UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The role of the protein tyrosine phosphatase PRL-3 in regulating cell signaling in cancerous and non-cancerous… Bessette, Darrell Christopher 2010

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
24-ubc_2011_spring_bessette_darrell.pdf [ 5.08MB ]
Metadata
JSON: 24-1.0071524.json
JSON-LD: 24-1.0071524-ld.json
RDF/XML (Pretty): 24-1.0071524-rdf.xml
RDF/JSON: 24-1.0071524-rdf.json
Turtle: 24-1.0071524-turtle.txt
N-Triples: 24-1.0071524-rdf-ntriples.txt
Original Record: 24-1.0071524-source.json
Full Text
24-1.0071524-fulltext.txt
Citation
24-1.0071524.ris

Full Text

The Role of the Protein Tyrosine Phosphatase PRL-3 in Regulating Cell Signaling in Cancerous and Non-Cancerous Cell Lines  by  DARRELL CHRISTOPHER BESSETTE  B.Sc., Simon Fraser University, 1999 M.Sc., Simon Fraser University, 2003  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY  in  THE FACULTY OF GRADUATE STUDIES  (Pathology and Laboratory Medicine)  THE UNIVERSITY OF BRITISH COLUMBIA  (Vancouver)  December 2010  © Darrell Christopher Bessette, 2010  Abstract Protein tyrosine phosphorylation is an important mechanism that regulates complex intracellular signaling pathways that determine many cellular activities, such as proliferation, growth, and differentiation. Phosphorylation is regulated by the concerted activities of protein tyrosine kinases and protein tyrosine phosphatases (PTPs), which respectively add and remove phosphate to and from proteins. PRL-3 is a member of a novel subfamily of PTPs that possess prenylation motifs at the C-terminus, thus effecting membrane targeting of these proteins. High PRL-3 expression is an indicator of disease progression and poor prognosis in a number of carcinomas.  While some of the biology of PRL-3 has been determined, the molecular  mechanisms by which PRL-3 can impart enhanced malignancy to cells and tumours, and the means by which it is upregulated in these cancers are not well understood. I established three independent cell systems to investigate the role of PRL-3 in signaling: an inducible system expressing wild-type and catalytically-inactive Flag-tagged PRL-3 in HEK 293 (293) cells, a constitutive system expressing EGFP-tagged wild-type, catalytically-inactive and prenylationdeficient PRL-3 in LNCaP prostate carcinoma cells, and a shRNA-mediated stable knockdown of PRL-3 in LNCaP, C4-2 and DU-145 prostate carcinoma cells. I examined the effect of altering PRL-3 expression in these cell systems upon phenotypic characteristics of proliferation, migration, and invasion, and on modulating the expression or activation of a number of signaling molecules.  PRL-3 was found to promote invasion and/or migration of 293 and prostate  carcinoma cells. PRL-3 limited the proliferation of LNCaP cells, but had no effect on the proliferation of the other cell types. PRL-3 expression reduced E-cadherin expression in 293 cells but did not alter this or other EMT marker expression in prostate cancer cells. While PRL-3 expression had little effect on cell signaling in untreated 293 or prostate cancer cells,  ii  overexpression of PRL-3 in growth factor-stimulated 293 cells regulated Mek-dependent Erk activity.  PRL-3 localized at the plasma membrane with adherens junction proteins but did not  alter E-cadherin, β- catenin or α-catenin expression or interactions. These studies support the findings that PRL-3 imparts metastasis-associated properties upon different cell types but the molecular mechanisms behind this behaviour remain unresolved.  iii  Preface  The research carried out in this dissertation was approved by UBC Biohazards Committee (certificate numbers: H01-0093, H06-099, H08-0107).  iv  Table of Contents Abstract .......................................................................................................................................... ii Preface ........................................................................................................................................... iv Table of Contents .......................................................................................................................... v List of Tables .............................................................................................................................. viii List of Figures ............................................................................................................................... ix Acknowledgements ...................................................................................................................... xi Dedication .................................................................................................................................... xii Chapter 1: Introduction ............................................................................................................... 1 1.1: Protein phosphorylation ....................................................................................................... 1 1.2: Protein tyrosine phosphatases (PTP) superfamily ............................................................... 3 1.3: Phosphatase of regenerating liver (PRL)-3 .......................................................................... 6 1.3.1: PRL-3 structure and relationship to other PTP superfamily members ......................... 6 1.3.2: Description of the human and murine PRL-3 gene .................................................... 11 1.3.3: Regulation of expression and localization of PRL-3 .................................................. 13 1.3.4: PRL-3 expression in human cancers ........................................................................... 16 1.3.5: PRL-3 expression imparts metastasis-associated properties to cell lines ................... 24 1.3.6: PRL-3 and angiogenesis ............................................................................................. 29 1.3.7: The signaling pathways of PRL-3 .............................................................................. 30 1.3.8: Physiological and therapeutic regulation of PRL-3 .................................................... 40 1.4: Insulin-like growth factor (IGF)-I signaling ...................................................................... 43 1.5: Adherens junction complexes ............................................................................................ 44 1.6: Prostate cancer ................................................................................................................... 45 1.7: Rationale and hypothesis ................................................................................................... 46 Chapter 2: Materials and Methods ........................................................................................... 49 2.1: Generation and analysis of Flp-In™ T-Rex™ HEK 293 cells that inducibly express FLAG-PRL-3 wildtype and FLAG-PRL-3 mutant (C104S) .................................................... 49 2.1.1: Cell culture .................................................................................................................. 49 2.1.2: Generation of the FLAG-tagged PRL-3 wild-type and catalytically-inactive mutant C104S vectors ....................................................................................................................... 49 2.1.3: Generation of the FlpIn T-Rex-293-FLAG-PRL-3 cell lines ..................................... 51 2.1.4: 293 cell proliferation assay ......................................................................................... 52 2.1.5: 293 cell migration assay ............................................................................................. 53 2.1.6: 293 cell lysis, immunoprecipitation and immunoblotting .......................................... 53 2.1.7: Antibodies and reagents .............................................................................................. 54 2.2: Generation and analysis of LNCaP cells stably overexpressing EGFP-PRL-3 (wt) and EGFP-PRL-3 mutants (C104S, C170S).................................................................................... 55 2.2.1: Prostate and breast cancer cell lines and antibodies ................................................... 55 2.2.2: Generation of the pEGFP-C2-PRL-3 expression vectors ........................................... 56 2.2.3: Generation of stable GFP fusion protein LNCaP cell lines ........................................ 57 2.2.4: Immunofluorescence assay ......................................................................................... 58 2.2.5: GFP-PRL-3 LNCaP cell proliferation assay............................................................... 58 2.2.6: GFP-PRL-3 LNCaP cell lysis and immunoblotting ................................................... 58 2.2.7: Treatment of prostate cancer and MCF-7 cells to induce p53 expression .................. 59  v  2.3: Generation and analysis of prostate cancer cells with transient and stable PRL-3 knockdown ................................................................................................................................ 59 2.3.1: Cell lines and antibodies ............................................................................................. 59 2.3.2: Determination of proliferation and long-term survival in siRNA-treated PCa cells .. 59 2.3.3: Generation of stable PRL-3 knockdown prostate cancer lines .................................. 60 2.3.4: RNA isolation and semi-quantitative RT-PCR........................................................... 61 2.3.5: Immunohistochemistry and immunofluorescence ...................................................... 62 2.3.6: Determination of proliferation and long-term survival of the lentiviral shRNA transduced PCa cells ............................................................................................................. 63 2.3.7: shRNA-transduced prostate cancer cell migration and invasion ................................ 63 2.3.8: Colony formation assays ............................................................................................. 64 2.3.9: Prostate cancer cell lysis and immunoblotting ........................................................... 65 Chapter 3: The Role of FLAG-PRL-3 Expression in FlpIn TRex HEK 293T Cells ............ 66 3.1: Generation of inducible FLAG-PRL-3 cell lines ............................................................... 66 3.2: FLAG-PRL-3 expression does not affect proliferation but increases migration of 293 cells in a phosphatase-independent fashion ...................................................................................... 68 3.3: FLAG-PRL-3 expression does not affect global tyrosine phosphorylation or Erk or Akt cell signaling in unstimulated 293 cells .................................................................................... 70 3.4: Tyrosine phosphorylation in response to growth factor stimulation is unaffected by PRL-3 ................................................................................................................................................... 73 3.5: PRL-3 can modulate Erk but not JNK or p38 MAPK or Akt signaling in FlpIn TRex™ 293 cells upon IGF-I stimulation .............................................................................................. 75 3.6: FLAG-PRL-3 expression enhances Erk but not Akt activation in EGF-stimulated 293 cells ........................................................................................................................................... 78 3.7: Mek activity is required for PRL-3-mediated enhancement of Erk activation in IGF-I stimulated cells.......................................................................................................................... 79 3.8: Src activation is unaffected by FLAG-PRL-3 expression in IGF-I-stimulated 293 cells .. 81 3.9: Summary ............................................................................................................................ 83 Chapter 4. The Role of PRL-3 Overexpression in Prostate Cancer...................................... 86 4.1: Generation of GFP-PRL-3 overexpression LNCaP cell lines ........................................... 88 4.2: GFP-PRL-3 (wt) localizes to the cell membrane in a phosphatase-independent but prenylation-dependent fashion .................................................................................................. 90 4.3: GFP-PRL-3 expression decreases LNCaP cell proliferation ............................................. 93 4.4: GFP-PRL-3 colocalizes with components of the adherens junction in a phosphataseindependent but prenylation-dependent manner ....................................................................... 94 4.5: Expression of adherens junction proteins is unaltered by GFP-PRL-3 ............................. 98 4.6: EMT is unaltered by GFP-PRL-3 expression in LNCaP cells......................................... 100 4.7: GFP-PRL-3 expression does not alter Csk expression or Src signaling in LNCaP cells 101 4.8: Doxorubicin treatment of prostate and breast cancer cell lines leads to increased p53 protein expression but decreased PRL-3 protein expression .................................................. 102 4.9: CoCl2 treatment of prostate and breast cancer cell lines leads to increases in p53 protein but not in PRL-3 protein levels ............................................................................................... 105 4.10: Summary ........................................................................................................................ 108 Chapter 5: The Effect of PRL-3 Ablation on Prostate Cancer ............................................ 111 5.1: Transient ablation of PRL-3 does not affect cell proliferation but decreases long-term viability in prostate cancer cells .............................................................................................. 112  vi  5.2: Generation of stable PRL-3 shRNA prostate cancer cell lines ........................................ 115 5.3: Loss of PRL-3 expression increases LNCaP cell proliferation. ...................................... 118 5.4: Loss of PRL-3 results in decreased migratory and invasive abilities of prostate cancer cells ......................................................................................................................................... 121 5.5: Loss of PRL-3 expression in the PRL-3 shRNA DU-145 cell lines reduces anchorageindependent growth ................................................................................................................. 125 5.6: Loss of PRL-3 slightly decreases the expression of the cell cycle proteins p27Kip1 and p21Waf1/Cip1 in C4-2 cells ......................................................................................................... 126 5.7: Loss of PRL-3 has little effect on EMT-marker protein expression in prostate cancer cells ................................................................................................................................................. 128 5.8: PRL-3 knockdown does not alter total cellular tyrosine protein phosphorylation but reduces phosphorylated Src Tyr529 and Erk in prostate cancer cells .................................... 130 5.9: Loss of PRL-3 does not alter PTEN expression or Akt Ser473 phosphorylation............ 132 5.10: Summary ........................................................................................................................ 134 Chapter 6: Discussion .............................................................................................................. 139 6.1: The role of PRL-3 in controlling cell proliferation and the cell cycle ............................. 141 6.2: The role of PRL-3 in mediating EMT and the metastatic properties of migration and invasion ................................................................................................................................... 145 6.3: The role of PRL-3 in cell signaling pathways ................................................................. 151 6.4: The role of PRL-3 isoform 1 vs. isoform 2 – which is the active isoform? .................... 153 6.5: Conclusions...................................................................................................................... 157 References .................................................................................................................................. 158 Appendix 1: Supplemental figures .......................................................................................... 174  vii  List of Tables Table 1.1: Cancer related processes affected by altered PRL-3 expression. ................................ 26 Table 1.2: Molecular effects of altered PRL-3 expression. .......................................................... 38  viii  List of Figures Figure 1.1: Schematic structures and subgroups of representative PTPs. ...................................... 4 Figure 1.2: The sequence alignment of murine PRL-1, -2, and -3. ................................................ 7 Figure 1.3: Schematic diagram of the PRL proteins. ...................................................................... 8 Figure 1.4: Sequence alignment of full-length and truncated PRL-3 isoforms. ........................... 12 Figure 1.5: Sequence alignment between murine and human orthologs of PRL-3. ..................... 13 Figure 3.1: The expression of FLAG-PRL-3 (wt) and (C104S) can be induced with doxycycline in the transfected FlpIn TRex 293 cell lines. ................................................................................ 68 Figure 3.2: FLAG-PRL-3 expression in FlpIn TRex 293 cells does not affect cell prolilferation but increases migration to fibronectin........................................................................................... 70 Figure 3.3: FLAG-PRL-3 expression has little effect on protein phosphorylation in unstimulated cells. .............................................................................................................................................. 72 Figure 3.4: FLAG-PRL-3 expression does not affect total protein tyrosine phosphorylation in IGF-I stimulated 293 cells. ........................................................................................................... 74 Figure 3.5: The expression of FLAG-PRL-3 can modulate Erk, but not Akt, p38 or JNK phosphorylation in response to IGF-1........................................................................................... 76 Figure 3.6: PRL-3 expression enhances Erk but not Akt phosphorylation in EGF-stimulated cells. ....................................................................................................................................................... 79 Figure 3.7: PRL-3-mediated enhancement of Erk activation after IGF-1 stimulation operates through the MAPKK Mek1/2. ...................................................................................................... 80 Figure 3.8: IGF-I and/or PRL-3 expression do not affect Src activation in 293 cells. ................. 82 Figure 4.1: Cytoplasmic PRL-3 correlates with increased aggressiveness of prostate cancer tumours. ........................................................................................................................................ 87 Figure 4.2: Enrichment of high intensity GFP fluorescent cells in the GFP-PRL-3 cell lines by flow cytometry sorting. ................................................................................................................. 89 Figure 4.3: GFP-PRL-3 localizes to the cell membrane in a prenylation-dependent but phosphatase-independent manner. ................................................................................................ 91 Figure 4.4: The role of GFP-PRL-3 (wt) and its mutants in regulating LNCaP cell proliferation. ....................................................................................................................................................... 94 Figure 4.5: GFP-PRL-3 (wt) and (C104S) but not (C170S) can localize at the cell membrane with the adherens junction proteins E-cad, β-cat and α-cat. ......................................................... 95 Figure 4.6: GFP-PRL-3 does not alter expression or interactions of adherens junction proteins in LNCaP cells. ................................................................................................................................. 99 Figure 4.7: Expression of the mesenchymal markers FN and N-cad are unaffected by GFP-PRL3................................................................................................................................................... 100 Figure 4.8: GFP-PRL-3 does not alter Csk expression or Src activity in LNCaP cells. ............. 101 Figure 4.9: PRL-3 is not upregulated in prostate or breast cancer cells in response to doxorubicin (doxo). ......................................................................................................................................... 104 Figure 4.10: PRL-3 is not upregulated in prostate or breast cancer cells in response to cobalt chloride. ...................................................................................................................................... 107 Figure 5.1: Ablation of PRL-3 by siRNA decreases long-term survival of prostate cancer cell lines but does not affect proliferation. ........................................................................................ 113 Figure 5.2: Prostate cancer cell lines transduced with lentivirus particles containing PRL-3 shRNA have reduced PRL-3 expression..................................................................................... 116 Figure 5.3: Ablation of PRL-3 increases cell proliferation in LNCaP prostate cancer cells. ..... 119  ix  Figure 5.4: PRL-3 ablation reduces cell migration of prostate cancer cells. .............................. 123 Figure 5.5: PRL-3 ablation reduces cell invasion of prostate cancer cells. ................................ 124 Figure 5.6: Loss of PRL-3 reduces anchorage-independent growth of DU-145 prostate cancer cells. ............................................................................................................................................ 126 Figure 5.7: Ablation of PRL-3 protein expression slightly decreases p27Kip1 and p21Waf1/Cip1 protein expression in C4-2 cells. ................................................................................................. 128 Figure 5.8: EMT protein expression is minimally affected by PRL-3 knockdown in prostate cancer cell lines. .......................................................................................................................... 129 Figure 5.9: PRL-3 knockdown does not change total cellular protein tyrosine phosphorylation in prostate cancer cell lines. ............................................................................................................ 131 Figure 5.10: Erk phosphorylation is reduced in PRL-3 knockdown prostate cancer cell lines. . 132 Figure 5.11: Ablation of PRL-3 does not change phosphorylation of Ser473 of Akt or PTEN expression in prostate cancer cell lines. ...................................................................................... 133 Figure 6.1: Schematic of hPRL-3 domains in the full-length isoform vs. the truncated isoform. ..................................................................................................................................................... 154  x  Acknowledgements I wish to thank my supervisor, Dr. Catherine Pallen for providing me the opportunity to conduct doctoral research in her laboratory. Her excellent mentorship, patience and guidance were invaluable in my studies. I would like to thank my the chair of my supervisory committee, Dr. Wan Lam and the members of my committee, Dr. Marcel Bally and Dr. Michael Cox for their invaluable insight and advice on my project. I would like to thank the members of my lab for encouragement and advice and especially Dr. Jing Wang for being an excellent technical resource and keeping lab operations smooth. I would like to thank Dr. Michael Anglesi, for providing technical advice, Michael Butt for assistance with microscopy, and Dr. Sandra Dunnn and Dr. Qi Zeng for providing reagents used in my thesis. This research was funded in part by a CIHR Canada Graduate Scholarship, a Foundation for Prostate Disease Fellowship and a Roman M. Babick Medical Fellowship in Cancer Research.  xi  Dedication  To my beautiful wife, Pam.  xii  Chapter 1: Introduction  The development of multi-cellular organisms relies on modulating many different cellular signals to progress from an undifferentiated single cell to a complex, highly differentiated individual. Cell signaling refers to the process by which input signals are analyzed and responded to by the target cell(s).  This signaling may control one or more of many  important cellular events, regulating such processes as cell proliferation, growth, differentiation, apoptosis and cell polarity. In addition to regulating organismal development, cell signaling is critical for cellular homeostasis, responding to environmental stresses and coping with disease, both pathogenic and intrinsic (i.e. cancer). The misregulation of signaling processes can lead to cellular imbalances and disease, such as cancer.  Understanding the developmental and  homeostasis contexts of cellular signaling and the components involved is crucial for understanding both development and disease.  1.1: Protein phosphorylation Cell signaling pathways may exploit several different mechanisms, either individually or in combination with other mechanisms, to elicit the desired (or undesired, in the case of disease processes) responses. For example, binding of hormone to the androgen receptor forms a transcriptional complex that can directly regulate androgen-responsive genes through binding these genes’ promoters (Klocker et al., 1994). The Notch signaling pathway involves proteolytic cleavage of the transmembrane protein Notch subsequent to binding with its substrate(s), with the resulting intracellular fragment translocating to the nucleus and initiating transcription of target genes (Greenwald, 1998). Another form of signaling that is used is the regulation of  1  membrane polarity, through the action of ion pumps and voltage-gated ion channels which generate action potentials and mediate a number of cellular responses, including sensory neuron excitation (Resta and Becchetti, 2010; Swartz, 2008). A major class of cell signaling is through the regulation of phosphorylation of proteins through the coordinated actions of protein kinases and protein phosphatases (Hardie, 1990). The addition of phosphate groups to proteins is mediated by the enzymatic activity of a superfamily of proteins termed kinases. The principle of protein phosphorylation is the transfer of the terminal phosphate group of adenosine 5’ triphosphate (ATP) to a hydroxyl group of an amino acid residue in a protein, releasing an adenosine 5’ diphosphate (ADP) molecule. This imparts a net negative charge to the protein and can affect substrate binding or conformational changes in the target protein. There are two main classes of kinases – protein tyrosine kinases (PTKs) which phosphorylate tyrosine residues (Wilks, 1990) and serine/threonine kinases which phosphorylate serine and threonine residues (Blackshear et al., 1988; Edelman et al., 1987). These classes of enzymes are often found in the same pathway and may be activated in a “cascading” mechanism.  An external stimulus activates cell surface receptors which then  activate upstream kinases by phosphorylation, leading to activation of downstream kinases by phosphorylation and, finally, activation of the cellular response, frequently through activation of transcriptional regulatory proteins (Ahn et al., 1992; Karin, 1992). Removal of phosphate from phosphorylated proteins is accomplished by the activity of a superfamily of proteins called protein phosphatases (PPs).  These enzymes catalyze the  hydrolysis of the phosphate group from the phosphoprotein (Barrett et al., 1969; Sparks and Brautigan, 1986).  There are two families of protein phosphatases – the serine/threonine  phosphatases (Cohen, 1989) and the protein tyrosine phosphatases (PTPs) (Fischer et al., 1990;  2  Wang et al., 2003). The serine/threonine phosphatase family is comprised of four members which appear to dephosphorylate all cytosolic and nuclear proteins phosphorylated by serine/threonine kinases (Hardie, 1990). In contrast, the PTP superfamily is comprised of over 100 members (Dewang et al., 2005; Wang et al., 2003). Initially, the protein phosphatases were believed to be “housekeeping” enzymes, returning cells to the “resting” state that occurred before cell stimulation and the activation of kinases. However, it has become apparent that protein phosphatases may induce signaling pathways on their own merit (Tonks and Charbonneau, 1989).  1.2: Protein tyrosine phosphatases (PTP) superfamily The founding member of the PTP superfamily, PTP-1was purified from human placental tissue as two different polypeptides, termed 1A and 1B (Tonks et al., 1988). Subsequently a group cloned the cDNA of PTP-1 from a human placental cDNA library and found that a longer version of PTP-1 was encoded by the cDNA, and the PTP was termed PTP-1B (Chernoff et al., 1990). Since the initial isolation and characterization of PTP-1B, the family of PTPs has grown extensively to encompass at least 107 members (Alonso et al., 2004). A key feature of the PTPs is the active site sequence C(X)5R, termed the PTP signature motif, in the catalytic domain (Zhang, 2003). There are four subfamilies of PTP grouped on the basis of their structure and substrate specificity: the classical PTPs, the dual-specific phosphatases (DSPs), the cdc25 phosphatases, and the low molecular weight (LMW) PTPs (Fig. 1.1) (Alonso et al., 2004; Wang et al., 2003).  3  Figure 1.1: Schematic structures and subgroups of representative PTPs. PTP, protein tyrosine phosphatase; DSP, dual specific phosphatase; LMW, low molecular weight; ER, endoplasmic reticulum; SH2, src homology 2; PEST, polypeptide sequences enriched in proline (P), glutamate (E), serine (S), and threonine (T); FERM, four-point one, ezrin, radixin, moesin; PDZ, PSD-95, Dlg, and ZO-1 homology; FN, fibronectin; Ig, immunoglobulin; MAM, meprin- A5-mu; CH2, cdc25 homology. Modified from Wang et al. (2003).  4  The classical, phosphotyrosine-specific PTP subfamily is comprised of ~40 members and is further subdivided into intracellular/nonreceptor PTPs (NRPTPs) and transmembrane/receptortype (RPTPs) (Paul and Lombroso, 2003). The RPTPs contain an extracellular domain, a transmembrane domain and one or two cytoplasmic PTP domains. The majority of RPTPs have two cytoplasmic PTP domains and the more membrane proximal domain, D1, has most, if not all, of the phosphatase activity (Wang et al., 2003). The function of the membrane distal PTP domain, D2, remains speculative but it may be involved in regulation of the D1 catalytic activity through intramolecular and intermolecular interactions between the D1 and D2 domains of the RPTPs (Wang et al., 2003). Another possible role for the D2 domain is substrate targeting (Andersen et al., 2001).  The NRPTPs are typified by the founding member of the PTP  superfamily, PTP-1B. The NRPTPs contain a single catalytic domain and various amino or carboxyl terminal extensions that may have targeting or regulatory functions (Wang et al., 2003). The non-classical subfamilies of PTPs comprise the remainder of the identified PTP genes. The DSPs are a structurally and functionally diverse class of PTPs. The family is further subdivided into four groups: VH-1 like, Cdc14-like, phosphoinositide phosphatases, and mRNA 5’-triphosphatases (Wang et al., 2003). The DSPs can remove phosphate groups from both tyrosine and serine/threonine residues. The Cdc25 phosphatases (Cdc25A, B, and C) remove the inhibitory phosphates from Thr14 and Tyr15 in cyclin-dependent kinases to activate the kinases and drive progression through the cell cycle (Dunphy and Kumagai, 1991; Millar and Russell, 1992). Even though the Cdc25 phosphatases are DSPs, they have no significant amino acid sequence similarity with the other PTP subfamilies, apart from the signature PTP motif. The LMW PTPs are ~18 kDa proteins with no significant homology to the other PTPs. LMW PTPs regulate several receptor tyrosine kinase signaling pathways, including those of the receptors for  5  vascular endothelial growth factor (VEGF) (Huang et al., 1999), fibroblast growth factor (FGF) (Park et al., 2002), and platelet-derived growth factor (PDGF) (Chiarugi et al., 2002). The active site motif in the LMW PTPs is near the N-terminus, whereas it is found near the C-terminus of the catalytic domain in all other PTPs (Wang et al., 2003). The mechanism of catalysis utilized by the PTP superfamily requires two sequence motifs, the P-loop and the WPD loop (Barford et al., 1998; Denu et al., 1996; Zhang, 1998). The dephosphorylation of phosphoproteins by PTPs is accomplished by insertion of the substrate phosphoamino acid into the catalytic cleft or pocket, where it is in close proximity to the essential cysteine of the P-loop located at the base of the pocket. The cysteine residue targets the substrate thioester with a nucleophilic attack, resulting in the formation of a covalent thiophosphate intermediate. The phosphate transfer is facilitated by protein donation from an essential general acid (aspartate in the WPD loop) to the phenolic oxygen of the phosphoamino acid of the substrate. The phosphate is released by the catalytic aspartic acid acting as a general base to transfer the phosphate to water. Substrate specificity is maintained, in part, by the size and shape of the catalytic cleft, though PTP localization and regulation of PTP activity through post-translational mechanisms are involved as well. (Soulsby and Bennett, 2009).  1.3: Phosphatase of regenerating liver (PRL)-3 1.3.1: PRL-3 structure and relationship to other PTP superfamily members The founding of the PRL subgroup of PTPs began with the discovery of PRL-1, which was identified as a protein that showed increased levels of expression in regenerating liver, hence the name phosphatase of regenerating liver-1 (Diamond et al., 1994; Mohn et al., 1991). Susequently, two other PRLs were identified, PRL-2 and PRL-3, that are closely related to PRL3 in sequence and structure (Cates et al., 1996; Montagna et al., 1995; Zeng et al., 1998). Human  6  PRL-1 (also known as PTP4A1 and PTPCAAX1) and PRL-2 (also known as PTP4A2, PTPCAAX2, and OV-1) shared the closest amino acid identity at 87%. PRL-1 and PRL-3 (also known as PTP4A3 and PRL3) display 76% sequence identity and PRL-2 and PRL-3 share 75% sequence identity (Figure 1.2). The PRLs are members of the PTP superfamily and are most closely related in sequence to PTEN and Cdc14 DSPs (Zeng et al., 1998). The PRLs have a central catalytic domain with the signature PTP active site sequence CX5R. The short C-terminal region of the PRLs contains a polybasic region immediately preceding a prenylation motif, or CAAX box, at the tail (Fig. 1.3). The prenylation motif is important for membrane localization and intracellular localization (Sun et al., 2005; Zeng et al., 2000). The presence of a CAAX box in the PRLs is unique among the PTPs, however, a polybasic region and adjacent CAAX box are found in the members of the Ras superfamily of GTPases.  Figure 1.2: The sequence alignment of murine PRL-1, -2, and -3. Alignment of amino acid sequence of murine PRL-3 with PRL-2 and PRL-1. Identical residues are shaded. The WPD loop is in bold and the P-loop is underlined.  1 1 1  MARMNRPAPVEVSYRHMRFLITHNPSNATLSTFIEDLKKYGATTVVRVCEVTYDKTPLEK ---MNRPAPVEISYENMRFLITHNPTNATLNKFTEELKKYGVTTLVRVCDATYDKAPVEK MARMNRPAPVEVTYKNMRFLITHNPTNATLNKFIEELKKYGVTTIVRVCEATYDTTLVEK  PRL-3 PRL-2 PRL-1  61 58 61  DGITVVDWPFDDGAPPPGKVVEDWLSLLKAKFYNDPGSCVAVHCVAGLGRAPVLVALALI EGIHVLDWPFDDGAPPPNQIVDDWLNLLKTKFREEPGCCVAVHCVAGLGRAPVLVALALI EGIHVLDWPFDDGAPPSNQIVDDWLSLVKIKFREEPGCCIAVHCVAGLGRAPVLVALALI  PRL-3 PRL-2 PRL-1  121 118 121  ESGMKYEDAIQFIRQKRRGAINSKQLTYLEKYRPKQRLRFKDPHTHKTRCCVM ECGMKYEDAVQFIRQKRRGAFNSKQLLYLEKYRPKMRLRFRDTNGH---CCVQ EGGMKYEDAVQFIRQKRRGAFNSKQLLYLEKYRPKMRLRFKDSNGHRNNCCIQ  PRL-3 PRL-2 PRL-1  7  Crystal structures of PRL-1 (Jeong et al., 2005; Sun et al., 2005) and NMR solution structures for PRL-3 (Kim et al., 2004; Kozlov et al., 2002) have been reported. Also, a molecular model has been reported for PRL-2 from resonance assignments and corresponding secondary structure (Kozlov et al., 2002; Zhou et al., 2003).  As expected, the secondary  structures and tertiary folding of the PRLs are highly similar to one another and place them in the DSP category of PTPs, as was suggested by sequence alignments. Overall, however, the PRLs have low sequence identity with other DSPs (<30%) and have the closest structural homology to the DSPs VHR, KAP, and PTEN (PRL-1 and -3), and to Cdc14 and to MKP (PRL-1) (Jeong et al., 2005; Kozlov et al., 2002; Sun et al., 2005). Both the WPD loop and the P-loop are present in the PTP domains of the PRLs (Fig. 1.3), but they display differences that are likely to affect PRL activity and substrate specificity.  Figure 1.3: Schematic diagram of the PRL proteins. PRL-1, -2, and -3 share a homologous domain structure. The catalytic of PTP domain is required for enzymatic activity, requiring the WPD loop residues for phosphate transfer and the CX5R active site or P-loop residues. A polybasic domain and a prenylation motif (PRL-1, CCIQ; PRL-2, CCVQ; PRL-3, CCVM) are important for regulating subcellular localization of the PRL proteins.  8  The structures of PRL-1 and PRL-3 reveal that the active site pocket is shallow (Jeong et al., 2005; Kim et al., 2004; Kozlov et al., 2002; Sun et al., 2005), which is typical of DSPs. The shallow pocket allows access for the shorter side chains of phosphoserine and phosphothreonine to the catalytic cysteine at the base of the pocket, ensuring that both these and the longer side chain of phosphotyrosine can be dephosphorylated by the DSPs. However, the active site pocket of the PRLs is unusually wide and shallow even compared to the other DSPs. This is due to the presence of shorter and flatter, or less protruding, loops around the active site (Jeong et al., 2005; Kim et al., 2004; Kozlov et al., 2002; Sun et al., 2005). Possibly, the PRLs can accommodate a broad range of substrates and that substrate specificity must be determined by other features. For example, PRL-1 has been determined to form homotrimers, both in vitro and in vivo (Jeong et al., 2005) and there may be important structural determinants in the trimers that impart substrate specificity. PRL-3 has been postulated to also form trimers, as it displays a high structural similarity to PRL-1.  In addition to affecting the topology of the active site pocket, the  protrusions around the active site may influence substrate specificity (Jeong et al., 2005; Kozlov et al., 2002; Sun et al., 2005), as similar secondary structures in DSPs participate in substrate interactions. The X5 residues (VAGLG) of the P-loop of the PRLs are extremely hydrophobic and may also play a role in substrate specificity (Jeong et al., 2005; Kozlov et al., 2002). The WPD loop of the PRLs has the amino acid sequence WPFDD. Site-directed mutagenesis studies have indicated that it is the more C-terminal Asp (Asp72 in PRL-1 and PRL-3) that plays the role of general acid proton donor during the formation of the phosphoenzyme intermediate (Kozlov et al., 2002; Wang et al., 2002). Amino acid substitutions among the PRLs are non-conserved on the opposite surface of the molecule to the active site, but are either identical or conserved on the surface surrounding the active site of PRLs (Jeong et al., 2005; Kozlov et al., 2002). This  9  suggests that these non-conserved faces of the PRL molecules impart homolog-specific substrate specificity. PRLs have low in vitro catalytic activity and several differences between the PRLs and the classical PTPs may account for this (Kozlov et al., 2002; Sun et al., 2005). An Ala is found immediately distal to the invariant Arg of the CX5R motif instead of the usual, highly conserved Ser/Thr residue normally found in this position. This results in the loss of the Ser/Thr hydroxyl group that facilitates the breakdown of the phosphoenzyme intermediate during PTP-mediated catalysis (Denu and Dixon, 1995; Zhang et al., 1995). The function of the supporting hydroxyl group may be provided by a number of mechanisms. It may be supplied in trans, either by the substrate or by PRL-3 interacting proteins. Such interactions may induce conformation changes in the PRLs that promote effective substrate desphosphorylation. Alternatively, the formation of a trimer, as described by Jeong et al., (2005) may be necessary to provide appropriate hydroxyl acceptor functional groups to support catalytic activity. This latter possibility may also explain the low catalytic activity of the PRLs in vitro, as it was observed that PRL-1 is less efficient at forming homotrimers in vitro than in vivo (Jeong et al., 2005). PRL activities may be affected by two types of interactions that were discovered through structure-activity studies. As discussed briefly above, PRL-1 was found to crystallize as a trimer, and to be able to form dimers, trimers and higher oligomers in vivo (Jeong et al., 2005; Sun et al., 2005). The residues that are involved in the interactions between PRL-1 monomers are identical or conserved among the PRLs, suggesting that PRL-2 and PRL-3 may also form oligomers. Consistent with this hypothesis, dimers and trimers form upon cross-linking of recombinant PRL-3 and intermolecular PRL-3 association is detectable in vivo (Sun et al., 2007). A reversible disulfide bond is formed by the intramolecular interaction between Cys49 and the  10  catalytic cysteine (Cys104) within both PRL-1 and PRL-3 (Jeong et al., 2005; Kozlov et al., 2002; Orsatti et al., 2009; Sun et al., 2005; Yu et al., 2007), as well. This suggests that the PRLs may be subject to redox regulation, as appears to be the case for many other PTPs (den Hertog et al., 2005). These redox reactions inactivate the PTPs and are reversible, as long as oxidation does not proceed to a sulfinic or sulfonic acid state. Interestingly, it has been found that in PTP1B, oxidation of the catalytic cysteine rapidly converts to a cyclic sulfenamide form through reaction with the adjacent serine and this form is resistant to further oxidation (den Hertog et al., 2005). As the PTP catalytic site is highly conserved, cyclic sulfenamide formation may be a general mechanism of oxidation.  However, as the PRLs are missing the serine/threonine  adjacent to the catalytic cysteine, the PRLs may be more sensitive to oxidation and more easily irreversibly inactivated. Another mechanism of cysteine inactivation and inhibition of PRL-3 enzymatic activity was reported by Orsatti et al. (2009) and involved a cation mediated conversion of cysteine to glycine after protein purification through a nickel column. These data suggest that regulation of PRL-3 activity can be modulated through several mechanisms. 1.3.2: Description of the human and murine PRL-3 gene The human PRL-3 (hPRL-3) gene is found on the long arm of chromosome 8 (8q) and spans 9613 nucleotides (http://www.ncbi.nlm.nih.gov/gene/11156).  The hPRL-3 coding  sequence is comprised of five exon sequences to produce full-length PRL-3 cDNA. There is an alternatively spliced transcript that can be formed through the deletion of exon four, resulting in a PRL-3 protein that is 25 amino acids shorter than the full-length protein product (i.e. 148 residues vs. 173 residues) (Fig. 1.4A). The murine PRL-3 (mPRL-3) genomic locus is similar in size to the hPRL-3 locus, spanning 10354 nucleotides and also encodes five coding exons, but the  mPRL-3  gene  has  an  additional  exon  in  the  5’  untranslated  region  11  (www.ncbi.nlm.nih.gov/gene/19245). The mPRL-3 gene has a splice variant as well, however, a twenty amino acid fragment is absent from the N-terminal start of the mPRL-3 protein, rather than from the C-terminal proximal region as in the hPRL-3 protein (Fig. 1.4B).  