This page has been archived and is no longer updated
Centrosome aberrations: cause or consequence of cancer progression?
Author: Erich A. Nigg
Keywords
Keywords for this Article
Add keywords to your Content
Save
|
Cancel
Share
|
Cancel
Revoke
|
Cancel
Rate & Certify
Rate Me...
Rate Me
!
Comment
Save
|
Cancel
Flag Inappropriate
The Content is
Objectionable
Explicit
Offensive
Inaccurate
Comment
Flag Content
|
Cancel
Delete Content
Reason
Delete
|
Cancel
Close
Full Screen
"� 2002 Nature Publishing Group REVIEWS Centrosomes are the microtubule-organizing centres of animal cells 1?3 . By controlling the number, polarity and distribution of microtubules, they coordinate all micro- tubule-related functions. These include cell shape, polarity, adhesion and motility, as well as the intracellu- lar transport and positioning of organelles. Furthermore, centrosome function is crucial for chro- mosome segregation and CYTOKINESIS. Although certain female germ cells can assemble bipolar SPINDLES in the absence of centrosomes, the number of centrosomes that are present in somatic cells determines the number of spindle poles. Normally, the two centrosomes that are present at the onset of mitosis instruct the formation of a bipolar spindle. Extra copies frequently result in the formation of multipolar spindles, and failure of centro- somes to separate results in monopolar ASTERS. Aberrations in the number of centrosomes therefore almost inevitably cause chromosome missegregation. Centrosomes also determine the positioning of the CLEAVAGE PLANE during cytokinesis, which is essential for asymmetric divisions and morphogenesis. Considering the multitude of cellular properties that depend on accurate centrosome function, it is not surprising that the structure and number of these organelles is tightly regulated throughout the cell cycle 2,4,5 . Conversely, there is evidence that the centrosome contributes to cell-cycle regulation and checkpoints 3,6?9 . As early as 1914, Theodor Boveri proposed a direct link between centrosomal abnor- malities and both the aneuploidy (BOX 1) and loss of tissue architecture that are typical of human tumours 10 . Sparked by the demonstration that cen- trosomal abnormalities are frequent in many com- mon cancers 11?13 , interest in this old hypothesis has staged an impressive comeback. So how do centro- some aberrations in tumours occur, and how might they contribute to chromosomal instability (BOX 1) and other characteristic features of human tumours? The centrosome duplication cycle Discovered more than a century ago, the centrosome is a tiny organelle of surprising structural complexity (BOX 2). A single centrosome consists of two centrioles that are surrounded by amorphous pericentriolar material (PCM). Centrioles are important in the assembly of the PCM and the anchoring of microtubules, but the nucle- ation of microtubules occurs from within the PCM, where ?-TUBULIN RING COMPLEXES act as nucleation tem- plates 14 (BOX 2). In humans and most other mammalian species, the sperm contributes the centrosome to the zygote 15 . Throughout development and adult life, this single centrosome then needs to be duplicated once, and only once, in every cell cycle. On the basis of early mor- phological studies, the centrosome duplication cycle can be subdivided into several distinct steps (BOX 3). Our understanding of the regulation of these steps remains CENTROSOME ABERRATIONS: CAUSE OR CONSEQUENCE OF CANCER PROGRESSION? Erich A. Nigg Many human tumours show centrosome aberrations, indicating an underlying deregulation of centrosome structure, duplication or segregation. Centrosomes organize microtubule arrays throughout the cell cycle, thereby influencing both tissue architecture and the accuracy of chromosome segregation. But what are the origins of centrosomal abnormalities in tumours, and what impact do they have on the generation of invasive, genetically unbalanced cells during cancer progression? CYTOKINESIS The process of cytoplasmic division. SPINDLE A dynamic bipolar array of microtubules that is assembled during mitosis and meiosis to segregate chromosomes. ASTERS Radial microtubule arrays with minus ends that are usually tethered to centrosomes (or assemblies of centrosomal proteins) and plus ends that extend towards the periphery. CLEAVAGE PLANE The plane of cell division ? defined by the assembly of a contractile actomyosin ring at the cell cortex. NATURE REVIEWS | CANCER VOLUME 2 | NOVEMBER 2002 | 1 Max Planck Institute of Biochemistry, Department of Cell Biology, Am Klopfersitz 18a, D-82152 Martinsried, Germany. e-mail: nigg@biochem.mpg.de doi:10.1038/nrg924 � 2002 Nature Publishing Group ?-TUBULIN RING COMPLEX A ?-tubulin-containing multiprotein complex that acts as a ring-shaped template for microtubule nucleation in metazoan organisms. 2 | NOVEMBER 2002 | VOLUME 2 www.nature.com/reviews/cancer REVIEWS duplication and DNA replication require the hyper- phosphorylation of the retinoblastoma (RB) protein and the activation of cyclin-dependent kinase 2 (CDK2) 19?22 . This ensures one level of coordination between these two key S-phase events, but also implies that mutational inactivation of the retinoblastoma pathway in human cancers will potentially deregulate both DNA replication and centrosome duplication. Considering that the loss of coordination between the centrosome cycle and the chromosome cycle is likely to constitute an important primary cause of numerical chromosomal instability in human tumours, it is incomplete, but it is clear that phosphorylation has a key role 4,5 . Furthermore, there is increasing evidence for an important contribution of ubiquitin-dependent proteolysis in the regulation of centrosome biology 16,17 . From the perspective of tumorigenesis, one of the key issues to be resolved concerns the coordination of the centrosome and chromosome duplication cycles (FIG. 1). Although these two cycles can be dissociated experimentally during the rapid early nuclear divi- sions in the embryos of some species 18 , in human somatic cells they were shown to be linked through the retinoblastoma pathway 19 . Both centrosome Box 1 | Aneuploidy and chromosomal instability Most aggressive human cancers are characterized by an inherent instability of their genomes, a phenotype termed genomic (or genetic) instability 41,82?84 . Whether genomic instability is strictly required for tumorigenesis remains subject to debate 85,86 , but it almost certainly favours both the adaptation of developing tumours to changing physiological conditions and the emergence of therapy-resistant cells. The most common type of instability ? chromosomal instability ? is visible at the cytological level. It is present in most, if not all, classes of solid tumour and so constitutes the most conspicuous hallmark of cancer 87 . As revealed by molecular cytogenetic methods, such as comparative genomic hybridization (CGH), multiplex fluorescence in situ hybridization (M-FISH) or spectral karyotyping (SKY), chromosomal instability is characterized by losses or gains of whole chromosomes (aneuploidy), as well as chromosome rearrangements. Importantly, the term ?chromosomal instability? describes a rate of change, whereas the term ?aneuploidy? merely refers to a state 83 . Relatively rare genomic instabilities occur at the level of the nucleotide sequence. These result from mutational inactivation of pathways that are involved in mismatch or nucleotide-excision repair 83 . By contrast, the molecular mechanisms that give rise to chromosomal instability remain largely unknown. Depending on the cause of instability, aberrant karyotypes might be dominated by either chromosome number aberrations or structural aberrations (such as amplifications, deletions and translocations). Many chromosome rearrangements might reflect telomere dysfunction, but chromosome number aberrations are likely to arise through different mechanisms. Centrosomal abnormalities almost certainly represent one important cause of chromosome missegregation 46,69,70 . Additional plausible causes include inappropriate chromosome condensation or cohesion, deregulated mitotic progression, an impairment of the spindle-assembly checkpoint, cytokinesis malfunction, endoreduplication or cell fusion 80,83,88,89 . Summary ? The centrosome nucleates microtubules; it is important for cell shape, motility and division. During S phase of the cell cycle, the single centrosome that is present in a G1-phase cell is duplicated. The two centrosomes then set up the poles of the mitotic spindle and each incipient daughter