Search

Radial glial progenitor cells (RGPs) are the major neural progenitor cells that generate neurons and glia in the developing mammalian cerebral cortex1-4. In RGPs, the centrosome is positioned away from the nucleus at the apical surface of the ventricular zone of the cerebral cortex5-8. However, the molecular basis and precise function of this distinctive subcellular organization of the centrosome are largely unknown. Here we show in mice that anchoring of the centrosome to the apical membrane controls the mechanical properties of cortical RGPs, and consequently their mitotic behaviour and the size and formation of the cortex. The mother centriole in RGPs develops distal appendages that anchor it to the apical membrane. Selective removal of centrosomal protein 83 (CEP83) eliminates these distal appendages and disrupts the anchorage of the centrosome to the apical membrane, resulting in the disorganization of microtubules and stretching and stiffening of the apical membrane. The elimination of CEP83 also activates the mechanically sensitive yes-associated protein (YAP) and promotes the excessive proliferation of RGPs, together with a subsequent overproduction of intermediate progenitor cells, which leads to the formation of an enlarged cortex with abnormal folding. Simultaneous elimination of YAP suppresses the cortical enlargement and folding that is induced by the removal of CEP83. Together, these results indicate a previously unknown role of the centrosome in regulating the mechanical features of neural progenitor cells and the size and configuration of the mammalian cerebral cortex.

Copyright © 2020 Chong et al. Subdistal appendages (sDAPs) are centriolar elements that are observed proximal to the distal appendages (DAPs) in vertebrates. Despite the obvious presence of sDAPs, structural and functional understanding of them remains elusive. Here, by combining super-resolved localization analysis and CRISPR-Cas9 genetic perturbation, we find that although DAPs and sDAPs are primarily responsible for distinct functions in ciliogenesis and microtubule anchoring, respectively, the presence of one element actually affects the positioning of the other. Specifically, we find dual layers of both ODF2 and CEP89, where their localizations are differentially regulated by DAP and sDAP integrity. DAP depletion relaxes longitudinal occupancy of sDAP protein ninein to cover the DAP region, implying a role of DAPs in sDAP positioning. Removing sDAPs alter the distal border of centrosomal g-tubulins, illustrating a new role of sDAPs. Together, our results provide an architectural framework for sDAPs that sheds light on functional understanding, surprisingly revealing coupling between DAPs and sDAPs. Ministry of Science and Technology, Taiwan (107-2112-M-001-037, 107-2313-B-001-009, 108-2313-B-010-001, 108-2628-B-010-007, 108-2638-B-010-001 -MY2); Academia Sinica (2317-1040300); National Institutes of Health (GM088253).

Centriole-to-centrosome conversion (CCC) safeguards centriole homeostasis by coupling centriole duplication with segregation, and is essential for stabilization of mature vertebrate centrioles naturally devoid of the geometric scaffold or the cartwheel. Here we identified PPP1R35, a putative regulator of the protein phosphatase PP1, as a novel centriolar protein required for CCC. We found that PPP1R35 is enriched at newborn daughter centrioles in S or G2 phase. In the absence of PPP1R35, centriole assembly initiates normally in S phase, but none of the nascent centrioles can form active centrosomes or recruit CEP295, an essential factor for CCC. Instead, all PPP1R35-null centrioles, although stable during their birth in interphase, become disintegrated after mitosis upon cartwheel removal. Surprisingly, we found that neither the centriolar localization nor the function of PPP1R35 in CCC requires the putative PP1-interacting motif. PPP1R35 is thus acting upstream of CEP295 to induce CCC for proper centriole maintenance.

SUMMARY PLK4 is the major kinase driving centriole duplication. Duplication occurs only once per cell cycle, forming one new (or daughter) centriole that is tightly engaged to the preexisting (or mother) centriole. Centriole engagement is known to block the reduplication of mother centrioles, but the molecular identity responsible for the block remains unclear. Here, we show that the centriolar cartwheel, the geometric scaffold for centriole assembly, forms the identity of daughter centrioles essential for the block, ceasing further duplication of the mother centriole to which it is engaged. To ensure a steady block, we found that the cartwheel requires constant maintenance by PLK4 through phosphorylation of the same substrate that drives centriole assembly, revealing a parsimonious control in which “assembly” and “block for new assembly” are linked through the same catalytic reaction to achieve homeostasis. Our results support a recently deduced model that the cartwheel-bound PLK4 directly suppresses centriole reduplication., Graphical Abstract

