InteractiveFly: GeneBrief

Centrobin: Biological Overview | References

BIOLOGICAL OVERVIEW

During interphase in Drosophila neuroblasts, the Centrobin (Cnb)-positive daughter centriole retains pericentriolar material (PCM) and organizes an aster that is a key determinant of the orientation of cell division. This study shows that daughter centrioles depleted of CNB cannot fulfill this function whereas mother centrioles that carry ectopic CNB can. CNB co-precipitates with a set of centrosomal proteins that include gamma-Tub, Ana2, Cnn, Sas-4, Asl, DGRIP71, Polo and Sas-6. Following chemical inhibition of Polo or removal of three Polo phosphorylation sites present in Cnb, the interphase microtubule aster is lost. These results demonstrate that centriolar Cnb localization is both necessary and sufficient to enable centrioles to retain PCM and organize the interphase aster in Drosophila neuroblasts. They also reveal an interphase function for Polo in this process that seems to have co-opted part of the protein network involved in mitotic centrosome maturation (Januschke, 2013).

Since first described in germline stem cells, unequal mother-daughter centriole behaviour has become a major issue in stem cell self-renewing asymmetric division that has now been reported to occur in mice and in other cell types in Drosophila melanogaster including neuroblasts. Drosophila neuroblasts are neural precursors that generate the flys central nervous system through repeated rounds of self-renewing asymmetric divisions. Neuroblast asymmetric division is driven by the polarized localization of protein complexes at the apical and basal cortex and by the orientation of the mitotic spindle along the apico-basal axis. As a result, the apical and basal sides of the cortex are cleaved apart into the renewed neuroblast and the smaller differentiating ganglion mother cell (GMC), respectively. Dysfunction of several of the molecules involved in these processes results in tumour growth (Januschke, 2013).

Neuroblast cortical polarity and mitotic spindle alignment are tightly linked to the centrosome cycle of these cells in which centriole splitting is a very early interphase event and mother and daughter centrosomes exhibit significant differences in structure, function and fate (Januschke, 2011; Januschke, 2010; Rebollo, 2007; Rusan, 2007). In Drosophila neuroblasts, the mother centriole, which has no significant microtubule organizing activity and migrates extensively during interphase, is inherited by the GMC after mitosis (Conduit, 2010; Januschke, 2011; Rebollo, 2007; Rusan, 2007; Rebollo, 2009). In contrast, the daughter centriole, which retains high microtubule organizing activity and remains rather stable near the apical cell cortex throughout interphase, is retained by the neuroblast at mitosis. Thus, unlike most Drosophila cells where centrosomes are rather feeble microtubule organizing centres (MTOCs) during interphase, Drosophila neuroblasts posses a prominent interphase microtubule aster that is organized by the cortex-bound daughter centriole. Drosophila centrioles do not seem to have age-dependent ultrastructural attributes such as the satellites and appendages observed in vertebrates. However, Drosophila CNB, similarly to its homologues in mouse and human cells, binds only to the daughter centriole, revealing a molecular dimorphism that could contribute to daughter-centriole-specific behaviour in Drosophila larval neuroblasts (Januschke, 2013).

This study shows that in Drosophila neuroblasts, daughter centrioles without CNB behave like mother centrioles, and mother centrioles with ectopically localized CNB behave like daughter centrioles. Moreover, by co-immunoprecipitation and tandem mass-spectroscopy (MS/MS) this study shows that CNB co-precipitates with γ-Tub, Ana2, Cnn, Sas-4, Asl, Asl, DGRIP71, Polo and Sas-6. These centriolar and PCM proteins are part of a well-characterized network that drives PCM accumulation during mitotic centrosome maturation (see Gopalakrishnan, 2011). This study also shows that CNB phosphorylation by Polo is essential to maintain the interphase aster, hence revealing a requirement for Polo function during interphase (Januschke, 2013).

In neuroblasts expressing YFP-CNB-PACT (The PACT domain is a conserved centrosomal targeting motif in the coiled-coil proteins AKAP450 and pericentrin), both mother and daughter centrioles carry CNB and behave as daughter centrioles do in wild-type neuroblasts (see Centrobin function and graphical summary). Interestingly, ectopic activation of the interphase PCM retention program on both centrioles forces a canonical centrosome cycle in these cells whereby one centrosome-aster complex splits in two that segregate away from each other. In these cells, spindle assembly frequently occurs at an angle with respect to the apico-basal axis, but eventually rotates to align with it as in delaminating neuroblasts in the Drosophila embryo. In neuroblasts mutant for Cnb or expressing YPF-CNB^{T4A T9A S82A}, the two centrioles behave as mother centrioles do in wild-type neuroblasts; they retain little or no PCM during interphase, but centrosomes mature at mitosis entry. In all three cases the spindle aligns with the apico-basal axis. CNB depletion, similarly to CNB ectopic localization on the two centrioles, randomizes age-dependent centriole fate. Apico-basal orientation can change from one cycle to the next in CNB-mutant cells, but not in neuroblasts that express CNB-PACT, strongly suggesting that the orientation of cortical polarity is critically dependent on daughter centriole behaviour and that whether this is provided by one or two centrioles seems to be irrelevant as long as a cortex-bound microtubule network is maintained during interphase (Januschke, 2013).

The results show that interphase PCM retention in neuroblasts requires Polo to act on CNB-decorated centrioles to mediate Cnn and PCM recruitment. Which centriole and where exactly on the centriole CNB is positioned do not seem to be essential because ectopic localization of CNB by the PACT domain works just as well. Indeed, the interaction of CNB with proteins such as Ana2 and Sas6 that are essential to build the centriolar cartwheel, and with Cnn and Polo that are key regulators of PCM assembly, strongly suggests that CNB might serve its function by providing a critical bridge between centrioles and PCM. It also suggests that the neuroblast and daughter-centriole-specific process of PCM retention by the CNB-containing centriole has co-opted at least some of the proteins that drive PCM enrichment on centrosomes during mitosis in many cell types. Nevertheless, the fact that mitotic centrosome maturation does not require CNB and is hardly affected by 20 nM of the Polo kinase inhibitor BI2536 strongly suggests that important differences apply between interphase PCM retention in Drosophila neuroblasts and the ubiquitous process of mitotic centrosome maturation (Januschke, 2013).

Interphase PCM retention by CNB-containing centrioles also requires Pins, but seems to require neither Mud, nor the Khc-73/Dlg pathway (Januschke, 2010). Centriole migration to the cortex shortly after mitosis in neuroblasts mutant for pins suggests that Pins function is required at a later stage. Such a function is unlikely to require cortical Pins because CNB-decorated centrioles detached from the cortex by microtubule poisons efficiently retain PCM and, indeed, no evidence of Pins cortical localization during interphase has been reported in Drosophila larval neuroblasts. The seemingly normal mitotic centrosome maturation in >pins-mutant neuroblasts further substantiates that despite shared molecular factors, important differences apply between interphase PCM retention in Drosophila neuroblasts and mitotic centrosome maturation (Januschke, 2013).

Neither cell size asymmetry nor Mira cortical polarization seem perturbed by CNB depletion or by CNB ectopic localization in both centrioles. However, given the extraordinary complexity of cell fate determination, which is only partially understood, these observations cannot substantiate any solid conclusion on the effect that CNB dysfunction might have on cell fate determination and brain development. CNB is a ubiquitous protein that is one of the two centrioles in different cell types in Drosophila. This study also found that some CNB allelic combinations are lethal, which strongly suggests that CNB might have functions in other tissues. In vertebrates, Centrobin has been reported to bind to the tumour suppressor BRCA2 (Zou, 2005), and some common genetic variants of human Centrobin have been associated with breast cancer susceptibility (Wang, 2012; Januschke, 2013 and references therein).

A recent study has revealed that in human neuroblastoma cell lines, the sister cell that retains the NuMA crescent, which is thought to have more self-renewal potential, frequently inherits the daughter centrosome (Izumi, 2012). These results suggest the tantalizing possibility that the age-dependent centrosome segregation demonstrated in Drosophila neuroblasts might also occur in mammals and may have relevance in human disease (Januschke, 2013).

Drosophila neuroblasts retain the daughter centrosome

During asymmetric mitosis, both in male Drosophila germline stem cells and in mouse embryo neural progenitors, the mother centrosome is retained by the self-renewed cell; hence suggesting that mother centrosome inheritance might contribute to stemness. This hypothesis was tested in Drosophila neuroblasts (NBs) tracing photo converted centrioles and a daughter-centriole-specific marker generated by cloning the Drosophila homologue of human Centrobin. This study shows that upon asymmetric mitosis, the mother centrosome is inherited by the differentiating daughter cell. These results demonstrate maturation-dependent centrosome fate in Drosophila NBs and that the stemness properties of these cells are not linked to mother centrosome inheritance (Januschke, 2011).

