Highlights d Centriolar Plk4 levels oscillate and act as a switch for centriole biogenesis d Oscillations may be generated via an Asl/Plk4 delayed negative feedback loop d Plk4 oscillations are entrained and phase-locked by the Cdk/ Cyclin oscillator (CCO) d Plk4 oscillations can drive centriole biogenesis even when the CCO is perturbed

SummaryCentrosomes and cilia are organized by a centriole pair comprising an older mother and a younger daughter. Centriole numbers are tightly regulated, and daughter centrioles (which assemble in S phase) cannot themselves duplicate or organize centrosomes until they have passed through mitosis. It is unclear how this mitotic “centriole conversion” is regulated, but it requires Plk1/Polo kinase. Here we show that in flies, Cdk1 phosphorylates the conserved centriole protein Sas-4 during mitosis. This creates a Polo-docking site that helps recruit Polo to daughter centrioles and is required for the subsequent recruitment of Asterless (Asl), a protein essential for centriole duplication and mitotic centrosome assembly. Point mutations in Sas-4 that prevent Cdk1 phosphorylation or Polo docking do not block centriole disengagement during mitosis, but block efficient centriole conversion and lead to embryonic lethality. These observations can explain why daughter centrioles have to pass through mitosis before they can duplicate and organize a centrosome.

No abstract

Sas-6, Ana2 and Sas-4 self-organise into macromolecular structures that can be used to probe centriole/centrosome assembly

SummaryCentrioles are small barrel-shaped structures that form centrosomes and cilia [1]. Centrioles assemble around a central cartwheel comprising the Sas-6 and Ana2/STIL proteins. The amino termini of nine Sas-6 dimers form a central hub of ∼12 nm radius from which nine dimer spokes radiate, placing the Sas-6 carboxyl termini at the outer edge of the ∼60 nm radius cartwheel [2]. Several centriole proteins are distributed in a toroid around the cartwheel, and super-resolution light microscopy studies have measured the average radii of these ∼100–200 nm radius toroids with a ‘precision’ — or standard deviation (s.d. or 1σ) — of ±∼10–40 nm. The organization of Ana2/STIL within the cartwheel, however, has not been resolvable. Here, we develop methods to calculate the average toroidal radius of centriolar proteins in the ∼20–60 nm range with a s.d. of just ±∼4–5 nm, revealing that the amino and carboxyl termini of Ana2 are located in the outer cartwheel region.