Figure 1.4: Sequence alignment of full-length and truncated PRL-3 isoforms. A. Amino acid alignment of isoform 1 (form 1) and isoform 2 (form 2) of hPRL-3. B. Amino acid alignment of isoform 1 (form 1) and isoform 2 (form 2) of mPRL-3. The Ploop is underlined in both alignments.  A 1 1  MARMNRPAPVEVSYKHMRFLITHNPTNATLSTFIEDLKKYGATTVVRVCEVTYDKTPLEK MARMNRPAPVEVSYKHMRFLITHNPTNATLSTFIEDLKKYGATTVVRVCEVTYDKTPLEK  form 1 form 2  61 61  DGITVVDWPFDDGAPPPGKVVEDWLSLVKAKFCEAPGSCVAVHCVAGLGRAPVLVALALI DGITVVDWPFDDGAPPPGKVVEDWLSLVKAKFCEAPGSCVAVHCVAGLGR----------  form 1 form 2  121 111  ESGMKYEDAIQFIRQKRRGAINSKQLTYLEKYRPKQRLRFKDPHTHKTRCCVM ---------------KRRGAINSKQLTYLEKYRPKQRLRFKDPHTHKTRCCVM  form 1 form 2  1 1  MARMNRPAPVEVSYRHMRFLITHNPSNATLSTFIEDLKKYGATTVVRVCEVTYDKTPLEK -------------------MVRHLQVSEALAGNLQDLKKYGATTVVRVCEVTYDKTPLEK  form 1 form 2  61 42  DGITVVDWPFDDGAPPPGKVVEDWLSLLKAKFYNDPGSCVAVHCVAGLGRAPVLVALALI DGITVVDWPFDDGAPPPGKVVEDWLSLLKAKFYNDPGSCVAVHCVAGLGRAPVLVALALI  form 1 form 2  121 102  ESGMKYEDAIQFIRQKRRGAINSKQLTYLEKYRPKQRLRFKDPHTHKTRCCVM ESGMKYEDAIQFIRQKRRGAINSKQLTYLEKYRPKQRLRFKDPHTHKTRCCVM  form 1 form 2  B  The human and mouse isoform 1 proteins are highly similar in sequence, exhibiting over 95% identity (Fig. 1.5). Only six amino acid substitutions have occurred during the evolution from mouse to man. Three of the six substitutions are highly conserved changes (from mouse to human): R15K, S26T, and L88V. The other three substitutions are non-conserved in nature and are a tandem tripeptide alteration. The mouse sequence Tyr-Asn-Asp is divergent from the  12  human sequence Cys-Glu-Ala. This tripeptide is immediate upstream of the P-loop (residues 9395) and may affect substrate specificity. However, the homologous region in PRL-1 was not reported to be important for intramolecular interactions or trimerization so these residues may not affect molecular interactions (Jeong et al., 2005). This suggests that the murine and human PRL-3 proteins are functionally similar.  Figure 1.5: Sequence alignment between murine and human orthologs of PRL-3. The amino acid sequences of mPRL-3 and hPRL-3 are displayed. There are 6 mismatches out of 173 amino acid residues (96.5% identity): three conserved amino acid substitutions (yellow boxes) and three non-conserved amino acid substitutions (shaded). The P-loop is underlined. 1 1  MARMNRPAPVEVSYRHMRFLITHNPSNATLSTFIEDLKKYGATTVVRVCEVTYDKTPLEK MARMNRPAPVEVSYKHMRFLITHNPTNATLSTFIEDLKKYGATTVVRVCEVTYDKTPLEK  mPRL-3 hPRL-3  61 61  DGITVVDWPFDDGAPPPGKVVEDWLSLLKAKFYNDPGSCVAVHCVAGLGRAPVLVALALI DGITVVDWPFDDGAPPPGKVVEDWLSLVKAKFCEAPGSCVAVHCVAGLGRAPVLVALALI  mPRL-3 hPRL-3  121 121  ESGMKYEDAIQFIRQKRRGAINSKQLTYLEKYRPKQRLRFKDPHTHKTRCCVM ESGMKYEDAIQFIRQKRRGAINSKQLTYLEKYRPKQRLRFKDPHTHKTRCCVM  mPRL-3 hPRL-3  1.3.3: Regulation of expression and localization of PRL-3 PRL-3 is expressed primarily in the heart and skeletal muscle, with lower levels of PRL-3 expression detectable in a number of other tissues (Matter et al., 2001; Zeng et al., 1998). This is in contrast with PRL-1 and -2 protein expression, which were found to be quite widespread, with PRL-2 levels especially high in most tissues examined (Dumaual et al., 2006). The more limited expression of PRL-3 may explain why enhanced PRL-3 expression is associated with quite a number of cancer types. Increased PRL-3 expression has been identified in a number of cancer cell lines, for example, in colorectal, gastric, and breast carcinoma cell lines (Kato et al., 2004; Miskad et al., 2004; Radke et al., 2006; Rouleau et al., 2006). Possibly, tissues that do not  13  express or only express low levels of PRL-3 are unable to cope with elevated levels of PRL-3 activity and thus are susceptible to the tumourigenic effects of PRL-3. In cells, the PRLs are typically associated with the plasma membrane and early endosome (Zeng et al., 2000).  All the PRLs possess a consensus C-terminal CAAX sequence for  prenylation (Cates et al., 1996; Zeng et al., 2000), where C is cysteine, A is an aliphatic residue and X is any amino acid; this sequence in PRL-3 is CCVM. The Cys of the CAAX sequence is post-translationally modified by addition of a lipid molecule that commonly targets proteins to the membrane. Isoprenoid modification of the PRLs has been shown directly in vivo for PRL-2 (Si et al., 2001). Indirect evidence for farnesylation of the PRLs comes from the observation of relocalization of PRL-1, -2, and -3 to the nucleus upon treatment of cells with the farnesyltransferase inhibitor (FTI) (Wang et al., 2002; Zeng et al., 2000). Deletions or mutations of the CAAX sequence that block isoprenoid attachment result in relocalization of the PRL-3 to the cytoplasm and/or nucleus and not to the plasma membrane/early endosome (Si et al., 2001; Sun et al., 2007; Wang et al., 2002; Zeng et al., 2000). PRL action and functions can also be affected by the loss of prenylation. For example, expression of a PRL-3 with a mutant CAAX sequence (CCVM to SCVM: C170S) in SW480 colon carcinoma cells inhibits the migration and invasion properties of these cells. Forced expression of Myc-tagged PRL-3 and various CAAX mutants in B16F1 mouse melanoma cells resulted in reduced adhesion and migration to fibronectin in the PRL-3 mutants with deleted or mutated C170 compared to wild type PRL-3 (Song et al., 2009). Prenylation was also shown to be required to induce PRL-3-mediated metastasis in a tail vein assay in mice, as the CAAX mutants did not display enhanced metastasis compared to mock controls (Song et al., 2009). The PRL-3 CAAX sequence is unique among the PRLs as it has two adjacent Cys residues (CCVM). However, mutation of the second Cys  14  (C171S) did not alter the effect of wild type PRL-3 on adhesion, migration or metastasis of B16F1 mouse melanoma cells (Song et al., 2009). The highly conserved polybasic region adjacent to the CAAX sequence also promotes membrane localization of PRL-1, and likely PRL-2 and -3. Partial or complete substitution of the six basic residues of this motif abolishes the association of PRL-1 with the plasma membrane and interferes with PRL-1-mediated promotion of migration and proliferation of HEK 293 cells (Sun et al., 2007). The polybasic signal may also serve as a nuclear localization signal in the absence of PRL prenylation as mutation of the six basic residues in PRL-1 results in cytoplasmic localization of PRL-1, while deletion or mutation of the PRL-1 CAAX box results in nuclear localization (Sun et al., 2007). The positively charged basic residues in the polybasic motif of PRL-1 can also associate with phosphoinositides (strongest interactions with PI(3)P, PI(4)P, and PI(5)P) and this may target PRL-1 to regions of the membrane enriched in these and other negatively charged phospholipids (Sun et al., 2007). This is a non-enzymatic interaction as PRL1 does not exhibit phospholipid phosphatase activity, unlike the related DSP PTEN. The high degree of conservation of the PRLs, both structurally and in sequence, strongly suggests that PRL-2 and PRL-3 also utilize the polybasic domain to facilitate membrane targeting in the presence of prenylation and nuclear targeting in the absence of prenylation. The in vivo subcellular localization of endogenous PRL-3 is less well defined. PRL-3 is expressed in tumours from a number of different tissues (discussed below in section 1.3.4), where it has been determined to be predominantly cytoplasmic in nature, though the limits of the techniques used may preclude identifying membrane subcompartments within the cytoplasm in which PRL-3 resides. An examination of tissue sections derived from entire 1-week old mice revealed that PRL-3 was only found in differentiated epithelial villus cells of the small intestine,  15  but not in the proliferating crypt cells of the villi in small intestine or in the other tissues examined (Zeng et al., 2000). Further examination of the PRL-3 localization in the villus cells revealed that PRL-3 was present in punctuate cytoplasmic structures and in a perinuclear region (Zeng et al., 2000). It has also been observed that PRL-3 is expressed in both the cytoplasm and nucleus of prostate cells of non-cancerous basal prostate hyperplasia (BPH) and in progressively aggressive prostate carcinoma up to Gleason score 5 (G5) grade (Wong, 2005b).  Further  research is required to determine whether the subcellular distribution of PRL-3 is important for its function in vivo and what regulates this distribution. 1.3.4: PRL-3 expression in human cancers PRL-3 gained widespread interest and attention following a report from the Vogelstein lab that PRL-3 expression was dramatically upregulated in metastases from colorectal carcinomas (CRCs).  Using serial analysis of gene expression (SAGE) library-based gene  expression profiling, 144 transcripts were identified that were present at higher levels in colonic epithelial cells isolated from liver metastases than in non-metastatic colorectal tumour or normal colon epithelium (Saha et al., 2001). Remarkably, further examination of 38 of those transcripts revealed that only PRL-3 was consistently upregulated in all the metastatic samples. PRL-3 expression clearly correlated with CRC progression, as analysis of matched samples from six patients found that PRL-3 expression was undetectable in normal colon epithelia, rose to intermediate levels in advanced primary tumours and increased to significantly higher levels in CRC liver metastasis. This key study raised the possibility that PRL-3 may be an excellent marker and target for metastatic CRC and stimulated further studies into the prevalence of PRL-3 in other cancers and metastases. Several studies have now been conducted and demonstrate roles  16  for PRL-3 in colorectal, breast and gastric carcinomas as well as other cancers, and the results of these studies are described below. Several studies have since confirmed that PRL-3 expression is elevated in 11-45% of primary CRCs using in situ hybridization and/or immunohistochemistry to detect PRL-3 mRNA or protein (Bardelli et al., 2003; Hatate et al., 2008; Kato et al., 2004; Li et al., 2005; Mollevi et al., 2008; Peng et al., 2004; Wang et al., 2007b). PRL-3 is also elevated in CRC metastases, not only in the liver, but also in secondary CRC lesions found in the brain, lung, peritoneum, ovary and lymph nodes (Bardelli et al., 2003; Hatate et al., 2008; Kato et al., 2004; Li et al., 2005; Mollevi et al., 2008; Wang et al., 2007b). The level of PRL-3 expression in primary colorectal tumours has prognostic significance in predicting the development of liver and lung metastasis. A study of 177 primary colorectal tumours showed ~45% scoring positive for high PRL-3 transcript. However, if the 177 cases were divided into those with or without distant liver and lung metastasis, elevated PRL-3 expression was found in a significantly higher proportion of patients with these metastases (liver, 84.4%; lung, 88.9%) than those without (liver, 35.9%; lung, 42.3%) (Kato et al., 2004). In contrast, PRL-3 expression does not correlate with lymph node metastases. Another study, in which PRL-3 protein expression in 88 primary colorectal tumours was examined, found that PRL-3 positivity in primary tumours (~23.9% of the total) significantly correlated with the development of liver (66.7%) but not lymph node (28.6%) metastases (Li et al., 2005). A third independent study reported that PRL-3 was expressed in 16.3% of 49 primary adenocarcinomas (Wang et al., 2007b). Fourteen of the 49 patients had lymph node metastases and all the metastatic samples had higher PRL-3 expression compared to any primary tumour, but only in 2 of those 14 cases did the primary tumour also express PRL-3. This is in accordance with the other findings that PRL-3 expression in primary tumours does not  17  correlate with lymph node metastasis. However, two other studies challenge that result. In a study of CRC with the primary tumours graded from Dukes’ A to D, the number of tumours which expressed PRL-3 and the intensity of that expression increased as the cancer progressed ( PRL-3 staining: Dukes’ A, faint: 23/26, weak: 3/26, strong: 0/26; Dukes’ B, faint: 25/30, weak: 5/30, strong: 0/30; Dukes’ C, faint: 6/30, weak: 6/30, strong: 18/30; Dukes’ D, faint: 0/21, weak: 1/21, strong: 20/21) (Hatate et al., 2008). However, in patients with Dukes’ C primary tumour, similar patterns of strong/weak/faint PRL-3 were observed in patients with metachronous liver metastasis (9/4/3) and in those without (9/2/3), suggesting that upregulated PRL-3 expression does not promote metachronous liver metastasis in Dukes’ C CRC patients. In contrast, a large proportion of synchronous liver metastasis occurs in patients with Dukes’ D CRC who have lymph node metastasis (18/21), and the expression of PRL-3 is more closely associated with lymph node metastasis than synchronous liver metastasis. This suggested that PRL-3 expression is significantly involved in lymph node metastasis but not directly involved in liver metastasis. Another study found that PRL-3 expression in primary CRC correlated with both liver and lymph node metastases (Mollevi et al., 2008). Most patients with high PRL-3 expression in the primary colorectal tumour eventually develop liver metastases (~73.7%) while most patients with low PRL-3 expression in the primary colorectal tumour do not (~78.6%). Similarly, 24 of 38 (~63.2%) patients with high PRL-3 expression in the primary colorectal tumour exhibit vascular or lymphatic invasion compared to 17 of 42 (40.5%) patients with low PRL-3 expression in the primary tumour. Interestingly, the expression of both PRL-3 isoform 1 and 2 was detected in the primary colorectal tumours by quantitative real time RT-PCR, though expression of isoform 1 was elevated to a greater extent than that of isoform 2 (Mollevi et al., 2008). High expression of PRL-3 in primary CRC significantly correlated with venous invasion  18  as well (Kato et al., 2004). In terms of patient outcome, high expression of PRL-3 in CRC was predictive for reduced metastasis-free survival and shorter overall survival after surgical resection of the primary tumour (Kato et al., 2004; Peng et al., 2004). PRL-3 is thus a potentially useful marker to predict CRC aggression and outcome. Another common form of cancer in which PRL-3 expression has been investigated is breast carcinoma.  Several studies using immunohistochemical detection of PRL-3 found  positive or increased PRL-3 protein expression in a subset of tumours. In one study, high PRL-3 expression in 135 ductal carcinomas in situ was determined to occur in 41.5% of samples and in 25.2% and 16.2% of two sets (147 and 99) of samples of invasive breast carcinoma (Radke et al., 2006). A high level of PRL-3 expression was reported in 34.8% of 382 operable primary breast cancers (Wang et al., 2006), while a third study reported that 70.7% of 82 patients with invasive breast cancer showed a high level of PRL-3 expression (Hao et al., 2010). All three studies reported an inverse link between PRL-3 expression and survival and while two studies did not observe a correlation of PRL-3 expression with lymph node involvement (Radke et al., 2006; Wang et al., 2006), the third study reported that 78.7% of 63 patients with lymph node metastasis had high PRL-3 expression in the primary tumour and a statistically significant correlation of PRL-3 expression to lymph node metastasis (Hao et al., 2010). Patients with PRL-3 positive tumours had a significantly reduced disease-free survival (DFS) (Hao et al., 2010; Wang et al., 2006) or exhibited a trend, albeit one lacking statistical significance, towards shorter DFS (Radke et al., 2006). In addition, PRL-3 expression significantly correlated with the more frequent occurrence of distant metastasis in samples from patients with a longer follow-up time (Radke et al., 2006) and inversely correlated with overall survival (Hao et al., 2010). Surprisingly, one study found that PRL-3 expression was significantly correlated with DFS in node-positive  19  but not node-negative patients (Radke et al., 2006), while the second study found a significant association between high PRL-3 expression and reduced DFS survival in node-negative but not in node-positive patients (Wang et al., 2006). The conclusions of the third and most recent study are in agreement with the report by Radke et al. (2006), in that PRL-3 expression and DFS were significantly correlated in node-positive but not node-negative patients (Hao et al., 2010). Further studies are still required to resolve this discrepancy. Several groups have investigated the role of PRL-3 expression in gastric carcinoma. In one study that examined 94 primary gastric carcinomas and 54 matched lymph node metastases, PRL-3 expression was detected in 68% of the primary tumours (Miskad et al., 2004). PRL-3 expression was detected in ~81.5% of primary tumours with matched lymph node metastasis but in only 50% of the primary tumours that did not have nodal metastasis. PRL-3 expression (low and high) was also closely associated with lymphatic invasion, extent of lymph node metastasis and tumour stage. Expansion of this study by the same group (Miskad et al., 2007) demonstrated that high PRL-3 expression was found in 36.2% of the primary gastric tumours, with 55.6% of the node-positive and 10% of the node-negative tumours exhibiting high PRL-3 expression. High PRL-3 expression was also evident in 74.1% of the lymph node metastases in this sample set. In an independent study of 639 gastric carcinoma tumours with 89 matched peritoneal metastases, PRL-3 was found to be expressed in 70.4% of the primary tumours (Li et al., 2007b). PRL-3 was expressed in 80.9% of the cancers with peritoneal metastasis and in 68.7% of those without peritoneal metastasis.  PRL-3 expression was higher in peritoneal metastases than  primary tumours and was associated with tumour stage, lymphatic invasion, extent of lymph node metastasis, and peritoneal metastasis. Patients with PRL-3-positive tumours had poorer survival than those with PRL-3-negative tumours at all stages (I, II, III, and IV) of gastric cancer.  20  Another group examined PRL-3 expression in 137 primary lesions of gastric carcinoma and 107 matched lymph node metastases and found that high PRL-3 expression was more frequently detected in the lymph node metastases than in the matched primary lesion (72.9 vs. 47.7%) (Wang et al., 2008). In this cohort of patients, overall survival was significantly less in patients with high PRL-3 than in those with moderate/low expression. The latest study to examine PRL3 expression in gastric cancer examined 173 primary tumours with 83 lymph node metastases (Ooki et al., 2009). PRL-3 expression was found in 45% of the primary tumours but was found in 80% of the lymph node metastases. PRL-3 expression correlated with lymphatic and vascular invasion, extent of lymph node metastasis and progression of gastric cancer. Interestingly, PRL3 expression negatively correlated with survival in node-negative patients, but not in nodepositive patients after curative resection.  PRL-3 expression in stage I gastric carcinoma  inversely affected patient outcome, and was associated with the presentation of characteristics of stage II disease from a prognostic point of view. Overall, the above findings indicate that PRL-3 expression is enhanced during the progression and metastasis of gastric carcinoma and may prove to be a useful indicator of aggressiveness of this cancer and associated outcome. PRL-3 expression has been examined in several other cancers including liver, ovarian, rectal, esophageal squamous cell, and nasopharyngeal carcinomas, squamous cell carcinoma of the cervix, intrahepatic cholangiocarcinoma and gliomas. In liver cancer, PRL-3 mRNA levels were increased in liver carcinomas compared to normal liver samples, as detected by Northern blot analysis (Wu et al., 2004). No difference was found in PRL-3 transcript expression between primary stage III tumours and matched peritoneal metastases of five patients with ovarian cancer (Polato et al., 2005).  However, PRL-3 transcript expression significantly correlated with  progression of ovarian cancer from stage I to stage III, suggesting that PRL-3 may correlate with  21  invasion rather than metastasis in ovarian cancer. Interestingly, in a study of patients with rectal cancer treated either with or without radiotherapy, PRL expression at the invasive margin was related to distant recurrence and poor survival in the radiotherapy-treated group but not in the untreated patients (Wallin et al., 2006). This suggests that PRL expression was selected for by radiotherapy treatment. However, the antibody used for immunohistochemistry in this study could detect all three PRL homologs, so it is not known which of the homologs is overexpressed in these tumours. The expression of PRL-3 mRNA was examined in 40 cases of esophageal squamous cell carcinoma (ESCC) with 21 cases of matched lymph node metastasis (Liu et al., 2008). PRL-3 expression was significantly higher in ESCC than in normal esophageal tissue and significantly higher in ESCC with lymph node metastasis than in ESCC without metastasis. Interestingly, PRL-1 mRNA expression was also greater in ESCC compared to normal tissue and in ESCC with lymph node metastasis than in ESCC without metastasis. In both cases, expression of PRL1 or -3 correlated with later stage disease. In another study that used immunohistochemistry to examine PRL-3 protein expression in ESCC, 79% of 88 cases of ESCC exhibited PRL-3 expression, and PRL-3 expression was an independent predictor of lymph node metastasis (Ooki et al., 2010). Immunohistochemical analysis was used to examine PRL-3 expression in 174 nasopharyngeal carcinomas (NPC), and 55.7% of the NPC tumours exhibited overexpression of PRL-3 (Zhou et al., 2009). Overexpression of PRL-3 in NPC correlated with distant metastasis, clinical stage of disease and poor survival. Intrahepatic cholangiocarcinoma (ICC) is a cancer of the bile ducts and PRL-3 expression was examined in 102 ICC samples with 62 matched lymph node metastases (Xu et al., 2010). PRL-3 expression was greater in ICC than in normal intrahepatic bile duct tissue, with 47.1% of  22  tumours exhibiting high PRL-3 expression. High PRL-3 expression was more prevalent in the lymph node metastases (80.6%) and correlated with tumour stage, vascular invasion and lymph node metastasis. High PRL-3 expression also correlated with poor overall survival and was an independent prognostic marker of overall survival. PRL-3 expression was absent from normal brain tissue and grade 1 tumours but correlated with late stage disease in glioma, with a greater incidence in grade 3 and 4 tumours than in grade 2 tumours (Kong et al., 2007). PRL-3 was detected by immunohistochemistry in squamous cell carcinoma of the cervix (SCC) (Ma and Li, 2010). PRL-3 expression was found to be higher in SCC than in normal cervical epithelia or in moderate-severe dysplasia and was found to be significantly higher in lymph node metastases than in primary SCC. Cytoplasmic PRL-3 expression was also shown to be correlated with aggressive prostate cancer while no such association was observed with nuclear-localized PRL-3 expression (Wong, 2005b). Surprisingly, PRL-3 transcript expression was downregulated during cancer progression in a limited study of non-small cell lung carcinoma (NSCLC) (Yamashita et al., 2007). PRL-3 expression was strikingly reduced (five- to tenfold) in four of the five sample pairs of lung metastasis and normal lung tissue and was somewhat reduced (~20%) in the primary tissues compared to normal tissue. Thus, it appeared that PRL-3 expression is down-regulated during metastatic progression of non-small cell lung carcinoma. However, this study only examined five patients. A more comprehensive investigation examined the expression of PRL-3 in 94 patients with NSCLC (Ming et al., 2009). High PRL-3 expression was observed in NSCLC and this correlated with advanced clinical stage, distant metastasis, lymph node metastasis and poor post-operative survival. PRL-3 overexpression was also associated with vascular endothelial growth factor (VEGF) and VEGF-C expression, microvessel density and lymph vessel density.  23  These findings support a proangiogenic (or vessel-formation) role for PRL-3. These results are more in line with those from investigations of PRL-3 expression in other cancers, and may more accurately reveal the nature of the role of PRL-3 in NSCLC than did the first study, which may be an anomaly due to the small sample size. All together, these investigations reveal that PRL-3 can promote advanced stage disease and/or metastasis in cancer. Furthermore, PRL-3 expression correlates with poor overall survival in a number of examined cancers. PRL-3 may prove to be a valuable prognostic marker for disease progression and outcome and would be worthwhile investigating as a therapeutic target. 1.3.5: PRL-3 expression imparts metastasis-associated properties to cell lines In addition to its association with aggressive cancer and metastasis, overexpression of PRL-3 in many cell lines can impart metastasis-associated properties to the cells, while ablation of PRL-3 message can reduce these cellular.  The properties affected by PRL-3 include  transformation, migration, invasion, tumourigenesis and metastasis. Some examples of these studies are detail below, and the findings of these and additional investigations are summarized in Table 1. The first study to demonstrate that PRL-3 overexpression promoted the tumorigenic properties and metastatic ability of cells was conducted by Zeng et al. (2003) using a Chinese hamster ovary (CHO) cell line. Stable expression of Myc-tagged PRL-3 enhanced cell migration in wound healing and Transwell assays and increased invasion into Matrigel, compared to β-galtransfected control cells. Myc-PRL-3-expressing cells injected into the tail vein of mice induced the formation of lung and some liver metastases, whereas injection with control cells did not result in metastatic lesions. Interestingly, PRL-1 overexpression in CHO cells also enhanced migration and invasion and tail vein injection of PRL-1-expressing cells gave rise to metastatic  24  lesions at the same frequency as the PRL-3-expressing cells (Zeng et al., 2003b).  The  phosphatase activity of PRL-3 was required to bring about these changes in cell behavior. CHO cells expressing the catalytically inactive mutant of EGFP-PRL-3 (C104S) exhibited reduced motility compared to wild-type EGFP-PRL-3-expressing cells (Zeng et al., 2003b). Also, the wild-type PRL-3-expressing cells formed more colonies in soft agar and larger xenograft tumours in mice compared to cells expressing the inactive PRL-3 mutant, and gave rise to lung metastases from the xenograft tumours, unlike the PRL-3 (C104S) cells (Guo et al., 2004). These findings indicate that the PRL-3 up-regulation seen in human tumours likely plays an early causal role in tumour progression and metastasis rather than simply being a consequence of these processes, and that PRL-3 enzymatic activity is required for this behavior. Different levels of PRL-3 expression were found in two related melanoma cell lines, with an approximately threefold higher PRL-3 transcript level in the highly metastatic B16-BL6 cell sub-line compared to the parental B16 cell line (Wu et al., 2004). Stable transfection of B16 cells with PRL-3 led to increased rates of migration and invasion to levels comparable to those of the B16-BL6 cells. PRL-3-expressing B16 cells also displayed enhanced cell adhesion to and spreading on fibronectin and adhesion to laminin, while adhesion to collagen was decreased, compared to parental B16 control cells. Cell proliferation rates were increased in B16-PRL-3 cells which correlated with increased in vivo tumour growth and the formation of more metastases in the liver and lung in in vivo metastasis assays. PRL-3 catalytic activity was required for its ability to promote migration. Cell motility was abolished by general PTP inhibitors, and was not observed in B16 cells expressing the catalytically defective PRL-3 (D72A) or the inactive PRL-3 (C104S) mutants. Ablation of PRL-3 expression in the transfected B16-PRL-3 cells eliminated the increased migration, adhesion and proliferation of these cells  25  Table 1.1: Cancer related processes affected by altered PRL-3 expression.  Cell process Proliferation  Overexpression ↑  No effect  ↓  Colony formation in soft agar  ↑  ↓  Migration  ↑  ↓  Migration  ↓  Knockdown  B16 (Wu et al., 2004) HEK 293 (Liang et al., 2007; Matter et al., 2001) TE5 (Ooki et al., 2010) SW480 (Semba et al., 2010) DLD-1 (Rouleau et al., 2006) B16-BL6 (Qian et al., 2007) DLD-1 (Kato et al., 2004) SGC7901 (Li et al., 2006) HCT116 (Polato et al., 2005) INA-6 (Fagerli et al., 2008) 5-8F (Zhou et al., 2009) HONE1 (Zhou et al., 2009) A2780 (Polato et al., 2005) IGROV-1 (Polato et al., 2005) SKOV-1 (Polato et al., 2005) SGC7901 (Wang et al., 2008) A549 (Ming et al., 2009) SH101-P4 (Matsukawa et al., 2010) TE8 (Ooki et al., 2010) TE10 (Ooki et al., 2010) TE11 (Ooki et al., 2010) TE14 (Ooki et al., 2010) SW480 (Semba et al., 2010) HEK 293 (Liang et al., 2007; Matter et al., 2001) CHO (Zeng et al., 2003b) TE5 TE8 (Ooki et al., 2010) TE10 (Ooki et al., 2010) TE11 (Ooki et al., 2010) TE14 (Ooki et al., 2010) HEK 293 (Liang et al., 2007; Matter et al., 2001) B16 (Wu et al., 2004) CHO (Zeng et al., 2003b) SW480 (Fiordalisi et al., 2006) LoVo (Peng et al., 2009) DLD-1 (Wang et al., 2007a) B16-BL6 (Qian et al., 2007) DLD-1 (Kato et al., 2004) SGC7901 (Li et al., 2006) (Wang et al., 2008) INA-6 (Fagerli et al., 2008) 5-8F (Zhou et al., 2009) HONE1 (Zhou et al., 2009)  26  Cell process  Overexpression  Knockdown A549 (Ming et al., 2009)  Invasion  ↑  CHO (Zeng et al., 2003b) SW480 (Fiordalisi et al., 2006) B16 (Wu et al., 2004) DLD-1 (Rouleau et al., 2006) LoVo (Peng et al., 2009) TE5 (Ooki et al., 2010)  ↓  Tumor formation  ↑  Metastasis  ↓ ↑ ↓  B16-BL6 (Qian et al., 2007) SGC7901 (Li et al., 2006) (Wang et al., 2008) MCF-7 (Rouleau et al., 2006) 5-8F (Zhou et al., 2009) HONE1 (Zhou et al., 2009) A549 (Ming et al., 2009) SH101-P4 (Matsukawa et al., 2010) TE8 (Ooki et al., 2010) TE10 (Ooki et al., 2010) TE11 (Ooki et al., 2010) TE14 (Ooki et al., 2010) CHO (Zeng et al., 2003b) B16 (Wu et al., 2004) B16-BL6 (Qian et al., 2007) CHO (Zeng et al., 2003b) B16 (Wu et al., 2004) LoVo (Peng et al., 2009) B16-BL6 (Qian et al., 2007) DLD-1 (Kato et al., 2004) SGC7901 (Li et al., 2006) SH101-P4 (Matsukawa et al., 2010)  The overexpression or ablation (knockdown) of PRL expression in various cell lines enhances (↑), has no effect, or has a negative effect (↓) on a number of cellular properties. (A549, lung carcinoma cells; A2780, IGROV-1, and SKOV-1, ovarian carcinoma cells; B16, mouse melanoma cells; B16-BL6, a more metastatic subline of B16; CHO, Chinese hamster ovary cells; DLD-1, LoVo, and SW480, colorectal adenocarcinoma cells, HCT116, colorectal carcinoma cells; HEK 293, human embryonic kidney cells; MCF-7, mammary adenocarcinoma cells;, SGC7901 and SH101-P4, gastric adenocarcinoma cells; INA-6, myeloma cells;, 5-8F and HONE1, nasopharyngeal carcinoma cells; TE5, TE8, TE10, TE11, and TE14, esophageal squamous cell carcinoma cells)  27  associated with PRL-3 expression.  In a subsequent study, siRNA-mediated reduction of  endogenous PRL-3 in the metastatic B16-BL6 melanoma cells decreased cell migration and invasion, inhibited tumour formation, prevented lymph node metastasis and significantly prolonged the survival of mice after injection of the cells (Qian et al., 2007). This raises the possibility that specific targeting of PRL-3 may prove to be an effective treatment of PRL-3expressing tumours. It is interesting that so many different types of carcinoma cell lines are sensitive to alterations in the levels of PRL-3 expression. In particular, the loss or reduction of cancerassociated properties in cells treated to knockdown PRL-3 indicates that endogenous PRL-3 plays a conserved and critical role in these events. Most of the gains in or losses of the cancerassociated cell behavior in the different cell types correlated with the respective gain or loss of PRL-3 expression. The notable exception is that proliferative rates were not always affected by changes in PRL-3 expression in some cell lines, even though other endpoint changes were occurring in the same cells (Fagerli et al., 2008; Kato et al., 2004; Li et al., 2006; Ming et al., 2009; Qian et al., 2007; Rouleau et al., 2006; Zhou et al., 2009). This suggests that PRL-3 may not be an essential regulator of cell cycle progression (Qian et al., 2007) and that the ability of PRL-3 to promote tumour formation and metastasis may be due to other effects. These are most likely related to endpoints such as migration and invasion that can affect the interaction of the cancer cells with their local environment. In conjunction with the elevated levels of PRL-3 expression that is correlated with tumour progression and metastasis, the results of the experiments discussed above strongly indicate that PRL-3 plays a key functional role in the processes of tumourigenesis and metastasis. These experiments are also highly encouraging as to the suitability of PRL-3 as a target for the development of anti-cancer therapeutics.  28  1.3.6: PRL-3 and angiogenesis For aggressive cancers to form distant metastases the tumours cells requires the ability to promote a pro-angiogenic switch and recruit new vasculature. Several lines of evidence suggest that PRL-3 has a role in tumour-related angiogenesis. A comparison of gene expression profiles of normal and malignant colorectal endothelial cells detected increased numbers of ESTs for PRL-3 in the tumour vasculature (St Croix et al., 2000). In addition, a proportion of the PRL-3 expression detected in colorectal cancer metastasis is localized to tumour vasculature, in both endothelial cells and the smooth muscle cells surrounding larger vessels (Bardelli et al., 2003). When injected into the tail vein of nude mice, PRL-3-expressing CHO cells form micro- and macro-metastatic solid tumors in the lung that are tightly surrounded by vasculature with some of these tumors forming within established blood vessels (Guo et al., 2004). Elevated PRL-3 expression is detected in cases of colorectal cancer with intensive venous infiltration and distant metastasis compared to cases without venous invasion with the cancer cells forming the intravenous tumour emboli showing strikingly increased expression (Kato et al., 2004). PRL-3 expression is increased ~10-fold in breast tumour vasculature compared to normal vasculature and leads in an enhancement of cellular migration and tube formation in human microvascular endothelial cells (HMVEC) (Parker et al., 2004). Human umbilical vascular endothelial cells (HUVEC) or HMVEC treated with phorbal esters mimic the malignant phenotype of increased invasion, tube formation and growth factor-induced proliferation and show increased levels of PRL-3 expression (Rouleau et al., 2006).  Also, HMVEC cells infected with adenovirus  containing PRL-3 undergo tube formation.  CHO and DLD-1 cells expressing PRL-3 recruit  HUVECs to embed within the PRL-3-expressing cells in a co-culture system and PRL-3expressing DLD-1 cells promote vascular formation by HUVECs in vitro (Guo et al., 2006). This  29  proangiogenic effect of PRL-3 is likely mediated in part by downregulation of the antiangiogenic cytokine IL-4 (Guo et al., 2006)]. Interestingly, PRL-3 may be important in normal blood vessel formation as it is expressed in fetal heart, in developing blood vessels in normal colonic tissue and colorectal cancer tissue, and in developing red blood cells but not in mature blood vessels or red blood cells (Guo et al., 2006).  PRL-3 expression was significantly  associated with increased microvessel density in hepatocellular carcinoma (Zhao et al., 2008). PRL-3 expression in NSCLC correlated with upregulated VEGF, the key growth factor that induces angiogenesis, and VEGF-C, which stimulates lymphangiogenesis, and increased microvessel density and lymph vessel density. As well, blocking endogenous PRL-3 expression in A549 lung carcinoma cells reduced expression of VEGF and VEGF-C (Ming et al., 2009). The up-regulation of PRL-3 expression in tumor vasculature (Bardelli et al., 2003; St Croix et al., 2000), the development of highly vascularized tumors and metastases from PRL-3 expressing cells in nude mice (Guo et al., 2004; Guo et al., 2006; Kato et al., 2004), the ability of overexpressed PRL-3 to induce tube formation in HMVEC cells (Rouleau et al., 2006) and recruit endothelial cells (Guo et al., 2006), correlation of PRL-3 expression and increased vasculature in hepatocellular carcinoma and NSCLC (Ming et al., 2009; Zhao et al., 2008), and the finding that reducing endogenous PRL-3 in A549 cells results in reduced VEGF and VEGFC expreesion are exciting discoveries that indicate that elevated PRL-3 expression may promote angiogenesis as a facet of tumor establishment and progression. 1.3.7: The signaling pathways of PRL-3 The studies involving inactive mutant forms of PRL-3 have provided clear evidence that many of the cellular effects of PRL-3 require catalytic activity. The substrate(s) of PRL-3 has remained a key unanswered problem in understanding its mechanism of action in cell signaling.  30  Recently, several groups have uncovered links between PRL-3 and the molecular mediators of cell-cell and cell-matrix interactions (Table 2), consistent with the effects of up- or downregulating PRL-3 expression on cell migration and invasion. I discuss several of the reports below. PRL-3 can bind to integrin α1 and inducible expression of PRL-3 in HEK 293 cells reduced levels of integrin β1 tyrosine phosphorylation (Peng et al., 2006). This suggests that the interaction with integrin α1 positions PRL-3 to catalyze dephosphorylation of integrin β1, though no evidence was presented to show that integrin β1 is a direct substrate of PRL-3. PRL-3 expression in these cells also increased Erk1/2 phosphorylation, suggesting that PRL-3 increases Erk activity. Subsequent work by the same investigators demonstrated that PRL-3 can also bind integrin β1 in vivo in LoVo colon carcinoma cells (Peng et al., 2009). PRL-3 expression promoted migration and invasion of the LoVo cells and induced metastasis formation after injection of PRL-3-expressing cells into the tail veins of mice. PRL-3 expression also increased the activity of Erk1/2 and the expression and the activity of matrix metalloproteinase- (MMP-) 2, an important component of cell invasion which is required to digest the proteins of the extracellular matrix (Chang and Werb, 2001; John and Tuszynski, 2001). Interestingly, integrin β1 knockdown blocks the effects of PRL-3 on migration, invasion, metastasis and Erk1/2 and MMP-2 activity.  Furthermore, PRL-3-mediated migration and invasion are blocked by  inhibiting Erk1/2 or MMP-2 activity. These results provide evidence that PRL-3 acts in a signaling pathway in which dephosphorylation of integrin β1 leads to increased Erk1/2 and MMP-2 activity to promote migration, invasion and metastasis. PRL-3 overexpression in DLD-1 colon carcinoma cells results in enhanced phosphorylation and activation of the serine/threonine protein kinase Akt, and the Akt substrate  31  GSK-3β is phosphorylated and inactivated (Wang et al., 2007a). Both of these events requires the activity of phosphotidylinositol-3 kinase (PI3K), an upstream activator of Akt, as treatment of the cells with a PI3K inhibitor abolishes the enhanced Akt activation. PRL-3 appears to enhance Akt activity by downregulating phosphatase and tensin homolog (PTEN) protein expression (Wang et al., 2007a). PTEN, a well characterized tumour suppressor, is a DSP and lipid phosphatase that can dephosphorylate phosphatidylinositol at the 3-position of the inositol ring. A key substrate for PTEN is PI(3,4,5)P3, and PTEN opposes PI3K activity through dephosphorylation of this substrate.  Thus downregulation of PTEN can promote PI3K-  dependent endpoints, such as Akt activity. The PI3K/Akt pathway is an important signaling pathway in cell survival, and is frequently activated in human cancers (Altomare and Testa, 2005; Dillon et al., 2007; Nicholson and Anderson, 2002). PI3K/Akt signaling is also linked to epithelial-mesenchymal transitions (EMT), a process by which epithelial cells are converted to mesenchymal cells and is important in embryonic development and tumour progression, among other things (Guarino et al., 2007; Thiery and Sleeman, 2006).  EMT is characterized by several cellular and molecular changes and is  frequently associated with cell motility and invasiveness. The downregulation of the epithelial marker protein E-cadherin (E-cad), a critical component of the adherens junction (AJ) complex that mediates epithelial cell-cell attachment, is a prominent feature of EMT. E-cad expression was reduced in PRL-3-expressing DLD-1 cells, as was expression of two other epithelial marker proteins, γ-catenin, and integrin β3 (Wang et al., 2007a). The expression of the mesenchymal marker proteins Snail and fibronectin was increased in the DLD-1 cells overexpressing PRL-3. The alterations in the levels of the EMT marker proteins indicate that PRL-3 induces EMT in DLD-1 cells. There are also several points of convergence among the proteins that PRL-3 was  32  reported to regulate in DLD-1 cells. Snail is a well-known transcriptional repressor of E-cad and activity of GSK-3β that targets Snail for degradation, thereby reducing its protein expression (Barbera et al., 2004; Guarino et al., 2007; Julien et al., 2007; Larue and Bellacosa, 2005; Zhou et al., 2004). Furthermore, E-cad and PTEN protein expression undergo coincident increases and decreases, and siRNA-mediated downregulation of E-cad reduces PTEN protein level (Li et al., 2007a). However, as PTEN is reduced in PRL-3-expressing DLD-1 cells in a PI3K-independent manner, it is unlikely to be the upstream initiator of a potential E-cad-linked reduction of PTEN. Other signaling molecules that regulate Snail, E-cad and/or PTEN protein levels could be targeted by PRL-3. Other molecular consequences of PRL-3 expression were reported by Wang et al. (2007). For example, the levels of paxillin and vinculin are reduced in PRL-3-expressing HeLa and CHO cells (Wang et al., 2007a). Both of these proteins are component of focal adhesions, structures composed of multi-protein complexes that mediate the attachment of cells to the extracellular matrix. The formation and disruption of focal adhesions are required for cell movement. Indeed, focal adhesions are reduced in PRL-3-expressing cells (Wang et al., 2007a). Vinculin can interact with AJ complexes via association with α-catenin as well as reside in focal adhesions (Watabe-Uchida et al., 1998). PTEN also interacts with AJ complexes by association with the membrane-associated guanylate kinase (MAGUK) protein MAGI-2, a protein that binds βcatenin (Kawajiri et al., 2000). Interestingly, vinculin expression can regulate the interaction of β-catenin with MAGI-2, and this in turn regulates PTEN expression (Subauste et al., 2005) and loss of vinculin downregulates PTEN. Vinculin regulates PTEN at the post-transcriptional level by affecting its proteosome-mediated proteolytic degradation (Subauste et al., 2005). PTEN levels are also reduced in cells incubated with an antibody that blocks E-cad-mediated AJ  33  complex formation. Together these results indicate that AJ integrity is important for PTEN protein stability, and that PRL-3-induced downregulation of PTEN may reflect a PRL-3mediated disruption of AJ through an as yet unidentified mechanism. The stable expression of PRL-3 in HEK 293 cells induces phenotypic changes in the cell characteristic of epithelial-mesenchymal transitions (EMT) and in the activation of the tyrosine kinase Src (Liang et al., 2007). This led to increased tyrosine phosphorylation of several cellular proteins, including the activation of some direct substrates of Src (Stat3 and p130Cas) and downstream targets of Src (Erk1/2). Src plays key roles in multiple receptor-mediated pathways and has been implicated in EMT and tumourigenesis in breast and colorectal cancers. Src activity is regulated through tyrosine phosphorylation of the C-terminal residue Tyr529 which acts through inhibitory intramolecular interactions when phosphorylated. Phosphorylation of Src at Tyr529 is mediated by the C-terminal Src kinase (Csk). PRL-3 reduced levels of phospho-Src Tyr529 and this was found to be due to PRL-3-mediated downregulation of Csk.  