Excessive R-Loops Trigger an Inflammatory Cascade Leading to Aberrant Hematopoietic Stem and Progenitor Cell Expansion

Mutations in DEAD-box Helicase 41 (DDX41) are observed in patients with myelodysplastic syndromes (MDS) with inferior overall survival, implicating this factor in disease pathogenesis. DDX41 is an understudied factor with links to inflammation and RNA processing, but whose function in hematopoiesis is unknown. Using a novel zebrafish model of Ddx41-deficiency, we unveiled a critical role for Ddx41 in regulating hematopoietic stem and progenitor cells (HSPCs). Mutants for ddx41 have more HSPCs than their siblings as measured by in situ hybridization for the HSPC marker runx1 as well as quantification of cd41:gfp⁺ and runx1:mcherry⁺ HSPCs by flow cytometry. To uncover potential mechanisms driving HSPC expansion in ddx41 mutants, we performed RNA-seq analysis of purified HSPCs from ddx41 mutants and sibling controls. Expression of inflammatory target genes was significantly elevated in ddx41 mutants including those regulated by the transcription factor NFκB (nuclear factor kappa-light-chain-enhancer of activated B cells). As inflammatory signaling and NFκB signaling in particular is elevated in MDS, we next explored Ddx41 regulation of NFκB signaling. Using an in vivonfkb:gfp reporter, we demonstrated that NFκB signaling was indeed elevated in ddx41 mutants. DDX41 is implicated in the STING (STimulator of INterferon Genes) and TBK1 (TANK binding kinase 1) pathway that mediates type I interferon production in response to foreign or damaged endogenous DNAs. By lowering pathway activity via knockdown of sting or pharmacological inhibition of TBK1, we determined that the STING/TBK1 pathway triggered excessive NFκB activation in ddx41 mutants. Moreover, the elevated inflammatory signaling was detrimental to normal HSPC homeostasis as lowering STING/TBK1 pathway activation significantly reduced HSPC expansion in ddx41 mutants, as measured by runx1in situ hybridization and cd41:gfp flow cytometric quantification. R-loops, nucleic acid structures consisting of RNA:DNA hybrids and displaced ssDNAs, can activate the STING/TBK1 pathway via the nucleic acid sensor cGAS (cyclic GMP-AMP synthase). Recent proteomic analysis showed that DDX41 can bind R-loops. Via immunofluorescence quantification of RNA:DNA hybrid levels in ddx41 mutants versus sibling controls, we revealed that Ddx41 acts as a suppressor of R-loop accumulation. Using a RNASEH1-GFP transgenic zebrafish line that permits inducible depletion of R-loops, we revealed that excess R-loops in ddx41 mutants promote HSPC expansion as measured by runx1in situ hybridization and runx1:mcherry flow cytometric quantification. Moreover, using RT-qPCR analysis of NFκB target genes, we determined that R-loop depletion dampened NFκB activation in ddx41 mutants, suggesting the R-loop-mediated effects on HSPC numbers was via inflammatory signaling. We deciphered cGAS activity was critical for this effect as both nfkb:gfp reporter activity and HSPC levels in ddx41 mutants were diminished following treatment with a pharmacological inhibitor of cGAS. Our data demonstrate that DDX41 insufficiency triggers an R-loop-mediated inflammatory cascade leading to aberrant HSPC expansion and is the first study to delineate a functional consequence for R-loops in HSPC biology. Furthermore, recent studies showed R-loop accumulation in other common forms of MDS suggesting our novel in vivo findings are more broadly applicable to the MDS pathogenesis.