Fig 1.
Production of PAH-targeted piglets.
A) TALENs were designed to target Exon 8 of sus scrofa (ss) PAH. The location of the right monomer was strategically placed to contain a mismatch following a successful R408W HDR event. B) Sequence alignment of human (hs, top) and wild type pig (ss, second) shows the high similarity surrounding the target R408W mutation (red box). A homology template (HT, third) was designed to produce the R408W mutation in addition to introducing the 5 SNPs (black boxes) needed to “humanize” the sequence (hR408W) around the mutation to allow for targeting with human-translatable gene editing reagents. Lastly, genotype confirmation verified all piglets were positive for R408W and the 5 humanizing SNPs. C) Picture of wild type Ossabaw piglet (left) and Ossabaw PAHhR408W/hR408W founder piglet (middle left, No. 1769) showing hypopigmentation with hR408W mutation consistent with that observed in PKU mouse models. Additionally, large white/landrace PKU founders (PAHhR408W/hR408W, middle right) and PKU founders gestated on NTBC (PAHhR408W/A403GfsX47, far right) are also presented. D) Birth weights of these PAH-targeted piglets are lower than historical values for wildtype piglets of their respective genetic background strains in the absence of NTBC during gestation. * p < 0.05 compared to untreated PAHhR408W/hR408W in the same background.
Table 1.
Genotypes of all PAH-targeted piglets.
Fig 2.
Genotype of compound heterozygous hR408W/pA403GfsX47 pigs.
A) Sanger sequencing of TOPO clones from Piglet Nos. 22 (Charm) and 23 (Lucky). Only the non-R408W allele is shown, revealing an indel consisting of a 21 base pair deletion with a 16 base pair insertion. B) Translation of the indel cDNA and alignment the wild type and R408W shows the frame shift mutation begins at amino acid A403 resulting in a stop codon 47 amino acids downstream. This allele does not produce functional PAH, as predicted by this translation.
Fig 3.
PAH-targeted phenotype characterization.
A) Yorkshire founder PKU piglets maintained on NTBC were analyzed at birth for circulating phenylalanine and tyrosine levels, compared to the sow (assayed a C-section). Wild type tyrosine level is represented for comparison due to the elevated levels present in the sow, as maintained on NTBC. B) Two piglets (Lucky and Cornflake) were maintained continuously to characterize the R408W/x metabolic phenotype. Tyrosine levels (blue) were elevated at birth due to NTBC administration during gestation, and showed muted responsiveness dietary phenylalanine available during development (green). C) Serum phenylalanine levels (blue) were well correlated with dietary phenylalanine (green), indicative of disrupted phenylalanine metabolism, modeling human disease. Asterisk represents timing of clinical chemistry analysis in the next panel. D) Liver enzyme analysis of the long term piglet (Lucky) showed healthy ALP and AST levels while maintained on phenylalanine-restricted diet. E) PAH activity was assayed from liver homogenates of animals harvested within 1 week of birth demonstrating nearly complete ablation of activity despite expression, indicating loss of function from the hR408W mutation. * p<0.05 compared to WT sow (A) or age matched control (E).
Table 2.
Brain amino acid profiles in PAHR408W/R408W piglets.
Fig 4.
Histology of WT and Pah KO livers in mouse and pig.
Standard H&E and Masson’s trichrome sections show no appreciable difference between WT and Pahenu2/enu2 mice or WT and PAHR408W/R408W pigs. Similarly, immunohistochemistry showed similar levels and ubiquitous distribution of expression of the wild type and mutant protein in both models/species.
Fig 5.
In vitro targeting of humanized PAHR408W and correction by HDR.
A) Sequence alignment of humanized pig PAHhR408W (top) showing the guide RNA for the RNPs (grey arrow) as well as a homology template modified to revert the mutation and disrupt the PAM to prevent re-cutting (bottom). The location of the R408W mutation (red letters) and engineered BsaI site are also indicated. B) T7 endonuclease assay of PAH PCR products derived from hR408W fibroblasts, fibroblasts treated with RNPs containing the PAH gRNA, and a positive control for T7 cutting. The presence of the lower/lighter bands indicate mismatched double-stranded DNA indicative of NHEJ resulting from the RNPs. C) Quantitation of amplicon sequences derived from unedited (blue) and RNP treated (red) PCR of the PAHhR408W locus as well as the top 12 predicted off-target cutting sites for the gRNA employed. Only the PAH locus showed indels above background detection levels when compared to untreated samples. D) RFLP with BsaI was negative for untreated hR408W fibroblasts, while the diagnostic bands were present when cells were co-transfected with the RNP and the single stranded oligo template for HDR. Positive control was based on a synthesized dsDNA encoding the intended HDR product. Relative position of the BsaI cut site is diagramed below the gel for reference. E) HDR-specific PCR showed the presence of the anticipated product in co-transfected cells, but not untreated hR408W fibroblasts. The positive control from D, synthetic DNA created to produce the target HDR sequence for analysis, was also evaluated. F) The predominant sequence in untreated cells was R408W, which is reduced in both RNP and RNP/HT co-transfected cells. G) Pie chart indicating the relative presence of unedited (R408W), NHEJ, HDR (W408R and BsaI site), and incomplete HDR (not all corrections present) species in PAH locus amplicon sequencing from panel F. Gel images presented in B, D, and E are from a single gel each that was abbreviated for presentation as indicated by black vertical lines.
Table 3.
Off target analysis sites evaluated for indel formation after CRISPR editing.