This  downregulation of Csk was subsequently found to be due a translational control mechanism of PRL-3 activity (Liang et al., 2008). The PRL-3-mediated inhibition of Csk protein synthesis was due to increased phosphorylation of elongation initiation factor 2 (eIF2) in PRL-3-expressing cells. The mode of PRL-3-induced Src activation is a strikingly similar to the PRL-3-mediated downregulation of PTEN and activation of Akt (Wang et al., 2007a). The phosphorylation of the important scaffold molecule p130Cas was increased in the PRL-3-expressing HEK cells (Liang et al., 2007).  p130Cas is a substrate of Src, and its  phosphorylation stimulates relocalization to newly formed focal adhesion.  p130Cas is a  mediator of cell migration that associates with Crk and DOCK180 at sites of adhesion to promote cell motility (Defilippi et al., 2006). The association of p130Cas with the focal adhesion  34  protein vinculin was also increased by PRL-3 expression, suggesting that PRL-3 promotes the localization of p130Cas to focal adhesions (Liang et al., 2007). Fak tyrosine phosphorylation and association with vinculin was unaltered by PRL-3 expression. In contrast to the findings of Wang et al. (2007), vinculin expression was not altered by PRL-3 expression in HEK 293 cells. The overexpression of Csk in PRL-3-expressing cells blocked p130Cas phosphorylation and association with vinculin, as well as eliminating the enhanced proliferation and migration, indicating that Src activity was responsible for the PRL-3-mediated alterations.  PRL-3  overexpression in HEK 293 cells appears to promote certain Src-mediated focal adhesion protein modifications that promote cell spreading and migration. This is in contrast to the effect of PRL3 expression in reducing focal adhesions and focal adhesion proteins in HeLa and CHO cells (Wang et al., 2007a). The contrast in the behavior of PRL-3 from the two reports could be due to cell types and conditions and likely reflects different aspects of the ability of PRL-3 to promote focal adhesion turnover, as both enhanced focal adhesion formation and disassembly are important processes in promoting cell migration. The members of the Rho family of GTPases, Rho, Rac, and Cdc42, are critical regulators of the actin cytoskeleton in cell migration. They regulate actin polymerization and stress fiber assembly, focal adhesion formation, and motile cell structures such as lamellipodia and filopodia. The overexpression of PRL-3 in the colon carcinoma cell lines SW480 and DLD-1, and in CHO cells, results in altered Rho GTPase activity (Fiordalisi et al., 2006; Wang et al., 2007a). In SW480 cells, PRL-3 expression promotes a four- to fivefold in increase in RhoA and RhoC activity and a 70% decrease in Rac activity, but no change in Cdc42 activity (Fiordalisi et al., 2006). This activity of PRL-3 was upstream of Rho and the Rho upstream activating kinase ROCK, as an inhibitor of ROCK blocked PRL-3-mediated alterations in Rho activity. PRL-3  35  phosphatase was required to promote Rho activation and SW480 cell invasion, as the inactive mutants PRL-3 (C104S) and (D72A) could not promote these effects. Interestingly, the mutants could promote cell motility and activation of the serum response element (SRE) transcription factor, though to a lesser extent than wild-type PRL-3. Wang et al. (2007) also observed effects of PRL-3 expression on RHO GTPase activity, although opposite of those reported by Fiordilisi et al. (2006). In CHO and DLD-1 cells, PRL-3 expression reduced F-actin levels and caused a reduction in RhoA and Rac1 activities. As discussed above, the distinct findings of the two groups may reflect that it is the process of enhanced focal adhesion turnover that promotes cell migration and invasion. PRL-3 has been shown to interact with several other proteins involved in EMT and/or invasion. The cytoskeletal intermediate filament protein keratin 8 (or cytokeratin 8, Krt8), an epithelial marker protein, is a direct target of PRL-3, and has been shown to associate with PRL3 in vivo in SW480 cells (Mizuuchi et al., 2009).  This interaction resulted in reduced  phosphorylation of Ser73 and Ser431 of Krt8. PRL-3 and Krt8 also co-localized to lamellipodia and membrane ruffles. Another protein shown to bind to PRL-3 in vivo is Ezrin, a member of the ERM family of proteins which link the actin cytoskeleton to the plasma membrane (Forte et al., 2008). Overexpression of PRL-3 in HCT116 cells resulted in reduced Tyr phosphorylation and Thr567 phosphorylation, however, ablation of PRL-3 in HCT116 cells only resulted in enhanced Thr567 phosphorylation. Ezrin is activated through phosphorylation of its Thr567 by ROCK (Matsui et al., 1998), and promotes cell proliferation, migration and invasion (Bretscher et al., 1997; Wick et al., 2001). The regulation of Ezrin phosphorylation is another potential site of action in which PRL-3 may mediate cell migration and invasion. Another potential EMT protein that PRL-3 can interact with is Cadherin-22 (CDH22) (Liu et al., 2009). PRL-3 can bind  36  directly to CDH22 in SW480 and SW620 colon carcinoma cells and PRL-3 expression results in reduced CDH22 expression in SW480 cells. PRL-3 expression in these cells also results in changes in the expression and activity of other AJ proteins. β-catenin expression is upregulated in PRL-3-expressing SW480 cells and accumulates in the nucleus, where β-cat may interact with the transcription co-factor TCF in order to mediate Wnt pathway actions. The knockdown of endogenous PRL-3 in SW480 reduces α-catenin but increases γ-catenin expression. The AJ proteins are mediators of cell-cell contact and include E-cad or other cadherin family members as the central component (see description in Section 1.5 below). Together, these studies support the findings that PRL-3 is involved in mediating EMT and can target the process through a number of different signalling molecules/pathways, some of which overlap in function. The expression of the microtubule dynamics regulatory protein stathmin was found to be reduced in SW480 cells after PRL-3 expression was knocked down for 24 or 48 hours (Zheng et al., 2010). PRL-3 binds stathmin in vivo in SW480 and LoVo cells, as well as in primary CRC tumours with or without corresponding metastases.  Partial co-localization of PRL-3 and  stathmin in SW480, LoVo, HT29 and SW620 colon carcinoma cell lines was also observed. Both PRL-3 and stathmin overexpression in SW480 cells resulted in reduced acelylated tubulin, indicating that both proteins play a role in destabilizing microtubules. Interestingly, stathmin promoted cell migration, adhesion to fibronectin and colony formation in SW480, LoVo, HT29, and SW620 cells, similar to some of the phenotypes that PRL-3 has been shown to display in these and other cell types. PRL-3 has also been shown to bind in vivo to the nucleolar specific protein, nucleolin, in SW480 cells (Semba et al., 2010).  PRL-3 expression reduced  phosphorylated nucleolin on Thr76 and Thr84, and was required for appropriate nucleolar assembly of nucleolin.  37  Together, these studies reveal that PRL-3 may interact with a number of different proteins and control a number of different signaling pathways.  The identification of several  distinct binding partners/substrates (integrin α1, integrin β1, Ezrin, stathmin, and nucleolin) for PRL-3 corresponds with the ability of DSPs to bind several different substrates due to the shallowness of the catalytic pocket. Indeed, as mentioned above, PRL-3 has an even shallower catalytic pocket than most DSPs, and its ability to bind different members from different families of proteins suggest that PRL-3 exhibits promiscuous substrate binding ability.  It will be  Table 1.2: Molecular effects of altered PRL-3 expression.  Signaling Protein Target pathway/ Process Integrin/Src Integrin α1 signalling Integrin β1 Integrin β3 Src Csk Fak p130Cas  paxillin vinculin  Fibronectin Erk1/2  Stat3  Effect  Reference  Binds in vivo in HEK 293 Tyrosine phosphorylation reduced in HEK 293, LoVo Binds in vivo in LoVo Reduced in DLD-1 Increased Src activity in HEK 293 Reduced in HEK 293 No effect in HEK 293 Enhanced tyrosine phosphorylation in HEK 293 Reduced tyrosine phosphorylation in Ang-II-stimulated HEK 293 Reduced expression and reduced Tyr-31 phosphorylation in HeLa Decreased localization to focal adhesions in HeLa, CHO; enhanced binding to p130Cas in HEK 293 Upregulated in DLD-1 Increased activity in HEK 293  (Peng et al., 2006) (Peng et al., 2006; Peng et al., 2009) (Peng et al., 2009) (Wang et al., 2007a) (Liang et al., 2007) (Liang et al., 2007) (Liang et al., 2007) (Liang et al., 2007) (Matter et al., 2001) (Liang et al., 2007) (Liang et al., 2007)  (Liang et al., 2007; Peng et al., 2006) Reduced activity in A549 after PRL-3 (Ming et al., 2009) ablation Increased activity in HEK 293 (Liang et al., 2007)  38  Signaling pathway/ Process PI3K/Akt Signaling  Protein Target  Effect  PTEN Akt  Reduced in DLD-1 Increased activity in HEK 293 Transient increased activity in HT1080 Decreased activity in DLD-1, SW480  GSK-3β EMT  E-cad Cadherin-22 β-catenin α-catenin γ-catenin  EMT  Vimentin Snail  Keratin 8  Rho family GTPases  Rho  Rac  Cdc42  Reference  (Wang et al., 2007a) (Wang et al., 2007a) (Basak et al., 2008) (Liu et al., 2009; Wang et al., 2007a) Reduced in DLD-1, SW480 (Liu et al., 2009; Wang et al., 2007a) Binds in vivo in SW480, SW620 (Liu et al., 2009) Reduced in SW480 (Liu et al., 2009) Increased expression and nuclear (Liu et al., 2009) accumulation in SW480 Reduced in SW480 after PRL-3 (Liu et al., 2009) knockdown Increased in SW480 after PRL-3 (Liu et al., 2009) knockdown Reduced in DLD-1 (Liu et al., 2009) Upregulated in SW480 (Liu et al., 2009) Upregulated in DLD-1 (Wang et al., 2007a) Reduced in SW480 after PRL-3 (Liu et al., 2009) knockdown Binds in vivo in SW480, LoVo, CRC (Mizuuchi et al., tumours; reduced expression after PRL-3 2009) Increased activity in SW480, reduced in (Fiordalisi et CHO 2006; Wang et 2007b) Decreased Rho-A and –C expression in (Ming et al., 2009) A549 after blocking PRL-3 with antiPRL-3 antibodies Reduced activity in SW480, CHO (Fiordalisi et 2006; Wang et 2007b) Activity unaffected in SW480 (Fiordalisi et 2006)  al., al.,  al., al., al.,  39  Signaling pathway/ Process Cell cycle  Protein Target  Effect  Reference  p53  Upregulated in MEF Reduced expression and activity in HCT116; expression upregulated in HeLa after PRL-3 knockdown Upregulated in HT1080; upregulated in MEFs after PRL-3 knockdown Reduced p21 activity in HCT116 Upregulated in HT1080 Upregulated in HCT116 Increased phosphorylation in HCT116 Binds in vivo in HCT116; reduces phosphorylated Tyr and phospho-Thr567 Bind in vivo in SW480; reduces phospho-Ser73 and phospho-Ser431 knockdown in SW480 Binds in vivo in SW480; reduced phospho-nucleolin Thr76/Thr84; reduced nucleolar assembly of nucleolin Reduces phosphorylation in THP-1 Reduces expression in THP-1 cells  (Basak et al., 2008) (Min et al., 2010)  p21  Other  p27 PIRH2 MDM2 Ezrin  Stathmin Nucleolin  p38 LITAF  (Basak et al., 2008) (Min et al., 2010) (Basak et al., 2008) (Min et al., 2010) (Min et al., 2010) (Forte et al., 2008)  (Zheng et al., 2010) (Semba et al., 2010)  (Tang et al., 2009) (Tang et al., 2009)  The molecular effects of PRL overexpression or ablation were investigated in various cultured cell lines (A549, lung carcinoma cells; CHO, Chinese hamster ovary cells; DLD1, SW480, SW620, LoVo and HCT116 colorectal adenocarcinoma cells, HEK 293, human embryonic kidney cells; HeLa, cervical adenocarcinoma cells; HT1080, fibrosarcoma cells; MEF, mouse embryonic fibroblasts; THP-1, macrophage cells).  interesting to determine the regions of these PRL-3 substrates that PRL-3 binds and to generate a consensus binding sequence(s) for PRL-3. This would also facilitate generating PRL-3-specific therapeutics to treat PRL-3-expressing cancers. 1.3.8: Physiological and therapeutic regulation of PRL-3 As increases in PRL-3 expression correlates with tumour aggressiveness and can impart tumourigenic or metastatic properties upon cell lines, knowledge of the process that regulate PRL-3 expression is critical to understanding, and perhaps controlling, the molecular processes  40  underlying the pathogenesis and spread of cancer. Several reports have emerged providing evidence of physiological regulation or therapeutic regulation of PRL-3 expression. Although gene amplification is a common mechanism for upregulating protein expression in cancer, little gene amplification of PRL-3 is observed in PRL-3-expressing cancers. For example, PRL-3 gene amplification was not detected in late stage ovarian tumours that overexpressed PRL-3 (Polato et al., 2005). Also, gene amplification is only present in a limited subset of PRL-3-expressing primary CRCs or their metastases (Bardelli et al., 2003; Saha et al., 2001), even though a high frequency (67-100%) of CRC liver metastases have upregulated PRL3 (Bardelli et al., 2003; Kato et al., 2004; Peng et al., 2004; Saha et al., 2001). These findings suggest that PRL-3 is likely to be upregulated in human cancers mainly through transcriptional or post-transcriptional mechanisms. PRL-3 expression has been shown to be regulated through a number of means. Phorbol 12-myristate 13-acetate (PMA) exposure dramatically increases PRL-3 expression in human endothelial cells (Rouleau et al., 2006). PRL-3 is downregulated following denervation of rat skeletal muscle (Magnusson et al., 2005). More recently, PRL-3 has been shown to be a direct target of p53 transcriptional activity and is upregulated in response to DNA-damaging events in a p53-mediated process (Basak et al., 2008). Translation of PRL-3 is regulated by the poly(C)binding protein 1 (PCBP1), which is a member of the hnRNP family of RNA- and/or DNAbinding proteins (Makeyev and Liebhaber, 2002).  PCBP1 is known to participate in the  regulation of RNA transcription, pre-mRNA processing, maturation, and mRNA export. PCBP1 has been reported to play crucial roles in a broad-spectrum of transcriptional and translational events (Huo and Zhong, 2008). It is likely that there are several more transcriptional regulators  41  of PRL-3 expression as well, as consensus sequences of a number of transcription factors is found in the 5’-UTR of the PRL-3 promoter (Wong, 2005a). Due to the association of PRL-3 with cancer metastasis, PRL-3 is potentially an attractive target for therapeutic regulation in disease. Several groups have targeted PRL-3 activity to attempt to block PRL-3 effects in cancer cell lines (Ahn et al., 2006; Choi et al., 2006; Dursina et al., 2006; Pathak et al., 2002; Wang et al., 2009a). Pentamidine, an anti-protozoan drug, was the first compound tested (Pathak et al., 2002). It has a selective inhibitory effect on several PTPs, including PTP1B, MAPK phosphatase (MKP)-1 and the PRLs.  Pentamidine (10 μg/ml)  completely inhibits recombinant PRL-1, -2 and -3 activity in vitro through an irreversible mechanism, and the growth of several cancer cell lines is arrested by pentamidine. Though other inhibition of PTPs may also be involved, further testing and development may indicate that pentamidine to be a useful therapeutic of cancers expressing PRLs, especially as it has already been used clinically as a treatment for leishmaniasis, Gambian Trypanosomiasis and pneumonia caused by Pneumocystis carinii (Sands et al., 1985).  High-throughput screening revealed  rhodanine to be a PRL-3 inhibitor (Ahn et al., 2006). Two rhodanine derivatives, one with a 2bromobenzyl substituent and the other with a 2-chloro-6-fluorobenzyl substituent, exhibited ~30to 60-fold stronger inhibition of PRL-3 than pentamidine. There were also more effective at inhibiting the invasive ability of B16F10 mouse melanoma cells, though the toxicity of these compounds and specificity for PRL-3 has not been tested. The first natural inhibitors of PRL-3 activity were found in a screen of a methanol extract of young branches of Taxus cuspidate, which is the genus of the source of the anti-cancer drug paclitaxel (Wani et al., 1971). The biflavinoids ginkgetin and sciadopitysin, compounds with a number of known biological activities, were the PRL-3 inhibitors in the plant extract. They exhibited in vitro IC50 values  42  similar to pentamidine, but specificity to PRLs and among other PTPs is unknown. Another natural inhibitor of PRL-3 is curcumin (Wang et al., 2009a), a polyphenol derived from dietary spice turmeric that possesses wide-ranging anti-inflammatory and anticancer properties (Sharma et al., 2005). Curcumin drasticially downregulated PRL-3 expression from a number of cell lines and reduced a number of PRL-3-mediated phenotypic effects in B16BL6 mouse melanoma cells, such as migration, proliferation and adhesion to fibronectin. Curcumin treatment significantly inhibited tumour growth and spontaneous metastasis after injection of B16BL6 cells into the right footpads of mice. These effects of curcumin were mostly PRL-3-specific as PRL-1 and -2 expression for the most part was not downregulated by curcumin, except for a mild reduction (30%) of PRL-1 at high dose (40 μM) of curcumin, a dose which reduces PRL-3 expression ~90%. As drugs from natural products sometimes display reduced side effects and toxicity, the development of the biflavinoids and curcumin may provide safe and potent PRL-3 inhibitors for therapeutic use.  1.4: Insulin-like growth factor (IGF)-I signaling PRL-3 expression has been shown to alter the expression or activity of several molecules, but a role for PRL-3 has not been reported in growth factor signaling. Canonical growth factor signaling involves the binding of an extracellular ligand to the extracellular domain of a transmembrane receptor tyrosine kinase which results in a phosphorylation cascade of effector proteins leading to changes in gene transcription. There are quite a number of growth factor signaling pathways and they share some, but not all effector proteins and features. One key growth factor pathway is the insulin-like growth factor (IGF)-I signaling pathway.  The ligand IGF-I binds to the IGF-I receptor (IGF-IR), a heterodimer molecule  consisting of an extracellular α-chain covalently bound to the extracellular domain of the  43  transmembrane β-chain, which has cytoplasmic tyrosine kinase activity. IGF-I binding to the αchain results in crosslinking two heterodimers and activation of autophosphorylation of Tyr residues on the cytoplasmic tails of the β-chains (Leroith et al., 1995). The phosphorylation of the cytoplasmic tails recruits signaling and scaffolding molecules to the IGF-IR complex and initiates signaling cascades (Blakesley et al., 1996). For example, the IGF-IR substrates insulin related substrate (IRS)-1 and -2 are phosphorylated and subsequently recruit SH2-domain containing proteins such as PI3K kinase, which in this case, leads to activation of the Akt pathway. Likewise, IGF-IR can phosphorylate Shc, which in turn recruits Grb2. This results in association with the guanine nucleotide exchange protein Sos and leads to activation of the Ras/Raf/MAPK pathway. Activation of the downstream pathways mediate cell processes such as apoptosis, proliferation, differentiation and migration, and IGF-I signaling is disrupted in numerous cancers and in other diseases (Brodt et al., 2000; Gross and Yee, 2003; Lewis et al., 1993; Wu et al., 2006).  1.5: Adherens junction complexes Adherens junctions (AJs) are multi-protein complexes that mediate cell-cell contacts. The canonical AJ complex consists of the transmembrane receptor E-cad, β-catenin or γ-catenin, p120-catenin and α-catenin. The extracellular domain of E-cad can homodimerize in trans with the extracellular domain of E-cad from an adjacent cell (Meng and Takeichi, 2009). β-catenin and p120-catenin bind to distinct domains on the cytoplasmic tail of E-cad and link, respectively, to the actin cytoskeleton through association with α-catenin and EPLIN, which bundles F-actin filaments, or to microtubule minus end through association with PLEKHA7 and Nezra. Through these interactions the actin cytoskeleton and microtubules are linked to the cell membrane AJ complexes of adjacent cells, creating tension to form tight junctions between the cells.  44  Disruption of AJ complexes are required for cell-cell scattering and is associated with cell movement.  1.6: Prostate cancer PRL-3 has been shown to correlate with aggressiveness, increased metastases, increased angiogenesis and lower survival in a number of human cancers. These cancers have included the common breast, colon and lung carcinomas. However, little has been reported with respect to prostate cancer. In a small study, PRL-3 expression was found to be upregulated in 5/53 prostate tumours (Wang et al., 2010). Previous unpublished work in the Pallen lab has shown that cytoplasmic PRL-3, but not nuclear PRL-3 expression, correlates with increasing tumour stage of prostate cancer (Wong, 2005b). Prostate cancer (PCa) is the most common form of cancer in North American men and the second most frequent cause of cancer death, behind lung cancer, for men in Canada and the USA. While most men with PCa have a slow-growing, non-aggressive form and, indeed, may die of other causes without ever knowing they have PCa, about one-third of PCa is aggressive and requires treatment. The development of efficient and accurate diagnostic tests for PCa increased it detected incidence over the last several decades, though its incidence rate has plateaued for the last several years. The mortality rate decreased over the decade from 1995 to 2004. As the death rate for PCa is so great (~4300 men in Canada in 2010), better understanding of the molecular mechanisms behind PCa would facilitate the discovery of new targets for therapeutic treatments. Several cell lines have been generated as models of PCa disease. Some examples include LNCaP, C4-2, DU-145 and PC-3 cells. The LNCaP cell line was derived from a lymph node metastatic lesion of prostatic adenocarcinoma. LNCaP cells express the androgen receptor (AR)  45  and are androgen responsive, and are positive for p53 expression but negative for PTEN expression (van Steenbrugge et al., 1989). The C4-2 cell line is a sub-line of LNCaP cells (Wu et al., 1994) which also express the AR, but are not responsive to androgen, and remain p53 positive and PTEN negative. These cells were generated by coculturing LNCaP cells with the bone stromal cell line MS to form carcinomas in castrated mice hosts.  The androgen  independent carcinomas were extracted and cultured in vitro to generate new cell lines. One of these cell lines, C4, was then cocultured with MS cells, in castrated mice hosts and the above process was repeated. The resulting cell line was designated C4-2. The DU-145 cell line was derived from a brain metastatic lesion from prostatic adenocarcinoma. DU-145 cells have little or no AR expression, express a mutant, inactive p53 protein but express intact PTEN. PC-3 cells are derived from a bone metastasis of prostate adenocarcinoma. The PC-3 cells do not express the AR, p53 or PTEN.  1.7: Rationale and hypothesis PRL-3 is associated with aggressiveness and metastases of a number of cancers. At the onset of this study, little was known of the signaling mechanisms in which PRL-3 could mediate these effects. Early research with overexpression of PRL-3 in cell systems revealed that PRL-3 could promote metastasis-associated properties in non-cancerous cells (Zeng et al., 2003b). This led to the formation of my initial hypothesis for this study: PRL-3 functions in specific signaling pathways that control cell processes key to the progression and metastasis of human cancers. I developed the following aims to investigate my hypopthesis: 1. To develop inducible expression systems of PRL-3 in HEK 293 cells to investigate the role of PRL-3 in non-cancer cell signaling.  46  2. To identify alterations in cell signaling dependent upon PRL-3 activity. I generated the inducible Flp-In T-Rex 293 PRL-3-expressing cell system to determine what alterations in HEK 293 cell signaling that PRL-3 expression would promote focusing upon signaling molecules important in cell adhesion, migration and invasion. As my research progressed, further reports surfaced on the activity of PRL-3 in cell signaling and in cancer. Research in our lab also indicated that cytoplasmic PRL-3 expression in prostate cancer significantly correlated with prostate cancer aggressiveness and poor survival (Wong, 2005b). As a role for PRL-3 in prostate cancer had not been described in the literature, I decided to investigate the role of PRL-3 in prostate cancer cell signaling with the following hypothesis in mind: PRL-3 promotes prostate cancer aggressiveness by activating pathways that are central to tumourigenesis and metastasis. I developed several aims to investigate the above hypothesis: 1. To generate a PRL-3 overexpression system in LNCaP prostate cancer cells. 2. To generate stable knockdown of endogenous PRL-3 in prostate cancer cell lines. 3. To determine whether alterations in PRL-3 expression can regulate cancerassociated properties in prostate cancer cells. 4. To investigate the role of PRL-3 in EMT in prostate cancer cells. I generated EGFP-PRL-3 expression plasmids which were stably transfected into LNCaP cells to investigate the role of overexpressed PRL-3 and examine its localization and potential association with adherens junction proteins in prostate cancer cells. I also established stable knockdown of endogenous PRL-3 in LNCaP, C4-2 and DU-145 cells which were used to  47  examine the effect of loss of PRL-3 on prostate cancer cell phenotypic properties and upon some of the signaling molecules that may mediate these processes.  48  Chapter 2: Materials and Methods 2.1: Generation and analysis of Flp-In™ T-Rex™ HEK 293 cells that inducibly express FLAG-PRL-3 wildtype and FLAG-PRL-3 mutant (C104S) 2.1.1: Cell culture For these studies, I chose to use Human Embryonic Kidney 293 (HEK-293) cells transfected with the Flp-In™ T-Rex™ Expression System (Flp-In™ T-REx™ 293 Cell Line, Invitrogen). This expression system uses a tetracycline-based repressor to control transgene expression. Two vectors are stably transfected into the HEK-293 cells - the pFRT/lacZeo vector, which encodes the lacZ-Zeocin fusion and contains a single integrated Flp Recombination Target (FRT) site, and the pcDNA6/TR vector which encodes the Tet repressor protein. The Flp-In™ TRex™ 293 cells were maintained in Dulbecco’s Modified Eagle Medium (DMEM, Gibco), containing 10% fetal bovine serum (FBS) and penicillin/streptomycin. Cells transfected with the above vectors were maintained with the addition of 100µg/ml Zeocin (Invitrogen), to select for pFRT/lacZeo, and 15µg/ml blasticidin (Invitrogen), to select for pcDNA6/TR. The Flp-In T-Rex 293 cells were maintained in a humidified incubator at 37ºC with 5% CO2. These are the “parental” cells in this study and will be referred to as the 293 cells. 2.1.2: Generation of the FLAG-tagged PRL-3 wild-type and catalytically-inactive mutant C104S vectors PRL-3 (wt) and catalytically-inactive PRL-3 (C104S) were subcloned from pXJ40-mycPRL-3 (wildtype, wt) and pXJ40-myc-PRL-3 (C104S) vectors (Zeng et al., 2000), respectively, by restriction digest using BamHI (5’-enzyme) and XhoI (3’-enzyme). The fragment was cloned into the BamHI and XhoI sites of the pXJ40-FLAG vector (gift from Dr. Esther Verheyen,  49  Burnaby, Canada) to generate in-frame fusion FLAG-PRL-3 (wt) or (C104S) transgenes. After selection of correct pXJ40-FLAG-PRL-3 (wt) or (C104S) plasmids using standard molecular biology techniques, sequencing was conducted to confirm correct insert orientation and sequence. In sequencing the above plasmids, it was discovered that there was a sequence mismatch in the PRL-3 (wt) sequence (but not in the PRL-3 (C104S) sequence) compared to sequence data published in the National Center for Biotechnology Information (NCBI) public database (http://www.ncbi.nlm.nih.gov). The amino acid at residue 101 was leucine in the pXJ40-FLAGPRL-3 (wt) vector but should be an alanine based on the sequence deposited at the NCBI database. To rectify this discrepancy, the pXJ40-FLAG-PRL-3 (wt) vector was subjected to sitedirected mutagenesis using the QuikChange II Site Directed Mutagenesis Kit (Agilent Technologies) following the manufacturer’s protocol. Briefly, the forward mutagenesis primer (PRL3WT-F: 5’- ccgggaagctgcgtacttgtgcactgttgtggc – 3’) and the reverse mutagenesis primer (PRL3WT-R: 5’- ggcacaacagtgcacaagtacgcagcttcccgg -3’) (the mismatch residues are highlighted in bold) were used in a PCR reaction with an aliquot of the pXJ40-FLAG-PRL-3 (wt) vector. PCR was conducted in an Mastercycler® Thermal Cycler (Eppendorf). The PCR products were treated with the restriction enzyme DpnI to digest the parental methylated and hemimethylated DNA then transformed into competent E. coli cells. Standard molecular biology techniques were used to isolate transformed cells with the correct pXJ40-FLAG-PRL-3 (wt) vectors which were then sequenced to confirm the change from the leucine to alanine after mutagenesis.  50  2.1.3: Generation of the FlpIn T-Rex-293-FLAG-PRL-3 cell lines FLAG-PRL-3 (wt) or (C104S) fusion genes were excised from the pXJ40-FLAG-PRL (wt) or (C104S) vector, respectively, with the restriction enzymes BamHI and XhoI and the fragments were blunt-ended using Klenow fragment. The expression vector pcDNA5/FRT/TO (Invitrogen) was digested at its multiple cloning site with the restriction enzyme EcoRV. The vector and the fusion gene fragments were ligated and selection of the correct pcDNA5/FRT/TOFLAG-PRL-3 (wt) or (C104S) plasmids was conducted using standard techniques.  The  pcDNA5/FRT/TO vector has a tetracycline-inducible promoter upstream of the multiple cloning site, allowing for controlled expression of FLAG-PRL-3 (wt) or (C104S). The advantage of the Flp-In™ T-Rex™ expression system is the use of the FRT recombination sites in the pcDNA5/FRT/TO expression vector and in the pFRT/lacZeo to mediate the site-specific recombination of the target fusion gene into the genome. The Flp recombinase protein, encoded by the pOG44 vector (Invitrogen), mediates DNA recombination between the two vectors, leading to loss of lacZeo (and Zeocin resistance) and introduction of the gene of interest. Every recombination event occurs at the same site in the genome (where the pFRT/lacZeo is inserted) allowing for “polyclonal” selection of transfected cells and ensuring that each separate vector is inserted into the same site. This latter property reduces variable transgene expression between independent transfection events due to genome siting. pcDNA5/FRT/TO FLAG-PRL-3 (wt) or (C104S) vectors were co-transfected with the pOG44 vector into the 293 cells following the manufacturer’s protocol.  Briefly, 9 µg of  expression plasmid DNA was mixed with 1 µg of pOGG44 in Opti-MEM media (Gibco) while 10 µl lipofectamine was diluted in Opti-MEM. The DNA and lipofectamine were mixed and incubated at room temperature for 30 minutes before the mixture was added to the 293 cells.  51  The cells were cultured for 24 hours in DMEM media containing FBS but not selection antibiotics. The media was refreshed and blasticidin (but not Zeocin) was added to a final concentration of 15 µg/ml. After another 48 hours, the cells were split to ~ 20-25% confluency, allowed to attach for 2-3 hours and fresh media containing blasticidin (15 µg/ml) and hygromycin B (50 µg/ml; the selection antibiotic for the pcDNA5/FRT/TO vector) were added to the cells. The cells were cultured under selection until surviving cells formed single colonies that were visible in the culture dish. The colonies were pooled, assessed for FLAG-PRL-3 (wt) or (C104S) expression, and used for further studies. 2.1.4: 293 cell proliferation assay 293 parental, FLAG-PRL-3 (wt)-expressing or FLAG-PRL-3 (C104S)-expressing cells were seeded into 10-cm dishes and grown overnight in the presence of 1 µg/ml of doxycycline (Sigma) to induce transgene expression. Doxycycline acts in a similar fashion as tetracycline to induce expression from the pcDNA5/FRT/TO-FLAG-PRL-3 (wt) or (C104S) transgenes, but is more stable.  The next day (after ~16 hours), cells were collected using 0.25% trypsin  (Invitrogen), counted and seeded in triplicate onto 96-well plates at a density of 5x103 cells/well in full media containing 1 µg/ml of doxycycline. Cells were grown for 24, 48, 72 or 96 hours before being treated with the CellTiter 96® AQueous Non-Radioactive Cell Proliferation Assay (MTS:  3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-  tetrazolium, inner salt) kit (Promega) following the manufacturer’s protocol. Cells seeded onto 96-well plates and immediately assayed with the MTS reagent were treated as “0” hour. Cells were incubated with MTS for 90 minutes at 37°C before the absorbance at 490 nm was read on an Enspire 2300 Multilabel Reader (Perkin-Elmer) plate reader. Wells containing media plus MTS Assay reagent alone were used as blank controls.  52  2.1.5: 293 cell migration assay 293 parental, FLAG-PRL-3 (wt)-expressing or FLAG-PRL-3 (C104S)-expressing cells were seeded into 10-cm dishes and grown overnight in 1 µg/ml of doxycycline to induce transgene expression. The next day, cells were collected using 0.25% trypsin, counted and seeded into the top chamber of Transwell inserts (8.0 µm pore, Corning). The Transwell membranes were previously coated on the underside with 20 µg/ml of fibronectin (Sigma). Cells (8x104) suspended in 100 µl serum-free media (SFM) were added to the top chamber, 500 µl of SFM was added to the bottom chamber and the chambers incubated for 60 minutes at 37°C. The inserts were rinsed in PBS, fixed by incubation in 100% methanol for 30 minutes at 4°C and stained with Karyomax Giemsa Stain Solution (Invitrogen) for 45 minutes at room temperature. The top side of the membranes was wiped with a cotton swab to remove non-migrated cells and rinsed several times in PBS to reduce background Giemsa staining. The membranes were excised from the insert and mounted on microscope slides. Photomicrographs were taken at 200x magnification on a Leica DMIL microscope and images collected using a Leica DFC320 digital camera and the Openlab 4.0.2 software.  The number of cells that migrated to the  underside of the membrane in six fields of view was counted. 2.1.6: 293 cell lysis, immunoprecipitation and immunoblotting 293 parental, FLAG-PRL-3 (wt)-expressing or FLAG-PRL-3 (C104S)-expressing cells were harvested in modified RIPA buffer (50 mM Tris-Cl, pH 7.5, 150 mM NaCl, 5 mM EDTA, pH 8.0, 1% NP-40, 0.5% sodium deoxycholate, 0.1% sodium dodecyl sulphate (SDS), 50 mM NaF, 1 mM phenylmethanesulphonyl fluoride (PMSF), 10 µg/ml aprotinin, 10 µg/ml leupeptin, 5mM sodium vanadate) and clarified by centrifugation at 14000 rpm at 4°C in a Hettich Mikro 200R centrifuge (Sigma). The protein concentration of cell lysates was determined using the  53  Bio-Rad Protein Determination Reagent (Bio-Rad Laboratories) and absorbance measurements made at 595 nm (UltraSpec, Pharmacia Biotech). For immunoprecipitations (IPs), 500 µg of lysate was incubated with 1-2 µg of antibody for 1-2 hours on ice, then 40 µl of Protein A/G agarose beads (Santa Cruz Biotechnology) was added and the mixture rotated overnight at 4ºC. After three wash steps in 1 ml of modified RIPA buffer, the immunoprecipitated proteins were eluted by boiling for five minutes in 20 µl of 2X Laemmli buffer (4% SDS, 125mM Tris-Cl, pH 6.8, 20% glycerol, 0.004% bromophenol blue, 10% β-mercaptoethanol). For immunoblotting, either the entire IP or 20-50 µg of cell lysate standardized to equal volumes was resolved by electrophoresis using the Bio-Rad Mini-Protean 3 electrophoresis and Trans-blot system (BioRad Laboratories), and proteins were then electrotransferred from the gel onto polyvinylidene fluoride (PVDF) membranes. Membranes were blocked with either 3% bovine serum albumin (BSA) or 5% skim milk in phosphate-buffered saline (PBS) containing 0.1% Tween-20 (PBS-T) for 1 hour at room temperature or overnight at 4ºC.  Antibodies were diluted in PBS-T  containing 1% BSA or 5% skim milk and incubated with the membranes for 2 hours at room temperature or overnight at 4ºC.  After several washes in PBS-T, appropriate secondary  antibody, conjugated to horse radish peroxide (HRP) and diluted in PBS-T containing 1% BSA, was incubated with the membranes for 1-2 hours at room temperature. Membranes were washed and proteins detected using the enhanced chemiluminescence (ECL) system with film development in an X-OMAT machine (Kodak). 2.1.7: Antibodies and reagents The following antibodies were used in the above experiments: rabbit anti-FLAG (M2), rabbit anti-actin, goat anti-mouse-HRP and goat anti-rabbit-HRP (Sigma); mouse anti-PTEN and rabbit anti-IGFR-Iβ (Santa Cruz); rabbit anti-phospho-Thr202/Tyr204-Erk1/2 (P-Erk), rabbit  54  anti-Erk1/2 (Erk), rabbit anti-phospho-Ser217/221-Mek1 (P-Mek), rabbit anti-Mek1 (Mek), rabbit anti-phospho-Akt Ser473 (P-Akt S473),  rabbit anti-Akt, rabbit anti-phospho-  Thr183/Tyr185-JNK (P-JNK), rabbit anti-JNK, rabbit anti-phospho-Thr180/Tyr182-p38 (P-p38) and rabbit anti-p38 (Cell Signaling); mouse anti-Csk and mouse anti-E-cadherin (BD Biosciences); rabbit anti-phospho-Src Tyr529 (P-Src Y529) and rabbit anti-phospho-Src Tyr418 (P-Src Y418) (Biosource); mouse anti-phosphotyrosine (4G10, Millipore), mouse anti-Src (Abcam). Insulin-like growth factor-I (IGF-I) was from Sigma and the Mek inhibitor U0126 was from Cell Signaling.  2.2: Generation and analysis of LNCaP cells stably overexpressing EGFP-PRL-3 (wt) and EGFP-PRL-3 mutants (C104S, C170S) 2.2.1: Prostate and breast cancer cell lines and antibodies LNCaP, DU-145 and PC-3 prostate cancer cell lines (gifts from Dr. Michael Cox, Vancouver, Canada) were maintained in RPMI 1640 media (Gibco) containing 10% FBS and penicillin/streptomycin in a humidified 37°C incubator at 5% CO2. LNCaP prostate cancer cells are derived from a metastatic site at the left supraclavicular lymph node, are androgen sensitive and have low tumorigenic potential (www.atcc.org). DU-145 prostate cancer cells are derived from a brain metastasis, are insensitive to androgen and have high tumorigenic potential (www.atcc.org). PC-3 prostate cancer cells are grade IV adenocarcinoma cells derived from a bone metastasis and have a high tumorigenic potential (www.atcc.org).  The MCF-7 breast  carcinoma cell line (ATCC), an adenocarcinoma derived from a pleural effusion metastasis (www.atcc.org), was maintained in phenol red-free RPMI 1640 media (Gibco) containing 10% FBS and penicillin/streptomycin in a humidified 37°C incubator at 5% CO2. Rabbit anti-actin, rabbit anti-fibronectin, rabbit anti-β-catenin and rabbit-PRL-3 antibodies were from Sigma. 55  Mouse anti-E-cadherin, mouse anti-N-cadherin and mouse anti-Csk antibodies were from BD Biosciences. Mouse anti-p53 and rabbit anti-p21cip1/Waf1 antibodies were from Santa Cruz. The 4G10 antibody was from Millipore and rabbit anti-phospho-Src Tyr418 was from Biosource. Mouse anti-v-Src and rabbit anti-α-catenin were from Chemicon. Rabbit anti-phospho-Ser139 H2A.X (phospho-H2AX) was from Cell Signaling Technologies. Alexa Fluro-594 goat antimouse and anti-rabbit secondary antibodies were from Molecular Probes. 2.2.2: Generation of the pEGFP-C2-PRL-3 expression vectors I used the pEGFP-C2 mammalian expression vector (Clontech) to create GFP-PRL-3 fusion protein expression vectors. The pEGFP-C2 vector encodes a red-shifted mutant form of GFP that has brighter fluorescence and higher expression in mammalian cells than the wildtype GFP. The GFP mut1 variant has an excitation maximum of 488 nm and an emission maximum of 507 nm. PRL-3 (wt), (C104S) and the prenylation-deficient mutant (C170S) were subcloned into the multiple cloning site of pEGFP-C2 using the EcoRI and BamHI restriction sites to create in-frame fusions of GFP-PRL-3 (wt), (C104S) and (C170S), respectively. The PRL-3 fragments were generated by PCR of the vectors pGEX-PRL-3 (wt), (C104S) and (C170S) using primers that incorporate an EcoRI restriction site in the 5’-primer and a BamHI restriction site in the 3’primer. After selection of the plasmids using standard molecular biology techniques, the fusion genes were sequenced to confirm correct orientation, proper in-frame insertion and sequence of the PRL-3 fragments. Henceforth, the parental vector will be referred to as GFP, and the fusion gene vectors pEGFP-C2-PRL-3 (wt), (C104S) and (C170S) will be referred to as GFP-PRL-3, GFP-PRL-3 (C104S) and GFP-PRL-3 (C170S), respectively.  56  2.2.3: Generation of stable GFP fusion protein LNCaP cell lines LNCaP cells were transfected with the GFP vectors using lipofectamine (Invitrogen) following the manufacturer’s protocol. Briefly, cells were seeded onto 60-mm dishes so as to be ~50% confluent at the time of transfection. The next day, 3 µg of DNA was diluted in 300 µl of Opti-MEM I Reduced Serum Medium (Gibco) without serum and combined with a solution of 6 µl of lipofectamine diluted in 300 µl of Opti-MEM without serum.  The complexes were  incubated for 30 minutes at room temperature then 600 µl of Opti-MEM was added to the complexes. Meanwhile, the growth medium was removed from the cells and 1.2 ml of OptiMEM added to each dish. The 1.2 ml of diluted complexes was added to each dish and mixed by gentle rocking.  The cells were incubated overnight (~16 hours) at 37°C and the medium  replaced with fresh complete growth medium (i.e. RPMI, 10% FBS, penicillin/streptomycin). At 72 hours post-transfection, the cells were passaged 1:10 and treated with 800 µg/ml G418 (Gibco) to select for colonies of cells containing the GFP vectors. Post-transfection and selection, the colonies were pooled and maintained in complete medium using 600 µg/ml G418 to keep selection pressure on the cells. The transfected cells were sorted using a BD FACSAria to select for high GFP intensity and, presumably, higher fusion gene expression, using the GFP fluorescence.  Briefly, 2x107 cells were collected,  resuspended in PBS and passed through a 25 µm sieve to break up cell clumps. The fluorescent compound 7-Aminoactinomycin D (7-AAD) was added to the cells to separate live and dead cells in the FACS analysis. After a first round of cell sorting, the cell lines were passaged and a second round of cell sorting was performed. Further experiments were conducted using these two-step sorted cells.  57  2.2.4: Immunofluorescence assay GFP-, GFP-PRL-3 (wt)-, C104S)- and (C170S)-expressing LNCaP cells were seeded into 6-well dishes containing sterile coverslips at a density of 1x105 cells/well and incubated at 37°C for 24 hours. The cells were then rinsed with PBS and fixed for 30 minutes in 100% methanol at 4°C. For examination of only GFP fluorescence in cells, the coverslips were mounted on microscope slides using VectorShield mounting medium containing DAPI (Vector Laboratories) and examined at 400X using a Leica DM4000B microscope and the images collected using a Qimaging digital camera and OpenLab 4.0.2 software.  