USF Libraries
USF Digital Collections

Molecular mechanism of Aurora-A kinase in human oncogenesis

MISSING IMAGE

Material Information

Title:
Molecular mechanism of Aurora-A kinase in human oncogenesis
Physical Description:
Book
Language:
English
Creator:
He, Lili
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:

Subjects

Subjects / Keywords:
E2F3
CHFR
CNTD2
Centrosome
Cell cycle
Dissertations, Academic -- Pathology and Cell Biology -- Doctoral -- USF   ( lcsh )
Genre:
non-fiction   ( marcgt )

Notes

Summary:
ABSTRACT: Aurora-A is a mitotic kinase, which regulates cell cycle progression through modulating centrosome function. Aurora-A expression is frequently altered in human malignancies. The discrepancy between overexpression and amplification suggests that elevated Aurora-A level is likely to be regulated also by transcriptional and/or translational modifications. In this study, we have demonstrated: 1) transcriptional regulation of Aurora-A by E2F3; 2) feedback regulation between tumor suppressor CHFR and Aurora-A; 3) CNTD2 as a novel Aurora-A partner and oncogene to activate Aurora-A and induce transformation. Aurora-A expression and activity are cell cycle regulated. The mechanism of Aurora-A upregulation at onset of mitosis is largely unknown. We demonstrated, for the first time, that transcription factor E2F3 directly binds to Aurora-A promoter and tightly regulates Aurora-A expression during G2/M phase.Moreover, expression of E2F3 considerably correlates with Aurora-A level in human ovarian cancer, indicates that E2F3 is a causal factor for Aurora-A overexpression. Thus, E2F3-Aurora-A axis could be an important target for cancer intervention. The frequent inactivation of prophase checkpoint CHFR caused by DNA methylation or mutation has been reported in human cancers. We showed that CHFR is hypermethylated in ovarian carcinoma. Aurora-A phosphorylates CHFR on Ser-218 and Ser-337 in vivo and in vitro, which inhibits CHFR ubiquitin ligase activity. The feedback regulation loop between Aurora-A and CHFR could play a critical role in regulation of cell cycle progression, imbalance of which may contribute to human oncogenesis. Using yeast 2-hybrid screening, we identified a splicing form of CNTD2 as Aurora-A interacting protein. CNTD2 locates to chromosome 19q13.2 AKT2 amplicon.CNTD2 is amplified and overexpressed in human ovarian, pancreatic and lung cancer cell lines and primary tumors. CNTD2 colocalizes and interacts with Aurora-A in the centrosome. CNTD2 expression induces Aurora-A and cdc2 kinase activity, G2/M progression, and malignant transformation. These data indicate that CNTD2 is an oncogene and could play a pivotal role in Aurora-A activation during the cell cycle and that disruption of CNTD2-Aurora-A axis may represent a potential means to antitumor drug discovery.
Thesis:
Dissertation (Ph.D.)--University of South Florida, 2008.
Bibliography:
Includes bibliographical references.
System Details:
Mode of access: World Wide Web.
System Details:
System requirements: World Wide Web browser and PDF reader.
Statement of Responsibility:
by Lili He.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 116 pages.
General Note:
Includes vita.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 002021681
oclc - 428816627
usfldc doi - E14-SFE0002582
usfldc handle - e14.2582
System ID:
SFS0026899:00001


This item is only available as the following downloads:


Full Text

PAGE 1

Molecular Mechanism of Aurora-A Kinase in Human Oncogenesis by Lili He A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Pathology and Cell Biology College of Medicine University of South Florida Major Professor: Jin Q. Cheng, M.D., Ph.D. Santo V. Nicosia, M.D. Domenico Coppola, M.D. Patricia A. Kruk, Ph.D. Jie Wu, Ph.D. Date of Approval: July 7, 2008 Keywords: E2F3, CHFR, CNTD2, cen trosome, cell cycle, cancer Copyright 2008 Lili He

PAGE 2

This work is dedicated to my mother, Cuiqing Xu, and my father, Chunnian He.

PAGE 3

ACKNOWLEDGEMENTS A great many people have contributed to this wo rk. I owe my thankfulne ss to all of those people who have made this dissertation possible. My deepest gratitude is to my mentor, Dr Jin Q. Cheng. It is him who gave me this great study opportunit y. His guidance was paramount in providing a well rounded experience consistent with my long-term career goals. His illustrious accomplishments in cancer research confirm him a successful sc ientist. His dedication and vigorousness inspire people. My appreciation goes to Department Chair Dr. Santo V. Nicosia for accepting me into the ovarian cancer research program. W ithout guidance and help from Dr. Nicosia, I could not complete my work. Many thanks to my committee members, Dr Domenico Coppola, Dr. Patricia A. Kruk, Dr. Jie Wu, whose insightful advice, unfailingly patience and encouragement allowed me to finish this dissertation. I want to give special acknowledgement to Dr. Hua Yang and Dr. Mei Sun. They enlightened me in the research field. I am thankful to the current and former members of Dr. Cheng’s lab, for their various forms of support during my graduate study. I am also grateful to Dr. Yihong Ma, Dr. Yongping Cui, Dr. Yu Pan, Dr. Yingtao Zhang and Dr. Zhong Zheng; Dr Kruk’s lab and Dr. Bai’s lab, and all the people w ho have ever offered priceless help. I would like to give my appr eciation to staff of Office of Graduate & Postdoctoral Affairs, Dr. Abdul S. Rao, Dr. Michael J. Ba rber, Dr. Eric S. Bennett, Mr. Thomas E. Moore-Prizon, Ms Kathryn Zahn, Ms Susan Chapman and Ms Franjesca Jackson. Last but not least, I than k my family and my friends who have comforted and sustained me all the way through my graduate study. I deeply apprec iate their belief in me.

PAGE 4

i TABLE OF CONTENTS LIST OF TABLES...............................................................................................................v LIST OF FIGURES...........................................................................................................vi ABSTRACT.....................................................................................................................vi ii CHAPTER I INTRODUCTION..........................................................................................1 Aurora Kinases.........................................................................................................1 Structure.......................................................................................................1 Localization and Timing of Activation........................................................2 Chromosome Location and Invol vement in Human Cancer........................4 Aurora-A, but not Aurora-B and -C, Induces Transformation, Centrosome Amplification and Genetic Instability................................5 Aurora-A Kinase Activity is Re quired for Its Oncogenic Activity..............5 Aurora-A and Mitosis..............................................................................................6 Centrosome Maturation................................................................................8 Centrosome Separation................................................................................9 Mitotic Commitment..................................................................................10 Bipolar Spindle Assembly.........................................................................10 Chromosome Alignment on the Metaphase Plate......................................12 Cytokinesis.................................................................................................12 Aurora-A Signaling................................................................................................13 p53..............................................................................................................13 c-Myc.........................................................................................................14 AR and ER.................................................................................................15 Aurora-A and Tumorigenesis................................................................................16 Centrosome Amplification and Aneuploidy..............................................16 Checkpoint Disruption and Chromosome Instability.................................17 Polymorphism............................................................................................18 Aurora-A Inhibitors...............................................................................................18 References..............................................................................................................20 CENTRAL HYPOTHESI S AND OBJECTIVES.............................................................27

PAGE 5

ii CHAPTER II IDENTIFICATION OF AUROR A-A AS A DIRECT TARGET OF E2F3 DURING G2/M C ELL CYCLE PROGRESSION.............................................28 Abstract..................................................................................................................28 Introduction............................................................................................................29 Materials and Methods...........................................................................................31 Cell Lines, Culture and Transfection.........................................................31 Synchronization and Cell Cycle Analyses.................................................31 Plasmids.....................................................................................................31 Luciferase Assay........................................................................................32 Northern and Western Blotting Analyses..................................................32 shRNA Lentiviral Infection.......................................................................33 Chromatin Immunoprecipitation (ChIP) Assay.........................................33 Results....................................................................................................................35 Expression of the Members of E2F Family and Aurora-A during the Cell Cycle............................................................................35 Ectopic Expression of E2F3 Induces and Knockdown of E2F3 Decreases Aurora-A Protein and mRNA Levels..................................35 Knockdown of E2F3 Delays G2/M Entry and Reduces Aurora-A Expression during the Cell Cycle.........................................................38 E2F3 Binds to and Transactivates Aurora-A Promoter during G2/M Phase.....................................................................................................38 Correlation of Expression of E2F3 and Aurora-A.....................................43 Discussion..............................................................................................................46 References..............................................................................................................49 CHAPTER III AURORA-A FEEDBACK RE GULATION OF CHFR, A TUMOR SUPPRESSOR FREQUENTLY DOWN REGULATED IN OVARIAN CANCER...............................................................................................................52 Abstract..................................................................................................................52 Introduction............................................................................................................53 Materials and Methods...........................................................................................55 Cell Lines, Transfection and Treatment.....................................................55 Plasmids.....................................................................................................55 RT-PCR......................................................................................................56 Western Blotting Analysis and Immunoprecipitation................................56 In vivo [32P] Pi Labeling.............................................................................57 In vitro Aurora-A Kinase Assay................................................................57 2-Dimensioanl Phosphopeptide Mapping and Edman Degradation..........58 In vitro Ubiquitination Assay.....................................................................59

PAGE 6

iii Results....................................................................................................................60 Downregulation of CHFR in Human Ovarian Cancer and the Correlation Between Expressi on of CHFR and Aurora-A...................60 Aurora-A Phosphorylates CHFR in vivo and in vitro ................................61 Mapping Phosphorylation Site s of CHFR by Aurora-A............................63 Mutation of the Phosphorylation Si tes Abolishes CH FR E3 Activity.......66 Discussion..............................................................................................................69 References..............................................................................................................71 CHAPTER IV CYCLIN N-TERMINAL DOMAIN CONTAINING 2, CNTD2, IS AN ONCOGENE THAT INTERACTS WITH AND ACTIVATES AURORA-A..........................................................................................................74 Abstract..................................................................................................................74 Introduction............................................................................................................75 Materials and Methods...........................................................................................77 Cell Culture, Transfec tion and Treatment..................................................77 Plasmid Constructions................................................................................77 Yeast Two-Hybrid Screen..........................................................................78 Western Blotting and Immunoprecipitation...............................................78 Northern and Southern Blot.......................................................................79 Indirect Immunofluorecsence (IIF)............................................................79 In vitro Aurora-A, p34cdc2 Kinase Assay.................................................80 Soft Agar Assay and Tumorigenicity in Nude Mice..................................81 Results....................................................................................................................82 Identification of CNTD2 as an Aurora-A Interacting Protein...................82 CNTD2 Colocalizes with Aurora-A in Centrosome..................................82 CNTD2 Stabilizes Aurora-A Protein a nd Stimulates Aurora-A Kinase Activity.................................................................................................85 CNTD2 Promotes G2/M Progression and Activates cdc2 Kinase Activity.................................................................................................87 CNTD2 is Ubiquitously Expressed in Normal Tissues and is Frequently Altered in Human Cancers...................................................................89 Ectopic Expression of CNTD2 Indu ces Malignant Transformation..........95 CNTD2 Expression Induces Cent rosome Amplification and Genomic Instability..............................................................................97 Discussion..............................................................................................................99 References............................................................................................................101 CHAPTER V DISCUSSI ON AND CONCLUSION......................................................106

PAGE 7

iv References............................................................................................................111 APPENDICES.................................................................................................................115 Appendix I. List of Publications..........................................................................116 ABOUT THE AUTHOR.......................................................................................End Page

PAGE 8

v LIST OF TABLES Table 1. Aurora substrates/interactin g proteins and mitotic processes................................7 Table 2. Summary of expression of E2F3 and Aurora-A in ovarian cancer......................45 Table 3. CNTD2 expression in ovarian tumors and normal tissues..................................95 Table 4. Association of CNTD2 expre ssion with tumor histological types......................95 Table 5. Correlation of CNTD2 expressi on level with tumor grade and stage..................95 Table 6. Tumorigenicity of CLB6 and control transfected 3Y1 cells................................97

PAGE 9

vi LIST OF FIGURES Figure 1.1. Aurora kinase family.......................................................................................2 Figure 1.2. Localization of Aurora-A and -B kinases during the cell cycle......................3 Figure 1.3. Aurora-A regulation of centrosomal microtubule assembly...........................9 Figure 1.4. Regulation of Aurora -A activity by Ran-GTP and TPX2.............................11 Figure 1.5. Mechanisms of Aurora-A promoted tumorigenesis......................................16 Figure 2.1. The correlation of expression of E2F family proteins and Aurora-A during the cell cycle................................................................................................................. 36 Figure 2.2. E2F3 regulates Aurora-A e xpression at protein and mRNA levels..............37 Figure 2.3. Knockdown of E2F3 results in m itosis entry delay and decreased Aurora-A expression during the cell cycle...................................................................................39 Figure 2.4. Aurora-A promot er is regulated by E2F3......................................................40 Figure 2.5. E2F3 binds to Aurora-A prom oter in a cell cycle dependent manner...........42 Figure 2.6. Expression of E2F3 increas es Aurora-A density in centrosome...................43 Figure 2.7. Overexpression of E2F3 correlate s with Aurora-A level in human ovarian tumors......................................................................................................................... ..45 Figure 3.1. Frequent downregulatio n of CHFR in ovarian cancer..................................61 Figure 3.2. CHFR is ph osphorylated by Aurora-A..........................................................62 Figure 3.3. Define Aurora-A phosphorylation sites of CHFR.........................................64 Figure 3.4. Confirm Aurora-A phosphorylation sites of CHFR......................................66 Figure 3.5. Aurora-A inhibits CHFR E3 activity through phosphoryl ation of Ser218 and Ser337......................................................................................................................... ..68 Figure 4.1. CNTD2 sequences.........................................................................................83 Figure 4.2. Interaction and coloca lization of CNTD2 and Aurora-A..............................84 Figure 4.3. CNTD2 stabilizes Au rora-A protein and induces Aurora-A kinase activity.86

PAGE 10

vii Figure 4.4. CNTD2 promotes G2 /M progression and stimulates cdc2 kinase activity...88 Figure 4.5. CNTD2 expression in normal tissues and human cancer cell lines...............90 Figure 4.6. CNTD2 is amplifie d in human cancer cell lines...........................................91 Figure 4.7. CNTD2 is overexpressed in human cancer...................................................93 Figure 4.8. Overexpression of CNTD2 in ova rian tumors is associated with poor prognosis...................................................................................................................... 94 Figure 4.9. Ectopic expression of CNTD2 transforms 3Y1 cells....................................96 Figure 4.10. CNTD2 induces cen trosome amplification and genomic instability.............98 Figure 5.1. Working scheme of molecu lar mechanism of Au rora-A in human oncogenesis................................................................................................................110

PAGE 11

viii Molecular Mechanism of Aurora-A Kinase in Human Oncogenesis Lili He ABSTRACT Aurora-A is a mitotic kinase, which regulat es cell cycle progre ssion through modulating centrosome function. Aurora-A expression is fr equently altered in human malignancies. The discrepancy between overexpression and amplification suggests that elevated Aurora-A level is likely to be regulated al so by transcriptional and/or translational modifications. In this study, we have dem onstrated: 1) transcriptional regulation of Aurora-A by E2F3; 2) feedback regulation between tumor suppressor CHFR and AuroraA; 3) CNTD2 as a novel Aurora-A partner and oncogene to activate Aurora-A and induce transformation. Aurora-A expression and activity are ce ll cycle regulated. The mechanism of Aurora-A upregulation at onset of mitosis is largely unknown. We demonstrated, for the first time, that transcription factor E2F3 dire ctly binds to Aurora-A promoter and tightly regulates Aurora-A expression during G2/M phase. Moreover, expression of E2F3 considerably correlates with Aurora-A level in human ovarian cancer, indicates that E2F3 is a causal factor for Aurora-A overexpressi on. Thus, E2F3-Aurora-A axis could be an important target for cancer inte rvention. The frequent inactiva tion of prophase checkpoint CHFR caused by DNA methylation or mutation ha s been reported in human cancers. We showed that CHFR is hypermethylated in ovarian carcinoma. Aurora-A phosphorylates CHFR on Ser-218 and Ser-337 in vivo and in vitro which inhibits CH FR ubiquitin ligase activity. The feedback regulation loop between Aurora-A and CHFR c ould play a critical

PAGE 12

ix role in regulation of cell cy cle progression, imbalance of which may contribute to human oncogenesis. Using yeast 2-hybrid screening, we identified a splici ng form of CNTD2 as Aurora-A interacting protein. CNTD2 lo cates to chromosome 19q13.2 AKT2 amplicon. CNTD2 is amplified and overexpressed in human ovarian, pancreatic and lung cancer cell lines and primary tumors. CNTD2 colocaliz es and interacts with Aurora-A in the centrosome. CNTD2 expression induces Auro ra-A and cdc2 kinase activity, G2/M progression, and malignant transformation. Th ese data indicate that CNTD2 is an oncogene and could play a pivotal role in Aurora-A activation during the cell cycle and that disruption of CNTD2-Aurora-A axis ma y represent a potential means to antitumor drug discovery.

PAGE 13

1 CHAPTER I INTRODUCTION: MOLECULAR MECH ANISM OF AURORA-A KINASE IN HUMAN ONCOGENESIS Aurora Kinases Homologues of Aurora/Ipl1-related kinase ha ve been reported in various organisms including yeast, nematodes, fruit flies and ve rtebrates (1-3). In mammals, this subfamily of serine/threonine kinases comprises thr ee members: Aurora-A, -B and -C (4). Drosophila melanogaster and Caenorhabditis elegans express homologues of Aurora-A and -B, whereas Saccharomyces cerevisiae and Schizosaccharomyces pombe have only one Aurora kinase gene (5,6) suggesting that th e functions of Auroras have diverged throughout evolution. Structure All Aurora kinases share similar structures, with their catalytic domains flanked by very short (15–20 residues) C-termin al tails and N-terminal domains of variable lengths (39– 129 residues). The overall homology between thes e three isoforms in human is about 80% at amino acid level. The central catal ytic kinase domain is considered highly conserved. However, it is interesting to not e that Aurora-C and -B share 77.6% amino acid sequence identity, while Aurora-C and -A share about 66.5% sequence identity (7), suggesting a functional relations hip between Aurora-B and -C. The N-terminal and the Cterminal sequences of the three members ar e quite different. The C-terminal domain of human Aurora-B shares 53% and 73% sequenc e similarity to huma n Auroras-A and -C, respectively. The N-terminals of Auroras-A, -B and -C share low sequence conservation, which may determine the sele ctivity during protein–protein interactions. Aurora-A N-

PAGE 14

2 terminal domain serves as a localizati on domain, which targets the protein to the centrosome in a microtubule dependent manner. The unique A-box motif at the Nterminal of Aurora-A is required for pr oteolysis of the protein. Furthermore, phosphorylation of Ser 51 at N-terminal decide s the timing of Aurora-A destruction (8). (see Fig. 1.1) Fig. 1.1. Aurora kinase family. Diagrammatic representation of the domain structure of three Aurora family members. The percen tages indicate the degree of the identity between Aurora-A, Aurora-B and Aurora-C. Localization and Timing of Activation Despite these similarities, the three mammalia n Aurora family members differ in their expression patterns, subce llular localization, and timi ng of activity. (see Fig. 1.2) Aurora-A is first detected at the centrosome during late S phase. The activation of a small proportion of Aurora-A at centroso mes was first evident before chromosome condensation at late G2 phase (9). Both th e amount and activity of Aurora-A rapidly increase in the centrosome, and a fraction of active Aurora-A subsequently translocates into the nucleus coincident with chromatin condensati on at prophase. After nuclearenvelope breakdown (NEBD), activated Aurora -A is observed at th e spindle poles and bipolar spindles during prometaphase and me taphase. The amount of Aurora-A starts to decrease at the metaphase-anaphase transi tion, whereas a small fraction of Aurora-A remains on the centrosomes and the spindles at the onset of anaphase and telophase

PAGE 15

3 (10,11). Aurora-A is degraded by the cdh1/Fizzy -related form of the anaphase-promoting complex/cyclosome (APC/C). At the final stag e of cytokinesis, most of the Aurora-A protein becomes undetectable. Aurora-B whose activity appears to reach ma ximal levels later in mitosis, displays the dynamic properties of a chromoso mal passenger protein. It first associates with centromeres/kinetochores — the sites on chromosomes where microtubules attach — then relocalizes to the midzone of the centr al spindle, and finally concentrates at the midbody between dividing cells (12). Fig. 1.2. Localization of Aurora-A and -B kinases during the cell cycle. Nature Reviews Cancer 5, 42-50. January 2005.

PAGE 16

4 In line with these distinct localizations Aurora-A is implicated primarily in centrosome maturation and spindle assembly (5,6), whereas Aurora-B is proposed to regulate chromosome condensation and cohe sion, kinetochore assembly and bipolar chromosome attachment, the spindle ch eckpoint and the coordination between chromosome segregation and cytokinesis (13-16). Aurora-C has been described only in mammals, wh ere it is expressed in testis and certain tumor cell lines and, like Aurora-B, f unctions as a chromosomal passenger, which co-localizes with Aurora-B firs t to centromeres and then to the midzone of mitotic cells. It has been shown recently that Aurora-C c ooperates with Aurora-B to regulate mitotic chromosome segregation and cytokinesis (17). Chromosome Location and Involvement in Human Cancer Human Aurora-A is located at chromosome 20q13.2, which is commonly amplified in various epithelial malignant tumors, includ ing breast, colon, bladder, ovarian and pancreatic cancers (18-22). In breast cancer in part icular, the 20q11-q13 region is amplified in 12-18% of primary tumors. The 20q1 3 amplification is also detected in 5% (4/72) of gastric cancer from 3.5to 6.3fol d, and in 52% (41/79) of primary colorectal tumors. The levels of Aurora-A mRNA and protei n are also increased in those tumors. However, alterations of Aurora -A at mRNA and/or protein levels are much common than the gene amplification. For example, amplification of Aurora-A was detected in only 3% of hepatocellular carcinoma s (HCCs) while more than 60% of HCCs overexpressed Aurora-A mRNA and protein (23). Disc repancy between amplification and overexpression rates was also reported in br east cancer (24), gastric cancer (25) and ovarian cancer (21). Therefore, Aurora-A ove rexpression is likely to be regulated not only by gene amplification, but also by othe r mechanisms such as transcriptional activation and suppression of protein degradation. Unlike Aurora-A Aurora-B coding gene, which is mapped to chromosome 17p13.1, is not amplified in human tumors. Ho wever, both mRNA and protein levels of Aurora-B are frequently increased in various human tumors, including colorectal cancers

PAGE 17

5 (26,27). Furthermore, exogenous overexpression of Aurora-B in Chinese hamster embryo cells causes chromosome separation defects dur ing mitosis and increa sed invasiveness of cells in vivo indicating a role for Aurora -B in tumorigenesis (28). A gene encoding Aurora-C is localized at chromosome 19q13.43, which is the region frequently deleted or rearranged in several tumor tissues (29). However, the involvement of Aurora-C in cancer development is still unclear. Aurora-A, but not Aurora-B and -C, Induces Transformation, Centrosome Amplification and Genetic Instability Ectopic expression of Aurora-A, but not Auro ra-B and -C, in Rat1 or NIH3T3 cells results in malignant phenotype, including growin g in soft agar and forming tumor in nice mouse, demonstrating that Au rora-A is an oncogene. Overex pression of Aurora-A leads to disrupt cell-cycle checkpoint functions. Normal cells have cell-cycle checkpoints that monitor genomic integrity and prevent cells from dividing in th e presence of DNA damage. Recent studies have shown that DNAdamage signals inhibit Aurora-A kinase activity to induce cell-cycle arrest at the end of the G2 phase and that, conversely, overexpression of Aurora-A disrupts the DNA-damage-induced G2 checkpoint (30). Additionally, Aurora-A overexpr ession has been found to overri de the spindle checkpoint activated by the chemotherapeutic agent paclitaxel (Taxol), allowing cells to inappropriately enter anaphase despite de fective spindle formation and to become resistant to paclitaxel-induced apoptosis (3 1). Many experiments have demonstrated the requirement for Aurora-A in centrosome regulation. In Drosophila Aurora-A mutants show several centrosome defects that were confirmed by RNA-mediated interference (RNAi) (32). Aurora-A overexpression in cell li nes leads to centrosome amplification, multipolar spindle and polyploidy. These ch eckpoint defects and genetic instability, caused by Aurora-A overexpression, mi ght contribute to transformation. Aurora-A Kinase Activity is Requ ired for Its Oncogenic Activity Transient expression of active Aurora-A in the near diploi d human breast epithelial cell line MCF10A leads to aberrant chromosome segregation and an increase in ploidy (24).

PAGE 18

6 In contrast, a C-terminal-deleted mutant of Aurora-A induces no effect in the ploidy state of transfected gastric cancer cells, but inhi bits cell prolifera tion (25). However, surprisingly, an inactive Auro ra-A kinase produces virtuall y identical phenotypes to the wild-type active kinase when overexpressed in Chinese hamster ovary cells (33). Both transfected cells exhibit extra copies of cen trosomes. However such inactive Aurora-A is unable to transform Rat1 or NIH3T3 cells, str ongly indicating that (1) a gain of Aurora-A kinase activity is necessary to transform cells and that (2) centrosome amplification is not sufficient to provoke tran sformation (19). Thus, overe xpression of the Aurora-A kinase is able to induce centrosome amplifica tion, but its kinase activity is not necessary to obtain this phenotype. It is possible that the accumulation of the protein leads to a disorganization of the centroso me machinery unrelated to its catalytic function but caused by a steric accumulation that interferes with the centrosomal functions. In contrast, the kinase activity of Aurora -A is necessary to obtain the transformed phenotype by overexpression in cultured cells Thus centrosome amplification induced by Aurora-A overexpression is not suffi cient for its oncogenic activity. Under overexpression, Aurora-A might phosphorylate s ubstrates unrelated to the centrosome function but involved in oncogenesi s. Furthermore, this suggests that kinase activity of Aurora-A is essential for its oncogenic activit y. In addition to elevat ed levels of mRNA and protein, we have demonstr ated frequent activation of Aurora-A in human prostate and ovarian cancers, indicating that activa tion of Aurora-A may be important for development of these subsets of tumors. Aurora-A and Mitosis In the process of separating replicated gene tic materials into tw o daughter cells during mitosis, cells undergo centrosome matura tion, chromosome condensation, nuclearenvelope breakdown, centrosome separation, bipolar-spindle assembly, chromosome segregation and cytokinesis. St udies in lower organisms have shown that these events are controlled by phosphorylation events perfor med by several serine /threonine kinases, known as mitotic kinases (4). Mitotic kinase s include cyclin dependent kinase 1 (CDK1;

PAGE 19

7 also known as p34cdc2), polo-like kinase s, NIMA-related kinases, WARTS/LATS1related kinases and Aurora/Ipl1-related kinase s. The structure of these enzymes has been well conserved through evolution. Elucidat ion of how these mitotic kinases are coordinated will provide a clue to the regulation of cell division. In human cells, Aurora-A levels and kina se activity increase during late G2 to M phase, and it undergoes dynamic changes in s ubcellular localization throughout the cell cycle. This dynamic localization pattern indicates that Aurora -A is involved in various mitotic events. In fact, several substrates and/or interacting proteins of Aurora-A identified recently have been shown to have a crucial role in mitotic processes (Table 1.). Table 1. Aurora substrates/interacti ng proteins and mitotic processes. Interacting protein that is also Aurora-A substrate.

