Truseq chip library preparation kit

Chromatin immunoprecipitation-sequencing was performed on pooled samples. Briefly, mammary tissues were dissected into small pieces and crosslinked using formaldehyde (1% w/v, methanol-free) followed by homogenization and chromatin preparation using a Covaris sonicator (Covaris, Woburn, MA, United States). Chromatin was immunoprecipitated using an appropriate antibody, washed, reverse-crosslinked, and DNA isolated using the phenol-chloroform method. Precipitated DNA was QC’ed by quantitative PCR using Mouse ChIP Control qPCR Primer Sets (Active Motiff, Carlsbad, CA United States). For ChIP-seq, 10 ng precipitated DNA was used to prepare the library using Illumina TruSeq ChIP library preparation kit following the manufacturer recommendations. Libraries were quantified for cluster generation using KAPA Library Quantification Kit (Kapa Biosystems, Wilmington, MA, United States). Sequencing was performed using Illumina HiSeq2500 using 50-bp single-end rapid run format. Sequencing quality was initially checked by running FastQC. Sequencing reads were aligned against mm9 by using Bowtie2. BAM files were normalized by depth after removal of PCR duplicates and blacklisted regions. Peaks were called using MACS2 with default parameters. For heatmap and metagene plots, ngs.plot program was used. UCSC genome browser was used to visualize individual gene tracks.

For Chromatin Isolation by RNA Purification (ChIRP), we followed a previously described protocol^{35 (link)}. Briefly, 20 million cells were harvested and fixed in 1% glutaraldehyde solution (Sigma-Aldrich, G5882) for each reaction. ChIRP was performed using biotinylated oligo probes designed against mouse MaTAR25 using the ChIRP probe designer (Biosearch Technologies). Independent even and odd probe pools were used to ensure lncRNA-specific retrieval (refer to Supplementary Data 8 for odd and even sequences targeting MaTAR25, and probes targeting mouse Ppib transcripts which were used as negative controls). ChIRP-Seq libraries were constructed using the Illumina TruSeq ChIP Library Preparation Kit (Illumina, IP-202-1012). Sequencing libraries were barcoded using TruSeq adapters and sequenced on Illumina NextSeq instruments.

To obtain samples for the ChiP-seq experiments biological duplicates of primary B cells expressing either C/EBPα^WT ER or C/EBPα^R35A ER were induced for 3 hr with E2 and 100,000 cells per sample processed by ChIPmentation, as described (Schmidl et al., 2015 (link)) (see also: https://www.nature.com/articles/nmeth.3542). Three µL of anti-C/EBPa antibody (Santa Cruz sc-61 14AA) or anti-PU1 (Santa Cruz sc-352 T21) were used per immunoprecipitation. Tagmentation of immobilized C/EBPa or PU1-enriched chromatin was performed for 5 min at 37 degrees in a 25 µL transposition reaction mix. Libraries were prepared with an Illumina TruSeq ChIP Library Preparation Kit and after quality control on Bioanalyzer sequenced (1 × 50 bp) on an Illumina HiSeq2500 instrument.
ChIP-seq data were processed similarly to ATAC-seq data (refer to ‘determination of differentially accessible ATAC peaksanalysis of ATAC-seq data’ section) with the following modifications: bamCoverage was used with: --binSize 1 --normalizeUsing RPKM --effectiveGenomeSize 215057000. Peak calling for each experiment was conducted using macs2 callpeak and the options -f BAMPE -g mm.

Coding sequences for HA-tagged HYH, GBF1, GBF2 and GBF3 were amplified by PCR from full-length cDNA clones using gene-specific primers listed in Table S3 and inserted into pEU-His vectors using appropriate restriction enzymes creating pEU-His-HYH-HA, pEU-His-GBF1-HA, pEU-His-GBF2-HA and pEU-His-GBF3-HA. A series of gDB-seq procedures including protein synthesis, immunoprecipitation and sequencing by MiSeq (Illumina, Inc., San Diego, CA, USA), mapping of sequenced reads onto the genome by Bowtie 2 [33 (link)] and peak detection by MACS2 (q-value < 0.01) [34 (link)] have been described previously [6 (link)]. For constructing the gDB-seq library, a KAPA Hyper Prep Kit (Roche, Basel, Switzerland) was used instead of a TruSeq ChIP Library Preparation Kit (Illumina, Inc., San Diego, CA, USA). Binding motif sequences were predicted from gDB-seq data using the GADEM program [35 (link)]. JBrowse was used for visualization of the peak position of gDB-seq [36 (link)]. Promoters were defined as regions from 500 nt upstream to 200 nt downstream of the 5′ ends of genes.

ChIP experiments were performed following manufacture instructions using iDeal ChIP-seq kit for transcription factors and for histones (Diagenode) with respectively a rabbit polyclonal anti-FLI1 antibody (Ab15289, Abcam) and a rabbit polyclonal anti-H3K27ac (Ab4729, Abcam). The Ewing sarcoma A673 cell line was obtained from the American Type Culture Collection (ATCC) and the Ewing sarcoma TC-71 cell line was obtained from the German Collection of Microorganisms and Cell Cultures (DSMZ). STR-profiling proved each cell line matched with the reference profile provided by ATCC and DSMZ, respectively; and cells were routinely tested for mycoplasma contamination by PCR. Briefly, the EWS cell lines were fixed for 10 min with 1% of methanol-free formaldehyde (28908, Thermo-Scientific). Chromatin was sonicated (Bioruptor, Diagenode) for 20 cycles (30-sec on, 30-sec off) set at position “high” to generate DNA fragments with an average size around 150–300pb. For ChIP sequencing, the libraries were generated using TruSeq ChIP library preparation kit (Illumina) and sequenced on Illumina HiSeq 2500 (single end, 100 bp). Reads were aligned to hg19 reference genome with Bowtie2^{42 (link)}. Peaks were called with MACS2^{43 (link)} with the option narrow for FLI1 ChIP-seq and broad for H3K27ac ChIP-seq. For each cell line, ChIP-seq results were normalized according to their input sample.