Molecular details of centriole assembly emerge from new research
The centriolar protein SAS-6 of C. elegans assembles into a spiral oligomer with 4.5-fold symmetry per turn (viewed from above) (Click to enlarge)
A new paper in eLife, from John Vakonakis and colleagues, sheds light on an essential component of centriole assembly.
Dr Vakonakis and DPhil student Kacper Rogala, collaborating with Pierre Gönczy in Lausanne and colleagues in Oxford, have uncovered the architecture of SAS-5, a key centriolar protein, and provide new clues as to how it might function (1).
All animal cells contain structures known as centrioles. These are essential for cell division, when the two existing centrioles duplicate and organise the mitotic spindle scaffold to ensure that daughter cells receive a proper complement of chromosomes. Centrioles are also important for flagellar movement and for sensing by cilia.
Centriole assembly is tightly regulated. Studies in C. elegans were the first to identify the handful of proteins essential for this process. One of these proteins, SAS-6, is critical for establishing the 9-fold radial symmetry of centrioles that is conserved across organisms, but whose underlying construction may be different. Whilst C. elegans SAS-6 assembles to a spiral structure, in other organisms, including vertebrates and Drosophila, SAS-6 forms a cartwheel structure.
Comparing the assembly to Lego brick construction, Dr Vakonakis says this curious difference in underlying structure is important. 'We are interested in looking at what these differences tell us about centriolar architecture and the mechanism of assembly.'
In C. elegans, SAS-6 associates with another protein called SAS-5. The interaction is essential for centriole formation, but the role that SAS-5 plays is unclear. Studies show SAS-5 shuttling rapidly between the cytoplasm and centrioles throughout the cell cycle, suggesting that it acts as a protein transporter.
Schematic representation of two views of the SAS-5 Implico domain crystallographic structure (Click to enlarge)
Detailed studies of SAS-5 have been difficult because the protein forms artefactual aggregates. Now, Dr Vakonakis and colleagues have generated full-length SAS-5 in vitro for the first time. Their paper in eLife combines structural studies of SAS-5 in Oxford with functional assays of the protein in worm embryos in Lausanne.
The group identified two domains in SAS-5 using X-ray crystallography and biophysical methods - a coiled-coil domain and a novel globular Implico domain. SAS-5 proteins interact with each other via these domains, forming higher-order oligomers with a trimeric coiled-coil and a dimeric Implico domain.
To determine the role of this interaction, the researchers disrupted either domain and assayed the mutant proteins in worm embryos. They found that the embryos failed to form new centrioles indicating that SAS-5 oligomerisation is required for centriole duplication. Furthermore, it was possible to disconnect SAS-5's role in protein shuttling and centriole formation, indicating that SAS-5 may have an additional, perhaps structural, role.
Their work is supported by an independent study from Jordan Raff and Susan Lea at the Dunn School of Pathology. Their accompanying paper in eLife describes how the functional orthologue of SAS-5 in Drosophila, Ana2, can also form oligomers and that this is required for centriole duplication (2).
Cartoon showing the first step in creating the central tube in a C. elegans centriole (Click to enlarge)
Dr Vakonakis thinks that, despite their different underlying structures, orthologues of SAS-5 such as Ana2 function in the same way in other organisms. 'Ana2 has only a coiled-coil domain and it appears to form tetramers. But the self-assembly theme holds true across organisms.'
He suggests that large assemblies of SAS-5 may function in assisting SAS-6 organisation by seeding SAS-6 self-assembly. 'But to fully understand what SAS-5 does in conjunction with SAS-6, we need to find out how full-length SAS-5 and SAS-6 proteins, in their proper oligomerisation states, fit together.
This, he adds, will require a multipronged approach bringing in microscopy and also functional assays.