• unlimited access with print and download
    $ 37 00
  • read full document, no print or download, expires after 72 hours
    $ 4 99
More info
Unlimited access including download and printing, plus availability for reading and annotating in your in your Udini library.
  • Access to this article in your Udini library for 72 hours from purchase.
  • The article will not be available for download or print.
  • Upgrade to the full version of this document at a reduced price.
  • Your trial access payment is credited when purchasing the full version.
Buy
Continue searching

Cyclin E-dependent mechanisms linking centrosome duplication with DNA replication and cell cycle progression

ProQuest Dissertations and Theses, 2009
Dissertation
Author: Rebecca L Ferguson
Abstract:
Centrosome duplication and DNA replication are fundamentally similar processes in that both occur once-and-only-once during each cell cycle. Both events initiate at the GUS transition and require an increase in Cdk2 activity. It has long been evident that the centrosome duplication and DNA replication cycles must not only be highly regulated but also temporally coordinated to prevent inappropriate re-replication. Abnormal centrosome numbers are described in virtually all human malignancies and have been shown to be a probable cause of chromosomal instability. Previous work in this laboratory defined a 20 amino acid modular centrosomal localization signal (CLS) in cyclin E that is both necessary and sufficient for centrosomal localization. Furthermore, it was shown that expression of the CLS displaces both endogenous cyclins E and A from centrosomes and prevents BrdU incorporation into DNA. To identify interactions mediated by the CLS, a HeLa cDNA library was screened using a bacterial two-hybrid assay with the CLS as bait. In this screen MCM5, an essential DNA replication factor, was isolated. Co-immunoprecipitation experiments and GST-pulldown assays verified a direct, in vivo interaction between MCM5 and cyclin E that is dependent on a wild-type CLS but independent of Cdk2. Interaction with cyclin E accounts for MCM5 localization on centrosomes in mammalian cells, and ectopic expression of MCM5 inhibits centrosome over-duplication in S phase-arrested CHO cells. MCM5 and cyclin A also interact in a manner dependent on a functional CLS domain but not on Cdk binding. Expression of the cyclin E CLS in G1/S synchronized cells inhibits S phase due to a lack of chromatin loading of proteins essential for the initiation of DNA replication. Rescue of DNA replication was evident when active cyclin E/Cdk2 was re-targeted to centrosomes. Taken together, the work presented here reveals novel mechanisms of communication between the centrosome and nuclear replication cycles in which elements of each cycle can directly regulate events in the other.

TABLE OF CONTENTS CHAPTER I. INTRODUCTION 1 1.1 Cell Cycle 1 1.1.1 Phases of the cell cycle 1 1.1.2 Cell cycle control 6 1.1.2.1 Cyclin-dependent kinase 6 1.1.2.2 Regulating the regulators 12 1.1.3 Checkpoints 15 1.2 Eukaryotic DNA Replication 19 1.2.1 Origin recognition and formation of pre-replication complexes 19 1.2.1.1 Origin recognition compex (ORC) 19 1.2.1.2 Cell division cycle homolog 6 (Cdc6) 22 1.2.1.3 Chromatin licensing and DNA replication factor 1 (Cdtl) 22 1.2.1.4 Minichromosome maintenance complex 22 1.2.2 Initiation of DNA replication 29 1.2.3 Preventing re-replication of DNA 34 1.3 Centrosomes 36 1.3.1 Centrosome structure 3 6 1.3.2 Centrosome function 39 1.3.2.1 Microtubule-organizing center (MTOC) 3 9 1.3.2.2 Signaling platform 40 vii

1.3.3 Centrosome cycle 41 1.3.3.1 Control of centrosome duplication 43 1.3.3.2 De novo centrosome formation 45 1.3.3.3 Coordination with DNA replication 46 1.3.3.4 Centrosome substrates 48 1.3.3.5 Cdk2 and cyclin E are dispensable for normal duplication 49 1.3.3.6 Cancer and disease 51 1.4 Centrosome Control of the Cell Cycle 52 1.4.1 Gl-S phase transition 52 1.4.2 G2-M phase transition 54 1.4.3 Metaphase-anaphase transition 55 1.4.4 Cytokinesis 56 1.5 Overview 57 1.6 Thesis Aims 58 II. MATERIALS AND METHODS 61 2.1. Bacterial Two-Hybrid 61 2.2 Sodium Dodecylsulfate Polyacrylamide Gel Electrophorsis (SDS-PAGE) 62 2.3 Western Blotting 62 2.4 DNA Manipulation 63 2.4.1 Plasmid transformation 63 2.4.2 Plasmid purification 64 2.4.3 Agarose gel electrophoresis 65 2.4.4 Restriction endonuclease digestion 65 vin