SummaryVertebrate centrioles lose their geometric scaffold, the cartwheel, during mitosis, concurrently with gaining the ability to recruit the pericentriolar material (PCM) and thereby function as the centrosome. Cartwheel removal has recently been implicated in centriole duplication, but whether “cartwheel-less” centrioles are intrinsically stable or must be maintained through other modifications remains unclear. Here, we identify a newborn centriole-enriched protein, KIAA1731/CEP295, specifically mediating centriole-to-centrosome conversion but dispensable for cartwheel removal. In the absence of CEP295, centrioles form in the S/G2 phase and lose their associated cartwheel in mitosis but cannot be converted to centrosomes, uncoupling the two events. Strikingly, centrioles devoid of both the PCM and the cartwheel progressively lose centriolar components, whereas centrioles associating with either the cartwheel or PCM alone can exist stably. Thus, cartwheel removal can have grave repercussions to centriole stability, and centriole-to-centrosome conversion mediated by CEP295 must occur in parallel to maintain cartwheel-less centrioles for duplication.

The transition zone is a specialized compartment found at the base of cilia, adjacent to the centriole distal end, where axonemal microtubules are heavily crosslinked to the surrounding membrane to form a barrier that gates the ciliary compartment. A number of ciliopathy molecules have been found to associate with the transition zone, but factors that directly recognize axonemal microtubules to specify transition zone assembly at the cilia base remain unclear. Here, through quantitative centrosome proteomics, we identify an axoneme-associated protein, CEP162 (KIAA1009), tethered specifically at centriole distal ends to promote transition zone assembly. CEP162 interacts with core transition zone components, and mediates their association with microtubules. Loss of CEP162 arrests ciliogenesis at the stage of transition zone assembly. Abolishing its centriolar tethering, however, allows CEP162 to stay on the growing end of the axoneme and ectopically assemble transition zone components at cilia tips. This generates extra-long cilia with strikingly swollen tips that actively release ciliary contents into the extracellular environment. CEP162 is thus an axoneme-recognition protein pre-tethered at centriole distal ends before ciliogenesis to promote and restrict transition zone formation specifically at the cilia base.

The distal appendages (DAPs) of centrioles have been proposed to anchor cilia to the plasma membrane, but their molecular composition, assembly, and exact function in ciliogenesis remain poorly understood. Using quantitative centrosome proteomics and superresolution microscopy, we identified five DAP components, including one previously described (CEP164), one partially characterized (CEP89 [ccdc123]), and three novel (CEP83 [ccdc41], SCLT1, and FBF1) DAP proteins. Analyses of DAP assembly revealed a hierarchy. CEP83 recruits both SCLT1 and CEP89 to centrioles. Subsequent recruitment of FBF1 and CEP164 is independent of CEP89 but mediated by SCLT1. All five DAP components are essential for ciliogenesis; loss of CEP83 specifically blocks centriole-to-membrane docking. Undocked centrioles fail to recruit TTBK2 or release CP110, the two earliest modifications found on centrioles prior to cilia assembly, revealing centriole-to-membrane docking as a temporal and spatial cue promoting cilia initiation.

Vertebrate cells can initiate ciliogenesis from centrioles at the cell center, near the Golgi, forming primary cilia confined or submerged in a deep narrow pit created by membrane invagination. How or why cells maintain submerged cilia is unclear. Here, by characterizing centriole sub-distal appendages (sDAP) in cells exclusively growing submerged cilia, we found that a group of sDAP components localize to the centriole proximal end through the cohesion factor C-Nap1, and that sDAP functions redundantly with C-Nap1 for submerged cilia maintenance. Loss of sDAP and C-Nap1 has no effect on cilia assembly, but disrupts stable Golgi-cilia association, and allows normally submerged cilia to fully surface, losing the deep membrane invagination. Intriguingly, unlike submerged cilia (stationary), surfaced cilia actively respond to mechanical stimuli with motions, and can ectopically recruit hedgehog signaling components in the absence of agonist. We propose that spatial control of ciliogenesis uncouples or specifies sensory properties of cilia.

Centriole or centrosome number in cycling cells is strictly maintained through coordinated duplication and segregation. Duplication is limited to once only per cell cycle by separating the assembly event that occurs in S/G2 phase from the two licensing events, centriole disengagement and centriole-to-centrosome conversion, both of which occurs in mitosis. In addition to duplication licensing, centriole-to-centrosome conversion also enables centrioles to associate with spindle poles and thereby to segregate equally during cell division. Centriole disengagement and centriole-to-centrosome conversion thus constitute the major regulatory module ensuring centrosome homeostasis in cycling cells. Using Xenopus egg extracts and purified engaged centrioles, we here describe an in vitro assay allowing us to synchronously induce the initiation of centriole disengagement and centrosome formation, pause the reaction anytime during the process, and more importantly, preserve "reaction intermediates" for further analyses.

