Bone Marrow Microenvironment Dictates the Progression of Clonal Hematopoiesis through Stromal-Macrophage Cell-Cell Interactions

Clonal hematopoiesis (CH) was first recognized as a distinct category of precursor myeloid disease state by WHO in 2022. Commonly originates from somatic mutations in one of three genes (TET2, ASXL1, DNMT3A) within HSPCs, CH leads to the clonal expansion of mutant hematopoietic cells and the development of cytopenia and myeloid-biased hematopoiesis. The occurrence of CH rises with age and impacts over 10% of individuals aged 70 and above .CH elevates the risk of developing myeloid malignancies by nearly13 folds. Many studies have linked TET2 deficiency to cytokine expression and inflammatory environment, including in the bone marrow. Tet2 knockout mice display increased myeloid skewing and develop age-dependent CH phenotypes, which eventually progresses to chronic myelomonocytic leukemia (CMML)-like disease and exhibits high levels of inflammatory cytokine and chemokine in serum. A proliferative advantage for Tet2^-/- murine and TET2-mutant human HSPCs was observed in vitro with the presence of the proinflammatory cytokine TNFα. These findings imply that TET2 mutations promote clonal dominance by endowing HSPCs with a cell-intrinsic resistance to inflammatory cytokines, simultaneously contributing to the creation of a conducive inflammatory environment.

To investigate how extrinsic inflammatory cues from non-hematopoietic stromal cells contribute to disease progression, we transplanted Tet2^-/- HSPCs into lethally-irradiated Gsdmd whole-body knockout (Gsdmd^-/-) mice and their WT littermates (five-month-old donors to five-month-old recipients). In this model, Gsdmd^-/- mice provides a Gsdmd-null bone marrow microenvironment. Eight months post-transplantation, we found that, unlike their WT counterparts, Gsdmd^-/- mice failed to develop CH to MDS transformation upon receiving Tet2^-/- HSPCs. The histological features of the bone marrow and spleen were significantly retained in Gsdmd^-/- recipients. We observed that loss of Gsdmd in the bone marrow microenvironment significantly reverted anemia and monocytosis. In addition, we conducted single-cell RNA-seq (scRNA-seq) on bone marrow mononuclear cells from these mice. Our findings revealed a notable increase in inflammatory macrophages and exhausted T cells within the ‘Tet2^-/- to WT’ group. Remarkably, this phenotype was significantly mitigated in the Gsdmd^-/- recipients. These data collectively indicate that Gsdmd in the non-hematopoietic bone marrow microenvironment plays a critical role in the progression of CH in this model.

Xenium Spatial Transcriptomic Analyses provided further insights into the cellular architecture and interactions within the bone marrow environment. Being consistent with our scRNA-seq data, we observed a marked increase of inflammatory macrophages in ‘Tet2^-/- to WT’ group. Notably, the Xenium analysis revealed a distinct cell-cell interaction between inflammatory macrophages and non-hematopoietic stromal cells. This interaction was exclusively observed in ‘Tet2^-/- to WT’ group, whereas it was abolished in the ‘Tet2^-/- to Gsdmd^-/-’ group, suggesting a specific mechanistic pathway through which Gsdmd protein in non-hematopoietic stromal cells may facilitate the communication between inflammatory macrophages and stromal cells. This finding implies that Tet2 mutation in hematopoietic cells alone is insufficient to drive this interaction; rather, it is the presence of Gsdmd in the non-hematopoietic stromal cells that appears to be the key in mediating this communication, highlighting a novel dimension of the inflammatory response regulated by the bone marrow microenvironment.

Additionally, we have developed GSDMD knockout iPSCs to construct GSDMD-null human bone marrow organoids. GSDMD knockout did not affect the formation and composition of the bone marrow organoids. Preliminary data indicate that CD34⁺ cells from MDS patients with a TET2 loss-of-function mutation exhibit significantly increased myeloid differentiation when engrafted into WT organoids. Conversely, this myeloid-skewed differentiation is notably reduced in GSDMD-null organoids. These data highlight the important role of GSDMD in bone marrow microenvironment in human context.

By employing both murine models and human bone marrow organoids, we demonstrated that the absence of GSDMD in the bone marrow stromal cells significantly impedes the development of CH and its associated inflammatory responses.