Media for Suspension-Based Viral-Vector Manufacturing

Image Source: ©Artur Wnorowski/Stock.Adobe.com

Viral-vector production generally proceeds via cell culture in human embryonic kidney (HEK) 293 cells. High-performance media and feeds for use in high-density HEK293 suspension cultures have been developed previously for applications outside of the viral-vector space.

This foundational knowledge, says Jonathan Zmuda, director of Thermo Fisher Scientific’s Viral-Vector Production Systems, has been key to further developing media designed to meet the specific needs associated with viral-vector production.

Viral-vector production is fundamentally different than protein expression, however, and there are different media formulation requirements. Issues that must be addressed include the diversity of viral-vector types and the many different cell lines being developed for suspension-based viral-vector production, which are adapted from adherent HEK293 cell lines. Rapidly evolving transient transfection technology also creates uncertainty for media developers.

Need to support rapid, high-cell-density growth and high titers

As is the case with any cell-culture process, optimal viral-vector performance requires robust HEK293 cell growth and high viral-vector productivity. Media should support rapid (low doubling time), high-density cell growth and high titer production, according to Michael Shen, head of viral-vector process development at MilliporeSigma.

Particularly for suspension-based lentivirus (LV) and adeno-associated virus (AAV) production with HEK293 cell lines, Shen notes that media must be formulated to be compatible with plasmid transfection for transient expression of viral vectors. “Some HEK293 cell lines tend to clump in suspension and require the use of anti-clumping supplements (like dextran sulfate); however, some of these supplements interfere with transfection and must be removed or diluted prior to transfection,” says Kimberly Schrag, viral-vector-therapies and beyond CHO media development with MilliporeSigma.

In addition, observes Nan Lin, director of new technologies and service applications for Cytiva, optimal viral-vector processes ideally support production of various viral-vector types (LV, adenovirus [AV], AAV) and generate products with desirable critical quality attributes (CQAs), such as full:partial:empty capsid ratios and biological activity. She also notes that unlike with protein expression, high cell density may not always translate to high viral titer, and therefore optimized media must achieve a balance between titer and growth, with viral titer weighted more heavily.

Media for AAV production with insect cell lines such as Sf-9 or Sf-RVN, meanwhile, have the same requirements for rapid cell growth and high productivity as insect media used for recombinant protein and virus-like particles production, according to Schrag.

Impacts of adherent adaptation

Suspension-based cell culture for viral-vector production generally takes place using adherent cell lines that have been adapted to perform under suspension conditions. This fact is important, because the composition of suspension culture media and adherent culture media is drastically different in every component category, according to Lin.

Generally, adaptation of adherent HEK293 cells into chemically defined suspension media can occur quite readily, notes Zmuda—as long as care is taken to perform the adaptation at relatively high cell densities to not stress the cells during the process. “Adaptation from adherent to suspension culture changes the nutritional needs of the cells, and this transition is dependent on the medium and process used,” he adds.

Lin outlines the key differences. First, the concentrations of components are typically higher in several important categories for suspension media. Second, chemically defined and protein-free (without recombinant growth factors) expression can be achieved more easily with suspension culture media. Third, adherent media need to incorporate specific components that support cell attachment to growth surfaces such as microcarriers or roller bottles that are not needed in suspension media. Fourth, suspension media development can also address the need for perfusion or intensified fed-batch processes that many companies employ for viral-vector production.

Compared to adherent-based cell culture, suspension-based viral-vector production in HEK293 cells takes place with significantly higher cell densities and production volumes, Zmuda comments. In addition, he emphasizes that chemically defined media for suspension cell culture must include the same nutrients or suitable synthetic alternatives found in the animal serum that is present in the leaner adherent media formulations.

Media for suspension cell culture must also be formulated to address the shear stress that can be caused by mixing during these processes, which is not an issue for adherent processes, Schrag observes. “Poloxamer 188 is a common component for reducing the impact of shear stress for protein expression, but it can interfere with some transfection regents and therefore is one component that tends to be present in suspension media formulations for viral-vector production at lower concentrations than in media for mAb [monoclonal antibody] production in stable Chinese hamster ovary (CHO) cell lines,” she says.

