Prévia do material em texto

<p>Cell proliferation and cell cycle control: a mini review</p><p>C.H. GOLIAS, A. CHARALABOPOULOS, K. CHARALABOPOULOS</p><p>Department of Physiology, Clinical Unit, Medical Faculty, University of Ioannina, Ioannina, Greece</p><p>SUMMARY</p><p>Tumourigenesis is the result of cell cycle disorganisation,</p><p>leading to an uncontrolled cellular proliferation. Specific</p><p>cellular processes-mechanisms that control cell cycle pro-</p><p>gression and checkpoint traversation through the inter-</p><p>mitotic phases are deregulated. Normally, these events are</p><p>highly conserved due to the existence of conservatory</p><p>mechanisms and molecules such as cell cycle genes and</p><p>their products: cyclins, cyclin dependent kinases (Cdks),</p><p>Cdk inhibitors (CKI) and extra cellular factors (i.e. growth</p><p>factors). Revolutionary techniques using laser cytometry</p><p>and commercial software are available to quantify and</p><p>evaluate cell cycle processes and cellular growth. S-phase</p><p>fraction measurements, including ploidy values, using his-</p><p>tograms and estimation of indices such as the mitotic index</p><p>and tumour-doubling time indices, provide adequate infor-</p><p>mation to the clinician to evaluate tumour aggressiveness,</p><p>prognosis and the strategies for radiotherapy and che-</p><p>motherapy in experimental researches.</p><p>Keywords: Cell proliferation; cell cycle; cell cycle genes;</p><p>cyclins; cyclin dependent kinases; cyclin dependent kinases</p><p>inhibitors</p><p>� 2004 Blackwell Publishing Ltd</p><p>Uncontrolled proliferation is a hallmark of cancer cells. Molecu-</p><p>lar analysis of human tumours and animal models has</p><p>provided a clear basis for the understanding of the cellular</p><p>processes that govern cell cycle progression, in normal and</p><p>tumour cells. Many cell cycle regulators controlling the cor-</p><p>rect entry and progression through the cell cycle are altered in</p><p>tumours. In fact, most, if not all, human cancers show a</p><p>deregulated control of G1 phase progression, a period when</p><p>cells decide whether to start proliferation or to stay quiescent</p><p>(1).</p><p>In addition, human neoplasms develop following the pro-</p><p>gressive accumulation of genetic and epigenetic alterations to</p><p>oncogenes and tumour-suppressor genes (2). Such genes posi-</p><p>tively controlling cell cycle checkpoints can be targets for</p><p>oncogenic activation in cancer, whereas negative regulators</p><p>such as tumour-suppressor genes are targeted for inactivation</p><p>(3). Alterations to genes whose protein products, in some</p><p>cases, allow interaction with the environment external to the</p><p>cell, for example, with growth factors (2) confer a growth</p><p>advantage to cancer cell.</p><p>Genes that are altered in neoplasia affect three major bio-</p><p>logic pathways that normally regulate cell growth and tissue</p><p>homeostasis: the cell cycle, apoptosis and differentiation all</p><p>three functioning as an integrated molecular network, and</p><p>perturbations in one pathway can have profound conse-</p><p>quences on another (2).</p><p>Understanding the molecular details of the cell cycle regu-</p><p>lation and checkpoint abnormalities in cancer and manipu-</p><p>lation of these control mechanisms offers insight into</p><p>potential therapeutic strategies.</p><p>CELL CYCLE CONTROL</p><p>More than 50 years have passed since Howard and Pele in</p><p>1951 first described the cell cycle and its phases (Figure 1).</p><p>Nevertheless, there are only more recent studies that have</p><p>revealed that the cell cycle is a highly conserved and ordered</p><p>set of events, culminating in cell growth and division. Cell</p><p>cycle is tightly controlled by many regulatory mechanisms</p><p>that either permit or restrain its progression (4). Many of</p><p>the intermitotic events initiating DNA synthesis and mitosis</p><p>have become clear. The main families of regulatory proteins</p><p>that play key roles in controlling cell-cycle progression are the</p><p>cyclins, cyclin dependent kinases (Cdks), their substrate pro-</p><p>teins, the Cdk inhibitors (CKI) and the tumour-suppressor</p><p>gene products p53 and pRb (4). These families comprise the</p><p>basic regulatory machinery responsible for catalysing cell cycle</p><p>transition and checkpoint traversation.</p><p>CELL CYCLE GENES</p><p>Recent advantages in the field of cell cycle progression have</p><p>extended our knowledge of the mechanism(s) of cellular</p><p>Correspondence to:</p><p>K. A. Charalabopoulos, MD, PhD, Associate Professor, Department</p><p>of Physiology, Clinical Unit, Medical Faculty, University of</p><p>Ioannina, 13, Solomou str. 452 21 Ioannina, Greece</p><p>Tel.: 13-26510-97574</p><p>Fax: 13-26510-97850</p><p>Email: kcharala@cc.uoi.gr</p><p>ª 2004 Blackwell Publishing Ltd Int J Clin Pract, December 2004, 58, 12, 1134–1141</p><p>REVIEW d o i : 1 0 . 1 1 1 1 / j . 1 3 6 8 - 5 0 3 1 . 2 0 0 4 . 0 0 2 8 4 . x</p><p>proliferation and have placed it on a solid molecular basis. We</p><p>have considered two large groups of factors involved in cell</p><p>cycle control: extra cellular (environmental signals, i.e. growth</p><p>factors) and intracellular (oncogenes and anti-oncogenes) the</p><p>last indicating may be the most promising areas of investiga-</p><p>tion (5). Moreover, a great deal of work has focused on how</p><p>oncogenes and tumour-suppressor genes regulate the cell cycle</p><p>during normal development and in cancer. Recent work in</p><p>model organisms have demonstrated how oncogenes affect</p><p>the cell cycle by promoting growth. These studies also sug-</p><p>gested how growth and cell cycle progression may be coupled</p><p>(6).