1.7.1 (25) and aligned against the human genome (v. hg19) using bwa-mem v0.7.12 (26). Duplicate reads were marked using picard v1.128 (27) and base recalibration and local indel realignment performed with GATK v 3.1.1 (28). Reads with a mapping quality <30 were removed by samtools (29). Somatic single nucleotide variants (SNV) were detected from the intersection of at least two variant callers of GATK Unified Genotyper, VarScan v2.3 (–strand-filter 1 –min-var-freq 0.1; ref. 30) and muTect v2.7 (31). Variants were filtered using the following criteria: (i) total coverage below 10X in tumor or germline; (ii) overlap with annotated repetitive sequences by repeatmasker (32); (iii) overlap with low complexity sequences (33); (iv) present on the Exome Variant Server vESP6500 [Exome Variant Server, NHLBI GO Exome Sequencing Project (ESP), Seattle, WA; URL: http://evs.gs.washington.edu/EVS/, accessed September, 2013] or in the 1,000 genome database (34); (v) overlap with artifact-generating regions corresponding to gene CDC27 (17:45195069-45266788 in hg19; refs. 35, 36); and (vi) overlap with genomic regions of mucins, ryanodine receptors, olfactory receptors and other genes that cause common and likely nonsignificant mutations due to gene length and late replication time (37). Mutant allele fractions that incorporate CN and cellularity estimates were calculated as described (38). Impacts of mutations within protein-coding regions were assessed using Variant Effect Predictor v73 (39). Nonsynonymous mutations on YAP1 refer to transcript ENST00000282441. Estimates of allele-specific CN and tumor cellularity were obtained using Sequenza (40) with a median normalization method and with gender setting. To improve segmentation of focal regions, the gamma parameter was set to 20.

Inference of genome doubling

A test for genome doubling was developed by comparing two branching processes: (i) stochastic acquisition of aneuploidy and a genome doubling event, and (ii) stochastic acquisition of aneuploidy only. To account for model complexity, model selection was performed by choosing the model that minimized the Akaike Information Criterion (AIC). AIC = 2k − 2log(L), where L is the likelihood and k the number of parameters in each model (k = 2 for aneuploidy only and k = 3 for aneuploidy and genome doubling). The branching process with the lowest AIC value was the one of higher likelihood.

Analysis of TCGA data

TCGA mutation profiles for genes in the Hippo pathway were retrieved using the query function within cbioportal across the melanoma (SKCM) cohort (41, 42). Spearman correlation P values were computed using the AS 89 algorithm as implemented in the cor.test method in R. For CN versus mRNA correlations, the alternative hypothesis was one of positive association; whereas for methylation versus mRNA correlations, the alternative hypothesis was one of negative association.

Cloning

GFP-luciferase (GFP-LUC) labeled dox-inducible shRNA was created on the backbone of a GFP-tagged doxycycline (dox)-inducible shRNA vector (43) by replacing the GFP-coding sequence with a GFP–LUC fusion coding sequence (44) using BlpI and KpnI restriction sites. Prior to ligation, sense and antisense oligonucleotides that code for scrambled (SCR) or YAP-targeting shRNAs (shYAP-3-F: tcccCGGCCATGCTGTCCCAGATGAAtttcaagagaTTCATCTGGGACAGCATGGCCTtttttt, shYAP-3-R: tcgagaaaaaaCGGCCATGCTGTCCCAGATGAAtctcttgaaaTTCATCTGGGACAGCATGGCCT, shYAP-4-F: tcccCGGAGATGGAATGAACATAGAAtttcaagagaTTCTATGTTCATTCCATCTCCTttttttc, shYAP-4-R: tcgagaaaaaaCGGAGATGGAATGAACATAGAAtctcttgaaaTTCTATGTTCATTCCATCTCCT, SCR-F: tcccAGTACTGCTTACGATACGGtttcaagagaCCGTATCGTAAGCAGTACTttttttc, SCR-R: tcgagaaaaaaAGTACTGCTTACGATACGGtctcttgaaaCCGTATCGTAAGCAGTACT) were annealed at 96°C before insertion between PacI restriction sites.

YAP2L, YAP2L2SA, YAP2L5SA, and YAP2L7SA cDNAs, each with an N-terminal Myc tag, were synthesized by Biomatik. These cDNAs were subcloned into the retroviral vector pMSCV-pBabeMCS-IRES-GFP, which was a gift from Dr. Martine Roussel & Charles Sherr (Addgene plasmid #33336).

Viral transduction

To make lentivirus for delivery of shRNA-GFP-LUC, HEK293T packaging cells were plated in DMEM-based growth media to reach 80% confluence overnight (∼114,000 cells/cm2). On the next day, 1 mL of 0.94 mmol/L PEI and 130 mmol/L NaCl was added to 1 mL of 150 mmol/L NaCl containing packaging plasmid pCD/NL-BH*ΔΔΔ 3.4 μg, envelop plasmid pLTR-G 1.7 μg and shRNA vector 5.1 μg) while vortexing. Following 10-minute incubation at room temperature, the plasmid-PEI complex was added to 293T cells in UltraCULTURE media (Lonza) containing 1% l-glutamine. Virus-containing supernantants were collected 48 and 72 hours posttransfection. For stable cell transduction, equal titers of lentivirus expressing scrambled or YAP-targeting shRNA were added to target cells in six-well plates with 1 μg/mL protamine sulphate and centrifuged at 500 × g and 25°C for 2 hours. Transduced cells were washed and collected without intervening in vitro culture for immediate subcutaneous injection into NOD/SCID Il2rγ null (NSG) mice.

PDX assays

All mouse experiments were performed according to protocols approved by the Peter MacCallum Cancer Centre Animal Experimentation Ethics Committee (approvals E421 and E526). PDX melanomas (PDX-1 from patient 332, PDX-2 from patient 233, and PDX-3 from patient 223) had been established (45) from patient melanomas previously genotyped for BRAF mutations. For YAP-depletion in PDX melanomas, cells isolated freshly from PDX melanomas were transduced with dox-inducible shRNA-GFP-LUC and transplanted into NSG mice as above. GFP+ cells from these tumors were isolated by flow cytometry and re-injected into new NSG mice. When tumors reached ∼10 mm in diameter, dox (40 μg in 100 μL/mouse) (or vehicle) was injected intraperitoneally three times in the first week. Dox was also added to drinking water (2 mg/mL) until experimental endpoint. Tumors were harvested for flow cytometric and molecular analyses as their diameters approximated 20 mm. Melanoma cells in shYAP3-GFP-LUC- and shSCR-GFP-LUC-transduced tumors were typically heterogeneous for GFP expression despite being established from purified GFP+ cells, consistent with the presence of contaminating nontransduced GFP− cells in the inoculate, or subclonal gene silencing during resultant tumor growth.

QPCR and soft agar assays

NIH-3T3, RPE, and MeWo cell lines that stably expressed YAP2L or YAP2L variants were generated as described previously (46). QPCR and soft agar assays were performed as described previously (46, 47). Primers sequences for CYR61 and CTGF were as described previously (48).

MTT cell viability assay

shYAP3-GFP-LUC- and shSCR-GFP-LUC-transduced melanoma cells from PDX tumors were seeded in 96-well plates, such that control cells became 80% confluent by day 5. Dox was added to cells at seeding at required concentrations. Following addition to each well of 0.2 mg/mL MTT, the colorimetric reaction was allowed to develop for 2 hours prior to overnight incubation in 5% SDS + 0.05 mmol/L HCl. Absorbance was measured at 550 nm on a compatible plate reader.

