KLF1 Acts As a Pioneer Transcription Factor Via SMARCA4 to Open Chromatin and Facilitate Redeployment of an Enhancer Complex Containing GATA1 and SCL

Abstract

Introduction: Sequence specific transcription factors (TFs) bind to short DNA motifs in vitro; but, only a tiny fraction of possible binding sites within the genome are bound. Factors that influence in vivo binding are relative affinity for the specific motif at the site, co-operative interactions with additional TFs that bind to adjacent motifs, and physical impediments provided by chromatin. Pioneer TFs are a special class of TF that can interact with 'silent' chromatin, increasing local accessibility. This can facilitate the binding of other TFs that do not possess pioneering activity (settler TFs). KLF1 is an essential erythroid-specific TF that binds to ~1000 enhancers, often in co-operation with other essential TFs such as GATA1 and SCL/TAL1. Aim: To determine whether KLF1 acts as a pioneer TF and, if so, address the mechanism. Metods: We employed the erythroid cell line, K1ER1, to study the immediate effects of KLF1 on chromatin re-organization following addition of 4-OHT. This erythroid cell line is null for Klf1 and stably expresses an inducible KLF1-ERTM transgene. We performed ATAC-seq, and ChIP-seq for KLF1, GATA1, TAL1, NF-E2 and SMARCA4 in the presence or absence of 4-OHT. We generated a deletion series of KLF1 in stably expressing K1ER clones to map the domains responsible for chromatin remodeling and we examined the neo-morphic DNA-binding mutants, nan and CDA-IV, to determine whether ectopic binding led to ectopic pioneering. We validated the cell line results by ATAC-seq and ChIP-seq in Klf1-/- versus wild type fetal liver erythroid cells. We employed to SMARCA4 inhibitor, BRM014, to test its requirement for erythroid gene expression. Results: Using ATAC-seq and ChIP-seq we show KLF1 binding induces an increase in local chromatin accessibility that facilitates assembly of a complex of settler TFs including GATA1 and SCL/TAL1, but not p45-NF-E2. Interestingly, KLF1 rapidly induces relocation of GATA1 in local neighborhoods, including GATA1 'desertion' of certain sites. The pioneering function occurs at ~300 key erythroid enhancers and super-enhancers such R2 at -26kb in the a-globin LCR, an important E2f2 intronic enhancer, and dematin/Epb4.9 gene enhancers. We confirmed these findings in primary erythroid cells from Klf1-/- fetal liver. We show neo-morphic mutations in the DNA-binding domain of KLF1 (Nan and CDA-IV mutations) lead to ectopic pioneering in some but not all cases of ectopic binding. We show a N-terminal domain of KLF1 that is distinct from the transactivation domains (TADs) is responsible for SMARCA4/Brg1 recruitment (Fig 1), the likely effector of chromatin remodeling. Inhibition of SMARCA4 using specific inhibitors2 leads to inhibition of induction of some KLF1 target genes such as dematin/Ebp4.9, but not others such as beta-globin. Conclusions: KLF1 is a pioneer TF that recruits GATA1, TAL1 and likely an enhancer complex that includes LMO2 and LDB1 to key erythroid enhancers and super enhancers. KLF1 can induce GATA1 to abandon certain bound sites within some genome neighborhoods but not globally. Pioneering occurs in part by recruitment of SMARCA4 via an N-terminal domain that is distinct from the TADs, and this is essential for expression of subset of KLF1 target genes. This work has implications for how KLF1 works to direct erythroid versus lineage trajectories from bi-potential progenitors, and how KLFs reprogram cells generally.