Self-inactivating, all-in-one AAV vectors for precision Cas9 genome editing via homology-directed repair in vivo

Abstract

Adeno-associated virus (AAV) vectors are important delivery platforms for therapeutic genome editing but are severely constrained by cargo limits. Simultaneous delivery of multiple vectors can limit dose and efficacy and increase safety risks. Here, we describe single-vector, ~4.8-kb AAV platforms that express Nme2Cas9 and either two sgRNAs for segmental deletions, or a single sgRNA with a homology-directed repair (HDR) template. We also use anti-CRISPR proteins to enable production of vectors that self-inactivate via Nme2Cas9 cleavage. We further introduce a nanopore-based sequencing platform that is designed to profile rAAV genomes and serves as a quality control measure for vector homogeneity. We demonstrate that these platforms can effectively treat two disease models [type I hereditary tyrosinemia (HT-I) and mucopolysaccharidosis type I (MPS-I)] in mice by HDR-based correction of the disease allele. These results will enable the engineering of single-vector AAVs that can achieve diverse therapeutic genome editing outcomes.

Fig. 1. In vitro design and Nanopore sequencing validation of dual-sgRNA AAV:Nme2Cas9 vectors.

a Schematic of four different designs of dual-sgRNA AAV:Nme2Cas9 vector plasmids containing two sgRNA expression cassettes. b Alkaline agarose gel electrophoresis of viral DNA isolated from the four dual-sgRNA AAV:Nme2Cas9 vectors presented in (a) after packaging in AAV8. DNA size markers are indicated. Red arrows indicate full-length non-truncated viral genomes. This analysis was performed once. c, d Alignments of Nanopore sequencing reads representing DNAs extracted from Dual-sgRNA:Design 1 (c) and Dual-sgRNA:Design 4 (d) rAAV8 vectors. Diagrams of references displaying notable construct domains are shown above each alignment. Regions upstream and downstream of the vector sequences (beyond ITRs) represent the bacterial plasmid backbone. The alignment summary shows read coverage. Read alignments displayed in IGV show matching nucleotides in gray. Nucleotide mismatches are depicted in green (A), red (T), blue (C), or brown (G); gaps are in black dashes; and inserts are in purple. Start and end position alignment counts (orange and blue traces, respectively) are displayed below each alignment. The percentages of read start and end peaks are displayed for prominent peaks. e Schematic diagram of the mouse Hpd gene. The first target site is represented by half arrowhead in red (gRNA-I) and the second target site is represented in green (gRNA-II). SMRT sequencing primers are also highlighted in gray, while UDiTaS primers are in orange.

Fig. 2. In vivo gene editing using dual-sgRNA rAAV:Nme2Cas9 vectors.

a In vivo experimental plan to validate rAAV:Nme2Cas9 Dual-sgRNA:Designs 1 and 4 to disrupt the Hpd gene by tail vein injections of AAV8 vectors in adult Fah^neo/neo and C57BL/6 mice. b Complete rescue of body weights after Dual-sgRNA:Designs 1 and 4 in Fah^neo/neo mice. c Quantification of editing events in C57BL/6 (left) and Fah^neo/neo (right) mice by SMRT sequencing analysis showing efficiencies of segmental deletion (orange) and inversion (blue) outcomes. d Bar graph showing the percentages of indels recorded in full-length (3.6 kb), UMI-corrected SMRT reads after Hpd editing by AAV8 delivery of Dual-sgRNA:Designs 1 and 4 in C57BL/6 (left) and Fah^neo/neo (right) mice. The plot indicates indels recorded only at gRNA-I (in red), indels only at gRNA-II (in dark green), and indels at both gRNA-I and gRNA-II (yellow) as measured by SMRT sequence analysis. e rAAV fragment integration as detected by SMRT sequencing analysis. f Quantitation of editing events in UMI-corrected UDiTaS analysis reads after Hpd editing by AAV8 delivery of Dual-sgRNA:Design 4 in C57BL/6 and Fah^neo/neo mice showing mean efficiencies of segmental deletion, inversion, and AAV fragment integration. g mRNA levels from RNA-seq (transcripts per million) of Hpd wild-type mRNA in the livers of C57BL/6 mice and Fah^neo/neo mice. h Volcano plots showing differentially expressed genes with false discovery rate ≤5% after adjusting for multiple comparisons (red; analysis detailed in Methods) between Dual-sgRNA:Design 1 and PBS-treated Fah^neo/neo mice (left) and Dual-sgRNA:Design 4 and PBS-treated Fah^neo/neo mice (right). i Total HPD protein knockout as shown by anti-HPD Western blot using total protein collected from mouse liver homogenates. This analysis of samples from multiple mice was performed once. j Representative images of immunostaining for HPD in liver tissues in Fah^neo/neo mice injected with PBS (left) or Dual-sgRNA:Design 4 (right). Scale bar is 100 μm. All mice in the cohort were examined once by this method (see Supplementary Fig. 2). Data were presented as mean values ± s.e.m. Sample size in panels b–g: (n = 3 PBS-injected C57BL/6 mice; n = 4 Dual-sgRNA.Design 1 or 4 injected C57BL/6 mice; n = 1 in pre-NTBC Fah^neo/neo cohort; n = 7 in post-NTBC Dual-sgRNA.Design 1 or 4 injected Fah^neo/neo cohort).

