A chance to cut (the genome) is a chance to cure

In this issue of Blood,Antoniani et al identify an innovative genome editing approach to induce fetal hemoglobin (HbF), which may eventually lead to therapeutic strategies for ameliorating or curing sickle-cell disease (SCD) and β-thalassemia.¹ 

Significant advances have been made in deciphering the molecular underpinnings of hemoglobin switching. These findings hold substantial promise for being able to identify improved approaches for HbF induction to treat SCD and β-thalassemia.²  However, for patients with these diseases, treatment remains predominantly palliative, with allogeneic hematopoietic stem cell (HSC) transplantation being the only curative therapy available. Experimental gene therapy has shown promise, but these approaches have a number of limitations, including concerns about the inability to produce sufficient hemoglobin by randomly integrating lentiviral transgenes. With the recent explosion of genome editing tools, including clustered regularly interspaced short palindromic repeats (CRISPR) and CRISPR-associated protein 9 (Cas9) (CRISPR/Cas9), there is potential for endogenous correction of pathogenic mutations, which could overcome the challenges present with HSC transplantation or gene therapy.

The hematopoietic system, in particular, is uniquely poised to host major advances in genome editing. Pursuing this strategy in HSCs circumvents many current obstacles facing genome editing in other tissue types. HSCs can be manipulated ex vivo and then delivered effectively to patients.²  Groundbreaking studies have illustrated how disruption of endogenous genes in human HSCs using genome editing can be therapeutically valuable, and genetic studies have been critical in guiding these approaches. One important example is the inactivation of the HIV coreceptor CCR5 to treat individuals with acquired immunodeficiency syndrome, a strategy that was motivated by identifying individuals resistant to HIV infection who had natural deletions in CCR5.² 

In the case of SCD and β-thalassemia, correction of the causal mutations could be achieved using genome editing. However, such approaches are challenging to perform in bulk HSC populations, where the majority of genome editing events result in gene disruption.^3,4  This has motivated the search for alternative strategies that take advantage of the high frequency of genomic disruption events. For example, an erythroid enhancer of the key regulator of HbF, BCL11A,⁵  has been a target of potential genome editing-based approaches.^6,7  There have also been efforts to recapitulate naturally occurring hereditary persistence of fetal hemoglobin (HPFH) deletions found in rare individuals. CRISPR/Cas9-mediated recreation of a 12.9-kb Sicilian HPFH deletion or a short 13-bp deletion in the γ-globin promoters in human hematopoietic progenitors has enabled effective HbF induction.^8,9 

Although these studies are promising, the actual effectiveness of such approaches remains uncertain. These deletions have only been reported in 1 or a few patients, and therefore, the extent of HbF induction that could be achieved by recreating the deletions in vivo is unclear. Several years ago, we had analyzed a large group of HPFH and δβ-thalassemia deletions, with the former having higher HbF levels than the latter, which are insufficient to compensate for globin chain imbalance.¹⁰  These mapping studies had allowed us to identify a 3.5-kb region upstream of the δ-globin gene that was proposed as a potential silencer region (see figure). This region was intact in individuals with lower HbF levels because of δβ-thalassemia deletions but was absent in those with higher HbF levels because of HPFH deletions. Interestingly, this region appears to be bound by BCL11A and its cofactors,^5,10  suggesting a potential mechanism of action. However, the functionality of these regions has remained untested.

In the present study, Antoniani et al directly address this question of functionality by examining how specific genomic modifications in the β-globin locus can result in HbF induction. Specifically, the authors use a combination of deletional mapping and epigenomic data to select 3 target regions for disruption with CRISPR/Cas9 genome editing: (1) the 3.5-kb silencer region itself; (2) a 7.2-kb deletion (the known “Corfu” deletion), which includes the silencer region and the δ-globin gene; and (3) a 13.6-kb deletion that removes the silencer region, as well as the δ- and β-globin genes (see figure). Interestingly, the authors observe that the 3.5-kb and 7.2-kb deletions caused little or no HbF induction, whereas deletion of the 13.6-kb region led to a robust elevation of HbF, suggesting that an element of gene competition between the fetal and adult globin genes may limit the degree of HbF induction that can be achieved. Moreover, the authors demonstrate that the creation of the 13.6-kb deletion alters chromatin conformation patterns and can ameliorate the sickling phenotype in SCD patient–derived primary erythroid cells. A fascinating observation is that inversion of this 13.6-kb region appears to be as effective for HbF induction as deletion of this region, suggesting that conformation of this region has a critical role in its activity.

This work is exciting and raises a number of important questions for future studies. The reason why removal of the silencer region with or without the adult globin genes can have such a variable impact on HbF levels will require further study. In addition, the exact necessity of specific elements within the targeted 13.6-kb region will require further mechanistic dissection. Finally, the reason why inversion of this region is as effective as deletion is unclear with current models of globin gene regulation, emphasizing the critical need for further studies. In addition to the mechanistic implications, the findings reported here suggest a highly effective and potentially valuable genome editing strategy for treatment of SCD and β-thalassemia. Further preclinical studies will be needed, such as through the use of improved immunodeficient xenograft models, but the work reported here is promising and may even be more effective than other proposed editing strategies that are under clinical consideration.^6,9  This paper nicely illustrates how by utilizing experiments of nature, we can better understand the molecular basis for a wide range of biological processes and define effective therapeutic opportunities.