The correlation between mother versus daughter centrosome segregation and asymmetric cell fate was first documented in Drosophila male germline stem cells (mGSCs), in which the mother centrosome is anchored near the niche and is retained by the stem cell, whereas the daughter centrosome migrates to the opposite side of the cell and is segregated into the differentiating gonial cell after mitosis (Yamashita, 2007). More recently, mother centrosome retention has also been demonstrated during asymmetric mitosis in radial glia progenitors in the developing mouse neocortex. In Drosophila mGSCs, the possible functional relevance of such stereotyped centrosome behaviour is still unclear, although tantalizing hypotheses have been put forward. In the developing mouse neocortex, centriole maturation has been shown to be required for maintaining radial glia progenitors (Januschke, 2011 and references therein).

Asymmetric centrosome behaviour has also been documented in Drosophila neuroblasts (NBs). NBs are stem-cell-like precursors that generate the flys central nervous system through a series of asymmetric divisions giving rise to a self-renewed NB and a differentiating ganglion mother cell (GMC), which typically divides only once into a pair of neurons or glia. NB asymmetric division is largely driven by cortical polarity. It starts by the apical accumulation of a number of protein complexes including the Par complex (Par-3, Par-6, aPKC) as well as Mushroom body defect, recruited by Partner of Inscuteable, which associates with Gαi, Inscuteable and Par-3. Apical complexes control the clustering at the basal cortex of cell fate determinants, such as Numb, Prospero and Brain Tumour, through their adaptor proteins Partner of Numb and Miranda. The apical cortex also controls the orientation of the mitotic spindle along the apico-basal axis, which is essential to position the cytokinesis furrow, such that cleavage segregates the apical and basal sides of the cortex into the newly formed NB and GMC, respectively. NB asymmetric division is also controlled by other auxiliary modulators (Januschke, 2011 and references therein).

Mosaic clones mutant for any of the known cell fate determinants overgrow in the larval brain and develop as malignant neoplasms upon transplantation into wild-type adult flies, showing that correct execution of the asymmetric division programme in Drosophila NBs is crucial to prevent tumour growth. Centrosomes are important in this process and larval brains that have cells without centrosomes or with more than two centrosomes per cell are highly prone to developing tumours (Januschke, 2011 and references therein).

NB asymmetry is not limited to cortical polarity at mitosis. Early in interphase, the NB centrosome splits in two, which display significant structural and functional differences. One of the resulting centrosomes retains most of the pericentriolar material (PCM) and remains at the apical cortex organizing the main microtubule network. The other centrosome has little, if any, PCM and microtubule organizing activity, and migrates extensively across the cytoplasm for most of the interphase. At mitosis, the apical centrosome is retained by the NB and the other centrosome is inherited by the GMC. Remarkably, at the time of splitting, the GMC-fated centrosome does contain a significant amount of PCM, as revealed by centrosomin fused to GFP which is shed off just before it starts to move. These centrosome asymmetries are thought to have a main role in the orientation of cortical polarity and mitotic spindle alignment (Januschke, 2011 and references therein).

Unlike centrioles in vertebrate cells where centriolar satellites, appendages and cartwheel span correlate with centriole maturation, no ultrastructural dimorphism has been reported between mother and daughter centrioles in Drosophila. Mother or daughter centriole-specific molecular markers are also lacking in Drosophila. Consequently, it is still unclear if the conspicuously unequal size, activity and fate of Drosophila NBs centrosomes correlate with centriole maturation. It is also unclear if at the time of splitting each centrosome contains a full diplosome or a single centriole, as is the case in Drosophila embryos and in some human cell lines (Januschke, 2011 and references therein).

This study shows by tracing photo-converted centrioles that centrosome splitting occurs before centriole duplication and that the centrosome that loses PCM is motile during interphase. This centrosome is also inherited by the differentiating GMC and contains the mother centriole. The cloning of the Drosophila homologue of Centrobin is also reported. In vertebrate cell lines, Centrobin specifically accumulates on the daughter centriole. It was found that at the time of centrosome splitting, a yellow fluorescent protein (YFP) fusion to CNB stays on the centrosome that remains at the apical cortex, which is eventually inherited by the NB, and is excluded from the centrosome that breaks away and is fated to the GMC. From these observations, it is concluded that centrosome maturation and fate are tightly correlated during asymmetric mitosis in Drosophila NBs, in which the cell that retains stemness inherits the daughter centriole. It is also concluded that in spite of the lack of ultrastructural dimorphism, molecular dimorphism exists between mother and daughter Drosophila centrioles (Januschke, 2011).

To investigate if mother-versus-daughter centriole differences determine centriole fate during NB asymmetric division, a direct approach was devised based on photo-conversion of a pancentriolar marker in the apical centrosome. There are two expected outcomes of such an experiment (see Photo-conversion of a centriolar marker in the apical centrosome). If at the time of photo-conversion, once the basal centrosome has moved away, the apical centrosome contains two centrioles, then, in the next cell cycle, both the apical and basal centrosomes will carry the converted signal). Maturation-dependent centrosome fate will become apparent in the following cycle by the consistent segregation of the converted signal, which traces the mother centriole to either the apical centrosome, which is retained in the NB, or the basal centrosome, which is fated to the GMC. Alternatively, if at the time of photo-conversion the apical centrosome contains only one centriole, the hypothetical maturation-dependent centrosome fate will be revealed in the next cell cycle when, upon centrosome splitting, only the apical or the basal centrosome will carry converted signal. In such case, the mother centriole will be systematically fated to either the NB or the GMC, respectively (Januschke, 2011).

A time-lapse series including the most representative time points of a centrosome photo-conversion experiment using the pancentriolar marker PACT-d2Eos is presented. After the basal centrosome has started to move away from the apical centrosome, a pulse of 405 nm light is shed on a manually defined rectangular region of interest which overlaps and slightly exceeds the size of the apical centrosome, causing the photo-conversion of the centriole marker from green to red. The converted signal remains stably associated with the apical centrosome, which is retained by the NB after the ensuing mitosis, and migrates to the apical cortex in early interphase. When the basal centrosome breaks away and starts its characteristic migration through the cytoplasm, the photo-converted signal, which traces the mother centriole, goes with it and is subsequently delivered into the GMC at mitosis. Identical results were obtained in a total of 20 photo-conversion experiments (Januschke, 2011).

Three main conclusions can be derived from these observations. First, at the time when apical and basal centrosomes split in Drosophila NBs, the non-motile apical centrosome contains only one centriole. Second, segregation of mother and daughter centrosomes is highly correlated with asymmetric cell fate. Finally, these results show that NBs retain the daughter rather than the mother centriole (Januschke, 2011).

To independently corroborate these conclusions, it was decided to circumvent the current lack of molecular markers by cloning the Drosophila homologue of the human daughter-centriole-specific protein Centrobin (Jeong, 2007; Zou, 2005). By homology search, open reading frame CG5690, predicted to encode a 78.7 kDa protein of 689 aa (Flybase), was identified as the putative Drosophila centrobin (cnb) gene. Transgenic flies were generated constitutively co-expressing YFP-CNB, and a fusion between mKATE and the pancentriolar marker ASL47, and the localization of both fluorescent probes was followed in larval NBs. Although, as reported for other centriolar proteins, one or two randomly located YFP-CNB aggregates are often found in these cells, a clear pattern of YFP-CNB signal segregation emerges. Centrosome-bound YFP-CNB can be observed without interruption from telophase, up to the time of centrosome splitting, early in the following interphase. Then, when apical and basal centrosomes break apart, YFP-CNB remains bound to the apical non-motile centrosome, and cannot be detected over the migrating basal centrosome. Identical results were obtained in a total of 16 NBs (Januschke, 2011).

Retention of the daughter-centriole marker YFP-CNB by the apical centrosome and the lack of YFP-CNB signal in the motile centrosome corroborates further the conclusions derived from the photo-conversion experiments. In addition, the age-dependent centriole-specific localization of YFP-CNB provides the first instance of molecular asymmetry between mother and daughter Drosophila centrioles (Januschke, 2011).

Tracing mother and daughter centrioles with photo-converted PACT-d2Eos and YFP-CNB, respectively, it was shown that apical and basal Drosophila NB centrosomes do not have fully formed diplosomes when they split, and that mother-versus-daughter centriole segregation tightly correlates with cell fate; contrary to what has been generally assumed, the daughter centrosome is fated to be retained by the NB and the mother centrosome is fated to the GMC (Januschke, 2011).

The functional relevance of specific centrosome retention by the stem cell in Drosophila remains to be ascertained. Both in mGSCs and in NBs, a microtubule-dependent mechanism hooks one centrosome to the region of the cortex, proximal to the hub in the germ line and apical in NBs38; that is, retained by the stem cell after mitosis. Preferential retention of the mother centrosome in the germ line might therefore simply be the inescapable consequence of carrying more PCM and having greater microtubule-organizing activity. This is less likely in NBs where the daughter centriole is retained, while the mother centriole breaks away carrying little, if any, PCM. That implies the removal of PCM from around the mother centriole and the retention of PCM by the daughter centriole. The shedding of PCM by the GMC-fated centrosome as it splits from the apical centrosome has been previously documented (Januschke, 2011).

