p21-ACTIVATED KINASE: A NOVEL THERAPUETIC TARGET IN THYROID CANCER
DISSERTATION
Presented in Partial Fulfillment of the Requirements for
the Degree Doctor of Philosophy in the Graduate
School of The Ohio State University
By
Leonardo M. Porchia, M.S.
* * * * *
The Ohio State University 2007
Dissertation Committee:
Professor Matthew D. Ringel, Advisor Approved by Professor Robert Brueggemeier
Professor Ching-Shih Chen
Professor Lawrence S. Kirschner Advisor Ohio State Biochemistry Program
ABSTRACT
Follicular cell derived thyroid cancer (i.e. follicular, papillary and anaplastic thyroid
cancer) is the most common endocrine malignancy. While patients with diagnosed with
early stage disease have an excellent prognosis, patients with invasive or metastatic
thyroid cancer have poor survival rates. Because progressive thyroid cancer is
unresponsive to chemotherapy, there is a critical need to identify novel therapeutic
targets. Genetic alterations that result in enhanced activation of the RAS-RAF-MEK and
PI3K-AKT pathways occur in more than 50% of papillary (PTC), follicular (FTC), and
anaplastic (ATC) thyroid cancers. However, the key regulators of thyroid cancer invasion and metastases are less certain. Many lines of evidence suggest important roles for PI3K signaling and the process of epithelial-to-mesenchymal transition (EMT) in thyroid cancer progression. Thus, we are working to develop inhibitors of these pathways for thyroid cancer.
OSU-03012 (OSU) is a celecoxib derivative that was optimized to inhibit PDK-1, a key signaling kinase in the PI3K cascade. NPA (papillary), WRO (follicular), and ARO
(anaplastic) thyroid cancer cell lines were used to study the effects of OSU on thyroid cancer cells in vitro. OSU inhibited proliferation and induced cytotoxicity at doses sufficient to inhibit PDK-1- mediated AKT phosphorylation. Unexpectedly, OSU inhibited cell motility in NPA cells at doses below its IC50 for PDK-1 and below those
ii sufficient to reduce AKT phosphorylation, suggesting that that the inhibition of migration
by OSU might be due to another mechanism. p21 activated kinases (PAK) are master
regulators of cell motility and EMT that are regulated both by PDK1 and other signaling
cascades. Phospho-PAK levels were reduced with 1 μM of OSU in motile NPA cells and
WRO cells. PAK-dependent phosphorylation of vimentin, a key regulator of thyroid cell
EMT, was decreased at similar doses, consistent with reduced activity in these cells.
Subsequent in vitro kinase assays demonstrated that OSU competitively inhibits ATP binding to PAK. Finally, overexpression of constitutively active-PAK1 rescued the anti- migratory effects of OSU. These data demonstrate that PAK is a novel target of and that
OSU might be a therapeutic option for tumors with PAK-dependent invasion and motility.
Because of these findings, and because prior work by our group suggested that signaling through PAK might be dysregulated in thyroid cancer invasion, we undertook a study to determine if PAK expression or PAK activity were altered in thyroid cancer.
Protein and total RNA was isolated from ten PTC tissue samples and from the normal tissue in the opposite lobes for Western blot and quantitative RT-PCR of all PAK isoforms. Of the six isoforms, PAK2 and PAK4 mRNA and protein expression levels were increased in the majority of PTCs. In addition, the levels of phosphorylated PAK were higher in 8 of 10 PTCs. Finally, molecular inhibition of PAK using cDNAs designed to inhibition PAKs1, 2, and 3 confirmed that thyroid cancer cells display PAK- dependent cell migration. Taken together, these data support a role for PAK activity in thyroid cancer and suggest it might be an appropriate therapeutic target for this disease in patients.
iii The most common mutation of PTC is BRAF V600E. Expression of this mutant
BRAF induces thyroid cell invasion and is associated with a more aggressive form of
PTC clinically. PAK is known to functionally phosphorylate C-RAF, a closely related
RAF family protein member, at serine 338. Nine potential PAK phosphorylation sites were found in the B-RAF protein sequence. One site aligned to serine 338 in C-RAF, suggesting this site might represent a PAK phosphorylation site on BRAF. Subsequent experiments demonstrated that B-RAF and PAK co-immunoprecipitate, suggesting they interact directly. This interaction was identified both in overexpression and endogenous systems. Further studies are planned to clarify if the BRAF-PAK interactions are functional and to identify specific phosphorylation sites
Histone deacetylase inhibitors (HDACi) are a class of agents with broad cellular effects. HDAC42 is a novel HDACi that has been shown to be cytotoxic in a number of cell systems. This agent has also been shown to regulate expression and function of signaling molecules, including AKT. HDAC42 inhibited thyroid cancer cell line proliferation, and was cytotoxic at 48 hours. Distinct from OSU, this compound did not appear to inhibit migration. Treatment with HDAC42 increased acetyl-histone 4 at 24 hours and the level of AKT expression and activation were inhibited. However, while levels of PAK protein were reduced after 48 hours of exposure; levels of phospho-PAK at
24 hours were unaltered and increased at 48 hours. Based on these results, we hypothesize that the lack of PAK inhibition may be responsible for the inability of
HDAC42 to block migration, suggesting that the combination of PAK and HDAC inhibitors could be a novel treatment for metastatic cancer.
iv In summary, these results are the first to demonstrate that PAK is a novel target of
OSU, and its inhibition is at least partially responsible for in vitro effects on migration.
They are also the first to determine that PTCs are characterized by overexpression of
PAKs 2 and -4, which PAK activity is frequently enhanced in PTC and is functionally involved in PTC cell biology in vitro. The additional finding that BRAF and PAK physically interact will lead to new experiments determining if there is an important role for interactions between these signaling molecules in PTC biology. Finally, the finding
that HDAC42 increases phospho-PAK levels suggest that combination therapy with this
OSU and this compound may be a rationale approach to thyroid cancer therapy. These
results suggest that PAK may represent a novel therapeutic target for thyroid cancer, and
that compounds based on OSU may be optimized to inhibit this pathway.
v
Dedicated to my mother and father and family
vi
ACKNOWLEDGMENTS
• I would like to acknowledge Dr. Ringel for his guidance, immense understanding
and support, and providing me the ability to work in his lab.
• I would also like to acknowledge Dr. Chen for his guidance and support and
originally giving the opportunity to study signal transduction and get a better
prospective in a drug discovery lab.
• Dr. Jiuxiang Zhu, whose help with planning experiments and analyzing data and
leading me in the beginning.
• Colleagues and friends: Dr. Sam Kulp, Dr. Motoyasu Saji, Dr. Allan Espinosa,
Marcy Geurra, Dr. Motoo Shinohara, Dr. Jun Yea Chung, Yu-Chieh Wang, Ho-
Pi Lin, Ya-ting Wang, for your help with experiments, analyzing data, and
making lab an enjoyable experience.
• Yunlong Zhang for his help with the computer modeling.
• Susanna Pearce for helping edit and revising my writings.
vii
VITA
1998 – 2002 B.S Biochemistry Otterbein College, Westerville Ohio
2002 – 2006 M.S. Biochemistry The Ohio State University, Columbus, Ohio
2002 – Present Graduate Teaching and Research Associate College of Chemistry, The Ohio State University College of Pharmacy, The Ohio State University College of Medicine, The Ohio State University
PUBLICATIONS
1. Porchia LM, Guerra M, Wang YC, Zhang Y, Espinosa AV, Shinohara M, Kulp SK, Kirschner LS, Saji M, Chen CS, Ringel MD. “OSU03012, A Celecoxib Derivative, Directly Targets p21 Activated Kinase.” Mol Pharmacol. 2007 Aug 2; [Epub] 2. Espinosa AV, Porchia L and Ringel MD. “Targeting BRAF in thyroid cancer.” British Journal of Cancer. 2007 Jan 15:96(1): 16-20.
FIELDS OF STUDY
Major Field: Biochemistry
viii
TABLE OF CONTENTS
Page Abstract...... ii
Dedication...... iii
Acknowledgments...... vii
Vita...... viii
List of Tables...... xii
List of Figure...... xiii
Abbreviations...... xv
Chapter 1 – Introduction...... 1 1.1 Normal Thyroid Function...... 2 1.2 Thyroid Cancer ...... 3 1.2.1 Oncogenic Causes of Thyroid Cancer...... 5 1.2.1.1 RET/PTC and B-RAF Alteration Lead to the Development of PTC...... 5 1.2.1.1.1 RET Activity Causes PTC...... 6 1.2.1.1.2 Mutations in B-RAF Lead to PTC...... 7 1.2.1.2 AKT Activity and PPARγ/PAX8 Fusion Proteins Cause FTC...... 8 1.2.1.2.1 Increase AKT Activity Leads to FTC...... 8 1.2.1.2.2 PAX8/PPARγ Leads to FTC Development...... 10 1.2.1.3 The Development of Anaplastic Thyroid Cancer...... 11 1.2.2 Thyroid Cancer Staging and Treatment Option...... 12 1.3 AKT Signaling Pathway...... 13 1.3.1 Mechanism of Activation...... 14 1.3.1.1 PI3K Coverts PIP2 into PIP3...... 14 1.3.1.2 PDK-1 Regulates AKT Activity...... 14 1.3.1.3 AKT Controls Cell Proliferation and Motility...... 15 1.3.2 AKT Isoforms and Functions...... 16 1.3.2.1 AKT1 Regulates Cell Growth and Angiogenesis...... 17
ix 1.3.2.2 AKT2 Regulates Glucose Metabolism...... 17 1.3.2.3 AKT3 Regulates Cell Size and Proliferation...... 18 1.3.3 Target and Function of the AKT pathway...... 18 1.3.3.1 Up-Regulation of AKT Increase Cell Proliferation...... 18 1.3.3.2 AKT Prevents Apoptosis...... 19 1.3.3.3 AKT Regulates Cell Motility...... 21 1.4 RAS/RAF Signaling Pathway...... 22 1.4.1 RAF Mechanism of Activation...... 22 1.4.2 RAF Isoforms Share Sequence Homology and Similar Effectors...... 25 1.4.2.1 C-RAF...... 26 1.4.2.2 B-RAF...... 27 1.4.2.3 A-RAF...... 28 1.4.3 B-RAF Regulates Proliferation, Apoptosis and Cell Motility...... 28 1.5 Steps Required For Cell Motility...... 30 1.5.1 Five Steps for Cell Motility...... 31 1.6 p21-Activated Kinase Signaling Pathway: A Key Regulator of Cell Motility...... 34 1.6.1 Mechanism for p21-Activated Kinase Activation...... 36 1.6.1.1 Key Phosphorylation Sites and Phosphatases...... 37 1.6.1.2 GTPase (RAC1 or CDC42)–Dependent Activation of PAK...... 38 1.6.1.3 GTPase-Independent Activation of PAK...... 39 1.6.2 Isoforms of PAK – Functions and Key Characteristics...... 41 1.6.2.1 PAK1...... 41 1.6.2.2 PAK2...... 42 1.6.2.3 PAK3...... 42 1.6.2.4 PAK4...... 43 1.6.2.5 PAK5...... 44 1.6.2.6 PAK6...... 44 1.6.3 Targets and Function of PAK Proteins...... 45 1.6.3.1 PAK Increase Cell Proliferation...... 45 1.6.3.2 PAK Decreases Apoptosis Through Multiple Mechanisms...... 46 1.6.3.3 PAKs are an Integral Protein in Cell Motility...... 48 1.7 Conclusion...... 50
Chapter 2 - OSU-03012, a Celecoxib Derivative, Directly Targets p21-Activated Kinase...... 60 2.1 Introduction...... 60 2.2 Materials and Methods...... 62 2.3 Results...... 68
x 2.3.1 OSU-03012 Inhibits Thyroid Cancer Cell Proliferation and Induces Apoptosis...... 68 2.3.2 OSU-03012 Inhibits AKT and PAK Phosphorylation in Thyroid Cancer Cells...... 69 2.3.3 OSU-03012 Directly Inhibits PAK Activity via Competitive Inhibition of ATP Binding...... 70 2.3.4 OSU-03012 Inhibits Thyroid Cancer Cell Migration in a PAK-Dependent Manner...... 72 2.4 Discussion...... 73
Chapter 3 - p21-Activated Kinases are Upregulated in Human Papillary Thyroid Cancer...... 83 3.1 Introduction...... 83 3.2 Methods and Materials...... 87 3.3 Results...... 90 3.3.1 PAK Isoform Specific Primer Design...... 90 3.3.2 PAK Expression in Thyroid Cancer Cell Lines...... 91 3.3.3 Dominant Negative PAK Decrease NPA Cell Motility...... 92 3.3.4 PAK Expression in Human Papillary Thyroid Tumors...... 93 3.4 Discussion...... 94
Chapter 4 - p21-Activated Kinase Interacts with B-RAF...... 102 4.1 Introduction...... 102 4.2 Materials and Methods...... 105 4.3 Results...... 107 4.3.1 Identification of Possible PAK Phosphorylation Sites on B-RAF...... 107 4.3.2 PAK and B-RAF Co-Precipitate...... 108 4.4 Discussion...... 109
Chapter 5 - HDAC42 Decreases Thyroid Cancer Proliferation...... 116 5.1 Introduction...... 116 5.2 Materials and Methods...... 121 5.3 Results...... 124 5.3.1 Effect of Histone Deacetylase Inhibitor (HDAC42) on Thyroid Cancer Cell Viability...... 124 5.3.2 HDAC42 Regulates AKT and PAK1 Protein Expression...... 125 5.3.3 HDAC42 has a Modest Affect on Thyroid Cancer Cell Migration in vitro...... 127 5.4 Discussion...... 127
Chapter 6 - Conclusion and Future Directions...... 135 Bibliography...... 140
xi
LIST OF TABLES
Table Page
1.1 Thyroid Cancer Statistics...... 53
1.2 Effectors of PAK...... 58
3.1 PAK isoform specific primer design...... 97
3.2 PAK isoform gene and protein expression in Human thyroid cancer cell lines...... 99
3.3 Summary of PAK isoform gene and protein expression in Human Thyroid Papillary carcinomas...... 101
xii
LIST OF FIGURES
Figure Page
1.1 - Structure and function of the thyroid...... 52
1.2 - PI3K/PDK-1/AKT signaling pathway...... 54
1.3 - Activation of RAF kinase...... 55
1.4 - 5 Steps for cell motility...... 56
1.5 - PAK isoforms and method of activation...... 57
1.6 - PAK interacts with the PI3K/PDK-1/AKT signaling pathway and RAS/RAF signaling pathway...... 59
2.1 - OSU-03012 decreases cell viability and increases cells in S-phase...... 77
2.2 - OSU-03012 decreases AKT and PAK phosphorylation...... 78
2.3 - OSU-03012 directly inhibits PAK through direct competition with ATP...... 79
2.4 - Molecular modeling of potential OSU-03012 interactions with PAK 1...... 80
2.5 - OSU-03012 inhibition of NPA cell motility is partially PAK-dependent...... 81
2.6 – OSU-03012 inhibits PDK-1 and PAK decreasing cell motility...... 82
3.1 – PAK primer specificity and selectivity...... 98
3.2 – p21-inhibitory domain (PID) decreases NPA cell migration...... 100
4.1 – PAK interacts with B-RAF...... 112
4.2 - Possible PAK phosphorylation sites for B-RAF...... 113
4.3 - Comparison of RAF isoforms...... 114
xiii 4.4 - PAK and B-RAF co-immunoprecipitate together...... 115
5.1 - Structure of HDAC42...... 130
5.2 - HDAC42 is cytotoxic in thyroid cancer cell lines...... 131
5.3 - HDAC42 induces apoptosis in thyroid cancer cell lines...... 132
5.4 - HDAC42 causes an increase in the acetylation of Histone 4, and alters both AKT and PAK protein expression...... 133
5.5 - HDAC42 does not inhibit NPA cell migration...... 134
xiv
ABBREVIATIONS
4E-BP1 4E binding protein 1 ACAP1 ADP-ribosylation factor directed GTPase-activating protein ADF Actin depolymerizing factor ADP Adenosine 5'-diphosphate AID Autoinhibitory Domain AKAP9-B- AKAP9 and B-RAF fusion protein RAF AKT Protein Kinase B Apaf-1 Apoptotic protease activating factor-1 APE AKT phosphorylation enhancer AR Androgen Receptor arp2/3 actin-related protein-2/3 Asp asparatic acid ATC Anaplastic Thyroid Cancer ATP 5'-adenosine triphosphate BAD Bcl-XL/Bcl-2-associated death promoter BAX BCL-2-associated X protein Bcl-2 B-cell lymphoma leukemia-2 Bcl-xL bcl-xLong BID BH3-interacting domain death agonist BIM Bcl-2 interacting mediator of cell death BTK Bruton tyrosine kinase CA-PAK Constitutively Active p21-Activated Kinase CAPZ beta-actinin, capping protein Caspase cysteine-dependent aspartate-directed protease CDC42 Cell division cycle 42 CDK Cyclin Dependent Kinases cDNA Complementary deoxyribonucleic acid COX-2 cyclooxygenase II CR1/-2/-3 Conserved region 1/ conserved region 2/conserved region 3 CRD Cysteine-rich Domain c-SIS cellular sis gene CTMP Carboxyl-terminal modulator protein cys/his Cysteine/histine DLC1 dynein light chain 1 DMSO dimethylsulfoxide
xv DNA Deoxyribonucleic Acid DNase I Deoxyribonuclease I EGFR Epidermal growth factors receptor ELISA Enzyme Linked Immuno Sorbant Assay EMT Epithelial-mesenchymal transition/transformation ER Estrogen Receptor ERK Extracellular signal Regulated Kinase FAK Focal Adhesion Kinase FAS Apoptosis Stimulating Fragment (programmed cell death) FASL Fas ligand FKHR forkhead transcription factors FTC Follicular Thyroid Cancer FYVE Fab1p, YOTB, Vac1p and EEA1 GAPDH Glyceraldehyde phosphate dehydrogenase GDP Guanosine 5'-diphosphate GFP Green Fluorescence Protein Grb-2 Growth Factor Receptor-Bound protein 2 GSK3 Glycogen Synthase Kinase 3 GTP Guanosine 5'-triphosphate GTPase Guanosine triphosphatase HAT Histone acetyl transferase HDAC histone deacetylase HEK293 Human embryonic kidney 293 HER2/nue Human epidermal growth factor receptor 2/ ErbB-2 H-RAS Harvey ras HSP90 Heat shock protein 90 I κB inhibitory kappa B IAP-1/2 Inhibitor of apoptosis protein-1/2 IgG Immunoglobulin G IKK I kappa B kinase ILK Integrin Linked Kinase JAK Janus Kinase K229R Kinase-dead p21-activated kinase 1 K-RAS Kirsten ras LIMK LIM kinase MAPK mitogen-activated protein kinase MBP Myelin Basic Protein MDM murine double minute clone 2 MEK MAP kinase/ERK kinase or MAP kinase kinase MLC Myosin Light Chain MLCK Myosin Light Chain Kinase MMP Matrix metalloproteinase mTOR Mammalian Target of Rapamycin 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-2H-tetrazolium MTT bromide
xvi NAD+ nicotinamide adenine dinucleotide NCBI National Center for Biotechnology Information NF-κB Nuclear Factor kappa B NLS Nuclear Localization Signal Op18 Oncoprotein 18 p110 PI3K (kinase domain) p21CIP/WAF1 cyclin-dependent kinase inhibitor p21 p27KIP1 p27Kip1 cell cycle inhibitor p35/CDK5 cyclin-dependent kinase regulator p35 p47 PHOX NADPH oxidase, p47 p70S6K p70 ribosomal S6 kinase p85 regulatory subunit of PI3K p90RSK p90 ribosomal S6 kinase PAK p21-activated kinase PAX8 paired-box-containing 8 PBD p21-binding domain PCR Polymerase Chain Reaction PDGFR Platelet Derived Growth Factor Receptor PDK-1 3'-phosphoinositide-dependent kinase-1 PDK-2 3'-phosphoinositide-dependent kinase-2 pEGFP enhanced green fluorescent protein expressing plasmid PH-Domain Pleckstrin Homology domain PHLPP PH domain leucine-rich repeat protein phosphatase PI3K Phosphoinositide 3 kinase PI3KCA Constitutively Active PI3K PID PAK autoinhibitory Domain PIP2 Phosphatidylinositol-4,5-bisphosphate PIP3 Phosphatidylinositol-3,4,5-trisphosphate PIX PAK-interacting exchange factor PKA Protein Kinase A PKC Protein Kinase C POPX1/2 partner of PIX1/2 PP1 Protein Phosphatase 1 PP2A Protein Phosphatase 2A Peroxisome-proliferator-activated receptors gamma/paired-box- PPARγ/PAX8 containing 8 fusion protein PPARγ Peroxisome-proliferator-activated receptors gamma PRK1/2 protein kinase C-related protein kinases-1/2 PTC Papillary Thyroid Cancer Phosphatase and Tensin Homologue Deleted on Chromosome PTEN Ten qRT-PCR quantitative real-time PCR RAF rapidly growing fibrosarcomas RAP RAS-related protein RAS rat sarcoma virus.
xvii Rb Retinoblastoma RBD RAS-binding domain RET receptor tyrosine kinase RET chimeric oncoproteins found in human papillary thyroid RET/PTC carcinomas RHO Ras-homologous RNA ribonucleic acid RNase A Ribonuclease A ROCK Rho-associated coiled coil-containing protein kinase RPMI-1640 Roswell Park Memorial Institute-1640 SAHA Suberoylanilide hydroxamic acid SGK Serum and glucocorticoid-inducible kinase SH3 Src Homology 3 SIR Silent Information Regulator siRNA small interfering RNA Sirt sirtuin SOS Son of Sevenless SRC sarcoma virus STAT3 signal transducer and activator of transciption-3 T3 triiodothyronine T4 Thyroxine TGFβ Transforming Growth Factor Beta TNM system Tumor Node Metastasis staging system TSA Trichostatin A TSC Tuberous sclerosis complex V600E B-RAF with activate mutation at valine 600 to glutamate VEGFR Vascular Endothelial Growth Factor Receptors VPA Valproic Acid WASP Wiscott Aldrich Syndrome protein
xviii
CHAPTER 1
INTRODUCTION
The purpose of this introductory section is to give a summary of thyroid function, and the oncogenic alterations that are associated with certain forms of thyroid cancer. For example, papillary thyroid cancers are typically associated with RET/PTC rearrangements and B-RAF mutations. Conversely, follicular thyroid cancers are associated with decreased Phosphatase and tensin homologue deleted on chromosome 10
(PTEN) activity, PPARγ/PAX8 rearrangements, and RAS mutations. Even though different mutations lead to different forms of thyroid cancer, they activate similar pathways in cell motility. The mechanisms for cell motility will be discussed in detailed later, focusing on p21-activated kinases (PAK) as key proteins involved in cell motility.
Furthermore, data reviewing the broader role of PAKs will also be examined.
To date, an estimated 10.5 million Americans have cancer or are cancer survivors.
For 2007 alone, the American Cancer Society estimates that there will be 1,444,920 new cases of, and 559,650 deaths, from cancer, making it one of the leading causes of death in
America (3). There are many different forms of cancer (breast, prostate, lung, ovarian, thyroid, etc.) some of which are treatable and some of which are not, depending on the stage the disease and form of the disease (3). One of the more common factors, leading to
1 a poor prognosis for nearly all solid tumors, is the development of distant metastasis.
When a tumor has metastasized, the cancer cells have developed the ability of detach
themselves from the primary tumor, either by moving into the surrounding tissue or by
utilizing either the lymphatic system or circulatory system to form a secondary site
(typically in the lungs, bones or liver) (3). For this reason, understanding the principles of cell motility, and identifying the initiating events which lead to the metastasis of cancer, would aid in the development of anti-metastatic drugs. One form of cancer that is of particular interest is thyroid cancer, due to its high incidence rate, and the stark variation in survival rates between the early stages of the disease and the metastatic stage.
1.1 Normal Thyroid Function
The thyroid is one of the key glands in the endocrine system. It is located below the
larynx, weighs typically less than one ounce, and has a rich vascular supply. The function of the thyroid is to intake circulating iodine, using the sodium-iodine symporter, and to convert iodine and tyrosine into thyroid hormones: thyroxine (T4) and triiodothyronine
(T3) (7). These hormones control many different processes, including metabolism, body temperature, the production of proteins, and a cell’s sensitivity to other hormones(1,7).
The thyroid is composed of follicles, which are comprised of a central core of colloid protein. That protein is itself is surrounded by a single spherical layer of follicular cells
(also known as thyroid epithethial cells) (1,7). Inside the colloid are large quantities of a
660 kDa dimeric protein called thyroglobin, whose function is to store the thyroid hormones until they are needed (1,7). Also attached to these follicles are C cells (also
2 known as parafollicular cells), whose function is to secrete calcitonin, a 32 amino acid
polypeptide hormone. The secretion of calcitonin is involved in calcium and phosphorus
metabolism (1,7) (Figure 1.1).
1.2 Thyroid Cancer
Since 1950, the incidence of thyroid cancer has increased by 240%, with an estimated
33,550 new cases expected in 2007, alone, making thyroid cancer the most common form
of human endocrine cancer (8). Out of the 33,550 new cases expected, 25,480 will occur
in women, and 8,070 will occur in men. According to the American Cancer Society, an
estimated 1,600 individuals will die from the disease in 2007 (3). The problem is that
patients with the early stage of the disease many not experience any symptoms. Which
means that, in those cases, the disease is often allowed to advance to a more problematic stage. Typical symptoms of thyroid cancer include a lump or nodule on the neck, throat or neck pain, and/or difficulty swallowing or breathing (1).
There are many different histological types of thyroid cancers (Table 1.1). The most common, accounting for 80-90% of all cases, is papillary thyroid carcinoma (PTC), which is derived from the follicular cells (1,7). Typically, the onset of the disease is between 30–50 years of age, with a female to male ratio of 3 to 1. Out of all the factors
that are associated with cancer development, radiation exposure has the highest
correlation to the onset of papillary thyroid cancer (1). When the papillary form
metastasizes, it typically spreads to cervical lymph nodes, and only rarely metastasizes to
the lungs or bones (1,7). This form of thyroid cancer has a high survival rate overall,
except for those patients with large, invasive cancers or distant metastasis.
3 The second most common type of thyroid cancer is follicular (FTC), which is also
derived from the follicular cells, and accounts for ~15% of all the cases of thyroid cancer
(1,7). Unlike the papillary form, it is not associated with radiation exposure. It is also
more aggressive than the papillary form, typically metastasizing through the circulatory
system. This leads to a more distant metastasis, such as to the lung, bones, bladder, brain,
liver, or skin (1). This form of the disease is present in an older age group, generally
afflicting those who are 40 – 60 years of age. The cure rate is ~70%.
The third most common form of thyroid cancer is medullary, which accounts for
~5% of all cases (1,7). Medullary thyroid cancers are derived from the C-cells, instead of
the follicular cells, as are PTC sans FTCs. Since we are only interested in only the forms
that are derived from follicular cells, we will not be discussing this form any further.
Anaplastic carcinomas (ATC) are thought to be derived from either papillary or
follicular carcinomas, but tend to arise more often in the papillary form, which is believed
to be more of a precursor of the anaplastic form of the disease. This form is the least
common, accounting for only 0.5–1.5% of all cases (1,7). It is the most aggressive form
of thyroid cancer, with only a 10-12 month mean survival rate (1). Its onset has been
associated with radiation exposure (1). Typically, it metastasizes locally to both the
cervical region and the lymph nodes, and distantly to either the lung or bones (1,7). Due
to its rapid growth-rate, surgical removal of the disease is rarely effective, and tends to
metastasize quickly. Onset age is around 65 years, and found more often in males than
females by a 2 to 1 ratio (1). There are no real treatment options for anaplastic thyroid
cancer. The survival rate increases to 10% at 3 years after diagnosis, with radiation and
chemotherapy (1).
4 1.2.1 Oncogenic Causes of Thyroid Cancer
Over thirty different oncogenes have been discovered in the human genome. Oncogenes are genes that when expressed are sufficient to induce malignant transformation. Some of the most common classes of oncogenes include growth factors (e.g. c-SIS) (9,10), receptor tyrosine kinases [e.g. EGFR (11,12), PDGFR (13-15), VEGFR (16), and
HER2/neu (17)], cytoplasmic tyrosine kinases [e.g. Src- (18) and BTK-families(19)], serine/threonine kinases [e.g. RAF(20,21), AKT (22,23), cyclin dependent kinases (24)], regulatory GTPase [e.g. RAS (25)], and transcription factors [e.g. MYC (26)]. Many different types of genetic alterations and dysregulation can either increase or decrease the activity of any of these proteins (such as chromosomal translocations, deletions, mutations, up- and down-regulation of gene transcription). Here we will focus on those genetic alterations that are involved in the development of thyroid cancer. Usually it takes more than one event to promote oncogenesis, but certain forms of cancer are associated with specific oncogenes. Clinical and experimental data suggests that most of the mutations associated in the formation of thyroid cancer activate the RAS/RAF (20,21,25) and phosphatidylinositol-3-kinase (PI3K)/AKT(22,23) pathways (Table 1.1).
1.2.1.1 RET/PTC and B-RAF Alteration Leads to the Development of PTC
RET/PTC rearrangements and B-RAF mutations are two of the most common oncogenic events found in PTC. The most prevalent B-RAF mutation found in PTC is
V600E. For this reason, both RET/PTC rearrangements and B-RAF mutations will require further focus. We will begin by discussing the mechanism for PTC development.
5 1.2.1.1.1 RET Activity Causes PTC
RET is a receptor tyrosine kinase that is expressed in the C-cells of the thyroid, but which is not normally expressed in the follicular cells (27,28). The RET protein contains three domains: an extracellular ligand-binding domain, a hydrophobic transmembrane domain, and an intracellular tyrosine kinase domain (29-31). Normally, a ligand binds to the receptor, causing the receptor to dimerize, followed by the autophosphorylation of certain key tyrosine residues (such as Y1062) (32). These phospho-tyrosines become binding sites for adaptor proteins, which then activate signaling pathways, including the
RAS/RAF and PI3K/ phosphoinositide-dependent kinase (PDK)-1/AKT pathways. In papillary carcinomas, the 3’ end of RET gene, containing the tyrosine kinase domain, breaks within intron 11, and translocates to the 5’ end of an unrelated gene expressed in follicular cells (33). This event causes the RET tyrosine kinase to become constitutively active, thereby increasing the activity of both the RAS-RAF and PI3K/PDK-1 signaling pathways.
There are fifteen RET rearrangements currently known in thyroid cancer, and of those fifteen, RET/PTC1 and RET/PTC3 are the most common in papillary thyroid cancer. RET/PTC1 is formed when RET is recombined with the 5’ region of the H4 gene
(28) and RET/PTC3 is formed when RET is recombined with the NCOA4 gene (34).
RET/PTC1 and RET/PTC3 have been shown to transform thyroid cells in vitro (35) and in vivo (36). RET activity has been shown to decrease the expression of both thyroglobulin (37) and the sodium-iodine symporter (38). As mentioned above, B-RAF
6 is a downstream effector of RET signaling and thyroid cells (that express RET/PTC rearrangements); when B-RAF expression was decreased, the effects of RET/PTC were reversed, supporting a key role for BRAF in RET/PTC effects (39).
The major cause of RET/PTC rearrangement is radiation (40,41). In adult sporadic papillary cancer, RET/PTC is prevalent in ~20% of all cases, and is even higher,
with a 40-70% prevalence, in children. Interestingly, those children from the Chernobyl
incident who developed papillary thyroid cancer had an ~80% frequency of RET/PTC
rearrangements 5 years after the incident occurred.
1.2.1.1.2 Mutations in B-RAF Leads to PTC
B-RAF has been shown to have increased activity in melanomas (42) and in papillary
thyroid cancer (20) in particular. Many B-RAF mutations have been characterized, but
the most common of those mutations is the valine to glutamic acid on residue 600
(V600E), which converts the kinase into its constitutively active form. Normally, the kinase is kept in its inactive conformation by hydrophobic interactions which transpire between the activation loop and the ATP binding site. However, when the V600E is present, these interactions are aberrant (43).
The BRAF V600E mutation is found in ~45% of those cases of adult sporadic papillary cancer (44). Other mutations that are found in B-RAF include K601E (45) and an insertion of a valine residue 599 (46). Children have a less frequent prevalence for these mutations, ranging from 0-12 % of all juvenile cases (47,48). The activity of B-
RAF has been shown to inhibit radioactive iodine uptake, which is associated with poor treatment response (49). Using a transgenic mouse model in which V600E B-RAF is
7 specifically expressed in the thyroid, a correlation between aggressive thyroid disease and
B-RAF activity was shown. The tumors were able to invade the blood vessels and the thyroid capsule, indicating more aggressive behavior (50).
A recent study has shown another mechanism for B-RAF activation. An in-frame inversion of chromosome 7 leads to an AKAP9-B-RAF protein. This protein, while it lacks the autoinhibitory domain of B-RAF, has the same protein kinase domain (51). This means that it is able to phosphorylate ERK and transform NIH 3T3 cells according to the study. Lastly, it was shown that B-RAF activity could actually be increased with this mechanism through an increase in copy number or with gene amplification (52).
1.2.1.2 AKT Activity and PPARγ/PAX8 Fusion Proteins Cause FTC
Heightened AKT activity has been associated with the development of FTC. Many
mechanisms can lead to increased AKT, such as an increase in the AKT gene and/or
protein expression, a decrease in PTEN activity, and an increased in PI3K activity. Here
we will discuss those mechanisms for FTC development.
