Fig. 1.

CRISPR–Cas9 whole-genome screening in HEL cells to identify regulators of erythropoiesis. a CRISPR–Cas9 screen design for enrichment of sgRNAs promoting the CD235a-/low state. b Individual genes retested by flow cytometry for CD235a surface expression on day 10 post-transduction in HEL cells. Cells were transduced with individual gene retests pools of four sgRNAs (see Methods for details). A modified z-score cutoff for a p-value < 0.01 was used to define a positive hit, with all scoring genes indicated within the box. For the m⁶A MTase complex, only METTL14 scored in the primary screen. We found that the sgRNAs targeting METTL3 and WTAP in the screening library were not effective and substituted sgRNAs from the human CRISPR Brunello lentiviral pooled library for METTL3 and WTAP. LDB1 and GFI1B each scored with 1 sgRNA in the initial screen and new sgRNAs were generated, also from the Brunello library. c Diagram of the three primary categories of screen hits. Top panel shows genes with multiple sgRNA hits in bold; all others have a single sgRNA scoring. Middle and bottom panels: genes validated by secondary individual gene tests highlighted in color; solid lines indicate 2 or more sgRNAs scored from primary screen, while dashed lines indicate 0 or 1 sgRNAs scored from primary screen. d Representative flow cytometry results for positive retest hits in HEL cells. Cells were transduced with lentiCRISPRv2-mCherry virus and assayed by FACS 7–9 days later. (n = 3–5 biological replicates). e Representative flow cytometry results for METTL3 rescue experiments utilizing full-length WT METTL3 or a catalytically inactive mutant METTL3, in sgMETTL3-KO HEL cells. (n = 2 biological replicates, mean ± SEM. Student’s t-test two-sided, %CD235a: NTC 89 ± 2.2; sgMETTL3 71.4 ± 1.9, p = 0.026; sgMETTL3/WT METTL3 88.7 ± 1.8; sgMETTL3/Mut METTL3 67.1 ± 1.5, p = 0.01). f Western blot for METTL3 sgRNA KO and expression levels following rescue with WT METTL3 or the catalytically inactive mutant