Both base editing and prime editing emerged from the Liu lab as alternatives to standard CRISPR-Cas9 — alternatives that avoid double-strand breaks while still making precise changes. They’ve quickly become the methods of choice for many edits, especially in therapeutic contexts.
Why avoid double-strand breaks?
Standard CRISPR-Cas9 cuts both DNA strands and relies on cellular repair pathways. NHEJ creates indels (good for knockouts, bad for precise edits), and HDR is inefficient. DSBs also produce unwanted outcomes: large deletions, translocations, p53 activation, and genotoxic stress. Base and prime editors sidestep these issues.
Base editing: precise base swaps
Base editors fuse a catalytically impaired Cas9 (nickase or dead Cas9) to a deaminase enzyme. The result: a complex that binds DNA, exposes a small “editing window” of single-stranded DNA, and chemically converts one base to another — without ever fully cutting the DNA.
Two main classes:
- Cytosine base editors (CBEs): Convert C → T (and the complementary G → A). Built around APOBEC or similar cytidine deaminases
- Adenine base editors (ABEs): Convert A → G (and the complementary T → C). Built around an evolved adenosine deaminase
Together, CBEs and ABEs can install all four transition mutations.
Prime editing: most edits, no donor needed
Prime editing uses a Cas9 nickase fused to a reverse transcriptase, guided by a prime editing guide RNA (pegRNA) that contains:
- A spacer that targets the genomic site (like a normal sgRNA)
- A primer binding site (PBS) — anneals to the nicked DNA
- A reverse transcriptase template (RT template) — encodes the desired edit
The complex nicks one DNA strand, the reverse transcriptase extends from the nicked end using the pegRNA as a template, and the new sequence — containing the edit — is incorporated into the genome.
Prime editing can install all 12 possible base substitutions plus small insertions and deletions (up to ~50 bp routinely, longer with newer designs).
Comparison
| Feature | Base editing | Prime editing |
|---|---|---|
| Edit types | Transitions only (C↔T, A↔G) | All substitutions, small indels |
| Editing window | ~5 nt within target | Defined by pegRNA |
| Bystander edits | Common in window | Rare |
| Off-target editing | Higher (RNA editing too) | Lower |
| Efficiency | Typically 30–80% | Typically 5–40% |
| Construct size | Smaller | Larger |
| Design complexity | Lower | Higher (pegRNA tuning) |
When to use each
Base editing is best when…
- You need a transition mutation (C↔T or A↔G) and there are no bystander cytosines/adenines in the editing window
- High efficiency matters more than absolute precision
- You’re editing in vivo where construct size is constrained (AAV)
- You need to silence stop codons or introduce them
Prime editing is best when…
- You need a transversion (C↔G or A↔T) — base editors can’t do these
- You need a small insertion or deletion at a defined position
- You need to avoid bystander edits
- You’re correcting a precise pathogenic variant
Recent improvements
- PE3, PE4, PE5 prime editing systems: Use additional nicks or DNA repair modulation to improve efficiency
- Twin prime editing: Two pegRNAs flanking a region for larger insertions, deletions, or replacements
- Engineered base editors: Narrower editing windows, reduced off-target activity, expanded compatibility
- Smaller variants: SaCas9-based base and prime editors fit in a single AAV
Therapeutic implications
Several base editing and prime editing programs are now in clinical trials, including for sickle cell disease, transthyretin amyloidosis, and familial hypercholesterolemia. The lower DSB burden makes these editors particularly attractive for in vivo therapeutics where chromosomal stability is critical.
Choose base editing for high-efficiency transition edits where bystander positions are clean. Choose prime editing for any precision edit outside the base editing scope, especially transversions and small indels.
