Optimizing the harvest of viral particles is important for the yield and purity of the final product.
Despite unique manufacturing challenges, gene therapy holds great promise to prevent, treat, or cure certain inherited disorders and diseases. As such, the number of therapeutics in development continues to grow; FDA and the European Medicines Agency predict they will each approve 10 to 20 gene and cell therapies per year by 2025 (1).
Viruses such as adeno-associated virus (AAV) and lentivirus serve as the primary delivery vehicle for treatment, either by replacing, truncating, or turning off defective genes or by introducing new genes to treat or cure a disease (2). Production of recombinant AAV vectors is a process that has made many strides but remains complex and expensive at every stage.
Further attempts to improve the process are hindered because running large-scale reactors for process development is cost and time prohibitive. Therefore, bench-scale analysis may have process developers inadvertently limit the selection of materials based on performance at pilot scale before relevant large-scale conditions are tested (3).
Chemical lysis is one such process step where scale is a critical parameter, presenting unique challenges that current cell lysis reagents fail to address. Yet finding a way to optimize the harvest of viral particles is crucial as it directly impacts the yield and purity of the final product (1).
The AAV harvest step, including cell lysis, is the transition point between upstream and downstream processing. After viral particles are produced intracellularly for non-lytic particles, the host cells must be lysed by disrupting the lipid bilayer of each cell to release the vectors and remove host cell DNA.
Historically, the following chemical and mechanical methods have been employed to achieve this, each with its own advantages and drawbacks:
High-pressure homogenization. Cells are subjected to intense shear forces by forcing them through a small orifice to break down cell membranes (4). Though a quick method, those shear forces can cause aggregation and/or precipitation and are challenging to scale.
Freeze-thaw cycles. This method involves repeatedly freezing and thawing the cells, causing ice crystals to form and disrupt the cell membrane. While effective, it is not scalable and can be time-consuming (4).
Sonication. Using ultrasonic waves to break cells open, sonication is another physical method to achieve cell lysis (5). However, it is not scalable and can generate heat, potentially damaging sensitive viral particles.
Mechanical disruption. Techniques such as bead milling physically grind cells to release their contents. This method can be efficient but may also introduce shear stress, leading to the degradation of viral vectors (4).
Detergents. Reagents such as octoxynol-9 have been the standard for chemical lysis. These detergents form micelles that solubilize membrane proteins, releasing intracellular components (5). However, octoxynol-9 is now on the “substance of very high concern” list due to its environmental impact and potential health risks (6). Numerous replacement detergents are being considered, but there may be challenges related to reactor volume consumption and viral particle quality.
Alkali. A lysis buffer often containing sodium hydroxide (NaOH) and sodium dodecyl sulfate (SDS) is added to break down the cell wall, disrupt DNA hydrogen bonding, and trigger denaturation (4). This method can be time-consuming and challenging to automate, and the resultant high pH can cause instability.
Each of these traditional methods presents specific challenges when applied to large-scale production. The primary issues include the need for large volumes of reagents, process-intensive handling, and the risk of damaging viral particles, which can reduce the efficiency of the overall production process.
When scaling a process for commercial production, process development should evaluate several critical attributes when selecting a cell lysis reagent. Determining critical parameters of quality (Figure 1) for chemical lysis agents can serve as a framework to analyze and validate a cell lysis solution that may improve the harvest process and vector quality. Critical attributes to consider are outlined in the following paragraphs.
Efficiency in lysis and recovery
A high-quality cell lysis solution must effectively lyse cells, releasing the viral particles efficiently. This includes performing well across different cell types, serotypes, and cell densities, ensuring maximum yield and productivity. In addition, given the difficulty in maintaining temperature at large-scale production or purposeful changing of temperature as the downstream process begins, the lysis solution should demonstrate robust performance across a range of temperatures. This ensures reliable cell lysis under variable conditions and allows flexibility in the design space.