PAGE 20

8 Centrosome Maturation The centrosome is the primary microtubule nucl eating centre within most animal cells at all stages of the cell cycle, but on entry in to mitosis, microtubule assembly at the centrosome dramatically increases in a process termed maturation. This activation enables microtubules to grow out from the centrosome, interact with chromosomes and form the bipolar spindle required for chromo some segregation. At the same time, many proteins are recruited to centrosomes, including the tubulin ring complex and other microtubule regulators. Aurora-A has been shown to be esse ntial for the process of centrosome maturation in various organisms such as Caenorhabditis elegans and Drosophila melanogaster (32,34,35). Also, in cultured human cells Aurora-A depletion results in the inhibition of centrosome maturation (9). The discovery of phosphorylation of a conserved centrosomal protein TACC (T ransforming A cidic C oiled-C oil) by Aurora-A could partly explain the molecular nature of Aurora-A regulation of cen trosomal microtubule assembly (32,36,37). Phosphorylation by Aurora-A recruits TACC, and consequentially its binding partner Msps/XMAP215 to the centrosome in mitosi s (38). Targeting of TACC–Msps/XMAP215 to the centrosome may enhance the ac tivity of Msps/XMAP215 in stabilizing microtubule plus ends (1). In another model, the phosphorylation is proposed to allow the complex to stabilize the minus ends of microtubules nucleat ed at the centrosome (2) (see Fig. 1.3). Other than for microtubule assembly, C. elegans homologue of Aurora-A kinase AIR-1 is required for centrosome maturation by recruiting centrosomal -tubulin and two other PCM (pericentriolar material) compone nts, ZYG-9 and CeGrip, as embryos enter mitosis (34).

PAGE 21

9 Fig. 1.3. Aurora-A regulation of ce ntrosomal microtubule assembly. MT, microtubules; -TuRC, -tubulin ring complex. JCB, Volume 170, Number 7, 1039-1046. 2005. Centrosome Separation Depletion of AIR-1 in C. elegans doesn’t affect centrosome separation before nuclearenvelope breakdown (NEBD); however, subseque nt to NEBD the tw o asters collapse back together and remain unseparated, indica ting that AIR-1 is not necessary for initial separation of centrosomes but is required to maintain centrosome separation during spindle assembly (34). In human cells (HeLa cells), micro-injection of affinity-purified Aurora-A antibody at late G2 phase inhibits separation of centriole pairs at prophase. Meanwhile, abnormally organized spindles with no centrioles at one po le and two pair of centrioles at the other are frequently noticed (10). It was previously shown that the Xenopus Aurora-related kinase Eg2 phosphorylates the kinesin-li ke protein XlEg5, which is required for centrosome separation (39). Therefore it is reasonable to speculate that human Aurora-A might also regulate separation of centriol e pairs through the phosphorylati on of kinesin-like motors during prometaphase.

PAGE 22

10 Mitotic Commitment Analysis of HeLa cells has revealed that th e initial active form of Aurora-A is first detected in late G2 phase at the centrosome. Application of RNA interference (RNAi) to synchronized cells has shown that this activ ation is required for the recruitment of CDK1–cyclin B1 to the centrosome, where it is fully activated and commits cell to mitosis. A LIM protein called Ajuba was identi fied as a substrate and an activator for Aurora-A during mitotic commitment. Depletio n of Ajuba prevents activation of AuroraA at centrosomes in late G2 phase and inhibits mitotic entry (9). Additionally, it has recently been dem onstrated that Aurora-A phosphorylates Ser353 of the phosphatase CDC 25B — an activator of cyc lin-dependent kinases — at centrosomes during late G2 phase, also c ontributing to the onset of mitosis (40). However, the activation of CDK1–cyclin B1 is in turn required for full activation of Aurora-A at the centrosome and nucleus (9 ). These findings indicate that the coactivation of Aurora-A and CDK1 at late G2 centrosomes constitutes an essential early event in the progression of cells towards mitosis. Bipolar Spindle Assembly The role of Aurora-A in mito tic-spindle assembly was init ially revealed through studies in Xenopus egg extracts (41), where the phospho rylation and activation of Eg2 ( Xenopus Aurora-related kinase) stimulated by the RanGTP-TPX2 pathway is essential for spindle assembly; a catalytically inactive pEg2 stops the assembly of bipolar mitotic spindles (42-44). TPX2 (T argeting P rotein for X KLP2) was isolated from Xenopus as a microtubule-associated protei n, which is required for the recruitment of XKLP2 ( Xenopus kinesin-like protein) to microtubul es (45). GTP-binding Ran (the small GTPase) is known to be required for the i nduction of spindle formation by chromosomes in M phase. And the action of Ran-GTP in spindle formation requires TPX2 (46). Depletion of TPX2 from Xenopus egg extracts resulted in the formation of less compact spindles and a range of pole (47). Additiona lly, in mammalian cells, depletion of TPX2 using small interfering RNA caused the fo rmation of multipolar spindles (48).

PAGE 23

11 Furthermore, TPX2 was found to interact with Aurora-A and to be not only a substrate but also an activator for Aurora-A. Inhibi tion of the interacti on between TPX2 and Aurora-A prevents Aurora-A activation and recruitment to microtubules (49,50). Earlier evidences established that de pletion of Aurora-A disrupts the stability of the mitotic spindle in Xenopus eggs and causes the formation of multipolar spindles in mammalian cells, similar to depletion of TPX2 (41,51). Th is indicates that th e interaction between TPX2 and Aurora-A could be an important prerequisite for efficient mitotic-spindle formation. (see Fig. 1.4) Fig. 1.4. Regulation of Aurora-A activity by Ran–GTP and TPX2. Nature Reviews Molecular Cell Biology 4, 842-854. November 2003.

PAGE 24

12 Chromosome Alignment on the Metaphase Plate Studies have shown that Aurora-B is requi red for the correct kinetochore-microtubule attachment between sister chromatids to sp indle poles (52,53). The proper interaction generates the amphitelic attachment state, which is required for accurate chromosome alignment and segregation (12). However, it ha s been recently shown that Aurora-A is also directly involved in metaphase ch romosome alignment, by phosphorylating a kinetochore-specific histone H3 variant CE NP-A, which is crucial for kinetochore assembly and function (54). CENP-A is phos phorylated on the amino-terminal serine (Ser7) by Aurora-A, and this phosphoryla tion is important for not only the proper attachment of microtubules to the kinetoch ore but also the consequent chromosome alignment and segregation. Given that posttranslational modifications of the aminoterminal tail domain of core histones ar e known to regulate chromatin structure and function (55), phosphorylation of Ser7 of CENP -A by Aurora-A might be a crucial event for kinetochore function. Although Aurora-B is also found to phosphorylate CENP-A at the same site (56), it is be lieved that Aurora-A phosphorylat ion is an earlier and more important event. Ser7 of CENP-A is firs t phosphorylated during prophase in a manner dependent on Aurora-A, and that this reacti on is required for the subsequent Aurora-Bdependent phosphorylation of CE NP-A as well as the recruitment of Aurora-B to the inner centromere, where Aurora-B mainta ins the phosphorylation of CENP-A from prometaphase through metaphase (54). The phosphorylation of CENP-A on Ser-7 is required to initiate kinetochore assembly by recruiting several components important for the establishment of kinetochore-microtubule connections. The two Aurora kinases therefore regulate kinetochore functions by phosphorylating this common substrate. Cytokinesis Both Aurora-A and Aurora-B seem to be invol ved in cytokinesis. Inhibition of Aurora-B function results in cytokinesis failure in mammalian cells (13,57-60). Inhibition of Aurora-A by antibody microinjection in meta phase HeLa cells, which have completed centrosome separation, bipolar spindle assembly and chromosome alignment, causes cytokinesis failure; chro mosomes are able to segregate, but the daughter cells eventually

PAGE 25

13 fuse to form binucleate cells (10). It is t hus possible that Aurora-A has some direct function in cytokinesis. The other possibility is that Aurora-A leads to cytokinesis defect by dysregulation of centriole behavior, whic h is proposed tightly connected with the completion of cell division. Interestingly, overexpression of Aurora-A also impairs cytokinesis, resulting in formati on of multinucleated cells (31,33). As both the inhibition and the increas ed activity of Aurora-A lead to multinucleated cell formation, proper timing of activation and subsequent inactivation of this kinase is required for proper cytokinesis. The reduction of Aurora -A levels at late mitosis is mediated by ubiquitin-proteasomemediated degradation, which is promoted by the anaphase-promoting complex (APC/C) activated by Cdh1 (61,62). This rapid degradation of Aurora-A resulting in Auro ra-A inactivation and dephosphorylation of the specific substrates, might be requir ed for completion of cytokinesis. Aurora-A Signaling Aurora-A regulates a handful of important signal molecules, including cross-talk with p53 and c-Myc pathways, leading to cell cy cle progression, cell pr oliferation and cell survival. p53 Aberrations in centrosome nu mbers have long been implicated in aneuploidy and tumorigenesis. Overexpression of Aurora-A kinase causes centrosome amplification in cultured cells. It does not de regulate centrosome duplicati on but gives rise to extra centrosomes through defects in cell division a nd consequent tetraplo idization (33). Over expression of other mitotic kinases (Pololike kinase 1 and Aurora-B) also causes multinucleation and concomita nt increases in centros ome numbers (63,64). The accumulation of extra copies of centrosomes is strikingly enhanced in p53–/– cells, consistent with the notion that p53 arrests te traploid cells as part of a G1 checkpoint (65,66). A study in yeast and human cell lines (HEK293, H1299) showed that Aurora-A interacts with p53 through Aurora-Box in vivo and in vitro (67). And the interaction

PAGE 26

14 resulted in the suppression of Aurora-A oncogenic activity regarding centrosome amplification and cellular tran sformation by p53 in a transactivation-independent manner. Our lab has demonstrated, on the other hand, that p53 is phosphorylat ed and regulated by Aurora-A (68). Unlike most identified phos phorylation sites of p53 that positively associate with p53, the phosphorylation of p53 by Aurora-A at Ser-215 abrogates p53 DNA binding and transactivation activity. Do wnstream target genes of p53, such as p21Cip/WAF1 and PTEN were inhibited by Aurora-A in a Ser-215 phosphorylationdependent manner. As a result, Aurora-A overr ides the apoptosis and cell cycle arrest induced by cisplatin and -irradiation, respectively. However, the effect of Aurora-A on p53 DNA binding and transactivation activity was not affected by phosphorylation of Ser-315, a previously identified Aurora-A phos phorylation site of p53. We have also reported that Aurora-A protects ovarian cancer cells from apoptosis induced by chemotherapeutic agent and activates Akt pathway in a p53-dependent manner (69). c-Myc Aurora-A plays a pivotal role in transf ormation, however, the molecular mechanism by which Aurora-A induces ovarian and breast cell transformation remains elusive. We found that Aurora-A induces telomerase activity and human telomerase reverse transcriptase (hTERT) by up-regulation of cMyc (70). Ectopic expr ession of Aurora-A activity in human ovarian and breast ep ithelial cell lines HIOSE118 and MCF-10A induces telomerase activity. The mRNA and prom oter activities of hTERT are stimulated by Aurora-A. Furthermore, c-Myc binding si tes of hTERT promoter are required for Aurora-A-induced hTERT promoter activ ity. Ectopic expression of Aurora-A upregulates c-Myc. Knockdown of c-Myc by RNA interference attenuates Aurora-Astimulated hTERT expression and telomerase activity. Regarding the mechanism of Aurora-A upregulation of c-Myc, our data (unpublished) show that Aurora-A inactivates GSK3 through a phosphorylationdependent manner, which leads to -catenin cytoplasmic accumulation, nuclear translocation and subsequently activates cMyc transcription by forming complex with TCF/LEF transcription factors. We belie ve that knockdown of c-Myc may reduce

PAGE 27

15 Aurora-A-induced transformation and genomic instability, but not necessarily abrogate Aurora-A oncogenic activity. It has been well documented that c-Myc plays a critical role in human oncogenesis by regulation of ce ll proliferation and differentiation. Thus, Aurora-A up-regulation of c-Myc may play an essential role in Aurora-A-induced malignant transformation in addition to its G2/M cell cycle control. AR and ER We have recently shown frequent activation/ overexpression of Aurora-A kinase in human primary prostate tumors (unpublished data). Constitutively active Aurora-A exhibits more oncogenic activity. Therefore, activation and/or overexpression of Aurora-A could play an important role in prostate cancer deve lopment. In addition, our preliminary study shows that Aurora-A phosphorylates androgen receptor (AR) in vitro and in vivo and activates AR transactivation activity, which might contribute to androgen-independent tumor growth. According to our data, Aurora-A activation of AR could be phosphorylation dependent. Ectopic expression of Aurora-A in prostate cancer cells may lead to more aggressive phenotype a nd androgen-independent growth; whereas knockdown of Aurora-A could interf ere with androgen dependence. As Aurora-A is frequently altered in breast cancer and induces telomerase activity and tamoxifen resistance in mammary epit helia (unpublished data); we reasoned that Aurora-A could regulate ER activity. Reporter assay showed that ERE-Luc activity is induced by Aurora-A in a dosedependent manner in only ER positive cells. Further, Aurora-A exhibits synergisti c effect with estrogen on ER transactivation activity. In addition, we have demonstrated ER is phosphorylated by Aurora-A in vitro and in vivo (data not shown). We have also observed that tamoxife n downregulated Bcl-2 is overridden by Aurora-A, especially constitutiv ely active Aurora-A. These data indicate important function of Aurora-A besides cell cycle control.

PAGE 28

16 Aurora-A and Tumorigenesis Several mechanisms exist by which Aurora -A overexpression might contribute to tumorigenesis. (see Fig. 1.5) Fig. 1.5. Mechanisms of Aurora-A promoted tumorigenesis. Nature Reviews Cancer 5, 42-50. January 2005. Centrosome Amplification and Aneuploidy Centrosome alterations, including centrosome amplification, are found in brain, breast, lung, colon, prostate, pancreas, bile duct, h ead and neck tumors and lymphoma (71-77). Overexpression of Aurora-A causes centrosome amplification in both cell culture and rat

PAGE 29

17 mammary models (78,79). However, the phenomenon seems not due to centrosome duplication but a result of cytokinesis failure and conse quent multinucleation (33). Since both Aurora-A and Aurora-B are shown to be in volved in cytokinesis, it is possible that Aurora-A causes cytokinesis failure directly, or by affecting Aurora-B or other proteins that regulate cytokinesis. The other possibility is that Au rora-A leads to cytokinesis defect by dysregulation of centriole behavior, which is proposed tightly connected with the completion of cell division. Aneuploidy is thought to be a marker of tumor progression and prognosis. Correlation between Aurora-A overexpression a nd aneuploidy exists in gastric and colon cancer (25,78). Aurora-A-induced polyploid cells would be a rrested at the following G1 phase by a post-mitotic checkpoint depe ndent on the p53/Rb pathway. The G1 checkpoint is required for the cell to tri gger apoptosis. However, when these is no functional p53 present, as in p53-/cells, th e hyperploidy is not detected and cells go on DNA replication and cell division, which mi ght eventually cause aneuploidy and abnormal centrosome number (33,80). Checkpoint Disruption and Chromosome Instability Overexpression of Aurora-A tends to disr upt cell-cycle checkpoint functions. Normal cells have cell-cycle checkpoint s that monitor genomic integr ity and prevent cell dividing in the presence of DNA damage. Recent studi es have shown that DNA-damage signals inhibit Aurora-A kinase activity to induce cell-cycle arrest at the end of the G2 phase and that, conversely, overexpression of Auro ra-A disrupts the DNA-damage-induced G2 checkpoint (30). Additionally, Aurora-A ove rexpression has been found to override the spindle checkpoint, allowing cells to in appropriately enter anaphase despite defective spindle formation. Upon activation of the spindle checkpoint, Bub and Mad proteins become localized to unattached kinetochores. The con centration of these checkpoint proteins on the kinetochore facilita tes the assembly of an active checkpoint complex, which consists of Cdc20, BubR1, Mad2 and Bub3 (81,82). This complex has the ability to inhibit the function of anaphase-promoting complex/ cyclosome (APC/C). Overexpression of

PAGE 30

18 Aurora-A leads to a dramatic accumulation in the Cdc20 complex, where BubR1 level is decreased. At the same time, Mad2 and Bub3 can not be efficiently recruited to Cdc20 complex while the expression of both proteins is not changed (83). The dissociation of these checkpoint proteins from Cdc20 results in the prematurely activation of APC/C and bypass of the checkpoint function. Aurora-A plays a critical role in cell cycle checkpoints to ensure the faithful duplication and transmi ssion of genetic materi als. Perturbation of this regulatory process will endanger geneti c stability and possibly contribute to tumorigenesis. Polymorphism Aurora-A was recently identified as a ca ndidate low-penetrance tumor-susceptibility gene. The involvement of two polymorphisms of Aurora-A (Phe31Ile and Val57Ile) in human tumor susceptibility was analyzed. The Ile31 allele of Phe31Ile was found to be preferentially amplified in human colon a nd ovarian cancers (84,85), and the combination of the two polymorphisms (31Ile and 57Val) wa s related to a significant twofold excess in the risk of breast cancer (86). This ge netic evidence also su pports the role of Aurora-A as a tumor modifier gene. Aurora-A Inhibitors The evidence linking Aurora overexpression and malignancy has stimulated interest in developing Aurora inhibitors for cancer therapy. Aurora-A is considered the principal member of the Aurora family with oncogenic ac tivity. Studies have shown a specific role of Aurora-A in cancer development and progression. However, an in vivo study of Aurora inhibitors demonstrated that it is not Aurora-A but Aurora -B inhibition induces tumor regression. Dual inhibition of Aurora-A and -B results in phe notypes identical to inactivation of Aurora-B alone (87). Inactivation of Au rora-B bypasses the requirement for Aurora-A and leads to polyploidy, indicating that Aurora-B is responsible for mitotic arrest in the absence of Aurora-A (88). Ther efore, it is possible that the better antitumor drug should inhibit all Aurora members instead of only one.

PAGE 31

19 Many Aurora-selective small-molecule inhibitors are currently undergoing preclinical and clini cal assessment. VX-680 is a pyrimid ine derivative, designed against the ATP-binding site, with affinity for Aurora-A, -B, and -C. VX-680 prevents cytokinesis but allows cells to progress through the other stages of mitosis, which leads to polyploidy and, in some cancer cell lines, massive apoptosis. In preclinical models, VX680 blocked tumor growth in three xenograft mode ls, including leukemia, colon and pancreatic (87). AZD1152 is a quinazolin e prodrug, which is converted in plasma into its active metabolite AZD1152-HQPA, which in turn has high affinity for Aurora-B and -C. The effects of AZD1152-HQPA in cancer cells are comparab le with VX-680 (89). In preclinical models, AZD1152 significantly inhi bited the growth of human tumor xenografts. PHA-739358 is a pan-Aurora (A, B, and C) inhibitor with documented antitumor activity in multiple tumor xenograft models, which have shown sustained tumor growth inhibition after discontinuation of treatment. MLN8054 was the first orally available Aurora kinase inhibitor and the first Aurora-A–selective inhibitor in clinical trials. This compound induces mitotic accumulation and spindle defects and inhibits proliferation in multiple human cancer cell lines (90). Tumor growth in nude mice was inhibited after oral admini stration at well-tolerated doses, and the inhibition was sustained after discontinuation of treatment. At least one phase I clinical trial has been completed for each of the inhibitors mentioned above. Although no objective tumor responses were observed, disease stabilization was achieved in some patients. Nonhematologic toxicities were mild. How Aurora kinase targeting interacts with chemothera py and radiation at the molecular level remains to be determined, but these results are very encouraging and pave the way for clinical studies with Aurora kinase in hibitors in the near future.

PAGE 32

20 References 1. Ke, Y. W., Dou, Z., Zhang, J., and Yao, X. B. (2003) Cell Res 13 (2), 69-81 2. Scrittori, L., Hans, F., Angelov, D., Ch arra, M., Prigent, C., and Dimitrov, S. (2001) J Biol Chem 276 (32), 30002-30010 3. Hsu, J. Y., Sun, Z. W., Li, X., Reuben, M., Tatchell, K., Bishop, D. K., Grushcow, J. M., Brame, C. J., Caldwell, J. A., Hunt, D. F., Lin, R., Smith, M. M., and Allis, C. D. (2000) Cell 102 (3), 279-291 4. Nigg, E. A. (2001) Nat Rev Mol Cell Biol 2 (1), 21-32 5. Chan, C. S., and Botstein, D. (1993) Genetics 135 (3), 677-691 6. Petersen, J., Paris, J., Willer, M., Philippe, M., and Hagan, I. M. (2001) J Cell Sci 114 (Pt 24), 4371-4384 7. Tseng, T. C., Chen, S. H., Hsu, Y. P., and Tang, T. K. (1998) DNA Cell Biol 17 (10), 823-833 8. Kitajima, S., Kudo, Y., Ogawa, I., Tatsuka, M., Kawai, H., Pagano, M., and Takata, T. (2007) PLoS ONE 2 (9), e944 9. Hirota, T., Kunitoku, N., Sasayama, T ., Marumoto, T., Zhang, D., Nitta, M., Hatakeyama, K., and Saya, H. (2003) Cell 114 (5), 585-598 10. Marumoto, T., Honda, S., Hara, T., Nitta, M., Hirota, T., Kohmura, E., and Saya, H. (2003) J Biol Chem 278 (51), 51786-51795 11. Bischoff, J. R., and Plowman, G. D. (1999) Trends Cell Biol 9 (11), 454-459 12. Andrews, P. D., Knatko, E., Moore, W. J., and Swedlow, J. R. (2003) Curr Opin Cell Biol 15 (6), 672-683 13. Giet, R., and Glover, D. M. (2001) J Cell Biol 152 (4), 669-682 14. Adams, R. R., Maiato, H., Earnsh aw, W. C., and Carmena, M. (2001) J Cell Biol 153 (4), 865-880 15. Murata-Hori, M., and Wang, Y. L. (2002) Curr Biol 12 (11), 894-899 16. Kallio, M. J., McCleland, M. L., Stukenberg, P. T., and Gorbsky, G. J. (2002) Curr Biol 12 (11), 900-905

PAGE 33

21 17. Sasai, K., Katayama, H., Stenoien, D. L., Fujii, S., Honda, R., Kimura, M., Okano, Y., Tatsuka, M., Suzuki, F., Nigg, E. A., Earnshaw, W. C., Brinkley, W. R., and Sen, S. (2004) Cell Motil Cytoskeleton 59 (4), 249-263 18. Miyoshi, Y., Iwao, K., Egawa, C., and Noguchi, S. (2001) Int J Cancer 92 (3), 370-373 19. Bischoff, J. R., Anderson, L., Zhu, Y., Mo ssie, K., Ng, L., Souza, B., Schryver, B., Flanagan, P., Clairvoyant, F., Gint her, C., Chan, C. S., Novotny, M., Slamon, D. J., and Plowman, G. D. (1998) EMBO J 17 (11), 3052-3065 20. Sen, S., Zhou, H., Zhang, R. D., Yoon, D. S., Vakar-Lopez, F., Ito, S., Jiang, F., Johnston, D., Grossman, H. B., Ruifrok, A. C., Katz, R. L., Brinkley, W., and Czerniak, B. (2002) J Natl Cancer Inst 94 (17), 1320-1329 21. Gritsko, T. M., Coppola, D., Paciga, J. E., Yang, L., Sun, M., Shelley, S. A., Fiorica, J. V., Nicosia, S. V., and Cheng, J. Q. (2003) Clin Cancer Res 9 (4), 14201426 22. Li, D., Zhu, J., Firozi, P. F., Abbruzzese, J. L., Evans, D. B., Cleary, K., Friess, H., and Sen, S. (2003) Clin Cancer Res 9 (3), 991-997 23. Jeng, Y. M., Peng, S. Y., Lin, C. Y., and Hsu, H. C. (2004) Clin Cancer Res 10 (6), 2065-2071 24. Zhou, H., Kuang, J., Zhong, L., Kuo, W. L ., Gray, J. W., Sahin, A., Brinkley, B. R., and Sen, S. (1998) Nat Genet 20 (2), 189-193 25. Sakakura, C., Hagiwara, A., Yasuoka, R., Fujita, Y., Nakanishi, M., Masuda, K., Shimomura, K., Nakamura, Y., Inazawa, J., Abe, T., and Yamagishi, H. (2001) Br J Cancer 84 (6), 824-831 26. Katayama, H., Ota, T., Jisaki, F., Ueda, Y., Tanaka, T., Odashima, S., Suzuki, F., Terada, Y., and Tatsuka, M. (1999) J Natl Cancer Inst 91 (13), 1160-1162 27. Tatsuka, M., Katayama, H., Ota, T., Ta naka, T., Odashima, S., Suzuki, F., and Terada, Y. (1998) Cancer Res 58 (21), 4811-4816 28. Ota, T., Suto, S., Katayama, H., Han, Z. B., Suzuki, F., Maeda, M., Tanino, M., Terada, Y., and Tatsuka, M. (2002) Cancer Res 62 (18), 5168-5177

