Back

Pervasive Horizontal Transfer of Adeno-Associated Virus Capsid Genes.

Gifford, R. J.

2025-03-17 evolutionary biology
10.1101/2025.03.15.643461 bioRxiv
Show abstract

Adeno-associated viruses (AAVs) are non-pathogenic DNA viruses with potent gene delivery capabilities, making them essential tools in gene therapy and biomedical research. Despite their therapeutic importance, key aspects of AAV natural biology remain obscure, complicating efforts to explain rare AAV-associated diseases and optimize gene therapy vectors. By analyzing sequence data from virus isolates and endogenous viral elements (EVEs), I reveal a striking evolutionary pattern: while AAV sub-lineages, defined by the replication-associated (rep) gene, have broadly co-diverged with host groups over millions of years, capsid (cap) diversity has been shaped by extensive recombination. In particular, one capsid lineage, Mammalian-wide (M-wide), has spread horizontally across diverse rep lineages and host taxa through multiple recombination events. Furthermore, several AAVs with M-wide capsids - including AAV-4, AAV-12, and bovine AAV (BAAV) - originate from historical adenovirus (Ad) stocks, raising the possibility that laboratory conditions contributed to capsid transfer. Distinguishing natural from laboratory-driven recombination is essential for understanding AAV ecology and its implications for gene therapy. A systematic sequencing effort in human and primate populations is needed to assess the extent of recombinant capsid acquisition, determine the impact of laboratory-driven recombination on circulating AAV diversity, and track ongoing recombination events that could affect vector safety and efficacy.

Published in Proceedings of the National Academy of Sciences (predicted rank #9) · training set

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.