1.2.1.2.1 Increase AKT Activity Leads to FTC
The up-regulation of AKT is prevalent in prostate (53), breast (54), pancreatic (55), and
non-small cell lung cancer (56). In many tumor types, there is a correlation between AKT
activity, tumor size and invasion (57-60). AKT can be up-regulated by 1) increased
protein expression (61), 2) increased PI3K activity (61), 3) decreased PTEN activity (62),
and 4) through enhanced interactions with HSP90 (63). Increased PI3K activity converts
phosphatidylinositide-4,5-bisphosphate (PIP2) to phophatidylinositide-3,4,5-triphosphate
8 (PIP3), thus giving the pleckstrin homology (PH)-domain of AKT a site to become
localized and activated (61) When localized to the membrane via its PH-domain, AKT becomes activated after being phosphorylated by PDK-1 at threonine 308 and at serine
473 by other kinases (64). Once active, AKT increases cell proliferation, cell cycle
progression, aids in cell motility and cytoskeletal stability, inhibits apoptosis, and
increases energy metabolism (22,64). PTEN normally dephosphorylates PIP3, preventing
AKT localization and activation at the cell membrane (62,65). More about the
PI3K/PDK1/AKT pathway will be mentioned further on.
In thyroid carcinomas, AKT1 and AKT2 have increased expression when compared to normal tissue (66). AKT3 remains unchanged. Typically, AKT is the most
increased in follicular carcinomas (66). Consistent with this, almost all tumors that form
from Cowden’s syndrome are caused by the loss of PTEN expression (62). Several of
groups have confirmed that sporadic FTC has an increase AKT1 and AKT2 expression
compared with normal tissue. Some studies have demonstrated that increased AKT
activity is linked to an increased expression in the PI3KCA gene (67,68), active RAS mutations (69), a decreased PTEN expression, or the expression of PAX8/PPARγ fusion protein (69). Thus, AKT activation is particularly common in FTC and is often involved in early oncogene signaling.
In PTC, it is believed that PI3K/PDK-1/AKT is more crucial to tumor progression than tumor formation. As mentioned above, PTCs are the result of either RET/PTC
rearrangements or mutations in B-RAF and RAS. Interestingly, rearrangements in
RET/PTC can also activate the PI3K/PDK-1/AKT pathway, as well as the RAS/RAF
pathway. Conversely, mutations in B-RAF do not typically activate the PI3K/PDK-
9 1/AKT pathway. These observations were verified by analyzing the cDNA microarrays
of human thyroid carcinomas (70). Patients with RET/PTC rearrangements showed an
increased gene expression of those genes which are controlled by both pathways, whereas
patients with B-RAF mutations did not. Overall, in vitro AKT activation led to an
increased cellular proliferation and increased migration in many cellular systems (57).
Unfortunately, AKT activation does not seem to be a precursor for PTC development, but
is required for further tumor proliferation and invasion. However, the exact role of AKT
in PTC development is not yet fully understood.
1.2.1.2.2 PAX8/PPARγ Leads to FTC Development
The last major thyroid cancer initiating genetic event, found mainly in FTC, was only recently deduced: a gene rearrangement between the PAX8 protein and the PPARγ
protein (71). This fusion protein is found in ~20% of all FTCs and is correlated to
malignant thyroid tumors (71,72). Low levels of PAX8/PPARγ expression have been
reported in follicular adenomas, and found at higher levels in FTCs (73), suggesting it
may have a early role in follicular cancer development. PAX8/PPARγ is not found in
PTCs (71-73). Interestingly, the PAX8/PPARγ and RAS mutations are never found in the
same tumor (74), indicating that the PAX8/PPARγ rearrangement is a different
mechanism of tumorgenesis. The rearrangement occurs between the 5’ region of the
PAX8 gene, which lacks the transactivation domain, and the exon1 of the PPARγ gene,
which is un-altered and full length (75).
10 The PAX8 promoter is active in follicular cells, and increases both expression and activity of the PPARγ protein (76). PAX8 is a transcription factor which is essential for normal thyroid development (77). PPARγ is involved in cell differentiation and proliferation (78). The fusion protein enables cells to reduce the rate of apoptosis, and to increase proliferation via cell cycle kinetics (79). The exact mechanism of PAX8/PPARγ
on tumorgenesis is not yet fully understood. PAX8 is required for thyroid development,
and its expression is highly controlled. Therefore, it is believed that the fusion protein
may decrease the typical effect which PAX8 would have on normal thyroid growth.
PPARγ is present in thyroid cells, and the fusion protein inhibits, or down-regulates,
wild-type expression of PPARγ. Recent studies have shown that PPARγ controls the
expression of PTEN (80-82), a regulator of the PI3K pathway. Thus, it has been proposed
that PPARγ/PAX8 may decrease PPARγ function, thereby decreasing PTEN expression
and increasing AKT activity.
1.2.1.3 The Development of Anaplastic Thyroid Cancer
Anaplastic thyroid cancer (ATC) is the fastest growing follicular cell cancer derivative,
and has the worst prognosis. The p53 mutation is one of the more typical initiating
events, and is the best characterized in ATC (83). It is believed that many ATCs develop
from PTC, due to the high coexistence of PTC in ATC tumors. Interestingly, B-RAF
mutations are also found in ATC (84). This further suggests that PTCs are the precursor
of ATC. It was shown that 80% of all ATCs contain a B-RAF V600E mutation, as well
as a p53 mutation (85). PI3KCA mutations that cause an increase in AKT activity are
also found in ATCs (68,86). Lastly, N-RAS was also found to be mutated in ATC (25). 11
These oncogenic events in ATC overlap with oncogenic events found in both FTCs and
PTCs, supporting the concept that ATCs develop from more differentiated PTCs and
FTCs.
1.2.2 Thyroid Cancer Staging and Treatment Options
Treatment options for thyroid cancer are dependent on the stage in which the cancer has
developed. The staging of a tumor is defined by the TNM system, which will not be
described here in detailed. Generally, the early stages (1 and 2) are highly treatable, with
survival rates of 100 % and 98%, respectively, at 5 years. Stage 3 has a survival rate of
82% at 5 years. Unfortunately, stage 4 has only a 5-year survival rate of 40% (1,3,7).
Current treatments for all forms of thyroid cancer include the surgical removal of the thyroid. The surgeon may remove part of the thyroid (a lobectomy or partial thyroidectomy) or remove the entire thyroid (a total thyroidectomy). At the time of surgery, the lymph nodes will be checked and removed if necessary. Another form of treatment is radiotherapy with I-131. Radiotherapy utilizes the sodium-iodine symporter to uptake radioactive iodine (I-131), and the radiation destroys the thyroid cells. There are minimal side-effects with this form of treatment. Unfortunately, ATCs grow too rapidly, and usually lose their uptake abilities, for this form of treatment to be very effective. With this form of cancer, an alternative option is to use either external radiotherapy or hormone therapy, which can be utilized to treat PTCs and FTCs as well.
Lastly, there is chemotherapy. Chemotherapy uses chemical compounds to inhibit the cellular process within the cell in order to stop cell proliferation and motility (metastasis).
12 Some such compounds include bleomycin, adriamycin, and cisplatinum. Unfortunately,
these compounds are ineffective against thyroid cancer. Currently, there are a number of new compounds that are better tailored to inhibit those pathways which are overexpressed
in specific forms of cancer.
1.3 AKT Signaling Pathway
The PI3K/PDK-1/AKT and the RAS/RAF pathway have been previously shown to be
integral to the development of thyroid cancer. Furthermore, these pathways have been
implicated in proliferation, anti-apoptosis, and cell motility in not just thyroid, but in
other types of cancers. Follicular thyroid cancer has an increased PI3K/PDK-1/AKT
pathway. This activation can be caused by altered expression of PTEN, through
mutations by the PPARγ/PAX8 fusion protein, or by other mechanisms. Here we will
overview the activation of the PI3K/PDK-1/AKT signaling pathway, distinguish between
the multiple AKT isoforms, and consider how this pathway controls proliferation, anti-
apoptosis and cell motility.
The PI3K/PDK-1/AKT pathway regulates many cellular processes, such as
transcription, translation, proliferation, growth, and survival (87). Altered expression of
this pathway has been shown to cause diverse disease such as cancer (88,89) and diabetes
mellitus (90). Many proteins of the PI3K pathway are located in the cytoplasm, but some
have also been shown to be present in the nucleus (91). The full importance of why this
pathway is sometimes activated in the nucleus remains unclear.
13 1.3.1 Mechanism of Activation
1.3.1.1 PI3K Converts PIP2 into PIP3
PI3K is a heterodimer that contains a p85 subunit (regulatory domain) and a p110 subunit
(kinase domain) (92). Upon ligand binding to a receptor tyrosine kinase, the receptor dimerizes, and autophosphorylation occurs at tyrosine residues. The p85 domain of PI3K binds to the phospho-tyrosine, bringing the p110 subunit in close proximity to the cellular membrane. When this occurs, the p110 subunit converts phosphatidylinositol-4,5- bisphosphate (PIP2) to phosphatidylinositide-3,4,5-triphosphate (PIP3) (92) (figure 1.2).
Another way the p110 subunit can be activated is by binding to activated RAS (93) and a
G-protein coupled receptor (93). It can also be activated by mutations in the p110α
(PI3KCA) (94). Upon conversion to PIP3, proteins containing either the FYVE or PH-
domain bind to the phospholipids (95). PH-domains, found in PDK1, AKT and other
proteins, can also bind to phosphatidylinositol-3,4-bisphosphate. PTEN, a phosphatase,
removes the phosphate group on the 3’ position of PIP3, which then prevents PH domains from binding (65,96). SH2-containing inositol phosphatases remove the phosphate group from the 5’ position, also regulating the PI3K/PDK-1/AKT pathway (97).
1.3.1.2 PDK-1 Regulates AKT Activity
A link between PI3K and many of its downstream targets [such as AKT (98) and
p70S6K (99)] is PDK-1. This serine/threonine kinase was discovered to phosphorylate
AKT in its activation loop, and to regulate AKT activity (100). PDK1 phosphorylates
several other proteins at the cell membrane: p70S6K, PRK1/2 (101), PAK (102), SGK
(103), p90RSK, certain members of PKC kinase family, and certain members of the AGC
14 kinase family (104) (figure 1.2). Through activation of these proteins via
phosphorylation, PDK1 has been shown to regulate glucose uptake, protein synthesis, and
actin reorganization, as well as to inhibit pro-apoptotic proteins.
PDK1, a 63 kDa protein, has both an N-terminal kinase domain and a C-terminal
PH-domain (91). Similar to AKT, PDK-1 localizes to the cell membrane through its PH
domain via PIP3. At the cell membrane, PDK-1 phosphorylates AKT at threonine 308,
which is located in the activation loop (T-loop) (100). PDK-1 can also become active
even without PI3K activity. PDK-1 dimerizes, and then autophosphorylates at serine 241
(located in the T-loop), allowing PDK-1 to be active and free from membrane-associated activity (105). This stage of PI3K signaling has been shown to be negatively regulated by protein—protein interactions. For example, the p107 subunit of Rb, when overexpressed in rat fibroblast, decreased phospho-p70S6K and –AKT. This decrease was
shown to be due to the ability of p107 to prevent PDK-1 from accessing the cell membrane (106).
1.3.1.3 AKT Controls Cell Proliferation and Motility
AKT, the most studied target of the PI3K pathway, will be the focus of this section. AKT
is a 57 kDa serine/threonine kinase (107). There are three known isoforms of AKT:
AKT1, AKT2, and AKT3. AKT2 and AKT3 share an 81% and 83% sequence homology
to AKT1, respectively (108). All of the AKT isoforms are structurally similar and contain
an N-terminal PH-domain, a central kinase domain, and a C-terminal regulatory domain.
All of the AKT isoforms also share the same modes of activation. Once PI3K converts
PIP2 into PIP3, the PH-domain of AKT binds to PIP3 causing a conformational change of
15 AKT, which exposes two critical phosphorylation sites required for maximal activity: serine 473 and threonine 308 (87). The latter is phosphorylated by PDK1, which leads to the stabilization of the active conformation (figure 1.1). The serine 473 kinase remains unknown. Recent studies suggest that either DNA-dependent kinase (109) or integrin- linked kinase (ILK) could be responsible for serine 473 (110). However, it is also possible that serine 473 is phosphorylated by PKC βII (111) or mTOR-rictor complex
(112), or a PDK2 that remains elusive or even AKT (113). CTMP inhibits AKT activation, by binding to AKT and thereby preventing it from localizing to the cellular membrane. This, in turn, prevents phosphorylation at both threonine 308 and serine 473 from occurring (114). Once activated, AKT travels to different compartments of the cell, to either up- or down-regulate protein activity.
1.3.2 AKT Isoforms and Functions
AKT plays a central role in the PI3K signaling pathway. In many forms of cancer, certain mutations lying upstream from AKT (such as, RET/PTC rearrangement, mutations in
PI3KCA (115), and increase expression of PI3K) cause the increased activation of AKT.
Recently, a mutation in the PH domain of AKT (E17K) was shown to increase AKT activity (116). In thyroid cancer, it was shown that AKT1 and AKT2 were overexpressed compared with normal tissue. Furthermore, the metastatic regions compare with initial thyroid tumor, had an increase of active AKT. Here we are going to explain AKT isoforms and their function in thyroid cancer development and cell motility.
16 1.3.2.1 AKT1 Regulates Cell Growth and Angiogenesis
The first cloned isoform of AKT was AKT1, which is also known as AKTα, PKB or PKBα. The AKT1 gene is overexpressed in gastric carcinomas (107), in glioblastomas
(117) and in thyroid carcinomas (66). Through the use of knockout mice, the normal physiological roles of AKT1 were deduced. The reports indicated that mice that lacked
AKT1 were smaller in size, when compared to their litter mates who did have AKT1 expression, and experienced increased apoptosis. ~40% of the mice who had the AKT1 knockout died in early embryonic development. This is believed to be due to a smaller placental development, which results in a reduced intake of the embryonic nutrients required to normal development (118). Also, the vascularization of the placenta was decreased (119). These results suggest that AKT1 is required for cell growth and angiogenesis.
1.3.2.2 AKT2 Regulates Glucose Metabolism
The next cloned isoform of AKT is AKT2, which is also known as AKTβ or
PKBβ. The AKT2 gene was found to be overexpressed in thyroid (66), ovarian (120),
pancreatic (121), and breast cancers (120). Mice that had AKT2 knockout are normal in
size, when compared to their litter mates, but developed diabetes mellitus. Another effect
of AKT2 knockout expression was the loss of adipose tissue during 5-8 months of age
(118). These results suggest that AKT2 has an integral role in glucose metabolism and
adipogenesis (122). qRT-PCR confirmed that AKT2 was the most expressed isoform of
AKT present in tissue and organs that responded to the insulin stimulus (123).
17 1.3.2.3 AKT3 Regulates Cell Size and Proliferation
The last of the cloned isoforms of AKT is AKT3, which is also known as AKTγ
or PKBγ. The AKT3 gene is overexpressed in both breast and prostate cancer (124). Mice
with AKT3 knockouts developed normally, and experienced no neonatal mortality rate.
AKT3 is typically expressed in the neuronal tissue and in the brain. Mice with AKT3 knockouts had an average brain mass that was 25% smaller than that of their litter mates.
Within the brain of these mice, all of the key brain regions were underdeveloped, due to a decrease in cell size and number (118). These results suggest that AKT3 is responsible
for postnatal brain development.
1.3.3 Target and Function of the AKT Pathway
1.3.3.1 Up-Regulation of AKT Increase Cell Proliferation
AKT increases cell proliferation through many mechanisms. One of the first studies that
investigated the function of AKT demonstrated that AKT phosphorylated glycogen
synthase kinase 3 (GSK3) at serines 9 and 21, thereby inhibiting GSK3 activity (125).
Normally, GSK3 phosphorylates of β–catenin, leading to β–catenin degradation (126).
When AKT prevents GSK3 activity, the unphopshorylated β-catenin enters the nucleus.
In the nucleus, β–catenin binds to its transcription factors causing an increase in cyclin D
expression (127). An increased cyclin D expression helps regulate Rb
hyperphosphorylation, and instigates cell cycle progression and proliferation (128). This is one mechanism of how AKT regulates cell proliferation.
AKT activity can also generate increased proliferation, by causing certain proteins to be retained in the cytoplasm. p21WAF1/CIP1 and p27KIP1, which are cyclin-dependent 18 kinase inhibitors that bind to cyclin/cdk complexes, stop cell cycle progression. AKT phosphorylates p21WAF1/CIP1 (129) and p27KIP1 (130), and retains them in the cytoplasm.
For example, AKT phosphorylates p27KIP1 at threonine 157 in its nuclear localization sequence (NLS), thereby preventing p27KIP1 from returning to the nucleus (131). Located in the cytoplasm, p27KIP1 can not bind to either cyclin A or cyclin E and cell proliferation remains uninhibited. It should be noted that this phosphorylation does not affect the normal binding function of these proteins.
Lastly, AKT can increase cell proliferation by altering protein syntheses.
Tuberous sclerosis 2 (TSC2), a growth suppressor, forms a complex with TSC1. This complex normally inhibits p70S6K, an activator of translation, and instead activates eukaryotic 4E-binding protein 1 (4E-BP1), an inhibitor of translation. AKT phosphorylates TSC2, and prevents the TSC2/TSC1 complex from forming, thus attenuating the inhibition of protein translation (132). AKT has been shown to directly phosphorylate p70S6K and mTOR, activating these molecules. Activated mTOR promotes cyclin D mRNA and inhibits 4E-BP1 (133). These are some of the many effects of AKT on cell proliferation, making it an interesting target for drug treatment design.
1.3.3.2 AKT Prevents Apoptosis
As mentioned above, AKT affects many cellular responses. One characteristic of cancer is its ability to evade apoptosis (programmed cell death). Here we will mention a few of the mechanisms for how AKT is an anti-apoptotic protein. The overexpression of
19 AKT leads to resistance of apoptosis caused by a number of death-inducing situations:
oxidative stress, irradiation, ischemic shock, the removal of extracellular signaling factors
and treatment with chemotherapeutic drugs (134).
AKT controls pro-apoptotic proteins by phosphorylating them, and thus inhibiting
their function. Normally, BAX (a pro-apoptotic protein) is bound to Bcl-xL (an anti-
apoptotic protein) thus preventing BAX from making the mitochondria membrane
permeable to cytochrome C release. BAD (a pro-apoptotic protein) will displace BAX
under apoptotic conditions, and allows cytochrome C to bind to both Apaf-1 and Caspase
9, thereby forming the apopsome (135). AKT phosphorylates BAD at serine 136 (136),
and attenuates BAD activity though an increase in its binding to 14-3-3(137), which
sequesters it. Another mechanism of how AKT phosphorylation regulates protein
function is through MDM-2. AKT phosphorylates MDM-2, causing it to translocate to the nucleus. MDM-2 then binds to p53 (a tumor suppressor) and destabilizes it, leading to cell survival (138).
AKT prevents apoptosis through the regulation of transcription factors. The forkhead family of transcription factors increases the expression of pro-apoptotic genes, such as FasL (a death receptor ligand) and BIM (a pro-apoptotic protein). Through phosphorylating AKT, there is a reduction in the nuclear import of forkhead transcription factors, thus reducing the expression of FKHR-dependent genes (87). AKT activates the cyclic AMP-response element-binding protein (139), which promotes the expression of the anti-apoptotic gene Bcl-2 (140), as well as AKT (141). Thus, AKT can increase anti- apoptotic potential by either increasing or decreasing gene transcription.
20 1.3.3.3 AKT Regulates Cell Motility
Cancer cell motility and invasion decreases the potential of patient survival. AKT’s role
in cell motility is currently under investigation, but Vasko et al. have shown that AKT1
regulates thyroid cell motility (69). Through the use of dominant negative AKT isoforms,
siRNA knockdown of expression, and animal models, those effectors of AKT that aid in
cell motility are currently being deduced. Unfortunately, AKT’s role in cell motility is
complex. Depending on the cell type, certain isoforms have shown to increase migration,
such as AKT2 in breast cancer cell lines (142,143), and AKT1 in both fibroblast and
thyroid cancer cell lines (69). Whereas the other AKT isoforms decrease cell motility,
such as with AKT1 in breast cancer cell lines. The exact mechanism for cell motility will
be covered later.
Girdin, also known as the AKT phosphorylation enhancer (APE), is found at the
leading edge of the cell. When activated by AKT, girdins aid in stress fiber and in
lamellipodia formation by binding to the actin cytoskeleton (144). ADP-ribosylation factor directed GTPase-activating protein (ACAP1) function is to promote integrin β1
recycling, thus increasing migration. The rate of integrins moving from the lagging edge
to the leading edge is an important factor in cell motility. The slower the rate of integrin
translocation, slower cell motility. ACAP1 was shown to be a substrate for AKT, so that
when AKT expression was decreased, ACAP1 was hypo-phosphorylated, thereby
resulting in decreased migration (145). A newly discovered function of both GSK3 and
the GSK-3-interacting protein (h-prune) demonstrates that these proteins inhibit focal
21 adhesion kinases (FAK), the recycling of both integrin ανβ3 and integrin α5β1, cause
disassembly of paxillin, and the activation of RAC (146,147). AKT attenuated GSK3
affect and increased cell motility.
AKT was shown to be up-regulated in both papillary and follicular thyroid
cancers, through a decrease in PTEN activity, an increase in PI3K and/or RET/PTC receptor rearrangements. It was also shown that AKT regulates cell proliferation, apoptosis and migration, despite the lack of known active mutations of AKT. This would suggest that the inhibition of either AKT or PDK-1 might represent a possible method for inhibiting the migration of thyroid cancer.
1.4 RAS/RAF Signaling Pathway
As mentioned above, the RAS/RAF signaling pathway is frequently activated in thyroid cancer. Typically, B-RAF mutations are found in PTC, and RAS activation (which also activates RAF) is found in FTC. Here we will overview the activation of the RAS/RAF signaling pathway, distinguish between the multiple RAF isoforms and examine RAF kinases control of cell proliferation, anti-apoptosis and motility.
1.4.1 RAF Mechanism of Activation
For RAF to become active, a cascade of requisite events must first occur. RAS binding, the phosphorylation of key residues in the regulatory domain, and the scaffolding proteins all contribute to RAF activation. There are three isoforms of RAS: H-, K-, and
N-RAS. Each of these isoforms binds to RAF’s regulatory domain with different affinity, but all three RAF isoforms bind to RAS with equal affinity. For example, K-RAS is a
22 better activator of C-RAF then N-RAS, but C-RAF, B-RAF and A-RAF all bind to K-
RAS equally (148). Mutations in N-RAS have been shown to be present in FTC as well
as follicular-variant PTC (149,150).
Typically, a growth factor binds to a receptor, causing the receptor to become active (for example, as with receptor tyrosine kinases). Next, accessory proteins bind to
the receptor, aiding in RAS localization to the cell membrane, such as with Grb-2 and
SOS (151-153) (figure 1.3). These proteins help RAS activation by causing a GDP-GTP switch, thereby forcing RAS into a new active conformation. But the localization of RAS
to the membrane can also occur through prenylation, franesylation, or
geranylgeranylation. Prenylation of N-RAS localizes to the Golgi membrane, whereas
prenylation of H-RAS localizes to the endoplasmic reticulum (154). Normally, C-RAF is
found as a 300-to-500 kDa protein complex consisting of HSP90 and dimerized 14-3-3
(155). The active RAS binds to RAF via the RAS-binding domain (RBD) and the
cysteine-rich domain (CRD), which causes the displacement of 14-3-3, which exposes
the certain phosphorylated residues. PP2A dephosphorylates serine 259, leading to
maximal kinase activity, and preventing 14-3-3 from binding (156) (figure 1.3). Other
phosphorylations are also required for complete RAF activity, but will be mentioned
later. All phosphorylation sites mentioned will pertain to C-RAF unless otherwise noted.
RAF kinases are highly regulated. The phosphorylation of serine 338 (157),
tyrosine 341 (158), threonine 491 (159), and serine 494 (159) are all required for C-RAF
activity. Serine 338 has been shown to be phosphorylated by PAK (157,160). In B-RAF, this residue is constitutively phosphorylated, but the kinase responsible for this phosphorylation remains unclear. Some studies suggest that it is PAK, AKT, or an
23 unknown kinase, but it is most likely regulated by the PI3K pathway. Tyrosine 341 is
phosphorylated by both Src and JAK (161), which aid in RAS binding. In B-RAF, this
residue is an aspartic acid, thus decreasing the necessity for these other pathways to be
activated. The last two phosphorylation sites (threonine 491 and serine 494) are located in
the activation loop (159). These phosphorylations prevent the kinase domain from
binding back to the regulatory domain. Initially, PKC was shown to activate C-RAF, and
it was thought to be through serine 499 (162). But changing serine to alanine did not
obliterate the effect that PKC has on RAF (163). Suggesting other kinases regulate RAF activation.
RAF activity can be regulated by phosphatases. For example, the expression of
PP1 and PP2A can lead to an increased C-RAF activity, as seen in C. elegans (164).
Serine 259 inhibits RAF kinase activity by binding to 14-3-3 (165). Jaumot and Hancock showed that PP1 and PP2A dephosphorylate serine 259 (166). Thus, the inhibition of
PP1 and PP2A leads to an increased presence of RAF/14-3-3 complexes, as well as decreased stability and activity in the RAF kinase (as measured by a decrease of phosphorylated MEK). This data suggests that phosphatases increase RAF activity.
Phosphorylation of other sites also negatively controls RAF, such as with serine
43 (which is located in RBD) (167). PKA activity reduced RAF activity, and was once
believed to work through the phosphorylation of serine 43 (168). But this notion was
abandoned when serine was changed to alanine, and failed to prevent PKA’s effect on
RAF (169). Furthermore, B-RAF does not contain serine at this site, and is sensitive to
24 PKA (169). Since B-RAF is sensitive to PKA and lacks a corresponding serine43 site,
this suggests that PKA regulates RAF kinases through other possible phosphorylation site
or the regulation of other protein known to activate RAF.
Another mode of RAF inhibition is the phosphorylation of serine 259 and serine
621. These phosphorylations allow the 14-3-3 dimer to bind, causing RAF to stay in an
autoinhibited conformation state (170). It has also been shown that Akt phosphorylates
serine 259 (171,172). Interestingly, B-RAF contains two other sites that are
phosphorylated by Akt: serine 428 and threonine 439 (172). Mutating serine 259 to
alanine failed to inhibit B-RAF activity, as was seen in C-RAF and A-RAF, but mutating
all three (serine 259, serine 428 and threonine 439) were sufficient to inhibit kinase
activity (172). Lastly, serine 364, which is unique to B-RAF, is phosphorylated by SGK,
which inhibits B-RAF kinase activity (173). This unique site suggests a potential method
for inhibiting specifically B-RAF, over other RAF isoforms. Later, in Chapter 5, it will be
explained that PAK could be a potential kinase for the phosphorylation of serine 364 in
B-RAF as well. Due to the unique role of B-RAF in PTC, such a result would be of
particular importance for drug design.
1.4.2 RAF Isoforms Share Sequence Homology and Similar Effectors
The RAF kinases function as key regulators for the MAPK cascade, which plays a key
role in thyroid cancer cell biology. Of the RAF proteins, B-RAF has been shown to play a
critical role in the development of PTC. The RAF kinases were initially discovered as a
viral oncogene in mice that induced rapidly growing fibrosarcomas or rapidly accelerated fibrosarcoma: v-raf, (also known as v-mil) (174,175). The cellular
25 homologues of RAF are known as c-raf & c-mil. The discovery that RAF and MYC are co-expressed in retroviruses has led investigators to believe that there maybe a switch from tyrosine phosphorylation to serine/threonine phosphorylation, occurring when growth factor signal moves from the peripheral to the interior of the cell (176,177). The
RAF proteins are serine/threonine kinases that convert a growth signal from a tyrosine receptor, which satisfied the investigator’s hypothesis.
There are three isoforms of RAF: C-RAF, B-RAF (178), and A-RAF (179) (C-
RAF being the first discovered). These three isoforms share a high level of sequence homology (180) and are composed of an N-terminal regulatory domain, an activation loop, and a C-terminal kinase domain. All three isoforms contain three conserved regions: CR1, CR2 and CR3 (170). The CR1 region contains the RBD and the CRD, and has been implicated in the regulation of RAF kinase activity (170,181). The CR2 region is serine and threonine rich. It has been observed that this region is phosphorylated by a number of different kinases, and also that this region controls protein-protein interaction.
The effects of these interactions help regulate its localization and the activity of the kinase. CR2 has also been hypothesized to lead to substrate recognition (180). The CR3 region contains the kinase domain, which is also regulated by phosphorylation (181).
RAF kinases are implicated in cell proliferation, motility, regulation apoptosis, as well as transcription.
1.4.2.1 C-RAF
C-RAF was the first RAF kinase that was discovered. It is also known as RAF-1 or c-
RAF-1. Its gene is located at 3p25 (182), and recent studies have shown that C RAF is
26 ubiquitously expressed (182). Unlike B-RAF, C-RAF and A-RAF do not undergo splicing that result in multiple variants. The RAF kinases recognize many of the same targets. For example, in mice that have C-RAF knockout, B-RAF was still able to activate MEK1 and MEK2 (183). C-RAF has been shown to localize to the mitochondria, and to phosphorylate BAD, thus implicating C-RAF in the prevention of apoptosis (184).
The RHO family can also activate C-RAF by PAK (through the phosphorylation of ser338) (157,160).
1.4.2.2 B-RAF
B-RAF plays a central role in PTC. Its gene is located on 7q32 (182) and is expressed in neurons, testis, spleen and hematopoietic tissue (182). There are two splice sites in B-
RAF that give rise to 10 distinct B-RAF isoforms: exon 8b and 10 (185). B-RAF can be activating by RAS, but also by RAP (186), which is another small GTPase. The effector domains of RAS and RAP are identical, but their modes of activation are different. In many cancers, including PTC, a mutation of V600E leads to a constitutively active B-
RAF (187). This mutation is located between two key phosphorylation sites located in the activation loop of B-RAF (43). It is believed that this mutation alleviates the dependence for the surrounding phosphorylation and causes the kinase domain to attain an active conformation. As mention above, B-RAF shares similar effectors as C-RAF, but has a higher basal level of kinase activity.
27 1.4.2.3 A-RAF
The least is known about A-RAF. It was originally discovered with a v-raf probe used to screen the mouse spleen cDNA library. Its gene is located on Xp11 (182), and typically expressed in the kidney, ovary, prostate and epididymis (182). A-RAF can activate
MEK1, but not MEK2 (188). The kinase domains of all three RAF isoforms share a high level of homology, as well as the RBD and CRD. But C-RAF and B-RAF activate MEK1 and MEK2. This suggests that substrate recognition depends on protein interaction, with possibly CR2. A-RAF has been found to localize to the mitochondria (189), and performs a similar function as C-RAF in preventing apoptosis.
1.4.3 B-RAF Regulates Proliferation, Apoptosis and Cell Motility
RAF kinases regulate a number of cellular processes, including metabolism, apoptosis, cell cycle, and motility. Here we will focus on some of the more cancer pertinent cellular activities of B-RAF. For example, B-RAF regulates cell proliferation through the MAPK pathway. Studies show that mice that are heterozygous for B-RAF knockout are 50% smaller than their litter mates on average, and fail to develop a normal liver, as well as vascular tissue. The homozygous B-RAF knockout was embryonic lethal (E12 days). In the C-RAF knockout mice, B-RAF was still able to activate MEK, compensating for the loss of C-RAF activity (183). These data suggest that the RAF kinases are required for embryonic development, and that they can compensate each other. B-RAF increases cell proliferation through the phosphorylation of MEK at serines 217 and 221, which in turn phosphorylate ERK (190). B-RAF can also regulate proliferation through NF-κB (191),
Rb (192), and Bcl-2 (193). In cell lines, the use of RAF and MEK inhibitors caused an
28 increase in cell cycle arrest at Go/G1. Unfortunately, the exact mechanism by which B-
RAF regulates the cell cycle remains unclear. It is nonetheless safe to conclude that, RAF
regulates cell proliferation through a array of different targets (both known and unknown).
B-RAF inhibits apoptosis by phosphorylating key pro- and anti-apoptotic proteins. It regulates the Bcl-2 family through the phosphorylation of both Bcl-2 and
BAD (184,193). B-RAF also activates NF-κB, leading to an increase in anti-apoptotic
proteins (such as IAP-1/2) (194). One mechanism B-RAF uses to inhibit apoptosis is
believed to be via phosphorylation of IKK2, which itself phosphorylates I κB, thus
causing the I κB-NF- κB complex to dissociate, and leading to I κB degradation (195).
This allows NF- κB to translocate to the nucleus, and increases the expression of anti-
apoptotic proteins. This data suggests that the RAF kinases can inhibit proteins through
phosphorylation.
B-RAF also regulates proteins that are associated with cell migration and
invasion. MEFs’ lacking B-RAF displayed a reduced number of actin stress fibers, thus
weakening cellular movement. ROCKII expression was also decreased while cofilin
activity was increased (196). These data suggests that B-RAF regulates cell movement by
interfering with actin organization. Knockdown of B-RAF expression, using siRNA,
indicated that BRAF increases invasion by directly augmenting the expressions of matrix
metalloproteinase (MMP) -2 and β-integrin (197). These data suggest that B-RAF regulates cell motility, as well as offers a possible target for the inhibition of cancer
metastasis.
29 1.5 Steps Required for Cell Motility
Cellular movement is essential for tumor invasion and/or metastasis. Therefore, the key regulators of this process offer exciting potential as therapeutic targets for late-stage cancer (198). The re-organization of the cytoskeleton, and the control of structural proteins involved in cell movement or cell division, are highly regulated processes (199).
External stimuli, such as growth hormones or physical stress, can cause cells to undergo membrane ruffling, as seen with wound healing, morphogenesis and cancer (200-202). In addition, internal stimuli can instigate cell motility. The development of a distant metastasis is a critical determinant for the prognosis of cancer, and currently inhibitors are being developed to restrict proteins that are integral regulators of cell motility and structure.
The basic structure of the cell is composed of the cytoskeleton, which consists of three polymer systems: actin filaments (actin/myosin), intermediate filaments (vimentin, lamin and desmin) and microtubules (α-, β-, γ-tubulin) (200). Normally, an immobile cell
is flattened around the edge, and has its nucleus in the center, surrounded by organelles.
