Project description:The prevailing model of therapeutic resistance in cancer posits that malignant cells acquire genetic mutations in the context of therapeutic pressure that negate the efficacy of the drug. This mutation centric view of therapeutic resistance has recently been challenged by emerging evidence that shows a paucity of genetic diversity in malignant cells that evade therapeutic challenge emphasizing the role of epigenetic evolution in drug resistance. We have previously established that therapeutic resistance emerges in the absence of genetic adaptation from leukemia stem cells and whilst the clinical relevance of these findings has been established, what remains unanswered is whether transcriptional programs that confer resistance to therapies are pre-existing or acquired. It is also unclear how stable this adaptive transcriptional response is and whether it can be reversed. Here, we utilize a unique leukaemia model to highlight the key principles of epigenetic resistance. Using single cell RNA-seq with indexed flow sorting, we demonstrate that a form of Lamarckian evolution, namely transcriptional plasticity, drives epigenetic resistance. With a CRISPR-Cas9 screen we identify reciprocal regulators of the enhancer modification H3K4me1/2, LSD1 and MLL4, as important modulators of the resistant cell state. Knockdown or catalytic inhibition of LSD1 causes reversion to a drug sensitive state that is dependent on the lineage determining pioneer transcription factor Pu.1. Mechanistically, we establish that inhibition of LSD1 results in newly formed enhancers marked by H3K4me1/2, H3K27ac and Pu.1 binding, which facilitates the recruitment of transcriptional co-activators and activation of a differentiation gene expression program. We show that this enhancer-remodeling does not result in a simple linear reversion to the original chromatin state of drug naive cells, rather it reinstates the cells dependence on transcriptional co-activators including Med1 and Brd4 due to the formation of new enhancers with enhancer-promoter loops to canonical Brd4 target genes. Together these findings provide new insights into the molecular mechanisms that facilitate epigenetic resistance to cancer therapies and highlight new opportunities for therapeutic intervention to overcome transcriptional plasticity.

Project description:Clinical responses to kinase inhibitor therapy in acute myeloid leukemia (AML) are limited by the inevitable development of resistance. A major contributor to resistance is early epigenetic adaptation, leading to persistence of a small number of leukemia cells. Here we show that inhibition of the epigenetic regulator lysine-specific deme-thylase 1 (LSD1) augments the response to inhibitors of the FLT3 kinase in AML. We demonstrate that combined FLT3 and LSD1 inhibition results in synergistic cell death of FLT3-mutant AML cells via proliferative arrest and apoptosis. The drug combination synergistically activates a pro-differentiative epigenetic and transcriptional program while simultaneously suppressing the activity of MYC target genes. High resolution multi-modal epigenetic analyses revealed that combined FLT3 and LSD1 resulted in the suppression of MYC-bound promoters and the activation of PU.1-bound enhanc-ers. Forced expression of MYC partially abrogated the drug effect, and regulon en-richment analysis in primary AML samples nominated STAT5 as a putative regulator of MYC gene expression. STAT5 is highly bound to the MYC blood super-enhancer and inhibition of FLT3 results in a loss of STAT5 binding and a loss of super enhancer acti-vation. Furthermore, knockdown of STAT5 augments LSD1-inhibitor induced cell death. LSD1 inhibition also directly represses MYC target genes which show specific accumulation of the repressive LSD1 substrate H3K9me1. We validated these findings in 67 primary AML samples including 19 FLT3-ITD positive AML samples, with the vast majority demonstrating improved responses with the drug combination. High MYC regulon activity was a predictor of response to the drug combination and RNA-seq on drug treated AML samples revealed suppression of MYC target genes. Finally, single cell ATAC seq on primary AML blasts treated ex-vivo with combined FLT3 and LSD1 inhibition results in a shift from MYC super enhancer-high to a MYC super enhancer-low cell state. Collectively, these studies provide preclinical rationale for the investiga-tion of dual FLT3 and LSD1 inhibition in clinical trial.

Dataset Information

Targeting enhancer switching overcomes non-genetic drug resistance in acute myeloid leukaemia [RNA-seq]

Dataset's files

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets