A human monoclonal antibody with neutralizing activity against highly divergent influenza subtypes

Abstract

The interest in broad-range anti-influenza A monoclonal antibodies (mAbs) has recently been strengthened by the identification of anti-hemagglutinin (HA) mAbs endowed with heterosubtypic neutralizing activity to be used in the design of "universal" prophylactic or therapeutic tools. However, the majority of the single mAbs described to date do not bind and neutralize viral isolates belonging to highly divergent subtypes clustering into the two different HA-based influenza phylogenetic groups: the group 1 including, among others, subtypes H1, H2, H5 and H9 and the group 2 including, among others, H3 subtype. Here, we describe a human mAb, named PN-SIA28, capable of binding and neutralizing all tested isolates belonging to phylogenetic group 1, including H1N1, H2N2, H5N1 and H9N2 subtypes and several isolates belonging to group 2, including H3N2 isolates from the first period of the 1968 pandemic. Therefore, PN-SIA28 is capable of neutralizing isolates belonging to subtypes responsible of all the reported pandemics, as well as other subtypes with pandemic potential. The region recognized by PN-SIA28 has been identified on the stem region of HA and includes residues highly conserved among the different influenza subtypes. A deep characterization of PN-SIA28 features may represent a useful help in the improvement of available anti-influenza therapeutic strategies and can provide new tools for the development of universal vaccinal strategies.

Figure 1. PN-SIA28 Neutralizing activity.

Influenza hemagglutinin unrooted phylogenetic tree of the different viral strains tested in neutralization assays with PN-SIA28. Viral isolates belonging to group 1 and group 2 are presented in the box B and box A, respectively. A green ‘+’ indicates positive neutralizing activity, a red ‘−’ indicates negative neutralizing activity. As reported in the text, PN-SIA28 is able to neutralize all of the group 1 strains and is also able to neutralize all of the H3N2 isolates spanning 1968 and 1975. *Recombinant HA from (H1N1) A/South Carolina/1/1918 pandemic strain was previously shown to be bound by PN-SIA28 , . Analogously, recombinant HA from H5N1 A/Cygnus Olor/Italy/742/2005 was recognized by PN-SIA28 (data not shown). # H1N1 A/New Caledonia/20/1999 was previously shown to be neutralized by PN-SIA28 as Fab fragment , .

Figure 2. Sequence conservation in hemagglutinin groups and subtypes.

Boxes indicate mutated residues which decrease PN-SIA28 binding to mutated HAs. Circles on the top indicate PN-SIA28 percent binding to each HA alanine mutants compared to binding to wild-type HA: red 25% binding, yellow 50–75% binding. Sequence numbering is based on H1N1 A/PR/8/34 coding region (GenBank accession number ABO21709). Neutralizing activity of PN-SIA28 against each strain is highlighted by a green ‘+’ or a red ‘−’ on the left, indicating neutralizing activity and no neutralizing activity, respectively. *Recombinant HA from H1N1 A/South Carolina/1/1918 pandemic strain was previously shown to be bound by PN-SIA28 , ; analogously, recombinant HA from H5N1 A/Cygnus Olor/Italy/742/2005 was recognized by PN-SIA28 (data not shown). # H1N1 A/New Caledonia/20/1999 was previously shown to be neutralized by PN-SIA28 as Fab fragment , .

Figure 3. Position of the residues on the surface of (H1N1) A/PR/8/34 influenza hemagglutinin trimer decreasing the binding of PN-SIA28.

White areas indicate the HA1 domains of the HA trimer, whereas cyan areas indicate the HA2 domains. Blue regions depict the mutated residues inhibiting PN-SIA28 binding to HA according to Figure 2.

Figure 4. Comparison of the core of PN-SIA28 epitope on the surface of H1 and H3 trimers.

(A) (H1N1) A/PR/8/34 HA: white areas indicate the HA1 domains of H1 trimer whereas cyan areas indicate its HA2 domains. (B) (H3N2) A/Aichi/2/68 hemagglutinin: grey areas indicate the HA1 domains of H3 trimer whereas white areas indicate its HA2 domains. On both pictures, red regions indicate the residues 25 and 45 on HA1 and residues 361 and 362 on HA2 that, when mutated, were able to inhibit PN-SIA28 binding both to H1 and H3 HAs.

Figure 5. Sequence analysis of H3N2 HA1-hemagglutinins of viral isolates neutralized and not neutralized by PN-SIA28.

(A) Aminoacidic sequences of H3N2 strains studied in the paper and in proximity to the region identified on H1N1 hemagglutinin were aligned. Differences among solvent exposed residues close to PN-SIA28 epitope core are boxed. The yellow circle on Lys57Ala mutant indicate 50–75% PN-SIA28 binding compared to wild-type HA. (B) (H3N2) A/Aichi/2/68 hemagglutinin trimer: grey areas indicate the HA1 domains whereas white areas indicate HA2 domains. Red regions indicate the residues able to inhibit PN-SIA28 binding to H3-HA. Yellow circle define the residues involved in the interaction of PN-SIA28 with both, H1 and H3 HAs.