Gene editing as an alternative to retroviral vectors for Wiskott-Aldrich syndrome gene therapy

Resumen

One of the safest and most efficient treatments for monogenic diseases affecting the hematopoietic system is gene therapy using lentiviral vectors (LVs) and gammaretroviral vectors (GVs). However, although latest generation LVs and GVs are highly efficient and safe, both are retroviral vectors which integrate randomly in transcriptional active sites. This uncontrollable integration generates risk of insertional mutagenesis, which can produce undesirable effects on the modified cells. Gene editing has been proposed as a safer alternative, whose objective is the restoration of the endogenous levels of the wilf type gene or the ectopic insertion of a therapeutic gene in the mutated locus or in a safe locus (safe harbor). The most potent gene editing strategies are based on the use of specific nucleases that produce double strand brakes (DSBs) in a specific site in the genome. Non homologous end joining (NHEJ) or homologous recombination (HR) repair of these DSBs allow editing the genome in different ways with a final objective: restore de normal function of the mutated gene. Our final aim was to develop a gene edition tool for efficient genetic rescue of human hematopoietic stem cells (hHSCs) from Wiskott-Aldrich syndrome (WAS) patients. In this work, we have designed different WAS-specific CRISPR/Cas9 systems and compared their efficacy and specificity with homodimeric and heterodimeric WAS-specific ZFNs directed to the exactly the same region, using nucleofection and integrative-deficient lentiviral vectors (IDLVs) as delivery systems in K562 cells. Our results showed that, although the CRISPR/Cas9 system can be as efficient as the best ZFNs for gene disruption (by NHEJ); the outcome depends on the delivery system. Indeed, while both systems achieved similar results using nucleofection, IDLV delivery of ZFNs outperformed the IDLV delivered CRISPR/Cas9 systems. This is true even for our optimized CRISPR system by the co-expression of the guide RNA (gRNA) and Cas9 in the same vector (all in one system), the inclusion of the WPRE (woodchuck hepatitis virus post-transcriptional regulatory element) element or the insertion of insulators. Our studies also showed that our WAS-specific CRISPR/Cas9 system has a similar specificity to the WAS-specific heterodimeric ZFNs and is higher to the WAS-specific homodimeric ZFNs. These results were obtained thanks to the analysis of DSBs levels detected (γH2AX staining) in the edited cells with the different nucleases. Interestingly, when we analyzed the efficacy and specificity of the different systems for genome edition of the WAS locus by HR (insertion of a expression cassette in the WAS locus), we observed a similar pattern to that obtained with NHEJ; heterodimeric ZFNs and CRISRP/Cas9 system show a similar specificity, achieving around a 80% of donor DNA integration in the WAS locus (in target). However, homodimeric ZFNs show a low specificity, around a 55% of donor insertion outside the target site. Data that confirmed the results obtain by the analysis of specificity by γH2AX staining. All the first analyses were done in K562 cells with the purpose of optimization and comparison in an easier cellular model. However, the target cells for WAS gene therapy are hematopoietic stem cells (HSCs) and/or T cells. We decided to study the efficiency on the ZNFs and CRISPR/Cas9 system to edit WAS in these cells type. However, none of these system were able to achieve the desirable levels of cutting in the WAS locus in HSC or T cells. In the last part of this thesis, we decided to study other delivery system for CRISPR. In collaboration with Els Verhoeyen group and Theophile Ohmann group, VLPs (virus like particles) or nanoblades were developed against the WAS locus. Nanoblades derived from MLV and HIV-1 were produced, incorporating Cas9 and gRNA and pseudotyped with different envelops. The WAS-nanoblades based on the MLV pseudotyped with the modified Baboon endogenous retroviral (BRL) and VSVg envelopes were the ones that presented the best efficiency, being able to edit around a 30% of the WAS locus both in HSC cells and T cells. There is still need to investigate if the ZFNs-nanoblades could improve the efficiency and specificity of the Cas9-nanoblades. Nevertheless, the CRISPR/Cas9 nanoblades have shown to be a new and strong system for future therapeutic used, due to their high efficiency in editing the genome in primary cells such as HSC and T cells.