PIEZO1 Gain-of-Function Variants Lead to Alterations in Late-Stage Erythropoiesis By Enhancing Enucleation Rate

PIEZO1 is a mechanosensitive cation channel that plays a crucial role in various physiological processes as a mechanical force sensor. Gain-of-function (GoF) mutations in PIEZO1 lead to dehydrated hereditary stomatocytosis (DHS) or hereditary xerocytosis by slowing PIEZO1 inactivation kinetics. DHS patients exhibit a range of clinical presentations, including mild to severe hemolytic anemia, as well as hepatic iron overload. Erythroid progenitor cells from DHS patients show mutation-dependent alterations in erythroid differentiation. Both constitutive and macrophage-specific GoF Piezo1 mice demonstrated that PIEZO1 is a key regulator of macrophage phagocytic activity and subsequent erythrocyte turnover. In this study, we investigated the specific role of PIEZO1 in an in vitro model of isolated erythropoiesis. We first generated an erythroid model of DHS using Human Umbilical cord blood-Derived Erythroid Progenitor-2 (HUDEP2) cells, by inserting the PIEZO1 GoF variant, R2456H, in the heterozygous state by CRISPR/Cas9. The resulting cell line, HUDEP2-PIEZO1-KI (PIEZO1-KI), was induced to erythroid differentiation for 10 days by erythropoietin and compared to HUDEP2-WT cells. Flow cytometry analysis revealed that PIEZO1-KI cells presented an altered expression of late erythroid differentiation markers. We observed a significant increase in GPA+ cells (a marker of late erythroblasts) in PIEZO1-KI cells compared to WT cells and no differences in CD36+/CD49d+ cells (early erythroid markers). Hoechst staining showed an increased enucleation rate in PIEZO1-KI cells compared to PIEZO1-WT cells and an increase in the differentiated population (Hoechst-/GPA+). Erythroid population analysis showed a significant rise in intermediate precursors percentage (CD36moderate/low, CD49dmoderate/low, GPAmoderate/high) in PIEZO1-KI cells compared to WT (pValue by ANOVA - Tukey's correction for multiple comparison). To characterize the molecular basis of the cellular phenotype observed, RNAseq was performed on day 0, 7 and 10 of differentiation. We first focused on genes physiologically regulated during differentiation of HUDEP2-WT cells (359 genes). Among them, we selected both genes with the same regulation trend (List_1: n=258) and with a different regulation trend in PIEZO1-KI cells compared to WT (List_2: n=101). Gene ontology on genes of list_1 revealed, as expected, “erythrocyte development” as the most enriched biological process (BP). Particularly, single gene analysis revealed that three genes, although presenting the same regulation trend, are significantly downregulated in PIEZO1-KI cells at day 10 of differentiation: ALAS2 (heme formation), DMTN (erythrocyte shape regulation), and BPGM (hemoglobin oxygen affinity regulation and glycolysis). Three other genes involved in glycolysis were found to have different regulation trends: PFKM (involved in the first step of glycolysis) was suppressed during differentiation, while PGAM1 and ENO3 (conversion of 3-phosphoglycerate to 2-phosphoglycerate and to phosphoenolpyruvate) were upregulated. Taking together, these data demonstrate that the PIEZO1 GoF variant influences the later stages of erythroid differentiation in vitro by affecting genes involved in erythropoiesis and glycolysis. These in vitro findings link several components of glycolysis to changes in erythroid differentiation caused by a prototypical PIEZO1 GoF mutation in DHS. These findings suggest future studies should investigate the potential benefits of modulating glycolysis in DHS models.