In a moving cell, however, the cell is polarized. The nucleus is located more toward the
lagging edge of the cell with its organelles, and the leading edge forms a pseudopod
which is devoid of organelles. Filopodia, lamellipodia and membrane ruffles are
protrusions in the membrane that are seen in moving cells (2). The process of cellular
movement can be defined into three basic steps: protrusion, adhesion and contraction. A
more detailed depiction of the process defines cellular movement into five distinct steps –
1) the generation of membrane protrusion in the forward moving direction, 2) the formation of adhesion sites in the extended membrane (called focal complexes), 3) the
30 maturation of focal complexes into more stable focal adhesion, 4) the forward movement
of the cell by contractile forces, and 5) the detachment of the cell from the lagging edge.
We will discuss each of these five steps and consider those proteins that are associated
with each process (Figure 1.4).
1.5.1 Five Steps for Cell Motility
1. Generation of membrane protrusion
The first step of cellular movement is the extension of the leading edge. Actin filaments
regulate the morphology of the cell membrane. For example, lamellipodia are comprised
of diagonally-oriented actin filaments, and protruding from lamellipodia are radial actin
bundles which form filopodia (2). In both of these situations, a variety of proteins are
associated with actin. Actin organization is highly regulated. For example, Nobes and
Hall showed that RAC and CDC42 signaling leads to the formation of lamellipodia and
filopodia, and that RHO signaling lead to the formation of stress fiber bundles (203-206).
The link between RAC and CDC42s’ effect on membrane morphology is PAK; RHO
utilizes ROCK for its effect (203-206). Both ends of the actin filament can bind
monomeric actin, and elongate. The faster polymerizing end of the actin filament is
known as the plus end, whereas the slower polymerizing end is known as the minus end
(207). Under normal cell conditions, the actin monomer concentrations are 0.1 μM and
0.6 μM for the plus and minus ends, respectively (207). If the concentration were to increase higher than 0.1 μM at the plus end, then polymerization would occur and the filament would elongate. If the concentration were to decrease below 0.1 μM, then we would see the depolymerization and shrinking of the filament. This effect is seen in the 31 minus end as well, only with 0.6 μM instead of 0.1 μM. For example, a concentration of
0.4 μM would cause the plus end to elongate and the minus end to shrink. The addition of
actin monomers to the actin filament occurs at the plus end (the end where the membrane
is moving forward), and the removal of actin monomers occurs at the minus end (the
lagging edge). The addition of actin monomers to the leading edge, and removal of
action monomers from the lagging edge, gives rise to the “treadmill” movement of actin
(2).
The process of actin polymerization requires numerous proteins that are involved in the activation of polymerization, in cross linking, in stabilization, and in the severing of filaments. WASP and PAK are both examples of activating proteins that cause the elongation and the cross linking of actin filaments, through the use of nucleator proteins such as the arp2/3 complex (208,209). Profilin and thymosineβ-4 bind to globular actin in
order to maintain a constant actin monomer pool (209,210). Fimbrin, scruin, filamin, and
fascin are able to crosslink actin filaments with bundle actin filaments, increasing the complexity of the actin network (209,211). Capping proteins, such as CapZ, bind to the filament ends to regulate the length of the actin filament (209). Cofilin, as well as gelsolin and severin, an actin depolymerization factor (ADF), breaks actin filaments
apart, thus causing the formation of new plus ends (209,212). All of these proteins work
together to generate the leading edge of the cell.
2. Formation of Adhesion Sites (Focal Complexes)
An important step in cell motility is the formation of adhesion sites to allow for surface
attachment. Integrins link the cytoskeleton to the extracellular matrix on the outside of 32 the cell (213). Actin filaments then bind to integrin proteins that transverse the cell membrane, via actin adapter complexes that are composed of α-actinin, talin and vinculin
(214-216). There are over 50 different proteins that make up the adhesion sites (217).
Initially, small focal complexes form in the lamellipodia and filopodia. They only are
present for 1-2 minutes, where they can either dissolve or become more stable focal
adhesion sites (2).
3. Maturation of focal complexes into more stable focal adhesion
As mentioned above, focal complexes are found in the lamellipodia, and are regulated by
both RAC and CDC42. Located behind the lamellipodia and the focal complexes are
focal adhesions. Focal adhesion dynamics are regulated by RHO (204). Experimental
evidence has shown that what causes focal complexes to mature into more stable focal
adhesion is the ratio between RAC/CDC42 to RHO (2). Mechanical stress aids in the
development of focal adhesions. Actin filaments bind to integrins, as well as other
accessory proteins, in order to develop focal complexes. After this, myosin binds to the
actin, which causes the integrin complex to become more durable and more dense,
developing it into a focal adhesion (2). The binding of myosin also forms a contractile
bundle, which later provides the mechanical force necessary for cellular movement.
4. The cell moves forward by the actomyosin motor
The cell moves itself forward via the contraction of actin and myosin, which is performed
by the “actomyosin motor”. Myosin is divided into three regions: the head, neck and tail
(207). The head is where myosin attaches to actin, and both the head and neck regions are
33 what create the contracting force. The tail region binds to other myosin molecules,
vesicles, filaments, etc. The actual mechanism for myosin contraction can be divided into three steps: the binding to the substrate, the “power stroke” and the unbinding. These steps leads to the cell pulling forward (207).
5. The rear of the cell detaches
The precise mechanism for cellular detachment has yet to be fully characterized. Some
researchers believe that proteases, such as calpain, might break apart the adhesion
complexes, thereby weakening the cell’s hold on the surface (218). This would suggest
that the inhibition of certain proteases might inhibit cell motility. Another theory about
the mechanism for cellular detachment is that the contractile force pulls the cell away
from the surface, thus leaving parts of the cell’s surface behind (2). Some studies have
shown that parts of the cell membrane remain attached to the surface when the cell moves
(2,201). If this model is correct, then inhibition of this step would be difficult because it
would mean that cellular release is also controlled by the actomyosin motor.
1.6 p21-Activated Kinase Signaling Pathway: A Key Regulator of Cell Motility
Both the PI3K/PDK-1/AKT pathway and the RAS/RAF pathway have been shown to be up-regulated in thyroid cancer. These pathways activate specific proteins that lead to similar cellular responses, such as increased cell proliferation, the inhibition of apoptosis, and increased cell motility. Cross-talk between these two pathways is not uncommon. A protein of particular interest, which is regulated by both pathways, and that regulates effectors of both pathways is PAK. This protein was initially discovered during screening
34 RHO family binding partners in the cytosolic protein extract of a rat’s brain (219). PAK, a serine/threonine kinase that plays an integral role in cellular motility, also aids in both cellular proliferation and anti-apoptotic signaling. As indicated in the family name, these proteins are activated by the small GTPases; CDC42 and RAC1, but not by RHO
(220,221). All PAK proteins contain an N-terminal regulatory domain, which itself contains a p21-binding domain (PBD) and at least two Src homology 3 (SH3) binding motifs, and a C-terminal kinase domain. The kinase-preferred phosphorylation sequence is –K/R-R-X-S/T-, and it phosphorylates either serine or threonine residue. This sequence is believed to be common for most of the PAK isoforms, but some variations can be found. The -3 position is typically a basic residue, either an arginine (R), which is more common, or a lysine (K). The -2 position is believed to contain an arginine (R), an the -1 site typically any amino acid (5,222,223).
The six known PAK isoforms are separated into two groups: Group 1 (which contain PAK1, PAK2 and PAK3) (5) and Group 2 (which contain PAK4, PAK5 and
PAK6) (6) (Figure 1.5A). Group 1 PAKs, also known as classical PAKs, all contain an autoinhibitory domain (AID), and high sequence and structure homology. The AID region and PBD region overlap, suggesting that GTPases cause a conformational change in both elements, thereby increasing activity. At the amino acid level, the regulatory domains of PAK2 and PAK3 are 88% and 90% similar to PAK1, respectively, and their kinase domain are 93% and 95% similar to PAK1, respectively (6,224). Group 1 PAKs’ show increased activity when bound to the active form of members of the RHO GTPase family, [RAC1 and CDC42, but not RHO (219)]. They contain multiple SH3 domains, which bind to the –P-X-X-P- motif, thus increasing protein activity and substrate
35 reorganization (225,226). Group I PAKs’ also contain a modified SH3 domain, -P-X-P-, which binds to PIX, and aids in regulation of function PAK (227). It is worth noting,
however, that there is an ED-rich region of PAK for which the function remains elusive.
Also, there is a critical threonine residue (threonine 423), located in the kinase domain’s
T-loop, which is required for the maximal activation of PAK (228,229). Mutation of this
threonine-to-glutamic acid resulted in a constitutively active kinase, which was
independent of both RAC and CDC42 (230,231).
Group 2 PAKs, or non-classical PAKs, are structurally different than Group 1 in
their regulatory domain, and with the exception of PAK5 (which does contain a different
AID (232)), lack an AID. The regulatory domain of PAK5 and PAK6 are 72% and 60% similar to that of PAK4, respectively, and their kinase domains are 75% and 75% similar to PAK4, again, respectively (6). When Group 2 PAKs are compared to PAK1, there is less than 40% homology in the regulatory domain and they have less than 54% homology in the kinase domain (6,233). Instead of having a critical threonine residue in the T-loop, all group 2 PAKs contain a serine instead (233). Mutating the serine residue into a glutamic acid, for PAK4 and PAK6, did not make a constitutively active kinase (6,234).
Regulation of group 2 PAKs is not augmented by GTPase binding, since they lack an
AID (with the exception of PAK5), but is due to the interaction with, and expression of other proteins.
1.6.1 Mechanism for p21-Activated Kinase Activation
There are six known PAK isoforms, each of which contains similar structural elements and motifs. The PBD and the AID (only found in PAK isoforms 1, 2, 3 and 5) are located
36 in the N-terminal, which controls the regulation of the kinase domain. The AID has been shown to interact with the kinase domain via a three-helix structure (235). PAKs are serine/threonine kinases that become activated when bound to either RAC1 or CDC42 complex with GTP (219). From here on out active GTP bound to either RAC or CDC42 will be referred as either RAC or CDC42, respectively, and the inactive form will be referred as either GDP-RAC or GDP-CDC42, respectively.
1.6.1.1 Key Phosphorylation Sites and Phosphatases
Many kinases are controlled by protein modification (mainly phosphorylations, but also acetylation and the incorporation of lipids and/or carbohydrates). PAK1 alone contains multiple phosphorylation sites that regulate its activity. The most critical of these sites is threonine 423. Located in the T-loop, threonine 423 can be phosphorylated by either
PDK-1 (102) or active PAK (229). This allows the kinase to become active, and prevents the AID from re-associating. Serine 144 is auto-phosphorylated by PAK1 in order to prevent the kinase from reverting back to its inactive state (236). A serine residue in the
N-terminal, serine 21, is phosphorylated by AKT, inhibiting the Nck adaptor protein from binding to PAK1 (5,237). It is believed that this phosphorylation targets certain PAKs to a GTPases-independent mode of activation, and therefore to different cellular responses.
The final site of interest is threonine 212. Either p35/Cdk5 or cyclinB/cdc2 can
phosphorylate this residue, thereby decreasing PAK activation and activity (238,239).
Kinases have been shown to activate PAK through the phosphorylation of certain key residues. Phosphatases decrease PAK activity through the removal of phosphate groups. Three such phosphatase are known to inhibit PAK function: protein-
37 serine/threonine phosphatase 2A (PP2A) (240), partner of PIX1 (POPX1) and partner of
PIX2 (POPX2) (241). POPX1 and POPX2 were both shown to remove the phosphate at
threonine 423, leading to a decrease in the kinase activity of PAK1. It is believed that the
POPX phosphatases are localized to PAK through interactions with PIX proteins (241).
PIX bind to PAK1 through the use of a modified SH3 domain and aid in the transfer of
GTP to GDP (227). PIX1 was shown to increase PAK activation, as where PIX2
inhibited it (242).
1.6.1.2 GTPase (RAC1 or CDC42)–Dependent Activation of PAK
There are several ways to activate Group 1 PAKs. The primary method is through the
binding of GTPases RAC1 and CDC42. This binding actually increases kinase activity
more than 100 fold (236,243). As mentioned above, PAKs contain SH3-binding domains.
In PAK1, the first SH3 domain binds with the adaptor protein Nck (226) and the second
SH3 domain binds Grb2 (225). Nck and Grb2 have been implicated in increased activity
of PAK1 because they localize PAK1 to the cell membrane, where the activation of PAK
takes place. Another way of localizing PAK to the cell membrane is through G-proteins.
A G-protein βγ binding site also exists at the C-terminal end, which has been shown to
inhibit kinase activity (244). All this suggests that regulation of PAK is a very complex
process.
Normally, inactive PAKs are homodimers in a trans-configuration (235) (Figure
1.5B). The PBD/AID of one molecule is bound to the kinase region of the other
molecule. The regulatory switch for group 1 PAKs is the PBD and AID (235), both of
which are located in the N-terminal domain, and share an overlapping region. The PBD is
38 comprised of residues 67-113, and the AID is comprised of 83-149 for PAK1 (235).
When either RAC or CDC42 binds to the PBD, there is a conformational change in the
PBD/AID (245), causing the three helices of AID to adopt a different conformation,
increasing the Kd (normally at 90nM) (5). These changes release the kinase domain from
one PAK molecule and the kinase reorders itself into a more preparatory state for the
activation of the kinase domain. The threonine 423 residue, located in the T-loop, is phosphorylated to prevent it from reverting back to its inhibited state, and to activate the kinase (5). This site has been shown to be autophosphorylated by PAK in vitro (229).
Threonine 423 can also be phosphorylated by PDK-1 (102).
The final step required for the retention of kinase activity is the phosphorylation
of certain serine residues, such as serine 144, which also prevents the kinase from
reverting back to an inactive state (246,247). It should be noted that GTPase binding
increases PAK activation, but may also provide an allosteric mechanism which prevents
the kinase from reverting back to its inactive state (248). RAC and CDC42 are prenylated proteins (249) located in the cell membrane, increasing their activity. It is postulated that the prenylation of RAC aids in PAK activation, by causing PAK to be in close proximity
to PDK-1. After PAK activation has occurred, GTPase activity is no longer required. This
notion was proven by the fact that PAK3 has been shown to dissociate from CDC42 after
activation (219).
1.6.1.3 GTPase-Independent Activation of PAK
For Group 1 PAKs, binding with either RAC or CDC42 increases activity in a GTPases-
dependent mechanism. But for Group 2 PAKs, is has been shown that activity does not
39 increase upon binding with RAC or CDC42 (6). As mentioned above, Group 1 PAKs have a critical threonine residue, located in the T-loop for maximal activation. In Group 2
PAKs, this residue is a serine, and mutation of it failed to increase activity (233). In addition to this, the Group 1 PAKs inhibit each other via trans-homodimer, and the binding of either RAC or CDC42 released these molecules. Since, PAK4 and PAK6 both lack an AID, other mechanisms for activation need to be explored. The excised catalytic domains of both PAK4 and PAK6 have shown greater activity than the full length protein, suggesting the N-terminal region may be regulated by other intramolecular mechanisms (233,250).
Recent studies have revealed that PAK also performs non-kinase functions. Under certain physiological conditions, PAK proteins can act more like a scaffolding protein than a kinase. For example, PAK can act as a link between the RAC mediated activation of guanylyl cyclase activity, which is responsible for lamellipodia formation (251). Using a recombinant active PAK protein, guanylyl cyclase experienced increased activity.
However, in as analysis of guanylyl cyclase, the researchers failed to detect that any of the proteins were phosphorylated. This suggests an alternative mechanism of PAK was needed in order to affect other proteins. What was discovered was that when RAC bound to PAK, this causes a conformational change in PAK, which then causes a conformational change in guanylyl cyclase, increasing it activity (251). This kinase- independent function of PAK is also believed to be responsible for membrane ruffling
(252). Another GTPase independent mechanism is how PAK6 regulates androgen receptor (AR) and estrogen receptor (ER) transcriptional activity (250,253). This will be discussed shortly.
40 1.6.2 Isoforms of PAK – Functions and Key Characteristics
1.6.2.1 PAK1
PAK1 has been the most studied isoform out of the six known PAK isoforms, and its function overlaps with other PAK isoforms as will be mentioned later. Normally, PAK1 is expressed in the spleen, breast, thyroid, brain, and muscles. PAK1 has been shown to inhibit apoptosis, through the phosphorylation of BAD at both serine 112 and serine 136
(230). Found in the leading edge of cells, PAK1 regulates cellular structure and movement through the regulation of the myosin light chain kinase (MLCK) (254), the phosphorylation of myosin light chain (255), activation of Desmin (256), activation of merlin (257) and activation of LIMK (258). A constitutively active PAK1 (CA-PAK) also induces membrane ruffles, filopodia, localization at focal adhesion sites (later found to be due to interactions with focal adhesion kinases), and the retraction of the lagging edge (224). A dominant negative effect, through the use of either the p21 inhibitory domain (PID) or kinase dead PAK (K299R), leads to a decrease in membrane microspikes, in focal adhesion, and in stress fibers (259). The PID diminished cell motility by decreasing the formation of lamellipodia in a random manner, and the CA-
PAK increased cell motility. These data suggests that PAK1 is an integral protein for cell motility (260). Lastly, PAK1 has been shown to regulate gene transcription, through the phosphorylation of the ER at serine 305, located in the activation function-2 domain
(261).
41 1.6.2.2 PAK2
PAK2 is ubiquitously expressed, and shares some similar functions with PAK1. But the greatest difference between PAK2 and the other PAK isoforms is that PAK2 contains a
Caspase 3 cleavage site, located between the regulatory domain and the kinase domain, at
Asp212. This causes a 28 kDa regulatory fragment, as well as a 38 kDa constitutively active kinase fragment (262,263). This gives PAK2 a dual function in the apoptotic pathway. PAK2 is both an anti- and a pro-apoptotic protein. Once Caspase 3 cleaves
PAK2, the kinase domain becomes constitutively active (263). More about the PAK2 apoptotic function will be mentioned later.
1.6.2.3 PAK3
PAK3 is the least studied of the classical PAKs. It is normally expressed in the brain and in the testis. PAK3 is implicated in X-linked mental retardation, which is caused by point mutations, and is the only PAK isoform associated with a human genetic disease. One mutation of PAK3 is located in the PBD (R67C) region, and obliterates GTPases binding
(264,265). Another mutation of PAK3 (R419Stop) generates a truncated kinase-dead version of PAK3 (265), and yet another mutation (A365E) is located in the highly conserved region of the in kinase domain (265,266). PAK3 has been identified as a key regulator of both synapse formation and plasticity in the hippocampus (265). In patients with Alzheimer disease, apoptosis and DNA synthesis are both activated by PAK3 when it interacts with the amyloid precursor protein in neuronal tissue. A variant of PAK3 leads to the resistance to both apoptosis and DNA synthesis (267). Lastly, PAK3 has been shown to phosphorylate C-RAF at serine 338, aiding in RAF activation (160).
42
1.6.2.4 PAK4
PAK4, the most studied of the Group II PAKs, seems to be ubiquitously expressed, but is
expressed at even higher levels in the prostate, testis, and colon. PAK4 preferentially
binds CDC42 over RAC, and a CDC42 mutant that eliminated binding to PAK1, PAK2
and PAK3 was still able to bind to PAK4. Normally, PAK4 is located around the nucleus,
but upon binding with CDC42, PAK4 relocates to the brefeldin A-sensitive compartment
of the Golgi membrane (233). Also, the expression of PAK4 leads to sustained filopodia
formation, suggesting that PAK4 is an initiator of membrane morphology (233). In a soft agar assay, the overexpression of PAK4 led to colony formation, and no longer has anchorage dependent growth, suggesting that PAK4 expression may cause cell transformation (268). Chernoff et al. showed that active PAK4 also caused a decrease in actin bundles, stress fiber formation and focal adhesion, via a LIM kinase-dependent mechanism (269).
PAK4’s effect on both apoptosis and protein expression has also been studied.
PAK4, as well as other PAK isoforms, phosphorylates BAD on serine 112, inhibiting apoptosis (270). In addition to this, PAK4 is also known to increase AKT activity, which also leads to increased phosphorylation of BAD, thereby preventing BAD activity (270).
When a kinase-dead PAK4 isoform was transfected into cells, RAS expression was decreased (268), suggesting a possible mechanism of regulation of the RAS/RAF pathway.
43 1.6.2.5 PAK5
PAK5 is expressed only in the brain, but appears to be constitutively active (271). As
mentioned above, PAK5 contains an AID (232), which is about 120 amino acids long.
PAK5 contains a nuclear export, a nuclear import and a mitochondrial localization
sequence, all of which cause PAK5 to cycle between the mitochondria and the nucleus
(272). In the mitochondria, PAK5 phosphorylates BAD, preventing apoptosis (273).
When excluded from the mitochondria, PAK5 can no longer phosphorylate BAD and can no longer prevent apoptosis. PAK5 is activated by CDC42, but also by RHO-D and
RHO-H (274). When overexpressed in NIE-115 cells, PAK5 was able to cause neurite- like projections, filopodia, and dendritic spines (271). Interestingly, PAK5 expression is not associated with cancer. Overexpression of PAK5 leads to an increased Jun N-terminal protein kinase, which normally increases apoptosis, but PAK5 also phosphorylates BAD, leading to decreased apoptosis, giving conflicting roles for PAK5 (275). Tau stabilizes microtubules, and is negatively regulated by phosphorylation by MARK2. PAK5 inhibits
MARK proteins by direct interactions via their catalytic domains, not requiring phosphorylation. PAK5 sequesters and obliterates MARK function on tau. Thus, PAK5 aides in microtubule stabilization by preventing down regulation of tau (276).
1.6.2.6 PAK6
PAK6 functions differently than the other PAKs. It is expressed at high levels in the
brain, testis, breast, kidney, prostate and placenta. Normally found in the cytoplasm,
PAK6 translocates to the nucleus upon ligand binding and suppresses gene expression
associated AR-mediated transcription. PAK6 binds to both the AR and the ER (250). AR
44 consists of three domains, the N-terminal transactivation domain, a DNA-binding domain, hinge, and a ligand-binding domain (277). When PAK6 directly binds to AR, it prevents AR from binding to other transcription co-activators. Similarly, by phosphorylating the AR DNA-binding domain, PAK6 prevents AR from translocating into the nucleus (253). Like other Group II PAKs, PAK6 activity is unaltered upon the binding of GTPases (6).
1.6.3 Targets and Function of PAK proteins
In recent years, the effects of PAK signaling on cell motility, apoptosis, and cell proliferation have started to be elucidated. Here we focus on some of the ways PAK affects these cellular responses and the proteins that are associated (Table 1.2). Here we will review certain effects that have been associated with thyroid cancer, suggesting the potential importance of the PAK signaling pathway.
1.6.3.1 PAK Increases Cell Proliferation
PAKs have been shown to increase cell proliferation. For example, a constitutively active
PAK4 is able to increase cell proliferation in an anchorage-independent way in NIH-3T3 cells (268). Activation of the MAPK signaling pathway has led to an increased proliferation in many cell lines and cancers. PAK1 and PAK3 mediated phosphorylation of C-RAF at serine 338 is essential for maximal activity (160,184). Also, PAK1 can activate MEK, via the phosphorylation on serine 298, which is also downstream of both
C-RAF and B-RAF (278,279). RAF can then activate MEK, which in turn activate ERK, or MEK can activate ERK directly, which leads to increased survival and proliferation.
45 Both of these pathways are activated in thyroid cancers, but not usually within the same cancer type. This activation (PAK phosphorylation of C-RAF) could be a major link between the RAS/RAF pathway and the PI3K/PDK1 pathway.
During the process of cell division, PAKs, have been shown to regulate microtubule formation (280), which is necessary for chromatids to separate, possibly through Aurora A and PAK (281-283). PAKs are affected by cyclin dependent kinases present in mitosis that may be required to coordinate the actin cytoskeleton, and to aid in cell division. Cyclin B1/CDC2, and/or p35-bound CDK5, phosphorylate PAK1 on threonine 212 at different times of the cell cycle (284). PAK2 and PAK3 do not have this site. This phosphorylation did not change the kinetics of the kinase activity, but allowed an, as of yet, unidentified substrate to act with PAK1. Phospho-PAK1 threonine 212 is localized to the microtubule organizing center (284). The effect of PAK on microtubule stability suggests that PAK decreases stability, possibly allowing the microtubules to retract pulling the chromosomes apart. All of these points suggest that PAK aids in the regulation of cell proliferation. A characteristic of cancer is increased cellular proliferation, and that PAK functions suggest a mechanism its involvement in that increased proliferation.
1.6.3.2 PAK Decreases Apoptosis Through Multiple Mechanisms
Evading apoptosis is a critical characteristic of tumorgenesis. NF-κB is known to increase cell survival by increasing the expression of proteins that are anti-apoptotic, and
BRAF (V600E) has been shown to activate NF-κB in thyroid cancer (194). PAK1 can also promote anti-apoptosis through the activation of NF-κB (285). As seen in 46 Helicobacter pylori, PAK1 activates NF-κB kinase, which in turn activates NF-κB (286).
Thus, PAK could also regulate NF-κB’s effect in thyroid cancer. Also, through its kinase activity, PAK can sequester forkhead transcription factors (FKHR) to the cytosol, preventing FKHR from increasing the transcription of pro-apoptotic proteins (287).
These examples provide proof that PAK can regulate apoptosis by regulating transcription.
PAK can regulate apoptosis by regulating protein function. For example, PAK1 inhibits apoptosis by leading to the inactivation and degradation of dynein light chain 1
(DLC1), in breast cancers (288). DLC1 is able to promote cell survival by binding to
BIM, which keeps BIM from binding to Bcl-2. When it receives apoptotic signals, the
DLC1-BIM complex dissociates, and BIM is able to inhibit Bcl-2 (289). PAK1 can also bind to the DLC1-BIM dimer to phosphorylate both proteins, thus leading to their inactivation and degradation (288). Also, PAK family members phosphorylate BAD at serine 112, thus preventing the Bcl-2 family from inducing apoptosis (230,270,273,290).
These example shows that PAK can regulate apoptosis by regulating protein function.
Interestingly, PAK2 has some unique features with respect to apoptosis. PAK2 is cleaved by Caspase 3 at asparate 212, which is located between the regulatory domain and the kinase domain, causing the kinase domain to become constitutively active
(262,263). Interestingly, this event transpires late in the apoptotic process. Normally, full length PAK2 prevents apoptosis by phosphorylating BAD (290), as seen with PAK1. But the cleaved kinase product (PAK2p34) is constitutively active, and therefore aids in programmed cell death. A recombinant PAK2p34 increased cell death in Jurkat cells
(262), HeLa (291), and CHO (291). Conversely a dominant negative PAK2 postponed 47 onset of the apoptosis induced by the Fas receptor signaling pathway (292). Thus, PAK2
appears to have a dual function in the regulation of apoptosis. PAK2 is ubiquitously
expressed, and therefore is present in all thyroid tissue, which means that it represents a
potential mechanism for resistance to apoptosis in thyroid cancer.
1.6.3.3 PAKs are an Integral Protein in Cell motility
PAKs effect on cell motility is the most studied of the cellular responses
performed by this signaling pathway. Initially, PAK1 was found to localize to both
cortical actin structures and focal adhesions, which are found in the lamellipodia,
filopodia and membrane ruffles at the leading edge of the cell. Transfection of either a
constitutively active PAK1 or RAC leads to the formation of lamellipodia. Interestingly, the transfection of either PAK2 or PAK4 leads to different membrane structures. In all cases, these PAK-transfected developed increased motility in vitro. The expression of the
PID (which inhibits PAK1, 2 and 3 kinase activity) leads to the loss of both focal adhesion and of microspikes induced by RAC and CDC42. Expression of a kinase-dead
PAK1 generates multiple lamellipodia, which develops in varying directions, which in turn causes an overall decrease in directional cell motility. (For review see Bokoch ref.
[(5)]) These results suggest that PAK proteins play an integral role in cellular movement and in membrane morphology.
At non-mitotic times in the cell cycle, PAK also effects control of microtubules.
The mechanism PAK uses to alter the microtubule dynamic is through interaction with stathmin/Op18. Normally, Op18 destabilizes microtubules by binding to the αβ tubulin dimers, preventing their polymerization (293). But PAKs phosphorylate Op18 at a critical
48 residue, serine 16, which reduces Op18’s destabilizing effect on microtubules (294). This stabilizes the leading edge of the cells, thus showing how PAK aids in cell migration.
Another mechanism PAKs use to regulate the highly-organized process of cellular movement is through actin filament formation. When a PAK molecule (either PAK1,
PAK2 or PAK4) becomes active, it phosphorylates many targets associated with actin organization (such as the myosin light chain kinase (254), ARP2/3 (295) and LIMK 1/2
(258). LIMK 1/2 regulates actin organization by controlling of cofilin, an actin depolymerization protein (296). PAK activates LIMK 1/2 by phosphorylating a threonine residue (LIMK1 = thr508), which is located in the T-loop. PAK thereby increases LIMK activity by 10-fold (5). Active LIMK then phosphorylates cofilin on serine 3, rendering the cofilin inactive, which allows actin filaments to form and increase stability (296).
PAKs interaction with LIMK1/2 therefore makes it a crucial protein in the regulation of cell motility.
Another mechanism PAK uses to regulate cell motility is through interaction involving myosin light chain regulation. Myosin is the mechanical protein found in the actomyosin motor. Myosin contains an ATPase function that utilizes the released energy from ATP hydrolysis to physically move the myosin head, there by pulling on the actin, and leading to contraction. Myosin is made up of two myosin-heavy chains, which contain the actin-dependent ATPase function as well as two myosin-light chains, a regulatory chain and an essential chain. (For review of the actomyosin motor, see ref
[(297)]) A critical step in the functioning of the actomyosin motor is the phosphorylation of both serine 19 and threonine 18 on the myosin regulatory chain. It has been shown that
PAK1 and PAK2 phosphorylate only serine 19 (298,299), but another kinase was
49 discovered that phosphorylates both sites: the MLCK (254). PAK2 phosphorylates
MLCK on serines 439 and 991, inhibiting MLCK’s function (300). This allows for the formation of stress fibers and focal adhesion, as well as prevents the contraction of the leading edge toward the center of the cell. The lack of PAK at the side and rear of the cell, where contraction occurs, supports the claim that PAK prevents contraction and increases adhesion at the leading edge in a specific manner.
Lastly, PAKs regulate cell motility by regulating Filamin A (301). The function of
Filamin A is to cross-link actin filaments (302). Originally, PAK1 and Filamin A were shown to interact through a two-hybrid screen (301). Filamin A binds to PAK1 through the PDB, thus increasing PAK kinase activity. In turn, PAK phosphorylates Filamin A at serine 2152, and causes both membrane ruffling and the inhibition of cofilin (301) (most likely through LIMK). The exact mechanism how Filamin A causes membrane ruffle and the inhibition of cofilin remains unclear. However, PAK regulates actin crosslinking by regulating Filamin A, which further emphasizes the importance of PAK on cell motility.
1.7 Conclusion
PAK has been shown to play a critical role in the proliferation, apoptosis and motility of cells. Unsurprisingly, many forms of cancer display an elevated PAK expression. In the case of thyroid cancer, the PI3K signaling pathway is increased due to activating mutations in RET/PTC or PI3KCA. These mutations increase PDK-1 activity, which in turn would be predicted to enhance PAK activity. PAK has been shown to activate C-RAF and MEK, increase cell proliferation and motility, pathway known to be important in thyroid cancer development. (Figure 1.6) PAK therefore controls those
50 functions that are integral for metastasis: it regulates cell proliferation through mitosis, it inhibits apoptotic proteins, and manages cellular motility. Given the low patient survival rate found in those forms of thyroid cancer that are associated with metastasis, this suggests that PAK could represent a possible target for thyroid cancer therapy.
51
Figure - 1.1 Structure and function of the thyroid. The thyroid, a two lobe gland, is found at the base of the neck. The purpose of the thyroid is to uptake iodine and to produce T3 and T4. It is comprised of follicles, which are follicular cells surrounding the colloid. The colloid stores the T3 and T4, until needed. Adapted from [(1)]
52
Table 1.1: Thyroid Cancer Statistics.
53
Figure - 1.2 PI3K/PDK-1/AKT activation. Binding a ligand to the RTK causes the PI3K to localize to the cell membrane, where it converts PIP2 to PIP3. PDK1 and AKT are then localized to the cell membrane through the preferential binding of their PH-domain to PIP3. From there, PDK1 either activates AKT, which then acts upon it is effectors, or it activates other proteins.
54 Figure 1.3 - Activation of RAF Kinase. RTK autophosphorylates, allowing GAP proteins to bind. This aids in the transfer of GTP to RAS, thus activating it. Activated RAS then binds to RAF, causing it to dissociate from 14-3-3. PP2A removes a phosphate from serine 259. Other phosphorylations by PAK (as well as other proteins) aid in the activation of RAF.
55 Figure 1.4 - 5 steps for cell motility. 1) The cell uses its actin network to form lamellipodia and/or filopodia. 2) The cell attaches itslef to the surface with structures called focal complexes. 3) The focal complexes mature into focal adhesion. 4) The mechanical actomyosin motor drives the cell to move in a unidirectional manner. 5) The lagging edge of the cell detaches from surface. Adapted from [(2)]
56 Figure 1.5 - PAK isoforms and their method of activation. There are six known isoforms of PAK. A) Comparison of each PAK isoform. Green represents the SH3 binding domains, blue represents the PBD, yellow represents the AID, purple represents ED-rich region, light blue shows the modified SH3 binding domain, red shows the kinase domain and dark blue depicts βγ complexes binding domain. Black triangles are important phosphorylation site with corresponding kinase located below. B) Activation of PAK isoforms. PAK is in a transhomodimer in its inactive form. When active RAC1 or CDC42 bind to the PBD, a conformational change in the AID occurs, causing the PAK proteins to dissociate. PDK1 phosphorylates a critical residue (threonine 423) in the T-loop for maximal activation. Which is then followed by other phosphorylations that prevent PAK from reverting back into its inactivate form. Adapted from [(5,6)]
57
Table 1.2: Effectors of PAK.
58
Figure 1.6 - PAK interacts with the PI3K/PDK-1/AKT signaling pathway and RAS/RAF signaling pathway. A link between PI3K/PDK-1 and RAS/RAF signaling pathway is shown through PAK. PAK is activated by PDK-1 and PAK also activated C-RAF. Ultimately, PAK, AKT, RAF, MEK, and ERK all aid in the regulation of cell motility. With, PAK being a central node in above mentioned regulation of cell motility, we expect that PAK would a possible therapeutic target.