PAGE 34

22 29. Kimura, M., Matsuda, Y., Yoshioka, T., and Okano, Y. (1999) J Biol Chem 274 (11), 7334-7340 30. Marumoto, T., Hirota, T., Morisaki, T ., Kunitoku, N., Zhang, D., Ichikawa, Y., Sasayama, T., Kuninaka, S., Mimori, T., Tamaki, N., Kimura, M., Okano, Y., and Saya, H. (2002) Genes Cells 7 (11), 1173-1182 31. Anand, S., Penrhyn-Lowe, S., and Venkitaraman, A. R. (2003) Cancer Cell 3 (1), 51-62 32. Giet, R., McLean, D., Descamps, S., Lee, M. J., Raff, J. W., Prigent, C., and Glover, D. M. (2002) J Cell Biol 156 (3), 437-451 33. Meraldi, P., Honda, R., and Nigg, E. A. (2002) EMBO J 21 (4), 483-492 34. Hannak, E., Kirkham, M., Hyman, A. A., and Oegema, K. (2001) J Cell Biol 155 (7), 1109-1116 35. Berdnik, D., and Knoblich, J. A. (2002) Curr Biol 12 (8), 640-647 36. Bellanger, J. M., and Gonczy, P. (2003) Curr Biol 13 (17), 1488-1498 37. Conte, N., Delaval, B., Ginestier, C., Ferrand, A., Isnardon, D., Larroque, C., Prigent, C., Seraphin, B., Jacquemi er, J., and Birnbaum, D. (2003) Oncogene 22 (50), 8102-8116 38. Lee, M. J., Gergely, F., Jeffers, K., P eak-Chew, S. Y., and Raff, J. W. (2001) Nat Cell Biol 3 (7), 643-649 39. Giet, R., Uzbekov, R., Cubizolles, F., Le Guellec, K., and Prigent, C. (1999) J Biol Chem 274 (21), 15005-15013 40. Dutertre, S., and Prigent, C. (2003) Mol Interv 3 (3), 127-130 41. Roghi, C., Giet, R., Uzbekov, R., Morin, N., Chartrain, I., Le Guellec, R., Couturier, A., Doree, M., Philippe, M., and Prigent, C. (1998) J Cell Sci 111 ( Pt 5) 557-572 42. Tsai, M. Y., Wiese, C., Cao, K., Marti n, O., Donovan, P., Ruderman, J., Prigent, C., and Zheng, Y. (2003) Nat Cell Biol 5 (3), 242-248 43. Kalab, P., Pu, R. T., and Dasso, M. (1999) Curr Biol 9 (9), 481-484 44. Carazo-Salas, R. E., Guarguaglini, G., Gr uss, O. J., Segref, A., Karsenti, E., and Mattaj, I. W. (1999) Nature 400 (6740), 178-181

PAGE 35

23 45. Wittmann, T., Boleti, H., Antony, C., Karsenti, E., and Vernos, I. (1998) J Cell Biol 143 (3), 673-685 46. Karsenti, E., and Vernos, I. (2001) Science 294 (5542), 543-547 47. Wittmann, T., Wilm, M., Karsenti, E., and Vernos, I. (2000) J Cell Biol 149 (7), 1405-1418 48. Garrett, S., Auer, K., Compton, D. A., and Kapoor, T. M. (2002) Curr Biol 12 (23), 2055-2059 49. Kufer, T. A., Sillje, H. H., Korner, R., Gruss, O. J., Meraldi, P., and Nigg, E. A. (2002) J Cell Biol 158 (4), 617-623 50. Eyers, P. A., Erikson, E., Chen, L. G., and Maller, J. L. (2003) Curr Biol 13 (8), 691-697 51. Giet, R., and Prigent, C. (2000) Exp Cell Res 258 (1), 145-151 52. Ditchfield, C., Johnson, V. L., Tighe, A ., Ellston, R., Haworth, C., Johnson, T., Mortlock, A., Keen, N., and Taylor, S. S. (2003) J Cell Biol 161 (2), 267-280 53. Hauf, S., Cole, R. W., LaTerra, S., Zi mmer, C., Schnapp, G., Walter, R., Heckel, A., van Meel, J., Rieder, C. L., and Peters, J. M. (2003) J Cell Biol 161 (2), 281294 54. Kunitoku, N., Sasayama, T., Marumoto, T ., Zhang, D., Honda, S., Kobayashi, O., Hatakeyama, K., Ushio, Y., Saya H., and Hirota, T. (2003) Dev Cell 5 (6), 853864 55. Cheung, P., Allis, C. D., and Sassone-Corsi, P. (2000) Cell 103 (2), 263-271 56. Zeitlin, S. G., Shelby, R. D., and Sullivan, K. F. (2001) J Cell Biol 155 (7), 11471157 57. Severson, A. F., Hamill, D. R., Carter, J. C., Schumacher, J., and Bowerman, B. (2000) Curr Biol 10 (19), 1162-1171 58. Murata-Hori, M., Fumoto, K., Fukuta, Y ., Iwasaki, T., Kikuchi, A., Tatsuka, M., and Hosoya, H. (2000) Journal of biochemistry 128 (6), 903-907 59. Goto, H., Yasui, Y., Kawajiri, A., Nigg, E. A., Terada, Y., Tatsuka, M., Nagata, K., and Inagaki, M. (2003) J Biol Chem 278 (10), 8526-8530

PAGE 36

24 60. Kawajiri, A., Yasui, Y., Goto, H., Tatsuka, M., Takahashi, M., Nagata, K., and Inagaki, M. (2003) Mol Biol Cell 14 (4), 1489-1500 61. Honda, K., Mihara, H., Kato, Y., Ya maguchi, A., Tanaka, H., Yasuda, H., Furukawa, K., and Urano, T. (2000) Oncogene 19 (24), 2812-2819 62. Taguchi, S., Honda, K., Sugiura, K., Ya maguchi, A., Furukawa, K., and Urano, T. (2002) FEBS Lett 519 (1-3), 59-65 63. Mundt, K. E., Golsteyn, R. M., La ne, H. A., and Nigg, E. A. (1997) Biochem Biophys Res Commun 239 (2), 377-385 64. Terada, Y., Tatsuka, M., Suzuki, F., Yasuda, Y., Fujita, S., and Otsu, M. (1998) Embo J 17 (3), 667-676 65. Khan, S. H., and Wahl, G. M. (1998) Cancer Res 58 (3), 396-401 66. Andreassen, P. R., Lohez, O. D., Lacroix, F. B., and Margolis, R. L. (2001) Mol Biol Cell 12 (5), 1315-1328 67. Chen, S. S., Chang, P. C., Cheng, Y. W., Tang, F. M., and Lin, Y. S. (2002) Embo J 21 (17), 4491-4499 68. Liu, Q., Kaneko, S., Yang, L., Feldman, R. I., Nicosia, S. V., Chen, J., and Cheng, J. Q. (2004) J Biol Chem 279 (50), 52175-52182 69. Yang, H., He, L., Kruk, P., Nicosia, S. V., and Cheng, J. Q. (2006) Int J Cancer 119 (10), 2304-2312 70. Yang, H., Ou, C. C., Feldman, R. I., Nicosi a, S. V., Kruk, P. A., and Cheng, J. Q. (2004) Cancer Res 64 (2), 463-467 71. Lingle, W. L., Lutz, W. H., Ingle, J. N ., Maihle, N. J., and Salisbury, J. L. (1998) Proc Natl Acad Sci U S A 95 (6), 2950-2955 72. Weber, R. G., Bridger, J. M., Benne r, A., Weisenberger, D., Ehemann, V., Reifenberger, G., and Lichter, P. (1998) Cytogenetics and cell genetics 83 (3-4), 266-269 73. Pihan, G. A., Purohit, A., Wallace, J., Ma lhotra, R., Liotta, L., and Doxsey, S. J. (2001) Cancer Res 61 (5), 2212-2219 74. Sato, N., Mizumoto, K., Nakamura, M., Nakamura, K., Kusumoto, M., Niiyama, H., Ogawa, T., and Tanaka, M. (1999) Clin Cancer Res 5 (5), 963-970

PAGE 37

25 75. Kuo, K. K., Sato, N., Mizumoto, K., M aehara, N., Yonemasu, H., Ker, C. G., Sheen, P. C., and Tanaka, M. (2000) Hepatology (Baltimore, Md 31 (1), 59-64 76. Gustafson, L. M., Gleich, L. L., Fukasawa, K., Chadwell, J., Miller, M. A., Stambrook, P. J., and Gluckman, J. L. (2000) The Laryngoscope 110 (11), 17981801 77. Neben, K., Ott, G., Schweizer, S., Kalla, J., Tews, B., Katzenberger, T., Hahn, M., Rosenwald, A., Ho, A. D., Muller-Hermelink, H. K., Lichter, P., and Kramer, A. (2007) Int J Cancer 120 (8), 1669-1677 78. Lentini, L., Amato, A., Schillaci, T., and Di Leonardo, A. (2007) BMC Cancer 7 212 79. Goepfert, T. M., Adigun, Y. E., Zhong, L ., Gay, J., Medina, D., and Brinkley, W. R. (2002) Cancer Res 62 (14), 4115-4122 80. Fujiwara, T., Bandi, M., Nitta, M., Ivanova, E. V., Bronson, R. T., and Pellman, D. (2005) Nature 437 (7061), 1043-1047 81. Sudakin, V., Chan, G. K., and Yen, T. J. (2001) J Cell Biol 154 (5), 925-936 82. Tang, Z., Bharadwaj, R., Li, B., and Yu, H. (2001) Dev Cell 1 (2), 227-237 83. Jiang, Y., Zhang, Y., Lees, E., and Seghezzi, W. (2003) Oncogene 22 (51), 82938301 84. Ewart-Toland, A., Briassouli, P., de Koning, J. P., Mao, J. H., Yuan, J., Chan, F., MacCarthy-Morrogh, L., Ponder, B. A., Nagase, H., Burn, J., Ball, S., Almeida, M., Linardopoulos, S., and Balmain, A. (2003) Nat Genet 34 (4), 403-412 85. Dicioccio, R. A., Song, H., Waterfall, C ., Kimura, M. T., Nagase, H., McGuire, V., Hogdall, E., Shah, M. N., Luben, R. N., Easton, D. F., Jacobs, I. J., Ponder, B. A., Whittemore, A. S., Gayther, S. A., Pharoah, P. D., and Kruger-Kjaer, S. (2004) Cancer Epidemiol Biomarkers Prev 13 (10), 1589-1594 86. Egan, K. M., Newcomb, P. A., Ambros one, C. B., Trentham-Dietz, A., TitusErnstoff, L., Hampton, J. M., Kimura, M. T., and Nagase, H. (2004) Carcinogenesis 25 (11), 2149-2153

PAGE 38

26 87. Harrington, E. A., Bebbington, D., Moore, J., Rasmussen, R. K., Ajose-Adeogun, A. O., Nakayama, T., Graham, J. A., De mur, C., Hercend, T., Diu-Hercend, A., Su, M., Golec, J. M., and Miller, K. M. (2004) Nat Med 10 (3), 262-267 88. Yang, H., Burke, T., Dempsey, J., Diaz, B., Collins, E., Toth, J., Beckmann, R., and Ye, X. (2005) FEBS Lett 579 (16), 3385-3391 89. Wilkinson, R. W., Odedra, R., Heaton, S. P., Wedge, S. R., Keen, N. J., Crafter, C., Foster, J. R., Brady, M. C., Bigley, A ., Brown, E., Byth, K. F., Barrass, N. C., Mundt, K. E., Foote, K. M., Heron, N. M., Jung, F. H., Mortlock, A. A., Boyle, F. T., and Green, S. (2007) Clin Cancer Res 13 (12), 3682-3688 90. Manfredi, M. G., Ecsedy, J. A., Meetze, K. A., Balani, S. K., Burenkova, O., Chen, W., Galvin, K. M., Hoar, K. M., Huck, J. J., LeRoy, P. J., Ray, E. T., Sells, T. B., Stringer, B., Stroud, S. G., Vos, T. J., Weatherhead, G. S., Wysong, D. R., Zhang, M., Bolen, J. B., and Claiborne, C. F. (2007) Proc Natl Acad Sci U S A 104 (10), 4106-4111

PAGE 39

27 CENTRAL HYPOTHESIS AND OBJECTIVES Central Hypothesis Upregulation/activation of Au rora-A by oncogenic factors and inactivation of tumor suppressor by Aurora-A contribute to Aurora-A-induced carcinogenesis. Objectives To understand the molecular mechanis m of Aurora-A in human cancers: 1). Aurora-A is transcriptionally regulated by E2F3; 2). Identification of tumor suppressor CHFR as an Aurora-A substrate; 3). Characterization of a nove l oncogene CNTD2 as an Auro ra-A interacting protein.

PAGE 40

28 CHAPTER II IDENTIFICATION OF AURORA-A AS A DIRECT TARGET OF E2F3 DURING G2/M CELL CYCLE PROGRESSION Abstract Aurora-A is a centrosome kinase and plays a pivotal role in G2/M cell cycle progression. Expression of Aurora-A is cell cycle-depende nt. Levels of Aurora-A mRNA and protein are low in G1/S, accumulate during G2/M, and decrease rapidly after mitosis. Previous studies have shown regulation of Aurora-A protein level during th e cell cycle through ubiquitin-proteasome pathway. However, transc riptional regulation of Aurora-A remains largely unknown. Here, we demonstrate th at E2F3 modulates Aurora-A mRNA expression during the cell cycle. Ectopic expression of E2F3 induces Aurora-A expression. Stable knockdown of E2F3 decreases mRNA and protein levels of Aurora-A and delays G2/M entry. Further, E2F3 di rectly binds to Aurora-A promoter and stimulates the promoter activity. Deletion a nd mutation analyses of Aurora-A promoter revealed that a region located 96-bp upstream of transcription initiation site is critical for the activation of the promoter by E2F3. In addition, expres sion of E2F3 considerably correlates with the protein le vel of Aurora-A in human ovarian cancer examined. These results indicate for the first time that Auro ra-A is transcriptionally regulated by E2F3 during the cell cycle and that E2F3 is a causa l factor for upregulati on of Aurora-A in a subset of human ovarian cancer. Thus, E2F3-Aur ora-A axis could be an important target for cancer intervention. Key words : cell cycle, G2/M, transcription, promoter activity, ovarian cancer

PAGE 41

29 Introduction Aurora family of serine/thre onine protein kinases is evolutionally conserved from human, Drosophila to yeast (1,2). In mammals, the Au rora kinase family comprises three members: Aurora-A, Aurora-B and Aurora-C. They share similar structures, with their highly conserved catalytic domains flanked by ve ry short C-terminal tails and N-terminal domains of variable lengths. The N-terminal domains of the three Auroras share low sequence conservation, which may determin e selectivity during protein–protein interactions. As a mitotic kinase, activated Aurora-A is required for mitotic entry, centrosome maturation and chromosome segregation (1,2). Aurora-A protein localizes to centrosomes during interphase and to both spindle pol es and spindle microtubules during early mitosis. Ectopic expression of Aurora-A l eads to an increase in centrosome number, causes catastrophic loss or gain of chromo somes, and then results in malignant transformation. Aurora-A kinase acti vity is regulated by phosphorylation, dephosphorylation and by association with a numb er of proteins such as HEF1, TPX2, or Bora (3-5). Aurora-A is located at chromosome 20q13.2, which is commonly amplified in various epithelial malignant tumors, includ ing breast, colon, bladder, ovarian and pancreatic cancers, and the levels of Auro ra-A mRNA and protein are also increased in those tumors (6-10). However, alterations of Aurora-A at mRNA and/or protein levels are much more common than the gene amplifica tion. Discrepancy between amplification and overexpression rates was also reported in br east cancer, gastric cancer and ovarian cancer (9,11,12). Therefore, Aurora-A overexpression is likely to be regulated not only by gene amplification, but also by other mechanisms such as transcriptional activation and/or suppression of protein degradation. Expressi on of Aurora-A is ce ll cycle regulated. Aurora-A mRNA and protein levels begin to accumulate at G2 to M phase. After metaphase, Aurora-A protein is rapidly degraded by the ubiquitin-mediated proteolysis, which is promoted by the hCdh1-activated anaphase-promoting complex/cyclosome

PAGE 42

30 (APC/C). However, transcripti onal regulation of Aurora-A du ring the cell cycle is largely unknown. It has been well documented that E2F fa mily of transcrip tion factors play important roles in the cont rol of expression of genes at DNA duplication as well as further cell cycle progression (13,14). In addi tion, previous studies have demonstrated that a substantial number of E2F-induced gene s are normally regulated at G2 of the cell cycle, encoding proteins know n to function in mitosis (1 5-18). Moreover, studies in Drosophila have provided evidence for a conne ction between E2F activity and the control of mitotic activities (19). A recent report shows that both E2F1 and E2F3 are required for cells to enter the S phase from a quiescent state, whereas only E2F3 is necessary for the S phase in growing cells ( 20). The acute loss of E2F3 activity affects the expression of genes encoding DNA replicati on and mitotic activities, whereas loss of E2F1 affects a limited number of genes that are distinct from those regulated by E2F3 (21). Here we report that E2F3 upregulates Aurora-A at transcriptional level during G2 and the onset of mitosis. Among several E2F3 response elements, the region between -96 to transcriptional initi ation site is responsible for E2F3 induction of Aurora-A promoter activity. Knockdown of E2F3 reduc es Aurora-A expression and delays mitotic cell cycle progression.

PAGE 43

31 Materials and Methods Cell Line, Culture and Transfection HeLa, HEK293, NIH3T3 and ovarian cancer ce ll lines were obtained as previously described (22,23) and cultured in Dulbecco ’s modified Eagle’s medium (Sigma) containing 10% fetal bovine serum (FBS). Cell transfection wa s performed with Lipofectamine Plus Reagent (Invitrogen). Synchronization and Cell Cycle Analyses Cell synchronization at G1/S phase was perf ormed using double-thymidine blocking (2). Briefly, HeLa cells were grow n in 60-mm plates and thymidine (Sigma) was added to the culture medium at a final concentration of 2 mM for 12 h. Following two times washing with serum free medium, cells were released in fresh culture medium containing 24 mM 2-deoxycytidine. After 9 h incubation, the second thymidine block was initiated, and completed after 14 h. Cells were released fr om the block by washing in warm PBS and replacing with complete culture medium. At different time points, cells were fixed in 70% ethanol. The fixed cells were rinsed w ith PBS and then stained with the propidium iodide (PI) in a solution c ontaining Triton X-100 and RNase A (Sigma). Cell synchrony at various cell cycle stages was monitored by flow cytometry. To synchronize cells to Mphase, HeLa cells were incubated in DM EM medium containi ng 50 nM/ml nocodazole for 14 h. Whole cell lysates and total RNA were isolated from parallel experiments for Western and Northern blotting analyses. Plasmids pGL3-Aurora-A/-1486/+354, pGL3-Aurora-A/-4 15, pGL3-Aurora-A/-189 reporters were kindly provided by Dr. Y. Ishigatsubo (Yokohama City University Sc hool of Medicine). To create deletion mutants, DNA fragments corresponding to -124/+354, -96/+354 and 49/+354 were amplified by PCR using pGL3-Aur ora-A/-1486 as template, respectively. The PCR products were cloned in to pGL3-basic vector at Mlu IBgl II sites and confirmed by DNA sequencing. The oligonucleotid e primers used were as follows: sense

PAGE 44

32 -124 5’-CGACGCGTTGGGACTGCCACAGGTCTGG-3'), -96 (5’CGACGCGTTGGCTCCACCACTTCCGG-3’), -49 (5’CGACGCGTGTGTGCGCCCTTAAACGCGAC-3’); and antisense +354 (5’GAAGATCTCTCTAGCTGTAATAAGTAAC-3’). Aurora-A promoter mutant constructs we re generated using QuikChange (Multi) Site-Directed Mutagenesis K it (Stratagene), including pGL3 -Aurora-A/-1486mtA, pGL3Aurora-A/-1486mtB and pGL3-Aurora-A/1486mtAB pGL3-Aurora-A/-96mtC, pGL3Aurora-A/-96mtD and pGL3-Aurora-A/-96mtCD. Primers used for mutagenesis reactions were: mutation A, sense 5’GCGCACGCTGAAAGGGATCCAAGCCGACCGCTGCGCTATCG-3’ and antisense 5’-CGATAGCGCAGCGGTCGGCTTGGATCCCTTTCAGCGTGCGC-3’; mutation B, sense 5’-CACTCTCTCTTGCTTTTCTATCCA TCTTACTTACTGGC-3’ and antisense 5’-GCCAGTAAGTAAGATG GATAGAAAAGCAAGAGAGAGT G-3’; mutation C, sense 5’-GGCTCCACCACTTCATGGTTCTTAGGGAGC-3’ and antisense 5’GCTCCCTAAGAACCATGAAGTGGTGGAGCC3’; mutation D, sense 5’GAGCAAGTCGCCTGCATGCGGTGTGCGC CCTT-3’ and antisense 5’AAGGGCGCACACCGCATGCAGGCGACTTGCTC-3’. Luciferase Assay Cells were cultured in 24-well plates and transfected with pGL3-Aurora-A reporters, galactosidase and the plasmids indicated in the figure legend. The amount of DNA in each transfection was kept constant by th e addition of empty vector. After 36 h of transfection, luciferase activity was measur ed using luciferase kit (Promega). The galactosidase activity was measured by us ing Galato-Light (Tropix). Transfection efficiency was normalized by -galactosidase expression. Luciferase activity was expressed as relative luciferase activity. Th e experiments were repeated three times. Northern and Western Blotting Analyses Total RNA was isolated from cells using Tr izol Reagent (Invitrogen). Northern blot analysis was performed as previously descri bed (24). Briefly, 20 g total RNA from each

PAGE 45

33 sample was separated on 1.0% denature agar ose gel. After transf erring to GeneScreen Plus Membrane (PerkinElmer) and prehybrid ization, the membrane was hybridized with [32P]dCTP-labeled Aurora-A cDNA probe in Ex press-Hybridization Solution (Clontech). For Western blot, equal amount of protei n was separated by SDS-PAGE gel, and electroblotted onto membranes. Following bl ocking in TBS-T containing 5% milk, the membranes were hybridized with appropriate antibodies indicated in figure legend. shRNA Lentiviral Infection Knockdown of E2F3 was carried out by infection of HeLa cells with lentiviruses (pLKO.1-shRNA, Sigma) expressing five di fferent shRNAs of human E2F3, four of which harmonize coding sequence of E2F3 and the other complements the 3’untranslation region. Briefly, He La cells were plated in 60 -mm plates. Following culture for 24 h and addition of polybrene, cells were infected pLKO.1-shRNA/E2F3 viruses at multiplicity of infection (MO I) of 100. After swirling the plat e, the cell-viral particle mixture was incubated at 37 C overnight and then repla ced with complete culture medium. For transient experime nts, the cells were harveste d after 48 h of infection and assayed by Western Blot. Stab le knockdown of E2F3 cell lines were established by selection with puromycin (5 g/ml). Chromatin Immunoprecipitation (ChIP) Assay ChIP assay was performed essentially as previously described (25). Solubilized chromatin was prepared from a total of 2 x 107 asynchronously growing HEK293 cells that were transfected with E2F3. The chro matin solution containing cross-linked binding proteins was diluted 10-fold with ChIP dilution buffer (1.1% Triton X-100, 1.2 mM EDTA, 167 mM NaCl, 16.7 mM Tris-HCl, pH 8.1, 0.01% SDS, protease inhibitors), and precleared with protein-A/G-agarose beads bl ocked with 2 g of sheared salmon sperm DNA and preimmune serum. The precleared chro matin solution was divided and utilized in immunoprecipitation assays with either an anti-E2F 3 antibody or an anti-actin antibody. Following wash, the antibody-proteinDNA complex was eluted from the beads by incubating the pellets in 1% SDS, 0.1 M NaHCO3 at room temperature for 20 min.

PAGE 46

34 The protein-DNA cross-linking was reverse d, and protein and R NA were removed by incubation with 10 g of Proteinase K a nd 10 g of RNase A at 67 C for 6 h to overnight. Purified DNA was subjected to PCR with primers specific for putative E2F3binding sites within the Aurora -A promoter. The sequences of the PCR primers used are as follows: site A, sense (5’-GAATCC TGCCCAATCTACCGCTCC-3’) and antisense (5’-GAAAAGCAAGAGAGAGTGGGACCG-3’); site B, sense (5’GCTATCGATCGGTCCCACTCTCTC-3’) and antisense (5’CTTGAGTCGCGTTTAAGGGCGCAC-3’); site C+D, sense (5’CGACGCGTTGGCTCCACCACTTCCGG-3’) and antisense (5’CCAGGAGCTCAGCCGTTAGAATTCAAAGG-3’).

PAGE 47

35 Results Expression of the Members of E2F Fam ily and Aurora-A during the Cell Cycle While mRNA and protein levels of Aurora-A are elevated at G2/M phase, the underlying molecular mechanism remains elusive. E2F fa mily members have been shown to play a critical role in cell cycle progression thr ough transcriptional regul ation of a number of cell cycle-associated genes. To examine if Aurora-A is regulated by E2F family, we compared the expression patterns of E2F fam ily and Aurora-A proteins during the cell cycle (Fig. 2.1A). HeLa cells were sync hronized with double-thymidine blocking and released for various time points. The ce ll cycle progression was monitored by flow cytometry. Before releasing, over 60% of the total cell population was at S phase. About 6 hours, cells entered G2/M. After 16 hours, th e majority of the cells were in G1 phase (Figs. 2.1B and 2.1C). E2F1 and E2F2 protein levels were high in G1/S and decreased at G2/M phase (Fig. 2.1A). E2F4 and E2F5 prot eins distributed to the whole cell cycle. Notably, E2F3 protein was elevated in S and maintained through G2/M phases and decreased in G1 phase. Recent studies have shown that E2F3 but not other E2F family members regulates the expression of genes that are involved in G2/M (21), implying that Aurora-A could be transcri ptionally regulated by E2F3. Ectopic Expression of E2F3 Induces and K nockdown of E2F3 Decreases Aurora-A Protein and mRNA Levels To determine if E2F3 transcriptionally regulates Aurora-A, we transfected E2F3 into three cell lines, which include two human epithelial cell lines Hela and A2780S, and a mouse fibroblast NIH3T3 (Fig. 2.2A). Imm unoblotting analysis showed that E2F3 upregulated Aurora-A protein expression in a dose-dependent manner (Fig. 2.2A). Moreover, mRNA levels of Aurora-A was al so induced by ectopic expression of E2F3 (Fig. 2.2B). To further demonstrate Aurora -A expression is controlled by E2F3, we knocked down E2F3 by infection of HeLa cells with the lentiviruses (pLKO.1E2F3/shRNA) expressing 5 different E2F3 shRNAs, 4 of which are corresponding to the coding region and the other matches the 3 ’-untranslated region. Fig. 2.2C showed that

PAGE 48

36 E2F3 was considerably reduced in the cells infected with three individual shRNA as well as their combination. After selection with puromycin, st able E2F3-knockdown pool cells were obtained. Expression of Aurora-A was si gnificantly reduced in these cells both at interphase and mitosis (Fig. 2.2D). Fig. 2.1. The correlation of expression of E2 F family proteins and Aurora-A during the cell cycle. HeLa cells were synchronized with double-thymidine block (DTB) and released for indicated time points. A portion of cells for each time point were lysed and immunoblotted with indicated antibodies ( A ). The rest were subjected to flow cytometry ( B and C ).

PAGE 49

37 Fig. 2.2. E2F3 regulates Aurora A exp ression at protei n and mRNA levels. ( A and B ) Ectopic expression of E2F3 induces Auro ra-A expression. HeLa, NIH3T3 and A2780S cells were transfected with increasing am ount of E2F3. Following 36 h incubation, cells were subjected to immunoblotting analysis with indicated antibodi es (A) and RT-PCR assay for expression of Aurora-A (B). ( C and D ) Stable knockdown of E2F3 reduces Aurora-A expression. HeLa cells were infect ed with individual lentivirus expressing 5 shRNAs targeting different re gions of E2F3 and all 5 shRNAs together. After selection with puromycin, stable knockdow n clonal cells were immunobl otted with anti-E2F3 (top) and -actin (bottom of panel C) antibo dies. E2F3-stably knockdown cells were synchronized with and without nocodazole a nd blotted with indica ted antibodies (D). Relative Aurora-A mRNA or protein levels were normalized to actin and quantified by ImageJ software (NIH).

