Cell lines and culture. The mouse ASPS cell line ASPS17 was established from the tumors induced in embryonic mesenchymal cells expressing ASPSCR1-TFE3 as previously described17. The human ASPS cell lines, ASPS-KY and ASPS-1, are described previously44,45. Mouse pericytes were purified from the dpc 17 mouse embryo mesenchyme by sorting the PDGFRB+, CD105+, and CD31- fraction, and the cells were immortalized by introduction of the SV40 large T antigen. All cell lines were grown in DMEM supplemented with 10% FBS.
Allograft transplantation studies. Tumor cell transplantation experiment was performed by injecting 5 x 106 ASPS cells mixed in Matrigel (Corning) into the subcutaneous regions of 10- to 12-weeks old Balb/c nude mice as described previously17. JQ1 and THZ1 were intraperitoneally administrated 50 mg/kg daily for two weeks. Mice were carefully observed daily, and were euthanized 17 days after transplantation (n = 10 per each group). All animal experiments described in this study were performed in strict accordance with standard ethical guidelines and were approved by the animal care committee at the Japanese Foundation for Cancer Research under licenses 10-05-9 and 0604-3-13.
Immunoblotting. Cells were lysed in RIPA buffer and precleaned by centrifugation at 10,000g for 10 min at 4°C. Protein concentrations were measured by the DC protein assay (Bio-Rad). Equal amout of protein lysates were boiled for 5 min in sample buffer (0.5 M dithiothreitol, 25% glycerol, 2% SDS, 60 mM Tris-HCl pH 6.8 and bromophenol blue. The samples were separated by SDS-PAGE on SuperSep gels (Fuji Film) and transferred onto nitrocellulose membrane (Amersham). Immunoblots were probed with primary antibodies in 5% skim milk overnight at 4°C and respective secondary antibodies for 1 h at room temperature. Primary antibodies used were ASPSCR1 (Sigma-Aldrich, HPA026749), Cas9 (Novus Biologicals, NBP2-36440), FLAG-tag (Sigma-Aldrich, F3165), Myc-tag (Santa Cruz Biotechnology, sc-40), GFP (Merck Millipore, MAB3580), and mCherry (Cell Signaling, 43590).
Immunofluorescence. Mouse ASPS cells were fixed with 4% paraformaldehyde and were subjected to immunofluorescence using the specific antibodies and the respective fluorochrome-labeled secondary antibodies. Immunofluorescent images were photographed with a Zeiss LSM 770 laser scanning microscope with a 60x objective (Zeiss) and LSM Software ZEN 2009 (Zeiss). Captured images were analyzed using ImageJ. The mean distance of Vwf from the nucleus was calculated in 10 different cells for each condition per experiment by tracing lines (6 in each cell) from the nuclear membrane to the most distal fluorescent foci. For evaluation of 3D-culture in the microfluidic device, fluorescence and immunofluorescence images were captured using Olympus IX71 and Olympus FV3000, and recorded every other day and quantified as described in the figure legends. Antibodies used were VWF (DAKO, N1505), FLAG-tag (Sigma-Aldrich, F7425), and goat anti-rabbit IgG conjugated with RRX (Jackson Immunoresearch, 111-295-144).
Quantitative RT-PCR (qRT-PCR). Total RNA extraction, reverse transcription and RNA quantification were performed by standard methods. Conventional RT-PCR and real-time quantitative RT-PCR (Q-RT-PCR) were performed with a Gene Amp 9700 thermal cycler (Applied Biosystems) and a 7500 Fast Real-Time PCR System (Applied Biosystems), respectively. The sequences of the oligonucleotide primers are shown in Supplementary Table 11.
