Precision Genome Editing Strategies to Treat Sickle Cell Disease and CDKL5 Deficiency Disorder

Abstract

There are over 70,000 known mutations that cause genetic disease. Genome editing technologies could enable the life-long correction of such pathogenic mutations, potentially improving the lives of thousands of rare disease patients worldwide. Nuclease-based genome editing tools lead to double strand breaks which result in stochastic indel formation. Therefore, these tools lack the precision to exactly correct many pathogenic mutations. Furthermore, nuclease-based editing can lead to undesired consequences such as p53 activation, large deletions, and chromothripsis among other unwanted outcomes. Development of genome editing strategies that can precisely correct pathogenic mutations and do not rely on double strand break formation could be transformative for the study and therapeutic landscape of genetic disorders. My doctoral studies have focused on developing clinically relevant precision genome editing strategies for two rare diseases: sickle cell disease (SCD), a devastating inherited blood disorder, and for CDKL5 deficiency disorder (CDD), a catastrophic early-onset pediatric seizure disorder. First, I describe the development of a clinically relevant base editing strategy for SCD. While adenine base editors (ABEs) cannot directly correct the pathogenic A•T-to-T•A mutation that causes the disease, base editors can convert the pathogenic sickle allele (HBBS) into a benign and naturally-occurring Makassar allele (HBBG). We utilized ABE8e-NRCH, a bespoke ABE with both an evolved SpCas9 nickase and an evolved adenine deaminase domain. With ex vivo electroporation of ABE8e-NRCH mRNA and a synthetic single guide RNA (sgRNA) into hematopoietic stem and progenitor cells (HSPCs) from sickle cell disease patients, we were able to achieve up to 80% conversion of HBBS to HBBG. We transplanted these edited cells into irradiated mice and found that sixteen weeks post-transplantation, Makassar β-globin comprised 79% of all β-globins in the blood, and that hypoxic sickling was significantly reduced. Additionally, mice that were transplanted with ABE8e-NRCH base-edited murine HSPCs had a statically significant reduction in hallmark sickle cell pathology. Furthermore, secondary transplantation of treated HSPCs revealed HBBS-to-HBBG correction was durable in long-term, bone marrow-repopulating hematopoietic stem cells. Compared to Cas9 nuclease treatment, base edited HSPCs avoided large deletions and activation of the p53-mediated DNA damage response. Our study demonstrated that ex vivo base editing represents one clinically relevant precision genome editing strategy that could be curative for SCD. Second, I describe the development of a clinically relevant prime editing strategy to directly revert the sickle cell allele back to wild-type hemoglobin. We used prime editing tools with improved architectures, namely PEmax and an engineered pegRNA (epegRNA). With ex vivo electroporation of PEmax mRNA and synthetic nicking sgRNAs, we were able to achieve up to 41% correction in HSPCs from sickle cell patients. Seventeen weeks after transplantation into immunodeficient mice, the prime-edited patient HSPCs retained HBBS-to-HBBA conversion and lineage potential. In addition, 28-43% of β-globins expressed in human reticulocytes isolated post-transplantation were wild-type adult hemoglobin, proportional to initial levels of editing. Edited cells also had a statically significant reduction in hypoxic sickling. Our experimentally unbiased off-target editing analysis at over 100 sites revealed minimal off-target editing. Compared to strategies that utilize Cas9 nuclease-mediated homology directed repair (HDR) to correct HBBS, our prime editing strategy has unique advantages like not requiring DSB formation or the delivery of an exogenous donor DNA template. Our study provides evidence that one-time ex vivo prime editing of patient HSPCs is a viable and clinically relevant precision genome editing strategy for the correction of the sickle cell allele. Finally, I describe the development of a prime editing strategy to correct various mutations causing CDKL5 deficiency disorder (CDD). We used evolved prime editors (PE6 variants) to screen correction strategies for two causative CDD mutations in patients, c.1412delA and c.826-1 G > A. For c.1412delA, we developed and tested corrective prime editing strategies in several preclinical models—both a murine and a human cell line with the pathogenic mutation, and primary patient fibroblasts, with future studies planned for the relevant mouse model and cortical organoids differentiated from patient-derived iPSCs. Using an evolved SpCas9 (SpCas9-NRCH) DNA-targeting domain, we can target the disease allele in a murine cell line and achieve 38% correction with PE6b-NRCH and PE6c-NRCH. In a homozygous c.1412delA-containting HEK293T line, we achieved to 33% and 30% correction of the pathogenic using PE6c and PE6d respectively. PE6c editing in primary fibroblasts from the patient with the c.1412delA mutation reached 28% correction with a high degree of DNA specificity. Furthermore, we observed up to 9% correction for the c.826-1 G > A mutation with PE6c in initial prime editing screens, and nearly 30% installation of replacement of a hotspot region of exon 8 that harbors a variety of CDKL5 pathogenic mutations. Our studies demonstrate the potential use of prime editing to correct multiple causative CDD mutations.