cell receives one centrosome. ? The duplication and segregation cycles of centrosomes and chromosomes need to be coordinated to avoid chromosome missegregation or ploidy changes. The retinoblastoma pathway has been identified as one important link between centrosome duplication and chromosome replication. ? Many tumours display numerical and structural centrosome aberrations. Extra copies of centrosomes could, in principle, arise through overduplication within a single cell cycle, through aborted cell division, cell fusion or de novo genesis. A growing body of evidence points to aborted division as an important cause of excessive centrosome numbers. ? Cells that lack a functional p53 pathway are proposed to acquire multiple centrosomes through failure of a G1-phase checkpoint that should eliminate cells after aborted division. However, it has also been argued that p53 regulates centrosome duplication. ? Centrosome aberrations can give rise to chromosomal instability and altered tissue architecture. Importantly, centrosome aberrations and chromosomal instability are expected to enhance each other. ? Most multipolar divisions cause severe chromosome missegregation and therefore constitute lethal events. Occasionally, however, they might give rise to cells with chromosomal compositions that favour survival in the microenvironment of the tumour. In tumour cells, genes that are involved in alternative mechanisms for spindle formation might be upregulated or re-expressed. This might cause several centrosomes to coalesce and allow the formation of bipolar spindles, in spite of excessive centrosome numbers. ? A better understanding of the origins and consequences of centrosome aberrations could lead to the development of novel diagnostic, prognostic or therapeutic approaches. � 2002 Nature Publishing Group NATURE REVIEWS | CANCER VOLUME 2 | NOVEMBER 2002 | 3 REVIEWS with high-risk human papillomavirus (HPV-16/HPV- 18) infection 30?32 . This is a particularly interesting observation, as it offers a unique opportunity to explore the generation of centrosomal abnormalities in response to the expression of the HPV oncoproteins E6 and E7 (see below). Numerical centrosome aberrations are frequently accompanied by structural irregularities. These include increases in centrosome size (FIG. 2), the formation of ACENTRIOLAR BODIES and alterations in the phosphorylation state of PCM components 11,12,18,23,26,31,33?35 . Presumably, these alterations reflect deregulation of the expression and activity of centrosomal proteins. Support for this important to achieve a better understanding of the pathways that synchronize the two cycles in somatic cells. Centrosome aberrations in human tumours Centrosomal abnormalities are very common not only in tumour-derived cell lines and animal tumour models, but also in both primary and metastatic human tumours (FIG. 2). Extra copies of centrosomes (supernumerary centrosomes) have been described for nearly all cancers that have been surveyed, including brain, breast, bile duct, colon, head and neck, lung, pancreas and prostate cancers 11?13,23?29 . Increased centrosome numbers have also been reported for cervical cancers that are associated ACENTRIOLAR BODIES Assemblies of centrosomal proteins that can form in the absence of centrioles. Box 2 | Centrosome structure and function The centrosome is a relatively small organelle ? its diameter is ~1 �m ? that comprises a pair of centrioles that are embedded in a proteinaceous matrix of pericentriolar material (PCM) (see figure; adapted from REFS 90,91). Centrioles are cylindrical structures that are made up of nine triplet microtubules, but they are unequal in that only one (the older of the two) carries appendages that are close to its distal end (see figure). During S-phase of the cell cycle, procentrioles assemble next to the proximal ends of both parental centrioles (not shown). (Occasionally, parental centrioles and procentrioles are also referred to as mother and daughter centrioles, respectively.) The complete maturation of a centriole ? that is, the time from procentriole formation to the acquisition of appendages ? requires about 1.5 cell cycles. Centrioles are closely related to the basal bodies underlying cilia and flagella 1,3,92 . They contribute to PCM assembly and to the anchoring of microtubules, primarily via their appendages 1 . Centrioles are not strictly required for spindle formation, but are essential for the formation of spindle asters. In turn, interactions between astral microtubules and the cortex are crucial for spindle positioning, asymmetric cell divisions and morphogenesis during development 72 . Under the electron microscope, the PCM appears as an amorphous, electron-dense cloud (see figure). The complete inventory of centrosomal components has not yet been established, but several dozen proteins have been reported to localize to the centrosome, either transiently or throughout the cell cycle. Prominent among the PCM components are ?-tubulin ring complexes, which act as templates for microtubule nucleation 14 . In addition, the PCM harbours several large proteins with predicted coiled-coil domains (for example, AKAP450, kendrin/pericentrin, C-NAP1/CEP250, ninein and CEP135), indicating that these components perform structural functions 1?3 . Furthermore, several protein kinases, phosphatases, components of the ubiquitin-dependent proteolytic machinery and microtubule-dependent motors associate permanently or transiently with centrosomes 5,8 . Although many of these activities might control centrosome function, others might use the centrosome as a structural platform to enhance the efficiency of reactions that are crucial for cell-cycle progression. Remarkably, most centrosomal proteins that have been studied so far also exist in a soluble, cytoplasmic pool, indicating that centrosomes are highly dynamic structures. Centrosomes organize the microtubule network throughout the cell cycle. During interphase, microtubule arrays direct the transport of membranous vesicle and organelles. Moreover, by interacting with the intermediate filament and actomyosin networks, they also influence cell shape, polarity and motility 93,94 . During mitosis, microtubules are indispensable for the formation of the spindle apparatus. In higher plants and specialized animal cells, notably female germ cells, spindles can form in the absence of centrosomes 75 . However, in most dividing animal cells, centrosomes instruct the formation of a bipolar spindle, and they determine the positioning of the contractile ring during cytokinesis. Furthermore, recent studies indicate that centrosomes are also required for abscission, the final stage of cell division 95 , and for the subsequent G1 to S transition 6,76 . Appendages Longitudinal sections through (parental) centrioles Pericentriolar material (PCM) Cross-section through centriole, showing microtubule cylinder and appendages 0.2 �m � 2002 Nature Publishing Group 4 | NOVEMBER 2002 | VOLUME 2 www.nature.com/reviews/cancer REVIEWS chromosome segregation were also seen following overexpression of TACC and CEP135 (REFS 36,38),but not following overexpression of C-NAP1 (REF. 37). This illustrates that different types of PCM assemblies can exert diverse effects on centrosome function and cytoskeletal dynamics. For most tumours, the functional consequences of structural centrosomal abnormalities remain to be explored. A recent survey that was performed on differ- ent types of breast cancer failed to reveal a correlation idea comes from the demonstration that the overexpres- sion of certain PCM components (such as pericentrin, TACC, CEP135 and C-NAP1) in cultured cells gives rise to structural centrosomal abnormalities that closely resemble those seen in tumours 23,36?39 . Furthermore, the overexpression of pericentrin in pri- mary prostate epithelial cells reproduced several phe- notypic characteristics of prostate tumours, notably increased genomic instability and loss of cellular archi- tecture 23 . Adverse effects on spindle formation and Box 3 | The centrosome cycle The centrosome duplication cycle can be subdivided into several discrete steps (see figure). During mitosis, the centrosome at each pole of the mitotic spindle contains a pair of centrioles. These two centrioles usually display a conspicuous orthogonal orientation, indicating that they are tightly connected. At the end of mitosis, this orthogonal association is lost during a process that is referred to as centriole disorientation. This step might relate to the final separation (abscission) of the two incipient daughter cells 95 . In addition, it might be required for the subsequent duplication step 21 , or for the re- establishment of a linker structure between the two parental centrioles 96 . Centriole duplication then occurs during S phase. At the morphological level, this event is characterized by the formation of procentrioles at the proximal end of each parental centriole. So, duplication is semi-conservative from the perspective of the whole centrosome, but conservative from the perspective of the centriole 97 . How centriole duplication is brought about remains a mystery, but the recent establishment of in vitro assays for centrosome duplication might hopefully provide new opportunities for studying this fundamental problem 98,99 . Procentrioles then elongate until they reach their maximal length, but, importantly, the two centriole doublets continue to function as a single microtubule-organizing centre until late G2. At the G2?M transition, centrosome maturation occurs. This process involves the exchange of several PCM components and culminates in the recruitment of additional ?