Rxivist logo

Do maternal antibodies facilitate hemagglutinin imprinting to influenza A viruses encountered early in childhood?

By Katelyn M Gostic, Monique Ambrose, Michael Worobey, James O. Lloyd-Smith

Posted 22 Feb 2017
bioRxiv DOI: 10.1101/110981

In a recent study, we showed evidence for childhood HA imprinting, a phenomenon in which children develop preferential, lifelong immune memory against zoonotic influenza viruses with hemagglutinin (HA) antigens in the same phylogenetic group as the first influenza virus encountered in childhood (Gostic et al., Science, 2016). Although our original study showed strong, population-level HA imprinting effects, it did not resolve the underlying immunological mechanisms. Similar immune imprinting phenomena, where individuals preferentially recall immune responses primed early in life, also influence seasonal influenza epidemiology via antigenic seniority and original antigenic sin, yet the mechanisms underlying all these childhood immune imprinting phenomena remain poorly understood. A recent letter from Dr. H. Lemke (Lemke, Science, 2017) suggested that these childhood imprinting effects might be mediated by the combined action of maternal antibodies (mAbs) and influenza antigen. In other words, that imprinting may require that children are exposed to influenza A virus in the first year of life, while maternal antibodies are still present. Our response (Gostic et al. Science, 2017) argued that this hypothesis seems qualitatively inconsistent with epidemiological and virological data on influenza. However to be thorough, we performed model comparison to test formally whether epidemiological data better supported our originally published models (where primary exposure at any age from 0 to 12 years could induce HA imprinting) or two alternative models that include maternal effects. In maternal effects model 1, we assumed that imprinting could only occur in the first year of life. In maternal effects model 2, we assumed that primary influenza exposures throughout childhood could induce HA imprinting (as in the original study), but primary exposures in the presence of maternal antibodies might enhance imprinting protection. Here, we report the results of these quantitative analyses, and our methods, and show that the addition of maternal effects does not improve our existing models' predictions of birth year-specific pandemic influenza risk.

Download data

  • Downloaded 482 times
  • Download rankings, all-time:
    • Site-wide: 47,873
    • In immunology: 1,274
  • Year to date:
    • Site-wide: 45,575
  • Since beginning of last month:
    • Site-wide: 92,155

Altmetric data

Downloads over time

Distribution of downloads per paper, site-wide


Sign up for the Rxivist weekly newsletter! (Click here for more details.)