The centrosome is the major microtubule organiser in animal cells, participating in a variety of processes from cell migration to cell division. Abnormalities in the number and structure of centrosomes are commonly found in many cancers. Recent studies highlight the role of a group of molecules in the control of centrosome number and shed light on the role of centrosomes in human disease. Here, we give a brief overview of developments in this field.

CENTROSOMES AND CENTRIOLES
The centriole is a eukaryotic organelle involved in a variety of processes, from the movement of spermatozoa, the sensing of light by our eyes and the cell division apparatus in most of our cells [Figure 1]. This structure is made of tubulin filaments, in addition to many other proteins. The centriole sets up the foundations for the skeleton of cilia and flagella, structures involved in sensation and movement of cells [Figure 1]. The centriole also participates in the formation of the centrosome, the major microtubule organising centre (MTOC) in animal cells, which coordinates cell
division, motility and polarity [Figure 1].

The centrosome is comprised of two centrioles, the older one, called mother, and a younger one, called daughter, that are surrounded by an electron-dense matrix, the pericentriolar material (PCM) [Figure 1]. Centriolar characteristics determine most properties of the centrosome, such as stability, capacity to reproduce, dynamicity and polarity. The PCM harbours molecules that anchor and nucleate cytoplasmic MTs in interphase and mitosis [1]. The structure of the centriole is amazingly conserved throughout the eukaryotic tree of life [1]. Centriole nine-fold symmetry is initially established by at least two molecules, SAS-6 and Bld10/CEP135, and centriole microtubules are then recruited, in part, through the action of a another molecule known as SAS-4/CPAP [1, 2].

CENTROSOME NUMBERS IN DIFFERENT CELL TYPES
The majority of multicellular eukaryotes have cells with either one or two centrosomes, each with two centrioles as described above. The number of centrioles in a cell is normally controlled through a canonical duplication cycle [Figure 2]. In this cycle, one new centriole forms orthogonally to each existing one, giving rise to two centrosomes. This happens in coordination with the DNA cycle in S phase. Thus, when the cell enters mitosis it is equipped with two centrosomes, each harbouring two centrioles, which nucleate and anchor microtubules into the mitotic spindle, the physical structure that allows equal segregation of the genetic material to the two daughter cells [Figure 2]. However, not all the cells follow the same arithmetics. Some cells have many centrioles and others have none. Many ciliated cells, such as those in vertebrate respiratory and reproductive tracts, can have 200-300 cilia per cell. This requires the generation of multiple centrioles, each forming one cilium. Other cells show centriolar loss. For example, centrioles disappear during oogenesis in several species. They are also lost during sperm maturation in the mouse (but not in most other mammals), and in muscle differentiation when myoblasts give rise to syncitial myotubes. Here, MTs are nucleated from sites associated with the nuclear envelope. Centrioles can appear de novo in cells that do not have these structures. That is the case in some parthenogenic species such as insects, during early stages of mouse embryonic development, and in some experimental conditions such as in human cells after
centrosome ablation [1].

CENTROSOME NUMBER, STRUCTURE AND HUMAN DISEASE
Control of centriole structure and number is very important as their dysfunction is associated with a variety of human diseases [3]. Aberrations of cilia cause several diseases, such as cystic kidneys, inversions of body symmetry, infertility and retinal degeneration [1]. Recent work also suggest a role for cilia in the reception of environmental signals and repression of tumourigenesis[4]. Centrosomes are also associated with tumourigenesis. It was Theodor Boveri, the zoologist who first described centrosomes at the end of the nineteenth century, who proposed abnormal centrosome numbers could lead to multipolar cell division. Abnormal divisions would cause the aneuploidy commonly observed in cancer [5].

Indeed, cells from many cancers have multiple and abnormal centrosomes, which often correlate with poor clinical outcome [6]. Moreover, numerical and structural centrosome aberrations are an early event in many cancers such as colorectal, uterine cervix, prostate and female breast carcinomas, and haematological malignancies [6]. However, it is difficult to unravel whether centrosome changes are the cause or consequence of cancer. Recent studies point to centrosome changes having the potential to be at the origin of certain cancers. For example, loss of p16INK4a, which is associated with cancer, can lead to centrosome amplification [7]. Moreover, changes in SAK levels, a molecule that triggers centrosome biogenesis (see below), can lead to changes in centrosome number in mice and flies, and lead to
tumourigenesis in both [8-9].

