Protein language models-assisted optimization of a uracil-N-glycosylase variant enables programmable T-to-G and T-to-C base editing

Current base editors (BEs) use DNA deaminases, including cytidine deaminase in cytidine BE (CBE) or adenine deaminase in adenine BE (ABE), to facilitate transition nucleotide substitutions. Combining CBE or ABE with glycosylase enzymes can induce limited transversion mutations. Nonetheless, a critical demand remains for BEs capable of generating alternative mutation types, such as T>G corrections. In this study, we leveraged pre-trained protein language models to optimize a uracil-N-glycosylase (UNG) variant with altered specificity for thymines (eTDG). Notably, after two rounds of testing fewer than 50 top-ranking variants, more than 50% exhibited over 1.5-fold enhancement in enzymatic activities. When eTDG was fused with nCas9, it induced programmable T-to-S (G/C) substitutions and corrected db/db diabetic mutation in mice (up to 55%). Our findings not only establish orthogonal strategies for developing novel BEs but also demonstrate the capacities of protein language models for optimizing enzymes without extensive task-specific training data. [Display omitted] •nCas9 with engineered UNGs enable transversion base editing without deamination•PLMs were used to predict enzymatic variant activities•Using the PLMs, an efficient T>S (G or C) base editor, TSBE3, was developed•TSBE3 effectively corrected a diabetic mutation (Leprdb) in murine embryos He et al. utilized protein language models (PLMs) to engineer an enhanced UNG variant, eTDG, targeting thymine. Accurate predictions allowed the validation of over 80% of high-fitness variants. This enabled the development of TSBE3, a tool for efficient T>G or C substitutions in cell lines, T cells, and mouse embryos.