Orthogonal Transposons for Iterative Genome Engineering of Mammalian Cells.

This paper introduces a robust, iterative genome engineering framework for mammalian cells using three mutually orthogonal Leap-In transposase systems to sequentially generate a glutamine synthetase-deficient CHO host, integrate therapeutic antibodies, and modulate glycosylation, thereby enabling the predictable and stable development of complex biopharmaceutical manufacturing lines.

The Gist

Imagine you are trying to build a high-tech factory inside a tiny, living cell to produce life-saving medicines. For decades, scientists have used a specific type of cell (called a CHO cell) as their factory floor. But as the medicines we need have become more complex—like custom-made antibodies or drugs that need very specific "wrapping" (sugar coats)—the old tools for building these factories have started to break.

This paper introduces a revolutionary new way to build these cellular factories, using a system the authors call "Orthogonal Transposons."

Here is the simple breakdown using a few creative analogies:

1. The Problem: The "One-Key" Lock

Imagine you have a house (the cell) and you want to install three different security systems: a new front door, a smart garage, and a high-tech alarm.

• The Old Way: You use the same key to install all three. The problem? When you go to install the second system, the key accidentally unlocks and removes the first one. By the time you finish, your house is a mess, and nothing works. In biology, this is called "cross-mobilization," where the tool used to insert a gene accidentally deletes the genes you already put there.

2. The Solution: The "Master Key" Set

The researchers at ATUM developed a toolbox of three completely different keys (transposases) that fit three completely different locks (transposons).

• Key A only opens Lock A.
• Key B only opens Lock B.
• Key C only opens Lock C.

Because they are "orthogonal" (meaning they don't interfere with each other), you can use Key A to install a door, then Key B to install a garage, and Key C to install an alarm. None of them will touch or break the others. This allows for iterative engineering—building the factory step-by-step without breaking what you've already built.

3. The Three-Step Construction Project

The paper describes a real-world test where they built a "super-cell" in three distinct steps:

• Step 1: The Foundation (The Metabolic Switch)

  • Goal: They needed the cell to rely on a specific nutrient (glutamine) so they could control its growth.
  • Action: They used Transposon A to turn off the cell's natural ability to make its own food.
  • Result: A "starved" cell that can only survive if you feed it the right nutrient. This acts as a safety switch.

• Step 2: The Product (The Medicine)

  • Goal: Now, they needed the cell to actually make the medicine (an antibody).
  • Action: They used Transposon B to insert the instructions for the antibody and a backup supply of the food the cell needs.
  • Result: The cell is now healthy, hungry for the specific nutrient, and churning out the medicine. They managed to insert 33 copies of the medicine-making instructions!

• Step 3: The Polish (The Sugar Coat)

  • Goal: The medicine needed a specific "sugar coat" to work better in the human body (specifically, to remove a sugar called fucose).
  • Action: They used Transposon C to turn off the machine that adds that specific sugar.
  • Result: The cell now produces the medicine with a "perfect" sugar coat, without messing up the 33 copies of the medicine instructions or the metabolic switch from Step 1.

4. The "WYSIWYG" Guarantee

In computer terms, WYSIWYG means "What You See Is What You Get."

• In the old days, when scientists tried to edit cells, the cell's internal repair crew would often mangle the instructions, creating a "broken" version of the medicine.
• With this new Leap-In system, the instructions go into the cell exactly as designed. If you design a perfect blueprint on your computer, the cell builds that exact blueprint. No surprises, no broken parts.

5. Why This Matters

• Stability: They tested these cells for over 240 generations (like running a factory for years). The "keys" never got mixed up, and the factory never broke down.
• Speed: This allows scientists to build complex, custom cell lines much faster, getting new drugs to patients sooner.
• Safety: Because they can map exactly where every piece of DNA landed (using a technique called TLA), they can prove to regulators that the cell line is safe and consistent.

In a nutshell: This paper proves that by using a set of "non-interfering" genetic tools, scientists can now build complex, multi-layered cellular factories step-by-step, ensuring that every layer stays exactly as they designed it. It's the difference from trying to build a skyscraper with a sledgehammer versus using a precision 3D printer.

Technical Summary

1. Problem Statement

The biopharmaceutical industry is shifting toward complex biotherapeutics (e.g., multispecific antibodies, ADCs, and glycoengineered drugs) that require precise control over gene dosage, chain ratios, and metabolic backgrounds. However, current genome engineering tools for Chinese Hamster Ovary (CHO) cells face significant limitations:

• Low Efficiency: Homology-directed repair (HDR) in nuclease-mediated editing is inefficient.
• Instability: Random integration methods often lead to genomic instability and unpredictable expression.
• Iterative Limitations: Traditional transposon systems (e.g., piggyBac, Sleeping Beauty) lack orthogonality. Re-expressing the same transposase for a second engineering step risks "cross-mobilization," where the enzyme excises or rearranges previously integrated transgenes, destroying the engineered cell line's integrity.

