Enhancing the production of adeno-associated virus (AAV)2 and AAV9 with high full capsid ratio in HEK293 cells through design-of-experiment optimization of triple plasmid ratio

Abstract

The recombinant adeno-associated virus (rAAV) vectors used in gene therapy are usually produced by transfecting three different plasmids (Adenoviral helper plasmid (pHelper), AAV rep/cap plasmids (pRepCap), and Transgene plasmid (pAAV-GOI)) into human embryonic kidney 293 (HEK293) cells. However, the high proportion of unwanted empty capsids generated during rAAV production is problematic. To simultaneously enhance the genome titer and full capsid ratio, the ratio of the three plasmids transfected into HEK293 cells was optimized using design-of-experiment (DoE). AAV2 and AAV9, which have different production kinetics, were selected as cell-associated and secreted model AAVs, respectively. In 125 mL Erlenmeyer flasks, the genome titers of rAAV2 and rAAV9 at DoE-optimized plasmid weight ratios (pHelper:pRep2Cap2:pAAV-GOI = 1:3.52:0.50 for rAAV2 and pHelper:pRep2Cap9:pAAV-GOI = 1:1.44:0.27 for rAAV9) were 2.23-fold and 2.26-fold higher than those in the widely used plasmid weight ratio (1:1:1), respectively. In addition, compared with the plasmid ratio of 1:1:1, the relative VP3 band intensities of rAAV2 and rAAV9, which represent the relative empty capsid ratios, were reduced by 26% and 25%, respectively, at the DoE-optimized plasmid ratio. Reduced empty capsid ratios in the DoE-optimized plasmid ratios were also confirmed using transmission electron microscopy (TEM). Taken together, regardless of the AAV serotype, DoE-aided optimization of the triple plasmid ratio was found to be an efficient means of improving the production of rAAV with a high full capsid ratio.

Keywords: adeno-associated virus; design-of-experiment; full capsid ratio; human embryonic kidney cells; triple plasmid ratio; viral genome titer.