CRISPR/Cas systems present a powerful gene-editing tool with the potential for widespread therapeutic use; however, current methods of in vivo delivery such as adeno-associated viruses (AAV) may stimulate an immune response, creating the need for an alternative for delivery of CRISPR/Cas9. Exosomes are small vesicles that are released by cells and serve as a delivery system for RNA, proteins, and various molecules to other cells. The focus of this project was to use exosomes as a delivery system for Cas9, exploiting their high uptake by target cells and their ability to avoid the immune system in vivo. Porcine fetal fibroblasts (PFF) were grown to 80% confluency; after 48h, exosomes were isolated and concentrated from conditioned media by filtration with a 0.22-μm filter followed by 100-kDa molecular weight cutoff filter. Transmission electron microscopy, Western blotting for presence of CD81, and an uptake assay for exosomes stained with the lipophilic dye DiI (Invitrogen/Thermo Fisher Scientific, Waltham, MA, USA) were used to characterise isolated exosomes, and average particle size was evaluated by NanoSight (Salisbury, United Kingdom). After characterisation, exosomes were loaded with Cas9 (PNA Bio, Newbury Park, CA, USA) using sonication, incubation with saponin, or extrusion. For each method of loading, 1.0×1011 exosomes and 500ng of Cas9 were used. For sonication, exosomes and Cas9 were sonicated 4 times: 4s on/2s off, left on ice for 2min, and then repeated for 4 more cycles. Loaded exosomes were then incubated at 37°C for 20min. For incubation with saponin, 100μL of 0.6% saponin solution was made in PBS, mixed with exosomes and Cas9, and then incubated on a shaker at 800 rpm for 20min. For extrusion, exosomes and Cas9 were extruded (Avanti Polar Lipids, Alabaster, AL, USA) 10, 15, or 20 times through a 0.22-μm filter. To evaluate efficiency of Cas9 loading into exosomes, loaded exosome samples were split in half, with one-half receiving a proteinase K digest (100μg mL−1) to remove free Cas9 and the other receiving no treatment. Proteinase K-treated and untreated samples were then compared side by side on Western blot staining for Cas9. ImageJ software (National Institutes for Health, Bethesda, MD, USA) was used to quantify band intensity and loading efficiency. With optimal conditions, our preliminary results show loading efficiency for sonication and saponin to be 16.7 and 19.2%, respectively, whereas loading by extrusion was undetectable. For CRISPR/Cas targeting, transgenic PFF carrying one copy of H2B-GFP were used to test delivery of ribonucleotide protein complex (RNP). To verify efficiency of the guide (g)RNA targeting green fluorescent protein (GFP), cells were nucleofected with Cas9 and gRNA. The DNA was extracted, PCR amplified, and sequenced (Eton Bioscience, San Diego, CA, USA) and then evaluated for indels with TIDE, resulting in a 53.2% cleavage efficiency. Next, exosomes will be loaded with RNP to knockout GFP in H2B-GFP cells, and targeting efficiency will be evaluated by flow cytometry and TIDE. We hypothesise that based on loading efficiency and target cell uptake, exosomes will present a safe and efficient method for in vitro and in vivo delivery of Cas9.
The financial support of the Comparative Medicine Institute is gratefully acknowledged.
CRISPR-Cas9 in gene therapy: much control on breaking, little control on repairing