For co-localization studies using  antibodies to detect protein markers, the cells were first blocked with 1% BSA in PBS-T for 1-2 hours at room temperature then incubated at 4°C overnight with antibody diluted in PBS containing 1% BSA. The coverslips were rinsed several times in PBS and incubated for 1-2 hours at room temperature in the dark with secondary antibodies conjugated to the fluorescent molecule Alexa Fluor-594 diluted in PBS containing 1% BSA. The coverslips were rinsed several times in PBS and mounted on microscope slides using VectorShield mounting medium containing DAPI. Photomicrographs were taken as described above. 2.2.5: GFP-PRL-3 LNCaP cell proliferation assay GFP-, GFP-PRL-3 (wt)-, C104S)- and (C170S)-expressing LNCaP cells were seeded in triplicate on 96-well dishes at a density of 2.5x103 cells/well and incubated for 24, 48, 72 or 96 hours at 37°C. Otherwise, the protocol was the same as in section 2.1.4. 2.2.6: GFP-PRL-3 LNCaP cell lysis and immunoblotting GFP-, GFP-PRL-3 (wt)-, C104S)- and (C170S)-expressing LNCaP cells were grown to 70-80% confluence and harvested in modified RIPA buffer.  The lysates were treated as  described in Section 2.1.6.  58  2.2.7: Treatment of prostate cancer and MCF-7 cells to induce p53 expression LNCaP, DU-145, PC-3 and MCF-7 cells were seeded onto plates so as to be ~70% confluent the next day. Doxorubicin or cobalt chloride was added to the cells at 0.2 µg/ml or 600 µg/ml, respectively, and the cells incubated for 24 hours at 37°C. Cells were then washed, lysed and analyzed for protein expression by immunoblotting.  2.3: Generation and analysis of prostate cancer cells with transient and stable PRL3 knockdown 2.3.1: Cell lines and antibodies DU-145 and LNCaP cells were as described in section 2.2.1. The cell line C4-2 is a LNCaP sub-line that was derived by co-culture of LNCaP cells with M2 bone stromal cells in castrated mouse hosts. The less tumorigenic LNCaP cells were progressed to the more highly tumorigenic C4-2 cells in a two-step process (Wu et al., 1994). C4-2 cells were maintained as described above for the LNCaP cells. Rabbit anti-actin, rabbit anti-fibronectin, rabbit-PRL-3 antibodies goat anti-mouse-HRP and goat anti-rabbit-HRP were from Sigma. Mouse anti-Ecadherin and mouse anti-Csk antibodies were from BD Biosciences. Rabbit anti-P-Erk, rabbit anti-Erk, rabbit anti-P-Akt S473, rabbit anti-Akt and mouse anti-cytokeratin 8/18 were from Cell Signaling Technologies. Mouse anti-PTEN, mouse anti-p53, rabbit anti-p21cip1/Waf1 and mouse anti-p27Kip1 antibodies are from Santa Cruz. The 4G10 antibody was from Millipore and rabbit anti-P-Src Y529 was from Biosource. Mouse anti-v-Src was from Chemicon. 2.3.2: Determination of proliferation and long-term survival in siRNA-treated PCa cells DU-145 and LNCaP cell lines were seeded into 96 well plates at 1x103 cells/well and 2.5x103 cells/well, respectively, in 80µl of media/well and allowed to adhere overnight.  59  Duplicate wells were set up for each transfection/condition. The next day (~16 h later) PRL-3 siRNA (Qiagen, cat. #: 1027400-21S2O1P3, sequence: 5’- CACCTTCATTGAGGACCTGAA3’) or AllStar Negative Control siRNA, a proprietary, validated negative control siRNA with minimal nonspecific effects on gene expression and phenotype (Qiagen, cat. #: 1027281), were transfected into cells using RNAiMax (Invitrogen) following the manufacturer’s protocol. Briefly, siRNA was diluted in media and mixed with RNAiMAX diluted 1: 50 in media in a 1:1 ratio. After a 30 minute incubation at room temperature, 20 µl of the siRNA transfection mix was added to each well to give a final siRNA concentration of 100 nM. At each time point after transfection (0, 24, 48, 72, 96, and 120 h), the cells were treated with the MTS Assay as described in section 2.1.4. 2.3.3: Generation of stable PRL-3 knockdown prostate cancer lines DU-145, LNCaP and C4-2 cells were seeded in a 12-well plate so that they would reach ~50% confluency for the viral infection the next day (24 hours hence). Media was removed from cells and replaced with media containing 5 µg/ml of polybrene (Santa Cruz Biotechnology). Lentiviral particles containing a proprietary, control, scrambled shRNA sequence (Santa Cruz Biotechnology, cat. # sc-108080) or a pool of three different PRL-3 shRNA plasmids (Santa Cruz  Biotechnology,  cat.  #  sc-39156-V,  A:  CCTTCATTGAGGACCTGAAtt;  B:  CCTCTAGCCTGTTTGTTGTtt; C: CTGTTCTCGGCACCTTAAAtt) were added to the appropriate wells at 105 particles/well or approximately 1 particle/cell. The culture media was replaced 24 h later and the cells were incubated for another 24 hours before being expanded into a 10-cm dish. After 48 h, 2 µg/ml puromycin dihydrochloride was added to the media to enable selection of stable clones. After 7 days of selection, the surviving cells were pooled and examined for loss of PRL-3 expression.  60  2.3.4: RNA isolation and semi-quantitative RT-PCR PRL-3 and control shRNA-transduced DU-145, LNCaP and C4-2 cells, or LNCaP and DU-145 cells treated with scrambled or PRL-3 siRNA, were grown to ~75% confluency in sixwell plates and lysed using the RNeasy Minikit (Qiagen) following the manufacturer’s protocol. Briefly, 350 µl of lysis buffer was added to the wells, cells were collected with a cell scraper and transferred to a Qiashredder column (Qiagen).  After a 1 minute spin at full speed in an  Eppendorf tabletop centrifuge, 325 µl of 70% ethanol was mixed with the flowthrough and the entire volume transferred to an RNeasy column. The RNA was bound to the column with a 15 second spin in the centrifuge and the column was washed once with Buffer RPE and twice with buffer RE. The RNA was eluted with 30 µl of RNase-free water and RNA concentration was determined by spectrophotometry at 260 nm. RT-PCR was conducted using the One-Step RT-PCR kit (Qiagen) following the manufacturer’s protocol. Briefly, 50 µg of RNA was mixed with a final concentration of 1x buffer mix, 1x Q-solution, 400 µM of each dNTP, 0.6 µM of the forward and reverse primers and the enzyme mix in a final volume of 25 µl. The forward PRL-3 primer sequence was: 5’ttcgaattcatggctcggatgaaccgcccg  -3’.  The  reverse  PRL-3  primer  was:  5’-  cccggtgctgcgttatgtagggatccacc -3’. These PRL-3 primers will amplify both the full-length active PRL-3 transcript (519 bases) and a shorter, alternatively-spliced, inactive PRL-3 transcript (444 bases) (www.ncbi.nlm.nih.gov).  The forward actin primer sequence was: 5’-  ggccacggctgcttc -3’, and the reverse actin primer was: 5’- gttggcgtacaggtctttgc -3’. The PRL-3 PCR was conducted on a Mastercycler Thermal Cycler with the following conditions: 30 minutes at 50°C; 15 minutes at 95°C; 30 cycles of 94°C for 30 seconds, 55°C for 30 seconds,  61  72°C for 90 seconds; then a final extension at 72°C for 7 minutes. The actin PCR was conducted in the same manner except only 26 cycles were used.  2.3.5: Immunohistochemistry and immunofluorescence PRL-3 and control shRNA-transduced DU-145, LNCaP and C4-2 cells were grown to 60-80% confluency on chamber slides (Thermo Fisher Scientific) in full media under puromycin selection. After a brief rinse in PBS, cells were fixed in 4% formaldehyde in PBS for 20-30 minutes at room temperature. The slides were rinsed in PBS and incubated for one hour in blocking solution (1% BSA in PBS-T). For immunohistochemistry, rabbit anti-PRL-3 antibody (1:200 in PBS-T with BSA) generated and validated by our lab (Wong, 2005a) was incubated with the cells overnight at 4°C. The slides were rinsed several times in PBS and incubated for 60 minutes at room temperature with HRP-conjugated goat anti-rabbit IgG diluted 1:200 in PBS-T. The slides were rinsed several times in PBS and treated with the NovaRED substrate kit for peroxidase (Vector Laboratories), following the manufacturer’s protocol, until PRL-3 staining was detected (about 15 minutes).  In some experiments, slides were counterstained with Mayer’s hematoxylin  solution (1g/L, Sigma) for 2-5 minutes. The slides were examined at 100X and 400X using a Leica DM4000B microscope and the images collected using a Qimaging digital camera and OpenLab 4.0.2 software. F-actin was detected using Alexa Fluor-594-conjugated phalloidin (Molecular Probes) in immunofluorescence studies. The phalloidin was diluted in PBS (1.5:80) and incubated with the cells overnight at 4°C after which the slides were rinsed several times in PBS. The slides were  62  examined at 400X using an Olympus IX81-ZDC microscope and the images collected using a Photometrics CoolSnap HQ2 digital camera and Metamorph 7.6 software.  2.3.6: Determination of proliferation and long-term survival of the lentiviral shRNA transduced PCa cells The DU-145 (1x103 cells/well), LNCaP (2.5x103 cells/well) and C4-2 (2.5x103 cells/well) shRNA knockdown cell lines were seeded, in triplicate for each condition, onto 96 well plates and allowed to adhere overnight. At each time point after seeding (24, 48, 72, and 96), the cells were treated with the MTS reagent as described in section 2.1.4. For LNCaP and C4-2 shRNA knockdown cells grown in charcoal-stripped FBS (CSS), cells were cultured in media containing CSS. These cells were then treated as above, with the exception that they were seeded into 96 well plates in media containing CSS. 2.3.7: shRNA-transduced prostate cancer cell migration and invasion The LNCaP, C4-2 and DU-145 shRNA knockdown cell lines were subjected to Boyden chamber assays to determine the effect of PRL-3 knockdown on migration using the protocol from section 2.1.5 with the following differences.  DU-145 (2.5x104), C4-2 (2.5x104) and  LNCaP (5x104) shRNA-transduced cells were resuspended in 100 µl of serum free RPMI containing 2 µg/ml of puromycin and added to the top chamber of the Transwell insert. RPMI containing 10% FBS and 2 µg/ml of puromycin (500 µl) was added to the bottom well of the Transwell insert and the cells incubated at 37°C for 24 or 48 hours. After the fixation in 4% paraformaldehyde, membranes were mounted on microscope slides in VectorShield mounting media containing DAPI to visualize nuclei. Photomicrographs were taken at a magnification of 400x on an Leica DM4000B microscope and the images collected using a Qimaging digital  63  camera and the OpenLab 4.0.2 software. The number of cells that migrated to the underside of the membrane was counted in six fields of view. A similar protocol was used to determine the effect of PRL-3 knockdown on the prostate cancer cell line invasion phenotypes. For this, test cells (DU-145 and C4-2 lines at 5x104 cells per well, LNCaP lines at 1x105 cells per well) were added to the top chamber of BD Biocoat Matrigel Invasion Chambers using (8 µm pore membrane, BD Biosciences, Canada) in 100 µl of serum free RPMI containing 2 µg/ml of puromycin. The top side of the membrane is coated with a layer of Matrigel to mimic extracellular matrix proteins as an invasion substrate. RPMI containing 10% FBS and 2 µg/ml of puromycin (750 µl) was added to the bottom well and the cells incubated for 24 hours at 37°C. After the incubation the wells were rinsed in PBS, and the top side of the membrane was wiped with cotton swabs to remove cells and Matrigel. The cells on the bottom side of the membrane were fixed in cold methanol for at least 30 minutes. The cells were then stained with Karyomax Giemsa Stain (Invitrogen) and destained with several washes in PBS. The membranes were mounted on microscope slides, photomicrographs were taken at a magnification of 200x on an Leica DMIL microscope and the images collected using a Leica DFC320 digital camera and the OpenLab 4.0.2 software. The number of cells that invaded through the Matrigel to the underside of the membrane was counted in six fields of view. 2.3.8: Colony formation assays To assess anchorage-independent growth, the shRNA PCa cell lines were subjected to colony formation assays in soft agar. Six-well plates were coated with 0.6% agar containing full media (RPMI, 10% FBS, penicillin/streptomycin and 2 µg/ml of puromycin). Cells were mixed with 0.3% agar containing full media as above and plated in triplicate wells (3x103cells/ml/well). The cells were grown at 37°C and fed every 4-5 days with 0.5 ml of full media. Colonies were  64  allowed to form over two weeks (DU-145 cell lines) or four weeks (LNCaP and C4-2 cell lines). Colonies were stained with 0.005% crystal violet, those over 25 µm (DU-145) or 10 µm (LNCaP and C4-2) in size were counted, and the average colony number/well from 3 wells was calculated. 2.3.9: Prostate cancer cell lysis and immunoblotting The LNCaP, C4-2 and DU-145 shRNA knockdown cells were grown to 75-80% confluency, rinsed with cold PBS and lysed in modified RIPA buffer and clarified by centrifugation. The lysates were treated as described in Section 2.1.6. Immunoblotting was performed as described (Section 2.1.6) or with the following method.  Proteins were  electrotransferred from the gel onto nitrocellulose membranes (Bio-Rad Laboratories) for detection using chemifluorescence on the Odyssey Infrared Imaging System (LI-COR Biosciences). Nitrocellulose membranes were blocked with either 5% skim milk in PBS, for pan-specific antibodies, or Odyssey Blocking Buffer (LI-COR Biosciences) containing 50 mM sodium fluoride or 0.1 mM sodium vanadate, for phosphotyrosine or phosphoserine/threonine antibodies, respectively. The following secondary antibodies were used: IRdye800 goat antimouse IgG (cat. #: CA610-132-121) and IRdye800 sheep anti-rabbit IgG (cat. #: CA611-632122) from VWR Canlabs; and DyLight 680 goat anti-rabbit IgG (cat. #: PI35569) and DyLight 680 goat anti-mouse IgG (cat. #: PI35519) from Thermo Fisher Scientific. The secondary antibodies are highly cross-absorbed to minimize cross reactivity between mouse and rabbit IgG, thus enabling multiplexing immunoblotting.  These secondary antibodies emit light at 680 or  800 nm and are detected on the 700 or 800 nm channels, respectively, of the Odyssey Infrared Imaging System. Immunoblot images were obtained and analyzed using Odyssey V3.0 software.  65  Chapter 3: The Role of FLAG-PRL-3 Expression in FlpIn TRex HEK 293T Cells  Upregulated PRL-3 expression was shown to correlate with colon carcinoma metastases in a global gene expression screen to search for differences in gene expression between primary colon tumors and their metastatic derivatives (Saha et al., 2001).  Subsequently, it was  demonstrated that overexpression of PRL-3 in CHO cells increased their ability to migrate and to invade into Matrigel, supporting the hypothesis that PRL-3 is a metastatis-promoting protein (Zeng et al., 2003b). However, at this time, little was known of the molecular and signaling mechanisms by which PRL-3 enhanced or conferred metastatic properties. One early report indicated a potential role for PRL-3 in angiotensin (Ang)-II signaling in HEK293 cells, as an inhibitor of calcium mobilization stimulated by Ang-II (Matter et al., 2001). To investigate how PRL-3 may modulate cell signaling I generated mammalian expression constructs for overexpression of PRL-3 and an inactive mutant that has the catalytic cysteine mutated to a serine (PRL-3 (C104S)).  3.1: Generation of inducible FLAG-PRL-3 cell lines The study examining the role of PRL-3 in migration and invasion found that high expression of PRL-3 (wt) in CHO cells led to cell toxicity (Zeng et al., 2003b). To circumvent this problem I introduced FLAG epitope-tagged PRL-3 (wt) and (C104S) constructs into a system that prevents transgene expression unless the cells are incubated with a factor that relieves repression of the promoter region in the expression vector. The Flp-In™ T-Rex™ inducible expression vector system controls expression of the gene of interest by using Tet-  66  repressor promoters.  Tetracycline or the more stable analogue doxycycline binds the Tet-  repressor, causing a conformational change and release of transcriptional repression of the gene of interest. The system uses Flp recombination target (FRT) sites to mediate site-specific DNA recombination of the gene of interest with the integrated Tet-repressor controlled vector. Commercially available Flp-In T-Rex 293 cells (henceforth referred to as 293 cells; Invitrogen) were used to facilitate generation of the FLAG-PRL-3 (wt) and (C104S) inducible expression cell lines (see Materials and Methods, sections 2.1.2 and 2.1.3, for a detailed description of the cloning and transfection). I incubated the 293 cells and the Flp-In™ T-Rex™ 293-FLAG-PRL-3 (wt) and (C104S) (henceforth referred to as FLAG-PRL-3 (wt) or (C104S), respectively) with doxycycline to confirm that FLAG-PRL-3 expression is controlled by doxycycline induction. Increasing amounts of doxycycline for 24 hours led to induction of transgene expression in the FLAG-PRL-3 (wt)- and (C104S)-expressing, but not the parental, 293 cells (Fig 3.1A). Maximal FLAG-PRL-3 (wt) expression was observed using 1000 ng/ml of doxycline; however, only 10 ng/ml of doxycycline was required for maximum FLAG-PRL-3 (C104S) expression.  I  conducted a short time course with 1 µg/ml of doxycycline to determine the kinetics of FLAGPRL-3 (wt) and (C104S) protein expression (Fig. 3.1B). FLAG-PRL-3 (wt) protein expression peaked at 16 hours and remained constant to 24 hours while FLAG-PRL-3 (C104S) protein expression continued to increase to 24 hours. Interestingly, there is a several fold greater FLAGPRL-3 (C104S) than FLAG-PRL-3 (wt) protein expression upon doxycycline treatment of the cells. As both the FLAG-PRL-3 (wt) and (C104S) vectors integrate into the same genomic site during transfection, the lesser FLAG-PRL-3 (wt) expression is not due to genomic-siting effects that may decrease transcriptional activity. It is probable that the cells limit the amount of FLAG-  67  PRL-3 (wt) protein through translational or degradation mechanisms, perhaps in order to limit toxicity associated with high levels of the active phosphatase.  Figure 3.1: The expression of FLAG-PRL-3 (wt) and (C104S) can be induced with doxycycline in the transfected FlpIn TRex 293 cell lines. A. Cells were incubated with 0 ng/ml (lane 1), 1 ng/ml (lane 2), 10 ng/ml (lane 3), 100 ng/ml (lane 4) or 1000 ng/ml (lane 5) doxycycline for 24 hours and the lysates probed for FLAG-PRL-3 (wt) or (C104S) expression using the anti-FLAG antibody. B. Cells were incubated with 1000 ng/ml doxycycline for 0 (lane 1), 2 (lane 2), 6 (lane 3), 16 (lane 4) or 24 hours (lane 5) and the lysates probed for FLAG-PRL-3 (wt) or (C104S) expression using the anti-FLAG antibody. Lysates were reprobed for actin as a loading control.  3.2: FLAG-PRL-3 expression does not affect proliferation but increases migration of 293 cells in a phosphatase-independent fashion A role for PRL-3 in cell proliferation and cell migration had previously shown in other cell systems ((Matter et al., 2001; Zeng et al., 2003b). To determine whether the induction of FLAG-PRL-3 (wt) expression would lead to similar phenotypes in the 293 inducible expression system, I examined cell proliferation and migration after doxycycline treatment.  68  Cells were grown overnight in the presence of 1 µg/ml of doxycycline to induce transgene expression, then seeded into 96-well plates in the continuing presence of doxycycline and assayed for proliferation at 24, 48, 72 and 96 hours. The proliferation of FLAG-PRL-3 (wt) cells was not changed compared to parental controls (Fig. 3.2A).  The expression of the  catalytically-inactive mutant also had no effect on the proliferation of the 293 cells (Fig. 3.2A). As the catalytically-inactive mutant is unable to release substrates due to its inability to dephosphorylate (Neel and Tonks, 1997), it is likely that it acts as a substrate trapping mutant. If this is indeed the case, then the lack of effect of FLAG-PRL-3 (C104S) on cell proliferation indicates that sequestering PRL-3 substrates does not affect 293 cell proliferation. I conducted Boyden chamber assays using Transwell inserts to address the effect of FLAG-PRL-3 (wt) and (C104S) on 293 cell migration. Cells were treated overnight with doxycycline to induce PRL-3 transgene overexpression and seeded into Transwell inserts coated on the underside with fibronectin (FN).  Compared to parental 293 cells, FLAG-PRL-3  overexpression resulted in a five-fold increase in migration of cells to the underside of the insert membrane (293 cells, 32 ± 25 cells; FLAG-PRL-3 (wt) cells, 148 ± 27 cells, p<0.0005, n=3) (Fig. 3.2B).  Interestingly, this effect was not dependent upon PRL-3 catalytic activity as  expression of FLAG-PRL-3 (C104S) induced a four-fold increase in migration over control parental cells (131 ± 46 cells, p<0.005 vs. 293 cell controls, n=3) (Fig. 3.2B). There was no significant difference between the migratory abilities of the FLAG-PRL-3 (wt) and (C104S) cells in this assay. This phosphatase-independent function of PRL-3 in promoting migration has also been reported in SW480 colon carcinoma cells (Fiordalisi et al., 2006). This may indicate that PRL-3 can act to recruit signaling molecules involved in promoting migration through a noncatalytic mechanism, perhaps as a scaffolding protein.  69  Figure 3.2: FLAG-PRL-3 expression in FlpIn TRex 293 cells does not affect cell prolilferation but increases migration to fibronectin. Cells were incubated overnight with 1 µg/ml doxycycline to induce transgene expression. A. To measure proliferation, the cells were seeded into 96-well plates at 5x103 cells/well and cultured. At the indicated times they were incubated with MTS reagent for 90 minutes and the absorbance at 490nm determined. 293, ♦; FLAG-PRL-3 (wt), ■; FLAG-PRL-3 (C104S), ●. B. To analyze cell migration, cells (8x104/well) were seeded into Transwell chambers with the underside of the membrane coated with 20mg/ml of fibronectin after 60 minutes incubation in media-containing well. The inserts were removed and analyzed for cells that migrated to the underside of the membrane. The total cells counted in six fields of view are indicated as the mean ± S.D. (* p < 0.005 vs control; ** p < 0.0005 vs control; no statistically significant difference between FLAG-PRL-3 (wt) and (C104S), n = 3).  3.3: FLAG-PRL-3 expression does not affect global tyrosine phosphorylation or Erk or Akt cell signaling in unstimulated 293 cells The migration-promoting ability of FLAG-PRL-3 (wt) and (C104S) 293 cells suggests that this property could be related to the PRL-3-linked tumour metastatis. I wished to determine what cell signaling pathways could be involved in controlling PRL-3 pro-metastatic phenotypes. The examination of total cellular protein phosphorylation can provide details of whether overexpression of a protein affects cell signaling in a global manner. To this end cells were treated overnight (~16 hours) with 1 µg/ml of doxycycline to induce FLAG-PRL-3 (wt) or (C104S) expression and cell lysates were probed for protein tyrosine phosphorylation using anti-  70  phosphotyrosine antibodies. Overexpression of either FLAG-PRL-3 (wt) or (C104S) in 293T cells did not result in detectable alterations in protein tyrosine phosphorylation using two different types of antibodies (Fig. 3.3A and data not shown). This is in contrast to a recent study which reported robust enhancement of global protein tyrosine phosphorylation in PRL-3expressing HEK-293 cells (Liang et al., 2007). Two critical cell signaling molecules are the Erk-1/2 (Erk) mitogen-activated protein kinases (MAPK) and the serine/threonine protein kinase Akt. Activation of Erk is frequently associated with cell proliferation, differentiation and anti-apoptotic effects (Cobb et al., 1994b). The serine/threonine kinase Akt was first identified as an oncogene (Staal, 1987) and has also been implicated in cell survival, cell cycle control, metabolism and angiogenesis (Carnero, 2010; Mazure et al., 1997). To determine if Erk and Akt activation/signaling were affected by upregulated PRL-3 expression, cells were cultured overnight (~16 hours) in the presence of 1 µg/ml of doxycycline and the lysates were subjected to immunoblotting for these phospho-proteins.  Levels of  phosphorylated Erk-1/2 did not change in FLAG-PRL-3 (wt) or (C104) cells relative to the 293 control cells (Fig. 3.3B). Similarly, no change was detected in the levels of phospho-Ser473 of Akt, a key residue which is phosphorylated in activated Akt, upon FLAG-PRL-3 (wt) or (C104S) expression (Fig. 3.3B). During the course of this study, two reports were published that identified roles for PRL3 in regulating specific molecules/cell signaling. PRL-3 expression in HEK 293 cells was shown to reduce the expression of C-terminal Src kinase (Csk), a kinase that phosphorylates the inhibitory tyrosine residue of Src (Tyr529) to reduce Src activity (Liang et al., 2007). Also, another group demonstrated that PRL-3 expression in the DLD-1 colon carcinoma cell line  71  Figure 3.3: FLAG-PRL-3 expression has little effect on protein phosphorylation in unstimulated cells. A. Lysates from doxycycline-treated (24 hours) cells were prepared: lane 1, 293; 2, FLAG-PRL-3 (wt); 3, FLAG-PRL-3 (C104S). These lysates were probed with the antiphosphotyrosine antibody 4G10 to determine the extent of protein tyrosine phosphorylation, and reprobed for actin as a loading control. B. Cell lysates were probed for phospho-Erk and phospho-Akt Ser473, then reprobed for Erk and Akt. C. Cell lysates were probed for Csk, PTEN and E-cad expression and reprobed for actin. They were further examined for phospho-Src-Tyr529 and reprobed for Src. Data are representative of at least three independent experiments.  72  increased Akt phosphorylation by downregulating phosphatase and tensin homologue deleted on chromosome 10 (PTEN), an important antagonist of phosphatidylinositol-3 kinase (PI3K) and thus of the PI3K substrate Akt (Wang et al., 2007a). The same report demonstrated that PRL-3 overexpression promoted epithelial-mesenchymal transition (EMT), as evidenced by decreasing protein expression of the epithelial marker, E-cadherin (E-cad). I sought to determine whether PRL-3 expression would induce similar effects in the 293 cell system. Cells were cultured overnight (~16 hours) in the presence of 1 µg/ml of doxycycline and the lysates were examined for the expression and/or phosphorylation status of the above mentioned proteins. Levels of Csk and phospho-Src Tyr529 were unchanged by expression of FLAG-PRL-3 (wt) or (C104S) (Fig. 3.3C). PTEN protein expression was also unaltered by expression of FLAG-PRL-3 (wt) or (C104S) (Fig. 3.C). Interestingly, though E-cadherin, (Ecad) expression was reduced in both FLAG-PRL-3 (wt)-expressing (0.56 ± 0.10, p< 0.0005, n=3) and, to a greater extent, FLAG-PRL-3 (C104S)-expressing (0.31 ± 0.19, p< 0.005, n=3) cells (Fig. 3.3C). This suggests that PRL-3 may promote EMT in this cell system and moreover does so in a phosphatase-independent manner. However, this is not associated with altered Csk or PTEN protein levels, or with PRL-3-dependent activation of Erk or Akt.  3.4: Tyrosine phosphorylation in response to growth factor stimulation is unaffected by PRL-3 Growth factor stimulation of cells can enhance cellular tyrosine phosphorylation and activation of signaling molecules such as Erk and Akt. As little effect of FLAG-PRL-3 was observed in cells growing in normal media and the role of PRL-3 in growth factor signaling has not been studied, I examined several different stimuli to determine whether PRL-3 would modulate signaling through these pathways.  73  Figure 3.4: FLAG-PRL-3 expression does not affect total protein tyrosine phosphorylation in IGF-I stimulated 293 cells. A. Doxycline-treated (~16 hours) and then serum-starved (24 hours) cells were incubated with 100 ng/ml of IGF-I for the times indicated and lysates probed with the anti-phosphotyrosine antibody 4G10, to determine the extent of protein tyrosine phosphorylation, and reprobed for actin as a loading control. B. Cell lysates were then probed for IGF-IRβ to determine if IGF-IR was expressed in these cells and if FLAGPRL-3 (wt) or (C104S) could modulate this expression. C. To confirm that the IGF-IRβ was being activated in this assay, lysates from doxycline-treated (~16 hours) and then serum-starved (24 hours) cells that were untreated (-) or treated for 10 minutes (+) with IGF-I were immunoprecipitated with anti-IGF-IRβ antibody, probed with 4G10 and reprobed for IGF-IRβ.  Prl-3 transcript is most highly expressed in skeletal muscle tissue, with high levels also found in cardiac tissue (Matter et al., 2001; Zeng et al., 1998). IGF-I signaling is also important  74  in these two tissues (Delafontaine, 1995; Monier et al., 1983) and can lead to activation of both Erk and Akt (Alessi et al., 1996; Webster et al., 1994). Therefore, the potential role of FLAGPRL-3 expression in IGF-1 signaling was examined in greater detail. Cells were cultured overnight (~16 hours) in the presence of 1 µg/ml of doxycycline to induce FLAG-PRL-3 (wt) and FLAG-PRL-3 (C104S) expression, then starved of serum to reduce background tyrosine phosphorylation. Cells were then stimulated by either reintroduction of serum or insulin-like growth factor (IGF)-I for 10 minutes. Expression of FLAG-PRL-3 (wt) or (C104S) did not alter the cellular protein tyrosine phosphorylation that was induced with either stimulation (Fig. 3.4A and data not shown).  The IGF-I receptor is a heterodimer  consisting of an α-subunit and a β-subunit. It is activated by binding the extracellular ligand IGF-I resulting in tyrosine phosphorylation of the β-subunit. The level of expression of the IGFI receptor β-subunit (IGF-IRβ) was examined to determine whether FLAG-PRL-3 could alter its expression. Neither FLAG-PRL-3 (wt) nor (C104S) expression affected IGF-IRβ expression (Fig. 3.4B), nor altered its tyrosine phosphorylation (Fig. 3.4C).  3.5: PRL-3 can modulate Erk but not JNK or p38 MAPK or Akt signaling in FlpIn TRex™ 293 cells upon IGF-I stimulation FLAG-PRL-3 expression did not alter total cellular protein tyrosine phosphorylation nor that of IGF-IRβ in the 293 cell stimulated with IGF-I. As mentioned above, however, IGF-I signaling activates both Erk and Akt, so experiments were performed to determine the effect of FLAG-PRL-3 on IGF-I stimulated Erk and Akt activation. Cells were treated overnight (~16 hours) with 1 µg/ml of doxycycline, serum starved for 24 hours and stimulated for 10 minutes with 100 ng/ml of IGF-I. FLAG-PRL-3 (wt) or (C104S) expression did not have a significant effect on the activity of Akt as evidenced by the lack of  75  change in the levels of phospho-Akt Ser473, indicating that PRL-3 does not play a role in IGF-Istimulated Akt signaling in 293 cells (Fig. 3.5A). In contrast, it was observed that FLAG-PRL-3 (wt) could modulate the phosphorylation of Erk in IGF-1-stimulated 293 cells (Fig. 3.5B, C). Interestingly, the cell density of the cultured cells determined the manner in which FLAG-PRL-3 expression affected Erk phosphorylation. If cells were at a lower density (~40-60% confluent) at the time of IGF-1 stimulation, FLAG-PRL3 (wt) expression resulted in a nearly two-fold increase in Erk-1/2 activity (1.87 ± 0.43 vs. 1.00 for 293 control cells, p<0.001, n=6) (Fig 3.5B). This trend was also observed for 293 cells expressing FLAG-PRL-3 (C104S), but not at statistically significant levels, suggesting that FLAG-PRL-3 phosphatase activity is partially dispensable for these effects (Fig. 3.5B). However, in cells cultured to near confluence (~80-90% confluent) at the time of IGF-1 stimulation, FLAG-PRL-3 (wt) expression led to a ~40% decrease in Erk-1/2 phosphorylation  Figure 3.5: The expression of FLAG-PRL-3 can modulate Erk, but not Akt, p38 or JNK phosphorylation in response to IGF-1. Cells were seeded at low density (A, B, D, E; 3.0x105 cells/ml) or high density (C, 7.5x105 cells/ml) and treated for at least 16 hours with 1 µg/ml of doxycycline to induce PRL-3 expression. The cells were then serum starved for 24 hours and stimulated with 100 ng/ml of IGF-I (+) for 10 minutes or left untreated (-). A. Lysates were probed with phospho-Akt Ser473 and reprobed with Akt. Neither FLAG-PRL-3 (wt) nor (C104S) expression altered Akt-Ser473 phosphorylation after IGF-1 treatment compared to 293 control cells. B. Lysates were probed for phospho-Erk and reprobed for Erk. At low cell density, FLAG-PRL-3 expression enhanced phosphorylated Erk compared to controls (* p<0.0005, n=7). C. Lysates were probed for phospho-Erk and reprobed for Erk. At high cell density, FLAG-PRL-3 (wt) expression reduced phosphorylated Erk (*p<0.001, n=6). D. The lysates were probed with phospho-p38 and reprobed for p38. p38 phosphorylation was unchanged by FLAG-PRL-3 after IGF-1 treatment. E. Lysates were probed for phospho-JNK and reprobed for JNK. JNK phosphorylation was not altered in FLAG-PRL-3-expressing cells after IGF-1 treatment. Fold induction was calculated by standardizing phosphoprotein levels to pan protein levels and subtracting the value from the untreated cells from that of the IGF-I-treated cells. PRL-3 (wt)expressing cells were compared relative to the parental 293 cells.  76  (0.59 ± 0.21 vs. 1.00 for 293 control cells, p<0.0005, n=7) (Fig. 3.5C). I examined two other MAPKs, p38 and c-jun N-terminal kinase (JNK) for alterations in their  phosphorylation  states after IGF-I stimulation of presence of FLAG-PRL-3 (wt)-expressing cells. The p38 phosphorylation did not change in response to IGF-I stimulation of 293 control serum-starved cells and this was not altered by FLAG-PRL-3 (wt) expression (Fig. 3.5D). Similarly, JNK phosphorylation did not increase after IGF-I stimulation of serum-starved 293 cells, and FLAGPRL-3 (wt) expression did not alter this (Fig. 3.5E).  77  Together, these results suggest that FLAG-PRL-3 expression can specifically affect the activation of the Erk MAPK pathway after IGF-1 stimulation in 293 cells, with the effect of PRL-3 being dependent on cell density. This function of FLAG-PRL-3 may also be partly dispensable with respect to its phosphatase activity, as there was a trend towards similarly affected Erk-1/2 phosphorylation in the FLAG-PRL-3 (C104S)-expressing 293 cells.  3.6: FLAG-PRL-3 expression enhances Erk but not Akt activation in EGFstimulated 293 cells FLAG-PRL-3 was shown to modulate IGF-I-stimulated Erk phosphorylation in a cell density-dependent manner (Fig. 3.6B, C). FLAG-PRL-3 was also observed to decrease Erk phosphorylation in response to serum stimulation in cells cultured to a high density (data not shown). As many growth factors are found in serum, I wished to determine if FLAG-PRL-3 could modulate Erk activation in response to other growth factors. As the epidermal growth factor (EGF) signaling pathway is a key cell signaling pathway with roles in proliferation, differentiation, survival and oncogenesis (Merlino, 1990), I examined the role of FLAG-PRL-3 in EGF signaling. Cells were treated with 1 µg/ml of doxycycline overnight (~16 hours), serum starved and stimulated with 50 ng/ml of EGF. Immunoblotting cell lysates for phospho-Erk demonstrated that EGF stimulated Erk phosphorylation in the 293 cells, and that this was further enhanced by PRL (wt) expression but not (C104S) expression (Fig. 3.6A). EGF stimulation did not result in appreciable Akt phosphorylation at Ser-473 in 293 cells, and this level of Akt phosphorylation was not altered by the expression of FLAG-PRL-3 (wt) or (C104S) (Fig 3.6B). This suggests that control of Erk phosphorylation by PRL-3 expression may be a common feature of several growth factor signaling pathways.  78  Figure 3.6: PRL-3 expression enhances Erk but not Akt phosphorylation in EGFstimulated cells. Cells were seeded at low density (3.0x105 cells/ml) and treated for at least 16 hours with 1µg/ml of doxycycline to induce PRL-3 expression. The cells were then serum starved for 24 hours and stimulated with 50 ng/ml of EGF for 10 minutes (+) or left untreated (-). A. Lysates were probed for phospho-Erk and Erk. FLAG-PRL-3 (wt) expression lead to increased levels of phospho-Erk compared to 293 control cells. B. Lysates were probed for phospho-Akt Ser473 and Akt. Ser473 phosphorylation of Akt was not altered by EGF-stimulation nor was it altered by FLAG-PRL-3 expression.  3.7: Mek activity is required for PRL-3-mediated enhancement of Erk activation in IGF-I stimulated cells The Erk MAPKs are part of a signaling cascade of protein kinases comprising both dualspecificity kinases and serine/threonine kinases. The MAPKKK, Raf-1, is activated by the small GTPase Ras and, in turn, activates the MAPK, Mek-1/2 (Mek). Mek is a dual-specificity threonine/tyrosine kinase responsible for the phosphorylation of threonine and tyrosine residues and activation of Erk-1/2 (Guan, 1994). Erk activity is transient and, in addition to its activation by Mek-1/2 activity, is tightly regulated by control of its phosphorylation by phosphatases,  79  including the dual-specificity phosphatase (DUSP)/MAPK phosphatase (MKP)-1 (Sun et al., 1993).  Figure 3.7: PRL-3-mediated enhancement of Erk activation after IGF-1 stimulation operates through the MAPKK Mek1/2. A. 293 (lane 1), FLAG-PRL-3 (wt) (lane 2) and FLAG-PRL-3 (C104S) (lane 3) cells were cultured in full medium for 24h in the presence of 1 µg/ml of doxycycline and lysates were probed as indicated. FLAG-PRL-3 (wt) or (C104S) expression did not alter Mek-1/2 Ser217/Ser221 phosphorylation compared to 293 control cells. B. Cells were seeded at low density (3.0x105 cells/ml) and pre-treated for at least 16 hours with 1 µg/ml of doxycycline to induce PRL-3 expression. The cells were then serum starved for 24 hours and then treated with (+) or without (-) 100 ng/ml IGF-I for 10 minutes. Lysates were probed for phospho-(Ser217/Ser221) Mek and for Mek. Mek phosphorylation is increased in FLAG-PRL-3 (wt) and, to a lesser extent, FLAG-PRL-3 (C104S) cells compared to 293 control cells. C. Cells were treated as in B, but pre-treated with the Mek inhibitor 10 µM U0126 (+) or left untreated (-) for 60 minutes before IGF-1 stimulation. Lysates were probed for phospho-Erk and Erk. D. 293 (lane 1), FLAGPRL-3 (wt)- (lane 2) and FLAG-PRL-3 (C104S)-expressing (lane 3) cells were cultured in full medium for 24h in the presence of 1 µg/ml of doxycycline and lysates were probed for Mkp-1 and actin.  I examined Mek-1/2 phosphorylation and DUSP-1/MKP-1 (MKP-1) expression in response to FLAG-PRL-3 (wt) and (C104S) expression to determine how PRL-3 might regulate  80  Erk in response to IGF-I stimulation of 293 cells. FLAG-PRL-3 (wt) expression in 293 cells resulted in an enhancement of Mek1/2 activity in IGF-1-stimulated cells but not in unstimulated cells as evidenced by an increase in Mek Ser217/Ser221 phosphorylation (Fig. 3.7A, B). A slight increase in phosphorylated Mek1/2 was observed in IGF-I-stimulated 293 cells expressing FLAG-PRL-3 (C104S) compared to 293 control cells, albeit to a lesser extent than that observed in FLAG-PRL-3 (wt) 293 cells (Fig. 3.7B). In further experiments, I verified that the IGF-Istimulated phosphorylation of Erk in the parental 293 cells could be blocked by U0126 (Fig. 3.7C) and PD98059 (data not shown), small molecule inhibitors of Mek (Favata et al., 1998; Dudley et al., 1995). IGF-1-stimulated Erk phosphorylation was also blocked in FLAG-PRL-3 (wt) and (C104S)-expressing cells pretreated with U0126 and PD98059 (Fig. 3.7C and data not shown), strongly suggesting that FLAG-PRL-3-mediated enhanced Erk activation occurs through Mek. The argument that PRL-3 acts upstream of Erk is strengthened by the finding that Mkp-1 expression is unchanged in unstimulated or IGF-1 stimulated cells (Fig. 3.7D).  3.8: Src activation is unaffected by FLAG-PRL-3 expression in IGF-I-stimulated 293 cells IGF-I signaling has been shown to activate Src family kinases (SFKs) in several systems. For example, IGF-I signaling activates SFKs and is required for Erk activity in 3T3-L1 preadipocytes ((Boney et al., 2001), colon carcinoma cells (Sekharam et al., 2003), and the pancreatic carcinoma cell line ASPC-1 (Zeng et al., 2003a). IGF-I-induced oligodendrocyte progenitor proliferation requires Src-like tyrosine kinase activity (Cui and Almazan, 2007). Src activity was also found to be regulated by PRL-3 overexpression in HEK293 cells (Liang et al., 2007). Therefore, the activation of Src in response to IGF-I stimulation in the 293 parental and  81  the FLAG-PRL-3-expressing cells 293 cells was examined (Fig. 3.7).  Cells treated with  doxycycline were stimulated with IGF-I and Src was immunoprecipitated from the lysates. Two tyrosine residues of Src, Tyr529 and Tyr418, critically regulate its activation. The kinase Csk phosphorylates Tyr529 in the C-terminal tail region of Src to inhibit Src activity (Bergman et al., 1992; Okada et al., 1991). Conversely, when Src inhibition is relieved by dephosphorylation of Tyr529, autokinase activity leads to phosphorylation of Tyr418 and full activation of Src (Boerner et al., 1996; Brown and Cooper, 1996; Kmiecik et al., 1988; Koegl et al., 1995; Moarefi et al., 1997; Weijland et al., 1996). I found that IGF-I treatment induced little or no alteration of Src phosphorylation at either Tyr529 or Tyr418 in the parental 293 cells (Fig 3.8A). This was also the case in FLAG-PRL-3 (wt) or (C104S) expressing 293 cells (Fig 3.8A). Furthermore, FLAG-PRL-3 (wt) or (C104S) expression did not alter basal or IGF-I-stimulated phosphorylation of Tyr529 and Tyr418 compared to control 293 cells. This indicates that FLAG-PRL-3 does not affect Src activation in IGF-1-stimulated 293 cells. Src has been implicated in Mek/Erk activation in several fibroblast  Figure 3.8: IGF-I and/or PRL-3 expression do not affect Src activation in 293 cells. Cells were seeded at low density (3.0x105 cells/ml) and treated for at least 16 hours with 1µg/ml of doxycycline to induce PRL-3 expression. The cells were then serum starved for 24 hours and stimulated with 100 ng/ml of IGF-1 for 10 minutes (+) or left untreated (-). The lysates were immunoprecipitated for Src and probed for phospho-Src Y529 or Y418 and for Src.  82  and v-Src transformed cell lines (Gardner et al., 1993; Macdonald et al., 1993; Troppmair et al., 1994). However, my results indicate that PRL-3 must mediate utilize some other mechanism to mediate increased Mek/Erk activation in response to IGF-I stimulation.  3.9: Summary To investigate the role of PRL-3 in cell signaling related to its reported metastasispromoting function, I generated 293 cell lines with inducible expression of FLAG-PRL-3 (wt) or (C104S) under the control of doxycycline.  Expression of FLAG-PRL-3 did not alter  proliferation of these cells but did increase the rate of cell migration to the ligand fibronectin (Fig. 3.2A, B). Interestingly, this indicates that PRL-3 catalytic activity was not required in promoting migration of 293 cells (Fig. 3.2B), an observation that was reported previously in SW480 cells (Fiordilisi et al., 2006). At the start of this study, virtually nothing was known of the cell signaling molecules and pathways regulated by PRL-3. However, during the course of this work, it was reported that overexpression of PRL-3 could reduce Csk expression and thus activate Src (Liang et al., 2007), and could downregulate PTEN expression (Wang et al., 2007a). In contrast, I found that PRL-3 expression in the 293 cells did not alter Csk or PTEN expression (Fig. 3.3C). Total cellular protein tyrosine phosphorylation was also unchanged, as was phosphorylation of Erk and Akt. PRL-3 expression reduced E-cad expression in 293 cells (Fig. 3.3C), a result consistent with observations in PRL-3-overexpressing in DLD-1 cells (Wang et al., 2007a). However, phosphatase-dead (C104S) PRL-3 expression in 293 cells also led to reduced E-cad expression (Fig. 3.3C), while PRL-3 (C104S)-expressing DLD-1 cells did not exhibit reduced E-cad levels.  83  The effect of PRL-3 overexpression on cells stimulated with growth factors has not been extensively investigated.  