We report a genome-editing strategy to correct compound heterozygous mutations, a common genotype in patients with recessive genetic disorders. Adeno-associated viral vector delivery of Cas9 and guide RNA induces allelic exchange and rescues the disease phenotype in mouse models of hereditary tyrosinemia type I and mucopolysaccharidosis type I. This approach recombines non-mutated genetic information present in two heterozygous alleles into one functional allele without using donor DNA templates.
Subscribe to Journal
Get full journal access for 1 year
only $8.25 per issue
All prices are NET prices.
VAT will be added later in the checkout.
Tax calculation will be finalised during checkout.
Rent or Buy article
Get time limited or full article access on ReadCube.
All prices are NET prices.
Sequence Read Archive
Bergman, A.J. et al. Hum. Mutat. 12, 19–26 (1998).
St-Louis, M. & Tanguay, R.M. Hum. Mutat. 9, 291–299 (1997).
Li, P., Wood, T. & Thompson, J.N. Genet. Med. 4, 420–426 (2002).
Choi, P.S. & Meyerson, M. Nat. Commun. 5, 3728 (2014).
Blasco, R.B. et al. Cell Rep. 9, 1219–1227 (2014).
Torres, R. et al. Nat. Commun. 5, 3964 (2014).
Vanoli, F. et al. Proc. Natl. Acad. Sci. USA 114, 3696–3701 (2017).
Spraggon, L. et al. J. Pathol. 242, 102–112 (2017).
Grompe, M. Semin. Liver Dis. 21, 563–572 (2001).
Bunge, S. et al. Biochim. Biophys. Acta 1407, 249–256 (1998).
Wang, D. et al. Mol. Genet. Metab. 99, 62–71 (2010).
Ghezraoui, H. et al. Mol. Cell 55, 829–842 (2014).
Simsek, D. et al. PLoS Genet. 7, e1002080 (2011).
Yin, H. et al. Nat. Biotechnol. 32, 551–553 (2014).
Yin, H. et al. Nat. Biotechnol. 34, 328–333 (2016).
Yang, Y. et al. Nat. Biotechnol. 34, 334–338 (2016).
Wang, M.J., Chen, F., Lau, J.T.Y. & Hu, Y.P. Cell Death Dis. 8, e2805 (2017).
Nussenzweig, A. & Nussenzweig, M.C. Cell 141, 27–38 (2010).
Borel, F. et al. Mol. Ther. 25, 2477–2489 (2017).
Nygaard, S. et al. Sci. Transl. Med. 8, 342ra79 (2016).
Grompe, M. et al. Genes Dev. 7, 2298–2307 (1993).
Aponte, J.L. et al. Proc. Natl. Acad. Sci. USA 98, 641–645 (2001).
Yokoyama, T. et al. Nucleic Acids Res. 18, 7293–7298 (1990).
Clarke, L.A. et al. Hum. Mol. Genet. 6, 503–511 (1997).
Bartlett, J.S. et al. Proc. Natl. Acad. Sci. USA 93, 8852–8857 (1996).
Gao, G. & Sena-Esteves, M. in Molecular Cloning edn. 4 (eds. Green, M.R. & Sambrook, J.R.) 1209–1330 (Cold Spring Harbor Laboratory Press, Cold Spring Harbor, 2012).
Zhang, Z., Theurkauf, W.E., Weng, Z. & Zamore, P.D. Silence 3, 9 (2012).
Wang, D. et al. Mol. Genet. Metab. 105, 116–125 (2012).
Sánchez-Rivera, F.J. et al. Nature 516, 428–431 (2014).
Goecks, J., Nekrutenko, A., Taylor, J. & Galaxy Team. Genome Biol. 11, R86 (2010).
Giardine, B. et al. Genome Res. 15, 1451–1455 (2005).
Blankenberg, D. et al. Curr. Prot. Mol. Biol. 89, 19.10 (2010).
Robinson, J.T. et al. Nat. Biotechnol. 29, 24–26 (2011).
We thank M. Grompe (Oregon Health & Science University) for providing the homozygous Fahneo/neo and FahPM/PM mice, L. Luo (Stanford University) for providing the pGT and pTG plasmids (Addgene, #36885 and #36887), and E. Hendrickson (University of Minnesota) for providing the HCT116 and LIG4−/− cells. This work was supported by grants from the National Institutes of Health to G.G. (1P01AI100263, 1R01NS076991, 5P01HD080642, R01AI12135) and to W.X. (DP2HL137167, P01HL131471), a grant from the National High Technology Research and Development Program (“863” Program) of China to G.G. (2012AA020810), and a grant to W.X. from Hyundai Hope on Wheels. The authors thank H. Yin and the members of Gao laboratory and Xue laboratory for helpful discussions.