Minimal impact on vector integrity
A high-quality cell lysis solution must preserve the structural integrity of viral particles during the lysis process. Damage to viral vectors can significantly reduce the efficacy of gene therapy treatments by impacting infectivity, making this feature paramount.
Ease of use
A ready-to-use format reduces preparation time and minimizes the potential for inconsistencies, which is particularly important in large-scale operations. Low viscosity and stability at room temperature enhance handling and storage efficiency.
Compatibility with downstream processing
The solution should be easily cleared during downstream processing to prevent contamination of the final product. Compatibility with endonucleases and other processing steps is essential to maintain product purity and quality.
Regulatory compliance
Compliance with environmental, health, and safety (EHS) regulations, as well as REACH regulations, is essential. The reagent should not include harmful substances that could pose risks during manufacturing or potentially impact regulatory approval. Another more recent consideration is whether the process can inactivate adventitious virus which is predominantly accomplished using detergents and alkali solutions (7). While octoxynol-9 has historically been used for viral inactivation, there is a question of whether replacements have similar properties.
During process development, a critical experimental design question is the choice of the smallest, large-scale-relevant vessel for testing. Research at smaller scales does not necessarily replicate the complexity of a larger reactor. There are variables and conditions that, while irrelevant at small scale, will have a significant impact at production scale.
Often initial testing occurs in shaker flasks which agitate cells differently from reactors that use impellers. This difference may impact how lysis agents disrupt cell membranes and interact with particles but can be masked by the insensitive nature of downstream processing at small scale. Additionally, the effect of time and temperature on cell lysis may be miscalculated during process development. Manufacturers must consider that with larger volumes, the lysis agent will be in contact with particles for hours after the lysis step and at lower temperatures.
These conditions must be accounted for; however, early testing at full scale is labor and cost prohibitive. To maximize process efficiency at scale, a robust process development strategy will ensure that reagents and materials exceed minimum process requirements at the bench.
The following study results, based on the outlined framework of quality parameters, can help inform large-scale process behavior of detergent-based cell lysis solutions.
Cell viability and lysis efficiency
A cell lysis solution should be capable of lysing cells to homogeneity compared to other commonly used detergents at typical usage concentrations and within a reasonable duration of time. Lysis kinetics provide evidence of robustness and process flexibility, as well as minimal damage to the viral particles.
Flexibility no matter the density. New processes may utilize denser cell concentrations than those in use today. An ideal cell lysis solution maintains its performance across a wide range of cell densities, ensuring consistent lysis and high yield regardless of scale.
This experiment specifically tested human embryonic kidney (HEK) cell densities of 5, 10, 20 and 40e6 cells/mL following one hour of cell lysis (Figure 2). Maintaining equivalent performance over a range of densities offers a large operating window for greater flexibility to adjust to future harvest requirements.
Temperature stability. At production scale, temperature shifts are common before the detergent is removed. This is another key example where scale consideration is crucial, as the detergent is often only analyzed at elevated temperatures during process development.
This study (Figure 3) demonstrates the performance variability of lysis reagents at different temperatures, a critical factor for large-scale processing where temperature variations are common.
Viral particle protection and yield
Once the viral particles are released from the cell during the lysis process, the detergent lysis step must have minimal effect on the integrity or yield of the released viral particles—particularly from shear stress due to agitation.
Higher recovery rates of intact viral particles demonstrate that damage from shear stress can be minimized, ensuring higher quality and yield of the final product (Figure 4).
Viral particle yield and the impact of serotype. The impact of cell lysis agents may be more significant depending on the serotype of the viral vectors being produced (Figure 5). Performance across AAV serotypes ensures that a cell lysis solution can be effectively used for different types of gene therapy vectors, enhancing its versatility and reliability in production settings.
Viral particle infectivity and adventitious viral inactivation. Viral particles that are released during the cell lysis process often consist of product-related impurities that inhibit the production of quality, potent vectors. A full capsid with intact viral genomic DNA containing the transgene of interest is the desired outcome; however, it is possible to produce empty capsids, partial capsids that may contain an incomplete viral genome, or a capsid that contains contaminant DNA.
