Cell culture

Immortalized melanocyte (Mel-ST) cells were a gift from the lab of Dr. Robert Weinberg. Mel-ST cells, and all derivative cell lines generated in this study, were grown in DMEM media containing 5% fetal bovine serum (FBS), 100 IU/mL penicillin, and 100 µg/mL streptomycin. Human Embryonic Kidney 293 (HEK293A) cells, and all derivative cell lines, were grown in DMEM media containing 10% FBS, 100 IU/mL penicillin, and 100 µg/mL streptomycin. hTERT-BJ fibroblasts, and all derivative cell lines, were grown in DMEM:F12 media containing 10% FBS, 100 IU/mL penicillin, and 100 µg/mL streptomycin. BrafV600E/Pten−/− mouse tumor cells (D4M.3 A) were a generous gift of Dr. Constance Brinckerhoff. D4M.3 A mouse tumor cells were maintained in DMEM:F12 media containing 5% FBS. Primary adult epidermal melanocytes were purchased from ATCC (PCS-200-013) and maintained in Dermal Cell Basal Medium (ATCC PCS-200-030) supplemented with an Adult Melanocyte Growth Kit (ATCC PCS-200-042). All FBS used in these studies was confirmed to either be naturally absent of tetracyclines or below 20 ng/mL by the manufacturer. All cells were maintained at 37 °C with 5% CO 2 atmosphere and maintained at sub-confluent levels for passaging and all experiments. Cultures were regularly checked for mycoplasma contamination utilizing a PCR detection kit (G238, ABM) or Hoechst staining. Bright-field images of tissue culture cells were captured on an Echo Revolve Hybrid Microscopy system at ×10 or ×20 (Echo Laboratories).

Cell-line generation

To generate the BRAFV600E doxycycline-inducible system Mel-ST, HEK293A, or hTERT-BJ fibroblasts were first infected with lentivirus generated from pLenti CMV TetR Blast (Tet Repressor) and selected. Following selection, cells were infected with lentivirus generated from pLenti CMV/TO BRAFV600E Neo or pLenti CMV/TO BRAF Neo, selected, and single-cell cloned to establish cell lines which demonstrated no basal expression at baseline and strong induction after doxycycline addition. The expression of BRAFV600E was confirmed using two different mutant-specific antibodies (VE1 and RM8 clones). To generate stably expressing H2B-GFP lines, cells were infected with lentivirus generated from pLenti H2B-GFP Blast. Mel-ST cells stably expressing empty vector (pLVX Puro), YAP-5SA (pBABE YAP-5SA Puro), and TAZ-4SA (pLVX Flag TAZ-4SA Puro) were generated via viral infection followed by selection. Tetraploid cells were generated by treating asynchronous cells with 4 µM DCB for 16 h, followed by gentle washing to remove drug (5 × 5 min); completion of cytokinesis was confirmed by phase-contrast imaging.

Viral infections and transfections

Mel-ST, HEK293A, or BJ Fibroblasts were infected for 12–16 h with virus-carrying genes of interest in the presence of 10 µg/mL polybrene, washed, and allowed to recover for 24 h before selection or single-cell cloning. Short-term viral infection of primary melanocytes was carried out similarly, but with 2 µg/mL of polybrene. All RNAi transfections were performed using 25–50 nM siRNA with Lipofectamine RNAi MAX according to the manufacturer’s instructions. Briefly, cells were seeded in 6 or 12-well plates either by addition of reverse transfection mixture overnight for 18 h, which was then washed with PBS and replaced with fresh media, or forward transfection mixture for 4 h, which was then replaced with fresh media. Cells were then incubated for 48–72 h prior to lysis at sub-confluent levels.

Plasmid generation

Plasmids encoding the Tetracycline Repressor, pLenti CMV TetR Blast (716-1), was a gift from Eric Campeau & Paul Kaufman (Addgene Plasmid #17492). To create pLenti CMV/TO BRAF Neo and pLenti CMV/TO BRAFV600E Neo, we performed Gateway cloning using Gateway LR Clonase II (Invitrogen) according to manufacturer instructions to insert BRAF or BRAFV600E into pLenti CMV/TO Neo DEST (685-3) using pENTR BRAF or pENTR BRAFV600E. pENTR BRAF and pENTR BRAFV600E were gifts from Craig Ceol and pLenti CMV/TO Neo DEST (685-3) was a gift from Eric Campeau & Paul Kaufman (Addgene plasmid #17292). To generate pLenti CMV/TO NRASQ61R Neo we performed Gateway cloning to insert NRASQ61R from the donor vector pDONR223 NRASQ61R into the destination vector pLenti CMV/TO Neo Dest. pDONR223 NRASQ61R was a gift from Jesse Boehm, William Hahn, and David Root (Addgene plasmid #81652).

