As the field of gene therapy continues to evolve, there is a growing need for innovative methods that can address these challenges. Here, a unique method is presented, which streamlines the process of generating high-yield and high-purity AAV vectors using a cell factory platform, meeting the quality standards for in vivo studies.
Preclinical gene therapy research, particularly in rodent and large animal models, necessitates the production of AAV vectors with high yield and purity. Traditional approaches in research laboratories often involve extensive use of cell culture dishes to cultivate HEK293T cells, a process that can be both laborious and problematic. Here, a unique in-house method is presented, which simplifies this process with a specific cell factory (or cell stacks, CF10) platform. An integration of polyethylene glycol/aqueous two-phase partitioning with iodixanol gradient ultracentrifugation improves both the yield and purity of the generated AAV vectors. The purity of the AAV vectors is verified through SDS-PAGE and silver staining, while the ratio of full to empty particles is determined using transmission electron microscopy (TEM). This approach offers an efficient cell factory platform for the production of AAV vectors at high yields, coupled with an improved purification method to meet the quality demands for in vivo studies.
Adeno-associated virus (AAV) vectors have become an indispensable tool in gene therapy research, offering a unique combination of efficacy and safety for gene delivery1. Traditional methods for generating AAVs in laboratory settings have been pivotal in advancing our understanding and application of gene therapy2. However, these methods, while foundational, exhibit certain limitations and challenges, especially in terms of yield, time efficiency, and the quality of the vectors produced, notably the ratio of full to empty particles3.
The conventional procedure for AAV production primarily involves the transfection of HEK293 cells4. This process, typically conducted in cell culture dishes, requires the cells to be transfected with a plasmid containing the gene of interest along with helper plasmid and AAV capsid plasmid5,6. Following transfection, the cells produce AAV particles, which are then harvested and purified5,6. The purification process often involves ultracentrifugation, a critical step in obtaining high-purity AAV vectors7. Ultracentrifugation, particularly using cesium chloride (CsCl) or iodixanol gradient, is a standard method for separating AAV particles from cellular debris and other impurities8. This step is crucial for achieving the desired purity and concentration of AAV vectors, which directly impacts their efficacy in gene delivery8. Despite its widespread use, traditional ultracentrifugation has its drawbacks. For example, the yield of AAV vectors from this method can be variable and often low, which poses significant challenges when large quantities of high-titer vectors are needed, particularly for in vivo studies or large animal models9.
Another critical aspect of AAV vector quality is the ratio of full to empty particles10. AAV preparations often contain a mixture of these particles; however, only the full particles contain the therapeutic genetic material. The presence of a high proportion of empty particles can significantly reduce the efficiency of gene delivery10. Assessing and optimizing the ratio of full to empty particles is thus a key parameter in evaluating the efficacy of AAV vectors. Traditional methods, while capable of producing AAV vectors, often struggle to control this ratio consistently, leading to variations in vector potency10.
Here, a unique method is presented, which streamlines the process of generating high-yield and high-purity AAV vectors using a cell factory platform free from using labor-intensive HEK293T cell cultures in cell dishes, integrating polyethylene glycol/aqueous two-phase partitioning with iodixanol gradient ultracentrifugation. The AAV vector purity is confirmed via SDS-PAGE and silver staining, and the full-to-empty particle ratio is determined using transmission electron microscopy (TEM), meeting the quality standards for in vivo studies11.
The details of the reagents, plasmids, and equipment used in the study are listed in the Table of Materials. The composition of the buffers used is provided in Supplementary File 1.
1. Plasmid preparation
2. Preparing HEK293T cells
3. Triple-transfection of AAV plasmids
4. Harvesting AAV vectors
5. AAV extraction
6. AAV purification by iodixanol gradient ultracentrifugation
7. Second round iodixanol gradients ultracentrifugation
NOTE: This step is optional. This step is to reduce the empty AAV ratio for a higher quality full AAV capsid.
8. AAV dialysis and concentration
9. AAV virus titration
NOTE: TaqMan quantitative polymerase chain reaction (qPCR) was used to titrate purified AAV.
10. Quality control of AAV
NOTE: AAV viruses were characterized for purity by SDS-PAGE silver stain using an SDS-PAGE gel and stained using a commercially available staining kit.
In this detailed step-by-step protocol, a standardized platform is demonstrated to make high-titer and high-quality AAV virus with the CF10 in a large-scale research lab setting. Compared with conventional cell culture dishes, the CF10 provides a convenient way to culture large amounts of cells and produce AAV virus (Figure 1). Several culture conditions were tested to determine whether cells in an optimal environment can promote viral production. A low glucose DMEM supplemented with 10 mM HEPES and 2% FBS showed the best AAV production.
Several protocols were tested to purify AAVs. Most procedures have low virus yields and impurity of AAV capsids. Here, a revised purification protocol was developed, combining the AAV from both cell pellets and culture media (Figure 2). We have found that 80% of AAV was in the cells, and another 20% of AAV was in the culture media, which were secreted from the cells. Both parts of AAV were treated with DNase to remove the free DNA. Sodium deoxycholate was used to further release AAV from the cells. AAVs were then extracted with chloroform extraction followed by an aqueous two-phase partitioning. These steps allowed most protein contaminants to be removed. AAV remains soluble in the ammonium sulfate phase.