Alamar Blue assay

Cells were seeded in 96-well plates and transiently transfected with siGENOME SMARTpool siRNAs (GE Dharmacon; siYAP: GGUCAGAGAUACUUCUUAA, CCACCAAGCUAGAUAAAGA, GAACAUAGAAGGAGA GGAG, GCACCUAUCACUCUCGAGA; siTAZ: GACAUGAGAUCCAUCACUA, GGACAAACACCCAUGAACA, AGGAACAAACGUUGACUUA, AAGCCUAGCUCGUGGCGGA; siTEAD1: CACAAGACGUCAAGCCUUU, GAAAGGUGGCUUAAAGGAA, CGAUUUGUAUACCGAAUAA, CCCAAUGUGUGAAUAUAUG; siTEAD2: CGAAGGAAAUCAAGGGAAA, GCAGUUGAUUCUUACCAGA, GGAAUGAACUGAUCGCCCG, GGAAGACCCGAACUCGAAA), OTP control siRNAs (ON-TARGETplus Non-Targeting Pool siRNA) or as mock using Lipofectamine RNAiMAX transfection reagent (Invitrogen) without siRNA. Media was changed after 24 hours and cells incubated for a further 72 hours. For drug treatment, cells were seeded in 96-well plates for 24 hours and verteporfin (Novartis) was added and incubated with cells for 72 hours. Before measurement, Alamar Blue was added to each well and incubated at 37°C for 2 hours. Fluorescence was read using a POLARstar OPTIMA (BMG Labtech) at 540/610 nm.

Statistical analysis

Statistical analyses were performed with GraphPad Prism (GraphPad Software).

Results

YAP expression in human melanocytic neoplasia

To resolve outstanding questions regarding YAP expression in melanoma, we performed immunofluorescence labeling on human tissue samples [benign and dysplastic nevi (n = 18), primary cutaneous melanomas (n = 13)] from a pathology archive. Normal skin (n = 4) was used to evaluate YAP expression in normal melanocytes in parallel. To identify YAP expression in melanocytic cells, tissue sections were co-labeled with anti-melan A antibodies. As marked variation in YAP signal in adjacent normal skin was noted amongst tissue sections from different patients [Nallet-Staub and colleagues (24) and not shown], we used isotype controls for each section and image analysis software to normalize melanocytic YAP expression relative to YAP expression in adjacent normal keratinocytes (49).

Compared with controls, YAP was detectable at low levels in a subset of melanocytes (Fig. 1A) and increased in most nevi and melanomas (Fig. 1A). Although heterogeneity in YAP expression was apparent in most lesions, the majority of nevi and melanomas had abundant cells with YAP expression levels that were greater (P < 0.01, z-score) than average levels in normal melanocytes (Fig. 1B). In parallel studies (Supplementary Fig. S1) of melanoma cell lines representing mutant BRAF, mutant NRAS and BRAF/NRAS double-wild-type genotypes, immunoblotting of YAP and TAZ showed ubiquitous expression of YAP, with phospho-S127-YAP (phosphorylation mediated by the Hippo pathway) levels proportional to total YAP. TAZ was also expressed in most cell lines, with no evidence of reciprocity of expression with YAP. YAP expression was also detected by western blotting in 18/19 PDX melanomas (not shown).

Figure 1.
Figure 1.

YAP protein expression and subcellular distribution in melanocytic neoplasia. A, Co-immunofluorescence evaluation of melan-A (red; for identifying melanocytic cells) and YAP (green) expression in FFPE sections of normal human skin (top row), nevi (middle), and melanomas (bottom). In normal skin panels, top left insets show negative labeling with YAP isotype antibody in melan-A+ cells in a melanoma and bottom right insets show magnified regions of normal basal epidermis, highlighting melan-A+ melanocytes with no (white arrows) or low (yellow arrows) YAP expression. DAPI (blue) = nuclei. B, Left: Quantitation by Metamorph of YAP expression in individual melan-A+ cells in four normal human skin samples and melanocytic lesions (9 benign nevi, 9 dysplastic nevi, and 13 melanomas). Expression values in each cell (shown as dots) were normalized to the average YAP signal in adjacent but separate normal basal keratinocytes. Red line indicates the level of normalized YAP expression above which expression values were greater (P < 0.01, two-tailed z-score) than the mean expression level across melanocytes in normal skin. Right: The proportions of samples for each tissue (N, normal skin; BN, benign nevus; DN, dysplatic nevus; M, melanoma) in which <10% or >10% cells displayed increased YAP expression. C, Left: Quantitation by Metamorph of YAP protein distribution in individual melan-A+ cells in normal human skin, nevi, and melanomas of different stage (1–4). Values for each imaged cell (shown as dots) indicate the nuclear:cytoplasmic (N:C) ratio of YAP signal in that cell. Red line indicates the ratio above which values were greater (P < 0.01, two-tailed z-score) than the mean ratio across melanocytes in normal skin. Right: The proportions of samples for each tissue (N, normal skin; BN, benign nevus; DN, dysplatic nevus; M, melanoma) in which <10% or >10% cells displayed increased (P < 0.01, two-tailed z-score) YAP N:C ratios compared with normal melanocytes.

Consistent with published data (24), YAP was distributed in patient melanomas across the nucleus and cytoplasm within melan A+ cells. As an increased YAP nucleus:cytoplasm (N:C) expression ratio might indicate altered Hippo pathway activity, we used image analysis to calculate YAP N:C ratios in melan A+ cells. For these studies, tumors from metastatic sites were also evaluated. YAP N:C ratios were increased compared with normal melanocytes in substantial proportions of lesions, and more melanocytic cells in nevi and in early-stage cutaneous melanomas had higher ratios compared with cells from metastatic disease (Fig. 1C). These analyses indicate that YAP is upregulated and enriched in nuclei across a spectrum of neoplastic melanocytic lesions.

Somatic hypermutation of YAP in a cutaneous melanoma

Although we (Fig. 1) and others (24) found YAP protein expression elevated in melanocytic neoplasia, the mechanisms of its upregulation and/or activation are unclear. In evaluating this via genotyping studies, we surprisingly detected multiple YAP mutations that caused seven independent serine to alanine substitutions in a primary cutaneous melanoma from a patient (CA-04) whose melanoma underwent whole exome sequencing (Fig. 2A). All but one serine to alanine substitution resulted from SNVs (TCG to GCG for Ser61Ala, TCC to GCC for Ser127Ala, and TCT to GCT for Ser128Ala, Ser131Ala, Ser163Ala, and Ser164Ala). Ser397 was mutated to Ala397 by a dinucleotide substitution (AGT to GCT). Strikingly, the majority of these serines are key regulatory residues phosphorylated by the central Hippo pathway kinases Large Tumor Suppressor 1 and 2 (LATS1 and LATS2), which repress YAP (50, 51). One of them, S61, is also phosphorylated repressively by AMP-activated protein kinase (AMPK; refs. 52, 53; Fig. 2A). Inspection of reads indicated these variants had a common clonal origin. As far as we are aware, these are the first predicted spontaneous somatic activating mutations reported to affect YAP in any human cancer. We termed this hypermutated YAP allele, YAP-7SA.

Figure 2.
Figure 2.

Hypermutation of YAP in a human melanoma. A, Depiction of hypermutation at seven different residues in the YAP1 gene of patient CA-04, identified through whole exome sequencing of a primary cutaneous melanoma. TEAD-binding (TBD), WW and transactivation (TA) domains are functional domains in the YAP protein. Bolded mutations indicate validated LATS1/2 or AMPK binding sites. B, B-allele frequency (top), depth ratio (middle), and CN estimate (bottom) plots of chromosomes (chr) 7 and 11 derived from whole exome sequencing of the primary cutaneous melanoma from patient CA-04. Boxes show regions including mutations in RAC1 and BRAF (on chr 7) and YAP on (chr 11), indicating focal amplification of RAC1 P29S but not of BRAF V600E or the YAP1 hypermutant. LBAF, lesser B-allele frequency. C, B-allele frequency (top), depth ratio (middle), and CN estimate (bottom) plots derived from whole exome sequencing of the primary cutaneous melanoma from patient CA-04.

