P1329: regional compartmentalization of the bone marrow stroma is important for responses to inflammation and cancer

Topic: 23. Hematopoiesis, stem cells and microenvironment Background: Hematopoietic stem and progenitor cells (HSPC) reside in the bone marrow (BM), where their activity is regulated by interactions with stromal cells. Previous studies reveal that the BM niche is partitioned into two major compartments, with activated hematopoietic stem cells and myeloid progenitors primarily found in the central marrow in association with sinusoidal endothelial cells (SEC) and leptin receptor (LepR)+ mesenchymal stromal cells (MSC), and deeply quiescent HSCs and lymphoid progenitors enriched in the endosteal niche composed of Sca-1+ MSCs, osteoprogenitors (OPr), and arterial endothelial cells (AEC). Methods are lacking to profile and prospectively isolate stromal populations present in each location, and to investigate how spatial compartmentalization of the niche affects HSPC responses to inflammation or cancer. Aims: We sought to develop a method to profile the two major BM compartments in the same mice by flow cytometry, permitting systematic evaluation and prospective isolation of stromal populations. We then asked how inflammatory and neoplastic challenges remodel different compartments of the BM niche to affect HSPC activity. Methods: We optimized methods to isolate cells from each BM compartment by enzymatic digestion of a marrow plug from a single femur to obtain central marrow populations, or bone chips from 8 crushed bones to obtain endosteal cells. We used key surface markers to design a flow cytometric gating scheme to identify all major stromal populations and isolated these cells for functional assays and molecular profiling by single cell RNA sequencing (scRNAseq). We induced inflammation by injection of the toll-like receptor (TLR)3 ligand poly(I:C) and evaluated the effect of cancer by injecting MLL-AF9 blasts to model acute myeloid leukemia. Results: We successfully developed a method to profile stromal cells in both BM compartments, defining 6 populations: LepR+ MSC-L and SEC in central marrow, and Sca-1+ MSC-S, multipotent OPr (mOPr), and committed OPr in the endosteum. Using functional assays ex vivo and plate-based scRNAseq, we inferred that MSC-L generate adipocytes and contribute to osteogenesis, whereas MSC-S and downstream OPrs produce osteoblasts, fibroblasts, and chondroblasts at the endosteum. We validated this annotation extensively against previously published scRNAseq profiles of the murine BM niche and with fluorescent reporter mice. While infiltration of AML blasts caused global expansion of all types of stromal cells in both BM compartments, exposure to poly(I:C) caused a more selective effect in the central marrow, inducing MSC-L to adopt an inflammatory state marked by surface expression of Sca-1. Using scRNAseq, we showed that these inflammatory (i)MSC-L downregulated genes related to stromal differentiation and HSC trophic factor production and upregulated inflammatory chemokine genes, which may contribute to the HSPC response to poly(I:C). We also performed motif enrichment among differentially expressed genes to identify transcriptional drivers of the iMSC-L state, finding that iMSC-L activated an IRF7-driven inflammatory program that superseded their Pten-mediated homeostatic functions. Summary/Conclusion: Our data establish a novel method to investigate BM niche regionalization and show that different perturbations exert selective effects in specific marrow locations. In particular, we identify MSC-L as a first responder in poly(I:C)-driven inflammation, resulting in local overproduction of chemokines in the central marrow. We are currently investigating whether this phenomenon is shared in other inflammatory contexts and evaluating its importance for hematopoietic responses to inflammation. Keywords: Mesenchymal cells, AML, Inflammation, Bone marrow stroma