Centriole stability may also be aided by post-translational modifications of centriolar tubulin, such as glutamylation reviewed in [ 5 , 6 ]. Centriole size is highly stable and homogeneous after reaching final length, suggesting that a length-maintenance mechanism must exist. The centriole component CP is a strong candidate for this function: it localizes to centriole tips and its depletion leads to the formation of abnormally long centrioles that might fragment originating abnormal mitotic spindles reviewed in [ 5 , 6 ].
Several centrosome components have been identified recently, through proteomic studies or functional genomic analysis and their localization and function characterized [ 7 — 9 ]. From these studies, a new view, of a highly organized PCM is emerging, where different domains might be involved in separate functions and are regulated differently through the cell cycle [ 10 — 14 ]. The size and organization of the PCM is likely to impinge on centrosome function and is determined by the intrinsic properties of its components size, shape and protein domains, amongst others , their availability and their regulation by kinases [ 5 , 9 , 15 ].
How this all works to ensure centrosome function is poorly understood and is an important avenue of research for the future. Most eukaryotic cells do have a microtubule cytoskeleton but this can be organized in many different ways by MTOCs, which need not be centrosomes. Several species do not have centrosomes. In many cells in Drosophila, centrosomes are inactive in interphase and only become active in mitosis [ 17 ].
Centrosomes are absent in many species of fungi and seed plants, as well as in many classes of protists, and in these species the specific genes encoding the proteins responsible for the nine-fold symmetry of centrioles, appendage formation, microtubule stability and length regulation have been lost [ 18 , 19 ] Figure 2.
Within animals, the flatworm Planaria , despite making centrioles that assemble cilia, does not have centrosomes [ 20 ]. Moreover, even within mammals there are cases of acentriolar cells: female oocytes lack centrioles Figure 3 c and the mouse embryo develops with no centrioles until the cell stage [ 21 ]. Commonly, in differentiated animal cells the centrosome is no longer the major MTOC and is inactive.
That is the case for muscle cells, epithelial cells and neurons. In those cell types, upon differentiation, the centrosome often loses PCM components, which delocalize to other parts of the cell such as the cytoplasmic membrane and the nuclear envelope, which then behave as the MTOC reviewed in [ 16 , 22 ]. Furthermore, centrosomes are not essential for mitotic spindle assembly, even in cells that normally have them. Moreover, mutant flies that do not assemble centrosomes from centrioles or that do not have centrioles can develop to adulthood [ 23 , 24 ].
In those flies, somatic cell division is fine although some defects are observed in asymmetric cell division and cytokinesis [ 24 ]. In summary, centrosomes contribute to mitotic fidelity, cytokinesis and asymmetric cell division, but this is not essential for the development of the flies reviewed in [ 16 ]. Unikonts include eukaryotic cells that, for the most part, have a single emergent flagellum divided into opisthokonts propel themselves with a single posterior flagellum; metazoans, fungi and choanoflagellates and Amoebozoa.
Bikonts include eukaryotic organisms with two emergent flagella. Branch color code: purple, opisthokonts; blue, Amoebozoa; green, plants; yellow, alveolates; orange, stramenopiles; rose, Rhizaria; brown, excavates and discicristates. Adapted with permission [ 19 ]. Originally published in J Cell Biol. Regulation of centrosome number. Procentriole formation begins in S phase orthogonally to its mother. CDK2 activity may be necessary for speeding up procentriole formation and elongation, thus coordinating this event with DNA replication.
In G2, the daughter centriole reaches full elongation and maturation with the recruitment of several molecules to the pericentriolar material PCM. CDK1 activity increases in G2 regulating a variety of molecules and processes needed for entry into mitosis, such as changes in microtubule dynamics.
Through the concerted action of molecules such as the kinase Nek2, the two centrosomes separate. The mitotic spindle segregates the chromosomes equally to the two daughter cells. When a cell exits mitosis, the centrioles within the centrosome disengage. That process may allow recruitment or activation of molecules necessary for duplication. Hundreds of centrioles are formed either around a pre-existing mother centriole 1 or a deuterosome 2.
These centrioles migrate and dock to the cell membrane, where they nucleate cilia. Centrioles disappear during oogenesis in many animal species. Female meiosis is acentriolar. After egg activation multiple centrioles arise de novo and join the female pronucleus resulting from meiosis. In the absence of fertilization, those MTOCS set up the first mitotic spindle in the unfertilized egg.