-tubulin ring complexes ? a prerequisite for increased microtubule-nucleating activity. In response to the activation of microtubule-dependent motor proteins, centrosomes then separate from each other and instruct the formation of the two spindle poles. As a result, each incipient daughter cell again inherits one centrosome. + Centrosome maturation Centriole elongation Centriole duplication Centrosome separation The centrosome cycle M G1 S G2 Centriole disorientation � 2002 Nature Publishing Group NATURE REVIEWS | CANCER VOLUME 2 | NOVEMBER 2002 | 5 REVIEWS of the uterine cervix, breast and prostate, centroso- mal abnormalities are common not only in highly advanced, invasive cancers, but can also be detected in low-grade tumours and in situ carcinomas 23,34,35 . Similarly, centrosome defects were found to represent an early event in the evolution of malignant pheno- types in ORGANOTYPIC CULTURE and animal models 42?45 . These studies support the hypothesis that centroso- mal abnormalities constitute an important cause of chromosomal instability, rather than a secondary consequence of late-stage tumorigenesis. So, it is attractive to evaluate the utility of centrosomal mark- ers as prognostic indicators for the development of aggressive forms of cancer 23,27,28 . Origins of centrosome aberrations? Supernumerary centrosomes can arise through funda- mentally distinct mechanisms 2,10,46 . As outlined in FIG. 3a, they might reflect several rounds of centrosome dupli- cation within the same cell cycle (model I); however, as centrosome duplication requires several hours, this mechanism is expected to depend on a substantial delay in cell-cycle progression. Certain tumour-derived cell lines (for example, U2OS osteosarcoma cells) can indeed be induced to undergo several rounds of centro- some reduplication in vitro, provided that DNA repli- cation is arrested for many hours by drugs such as hydroxyurea or aphidicolin 19,47 . Interestingly, however, other tumour-derived cell lines (for example, HeLa) arrest both centrosome duplication and DNA replica- tion in response to the same drugs 48 . What determines this response is an important unresolved question 49 .In particular, it would be important to know whether cells normally possess a mechanism that limits centrosome duplication to once per cell cycle. If such a mechanism exists, one might expect it to be mutated in those tumour cell lines that re-duplicate centrosomes follow- ing inhibition of S-phase progression. A second plausible scenario for the generation of cells with supernumerary centrosomes invokes an aborted cell division (FIG. 3a; model II). Cell-division fail- ure can have several distinct primary causes, including between microtubule-nucleation capacity and centroso- mal abnormalities 34 . This supports the view that the microtubule-nucleation ability of structurally aberrant centrosomes might be either reduced or enhanced, depending on the identity and modification state of the overexpressed PCM components. As different types of structural centrosomal abnormalities influence cellular properties in different ways, it might be rewarding to search for correlations between the overexpression of par- ticular centrosomal proteins and clinical parameters that are associated with the corresponding tumours. Such cor- relations might constitute valuable prognostic indicators. Chromosomal aberrations are particularly com- mon in advanced, highly invasive cancers, and are increasingly used as a prognostic marker for tumour progression 40,41 . A similar situation might hold true for centrosomal abnormalities. As shown for lesions ORGANOTYPIC CULTURE The in vitro maintenance and growth of tissue explants and multicellular cultures that mimic cell interactions within tissues. Figure 1 | The cell cycle: a tale of two cycles. A schematic comparison of a | the centrosome cycle and b | the chromosome cycle. Both the centrosome and the complete genome need to be duplicated once, and only once, in every cell cycle. Loss of coordination between the two cycles inevitably leads to chromosome missegregation or changes in ploidy. G1 a The centrosome cycle b The chromosome cycle M S G2 G2 G1 M S Figure 2 | Centrosomal abnormalities in human tumours. a,b | Normal (a) and tumour (b) colon tissues from the same patient were stained with antibodies against cytokeratin 20 (red) to identify epithelial cells, pericentrin (green) to label centrosomes, and with Hoechst 33342 (blue) to label nuclei. Normal crypt epithelial cells (a) have apical centrosomes and basal nuclei, with approximately one centrosome per nucleus. The aneuploid tumour (b) has amplified centrosomes that are larger and more numerous than those in the normal tissue. Moreover, cellular organization is disturbed. Bar denotes 10 �m. Images kindly provided by Vivian Negron and Wilma Lingle (Mayo Clinic, Rochester, Minnesota, USA). c?e | A human prostate tumour (d,e) and adjacent tissue (c) were sectioned, stained for ?-tubulin (brown), and processed for immunoperoxidase. Nuclei were stained with hematoxylin (purple). Compared with the bipolar spindle of a normal mitotic cell (c), the spindles in many dividing tumour cells are multipolar, with much larger ?-tubulin-positive poles (d,e). Bar denotes 10 �m. Images kindly provided by German Pihan and Stephen Doxsey (University of Massachusetts, Worcester, Massachusetts, USA). ab cde � 2002 Nature Publishing Group 6 | NOVEMBER 2002 | VOLUME 2 www.nature.com/reviews/cancer REVIEWS of the primary cause of the failed division, the resulting G1 cell will contain not only twice the normal amount of DNA, but also twice the normal number of centro- somes. As discussed further below, the subsequent fate of such a cell ? cell-cycle arrest, apoptosis or re-entry into S phase ? seems to depend on the presence or absence of a functional p53 checkpoint pathway. A third scenario for the generation of supernumer- ary centrosomes is based on cell fusion (FIG. 3a;model III). Although this mechanism has not yet received much attention in the context of tumorigenesis, fusion- induced centrosome amplification has been observed following ectopic expression of the RAD6 ubiquitin- conjugating enzyme in human breast epithelial cells 51 . Clearly, this mechanism could be important in cells that have been infected by viruses with fusogenic activities. A fourth possible mechanism, not illustrated in FIG. 3, relates to the de novo formation of centrioles. The exis- tence of pathways for de novo assembly of centrioles and basal bodies was long thought to be restricted to highly specialized cell types (such as multiciliated epithelial cells), but a recent study on the consequences of centrosome ablation by laser microsurgery indicates that many vertebrate somatic cells are also able to form centrioles de novo (REF. 109). This provocative finding indicates not only that mechanisms allowing the de novo formation of centrioles are more widespread than previously suspected, but also that these mecha- nisms are normally suppressed by existing centrioles. This implies that unscheduled activation of de novo assembly pathways could contribute to excessive centri- ole numbers in cancer cells. The four described mechanisms for the generation of supernumerary centrosomes are not mutually exclu- sive, and the available evidence is not sufficient to definitively favour one mechanism over another. However, supernumerary centrosomes have been observed in response to