SPELL reduces internal complementarity in prime editing guide RNAs to ease design optimization
Background: Internal guide RNA pairing can affect editing efficiency
Reducing internal complementarity in prime editing guide RNAs (pegRNAs) can improve editing efficiency. SPELL uses this principle and achieved efficiencies close to those of existing designs across several targets without prior PBS optimization. Published in Nucleic Acids Research, the study assessed guide RNA design parameters through literature analysis and experiments in human-derived cells.
Prime editing changes DNA sequences using information encoded in a guide RNA. The primer binding site (PBS) pairs with the 3′ end of the nicked DNA strand to help initiate reverse transcription. However, complementary pairing between the PBS and the spacer region of the same pegRNA can interfere with editing. Choosing PBS length therefore requires considering internal RNA interactions as well as binding to DNA.
The researchers analyzed existing data on about 3,000 pegRNAs and end-protected pegRNAs (epegRNAs). Contrary to prior observations that end protection shortens the optimal PBS length, this analysis found no clear difference between the two types in the PBS length associated with the highest mean editing efficiency. The findings indicate that length criteria derived from a particular target are difficult to generalize.
Key findings: Mismatch designs improved efficiency, while SPELL simplified design
The researchers tested more than 300 modified pegRNAs designed to weaken PBS–spacer interactions. In a plasmid-based fluorescent reporter system in HEK293T cells, certain single-nucleotide spacer mismatches increased editing efficiency to as much as 7.2 times that of the corresponding mismatch-free control. The effect was more pronounced at non-optimal PBS lengths, and not every mismatch was beneficial. Among 46 designs with a single mismatch at a PAM-distal position in the spacer region complementary to the PBS, 24 significantly increased efficiency and one reduced it. The PAM is an adjacent sequence required for Cas9 to recognize its target.
SPELL aims to reduce the burden of extensively optimizing individual combinations. It introduces a single-nucleotide deletion into PBSs 17–20 nucleotides long; in the main comparison at genomic targets, nucleotide 13 of a 20-nucleotide PBS was deleted. This is a guide RNA design change and is distinct from a deletion intended to be introduced into the target DNA.
Across 25 pegRNAs analyzed in HEK293T cells, the overall comparison found no significant difference in editing activity between SPELL and short-PBS designs. This did not establish statistical equivalence. Of the seven design sets included in the comparison, three showed significantly lower efficiency than the existing designs. In the human embryonic stem cell line HuES9, efficiency at the GBA1 target was 1.5 times that of the literature-based design, while no significant difference was found at LRRK2.
Meaning and outlook: An initial design candidate that still requires target-specific validation
SPELL’s main contribution is showing that a relatively simple design—a single-nucleotide deletion in a long PBS—achieved performance close to existing PBS designs across several targets. Combining it with spacer mismatches produced higher efficiency than the optimal PBS at some targets, but finding the best combination required extensive optimization. The maximum 7.2-fold improvement in particular mismatch experiments should therefore not be interpreted as a universal effect of SPELL.
The mechanism is not fully understood. The inhibitory effect of internal complementarity did not always manifest as reduced Cas9 cleavage activity, and the researchers suggested that later steps, such as reverse-transcription initiation, could be affected. Under the tested conditions, there were no signs of increased unintended insertions or deletions or increased editing at some known off-target sites. However, the study did not establish comprehensive safety or therapeutic efficacy in patients. Performance with protein–RNA complex (RNP) delivery also requires further systematic evaluation.
The study design and results can be found in the provided full text of the paper.
The length and sequence of the primer binding site (PBS) are critical for efficient prime editing, and its intramolecular complementarity with the prime editing guide RNA (pegRNA) spacer is a major drawback. We investigated the effects of these factors by literature analyses and by testing over 300 modified pegRNAs with weakened PBS-spacer interactions. It has been suggested that the effective PBS length for plasmid-delivered pegRNAs without end protection is considerably longer than what efficient priming requires due to exonuclease digestion of the PBS ends; however, analysing literature data of over 3000 pegRNAs revealed no significant shift in the optimal PBS length for epegRNAs compared to conventional pegRNAs. We also found improvement in editing efficiency with up to seven-fold when mismatches were introduced in the spacer or PBS sequence disrupting complementarity, although this effect is more pronounced with non-optimal PBS lengths. A combination of spacer mismatches and PBS deletions led to further editing improvements, even compared to the optimal PBS, although finding the best combination requires extensive optimization. Here, we achieved near-optimal editing efficiency in the majority of cases without the need for prior pegRNA optimization by using SPELL ("Streamlined Prime Editing with fixed-Length PBS Leverage"), a prime editing approach that employs a 17-20 nucleotide-long PBS with a single nucleotide deletion.
When designing pegRNAs for a new target in cell-based gene-function studies, SPELL could be considered as an initial candidate to test alongside existing PBS designs. One approach is to check whether a long PBS with a single-nucleotide deletion provides adequate editing efficiency before broadly exploring multiple PBS lengths. Because performance was lower at some targets, editing efficiency and unintended insertions or deletions should be compared for each target. Performance should also be reassessed when the delivery method or cell type changes. These results support the possibility of reducing the burden of design exploration, but they do not establish the actual amount of time or cost saved, shorter drug development timelines, or clinical safety.