PAGE 50

38 Knockdown of E2F3 Delays G2/M Entry and Reduces Aurora-A Expression during the Cell Cycle Having demonstrated E2F3 transcriptional regul ation of Aurora-A, we then examined the effects of knockdown of E2F3 on G2/M progr ession and Aurora-A expression during the cell cycle. pLKO.1-shRNA vector infected and E2F3 stable knockdown HeLa cells were synchronized by double-thymidine block. Flow cy tometry analysis revealed that cells with knockdown of E2F3 delays to enter G2/M phase as compared to the pLKO.1 vectorinfected cells (Fig. 2.3A). In consistence w ith cell cycle change, elevation of Aurora-A mRNA and protein levels was approximat ely 2 or 3 h delayed by knockdown of E2F3. Moreover, the mRNA and protein levels of Au rora-A at G2/M phase were also reduced in E2F3-knockdown cells as compared to c ontrol shRNA-treated cells (Figs. 2.3B and 2.3C). However, Aurora-A protein underwent degradation at late M phase in both cell lines (Fig. 2.3C). These data further suggest that Aurora-A is transcriptionally regulated by E2F3 during the cell cycle. E2F3 Binds to and Transactivates Aurora-A Promoter during G2/M Phase We next examined whether the Aurora-A promoter is transactivated by E2F3. HeLa cells were transfected with pGL3-Aurora-A/1486/+354-Luc reporter and increasing amounts of E2F3. Luciferase reporter assay revealed that Auro ra-A promoter act ivity was induced by E2F3 in a dose-dependent manner (Fig. 2.4A). Moreover, basal level of Aurora-A promoter activity, especially at mitosis, was reduced in E2F3-knockdown cells as compared to pLKO.1 vector-inf ected HeLa cells (Fig. 2.4B). Sequence analysis showed 4 put ative E2F-binding elements [TT(C/G)GCGC(C/G)] w ithin the promoter (Fig. 2.4C). To define the response region(s) of the promoter to E2F3, we created a series of deletion mu tants of Aurora-A promoter. Reporter assay showed that a mutant with deletion from -1486 to -415 significantly deceased E2F3-induced promoter activity while it still contains all 4 putative E2Fbinding elements, implying that this region is of transactivation function. Moreover, the deletion of 2 distal E2F-bindi ng sites (i.e., from -1486 to 189; A and B) further reduced the promoter activity. However, promoter activ ity of the mutants with additional deletion

PAGE 51

39 of either from -189 to -124 or from -189 to -96, both of which retain the 2 E2F-response elements (C and D) proximal to the transcri ptional starting site, was significantly induced by E2F3. The promoter activity was complete ly abrogated by further deletion of the proximal E2F-response elements (Fig. 2.4C). Th ese results suggest th at the region from 189 to -96 contains a repressi on element(s) and that the a ll four E2F-binding sites are response to E2F3. However, the two binding elements proximal to the transcriptional start site are sufficient for E2F3 trans activation of the Aurora-A promoter. Fig. 2.3. Knockdown of E2F3 results in mitotic cell entry delay and decreased Aurora-A expression during the cell cycle. Control shRNA and stable E2F3-shRNA infected HeLa cells were synchronized with double-thymidine block and released for indicated time points. Cells were analyzed by flow cytometry ( A ), Northern ( B ) and Western ( C ) blotting with indicate d probe and antibodies.

PAGE 52

40 Fig. 2.4. Aurora-A promoter is regulated by E2F3. ( A ) Ectopic expression of E2F3 induces Aurora-A promoter activity. NIH3T3 cells were transiently transfected with indicated plasmids. Following 48 h incubati on, luciferase activity was measured and normalized to -gal. Results are the mean SEM of three independent experiments performed in triplicate (top). Bottom panel shows expression of transfected E2F3. ( B ) Knockdown of E2F3 reduces ba sal level of Aurora-A prom oter activity. Aurora-A-Luc and -gal were introduced into pLKO.1 contro l shRNAand E2F3-shRNA-infected HeLa cells. After 36h incubation, cells were treate d with and without nocodazole for 12 h and then subjected to luciferase assay. ( C ) Deletion mapping of E2F3 response elements in Aurora-A promoter. NIH3T3 cells were tran sfected with indicated Aurora-A deletion mutants and different amounts of E2F3 plasmid. E2F3 binding sites were labeled as “A – D”. Luciferase assay was perfor med after 36 h of transfection.

PAGE 53

41 To determine whether E2F3 could direct ly bind to the E2F-binding sites of the Aurora-A promoter in vivo and to further define the E2F3 response elements in the promoter, we carried out ChIP assay, whic h detects specific genomic DNA sequences that are associated with a particular transcri ption factors in intact cells. HeLa cells were transfected with wild type E2F3. Following 36 h inc ubation, unsynchronized cells were immunoprecipitated with an anti-E2F 3 antibody. The E2F3 bound chromatin was subjected to PCR using 3 pair s of oligonucleotide primers th at amplify regions spanning each distal and 2 proximal E2F3-binding sites within the Aurora-A promoter because 2 proximal E2F-binding sites are so close that could not be separated by ChIP assay. As shown in Figs. 2.5A and 2.5B, the anti-E2F3 antibody pulled down a ll E2F-binding sites. In contrast, immunoprecipitation with an irre levant antibody (e.g. anti-Actin) resulted in the absence of bands in these sites. These resu lts indicate that E2F3 directly binds to the Aurora-A promoter. By mutation of individu al or combinational E2F3 binding sites (CG AT) in Aurora-A promoter, we further demons trated that the 2 E2F-response elements proximal to transcriptional start site are requir ed for E2F3 transactivation of the Aurora-A promoter (Figs. 2.5A and 2.5B). Since mRNA and protein levels of Aurora-A are low at G1 and gradually increase during G2/M phase, we next examined if E2F3 binding to and activa tion of the Aurora-A promoter is cell cycle depende nt. HeLa cells were transfec ted with pGL3-Aurora-A or pGL3-vector. Following synchronization w ith double-thymidine block, ChIP and luciferase reporter assays were performed at different phases of th e cell cycle. Fig. 2.5C shows that E2F3 barely interacts with Au rora-A promoter during G1/S phase. The binding and transactivation ac tivities of E2F3 toward th e Aurora-A promoter were significantly increased upon ce ll entering G2/M phase. Thes e data further support the notion that E2F3 plays a pivotal role in Aurora-A expression during G2/M phase.

PAGE 54

42 Fig. 2.5. E2F3 binds to Aurora-A promot er in a cell cycle dependent manner. ( A and B ) E2F3 interacts with Aurora-A promoter in vivo ChIP assay was performed as

PAGE 55

43 described in Materials and Me thods with unsynchronized HeLa cells (left panels). E2F3 binding sites C and D are too close to sepa rate by ChIP assay. Luciferase assay was performed with Aurora-A-Luc reporter plas mids containing E2F3-binding site mutation (right panels). ( C ) E2F3 binds to Aurora-A promot er during G2/M phase. HeLa cells were synchronized with double-thymidine bloc k and released for indicated times. Cell cycle was monitored with flow cytometry (t op) and ChIP assay was performed for each time point (panel 2). Aurora-A RNA levels (p ane 3) and promoter activity (bottom) are correlated with E2F3 binding to Aurora-A promoter. Correlation of Expression of E2F3 and Aurora A To further demonstrate E2F3 regulation of Aurora-A, we transfected GFP-E2F3 into HeLa cells. After 48 h of transfection, cells were immunostained with anti-Aurora-A antibody. As shown in Fig. 2.6, cells expressing GFP-E2F3 exhibited higher density and clearer centrosome than the cells that did not, further indicating E2F3 upregulation of Aurora-A. Fig. 2.6. Expression of E2F3 increase s Aurora-A density in centrosome. HeLa cells were transfected with GFP-E2F3 and imm unostained with anti-Aurora-A antibody. Two cells are shown in the picture (arrows), one cell expresses E2F3 (gr een) and the other one does not.

PAGE 56

44 We and others have previously show n amplification and overexpression of Aurora-A in ovarian cancers (9,26). However, the freque ncy of elevated Aurora-A protein and/or mRNA is much higher than its change at DNA level (e.g. ~l5%), suggesting the mechanism of activating transcri ption and/or translation of Aurora-A in ovarian cancer cells. While overexpression of E2F3 has been detected in different tumors (27-29), the E2F3 status in ovarian cancer has not been well documented. Thus, we reasoned that E2F3 could be elevated in ova rian cancer and might be a causal factor of upregulation of Aurora-A. To this end, we ex amined protein levels of E2F3 and AuroraA in human ovarian cancers. Im munoblotting analyses were performed in total 8 ovarian cancer cell lines and 72 microdissected ovarian tumor specimens (Figs. 2.7A and 2.7B). Elevated levels of E2F3 and Aurora-A we re detected in 43 of 72 (60%) and 41 of 72 (57%) primary tumors, respectively, as well as majority of ovari an cancer cell lines examined. Notably, 78% (32/41) of tumors w ith high levels of Aurora-A overexpress E2F3 (Fig. 2.7C). Co-upregulation of E2F3 and Aurora-A seems to be predominantly observed in serous ovarian carcinomas and clear cell tumors although the number of cases is relatively small (Fig. 2.7D). These data indicate that in a la rge subset of ovarian tumors elevated Aurora-A might result from expression of high level of E2F3 and further support the findings of biochemical and func tional links between E2F3 and Aurora-A.

PAGE 57

45 Fig. 2.7. Overexpression of E2F3 correlates with Aurora-A level in human ovarian tumors. ( A and B ) Western blot analysis. Representa tive ovarian cancer cell lines (A) tumor and normal ovarian tissue (B) lysates were analyzed by Western blot with indicated antibodies. Intensity of E2F3 and Aurora-A were quantified via ImageJ software. The overexpression of E2F3 and Aurora-A in tumor samples was scored based on the average values of the normal tissues from three independent experiments. Table 2. Summary of expression of E2F3 and Aurora-A in ovarian cancer.

PAGE 58

46 Discussion In this report, we investigated the tran scriptional regulation of Aurora-A by E2F3. Ectopic expression of E2F3 induced mRNA and protein levels of Aurora-A whereas knockdown of E2F3 decreased Aurora-A expr ession and resulted in mitotic cell cycle delay, which resembles the mitotic arrest phenotypes in Aurora-A siRNA treated cells (4). Notably, chromatin immunoprecipitation revealed that E2F3 bound to Aurora-A promoter in vivo and the interaction is cel l-cycle dependent, i.e., primarily occurred during G2/M phase. Moreover, co-overexpressi on of Aurora-A and E2F3 was frequently detected in ovarian cancer. These findings ar e important for several reasons. First, they provide a mechanistic unders tanding of transcriptional regulation of the Aurora-A expression during the cell cycle. Second, a di rect link between Aurora-A and the E2F3 has now been established. Finall y, this is the first demonstrat ion of co-alteration of E2F3 and Aurora-A in human ovarian cancer and el evated E2F3 could be a causal factor for deregulation of Aurora-A in this disease. In mammalian cells, E2F family is com posed of 10 distin ct gene products encoded by eight independent loci that can be divided into three subfamilies based on their sequence homology – the E2F1–3 genes, the E2F4 and 5 genes and the E2F6–8 genes. The E2F1–3 genes have been shown to be tightly regulated during the cell cycle, whereas the E2F4–8 genes are constitutivel y expressed. Functionally, E2F1–3 act as positive regulators of transcription whose accumulation is tightly regulated and in most cell types correlates with increased cell pro liferation (30-32), wh ereas E2F4 and E2F5, when bound to p130 or Rb, act as transcriptio nal repressors (33). The E2F6–8 proteins appear to function as Rb-inde pendent transcriptio nal repressors (34,35). In addition, the E2F3 locus expresses two distinct transcripts, full-length E2F3a and N-terminal-truncated E2F3b transcribed from an intronic promoter within the E2F3 locus (36). E2F3a expression is cell cycle regulated whereas E2F3 b is expressed equiva lently in quiescent and proliferating cells, and may have an oppos ing role to E2F3a in cell cycle control. Accumulated evidence shows th at E2F3 (e.g. E2F3a) regulates S and G2/M cell cycle progression (15,16). Gene expres sion microarray analyses have revealed that E2F3

PAGE 59

47 regulates many of the DNA replication, mitotic and cell cycle regul atory genes (15,21). A previous report has shown that E2F3 re gulates cyclin B1, cyclin A2 and cdc2 transcription (18). We demonstrated, in the present study, that E2F3 directly binds to Aurora-A promoter and tightly regulates Auro ra-A expression during the G2/M phase. Previous studies have shown that Auro ra-A is transcrip tionally regulated by E4TF1, a member of the Ets family, and GAB P, the Ets-related transcription factor GABP (37,38). E4TF1 and GABP bind to the same DNA-binding motif (CTTCCGG; -85 to -79) of the human Aurora-A promoter to induce Aurora-A promoter activity and transcription. The transactiv ation of Aurora-A by GABP is regulated through interaction with TRAP220/MED1, an evol utionarily conserved multisubunit coactivator that plays a central role in regulating tran scription from protein-encoding genes (37,38). Tanaka et al. cloned human Aurora-A promoter and id entified 2 E2F binding elements which correspond to the binging sites A and B in Fig. 2.4C. The findings presented here show that while E2F3 binds to these 2 sites, thei r mutations still respond to E2F3 (Fig. 2.5A). This led us to identify 2 more E2F binding mo tifs proximate to transcriptional start site (Fig. 2.4C). ChIP and reporter assays show that they are required for the transcriptional activation of Aurora-A gene by E2F3 (Figs. 2.5 B and 2.5C). Previous reports have demons trated that E2F3 is fre quently overexpressed in a variety of types of human mali gnancy and whose alteration is associated with late stage and high grade tumors. However, alterations of E2F3 in human ovarian cancer have not been well documented while a gene expre ssion array study shows upr egulation of E2F3 in ovarian tumor (39). Previously, we have shown frequent overexpr ession of Aurora-A in ovarian cancer. In the present study, we demonstrated frequent co-existence of elevated levels of E2F3 and Aurora-A in human primary ovarian carcinoma. Strong association of elevated E2F3 expression and Aurora-A in ovarian tumors underscored the clinical significance of the E2F3-Aurora-A signa ling axis. It is very likely that elevated E2F3 is one of the major transcriptional factor s that contribute to upregulation of AuroraA in other human primary tumors. As described above, E4TF1 and GABP are known transcriptional regulato rs of Aurora-A (37,38), however, th eir role in Aurora-A induction in human cancer are unclear.

PAGE 60

48 Importance of Aurora-A and E2F3 in oncogenesis has been well established by their alterations in human neoplasms and thei r capacity to induce cell transformation (2729,40). Based on our current observations, we be lieve that Aurora-A, as a mitotic E2F3 target gene, could mediate E2F3 oncogenic f unction. Pharmacological agents that inhibit E2F3 or its downstream mediator Aurora-A ma y lead to inhibition of tumor development underscoring the significance of the E2F3-Aurora-A axis as an attractive target for cancer therapy.

PAGE 61

49 References 1. Chan, C. S., and Botstein, D. (1993) Genetics 135 677-691 2. Petersen, J., Paris, J., Willer, M., Philippe, M., and Hagan, I. M. (2001) J Cell Sci 114 4371-4384 3. Pugacheva, E. N., and Golemis, E. A. (2005) Nat Cell Biol 7 937-946 4. Kufer, T. A., Sillje, H. H., Korner, R., Gruss, O. J., Meraldi, P., and Nigg, E. A. (2002) J Cell Biol 158 617-623 5. Hutterer, A., Berdnik, D., Wirtz-Peitz F., Zigman, M., Schleiffer, A., and Knoblich, J. A. (2006) Dev Cell 11 147-157 6. Miyoshi, Y., Iwao, K., Egawa, C., and Noguchi, S. (2001) Int J Cancer 92 370373 7. Bischoff, J. R., Anderson, L., Zhu, Y., Mossi e, K., Ng, L., Souza, B., Schryver, B., Flanagan, P., Clairvoyant, F., Ginther, C., Chan, C. S., Novotny, M., Slamon, D. J., and Plowman, G. D. (1998) EMBO J 17 3052-3065 8. Sen, S., Zhou, H., Zhang, R. D., Yoon, D. S., Vakar-Lopez, F., Ito, S., Jiang, F., Johnston, D., Grossman, H. B., Ruifrok, A. C., Katz, R. L., Brinkley, W., and Czerniak, B. (2002) J Natl Cancer Inst 94 1320-1329 9. Gritsko, T. M., Coppola, D., Paciga, J. E., Yang, L., Sun, M., Shelley, S. A., Fiorica, J. V., Nicosia, S. V., and Cheng, J. Q. (2003) Clin Cancer Res 9 14201426 10. Li, D., Zhu, J., Firozi, P. F., Abbruzzese, J. L., Evans, D. B., Cleary, K., Friess, H., and Sen, S. (2003) Clin Cancer Res 9 991-997 11. Zhou, H., Kuang, J., Zhong, L., Kuo, W. L ., Gray, J. W., Sahin, A., Brinkley, B. R., and Sen, S. (1998) Nat Genet 20 189-193 12. Sakakura, C., Hagiwara, A., Yasuoka, R., Fujita, Y., Nakanishi, M., Masuda, K., Shimomura, K., Nakamura, Y., Inazawa, J., Abe, T., and Yamagishi, H. (2001) Br J Cancer 84 824-831 13. Nevins, J. R. (1998) Cell Growth Differ 9 585-593 14. Dyson, N. (1998) Genes Dev 12 2245-2262

PAGE 62

50 15. Ishida, S., Huang, E., Zuzan, H., Spang, R., Leone, G., West, M., and Nevins, J. R. (2001) Mol Cell Biol 21 4684-4699 16. Polager, S., Kalma, Y., Berkovi ch, E., and Ginsberg, D. (2002) Oncogene 21 437-446 17. Ren, B., Cam, H., Takahashi, Y., Volk ert, T., Terragni, J., Young, R. A., and Dynlacht, B. D. (2002) Genes Dev 16 245-256 18. Zhu, W., Giangrande, P. H., and Nevins, J. R. (2004) EMBO J 23 4615-4626 19. Neufeld, T. P., de la Cruz, A. F., J ohnston, L. A., and Edgar, B. A. (1998) Cell 93 1183-1193 20. Leone, G., DeGregori, J., Yan, Z., Jakoi L., Ishida, S., Williams, R. S., and Nevins, J. R. (1998) Genes Dev 12 2120-2130 21. Black, E. P., Hallstrom, T., Dressman, H. K., West, M., and Nevins, J. R. (2005) Proc Natl Acad Sci U S A 102 15948-15953 22. Cheng, J. Q., Altomare, D. A., Klein, M. A., Lee, W. C., Kruh, G. D., Lissy, N. A., and Testa, J. R. (1997) Oncogene 14 2793-2801 23. Yuan, Z. Q., Feldman, R. I., Sussman, G. E., Coppola, D., Nicosia, S. V., and Cheng, J. Q. (2003) J Biol Chem 278 23432-23440 24. Cheng, J. Q., Godwin, A. K., Bellacosa, A., Taguchi, T., Franke, T. F., Hamilton, T. C., Tsichlis, P. N., and Testa, J. R. (1992) Proc Natl Acad Sci U S A 89 92679271 25. Wells, J., and Farnham, P. J. (2002) Methods 26 48-56 26. Tanner, M. M., Grenman, S., Koul, A., Johannsson, O., Meltzer, P., Pejovic, T., Borg, A., and Isola, J. J. (2000) Clin Cancer Res 6 1833-1839 27. Feber, A., Clark, J., Goodwin, G., Dods on, A. R., Smith, P. H., Fletcher, A., Edwards, S., Flohr, P., Falconer, A., Roe, T., Kovacs, G., Dennis, N., Fisher, C., Wooster, R., Huddart, R., Foster, C. S., and Cooper, C. S. (2004) Oncogene 23 1627-1630 28. Foster, C. S., Falconer, A., Dodson, A. R., Norman, A. R., Dennis, N., Fletcher, A., Southgate, C., Dowe, A., Dearnaley, D., Jhavar, S., Eeles, R., Feber, A., and Cooper, C. S. (2004) Oncogene 23 5871-5879

PAGE 63

51 29. Cooper, C. S., Nicholson, A. G., Foster, C., Dodson, A., Edwards, S., Fletcher, A., Roe, T., Clark, J., Joshi, A., Norman, A., Feber, A., Lin, D., Gao, Y., Shipley, J., and Cheng, S. J. (2006) Lung Cancer 54 155-162 30. Attwooll, C., Lazzerini Denchi, E., and Helin, K. (2004) EMBO J 23 4709-4716 31. DeGregori, J., Leone, G., Miron, A., Jakoi, L., and Nevins, J. R. (1997) Proc Natl Acad Sci U S A 94 7245-7250 32. Lukas, J., Petersen, B. O., Holm, K., Bartek, J., and Helin, K. (1996) Mol Cell Biol 16 1047-1057 33. Leone, G., Nuckolls, F., Ishida, S., Adam s, M., Sears, R., Jakoi, L., Miron, A., and Nevins, J. R. (2000) Mol Cell Biol 20 3626-3632 34. Ogawa, H., Ishiguro, K., Gaubatz, S., Livi ngston, D. M., and Nakatani, Y. (2002) Science 296 1132-1136 35. Logan, N., Graham, A., Zhao, X., Fish er, R., Maiti, B., Leone, G., and La Thangue, N. B. (2005) Oncogene 24 5000-5004 36. He, Y., Armanious, M. K., Thomas, M. J., and Cress, W. D. (2000) Oncogene 19 3422-3433 37. Tanaka, M., Ueda, A., Kanamori, H., Ideguchi, H., Yang, J., Kitajima, S., and Ishigatsubo, Y. (2002) J Biol Chem 277 10719-10726 38. Udayakumar, T. S., Belakavadi, M., Choi, K. H., Pandey, P. K., and Fondell, J. D. (2006) J Biol Chem 281 14691-14699 39. Lu, K. H., Patterson, A. P., Wang, L., Ma rquez, R. T., Atkinson, E. N., Baggerly, K. A., Ramoth, L. R., Rosen, D. G., Liu, J., Hellstrom, I., Smith, D., Hartmann, L., Fishman, D., Berchuck, A., Schmandt, R., Whitaker, R., Gershenson, D. M., Mills, G. B., and Bast, R. C., Jr. (2004) Clin Cancer Res 10 3291-3300 40. Xu, G., Livingston, D. M., and Krek, W. (1995) Proc Natl Acad Sci U S A 92 1357-1361

PAGE 64

52 CHAPTER III AURORA-A FEEDBACK REGULATION OF CHFR, A TUMOR SUPPRESSOR FREQUENTLY DOWNREGULA TED IN OVARIAN CANCER Abstract CHFR is a tumor suppressor gene which enc odes a forkhead-associated (FHA) and ringfinger (RF) domains containing protein and plays an important role in G2/M cell cycle checkpoint. Previous studies s how that CHFR exerts tumor suppressor function via its E3 ligase activity by targeting Au rora-A and polo-like kinase degradation. Hypermethylation and mutation of CHFR have been detected in human cancer from different organs including stomach, breast, colon and lung. Ho wever, molecule(s) that regulate CHFR function and alteration of CHFR in human ovarian cancer remain largely unknown. Here we demonstrate that Aurora-A phosphoryl ates CHFR at serine-218 and serine-337 in vitro and in vivo Phosphorylated CHFR reduces its E3 activity and tumor suppressor function. Further, frequent downregulation of CHFR was detected in human ovarian cancer cell lines and primary tumors. Expre ssion of the CHFR could be restored by treatment with methyltransferase inhibitor 5-aza-2'-deoxycytidine and histone deacetylase inhibitor TSA. Three of 4 cell lines and 9 of 17 tumors with downregulation of CHFR exhibited higher levels of Aurora-A, supporting the notion that CHFR is an E3 ligase of Aurora-A. These findings indicate that CHFR is a substrate of Aurora-A and that epigenetic inactivation of CHFR is a recurre nt event in ovarian can cer and could play a pivotal role in ovarian oncogenesis. Key words : tumor suppressor, E3 ligase, hype rmethylation, phosphorylation, ovarian cancer

PAGE 65

53 Introduction The segregation of chromosomes at mitosi s involves several processes, including chromosome condensation in prophase, ch romosomal alignment on the spindle in metaphase, and sister-chromatids separa tion in anaphase (1-3). CHFR (ch eckpoint with f orkhead and r ing finger) has been discovered as a prophase ch eckpoint protein. Activation of CHFR delays chromosome c ondensation and nuclear envelope breakdown in response to mitosis stress induced by taxol or nocodazole (4). The CHFR protein contains an N-terminal forkhead-associate d (FHA) domain, a central ring finger (RF) region, and a C-terminal cysteine-rich (CR) motif. The FHA domain is conserved in several checkpoint genes, which plays a ro le in recognizing phos phorylated proteins. Deletion of the FHA domain has been shown to attenuate CHFR function; the CR domain is also required for the checkpoint function (4). The RF domain is required for CHFR ubiquitin ligase activity. It ha s been shown that CHFR medi ates the degradation of pololike kinase 1(PLK1), Aurora-A and CHFR itself. Ubiquitination of PLK1 by CHFR delays the activation of Cdc25C and the inac tivation of Wee1, leadi ng to delay of Cdc2 activation, which provides a molecular m echanism that CHFR executes checkpoint function by ubiquitin-dependent proteolysis (5). However, another study suggests that CHFR delays cells in early prophase by inhibitin g entry of cyclin B1 in the nucleus (6). CHFR localizes to chromosome 12q24.33, wh ere is frequently deleted in human cancers (7-9). The frequent inactivation of CHFR which caused by DNA methylation or mutation has been reported in human primar y tumor and cancer cell lines of lung, colon, esophagus, gastric, brain, bone a nd breast (10-14). CHFR-deficie nt mice are cancer-prone and develop spontaneous tumors (15). Aurora-A kinase is required for m itotic entry, centrosome maturation and chromosome segregation (16,17). Aurora-A expr ession and kinase activ ity are cell cycle regulated. Aurora-A mRNA and protein levels begin to accumulate at G2/M transition. After executing its mitotic kinase function through metaphase, Aurora-A protein is rapidly degraded by hCdh1-activated anapha se-promoting complex/cyclosome (APC/C) (18). The precise timing of activation and de gradation is the key feature of Aurora-A

PAGE 66

54 regulation. A previous report ha s demonstrated that CHFR se rves as the E3 ligase for Aurora-A proteasome degradation (15). In th e present study, we show that CHFR is a substrate of Aurora-A and is downregulated in ovarian cancer. Aurora-A phosphorylates CHFR leading to inhibition of its E3 ligas e activity. We previous showed frequent upregulation of Aurora-A in human ovarian cancer and demonstrated in this report inverse correlation of Aurora-A and CHFR expression in this malignancy. Thus, these results uncover a feedback regulation loop between Aurora-A and CHFR and suggest potential importance of CHFR-Aurora-A axis in ovarian car cinogenesis.