Plk1-dependent modification of centrioles early in mitosis is necessary for accurate centriole duplication and segregation., Centrioles are self-reproducing organelles that form the core structure of centrosomes or microtubule-organizing centers (MTOCs). However, whether duplication and MTOC organization reflect innate activities of centrioles or activities acquired conditionally is unclear. In this paper, we show that newly formed full-length centrioles had no inherent capacity to duplicate or to organize pericentriolar material (PCM) but acquired both after mitosis through a Plk1-dependent modification that occurred in early mitosis. Modified centrioles initiated PCM recruitment in G1 and segregated equally in mitosis through association with spindle poles. Conversely, unmodified centrioles segregated randomly unless passively tethered to modified centrioles. Strikingly, duplication occurred only in centrioles that were both modified and disengaged, whereas unmodified centrioles, engaged or not, were “infertile,” indicating that engagement specifically blocks modified centrioles from reduplication. These two requirements, centriole modification and disengagement, fully exclude unlimited duplication in one cell cycle. We thus uncovered a Plk1-dependent mechanism whereby duplication and segregation are coupled to maintain centriole homeostasis.

SummaryIt has been proposed that separase-dependent centriole disengagement at anaphase licenses centrosomes for duplication in the next cell cycle. Here we test whether such a mechanism exists in intact human cells. Loss of separase blocked centriole disengagement during mitotic exit and delayed assembly of new centrioles during the following S phase; however, most engagements were eventually dissolved. We identified Polo-like kinase 1 (Plk1) as a parallel activator of centriole disengagement. Timed inhibition of Plk1 mapped its critical period of action to late G2 or early M phase, i.e., prior to securin destruction and separase activation at anaphase onset. Crucially, when cells exited mitosis after downregulation of both separase and Plk1, centriole disengagement failed completely, and subsequent centriole duplication in interphase was also blocked. Our results indicate that Plk1 and separase act at different times during M phase to license centrosome duplication, reminiscent of their roles in removing cohesin from chromosomes.

Vertebrate centrioles normally propagate through duplication, but in the absence of preexisting centrioles, de novo synthesis can occur. Consistently, centriole formation is thought to strictly rely on self-assembly, involving self-oligomerization of the centriolar protein SAS-6. Here, through reconstitution of de novo synthesis in human cells, we surprisingly found that normal looking centrioles capable of duplication and ciliation can arise in the absence of SAS-6 self-oligomerization. Moreover, whereas canonically duplicated centrioles always form correctly, de novo centrioles are prone to structural errors, even in the presence of SAS-6 self-oligomerization. These results indicate that centriole biogenesis does not strictly depend on SAS-6 self-assembly, and may require preexisting centrioles to ensure structural accuracy, fundamentally deviating from the current paradigm. DOI: http://dx.doi.org/10.7554/eLife.10586.001, eLife digest Cells pass on their characteristics or “traits” to new generations in the form of DNA molecules. DNA provides the instructions to make proteins, which may then assemble into larger structures without using any external templates in a process called self-assembly. However, when a cell divides, DNA is not the only element that is passed on to the daughter cells; many large protein structures that have assembled in mother cells are also divided between the daughter cells. The daughter cells may then produce extra copies of these protein structures, but it is not known whether the pre-existing structures are involved in this process. Centrioles are complex structures made of proteins and play a crucial role in cell division. One of the main components of centrioles is a protein called SAS-6. Recent studies have shown that SAS-6 molecules can bind to each other to form “oligomers”. This process, which is called self-oligomerization, has been proposed to drive the formation of centrioles. Now, Wang et al. examine whether centrioles can form properly in cells when no other centrioles are present. The experiments show that centrioles can indeed form, but they are prone to structural errors. In contrast, centrioles that form in the presence of older centrioles are essentially free of errors. The experiments used human eye cells that were missing the gene that encodes SAS-6. These cells could not make centrioles, but when SAS-6 was re-introduced into these cells, new centrioles formed. Unexpectedly, re-introducing a mutant form of SAS-6 that cannot form oligomers into the cells still allowed new centrioles to form, which shows that self-oligomerization of SAS-6 is not essential for the assembly of centrioles. Together, Wang et al.’s findings challenge the idea that SAS-6 self-oligomerization is involved in the formation of centrioles, and suggest that preexisting centrioles may help to minimize errors in the formation of new centrioles. DOI: http://dx.doi.org/10.7554/eLife.10586.002

