Revealing the full potential of Adeno Associated Virus thanks to molecular biology

Adeno-associated virus (AAV) is a small and single strand DNA virus. In 1960, this non-enveloped entity was accidentally discovered in adenovirus culture. It was before we understood that AAV requires the presence of another "helper" virus in order to propagate. This family was named as “Dependovirus genus”.

Recombinant AAV presents many advantages for gene therapy delivery:

• Tropism: thanks to the serotype diversity of AAV, you can specifically target and transduce the required tissue.
• No pathogenicity. AAVs can be easily handled in a BSL-1 compliant environment.
• No integration and limited replication in the host genome. The recombinant form of AAV lacks two essential genes to integrate and replicate, compared to the wild-type strain.
• Long-term gene expression. AAV can persist in non-dividing cells.

In 2012, the first gene therapy product, Glybera, based on an AAV platform, was approved. Unfortunately, despite several promising clinical trials, AAV still faces major concerns. And not only due to its manufacturing costs. Let’s dive into these hindrances at a molecular biology level.

The genome of AAV comprises two ORF: the rep gene and the cap gene. They’re flanked by two-inverted terminal repeats or ITRs. The rep gene encodes, Rep78, Rep68, Rep52 and Rep40, in charge of AAV genome replication and virion assembly. The cap gene encodes the three capsid proteins VP1 (virion protein 1), VP2 and VP3.

Since only ITR elements are essential for AAV propagation, 96% of the entire AAV genome can be removed and replaced by your custom trio “promoter, transgene and polyA tail”.

Great, huh? I guess you’re already thinking about the long transgene you’ve been wanting to test for so long. But, it’s not so simple.

The entire AAV genome is only 4.7kb, so the packaging capacity is highly limited. And this transgene size limitation impacts directly the type of applications of AAV in gene therapy.

So, what can we do? 

• One way to address the problem is to screen for small promoters and thus increase the space available for the transgene. A synthetic polyA can play the same role.
• Another method is to split the transgene between two AAV transfer plasmids and work on homologous recombination to have the appropriate combined expression at the end of the day. 

Once in the nucleus, AAV virions release their single-stranded genome, which is converted into a double-stranded DNA (dsDNA) template from which the transgene can be transcribed. This obligatory molecular step is initiated by the ITRs. Depending on cell machinery, this step can be delayed and the number of genome-containing particles highly impacted.

So, what can we do? 

• The need for rAAV second-strand synthesis after infection can be overcome by mutating one of the wild-type ITRs.
• Alternatively, you can try to generate a self-complementary adeno-associated virus or scAAV. ScAAV contains complementary sequences that are capable of spontaneously annealing. 

The majority of transgenes are derived from natural gene sequences; the codons within them are not optimized. Consequently, the transcription and transduction are not really effective.

So, what can we do? 

Based upon species-specific bias in codon use and tRNA abundance, codon optimization can solve the issue. But performing site-directed mutagenesis can be time consuming. So, prefer the alternative and cost-effective strategies of in-vitro tRNA expression, or de novo gene synthesis.

Toll-like receptor 9 (TLR9) induces immediate innate immune response after cell infection by AAV. After this, AAV transgene has a tendency to induce an adaptive immune response, including a Cytotoxic T lymphocytes (CTLs) mediated cytotoxicity.

So, what can we do?

• The use of a tissue-specific promoter can limit AAV transduction to target cells.
• USP6 glycoprotein from CMV and ICP47 from HSP can inhibit MHC class I. Fusing the genes coding these peptides to the AAV transgene allows to evade CTL-mediated cytotoxicity.
• Engineering the transgene cassette - by eliminating CpG residues or adding (TTAGGG)-like sequence derived from telomeres - to interfere with the innate immune response are both good strategies. 

So far in our article, we have only focused our attention on vector genome. But do you remember that AAV also needs the help of another virus? In fact, transgene is not sufficient alone to generate AAV virions and other elements are necessary: Rep and Cap from the AAV, but also the genome of the Adenovirus.

So, what can we do?

• We can opt for a two-plasmid strategy. One AAV helper, containing Rep and Cap. Another one, the Ad helper, containing genes from adenovirus (E4, E2a and VA) that mediate AAV replication.
• A one-plasmid strategy. Rep/Cap and the adenovirus helper genes are combined into a single plasmid.

Challenges related to AAV production are numerous (packaging and cell line production, adaptation of HEK293 to suspension mode, large-scale transfection for clinical purposes…), but we have made the choice to focus only on plasmid optimization today. The AAV genome hasn't finished to surprise us and viral gene therapy is still in its infancy.