PAGE 67

55 Materials and Methods Cell Lines, Transfection and Treatment The human epithelial ovarian cancer cell lines were cultured in appropriate mediums supplemented with 10% fetal bovine seru m: OV3, OV5, OV8, A2780S, IGROV, OV90 and C13 were grown in RPMI 1640 medium ; OV2 was cultured in 50:50 medium M199:MCDB105; RMUG-S was grown in F-12 medium; UTOV1, UTOV2, UTOV3A and UTOV4 were cultured in DMEM. All th e media were supplemented with 2 mM Lglutamine, 1% nonessential amino acid s, 100 IU/ml penicillin and 100 IU/ml streptomycin. Human embryonic kidney (H EK) 293 cells were cultured in DMEM containing 10% FBS. Cells were seeded in 60-mm Petri dishes at a density of 0.5 106 cells/dish. Cells were grown overnight to r each log growth phase before transfection. Transfection was performed using LipofectAMINE 2000 (Invitrogen). To restore endogenous CHFR expression, the cells were treated methyltransferase inhibitor 5-aza-2'-deoxycytidin e (5-aza) at concentration of 5 M for 4 days and HDAC inhibitor Trichostatin-A (TSA, 100 ng/ml) fo r last 24 hours. Cycloheximide (CHX) was used to block protein synthe sis at a concentration of 50 g/ml. Proteasome inhibitor MG132 was administered to cells at 10M for 6 hours in ubiquitination assay. Plasmids pcDNA3-myc-CHFR was kindly provided by Dr. J unjie Chen. Different truncated forms of GST-CHFR were created by PCR amplification using pcDNA3-myc-CHFR as a template. The PCR products were digested by Bam HI and Xh oI and cloned into pGEX4T-1 vector. The accuracy of resu lting constructs was confirmed by DNA sequencing. The oligonucleotide primers used were as follows: CHFR1 (sense), 5’CGCGGATCCAAGCTTATG GAGCGGCCCGAGGAAGG CAAG-3’; CHFR110 (sense), 5’-CGCGGATCCAAGCTT AATGAACCGGAACACAACGTGGCA-3’; CHFR124 (antisense), 5’TGCTCTAGCCTCGAGACTTAAAGATTCATAGAGGTATGC-3’; CHFR279 (sense), 5’-CGCGGATCCAAGCTTAGAAATGCCCAAACCGTCCACGAG-3’; CHFR281

PAGE 68

56 (antisense), 5’-TGCTCTAGCCTCG AGGGCATTTCTACGCGGTTGTGCGAC-3’; CHFR376 (antisense), 5’TGCTCTAGCCTCGAGCACATC TTCTTCACTGCGACTCTT-3’. CHFR mutant plasmids were generate d by QuikChange (Multi) Site-Directed Mutagenesis Kit (Stratagene), includi ng pCDNA3-myc-CHFR/S218A, pCDNA3-mycCHFR/S337A, pCDNA3-myc-CHFR/S218S 337A; and GST-CHFR-C2/S218A, GSTCHFR-C2/S337A, GST-CHFR-C2/S218S337A. Prim ers used for mutagenesis reactions were: CHFR-S218A-F, 5’-CCTAAAGGAGC TGGTCCCTCTGTG-3’; CHFR-S218A-R: 5’-CACAGAGGGACCAGCTCCTTTAGG -3’; CHFR-S337A-F, 5’ATGGAGCGCTCGGCCCTGTGTCCT-3’; CHFR-S337A-R, 5’AGGACACAGGGCCGAGCGCTCCAT-3’. RT-PCR Total RNA was isolated from cells or tissues using Trizol Reagent (Invitrogen). 1-2 g RNA was reverse transcripted by M-MLV reve rse transcriptase (Promega) at 37C for 1.5 hour. The sequential PCR reaction to amp lify CHFR from the cDNA was carried out using Taq polymerase (Promega) with 30 cycles of denaturation 30 seconds at 95C; annealing 30 seconds at 60C; and extensi on 60 seconds at 72C. The primers used for PCR were: CHFR-RT-F, 5’-GGCGAGAGC GTTCTCCAGTTG-3’; CHFR-RT-R, 5’GCATGTCAGCGTCTCCTCCATCTTG-3’. Western Blotting Analysis and Immunoprecipitation Western blotting analysis was performed as pr eviously described (19). Briefly, cells and tumor tissues were lysed in RIPA buffer (50 mM NaCl, 0.5% (w/v) DOC, 50 mM TrisHCl (pH 8.0), 1% (v/v) NP-40, 0.1% (w/v) SD S, 1g/ml aprotinin, 1g/ml leupeptin, 0.5mM Na3VO4 and 1mM PMSF). Equal amounts of proteins were separated in SDSpolyacrylamide gel and electr oblotted onto Nitrocellulose (Amersham) membranes. Following blocking in TBS-T containing 5% m ilk, the membranes were hybridized with appropriate antibodies indicated in figure le gend. Detection of an tigen-bond antibody was carried out using ECL Western Blot ting Detection System (Amersham).

PAGE 69

57 For immunoprecipitation (IP), cells were lysed in TNEN buffe r containing 50 mM Tris-HCl (pH 7.5), 150 mM NaCl, 2 mM ED TA, 0.5% (v/v) NP40, 0-0.3% (v/v) Triton X-100, protease and phosphatase inhibitors (1g/ml aprotinin, 1g/ml leupeptin, 0.5mM Na3VO4 and 1mM PMSF). The lysates were pr ecleared with 25l pr otein A-protein G (2:1) agarose beads at 4C for 30 min. Followi ng removal of the b eads by centrifugation, the lysates were incubated with 30l antibody conjugate protein A:G (2:1) agarose beads at 4C for 4 hours. The beads were washed 3 times with lysis buffer and subjected to immunoblotting analysis with the an tibodies indicated in figure legend. In vivo [32P] Pi Labeling HEK293 cells were co-transfected with my c-CHFR and HA-Aurora-A or pcDNA3. After 24 hours incubation, cells was serum starved for overnight and then incubated with phosphate-free and serum-free medium for 2 hours. [32P] Pi was added to the same medium at final concentration of 0.5 mCi/ ml. Following 4 hours incubation, cells were lysed and subjected to immunoprecip itation with anti -Myc antibody. The immunoprecipitates were separated by SDSPAGE and transferred to membranes. Phosphorylated CHFR band was examined by autoradiography. The expression of transfected CHFR and Aurora -A was detected with anti -Myc and -HA antibodies, respectively. In vitro Aurora-A Kinase Assay Aurora-A kinase assay was performed as pr eviously described (20). Briefly, reactions were carried out in a kinase buffe r in the presence of 10 Ci of [ -32P] ATP (Perkin Elmer Life Sciences). Recombinant Histone H3 or purified GST-CHFR protein was used as substrate. After incubation at 30C fo r 30 min, the reaction was stopped by adding protein loading buffer. The proteins were separated by SDS-PAGE, and the amounts of incorporated radioac tivity were determined by autoradiography.

PAGE 70

58 2-Dimentional Phosphopeptide Mapping and Edman Degradation Following in vitro kinase assay, Aurora-A-phosphor ylated CHFR was extracted from the SDS-polyacrylamide gel by cutting the protei n band and incubating in a buffer containing 50 mM ammonium bicarbonate, 1% 2mercaptoethanol, and 0.2% SDS for at least 90 min duplicated at 37C. After chloro form precipitation and oxidation with performic acid, the sample was evaporated and resuspende d by ammonium bicarbonat e. The protein was digested with 10 ng of L-1-tosylamido-2-phenylethyl chloromethyl ketone -treated trypsin at 37 C for overnight, followed by an additional overnight incubation at 37C in the presence of a fresh 10 ng of the same protei nase. The trypsin digests were washed for 4 times, lyophilized, and resuspended in deionized water. Phosphopeptides of CHFR were resolved at the first-dimension by electrophoresis on the cellulose layer of TLC plate for 30 min at 1.0 kV in 88% formic acid/glacial acetic acid/water (1:3.1:35.9; pH 1.9) and 4.7 ( n -butanol:pyridine:glaci al acetic acid:water = 2:1:1:36), and then for 20 min in a buffer pH 8.9 (10 g ammonium carbonate in 1 liter water) using the HTLE-7000 system. The phosphope ptides were further separated at second-dimension by chromatography in Phosphochromatography buffer ( n butanol:pyridine:glacial acet ic acid:water = 5:3.3:1:4). The phosphopeptides were visualized by autoradiography. For phosphoamino acid analysis, the phosphope ptides were recovered from the TLC plates and hydrolyzed in 6N HCl at 110C for 60 min. Following acid hydrolysis, phosphoamino acids were separated by two-dime nsional electrophoresis on TLC plates. The electrophoresis was carried out for 20 min at 1.5 kV in pH 1.9 buffer for the first dimension and for 16 min at 1.3 kV in pH 3.5 buffer (pyridine:glac ial acetic acid:water = 1:10:189) for the second dimens ion. Phosphoamino acid standards, which were mixed with each sample, were visualized by st aining with 0.25% w/v ninhydrin in acetone. For Edman degradation, the phosphopeptides were eluted from the cellulose layer of the TLC plates by incubation in deionize d water for 30 min at room temperature; second elution may be applied if necessary. The purified phosphopeptides were subjected to manual Edman degradation to determine at which position the phosphorylated amino acid was present in the peptide. At pH 8.0 to 9.0, phenylisothiocyanate reacted with the

PAGE 71

59 free amino group of the most amino-terminal amino acid to form a corresponding phenylthiocarbamyl (PTC) peptide. During each cycle of Edman degradation, treatment of trifluoroacetic acid (TFA) resulted in the cleavage and release of the derivatized amino-terminal amino acid, and a sample from the reaction mixture was withdrawn. If a phosphoserine or phosphothreonine residue is present, a -elimination during the cyclization released free phosphate. The free phosphate was separated from the peptide by electrophoresis, and the phos phoamino acid was determined. In vivo Ubiquitination Assay The cell ubiquitination assay was carried out as previously described (21). HEK293 cells were transfected with combinations of His6-Ubiquitin, HA-PLK1, myc-CHFR and HAAurora-A using calcium phosphate. After 36 h of transfection, cells we re collected into two aliquots, one of which was used for West ern blot analysis to confirm expression of transfected plasmids and the other wa s subjected to purification of His6-tagged protein using Ni2+-nitrilotriacetic acid (NTA) conjugated agarose beads (Qiagen). Briefly, the cell pellet was lysed in buffer A (6 M guanidinium-HCl, 0.1 M Na2HPO4/NaH2PO4 [pH 8.0], 0.01 M Tris-HCl [pH 8.0] 10 mM imidazole, 10 mM -mercaptoethanol) and incubated with Ni2+-NTA beads for 6 hours at room temperature. The beads were sequentially washed with buffer A, buffer B [8 M urea, 0.1 M Na2PO4/NaH2PO4 [pH 8.0], 0.01 M Tris-HCl (pH 8.0) 10 mM imidazole, 10 mM -mercaptoethanol], buffer C [8 M urea, 0.1 M Na2PO4/NaH2PO4 (pH 6.3), 0.01 M Tris-HCl (pH 6.3), 10 mM imidazole, 10 mM -mercaptoethanol], 0.2% Triton X-100, and then buffer C. The beads bound proteins were eluted with buffer D [200 mM imidazole, 0.15 M Tris-HCl (pH 6.7), 30% glycerol, 0.72 M -mercaptoethanol, 5% SDS] at r oom temperature. The eluted proteins were analyzed with Western blot in the presence of ubiquitin-conjugated PLK1 using anti-HA antibody.

PAGE 72

60 Results Downregulation of CHFR in Human Ovari an Cancer and the Correlation between Expression of CHFR and Aurora-A While epigenetic inactivation of CHFR has been observed in various types of human malignancy (22-26), its alterati on in ovarian cancer remains el usive. Since CHFR resides in chromosome 12q24 region, which is frequent ly deleted in ovarian cancer (27), we examined expression of CHFR in both ovari an cancer cell lines and primary tumors by semi-quantitative RT-PCR. Fig. 3.1A shows th at CHFR was downregulated in 8 of 17 ovarian cancer cell lines and 17 of 30 primar y ovarian tumors examined. Undetectable CHFR was observed in 4 cell lines and 9 primary tumors. To determine if DNA methylation is responsible for CHFR downregulation in ovarian cancer, we treated 4 ovarian cancer cell lines, which do not express CHFR, with combined 5-aza (methyltransferase inhibitor) and TSA (HDAC inhibitor). CHFR expression was restored in C13 cells (Fig. 3.1B) but not the other 3 cell lines (data not shown), suggesting that DNA methylation is not the only mechanism fo r downregulation of CHFR. Nevertheless, these results indicate that de regulation of CHFR is a recu rrent event in human ovarian cancer. We and others have previously show n amplification and overexpression of Aurora-A in 15% and 50% ovarian cancer s, respectively (28), suggesting that deregulation of Aurora-A at transcriptional or /and translational levels occurs in a large fraction of tumors. Since CHFR is an E3 ligase of Aurora-A ( 15), downregulation of CHFR could be a causal factor for upregulation of Aurora-A By comparing expression levels of CHFR and Aurora-A, we found an inverse correlation between Aurora-A and CHFR expression in a subset of ovarian cancer (Fig. 3.1C). For instance, C13 and SW626 cells do not express CHFR, but have high levels of Aurora-A, whereas SKOV3 and OV90 express CHFR but not Aurora-A (Fig. 3.1C) However, it is noted that a subset of ovarian cancers exhibit high levels of both Aurora-A and CHFR, s uggesting a feedback regulation between Aurora-A and CHFR.

PAGE 73

61 Fig. 3.1. Frequent downregulation of CHFR in ovarian cancer. ( A ) RT-PCR analysis of CHFR mRNA levels in ovarian cancer ce ll lines and primary ovarian tumors. Total RNA was isolated and subjected to RT-PCR an alysis for expression of CHFR (top) and -actin (bottom). ( B ) Restoration of CHFR in C13 cel ls by demethylation. C-13 and OV3 cells were treated with 5-Aza for 72 hours a nd together with TSA for another 24 hours. ( C ) Correlation between expression of Aurora -A and CHFR. Indica ted ovarian cancer cells were immunoblotted with an ti-Aurora-A (top panel) and -actin (panel 2) antibodies. RNAs from these cells were subjected to RT-PCR analysis for expression of CHFR (panel 3) and -actin control (bottom panel). Aurora-A Phosphorylates CHFR in vivo and in vitro To determine if Aurora-A phosphorylates CHFR, we carried out in vivo labeling experiment. HEK293 cells were co-transfect ed with Myc-CHFR together with/without HA-Aurora-A. Following 48 h incubation, cell s were labeled with orthophosphate [32P] Pi and immunoprecipitated with anti-Myc antibody. The immunoprecipitated Myc-CHFR

PAGE 74

62 was separated by SDS-PAGE and the phosphor ylated CHFR protein was revealed by autoradiography. Fig. 3.2A shows that phosphoryl ation level of CHFR was significantly increased in cells cotransfected with Aurora -A/CHFR as compared to cells treated with CHFR alone. We next carried out in vitro kinase assay. Immunopr ecipitated CHFR was incubated with or without recombinant Auro ra-A protein (Upstate). As shown in Fig. 3.2B, CHFR phosphorylation was detected only in the presence of Aurora-A kinase. These findings demonstrate that CH FR is phosphorylated by Aurora-A in vitro and in vivo Fig. 3.2. CHFR is phosphorylated by Aurora-A. ( A ) In vivo labeling. HEK293 cells were transfected with myc-CHFR and HA-Aurora-A or vector, labeled by [32P] Pi and immunoprecipitated with anti-Myc antibody. Th e immunoprecipitates we re separated in SDS-PAGE and the phosphorylat ed CHFR was revealed by autoradiography (top panel). Expression of Aurora-A and CHFR as well as even loading were confirmed by Western blot analysis with anti-HA (panel 2), -Myc (pan el 3) and -actin (bottom panel) antibodies. ( B ) In vitro kinase assay. Myc-CHFR was introduced to HEK293 cells and immunoprecipitated with anti-myc anti body. The immunoprecipitated Myc-CHFR was incubated with recombinant Aurora-A protein in kinase buffer for 30 min. The reactions were stopped by protein loading buffer and resolved by SDS-PAGE (top panel). Bottom panel is the gel stained with Coomassie blue. Above phosphorylation levels of CHFR were quantified by ImageJ.

PAGE 75

63 Mapping Phosphorylation Sites of CHFR by Aurora-A To define the Aurora-A phosphorylation site (s) of CHFR, we generated a series of glutathione S-transferase (GST)-CHFR tr uncated fusion proteins (Fig. 3.3A). In vitro kinase assay revealed that CHFR fragment s C2 (aa1-376) and C4 (aa110-376) were highly phosphorylated by Aurora-A; but mini mal or no phosphorylation was observed in C1 (aa1-124), C3 (aa110-281) and C5 ( aa279-376). This indicates that Aurora-A phosphorylation of CHFR site(s) must re side in a region between aa110 to aa376. To map the phosphorylation sites of CHFR by Aurora-A, we performed 2Dimensional Phosphopeptide Mapping analysis. In vitro kinase assay was carried out by incubation of GST-CHFR-C2 protein with recombinant Aurora-A. The phosphorylated GST-CHFR-C2 was extracted from SDS-PAGE gel. After digestion with trypsin, which cleaves the peptides after R/K, the phosphope ptides were separated on TLC plates. As shown in Fig. 3.3B, we identified 2 majo r phosphopeptides A and B. To determine whether the phosphorylated resi due is serine or threonine we performed Phosphoamino Acid Analysis. The phosphopeptides were r ecovered from the TLC plates. Following acid hydrolysis, phosphoamino acids were se parated by two-dimensional thin-layer electrophoresis on TLC plates. The plates were dried in 80C for 30 minutes, sprayed with ninhydrin in acetone and reheated for 5 minutes to visualize the phosphoamino acid standards. Following autoradiograp hy, align the film with the plate and both the residues were identified to be serine (Fig. 3.3C). The phosphopeptides isolated from TLC plates were subjected to manual Edman degradation to determine the position of the phosphorylated amino acid in the peptide. Basi cally, the labeled peptides were cleaved one by one from the very amino-terminal of the peptide without disrupting the peptide bonds between other residues. The [32P]-labeled phosphate group was released where the phospho-residue was cleaved, and form a separate spot from the peptide and closer to the anode, because the phosphate is smaller and negatively charged. As shown in Fig. 3.3D, both the phosphopeptides released the phosphate at the second cycle. Judging from the amino acid sequence of CHFR-C2, the e ndoproteinase trypsin-digested peptides that contain serine or threonine at position 2 ar e serine-218 and serine-337, respectively (Fig. 3.3D).

PAGE 76

64

PAGE 77

65 Fig. 3.3. Define Aurora-A phos phorylation sites of CHFR. ( A ) Different CHFR fragments were fused to GST (top). In vitro Aurora-A kinase a ssay was carried out by incubation of GST-CHFR fusion proteins with recombinant Aurora-A in a kinase buffer containing10 Ci of [ -32P] ATP. After 30 min incubation, the reactions were separated in SDS-PAGE and exposed to X-ray film. Hist one H3 was used as a positive control and GST-protein as a negative control (middle) The gel was subsequently stained with Coomassie blue (bottom). ( B ) Aurora-A phosphorylates CHFR at 2 sites identified by two-dimensional phosphopeptide mapping. Th e phosphorylated GST-CHFR-C2 protein, which was extracted from SDS-PAGE menti oned in A, was digested by trypsin and separated on TLC plate by electrophoresis at fi rst-dimension and chromatography at second-dimension. Each spot on the pl ate represents a phosphopeptide. ( C ) Aurora-A phosphorylation of CHFR on serine residues. Phosphopeptides were recovered from TLC plates. Following acid hydrolysis, phosphoamino acids were separated by twodimensional electrophoresis on TLC plates. Phosphoamino acid standards, which were mixed with each sample, were visualized by staining with 0.25% w/v ninhydrin in acetone. ( D ) The phosphopeptides were eluted from TLC plates and purified for manual Edman degradation. During each cycle of Edma n degradation, the most amino-terminal amino acid was cleaved, and a sample from the reaction mixture is withdrawn. The free phosphate, which can be separated from the peptide by electrophoresis, was released where a phosphoserine or phosphothr eonine residue is present. To confirm this result, we convert ed Ser218 and Ser337 to alanine. In vitro kinase assay, using wild-type GST-CHFR-C2, C2-S218A, C2-S337A and C2-S218/337A as substrates, shows that mutation of either si te or both significantly reduces Aurora-A phosphorylation of CHFR (Fig. 3.4A). Furt her, immunoprecipitation and immunoblotting analysis with anti-pS/T anti body revealed that Aurora-A pho sphorylation of CHFR also decreased in myc-CHFR singl e or double A-form mutants as compared to wild-type CHFR (Fig. 3.4B). Therefore, Ser-218 a nd Ser-337 of CHFR are major residues targeted by Aurora-A.

PAGE 78

66 Fig. 3.4. Confirm Aurora-A phosphorylation sites of CHFR. ( A ) In vitro kinase assay using GST fused wild type CHFR, CHFR-S218A, -S337A a nd -S218/337A as substrates. Recombinant Histone H3 was used as positive control (top). Bottom panel is Coomassie blue staining showing equal amount of GS T-CHFR protein used in the reaction. ( B ) Western blot analysis. HEK293 cells were transfected with Myc-tagged wild type, S218A, S337A and S218/337A CHFR expressing constr ucts together with HA-Aurora-A or control vector. Following 48 h incubation, th e cells were lysed and immunoprecipitated using anti-Myc antibody. The immunoprecip itates were immunoblotted with antiphosphor-Ser/Thr antibody (top). Expression of the transfecte d CHFR and Aurora-A was shown in panels 2 and 3. Relative phosphoryla tion levels of CHFR were normalized to expression and quantified by ImageJ. Mutation of the Phosphorylation Si tes Abolishes CHFR E3 Activity Since serine-218 is close to RF domain a nd seine-337 locates with in the RF domain, which is required for E3 activity (29), we reasoned that Aurora-A phosphorylation of these 2 sites will decrease E3 activity of CHFR. In vivo ubiquitination assay was performed using PLK1 as readout. HEK293 ce lls were transfected with HA-PLK1, HisUbiquitin, Myc-CHFR and HA-Aurora-A. Following treatment with a proteasome

PAGE 79

67 inhibitor MG132 for 6 hours, cells were lysed and subjected to Ni+-NTA-agarose beads to pull-down His-tagged ubiqui tin. The ubiquitin conjugated PLK protein was detected by western blotting with anti-HA antibody. The re sult showed that a po rtion of PLK1 was conjugated to ubiquitin in vivo. Ectopic expression of CHFR stimulated PLK1 polyubiquitination. Addition of Au rora-A reduced CH FR-induced PLK1 degradation (Fig. 3.5A), suggesting that Aurora-A negatively regulates CHFR ubiquiti n ligase function. We next examined if Aurora-A inhib ition of CHFR E3 activity depends on phosphorylation of serine-218 and serine337. We created CHFR-S218/337D and CHFRS218/337A forms by converting serine residu es to aspartic acids and alanines, respectively. HEK293 cells were tran sfected CHFR, CHFR-S218/337D or CHFRS218/337A together with/without Aurora-A and then analyzed by Western blot for PLK1 expression. As expected, wild type CHFR reduced PLK1 protein level, which was abrogated by expression of Aurora-A Further, nonphosphorylatable CHFR-S218/337A decreased PLK1 expression and Aurora-A fa iled to rescue it, whereas expression of phospho-mimic CHFR was unable to reduce PLK1 protein (Fig. 3.5B). Thus, we conclude that Aurora-A inhibits CHFR function through a phosphorylation-dependent mechanism.

PAGE 80

68 Fig. 3.5. Aurora-A inhibits CHFR E3 acti vity through phosphorylation of Ser218 and Ser337. ( A ) In vivo ubiquitination assay. HEK293 ce lls were transfected with indicated plasmids and treated with MG132. The cells were lysed and incubated with Ni2+-NTA beads to pull-down His6-tagged protein. The ubiquitin conjugated PLK1 was evaluated by Western blot an alysis with anti-HA antibody. ( B ) Western blot. HEK293 cells were transfected with indicated expr ession constructs. Following 48 h incubation, cells were lysed and immunoblotted with indi cated antibodies. The relative PLK1 level was normalized to actin and quantified by ImageJ.

PAGE 81

69 Discussion In this study, we demonstrated that CHFR is a substrate of Aurora-A kinase. Aurora-A phosphorylates CHFR at Ser218 and Ser337 and inhibits CHFR E3 activity. As a result, its downstream targets, such as PLK1, become stable. Further, we also showed frequent downregulation of CHFR and i nverse correlation of expressi on of Aurora-A and CHFR in human ovarian cancer, which supports the notion that CHFR is an E3 ligase of AuroraA (15). However, co-expression of CHFR a nd elevated Aurora-A was observed in a subset of ovarian cancers, suggesting th at Aurora-A might phosphorylate CHFR and inhibit its tumor suppressor function in these tumors. Ta ken together, these findings indicate a feedback regulation loop between Aurora-A and CHFR. Previous studies have primarily focuse d on genetic alteration and normal cellular function of CHFR. Posttran slational regulation of CH FR remains largely unknown. A previous report shows that Akt/PKB phosphor ylates CHFR at Thr-39 in FHA domain after DNA damage, which led to inhibition of E3 activity and shortening G2 arrest induced by DNA damage (30). In this study, we showed that Aurora-A inhibits E3 activity of CHFR by phosphorylation of Ser218 and Ser337, which are adjacent and within RF region, respectively. Taken together these data indicate that posttranslational modification of CHFR plays a critical ro le in regulation of its function. It has been suggested that R/K/N-R-X-S/T-B or R/KXT/S -I/L/V is Aurora kinase phosphorylation consensus sequence, X repres ents any amino acid and B represents hydrophobic residues with the exception of Pro (31). However, a number of Aurora-A substrates identified so far do not contain this motif. While the serine-218 [PKGS(218)G] of CHFR is not perfect match either seque nce and the serine-337 [ERSS(337)L] does not comply with R/K/N-R-X-S/T-B, th ey are phosphorylated by Aurora-A in vitro and in vivo Mutation of these 2 sites to aspartic acids, which mimic Aurora-A phosphorylation, reduces E3 activity of CHFR and resemble the effect of Aurora-A phosphorylation of CHFR, whereas CHFR-S218/377A retains its E3 activity th at was not inhibited by Aurora-A. Therefore, CHFR is a physiological substrate of Aurora -A and the Aurora-A consensus phosphorylation motif could be more variable than orig inally expected.