2.4.5 Polymerase chain reaction (PCR) 65 2.4.6 Oligonucleotide primers 67 2.5 Tissue Culture and Stable Cell Lines 69 2.5.1 Cell lines 69 2.5.2 Flp-In TRex stable cell lines 70 2.5.3 Transfection 71 2.6 GST Pull-Down Assay 72 2.7 Co-Immunoprecipitation 73 2.7.1 Endogenous immunoprecipitation 73 2.7.1.1 CyclinE 73 2.7.1.2 Cyclin A 74 2.7.2 Myc immunoprecipitation 75 2.8 Immunofluorescence Analysis 76 2.8.1 BrdU incorporation 76 2.8.2 Indirect immunofluorescence localization and centrosomes 77 2.9 Centrosome Duplication Assay 78 2.10 Chromatin Extraction Assay 78 III. CYCLIN E-DEPENDENT LOCALIZATION OF MCM5 REGULATES CENTROSOME DUPLICATION 80 3.1 Introduction 80 3.2 Results 83 3.2.1 Verification of two-hybrid results 83 3.2.2 In vivo direct interaction between cyclin E and MCM5 83 IX

3.2.3 Interaction between cyclin E and MCM5 is dependent on the cyclin E CLS but independent of Cdk2 85 3.2.4 Amino acids 533-569 of MCM5 mediate interaction with cyclin E 87 3.2.5 MCM5 is centrosomally localized and displaced by expression of the CLS of cyclin E 92 3.2.6 Centrosomal localization of MCM5 is mediated by interaction with cyclin E 92 3.2.7 MCM5 inhibits centrosome amplification without disrupting cyclin E localization 100 3.2.8 MCM5-mediated inhibition of centrosome duplication does not require binding to other MCM family members 105 3.3 Discussion 109 IV. CYCLIN A DIRECTLY BINDS AND RECRUITS MCM5 TO CENTROSOMES 118 4.1 Introduction 118 4.2. Results 120 4.2.1 Cyclin A and MCM5 directly interact in vivo 120 4.2.2 Amino acids 533-569 of MCM5 mediate interaction with cyclin A 123 4.2.3 Interaction between cyclin A and MCM5 is dependent on the intact CLS of cyclin A but independent of Cdk binding 125 4.2.4 MCM5 centrosome localization is supported by interaction with cyclin A 127 4.3 Discussion 131 V. CENTROSOMAL LOCALIZATION OF CYCLIN E/CDK2 IS REQUIRED FOR INITIATION OF DNA SYNTHESIS 13 8 5.1. Introduction 138 5.2. Results 140 x

5.2.1 Expression of the cyclin E CLS inhibits BrdU incorporation in cells arrested at the Gl/S boundary 140 5.2.2 Targeting cyclin E to the centrosome restores BrdU incorporation in CLS expressing cells 140 5.2.3 Rescue of BrdU incorporation requires centrosomally-localized Cdk2 activity 146 5.2.4 Initiation of DNA replication is inhibited in cells expressing the cyclin E CLS 146 5.3 Discussion 151 VI. CONCLUSION AND FUTURE DIRECTIONS 156 REFERENCES 166 APPENDIX 198 A. IN VITRO KINASE ASSAYS 198 B. DNA REPLICATION ASSAY IN XENOPUS EMBRYOS 203 C. TWO-HYBRID RESULTS 207 XI

LIST OF FIGURES FIGURE 1.1 The cell cycle and checkpoints 2 1.2 The stages of mitosis and cytokinesis 5 1.3 Expression of the cyclins is temporally regulated 8 1.4 Phosphorylation of pRb is essential for cell cycle progression 9 1.5 Summary of known cyclin E functions 11 1.6 Regulation of Cdk activity by Cdk inhibitors 14 1.7 General outline of the DNA damage response pathway 16 1.8 Cell cycle checkpoints 17 1.9 Formation of the pre-replication complex 20 1.10 Features of the MCM family proteins 25 1.11 MCM complex architecture and structure 27 1.12 Formation of the pre-initiation complex 3 0 1.13 Centrosome structure 37 1.14 The centrosome duplication cycle 42 3.1 MCM5 directly interacts with cyclin E in vivo 84 3.2 Interaction between cyclin E and MCM5 is CLS-dependent but Cdk2-independent 86 3.3 Amino acids 533-569 within MCM5 mediate interaction with cyclin E 88 3.4 Evolutionary conservation of the MCM5 domain that interacts with cyclin E 90 3.5 The mcm5A640-706 is defective in DNA replication 91 xn

