Immune evasive DNA donors and recombinases license kilobase-scale writing
Summary
This Nature paper describes INSTALL: a set of methods that combine immune‑evasive DNA donors (including circular single‑stranded DNA and CpG‑minimised oligodeoxynucleotides), engineered recombinases (Bxb1 variants and others) and delivery strategies (LNPs or nucleofection) to enable efficient, kilobase‑scale, double‑strand‑break‑free insertion of large DNA cargos into mammalian genomes.
The authors show biochemical and cellular demonstrations of high integration efficiencies using cssDNA and optimised donors (INSTALL‑1, INSTALL‑2, INSTALL‑2e), compatibility with prime and click editing to preinstall attachment sites, and successful delivery in primary human cells (T cells, hepatocytes, iPSCs, K562) and in vivo in neonatal and adult mice. Crucially, minimising immune triggers in the donor DNA and tuning recombinase design reduces innate immune activation (TLR9, cGAS–STING) and improves tolerability in vivo. The study includes long‑read sequencing and genome‑wide assays to assess on‑ and off‑target integration and provides extensive supplementary data, protocols and sequence tables.
Key Points
- INSTALL uses circular single‑stranded DNA (cssDNA) and CpG‑minimised oligodeoxynucleotides (oDNA) as immune‑evasive donors to support kilobase‑scale insertions without double‑strand breaks.
- Engineered recombinases (Bxb1 variants with NLS and stabilising tags) increase integration efficiency and can be delivered as mRNA encapsulated in LNPs or via nucleofection.
- The method is compatible with preinstalling attP sites using prime or click editing, enabling a two‑step WRITE workflow (install site, then integrate cargo).
- INSTALL achieves high integration efficiencies in primary human cells (including T cells and hepatocytes), iPSCs and K562 cells and demonstrates full‑length integrations verified by long‑read sequencing.
- In vivo delivery (neonatal and adult mice) via LNPs yielded integration with reduced innate immune responses when using CpG‑minimised donors (INSTALL‑2e), improving tolerability versus conventional dsDNA donors.
- Approach is compatible with transposon systems (PiggyBac), prime editing, and other emerging genome‑writing tools — offering flexibility for therapeutic and research applications.
- Authors provide extensive datasets, supplementary tables, and sequencing accessions (PRJNA1296083) and discuss potential IP and competing interests (patent applications, industry affiliations).
Content summary
The paper presents a pragmatic route to write kilobase‑scale DNA into mammalian genomes while avoiding major innate immune activation that typically accompanies large DNA donors. The team developed and compared several donor formats (plasmid dsDNA, cssDNA produced by phagemid or enzymatic nick‑and‑digest methods, and optimised oligodeoxynucleotides with reduced CpG content). They show that cssDNA and CpG‑minimised oDNA achieve substantially improved integration when paired with recombinases delivered as mRNA.
Recombinase engineering (different NLS configurations, activity‑enhancing mutations and stabilising tags) and optimisation of dosing and cell‑cycle conditions further boosted efficiency. The work includes both in vitro biochemical reaction evidence and cellular data (ddPCR, flow cytometry, long‑read sequencing). Importantly, LNP delivery of recombinase mRNA with immune‑evasive donors enabled in vivo integrations in mice with markedly reduced inflammatory signatures compared with standard dsDNA donors.
The authors also test combinatorial workflows: writing attP landing pads via prime‑ or click‑editing, then performing one‑pot integration; combining INSTALL donors with PiggyBac transposase for alternative insertion routes; and conducting genome‑wide UDiTaS‑style analyses to assess off‑target events. They highlight translational potential for mutation‑agnostic therapeutic strategies and provide detailed methods, supplementary data and sequence resources.
Context and relevance
This paper sits at the intersection of genome writing, delivery engineering and immunology. It addresses two persistent barriers to therapeutic genome insertion of large cargos: (1) low efficiency of safe, site‑specific kilobase‑scale insertion without double‑strand breaks; and (2) innate immune sensing of exogenous DNA that limits tolerability in primary cells and in vivo. By reducing donor immunogenicity and improving recombinase function/delivery, the authors advance practical approaches for inserting therapeutic transgenes or exons at scale in clinically relevant cell types.
The work is highly relevant to groups working on gene therapy, cell engineering, and synthetic biology — especially teams aiming to deliver multi‑kilobase corrective cassettes (whole exons, reporters, or regulatory modules) to primary cells or directly in vivo. The extensive in vivo tolerability data and off‑target assessment make the study particularly useful for translational planning.
Author style
Punchy: this is a substantial step towards making large, site‑specific insertions feasible in primary cells and in vivo without provoking strong innate immune responses. If you care about moving genome writing from proof‑of‑concept to the clinic, the methodological details and in vivo tolerability data are worth digging into.
Why should I read this?
Short version: they figured out how to sneak large DNA cargo into genomes without setting off the cellular alarm bells — and they show it works in real primary cells and in mice. If you work on gene therapy, T‑cell engineering or any project that needs reliable kilobase‑scale insertions, this paper saves you the trial‑and‑error and points to donor formats, recombinase tweaks and delivery recipes that actually behave in vivo.