Precedence for centrosome transmitted cell fate information exist in other species, but not in Drosophila. There are, however, published data on the effect of switching centrosome fate by transient microtubule depolymerization. Upon disassembly of the interphase aster of the NB by treatment with microtubule poisons, the apical centrosome loses connection with the cortex and moves deeper inside the cell so that at mitosis onset, the position of both centrosomes is essentially randomized. Upon recovery of microtubule dynamics, mitosis resumes and asymmetric cell division takes place. In some cases, centrosome fate is switched so that the centrosome originally destined to the GMC is retained by the NB, and vice versa, yet mitosis generates an NB and a GMC that are morphologically normal, strongly suggesting that the identity of the resulting NB and GMC is not severely compromised. It is unknown, however, whether such switch in centriole fate might have long-term developmental consequences (Januschke, 2011).

Centriole-maturation-dependent segregation in Drosophila NBs is somewhat reminiscent of spindle pole body (SPB) segregation in budding yeast. Similar to centrioles in animals, the SPB duplicates once per cell cycle. The new SPB remains connected to the old one by the bridge, which is cleaved in S phase. The two SPBs then separate to form the opposite poles of the spindle. In unperturbed cells, the old SPB always migrates into the bud. Old and new SPBs are also functionally and biochemically distinct. For instance, the Bfa1p-Bub2p GTPase-activating protein complex, which is an integral part of the spindle position checkpoint, specifically binds to the old SPB. The SPC inhibits the mitotic exit network) until the nucleus has migrated into the bud. Also the antigen-presenting cell-related molecule Kar9 localizes only to the old SPB due to the activity of the new SPB-resident Clb4/Cdc28 kinase. Interestingly, SPB fate can also be artificially switched, in which case, Bfa1p still localizes to the SPB that enters the bud independently of whether it is the old or the new. These data show that in budding yeast biochemical differences between the old and new SPBs are functionally relevant (Januschke, 2011 and references therein).

As in yeast where the old SPB is inherited by the long-lived bud, the mother centriole in Drosophila is inherited in testes by the mGSC, which is the most long-lived of the two mGSC daughters, and in larval brains by the GMC whose daughters outlive the NB. Thus, what appears to be a diametrically different behaviour between mGSCs and NBs using stemness as criteria is consistent, when the expected lifespan of each cell type is considered (Januschke, 2011).

The centriolar protein Bld10/Cep135 is required to establish centrosome asymmetry in Drosophila neuroblasts

Centrosome asymmetry has been implicated in stem cell fate maintenance in both flies and vertebrates and Drosophila neuroblasts, the neural precursors of the fly's central nervous system, contain molecularly and physically asymmetric centrosomes, established through differences in pericentriolar matrix (PCM) retention. For instance, the daughter centriole maintains PCM and thus microtubule-organizing center (MTOC) activity through Polo-mediated phosphorylation of Centrobin (Cnb). The mother centriole, however, quickly downregulates PCM and moves away from the apical cortex, randomly migrating through the cytoplasm until maturation sets in at prophase. How PCM downregulation is molecularly controlled is currently unknown, but it involves Pericentrin (PCNT)-like protein (plp) to prevent premature Polo localization and thus MTOC activity. This study reports that the centriolar protein bld10, the fly ortholog of Cep135, is required to establish centrosome asymmetry in Drosophila neuroblasts through shedding of Polo from the mother centrosome. bld10 mutants fail to downregulate Polo and PCM, generating two active, improperly positioned MTOCs. Failure to shed Polo and PCM causes spindle alignment and centrosome segregation defects, resulting in neuroblasts incorrectly retaining the older mother centrosome. Since Cep135 is implicated in primary microcephaly, it is speculated that perturbed centrosome asymmetry could contribute to this rare neurodevelopmental disease (Singh, 2014).

In a gene candidate approach to identify molecules required for centrosome asymmetry in Drosophila neuroblasts, this study identified bld10/Cep135 as a potential centrosome dematuration regulator. bld10 is a ubiquitous centriolar protein, localizing to centrioles in Drosophila larval neuroblasts and other cell types. To investigate centrosome asymmetry, live imaging experiments were performed in intact third-instar larval brains, labeling centrosomes with the centriolar markers DSas6::GFP or DSas4::GFP and mCherry::Jupiter. jupiter encodes for a microtubule binding protein, sharing properties with several structural microtubule-associated proteins (MAPs), and is ideally suited to visualize microtubule dynamics and microtubule-organizing center (MTOC) activity. In agreement with previous findings, it was found that wild-type (WT) interphase neuroblasts contained one apical MTOC only. The second MTOC appeared during prophase in close proximity to the basal cortex. By prometaphase, both MTOCs reached maximal activity and intensity. However, in bld10 mutant interphase neuroblasts (bld10^c04199/Df(3L)Brd15, two centrosomes of similar size and MTOC activity were observed close together on the apical cortex. The two centrosomes progressively separated from each other until they reached their respective positions on the apical and basal cortex by prometaphase. Thus, in contrast to the wild-type, bld10 mutant neuroblasts show symmetric centrosome behavior. bld10's centrosome asymmetry defect could be rescued with bld10::GFP, and immunohistochemistry experiments confirmed the live imaging results (Singh, 2014).

The bld10^c04199 allele is predicted to produce a truncated protein, retaining bld10's N terminus. A new N-terminal deletion allele (bld10ΔN was generated that showed the same centrosome asymmetry phenotype. In addition, neuroblasts were found containing monopolar and multipolar spindles, which are not observed with the bld10^c04199 allele. This suggests that bld10^c04199 is a separation-of-function allele, specifically disrupting centrosome asymmetry. Unless otherwise noted, all of the experiments described in the following sections were performed with the bld10^c04199 allele (Singh, 2014).

The lack of centrosome asymmetry in bld10 mutant neuroblasts could be due to aberrant centriole migration. For example, the mother centriole could either fail to migrate through the cytoplasm or migrates back to the apical cortex to mature. This hypothesis was tested, measuring centriole migration as a function of time, and it was observed that centriolar migration in wild-type and bld10 mutant neuroblasts occur in two distinct phases: (1) centrioles steadily separated from each other, followed by (2) a sudden increase in intercentriolar distance, which peaked when centrioles reached a separation distance of ~4-6 μm in the wild-type and bld10 mutants. Centrioles in bld10 mutants did not require more time to reach this threshold distance and did not return to the apical cortex to mature. It is concluded that bld10's centrosome asymmetry defect is not due to aberrant centriole migration (Singh, 2014).

To get mechanistic insight into the bld10 phenotype, live imaging was used to measure the dynamic localization of three GFP-tagged pericentriolar matrix (PCM) markers: γ-tubulin (γ-Tub), Mini spindles (Msps; CKAP5 in vertebrates) and centrosomin (Cnn; CDK5Rap2 in vertebrates). Wild-type neuroblasts showed robust localization of γ-Tub, Msps, and Cnn to the apical centrosome during interphase. After centrosome splitting, all three PCM markers were downregulated from the basal centrosome (shedding phase) but reaccumulated during prophase (maturation phase. bld10 mutant neuroblasts also correctly localized γ-Tub, Msps, and Cnn to the apical interphase centrosome. However, similar to the MTOC marker Jupiter, γ-Tub, Msps, and Cnn were not downregulated from the separating centriole. Centrosome size was measured and a centrosome asymmetry index was plotted, starting at centrosome splitting until metaphase. Wild-type centrosomes developed a clear size asymmetry during the shedding phase and reduced it during the maturation phase. bld10 mutant centrosomes stayed similar in size, manifested in an asymmetry index below 1.5. Centrosome size and intensity measurements also revealed that in most wild-type neuroblasts, γ-Tub, Msps, and Cnn were removed from the basal centrosome ~15 min after centrosome splitting. Basal wild-type centrosomes were essentially devoid of γ-Tub after that time, whereas bld10 mutants contained equal amounts of this PCM marker. Changes in centrosome size were further compared, and it was found that wild-type apical centrosomes predominantly grew, whereas basal centrosomes increased (maturation phase) and decreased (shedding phase) their size to almost the same extent. bld10 mutant centrosomes were able to enlarge but showed very little size reduction, comparable to apical wild-type centrosomes. It is concluded that bld10 mutant centrosomes are able to mature but fail to downregulate the PCM markers γ-Tub, Msps, and Cnn (Singh, 2014).

