Abstract
Stimulation of hematopoietic stem and progenitor cells (HSPCs) with the inflammatory mediator Prostaglandin E2 (PGE2) enhances self-renewal and stem cell engraftment following transplantation. Currently, the long-acting derivative of PGE2, 16,16-dimethyl-PGE2 (dmPGE2) is in its fourth clinical trial to improve HSC engraftment and reduce graft versus host disease. To understand the effect of dmPGE2, we assessed genome-wide chromatin reorganization and gene expression changes in human CD34+ HSPCs after 2 hours of dmPGE2 treatment, the time period of treatment in the clinical trials. Enhancers are known to regulate gene expression changes in specific environmental contexts such as stress or inflammation, however the regulatory principles by which subsets of enhancers become activated are poorly understood. Here, we mapped active enhancers by ChIP-seq for H2K27ac and found that dmPGE2 activates a discrete set of enhancers in HSPCs. To investigate enhancer chromatin remodeling, we performed micrococcal nuclease digestion followed by high-throughput sequencing (MNase-seq) to map the occupancy and position of nucleosomes. We found that, contrary to the predominant assumption that open chromatin structures are essentially nucleosome-free, MNase-accessible nucleosomes are retained at inducible enhancers following dmPGE2 stimulation. Through ATAC-seq analysis we mapped changes in open chromatin and found that induced enhancers gain chromatin accessibility following stimulation while maintaining their nucleosome configuration. Surprisingly, this indicates that nucleosomes present at the center of dmPGE2-responsive enhancers play an important function in enhancer accessibility and activity. We then correlated enhancers with gene expression changes by performing RNA-seq and found that genes associated with dmPGE2-induced enhancers display higher gene expression changes after stimulation compared to genes associated with non-responsive enhancers. Transcripts upregulated after dmPGE2 treatment include previously identified regulators of self-renewal and migration such as NR4A2, EGR1 and CXCR4. Moreover, inflammatory chemokines including CXCL2 and CXCL8 as well as members of the activating protein 1 (AP-1) transcription factor gene family such as FOS, FOSL2 and JUNB are increasingly expressed upon stimulation. The gene expression profile included a signature implying CREB as the main transcription factor responsible for the acute dmPGE2 response. Western blot revealed dmPGE2-mediated activation of the signaling transcription factor CREB through phosphorylation in HSPCs. Using ChIP-seq, we found increased genomic binding of phospho-CREB (pCREB) after dmPGE2 treatment in the enhancers. Surprisingly, the binding of pCREB coincided directly with variant histone H2A.Z containing labile nucleosomes in enhancers. We validated the interaction between pCREB and H2A.Z on chromatin in dmPGE2-responsive U937 cells through chromatin fractionation followed by complex immunoprecipitation. This suggests that labile nucleosomes provide sufficient DNA access to allow for binding of pCREB at enhancers. Taken together, our study proposes a novel model for stimulus-mediated activation of enhancers by the inflammatory mediator dmPGE2. dmPGE2 induces the phosphorylation of CREB and subsequently leads to a specific interaction of pCREB with previously deposited H2A.Z-rich nucleosomes at inducible enhancers who regulate genes that promote HSPC fate. This new mechanism of variant histone deposition followed by the interaction with a signaling transcription factor at enhancers supports a rapid inducible response from the environment.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.