Impact of cultivation conditions on N‐glycosylation of influenza virus a hemagglutinin produced in MDCK cell culture. (4th February 2013)

• Record Type: Journal Article
• Authors: Rödig, Jana Verena; Rapp, Erdmann; Bohne, Jana; Kampe, Michael; Kaffka, Helene; Bock, Andreas; Genzel, Yvonne; Reichl, Udo
• Is Part Of: Biotechnology and bioengineering. Volume 110:Number 6(2013:Jun.)
• Page Start: 1691
• Page End: 1703
• Publication Date: 2013-02-04
• Subjects: Biotechnology -- Periodicals; Bioengineering -- Periodicals; 660.6
• DOI: 10.1002/bit.24834
• Languages: English
• ISSNs: 0006-3592

Abstract

<abstract abstract-type="main" xml:lang="en"> <title>Abstract</title> <p>Manufacturers worldwide produce influenza vaccines in different host systems. So far, either fertilized chicken eggs or mammalian cell lines are used. In all these vaccines, hemagglutinin (HA) and neuraminidase are the major components. Both are highly abundant glycoproteins in the viral envelope, and particularly HA is able to induce a strong and protective immune response. The quality characteristics of glycoproteins, such as specific activity, antigenicity, immunogenicity, binding avidity, and receptor‐binding specificity can strongly depend on changes or differences in their glycosylation pattern (potential <italic>N</italic>‐glycosylation occupancy as well as glycan composition).</p> <p>In this study, capillary gel electrophoresis with laser‐induced fluorescence detection (CGE‐LIF) based glycoanalysis (<italic>N</italic>‐glycan fingerprinting) was used to determine the impact of cultivation conditions on the HA <italic>N</italic>‐glycosylation pattern of Madin–Darby canine kidney (MDCK) cell‐derived influenza virus A PR/8/34 (H1N1). We found that adaptation of adherent cells to serum‐free growth has only a minor impact on the HA <italic>N</italic>‐glycosylation pattern. Only relative abundances of <italic>N</italic>‐glycan structures are affected. In contrast, host cell adaptation to serum‐free suspension growth resulted in significant changes in the HA <italic>N</italic>‐glycosylation pattern regarding the presence of specific <italic>N</italic>‐glycans as well as their abundance. Further controls such as different suppliers for influenza virus A PR/8/34 (H1N1) seed strains, different cultivation scales and vessels in standard or high cell density mode, different virus production media varying in either composition or trypsin activity, different temperatures during virus replication and finally, the impact of β‐propiolactone inactivation resulted—at best—only in minor changes in the relative <italic>N</italic>‐glycan structure abundances of the HA <italic>N</italic>‐glycosylation pattern.</p> <p>Surprisingly, these results demonstrate a rather stable HA <italic>N</italic>‐glycosylation pattern despite various (significant) changes in upstream processing. Only the adaptation of the production host cell line to serum‐free suspension growth significantly influenced HA <italic>N</italic>‐glycosylation regarding both, the type of attached glycan structures as well as their abundances. Biotechnol. Bioeng. 2013; 110: 1691–1703. © 2013 Wiley Periodicals, Inc.</p> </abstract>