Flagellar motility has arisen independently three times in evolution, in bacteria, archaea, and eukaryotes. Just as birds, bats and bees all have wings, which are analogous but not homologous structures, bacteria, archaea and eukaryotes have flagella that have no homology with each other. While the flagellar motility of protozoa was first observed by Leeuwenhoek in the 17 th century, it took more than three centuries to understand the structural basis for this motility. In prokaryotes (bacteria and archaea) flagellar filaments are typically a homopolymer of a single protein, bacterial or archaeal flagellin. In a helical polymer all subunits are in an equivalent environment (ignoring the ends of such filaments), but bacterial and archaeal flagellar filaments form supercoils that are needed to generate thrust when rotated. How such a homopolymer can supercoil, where subunits are not in equivalent environments, has been a puzzle that was “solved” about 50 years ago by postulating the existence of mixtures of two states in the bacterial filaments. Taking advantage of great advances in cryo-EM, we have been able to generate atomic structures for supercoiled bacterial and archaeal flagellar filaments, which show that the two-state model for bacteria was deeply flawed. We have provided structural insights into how archaeal flagellin evolved from an archaeal type IV pilin, and our studies of type IV pili in both bacteria and archaea have helped elucidate the incredible diversity of these filaments that explains their abundance.
Egelman et al. (2026) studied this question.