Persistence of CRISPR/Cas9-Edited Hematopoietic Stem and Progenitor Cells and Reactivation of Fetal Hemoglobin in Nonhuman Primates

Beta-thalassemia and sickle cell disease are monogenic disorders that are currently treated by allogeneic bone marrow (BM) transplantation although the challenges of finding a suitable matched-donor and the risk of graft vs host disease have limited the adoption of this otherwise curative treatment. A potentially promising approach for hemoglobinopathies aims to reactivate fetal hemoglobin (HbF) as a substitute for defective or absent adult hemoglobin by modifying the patient's own hematopoietic stem and progenitor cells (HSPCs). Here, we evaluated CRISPR/Cas9-induced small deletions in HSPCs that are associated with hereditary persistence of fetal hemoglobin (HPFH) using our nonhuman primate (NHP) stem cell transplantation and gene therapy model.

The CRISPR/Cas9 nuclease platform was employed to recapitulate a natural genetic alteration identified in individuals with HPFH, consisting of a 13-nucleotide (nt) deletion in the gamma globin gene promoter. A first cohort of three rhesus macaques received 70-75% HPFH-edited BM-derived CD34+ HSPCs. All animals showed rapid hematopoietic recovery and peripheral blood (PB) editing levels stabilized at 12-30% for at least a year post transplantation (Figure 1). HbF production, determined by circulating F-cells, persisted at frequencies of 8-22% and correlated with in vivo PB editing. Robust engraftment of gene-edited HSPCs in the BM compartment was observed in all animals, with no measurable off-target activity or clonal expansion.

We have recently shown, that the CD34+CD90+CD45RA- phenotype is exclusively required for short- and long-term multilineage reconstitution, significantly reduces the target cell number for gene therapy/editing and is conserved between human and NHP hematopoietic cells (Radtke et al., STM, 2017). To explore this cell population further, we transplanted a second cohort of three animals by sort-purifying and solely editing this hematopoietic stem cell (HSC)-enriched CD34+CD90+CD45RA- phenotype, thus reducing the number of target cells by over 10-fold without impacting hematopoietic recovery, engraftment, or HbF reactivation. In vivo levels of gene-edited PB started at less than 5% because of the high number of co-infused unmodified progenitor cells, but rapidly increased to about 50% within 1 week (Figure 1) and stabilized at levels comparable to the CD34 cohort. This data supports our interpretation that CD34+CD90+CD45RA- cells are the main cell population relevant for long-term reconstitution and an excellent target for improved and efficient gene therapy/editing.

These results demonstrate robust engraftment and persistence of CD34+ HPSCs as well as HSC-enriched CD34+CD90+CD45RA- cells that have been CRISPR/Cas9-edited at the 13nt-HPFH site, with marked and stable HbF reactivation and no overt adverse effects in a NHP transplantation and gene therapy model. Most importantly, we validated our refined CD90+ target which reduces the need for editing reagents by 90% without compromising the gene modification and engraftment efficiencies. These are the first data in a clinically relevant large animal model to demonstrate the feasibility and clinical applicability of CRISPR/Cas9-mediated fetal hemoglobin reactivation. The successful targeting and engraftment of our HSC-enriched population should also have significant implications for gene therapy and editing of other genetic diseases.

Figure 1:Tracking of HPFH editing in transplanted animals. A) Editing efficiency was longitudinally determined by next generation sequencing of the targeted locus in PB white blood cells from 2 cohorts of transplanted rhesus animals. Frequency is represented as the proportion of all sequence reads containing an edited locus. B) Normalized frequency of the desired 13nt-HPFH deletion in the same animals as shown in A).

Figure.

View largeDownload slide

Figure.

View largeDownload slide

Close modal