A viable alternative for editor delivery

In this issue of Blood, Vavassori et al demonstrate that ex vivo lipid nanoparticle (LNP)-mediated delivery of gene editors to T cells and hematopoietic stem and progenitor cells (HSPCs) results in lower cytotoxicity, reduced stress responses, and improved cell yield compared with standard electroporation of editing agents.¹ 

Monogenic hematological disorders such as hemoglobinopathies and immunodeficiencies afflict millions of people worldwide, diminishing their life expectancy and quality of life. Ex vivo gene editing of hematopoietic cells has curative potential in these inherited disorders as well as a host of acquired conditions, with the first CRISPR-based therapy for sickle cell disease under consideration for regulatory approval and an in-human proof-of-concept established for the use of gene-edited chimeric antigen receptor expressing T cells.^2,3 Ex vivo transfection of gene-editing agents is most commonly accomplished via electroporation, in which an electric pulse is used to create temporary pores in the cell membrane through which exogenous materials can enter. Despite its utility, electroporation is also associated with significant cytotoxicity secondary to oxidative damage to membrane lipids, ion imbalances, and mitochondrial damage.⁴ This has motivated the exploration of alternative delivery methods for ex vivo editing of hematopoietic cells, including the use of membrane-disruptive polymers⁵ and peptides.^6,7 LNPs have emerged as a leading nonviral nanoparticle platform for therapeutic genome editor delivery. LNPs are composed of 4 key components: an ionizable lipid to promote endosomal escape following particle uptake, a cationic lipid to assist in nucleic acid cargo complexation, cholesterol, and a polyethylene glycol–lipid conjugate to improve stability; all are combined with nucleic acid cargo generally via microfluidic mixing. LNPs have shown clinical safety and efficacy for in vivo delivery of gene-editing nucleic acids (Cas9 mRNA and guide RNA) to the liver⁸ but remain little explored for delivery to hematopoietic cells, including for ex vivo applications.

In this study, Vavassori et al assess the aggregate impacts of electroporation, response to the editing agents including the DNA donor template, and nuclease-induced genomic double-strand breaks (DSBs) on HSPC and T-cell clonogenicity, viability, and phenotype by leveraging LNPs as an alternative delivery method to electroporation. The authors first evaluate the effects of electroporation on human CD4⁺ T cells by cotreating T cells with Cas9 ribonucleoprotein (RNP) and an adeno-associated virus (AAV)-based donor template. Transcriptomic and proteomic analysis comparing untreated cells to electroporated cells (even those only electroporated, without editing agent treatment) showed an upregulation of genes and proteins associated with apoptosis, inflammation, and DNA damage, specifically the p53 pathway, which promotes cell cycle arrest and death. Next, the authors compared the efficiency of LNP-mediated mRNA-encoded Cas9 delivery to electroporation of RNP. Interestingly, although the editing efficiency measured in LNP-treated T cells was generally lower, the yield of edited cells was higher secondary to the lower cytotoxicity of LNPs. This relationship held true for both nonhomologous end joining and homology-directed repair editing outcomes at multiple gene targets. As compared with electroporation, LNP treatment was associated with an antiapoptotic rather than proapoptotic gene expression signature, in alignment with the reduced cytotoxicity of LNPs. Similar results were observed in both cord and peripheral blood–derived HSPCs. Despite lower editing efficiencies compared with electroporation, LNP-mediated Cas9 delivery yielded comparable or greater yields in edited cells due to their lower toxicity. On top of relative reductions in apoptosis- and inflammation-related protein levels, LNP-treated HSPCs also demonstrated improved clonogenic potential ex vivo and noninferior repopulation in vivo compared with electroporated cells.

The data presented by the authors suggest that along with inducing upregulation of apoptosis- and inflammation-related genes, electroporation additionally activates p53-mediated DNA damage repair responses that create further cytotoxic effects. However, unlike in HSPCs,⁹ transient inhibition of p53 in CD4⁺ T cells via coelectroporation of GSE56 mRNA did not ameliorate cytotoxicity, leaving open the possibility that other, yet-to-be-characterized pathways may also play a role in T-cell responses to electroporation. The authors also note that LNP-treated T cells and HSPCs show a cholesterol-mediated transcriptional response, which they posit is related to intracellular loading of exogenous lipids and subsequent induction of an inflammatory state. This effect points toward a new mode of toxicity that must be accounted for when LNPs are used for ex vivo editor delivery. The authors address this by replacing LNP-containing media with fresh media after a brief (<24 hours) incubation. Importantly, the authors also note that the induction of DSBs by Cas9 nuclease editing and exposure to AAV as donor template are key additional drivers of p53-dependent DNA damage responses beyond the delivery method itself. Next-generation editing technologies, such as base editing and prime editing, that primarily rely on single-strand breaks and do not necessitate AAV coinfection may further mitigate these undesired reactions in edited cells.

Altogether, the work of Vavassori et al puts forward LNPs as a “more viable” alternative for ex vivo gene editing of hematopoietic cells that holds promise in a variety of potential clinical applications (see figure). By offering a less toxic approach as compared with electroporation, LNPs have potential to improve the safety and efficiency of ex vivo cell engineering. Further development of LNP-based delivery systems for in vivo hematopoietic cell gene editing, for which murine proof of concept has already been established,¹⁰ might ultimately unlock the full therapeutic potential of hematopoietic cell gene editing, though challenges may include specifically targeting hematopoietic cells of interest, detargeting hepatic and reticuloendothelial sinks, maximizing editing efficiency and precision, minimizing immunogenicity, allowing repeated dosing, and simplifying manufacturing. Enabling delivery of more diverse cargo including proteins and DNA templates might further boost the specificity and scope of therapeutic edits. Until this comes to fruition, ex vivo gene editing using LNPs delivering RNAs may be a valuable stepping stone toward the continued translation of gene-edited hematopoietic cell therapies into the clinic.

Conflict-of-interest disclosure: D.E.B. is an inventor of patents related to therapeutic gene editing and has consulted for Kytopen. F.E. declares no competing financial interests.