59
CHAPTER 2
OSU-03012, A CELECOXIB DERIVATIVE, DIRECTLY TARGETS p21-
ACTIVATED KINASE
2.1 Introduction
PDK1 is an important regulator of PI3K/AKT cell signaling (108). Downstream effectors of PDK-1, such as AKT (303), p70S6K (304), PAK (102), and atypical forms of PKC
(305) play a critical role in the regulation of cellular proliferation, metabolism, apoptosis,
and motility (306). Overactivation of the PDK-1-regulated pathways commonly occurs in
cancer and it has been associated with aggressive clinical behavior in some tumor types
(307). Therefore, there has been an interest in developing inhibitors of PDK-1 and its effectors for cancer therapy.
OSU-03012, (2-amino-N-{4-[5-(2-phenanthrenyl)-3-(trifluoromethyl)-1H- pyrazol-1-yl]-phenyl} acetamide), is a recently developed derivative of celecoxib that competitively inhibits PDK-1 at low micromolar concentrations in vitro (4). OSU-03012 has also been shown to induce cell death in an AKT-dependent manner both in prostate and breast cancer cell lines, and display broad anti-tumor activity in vitro. Finally this agent inhibits the growth of prostate and breast cancer xenografts in vivo, and is well- tolerated (4,308). This suggests that inhibition of PDK-1 using OSU-03012 might have potential for anti-cancer therapy.
60 The global process of epithelial-to-mesenchymal transition (EMT) is both
common and mechanistically important in solid tumor progression (309-312). Regulatory
pathways for EMT therefore represent important potential therapeutic targets for the
stabilization and/or reversal of progressive cancer. PAKs are direct targets of PDK-1
(102) that function as regulators for both cell motility (313) and EMT (314). As such, the inhibition of PAKs should also cause a concurrent reduction in EMT and cell motility.
Therefore, the inhibition of PDK-1-mediated PAK activation represents a potential therapeutic approach to cancer therapy.
The activation of PAK is required for the formation and stabilization of the
leading edges of cells that result in directional motility (315). The generation of dynamic
motility-related structures at focal adhesions is regulated in part by small GTPases of the
Rho family, such as RAC and CDC42 (316-318), as well as other pathways. Both GTP-
bound RAC and CDC42 bind to specific sites in the N-terminal PBD of PAK resulting in
the subsequent release of PAK from its auto-inhibited state (319). This is maintained by
binding to PAK interacting exchange factor (PIX). Once released from PIX binding, PAK
requires phosphorylation of several residues for activation, including
autophosphorylation at serine 144 (236,247) and phosphorylation by PDK1 at threonine
423 (102,320), the latter of which is critical for maximal activation of the protein.
Threonine 423 has also been reported to be a PAK autophosphorylation site (229,321).
In the present study, we demonstrate that OSU-03012 inhibits both AKT and
PAK activity, and is cytotoxic in three thyroid cancer cell lines. Unexpectedly, the ability
of OSU-03012 to reduce levels of phosphorylated PAK occurred at concentrations below those required to block PDK1-mediated AKT phosphorylation in two of the examined
61 lines. Subsequent studies have demonstrated that OSU-03012 directly inhibits PAK kinase activity with an IC50 of ~1 μM. Overexpression of constitutively activated PAK1 cDNA partially rescued the ability of the most motile cell line (NPA) examined to migrate in the presence of OSU-03012, suggesting that PAK inhibition, either directly or indirectly, is involved in the biological effects of the compound in these cells.
2.2 Materials and Methods
Reagents and Vectors – Anti-AKT (#9272), anti-phospho-AKT (Ser473) (#9277), anti-
PAK1/2/3 (#2604), anti-phospho-thr423 PAK (#2601), and anti-Myc (#2276) antibodies were purchased from Cell Signaling Technology (Beverly, MA). Anti-PARP antibody was obtained from Santa Cruz Biotechnology (sc-8007, Santa Cruz, CA), anti-Vimentin
(#6630) was obtained from Sigma-Aldrich (St. Louis, MO.), anti-phospho ser56-vimentin
(#D076-3) was obtained from Medical & Biological Laboratories (Woburn, MA), and anti-GAPDH (#N8300-250) was obtained from Novus Biologicals (Littleton, CO).
Expression vectors containing myc-tagged constitutively activated (CA) PAK1 cDNA was the generous gift of Dr. Jonathan Chernoff (Fox Chase Cancer Center, Philadelphia,
PA) (224,231). All other reagents were obtained from Sigma-Aldrich unless otherwise noted.
Cell Culture and transient transfection of plasmids - Human thyroid carcinoma NPA
(papillary) (322), WRO (follicular) (323) and ARO (anaplastic) cell lines (322) were obtained from Dr. Guy Juillard (University of California, Los Angeles, Los Angeles, CA) and were grown in RPMI 1640 medium (Invitrogen, Carlsbad, CA), supplemented with
62 10% fetal bovine serum (FBS), 100 mM L-Glutamine, and 100 μM Non-Essential Amino
Acids (Invitrogen). Once the cells achieved 70% confluence, the cells were washed, trypsinized, and re-plated. The growth medium was replaced with RPMI 1640 containing either 0.2% or 5% FBS as noted for individual experiments for 24 hours. This medium was aspirated, and replaced with fresh RPMI 1640 containing the same FBS concentration as well as either OSU-03012 or vehicle. For transfection studies, NPA cells were seeded 16 hours before transfection at ~20% confluency. Cells were transfected using Lipofectamine-plus transfection reagent (Invitrogen) according to the manufacturer’s protocol. Transfection efficiency was estimated by co-transfection with pEGFP (Clontech).
Cell Proliferation - NPA, WRO and ARO cells were seeded into six-well plates at
150,000 cells/well in RPMI 1640 containing 10% FBS. After the 24-hour attachment
period, cells were treated with the indicated concentration of either OSU-03012 or DMSO
vehicle in RPMI 1640 containing 5% FBS. At different time intervals, cells were
harvested by trypsinization, and were counted using a Coulter counter model Z1 D/T
(Beckman Coulter, Fullerton, CA). Triplicates were performed in all experiments, and experiments were performed on three separate occasions. The concentration of OSU-
03012 required to inhibit growth by 50% (GI50) was calculated using the guidelines
suggested by the NIH.
Cell Viability Analysis - The effect of OSU-03012 on cell viability was assessed by
using the 3-(4, 5-dimethylthiazol-2-yl)-2, 5-diphenyl-2H-tetrazolium bromide (MTT) assay. Triplicates were performed in each individual experiment, and experiments were
63 repeated on at least three separate occasions. Cells were grown in 96-well plates for 24 hours, and were exposed to various concentrations of OSU-03012 dissolved in DMSO
(final concentration 0.1%) in RPMI-1640 containing 5% FBS. The medium was removed, replaced by 200 µl of 0.5 mg/ml of MTT in RPMI 1640 with 10% FBS, as per
the manufacturer’s recommended protocol (T.C.I. America, Portland OR). Cells were
then incubated in the CO2 incubator at 37°C for 2 hours. Supernatants were removed from
the wells, and the reduced MTT dye was solubilized in 200 µl/well DMSO. Absorbance
at 570 nm was determined on a plate reader. The inhibitory concentration (IC50) was
calculated using CalcuSyn (Ver. 1.2 Biosoft).
Cell Cycle Analysis - NPA, WRO and ARO cells were treated with OSU-03012 for 24
hours in RPMI 1640 containing 5% FBS. All cells were collected, and 1x106 cells were
centrifuged, resuspended in cold 70% ethanol, and stored at -20°C until analysis. Washed
cells were stained in 0.1% Triton X-100 in PBS with both 1 µg/ml propidium iodide
(Sigma-Aldrich) and RNase A (Invitrogen). Flow cytometry was performed using a flow
cytometer (BD FACSCalibur, BD Biosciences) with a 610 long pass filter for data
collection. Data was filtered, and cell cycle phases were quantified using the Modfit
program (Verity Software, Bowdoin, ME). Duplicates were performed in all experiments,
and experiments were performed on three separate occasions.
PAK Kinase Assay - Recombinant active PAK 1 (Cell Signaling), with various
concentrations of either OSU-03012 or 0.1% DMSO, was incubated in either the
presence or absence of MBP (Upstate) in a 1x kinase buffer (Cell Signaling). The
64 addition of [γ-32P]-ATP (2.5 μCi/reaction or 5 μCi/100 μM cold-ATP) (GE Healthcare,
Piscataway, NJ) started the reaction, which was incubated at 30oC for 30 min. The
reaction was terminated by the addition of EDTA. The phosphorylated MBP was
separated from the residual [γ-32P]-ATP using P81 phosphocellulose paper, and was
quantified by a scintillation counter after three washes with 0.75% phosphoric acid.
Results were also verified following visualization of 32P-labelled MBP following
separation on 12% polyacrylamide gel electrophoresis and autoradiography. For the
Lineweaver-Burk plot analysis, the ratio of radioactive ATP to non-radioactive ATP was
unaltered. The IC50 values are the means of five independent experiments, quantified
using p81 phosphocellulose paper performed in duplicate and were calculated using
CalcuSyn (Ver. 1.2 Biosoft). Additional experiments (n=3) were performed as described
above using OSU-03013, another celecoxib derivative with PDK-1 inhibitory effects (4).
Protein Extraction and Immunoblotting – Protein extraction and immunoblotting were
performed as previously described in detail (69,324). After cells were washed with ice-
cold PBS, cells were collected, centrifuged for 5 min at 500x g, and washed with 500 μl of ice-cold TBS. Either cell lysis or M-PER buffer (Pierce Biotechnology, Inc., Rockford,
IL) supplemented with 20 mM 4-amidino-phenyl methane-sulfonyl fluoride, 0.3 µM
Okadaic acid and 1 µg/ml each of aprotinin, pepstatin, and leupeptin, were added to the
tubes, which were incubated on ice for 15 min. The cells were centrifuged at 16,000 x g
for 10 min at 4°C, and the supernatant was saved and stored at -80°C. Protein
65 concentrations were calculated using a Micro BCA Protein Assay Kit (Pierce). Relative
quantitation of proteins was determined using Imagegauge software (Fuji Photo Film Co.,
LTD, Tokyo, Japan).
Apoptosis Analysis - To assess apoptosis, a Cell Death Detection ELISA kit (Roche
Diagnostics) for nucleosome detection, and an immunoblot analysis for cleaved poly
(ADP-ribose) polymerase (PARP), were performed. The ELISA was performed
according to the manufacturer’s instructions. In brief, 8 x 105 of NPA, WRO and ARO
cells were cultured in a T-25 flask for 24 hours before treatment. Cells were treated with
the DMSO vehicle or OSU-03012 at the indicated concentrations in RPMI 1640
containing 5% FBS for 24 hours. The cells were then collected, and cell lysates
equivalent to 1 x 104 cells were used in the ELISA. Triplicates were performed in all
experiments, and experiments were performed on three separate occasions. For the PARP
cleavage assay, cellular proteins from cells treated with OSU-03012, in the same
conditions as described above, were used to detect fragmented PARP by immunoblotting
with anti-PARP antibody.
Migration assays - Migration assays were performed as previously described in detail
(69,324). NPA cells were grown in RPMI 1640 containing 10% FBS. After 24 hours, the medium was aspirated, and RPMI 1640 containing 0.2% FBS was added for 24 hours.
The cells were trypsinized for 2-5 min, washed, and resuspended in RPMI-1640 containing 0.2% FBS. The cell concentration was calculated by hemocytometer. 400 μl of 0.2% FBS medium was added into 24-well plate wells, and a Boyden chamber (8 μm pores) was inserted into each well. 9 x105 NPA cells were added to each insert, and the 66 o NPA cells were allowed to attach in an incubator at 37 C 5% CO2 for 30 minutes. The inserts were switched to a new well containing RPMI 1640 containing 5% FBS. OSU-
03012 or vehicle was added to the upper chamber, and the chamber was incubated for 24 hours. Cells were visualized by Diff-Quik (Dade Behring Inc., Newark, DE) staining before and after swiping the top of the membranes, allowing for determination of total cells, as well as the number of cells on the bottom of the membrane, as previously described. Experiments were performed on at least three separate occasions.
Computer modeling – The three-dimensional structure of OSU-03012 was both generated and minimized, using MMFF94 force field, by SYBYL 7.1 software (Tripos
Associates, St. Louis, MO; 2002). The crystal structure of PAK1 kinase with two mutations [entry code 1YHV, K299R, T423E; Lei, M.; Robinson, M. A.; Harrison, S. C.
(319)] was retrieved from the RCSB Protein Data Bank (325). The heteroatom entries were deleted, and residue 299 was modified from Arg into Lees in 1YHV. The protein structure was then subjected to the addition of polar hydrogens, Kollman charges, and solvation parameters using the molecular modeling software “AutoDock Tools” as described (326). Gasteiger charges and rotatable bonds of OSU-03012 were assigned with default parameters. The structure of ATP was extracted from the crystal structure of phosphorylase kinase (PDB accession code 2PHK), and was assigned Gasteiger charges.
The torsions of ATP were defined as 0. The grid parameter file of 1YHV was generated on the well-defined ATP binding domain, and the default parameters in AutoDock Tools were used. The docking parameter files of both OSU-03012 and ATP were defined with
GA runs [100], population size [150], and evaluations [50,000,000] using the Lamarckian
67 genetic algorithm. Other default parameters were used. Docking was performed with
AutoDock 3.05 (327) using the Pentium 4 Cluster at the Ohio Supercomputer Center
(internet address: http://www.osc.edu).
Statistical Analysis - All quantitative data are represented as mean +/- S.D. Analysis was
performed using GB-STAT v10.0. The significance of the differences between groups
was examined by both ANOVA and post-hoc analysis, or by Dunnett’s Test if the
distribution was normal. Statistical significance had a P-value of <0.05.
2.3 Results
2.3.1 OSU-03012 Inhibits Thyroid Cancer Cell Proliferation and Induces Apoptosis
To investigate whether OSU-03012 inhibited thyroid cancer cell proliferation, NPA,
WRO, and ARO cells derived from poorly differentiated human papillary, follicular, and
anaplastic thyroid cancer, respectively, were treated with OSU-03012 at 2.5-10 μM for 1,
2, and 3 days. By comparison to vehicle-treated cells, OSU-03012 inhibited proliferation
of all cell lines. Moreover, OSU-03012 appeared to be cytotoxic at higher concentrations,
as shown by the decrease in cell number below the original number that was plated
(Figure 2.1A). This cell counting data was confirmed using an MTT assay (data not
shown). The GI50 and IC50 concentrations for after 24 hours of treatment were 1.29 ± 0.10
and 3.18 ± 0.80 for NPA, 2.16 ± 0.09 and 6.53 ± 1.57 for WRO, and 2.10 ± 0.22 and 5.91
± 0.92 μM for ARO cells, respectively.
68 To determine the effects of OSU-03012 on thyroid cancer cell cycle progression,
as well as to compare its effects with PI3 kinase inhibition, NPA, WRO and ARO cells
were treated with either OSU-03012 or LY2940012, a PI3 kinase inhibitor, or DMSO for
24 hours. Cells were labeled with propidium iodide, and flow cytometry was performed.
LY294002 induced cell cycle arrest in G1/S phase beginning at 12.5 μM (data for 25 μM shown in Figure 2.1B), as been previously reported (328). In contrast, OSU-03012 treatment resulted in an increase of cells in S phase without an increase of cells in G2 in a dose-responsive manner beginning a 1 μM (Figure 2.1B). These data suggest that the anti-proliferative mechanisms between these two agents differ.
To determine whether the cytotoxic effects of OSU-03012 (Figure 2.2A) were associated with apoptosis, Western blots were performed to detect PARP cleavage. OSU-
03012 treatment induced PARP cleavage in all three cell lines (Figure 2.1C), confirming that OSU-03012 induces apoptosis in these cells. The amount of PARP cleavage varied between the three cell lines, and correlated with the quantitative results (detecting
nucleosome release by ELISA, data not shown). Thus, it seems likely that apoptosis is
not responsible for all of the OSU-03012-induced cell death in thyroid cancer cells.
Nearly 100% of NPA and ARO cells died after exposure to 10 μM OSU-03012 (see
Figure 2.1A).
2.3.2 OSU-03012 Inhibits AKT and PAK Phosphorylation in Thyroid Cancer Cells
NPA, WRO, and ARO cells were treated with either OSU-03012 or vehicle for 24 hours.
Western blots were performed using anti-threonine 308 phospho-AKT, anti-total AKT,
and anti-GAPDH primary antibodies. The anti-phospho-threonine 308 AKT antibody was 69 chosen because it is a PDK1-specific phosphorylation site. OSU-03012 treatment resulted in reduced AKT phosphorylation by PDK1, which is consistent with its known inhibitory effect on this kinase (Figure 2.2). The dose that was required varied among the cell lines.
The ability of OSU-03012 to inhibit PDK1-dependent PAK phosphorylation was assessed by Western blot for threonine 423 phospho-PAK (Figure 2). In WRO and NPA cells, OSU-03012 reduced threonine 423 phospho-PAK levels at lower concentrations than it did AKT phosphorylation, by comparison to vehicle alone (zero concentration lane). Vimentin is a downstream target of PAK, involved in both intermediate filament organization and cell motility, which is expressed in NPA cells, but not WRO or ARO cells (329). Using an antibody for the PAK-dependent phosphorylation of vimentin at the serine 56 residue, OSU-03012 reduced vimentin phosphorylation in concert with the reduction of PAK phosphorylation, which is consistent with reduced PAK activity
(Figure 2.2C). These results suggest that OSU-03012 inhibition of PAK kinase activity could be possibly through a direct inhibition mechanism.
2.3.3 OSU-03012 Directly Inhibits PAK Activity via Competitive Inhibition of ATP
Binding
Because OSU-03012 inhibited PAK phosphorylation at lower concentrations than PDK-
1-dependent AKT phosphorylation in two of the cell lines, and since the threonine 423 in
PAK has been reported to be both a PDK1 and a PAK autophosphorylation site (102), we questioned whether OSU-03012 might also directly inhibit PAK. Recombinant PAK1 protein was incubated with MBP, a known substrate of PAK1, in the presence of increasing concentrations of OSU-03012. Representative autoradiographic data from one
70 of several initial dose-response experiments is shown in Figure 2.3A. These experiments
suggested that the IC50 of OSU-03012 for PAK was between 500 nM and 1 μM. To
define the IC50, five subsequent PAK kinase assays were performed using p81 paper for
quantitation (Figure 2.3B). Using this assay, we found that OSU-03012 reduced PAK
activity with an IC50 of 1.03 ± 0.59 μM. Subsequent kinetic experiments demonstrated an
inverse relationship between the effects of OSU-03012 on PAK activity and on ATP
concentration. The Km of PAK for ATP was calculated to be 27 μM. The Lineweaver-
Burk plot of these data suggests that OSU-03012 is a competitive inhibitor of ATP
binding to PAK (Figure 2.3C). OSU-03013 (4-[5-(2-phenanthrenyl)-3-(trifluoromethyl)-
1H-pyrazol-1-yl]-phenyl-guanidine) is another derivative of celecoxib that is structurally
similar to OSU-03012 and also inhibits PDK-1 (4,330). Similar to OSU-03012, the OSU-
03013 inhibited PAK kinase activity with an IC50 of 0.81 ±0.56 μM. Further cellular
studies were not pursued with this agent due to toxicity in vivo (C-S Chen, unpublished
observations).
We next performed molecular modeling, to determine potential binding sites for
OSU-03012 in the ATP binding domain of PAK1. Since the crystal structure of PAK1 complexed with ligand is not available, we modified the double mutant 1YHV (K299R,
T432E) for which a crystal structure is known, to undertake docking. This model was chosen because it has been demonstrated to adopt an essentially active conformation
(235). We modified the Arginine 299 back to the wild-type Lysine 299 in 1YHV and hypothesized that this would represent the active conformation of PAK1. To test this hypothesis, and to compare the binding modes of both ATP and OSU-03012, the coordinates of ATP in 2PHK were retrieved (see methods) and were rigidly docked to 71 the modified PAK1 structure with no rotatable bonds allowed in ATP. The binding mode
of ATP is shown in Figure 2.4A, in which ATP is docked into the ATP binding pocket of
PAK1. As in other kinases, ATP has been predicted to form two hydrogen-bonds with
Alanine 297 and Lysine 299. In addition to this, the 5’-OH group of the ATP sugar
moiety forms a hydrogen-bond with Glutamate 315. An ionic bond links the α-phosphate
group of ATP with the terminal amino group of Lys299.
The docking results suggested a potential site for OSU-03012 interaction with the
ATP binding domain of PAK1 (Figure 2.4B). Only one potential docking mode was
found, and this mode was similar to that previously reported for the interaction of OSU-
03012 with PDK1 (4). In this model, two hydrogen bonds are predicted to form between
the terminal amino group of OSU-03012 and both Alanine297 and Valine 342. The
phenanthrene ring of OSU-03012 interacts with residues Leucine 311, Isoleucine 312,
and Lysine 308 through hydrophobic interactions. In addition, the model predicts an
important π-cation interaction between the protonated terminal amino group of Lys299
and the 2-phenyl group in OSU-03012 with a distance of about 4Ǻ (331).
2.3.4 OSU-03012 Inhibits Thyroid Cancer Cell Migration in a PAK-Dependent
Manner
Due to the central role of PAK in cell motility, we next investigated whether OSU-03012
would be able to inhibit thyroid cancer cell migration. For these experiments, we focused
on the NPA cells, because they are more motile than the other cell lines in this assay
system, and demonstrate at mesenchymal phenotype (329). Initial experiments
72 demonstrated a statistically significant reduction in migration, detected between 1 and 2.5
µM of OSU-03012 compared with DMSO control (data not shown, but see control vector experiments in Figures 2.5A and B).
To determine whether this effect was PAK-dependent, NPA cells were transfected with a constitutively activated PAK cDNA (CA-PAK) or control vector. Expression of
CA-PAK was confirmed by Western blot (Figure 2.5D). Cell migration was examined in the presence of either OSU-03012 (1 to 5 μM) or vehicle for 24 hours. 1, 2.5 and 5 µM of OSU-03012 reduced the migration of vector-transfected cells in comparison to vehicle-treated cells (P<0.05, Figure 2.5A and B). In preliminary experiments, lower concentrations of OSU-03012 (<1 μM) minimally inhibited the migration of NPA cells, but this was not statistically significant (data not shown). Overexpression of CA-PAK in
NPA cells reduced the ability of OSU-03012 to block cell migration, as demonstrated by a shift in the dose-response, so that only the 5 µM dose blocked migration (Figure 2.5B; p<0.05), consistent with a partial rescue effect. Moreover, the level of cell migration in
CA-PAK transfectants was statistically different from that of the controls at both the 1 and 2.5 µM doses (p<0.01). Transfection efficiency was approximately 30%, ,as estimated by co-transfection of a GFP-expressing plasmid.
2.4 Discussion
In the present study, we demonstrate that the celecoxib derivative OSU-03012 is a direct inhibitor of PAK kinase. OSU-03012 was developed from celecoxib and established to be a PDK-1 inhibitor at low micromolar concentrations. OSU-03012 was able to inhibit proliferation and induce apoptosis in an AKT-dependent manner based on rescue 73 experiments using constitutively activated AKT1 (4,332). Consistent with these data, we
demonstrate that OSU-03012 is able to decrease proliferation and increase apoptosis in three thyroid cancer cell lines in concert with inhibition of PDK1-dependent AKT phosphorylation. Moreover, we demonstrate that OSU-03012 appears to induce S-phase arrest in these lines, in contrast to the effects of the PI3 kinase inhibitor, LY294002. The precise mechanism for this effect is currently being investigated, by performing rescue experiments in a larger number of cell lines, but could include the inhibition of PDK-1,
PAK, and/or targets that have yet to be defined and which might vary between cell systems.
As shown in Figure 2.3, OSU-03012 directly inhibits PAK1 kinase activity through competitive inhibition of ATP binding. This finding is supported by computer- based molecular modeling, which suggests a potential binding site for OSU-03012 in the
ATP binding region of PAK 1 (Figure 2.4). It should be noted that the AutoDock software used predicts a higher binding affinity for OSU03012 than ATP, which is consistent with the experimental data by comparing the Km of PAK1 for ATP (27 μM) and the IC50 of OSU-03012 (1.03 μM). These data, in addition to the similarity between
the predicted docking modes of OSU-03012 for PAK1 and PDK-1, suggest a common
mechanism of action that might account for its multikinase inhibitory effects. In addition,
OSU-03012 was submitted to a commercial kinase profiling service (Upstate) to perform
a screening test to evaluate its specificity in a panel of kinases, including AKT,
CDK7/cyclin H, IKK, PKA, p70S6K, PKCγ, MEK1, MAPK2, PDGFRα, c-RAF, and c-
Src. None of these kinases were inhibited by OSU-03012, at 10 µM (data not shown).
74 While it is recognized that this screening test is not exhaustive, and that OSU-03012 is
probably active against other kinases, the absence of activity against this panel of kinases
at 10 µM suggests a degree of target specificity for OSU-03012.
Because PAKs play an important role in cell motility, and because OSU-03012 blocks both PDK1 and PAK activity, we hypothesized that the ability of OSU-03012 to inhibit cell motility would depend in part on its ability to inhibit PAK. We focused on the
NPA cells, due to their high migratory capacity in Boyden chamber assays. Indeed, the ability of OSU-03012 to inhibit motility in these cells appears to depend partially on its inhibition of PAK activity, as determined in the rescue experiments using the constitutively activated PAK cDNA. These data are further supported by the observation that OSU-03012 inhibits PAK phosphorylation of vimentin, a key regulator of EMT and migration in these cells (329), at similar doses to the anti-motility effects of the compound. Thus, it appears that the effect of OSU-03012 on migration depends, at least in part, on PAK inhibition in NPA cells, which could occur both by inhibition of PAK and PDK1. The rescue experiments with the PAK1 constructs serve as “proof of principle” to demonstrate that the inhibition of PAK is involved in the biological effects of OSU-03012 in NPA cells. It is not yet clear whether the anti-PAK effect in NPA cells primarily occurs via the inhibition of PDK1, PAK, or both. In addition, the rescue experiments with PAK1 do not exclude other OSU-03012 targets from playing a role in its effects. Further studies are required to determine both the relative and more generalized importance of PAK inhibition in comparison to other OSU-03012-targeted pathways on migration, proliferation, and/or apoptosis in multiple cell lines with different genetic backgrounds (Figure 2.6).
75 There are six known isoforms of PAK. PAKs 1-3 share similar modes of activation (GTPase-dependent), as well as a high degree of sequence homology in both the regulatory and kinase domains. By contrast, PAKs 4-6 have GTPase-independent modes of activation, and share less sequence homology with PAKs 1-3 [reviewed in
(333)]. The ability of OSU-03012 to block isoform-specific kinase activity is still an area of active investigation, but it seems unlikely, based upon the sequence homology of
PAKs 1-3 in the kinase domain.
The extensive literature demonstrating that PAK activity plays a role in both directional cell motility and EMT is supported by recent findings in clinical samples. For example, in breast cancer, increased expression and phosphorylation of PAK are associated with aggressive clinical behavior, as well as the loss of response to anti- estrogen therapies (334,335). In addition to this, the ability of PAK1 to induce breast cancer in vivo was recently demonstrated (336). Moreover, PAK overexpression and activation has been implicated in the invasion and progression of a variety of others cancers (313). Thus, PAKs appear to have a broad functional role in cancer. However, the role of PAK in thyroid cancer has not yet been defined.
In summary, we have established that OSU-03012, a derivative of celecoxib, inhibits PAK kinase. The effect of the compound on PAK activity occurs at similar or lower concentrations than PDK-1, based on in vitro kinase and cell-based assays. Both of these activities may be fundamental to the biological effects of OSU-03012. These data suggest that OSU-03012, or compounds with similar mechanisms of action, may represent new therapeutic tools for the treatment of progressive malignancies which are associated with PAK overactivation.
76
Figure 2.1 - OSU-03012 decreases cell viability and increases cells in S- phase. A) NPA, WRO and ARO cells were treated with increasing doses of OSU-03012 for 0, 24, 48 and 72h and then counted. OSU-03012 caused a decrease in cell numbers in each cell line (*P<0.01). The Y-axis represents arbitrary units, with the number of cells normalized to day 0 (pre-treatment), which is set at 1.0. Mean values and standard deviations are shown. B) NPA cells were treated with 1 and 2.5 μM OSU-03012 or 25 µM LY294002 for 24h. Cells were collected and stained with propidium iodine, then analyzed by flow cytometry. OSU-03012 caused a dose-related increase in the number of cells in the S-phase, while LY294002 blocked G1/S transition. Representative data are shown. Similar results were noted in replicate experiments and with the ARO and WRO cell lines (not shown). C) Western blot for PARP demonstrates PARP cleavage in all three cell lines, beginning at 5μM for NPA cells and 7.5 μM for both WRO and ARO cells.
77
Figure 2.2 - OSU-03012 decreases AKT and PAK phosphorylation. ARO, WRO, and NPA cells were incubated with either OSU-03012 or DMSO alone (0 μM) at increasing concentrations for 24h. Whole cell lysates were isolated, and Western blots were performed. Representative data are shown and demonstrates that OSU-03012 inhibits PDK-1-dependent threonine 308 AKT phosphorylation and threonine 423 PAK phosphorylation. In NPA cells, the only cell line that expresses vimentin, OSU-03012 also reduced PAK-dependent vimentin phosphorylation at serine 56. These data are representative of experiments performed in triplicate.
78
Figure 2.3 - OSU-03012 directly inhibits PAK through direct competition with ATP. A) An in vitro recombinant PAK1 kinase assay, using MBP as substrate, demonstrates a dose-dependent reduction of p-MBP levels by OSU-03012, as determined by autoradiography. The inhibitory effect occurs at doses above 0.5 μM. Data from one representative experiment is shown. B) Five quantitative assays using P81 paper were performed to define the inhibitory range of OSU-03012 for PAK activity. Using this data, the IC50 was calculated to be 1.03 ±.59 μM. Results from all experiments are shown as means values, with standard deviations as error bars. C) A Lineweaver-Burk plot of the competition of OSU-03012 with ATP, utilizing a PAK1 kinase assay. One representative experiment is shown. Substrate specificity was determined using an ATP concentration range of 5-100 μM in the presence of increasing concentrations of OSU-03012. The Km of PAK for ATP was determined to be 27 ± 6 μM in multiple experiments (n=3).
79
Figure 2.4 - Molecular modeling of potential OSU-03012 interactions with PAK 1. A) Using the crystal structure of 1YHV (K229R, T432E) PAK1, the predicted structure for the wild-type PAK1 was obtained. The predicted docking of ATP into the binding pocket is demonstrated. B) The predicted docking of OSU-03012 into the PAK ATP binding domain is demonstrated. The PAK1 structure (modified 1YHV) is shown in ribbon form, and the ATP (panel A) and OSU-03012 (panel B) structures are presented as stick and ball structures, and colored by atom types. The important amino acid residues are labeled with name and number. Hydrogen bonds are indicated in yellow. π-Cation interactions are depicted in green and ionic interactions are in white. Distances for these interactions are labeled in yellow. In the structures of ATP and OSU-03012, gray is carbon; light blue/white is hydrogen; blue is nitrogen; red is oxygen; green is fluorine; and pink represents phosphorus. The chemical structures of ATP and OSU-03012 are shown. OSU-03012 chemical name is: 2-amino-N-{4-[5-(2- phenanthrenyl)-3-(trifluoromethyl)-1H-pyrazol-1-yl]-phenyl} acetamide (4). This work was done in collaboration with Yunlong Zhang.
80
Figure 2.5 - OSU-03012 inhibition of NPA cell motility is partially PAK- dependent. NPA cells were transiently transfected with either CA PAK or vector, and were exposed to either vehicle or OSU-03012. Panels A and B: OSU-03012 inhibited migration of the vector-transfected cells at 1, 2.5 and 5 µM concentrations (*P<0.05 vs. vehicle) while migration of the CA-PAK- transfected cells was blocked by only 5 µM OSU-03012 (*P<0.05 vs. vehicle). The percentage of migration at the 1 and 2.5 µM doses was greater for the CA- PAK-transfectants than the control transfectants (P<0.01). Mean values and standard deviations are shown (Panel B) from experiments performed in duplicate on three separate occasions. Panel C: OSU-03012 effects on cell proliferation were not rescued by the expression of CA-PAK. Results (mean and standard deviations) are from experiments performed in triplicate on three separate occasions, and were non-significant at all data points. D) Expression of CA-PAK was confirmed by Western blot using anti-myc antibody, and by the presence of a slower migrating band using anti-PAK1/2/3 (arrow).
81
Figure 2.6 – OSU-03012 inhibits PDK-1 and PAK decreasing cell motility. OSU- 03012 was initially shown to be a PDK-1 inhibitor. Further analysis have shown that OSU03012 also inhibits PAK directly by competitively binding to the ATP-binding site. These results suggested that OSU-03012’s mechanism to decrease migration was due to it ability to inhibit PAK. Further suggesting that PAK is an integral protein in cell motility and thyroid cancer development.
82
CHAPTER 3
p21 ACTIVATED KINASES ARE UPREGULATED IN
HUMAN PAPILLARY CANCER
3.1 Introduction
Cancer is the product of the deregulation of signaling pathways. Deregulation can occur
due to an increase in gene expression, mutations that increase/decrease enzymatic
activity, and/or genetic rearrangements that develop new chimeric proteins. Investigation
of PI3K/PDK-1/AKT, one pathway that has been studied, has shown that multiple
different events along a single pathway can lead to similar responses (such as increased
proliferation, evasion of apoptosis, and metastasis). For example in thyroid cancer, PTEN
null mutations, that renders this protein inactive, causes an abundance of PIP3 to accumulate in the cell membrane, thus increasing PDK-1 and AKT activity(337). Other events can turn on multiple pathways (PI3K/PDK-1/AKT and RAS/RAF), as with
RET/PTC rearrangements (329,338,339). This particular rearrangement is known to activate AKT and RAS/RAF signaling pathways. Other pathways, such as Wnt
(340,341), NF-κB (342), Notch (343), Hedgehog (344), and TGFβ (26,345) are also
83 altered in cancer development. Understanding how these pathways are activated and
interact with each other in vivo would aid in the development of therapies to treat many
types of cancer.