The remaining MTOCs disappear. The chromatin pathway generates microtubules close to the chromosomes, a process that can depend on RanGTP or a molecular complex called CPC chromosome passenger complex. Moreover, microtubules can be nucleated from pre-existing microtubules, through a molecular complex called augmin.
Finally, the nuclear envelope may also contribute to microtubule nucleation reviewed in [ 25 ]. Therefore, it would appear that centrosomes are not always necessary for spindle assembly and cell division. Centrosomes are important for specialized cell divisions.
For example, in Drosophila , adult males with no centrosomes show highly abnormal meiotic divisions [ 26 ]. Moreover, eggs from mothers that are mutant for centriole proteins arrest very early in embryonic development after only a few abnormal mitoses, showing that centrioles are necessary for syncytial mitoses [ 26 , 27 ].
Moreover, asymmetric cell divisions can also be abnormal in the absence of centrosomes reviewed in [ 16 ]. The same is true in other organisms, such as the Caenorhabditis elegans embryo and fission yeast, where the centrosome and its equivalent, the spindle pole body, are essential for bipolar spindle assembly and cytokinesis, respectively reviewed in [ 16 , 26 ].
A large debate exists on whether this is just an epiphenomenon reviewed in [ 4 ]. Findings, some dating back almost 40 years, that male and female meiotic cells can form bipolar acentrosomal spindles challenged the notion that centrosomes are needed for the establishment of spindle bipolarity Dietz ; Steffen et al. More recent work with Xenopus egg extracts has revealed that bipolar spindles will assemble from initially randomly oriented microtubules assembled in the vicinity of chromatin, be it chromosomes or beads coated with DNA fragments Heald et al.
Spindle organization is achieved through the bundling and movement of these microtubules into bipolar arrays by the action of microtubule-based motor proteins coupled with the bundling action of the nuclear mitotic apparatus protein NuMa; Gaglio et al.
Together, these findings have raised the possibility that centrosomes are simply microtubule donors or are perhaps completely irrelevant for spindle pole formation. However, when observations of spindle assembly in a wide variety of dividing cells are considered together, one finds that the ability of a cell to form acentrosomal spindle poles is system dependent and that centrosomes, when present, act in a dominant fashion to determine spindle polarity Heald et al.
The alternative mechanism for bipolar spindle assembly is revealed only when centrosomes are not present. To further complicate the story, the existence of stable cell lines in which many or all of the cells contain multiple centrosomes has challenged the notion that multiple centrosomes always produce a multipolar cell division. Similarly, one can ask how the cells of aggressive human tumors containing multiple centrosomes can propagate without catastrophic loss of chromosomes and consequent loss of viability Lingle et al.
We submit that these observations do not necessarily constitute a challenge to the notion that the number of centrosomes determines spindle polarity in normal cells.
These cells have undergone clonal selection for compensatory mechanisms that are able to deal with the presence of multiple centrosomes. Possibly these cells have up-regulated the levels of microtubule bundling proteins, such as NuMa, so that the centrosomes are constrained from separating to form a multipolar spindle. In effect, the alternative mechanisms for spindle bipolarization, described above, now dominate.
At the end of mitosis, each daughter cell inherits a single centrosome, and by the onset of the next mitosis, it contains just two centrosomes.
This precise doubling of the interphase centrosome in preparation for mitosis is called centrosome duplication or reproduction. In higher animal cells, centrosome reproduction consists of four morphological events: 1 centriole splitting, 2 centriole duplication, 3 centrosome disjunction, and 4 daughter centrosome separation Fig. Schematic representation of the centrosome reproduction cycle; centrioles are represented as barrels, and the pericentriolar material is not shown.
Centriole duplication: Short, annular daughter centrioles, called procentrioles, are assembled at right angles to the parent centrioles. The procentrioles elongate during S phase or G 2 and may not reach their full length until the following G 1. Centrosome disjunction: The duplicated centrosome disjoins during G 2 into two functionally separate centrosomes, each containing a mother-daughter pair of centrioles.
Centrosome separation: The sister centrosomes physically separate from each other through the action of microtubule-based motor proteins. Centriole splitting, or centriole disorientation, has been defined as the detachment and loss of orthogonal relationship between the mother-daughter centriole pair, observed during late G 1 Kuriyama and Borisy ; Wheatley Although this event has been commonly said to be the leading event in centrosome reproduction, the mother and daughter centrioles can widely separate from each other as early as telophase in some cultured cells Mack and Rattner ; Piel et al.
