STRC Research

Could We Fix Just the One Letter?

COMPUTATIONAL

Instead of replacing the entire STRC gene (5325 bp), what if we could correct just the single mutated base? Misha's maternal allele carries c.4976A>C. STRC is on the minus strand — the coding strand mutation is A→C, which requires a C→A correction. Every standard base editor is ruled out. Only prime editing (PE3/PE3b) can make this specific change. PAM survey confirmed: one optimal PAM at 14 nt from the variant.

Michael's mutation: one wrong letter

Healthy: ...AATTTACAGTG...

Michael: ...AATTTCCAGTG...

We need to change C back to A. This is called a C>A transversion.

CBE (Cytosine Base Editor)

What it does:

C → T only

We need C→A. CBE can only do C→T.

Cannot fix this variant

ABE (Adenine Base Editor)

What it does:

A → G only

Wrong direction entirely. Irrelevant here.

Cannot fix this variant

Prime Editor

What it does:

any → any (all 12 substitutions)

C→A included. Installs any base via reverse transcription from a template encoded in the pegRNA. Does not require HDR — works in post-mitotic OHCs.

✓ Only PE3/PE3b can fix this

PAM survey complete: 1 optimal PAM confirmed

STRC is on the minus strand (chr15, strand −1). The coding-strand mutation is C→A (equivalent to G→T on the + strand). Fetched genomic sequence from Ensembl REST API (chr15:43600500-43600620, GRCh38). Searched all NGG PAMs on both strands within 30 bp. Found 5 PAMs total: 1 in the optimal window, 4 in extended range.

chr15:43600521-43600581 (GRCh38)
CCCAGCTCCCCACCTGCTATGGTGCCCCAATTT[C]AGTGAAGATCTCAGG
..........................PAM↑....↑variant
..........................4bp apart

Best PAM: TGG at chr15:43600540 (+strand), 14 nt from the variant — optimal range (3–16 nt). Nick at chr15:43600537. PE3b: nick the non-edited strand with a second guide to reduce indels. 4 additional NGG PAMs on the minus strand at 22–25 nt (extended window, lower efficiency). Editor compatibility verified: CBE, ABE, CGBE, and ACBE are all ruled out by base identity at the mutant position. Only PE3/PE3b is mechanistically viable.

Cochlear Precision Editing: Proven (2025)

Precise gene editing has been demonstrated in cochlear OHCs with functional hearing recovery:

Zhang et al. 2025 (Nature Communications): ABE (SchABE8e) delivered via Anc80L65 AAV to neonatal mice. Targeted a stop codon in POU4F3 (hair cell transcription factor). Near-complete hearing recovery sustained 4+ months. This is the most relevant proof: same AAV serotype used for STRC gene therapy, same cells, functional rescue.

Chen et al. 2024 (Nature Biotechnology): Dual-AAV split-intein prime editor in adult mouse brain. Up to 42% editing in post-mitotic cortical neurons. Also 35% in iPSC-derived cardiomyocytes (Chemla 2025). OHCs are similarly post-mitotic — these are the best efficiency benchmarks available.

No one has prime-edited a cochlear OHC yet. But the full chain is proven in pieces: AAV reaches OHCs (Fang 2021, Iranfar 2026), ABE in OHCs restores hearing (Zhang 2025), and dual-AAV PE achieves ~42% in post-mitotic neurons (Chen 2024). The specific edit for Misha requires PE or ACBE — both are the remaining gaps to close.

Reality check: No one has prime-edited a cochlear OHC. This is a genuine gap. Efficiency in OHCs is unknown — best analogues are cortical neurons (42%, Chen 2024) and cardiomyocytes (35%, Chemla 2025). PE3 requires dual-AAV split-intein (established in retina and brain, not yet cochlea). The PAM survey is done: a TGG PAM at 14 nt from the variant is confirmed optimal. The pegRNA can be designed. The path is technically clear. What's missing: efficiency data in OHCs specifically. Gene replacement (dual-AAV STRC cDNA) covers both alleles and is further clinically advanced. These are parallel tracks, not alternatives.