HI assays were carried out for all of the point mutants and the fold differences between their titre and that of WT Newmarket/79 virus are shown in Table 2. The majority of the mutations had little antigenic effect and none reduced recognition by the Newmarket/79 or Fontainebleau/79 antisera by four fold or more, suggesting that a loss of antigenic recognition required more than one amino RI-2 change. This was unsurprising, as Fontainebleau/79 serum also recognised Sussex/89 WT virus. However, three individual changes resulted in increased titres to one or more ferret serum. In more detail, changes at positions 46, 55, 140, 163, 172, 187, 196, 207, 213, 260, 267 and 310 had little or no effect (a difference of 2-fold or less in titre against most sera). Of the remaining mutations, D159N was the only change to have a pronounced effect on the reactivity to the vaccine sera, resulting in more than a 4-fold increase in titre to Fontainebleau/79 but having no effect on the titre against Newmarket/79. This was consistent with the difference in HA1 sequence at position 159 between the two vaccine strains (Fig. 1). Newmarket/79/D159N also showed a marked increase in titre to antisera against the Kentucky and Florida sub-lineage viruses Newmarket/1/93 and Kentucky/97, increasing the titre by 90-fold and 11-fold respectively. However, this mutation had little effect on antisera raised to the Eurasian viruses including Sussex/89 and Newmarket/2/93, despite all these viruses having asparagine (N) at position 159. This suggests that this particular mutation did not make a significant contribution to the antigenic differences between the 1979 and 1989 viruses but may be important for evolution of the American/Kentucky sub-lineage. The D159N substitution occurred on the top of the HA trimer, close to the receptor binding site but across the trimer interface (Fig. 2 and Fig. 4). In contrast, the most obvious increase in titre against the Eurasian antisera was observed for the single and triple mutants containing the substitution N189K. The difference was slightly more pronounced with the triple mutant T187S/N189K/V196I (Table 2). However, the single substitution at 189 was sufficient to cause a marked and specific increase in recognition by all four of the Eurasian antisera whilst having little effect on titres against American sera. This amino acid maps to the 190 loop of the receptor-binding site (Fig. 4). In contrast to the pattern seen for N189K, the P227S mutation showed more complex changes in antigenicity; this substitution specifically increased recognition by two of the Eurasian antisera, Yvelines/89 and Hong Kong/92, by more than 10 fold and also increased the titre against the American antiserum to Newmarket/1/93. This mutation also caused a subtle increase (three fold) in recognition by antiserum to the vaccine strain Fontainebleau/79 but not Newmarket/79. This mutation occurs within the 220 loop region of the receptor-binding site (Fig. 4). Taken together, these data suggest that, like residue 159, positions 189 and 227 may also be antigenically important for EIV.
↧