Elsevier

Vaccine

Volume 17, Issues 15–16, 9 April 1999, Pages 1942-1950
Vaccine

Characterization of high-growth reassortant influenza A viruses generated in MDCK cells cultured in serum-free medium

https://doi.org/10.1016/S0264-410X(98)00464-2Get rights and content

Abstract

In the present study reassortant influenza A viruses of both the H1N1 and H3N2 type were generated in Madin Darby Canine Kidney cells grown in the absence of fetal bovine serum (MDCK-SF1 cells). To this end, MDCK-SF1 cells were simultaneously infected with one of the high-growth laboratory strains A/Puerto Rico/8/34 (H1N1) or A/Hong Kong/2/68 (H3N2) and recent H3N2 and H1N1 vaccine strains, respectively. Reassortant viruses obtained from these mixed infections were genetically characterized by RT-PCR and restriction enzyme analysis and their growth properties were compared to those of the corresponding field strains. Reassortant H3N2 viruses inherited the matrix and polymerase pa gene whilst H1N1 reassortant viruses inherited the matrix and polymerase pb1 gene of the high-growth parent. Reassortant viruses generally gave higher viral yields, as measured by a haemagglutination assay, than their wild type counterparts. The procedure followed results in the generation of high-growth reassortant viruses in weeks. The use of MDCK-SF1 cells together with these reassortants for generating influenza virus antigens can significantly speed up the vaccine production procedure.

Introduction

Influenza viruses, members of the family Orthomyxoviridae, are the causative agents of annual epidemics of acute respiratory disease. Influenza epidemics are associated with considerable morbidity and mortality, especially in people at risk, i.e. people suffering from heart or lung diseases, diabetics or a malfunction of the immune system. Vaccination is the most effective way to prevent the often fatal complications in these patients during influenza virus outbreaks.

Influenza vaccines contain the surface glycoproteins (haemagglutinin and neuraminidase) of the influenza viruses expected to circulate in the human population in the upcoming season. Currently, trivalent influenza vaccines are used that contain representative strains of influenza A (H1N1 and H3N2) and influenza B viruses. Due to antigenic drift, the antigenic epitopes of the surface glycoproteins continuously change which necessitates yearly adjustments of the vaccine. These changes are monitored by over a hundred National Influenza Centres worldwide. Based on the information collected, the World Health Organization (WHO) in Geneva yearly recommends the composition of the influenza vaccine, usually in February. This leaves the vaccine manufacturers approximately 6 months to produce and register the vaccine for use in the Northern hemisphere. Vaccine production requires embryonated chicken eggs to propagate the influenza virus antigens. In case of an urgent worldwide demand for large quantities of vaccine, for example with the emergence of a novel subtype of influenza A virus (due to an antigenic shift), vaccine production may be hampered by the limited availability of eggs. Recently, an influenza A virus of the H5N1 subtype was isolated from humans[1]. Fortunately, human infections with this virus appeared to be restricted to 18 hospitalized cases. If, however, this virus would have managed to spread in the human population, it could have initiated a new influenza pandemic[2]. In that case it would have been difficult, if not impossible, to obtain sufficient embryonated eggs for rapid production of a new vaccine. The availability of embryonated eggs also can become a problem when flocks of chickens supplying the eggs become the victim of avian diseases or when the embryos themselves are sensitive to the influenza virus used. The time limits on the one hand and the use of eggs on the other hand render influenza vaccine production unflexible and require long-term planning by vaccine manufacturers.

The use of continuous cell lines like Madin Darby Canine Kidney (MDCK) cells for influenza virus antigen production can be considered an attractive alternative to embryonated chicken eggs for several reasons. Firstly, cell cultures are maintained easily and can be expanded in a relatively short time and therefore will allow initiation and scaling up of antigen production at any time. Secondly, human viruses propagated in MDCK cells usually resemble the original human isolate more closely than do viruses propagated in eggs[3]. Also, egg-derived vaccine strains often constitute a heterogenous population of influenza viruses4, 5, 6. Due to adaptation to avian cells, these viruses undergo mutations in their surface glycoproteins which can render the ultimate vaccine less effective7, 8, 9. Thirdly, the use of MDCK cell derived influenza vaccines would overcome allergic reactions in vaccinated individuals sensitive to egg proteins and, finally, from an ethical point of view, it would reduce the amount of animals used for the benefit of human medicine.

