The mechanism of neutralization of a type A influenza virus by polyclonal IgM was similar for both tracheal epithelial and BHK cells. Maximum neutralization was only 90 % and most (70 %) of the virus failed to attach to inoculated cells. The remainder attached to Nacetylneuraminic acid receptors but was not internalized. IgM aggregated virus, but only at an IgM:virus ratio below the level required for neutralization. Failure to detect any loss of infectivity associated with aggregation suggested that aggregates were unstable. Monoclonal polymeric IgA neutralized virus more efficiently on BHK cells (99-9%) than the equivalent amounts of IgM (90 %). Otherwise the mechanisms of IgA and IgM neutralization were similar, except that IgA-induced aggregation was coincident with loss of infectivity and may thus have contributed to it. However, IgA-neutralized virus attached to tracheal epithelial cells more efficiently than infectious virus, initially using a neuraminidase-sensitive receptor, but then becoming neuraminidase-resistant. Whether the latter IgA-virus complexes were internalized or attached to a neuraminidase-resistant receptor is not known. This use of differentiated murine cells with murine IgA gave neutralization data that differed qualitatively from those obtained with the same antibody and undifferentiated hamster cells.

SUMMARYIt has long been known that the binding of antibodies to viruses can result in a loss of infectivity, or neutralization, but little is understood of the mechanism or mechanisms of this process. This is probably because neutralization is a multifactorial phenomenon depending upon the nature of the virus itself, the particular antigenic site involved, the isotype of immunoglobulin and the ratio of virus to immunoglobulin (see below). Thus not only is it likely that neutralization of one virus will differ from another but that changing the circumstances of neutralization can change the mechanism itself. To give coherence to the topic we are concentrating this review on one virus, influenza type A which is itself well studied and reasonably well understood [1–3]. Reviews of the older literature can be found in references 4 to 7.

No abstract

Quantitative relationships between neutralization, aggregation and attachment to monolayers of chick embryo fibroblast (CEF) cells have been studied using a constant amount of influenza A/fowl plague virus/Rostock/34 (H7N1) and varying amounts of purified mouse polyclonal IgM directed against the haemagglutinin, the major viral neutralization antigen. There are two major types of interaction. (i) At low concentrations of IgM there is aggregation of virus, but no neutralization provided that the aggregates are dispersed by vortexing and dilution. Maximum aggregation occurs at less than seven molecules of IgM per virion and the IgM is probably bound in the 'staple' or 'crab' conformation at these concentrations. (ii) At higher concentrations there is neutralization and this coincides with inhibition of attachment of virus to CEF cells. Neutralization of 50 % infectivity requires about 35 molecules of IgM per virion. The maximum neutralization observed was only 87%. Quantitative data and electron microscopy observations suggest that molecules of IgM at the higher concentrations adopt a planar stance approximately perpendicular to the viral surface. It appears that IgM neutralizes fowl plague virus in vitro primarily by interfering with its attachment to cells; the fraction of neutralized virus that does attach is known not to be internalized.