Adeno-associated virus (AAV) is a popular viral vector used for gene delivery and has rapidly overtaken other delivery systems as the vector of choice for in vivo applications and therapeutics development. This is largely due to its enhanced safety profile, owing to its lack of pathogenicity and low immunogenicity, as well as broad tissue tropism enabled by the diversity of naturally occurring serotypes. Throughout this article, we will discuss how AAV became so widely used, particularly for therapeutics development, and explore how AAV’s natural biology can be harnessed for efficient gene delivery, highlighting its key advantages and limitations as a viral vector.

Basics of AAV

AAV was first identified in the 1960s as a contaminant in adenovirus preparations. Follow-up studies revealed that this novel parvovirus could not replicate efficiently in human cells without the presence of a "helper" virus. This dependence led to its classification as a dependovirus, with successful propagation achieved only through coinfection with other viruses, such as adenovirus or herpes simplex virus, which provide additional factors essential for replication.

Further research uncovered that AAV is a small, non-enveloped virus with a single-stranded DNA (ssDNA) genome of about 4.7 kb enclosed within a protein capsid (Figure 1A). The two main open reading frames (ORFs) within the AAV genome, Rep and Cap, encode non-structural replication proteins (Rep78, Rep68, Rep52, and Rep40) and structural capsid proteins (VP1, VP2, and VP3), respectively (Figure 1B). The structural proteins assemble into 60 subunits (5x VP1, 5x VP2, and 50x VP3) which make up the icosahedral capsid. Recently, additional overlapping ORFs within the cap gene were identified, encoding assembly-activating protein (AAP), which facilitates capsid assembly, and membrane-associated accessory protein (MAAP), which promotes viral egress. Flanking the genome are the inverted terminal repeats (ITRs). These are highly structured elements, which self-anneal to form a T-shaped hairpin, containing sequences essential for replication and packaging of the viral genome into new virions (Figure 1B).

The AAV life cycle begins with virions binding to receptors on the host cell surface, followed by internalization via clathrin-mediated endocytosis or other forms of endocytosis. Following endosomal escape, the virus is either targeted to the proteasome for degradation or trafficked to the nucleus, where it undergoes uncoating and releases its ssDNA genome. The genome is then converted into double-stranded DNA (dsDNA) by the host cell machinery. If helper viruses are present, such as adenovirus or herpes simplex virus, the life cycle then continues into a productive (lytic) phase. The dsDNA is used as a template for transcription, producing mRNA, which is exported to the cytoplasm for translation into viral proteins. Rep proteins are expressed first, as these are essential for viral replication and efficient transcription, prior to production of the capsid proteins VP1, VP2, and VP3, which are transported into the nucleus for the formation of empty capsids. The dsDNA is also used as a template for production of more copies of genomic ssDNA, which can then be packaged into empty capsids, and the newly formed virions exit the nucleus and travel to the host membrane where they are released from the cell. Without helper viruses, the wild-type (WT) AAV remains latent within the host, persisting as episomal circular monomers or concatemers, until coinfection with a helper virus triggers reactivation and lytic infection (Figure 2).

Recombinant AAV

Due to AAV’s relatively simple genomic and virion structure, it can be easily modified to deliver user-selected cargo by replacing the native ssDNA genome with a user-defined DNA sequence of up to approximately 4.2 kb in length, inserted between the ITRs (Figure 3).

Packaging of a user-selected sequence into AAV can be achieved using two main approaches: triple transfection or the baculovirus-based system for large-scale manufacturing. In either case, the production of recombinant virions relies on supplying the necessary viral proteins required for replication and capsid assembly, along with the user-selected sequence to be packaged. 

The triple-transfection approach is most commonly used and involves the co-transfection of three plasmids that provide the genes and regulatory elements necessary for AAV replication, packaging, and capsid assembly. These include: (1) the transfer plasmid, which encodes the user-selected gene of interest (GOI); (2) the RepCap/packaging plasmid, which provides the replication proteins (Rep) required for genome replication and capsid proteins (Cap) that determine the serotype for targeting; and (3) the helper plasmid, which supplies adenoviral factors (E4, E2A, and VA) essential for efficient replication (Figure 4A). Following transfection of these plasmids into a packaging cell line which stably expresses helper factors E1A/E1B, viral particles are harvested, concentrated, and purified for downstream applications (Figure 4B).

Several plasmid design considerations, particularly for the transfer and Rep/Cap plasmids, are important, depending on the intended use of your AAV vector. For the transfer plasmid, the promoter, the transgene, and several components within the vector backbone can be optimized to ensure your AAV vector is primed for optimal performance (Figure 4A).