Immunofluorescence and confocal microscopy

Cells were plated on glass coverslips, treated as indicated, washed in 1× phosphate-buffered saline (PBS) (Boston Bioproducts), and fixed in 4% paraformaldehyde for 10 min. Cells were then washed in PBS-0.01% Triton X-100, extracted in PBS-0.2% Triton X-100 for 10 min, blocked in Tris-buffered saline (TBS)-bovine serum albumin (BSA) (10 mM Tris, pH 7.5, 150 mM NaCl, 5% bovine serum antigen, 0.2% sodium azide) for 1 h, and incubated with primary antibodies diluted in TBS-BSA for 1 h at RT or overnight at 4 °C in a humidified chamber. Primary antibodies were visualized using species-specific fluorescent secondary antibodies (Molecular Probes, Alexa Fluor secondaries, 488 nm, 568 nm, 1:500) and DNA was detected with 2.5 µg/mL Hoechst. F-actin was visualized using rhodamine-conjugated phalloidin (1:2000, Molecular Probes, R415). Immunofluorescence images for analysis were collected on a Nikon Ti-E inverted microscope equipped with a Zyla 4.2 PLUS (Andor) and X-Cite 120 LED light source at the same exposure. Confocal immunofluorescence images were collected on a Nikon Ti-E inverted microscope equipped with a C2 + laser scanning confocal head with 405 nm, 488 nm, 561 nm, 640 nm laser lines. Z-stacks were acquired with a series of 0.5–1 µm optical slices which were then converted into a single, max-intensity projected image. Images were analyzed using NIS-Elements Advanced Research (AR) and ImageJ (Version 1.51). To assess YAP localization, two small square regions of interest were drawn at random in individual cells with one in the nucleus, and one in the cytoplasm. The background-corrected, mean fluorescence intensity of YAP was subsequently measured in these regions of interest and a nuclear to cytoplasmic ratio was determined. To assess stress fiber quantity, images were background-corrected, contrast normalized, and then fibers obvious to the naked eye were counted.

Live-cell imaging

Cells stably expressing H2B-GFP were grown on glass-bottom 12-well tissue culture-treated dishes (Cellvis) and treated with drugs of interest. Immediately post-treatment imaging was performed on a Nikon Ti-E inverted microscope equipped with the Nikon Perfect Focus system. The microscope stage was enclosed within a temperature and atmosphere-controlled environment at 37 °C and 5% humidified CO 2 . Fluorescent or bright-field images were captured every 5–10 min with an ×10 or 20×0.5 NA Plan Fluor objective at multiple locations for 72–96 h. All captured images were analyzed using NIS-Elements AR software. The mitotic length was calculated by counting the duration from nuclear envelope breakdown to anaphase onset.

