Crispant mice with mutated Tyrp1 display the expected recessive brown coat colour phenotype

A. Schematic of the Tyrp1 gene with the predominant TYRP1 protein isoform below (exons and protein domains drawn to scale). The position of the Cas9 cut site is indicated by the scissors and arrow. Key: 5’ and 3’ UTRs: white; open reading frame (ORF): black; targeted exon: red; excised propeptide: white; tyrosinase domain: brown; transmembrane domain “TM”: red; cytoplasmic domain “Cyt”: orange; and uncharacterized protein regions: grey. B. Photograph of founders with a Tyrp1+/+control. C. Schematic of exon 3 of the Tyrp1 gene showing the location of the Cas9 cut site, genotyping and qPCR primers (drawn to scale). D. PCR amplification around the targeted site in all four Typr1 founders, beside a wild-type reference (615 bp). E. qPCR quantification around the 5’ and 3’ genotyping primer annealing sites in founders #4084 and #4086, normalised to three wild-type controls (mean ± SEM). F. DAS analysis of #4084 (percentage abundance of indels present in >0.5% reads are shown). G. Schematic of the two alleles in the compound heterozygous founder, #4086 (drawn to scale, deleted regions shown in pink). H. Schematic of the protein sequence resulting from the in-frame deletion in #4086 (c.636_647del; p. D212_H215del) and the published Tyrp1b-Btlr allele.

Overview of zygote EP experiments

Crispant mice with mutated Slc45a2 show phenotypic variability

A. Schematic of the Slc45a2 gene with the S45A2 protein below (exons and protein domains drawn to scale). The position of the Cas9 cut site is indicated by the scissors and arrow. Key: 5’ and 3’ UTRs: white; open reading frame (ORF): black; targeted exon: red; transmembrane domains “TM”: red; cytoplasmic domains “Cyt”: orange; extracellular domains “Ex”: beige; and disordered “Dis”: black bands. B. Photographs of selected founders with a full light grey coat (left), patchy grey and black coat (left and right) and black coat (right). C-D. PCR amplification around the targeted site in light grey and patchy founders (C) and black founders (D), beside a wild-type reference (614 bp). E. qPCR quantification around the 5’ and 3’ genotyping primer annealing sites in compound heterozygous black founders (#4062, #4064 and #4079), normalised to three wild-type controls (mean ± SEM). F. Protein sequence resulting from the published truncated Slc45a2uw-10Btlr allele and in-frame indels in the compound heterozygous black founders. G. Schematic of the two alleles in the compound heterozygous founder with genome damage, #4062 (drawn to scale, deleted regions and the inserted region of intron 1 are shown in pink).

Crispant mice display recessive visible phenotypes

A-D. Schematic of the Fgf5 (A), Tgm3 (B), Bmp5 (C) and Pax1 (D) genes with their predominant/sole protein isoforms below (exons and protein domains drawn to scale). The position of the Cas9 cut site is indicated by the scissors and arrow. Key: 5’ and 3’ UTRs: white; open reading frame (ORF): black; targeted exon: red; excised propeptide: white; FGF domain: gold; N- and C-terminal Transglutaminase domains: green; calcium (Ca2+) binding sites: teal; signal peptide: cyan; RxxR motif: blue; transforming growth factor beta (TGFβ) like domain: navy; PAI-RED box domain first/second portion: purple/magenta and uncharacterised protein regions: grey. E-H. Photograph of one representative founder for each gene: Fgf5 (E), Tgm3 (F), Bmp5 (G) and Pax1 (H). I. Comparison of percentage viability when targeting one autosomal endogenous gene, two autosomal endogenous genes, or one autosomal endogenous gene in combination with X-linked EGFP. J. Comparison of percentage viability when targeting Tyrp1, Fgf5, Tgm3, and Bmp5 alone or in combination with X-linked EGFP. K. Comparison of sex ratio of founders obtained when targeting one autosomal endogenous gene alone or in combination with X-linked EGFP. L. Comparison of estimated percentage viability when targeting one autosomal endogenous gene in combination with X-linked EGFP in male founders (three target sites per genome) and female founders (four target sites per genome). To calculate percentage viability in the endogenous gene + EGFP experiments, half of the 117 zygotes were assumed to be from each sex: males 12/58.5 = 20.5% and females 5/58.5 = 8.5%. I-L. Pairs of ratios were compared using Fisher’s Exact tests: **** P < 0.0001, *** P <0.001, ** P < 0.01, * P < 0.05 and NS (not significant) P > 0.05.

