Hemagglutinin (HA) crystal structure
HA structure showing mutating amino acids1

Anyone who’s taken a virology class, and many who haven’t, know about “antigenic drift” and “antigenic shift”. These are usually used to explain influenza virus changes over time (although of course the same concepts apply to many other viruses). Antigenic shift refers to large, abrupt changes in the virus;2 antigenic drift refers to smaller changes. Antigenic shifts are associated with pandemic influenza, as the pre-existing immune responses to influenza from previous years aren’t protective any more (because the virus has shifted away from them). Antigenic drift doesn’t let the virus escape immune control altogether, but does give the virus an advantage in infecting people — presumably, people with strong responses are still protected, but those whose exposure was maybe a few years ago, or who happened to make a weaker antibody response, to the previous virus, would be susceptible to the new virus but protected against the original. Antigenic drift happens all the time, and new drifted variants of influenza take over every year or two, which is why we need new seasonal flu vaccines on a regular basis.

What drives antigenic drift? The simple answer is that it’s driven purely by immunity. According to this notion, viruses that are resistant to being neutralized by antibody are most able to replicate and transmit to new hosts. Jon Yewdell’s group, however, has just revisited antigenic drift,1 and propose a somewhat different model: Antigenic drift is the result of viruses cycling between immune and non-immune hosts, and it’s almost a side effect of the way the virus interacts with cells.  (Unusually for Jon, I don’t htink he coined any new and exciting acronyms for his new model.)

Model for antigenic drift
Model for antigenic drift selection (click for larger version)1

Normally, influenza virus doesn’t “want” to bind very strongly to its cellular target, 3 because then it’s harder for newly-formed viruses to escape from the cell (because it binds to the receptors on the way out, as well). But in the presence of neutralizing antibody, the virus needs to bind more strongly to the receptor to overcome the effects of antibody binding. Yewdell’s group argue that it’s this whip-saw effect that pushes long-term changes in antigenicity:

Thus, antigenic drift can be a by-product of Darwinian selection for mutations that optimize host cell receptor binding during influenza A virus transmission between immune (increased receptor binding) and naïve individuals (decreased receptor binding). 1

One difference between Jon’s model and the standard concept is that with the latter, you’d expect that there would be more antigenic drift as immunity increases among the population. Yewdell’s model, though, predicts that sequential passage through immune and non-immune individuals drives antigenic drift. This actually leads to an important prediction:

In our model, antigenic drift is accelerated by sequential passage of influenza A virus between immune and nonimmune individuals, which in the human population are nearly all children. Therefore, decreasing the naïve population size by increasing pediatric influenza A virus vaccination rates will likely slow antigenic drift and temporally extend the effectiveness of influenza vaccines.1

(My emphasis.)  They also point out that, because changes in antigenicity run in parallel with changes in receptor binding affinity, antigenic drift can itself could be pushing other changes in virus personality. That’s because the drift changes push changes in receptor binding, which in turn alters the cells to which the virus can interact; and changing the cells that the virus infects will inevitably change the nature of an infection as well. For example, does the virus best infect cells of the upper respiratory tract — leading to coughs and sniffles — or the lower respiratory tract — leading to pneumonia.


  1. Hensley, S., Das, S., Bailey, A., Schmidt, L., Hickman, H., Jayaraman, A., Viswanathan, K., Raman, R., Sasisekharan, R., Bennink, J., & Yewdell, J. (2009). Hemagglutinin Receptor Binding Avidity Drives Influenza A Virus Antigenic Drift Science, 326 (5953), 734-736 DOI: 10.1126/science.1178258[][][][][]
  2. “Changes” here mean changes in the ways the immune system responds to the virus — hence, “antigenic” changes. In practice, the changes are antibody-based changes, although in principle antigenic shift and drift could also refer to T cell-based recognition.[]
  3. The influenza hemagglutinin, HA, protein binds to sialic acid on cells of the respiratory system.[]