Plasmids
pLU-EF1aL-rtTA3-iCherry was obtained from Meenhard Herlyn from the Wistar Institute in Philadelphia, USA. TetO-Ascl1-puro was a gift from Marius Wernig (Addgene plasmid # 97329; http://n2t.net/addgene:97329; RRID: Addgene_97329) [11]. The ORFs of murine Brn2, Myt1L and NeuroD1 were cloned into our own doxycycline-inducible lentiviral vector system by GenScript (Hong Kong).
Cell culture
All parental melanoma cell lines (HT144 (RRID:CVCL_0318), A375 (RRID:CVCL_0132), SKMel103 (RRID:CVCL_6069) and MA1[12]) were cultivated in DMEM (Gibco, Life Technologies) supplemented with 10% FCS (Biochrom), 1% penicillin (100 units/mL, Sigma-Aldrich) and streptomycin (100 µg/mL, Sigma-Aldrich) (pen/strep), 1% (v/v) of non-essential amino acid solution (NEAA,10 mM, Sigma-Aldrich) and 0.1% (v/v) β-mercaptoethanol (Gibco, Life Technologies). All melanoma cell lines were cultured in a humidified incubator with 5% CO2 at 37°C. Cells were passaged using a 21 mM trypsin solution (Sigma-Aldrich) when reaching 70–90% confluency. If not otherwise specified, transdifferentiated cells were also cultured in the same medium as the parental cells.
Production of lentiviral particles and transduction
HEK293T cells (RRID:CVCL_0063) were used for lentiviral particle production. X-tremeGENE® transfection reagent (Roche) was used according to the manufacturer ‘s instructions to transfect the HEK cells. 1 mL of filtered supernatant containing viral particles was used for transducing melanoma cells in one well of a 6-well plate. 48h after transduction, the cells were washed with PBS and fresh medium without virus.
Generation of subclones harboring all four neuron-specific expression constructs
HT144, A375, SKMel103 and MA1 melanoma cells were transduced with pLU-EF1aL-rtTA3-iCherry. The cells were then selected for mCherry fluorescence using cell sorter BD FACSAria from the Flow Cytometry Core Facility of the DKFZ. Successfully transduced cells were hereafter termed Melanoma-M2. Then the cells were sequentially transduced with the expression constructs for the factors BRN2, MYT1L and NEUROD1 and afterwards treated with 3 different antibiotics (blasticidin, hygromycin and G418) in order to select for cells transduced with the respective constructs. The selected cells were transduced with an expression construct coding for ASCL1 and after induction with doxycycline (1 µg/ml) subjected to selection with puromycin. Melanoma-M2 cells infected with and selected for all four transcription factors were named Melanoma-4F (Fig. 1A).
Transdifferentiation of melanoma cells
The melanoma cells that were successfully selected for all four factors were used for the transdifferentiation experiments. The cells were grown in DMEM supplemented with 10% FCS, 1% pen/strep, 1% (v/v) NEAA, 0.1% (v/v) β-mercaptoethanol and doxycycline (1 µg/ml). Doxycycline induced the expression of the transdifferentiation factors ASCL1, BRN2, MYT1L and NEUROD1. The medium was changed every 2–3 days. After a minimum of 4 days, cells were used for functional studies or molecular analyses.
Quantitative real-time PCR
Total RNA was extracted using the RNeasy Mini Kit (QIAGEN) according to the manufacturer’s instructions. NanoDrop ND-1000 spectrophotometer was used to measure the quality and quantity of RNA samples. 500 ng of RNA were used to synthesize cDNA with the RevertAid First Strand cDNA Synthesis Kit (Thermo Fisher Scientific) according to the manufacturer’s instructions. The synthesized cDNA was diluted 1:10 prior to qRT-PCR analysis. qRT-PCR was performed using the ABI® 7500 Real-Time PCR system and the QuantiNova SYBR Green Kit (Qiagen). The relative expression of each gene was normalized to 18S expression. Gene expression was analyzed using the ΔΔCt method. After 7 days of treatment with doxycycline, M2 and 4F cells were harvested and analyzed for the expression of melanocytic and neuronal markers. Moreover, their proliferative and metastatic capacity were examined.