D.W. and G.G. have submitted a patent application concerning the methodology described in this study. P.D.Z. and G.G. are scientific co-founders of Voyager Therapeutics and hold equity in the company. P.D.Z. and G.G. are inventors on patents with potential royalties licensed to Voyager Therapeutics and other biopharmaceutical companies.
Integrated supplementary information
(a) Schematics showing the constructs packaged into rAAV vectors. U1a: murine U1a small nuclear RNA promoter; RBGpA: rabbit beta-globin polyadenylation signal; CMV/CB: cytomegalovirus enhancer fused with chicken beta-actin promoter. (b) T7EI nuclease assay to detect gene editing in GreenGo cells by three sgRNAs targeting Fah. Gray arrow head indicates T7EI cleavage bands that reflect gene editing. Black arrow head indicates uncleaved PCR products. Open arrow heads indicate T7EI cleavage events at a dinucleotide polymorphic site within the Fah gene PCR products, which are non-specific to the targeted gene editing. SgFah_1 shows the strongest signal of gene editing and therefore is chosen for the following experiments. Full-length gel image is shown in Supplementary Fig. 13b. (c) Sanger sequencing reads of TOPO clones of PCR products derived from the gene editing experiment using GreenGo cells and sgFah_1. WT sequence is shown in bold at the top, and the protospacer adjacent motif (PAM) of sgRNA is boxed (antisense for Fah). The number of TOPO clones obtained is indicated by “xN” if it is more than one. Deletion is indicated by a dash. (d) T7EI nuclease assay showing gene editing in mouse liver at the Fah gene (left gel image) and Aspa gene (right gel image). Mice 1 and 2 were treated with rAAV9-SpCas9 and scAAV8-sgFah, and mice 3 and 4 were treated with rAAV9-SpCas9 and scAAV8-sgAspa. All mice were compound heterozygous HT1 mice that were treated at postnatal day 1 (P1), maintained on NTBC water, and euthanized at five weeks old. Full-length gel image is shown in Supplementary Fig. 13c. (e) Stacked histogram showing the counts of wild-type (WT) and indel reads by TOPO cloning and Sanger sequencing of the amplicons used in the T7EI assay. Each stacked bar represents one mouse. (f) Sanger sequencing reads of the TOPO clones shown in (e). Underlined sequence indicates insertion in between, and the inserted DNA sequence is shown in parenthesis. The sequences of some long insertions are not shown, but the length is indicated instead. Sequence that is not aligned to WT sequence is colored in red. All experiments were performed once.
Supplementary Figure 2 Naïve compound heterozygous HT1 mice receiving no rAAV treatment showed body weight loss after NTBC withdrawal.
All mice (n=4) were maintained on NTBC water until five weeks old and monitored for body weight after NTBC withdrawal. The body weight measured immediately prior to NTBC withdrawal was normalized to 100% for individual mice. Note that all mice lost >20% body weight within 28 days, which is comparable to the mice treated with rAAV9-SpCas9 and scAAV8-sgAspa as control (Fig. 2a).
Supplementary Figure 3 Some compound heterozygous HT1 mice were rescued after two rounds of NTBC withdrawal.
(a) Body weight of three mice treated with rAAV9-SpCas9 and scAAV8-sgFah after NTBC withdrawal at five weeks old. See Fig. 1c for experimental timeline. These mice lost >20% body weight after the first NTBC withdrawal, and were put back onto NTBC water to further stimulate the expansion of corrected hepatocytes (blue squares). NTBC was withdrawn again when the body weight was stabilized above 100% (yellow squares). Note that, after the second round of NTBC withdrawal, the body weight was stabilized at approximately 120% within two weeks, reflecting rescue in these mice. (b) Representative FAH IHC of liver section from mice that were rescued after two rounds of NTBC withdrawal (n=3 mice). Note that the Fah-positive hepatocytes are in the process of repopulating the liver, but are fewer than the ones observed in the mice that were rescued by only one round of NTBC withdrawal (Fig. 2b).
Supplementary Figure 4 Frequency of FAH restoration and SpCas9 delivery in the liver before NTBC withdrawal.
(a) FAH immunohistochemistry (IHC) of liver sections from five Fahneo/PM mice treated with rAAV9-SpCas9 and scAAV8-sgFah at postnatal day 1 (P1). These mice were maintained on NTBC until 10 weeks old, when they were euthanized. Each section is from an individual mouse. (b) Quantification of FAH+ pixels in the IHC sections shown in (a), and its correlation with SpCas9 genome copies (GCs) (left panel) or SpCas9 expression (right panel) in the liver. Mouse ID is labeled in the IHC images, and in the dot plots. P values are calculated for Pearson’s correlation.