deregulation (either knockout or overexpression) of several gene products that are implicated in human cancer (TABLE 1). It is striking that few, if any, of these genes are ostensibly involved in the regulation of the centrosome cycle. Although it is possi- ble that future studies will reveal a link between these genes and centrosome duplication, at present it seems more straightforward to postulate that the deregulation of the genes listed in TABLE 1 could interfere with cell- cycle progression. A number of distinct primary defects, such as the persistence of unrepaired DNA damage, errors in chromosome structure or deregu- lated mitotic progression, could interfere with success- ful cell division. This would then lead to a similar ter- minal phenotype that is characterized by an increase in both chromosome and centrosome numbers (FIG. 3a; model II). So, aborted mitoses might constitute an important primary cause of numerical centrosome aberrations in tumours 48 . One way to further explore the relative importance of different mechanisms for generating extra copies of centrosomes in tumours is to examine the ploidy of the cells that show these abnormalities. Whereas a mecha- nism based on overduplication should initially produce the persistence of unrepaired DNA damage or the deregulation of pathways that coordinate mitotic pro- gression and cytokinesis. Another important reason for aborting division relates to the spindle-assembly check- point. This checkpoint delays the separation of sister chromatids (anaphase onset) until all chromosomes have undergone correct bipolar attachment on the spin- dle apparatus 50 . Malfunction of this checkpoint, or adaptation to a prolonged checkpoint arrest (mitotic slippage), will result in aberrant mitotic exit. Regardless Model I: overduplication Model II: aborted division Model III: cell fusion a b Centrosome overduplication (Model I) Diploid genome Aberrant mitosis Chromosome missegregation Aborted division (Model II) Centrosome amplification Tetraploidization Polyploid genome Chromosome missegregation Figure 3 | Centrosome amplification. a | Mechanisms of centrosome amplification. Three plausible models for the generation of supernumerary centrosomes. A fourth model ? de novo assembly of centrioles ? is not indicated. For the sake of simplicity, all supernumerary centrosomes are shown in clusters, although scattered distributions might also be generated. Model I: deregulated centrosome duplication. Supernumerary centrosomes arise through several rounds of duplication within a single S phase. Model II: failure to complete cell division. As a result of an aborted mitosis, a tetraploid (or near-tetraploid) cell contains two centrosomes that are already in G1. Model III: cell fusion. Depending on the cell-cycle stages of the fusion partners, the products of such fusions will display different centrosome/genome ratios. Note that the products of fusion and aborted cell division will first be multinucleated, but often form single polyploid nuclei after subsequent mitoses. b | Centrosome amplification and ploidy. Centrosome overduplication during a prolonged S phase will give rise to supernumerary centrosomes in a diploid cell. In striking contrast, an aborted mitosis will generate supernumerary centrosomes that are concomitant with an increase in ploidy. Although supernumerary centrosomes are expected to cause chromosome missegregation in all dividing cells (regardless of ploidy), the likelihood of generating viable, potentially harmful progeny (in the form of hyperdiploid cells) is enhanced when segregating chromosomes of a tetraploid rather than a diploid genome. So, the combination of supernumerary centrosomes with tetraploidy sets the stage for chromosome missegration and chromosomal instability. � 2002 Nature Publishing Group NATURE REVIEWS | CANCER VOLUME 2 | NOVEMBER 2002 | 7 REVIEWS Remarkably, both of these phenotypes were exacerbated in a p53 ?/? background 48 . Therefore, centrosome ampli- fication in p53 ?/? cells does not necessarily imply a role for p53 in the regulation of centrosome duplication, but instead might reflect the involvement of a p53-dependent checkpoint in the elimination of cells that emerge from aborted divisions 48,58?63 . Several additional arguments support the view that the absence of p53 favours the emergence of supernu- merary centrosomes through an indirect, checkpoint- related mechanism. Centrosome amplification is not an inevitable consequence of p53 deficiency in vivo 64 , indi- cating that the elimination of p53 is not in itself suffi- cient to deregulate the centrosome cycle. Furthermore, the targeted inactivation of p53 in diploid human cells did not cause aneuploidy, although it favoured the for- mation of tetraploid cells 65 . It is also interesting to con- sider the generation of supernumerary centrosomes by the HPV-encoded oncoproteins, E6 and E7 (REF. 30). Whereas E7 primarily targets the retinoblastoma gene product (see below), E6 causes the ubiquitin-dependent degradation of p53. Yet, overexpression of E6 in primary human keratinocytes failed to exert a rapid effect on centrosome duplication, but instead produced centro- some amplification in conjunction with multinucle- ation 66,67 . Similarly, when expressed in a lung cancer cell line, E6 did not cause chromosomal instability unless mitotic-spindle formation was transiently abrogated 68 . Centrosome amplification and the RB pathway. Con- sidering that both DNA replication and centrosome duplication are regulated through the RB pathway, it is attractive to speculate that mutational inactivation of this pathway ? a common event in human tumours ? could set the stage for centrosome overduplication 19 . However, although the loss of RB function might create permissive conditions for centrosome overduplication, supernumerary centrosomes in (near-) diploid cells, numerical centrosome aberrations arising through aborted mitoses should be accompanied by an approxi- mate doubling of chromosome content (FIG. 3b). Remarkably, a growing body of evidence indicates that tetraploidization frequently precedes aneuploidy in solid human tumours 52?54 . This is in line with a model in which aborted divisions give rise simultaneously to tetraploidy and supernumerary centrosomes (FIG. 3b). Centrosome amplification and the p53 pathway. Much of the renewed interest in a possible link between centro- somal abnormalities and tumorigenesis was stimulated by the demonstration that the loss of the p53 tumour suppressor results in supernumerary centrosomes 55 . p53 is a transcription factor that causes cell-cycle arrest or apoptosis in response to DNA damage. A significant pro- portion (~10?30%) of p53 ?/? mouse embryo fibroblasts cultured in vitro have supernumerary centrosomes, and increased centrosome numbers have also been observed in mouse models that have impaired p53 pathways 13,29,56 . Furthermore, supernumerary centrosomes have been described following deletion of two p53 targets ? the CDK2 inhibitor WAF1 (also known as p21) and GADD45 ? and following overexpression of MDM2/ HDM2, a ubiquitin-ligase and negative regulator of p53 (TABLE 1). So, it is well established that the loss of a func- tional p53 pathway favours the appearance of cells with supernumerary centrosomes, both in tissue culture and in tumours 29,55 . But what is the link between p53 function and the centrosome duplication cycle? It has been argued that loss of p53 causes centro- some overduplication within a single S phase 55,57 . However, a recent study favours an alternative interpre- tation 48 . Overexpression of Aurora-A and other mitotic kinases was shown to cause centrosome amplification by interfering with the successful completion of cell division, giving rise to cells that were characterized by both centrosome amplification and polyploidy. Table 1 | Genes implicated in centrosome amplification* Gene Proposed function References p53 pathway p53 (knockout) Cell-cycle checkpoint 55 WAF1 (antisense) p53 target/CDK inhibitor 100 Gadd45 (knockout) p53 target/checkpoint 101 Mdm2 (overexpression) Ubiquitin-ligase for p53 29 DNA-repair pathway ATR (gene duplication) Protein kinase/checkpoint 102 Brca1 (knockout) DNA recombination 103 Brca2 (knockout) DNA recombination 104 XRCC2/3 (mutation) Recombination/repair 71 Protein degradation Tsg101 (knockout) Ubiquitylation 105 Skp2 (knockout) Ubiquitylation 106 RAD6 (overexpression) Ubiquitylation/DNA repair 51 Mitosis Aurora-A (overexpression) Protein kinase 107 Survivin (antisense) Cytokinesis? 