A base editing strategy using mRNA-LNPs for in vivo correction of the most frequent phenylketonuria variant

Abstract

The c.1222C>T (p.Arg408Trp) phenylalanine hydroxylase (PAH) variant is the most frequent cause of phenylketonuria (PKU), an autosomal recessive disorder characterized by accumulation of blood phenylalanine (Phe) to neurotoxic levels. Here we devised a therapeutic base editing strategy to correct the variant, using prime-edited hepatocyte cell lines engineered with the c.1222C>T variant to screen a variety of adenine base editors and guide RNAs in vitro, followed by assessment in c.1222C>T humanized mice in vivo. We found that upon delivery of a selected adenine base editor mRNA/guide RNA combination into mice via lipid nanoparticles (LNPs), there was sufficient PAH editing in the liver to fully normalize blood Phe levels within 48 h. This work establishes the viability of a base editing strategy to correct the most common pathogenic variant found in individuals with the most common inborn error of metabolism, albeit with potential limitations compared with other genome editing approaches.

Keywords: CRISPR; base editing; gene editing; genome editing; inborn error of metabolism; metabolic disease; phenylketonuria; prime editing; rare disease.

Figure 1

Base editing to correct PAH c.1222C>T variant in human hepatocytes in vitro (A) Schematic of the genomic site of the PAH c.1222C>T variant, adapted from the UCSC Genome Browser (GRCh38/hg38). The vertical blue bar outlined by the orange box indicates the G altered to A (in orange) by the variant on the antisense strand. The arrows indicate the sites of potential bystander editing. The horizontal green bars indicate protospacer (thick) and PAM (thin) sequences targeted by the PAH1 through PAH6 gRNAs. (B) Corrective PAH c.1222C>T editing (determined from genomic DNA) following transfection of cells with plasmids encoding adenine base editor/gRNA combinations (n = 2 biological replicates, one each from two PAH c.1222C>T homozygous HuH-7 cell lines; controls, n = 1), calculated as the proportion of aligned sequencing reads with the indicated type of edits. “Correction only” refers to reads in which the c.1222C>T adenine variant is edited to guanine, with or without base editing of the adjacent synonymous adenine, with no base editing of any other adenines; “unwanted bystander editing” refers to reads in which the c.1222C>T adenine variant is edited to guanine, along with base editing of one or more nonsynonymous adenines. (C) Dose-response study with PAH c.1222C>T homozygous HuH-7 cells treated with SpRY-ABE8.8/PAH4 LNPs (n = 3 biological replicates). (D) Dose-response study with PAH c.1222C>T homozygous HuH-7 cells treated with GFP LNPs (n = 2 to 3 biological replicates).

Figure 2

Base editing to correct PAH c.1222C>T variant in humanized mice (A) Changes in blood phenylalanine levels in homozygous PKU mice following treatment with 5-mg/kg dose of SpRY-ABE8.8/PAH4 LNPs (n = 3 animals) or with 2.5-mg/kg dose of LNPs (n = 3 animals), comparing levels at various timepoints up to 7 days following treatment to levels in vehicle-treated homozygous PKU control (n = 4 animals) and vehicle-treated heterozygous non-PKU control (n = 4 animals) age-matched (approximately 8 weeks of age) colonymates (one blood sample per time point). (B) Corrective PAH c.1222C>T editing (determined from genomic DNA) in each of eight liver samples (two samples each from the four lobes) from each treated mouse, calculated as the proportion of aligned sequencing reads with the indicated type of edits. “Correction only” refers to reads in which the c.1222C>T adenine variant is edited to guanine, with or without base editing of the adjacent synonymous adenine, with no base editing of any other adenines; “unwanted bystander editing” refers to reads in which the c.1222C>T adenine variant is edited to guanine, along with base editing of one or more nonsynonymous adenines. (C) Standard CRISPResso output for a liver sample from the LNP-treated homozygous PKU mouse with the highest level of editing. The codons in the vicinity of the c.1222C>T variant site are indicated; the top-listed amino acid is the baseline identity of the codon, and the bottom-listed amino acid is the one that results from base editing of the adenine in the codon. Lines in graphs = mean values.

Figure 3

Assessment of off-target editing On-target or off-target editing at top in silico-nominated candidate sites calculated as the proportion of aligned sequencing reads with ≥1 adenine base edited to guanine within the editing window at each site in PAH c.1222C>T homozygous HuH-7 cells that underwent treatment with SpRY-ABE8.8/PAH4 LNPs at a dose of 10,000 fg/cell (n = 3 treated and 3 untreated biological replicates), the highest dose shown in Figure 1C. Sites with unsuccessful sequencing are omitted. Refer to Table S2 for candidate site sequences and numerical values.