Monoclonal antibodies to the glycoprotein of vesicular stomatitis virus (New Jersey serotype): a method for preliminary mapping of epitopes

Abstract

Of the nine antigenic determinants on the glycoprotein (G) of the New Jersey serotype of vesicular stomatitis virus (VSV) identified by competitive binding of 25 monoclonal antibodies (MAbs), those relegated to epitopes I, II, III, and IV exhibited no significant ability to neutralize virus infectivity but some nonneutralizing MAbs cross-reacted by ELISA with the G protein of VSV-Indiana. High-titered neutralization of homotypic virus was exhibited by epitope V, VI, and VII MAbs but quite variable neutralizing activity was found among MAbs of epitope "family" VIII and particularly the heterogeneous epitope "family" IX. Peptide mapping of the epitopes was not feasible because most MAbs would not bind by Western blotting to G protein under standard conditions of proteolysis or disulfide bond reduction. Therefore, a technique was devised for roughly locating epitopes by protease footprinting of G protein partially protected by individual MAbs complexed with staphylococcal protein A-Sepharose beads. Under these conditions, MAbs to all nine epitopes protected a similar 12-kDa fragment of the G protein from proteolysis by Staphylococcus aureus V8 protease. N-Terminal amino acid sequencing mapped two of these 12-kDa peptide footprints to a position on the G protein extending from amino acid 219 to about 100 amino acids downstream. Although MAbs to only one epitope bound to the 12-kDa fragment by Western blotting, these data suggest, but do not prove, that all nine epitopes of the undenatured VSV-New Jersey G protein are clustered at the middle 20% of a highly structured protein. This method may help to identify the general regions for epitopes on complex proteins of as yet unknown three-dimensional structure.