</p><p>Cell cycle genes function is targeted basically on controlling</p><p>accelerating and braking mechanisms, which act on the</p><p>engine during the cycle. It is of high importance to focus on</p><p>checkpoint or tumour-suppressor pathways as transduction</p><p>systems of negative signals that may induce cell cycle-braking</p><p>operation. They prevent any important cycle transition at the</p><p>initiation of proliferation, replication, mitosis, etc. until the</p><p>DNA and other cellular conditions make such a progression</p><p>safe.</p><p>Most of what we know about cell cycle regulation originally</p><p>came from lower organisms, including yeast. One of the first</p><p>genes to be identified as an important cell cycle regulator in</p><p>yeast is cdc2/cdc28. Activation of this gene requires association</p><p>with a regulatory subunit called cyclin A. It is now known</p><p>that sequential activation and inactivation of cyclin-dependent</p><p>kinases (cdks) is the primary means of cell cycle regulation</p><p>(Figures 2 and 3).</p><p>The checkpoint or tumour-suppressor protein p53 is one of</p><p>the proteins encoded by checkpoint gene (tumour-suppressor</p><p>gene). Mutations in that gene are present in more than half of</p><p>all human tumours (7). The p53 gene product is an import-</p><p>ant cell cycle check-point regulator at both the G1/S and G2/M</p><p>check points but does not appear to be important at the</p><p>mitotic spindle check point because gene knockout of p53</p><p>does not alter mitosis. The p53 tumour-suppressor gene is the</p><p>most frequently mutated gene in human cancer, indicating its</p><p>important role in conservation of normal cell cycle progression.</p><p>Specifically, the amino-terminal sequences of p53 function as</p><p>a transcriptional activation domain and the carboxy-terminal</p><p>sequences appear to be required for p53 to form dimers and</p><p>tetramers with itself. p53 has been shown to activate transcrip-</p><p>tion of number of genes with roles in the control of the cell</p><p>cycle, including WAF1/CIP1/p21 (which encodes a regulator of</p><p>Cdk activity) (8), GADD45 (a growth arrest DNA damage-</p><p>inducible gene), MDM2 (as note above, encoding a protein that</p><p>is a known negative regulator of p53) (9) (Figures 2 and 3). One</p><p>of p53’s essential roles is to arrest cells in G1 after genotoxic</p><p>damage, to allow for DNA repair prior to DNA replication and</p><p>cell division. In response to massive DNA damage, p53 triggers</p><p>the apoptotic cell death pathway. Tumour cells lacking normal</p><p>p53 function do not arrest in G1 and are more likely to progress</p><p>into S or G2/M and die, although different cells and different</p><p>drugs appear to have different requirements for this cell-killing</p><p>10 h</p><p>1 h</p><p>4 h</p><p>9 h</p><p>G1</p><p>G2</p><p>M</p><p>ito</p><p>si</p><p>s</p><p>S</p><p>Interphase</p><p>Cell Division</p><p>18–24 h</p><p>Complete cycle</p><p>s</p><p>Figure 1 Cell cycle and its phases</p><p>S-phase entry</p><p>P</p><p>Rb</p><p>p107</p><p>Rb</p><p>p107</p><p>P</p><p>P</p><p>E2FE2F</p><p>S-phase</p><p>gene</p><p>expression</p><p>INK4</p><p>CIP (p21)</p><p>CDK4/6</p><p>Cyclin E</p><p>Cyclin D</p><p>CDK2</p><p>S-phase</p><p>entry</p><p>Figure 2 Schematic presentation of various substances involved in</p><p>cell cycle proliferation and control, regarding the S-phase entry</p><p>(details in the text)</p><p>p16</p><p>p53</p><p>p21</p><p>Rb</p><p>Rb</p><p>E2F</p><p>E2F</p><p>p27</p><p>cyclin D</p><p>cyclin E</p><p>cyclin A,B</p><p>cdk2</p><p>cdk4.6cdc2 G0M</p><p>S</p><p>G1G2</p><p>PO4 PO4</p><p>PO</p><p>4</p><p>PO</p><p>4</p><p>PO</p><p>4</p><p>Figure 3 The cell division cycle and its control</p><p>CELL PROLIFERATION AND CELL CYCLE CONTROL 1135</p><p>ª 2004 Blackwell Publishing Ltd Int J Clin Pract, December 2004, 58, 12, 1134–1141</p><p>effect (10). The p53 protein functions as a transcription factor</p><p>by binding specific DNA sequences and regulating transcription</p><p>from promoters containing those sequences. The transcription</p><p>of a number of genes can be affected by activation of p53.</p><p>However, the ability of p53 to directly increase expression of</p><p>p21CIP1 is probably important for p53-dependent G1 cell</p><p>cycle arrest, observed upon DNA damage. Recent studies of</p><p>the function of the wild type p53 demonstrated that its anti-</p><p>proliferative effect is mediated by stimulation of a 21 kDa</p><p>protein (p21cip/waf1),a CKI, that inhibits cyclin-dependent</p><p>kinase activity, including cyclin D/CDK4 or 6 as well as cyclin</p><p>E/CDK2 and thereby cell division (11). In summary, DNA</p><p>damage generates a signal that can activate p53 by post-transla-</p><p>tional modification. Increased p53 activity up-regulates</p><p>p21CIP1, which prevents activation of CDKs, required for</p><p>the G1 to S transition. This negative cell cycle controller effect</p><p>may explain why the wild type p53 gene can suppress the</p><p>transformation of cells by activated oncogenes, thereby inhibit-</p><p>ing the growth of malignant cells in vitro and suppressing the</p><p>tumourigenic phenotype in vivo (11). An American large case</p><p>control study of lung cancer that included 482 cases and 510</p><p>controls demonstrated that the mutated p53 tumour-suppressor</p><p>gene, pro/pro genotype is associated with increased lung cancer</p><p>risk.