Histopathology and immunohistochemistry. For light microscopic analysis, tumor tissues were fixed with 3% formaldehyde, paraffin embedded, and stained with hematoxylin and eosin (H&E) using standard techniques. Antibodies used were mouse CD31 (Cell Signaling, 77699), PDGFRB (R & D Systems, BAF1042), RAB27A (Cell Signaling, 69295), SYTL2 (Santa Cruz Biotechnology, sc393847), VWF (Santa Cruz Biotechnology, sc365712), human CD31 (Abcam, ab28346), and FLAG-tag (Sigma-Aldrich, F3165). Heat-mediated antigen retrieval was performed in Tris-EDTA buffer at pH 6.0. Immunohistochemical staining was performed using the Simple Stain MAX-PO kit (Nichirei Bioscience), the Histofine SAB-PO (R) kit (Nichirei Bioscience).
Gene expression profiling. ASPS cell pellets were processed to extract total RNeasy Mini Kit (Qiagen). RNA quality was assessed using the Bioanalyzer 2100 (Agilent). The murine Genome HT MG-430 PM Array and the human HG-U133 + PM Array (Affymetrix, Santa Clara, CA, USA) were hybridized with aRNA probes generated from the total RNA samples. After staining with streptavidin-phycoerythrin conjugates, arrays were scanned using an Affymetrix GeneAtlas Scanner.
Microarray data processing and analysis. The data were analyzed with Microarray Suite version 5.0 (MAS 5.0) using Affymetrix default analysis settings and global scaling as normalization method. The trimmed mean target intensity of each array was arbitrarily set to 100. Microarray data were analyzed using GeneSpring GX 12.6 (Agilent). Gene pathway analysis was performed using Gene Set Enrichment Analysis (GSEA) was performed using GSEA-P 2.0 software46, and Ingenuity Pathway Analysis (Qiagen).
Chromatin immunoprecipitation (ChIP)-Sequencing. ChIP-Seq was performed using the method previous described with modifications47. A total of 5 x 106 ASPS cells per immunoprecipitation were cross-linked with 1% formaldehyde for ten minutes at room temperature. Chromatin was sheared in sodium dodecyl sulfate (SDS) lysis buffer containing 1% SDS, 10 mM EDTA, and 50 mM Tris pH 8.0 to an average size of 400 to 500 bp using a Covaris S220 sonicator for 15 min. ChIP was performed with 5 μg anti-histone H3K27ac (Active Motif, 39133), anti-H3K4me3 (Abcam, ab8580), anti-H3K27me3 (Millipore, 07-449), anti-FLAG (Sigma-Aldrich, F7425), anti-ASPSCR1 (Sigma-Aldrich, A026749), anti-BRD4 (Bethyl Laboratories, A301-985A100) antibodies. The antibody-bound protein/DNA complexes were immunoprecipitated using protein G magnetic beads. Immunoprecipitated DNA was then purified and subjected to secondary sonication to an average size 150 to 350 bp. Libraries were prepared according to instructions accompanying the ThruPLEX DNA-Seq kit (Rubicon Genomics). The ChIP DNA was end modified and adapters were ligated. DNA was PCR amplified with Illumina primers and Illumina-compatible indexes were added. The library fragments of approximately 300-500 bp were band-isolated from an agarose gel. The purified DNA was sequenced on an Illumina MiSeq next-generation sequencer following the manufacturer protocols.
ChIP-seq data analysis. Base calls were performed using Bowtie 2 (http://bowtie-bio.sourceforge.net/bowtie2/index.shtml). ChIP-Seq reads were aligned to the mm9 (https://www.ncbi.nlm.nih.gov/assembly/GCF_000001635.18) or hg19 (https://www.ncbi.nlm.nih.gov/assembly/GCF_000001405.13/) genome assembly using samtools 1.2 (http://www.htslib.org). Peak calling was performed using MACS1.4 (http://liulab.dfci.harvard.edu/MACS). Peak distribution was calculated by Cistrome (http://cistrome.org/ap/root). Neighbor genes on enriched genomic regions were determined using by Nucleus (https://rias.rhelixa.com). The genomic distributions of DNA-binding peaks were visualized by NGSplot (https://anaconda.org/bioconda/r-ngsplot). DNA-binding of each ChIP-seq data was visualized using IGV_2.3.80 (http://software.broadinstitute.org/software/igv). The de novo motif enrichment was performed using HOMER v 4.11.1 (http://homer.ucsd.edu/homer/motif). Super-enhancers were identified using the method previously described with the ROSE program (http://younglab.wi.mit.edu/super_enhancer_code.html). Gene ontology analysis for nearby genes on super-enhancers were performed by GREAT version 4.0.4 (http://great.stanford.edu/public/html/).