The same has been observed with other mutants in molecules involved in centrosome biogenesis and function in flies [8]. Although Boveri predicted those cancers would result from abnormal chromosome segregation, high chromosome
instability was not observed in the cases described above [8, 9].
Cancer cells might evolve mechanisms of avoiding abnormal cell division, since ultimately that could lead to cell death. One such mechanism is that extra centrosomes can ‘‘cluster’’ together during mitosis, forming bipolar mitotic spindles [10]. If there is not much aneuploidy, how do supranumerary centrosomes lead to cancer? These papers suggest an alternative model based on stem cell homeostasis [8, 9]. Since asymmetric cell division may be affected when centrosome number and function is changed, perhaps this could lead to an expansion of cells with higher proliferative potential in the stem cell compartment. Indeed, recent evidence suggests alterations in adult stem cells could be at the origin of certain tumour types.

CAUSES OF CHANGES IN CENTROSOME NUMBERS
Changes in centrosome number may arise due to a variety of causes. Changes may obviously occur if the centrosome cycle is uncoupled from the chromosome cycle, for example if DNA synthesis is delayed but centrosome biogenesis continues, or if centrosomes start to be formed de novo, with no control on their number [1]. Secondly, problems in cell division, such as aborted cytokinesis, also lead to an accumulation of centrosomes and chromosomes. Finally, cell fusion could also lead to a similar result [6]. Here we will discuss briefly the molecules that control the centrosome cycle and how some of those may be related to cancer. In mammalian and insect cells, the protein kinase SAK (also known as PLK4) is at the head of a pathway for centriole duplication, originally discovered in C. elegans, comprising several coiled-coil proteins, such as SAS-6 and SAS-4/CPAP [1]. While lack of SAK impairs centrosome duplication, its over-expression leads to abnormally high centrosome numbers [11-13]. SAK can trigger the formation of many centrioles even in circumstances where these structures were first absent [12], strongly suggesting that centriole assembly is a template-free process, locally triggered by a threshold of SAK activity [14]. Little is known about the control of SAK activity and how it is coordinated with the cell cycle, a process that we are interested in since changes in its levels are associated with cancer.

Three other molecules have been suggested to play a role in the coordination between centrosome duplication and DNA replication. It is thought that CDK2/cyclin E promotes both DNA replication and centrosome duplication in S phase, coordinating both processes [15, 16]. CDK1, a major player on G2/M transition may also have a direct or indirect role. For example in yeast and flies, increase of mitotic CDK
activity has an inhibitory role over spindle pole body and centrosome duplication, respectively [1, 17]. This may explain why there is no centrosome duplication in mitosis. Finally, separase, the protease that triggers sister chromatid separation, has also been shown to trigger mother and daughter centriole disengagement in mitosis, an event that allows centrioles to duplicate in the next cell cycle [18] [Figure 2].

Other molecules, linked to cancer, might also regulate centrosome number. This is the case for nucleophosmin (NPM), p53 and Brca1. Nucleophosmin is a phosphoprotein that shuttles between the nucleus and the centrosome. When present on the centrosome NPM can prevent centrosome duplication being dissociated from the centrosome after phosphorylation by CDK2/Cyclin E, which coincides with G1/S transition. Over-expression of a non-phosphorylatable form of nucleophosmin, which is bound to the centrosome, inhibits disengagement and consequent duplication, whereas its depletion results in centrosome amplification [19]. Recently, heterozygocity for nucleophosmin was shown to accelerate oncogenesis, namely in a haematological disorder with features of human
myelodysplastic syndromes [20].

Loss or mutational inactivation of p53 leads to centrosome amplification and out of schedule duplication[21]. It is yet not clear how p53 normally prevents centrosome amplification, although its been suggested that it may act directly through p21, a CDK2 inhibitor. Alternatively, p53 has also been implicated in maintaining the integrity of cellular checkpoints, which prevent centrosome amplification [21]. Brca1 is another tumour suppressor that has been linked to changes in centrosomes. Germline mutations of BRCA1 have been shown to predispose women to ovarian and breast cancer, associated with changes in centrosome number. Several activities related to Brca1 may link it to centrosome function. Clearly, it is associated with a variety of cell cycle checkpoints. Additionally, it was demonstrated that γ tubulin, a major component of centrosomes, is ubiquitinated by BRCA1 [23]. Since γ tubulin is also involved in centriole and centrosome formation, its regulation by Brca1 may be important for control of centrosome number. Clearly, in recent years there has been a great understanding of the list of molecules that regulate centrosome formation. Further studies are now needed to understand how those molecules may regulate the centrosome cycle and may lead to tumourigenesis.