</p><p>Another fundamental cell cycle regulator is a gene located</p><p>on chromosome 13q14, encoding retinoblastoma protein</p><p>(pRb) (Figures 2 and 3). Knudson predicted its existence, in</p><p>1971, based on observations of families with inherited</p><p>retinoblastoma (12). Examination of small-cell lung cancer</p><p>(SCLC) cytogenetics reveals a frequent loss of heterozygosity</p><p>at 13q, with more detailed molecular analysis pinpointing</p><p>these abnormalities to the Rb locus (13,14). In a study,</p><p>gross DNA abnormalities in the Rb gene were found in</p><p>approximately 20% of SCLC cell lines, while mRNA</p><p>expression was absent 60% of cell lines (14). Rb gene</p><p>functions as a tumour suppressor in SCLC in a similar</p><p>fashion as in retinoblastoma. Although, its exact function in</p><p>normal cells is not clear, changes in its degree of</p><p>phosphorylation have been correlated with the cell cycle,</p><p>suggesting a direct role in this process (15). Cross talk</p><p>pathways between p53-pRb nuclear proteins are being</p><p>delineated, implying potential links between p53 and pRb</p><p>in cell cycle control, apoptosis and tumour progression (16).</p><p>It is a fact that pRb is a master regulator for transcription</p><p>(17). PRbs’ presence is critical for G1 to S phase transition in</p><p>the cell cycle. It interacts with the E2F family of cell cycle</p><p>transcription factors to repress gene transcription required for</p><p>this transition (18) (Figures 2 and 3). Cyclin D is able to form</p><p>heteromeric complexes with the protein product (pRb) of the</p><p>retinoblastoma susceptibility gene. Similarly, cyclin E has</p><p>been shown to associate with the transcription factor E2F,</p><p>as well as with the Rb-related protein p107 (p107) (Figure 2).</p><p>These associations provide mechanisms by which catalytic</p><p>subunits (cdk4, cdk2) are targeted to specific substrates.</p><p>When constitutively overexpressed, pRb accumulates pri-</p><p>marily in the hypophosphorylated form and induces an arrest</p><p>of proliferation.This arrest could be overcome by co-expression</p><p>of cyclin A or E, in which case, the majority of pRb was seen</p><p>to be hyperphosphorylated (19). These findings combined</p><p>with the additional observation that pRb is first phosphorylated</p><p>(and it is assumed active) some hours before replicative DNA</p><p>synthesis is first detectable (20), provide the basis for a model</p><p>for G1 progression in animal cells. In this model, sequential</p><p>activation of different cyclin/cdk complexes would drive the</p><p>transition form a G1 state, in which active, hypophosphoryl-</p><p>ated pRb represses transcription from proliferation-associatd</p><p>gene promoters, to a state, in which Rb repression of E2F-</p><p>mediated transcription is relieved by phosphorylation of Rb</p><p>(21,22). Moreover, pRb not only regulates the activity of</p><p>certain protein, encoding genes but also the activity of RNA</p><p>polymerase pol I and pol II transcription. This protein appears</p><p>to be the major player in a regulatory circuit in the late G1</p><p>phase, the so-called restriction point (23). Members of the pRb</p><p>family form part of a signal transduction pathway called the Rb</p><p>pathway, which is important in cell cycle regulation, and have</p><p>roles in growth progression, differentiation and apoptosis in</p><p>different organisms and cell types (24). Several studies have</p><p>identified 97 genes as physiological targets of the pRB pathway,</p><p>providing further insights into how this pathway controls</p><p>proliferation (25). pRb through its interaction with E2F also</p><p>regulates genes that control apoptosis.</p><p>Over the last 10 years, the appreciation that the p53 and pRb</p><p>pathways are interconnected has grown substantially. The</p><p>checks and balances that exist between pRb and p53 involve</p><p>the regulation of the G1/S transition and its checkpoints and</p><p>much of this is under the control of the E2F transcription</p><p>family. From the existing mouse models, we also know that</p><p>proliferation, cell death and differentiation of distinct tissues are</p><p>also intimately linked through entrance and exist from the cell</p><p>cycle and thus through pRb and p53 pathways (26).</p><p>The cell is maintained by phosphorylation and dephos-</p><p>phorylation of the cell cycle gene products by cyclin/Cdk</p><p>complexes, the last being a group of proteins present at the</p><p>interphase cell.</p><p>CYCL INS</p><p>The cyclins are a family of proteins that are centrally involved</p><p>in cell cycle regulation and which are structurally identified by</p><p>conserved ‘cyclin box’ regions (27). They are 56 kDa proteins</p><p>first discovered in rapidly dividing cells and are implicated in</p><p>the mitosis of all eukaryotes (Figures 2 and 3).