108 *Note that in some cases (for example, XRCC2/3), centrosome ?amplification? might primarily reflect fragmentation, rather than a true numerical aberration. Figure 4 | The RB pathway and centrosome duplication. This speculative model proposes that overduplication of centrosomes within the same cell cycle depends on at least two (and possibly three) events. a | The mutational inactivation of the RB pathway might create a permissive environment for centrosome overduplication. b | A delay in S-phase progression could then provide the time that is required for several rounds of centriole duplication. This could result, for instance, from activation of an intra S-phase checkpoint (for example, in response to chemotherapy or ?-irradiation). c | If cells possess a pathway that normally prevents centrosome reduplication within the same cell cycle, this block to reduplication would also have to be relieved. Different human tumour cells might or might not possess such a pathway, and this could influence the outcome of therapeutic intervention. a c b RB pathway Block to reduplication? S phaseG1 G2 Prolongation of S phase � 2002 Nature Publishing Group 8 | NOVEMBER 2002 | VOLUME 2 www.nature.com/reviews/cancer REVIEWS theoretically arise as a result of failed centrosome dupli- cation, but most frequently reflect defects in centrosome separation. Analyses of human tumours have revealed a strong positive correlation between centrosomal abnormalities and chromosome number aberra- tions 12,24,26,30,32,34,45 . However, correlative evidence does not establish causality, and so far it has not been pos- sible to directly show that centrosome abnormalities constitute a frequent primary cause of aneuploidy. Yet there is no doubt that the two phenotypes enhance each other; centrosome aberrations will foster chro- mosome missegregation, regardless of whether they arose through deregulation of the centrosome cycle or as a consequence of another primary event. Even those supernumerary centrosomes that result from aborted mitoses do not merely constitute innocent bystanders (and potentially useful markers) of tetraploidization. Instead, the presence of extra copies of centrosomes in tetraploid cells inevitably sets the stage for enhanced chromosome missegregation 48 . Centrosome aberrations and tissue architecture. So far, most studies on the link between centrosomes and tumours have focused on the impact of centrosomal abnormalities on the stability of the genome. But con- sidering that loss of cell polarity and tissue architecture is an important aspect of tumour progression, other facets of centrosome function should not be neglected 23,34 . In particular, centrosome positioning determines both the orientation of the cleavage plane and the (a)symmetry of cell division, two parameters that are absolutely crucial in epithelia and stem-cell compartments 72?74 . Investigations into the impact of centrosomal abnormalities on tissue organization might therefore prove rewarding and will hopefully lead to a better understanding of the cytoskeletal changes that underlie the acquisition of an invasive phenotype. Compensation for extra centrosomes? Dysfunctional or supernumerary centrosomes will either impede cell division or cause multipolar divisions, which will most frequently lead to mitotic catastrophe. Neither of these phenotypes would be expected to favour the clonal expansion of a tumour cell. So why are centroso- mal abnormalities so common in tumours? Why are they not eliminated by negative selection? Two observations might, together, provide an answer to these questions. First, observation of established tumour cell lines in vitro reveals that cells with supernumerary centro- somes are usually present at comparatively low levels, amounting to just a small proportion (~1?15%) of the total cell population. These levels are fairly constant and almost certainly reflect a steady-state situation, which is determined by the rate at which cells with supernumer- ary centrosomes arise de novo, and the rate at which they die, due to either elimination by cell-cycle check- points or mitotic catastrophe. So, under in vitro culture conditions, the acquisition of extra copies of centro- somes generally constitutes a disadvantage. This situa- tion is likely to be different in vivo, where the fraction of this alone is clearly not sufficient (FIG. 4). Several rounds of centrosome duplication could only occur in RB-defi- cient cells if S phase was sufficiently prolonged, for instance, in response to activation of a DNA-damage checkpoint. Studies on the E7 oncoprotein of HPV seem consistent with a role of the RB pathway in restraining centrosome duplication 43,67 . So, the available evidence indicates that the E6 and E7 oncoproteins use distinct mechanisms for generat- ing centrosome amplification and chromosomal insta- bility 31 . By interfering with p53, E6 might favour the survival of cells exiting aberrant mitoses. E7 might inactivate RB and thereby set the stage for centrosome overduplication. If these interpretations are correct, then the two cooperating oncoproteins of high-risk HPV would trigger two major mechanisms for centrosome amplification. Consequences of centrosome aberrations Centrosome amplification and genetic instability. Because centrosomes have a dominant role in the formation of the mitotic-spindle apparatus, centrosomal abnor- malities will almost inevitably cause the formation of abnormal spindles, with dire consequences for the integrity of the genome. Multipolar spindles are indeed common in tumours (FIG. 2). They often reflect excessive centrosome numbers 46,69,70 , but acentriolar bodies can also occasionally act as spindle poles. Such bodies can arise through the assembly of overexpressed PCM com- ponents (as discussed above), or through centrosome fragmentation 71 . So, caution should be exerted when interpreting data that are based exclusively on the use of antibodies against components of the PCM; definitive evidence for centrosome amplification will generally require the visualization of centrioles. Monopolar spin- dles have also been described in tumours. These could Figure 5 | Coalescence of centrosomes to two poles in a mitotic neuroblastoma cell. Centrosomes in a mitotic N115 neuroblastoma cell were stained with antibodies against centrin (orange). Microtubules were counter-stained with antibodies against ?-tubulin (green) and DNA was labelled with Hoechst dye (blue). Note that most centrosomes have assembled to two broad spindle poles (boxed), but at least one centrosome (arrowhead) has not (yet) coalesced. Image kindly provided by Martina Casenghi (Max Planck Institute of Biochemistry, Martinsried, Germany). � 2002 Nature Publishing Group NATURE REVIEWS | CANCER VOLUME 2 | NOVEMBER 2002 | 9 REVIEWS of cell-cycle deregulation. This by no means diminishes the importance of centrosome aberrations for tumori- genesis, as supernumerary centrosomes will foster chromosomal instability regardless of their origin. Whether this then implies that centrosome aberrations have a causal role in the development of cancer depends on the extent to which genomic instability is important for the outgrowth of malignant and drug- resistant cells 80 . Finally, it is important to emphasize that aberrant centrosomes could promote progression of cancers to more malignant forms not only through their impact on the stability of the genome, but also through their influence on tissue architecture. The explo- ration of this intriguing possibility has barely begun. From a clinical perspective, centrosomal abnormali- ties are interesting for their potential use as diagnostic or prognostic markers 23 . Furthermore, it might prove rewarding to explore centrosome-related processes for their potential exploitation in therapeutic approaches. First, cells bearing aberrant centrosomes might offer therapeutic windows for drugs that are directed at centrosomes or microtubules 33 . Second, if alternative mechanisms for bipolar spindle formation were indeed upregulated in tumours with supernumerary centro- somes, such mechanisms would constitute attractive targets for therapeutic intervention. Third, if regulatory pathways normally prevent the reduplication of centro- somes within the same cell cycle, the inactivation of such pathways should favour centrosome overduplication following treatment of tumours with agents that extend the duration of S phase. Conceivably, this could increase the frequency of multipolar mitoses to a level that is no longer compatible with the survival of progeny 81 . So, where do we go from here? One important