Western blot
Proteins were extracted from M2 and 4F cells after 7 days of doxycycline treatment, using RIPA buffer (Invitrogen) containing complete mini protease inhibitor cocktail (Roche). The protein concentration was measured using the Pierce BCA Protein Assay kit. The following primary antibody dilutions were used in 5% bovine serum albumin (BSA) in TBST: ASCL1 (a kind gift from Dr. Moritz Mall from the DKFZ, 1:100), BRN2 (Thermo Fisher Scientific Cat# MA5-35032, RRID:AB_2848937, 1:1000), MYT1L (a kind gift from Dr. Moritz Mall from the DKFZ, 1:1000), NEUROD1 (Thermo Fisher Scientific Cat# MA5-32626, RRID:AB_2809903, 1:1000), S100 (Abcam Cat# ab109252, RRID:AB_10862910, 1:1000), MITF (Thermo Fisher Scientific Cat# MA5-32554, RRID:AB_2809831, 1:1000), KI67 (Invitrogen, MA5-14520, RRID AB_10979488, 1:1000), TRP2 (Santa Cruz Biotechnology Cat# sc-74439, RRID:AB_1130818, 1:1000), Melan-A (Agilent Cat# M7196, RRID:AB_2335691, 1:1000) and α-actinin (Santa Cruz Biotechnology Cat# sc-17829, RRID:AB_626633, 1:5000). The secondary antibodies (Cell Signaling Technology Cat# 7076, RRID:AB_330924 and Cell Signaling Technology Cat# 7074, RRID:AB_2099233) were diluted 1:10.000 in 5% BSA in TBST and incubated at RT for 1h.
Light and fluorescent microscopy
To evaluate morphological alterations of 4F cells, M2 and 4F cells were seeded into 10cm culture dishes and treated with doxycycline for more than 7 days. On the fourth day, 4F cells were treated with palbociclib in order to enrich neuron-like cells. On days 4, 7 and 14, pictures were taken with a DM LS Light Microscope (Leica). Fluorescent images were acquired using the Zeiss LSM 700 from the Light Microscopy Core Facility at the DKFZ. The number of fluorescent cells (positive staining for neuron markers) or the number of granules per cell (gH2AX immunostaining detection) were counted using FIJI, with the script developed by the Light Microscopy Core Facility at the DKFZ.
Immunofluorescence for neuron markers
2x104 cells were seeded in 8-chamber culture slides (Falcon). After 7 days of doxycycline treatment, M2 and 4F cells were fixed in 4% paraformaldehyde (PFA) for 5 min on ice and an additional 10 min at RT and then permeabilized with 0.1% Triton X-100 for 10 min. Next, cells were preincubated for 1h with blocking solution (3% BSA) before administering the primary antibodies anti-b-tubulin III (Sigma-Aldrich Cat# T8578, RRID:AB_1841228) and anti-MAP2 (Sigma-Aldrich Cat# HPA012828, RRID:AB_1853946), respectively, and incubating overnight at 4°C. The following day, cells were washed and incubated with goat anti-rabbit IgG (H + L) Highly Cross-Adsorbed Secondary Antibody, Alexa Fluor Plus 647 (Thermo Fisher Scientific Cat# A32733, RRID:AB_2633282) and goat anti-mouse IgG (H + L), Superclonal™ Recombinant Secondary Antibody, Alexa Fluor™ 488 (Thermo Fisher Scientific, A28175) for 1h at RT in the dark. Nuclei were stained with 4, 6-diamidino-2-phenylindole (DAPI, Roche) for 15 min. Slide mounting was performed with fluorescence mounting medium (Dako).