A major draw-back in the use of MDCK cells has been the requirement to grow the cells in the presence of fetal bovine serum. The use of fetal bovine serum not only interferes with the activity of trypsin, necessary for propagation of most influenza A viruses, but its biological variation also complicates standardization of the culture conditions. Recently, a MDCK cell line (MDCK-SF1) that overcomes this draw-back has been developed as these cells grow in medium devoid of fetal bovine serum10, 11. MDCK-SF1 cell derived vaccines already have been extensively tested and shown to be equally effective as egg-derived vaccines[12].

Important factors in the production of an influenza vaccine are the growth properties of the virus and the amount of viral antigens (haemagglutinin and neuraminidase) that can be produced per unit of volume. Ideally, viruses should replicate to high titre in a short period of time and yield high amounts of viral antigens at the end of the process. Influenza A virus field strains, selected to be included in egg-based antigen production, are usually being genetically modified to generate a virus containing the genes encoding haemagglutinin and neuraminidase of the field strain while a high-growth phenotype is derived from a laboratory strain. Dual infection of embryonated chicken eggs with repression of replication of the laboratory strain results in a selective advantage of so called high-growth reassortants with the appropriate surface glycoproteins. This reassortment procedure, made possible by the segmented genome of influenza viruses, is currently carried out routinely for the egg-derived influenza vaccine viruses.

Here we describe the application of this approach for the generation of high-growth reassortant influenza A viruses that can be used for viral antigen production in MDCK-SF1 cells. Also, methods to characterize reassortant viruses are described.

Section snippets

Viruses, cells and sera

The H1N1 viruses A/Puerto Rico/8/34 (PR 34), A/Taiwan/1/86 (TW 86), A/Johannesburg/82/96 (JB 96) and A/Shenzhen/227/95 (SZ 95) as well as the H3N2 viruses A/Hong Kong/2/68 (HK 68), A/Wuhan/359/95 (WH 95), A/Nanchang/933/95 (NC 95) and A/Johannesburg/33/94 (JB 94) were obtained from the repository of the Dutch National Influenza Centre. Most of these vaccine strains were primarily obtained from the National Institute for Biological Standards and Control (NIBSC), Potters Bar, UK. All viruses were

Dual infections

For generation of H3N2 and H1N1 reassortant influenza A viruses, PR 34 and HK 68 were chosen as high-growth parents, respectively, because on comparison with a series of other viruses, PR 34 and HK 68 yielded the highest amounts of haemagglutinating units (HAU) when propagated in MDCK-SF1 cells. Replication of these viruses was completely inhibited by their respective antisera since no haemagglutinating units (HAU) could be measured in supernatants of cells infected with PR 34 or HK 68 only.

Discussion

MDCK-SF1 cells, a newly developed MDCK cell line, can be considered a serious candidate for production of tissue culture grown vaccines. In the present study it was shown that these cells allow genetic reassortment to take place after mixed infection resulting in potential vaccine seed strains with a high-growth phenotype.

The reassortant H3N2 and H1N1 influenza A viruses all contained the gene segment encoding the matrix proteins from the high-growth laboratory strains PR 34 or HK 68,

Acknowledgements

This work was supported by the Ministry of Economical Affairs (PBTS project): Senter BIO95004. Part of this work was supported by the SRVI: Foundation for Respiratory Virus Infections. The authors wish to thank Ger van der Water for continuous support.

References (23)

  • J.C de Jong et al.

    A pandemic warning?

    Nature

    (1997)
  • Cited by (45)

    • Purification of Vero cell derived live replication deficient influenza A and B virus by ion exchange monolith chromatography

      2014, Vaccine
      Citation Excerpt :

      Causative agents of the disease are enveloped viruses from the family of Orthomyxoviridae classified in three genera, influenza A, B, and C viruses [2]. Increase of demand for the vaccine and threat of pandemic, together with some other disadvantages of traditional egg based vaccines, have stimulated development and introduction of new production strategies [3–5]. Typically, downstream processes of influenza virus from allantoic fluids consist of clarification by centrifugation, concentration by ultra-filtration and purification by ultra-centrifugation [6].

    • Improving influenza virus backbones by including terminal regions of MDCK-adapted strains on hemagglutinin and neuraminidase gene segments

      2013, Vaccine
      Citation Excerpt :

      Although the production of HGRs in eggs has proved effective in the generation of many suitable vaccine viruses, it has proved difficult for some strains [6]. As an alternative approach, vaccine seed viruses can be isolated in qualified cell lines [7]. It is generally found that viruses isolated from mammalian cell culture do not acquire adaptive mutations in HA and NA that can alter antigenicity and potentially affect the efficacy of the vaccine, as has been documented for some strains isolated from eggs [8,9].

    View all citing articles on Scopus
    View full text