Our current research
Centrioles organize centrosomes which function in spindle formation, cell cycle progression, and  directing the organisation of the interphase cell. They also direct formation of cilia, which are critical for multiple signaling processes and motility.

For over a century, supernumerary centrosomes have been known to be found in cancer cells. Moreover, defective centrosomes prevent the correct development of the cortex of the brain in the microcephaly disease family and defective centrioles and cilia result in a large complex of inherited diseases collectively known as the ciliopathies.

We wish to understand the processes whereby the centriole duplicates to organize centrosomes and cilia.  
The centriole duplication cycle

Genetic studies in C. elegans first identified the canonical duplication pathway: SPD2 targets ZYG1 kinase to centrioles upstream of SAS5 and SAS6 that in turn act before SAS4. Studies from our lab and from Erich Nigg’s showed that the related Plk4 kinase drives canonical centriole duplication and de novo formation in human and Drosophila cells. Subsequently, we and others showed that Plk4 levels are regulated through autophosphorylation that enables binding to the F-box protein of SCF to mediate its destruction. We identified its other main partner to be Asterless (Asl); Asl targets Plk4 to centrioles in Drosophila, in human cells it shares that task with human Spd2 (Cep192).

We recently showed that Drosophila Plk4 phosphorylates Ana2, the fly counterpart of SAS5, to enable it to recruit its Sas6 partner in procentriole formation after centriole disengagement in telophase; this mechanism is conserved in human cells.

We now aim to determine how phosphorylation of Ana2 and Plk’s other substrates influence their interactions with both characterised and as yet poorly characterized core components of the centriole and how this ensures procentriole formation at a single site.  

 The microtubules of the procentriole are fully elongated by the time a cell is ready to enter into mitosis. It is now a fully fledged daughter but it is still unable to duplicate or to organize peri-centriolar material. The daughter centriole acquires these properties during passage through mitosis when it is converted into a centrosome.

We recently described the hierarchy of assembly of a complex that extends radially from the core of the centriole to its periphery enabling this conversion process in cultured fly cells. Key to this is the protein Ana1 which forms a complex with the Sas6-associated Cep135 to load them onto the centriole. Asl can then be recruited onto Ana1 and can then recruit both Plk4 and PCM. The main features of this process are conserved in human cells.

A similar part of Asl also interacts with Sas4 raising the new possibility that Ana1 might be a relay that passes Asl on to Sas4. Once conversion is completed, the daughter must disengage from its mother before procentriole formation can be initiated. Disengagement requires Separase, the protease promoting sister chromatid separation, and also Polo kinase through mechanisms that are so far poorly understood.

We would like to understand the roles of the Polo and Plk4 kinases in this process. 

Centriole to centrosome conversion in DMEL cells (from Dzhindzhev et al., 2014)
Sequential loading of Cep135, Ana1 and Asl during centriole-to-centrosome conversion (from Fu et al., 2016)
Supernumerary centrosomes in the mouse 

To address the long-standing question of what might multiple centrosomes be doing in tumour cells, we decided to make a mouse in which we could induce Plk4 expression and hence centrosome overduplication. We found that Plk4 over-expression leads to hyper-proliferation of the skin exacerbated by loss of p53; it also advances tumour formation in p53-/- mice.

Mice overexpressing Plk4 developed grey hair due to a loss of differentiated melanocytes and bald patches of skin associated with a thickening of the epidermis. We found that the balance of cell proliferation to differentiation was disturbed in the skin of these mice apparently because the over-duplication of centrosomes does not permit primary cilia to form. Because the primary cilia are necessary for signaling between cells, this perturbs cell differentiation.

We found that mice over-expressing Plk4 also showed hyperproliferation of pancreatic islets; these also become enlarged as a result of equal expansion of α- and β-cells, which exhibit centrosome amplification.
We now study the effects of Plk4 over-expression in pancreatic organoids established from these mice.  

Plk4 has multiple functions at the centrosome

Plk4’s main function is to drive centriole duplication. However, clues to its other functions in nucleating microtubules have come from studies of the oocytes and early embryos of the mouse, where centrioles have yet to be assembled and the spindle becomes assembled using acentriolar microtubule organizing centres (MTOCs).

We found that, in the early embryo, bipolar spindle formation requires Plk4 function in concert with its partner protein Cep152. However, in the oocyte, which also lacks centrioles, the MTOCs require not only Plk4 but also Aurora A to become fully active.
 We are currently studying this difference in spindle formation in these two acentriolar cell types. 
Plk4 over-expression (Plk4OE/Plk4OE; p53KO/p53KO) leads to loss of hair and its pigmentation (from Coelho et al., 2015)
(A and B) Mouse zygotes stained to reveal α-tubulin (green), Plk4 (red and monochrome, A′ and B′), and DNA (blue). Plk4 (red arrowheads) associates with cytoplasmic MTOCs in interphase and spindle poles in mitosis (B).  (C–G) Time-lapse series of zygotes expressing α-tubulin-mcherry (inverted black) and EGFP-Plk4 (red) showing multiple interphase MTOC-associated Plk4 bodies that coalesce at spindle pole in mitosis. Time is in hr:min  (from Coelho et al., 2013)
Schematic representation of early events in acentriolar spindle assembly in mouse (from Coelho et al., 2013)