It is not known whether these separated centrioles are still joined by an extensible linkage or have truly split well before the G 1 -S transition. Centriole duplication is first seen at the beginning of S phase or during S phase by the appearance of short daughter centrioles, or procentrioles, at right angles to and separated slightly from the two original centrioles. These procentrioles elongate during S and G 2 , reaching mature length in mitosis or the following G 1 Kuriyama and Borisy ; Lange et al.
The completion of centrosome reproduction occurs with centrosome disjunction at a variable time in G 2 , with pairs of mother-daughter centrioles going to each daughter centrosome Aubin et al. The disjunction of the sister centrosomes is correlated with the activity of Nek2, the mammalian homolog of the Asperillus protein NIMA, a cell cycle regulatory kinase that contributes to driving the cell into mitosis Fry et al. The kinase activity of Nek2 and or the phosphorylation of centrin may function to coordinate sister centrosome disjunction with entry into mitosis, thereby ensuring that the cell contains two spindle poles at the right time.
This is followed by daughter centrosome separation through the action of microtubule-based motor proteins Sharp and Scholey However, the extent to which aster separation occurs before the onset of mitosis can vary between cells, even those in the same population. In some cases, the two centrosomes remain close together until nuclear envelope breakdown, whereas in others, both asters are well separated around the nucleus before the end of prophase for review, see Rieder Control of centrosome duplication is exercised by limits that are intrinsic to the centrosome itself and by extrinsic controls imposed by changing cytoplasmic conditions during cell cycle progression.
Limits intrinsic to the centrosome determine the number of daughter centrosomes that arise from the parent centrosome; cytoplasmic controls determine when the centrosome duplicates in relation to the progression of nuclear events such as DNA synthesis and mitosis. Functional studies of living cells coupled with serial section electron microscopy indicate that there is a counting mechanism within each centrosome that limits the number of daughters that can arise from the parent centrosome.
Zygotes contain enough centrosomal subunits at fertilization to assemble many complete centrosomes Gard et al. This specificity of the duplication process appears to be determined by the cycle of centriole disjunction and centriole duplication. Because centrioles act to localize the PCM, as discussed earlier, the number of centriole pairs determines the number of spindle poles.
The evidence for this assertion originated with the remarkable finding of Mazia et al. When mitosis is prolonged by any of several independent methods, the two spindle poles split during mitosis to yield four functional poles that will not further subdivide even when mitosis is prolonged to 20 times its normal duration Mazia et al.
Ultrastructural analysis of such tetrapolar spindles reveals that each pole contains only one centriole, confirming that the centrosomes have split, not duplicated Fig. After the cell divides into four, these half centrosomes each assemble a daughter centriole, thus becoming normal centrosomes with full reproductive capacity.
However, they do not undergo centriole splitting or centrosome disjunction, and each cell assembles a monopolar spindle at next mitosis Fig. In effect, the two mitotic centrosomes with normal reproductive capacity have subdivided into four centrosomes with half the normal reproductive capacity and half the complement of centrioles. If a daughter cell with a monopolar spindle remains in mitosis longer than normal, as often happens, the centrosome of the monopolar spindle will split to give two functional spindle poles with one centriole apiece.
These poles undergo centriole duplication but not centrosome disjunction during interphase, and monopolar spindles are once again formed at the following mitosis. These observations reveal that each centriole can organize a functional spindle pole, but normally the mother and daughter centrioles remain physically associated with each other, thereby forming only one spindle pole.
The importance of centrioles in the control of centrosome number was further substantiated by the finding that sea urchin zygotes, from which the centrioles were removed, would form a single MTOC.
Thus, the mechanism that determines the doubling of a spindle pole in preparation for mitosis is not part of the PCM. Although all of this work was performed on sea urchin zygotes, these observations are not peculiar to embryonic cells. When mitosis is prolonged in cultured cells by transfection with a nondegradable cyclin B construct, the spindle poles double from two to four Gallant and Nigg The refractile sphere in the upper right portion of the zygote is a drop of oil used to cap the micropipet.
Serial semithick section ultrastructural analysis of similar zygotes reveals that each spindle pole contains only one centriole.
Diagrammatic representation of the experimental manipulation of the reproductive capacity of spindle poles in zygotes. During prolonged prometaphase, the centriole pairs and centrosomes split without duplicating, and the four spindle poles separate from each other, each containing a single centriole.
