Quantitative analysis of centriolar satellites and control of ciliogenesis by centrosome associated kinases

Abstract

Centrioles nucleate two types of cellular structures: the centrosome and the cilium. Because of the signaling activity that takes place in centrosomes and cilia, these centriole based organelles are central in maintaining cellular homeostasis, control cell fate, growth, and cell cycle progression. Centrosomes are composed of a pair of centrioles enveloped by a protein mesh that nucleates microtubules and serves as the cell’s microtubule organizing center. The older centriole of the pair is called the mother centriole and is deco- rated at its distal tip with a nine-fold symmetric ring of proteinaceous appendages. Cilia are antenna-like pro- jections of the plasma membrane that surround a microtubule-based structure called axoneme which is nu- cleated by the mother centriole. Ciliogenesis starts at the G1/G0 phase of the cell cycle by a mechanism that is not entirely understood. For cilia to emerge, it is necessary that a mature centriole contacts a membrane or vesicle and attaches to it. After that, for the axoneme to grow, there is a need to remove from the mother centriole the inhibitory factors that prevent abnormal centriole size in proliferating cells. The overall aim of my work was to shed light onto some of the regulatory events controlling the onset of cilia formation. For this, I investigated the role of two microtubule-associated kinases and developed methods that allowed the analy- sis of centrosome/cilia associated components using large fluorescence-based datasets. Our lab has identified MARK4 and TTBK2, among others, as positive regulators of cilia formation. I show that MARK4 is associated with purified centrosomes and regulates the levels of the centrosomal pro- tein ODF2 at the sub-distal appendages of the mother centriole. Using fluorescence microscopy and in vivo pull downs, I show that TTBK2 interacts with the distal appendage component Cep164 via the N-terminal WW-domain of Cep164. Although MARK4 and TTBK2 seem to be associated with different appendage pro- teins, previous studies have shown that MARK4 and TTBK2 are necessary for the removal of CP110 and Cep97 (inhibitory factors of axoneme extension) from the distal tip the mother centriole. Presented here is the finding that MARK4 interacts with Cep97 in vivo, and it phosphorylates CP110 in vitro. Since the deple- tion of both MARK4 and TTBK2 causes bigger deficiencies in ciliogenesis and CP110 removal from the mother centriole than the single depletion of either kinase, I could not assign them a hierarchy of importance in the same pathway for cilia biogenesis. The most satisfactory explanation for the results is that MARK4 and TTBK2 action on ciliogenesis is exerted in separate and synergistic pathways. In the vicinity of centrosomes and cilia, protein complexes, named centriolar satellites, are thought to play crucial roles in centriole duplication and cilia formation. Recently it was found that autophagic degrada- tion of the OFD1 protein pool that localizes to the centriolar satellites sets ciliogenesis in motion. Here, I re- port that MARK4 is required for autophagic activation in human retinal epithelial cells. I then pursued the idea that the defect of MARK4 depleted cells to ciliate could be due to persistent OFD1 at the centriolar satellites. I postulated that if this hypothesis was correct, co-depleting ODF1 and MARK4 should rescue the ciliogene- sis defect observed upon a single MARK4 depletion. Indeed, I found that cells lacking MARK4 and OFD1 had a higher probability to ciliate than cells lacking MARK4 alone. This finding involves MARK4 in the au- tophagy-induced OFD1 degradation from the centriolar satellites at the onset of ciliogenesis. Little is known about how the spatial distribution of the centriolar satellites around the centrosome is determined. My aim was to develop quantitative methods to compare centriolar satellite behavior based on the information gathered through the analysis of large datasets of fluorescence microscopy images. With the data obtained, I present two methods that proved well-suited for comparing the centriolar satellite organiza- tion around the centrosome of a subset of satellite proteins, namely PCM1, OFD1, and Cep290. My data revealed that during the serum starvation process used to induce ciliogenesis in cultured cells, a drastic re- organization of satellite particles loaded with OFD1 occurs. However, the same does not happen with centri- olar satellite particles containing either the scaffold protein PCM1 or Cep290. My observations indicate that the protein composition of centriolar satellite particles is regulated according to the proliferative state of the cell. Together, my work contributed to the understanding of MARK4 and TTBK2 function during ciliogene- sis. Furthermore, I anticipate that the development of the quantitative tools for the analysis of centriole satel- lites will allow the systematic study of these protein complexes that has been lacking so far.