Since 1950, the incidence of thyroid cancer has increased by 240%, with an
expected 33,550 new cases in 2007, making it the second most rapidly rising form of cancer in the United States (3,8). The treatment for the disease in its early stages has a
high success rate, but patients with extensive invasion and/or distant metastasis have only
a 40% survival rate at 5 years. The ability to deduce pathways that are deregulated in
thyroid cancer and in metastatic locations would aid in understanding the progression of
the disease and those factors that influence metastasis, and may help in the development
of new compounds to inhibit the metastatic disease.
PTC is the most common form of thyroid cancer, accounting for 85% of
occurrences, which makes it responsible for the majority of patients with metastatic
disease. It is also one of the major precursors for the anaplastic form of the disease. In
80% of all cases, the initiating events leading to the development of PTCs have been
linked to mutations in B-RAF and RET/PTC rearrangements (20). Both of these
mutations cause an increase in ERK and AKT signaling, which tends to enhance growth.
PTC typically metastasizes to the lymphatic system, but the events leading to a more
aggressive disease have yet to be fully understood (1,7).
FTC is less common than PTC. Instead of metastasizing through the lymphatic system, FTCs typically metastasize through the circulatory system (1,7). Different from
PTCs, FTCs are associated instead with mutations in RAS genes (25), PPARγ/PAX8
rearrangements (74), and inactivation of PTEN (346). Notably, mutations in RAS cause
84 similar effects as RET/PTC rearrangements. Also, decreased PTEN activity in FTC
increases the activity of PDK-1 and AKT. Thus the upstream mutations lead to similar
downstream effects in both PTCs and FTCs.
With cross-talk between many pathways at further downstream targets (such as
PI3K/PDK-1/AKT and RAS/RAF/MEK), investigation into points of convergence for
these pathways may lead to important potential chemotherapeutic targets. When PI3K
converts PIP2 to PIP3, PDK-1 activity is increased, leading to activation of AKT and
other downstream targets. Well-characterized targets of PDK-1 include AKT (91),
p70S6K (99), and PAK (102). PDK-1 phosphorylates p70S6K at threonine 229 (99), and
AKT activates p70S6K at serine 389 through the mTOR pathway (133). Also, PDK-1
phosphorylates PAK at a critical residue, threonine 423 in PAK1 (102), whereas AKT
has been shown to phosphorylate serine 21 which decreases PAK’s dependence on
membrane localization for activity (237). RAS, a small GTPase, also can activate PI3K
(347) and RAF. RAS activation of PI3K leads to increased PAK activity through PDK-1, but a more describe target would be RAF for RAS. As mentioned before, RAS can activate RAF, but PAK1 has been known to activate C-RAF (157,160,184). Therefore, one point where these pathways converge, PI3K/PDK-1/AKT and RAS/RAF/MEK, is
PAK.
p21-activated kinases are serine/threonine kinases that are primarily activated by small GTPases, CDC42 and RAC1(5). PAKs are divided into two subfamilies: Group I
(classical PAK) and Group II (non-classical PAK). The Group I PAKs (PAK1, PAK2, and PAK3) contain an N-terminal regulatory domain and a C-terminal kinase domain(5).
The regulatory domain contains a p21-binding domain (PBD), where CDC42 and RAC1
85 bind(5). Part of the PBD is the autoinhibitory domain (AID), which binds to the kinase
domain of another PAK molecule (5). In the inactive form, two PAK molecules bind in a
trans-homodimer(236,248), maintained by PAK interacting exchange factor
(PIX)(227,242). Upon binding CDC42 or RAC1, there is a conformation change in the
AID, which releases the kinase domain (5,229,236,248). From there, either PDK-1 or
PAK phosphorylates threonine 423 and kinase activity is increased and maintained by other PAK autophosphorylations (5,102). The Group II PAKS are similar to Group I
PAKs, but lack an AID (5) (except in PAK5(232)). The Group II PAKs (PAK4, PAK5, and PAK6) can be activated by both GTPase-dependent and GTPase-independent mechanisms (5). It has been shown that in breast, colon, and ovarian cancer, PAK1 expression is increased, whereas in bladder and T-cell lymphoma there is an increase in
PAK1 gene expression (313). In pancreatic cancer, there is increased gene and protein expression of PAK4 (313).
PAK has a central role in multiple pathways, and also acts as a link between
PI3K/PDK-1/AKT and RAS/RAF, two pathways that have been shown to be integral in the development of thyroid cancer. Here, we assessed the gene and protein expression of
PAK isoforms in three thyroid cancer cells, which represented the papillary, follicular and anaplastic forms of the disease. We then assessed the PAK expression in 10 papillary thyroid human tumor samples and compared them to normal thyroid tissue. What we found was that all the cell lines expressed PAK1, PAK2, and PAK4 proteins. We found that the follicular and anaplastic forms also expressed RNA for PAK6, but protein expression was not detected. The human tumor samples showed that the gene expression of PAK1, PAK2 and PAK4 were increased, and the protein lysates corresponded to the
86 gene expression profiles. Importantly, levels of phosphorylated PAK (threonine 423)
were increased in the thyroid cancer samples, consistent with increased activity. Because
of this, we hypothesized that PAK activity maybe involved in thyroid cancer biology.
3.2 Materials and Methods
Reagents and Vectors – Anti-PAK1 (#2602), anti-PAK2 (#2608), anti-PAK3 (#2609),
anti-PAK4 (#3242), and anti-phospho-thr423 PAK (#2601) antibodies were purchased
from Cell Signaling Technology (Beverly, MA). Anti-PAK5 (#ST1097) was purchased from Calbiochem (San Diego, CA). Anti-PAK6 (#B50227) was purchased from
Stratagene (La Jolla, CA). Anti-GAPDH (#N8300-250) was obtained from Novus
Biologicals (Littleton, CO). DNase I (#18047-019) and TRIzol® Reagent (#15596-026)
was purchased from Invitrogen (Carlsbad, CA). MultiScribeTM Reverse transcriptase
(#4311235), Power SYBR® Green PCR Master Mix (#4367659) and TaqMan®
Universal PCR Master Mix (#4304437) were purchased from Applied Biosystems (Foster
City, CA). Expression vectors containing GFP-tagged control, p21 binding inhibitory domain (PID), myc-tagged PAK1, HA-tagged PAK2, HA-tagged PAK3, HA-tagged
PAK4, HA-tagged PAK5 and HA-tagged PAK6 cDNAs was the generous gift of Dr.
Jonathan Chernoff (Fox Chase Cancer Center, Philadelphia, PA)(224,231). All other reagents were obtained from Sigma-Aldrich unless otherwise noted.
Cell Culture and transient transfection of plasmids - Human thyroid carcinoma NPA
(papillary)(322), WRO (follicular) (323) and ARO (anaplastic)(322) cell lines were obtained from Dr. Guy Juillard (University of California, Los Angeles, Los Angeles, CA)
87 and were grown in RPMI 1640 medium (Invitrogen, Carlsbad, CA) supplemented with
10% fetal bovine serum (FBS), 100 mM L-Glutamine, and 100 μM Non-Essential Amino
Acids (Invitrogen). Once the cells achieved 70% confluence, the cells were washed, trypsinized, and re-plated. The growth medium was replaced with RPMI 1640, containing either 0.2% or 5% FBS as indicated for individual experiments for 24 hours. For transfection studies, NPA cells were seeded 16 hours before transfection at ~20% confluency. Cells were transfected, using Lipofectamine-plus transfection reagent
(Invitrogen), according to manufacturer’s protocol. Transfection efficiency was estimated by transfection with pEGFP (Clontech).
Protein and RNA Isolated from 10 PTC Patients - Human tissues samples were the generous gift from Dr. Albert de la Chappelle. They were obtained using an Institutional
Review Broad approved protocol.
Protein Extraction and Immunoblotting – Protein extraction and immunoblotting were performed as follows (69,324). After cells were washed with ice-cold PBS, cells were collected, centrifuged for 5 min at 500x g, and washed with 500 μl ice-cold PBS. M-PER buffer (Pierce Biotechnology, Inc., Rockford, IL) supplemented with 0.3 µM Okadaic acid and 1 µg/ml each of aprotinin, pepstatin, leupeptin, and 20 mM 4-amidino-phenyl methane-sulfonyl fluoride were added to the tubes which were incubated on ice for 15
min. The cells were centrifuged at 16,000 x g for 10 min at 4°C, and the supernatant was
saved and stored at -80°C. Protein concentrations were calculated using Micro BCA
Protein Assay Kit (Pierce).
88 RNA isolation – RNA was isolated from the thyroid cancer cell lines as per the
manufactor’s instructions. Briefly, the cells were washed with PBS twice, followed by
TRIzol® RNA isolation. The RNA was washed with 75% ethanol and air dried. The
RNA was resuspended with DEPC-water, and the concentration was calculated on a
BioMate3 from Thermo Electron Corporation.
Primer Design – All six PAK isoform sequences were retrieved from NCBI, and aligned
using Dialign 2.0 (348). Regions where the isoforms had the least similarity were
selected. The regions were then inputted into Primer ExpressTM 1.5, and the Tm and GC- content was calculated to be near 60oC and 55%, respectively. Then the primers sequences were placed into an nBLAST search to verify that there were no unspecific results. Selectivity, specificity and sensitivity of the PCR primers were assessed, using plasmid DNA and RNA collected from HEK293 cells transfected with PAK1, PAK2,
PAK3, PAK4, PAK5, and PAK6 cDNA,
Real Time-PCR – RNA from thyroid cancer cell lines, HEK transfected with PAK
isoforms, human thyroid tumor tissue and normal tissue were treated with DNase I per the manufacture’s protocol. The DNA-free RNA was then reverse-transcribed using the
MultiScribeTM Reverse transcriptase per the manufacture’s protocol on an Applied
Biosystems 2720 Thermal Cycler. Gene expression for each PAK isoform was calculated
by qualitative real time-PCR, using the ABI SYBR Green PCR core reagents on a
Stratagene Mx3000P quantitative PCR system. 18s was used as an input control.
89 Migration assays - Migration assays were performed as described (69,324). NPA cells
were grown in RPMI 1640, containing 10% FBS. After 24 hours, the medium was
aspirated, and RPMI 1640 containing 0.2% FBS was added for another 24 hours. The
cells were trypsinized for 2-5 min, washed, and resuspended in RPMI-1640 containing
0.2% FBS. The cell concentration was calculated by hemocytometer. 400 μl of 0.2% FBS
medium was added into 24-well plate wells, and a Boyden chamber (8 μm pore) was
inserted into each well. 9 x105 NPA cells were added to each insert, and the NPA cells
o were allowed to attach in an incubator at 37 C 5% CO2 for 30 minutes. The inserts were
switched to a new well containing RPMI 1640 containing 5% FBS. The cells were
visualized by Diff-Quick® staining before and after swiping the top of the membranes, allowing for determination of total cells and the number of migrated cells (cells on the bottom of the membrane). Experiments were performed on at least two occasions.
Statistical Analysis - All quantitative data are represented as mean plus/minus S.D.
Analysis was performed using MS EXCEL, T-test. Statistical significance was a P-value of <0.05.
3.3 Results
3.3.1 PAK Isoform Specific Primer Design
PAK1, PAK2 and PAK4 have been shown to be up-regulated in many forms of cancer
(313). In thyroid cancer, PAK isoform expression has yet to be investigated. Primers were designed that would specifically detect each PAK isoform. The cDNA sequences
from NCBI were aligned using Dialign2. Regions where the PAK isoforms share 90 minimal sequence similarity were chosen. Primers were designed using Primer Express
1.5, and a Tm around 60oC was calculated for each primer (See table 3.1). The primer
sequences were then compared to other human sequences to check for specificity using nBLAST from the NCBI. The results indicated no nonspecific targets. To further verify the specificity and selectivity of the primers, 100 pg of plasmid for each PAK isoform was placed into a PCR reaction. The primers amplified a band that corresponded to the proper amplicon size in the present of the correct plasmid and no band was detected in the presence of the incorrect plasmid (figure 3.1A). PAK4 primers did not amplify any band. The reason for this may be because the primers were designed for a region that was located before the start codon, which the plasmid lacked. PAK4 has six variants, and the primers were designed to detect all six variants. Other experiments with PC3 cells, which were shown to express endogenous PAK4, amplified the proper bands. Restriction enzyme digestion of the PCR reaction product further implicated the specificity of the primers (data not shown). Next, the primers were tested against RNA that was collected from HEK293 transiently transfected with each PAK isoform plasmid. In each case, the transfection led to a lower threshold cycle when compared to control transfected HEK293 cells that expressed endogenous PAK isoforms (figure 3.1B), thus reassuring the specificity and selectivity of the primers.
3.3.2 PAK Expression in Thyroid Cancer Cell Lines
To assess which PAK isoforms are present in thyroid cancer, we utilized the primers to analyze PAK expression in thyroid cancer cell lines. The RNA from NPA (papillary),
WRO (follicular) and ARO (anaplastic) was collected and assessed for PAK expression.
91 PAK1, PAK2, and PAK4 were expressed in all three cell lines. PAK3 and PAK5 were not able to be detected. Interestingly, PAK6 was expressed in WRO and ARO, but not in
NPA (Table 2).
The three cell lines were grown to 40% confluency, a concentration that has been shown to have the greatest levels of PAK activity (unpublished data). Protein was collected from each of the cell lines, and probed for all PAK isoforms using isoform specific antibodies. HEK293 cells that were individually transiently transfected with cDNAs for each PAK isoform were used as a control. PAK1, PAK2, and PAK4 protein expression was detected in all three cell lines (Table 2). PAK6 protein expression was not detected in WRO and ARO, where we were able to detect gene expression. PAK3 and
PAK5 protein expression was undetectable, which correlated well with the gene expression data.
3.3.3 Dominant Negative PAK Decrease NPA Cell Motility
PAK proteins have been implicated in the regulation of many proteins that are associated with cellular structure and movement, such as myosin (255), LIM kinase 1 &2 (258,269),
FAK (349-351), and vimentin (352-354). To investigate the importance of PAK function on thyroid cancer cell motility, we used NPA cells, due to their high degree of movement on a Boyden Chamber assay. The cells were initially transfected with either GFP-control or p21-binding inhibitory domain (PID), a peptide derived from the N-terminus that is able to suppress the activity of the class I full-length PAK kinases or PID-L107F, a variant that binds more weakly to PBD. All plasmids contained a GFP insert, and transfection efficiency was calculated by counting GFP-containing cell versus the number
92 total cells, which was ~20-25%. After transfection, the cells were pre-treated in a 0.2%
FBS media, and were then plated for migration. The cells were allowed to migrate for 24
hours, and were visualized by staining with DiffQuick®. After normalization to the
control or normalizing to PID-L107F, it appeared that the cells containing the PID had a
decrease in migration of about 35% and 20%, respectively (figure 3.2). These data
suggests that PAK protein expression may be pertinent to thyroid cancer metastasis.
3.3.4 PAK Expression in Human Papillary Thyroid Tumors
To determine if PAK expression is up-regulated in thyroid cancer, we received RNA and
protein from ten patients with papillary thyroid cancer and the corresponding normal thyroid tissue. PAK gene expression was determined using the qRT-PCR with the aforementioned primers, utilizing 18s as an internal control. For PAK1, 50% of the patients had an increase, while 40% of the patients had a decrease. Interestingly, PAK2 and PAK4 were increased in seven of the ten patients. The expression of PAK5 was minimal and out of detectable range. PAK3 and PAK6, which were not detected in the cell line data, were detected with modest fluorescence. Eight out of ten patients had decreased expression of the PAK3 gene and 6 out of 10 had decreased expression of the
PAK6 gene (Table 3.3).
The protein from each pair of patient samples was assayed for p-PAK thr423,
PAK1, PAK2, PAK3, and PAK4 expression, where GAPDH acted as an internal control.
Due to the amount of protein collected, PAK5 and PAK6 were not assayed. We found a good correlation to the gene expression data. Eight out of ten patients seemed to have an increase in PAK2 expression, where nine out of ten patients had an increase in PAK4
93 protein expression. PAK1 protein expression was increased in 4 samples and decreased in
six samples, which are similar to the gene expression results. PAK3, which was found in
the patient gene expression, was not detected (Table 3.3). The level of phosphorylated-
PAK thr423, an indicator of PAK activity, was compared to normal tissue; similar results
were seen as with PAK2, with seven out of ten patients having an increase in levels.
When the phospho-PAK expression was normalized to PAK2 expression, it seemed that
most PTCs had a stable or reduced ratio versus normal tissue. These results demonstrate
that PAK2, and PAK4 levels are increased in human thyroid cancers when compared to
normal thyroid tissue. PAK activity was also increase in most PTC, and that they are
more active than the normal tissue expression. However, because the ratio of total-to-
active PAK did not increase, it appears that the increase in PAK activity is due to an
increase PAK expression rather than constitutive activation.
3.4 Discussion
p21-activates kinases have been reported to increase proliferation, as well as evade
apoptosis and increase cell motility (5). Recently, PAK1 has been implicated in resistance to tamoxifen in estrogen receptor positive breast cancer (334,355,356). PAK1 expression has also been shown to increased in many forms of cancers; breast, colon, bladder, and
brain (313). PAK4 overexpression has been assessed in pancreatic cancer (313). PAK1
and PAK2 have been shown to prevent apoptosis, but under certain conditions, PAK2 can
promote apoptosis (262,263,290). PAK6 interacts with androgen-receptor to inhibit its
function (6,250,253). Lastly, PAK1, PAK2, and PAK4 have been shown to regulate cell
motility, through multiple proteins, such as LIMK 1 and 2(258), Merlin (257), Desmin
94 (256), and Myosin light chain kinase (255). Because of these data as well as its potential role as a central mode in thyroid cancer cell motility, we decided to look at PAK expression in thyroid cancer.
We found that certain PAK isoforms are expressed in all of the thyroid cancer cell lines, including PAK1, PAK2, and PAK4. PAK3 and PAK5, which are normally found in brain and neurons(6,160,232,273,275,276), were not found in the thyroid cell lines.
Interestingly, PAK6 mRNA levels were detected in WRO and ARO cells, but not in NPA cells. ARO and WRO have a similar rounded morphology and express PAK6, while
NPA, which has a more fibroblast shape, did not. We initially thought that PAK6 could regulate certain transcription factors that could lead to expression of proteins that are associated with EMT, but we were not able to find PAK6 protein expression in these samples, making this unlikely.
The importance of PAK in cell migration has been reported by many groups. We wanted to investigate if PAKs play an important role in thyroid cancer motility. A peptide that has been shown to inhibit class I PAKs (PAK1, PAK2, and PAK3) was utilized to inhibit PAK function, and its effects on migration were assayed. We demonstrated decrease NPA cell motility with the expression of this peptide. The data suggests that
PAK1 and PAK2 could have an integral role in cell motility, but the peptide doesn’t inhibit PAK4, PAK5, and PAK6. PAK4 was expressed in NPA, so we have not excluded the possibility that PAK4 could be a mechanism of thyroid cancer motility. Experiments using siRNA for each expressed isoform of PAK are currently under way.
Furthermore, the nuclear localization of PAK isoforms in breast cancer has shown a resistance to tamoxifen (334,355,356). PAK in the nucleus has been implicated in
95 transcription regulation (335,356,357). Thus we hope to investigate whether PAK nuclear
expression in thyroid cancer might affect either metastatic disease or the resistance to
iodine treatment by the regulation of the sodium-iodine symporter. Preliminary results suggest that a decrease in nuclear PAK1 increases cell motility (data not shown).
In human PTC samples, we found that PAK2 and PAK4 had an increase in expression when compared to normal tissue. This overexpression correlated with an increase in phosphorylated PAK suggests the overexpression might be functionally involved in thyroid cancer. Which isoform, or isoforms, is most important to thyroid cancer development, and its progression into metastatic disease, is currently under investigation. The increase in PAK activity provides another way to connect the similar effects that are seen in PTCs and FTCs. We are currently mapping out the possible interactions between these two forms of cancer that include PAK signaling pathways.
96
Table 3.1: PAK isoform specific primer design.
97
Figure 3.1 – PAK primer specificity and selectivey. A) 100 pg of plasmid with a PAK isoform specific insert was PCR for 50 cycles. Each primer set amplified only the corresponding vector, except for PAK4, whose primer location is upstream of the start codon, which is not part of the insert sequence. PAK4 primer specificity was later verified with HEK293 RNA. B) RNA from HEK293 transfected with each PAK isoform was collected, reverse transcribed and amplified by PCR (quantitatively). For each primer set, there was at lease two cycle difference compared to vector control RNA except for PAK4 samples.
98
Table 3.2: PAK isoform gene and protein expression in Human thyroid cancer cell lines.
99
Figure 3.2 – p21-inhibitory domain (PID) decreases NPA cell migration. NPA were transfected with either GFP-vector (control), wild type (WT)-PID, or mutated (L107F)-PID. Transfection efficiency was about 20-25%, assessed by GFP fluorescence. Presence of the PID decrease migration was compared to either ther control or to a PID with decrease binding affinity (L107F).
100
Table 3.3: Summary of PAK isoform gene and protein expression in Human Thyroid
Papillary carcinomas.
101
CHAPTER 4
p21-ACTIVATED KINASES INTERACT WITH B-RAF
4.1 Introduction
Papillary thyroid cancers have been demonstrated to have increased PI3K activity, through RET/PTC rearrangements. This in turn, leads to an increase in AKT and PDK-1 activity (339,358). Similarly RET/PTC rearrangements have been shown to activate RAS
(359) and RAF (21). Finally, B-RAF activating mutations are the most common genetic abnormality in PTC (20,21,25,39,44,45). Thus, both the PI3K and the RAS/RAF pathways are heavily implicated in tumorgenesis and metastasis in thyroid cancer
(20,23). The mechanism of how these two different pathways to respond similarly to regulate cancer cell behavior is being deduced. One of the points where these two pathways merge, and merits further investigation, is PAK. (Figure 4.1)
More than 25 substrates for the PAK family of kinases have been identified thus far, and even with more substrates yet to be identified, the importance of PAK proteins for both normal cellular function and cancer development is being clarified (5). PAK1 and PAK2 share an ~90% sequence homology in their kinase domains, which suggests that they would recognize similar substrates (6,224). PAK4 shares only an ~50% sequence homology to PAK1, and may not phosphorylate similar substrates (6). Human
102 papillary thyroid carcinomas were shown to overexpress PAK2 and PAK4 in ~50, 80 and
90 % of cases, respectively (data from chapter 3). PAK inhibition has been shown to
inhibit thyroid cancer motility in vitro (data from chapter 2 and 3). Because of this,
finding targets for PAK proteins uniquely involved in thyroid cancer may lead us to
better understand how signal transduction leads to both tumorgenesis and to cell motility
in this disease.
The mode for PAK activation is described in previous chapters. PDK-1 has been
shown to phosphorylate PAK at threonine 423 (157,160), a residue critical for maximal
kinase activity, and AKT can phosphorylate PAK at serine 21 which is believed to aid in
membrane localization, as well as increase the activation of PAK (157,160). Once PAK
proteins have been activated, there are numerous targets that are associated with
proliferation, evasion of apoptosis and increased cell motility. One key target of PAK is
RAF-1 (CRAF) at serine 338 which is one of the two critical residues (serine 338 and
tyrosine 341) for RAF activation and kinase activity (157,160). The phosphorylation of
RAF by PAK leads to increased RAF kinase activity in overexpressed conditions, as well
as typical physiological conditions (157). RAF-1 activity leads to enhanced MEK and
ERK phosphorylation (359). Recently, it has been shown that PAK phosphorylates serine
298 on MEK, which leads to autophosphorylation and the activation of MEK1 (278,279).
This may explain the high level of MEK activity even in the absence of RAS and RAF mutations reported in some thyroid cancer. Because of this, we hypothesized that PAK maybe responsible for some of the overlapping cellular responses that the PI3K/PDK-
1/AKT and RAS/RAF pathways generate, particularly in relation to cell motility.
103 PAKs recognize a specific phosphorylation sequence. Originally, some researchers had postulated that the PAK phosphorylation sequence may be –K-R-E-S/T- from identification of known PAK targets (5). But recent studies suggest the consensus target sequence is more complex. Shaw showed that a basic amino acid (usually arginine) is preferred in the -2 position, and at the -3 position is either an arginine or lysine (223).
Rennefahrt et al. confirmed these results and deduced even more about the consensus sequence (222). Their results suggested that PAK proteins prefer to phosphorylate serine over threonine, and do not phosphorylate tyrosine residues. Also, they discovered that for
PAK1, the preferred target sequence is –R-R-X-S/T-hy-(Y/phospho-Y)-, the preferred target sequence for PAK2 is –(R/K)-R-(R/K)-S/T-hy-(Y/phospho-Y)- and the preferred target sequence for PAK4 is –(R/K)-R-(R/K)-S/T-hy-(A>C>Y>V>S)-S- where the bold amino acids represents the site of PAK phosphorylation, and “hy” equals a hydrophobic residue(222). Many of the known targets for PAK contain features of the aforementioned sequence, such as p47PHOX (-RRNSVRF-) (360) and prolactin (-RRDSHK-)(361). But
other known targets for PAK do not mimic the proposed sequence, such as Caldesmon (-
VRNIKSNWE- and –TAGLKGVSSRINKE-) (362) and regulatory myosin light chain (-
KKRPQRATSNVFAM-)(298).
As mentioned above, PAK phosphorylates RAF-1, thus increasing RAF kinase
activity. Due to the large regions of sequence homology between all three RAF isoforms,
we hypothesized that B-RAF may also be phosphorylated by PAK. We are particularly
interested in this question due to the critical role of B-RAF in thyroid tumorgenesis and
cell biology. Here, we compared the sequences of all of the RAF isoforms. Furthermore,
104 we investigated whether PAK and B-RAF interact with each other. The results indicate
direct interactions occur, but the functional effects of this interaction remain unclear.
4.2 Materials and Methods
Reagents and Vectors – Anti-Myc (#2278), anti-PAK1 (#2602), anti-phospho-MEK1/2
(ser298) (#9128), anti-phospho-MEK1/2 (ser217/221) (#9121), anti-MEK1/2 (#9122),
and anti-phospho-AKT (Ser473) (#9277) antibodies were purchased from Cell Signaling
Technology (Beverly, MA). Anti-B-RAF antibody was obtained from Santa Cruz
Biotechnology (sc-95284, Santa Cruz, CA). Rabbit IgG (#12-370), mouse IgG (#12-371), and Protein G Agarose (#16-266) were obtained from Upstate (Chicago, IL). Anti-
GAPDH (#N8300-250) was obtained from Novus Biologicals (Littleton, CO). Expression vectors containing myc-tagged wild type PAK1 cDNA was the generous gift of Dr.
Jonathan Chernoff (Fox Chase Cancer Center, Philadelphia, PA) (224,231) and a constitutively active B-RAF (V600E) was a gift from Dr. James Fagin (187). All other reagents were obtained from Sigma-Aldrich, unless otherwise noted.
RAF Sequence analysis – The protein sequences for each RAF isoform was gathered
from the NCBI database: A-RAF (NP_001645), B-RAF (NP_004324), and C-RAF
(NP_002781). The B-RAF sequence was visually examined for either serine or threonine
residue, which had an arginine residue at its -2 position. Dialgin2 was utilized to
compare RAF sequences. The program was assessed via the internet from the following
address: http://bioweb.pasteur.fr/seqanal/interfaces/dianlign2.html.
105 Cell Culture – HEK293 were grown in DMEM medium (Invitrogen, Carlsbad, CA)
supplemented with 10% fetal bovine serum (FBS), 100 mM L-Glutamine, and 100 μM
Non-Essential Amino Acids (Invitrogen). Once the cells achieved 70% confluence, the cells were washed, trypsinized, and re-plated.
Protein Extraction and Immunoblotting – Protein extraction and immunoblotting were performed as previously described (69,324). After the cells were washed with ice-cold
PBS, the cells were collected, centrifuged for 5 min at 500x g, and washed with 500 μl
ice-cold TBS. Cell lysis buffer supplemented with 0.3 µM Okadaic acid and 1 µg/ml each
of aprotinin, pepstatin, leupeptin, and 20 mM 4-amidino-phenyl methane-sulfonyl
fluoride were added to the tubes, which were then incubated on ice for 15 min. The cells
were centrifuged at 16,000 x g for 10 min at 4°C and the supernatant was saved and
stored at -80°C. Protein concentrations were calculated using Micro BCA Protein Assay
Kit (Pierce).
Co-immunoprecipitation of B-RAF and PAK1
Protein G-agarose beads (Upstate) were washed with cell lysis buffer 3x. 300 μg of
protein lysate was diluted to a concentration of 1 μg/μl. The Protein G-agarose beads
were then incubated for 0.5 hours at 4oC. The solution was centrifuged at 14,000g at 4oC
for 10 minutes. The supernatant was collected and transferred to newly chilled
microcentrifuge tubes. 2 μg of either B-RAF antibody or mouse IgG was added, and
106 incubated overnight at 4oC. Then Protein G-agarose beads were added and incubated for
another 2 hours. The beads were collected by centrifuging 14,000g at 4oC for 30 seconds.
The pellet was washed 3x and analyzed by western blotting.
4.3 Results
4.3.1 Identification of Possible PAK Phosphorylation Sites on B-RAF
With similar phenotypes for follicular and papillary thyroid carcinomas, the identification
of potential targets for these enzymes which cross-talk between major pathways
associated with tumorgenesis is of particular interest. PAK1 and PAK3 have been shown
to phosphorylate C-RAF at a critical residue, serine 338 (157,160). However potential
phosphorylation sites for PAK in B-RAF are unknown. We therefore set out to identify
possible PAK phosphorylation sites in B-RAF. Using the principle given by Rennefahrt
et al., most PAK isoforms phosphorylate either serine or threonine, and have an arginine at the -2 position (222). The B-RAF sequence was retrieved from NCBI database
(NP_004324). The sequence was then scanned for serine and threonine residues which had an arginine located at the -2 position. Nine possible candidates were identified; serine
180, 274, 364, 428, 466, and 605 and threonine 241, 508, and 560 (figure 4.2). For
PAK1, PAK2 and PAK4, an arginine located at the -3 position was even more favorable.
Because of its presence suggests serine 446 as a possible site for PAK phosphorylation.
A comparison of the three RAF isoforms has shown a high sequence homology.
The most predicted phosphorylation sites in B-RAF were conserved with A-RAF
(NP_001645) and C-RAF (NP_002781) (figure 4.3). Only, three sites did not share a high sequence homology: serine 180 (located in the RAS binding domain), serine 364
107 (located in conserved region 2), and serine 428 (located between CR2 and kinase domain)
(246) (figure 4.2). All of the other possible phosphorylation sites had a high level of
sequence homology. Interestingly, PAK phosphorylates C-RAF at serine 338, whose
sequence is –QRDSSYYWE-, whereas the corresponding sequence in B-RAF is –
RRDSSDDWE-. As mentioned before, an arginine at the -3 position is an even more preferred target sequence for PAK. These alignments suggest that PAK may phosphorylate B-RAF.
4.3.2 PAK and B-RAF Co-Precipitate
To test whether PAK and B-RAF co-precipitate, HEK293 cells were transiently transfected with cDNA for either the wild type PAK1 or for mutated B-RAF V600E or both. Expression was monitored by western blot (figure 4.4A) and both B-RAF and
PAK1 protein levels were increased with transfection. The constitutively active B-RAF increased the phosphorylation of phospho-MEK1/2 (serine 217 and 221) at the B-RAF phosphorylation site as predicted. PAK protein expression was also increased, but did not increase phopho-MEK1/2 (serine 298), likely because there was no activating mutation in the PAK cDNA. Because PDK-1 and AKT both are upstream from PAK, phospho-AKT
(serine 473) was utilized to monitor stimulation of the PAK signaling pathway, and
GAPDH was used as a loading control.
Proteins were collected from transfected cells, and immunoprecipitated under non-denaturing conditions. Mouse Anti-B-RAF IgG was used to pull down B-RAF. Even with low immunoprecipitate efficiency, we were still able to detect PAK1 (figure 4.4B).
The transfection control lane showed minimal basal interaction, and when PAK1 was
108 transfected into HEK293 cell, there was no increase in co-immunoprecipitate. But when
B-RAF was transfected into HEK293, there was a dramatic increase in
immunoprecipitated PAK1, suggesting an interaction between B-RAF and PAK1.
4.4 Discussion
PAK regulates many proteins associated with cell motility and survival (5). Initially,
PAK was discovered to be a regulator of cell motility, but recently, PAK has also been
implicated in cell survival, mitosis and differentiation. Currently, there are over 25
known targets of PAK, ranging from the myosin light chain to BAD (184,230,237,273) to
vimentin (352-354). Knowing that PAK1 aids in the activation of C-RAF, we looked at
B-RAF as a possible target. B-RAF activity and/or expression have been shown to be
increased in many forms of cancer but has particular relevance for thyroid cancer (42). B-
RAF also regulates many proteins that are associated with both cell survival, and cell
motility, for example, MEK1/2 (278,279) and NF-κB (195).
We initially analyzed the B-RAF sequence for potential targets for PAK
phosphorylation. Using the assumption that the -2 and possibly the -3 position would
contain an arginine residue, we identified nine possible targets with either serine or
threonine residues as potential PAK phosphorylation sites. Only one, serine 446, had the
arginine located at both the -2 and the -3 position. Other studies have shown that arginine
is the most common and important factor for PAK substrate reorganization But some
studies, have also shown that PAK can phosphorylate other substrates as well, which do
not adhere to the aforementioned assumptions. Even though the initial screening only
generated nine possible residues, other phosphorylation sequences/sites will be
109 investigated at a later time. Using a radioactive PAK kinase, we will detect the
incorporation of 32P to B-RAF. B-RAF mutants are being developed to identify which of
the nine possible PAK phosphorylation sites are acted upon by PAK.
