Comparative Analysis of CRISPR-Cas9, Lentiviral Transduction, and Base Editing for Sickle Cell Disease Therapy in a Murine Model

Abstract

Introduction

Sickle cell disease (SCD) is a red blood cell (RBC) disorder resulting from a point mutation in the β-globin gene resulting in sickle hemoglobin (HbS) polymerization under hypoxic conditions¹. The FDA approved Casgevy® and Lyfgenia ^TM for patients with SCD^2,3. These options have both led to similar promising clinical outcomes, making it unclear which approach is best for patients at this time^4,5. Utilizing a base editing approach may reduce the risk of insertional mutagenesis and large-scale genomic rearrangements. This approach converts HbS into Makassar hemoglobin (HbG), a benign naturally occurring variant^6,7. Herein, we sought to study the differences between each gene therapy approach described above in an immunocompromised mouse model.

Methods

We optimized conditions ex vivo with all three editing techniques in both healthy and sickle CD34+ hematopoietic stem and progenitor cells (HSPCs) using bluebird bio's BB305 β^A-T87Q-globin vector for the transduction group, CRISPR-Cas9 ribonucleoprotein to target the erythroid specific BCL11A +58 enhancer, or an adenine base editor mRNA to convert HbS to HbG-Makassar. Edited SCD CD34+ HSPCs were infused into busulfan conditioned NBSGW mice (n = 5 mice for each technique, including a non-edited group as a control and a group infused with a 1:1:1 mixture of all three products).

Results

Ex vivo analysis of the infused cells showed ∼95% editing of BCL11A enhancer, ∼75% base editing at β-globin, and a vector copy number (VCN) of ∼0.9 copies/diploid genome. At 16 weeks of transplantation, bone marrow (BM) analyses showed similar human CD45+ cell engraftment frequencies across all groups (22-94%). In engrafted mice, there was 95.8% BCL11A editing and 62.8% β-globin base editing. VCN in mice infused with transduced cells were 2 copies/diploid genome. In the competitive transplantation group, there was 8% BCL11A enhancer editing, 26% base editing and a VCN of 0.3 copies/diploid genome detected (Figure 1). Higher F-cells (%) in BCL11A-edited and the competitive transplantation groups were detected in comparison to the remainder of groups (p<0.05). Using RP-HPLC and mass spectrometry, we found 73% HbG production in the base editing group, 62% γ-globin production in BCL11A edited group, 38% β^A-T87Q production in the BB305 group (Figure 2). An anti-sickling assay showed a 65%, 30%, 50%, and 63% reduction in BM RBC sickling in Makassar, BCL11A, BB305, and Mix groups, respectively (Figure 3).

Conclusions

This study compared three gene therapy approaches - CRISPR-Cas9, lentiviral transduction, and base editing - in an immunocompromised mouse model. All methods showed significant therapeutic potential, with the increase in respective globins and reduced sickling. In the competitive transplantation setting, both base editing and BB305 transduction showed superior results in comparison to CRISPR-Cas9 mediated BCL11A enhancer editing.