For example, PRL-3 overexpression blocked Ang-II-stimulated  calcium mobilization and p130Cas phosphorylation in HEK293 cells (Matter et al., 2001); however, little else is known about the role of PRL-3 in response to other signaling molecules. In IGF-I-stimulated 293 cells, PRL-3 expression can modulate Erk phosphorylation (Fig. 3.5B, C). When the 293 cells were cultured to about 60% confluence before IGF-I stimulation they displayed increased Erk phosphorylation upon IGF-1 treatment in the presence of PRL-3 (Fig. 3.5B), while cells cultured to near confluence displayed decreased Erk phosphorylation in PRL3-expressing cells in response to IGF-I stimulation (Fig. 3.5C). The function of PRL-3 in regulating the Erk response to IGF-I stimulation was mediated through the Erk kinase, Mek (Fig. 3.7A). Mek was phosphorylated in response to IGF-I stimulation and this phosphorylation was enhanced by PRL-3 expression.  Also, inhibition of Mek kinase with U0126 blocked Erk  phosphorylation resulting from IGF-I stimulation in the presence or absence of PRL-3 expression (Fig. 3.7C). Other kinases were examined for activation after IGF-I stimulation, and for regulation by PRL-3 expression. The JNK and p38 MAPKs were not phosphorylated in response to IGF-I in the 293 cells, and their phosphorylation status was not altered by PRL-3 expression. Lastly, unlike events in some other cell types (Boney et al., 2001; Sekharam et al., 2003; Zeng et al., 2003a), Src phosphorylation was not regulated by IGF-I stimulation and nor by the enhanced expression of PRL-3 in 293 cells (Fig. 3.8). Overall, these data indicate that PRL-3 promotes migration in a phosphatase-independent manner, but does not regulate Csk or PTEN expression in the Flp-In T-Rex 293 cells, unlike previously reported research (Liang et al., 2007; Wang et al., 2007a). PRL-3 does mediate the  84  response of Erk MAPK in response to growth factor signaling, specifically, IGF-I- and EGFmediated signaling pathways. This suggests that in the 293 cell system, exogenous PRL-3 is capable of modulating immediate cell responses to specific external stimuli received by quiescent/synchronized cells (after serum stimulation), but does not affect the Src or Akt pathways in cells that are proliferating/asynchronous in the presence of multiple stimuli (presence of serum). Thus, the regulation of 293 cell migration by PRL-3 must occur by some other means.  85  Chapter 4. The Role of PRL-3 Overexpression in Prostate Cancer  As there was a lack of strong phenotypes in the FLAG-PRL-3 293 cells and much of the cell signaling that had been reported during my studies was not observed in my system, I chose to generate a new expression system. Research literature released during the course of the initial research of this study suggested several new avenues of investigation. In addition to the roles reported for PRL-3 in Csk and PTEN regulation (Liang et al., 2007; Wang et al., 2007a), further evidence for a role in E-cad signaling was revealed by the report that ezrin (Ezrin/Radixin/Meosin, ERM family) is a substrate for PRL-3 (Forte et al., 2008). ERM proteins act as regulatory bridges between cell surface receptors and the actin cytoskeleton. Ezrin has been shown to interact with E-cad and β-catenin (β-cat) and play a role in cell-cell and cell-matrix adhesion in colorectal cancer cells (Hiscox and Jiang, 1999). Coupled with the findings that PRL-3 expression can decrease E-cad expression and promote EMT (Wang et al., 2007a), I designed a system that could investigate possible interactions between PRL-3 and components of cell-cell adhesion. To this end, I chose the pEGFP-C2 expression vector to express GFP-PRL-3 fusion proteins which facilitates immunofluorescence studies. PRL-3 has been shown to promote disease aggressiveness in a number of cancers, including colorectal, ovarian and gastric cancers (Bessette et al., 2008). Interestingly, however, a role for PRL-3 in prostate cancer had not been described in the available literature. Research conducted by a previous graduate student in Dr. Pallen’s laboratory revealed a link between cytoplasmic PRL-3 expression and prostate cancer aggressiveness (Fig 4.1) (Wong, 2005b). Increased cytoplasm expression of PRL-3 was correlated with more aggressive stages of prostate cancer compared to basal prostate hyperplasia (BPH; Fig. 4.1B).  I chose the low to  86  Figure 4.1: Cytoplasmic PRL-3 correlates with increased aggressiveness of prostate cancer tumours. A. A prostate tumor microarray was fixed, permeabilized and blocked with PBS-T containing 1% BSA. It was incubated with PRL-3 antibody overnight at 4°C, then HRPconjugated secondary antibody before being treated with NovaRed Substrate and hematoxylin for immunohistochemistry detection of PRL-3 (brown staining) and cell nuclei, respectively. B. Mean tumor scores from quadruplicate cores for cytoplasmic PRL-3 in basal prostate hyperplasia (BPH) or Gleason grade 3 (G3; p<0.05, n=34), grade 4 (G4; p<0.05, n=16) or grade 5 (G5; p<0.0005, n=21 vs. BPH; p<0.05, n=21 vs. G3) primary prostate cancer tumors. For scoring, 0 = no PRL-3 expression, 1 = low PRL-3 expression, 2 = moderate PRL-3 expression, 3 = high PRL-3 expression. Data was analyzed using student’s T-test (unpublished data, Wong and Pallen).  87  the low to moderately tumorigenic LNCaP prostate cancer cell line as the cell line for my GFPPRL-3 expression system to examine the role of PRL-3 in prostate cancer cell signaling. A very recent report has added weight to the prospect that PRL-3 has clinical significance in prostate cancer as it was shown that PRL-3 was expressed in primary prostate cancer with 5/53 tumours (~10%) displaying PRL-3 expression by immunohistochemistry (Wang et al., 2010). PRL-3 expression was also detected by immunoblotting in at least one tumour.  4.1: Generation of GFP-PRL-3 overexpression LNCaP cell lines In this investigation, I wished to determine the role of PRL-3 in cell signaling, in general, and in cell-cell adhesion signaling in particular. I focused on the adherens junction proteins, Ecad, β-cat and α-catenin (α-cat) and, as I wished to investigate potential PRL-3 interactions with the above mentioned proteins, chose to use a fluorescence-based expression system. PRL-3 (wt), a catalytically-inactive (C104S) and a prenylation-deficient mutant (C170S) were subcloned into the pEGFP-C2 vector to generate GFP-PRL-3 (wt), (C104S) and (C170S) expression plasmids. Prenylation is a post-translational modification that involves the incorporation of a lipid moiety into the target cysteine in found in a signature CAAX box, where C is cysteine, A is any aliphatic amino acid and X is any amino acid (Schaber et al., 1990; Schmidt et al., 1984). The C170S mutation of PRL-3 has been shown to affect the cellular localization of PRL-3 as does inhibitors of farnesyltransferase, which catalyzes the incorporation of the farnesyl isoprenoid onto cysteine residues of the CAAX box (Zeng et al., 2000). These constructs, as well as the empty vector control (henceforth referred to as GFP) were stably transfected into the LNCaP cell line. Cells were cultured in the presence of the antibiotic G418 (800 µg/ml) to select for the formation of single colonies and the resulting  88  Figure 4.2: Enrichment of high intensity GFP fluorescent cells in the GFP-PRL-3 cell lines by flow cytometry sorting. GFP cells were collected by trypsinization, passed through a 25 µm filter and resuspended in PBS containing 1% FBS and 20 µg/ml of 7-AAD. The cells were then sorted using a BD FACSAria flow cytometer and the GFP channel. A. High intensity GFP fluorescence and low intensity 7-AAD fluorescence were gated to select viable cells with high GFP transgene expression (lower right corner). Using parental untransfected LNCaP cells it was determined that only 0.7% of the cells/particles in this region are nonGFP expressing. B, C, D, E. After culturing the first set of sorted cell lines, the process was repeated as above and the cells recultured. A sample of each cell line was analyzed by regular flow cytometry to determine the percentage of cells with high intensity GFP fluorescence (B, GFP, 78%; C, GFP-PRL-3 (wt), 83%; D, GFP-PRL-3 (C104S), 87%; E, GFP-PRL-3 (C170S), 87%).  89  colonies were pooled to form a polyclonal population of LNCaP cells stably expressing GFP, GFP-PRL-3 (wt), (C104S) or (C170S) fusion proteins. It was quickly determined that there were variable levels of protein expression between the different cell lines based on the intensity and frequency of detectable GFP fluorescence (data not shown). I sought to overcome this problem by using the GFP fluorescence of the fusion proteins to sort for high GFP intensity cells which would, presumably, have higher protein expression, using a BD FACSAria flow cytometer. The cell lines were cultured in medium containing G418 (600 µg/ml), resuspended in PBS containing 1% FBS and passed through a 25 µm filter to remove cell clumps. To identify and exclude dead cells on the flow cytometer, cells were treated with 7-AAD (20 µg/ml), a fluorescent, DNA-binding molecule that cannot pass through intact membranes and thus only fluorescently mark dead cells (Schmid et al., 1992). Cells with low intensity 7-AAD fluorescence and high intensity GFP fluorescence were FACS-sorted, collected and cultured (Fig. 4.2A). After further culturing of the cells, a sample was examined by flow cytometry to confirm that the high intensity cells comprised a stable majority of the polyclonal population. This sorting step was repeated once more for a total of two sorts. This process enriched the GFP, GFP-PRL-3 (wt), (C104S) and (C170S) LNCaP cell lines (by two-fold on average for all cell lines) for high intensity GFP fluorescence to 78%, 83%, 87% or 87% of the cell populations, respectively (Fig. 4.2B, C, D, E).  4.2: GFP-PRL-3 (wt) localizes to the cell membrane in a phosphatase-independent but prenylation-dependent fashion Wild-type and mutant GFP-PRL-3 subcellular localization and protein expression levels were determined in the established LNCaP cell lines. Cells were seeded on sterile coverslips in six-well plates and cultured overnight (24 hours). After fixing and permeabilization of the cells,  90  Figure 4.3: GFP-PRL-3 localizes to the cell membrane in a prenylation-dependent but phosphatase-independent manner. A. The GFP- and GFP-PRL-3-expressing LNCaP cells were cultured on glass coverslips, fixed in methanol and mounted using DAPI-containing mounting medium. GFP and GFP-PRL-3 (C170S) are present in the cytoplasm and the nucleus with more defined nuclear localization of the latter observed (top left and bottom right panels). GFP-PRL-3 (wt) and (C104S) are present in the cytoplasm and enriched at the cell membrane (red arrows, top right and bottom left panels). B. Cells (1x106) were freshly seeded into 10cm dishes, cultured for 48 hours and the lysates probed for GFP (lane 1), GFP-PRL-3 (wt) (lane 2), GFP-PRL-3 (C104S) (lane 3) and GFP-PRL-3 (C170S) using anti-GFP antibody, and for actin as a loading control.  91  the coverslips were mounted on microscope slides with mounting medium containing DAPI to stain cell nuclei. The cells were examined under fluorescent microscopy to determine the localization of GFP or the GFP-PRL-3 (wt), (C104S) or (C170S) fusion proteins (Fig. 4.3A). GFP expression was non-localized and found throughout the cytoplasm and nucleus of the cell, although not at the cell membrane (Fig. 4.3A, top left). GFP-PRL-3 (wt) (Fig. 4.3A, top right) and phosphatase-dead GFP-PRL-3 (C104S) (Fig. 4.3A, bottom left) were localized to the cytoplasm and were enriched at the cell membrane, especially at regions of cell-cell contact (red arrows), but are excluded from the nucleus. GFP-PRL-3 (C170S) subcellular localization was similar to that of GFP. It was found in the cytoplasm and nucleus but not at the cell membrane or at other detectable membrane structures (Fig. 4.3A, bottom right). This suggests that GFPPRL-3 localization to the cell membrane requires prenylation on the first cysteine (C170) of the C-terminal CAAX box, but is not dependent on phosphatase activity. These results show that blocking prenylation of GFP-PRL-3 in the (C170S) mutant in LNCaP cells relocalizes PRL-3 distribution from cell membranes to the nucleus, reported with Myc-PRL-3 in CHO cells (Zeng et al., 2000). To examine GFP-PRL-3 protein expression levels, cells (1x106) were freshly seeded into 10-cm plates and cultured for 48 hours. Lysates were probed with GFP antibody to examine protein expression (Fig. 4.3B). Interestingly, GFP-PRL-3 (wt) was expressed to a lesser extent than GFP-PRL-3 (C104S) or (C170S). This pattern of expression was similar to that seen in the FLAG-PRL-3 (wt) 293 cells compared to the (C104S) cells. In the LNCaP GFP cell lines, the level of GFP-PRL-3/GFP protein expression follows the order: GFP-PRL-3 (C170S) > GFPPRL-3 (C104S) ≈ GFP > GFP-PRL-3 (wt). These results suggest that excessive wildtype PRL-3  92  may be somewhat toxic to at least some cell lines, including LNCaP (and 293) cells, resulting in the activation of mechanisms to reduce the load of exogenous PRL-3 protein.  4.3: GFP-PRL-3 expression decreases LNCaP cell proliferation With the establishment of the GFP-PRL-3 LNCaP cell lines, I sought to determine whether PRL-3 could promote the proliferation of these prostate cancer cells as it does in some other cancer cell lines, such as ovarian, gastric, lung and esophageal squamous cancer lines (Matsukawa et al., 2010; Ming et al., 2009; Ooki et al., 2010; Polato et al., 2005; Wang et al., 2008). Cells (2.5x103/well) were seeded in triplicate wells of 96-well plates and cultured for 24, 48, 72 or 96 hours and then assayed for proliferation. GFP-PRL-3 (wt) expression decreased cell proliferation by ~ 15% compared to GFP control cells at 24 (0.85 ± 0.018, p<0.005, n=3), 48 (0.85 ± 0.034, p<0.05, n=3) and 72 (0.88 ± 0.021, p<0.05, n=3) hours but did not significantly alter proliferation at 96 hours compared to GFP control cells (Fig 4.4A).  LNCaP cell  proliferation was unaffected by GFP-PRL-3 (C104S) expression between 24 and 72 hours, but was drastically reduced at 96 hours (0.70 ± 0.016, p< 0.01, n=3) suggesting that long-term survival in confluent culture may be reduced in these cells (Fig 4.4B). GFP-PRL-3 (C170S) expression reduced LNCaP cell proliferation at all time points (24h, 0.64 ± 0.17, p<0.00005; 48h, 0.63 ± 0.13, p<0.05; 72h, 0.58 ± 0.10, p<0.005; 96h, 0.68 ± 0.12, p<0.05; n = 3) (Fig 4.4C). These results indicate that both catalytic activity and subcellular localization of PRL-3 are important in the regulation of LNCaP cell proliferation, with plasma membrane displacement of PRL-3 leading to reduced proliferative ability of the cells. It also indicates that excessive functional PRL-3 may slightly inhibit LNCaP cell proliferation during the initial stages of cell growth to confluence, but does not exert major effects on the proliferative ability of these cells.  93  Figure 4.4: The role of GFP-PRL-3 (wt) and its mutants in regulating LNCaP cell proliferation. The GFP- and GFP-PRL-3-expressing LNCaP cells were seeded (2.5x103/well) in triplicate wells in 96-well dishes and incubated for the times indicated. MTS reagent was added to each well, incubated for 90 minutes at 37°C and the absorbance at 490 nm determined. A. GFP- (♦) and GFP-PRL-3 (wt)- (■) expressing LNCaP cells (*p<0.005, **p<0.05). B. GFP- (♦), and GFP-PRL-3 (C104S)- (■) expressing LNCaP cells (*p<0.01). C. GFP- (♦) and GFP-PRL-3 (C170S)- (■) expressing LNCaP cells (*p<0.00005, **p<0.005, ***p<0.05). The averages ± S.D. of three independent experiments are shown above.  4.4: GFP-PRL-3 colocalizes with components of the adherens junction in a phosphatase-independent but prenylation-dependent manner Cell-cell contacts are mediated, in part, by large protein complexes called adherens junctions (AJ) which link to the actin cytoskeleton. The components of these junctions include the cadherin cell membrane receptor family, which forms heterodimers with other cadherin members on adjacent cells, and β-cat and p120-catenin (p120-cat), which bind to the cadherin molecules (Meng and Takeichi, 2009). α-cat links the cadherins to the actin cytoskeleton through interactions with β-cat and F- (Drees et al., 2005; Yamada et al., 2005). In epithelial cells, the most studied cadherin is E-cad, which is also an important marker of EMT (reviewed in  94  Berx and van Roy, 2009). As PRL-3 can downregulate E-cad levels in DLD-1 colon carcinoma cells (Wang et al., 2007a) and in my FLAG-PRL-3 293 cell system (Fig. 3.3), and PRL-3 is a membrane-localized protein (Fig 4.3A), I wished to determine if GFP-PRL-3 would localize to AJs in LNCaP cells. Cells were seeded on sterile coverslips in six-well plates and cultured overnight. After fixing and blocking of the cells, they were incubated overnight at 4°C with antibodies against Ecad, β-cat or α-cat, rinsed in PBS and then incubated with Alexa594-conjugated secondary antibodies. The coverslips were mounted on microscope slides with mounting media containing DAPI for nuclei staining and examined by fluorescent microscopy. In the GFP control cells, expression of E-cad (red) was predominantly at the cell membrane, while GFP fluorescence (green) is predominantly localized to the cytosol (Fig. 4.5A, top left). The localization of E-cad was unaltered by expression of GFP-PRL-3 (wt), (C104S) or (C170S) (Fig 4.5A, top right, bottom left, bottom right, respectively). GFP-PRL-3 (wt) co-localized with E-cad at the cell membrane as evidenced by the yellow false colour image representing overlapping E-cad (red) and GFP (green) fluorescence.  Consistent with and confirming the respective plasma  membrane-enriched and nuclear/non-plasma membrane localizations of GFP-PRL-3 (C104S)  Figure 4.5: GFP-PRL-3 (wt) and (C104S) but not (C170S) can localize at the cell membrane with the adherens junction proteins E-cad, β-cat and α-cat. The GFP- and GFP-PRL-3-expressing cells were cultured on glass coverslips, fixed in methanol and blocked with 1% BSA in PBS-T. Antibodies for E-cad, β-cat or α-cat were diluted with PBS-T containing 1% BSA and incubated overnight at 4°C. After incubating the coverslips with Alexa594 fluorescent secondary antibodies, they were mounted using DAPI-containing mounting medium. A. E-cad co-localizes to the cell membrane with GFP-PRL-3 (wt) and (C104S) but not GFP or GFP-PRL-3 (C170S). B. β-cat co-localizes to the cell membrane with GFP-PRL-3 (wt) and (C104S) but not GFP or GFP-PRL-3 (C170S). C. α-cat co-localizes to the cell membrane with GFP-PRL-3 (wt) and (C104S) but not GFP or GFP-PRL-3 (C170S).  95  96  and (C170S) described above (Section 4.2), the inactive C104S mutant co-localized at the membrane with E-cad whereas C170S did not. The subcellular localization of β-cat was similar to that of E-cad. β-cat was present at the plasma membrane and its localization was unaltered by GFP-PRL-3 expression (Fig 4.5B). Both GFP-PRL-3 (wt) and (C104S) colocalized with β-cat at the membrane (Fig 4.5B, top right, bottom left) while GFP-PRL-3 (C170S) does not (Fig. 4.5B, bottom right), again indicating that prenylation is required for PRL-3 to be located at the membrane. In Wnt signaling, activation of the Wnt pathway acts to stabilize cytoplasmic β-cat and leads to its translocation to the nucleus where β-cat acts as a nuclear co-factor (Huber et al., 1996; Schneider et al., 1996). As GFPPRL-3 expression did not lead to the subcellular redistribution of β-cat in the presence of GFPPRL-3, it is likely that PRL-3 does not activate β-cat function to promote Wnt-signaling-like effects. α-cat does not bind to E-cad directly, but associates with the AJ complex through βcatenin. More specifically, it appears that monomeric α-cat associates with β-cat, while dimeric α-cat preferentially binds actin, and the control of homodimer formation of α-cat coordinates the interaction of the cytoskeleton with the AJ (Drees et al., 2005; Yamada et al., 2005). In GFP control cells, α-cat localized at the cell membrane and this was not altered by GFP-PRL-3 expression (Fig. 4.5C). GFP-PRL-3 (wt) and (C104S) colocalized with α-cat at the membrane (Fig 4.5C, top right, bottom left) while GFP-PRL-3 (C170S) did not (Fig. 4.5C, bottom right). These results also suggest that PRL-3 is unlikely to play a role in controlling the link between AJs and the actin cytoskeleton through α–cat.  97  4.5: Expression of adherens junction proteins is unaltered by GFP-PRL-3 GFP-PRL-3 (wt) and (C104S) localize with the AJ proteins at the cell membrane, however this could reflect incidental association of diffusely organized membrane-bound proteins, rather than specific protein-protein interactions. The distribution of E-cad, β-cat and αcat is not altered by the expression of GFP-PRL-3 (wt), (C104S) or (C170S) (Fig. 4.6A, B, C). There also does not appear to be a change in the expression levels of the AJ proteins when fluorescent intensities are compared, qualitatively.  However, fluorescent intensity in  immunofluorescence is not usually the best tool for determining protein expression levels. I thus determined the effect of GFP-PRL-3 upon the protein expression of E-cad, β-cat and α-cat by immunoblotting. Freshly passaged cells were seeded onto 10-cm dishes, cultured and the lysates collected. Neither β-cat nor α-catenin expression was altered by expression of GFP-PRL-3 (wt), (C104S) nor (C170S) compared to GFP expression in the LNCaP cell lines (Fig. 4.6A). E-cad expression was also unaltered by the expression of the transgene constructs in LNCaP cells (Fig. 4.6B). These results, together with the observed lack of effects of wild-type or mutant GFP-PRL-3 expression on the localization and molecular interactions of key AJ proteins, indicate that PRL-3 may not play a role in regulating E-cad, β-cat or α-cat distribution or expression in LNCaP cells. They also suggest that expression of PRL-3 is insufficient to promote EMT, as E-cad expression is unchanged in the presence of GFP-PRL-3 (wt). To determine whether GFP-PRL-3 can directly interact with the AJ proteins or affect the stability of the adherens complex, co-immunoprecipitation (co-IP) studies were conducted. IP of GFP-PRL-3 (wt), (C104S) or (C170S) using anti-PRL-3 antibody did not co-IP E-cad, β-cat or α-cat (data not shown). The reciprocal experiment, the IP of E-cad, did not IP GFP-PRL-3 (wt), (C104S) or (C170S), either (data not shown). β-cat and α-cat were co-IP’ed with E-cad in the  98  GFP-expressing LNCaP cells, and these interactions were unaltered in the GFP-PRL-3 (wt)-, (C104S)- or (C170S)-expressing cells (Fig. 4.6C, data not shown). Together, these results indicate that GFP-PRL-3 does not directly associate with the AJ proteins, nor does GFP-PRL-3 alter the AJ complex consisting of E-cad, β-cat and α-cat.  Figure 4.6: GFP-PRL-3 does not alter expression or interactions of adherens junction proteins in LNCaP cells. The GFP- and GFP-PRL-3-expressing cells (1x106) were freshly seeded into 10-cm dishes, cultured for 48 hours and the lysates probed for protein expression by immunoblotting. A. Lysates were probed with α-cat and β-cat antibodies and reprobed for actin as a loading control. B. Lysates were probed with E-cad antibodies and reprobed for actin as a loading control. GFP and GFP-PRL-3 expression was detected with GFP antibody. C. Lysates were immunoprecipitated with E-cad antibodies and the antibody-protein complexes were probed for E-cad and β-cat. The graph represents the mean ± S.D. of three independent experiments.  99  4.6: EMT is unaltered by GFP-PRL-3 expression in LNCaP cells In addition to its role in AJ, E-cad is an epithelial marker downregulated in many aggressive cancers and is associated with EMT (Berx and van Roy, 2009). As E-cad expression decreases during EMT, PRL-3 expression should decrease E-cad levels if PRL-3 promotes EMT in LNCaP cells. As seen in Fig 4.6B, however, E-cad expression is unaltered by GFP-PRL-3 expression. To further confirm that GFP-PRL-3 cannot induce EMT in the LNCaP cells, the expression of the mesenchymal marker proteins fibronectin (FN) and N-cadherin (N-cad), was  Figure 4.7: Expression of the mesenchymal markers FN and N-cad are unaffected by GFP-PRL-3. The GFP- and GFP-PRL-3-expressing cells (1x106) were freshly seeded into 10-cm dishes, cultured for 48 hours and the lysates probed with fibronectin (FN) or N-cad antibodies and reprobed with actin antibody as a loading control.  examined. The expression of these marker proteins is upregulated during EMT (Nieman et al., 1999; Thiery and Sleeman, 2006) Neither FN nor N-cad expression was altered by GFPPRL-3 in LNCaP cells (Fig. 4.7), suggesting that PRL-3 expression does not lead to EMT in LNCaP cells.  100  4.7: GFP-PRL-3 expression does not alter Csk expression or Src signaling in LNCaP cells PRL-3 overexpression downregulates Csk expression in HEK 293 cells (Liang et al., 2007) through a translational control mechanism (Liang et al., 2008). As Csk is a negative regulator of Src, PRL-3 expression also indirectly leads to Src activation. PRL-3 expression has also been shown to reduce PTEN expression (Wang et al., 2007a). As neither Csk nor PTEN were downregulated in the 293 cell system in response to PRL-3 expression, I wanted to examine whether GFP-PRL-3 could control Csk and PTEN expression in LNCaP cell; however, the  Figure 4.8: GFP-PRL-3 does not alter Csk expression or Src activity in LNCaP cells. The GFP- and GFP-PRL-3-expressing cells (106) were freshly seeded into 10-cm dishes, cultured for 48 hours and the lysates probed with Csk (FN) or phosphor-Src-Tyr418 antibodies and reprobed with actin or Src antibodies as a loading control.  101  LNCaP genome encodes a PTEN frameshift mutant which does not express the protein product (Vlietstra et al., 1998), so PTEN expression was not examined. Fresh cells were cultured for 48 hours, lysed and probed for Csk expression or for Src phosphorylated at Tyr418 (P-Src Y418), a measure of active Src. Neither Csk expression nor levels of P-Src Y418 were altered by expression of GFP-PRL-3 (wt), (C104S) or (C170S) compared to that in cells expressing GFP (Fig. 4.8). This indicates that Src activity is unaffected by overexpressed PRL-3 in LNCaP cells growing in normal media conditions.  4.8: Doxorubicin treatment of prostate and breast cancer cell lines leads to increased p53 protein expression but decreased PRL-3 protein expression Overexpression of GFP-PRL-3 did not alter Src activity, EMT or adherens junction protein interactions and expression in LNCaP cells. However, GFP-PRL-3 (wt) and (C170S) decreased the rate of LNCaP cell proliferation, while GFP-PRL-3 (C104S) led to an apparent decrease in cell survival after 96 hours incubation. Due to the limited effects of GFP-PRL-3 expression in the above studies, the regulation of endogenous PRL-3 was examined. A recent study indicated that treatment of mouse embryonic fibroblasts (MEFs) for 24 hours with the DNA-damaging agent doxorubicin (doxo) could potently upregulate p53 expression and result in an increase in PRL-3 mRNA and protein expression (Basak et al., 2008). LNCaP and MCF-7 cells express wild-type p53 (Carroll et al., 1993; Thompson et al., 1990), DU145 cells express an inactive, mutant form (Carroll et al., 1993), while PC-3 cells do not express p53 protein (Rubin et al., 1991).  To address whether doxo treatment of prostate  (LNCaP, DU-145 and PC-3) and breast (MCF-7) cancer cell lines could likewise upregulate p53 expression and lead to upregulation of PRL-3 protein, cells were treated overnight with 0.2 g/ml doxo.  102  As a positive control, MEF cells were initially examined for their ability to upregulate PRL-3 in response to doxo-treatment as reported in Basak et al (2008). Surprisingly, the “wildtype” MEFs tested did not express p53 protein, either with or without doxo treatment (Fig. 4.9A). While PRL-3 was expressed, its expression changed very little if at all in response to doxo treatment. Two wildtype MEF lines available from ATCC (Src+/+ and Fak+/+) were also tested for their response to doxo treatment. They too did not express p53 in the presence or absence of doxo (data not shown), nor did they show a change in PRL-3 expression. This suggests that the generation of immortalized MEFs may lead to different gene expression profiles that may confound the interpretation of experimental results. Doxo treatment greatly enhanced expression of p53 in both LNCaP and MCF-7 cells (11.2 ± 2.1 fold, p<0.000005, n=4; 23.6 ± 5.2 fold, p<0.00001, n=3; respectively) relative to control, untreated cells. The DU-145 cells did not exhibit a change in p53 expression in response to doxo (0.998 ± 0.091), while as expected, p53 was undetectable in the PC-3 cells (Fig. 4.9A, B). However, in contrast to the study by Basak et al. (2008), PRL-3 expression did not increase after doxotreatment, even after robust p53-induction (Fig. 4.9A). Rather, PRL-3 expression decreased in all lines tested (Fig. 4.9C), whether p53 was absent (PC-3: 0.68 ± 0.11, p<0.0005, n=3), unchanged (DU-145: 0.68 ± 0.12, p<0.0005, n=4), or increased (LNCaP: 0.74 ± 0.11, p<0.0005, n=3; MCF-7: 0.69 ± 0.11, p<0.0005, n=3). The results show that treating LNCaP and MCF-7 cells with doxo to upregulate p53 led to decreased PRL-3 protein expression rather than increased expression as reported in MEFs (Basak et al., 2008).  Expression of p21 is upregulated by both p53-dependent and –  independent mechanisms (Cox, 1997). The expression of phospho-H2AX is a marker of doublestranded breaks associated with doxorubicin, among other (Banath and Olive, 2003), and is an  103  early apoptotic marker (Rogakou et al., 2000). Protein expression of the cell cycle inhibitor p21Cip1/Waf1 (p21) was examined to determine whether downstream responses of p53 remain  Figure 4.9: PRL-3 is not upregulated in prostate or breast cancer cells in response to doxorubicin (doxo). A. Cells were incubated with (+) or without (-) 0.2 mg/ml doxo for 24 hours and probed for p53 and PRL-3 protein expression. Actin was probed as a loading control. B. After doxo treatment, p53 expression is increased in LNCaP cells and MCF-7 cells, but remains unchanged in DU-145 cells. C. PRL-3 expression is reduced in LNCaP cells, DU-145 cells, PC-3 and MCF-7 cells cells. D. The expression levels of the cell cycle inhibitor protein, p21Cip1/Waf1 (p21), and the apoptotic marker protein, phospho-H2AX (PH2AX) in response to doxo-treatment were assessed by probing with the appropriate antibodies as indicated. Results are representative of three independent experiments.  104  intact and the apoptotic marker phospho-H2AX was examined to confirm that the DNA damage caused by doxorubicin was occurring. The prostate cancer cell lines LNCaP and PC-3, and the MCF-7 cell line, displayed small increases in p21 protein in response to doxo treatment, though the DU-145 prostate cancer cell line showed decreased p21 protein expression (Fig. 4.9D). The apoptotic marker P-H2AX was strongly upregulated in all of the prostate cancer cell lines, with very weak, but detectable upregulation in the MCF-7 cell line after doxo treatment (Fig. 4.9D). The doxo-treatment of the prostate cancer and MCF-7 cells are causing DNA damage and apoptosis, as indicated by the increase in phospho-H2AX, suggesting that p53 should also be activated.  These results also indicate that transcriptional control of p21 by p53 is likely  occurring in doxo-treated cell lines, though the p21Cip1/Waf1 increase in PC-3 cells must be due to p53-independent processes.  4.9: CoCl2 treatment of prostate and breast cancer cell lines leads to increases in p53 protein but not in PRL-3 protein levels To address whether other stimuli known to increase p53 expression might result in the concomitant upregulation of PRL-3 protein, the chemical mimic of hypoxia, CoCl2, was used to treat the cell lines. CoCl2 is an inhibitor of hydroxylase enzymes, resulting in stabilization of the hypoxia-induced factor-1HIF-1), the key transcription factor in the oxidative stress response (Dachs and Stratford, 1996; Wang and Semenza, 1993). Both hypoxic and CoCl2 treatment of cells lead to increased p53 expression (Graeber et al., 1994). After overnight treatment of the cells with CoCl2, p53 expression was greatly increased in both LNCaP and MCF-7 cells (14.4 ± 3.32 fold, p<0.005, n=3; 9.81 ± 0.754 fold, p<0.00005, n=3; respectively) relative to control, untreated cells. While p53 was undetectable in the PC-3  105  cells, as expected, the DU-145 cells displayed a CoCl2-induced decrease in p53 (0.34 ± 0.054, p<0.00005, n=3) (Fig. 4.10A, B). However, in contrast to the report by Basak et al. (2008) but similar to my results with doxorubicin-treated prostate cancer cells (Fig. 4.9A, C), PRL-3 expression did not increase after CoCl2 treatment, even after robust p53 induction (figure 4.10A, C). PRL-3 expression was essentially unchanged in LNCaP cells but decreased in MCF-7 cells (0.30+/-0.11, p<0.0005, n=3) (Fig. 4.10C). PRL-3 expression was also decreased in PC-3 cells (0.73+/-0.022, p<0.00005, n=3) and showed a trend to reduced levels in DU-145 cells, albeit one that was not statistically significant (0.84+/-0.11, p=0.0592, n=3). The results clearly show that upregulating p53 expression via CoCl2-treatment in the prostate cancer and MCF-7 cell lines does not result in a corresponding upregulation of PRL-3 in these lines, and, indeed, may even lead to decreased levels of PRL-3. To confirm that DNA damage was occurring and that p53 downstream effector signaling was intact, the protein expression of P-H2AX and p21 was examined (Fig. 4.10D). The prostate cancer cell lines LNCaP and PC-3 both displayed small increases in p21 protein in response to CoCl2 treatment, though the DU-145 (0.65 ± 0.14, p<0.05, n=3) and MCF-7 (0.42 ± 0.2, p<0.01, n=3) cell lines showed decreased p21 protein expression.  The apoptotic marker P-H2AX was strongly  upregulated in the LNCaP and DU-145 cell lines, with barely detectable upregulation in the MCF-7 cell line after CoCl2 treatment (Fig. 4.10D). Phospho-H2AX was undetectable in either CoCl2 treated or untreated PC-3 cells. As with the responses to doxo treatment, these results suggest that the p53 target, p21 can be induced in these cells in response to stress, specifically, in response to CoCl2 treatment. Hypoxia increases P-H2AX in a HIF-1-independent fashion (Hammond et al., 2003) and, a recent report indicates that CoCl2 treatment of A549 lung carcinoma cells did not alter P-H2AX  106  foci formation (Schults et al., 2010). However, I observed that P-H2AX expression levels were increased by CoCl2 treatment of LNCaP and DU-145 prostate cancer cells. Interestingly, cobalt chloride treatment of cells can induce apoptosis or block other pro-apoptotic molecules, depending on the cell type (Ishino et al., 1999; Manome et al., 1999; Terasaka et al., 2000; Zou et al., 2001).  Figure 4.10: PRL-3 is not upregulated in prostate or breast cancer cells in response to cobalt chloride. A. Cells were incubated with (+) or without (-) 600 mM CoCl2 for 24 hours and probed for p53 and PRL-3 protein expression. Actin was probed as a loading control. B. p53 expression is enhanced in MCF-7 cells (*p<0.00005) and in LNCaP cells (**p<0.005), but decreased in DU-145 cells (*p<0.00005). C. PRL-3 expression is unchanged in LNCaP and DU-145 cells, but decreased in MCF-7 (*p<0.00005) and PC-3 (*p<0.0005) cells. D. The expression levels of the cell cycle inhibitor protein, p21Cip1/Waf1 (p21), and the apoptotic marker protein, phospho-H2AX (P-H2AX), after CoCl2 treatment were assessed by probing with the appropriate antibodies, as indicated.  107  4.10: Summary Previous research in Dr. Pallen’s lab indicated that increased cytoplasmic PRL-3 expression correlated with more aggressive prostate cancer tumours (Wong, 2005b). The GFPPRL-3 expression system was generated in LNCaP cells to identify the mechanisms by which this effect of PRL-3 may be mediated. This system also provided a means to determine PRL-3 subcellular localization using the fluorescence of the GFP portion of the fusion proteins. PRL-3 expression slightly decreased LNCaP proliferation and this effect was greatly enhanced in cells expressing the prenylation-deficient mutant (C170S) (Fig. 4.4A, C). PRL-3 phosphatase activity was partially dispensable for this effect on proliferation as expression of the PRL-3 (C104S) mutant only decreased cell numbers after 96 hours of proliferation (Fig. 4.4). EMT is a process which involves the loss of cell-cell contracts and promotes motility/migratory phenotypes in cells. As AJs regulate cell-cell contact and PRL-3 has been shown to downregulate expression of the AJ protein E-cad, the role of overexpressed PRL-3 in AJ signaling was investigated. GFP-PRL-3 could colocalize with E-cad, β-cat and α-cat at the cell membrane (Fig.4.5). Phosphatase activity was not required for this juxtaposition of PRL-3 and the AJs, but prenylation was required. Subcellular localization of the AJ proteins was unaffected by PRL-3 or its mutants. The expression of the AJ proteins was also unaltered by PRL-3 expression as detected in immunoblotting experiments (Fig. 4.6A, B). PRL-3 did not directly associate with the AJ proteins (data not shown), nor did PRL-3 affect the association between E-cad and β-cat or α-cat (Fig. 4.6C and data not shown). This indicates that the mechanism by which PRL-3 mediates E-cad downregulation does not occur in LNCaP cells. Further examination of the role of PRL-3 in EMT revealed that PRL-3 overexpression did not alter fibronectin nor N-cad expression, indicating that PRL-3 may not induce EMT in LNCaP  108  cells. Although a role for PRL-3 has been reported in Csk regulation and Src (Liang et al., 2007; Liang et al., 2008), PRL-3 did not affect Csk expression or Src activity in the LNCaP cells, similar to what was observed in the 293 cell system. Concurrent to this study, a report indicated that PRL-3 was a p53-responsive gene, with the addition of the DNA-damaging agent doxorubicin, which upregulates p53, resulting in increased endogenous PRL-3 expression (Basak et al., 2008). This was observed in mouse embryonic fibroblasts and led to cell cycle arrest, a result that was also observed in HT1080 fibrosarcoma cells, but not RKO colon carcinoma nor U2OS osteosarcoma cells. With the aim of investigating whether endogenous, rather than exogenous, PRL-3 had effects on prostate cancer cells, I sought conditions that would regulate PRL-3 in prostate cancer cells. In cells that express wild-type p53 (LNCaP and the MCF-7 breast carcinoma cell line), doxo treatment resulted in robust p53 expression (Fig. 4.8A, B), however, endogenous PRL-3 expression was decreased (Fig. 4.9A, C). PRL-3 expression was also decreased in DU-145 and the p53-negative PC-3 cells. Cobalt chloride treatment is also capable of upregulating p53 expression (Graeber et al., 1994), and this treatment upregulated p53 in LNCaP and MCF-7 cells (Fig. 4.10A, B). PRL3 expression was not upregulated by cobalt chloride; rather, it was decreased in MCF-7 and PC-3 cells (Fig. 4.10A, C). Together, these results that PRL-3 expression is not responsive to p53 signaling in LNCaP or MCF-7 cells. Other than a small effect on cell proliferation, exogenous PRL-3 did not regulate AJs, EMT or Csk expression in LNCaP cells. PRL-3 was also not upregulated by p53 after doxo or cobalt chloride treatment. This suggests that the mechanisms by which PRL-3 may act in prostate cancer cells may be distinct from those reported in other cell types, sensitive to PRL-3  109  dosage effects, and/or that overexpression of PRL-3 may be unable to further enhance endogenous PRL-3 actions in these cells.  110  Chapter 5: The Effect of PRL-3 Ablation on Prostate Cancer  PRL-3 overexpression produced limited effects in both the FLAG-PRL-3 293 and GFPPRL-3 LNCaP cell systems. While endogenous PRL-3 protein was undetectable in the 293 cells, it could be detected in the prostate cancer cells lines and the MCF-7 breast carcinoma (see Fig. 4.9A). The role endogenous PRL-3 plays in the GFP-PRL-3 LNCaP cell system may be sufficient to exert its effects and the addition of exogenous PRL-3 may not be capable of increasing these effects. Indeed, GFP-PRL-3 (wt) expression is less than that of the mutant constructs (C104S and C170S, Fig. 4.3B), and this could be a result of the cells trying to adjust to excessive PRL-3 protein and limiting transcription of GFP-PRL-3 (wt) from the exogenous promoter. However, the lack of effects exhibited by the catalytically-inactive and prenylationdeficient mutants of GFP-PRL-3 on signaling suggests that, at least in terms of adherens junction proteins, EMT and Src signaling, these mutants cannot affect wildtype, endogenous PRL-3 function or that PRL-3 does not play a role in these phenotypes in LNCaP cells. The effect of GFP-PRL-3 (C170S) on reducing LNCaP cell proliferation does suggest that improper localization of GFP-PRL-3 can impair some PRL-3 functions. To determine whether endogenous PRL-3 has a functional role in prostate cancer cells, transient PRL-3 siRNA treatment or stable transduction of PRL-3 shRNA in prostate cancer cell lines was conducted to reduce endogenous PRL-3 protein. The LNCaP and DU145 cells were used for both the siRNA experiments and the shRNA experiments as the two cell lines represent opposite ends of the tumourigenicity scale, with LNCaP cell exhibiting lower tumourigenicity than DU-145 cells (Ablin and D, 2007). The more malignant subline of LNCaP, called C4-2  111  cells (Wu et al., 1994) was included in the experiments that generated stable PRL-3 shRNA cell lines. The C4-2 cells were generated by two-step process.  Subcutaneous co-injection of  LNCaP and MS cells, the latter a non-tumorigenic bone-stromal cell line, in castrated adult mouse hosts resulted in the formation of prostate-specific antigen (PSA) producing tumours. The C4 subline derived from these tumours was co-injected with MS cells into castrated hosts. The C4-2 subline was derived from the resulting tumours. It was found to be more tumourigenic than LNCaP cells and could form tumours in xenograft assays in castrated hosts in an androgenindependent manner, although still exhibiting androgen responsiveness in culture vis-à-vis proliferation and transcription of androgen-responsive genes (Wu et al., 1994), (Snoek et al., 2009).  5.1: Transient ablation of PRL-3 does not affect cell proliferation but decreases long-term viability in prostate cancer cells To address whether reducing endogenous PRL-3 expression would alter prostate cancer cell signaling, siRNA treatment of LNCaP and DU-145 cells was performed. LNCaP (7.5x105) or DU-145 (3x105) cells were seeded into 6-well plates and cultured overnight. The next day (~16 hours later) PRL-3 siRNA or control, scrambled siRNA was transfected into the cells to a final concentration of 100 nM and the cells cultured for another 48 hours. The cells were lysed, RNA collected and subjected to RT-PCR in order to assess the degree of reduction of PRL-3 transcript (Fig. 5.1A). RT-PCR of the scrambled siRNA-transfected LNCaP and DU-145 cell lines demonstrated that both the full-length isoform (519 bases) and the truncated isoform (444 bases) of PRL-3 were detectable in the prostate cancer cells. Treatment of LNCaP and DU-145 cells with PRL-3-specific siRNA resulted in ~85-95% reduction of PRL-3 isoform 1. PRL-3  112  isoform 2 was reduced by about 70-80% in the LNCaP and DU-145 cells. PRL-3 protein expression was also decreased in PRL-3 siRNA-treated cells (data not shown), however, technical problems with PRL-3 antibodies prevented continued use of the antibodies to examine endogenous protein levels in the prostate cancer cells. The effect of loss of endogenous PRL-3 on LNCaP and DU-145 cell proliferation was then examined. LNCaP (2.5x103) and DU-145 (1x103) cells were seeded onto 96-well plates and incubated overnight. Scrambled or PRL-3 siRNA was then added to a final concentration of 100 nM, the cells were incubated for 24, 48, 72, 96 or 120 hours, then assessed for cell viability using MTS reagent as a measure of cell proliferation or survival. Cells that were left untreated for siRNA were examined for proliferation without further incubation (0 hour). It was observed that both LNCaP and DU-145 cells exhibited increasing cell proliferation until 72 hours of culture post-siRNA-treatment, at which point cell density in the wells reached ~100% confluence and the cells stopped proliferating.  