The precise trafficking and spatial organization of signaling molecules within cells is critical for many fundamental cellular processes. Two interconnected microtubule-based organelles, the centrosome and primary cilium, have been making headlines recently due to their role as central “hubs” for coordinating such signaling events. The centrosome is the major microtubule-nucleating center in animal cells, which polarizes microtubule arrays and thereby directs microtubule-based trafficking toward itself and its associated structure, the primary cilium. [1]. The primary cilium is a tiny hair-like sensory organelle that is templated by one of two centrioles, core elements of the centrosome, and protrudes above the apical surface of almost every cell in the human body (Figure 1). Together, the centrosome and cilium mediate the initiation and transmission of extracellular signals to the interior of the cell, thus controlling many aspects of cell physiology [2], [3]. Defects in the structure and/or function of these organelles result in human disease conditions termed “ciliopathies,” a heterogeneous group of disorders with phenotypes including cystic kidneys; digit, bone, and brain anomalies; infertility; and even cancer [4], [5]. Figure 1 Schematic of centriolar satellites undergoing microtubule-dependent movement towards the centrosome and cilium. Human diseases associated with centrosome-cilium aberrations generally result from defects in the proper assembly of the centrosome or cilium, or the trafficking of proteins to these structures. It is now well established that movement of proteins to and from the centrosome-cilium complex is facilitated by centriolar satellites, non-membranous 70–100 nm cytoplasmic granules that concentrate around the centrosome in animal cells (Figure 1) [6]. Centriolar satellites are generally defined by PCM1, a large scaffolding protein reported to function in dynein-dependent, microtubule-based trafficking of centrosomal and ciliary proteins [7], [8]. Although the proteome of centriolar satellites remains unknown, a number of proteins have now been shown to associate with PCM1, many of which are mutated in patients with ciliopathies [9]–[14]. However, the functional significance of centriolar satellites in mammalian development and ciliogenesis in vivo remains unclear. In this issue, Hall and colleagues [15] address the role and requirement of Azi1, a conserved satellite protein, in regulating cilium function during development. Azi1 was originally identified as a centrosomal protein [16] and localized to centriolar satellites [17]. Hall et al. spatially refined the localization pattern of mouse Azi1, and discovered that Azi1 is fairly dynamic and undergoes microtubule-based trafficking to and from the centrosome. They noted higher levels of Azi1 around centrioles and surrounding centriolar satellites of ciliated cells compared to non-ciliated cells, and identified a further pool of Azi1 at the transition zone, an area at the base of cilia involved in regulating protein entry into the ciliary space [18]. Together, these results show a redistribution of Azi1 upon initiation of ciliogenesis, hinting at a functional role in regulating ciliary protein trafficking. Indeed, Azi1 was previously shown to be essential for ciliogenesis in zebrafish and flies [19], [20], and a siRNA screen in human cells also identified Azi1 as necessary for cilium formation [21]. Similarly, Hall et al. find that knockdown of Azi1 in mouse cells using siRNA, which they refer to as “acute” loss of the protein, results in a significant proportion of cells that lack primary cilia. Given the high conservation of Azi1 among ciliated organisms, the loss-of-function phenotypes from various model organisms, as well as the siRNA results, it was safe to predict that Azi1 would have a central role in mammalian cilia biology in vivo. Thus, the authors generated mouse mutants null for Azi1 to test the role of this protein (and centriolar satellites) in mammalian development. Remarkably, Azi1 null mice are viable, and display none of the gross abnormalities commonly associated with either primary or motile cilia dysfunction. Examination of primary mouse embryonic fibroblasts (MEFs) isolated from Azi1 null mice showed that formation and compartmentalization of primary cilia appear normal, with normal distribution of ciliary membrane and transition zone proteins. Similarly, there were no gross defects with regard to centrosome duplication or centriolar satellite composition and distribution in cells lacking Azi1. Ultrastructure analysis of motile multiciliated tracheal epithelia also failed to detect any defects in centrioles/basal bodies and cilia. Collectively, and in stark contrast with the acute Azi1 loss in human and mouse cells, primary and motile cilia structure and function appear grossly normal in Azi1 null animals in vivo, as well as in Azi1 null MEFs grown in vitro. The difference in phenotypes observed following acute knockdown of Azi1 versus permanent deletion of the gene (referred to as “chronic” loss by the authors) is suggestive of a compensation mechanism that, in essence, rescues the defects caused by loss of Azi1. Intriguingly, while such compensation was seen in most tissues, Azi1 null male mice were infertile, exhibiting post-meiotic defects in spermatogenesis. Sperm cells lacked flagella and any remaining axonemes were truncated and immotile, likely due to mistrafficking of proteins required for flagellar assembly. This indicates that Azi1 (and by extension centriolar satellites) plays a critical role in regulating trafficking of proteins in the highly specialized sperm cells that, unlike other ciliated cell types, cannot compensate for loss of Azi1. This tissue-specific phenotype is reminiscent of human mutations in a number of genes required for centrosome assembly and duplication, where the predominant phenotype appears to be microcephaly [22]–[26]. These latter findings suggest that, during development, most tissues can compensate and overcome mutations in centrosomal proteins, except for neural progenitor cells, which appear most sensitive to these perturbations. The findings by Hall et al. similarly indicate that sperm cells may be especially sensitive to mutations in centriolar satellite genes. The functional compensation reported by Hall et al. highlights the complexity and challenge in understanding cilia assembly and function in the context of a developing organism. The basis of the compensation, at this point, is unclear, as it cannot be simply attributed to genetic redundancy, but a mechanistic understanding of the process will be critical if more cases like Azi1 are discovered in the future. Interestingly, the Azi1 null MEFs isolated by Hall et al. are shown to be “active” in compensating for the loss of Azi1; it will be very useful to test whether the compensation can be reversed (or switched on/off) by inducibly expressing exogenous Azi1 in these cells. One interesting strategy would be to utilize the Azi1 null MEFs as an effective in vitro system to understand how functional compensation works in this case, or with centriolar satellites in general. Moreover, this study also highlights the importance of functional follow-up studies of siRNA-based data and cautions against direct extrapolation of transient loss-of-function phenotypes to the genetic in vivo phenotype. On the other hand, transient loss-of-function studies may expose roles of genes that would otherwise be overlooked due to compensation in long-term mutagenesis studies in vivo. As such, it will be important to combine both approaches in future studies.