We identified one phosphorylation site of particular interest: serine 446. Serine
446 of B-RAF corresponds to C-RAF serine 338, which is phosphorylated by PAK1 in
vitro and in vivo. Phosphorylation of serine 338 is a critical step in the activation of C-
RAF. The 3-dimenisonal structures of the kinase domains (CR3) of B-RAF and C-RAF,
where the critical serine residue is found, are very similar. This would suggest that their
kinase activity is regulated by similar methods of activation. The surrounding sequences
for B-RAF were similar to C-RAF except at the +2 and +3 position. C-RAF contains two
tyrosine residues, whose phosphorylation status upon PAK phosphorylation is currently
unknown, whereas B-RAF contains two aspartic acid residues. Rennefahrt et al. reported
that PAK1 and PAK2 could recognize, and sometimes prefer, a substrate containing
phospho-tyrosine located at the +2 position (222). Furthermore, the +3 position effect on
substrate recognition is less important. The importance of these differences is currently,
under investigation.
Serine 364, unique to B-RAF, is phosphorylated by SGK, which inhibits B-RAF
kinase activity. It was reported that this single phosphorylation was able to inhibit B-RAF
(173). This unique site suggests a specific way of inhibiting B-RAF compared with the other RAF isoforms. Serine 364 is one of the proposed PAK phosphorylation sites. But we hypothesized that phosphorylation by PAK would lead to increase RAF kinase function. If PAK phosphorylates this site, then PAK may also negatively regulate B-RAF activity.
110 Lastly, we assessed the physical interaction between PAK and B-RAF (Figure
4.4). We initially used a constitutively active B-RAF cDNA (V600E), allowing B-RAF to assume an open conformation, thus limiting B-RAF’s dependence on certain cellular conditions, such as RAS activation. In this experiment, the increase in B-RAF protein expression, correlated to an increase in co-immunoprecipitated PAK. One possible reason for this result is B-RAF acts upon PAK. To date, there is not any information suggesting that the RAF kinases phosphorylate or regulate PAK kinases. If PAK could
phosphorylate serine 364 and B-RAF could regulate PAK activity, this would suggest a
negative feedback loop into B-RAF activation.
111
Figure 4.1 – PAK interacts with B-RAF. PAK was shown to interact with C- RAF and phosphorylate serine 338. After comparison of B-RAF and C-RAF, the corresponding phosphorylation site in B-RAF has high sequence homology. It was proposed that it is possible that PAK phosphorylates B-RAF, aiding in B-RAF activation, as seen with C-RAF. We discovered that PAK does interact with B- RAF.
112 Figure 4.2 Possible PAK phosphorylation sites for B-RAF. The protein sequence of B-RAF was retrieved from the NCBI (NP_004324) and serine (S) or threonine (T) that had an arginine (R) residue located -2 position were considered possible targets (red). The corresponding C-RAF site was also found to be a possible target (blue).
113
Figure 4.3 Comparison of RAF isoforms. The RAF isoforms’ protein sequence was retrieved from the NCBI: C-RAF (NP_002781), A-RAF (NP_001645) and B- RAF (NP_004324). They were aligned using Dialign2. The red arrows correspond to possible PAK phosphorylation sites, and blue arrow represents the known C- RAF phosphorylation site.
114
Figure 4.4 PAK and B-RAF co-immunoprecipitate together. Either wild type PAK-MYC tagged or constitutively active B-RAF-MYC tagged (BRAF) were transfected into HEK293 cells individually or together. A) Immunoblotting shows that the proteins are present and active. B) B-RAF was precipitated with anti-B- RAF antibody and the proteins were assessed by immunoblotting, showing that PAK was present when B-RAF was IP.
115
CHAPTER 5
HDAC42 DECREASES THYROID CANCER PROLIFERATION
5.1 Introduction
Multiple methodologies have been implemented in the prevention of cancer metastasis.
Most are designed to inhibit pathways that have a direct effect on cell motility (such as
actin filament formation and microtubule assembly) through the use of small molecule
inhibitors (20,363-365). But few have used epigenetic drugs as a mean to prevent
metastasis. Protein production is a highly controlled process: DNAÆRNAÆprotein. The
transcription of DNA to RNA is regulated by a number of different mechanisms,
including the methylation of CpG islands, the activation/deactivation of transcription
factors, and the acetylation of histones, just to mention a few (366,367). The function of
histones is to organize and regulate DNA. When a histone is acetylated on key lysine
residues (the ε-amino group), the lysines lose their positive charge, and are no longer able
to tightly hold the negatively charged DNA backbone (368). This leaves the DNA in an
open conformation for transcription factor binding (369). After HDACs have removed
the acetyl-groups from the histones, the DNA binds tightly to those histones, silencing
transcription (368). Histone acetyltransferase (HAT) adds acetyl-groups to histones,
116 whereas histone deacetylase (HDAC) remove the acetyl-group from histones. HDAC has other properties and has been shown to directly interact with key proteins implicated in cancer, such as with p53 and NF-κB (370) (371).
HDACs are typically divided into three classes. Class I HDACs are homologues of yeast RPD3, and consist of HDAC 1, 2, 3 and 8 (372,373). Class II HDACs are similar to yeast Hda1 and consists of HDAC 4, 5, 6, 7, 9 and 10. Their molecular weight is about twice the size of Class I HDACs. Both Class I and Class II HDACs utilize a zinc divalent mechanism (372,373). The Class III HDACs are homologues of yeast Sir2 and are comprised of Sirt1-Sirt7 (372). The Class III Sirts use NAD+ instead of zinc ion to remove the acetyl-group from their target protein. In thyroid cancer, SIRT7 (a Class III
HDAC) was found to be up-regulated in thyroid carcinomas, but not in thyroid adenomas
(374,375).
Non-histone targets of HAT/HDAC regulation have also been described; some of these are involved in cell motility. Studies have shown that α-tubulin is acetylated on lysine 40, regulating cell motility and cell structure by stabilizing microtubules (376,377).
HDAC6 was also shown to specifically deacetylate tubulin and not histone (378).
HDAC6 and SIRT2 remove the acetyl group from the lysine 40 (379), and the overexpression of HDAC6 causes the increased migration of NIH-3T3 cells (380).
Trichostatin A, a histone deacetylase inhibitor (HDACi), inhibits both actin filament formation and reorganization, which leads to the decreased migration of hepatic stellate cells (381). These data suggest an important role in cell motility.
HDAC function has been shown to correlate with cancer invasion. For example, in gastric cancer, histone 4 is hypo-acetylated when compared to normal tissue, and 117 displays even lower levels of acetylation in those areas of invasion (382). Functionally,
HDAC6 was shown to regulate cell motility by decreasing the level of acetyl-cortactin.
Cells that had more hypo-acetylated cortactin moved slower, than when cortactin was
hyper-acetylated. It was hypothesized that when cortactin was charged, it bound more
strongly to F-actin (383). HDAC6 was also shown to regulate cell motility by decreasing
focal adhesion turnover (384). For these reasons, inhibition of HDAC could help prevent
cancer metastasis, by reducing cell motility.
Histone deacetylase inhibitors (HDACi) increase not only the quantity of
acetylated proteins (namely histones) but also other proteins, as well as transcription
factors and structural proteins. HDACi can be divided into multiple classes, which
include both natural and synthetic compounds. There are short-chain fatty acids, which
include valproic acid and AN-9, there are there are cyclic and non-cyclic hydroxamates,
which include SAHA and TSA, and there are cyclic peptides (or tetrapeptides), which
include FK228, benzamides, ketones (370). HDACi’s have been shown to induce cell
cycle arrest, and inhibit growth, by increasing expression of p21WAF1/CIP1 and p27KIP1 while inhibiting cyclins A and D (385). HDACi can also cause apoptosis. For example, the acetylation of Ku70 causes cancer cells to be more sensitive to apoptosis (386).
Several studies have begun to assess the effects of HDACi on motility. One specific
HDACi, valproic acid, inhibits cell motility in bladder cancer, but did not prevent invasion in the prostate cancer cell lines DU-145 and PC-3 (387). Uchida et al. showed that HDAC inhibitors led to increased cell motility in human endometrial
118 adenocarcinoma cell lines (388). With three different types of cells yielding different
responses to cell migration, these results indicate that the exact mechanism for the control
of cell motility is poorly understood.
In many types of cancer, HDACi have shown promise as a possible form of
treatment. HDACi’s have been shown to cause cell cycle arrest, cytodifferentiation, and
apoptosis. As mentioned above, some forms of thyroid cancers have been shown to
increase SIRT7 expression. Treatment of thyroid cancer with HDACi has already shown
some promise. For example, depsipeptide (FR901228) and TSA both increased the sodium-iodine symporter expression in anaplastic thyroid carcinoma cell lines, further
sensitizing those cell lines to radioactive iodine treatment (389,390). Also, treatment with
TSA caused increased cell cycle arrest and apoptosis in anaplastic thyroid carcinomas
(391). One study showed that this was due to a decreased expression of Cyclin A, as well as activation of the intrinsic apoptotic pathway (392). The expression of p21CIP/WAF1 increased in the follicular cell lines WRO and FRO when treated with depsipeptide (393).
Depsipeptide also increased p53 expression in the anaplastic cancer cell line SW-1736
(394). The effects of TSA, VPA, and depsipeptide on cellular migration have not yet been assessed. These studies, and their implications, provide the bases to test whether
HDAC42, a novel HDACi, would be effective in thyroid cancer treatment.
Currently, a number of groups are working to develop more potent HDACi. One study, in particular, helped in the development of more potent HDACi. This study uncovered a distinctive mechanism for how TSA and SAHA bind to a bacterial HDAC homologue (395). The catalytic site displayed a tube-like pocket with a Zn2+ cation at its
end, located about 10Å from the opening. The Zn2+ cation was coordinated with two His-
119 Asp charged relay systems, which aided the deacetylation mechanism (395). This structure, complexed with TSA and SAHA, allow the HDACi to be characterized into three motifs (Zn2+-chelating region, an aliphatic chain, and the polar planar cap), which interacted with specific regions in the binding pocket. This discovery has led to the development of more potent HDAC inhibitors. Lu, Q. et al. used this discovery to develop HDAC42, a new Zn2+-chelating containing molecule attached to a polar cap via an aromatic linker (Figure 5.1) (396,397).
Histone deacetylase inhibitors (HDACi) have become a much more widely studied form of treatment for many different types of cancer in clinical trials (398). The results from these clinical trials suggest that treatment with HDACi may lead to tumor regression in patients who are in an advanced stage of the disease, with minimal side- effects. Because of this, and given the aforementioned study by Finnin et al. (395), we hypothesized that HDAC42 might inhibit thyroid cancer cell line proliferation and migration. Our study found that HDAC42 decreased thyroid cancer proliferation in cell lines by inducing apoptosis. We also found that HDAC42 decreased total protein expression of AKT and PAK at 48 hours. Interestingly, phospho-AKT was also decreases, whereas phospho-PAK was actually increased, in a dose-dependent manner.
Lastly, we found that HDAC42 did not alter thyroid cancer migration in vitro, using a
Boyden chamber assay. These results suggest that HDAC42 could possibly be used as a form of treatment for thyroid cancer proliferation, but that inhibition of migration would require combination therapy.
120 5.2 Materials and Methods
Reagents and Vectors – Anti-AKT (#9272), anti-phospho-AKT (Ser473) (#9277), anti-
PAK1 (#2602) and anti-phospho-PAK (thr423) (#2601) antibodies were purchased from
Cell Signaling Technology (Beverly, MA). Anti-β-tubulin antibody was obtained from
Santa Cruz Biotechnology (sc-9401, Santa Cruz, CA). Anti-acetyl-Histone 4 antibody
was obtained from Upstate (Lys5, 8, 12 and 16) (#06-598, Chicago, IL), anti-β-actin
(#691001) was obtained from Medical & Biological Laboratories (Woburn, MA), and
anti-GAPDH (#N8300-250) was obtained from Novus Biologicals (Littleton, CO). All
other reagents were obtained from Sigma-Aldrich, unless otherwise noted.
Cell Culture - Human thyroid carcinoma NPA (papillary) (322), WRO (follicular) (323)
and ARO (anaplastic) (322) cell lines were obtained from Dr. Guy Juillard (University of
California, Los Angeles, Los Angeles, CA). They were grown in a RPMI 1640 medium
(Invitrogen, Carlsbad, CA) supplemented with 10% fetal bovine serum (FBS), 100 mM
L-Glutamine, and 100 μM Non-Essential Amino Acids (Invitrogen). Once the cells
achieved 70% confluence, the cells were washed, trypsinized, and re-plated. The growth medium was replaced with RPMI 1640, containing either 0.2% or 5% FBS (as noted for individual experiments), for 24 hours. Afterwards, the medium was aspirated and
replaced with fresh RPMI 1640, containing the same FBS concentration as before, and containing either HDAC42 or vehicle.
Cell Proliferation - NPA, WRO and ARO cells were seeded into 24-well plates at
~50,000 cells/well in RPMI 1640 containing 10% FBS. After the 24-hour attachment
121 period, cells were treated with the indicated concentration of either HDAC42 or DMSO
vehicle in RPMI 1640 containing 5% FBS. At different time intervals, cells were
harvested by trypsinization, and counted using a Coulter counter model Z1 D/T (Beckman
Coulter, Fullerton, CA). Triplicates were performed in all experiments, and experiments
were performed on three separate occasions. The concentration of HDAC42 that was required to inhibit growth by 50% (GI50) was calculated using the guidelines suggested
by the NIH.
Cell Viability Analysis - The effect of HDAC42 on cell viability was assessed using the
3-(4, 5-dimethylthiazol-2-yl)-2, 5-diphenyl-2H-tetrazolium bromide (MTT) assay.
Triplicates were performed in each individual experiment, and those experiments were
repeated on at least three separate occasions. Cells were grown in 96-well plates for 24 hours, and were exposed to various concentrations of HDAC42 dissolved in DMSO
(final concentration 0.1%) in RPMI-1640 containing 5% FBS. The medium was then
removed, and replaced by 200 µl of 0.5 mg/ml of MTT in RPMI 1640 with 10% FBS, as
per the manufacturer’s recommended protocol (T.C.I. America, Portland OR). The cells
were then incubated in a CO2 incubator at 37°C for 2 hours. Supernatants were removed from the wells, and the reduced MTT dye was solubilized in 200 µl/well DMSO.
Absorbance at 570 nm was determined on a plate reader. The inhibitory concentration
(IC50) was calculated using CalcuSyn (Ver. 1.2 Biosoft).
Protein Extraction and Immunoblotting – Protein extraction and immunoblotting were
performed as previously described in detail (69,324). After the cells were washed with
ice-cold PBS, they were collected, centrifuged for 5 min at 500x g, and then washed with
122 500 μl ice-cold TBS. Either cell lysis or M-PER buffer (Pierce Biotechnology, Inc.,
Rockford, IL), supplemented with 0.3 µM Okadaic acid and 1 µg/ml each of aprotinin, pepstatin, leupeptin, and 20 mM 4-amidino-phenyl methane-sulfonyl fluoride, were added to the tubes, which were then incubated on ice for 15 min. The cells were
centrifuged at 16,000 x g for 10 min at 4°C, and the supernatant was saved and stored at -
80°C. Protein concentrations were calculated using Micro BCA Protein Assay Kit
(Pierce).
Apoptosis Analysis - To assess apoptosis, a Cell Death Detection ELISA kit (Roche
Diagnostics) was used for nucleosome detection. The ELISA was performed according to
the manufacturer’s instructions. In brief, 8 x 105 NPA, WRO and ARO cells were
cultured in a T-25 flask for 24 hours before treatment. Cells were treated with either the
DMSO vehicle or HDAC42 at the indicated concentrations in RPMI 1640 containing 5%
FBS for 24 hours. They were then collected, and cell lysates equivalent to 1 x 104 cells
were used in the ELISA. Triplicates were performed in all experiments.
Migration assays - Migration assays were performed as previously described in detail
(69,324). NPA cells were grown in RPMI 1640 containing 10% FBS. For the 24 hour
data, the medium was aspirated, and RPMI 1640 containing 0.2% FBS was added for 24
hours. For the 48 hour data, either HDAC42 or vehicle was added to the 0.2% FBS
media. The cells were trypsinized for 2-5 min, washed, and resuspended in RPMI-1640
containing 0.2% FBS. The cell concentration was calculated by hemocytometer. 400 μl
of 0.2% FBS medium was added to 24-well plate wells, and a Boyden chamber (8 μm
pore) was inserted into each well. 9 x105 NPA cells were added to each insert, and the 123 o NPA cells were allowed to attach while incubating at 37 C 5% CO2 for 30 minutes. The
inserts were switched to a new well containing RPMI 1640 containing 5% FBS. Either
HDAC42 or vehicle was added into the upper chamber, and the chamber was incubated
for 24 hours. Cells were visualized by Crystal Violet staining, before and after swiping
the top of the membranes, allowing for determination of total cells and of the number of
cells on the bottom of the membrane, as previously described. Experiments were
performed on at least two occasions.
Statistical Analysis - All quantitative data is represented as a mean plus/minus S.D.
Analysis performed using MS EXCEL. The significance of the variations between groups
was examined using a Student’s T-Test. Statistical significance had a P-value of <0.05.
5.3 Results
5.3.1 Effect of Histone Deacetylase Inhibitor (HDAC42) on Thyroid Cancer Cell
Viability
HDACi’s have been shown to decrease cell proliferation in multiple cell lines. To test
whether HDAC42 might be able to inhibit NPA (papillary thyroid carcinoma), WRO
(follicular thyroid carcinoma) and ARO (anaplastic thyroid carcinoma), these cell lines
were plated and treated. Periodically, the cells were counted over a three days. The
results indicated that HDAC42, during the first 24 hours, was not cytotoxic. The cell
number didn’t decrease below the plated amount for all three cell lines. After another 24
hours, concentrations greater than 1 μM appeared to be cytotoxic (figure 5.2A).
Calculated GI50 and GI0 were 373.8 ± 107.3 and 922.1 ± 73.5 for NPA, 428.2 ± 98.2 and
124 1040.0 ± 88.1 for WRO, 340.7 ± 10.3 and 829.2 ± 192.2 nM for ARO, respectively.
Interestingly, cell viability, which was assessed by mitochondrial activity, suggested that
at 1 μM, there was a decrease in proliferation by ~40%, ~20% and ~30% for ARO, NPA
and WRO, respectively (figure 5.2B). The IC50 for 48 hours were 541.5 ± 214.7 nM for
NPA, 548.7 ± 149.0 nM for WRO, and 535.9 ± 72.7 nM for ARO.
The discrepancy between the cell counting data (showing minimal change at 24
hours), and the decrease in mitochondrial activity with the increase in drug treatment
(again at 24 hours), was thought to be potentially caused by mitochondrial instability.
This was due to the apoptotic signaling pathway, which caused cytochrome C to release.
WRO and NPA were utilized to assess whether the cytotoxic effect was due to apoptosis.
One of the markers of apoptosis is the cleavage of the polynucleosome into mono- and
oligo-nucleosome fragments. Using an ELISA design to detect the cleaved nucleosome,
WRO and NPA cells were treated for 48 hours, and histone cores were collected per the manufactors’ protocol. Preliminary results indicated that, as HDAC42 there is a dose- dependent raise in nucleosome fragmentation (figure 5.3). These data, along with the cell viability, indicates that HDAC42 prevents cell proliferation at least in part by initiating apoptosis.
5.3.2 HDAC42 Regulates AKT and PAK1 Protein Expression
HDAC inhibitors have been previously shown to control protein activity, either through the alteration of protein expression, or though the expression in regulators of certain pathways. For example, in one study, 293 T cells were treated with TSA, which up- regulated PTEN expression, thus leading to a reduced level of activation for AKT (399).
125 To determine whether HDAC42 can specifically increase acetyl-histone 4 in a dose
dependent manner, we treated NPA, WRO and ARO cells with HDAC42 for 24 hours,
and the proteins were collected. Antibodies specific for acetylated histone 4 were used,
and with the increase in HDAC42, there was a concurrent increase in acetyl-histone
expression (figure 5.4A).
In many cell lines, a link has been proven between the increase in expression of
those proteins that are associated with migration and the development of metastasis.
Using NPA cells, due to of their ability to migrate in vitro, cells were treated with
HDAC42 for 24, 48 and 72 hours. Proteins were collected, and the expressions of both
AKT and PAK were assessed. At 24 hours, both the AKT and PAK protein levels remained unaltered, but at 48 hours (at 500 nM), a noticeable decrease in total AKT and
PAK expression was detected. AKT activity was assessed by probing for phospho-AKT
at serine 473. At 24 hours phospho-AKT levels were inhibited at 500 nM and the effect
was still present at 48 hours (figure 5.4B). Because the activation of PAK requires the
phosphorylation of threonine 423 for maximal activity, we used an antibody specific for
that site, and found that treatment with HDAC42 led to a dose-dependent increase in phospho-PAK at both 24 and 48 hours. These results suggest that HDAC42 down- regulates the expression of both AKT and PAK1 after 24 hours. Interestingly, phospho-
AKT was reduced at 48 hours, whereas phospho-PAK was increased, which suggests that
the regulation of these proteins may be uniquely affected by HDAC42 treatment.
126 5.3.3 HDAC42 Has a Modest Affect on Thyroid Cancer Cell Migration in vitro
The importance of AKT in thyroid cancer migration is well documented. Saji et al. have
shown that the inhibition of AKT1 by siRNA leads to a decrease migration of NPA cells in vitro (324). In previous chapters, we have shown the importance of PAK on NPA cell migration. Since active AKT was reduced at 24 hours, it was expected that NPA migration would also drop at 24 hours. Interestingly, NPA migration did not decrease as expected with HDAC42 treatment at 24 hours (figure 5.5A). To assess the effect of
HDAC42 on migration at 48 hours, the cells were treated with HDAC42 for 24 hours, and then plated into a Boyden chamber and migrated for 24 hours (equaling 48 hours of total treatment). The results indicated that at 1 and 2.5 μM of HDAC42, migration decreased by ~20% and ~30%, respectively (Figure 5.5B). The protein levels of PAK and
AKT was reduced at 48 hours, which would suggest a reduction in NPA migration.
Interestingly, PAK activity was elevated with HDAC42 at both 24 and 48 hours, where there wasn’t any inhibition of migration. These data suggests a possible restrictive role for PAK in the effects of HDAC42 on NPA cell migration
5.4 Discussion
HDAC inhibitors have been shown to keep histone deacetylase from removing the acetyl
group that is attached to lysine on histones, thus increasing gene expression. Some of the
proteins whose expressions are increased in thyroid cancer treatment with various
HDACi’s include p53 (a tumor suppressor protein)(394), sodium-iodine symporter
(389,390), and p21CIP/WAF1 (a cyclin dependent kinase inhibitor) (393). The effects of
HDAC42 (a new Zn-chelating containing molecule attached to a polar cap via an
127 aromatic linker), on both thyroid cancer cell viability and motility were assessed.
Treatment with HDAC42 led to a reduction in cell proliferation after 24 hours, but did not become cytotoxic until after 24 hours (as assessed by cell numeration). Cell viability, assessed by mitochondrial activity, revealed that at concentrations less than or equal to 1
μM, decreased activity was present at 24 hours and continued to decrease at later time points.
HDAC inhibitors have also been shown to increase pro-apoptotic Bcl-2 family proteins, which can increase apoptosis in many cell lines (400). Some of the initiating events in apoptosis include Caspase activation and PARP deactivation, cytochrome C release (caused by pro-apoptotic Bcl-2 family proteins), and ultimately the degradation of the nucleosome. Cleavage of the nucleosome caused by HDAC42 was shown to be dose- dependent. Currently, we are investigating possible mechanisms that would allow
HDAC42 to cause apoptosis. One pathway of interest is the Bcl-2 family, due to recent reports that HDAC42 may increase both BAD and BIM expression, which in turn would lead to decreased mitochondrial activity (401,402).
AKT1 in thyroid cancer cells is thought to regulate cell motility, whereas AKT2 and AKT3 seem to have minimal influence (324). In NPA cells, HDAC42 decreased active AKT at 24 and 48 hours, but the cells are still able to migrate in a Boyden chamber assay at 24 hours, and were minimally inhibited at 48 hours. This suggests that other pathways may be more instrumental in regulating the reorganization of the cytoskeleton.
One pathway associated with cytoskeleton reorganization is PAK. With NPA cells treated with HDAC42, phospho-PAK at threonine 423 was increased at 24 and 48 hours in a dose-dependent manner. This increase occurred despite a reduction of total PAK levels at
128 48 hours. It is believed that the increase in active PAK may lead to the inability of
HDAC42 to inhibit NPA cell migration. Since phospho-PAK was increased, it might be beneficial to inhibit PAK activity along with HDAC inhibition. HDAC42 caused an increase in phospho-PAK at threonine 423 in a dose-dependent manner, suggesting that a possible combination study with a PDK-1 inhibitor (such as OSU-03012) could hinder the increased phospho-PAK response caused by HDAC42 treatment. Also, the notion that other proteins, and not PAK, may factor into HDAC42’s inability to inhibit NPA migration is being considered.
HDAC inhibitors affect a number of different proteins, making it more difficult to find a specific mechanism for their inhibition of migration and proliferation. AKT requires PDK-1 to phosphorylate threonine 308, and PDK-2 to phosphorylate serine 473.
Some researchers theorize that the phosphorylation of threonine 308 a prerequisite of the phosphorylation of serine 473, so it is possible that the reduction of serine 473 is due to the regulation of threonine 308 phosphorylation. We are currently trying to deduce whether PDK-1 activity and expression is affected by HDAC42 treatment. However
PAK, which also contains a PDK-1 phosphorylation site (threonine 423) actually increased with HDAC42 treatment, suggesting that the PDK-1 activity was normal.
Another mechanism that might explain this discrepancy comes from the regulation of
AKT and PAK by phosphatases. AKT is regulated by PHLPP (403), whereas PAK is regulated by both POPX1 and POPX2 (241). Since AKT and PAK are regulated by different phosphatases, it would be interesting to see whether the expression of these phosphatases, and their activity when treated with HDAC42, would be altered.
129 Figure 5.1 - Structure of HDAC42
130
Figure 5.2 - HDAC42 is cytotoxic in thyroid cancer cell lines. A) NPA, WRO and ARO cells were treated with increasing doses of HDAC42 for 0, 24, 48, and 72h, and then counted. HDAC42 caused a decrease in cell numbers in each cell line after 48 hours. The Y-axis represents arbitrary units with the number of cells normalized to day 0 (pre-treatment), which is set at 1.0. Mean values and standard deviations are shown. B) MTT assay for mitochondrial activity showed similar results.
131
Figure 5.3 - HDAC42 induces apoptosis in thyroid cancer cell lines. NPA and WRO cells were treated with HDAC42 for 48 hours. The cell lysates were collected and probed for cleaved nucleosomes by Roche Cell Death ELISA ®. As HDAC42 increased in concentration, there was a concurrent increase in nucleosome cleaved signal.
132
Figure 5.4 - HDAC42 causes an increase in the acetylation of Histone 4, and alters both AKT and PAK protein expression. A) NPA, WRO and ARO cells were treated with HDAC42 for 24 hours. Protein was collected and probed for acetyl-histone 4, which increased in a dose-dependent manner. B) NPA cells were treated for 24, 48 and 72 hours, and then probed for the expression/activation of key proteins associated with migration. Loading was monitored by poncues S staining. At 24 hours, AKT and PAK remained unaltered, but AKT activation was reduced whereas PAK activation was elevated. At 48 hours, AKT expression was reduced in a dose-dependent manner. Similar results were seen with total PAK. Again, AKT activation was diminished, whereas PAK activation was elevated, at 48 hours.
133
Figure 5.5 - HDAC42 does not NPA cell inhibit migration. A) NPA cells were placed into a Boyden Chamber with HDAC42 for 24 hours. Treatment did not decrease migration. B) NPA cells were initially treated for 24 hours with HDAC42, and were then placed in a Boyden Chamber with HDAC42 for another 24 hour period (simulating 48 hours of treatment). Again, the treatment had a minimal effect on migration, with slight reduction at 2.5 μM.
134
CHAPTER 6
CONCLUSION AND FUTURE DIRECTIONS
A variety of initiating events can lead to the onset of cancer. Key events in thyroid cancer include gene rearrangements (RET/PTC and PPARγ/PAX8) (34,71-73,77,79,329,338), mutations that bring about either constitutively active enzymes (B-RAF (V600E))
(20,21,25,44,45)or inactive enzymes (PTEN-null) (62,66,337,346), increase gene and protein expression (AKT and PAK)(22,23,66), and deregulation of the cell cycle (p53 and p27kip1)(26,322,328). In thyroid cells, these events can lead to the development of several different forms of cancer: papillary, follicular, medullary and anaplastic carcinomas. Most forms of thyroid cancer can be treated with surgery and radioiodine therapy, but for more aggressive types of thyroid cancers and other tumors, these methods fail to eradicate the disease (1). The key determinate for a poor prognosis is the development of local and distant metastasis.
Many have shown that AKT1 was important in thyroid cancer cell line motility, and that both AKT1 and AKT2 are increased in thyroid carcinomas when compared with adjacent normal thyroid tissue (23,69,324). AKT is activated directly by PDK-1, through the phosphorylation of thr308 (100). It was believed that the inhibition of PDK-1 might lead to a decrease in cell motility. OSU-03012 was developed from celecoxib to be a
135 PDK-1 inhibitor, devoid of its COX-2 inhibition (4). OSU-03012 was shown to be
cytotoxic and induced apoptosis. Treatment with OSU-03012 led to a decrease in thyroid
cell motility, but at a dose concentration less than what was required to inhibit AKT
activation by PDK-1. These results suggested that OSU-03012 effect in inhibiting
migration was through a different mechanism. Further analysis of different PDK-1 targets suggested that OSU-03012 inhibited PAK directly, and at a dose that was comparable to
the effects on migration. In an in vitro kinase assay verified that OSU-03012 directly
inhibited PAK by competitively binding to the AKT-binding pocket. Better PAK inhibitors are currently being developed using OSU-03012 as a parent compound. Also,
other potential targets for OSU-03012, those particularly associated with migration, are
currently being assessed.
Because of the finding that PAK proteins play an important role in migration (5),
further analysis of PAK’s influence on the development of thyroid cancer was
investigated. Primers and antibodies designed to detect individual PAK isoforms were
utilized to assess which isoforms were present in thyroid cancer cell lines. The results
suggested that PAK1, PAK2 and PAK4 are present. Interestingly, PAK6 gene was found
to be expressed in the follicular and anaplastic cell line and not in the papillary cell line.
PAK3 and PAK5 were not detected. RNA and protein from both normal and cancerous
tissue was collected from 10 different patients with papillary thyroid carcinomas. The
data suggested that PAK2 and PAK4 were up-regulated in at least 7 cases, and that
phospho-PAK was increased in 7 out of 10 cases. Thus, it is believed that PAK2 and
PAK4 may be important in proliferation and migration of thyroid cancer. The
mechanisms for the up-regulation of both PAK2 and PAK4 are being studied. PAK1 was
136 up-regulated in 5 out of 10 cases, suggesting its role may not be as important as that of
PAK2 and PAK4. However, PAK2 and PAK4 shared many similar effectors as PAK1.
PAK3 gene expression was modestly detected, and 8 out of 10 cancers had a decrease in expression, yet protein expression was not detected. The gene expression of PAK5, which was barely detectable, and that of PAK6, which was modestly detectable, were both reduced when compared to normal tissue in 7 and 6 out of 10 cases, respectively.
The expression of PAK3 and of PAK5, are found mainly in neuronal tissue (6), and it wasn’t expected to be found in thyroid tissue. We are currently looking to determine if genes regulated by PAK6 expression are altered.
Due to overlapping effects from different signaling pathways, we investigated whether the RAS/B-RAF and PI3K/PDK-1 pathways had a point where they crosstalked.
Interestingly, PAK is known to phosphorylate a critical residue on C-RAF (serine 338)
(157,160), and scanning of possible phosphorylatable serine and threonine on B-RAF led to nine possible PAK phosphorylation sites. When the sequences of all the RAF isoforms were compared, serine 446 of B-RAF aligned with serine 338 of C-RAF. Experimentally,
PAK and B-RAF co-immunoprecipitated in vitro. Experiments are currently being planned to investigate whether PAK phosphorylates B-RAF or if B-RAF phosphorylates
PAK as well as to determine which site is being phosphorylated. The implication of this phosphorylation may suggest why we also see similar effects in cancers that lack B-RAF activating mutations.
HDAC inhibitors have been shown to alter the expression of key proteins. HDACi increased the expression of p21WAF1/CIP1, p27KIP1, BAD, BID, BAX, Apaf-1, caspase-3 and caspase-9, leading to an increase in apoptosis, as well as decreasing the expression of
137 Bcl-2 and Bcl-xL, and preventing their anti-apoptotic functions (400). With an increased
expression of PAK isoforms in thyroid cancer, it was postulated that an HDAC inhibitor
might alter the expression of PAK isoforms and alter thyroid cancer biology. Treatment of HDAC42 was cytotoxic, and induced apoptosis at 48 hours. At 24 hours, HDAC42 did not decrease AKT or PAK expression. Interestingly, while phospho-AKT was decreased, phospho-PAK levels increased, at 24 hours. This suggests that the expressions of certain phosphatases which regulate these proteins activities [PHLPP for AKT (403) and
POPX1/2 for PAK (241)] are not similar. Currently, the expressions of PAK, AKT,
PHLPP, POPX1 and POPX2 are being investigated. Again, phospho-AKT was decreased at 48 hours, while phospho-PAK remained elevated. Functionally, HDAC42 had a minimal effect on migration correlating with the increase in PAK activation at both 24 and 48 hours. These data suggest that PAK may have a dominant effect on migration in
NPA cells. We plan to combine HDAC42 with OSU-03012 and molecular inhibitors and study it synergistic effects due to the increase level of PAK activity.
In thyroid cancer, like all solid tumors, the development of metastasis leads to a poor prognosis. PAK, an integral protein in cell motility that also regulates both cell proliferation and apoptosis, was shown to be up-regulated in thyroid cancer as well as in breast, pancreatic and bladder cancers. PAK can interact with C-RAF to increases its activity, and has also been suggested to interact with B-RAF. For these reasons, we believe that the inhibition of PAK may aid the in prevention of metastasis. Inhibition of
PAK activation and activity by OSU-03012 decreased both cell proliferation and migration. HDAC42 decreased proliferation at 48 hours; however, HDAC42 could not
138 inhibit migration at either 24 or 48 hours. Phospho-PAK was increased with treatment at those time points. Ultimately, the whole of our data suggests that p21-activated kinases regulate thyroid cancer metastasis, and may represent a viable therapeutic target.