Compound heterozygous crispants enable genotype – phenotype associations

A. Photograph of Tyrp1 founder #4728 showing the partial brown coat colour phenotype with black patches. B. Percentage of unedited sequence in all founders that had >0% unedited sequence predicted by Sanger deconvolution software (DECODR and ICE); and quantified by DAS in selected founders. C. Schematic of the large deletions in Bmp5 founders, #3722 and #4727 (drawn to scale, deleted regions shown in pink). D. Schematic of the mouse chromosome 9 showing the Myo5a (“dilute”) gene 700 kb upstream of Bmp5. E. Photograph of Bmp5 founder #3725 showing patches of grey fur. F. qPCR quantification at the upstream Myo5a locus and around the Bmp5 5’ and 3’ genotyping primer annealing sites in gDNA derived from the spleen and grey fur patch of founder #3725), normalised to three wild-type controls (mean ± SEM). G. Graph showing the number of founders generated for each gene that were compound heterozygous for an in-frame indel and frameshift/large deletion allele (black); compound heterozygous for two frameshift indels or one frameshift indel and one large deletion (grey); and with genetic mosaicism (white). H-J. Protein sequence schematic for the in-frame deletions in the compound heterozygous Fgf5 (H), Tgm3 (I) and Bmp5 (J) founders. See Figure 3 for colour key.

CRISPR/Cas9 mutagenesis can generate an allelic series of non-null mutations

A. Schematic of the Adamts20 gene with the ADAMTS20 protein (exons and protein domains drawn to scale). The position of the Cas9 cut site is indicated by the scissors and arrow. Key: 5’ and 3’ UTRs: white; open reading frame (ORF): black; targeted exon: red; excised propeptide: white; Reprolysin: green; Disintegrin: blue; TSP repeats: purple and orange. Protein changes resulting from published alleles are shown below. Key: nonsense mutations: black; and missense mutations: colour matches domain. B. Photographs of selected founders showing the full belted phenotype (upper), partial belt from mosaicism (lower left) and partial belt from hypomorphic allele (lower right). C. Schematic of the protein sequence resulting from the truncating indels and in-frame deletions in compound heterozygous founders (“het”) and mosaic founders (“mos”) carrying one short indel and more than one large deletion; and the published Adamts2Bt-Bei1and Adamts20m5Btlr alleles. D. Schematic of the genome damage in the compound heterozygous founders, #2984 and #2986 (drawn to scale, deleted regions shown in pink).

Genotyping analysis of Tyrp1 founders

A. DECODR deconvolution of Sanger sequencing chromatograms of the genotyping amplicon in Tyrp1 founders. B. ICE deconvolution of Sanger sequencing chromatogram of the genotyping amplicon in Tyrp1 founder #4084.

ELF-CLAMP analysis of the compound heterozygous Tyrp1 founder

A. Schematic of the ELF-CLAMP experimental workflow. B. Screenshot of ELF-CLAMP sequencing reads viewed in IGV for Typr1 founder #4086 are shown below a schematic of the Tyrp1 gene showing the location of the Cas9 cut site, genotyping and qPCR primers and the viewpoints used for ELF-CLAMP.

Genotyping analysis of Slc45a2 founders

A. Sequence alignment of the complex deletion-insertion present in Slc45a2 founder #4062 compared to the wild-type (reference) sequence. B. DECODR deconvolution of Sanger sequence chromatograms of the genotyping amplicon in Slc45a2 founders #4064 and #4079.

Crispant mice with two mutated genes display both phenotypes

A. Photographs of founders in which two endogenous genes have been mutated. From left: #5485 (Tyrp1 + Tgm3), #5486 (Tyrp1 + Bmp5) and #2983-6 (Tgm3 + Adamts20). B. PCR amplification around the Adamts20 target site in all four Tgm3 + Adamts20 founders, beside a wild-type reference (437 bp). C. qPCR quantification around the 5’ and 3’ Adamts20 genotyping primer annealing sites and more distal sites in intron 21 and exon 24 (∼1.5 kb away from the targeted site) in founders #2983, #2984 and #2986), normalised to three wild-type controls (mean ± SEM). D. DECODR deconvolution of Sanger sequence chromatograms of the Adamts20 genotyping amplicon in founder #2983 (mosaic for c.2798_2814del; p.H933PfsX9 and large 5’ and/or 3’ deletions) and #2984 (compound heterozygous for c.2804_2805insG; p.G935GfsX13 and a large 5’ deletion) and sequence alignment of the Adamts20 genotyping amplicon in founder #2986 (compound heterozygous for c.2803delGinsAA; p.G935KfsX13 and a large 5’ - 3’ deletion).

Mutations show high concordance across germ layers

A. PCR amplification around the targeted site in all five Pax1 founders (tail), beside a wild-type reference (630 bp). B. qPCR quantification around the 5’ and 3’ genotyping primer annealing sites in founders #4081, #4087 and #4088), normalised to three wild-type controls (mean ± SEM). C. Genotypic analysis of all five Pax1 founders using DNA derived from tail (largely ectoderm), spleen (mesoderm) and liver (endoderm). Graph shows the percentage of each repair product. inDelphi predicted two products, plotted by predicted frequency (percentage). D. Sequence alignment and ICE deconvolution of Sanger sequence chromatogram from the genotyping amplicon in founder #4088, using tail gDNA. E. DECODR deconvolution of Sanger sequence chromatogram of founder #4081 (compound heterozygous for c.339_343del; p.R113RfsX12 and a large 3’ deletion), using tail gDNA.