As such, it was determined that the assay measured cell  proliferation up to 72 hours and measured the ability of cells to survive under the stress of high density (i.e. cell crowding, nutrient depletion) at 96 and 120 hours.  Figure 5.1: Ablation of PRL-3 by siRNA decreases long-term viability of prostate cancer cell lines but does not affect proliferation. Semi-quantitative RT-PCR of PRL-3 from LNCaP (lanes 1, 2) or DU-145 (lanes 3, 4) transfected with either scrambled siRNA (lanes 1, 3) or PRL-3 siRNA (lanes 2, 4) for 48 hours. To measure cell proliferation, the LNCaP (2.5x103) (B) and DU-145 (1x103) (C) cells were seeded into 96-well plates at and cultured overnight. They were then treated with 100 nM scrambled (Scr, ●) or PRL-3 (■) siRNA. At the indicated times the cells were incubated for 90 minutes with MTS and the absorbance at 490 nm was determined. The mean ± the S.D. of three independent experiments is shown. D. To measure cell survival, the LNCaP (2.5x103) and DU-145 (1x103) cells were seeded into 96-well plates at and cultured overnight. They were then treated with 100 nM scrambled (Scr, ●) or PRL-3 (■) siRNA. At the indicated times the cells were incubated for 90 minutes with MTS and the absorbance at 490nm was determined. PRL-3 siRNA treatment decreases cell survival of LNCaP and DU-145 cells cultured for 96 or 120 hours compared to  113  scrambled siRNA controls (* p<0.00005, n=6; ** p<0.005, n=6; *** p<0.00005, n=5; § p<5x10-7, n=5).  114  Neither the LNCaP cells nor the DU-145 cells showed alterations in cell proliferation after PRL-3 knockdown at early time points (Fig. 5.1B, C). However, the reduction in PRL-3 expression resulted in decreased cell viability of both the LNCaP (96h: 0.72 ± 0.10, p<0.00005, n=6; 120h: 0.57 ± 0.16, p<0.00005, n=5) and DU-145 (96h: 0.69 ± 0.12, p<0.005, n=6; 120h: 0.43 ± 0.11, p<5x10-7, n=5) cells, 96 and 120 hours post-siRNA transfection, compared to the scrambled siRNA-transfected cells (Fig. 5.1D). This suggests that PRL-3 may be involved in maintaining cell survival in the presence of stressors, such as cell crowding, nutrient deficiencies and microenvironment changes (i.e. acidification of media).  5.2: Generation of stable PRL-3 shRNA prostate cancer cell lines The initial study of the effect of reducing PRL-3 expression in prostate cancer through siRNA suggested that PRL-3 may play a role in prostate cancer cell responses to stress but not proliferation. I wished to investigate how reducing endogenous PRL-3 would affect other cancer phenotypes, such as cell migration, invasion into Matrigel and anchorage-independent growth. I also wished to determine the role of PRL-3 ablation on the expression and activation of the putative PRL-3 signaling targets, namely, Csk expression, EMT-marker protein expression, and p53 and the cell cycle proteins p21 and p27Kip1. To facilitate these experiments, prostate cancer cell lines with stable reduction of PRL-3 expression were generated. LNCaP, C4-2 and DU-145 prostate cancer cells were transduced with lentiviral vectors containing scrambled control or PRL-3 shRNA sequences to generate stable knockdown cell lines. Formation of single colonies was selected for by treatment with the antibiotic puromycin (2 µg/ml) and the colonies were pooled to form a polyclonal population of LNCaP, C4-2 or DU145 cells expressing scrambled or PRL-3 shRNA. RT-PCR was performed to examine PRL-3  115  Figure 5.2: Prostate cancer cell lines transduced with lentivirus particles containing PRL-3 shRNA have reduced PRL-3 expression. A. Semi-quantitative RT-PCR of PRL-3 from stable LNCaP (lanes 1, 2), C4-2 (lanes 3, 4) or DU-145 (lanes 5, 6) transduced with lentiviral particles containing either scrambled shRNA (lanes 1, 3, 5) or PRL-3 shRNA (lanes 2, 4, 6). B. The mean relative expression of PRL-3 in PRL-3 shRNA vs. scrambled shRNA LNCaP, C4-2 and DU-145 prostate cancer cell lines (*p<0.0005, n=3). Cells were cultured on chamber slides, fixed with 4% paraformaldehyde and incubated for 1-2 hours in PBS-T containing 1% BSA. The cells were then incubated overnight at 4°C with PRL-3 antibody, rinsed in PBS and incubated with HRP-conjugated secondary antibody. The PRL-3 protein expression was visualized with VectorRed (red-brown for PRL-3) (C, D) and counterstained with hematoxylin (D) to detect cell nuclei (purple) (C, 100X, scale bar = 50 μm; D, 400X magnification, scale bar 25 μm).  116  knockdown in the PRL-3 shRNA cell lines compared to scrambled shRNA cell lines (Fig. 5.2A) and knockdown of isoform 1 of PRL-3 was quantified (Fig. 5.2B).  I only focused on  determining the values of isoform 1 knockdown, as I used isoform 1 in my overexpression experiments. Relative to control cells, in PRL-3 shRNA cells, PRL-3 isoform 1 expression was 0.36 ± 0.12 (p<0.0005, n=3) in LNCaP cells, 0.41 ±.079 (p<0.00005) in C4-2 cells, and 0.23 ± 0.038 (p<5x10-7, n=3) in DU-145 cells. Relative to control cells, the expression of PRL-3 isoform 2 was 0.44 in LNCaP cells, 0.63 in C4-2 cells and 0.32 in DU-145 cells. Interestingly, in the DU-145 cells, isoform 1 and 2 of PRL-3 exhibited similar expression levels. In the LNCaP and C4-2 cells, isoform 2 was expressed to a much lower extent than isoform 1 (Fig. 5.2A, lanes 1, 3, 5). The functional significance of this expression is unknown, as no research has been specifically conducted to determine the functional differences of the two isoforms. It should also be noted that the PRL-3 expression in control C4-2 cells consistently appeared to be greater than in the control LNCaP cells (Fig. 5.2), suggesting that PRL-3 expression was increased by the process of generating the more tumorigenic C4-2 subline from LNCaP cells. As the available anti-PRL-3 antibodies did not perform well or consistently in immunoblotting, immunohistochemistry was utilized to determine whether PRL-3 protein expression was reduced in the PRL-3 shRNA prostate cancer cell lines. Scrambled or PRL-3 shRNA LNCaP, C4-2 or DU-145 cells were seeded into chamber slides and cultured overnight. They were fixed, blocked and incubated overnight with the same anti-PRL-3 antibody validated for immunohistochemical detection of PRL-3 in a prostate tumour microarray (Wong, 2005b). The cells were then incubated with horseradish peroxidase (HRP)-conjugated secondary antibody and stained with VectorRed staining solution.  Some experiments included  counterstaining with hematoxylin to detect nuclei.  117  At low magnification (100X), reduced staining or PRL-3 was observed fin the PRL-3 shRNA-transduced cells compared to the scrambled shRNA-transduced cells (Fig. 5.2C). This was more readily apparent at high magnification (400X) (Fig. 5.2D). PRL-3 protein expression was mostly absent in the PRL-3 shRNA LNCaP and C4-2 cells and greatly reduced in the PRL-3 shRNA DU-145 cells. Interestingly, PRL-3 expression appeared to localize to the nuclei, as well as to the cytosol and cell membrane, in the scrambled shRNA transduced DU-145 and C4-2 cells, but not the LNCaP cells. This nuclear expression is completely lost in the PRL-3 shRNA cells. The C4-2 and DU-145 cells have greater tumourigenicity than LNCaP cells and the nuclear localized endogenous PRL-3 may play a role in this effect. It is important to note, though, that nuclear localization of PRL-3 in the prostate tumour microarray did not correlate with disease progression (Wong, 2005b).  5.3: Loss of PRL-3 expression increases LNCaP cell proliferation. The effect of PRL-3 knockdown upon LNCaP, C4-2 and DU-145 cells was examined in the shRNA cell lines.  Scrambled or PRL-3 shRNA-transduced LNCaP (2.5x103), C4-2  (2.5x103), and DU-145 (1x103) prostate cancer cells were seeded onto 96-well plates and incubated for 24, 48, 72 or 96 or 120 hours. In all scrambled shRNA-transduced cell lines grown in FBS, the cells proliferated for at least 96 hours (Fig. 5.3A, C, E). The PRL-3 shRNA LNCaP cells displayed greater proliferation than the scrambled shRNA LNCaP cells at all time points (24h: 1.41 ± 0.018, p<0.005, n=3; 48h: 1.41 ± 0.040, p<0.05, n=3; 72h: 1.36 ± 0.12, p<0.01, n=3; 96h: 1.38 ± 0.093, p<0.005, n=3). Proliferation of the C4-2 and DU-145 cell lines was unaltered in the PRL-3 shRNA cells compared to the control cells (Fig. 5.3C, E). Growth of LNCaP cells are an androgen-sensitive cell line, while the C4-2 subline is androgen-insensitive.  To determine whether the presence of androgen is necessary for  118  proliferation of the control and PRL-3 knockdown cell lines, the proliferation assays for LNCaP and C4-2 cell lines were repeated using medium containing 5% charcoal-stripped fetal bovine serum (CSS), which is greatly reduced for androgens (www.invitrogen.com). Scrambled shRNA-transduced LNCaP cells displayed a short increase in proliferation up to 48 hours when grown in CSS, after which cell numbers slowly decreased (Fig. 5.3B). This was somewhat attenuated in the PRL-3 shRNA-transduced LNCaP cells (24h: 1.15 ± 0.085, p<0.05; 72h: 1.15 ± 0.053, p< 0.05; 1.26 ± 0.014, p<0.00005; n=3).  Scrambled shRNA-  transduced C4-2 cells cultured in CSS also displayed a short increase in proliferation until 48 hours, and then reached a stationary growth phase that was maintained to 96 hours (Fig. 3.5D). PRL-3 knockdown did not affect the early proliferative or initial stationary growth phases, but did result in a somewhat increased level of cell proliferation at 96 hours (1.23 ± 0.030, p<0.05, n=3). This suggests that a very slow rate of proliferation was still occurring in the PRL-3 shRNA-transduced cells.  Figure 5.3: Ablation of PRL-3 increases cell proliferation in LNCaP prostate cancer cells. LNCaP (2.5x103; A, B), C4-2 (2.5x103; C, D) or DU-145 (1x103; E) scrambled (Scr, ●) or PRL-3 (■) shRNA cells were seeded into triplicate wells of 96 well plates in either media containing full serum (FBS; A, C, E) or charcoal-stripped serum (CSS; B, D) and incubated for 24, 48, 72 or 96 hours. MTS reagent was added to each well, incubated for 90 minutes at 37°C and the absorbance at 490 nm determined. The mean ± S.D of three independent experiments is displayed. A. PRL-3 knockdown increases cell proliferation in LNCaP cells grown in FBS (* p<0.005, ** p<0.05, *** p<0.01). B. PRL-3 knockdown increases cell proliferation in LNCaP cells grown in CSS (* p< 0.05, ** p<0.00005). C. Proliferation of C4-2 cells grown in FBS is unchanged. D. PRL-3 knockdown leads to a small increase in cell survival at 96 hours in C4-2 cells grown in CSS (* p<0.05). E. DU-145 cell proliferation is unchanged by PRL-3 knockdown.  119  120  The proliferation of several cell lines has been shown to be unaffected by ablating PRL-3 expression (Polato et al., 2005; Zhou et al., 2009). This was also observed in the PRL-3 shRNAtransduced DU-145 and C4-2 cell lines grown in media containing FBS. Interestingly, the PRL3 shRNA-transduced LNCaP cells cultured in FBS displayed an approximate 40% increase in cell proliferation compared to the scrambled control cells, suggesting that PRL-3 may be antiproliferative in LNCaP cells. This is in contrast with the lack of effect of transient siRNAmediated PRL-3 knockdown on LNCaP cell proliferation (Fig. 5.1B). However, the increased proliferation of the stable shRNA-LNCaP cell line is consistent with the slightly reduced proliferation of GFP-PRL-3 (wt)-expressing LNCaP cells (Fig. 4.4A). The C4-2 cell line was reported to be androgen-independent for growth of xenografts in castrated mice hosts (Wu et al., 1994), unlike its parental line, the LNCaP cells. However, the data shown here indicate that C42 cells require androgen for prolonged proliferation as incubation of scrambled shRNA C4-2 control cells in CSS results in a stationary growth phase at 72 to 96 hours. The same cells grown in FBS, which contain androgens, do not show a retarded rate of cell proliferation during the time period tested. The reason for this is unknown, and may be due to accumulated mutations in the stock of C4-2 cells used in this study.  5.4: Loss of PRL-3 results in decreased migratory and invasive abilities of prostate cancer cells The ability to promote cell migration and, especially, invasion, is a key property of metastasis-associated genes (Kopfstein and Christofori, 2006).  PRL-3 has been shown to  promote migration and/or invasion in a number of cell lines, including CHO cells (Zeng et al., 2003b) and 293 cells (Fig. 3.2B). However, its role in promoting migration or invasion in prostate cancer cells had not been studied. The ability of the PRL-3 shRNA-transduced LNCaP,  121  C4-2 and DU-145 cells to migrate in Boyden chamber assays or to invade into Matrigel was assessed. Scrambled or PRL-3 shRNA-transduced LNCaP (5x104), C4-2 (2.5x104) and DU-145 (2.5x104) cells were seeded in serum free medium (SFM) into the top chamber of a Transwell insert and allowed to migrate for 24 or 48 hours, using full serum (10%) in the bottom chamber of the Transwell plate as a chemoattractant. After fixation and staining, cell were photographed and counted in six fields of view. PRL-3 shRNA-transduction of cells reduced migration significantly in all three prostate cancer lines at 24 and LNCaP and C4-2 cells at 48 hours (Fig. 5.4A, B). After 24 hours, PRL-3 shRNA LNCaP cell migration was 0.65 ± 0.14 (p<0.005, n=4) relative to control cells; PRL-3 shRNA C4-2 cell migration was 0.53 ± 0.16 (p<0.01, n=3) relative to control cells; and PRL-3 shRNA DU-145 cell migration was 0.58 ± 0.14 (p<0.001, n=5). After 48 hours, PRL-3 shRNA LNCaP cell migration was 0.73 ± 0.083 (p<0.05, n=4) relative to control cells; and PRL-3 shRNA C4-2 cell migration was 0.56 ± 0.21 (p<0.05, n=3) relative to control cells. These data suggest that PRL-3 may act to facilitate migration in the LNCaP, C4-2 and DU-145 cells, as partial loss (up to 75%, Fig. 5.2B) of PRL-3 expression reduced their migratory ability by up to 45%. As mentioned, invasive ability is a key property acquired by metastatic cells (Kopfstein and Christofori, 2006). As migration is reduced in the PRL-3 shRNA-transduced prostate cancer cell lines, the effect of ablating PRL-3 expression upon cell invasion into Matrigel was examined.  Matrigel is a solubilized basement membrane preparation extracted from the  Engelbreth-Holm-Swarm (EHS) mouse sarcoma, a tumor rich in extracellular matrix proteins. Its major component is laminin, followed by collagen IV, heparan sulfate proteoglycans, and entactin/nidogen (Kleinman et al., 1986; Kleinman et al., 1982). It also contains a number  122  of growth factors, including TGF-beta, epidermal growth factor, insulin-like growth factor and fibroblast growth factor and other growth factors which occur naturally in the EHS tumor (McGuire and Seeds, 1989; Vukicevic et al., 1992). Matrigel is frequently used to assess the ability of cells to invade in in vitro assays (Albini et al., 1987; Terranova et al., 1986).  Figure 5.4: PRL-3 ablation reduces cell migration of prostate cancer cells. To analyze cell migration, stable LNCaP (5x104; lanes 1, 2), C4-2 (2.5x104; lanes 3, 4) or DU-145 (2.5x104; lanes 5, 6) transduced with lentiviral particles containing either scrambled shRNA (lanes 1, 3, 5) or PRL-3 shRNA (lanes 2, 4, 6) were diluted in SFM and added to the top chamber of a Transwell insert (8.0 µm). Media containing 10% FBS was added to the wells as a chemoattractant for the cells and they were incubated for 24 (A) or 48 hours (LNCaP, C4-2 only; B). A. PRL-3 shRNA-transduced prostate cancer cells display decreased cell migration at 24 hours (* p<0.005, ** p<0.01, *** p<0.001). B. PRL-3 shRNA-transduced LNCaP and C4-2 cells exhibit decreased cell migration after 48 hours (* p<0.05). The mean total number of cells counted in six fields ± S.D calculated from three independent experiments is displayed.  Scrambled and PRL-3 shRNA-transduced LNCaP (5x104), C4-2 (1x105) and DU-145 (5x104) were seeded into the top chamber of a Transwell insert coated with Matrigel. Media  123  containing serum as a chemoattractant was added to the bottom chamber of the well and the cells incubated for 24 hours. Loss of PRL-3 decreased the invasive ability of LNCaP (0.45 ± 0.15, p<0.05, n=3), C4-2 (0.37 ± 0.24, p<0.05, n=3) and DU-145 (0.51 ± 0.12, p<0.005, n=3) cells compared to control cells (Fig. 5.5). As with the ability to promote migration, this suggests that  Figure 5.5: PRL-3 ablation reduces cell invasion of prostate cancer cells. To analyze cell invasion, stable LNCaP (1x105; columns 1, 2), C4-2 (5x104; columns 3, 4) or DU-145 (5x104; columns 5, 6) transduced with lentiviral particles containing either scrambled shRNA (columns 1, 3, 5) or PRL-3 shRNA (columns 2, 4, 6) were diluted in SFM and added to the top chamber of a Transwell insert (8.0 µm) coated with Matrigel. Media containing 10% FBS was added to the bottom wells as a chemoattractant for the cells. After 24 hours, the cells that had invaded through the Matrigel to the underside of the insert were counted. The mean ± S.D. of three independent experiements is shown. * p<0.05, ** p<0.005; n=3  PRL-3 has a role in promoting invasion in prostate cancer cells. It should also be noted that the reported increased tumourigenicity of the C4-2 cells compared to the LNCaP parental line correlated with the observed relative migration and invasion abilities of these lines  124  (scrambled control cells). In both assays, the number of C4-2 cells used was one-half of the LNCaP cells used, but the ability of the C4-2 cells to promote migration or invasion was equal to or greater than that of the LNCaP cells, respectively.  5.5: Loss of PRL-3 expression in the PRL-3 shRNA DU-145 cell lines reduces anchorage-independent growth Another in vitro assay that is conducted to examine tumourigenicity of cells is the soft agar assay, which is used to assess anchorage-independent growth. The ability of cells to grow in the absence of attachment to the extracellular matrix is a hallmark of tumorigenic cells (Hanahan and Weinberg, 2000). A soft agar assay was performed with the PRL-3 shRNA prostate cancer cell lines to determine the role of PRL-3 on anchorage-independent growth in these cells. Scrambled or PRL-3 shRNA-transduced LNCaP, C4-2 or DU-145 cells (3x103) were resuspended in media containing 0.3% agar and plated on media containing 0.6% agar in six well plates. The plates were incubated for 2 (DU-145) or 5 (LNCaP and C4-2) weeks. Minimal colony formation was observed in the LNCaP or C4-2 cell lines, whether expressing scrambled or PRL-3 shRNA (data not shown). Scrambled shRNA-transduced DU-145 cells formed small, tight colonies of ≥25 µm in diameter after two weeks (16 ± 3.3 cells/well). Loss of PRL-3 resulted in an over 50% (0.45 ± 0.15, p<0.005, n=5) decrease in the number of colonies (≥25 µm) per well (7.3 ± 2.4 cells/well).  This suggests that PRL-3 expression promotes the  anchorage-independent growth of DU-145 cells.  125  Figure 5.6: Loss of PRL-3 reduces anchorage-independent growth of DU-145 prostate cancer cells. To analyze anchorage-independent growth, 3x103 DU-145 transduced with lentiviral particles containing either scrambled shRNA or PRL-3 shRNA were resuspended in media containing 0.3% agar and plated in 6-well plates coated with 0.6% agar. The plates were incubated for 14 days, stained with 0.005% crystal violet and colonies ≥ 25 µm in diameter were counted. A. Representative wells from scrambled or PRL-3 shRNA-transduced DU-145 are shown. B. The average number of colonies per well ± S.D. from five independent experiments (p<0.005).  5.6: Loss of PRL-3 slightly decreases the expression of the cell cycle proteins p27Kip1 and p21Waf1/Cip1 in C4-2 cells In the study reporting that p53 can upregulate PRL-3, it was also reported that knockdown of PRL-3 by shiRNA led to upregulation of p53 in the absence of other treatments (Basak et al., 2008). As p53 did not upregulate PRL-3 in the prostate cancer cell lines, the effect of PRL-3 knockdown upon p53 expression was examined.  Scrambled or PRL-3 shRNA-  transduced LNCaP, C4-2 or DU-145 cells were cultured for 48 hours and the lysates examined for p53 expression. While there was a trend to decreased p53 expression in the PRL-3 shRNA-  126  transduced LNCaP and C4-2 cells compared to scrambled controls, these differences were not significant (Fig. 5.7A, B). The expression of p53 was unaltered in the DU-145 cells (Fig. 5.7A, B). Interestingly, PRL-3 expression causes cell cycle arrest in late G1 stage and ablation of PRL-3 by shRNA-treatment causes a p53-mediated cell arrest (Basak et al., 2008). Cell cycle arrest at the G1 checkpoint can be mediated through p53 regulation of p21Cip1/Waf1 (p21) expression, an inhibitor of the cyclin D/cyclin-dependent kinase (CDK)-4 complex (el-Deiry et al., 1993; Harper et al., 1993; Xiong et al., 1992). Another protein involved in mediating G1 arrest is p27Kip1 (p27) and p27 expression is also regulated by p53 (Polyak et al., 1994). Overexpression of PRL-3 in HT1080 cells leads to upregulation of both the p21 and p27 proteins and knockdown of PRL-3 can increase p21 expression in MEFs (Basak et al., 2008). The expression of p21 or p27 protein was examined to determine whether PRL-3 can regulate p21 and p27, and thus cell cycle control, in prostate cancer cells. Scrambled or PRL-3 shRNA-transduced LNCaP, C4-2 or DU-145 cells were cultured for 48 hours and the lysates examined for p21 and p27 expression. Neither p21 nor p27 protein expression was altered by loss of PRL-3 in LNCaP or DU-145 cells (Fig. 5.7A, B). However, both p27 (0.85 ± 0.069, p<0.01, n=4) and p21 (0.90 ± 0.064, p<0.05, n=4) protein expression exhibited minor but statistically significant reductions in the PRL-3 shRNA-transduced compared to the scrambled control C4-2 cells. These results indicate that PRL does not function in mediating p53 signaling in LNCaP or DU-145 cells. PRL-3 does mediate p27 and p21 expression in C4-2 cells and may be involved in cell cycle control in these cells. However, this role for PRL-3 appears to be minor, as p27 and p21 protein expression is only reduced by 15% or 10%, respectively.  127  Figure 5.7: Ablation of PRL-3 protein expression slightly decreases p27Kip1 and p21Waf1/Cip1 protein expression in C4-2 cells. A. Lysates from LNCaP (lanes 1, 2), C4-2 (lanes 3, 4) or DU-145 (lanes 5, 6) transduced with lentiviral particles containing either scrambled shRNA (lanes 1, 3, 5) or PRL-3 shRNA (lanes 2, 4, 6) were probed with antibodies as indicated. B. The relative ratio of p53, p27Kip1 or p21Waf1/Cip1 (all vs. actin) protein expression from PRL-3 knockdown cell lines compared to the control, scrambled shRNA cell lines was quantified (n=4). The mean ± S.D. of four independent experiments is shown (*p<0.01, **p<0.05).  5.7: Loss of PRL-3 has little effect on EMT-marker protein expression in prostate cancer cells PRL-3 overexpression may lead to changes in EMT marker protein expression (Fig. 3.3C) (Wang et al., 2007a).  However, GFP-PRL-3 overexpression did not affect E-cad,  fibronectin or N-cad expression in LNCaP cells (Fig. 4.6B, 4.7). As it is possible endogenous PRL-3 can promote EMT to an extent that exogenous PRL-3 cannot increase, the shRNAtransduced prostate cancer cell lines were used to determine if loss of PRL-3 altered EMT marker expression.  128  Figure 5.8: EMT protein expression is minimally affected by PRL-3 knockdown in prostate cancer cell lines. A. Lysates from LNCaP (lanes 1, 2), C4-2 (lanes 3, 4) or DU-145 (lanes 5, 6) transduced with lentiviral particles containing either scrambled shRNA (lanes 1, 3, 5) or PRL-3 shRNA (lanes 2, 4, 6) were probed with antibodies as indicated (FN = fibronectin, E-cad = E-cadherin, Cytok = cytokeratin. B. The relative ratio of FN, E-cad or Cytokeratin8/18 (all vs. actin) protein expression from PRL-3 knockdown cell lines compared to the control, scrambled shRNA cell lines was quantified (n=4). FN expression is 72% relative to control in LNCaP PRL-3 shRNA cells (inset). E-cad expression is 85% relative to control in C4-2 PRL-3 shRNA cells (inset). The mean ± S.D. of four independent experiments is shown. (* p<0.05).  Scrambled or PRL-3 shRNA-transduced LNCaP, C4-2 or DU-145 cells were cultured for 48 hours and the lysates examined for expression of the epithelial markers E-cad and cytokeratin-8/18 and the mesenchymal marker fibronectin. Interestingly, very minimal affects were observed in the levels of these proteins in the PRL-3 shRNA-transduced compared to the scrambled shRNA-transduced cell lines. Fibronectin expression is low in the LNCaP and DU145 cells but much higher in the C4-2 cells. Loss of PRL-3 in the PRL-3 shRNA LNCaP cells  129  reduced fibronectin expression (0.72 ± 0.16, p<0.05, n=4) compared to control cells. Fibronectin expression was unchanged in C4-2 and DU-145 cells (Fig. 5.8A, B).  Conversely, E-cad  expression decreased slightly (but still statistically significant) in PRL-3 shRNA-transduced C42 cells (0.85 ± 0.084, p<0.05, n=4) compared to control cells (Fig. 5.8A, B). The extremely low level of E-cad in DU-145 cells was unchanged upon PRL-3 ablation, but PRL-3 shRNAtransduced LNCaP cells exhibit a trend of reduced E-cad expression, albeit at a non-specific level (0.84 ± 0.14, p=0.058, n=4). The level of the epithelial markers cytokeratin-8 and 18 (Oshima et al., 1996) was unaltered by loss of PRL-3 in all the prostate cancer cell lines (Fig. 5.8A, B). Together, these results indicate PRL-3 expression does not play a key role in EMT in prostate cancer.  5.8: PRL-3 knockdown does not alter total cellular tyrosine protein phosphorylation but reduces phosphorylated Src Tyr529 and Erk in prostate cancer cells PRL-3 has been reported to increase total cellular protein tyrosine phosphorylation in HEK293 cells (Liang et al., 2007); however, overexpression of FLAG-PRL-3 in the 293 cell system did not change cellular protein tyrosine phosphorylation (Fig. 3.3). The increase in protein tyrosine phosphorylation may be due to the role of PRL-3 in downregulating Csk and thus, increasing Src activity (Liang et al., 2007). The role of PRL-3 in regulating total cellular protein tyrosine phosphorylation, as well as Csk expression, Src activity and Erk phosphorylation was examined. Scrambled or PRL-3 shRNA-transduced LNCaP, C4-2 or DU-145 cells were cultured for 48 hours and the lysates examined for protein tyrosine phosphorylation using the monoclonal 4G10 antibody. Lysates were also probed for expression of Csk and for phosphorylation of Src at the inhibitory Tyr529 residue and the phosphorylation of Erk.  Robust tyrosine  130  phosphorylation was exhibited by many proteins in the scrambled shRNA-transduced prostate cancer cells (Fig. 5.9). The tyrosine phosphorylation of these proteins was unaltered by the loss of PRL-3 in the PRL-3 shRNA transduced cells (Fig. 5.9). Csk expression was unaltered by loss of PRL-3 in the prostate cancer cell lines (Fig. 5.10A, B). Phosphorylation of Src Tyr529 was unaltered in the PRL-3 shRNA-transduced LNCaP and C4-2 cells compared to the scrambled control cells (Fig. 5.10A, B).  Src and its phosphorylation at Src Tyr529 were  marginally detectable in the control and PRL-3 knockout DU-145 cells (Fig. 10A). Finally,  Figure 5.9: PRL-3 knockdown does not change total cellular protein tyrosine phosphorylation in prostate cancer cell lines. Lysates from LNCaP (lanes 1, 2), C4-2 (lanes 3, 4) or DU-145 (lanes 5, 6) transduced with lentiviral particles containing either scrambled shRNA (lanes 1, 3, 5) or PRL-3 shRNA (lanes 2, 4, 6) were probed with anti-phosphotyrosine (4G10) antibody and reprobed with actin antibody.  PRL-3 shRNA-transduced prostate cells reduced Erk phosphorylation in DU-145 cells (DU-145: 0.83 ± 0.091, p<0.01, n=4) but Erk phosphorylation was extremely low in C4-2 and LNCaP  131  cells, both in control and PRL-3 shRNA-transduced cells. Together, these results indicate that PRL-3 may function to regulate Erk in DU-145 but not LNCaP or C4-2 cells but does not alter Csk expression or Src activity in any of the prostate cancer cell lines examined in this study.  Figure 5.10: Erk phosphorylation is reduced in PRL-3 knockdown prostate cancer cell lines. A. Lysates from LNCaP (lanes 1, 2), C4-2 (lanes 3, 4) or DU-145 (lanes 5, 6) transduced with lentiviral particles containing either scrambled shRNA (lanes 1, 3, 5) or PRL-3 shRNA (lanes 2, 4, 6) were probed with Csk, phospho-Src Y529, or phospho-Erk antibodies and reprobed with actin, Src or Erk antibodies, respectively. * The top band detected by the phospho-Src Tyr529 antibody does not comigrate with Src. B. The relative ratio of Csk (vs. actin), phospho-Src Y529 (vs. Src) or P-Erk-1/2 (vs. Erk-1/2) protein expression from PRL-3 knockdown cell lines compared to the control, scrambled shRNA cell lines was quantified, where phosphor-protein or protein expression was consistently detected. The mean ± S.D. of four independent experiments is shown. (* p<0.01).  5.9: Loss of PRL-3 does not alter PTEN expression or Akt Ser473 phosphorylation PRL-3 downregulates PTEN protein expression in DLD-1 colon carcinoma cells and this results in activation of Akt as evidenced by increased Akt Ser473 phosphorylation (Wang et al.,  132  2007a). PRL-3 has also been reported to transiently upregulate Akt Ser473 phosphorylation in HT1080 cells (Basak et al., 2008).  In contrast, PTEN expression and Ser473 phosphorylation  of Akt were not regulated by PRL-3 in the 293 system (Fig. 3.3). The effect of PRL-3 on PTEN expression was not examined in the GFP-PRL-3-expressing LNCaP cells, as PTEN is not expressed in LNCaP cells due to a nonsense mutation (Vlietstra et al., 1998). C4-2 cells are a subline of LNCaP cells and do not express PTEN either (Fig. 5.11A). The phosphorylation of Akt at Ser473 was examined in the PRL-3 knockdown prostate cancer cells, as was PTEN expression in the DU-145 shRNA cell lines.  Figure 5.11: Ablation of PRL-3 does not change phosphorylation of Ser473 of Akt or PTEN expression in prostate cancer cell lines. A. Lysates from LNCaP (lanes 1,2), C4-2 (lanes 3,4) or DU-145 (lanes 5,6) transduced with lentiviral particles containing either scrambled shRNA (lanes 1,3,5) or PRL-3 shRNA (lanes 2,4,6) were probed with antibodies as indicated (long exp. = longer exposure to visualize phospho-Akt in DU-145 cells). B. The relative ratio of PTEN (vs actin) and phospho-Akt S473 (vs. Akt) protein expression in PRL-3 knockdown cell lines compared to the control, scrambled shRNA cell lines was quantified (n=4).  133  Scrambled or PRL-3 shRNA-transduced LNCaP, C4-2 or DU-145 cells were cultured for 48 hours and the lysates examined for PTEN expression or phosphorylation of Akt Ser473. Akt phosphorylation at Ser473 was unaffected by loss of PRL-3 in the three prostate cancer cell lines (Fig. 5.11A, B). The PRL-3 shRNA-transduced DU-145 cells did not exhibit altered PTEN expression compared to the scrambled control cells (Fig. 5.11A, B). As expected, PTEN was not detected in the LNCaP or C4-2 cells. These results indicate that PRL-3 does not regulate PTEN or Akt in these prostate cancer cells.  5.10: Summary The GFP-PRL-3 expression system in LNCaP cells did not provide information about the potential function of PRL-3 in promoting prostate cancer aggressiveness. As endogenous PRL-3 was present in LNCaP cells, it was posited that the exogenous PRL-3 was unable to increase the effects of endogenous PRL-3 protein. It was also posited that the mutated forms of GFP-PRL-3 (C104S and C170S) did not function in a manner to enhance or reduce endogenous PRL-3 activity, except in the process of regulating cell proliferation. To that end transient siRNA and stable shRNA experiments were conducted to examine whether loss of PRL-3 would affect prostate cancer cells. Transient reduction of PRL-3 expression by siRNA was performed to assess how loss of PRL-3 would affect cell proliferation of LNCaP and DU-145 prostate cancer cell lines. Even with robust knockdown of PRL-3 expression (Fig. 5.1A), neither LNCaP nor DU-145 cells exhibited altered proliferation up to 72 hours compared to scrambled siRNA control cells (Fig. 5.1B, C). Interestingly, at 96 and 120 hours, when the cells reach confluence in the wells, both LNCaP and DU-145 cells had reduced survival due to PRL-3 ablation. Survival was reduced, compared to scrambled siRNA-treated cells, by as much as 57% in the DU-145 cells and 45% in  134  the LNCaP cells at (120 hours) (Fig. 5.1D). This indicates that PRL-3 is a pro-survival factor in response to certain cell stresses. To facilitate further investigations into the effect of PRL-3 knockdown on prostate cancer cells, lentiviral vectors that express either scrambled shRNA sequences or PRL-3 shRNA sequences were transduced into LNCaP, C4-2 or DU-145 cells. A stable reduction of 50-75% in the PRL-3 transcript level was achieved in the cells (Fig. 5.2A,B) and this mirrored by the reduced protein expression detected in the cells by immunohistochemistry (Fig. 5.2C, D). Ablation of PRL-3 led to an increase in the rate of cell proliferation in LNCaP cells grown in the presence of serum and in the ability of the LNCaP cells to survive in androgen-depleted culture conditions (Fig. 5.3A, B). PRL-3 knockdown did not alter cell proliferation in C4-2 or DU-145 cells grown in the presence of serum, though it did result in enhanced of C4-2 cell proliferation/survival in androgen-depleted culture conditions at extended culturing times (96 hours; Fig. 3.3 C, D, E). PRL-3 has been reported to increase cell migration in a number of cell systems, including CHO cells (Zeng et al., 2003b), SG7901 gastric carcinoma cells (Li et al., 2006), human myeloma cells (Fagerli et al., 2008) and Flp-In T-Rex 293 cells (Fig. 3.2). Ablation of PRL-3 reduced cell migration in the shRNA-transduced cell lines, indicating that PRL-3 promotes migration in these prostate cancer cell lines (Fig. 5.4A, B). Loss of PRL-3 also reduced invasion in Matrigel by the PRL-3 shRNA-transduced prostate cancer cell lines, indicating the PRL-3 promotes cell invasive ability (Fig 5.5). A further observation in support of a role for PRL-3 in tumorigenesis and metastasis of prostate cancer is that loss of PRL-3 from DU-145 cells reduces colony formation in assays of anchorage-independent growth ability (Fig. 5.6A, B).  135  Overexpression and loss of PRL-3 was reported to promote cell cycle arrest in MEFs through a p53-dependent mechanism (Basak et al., 2008). The role of PRL-3 in regulating the cell cycle in prostate cancer cells was investigated by examining the expression of the cell cycle inhibitor proteins p21 and p27 in the stable knockdown lines. A minor role for PRL-3 was indicated in the regulation of the expression of the cell cycle proteins p21 and p27 in C4-2 cells, but not the LNCaP or DU-145 cells. However, unlike observations in MEFs (Basak et al., 2008), PRL-3 knockdown does not alter p53 expression in prostate cancer cells (Fig. 5.7). These results, together with the finding that PRL-3 expression is not regulated by p53 expression (Fig. 4.9A, 4.10A), suggest that PRL-3 is not involved or only minimally involved in the p53 cell cycle control pathway in prostate cancer cells. EMT is an key mechanism by which tumorigenic cells gain the metastatic properties of migration and invasion. PRL-3 has been reported to promote EMT and alter EMT marker protein expression (Wang et al., 2007a). Overexpression of PRL-3 in the 293 cell system downregulated E-cad expression (Fig. 3.3C) but PRL-3 overexpression did not affect E-cad, Ncad or fibronectin expression in the GFP-PRL-3 LNCaP cell system, suggesting that PRL-3 may not have a role in EMT in LNCaP cells. The role of PRL-3 in altering EMT marker protein expression in prostate cancer cells was examined in the PRL-shRNA cell lines. The effect of PRL-3 loss on EMT marker expression was minimal. Fibronectin expression was partially reduced in PRL-3 shRNA-transduced LNCaP cells and E-cad expression was partially reduced in PRL-3 shRNA-transduced C4-2 cells, while cytokeratin 8 and 18 levels were unaffected (Fig 5.8A, B). Together, these findings indicate that the loss or gain of PRL-3 protein expression does not affect the expression of several EMT marker proteins. Interestingly, the C4-2 cells had  136  a much greater expression level of fibronectin compared to the parental LNCaP cell line (Fig. 5.8A). Overexpression of PRL-3 had shown little effect on the investigated cell signaling pathways in the FLAG-PRL-3-expressing 293 cells and GFP-PRL-3-expressing LNCaP cells, including little to no alterations in the expression or phosphorylation of molecules that PRL-3 has been reported to regulate (Liang et al., 2007; Wang et al., 2007a). To determine whether endogenous PRL-3 functions in Csk/Src or PTEN/Akt signaling, the effect of PRL- ablation on total protein tyrosine phosphorylation, Src and Erk phosphorylation, and the Akt pathway was examined in the PRL-3 shRNA-transduced prostate cancer cell lines. Loss of PRL-3 did not alter levels of protein tyrosine phosphorylation in the prostate cancer cell lines (Fig. 5.9). Csk expression was unchanged by loss of PRL-3 in the prostate cancer cell lines and, similarily, the phosphorylation of Tyr529 was unchanged in LNCaP or C4-2 cells transduced with PRL-3 shRNA compared to control cells (Fig. 5.10A, B). Phospho-Src Tyr529 was undetectable in DU145 cells and Src protein was marginally detectable. The levels of phosphorylated Erk were decreased in the PRL-3 shRNA-transduced DU-145 cells, but while Erk was expressed in LNCaP and C4-2 cells, phosphorylated Erk was almost undetectable and unquantifiable (Fig. 5.10A, B). PTEN was not expressed in LNCaP or C4-2 cells and its expression was unchanged by PRL-3 knockdown in DU-145 cells (Fig. 5.11A, B). Akt Ser473 phosphorylation was also unchanged by loss of PRL-3 in the shRNA cell lines (Fig 5.11A, B). Together, these results indicate that PRL-3 promotes Src activity in a Csk-independent manner in prostate cancer cells. The results also suggest that PRL-3 promotes Erk, but not affect Akt activity in prostate cancer cells.  137  PRL-3 expression correlates with tumour aggressiveness in prostate (Wang et al., 2010; Wong, 2005b). I have shown that PRL-3 can promote the metastasis-associated properties of migration, invasion and anchorage-independent growth in prostate cancer cells. This is not accomplished through promotion of EMT or through affecting the cell cycle, but PRL-3 does promote Erk activity in DU-145 cells. My findings support the concept that PRL-3 is a novel marker for cancer aggressiveness and represents a therapeutic target for cancer, but clearly indicate that the substrates and/or targets, and overall signaling pathway(s) by which PRL-3 promotes metastatis in prostate cancer cells remain unclear.  138  Chapter 6: Discussion  PRL-3 was reported to be a metastasis-associated protein tyrosine phosphatase (Bessette et al., 2007; Saha et al., 2001) and is associated with advanced tumour stage, metastasis and/or poor prognosis in an expanding field of human cancers. PRL-3 expression has been correlated with poor prognosis and/or metastasis in colon (Kato et al., 2004; Saha et al., 2001; Wang et al., 2007b), ovarian (Polato et al., 2005; Ren et al., 2009), breast (Hao et al., 2010; Radke et al., 2006; Wang et al., 2006; Wong, 2005b), gastric (Li et al., 2007b; Miskad et al., 2007; Wang et al., 2009b), myeloma (Fagerli et al., 2008), esophageal squamous cell (Liu et al., 2008; Ooki et al., 2010), cervical squamous cell (Ma and Li, 2010), intrahepatic cholangiocarcinoma (Xu et al., 2010) , nasopharyngeal (Zhou et al., 2009), hepatocellular (Zhao et al., 2008) and prostate carcinomas (Wang et al., 2010; Wong, 2005b). There are conflicting reports for the role of PRL3 in non-small cell lung carcinoma. An initial report indicated that loss of PRL-3 expression was associated with metastatic disease (Yamashita et al., 2007) with two subsequent reports indicating that the expression of PRL-3 correlated with advanced stage disease and metastasis, as well as with poor prognosis (Ming et al., 2009; Zhang et al., 2010). However, the initial report only examined five tumour samples, while the latter studies examined larger sets of clinical samples. The expanding clinical research indicates that PRL-3 functions as a metastasis-promoting gene in many forms of cancer. This together with the poor prognosis for survival associated with PRL-3 expression provides a strong incentive to develop tools to target PRL-3 as a marker of aggressive disease and for developing therapeutics that target PRL-3 function. To facilitate  139  development of these tools, information about the breadth of cellular functions of PRL-3 and the signaling mechanisms used by PRL-3 to mediate these functions is necessary. At the onset of this study, little was known of PRL-3 signaling and functions. A group reported that PRL-3 promoted metastasis of colon carcinoma (Saha et al., 2001) and that PRL-3 overexpression promoted migration and invasion of CHO cells (Zeng et al., 2003b). A role for PRL-3 in angiotensin signaling in HEK293 cells was also described, revealing that PRL-3 expression could block Ang-II-stimulated calcium mobilization and p130Cas phosphorylation (Matter et al., 2001). I hypothesized that PRL-3 functions in specific signaling pathways that control cell processes key to the progression and metastasis of human cancers and, as such, I sought to identify the alterations of cell signaling dependent upon PRL-3 activity. Over the course of this study, I used three independent cell systems to investigate PRL-3 signaling. I generated an inducible expression system using Flp-In T-Rex 293 cells and FLAGepitope-tagged mPRL-3, a fluorescent expression system comprised of EGFP-tagged mPRL-3 in LNCaP cells, and a stable PRL-3 knockdown system using lentivirally transduced PRL-3 shRNA in three prostate cancer cell lines. I used these gain- and loss-of-function expression systems to investigate the role of PRL-3 in signaling in HEK293 and prostate cancer cells, respectively. Briefly, these studies indicate that PRL-3 did not have a role regulating the proliferative ability of 293 cells, or the C4-2 or DU-145 prostate cancer cell lines. PRL-3 had an inhibitory effect on LNCaP proliferation. PRL-3 did promote metastasis-associated properties in 293 and prostate cancer cells. Specifically, PRL-3 promoted migration of 293 cells and migration and invasion of LNCaP, C4-2 and DU-145 cells. PRL-3 also promoted anchorage-independent growth of DU-145 cells and my preliminary data suggest that PRL-3 may also promote anchorage-independent growth of 293 cells (data not shown). The role of PRL-3 in promoting  140  migration and invasion has been a recurring theme in all cell types studied thus far. A major role was not identified for PRL-3 in Csk/Src, PTEN/Akt or EMT signaling in the 293 or prostate cancer cell lines. Erk activity was regulated by PRL-3 in IGF-I- or EGF-stimulated 293 cells.  6.1: The role of PRL-3 in controlling cell proliferation and the cell cycle PRL-3 has effects on proliferation that are dependent upon cell type. PRL-3 has been shown to promote proliferation of HEK293 cells, esophageal squamous cell carcinoma (ESCC), ovarian cancer cell lines, gastric carcinoma cells, nasopharyngeal carcinoma and the A549 lung carcinoma cell lines (Cai et al., 2008; Li et al., 2006; Liang et al., 2007; Matsukawa et al., 2010; Matter et al., 2001; Ming et al., 2009; Ooki et al., 2010; Polato et al., 2005; Wang et al., 2008; Zhou et al., 2009). Overexpression of PRL-3 increased the proliferation of HEK293 cells, and this was blocked by potassium bisperoxo (bipyridine) oxovanadate V-mediated inhibition of phosphatase activity (Matter et al., 2001). Forced expression of PRL-3 in the ESCC line TE5, which expresses low levels of endogenous PRL-3, increased the rate of cell proliferation (Ooki et al., 2009). The reciprocal experiment of siRNA-mediated knockdown of endogenous PRL-3 from the high expressor ESCC line TE11 resulted in decreased proliferation of the TE11 (Ooki et al., 2009).  