PAGE 82

70 Aurora-A exerts its mitotic function thr ough regulation of a number of proteins (32-36), which include histone H3, TACC3, Eg5, CPEB and TPX2. We and others have previously shown Aurora-A phosphorylation of p53 which leads to inactivation of p53 and G2/M cell cycle progression (20). CHFR ha s been shown to mediate a delay in cell cycle progression early in mitosis in respons e to microtubule stress. Identification of Aurora-A inhibition of CHFR provides addi tional molecular mechanism for Aurora-A function in control of G2 /M cell cycle progression. Several groups have shown that CHFR mRNA expression is lost or decreased in primary lung, colon, esophageal, gastric, brain, and breast tumors as well as cancer cell lines. The best characterized means of expression loss is promoter hypermethylation, which occurs in a subset of tumors and cell lines and the frequency of which seems to be dependent on the tissue of origin (10-14). A r ecent report performed DNA sequence and methylation specific PCR analyses and show ed no mutation and hypermethylation in 48 ovarian cancers examined (37). However, we found frequent downr egulation of CHFR mRNA in both ovarian cancer cell lines and primary tumors by RT-PCR (Fig. 3.1). The hypermethylation was only detected in 1 of 4 cell lines, which is consistent with the previous findings that only a fraction of tu mors with loss of CHFR expression are due to promoter hypermethylation. Further investig ation is required for determining the mechanism of downregulation of CHFR in human cancer including ovarian carcinoma.

PAGE 83

71 References 1. Zegers, M. M., Zaal, K. J., van, I. S. C., Klappe, K., and Hoekstra, D. (1998) Mol Biol Cell 9 (7), 1939-1949 2. Pines, J., and Rieder, C. L. (2001) Nat Cell Biol 3 (1), E3-6 3. Mitchison, T. J., and Salmon, E. D. (2001) Nat Cell Biol 3 (1), E17-21 4. Scolnick, D. M., and Ha lazonetis, T. D. (2000) Nature 406 (6794), 430-435 5. Kang, D., Chen, J., Wong, J., and Fang, G. (2002) J Cell Biol 156 (2), 249-259 6. Summers, M. K., Bothos, J., and Halazonetis, T. D. (2005) Oncogene 24 (16), 2589-2598 7. Rickert, C. H., Dockhorn-Dworniczak, B ., Busch, G., Moskopp, D., Albert, F. K., Rama, B., and Paulus, W. (2001) Acta Neuropathol 102 (6), 615-620 8. Rutherford, S., Hampton, G. M., Frierson, H. F., and Moskaluk, C. A. (2005) Lab Invest 85 (9), 1076-1085 9. Natrajan, R., Williams, R. D., Hing, S. N ., Mackay, A., Reis-Filho, J. S., Fenwick, K., Iravani, M., Valgeirsson, H., Grigor iadis, A., Langford, C. F., Dovey, O., Gregory, S. G., Weber, B. L., Ashwort h, A., Grundy, P. E., Pritchard-Jones, K., and Jones, C. (2006) J Pathol 210 (1), 49-58 10. Mizuno, K., Osada, H., Konishi, H., Ta tematsu, Y., Yatabe, Y., Mitsudomi, T., Fujii, Y., and Takahashi, T. (2002) Oncogene 21 (15), 2328-2333 11. Corn, P. G., Summers, M. K., Fogt, F., Vi rmani, A. K., Gazdar A. F., Halazonetis, T. D., and El-Deiry, W. S. (2003) Carcinogenesis 24 (1), 47-51 12. Shibata, Y., Haruki, N., Kuwabara, Y ., Ishiguro, H., Shinoda, N., Sato, A., Kimura, M., Koyama, H., Toyama, T., Ni shiwaki, T., Kudo, J., Terashita, Y., Konishi, S., Sugiura, H., and Fujii, Y. (2002) Carcinogenesis 23 (10), 1695-1699 13. Kang, H. C., Kim, I. J., Park, J. H., Shin, Y., Park, H. W., Ku, J. L., Yang, H. K., Lee, K. U., Choe, K. J., and Park, J. G. (2004) Oncol Rep 12 (1), 129-133 14. Erson, A. E., and Petty, E. M. (2004) Mol Carcinog 39 (1), 26-33

PAGE 84

72 15. Yu, X., Minter-Dykhouse, K., Malureanu, L ., Zhao, W. M., Zhang, D., Merkle, C. J., Ward, I. M., Saya, H., Fang, G., van Deursen, J., and Chen, J. (2005) Nat Genet 37 (4), 401-406 16. Chan, C. S., and Botstein, D. (1993) Genetics 135 (3), 677-691 17. Petersen, J., Paris, J., Willer, M., Philippe, M., and Hagan, I. M. (2001) J Cell Sci 114 (Pt 24), 4371-4384 18. Taguchi, S., Honda, K., Sugiura, K., Ya maguchi, A., Furukawa, K., and Urano, T. (2002) FEBS Lett 519 (1-3), 59-65 19. Dan, H. C., Sun, M., Yang, L., Feldman, R. I ., Sui, X. M., Ou, C. C., Nellist, M., Yeung, R. S., Halley, D. J., Nicosia, S. V., Pledger, W. J., and Cheng, J. Q. (2002) J Biol Chem 277 (38), 35364-35370 20. Liu, Q., Kaneko, S., Yang, L., Feldman, R. I., Nicosia, S. V., Chen, J., and Cheng, J. Q. (2004) J Biol Chem 279 (50), 52175-52182 21. Pan, Y., and Chen, J. (2003) Mol Cell Biol 23 (15), 5113-5121 22. Yanokura, M., Banno, K., Kawaguchi, M., Hirao, N., Hirasawa, A., Susumu, N., Tsukazaki, K., and Aoki, D. (2007) Oncol Rep 17 (1), 41-48 23. Banno, K., Yanokura, M., Kawaguchi, M., Kuwabara, Y., Akiyoshi, J., Kobayashi, Y., Iwata, T., Hirasawa, A., Fujii, T., Susumu, N., Tsukazaki, K., and Aoki, D. (2007) Int J Oncol 31 (4), 713-720 24. Gong, H., Liu, W., Zhou, J., and Xu, H. (2005) J Huazhong Univ Sci Technolog Med Sci 25 (3), 240-242 25. Kobayashi, C., Oda, Y., Takahira, T., Izumi, T., Kawaguchi, K., Yamamoto, H., Tamiya, S., Yamada, T., Iwamoto, Y., and Tsuneyoshi, M. (2006) Mod Pathol 19 (4), 524-532 26. Sakai, M., Hibi, K., Kanazumi, N., Nomo to, S., Inoue, S., Takeda, S., and Nakao, A. (2005) Hepatogastroenterology 52 (66), 1854-1857 27. Thompson, F. H., Emerson, J., Alberts, D., Liu, Y., Guan, X. Y., Burgess, A., Fox, S., Taetle, R., Weinstein, R., Makar, R., and et al. (1994) Cancer genetics and cytogenetics 73 (1), 33-45

PAGE 85

73 28. Gritsko, T. M., Coppola, D., Paciga, J. E., Yang, L., Sun, M., Shelley, S. A., Fiorica, J. V., Nicosia, S. V., and Cheng, J. Q. (2003) Clin Cancer Res 9 (4), 14201426 29. Freemont, P. S. (2000) Curr Biol 10 (2), R84-87 30. Shtivelman, E. (2003) Mol Cancer Res 1 (13), 959-969 31. Ferrari, S., Marin, O., Pagano, M. A., Meggio, F., Hess, D., El-Shemerly, M., Krystyniak, A., and Pinna, L. A. (2005) Biochem J 390 (Pt 1), 293-302 32. Crosio, C., Fimia, G. M., Loury, R., Kimura, M., Okano, Y., Zhou, H., Sen, S., Allis, C. D., and Sassone-Corsi, P. (2002) Mol Cell Biol 22 (3), 874-885 33. Kinoshita, K., Noetzel, T. L., Pelletier L., Mechtler, K., Drechsel, D. N., Schwager, A., Lee, M., Raff, J. W., and Hyman, A. A. (2005) J Cell Biol 170 (7), 1047-1055 34. Giet, R., Uzbekov, R., Cubizolles, F., Le Guellec, K., and Prigent, C. (1999) J Biol Chem 274 (21), 15005-15013 35. Ma, C., Cummings, C., and Liu, X. J. (2003) Mol Cell Biol 23 (5), 1703-1716 36. Kufer, T. A., Sillje, H. H., Korner, R., Gruss, O. J., Meraldi, P., and Nigg, E. A. (2002) J Cell Biol 158 (4), 617-623 37. Ludwig, A. H., Bujko, M., Bidzinsk i, M., and Kupryjanczyk, J. (2007) Cancer Detect Prev 31 (3), 257-261

PAGE 86

74 CHAPTER IV CYCLIN N-TERMINAL DOMAIN CONTAINI NG 2, CNTD2, IS AN ONCOGENE THAT INTERACTS WITH AND ACTIVATES AURORA-A Abstract Cyclin N-terminal domain containing 2, CNTD 2, is an uncharacteri zed cyclin. Here, we identify CNTD2 as an Aurora-A interacti ng protein. CNTD2 colocalizes with Aurora-A in the centrosome. Interaction between CN TD2 and Aurora-A leads to activation of Aurora-A and cdc2 kinase, which promot es G2/M cell cycle progression. Further, CNTD2 resides at chromosome 19q13.2 and is amplified and overexpressed in human cancer cell lines and primary tumors of ova ry, pancreas, breast, prostate and lung. Alterations of CNTD2 appear to be asso ciated with poor clinic outcome. Ectopic expression of CNTD2 murine fibroblasts results in centrosome amplification, genomic instability and malignant phenotype. These data suggest that CNTD2 is a key regulator of Aurora-A and could play a pivotal role in cell cycle progression and oncogenesis. Key words : amplicon, cyclin, cell cycle, cdc2, oncogenesis

PAGE 87

75Introduction Aurora-A is a mitotic kinase, which locali zes to centrosome and regulates cell cycle progression through modulating centrosome fu nction. Aurora-A is amplified and/or overexpressed in a range of human cancers, including breast, ovaria n, colon, bladder and pancreatic cancers (1-5). Aurora-A overexpre ssion in cultured cells leads to centrosome amplification, multipolar spi ndle and polyploidy, resulti ng in cell-cycle checkpoint defects and genetic instability, which contri butes to malignant transformation. Aurora-A executes regulatory and transforming func tion through its kinase activity, which is regulated by phosphorylation and dephosphorylati on as well as associat ion with a number of proteins such as HEF1, TPX2, or Bora (6-8). Genetic and chromosomal abnormalities, including amplification, deletion and aneuploidy in chromosomes, are common feat ures associated with cancer development and progression (9,10). Amplified chromosomal regions, known as amplicons, have been implicated in a wide variety of cance rs. Well-known amplicons in cancer include chromosome 17q12 containing ErbB-2 ( 11,12), chromosome 7q11.2 containing EGFR (13), 20q13 containing Aurora-A (14) chromosome 19q13.1-q13.2 containing AKT2 (15), chromosome 11q13 containing CCN D1 and EMS1 (16,17), chromosome 12q13–14 containing MDM2 (18,19), N-Myc amplifi cation on chromosome 2p24 (20-22), newly discovered co-amplification of PI3KCA and ZASC1 on 3q26.3 (23) and chromosome 19q12 containing CCNE1 (24). Studies on thes e genes not only have provided new insights on how cancer develops but also have significant translational implications. For example, ErbB-2 and EGFR are the molecu lar targets for the humanized antibodies Trastuzumab (Herceptin) and Matuzumab, which are used for the treatment of breast and lung cancer, respectively (25,26). The two major processes common to all cell cycles are S phase, when chromosomes are replicated, and M phase, when the replicated chromosomes are segregated into two daughter cells. In most cell cycles, an interval of time, G1 phase, separates the previous cell di vision from the beginning of th e next S phase. It is now firmly established that progression of the ce ll cycle — that is, transitions between one

PAGE 88

76phase of the cycle and the ne xt — are controlled by cyclin -dependent kinases (CDKs). The CDK regulating G2/M transition is CDK1 /Cdc2, which is the major mitotic kinase. Cdc2 interacts with B-type cyclins to form a stoichiometric complex, known as mitosispromoting factor (MPF). Binding of the cyclin subunit is required for the phosphorylation and activation of the Cdc2 subunit by cdk-ac tivating kinase (CAK). Two B-type cyclins, B1 and B2, have been identified in mamma ls. Cyclin B1 was first cloned at 1989, and found predominantly expressed in the G2/M phase of cell division. In this study, we identified hypotheti cal protein FLJ13625 as an Aurora-A interacting protein by yeast tw o-hybrid screening. The coding region of this protein maps to chromosome 19q13 amplicon. Sequence analysis showed that this protein contains a cyclin domain (CD) at C-terminus, and it is homologous to yeast B-t ype cyclin 6 (clb6). The protein is identical to human cyclin N-terminus Domain containing protein 2 (CNTD2), only with a cyclin domain th at is 30 amino acids longer than CNTD2. Therefore, we consider it as an alterna tive splice form of CNTD2. Like Aurora-A, CNTD2 localizes to the centr osome and activates Aurora-A and cdc2. Further, the CNTD2 gene is amplified and overexpressed in human cancer. Ectopic expression of CNTD2 is able to cause centrosome amplif ication, genomic instability and oncogenic transformation in 3Y1 cells.

PAGE 89

77Materials and Methods Cell Culture, Transfection and Treatment Human cancer cell lines (MDA-MB-231, MDA-MB-435s, MDA-MB-453, MDA-MB468, T47D, UTOV2, UTOV3A, UTOV5, UT OV7, DU145, D32, U118, PANC-1, Colo357, SW480 and HeLa) and HEK293 cells were maintained in Dulbecco modified Eagle medium (DMEM). Cells cultured in RPMI 1640 medium are OV3, OV8, OV2008 C13, MCF7, LNCaP, PC3, AsPC-1 and A549 ce lls. Ovarian cancer SW626 cells were cultured in 50:50 medium M199:MCDB 105; A2780S, A2780CP and RMUG-S were grown in DMEM/F-12 medium. MCF10A cells were maintained in Mammary Epithelial Growth Medium (MEGM). HCT116 was culture d in McCoy’s 5A medium. All culture mediums were supplemented with 10% fetal bov ine serum (FBS). Rat fibroblast cell line 3Y1 was cultured in DMEM containing 10% calf serum. Transfection was performed with LipofectAMINE 2000 reagent (Invitrogen) For protein degradation assay, cells were treated with cyclohex imide (CHX) at a concentra tion of 50 g/ml for 6 hours. Plasmid Constructions For yeast two-hybrid screening, the N-term inal portion of Aurora-A encoding amino acids 1 to 110 was cloned into the EcoR1 and BamH1 sites of pJK202 to create the bait pNLexA-Aurora-AN. The oligonucleotide prim ers used were as follows: forward, 5 ATGGACCGATCTAAAGAAAACTGC-3 and reverse 5’AGGATTATTTTCAGGTGCCGATG. The coding sequence of CNTD2 was ligated to pGEX4T-1 vector to generate GST-CNTD2 fusion protein (Pharmacia). HA-CNTD2 was constructed by cloning CNTD2 into EcoR1/Not1 of pHM6 vector ; Flag-CNTD2 was cloned by insertion of EcoR1/BamH1 digested CNTD2 into p3 FLAG/CMV10 and/or pIRESflagEGFP2 vectors, respectively. Inducible CNTD2 was obtained by cutting out flag-CNTD2 sequence from pIRES2-flag-CNTD2 and ligated into pTRE2hyg2-HA vector through PvuII site. The primers were: H_CNTD2-1 (forward), 5’-

PAGE 90

78CGCGGATCCGCGAATTCCATGCTGG TGAGAGGCAGGGACCAG-3’, and H_CNTD2-2 (reverse), 5’-CGCGGATCCGCGGCCGCTTACTCCCTGCCCTGAAAG3’. Yeast Two-Hybrid Screen A genetic screen using the yeast two-hybrid system was performed as previously described (27). Briefly, yeas t strain EGY191, which harbor s the LexAop-Leu2 reporter gene, was transformed with bait plasmi d pNLexA-Aurora-AN and reporter plasmid pSH18-34 and subsequently transformed w ith a human brain interaction library. Approximately 2 106 primary library transformants were obtained and plated on Ura His Trp Leu galactose-raffinose plates. Candida te clones were identified by their ability to grow on Ura His Trp Leu galactose-raffinose plates, but not on Ura His Trp Leu glucose plates, and their ability to yield blue colonies on Ura His Trp X-Gal (5-bromo-4-chloro-3-indolyl-D-galactopyranoside)-galactose -raffinose plates, but not on Ura His Trp glucose plates. KC8 cells were tr ansformed with plasmids isolated from positive colonies. The specificity of interaction of candidate plasmids with pNLexA-Aurora-AN was tested by retransf ormation of positive clones into yeast harboring either the aurora-A bait or several unrelated bait plasmids. Nucleotide sequence analysis of cDNA inserts was performed using an Applied Biosystems automated sequencer. Western Blotting and Immunoprecipitation Cells were lysed for Western blotting analys is in RIPA buffer (50 mM NaCl, 0.5% (w/v) DOC, 50 mM Tris-HCl (pH 8.0), 1% (v/v) NP -40, 0.1% (w/v) SDS, 1g/ml aprotinin, 1g/ml leupeptin, 0.5mM Na3VO4 and 1mM PMSF). The proteins were resolved under denaturing condition by SDS-polyacrylamide ge l and transferred onto Nitrocellulose (Amersham) membranes. The membranes were blocked and then incubated with appropriate antibodies indi cated in figure legends. For immunoprecipitation (IP), cells were lysed in TNEN buffe r containing 50 mM Tris-HCl (pH 7.5), 50-150 mM NaCl, 2 mM EDTA, 0.5% (v/v) NP40, 0-0.3% (v/v)

PAGE 91

79Triton X-100, protease and phosphatase inhibito rs (1g/ml aprotinin, 1g/ml leupeptin, 0.5mM Na3VO4 and 1mM PMSF). Precleared-lysates were incubated with 30l antibody conjugate protein A:G (2:1) agarose beads at 4C overnight. Following 3 times washing, the precipitates were denatured and subj ected to Immunoblotting with the antibodies indicated in figure legends. Northern and Southern Blot Total RNA was isolated from cells using Tr izol Reagent (Invitrogen). Northern blot analysis was performed as previously descri bed (15). Briefly, 20 g total RNA from each sample was separated on 1.0% denature agar ose gel. After transf erring to GeneScreen Plus Membrane (PerkinElmer) and prehybrid ization, the membrane was hybridized with [32P]dCTP-labeled Aurora-A cDNA probe in Ex press-Hybridization Solution (Clontech). For Southern blot, 10 g of genomic DNA extracted from cancer cell lines was digested by EcoR I in 100 l volume overnight then precipitated with 2.5 folds of icecold 100% ethanol. The digested and purif ied DNA was subjected to 0.8% agarose gel electrophoresis at 40 volts for 16 hours. Afte r transfer, the membrane was hybridized with radioactive labeled CNTD2 probe. Th e probe was obtained from PHM6-CNTD2 by restriction enzyme digestion with EcoR I a nd Not I. The cDNA fragment was gel purified and resuspended in TE buffer, then labeled with [32P]-dCTP using Prime It II Random Primer Labeling Kit (Stratagene). Indirect Immunofluorescence (IIF) Cells were seeded on 2 cm 2 cm coverslips in 6-well plates, the coverslips were pretreated with polyL -Lysine as needed. After expressi on of transfected proteins or appropriate treatment, cells were fixed and penetrated by 4% paraformaldehyde + 0.2% Triton X-100 in PBS/Mg (0.5 mM MgCl2) 10 min at room temperature. Following 4 times wash in PBS/Mg, cells were stored at 4C or proceeding to blocking. Non-related antigens were blocked by incubating in K nudsen Modified Blocking buffer (0.5% BSA, 0.5% NP-40, 1 mM MgCl2, 1 mM NaN3 in 1PBS) 1 hour at room temperature. After removing the blocking buffer, cells were inc ubated with diluted pr imary antibodies for 2

PAGE 92

80hour at room temperature or 4C overnight. Multiple primary antibodies of different origins were combined. Fluorescence conjuga ted secondary antibodies were applied at appropriate dilutions after washing off the primary antibodies. Following final washes by PBS/Mg, coverslips were mounted onto glass slides using VECTASHIELD mounting medium (with DAPI) (VECTOR), which stains DNA while mounting. A ll the steps were performed in humidified chamber and in dark if fluorescence was present. The slides were kept in dark at 4C before vi ewed by ZEISS automated fluorescent scope. In vitro Aurora-A, p34cdc2 Kinase Assay Aurora-A kinase assay was performed as pr eviously described (28). Briefly, reactions were carried out in the presence of 10 Ci of [ -32P] ATP (Perkin Elmer Life Sciences) and 1 nM cold ATP (Invitrogen) in 30 l kinase buffer containing 16.5 mM MOPS (pH 7.0), 0.375 mM EDTA, 2.25 mM EGTA, 33.75 mM MgCl2, 225 M ATP (cold), 11.25 mM -glycerophosphate, 0.45 mM Na3OV4, 0.5 mM DTT, 0.0125% (v/v) 2mercaptoethanol, and 0.625% (v/v) glycerol. 4 g recombinant Histone H3 used as substrate. After incubation at 30C fo r 30 min, the reaction was stopped by adding protein loading buffer and denaturing. The proteins were separated by SDS-PAGE, and the amounts of incorporated radioactivity were determined by autoradiography. p34 cdc 2 kinase activity was measured as previously described (29). Briefly, transfected Hela cells were lysed in bu ffer containing 50 mM Tr is-HCl (pH 8.0), 0.5% Nonidet P-40, 2 mM EDTA, 137 mM NaCl, 10% glycerol, 2 mM sodium vanadate, 100 mM leupeptin, and 1 mM phenylmethylsulfonyl fluoride. 500 g of protein were incubated with anti-cyclin B antibody conjugate d protein A/G-agarose beads in 1 ml lysis buffer at 4 C for at least 4 hours. Follo wed 3 times washing, beads-bond cdc2/cyclin B complexes were incubated at 30 C with [ -32P]-ATP and Histone H1 in a buffer containing 100 mM NaCl, 10% Trito n X-100, 50 mM ATP and 10 mM MgCl2. The reaction product was resolved by SDSPAGE and exposed to autoradiography.

PAGE 93

81Soft Agar Assay and Tumorigenicity in Nude Mice Soft agar assay was performed as previous described (30). 2105 of CNTD2-transfected 3Y1 cells were suspended in 5 ml of 0.33% (w/v) agar containing DMEM/20% fetal calf serum, overlaid onto a 10 ml volume 0.66% agar solution in 10 cm plates. The cultures were fed twice a week for 4 weeks. Tumor formation in 6-week old NOD SCID mice (Jackson Laboratory) was tested by subcutaneous inoculation and assessed fo r 6 weeks. 3Y1 cells transfected with CNTD2 or control vector were each injected into 8 animals at 5106 cells per mouse. Tumor measurements were made with linear calipers in two ort hogonal directions by the same observer.

PAGE 94

82Results Identification of CNTD2 as an Aurora-A Interacting Protein In order to understand Aurora-A function and its involvement in molecular pathways, we performed yeast two-hybrid scr eening using N-terminal region of Aurora-A as bait. Two overlapping clones were isolated from a human brain library. Sequence analysis revealed that it encodes a longer splicing form of huma n cyclin N-terminal domain containing 2 (CNTD2), with extra 30 amino acids that inse rt between residues 89 and 90 of CNTD2. CNTD2 is an uncharacterized B-type cyclin which contain a C-terminal cyclin domain and an N-terminal CDK5 activator homo logy region (Fig. 4.1A). The amino acid sequence differs substantially from those of the well characterized cyclins but is similar to hypothetical protein LOC779739 in Xenopus tr opicalis (similarity 70%; identity 50%) and clb6 in Saccharomyces cerevisiae [similarity, 60%; identit y, 42%; (Fig. 4.1B)]. To confirm the interaction between Aurora-A and CNTD2, we performed coimmunoprecipitation experiment in HEK293 cells that were tr ansfected with HAtagged-CNTD2 and/or GFP-Aurora-A. Antibody against GFP was able to immunoprecipitate CNTD2, which is detected by HA antibody, from cells co-transfected with both Aurora-A and CNTD2, but not cells transfected with either Aurora-A or CNTD2 alone. In addition, HA-CNTD2 also co -precipitated with GFP-Aurora-A (Fig. 4.2A). Furthermore, antibodies against endoge nous Aurora-A or CNTD2 identified the physical interaction of between Aurora-A and CNTD2 (Fig. 4.2B). CNTD2 Colocalizes with Aurora-A in Centrosome Since Aurora-A is a centrosome kinase and interacts with CNTD2, we next examined if CNTD2 confine to centrosome and colocali zes with Aurora-A. HeLa cells were transfected with GFP-tagge d-CNTD2 and then immunostained with antibody against tubulin, a centrosome marker. As shown in Fig. 4.2C, CNTD2 localizes to centrosome during the cell cycle. Co-immunofluorescence st aining of HeLa cells with anit-CNTD2 and -Aurora-A antibodies revealed that CNTD2 and Aurora-A co-localize in the centrosome. These observations further in dicate CNTD2 interaction with Aurora-A.

PAGE 95

83 Fig. 4.1. CNTD2 sequences. ( A ) Domain structure of hu man CNTD2 (longer splicing form). CDK5_a, CDK5 activator homol ogy region; CYCLIN, cyclin domain. (B) Sequence alignment of human CNTD2 and yeast clb6 by ClustalW2 program. Cyclin domains of CNTD2 and clb6 were highlighted in purple and grey color, respectively.

