High-Titer Hepatitis C Virus Production in a Scalable Single-Use High Cell Density Bioreactor.

Hepatitis C virus (HCV) infections pose a major public health burden due to high chronicity rates and associated morbidity and mortality. A vaccine protecting against chronic infection is not available but would be important for global control of HCV infections. In this study, cell culture-based HCV production was established in a packed-bed bioreactor (CelCradle™) aiming to further the development of an inactivated whole virus vaccine and to facilitate virological and immunological studies requiring large quantities of virus particles. HCV was produced in human hepatoma-derived Huh7.5 cells maintained in serum-free medium on days of virus harvesting. Highest virus yields were obtained when the culture was maintained with two medium exchanges per day. However, increasing the total number of cells in the culture vessel negatively impacted infectivity titers. Peak infectivity titers of up to 7.2 log10 focus forming units (FFU)/mL, accumulated virus yields of up to 5.9 × 1010 FFU, and a cell specific virus yield of up to 41 FFU/cell were obtained from one CelCradle™. CelCradle™-derived and T flask-derived virus had similar characteristics regarding neutralization sensitivity and buoyant density. This packed-bed tide-motion system is available with larger vessels and may thus be a promising platform for large-scale HCV production.

Semi-continuous Propagation of Influenza A Virus and Its Defective Interfering Particles: Analyzing the Dynamic Competition To Select Candidates for Antiviral Therapy.

Defective interfering particles (DIPs) of influenza A virus (IAV) are naturally occurring mutants that have an internal deletion in one of their eight viral RNA (vRNA) segments, rendering them propagation-incompetent. Upon coinfection with infectious standard virus (STV), DIPs interfere with STV replication through competitive inhibition. Thus, DIPs are proposed as potent antivirals for treatment of the influenza disease. To select corresponding candidates, we studied de novo generation of DIPs and propagation competition between different defective interfering (DI) vRNAs in an STV coinfection scenario in cell culture. A small-scale two-stage cultivation system that allows long-term semi-continuous propagation of IAV and its DIPs was used. Strong periodic oscillations in virus titers were observed due to the dynamic interaction of DIPs and STVs. Using next-generation sequencing, we detected a predominant formation and accumulation of DI vRNAs on the polymerase-encoding segments. Short DI vRNAs accumulated to higher fractions than longer ones, indicating a replication advantage, yet an optimum fragment length was observed. Some DI vRNAs showed breaking points in a specific part of their bundling signal (belonging to the packaging signal), suggesting its dispensability for DI vRNA propagation. Over a total cultivation time of 21 days, several individual DI vRNAs accumulated to high fractions, while others decreased. Using reverse genetics for IAV, purely clonal DIPs derived from highly replicating DI vRNAs were generated. We confirm that these DIPs exhibit a superior in vitro interfering efficacy compared to DIPs derived from lowly accumulated DI vRNAs and suggest promising candidates for efficacious antiviral treatment. IMPORTANCE Defective interfering particles (DIPs) emerge naturally during viral infection and typically show an internal deletion in the viral genome. Thus, DIPs are propagation-incompetent. Previous research suggests DIPs as potent antiviral compounds for many different virus families due to their ability to interfere with virus replication by competitive inhibition. For instance, the administration of influenza A virus (IAV) DIPs resulted in a rescue of mice from an otherwise lethal IAV dose. Moreover, no apparent toxic effects were observed when only DIPs were administered to mice and ferrets. IAV DIPs show antiviral activity against many different IAV strains, including pandemic and highly pathogenic avian strains, and even against nonhomologous viruses, such as SARS-CoV-2, by stimulation of innate immunity. Here, we used a cultivation/infection system, which exerted selection pressure toward accumulation of highly competitive IAV DIPs. These DIPs showed a superior interfering efficacy in vitro, and we suggest them for effective antiviral therapy.

Towards integrated production of an influenza A vaccine candidate with MDCK suspension cells.

