Gene Therapy for Inherited Blood Disorders

National Heart, Lung, And Blood Institute

Investigators

Linked publications, trials & patents

Abstract

Objective 1: Develop a targeted preparative regimen for HSPC gene therapy We hypothesized that cMPL might be a relevant antigen for an antibody-based targeted depletion of human HSPCs and provide the basis for a safer conditioning regimen prior to transplant. To investigate this possibility, we produced a recombinant anti-cMPL bivalent (bi) single-chain fragment variable (scFV) fused with diphtheria toxin (DT) truncated at residue 390 (DT390-biscFV(cMPL). We first confirmed the cMPL-dependent cytotoxic effects of the DT390-biscFV(cMPL) conjugate in a HEK293A cell line engineered to express the human cMPL receptor. Next, we assessed DT390-biscFV(cMPL) for its ability to inhibit growth of human CD34+ cells in vitro. Consistent with a cMPL dependent cytotoxic effect, increased cellular death was measured in populations expressing higher densities of cMPL receptors, suggesting preferential targeting of the most primitive hematopoietic compartment. We then assessed whether DT390-biscFV(cMPL) could safely deplete human HSPCs in vivo in a humanized NBSGW mouse model. At 12 weeks post-transplantation, engrafted animals received a single maximum tolerated dose (1.2 mg/kg) of DT390-biscFV(cMPL) or vehicle. We observed a gradual decline in HSPC activity, with a peak 2.6-fold reduction in frequency of human CD45+CD13+ cells at 6 weeks following administration of DT390-biscFV(cMPL) compared to untreated animals (p = 0.003). To further evaluate safety and efficacy in vivo in a more clinically-relevant model, we administered single doses of 0.2, 0.4, 0.6 or 0.8 mg/kg DT390-biscFV(cMPL) by intravenous infusion into three independent rhesus macaques. We selectively depleted >90% cMPL+CD34+ cells at all doses tested. Notably, the drug displayed a short serum half-life (33-163 minutes), thus conferring a distinct advantage for pre-transplant conditioning applications. The overall safety profile was favorable. Objective 2: Evaluate the impact of post-transplant G-CSF administration on gene-edited HSPCs. G-CSF is commonly administered to hasten recovery from chemotherapy-induced neutropenia after autologous transplantation. However, in patients pre-treated with chemotherapeutic agents, G-CSF was shown to exacerbate HSPC toxicity triggered by these drugs by prompting their differentiation or senescence. We reasoned that G-CSF might also negatively impact HSPCs treated with DNA-damaging programmable nucleases and potentially reduce their engraftment in vivo. To address this question, human CD34+ cells were gene-edited by electroporation of sgRNA/Cas9 RNP complexes and subsequently transplanted into NSG mice. We subcutaneously injected G-CSF or a vehicle control solution once daily for the first 14 days after cell infusion and compared hematopoietic reconstitution between experimental groups. From week 10 post-transplantation, administration of G-CSF resulted in a 3- to 4-fold reduction in PB human cell engraftment compared to untreated mice and a marked reduction in human repopulating activity within the spleen and marrow at the endpoint (22 weeks) analysis. In limiting-dilution secondary transplant experiments, the overall frequency of HSPCs with long-term repopulating ability was reduced by 9.7-fold with G-CSF administration post-transplant (p = 0.011). Notably, vehicle- and G-CSF-treated mice revealed no difference in human cell engraftment after transplantation of unmanipulated HSPCs or HSPCs electroporated with Cas9 alone or HSPCs transduced with lentivirus vectors expressing GFP. Collectively, our data suggest that G-CSF use post-transplant significantly impairs long-term engraftment of CRISPR-Cas9 gene-edited HSPCs. Objective 3: Develop safe and efficient HSPC gene editing approaches HDR-based gene editing of human HSPCs. To facilitate HDR-mediated targeted integration of large DNA cargoes for therapeutic applications, we have evaluated the use of vectors based on baculovirus (BV) as alternative gene delivery vehicles in human HSPCs. Baculovirus vectors are capable of packaging inserts of at least 38 kb in size. We constructed VSV-G pseudotyped BV vectors harboring a copGFP reporter flanked by 4kb sequences homologous to the ITGB2 genomic locus implicated in LAD-1. To evaluate vector functionality, we first transduced 293A and K562 cell lines as well as primary human HSPCs. Exogenous expression of copGFP was robust in 293A cells, but approximately 7-fold lower in the hematopoietic cell line K562 and mostly undetectable in CD34+ and CD34+CD38- cell populations, independent of vector MOI or duration of transduction. We uncovered an early innate immune block to BV transduction in K562 cells and primary human HSPCs mediated by the cGAS-STING cytosolic DNA sensing pathway. The transduction blockade could be overcome in part by a brief (45-minute) pre-treatment with H-151, a potent small molecule inhibitor of STING, in combination with zVAD and Nec-1 compounds to limit activation of apoptotic and necrotic pathways. We next evaluated the ability of electroporated sgRNA/Cas9 RNPs to direct the integration of a BV-packaged copGFP reporter gene construct at the ITGB2 locus in K562 cells. The levels of GFP expression observed in cells treated with BV+RNP stabilized by day 21, with an average of 12% GFP+ cells observed during the remaining duration of the time-course. Stable targeted integration at the ITGB2 locus was confirmed by in-out PCR amplification and ddPCR of genomic DNA isolated from edited K562 cells at the end of culture. NHEJ-based gene editing of human HSPCs. Since DNA DSB lesions are predominantly repaired by the NHEJ mechanism in HSPCs, we hypothesized that an NHEJ-based approach to gene addition in HSPCs might facilitate efficient site-specific transgene integration in these cells and provide an alternative to HDR-mediated gene editing methods. In this study, we utilized a homology-independent targeted integration (HITI)-based approach to achieve robust site-specific transgene integration in human adult CD34+ HSPCs. As proof-of-concept, a reporter gene was targeted to a clinically relevant genetic locus (ITGB2) using a rAAV6 vector and sgRNA/Cas9 RNP complexes. We demonstrate high levels of stable HITI-mediated genome editing (21%) in repopulating HSPCs after transplant into immunodeficient mice. Our study demonstrates that HITI-mediated genome editing provides an effective alternative to HDR-based transgene integration in CD34+ HSPCs for the treatment of monogenic diseases affecting the hematopoietic system. This work was published Molecular Therapy 2021, and J. Clin. Med. 2021. Objective 4. Assess genotoxicity related to gene editing of human HSPCs To address the significance of off-target CRISPR/Cas9 activity for the treatment of inherited blood disorders, we performed Cas9 RNP-based gene editing at two genetic loci (CXCR4 and AAVS1) in human mobilized PB CD34+ cells and characterized the genome-wide accumulation of post-editing somatic mutations using high-throughput WGS analysis of single-cell-derived HSPC clones. Total somatic variants, including indels, single nucleotide variants (SNVs), and structural variants (SVs), were compared to Cas9-treated and non-Cas9-treated control HSPC clones. Statistical analysis revealed no significant difference in the number of novel non-targeted indels among the samples. The median number of novel SNVs was slightly elevated in Cas9 RNP-recipient sample groups compared to baseline, but did not reach statistical significance. SVs were rare and demonstrated no clear causal connection to Cas9-mediated gene editing procedures. In sum, we found that the collective somatic mutational burden observed within Cas9 RNP-edited human HSPC clones is indistinguishable from naturally occurring levels of background genetic heterogeneity. This work was published Genes 2020.