PAGE 96

84 Fig. 4.2. Interaction and colocalization of CNTD2 and Aurora-A. ( A and B ) Coimmunoprecipitation. HEK293 cells were co-t ransfected with GFP-Aurora-A and HACNTD2, and immunoprecipitated with anti-HA and detected with anti-GFP antibody (top panel) or vise versa (second panel). Panels 3 and 4 show expression of transfected plasmids (A). PANC1 cells were lysed a nd immunoprecipitated with anti-AurA or antiCNTD2 antibody and immunoblotted with anti-CNTD2 or anti-AurA antibody, respectively (B). ( C ) Immunofluorescence staining. As shown in upper panel, GFPCNTD2 (green) transfected Hela cells were synchronized to mitosis and centrosomes was stained with anti-tubulin antibody followed by TRITCconjugated secondary antibody (red); in lower panel, mitotic arrested HeLa cells were stained with anti-CNTD2 and anti-

PAGE 97

85Aurora-A antibodies. CNTD2 was shown in re d (TRITC), Aurora-A was shown in green (FITC), and the nucleus was stained by DAPI. CNTD2 Stabilizes Aurora-A Protein and Stimulates Aurora-A Kinase Activity We next investigated the regulation between CNTD2 and Aurora-A. As Aurora-A is a serine/threonine protein kinase, we first exam ined if CNTD2 is a s ubstrate of Aurora-A. In vivo phosphorylation and in vitro kinase assays exhibi ted no phosphorylation of CNTD2 by Aurora-A (data not shown). However, ectopic expression of CNTD2 increases Aurora-A protein but not mRNA leve ls in a dose-dependent manner. In Tet-On inducible CNTD2 HeLa cells, Aurora-A pr otein level was upregulated upon CNTD2 expression induced by doxycycline (Fig. 4.3A). To determine if CNTD2 inhibits AuroraA degradation, HeLa cells were treated with CHX to block de-novo pr otein synthesis. As shown in Fig. 4.3B, without induction of CNTD 2, Aurora-A protein degradated rapidly with 20% decrease in an hour. Once CNTD2 was expressed, Aurora-A protein became more stable and reached the same degrad ation level (20%) until 6 hours after CHX treatment. Aurora-A kinase is activated during G2/M phase due to the fact that Aurora-A is accumulated and interacts with TPX2 through its C-terminal hydrophobic region (7). In addition, we have previously shown that Aurora -A kinase activity is closely associated with Aurora-A protein levels in human can cers (4). Thus, CNTD2 could activate AuroraA through upregulation or/and interaction with Aurora-A To this end, Aurora-A kinase activity was examined in both CNTD2-i nducible and -knockdown cells. Following treatment of Tet-On CNTD2 HeLa cells wi th doxycycline for 12 h and 24 h, Aurora-A was immunoprecipitated w ith anti-Aurora-A antibody. In vitro kinase assay showed an increase in Aurora-A kinase activity in bot h time points (Fig. 4.3C), when the Aurora-A protein expression was induced (Fig. 4.3A). Further, knockdown of CNTD2 in PNAC1 cells, in which CNTD2 is amplified (Fig. 4.5B), reduced Aurora-A kinase activity, especially in mitotic cells in which Aurora -A protein and activity reach the peak (Fig. 4.3D).

PAGE 98

86 Fig. 4.3. CNTD2 stabilizes Aurora-A protein and induces Aurora-A kinase activity. ( A ) Aurora-A is upregulated by CNTD2. HeLa -TetOn-CNTD2 cells were treated with doxycycline for indicated times and immunobl otted with indicated antibodies. ( B ) CNTD2 inhibits Aurora-A degradation. HeLa-T et-on-CNTD2 cells were treated with or without doxycycline overnight. Following addition of CHX fo r indicated times, cells were subjected to Western blot anal ysis with indicated antibodies. ( C ) CNTD2 induces Aurora-A kinase activity. Following trea tment with and without doxycycline, HeLaTetOn-CNTD2 cells were lysed and the e ndogenous Aurora-A was immunoprecipitated by anti-Aurora-A antibody and subjected to in vitro kinase assay using Histone H3 as substrate. ( D ) Knockdown of CNTD2 decreases Auro ra-A kinase activity. PANC1 cells were transfected with scramble or CN TD2 siRNA, and selected by puromycin. Endogenous CNTD2 protein was analyzed by We stern blot (right). Following treatment with or without nocodazole, cells were im munoprecipitated with anti-Aurora-A antibody

PAGE 99

87and subjected to in vitro Aurora-A kinase assay. The Auro ra-A protein and activity were normalized to expression and quantified by ImageJ software. CNTD2 Promotes G2/M Progression and Activates cdc2 Activity Since Aurora-A activates cdc2 activity and promotes G2/M progres sion by regulation of mRNA polyadenylation of cyclin B1 through phosphorylation of cytoplasmic polyadenylation element binding pr otein (CPEB) (31), we next examined the effects of CNTD2 on cell cycle progression and cdc2 activity. CNTD2 inducible HeLa cells were synchronized by double-thymidine block and si multaneously treated with doxycycline or DMSO at the second thymidine block. Cell cycle analysis showed that CNTD2-cells reached G2/M peak at 6 hours after release from the double-thymidine block as compared to control cells, in which G2/M accumulation ap peared at 9 hours of release (Fig. 4.4A). Further, knockdown of CNTD2 delayed G2 /M cell cycle progression (Fig. 4.4B). To investigate the effect of CNTD2 on cdc2 kinase activity, in vitro cdc2 kinase assay was performed in CNTD2-inducible and control-TetOn Hela cells. After treatment with doxycycline for 12 h and 24 h, cells were immunoprecipitated with anti-cyclin B1 antibody and cdc2 activity was measured by in vitro kinase assay using Histone H1 as substrate. Induced expression of CNTD2 in Hela-TetOn-CNTD2 cells significantly induced cdc2 kinase activity; whereas in control cells where there is no CNTD2 expression, cdc2 activity was not affect ed by doxycycline trea tment (Fig. 4.4C).

PAGE 100

88 Fig. 4.4. CNTD2 promotes G2/M progression and stimulates cdc2 kinase activity. ( A and B ) CNTD2 promotes G2/M progressi on. Following induction of CNTD2 by doxycycline, HeLa-TetOn-CNTD2 cells were synchronized to G1/S with doublethymidine block. After release for indicated times, cells were assayed with flow cytometry (A). PANC1 cells were stably tr ansfected with siRNA of CNTD2 and control

PAGE 101

89siRNA. After treatment with hydroxyurea for 16 cells were released for indicated times and subjected to flow cy tometry analysis (B). ( C ) CNTD2 induces cdc2 activity. HeLaTetOn-CNTD2 and control cells were trea ted with doxycycline and immunoprecipitated with anti-cyclin B antibody. The cdc2 kinase activity was assayed using Histone H1 as substrate and quantified by ImageJ. CNTD2 is Ubiquitously Expressed in Norm al Tissues and is Frequently Altered in Human Cancers Northern and Western blot analyses rev ealed ubiquitous expression of CNTD2 in different normal mouse tissues (Fig. 4.5A) a nd human cancer cell lines. Notably, two and three different transcripts were observed in the cell lines examined with elevated levels of CNTD2 in HL60 and Raji cell s (Fig. 4.5B). However, expression levels of CNTD2, unlike Aurora-A, is not changed dur ing the cell cycle (Fig. 4.5C). Genomic database analysis shows that CNTD localizes to chromosome 19q13.2 within the 19q13.1-13.2 amplicon (Fig. 4.6A). Amplification of chromosome 19q13 has been linked to different types of cancers in cluding pancreatic car cinoma (32-34), ovarian carcinoma (35-37), breast cancer (38) and other cancers (39-41). Inte restingly, CNTD2 is 32-kb centromeric to AKT2, an oncogene frequently amplified in human cancer. To investigate if CNTD2 is also altered in human cancer, we performed Southern blot analysis in 18 cancer cell lines and found that CNTD2 was amplified in PANC1, ASPC1 and OV3 cells (Fig. 4.6B), in which AKT2 is also amplified (15,32).

PAGE 102

90 Fig. 4.5. CNTD2 expression in normal tissues and human cancer cell lines. ( A ) CNTD2 protein expression in normal m ouse tissues shown by Western blot. ( B ) Northern blot membrane from Clontech was hybridi zed with CNTD2 probes. Human cancer cell lines examined: 1. HL60, 2. Hela S3, 3. K562, 4. MOLT-4, 5. Raji, 6. SW480, 7. A549, 8. G361. ( C ) Hela cells were synchronized and co llected at different time points for Northern, Western blot and cell cycle analyses.

PAGE 103

91 Fig. 4.6. CNTD2 is amplified in human cancer cell lines. ( A ) CNTD2 locates to chromosome 19q13.2 close to AKT2. ( B ) Southern blot analysis. DNAs from indicated cell lines were digested with EcoRI and el ectrophoresed in 0.8% agarose gel. After transferring, a membrane was hybridized with [32P]-dCTP labeled CNTD2 (top) and actin (bottom) probes.

PAGE 104

92We further examined the expression of CNTD 2 protein in human cancer cell lines. Immunoblotting analysis revealed an elevated CNTD2 protein level in 3 of 8 breast cancer, 5 of 10 ovarian cancer and 2 of 4 prostate cancer cel l lines examined (Fig. 4.7A). The majority of cell lines with overexpression of CNTD2 protein did not have change at DNA level, suggesting that amplification of CNTD2 is responsible for increase in its protein expression in a small subset of cell lines. To determine the expression level of CNTD2 in human primary tumors, we analyzed 18 pairs of lung cancer and nor mal samples for CNTD2 protein expression. Fifteen (83%) tumor samples showed increased CNTD2 level compared to the adjacent normal tissues (Fig. 4.7B). We further inves tigated the alteration of CNTD2 in ovarian cancer due to the fact that chromosome 19q13 is frequently amplified in this malignancy. Western blotting and immunohist ochemistry analyses reveal ed overexpression of CNTD2 in 35 of 65 (54%) ovarian tumors, whereas on ly 3 of 10 normal ovary samples expressed moderate to low levels of CNTD2 (Figs. 4.8A and 4.8B; Table 3). No significant correlation was found between CN TD2 protein levels and hist ological types (Table 4). However, more frequent overexpression of CN TD2 was observed in late stage and, more significantly, high grade tumors (Table 5). We also analyzed the association of the expression of CNTD2 with the ovarian cancer patients’ survival. Kaplan-Meier curves analysis of 65 patients with ovarian tumors demonstrated a statistically significant negative correlation between patient overall survival time and CNTD2 expression level ( P = 0.05; Fig. 4.8C). These results suggest th at CNTD2 is a candidate oncogene in chromosome 19q13 amplicon and could play a role in human oncogenesis and that its alterations are recurrent events in human cancer and associated with tumor progression and poor clinical outcome in ovarian carcinoma.

PAGE 105

93 Fig. 4.7. CNTD2 is overexpressed in human cancer. ( A ) Elevated levels of CNTD2 in human cancer cell lines. Cell lysates from i ndicated cell lines were immunoblotted with anti-CNTD2 (top) and -ac tin (bottom) antibodies. ( B ) Overexpression of CNTD2 in primary lung cancer. Tissue lysates from paired normal lung and tumor specimens were assayed by Western blot analysis with indicated antibodies.

PAGE 106

94 Fig. 4.8. Overexpression of CNTD2 in ovarian tumors is associated with poor prognosis. ( A ) Western blot. Human primary ovarian tumors were immunoblotted with anti-CNTD2 and -actin antibodies. ( B ) Immunohistochemical st aining. Representative tumors were immunostained with anti-CNTD2 antibody. ( C ) Overall survival in patients with high CNTD2 (n = 35) versus the rema ining patients (n = 30) was plotted by the Kaplan-Meier method. Statisti cal comparison of survival between groups with the logrank statistic suggests that patients with overexpression of CNTD2 in the tumor had poor survival compared with low and no expression (P = 0.018).

PAGE 107

95 Table 3. CNTD2 expression in ovarian tumors and normal tissues. Table 4. Association of CNTD2 expressi on with tumor histological types. Table 5. Correlation of CNTD2 expression level with tumor grade and stage. Ectopic Expression of CNTD2 Induces Malignant Transformation To determine CNTD2 oncogenic activity, we stably transfected 3Y1 cells with pHM6HA-CNTD2 and vector alone. pHM6 vector-tra nsfected 3Y1 grew in monolayer with fibroblast feature, whereas CNTD2 stably -transfected cells exhibited epithelial morphology, larger nucleus and lost their cont act inhibition (Fig. 4.9A). Further, CNTD2 cells grew much faster than control cells, s uggesting the role of CN TD2 in regulating cell proliferation (Fig. 4.9B).

PAGE 108

96Furthermore, soft agar assay was carried out to investigate anchorage-independent growth. Colonies of CNTD2-transfected cells appeared microscopically after 10 days and became visible to the naked eye after 20 days of incubation. Colonies were not observed in soft agar cultures of parent al 3Y1 cells or of cells transf ected with pHM6 vector (Fig. 4.9C). To investigate the tumorigenic activity of CNTD2, clonal cell lines were injected subcutaneously into nude mice. All mice inoculated with CNTD2 transfected cells formed tumors within 14-21 days following injection, whereas vect or-transfected 3Y1 cells did not develop tumors until over 2 m onths (Fig. 4.9D and Table 6). Western blot analysis revealed expressi on of CNTD2 protein only in pHM6-HA-CNTD2-transfected cells and tumors. Fig. 4.9. Ectopic expression of CNTD2 transforms 3Y1 cells. ( A ) Cell morphology. CNTD2and vector-transf ected 3Y1 cells were observed under microscope. ( B ) Cell growth curve. CNTD2-transfected and contro l 3Y1 cells were plat ed in 96-well plates and cell number was calculated daily for 6 days. ( C ) CNTD2-transfected but not control 3Y1 cells grew in soft agar. ( D ) Tumor formation in nude mi ce. CNTD2-transfected and control 3Y1 cells were subcutaneous ly injected into nude mice (5 106 cells/mouse; 8

PAGE 109

97mice/group). Tumor formation was observed in all the mice injected with CNTD2transfected cells and in 1 of 8 mice received mock cells. Western blot showed CNTD2 expression in dissected tumors. Table 6. Tumorigenicity of CNTD2 and control transfected 3Y1 cells. CNTD2 Expression Induces Centrosome Amplification and Genomic Instability Since CNTD2 localizes to centrosome and i nduces transformation, we hypothesized that overexpression of CNTD2 causes centrosome amplification and aberrations in chromosome partitioning at mitosis, leading to catastrophic loss or gain of chromosomes and resulting in either cell death or survival through mali gnant transformation. To test this hypothesis, we immunofluorescence-st ained CNTD2-transformed 3Y1 and control cells with anti-tubulin and -HA (e.g., CNTD2) anti bodies (Fig. 4.10A). Fig. 4.10 shows that control 3Y1 cells have e ither 1 (interphase) or 2 centrosomes (dividing). However, the number of centrosomes is significantly increased in CNTD2-transformed cells. Approximately 28% of CNTD2 transfectants revealed more than two centrosomes, compared with less than 3% of the vector-t ransfected cells showing a similar phenotype (Fig. 4.10B). Centrosome amplification is related to abnormal cell division, chromosome instability and cancer development (42). Chromo somal analysis showed that most (42 of 50) vector-transfected 3Y1 cells were near -diploid, whereas 39 of 50 (78%) metaphases from CNTD2 -transfected cells were polyploid. Am ong these 39 metaphase spreads, 23 were near-tetraploid and c ontained one or two small marker chromosomes, and the remaining 16 metaphases had chromosome counts ranging from 132 to more than 600

PAGE 110

98(Fig. 4.10C). One metaphase contained a ri ng chromosome and numerous (>50) small markers, fragments and double minutes. T hus, overexpression of CNTD2 induces genomic instability. Fig. 4.10. CNTD2 induces centrosome ampl ification and genomic instability. ( A ) HeLa cells were immunostained with anti-tubulin (left) and anti-CNTD2 (middle) antibodies. CNTD2 colocalizes with -tubulin (right panel). ( B ) CNTD2-transfected and control 3Y1 cells were immunostained with anti-tubulin antibody (left). The number of centrosome was counted in both CNTD2-transf ected and control cells, 200 events each (right). ( C ) Karyotype analysis of CNTD2-tr ansfected and control 3Y1 cells.

PAGE 111

99Discussion Using the amino-terminal region of Aurora-A as bait in a yeast two-hybrid screen, we have identified a novel B-type cyc lin, CNTD2. CNTD2 colocalizes and immunoprecipitates with Aurora -A. The interaction results in stabilization Aurora-A protein and activation of Aurora-A kina se. CNTD2 is conserved between yeast, Xenopus and human cells. CNTD2 locates to chromo some 19q13 amplicon. Amplification and/or overexpression of CNTD2 were detected in human cancer cell lines and primary tumors. Ectopic expression of CNTD2 induces tran sformation, centrosome amplification and genomic instability. These findings indicate th at CNTD2 functions as an oncogene and is associated with and regulates Aurora-A. In mammalian cells, the Aurora kinase fa mily has three members: Aurora-A, -B, and -C. Aurora-A expression up-regulates at m itosis and it localizes at centrosome and mitotic spindles. Aurora-A T 288 phosphorylation activat es its kinase activity in late G2 phase at centrosomes, which is prior to and required for the recruitment of cdc2-cyclin B1 to the centrosome. The activ ated cdc2-cyclin B1 then comm its the cell to mitosis (43). While CNTD2 interacts with Aurora-A and act ivates Aurora-A and cdc2 kinase activity, it does not form complex with cdc2 (data no t shown) and its prot ein level is relative stable through the cell cycle. Since N-terminal region (aa 1-110) of Aurora-A, which is required for interaction with CNTD2, is differe nt from those of Aurora-B and Aurora-C, we found no interaction between CNTD2 and Aurora-B or Aurora -C (data not shown). The kinase activity of Aurora-A is regulated by autocatalyt ic phosphorylation of Thr288 in its activatory T-loop (44). This auto catalytic activity of Aurora-A is facilitated by cofactors such as Bora, Ajuba, PAK1, HEF1 and TPX2 (6,8,43,45,46). The mechanism of cofactor-mediated Aurora-A activation is best understood in the case of TPX2. Co-crystallisation revealed that binding of TPX2 to Aurora-A not only induces it to adopt an active conformation but also prevents dephosphorylation of Thr288 by protein phosphatase 1 (PP1) (47). TPX2, Bora and Ajuba are also substrates of the kinase (7,8,43). Unlike these Aurora-A interaction proteins, CNTD2 increases Aurora-A protein stability and kinase activity but is not phosphorylated by Aurora-A.

PAGE 112

100 Several oncogenes have been iden tified from chromosome 19q13 amplicon, including AKT2, PAK4, pancreatic di fferentiation 2 (PD2), SERTAD3 and transcriptional regulator intersex-like ( IXL ) (15,48-51). We have previously shown amplification/overexpre ssion of AKT2 in human ovarian, pa ncreatic and breast cancers. Inhibition of AKT2 inhibits tumor growth and sensitizes cells to chemotherapeutic agentinduced cell death (15,52,53). In this report, we demonstr ated that CNTD2 resides between AKT2 and PAK4 and is amplified/ov erexpressed in human ovarian, pancreatic and lung cancers. Overexpressi on of CNTD2 is associated with late stage/high grade ovarian tumors and accompanied with poor cl inic outcome. Therefore, CNTD2 could be a potential therapeutic target for these mali gnancies. Future invest igations are required for characterization of the molecular mechan ism of CNTD2 stabiliz ation and activation of Aurora-A as well as the ro le of CNTD2 in oncogenesis in vivo

PAGE 113

101References 1. Miyoshi, Y., Iwao, K., Egawa, C., and Noguchi, S. (2001) Int J Cancer 92 (3), 370-373 2. Bischoff, J. R., Anderson, L., Zhu, Y., Mossi e, K., Ng, L., Souza, B., Schryver, B., Flanagan, P., Clairvoyant, F., Ginther, C., Chan, C. S., Novotny, M., Slamon, D. J., and Plowman, G. D. (1998) EMBO J 17 (11), 3052-3065 3. Sen, S., Zhou, H., Zhang, R. D., Yoon, D. S., Vakar-Lopez, F., Ito, S., Jiang, F., Johnston, D., Grossman, H. B., Ruifrok, A. C., Katz, R. L., Brinkley, W., and Czerniak, B. (2002) J Natl Cancer Inst 94 (17), 1320-1329 4. Gritsko, T. M., Coppola, D., Paciga, J. E., Yang, L., Sun, M., Shelley, S. A., Fiorica, J. V., Nicosia, S. V., and Cheng, J. Q. (2003) Clin Cancer Res 9 (4), 14201426 5. Li, D., Zhu, J., Firozi, P. F., Abbruzzese, J. L., Evans, D. B., Cleary, K., Friess, H., and Sen, S. (2003) Clin Cancer Res 9 (3), 991-997 6. Pugacheva, E. N., and Golemis, E. A. (2005) Nat Cell Biol 7 (10), 937-946 7. Kufer, T. A., Sillje, H. H., Korner, R., Gruss, O. J., Meraldi, P., and Nigg, E. A. (2002) J Cell Biol 158 (4), 617-623 8. Hutterer, A., Berdnik, D., Wirtz-Peitz F., Zigman, M., Schleiffer, A., and Knoblich, J. A. (2006) Dev Cell 11 (2), 147-157 9. Lengauer, C., Kinzler, K. W., and Vogelstein, B. (1998) Nature 396 (6712), 643649 10. Sieber, O. M., Heinimann, K., and Tomlinson, I. P. (2003) Nat Rev Cancer 3 (9), 701-708 11. Ross, J. S., and Fletcher, J. A. (1999) Semin Cancer Biol 9 (2), 125-138 12. Luoh, S. W. (2002) Cancer Genet Cytogenet 136 (1), 43-47 13. Eley, G. D., Reiter, J. L., Pandita, A., Pa rk, S., Jenkins, R. B., Maihle, N. J., and James, C. D. (2002) Neuro Oncol 4 (2), 86-94

PAGE 114

10214. Tanner, M. M., Tirkkonen, M., Kallioniem i, A., Collins, C., Stokke, T., Karhu, R., Kowbel, D., Shadravan, F., Hintz, M., Kuo, W. L., and et al. (1994) Cancer Res 54 (16), 4257-4260 15. Cheng, J. Q., Godwin, A. K., Bellacosa, A., Taguchi, T., Franke, T. F., Hamilton, T. C., Tsichlis, P. N., and Testa, J. R. (1992) Proc Natl Acad Sci U S A 89 (19), 9267-9271 16. Ormandy, C. J., Musgrove, E. A., Hui, R., Daly, R. J., and Sutherland, R. L. (2003) Breast Cancer Res Treat 78 (3), 323-335 17. Zaharieva, B. M., Simon, R., Diener, P. A., Ackermann, D., Maurer, R., Alund, G., Knonagel, H., Rist, M., Wilber, K ., Hering, F., Schonenberger, A., Flury, R., Jager, P., Fehr, J. L., Mihatsch, M. J., Ga sser, T., Sauter, G., and Toncheva, D. I. (2003) J Pathol 201 (4), 603-608 18. Oliner, J. D., Kinzler, K. W., Meltzer, P. S., George, D. L., and Vogelstein, B. (1992) Nature 358 (6381), 80-83 19. Reifenberger, G., Liu, L., Ichimura, K., Schmidt, E. E., and Collins, V. P. (1993) Cancer Res 53 (12), 2736-2739 20. Shiloh, Y., Korf, B., Kohl, N. E., Sakai, K., Brodeur, G. M., Harris, P., Kanda, N., Seeger, R. C., Alt, F., and Latt, S. A. (1986) Cancer Res 46 (10), 5297-5301 21. Schneider, S. S., Hiemstra, J. L., Zehnbaue r, B. A., Taillon-Miller, P., Le Paslier, D. L., Vogelstein, B., and Brodeur, G. M. (1992) Mol Cell Biol 12 (12), 55635570 22. Hiemstra, J. L., Schneider, S. S., and Brodeur, G. M. (1994) Prog Clin Biol Res 385 51-57 23. Lin, S. C., Liu, C. J., Ko, S. Y., Chang, H. C., Liu, T. Y., and Chang, K. W. (2005) J Pathol 206 (4), 417-422 24. Leung, S. Y., Ho, C., Tu, I. P., Li, R., So, S., Chu, K. M., Yuen, S. T., and Chen, X. (2006) Mod Pathol 19 (6), 854-863 25. Pegram, M. D., Lipton, A., Hayes, D. F., Weber, B. L., Baselga, J. M., Tripathy, D., Baly, D., Baughman, S. A., Twaddell, T., Glaspy, J. A., and Slamon, D. J. (1998) J Clin Oncol 16 (8), 2659-2671

PAGE 115

10326. Kollmannsberger, C., Schittenhelm, M., Honecker, F., Tillner, J., Weber, D., Oechsle, K., Kanz, L., and Bokemeyer, C. (2006) Ann Oncol 17 (6), 1007-1013 27. Gyuris, J., Golemis, E., Chertkov, H., and Brent, R. (1993) Cell 75 (4), 791-803 28. Liu, Q., Kaneko, S., Yang, L., Feldman, R. I., Nicosia, S. V., Chen, J., and Cheng, J. Q. (2004) J Biol Chem 279 (50), 52175-52182 29. Kao, G. D., McKenna, W. G., and Muschel, R. J. (1999) J Biol Chem 274 (49), 34779-34784 30. Cheng, J. Q., Altomare, D. A., Klein, M. A., Lee, W. C., Kruh, G. D., Lissy, N. A., and Testa, J. R. (1997) Oncogene 14 (23), 2793-2801 31. Sasayama, T., Marumoto, T., Kunitoku, N., Zhang, D., Tamaki, N., Kohmura, E., Saya, H., and Hirota, T. (2005) Genes Cells 10 (7), 627-638 32. Miwa, W., Yasuda, J., Murakami, Y., Ya shima, K., Sugano, K., Sekine, T., Kono, A., Egawa, S., Yamaguchi, K., Haya shizaki, Y., and Sekiya, T. (1996) Biochem Biophys Res Commun 225 (3), 968-974 33. Curtis, L. J., Li, Y., Gerbault-Seureau M., Kuick, R., Dutrillaux, A. M., Goubin, G., Fawcett, J., Cram, S., Dutrillaux, B ., Hanash, S., and Muleris, M. (1998) Genomics 53 (1), 42-55 34. Hoglund, M., Gorunova, L., Andren-Sandberg, A., Dawiskiba, S., Mitelman, F., and Johansson, B. (1998) Genes Chromosomes Cancer 21 (1), 8-16 35. Thompson, F. H., Nelson, M. A., Trent, J. M., Guan, X. Y., Liu, Y., Yang, J. M., Emerson, J., Adair, L., Wymer, J., Balfour C., Massey, K., Weinstein, R., Alberts, D. S., and Taetle, R. (1996) Cancer Genet Cytogenet 87 (1), 55-62 36. Bicher, A., Ault, K., Kimmelman, A., Gershenson, D., Reed, E., and Liang, B. (1997) Gynecol Oncol 66 (1), 36-40 37. Wang, Z. J., Churchman, M., Campbe ll, I. G., Xu, W. H., Yan, Z. Y., McCluggage, W. G., Foulkes, W. D., and Tomlinson, I. P. (1999) Br J Cancer 80 (1-2), 70-72 38. Muleris, M., Almeida, A., Gerbault-Seureau, M., Malfoy, B., and Dutrillaux, B. (1995) Genes Chromosomes Cancer 14 (3), 155-163

PAGE 116

10439. Marchio, A., Meddeb, M., Pineau, P., Dangl ot, G., Tiollais, P., Bernheim, A., and Dejean, A. (1997) Genes Chromosomes Cancer 18 (1), 59-65 40. Petersen, I., Langreck, H., Wolf, G., Schw endel, A., Psille, R., Vogt, P., Reichel, M. B., Ried, T., and Dietel, M. (1997) Br J Cancer 75 (1), 79-86 41. Beghini, A., Magnani, I., Roversi, G., Pie poli, T., Di Terlizzi, S., Moroni, R. F., Pollo, B., Fuhrman Conti, A. M., Cowell, J. K., Finocchiaro, G., and Larizza, L. (2003) Oncogene 22 (17), 2581-2591 42. Fukasawa, K. (2005) Cancer Lett 230 (1), 6-19 43. Hirota, T., Kunitoku, N., Sasayama, T ., Marumoto, T., Zhang, D., Nitta, M., Hatakeyama, K., and Saya, H. (2003) Cell 114 (5), 585-598 44. Littlepage, L. E., Wu, H., Andresson, T., Deanehan, J. K., Amundadottir, L. T., and Ruderman, J. V. (2002) Proc Natl Acad Sci U S A 99 (24), 15440-15445 45. Zhao, Z. S., Lim, J. P., Ng, Y. W., Lim, L., and Manser, E. (2005) Mol Cell 20 (2), 237-249 46. Eyers, P. A., Erikson, E., Chen, L. G., and Maller, J. L. (2003) Curr Biol 13 (8), 691-697 47. Glover, D. M. (2003) Mol Cell 12 (4), 797-799 48. Callow, M. G., Clairvoyant, F., Zhu, S., Schryver, B., Whyte, D. B., Bischoff, J. R., Jallal, B., and Smeal, T. (2002) J Biol Chem 277 (1), 550-558 49. Moniaux, N., Nemos, C., Schmied, B. M., Chauhan, S. C., Deb, S., Morikane, K., Choudhury, A., Vanlith, M., Sutherlin, M., Si kela, J. M., Hollingsworth, M. A., and Batra, S. K. (2006) Oncogene 25 (23), 3247-3257 50. Darwish, H., Cho, J. M., Loignon, M., and Alaoui-Jamali, M. A. (2007) Oncogene 26 (29), 4319-4328 51. Kuuselo, R., Savinainen, K., Azorsa, D. O., Basu, G. D., Karhu, R., Tuzmen, S., Mousses, S., and Kallioniemi, A. (2007) Cancer Res 67 (5), 1943-1949 52. Cheng, J. Q., Ruggeri, B., Klein, W. M ., Sonoda, G., Altomare, D. A., Watson, D. K., and Testa, J. R. (1996) Proc Natl Acad Sci U S A 93 (8), 3636-3641

PAGE 117

10553. Bellacosa, A., de Feo, D., Godwin, A. K., Bell, D. W., Cheng, J. Q., Altomare, D. A., Wan, M., Dubeau, L., Scambia, G., Ma sciullo, V., Ferrandina, G., Benedetti Panici, P., Mancuso, S., Neri, G., and Testa, J. R. (1995) Int J Cancer 64 (4), 280285

PAGE 118

106 CHAPTER V DISCUSSION AND CONCLUSION Aurora kinases represent a nove l family of serine/threonine kinases crucial for cell cycle control. The first Aurora kinase was discovered in Drosophila (1). Because its mutations resulted in a failure of centrosome separation, leading to the formation of monopolar spindles, it was given the name "Aurora," reminiscent of the North Pole. The mammalian Aurora family comprises thr ee related kinases that share the highest degree of sequence homology in their catalytic domains (2, 3). The Aurora-A and -B kinases have emerged as essential regulators of cell division. While Aurora-A is im plicated in regulating mitotic entry, centrosome maturation, and spindle a ssembly; Aurora-B is required for correct chromosome segregation and cytokine sis. For Aurora-C, expression seems to be restricted to normal testicular tissue, the normal function of which is not well documented until recent reports show that Aurora-C is al so a chromosomal passenger protein, and that it binds directly to INCENP and survivin in vitro (4-6). Experimental data suggest that inappropriately high or low levels of Aurora kinase activity are linked to genetic instability. Despite their sequence homology and common association with cycling cells, the subcellular distributi on, partners, and substrates and therefore functions of Aurora-A and -B are essentially nonoverlapping (7, 8). In this work, we focused on Aurora-A kinase. Through study of upstream regulator, downstream target and functional associating partner (F ig. 5.1), we provided new insights into better understa nding of the molecular mechan ism of Aurora-A kinase in human oncogenesis.

