Influenza virus intracellular replication dynamics, release kinetics, and particle morphology during propagation in MDCK cells

Abstract

Influenza viruses are respiratory pathogens and can cause severe disease. The best protection against influenza is provided by annual vaccination. These vaccines are produced in embryonated chicken eggs or using continuous animal cell lines. The latter processes are more flexible and scalable to meet the growing global demand. However, virus production in cell cultures is more expensive. Hence, further research is needed to make these processes more cost-effective and robust. We studied influenza virus replication dynamics to identify factors that limit the virus yield in adherent Madin-Darby canine kidney (MDCK) cells. The cell cycle stage of MDCK cells had no impact during early infection. Yet, our results showed that the influenza virus RNA synthesis levels out already 4 h post infection at a time when viral genome segments are exported from the nucleus. Nevertheless, virus release occurred at a constant rate in the following 16 h. Thereafter, the production of infectious viruses dramatically decreased, but cells continued to produce particles contributing to the hemagglutination (HA) titer. The majority of these particles from the late phase of infection were deformed or broken virus particles as well as large membranous structures decorated with viral surface proteins. These changes in particle characteristics and morphology need to be considered for the optimization of influenza virus production and vaccine purification steps. Moreover, our data suggest that in order to achieve higher cell-specific yields, a prolonged phase of viral RNA synthesis and/or a more efficient release of influenza virus particles is required.

Keywords: Influenza virus; Vaccine production; Virus particle morphology; Virus release.

Fig. 1

Virus production kinetics. MDCK cells were infected with the influenza virus PR8 at an MOI of 10 PFU per cell. The supernatant was harvested every 4 h and replaced with fresh media. Infections without medium exchange (incubated for 12, 24, and 36 h) served as controls. The number of total virus particles was calculated based on the HA titer and infectious virions were quantified by the TCID₅₀ assay. Depicted are the cumulative titers of total (a) and infectious virus particles per cell (b) as well as the virus release rates per hour (c). The first and the second dashed lines indicate time points when ≥95 % of the final titer of infectious and total particles were released, respectively. Control titers (upper panel) indicate the number of total virions released from infected cells without medium exchange. d Cell counts and cell viability. The cumulative numbers of detached cells are depicted. Numbers of adherent cells and viabilities of all cells (based on trypan blue staining) are derived from infected cells without medium exchange. Error bars indicate standard deviation of three independent experiments

Fig. 2

Virus particle morphology at early and late stages of virus production. Cells were infected with influenza virus PR8 at an MOI of 10 PFU per cell and supernatants were harvested and replenished with fresh medium every 4 h. Selected samples were subjected to negative stain transmission electron microscopy. a Virus particle morphology. Representative images of released particles at early (8–12 hpi) and late (20–24 hpi) stages of virus production are shown. Particles were classified as indicated by arrows: (I) intact and spherical with well-resolved surface spike proteins; (II) deformed or broken with distinct spike protein-decorated surface; (III) large membranous, of which many appear spike-decorated. Insets show higher magnifications of selected particles. Scale bar indicates 250 nm. b Quantitative results of particle classification. Particles of a total of 100 images per sample (yielding 345 and 314 particles for early and late stages, respectively) were subjected to manual classification analysis. Imaging data of one representative experiment is depicted (two independent experiments were analyzed)

Fig. 3

Dynamics of the intracellular viral RNA synthesis. MDCK cells were infected with the influenza virus PR8 at an MOI of 10 PFU per cell and harvested at indicated time points. Viral mRNA (a) as well as cRNA and vRNA (b) of the influenza virus genome segment 5 were measured by a strand-specific RT-qPCR. The data shown represent the mean of three independent experiments and error bars indicate the standard deviation

Fig. 4

Imaging cytometry of virus-infected cells. MDCK cells were infected with influenza PR8 virus at an MOI of 10 PFU per cell. At indicated time points, cells were fixed and stained for vRNPs and cell nuclei using DAPI (a) or for M1 and cell nuclei using 7-AAD (d). a Images of individual cells at selected time points after the infection. Each panel series shows the DAPI, vRNP, and bright field (BF) signal of representative cells. b, c Spatio-temporal analysis of virus replication. Image analysis was performed using the IDEAS software. Diagrams show the following fractions (%): b Infected cells (showing a vRNP signal) and apoptotic cells (showing nuclear fragmentation); c vRNPs in the nucleus (based on the amount of vRNP signal co-localized with the DAPI signal) and M1 in the nucleus (based on the proportion of M1 signal co-localized with the 7-AAD signal). d Images of individual cells at selected time points after the infection. Each panel series shows the 7-AAD, M1, and BF signal of representative cells. First and second dashed lines indicate time points when ≥95 % of the final titer of infectious and total virus particles were released, respectively. Error bars indicate standard deviations of three independent experiments

Fig. 5

Evaluation of cell cycle and infection status. MDCK cells were infected with the influenza PR8 virus at an MOI of 0.1. Infection and cell cycle analysis were performed using imaging flow cytometry by staining for vRNPs and DAPI, respectively. a Comparison of the fraction of infected cells in the total population and in each cell cycle phase. b Proportion of infected cells in the different cell cycle phases. c Percentage of non-infected cells in the different cell cycle phases in mock cultures