Fig. 6: Multilayered septin filament assembly observed by HS-AFM.

a Contrast adjusted HS-AFM frame of septin filaments on which second layer septin molecules start to adhere on the first layer filaments (false-color scale is adapted to enhance the strongly protruding features in blue; the dashed white the periodicity of the features, roughly perpendicular to the aligned first layer filaments). b Distance distribution of recruited second layer septins. Peaking at ~30 nm periodicity, matching the septin rod length and thus suggestive of an interaction with Cdc11 or Cdc10 (Inset: cartoon of second layer septin rods on the first layer filaments). c HS-AFM image sequence of a second-layer septin filaments assembly on top of a first layer by injection of fresh septin solution (in low salt buffer: 85 mM KCl). Following second-layer formation initiation of filament assembly for the formation of a third layer was observed (arrowhead in t = 310 min). d Septin multilayer assemblies can only be established within a rather narrow window of KCl concentration (50 mM to 150 mM KCl). Data are represented as mean ± s.d from ≥3 different experiments. e Second layer assembly is guided by the first layer serving as a template: ~80% of the second layer filaments align with the sub-layer filaments, independent of the second-layer filament surface coverage. Inset: Height section profile (along the dashed line in image t = 70 min) showing the two layers of ~4 nm in height, where the surface of the first layer is set to 0 nm. Data are represented as mean ± s.d from ≥3 different imaging areas from 1 representative time-lapse experiment. f Both packing orientation and filament pairing are strongly influenced by the first septin layer serving as template. Data are presented as mean ± s.d. from ≥3 different imaging areas from 1 representative time-lapse experiment.