Genomic Determinants of Hematopoietic Progenitor Sensitivity to Innate Immune Signaling

Abstract

Innate immune processes impact diverse hematopoietic stem and progenitor cell (HSPC) functions, yet how innate immune signaling networks are sensed by HSPC genomes and how HSPC genomic activity alters cell responsiveness to innate immune regulators are not fully understood. We demonstrated that E14.5 primary (-77 -/-) and immortalized (hi-77 -/-) GATA2-deficient fetal progenitors from Gata2 -77 enhancer deletion mice upregulate Interferon-gamma (IFNγ) and Toll-like receptor (TLR) signaling components (Johnson et al. JEM 2020) and are hypersensitive to IFNγ and TLR signaling (Tran et al. iScience 2023). Combinatorial signaling involving both pathways regulates genes that are not regulated by individual pathways and amplifies responses beyond that induced by single pathways. Innate immune-activated genes harbor motifs for signal-dependent transcription factors, including the ETS factor PU.1, and GATA2 deficiency elevates PU.1 activity to upregulate B-lineage and myeloid genes (Jung et al. JCI 2023). We hypothesize that PU.1 links innate immune signaling networks and HSPC genome control. In new multi-omic studies, we asked if PU.1 is required to sense exclusively IFNγ or TLR signaling, both pathways independently or combinatorially, or is not essential. We used hi-77 -/- progenitors lacking a PU.1 enhancer ( Spi1 URE-/-) with ~2-fold lower PU.1 vs. control hi-77 -/- progenitors. RNA-seq was conducted with hi-77 -/- and hi-77 -/-; Spi1 URE-/- progenitors, with or without IFNγ, TLR1/2 agonist Pam 3CSK 4, or both agents for 4 h. hi-77 -/-; Spi1 URE-/- progenitors lost the responsiveness of 27 TLR-activated genes and retained the responsiveness of 6 TLR-activated genes. 80% of IFNγ-activated genes (115 of 144) retained IFNγ-responsiveness when PU.1 was reduced. In hi-77 -/-; Spi1 URE-/- progenitors, 68.2% (148 of 217 genes) retained their responsiveness to combinatorial signaling. These studies revealed the disproportionate importance of PU.1 for TLR- vs. IFNγ-mediated transcriptional control and a PU.1 requirement for combinatorial signaling at the majority of loci. Elevated IFNγ and TLR1/2 signaling in GATA2-deficient fetal progenitors is associated with increased monocytic and reduced granulocytic progenitors (Johnson et al. JEM 2020; Blood Adv 2022). Ablation of Irf8, encoding an IFNγ pathway component, partially normalizes the granulocytic progenitor deficit (Johnson et al. Blood Adv. 2022). Since GATA2 deficiency creates crosstalk between IFNγ and TLR1/2 pathways to elevate cytokine/chemokine production (Tran et al . iScience 2023), we asked if attenuating TLR signaling reverses the differentiation defect. As MYD88 Innate Immune Signal Transduction Adaptor (MYD88) is crucial for TLR1/2 signaling, we generated -77 +/-; Myd88 -/- mice. Timed matings were conducted to obtain -77 +/+, -77 -/-, Myd88 -/-, and -77 -/-; Myd88 -/- E14.5 embryos. Myd88 ablation in -77 -/- embryos did not reverse the skewed granulocytic vs. monocytic progenitor ratio. To determine if MYD88 loss impacts -77 -/- progenitor function, we analyzed CMP and GMP populations from E14.5 embryos with CFU assays. In comparison with -77 +/+, -77 -/- CMP and Myd88 -/- CMP produced fewer CFU-G colonies (1.9-fold decrease, P = 0.0097 and 2.0-fold decrease, P = 0.007, respectively). Surprisingly, -77 -/-; Myd88 -/- CMP exhibited a further reduction in CFU-G colonies (5.2-fold decrease, P < 0.0001). The CFU-G colonies produced by -77 +/-; Myd88 -/- CMP resembled that of -77 +/+ CMP. Thus, GATA2 and MYD88 synergistically promote CFU-G production from CMP, and a single -77 allele maintains CMP activity to produce CFU-G without Myd88. Myd88 deletion did not affect the reduced CFU-G nor the increased CFU-M with -77 -/- GMP. We are testing if TLR1/2 signaling is not required to maintain balanced fetal liver progenitor populations or if another signaling adapter nullifies the impact of MYD88 loss. Our studies revealed a mechanism underlying the hypersensitivity of GATA2-deficient fetal progenitors to innate immune signaling, and MYD88 loss further decreases the reduced CFU-G generated from -77 -/- CMP. We are testing the hypothesis that GATA2 restricts assembly of PU.1-containing chromatin complexes and analyzing models to unveil functional ramifications of elevated innate immune signaling in single HSPCs.