Small extracellular vesicles (sEVs)-based gene delivery platform for cell-specific CRISPR/Cas9 genome editing

Abstract

Small extracellular vesicles (sEVs) are naturally occurring vesicles that have the potential to be manipulated to become promising drug delivery vehicles for on-demand in vitro and in vivo gene editing. Here, we developed the modular safeEXO platform, a prototype sEV delivery vehicle that is mostly devoid of endogenous RNA and can efficaciously deliver RNA and ribonucleoprotein (RNP) complexes to their intended intracellular targets manifested by downstream biologic activity. We also successfully engineered producer cells to produce safeEXO vehicles that contain endogenous Cas9 (safeEXO-CAS) to effectively deliver efficient ribonucleoprotein (RNP)-mediated CRISPR genome editing machinery to organs or diseased cells in vitro and in vivo. We confirmed that safeEXO-CAS sEVs could co-deliver ssDNA, sgRNA and siRNA, and efficaciously mediate gene insertion in a dose-dependent manner. We demonstrated the potential to target safeEXO-CAS sEVs by engineering sEVs to express a tissue-specific moiety, integrin alpha-6 (safeEXO-CAS-ITGA6), which increased their uptake to lung epithelial cells in vitro and in vivo. We tested the ability of safeEXO-CAS-ITGA6 loaded with EMX1 sgRNAs to induce lung-targeted editing in mice, which demonstrated significant gene editing in the lungs with no signs of morbidity or detectable changes in immune cell populations. Our results demonstrate that our modular safeEXO platform represents a targetable, safe, and efficacious vehicle to deliver nucleic acid-based therapeutics that successfully reach their intracellular targets. Furthermore, safeEXO producer cells can be genetically manipulated to produce safeEXO vehicles containing CRISPR machinery for more efficient RNP-mediated genome editing. This platform has the potential to improve current therapies and increase the landscape of treatment for various human diseases using RNAi and CRISPR approaches.

Keywords: CRISPR; Cas; Delivery; Genome Editing; sEVs.

Figure 1

Schematic representation of safeEXO-CAS sEVs.

Figure 2

SafeEXO-CAS characterization and preparation. A Percentage change of total exo-RNA (ng) normalized based on the protein in sEVs, isolated from siRNA silenced hnRNPA2B1, ALIX, and Drosha THP-1 monocytes. B Number of sEVs produced per μg of sEVs protein from hnRNPA2B1, ALIX, and Drosha knockout THP-1 monocytes. C-D Flow cytometric analysis of sEVs from hnRNPA2B1, ALIX, and Drosha knockout for the presence of sEV-enriched markers CD63 and CD81. SafeEXO-CAS sEVs were labeled with anti-CD81 or anti-CD63 magnetic beads and the levels of CD81 and CD63 were quantified by flow cytometry.

Figure 3

SafeEXO-CAS characterization and uptake. A Western blot analysis of sEV markers CD63 and TSG101, along with CAS9 and GRP94 (negative control) in safeEXO-CAS sEVs from THP-1 monocyte producer cells with and without ALIX knockout. B Size of safeEXO-CAS sEVs quantified by NanoSight. C Transmission electron micrograph of safeEXO-CAS sEVs. D Fluorescence microscopy of THP-1 cell line co-cultured with PKH26 and Di-8-ANEPPS labeled safeEXO-CAS after 1 hour. e Flow cytometry analysis of Di-8-ANEPPS labeled safeEXO-CAS uptake in THP-1 cell line after 1h co-culture. Experiments were repeated twice and data are represented as mean ± standard deviation (SD). ****p ≤ 0.0001, One-way ANOVA was used for the comparison of multiple groups, and Student's T test was used for pairwise comparison.

