RESUMO
Contamination by the parvovirus minute virus of mice (MVM) remains a challenge in Chinese hamster ovary (CHO) biopharmaceutical production processes. Although infrequent, infection of a bioreactor can be catastrophic for a manufacturer, can impact patient drug supply and safety, and can have regulatory implications. We evaluated engineering a CHO parental cell line (CHOZN® GS-/- ) to create a new host cell line that is resistant to MVM infection by modifying the major receptors used by the virus to enter cells. Attachment to a cell surface receptor is a key first step in the infection cycle for many viruses. While the exact functional receptor for MVM binding to CHO cell surface is unknown, sialic acid on the cell surface has been implicated. In this work, we used the zinc finger nuclease gene editing technology to validate the role of sialic acid on the cell surface in the binding and internalization of the MVM virus. Our approach was to systematically mutate genes involved in cell surface sialylation and then challenge each cell line for their ability to resist viral entry and propagation. To test the importance of sialylation, the following genes were knocked out: the CMP-sialic acid transporter, solute carrier family 35A1 (Slc35a1), the core 1-ß-1,3-galactosyltransferase-1 specific chaperone (Cosmc), and mannosyl (α-1,3-)-glycoprotein ß-1,2-N-acetylglucosaminyltransferase (Mgat1) as well as members of the sialyltransferase family. Slc35a1 is responsible for transporting sialic acid into the Golgi. Knocking out function of this gene in a cell results in asialylated glycan structures, thus eliminating the ability of MVM to bind to and enter the cell. The complete absence of sialic acid on the Slc35a1 knockout cell line led to complete resistance to MVM infection. The Cosmc and Mgat1 knockouts also show significant inhibition of infection likely due to their effect on decreasing cell surface sialic acid. Previously in vitro glycan analysis has been used to elucidate the precise sialic acid structures required for MVM binding and internalization. In this work, we performed the sequential knockout of various sialyltransferases that add terminal sialic acid to glycans with different linkage specificities. Cell lines with modifications of the various genes included in this study resulted in varying effects on MVM infection expanding on the knowledge of MVM receptors. MVM resistant host cell lines were also tested for the production of model recombinant proteins. Our data demonstrate that resistance against the MVM virus can be incorporated into CHO production cell lines, adding another level of defense against the devastating financial consequences of MVM infection without compromising recombinant protein yield or quality. Biotechnol. Bioeng. 2017;114: 576-588. © 2016 Wiley Periodicals, Inc.