(a) Body weight curves of homozygous Fahneo/neo (left panel) and homozygous FahPM/PM (right panel) mice receiving no treatment (black lines; n=5 Fahneo/neo mice and 3 FahPM/PM mice) or rAAV at postnatal day 1 (P1; red lines; n=4 Fahneo/neo mice and 4 FahPM/PM mice). The rAAV treatment was the same as performed in the compound heterozygous mice (Fig. 1). All mice were kept on NTBC water until five weeks old, when NTBC was withdrawn and body weight was recorded over time. The body weight immediately prior to NTBC withdrawal was normalized to 100% for individual mice. Note that, regardless of rAAV treatment, mice of the same strain lost >20% body weight during comparable timeframes, although body weight loss was more rapid in the Fahneo/neo mice than in the FahPM/PM mice. (b) Representative FAH IHC with liver sections from homozygous Fahneo/neo (top panels) and homozygous FahPM/PM (bottom panels) mice, receiving either no treatment (left panels) or rAAV treatment at P1 (right panels). N=3 mice per group. All mice were euthanized when the body weight dropped >20% after NTBC withdrawal. Note that the same rAAV treatment did not result in FAH-positive cells in homozygous mice, in contrast with the results seen in the compound heterozygous mice (Fig. 1d, 2b).
Supplementary Figure 6 rAAV treatment at young adult age rescued disease phenotype in compound heterozygous HT1 mice.
(a) Experimental timeline. All compound heterozygous mice were either left untreated with rAAV (ctrl) or treated at six weeks old with rAAV9-SpCas9 and scAAV8-sgFah (sgFah). All mice were maintained on NTBC water until 11 weeks old. (b) Body weight after NTBC withdrawal. The body weight immediately prior to NTBC withdrawal was set to 100% for individual mice. Note that all mice in the ctrl group (n=4) lost >20% body weight within four weeks, whereas all mice in the sgFah group (n=4) regained body weight after initial weight loss. (c) Representative liver FAH IHC in a sgFah-treated mouse prior to NTBC withdrawal. N=2 mice. The boxed region is enlarged to show FAH-positive hepatocytes (dark brown, indicated by arrow heads). (d) Representative liver FAH IHC in a sgFah-treated mouse after it was rescued from body weight loss by allelic exchange, and euthanized when gene-modified hepatocytes repopulated nearly the entire liver. N=4 mice. (e) Serum aspartate transaminase (AST) (top panel) and alanine transaminase (ALT) (bottom panel) levels in wild-type (WT) mice (n=8 mice) and compound heterozygous (comp het) mice receiving no rAAV (ctrl; n=4 mice) or rAAV for allelic exchange (sgFah; n=4 mice). Ctrl mice were euthanized when they lost >20% body weight. Mice in the sgFah group were euthanized when their body weight was stabilized above 100%. Serum samples were obtained immediately prior to euthanasia, and used for AST and ALT measurement. Each dot represents one mouse. Horizontal lines depict the mean values. Error bars are SD. One-way ANOVA was performed (p<0.001), followed by multiple comparisons. P values are adjusted for multiple comparisons.
Supplementary Figure 7 Partial correction of the biochemical defects in the heart of compound heterozygous Iduaneo/W392X mice of MPS-I.
(a) Genomic structure of the Iduaneo and IduaW392X alleles and the strategy to induce allelic exchange between the two alleles (drawn to scale, except for the neo insertion). Exons 6 to 11 (black boxes) are labeled. The two mutations (neo insertion in exon 7 and W392X mutation in exon 10) are labeled in red. Bar-coded primers used to generate amplicons for SMRT sequencing are labelled. Note that DW989 binds upstream of the neo insertion in exon 7; DW1005 binds to the loxP insertion in intron 9 of the IduaW392X allele; DW990 binds to downstream of the W392X mutation in exon 10. Yellow lightning bolt: Cas9/sgRNA targeting site. Orange triangle: loxP insertion in intron 9 of the IduaW392X allele. The bottom cartoon shows an overview of this set of experiments. Compound heterozygous mice received either no rAAV treatment (NoTr), or rAAV9-SpCas9 and scAAV9-sgIdua (rAAV) at six weeks old. All mice were euthanized at one year old. (b) Stacked histogram showing the counts of TOPO clones harboring either the unedited (WT) sequence, or sequences containing indel mutations (Indel) due to sgIdua-directed Cas9 cleavage in the heart. Each bar represents one mouse as described in (a). (c) Frequency of mutation-free amplicons derived from Idua genomic DNA determined by SMRT sequencing. The amplicons were generated using primers DW1005 and DW990 as shown in (a). N=3 mice per group. (d) Frequency of mutation-free amplicons derived from Idua cDNA determined by SMRT sequencing. The amplicons were generated using primers DW989 and DW990 as shown in (a). N=3 mice for group. (e) IDUA specific activity in the heart of Iduaneo/W392X mice receiving no treatment (NoTr; n=5 mice) or rAAV treatment (n=4 mice). Dashed line indicates 0.2% or 0.5% of the IDUA specific activity in the heart of wild-type (WT) mice. (f) Glycosaminoglycan (GAG) content in the heart of the same mice shown in (e). Dashed line indicates the GAG level in the heart of WT mice. In all dot plots, horizontal lines depict mean values and error bars are SD. P values are calculated for Student’s t-test (2-sided).