task for the future will be to explore the relative contribu- tion of different mechanisms to the generation of supernumerary centrosomes in the course of tumour development. This will undoubtedly require the establishment of appropriate animal models. Another important goal is to elucidate the regulatory circuits that control the centrosome cycle. A detailed molecu- lar understanding of the links that coordinate the duplication and segregation of centrosomes with the propagation of the genome will not only lead to a bet- ter appreciation of the role of centrosomes in human cancer, but might also provide a rational basis for the development of centrosome-related diagnostic or therapeutic applications. cells with centrosome abnormalities progressively increases with advancing tumour stages 23,34,45 . Although most of the multipolar divisions that occur in tumours probably reflect non-productive events, an occasional division might give rise to progeny with a genetic consti- tution that favours survival in a changing physiological environment. Selective pressure might arise, for instance, through increasing hypoxia or nutritional deprivation in a growing tumour mass, or through the presence of a chemotherapeutic drug. Second, tumour cells might divide successfully in spite of supernumerary centrosomes because of alterna- tive, centrosome-independent mechanisms for bipolar spindle formation 75,76 . This point is illustrated best by neuroblastoma-derived cell lines that harbour large numbers of centrioles, and yet undergo mostly bipolar divisions, with often unequal numbers of centrioles coa- lescing at the two poles 77?79 (FIG. 5). Considering that not all supernumerary centrosomes (or acentriolar bodies) of tumour cells are necessarily equivalent 33 , it is possible that only two pairs of centrioles are actually functional. Alternatively, however, it is attractive to speculate that certain tumours might compensate for the presence of supernumerary centrosomes by re-expressing or upreg- ulating genes that code for important components of an alternative, centrosome-independent pathway for spin- dle formation. As a result, supernumerary centrosomes would be forced to coalesce into two spindle poles 46,69 . This would then allow the tumour to expand through binary divisions, while at the same time maintaining genetic instability through occasional multipolar divi- sions. If this hypothesis were correct, upregulated alter- native mechanisms for bipolar spindle formation would constitute attractive drug targets. Conclusions and future directions The past several years have seen a surge of renewed interest in the hypothesis that centrosomal abnormali- ties might contribute to the development of cancer 10 . As yet, there is no genetic evidence to indicate that cen- trosomal abnormalities constitute a frequent cause of tumour initiation. However, centrosomal abnormalities are observed in early, pre-cancerous lesions, which sup- ports the view that they fuel tumour progression. In principle, supernumerary centrosomes in tumours could arise through deregulation of the centrosome cycle. However, recent evidence indicates that they might frequently constitute a secondary consequence 1. Bornens, M. Centrosome composition and microtubule anchoring mechanisms. Curr. Opin. Cell Biol. 14, 25?34 (2002). 2. Doxsey, S. Re-evaluating centrosome function. Nature Rev. Mol. Cell Biol. 2, 688?698 (2001). 3. Lange, B. M. H. Integration of the centrosome in cell cycle control, stress response and signal transduction pathways. Curr. Opin. Cell Biol. 14, 35?43 (2002). 4. Hinchcliffe, E. H. & Sluder, G. ?It takes two to tango?: understanding how centrosome duplication is regulated throughout the cell cycle. Genes Dev. 15, 1167?1181 (2001). 5. Meraldi, P. & Nigg, E. A. The centrosome cycle. FEBS Lett. 521, 9?13 (2002). 6. Hinchcliffe, E. H., Miller, F. J., Cham, M., Khodjakov, A. & Sluder, G. Requirement of a centrosomal activity for cell cycle progression through G1 into S phase. Science 291, 1547?1550 (2001). Cell microsurgery is used to study the role of the centrosome during cell-cycle progression. This provocative study points to a crucial centrosome- related function at the G1?S transition. 7. Pines, J. Four-dimensional control of the cell cycle. Nature Cell Biol. 1, E73?E79 (1999). 8. Rieder, C. L., Faruki, S. & Khodjakov,A. The centrosome in vertebrates: more than a microtubule-organizing center. Trends Cell Biol. 11, 413?419 (2001). 9. Sibon, O. C., Kelkar, A., Lemstra, W. & Theurkauf, W. E. DNA-replication/DNA-damage-dependent centrosome inactivation in Drosophila embryos. Nature Cell Biol. 2, 90?95 (2000). 10. Boveri, Th. Zur Frage der Entstehung maligner Tumoren, 1914 (English Translation: The Origin of Malignant Tumors, Williams and Wilkins, Baltimore, Maryland, 1929). A summary of the pioneering experiments that provide the foundations for our current thinking about the role of the centrosome in tumorigenesis. A lucid and prophetic treatise. 11. Lingle, W. L., Lutz, W. H., Ingle, J. N., Maihle, N. J. & Salisbury, J. L. Centrosome hypertrophy in human breast tumors: implications for genomic stability and cell polarity. Proc. Natl Acad. Sci. USA 95, 2950?2955 (1998). 12. Pihan, G. A. et al. Centrosome defects and genetic instability in malignant tumors. Cancer Res. 58, 3974?3985 (1998). 13. Weber, R. G. et al. Centrosome amplification as a possible mechanism for numerical chromosome aberrations in � 2002 Nature Publishing Group cerebral primitive neuroectodermal tumors with TP53 mutations. Cytogenet. Cell Genet. 83, 266?269 (1998). 14. Zheng, Y., Wong, M. L., Alberts, B. & Mitchison, T. Nucleation of microtubule assembly by a ?-tubulin- containing ring complex. Nature 378, 578?583 (1995). 15. Simerly, C. et al. The paternal inheritance of the centrosome, the cell?s microtubule-organizing center, in humans, and the implications for infertility. Nature Med. 1, 47?52 (1995). 16. Freed, E. et al. Components of an SCF ubiquitin ligase localize to the centrosome and regulate the centrosome duplication cycle. Genes Dev. 13, 2242?2257 (1999). 17. Wojcik, E. J., Glover, D. M. & Hays, T. S. The SCF ubiquitin ligase protein slimb regulates centrosome duplication in Drosophila. Curr. Biol. 10, 1131?1134 (2000). 18. Sluder, G. & Hinchcliffe, E. H. The coordination of centrosome reproduction with nuclear events during the cell cycle. Curr. Top. Dev. Biol. 49, 267?289 (2000). 19. Meraldi, P., Lukas, J., Fry, A. M., Bartek, J. & Nigg, E. A. Centrosome duplication in mammalian somatic cells requires E2F and Cdk2-cyclin A. Nature Cell Biol. 1, 88?93 (1999). 20. Hinchcliffe, E. H., Li, C., Thompson, E. A., Maller, J. L. & Sluder, G. Requirement of Cdk2?cyclin E activity for repeated centrosome reproduction in Xenopus egg extracts. Science 283, 851?854 (1999). 21. Lacey, K. R., Jackson, P. K. & Stearns, T. Cyclin-dependent kinase control of centrosome duplication. Proc. Natl Acad. Sci. USA 96, 2817?2822 (1999). 22. Matsumoto, Y., Hayashi, K. & Nishida, E. Cyclin-dependent kinase 2 (Cdk2) is required for centrosome duplication in mammalian cells. Curr. Biol. 9, 429?432 (1999). 23. Pihan, G. A. et al. Centrosome defects can account for cellular and genetic changes that characterize prostate cancer progression. Cancer Res. 61, 2212?2219 (2001). A very careful study on the relationship between centrosome anomalies and prostate tumours at various stages of progression. 24. Lingle, W. L. & Salisbury, J. L. Altered centrosome structure is associated with abnormal mitoses in human breast tumors. Am. J. Pathol. 155, 1941?1951 (1999). Detailed ultrastructural analysis of centrosomes in human breast tumours, revealing a strong association between the excess of pericentriolar material and abnormal mitoses. 25. Sato, N. et al. Centrosome abnormalities in pancreatic ductal carcinoma. Clin. Cancer Res. 5, 963?970 (1999). 26. Sato, N. et al. Correlation between centrosome abnormalities and chromosomal instability in human pancreatic cancer cells. Cancer Genet. Cytogenet. 126, 13?19 (2001). 27. Kuo, K. K. et al. Centrosome abnormalities in human carcinomas of the gallbladder and intrahepatic and extrahepatic bile ducts. Hepatology 31, 59?64 (2000). 28. Gustafson, L. M. et al. Centrosome hyperamplification in head and neck squamous cell carcinoma: a potential phenotypic marker of tumor aggressiveness. Laryngoscope 110, 1798?1801 (2000). 29. Carroll, P. E. et al. Centrosome hyperamplification in human cancer: chromosome instability induced by p53 mutation and/or Mdm2 overexpression. Oncogene 18, 1935?1944 (1999). 