</p><p>Cyclins are regulatory subunits of holoenzymecyclin</p><p>dependent kinase (CDK) complexes controlling progression</p><p>1136 CELL PROLIFERATION AND CELL CYCLE CONTROL</p><p>ª 2004 Blackwell Publishing Ltd Int J Clin Pract, December 2004, 58, 12, 1134–1141</p><p>through cell cycle checkpoints by phosphorylating and inacti-</p><p>vating target substrates (Figure 3). Immunohistochemistry</p><p>studies were used to detect the expression of cyclins (cyclin A,</p><p>cyclin B, cyclin D1, cyclin D3 and cyclin E) on hepatocellular</p><p>carcinomas (HCC) tissue microarrays. This demonstrated that</p><p>cyclins in different cell cycles were over expressed at varied</p><p>levels in HCC. In addition, the increased expression of cyclins</p><p>may shorten the tumour cell cycle phase, accelerate cell</p><p>proliferation and finally have a close relationship with HCC</p><p>aggressiveness (28).</p><p>Many hormones and growth factors that influence in</p><p>turn cellular growth through signal transduction pathways</p><p>modify cyclins activity. De-regulated cyclin activity in</p><p>transformed cells contributes to accelerated cell cycle pro-</p><p>gression. Analysis of transformed cells and cells undergoing</p><p>mitogen stimulated growth implicate proteins of the NF</p><p>Kappa B family in cell cycle regulation through actions on</p><p>the Cdk/CKI system. The mammalian members of this</p><p>family are Rel A (p65), NF Kappa B [1] (p50, p105), NF</p><p>Kappa B [2] (p52, p100), c-Rel and Rel-B. These proteins</p><p>exist in cytoplasmic complexes with inhibitory proteins of</p><p>the IKappa B family and translocate to the nucleus to act as</p><p>transcription</p><p>factors when activated. The best explored link</p><p>between NF Kappa activation and cell cycle progression</p><p>involves cyclin D1 (27). The relationship between CDK/</p><p>CKI system and cyclins will be thoroughly discussed later</p><p>in the text.</p><p>Cyclin A is particularly interesting among the cyclin family,</p><p>because it can activate two different cyclin-dependent kinases</p><p>(Cdks) and function in both S phase and mitosis (Figures 2</p><p>and 3). In S phase phosphorylation of components of the</p><p>DNA replication machinery such as CDC6 by cyclin A-Cdk</p><p>is believed to be important for initiation of DNA replication</p><p>and to restrict the initiation to only once per cell cycle (29).</p><p>In mitoses, the precise role of Cyclin A is obscure, but it may</p><p>contribute to the control of Cyclin B stability. Cyclin A starts</p><p>to accumulate during S phase and is abruptly destroyed before</p><p>metaphase in advance of cyclin B which persists until</p><p>metaphase. Scientists have studied the expression of cyclin</p><p>A, cyclin E in HCC and demonstrated that cyclin E was</p><p>over expressed in 16 of 45 cases (35.6%) while cyclin A</p><p>was over expressed in 21 of 45 cases (46.7%) in HCC</p><p>lesions (30).</p><p>Other cyclins are acting to regulate entry into S phase. Over</p><p>expression of cyclin D1 and disruption of cell cycle control in</p><p>G1, occurs frequently in human esophageal cancer. Fong Ly</p><p>et al. (31) study using transgenic mice (TG) demonstrated</p><p>that over expression of D1, in combination with zinc defi-</p><p>ciency (ZD) substantially increased fore stomach tumour</p><p>incidence in TG mice: 85% of TG mice vs. 14% of ZS</p><p>(zinc sufficient). Additionally 14% of ZD–TG mice</p><p>developed esophageal tumours and esophageal intestinal</p><p>metaplasia in 77 days. It has been shown that cyclin D2</p><p>gene is underexpressed in tumour cell compared to normal</p><p>cells. The same studies also demonstrated that in total, low</p><p>expression was seen in 48/109 (48%) tumours. In addition,</p><p>cyclin D2 low expression was seen in 10/39 (25%) cases of</p><p>sporadic breast cancer and in 38/70 (54%) of familiar breast</p><p>cancer (32).</p><p>Progression through each phase of cell cycle is delicately</p><p>controlled by the activity of different cyclin-dependent</p><p>kinases (Cdks) and their regulatory subunits, the cyclins,</p><p>which are being demonstrated previously. The Cdks control</p><p>the G1 to S phase transition whereas complexes such as Cdk2,</p><p>cyclin E are important for initiation of the S phase (33). Cdk-</p><p>cyclin complexes act on specific targets which belong to at</p><p>least two major regulatory networks: the Rb-related/E2F</p><p>pathway and complexes that are responsible for the initiation</p><p>of DNA replication (34).</p><p>MEASURING TUMOUR-CELL GROWTH</p><p>Kinetic studies of cell proliferation rates shed light on the</p><p>growth dynamics of cancer. Most such studies are based on</p><p>measurements of cell numbers that are evaluated in time</p><p>intervals of about 12 h. Studies of the initial tumour growth</p><p>with short measuring intervals are rare (35). In normal tissues</p><p>a steady state renewal system operates, which contain the</p><p>enormous proliferate potential and tumour growth is a man-</p><p>ifestation of a breakdown of normal regulatory feedback</p><p>mechanisms. Techniques are now available which can quan-</p><p>tify the rate constant for cell production in human tumours</p><p>and these measurements are being used to design specific</p><p>treatment for special tumours (36).