An in vivo selection method to optimize trans-splicing ribozymes

Group I intron ribozymes can repair mutated mRNAs by replacing the 3'-terminal portion of the mRNA with their own 3'-exon. This trans-splicing reaction has the potential to treat genetic disorders and to selectively kill cancer cells or virus-infected cells. However, these ribozymes have not yet been used in therapy, partially due to a low in vivo trans-splicing efficiency. Previous strategies to improve the trans-splicing efficiencies focused on designing and testing individual ribozyme constructs. Here we describe a method that selects the most efficient ribozymes from millions of ribozyme variants. This method uses an in vivo rescue assay where the mRNA of an inactivated antibiotic resistance gene is repaired by trans-splicing group I intron ribozymes. Bacterial cells that express efficient trans-splicing ribozymes are able to grow on medium containing the antibiotic chloramphenicol. We randomized a 5'-terminal sequence of the Tetrahymena thermophila group I intron and screened a library with 9 × 10⁶ ribozyme variants for the best trans-splicing activity. The resulting ribozymes showed increased trans-splicing efficiency and help the design of efficient trans-splicing ribozymes for different sequence contexts. This in vivo selection method can now be used to optimize any sequence in trans-splicing ribozymes.