30. Duensing, S. & Munger, K. Centrosome abnormalities, genomic instability and carcinogenic progression. Biochim. Biophys. Acta 1471, M81?M88 (2001). 31. Duensing, S. & Muenger, K. Human papillomaviruses and centrosome dupliation errors: modeling the origins of genomic instability. Oncogene 21, 6241?6248 (2002). 32. Skyldberg, B. et al. Human papillomavirus infection, centrosome aberration, and genetic stability in cervical lesions. Mod. Pathol. 14, 279?284 (2001). 33. Ghadimi, B. M. et al. Centrosome amplification and instability occurs exclusively in aneuploid, but not in diploid colorectal cancer cell lines, and correlates with numerical chromosomal aberrations. Genes Chromosom. Cancer 27, 183?190 (2000). 34. Lingle, W. L. et al. Centrosome amplification drives chromosomal instability in breast tumor development. Proc. Natl Acad. Sci. USA 99, 1978?1983 (2002). 35. Pihan, G. A., Wallace, J., Zhou, Y. & Doxsey, S. Centrosome abnormalities and chromosome instability occur together in precancerous lesions. Proc. Natl Acad. Sci. USA (in the press). 36. Gergely, F. et al. The TACC domain identifies a family of centrosomal proteins that can interact with microtubules. Proc. Natl Acad. Sci. USA 97, 14352?14357 (2000). 37. Mayor, T., Hacker, U., Stierhof, Y. D. & Nigg, E. A. The mechanism regulating the dissociation of the centrosomal protein C-Nap1 from mitotic spindle poles. J. Cell Sci. 115, 3275?3248 (2002). 38. Ohta, T. et al. Characterization of Cep135, a novel coiled- coil centrosomal protein involved in microtubule organization in mammalian cells. J. Cell Biol. 156, 87?100 (2002). 39. Purohit, A., Tynan, S. H., Vallee, R. & Doxsey, S. J. Direct interaction of pericentrin with cytoplasmic dynein light intermediate chain contributes to mitotic spindle organization. J. Cell Biol. 147, 481?492 (1999). 40. Mitelman, F. Recurrent chromosome aberrations in cancer. Mutat. Res. 462, 247?253 (2000). 41. Ried, T., Heselmeyer-Haddad, K., Blegen, H., Schrock, E. & Auer, G. Genomic changes defining the genesis, progression, and malignancy potential in solid human tumors: a phenotype/genotype correlation. Genes Chromosom. Cancer 25, 195?204 (1999). 42. Duensing, S. et al. Centrosome abnormalities and genomic instability by episomal expression of human papillomavirus type 16 in raft cultures of human keratinocytes. J. Virol. 75, 7712?7716 (2001). 43. Duensing, S., Duensing, A., Crum, C. P. & Munger, K. Human papillomavirus type 16 E7 oncoprotein-induced abnormal centrosome synthesis is an early event in the evolving malignant phenotype. Cancer Res. 61, 2356?2360 (2001). 44. Goepfert, T. M. et al. Centrosome amplification and overexpression of Aurora-A are early events in rat mammary carcinogenesis. Cancer Res. 62, 4115?4122 (2002). 45. Shono, M. et al. Stepwise progression of centrosome defects associated with local tumor growth and metastatic process of human pancreatic carcinoma cells transplanted orthotopically into nude mice. Lab. Invest. 81, 945?952 (2001). 46. Brinkley, B. R. Managing the centrosome numbers game: from chaos to stability in cancer cell division. Trends Cell Biol. 11, 18?21 (2001). 47. Balczon, R. et al. Dissociation of centrosome replication events from cycles of DNA synthesis and mitotic division in hydroxyurea-arrested Chinese hamster ovary cells. J. Cell Biol. 130, 105?115 (1995). 48. Meraldi, P., Honda, R. & Nigg, E. A. Aurora-A overexpression reveals tetraploidization as a major route to centrosome amplification in p53 ?/? cells. EMBO J. 21, 483?492 (2002). Overexpression of Aurora-A and other mitotic kinases is shown to cause centrosome amplification, not by deregulating centrosome duplication as previously thought but, instead, through defects in cytokinesis that result in transiently tetraploid cells. This phenotype is enhanced in p53 ?/? cells. 49. Balczon, R. C. Overexpression of cyclin A in human HeLa cells induces detachment of kinetochores and spindle pole/centrosome overproduction. Chromosoma 110, 381?392 (2001). 50. Millband, D. N., Campbell, L. & Hardwick, K. G. The awesome power of multiple model systems: interpreting the complex nature of spindle checkpoint signaling. Trends Cell Biol. 12, 205?209 (2002). 51. Shekhar, M. P., Lyakhovich, A., Visscher, D. W., Heng, H. & Kondrat, N. Rad6 overexpression induces multinucleation, centrosome amplification, abnormal mitosis, aneuploidy, and transformation. Cancer Res. 62, 2115?2124 (2002). 52. Galipeau, P. C. et al. 17p (p53) allelic losses, 4N (G2/tetraploid) populations, and progression to aneuploidy in Barrett?s esophagus. Proc. Natl Acad. Sci. USA 93, 7081?7084 (1996). 53. Shackney, S. E. et al. Model for the genetic evolution of human solid tumors. Cancer Res. 49, 3344?3354 (1989). 54. Southern, S. A., Evans, M. F. & Herrington, C. S. Basal cell tetrasomy in low-grade cervical squamous intraepithelial lesions infected with high-risk human papillomaviruses. Cancer Res. 57, 4210?4213 (1997). 55. Fukasawa, K., Choi, T., Kuriyama, R., Rulong, S. & Vande Woude, G. F. Abnormal centrosome amplification in the absence of p53. Science 271, 1744?1747 (1996). An influential study that contributed greatly to the revival of interest in the possible contribution of centrosome aberrations to carcinogenesis. 56. Levine, D. S., Sanchez, C. A., Rabinovitch, P. S. & Reid, B. J. Formation of the tetraploid intermediate is associated with the development of cells with more than four centrioles in the elastase-simian virus 40 tumor antigen transgenic mouse model of pancreatic cancer. Proc. Natl Acad. Sci. USA 88, 6427?6431 (1991). 57. Tarapore, P., Horn, H. F., Tokuyama, Y. & Fukasawa, K. Direct regulation of the centrosome duplication cycle by the p53-p21Waf1/Cip1 pathway. Oncogene 20, 3173?3184 (2001). 58. Andreassen, P. R., Lohez, O. D., Lacroix, F. B. & Margolis, R. L. Tetraploid state induces p53-dependent arrest of nontransformed mammalian cells in G1. Mol. Biol. Cell 12, 1315?1328 (2001). 59. Borel, F., Lohez, O. D., Lacroix, F. B. & Margolis, R. L. Multiple centrosomes arise from tetraploidy checkpoint failure and mitotic centrosome clusters in p53 and RB pocket protein-compromised cells. Proc. Natl Acad. Sci. USA. 99, 9819?9824 (2002). 60. Casenghi, M. et al. p53-independent apoptosis and p53- dependent block of DNA rereplication following mitotic spindle inhibition in human cells. Exp. Cell Res. 250, 339?350 (1999). 61. Khan, S. H. & Wahl, G. M. p53 and pRb prevent rereplication in response to microtubule inhibitors by mediating a reversible G1 arrest. Cancer Res. 58, 396?401 (1998). 62. Lanni, J. S. & Jacks, T. Characterization of the p53- dependent postmitotic checkpoint following spindle disruption. Mol. Cell. Biol. 18, 1055?1064 (1998). 63. Minn, A. J., Boise, L. H. & Thompson, C. B. Expression of Bcl-x L and loss of p53 can cooperate to overcome a cell cycle checkpoint induced by mitotic spindle damage. Genes Dev. 10, 2621?2631 (1996). 64. Goepfert, T. M. et al. Progesterone facilitates chromosome instability (aneuploidy) in p53 null normal mammary epithelial cells. FASEB J. 14, 2221?2229 (2000). 65. Bunz, F. et al. Targeted inactivation of p53 in human cells does not result in aneuploidy. Cancer Res. 62, 1129?1133 (2002). 66. Duensing, S. et al. The human papillomavirus type 16 E6 and E7 oncoproteins cooperate to induce mitotic defects and genomic instability by uncoupling centrosome duplication from the cell division cycle. Proc. Natl Acad. Sci. USA 97, 10002?10007 (2000). 67. Duensing, S., Duensing, A., Crum, C. P. & Munger, K. Human papillomavirus type 16 E7 oncoprotein-induced abnormal centrosome synthesis is an early event in the evolving malignant phenotype. Cancer Res. 61, 2356?2360 (2001). An interesting studying indicating that the two oncoproteins that are encoded by HPV-16 induce numerical centrosome aberrations by distinct mechanisms. 68. Haruki, N. et al. Persistent increase in chromosome instability in lung cancer: possible indirect involvement of p53 inactivation. Am. J. Pathol. 159, 1345?1352 (2001). 69. Lingle, W. L. & Salisbury, J. L. The role of the centrosome in the development of malignant tumors. Curr. Top. Dev. Biol. 49, 313?329 (2000). 70. Pihan, G. A. & Doxsey, S. J. The mitotic machinery as a source of genetic instability in cancer. Semin. Cancer Biol. 9, 289?302 (1999). 71. Griffin, C. S., Simpson, P. J., Wilson, C. R. & Thacker, J. Mammalian recombination-repair genes XRCC2 and XRCC3 promote correct chromosome segregation. Nature Cell Biol. 2, 757?761 (2000). 72. Knoblich, J. A. Asymmetric cell division during animal development. Nature Rev. Mol. Cell Biol. 2, 11?20 (2001). 