Domain-Focused CRISPR-Cas9 Screen Identifies Hri Kinase As a Regulator of Fetal Hemoglobin in Adult Human Erythroid Cells

Increasing fetal hemoglobin (HbF) levels in adult red blood cells provides clinical benefit in patients with sickle cell disease (SCD) and some forms of β-thalassemia. While developing genome editing and gene replacement approaches promise long term HbF elevation in SCD patients, there remains a strong need for pharmacologic treatment of these diseases. Two transcription factors, BCL11A and LRF (ZBTB7A) and their co-regulators mediate most of fetal globin transcriptional silencing in adult erythroid cells. However, transcription factors are inherently challenging to inhibit with small molecules.

To identify potentially druggable HbF regulators, we carried out a CRISPR-Cas9 based genetic screen in a human erythroid cell line targeting protein kinases as they are inherently amenable to inhibition by small molecules. Most CRISPR-screens published to date have made use of sgRNAs targeting early exons to induce frameshift mutations. It has recently been shown that targeting sgRNAs to functional protein domains significantly improves screening efficiencies, as both in-frame and frameshift mutations contribute to generating hypomorphic alleles. We thus designed a sgRNA library targeting almost all annotated kinase-domains using a newly optimized sgRNA-scaffold that improves on-target editing activity. The screen uncovered the heme-regulated inhibitor HRI (also known as EIF2AK1) as an HbF repressor. HRI is an erythroid specific kinase known to phosphorylate the translation initiation factor eIF2α. eIF2α phosphorylation generally tempers protein synthesis, but select transcripts with upstream Open Reading Frames (uORFs) in their 5'UTR are known to escape this inhibition.

HRI depletion increased the fraction of HbF+ cells and elevated γ-globin (the fetal β-type globin chain) mRNA and protein levels. These effects were surprisingly specific as evidenced by whole proteome mass spectrometry and RNA-seq. Importantly, HRI depletion did not adversely affect cell growth or maturation. Similar results were obtained in primary CD34+ derived erythroid cells that were HRI depleted. HRI deficiency in erythroid cultures from patients with SCD reduced cell sickling, suggesting that biologically relevant HbF levels might be achieved by targeting HRI.

Mechanistically, HRI loss strongly reduces BCL11A expression post-transcriptionally without impacting LRF production. BCL11A appears to be a major effector of HRI function since forced expression of BCL11A in HRI-depleted cells restores HbF repression to a significant degree.

Finally, we explored whether pharmacologic HbF inducers that act via distinct mechanisms might further amplify HbF production. Pomalidomide has recently been shown to lower BCL11A mRNA levels and raise HbF production. Our proof-of-concept experiments demonstrate strong cooperativity of HRI depletion and pomalidomide treatment in HbF induction. HRI emerges as an erythroid specific target for HbF induction alone or in combination with mechanistically distinct HbF inducers.