Adeno-associated virus: from defective virus to effective vector

Abstract

The initial discovery of adeno-associated virus (AAV) mixed with adenovirus particles was not a fortuitous one but rather an expression of AAV biology. Indeed, as it came to be known, in addition to the unavoidable host cell, AAV typically needs a so-called helper virus such as adenovirus to replicate. Since the AAV life cycle revolves around another unrelated virus it was dubbed a satellite virus. However, the structural simplicity plus the defective and non-pathogenic character of this satellite virus caused recombinant forms to acquire centre-stage prominence in the current constellation of vectors for human gene therapy. In the present review, issues related to the development of recombinant AAV (rAAV) vectors, from the general principle to production methods, tropism modifications and other emerging technologies are discussed. In addition, the accumulating knowledge regarding the mechanisms of rAAV genome transduction and persistence is reviewed. The topics on rAAV vectorology are supplemented with information on the parental virus biology with an emphasis on aspects that directly impact on vector design and performance such as genome replication, genetic structure, and host cell entry.

Figure 1

Transmission electron microscopy of AAV2 and Ad5 particles in human cells. (A) AAV2 and Ad5 particles in the nucleus of a HeLa cell at 48 hours after co-infection. Magnification: × 15,000. (B) AAV2 virions in a HeLa cell at 48 hours after co-infection with Ad5. Magnification: × 40,000.

Figure 2

Secondary structure of the AAV2 ITR. The AAV2 ITR serves as origin of replication and is composed of two arm palindromes (B-B' and C-C') embedded in a larger stem palindrome (A-A'). The ITR can acquire two configurations (flip and flop). The flip (depicted) and flop configurations have the B-B' and the C-C' palindrome closest to the 3' end, respectively. The D sequence is present only once at each end of the genome thus remaining single-stranded. The boxed motif corresponds to the Rep-binding element (RBE) [119] where the AAV Rep78 and Rep68 proteins bind. The RBE consists of a tetranucleotide repeat with the consensus sequence 5'-GNGC-3'. The ATP-dependent DNA helicase activities of Rep78 and Rep68 remodel the A-A' region generating a stem-loop that locates at the summit the terminal resolution site (trs) in a single-stranded form [120,121]. In this configuration, the strand- and site-specific endonuclease catalytic domain of Rep78 and Rep68 introduces a nick at the trs. The shaded nucleotides at the apex of the T-shaped structure correspond to an additional RBE (RBE') [121] that stabilizes the association between the two largest Rep proteins and the ITR.

Figure 3

Schematic representation of the AAV DNA replication model. AAV DNA replication is thought to involve a self-priming single-strand displacement mechanism that is initiated by DNA polymerisation at the 3' hairpin primer of input single-stranded genomes. This leads to the formation of linear unit-length double-stranded molecules (duplex monomers, DMs) with one covalently closed end. These structures are resolved at the terminal resolution site (trs) by site-specific nicking of the parental strand opposite the original 3' end position (i.e., at nucleotide 125). The newly generated free 3' hydroxyl groups provide a substrate for DNA polymerases that unwind and copy the inverted terminal repeat (ITR). Finally, the palindromic linear duplex termini can renaturate into terminal hairpins putting the 3' hydroxyl groups in position for single-strand displacement synthesis. Next, single-stranded genomes and new DM replicative forms are made. When nicking does not occur, elongation proceeds through the covalently closed hairpin structure generating linear double-length double-stranded molecules (duplex dimers, DDs) with either a head-to-head or a tail-to-tail configuration. The DD replicative intermediates can be resolved to DMs through the AAV ITR sequences located at the axis of symmetry.

Figure 4

Overview of the initial recombinant AAV production system. The generation of the first infectious clones of AAV permitted functional dissection of the virus genome. This allowed the construction of plasmids encoding rAAV genomes in which the minimal complement of wild-type sequences necessary for genome replication and packaging (i.e., the AAV ITRs) frame a gene of interest (transgene) instead of the AAV rep and cap genes. When these constructs are transfected into packaging cells together with a rep and cap expression plasmid they lead to the production of rAAV particles. Helper activities required for the activation and support of the productive phase of the AAV life cycle were originally introduced by infection of the packaging cells with wild-type Ad as depicted. Current transfection-based production methods make use of recombinant DNA encoding the helper activities instead of Ad infection. Cellular DNA polymerase activities together with the Rep78 and Rep68 proteins lead to the accumulation of replicative intermediates both in the duplex monomer (DM) and duplex dimer (DD) forms. A fraction of this de novo synthesized DNA is incorporated in the single-stranded format into preformed empty capsids most likely through the catalytic activities of the Rep52 and Rep40 proteins. The resulting infectious rAAV virions are released from the producer cells together with helper Ad particles. Sequential heat treatment and buoyant density centrifugation allows the selective elimination of the helper virus from the final rAAV preparation.

Figure 5

Diagram of the recombinant AAV split gene principle. An expression unit corresponding to a large gene is roughly divided in two halves. One of them consists of a promoter (solid box with arrowhead), the 5' half of the gene (open box) and a splice donor site (SD) while the other encodes a splice acceptor sequence (SA), the 3' portion of the gene (shaded box) and a polyadenylation signal (solid box). These fragments are independently cloned between two AAV ITRs. Vector stocks are then generated from the resulting shuttle plasmids and are used to co-transduce target cells. Head-to-tail heterodimerization via intermolecular recombination between the two vector DNA molecules restores the full-length expression unit and results in the synthesis of the desired protein after the splicing of the intervening AAV ITR sequences from the primary transcript.

Figure 6

Diagram of the generation and transduction of a self-complementary AAV vector as compared to that of a conventional recombinant AAV. Left panel: According to the AAV DNA replication scheme, full-length rAAV genomes of both polarities are generated from duplex monomeric (DM) and duplex dimeric (DD) replicative intermediates and individually packaged in AAV capsids. In the nucleus of transduced cells the single-stranded genomes can either be a target for degradation or be converted into transcriptionally active double-stranded templates. The single-to-double strand DNA conversion depends on complementary chain synthesis or on the recruitment of a complementary genome (i.e., intermolecular hybridization). Right panel: According to the same replication model, a rAAV genome with roughly half the size of the wild-type AAV DNA and with one trs mutated, generates DD replicative intermediates with an inverted repeat configuration containing wild-type ITRs at the extremities and mutated ITRs at the axis of symmetry. Single-stranded molecules derived from these DNA structures are packaged in AAV capsids. After uncoating in the target cell nucleus, these molecules can readily fold into double-stranded templates through intramolecular base pairing due to their self-complementary nature (i.e., intramolecular hybridization).