Genotyping analysis of Tyrp1 and Tgm3 founders with predicted unedited sequence

A. PCR amplification around the targeted site in the two additional Typr1 founders with predicted unedited sequence (#4728 and #5486), beside a wild-type reference (615 bp). B. qPCR quantification around the 5’ and 3’ genotyping primer annealing sites in founders #4728 and #5486, normalised to three wild-type controls (mean ± SEM). C. DECODR deconvolution of Sanger sequence chromatograms of the genotyping amplicon in the Tyrp1 founders. D. DAS analysis of #4728 and #5486 (percentage abundance of indels present in >0.5% reads are shown). E. PCR amplification around the targeted site in the two Tgm3 founders with predicted unedited sequence (#4136 and #4723), beside a wild-type reference (609 bp). F. qPCR quantification around the 5’ and 3’ genotyping primer annealing sites in founders #4136 and #4723, normalised to three wild-type controls (mean ± SEM). G. DECODR deconvolution of Sanger sequence chromatograms of the genotyping amplicon in Tgm3 founders. H. DAS analysis of #4136 and #4723 (percentage abundance of indels present in >0.5% reads are shown).

High microhomology of Tgm3 cut site results in founders that are heterozygous for the predicted repair product

A. InDelphi prediction of repair products at the Tgm3 cut site. High microhomology promotes an 8 bp deletion: c.739-746del p.P247LfsX4. B. PCR amplification around the targeted site in the four compound heterozygous Tgm3 founders (#3735, #3736, #3737 and #3738), beside a wild-type reference (609 bp). C. qPCR quantification around the 5’ and 3’ genotyping primer annealing sites in these founders, normalised to three wild-type controls (mean ± SEM). D. DECODR deconvolution of Sanger sequence chromatograms of the genotyping amplicon in founders #3735, #3736 and #3738. E. Sequence alignment of Sanger sequencing of the genotyping amplicon in founder #3737.

Genotyping analysis of compound heterozygous Tyrp1, Fgf5 and Bmp5 founders carrying two knock-out alleles

A. PCR amplification around the targeted site in compound heterozygous founders carrying two knockout alleles: Tyrp1 #4730; Fgf5 #3713 and #3724; and Bmp5 #3715, #3719 and #4733. B. qPCR quantification around the 5’ and 3’ genotyping primer annealing sites (and the nested sequencing primer annealing site for Bmp5) in founders, normalised to three wild-type controls (mean ± SEM). C. DECODR deconvolution of Sanger sequence chromatograms of the genotyping amplicon in these founders. For genotyping analysis of Pax1 founder #4081, see Figure S5; and of Tgm3 founders #3735, #3736, #3737 and #3738, see Figure S7.

Genotyping analysis of compound heterozygous Fgf5, Tgm3 and Bmp5 founders carrying one in-frame allele

A. PCR amplification around the targeted site in compound heterozygous founders carrying an in-frame allele: Fgf5 #3728; Tgm3 #4724 and Bmp5 #4727 and #4734. B. qPCR quantification around the 5’ and 3’ genotyping primer annealing sites (and the nested sequencing primer annealing site for Bmp5) in these founders, normalised to three wild-type controls (mean ± SEM). C. DECODR deconvolution of Sanger sequence chromatograms of the genotyping amplicon in these founders. A schematic of the genome damage in Tgm3 founder #2724 (drawn to scale, deleted regions shown in pink). For a schematic of the genome damage in Bmp5 founder #4727, see Figure 4C. For genotyping analysis of Tyrp1 founder #4086, see Figures 1 and S1-2; and of Slc45a2 founders #4062, #4064 and #4079, see Figures 2 and S3.

Genotyping analysis of Adamts20 founders carrying an in-frame allele

A. PCR amplification around the targeted site in Adamts20 founders carrying an in-frame allele: #3748, #3744 and #3745. B. qPCR quantification around the 5’ and 3’ genotyping primer annealing sites and more distal sites in intron 21 and exon 24 (∼1.5 kb away from the targeted site) in these founders, normalised to three wild-type controls (mean ± SEM). C. Alignment and DECODR deconvolution of Sanger sequence chromatograms of the genotyping amplicon in these founders. The 42 bp deletion in #3744 is flanked by 6 bp of microhomology (boxed).

Structure of genes used in this study

*Current gene structure annotation means numbering differs from cited publications.

Founders from single guide zygote EP genotyping results

* Founders in photographs. † DECODR results. ‡ Other software

Mutagenesis is consistent across all three germ layers

† DECODR results. ‡ Other software.

Founders from two guides (endogenous gene + EGFP) zygote EP genotyping results

* Founders in photographs. † DECODR results. ‡ Other software

Founders from two guides (two endogenous genes) zygote EP genotyping results

Founders in photographs. † DECODR results. ‡ Other software

Founders from Adamts20 zygote EP genotyping results

Founders in photographs. † DECODR results. ‡ Other software

sgRNA sequences

Primers

Oligos for ELF-CLAMP