Objective 1: Develop a FV-based gene therapy approach for the treatment of subjects with LAD-1. 1.1 Pre-clinical studies to evaluate the efficacy of FV in human LAD-1 CD34+ cells. We have conducted extensive pre-clinical studies to investigate the efficacy of clinical grade FV expressing the human CD18 cDNA (FV-hCD18) in HSPCs collected from subjects with a molecularly confirmed diagnosis of LAD-1. Cells were transduced ex vivo with FV-hCD18 for 16 hours. Flow cytometry of CD34+ cells cultured for 3 days after transduction demonstrated CD18+ cell surface expression in 39-42% of cells. Genetic correction of HSPCs from LAD-1 patients restored the chemotactic function of neutrophils differentiated from these progenitor cells in vitro. Transplantation of FV-hCD18-transduced LAD-1 HSPCs into immuno-deficient (NSG) mice resulted in high-level, clinically relevant gene marking levels in vivo. The average percentages of human cells expressing CD18 in the murine BM 5 months after transplantation were 36.0 3.9%. Quantitative PCR analysis of vector integrants within engrafted human cells indicated a single integration event occurred in most of long-term repopulating HSPCs. Flow cytometry-based lineage analysis of BM from mice transplanted with CD34+ cells transduced with FV-hCD18 revealed human CD18+ cells in both CD13+ myeloid and CD20+ lymphoid compartments. Using next-generation sequencing technology, a total of 101 unique integration sites were recovered in repopulating cells and revealed a polyclonal pattern of integration with no evidence of insertional mutagenesis or tumorigenicity five months after transplantation. 1.2 First-in-human clinical trial testing safety/efficacy of FV for gene therapy of patients with LAD-1. Based on pre-clinical evidence of safety and efficacy discussed above, we have designed a first-in-human phase I/II gene therapy clinical trial using FV for the gene therapy of LAD-1. A pre-IND type B meeting was held with members of the Center for Biologics Evaluation and Research (CBER) within the FDA. Regulatory review and accrual are currently on hold due to delays in production of GMP grade FV. Objective 2: Develop and evaluate safety of CRISPR/Cas9-based strategies for permanent site-specific delivery of a therapeutic gene in human HSPCs. 2.1 Evaluate off-target Cas9 activity in human HSPCs We performed CRISPR-Cas9-based genome editing in human HSPCs and assessed the acquisition of de novo somatic mutations in an unbiased, genome-wide manner via high-throughput, whole-genome sequencing (WGS) of single-cell-derived HSPC clones. Importantly, to distinguish between naturally occurring spontaneous somatic mutation, potential cell culture-induced mutagenesis, and bona fide Cas9-mediated genetic alterations, the experimental design included a parallel set of control treatments applied to samples of the same bulk HSPC population. HSPC samples were either i) mock electroporated, ii) electroporated in the absence of effector molecules, iii) electroporated with recombinant Cas9 alone, or iv) electroporated with ribonucleoprotein complexes (RNPs) consisting of Cas9 bound to single guide RNAs (sgRNAs) targeting either the cxcr4 gene or the safe harbor locus AAVS1. Experimentally manipulated cells were cloned by limiting dilution and single-cell isolates were expanded in vitro to obtain sufficient genomic DNA for WGS library preparation and high-throughput sequencing. We employed somatic variant calling algorithms to identify de novo insertions/deletions (indel) mutations, single nucleotide variants (SNVs) and structural variants (SVs) captured within the WGS datasets for each treatment group relative to the un-treated, bulk HSPC reference sequence. Consistent with previous studies investigating CRISPR-Cas9 off-target activity in mouse embryos and human iPSCs, our WGS-based analysis of potential Cas9 RNP-associated off-target mutational events in primary human CD34+ HSPCs suggests that genome-editing utilizing electroporated Cas9 RNP complexes does not result in a significant, Cas9-specific mutational burden within recipient cells. This observation has relevance to current Cas9-based ex vivo genome-editing strategies. 2.2 Homology-independent targeted integration (HITI). We have utilized a novel NHEJ-based approach to Cas9-mediated transgene knock-in, known as HITI, to achieve robust site-specific transgene integration within human CD34+ HSPCs. As proof-of-concept, a copGFP expression cassette was targeted to the genetic locus ITGB2 encoding the beta-2 integrin subunit CD18. First, a HITI donor template bearing a CMV promoter-driven copGFP reporter gene flanked by 20-nt ITGB2-specific sgRNA target sequences (designated ITGB2-ts) was constructed and packaged within recombinant adeno-associated virus serotype 6 (rAAV6) capsids. Mobilized human CD34+ HSPCs isolated from healthy donors were transduced with rAAV6-copGFP and electroporated with pre-formed, ITGB2-targeted Cas9/sgRNA ribonucleoprotein (RNP) complexes. Upon nuclear entry, the ITGB2-ts-flanked reporter cassette and the endogenous ITGB2 target gene are concomitantly cleaved by Cas9, thus promoting NHEJ-mediated transgene insertion at the site of the Cas9-induced chromosomal double-strand break. To determine the optimal time of rAAV transduction, HSPCs were transduced with rAAV6-copGFP at either 48 hr or 36 hr pre- or 0.5 hr post-RNP electroporation. Gene edited cells were cultured for up to 28 days post-electroporation and periodically sampled for flow cytometry and genomic DNA (gDNA) extraction. Transduction of HSPCs prior to electroporation resulted in enhanced cellular viability compared to post-electroporation transduction. Flow cytometry revealed efficient rAAV transduction at 4 days post-transduction (range 18-46% copGFP+ cells). The percentage of cells expressing copGFP slowly decreased over the extended culture period, stabilizing at approximately 5-10 percent of the bulk cell population at >2 weeks post-electroporation. Site-specific transgene integration was confirmed by PCR analysis of bulk cell gDNA using reporter- and flanking gene-specific primer pairs. To estimate the frequency of transgene integration, CD34+ cells were plated in a CFU assay and transplanted into NSG mice. Approximately 12% of colonies and similar percentages of human cells at 5 months after transplantation demonstrated copGFP expression. Integration junction-specific PCR analysis of gDNA derived from GFP+ colonies and post-transplant human CD45+ cells confirmed that reporter gene expression was attributable to integrated donor template sequences, as opposed to rAAV-copGFP episomes or random transgene integration events. In summary, HITI-based transgene knock-in provides an effective alternative to HDR-mediated donor template recombination in human CD34+ HPSCs. Objective 3: Develop a targeted, non-genotoxic conditioning regimen based on anti-c-MPL antibodies conjugated to immunotoxin. In collaboration with Dr. Zhirui Wang and Dr. Diogo Magnani, proof-of-concept experiments have been initiated in non-human primates (NHP). Construction of anti-c-MPL monomeric and dimeric single-chain variable fragment (scFv) fused with the diphtheria toxin fragment 390 (DT390) is underway. This recombinant toxin fragment is safer because it lacks the native toxin-binding domain, therefore, preventing internalization of toxin in off-target cells. The resulting scFv-DT390 (69.55 kDa) and svFv2-DT390 (96.5 KDa) will be produced using the yeast Pichia pastoris expression system, as previously described for anti-CD3-DT390. Evaluation of in vitro and in vivo rhesus macaque HSPC target cytotoxicity and pharmacokinetic studies will be performed in FY20. Pending success of these proof-of-concept experiments, options will be explored to develop this concept clinically.