The results suggest two possible mechanisms for centrosome asymmetry: (1) bld10 could prevent premature mother centrosome maturation by blocking the precocious accumulation of PCM proteins. (2) Alternatively, bld10 could promote PCM shedding right after centrosomes separate, thereby preventing the basal mother centrosome to prematurely become an MTOC. An in vivo pulse-chase labeling experiment was devised to distinguish between these two possibilities. To this end, Cnn at its endogenous locus was tagged with the photoconvertable fluorescent protein mDendra2. If bld10 blocks premature PCM accumulation, mother centrioles should quickly shed photoconverted Cnn and prematurely reaccumulate unconverted Cnn in bld10 mutants. Vice versa, if PCM shedding is compromised, it should be possible to follow the photoconverted centrosomes from the moment they separate until telophase. It was found that apical wild-type daughter centrioles retained the majority of photoconverted Cnn::mDendra2 from early interphase until prophase (possibly longer), indicating that very little Cnn protein gets exchanged. The basal mother centriole, however, lost photoconverted Cnn::mDendra2 within approximately 10-15 min after centriole separation, confirming that Cnn is shed quickly. Interestingly, bld10 centrioles retained photoconverted Cnn::mDendra2 for at least 45 min after separation. In some cases, one of the centrioles decorated with photoconverted Cnn::mDendra2 was even inherited by the newly formed GMC. It is conclude that (1) on the apical centrosome, Cnn protein turnover is absent or significantly reduced during interphase, that (2) on the basal centrosome, Cnn is shed quickly and replaced with new Cnn when maturation sets in, and that (3) bld10's centrosome asymmetry defect is not due to premature centrosome maturation. Instead, separating basal centrioles fail to shed Cnn in particular and possibly PCM proteins in general (Singh, 2014).

To elucidate the molecular mechanism underlying PCM shedding, the relationship was analyzed between bld10, Centrobin (Cnb), and Pericentrin (PCNT)-like protein (plp). Recently, it was shown that Cnb is necessary and sufficient for PCM retention on the apical daughter neuroblast centrosome. However, gain- and loss-of-function experiments with Cnb did not perturb bld10's localization. Similarly, as in the wild-type, Cnb was localized asymmetrically in bld10 mutants. plp mutants fail to downregulate γ-Tub on the mother centrosome. In plp mutant neuroblasts, the basal centrosome retained Cnn and MTOC activity during interphase. Interestingly, photoconversion experiments showed that similar to bld10, Cnn shedding from the basal centrosome was compromised in plp mutants. However, plp localization is not perturbed in bld10 mutant neuroblasts, and bld10 was normally localized in plp mutants. Knockdown of plp in bld10 mutants did not enhance bld10 PCM shedding phenotype, but due to the occurrence of additional phenotypes (fragmented or multiple centrosomes), the shedding phenotype could also be partially masked. In sum, it is concluded that bld10 is regulating centrosome asymmetry independently of Cnb and that plp is also required to shed Cnn (Singh, 2014).

Since the mitotic kinase Polo has been implicated in PCM retention during interphase, Polo localization dynamics were analyzed in wild-type and bld10 mutant neuroblasts. Recently, it was reported that Polo localizes to the apical centrosome during interphase and is only detectable at the basal centrosome during prophase, when maturation sets in. A Polo::GFP protein trap line was used and it was confirmed that Polo is stably localized to the apical interphase centrosome. Surprisingly, weak Polo was also found on the separating mother centrosome. Subsequently, Polo disappeared from the basal mother centriole within 10 min, comparable to Cnn, γ-Tub, and Msps shedding times. With a genomic Polo::GFP transgene, showing lower fluorescence intensity, Polo was found to be localized on both centrosomes in bld10 mutants. These data suggest that in wild-type neuroblasts, Polo is not just recruited onto the basal mother centrosome by prophase as previously reported, but is also subject to shedding during interphase. Polo is required for PCM retention since in bld10 mutant neuroblasts treated with the Polo inhibitor BI2536, both centrosomes lose MTOC activity. Thus, it is concluded that shedding of Polo is a requirement for the subsequent shedding of Cnn, γ-Tub, and Msps, enabling basal mother centrosome dematuration and the establishment of centrosome asymmetry (Singh, 2014).

Finally, the consequences were analyzed of disrupted centrosome asymmetry. The daughter centriole was labeled with Cnb::YFP ] and centrosome segregation was assayed. It was confirmed that wild-type neuroblasts faithfully retain the Cnb* daughter centriole, whereas the Cnb- centrosome segregates into the GMC. bld10 mutants showed correct asymmetric Cnb localization, but ~45% of bld10 mutant neuroblasts wrongly retained the mother centriole and segregated the daughter centriole into the GMC. Cnb+ centrosomes are usually bigger in wild-type and bld10 mutant neuroblasts, but centrosome segregation is independent of MTOC activity and size since bld10 mutant neuroblasts often retained the smaller centrosome. It is concluded that centrosome asymmetry is required for faithful centrosome segregation (Singh, 2014).

Since bld10 mutant neuroblasts have mispositioned MTOCs in relation to the apical-basal division axis, spindle orientation was analyzed. Immunohistochemistry experiments showed that bld10 mutant neuroblast spindles deviate from the regular orientation range, with isolated cases of extreme misalignment. Metaphase spindles, aligned orthogonally to the apical-basal polarity axis, can induce symmetric neuroblast divisions, resulting in an increase of the neural stem cell pool. However, neuroblast numbers were unchanged in bld10 mutants compared to control brains, and symmetric neuroblast divisions were not found with live imaging. Instead, time-lapse experiments showed that bld10 mutant centrosomes prematurely formed misaligned bipolar spindles. Spindle rotation during metaphase corrected this misalignment. Apical-basal polarity is a prerequisite for correct spindle orientation, but apical and basal polarity markers localized normally in bld10 mutants. Similarly, the spindle orientation regulators, Partner of Inscuteable (Pins; LGN/AGS3 in vertebrates) and the NuMA ortholog Mud, were also correctly localized. It is concluded that controlled PCM shedding and maturation is required for correct centrosome positioning but backup mechanisms exist, correcting for misaligned metaphase spindles (Singh, 2014).

Many cell types, including stem and progenitor cells, contain asymmetric centrosomes and segregate them nonrandomly, suggesting a connection between centrosome asymmetry and cell fate. How centrosome asymmetry is regulated is currently not understood, but centrosome dematuration is a critical step in establishing centrosome asymmetry. It was found that the centriolar protein bld10/Cep135, known as a centriole duplication and elongation factor, is required to establish centrosome asymmetry. On the basis of these data, it is proposed that plp and bld10 induce Polo's removal from the mother centriole, triggering the shedding of PCM proteins such as Cnn, γ-Tub, and Msps. Polo has been reported to be closely associated with centrioles, ideally positioned to phosphorylate both centriolar and PCM proteins. Thus, it is proposed that Polo-mediated phosphorylation of PCM proteins maintains a stable interaction between the centriole and surrounding PCM (Singh, 2014).

How Polo localization is regulated is currently not known, and no direct molecular interaction was detected between bld10 and Polo. Although no centriolar markers were found to be mislocalized inbld10 mutants (at the resolution level of confocal microscopy), it is possible that structural centriole defects, as detected in bld10 mutant spermatocytes and wing disc cells, could affect PCM turnover rates. However, since bld10 is not asymmetrically localized, it is difficult to conceive how such defects specifically compromise the behavior of the mother but not the daughter centrosome (Singh, 2014).

Although perturbed centrosome asymmetry does not seem to undermine neuroblast polarity, the cell cycle, or physical and molecular asymmetric cell division, the possibility cannot be excluded that centrosome asymmetry could have long-term consequences currently beyond the ability to detect. Interestingly, defects in centrosome maturation or mutations in Cep135 can cause neurodevelopmental disorders such as primary microcephaly. It will be interesting to address the question whether lack of Cep135 is causing microcephaly due to compromised centrosome asymmetry and dematuration (Singh, 2014).

Centrobin is essential for C-tubule assembly and flagellum development in Drosophila melanogaster spermatogenesis

Centrobin homologues identified in different species localize on daughter centrioles. In Drosophila melanogaster sensory neurons, Centrobin (referred to as CNB in Drosophila) inhibits basal body function. These data open the question of CNB's role in spermatocytes, where daughter and mother centrioles become basal bodies. This study reports that in these cells, CNB localizes equally to mother and daughter centrioles and is essential for C-tubules to attain the right position and remain attached to B-tubules as well as for centrioles to grow in length. CNB appears to be dispensable for meiosis, but flagellum development is severely compromised in Cnb mutant males. Remarkably, three N-terminal POLO phosphorylation sites that are critical for CNB function in neuroblasts are dispensable for spermatogenesis. These results underpin the multifunctional nature of CNB that plays different roles in different cell types in Drosophila, and they identify CNB as an essential component for C-tubule assembly and flagellum development in Drosophila spermatogenesis (Reina, 2018).

Centrobin was initially identified in humans through a yeast two-hybrid screen for proteins that interact with the tumor suppressor BRCA2. Like human Centrobin (referred to as CNTROB in humans), its homologues in other species are components of daughter centrioles in different cell types. In mammals, CNTROB has been reported to be required for centriole duplication and elongation as well as for microtubule nucleation and stability through its binding to tubulin and its effect in stabilizing centrosomal P4.1-associated protein (CPAP) (Reina, 2018 and references therein).

Drosophila melanogaster Centrobin is a key determinant of mother/daughter centriole functional asymmetry in larval neuroblasts and type I sensory neurons. In neuroblasts, CNB functions as a positive regulator of microtubule organizing center (MTOC) activity during interphase: its depletion impedes daughter centrioles to assemble an MTOC, whereas mother centrioles carrying ectopic CNB become active MTOCs. This activity is dependent on CNB phosphorylation by PLK1/POLO on three conserved sites and involves the function of pericentrin-like protein (PLP) and BLD10/CEP135. In Drosophila type I sensory neurons, CNB functions as a negative regulator of ciliogenesis: CNB depletion enables daughter centrioles as functional basal bodies that template ectopic cilia, and mother centrioles modified to carry CNB cannot function as basal bodies (Reina, 2018).