139
BIBLIOGRAPHY
1. Endocrine Education, I. (2007) Thyroid Disease Manager. In.
2. Kaverina, I., Rottner, K., Vignal, E., Anderson, K., Stradal, T., Nakagawa, H., Krylyshkina, O., Hahne, P., Resch, G., Magid, N., and Small, V. (2003) A Video Tour on Cell Motility. In.
3. Society, A. C. (2007)
4. Zhu, J., Huang, J. W., Tseng, P. H., Yang, Y. T., Fowble, J., Shiau, C. W., Shaw, Y. J., Kulp, S. K., and Chen, C. S. (2004) Cancer Res 64(12), 4309-4318
5. Bokoch, G. M. (2003) Annu Rev Biochem 72, 743-781
6. Jaffer, Z. M., and Chernoff, J. (2002) Int J Biochem Cell Biol 34(7), 713-717
7. Clinic., E. W. a. t. N. E. S. (2007) How Your Thyroid Works: "A delicate Feedback Mechanism". In.
8. Air, M., Roman, S. A., Yeo, H., Maser, C., Trapasso, T., Kinder, B., and Sosa, J. A. (2007) Thyroid 17(3), 259-265
9. Doolittle, R. F., Hunkapiller, M. W., Hood, L. E., Devare, S. G., Robbins, K. C., Aaronson, S. A., and Antoniades, H. N. (1983) Science 221(4607), 275-277
10. Waterfield, M. D., Scrace, G. T., Whittle, N., Stroobant, P., Johnsson, A., Wasteson, A., Westermark, B., Heldin, C. H., Huang, J. S., and Deuel, T. F. (1983) Nature 304(5921), 35-39
140 11. Kasai, K., Hiraiwa, M., Suzuki, Y., Emoto, T., Banba, N., Nakamura, T., and Shimoda, S. (1987) Acta Endocrinol (Copenh) 114(3), 396-401
12. Normanno, N., De Luca, A., Bianco, C., Strizzi, L., Mancino, M., Maiello, M. R., Carotenuto, A., De Feo, G., Caponigro, F., and Salomon, D. S. (2006) Gene 366(1), 2-16
13. Tanaka, K., Nagayama, Y., Nakano, T., Takamura, N., Namba, H., Fukada, S., Kuma, K., Yamashita, S., and Niwa, M. (1998) Endocrinology 139(3), 852-858
14. Yang, J., Symes, K., Mercola, M., and Schreiber, S. L. (1998) Curr Biol 8(1), 11- 18
15. Zhang, H., Bajraszewski, N., Wu, E., Wang, H., Moseman, A. P., Dabora, S. L., Griffin, J. D., and Kwiatkowski, D. J. (2007) J Clin Invest 117(3), 730-738
16. Viglietto, G., Maglione, D., Rambaldi, M., Cerutti, J., Romano, A., Trapasso, F., Fedele, M., Ippolito, P., Chiappetta, G., Botti, G., and et al. (1995) Oncogene 11(8), 1569-1579
17. Mitteldorf, C. A., Leite, K. R., Meirelles, M. I., Gattas, G. J., and Camara-Lopes, L. H. (2004) Acta Cytol 48(2), 199-206
18. Liu, Z., Falola, J., Zhu, X., Gu, Y., Kim, L. T., Sarosi, G. A., Anthony, T., and Nwariaku, F. E. (2004) J Clin Endocrinol Metab 89(7), 3503-3509
19. Uckun, F., Ozer, Z., and Vassilev, A. (2007) Br J Haematol 136(4), 574-589
20. Espinosa, A. V., Porchia, L., and Ringel, M. D. (2007) Br J Cancer 96(1), 16-20
21. Wojciechowska, K., and Lewinski, A. (2006) Endocr Regul 40(4), 129-138
22. Kada, F., Saji, M., and Ringel, M. D. (2004) Curr Drug Targets Immune Endocr Metabol Disord 4(3), 181-185
23. Shinohara, M., Chung, Y. J., Saji, M., and Ringel, M. D. (2007) Endocrinology 148(3), 942-947 141 24. Pestell, R. G., Albanese, C., Reutens, A. T., Segall, J. E., Lee, R. J., and Arnold, A. (1999) Endocr Rev 20(4), 501-534
25. Fukushima, T., and Takenoshita, S. (2005) Fukushima J Med Sci 51(2), 67-75
26. Duh, Q. Y., and Grossman, R. F. (1995) Surg Clin North Am 75(3), 421-437
27. Fusco, A., Grieco, M., Santoro, M., Berlingieri, M. T., Pilotti, S., Pierotti, M. A., Della Porta, G., and Vecchio, G. (1987) Nature 328(6126), 170-172
28. Grieco, M., Santoro, M., Berlingieri, M. T., Melillo, R. M., Donghi, R., Bongarzone, I., Pierotti, M. A., Della Porta, G., Fusco, A., and Vecchio, G. (1990) Cell 60(4), 557-563
29. Knowles, P. P., Murray-Rust, J., Kjaer, S., Scott, R. P., Hanrahan, S., Santoro, M., Ibanez, C. F., and McDonald, N. Q. (2006) J Biol Chem 281(44), 33577-33587
30. Takahashi, M. (1988) IARC Sci Publ (92), 189-197
31. Takahashi, M., Buma, Y., Iwamoto, T., Inaguma, Y., Ikeda, H., and Hiai, H. (1988) Oncogene 3(5), 571-578
32. Coulpier, M., Anders, J., and Ibanez, C. F. (2002) J Biol Chem 277(3), 1991-1999
33. Knauf, J. A., Kuroda, H., Basu, S., and Fagin, J. A. (2003) Oncogene 22(28), 4406-4412
34. Santoro, M., Dathan, N. A., Berlingieri, M. T., Bongarzone, I., Paulin, C., Grieco, M., Pierotti, M. A., Vecchio, G., and Fusco, A. (1994) Oncogene 9(2), 509-516
35. Santoro, M., Melillo, R. M., Grieco, M., Berlingieri, M. T., Vecchio, G., and Fusco, A. (1993) Cell Growth Differ 4(2), 77-84
36. Jhiang, S. M., Sagartz, J. E., Tong, Q., Parker-Thornburg, J., Capen, C. C., Cho, J. Y., Xing, S., and Ledent, C. (1996) Endocrinology 137(1), 375-378
142 37. De Vita, G., Zannini, M., Cirafici, A. M., Melillo, R. M., Di Lauro, R., Fusco, A., and Santoro, M. (1998) Cell Growth Differ 9(1), 97-103
38. Trapasso, F., Iuliano, R., Chiefari, E., Arturi, F., Stella, A., Filetti, S., Fusco, A., and Russo, D. (1999) Eur J Endocrinol 140(5), 447-451
39. Melillo, R. M., Castellone, M. D., Guarino, V., De Falco, V., Cirafici, A. M., Salvatore, G., Caiazzo, F., Basolo, F., Giannini, R., Kruhoffer, M., Orntoft, T., Fusco, A., and Santoro, M. (2005) J Clin Invest 115(4), 1068-1081
40. Rabes, H. M., Demidchik, E. P., Sidorow, J. D., Lengfelder, E., Beimfohr, C., Hoelzel, D., and Klugbauer, S. (2000) Clin Cancer Res 6(3), 1093-1103
41. Smida, J., Salassidis, K., Hieber, L., Zitzelsberger, H., Kellerer, A. M., Demidchik, E. P., Negele, T., Spelsberg, F., Lengfelder, E., Werner, M., and Bauchinger, M. (1999) Int J Cancer 80(1), 32-38
42. Davies, H., Bignell, G. R., Cox, C., Stephens, P., Edkins, S., Clegg, S., Teague, J., Woffendin, H., Garnett, M. J., Bottomley, W., Davis, N., Dicks, E., Ewing, R., Floyd, Y., Gray, K., Hall, S., Hawes, R., Hughes, J., Kosmidou, V., Menzies, A., Mould, C., Parker, A., Stevens, C., Watt, S., Hooper, S., Wilson, R., Jayatilake, H., Gusterson, B. A., Cooper, C., Shipley, J., Hargrave, D., Pritchard-Jones, K., Maitland, N., Chenevix-Trench, G., Riggins, G. J., Bigner, D. D., Palmieri, G., Cossu, A., Flanagan, A., Nicholson, A., Ho, J. W., Leung, S. Y., Yuen, S. T., Weber, B. L., Seigler, H. F., Darrow, T. L., Paterson, H., Marais, R., Marshall, C. J., Wooster, R., Stratton, M. R., and Futreal, P. A. (2002) Nature 417(6892), 949- 954
43. Wan, P. T., Garnett, M. J., Roe, S. M., Lee, S., Niculescu-Duvaz, D., Good, V. M., Jones, C. M., Marshall, C. J., Springer, C. J., Barford, D., and Marais, R. (2004) Cell 116(6), 855-867
44. Ciampi, R., and Nikiforov, Y. E. (2005) Endocr Pathol 16(3), 163-172
45. Trovisco, V., Vieira de Castro, I., Soares, P., Maximo, V., Silva, P., Magalhaes, J., Abrosimov, A., Guiu, X. M., and Sobrinho-Simoes, M. (2004) J Pathol 202(2), 247-251
143 46. Carta, C., Moretti, S., Passeri, L., Barbi, F., Avenia, N., Cavaliere, A., Monacelli, M., Macchiarulo, A., Santeusanio, F., Tartaglia, M., and Puxeddu, E. (2006) Clin Endocrinol (Oxf) 64(1), 105-109
47. Kumagai, A., Namba, H., Saenko, V. A., Ashizawa, K., Ohtsuru, A., Ito, M., Ishikawa, N., Sugino, K., Ito, K., Jeremiah, S., Thomas, G. A., Bogdanova, T. I., Tronko, M. D., Nagayasu, T., Shibata, Y., and Yamashita, S. (2004) J Clin Endocrinol Metab 89(9), 4280-4284
48. Lima, J., Trovisco, V., Soares, P., Maximo, V., Magalhaes, J., Salvatore, G., Santoro, M., Bogdanova, T., Tronko, M., Abrosimov, A., Jeremiah, S., Thomas, G., Williams, D., and Sobrinho-Simoes, M. (2004) J Clin Endocrinol Metab 89(9), 4267-4271
49. Riesco-Eizaguirre, G., Gutierrez-Martinez, P., Garcia-Cabezas, M. A., Nistal, M., and Santisteban, P. (2006) Endocr Relat Cancer 13(1), 257-269
50. Knauf, J. A., Ma, X., Smith, E. P., Zhang, L., Mitsutake, N., Liao, X. H., Refetoff, S., Nikiforov, Y. E., and Fagin, J. A. (2005) Cancer Res 65(10), 4238-4245
51. Ciampi, R., Knauf, J. A., Kerler, R., Gandhi, M., Zhu, Z., Nikiforova, M. N., Rabes, H. M., Fagin, J. A., and Nikiforov, Y. E. (2005) J Clin Invest 115(1), 94- 101
52. Ciampi, R., Zhu, Z., and Nikiforov, Y. E. (2005) Endocr Pathol 16(2), 99-105
53. Pommery, N., and Henichart, J. P. (2005) Mini Rev Med Chem 5(12), 1125-1132
54. Liu, W., Bagaitkar, J., and Watabe, K. (2007) Front Biosci 12, 4011-4019
55. Michl, P., and Downward, J. (2005) Z Gastroenterol 43(10), 1133-1139
56. Kurie, J. M. (2004) Chest 125(5 Suppl), 141S-144S
57. Kim, D., Dan, H. C., Park, S., Yang, L., Liu, Q., Kaneko, S., Ning, J., He, L., Yang, H., Sun, M., Nicosia, S. V., and Cheng, J. Q. (2005) Front Biosci 10, 975- 987
144 58. Tokunaga, E., Kimura, Y., Oki, E., Mashino, K., Kataoka, A., Ohno, S., Kakeji, Y., Baba, H., and Y, a. M. (2005) Journal of Clinical Oncology, 2005 ASCO Annual Meeting Proceedings Vol 23(No 16S), 9500
59. Shin, E., Hong, S. W., Kim, S. H., and Yang, W. I. (2004) Yonsei Med J 45(2), 306-313
60. Vasko, V. V., Gaudart, J., Allasia, C., Savchenko, V., Di Cristofaro, J., Saji, M., Ringel, M. D., and De Micco, C. (2004) Eur J Endocrinol 151(6), 779-786
61. Lopiccolo, J., Granville, C. A., Gills, J. J., and Dennis, P. A. (2007) Anticancer Drugs 18(8), 861-874
62. Liaw, D., Marsh, D. J., Li, J., Dahia, P. L., Wang, S. I., Zheng, Z., Bose, S., Call, K. M., Tsou, H. C., Peacocke, M., Eng, C., and Parsons, R. (1997) Nat Genet 16(1), 64-67
63. Basso, A. D., Solit, D. B., Chiosis, G., Giri, B., Tsichlis, P., and Rosen, N. (2002) J Biol Chem 277(42), 39858-39866
64. Alessi, D. R., and Cohen, P. (1998) Curr Opin Genet Dev 8(1), 55-62
65. Suzuki, A., Hamada, K., Sasaki, T., Mak, T. W., and Nakano, T. (2007) Biochem Soc Trans 35(Pt 2), 172-176
66. Ringel, M. D., Hayre, N., Saito, J., Saunier, B., Schuppert, F., Burch, H., Bernet, V., Burman, K. D., Kohn, L. D., and Saji, M. (2001) Cancer Res 61(16), 6105- 6111
67. Wu, G., Mambo, E., Guo, Z., Hu, S., Huang, X., Gollin, S. M., Trink, B., Ladenson, P. W., Sidransky, D., and Xing, M. (2005) J Clin Endocrinol Metab 90(8), 4688-4693
68. Garcia-Rostan, G., Costa, A. M., Pereira-Castro, I., Salvatore, G., Hernandez, R., Hermsem, M. J., Herrero, A., Fusco, A., Cameselle-Teijeiro, J., and Santoro, M. (2005) Cancer Res 65(22), 10199-10207
145 69. Vasko, V., Saji, M., Hardy, E., Kruhlak, M., Larin, A., Savchenko, V., Miyakawa, M., Isozaki, O., Murakami, H., Tsushima, T., Burman, K. D., De Micco, C., and Ringel, M. D. (2004) J Med Genet 41(3), 161-170
70. Giordano, T. J., Kuick, R., Thomas, D. G., Misek, D. E., Vinco, M., Sanders, D., Zhu, Z., Ciampi, R., Roh, M., Shedden, K., Gauger, P., Doherty, G., Thompson, N. W., Hanash, S., Koenig, R. J., and Nikiforov, Y. E. (2005) Oncogene 24(44), 6646-6656
71. Kroll, T. G., Sarraf, P., Pecciarini, L., Chen, C. J., Mueller, E., Spiegelman, B. M., and Fletcher, J. A. (2000) Science 289(5483), 1357-1360
72. Nikiforova, M. N., Biddinger, P. W., Caudill, C. M., Kroll, T. G., and Nikiforov, Y. E. (2002) Am J Surg Pathol 26(8), 1016-1023
73. Marques, A. R., Espadinha, C., Catarino, A. L., Moniz, S., Pereira, T., Sobrinho, L. G., and Leite, V. (2002) J Clin Endocrinol Metab 87(8), 3947-3952
74. Reddi, H. V., McIver, B., Grebe, S. K., and Eberhardt, N. L. (2007) Endocrinology 148(3), 932-935
75. Poleev, A., Okladnova, O., Musti, A. M., Schneider, S., Royer-Pokora, B., and Plachov, D. (1997) Eur J Biochem 247(3), 860-869
76. Mascia, A., Nitsch, L., Di Lauro, R., and Zannini, M. (2002) J Endocrinol 172(1), 163-176
77. Pasca di Magliano, M., Di Lauro, R., and Zannini, M. (2000) Proc Natl Acad Sci U S A 97(24), 13144-13149
78. Fajas, L., Debril, M. B., and Auwerx, J. (2001) Nutr Metab Cardiovasc Dis 11(1), 64-69
79. Gregory Powell, J., Wang, X., Allard, B. L., Sahin, M., Wang, X. L., Hay, I. D., Hiddinga, H. J., Deshpande, S. S., Kroll, T. G., Grebe, S. K., Eberhardt, N. L., and McIver, B. (2004) Oncogene 23(20), 3634-3641
146 80. Chen, W. C., Lin, M. S., and Bai, X. (2005) Chin Med J (Engl) 118(17), 1477- 1481
81. Lee, Y. R., Yu, H. N., Noh, E. M., Kim, J. S., Song, E. K., Han, M. K., Kim, B. S., Lee, S. H., and Park, J. (2007) Int J Hematol 85(3), 231-237
82. Teresi, R. E., Shaiu, C. W., Chen, C. S., Chatterjee, V. K., Waite, K. A., and Eng, C. (2006) Int J Cancer 118(10), 2390-2398
83. Vecchio, G., and Santoro, M. (2000) Clin Chem Lab Med 38(2), 113-116
84. Begum, S., Rosenbaum, E., Henrique, R., Cohen, Y., Sidransky, D., and Westra, W. H. (2004) Mod Pathol 17(11), 1359-1363
85. Quiros, R. M., Ding, H. G., Gattuso, P., Prinz, R. A., and Xu, X. (2005) Cancer 103(11), 2261-2268
86. Hou, P., Liu, D., Shan, Y., Hu, S., Studeman, K., Condouris, S., Wang, Y., Trink, A., El-Naggar, A. K., Tallini, G., Vasko, V., and Xing, M. (2007) Clin Cancer Res 13(4), 1161-1170
87. Datta, S. R., Brunet, A., and Greenberg, M. E. (1999) Genes Dev 13(22), 2905- 2927
88. Nicholson, K. M., and Anderson, N. G. (2002) Cell Signal 14(5), 381-395
89. Testa, J. R., and Bellacosa, A. (2001) Proc Natl Acad Sci U S A 98(20), 10983- 10985
90. Zdychova, J., and Komers, R. (2005) Physiol Res 54(1), 1-16
91. Kikani, C. K., Dong, L. Q., and Liu, F. (2005) J Cell Biochem 96(6), 1157-1162
92. Cantley, L. C. (2002) Science 296(5573), 1655-1657
93. Wu, H., Yan, Y., and Backer, J. M. (2007) Biochem Soc Trans 35(Pt 2), 242-244 147 94. Gymnopoulos, M., Elsliger, M. A., and Vogt, P. K. (2007) Proc Natl Acad Sci U S A 104(13), 5569-5574
95. Pawson, T., and Nash, P. (2000) Genes Dev 14(9), 1027-1047
96. Simpson, L., and Parsons, R. (2001) Exp Cell Res 264(1), 29-41
97. Taylor, V., Wong, M., Brandts, C., Reilly, L., Dean, N. M., Cowsert, L. M., Moodie, S., and Stokoe, D. (2000) Mol Cell Biol 20(18), 6860-6871
98. Cohen, P., Alessi, D. R., and Cross, D. A. (1997) FEBS Lett 410(1), 3-10
99. Pullen, N., Dennis, P. B., Andjelkovic, M., Dufner, A., Kozma, S. C., Hemmings, B. A., and Thomas, G. (1998) Science 279(5351), 707-710
100. Anderson, K. E., Coadwell, J., Stephens, L. R., and Hawkins, P. T. (1998) Curr Biol 8(12), 684-691
101. Flynn, P., Mellor, H., Casamassima, A., and Parker, P. J. (2000) J Biol Chem 275(15), 11064-11070
102. King, C. C., Gardiner, E. M., Zenke, F. T., Bohl, B. P., Newton, A. C., Hemmings, B. A., and Bokoch, G. M. (2000) J Biol Chem 275(52), 41201-41209
103. Kobayashi, T., and Cohen, P. (1999) Biochem J 339 (Pt 2), 319-328
104. Mora, A., Komander, D., van Aalten, D. M., and Alessi, D. R. (2004) Semin Cell Dev Biol 15(2), 161-170
105. Casamayor, A., Morrice, N. A., and Alessi, D. R. (1999) Biochem J 342 (Pt 2), 287-292
106. Makris, C., Voisin, L., Giasson, E., Tudan, C., Kaplan, D. R., and Meloche, S. (2002) Oncogene 21(51), 7891-7896
107. Staal, S. P. (1987) Proc Natl Acad Sci U S A 84(14), 5034-5037 148 108. Vanhaesebroeck, B., and Alessi, D. R. (2000) Biochem J 346 Pt 3, 561-576
109. Feng, J., Park, J., Cron, P., Hess, D., and Hemmings, B. A. (2004) J Biol Chem 279(39), 41189-41196
110. Persad, S., Attwell, S., Gray, V., Mawji, N., Deng, J. T., Leung, D., Yan, J., Sanghera, J., Walsh, M. P., and Dedhar, S. (2001) J Biol Chem 276(29), 27462- 27469
111. Kawakami, Y., Nishimoto, H., Kitaura, J., Maeda-Yamamoto, M., Kato, R. M., Littman, D. R., Leitges, M., Rawlings, D. J., and Kawakami, T. (2004) J Biol Chem 279(46), 47720-47725
112. Sarbassov, D. D., Guertin, D. A., Ali, S. M., and Sabatini, D. M. (2005) Science 307(5712), 1098-1101
113. Toker, A., and Newton, A. C. (2000) J Biol Chem 275(12), 8271-8274
114. Maira, S. M., Galetic, I., Brazil, D. P., Kaech, S., Ingley, E., Thelen, M., and Hemmings, B. A. (2001) Science 294(5541), 374-380
115. Samuels, Y., Wang, Z., Bardelli, A., Silliman, N., Ptak, J., Szabo, S., Yan, H., Gazdar, A., Powell, S. M., Riggins, G. J., Willson, J. K., Markowitz, S., Kinzler, K. W., Vogelstein, B., and Velculescu, V. E. (2004) Science 304(5670), 554
116. Carpten, J. D., Faber, A. L., Horn, C., Donoho, G. P., Briggs, S. L., Robbins, C. M., Hostetter, G., Boguslawski, S., Moses, T. Y., Savage, S., Uhlik, M., Lin, A., Du, J., Qian, Y. W., Zeckner, D. J., Tucker-Kellogg, G., Touchman, J., Patel, K., Mousses, S., Bittner, M., Schevitz, R., Lai, M. H., Blanchard, K. L., and Thomas, J. E. (2007) Nature 448(7152), 439-444
117. Riemenschneider, M. J., Betensky, R. A., Pasedag, S. M., and Louis, D. N. (2006) Cancer Res 66(11), 5618-5623
118. Yang, Z. Z., Tschopp, O., Baudry, A., Dummler, B., Hynx, D., and Hemmings, B. A. (2004) Biochem Soc Trans 32(Pt 2), 350-354
149 119. Cho, H., Thorvaldsen, J. L., Chu, Q., Feng, F., and Birnbaum, M. J. (2001) J Biol Chem 276(42), 38349-38352
120. Bellacosa, A., de Feo, D., Godwin, A. K., Bell, D. W., Cheng, J. Q., Altomare, D. A., Wan, M., Dubeau, L., Scambia, G., Masciullo, V., Ferrandina, G., Benedetti Panici, P., Mancuso, S., Neri, G., and Testa, J. R. (1995) Int J Cancer 64(4), 280- 285
121. 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
122. Garofalo, R. S., Orena, S. J., Rafidi, K., Torchia, A. J., Stock, J. L., Hildebrandt, A. L., Coskran, T., Black, S. C., Brees, D. J., Wicks, J. R., McNeish, J. D., and Coleman, K. G. (2003) J Clin Invest 112(2), 197-208
123. Yang, Z. Z., Tschopp, O., Hemmings-Mieszczak, M., Feng, J., Brodbeck, D., Perentes, E., and Hemmings, B. A. (2003) J Biol Chem 278(34), 32124-32131
124. Nakatani, K., Thompson, D. A., Barthel, A., Sakaue, H., Liu, W., Weigel, R. J., and Roth, R. A. (1999) J Biol Chem 274(31), 21528-21532
125. Cross, D. A., Alessi, D. R., Cohen, P., Andjelkovich, M., and Hemmings, B. A. (1995) Nature 378(6559), 785-789
126. Kuure, S., Popsueva, A., Jakobson, M., Sainio, K., and Sariola, H. (2007) J Am Soc Nephrol 18(4), 1130-1139
127. Horvai, A. E., Kramer, M. J., and O'Donnell, R. (2006) Arch Pathol Lab Med 130(6), 792-798
128. Kato, J., Matsushime, H., Hiebert, S. W., Ewen, M. E., and Sherr, C. J. (1993) Genes Dev 7(3), 331-342
129. Zhou, B. P., Liao, Y., Xia, W., Spohn, B., Lee, M. H., and Hung, M. C. (2001) Nat Cell Biol 3(3), 245-252
150 130. Liang, J., Zubovitz, J., Petrocelli, T., Kotchetkov, R., Connor, M. K., Han, K., Lee, J. H., Ciarallo, S., Catzavelos, C., Beniston, R., Franssen, E., and Slingerland, J. M. (2002) Nat Med 8(10), 1153-1160
131. Shin, I., Yakes, F. M., Rojo, F., Shin, N. Y., Bakin, A. V., Baselga, J., and Arteaga, C. L. (2002) Nat Med 8(10), 1145-1152
132. Inoki, K., Li, Y., Zhu, T., Wu, J., and Guan, K. L. (2002) Nat Cell Biol 4(9), 648- 657
133. Nave, B. T., Ouwens, M., Withers, D. J., Alessi, D. R., and Shepherd, P. R. (1999) Biochem J 344 Pt 2, 427-431
134. Franke, T. F., Kaplan, D. R., and Cantley, L. C. (1997) Cell 88(4), 435-437
135. Adams, J. M., and Cory, S. (2007) Oncogene 26(9), 1324-1337
136. Blume-Jensen, P., Janknecht, R., and Hunter, T. (1998) Curr Biol 8(13), 779-782
137. Yang, H., Masters, S. C., Wang, H., and Fu, H. (2001) Biochim Biophys Acta 1547(2), 313-319
138. Limesand, K. H., Schwertfeger, K. L., and Anderson, S. M. (2006) Mol Cell Biol 26(23), 8840-8856
139. Du, K., and Montminy, M. (1998) J Biol Chem 273(49), 32377-32379
140. Pugazhenthi, S., Nesterova, A., Sable, C., Heidenreich, K. A., Boxer, L. M., Heasley, L. E., and Reusch, J. E. (2000) J Biol Chem 275(15), 10761-10766
141. Reusch, J. E., and Klemm, D. J. (2002) J Biol Chem 277(2), 1426-1432
142. Cheng, G. Z., Chan, J., Wang, Q., Zhang, W., Sun, C. D., and Wang, L. H. (2007) Cancer Res 67(5), 1979-1987
151 143. Irie, H. Y., Pearline, R. V., Grueneberg, D., Hsia, M., Ravichandran, P., Kothari, N., Natesan, S., and Brugge, J. S. (2005) J Cell Biol 171(6), 1023-1034
144. Enomoto, A., Murakami, H., Asai, N., Morone, N., Watanabe, T., Kawai, K., Murakumo, Y., Usukura, J., Kaibuchi, K., and Takahashi, M. (2005) Dev Cell 9(3), 389-402
145. Li, J., Ballif, B. A., Powelka, A. M., Dai, J., Gygi, S. P., and Hsu, V. W. (2005) Dev Cell 9(5), 663-673
146. Kobayashi, T., Hino, S., Oue, N., Asahara, T., Zollo, M., Yasui, W., and Kikuchi, A. (2006) Mol Cell Biol 26(3), 898-911
147. Roberts, M. S., Woods, A. J., Dale, T. C., Van Der Sluijs, P., and Norman, J. C. (2004) Mol Cell Biol 24(4), 1505-1515
148. Yan, J., Roy, S., Apolloni, A., Lane, A., and Hancock, J. F. (1998) J Biol Chem 273(37), 24052-24056
149. Garcia-Rostan, G., Zhao, H., Camp, R. L., Pollan, M., Herrero, A., Pardo, J., Wu, R., Carcangiu, M. L., Costa, J., and Tallini, G. (2003) J Clin Oncol 21(17), 3226- 3235
150. Vasko, V., Ferrand, M., Di Cristofaro, J., Carayon, P., Henry, J. F., and de Micco, C. (2003) J Clin Endocrinol Metab 88(6), 2745-2752
151. Philips, M. R. (2005) Biochem Soc Trans 33(Pt 4), 657-661
152. Nimnual, A., and Bar-Sagi, D. (2002) Sci STKE 2002(145), PE36
153. Tari, A. M., and Lopez-Berestein, G. (2001) Semin Oncol 28(5 Suppl 16), 142- 147
154. Chiu, V. K., Bivona, T., Hach, A., Sajous, J. B., Silletti, J., Wiener, H., Johnson, R. L., 2nd, Cox, A. D., and Philips, M. R. (2002) Nat Cell Biol 4(5), 343-350
155. Tzivion, G., Luo, Z., and Avruch, J. (1998) Nature 394(6688), 88-92 152 156. Adams, D. G., Coffee, R. L., Jr., Zhang, H., Pelech, S., Strack, S., and Wadzinski, B. E. (2005) J Biol Chem 280(52), 42644-42654
157. Zang, M., Hayne, C., and Luo, Z. (2002) J Biol Chem 277(6), 4395-4405
158. Fabian, J. R., Daar, I. O., and Morrison, D. K. (1993) Mol Cell Biol 13(11), 7170- 7179
159. Chong, H., Lee, J., and Guan, K. L. (2001) Embo J 20(14), 3716-3727
160. King, A. J., Sun, H., Diaz, B., Barnard, D., Miao, W., Bagrodia, S., and Marshall, M. S. (1998) Nature 396(6707), 180-183
161. Xia, K., Mukhopadhyay, N. K., Inhorn, R. C., Barber, D. L., Rose, P. E., Lee, R. S., Narsimhan, R. P., D'Andrea, A. D., Griffin, J. D., and Roberts, T. M. (1996) Proc Natl Acad Sci U S A 93(21), 11681-11686
162. Kolch, W., Heidecker, G., Kochs, G., Hummel, R., Vahidi, H., Mischak, H., Finkenzeller, G., Marme, D., and Rapp, U. R. (1993) Nature 364(6434), 249-252
163. Barnard, D., Diaz, B., Clawson, D., and Marshall, M. (1998) Oncogene 17(12), 1539-1547
164. Sieburth, D. S., Sundaram, M., Howard, R. M., and Han, M. (1999) Genes Dev 13(19), 2562-2569
165. Dumaz, N., and Marais, R. (2003) J Biol Chem 278(32), 29819-29823
166. Jaumot, M., and Hancock, J. F. (2001) Oncogene 20(30), 3949-3958
167. Morrison, D. K., Heidecker, G., Rapp, U. R., and Copeland, T. D. (1993) J Biol Chem 268(23), 17309-17316
168. Schramm, K., Niehof, M., Radziwill, G., Rommel, C., and Moelling, K. (1994) Biochem Biophys Res Commun 201(2), 740-747
153 169. Sidovar, M. F., Kozlowski, P., Lee, J. W., Collins, M. A., He, Y., and Graves, L. M. (2000) J Biol Chem 275(37), 28688-28694
170. Morrison, D. K., and Cutler, R. E. (1997) Curr Opin Cell Biol 9(2), 174-179
171. Zimmermann, S., and Moelling, K. (1999) Science 286(5445), 1741-1744
172. Guan, K. L., Figueroa, C., Brtva, T. R., Zhu, T., Taylor, J., Barber, T. D., and Vojtek, A. B. (2000) J Biol Chem 275(35), 27354-27359
173. Zhang, B. H., Tang, E. D., Zhu, T., Greenberg, M. E., Vojtek, A. B., and Guan, K. L. (2001) J Biol Chem 276(34), 31620-31626
174. Rapp, U. R., Goldsborough, M. D., Mark, G. E., Bonner, T. I., Groffen, J., Reynolds, F. H., Jr., and Stephenson, J. R. (1983) Proc Natl Acad Sci U S A 80(14), 4218-4222
175. Jansen, H. W., Ruckert, B., Lurz, R., and Bister, K. (1983) Embo J 2(11), 1969- 1975
176. Rapp, U. R., Cleveland, J. L., Storm, S. M., Beck, T. W., and Huleihel, M. (1986) Princess Takamatsu Symp 17, 55-74
177. Rapp, U. R., Heidecker, G., Huleihel, M., Cleveland, J. L., Choi, W. C., Pawson, T., Ihle, J. N., and Anderson, W. B. (1988) Cold Spring Harb Symp Quant Biol 53 Pt 1, 173-184
178. Sithanandam, G., Kolch, W., Duh, F. M., and Rapp, U. R. (1990) Oncogene 5(12), 1775-1780
179. Beck, T. W., Huleihel, M., Gunnell, M., Bonner, T. I., and Rapp, U. R. (1987) Nucleic Acids Res 15(2), 595-609
180. Daum, G., Eisenmann-Tappe, I., Fries, H. W., Troppmair, J., and Rapp, U. R. (1994) Trends Biochem Sci 19(11), 474-480
154 181. Cutler, R. E., Jr., Stephens, R. M., Saracino, M. R., and Morrison, D. K. (1998) Proc Natl Acad Sci U S A 95(16), 9214-9219
182. Beeram, M., Patnaik, A., and Rowinsky, E. K. (2005) J Clin Oncol 23(27), 6771- 6790
183. Mikula, M., Schreiber, M., Husak, Z., Kucerova, L., Ruth, J., Wieser, R., Zatloukal, K., Beug, H., Wagner, E. F., and Baccarini, M. (2001) Embo J 20(8), 1952-1962
184. Jin, S., Zhuo, Y., Guo, W., and Field, J. (2005) J Biol Chem 280(26), 24698- 24705
185. Barnier, J. V., Papin, C., Eychene, A., Lecoq, O., and Calothy, G. (1995) J Biol Chem 270(40), 23381-23389