</p><p>In the past 2 decades, the technology of laser cytometry</p><p>and the use of halogenated thymidine analogues, haloge-</p><p>nated pyrimidines (HP), bromodeoxyuridine and iodo-</p><p>deoxyuridine as proliferation labels, have allowed us to</p><p>quantify the rate of cell turnover in tissues and tumours,</p><p>in clinical samples as in laboratory models. The principal</p><p>studies have used injection of bromo or iododeoxyuridine</p><p>to measure cell production rates in vivo. Flow cytometry</p><p>(FCM) has been used estimate the S phase labelling index</p><p>(LI) and the S phase duration (Ts) and calculate the cell</p><p>production rate, represented by the potential-doubling</p><p>time (Tpot). This has allowed calculation of time depen-</p><p>dent indices of proliferation from single biopsies of HP</p><p>pulse labelled human tissues and tumours (37). However,</p><p>cell production rate measurements do not adequately</p><p>describe the biological aggressiveness of tumours. They</p><p>may be used to refine adjuvant strategies for radiotherapy</p><p>and chemotherapy in experimental research (38). Under-</p><p>standing of the rates of growth of human cancers is essen-</p><p>tial for understanding the spectrum of cancer behaviour</p><p>observed clinically.</p><p>CELL PROLIFERATION AND CELL CYCLE CONTROL 1137</p><p>ª 2004 Blackwell Publishing Ltd Int J Clin Pract, December 2004, 58, 12, 1134–1141</p><p>Tumour-Doubling Times</p><p>At the late 70s, Steel made an attempt and managed to collect</p><p>110 cases of breast cancer and found a mean value for the</p><p>volume-doubling time (Td) of primary tumours and second-</p><p>ary tumours. This method of measurement of the rate of</p><p>macroscopic growth was a subject to error from the inaccur-</p><p>acy of assessing tumour dimensions in vivo.</p><p>During the last decades, a better index of human prolifer-</p><p>ation was developed than the clinically observed (Td). The</p><p>potential-doubling time (Tpot) is defined as the time</p><p>necessary to double the number of proliferating tumour cells</p><p>in the absence of spontaneous cell loss. The in vivo measure-</p><p>ment of Tpot is a possible way to detect fast growing</p><p>tumours, which would be better controlled by accelerated</p><p>radiotherapy (39). According to Steel, the mathematical defi-</p><p>nition of Tpot is Tpot5 ln2/Kp, where Kp is the rate con-</p><p>stant of cell production. All the operative formulas, which</p><p>allow the estimation of Tpot from the flow cytometric data,</p><p>derive from this definition (40). In general, Tpot is believed</p><p>to be important in the evaluation of tumour aggressiveness</p><p>and therapy response.</p><p>Mitotic Index</p><p>The progression of cells through the M phase of the cell cycle</p><p>is observed by using the mitotic index (MI). MI is established</p><p>by various methods, e.g. microinjection of fluorescently</p><p>labelled alpha/beta tubulin. This tubulin incorporates effi-</p><p>ciently into the mitotic spindle apparatus and by analysing</p><p>various steps of spindle formation we may also visualise</p><p>different stages of mitosis (41). Assessment of the MI could</p><p>be performed either microscopically by determining the fre-</p><p>quency of mitotic cells or by automating the process with flow</p><p>cytometry. On the other hand, mitotic activity index (MAI) is</p><p>a strong prognostic factor for disease-free survival in breast</p><p>cancer. The MAI is lower in screen detected tumours correl-</p><p>ating with less aggressive biological behaviour (42). Patel</p><p>et al. (43) demonstrated the role of mitotic counts in the</p><p>grading and prognosis of the breast cancer. MAI emerged as</p><p>one of the most significant prognostic parameter followed by</p><p>overall grade in predicting tumour-free survival (TFS) for the</p><p>patients. Mitotic count also correlated well with overall grade</p><p>and lymph node status in predicting TFS. This parameter is</p><p>very useful where advanced studies like flow cytometry and</p><p>immunohistochemical studies of the cell proliferation marker</p><p>are not available.</p><p>CELL PROL IFERAT ION – S PHASE</p><p>MEASUREMENTS</p><p>The rate of cell proliferation is identified through the</p><p>calculation of the proportion of cells in various phases of</p><p>the cell cycle, the easiest being the S phase. Many methods</p><p>have been used, but more recently studies are based upon the</p><p>use of halogenated pyrimidines and S phase fraction calcu-</p><p>lated from DNA histograms, in combination with static or</p><p>flow cytometry.</p><p>HALOGENATED PYRIMID INES</p><p>The advent of halogenated pyrimidines (bromodeoxyuridine-</p><p>BrdU, idoxuridine-IdU) and antibodies to recognise them has</p><p>opened new horizons for the measurement of proliferation in</p><p>human tumours leaving behind methods such as thymidine</p><p>labelling index (TLI) (44). The essence of the BrdU- flow</p><p>cytometric technique is that cells are labelled with the thymi-</p><p>dine analogue BrdU. They are then allowed to progress</p><p>through the cell cycle in a BrdU-free</p><p>environment during</p><p>the post-labelling time period. At a post-labelling time shorter</p><p>than the length of the S phase (Ts), cells are fixed and</p><p>prepared for FCM-mediated analysis of BrdU and DNA</p><p>contents. From FCM-derived data, cell cycle kinetic param-</p><p>eters such as labelling index (LI),Ts and Tpot can be</p><p>calculated (45). Wilson et al. (46) study investigating the</p><p>potential of human solid tumours, in vivo, using BrdUrd-</p><p>FCM and therefore measuring LI, Ts and Tpot within 24 h of</p><p>sampling demonstrated that; both LI and Ts vary greatly</p><p>between tumours (Ts ranges from 5.8 to 30.7). However,</p><p>within this study of 26 evaluable patients, tumours of the</p><p>same tissue origin tended to have similar Ts values.