3.6 Endogenous MCM5 in localized on centrosomes 93 3.7 Cyclin E-interacting fragments of MCM5 are centrosomally localized 95 3.8 Cyclin E recruits MCM5 to centrosomes 98 3.9 Expression of MCM5 but not MCM2 inhibits centrosome duplication 103 3.10 Expression of MCM5 does not displace endogenous cyclin E from centrosomes 104 3.11 The zinc finger in MCM5 is essential for MCM family binding but not cyclin binding 106 3.12 A centrosomal function for MCM5 does not require MCM family members 108 3.13 Model for positive and negative regulatory roles of cyclin E/Cdk2 during DNA and centrosome replication 115 4.1 Space-filling models of the crystal structures of cyclin E and cyclin A 121 4.2 MCM5 directly interacts with cyclin A in vivo 122 4.3 Cyclin A interacts with amino acids 532-569 of MCM5 124 4.4 Interaction between MCM5 and cyclin A is mediated by the cyclin A CLS 126 4.5 Interaction between MCM5 and cyclin A requires an intact cyclin A CLS 128 4.6 Displacement of endogenous MCM5 from the centrosome by expression of the cyclin A CLS can be restored by co-expression of PACT-cyclin A 130 5.1 Expression of the CLS of cyclin E inhibits BrdU incorporation in asynchronous cells 141 5.2 Expression of the CLS of cyclin E inhibits incorporation of BrdU in double thymidine synchronized cells 142 5.3 The PACT domain mediates centrosomal localization independently of the CLS of cyclin E 144 5.4 Redirecting cyclin E to centrosomes restores DNA synthesis in cells expressing the CLS 145 5.5 DNA synthesis requires centrosomally localized Cdk activity 147 xiii

5.6 Schematic of biochemical chromatin fractionation method 148 5.7 Chromatin loading of DNA replication factors is impaired in cells expressing the CLS of cyclin E 150 5.8 Model for centrosomal action of cyclin E/Cdk2 in DNA synthesis 155 6.1 Model of communication between centrosome duplication and DNA replication 157 6.2 Endogenous Cdc45 and ectopically expressed HA-tagged Cdc45 are centrosomally localized 163 6.3 Cdc45 does not directly interact with cyclin A or cyclin E 164 A. 1 MCM5 is not a substrate for cyclinE/Cdk2 199 A.2 MCM5 does not inhibit cyclin E/Cdk2 activity 202 B.l Expression of the CLS of cyclin E does not inhibit DNA replication in Xenopus embryos 206 xiv

TABLE OF ABBREVIATIONS 3-AT aa AAA ACS AEBSF AKAP APC/C ARS ATP BrdU BSA C. elegans CAK CAT CCD Cdc20 Cdc25 Cdc45 Cdc6 Cdk Cdtl CG-NAP CHO CIN Cip CKI CLS CMG CP110 3-amino-1,2,4-triazole Amino acid ATPase associated with various cellular activities ARS consensus sequence 4-(2-Aminoethyl) benzenesulfonyl fluoride hydrochloride A-kinase anchor protein Anaphase-promoting complex/cyclosome Autonomously replicating sequence Adenosine 5'-triphosphate Bromodeoxyuridine Bovine serm albumin Caenorhabditis elegans Cdk activating kinase chloramphenicol acetyltransferase Charge-coupled device cell division cycle 20 gene product cell division cycle 25 gene product cell division cycle 45 gene product cell division cycle 6 gene product Cyclin-dependent kinase Chromatin licensing and DNA replication factor 1 Centrosome- and Golgi-localized PKN-associated protein Chinese hamster ovary cell line Chromosomal instability Cdk interacting protein Cyclin-dependent kinase inhibitor protein Centrosomal localization sequence Cdc45, MCM, Gins complex Centrosome protein 110 XV