Loss of PRL-3 expression through siRNA- or miRNA-mediated knockdown  decreased proliferation of several ovarian cancer cell lines expressing high PRL-3 (Polato et al., 2005), SGC7901 gastric carcinoma cells (Wang et al., 2008) and the A549 lung carcinoma cell line (Ming et al., 2009). However, PRL-3 does not affect cell proliferation of all cell types. The proliferation of the colon carcinoma cell lines DLD-1 and HCT-116 were unaffected by PRL-3 knockdown (Kato et al., 2004; Polato et al., 2005), nor did PRL-3 ablation affect cell proliferation of BL6-B16 mouse melanoma cells (Fagerli et al., 2008) or of the nasopharyngeal carcinoma cell lines 5-8F and HONE1 (Zhou et al., 2009).  141  Overexpression of PRL-3 in the 293 cell system did not alter proliferation over a period of 96 hours. The parental cells for the Flp-In T-Rex 293 system are HEK293 cells, which were also used in the study that examined the effects of PRL-3 expression in Ang-II signaling (Matter et al., 2001). In that system, PRL-3 was reported to increase cell proliferation, but proliferation was only examined over a period of 24 hours. The different outcomes in the 293 cell systems could be due to a number of factors. The clonal population selected as the overexpression system in Matter et al. may be inherently more proliferative. Even though there is a 96% amino acid identity between mouse (my studies) and human (Matter et al., 2001) PRL-3 clones (Fig. 1.5), there may be subtle differences in substrate and regulatory protein binding between the two species of PRL-3. Most notably, information in the Materials and Methods section of the Matter et al. report suggests that they used the truncated isoform 2 of PRL-3, not isoform1, as I used. This may be a source of significant experimental differences between research groups and is discussed in more detail below (section 6.4). PRL-3 overexpression had an inhibitory effect on proliferation of LNCaP cells, but loss of PRL-3 did not affect proliferation of C4-2 or DU-145 cells. Overexpression of GFP-PRL-3 in LNCaP cells slightly decreased proliferation (by ~15%), while loss of PRL-3 by shRNA(though, interestingly, not by siRNA-) mediated silencing resulted in a slightly enhanced proliferation rate of the LNCaP cells (by ~40%). Intriguingly, the prenylation-deficient mutant, GFP-PRL-3 (C170S), had a greater effect in reducing LNCaP cell proliferation than did wildtype PRL-3. The implication of this effect of the prenylation mutant is unclear, but suggests that regulation of the subcellular localization of PRL-3 is important in mediating certain PRL-3 effects, particularly the ability of PRL-3 to control cell proliferation.  Possibly, potential  substrates of PRL-3 that are involved in cell proliferation may be sequestered in domains distinct  142  from the subcellular compartments that prenylated PRL-3 localizes to.  Regulation of the  prenylation status of PRL-3 (or loss of the prenylation site in PRL-3 by mutation) may result in colocalization of PRL-3 with these potential substrates and, subsequently, PRL-3-mediated regulation of proliferation. A recent publication that supports the hypothesis, first proposed by Zeng et al. (2000), that PRL-3 may have localization-specific functions revealed that PRL-3 can interact with the nucleolar phosphoprotein nucleolin and target it for dephosphorylation (Semba et al., 2010). Loss of PRL-3 resulted in an increase of about 40% in the proliferation of LNCaP cells, supporting the finding that expression of PRL-3 limits LNCaP proliferation. Intriguingly, the expression of the catalytically-inactive PRL-3 mutant (C104S) promoted a trend, albeit a nonsignificant one, towards reduced LNCaP cell proliferation.  This suggests that GFP-PRL-3  (C104S) may not act as a loss-of-function mutant but as an altered function mutant, at least in terms of LNCaP cell proliferation. The GFP-PRL-3 (C104S) protein may bind substrates or factors required by endogenous PRL-3 for its action in the cell proliferation pathway. Alternatively, phosphatase activity may not be required or is partly dispensable for PRL-3mediated functions in LNCaP cell proliferation. Interestingly, the proliferation of the more tumourigenic and metastatic prostate cancer cell lines, C4-2 and DU-145, was not affected by PRL-3 knockdown. It is especially interesting that the proliferation of the LNCaP subline C4-2 was unaltered by loss of PRL-3, as this suggests that in the process of gaining metastatic ability, the C4-2 cells lost PRL-3-mediated control of proliferation. It is also possible that minimal PRL-3 protein expression is required to mediate prostate cancer proliferation, and the shRNA-treatment of the C4-2 and DU-145 cells was not complete enough for a difference in proliferation to be observed. This is unlikely as PRL-3  143  reduction was as great as 77% in DU-145 cells, though the PRL-3 siRNA-treatment of C4-2 cells only resulted in a 60% reduction in PRL-3 protein. Together these data indicate that PRL-3 can affect proliferation of different cell lines from the same type of carcinoma in a different manner. LNCaP cells are androgen-dependent and do not proliferate efficiently in media containing serum stripped of androgen. Ablation of PRL-3 in LNCaP cells cultured in androgendepleted medium promotes survival of the cells compared to controls. This suggests that PRL-3 misexpression may be pro-apoptotic in some cell lines under certain conditions. Interestingly, overexpression of PRL-3 in CHO cells was lethal (Zeng et al., 2003b), supporting the notion that PRL-3 may be pro-apoptotic in some cells. One method by which PRL-3 may control cell proliferation is through regulation of the cell cycle. PRL-3 overexpression upregulated p21 and p27 protein expression and caused cell cycle arrest in G1 in MEFs and HT1080 fibrosarcoma cells (Basak et al., 2008). Interestingly, loss of PRL-3 from unstressed MEFs resulted in increased p21 transcript levels as well (Basak et al., 2008). However, in HCT-116 colon carcinoma cells, forced PRL-3 expression resulted in decreased p21 and p53 reporter activity, indicating that PRL-3 functioned to reduced p21 and p53 levels or activity (Min et al., 2010). This suggests that regulation of the cell cycle inhibitor p21, and perhaps p27, may be cell type-dependent. It also suggests that upregulation of PRL-3 through a more physiological mechanism (i.e. through activity of p53) may represent a truer picture of the role of PRL-3 in the cell cycle in non-cancerous cells (i.e. MEFs). In my studies, loss of PRL-3 resulted in either no change in p21 or p27 expression (LNCaP and DU-145 cells) or a small (<15%) decrease in expression (C4-2 cells).  The  expression of p53 protein was not significantly altered in the PRL-3 shRNA prostate cancer cells, though there was a (statistically non-significant) trend for reduced p53 levels in LNCaP  144  and C4-2 cells upon PRL-3 knockdown. A preliminary examination of cell cycle protein status in the 293 system also did not reveal alterations in p53 or p21 protein expression in PRL-3 expressing cells (Fig. A.1). This suggests that PRL-3 mediated control of the cell cycle does not occur to a great extent in prostate cancer cell lines.  6.2: The role of PRL-3 in mediating EMT and the metastatic properties of migration and invasion A key phenotype exhibited by metastasis-capable cells is the epithelial-mesenchymal transition (EMT) (Boyer and Thiery, 1993). This involves the loss of cell-cell contacts and epithelial protein markers and the gain of cell motility, invasion and mesenchymal protein marker expression. Epithelial markers include E-cad, cytokeratins 8 and 18, integrin β3 and γcatenin, and some of these are downregulated in DLD-1 colon carcinoma cells expressing EGFPPRL-3 (Wang et al., 2007a). EGFP-PRL-3 expression in DLD-1 cells also upregulated the mesenchymal markers fibronectin and Snail (Wang et al., 2007a). PRL-3 overexpression resulted in reduced E-cad expression in the 293 system, but not in LNCaP cells. Examination of mesenchymal markers in the LNCaP cells overexpressing GFPPRL-3 did not reveal alterations in fibronectin or N-cadherin expression. Loss of PRL-3 in the prostate cancer cells revealed that PRL-3 has a limited role in mediating EMT in these cells. LNCaP cells stably expressing PRL-3 shRNA had reduced levels of fibronectin compared to controls, but also exhibited essentially unchanged levels of E-cad or cytokeratins 8 and 18. Conversely, PRL-3 shRNA-transduced C4-2 cells displayed slightly reduced levels of E-cad but no change in cytokeratin 8 and 18 or fibronectin levels.  However, the change in E-cad  expression was small in the C4-2 cells, only a 15% decrease compared to control cells, and this result may not be functionally important. This suggests that PRL-3 may play a partial role in  145  promoting some of the features of EMT in LNCaP cells, but has a minimal role in EMT in C4-2 cells. It should be noted that, while similar levels of E-cad and cytokeratin 8 and 18 expression are observed between LNCaP and C4-2 cells, there is much more fibronectin expressed (about 15-fold) in the C4-2 cells compared to the LNCaP cells. The increased expression of the mesenchymal marker fibronectin supports the definition of C4-2 cells as a more metastatic subline of the LNCaP cells. E-cad, cytokeratins 8 and 18, and fibronectin expression were unchanged in DU-145 cells stably expressing PRL-3 shRNA. However, it should be noted that minimal levels of fibronectin and E-cad, and low levels of cytokeratin 8, were detected in the scrambled shRNA-transduced DU-145 cells. The DU-145 cells display unstable heterogenous expression of E-cad (Rokhlin and Cohen, 1995) and, in view of the observation that PRL-3 shRNA does not alter this minimal E-cad expression, it is likely that this loss of PRL-3 expression does not relieve repression of the E-cad gene.  DU-145 cells do express cytokeratins 8 and 18, however, these keratins are  correlated with malignant phenotypes (Sherwood et al., 1990) and may not be altered by a potential EMT-promoting gene (i.e. PRL-3). A review of the literature revealed that EMT can be induced inLNCaP and DU-145 cells. EMT can be induced in LNCaP cells by HIF-1α (Luo et al., 2006), resulting in the downregulation of E-cad and the upregulation of the mesenchymal marker vimentin. The forced expression of Snail, a transcriptional repressor of E-cad (Cano et al., 2000), results in EMT in LNCaP cells (Odero-Marah et al., 2008). Also, co-treatment of LNCaP cells for over three days with TGF-β and EGF results in EMT (Odero-Marah et al., 2008) and an obscure report suggests that TGF-β alone can induce EMT in LNCaP cells (Zhu et al., 2008). To determine whether EMT could be induced in the LNCaP cells used in this study, I used cobalt chloride to induce  146  HIF-1α expression in cells cultured in the presence or absence of androgen and used TGF-β to stimulate cells.  While cobalt chloride treatment resulted in increased fibronectin, E-cad  expression was unchanged (Fig. A.2).  TGF-β-treated LNCaP cells did not alter E-cad,  cytokeratin 8 and 18, vimentin or fibronectin expression (data not shown).  EMT can be  upregulated in DU-145 cells by EGF (Gan et al., 2010) and treatment of DU-145 cells with the proteosome inhibitor NPI-0052 reverted the cells to an epithelial phenotype, inhibiting EMT (Baritaki et al., 2009). EMT was difficult to achieve in control cells and this suggests that the LNCaP cells are resistant to EMT and may not be the best model for determining the role of PRL-3 in EMT. The DU-145 and C4-2 cells have a more mesenchymal phenotype and this phenotype is unresponsive to loss of PRL-3. Loss of cell adhesion properties, especially of cell-cell contacts, is one mechanism of promoting migration and invasion. As E-cad and other EMT marker expression are for the most part unchanged in prostate cancer cells, the interaction of these proteins could be altered, thus disrupting cell adhesion. In addition to being an epithelial marker of EMT, E-cad is an important core component of AJ complexes, binding catenins and other molecules to form sites of cell-cell contacts. Investigation of the localization and interaction of the AJ proteins, β-cat and α-cat, in addition to E-cad, revealed that PRL-3 expression in LNCaP cells did not disrupt AJ complex interactions or subcellular localization.  This suggests that PRL-3 mediates migration and  invasion through a different mechanism than control of EMT or AJ proteins. As mentioned, PRL-3 expression did reduce E-cad in the 293 cell system and, interestingly, resulted in slightly decreased adhesion to fibronectin (Fig. A.3). The increased migration observed in the PRL-3expressing 293 cells may be mediated in part through the PRL-3-mediated regulation of EMT proteins, specifically, E-cad.  147  The Rho family of GTPases is a key group of proteins involved in promoting motility of cells during development and in tumourigenesis (Ridley et al., 1999). Upon activation and binding of GTP, these proteins promote formation of contractile actin-based filaments (Rho), lamellipodia and membrane ruffles (Rac) and filopodia (Cdc42). They are involved in the formation of focal adhesion complexes, multiprotein complexes that link the extracellular matrix (ECM)  with  the  actin  cytoskeleton  (Burridge  and  Chrzanowska-Wodnicka,  1996).  Overexpression of PRL-3 in SW480 colon carcinoma cells enhanced RhoA and RhoC activity but reduced Rac activity (Fiordalisi et al., 2006). Blocking PRL-3 expression using anti-PRL-3 antibodies reduced RhoA and RhoC activity in A549 lung carcinoma cells, suggesting that PRL3 promotes Rho activity in these cells (Ming et al., 2009). The role of PRL-3 in regulating Rho-family GTPases was not extensively examined in this study. A preliminary investigation of Rho family GTPase activity in the 293 cell system suggested that PRL-3 expression reduced Rac activity (Fig. A.4). However, it is unknown what effect on Rho-family signaling PRL-3 has in prostate cancer. As motility and invasion are regulated in part by PRL-3 in the LNCaP, C4-2 and DU-145 prostate cancer cells, it is possible that this is mediated by PRL-3 regulation of the Rho family. This remains a plausible line of investigation for future research of PRL-3 function in prostate cancer. As mentioned, the Rho family mediates formation of focal adhesion complexes. These complexes are initiated when the integrin heterodimer cell surface receptor binds an ECM ligand (Burridge et al., 1990; Chen et al., 1986; Stickel and Wang, 1988). Other molecules are then recruited to the complex, including vinculin (David-Pfeuty and Singer, 1980; Geiger et al., 1980), talin ((Burridge and Connell, 1983), Src (Rohrschneider, 1980) and Focal Adhesion Kinase (FAK) (Schaller et al., 1992; Zachary and Rozengurt, 1992).  FAK activation is  148  dependent upon integrin attachment to the ECM and colocalizes with integrin (Lipfert et al., 1992). The interaction of FAK with integrin is mediated through direct interaction with talin (Chen et al., 1995) which can also bind integrin directly (Horwitz et al., 1986). Src can bind FAK (Cobb et al., 1994a) and is recruited to the focal adhesion complex by FAK association upon integrin engagement (Schlaepfer et al., 1994).  The formation of short-term focal  complexes and long-term focal adhesions mediates cell locomotion (Geiger et al., 1984). Src can mediate cell migration in part through modulating FAK activity (Sankar et al., 1995). This represents another potential mechanism through which PRL-3 may mediate cell migration. As PRL-3 can upregulate Src activity by downregulating Csk (Liang et al., 2007), this suggests another mechanism by which PRL-3 may mediate cell migration and invasion in some cell types. Indeed, loss of PRL-3 in the PRL-3 shRNA prostate cancer cells resulted in a slight enhancement of phosphorylation of the inhibitory Tyr529 residue of Src, suggesting PRL-3 may activate Src in prostate cancer cells. However, the effect of PRL-3 upon other components of the focal adhesion complex was not examined in this system. PRL-3 overexpression in LNCaP cells did not appear to alter integrin β1 or α1 expression (data not shown).  Also, PRL-3  overexpression in the 293 cell system did not alter Csk expression or Src activity. Interestingly, a group using an independently generated inducible-expression system with the Flp-In T-Rex 293 cells reported that PRL-3 could associate with integrin α1 and decrease tyrosine phosphorylation of integrin β1 (Peng et al., 2006), and ablation of integrin β1 abolished PRL-3 induced migration of LoVo colon carcinoma cells (Peng et al., 2009). These results suggest that PRL-3 may promote migration in prostate cancer and colon carcinoma cells through regulation of the formation of focal adhesion complexes but this does not occur in all cell types.  149  Invasion is a key property that confers malignancy to cells and enables metastasis to occur. In addition to the ability to migrate, cell invasion into new tissues requires the ability to degrade the ECM to allow cellular locomotion through the substratum (Duffy, 1987). The matrix metalloproteinases (MMPs) have been established as key proteases necessary for the degradation of the ECM in the tumour environment to allow invasion (Stetler-Stevenson et al., 1993).  This suggests that PRL-3 could promote MMP expression or activity in cells.  Overexpression of PRL-3 in the 293 cell system did not alter MMP expression in these (Ng and Pallen, 2007); however, invasive ability was not examined in these cells, although PRL-3 expression did induce a large increase in cell migration to fibronectin in a phosphataseindependent manner. It would be useful to determine whether MMP activity or expression is altered in prostate cancer cells in response to PRL-3 overexpression or ablation. As invasion decreased 50-60% in the prostate cancer cells, it is possible that MMP activity is decreased in these cells. A report supporting this hypothesis indicates that PRL-3 expression promotes MMP2 activity, and PRL-3-induced invasion is blocked by inhibitors of MMP activity (Peng et al., 2009). A hallmark of EMT is the loss of cell-cell and cell-basement membrane adhesion. Interestingly, I observed that PRL-3 overexpressing cells appeared to have a somewhat decreased adhesive ability when plated onto tissue culture dishes, both in the course of normal passaging and in the set-up of various experiments (data not shown). Indeed, induction of wildtype PRL-3 decreased the ability of 293 cells to adhere to fibronectin at 60 minutes (see Appendix A.3), though not to bovine serum albumin, poly-L-lysine or laminin (data not shown). This phenomenon was not investigated further because only a minor change in adhesion was quantified (~15-20% reduction upon PRL-3 expression). Howewver, this is in accord with the  150  ability of PRL-3 to promote migration to fibronectin. Future research into the role of PRL-3 in modulating focal adhesions, Rho GTPase activity and migration and invasion is required to fully understand the role of PRL-3 in cellular adhesion and EMT.  6.3: The role of PRL-3 in cell signaling pathways As this study progressed, reports describing functions for PRL-3 in several potential signaling pathways emerged. PRL-3 was found to regulate the Src pathway by downregulation of Csk protein (Liang et al., 2007; Liang et al., 2008), the Akt pathway by downregulation of PTEN (Wang et al., 2007a) and an integrin β-1/Erk pathway (Peng et al., 2009). PRL-3 has been reported to interact directly with the membrane cytoskeleton linker protein Ezrin and dephosphorylate it at Thr567 (Forte et al., 2008), to bind the cadherin family member CDH22 (Liu et al., 2009), and to bind the nucleolar-specific phosphoprotein nucleolin and dephosphorylate it at Thr76 and Thr84 residues (Semba et al., 2010). A report has also shown that overexpressed PRL-3 can co-immunoprecipitate with p38 MAPK and results in p38 desphosphorylation and decreased LITAF (lipopolysaccharide-induced TNF-α factor) expression (Tang et al., 2009). PRL-3 overexpression did not alter Csk expression or phosphorylation of Src at Tyr529 in the 293 cell system or in LNCaP cells. Attenuation of PRL-3 expression by shRNA-treatment did not alter Csk or Src phosphorylation in the prostate cancer cell lines. This suggests that PRL3 does not mediate Src activity in 293 or prostate cancer cells.  As my studies did not  demonstrate effects of PRL-3 upon Src signaling similar to those reported by others, it remains to be seen how common a signaling mechanism is the PRL-3 regulation of Src activity through Csk.  151  PRL-3 has been shown to upregulate Akt activity in a number of contexts. PRL-3 expression reduces PTEN levels in DLD-1 cells, and the loss of this inhibitor of PI3-K leads to increased activity of the downstream kinase Akt (Wang et al., 2007a). Overexpression of PRL-3 in cells leads to a transient activation of Akt activity through an undetermined (Basak et al., 2008). However, I did not observe an alteration of Akt activity in any of the systems I examined. PTEN expression did not change in response to PRL-3 overexpression in the 293 cell system or upon loss of PRL-3 in the PRL-3 shRNA DU-145 cells. This indicates that the activation of Akt by PRL-3 is likely cell type-specific. Cursory examination of several other signaling molecules has not revealed other roles for PRL-3 in the cell systems. For example, PRL-3 does not regulate Ezrin protein expression in LNCaP cells; however, phosphorylation of Ezrin at Thr567 was undetectable, so the purported role of PRL-3 in mediating this dephosphorylation was unresolved. It is possible that the levels of endogenous PRL-3 are sufficient to mediate complete dephosphorylation of Ezrin, at least to undetectable levels. Also, FAK phosphorylation and activation in 293 or LNCaP cells was unaffected by exogenous PRL-3 (data not shown). This finding was in agreement with another study where FAK was not observed to be phosphorylated in HEK293 cells in response to PRL-3 overexpression (Liang et al., 2007). PRL-3 expression in the 293 cell system did modulate Erk phosphorylation in response to growth factor and other stimuli after cells were serum-starved. In cells treated with IGF-I or EGF, PRL-3 expression enhanced Erk phosphorylation by about 2-fold or 3-fold, respectively. However, this enhancement was dependent upon cell culturing conditions, at least in IGF-Istimulated cells. In IGF-1-stimulated cells at ~50% confluency, PRL-3 expression enhanced Erk phosphorylation, whereas IGF-I stimulation of PRL-3-expressing cells cultured to 80-90%  152  effected a reduction of Erk phosphorylation. Reduced Erk phosphorylation was also observed in near-confluent PRL-3-expressing cells stimulated with serum (Fig. A.5). The regulation of MAPK phosphorylation by PRL-3 in response to growth factor stimulation was specific to Erk, as neither p38 nor JNK phosphorylation was affected in IGF-I-stimulated cells. As neither p38 nor JNK responded strongly to IGF-I stimulation of the 293 cells, and both p38 and JNK are stress-responsive MAPKs (Galcheva-Gargova et al., 1994; Han et al., 1994), the role of PRL-3 in modulating p38 or JNK phosphorylation in response to hyperosmolarity was examined. PRL-3 expression was unable to alter the response of either p38 or JNK phosphorylation to osmotic shock (data not shown). The Akt pathway is strongly activated in IGF-I-stimulated (Alessi et al., 1996), however, the phosphorylation of Ser473, an indicator of Akt activity, was unaffected by PRL-3 expression in 293 cells. My data suggest that, in the pathways I examined, PRL-3 has only minor roles in signaling in cells proliferating in media containing a multitude of growth stimuli and co-factors (i.e. cells growing in serum). However, in cells serum-starved to decrease signaling and to synchronize cells, PRL-3 expression could enhance or reduce Erk phosphorylation in response to growth factor stimuli depending on the cell density at the time of stimulation. Thus, a key finding from investigating the role of PRL-3 in growth factor signaling in 293 cells is the dependence of cellular responses upon confluency of the cells. Future research should remain mindful of this cell density-dependent trait of PRL-3 signaling. Investigating the role of PRL-3 in growth factor signaling may also prove fruitful in future investigations.  6.4: The role of PRL-3 isoform 1 vs. isoform 2 – which is the active isoform? The  Prl-3  locus  produces  two  alternatively  spliced  isoforms  of  PRL-3  (www.ncbi.nlm.nih.gov). Isoform 1 of PRL-3 encodes a full-length cDNA that produces a  153  polypeptide of 173 amino acids. Isoform 2 of PRL-3 encodes a truncated cDNA that produces a polypeptide of 148 amino acids (Fig. 1.4). Isoform 2 retains the major phosphatase catalytic domain features, including the acidic loop (residues 68-75) and the PTP signature motif (residues 102-110) (Fig. 1.4). However, 25 residues distal to the PTP signature motif (residues 111-135) are absent in isoform 2 (Fig. 1.4). Structural analyses demonstrate that there are two α-helix domains contained within these residues which are structurally conserved among the PRL-3related dual-specificity PTPs PTEN, MKP-3, Cdc14 and the VHR sub-family phosphatases (Kim et al., 2004; Kozlov et al., 2004). It should be noted that both the murine and human PRL-3 loci may produce both isoforms. The truncation of 25 amino acids is a significant fraction (~14.5%) of the full-length protein and likely leads to changes in the regulation and signaling of PRL-3. PRL-1, a highly related homolog of PRL-3, was found to exist in trimer configuration, both in crystallization studies and in vivo studies (Jeong et al., 2005). As PRL-3 has 76% amino acid identity with PRL-1, it is likely PRL-3 can associate as a homotrimer and, possibly, as a heterotrimer with either PRL-1 and/or PRL-2. Interestingly, one face of the PRL-1 molecule that is involved in trimer formation (Jeong et al., 2005) is 80% absent (4 of 5 interacting residues) in isoform 2 of PRL-3. This suggests that potential trimer formation of PRL-3 is disrupted in the truncated isoform (Fig. 6.1)  Figure 6.1: Schematic of hPRL-3 domains in the full-length isoform vs. the truncated isoform. Isoform 1 (full-length) and isoform 2 (truncated) forms of PRL-3 are schematically diagrammed illustrating some of the major domains contained within the protein. The solid black line in isoform 2 represents the deleted amino acids. Note that most of the most C-terminal trimer interaction domain is absent from isoform 2 and that the deletion begins immediately distal to the P-loop Arg.  154  Little, if any, of the published research on PRL-3 has been confirmed by investigations from other independent research groups. In fact, in my own studies, few of the published effects of PRL-3 upon cellular signaling were confirmed in the systems I used to investigate PRL-3 signaling. The phenotypic properties that PRL-3 imparts are recurrently reported in the literature (i.e. migration, invasion, and tumour-promoting effects), suggesting that PRL-3 is an important regulator of these metastasis-associated phenomena. However, the lack of reproducibility of the signaling mechanisms described for PRL-3 suggests that the expression/knockdown systems and cell lines used in the literature produce specific effects that other expression/knockdown systems and/or cell lines may not. The observation that investigators have used either the full-length isoform 1 or the truncated isoform 2 of PRL-3 in their expression systems suggests that both isoforms have independent (and perhaps overlapping) functions in cell signaling.  155  Evidence to support the hypothesis that both isoforms 1 and 2 have functional effects in cells is two-fold. First, I observed that there was significant PRL-3 isoform 2 mRNA present in LNCaP, C4-2 and DU-145 cells. These mRNAs were reduced in response to PRL-3 siRNA or shRNA to levels similar to that of PRL-3 isoform 1 (Fig. 5.2). The effects observed in the PRL3 knockdown prostate cancer cells (and by extension, published observations using PRL-3 knockdown as an experimental model system) could be due to reduced PRL-3 isoform 1, isoform 2 or a combination of both. The second line of evidence that suggests both PRL-3 isoforms have functional responses is the observation that overexpression of PRL-3 isoform 2 in THP-1 macrophages reduced p38 phosphorylation and LITAF activity in response to LPS stimulation (Tang et al., 2009). Endogenous PRL-3 is expressed in primary macrophages, suggesting that the role of PRL-3 in LPS signaling is important in vivo. Interestingly, it was found that a peptide encoding PRL-3 residues 39-66 could mimic PRL-3 isoform 2 functions in response to LPSstimulation of macrophages (Tang et al., 2009). This region is found in both isoforms, and may be important in signaling from both forms of PRL-3. The observed results with overexpressed PRL-3 isoform 2 belies the suggestion that isoform 2 is inactive and unstable, as previously reported (Kozlov et al., 2004) or suggests that it is indeed intrinsically inactive but exerts cellular effects by disrupting endogenous PRL-3 trimers of isomer 1. Several lines of evidence suggest that PRL-3 regulation may be complex and intricate. Two isoforms are expressed and both appear functional in overexpression studies. There is potential for regulation of PRL-3 through trimerization, mediated by regions that are missing in isoform 2 (Jeong et al., 2005). In addition to being a site of prenylation, the C-terminal CAAX box appears to regulate the catalytic activity of PRL-3 (Pascaru et al., 2009). Further research is necessary to determine the activity and signaling functions of the PRL-3 isoforms in finer detail,  156  as well as to elucidate the roles of the different domains of PRL-3 in regulating its function. Full understanding of the roles of the PRL-3 isoforms would reveal the PRL-3 functions which are due to general PRL-3 signaling effects and those which are dependent upon particular domains of PRL-3. Hopefully, this would explain some of the ambiguity and non-reproducible results in the prevailing literature.  6.5: Conclusions In summary, PRL-3 promotes migration of 293 and prostate cancer cell lines, and promotes invasion of LNCaP, C4-2 and DU-145 prostate carcinoma cells. It has minimal effects on cell proliferation in 293, C4-2 and DU-145 cells, but may act to inhibit LNCaP cell proliferation. Endogenous PRL-3 expression promotes anchorage-independent growth of DU145 cells as well. While PRL-3 expression can modulate Erk signaling in response to IGF-I or EGF stimulation and downregulate E-cad in 293 cells, any alterations in signaling pathways examined in the prostate cancer cell lines were minor, if present at all. Together these results indicate that PRL-3 may be involved in the regulation of growth factor signaling in normal cells and confers metastatic properties upon prostate cancer cells. This suggests that PRL-3 is a valuable candidate as a marker for prostate carcinoma prognosis and a potential therapeutic target to treat prostate cancer. However, a far more complete understanding of the cellular regulation and actions of PRL-3 are essential to evaluate its utility as a therapeutic target.  157  References  Ablin, R. J., and D, M. M. (2007). Metastasis of prostate cancer, Vol 10). Ahn, J. H., Kim, S. J., Park, W. S., Cho, S. Y., Ha, J. D., Kim, S. S., Kang, S. K., Jeong, D. G., Jung, S. K., Lee, S. H., et al. (2006). Synthesis and biological evaluation of rhodanine derivatives as PRL-3 inhibitors. Bioorg Med Chem Lett 16, 2996-2999. Ahn, N. G., Seger, R., Bratlien, R. L., and Krebs, E. G. (1992). Growth factor-stimulated phosphorylation cascades: activation of growth factor-stimulated MAP kinase. Ciba Found Symp 164, 113-126; discussion 126-131. Albini, A., Iwamoto, Y., Kleinman, H. K., Martin, G. R., Aaronson, S. A., Kozlowski, J. M., and McEwan, R. N. (1987). A rapid in vitro assay for quantitating the invasive potential of tumor cells. Cancer Res 47, 3239-3245. Alessi, D. R., Andjelkovic, M., Caudwell, B., Cron, P., Morrice, N., Cohen, P., and Hemmings, B. A. (1996). Mechanism of activation of protein kinase B by insulin and IGF-1. EMBO J 15, 6541-6551. Alonso, A., Sasin, J., Bottini, N., Friedberg, I., Friedberg, I., Osterman, A., Godzik, A., Hunter, T., Dixon, J., and Mustelin, T. (2004). Protein tyrosine phosphatases in the human genome. Cell 117, 699-711. Altomare, D. A., and Testa, J. R. (2005). Perturbations of the AKT signaling pathway in human cancer. Oncogene 24, 7455-7464. Andersen, J. N., Mortensen, O. H., Peters, G. H., Drake, P. G., Iversen, L. F., Olsen, O. H., Jansen, P. G., Andersen, H. S., Tonks, N. K., and Moller, N. P. (2001). Structural and evolutionary relationships among protein tyrosine phosphatase domains. Mol Cell Biol 21, 7117-7136. Banath, J. P., and Olive, P. L. (2003). Expression of phosphorylated histone H2AX as a surrogate of cell killing by drugs that create DNA double-strand breaks. Cancer Res 63, 4347-4350. Barbera, M. J., Puig, I., Dominguez, D., Julien-Grille, S., Guaita-Esteruelas, S., Peiro, S., Baulida, J., Franci, C., Dedhar, S., Larue, L., and Garcia de Herreros, A. (2004). Regulation of Snail transcription during epithelial to mesenchymal transition of tumor cells. Oncogene 23, 7345-7354. Bardelli, A., Saha, S., Sager, J. A., Romans, K. E., Xin, B., Markowitz, S. D., Lengauer, C., Velculescu, V. E., Kinzler, K. W., and Vogelstein, B. (2003). PRL-3 expression in metastatic cancers. Clin Cancer Res 9, 5607-5615. Barford, D., Das, A. K., and Egloff, M. P. (1998). The structure and mechanism of protein phosphatases: insights into catalysis and regulation. Annu Rev Biophys Biomol Struct 27, 133-164. Baritaki, S., Chapman, A., Yeung, K., Spandidos, D. A., Palladino, M., and Bonavida, B. (2009). Inhibition of epithelial to mesenchymal transition in metastatic prostate cancer cells by the novel proteasome inhibitor, NPI-0052: pivotal roles of Snail repression and RKIP induction. Oncogene 28, 3573-3585. Barrett, H., Butler, R., and Wilson, I. B. (1969). Evidence for a phosphoryl-enzyme intermediate in alkaline phosphatase catalyzed reactions. Biochemistry 8, 1042-1047.  158  Basak, S., Jacobs, S. B., Krieg, A. J., Pathak, N., Zeng, Q., Kaldis, P., Giaccia, A. J., and Attardi, L. D. (2008). The metastasis-associated gene Prl-3 is a p53 target involved in cellcycle regulation. Mol Cell 30, 303-314. Bergman, M., Mustelin, T., Oetken, C., Partanen, J., Flint, N. A., Amrein, K. E., Autero, M., Burn, P., and Alitalo, K. (1992). The human p50csk tyrosine kinase phosphorylates p56lck at Tyr-505 and down regulates its catalytic activity. EMBO J 11, 2919-2924. Berx, G., and van Roy, F. (2009). Involvement of members of the cadherin superfamily in cancer. Cold Spring Harb Perspect Biol 1, a003129. Bessette, D. C., Qiu, D., and Pallen, C. J. (2008). PRL PTPs: mediators and markers of cancer progression. Cancer Metastasis Rev 27, 231-252. Bessette, D. C., Wong, P. C., and Pallen, C. J. (2007). PRL-3: a metastasis-associated phosphatase in search of a function. Cells Tissues Organs 185, 232-236. Blackshear, P. J., Nairn, A. C., and Kuo, J. F. (1988). Protein kinases 1988: a current perspective. FASEB J 2, 2957-2969. Blakesley, V. A., Scrimgeour, A., Esposito, D., and Le Roith, D. (1996). Signaling via the insulin-like growth factor-I receptor: does it differ from insulin receptor signaling? Cytokine Growth Factor Rev 7, 153-159. Boerner, R. J., Kassel, D. B., Barker, S. C., Ellis, B., DeLacy, P., and Knight, W. B. (1996). Correlation of the phosphorylation states of pp60c-src with tyrosine kinase activity: the intramolecular pY530-SH2 complex retains significant activity if Y419 is phosphorylated. Biochemistry 35, 9519-9525. Boney, C. M., Sekimoto, H., Gruppuso, P. A., and Frackelton, A. R., Jr. (2001). Src family tyrosine kinases participate in insulin-like growth factor I mitogenic signaling in 3T3-L1 cells. Cell Growth Differ 12, 379-386. Boyer, B., and Thiery, J. P. (1993). Epithelium-mesenchyme interconversion as example of epithelial plasticity. APMIS 101, 257-268. Bretscher, A., Reczek, D., and Berryman, M. (1997). Ezrin: a protein requiring conformational activation to link microfilaments to the plasma membrane in the assembly of cell surface structures. J Cell Sci 110 ( Pt 24), 3011-3018. Brodt, P., Samani, A., and Navab, R. (2000). Inhibition of the type I insulin-like growth factor receptor expression and signaling: novel strategies for antimetastatic therapy. Biochem Pharmacol 60, 1101-1107. Brown, M. T., and Cooper, J. A. (1996). Regulation, substrates and functions of src. Biochim Biophys Acta 1287, 121-149. Burridge, K., and Chrzanowska-Wodnicka, M. (1996). Focal adhesions, contractility, and signaling. Annu Rev Cell Dev Biol 12, 463-518. Burridge, K., and Connell, L. (1983). Talin: a cytoskeletal component concentrated in adhesion plaques and other sites of actin-membrane interaction. Cell Motil 3, 405-417. Burridge, K., Nuckolls, G., Otey, C., Pavalko, F., Simon, K., and Turner, C. (1990). Actinmembrane interaction in focal adhesions. Cell Differ Dev 32, 337-342. Cai, S. R., Wang, Z., Chen, C. Q., Wu, W. H., He, Y. L., Zhan, W. H., Zhang, C. H., Cui, J., and Wu, H. (2008). Role of silencing phosphatase of regenerationg liver-3 expression by microRNA interference in the growth of gastric cancer. Chin Med J (Engl) 121, 2534-2538. Cano, A., Perez-Moreno, M. A., Rodrigo, I., Locascio, A., Blanco, M. J., del Barrio, M. G., Portillo, F., and Nieto, M. A. (2000). The transcription factor snail controls epithelialmesenchymal transitions by repressing E-cadherin expression. Nat Cell Biol 2, 76-83.  159  Carnero, A. (2010). The PKB/AKT pathway in cancer. Curr Pharm Des 16, 34-44. Carroll, A. G., Voeller, H. J., Sugars, L., and Gelmann, E. P. (1993). p53 oncogene mutations in three human prostate cancer cell lines. Prostate 23, 123-134. Cates, C. A., Michael, R. L., Stayrook, K. R., Harvey, K. A., Burke, Y. D., Randall, S. K., Crowell, P. L., and Crowell, D. N. (1996). Prenylation of oncogenic human PTP(CAAX) protein tyrosine phosphatases. Cancer Lett 110, 49-55. Chang, C., and Werb, Z. (2001). The many faces of metalloproteases: cell growth, invasion, angiogenesis and metastasis. Trends Cell Biol 11, S37-43. Chen, H. C., Appeddu, P. A., Parsons, J. T., Hildebrand, J. D., Schaller, M. D., and Guan, J. L. (1995). Interaction of focal adhesion kinase with cytoskeletal protein talin. J Biol Chem 270, 16995-16999. Chen, W. T., Chen, J. M., and Mueller, S. C. (1986). Coupled expression and colocalization of 140K cell adhesion molecules, fibronectin, and laminin during morphogenesis and cytodifferentiation of chick lung cells. J Cell Biol 103, 1073-1090. Chernoff, J., Schievella, A. R., Jost, C. A., Erikson, R. L., and Neel, B. G. (1990). Cloning of a cDNA for a major human protein-tyrosine-phosphatase. Proc Natl Acad Sci U S A 87, 2735-2739. Chiarugi, P., Cirri, P., Taddei, M. L., Giannoni, E., Fiaschi, T., Buricchi, F., Camici, G., Raugei, G., and Ramponi, G. (2002). Insight into the role of low molecular weight phosphotyrosine phosphatase (LMW-PTP) on platelet-derived growth factor receptor (PDGF-r) signaling. LMW-PTP controls PDGF-r kinase activity through TYR-857 dephosphorylation. J Biol Chem 277, 37331-37338. Choi, S. K., Oh, H. M., Lee, S. K., Jeong, D. G., Ryu, S. E., Son, K. H., Han, D. C., Sung, N. D., Baek, N. I., and Kwon, B. M. (2006). Biflavonoids inhibited phosphatase of regenerating liver-3 (PRL-3). Nat Prod Res 20, 341-346. Cobb, B. S., Schaller, M. D., Leu, T. H., and Parsons, J. T. (1994a). Stable association of pp60src and pp59fyn with the focal adhesion-associated protein tyrosine kinase, pp125FAK. Mol Cell Biol 14, 147-155. Cobb, M. H., Hepler, J. E., Cheng, M., and Robbins, D. (1994b). The mitogen-activated protein kinases, ERK1 and ERK2. Semin Cancer Biol 5, 261-268. Cohen, P. (1989). The structure and regulation of protein phosphatases. Annu Rev Biochem 58, 453-508. Cox, L. S. (1997). Multiple pathways control cell growth and transformation: overlapping and independent activities of p53 and p21Cip1/WAF1/Sdi1. J Pathol 183, 134-140. Cui, Q. L., and Almazan, G. (2007). IGF-I-induced oligodendrocyte progenitor proliferation requires PI3K/Akt, MEK/ERK, and Src-like tyrosine kinases. J Neurochem 100, 1480-1493. Dachs, G. U., and Stratford, I. J. (1996). The molecular response of mammalian cells to hypoxia and the potential for exploitation in cancer therapy. Br J Cancer Suppl 27, S126-132. David-Pfeuty, T., and Singer, S. J. (1980). Altered distributions of the cytoskeletal proteins vinculin and alpha-actinin in cultured fibroblasts transformed by Rous sarcoma virus. Proc Natl Acad Sci U S A 77, 6687-6691. Defilippi, P., Di Stefano, P., and Cabodi, S. (2006). p130Cas: a versatile scaffold in signaling networks. Trends Cell Biol 16, 257-263. Delafontaine, P. (1995). Insulin-like growth factor I and its binding proteins in the cardiovascular system. Cardiovasc Res 30, 825-834.  160  den Hertog, J., Groen, A., and van der Wijk, T. (2005). Redox regulation of protein-tyrosine phosphatases. Arch Biochem Biophys 434, 11-15. Denu, J. M., and Dixon, J. E. (1995). A catalytic mechanism for the dual-specific phosphatases. Proc Natl Acad Sci U S A 92, 5910-5914. Denu, J. M., Stuckey, J. A., Saper, M. A., and Dixon, J. E. (1996). Form and function in protein dephosphorylation. Cell 87, 361-364. Dewang, P. M., Hsu, N. M., Peng, S. Z., and Li, W. R. (2005). Protein tyrosine phosphatases and their inhibitors. Curr Med Chem 12, 1-22. Diamond, R. H., Cressman, D. E., Laz, T. M., Abrams, C. S., and Taub, R. (1994). PRL-1, a unique nuclear protein tyrosine phosphatase, affects cell growth. Mol Cell Biol 14, 37523762. Dillon, R. L., White, D. E., and Muller, W. J. (2007). The phosphatidyl inositol 3-kinase signaling network: implications for human breast cancer. Oncogene 26, 1338-1345. Drees, F., Pokutta, S., Yamada, S., Nelson, W. J., and Weis, W. I. (2005). Alpha-catenin is a molecular switch that binds E-cadherin-beta-catenin and regulates actin-filament assembly. Cell 123, 903-915. Duffy, M. J. (1987). Do proteases play a role in cancer invasion and metastasis? Eur J Cancer Clin Oncol 23, 583-589. Dumaual, C. M., Sandusky, G. E., Crowell, P. L., and Randall, S. K. (2006). Cellular localization of PRL-1 and PRL-2 gene expression in normal adult human tissues. J Histochem Cytochem 54, 1401-1412. Dunphy, W. G., and Kumagai, A. (1991). The cdc25 protein contains an intrinsic phosphatase activity. Cell 67, 189-196. Dursina, B., Reents, R., Delon, C., Wu, Y., Kulharia, M., Thutewohl, M., Veligodsky, A., Kalinin, A., Evstifeev, V., Ciobanu, D., et al. (2006). Identification and specificity profiling of protein prenyltransferase inhibitors using new fluorescent phosphoisoprenoids. J Am Chem Soc 128, 2822-2835. Edelman, A. M., Blumenthal, D. K., and Krebs, E. G. (1987). Protein serine/threonine kinases. Annu Rev Biochem 56, 567-613. el-Deiry, W. S., Tokino, T., Velculescu, V. E., Levy, D. B., Parsons, R., Trent, J. M., Lin, D., Mercer, W. E., Kinzler, K. W., and Vogelstein, B. (1993). WAF1, a potential mediator of p53 tumor suppression. Cell 75, 817-825. Fagerli, U. M., Holt, R. U., Holien, T., Vaatsveen, T. K., Zhan, F., Egeberg, K. W., Barlogie, B., Waage, A., Aarset, H., Dai, H. Y., et al. (2008). Overexpression and involvement in migration by the metastasis-associated phosphatase PRL-3 in human myeloma cells. Blood 111, 806-815. Fiordalisi, J. J., Keller, P. J., and Cox, A. D. (2006). PRL tyrosine phosphatases regulate rho family GTPases to promote invasion and motility. Cancer Res 66, 3153-3161. Fischer, E. H., Tonks, N. K., Charbonneau, H., Cicirelli, M. F., Cool, D. E., Diltz, C. D., Krebs, E. G., and Walsh, K. A. (1990). Protein tyrosine phosphatases: a novel family of enzymes involved in transmembrane signalling. Adv Second Messenger Phosphoprotein Res 24, 273-279. Forte, E., Orsatti, L., Talamo, F., Barbato, G., De Francesco, R., and Tomei, L. (2008). Ezrin is a specific and direct target of protein tyrosine phosphatase PRL-3. Biochim Biophys Acta 1783, 334-344.  