Lipid nanoparticles enable high efficiency CRISPR HDR mediated gene insertions in primary human T cells

Stable gene insertion using non-viral delivery is essential to unlocking the next generation of safe and accessible cell and gene therapies. Currently, viral vectors carry high manufacturing costs and insertional mutagenesis concerns, while electroporation (EP) affects cell yield and viability. Lipid nanoparticles (LNPs) can be a promising alternative option for engineering cells for persistent expression of target genes, as they offer a favorable safety profile and are both cost-effective and scalable. In this work, we mapped out the critical parameters governing homology-directed repair (HDR) in primary T cells using LNP-mediated cargo delivery. Cas9 mRNA, a chemically synthesized guide RNA (sgRNA), and an ~100 nt single-stranded donor oligonucleotide (ssODN) were encapsulated within a novel LNP composition using a scalable production platform. A haemagglutinin (HA) epitope tag was knocked in at the CD5 locus as an easily quantified read-out for parameter optimization. CD3+ primary T cells from healthy donors were cultured in well-plates and the CRISPR LNPs added to the media in a one-step process, without further cell manipulation. Various parameters were identified and systematically varied, including (and not limited to) the length of cell activation, cell density, nucleic acid dose and the RNA/DNA molar ratios. Through multiple rounds of optimization, LNPs achieved on average 31 ± 7% HDR in n=5 T cell donors, detected through dual CD5/HA flow cytometry 4 days post-LNP administration. Viability of the cells remained high at 96 ± 5% at the time of HDR detection, relative to untreated controls. The aforementioned results reflect no added enhancers; however, when we tested various small molecules, such as NHEJ inhibitors, HDR rates further improved to over 50% in primary T cells. Finally, we compared the optimized LNP protocol to EP which resulted in similar HA+ frequencies. However, most notably, the yield of viable edited cells by LNP was an order of magnitude higher than EP owing to improved cell viability and proliferation. All together, this data demonstrates how LNPs can achieve clinically relevant knock-in frequencies and showcases the benefits of LNPs as a non-viral alternative for gene insertion. This approach for parameter optimization is applicable for a diverse set of therapeutic loci, providing a framework for the rapid application of LNPs to enable the next generation of T cell therapies.