</p><p>Melanomas had the shortest Ts (8.8 h), nine patients with</p><p>head and neck cancer had Ts from 5.8 to 18.8 (median</p><p>12.5 h). The longest Ts values were found in lung and</p><p>rectum. The striking feature was that 38% of the tumours</p><p>had a Tpot of 5 days or less.</p><p>S-PHASE FRACT ION</p><p>S phase fraction (SPF) measurement by flow cytometry is a</p><p>method measuring the proportion of cells in S phase. Stand-</p><p>ardised use of commercially available resolution provided</p><p>the ability of measuring comparable SPF results (47). Its</p><p>advantage over thymidine labelling is that SPF is able to</p><p>count many thousands of cells producing the results in</p><p>hours rather than weeks.</p><p>The SPF of a tumour cell population is often viewed as</p><p>general indicator of the clinical aggressiveness of that tumour.</p><p>Actually, the SPF of a cancer should be interpreted as no</p><p>more than an indicator of the mean duration of its mitotic</p><p>cycle. The SPF of a solid tumour is difficult to measure</p><p>accurately, but in principle, it is a powerful predictor of the</p><p>duration of the recurrence-free interval (48).</p><p>The prognostic significance of SPF and DNA ploidy in</p><p>breast cancer following enrichment of tumour cells by cyto-</p><p>keratin labelling has been demonstrated. Univariate analysis</p><p>in breast cancer detected the prognostic significance of DNA</p><p>1138 CELL PROLIFERATION AND CELL CYCLE CONTROL</p><p>ª 2004 Blackwell Publishing Ltd Int J Clin Pract, December 2004, 58, 12, 1134–1141</p><p>ploidy and SPF of Go/G1 peak of tumour cells for clinical</p><p>outcome (49). The SPF is calculated from the ploidy</p><p>histogram by the use of computer modelling systems to</p><p>determine the size and the compartment between the G0/</p><p>G1 and G2/M peaks.</p><p>Michels et al. (50) study showed that SPF is a valuable</p><p>predictor of survival and can be confidently assessed in multi-</p><p>ploid histograms and thus improves the yield of flow cyto-</p><p>metry. When combined with mitotic activity, the prognostic</p><p>impact of SPF is the same as that of lymph node status in</p><p>breast cancer. Tumours that are hypoploid and multiploid</p><p>have a significantly worse prognosis. Tumour ploidy can be</p><p>calculated from the same histogram that SPF is calculated.</p><p>Ploidy is an expression of the DNA content of the tumour</p><p>cells, which are classified as diploid if their DNA content is</p><p>similar to non-transformed cells, or aneuploid if it is grossly</p><p>abnormal.</p><p>A standardised flow cytometric study analysed SPF and</p><p>DNA ploidy in 633 T1T2 breast cancer cases. DNA aneu-</p><p>ploidy was observed in 61% of cases. Aneuploidy and high</p><p>SPF were associated with large tumour size, node involve-</p><p>ment, high histological grade and hormone receptor negativ-</p><p>ity. Patients with medium and high SPF values, in the overall</p><p>population, had shorter disease-free survival (DFS) than those</p><p>with low SPF. Ploidy had no significant value (51). In</p><p>another study, conducted in a total of 177 patients with</p><p>primary breast cancer, was demonstrated that 5-year survival</p><p>was twice worse in aneuploid tumours than those with diploid</p><p>ones. The survival rates after one operation, pre-operative and</p><p>post-operative radiation therapy was greatest in diploid than</p><p>in aneuploid tumours. Thus, DNA ploidy of the studied</p><p>neoplasm is of high informative value in predicting the course</p><p>of tumours process and in choosing treatment policy on an</p><p>individual basis (52).</p><p>CELL–CYCLE MEASUREMENTS</p><p>The prediction of tumour behaviour and response to treat-</p><p>ment has led to interest in the assessment of the proliferative</p><p>potential tumours. It is well known, for example, that pituitary</p><p>tumours are usually histologically benign but are capable</p><p>of aggressive growth and local invasion, although distant</p><p>metastasis are limited to the rare pituitary carcinoma. These</p><p>differences in tumour behaviour may determine both the</p><p>prognosis and also the effectiveness of treatment whether it</p><p>be surgery, drugs or radiotherapy. Immmunohistochemistry</p><p>using antibodies to Ki-67 and proliferating cell nuclear anti-</p><p>gen (PCNA) which are expressed in cells that have entered the</p><p>cell cycle, can be used to assess the proportion of the cells</p><p>from a tumour that are proliferating. The percentage of</p><p>positively stained nuclei (LI) may be helpful in predicting</p><p>appropriate management, as there is a relationship in many</p><p>tumours between LI, invasiveness and tumour recurrence</p><p>(53). Martinez-Arribas et al. (54) studied the proliferation of</p><p>181 breast cancers by means of FCM and Ki-67 LI, using the</p><p>MIBI antibody in an attempt to evaluate both methods pro-</p><p>spectively. They found a significant correlation between rising</p><p>DNA content and increasing Ki-67 index (r5 0.18;</p><p>p5 0022), as well as between the percentage of the cells in S</p><p>phase of the whole tumour population and Ki-67 (r5 0.22;</p><p>p5 0.0055). The conclusion was that the Ki-67 LI and S</p><p>phase fraction are significantly correlated. However, flow</p><p>cytometry provides additional indirect information on</p><p>tumour aggressiveness associated with DNA-ploidy. Further-</p><p>more, studies are needed to determine whether Ki-67 alone is</p><p>sufficient as a routinely applicable method (54).