A new genome editing approach that can insert DNA payloads of up to approximately 12 kilobases into defined genomic locations without generating double-strand breaks has been demonstrated in primary human blood stem cells and T cells, according to a study published September 16, 2026 in Nature. The work, led by researchers at Boston Children's Hospital and Dana-Farber Cancer Institute, addresses a persistent challenge in gene therapy: precisely introducing large genetic sequences while avoiding the potential genotoxicity associated with nuclease-induced DNA breaks.
The platform, called prime assembly, extends the logic of prime editing, which uses a Cas9 nickase fused to a reverse transcriptase to make targeted single-strand edits. Prime assembly uses engineered guide RNAs to generate complementary single-stranded DNA flaps at a genomic target, allowing exogenous donor DNA to anneal before endogenous repair machinery completes the integration. The researchers used the approach to assemble as many as four separate DNA fragments totaling approximately 12.1 kilobases at a single genomic site.
A key feature of the approach is its ability to function in non-dividing cells. Homology-directed repair (HDR), a conventional method for precise large-sequence insertion, is strongly dependent on cell-cycle state and performs poorly in resting cells. In primary CD3+ T cells from four healthy donors, prime assembly produced targeted integration at the TRAC locus, a site relevant to CAR-T cell engineering. In CD34+ hematopoietic stem and progenitor cells (HSPCs) from three healthy donors, integration was demonstrated at the IL2RG and AAVS1 loci, with performance matching or exceeding HDR under the conditions tested.
Pharmacological experiments also implicated Polθ-mediated end joining in the integration process. Inhibition of Polθ reduced integration efficiency, while inhibition of DNA-PK-dependent non-homologous end joining with AZD7648 increased it.
Beyond DNA insertion, the researchers used prime assembly to generate megabase-scale genomic rearrangements, including deletions, inversions, and interchromosomal alterations, suggesting potential applications in modeling structural genomic changes as well as therapeutic gene insertion.
The work is not an isolated technical advance. Several groups have reported related double-strand-break-free approaches for inserting large DNA sequences during 2026, indicating that the field is converging on complementary strategies for overcoming one of the major limitations of conventional prime editing.