(a) Diagram showing the circularization PCR strategy. Liver genomic DNA was digested with restriction enzymes (RE) Sph-I (light gray arrowhead) and Sca-I (dark gray arrowhead). This digestion generates unique DNA fragments from four Fah alleles, namely the original Fahneo and FahPM alleles, and the Fahneo-PM and FahWT alleles resulting from allelic exchange (not drawn to scale). Only exon 5 and exon 8 are labeled (black boxes). The other genomic sequence is shown as light gray bar. Red bar: neo insertion in exon 5; Orange line: G→A mutation in exon 8. Note that the neo insertion in the Fahneo allele disrupts the Sph-I site present in the WT Fah allele. The fragmented DNA ends were blunted, and individual DNA fragment underwent circularization by self-ligation. Yellow line: ligation junction. The resulting circular DNA was subjected to PCR using primers binding to common sequences present in all four Fah alleles (black arrows). PCR products were cloned into a TOPO vector, and the identity of amplicon in individual clones was determined by Sanger sequencing. (b) Representative chromatograms of Sanger sequencing results showing the circularization PCR amplicons that derived from the exchanged allele Fahneo-PM (top; n=4 clones) and FahWT (bottom; n=15 clones). The primers used in circularization PCR are labeled (DW886 and DW887). Note that the antisense sequence is shown.
Supplementary Figure 9 Targeting Fah intron 9 does not correct the HT1 phenotype in the compound heterozygous HT1 mice.
(a) Cartoon showing the hypothetical gene conversion in the compound heterozygous HT1 mice (compare with Fig. 1a). Gene conversion may generate a WT allele (top DNA repair outcome), thus leading to correction; it may also generate a double-mutation allele that is not functional (bottom DNA repair outcome). (b) Cartoon showing the design of an sgRNA targeting Fah intron 9 (orange lightning bolts) to test whether gene conversion can occur. Also see Fig. 1b for the sgRNA designed to target Fah intron 7 and to induce allelic exchange. Note that the distances between the G◊A mutation to either sgRNA targeting site are comparable. (c) Stacked histogram showing the counts of TOPO clones harboring either the unedited (WT) sequence, or sequences containing indel mutations (Indel) due to sgFah.Intron9-directed Cas9 cleavage in the liver. Each bar represents one mouse. These mice were treated with rAAV9-SpCas9 and scAAV8-sgFah.Intron9 at P1 (n=5) or adult age (6 weeks old, n=3). All compound heterozygous mice were maintained on NTBC until euthanasia. Regardless of the age of treatment, all mice were euthanized five weeks after treatment. (d) Agarose gel image showing the detection of reverse transcription-PCR products of Fah messenger RNA (mRNA) in liver lysate. M: DNA marker, sizes in base pairs (bp) are labeled; WT: wild-type mouse; comp het: compound heterozygous HT1 mice; C: control mouse receiving no treatment. The rest lanes include samples from the treated mice described in (c). Note that treatment with sgFah.Intron9 does not restore the expression of normal Fah mRNA (compare with Fig. 1e). Full-length gel image is shown in Supplementary Fig. 13d. (e) Representative images of FAH IHC of liver sections from the treated Fahneo/PM mice as described in (c). Note that FAH+ hepatocyte is not detected (compare with Fig. 1d and Supplementary Fig. 6c).