Seasonal influenza epidemics occur both in northern and southern hemispheres every year. Despite the differences in influenza virus surface antigens and virulence of seasonal subtypes, manufacturers are well-adapted to respond to this periodical vaccine demand. Due to decades of influenza virus research, the development of new influenza vaccines is relatively straight forward. In similarity with the ongoing coronavirus disease 2019 pandemic, vaccine manufacturing is a major bottleneck for a rapid supply of the billions of doses required worldwide. In particular, egg-based vaccine production would be difficult to schedule and shortages of other egg-based vaccines with high demands also have to be anticipated. Cell culture-based production systems enable the manufacturing of large amounts of vaccines within a short time frame and expand significantly our options to respond to pandemics and emerging viral diseases. In this study, we present an integrated process for the production of inactivated influenza A virus vaccines based on a Madin-Darby Canine Kidney (MDCK) suspension cell line cultivated in a chemically defined medium. Very high titers of 3.6 log10 (HAU/100 µl) were achieved using fast-growing MDCK cells at concentrations up to 9.5 × 106 cells/ml infected with influenza A/PR/8/34 H1N1 virus in 1 L stirred tank bioreactors. A combination of membrane-based steric-exclusion chromatography followed by pseudo-affinity chromatography with a sulfated cellulose membrane adsorber enabled full recovery for the virus capture step and up to 80% recovery for the virus polishing step. Purified virus particles showed a homogenous size distribution with a mean diameter of 80 nm. Based on a monovalent dose of 15 µg hemagglutinin (single-radial immunodiffusion assay), the level of total protein and host cell DNA was 58 µg and 10 ng, respectively. Furthermore, all process steps can be fully scaled up to industrial quantities for commercial manufacturing of either seasonal or pandemic influenza virus vaccines. Fast production of up to 300 vaccine doses per liter within 4-5 days makes this process competitive not only to other cell-based processes but to egg-based processes as well.

SARS-CoV-2 Production in a Scalable High Cell Density Bioreactor.

The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pandemic has demonstrated the value of pursuing different vaccine strategies. Vaccines based on whole viruses, a widely used vaccine technology, depend on efficient virus production. This study aimed to establish SARS-CoV-2 production in the scalable packed-bed CelCradleTM 500-AP bioreactor. CelCradleTM 500-AP bottles with 0.5 L working volume and 5.5 g BioNOC™ II carriers were seeded with 1.5 × 108 Vero (WHO) cells, approved for vaccine production, in animal component-free medium and infected at a multiplicity of infection of 0.006 at a total cell number of 2.2-2.5 × 109 cells/bottle seven days post cell seeding. Among several tested conditions, two harvests per day and a virus production temperature of 33 °C resulted in the highest virus yield with a peak SARS-CoV-2 infectivity titer of 7.3 log10 50% tissue culture infectious dose (TCID50)/mL at 72 h post-infection. Six harvests had titers of ≥6.5 log10 TCID50/mL, and a total of 10.5 log10 TCID50 were produced in ~5 L. While trypsin was reported to enhance virus spread in cell culture, addition of 0.5% recombinant trypsin after infection did not improve virus yields. Overall, we demonstrated successful animal component-free production of SARS-CoV-2 in well-characterized Vero (WHO) cells in a scalable packed-bed bioreactor.

High cell density perfusion process for high yield of influenza A virus production using MDCK suspension cells.

Similar to the recent COVID-19 pandemic, influenza A virus poses a constant threat to the global community. For the treatment of flu disease, both antivirals and vaccines are available with vaccines the most effective and safest approach. In order to overcome limitations in egg-based vaccine manufacturing, cell culture-based processes have been established. While this production method avoids egg-associated risks in face of pandemics, process intensification using animal suspension cells in high cell density perfusion cultures should allow to further increase manufacturing capacities worldwide. In this work, we demonstrate the development of a perfusion process using Madin-Darby canine kidney (MDCK) suspension cells for influenza A (H1N1) virus production from scale-down shake flask cultivations to laboratory scale stirred tank bioreactors. Shake flask cultivations using semi-perfusion mode enabled high-yield virus harvests (4.25 log10(HAU/100 µL)) from MDCK cells grown up to 41 × 106 cells/mL. Scale-up to bioreactors with an alternating tangential flow (ATF) perfusion system required optimization of pH control and implementation of a temperature shift during the infection phase. Use of a capacitance probe for on-line perfusion control allowed to minimize medium consumption. This contributed to a better process control and a more economical performance while maintaining a maximum virus titer of 4.37 log10(HAU/100 µL) and an infectious virus titer of 1.83 × 1010 virions/mL. Overall, this study clearly demonstrates recent advances in cell culture-based perfusion processes for next-generation high-yield influenza vaccine manufacturing for pandemic preparedness. KEY POINTS: • First MDCK suspension cell-based perfusion process for IAV produciton was established. • "Cell density effect" was overcome and process was intensified by reduction of medium use and automated process control. • The process achieved cell density over 40 × 106 cells/mL and virus yield over 4.37 log10(HAU/100 µL).