Adenine transversion editors enable precise, efficient A•T-to-C•G base editing in mammalian cells and embryos

Chen et al. (2024) introduce adenine-to-cytosine base editors (ACBEs) — fusion proteins combining mouse alkyladenine DNA glycosylase (mAAG) with nickase Cas9 and TadA-8e deaminase. Laboratory evolution of mAAG increased A→C/T conversion efficiency from baseline to 73% maximum, yielding precise A•T→C•G transversions with minimal Cas9-independent off-targeting. In mouse embryos, founder mice showed 44–56% average A→C edits with allelic frequencies reaching 100%.

Relevance to h07 (Misha’s STRC mutation): ACBE goes the WRONG direction for Misha’s c.4976A>C variant. Misha has a C on the coding strand at the mutant position; correction requires C→A (not A→C). ACBE would CREATE the mutation, not fix it. ACBE is relevant only for other DFNB16 patients with different variant classes, or as a size-advantage reference (ACBE is ~130 kDa vs PE3 ~180 kDa; single-AAV compatible).

Author note: This paper was cited in h07 notes as “Kim 2023.” PMID 37322276 belongs to Chen L et al. (2024, Nature Biotechnology), not Kim. There is no author named Kim in the author list. Published 2024 (epub June 2023 — explaining the “2023” year confusion). First author is Liang Chen (Dali Li lab + Liu lab collaboration).

Key finding

A→C transversion editors reach 73% (evolved) in cell culture, 44–56% in embryos. Smaller payload than PE3 favors single-AAV packaging. But the edit direction is A→C, which creates Misha’s STRC mutation (c.4976A>C) rather than correcting it.

Numbers that matter

ParameterValueUnitsSourceConditions
Max ACBE efficiency (evolved mAAG)73%AbstractCell culture, evolved variant, A→C conversion
Founder mouse embryo efficiency44–56% (average)AbstractMouse embryos, A→C edits
Embryo allelic frequency (max)100%AbstractBest-case allele
ACBE size~130kDaPaperEnables single-AAV delivery (vs PE3 ~180 kDa)
PE3 size~180kDaLiteratureRequires dual-AAV
Edit typeA•T → C•GtransversionAbstractCannot perform C→A correction

Limitations

  • Wrong edit direction for Misha’s STRC c.4976A>C variant: correcting C→A on coding strand requires prime editing, not ACBE.
  • No in vivo post-mitotic data in neuronal or cochlear cells reported.
  • Embryo data; somatic adult tissue efficiencies not shown in abstract.
  • Off-target profile in vivo not characterized in abstract.

Connections