Tissue staining

At the experimental endpoint mice were euthanized and mouse tumors or skin samples were dissected and immediately fixed in 4% paraformaldehyde (PFA) in PBS for 16 h at 4 °C. Tissues were then paraffin-embedded for sectioning and mounting. PFA-fixed paraffin-embedded tissue sections were cut at 5 µm and mounted onto positively charged coverslips (Colorfrost Plus, Thermo). Mounted tissue samples were deparaffinized using xylenes and rehydrated via an ethanol:water gradient. For hematoxylin and eosin staining (H&E), the tissues were incubated in hematoxylin (#14166, Cell Signaling Technology), rinsed in water, differentiated with 1% acid ethanol, blued with 0.1% sodium bicarbonate solution, rinsed in water, dehydrated, cleared, and mounted with Cytoseal XYL (Thermo Scientific). For immunohistochemistry (IHC) or immunofluorescence (IF), antigen unmasking was performed on rehydrated tissue sections using either a citric-acid retrieval buffer (Vector Labs) or Tris-EDTA retrieval buffer (10 mM Tris, 1 mM EDTA, 0.05% Tween-20). Heat-mediated antigen retrieval was performed using either a standard microwave (95 °C, 20 min) or Decloaking Chamber NxGen (Biocare Medical) (110 °C, 12 min), followed by cooling to room temperature (~30–60 min). Citric-acid retrieval was used for most antibodies; Tris-EDTA retrieval was used for MelanA antibodies. For IHC, tissue sections were washed, endogenous peroxidase activity was quenched using 3% hydrogen peroxide in PBS for 10 min, and then blocked for 1 h in 10% goat serum in TBS (Sigma-Aldrich). For IF, tissue sections were blocked for 1 h in 10% goat serum in TBS. Following serum block, if necessary, tissue was incubated with Rodent Block M (Biocare Medical) for 30 min to block endogenous mouse IgG prior to murine primary antibody addition. Primary antibodies were diluted in 10% goat serum in TBS and incubated overnight at 4 °C in a humidified chamber. Following primary addition for IHC, slides were washed with TBS-0.01% Tween-20, incubated with anti-rabbit or mouse SignalStain Boost IHC detection reagent (Cell Signaling Technology) for 30 min and then developed with SignalStain DAB substrate kit (Cell Signaling Technology) according to manufacturer’s instructions. Counterstaining was performed using hematoxylin, followed by dehydration, clearing, and mounting with Cytoseal XYL (Thermo Fisher). Images were captured at randomly selected points using a Nikon Ti-E inverted microscope equipped with a DS-Ri2 (Nikon). For IF, slides were incubated with species-specific fluorescent secondary antibodies (Molecular Probes) and 2.5 µg/mL Hoechst for 1 h at room temperature in a dark humidified chamber. Auto-fluorescence was quenched using Vector TrueVIEW according to the manufacturer’s instructions, and slides were mounted using Prolong Gold Antifade (Invitrogen). Images were captured at randomly selected points using a Nikon Ti-E inverted microscope equipped with a Zyla 4.2 PLUS (Andor) and X-Cite 120 LED light source. For all staining experiments, tissue-specific secondary controls were included in each staining experiment to ensure specificity and control for endogenous tissue pigmentation levels.

Protein extraction and immunoblotting

Cells were gently washed twice in ice-cold 1X PBS and lysed using ice-cold cell lysis buffer (50 mM Tris-HCl, 2% w/v SDS, 10% glycerol) containing 1X HALT (dual-phosphatase and protease inhibitor, Thermo Fisher). Lysates were sonicated at 20% amplitude for 20 s, diluted in 4X Sample Buffer (Boston BioProducts), and resolved using SDS gel electrophoresis. For mouse tumor samples, 10 mg of tumor tissue was dissected and placed into RIPA lysis buffer (Boston BioProducts) supplemented with one Pierce protease inhibitor tablet and 1X HALT (Thermo Fisher). Tissue was mechanically dissociated initially using a 7-cm pestle (Kimble) followed by further homogenization with a 20-guage needle and brief sonication for 20 s at 20 kHz. Tissue lysate was then centrifuged at 4 °C and >15,000×g for 10 min. The supernatant was collected, diluted in 4× sample buffer to 2×, and resolved using SDS gel electrophoresis. For phos-tag immunoblots, a phos-tag gel was prepared with diluted Phos-Tag™ Acrylamide reagent (Wako Chemicals, AAL-107) according to the manufacturer’s instructions utilizing manganese as the divalent cation. Proteins were then transferred onto PVDF membranes using a wet-tank transfer system (Bio-Rad), blocked for 1 h with TBS-0.1% Tween-20 containing 5% skim milk powder and then probed overnight at 4 °C with primary antibodies diluted in TBS-0.1% Tween-20 containing 1% skim milk powder. Bound antibodies were detected by incubating membranes with horseradish peroxidase-linked, species-specific, secondary antibodies (1:5000, Cell Signaling Technology) for 1 h, followed by 3 × 10 min washing in TBS-0.1% Tween-20, and then the addition of Clarity or Clarity Max ECL blotting substrate (Bio-Rad). Chemiluminescence acquisition was carried out using the Bio-Rad ChemiDox XRS + system and quantitative densitometry was measured using Image Lab (Bio-Rad).

Human nevi and melanoma samples

Human skin tissues with melanocytic nevi and melanoma were retrieved from archived material in the pathology laboratory at UMass Medical Center in compliance with all relevant ethical regulations and were determined to be exempt by the Institutional Review Board at UMass Medical Center (IRB H00007200). The tissue blocks were deidentified before tissue sections (10 µm) were obtained by the authors for IHC. These studies were reviewed in compliance with all relevant ethical regulations and were determined to be exempt by the Boston University School of Medicine Institutional Review Board (IRB H-37967).