186. Ohtsuka, T., Shimizu, K., Yamamori, B., Kuroda, S., and Takai, Y. (1996) J Biol Chem 271(3), 1258-1261
187. Mitsutake, N., Knauf, J. A., Mitsutake, S., Mesa, C., Jr., Zhang, L., and Fagin, J. A. (2005) Cancer Res 65(6), 2465-2473
188. Wu, X., Noh, S. J., Zhou, G., Dixon, J. E., and Guan, K. L. (1996) J Biol Chem 271(6), 3265-3271
189. Yuryev, A., Ono, M., Goff, S. A., Macaluso, F., and Wennogle, L. P. (2000) Mol Cell Biol 20(13), 4870-4878
190. Dhanasekaran, N., and Premkumar Reddy, E. (1998) Oncogene 17(11 Reviews), 1447-1455
191. Baumann, B., Weber, C. K., Troppmair, J., Whiteside, S., Israel, A., Rapp, U. R., and Wirth, T. (2000) Proc Natl Acad Sci U S A 97(9), 4615-4620
192. Wang, S., Ghosh, R. N., and Chellappan, S. P. (1998) Mol Cell Biol 18(12), 7487- 7498
155 193. Wang, H. G., Miyashita, T., Takayama, S., Sato, T., Torigoe, T., Krajewski, S., Tanaka, S., Hovey, L., 3rd, Troppmair, J., Rapp, U. R., and et al. (1994) Oncogene 9(9), 2751-2756
194. Palona, I., Namba, H., Mitsutake, N., Starenki, D., Podtcheko, A., Sedliarou, I., Ohtsuru, A., Saenko, V., Nagayama, Y., Umezawa, K., and Yamashita, S. (2006) Endocrinology 147(12), 5699-5707
195. Li, S., and Sedivy, J. M. (1993) Proc Natl Acad Sci U S A 90(20), 9247-9251
196. Pritchard, C. A., Hayes, L., Wojnowski, L., Zimmer, A., Marais, R. M., and Norman, J. C. (2004) Mol Cell Biol 24(13), 5937-5952
197. Sumimoto, H., Miyagishi, M., Miyoshi, H., Yamagata, S., Shimizu, A., Taira, K., and Kawakami, Y. (2004) Oncogene 23(36), 6031-6039
198. Price, J. T., and Thompson, E. W. (2002) Expert Opin Ther Targets 6(2), 217-233
199. Small, J. V., Rottner, K., Kaverina, I., and Anderson, K. I. (1998) Biochim Biophys Acta 1404(3), 271-281
200. Alberts, B., Johnson, A., Lewis, J., and al., E. (2002) Garland Science 4 edition
201. Fenteany, G., Janmey, P. A., and Stossel, T. P. (2000) Curr Biol 10(14), 831-838
202. Gimona, M., and Buccione, R. (2006) Int J Biochem Cell Biol 38(11), 1875-1892
203. Hall, A. (1992) Mol Biol Cell 3(5), 475-479
204. Hall, A. (2005) Biochem Soc Trans 33(Pt 5), 891-895
205. Nobes, C. D., and Hall, A. (1995) Biochem Soc Trans 23(3), 456-459
206. Nobes, C. D., and Hall, A. (1995) Cell 81(1), 53-62
156 207. Ananthakrishnan, R., and Ehrlicher, A. (2007) Int J Biol Sci 3(5), 303-317
208. Pollard, T. D., Blanchoin, L., and Mullins, R. D. (2000) Annu Rev Biophys Biomol Struct 29, 545-576
209. dos Remedios, C. G., Chhabra, D., Kekic, M., Dedova, I. V., Tsubakihara, M., Berry, D. A., and Nosworthy, N. J. (2003) Physiol Rev 83(2), 433-473
210. Yarmola, E. G., and Bubb, M. R. (2006) Trends Biochem Sci 31(4), 197-205
211. Matsudaira, P. (1994) Semin Cell Biol 5(3), 165-174
212. Ono, S. (2007) Int Rev Cytol 258, 1-82
213. Berrier, A. L., and Yamada, K. M. (2007) J Cell Physiol
214. Critchley, D. R. (2004) Biochem Soc Trans 32(Pt 5), 831-836
215. Ratnikov, B. I., Partridge, A. W., and Ginsberg, M. H. (2005) J Thromb Haemost 3(8), 1783-1790
216. Sepulveda, J. L., and Wu, C. (2006) Cell Mol Life Sci 63(1), 25-35
217. Zamir, E., and Geiger, B. (2001) J Cell Sci 114(Pt 20), 3583-3590
218. Wells, A., Huttenlocher, A., and Lauffenburger, D. A. (2005) Int Rev Cytol 245, 1-16
219. Manser, E., Leung, T., Salihuddin, H., Zhao, Z. S., and Lim, L. (1994) Nature 367(6458), 40-46
220. Martin, G. A., Bollag, G., McCormick, F., and Abo, A. (1995) Embo J 14(9), 1970-1978
157 221. Lim, L., Manser, E., Leung, T., and Hall, C. (1996) Eur J Biochem 242(2), 171- 185
222. Rennefahrt, U. E., Deacon, S. W., Parker, S. A., Devarajan, K., Beeser, A., Chernoff, J., Knapp, S., Turk, B. E., and Peterson, J. R. (2007) J Biol Chem 282(21), 15667-15678
223. Zhu, G., Fujii, K., Liu, Y., Codrea, V., Herrero, J., and Shaw, S. (2005) J Biol Chem 280(43), 36372-36379
224. Sells, M. A., Knaus, U. G., Bagrodia, S., Ambrose, D. M., Bokoch, G. M., and Chernoff, J. (1997) Curr Biol 7(3), 202-210
225. Puto, L. A., Pestonjamasp, K., King, C. C., and Bokoch, G. M. (2003) J Biol Chem 278(11), 9388-9393
226. Galisteo, M. L., Chernoff, J., Su, Y. C., Skolnik, E. Y., and Schlessinger, J. (1996) J Biol Chem 271(35), 20997-21000
227. Manser, E., Loo, T. H., Koh, C. G., Zhao, Z. S., Chen, X. Q., Tan, L., Tan, I., Leung, T., and Lim, L. (1998) Mol Cell 1(2), 183-192
228. Yu, J. S., Chen, W. J., Ni, M. H., Chan, W. H., and Yang, S. D. (1998) Biochem J 334 (Pt 1), 121-131
229. Zenke, F. T., King, C. C., Bohl, B. P., and Bokoch, G. M. (1999) J Biol Chem 274(46), 32565-32573
230. Schurmann, A., Mooney, A. F., Sanders, L. C., Sells, M. A., Wang, H. G., Reed, J. C., and Bokoch, G. M. (2000) Mol Cell Biol 20(2), 453-461
231. Sells, M. A., Boyd, J. T., and Chernoff, J. (1999) J Cell Biol 145(4), 837-849
232. Ching, Y. P., Leong, V. Y., Wong, C. M., and Kung, H. F. (2003) J Biol Chem 278(36), 33621-33624
158 233. Abo, A., Qu, J., Cammarano, M. S., Dan, C., Fritsch, A., Baud, V., Belisle, B., and Minden, A. (1998) Embo J 17(22), 6527-6540
234. Qu, J., Cammarano, M. S., Shi, Q., Ha, K. C., de Lanerolle, P., and Minden, A. (2001) Mol Cell Biol 21(10), 3523-3533
235. Lei, M., Lu, W., Meng, W., Parrini, M. C., Eck, M. J., Mayer, B. J., and Harrison, S. C. (2000) Cell 102(3), 387-397
236. Chong, C., Tan, L., Lim, L., and Manser, E. (2001) J Biol Chem 276(20), 17347- 17353
237. Tang, Y., Zhou, H., Chen, A., Pittman, R. N., and Field, J. (2000) J Biol Chem 275(13), 9106-9109
238. Nikolic, M., Chou, M. M., Lu, W., Mayer, B. J., and Tsai, L. H. (1998) Nature 395(6698), 194-198
239. Rashid, T., Banerjee, M., and Nikolic, M. (2001) J Biol Chem 276(52), 49043- 49052
240. Westphal, R. S., Coffee, R. L., Jr., Marotta, A., Pelech, S. L., and Wadzinski, B. E. (1999) J Biol Chem 274(2), 687-692
241. Koh, C. G., Tan, E. J., Manser, E., and Lim, L. (2002) Curr Biol 12(4), 317-321
242. Feng, Q., Albeck, J. G., Cerione, R. A., and Yang, W. (2002) J Biol Chem 277(7), 5644-5650
243. Bagrodia, S., and Cerione, R. A. (1999) Trends Cell Biol 9(9), 350-355
244. Wang, J., Frost, J. A., Cobb, M. H., and Ross, E. M. (1999) J Biol Chem 274(44), 31641-31647
245. Parrini, M. C., Lei, M., Harrison, S. C., and Mayer, B. J. (2002) Mol Cell 9(1), 73-83
159 246. Chong, H., Vikis, H. G., and Guan, K. L. (2003) Cell Signal 15(5), 463-469
247. Gatti, A., Huang, Z., Tuazon, P. T., and Traugh, J. A. (1999) J Biol Chem 274(12), 8022-8028
248. Buchwald, G., Hostinova, E., Rudolph, M. G., Kraemer, A., Sickmann, A., Meyer, H. E., Scheffzek, K., and Wittinghofer, A. (2001) Mol Cell Biol 21(15), 5179-5189
249. Molnar, G., Dagher, M. C., Geiszt, M., Settleman, J., and Ligeti, E. (2001) Biochemistry 40(35), 10542-10549
250. Yang, F., Li, X., Sharma, M., Zarnegar, M., Lim, B., and Sun, Z. (2001) J Biol Chem 276(18), 15345-15353
251. Guo, D., Tan, Y. C., Wang, D., Madhusoodanan, K. S., Zheng, Y., Maack, T., Zhang, J. J., and Huang, X. Y. (2007) Cell 128(2), 341-355
252. Frost, J. A., Khokhlatchev, A., Stippec, S., White, M. A., and Cobb, M. H. (1998) J Biol Chem 273(43), 28191-28198
253. Lee, S. R., Ramos, S. M., Ko, A., Masiello, D., Swanson, K. D., Lu, M. L., and Balk, S. P. (2002) Mol Endocrinol 16(1), 85-99
254. Sanders, L. C., Matsumura, F., Bokoch, G. M., and de Lanerolle, P. (1999) Science 283(5410), 2083-2085
255. Wirth, A., Schroeter, M., Kock-Hauser, C., Manser, E., Chalovich, J. M., De Lanerolle, P., and Pfitzer, G. (2003) J Physiol 549(Pt 2), 489-500
256. Ohtakara, K., Inada, H., Goto, H., Taki, W., Manser, E., Lim, L., Izawa, I., and Inagaki, M. (2000) Biochem Biophys Res Commun 272(3), 712-716
257. Xiao, G. H., Beeser, A., Chernoff, J., and Testa, J. R. (2002) J Biol Chem 277(2), 883-886
160 258. Edwards, D. C., Sanders, L. C., Bokoch, G. M., and Gill, G. N. (1999) Nat Cell Biol 1(5), 253-259
259. Zhao, Z. S., Manser, E., Chen, X. Q., Chong, C., Leung, T., and Lim, L. (1998) Mol Cell Biol 18(4), 2153-2163
260. Kiosses, W. B., Daniels, R. H., Otey, C., Bokoch, G. M., and Schwartz, M. A. (1999) J Cell Biol 147(4), 831-844
261. Rayala, S. K., Talukder, A. H., Balasenthil, S., Tharakan, R., Barnes, C. J., Wang, R. A., Aldaz, M., Khan, S., and Kumar, R. (2006) Cancer Res 66(3), 1694-1701
262. Rudel, T., and Bokoch, G. M. (1997) Science 276(5318), 1571-1574
263. Walter, B. N., Huang, Z., Jakobi, R., Tuazon, P. T., Alnemri, E. S., Litwack, G., and Traugh, J. A. (1998) J Biol Chem 273(44), 28733-28739
264. Allen, K. M., Gleeson, J. G., Bagrodia, S., Partington, M. W., MacMillan, J. C., Cerione, R. A., Mulley, J. C., and Walsh, C. A. (1998) Nat Genet 20(1), 25-30
265. Kreis, P., Thevenot, E., Rousseau, V., Boda, B., Muller, D., and Barnier, J. V. (2007) J Biol Chem 282(29), 21497-21506
266. Gedeon, A. K., Nelson, J., Gecz, J., and Mulley, J. C. (2003) Am J Med Genet A 120(4), 509-517
267. Boda, B., Alberi, S., Nikonenko, I., Node-Langlois, R., Jourdain, P., Moosmayer, M., Parisi-Jourdain, L., and Muller, D. (2004) J Neurosci 24(48), 10816-10825
268. 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
269. Chernoff, J. (1999) Nat Cell Biol 1(5), E115-117
270. Gnesutta, N., Qu, J., and Minden, A. (2001) J Biol Chem 276(17), 14414-14419
161 271. Dan, C., Nath, N., Liberto, M., and Minden, A. (2002) Mol Cell Biol 22(2), 567- 577
272. Cotteret, S., and Chernoff, J. (2006) Mol Cell Biol 26(8), 3215-3230
273. Cotteret, S., Jaffer, Z. M., Beeser, A., and Chernoff, J. (2003) Mol Cell Biol 23(16), 5526-5539
274. Wu, X., and Frost, J. A. (2006) Biochem Biophys Res Commun 351(2), 328-335
275. Pandey, A., Dan, I., Kristiansen, T. Z., Watanabe, N. M., Voldby, J., Kajikawa, E., Khosravi-Far, R., Blagoev, B., and Mann, M. (2002) Oncogene 21(24), 3939- 3948
276. Matenia, D., Griesshaber, B., Li, X. Y., Thiessen, A., Johne, C., Jiao, J., Mandelkow, E., and Mandelkow, E. M. (2005) Mol Biol Cell 16(9), 4410-4422
277. Gao, W., Bohl, C. E., and Dalton, J. T. (2005) Chem Rev 105(9), 3352-3370
278. Coles, L. C., and Shaw, P. E. (2002) Oncogene 21(14), 2236-2244
279. Frost, J. A., Steen, H., Shapiro, P., Lewis, T., Ahn, N., Shaw, P. E., and Cobb, M. H. (1997) Embo J 16(21), 6426-6438
280. Cau, J., Faure, S., Comps, M., Delsert, C., and Morin, N. (2001) J Cell Biol 155(6), 1029-1042
281. Andrews, P. D., Knatko, E., Moore, W. J., and Swedlow, J. R. (2003) Curr Opin Cell Biol 15(6), 672-683
282. Cotteret, S., and Chernoff, J. (2005) Dev Cell 9(5), 573-574
283. Zhao, Z. S., Lim, J. P., Ng, Y. W., Lim, L., and Manser, E. (2005) Mol Cell 20(2), 237-249
162 284. Banerjee, M., Worth, D., Prowse, D. M., and Nikolic, M. (2002) Curr Biol 12(14), 1233-1239
285. Frost, J. A., Swantek, J. L., Stippec, S., Yin, M. J., Gaynor, R., and Cobb, M. H. (2000) J Biol Chem 275(26), 19693-19699
286. Foryst-Ludwig, A., and Naumann, M. (2000) J Biol Chem 275(50), 39779-39785
287. Mazumdar, A., and Kumar, R. (2003) FEBS Lett 535(1-3), 6-10
288. Vadlamudi, R. K., Bagheri-Yarmand, R., Yang, Z., Balasenthil, S., Nguyen, D., Sahin, A. A., den Hollander, P., and Kumar, R. (2004) Cancer Cell 5(6), 575-585
289. Day, C. L., Puthalakath, H., Skea, G., Strasser, A., Barsukov, I., Lian, L. Y., Huang, D. C., and Hinds, M. G. (2004) Biochem J 377(Pt 3), 597-605
290. Jakobi, R., Moertl, E., and Koeppel, M. A. (2001) J Biol Chem 276(20), 16624- 16634
291. Lee, N., MacDonald, H., Reinhard, C., Halenbeck, R., Roulston, A., Shi, T., and Williams, L. T. (1997) Proc Natl Acad Sci U S A 94(25), 13642-13647
292. Rudel, T., Zenke, F. T., Chuang, T. H., and Bokoch, G. M. (1998) J Immunol 160(1), 7-11
293. Cassimeris, L. (2002) Curr Opin Cell Biol 14(1), 18-24
294. Wittmann, T., Bokoch, G. M., and Waterman-Storer, C. M. (2004) J Biol Chem 279(7), 6196-6203
295. Vadlamudi, R. K., Li, F., Barnes, C. J., Bagheri-Yarmand, R., and Kumar, R. (2004) EMBO Rep 5(2), 154-160
296. Meberg, P. J. (2000) Mol Neurobiol 21(1-2), 97-107
163 297. Geeves, M. A., Fedorov, R., and Manstein, D. J. (2005) Cell Mol Life Sci 62(13), 1462-1477
298. Chew, T. L., Masaracchia, R. A., Goeckeler, Z. M., and Wysolmerski, R. B. (1998) J Muscle Res Cell Motil 19(8), 839-854
299. Zeng, Q., Lagunoff, D., Masaracchia, R., Goeckeler, Z., Cote, G., and Wysolmerski, R. (2000) J Cell Sci 113 (Pt 3), 471-482
300. Goeckeler, Z. M., Masaracchia, R. A., Zeng, Q., Chew, T. L., Gallagher, P., and Wysolmerski, R. B. (2000) J Biol Chem 275(24), 18366-18374
301. Vadlamudi, R. K., Li, F., Adam, L., Nguyen, D., Ohta, Y., Stossel, T. P., and Kumar, R. (2002) Nat Cell Biol 4(9), 681-690
302. Popowicz, G. M., Schleicher, M., Noegel, A. A., and Holak, T. A. (2006) Trends Biochem Sci 31(7), 411-419
303. Meier, R., and Hemmings, B. A. (1999) J Recept Signal Transduct Res 19(1-4), 121-128
304. Alessi, D. R., Kozlowski, M. T., Weng, Q. P., Morrice, N., and Avruch, J. (1998) Curr Biol 8(2), 69-81
305. Dong, L. Q., Zhang, R. B., Langlais, P., He, H., Clark, M., Zhu, L., and Liu, F. (1999) J Biol Chem 274(12), 8117-8122
306. Storz, P., and Toker, A. (2002) Front Biosci 7, d886-902
307. Brader, S., and Eccles, S. A. (2004) Tumori 90(1), 2-8
308. Kucab, J. E., Lee, C., Chen, C. S., Zhu, J., Gilks, C. B., Cheang, M., Huntsman, D., Yorida, E., Emerman, J., Pollak, M., and Dunn, S. E. (2005) Breast Cancer Res 7(5), R796-807
309. Huber, M. A., Kraut, N., and Beug, H. (2005) Curr Opin Cell Biol 17(5), 548-558
164 310. Thiery, J. P. (2002) Nat Rev Cancer 2(6), 442-454
311. Thiery, J. P. (2002) Nat Rev Cancer 2(6), 442-454.
312. Huber, M. A., Kraut, N., and Beug, H. (2005) Curr Opin Cell Biol 17(5), 548- 558.
313. Kumar, R., Gururaj, A. E., and Barnes, C. J. (2006) Nat Rev Cancer 6(6), 459-471
314. Yang, Z., Rayala, S., Nguyen, D., Vadlamudi, R. K., Chen, S., and Kumar, R. (2005) Cancer Res 65(8), 3179-3184
315. Sells, M. A., Pfaff, A., and Chernoff, J. (2000) J Cell Biol 151(7), 1449-1458
316. Burridge, K., and Wennerberg, K. (2004) Cell 116(2), 167-179
317. Sahai, E., and Marshall, C. J. (2003) Nat Cell Biol 5(8), 711-719
318. Zhao, Z. S., and Manser, E. (2005) Biochem J 386(Pt 2), 201-214
319. Lei, M., Robinson, M. A., and Harrison, S. C. (2005) Structure 13(5), 769-778
320. King, C. C., Gardiner, E. M., Zenke, F. T., Bohl, B. P., Newton, A. C., Hemmings, B. A., and Bokoch, G. M. (2000) J Biol Chem 275(52), 41201-41209.
321. Zenke, F. T., King, C. C., Bohl, B. P., and Bokoch, G. M. (1999) J Biol Chem 274(46), 32565-32573.
322. Fagin, J. A., Matsuo, K., Karmakar, A., Chen, D. L., Tang, S. H., and Koeffler, H. P. (1993) J Clin Invest 91(1), 179-184
323. Estour, B., Van Herle, A. J., Juillard, G. J., Totanes, T. L., Sparkes, R. S., Giuliano, A. E., and Klandorf, H. (1989) Virchows Arch B Cell Pathol Incl Mol Pathol 57(3), 167-174
165 324. Saji, M., Vasko, V., Kada, F., Allbritton, E. H., Burman, K. D., and Ringel, M. D. (2005) Biochem Biophys Res Commun 332(1), 167-173
325. Berman, H. M., Westbrook, J., Feng, Z., Gilliland, G., Bhat, T. N., Weissig, H., Shindyalov, I. N., and Bourne, P. E. (2000) Nucleic Acids Res 28(1), 235-242
326. Goodsell, D. S., and Olson, A. J. (1990) Proteins 8(3), 195-202
327. Rosenfeld, R. J., Goodsell, D. S., Musah, R. A., Morris, G. M., Goodin, D. B., and Olson, A. J. (2003) J Comput Aided Mol Des 17(8), 525-536
328. Motti, M. L., Califano, D., Troncone, G., De Marco, C., Migliaccio, I., Palmieri, E., Pezzullo, L., Palombini, L., Fusco, A., and Viglietto, G. (2005) Am J Pathol 166(3), 737-749
329. Vasko, V. V., and Saji, M. (2007) Curr Opin Oncol 19(1), 11-17
330. Zhu, J., Huang, J. W., Tseng, P. H., Yang, Y. T., Fowble, J., Shiau, C. W., Shaw, Y. J., Kulp, S. K., and Chen, C. S. (2004) Cancer Res 64(12), 4309-4318.
331. Scrutton, N. S., and Raine, A. R. (1996) Biochem J 319 (Pt 1), 1-8
332. Kulp, S. K., Yang, Y. T., Hung, C. C., Chen, K. F., Lai, J. P., Tseng, P. H., Fowble, J. W., Ward, P. J., and Chen, C. S. (2004) Cancer Res 64(4), 1444-1451
333. Hofmann, C., Shepelev, M., and Chernoff, J. (2004) J Cell Sci 117(Pt 19), 4343- 4354
334. Holm, C., Rayala, S., Jirstrom, K., Stal, O., Kumar, R., and Landberg, G. (2006) J Natl Cancer Inst 98(10), 671-680
335. Rayala, S. K., Molli, P. R., and Kumar, R. (2006) Cancer Res 66(12), 5985-5988
336. Wang, R. A., Zhang, H., Balasenthil, S., Medina, D., and Kumar, R. (2006) Oncogene 25(20), 2931-2936
166 337. Sansal, I., and Sellers, W. R. (2004) J Clin Oncol 22(14), 2954-2963
338. Kim, D. W., Hwang, J. H., Suh, J. M., Kim, H., Song, J. H., Hwang, E. S., Hwang, I. Y., Park, K. C., Chung, H. K., Kim, J. M., Park, J., Hemmings, B. A., and Shong, M. (2003) Mol Endocrinol 17(7), 1382-1394
339. Miyagi, E., Braga-Basaria, M., Hardy, E., Vasko, V., Burman, K. D., Jhiang, S., Saji, M., and Ringel, M. D. (2004) Mol Carcinog 41(2), 98-107
340. Emami, K. H., and Corey, E. (2007) Cancer Lett 253(2), 170-179
341. Mohinta, S., Wu, H., Chaurasia, P., and Watabe, K. (2007) Front Biosci 12, 4020- 4033
342. Lee, C. H., Jeon, Y. T., Kim, S. H., and Song, Y. S. (2007) Biofactors 29(1), 19- 35
343. Bolos, V., Grego-Bessa, J., and de la Pompa, J. L. (2007) Endocr Rev 28(3), 339- 363
344. Lauth, M., and Toftgard, R. (2007) Curr Opin Investig Drugs 8(6), 457-461
345. Chang, C. F., Westbrook, R., Ma, J., and Cao, D. (2007) Front Biosci 12, 4393- 4401
346. Eng, C. (1999) Recent Prog Horm Res 54, 441-452; discussion 453
347. Sasaki, A. T., Chun, C., Takeda, K., and Firtel, R. A. (2004) J Cell Biol 167(3), 505-518
348. Morgenstern, B. (1999) Bioinformatics 15(3), 211-218
349. Jung, I. D., Lee, J., Lee, K. B., Park, C. G., Kim, Y. K., Seo, D. W., Park, D., Lee, H. W., Han, J. W., and Lee, H. Y. (2004) Eur J Biochem 271(8), 1557-1565
167 350. Lee, J., Jung, I. D., Chang, W. K., Park, C. G., Cho, D. Y., Shin, E. Y., Seo, D. W., Kim, Y. K., Lee, H. W., Han, J. W., and Lee, H. Y. (2005) Exp Cell Res 307(2), 315-328
351. Schober, M., Raghavan, S., Nikolova, M., Polak, L., Pasolli, H. A., Beggs, H. E., Reichardt, L. F., and Fuchs, E. (2007) J Cell Biol 176(5), 667-680
352. Goto, H., Tanabe, K., Manser, E., Lim, L., Yasui, Y., and Inagaki, M. (2002) Genes Cells 7(2), 91-97
353. Li, Q. F., Spinelli, A. M., Wang, R., Anfinogenova, Y., Singer, H. A., and Tang, D. D. (2006) J Biol Chem 281(45), 34716-34724
354. Tang, D. D., Bai, Y., and Gunst, S. J. (2005) Biochem J 388(Pt 3), 773-783
355. Jordan, V. C. (2006) J Natl Cancer Inst 98(10), 657-659
356. Rayala, S. K., and Kumar, R. (2007) Biomed Pharmacother
357. Singh, R. R., Song, C., Yang, Z., and Kumar, R. (2005) J Biol Chem 280(18), 18130-18137
358. Jung, H. S., Kim, D. W., Jo, Y. S., Chung, H. K., Song, J. H., Park, J. S., Park, K. C., Park, S. H., Hwang, J. H., Jo, K. W., and Shong, M. (2005) Mol Endocrinol 19(11), 2748-2759
359. Roberts, P. J., and Der, C. J. (2007) Oncogene 26(22), 3291-3310
360. Knaus, U. G., Morris, S., Dong, H. J., Chernoff, J., and Bokoch, G. M. (1995) Science 269(5221), 221-223
361. Tuazon, P. T., Lorenson, M. Y., Walker, A. M., and Traugh, J. A. (2002) FEBS Lett 515(1-3), 84-88
362. Foster, D. B., Shen, L. H., Kelly, J., Thibault, P., Van Eyk, J. E., and Mak, A. S. (2000) J Biol Chem 275(3), 1959-1965
168 363. Ara, T., and DeClerck, Y. A. (2006) Cancer Metastasis Rev 25(4), 645-657
364. Cicek, M., and Oursler, M. J. (2006) Cancer Metastasis Rev 25(4), 635-644
365. Porchia, L. M., Guerra, M., Wang, Y. C., Zhang, Y., Espinosa, A. V., Shinohara, M., Kulp, S. K., Kirschner, L. S., Saji, M., Chen, C. S., and Ringel, M. D. (2007) Mol Pharmacol
366. Beyersmann, D. (2000) Exs 89, 11-28
367. An, W. (2007) Subcell Biochem 41, 351-369
368. Kuo, M. H., and Allis, C. D. (1998) Bioessays 20(8), 615-626
369. Mayo, M. W., Denlinger, C. E., Broad, R. M., Yeung, F., Reilly, E. T., Shi, Y., and Jones, D. R. (2003) J Biol Chem 278(21), 18980-18989
370. Liu, T., Kuljaca, S., Tee, A., and Marshall, G. M. (2006) Cancer Treat Rev 32(3), 157-165
371. Glozak, M. A., Sengupta, N., Zhang, X., and Seto, E. (2005) Gene 363, 15-23
372. Gray, S. G., and Ekstrom, T. J. (2001) Exp Cell Res 262(2), 75-83
373. Marks, P. A., Miller, T., and Richon, V. M. (2003) Curr Opin Pharmacol 3(4), 344-351
374. De Nigris, F., Cerutti, J., Morelli, C., Califano, D., Chiariotti, L., Viglietto, G., Santelli, G., and Fusco, A. (2002) Br J Cancer 87(12), 1479
375. Frye, R. (2002) Br J Cancer 87(12), 1479
376. MacRae, T. H. (1997) Eur J Biochem 244(2), 265-278
377. Rosenbaum, J. (2000) Curr Biol 10(21), R801-803 169 378. Hubbert, C., Guardiola, A., Shao, R., Kawaguchi, Y., Ito, A., Nixon, A., Yoshida, M., Wang, X. F., and Yao, T. P. (2002) Nature 417(6887), 455-458
379. Dompierre, J. P., Godin, J. D., Charrin, B. C., Cordelieres, F. P., King, S. J., Humbert, S., and Saudou, F. (2007) J Neurosci 27(13), 3571-3583
380. Haggarty, S. J., Koeller, K. M., Wong, J. C., Grozinger, C. M., and Schreiber, S. L. (2003) Proc Natl Acad Sci U S A 100(8), 4389-4394
381. Rombouts, K., Knittel, T., Machesky, L., Braet, F., Wielant, A., Hellemans, K., De Bleser, P., Gelman, I., Ramadori, G., and Geerts, A. (2002) J Hepatol 37(6), 788-796
382. Ono, S., Oue, N., Kuniyasu, H., Suzuki, T., Ito, R., Matsusaki, K., Ishikawa, T., Tahara, E., and Yasui, W. (2002) J Exp Clin Cancer Res 21(3), 377-382
383. Zhang, X., Yuan, Z., Zhang, Y., Yong, S., Salas-Burgos, A., Koomen, J., Olashaw, N., Parsons, J. T., Yang, X. J., Dent, S. R., Yao, T. P., Lane, W. S., and Seto, E. (2007) Mol Cell 27(2), 197-213
384. Tran, A. D., Marmo, T. P., Salam, A. A., Che, S., Finkelstein, E., Kabarriti, R., Xenias, H. S., Mazitschek, R., Hubbert, C., Kawaguchi, Y., Sheetz, M. P., Yao, T. P., and Bulinski, J. C. (2007) J Cell Sci 120(Pt 8), 1469-1479
385. Johnstone, R. W., and Licht, J. D. (2003) Cancer Cell 4(1), 13-18
386. Subramanian, C., Opipari, A. W., Jr., Bian, X., Castle, V. P., and Kwok, R. P. (2005) Proc Natl Acad Sci U S A 102(13), 4842-4847
387. Chen, C. L., Sung, J., Cohen, M., Chowdhury, W. H., Sachs, M. D., Li, Y., Lakshmanan, Y., Yung, B. Y., Lupold, S. E., and Rodriguez, R. (2006) J Pharmacol Exp Ther 319(2), 533-542
388. Uchida, H., Maruyama, T., Ono, M., Ohta, K., Kajitani, T., Masuda, H., Nagashima, T., Arase, T., Asada, H., and Yoshimura, Y. (2007) Endocrinology 148(2), 896-902
170 389. Kitazono, M., Robey, R., Zhan, Z., Sarlis, N. J., Skarulis, M. C., Aikou, T., Bates, S., and Fojo, T. (2001) J Clin Endocrinol Metab 86(7), 3430-3435
390. Zarnegar, R., Brunaud, L., Kanauchi, H., Wong, M., Fung, M., Ginzinger, D., Duh, Q. Y., and Clark, O. H. (2002) Surgery 132(6), 984-990; discussion 990
391. Greenberg, V. L., Williams, J. M., Cogswell, J. P., Mendenhall, M., and Zimmer, S. G. (2001) Thyroid 11(4), 315-325
392. Catalano, M. G., Fortunati, N., Pugliese, M., Costantino, L., Poli, R., Bosco, O., and Boccuzzi, G. (2005) J Clin Endocrinol Metab 90(3), 1383-1389
393. Imanishi, R., Ohtsuru, A., Iwamatsu, M., Iioka, T., Namba, H., Seto, S., Yano, K., and Yamashita, S. (2002) J Clin Endocrinol Metab 87(10), 4821-4824
394. Kitazono, M., Bates, S., Fok, P., Fojo, T., and Blagosklonny, M. V. (2002) Cancer Biol Ther 1(6), 665-668
395. Finnin, M. S., Donigian, J. R., Cohen, A., Richon, V. M., Rifkind, R. A., Marks, P. A., Breslow, R., and Pavletich, N. P. (1999) Nature 401(6749), 188-193
396. Lu, Q., Wang, D. S., Chen, C. S., Hu, Y. D., and Chen, C. S. (2005) J Med Chem 48(17), 5530-5535
397. Lu, Q., Yang, Y. T., Chen, C. S., Davis, M., Byrd, J. C., Etherton, M. R., Umar, A., and Chen, C. S. (2004) J Med Chem 47(2), 467-474
398. Marks, P. A., Richon, V. M., Kelly, W. K., Chiao, J. H., and Miller, T. (2004) Novartis Found Symp 259, 269-281; discussion 281-268
399. Pan, L., Lu, J., Wang, X., Han, L., Zhang, Y., Han, S., and Huang, B. (2007) Cancer 109(8), 1676-1688
400. Mitsiades, C. S., Poulaki, V., McMullan, C., Negri, J., Fanourakis, G., Goudopoulou, A., Richon, V. M., Marks, P. A., and Mitsiades, N. (2005) Clin Cancer Res 11(10), 3958-3965
171 401. Gillespie, S., Borrow, J., Zhang, X. D., and Hersey, P. (2006) Apoptosis 11(12), 2251-2265
402. Sawa, H., Murakami, H., Ohshima, Y., Sugino, T., Nakajyo, T., Kisanuki, T., Tamura, Y., Satone, A., Ide, W., Hashimoto, I., and Kamada, H. (2001) Brain Tumor Pathol 18(2), 109-114
403. Gao, T., Furnari, F., and Newton, A. C. (2005) Mol Cell 18(1), 13-24
172