</p><p>CONCLUS IONS</p><p>During the last decade, the studies concerning the cell cycle</p><p>and its regulation have passed from the simpler eukaryotes to</p><p>the ones’ of fundamental importance, multi-cellular organ-</p><p>isms. These researches and others being in progress, promise</p><p>to illuminate the links between molecules provoking tumouri-</p><p>genesis and those involved in cell cycle progression giving</p><p>more evidence of their biochemical importance and clinical</p><p>application. Improving of the anti-cancer drug therapy is</p><p>directly combined with the understanding of the check-</p><p>point and the apoptotic mechanisms. Additionally, the iden-</p><p>tification of patients with early pre-neoplastic lesions and the</p><p>screening of samples, respectively, for genetic abnormalities</p><p>such as p53 or pRb mutations, may allow us the early identi-</p><p>fication of tumourigenesis and subsequently an opportunity</p><p>to apply a curative rather than a palliative therapy.</p><p>REFERENCES</p><p>1 Malumbres M, Carnero A. Cell cycle deregulation: a common</p><p>motif in cancer. Prog Cell Cycle Res 2003; 5: 5–18.</p><p>2 Corn BG, El-Deiry WS. Derangement of growth and differ-</p><p>entiation control in oncogenesis. Bioessays 2002; 24 (1): 83–90.</p><p>3 McDonald ER III, El-Deiry W. Checkpoint genes in cancer.</p><p>Ann Med 2001; 33 (2): 113–22.</p><p>4 Gali-Muhtasib H, Bakkar N. Modulating cell cycle: current</p><p>applications and prospects for future drug development. Curr</p><p>Cancer Drug Targets 2002; 2 (4): 309–36.</p><p>5 Baserga R, Porcu P, Sell C. Oncogenes, growth factors and</p><p>control of the cell cycle. Cancer Surv 1993; 16: 201–13.</p><p>6 Prober DA, Edgar BA. Growth regulation by oncogenes-new</p><p>insights from model organisms. Curr Opin Genet Dev 2001; 11</p><p>(1): 19–26.</p><p>7 Lopez-Saez JF, de la Torre C, Pincheira J, Gimenez-Martin G.</p><p>Cell proliferation and cancer. Histol Histopathol 1998; 13 (4):</p><p>1197–214.</p><p>8 el-Deiry WS, Tokino T, Velculescu VE et al. WAF1, a potential</p><p>mediator of p53 tumor suppression. Cell 1993; 75: 817–25.</p><p>CELL PROLIFERATION AND CELL CYCLE CONTROL 1139</p><p>ª 2004 Blackwell Publishing Ltd Int J Clin Pract, December 2004, 58, 12, 1134–1141</p><p>9 Kastan M, Zhan Q, El-Diery W et al. A mammalian cell cycle</p><p>checkpoint pathway utilizing p53 and GADD45 is defective in</p><p>ataxia-telangiectasia.</p><p>Cell 1992; 71: 587–98.</p><p>10 Brown JM, Wouters BG. Apoptosis, p53, and tumor cell</p><p>sensitivity to anticancer agents. Cancer Res 1999; 59: 1391–9.</p><p>11 Fan R, Wu MT, Miller D et al. The p53 codon 72 polymorph-</p><p>ism and lung cancer risk. Cancer Epidemiol Biomarkers Prev</p><p>2000; 9 (10): 1037–42.</p><p>12 Evans GA, Farrar WL. Retinoblastoma protein-phosphoryl-</p><p>ation does not require activation of p34cdc2 protein kinase.</p><p>Biochem J 1992; 287: 965–9.</p><p>13 Bonneta L. Open questions on p16. Nature 1994; 370: 180.</p><p>14 Ewen ME, Sluss HK, Sherr CJ, Matsushime H, Kato JY,</p><p>Livingstone DM. Functional interactions of the retinoblastoma</p><p>protein with mammalian D-type cyclins. Cell 1993; 73: 487–97.</p><p>15 Ferrel JJ, Wu M, Gerhart JC, Martin GS. Cell cycle tyrosine</p><p>phosphorylation of p34cdc2 and a microtubule-associated</p><p>protein kinase homolog in Xenopous oocytes and eggs. Mol</p><p>Cell Biol 1991; 11: 1965–71.</p><p>16 Cordon-Cardo C. Mutations of cell cycle regulators. Biological</p><p>and clinical implications for human neoplasia. Am J Pathol 1995;</p><p>147 (3): 545–60.</p><p>17 Chan HM, Shikama N, La Thangue NB. Control of gene</p><p>expression and the cell cycle. Essays Biochem 2001; 37: 87–96.</p><p>18 Harbour JW, Dean DC. Rb function in cell cycle regulation and</p><p>apoptosis. Nat Cell Biol 2000; 2 (4): 65–7.</p><p>19 Hinds PW, Mittnacht S, Dulic V, Arnold A, Reed SI, Weinberg RA.</p><p>Regulation of the retinoblastoma protein functions by ectopic</p><p>expression of human cyclins. Cell 1992; 70: 993–1006.</p><p>20 Mittnacht S, Lees JA, Desai D, Harlow E, Morgan DO, Weinberg RA.</p><p>Distinct subpopulations of the retinoblastoma protein show a</p><p>distinct pattern of phosphorylation. EMBO J 1994; 13: 118–27.</p><p>21 Dulic V, Lees E, Reed SI. Association of human cyclin E</p><p>with a periodic G1-S phase protein-kinase. Science 1992; 257:</p><p>1958–61.</p><p>22 Koff A, Giordano A, Desai D et al. Formation and activation of</p><p>a cyclin E-cdk2 complex during the G1 phase of the human cell</p><p>cycle. Science 1992; 157: 1689–94.</p><p>23 Herwig S, Strauss M. The retinoblastoma protein: a master</p><p>regulator of cell cycle, differentiation and apoptosis. Eur J Bio-</p><p>chem 1997; 246 (3): 581–601.</p><p>24 Claudio PP, Tonini T, Giordano A. The retinoblastoma family:</p><p>twins or distant cousins? Genome Biol 2002; 3 (9): 3012.</p><p>25 Vernel R, Helin K, Muller H. Identification of target genes of</p><p>the p16INK4A-pRB-E2F pathway. Biol Chem 2003; 278 (46):</p><p>124–37.</p><p>26 Yamasaki L. Role of the RB tumour suppressor in cancer. Cancer</p><p>Treat Res 2003; 115: 209–39.</p><p>27 Joyce D, Albanese C, Steer J, Fu M, Bouzahzah B, Pestell RG.</p><p>NF-Kappa and cell cycle regulation: the cyclin connection. Cyto-</p><p>kine Growth Factor Rev 2001; 12 (1): 73–90.