161  Galcheva-Gargova, Z., Derijard, B., Wu, I. H., and Davis, R. J. (1994). An osmosensing signal transduction pathway in mammalian cells. Science 265, 806-808. Gan, Y., Shi, C., Inge, L., Hibner, M., Balducci, J., and Huang, Y. (2010). Differential roles of ERK and Akt pathways in regulation of EGFR-mediated signaling and motility in prostate cancer cells. Oncogene 29, 4947-4958. Gardner, A. M., Vaillancourt, R. R., and Johnson, G. L. (1993). Activation of mitogenactivated protein kinase/extracellular signal-regulated kinase kinase by G protein and tyrosine kinase oncoproteins. J Biol Chem 268, 17896-17901. Geiger, B., Avnur, Z., Rinnerthaler, G., Hinssen, H., and Small, V. J. (1984). Microfilamentorganizing centers in areas of cell contact: cytoskeletal interactions during cell attachment and locomotion. J Cell Biol 99, 83s-91s. Geiger, B., Tokuyasu, K. T., Dutton, A. H., and Singer, S. J. (1980). Vinculin, an intracellular protein localized at specialized sites where microfilament bundles terminate at cell membranes. Proc Natl Acad Sci U S A 77, 4127-4131. Graeber, T. G., Peterson, J. F., Tsai, M., Monica, K., Fornace, A. J., Jr., and Giaccia, A. J. (1994). Hypoxia induces accumulation of p53 protein, but activation of a G1-phase checkpoint by low-oxygen conditions is independent of p53 status. Mol Cell Biol 14, 62646277. Greenwald, I. (1998). LIN-12/Notch signaling: lessons from worms and flies. Genes Dev 12, 1751-1762. Gross, J. M., and Yee, D. (2003). The type-1 insulin-like growth factor receptor tyrosine kinase and breast cancer: biology and therapeutic relevance. Cancer Metastasis Rev 22, 327336. Guan, K. L. (1994). The mitogen activated protein kinase signal transduction pathway: from the cell surface to the nucleus. Cell Signal 6, 581-589. Guarino, M., Rubino, B., and Ballabio, G. (2007). The role of epithelial-mesenchymal transition in cancer pathology. Pathology 39, 305-318. Guo, K., Li, J., Tang, J. P., Koh, V., Gan, B. Q., and Zeng, Q. (2004). Catalytic domain of PRL-3 plays an essential role in tumor metastasis: formation of PRL-3 tumors inside the blood vessels. Cancer Biol Ther 3, 945-951. Guo, K., Li, J., Wang, H., Osato, M., Tang, J. P., Quah, S. Y., Gan, B. Q., and Zeng, Q. (2006). PRL-3 initiates tumor angiogenesis by recruiting endothelial cells in vitro and in vivo. Cancer Res 66, 9625-9635. Hammond, E. M., Green, S. L., and Giaccia, A. J. (2003). Comparison of hypoxia-induced replication arrest with hydroxyurea and aphidicolin-induced arrest. Mutat Res 532, 205-213. Han, J., Lee, J. D., Bibbs, L., and Ulevitch, R. J. (1994). A MAP kinase targeted by endotoxin and hyperosmolarity in mammalian cells. Science 265, 808-811. Hanahan, D., and Weinberg, R. A. (2000). The hallmarks of cancer. Cell 100, 57-70. Hao, R. T., Zhang, X. H., Pan, Y. F., Liu, H. G., Xiang, Y. Q., Wan, L., and Wu, X. L. (2010). Prognostic and metastatic value of phosphatase of regenerating liver-3 in invasive breast cancer. J Cancer Res Clin Oncol 136, 1349-1357. Hardie, D. G. (1990). Roles of protein kinases and phosphatases in signal transduction. Symp Soc Exp Biol 44, 241-255. Harper, J. W., Adami, G. R., Wei, N., Keyomarsi, K., and Elledge, S. J. (1993). The p21 Cdk-interacting protein Cip1 is a potent inhibitor of G1 cyclin-dependent kinases. Cell 75, 805-816.  162  Hatate, K., Yamashita, K., Hirai, K., Kumamoto, H., Sato, T., Ozawa, H., Nakamura, T., Onozato, W., Kokuba, Y., Ihara, A., and Watanabe, M. (2008). Liver metastasis of colorectal cancer by protein-tyrosine phosphatase type 4A, 3 (PRL-3) is mediated through lymph node metastasis and elevated serum tumor markers such as CEA and CA19-9. Oncol Rep 20, 737743. Hiscox, S., and Jiang, W. G. (1999). Ezrin regulates cell-cell and cell-matrix adhesion, a possible role with E-cadherin/beta-catenin. J Cell Sci 112 Pt 18, 3081-3090. Horwitz, A., Duggan, K., Buck, C., Beckerle, M. C., and Burridge, K. (1986). Interaction of plasma membrane fibronectin receptor with talin--a transmembrane linkage. Nature 320, 531-533. Huang, L., Sankar, S., Lin, C., Kontos, C. D., Schroff, A. D., Cha, E. H., Feng, S. M., Li, S. F., Yu, Z., Van Etten, R. L., et al. (1999). HCPTPA, a protein tyrosine phosphatase that regulates vascular endothelial growth factor receptor-mediated signal transduction and biological activity. J Biol Chem 274, 38183-38188. Huber, O., Korn, R., McLaughlin, J., Ohsugi, M., Herrmann, B. G., and Kemler, R. (1996). Nuclear localization of beta-catenin by interaction with transcription factor LEF-1. Mech Dev 59, 3-10. Huo, L. R., and Zhong, N. (2008). Identification of transcripts and translatants targeted by overexpressed PCBP1. Biochim Biophys Acta 1784, 1524-1533. Ishino, A., Kusama, K., Watanabe, S., and Sakagami, H. (1999). Inhibition of epigallocatechin gallate-induced apoptosis by CoCl2 in human oral tumor cell lines. Anticancer Res 19, 5197-5201. Jeong, D. G., Kim, S. J., Kim, J. H., Son, J. H., Park, M. R., Lim, S. M., Yoon, T. S., and Ryu, S. E. (2005). Trimeric structure of PRL-1 phosphatase reveals an active enzyme conformation and regulation mechanisms. J Mol Biol 345, 401-413. John, A., and Tuszynski, G. (2001). The role of matrix metalloproteinases in tumor angiogenesis and tumor metastasis. Pathol Oncol Res 7, 14-23. Julien, S., Puig, I., Caretti, E., Bonaventure, J., Nelles, L., van Roy, F., Dargemont, C., de Herreros, A. G., Bellacosa, A., and Larue, L. (2007). Activation of NF-kappaB by Akt upregulates Snail expression and induces epithelium mesenchyme transition. Oncogene 26, 7445-7456. Karin, M. (1992). Signal transduction from cell surface to nucleus in development and disease. FASEB J 6, 2581-2590. Kato, H., Semba, S., Miskad, U. A., Seo, Y., Kasuga, M., and Yokozaki, H. (2004). High expression of PRL-3 promotes cancer cell motility and liver metastasis in human colorectal cancer: a predictive molecular marker of metachronous liver and lung metastases. Clin Cancer Res 10, 7318-7328. Kawajiri, A., Itoh, N., Fukata, M., Nakagawa, M., Yamaga, M., Iwamatsu, A., and Kaibuchi, K. (2000). Identification of a novel beta-catenin-interacting protein. Biochem Biophys Res Commun 273, 712-717. Kim, K. A., Song, J. S., Jee, J., Sheen, M. R., Lee, C., Lee, T. G., Ro, S., Cho, J. M., Lee, W., Yamazaki, T., et al. (2004). Structure of human PRL-3, the phosphatase associated with cancer metastasis. FEBS Lett 565, 181-187. Kleinman, H. K., McGarvey, M. L., Hassell, J. R., Star, V. L., Cannon, F. B., Laurie, G. W., and Martin, G. R. (1986). Basement membrane complexes with biological activity. Biochemistry 25, 312-318.  163  Kleinman, H. K., McGarvey, M. L., Liotta, L. A., Robey, P. G., Tryggvason, K., and Martin, G. R. (1982). Isolation and characterization of type IV procollagen, laminin, and heparan sulfate proteoglycan from the EHS sarcoma. Biochemistry 21, 6188-6193. Klocker, H., Culig, Z., Kaspar, F., Hobisch, A., Eberle, J., Reissigl, A., and Bartsch, G. (1994). Androgen signal transduction and prostatic carcinoma. World J Urol 12, 99-103. Kmiecik, T. E., Johnson, P. J., and Shalloway, D. (1988). Regulation by the autophosphorylation site in overexpressed pp60c-src. Mol Cell Biol 8, 4541-4546. Koegl, M., Courtneidge, S. A., and Superti-Furga, G. (1995). Structural requirements for the efficient regulation of the Src protein tyrosine kinase by Csk. Oncogene 11, 2317-2329. Kong, L., Li, Q., Wang, L., Liu, Z., and Sun, T. (2007). The value and correlation between PRL-3 expression and matrix metalloproteinase activity and expression in human gliomas. Neuropathology 27, 516-521. Kopfstein, L., and Christofori, G. (2006). Metastasis: cell-autonomous mechanisms versus contributions by the tumor microenvironment. Cell Mol Life Sci 63, 449-468. Kozlov, G., Cheng, J., Lievre, C., Banville, D., Gehring, K., and Ekiel, I. (2002). 1H, 13C and 15N resonance assignments of the human phosphatase PRL-3. J Biomol NMR 24, 169170. Kozlov, G., Cheng, J., Ziomek, E., Banville, D., Gehring, K., and Ekiel, I. (2004). Structural insights into molecular function of the metastasis-associated phosphatase PRL-3. J Biol Chem 279, 11882-11889. Larue, L., and Bellacosa, A. (2005). Epithelial-mesenchymal transition in development and cancer: role of phosphatidylinositol 3' kinase/AKT pathways. Oncogene 24, 7443-7454. Leroith, D., Werner, H., Neuenschwander, S., Kalebic, T., and Helman, L. J. (1995). The Role of the Insulin-like Growth Factor-I Receptor in Cancer. Annals of the New York Academy of Sciences 766, 402-408. Lewis, M. E., Neff, N. T., Contreras, P. C., Stong, D. B., Oppenheim, R. W., Grebow, P. E., and Vaught, J. L. (1993). Insulin-like growth factor-I: potential for treatment of motor neuronal disorders. Exp Neurol 124, 73-88. Li, J., Guo, K., Koh, V. W., Tang, J. P., Gan, B. Q., Shi, H., Li, H. X., and Zeng, Q. (2005). Generation of PRL-3- and PRL-1-specific monoclonal antibodies as potential diagnostic markers for cancer metastases. Clin Cancer Res 11, 2195-2204. Li, X., Deng, W., Lobo-Ruppert, S. M., and Ruppert, J. M. (2007a). Gli1 acts through Snail and E-cadherin to promote nuclear signaling by beta-catenin. Oncogene 26, 4489-4498. Li, Z., Zhan, W., Wang, Z., Zhu, B., He, Y., Peng, J., Cai, S., and Ma, J. (2006). Inhibition of PRL-3 gene expression in gastric cancer cell line SGC7901 via microRNA suppressed reduces peritoneal metastasis. Biochem Biophys Res Commun 348, 229-237. Li, Z. R., Wang, Z., Zhu, B. H., He, Y. L., Peng, J. S., Cai, S. R., Ma, J. P., and Zhan, W. H. (2007b). Association of tyrosine PRL-3 phosphatase protein expression with peritoneal metastasis of gastric carcinoma and prognosis. Surg Today 37, 646-651. Liang, F., Liang, J., Wang, W. Q., Sun, J. P., Udho, E., and Zhang, Z. Y. (2007). PRL3 promotes cell invasion and proliferation by down-regulation of Csk leading to Src activation. J Biol Chem 282, 5413-5419. Liang, F., Luo, Y., Dong, Y., Walls, C. D., Liang, J., Jiang, H. Y., Sanford, J. R., Wek, R. C., and Zhang, Z. Y. (2008). Translational control of C-terminal Src kinase (Csk) expression by PRL3 phosphatase. J Biol Chem 283, 10339-10346.  164  Lipfert, L., Haimovich, B., Schaller, M. D., Cobb, B. S., Parsons, J. T., and Brugge, J. S. (1992). Integrin-dependent phosphorylation and activation of the protein tyrosine kinase pp125FAK in platelets. J Cell Biol 119, 905-912. Liu, Y., Zhou, J., Chen, J., Gao, W., Le, Y., Ding, Y., and Li, J. (2009). PRL-3 promotes epithelial mesenchymal transition by regulating cadherin directly. Cancer Biol Ther 8, 13521359. Liu, Y. Q., Li, H. X., Lou, X., and Lei, J. Y. (2008). Expression of phosphatase of regenerating liver 1 and 3 mRNA in esophageal squamous cell carcinoma. Arch Pathol Lab Med 132, 1307-1312. Luo, Y., He, D. L., Ning, L., Shen, S. L., Li, L., and Li, X. (2006). Hypoxia-inducible factor1alpha induces the epithelial-mesenchymal transition of human prostatecancer cells. Chin Med J (Engl) 119, 713-718. Ma, Y., and Li, B. (2010). Expression of phosphatase of regenerating liver-3 in squamous cell carcinoma of the cervix. Med Oncol. Macdonald, S. G., Crews, C. M., Wu, L., Driller, J., Clark, R., Erikson, R. L., and McCormick, F. (1993). Reconstitution of the Raf-1-MEK-ERK signal transduction pathway in vitro. Mol Cell Biol 13, 6615-6620. Magnusson, C., Svensson, A., Christerson, U., and Tagerud, S. (2005). Denervation-induced alterations in gene expression in mouse skeletal muscle. Eur J Neurosci 21, 577-580. Makeyev, A. V., and Liebhaber, S. A. (2002). The poly(C)-binding proteins: a multiplicity of functions and a search for mechanisms. RNA 8, 265-278. Manome, H., Aiba, S., and Tagami, H. (1999). Simple chemicals can induce maturation and apoptosis of dendritic cells. Immunology 98, 481-490. Matsui, T., Maeda, M., Doi, Y., Yonemura, S., Amano, M., Kaibuchi, K., Tsukita, S., and Tsukita, S. (1998). Rho-kinase phosphorylates COOH-terminal threonines of ezrin/radixin/moesin (ERM) proteins and regulates their head-to-tail association. J Cell Biol 140, 647-657. Matsukawa, Y., Semba, S., Kato, H., Koma, Y., Yanagihara, K., and Yokozaki, H. (2010). Constitutive suppression of PRL-3 inhibits invasion and proliferation of gastric cancer cell in vitro and in vivo. Pathobiology 77, 155-162. Matter, W. F., Estridge, T., Zhang, C., Belagaje, R., Stancato, L., Dixon, J., Johnson, B., Bloem, L., Pickard, T., Donaghue, M., et al. (2001). Role of PRL-3, a human muscle-specific tyrosine phosphatase, in angiotensin-II signaling. Biochem Biophys Res Commun 283, 10611068. Mazure, N. M., Chen, E. Y., Laderoute, K. R., and Giaccia, A. J. (1997). Induction of vascular endothelial growth factor by hypoxia is modulated by a phosphatidylinositol 3kinase/Akt signaling pathway in Ha-ras-transformed cells through a hypoxia inducible factor1 transcriptional element. Blood 90, 3322-3331. McGuire, P. G., and Seeds, N. W. (1989). The interaction of plasminogen activator with a reconstituted basement membrane matrix and extracellular macromolecules produced by cultured epithelial cells. J Cell Biochem 40, 215-227. Meng, W., and Takeichi, M. (2009). Adherens junction: molecular architecture and regulation. Cold Spring Harb Perspect Biol 1, a002899. Merlino, G. T. (1990). Epidermal growth factor receptor regulation and function. Semin Cancer Biol 1, 277-284.  165  Millar, J. B., and Russell, P. (1992). The cdc25 M-phase inducer: an unconventional protein phosphatase. Cell 68, 407-410. Min, S. H., Kim, D. M., Heo, Y. S., Kim, H. M., Kim, I. C., and Yoo, O. J. (2010). Downregulation of p53 by phosphatase of regenerating liver 3 is mediated by MDM2 and PIRH2. Life Sci 86, 66-72. Ming, J., Liu, N., Gu, Y., Qiu, X., and Wang, E. H. (2009). PRL-3 facilitates angiogenesis and metastasis by increasing ERK phosphorylation and up-regulating the levels and activities of Rho-A/C in lung cancer. Pathology 41, 118-126. Miskad, U. A., Semba, S., Kato, H., Matsukawa, Y., Kodama, Y., Mizuuchi, E., Maeda, N., Yanagihara, K., and Yokozaki, H. (2007). High PRL-3 expression in human gastric cancer is a marker of metastasis and grades of malignancies: an in situ hybridization study. Virchows Arch 450, 303-310. Miskad, U. A., Semba, S., Kato, H., and Yokozaki, H. (2004). Expression of PRL-3 phosphatase in human gastric carcinomas: close correlation with invasion and metastasis. Pathobiology 71, 176-184. Mizuuchi, E., Semba, S., Kodama, Y., and Yokozaki, H. (2009). Down-modulation of keratin 8 phosphorylation levels by PRL-3 contributes to colorectal carcinoma progression. Int J Cancer 124, 1802-1810. Moarefi, I., LaFevre-Bernt, M., Sicheri, F., Huse, M., Lee, C. H., Kuriyan, J., and Miller, W. T. (1997). Activation of the Src-family tyrosine kinase Hck by SH3 domain displacement. Nature 385, 650-653. Mohn, K. L., Laz, T. M., Hsu, J. C., Melby, A. E., Bravo, R., and Taub, R. (1991). The immediate-early growth response in regenerating liver and insulin-stimulated H-35 cells: comparison with serum-stimulated 3T3 cells and identification of 41 novel immediate-early genes. Mol Cell Biol 11, 381-390. Mollevi, D. G., Aytes, A., Padulles, L., Martinez-Iniesta, M., Baixeras, N., Salazar, R., Ramos, E., Figueras, J., Capella, G., and Villanueva, A. (2008). PRL-3 is essentially overexpressed in primary colorectal tumours and associates with tumour aggressiveness. Br J Cancer 99, 1718-1725. Monier, S., Le Cam, A., and Le Marchand-Brustel, Y. (1983). Insulin and insulin-like growth factor I. Effects on protein synthesis in isolated muscles from lean and goldthioglucose-obese mice. Diabetes 32, 392-397. Montagna, M., Serova, O., Sylla, B. S., Feunteun, J., and Lenoir, G. M. (1995). A 100-kb physical and transcriptional map around the EDH17B2 gene: identification of three novel genes and a pseudogene of a human homologue of the rat PRL-1 tyrosine phosphatase. Hum Genet 96, 532-538. Neel, B. G., and Tonks, N. K. (1997). Protein tyrosine phosphatases in signal transduction. Curr Opin Cell Biol 9, 193-204. Ng, Y., and Pallen, C. J. (2007). PRL-3 does not alter MMP expression in 293 cells. Nicholson, K. M., and Anderson, N. G. (2002). The protein kinase B/Akt signalling pathway in human malignancy. Cell Signal 14, 381-395. Nieman, M. T., Prudoff, R. S., Johnson, K. R., and Wheelock, M. J. (1999). N-cadherin promotes motility in human breast cancer cells regardless of their E-cadherin expression. J Cell Biol 147, 631-644. Odero-Marah, V. A., Wang, R., Chu, G., Zayzafoon, M., Xu, J., Shi, C., Marshall, F. F., Zhau, H. E., and Chung, L. W. (2008). Receptor activator of NF-kappaB Ligand (RANKL)  166  expression is associated with epithelial to mesenchymal transition in human prostate cancer cells. Cell Res 18, 858-870. Okada, M., Nada, S., Yamanashi, Y., Yamamoto, T., and Nakagawa, H. (1991). CSK: a protein-tyrosine kinase involved in regulation of src family kinases. J Biol Chem 266, 2424924252. Ooki, A., Yamashita, K., Kikuchi, S., Sakuramoto, S., Katada, N., and Watanabe, M. (2009). Phosphatase of regenerating liver-3 as a prognostic biomarker in histologically node-negative gastric cancer. Oncol Rep 21, 1467-1475. Ooki, A., Yamashita, K., Kikuchi, S., Sakuramoto, S., Katada, N., and Watanabe, M. (2010). Phosphatase of regenerating liver-3 as a convergent therapeutic target for lymph node metastasis in esophageal squamous cell carcinoma. Int J Cancer 127, 543-554. Orsatti, L., Innocenti, F., Lo Surdo, P., Talamo, F., and Barbato, G. (2009). Mass spectrometry study of PRL-3 phosphatase inactivation by disulfide bond formation and cysteine into glycine conversion. Rapid Commun Mass Spectrom 23, 2733-2740. Oshima, R. G., Baribault, H., and Caulin, C. (1996). Oncogenic regulation and function of keratins 8 and 18. Cancer Metastasis Rev 15, 445-471. Park, E. K., Warner, N., Mood, K., Pawson, T., and Daar, I. O. (2002). Low-molecularweight protein tyrosine phosphatase is a positive component of the fibroblast growth factor receptor signaling pathway. Mol Cell Biol 22, 3404-3414. Parker, B. S., Argani, P., Cook, B. P., Liangfeng, H., Chartrand, S. D., Zhang, M., Saha, S., Bardelli, A., Jiang, Y., St Martin, T. B., et al. (2004). Alterations in vascular gene expression in invasive breast carcinoma. Cancer Res 64, 7857-7866. Pascaru, M., Tanase, C., Vacaru, A. M., Boeti, P., Neagu, E., Popescu, I., and Szedlacsek, S. E. (2009). Analysis of molecular determinants of PRL-3. J Cell Mol Med 13, 3141-3150. Pathak, M. K., Dhawan, D., Lindner, D. J., Borden, E. C., Farver, C., and Yi, T. (2002). Pentamidine is an inhibitor of PRL phosphatases with anticancer activity. Mol Cancer Ther 1, 1255-1264. Paul, S., and Lombroso, P. J. (2003). Receptor and nonreceptor protein tyrosine phosphatases in the nervous system. Cell Mol Life Sci 60, 2465-2482. Peng, L., Jin, G., Wang, L., Guo, J., Meng, L., and Shou, C. (2006). Identification of integrin alpha1 as an interacting protein of protein tyrosine phosphatase PRL-3. Biochem Biophys Res Commun 342, 179-183. Peng, L., Ning, J., Meng, L., and Shou, C. (2004). The association of the expression level of protein tyrosine phosphatase PRL-3 protein with liver metastasis and prognosis of patients with colorectal cancer. J Cancer Res Clin Oncol 130, 521-526. Peng, L., Xing, X., Li, W., Qu, L., Meng, L., Lian, S., Jiang, B., Wu, J., and Shou, C. (2009). PRL-3 promotes the motility, invasion, and metastasis of LoVo colon cancer cells through PRL-3-integrin beta1-ERK1/2 and-MMP2 signaling. Mol Cancer 8, 110. Polato, F., Codegoni, A., Fruscio, R., Perego, P., Mangioni, C., Saha, S., Bardelli, A., and Broggini, M. (2005). PRL-3 phosphatase is implicated in ovarian cancer growth. Clin Cancer Res 11, 6835-6839. Polyak, K., Lee, M. H., Erdjument-Bromage, H., Koff, A., Roberts, J. M., Tempst, P., and Massague, J. (1994). Cloning of p27Kip1, a cyclin-dependent kinase inhibitor and a potential mediator of extracellular antimitogenic signals. Cell 78, 59-66.  167  Qian, F., Li, Y. P., Sheng, X., Zhang, Z. C., Song, R., Dong, W., Cao, S. X., Hua, Z. C., and Xu, Q. (2007). PRL-3 siRNA inhibits the metastasis of B16-BL6 mouse melanoma cells in vitro and in vivo. Mol Med 13, 151-159. Radke, I., Gotte, M., Kersting, C., Mattsson, B., Kiesel, L., and Wulfing, P. (2006). Expression and prognostic impact of the protein tyrosine phosphatases PRL-1, PRL-2, and PRL-3 in breast cancer. Br J Cancer 95, 347-354. Ren, T., Jiang, B., Xing, X., Dong, B., Peng, L., Meng, L., Xu, H., and Shou, C. (2009). Prognostic significance of phosphatase of regenerating liver-3 expression in ovarian cancer. Pathol Oncol Res 15, 555-560. Resta, C. D., and Becchetti, A. (2010). Introduction to ion channels, In Integrins and Ion Channels, A. Becchetti, and A. Arcangeli, eds. (Springer), pp. 9-21. Ridley, A. J., Allen, W. E., Peppelenbosch, M., and Jones, G. E. (1999). Rho family proteins and cell migration. Biochem Soc Symp 65, 111-123. Rogakou, E. P., Nieves-Neira, W., Boon, C., Pommier, Y., and Bonner, W. M. (2000). Initiation of DNA fragmentation during apoptosis induces phosphorylation of H2AX histone at serine 139. J Biol Chem 275, 9390-9395. Rohrschneider, L. R. (1980). Adhesion plaques of Rous sarcoma virus-transformed cells contain the src gene product. Proc Natl Acad Sci U S A 77, 3514-3518. Rokhlin, O. W., and Cohen, M. B. (1995). Expression of cellular adhesion molecules on human prostate tumor cell lines. Prostate 26, 205-212. Rouleau, C., Roy, A., St Martin, T., Dufault, M. R., Boutin, P., Liu, D., Zhang, M., PuorroRadzwill, K., Rulli, L., Reczek, D., et al. (2006). Protein tyrosine phosphatase PRL-3 in malignant cells and endothelial cells: expression and function. Mol Cancer Ther 5, 219-229. Rubin, S. J., Hallahan, D. E., Ashman, C. R., Brachman, D. G., Beckett, M. A., Virudachalam, S., Yandell, D. W., and Weichselbaum, R. R. (1991). Two prostate carcinoma cell lines demonstrate abnormalities in tumor suppressor genes. J Surg Oncol 46, 31-36. Saha, S., Bardelli, A., Buckhaults, P., Velculescu, V. E., Rago, C., St Croix, B., Romans, K. E., Choti, M. A., Lengauer, C., Kinzler, K. W., and Vogelstein, B. (2001). A phosphatase associated with metastasis of colorectal cancer. Science 294, 1343-1346. Sands, M., Kron, M. A., and Brown, R. B. (1985). Pentamidine: a review. Rev Infect Dis 7, 625-634. Sankar, S., Mahooti-Brooks, N., Hu, G., and Madri, J. A. (1995). Modulation of cell spreading and migration by pp125FAK phosphorylation. Am J Pathol 147, 601-608. Schaber, M. D., O'Hara, M. B., Garsky, V. M., Mosser, S. C., Bergstrom, J. D., Moores, S. L., Marshall, M. S., Friedman, P. A., Dixon, R. A., and Gibbs, J. B. (1990). Polyisoprenylation of Ras in vitro by a farnesyl-protein transferase. J Biol Chem 265, 1470114704. Schaller, M. D., Borgman, C. A., Cobb, B. S., Vines, R. R., Reynolds, A. B., and Parsons, J. T. (1992). pp125FAK a structurally distinctive protein-tyrosine kinase associated with focal adhesions. Proc Natl Acad Sci U S A 89, 5192-5196. Schlaepfer, D. D., Hanks, S. K., Hunter, T., and van der Geer, P. (1994). Integrin-mediated signal transduction linked to Ras pathway by GRB2 binding to focal adhesion kinase. Nature 372, 786-791. Schmidt, R. A., Schneider, C. J., and Glomset, J. A. (1984). Evidence for post-translational incorporation of a product of mevalonic acid into Swiss 3T3 cell proteins. J Biol Chem 259, 10175-10180.  168  Schneider, S., Steinbeisser, H., Warga, R. M., and Hausen, P. (1996). Beta-catenin translocation into nuclei demarcates the dorsalizing centers in frog and fish embryos. Mech Dev 57, 191-198. Schults, M. A., Timmermans, L., Godschalk, R. W., Theys, J., Wouters, B. G., van Schooten, F. J., and Chiu, R. K. (2010). Diminished carcinogen detoxification is a novel mechanism for hypoxia-inducible factor 1-mediated genetic instability. J Biol Chem 285, 14558-14564. Sekharam, M., Nasir, A., Kaiser, H. E., and Coppola, D. (2003). Insulin-like growth factor 1 receptor activates c-SRC and modifies transformation and motility of colon cancer in vitro. Anticancer Res 23, 1517-1524. Semba, S., Mizuuchi, E., and Yokozaki, H. (2010). Requirement of phosphatase of regenerating liver-3 for the nucleolar localization of nucleolin during the progression of colorectal carcinoma. Cancer Sci 101, 2254-2261. Sharma, R. A., Gescher, A. J., and Steward, W. P. (2005). Curcumin: the story so far. Eur J Cancer 41, 1955-1968. Sherwood, E. R., Berg, L. A., Mitchell, N. J., McNeal, J. E., Kozlowski, J. M., and Lee, C. (1990). Differential cytokeratin expression in normal, hyperplastic and malignant epithelial cells from human prostate. J Urol 143, 167-171. Si, X., Zeng, Q., Ng, C. H., Hong, W., and Pallen, C. J. (2001). Interaction of farnesylated PRL-2, a protein-tyrosine phosphatase, with the beta-subunit of geranylgeranyltransferase II. J Biol Chem 276, 32875-32882. Snoek, R., Cheng, H., Margiotti, K., Wafa, L. A., Wong, C. A., Wong, E. C., Fazli, L., Nelson, C. C., Gleave, M. E., and Rennie, P. S. (2009). In vivo knockdown of the androgen receptor results in growth inhibition and regression of well-established, castration-resistant prostate tumors. Clin Cancer Res 15, 39-47. Song, R., Qian, F., Li, Y. P., Sheng, X., Cao, S. X., and Xu, Q. (2009). Phosphatase of regenerating liver-3 localizes to cyto-membrane and is required for B16F1 melanoma cell metastasis in vitro and in vivo. PLoS One 4, e4450. Soulsby, M., and Bennett, A. M. (2009). Physiological signaling specificity by protein tyrosine phosphatases. Physiology (Bethesda) 24, 281-289. Sparks, J. W., and Brautigan, D. L. (1986). Molecular basis for substrate specificity of protein kinases and phosphatases. Int J Biochem 18, 497-504. St Croix, B., Rago, C., Velculescu, V., Traverso, G., Romans, K. E., Montgomery, E., Lal, A., Riggins, G. J., Lengauer, C., Vogelstein, B., and Kinzler, K. W. (2000). Genes expressed in human tumor endothelium. Science 289, 1197-1202. Staal, S. P. (1987). Molecular cloning of the akt oncogene and its human homologues AKT1 and AKT2: amplification of AKT1 in a primary human gastric adenocarcinoma. Proc Natl Acad Sci U S A 84, 5034-5037. Stetler-Stevenson, W. G., Liotta, L. A., and Kleiner, D. E., Jr. (1993). Extracellular matrix 6: role of matrix metalloproteinases in tumor invasion and metastasis. FASEB J 7, 1434-1441. Stickel, S. K., and Wang, Y. L. (1988). Synthetic peptide GRGDS induces dissociation of alpha-actinin and vinculin from the sites of focal contacts. J Cell Biol 107, 1231-1239. Subauste, M. C., Nalbant, P., Adamson, E. D., and Hahn, K. M. (2005). Vinculin controls PTEN protein level by maintaining the interaction of the adherens junction protein betacatenin with the scaffolding protein MAGI-2. J Biol Chem 280, 5676-5681.  169  Sun, H., Charles, C. H., Lau, L. F., and Tonks, N. K. (1993). MKP-1 (3CH134), an immediate early gene product, is a dual specificity phosphatase that dephosphorylates MAP kinase in vivo. Cell 75, 487-493. Sun, J. P., Luo, Y., Yu, X., Wang, W. Q., Zhou, B., Liang, F., and Zhang, Z. Y. (2007). Phosphatase activity, trimerization, and the C-terminal polybasic region are all required for PRL1-mediated cell growth and migration. J Biol Chem 282, 29043-29051. Sun, J. P., Wang, W. Q., Yang, H., Liu, S., Liang, F., Fedorov, A. A., Almo, S. C., and Zhang, Z. Y. (2005). Structure and biochemical properties of PRL-1, a phosphatase implicated in cell growth, differentiation, and tumor invasion. Biochemistry 44, 1200912021. Swartz, K. J. (2008). Sensing voltage across lipid membranes. Nature 456, 891-897. Tang, X., Woodward, T., and Amar, S. (2009). A PTP4A3 peptide PIMAP39 modulates TNF-alpha levels and endotoxic shock. J Innate Immun 2, 43-55. Terasaka, H., Tamura, A., Takayama, F., Kashimata, M., Ohtomo, K., Machino, M., Fujisawa, S., Toguchi, M., Kanda, Y., Kunii, S., et al. (2000). Induction of apoptosis by dopamine in human oral tumor cell lines. Anticancer Res 20, 243-250. Terranova, V. P., Hujanen, E. S., and Martin, G. R. (1986). Basement membrane and the invasive activity of metastatic tumor cells. J Natl Cancer Inst 77, 311-316. Thiery, J. P., and Sleeman, J. P. (2006). Complex networks orchestrate epithelialmesenchymal transitions. Nat Rev Mol Cell Biol 7, 131-142. Thompson, A. M., Steel, C. M., Chetty, U., Hawkins, R. A., Miller, W. R., Carter, D. C., Forrest, A. P., and Evans, H. J. (1990). p53 gene mRNA expression and chromosome 17p allele loss in breast cancer. Br J Cancer 61, 74-78. Tonks, N. K., and Charbonneau, H. (1989). Protein tyrosine dephosphorylation and signal transduction. Trends Biochem Sci 14, 497-500. Tonks, N. K., Diltz, C. D., and Fischer, E. H. (1988). Purification of the major proteintyrosine-phosphatases of human placenta. J Biol Chem 263, 6722-6730. Troppmair, J., Bruder, J. T., Munoz, H., Lloyd, P. A., Kyriakis, J., Banerjee, P., Avruch, J., and Rapp, U. R. (1994). Mitogen-activated protein kinase/extracellular signal-regulated protein kinase activation by oncogenes, serum, and 12-O-tetradecanoylphorbol-13-acetate requires Raf and is necessary for transformation. J Biol Chem 269, 7030-7035. van Steenbrugge, G. J., Groen, M., van Dongen, J. W., Bolt, J., van der Korput, H., Trapman, J., Hasenson, M., and Horoszewicz, J. (1989). The human prostatic carcinoma cell line LNCaP and its derivatives. An overview. Urol Res 17, 71-77. Vlietstra, R. J., van Alewijk, D. C., Hermans, K. G., van Steenbrugge, G. J., and Trapman, J. (1998). Frequent inactivation of PTEN in prostate cancer cell lines and xenografts. Cancer Res 58, 2720-2723. Vukicevic, S., Kleinman, H. K., Luyten, F. P., Roberts, A. B., Roche, N. S., and Reddi, A. H. (1992). Identification of multiple active growth factors in basement membrane Matrigel suggests caution in interpretation of cellular activity related to extracellular matrix components. Exp Cell Res 202, 1-8. Wallin, A. R., Svanvik, J., Adell, G., and Sun, X. F. (2006). Expression of PRL proteins at invasive margin of rectal cancers in relation to preoperative radiotherapy. Int J Radiat Oncol Biol Phys 65, 452-458. Wang, G. L., and Semenza, G. L. (1993). General involvement of hypoxia-inducible factor 1 in transcriptional response to hypoxia. Proc Natl Acad Sci U S A 90, 4304-4308.  170  Wang, H., Quah, S. Y., Dong, J. M., Manser, E., Tang, J. P., and Zeng, Q. (2007a). PRL-3 down-regulates PTEN expression and signals through PI3K to promote epithelialmesenchymal transition. Cancer Res 67, 2922-2926. Wang, H., Vardy, L. A., Tan, C. P., Loo, J. M., Guo, K., Li, J., Lim, S. G., Zhou, J., Chng, W. J., Ng, S. B., et al. (2010). PCBP1 suppresses the translation of metastasis-associated PRL-3 phosphatase. Cancer Cell 18, 52-62. Wang, J., Kirby, C. E., and Herbst, R. (2002). The tyrosine phosphatase PRL-1 localizes to the endoplasmic reticulum and the mitotic spindle and is required for normal mitosis. J Biol Chem 277, 46659-46668. Wang, L., Peng, L., Dong, B., Kong, L., Meng, L., Yan, L., Xie, Y., and Shou, C. (2006). Overexpression of phosphatase of regenerating liver-3 in breast cancer: association with a poor clinical outcome. Ann Oncol 17, 1517-1522. Wang, L., Shen, Y., Song, R., Sun, Y., Xu, J., and Xu, Q. (2009a). An anticancer effect of curcumin mediated by down-regulating phosphatase of regenerating liver-3 expression on highly metastatic melanoma cells. Mol Pharmacol 76, 1238-1245. Wang, W. Q., Sun, J. P., and Zhang, Z. Y. (2003). An overview of the protein tyrosine phosphatase superfamily. Curr Top Med Chem 3, 739-748. Wang, Y., Li, Z. F., He, J., Li, Y. L., Zhu, G. B., Zhang, L. H., and Li, Y. L. (2007b). Expression of the human phosphatases of regenerating liver (PRLs) in colonic adenocarcinoma and its correlation with lymph node metastasis. Int J Colorectal Dis 22, 1179-1184. Wang, Z., Cai, S. R., He, Y. L., Zhan, W. H., Zhang, C. H., Wu, H., Peng, J. J., Xu, J. B., Zhang, X. H., Wang, L., and Song, W. (2009b). Elevated PRL-3 expression was more frequently detected in the large primary gastric cancer and exhibits a poor prognostic impact on the patients. J Cancer Res Clin Oncol 135, 1041-1046. Wang, Z., He, Y. L., Cai, S. R., Zhan, W. H., Li, Z. R., Zhu, B. H., Chen, C. Q., Ma, J. P., Chen, Z. X., Li, W., and Zhang, L. J. (2008). Expression and prognostic impact of PRL-3 in lymph node metastasis of gastric cancer: its molecular mechanism was investigated using artificial microRNA interference. Int J Cancer 123, 1439-1447. Wani, M. C., Taylor, H. L., Wall, M. E., Coggon, P., and McPhail, A. T. (1971). Plant antitumor agents. VI. The isolation and structure of taxol, a novel antileukemic and antitumor agent from Taxus brevifolia. J Am Chem Soc 93, 2325-2327. Watabe-Uchida, M., Uchida, N., Imamura, Y., Nagafuchi, A., Fujimoto, K., Uemura, T., Vermeulen, S., van Roy, F., Adamson, E. D., and Takeichi, M. (1998). alpha-Cateninvinculin interaction functions to organize the apical junctional complex in epithelial cells. J Cell Biol 142, 847-857. Webster, J., Prager, D., and Melmed, S. (1994). Insulin-like growth factor-1 activation of extracellular signal-related kinase-1 and -2 in growth hormone-secreting cells. Mol Endocrinol 8, 539-544. Weijland, A., Neubauer, G., Courtneidge, S. A., Mann, M., Wierenga, R. K., and SupertiFurga, G. (1996). The purification and characterization of the catalytic domain of Src expressed in Schizosaccharomyces pombe. Comparison of unphosphorylated and tyrosine phosphorylated species. Eur J Biochem 240, 756-764. Wick, W., Grimmel, C., Wild-Bode, C., Platten, M., Arpin, M., and Weller, M. (2001). Ezrin-dependent promotion of glioma cell clonogenicity, motility, and invasion mediated by BCL-2 and transforming growth factor-beta2. J Neurosci 21, 3360-3368.  171  Wilks, A. F. (1990). Structure and function of the protein tyrosine kinases. Prog Growth Factor Res 2, 97-111. Wong, P. C. (2005a) Protein tyrosine phosphatase PRL-3 expression and transcriptional regulation in cancer and metastasis, University of British Columbia, Vancouver. Wong, P. C., Pallen, C. J. (2005b). Cytoplasmic PRL-3 is associated with poor clinical trend in breast and prostate cancers. Wu, H. C., Hsieh, J. T., Gleave, M. E., Brown, N. M., Pathak, S., and Chung, L. W. (1994). Derivation of androgen-independent human LNCaP prostatic cancer cell sublines: role of bone stromal cells. Int J Cancer 57, 406-412. Wu, J. D., Haugk, K., Woodke, L., Nelson, P., Coleman, I., and Plymate, S. R. (2006). Interaction of IGF signaling and the androgen receptor in prostate cancer progression. J Cell Biochem 99, 392-401. Wu, X., Zeng, H., Zhang, X., Zhao, Y., Sha, H., Ge, X., Zhang, M., Gao, X., and Xu, Q. (2004). Phosphatase of regenerating liver-3 promotes motility and metastasis of mouse melanoma cells. Am J Pathol 164, 2039-2054. Xiong, Y., Zhang, H., and Beach, D. (1992). D type cyclins associate with multiple protein kinases and the DNA replication and repair factor PCNA. Cell 71, 505-514. Xu, Y., Zhu, M., Zhang, S., Liu, H., Li, T., and Qin, C. (2010). Expression and prognostic value of PRL-3 in human intrahepatic cholangiocarcinoma. Pathol Oncol Res 16, 169-175. Yamada, S., Pokutta, S., Drees, F., Weis, W. I., and Nelson, W. J. (2005). Deconstructing the cadherin-catenin-actin complex. Cell 123, 889-901. Yamashita, S., Masuda, Y., Matsumoto, K., Okumura, Y., Matsuzaki, H., Kurizaki, T., Haga, Y., Katafuchi, S., Murayama, T., Ikei, S., and Kawahara, K. (2007). Down-regulation of the human PRL-3 gene is associated with the metastasis of primary non-small cell lung cancer. Ann Thorac Cardiovasc Surg 13, 236-239. Yu, L., Kelly, U., Ebright, J. N., Malek, G., Saloupis, P., Rickman, D. W., McKay, B. S., Arshavsky, V. Y., and Bowes Rickman, C. (2007). Oxidative stress-induced expression and modulation of Phosphatase of Regenerating Liver-1 (PRL-1) in mammalian retina. Biochim Biophys Acta 1773, 1473-1482. Zachary, I., and Rozengurt, E. (1992). Focal adhesion kinase (p125FAK): a point of convergence in the action of neuropeptides, integrins, and oncogenes. Cell 71, 891-894. Zeng, H., Datta, K., Neid, M., Li, J., Parangi, S., and Mukhopadhyay, D. (2003a). Requirement of different signaling pathways mediated by insulin-like growth factor-I receptor for proliferation, invasion, and VPF/VEGF expression in a pancreatic carcinoma cell line. Biochem Biophys Res Commun 302, 46-55. Zeng, Q., Dong, J. M., Guo, K., Li, J., Tan, H. X., Koh, V., Pallen, C. J., Manser, E., and Hong, W. (2003b). PRL-3 and PRL-1 promote cell migration, invasion, and metastasis. Cancer Res 63, 2716-2722. Zeng, Q., Hong, W., and Tan, Y. H. (1998). Mouse PRL-2 and PRL-3, two potentially prenylated protein tyrosine phosphatases homologous to PRL-1. Biochem Biophys Res Commun 244, 421-427. Zeng, Q., Si, X., Horstmann, H., Xu, Y., Hong, W., and Pallen, C. J. (2000). Prenylationdependent association of protein-tyrosine phosphatases PRL-1, -2, and -3 with the plasma membrane and the early endosome. J Biol Chem 275, 21444-21452.  172  Zhang, P., Zhang, Z., Li, X., Lei, J., Su, K., Li, X., and Wang, X. (2010). [Expression and its relationship of PRL-3 and RhoC in non-small cell lung cancer]. Zhongguo Fei Ai Za Zhi 13, 598-601. Zhang, Z. Y. (1998). Protein-tyrosine phosphatases: biological function, structural characteristics, and mechanism of catalysis. Crit Rev Biochem Mol Biol 33, 1-52. Zhang, Z. Y. (2003). Mechanistic studies on protein tyrosine phosphatases. Prog Nucleic Acid Res Mol Biol 73, 171-220. Zhang, Z. Y., Palfey, B. A., Wu, L., and Zhao, Y. (1995). Catalytic function of the conserved hydroxyl group in the protein tyrosine phosphatase signature motif. Biochemistry 34, 1638916396. Zhao, W. B., Li, Y., Liu, X., Zhang, L. Y., and Wang, X. (2008). Evaluation of PRL-3 expression, and its correlation with angiogenesis and invasion in hepatocellular carcinoma. Int J Mol Med 22, 187-192. Zheng, P., Liu, Y. X., Chen, L., Liu, X. H., Xiao, Z. Q., Zhao, L., Li, G. Q., Zhou, J., Ding, Y. Q., and Li, J. M. (2010). Stathmin, a New Target of PRL-3 Identified by Proteomic Methods, Plays a Key Role in Progression and Metastasis of Colorectal Cancer. J Proteome Res. Zhou, B. P., Deng, J., Xia, W., Xu, J., Li, Y. M., Gunduz, M., and Hung, M. C. (2004). Dual regulation of Snail by GSK-3beta-mediated phosphorylation in control of epithelialmesenchymal transition. Nat Cell Biol 6, 931-940. Zhou, H., Gallina, M., Mao, H., Nietlispach, D., Betz, S. F., Fetrow, J. S., and Domaille, P. J. (2003). 1H, 13C and 15N resonance assignments and secondary structure of the human protein tyrosine phosphatase, PRL-2. J Biomol NMR 27, 397-398. Zhou, J., Wang, S., Lu, J., Li, J., and Ding, Y. (2009). Over-expression of phosphatase of regenerating liver-3 correlates with tumor progression and poor prognosis in nasopharyngeal carcinoma. Int J Cancer 124, 1879-1886. Zhu, G. F., Cheng, H. P., Wu, K. J., Zhang, L. L., Zhu, G. D., Zhang, D., and He, D. L. (2008). [Transforming growth factor beta upregulates the expression of invasion and metastasis associated proteins in prostate cancer LNCaP cell lines in vitro]. Zhonghua Nan Ke Xue 14, 238-241. Zou, W., Yan, M., Xu, W., Huo, H., Sun, L., Zheng, Z., and Liu, X. (2001). Cobalt chloride induces PC12 cells apoptosis through reactive oxygen species and accompanied by AP-1 activation. J Neurosci Res 64, 646-653.  173  Appendix 1: Supplemental figures  Figure A.1: PRL-3 expression has little to no effect on p21 and p53 expression in 293 cells. Parental and FLAG-PRL-3 (wt) cells were cultured in the presence of doxycycline (1 µg/ml) for 24 hours and then stimulated with doxorubicin (Doxo) (200 ng/ml) for 24 hours (in the presence of doxycycline). Lysates were probed with p53 and p21 antibodies, and with FLAG antibodies to detect FLAG-PRL-3 transgene expression. These lysates were reprobed for actin as a loading control. Both p53 and p21 are slightly reduced in untreated PRL-3-expressing cells. Similar effects were observed for p21 in two other experiments and for p53 in one other experiment.  174  Figure A.2: A partial EMT phenotype is gained by LNCaP cells treated with cobalt chloride or TGF-β. LNCaP cells were seeded onto 10-cm plates and cultured overnight (~16 hours). Cells were then treated with cobalt chloride (600 µM, A) or TGF-β (5 ng/ml, B) or left untreated (-) and then incubated for the times indicated. Lysates were probed for fibronectin (FN) or E-cad and reprobed for actin as a loading control. For cobalt chloride treatment, similar effects were observed in one other experiment.  175  Figure A.3: Expression of PRL-3 reduces 293 cell adhesion to fibronectin. Parental and FLAG-PRL-3 (wt) cells (1x105) were seeded in quadruplicate in 96-well plates that were precoated with fibronectin (20 µg/ml) and allowed to adhere for 60 minutes. Cells were gently rinsed in PBS, stained with 0.01% crystal violet for 30 minutes and then rinsed with PBS. The cells were destained for 30 minutes in 10% acetic acid. The absorbance at 600 nm of the destaining solution was determined. * p<0.05, n=3.  176  Figure A.4: Expression of PRL-3 reduces levels of GTP-bound Rac1 in serumstarved cells but increases GTP-bound Rac1 in growing cells. Parental and FLAG-PRL-3 (wt) cells (2.5x106) were seeded in 10-cm dishes and cultured overnight (~16 hours). After serum starving the cells for two hours (+) or leaving them in full serum, cells were lysed. Lysates were incubated with the p21-binding domain of PAK fused to GST to precipitate GTP-bound Rac1 and the complexes probed with antiRac1 antibody. Whole cell extracts were also probed with anti-Rac1 antibody to determine levels of Rac1 protein (both GDP- and GTP-bound). B. The ratio of GTPbound Rac1 vs. total Rac1 of PRL-expressing cells compared to parental cells is shown. * 0.60 ± 0.17, p<0.005, n=3; ** 2.71 ± 1.34, p<0.05, n=3.  177  Figure A.5: Expression of PRL-3 reduces levels of phosphorylated Erk in response to serum stimulation of 293 cells. Cells were cultured overnight (~16 hours) in 1 µg/ml of doxycycline then serum-starved for 24 hours. Cells were stimulated with 10% FBS for 10 minutes, and the lysates were probed for phospho-Erk and then reprobed for Erk as a loading control.  178  Figure A.6: Kinetworks screen of phosphoproteins in IGF-I-treated Flp-In T-Rex 293 PRL-3 cells vs. parental cells. Cells (3 x 106) were seeded onto 10-cm plates and grown for 24 hours in the presence of doxycycline (1 μg/ml). Cells were starved for 24 hours and then stimulated with 100 ng/ml IGF-I for 10 minutes. Lysates were sent to Kinexus to perform a Kinetworks Custom Profiling Immunoblot screen (KCPS 1.0) of a number of phosphoproteins. Immunoblots from 293 control cells (A) or PRL-3 (wt) 293 cells (B) treated with IGF-I. Lanes 1 and 21 are molecular marker lanes, lanes 2-20 were probed with various antibodies against phosphoproteins. C. The intensity of select bands was analyzed by densitometry. The counts per minute from the detection of the bands were normalized against background and graphed. The data labels represent the % difference of the phosphoprotein expression of the PRL (wt)-expressing cells vs. the 293 control cells. The lane of the corresponding phosphoprotein is indicated as is the molecular weight (in brackets) of select phosphoproteins. D. The complete list of the phosphoproteins and the sites recognized by the antibodies used in the screen. The lane and band number of each phosphoprotein from the respective trace (in this case, the 293 control cells treated with IGF-I) are listed. Unknown bands (Unclassified) with strong signal in response to imunoblotting with specific phosphoprotein antibodies are also indicated.  179  180  181  182  Figure A.7: Differential expression of mRNAs in PRL-3 (wt)-expressing vs. unexpressing 293 cells. PRL-3 (wt)-expressing 293 cellsCells (3 x 106) were seeded onto 10-cm plates and grown for 24 hours. Cells were then left untreated or treated for doxycycline (1 μg/ml) for 6 or 24 hours. In all cases, the cells were grown for a total of 24 more hours (i.e. untreated cells grown for 24 hours; 6 hour doxycycline treatment grown for 18 hours, then doxycycline added for 6 hours). Total mRNA was collected and sent to the Microarray Facility in the Centre for Applied Genomics at the Hospital for Sick Children in Toronto, Canada to perform the Human Genome U133 Plus 2.0 Array microarrary experiment using an Affymetrix GeneChip Scanner 3000. A. Summary of gene expression changes induced by six-hour or 24-hour PRL-3 expression compared to uninduced 293 cells (0 hour) or six-hour PRL-3 expression. Fold change is indicated. D, decreased gene expression; I, increased gene expression. B. Short descriptions of the genes that are regulated by PRL-3 expression in the 293 cell system.  183  184  185  186  187  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0071524/manifest

Comment

Related Items