2. Methodology

The authors utilized ATUM's Leap-In Transposase® platform to develop a "What You See Is What You Get" (WYSIWYG) iterative engineering paradigm.

• Orthogonal System Development:

  • Three distinct, mutually orthogonal transposase-transposon pairs (Systems A, B, and C) were engineered from three different organisms.
  • Design: Each transposase was optimized via machine learning for hyperactivity. Each transposon vector contains unique Inverted Terminal Repeats (ITRs) that are recognized only by their cognate transposase, preventing cross-reactivity.
  • Vector Architecture: Vectors included optimized promoters, UTRs, signal peptides, and codon-optimized sequences to maximize expression and stability.

• Iterative Engineering Workflow (Three-Step Process):

  1. Step 1 (Metabolic Engineering): Transfection of wild-type CHO-K1 cells with Transposon A (knocking down endogenous Glutamine Synthetase, GS) and Transposase A. This created a GS-deficient host (miCHO-GS) selectable via glutamine-free media.
  2. Step 2 (Therapeutic Expression): Transfection of the miCHO-GS host with Transposon B (encoding an IgG1 antibody heavy/light chain + a recombinant GS gene) and Transposase B. The recombinant GS rescued the host, allowing selection and high-titer antibody production.
  3. Step 3 (Glyco-engineering): Transfection of the antibody-expressing host with Transposon C (encoding a knockdown of FUT8, a fucosyltransferase) and Transposase C. This modulated the glycan profile to enhance Antibody-Dependent Cell-mediated Cytotoxicity (ADCC).

• Validation & Analytics:

  • Targeted Locus Amplification (TLA): Used in partnership with Solvias to map exact integration sites, count copy numbers, and verify structural integrity without prior knowledge of integration loci.
  • NGS: Confirmed sequence fidelity.
  • Functional Assays: Measured cell growth, viability, IgG1 titers (Protein A HPLC), and glycan profiles (HILIC).
  • Stability Studies: Clones were cultured for up to 240 generations to assess genetic and phenotypic stability.

3. Key Contributions

• Proof of Orthogonality: Demonstrated that three distinct transposase-transposon pairs can be used sequentially in the same cell line without cross-mobilization or rearrangement of previous integrations.
• WYSIWYG Paradigm: Established a workflow where the designed vector sequence is transferred 1-to-1 into the genome, eliminating the unpredictable truncations, concatemers, and mutations common in traditional random integration.
• Scalable Iterative Engineering: Successfully engineered a single cell line through three distinct functional modifications (metabolic selection, therapeutic expression, and glycan modulation) while maintaining high stability.
• Regulatory Readiness: Provided comprehensive molecular characterization (TLA/NGS) suitable for IND/BLA submissions, addressing a critical gap in cell line development for complex biologics.

4. Key Results

• Step 1 (GS Knockdown): Generated a stable miCHO-GS host with 6 distinct integration loci for Transposon A. The host failed to grow in glutamine-free media, confirming successful GS knockdown.
• Step 2 (Antibody Expression): The addition of Transposon B resulted in 33 integration loci. The cell line showed full rescue of the GS phenotype and achieved antibody titers exceeding several grams per liter.
  • Stability: After ~150 generations, all 39 integration sites (6 from A + 33 from B) remained structurally intact with no sequence alterations.
• Step 3 (Glyco-engineering): Introduction of Transposon C added 9 integration loci.
  • Glycan Profile: Total fucose levels dropped to <10%, with a corresponding increase in non-fucosylated G0, G1, and G2 species.
  • Productivity: Specific and volumetric productivity remained high and stable.
  • Total Stability: After 240 generations, all 48 integration sites (6A + 33B + 9C) were confirmed to be at their original chromosomal positions with perfect sequence fidelity. No phenotypic drift was observed.

5. Significance

• Accelerated Drug Development: This platform enables the rapid creation of complex, high-producing cell lines, significantly reducing the "speed-to-clinic" for next-generation biologics.
• Regulatory De-risking: The ability to precisely map and verify every integration site across multiple engineering steps provides the rigorous molecular characterization required for regulatory approval (IND/BLA).
• Future-Proofing Biomanufacturing: The orthogonal toolbox allows for the "retroactive" engineering of existing cell lines to improve product quality (e.g., changing glycosylation) or to build complex multi-chain molecules (e.g., bispecifics) without starting from scratch.
• Industry Validation: The authors note that as of February 2026, 50 INDs have been filed for molecules produced using Leap-In technology, validating the platform as a robust manufacturing engine rather than just a research tool.

In conclusion, this study establishes a robust, predictable, and scalable framework for iterative mammalian cell engineering, solving the critical bottleneck of maintaining genomic integrity during sequential modifications required for advanced biotherapeutics.