The bone marrow microenvironment (BMME) is indispensable for support of normal and malignant hematopoiesis. Mesenchymal stem/stromal cells (MSCs) give rise to multiple non-hematopoietic components that play a pivotal role in myelodysplastic syndromes (MDS). However, therapeutic targeting of MSCs has been challenging, in part due to the limited definition of the heterogeneity of MSC populations from human bone marrow. Using bone marrow from normal healthy controls obtained through aspirates rather than surgical specimens (normal bone marrow; NBM), we successfully identified two novel distinctive subsets of human MSCs by combining three MSC cell surface markers: CD271/ Nerve Growth Factor Receptor (NGFR), CD106/ Vascular Cell Adhesion Molecule-1 (VCAM-1) and CD146/ Melanoma Cell Adhesion Molecule (MCAM). We first isolated three BM-derived non-endothelial lineage (CD45/CD235ab/CD34-CD31-) cell populations from NBM: CD271+/CD146- (NGFR+/Lineage- cells: NLCs), CD271+/CD146+/CD106+ (NGFR+/VCAM1+/MCAM+/Lineage- NVML cells), and CD271+/CD146+/CD106- (NGFR+/MCAM+/Lineage-; NML cells). Gene profiling by RNA-seq revealed that NLCs and NVML cells represent distinct populations, whereas NML cells are a transitional population. Transcriptional analysis suggested that NVML cells, compared to NLCs, have stem cell characteristics, including quiescence and multipotentiality, and express higher levels of key genes critical for the support of hematopoietic stem and progenitor cells (HSPCs), including CXCL12 (SDF1A), Kit ligand (KITL) and Angiopoietin-1 (ANGPT1). This signature was compatible with BM-derived Leptin receptor+ MSCs with high capacity of supporting HSPCs based on previously reported single cell RNA-seq analysis in mice (Baryawno et al 2019, Cell). Consistent with their transcriptome, functional analysis showed that NVML cells had higher colony forming units (CFU-F) and tri-lineage differentiation capacity compared to NLCs. Normal NLCs or NVML cells were also co-cultured with NBM or human MDS cell line (MDS-L) in the absence of cytokines/chemokines to evaluate the ability of these novel human MSC populations to provide a supportive niche for HSPCs. NVML cells provided increased support for normal hematopoietic progenitors compared to baseline and NLCs (fold increase over baseline 1.78±0.78(NLCs, p<0.05) vs 2.71±1.71(NVML cells, p<0.001), NLCs vs NVML cells, p<0.05). In contrast, NLCs showed increased ability of supporting MDS-L (fold increase over baseline 2.35±1.35 (NLCs, p<0.01) vs 1.5±0.5 (NVML cells, p=0.26), NLCs vs NVML cells, p=0.053). We next profiled the transcriptome of NLCs and NVML cells from 17 patients with MDS in comparison with age-matched healthy controls (n=8, >50 years old). In spite of mutational heterogeneity in this MDS patient cohort, principal component analysis (PCA) revealed a distinctive shift of MDS-derived NVML cells compared to controls. For upregulated transcripts, pathways and gene ontology analyses of differentially expressed genes in NVML cells from MDS compared to controls identified signals related to DNA/RNA processing, while most of the down-regulated transcripts identified cytokines/chemokines critical for HSPCs. When the transcriptional profiles of NVML cells from MDS patients were compared to age-matched NVML controls, critical genes for HSPC support highly expressed in normal NVML (e.g. SDF1A, ANGPT1) were significantly down-regulated. Moreover, the transcriptional signature of NVML cells from MDS patients was similar to that of normal NLCs. Together these data suggest that, in spite of mutational heterogeneity in the hematopoietic clone, MDS may differentially impact these novel human marrow MSC subsets, decreasing normal HSPC support by disrupting NVML cells, which acquired the transcriptional signature of MDS-supportive normal NLCs. These data highlight a novel paradigm to enrich key human niche cells that provide support for HSPCs. Moreover, our data suggest that these human MSC cell subsets are differentially disrupted in MDS. Mutation specific effects are also likely to impact MDS-dependent disruption of the BMME. Therefore, our novel strategy to enrich critical human BM MSC subsets that differentially support normal and malignant hematopoiesis represents a new paradigm to accelerate therapeutic targeting of the BMME in MDS and other hematologic malignancies.
Read full abstract