Often considered as a final experimental check, infectivity can be affected by cell lysis agents resulting in a lower potency of therapeutic product. Detergent concentration and harshness of the agent can affect infectivity along with treatment time and salt concentration. As such, infectivity assays should be conducted earlier in the process when considering the selection of a cell lysis agent either through standard TCID50 or potency assays.
A secondary property of octoxynol-9 is its ability to inactivate adventitious virus, which is why it is commonly used in downstream processes for monoclonal antibody manufacturing. Various cell lysis agents may have differing capabilities to inactivate viruses (8). For example, polysorbate 20 at use concentrations is known to be ineffective at adventitious viral clearance (9). Therefore, testing this property through model viral systems should be considered to simplify viral clearance processes.
Ease of use and operational efficiency
Classic lysis reagents such as polysorbate 20 and octoxynol-9 are highly viscous. This characteristic presents challenges for their preparation and requires more significant dilution—taking up additional reactor space. There are a number of new lysis reagents entering the market that are low viscosity and are in a ready-to-use format which can streamline the harvest process. This “ease-of-use” property becomes more valuable at larger scale and therefore should be defined as an operational metric when selecting a lysis reagent even when testing is occurring at the benchtop scale.
Selecting the right cell lysis solution for large-scale viral vector production involves careful consideration of several critical features. Manufacturers should prioritize solutions that preserve vector integrity, comply with regulatory standards, are scalable and easy to use, demonstrate temperature stability, and are compatible with downstream processing. By focusing on these features, manufacturers can ensure efficient and consistent production of high-quality viral vectors, ultimately advancing the field of gene therapy. n
1. Scudellari, M. How Gene Therapy Overcame High-profile Failures. ScienceNews.org, March 22, 2022.
2. White, M.; Alsarraj, M. Quantifying AAV Viral Titer and Integrity with ddPCR. AmericanPharmaceuticalReview.com, Aug. 26, 2021.
3. Jiang, Z.; Dalby, P. A. Challenges in Scaling Up AAV-based Gene Therapy Manufacturing. Trends Biotechnol. 2023, 41 (10), 1268–1281.
4. Shehadul Islam, M.; Aryasomayajula, A.; Selvaganapathy, P. R. A Review on Macroscale and Microscale Cell Lysis Methods. Micromachines (Basel) 2017, 8 (3), 83.
5. Brown, R. B.; Audet, J. Current Techniques for Single-cell Lysis. J. R. Soc. Interface 2008, 5 Suppl 2 (Suppl 2), S131–S138. DOI: 10.1098/rsif.2008.0009.focus
6. ECHA, European Chemicals Agency, Proposal for Identification of a Substance as a CMR Cat 1A or 1B, PBT, Vpvb or a Substance of an Equivalent Level of Concern. 4-(1,1,3,3- Tetramethylbutyl)Phenol, Ethoxylated [Covering Well-Defined Substances and UVCB Substances, Polymers and Homologues]. (Annex XV Dossier, September 2012).
7. Barone, P. W.; Wiebe, M. E.; Leung, J. C.; et al. Viral Contamination in Biologic Manufacture and Implications for Emerging Therapies. Nat. Biotechnol. 2020, 38 (5), 563–572.
8. Farcet, J-B.; Kindermann, J.; Karbiener, M.; Kreil, T. R. Development of a Triton X-100 Replacement for Effective Virus Onactivation in Biotechnology Processes. Engineering Reports 2019, 1 (5), e12078.
9. Neveu, E.; Parcelier, A.; Avenier, N.; et al. Case Study: Evaluation of Several Process Cell Lysis Reagents as Replacements for Triton X-100 for rAAV Production. Yposkesi.com, Nov. 14, 2023.
Greg Swan is business development manager, cell & gene therapies, and Beth Kroeger-Fahnestock is director of new product introductions, biopharma; both at Avantor.