A dual-reference study design for understanding and improving AAV genome size analysis.

Recombinant adeno-associated virus (rAAV) is the leading platform of gene delivery for its long-lasting gene transformation and low immunogenicity. Characterization of the integrity and purity of the rAAV genome is critical to ensure clinical potency and safety. However, current rAAV genome characterization methods that can provide size assessment are either time-consuming or not easily accessible to general labs. Additionally, there is a lack of right reference standard for analyzing long single-stranded DNA (ssDNA) fragments. Here, we have developed an ssDNA assay on a microfluidic capillary electrophoresis platform using ssDNA reference standard. This assay provides size calling for ssDNA fragment, a detection sensitivity at ∼89 pg/µL (3 × 1010  GC/mL AAV) for 5.1 kb ssDNA fragment, and a turnaround time at ∼100 s per sample with a high throughput sample analyzing capability. Moreover, we have observed that the annealing of AAV ssDNA subsequent to its release from the capsid might introduce an additional double-stranded DNA (dsDNA) peak. This phenomenon is dependent on the sample processing workflow. To avoid the risk of mischaracterization, we recommend the use of dual-reference standards in combination with other orthogonal methods to have a comprehensive understanding of the rAAV genome size and integrity.

Purity and DNA content of AAV capsids assessed by analytical ultracentrifugation and orthogonal biophysical techniques.

Development and manufacturing adeno-associated virus (AAV)-based vectors for gene therapy requires suitable analytical methods to assess the quality of the formulations during development, as well as the quality of different batches and the consistency of the processes. Here, we compare biophysical methods to characterize purity and DNA content of viral capsids from five different serotypes (AAV2, AAV5, AAV6, AAV8, and AAV9). For this purpose, we apply multiwavelength sedimentation velocity analytical ultracentrifugation (SV-AUC) to obtain the species' contents and to derive the wavelength-specific correction factors for the respective insert-size. In an orthogonal manner we perform anion exchange chromatography (AEX) and UV-spectroscopy and the three methods yield comparable results on empty/filled capsid contents with these correction factors. Whereas AEX and UV-spectroscopy can quantify empty and filled AAVs, only SV-AUC could identify the low amounts of partially filled capsids present in the samples used in this study. Finally, we employ negative-staining transmission electron microscopy and mass photometry to support the empty/filled ratios with methods that classify individual capsids. The obtained ratios are consistent throughout the orthogonal approaches as long as no other impurities and aggregates are present. Our results show that the combination of selected orthogonal methods can deliver consistent empty/filled contents on non-standard genome sizes, as well as information on other relevant critical quality attributes, such as AAV capsid concentration, genome concentration, insert size length and sample purity to characterize and compare AAV preparations.