The centrosome is duplicated just once per cell cycle so that there is one centrosome at the beginning of the cell cycle and two in mitosis. Defects in this process can cause genome instability and even cancer. An in vitro study points to a strikingly similarity between the way cells control centrosome duplication and chromosome segregation. The pair of centrioles in the centrosome disengage at the end of mitosis, a process requiring separase, a protease involved in sister chromatid separation at mitosis. The separated centrioles are a substrate for duplication in the next cell cycle. Newly duplicated centrioles are paired, so they cannot duplicate again until passage through mitosis, when they are freed to duplicate by separase-mediated disengagement. This work reveals a common mechanism behind the two cellular processes that result in exact duplication once per cycle. Centrosome duplication needs to be tightly controlled to occur once and only once per cell cycle. The regulated timing of centriole disengagement in late mitosis is shown to allow one round of centrosome duplication in the next cycle. The centrosome organizes the microtubule cytoskeleton and consists of a pair of centrioles surrounded by pericentriolar material. Cells begin the cell cycle with a single centrosome, which duplicates once before mitosis. During duplication, new centrioles grow orthogonally to existing ones and remain engaged (tightly opposed) with those centrioles until late mitosis or early G1 phase, when they become disengaged1. The relationship between centriole engagement/disengagement and centriole duplication potential is not understood, and the mechanisms that control these processes are not known. Here we show that centriole disengagement requires the protease separase2 at anaphase, and that this disengagement licences centriole duplication in the next cell cycle. We describe an in vitro system using Xenopus egg extract and purified centrioles in which both centriole disengagement and centriole growth occur. Centriole disengagement at anaphase is independent of mitotic exit and Cdk2/cyclin E activity, but requires the anaphase-promoting complex and separase. In contrast to disengagement, new centriole growth occurs in interphase, is dependent on Cdk2/cyclin E, and requires previously disengaged centrioles. This suggests that re-duplication of centrioles within a cell cycle is prevented by centriole engagement itself. We propose that the ‘once-only’ control of centrosome duplication is achieved by temporally separating licensing in anaphase from growth of new centrioles during S phase. The involvement of separase in both centriole disengagement and sister chromatid separation would prevent premature centriole disengagement before anaphase onset, which can lead to multipolar spindles and genomic instability3,4.