RNA isolation and qRT-PCR

Total RNA from cultured cells was isolated using a Quick-RNA kit (Zymo Research). cDNA libraries were generated from RNA using the Superscript III kit and random hexamer primers (Invitrogen). Quantitative real-time PCR was performed using SYBR Green reagents in a StepOnePlus system (Applied Biosystems) according to manufacturer protocol. For each individual experiment, a technical triplicate was run which was then averaged to generate a single biological replicate. Primer sequences were as follows:

CYR61: forward, AGCCTCGCATCCTATACAACC; reverse, TTCTTTCACAAGGCGGCACTC

AMOTL2: forward, TTGGAATCTGCAAATCGCC; reverse, TGCTGTTCGTAGCTCTGAG

GAPDH: forward, GAGTCAACGGATTTGGTCG; reverse, CATTGATGGCAACAATATCCAC

Antibodies

The antibodies used herein categorized by technique and company.

Immunofluorescence (cells) : Santa Cruz Biotechnology: YAP 63.7, 1:250 (detects both YAP/TAZ, sc-101199).

Immunofluorescence (tissue): Cell Signaling Technologies: YAP (1A12) #12395, 1:100; YAP/TAZ (D24E4) #8418, 1:100; YAP (D8H1X) #14074, 1:100. Abcam: gp100 (ab137078), 1:250; SOX10 (Rabbit, ab180862), 1:250; SOX10 (Mouse, ab216020), 1:250.

Immunohistochemistry: Cell Signaling Technologies: YAP/TAZ (D24E4) #8418, 1:100; YAP (D8H1X) #14074, 1:100; GFP (D5.1, cross reacts with YFP) #2956, 1:100; phospho-p44/42 ERK1/2 (Thr202/Tyr204) #9101, 1:100; Abcam: gp100 (ab137078), 1:250; SOX10 (Rabbit, ab180862), 1:250; SOX10 (Mouse, ab216020), 1:250; MelanA (ab210546), 1:500. Dako: S100 (IS504), pre-diluted by the manufacturer.

Immunoblotting : Cell Signaling Technologies: B-Raf (D9T6S), 1:1000; phospho-p44/42 ERK1/2 (Thr202/Tyr204) #9101, 1:1000; p44/42 MAPK (ERK1/2) #9102, 1:1000; RSK1/RSK2/RSK3 (3D27) #9355, 1:1000; phospho-p90RSK (Ser380) (D3H11) #11989, 1:1000; GAPDH (14C10) #2118, 1:1000; YAP (D8H1X) #14074, 1:1000; LATS1 (C66B5) #3477, 1:1000; phospho-LATS1 (Thr1079) (D57D3) #8654, 1:1000; TAZ (E8E9G) #83669, 1:1000; TAZ (V386) #4883, 1:1000; phospho-S6 (Ser235/236) (D57.2.2E) XP #4858, 1:1000; PTEN (138G6) #9559, 1:1000; p53 (1C12) #2524, 1:1000; phospho-Chk1 (Ser345) #2341, 1:500. Abcam: Vinculin (ab18058), 1:4000. Invitrogen: BRAFV600E (RM8 Clone) #MA5-24661, 1:1000-2000. Bethyl Laboratories: LATS1/LATS2 (A300-479A), 1:1000. Santa Cruz Biotechnology: Chk1 (G-4) sc-8408, 1:500; S6 (C-8) sc-74459, 1:1000; N-Ras (F155) sc-31, 1:1000. Spring Biosciences: BRAFV600E (VE1 Clone) # E19290, 1:1000. Cytoskeleton: RhoA (ARH05), 1:500.

Soft-agar assays

In six-well dishes, sterile 2% noble agar stock solution in water was dissolved by heating to ~40–45 °C, mixed with warm media, plated at a final concentration of 0.6%, and allowed to cool and solidify at 4 °C. Following solidification, gels were warmed back to 37 °C. Next, cells were trypsinized, counted, 1 × 104 cells were plated in 0.3% noble agar and allowed to solidify at room temperature or briefly at 4 °C. Plates were maintained in a cell culture incubator for 2–4 weeks with feedings of 1.5 mL of 0.3% agarose solution weekly. All drugs were maintained at 2× concentration in underlays and independent experiments were done in technical triplicate. Total colonies per well were counted using phase-contrast imaging on an Echo Revolve Hybrid Microscopy system at ×10 (Echo Laboratories). For imaging, gels were stained for 20 min with 0.1% Crystal Violet, gently washed multiple times overnight, and imaged on a Chemi-Doc XRS + system under Fast Blast with 0.015-0.1 sec exposure.