(a) Cartoon showing the plasmid constructs of translocation reporter (drawn to scale). pGT harbors the N-terminal half of EGFP (EGFPN) and the ATG-less tdTomato (tdTomatoATG-less) separated by an intron (gray bar). pTG harbors the ATG and the C-terminal half of EGFP (EGFPC) separated by the same intron in pGT. Both cassettes are driven by the CAG promoter (arrow). Yellow lightning bolt: Cas9/sgRNA targeting site at the intron. (b) Representative fluorescence microscopy images of HCT116 cells and LIG4-deficient HCT116 cells (LIG4-/-) that are co-transfected with pGT, pTG, and pX330-sgGTTG (expressing SpCas9 and the sgRNA targeting the intron). N=2 independent experiments. (c) Flow cytometry analysis to quantify the GFP+tdTomato+ cell populations as shown in (b). Representative flow cytometry plots and a dot plot of three independent experiments are shown. (d) Representative fluorescence microscopy images of HCT116 and LIG4-/- cells transfected with a GFP-expressing plasmid alone to test transfection efficiency. N=2 independent experiments. (e) Flow cytometry analysis to quantify the GFP+ cell populations as shown in (d). Representative flow cytometry plots and a dot plot of three independent experiments are shown. In both dot plots, horizontal lines depict mean values and error bars are SD. P values are calculated by Student’s t-test (2-sided).
Supplementary Figure 11 Allelic exchange may occur between homologous chromosomes during cell division.
(a) Cartoon showing possible scenarios of allelic exchange in the Fahneo/PM mice during G2 phase of cell cycle, and segregation of the exchange alleles during mitosis. The Fahneo allele is in linkage with a wild-type Tyr gene (not shown), and the FahPM allele is in linkage with the TyrC mutant allele (also see Animal use section in Methods). The distance between the Tyr gene and the Fah gene is 2.8 megabase pairs (Mb). Homologous chromosomes are depicted by black or gray horizontal lines. The polyploidy nature of mouse hepatocytes is not reflected in the drawing for simplicity but may potentially enhance the exchange frequency per cell. The relative positions of the TyrC, Fahneo, and FahPM mutations are labeled and colored as purple, red, and orange, respectively. The distance between the mutations is not drawn to scale. Open circle: centromere; Yellow lightning bolt: Cas9/sgFah targeting sites at Fah intron 7. The number of chromatids that are simultaneously cut by Cas9 may vary. As a result, there are multiple DNA repair outcomes other than the depicted exchange event involving two of the four chromatids. However, the other events would not lead to distinguishable outcome in this analysis. During mitosis, the two exchanged chromatids may segregate to the same daughter cell (Z segregation) or different daughter cells (X segregation). Regardless of the segregation pattern, one daughter cell contains two mutant Fah alleles, and therefore is lost during selection after NTBC withdrawal. The other daughter cell (green background) contains one functional Fah allele due to allelic exchange, allowing for its expansion during selection after NTBC withdrawal. Note that following Z segregation, each of the TyrC, Fahneo, and FahPM mutations accounts for 50% of gene alleles in the surviving daughter cell, the same as in the mother cell; following X segregation, the surviving daughter cell contains two TyrC mutations, no Fahneo mutation, and one FahPM mutation. Therefore, expansion of the surviving daughter cell derived from Z segregation will not alter the frequency of each mutation in the liver, whereas expansion of the one derived from X segregation will lead to increase in the TyrC mutation and decrease in the Fahneo mutation. (b) Frequency of the TyrC, Fahneo, and FahPM mutations in the liver of Fahneo/PM mice. Centre values are mean and error bars are SD. The TyrC and FahPM mutations were quantified by SMRT sequencing, and the Fahneo mutation was quantified by ddPCR (see Methods for details). These mice were treated at P1 with either rAAV9-SpCas9 + scAAV8-sgFah (sgFah, n=3), or rAAV9-SpCas9 + scAAV8-sgAspa (sgAspa, n=3) as control, maintained on NTBC until 5 weeks old when NTBC was withdrawn from the drinking water. The sgAspa-treated mice showed body weight loss and were euthanized within five weeks after NTBC withdrawal (Fig. 2a, left panel). The sgFah-treated mice were rescued and euthanized between five to six weeks after NTBC withdrawal (Fig. 2a, right panel). Dashed line indicates 50% mutation frequency. Note that in the sgFah-treated mice, the TyrC mutation frequency increases to 60%, the Fahneo mutation decreases to 30%, and the FahPM mutation remains at approximately 50%.
(a) Guide RNA (5’ to 3’ direction) and protospacer adjacent motif (PAM) sequences used in this study. (b) DNA oligo sequences used in this study. (c) Sequences attached to PCR primers to generate barcoded amplicons for SMRT sequencing.
About this article
Cite this article
Wang, D., Li, J., Song, CQ. et al. Cas9-mediated allelic exchange repairs compound heterozygous recessive mutations in mice. Nat Biotechnol 36, 839–842 (2018). https://doi.org/10.1038/nbt.4219