The function of a daughter centriole protein like CNB as a negative regulator of ciliogenesis opens the question of CNB's role in Drosophila primary spermatocytes where after centrosome duplication all four centrioles, mothers and daughters alike become basal bodies that assemble axoneme-based cilium-like structures, which are the precursors of sperm flagella. To investigate this issue, this study examined CNB localization and function in spermatogenesis (Reina, 2018).

As in other cell types in Drosophila and other species, CNB localization in spermatogonial cells was asymmetric, restricted to only one of the two centrioles that were labeled with anti-SAS4 antibodies. However, after the last (fourth) round of spermatogonial mitosis, the resulting early primary spermatocytes presented CNB equally distributed on the two centrioles of each pair (Reina, 2018).

After centrosome duplication, all four centrioles of each spermatocyte grow significantly in length, migrate to the cell surface, and become basal bodies that template a short axoneme that forms a protrusion covered by the cell membrane. Recent studies refer to this protrusion as the cilium-like region (CLR) or the ciliary cap (Reina, 2018).

To determine CNB localization in the basal body-CLR complex, CNB was immunostained together with PLP, ANA1, and UNC. PLP signal was strongest on the proximal end of the basal body. ANA1 labelled the entire length of the basal body. UNC overlapped with ANA1 except on the PLP-positive proximal end, was strongest at the transition zone, and extended further distally into the axoneme. CNB signal was equal in all basal bodies; most concentrated on the proximal end of the ANA1 domain, largely overlapping with PLP. Like PLP, a very weak CNB signal could also be detected all along the basal body. These observations reveal some remarkable differences regarding CNB localization, which in turn suggest important functional differences as far as ciliogenesis is concerned between early spermatocytes and sensory neurons in Drosophila (Reina, 2018).

To determine whether CNB has a function in the basal body-CLR complex, trans-heterozygous PBac(RB)Cnb[e00267]/Df(3L)ED4284 (henceforth referred to as Cnb mutant) spermatocytes were examined. The length of YFP-Asterless (ASL) signal (a proxy for centriole length) was significantly shorter in Cnb mutant than in WT males. Longitudinal EM sections from centriole pairs in which both mother and daughter were cut along their entire length confirmed this observation and revealed that daughter basal bodies were shorter than their mothers in Cnb mutant spermatocytes. This was not the case in WT spermatocytes, where it was found that mother and daughter basal bodies were of equal length. EM sections also revealed that 10 out of 16 mutant axonemes sectioned longitudinally were irregularly shaped and had abnormal axonemes with doublets that were aberrantly interrupted halfway through the CLR (Reina, 2018).

Ultrastructural details were revealed by transversal EM sections. Unlike basal bodies in sensory neurons that are made of doublets, basal bodies in WT spermatocytes are made of triplets. At the transition zone, the outermost tubule (C-tubule) progressively lost protofilaments, hence becoming an open longitudinal sheet that extended beyond the transition zone along the axoneme. At the point where C-tubule remnants and B-tubules intersect, short radial projections could be observed (Reina, 2018).

In addition to this basic layout of triplets, doublets, and open sheaths, about half of basal body-CLR complexes in Drosophila WT spermatocytes present a luminal singlet microtubule independently of the Z position of the cross-section. The presence of this luminal singlet is as erratic in Cnb mutant as it is in WT males and does not represent a Cnb mutant phenotype (Reina, 2018).

Eleven basal body-CLR complexes were serially sectioned from two Cnb mutant males. In all 11 samples, C-tubules were either misplaced outwards, away from the nearly straight line defined by the A/B doublet, or lost, leaving only duplets behind. These abnormalities extended distally along the axoneme, where C-tubule-derived longitudinal sheets were often lacking in Cnb mutant spermatocytes. The short radial projections observed in WT axonemes also appeared to be lacking in Cnb mutant CLRs, but this observation must be interpreted with caution because of the technical difficulties of detecting such small structures that are located in regions of electron-dense material (Reina, 2018).

These observations reveal that CNB is required in Drosophila spermatocytes for centrioles to properly position and stabilize C-tubules, grow in length up to full-sized basal bodies, and template normal axonemes. The results identify CNB as one of the very few proteins known so far to be required for C-tubule assembly, including Δ-tubulin in Paramecium tetraurelia and Chlamydomonas reinhardtii along with UNC and SAS4 in Drosophila. Notably, as in Cnb mutant spermatocytes, basal bodies are also shorter than normal in Sas4F112A and Unc mutant spermatocytes, hence suggesting that short centrioles and lacking or mispositioned C-tubules may be causally linked (Reina, 2018).

Drosophila SAS4 and its human homologue CPAP are closely functionally related to the corresponding Centrobin homologues CNB/CNTRB. Mutants in the N-terminal helical motif of the PN2-3 domain of CPAP and SAS4 cause overelongation of newly formed centriolar/ciliary microtubules, and so does CNTRB overexpression by stabilizing and driving the centriolar localization of CPAP. Moreover, the mutants in PN2-3 domain of CPAP referred to above result in biciliated cells, and so does Cnb loss of function in Drosophila sensory neurons. In Drosophila, SAS4 and CNB coimmunoprecipitate from embryo extracts. These data strongly suggest that the role of CNB on basal body length may be SAS4 dependent (Reina, 2018).

The ultrastructural phenotypes that were observed in basal body-CLR complexes in Cnb mutant spermatocytes persist through meiosis and are present in early postmeiotic cells. However, no observable defects were observable in meiosis. Cnb mutant males assembled fairly normal meiotic spindles and generated normal cysts of postmeiotic onion-stage spermatids that presented a uniform nuclear size, which is a very sensitive readout of faithful meiotic chromosome segregation. These results strongly suggest that shortened centrioles and lacking C-tubules have no major observable consequences on meiosis progression (Reina, 2018).

However, soon after meiosis, when the basal body attaches to the nucleus and the axoneme starts to grow, pushing the old CLR caudally, Cnb mutant spermatids presented CLRs that were much longer than those from WT cells. In WT cells at this stage, CNB signal was strongest on the basal body proximal to UNC and was also detectable over the perinuclear plasm, a cap over the nuclear hemisphere where the basal body is engaged (Reina, 2018).

Cnb mutant phenotypes were also conspicuous at later stages of elongation. Transversal sections through the tails of WT elongating spermatids showed two mitochondrial derivatives flanking the axoneme, which was almost completely surrounded by a double membrane. Among the sample of 32 Cnb mutant elongating spermatids that were analyzed in detail, a wide range was found of phenotypic variability. Five presented a fairly WT 9 + 2 configuration, whereas the other 27 included cells that had no axoneme or axonemes with missing or highly disarrayed duplets and in which the surrounding membrane was widely open. Among the latter, six had some duplets that carried C remnants, and 16 did not. Interestingly, not all duplet-bound open sheaths observed in Cnb mutant cells were C remnants. Four axonemes were found that presented one to three triplets with an attached open sheath. No such structures have been reported in WT cells. The presence of a fraction of Cnb mutant spermatids with axonemes that presented C remnants suggests that a certain fraction of Cnb mutant primary spermatocytes retain C-tubules (Reina, 2018).

In addition to lacking and disarranged axonemes, Cnb mutant elongating spermatids also presented ectopic nuclei at sections where only tails were present in WT cysts. Scattered nuclei have also been reported in other Drosophila mutants like Unc and Sas4. Head-to-tail attachment defects have been documented in the sperm of a rat CNTROB mutant strain (hd rats) that was male sterile (Reina, 2018).

The results reveal that CNB depletion brings about a spectrum of mutant phenotypes at different stages of spermatogenesis that includes shorter centrioles, mispositioned or absent C-tubules, lack of C-tubule derivatives, abnormal or absent axonemes, and scattered nuclei along elongating spermatid bundles. It cannot be discarded that CNB may have independent functions accounting for each of these phenotypes. However, the apparent pleiotropic effect of CNB loss may simply reflect the downstream consequences of centrioles that are shorter than normal and present C-tubule defects (Reina, 2018).

The different, and in some aspects opposite, roles of CNB in basal body structure and ciliogenesis in type I sensory neurons and spermiogenesis are remarkable. In neurons, CNB inhibits basal body function on the centriole to which it binds that in WT cells is only the daughter centriole. In primary spermatocytes, however, CNB localizes to both mother and daughter centrioles, which in these cells are equally able basal bodies, and acts as a positive regulator of ciliogenesis. One conspicuous difference between these two cell types is that basal bodies in spermatocytes contain C-tubules, whose assembly, positioning, and stability requires CNB. In vertebrates, where centrioles also have C-tubules, a recent study has found that in addition to its primary localization to daughter centrioles, CNTROB associates with mother centrioles at the onset of ciliogenesis, and CNTROB loss causes defective axonemal extension. These results suggest that the different effect that CNB depletion has on ciliogenesis in Drosophila neurons and spermatocytes may reflect the doublet versus triplet structure of the corresponding basal bodies (Reina, 2018).