EdU assays

Mel-ST cells were seeded on coverslips a day before each collection at a density of 4.5 × 104 cells/well in a 12-well culture dish. Cells were treated with DMSO as a negative control or doxycycline to induce expression of BRAFV600E. EdU was added for a 30-min pulse at a final concentration of 10 µM and cells were then fixed with 4% PFA at different timepoints since the addition of doxycycline. Incorporated EdU was visualized using the Click-iT EdU kit from Invitrogen according to the manufacturer’s protocol (C10337), imaged via fluorescence microscopy on a Nikon Ti-E inverted microscope equipped with a Zyla 4.2 PLUS (Andor) and X-Cite 120 LED light source at the same exposure, and analyzed in ImageJ (version 1.51).

RhoA pull-down activation assay

Active, GTP-bound RhoA was assessed using the RhoA activation assay, bead pull-down format, from Cytoskeleton, Inc (BK036) according to manufacturer instructions. Briefly, dox-inducible BRAFV600E expressing Mel-ST cells were plated at 3 × 104 cells in six-well dishes and allowed to grow for 2–3 days. Prior to reaching 70% confluence, cells were serum-starved and treated ± doxycycline for 24 h, then stimulated with complete media for 6 h prior to ice-cold lysis. Following lysis, input was isolated from total lysate prior to flash freezing with liquid nitrogen and storage at −80 °C. Active RhoA pulldown was performed according to manufacturer instructions utilizing GST-tagged Rhotekin-RBD protein on agarose beads. Samples were then analyzed using immunoblot according to manufacturer protocols.

Drug treatments

The concentrations used for the MAPK pathway inhibitors (MEKi-1/2 and ERKi) were determined experimentally to be the doses at which phosphorylation of ERK1/2 and RSK1/2/3 returned to baseline despite the presence of oncogenic BRAF expression. The reagents used in these studies are as follows: MEKi-1: U0126 (Selleck Chemicals), 10 µM; MEKi-2: Trametinib (GSK1120212), 20 nM (Selleck Chemicals); ERKi: SCH772984, 20 nM (Selleck Chemicals); Hydroxyurea, 1 mM (Selleck Chemicals); doxycycline, 1 µg/mL (Sigma-Aldrich D9891); thymidine, 2.5 mM (Sigma-Aldrich T1895); RO-3306, 7 µM (Sigma-Aldrich SML0569)

Population doubling and viability assays

For population doubling assays, initially 1 × 105 cells (c initial ) were plated in 10-cm dishes. After 4 days of growth cells were trypsinized, counted, and 1 × 105 cells were plated again. Cells were trypsinized and counted again after 4 more days. The number of population doublings (pd) were calculated by inputting the counted number of cells (c final ) into the following equation: \({pd}={{\log }}_{2}(\frac{{c}_{{final}}}{{c}_{{initial}}})\). For viability assays, 4 × 103 cells per well were plated into a white-bottom 96-well dish in technical duplicate to quintuplicate, dependent upon the experiment. Cells were then treated with indicated siRNA and/or drugs for 4 days. The 96-well plate was then allowed to equilibrate at room temperature for 30 min followed by the addition of Cell Titer Glo (Promega) to each well according to the manufacturer's instructions. Viability was then assessed via luminescence using a BMG microplate reader and analyzed using BMG Labtech Optima v2.0R2 software. All technical replicates were averaged to generate a single biologic replicate.

Copy-number analysis

Log 2 copy-number values and/or putative copy-number GISTIC 2.0 values for LATS1 and LATS2 were downloaded from cBioPortal for the TCGA skin cutaneous melanoma (SKCM) subset of the PanCancer Atlas and Memorial Sloan Kettering Cancer Center (MSKCC), NEJM 2014 datasets. SKCM abbreviation represents skin cutaneous melanoma. Mutational status for BRAF and NRAS from the TCGA-SKCM dataset was also obtained from cBioPortal. Graphs were created in Prism 9.

