The phenylalanine-and-glycine repeats of NUP98 oncofusion form condensates that selectively partition transcriptional coactivators

Cancer-associated genetic aberrations, such as the nucleoporin 98 (NUP98) gene rearrangement detected in human leukemias, often produce condensates, a type of membrane-less biomolecular assemblies. How exactly the cancer-related condensation contributes to oncogenesis remains far from clear. Here, we investigate NUP98-PHF23, a leukemia-causing chimeric protein that fuses NUP98's sequence enriched in phenylalanine-and-glycine repeats (FG repeats, also known as intrinsically disordered region [IDR]) in-frame with PHF23's PHD finger, a domain that specifically 'reads' and binds H3K4 trimethylation (H3K4me3). Our integrated analyses using protein module mutagenesis, cell imaging, genomic profiling and condensation reconstitution collectively demonstrate a multifaced role for NUP98's FG repeats in driving fusion condensation while recruiting and co-mixing with a set of histone modifiers and chromatin remodelers, notably the MLL family of H3K4 methylation-'writing' enzymes. The H3K4me3-'reading' PHD finger and NUP98's IDR are cooperative in mediating efficient targeting of NUP98-PHF23 and its coactivators onto leukemic genes, leading to active transcription. Together, we show that NUP98-PHF23 coordinates a set of homotypic and heterotypic interactions (IDR:IDR, IDR:coactivator and PHD:histone) to organize formation of the chromatin-bound multi-component condensates, wherein a feedforward loop involving the 'reading' and 'writing' of H3K4 methylation acts to enforce an open chromatin landscape at leukemogenic loci, thereby driving oncogenic transformation. Overall design: In order to dissect requirement of NUP98-PHF23's protein modules for its biological functions, we generated two function-disruptive mutants — one containing the substitution of Phe in NUP98's FG repeats with Ala (the “FA” mutant), which was known to disrupt FG repeats-mediated LLPS, and the other containing a W362A mutation at PHF23's PHD finger, which was reported to abolish the H3K4me3 binding and suppresses leukemogenic transformation. Next, we generated HEK293FT cells with stable expression of NUP98-PHF23 (GFP-FLAG tagged), either wild-type (WT) or mutant. Cells were fixed in 1% formaldehyde (Thermo Scientific, 28908) for 10min, followed by quenching with 125mM glycine for 5min. Cells were then washed twice with cold PBS added with protease inhibitors (Sigma-Aldrich, 4693132001), and then subjected to resuspension and incubation in LB1 buffer (50 mM HEPES-KOH pH 7.5, 140mM NaCl, 1mM EDTA, 10% glycerol, 0.5% NP-40, 0.25% Triton X-100). The cell nuclei were collected for sonication using Bioruptor sonicator (Diagenode, B01020001; at high-energy setting for 45 cycles with 30s on and 30s off). Supernatant was collected after centrifugation (20,000g for 10min at 4?°C) for incubation with the dynabeads (Invitrogen, 11204D) that are pre-bound with antibodies for 8h at 4?°C. After a series of wash with LB1 buffer, LB2 buffer (10 mM Tris-HCl pH 8.0, 200mM NaCl, 1mM EDTA, 0.5mM EGTA), and LB3 buffer (10 mM Tris-HCl pH 8.0, 100mM NaCl, 1 mM EDTA, 0.5mM EGTA, 0.1% sodium deoxycholate, 0.5% N-lauroylsarcosine), the chromatin–protein complexes bound to beads were eluted, subject to reverse crosslink overnight at 65?°C, and treated with RNase (Roche, 11119915001; 1 h at 37?°C) and then protease K (Roche, 03115828001; 2h at 55?°C). The final DNA sample, as well as 1% of input chromatin, was recovered using PCR purification kit (Qiagen, 28106). The ChIP–seq library was prepared using NEBNext Ultra II kit (NEB, E7645L) following the manufacturer's instructions. ChIP–seq libraries were sequenced on the Nextseq 550 system using Nextseq 550 High Output Kit v2.5 (Illumina, 20024906). Drosophila spike-in chromatin for normalization was used for NUP98-PHF23 ChIP-seq as previously described(Active Motif spike-in ChIP–seq reagents, 53083 and 61686).