Dynamic centriolar localization of Polo and Centrobin in early mitosis primes centrosome asymmetry

Centrosomes, the main microtubule organizing centers (MTOCs) of metazoan cells, contain an older "mother" and a younger "daughter" centriole. Stem cells either inherit the mother or daughter-centriole-containing centrosome, providing a possible mechanism for biased delivery of cell fate determinants. However, the mechanisms regulating centrosome asymmetry and biased centrosome segregation are unclear. Using 3D-structured illumination microscopy (3D-SIM) and live-cell imaging, this study shows in fly neural stem cells (neuroblasts) that the mitotic kinase Polo and its centriolar protein substrate Centrobin (Cnb) accumulate on the daughter centriole during mitosis, thereby generating molecularly distinct mother and daughter centrioles before interphase. Cnb's asymmetric localization, potentially involving a direct relocalization mechanism, is regulated by Polo-mediated phosphorylation, whereas Polo's daughter centriole enrichment requires both Wdr62 and Cnb. Based on optogenetic protein mislocalization experiments, it is proposed that the establishment of centriole asymmetry in mitosis primes biased interphase MTOC activity, necessary for correct spindle orientation (Gallaud, 2020).

Centrosome asymmetry has previously been described to occur in asymmetrically dividing Drosophila neural stem cells (neuroblasts), manifested in biased interphase MTOC activity and asymmetric localization of the centrosomal proteins Cnb, Plp, and Polo and PCM proteins like Centrosomin. This study has shown that neuroblast centrosomes become intrinsically asymmetric by dynamically enriching centriolar proteins such as Cnb and Polo on the young daughter centriole during mitosis. This establishment of centriolar asymmetry is tightly linked to centriole-to-centrosome also called mitotic centriole conversion. In early prophase, Cnb and Polo colocalize on the existing mother centriole of the apical centrosome but from late prometaphase onward, Cnb is exclusively and Polo predominantly localized on the daughter centriole. Mechanistically, these dynamic localization changes could entail a direct or indirect translocation of Cnb and Polo from the mother to the daughter centriole. This model is partially supported for Cnb with FRAP data. Interestingly, Cnb behaves differently on the basal centrosome: the existing mother centriole does not contain any Cnb, appearing only on the forming daughter centriole in late prophase. This suggests a direct recruitment mechanism, which could also apply to the apical centrosome from anaphase onward. 3D-SIM, FRAP, and live-cell imaging data combined are most consistent with a model proposing that on the apical centrosome, a small pool of Cnb transfers from the mother to the daughter centriole during early mitosis. From anaphase onward, and from late prophase onward on the basal daughter centriole, Cnb levels increase through the recruitment of Cnb that was not previously associated with the mother centriole (Gallaud, 2020).

Cnb is phosphorylated by the mitotic kinase Polo and Polo-dependent phosphorylation of Cnb is necessary for its timely localization during mitosis. Interestingly, the data further suggest that Polo, which also becomes enriched on the daughter centriole during mitosis, is co-dependent with Cnb, while also requiring Wdr62. Polo's involvement in mitotic centriole conversion further suggests that the same molecular machinery cooperatively converts a maturing centriole into a centrosome for the next cell cycle while simultaneously providing it with its unique molecular identity (Gallaud, 2020).

The mechanisms generating 2 molecularly distinct centrioles during mitosis seem to directly influence the centrosome's MTOC activity in interphase; the 'Cnb+, high Polo' daughter centriole will retain MTOC activity during interphase whereas the 'Cnb-, low Polo' mother centriole, separates from its daughter in early interphase and becomes inactive. This model is in agreement with bld10 (Cep135) or plp mutants, which fail to down-regulate Polo from the mother centriole, resulting in the formation of 2 active interphase MTOCs. This is further supported by mislocalization data, showing that optogenetic manipulation of Polo and Cnb asymmetry specifically during mitosis impacts MTOC activity in the subsequent interphase. However, it cannot be excluded that MTOC asymmetry is also controlled independently of mitotic centrosome asymmetry establishment because optogenetic interphase manipulations of Polo and Cnb alone can also perturb biased MTOC activity (Gallaud, 2020).

Loss of Wdr62 or Cnb also affects asymmetric centriolar Polo localization. Yet, interphase centrosomes lose their activity in these mutants. wdr62 mutants and cnb RNAi neuroblasts both show low Polo levels in interphase. It is thus hypothesized that in addition to an asymmetric distribution, Polo levels must remain at a certain level to maintain interphase MTOC activity; high symmetric Polo results in 2 active interphase MTOCs, whereas low symmetric Polo results in the formation of 2 inactive centrosomes. Indeed, the optogenetic experiment triggered an increase in centriolar Polo levels upon blue-light induction, suggesting that both Polo levels and distribution influence MTOC activity. This hypothesis is strengthened by Cnn's capacity to oligomerize and form a scaffold, supporting PCM assembly upon phosphorylation by Polo. The more Polo is recruited, the more stable is the Cnn scaffold, supporting MTOC activity (Gallaud, 2020).

Centrobin is also enriched on the daughter centriole in mammals and preferentially becomes incorporated into the newly assembled daughter centriole in late G1 or early S phase. Centrobin remains localized at the daughter centrioles throughout the cell cycle. It is of interest to note that in mammalian cells, Centrobin becomes enriched on daughter centrioles during the G1-S transition and not during centriole-to-centrosome conversion. It is tempting to speculate that other kinases and mechanism regulate this translocation (Gallaud, 2020).

It was also observed that Cnb and Plp localize in a mutually exclusive manner, with Cnb localizing to the daughter centriole and Plp remaining on the mother centriole. In mammalian cells, similar mutual exclusion between the centriolar proteins Cep120 or Neurl4 and PCM proteins has been observed (Gallaud, 2020).

Taken together, the results reported in this study are consistent with a model, proposing that the establishment of 2 molecularly distinct centrioles is primed during mitosis and contributes to biased MTOC activity in the subsequent interphase. Wild-type neuroblasts unequally distribute a given pool of Cnb and Polo protein between the 2 centrioles so that the centriole inheriting high amounts of Cnb and Polo will retain MTOC activity. Furthermore, the dynamic localization of Polo and Cnb provides a molecular explanation for why the daughter-centriole-containing centrosome remains tethered to the apical neuroblast cortex and is being inherited by the self-renewed neuroblast. It remains to be tested why neuroblasts implemented such a robust machinery to asymmetrically segregate the daughter-containing centriole to the self-renewed neuroblast. More refined molecular and behavioral assays will be necessary to elucidate the developmental and postdevelopmental consequences of biased centrosome segregation. The tools and findings reported in this study will be instrumental in targeted perturbations of intrinsic centrosome asymmetry with spatiotemporal precision in defined neuroblast lineages (Gallaud, 2020).

Finally, the observations reported in this study further raise the tantalizing possibility that centriolar proteins also dynamically localize in other stem cells, potentially providing a mechanistic explanation for the differences in centriole inheritance across different stem cell systems (Gallaud, 2020).

Centrobin-tubulin interaction is required for centriole elongation and stability

Centrobin is a daughter centriole protein that is essential for centrosome duplication. However, the molecular mechanism by which centrobin functions during centriole duplication remains undefined. This study, carried out in mammalian cultured cells, shows that centrobin interacts with tubulin directly, and centrobin-tubulin interaction is pivotal for the function of centrobin during centriole duplication. Centrobin was found to be recruited to the centriole biogenesis site via its interaction with tubulins during the early stage of centriole biogenesis, and its recruitment is dependent on hSAS-6 but not centrosomal P4.1-associated protein (CPAP) and CP110. The function of centrobin is also required for the elongation of centrioles, which is likely mediated by its interaction with tubulin. Furthermore, disruption of centrobin-tubulin interaction led to destabilization of existing centrioles and the preformed procentriole-like structures induced by CPAP expression, indicating that centrobin-tubulin interaction is critical for the stability of centrioles. Together, this study demonstrates that centrobin facilitates the elongation and stability of centrioles via its interaction with tubulins (Gudi, 2011).

The coiled-coil protein centrobin is preferentially localized to the daughter centriole and is required for centriole duplication (Zou, 2005). Centrobin is recruited to the procentrioles at the beginning of S phase. During S, G2, and M phases, there are two centrobin-positive centrioles, the newly assembled procentrioles. After cell division, most G1 phase cells have one centrobin-positive centriole, the daughter centriole assembled in the previous cell cycle. Upon reentering S phase, centrobin on the daughter centriole assembled in the previous cycle becomes undetectable in the majority of the cells. In the absence of centrobin, no discernible centriole structures were assembled as demonstrated by EM analysis (Zou, 2005). Centrobin has also been reported to be a substrate of the kinase Nek2 and plays a role in stabilizing the microtubule network (Jeong, 2007). In addition, centrobin was found to regulate the assembly of functional mitotic spindles (Jeffery, 2010; Gudi, 2011 and references therein).