The same patient, whose melanoma was BRAF V600E mutant, also carried an activating RAC1 P29S mutation (Fig. 2B). CN analysis (Fig. 2B) indicated that this mutation was focally amplified on chromosome 7, resulting in a mutant allele frequency (MAF) of 96%. Furthermore, allele-specific CN calls indicated the RAC1 P29S mutation was homozygous. In contrast, the BRAF and YAP mutations were present at MAF's of 40% and 29% to 47%, respectively (Fig. 2B). Interestingly, 24% of the melanoma genome of patient CA-04 had a predicted CN of 4, with 2 copies per allele (Fig. 2C). A branching process-based analysis suggested that a combination of focal CN acquisition as well as a genome doubling event explained the CNAs observed (AIC score 210.6), rather than CN acquisition only (AIC score 279.1). This would be consistent with links between genome doubling, activation of RAC1 and Hippo pathway tumor suppression (54).

The YAP-7SA allele encodes a hyperactive YAP protein

To determine whether YAP-7SA is indeed a gain-of-function YAP allele, we synthesized cDNAs encoding for wild-type YAP2L, YAP2L-7SA, YAP2L-2SA, and YAP2L-5SA. YAPL-2SA and YAP2L-5SA are known hyperactive YAP alleles in which either the two major regulatory LATS1/2 phosphorylation sites (S127 and S397) are mutated or all five LATS1/2 sites (S127, S397, S61, S127, and S391) are mutated, respectively (55). We examined the different YAP alleles in two assays commonly used to explore YAP activity: quantitative RT-PCR of downstream YAP/TEAD target genes and soft agar assays.

Initially, we generated MeWo melanoma cell lines and nontransformed human retinal pigment epithelium (RPE) cell lines that stably expressed YAP2L, YAP2L-2SA, and YAP2L-7SA. Both YAP2L-2SA and YAP2L-7SA induced higher expression of the well-characterized YAP target genes CYR61 and CTGF than wild-type YAP2L in MeWo cells (Fig. 3A). Furthermore, YAP2L-7SA expression strongly increased anchorage-independent growth of both MeWo and RPE cells, compared with YAP2L-expressing cells (Fig. 3B and C).

Figure 3.
Figure 3.

YAP-7SA encodes a gain of function YAP protein. A, Quantification of expression of the YAP target genes CYR61 and CTGF by quantitative RT-PCR in MeWo cells stably expressing the indicated plasmids. B–D, Soft agar assays with MeWo cells (B), RPE cells (C), and NIH-3T3 cells (D) stably expressing the indicated plasmids. Data are represented as mean ± SEM from three biological replicates. *, P < 0.05; ***, P < 0.001; ****, P < 0.0001 (ANOVA plus Tukey's multiple comparison tests). Western blots were performed to determine YAP expression (actin blots serve as loading controls). Representative images are shown of soft agar cultures for each plasmid (CON, control; YAP, wild-type YAP; 2SA, YAP-2SA mutant; 5SA, YAP-5SA mutant; 7SA, YAP-7SA).

We also performed soft agar assays in murine NIH-3T3 cells, given that they are sensitive to YAP-mediated transformation (46, 55). Consistent with previous studies, expression of YAP2L-5SA increased the ability of NIH-3T3 cells to grow in an anchorage-independent fashion, compared with vector control cells. Consistent with our MeWo and RPE cell experiments, YAP2L-7SA also substantially increased the anchorage-independent growth of NIH-3T3 compared with wild-type YAP2L (Fig. 3D). Therefore, we conclude that YAP2L-7SA encodes a hyperactive version of the YAP protein.

Hippo pathway gene mutations in cutaneous melanoma

We also interrogated The Cancer Genome Atlas (TCGA; ref. 56) for SNVs and CNAs affecting Hippo pathway genes (Fig. 4A). Only one of 368 TCGA patient melanomas carried a YAP mutation, a Ser257Phe substitution of predicted medium-level functional impact. Similarly, only five melanomas had TAZ mutations. Thus, despite our discovery of a YAP hypermutant (Fig. 2A), activating point mutations in the YAP and TAZ oncogenes are rare in melanoma. In contrast, FAT genes 1 to 4, CRB genes 1 to 2, HMNC1, DCHS2, and TAOK2 were frequently mutated in TCGA melanomas, with notable mutational clustering in NF1/TP53 double-mutant tumors. This raises the possibility that YAP activation might occur in some melanomas via mutational inhibition of upstream Hippo pathway tumor suppressors.

Figure 4.
Figure 4.

Mutations and methylation changes in Hippo pathway genes in the TCGA melanoma cohort. A, Somatic coding mutations (top) and CN changes (bottom) affecting Hippo pathway genes identified in patients from the TCGA cohort (56), organized horizontally according to the BRAF-, NRAS-, NF1-, TP53-, RAC1- or PTEN-mutant melanoma subtype of each patient and vertically according the mutation frequency in each Hippo pathway gene. Mutations colored by type: amp, amplification; hemi, hemizygous; homo, homozygous; del, deletion. B, top: distributions of methylation values across 473 TCGA patients in 18 known/predicted oncogenes and 33 known/predicted tumor suppressor genes in the Hippo pathway. Bottom: Spearman coefficient values for correlations between methylation and mRNA expression for each gene. C, Methylation versus mRNA expression plots for YAP and TAZ. Each dot represents values for an individual patient. Blue dots, mutated gene; red dots, wild-type gene. r is the Spearman correlation coefficient; *, P < 0.001.

We also interrogated CNAs. Of the 367 TCGA melanomas in which CNAs were called, a high proportion contained CNAs affecting Hippo pathway genes (Fig. 4A, bottom). The most frequently amplified genes included HIPK2, CRB1, HMCN1, SCRIB, and STK4, whereas frequently deleted genes included GNAQ, CRB2, AMOTL1, and FAT3, with most of these genes having a statistically significant correlation between CN and expression (Supplementary Figs. S2 and S3). LATS1, a classical suppressor of YAP/TAZ, was the most frequently deleted gene (61% of cases) and its deletion was associated with reduced LATS1 mRNA expression (Supplementary Fig. S3). Surprisingly, despite evidence of elevated YAP expression in melanoma (Fig. 1; refs. 23, 24) and of a positive correlation between YAP mRNA and YAP gene CN (Supplementary Fig. S3), YAP was also commonly affected by deletion, undergoing heterozygous CN loss in 44% cases and amplification in only 11%. Increased YAP activity in melanoma therefore may not typically occur as a result of direct mutation or increased gene dose via CNAs.

YAP is hypomethylated in cutaneous melanoma

We also evaluated DNA methylation of Hippo pathway genes from 473 patients in the TCGA data (Fig. 4B and C). For each gene, methylation values were correlated with mRNA to estimate functional relationships between methylation and expression. Across 33 predicted tumor suppressor genes and 18 oncogenes, most evidenced a negative correlation between methylation and mRNA expression (Fig. 4B). A clear negative correlation (P < 0.001) was observed between methylation and mRNA expression of TAZ (Fig. 4C), suggesting that methylation is a mechanism by which this gene is regulated. In contrast, YAP was hypomethylated in every melanoma regardless of YAP mRNA levels (Fig. 4C), which nevertheless varied across the TCGA cohort.

YAP targeting in PDX melanomas

In functional studies, YAP facilitated invasion and clonogenicity in cell lines (24). However, YAP function has never been interrogated in PDX melanomas, which are more clinically relevant models. We therefore evaluated whether YAP targeting inhibits growth of PDX melanomas by introducing a GFP-LUC cassette (44) and validated (57) shYAP (or shSCR control) hairpins into a dox-inducible shRNA construct (Fig. 5A; ref. 43). In transduced shYAP-GFP-LUC A375 cells, dox induced knockdown of YAP (Fig. 5A), particularly from the shYAP3 hairpin that was used subsequently. To transduce PDX melanomas, we performed rapid spinfection (44) with lentivirus-packaged shYAP-GFP-LUC (or shSCR-GFP-LUC control) of cells freshly isolated (58) from two early passage BRAF V600E mutant PDX melanomas (PDX-1 and -2, Fig. 5B–D). Spinfected cells were immediately retransplanted into NOD/SCID IL2rγnull (NSG) mice and transduced GFP+ cells from resultant tumors were sorted into wells and treated with dox. In cells from PDX-1, dox impaired viability by three- to four-fold in shYAP-transduced cells (Fig. 5B). No effect of YAP knockdown was observed in PDX-2 cells.