CENTROSOME/CENTRIOLE ASSOCIATED DISEASE: FUTURE DEVELOPMENT AND DISEASE IMPLICATIONS
An understanding of how the centrosome cycle is regulated may provide markers for diagnosis and prognosis. Centrosome-targeting strategies may also lead to cancer treatments. Indeed, severe defects in mitosis, resulting from disrupted centrosome function, may lead to checkpoint arrest or to nonviable daughter cells [22]. For example, drugs that inhibit centrosome duplication or separation lead to the formation of monopolar spindles. Another class of possible drugs target centrosome clustering in cancer cells displaying multiple centrosomes. Recent studies have unravelled the role of different molecules necessary for clustering of the centrosomes, such as molecular motors [9, 10]. These studies have shown that targeting centrosome clustering has the advantage of only affecting cancer cells, since some of those motors are not essential in non-transformed cells [9, 10]. In conclusion, a molecular understanding of the centrosome cycle and its coordination with the chromosome cycle should help the development of centrosome-related diagnostic and therapeutic applications.

REFERENCES
1. Bettencourt-Dias M, Glover DM. Centrosome biogenesis and function: centrosomics brings new understanding. Nat Rev Mol Cell Biol 2007;8:451-463.
2. Strnad P, Gonczy P. Mechanisms of procentriole formation. Trends Cell Biol 2008;18:389-396.
3. Badano JL et al. The centrosome in human genetic disease. Nat Rev Genet 2005;6:194-205.
4. Plotnikova OV et al. Cell cycle-dependent ciliogenesis and cancer. Cancer Res 2008;68:2058-2061.
5. Boveri T. Concerning the Origin of Malignant Tumours by Theodor Boveri (1914). Translated and annotated by Henry Harris. Journal Cell Science 2008;121:1-84
6. Nigg EA. Origins and consequences of centrosome aberrations in human cancers. Int J Cancer 2006;119:2717-2723.
7. McDermott KM et al. p16(INK4a) prevents centrosome dysfunction and genomic instability in primary cells. PLoS Biol 2006;4:e51.
8. Castellanos E et al. Centrosome Dysfunction in Drosophila Neural Stem Cells Causes Tumors that Are Not due to Genome Instability. Curr Biol 2008;18:1209-14.
9. Basto R et al. Centrosome amplification can initiate tumorigenesis in flies. Cell 2008;133: 1032-1042.
10. Kwon M et al. Mechanisms to suppress multipolar divisions in cancer cells with extra centrosomes. Genes Dev 2008;22(16):2189-203.
11. Kleylein-Sohn J et al. Plk4-induced centriole biogenesis in human cells. Dev Cell 2007;13:190-202.
12. Rodrigues-Martins A et al. Revisiting the role of the mother centriole in centriole biogenesis. Science 2007;316:1046-1050.
13. Bettencourt-Dias M et al. SAK/PLK4 Is Required for Centriole Duplication and Flagella Development. Curr Biol 2005;15: 2199-2207.
14. Rodrigues-Martins A et al. From centriole biogenesis to cellular function: Centrioles are essential for cell division at critical developmental stages. Cell Cycle 2008;7:1-16.
15. Meraldi P et al. Centrosome duplication in mammalian somatic cells requires E2F and Cdk2-cyclin A. Nat Cell Biol 1999;1:88-93.
16. Hinchcliffe EH et al. Requirement of Cdk2-cyclin E activity for repeated centrosome reproduction in Xenopus egg extracts. Science 1999;283:851-854.
17. Simmons Kovacs LA et al. Intrinsic and Cyclin-dependent Kinase-dependent Control of Spindle Pole Body Duplication in Budding Yeast. Mol Biol Cell 2008;19:3243-3253.
18. Tsou MF, Stearns T. Mechanism limiting centrosome duplication to once per cell cycle. Nature 2006;442:947-951.
19. Budhu AS, Wang XW. Loading and unloading: orchestrating centrosome duplication and spindle assembly by Ran/Crm1. Cell Cycle 2005;4:1510-1514.
20. Grisendi S et al. Role of nucleophosmin in embryonic development and tumorigenesis. Nature 2005;437(7055):147-53.
21. Fukasawa K. Oncogenes and tumour suppressors take on centrosomes. Nat Rev Cancer 2007;7:911-924.
22. Nigg EA. Centrosome aberrations: cause or consequence of cancer progression? Nat Rev Cancer 2002;2:815-825.
23. Starita LM, Machida Y, Sankaran S, Elias JE, Griffin K, Schlegel BP, Gygi SP, Parvin JD. BRCA1-dependent ubiquitination of gamma-tubulin regulates centrosome number. Mol Cell Biol 2004;19:8457-66.

THE AUTHORS
Inês Bento, Inês Cunha-Ferreira,
Joana Borrego-Pinto &
Mónica Bettencourt-Dias
Cell Cycle Regulation Laboratory
Instituto Gulbenkian de Ciência
Rua da Quinta Grande, 6
P-2780-156 Oeiras
Portugal
sites.igc.gulbenkian.pt/ccr/
e-mail: mdias@igc.gulbenkian.pt