Engineered virus-like particles evolve for superior gene delivery efficiency

Breakthrough directed evolution system boosts virus-like particles for safer, more efficient gene editing, offering a powerful tool for next-generation therapeutic innovations.

Study: Directed evolution of engineered virus-like particles with improved production and transduction efficiencies. Image Credit: Dragon Claws / Shutterstock.com

A recent study published in the journal Nature Biotechnology discusses the development of a novel system for the directed evolution of engineered virus-like particles (eVLPs) with enhanced transduction and production capabilities.

What are eVLPs?

The ability to efficiently and safely deliver macromolecules into cells in vitro and in vivo is crucial for various emerging treatment modalities. Although adeno-associated virus (AAV) vectors can successfully deliver gene-editing agents in vivo, they are associated with several limitations.

Therefore, additional delivery methods are needed to overcome these limitations. VLPs comprise viral scaffolds that package and deliver cargo messenger ribonucleic acids (mRNAs), proteins, or ribonucleoproteins (RNPs). VLPs offer efficient transduction, tissue tropisms, reduced off-target editing, and transient cargo expression.

Previously, the current study’s authors developed eVLPs that enable efficient gene editing and protein delivery in vitro and in vivo. Within these systems, cargo proteins are fused to retroviral Gag proteins in eVLPs, which direct cargo localization into viral particles as they form. The cargo-Gag linker contains a sequence to be cleaved by a retroviral protease after particle formation, which subsequently releases the cargo inside the particles and into the transduced cells.

Study findings

The researchers developed a directed laboratory evolution system for eVLPs. Initially, the identity of eVLP variants was elucidated using barcoded single-guide RNAs (sgRNAs) to allow for the selection of desired variants with specific properties. The compatibility of the barcoded sgRNAs with functional eVLP production was subsequently developed, following which a 15-base pair barcode sequence was inserted into the tetraloop of the sgRNA scaffold.

Fourth-generation (v4) base-editor (BE)-eVLPs, which package an active adenine BE (ABE) RNP cargo, were used for the validation experiments. Standard v4 BE-eVLPs were produced by co-transfecting four plasmids into producer cells, which encoded the Gag-ABE fusion, Moloney murine leukemia virus (MMLV) Gag-Pro-Pol polyprotein, vesicular stomatitis virus G envelop protein, and sgRNA directing on-target base editing.

Thereafter, v4 eVLPs containing tetraloop or canonical-barcoded sgRNAs with four arbitrarily selected barcodes were produced and compared by measuring their base editing efficiencies. Barcoded eVLPs were found to exhibit potency comparable to standard eVLPs, whereas eVLPs with distinct barcoded sgRNAs had comparable potencies. Furthermore, eVLPs lacking the Gag-ABE fusion packaged 216-fold fewer sgRNAs than canonical v4 eVLPs.

Additional experiments indicated that barcoded sgRNAs could be used to label distinct eVLP variants and that barcodes enriched after selection identify variants with enhanced fitness.

This evolution system was then applied to mutate and select capsids with improved features. To this end, a barcoded eVLP capsid library containing 3,762 single-residue mutants of MMLV Gag protein capsid and nucleocapsid domains in the Gag-ABE cargo was generated.

This barcoded library was used to construct a library of barcoded eVLP producer cells. Lentiviral transduction of producer cells, followed by expansion of transduced cells, amplified the fraction of producer cells with a single barcode-capsid variant pair.

The barcoded eVLP capsid library was subject to two selections for improved production from producer cells and transduction of HEK293T cells, respectively. To this end, eVLP production was initiated from the barcoded producer cell library, with the resultant library of capsid variants purified.

Thereafter, eVLP-packaged sgRNAs were isolated, and barcodes present after this production selection were sequenced. The enrichment of eVLP production was estimated for each barcode sequence, which identified barcodes with increased enrichment as compared to those of canonical eVLP capsids. About 8% of capsid mutants in the library had higher production enrichment than the canonical eVLP capsid.

HEK293T cells were incubated with the purified barcoded eVLP capsid library. After six hours, sgRNAs transduced into target cells were isolated, and the eVLP transduction enrichment was calculated for each barcode sequence.

Only 0.7% of all capsid mutants had an average transduction enrichment higher than that of the canonical v4 eVLP capsid. Although most capsid mutants had worse transduction and production efficiencies than the canonical v4 eVLP capsid, no mutants exhibited improvements in their production or transduction, thus suggesting that distinct and competing mechanisms dictate their efficiencies.

Several mutants were selected for further analyses based on positive production or transduction selection enrichments. Mutants that improved one property without impairing the other were prioritized.

Introducing capsid mutations into the Gag-ABE construct did not increase potency. This led the researchers to assess the potency of capsid mutants in the Gag-ABE construct with a Q226P mutation, which was most robustly enriched in production selection in the Gag-Pro-Pol construct (Gag^Q226P-Pro-Pol).

Various capsid mutants increased BE delivery potency by up to three-fold as compared to v4 eVLPs. Five mutations, including C507V, C507F, A505W, D502Q, and R501I, had the highest potency; however, one mutant combination, GagC507V-ABE with GagQ226P-Pro-Pol, had 3.7-fold improved potency and was designated as the fifth generation (v5) BE-eVLPs.

The potencies of v4 and v5 BE-eVLPs were compared, which indicated that v5 BE-eVLPs had significantly higher base editing efficiencies and were more potent than v4 BE-eVLPs. Interestingly, the maximum editing efficiency achieved with v4 BE-eVLPs was attained with a 16-fold lower dose of v5 BE-eVLPs.

Conclusions

The researchers of the current study developed a directed evolution system for eVLPs with desired properties and employed this system to mutate/select eVLP capsid mutants with enhanced properties. Moreover, v5 eVLPs with enhanced cargo packaging and release, greater particle sizes, and higher delivery potency than v4 eVLPs were developed. Overall, this barcoded eVLP evolution method could support the development of future delivery vehicles that overcome the limitations associated with current gene editing systems.