C-terminal Dapi dCTP DDK dm dNTP DTT E. coli ECL EDTA EdU EGTA EM Emil FBS Go G, G2 GFP GINS GST Y-TURC Y-TUSC HA HCG HC1 HeLa Hepes HPV HRP HU Carboxy-terminal 4', 6-Diamidine-2-phenylindole dihydrochloride Deoxycytidine triphosphate Dbf4-dependent kinase Drosophila melanogaster Deoxyribonucleotide triphosphate Dithiothreitol Escherichia coli Enhanced chemiluminescence Ethylenediamine tetraacetic acid Ethynyl-deoxyuridine Ethyleneglycol bis (2-aminoethylether)-N'N' tetracetic acid Electron microscopy Early mitotic inhibitor 1 Fetal bovine seruma Null gap phase First gap phase Second gap phase Green fluorescence protein Go, Ichi Nii, San Glutithione-S-transferase Gamma tubulin ring complex Gamma tubulin small complex Hemagglutinin Human chorionic gonadotropin Hydrochloric acid Henrietta Lacks N-(2-hydroxyethyl) piperazine-N'-(2-ethanesulfonic acid) Human papillomavirus Horseradish peroxidase Hydroxyurea XVI

INK4 Kip LB M phase MAP MBT MCM MEF MMR mol MPF Mth MTOC Mytl NA NLS NP-40 NPM N-terminal ORC PACT PAGE PBS PCM PCNA PCR PKA PLK4 PMSG pRb Inhibitors of Cdk4 Kinase inhibitory protein Luria-Bertani broth Mitotic phase Mitogen activated protein Midblastula transition Minichromosome maintenance deficient Mouse embryonic fibroblasts Mailer's modified Ringer's solution mole Maturation promoting factor Methanobacterium thermoautotrophicum Microtubule organizing center Membrane-associated tyrosine-and threonine-specific cdc2-inhibitory kinase Numerical aperture Nuclear localization sequence Nonidet P40 Nucleophosmin Amino-terminal Origin recognition complex Pericentrin/AKAP450 centrosome targeting domain Polyacrylamide gel electrophoresis Phosphate-buffered saline Pericentriolar material Proliferating cell nuclear antigen Polymerase chain reaction Protein kinase A Polo-like kinase 4 Pregnant mare serum gonadotropin Retinoblastoma protein xvii

pre-IC pre-RC PVDF RPA S phase SAS6 sc SCF SDS Sf9 siRNA Sid TBST Thr TR Tris Tyr U20S UTR UV preinitiation complex prereplication complex Polyvinylidene fluoride Replication protein A DNA synthesis phase Spindle assembly abnormal 6 Sacchoromyces cerevisiae Skp-Cullin-F-box complex Sodium dodecylsulfate Spodoptera frugiperda 9 cell line Small-interfering RNA Synthetically lethal with Dpbl 1 Tris-buffered saline-tween 20 Threonine Tetracycline repressor Tris (hydroxymethyl) aminomethane Tyrosine U2 osteosarcoma cell line Untranslated region Ultraviolet xviii

CHAPTER I INTRODUCTION 1.1 Cell Cycle The function of the cell division cycle is to produce two genetically identical cells from one cell. Proper duplication and segregation of chromosomes and centrosomes into two daughter cells are the most fundamental processes necessary for the viability and integrity of life. More importantly, cell division must be achieved with extreme precision and reliability over thousands of generations. In order to achieve the required level of accuracy and fidelity, cell division is controlled by checkpoint surveillance mechanisms that prevent initiation and progression through each step in the cell cycle when genomic integrity is threatened. Mutations that inactivate or alter checkpoint responses often result in chromosome mutation, missegregation and aneuploidy, all of which are linked with uncontrolled cell growth, a hallmark of cancer. 1.1.1 Phases of the cell cycle The cell cycle is defined by two major distinct stages, interphase and M phase, both of which can be divided into subphases (Figure 1.1). Interphase consists of three morphologically indistinguishable subphases, Gap phase 1 (Gi), DNA synthesis (S) and Gap phase 2 (G2). Gi is the point at which mammalian cells enter the cell cycle upon completion of mitosis or from a quiescent state, termed G zero (Go), in response to mitogenic stimuli. Cells progressing through Gi increase in size, are highly active metabolically, and have a 2n or diploid DNA content. Additionally, cells in Gi prepare to duplicate multiple organelles including centrosomes, a small non-membranous organelle 1