73. Meads, T. & Schroer, T. A. Polarity and nucleation of microtubules in polarized epithelial cells. Cell Motil. Cytoskeleton 32, 273?288 (1995). 74. Reinsch, S. & Karsenti, E. Orientation of spindle axis and distribution of plasma membrane proteins during cell division in polarized MDCKII cells. J. Cell Biol. 126, 1509?1526 (1994). 75. Karsenti, E. & Vernos, I. The mitotic spindle: a self-made machine. Science 294, 543?547 (2001). 76. Khodjakov, A. & Rieder, C. L. Centrosomes enhance the fidelity of cytokinesis in vertebrates and are required for cell cycle progression. J. Cell Biol. 153, 237?242 (2001). This fascinating study examines ? using laser microsurgery ? the role of centrosomes during vertebrate cell-cycle progression. Following ablation of centrosomes, cells can still form bipolar mitotic spindles, but they frequently fail cytokinesis and they cannot reinitiate DNA synthesis. 77. Brinkley, B. R. et al. Tubulin assembly sites and the organization of cytoplasmic microtubules in cultured mammalian cells. J. Cell Biol. 90, 554?562 (1981). 78. Ring, D., Hubble, R. & Kirschner, M. Mitosis in a cell with multiple centrioles. J. Cell Biol. 94, 549?556 (1982). 79. Sharp, G. A., Weber, K. & Osborn, M. Centriole number and process formation in established neuroblastoma cells and primary dorsal root ganglion neurones. Eur. J. Cell Biol. 29, 97?103 (1982). 80. Jallepalli, P. V & Lengauer, C. Chromosome segregation and cancer: cutting through the mystery. Nature Rev. Cancer 1, 109?117 (2001). 81. Sato, N. et al. A possible role for centrosome overduplication in radiation-induced cell death. Oncogene 19, 5281?5290 (2000). 82. Mantel, C. et al. p21(cip-1/waf-1) deficiency causes 10 | NOVEMBER 2002 | VOLUME 2 www.nature.com/reviews/cancer REVIEWS � 2002 Nature Publishing Group NATURE REVIEWS | CANCER VOLUME 2 | NOVEMBER 2002 | 11 REVIEWS deformed nuclear architecture, centriole overduplication, polyploidy, and relaxed microtubule damage checkpoints in human hematopoietic cells. Blood 93, 1390?1398 (1999). 83. Hollander, M. C. et al. Genomic instability in Gadd45a- deficient mice. Nature Genet. 23, 176?184 (1999). 84. Smith, L. et al. Duplication of ATR inhibits MyoD, induces aneuploidy and eliminates radiation-induced G1 arrest. Nature Genet. 19, 39?46 (1998). 85. Xu, X. et al. Centrosome amplification and a defective G2- M cell cycle checkpoint induce genetic instability in BRCA1 exon 11 isoform-deficient cells. Mol. Cell 3, 389?395 (1999). 86. Tutt, A. et al. Absence of Brca2 causes genome instability by chromosome breakage and loss associated with centrosome amplification. Curr. Biol. 9, 1107?1110 (1999). 87. Xie, W., Li, L. & Cohen, S. N. Cell cycle-dependent subcellular localization of the TSG101 protein and mitotic and nuclear abnormalities associated with TSG101 deficiency. Proc. Natl Acad. Sci. USA 95, 1595?1600 (1998). 88. Nakayama, K. et al. Targeted disruption of Skp2 results in accumulation of cyclin E and p27(Kip1), polyploidy and centrosome overduplication. EMBO J. 19, 2069?2081 (2000). 89. Zhou, H. et al. Tumour amplified kinase STK15/BTAK induces centrosome amplification, aneuploidy and transformation. Nature Genet. 20, 189?193 (1998). 90. Li, F. et al. Pleiotropic cell-division defects and apoptosis induced by interference with survivin function. Nature Cell Biol. 1, 461?466 (1999). 91. Gray, J. W. & Collins, C. Genome changes and gene expression in human solid tumors. Carcinogenesis 21, 443?452 (2000). 92. Lengauer, C., Kinzler, K. W. & Vogelstein, B. Genetic instabilities in human cancers. Nature 396, 643?649 (1998). 93. Loeb, L. A. Mutator phenotype may be required for multistage carcinogenesis. Cancer Res. 51, 3075?3079 (1991). 94. Duesberg, P. & Rasnick, D. Aneuploidy, the somatic mutation that makes cancer a species of its own. Cell Motil. Cytoskeleton 47, 81?107 (2000). 95. Tomlinson, I., Sasieni, P. & Bodmer, W. How many mutations in a cancer? Am. J. Pathol. 160, 755?758 (2002). 96. Mitelman, F., Johansson, B. & Mertens, F. Catalog of Chromosome Aberrations in Cancer Vol. 2 (Wiley?Liss, New York, 1994). 97. Hartwell, L. H. & Kastan, M. B. Cell cycle control and cancer. Science 266, 1821?1828 (1994). 98. Nigg, E. A. Mitotic kinases as regulators of cell division and its checkpoints. Nature Rev. Mol. Cell Biol. 2, 21?32 (2001). 99. Bornens, M., Paintrand, M., Berges, J., Marty, M. C. & Karsenti, E. Structural and chemical characterization of isolated centrosomes. Cell Motil. Cytoskeleton 8, 238?249 (1987). 100. Paintrand, M., Moudjou, M., Delacroix, H. & Bornens, M. Centrosome organization and centriole architecture: their sensitivity to divalent cations. J. Struct. Biol. 108, 107?128 (1992). 101. Beisson, J. & Jerka-Dziadosz, M. Polarities of the centriolar structure: morphogenetic consequences. Biol. Cell 91, 367?378 (1999). 102. Fuchs, E. & Cleveland, D. W. A structural scaffolding of intermediate filaments in health and disease. Science 279, 514?519 (1998). 103. Schliwa, M., Euteneuer, U., Graf, R. & Ueda, M. Centrosomes, microtubules and cell migration. Biochem. Soc. Symp. 65, 223?231 (1999). 104. Piel, M., Nordberg, J., Euteneuer, U. & Bornens, M. Centrosome-dependent exit of cytokinesis in animal cells. Science 291, 1550?1553 (2001). The careful examination of the behaviour of centrosomes in living mitotic cells revealed a remarkable repositioning of the older centriole to the midbody, indicating a crucial role for a centrosome-dependent pathway at the final stage of cell division. 105. Mayor, T., Stierhof, Y. D., Tanaka, K., Fry, A. M. & Nigg, E. A. The centrosomal protein C-Nap1 is required for cell cycle- regulated centrosome cohesion. J. Cell Biol. 151, 837?846 (2000). 106. Kochanski, R. S. & Borisy, G. G. Mode of centriole duplication and distribution. J. Cell Biol. 110, 1599?1605 (1990). A very elegant study, demonstrating that tubulin incorporation into centrioles during each cell cycle is conservative, whereas the distribution of centrioles is semi-conservative. 107. Hinchcliffe, E. H. & Sluder, G. Centrosome reproduction in Xenopus lysates. Methods Cell Biol. 67, 269?287 (2001). 108. Piel, M. & Bornens, M. Centrosome reproduction in vitro: mammalian centrosomes in Xenopus lysates. Methods Cell Biol. 67, 289?304 (2001). 109. Khodjakov, A. et al. De novo formation of centrosomes in vertebrate cells arrested during S phase. J. Cell Biol. 158, 1171?1181 (2002). Acknowledgments I thank F. Barr, T. Mayer, P. Meraldi, H. Sillj� and C. Wilkinson for helpful comments on the manuscript and M. Bornens (Paris), M. Casenghi (Martinsried), S. Doxsey (Worcester) and W. Lingle (Rochester) for generously contributing material for the figures. My sincere apologies go to all authors whose primary work could not be cited due to space constraints. Online links DATABASES The following terms in this article are linked online to: Cancer.gov: http://www.cancer.gov/cancer_information/ bile duct cancer | brain cancer | breast cancer | cervical cancer | colon cancer | head and neck cancer | lung cancer | pancreatic cancer | prostate cancer GenBank: http://www.ncbi.nih.gov/Genbank/ E6 | E7 | HPV-16 LocusLink: http://www.ncbi.nlm.nih.gov/LocusLink/ AKAP450 | Aurora-A | CDK2 | C-NAP1 | GADD45 | kendrin | MDM2 | ninein | p53 | pericentrin | RAD6 | RB | WAF1 FURTHER INFORMATION Erich A. Nigg?s web site: http://www.biochem.mpg.de/nigg/home.shtml Mitelman Database of Chromosome Aberrations in Cancer: http://cgap.nci.nih.gov/Chromosomes/Mitelman Mitosis World: http://www.bio.unc.edu/faculty/salmon/lab/mitosis/mitosislabs.html Access to this interactive links box is free online. "
Add Content to Group
|
Bookmark
|
Keywords
|
Flag Inappropriate
share
Close
Digg
Facebook
MySpace
Google+
Comments
Close
Please Post Your Comment
*
The Comment you have entered exceeds the maximum length.
Submit
|
Cancel
*
Required
Comments
Please Post Your Comment
No comments yet.
Save Note
Note
View
Public
Private
Friends & Groups
Friends
Groups
Save
|
Cancel
|
Delete
Please provide your notes.
Next
|
Prev
|
Close
|
Edit
|
Delete
Genetics
Gene Inheritance and Transmission
Gene Expression and Regulation
Nucleic Acid Structure and Function
Chromosomes and Cytogenetics
Evolutionary Genetics
Population and Quantitative Genetics
Genomics
Genes and Disease
Genetics and Society
Cell Biology
Cell Origins and Metabolism
Proteins and Gene Expression
Subcellular Compartments
Cell Communication
Cell Cycle and Cell Division
Scientific Communication
Career Planning
Loading ...
Scitable Chat
Register
|
Sign In
Visual Browse
Close
Comments
CloseComments
Please Post Your Comment