Abstract 2764: Cas9 cytosine base editors for precision genome editing with high specificity and activity

Abstract Base editing technologies allow for the precise and efficient installation of defined nucleotide substitutions into a target genome without double strand breaks. These editing methods can be used to generate cell lines and animal models for the study of a diverse range of topics, from molecular mechanisms of pathogenesis to drug resistance and therapeutics. However, overexpression of base editors can lead to off-target editing, both genome-wide in a guide-independent fashion and enriched at sequences similar to the intended target. To counter this challenge, we have developed a suite of recombinant protein-based cytosine base editor (CBE) tools. This suite includes two separate CBE proteins with complementary editing windows, a protein format uracil glycosylase inhibitor (UGI), and a guide design tool to facilitate experimental planning and computationally predict editing outcomes. The first CBE, “Flexible, ” has a wide editing window, enabling editing over a large fraction of the target sequence and facilitating flexible placement of the guide RNA relative to the target cytosine residue. The second CBE, “Precision, ” has a narrowed editing window, enabling discrimination between the target cytosine and nearby, untargeted cytosines. We show that co-transfection with the UGI protein increases the proportion of substitutions that are C-to-T and the ratio of C-to-T editing to indel formation, thus reducing undesired editing outcomes. We further show that co-transfection with a transfection enhancer results in an increase in overall editing rates, and that the transfection enhancer and UGI can be combined to maximize the rate of C-to-T editing while minimizing undesired outcomes. We use in vitro editing experiments to characterize the editing profiles of our CBEs and identify a preference for cytosine residues preceded by thymine (TC dinucleotides) and unmethylated cytosine residues. Finally, we show that RNP delivery of Flexible and Precision CBEs results in lower rates of mismatch editing and genome-wide nonspecific editing than plasmid delivery of the same editors, demonstrating their improved specificity profiles. To facilitate use of these editors, we have developed a guide design tool that will identify candidate guide RNAs in a user-provided sequence, identify potential off-target sites, and design primers to amplify the on- and off-target sites. The tool further uses a machine learning model trained on experimental data to predict which cytosine residues are most likely to be edited, and a joint probability model to predict the potential bystander outcomes, enabling the user to predict their editing outcomes before performing the experiment. Together, these tools will facilitate precision genome editing experiments. Citation Format: Erin Brettmann, Stephen Beishir, Xiao Chen, Xiaokang Zhang, Rajagopalan Lakshmi Narasimhan, Ajaya Kumar Kotekar Manjunath, Daniel Taglicht, Jiajian Liu, Fuqiang Chen. 2764: Cas9 cytosine base editors for precision genome editing with high specificity and activity [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 2764.