Fig. 6: In vivo prime editing at Dnmt1 locus and correction of Pah^enu2 mice using full RNA–LNP delivery.

a, G-to-C editing rates (top) and indel rates (bottom) at the Dnmt1 locus in mice dosed twice with the full LNP–RNA approach using 2 mg kg⁻¹ PE7 mRNA + 2 mg kg⁻¹ pegRNA (n = 4). b, Pah^enu2 correction rates (top) and indel rates (bottom) of mice dosed three times with the full LNP–RNA approach using 2 mg kg⁻¹ PE7 mRNA + 2 mg kg⁻¹ pegRNA (n = 5). In a and b, NGS was performed on isolated hepatocytes, whole liver and tail lysates. c, Phe levels at the experimental end point of homozygous B6 Pah^enu2 mice dosed three times with 2 mg kg⁻¹ PE7 mRNA + 2 mg kg⁻¹ pegRNA (n = 5) from b. Dotted lines indicate recommended therapeutic thresholds for Phe levels in adults (600 μmol l⁻¹) or in children/during pregnancy (360 μmol l⁻¹)^29,57. d, Prime editing rates in a K562 reporter cell line harbouring a DNA cassette containing the 11 most common human pathogenic PKU mutations as well as the Pah^enu2 mutation and the Dnmt1 locus for comparison. The top three predicted pegRNAs (PRIDICT2.0) were tested. Values represent mean ± s.d. of three independent biological replicates.