G2-M Transition Influenced by: cell size, DNA replication and DNA damage Metaphase- Anaphase Transition Influenced by: chromosome attachment to spindle and tension mitogenic stimulation Gj Restriction Point (Start) Influenced by: growth factors, nutrients, cell size and DNA damage Figure 1.1 The cell cycle and checkpoints Mitogenic stimulation of quiescent cells in G0 phase causes cell cycle reentry into Gj phase. In Gl5 cells grow and prepare for entry into S phase. During S phase, the genome and centrosomes are replicated, followed by G2 phase and mitosis when the duplicated genome is segregated into two identical daughter cells. The cell cycle contains specific cell cycle checkpoints that monitor DNA damage, completion of DNA duplication, and segregation of DNA (boxes). Activation of a checkpoint can transiently halt cell cycle progression until conditions are favorable. 2

most commonly described as the major microtubule-organizing center (MTOC) in animal cells. Length of Gi can vary greatly as cells monitor both extracellular and intracellular conditions to determine suitability for commitment to cell cycle progression. If circumstances are unfavorable, cells delay progress through Gi and may even enter Go. However, Gi phase exhibits a restriction "R" point, defined as the point in late Gi after which cells enter S phase even in the absence of growth factors [1]. If properly stimulated by mitogenic factors and intracellular conditions such as cell size and DNA integrity are favorable, the cell transverses the R point and commits to S phase. S phase is the period of time in which cells are actively replicating chromosomes and centrosomes. An essential aspect of the cell division cycle is that cells accurately copy each chromosome and centrosome only once. Cells in S phase can be identified by labeling methods utilizing the incorporation of radioactive thymidine or a synthetic nucleoside analog such as bromodeoxyuridine (BrdU) into actively replicating DNA. During S phase, replication increases the DNA content from 2n to 4n; so cells during S phase have a range of DNA content. Once cells have completed DNA synthesis, they proceed into G2 phase. Similar to Gi, G2 is an intermediate phase when cells continue to grow, duplicate organelles, and synthesize proteins needed for the next cell cycle step, mitosis. This is the last opportunity before cell division to ensure DNA has been properly duplicated without mutations and is therefore sometimes described as a "safety gap" [2]. M phase consists of a series of dramatic events comprising mitosis, or nuclear division, followed by cytokinesis, or cytoplasmic division. The discovery of mitosis is attributed to Walther Flemming, who in 1882 identified and characterized through light 3

microscopy the four basic phases of the mitotic cycle; prophase, metaphase, anaphase and telophase [3]. However, present analysis includes a mid-transition step, prometaphase. These phases are largely defined by specific physical characteristics and movements of the chromosomes (Figure 1.2). Upon entry into prophase, chromosomes begin to condense into highly ordered structures termed chromatin fibers. Duplicated centrosomes separate and start to move to opposite sides of the nucleus. During late prophase, often termed prometaphase, the nuclear envelope breaks down and an elaborate protein complex, the kinetochore, assembles at centromeres on each sister chromatid. Each chromosome has two "sister" kinetochores positioned on opposite sides that bind microtubules emanating from each spindle pole, forming a bipolar attachment of each chromosome. By the end of prometaphase, centrosomes align at opposite poles of the cell and form the bipolar mitotic spindle. Chromosomes attached to spindle poles by highly dynamic kinetochore microtubules are pushed and pulled until pole-directed forces are balanced, allowing cells to reach metaphase, discernible by the alignment of chromosomes at the metaphase (or equatorial) plate. Metaphase accounts for a significant portion of the total time in mitosis as chromosome attachment and alignment is required for progression into anaphase and is a major point of cell cycle control. Cells can remain arrested in metaphase for hours or even days until kinetochores are fully attached and under equal tension with microtubules. Once chromosomes are aligned on the metaphase plate, cells enter anaphase. During anaphase, sister chromatids separate and move away from the metaphase plate 4

