Erythroid Dysplasia in Stag2 Deficient Murine Models Reveals Novel Erythropoietic Function for Stag2 Cohesin

The cohesin complex is a ring-shaped structure composed of SMC1, SMC3, RAD21, and either STAG1 or STAG2 and plays a crucial role in the topological control of gene expression. Given that STAG2 is recurrently mutated in myeloid malignancies, understanding the role of STAG2 in hematopoiesis could provide novel therapeutic insights. Our prior work identified a specific role for STAG2 in facilitating new short range chromatin loops that facilitate hematopoietic stem cell (HSC) fate commitment and lineage priming. Subsequently, we demonstrated that Stag2 loss results in increased HSC self-renewal and myelodysplasia, primarily due to loss of DNA accessibility and ensuing transcriptional dysregulation. After lineage priming and fate specification, erythropoiesis is characterized by genome wide repression, loss of DNA accessibility, and enucleation. In contrast to HSCs and myeloid progenitors, we demonstrate that loss of Stag2 cohesin alters erythropoiesis at the progenitor stage and during terminal erythroid differentiation (TED) with defects in chromosomal condensation and impaired nuclear extrusion.

To investigate TED, Stag2^WT and Stag2^∆ bone marrow cells were stained for Ter119/CD44 and analyzed via flow cytometry, which revealed fewer Ter119⁺ cells (p<0.001), decreased proerythroblasts (ProE, p<0.05), and increased orthochromatic erythroblasts (OrthoE, p<0.01). As Stag2^∆ mice are not anemic, we assayed serum erythropoietin (EPO) and splenic masses, and identified an increase in EPO and splenic mass (p<0.05 and 0.001, respectively) in Stag2^∆ mice suggesting orthogonal compensatory mechanisms. Flow cytometry confirmed splenic erythropoiesis with expansion of splenic Ter119⁺ erythroblasts in Stag2^∆ mice (p<0.05). Given that enucleation occurs during the OrthoE stage, Stag2^∆ cells were stained with the nuclear binding dye Syto16, revealing fewer enucleated cells relative to Stag2^WT controls (p<0.05). Imaging flow cytometry data of Stag2^WT and Stag2^∆ bone marrow reveals enlarged nuclei in Stag2^∆ erythroblasts, though most prominent in ProEs and polychromatic erythroblasts, suggestive of impaired chromatin condensation.

At the progenitor stage, we observe decreased erythroid maturation via scRNA-seq of murine Stag2^∆ Lin^- HSPCs. Furthermore, we observe decreased expression of Gypa and Tfrc in Stag2^∆ MEPs and CFU-Es. Interestingly, we also noted aberrations to critical components of erythroblast enucleation, which requires caspase-3 mediated cleavage of lamin B to permit nuclear opening and histone release. We observe decreased caspase-3 expression associated with Stag2 loss, suggesting alterations in critical enucleation machinery from the outset of erythropoiesis. Subsequent bulk ATAC-seq of sorted Stag2^WT and Stag2^∆ CFU-Es (Lin^-cKit⁺Sca1^-CD34^-Fcγ^-CD105⁺) showed an overall more relaxed chromatin state and hyperaccessibility with Stag2 loss, suggesting defective nuclear condensation from the earliest stages of erythropoiesis. Analysis of scATAC-seq of murine Stag2^WT and Stag2^∆ Lin^- HSPCs reveals failure of chromatin remodeling across the seven progressively mature erythroid populations observed during erythroid maturation. In support of this, we observe decreased accessibility at Casp3 with Stag2^∆, implicating enucleation pathways as Stag2-dependent in their transcriptional regulation. Investigations into whether correction of aberrant caspase-3 expression restores normal enucleation is underway to identify a potential common pathway by which Stag2 loss controls erythropoiesis. Overall, these data suggest that loss of Stag2 negatively affects chromosomal condensation, gene regulation, and nuclear extrusion, requiring compensatory EPO secretion and splenic erythropoiesis to maintain appropriate hemoglobin levels. These data expand the role of Stag2 from HSC function and lineage priming to terminal erythropoiesis and define a novel physiologic role for Stag2 in erythropoiesis.