Hematopoietic Stem and Progenitor Cells (HSPCs) reside in their niche, a structure that regulates the balance of cellular quiescence, self-renewal and commitment towards differentiated cells. This highly plastic niche is formed by several cellular players, mainly endothelial cells, osteoblasts, adipocytes, and a variety of stromal progenitor cells of which CXCL12-abundant reticular (CAR) cells have been best characterized. Adipocytes have been shown to inhibit hematopoietic progenitor proliferation, but pre-adipocytes can efficiently support the growth of hematopoietic cells in culture. This suggests that the same cellular lineage contains populations with divergent hematopoietic-supportive capacity.

In the first part of this thesis, we hypothesized that pharmacological modulation of the adipocytic differentiation axis could inhibit bone marrow (BM)-derived adipocyte maturation and improve hematopoietic progenitor support. Using a phenotypic screen, we identified calcipotriol, a vitamin D derivative, as a strong inhibitor of adipocyte differentiation. In vivo administration of calcipotriol decreased BM adipocyte size at day 18 post BM transplantation, increased total number of colony-forming units (CFUs) per leg and improved the survival of Cxcl12 haploinsufficient mice. Thus, we identified this vitamin D derivative as our top candidate to accelerate hematopoietic recovery in radio/chemotherapy-induced aplasia by targeting the maturation of adipocytes in the BM.

Then, we took advantage of a prospective cohort of patients to assess the association between bone health and hematopoietic output in the context of osteoporosis. Osteoporosis is characterized by reduced bone mineral density (BMD), microarchitectural deterioration and increased BM adipocytic content. We studied the association between BMD and differential blood counts. We found that significant associations between blood counts and BMD exist, but none was consistent across bone parameters or timepoints. This suggests that, for steady-state hematopoiesis, there is no clinically significant effect of the deteriorated bone microenvironment on blood production.

In the last part, to further investigate the composition of the HSPC-supportive microenvironment, we explored the simpler, albeit functional, extramedullary niche. We showed that the adrenal gland can be transformed into a hematopoietic supportive environment and used as a model to study the minimal components of a de novo niche. This induced adrenal niche supports HSPCs, including serially transplantable hematopoietic stem cells. Circulating HSPCs home into the adrenal cortex, where they are in proximity to CAR cells. These findings are supported by the presence of CXCL12+ cells in human adrenal myelolipoma, a benign tumor that contains extramedullary hematopoietic tissue. We envision our model as a novel tool to study the minimal components needed for hematopoietic support.

Thus, using a combination of clinical, translational and basic research, we discovered a new therapeutic candidate approach to improve hematopoietic support by using a vitamin D analog to modify the BM adiposity. We showed the resilience of the BM niche by observing no measurable effect of the osteoporotic niche on hematopoietic output in a large cohort of patients. Finally, we developed a model of de novo hematopoietic niche in the adrenal gland. Altogether, our work provides new strategies to improve our understanding of the hematopoietic niche.