An essential link between pyrimidine synthesis and ferroptosis suppression in erythroid differentiation Free

The human body produces 2-3 million red blood cells per second. Excessive death of erythroid precursors results in ineffective erythropoiesis and thus anemia, a common condition seen with bone marrow failure syndromes, cancer, and chemotherapy. The regulatory mechanisms underlying the death of cells in ineffective erythropoiesis are still largely unknown. We have used zebrafish as a powerful genetic model to uncover critical transcriptional and metabolic regulators that drive erythroid differentiation. Through a chemical screen, we previously discovered an essential role for dihydroorotate dehydrogenase (DHODH), a conserved mitochondrial pyrimidine de novo synthesis enzyme, in maintaining sufficient numbers of erythroid precursors during primitive erythropoiesis. DHODH inhibition, as a consequence, leads to starkly reduced embryonic globin expression and hemoglobin production. In cancer cells, DHODH was recently shown to protect from ferroptosis, which is a type of programmed cell death due to iron-dependent lipid peroxidation. In agreement, we now found that DHODH inhibition causes increased lipid peroxidation, which, as well as the erythroid block, could be rescued by several ferroptosis inhibitors with distinct mechanisms of action. Apoptosis inhibitors, in contrast, only marginally rescued the anemia induced by DHODH inhibitors. The anemia induced by DHODH inhibitors can be thus used as a model to better dissect how cell death leads to ineffective erythropoiesis, and how ferroptosis in particular intersects with metabolic features along the erythropoietic cell lineage. Giemsa staining of sorted erythroid precursors in the presence of the DHODH inhibitor brequinar revealed a larger proportion of less differentiated basophilic to the more differentiated polychromatic erythroblasts compared to vehicle control. Co-treatment with the ferroptosis inhibitor liproxstatin-1 partially rescued this erythroid differentiation block resulting in a larger proportion of polychromatic erythroblasts. To test whether the requirement of ferroptosis suppression for normal erythroid differentiation is conserved, we generated human CD34⁺-derived immortalized erythroid precursors carrying a CRISPR/Cas9-mediated knockout of the key ferroptosis suppressor GPX4. GPX4-depleted erythroid precursors showed increased levels of lipid peroxidation as well as a profound proliferation defect, suggesting an essential role for GPX4 in protecting erythroid precursors from ferroptosis. To identify the downstream mechanisms underlying the rescue of the DHODH-induced erythropoiesis block by ferroptosis inhibitors, we performed untargeted metabolomics experiments. Among the top 25 upregulated metabolites were several metabolites of the pyrimidine salvage pathway suggesting that under ferroptosis inhibition, metabolism is re-wired to allow salvage pyrimidine synthesis to compensate for otherwise lineage-essential de novo pyrimidine synthesis and drive lineage progression into mature erythroid differentiation stages. In agreement, the pyrimidine salvage metabolite uridine was able to rescue the erythroid differentiation block induced by DHODH inhibition, both in zebrafish and human primary CD34⁺-derived erythroid progenitors. Our results demonstrate that the differentiation of erythroid progenitors and precursors depends on the suppression of ferroptosis. We further reveal a novel tight link between pyrimidine metabolism and ferroptosis defense in erythropoiesis and uncover distinct adaptive plasticity of de novo and salvage nucleotide synthesis during erythroid differentiation, which, in the long-term, will open new avenues to treat anemia.