Optimizing Expression Systems for AAV Vectors

Viral vectors continue to grow as a major category for biomanufacturing processes based on the growing clinical programs for gene therapies. Optimizing the manufacturing process for viral vectors starts with the expression systems used. A look at the expression system used for manufacturing adeno-associated virus (AAV) vectors will highlight ways to troubleshoot problems in upstream processing.

The common denominator

The most common cell line used across the biopharma industry to produce AAV vectors is the human embryonic kidney 293 (HEK293) cell line, says Elie Hanania, PhD, vice-president, Process Development, Viral Vector Technologies, Avid Bioservices, and Pouria Motevalian, director of Process and Analytical Development, Viral Vector Services, Thermo Fisher Scientific.

AAV vectors are produced by transfecting three plasmids into HEK293 cells: Rep/Cap plasmid, helper plasmid, and gene of interest (GOI) plasmid. “The process is mediated by chemical reagents (transfection reagents), which can be pure lipid-based, mixed protein/lipid-based, or non-lipid based,” Hanania says.

After HEK293, the next widely used cell line to produce AAV vectors is Spodoptera frugiperda (SF) 9, says Motevalian, who also notes that there is additional cell line development work currently underway in the industry, where the focus is on novel producer cell lines.

Systems and alternatives

According to Hanania, AAV production via triple transfection of HEK293 cells can take place in either adherent or suspension cultures. “The former approach may have some scale limitations as well as having animal-derived components in the culture media (although there have been advancements with commercially available serum-free and/or chemically defined culture media). The majority of AAV manufacturing focuses on expansion of HEK293 cells in suspension (using serum-free and/or chemically defined culture media) with options to scale out and scale up the process depending on the desired amount of viral vector,” he explains.

High viral vector titer and overall yield are important aspects of any viral vector production, and to support these requirements, the production process needs to be optimized. Hanania points to cell line development as one of the key process components that has a significant impact on titer and yield, stating that clone selection and testing may be a simple approach to boost productivity, while more complex approaches, such as focusing on gene editing, have also been used to further increase AAV titer and yield.

“It is important to mention that, although HEK293 cells coupled with triple plasmid transfection is the go-to approach for AAV production for the majority of programs, there are other manufacturing upstream expression systems,” Hanania adds. Alternative upstream expression systems include, but are not limited to, co-infection of Sf9 insect cells with baculoviruses that harbor the Rep/Cap and GOI sequences, and infection of HEK293 (or HeLa) cells harboring the Rep/Cap genes and GOI with replicative adenovirus providing needed helper functions. “Both these approaches can support large-scale production of AAV with favorable cost of goods (GOG) reduction compared to the traditional transfection-based process,” he observes.

The use of yeast as a potential production system (with favorable COGs) is also being investigated at the R&D stage, Hanania says.

Meanwhile, other alternatives are also being explored. Motevalian notes that cell-free or synthetic biology options are being considered and implemented in R&D settings. Widespread commercial adoption, however, remains a work in progress.

Hanania says that, aside from focusing on generating AAV for cell and gene therapy applications, there is a need to consider the overall cost of producing these viral vectors. Traditional approaches, especially at commercial scale, are extremely expensive, and this results in costly therapies that have limited accessibility.

Significant strides have been made in the field of synthetic biology that have direct impact on the biomanufacturing of large molecules for advanced therapies, Hanania notes. “This concept is still in its emerging phase when it comes to AAV manufacturing,” he says. “Aside from the use of baculoviruses and herpes simplex viruses for AAV production, synthetic biology showed great strides when considering plasmid production. The chemical synthesis of doggybone DNA is being evaluated across several therapeutic programs and clinical applications.”

The focus on safety (i.e., decreased risk of adventitious agents and other contaminants) and efficacy (i.e., potency of the AAV as the intended therapeutic molecule) are two factors driving the evaluation of these new onsets in AAV manufacturing, Hanania says.

Process control strategies

Minimizing contamination risk is an important consideration in optimizing upstream processing. Process control strategies are needed to achieve the product’s safety profile. Currently, the industry employs a number of strategies, as applicable, to minimize process-related contamination, Motevalian says. “We have implemented stringent quality control for all raw materials, including cell culture media, plasmids, and reagents, to help ensure the reduction of contaminants,” he says.

Motevalian further explains that use of closed and automated systems for cell culture and vector production reduces the risk of contamination from the external environment, while maintaining cleanroom standards involves controlling air quality, temperature, and humidity. In addition, it is important to implement appropriate control and monitoring on critical upstream production process parameters, such as pH, which is known to influence cell growth and thus byprocess-related contaminants, such as host cell DNA and host cell protein (HCP).

“Aside from being potent, advanced gene therapy medicines need to adhere to the highest safety and quality standards. Accordingly, these need to be void of any adventitious viruses that could be introduced by source materials or during the production and purification processes,” stresses Hanania. This task is even more critical for a contract development manufacturing organization dealing with multiple product manufacturing jobs, where measures are in place to avoid any cross contamination, he adds.