The remaining contaminants were removed with a discontinuous iodixanol gradient ultracentrifugation (Figure 3A). The gradient was also helpful in removing the empty AAV capsids, especially with the second round of iodixanol gradient ultracentrifugation.
The purity of the AAV virus was determined by silver staining. When three major bands corresponding to AAV capsid protein, VP1, VP2, and VP3, with a purity greater than 90%, were obtained, the AAV virus was suitable for in vivo use (Figure 3B). The AAV full capsid to empty capsid ratio was accessed by TEM (Figure 3C). Only a full capsid with a transgene insert would allow the expression of transgene in the targeted tissue. A high portion of the empty capsid could also induce an immune response to the AAV capsid. These quality checks are necessary for each AAV virus that is produced and purified before use.
Figure 1: Schematic illustration of the AAV production by HEK293T cells triple-transfection method. Please click here to view a larger version of this figure.
Figure 2: Schematic illustration of the AAV purification. Please click here to view a larger version of this figure.
Figure 3: AAV purification by iodixanol gradient ultracentrifugation and AAV virus purity verification. (A) Iodixanol gradient layers and the position of the needle for harvesting. (B) Representative gel assessing capsid content and purity. M: molecular marker; R: reference AAV6 full capsid; S1: in-house made AAVDJ capsid with one round of ultracentrifugation; S2: in-house made AAVDJ capsid with two rounds of ultracentrifugation; S3: in-house made AAV6 capsid. (C) Electron microscopy image of AAV. Virus collected after purification. Viral capsids containing a viral genome appear as homogeneous white hexagons, while empty capsids appear as hexagons with a white rim but a dark center. Scale bar: 100 nm. Please click here to view a larger version of this figure.
Table 1: PEI calculator for AAV packaging. Please click here to download this Table.
Table 2: Raw AAV-PEG-Sulfate volume chart. Please click here to download this Table.
Table 3: Preparation of iodixanol gradient. Please click here to download this Table.
Table 4: Preparation of second round iodixanol gradient. Please click here to download this Table.
Table 5: AAV titration primers. Please click here to download this Table.
Supplementary File 1: Compositions of the buffers used for the study. Please click here to download this File.
Here, a technically advanced protocol for large-scale production of high-titer and high-quality AAV vectors using a cell factory platform (CF10) is introduced, representing a significant improvement over conventional cell culture dish methods. The use of cell factories simplifies the process of cultivating large volumes of cells, facilitating the production of AAV viruses more efficiently1. Also, by optimizing the culture conditions, particularly with low glucose DMEM supplemented with 10 mM HEPES and 2% FBS, a significantly enhanced viral production was confirmed, indicating the crucial role of the cellular environment in virus yield.
The revised purification protocol, which combines AAV from both cell pellets and culture media, addresses the common issue of low virus yields and impurity seen in many existing protocols. The steps of chloroform extraction and aqueous two-phase partitioning effectively remove most protein contaminants, with AAV remaining soluble in the ammonium sulfate phase. The improvement in both the yield and purity of AAV vectors using the PEG/aqueous two-phase partitioning combined with iodixanol gradient ultracentrifugation, as opposed to traditional gradient ultracentrifugation methods, may be attributed to enhanced initial separation with PEG/aqueous two-phase partitioning, refined purity with iodixanol gradient ultracentrifugation and reduction in contaminant co-purification12. First, the introduction of PEG/aqueous two-phase partitioning before ultracentrifugation significantly improves the initial separation of AAV particles from cellular debris and other contaminants. PEG, a high molecular weight polymer, when mixed with an aqueous solution, creates two distinct phases13. AAV vectors have a propensity to partition preferentially into one of these phases (commonly the PEG-rich phase), while many contaminants and impurities are partitioned into the other13. This selective partitioning effectively concentrates the AAV particles and removes a substantial portion of impurities even before ultracentrifugation, thereby increasing the yield and reducing the contaminant load entering the ultracentrifugation step13. Second, iodixanol gradient ultracentrifugation further refines the purity of AAV vectors. Iodixanol, a non-ionic, iso-osmolar gradient medium, allows for a more gentle and controlled separation compared to traditional CsCl gradients14. In this gradient, AAV particles migrate to a position in the gradient that corresponds to their buoyant density14. Importantly, this process is effective in separating full AAV capsids (containing the genetic payload) from empty capsids (lacking genetic material), which is a crucial determinant of vector quality. Iodixanol’s iso-osmolar nature also preserves the integrity of the AAV capsids better than hyperosmolar agents like CsCl, potentially leading to higher yields of intact, functional vectors14. Finally, traditional ultracentrifugation methods, especially those using CsCl gradients, can sometimes co-purify contaminants that have similar buoyant densities to AAV vectors15. By using PEG partitioning as a preliminary step, the load of such contaminants is greatly reduced before ultracentrifugation13. This reduction in contaminant load means that the iodixanol gradient can work more effectively and selectively in purifying AAV vectors, leading to higher purity15.
