Structure Determination of Microtubules and Pili: Past, Present, and Future Directions

Abstract

Historically proteins that form highly polymeric and filamentous assemblies have been notoriously difficult to study using high resolution structural techniques. This has been due to several factors that include structural heterogeneity, their large molecular mass, and available yields. However, over the past decade we are now seeing a major shift towards atomic resolution insight and the study of more complex heterogenous samples and in situ/ex vivo examination of multi-subunit complexes. Although supported by developments in solid state nuclear magnetic resonance spectroscopy (ssNMR) and computational approaches, this has primarily been due to advances in cryogenic electron microscopy (cryo-EM). The study of eukaryotic microtubules and bacterial pili are good examples, and in this review, we will give an overview of the technical innovations that have enabled this transition and highlight the advancements that have been made for these two systems. Looking to the future we will also describe systems that remain difficult to study and where further technical breakthroughs are required.

Keywords: cryo-EM; fibre; filament; microtubule; pilus; ssNMR.

FIGURE 1

A revolution in cryo-electron microscopy of MTs and pili. Graph showing the cumulative number of electron microscopy database (EMDB) depositions over time at different resolution levels as indicated; SP, single-particle cryo-EM, STA, sub-tomogram averaging. The emergence of new general hardware and software behind the revolution in cryo-EM is shown below the MT timeline, along with black arrows indicating the introduction of several image-processing pipelines for pseudo-helical single MTs in the top graph. Included data was based on the query title:MT AND status:REL or title:pilus AND status:REL at www.emdatasource.org.

FIGURE 2

Cryo-electron microscopy of MTs (A) Images of the lumenal face of undecorated MTs, centred on an inter-dimer interface, but also showing intra-dimer interfaces and lateral interfaces between protofilaments. Grey density is shown for reconstructions of the MT alone; top, EMDB:5193 at ∼8 Å resolution (Sui and Downing, 2010); and bottom; EMDB:7973 at 3.1 Å resolution (Zhang et al., 2018), with the atomic model for the undecorated GMPCPP MT PDB:6dpu (Zhang et al., 2018) fitted into each reconstruction (α-tubulin light blue, β-tubulin dark blue). Failure to resolve differences between α and β-tubulin is a symptom of the earlier study (top), but not the more recent study (bottom), as illustrated by poor (top) or good (top) density differentiation between α and β-tubulin’s S9-S10 loop (within red dashed oval) (B) Left; exemplar cryo-EM structures from single MTs. Coloured cryo-EM densities for MT-binding proteins are shown as indicated on the single MT alone cryo-EM density map EMDB:7973 (Zhang et al., 2018) coloured grey; CAMSAP1 CKK domain (Atherton et al., 2019), MT-binding repeat (MTBR) of Tau (Kellogg et al., 2018) and the motor domain (MD) of kinesin-13 (Benoit et al., 2018). Right; exemplar cryo-EM structures from axonemal doublet MTs. Coloured cryo-EM density for MT-binding proteins are shown as indicated on the bovine tracheal cilia doublet-MT cryo-EM density map EMDB:24664 (Gui et al., 2021); bovine tracheal cilia MT inner proteins (MIPs), surface and innter junction proteins and outer-dynein arm docking complex EMDB:24664 (Gui et al., 2021) and outer-arm dynein from Tetrahymena thermophila EMDB:22677 (Rao et al., 2021).

FIGURE 3

Cryo-electron microscopy of type IV pili. Images of T4P models derived from different resolutions of SP data STA data. Structures providing resolution breakthroughs are shown as example. SP Cryo-EM density and models of “thick” pili (N. meningitidis T4P at 12.5 Å (EMDB 1236; PDBID 2hil) (Craig et al., 2006), T. gonorrhoeae T4P at 6.0 Å (EMDB 8287; PDBID 5kua) (Kolappan et al., 2016) and the T. thermophilus PilA4 T4P at 3.2 Å (EMDB 10647; PDBID 6xxd) (Neuhaus et al., 2020)) and a thin “pilus” (T. thermophilus PilA5 T4P at 3.5 Å (EMDB 10648; PDBID 6xxe) (Neuhaus et al., 2020)) are shown. Cryo-EM density for the thinner T. thermophilus PilA5 T4P derived by STA at 32 Å is also presented (EMDB 3024) (Gold et al., 2015). Cartoon representation of the T. thermophilus PilA4 and PilA5 are also given as the first Cryo-EM backbone/sidechain resolved T4P structures (Neuhaus et al., 2020).