Epigenomic CRISPR/dCas9 screening. CRISPR/Cas9-based epigenomic screening was performed according to the method described with modifications48,49. Total of 494 SEs and active enhancers in which H3K27ac signals were reduced more than 40% by ASPSCR1-TFE3 loss were selected, and average 12.8 gRNAs per each enhancer were designed to target these enhancers using CRISPR direct (https://crispr.dbcls.jp) and GuideScan (http://www.guidescan.com) (Supplementary Table S8). 397 gRNAs for enhancers without ASPSCR1-TFE3 binding and 1,000 non-target gRNAs (https://www.addgene.org/pooled-library/zhang-mouse-gecko-v2/) were designed as negative controls. Gibson overhangs were fused to sense and antisense nucleotides corresponding to each gRNA. The gRNA library was synthesized by CustomeArray and inserted into pLV-U6-gRNA-UbC-DsRed-P2A-Bsr (Addgene) using Gibson Assembly (New England BioLabs). A high titer gRNA library was constructed in Endura ElectroCompetent cells (Lucigen), and 293FT cells were transfected with the purified library DNA, psPAX2 and pCMV-VSV-G. ASPS17/dCas9-KRAB cells were generated by infecting Lenti-Ef1a-dCas9-KRAB-Puro lentivirus (Addgene) and the cells were transduced with lentivirus bearing the gRNA library at 100x fold coverage. DsRed-positive cells were sorted by FacsAria II and selected using 2 µg/ml of Blasticidin four days after infection. Then, 1x106 infected cells were transplanted subcutaneously to nude mice 10 days after infection. Subcutaneous tumors were removed a month after transplantation and genomic DNA was extracted, PCR amplified, and subjected to target sequencing using screening primers listed in Supplementary Table 12. Sequencing reads were compared between pre-tranplanted and post-transplanted ASPS 17 cells using DESeq2, and enhancer loci that contain more than 50% of gRNAs reduced <0.5 with and adjusted p-value <0.05 were defined as enhancers required for in vivo ASPS growth.
Hi-C. The Hi-C libraries were constructed using an Arima-HiC Kit (Arima Genomics) according to the manufacturer’s instructions for Mammalian Cell Lines (A160134 v00) and Library Preparation (A160137 v00). In brief, one million cells were collected and crosslinked with 37% formaldehyde solution. DNA isolated from the crosslinked cells was digested with two restriction enzymes (^GATC and G^ANTC). After the incorporation of biotinylated nucleotides at the digested DNA ends, both ends were ligated with the spatially proximal ends. The ligated DNA was sheared into 200-600 bp fragments using the Focused-ultrasonicator M220 (Covaris) and the ligation junctions were enriched with streptavidin magnetic beads. The sequencing libraries from enriched DNA fragments were prepared with a TruSeq DNA PCR-Free Library Prep Kit (Illumina). The resulting libraries were amplified with 10 PCR cycles and purified with SPRI beads. The quality and concentration of the paired-end libraries were evaluated using the Qubit 4 Fluorometer (Thermo Fisher Scientific), the 2100 Bioanalyzer system (Agilent Technologies), and the 7900HT Fast Real-Time PCR System (Thermo Fisher Scientific). The final libraries were sequenced on the Illumina NovaSeq 6000 sequencer with a read length of 150 bp. Reads were processed with Juicer pipeline, using reference of hg38 for human samples and mm10 for mouse samples, yielding .hic files50. These files were further used for Juicebox visualization and downstream analysis, including finding loops and contact domains using Juicer tools. Loops were called by HICCUPS and those located in Chr1-22 (human), 1-19 (mouse), and X.