Figure 5.
Figure 5.

YAP depletion has potent activity against some but not all patient derived xenograft melanomas. A, top: Schematic of dox-inducible shYAP construct used to stably deplete YAP in melanoma cells via rapid lentiviral spinfection. Bottom: Western blot analysis indicating dox-inducible YAP depletion by two distinct shYAP constructs, but not scrambled controls (shSCR), in A375 melanoma cells. B, In vitro cell viability after 5 days, relative to input cells at day 0, by MTT assay in dox- vs. vehicle-treated cells purified freshly from two BRAF V600E mutant PDX melanomas (PDX-1 and PDX-2) that were initially grown from uncultured cells rapidly transfected with dox-inducible shYAP3- or shSCR-GFP- LUC (Fig. 4A, top). Transduced GFP+ and nontransduced GFP− cells for dox or vehicle treatment were isolated from each PDX melanoma by flow cytometry. Data points indicate values for each well, lines represent means. *, P < 0.01 (t test). C, In vivo YAP depletion in PDX melanomas (3151, 3155, 3160, and 3154) that were grown in NSG mice from uncultured cells (from PDX-1) transfected with dox-inducible shYAP3- [injected into LF (left flank)] or shSCR- (RF, right flank) GFP-LUC (Fig. 4A, top) and removed one week after dox or vehicle treatment of mice for Western blotting of lysates prepared from flow cytometrically isolated GFP+ cells from each tumor. D, Endpoint analysis of tumors from PDX-1, transduced as above and injected into the right flanks of NSG mice, followed by dox or vehicle treatment. Left: Bioluminescence imaging at endpoint. Middle: Representative flow cytometric analysis of shYAP3- or shSCR-GFP-LUC PDX melanomas from four mice treated ± dox. Right: Summary data of all mice treated in this way, showing marked depletion of GFP+ cells in shYAP3-GFP-LUC PDX melanomas transfected in dox-treated mice. *, P < 0.01 (t test). E, Endpoint flow cytometric analysis of PDX tumors from the PDX-3 line, manipulated and grown in NSG mice treated ± dox, as above. *, P < 0.01 (t test).

We therefore took GFP+ cells from PDX-1 and re-injected them into each flank (left: shYAP, right: shSCR) of NSG mice (3151, 3154, 3155, 3160). At tumor diameter ∼10 mm, mice were placed on dox water (or not) and 1 week later tumors were removed and GFP+ cells isolated by flow cytometry. Immunoblotting of lysates from these cells confirmed in vivo YAP knockdown (Fig. 5C). In parallel experiments using the same cells, single injections of either shYAP or shSCR cells were made into the right flanks of NSG mice. Following tumor formation and commencement of dox, shYAP tumors in dox-treated mice displayed lower luminescence than controls (Fig. 5D, left). At endpoint, tumors were assessed for GFP+ cells, which were markedly depleted in tumors grown from shYAP-GFP-LUC-transduced cells (Fig. 5D, middle and right). YAP dependency was similarly observed in another BRAF V600E mutant PDX melanoma, PDX-3 (Fig. 5E). These data indicate that YAP targeting has in vivo efficacy against uncultured cells in some but not all melanomas.

The Hippo pathway regulates melanoma progression independently of BRAF/NRAS genotype

Further to examine differential sensitivity of cutaneous melanomas to YAP targeting, we depleted YAP by siRNA in a panel of 10 melanoma cell lines representing different genotypes (BRAF mutant, NRAS mutant, BRAF/NRAS wild-type). Consistent with previous observations (Fig. 5B), sensitivity to YAP depletion was varied, if comparatively modest, in A375, HT144, and SK-MEL28 BRAF mutant cells (Fig. 6A). Variable sensitivity to YAP targeting was also observed in four NRAS mutant and three BRAF/NRAS double wild-type cell lines, with only some (C027-M1, HMCB, and CHL1) lines showing impressive sensitivity, even though YAP was efficiently depleted in each (Fig. 6A). These results suggest a role for YAP in the progression of some (but not all) melanomas that is independent of classical driver mutations.

Figure 6.
Figure 6.

YAP, TAZ, and TEAD promote viability in some but not all melanoma cell lines. A, Top: Cell viability in the indicated melanoma cell lines treated with YAP siRNA (Y), assessed by Alamar Blue and compared with mock treatment controls (C). Data are represented as mean ± SEM from three biological replicates. Bottom: Detection of YAP and actin by Western blot analysis in each transfected cell line. B, Top: Cell viability in the indicated melanoma cell lines treated with TEAD1 and TEAD2 siRNAs (T), assessed by Alamar Blue and compared with mock treatment controls (C). Bottom: Detection of TEAD1 and actin by western blot in each transfected cell line. C, Cell viability in the indicated melanoma cell lines treated with siYAP, siTAZ or siYAP+siTAZ, assessed by Alamar Blue and compared with mean mock treatment controls (C) in each line. Doses of siYAP were 50% of the doses used in Fig. 5A. Control, mock transfection without siRNA; siOTP, nontargeting siRNA control. Data are represented as mean ± SEM from three biological replicates. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001; and N/S indicates P > 0.05 (unpaired two-tailed t tests or ANOVA plus Tukey's multiple comparison tests).

To determine whether YAP's function in melanoma cell lines is linked to TEAD transcription factors, we depleted TEAD1 and TEAD2 by siRNA (Fig. 6B). High concordance was observed between cells that were sensitive to TEAD1/2 depletion and those sensitive to YAP depletion. We further treated the same cell lines with verteporfin, a drug that inhibits YAP and TEAD protein interactions (59), and observed a strong correlation between siYAP targeting and the EC50 of verteporfin (Supplementary Fig. S4). These results suggest that YAP promotes survival of select melanomas in partnership with TEAD transcription factors.

As TAZ is a paralogue of YAP, we also tested sensitivity to TAZ targeting by siRNA-depleting YAP or TAZ, or both proteins together. In these assays, siYAP dosage was reduced by 50% to standardize YAP deletion in YAP-only and combined YAP/TAZ targeting. In most cell lines in which YAP-only targeting displayed weak/modest effects (Fig. 6A), TAZ- or combined YAP/TAZ-depletion did not substantially improve anti-melanoma efficacy (Fig. 6A and C). The exception to this was C067-M1 cells, in which combined YAP/TAZ knockdown displayed moderate efficacy whereas single knockdowns were ineffectual. In two BRAF/NRAS double wild-type lines (HMCB, CHL1) that were sensitive to YAP targeting (Fig. 6A and C), TAZ targeting was ineffectual even combined with YAP deletion. In NRAS mutant CO27-M1 cells, YAP and TAZ depletion were modestly effective in isolation but synergistic in combination. These data show that only a subset of melanomas is sensitive to Hippo pathway targeting, and that cotargeting of YAP and TAZ only variably enhances anti-melanoma efficacy.

Discussion

We have established the central Hippo pathway oncoprotein, YAP, as a potential therapeutic target in human melanoma. Previous studies of cell lines found both anti-clonogenic and particularly anti-invasive effects of YAP targeting (23, 24, 60), as well as therapeutic synergism with combined YAP and MAPK pathway targeting (8). We revealed potent anti-melanoma effects of YAP targeting against uncultured, therapy-naïve melanoma cells grown as tumors in highly efficient PDX assays. However, these effects were not evident in all PDX melanomas. We confirmed this variability in sensitivity to YAP targeting by depleting YAP, with or without co-depletion of the YAP paralogue TAZ, in multiple melanoma cell lines. Although sensitivity to YAP or TAZ targeting was independent of classical BRAF/NRAS driver mutations, some lines were highly resistant even to combined YAP/TAZ targeting. Our data highlight the therapeutic potential of YAP/TAZ targeting in melanoma but indicate that elucidation of biomarkers other than YAP protein expression will be critical to predict efficacy.

