Evaluation of recombinant baculovirus clearance during rAAV production in Sf9 cells using a newly developed fluorescent-TCID50 assay.

Introduction: Recombinant adeno-associated virus (rAAV) vectors provide a safe and efficient means for in vivo gene delivery, although its large-scale production remains challenging. Featuring high manufacturing speed, flexible product design, and inherent safety and scalability, the baculovirus/Sf9 cell system offers a practical solution to the production of rAAV vectors in large quantities and high purity. Nonetheless, removal and inactivation of recombinant baculoviruses during downstream purification of rAAV vectors remain critical prior to clinical application. Methods: The present study utilized a newly developed fluorescent-TCID50 (F-TCID50) assay to determine the infectious titer of recombinant baculovirus (rBV) stock after baculovirus removal and inactivation, and to evaluate the impact of various reagents and solutions on rBV infectivity. Results and discussion: The results showed that a combination of sodium lauryl sulfate (SLS) and Triton X-100 lysis, AAVx affinity chromatography, low pH hold (pH3.0), CsCl ultracentrifugation, and NFR filtration led to effective removal and/or inactivation of recombinant baculoviruses, and achieved a log reduction value (LRV) of more than 18.9 for the entire AAV purification process. In summary, this study establishes a standard protocol for downstream baculovirus removal and inactivation and a reliable F-TCID50 assay to detect rBV infectivity, which can be widely applied in AAV manufacturing using the baculovirus system.

Systematic comparison of rAAV vectors manufactured using large-scale suspension cultures of Sf9 and HEK293 cells.

Recombinant adeno-associated virus (rAAV) vectors could be manufactured by plasmid transfection into human embryonic kidney 293 (HEK293) cells or baculovirus infection of Spodoptera frugiperda (Sf9) insect cells. However, systematic comparisons between these systems using large-scale, high-quality AAV vectors are lacking. rAAV from Sf9 cells (Sf9-rAAV) at 2-50 L and HEK293 cells (HEK-rAAV) at 2-200 L scales were characterized. HEK-rAAV had ∼40-fold lower yields but ∼10-fold more host cell DNA measured by droplet digital PCR and next-generation sequencing, respectively. The electron microscope observed a lower full/empty capsid ratio in HEK-rAAV (70.8%) than Sf9-rAAV (93.2%), while dynamic light scattering and high-performance liquid chromatography analysis showed that HEK-rAAV had more aggregation. Liquid chromatography tandem mass spectrometry identified different post-translational modification profiles between Sf9-rAAV and HEK-rAAV. Furthermore, Sf9-rAAV had a higher tissue culture infectious dose/viral genome than HEK-rAAV, indicating better infectivity. Additionally, Sf9-rAAV achieved higher in vitro transgene expression, as measured by ELISA. Finally, after intravitreal dosing into a mouse laser choroidal neovascularization model, Sf9-rAAV and HEK-rAAV achieved similar efficacy. Overall, this study detected notable differences in the physiochemical characteristics of HEK-rAAV and Sf9-rAAV. However, the in vitro and in vivo biological functions of the rAAV from these systems were highly comparable. Sf9-rAAV may be preferred over HEK293-rAAV for advantages in yields, full/empty ratio, scalability, and cost.