Improved Quantification of AAV Viral Genome Integrity using Nanoneedle Technology for Enhanced Molecular Specificity and Accuracy
Introduction
Recombinant adeno-associated virus (rAAV) vectors are commonly utilized in gene therapy applications; however, these vectors exhibit significant heterogeneity throughout the manufacturing process. Accurate quantification of capsid and genome titers, particularly the integrity of the viral genome, is crucial for ensuring the efficacy and safety of gene therapies. Current methodologies for characterizing AAV vectors include analytical ultracentrifugation (AUC), electrophoresis, and molecular assays, such as PCR and ELISA. However, these methods often fall short in distinguishing between full-length viral genomes and truncated or partial genome fragments, which impacts the interpretation of AAV vector quality.
Methods
We developed a novel viral genome assay utilizing a nanoneedle technology on the Tessie platform. Tessie enables accurate and highly reproducible quantification of full-length transgenes and offers molecular specificity to distinguish those from long partial genomes of similar sizes. In our study, Tessie’s performance was compared to established methods, including ELISA for capsid titer, droplet digital PCR for genome titer, and AUC and electrophoresis for assessing viral genome integrity. A reference AAV sample, potentially containing a high level of full-length genomes along with a certain level of contamination from long partial genomes, was analyzed using all platforms. This comparison aimed to evaluate Tessie’s ability and accuracy in identifying and quantifying distinct viral species within heterogeneous AAV preparations.
Results
The AUC characterization displayed a single peak (Fig. 1a), suggesting that the predominant species accounts for 70% of the total population. In contrast, electrophoresis revealed multiple side bands, highlighting the heterogeneous nature of the sample. When quantifying the area under the curve within the 3-5 kb size range, electrophoresis indicated that less than 11% of the total population corresponds to the full-length viral genome size (Fig. 1b).
Top row: The capsid assay demonstrates a dynamic range of 2E8 to 1E11 cp/ml, with an intra-assay coefficient of variation (CV) of 3.3%. The inter-assay CV, tested across 13 plates, 2 users, and 3 instruments, is 7%. Bottom row: The viral genome assay offers a tunable dynamic range, depending on the number of primer extension cycles after probes are mixed with the sample. The dynamic range spans from 9.8E5 to 1E11 vg/ml, with an intra-assay CV of 2.5%. The inter-assay CV, tested across 17 plates, 3 users, and 3 instruments, is 14%. Recovery in spike-in experiments returned expected quantification within 80-120%.
We hypothesize that the discrepancy between AUC and electrophoresis results arises from the presence of a significant population of long partial genomes that are similar in size to the full-length viral genome. On the Tessie platform, we developed two assays: the transgene assay, which specifically quantifies the titer of the full-length viral genome (4.3 kb), and the long partial transgene assay, which quantifies left-truncated partial transgenes (3.9 kb). Additionally, we quantified the capsid titer on the same platform, enabling us to calculate the empty/full/partial ratio. When accounting for the long partial genome populations, the empty/full ratio aligns well with the AUC characterization. Furthermore, the pure full-length population corresponds more closely with the electrophoresis results.
Our results demonstrate that traditional methods, such as AUC, can misclassify long partial genome subpopulations as full-length viral particles due to their limited ability to distinguish between full-length genomes and truncated transgene variants. In contrast, Tessie’s quantification method exhibited a high degree of correlation with molecular methods, including ELISA and ddPCR, and effectively reconciled the discrepancies observed between different biophysical methods, such as AUC and electrophoresis.
Conclusions
The Tessie platform offers a new methodology for both process development and product qualification in AAV-based gene therapies. Its ability to provide molecular specific quantification of viral genomes, including the identification and quantification of partial genomes, represents a substantial benefit over existing methodologies and has the potential to significantly improve the development and manufacturing of AAV-based gene therapies.
Improved Quantification of AAV Viral Genome Integrity using Nanoneedle Technology for Enhanced Molecular Specificity and Accuracy
Introduction
Recombinant adeno-associated virus (rAAV) vectors are commonly utilized in gene therapy applications; however, these vectors exhibit significant heterogeneity throughout the manufacturing process. Accurate quantification of capsid and genome titers, particularly the integrity of the viral genome, is crucial for ensuring the efficacy and safety of gene therapies. Current methodologies for characterizing AAV vectors include analytical ultracentrifugation (AUC), electrophoresis, and molecular assays, such as PCR and ELISA. However, these methods often fall short in distinguishing between full-length viral genomes and truncated or partial genome fragments, which impacts the interpretation of AAV vector quality.
Methods
We developed a novel viral genome assay utilizing a nanoneedle technology on the Tessie platform. Tessie enables accurate and highly reproducible quantification of full-length transgenes and offers molecular specificity to distinguish those from long partial genomes of similar sizes. In our study, Tessie’s performance was compared to established methods, including ELISA for capsid titer, droplet digital PCR for genome titer, and AUC and electrophoresis for assessing viral genome integrity. A reference AAV sample, potentially containing a high level of full-length genomes along with a certain level of contamination from long partial genomes, was analyzed using all platforms. This comparison aimed to evaluate Tessie’s ability and accuracy in identifying and quantifying distinct viral species within heterogeneous AAV preparations.
Results
The AUC characterization displayed a single peak (Fig. 1a), suggesting that the predominant species accounts for 70% of the total population. In contrast, electrophoresis revealed multiple side bands, highlighting the heterogeneous nature of the sample. When quantifying the area under the curve within the 3-5 kb size range, electrophoresis indicated that less than 11% of the total population corresponds to the full-length viral genome size (Fig. 1b).
Top row: The capsid assay demonstrates a dynamic range of 2E8 to 1E11 cp/ml, with an intra-assay coefficient of variation (CV) of 3.3%. The inter-assay CV, tested across 13 plates, 2 users, and 3 instruments, is 7%. Bottom row: The viral genome assay offers a tunable dynamic range, depending on the number of primer extension cycles after probes are mixed with the sample. The dynamic range spans from 9.8E5 to 1E11 vg/ml, with an intra-assay CV of 2.5%. The inter-assay CV, tested across 17 plates, 3 users, and 3 instruments, is 14%. Recovery in spike-in experiments returned expected quantification within 80-120%.
We hypothesize that the discrepancy between AUC and electrophoresis results arises from the presence of a significant population of long partial genomes that are similar in size to the full-length viral genome. On the Tessie platform, we developed two assays: the transgene assay, which specifically quantifies the titer of the full-length viral genome (4.3 kb), and the long partial transgene assay, which quantifies left-truncated partial transgenes (3.9 kb). Additionally, we quantified the capsid titer on the same platform, enabling us to calculate the empty/full/partial ratio. When accounting for the long partial genome populations, the empty/full ratio aligns well with the AUC characterization. Furthermore, the pure full-length population corresponds more closely with the electrophoresis results.
Our results demonstrate that traditional methods, such as AUC, can misclassify long partial genome subpopulations as full-length viral particles due to their limited ability to distinguish between full-length genomes and truncated transgene variants. In contrast, Tessie’s quantification method exhibited a high degree of correlation with molecular methods, including ELISA and ddPCR, and effectively reconciled the discrepancies observed between different biophysical methods, such as AUC and electrophoresis.
Conclusions
The Tessie platform offers a new methodology for both process development and product qualification in AAV-based gene therapies. Its ability to provide molecular specific quantification of viral genomes, including the identification and quantification of partial genomes, represents a substantial benefit over existing methodologies and has the potential to significantly improve the development and manufacturing of AAV-based gene therapies.