Overexpression of the alpha-2,6-sialyltransferase in MDCK cells increases influenza virus sensitivity to neuraminidase inhibitors

Abstract

No reliable cell culture assay is currently available for monitoring human influenza virus sensitivity to neuraminidase inhibitors (NAI). This can be explained by the observation that because of a low concentration of sialyl-alpha2,6-galactose (Sia[alpha2,6]Gal)-containing virus receptors in conventional cell lines, replication of human virus isolates shows little dependency on viral neuraminidase. To test whether overexpression of Sia(alpha2,6)Gal moieties in cultured cells could make them suitable for testing human influenza virus sensitivity to NAI, we stably transfected MDCK cells with cDNA of human 2,6-sialyltransferase (SIAT1). Transfected cells expressed twofold-higher amounts of 6-linked sialic acids and twofold-lower amounts of 3-linked sialic acids than parent MDCK cells as judged by staining with Sambucus nigra agglutinin and Maackia amurensis agglutinin, respectively. After transfection, binding of a clinical human influenza virus isolate was increased, whereas binding of its egg-adapted variant which preferentially bound 3-linked receptors was decreased. The sensitivity of human influenza A and B viruses to the neuraminidase inhibitor oseltamivir carboxylate was substantially improved in the SIAT1-transfected cell line and was consistent with their sensitivity in neuraminidase enzyme assay and with the hemagglutinin (HA) receptor-binding phenotype. MDCK cells stably transfected with SIAT1 may therefore be a suitable system for testing influenza virus sensitivity to NAI.

FIG. 1.

Binding of the linkage-specific lectins SNA and MAA to cell surface receptors of SIAT1-transfected and parent MDCK cells. Top panel, MDCK-SIAT1 cells (shaded profiles) and MDCK cells (open profiles) were incubated with DIG-labeled lectins followed by incubation with fluorescein isothiocyanate-labeled anti-DIG antibodies and subjected to FACS analysis. Bottom panel, preparations of cell plasma membranes adsorbed in the wells of a 96-well plate were incubated with DIG-labeled lectins followed by incubation with peroxidase-labeled anti-DIG-antibodies. The data represent ratios of absorbencies in the wells coated with membranes of MDCK-SIAT1 cells to those in wells coated with membranes of MDCK cells.

FIG. 2.

Scatchard plots for the binding of A/Memphis/14/96-M (H1N1) (top panel) and its egg-adapted mutant E1 (bottom panel) to plasma membranes from MDCK cells (closed boxes, solid line) and from MDCK-SIAT1 cells (open circles, dotted line). Data points represent results of two replicate experiments performed on the same plate as described in Materials and Methods.

FIG. 3.

Effect of oseltamivir carboxylate on the formation of virus foci in parallel cultures of MDCK (M) and MDCK-SIAT1 cells (ST) infected with the same virus multiplicity. Experiments were performed in 96-well plates as described in Materials and Methods using clinical virus isolates A/Sydney/5/97 (H3N2) (top panel), A/Memphis/14/96-M (H1N1) (middle panel), and B/Memphis/25/99 (bottom panel). The infection time was 24 h for the type A viruses and 48 h for the type B virus.

FIG. 4.

Comparison of clinical isolates (wt) and their drug-resistant variants with NA mutations (mut) by focus reduction assay. (a) A/Sydney/5/97-like virus (H3N2) and its variant with NA-substitution R292K assayed in MDCK-SIAT1 cells. Infection time was 40 h. (b) A/Wuhan/359/95-like virus (H3N2) and its variant with NA mutation E119V assayed in parallel in MDCK cells (two upper rows) and in MDCK-SIAT1 cells (two lower rows). Infection time was 48 h.