A striking aspect of our study is the discovery of a YAP allele with multiple point mutations causing seven serine to alanine substitutions in one patient's melanoma. Four of these serines are key residues that are phosphorylated by the central Hippo pathway kinases LATS1 and LATS2, which repress YAP. Mutation of these serines is known to alleviate repression of YAP in flies, mammalian cells, and mice (13, 14, 50, 51, 61, 62). Furthermore, mutation of these residues has a cumulative impact on YAP hyperactivity. Using both mRNA expression and soft agar assays in multiple cell lines, we showed that the hypermutant YAP allele we discovered codes for a highly active YAP protein.

Despite YAP being a potent oncogene, gain-of-function mutations are rare (16); in 368 TCGA melanoma samples, we identified only one YAP mutant melanoma. Another putative YAP gain-of-function mutation reported involved a translocation that fuses the YAP amino-terminus to the transcription factor TFE3 in epithelioid hemangioendothelioma, a rare sarcoma (22). There are several possible reasons for the paucity of oncogenic YAP mutations (16). YAP is subject to multiple layers of regulation, such that removal of just one layer is insufficient to hyperactivate YAP (55); whereas other oncogenes may be rendered oncogenic by a single amino acid change, this is not the case for YAP (63, 64). In mice, the Yap S118 mutation (equivalent to human YAP S127) does not cause obvious Yap hyperactivation phenotypes (63). This is likely explained by a negative feedback loop that activates upstream Hippo proteins that repress YAP (63, 64), such that multiple activating mutations in YAP may be required to overcome this feedback. We speculate that the first single nucleotide substitution in the melanoma from patient CA-04 was the YAP S127A mutation. Subsequently, there was strong evolutionary pressure to relieve this feedback and more LATS-phosphorylation sites in YAP were mutated. The most important of these was probably S397, given that mutations of S127 and S397 cause the strongest YAP hyperactivation (55). As the S397A change was delivered by dinucleotide substitution, this was probably a secondary event.

Another reason for the lack of YAP mutations in cancer is that YAP hyperactivation might not in isolation deliver a growth advantage. For example, YAP gain-of-function mutations might only be permitted in specific genetic or cellular contexts that support YAP hyperactivity. The YAP hypermutant we discovered in patient CA-04 occurred in tetraploid melanoma cells that possessed activating mutations in BRAF (V600E) and RAC1 (P29S), and where the RAC1 gene was also focally amplified. As GNAQ and GNA11 mutations were shown to activate YAP in a RAC1-dependent fashion (19), it is conceivable that RAC1 P29S could also activate YAP. Additionally, as tetraploidy was proposed to promote RAC1 and suppress RhoA, and thereby to activate Hippo pathway tumor suppression via activation of LATS kinases (54), it is plausible that the YAP hypermutant in patient CA-04 developed to overcome this. These possibilities suggest that sensitivity to YAP/TAZ targeting might be predicted by coexisting mutations that co-operate with classical oncogenic drivers in melanoma, such as BRAF and NRAS mutations, to promote YAP/TAZ activity.

Our data suggest that the expression and cellular distribution of YAP will not be useful biomarkers for predicting sensitivity to YAP targeting, as elevated YAP was found in most melanomas. YAP was also and similarly increased in benign nevi, the vast majority of which do not transform to invasive melanoma. As most nevi harbor activating BRAF mutations (65–67) whose oncogenicity is thought to be blocked by tumor suppressors such as p16INK4A (68), increased YAP expression in nevi might be linked to primary MAPK pathway activation in a manner that is usually unable co-operatively to induce malignant transformation in the face of intact tumor suppression.

Further to the paucity of SNVs directly affecting YAP and TAZ (Fig. 3A), we unexpectedly found that CNAs affecting these genes were frequently deletion events, rather than amplifications, with YAP particularly affected by hemizygous deletion in 44% of TCGA cases. Additionally, although TAZ DNA methylation was strongly negatively correlated with TAZ mRNA, suggesting epigenetic regulation of this gene, YAP was strikingly hypomethylated in every TCGA sample evaluated, regardless of mRNA levels. The infrequency of YAP methylation, YAP point mutations and YAP amplifications in the TCGA data (Fig. 4), as well as widespread intralesional heterogeneity in YAP expression across a spectrum of melanocytic lesions (Fig. 1), suggest dominant regulation of YAP activity via transcriptional, translational, and/or posttranslational mechanisms.

From a therapy perspective, our study suggests that targeting the Hippo pathway will be effective against some but not all melanomas. In future studies, it will be important to define the biomarkers that predict response to YAP inhibition so that patients in clinical trials are a priori likely to respond to anti-YAP therapies.

Disclosure of Potential Conflicts of Interest

G.V. Long is a consultant/advisory board member of Bristol-Myers Squibb, Novartis, Roche, Amgen, Pierre Fabre, Array, Merk, and Incycte. No potential conflicts of interest were disclosed by the other authors.

Authors' Contributions

Conception and design: X. Zhang, J.Z. Tang, K.F. Harvey, M. Shackleton

Development of methodology: X. Zhang, J.Z. Tang, Y. Zhang, G.V. Long, A.T. Papenfuss, M. Shackleton

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): X. Zhang, J.Z. Tang, Y. Zhang, P. Szeto, L. Yang, C. Mintoff, A. Colebatch, L. McIntosh, K.A. Mitchell, E. Shaw, H. Rizos, G.V. Long, G.A. McArthur, A.T. Papenfuss, M. Shackleton

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): X. Zhang, J.Z. Tang, I.A. Vergara, Y. Zhang, L. Yang, E. Shaw, H. Rizos, G.V. Long, A.T. Papenfuss, K.F. Harvey, M. Shackleton

Writing, review, and/or revision of the manuscript: X. Zhang, J.Z. Tang, I.A. Vergara, L. Yang, H. Rizos, G.V. Long, N. Hayward, G.A. McArthur, A.T. Papenfuss, K.F. Harvey, M. Shackleton

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): X. Zhang, M. Shackleton

Study supervision: X. Zhang, N. Hayward, A.T. Papenfuss, K.F. Harvey, M. Shackleton

Acknowledgments

We thank David Huang, Marco Herold, Christopher Schmidt and Karen Sheppard for cell lines and plasmids. J.Z. Tang was supported by a Victorian Cancer Agency (VCA) Early Career Fellowship (ECSG13001). K.F. Harvey was supported by a National Health and Medical Research Council (NHMRC) Senior Research Fellowship (1078220). A.T. Papenfuss was supported by the Lorenzo and Pamela Galli Charitable Trust and by an NHMRC Program Grant (1054618) and a NHMRC Senior Research Fellowship (1116955). M. Shackleton was supported by Pfizer Australia, NHMRC (628735), veski (200910), and VCA (CRF15007) Fellowships. This research was supported by the Cancer Council of Victoria (1102820), the Victorian Cancer Biobank, the Melbourne Melanoma Project, and the Victorian State Government Operational Infrastructure Support and Australian Government NHMRC Independent Research Institute Infrastructure Support Programs.

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.

Footnotes

  • Note: Supplementary data for this article are available at Molecular Cancer Research Online (http://mcr.aacrjournals.org/).

  • K.F. Harvey and M. Shackleton contributed equally and are co-senior authors of this article.

  • Mol Cancer Res 2019;17:1435–49

  • Received April 25, 2018.
  • Revision received October 25, 2018.
  • Accepted February 28, 2019.
  • Published first March 4, 2019.