In telophase, as the cell divides into four, the singlet centrioles replicate but do not separate. The result is the formation of monopolar spindles at the next mitosis. Each centrosome has the normal complement of two centrioles.
When a cleavage furrow fails to form, two monopolar spindles come together to assemble a functional bipolar spindle with poles that reproduce in a normal fashion. The single pole splits and a bipolar spindle forms when prometaphase is prolonged, as it often is in such cells.
Each sister aster contains only one centriole. After anaphase, the cell divides and the singlet centrioles duplicate but do not split during interphase. At the next mitosis, monopolar spindles are again assembled. The link between the cycle of centriole duplication and spindle pole reproduction, however, begs the question of how cells without centrioles manage to organize bipolar spindles during successive mitoses. Presumably, such cells use the alternate spindle assembly pathway, described earlier, by which microtubules nucleated in the vicinity of chromatin are sorted by microtubule-based motor proteins and then bundled in a bipolar array by crosslinkers such as NuMa for review, see Compton This means that spindle pole duplication does not exist in acentriolar cells; the poles of the spindle are formed anew at each mitosis.
The mechanism for the precise one-to-two duplication of centrioles is not understood. Specific precursor structures, such as an annular ring or a looped fiber containing 9 densely staining foci that later elaborate into triplet microtubules, have been described in ciliate basal body duplication Dippel ; Gould This templating hypothesis is supported by findings that zygotes have sufficient complete pools of centrosome subunits to make many centrosomes, yet daughter centrosomes form only one at a time at the parent centrosome Gard et al.
Also, when the centrosome is removed from zygotes Sluder et al. To the best of our knowledge, the only exceptions to this rule are found in the de novo formation of centrioles after parthenogenetic activation of sea urchin eggs Von Ledebur-Villiger ; Kallenbach and Mazia , the formation of multiple basal bodies from specialized generative structures during spermiogenesis in the water fern Marselia Hepler , and the assembly of multiple basal bodies from specialized generative structures in ciliated epithelia in higher animals Sorokin ; Dirksen For example acentriolar Naeglaria amoebae will assemble two basal bodies when they differentiate into a flagellated form after environmental stress Fulton and Dingle In addition, during mouse development, centrioles are not seen in the early mitotic divisions Szollosi et al.
Historically, the implied continuity of structure and pattern in centriole duplication has brought to mind DNA replication, the contemporary paradigm for a templated reproductive process in which information and copy number are under rigid control. Thus, it is not surprising that almost 40 years ago, workers started considering the notion that centrioles and basal bodies contain nucleic acids that serve either a genomic or structural role in the duplication process.
The thinking was that centrioles could be semiautonomous organelles with their own DNA genomes, much like mitochondria and chloroplasts. Alternatively, centrioles, like ribosomes, could contain RNA that would serve a structural role in the assembly of daughter centrioles.
These possibilities inspired searches, spanning many years, for nucleic acids physically located in centrioles and basal bodies. Because all of this work was fraught with serious technical problems and with initially promising but ultimately inconclusive observations, we will not review this field. The early work was reviewed by Fulton , and the later work has been reviewed by Marshall and Rosenbaum Suffice it to say, there is yet no credible evidence for the existence of DNA in centrioles or basal bodies, and claims to this effect have been discredited by the most recent experimentation.
Understandably, nobody has been able to rigorously differentiate between RNAs that are specific to the centriole and nuclear gene products, such as ribosomal, transfer, or messenger RNAs, which happen to lie within the cytoplasmic volume of the centriole or basal body. Thus, it appears that this seductive siren song about the role of nucleic acids in the control of centrosome duplication and function has been marginalized—for now.
The cell must ensure that the events of centrosome reproduction are properly coordinated with nuclear events in the cell cycle if it is to have just two centrosomes at the onset of mitosis.
Although much progress has been made over the past 15 years in defining the controls that ensure this essential coordination, we still do not have enough pieces of the puzzle to provide a complete picture. Because this field has progressed largely by testing a variety of hypothetical control strategies, we will review the potpourri of studies by category. By beating around the bush long enough, we hope to eventually define the outlines of the bush. This possibility was tested by using a micropipette to remove the nucleus and one centrosome from sea urchin zygotes Sluder et al.
These zygotes were ideal for this investigation because they contain large stores of mRNAs and proteins to support rapid early development without growth, and consequently, the cell cycle continues in the absence of the nucleus. The fact that all daughter centrosomes contained two centrioles indicated that centrosome reproduction was normal and complete. Thus, neither the presence of the nucleus nor its activities are required for repeated cycles of centrosome reproduction; temporal control of centrosome reproduction is under cytoplasmic control.