Cell proliferation assay
Cell viability was analyzed using the BrdU Cell Proliferation ELISA Kit (Abcam, ab126556). 5x103 melanoma cells were plated in flat bottom 96-well plates (Gibco Life Technologies). BrdU was added 8h before measurement. 24, 48, 72, and 96h after seeding of the cells, they were fixed using Fixing Solution. The fixed plates were stored at 4°C and tested within 2 weeks. The detector antibody, peroxidase goat anti-mouse IgG Conjugate, TMB Peroxidase Substrate and Stop Solution were added subsequently according to the manufacturer’s instructions. The plates were tested using Infinite M1000Pro Microplate Reader with an excitation wavelength of 450 nm and an emission wavelength of 590 nm. Cell Proliferation status was calculated from the resulting change in fluorescence intensity normalized to the value of the original melanoma cell group.
Colony formation assay
Parental cells, M2 cells treated with doxycycline for 4d (M2-Dox (+)4d), untreated 4F cells (4F-Dox (-)), 4F cells treated with doxycycline for 4d (4F-Dox (+)4d), and 4F cells treated with doxycycline for 4d and palbociclib for 3d (4F-Dox (+)4d-Pal (+)3d) cells were seeded at a density of 50–150 cells per well in a 6-well plate. Doxycycline was added to M2-Dox (+)4d, 4F-Dox (+)4d, and 4F-Dox (+)4d-Pal (+)3d cells, but no palbociclib was added. Old medium was removed and replaced with fresh medium every 3 days. After 5–10 days of induction, the M2-Dox (+), 4F-Dox (+), and 4F-Dox (+)-Pal (+) cells were given normal MEF medium without doxycycline. After 7 to 16 days, cell colonies were stained using crystal violet solution (0.5%). Relative colony area density was determined using ImageJ software (Fiji, RRID:SCR_003070).
Migration Assay
Parental cells, M2-Dox (+)4d, 4F-Dox (-), 4F-Dox (+)4d, and 4F-Dox (+)4d-Pal (+)3d cells were cultured at a density of 2x104 to 3x104 cells per chamber of an ibidi 2-well culture chamber (Gräfelfing, Germany) for 24h. After removing the insert, cell migration was monitored every 3h. Images were acquired 20-24h after removal of the insert. Acquired images were analyzed using TScratch software.
Invasion Assay
The Cultrex BME 24-well kit (3455–024-K, R&D systems, Minneapolis, Minnesota) was used for cell invasion assays with A375 and MA1. Cells were starved overnight and then seeded out (3x104 cells) in a BME-coated chamber with serum-free medium. Then, the number of invasive cells was measured by calcein staining. Due to the discontinuation of the production of the above kit, the Cell Invasion Assay (Basement Membrane), 96-well, 8 µm (Abcam, ab235697) was used for the SKMel103 and HT144 cell lines according to the manufacturer’s instructions. With both kits, we included parental melanoma cells, M2-Dox (+)4d, 4F-Dox (-), 4F-Dox (+)4d, and 4F-Dox (+)4d-Pal (+)3d cell groups.
Cell Apoptosis Assay
Cells were seeded in 6-well plates at a density of approximately 60%. The following day, M2-Dox (+)4d and 4F-Dox (+)4d cells were treated once with 3Gy of ionizing radiation. Untreated samples served as negative control. After 96 hours of incubation, adherent cells and cells in the supernatant were collected in tubes followed by washing with ice-cold PBS. Next, 1x105 cells were suspended in 1X Annexin V Binding buffer (100 µL), and 2.5 µL FITC Annexin V and 1 µL of DAPI was added to the cell suspension. Cell solution was gently vortexed and incubated at RT for 20 minutes avoiding light exposure. Then, 400 µL of 1X Annexin V Binding buffer were added to each tube and samples were analyzed by flow cytometry in the Flow Cytometry Core Facility at the DKFZ. FlowJo cell analysis software (RRID:SCR_008520) was used to analyze the data.
