Fig. 1: Design strategy for generating pH-dependent protein filaments.

a, Top: pH-dependent protein trimers are converted into monomers by addition of short loops connecting the subunits (blue). The connected monomers contain two or three extended hydrogen bond networks, each involving three buried histidine residues. The positions of the hydrogen bond networks are indicated by the red squares. Bottom: the top-down view shows the extensive hydrogen bond network with three central asparagines coordinated by histidine residues, which accept hydrogen bonds from the asparagines and donate hydrogen bonds to threonines on the outer helices. As each histidine makes two hydrogen bonds, protonation completely disrupts the networks. b, Computational sampling of alternative helical packing arrangements of subunits shown in a viewed from the side (top) and from above (bottom). Each subunit is shown in a different colour. The black arrows represent the rigid body transform that generates the helical assembly. c, Compact packing arrangement of fibre subunits in the lowest energy assemblies following sequence design. The helices are shown as rods and the pH-sensitive hydrogen bond networks are shown as sticks. Each subunit is shown in a different colour.