The finding that all the centrosomes reproduced in synchrony within a zygote indicates that the temporal control of their reproduction is exercised by a cyclical change in the state of the cytoplasm. The way in which the centrosome and nuclear cycles are coordinated might seem to be straightforward at first glance: The events of centrosome reproduction could be driven by cytoplasmic conditions particular to certain cell cycle stages.
However, the appealing simplicity of this notion was clouded by demonstrations that centrosomes repeatedly reproduce when the cell cycle is arrested in interphase by inhibitors of DNA synthesis or protein synthesis Sluder and Lewis ; Raff and Glover ; Gard et al. Such findings raised the possibility that the nuclear and centrosomal cycles are regulated by independent pathways Sluder et al.
Thus, it was important to experimentally define which cell cycle stages support centrosome reproduction and which do not. The approach used has been to arrest cells at various phases of the cell cycle and then to determine whether the centrosome will reproduce one or more times without further experimental intervention.
This phase of the cell cycle does not support centrosome reproduction. The centrosome does not duplicate as long as the cell is in this quiescent phase of the cell cycle Tucker et al. Whether or not centrosome reproduction begins during G 1 , when the cell has passed the restriction point and is committed to prepare for division, appears to vary among different cell systems. Before cell division, the centrosome duplicates and then, as division begins, the two centrosomes move to opposite ends of the cell.
Proteins called microtubules assemble into a spindle between the two centrosomes and help separate the replicated chromosomes into the daughter cells.
The centrosome is an important part of how the cell organizes the cell division. The major function of the centrosome is organization of microtubules in the cell, thereby controlling cellular shape, polarity, proliferation, mobility and cell division. During S-phase, the centrosome is replicated in a semi-conservative manner, resulting in formation of one daughter centriole next to each of the parental centrioles. As the cell approaches mitosis, the two centrosomes, each containing a parental centriole and a maturing procentriole, move to opposite ends of the cell.
At the same time, the amount of surrounding PCM proteins increase, enabling nucleation of more microtubules. When the nuclear membrane breaks down, microtubules originating from each of the centrosomes can interact with kinetochores on the replicated sister chromatids, forming the characteristic mitotic spindle.
The intricate spindle apparatus mediates separation of sister chromatids to opposite ends of the cell, and upon cytokinesis each of the daughter cells is provided with one set of chromosomes and one centrosome. The parental centriole, i. Centriolar satellites have long been considered as vehicles for protein trafficking to and from the centrosome and cilia, thus playing a role in dynamic regulation of protein composition in these organelles Tollenaere MA et al.
Indeed, several proteins that localize to centriolar satellites have been implicated in centrosome replication, maturation and separation. However, in recent studies, centriolar satellites have also emerged as regulators of multiple other cellular processes, such as protein degradation and autophagy, some of which are independent of centrosomes and cilia.
Similarly, centrosomes and cilia are not fully dependent on centriolar satellites. As key regulators of chromosome segregation and cell cycle progression, abnormalities in number, size and morphology of the centrosome, and mutations in genes encoding protein that localize to centrosomes, is commonly observed in cells undergoing tumorigenesis, but also in some other diseases Badano JL et al. Gene Ontology GO -based analysis of genes encoding proteins that localize to centrosomes or centriolar satellites shows enrichment of terms describing functions that are well in-line with existing literature.
The most highly enriched terms for the GO domain Biological Process are related to mitosis and cytokinesis, cell cycle progression, endocytosis, organization of the microtubule cytoskeleton, and organization of organelles Figure 3a. Enrichment analysis of the GO domain Molecular Function reveal enrichment of terms describing binding to microtubules and motor proteins, as well as motor activity Figure 3b. Figure 3a. Gene Ontology-based enrichment analysis for the centrosome proteome showing the significantly enriched terms for the GO domain Biological Process.
Figure 3b. Gene Ontology-based enrichment analysis for the centrosome proteome showing the significantly enriched terms for the GO domain Molecular Function. The network plot shows that the most common locations shared with the centrosome and centriolar satellites are the cytoplasm, nucleus and vesicles.
Dual localizations with nucleoplasm and cytosol are overrepresented, while dual localizations with the Golgi apparatus and nucleoli are underrepresented. Figure 4.
0コメント