Science Spotlight

Deciphering how influenza escapes neutralizing antibodies

From the Bloom Laboratory, Basic Sciences Division

The CDC recommends an annual flu shot to protect against infection by influenza virus during "flu season" which can start in October and last until May1. It is the rapid evolution of the influenza virus that necessitates getting a yearly vaccination.  Influenza's most abundant surface protein is hemagglutinin (HA) and it has been estimated that over two amino-acid changes occur in the protein per year2,3.  In the Bloom Laboratory (Basic Sciences Division), scientists are researching which amino-acid changes are possible in rapidly evolving viruses such as HIV and influenza and which amino acids are most important for shaping viral evolution in nature.  Their research could aid vaccine-design efforts.

When a person becomes infected with a virus, some antibodies can recognize proteins on the surface of the virus as "non-self". Cells in the immune system such as neutrophils and macrophages are then recruited to destroy the virus.  Another class of antibodies called neutralizing antibodies can block the virus's ability to infect cells without needing the help of cells from the immune system.  In a recent publication in PLOS Pathogens, scientists in the Bloom Lab present a new method for mapping amino-acid changes in influenza that confer resistance to several neutralizing antibodies.

In classic approaches to studying viral evolution under antibody selection, scientists have incubated the virus, growing in cells in a dish, with a panel of neutralizing antibodies and recorded which amino-acid changes allow the virus to infect cells, proliferate, and become most abundant following the treatment.  One issue with this selection method is that it begins with one specific strain of the virus and therefore can only predict changes that can occur starting from that initial version of the virus.  An additional issue with the classic method is that it only reveals a single amino-acid mutant in each experiment, giving an incomplete view of the full set of mutations conferring resistance. Outside the lab, however, there are many strains of influenza. In order to capture that diversity, scientists in the Bloom Lab have developed methods to create libraries of variant influenza viruses.  Mike Doud, a graduate student in the University of Washington Genome Sciences program, aimed to leverage these methods to subject a library of influenza virus variants to different neutralizing antibodies and then measure which amino-acid changes are most commonly enriched following selection.

The researchers chose to use a strain of influenza that is closely related to the strain used in classic escape-mutant studies, H1N1 (A/WSN/1933), in order to compare their results with those from the traditional method.  Using PCR, they created a library of every possible amino-acid variant at each position of the influenza surface protein HA and then transfected the influenza genome along with reverse-transcription machinery into mammalian cells.  The virus replicates and is packaged into virions inside the cells. They then incubated the resulting virus library with or without neutralizing antibodies, added virus to the cells, and compared the sequences of the viral RNAs in each dish of cells to identify mutations that are enriched specifically during antibody neutralization.

diagram of mutational antigenic profiling workflow
A. Schematic of the mutational antigenic profiling strategy. The enrichment of specific variants of a virus following antibody selection can be measured by deep sequencing the viral RNA following infection. B. An example of which amino-acids are enriched at a single site following selection. Only positively selected residues are shown. The taller the letter representing the amino-acid, the stronger the differential selection. Graphic adapted from the publication.

The investigators found that a unique set of amino-acid mutations confer escape at each epitope site (the site targeted by the antibody), with some epitope sites exhibiting many different mutational routes to escape, and other epitope sites only escaping through a limited set of amino-acid mutations.

diagram of sites of differential selection by neutralizing antibodies
“Escape mutants” enriched following selection under one monoclonal antibody. (left) Logo plots of amino-acids enriched at each site. (Right) The location of the escape mutations projected onto the crystal structure of HA. Image adapted from the publication.

Remarkably, they found that two different antibodies that are known to target an overlapping region on the HA protein actually selected for different amino-acid variants of the HA protein.  Their results suggest that the evolutionary pressures of antibodies can be incredibly specific, as similar antibodies targeting the same epitope site in HA can have both common and unique sets of escape mutations.

Their method, which they called mutational antigenic profiling, could pave the way for studying the effects of the immune system on viral protein evolution.  Said principal investigator Jesse Bloom, “This technique should be of great value in understanding what we should make of the genetic variation that we see in viral genomes.”


Doud MB, Hensley SE, Bloom JD.  2017.  "Complete mapping of viral escape from neutralizing antibodies."  PLOS Pathogens. 13(3):e1006271. doi: 10.1371/journal.ppat.1006271

Additional citations:


2. Smith DJ, Lapedes AS, de Jong JC, Bestebroer TM, Rimmelzwaan GF, Osterhaus AD, et al.  2004.  "Mapping the antigenic and genetic evolution of the influenza virus."  Science.

3. Bedford T, Riley S, Barr IG, Broor S, Chadha M, Cox NJ, et al.  2015.  "Global circulation patterns of seasonal influenza viruses vary with antigenic drift."  Nature.


Funding for this research was provided by the National Institutes of Health (National Institute of General Medical Sciences and the National Institute of Allergy and Infectious Diseases).  JB is supported by the Simons Foundation and Howard Hughes Medical Institute.