PAGE 119

107 Aurora-A alteration is found in variou s human malignancies, at mRNA and/or protein levels more than DNA copy number chan ge (9-11). The discrepancy suggests that Aurora-A overexpression is lik ely to be regulated not only by gene amplification, but also by other mechanisms such as transcriptional or translational activati on and suppression of protein degradation. However, transcripti onal regulation of Auro ra-A during the cell cycle is largely unknown. We demonstrated, for the first time, that transcription factor E2F3 directly binds to Aurora-A promoter and tightly regulates Aurora-A expression during G2/M phase, which provides a mechanism for regulation of Aurora-A at transcription level during the cell cycle. Among the 8 E2F family members identified so far, the transcriptional activators E2F1–3 have been shown to be cell cycle re gulated; E2F4 and E2F5 negatively regulate cell cycle with decisions of cell differen tiation and quiescence; and E2F6-8 are Rbindependent transcriptional repressors. Gene expression microarray analyses have revealed that E2F3 regulates not only DNA rep lication for S phase entry, but also mitotic regulatory factors such as cyclin B1, cyc lin A2 and cdc2 during G2/M progression (1214). Our data show that ectopic expression of E2F3 induces mRNA and protein levels of Aurora-A whereas knockdown of E2F3 decreas es Aurora-A expression and mitotic cell cycle delay. Notably, chromatin immunoprecipita tion reveals that E2F3 binds to AuroraA promoter in vivo and the interaction primarily occurs during G2/M phase. Moreover, we demonstrated frequent co-exi stence of elevated levels of E2F3 and Aurora-A in human primary ovarian carcinoma. Importance of Aurora-A and E2F3 in oncogenesis has been well established by their alterations in human neoplasms and their capacity to induce cell tran sformation (15-18). We obser ved strong association of elevated E2F3 expression and Aurora-A in ovarian tumors underscoring the clinical significance of the E2F3-AuroraA signaling axis. E2F3 could also be one of the major transcriptional factors that contribute to upregulation of Aurora-A in other human primary tumors. Aurora-A, as a mitotic E2F3 target gene, could mediate E2F3 oncogenic function. Thus, E2F3-Aurora-A axis represents an attractive target for cancer therapy.

PAGE 120

108 CHFR (ch eckpoint with f orkhead and r ing finger) has been discovered as an early mitosis checkpoint. Unlike the spindle checkpoi nt, the CHFR checkpoint is not essential for cell proliferation, which can be explained by the high frequency of CHFR inactivation, either by promoter methylat ion or in a few cases by mi ssense mutations. The overall CHFR inactivation rate is ranging from 15 to 50% in various tumor types (19-26), which is much higher than the frequency of in activation of all spindle checkpoint genes combined. It has been shown that CHFR mediates the degrad ation of polo-like kinase 1(Plk1), Aurora-A and CHFR itself thr ough its ring finger domain which possesses ubiquitin ligase activity. CHFR also c ontains a FHA domain, which could bind phosphorylated peptides (19, 27), suggesting th at protein kinases might act upstream of CHFR and regulate its activity. A previous report shows that Ak t/PKB phosphorylates CHFR at Thr-39 in FHA domain after DNA dama ge, which led to inhibition of CHFR E3 activity and shortening G2 arrest induced by DNA damage (28). We demonstrated in this study that CHFR is a substrate of Aurora-A kinase. Aurora-A phosphorylates CHFR at Ser218 and Ser337 and inhibits its E3 activity. We also showed that frequent downregul ation of CHFR and i nverse correlation of expression of Aurora-A and CHFR in huma n ovarian cancer, whic h supports the notion that CHFR is an E3 ligase of Aurora-A (29). However, co-expression of CHFR and elevated Aurora-A was observed in a subset of ovarian cancers, which is consistent with the previous findings that only a fraction of tumors with loss of CHFR expression are due to promoter hypermethylation, suggesting th at Aurora-A might phosphorylate CHFR and inhibit its tumor suppressor function in these tumors. Ta ken together, these findings indicate a feedback regulation loop between Aurora-A and CHFR. Aurora-A exerts its mitotic function thr ough regulation of a number of proteins (30-34), which include histone H3, TACC3, Eg5, CPEB and TPX2. We and others have previously shown Aurora-A phosphorylation of p53 which leads to inactivation of p53 and G2/M cell cycle progression (35). CHFR ha s been shown to mediate a delay in cell cycle progression early in mitosis in respons e to microtubule stress. Identification of

PAGE 121

109 Aurora-A inhibition of CHFR provides addi tional molecular mechanism for Aurora-A function in control of G2 /M cell cycle progression. Aurora-A has well-established but perhaps not yet fully understood roles in centrosome maturation and duplication, mitotic entry and bipolar spindle assembly. By the G2 phase of the cell cycle through anaphase, it can be de tected in the centrosomes, spindle poles and spindle microtubules. Following initial activation by the LIM protein Ajuba in G2, Aurora-A phosphorylates and recruits several microtubule-associated proteins to the centrosome to promote maturation. Cyclin-dependent kinases (CDKs) require not only phosphorylation of the equivalent threonine (Thr160CDK) but also the binding by a partner protein, cyclin A, to be fully activated (36). Aurora-A might also rely on a similar mechanism. After nuclear e nvelope breakdown, Aurora-A is localized to spindle microtubules and fully activated by its interacting non-enzym atic protein TPX2, which plays an as yet not fully defined role in the Ran-spindle assembly process (37, 38). The activation of Aurora-A is al so facilitated by other cofact ors/substrates, with different regulatory mechanism and cellular functions. We identified CNTD2, a novel mammalian B-type cyclin, as a cofactor of Aurora-A. CNTD2 colocalizes and immunoprecip itates with Aurora-A at centrosome. The interaction results in stabilization Au rora-A protein and activation of Aurora-A kinase. Unlike other Aurora-A in teraction proteins TPX2, Bora and Ajuba, which are also substrates of the kinase (34, 39, 40), CNTD2 increases Aurora-A pr otein stability and kinase activity but is no t phosphorylated by Aurora-A. Aurora-A is required for the activation a nd recruitment of cdc2-cyclin B1 to the centrosome. The activated cdc2-cyclin B1 th en commits the cell to mitosis (40). While CNTD2 interacts with Aurora-A and activates Aurora-A and cdc2 kinase activity, it did not form complex with cdc2. Other than the role in Aurora-A regulation, CNTD2 is an oncogene, which is concluded from the fact s that ectopic expression of CNTD2 induces malignant transformation, centrosome am plification and genomic instability. CNTD2 locates to chromosome 19q13 amplicon, which contains several other oncogenes AKT2 PAK4 pancreatic differentiation 2 ( PD2 ), SERTAD3 and

PAGE 122

110 transcriptional regulator intersex-like ( IXL ) (41-45). We have previously shown amplification/overexpre ssion of AKT2 in human ovarian, pancreatic and breast cancers. Inhibition of AKT2 reduces tumor growth and sensitizes cells to chemotherapeutic agentinduced cell death (41, 46, 47) In this report, we demonstrated that CNTD2 is amplified/overexpressed in human ovarian, panc reatic and lung cancers. Overexpression of CNTD2 is associated with late stage/high grade ovarian tumors and accompanied with poor clinic outcome. Therefore, CNTD2 could be a potential therapeutic target or cancer progression/prognosis marker for these malignancies. The molecular mechanism of CNTD2 regulation of Aurora-A protein stab ility and kinase activity remains to be investigated. Fig. 5.1. Working scheme of molecular mechanism of Aurora-A in human oncogenesis. This dissertation has demonstrated that transcripti onal upregulation by E2F3, interaction and activation by CNTD2 of Aurora-A; and Au rora-A feedback regulation of CHFR may contribute to Aurora-A-induced tumorigenesis.

PAGE 123

111 References 1. Glover, D. M., Leibowitz, M. H., McLean, D. A., and Parry, H. (1995) Cell 81 (1), 95-105 2. Hannak, E., Kirkham, M., Hyman, A. A., and Oegema, K. (2001) J Cell Biol 155 (7), 1109-1116 3. Marumoto, T., Honda, S., Hara, T., Nitta, M., Hirota, T., Kohmura, E., and Saya, H. (2003) J Biol Chem 278 (51), 51786-51795 4. Sasai, K., Katayama, H., Stenoien, D. L., Fujii, S., Honda, R., Kimura, M., Okano, Y., Tatsuka, M., Suzuki, F., Nigg, E. A., Earnshaw, W. C., Brinkley, W. R., and Sen, S. (2004) Cell Motil Cytoskeleton 59 (4), 249-263 5. Li, X., Sakashita, G., Matsuzaki, H., Sugimoto, K., Kimura, K., Hanaoka, F., Taniguchi, H., Furukawa, K ., and Urano, T. (2004) J Biol Chem 279 (45), 4720147211 6. Yan, X., Cao, L., Li, Q., Wu, Y., Zhang, H., Saiyin, H., Liu, X., Zhang, X., Shi, Q., and Yu, L. (2005) Genes Cells 10 (6), 617-626 7. Carmena, M., and Earnshaw, W. C. (2003) Nat Rev Mol Cell Biol 4 (11), 842-854 8. Keen, N., and Taylor, S. (2004) Nat Rev Cancer 4 (12), 927-936 9. Gritsko, T. M., Coppola, D., Paciga, J. E., Yang, L., Sun, M., Shelley, S. A., Fiorica, J. V., Nicosia, S. V., and Cheng, J. Q. (2003) Clin Cancer Res 9 (4), 14201426 10. Zhou, H., Kuang, J., Zhong, L., Kuo, W. L ., Gray, J. W., Sahin, A., Brinkley, B. R., and Sen, S. (1998) Nat Genet 20 (2), 189-193 11. Sakakura, C., Hagiwara, A., Yasuoka, R., Fujita, Y., Nakanishi, M., Masuda, K., Shimomura, K., Nakamura, Y., Inazawa, J., Abe, T., and Yamagishi, H. (2001) Br J Cancer 84 (6), 824-831 12. Ishida, S., Huang, E., Zuzan, H., Spang, R., Leone, G., West, M., and Nevins, J. R. (2001) Mol Cell Biol 21 (14), 4684-4699 13. Black, E. P., Hallstrom, T., Dressman, H. K., West, M., and Nevins, J. R. (2005) Proc Natl Acad Sci U S A 102 (44), 15948-15953

PAGE 124

112 14. Zhu, W., Giangrande, P. H., and Nevins, J. R. (2004) EMBO J 23 (23), 4615-4626 15. Feber, A., Clark, J., Goodwin, G., Dods on, A. R., Smith, P. H., Fletcher, A., Edwards, S., Flohr, P., Falconer, A., Roe, T., Kovacs, G., Dennis, N., Fisher, C., Wooster, R., Huddart, R., Foster, C. S., and Cooper, C. S. (2004) Oncogene 23 (8), 1627-1630 16. Foster, C. S., Falconer, A., Dodson, A. R., Norman, A. R., Dennis, N., Fletcher, A., Southgate, C., Dowe, A., Dearnaley, D., Jhavar, S., Eeles, R., Feber, A., and Cooper, C. S. (2004) Oncogene 23 (35), 5871-5879 17. Cooper, C. S., Nicholson, A. G., Foster, C., Dodson, A., Edwards, S., Fletcher, A., Roe, T., Clark, J., Joshi, A., Norman, A., Feber, A., Lin, D., Gao, Y., Shipley, J., and Cheng, S. J. (2006) Lung Cancer 54 (2), 155-162 18. Xu, G., Livingston, D. M., and Krek, W. (1995) Proc Natl Acad Sci U S A 92 (5), 1357-1361 19. Mizuno, K., Osada, H., Konishi, H., Ta tematsu, Y., Yatabe, Y., Mitsudomi, T., Fujii, Y., and Takahashi, T. (2002) Oncogene 21 (15), 2328-2333 20. Shibata, Y., Haruki, N., Kuwabara, Y ., Ishiguro, H., Shinoda, N., Sato, A., Kimura, M., Koyama, H., Toyama, T., Ni shiwaki, T., Kudo, J., Terashita, Y., Konishi, S., Sugiura, H., and Fujii, Y. (2002) Carcinogenesis 23 (10), 1695-1699 21. Bertholon, J., Wang, Q., Falett e, N., Verny, C., Auclair, J., Chassot, C., Navarro, C., Saurin, J. C., and Puisieux, A. (2003) Oncogene 22 (55), 8956-8960 22. Corn, P. G., Summers, M. K., Fogt, F., Vi rmani, A. K., Gazdar A. F., Halazonetis, T. D., and El-Deiry, W. S. (2003) Carcinogenesis 24 (1), 47-51 23. Mariatos, G., Bothos, J., Zacharatos, P., Summers, M. K., Scolnick, D. M., Kittas, C., Halazonetis, T. D., and Gorgoulis, V. G. (2003) Cancer Res 63 (21), 71857189 24. Satoh, A., Toyota, M., Itoh, F., Sasaki, Y ., Suzuki, H., Ogi, K., Kikuchi, T., Mita, H., Yamashita, T., Kojima, T., Kusano, M., Fujita, M., Hosokawa, M., Endo, T., Tokino, T., and Imai, K. (2003) Cancer Res 63 (24), 8606-8613

PAGE 125

113 25. Toyota, M., Sasaki, Y., Satoh, A., Ogi, K., Kikuchi, T., Suzuki, H., Mita, H., Tanaka, N., Itoh, F., Issa, J. P., Jair, K. W., Schuebel, K. E., Imai, K., and Tokino, T. (2003) Proc Natl Acad Sci U S A 100 (13), 7818-7823 26. Erson, A. E., and Petty, E. M. (2004) Mol Carcinog 39 (1), 26-33 27. Natrajan, R., Williams, R. D., Hing, S. N ., Mackay, A., Reis-Filho, J. S., Fenwick, K., Iravani, M., Valgeirsson, H., Grigor iadis, A., Langford, C. F., Dovey, O., Gregory, S. G., Weber, B. L., Ashwort h, A., Grundy, P. E., Pritchard-Jones, K., and Jones, C. (2006) J Pathol 210 (1), 49-58 28. Shtivelman, E. (2003) Mol Cancer Res 1 (13), 959-969 29. Yu, X., Minter-Dykhouse, K., Malureanu, L ., Zhao, W. M., Zhang, D., Merkle, C. J., Ward, I. M., Saya, H., Fang, G., van Deursen, J., and Chen, J. (2005) Nat Genet 37 (4), 401-406 30. Crosio, C., Fimia, G. M., Loury, R., Kimura, M., Okano, Y., Zhou, H., Sen, S., Allis, C. D., and Sassone-Corsi, P. (2002) Mol Cell Biol 22 (3), 874-885 31. Kinoshita, K., Noetzel, T. L., Pelletier L., Mechtler, K., Drechsel, D. N., Schwager, A., Lee, M., Raff, J. W., and Hyman, A. A. (2005) J Cell Biol 170 (7), 1047-1055 32. Giet, R., Uzbekov, R., Cubizolles, F., Le Guellec, K., and Prigent, C. (1999) J Biol Chem 274 (21), 15005-15013 33. Ma, C., Cummings, C., and Liu, X. J. (2003) Mol Cell Biol 23 (5), 1703-1716 34. Kufer, T. A., Sillje, H. H., Korner, R., Gruss, O. J., Meraldi, P., and Nigg, E. A. (2002) J Cell Biol 158 (4), 617-623 35. Liu, Q., Kaneko, S., Yang, L., Feldman, R. I., Nicosia, S. V., Chen, J., and Cheng, J. Q. (2004) J Biol Chem 279 (50), 52175-52182 36. Pavletich, N. P. (1999) Journal of molecular biology 287 (5), 821-828 37. Fu, J., Bian, M., Jiang, Q., and Zhang, C. (2007) Mol Cancer Res 5 (1), 1-10 38. Andrews, P. D. (2005) Oncogene 24 (32), 5005-5015 39. Hutterer, A., Berdnik, D., Wirtz-Peitz F., Zigman, M., Schleiffer, A., and Knoblich, J. A. (2006) Dev Cell 11 (2), 147-157

PAGE 126

114 40. Hirota, T., Kunitoku, N., Sasayama, T ., Marumoto, T., Zhang, D., Nitta, M., Hatakeyama, K., and Saya, H. (2003) Cell 114 (5), 585-598 41. Cheng, J. Q., Godwin, A. K., Bellacosa, A., Taguchi, T., Franke, T. F., Hamilton, T. C., Tsichlis, P. N., and Testa, J. R. (1992) Proc Natl Acad Sci U S A 89 (19), 9267-9271 42. Callow, M. G., Clairvoyant, F., Zhu, S., Schryver, B., Whyte, D. B., Bischoff, J. R., Jallal, B., and Smeal, T. (2002) J Biol Chem 277 (1), 550-558 43. Moniaux, N., Nemos, C., Schmied, B. M., Chauhan, S. C., Deb, S., Morikane, K., Choudhury, A., Vanlith, M., Sutherlin, M., Si kela, J. M., Hollingsworth, M. A., and Batra, S. K. (2006) Oncogene 25 (23), 3247-3257 44. Darwish, H., Cho, J. M., Loignon, M., and Alaoui-Jamali, M. A. (2007) Oncogene 26 (29), 4319-4328 45. Kuuselo, R., Savinainen, K., Azorsa, D. O., Basu, G. D., Karhu, R., Tuzmen, S., Mousses, S., and Kallioniemi, A. (2007) Cancer Res 67 (5), 1943-1949 46. Cheng, J. Q., Ruggeri, B., Klein, W. M ., Sonoda, G., Altomare, D. A., Watson, D. K., and Testa, J. R. (1996) Proc Natl Acad Sci U S A 93 (8), 3636-3641 47. Bellacosa, A., de Feo, D., Godwin, A. K., Bell, D. W., Cheng, J. Q., Altomare, D. A., Wan, M., Dubeau, L., Scambia, G., Ma sciullo, V., Ferrandina, G., Benedetti Panici, P., Mancuso, S., Neri, G., and Testa, J. R. (1995) Int J Cancer 64 (4), 280285

PAGE 127

115 APPENDICES

PAGE 128

116 Appendix I List of Publications Chapter II of this dissertation has been submitted to Journal of Biological Chemistry, 2008. Chapter III and IV of this dissert ation are in manuscript preparation. Other publications: Cheng GZ, Park S, Shu S, He L Kong W, Zhang W, Yuan Z, Wang LH, Cheng JQ. Advances of AKT pathway in human oncogene sis and as a target for anti-cancer drug discovery. Curr Cancer Drug Targets 2008 Feb;8(1):2-6. Yang H, Kong WW, He L Zhao JJ, O'Donnell J Wang JW, Wenham RM, Coppola D, Kruk PA, Nicosia SV, Cheng JQ. MicroRNA expression profiling in human ovarian cancer: miR-214 induces cell survival and cisplatin resistance by targeting PTEN. Cancer Res. 2008 Jan 15;68(2):425-33. Yang H, He L Kruk PA, Nicosia SV, Cheng JQ. Au rora-A induces cell survival and chemoresistance by activation of Akt through a p53-dependent manner in ovarian cancer cells. Int J Cancer 2006 Nov 15;119(10):2304-12. Kim D, Dan HC, Park S, Yang L, Liu Q, Kaneko S, Ning J, He L Yang H, Sun M, Nicosia SV, Cheng JQ. AKT/PKB signaling mech anisms in cancer and chemoresistance. Front Biosci 2005 Jan 1;10:975-87.

PAGE 129

ABOUT THE AUTHOR Lili He received her Bachelor’s Degree at Be ijing Institute of Technology, China in 2003. Ms He entered the Ph.D. program at University of South Florida Colle ge of Medicine in April 2004, and received her Master Degree of Science at USF in 2006. During her graduate study, Ms He involved in a range of research projects and published several papers in peer-reviewed scientific journals. She is a associate member of American Association of Cancer Research. She is also an active member of AMSGS (Association of Medical Science Graduate Students) at USF.


xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 002021681
003 fts
005 20090731122848.0
006 m||||e|||d||||||||
007 cr mnu|||uuuuu
008 090731s2008 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0002582
035
(OCoLC)428816627
040
FHM
c FHM
049
FHMM
090
RB37 (Online)
1 100
He, Lili.
0 245
Molecular mechanism of Aurora-A kinase in human oncogenesis
h [electronic resource] /
by Lili He.
260
[Tampa, Fla] :
b University of South Florida,
2008.
500
Title from PDF of title page.
Document formatted into pages; contains 116 pages.
Includes vita.
502
Dissertation (Ph.D.)--University of South Florida, 2008.
504
Includes bibliographical references.
516
Text (Electronic dissertation) in PDF format.
520
ABSTRACT: Aurora-A is a mitotic kinase, which regulates cell cycle progression through modulating centrosome function. Aurora-A expression is frequently altered in human malignancies. The discrepancy between overexpression and amplification suggests that elevated Aurora-A level is likely to be regulated also by transcriptional and/or translational modifications. In this study, we have demonstrated: 1) transcriptional regulation of Aurora-A by E2F3; 2) feedback regulation between tumor suppressor CHFR and Aurora-A; 3) CNTD2 as a novel Aurora-A partner and oncogene to activate Aurora-A and induce transformation. Aurora-A expression and activity are cell cycle regulated. The mechanism of Aurora-A upregulation at onset of mitosis is largely unknown. We demonstrated, for the first time, that transcription factor E2F3 directly binds to Aurora-A promoter and tightly regulates Aurora-A expression during G2/M phase.Moreover, expression of E2F3 considerably correlates with Aurora-A level in human ovarian cancer, indicates that E2F3 is a causal factor for Aurora-A overexpression. Thus, E2F3-Aurora-A axis could be an important target for cancer intervention. The frequent inactivation of prophase checkpoint CHFR caused by DNA methylation or mutation has been reported in human cancers. We showed that CHFR is hypermethylated in ovarian carcinoma. Aurora-A phosphorylates CHFR on Ser-218 and Ser-337 in vivo and in vitro, which inhibits CHFR ubiquitin ligase activity. The feedback regulation loop between Aurora-A and CHFR could play a critical role in regulation of cell cycle progression, imbalance of which may contribute to human oncogenesis. Using yeast 2-hybrid screening, we identified a splicing form of CNTD2 as Aurora-A interacting protein. CNTD2 locates to chromosome 19q13.2 AKT2 amplicon.CNTD2 is amplified and overexpressed in human ovarian, pancreatic and lung cancer cell lines and primary tumors. CNTD2 colocalizes and interacts with Aurora-A in the centrosome. CNTD2 expression induces Aurora-A and cdc2 kinase activity, G2/M progression, and malignant transformation. These data indicate that CNTD2 is an oncogene and could play a pivotal role in Aurora-A activation during the cell cycle and that disruption of CNTD2-Aurora-A axis may represent a potential means to antitumor drug discovery.
538
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
590
Advisor: Jin Q. Cheng, Ph.D.
653
E2F3
CHFR
CNTD2
Centrosome
Cell cycle
690
Dissertations, Academic
z USF
x Pathology and Cell Biology
Doctoral.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.2582