Several key ingredients in chemically defined media

To support high cell viability and rapid cell growth prior to transfection, chemically defined media should contain optimized concentrations of various components. Key ingredients according to Schrag include both essential and non-essential amino acids, vitamins, trace elements/minerals, pH-buffering agents, and glucose. MilliporeSigma recommends supplementing with L-glutamine for growth and viral vector production, with optimal concentrations typically around 6-8 mM.

Lin specifically highlights lipids and/or cholesterol as being important for viral-vector production in HEK293 due to their critical roles in cholesterol biosynthesis and cholesterol homeostasis.

For large-volume suspension cell-culture (1000 to 2000-L range), Zmuda recommends the use of non-liquid versions of culture medium such as Thermo Fisher Scientific’s Gibco Advanced Granulation Technology, which allows for the reconstitution of a granulated formulation for large-scale applications.

Need to Enable Transfection

One challenge to media development for viral-vector production is the need to enable transfection while also supporting high-density cell growth, as some components that support the latter interfere with the former, according to Schrag. CHO media formulations in particular are often not compatible with transfection reagents, she adds.

A balance of salts and the removal of transfection-interfering components are therefore important for enabling high transfection efficiencies and promoting high titers post-transfection, Schrag says. In addition to reduced polaxamer 188 content, she notes that most viral-vector media are hydrolysate-free, as hydrosylates impact lot-to-lot variability with respect to cell growth and titer and some also interfere with transfection (HEK293) and infection (Sf-RVN/Sf-9) reactions. In addition, dextran sulfate interferes with infection processes and should be reduced or removed from insect media formulations, according to Schrag.

The length of the cell-culture process may also impact media and feed demands. Some transfection processes are relatively short (two to three days) compared to the time required for mAb and protein production (five to seven days). The latter typically require additional feed supplementation. For shorter viral-vector manufacturing runs, however, media formulations are ideally developed to avoid any need for additional feeding steps, according to Zmuda.

Depletion of media components that support cell culture during virus production does occur for high-density transfections and processes that require up to five days for dilution, transfection, and virus production, according to Schrag. These challenges can be overcome by supplementing with approximately 5–12% of a concentrated CHO feed 24 hours post-transfection (HEK293) and 24 hours post-infection (Sf-RVN/Sf-9). “Supplementation post-transfection/post-infection can significantly increase AAV titers for both HEK293 and insect platforms under these conditions, with commercially available CHO, feeds working equally well as HEK293- and insect-specific feeds,” she contends.

Media optimization is, emphasizes Schrag, only one aspect of the development of high-titer viral-vector production processes. “Transfection optimization can bring the biggest improvements in titer and percent-full capsids, and it is worth the time to perform a small transfection screen while evaluating new media formulations,” she states.

Flexibility essential

Ensuring that media formulations are sufficiently robust to allow for flexibility in the manufacturing process without negatively impacting other aspects of the production system (i.e., transfection efficiency, downstream processing, and regulatory compliance) is another important challenge. Zmuda particularly points to seed-train scale-up, for which different strategies are employed to get the cells to reach the requisite density for transfection.

“Some groups will seed production reactors at a lower cell density and grow the cells for three to four days prior to transfection, while others will want to seed the cells for only one to two days prior to transfection. In addition, some groups will want to add fresh medium before transfection, while others will want to simply grow the cells up to density and then transfect directly,” says Zmuda.

“Given these differences in methodologies and HEK293 lineages and taking into account the relatively short duration of viral vector production runs, it can be argued that media robustness may be as important during the seed train as in the production run itself,” Zmuda concludes.

Pairing media with cell lines and viral-vector types

Improving transfection efficiency, particularly for large-scale reactions, is a major focus of viral-vector manufacturers as they work to meet growing demand. Many have developed platform processes leveraging proprietary clonal cell lines. “Long-term cultivation and subcloning of cells have been observed to result in karyotypic drift and phenotypic changes. In this sense, subcloning HEK293 host cell lines can be an effective approach to identify host cells with higher transfection efficiency, higher viral vector productivity, or more robust cell growth. Indeed, increasing numbers of researchers, Cytiva included, have conducted subcloning studies in HEK293 and identified subclones with such desirable characteristics,” Lin observes.