Copy-number analysis by ultra-low-pass whole-genome sequencing

At the experimental endpoint, mice were euthanized and mouse tumors or skin samples were dissected, aliquoted, and immediately flash-frozen in liquid nitrogen and stored at −80 °C. For genomic DNA isolation, a Monarch Genomic DNA purification kit was utilized following the manufacturer’s instructions. In brief, tumor or skin samples on ice were mixed with lysis buffer and proteinase K, vortexed, then incubated at 56 °C in a thermal mixer set to 1400 rpm agitation for 60 minutes. Lysed samples were then centrifuged for 3 min at >12,000×g. The supernatant was then mixed with RNase A, incubated for 5 minutes at 56 °C with full-speed agitation and then bound to the gDNA purification according to manufacturer instructions. gDNA was then eluted and stored at −20 °C until delivery to the Molecular Biology Core Facilities (MBCF) at Dana-Farber Cancer Institute (DFCI).

Library preparation and sequencing

gDNA was fragmented to 200 bp on a Covaris M220 instrument according to the manufacturer’s protocol. Libraries were prepared using Swift S2 Acel reagents on a Beckman Coulter Biomek i7 liquid handling platform from approximately 200 ng of DNA according to the manufacturer’s protocol with 14 cycles of PCR amplification. Finished libraries were quantified by Qubit fluorometer and fragment size distribution was evaluated by Agilent TapeStation 2200. Library pools were further evaluated for quality and balance with shallow sequencing on an Illumina MiSeq. Subsequently, libraries were sequenced targeting a depth of ~1X genome coverage on an Illumina NovaSeq6000 with paired-end 150 bp reads by the Molecular Biology Core Facilities at Dana-Farber Cancer Institute.

Copy-number analysis

Raw sequencing data (FASTQ files) were aligned to human and mouse reference genomes, GRCh37 and MM10 respectively, using BWA MEM (v.0.7.12). Both human and mouse alignment files were sorted by read names using Samtools (version 1.9). Software package Disambiguate (v. 1.0) was applied to sorted alignment files to infer the most likely source of each sequenced read (human or mouse genomes). Disambiguate output (BAM files) that comprised deconvoluted or “cleaned” human reads was used for further analysis. Pre-computed bin annotations for human genome (GRCh37) and bin sizes of 100 Kb, 500 Kb, and 1 Mb were obtained with QDNASeq package (v. 1.28.0), along with GC-content and mappability profiles required for data normalization and bias correction. Normalized copy-ratios were estimated for each deconvoluted alignment file using QDNASeq capabilities. CBS algorithm as implemented in DNAcopy package was used to produce genome segmentation profiles. CNVKit package was used to generate ideogram representations of per-chromosome copy-number profiles.

Gene expression analysis

Publicly available microarray dataset GSE61750 was obtained from the Gene Expression Omnibus (GEO) (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=gse61750). For GSEA, indicated groups were normalized via meandiv, probes collapsed to max probe, and a weighted enrichment metric (log 2 Ratio of Classes) was utilized with 1000 permutations to determine enrichment in indicated gene sets. The YAP/TAZ target score gene set for VAM and GSEA was derived from ref. 36 and ref. 37. Gene set can be found in Supplementary Table 1. The Hippo component gene set was derived from Reactome with additional genes added and can be found in Supplementary Table 2.

Single-cell RNAseq analysis

Count matrix of GSE154679 was downloaded from the GEO repository (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE154679)25. Cells that had less than 200 expressed genes, more than 4000 expressed genes or >17% of mitochondrial gene were removed from analysis. A total of 35,214 cells were retained for downstream analysis. Normalization, dimensionality reduction, cell clustering and data visualization were analyzed with Seurat package81. PCA dimension reduction was performed using top 2000 highest variance genes. The top 15 principal components were utilized to calculate the k-nearest neighbors of each cell. Cell clusters were determined using Louvain algorithm at a resolution = 0.5, which was high enough to obtain clusters associated with cell lineage identity. We used UMAP to visualize and confirm cell clustering. Melanocyte cell population was identified based on the expression of canonical gene markers. To identify subclusters within melanocyte cell populations, we re-analyzed cells within melanocyte cell clusters using the same analysis pipeline described above. To quantify gene sets activity in each cell, gene set testing of scRNA-seq data was performed following VAM pipeline35. Data Normalization and highest variable genes were analyzed using pipeline described above. VAM method was executed using vamForSeurat() function on the melanocyte cell population for each gene set. Graphs were created in Seurat or data was imported into Prism 9 for graph creation.