gH2AX immunostaining
1-1.5x105 M2 and 4F cells were seeded in 12-well plates on cover slips. After doxycycline treatment for 4d, M2 and 4F cells were subjected once to 3Gy of ionizing radiation. The untreated sample served as negative control. After 0.5, 1, 6, and 24h, cells were fixed in 4% paraformaldehyde (PFA) for 15 min at RT and then permeabilized with 0.2% Triton X-100 for 8 min. Next, cells were preincubated for 1h with blocking solution (5% BSA in 0.05% Triton PBS) before incubation with the primary antibody anti-phospho-Histone H2AX (Ser139) (Millipore Cat# 05-636, RRID:AB_309864) overnight at 4°C. The following day, cells were washed and incubated with goat anti-mouse IgG (H + L), Superclonal™ Recombinant Secondary Antibody, Alexa Fluor™ 488 (Thermo Fisher Scientific, A28175) for 1h at RT in the dark. Nuclei were stained with 4, 6-diamidino-2-phenylindole (DAPI, Roche) for 15 min. Slide mounting was performed with fluorescence mounting medium (Dako).
Mouse experiments
Tumorigenicity and metastatic potential of transdifferentiated HT144-4F and A375-4F were determined in vivo by injecting 2-5x106 cells subcutaneously into the flanks of NOD scid gamma (NSG) mice. The expression of the transdifferentiation factors was maintained by adding doxycycline to the drinking water (2 mg/ml, Sigma-Aldrich, Cat. No. D9891). The tumor size was measured every day. When the size reached 1.5 cm in diameter, mice were euthanized. For the in vivo metastasis assay, 1.5-5×106 cells were injected into the tail veins of NSG mice. 6 weeks later, the mice were sacrificed and lung and liver were collected. Lung and liver samples were fixed and the metastases were counted.
Immunohistochemistry
Hematoxylin-eosin (HE) staining was done for each paraffin sample of subcutaneous tumors, livers or lungs. For immunohistochemistry, 4µm-thick sections of the sample were stained with target antibodies S100 (Agilent Cat# Z0311, RRID:AB_10013383), KI67 (Abcam Cat# ab16667, RRID:AB_302459), MelanA (Novus Cat# NBP1-30151, RRID:AB_1987285) according to the manufacturers' instructions. Simultaneously, a negative control without the first antibody and a verified positive control were stained to avoid false positive or negative results.
RNA Sequence profiling
Total RNA from three independent experiments was extracted using RNeasy mini kit (QIAGEN) according to the manufacturer’s protocol. NanoDrop ND-1000 spectrophotometer was used to determine the quality and quantity of the RNA. The extracted RNAs were sent to BGI Company (Hong Kong) for sequencing. They constructed the Strand Specific RNA library (Eukaryote, oligo dT, mRNA enrichment), sequenced using DNBSEQ platform (Paired End 100bp sequencing, 20M clean read pairs (4G data) per sample), and filtered the raw data with adapter sequences or low-quality sequences. This filter step was completed by SOAPnuke software (RRID:SCR_015025) [13] In Galaxy platform [14], reads from each sample replicate were mapped to the hg19 human genome using HISAT2. Then, uniquely mapped reads were processed with featureCounts and DESeq2 (RRID:SCR_000154) for differential gene expression analysis [15–17]. IPA software was used for the generation of graphs depicting differentially expressed genes.
DNA Methylation Assay
DNA samples were extracted from M2-Dox (+)7d and 4F-Dox (+)7d cell lines with QIAGEN DNeasy Blood & Tissue Kit, and sample quality was assessed by NanoDrop ND-1000 spectrophotometer. The microarray unit of the DKFZ Genomics core facility conducted gene expression profiling using Illumina Infinium MethylationEPIC BeadChip. DNA methylation processing, including QC, probe filtering, normalization, dimensionality reduction analysis and differentially methylated probes analysis, have been performed using ChaMP package (version 2.18.3) in R environment (version 4.0.1). CNV analysis was performed using minfi (v1.34.0) and conumee (v 1.9.0–1) packages.
Statistical Analysis
Experiments were conducted at least in biological triplicates. Data were displayed as mean ± SD and unpaired t test was used for analysis. Except sequencing data, all statistical analyses were performed with GraphPad Software (RRID:SCR_002798).