Figure 4

SafeEXO-CAS-mediated non-homologous end joining (NHEJ) genome editing. A Fluorescence microscopy and B FACS analysis of GFP expressing HEK293T cells treated with safeEXO-CAS, GFP-sgRNA/CAS9 ribonucleoprotein (RNP), and safeEXO-CAS loaded with GFP-sgRNA. C T7 endonuclease assay against the EMX1 in Cas9 expressing THP-1 cells and in THP-1 cells. Cells were treated with different concentrations of safeEXO with CAS9/EMX1 RNP (200ug and 100ug) or safeEXO-CAS loaded with EMX1-sgRNA (200ug and 100ug). EMX1-sgRNA plasmid and EMX1-sgRNA as an RNP were used as positive control. D The percentages of indel induction in EMX1 gene were quantified in negative control, EMX1-sgRNA plasmid, safeEXO loaded with CAS9/EMX1 RNP and safeEXO-CAS loaded with EMX1-sgRNA based on deep sequencing of EMX1 locus.

Figure 5

SafeEXO-CAS-mediated homology-directed repair (HDR) editing. A Schematic representation of sEVs HDR experimental plan in HEK293T cells. B SafeEXO-CAS sEVs co-loaded with sgRNA targeting human DDX3 at its start codon and donor ssDNA encoding the FLAG-tag protein surrounded by homology arms to the DDX3 start codon (top). SafeEXO-CAS sEVs were also co-loaded with siRNAs targeting DCLRE1C and XRCC5 to prevent NHEJ (bottom). HEK293T cells were treated with sEVs containing three different concentrations (0.5uM-3uM) of ssDNA donor template. The FLAG-tag insertion was confirmed using the forward primer from FLAG-tag and reverse primer from the DDX3 locus. C HEK293T cells were treated with safeEXO-CAS sEVs co-loaded with sgRNA targeting human GFP at its start codon and a ssDNA template encoding the FLAG-tag protein and bearing homology arms to the GFP start codon (top). SafeEXO-CAS sEVs were also co-loaded with siRNAs targeting DCLRE1C and XRCC5 to prevent NHEJ (bottom). HEK293T cells were treated with sEVs containing two different concentrations (0.1uM-3uM) of ssDNA donor template. The FLAG-tag insertion was confirmed using the forward primer from FLAG-tag and reverse primer from GFP locus.

Figure 6

A Schematic representation of in vitro and in vivo safeEXO-CAS-ITGA6 lung targeting. B Flow cytometry analysis using an ITGA6 (CD49f) antibody of safeEXO-CAS-ITGA6 and untargeted safeEXO-CAS. C Flow cytometry analysis of the fluorescently labeled safeEXO-CAS-ITGA6 uptake by primary lung epithelial cells, monocytes and unstained cells after 6h coculture. D ExoGlow labeled safeEXO-CAS-ITGA6 sEVs or untargeted safeEXO-CAS sEVs uptake by different mouse organs (brain, lung, heart, kidney, liver and stomach) using optical imaging of organs 15 min after sEVs injection. E Fold change of fluorescence intensity normalized to organ area from different organs (n=8 total). F Percentage of sEVs uptake was quantified by flow cytometry in lung epithelial cells (EpCAM⁺) of mice receiving safeEXO-CAS-ITGA6 or untargeted safeEXO-CAS control sEVs (n=6 per group). Data are represented as mean ± standard deviation (SD). MFI, mean fluorescent intensity, *p ≤ 0.05, Student's T test was used for pairwise comparison.

Figure 7

A Schematic representation of safeEXO-CAS-ITGA6 mediated EMX1 editing in vivo. B Percentage of indel frequency in EMX1 in lung and C liver (n=8 per group). D Representative immune fluorescence of EMX1 expression in lung tissue of safeEXO-CAS-ITGA6 and safeEXO-CAS. E-J Percentage of immune subtypes frequency in CD8⁺T cells, B cells, monocytes, NK cells, neutrophils, and CD⁺4 T cells (n=6 per group). Data are represented as mean ± standard deviation (SD).