For the Rep/Cap plasmid, inclusion of different Cap genes encoding distinct capsid proteins produces different AAV serotypes, each displaying unique tissue tropisms. A variety of naturally occurring and engineered serotypes with known tissue tropisms can be selected depending on the desired target tissue (Table 1). For example, some naturally occurring serotypes, such as AAV9, have broad targeting capabilities, while engineered serotypes, such as AAV-PHP.eB and AAV-PHP.S, have more specific tropisms, targeting cells within the central nervous system (CNS) and peripheral nervous system (PNS), respectively.

When pre-existing serotypes are suboptimal for your desired application, novel AAV serotypes can be engineered through capsid evolution. This involves generating and screening large libraries of capsid variants to identify those with improved targeting and delivery. This approach addresses key challenges in AAV-based therapeutics, including improved targeting of difficult sites, such as the CNS and retina, and evasion of host immunity to pre-existing serotypes, which can otherwise reduce transduction efficiency and cause adverse effects. 

AAV applications

AAV can transduce diverse cell types and is used for a broad range of in vitro applications where transient expression is desired, including overexpression, shRNA knockdown, CRISPR, and library screening. While AAV can support prolonged expression in non-dividing cells, integrating viruses like lentivirus are typically used for long-term expression in dividing cells, though they carry a risk of insertional mutagenesis, which may disrupt non-target genes (Table 2).

Generally, for in vivo delivery and therapeutic development, AAV is the vector of choice due to its enhanced safety profile compared to other viral vector systems (Table 2). Unlike lentivirus, AAV replicates episomally and does not integrate into the host genome, greatly reducing the risk of insertional mutagenesis which may cause cancer. Compared to adenovirus, which is also non-integrating, AAV exhibits much lower immunogenicity and pathogenicity in vivo. Overall, AAV vectors carry a lower risk of side effects and toxicity than other viral vectors. This, combined with the potential to engineer novel variants with enhanced immune evasion and tissue targeting, makes AAV a promising vector for developing new cell and gene therapies.

Several AAV-based gene therapies are already FDA-approved, whereby AAV is used to deliver a functional copy of a specific gene to treat disorders such as inherited blindness, spinal muscular atrophy, Duchenne's muscular dystrophy, and hemophilia. This highlights AAV’s proven track record for safety and efficacy, and also means regulatory and manufacturing processes are already in place, making the transition to clinic a much smoother process.

	AAV	Adenovirus	Lentivirus	MMLV
Stable or transient	Transient expression	Transient expression	Stable long-term expression	Stable long-term expression
Tropism	Depending on serotype	Narrower	Broad	Broad
Cargo capacity	~4.7 kb (4.2 kb)	~36 kb (7.5-33 kb)	~9.2 kb (6.4 kb)	~8 kb (5.5 kb)
Popular applications	In vivo preclinical and clinical	In vivo	In vitro and ex vivo	In vitro

Conclusion

AAV has become the vector of choice for therapeutic applications, due to its enhanced safety profile compared to other viral vectors and flexible cell and tissue targeting due to the existence of many different serotypes. Through considered vector design, AAV can be harnessed as a highly efficient gene delivery tool for a broad range of applications and holds significant potential for the future of therapeutics development. By utilizing approaches such as capsid evolution, novel serotypes with enhanced properties can be generated, extending AAV’s therapeutic reach to target notoriously difficult sites, such as the brain and retina. AAV’s limited cargo capacity can be addressed through strategies like promoter screening, which reduce vector component size and create more space for the transgene, thereby expanding the range of potential therapeutic gene targets. 

At VectorBuilder, we have invested significant time and resources into the research and development of AAV vectors to enable the production of high-titer, high-quality viruses with enhanced targeting and expression. We offer a wide range of AAV solutions to support you from vector design all the way through to GMP manufacturing and clinical studies, and our design team is always on hand to help with all your AAV needs.

References

Wang, JH., Gessler, D.J., Zhan, W. et al. Adeno-associated virus as a delivery vector for gene therapy of human diseases. Sig Transduct Target Ther 9, 78 (2024).

FDA U.S. Food and Drug Administration. Approved Cellular and Gene Therapy Products.

Xingjian Bai, Jack F Hong, Shan Yu, David Y Hu, Amy Y Chen, Constance A Rich, Silk J Shi, Sandy Y Xu, Daniel M Croucher, Kristofer J Müssar, Daniel W Meng, Jane L Chen, Bruce T Lahn. Prevalence of errors in lab-made plasmids across the globe, Nucleic Acids Research, Volume 53, Issue 14, 12 August 2025.

Lopez-Gordo, E.; Chamberlain, K.; Riyad, J.M.; Kohlbrenner, E.; Weber, T. Natural Adeno-Associated Virus Serotypes and Engineered Adeno-Associated Virus Capsid Variants: Tropism Differences and Mechanistic Insights. Viruses 2024, 16, 442.