Chromatin accessibility was assayed using an adaptation of the assay for transposase accessible-chromatin (ATAC)-seq (Buenrostro et al., 2013). Briefly, 5 × 106 cells were harvested, washed with PBS and nuclei were isolated using 1 mL HS Lysis buffer (50 mM KCl, 10 mM MgSO4.7H20, 5 mM HEPES, 0.05 % NP40 (IGEPAL CA630)), 1 mM PMSF, 3 mM DTT) for 1 min at room temperature. Nuclei were centrifuged at 1,000 × g for 5 min at 4°C, followed by a total of three washes with ice-cold RSB buffer (10mM NaCl, 10mM Tris (pH 7.4), 3mM MgCl2), to remove as much of contaminating cytoplasmic and mitochondrial material as possible. Nuclei were then counted, and 5×104 nuclei were resuspended in Tn5 reaction buffer (10mM TAPS, 5mM MgCl2, 10% dimethylformamide) and 2 µl of Tn5 transposase (25µM) made in house according to the previously described protocol (Picelli et al., 2014). Nuclei were then incubated for 30 min at 37°C, before isolation and purification of tagmented DNA using QiaQuick MinElute columns (Qiagen). ATAC-seq libraries were prepared by PCR amplification using custom made Illumina barcodes previously described (Buenrostro et al., 2013) and the NEBNext® High-Fidelity 2X PCR Master Mix with 8-10 cycles. Libraries were purified with two rounds of AMPure XP bead cleanup (1.5X beads:sample), followed by quantification by qPCR using SensiMix SYBR (Bioline) and KAPA Library Quantification DNA standards (KAPA Biosystems). ATAC-seq libraries were sequenced on Illumina NextSeq500 using 80bp paired-end reads in biological triplicate.