References

  1. 1.↵
    1. Curtin JA,
    2. Fridlyand J,
    3. Kageshita T,
    4. Patel HN,
    5. Busam KJ,
    6. Kutzner H,
    7. et al.
    Distinct sets of genetic alterations in melanoma. N Engl J Med 2005;353:2135–47.
    OpenUrlCrossRefPubMed
  2. 2.↵
    1. Larkin J,
    2. Ascierto PA,
    3. Dreno B,
    4. Atkinson V,
    5. Liszkay G,
    6. Maio M,
    7. et al.
    Combined vemurafenib and cobimetinib in BRAF-mutated melanoma. N Engl J Med 2014;371:1867–76.
    OpenUrlCrossRefPubMed
  3. 3.↵
    1. Long GV,
    2. Stroyakovskiy D,
    3. Gogas H,
    4. Levchenko E,
    5. de Braud F,
    6. Larkin J,
    7. et al.
    Combined BRAF and MEK inhibition versus BRAF inhibition alone in melanoma. N Engl J Med 2014;371:1877–88.
    OpenUrlCrossRefPubMed
  4. 4.↵
    1. Van Allen EM,
    2. Wagle N,
    3. Sucker A,
    4. Treacy DJ,
    5. Johannessen CM,
    6. Goetz EM,
    7. et al.
    The genetic landscape of clinical resistance to RAF inhibition in metastatic melanoma. Cancer Discov 2014;4:94–109.
    OpenUrlAbstract/FREE Full Text
  5. 5.↵
    1. Grimaldi AM,
    2. Simeone E,
    3. Ascierto PA
    . The role of MEK inhibitors in the treatment of metastatic melanoma. Curr Opin Oncol 2014;26:196–203.
    OpenUrl
  6. 6.↵
    1. Hodi FS,
    2. O'Day SJ,
    3. McDermott DF,
    4. Weber RW,
    5. Sosman JA,
    6. Haanen JB,
    7. et al.
    Improved survival with ipilimumab in patients with metastatic melanoma. N Engl J Med 2010;363:711–23.
    OpenUrlCrossRefPubMed
  7. 7.↵
    1. Larkin J,
    2. Chiarion-Sileni V,
    3. Gonzalez R,
    4. Grob JJ,
    5. Cowey CL,
    6. Lao CD,
    7. et al.
    Combined nivolumab and ipilimumab or monotherapy in untreated melanoma. N Engl J Med 2015;373:23–34.
    OpenUrlCrossRefPubMed
  8. 8.↵
    1. Lin L,
    2. Sabnis AJ,
    3. Chan E,
    4. Olivas V,
    5. Cade L,
    6. Pazarentzos E,
    7. et al.
    The Hippo effector YAP promotes resistance to RAF- and MEK-targeted cancer therapies. Nat Genet 2015;47:250–6.
    OpenUrlCrossRefPubMed
  9. 9.↵
    1. Johannessen CM,
    2. Johnson LA,
    3. Piccioni F,
    4. Townes A,
    5. Frederick DT,
    6. Donahue MK,
    7. et al.
    A melanocyte lineage program confers resistance to MAP kinase pathway inhibition. Nature 2013;504:138–42.
    OpenUrlCrossRefPubMed
  10. 10.↵
    1. Shalem O,
    2. Sanjana NE,
    3. Hartenian E,
    4. Shi X,
    5. Scott DA,
    6. Mikkelsen TS,
    7. et al.
    Genome-scale CRISPR-Cas9 knockout screening in human cells. Science 2014;343:84–7.
    OpenUrlAbstract/FREE Full Text
  11. 11.↵
    1. Johnson R,
    2. Halder G
    . The two faces of Hippo: targeting the Hippo pathway for regenerative medicine and cancer treatment. Nat Rev Drug Discov 2014;13:63–79.
    OpenUrlCrossRefPubMed
  12. 12.↵
    1. Tapon N,
    2. Harvey KF,
    3. Bell DW,
    4. Wahrer DC,
    5. Schiripo TA,
    6. Haber D,
    7. et al.
    Salvador promotes both cell cycle exit and apoptosis in Drosophila and is mutated in human cancer cell lines. Cell 2002;110:467–78.
    OpenUrlCrossRefPubMed
  13. 13.↵
    1. Camargo FD,
    2. Gokhale S,
    3. Johnnidis JB,
    4. Fu D,
    5. Bell GW,
    6. Jaenisch R,
    7. et al.
    YAP1 increases organ size and expands undifferentiated progenitor cells. Curr Biol 2007;17:2054–60.
    OpenUrlCrossRefPubMed
  14. 14.↵
    1. Dong J,
    2. Feldmann G,
    3. Huang J,
    4. Wu S,
    5. Zhang N,
    6. Comerford SA,
    7. et al.
    Elucidation of a universal size-control mechanism in Drosophila and mammals. Cell 2007;130:1120–33.
    OpenUrlCrossRefPubMed
  15. 15.↵
    1. Zhang L
    . Control of growth and beyond: a special issue on Hippo signaling. Acta Biochim Biophys Sin (Shanghai) 2015;47:1.
    OpenUrlCrossRefPubMed
  16. 16.↵
    1. Harvey KF,
    2. Zhang X,
    3. Thomas DM
    . The Hippo pathway and human cancer. Nat Rev Cancer 2013;13:246–57.
    OpenUrlCrossRefPubMed
  17. 17.↵
    1. Sekido Y,
    2. Pass HI,
    3. Bader S,
    4. Mew DJ,
    5. Christman MF,
    6. Gazdar AF,
    7. et al.
    Neurofibromatosis type 2 (NF2) gene is somatically mutated in mesothelioma but not in lung cancer. Cancer Res 1995;55:1227–31.
    OpenUrlAbstract/FREE Full Text
  18. 18.↵
    1. Lomas J,
    2. Bello MJ,
    3. Arjona D,
    4. Alonso ME,
    5. Martinez-Glez V,
    6. Lopez-Marin I,
    7. et al.
    Genetic and epigenetic alteration of the NF2 gene in sporadic meningiomas. Genes Chromosomes Cancer 2005;42:314–9.
    OpenUrlCrossRefPubMed
  19. 19.↵
    1. Feng X,
    2. Degese MS,
    3. Iglesias-Bartolome R,
    4. Vaque JP,
    5. Molinolo AA,
    6. Rodrigues M,
    7. et al.
    Hippo-independent activation of YAP by the GNAQ uveal melanoma oncogene through a trio-regulated rho GTPase signaling circuitry. Cancer Cell 2014;25:831–45.
    OpenUrlCrossRefPubMed
  20. 20.↵
    1. Van Raamsdonk CD,
    2. Bezrookove V,
    3. Green G,
    4. Bauer J,
    5. Gaugler L,
    6. O'Brien JM,
    7. et al.
    Frequent somatic mutations of GNAQ in uveal melanoma and blue naevi. Nature 2009;457:599–602.
    OpenUrlCrossRefPubMed
  21. 21.↵
    1. Yu FX,
    2. Luo J,
    3. Mo JS,
    4. Liu G,
    5. Kim YC,
    6. Meng Z,
    7. et al.
    Mutant Gq/11 promote uveal melanoma tumorigenesis by activating YAP. Cancer Cell 2014;25:822–30.
    OpenUrlCrossRefPubMed
  22. 22.↵
    1. Antonescu CR,
    2. Le Loarer F,
    3. Mosquera JM,
    4. Sboner A,
    5. Zhang L,
    6. Chen CL,
    7. et al.
    Novel YAP1-TFE3 fusion defines a distinct subset of epithelioid hemangioendothelioma. Genes Chromosomes Cancer 2013;52:775–84.
    OpenUrlCrossRefPubMed
  23. 23.↵
    1. Menzel M,
    2. Meckbach D,
    3. Weide B,
    4. Toussaint NC,
    5. Schilbach K,
    6. Noor S,
    7. et al.
    In melanoma, Hippo signaling is affected by copy number alterations and YAP1 overexpression impairs patient survival. Pigment Cell Melanoma Res 2014;27:671–3.