Fig. 3. In vitro design and validation of all-in-one, self-inactivating rAAV:HDR vectors.

a Schematic of four different vector designs of rAAV:HDR constructs containing an sgRNA expression cassette, Nme2Cas9, and a donor DNA (<500 bp) with and without flanking sgRNA target sites. b Agarose gel electrophoresis (1% agarose) of linearized AAV:HDR plasmids. Asterisks indicate self-inactivated AAV:HDR plasmids that were cloned using anti-CRISPR-protein-expressing E. coli DH5α. DNA size marker size is 1 kb. Multiple plasmids were analyzed in this fashion three separate times, with consistent results. c Schematic of conventional E. coli DH5α and the recombined, anti-CRISPR-protein-expressing derivative used to successfully clone the self-targeting rAAV:HDR:cleaved plasmids. d Efficiencies of NHEJ and HDR events depicted as percentages of mCherry- and GFP-positive cells, respectively, obtained after transfection of AAV:Nme2Cas9:sgRNA and dsDNA GFP donor in trans), or of rAAV:HDR constructs in TLR-MCV1 HEK293T cells. The sample size represents independent transfection experiments (n = 2 of Nme2Cas9, sgRNA, and dsDNA donor in trans delivery; while n = 4 biological replicates of in cis delivery of AAV:HDR plasmids Designs A, B, C, and D). e Schematic of modified rAAV packaging transfection system using anti-CRISPR protein plasmid to block self-targeting by Nme2Cas9 expression in packaging cells during production.

Fig. 4. rAAV:HDR vectors rescue liver disease phenotypes in a mouse model of HT-I.

a Schematic of rAAV:HDR:uncleaved and -cleaved vectors (left) and control non-cognate donor or spacer vectors (right) to correct the Fah point mutation in Fah^PM/PM mice (top). b In vivo experimental regimen to correct the Fah mutation by injection of AAV8 through the tail vein in adult Fah^PM/PM mice. c Complete rescue of average body weight loss in Fah^PM/PM mice in treated cohorts injected with rAAV:HDR:uncleaved (red) and -cleaved (green), as compared to control cohorts injected with PBS (black), rAAV:FahDonor:ncSpacer (brown) or rAAV:ncDonor:FahSpacer (blue). Data shown represent the average body weight of each cohort. d Bar graph showing the percentage distribution of NHEJ, HDR, and imprecise NHEJ:HDR mix at Fah in livers of mice 6 weeks after NTBC withdrawal, as measured by NGS sequencing of PCR amplicons from genomic DNA. e Representative images of immunostaining for FAH in liver tissues of negative control and treated cohorts, as indicated below each panel. Scale bar, 100 μm. All mice in each cohort were examined once by this method (see Supplementary Fig. 3). f Levels of Fah mRNA from RNA-seq (transcripts per million) in the livers of Fah^PM/PM mice. g rAAV copy numbers in HT-I mouse liver tissues. rAAV copies are significantly reduced in rAAV:HDR:cleaved cohorts after NTBC withdrawal using the [U6<>FahDonor] primer pair (left) and [FahDonor<>U1a] primer pair (right). h qRT-PCR analyses with total RNA showing significantly reduced Nme2Cas9 mRNA and sgRNA expression in the rAAV:HDR:cleaved cohort after NTBC withdrawal. Data were presented as mean values ± s.e.m. Sample sizes in panels c, d, f–h: (n = 3 in PBS, rAAV:FahDonor:ncSpacer, rAAV:ncDonor:FahSpacer and pre-NTBC withdrawal rAAV:HDR:cleaved and -uncleaved cohorts; n = 7 in post-NTBC withdrawal rAAV:HDR:cleaved and -uncleaved cohorts). p values are calculated using Student’s t-test (two-sided).

Fig. 5. Reduced clinical disease manifestation in a mouse model of mucopolysaccharidosis type I treated with rAAV:HDR vectors.

a Schematic of rAAV:HDR:uncleaved, -cleaved, and control non-cognate spacer vectors to correct the Idua point mutation in MPS-I Idua^W392X mice. b In vivo experimental regimen to correct the Idua mutation by injection of rAAV9 through the facial vein in neonate MPS-I mice. c IDUA specific activity in liver lysates of healthy mice compared to rAAV:IduaDonor:ncSpacer mice (negative control) and treated rAAV:HDR uncleaved/cleaved cohorts. The dashed lines indicate 0.2% (therapeutic threshold) and 3% of the IDUA activity detected in lysates from WT mice. d Level of glycosaminoglycan (GAG) accumulation in the livers of the same mice shown in (c). e β-d-glucuronidase specific activity in liver lysates of healthy mice compared to rAAV:IduaDonor:ncSpacer mice (negative control) and treated rAAV:HDR uncleaved/cleaved cohorts. f Bar graph showing the percentage of NHEJ, HDR, and imprecise NHEJ:HDR mix at Idua in the liver as measured by NGS sequencing of PCR amplicons from genomic DNA. g qRT-PCR data showing increase in the relative Idua mRNA in the liver normalized to Gapdh mRNA. h AAV copy numbers in MPS-I mouse liver tissues using the [sgRNA<>U6] [U6<>IduaDonor], [IduaDonor<>U1a], and [U1a<>Nme2Cas9] primer pairs. i, j Animals treated with rAAV:HDR:cleaved vectors exhibit reduced Nme2Cas9 mRNA (i) and sgRNA (j) expression levels as measured by qRT-PCR, though reductions do not reach statistical significance. Data in panels c–h are presented as mean values ± s.e.m. (n = 3 mice in the WT cohort, n = 4 mice in the rAAV:IduaDonor:ncSpacer and rAAV:HDR:uncleaved cohorts, and n = 5 mice in the rAAV:HDR:uncleaved cohort). p values are calculated using Student’s t-test (two-sided).