Interphase prophase metaphase anaphase cytokinesis Figure 1.2 The stages of mitosis and cytokinesis Interphase precedes mitosis and is the period of time during which cells grow and duplicate both centrosomes (yellow) and chromosomes (red). During G2 phase, duplicated chromosomes are decondensed and not visible as distinct structures. As cells enter prophase, the chromosomes begin to condense, centrosomes separate and move to opposite sides of the nucleus, and the nuclear membrane disassembles. As prophase progresses, chromosome condensation and nuclear envelope breakdown are completed and duplicated centrosomes begin to form the bipolar spindle. In late prophase, or prometaphase, chromosomes move towards the equator of the cell and in metaphase, chromosomes are aligned on the metaphase plate. During anaphase, the sister chromatids separate and are segregated to opposite poles. Telophase is defined by the reformation of nuclear envelopes around the daughter nuclei. Cytokinesis begins as a cleavage furrow which continues to contract and forms two separate daughter cells. Figure based on Jackson et al., 2007. Immunofluorescent images (lower) reprinted with permission from Dr. William Earnshaw. Green: a-tubulin; red:DNA 5

toward each respective pole. At the end of anaphase, a complete set of chromosomes has assembled at each pole of the cell and cells enter the final stage of mitosis, telophase. During telophase a nuclear envelope reforms around each set of chromosomes and forms two nuclei within one cell. The chromosomes decondense and kinetochore microtubules depolymerize while polar fibers lengthen, elongating the cell. Once telophase is achieved, M phase is completed by a final step, cytokinesis. The beginning of cytokinesis is evident by the formation of an indentation or cleavage furrow. This process entails the formation of a contractile ring composed of short actin bundles and the force of myosin motor proteins. The ring contracts until the original membrane is separated into two, and cell division is complete. 1.1.2 Cell cycle control 1.1.2.1 Cyclin-dependent kinase (Cdk) The driving force for promoting forward movement through the cell cycle largely derives from the temporal activation and inactivation of multiple cyclin-dependent kinases (Cdks) [4, 5]. Like other protein kinases, Cdks catalyze the covalent attachment of phosphate groups from ATP to protein substrates. This reversible modification results in changes in the substrate that can include enzymatic activity, interaction with other proteins, localization, or degradation. Cdks are composed of a catalytic subunit whose activation is dependent on binding to regulatory subunit proteins called cyclins. In mammalian cells, there are four main classes of regulatory cyclins (D, A, E and B) and four Cdk catalytic subunits (Cdk 6, 4, 2 and 1) directly involved in cell cycle control. Oscillation of Cdk activity depends on the availability of cyclins, resulting in the 6

formation of distinct Cdk-cyclin complexes that regulate the initiation and progression of specific cell cycle phases (Figure 1.3). Gi Cdk complexes include both cyclin D/Cdk4 and cyclin D/Cdk6 that are activated in early to mid-Gi phase in a mitogen-dependent manner. Mitogenic stimulation induces expression of the D-type cyclins, thereby coupling cell growth stimuli to entry into the cell cycle [6, 7]. The primary functional roles attributed to the activation of Gi Cdks are the phosphorylation and inactivation of the retinoblastoma tumor suppressor protein (pRb) and inactivation of the Cdk inhibitor, p27Kipl [8, 9]. Active pRb binds and inhibits transcription factors of the E2F family that are known to regulate numerous genes necessary for the Gi/S transition and DNA replication (Figure 1.4) [10, 11]. Phosphorylation of pRb by Gi Cdks causes the dissociation and activation of E2F, thereby initiating transcription of genes essential for cell cycle progression [12]. The increase in cyclin D protein binds Cdk4/6 and sequesters the cellular pool of p27K'pl from cyclin E/Cdk2 complexes, resulting in Cdk2 activation [13, 14]. One of the many genes whose expression is upregulated after cyclin D-mediated E2F release from pRb is cyclin E, which binds and activates the Gi/S kinase, Cdk2 [15]. Activated cyclin E/Cdk2 complexes form a classic positive feedback loop by continuing to phosphorylate and inactivate pRb, causing full release and activation of E2F and transcription of genes critical for S phase progression, including cyclin A and cyclin E itself [11, 16, 17]. S phase-promoting roles regulated by cyclin E/Cdk2 include initiation of DNA synthesis via Cdc6 and PCNA binding, chromatin loading of minichromosome maintenance proteins (MCM), recruitment of Cdc45 and polymerase ct-primase, activation of the phosphatase Cdc25A, inactivation of Cdk inhibitors such as p21Wa' and 7