“One of the most recent measures introduced in the AAV manufacturing scheme is nano filtration. Being 20-25 nanometer in size, AAV will pass through nanofilters (with a molecular cutoff of 50 nanometers) while larger viruses (e.g., adenovirus, herpes simplex virus, baculovirus, or others), if present, will be retained by such a filter,” Hanania explains.

Hanania clarifies that, although 0.2-micron sterilizing-grade filters are used at the end of the drug substance and drug product steps, these filters are intended to ensure a sterile end product “void of bacterial and microbialcontamination, but not necessarily of other viruses,” which is why a virus-reducing filter is required during the downstream purification of AAV vectors.

Other steps used to ensure the inactivation of other viruses introduced during the AAV manufacturing process focus on low pH, lysis reagents, and/or heat, Hanania further explains. “Such measures will inactivate enveloped viruses and are typically performed in addition to multistep chromatography designed to separate the target AAV from other contaminants. Of course, employing single-use materials and supplies (pre-sterilized) is becoming the norm during bioprocessing of viral vectors.”

Hanania also emphasizes that host cell DNA and HCPs, as well as other process-related contaminants, may become partially packaged in the AAV particle and cause safety issues, including immunogenic response or the introduction of potential oncogenic sequences with long-term negative impacts. “Even though initial levels of such pollutants may be low, the final product, which typically undergoes hundreds of fold concentration, will harbor concentrated levels of such contaminants with a heightened safety risk,” he states. Therefore, robust processing needs to eliminate, or at best reduce, such contaminants. Endonucleases are commonly used to reduce host cell DNA, while clarification filters, proteases, and multistep chromatography are typically used to reduce other introduced or unwanted contaminants.

Further, extensive testing of all raw and starting materials must be factored into the production process to prevent impurities from being introduced into the manufacturing process at the start. “Such testing needs to span materials (biological and chemical), supplies, and any product contacting surfaces and equipment,” says Hanania.

Meanwhile, to prevent cross contamination across multiple batches, facility design comes into play. According to Hanania, robust cleaning measures—with validation—must be applied to a facility’s design, in addition to physical segregation of the different products (i.e., designated suits). “Using vaporized hydrogen peroxide or ionized hydrogen peroxide as part of the change-over process will further reduce cross contamination risks. It is expected that controlled access is implemented to limit personnel-contributed cross contamination,” he says.

Future of AAV vector manufacturing

Looking to the future, the utilization of synthetic biology is anticipated to become central to AAV manufacturing because large amounts of the viral vector will be required for systemic applications, Hanania observes. “System-level genome engineering will advance the field and ensure higher productivity and yield without the constant need of scaling up the manufacturing process,” he states, adding that artificial intelligence, machine learning, conceptual computer designs, and in-silico models will support the advancement of synthetic biology and its utilization in AAV manufacturing.

Beyond the growing use of synthetic biology, better and sustainable producer cell lines are expected to become available, Hanania further notes. “Although this effort started more than a decade ago, the outcome has not been as successful as producer cell lines for retroviruses (cytotoxicity of the helper and Rep genes),” he states.

Meanwhile, the demand for large quantities of viral vectors will become a bottleneck as there is expected to be more systemic applications for gene therapies as well as developers targeting more common disorders, including cancers and neurodegenerative diseases, Hanania says. Effort will focus on optimizing production and purification processes, engineering better starting (and raw) materials, and improving manufacturing platforms. “Although template processes are highly desirable from a manufacturing standpoint, it is clear that the outcome is not optimal across the different AAV serotypes and among the various transgenes. Continuous processing with intensified culture approached may be a good foundation for scaling up targeted advanced therapeutics,” Hanania explains.

From a process standpoint, Motevalian expects that continuous manufacturing with intensified processing enabled through adoption and implementation of process analytical technologies (such as Raman) for real-time monitoring and ultimately release will become more standard. “Concerning facility design, the cell and gene therapy field is progressing towards digitized modular manufacturing facilities, providing flexibility for different production scales and types,” he states.

Hanania points to safety and quality, which will be at the forefront of approved AAV gene therapy products. As such, improved analytical assays will need to be in place. These assays will need to have better sensitivity and higher specificity to characterize these AAV gene therapy products and identify (as well as quantify) contaminants. “Batch to batch consistency is a must,” he stresses.

While high demand for AAV gene therapies may be met by scaling up, there is a limit in the current equipment and approaches to achieve adequate scale up. “It seems gene and cell therapy trajectory are following the early manufacturing pathway for antibodies and recombinant proteins. The latter focused initially on scaling to larger volumes (multiple 10,000 L bioreactors) before considering cell line development, which resulted in high productivity (> 8 g/L of monoclonal antibodies),” explains Hanania. “This negated the need to seek such giant bioreactors and made the scaling process better and more streamlined. Accordingly, scaling up for cell and gene therapy products needs to incorporate process improvements, better reagents and supplies, state-of-the-art analytics, and new designed platforms. Strides across all these have started and will continue in the next several years,” he concludes.

About the author

Feliza Mirasol is the science editor for BioPharm International.

Article details

BioPharm International®
Vol. 37, No. 7
July/August 2024
Pages: 18–20

Citation

When referring to this article, please cite it as Mirasol, F. Optimizing Expression Systems for AAV Vectors. BioPharm International 2024, 37 (7), 18–20.