Centrioles are predominantly composed of α/β-tubulins. Centriole elongation, at least visually, is the process of assembling the nine microtubule triplets by adding α/β-tubulin dimers to an undefined initiating template structure. This study demonstrates that centrobin interacts directly with the core components of the centriole, α-tubulins via its C-terminal 139 residues (centrobin-TuBD). Centrobin-TuBD exhibited a clear centrosomal localization in addition to a diffused cytoplasmic and nuclear localization. It is noticeable that although centrobin-TuBD can bind strongly to tubulins, no detectable microtubular localization of centrobin-TuBD or adverse effect on microtubule nucleation was observed. Although surprising, this finding correlates with previous observations that endogenous centrobin is not clearly detectable on microtubules (Zou, 2005). Jeong (2007) had reported that centrobin is detectable in association with the roots or initiating points of microtubules in U2OS and MCF7 cells but not in HeLa cells. It is very likely that the tubulins at the initiating points of microtubules and tubulins at the centrosomes share similar conformation with the centrosomal tubulins, in which their centrobin-binding domain is exposed. When the tubulins are assembled into microtubules, the centrobin-binding domain is no longer accessible (Gudi, 2011).

Importantly, overexpression of centrobin-TuBD can disrupt the endogenous full-length centrobin-tubulin interaction, providing an a valuable tool to dissect the function of centrobin during centrobin duplication. Using centrobin-TuBD, it was demonstrated that centrobin is recruited to centrioles and facilitates centriole elongation via its interaction with centrosomal tubulins. Moreover, it was found that centrobin-TuBD overexpression destabilized the existing mother centrioles in addition to inhibiting the assembly of new centrioles, indicating that centrobin is required for the stability of centrioles. Previous findings (Zou, 2005) indicated that, in asynchronized cells, centrobin is mainly found on the daughter centrioles. Even in HU-treated cells, in which the degradation or displacement of centrobin from the daughter centriole is inhibited, there is at least one or two mother centrioles exhibiting no centrobin staining, which will suggest that centrobin should not be required for the stability of these mature centrioles. Two possible scenarios are proposed to explain these conflicting findings. First, centrobin is indeed required for the stability of both daughter and mature mother centrioles. On the mature mother centrioles, centrobin is still there to maintain their stability but becomes undetectable because additional mother centriole proteins block the access of centrobin antibody. Second, centrobin is only required to maintain the stability of the daughter centrioles. Once the daughter centrioles mature to become mother centrioles by recruiting additional mother centriole proteins and extensive modification of centriolar tubulins, centrobin is no longer required for their stability. Centrobin is then either degraded or displaced from the mother centrioles by the mother centriole proteins or the tubulin modifications. However, because of the small size of centrobin-TuBD and its presence at high concentration, it can still access the centrobin-binding domain on the tubulins and competes with the mother centriole proteins. Consequently, centrobin-TuBD will displace the mother centriole proteins and lead to destabilization of the mother centrioles. Although current data cannot distinguish between these two scenarios, it is speculated that the first scenario is more plausible because centrobin can indeed be present on the mother centriole as is evident in HU-treated cells. One seemingly conflicting piece of evidence against this hypothesis is that centrobin depletion inhibited the centriole duplication but did not destabilize the mother centrioles. The likely explanation is that centrobin assembled into the mother centrioles is stabilized and is impervious to depletion, as are most cellular structural proteins. Hence, centrobin depletion cannot destabilize the mother centriole, whereas centrobin-TuBD can displace the centrobin on the mother centrioles and lead to their destabilization. However, no direct evidence for this is available, and further studies are required to prove this scenario. Furthermore, in the rescue experiments, the existing centrioles were still destabilized in 5% of cells and were not rescued by the wild-type centrobin expression; therefore, it is possible that both scenarios can coexist and account for the observed destabilization of existing centrioles (Gudi, 2011).

The key findings from this study suggest that centrobin functions at least at three stages of the centriole duplication pathway (see Model of centrobin function during the centriole duplication process). At the beginning of centriole biogenesis, hSAS-6 is first recruited to the proximal end of the mother centrioles in the G1/S phase. Centrobin is then recruited and likely participates in the undefined centriole initiation structure formation along with the other proposed centriole initiation proteins, including CPAP, CEP135, γ-tubulin, and CP110. During the elongation of procentrioles, centrobin may function as a scaffold protein via its interaction with tubulins to facilitate the addition of tubulin dimers to the centriole initiation complex. Centrobin also acts to stabilize the newly assembled daughter centrioles before their maturation. Whether centrobin is required to maintain the stability of the mature mother centriole remains to be determined. The function of centrobin during centriole elongation and its function to maintain the stability of the centriole are likely both mediated by its ability to bind to tubulins (Gudi, 2011).

CPAP also has the ability to bind to microtubules, and regulation of its cellular levels is required to maintain the centriole length. It has been suggested that as a result of its tubulin binding property, CPAP might act as a scaffold for tubulin addition during procentriole biogenesis. Because the property of tubulin binding is shared by CPAP and centrobin and both are required for centriole biogenesis, it would be interesting to study whether centrobin and CPAP cooperate to facilitate the assembly of the microtubule triplets that form the main structure of centrioles. So far, there is no convincing evidence indicating the existence of a lower eukaryotic centrobin homologue. If centrobin indeed cooperates with CPAP for centriole elongation, it will indicate that centrobin is functionally similar to C. elegans SAS-5. In summary, it is concluded that centrobin-tubulin interaction is pivotal for centrobin recruitment to the centriole biogenesis site, centriole elongation, and stabilization of nascent centrioles until maturation (Gudi, 2011).

The interphase microtubule aster is a determinant of asymmetric division orientation in Drosophila neuroblasts

During asymmetric mitosis, both in male Drosophila germline stem cells and in mouse embryo neural progenitors, the mother centrosome is retained by the self-renewed cell; hence suggesting that mother centrosome inheritance might contribute to stemness. This hypothesis was tested in Drosophila neuroblasts (NBs) tracing photo converted centrioles and a daughter-centriole-specific marker generated by cloning the Drosophila homologue of human Centrobin. Upon asymmetric mitosis, the mother centrosome is inherited by the differentiating daughter cell. These results demonstrate maturation-dependent centrosome fate in Drosophila NBs and that the stemness properties of these cells are not linked to mother centrosome inheritance (Januschke, 2010).

Tracing mother and daughter centrioles with photo-converted PACT-d2Eos and YFP-CNB, respectively, it was shown w that apical and basal Drosophila NB centrosomes do not have fully formed diplosomes when they split, and that mother-versus-daughter centriole segregation tightly correlates with cell fate; contrary to what has been generally assumed, the daughter centrosome is fated to be retained by the NB and the mother centrosome is fated to the GMC (Januschke, 2010).

The functional relevance of specific centrosome retention by the stem cell in Drosophila remains to be ascertained. Both in mGSCs and in NBs, a microtubule-dependent mechanism hooks one centrosome to the region of the cortex, proximal to the hub in the germ line1 and apical in NBs; that is, retained by the stem cell after mitosis. Preferential retention of the mother centrosome in the germ line might therefore simply be the inescapable consequence of carrying more PCM and having greater microtubule-organizing activity. This is less likely in NBs where the daughter centriole is retained, while the mother centriole breaks away carrying little, if any, PCM. That implies the removal of PCM from around the mother centriole and the retention of PCM by the daughter centriole. The shedding of PCM by the GMC-fated centrosome as it splits from the apical centrosome has been previously documented (Januschke, 2010).

Precedence for centrosome transmitted cell fate information exist in other species48, but not in Drosophila. There are, however, published data on the effect of switching centrosome fate by transient microtubule depolymerization. Upon disassembly of the interphase aster of the NB by treatment with microtubule poisons, the apical centrosome loses connection with the cortex and moves deeper inside the cell so that at mitosis onset, the position of both centrosomes is essentially randomized. Upon recovery of microtubule dynamics, mitosis resumes and asymmetric cell division takes place. In some cases, centrosome fate is switched so that the centrosome originally destined to the GMC is retained by the NB, and vice versa, yet mitosis generates an NB and a GMC that are morphologically normal, strongly suggesting that the identity of the resulting NB and GMC is not severely compromised38. It is unknown, however, whether such switch in centriole fate might have long-term developmental consequences (Januschke, 2010).

Centriole-maturation-dependent segregation in Drosophila NBs is somewhat reminiscent of spindle pole body (SPB) segregation in budding yeast49. Similar to centrioles in animals, the SPB duplicates once per cell cycle. The new SPB remains connected to the old one by the bridge, which is cleaved in S phase. The two SPBs then separate to form the opposite poles of the spindle50. In unperturbed cells, the old SPB always migrates into the bud49. Old and new SPBs are also functionally and biochemically distinct. For instance, the Bfa1p-Bub2p GTPase-activating protein complex, which is an integral part of the spindle position checkpoint, specifically binds to the old SPB. The SPC inhibits the mitotic exit network) until the nucleus has migrated into the bud. Also the antigen-presenting cell-related molecule Kar9 localizes only to the old SPB due to the activity of the new SPB-resident Clb4/Cdc28 kinase. Interestingly, SPB fate can also be artificially switched, in which case, Bfa1p still localizes to the SPB that enters the bud independently of whether it is the old or the new. These data show that in budding yeast biochemical differences between the old and new SPBs are functionally relevant. (Januschke, 2010).