</p><p>28 Zhu MH, Ni CR, Zhu Z, Li FM, Zhang SM. Immunohisto-</p><p>chemical demonstration of cyclins A,B,D1,D3 and E in hepato-</p><p>cellular carcinomas using tissue microarrays. Zhonghua Bing Li</p><p>Xue Za Zhi 2003; 32 (5): 440–3.</p><p>29 Yam CH, Fung TK, Poon RY. Cyclin A in cell cycle control and</p><p>cancer. Cell Mol Life Sci 2002; 59 (8): 1317–26.</p><p>30 Zhou Q, He Q, Liang LJ. Expression of p27,cyclin E and cyclin</p><p>A in hepatocellular carcinoma and its clinical significance. World</p><p>J Gastroenterol 2003; 9 (11): 2450–4.</p><p>31 Fong LY, Mancini R, Nakagawa H, Rustgi AK, Huebner K.</p><p>Combined cyclin D1 overexpression and zinc deficiency disrupts</p><p>cell cycle and accelerates mouse forestomach carcinogenesis.</p><p>Cancer Res 2003; 63 (14): 4244–52.</p><p>32 Fischer H, Chen J, Skoog L, Lindoblom A. Cyclin D2 expres-</p><p>sion in familial and sporadic breast cancer. Oncol Rep 2002;</p><p>9 (6): 1157–61.</p><p>33 Lee MH, Yang HY. Regulators of G1 cyclin-dependent kinases</p><p>and cancers. Cancer Metastasis Rev 2003; 22 (4): 435–49.</p><p>34 Gutierrez C, Ramirez-Parra E, Castellano MM, del Pozo JC.</p><p>G(1) to S phase transition: more than a cell cycle engine switch.</p><p>Curr Opin Plant Biol 2002; 5 (6): 480–6.</p><p>35 Sedivy R, Thurner S, Budinsky AC, Kostler WJ, Zielinski CC.</p><p>Short-term rhythmic proliferation of human breast cancer cell</p><p>lines: surface effects and fractal growth patterns. J Pathol 2002;</p><p>197 (2): 163–9.</p><p>36 Watson JV. Tumour growth dynamics. Br Med Bull 1991;</p><p>47 (1): 47–63.</p><p>37 Rew DA, Wilson GD. Cell production rates in human tissues</p><p>and tumours and their significance. Part 1: an introduction to</p><p>the techniques of measurement and their limitations. Eur J Surg</p><p>Oncol 2000; 26 (3): 227–38.</p><p>38 Rew DA, Wilson GD. Cell production rates in human tissues</p><p>and tumours and their significance. Part II: clinical data. Eur J</p><p>Oncol 2000; 26 (4): 405–17.</p><p>39 Dubray B, Maciorowsky Z, Cosset JM, Terry NH. Clinical value</p><p>of the potential doubling time (Tpot) measured by flow cyto-</p><p>metry. Bull Cancer 1995; 82 (5): 331–8.</p><p>40 Bertuzzi A, Gandolfi A, Sinisgalli C, Starace G, Ubezio P. Cell</p><p>loss and the concept of potential doubling time. Cytometry 1997;</p><p>29 (1): 34–40.</p><p>41 Schwindling SL, Faust M, Montenarh M. Determination of the</p><p>mitotic index by microinjection of fluorescently labelled tubulin.</p><p>Eur J Cell Biol 2002; 81 (3): 169–74.</p><p>42 Groenendijk RP, Bult P, Noppen CM, Boettes C, Ruers TJ,</p><p>Wobbes T. Mitotic activity index in interval breast cancers. Eur J</p><p>Surg Oncol 2003; 29 (1): 29–31.</p><p>43 Patel C, Sidhu KP, Shah MJ, Patel SM. Role of the mitotic</p><p>counts in the grading and prognosis of the breast cancer. Indian J</p><p>Pathol Microbiol 2002; 45 (3): 247–54.</p><p>44 Bergstrom C, Begg A, Palmqvist R, Waites A, Denekamp J.</p><p>Labelling indices in human tumours: to apply or not – that is</p><p>the question. Br J Cancer 1999; 80 (10): 1635–43.</p><p>45 Johansson MC, Johansson R, Baldetorp B, Oredsson SM. Com-</p><p>parison of different labelling index formulae used on bromodeo-</p><p>xyuridine-flow cytometry data. Cytometry 1998; 32 (3): 233–40.</p><p>46 Wilson GD, McNally NJ, Dische S et al. Measurement of cell</p><p>kinetics in human tumours in vivo using bromodeoxyuridine</p><p>incorporation and flow cytometry. Br J Cancer 1988; 58 (4):</p><p>423–31.</p><p>47 Jourdan ML, Ferrero-Pous M, Spyratos F, Romain S, Martin PM,</p><p>Chassevent A. Flow cytometric S – phase fraction measurement in</p><p>breast carcinoma: Influence of software and histogram resolution.</p><p>Cytometry 2002; 48 (2): 66–70.</p><p>1140 CELL PROLIFERATION AND CELL CYCLE CONTROL</p><p>ª 2004 Blackwell Publishing Ltd Int J Clin Pract, December 2004, 58, 12, 1134–1141</p><p>48 Johnson HA. on the clinical significance of the S phase fractions</p><p>of tumours. Ann Clin Lab Sci 1994; 24 (5): 431–5.</p><p>49 Wimberger P, Hillemanns P, Kapsner T, Hepp H, Kimming R.</p><p>Evaluation of prognostic factors following flow-cytometric Dann</p><p>analysis after cytokeratin labelling: I. Breast Cancer. Ann Cell</p><p>Pathol 2002; 24 (4–5): 135–45.</p><p>50 Michels JJ, Duigou F, Marnay J, Denoux Y, Deloizier T,</p><p>Chasle J. Flow cytometry in primary carcinomas. Prognostic</p><p>impact of multiploidy and hypoploidy. Cytometry 2003; 55B</p><p>(1): 37–45.</p><p>51 Chassevent A, Jourdan ML, Romain S et al. S-phase fraction and</p><p>DNA ploidy in 633, T1T2 breast cancers. Clin Cancer Res 2001;</p><p>7: 909–17.</p><p>52 Nikolaeva TG, Dobrynin I, Letiagin VP, Prorokov V, Matiakin EG,</p><p>Davydov MI. The experience in the use of DNA flow cytometry</p><p>to predict the natural history of neoplasms. Vestn. Ross Akad</p><p>Med Nauk 2002; 1: 45–9.</p><p>53 Turner HE, Wass JA. Are markers of proliferation valuable in</p><p>the histological assessment of the pituitary tumours? Pituitary</p><p>1999; 1 (3–4): 147–51.</p><p>54 Martinez-Arribas F, Nunez MJ, Piqueras V et al. Flow cytometry</p><p>vs Ki67 labelling index in breast cancer: a prospective evaluation</p><p>of 181 cases. Anticancer Res 2002; 22 (1A): 295–8.</p><p>Paper received February 2004, accepted May 2004</p><p>CELL PROLIFERATION AND CELL CYCLE CONTROL 1141</p><p>ª 2004 Blackwell Publishing Ltd Int J Clin Pract, December 2004, 58, 12, 1134–1141</p>