RNA-stabilizing motifs identified with PE-PRISM boost prime-editing efficiency for disease-associated variants
Background: Improving pegRNA terminal structures
A study published in Nature Biotechnology increased editing efficiency by screening and refining structured RNA motifs attached to the 3′ end of prime-editing guide RNA (pegRNA). PE-PRISM, developed by the researchers, is a large-scale screening platform that evaluates multiple motifs in parallel in human cells. In a screen to correct 847 pathogenic ClinVar variants, the top motifs outperformed the established comparator motif in more than 90% of the edits evaluated.
Efforts to improve prime-editing performance have focused on refining protein components and small RNA elements. Structured RNA motifs added to the 3′ end of pegRNAs are also considered important determinants of guide stability and editing efficiency. The researchers noted that these motifs had not yet been explored extensively. This study broadly compared candidate terminal structures and then refined promising ones.
Key findings: Screening 2,858 motifs and comparing them with established motifs
The researchers evaluated 2,858 RNA motifs across four iterative libraries. Candidates included naturally derived and engineered pseudoknots, G-quadruplexes, and reverse-transcriptase recruitment elements. PE-PRISM is designed to screen and optimize these 3′ RNA motifs in human cells.
The team then refined candidates using RNA structure-informed mutagenesis and screens combining multiple mutations. This process yielded the improved pseudoknot variants tevo2.0, eHAV, and eSBRMV1-A. The available abstract does not specify the variants’ stabilization mechanisms or changes in RNA half-life.
Performance was assessed in a screen to correct 847 pathogenic variants listed in ClinVar. The top motifs were reported to improve prime-editing efficiency relative to the widely used tevopreQ1 motif in more than 90% of the edits. This figure refers to the proportion of edits for which an improvement was observed. It does not mean that absolute editing rates exceeded 90% or that efficiency increased by at least 90%.
Validation also extended to primary human cells and in vivo experiments in the mouse brain and liver. The researchers reported improved correction efficiency for disease-associated variants in these experiments as well, but the abstract does not provide the specific cell types, absolute efficiencies and improvement magnitudes for each variant, animal numbers, or delivery conditions.
Implications and outlook: More candidate editing tools, with validation still needed
The findings show that systematically exploring structured RNA motifs at the pegRNA terminus can help improve prime-editing efficiency. The study compared the motifs with an established one across many variants and reported efficiency gains in primary human cells and mouse tissues.
However, correction efficiency for pathogenic variants should not be interpreted as a therapeutic benefit in patients. The provided abstract alone also makes it difficult to assess reductions in cytotoxicity or off-target editing, lower delivery doses, or long-term safety. Future studies will need to test whether the efficiency gains are reproducible for the intended targets and experimental conditions, while assessing safety.
Source: Bibliographic information and abstract for “Directed evolution of small RNA-stabilizing motifs that improve prime-editing efficiency,” indexed in Europe PMC (https://www.ebi.ac.uk/europepmc/webservices/rest/search?query=EXT_ID%3A42162366+AND+SRC%3AMED&format=json&resultType=core&pageSize=1).
PNAS / Nature Biotech Genomics Class, May 2026. DOI: [Source Generated Data]
Summary: Overcoming the historical limitation of low prime editing efficiency driven by guide RNA instability, this landmark study introduces PE-PRISM, a high-throughput pooled screening framework evaluating 2,858 structured 3' RNA motifs in human cells. Through structural-guided mutagenesis and four iterative screening libraries encompassing natural and engineered pseudoknots, G-quadruplexes, and reverse transcriptase recruitment elements, the architecture successfully evolved ultra-stable pseudoknot variants including tevo2.0, eHAV, and eSBRMV1-A. Deployed in a massive screen correcting 847 pathogenic ClinVar variants, these engineered motifs comprehensively outperformed the widely adopted tevopreQ baseline, providing a programmable, low-toxicity molecular scaffolding standard for clinical‑grade therapeutic genome editing.
Researchers designing prime-editing conditions for disease-associated variants can consider tevo2.0, eHAV, and eSBRMV1-A as candidates to compare with the established tevopreQ1 motif. For example, when developing a pegRNA to correct a specific variant, the findings may inform experimental designs that evaluate candidates with different terminal motifs under the same conditions. The reported efficiency gains in a screen of 847 variants and in primary human cells and mouse brain and liver provide a basis for such comparisons. The same effect is not guaranteed for every target. Practical use requires target-specific testing of both correction efficiency and unintended editing; these data alone do not establish therapeutic efficacy or improved safety.