Murine studies

All animal experiments were conducted according to protocols approved by the Institutional Animal Care and Use Committee (IACUC) at Boston University (Protocol # PROTO201800236). Per Boston University IACUC protocol the maximum allowable mouse tumor size is 20 mm in diameter which was not exceeded. For mouse studies, a mix of male and female mice of the C57BL/6 strain were utilized under standard housing conditions (temperature: 68–79 °F, humidity: 30–70%, light cycles: 12 h on, 12 h off). All other mouse strains were sourced from Jackson Laboratory: Tyr::CreERT2 (Jax # 012328), Lats1f/f (Jax # 024941), Lats2f/f (Jax # 025428), BRAFV600E (Jax # 017837), R26-YFPLSL (Jax # 006148) and were crossed in-house. 4-hydroxytamoxifen (4-HT [Sigma, H7904]) was dissolved in methanol to a concentration of 5 mg/ml. To induce knockout in Lats1/2−/−, and BrafV600E/Lats1/2−/−, mice aged 8–12 weeks had an ~2 cm2 shaved on the right flank using surgical clippers and sufficient 4-HT solution to wet the skin was applied once daily to the shaved area for 3 consecutive days. For BrafV600E mice, a higher concentration of 4-HT was required; 4-HT was dissolved in dimethylsulfoxide (DMSO) to a concentration of 25 mg/ml, and topical administration was performed as above. For the depilation experiments using Lats1/2−/− mice, a 2 cm2 was shaved on both the left and right flank of each mouse, and 5 mg/ml 4-HT was topically applied to each square for 3 consecutive days as above on the right flank, as well as on an approximately 2 cm length of the tail for 3 consecutive days. Two days after the final application, chemical depilation was performed by application of Nair hair remover onto the right flank for 15 s, followed immediately by wiping with a damp tissue to prevent irritation. For systemic knockout experiments, mice were injected with 20 mg/ml tamoxifen (Sigma, T5648) dissolved in corn oil to induce deletion in all Tyrosinase expressing cells. 8–12-week-old mice were injected daily with 100 µl tamoxifen for 5 consecutive days. Tumor development was measured weekly using calipers.

Zebrafish studies

Zebrafish were handled in accordance with protocols approved by the University of Massachusetts Medical School IACUC (Protocol A-2171). Constructs Pmitfa:EGFP:pA and Pmitfa:YAP-5SA:pA were used in the miniCoopR assay as previously described65. Briefly, mitfa(lf) mutant animals were bred, and single-cell stage embryos were injected with 25 pg of a single construct and 25 pg of Tol2 transposase mRNA. Successful Tol2-mediated integration of the construct into the genome rescued the mitfa(lf) phenotype, enabling melanocyte development. Equal numbers of male and female zebrafish were used. Melanocyte rescue was scored at 4–5 days of development, and rescued animals were grown to adulthood and monitored weekly for the presence of melanomas.

Statistical analysis

All quantitative data are presented as mean ± SEM, unless otherwise indicated. The number of samples (n) represents the number of biologic replicates or animals in study, unless otherwise indicated. Prism 9 was used for all statistical analyses and for the creation of most graphs.

RNAi sequences

The siRNA’s used in this study are as follows:

Non-targeting #1 (Control siRNA) (Dharmacon):

UGGUUUACAUGUCGACUAA

LATS1 ON-TARGETplus SMARTpool (Dharmacon):

GGUGAAGUCUGUCUAGCAA; UAGCAUGGAUUUCAGUAAU

GGUAGUUCGUCUAUAUUAU; GAAUGGUACUGGACAAACU

LATS2 ON-TARGETplus SMARTpool (Dharmacon):

GCACGCAUUUUACGAAUUC; ACACUCACCUCGCCCAAUA

AAUCAGAUAUUCCUUGUUG; GAAGUGAACCGGCAAAUGC

YAP1 ON-TARGETplus SMARTpool (Dharmacon)

GCACCUAUCACUCUCGAGA; UGAGAACAAUGACGACCAA

GGUCAGAGAUACUUCUUAA; CCACCAAGCUAGAUAAAGA

WWTR1 ON-TARGETplus SMARTpool (Dharmacon)

CCGCAGGGCUCAUGAGUAU; GGACAAACACCCAUGAACA

AGGAACAAACGUUGACUUA; CCAAAUCUCGUGAUGAAUC.

Reporting summary

Further information on research design is available in the Nature Research Reporting Summary linked to this article.