    OpenUrl
  24. 24.↵
    1. Nallet-Staub F,
    2. Marsaud V,
    3. Li L,
    4. Gilbert C,
    5. Dodier S,
    6. Bataille V,
    7. et al.
    Pro-invasive activity of the Hippo pathway effectors YAP and TAZ in cutaneous melanoma. J Invest Dermatol 2014;134:123–32.
    OpenUrlCrossRefPubMed
  25. 25.↵
    1. Martin M
    . Cutadapt removes adapter sequences from high-throughput sequencing reads. EMBnet.journal 2011;17:10–2.
    OpenUrlCrossRefPubMed
  26. 26.↵
    1. Li H,
    2. Durbin R
    . Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinformatics 2009;25:1754–60.
    OpenUrlCrossRefPubMed
  27. 27.↵
  28. 28.↵
    1. DePristo MA,
    2. Banks E,
    3. Poplin R,
    4. Garimella KV,
    5. Maguire JR,
    6. Hartl C,
    7. et al.
    A framework for variation discovery and genotyping using next-generation DNA sequencing data. Nat Genet 2011;43:491–8.
    OpenUrlCrossRefPubMed
  29. 29.↵
    1. Li H,
    2. Handsaker B,
    3. Wysoker A,
    4. Fennell T,
    5. Ruan J,
    6. Homer N,
    7. et al.
    The sequence alignment/map format and SAMtools. Bioinformatics 2009;25:2078–9.
    OpenUrlCrossRefPubMed
  30. 30.↵
    1. Koboldt DC,
    2. Zhang Q,
    3. Larson DE,
    4. Shen D,
    5. McLellan MD,
    6. Lin L,
    7. et al.
    VarScan 2: somatic mutation and copy number alteration discovery in cancer by exome sequencing. Genome Res 2012;22:568–76.
    OpenUrlAbstract/FREE Full Text
  31. 31.↵
    1. Cibulskis K,
    2. Lawrence MS,
    3. Carter SL,
    4. Sivachenko A,
    5. Jaffe D,
    6. Sougnez C,
    7. et al.
    Sensitive detection of somatic point mutations in impure and heterogeneous cancer samples. Nat Biotechnol 2013;31:213–9.
    OpenUrlCrossRefPubMed
  32. 32.↵
    1. Smit AFA,
    2. Hubley R,
    3. Green P
    . RepeatMasker Open-4.0. 2013–2015. Available from: http://www.repeatmasker.org/.
  33. 33.↵
    1. Li H
    . Toward better understanding of artifacts in variant calling from high-coverage samples. Bioinformatics 2014;30:2843–51.
    OpenUrlCrossRefPubMed
  34. 34.↵
    1. Auton A,
    2. Brooks LD,
    3. Durbin RM,
    4. Garrison EP,
    5. Kang HM,
    6. et al.
    1000 Genomes Project Consortium, Auton A, Brooks LD, Durbin RM, Garrison EP, Kang HM, et al. A global reference for human genetic variation. Nature 2015;526:68–74.
    OpenUrlCrossRefPubMed
  35. 35.↵
    1. Li MX,
    2. Kwan JS,
    3. Bao SY,
    4. Yang W,
    5. Ho SL,
    6. Song YQ,
    7. et al.
    Predicting mendelian disease-causing non-synonymous single nucleotide variants in exome sequencing studies. PLos Genet 2013;9:e1003143.
    OpenUrlCrossRefPubMed
  36. 36.↵
    1. MacArthur DG,
    2. Balasubramanian S,
    3. Frankish A,
    4. Huang N,
    5. Morris J,
    6. Walter K,
    7. et al.
    A systematic survey of loss-of-function variants in human protein-coding genes. Science 2012;335:823–8.
    OpenUrlAbstract/FREE Full Text
  37. 37.↵
    1. Lawrence MS,
    2. Stojanov P,
    3. Polak P,
    4. Kryukov GV,
    5. Cibulskis K,
    6. Sivachenko A,
    7. et al.
    Mutational heterogeneity in cancer and the search for new cancer-associated genes. Nature 2013;499:214–8.
    OpenUrlCrossRefPubMed
  38. 38.↵
    1. Li Y,
    2. Xie X
    . Deconvolving tumor purity and ploidy by integrating copy number alterations and loss of heterozygosity. Bioinformatics 2014;30:2121–9.
    OpenUrlCrossRefPubMed
  39. 39.↵
    1. McLaren W,
    2. Pritchard B,
    3. Rios D,
    4. Chen Y,
    5. Flicek P,
    6. Cunningham F
    . Deriving the consequences of genomic variants with the Ensembl API and SNP Effect Predictor. Bioinformatics 2010;26:2069–70.
    OpenUrlCrossRefPubMed
  40. 40.↵
    1. Favero F,
    2. Joshi T,
    3. Marquard AM,
    4. Birkbak NJ,
    5. Krzystanek M,
    6. Li Q,
    7. et al.
    Sequenza: allele-specific copy number and mutation profiles from tumor sequencing data. Ann Oncol 2015;26:64–70.
    OpenUrlCrossRefPubMed
  41. 41.↵
    1. Cerami E,
    2. Gao J,
    3. Dogrusoz U,
    4. Gross BE,
    5. Sumer SO,
    6. Aksoy BA,
    7. et al.
    The cBio cancer genomics portal: an open platform for exploring multidimensional cancer genomics data. Cancer Discov 2012;2:401–4.
    OpenUrlAbstract/FREE Full Text
  42. 42.↵
    1. Gao J,
    2. Aksoy BA,
    3. Dogrusoz U,
    4. Dresdner G,
    5. Gross B,
    6. Sumer SO,
    7. et al.
    Integrative analysis of complex cancer genomics and clinical profiles using the cBioPortal. Sci Signal 2013;6:pl1.
    OpenUrlAbstract/FREE Full Text
  43. 43.↵
    1. Herold MJ,
    2. van den Brandt J,
    3. Seibler J,
    4. Reichardt HM
    . Inducible and reversible gene silencing by stable integration of an shRNA-encoding lentivirus in transgenic rats. Proc Natl Acad Sci U S A 2008;105:18507–12.
    OpenUrlAbstract/FREE Full Text
  44. 44.↵
    1. Quintana E,
    2. Piskounova E,
    3. Shackleton M,
    4. Weinberg D,
    5. Eskiocak U,
    6. Fullen DR,
    7. et al.
    Human melanoma metastasis in NSG mice correlates with clinical outcome in patients. Sci Transl Med 2012;4:159ra49.
    OpenUrl
  45. 45.↵
    1. Quintana E,
    2. Shackleton M,
    3. Sabel MS,
    4. Fullen DR,
    5. Johnson TM,
    6. Morrison SJ
    . Efficient tumour formation by single human melanoma cells. Nature 2008;456:593–8.
    OpenUrlCrossRefPubMed
  46. 46.↵
    1. Zhang X,
    2. Milton CC,
    3. Humbert PO,
    4. Harvey KF
    . Transcriptional output of the Salvador/warts/hippo pathway is controlled in distinct fashions in Drosophila melanogaster and mammalian cell lines. Cancer Res 2009;69:6033–41.
    OpenUrlAbstract/FREE Full Text
  47. 47.↵
    1. Poon CL,
    2. Zhang X,
    3. Lin JI,
    4. Manning SA,
    5. Harvey KF
    . Homeodomain-interacting protein kinase regulates hippo pathway-dependent tissue growth. Curr Biol 2012;22:1587–94.
    OpenUrlCrossRefPubMed
  48. 48.↵
    1. Meng Z,
    2. Moroishi T,
