The foundational adenine base editors (e.g. ABE7.10) enable programmable C•G to T•A point mutations but editing efficiencies can be low at challenging loci in primary human cells. Here we further evolve ABE7.10 using a library of adenosine deaminase variants to create ABE8s. At NGG PAM sites, ABE8s result in ~1.5x higher editing at protospacer positions A5-A7 and ~3.2x higher editing at positions A3-A4 and A8-A10 compared with ABE7.10. Non-NGG PAM variants have a ~4.2-fold overall higher on-target editing efficiency than ABE7.10. In human CD34+ cells, ABE8 can recreate a natural allele at the promoter of the γ-globin genes HBG1 and HBG2, with up to 60% efficiency, causing persistence of fetal hemoglobin. In primary human T cells, ABE8s achieve 98-99% target modification which is maintained when multiplexed across three loci. Delivered as mRNA, ABE8s induce no significant levels of sgRNA-independent off-target adenine deamination in genomic DNA and very low levels of adenine deamination in cellular mRNA.

Summary Embryonic development relies on the capacity of progenitor cells to appropriately respond to inductive cues; a cellular property known as developmental competence. Here we report that epigenetic priming of enhancers signifies developmental competence during endodermal lineage diversification. Chromatin mapping during pancreatic and hepatic differentiation of human embryonic stem cells revealed the en masse acquisition of a poised chromatin state at enhancers specific to endoderm-derived cell lineages in gut tube intermediates. Experimentally, the acquisition of this poised enhancer state predicts the ability of endodermal intermediates to respond to inductive signals. Furthermore, these enhancers are first recognized by the pioneer transcription factors FOXA1 and FOXA2 when competence is acquired, while subsequent recruitment of lineage-inductive transcription factors, such as PDX1, leads to enhancer and target gene activation. Together, our results identify the acquisition of a poised chromatin state at enhancers as a mechanism by which progenitor cells acquire developmental competence.

The foundational adenine base editors (e.g. ABE7.10) enable programmable C•G to T•A point mutations but editing efficiencies can be low at challenging loci in primary human cells. Here we further evolve ABE7.10 using a library of adenosine deaminase variants to create ABE8s. At NGG PAM sites, ABE8s result in ∼1.5x higher editing at protospacer positions A5-A7 and ∼3.2x higher editing at positions A3-A4 and A8-A10 compared with ABE7.10. Non-NGG PAM variants have a ∼4.2-fold overall higher on-target editing efficiency than ABE7.10. In human CD34+ cells, ABE8 can recreate a natural allele at the promoter of the γ-globin genes HBG1 and HBG2, with up to 60% efficiency, causing persistence of fetal hemoglobin. In primary human T cells, ABE8s achieve 98-99% target modification which is maintained when multiplexed across three loci. Delivered as mRNA, ABE8s induce no significant levels of sgRNA-independent off-target adenine deamination in genomic DNA and very low levels of adenine deamination in cellular mRNA.

Base editors are fusions of a deaminase and CRISPR-Cas ribonucleoprotein that allow programmable installment of transition mutations without double-strand DNA break intermediates. The breadth of potential base editing targets is frequently limited by the requirement of a suitably positioned Cas9 protospacer adjacent motif. To address this, we used structures of Cas9 and TadA to design a set of inlaid base editors (IBEs), in which deaminase domains are internal to Cas9. Several of these IBEs exhibit shifted editing windows and greater editing efficiency, enabling editing of targets outside the canonical editing window with reduced DNA and RNA off-target editing frequency. Finally, we show that IBEs enable conversion of the pathogenic sickle cell hemoglobin allele to the naturally occurring HbG-Makassar variant in patient-derived hematopoietic stem cells.

Cytosine base editors (CBEs) enable programmable genomic C·G-to-T·A transition mutations and typically comprise a modified CRISPR–Cas enzyme, a naturally occurring cytidine deaminase, and an inhibitor of uracil repair. Previous studies have shown that CBEs utilizing naturally occurring cytidine deaminases may cause unguided, genome-wide cytosine deamination. While improved CBEs that decrease stochastic genome-wide off-targets have subsequently been reported, these editors can suffer from suboptimal on-target performance. Here, we report the generation and characterization of CBEs that use engineered variants of TadA (CBE-T) that enable high on-target C·G to T·A across a sequence-diverse set of genomic loci, demonstrate robust activity in primary cells and cause no detectable elevation in genome-wide mutation. Additionally, we report cytosine and adenine base editors (CABEs) catalyzing both A-to-I and C-to-U editing (CABE-Ts). Together with ABEs, CBE-Ts and CABE-Ts enable the programmable installation of all transition mutations using laboratory-evolved TadA variants with improved properties relative to previously reported CBEs.