While base media formulations tend to be a good starting point for divergent HEK293 clones, the optimal pairing of media components with a particular clone is always critical to optimize a production process, states Gino Stolfa, a senior staff scientist in Thermo Fisher Scientific’s Biologicals and Chemicals Division. “HEK293 cell lines used in the field come from various lineages and have diverse requirements for growth and productivity. The diversity of these cells has resulted in a wide variety of media, feeds, and supplements to address differences in lineage,” he adds.

Different media formulations may work better for specific HEK293 cell lines, Schrag agrees, but the development of media using multiple HEK293 cell lines can result in media that is effective regardless of cell origin, as was done with MilliporeSigma’s most recently launched HEK293 medium, which was developed with three diverse HEK293 cell lines, according to Schrag. “Rather than having multiple basal formulations for multiple cell lines/AAV serotypes, it may be more efficient to have a single basal formulation that supports good cell growth and transfection efficiency and then use a feed to drive higher titers,” she says.

It can also be challenging to manage different media formulations and supplements for different viral-vector types, according to Lin. It becomes even more difficult if different media and supplements are required for different AAV serotypes or AAVs with engineered capsids. “Handling many different products from different media companies from CGMP [current good manufacturing practice] manufacturing, regulatory, and procurement aspects can post high complexity,” he says. Lin does note that a “plug-and-play” platform approach, which is common for CHO bioprocessing, can be a good strategy for streamlining viral vector processes. “A common media with various supplements to choose from for each viral-vector type can be a good strategy for companies with a viral vector-filled pipeline,” he concludes.

Managing Uncertainty

Development of media for viral-vector production, meanwhile, requires selection of appropriate analytical targets and techniques, which can be challenging given that industry understanding of viral-vector CQAs is continuously evolving, according to Schrag.

“We are constantly looking to add new analytical assays to improve our product development and protocols for customers,” she says. She also comments that analytical assays for media development need to be adaptable to mid- to high-throughput testing. MilliporeSigma has, for instance, implemented a high-throughput digital droplet polymerase chain reaction assay for titer and a mid-throughput testing solution for determination of percent full capsids.

Recognizing the impact of changes to media formulations is equally important, according to Zmuda. “It cannot be emphasized enough that every change in a viral vector production process can have unintended consequences on the upstream process, the downstream process, and/or CQAs. It is important that any changes are characterized carefully in these key areas to ensure that a change in one part of the process does not cause unexpected negative impacts elsewhere,” he states.

Strategic approaches to media development

The nature of the viral-vector production process dictates to some degree how media formulations are developed. “Understanding cell metabolism in a process is critical to developing fit-for-purpose media and supplements. The short process time and cell diversity associated with viral-vector production require the user to leverage design-of-experiment approaches, high-throughput methods, and -omics to identify which components drive growth and productivity for each lineage or process,” Stolfa contends.

MilliporeSigma has, for instance, a proprietary media development process based on multivariant data analysis of measured cell responses to media component concentration changes. The specific responses used to develop HEK293 media for AAV production were cell growth, transfection efficiency, and AAV CQAs of multiple HEK293 cell lines. “With this approach, we know which components impact each response (e.g., titer, growth, percent full capsids) and are able to optimize the medium to meet project goals,” she says.

Stolfa reiterates that the media development process must include assessment of impacts to product quality; considering titer alone is insufficient. “The highest titer may come at the expense of percent full capsids or reduced infectivity, both of which are counter-productive,” he remarks.

Opportunities for further improvement

There is always room for improvement of any type of cell-culture process, and viral-vector production is no exception. “Viral production mechanisms remain understudied by academic and industry researchers alike. As more knowledge is gained, it will be applied to the development of more effective media and supplements,” states Lin.

For instance, as media formulations change, the cells themselves also adapt, which can, at times, lead to new and even unexpected phenotypes that may have positive attributes that can then be iterated upon further, according to Stolfa. He also observes that given the expectation that stable viral-vector production will eventually complement traditional transient approaches, additional opportunities will arise for the development of new production media and supplements to support longer production runs and/or intensified processes, perfusion, and continual harvest.

About the author

Cynthia A. Challener, PhD, is a contributing editor to BioPharm International.

Article details

BioPharm International
Vol. 36, No. 6
June 2023
Pages: 18-21

Citation

When referring to this article, please cite it as Challener, C.A. Media for Suspension-Based Viral-Vector Manufacturing. BioPharm International 2023 36 (6).