cyclin B/Cdkl cyclin A/Cdk2 cyclin A/Cdkl cyclin E/Cdk2 cyclin D/Cdk4 cyclin D/Cdk6 B cyclin B G, phase S phase G2 phase mitosis Figure 1.3 Expression of the cyclins is temporally regulated Cell cycle progression is controlled by the activity of cyclin-dependent kinases and periodic expression of the activating cyclin subunits. In response to mitogenic stimulation, quiescent cells enter the cell cycle at Gj phase and Cdk 6/4 are activated upon synthesis of cyclin D, thereby promoting progression through Gj. Transition from G, to S phase and initiation of DNA synthesis is triggered by a dramatic increase in cyclin E expression in late G[. Cyclin E is quickly degraded in mid-S phase. Cyclin A is up-regulated and persists until prometaphase of mitosis. Cyclin B/Cdkl complexes begin to accumulate in late S/early G2 phases but are held inactive until the initiation of mitosis by inhibitory phosphorylation. Activation in late G2 drives cells into mitosis while degradation in mid-mitosis promotes mitotic exit. 8

Mitogens I Cdk2 \ cyclin E/ J E2F J Cdk4/6 N^c yc l JnP/ > Q p RBj E2F j B DNA repair Rad51 BRCA1 CtIP rir51 RPA s ^ / Cell-cycle pl07 Cdc25A Cyclin A Cyclin E Cdkl Cdk2 E2F \ % * 1 Mitosis TTK Cdc25C Smc4 Smc2 Securin CI iNP-E DNA replication MCM4 MCM5 MCM6 Orel Cdc6 Thymidine kinase DNA a-polymerase PCNA RPA Figure 1.4 Phosphorylation of pRb is essential for cell cycle progression A) A simplified schematic of the pRB-E2F pathway. In G0 and early G„ pRb binds and inhibits E2F. Phosphorylation of pRB by both cyclin D-associated kinases and cyclin E/Cdk2 during G, leads to the release of E2F transcription factors. Figure based on Baker and McKinnon, 2004. B) Liberation of the E2F transcription factors induces the upregulation of numerous genes that mediate cell cycle progression and DNA repair. The list of genes is not comprehensive as many other genes in each category have been identified. 9

p27Kipl, and initiation of histone synthesis (Figure 1.5) [18-22]. Additionally, cyclin E/Cdk2 has been strongly implicated in regulating centrosome duplication, which is initiated near the Gi/S transition, and early experiments inhibiting or depleting either Cdk2 or cyclin E block duplication [23-25]. A crucial role for cyclin E/Cdk2 during centrosome replication is supported by the identification of essential centrosome-specific substrates including nucleophosmin/B23 and CP110 [26-28]. Cyclin A gene expression, first detected at or shortly after the Gi/S transition, is mediated by the transcription factor E2F and can be induced by expression of either cyclin D or cyclin E [29]. Cyclin A activates two different Cdks, Cdk2 and Cdkl that mediate roles in both S phase progression and mitotic entry, respectively [30, 31]. Cyclin A peaks during G2 and is degraded by the anaphase promoting complex/cyclosome (APC/C) slightly ahead of cyclin B in prometaphase. During S phase, cyclin A/Cdk2 associates with chromatin-bound replication complexes and is essential for late DNA replication origin firing [32, 33]. Cyclin A/Cdk2 additionally inhibits the assembly of new replication complexes on chromatin, preventing reinitiation of DNA synthesis [34, 35]. Cyclin A/Cdk2 has also been implicated in the regulation of centrosome duplication [36, 37]. Furthermore, cyclin A/Cdkl regulates entry into mitosis, as inhibition or down- regulation of cyclin A delays entry into mitosis [38-40]. To date, most studies of cyclin A have focused primarily on its S phase functions when bound to Cdk2 and the precise role and potential substrates of cyclin A/Cdkl during the G2/M transition remain to be elucidated. However, in the absence of significant cyclin B/Cdkl activity, exogenous cyclin A/Cdkl induces chromosome condensation, indicating a direct regulation of early 10

