LBNL Masthead A-Z IndexBerkeley Lab mastheadU.S. Department of Energy logoPhone BookJobsSearch
Toward Design of a Universal Flu Vaccine Print

Worldwide, influenza causes substantial deaths and yearly economic burdens, but the highly changeable nature of the flu virus complicates the production of an effective vaccine. The Centers for Disease Control and Prevention (CDC) estimates that the effectiveness of this year's flu vaccine is about 62%. For comparison, this number for childhood vaccines is routinely well over 90%. One factor in determining the vaccine's effectiveness is how closely related the viruses used in the vaccine are to the viruses circulating that year. An international team of researchers, working at synchrotron facilities including the ALS, has solved the crystal structures of antibodies that protect against broad classes of influenza strains. Greater understanding of these antibody structures may aid in the eventual development of a universal vaccine, protecting against all types of influenza viruses and thus eliminating the guesswork that currently limits vaccine effectiveness.

Pinning Down
a Moving Target

Everyone has been hit by the flu at one time or another during their lives; in the very young and elderly the illness can be deadly. Vaccines are available, but one of the difficulties in producing them is that the virus is a moving target: influenza is caused not by just one organism, but by a number of genetically distinct forms ("strains") that are constantly changing. Before each flu season, organizations such as the FDA, WHO, and CDC collect virus samples from around the world and try to predict which flu strains will most likely get people sick that year.

In this study, Dreyfus et al. have determined the structures of several human antibodies that can neutralize a wide range of influenza strains. This structural information, in combination with electron microscope studies, allowed the researchers to understand how these antibodies attach themselves to regions of the virus that the various strains have in common. The persistence of such "conserved" regions across strains suggests that they are pivotal to a virus's survival and are thus less likely to mutate successfully. Understanding how antibodies fixate on the unchanging portions helps pin down this moving target and brings us one step closer to a "universal" flu vaccine.

Influenza has two main types: influenza A and influenza B. Influenza B includes two genetically distinct forms—the Yamagata and Victoria lineages. Influenza A contains 17 subtypes of hemagglutinin (H1–H17) and 9 subtypes of neuraminidase (N1–N9). Hemagglutinin (HA) mediates the binding of the virus to target cells, and neuraminidase (NA) is involved in the release of progeny virus from infected cells. Thus, these proteins are targets for antiviral drugs, in addition to being antigens to which antibodies can be raised. Influenza A viruses are classified into subtypes, such as H1N1 or H3N2, based on antibody responses to HA and NA.

Human antibodies are large proteins that locate and neutralize viruses using elements that recognize specific regions (epitopes) on the HA surface of the viruses. Researchers from the Crucell Vaccine Institute in the Netherlands, using human cells from volunteers recently vaccinated against the flu, discovered three antibodies—CR8033, CR8071, and CR9114—that bind to epitopes on the HA from various lineages and that protected broadly against the flu. CR8033 and CR8071 were found to bind to epitopes on both strains of influenza B lineages, whereas CR9114 additionally bound to lineages of influenza strain A.

To understand these differences, both crystal structures and electron microscopy (EM) reconstructions were determined by researchers at The Scripps Research Institute in collaboration with Crucell. The crystal structures, determined using crystallography techniques at synchrotron x-ray beamlines, including ALS Beamline 5.0.2, provided atomic-level views of portions of the antibodies alone and/or in complex with the HA of the virus. These structures were then fitted into several low-resolution EM maps of the antibody–HA complexes.

A section of a virus strain B/Brisbane/60/2008 (Victoria strain) with the CR9114, CR8071, and CR8033 binding regions (epitopes) colored according to conserved regions across all influenza B virus sequences: red is 98% conserved, orange is 75–98%, yellow is 50–75%.


The fitting of the crystal structures of the Fab (fragment antigen-binding) regions of CR8033, CR8071, and CR9114 into the reconstructions of the antibody–virus HA complex generated from electron microscopy data (gray mesh). Side and overhead views are shown on the top and bottom, respectively.

It was found that the three antibodies bind to distinct regions of the viruses. Structures of CR8033 were found to bind to HA from both types of influenza B lineages. In both cases, contacts are made through the heavy chain portion of the antibody and are cross-reactive because the HA binding pocket of influenza B is very conserved. The antibody CR9114 binds to an epitope that is similar to that of a previously characterized human antibody, FI6, which broadly neutralizes strain A viruses. However, FI6 shows large structural differences from CR9114, indicating that, while they bind to a similar region on various viruses, they employ different strategies for neutralizing those viruses.

High-resolution crystal structure of the Fab section of CR9114 in complex with the HA segment of virus A/Vietnam/1203/2004 (H5N1).

The crystal structures showed how CR9114 likely accomplishes its cross-group neutralizing ability: the antibody recognizes residues (amino acids) that are highly conserved in the different types of influenza virus and can accommodate some of the variable residues and avoid or push away glycans (carbohydrate chains) on the periphery, which vary between subtypes and strains and normally protect the HA surface. Since antibodies similar to CR9114 are likely present in the human population, this suggests that one possible vaccination strategy against both A and B strains would be to trigger production of CR9114-type antibodies.



Research conducted by: C. Dreyfus, R. Khayat, D.C. Ekiert, J.H. Lee, Z. Metlagel, A.B. Ward, and I.A. Wilson (The Scripps Research Institute); N.S. Laursen (The Scripps Research Institute and Aarhus University, Denmark); T. Kwaks, D. Zuijdgeest, M.V. Bujny, M. Jongeneelen, R. van der Vlugt, M. Lamrani, H.J.W. M. Korse, E. Geelen, Ö. Sahin, M. Sieuwerts, J.P.J. Brakenhoff, R. Vogels, W. Koudstaal, J. Goudsmit, and R.H.E. Friesen (Crucell Vaccine Institute, Netherlands); and O.T.W. Li, L.L.M. Poon, and M. Peiris (University of Hong Kong).

Research funding: University Grants Committee, Hong Kong; Achievement Rewards for College Scientists (ARCS) Foundation; National Institutes of Health; Danish Council for Independent Research, Natural Sciences; and the Skaggs Institute. Operation of the ALS is supported by the U.S. Department of Energy, Office of Basic Energy Sciences.

Publication about this research: C. Dreyfus, N.S. Laursen, T. Kwaks, D. Zuijdgeest, R. Khayat, D.C. Ekiert, J.H. Lee, Z. Metlagel, M.V. Bujny, M. Jongeneelen, R. van der Vlugt, M. Lamrani, H.J.W. M. Korse, E. Geelen, Ö. Sahin, M. Sieuwerts, J.P. J. Brakenhoff, R. Vogels, O.T.W. Li, L.L.M. Poon, M. Peiris, W. Koudstaal, A.B. Ward, I.A. Wilson, J. Goudsmit, and R.H.E. Friesen, "Highly conserved protective epitopes on influenza B viruses," Science 337, 1343 (2012).

ALS Science Highlight #263


ALSNews Vol. 338