Transcription Pausing Regulates Hematopoietic Stem Cell Emergence through Fine-Tuning of Signaling Pathways

Regulation of RNA polymerase II (Pol II)-mediated gene transcription is a fundamental mechanism to control cell fate during developmental organogenesis, including the emergence of hematopoietic stem cells (HSCs) in vertebrate embryos. Increasing evidences in recent years have demonstrated that transcriptionally engaged Pol II often pauses shortly after initiation by pausing factors such as DSIF and NELF, and requires positive elongation factors such as P-TEFb for pause release. How this pause-to-elongation transition contributes to HSC development is largely unknown. In a zebrafish spt5 mutant, we detected a dramatic loss of HSCs in early embryos. Spt5 is a key subunit of DSIF, which plays a dual role in both Pol II pausing and elongation. The mutation of the zebrafish spt5 mutant has been reported to specifically disrupt the pausing function of DSIF. Similarly, reduction of HSCs was also detected in embryos lacking the other pausing factor NELF. Consistent with the loss-of-pausing in Spt5- and NELF-deficient embryos, we found that the HSC defect could be rescued by inhibiting Pol II pause-to-elongation transition with either P-TEFb inhibitors or knockdown of the P-TEFb subunit cdk9. Intriguingly, we identified that two essential signaling pathways involved in HSC development, the TGFβ signaling pathway and the JAK-STAT mediated proinflammation pathway, were oppositely regulated by Pol II pausing. In spt5 mutant embryos, the TGFβ signaling pathway is upregulated with an enhanced transcription elongation at several TGFβ signaling regulators, as revealed by Pol II ChIP assays; whereas the transcription of multiple genes involved in proinflammation signaling pathways, especially the genes regulating the interferon signaling, are downregulated. Pol II ChIP and nucleosome mapping studies revealed an increase of nucleosome occupancy at the promoters of proinflammation genes, leading to attenuated transcription initiation. These results suggest that paused Pol II can compete with nucleosomes for occupancy on some promoters to maintain accessible chromatin thus the basal transcription level of these genes. Consistent with the change of signaling activity, the HSC defect in mutant embryos could be effectively rescued by TGFβ inhibitors or by activation of JAK-STAT, the downstream effector of IFNγ signaling. Thus Pol II pausing affects HSC emergence through differentially regulating the transcription of distinct signaling regulators: while they negatively regulate the HSC inhibitory signaling (such as TGFβ) by preventing premature transcription elongation, they are also required to maintain the basal transcription of HSC-promoting signaling (such as the inflammation signaling) by preventing promoter nucleosome reassembly. In summary, our study suggests that Pol II pausing/elongation regulators and developmental signals converge at the paused Pol II to regulate HSC emergence in vertebrate embryos.