Specific gangliosides function as host cell receptors for Sendai virus.

The ability of specific gangliosides to function as host cell receptors for Sendai virus was investigated by using Madin-Darby bovine kidney cells which become resistant to infection upon treatment with Vibrio cholerae sialidase. Sialidase-treated cells were incubated for 20 min at 37 degrees C with individual, highly purified gangliosides containing homogeneous carbohydrate moieties and then inoculated with virus for 10 min. Susceptibility of the cells to infection was monitored by hemagglutination titer of the virus produced 48 hr after inoculation. Incubation of the cells with gangliosides containing the sequence NeuAc alpha 2,3Gal beta 1,3GalNAc (i.e., GD1a, GT1b, and GQ1b) fully restored susceptibility to infection to the cells. However, the ganglioside GQ1b in which the sequence ends with two sialic acids in a NeuAc alpha 2,8NeuAc linkage instead of a single sialic acid as in GD1a and GT1b, was effective as a receptor at a concentration 1/100th that of any of the other gangliosides tested. Incubation with gangliosides similar in structure to GD1a, GT1b, and GQ1b but lacking the sialic acid attached to the terminal galactose (i.e., GM1 and GD1b) had no effect. The results from control experiments in which gangliosides were incubated at 0 degrees C with cells or in which trypsin was used to remove gangliosides adsorbed to cells were consistent with the premise that the gangliosides must actually insert into the cellular membrane to function as Sendai virus receptors. Addition of 4 X 10(6) molecules of 14C-labeled GD1a per cell made the cells fully susceptible to infection. Analysis of the ganglioside content of cell membranes showed that gangliosides GD1a, GT1b, and GQ1b are natural components of these cells and are present in quantities sufficient to act as receptors. These results demonstrate that gangliosides with the proper carbohydrate sequence, such as GD1a, GT1b, and GQ1b, function as natural receptors for Sendai virus in host cells.