Increasing Cas9-mediated homology-directed repair efficiency through covalent tethering of DNA repair template

Abstract

The CRISPR-Cas9 system is a powerful genome-editing tool in which a guide RNA targets Cas9 to a site in the genome, where the Cas9 nuclease then induces a double-stranded break (DSB). The potential of CRISPR-Cas9 to deliver precise genome editing is hindered by the low efficiency of homology-directed repair (HDR), which is required to incorporate a donor DNA template encoding desired genome edits near the DSB. We present a strategy to enhance HDR efficiency by covalently tethering a single-stranded oligodeoxynucleotide (ssODN) to the Cas9-guide RNA ribonucleoprotein (RNP) complex via a fused HUH endonuclease, thus spatially and temporally co-localizing the DSB machinery and donor DNA. We demonstrate up to a 30-fold enhancement of HDR using several editing assays, including repair of a frameshift and in-frame insertions of protein tags. The improved HDR efficiency is observed in multiple cell types and target loci and is more pronounced at low RNP concentrations.

Fig. 1

Selective covalent attachment of ssDNA to Cas9-PCV. a Schematic of Cas9 fused to the HUH endonuclease PCV with a covalently attached ssODN. b SDS-PAGE of Cas9 variants reacted with an Alexa 488 fluorescently labeled ssDNA containing the PCV recognition sequence. The top panel is the coomassie stained gel, and the bottom panel is the identical fluorescently imaged gel. PCV is fused to either the carboxyl (Cas9-PCV) or amino (PCV-Cas9) terminus of Cas9. Cas9-PCV (Y96F) represents catalytically inactive PCV (Y96F) fused to Cas9. c SDS-PAGE gel shift assay of Cas9 reacting with two ssDNA templates containing the PCV recognition sequence of differing lengths in a 1:1 ssDNA:Cas9 molar ratio

Fig. 2

Covalent attachment of ssODN to PCV fusions of Cas9 enhances insertion of a peptide tag. a Schematic of split luciferase insertion. The C-terminus of nanoluciferase (HiBiT) is encoded on the 200 bp ssODN along with the 5′ PCV recognition sequence and targeted to the 3′-end of GAPDH. b Assaying luminescence using different Cas9 variants when inserting HiBiT into GAPDH in HEK-293T cells. PCV is fused to either the amino (PCV-Cas9) or carboxyl (Cas9-PCV) terminus of Cas9. Transfections were performed with ssODN lacking the PCV recognition sequence (PCV- ssODN) or ssODN containing the PCV recognition sequence (PCV+ ssODN). Units are displayed in relative light units (RLU) normalized to Cas9. c Targeting the GAPDH locus in U2-OS cells. d Targeting a locus in vinculin in HEK-293T cells using an ssODN containing the PCV recognition sequence. e Fold change in RLU compared to Cas9 when varying the RNP concentration with equimolar ssODN. f HDR frequency and HDR/indel ratio from deep sequencing of the GAPDH locus. Graphs represent data from one of multiple independent experiments exhibiting similar results using distinct samples. The data with error bars are shown as mean+/− SD (n = 3). Significance calculated using 2-tailed Student’s t test: **P < 0.01, ***P < 0.001

Fig. 3

Covalent tethering enhances HDR efficiency in a fluorescent reporter cell line. a HEK-293T cells stably expressing a mutant mCherry-GFP reporter are edited by HDR through a frameshift correction, restoring mCherry activity. b Representative microscopy images of fluorescent reporter editing. Scale bars are 50 µm. c The percent of mCherry positive cells determined by flow cytometry at two different RNP concentrations using a 175 bp ssODN containing the PCV recognition sequence. d RNP transfections at 3 pmol in the presence or absence of ssODN. Data are shown as mean+/− SD (n = 3). For (c) and (d), the statistical significance of % mCherry positive cells between PCV-fusions of Cas9 and Cas9 alone was < 0.001, calculated using 2-tailed Student’s t test