Bone marrow failure (BMF) in Fanconi Anemia (FA) results from exhaustion of hematopoietic stem cells (HSC), but the physiological role of FA proteins in HSC pool integrity remains unknown. Herein we demonstrate that FANCD2, a core component of the FA pathway, counters replication stress during developmental HSC expansion in the fetal liver (FL). Rapid rates of proliferation and FANCD2 deficiency result in excess RPA-coated ssDNA, and provoke pChk1 activation and Cdkn1a(p21) nuclear localization in fetal Fancd2-/- HSC. Checkpoint mediated S-phase delays induced by Cdkn1a(p21) are rescued by Tgf-β inhibition, but pChk1 activation is further aggravated. Our observations reveal the mechanism and physiological context by which FANCD2 safeguards HSC pool formation during development.

Bone marrow (BM) failure is the principal source of morbidity and mortality in Fanconi Anemia (FA) patients. Recessively inherited germline mutations in one of 25 genes lead to deficits by in a pathway central to DNA crosslink repair. Functionally, FA proteins protect adult hematopoietic stem cells (HSC) from p53 mediated apoptosis elicited by alkylating agents, a range of experimental inflammatory cues or aldehyde exposure. However, these mechanisms do not seem to account for depleted hematopoietic stem and progenitor cell pools in very young FA patients, or the spontaneous, non-apoptotic and p53-independent fetal HSC deficits observed in murine models. Building on our previous observation of a quantitatively constrained fetal HSC pool in FA mice (Fancd2-/-), the current experiments reveal the specific developmental timeframe for the onset of stem cell deficits during HSC expansion in the fetal liver (FL). Cell cycle studies using an EdU/BrdU pulse chase protocol reveal delays in S-phase entry and progression at E13.5. Building on the role of FA proteins (FANCM, FANCI and FANCD2) in countering experimental replication stress (RS) in cell line models, we reasoned that rapid rates of proliferation required during expansion in the FL may similarly confer RS on the FA HSC pool. Experiments in E13.5 FL HSC confirmed the predicted increase in single stranded DNA and accumulation of nuclear replication associated protein (pRpa), along with activation of pChk1, a critical cell cycle checkpoint in cells under RS. For comparison, pChk1 in unperturbed adult cells was no different between Fancd2-/- and WT. The data are also consistent with gains in RAD51 and alkaline comet assays we previously published (Yoon et al., Stem Cell Reports 2016). The cell cycle regulator Cdkn1a (p21) is considered a canonical downstream component of the p53 response in adult FA HSC, but it also performs p53 independent functions in the RS response that coincide with its role in the nucleus. Here, we observed an increase in nuclear localization of p21 in Fancd2-/- FL HSC. TGF-β is a critical developmental morphogen that regulates p21 activity, and TGF-β inhibitors can partially reverse adult FA HSC function along with suppression of NHEJ mediated DNA repair. To test regulation of p21 in fetal HSC under RS, we first treated WT FL HSC with aphidicolin to experimentally simulate RS and found that SD208, a small molecule TGF-β-R1 inhibitor, completely rescued the p21 nuclear localization. We then went on to demonstrate that pharmacological inhibition of TGF-β signaling also reversed the nuclear p21 translocation in FA FL HSC (under physiological RS) and functionally rescued the primitive myeloid progenitor colony formation (CFU-GEMM) in vitro. Altogether, our data show that HSC deficits in FA first emerge in the fetal liver, where rapid fetal expansion causes RS that elicits pChk1 activation and nuclear p21 translocation, which restrain cell cycle progression and act as principal mechanisms limiting fetal HSC pool size in FA. Our experiments suggest a central and p53-independent role for p21 in fetal FA HSC regulation. Detailed knowledge of the physiological role of FA proteins in fetal phenotype HSC has the potential to lead to new therapies that delay or rescue hematopoietic failure in FA patients. Disclosures No relevant conflicts of interest to declare.

Compartmental bone marrow (BM) inflammation has been linked to acute myeloid leukemia (AML) progression and hematopoietic dysfunction, yet the underlying mechanisms remain unclear. MicroRNAs (miRs) are capable of broadly deregulating cellular gene expression programs through simultaneous, dose-dependent action on a panel of target genes. MicroRNA-155, a widely known proinflammatory miRNA is overexpressed in AML patients, in particular those with Flt3-ITD. We recently observed that miR-155 is highly abundant in AML patient plasma-derived extracellular vesicles (AML-EV) and identified inflammation as one of the most upregulated gene ontology (GO) categories in a proteomic screen of hematopoietic stem and progenitor cells (HSPCs) exposed to AML-EV. Here, we investigate whether AML-derived miR-155 coordinately regulates inflammatory signaling in the leukemic niche. To better model compartmental inflammation in AML, we first established a highly penetrant, immunocompetent, fully congenic system. Using the monoblastic AML cell line C1498 (CD45.2+), we were able to reliably graft animals without myeloablation, thereby avoiding unintended changes in BM niche function. Across multiple experimental cohorts, we then analyzed the BM plasma secretome and observed dynamic changes in the level of several inflammatory chemokines and cytokines. At sacrifice, flow cytometric analysis of the residual HSPC (CD45.1+) showed concurrent upregulation of key inflammatory response genes (ISG15, STAT1, CXCL10) and evidence of replicative stress (γ-H2AX) in residual HSPC. Consistent with potential trafficking of miR-155 via AML-EV, we next confirmed the enrichment of miR-155 in C1498-EVs, observed gains in miR-155 in BM plasma EVs and demonstrated increased levels in HSPCs following EV uptake. To determine if miR-155 could function as a regulatory hub under inflammatory conditions we utilized four separate target prediction databases (miRDB, TargetScan, miRWalk, and miRanda) to generate miR-155 target gene set. With an initial list of 360 potential gene targets that met a minimum threshold prediction score, we queried six publicly available RNA-Seq datasets to focus on genes belonging to inflammation or DNA damage repair GO groups highly expressed in HSPCs. This yielded a final set of 9 candidate targets (Wee1, E2f2, Rps6ka3, Rps6ka5, Csflr, Mgme1, Rad51, Rad51b, Ercc1). Reasoning that miR-155 target sets are involved in niche inflammation and genotoxicity in AML, these computationally predicted targets will undergo testing and validation for involvement in inflammatory and genotoxic stress. Chronic inflammation in the AML niche contributes to disease persistence and failure of residual hematopoiesis. Our study will determine the role of miR-155 as a regulatory hub for compartmental inflammatory signaling, and simultaneously serve as a potential opportunity for novel therapeutic target discovery in leukemic BM. Disclosures No relevant conflicts of interest to declare.