Date Published: April 21, 2009
Publisher: Public Library of Science
Author(s): Hui-Ling Yen, J. S. Malik Peiris
Abstract: Hui-Ling Yen and J. S. Malik Peiris discuss a study in PLoS Medicine that provides new information on the human antibody repertoire generated in response to H5N1 influenza virus infection.
Partial Text: Antibodies are a major component of specific immune protection against influenza and remain the established immune correlate of protection for influenza vaccines. The importance of humoral immunity against influenza infection is further highlighted by the apparent success of passive immunotherapy with convalescent sera during the 1918 Spanish influenza pandemic, and more recently by anecdotal reports of treating H5N1 human infection with convalescent sera , . Human monoclonal antibodies to H5N1 viruses have been generated from immortalized human memory B cells obtained from patients who recovered from H5N1 disease  or with combinational antibody library technologies . Some of these antibodies have broad H5N1 cross-clade reactivity ,  or cross-subtype reactivity to H1 viruses , and are effective in suppressing H5N1 virus disease in experimentally infected animals when administered prophylactically or therapeutically .
In the current issue of PLoS Medicine, Hana Golding and coauthors  use whole-genome-fragment phage display libraries (see Glossary) expressing fragments of a clade 1 H5N1 influenza virus (A/Vietnam/1203/04) and a random peptide phage display library to define the conformation-dependent epitopes of two neutralizing human monoclonal antibodies, one with reactivity restricted to clade 1 viruses and the other with capacity for broader cross-clade protection . They go on to define the H5N1 virus reactive antibody epitopes recognized in the convalescent sera from five patients with H5N1 disease collected between 54 and 182 days after hospitalization. H5N1-specific epitopes were identified in HA and NA surface glycoproteins as well as M2e, PB1-F2, and others. To differentiate potential cross-reactive antibody response elicited by previous exposure to H1N1 or H3N2 influenza viruses, control sera obtained from Vietnamese (n = 20) and US (n = 10) residents with no known exposure to H5N1 virus were also analyzed against the H5N1 whole-genome-fragment phage display library. Cross-reactive epitopes were identified in several H5N1 viral proteins, with strong reactions to peptides in HA and M1 and PA. This study provides much-needed information on the human antibody repertoire generated in response to H5N1 influenza virus infection, and these findings open up new avenues of research.
Further work is required to define which of these antibody epitopes elicit antibodies that protect against H5N1 infection, whether such protection spans many of the H5N1 clades and subclades, and whether some of these antibodies provide protection that might even extend to other influenza virus subtypes. It is expected that some of the epitopes in the HA are targets for the neutralizing antibodies; however, it is also important to evaluate whether the antibodies targeting epitopes in NA and M2e may facilitate clearance of H5N1 infection. The protective roles for cross-reactive antibodies targeting NA have been suggested previously . As the NA epitope identified by Golding and colleagues is located in proximity to the enzyme active site, it is possible that the interacting antibody can block NA enzymatic activity and thereby block viral release. Antibodies targeting M2e have been shown to be effective in animal models , and an M2e vaccine is currently being evaluated in clinical trials as an universal vaccine for influenza because of presumed broad subtype cross-reactivity induced by such antigens. M2e has previously been reported as being weakly immunogenic, and such antibodies detected after natural influenza infection are of low titer and of short duration . Golding and colleagues found that four H5N1 convalescent sera (collected 54–113 days post-admission) showed strong M2e antibody titers (≥2,500), while the fifth H5N1 convalescent serum collected at 182 days post-admission showed a low antibody titer (of 100). Whether this reflects the short duration of an M2e antibody response needs to be established. Surprisingly, sera from controls with no exposure to H5N1 virus but with high antibody titres to seasonal influenza viruses had no reactivity to these H5N1 M2e epitopes.