CRISPR/Cas9-mediated gene editing. Lentivirus plasmids containing short guided RNA (sgRNA) for Ccbe1, Pdgfb, Rab27a, Syngr1, Sytl2, and Vwf were introduced into the lentiCRISPRv2-puro vector (Addgene). The sgRNA sequences are listed in Supplementary Table 12. Knockdown efficiencies were confirmed by western blotting and/or RT-PCR.
ELISA. For the ELISA, 5 x 106 mouse ASPS cells were seeded and cultured overnight. The culture media were collected and immediately analyzed using the mouse Gpnmb (Abcam), Vwf (Abcam), and Pdgfb (Proteintech) ELISA kits according to manufacturers’ instructions.
Fluorescence recovery after photobleaching. Trafficking of Pdgfb, Gpnmb, and Angptl2 was evaluated by FRAP following previously described procedures51. ASPS17 and null cells were transfected with with plasmids containing, DsRed-tagged Pdgfb, Gpnmb, or Angptl2. RFP-tagged actin (Invitrogen) was used as a negative control of trafficking. FRAP images were acquired with an LSM880 confocal microscope equipped with a live cell chamber (set at 37°C) and ZEN software (Zeiss) with a 40X objective. Cells were excited with a 561 nm laser and the emission between 566 and 689 nm recorded. Images were acquired with 12 bits image depth and 512 × 512 resolution using a pixel dwell of ~1.52 μs. At least three (n ≥ 3) pre-bleaching images were collected and then the region of interest was bleached with 100% of laser power. The recovery of fluorescence was traced at every 5 min for 1 hr. For drug inhibition experiments, 10 µM of Nexinhib 20 (Tocris) was added into culture media 3 hr before bleaching. Fluorescence recovery was calculated as previously described51.
Fabrication of microfluidic device. The micropatterned master with 100 µm in height features was fabricated based on standard photolithography. SU-8 3050 (MicroChem) was coated and patterned on a 4 inch silicon wafer by exposing ultraviolet (UV) light to the top of a photomask on which the desired 3-channel patterns are printed. A mixture of PDMS (Sylgard 184, Dow-Corning) and its curing agent was mixed to a 10:1 weight ratio, poured onto the patterned master to a 5-mm thickness, degassed to remove air bubbles for 1-2 h, and cured in an oven at 80 °C for 3-4 h. The cured PDMS was removed from the wafer. The inlets (diameter: 2 mm), outlets (diameter: 6 mm), and center hole (diameter: 1 mm) for media filling and spheroid injection were punched using biopsy punches unless otherwise noted. After cleaning with adhesive tape, the PDMS device and glass coverslip were permanently bonded using oxygen plasma treatment (40 s, 50 sccm, 40 mW). Furthermore, the microfluidic devices were then stored in a 35mm dish and cured in an oven at 80 °C for more than 2 h. After sterilization under UV for 1-2 h, the microfluidic devices were ready for cell seeding.
ASPS spheroid formation. The prewarmed DMEM was added to the dish, and the ASPS cell suspensions were transferred into 15 mL tubes and centrifuged at 1000 rpm for 3 min. The cells were resuspended at desired density. Cell suspension was then added in a prime surface 96-well plate with U-shaped bottom well (Sumitomo Bakelite), which significantly caused self-aggregation of cells, namely a spheroid. Monoculture core spheroids were prepared with a density of 5.0 × 104 cells/mL in the IMDM medium for two days. Co-culture core-shell spheroids were initiated by core spheroid formation. After two days, the suspension culture of core spheroids was then replaced by pericytes suspension at a density of 7.5 × 104 cells/mL in the IMDM medium for shell formation for another day. Finally, the spheroids were introduced into a microfluidic device.