    3. Mottier-Pavie V,
    4. Plouffe SW,
    5. Hansen CG,
    6. Hong AW,
    7. et al.
    MAP4K family kinases act in parallel to MST1/2 to activate LATS1/2 in the Hippo pathway. Nat Commun 2015;6:8357.
    OpenUrlCrossRefPubMed
  49. 49.↵
    1. Schlegelmilch K,
    2. Mohseni M,
    3. Kirak O,
    4. Pruszak J,
    5. Rodriguez JR,
    6. Zhou D,
    7. et al.
    Yap1 acts downstream of alpha-catenin to control epidermal proliferation. Cell 2011;144:782–95.
    OpenUrlCrossRefPubMed
  50. 50.↵
    1. Zhao B,
    2. Wei X,
    3. Li W,
    4. Udan RS,
    5. Yang Q,
    6. Kim J,
    7. et al.
    Inactivation of YAP oncoprotein by the Hippo pathway is involved in cell contact inhibition and tissue growth control. Genes Dev 2007;21:2747–61.
    OpenUrlAbstract/FREE Full Text
  51. 51.↵
    1. Hao Y,
    2. Chun A,
    3. Cheung K,
    4. Rashidi B,
    5. Yang X
    . Tumor suppressor LATS1 is a negative regulator of oncogene YAP. J Biol Chem 2008;283:5496–509.
    OpenUrlAbstract/FREE Full Text
  52. 52.↵
    1. Wang LH,
    2. Baker NE
    . Salvador-warts-hippo pathway in a developmental checkpoint monitoring helix-loop-helix proteins. Dev Cell 2015;32:191–202.
    OpenUrlCrossRefPubMed
  53. 53.↵
    1. Mo JS,
    2. Meng Z,
    3. Kim YC,
    4. Park HW,
    5. Hansen CG,
    6. Kim S,
    7. et al.
    Cellular energy stress induces AMPK-mediated regulation of YAP and the Hippo pathway. Nat Cell Biol 2015;17:500–10.
    OpenUrlCrossRefPubMed
  54. 54.↵
    1. Ganem NJ,
    2. Cornils H,
    3. Chiu SY,
    4. O'Rourke KP,
    5. Arnaud J,
    6. Yimlamai D,
    7. et al.
    Cytokinesis failure triggers hippo tumor suppressor pathway activation. Cell 2014;158:833–48.
    OpenUrlCrossRefPubMed
  55. 55.↵
    1. Zhao B,
    2. Kim J,
    3. Ye X,
    4. Lai ZC,
    5. Guan KL
    . Both TEAD-binding and WW domains are required for the growth stimulation and oncogenic transformation activity of yes-associated protein. Cancer Res 2009;69:1089–98.
    OpenUrlAbstract/FREE Full Text
  56. 56.↵
    1. Akbani R,
    2. Akdemir Kadir C,
    3. Aksoy BA,
    4. Albert M,
    5. Ally A,
    6. Amin Samirkumar B,
    7. et al.
    Genomic classification of cutaneous melanoma. Cell 2015;161:1681–96.
    OpenUrlCrossRefPubMed
  57. 57.↵
    1. Zhang X,
    2. George J,
    3. Deb S,
    4. Degoutin JL,
    5. Takano EA,
    6. Fox SB,
    7. et al.
    The Hippo pathway transcriptional co-activator, YAP, is an ovarian cancer oncogene. Oncogene 2011;30:2810–22.
    OpenUrlCrossRefPubMed
  58. 58.↵
    1. Boyle SE,
    2. Fedele CG,
    3. Corbin V,
    4. Wybacz E,
    5. Szeto P,
    6. Lewin J,
    7. et al.
    CD271 expression on patient melanoma cells is unstable and unlinked to tumorigenicity. Cancer Res 2016;76:3965–77.
    OpenUrlAbstract/FREE Full Text
  59. 59.↵
    1. Liu-Chittenden Y,
    2. Huang B,
    3. Shim JS,
    4. Chen Q,
    5. Lee SJ,
    6. Anders RA,
    7. et al.
    Genetic and pharmacological disruption of the TEAD-YAP complex suppresses the oncogenic activity of YAP. Genes Dev 2012;26:1300–5.
    OpenUrlAbstract/FREE Full Text
  60. 60.↵
    1. Lamar JM,
    2. Stern P,
    3. Liu H,
    4. Schindler JW,
    5. Jiang ZG,
    6. Hynes RO
    . The Hippo pathway target, YAP, promotes metastasis through its TEAD-interaction domain. Proc Natl Acad Sci U S A 2012;109:E2441–50.
    OpenUrlAbstract/FREE Full Text
  61. 61.↵
    1. Oh H,
    2. Irvine KD
    . In vivo regulation of Yorkie phosphorylation and localization. Development 2008;135:1081–8.
    OpenUrlAbstract/FREE Full Text
  62. 62.↵
    1. Oka T,
    2. Mazack V,
    3. Sudol M
    . Mst2 and Lats kinases regulate apoptotic function of Yes kinase-associated protein (YAP). J Biol Chem 2008;283:27534–46.
    OpenUrlAbstract/FREE Full Text
  63. 63.↵
    1. Chen Q,
    2. Zhang N,
    3. Xie R,
    4. Wang W,
    5. Cai J,
    6. Choi KS,
    7. et al.
    Homeostatic control of Hippo signaling activity revealed by an endogenous activating mutation in YAP. Genes Dev 2015;29:1285–97.
    OpenUrlAbstract/FREE Full Text
  64. 64.↵
    1. Moroishi T,
    2. Park HW,
    3. Qin B,
    4. Chen Q,
    5. Meng Z,
    6. Plouffe SW,
    7. et al.
    A YAP/TAZ-induced feedback mechanism regulates Hippo pathway homeostasis. Genes Dev 2015;29:1271–84.
    OpenUrlAbstract/FREE Full Text
  65. 65.↵
    1. Pollock PM,
    2. Harper UL,
    3. Hansen KS,
    4. Yudt LM,
    5. Stark M,
    6. Robbins CM,
    7. et al.
    High frequency of BRAF mutations in nevi. Nat Genet 2003;33:19–20.
    OpenUrlCrossRefPubMed
  66. 66.↵
    1. Poynter JN,
    2. Elder JT,
    3. Fullen DR,
    4. Nair RP,
    5. Soengas MS,
    6. Johnson TM,
    7. et al.
    BRAF and NRAS mutations in melanoma and melanocytic nevi. Melanoma Res 2006;16:267–73.
    OpenUrlCrossRefPubMed
  67. 67.↵
    1. Shain AH,
    2. Bastian BC
    . From melanocytes to melanomas. Nat Rev Cancer 2016;16:345–58.
    OpenUrlCrossRefPubMed
  68. 68.↵
    1. Shain AH,
    2. Yeh I,
    3. Kovalyshyn I,
    4. Sriharan A,
    5. Talevich E,
    6. Gagnon A,
    7. et al.
    The genetic evolution of melanoma from precursor lesions. N Engl J Med 2015;373:1926–36.
    OpenUrlCrossRefPubMed
View Abstract
Back to top
View this article with LENS

Open full page PDF
Article Alerts
Email Article
Citation Tools
Somatic Hypermutation of the YAP Oncogene in a Human Cutaneous Melanoma
Xiaomeng Zhang, Jian Zhong Tang, Ismael A. Vergara, Youfang Zhang, Pacman Szeto, Lie Yang, Christopher Mintoff, Andrew Colebatch, Lachlan McIntosh, Katrina A. Mitchell, Evangeline Shaw, Helen Rizos, Georgina V. Long, Nicholas Hayward, Grant A. McArthur, Anthony T. Papenfuss, Kieran F. Harvey and Mark Shackleton
Mol Cancer Res July 1 2019 (17) (7) 1435-1449; DOI: 10.1158/1541-7786.MCR-18-0407
del.icio.us logo Digg logo Reddit logo Twitter logo CiteULike logo Facebook logo Google logo Mendeley logo

Jump to section

Advertisement