A CRISPR screen in enucleated human red cells identifies CLIC3 and VAMP8 as critical regulators of terminal erythropoiesis

Abstract During terminal erythroid differentiation, committed precursor cells proliferate and differentiate down the erythroid lineage, yielding billions of new reticulocytes every day. The final step involves a striking morphologic change wherein the erythroblast nucleus is expelled in a process termed enucleation. Enucleation is a complex process involving chromatin condensation, nuclear polarization, and clearance of organelles, yet a comprehensive understanding of its genetic determinants is lacking. To identify critical factors for erythroid enucleation, we designed and implemented a pooled CRISPR screen in enucleated, cultured red blood cells (cRBCs) derived from primary human CD34+ hematopoietic stem cells (HSPCs). We adapted a CROPseq-based lentiviral vector in which the sgRNA sequences are incorporated into mRNA exported to the cytoplasm, enabling their quantification in enucleated cells via RNA-seq. By coupling transduction of a custom erythroid-focused sgRNA library with nucleofection of Cas9 protein in HSPCs followed by ex-vivo erythropoiesis, we performed a pooled CRISPR screen for determinants of erythroid enucleation. Analysis of RNA-seq data from three biological replicates identified genes whose guides were differentially abundant in the enucleated vs. nucleated cRBCs on day 17. Two of the top candidates were the Chloride Intracellular Channel 3 (CLIC3) and Vesicle Associated Membrane Protein 8 (VAMP8), neither of which has a previously described role in enucleation. CLIC3 has been found to be overexpressed in several solid tumors, where it is associated with poor prognosis. Although it has been identified in the RBC proteome, its function in erythroid cells has not been described. We detected CLIC3 expression in all stages of differentiating primary erythroblasts and observed that the protein was primarily localized in the cytoplasm. Knock-down of CLIC3 led to an overall delay in erythroid differentiation and impaired enucleation, as assessed by single-cell proteomics and microscopy. This was associated with an early G0/G1 cell cycle arrest from day 7 of differentiation, suggesting that CLIC3 is important for proper cell cycle progression during erythroid maturation. Transcriptomic analysis of CLIC3-knockdown cells revealed a significant upregulation of the p53 pathway, including p21. Interestingly, CLIC3 has been shown to interact with the acetyltransferase NAT10 in bladder cancer, and this interaction has been proposed to promote tumor progression by altering the acetylation and stability of p21 mRNA. We observed that CLIC3 also interacts with NAT10 in erythroid cells, raising the possibility that the cell cycle dysregulation in CLIC3-depleted erythroblasts may involve similar epigenetic mechanisms. The other top candidate, VAMP8, is a SNARE protein involved in autophagosome-lysosome fusion. We confirmed that VAMP8 is expressed in the erythroid lineage and found that it co-localizes with lysosome markers. In contrast to CLIC3, depletion of VAMP8 in primary human CD34+ HSPCs resulted in an initially accelerated differentiation with an over-abundance of orthochromatic erythroblasts observed on day 13 relative to control cells. However, the knock-down cells displayed a specific defect at the step of enucleation, with a ~40% decrease in enucleated cells on day 17 compared to the control population. High-resolution imaging of VAMP8-depleted orthochromatic erythroblasts revealed significantly fewer cells with polarized nuclei on day 15 suggesting VAMP8 is necessary for this step of enucleation. Interestingly, our analysis of control cells undergoing nuclear expulsion revealed actin foci consistent with contractile actin rings (CAR). In contrast, VAMP8-depleted cells had overall less actin foci, and the formation of a CAR was not observed. Together these results could suggest a link between the autophagic-lysosomal pathway orchestrated by VAMP8, and the final cytoskeletal changes important for enucleation. Our work highlights the existence of undiscovered factors important for terminal erythroid differentiation and enucleation, and the potential of forward genetic screening in enucleated RBCs to shed light on this final, relatively understudied step of erythropoiesis. Using this tool, we have uncovered CLIC3 and VAMP8 as novel regulators of human erythropoiesis. This work provides both a new technology for screening in enucleated cells and new fundamental biological insights into erythropoiesis.