In vitro angiogenesis. A spheroid was transferred into fibrin-collagen gel (2.5 mg/mL fibrinogen) (Sigma-Aldrich), 0.15 U/mL aprotinin (Sigma-Aldrich), 0.2 mg/mL collagen type I (Corning), and thrombin 0.5 U/mL (Sigma-Aldrich) in D-PBS. The spheroid suspended in gel solution was then injected through the center hole into channel 2 without leakage into channels 1 and 3 and incubate at a 37 °C CO2 incubator for 15 min for gelation. Channels 1 and 3 were filled with EGM-2 and incubated overnight at the incubator to eliminate the bubbles at the boundary of gel and medium. For HUVEC adhesion at the gel interface, the HUVECs (5.0 × 106 cells/mL in the EGM-2) were injected into channel 1. With a 90° tilt device at a 37 °C CO2 incubator for 15 min, HUVECs adhered to the gel surface in the microfluidic device. This process was repeated for channel 3. After spheroid injection and HUVEC adhesion, the inlets and outlets of channels 1 and 3 were filled with EGM-2 and kept at 37 °C and 5% CO2 in an incubator. For culturing cells on the device, EGM-2 was replaced daily unless otherwise noted.
Human sarcoma specimens. Alveolar soft part sarcoma, synovial sarcoma, Ewing sarcoma, myxoid liposarcoma, dermatofibrosarcoma protuberans and solitary fibrous tumor surgical specimens were obtained from The Cancer Institute Hospital. Informed consent was obtained from donors, and the study was approved by Institutional Review Board at the Japanese Foundation for Cancer Research under license 2013-1155.
Statistical analysis. All data are representative results from at least three independent experiments unless otherwise specified in the figure legends. The mean ± SD of individual experiments is shown. Student t-test and one-way ANOVA statistical method were used.
Material availability. Plasmids generated in this study are available upon request (T.N.)
Data availability
Microarray and ChIP-seq data are accessible through the NCBI Gene Expression Omnibus (GEO) database (http://www.ncbi.nlm.nih.gov/geo), with the accession number GSE186789 and GSE189163, respectively. Hi-C data are available in the DDBJ Sequenced Read Archive under the accession numbers DRR330423 and DRR330424.
Code availability
We did not use any custom code or mathematical algorithm that is deemed central to the conclusions. All software and packages used are listed in the Reporting Summary and are publicly available.
References
44. Kamijyo, A & Shinoda, K. Establishment of human alveolar soft sarcoma cellline ASPS-KY. J. Jpn. Orthop. Assoc. 79 (2005).
45. Kenney, S. et al. ASPS-1 a novel cell line manifesting key features of alveolar soft part sarcoma. J. Pediatr. Hematol. Oncol.33, 360-368 (2011).
46. Subramanian, A., Kuehn, H., Gould, J., Tamayo, P. & Mesirov, J. P. GSEA-P: a desktop application for gene set enrichment analysis. Bioinformatics23, 3251-3253 (2007).
47. Shimizu, R. et al. EWS-FLI1 regulates a transcriptional program in cooperation with Foxq1 in mouse Ewing sarcoma. Cancer Sci. 109, 2907-2918 (2018).
48. Klann, T. S. et al. CRISPR-Cas9 epigenome editing enables high-throughput screening for functional regulatory elements in the human genome. Nat. Biotechnol.35, 561-568 (2017).
49. Klann, T. S., Crawford, G. E., Reddy, T. E. & Gersbach, C. A. Screening regulatory element function with CRISPR/Cas9-based epigenome editing. Methods Mol. Biol.1767, 447-480 (2018).
50. Durand, N. C. et al. Juicebox provides a visualization system for Hi-C contact maps with unlimited zoom. Cell Syst.3, 99-101 (2016).
51. Rademacher, D. J., Cabe, M. & Bakowska, J. C. Fluorescence recovery after photobleaching of yellow fluorescent protein tagged p62 in aggresome-like induced structures. J. Vis. Exp.145, e59288 (2019).