Abstract
β-hemoglobinopathies, the most common monogenetic blood disorders, result from defective function or inadequate production of β-globin. Unfortunately, accessible and efficient treatments remain limited. Previous studies demonstrated that β-hemoglobinopathies patients with increased naturally occurring β-like subunit, γ-globin, a component of fetal hemoglobin (HbF), exhibit milder disease severity. Investigating novel factors that upregulate γ-globin expression could provide new therapeutic avenues for patients affected with β-hemoglobinopathies.
In this study, we performed an unbiased genome-wide pooled CRISPR activation (CRISPRa) screen to identify novel γ-globin regulators. First, we generated a clonal HUDEP-2 cell line stably expressing the MPH activation complex (MS2-P65-HSF1). This cell line was then transduced with a lentiviral library expressing a VP64 activation element fused to a catalytically dead Cas9, a Blasticidin resistant gene, and one of 3 sgRNAs targeting virtually every gene in the human genome. On day 8 of differentiation, intracellular HbF staining was performed, and the 10% highest and lowest HbF-expressing cells (F-Cells) were sorted. sgRNAs were deep sequenced and compared between high and low γ-globin expressing cells. Expectedly, sgRNAs targeting the previously reported γ-globin repressor, BCL11A, were enriched in low γ-globin expressing cells, while sgRNAs targeting the γ-globin promotor were enriched in the high γ-globin expressing cells. Interestingly, this screen also identified several previously unknown putative γ-globin activators.
To validate our screen results, we first chose the top 25 candidate genes for which sgRNAs were enriched in high HbF-expressing cells. For each candidate gene, we generated lentiviral particles expressing an sgRNA that targets it, VP64 fused to a catalytically dead Cas9, and a Blasticidin resistant cassette and then transduced into HUDEP-2 cells expressing MPH. Differentiated HUDEP-2-MPH cells were analyzed for percent F-Cells by flow cytometry and HBG1/2 (which encode γ-globin) mRNA levels by RT-qPCR. Compared to control cells transduced with non-targeting sgRNAs, activation of HBG2, HIC1, CHCHD7, FXR2, ANKH and SKIL resulted in >2-fold increase in %F-Cells and HBG1/2 mRNA levels. Since HIC1 is paralogous to HIC2, we confirmed that the HIC1-activating sgRNA does not additionally activate HIC2. Furthermore, we showed that HIC1 activation results in reduced BCL11A mRNA, suggesting that HIC1 may induce HbF production by suppressing BCL11A, a hypothesis we are currently testing.
Next, we sought to validate our findings in primary human erythroid cells. To do so, we co-transduced primary human hematopoietic stem and progenitor cells (HSPCs) undergoing erythroid differentiation with two lentiviral constructs, one expressing MPH and a hygromycin resistant cassette and another expressing an sgRNA targeting a gene of interest, VP64-dCas9, and a Blasticidin resistance cassette. We found that, compared to control cells transduced with non-targeting sgRNAs, activation of HBG2, HIC1, CHCHD7, FXR2, ANKH and SKIL led to >3-fold increase in γ-globin mRNA levels, validating our HUDEP-2 findings described above. In additional preliminary results, we found that HIC1 and CHCHD7 cDNA overexpression in HSPCs resulted in increased %F-Cells and γ-globin mRNA. Further studies aimed at defining the mechanisms by which activation of the genes identified by our screen result in HbF induction are ongoing.
In summary, we identified novel regulators of γ-globin expression. Our findings have important implications for furthering our understanding of β-globin switching, with important implications for β-hemoglobinopathies.
