“…Nevertheless, our in vitro experiments, which used recombinant HA0 (either as soluble ectodomain or native, membrane-anchored HA on the surface of transiently-transfected cells), confirmed that head-stem interface antibodies must be able to recognize their epitope on some conformation of prefusion HA, because HA0 cannot transition to a stable postfusion structure 8 . Structural, biophysical, and computational approaches indicate that HA trimers transiently adopt conformations that expose epitopes normally occluded in the defined prefusion state 10–13 . Such transient fluctuations might explain how B cells and antibodies recognize ordinarily occluded epitopes 16,17,20,39–41 .…”