E2F Cdc6 PCNA NPM/B23 CP110 Histone biosynthesis \ SWI/SNF BAF155 Id2,Id3 Centrosome duplication Regulation of cyclin E/Cdk2 activation Initiation of DNA replication Chromatin remodeling and transcription Figure 1.5 Summary of known cyclin E functions Schematic of known substrates and their functional role. Cdk2/cyclin E regulates its own activation by phosphorylating and triggering the degradation of p21 and p27, while stimulating its own transcription by phosphorylation of pRb and release of E2F. Cdk2/cyclin E also upregulates the phosphatase Cdc25 A that activates Cdk2/cyclin E complexes. Additionally NPAT, a Cdk2/cyclin E substrate, acts as a transcriptional regulator of histone genes. Identification of centrosomal substrates, NPM/B23 and CP110, indicates Cdk2/cyclin E plays an essential role in regulating centrosome duplication. Finally, cyclin E has been shown to modulate the activity of chromatin remodeling proteins to maintain chromatin in a transcriptionally permissive state. Figure based on Moroy and Geisen, 2004. 11

prophase events [38, 39]. There is also evidence suggesting cyclin A is involved in activation of the major mitotic kinase, cyclin B/Cdkl [40, 41]. The key regulator of the M phase transition is cyclin B/Cdkl, also called maturation or M-phase promoting factor (MPF) [42, 43]. MPF initiates the major mitotic events from mid-prophase to anaphase including centrosome separation, nuclear envelope breakdown, spindle assembly, and chromosome condensation by as yet largely unidentified mechanisms [44-46]. Cyclin B/Cdkl activation begins in the cytoplasm and is particularly prominent at centrosomes [47, 48]. However, once a threshold of activation is achieved, MPF can provoke its own activation leading to a rapid increase in kinase activity [49-51]. In late prophase, cyclin B/Cdkl accumulates on centrosomes and in the nucleus and promotes the initiation and progression of mitosis [52, 53]. The completion of anaphase and subsequent exit from mitosis requires the inactivation of Cdkl via cyclin B degradation, as chemical inhibition or expression of a non-degradable form of cyclin B arrests cells in early anaphase [54, 55]. The loss of Cdkl activity leads to reassembly of the nuclear envelope, chromatin decondensation and disassembly of the mitotic spindle [56, 57]. Additionally, pre-replication complexes are reloaded onto chromatin, licensing DNA replication for the next mitotic cycle [58, 59]. 1.1.2.2 Regulating the regulators Regulation of Cdk activity is not solely dependent on cyclin binding, and Cdk complexes are subject to phosphorylation and dephosphorylation of specific residues. Activation can only be achieved after phosphorylation of the cyclin/Cdk on a conserved threonine residue that lies within the activation loop, or T loop of the kinase (Thr 172 in Cdk4/6, Thrl60 in Cdk2 and Thrl61 in Cdkl) [60, 61]. In its unphosphorylated form, 12

Full document contains 228 pages
Abstract: Centrosome duplication and DNA replication are fundamentally similar processes in that both occur once-and-only-once during each cell cycle. Both events initiate at the GUS transition and require an increase in Cdk2 activity. It has long been evident that the centrosome duplication and DNA replication cycles must not only be highly regulated but also temporally coordinated to prevent inappropriate re-replication. Abnormal centrosome numbers are described in virtually all human malignancies and have been shown to be a probable cause of chromosomal instability. Previous work in this laboratory defined a 20 amino acid modular centrosomal localization signal (CLS) in cyclin E that is both necessary and sufficient for centrosomal localization. Furthermore, it was shown that expression of the CLS displaces both endogenous cyclins E and A from centrosomes and prevents BrdU incorporation into DNA. To identify interactions mediated by the CLS, a HeLa cDNA library was screened using a bacterial two-hybrid assay with the CLS as bait. In this screen MCM5, an essential DNA replication factor, was isolated. Co-immunoprecipitation experiments and GST-pulldown assays verified a direct, in vivo interaction between MCM5 and cyclin E that is dependent on a wild-type CLS but independent of Cdk2. Interaction with cyclin E accounts for MCM5 localization on centrosomes in mammalian cells, and ectopic expression of MCM5 inhibits centrosome over-duplication in S phase-arrested CHO cells. MCM5 and cyclin A also interact in a manner dependent on a functional CLS domain but not on Cdk binding. Expression of the cyclin E CLS in G1/S synchronized cells inhibits S phase due to a lack of chromatin loading of proteins essential for the initiation of DNA replication. Rescue of DNA replication was evident when active cyclin E/Cdk2 was re-targeted to centrosomes. Taken together, the work presented here reveals novel mechanisms of communication between the centrosome and nuclear replication cycles in which elements of each cycle can directly regulate events in the other.