The purity and quality of the AAV vectors are rigorously assessed through silver staining and TEM11. The observation of three major bands corresponding to AAV capsid proteins VP1, VP2, and VP3, with a purity exceeding 90%, indicates the suitability of these AAV vectors for in vivo use. The TEM analysis for determining the full-to-empty capsid ratio is particularly crucial, as a high proportion of empty capsids can lead to reduced gene delivery efficiency and potential immune responses11. This quality check, although essential, adds to the procedural complexity and may require additional technical expertise.
In conclusion, the protocol offers significant technical advancements in the production of AAV vectors, particularly in terms of scalability and purity. However, the complexities associated with the purification process and the need for specialized equipment and expertise may still be a minor limitation for its application in certain research settings. Further refinement and simplification of these techniques could make this approach more accessible and widely applicable in the field of gene therapy research.
The authors have nothing to disclose.
TZ designed the experiments. TZ, VD, SB, and JP performed the experiments. TZ and VD generated data and analyzed the data. TZ and YX wrote the manuscript. TZ and GG revised the manuscript. This work was supported by UPMC Children's Hospital of Pittsburgh.
293T/17 cells | ATTC | CRL-11268 | |
(NH4)2SO4 | Millipore Sigma | 1.01217.1000 | |
0.5 M EDTA | MilliporeSigma | 324506-100ml | |
1 mL Henke-Ject syringe | Fisher Scientific | 14-817-211 | |
10% pluronic F68 solution | Fisher Scientific | 24-040-032 | |
10x Tris/Glycine/SDS Buffer | Biorad | 1610732 | |
1M HEPES | Fisher Scientific | 15-630-080 | |
2% Uranyl Acetate Solution | Electron Microscopy Sciences | 22400-2 | |
4%–20% Precast Protein Gels | biorad | 4561094 | |
40% PEG solution | Sigma | P1458-50ML | |
AAV6 reference full capsids | Charles River Laboratories | RS-AAV6-FL | |
Accutase Cell Detachment Solution | Fisher Scientific | A6964-100ML | |
Benzonase | Sigma | E1014-25KU | |
BioLite Cell Culture Treated Dishes 150 mm | Fisher Scientific | 12-556-003 | |
Centrifugal Filter Unit | MilliporeSigma | UFC905024 | |
Corning PES Syringe Filters | Fisher Scientific | 09-754-29 | |
Dialysis Cassettes, 10 K MWCO | Fisher Scientific | PI66810 | |
Disposable PES Filter Units 1 L 0.2 µm | Fisher Scientific | FB12566506 | |
Disposable PES Filter Units 1 L 0.45 µm | Fisher Scientific | FB12566507 | |
Disposable PES Filter Units 500 mL 0.2 µm | Fisher Scientific | FB12566504 | |
DMEM high glucose | Fisher Scientific | 10-569-044 | |
DMEM low glucose | Fisher Scientific | 10567022 | |
DNase | NEB | M0303S | |
DPBS 1x | Fisher Scientific | 14-190-250 | |
Fetal Bovin Serum (FBS) | Biowest | S1620 | |
Formvar/Carbon 300 Mesh, Cu | Electron Microscopy Sciences | FCF300-Cu-50 | |
glycerol | Sigma | G5516-1L | |
KCl | Sigma | P9541-500G | |
LB agar | Sigma | L2897-250G | |
LB broth | Fisher Scientific | BP9732-500 | |
MgCL2·6H2O | Sigma | M9272-100G | |
NEB stable competent cells | NEB | C3040H | |
Nest Biofactory 10 chamber | MidSci | 771302 | |
NucleoBond Xtra Maxi EF | Macherey-Nagel | 740424 | |
Opti-MEM | Fisher Scientific | 31-985-088 | |
OptiPrep Density Gradient Medium | Millipore Sigma | D1556-250ml | |
pAAV-CMV-GFP | Addgene | 105530 | |
pAAV-DJ | Cell BioLab | VPK-420-DJ | |
pAAV-RC6 | Cell BioLab | VPK-426 | |
pAdDeltaF6 | Addgene | 112867 | |
PEG 8000 | Promega | V3011 | |
PEI Max | Polysciences, Inc | 49553-93-7 | |
Pen-Strep | Fisher Scientific | 15-140-163 | |
Phenol red | Millipore Sigma | 1.07242.0100 | |
Pierce Silver Stain Kit | Thermo Fisher Scientific | 24612 | |
QuickSeal tube | Fisher Scientific | NC9144589 | |
Sodium Chloride | Sigma | 1162245000 | |
sodium deoxycholate | Millipore Sigma | D6750-100G | |
Taqman Fast Advanced Master Mix | Thermo Fisher Scientific | 4444557 | |
Type 70 Ti Fixed-Angle Titanium Rotor | Beckman Coulter | 337922 | |
Western Blotting Substrate | ThermoFisher | 32209 |