As in yeast where the old SPB is inherited by the long-lived bud, the mother centriole in Drosophila is inherited in testes by the mGSC, which is the most long-lived of the two mGSC daughters, and in larval brains by the GMC whose daughters outlive the NB. Thus, what appears to be a diametrically different behaviour between mGSCs and NBs using stemness as criteria is consistent, when the expected lifespan of each cell type is considered (Januschke, 2010).

PLP inhibits the activity of interphase centrosomes to ensure their proper segregation in stem cells

Centrosomes determine the mitotic axis of asymmetrically dividing stem cells. Several studies have shown that the centrosomes of the Drosophila melanogaster central brain neural stem cells are themselves asymmetric, organizing varying levels of pericentriolar material and microtubules. This asymmetry produces one active and one inactive centrosome during interphase. This study identifies pericentrin-like protein (PLP or cp309) as a negative regulator of centrosome maturation and activity. PLP is enriched on the inactive interphase centrosome, where it blocks recruitment of the master regulator of centrosome maturation, Polo kinase. Furthermore, it was found that ectopic Centrobin expression influenced PLP levels on the basal centrosome, suggesting it may normally function to regulate PLP. Finally, it is concluded that, although asymmetric centrosome maturation is not required for asymmetric cell division, it is required for proper centrosome segregation to the two daughter cells (Lerit, 2013).

A model is presented in which PLP functions as a negative regulator of centrosome maturation in interphase NBs by preventing the localization of Polo to the basal centrosome. Loss of PLP leads to the activation of both centrosomes in interphase and greatly reduces their mobility, resulting in the atypical apical positioning and inheritance of both centrosomes. Temperature shift experiments illustrate that plp^- mutants are sensitized to increased centrosome segregation and mitotic spindle defects. It is concluded that inhibition of basal/mother centrosome maturation during interphase is critical for proper centrosome partitioning and mitotic spindle organization in neural stem cells (Lerit, 2013).

It is unknown how PLP functions as a negative regulator, but one model may involve an indirect role for PLP. For example, PLP might normally promote early centriole separation to ensure efficient movement of the mother centriole away from the apical cortex to allow for its inactivation. Two experiments were performed to test this model. First, the timing of centriole separation was analyzed in cells exiting mitosis in WT and plp^- NBs, and it was found that centriole disengagement and the earliest signs of independent centriole movement are unaffected. Therefore, centrosomes remain apically positioned in early interphase in both WT and plp- NBs. Next, whether apical positioning of centrioles is sufficient for their activation was tested. Early interphase WT NBs were analyzed that contained two apical centrioles (<30^o separation) and no correlation was found between centrosome position and activity. Collectively, these observations argue that PLP does not influence early centrosome separation and that centrosome location is not critical for mother centrosome inactivation (Lerit, 2013).

Instead, the alternative hypothesis is favored that PLP directly regulates centrosome activity, which, in turn, influences centrosome position. One possible model is that PLP acts on the centrosome by physically shielding the docking of promaturation factors until mitotic entry. A biochemical modification and/or conformational change would then allow PLP to transition to a positive regulator in mitosis that scaffolds other PCM factors. Another possible direct function for PLP might be to differentially modulate the behavior of proteins at the apical and basal centrosomes. It will be critical to examine how the dynamics of other centrosome proteins are affected by the loss of PLP. Given the importance of interphase centrosome asymmetry in stem cell division, it is believed that an in depth understanding of the interplay between PCM proteins in both interphase and mitosis is a critical area of future research (Lerit, 2013).

Search PubMed for articles about Centrobin

Conduit, P. T. and Raff, J. W. (2010). Cnn dynamics drive centrosome size asymmetry to ensure daughter centriole retention in Drosophila neuroblasts. Curr Biol 20: 2187-2192. PubMed ID: 21145745

Gallaud, E., Ramdas Nair, A., Horsley, N., Monnard, A., Singh, P., Pham, T. T., Salvador Garcia, D., Ferrand, A. and Cabernard, C. (2020). Dynamic centriolar localization of Polo and Centrobin in early mitosis primes centrosome asymmetry. PLoS Biol 18(8): e3000762. PubMed ID: 32760088

Gopalakrishnan, J., Mennella, V., Blachon, S., Zhai, B., Smith, A. H., Megraw, T. L., Nicastro, D., Gygi, S. P., Agard, D. A. and Avidor-Reiss, T. (2011). Sas-4 provides a scaffold for cytoplasmic complexes and tethers them in a centrosome. Nat Commun 2: 359. PubMed ID: 21694707

Gudi, R., Zou, C., Li, J. and Gao, Q. (2011). Centrobin-tubulin interaction is required for centriole elongation and stability. J Cell Biol 193: 711-725. PubMed ID: 21576394

Izumi, H. and Kaneko, Y. (2012). Evidence of asymmetric cell division and centrosome inheritance in human neuroblastoma cells. Proc Natl Acad Sci U S A 109: 18048-18053. PubMed ID: 23064640

Januschke, J. and Gonzalez, C. (2010). The interphase microtubule aster is a determinant of asymmetric division orientation in Drosophila neuroblasts. J Cell Biol 188: 693-706. PubMed ID: 20194641

Januschke, J., Llamazares, S., Reina, J. and Gonzalez, C. (2011). Drosophila neuroblasts retain the daughter centrosome. Nat Commun 2: 243. PubMed ID: 21407209

Januschke, J., Reina, J., Llamazares, S., Bertran, T., Rossi, F., Roig, J. and Gonzalez, C. (2013). Centrobin controls mother-daughter centriole asymmetry in Drosophila neuroblasts. Nat Cell Biol 15: 241-248. PubMed ID: 23354166

Jeffery, J. M., Urquhart, A. J., Subramaniam, V. N., Parton, R. G. and Khanna, K. K. (2010). Centrobin regulates the assembly of functional mitotic spindles. Oncogene 29: 2649-2658. PubMed ID: 20190801

Jeong, Y., Lee, J., Kim, K., Yoo, J. C. and Rhee, K. (2007). Characterization of NIP2/centrobin, a novel substrate of Nek2, and its potential role in microtubule stabilization. J Cell Sci 120: 2106-2116. PubMed ID: 17535851

Lerit, D. A. and Rusan, N. M. (2013). PLP inhibits the activity of interphase centrosomes to ensure their proper segregation in stem cells. J Cell Biol 202: 1013-1022. PubMed ID: 24081489

Rebollo, E., Sampaio, P., Januschke, J., Llamazares, S., Varmark, H. and Gonzalez, C. (2007). Functionally unequal centrosomes drive spindle orientation in asymmetrically dividing Drosophila neural stem cells. Dev Cell 12: 467-474. PubMed ID: 17336911

Rebollo, E., Roldan, M. and Gonzalez, C. (2009). Spindle alignment is achieved without rotation after the first cell cycle in Drosophila embryonic neuroblasts. Development 136: 3393-3397. PubMed ID: 19762421

Reina, J., Gottardo, M., Riparbelli, M. G., Llamazares, S., Callaini, G. and Gonzalez, C. (2018). Centrobin is essential for C-tubule assembly and flagellum development in Drosophila melanogaster spermatogenesis. J Cell Biol 217(7):2365-2372. PubMed ID: 29712734

Rusan, N. M. and Peifer, M. (2007). A role for a novel centrosome cycle in asymmetric cell division. J Cell Biol 177: 13-20. PubMed ID: 17403931

Singh, P., Ramdas Nair, A. and Cabernard, C. (2014). The centriolar protein Bld10/Cep135 is required to establish centrosome asymmetry in Drosophila neuroblasts. Curr Biol 24: 1548-1555. PubMed ID: 24954048

Wang, H., Xie, Y. T., Han, J. Y., Ruan, Y., Song, A. P., Zheng, L. Y., Zhang, W. Z., Sajdik, C., Li, Y., Tian, X. X. and Fang, W. G. (2012). Genetic polymorphisms in centrobin and Nek2 are associated with breast cancer susxtibility in a Chinese Han population. Breast Cancer Res Treat 136: 241-251. PubMed ID: 23001753

Yamashita, Y. M., Mahowald, A. P., Perlin, J. R. and Fuller, M. T. (2007). Asymmetric inheritance of mother versus daughter centrosome in stem cell division. Science 315 (5811): 518-21. PubMed Citation: 17255513

Zou, C., Li, J., Bai, Y., Gunning, W. T., Wazer, D. E., Band, V. and Gao, Q. (2005). Centrobin: a novel daughter centriole-associated protein that is required for centriole duplication. J Cell Biol 171: 437-445. PubMed ID: 16275750

date revised: 24 August 2021

Home page: The Interactive Fly © 2006 Thomas Brody, Ph.D.

The Interactive Fly resides on the
Society for Developmental Biology's Web server.