A novel LRF degrader induces fetal hemoglobin expression in adult erythroblasts

Abstract

Abstract Elevating fetal hemoglobin (HbF) represents a validated therapeutic strategy for sickle cell disease (SCD) and β-thalassemia. Hydroxyurea is the only FDA-approved pharmacologic HbF inducer, but its efficacy is limited, and patient responses are variable. Emerging HbF-elevating gene therapies show great promise, but are complex, expensive, and globally inaccessible. Thus, there is a persistent need for effective and accessible pharmacologic intervention. Transcription factors BCL11A, NF1A/X and LRF (also known as ZBTB7A) are direct repressors of the fetal β-like globin genes HBG1 and HBG2. However, direct specific pharmacologic inhibition of these repressors has been challenging. Here, we report the development and preclinical characterization of YU1206, a cereblon-based molecular glue degrader of LRF. Using a dual luciferase LRF reporter system in K562 cells, we screened 1,138 compounds with scaffolds based on pomalidomide and thalidomide, established cereblon-targeting molecular glue degraders. Structure-activity relationship optimization of initial hits yielded YU1206 with improved selectivity and potency for LRF degradation. YU1206, but not a methylated analog which does not bind cereblon, induced dose-dependent degradation of LRF that was competitively inhibited by the cereblon ligand pomalidomide. In the human adult-stage erythroid cell line HUDEP-2, YU1206 degraded LRF in a dose-dependent manner, increased HbF positive cells (F-cells) from ~20% to ~60% and HbF protein levels from ~5% to ~40% of total hemoglobin as measured by HPLC. HBG1/2 mRNA production was increased proportionally, consistent with LRF's role as a transcriptional repressor. Proteome profiling confirmed reduction of LRF, with HBG1/2 emerging among the most highly induced proteins. Importantly, levels of erythroid transcription factors (e.g. GATA1, FOG1, p45 NF-E2 and KLF1) as well as erythroid maturation markers (e.g. BAND3, GLYCOPHORIN A, AHSP, ALAS2) remained unchanged, consistent with preserved erythroid maturation. In primary human CD34+ hematopoietic stem and progenitor cells (HSPCs)-derived erythroid cells, 200nM YU1206 degraded LRF, increasing F-cells from ~20% to ~65%. HbF protein levels rose from 5% to ~40% of total hemoglobin. ZBTB7A mRNA levels were unchanged, consistent with a post-translational mechanism of YU1206 function. Notably, at 200 nM concentration, YU1206 did not significantly impair proliferation and erythroid maturation as evidenced by cell morphology, normal CD71+CD235a+ distributions, and gene expression measurements. In primary erythroid precursor cells from SCD patients, YU1206 induced HbF to similar levels and markedly reduced cell sickling. To assess the activity of YU1206 in vivo, we crossed mice containing a humanized cereblon variant with Townes mice harboring human HBG and adult HBB genes. Following daily intraperitoneal YU1206 administration for two weeks, flow cytometry analysis of CD71+ and Ter119+ bone marrow cells revealed marked F-cell induction and enhanced HBG transcription, with effect sizes comparable to those observed in similar mouse models in which BCL11A was genetically depleted. YU1206 treatment did not significantly impair erythroid maturation, as evidenced by preserved peripheral blood counts, normal erythroid parameters, and normal CD71+Ter119+ distributions within the bone marrow. Previous studies demonstrated that complete genetic LRF loss impairs erythroid differentiation, raising initial concerns about LRF as a therapeutic target, yet YU1206 had little impact on erythroid maturation and viability in vitro and in mice. This may be explained by the residual LRF protein due to incomplete degradation. In conclusion, we identified YU1206 as novel molecular glue LRF degrader for HbF induction, highlighting the potential of future pharmacologically optimized LRF degraders for the treatment of SCD and β-thalassemia.