Identification of immunodominant neutralizing epitopes on the hemagglutinin protein of rinderpest virus

Abstract

The immunodominant epitopes on the hemagglutinin protein of rinderpest virus (RPV-H) were determined by analyzing selected monoclonal antibody (MAb)-resistant mutants and estimating the level of antibody against each epitope in five RPV-infected rabbits with the competitive enzyme-linked immunosorbent assay (c-ELISA). Six neutralizing epitopes were identified, at residues 474 (epitope A), 243 (B), 548 to 551 (D), 587 to 592 (E), 310 to 313 (G), and 383 to 387 (H), from the data on the amino acid substitutions of hemagglutinin protein of MAb-resistant mutants and the reactivities of MAbs against RPV-H to the other morbilliviruses. The epitopes identified in this study are all positioned on the loop of the propeller-like structure in a hypothetical three-dimensional model of RPV-H (J. P. M. Langedijk et al., J. Virol. 71:6155-6167, 1997). Polyclonal sera obtained from five rabbits infected experimentally with RPV were examined by c-ELISA using a biotinylated MAb against each epitope as a competitor. Although these rabbit sera hardly blocked binding of each MAb to epitopes A and B, they moderately blocked binding of each MAb to epitopes G and D and strongly blocked binding of each MAb to epitopes E and H. These results suggest that epitopes at residues 383 to 387 and 587 to 592 may be immunodominant in humoral immunity to RPV infection.

FIG. 1.

Relationships of diversity of amino acid sequences among morbilliviruses to positions associated with amino acid changes in MR mutants and conservation of those epitopes. (A) Deduced amino acid sequence alignment of the wild-type L strain, MR mutants, and other morbilliviruses. Dots indicate positions of sequence identity with the wild-type L strain. Sequences for comparison with L strains were from the following sources (GenBank accession number): LA (D82982), RBOK (Z30699), MV, Edmonston strain (K01711), and CDV, Onderstepoort strain (D00758). (B) Reactivities of MAbs to those viruses in IFA tests (16). Results of IFA tests are shown as the following ascitic fluid dilution endpoints: ++, ≧1:1000; +, <1:1000 to ≧1:100; −, <1;100.

FIG. 2.

The predicted positions of neutralizing epitopes on the H protein of L strain by hydropathy plots (A) and comparison of c-ELISA titers in five rabbits infected with RPV (B). The relative hydropathic index was calculated by the method of Chou and Fasman (2). Sera collected from five rabbits 4 weeks after infection with RPV were tested by c-ELISA with biotinylated anti-H MAbs 50 (to epitope group A), 30 (to B), 59 (to D), 47 (to E), 31 (to G), or E-1 (to H) as competitors. The serum titer was defined as the reciprocal of the highest serum dilution that reduced the specific absorbency by more than 50%. Columns indicate the titers of the individual rabbit samples. Results shown are representative of two experiments.

FIG. 3.

Locations of the neutralizing epitopes of RPV-H in a schematic illustration (A) and in the 3D structure model proposed by Langedijk et al. (9) (B). Stars indicate the predicted N-linked glycosylation sites. The six sheets in the β-propeller structure are shown as rectangles. Stem and transmembrane (TM) regions and the direction of the polypeptide chain (arrows) are indicated.